[01-Sep-2009 00:00:46] [connected at Tue Sep 1 00:00:46 2009]
[01-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[01-Sep-2009 06:52:31] kevin7kal_ is now known as kevin7kal
\n[01-Sep-2009 07:29:33] <etank> is there a way to get a graph /\ndata point of the return from SQL Transactions instead of just the\ntime it took the transaction to run?
[01-Sep-2009 08:52:25] <jb> morning..
[01-Sep-2009 08:53:05] <Apocalipse> hello
[01-Sep-2009 08:59:53] <rmatte> if he had waited another 10 minutes I would have been here (Apocalipse was messaging me)
[01-Sep-2009 09:17:34] <jb> hey rmatte
[01-Sep-2009 09:18:05] <rmatte> hey
[01-Sep-2009 09:18:14] <jb> need some report assistance
[01-Sep-2009 09:18:23] <jb>
[01-Sep-2009 09:18:51] <jb> i have a zenpack that graphs throughput of virtual servers on my F5 LTM's
[01-Sep-2009 09:19:05] <jb> one website may contain 4 virtual servers or so
[01-Sep-2009 09:19:09] <jb> which all have separate graphs
\n[01-Sep-2009 09:19:28] <jb> i want to make a combination graph\nthat takes the sum of all the virtual servers for one website's\nthroughput and graphs them on one graph
[01-Sep-2009 09:20:06] <rmatte> jb: hmmm, you're basically looking to make an aggregate report
[01-Sep-2009 09:20:24] <jb> ah ok
\n[01-Sep-2009 09:20:29] <rmatte> hold on, I'm just testing\nsomething then I'll see if I can find some reference material for you
[01-Sep-2009 09:20:36] <jb> thanks
[01-Sep-2009 09:38:48] <rmatte> jb: crap, I have a meeting that I'm late for apparently
[01-Sep-2009 09:38:51] <rmatte> I'll be back later
[01-Sep-2009 09:39:22] <jb> np
[01-Sep-2009 10:01:05] <cgibbons> hmmmm
[01-Sep-2009 10:56:42] <mrayzenoss> heading to a meeting, 2.4.5 is trickling out right now
[01-Sep-2009 11:16:33] crow is now known as ckrough
\n[01-Sep-2009 11:44:58] <jsm> so other than installing the dell\nmonitor zenpack and it's dependency, is there anything else i have to\ndo to see the newly modeled data other than re-modeling the device ? i\ndon't see any new data
[01-Sep-2009 11:45:22] <mrayzenoss> did you bind the template?
[01-Sep-2009 11:46:42] <jsm> yes, they are all bound to my /devices/server/linux class
[01-Sep-2009 11:47:03] <mrayzenoss> does it have a custom zCollectorPlugin?
[01-Sep-2009 11:48:22] <jsm> they are in the zcollectorplugins for that class
[01-Sep-2009 11:48:43] <mrayzenoss> and they're set?
[01-Sep-2009 11:48:53] <mrayzenoss> added to that device class, that is?
[01-Sep-2009 11:49:01] <mrayzenoss> you'll need to remodel after that
[01-Sep-2009 11:51:19] <jsm> if by set, you mean if i look in /Servers/Linux collector plugins, they are alll listed
[01-Sep-2009 11:51:42] <jsm> and the device has been remodelled
[01-Sep-2009 11:52:40] <mrayzenoss> I don't have any experience with that ZenPack, maybe someone else here has used it?
[01-Sep-2009 11:52:47] <mrayzenoss> or perhaps post on the forums?
[01-Sep-2009 11:52:48] <mrayzenoss> sorry
[01-Sep-2009 11:54:41] <jsm> It's somewhat disconcerting that so many of the community zenpacks have 0 documentation
[01-Sep-2009 11:54:49] <jsm> should make it a requirement from the author.
[01-Sep-2009 11:56:04]\n<jsm> Egor, while contributing a lot of code, from what i can\nfind, doesn't give much information on how to actually use the plugins.
[01-Sep-2009 11:59:58] <ckrough> Not that I disagree, but the argument usually becomes "it's open source, contribute".
[01-Sep-2009 12:01:06] <cwj> but first, use your magic divining rod to figure out how it works, so you can!
[01-Sep-2009 12:04:14]\n<jsm> the difference is in this case mray has the ability to\naccept projects.. if you accept a projects w/out documentation, you are\nbenefiting no one
[01-Sep-2009 12:05:03] <jsm> and i'd love to contribute, but that's hard when the damn thing doesn't work.
\n[01-Sep-2009 12:12:28] <aceazza> Can someone help me figure out\nhow to patch my zenoss server. I'm running 2.4.2 and I want to get to\nthe latest version. I've been searching the web but can't find much\ninstruction for applying patches. I did find something that says use\nzenpatch <patch_number> . Where can I download the patch files?\nThanks!
[01-Sep-2009 12:13:56] <ckrough> aceazza: Are you up on the issues with 2.4.3?
[01-Sep-2009 12:14:10] <aceazza> No
[01-Sep-2009 12:14:11] <ckrough> I think they are rolling out 2.4.5(?) currently
[01-Sep-2009 12:14:18] <ckrough> see the the /topic
[01-Sep-2009 12:14:25] <aceazza> Yeah, I saw 2.4.5 available for download
[01-Sep-2009 12:14:33] <ckrough> You may want to check the forums and see where everything is at before doing the upgrade
[01-Sep-2009 12:18:28] <jb> jsm.. are your dell boxes running openmanage?
[01-Sep-2009 12:18:36] <jsm> jsm, yes
[01-Sep-2009 12:18:43] <jsm> err, jb
[01-Sep-2009 12:18:49] <jb> what plugins are listed when you model the device?
[01-Sep-2009 12:19:30] <jb> i may help egor out by writing some documentation
[01-Sep-2009 12:19:40] <jb> but, theres not much to it really
[01-Sep-2009 12:20:00] <jsm> jb: http://pastebin.com/m10ca41ef
\n[01-Sep-2009 12:25:36] <tripitaka> Hi, my zenoss server is\nrunning very slow, what are the usual causes? It has 8GB of 16GB RAM\nfree, the CPU ( 2 quad core opterons) are usually around 70% idle, what\nelse should I be looking at?
[01-Sep-2009 12:26:05] <jb> jsm: and nothing in the hardware tab?
[01-Sep-2009 12:26:17] <tripitaka> (2.4.2 on rhel5.3 x86_64)
[01-Sep-2009 12:26:21] <jb> tripitaka: how many devices?
[01-Sep-2009 12:26:41] <tripitaka> just under 2,000
[01-Sep-2009 12:26:45] <jb> hmm
[01-Sep-2009 12:26:52] <jb> have you checked I/O?
[01-Sep-2009 12:27:16] <jb> check iowait..
[01-Sep-2009 12:27:24] <jb> what type of storage are you using for zenoss?
[01-Sep-2009 12:27:45] <tripitaka> internal SAS, 15k rpm
[01-Sep-2009 12:27:58] <tripitaka> top shows about 1% in wait
[01-Sep-2009 12:28:20] <jb> what type of raid?
[01-Sep-2009 12:28:33] <tripitaka> iostat shows up to 10,000 blockes written/sec
[01-Sep-2009 12:28:39] <tripitaka> raid 1+0, two mirrored disks
[01-Sep-2009 12:28:43] <jb> ah
[01-Sep-2009 12:28:45] <jb> k
[01-Sep-2009 12:29:06] <tripitaka> what's it writing to? mysql?
[01-Sep-2009 12:29:13] <jb> for events, yes
[01-Sep-2009 12:29:25] <jb> is EVERYTHING slow, or just certain things?
[01-Sep-2009 12:29:49] <aceazza> So is apt-get upgrade zenoss-stack the way to update on an Ubuntu system?
[01-Sep-2009 12:30:02] <jb> s
[01-Sep-2009 12:30:05] <tripitaka> depends, nearly all mouse clicks take a long time to respond, 30 - 90 seconds and then some timeouts
[01-Sep-2009 12:30:14] <tripitaka> but occasionally it's snappy
\n[01-Sep-2009 12:30:37] <tripitaka> modelling and device adds\noften fail with timeout connecting to zenhub, but work on retry
[01-Sep-2009 12:31:19] <jb> are you runnning core?
[01-Sep-2009 12:31:23] <tripitaka> stack
[01-Sep-2009 12:31:33] <jb> zenoss core, or enterprise?
[01-Sep-2009 12:31:38] <tripitaka> oh, core
[01-Sep-2009 12:31:42] <jb> k
[01-Sep-2009 12:32:01]\n<jb> im not sure.. i really don't have any experience with a\ncollector having that many devices.. i know that there are people on\nthe forums that do
[01-Sep-2009 12:32:03] <tripitaka> currently 2.4.2, I tried 2.4.3 last week but there were problems with processes dying
[01-Sep-2009 12:32:06] <jb> and several threads discussing performance
\n[01-Sep-2009 12:32:48] <tripitaka> about half the devices are in\nclasses I don't want to actively monitor - can I completely disable all\nmonitoring for a given class?
[01-Sep-2009 12:33:10]\n<tripitaka> I've tried removing a classes collectors, and\nunbinding all templates, hopefully that would do it
[01-Sep-2009 12:33:19] <tripitaka> but I suspect it's still pinging them
[01-Sep-2009 12:34:12] <jb> set zPingMonitorIgnore to "Yes"
[01-Sep-2009 12:34:19] <jb> on the classes that you dont want to ping
[01-Sep-2009 12:34:23] <aceazza> zenoss core
\n[01-Sep-2009 12:34:26] <tripitaka> hmm, yep, it's still snmp\npolling a device class after I've unbound the 'device' and set the\nzPingMonitorIgnore to 'yes'
[01-Sep-2009 12:34:31] <jb> as well as zSnmpMonitorIgnore
[01-Sep-2009 12:34:53] <tripitaka> I've set those in the class' zProperties, it's still walking them
[01-Sep-2009 12:35:02]\n<cgibbons> you can change hte production state of a device class\nas well; that's the first thing that the system does to see if a device\nshould be monitored by the various daemons or not
[01-Sep-2009 12:35:02] <jb> and zWmiMonitorIgnore if you need..
[01-Sep-2009 12:35:16] <jb> ah yeap
[01-Sep-2009 12:35:39] <tripitaka> oh, wait, I am a moron - I only thought I had
\n[01-Sep-2009 12:36:16] <cgibbons> what we do here is have a\nIgnored device class that has all the interesting bits disabled so the\ndevices in there are completely, well, ignored
\n[01-Sep-2009 12:36:40] <rmatte> hmmmm, am I crazy, or should\nZenoss be automatically detecting IpServices that are set to monitored\nwhen modeling?
[01-Sep-2009 12:39:10] <tripitaka> ok.. all boring systems set to a nonproduction state, and all ignores turned on..
[01-Sep-2009 12:41:10] <tripitaka> load average is dropping consistently
[01-Sep-2009 12:41:19] <tripitaka> yay
[01-Sep-2009 12:45:11] <jb>
[01-Sep-2009 12:45:15] <jb> rmatte: yeah
\n[01-Sep-2009 12:47:32] <rmatte> well, I have 1 IpService that's\nbeing picked up fine, and another 2 that aren't, during a modelling
\n[01-Sep-2009 12:47:50] <rmatte> all 3 ports are publicly\navailable on the device, and there's no difference in config on any of\nthe IpServices
[01-Sep-2009 12:47:59] <rmatte> so I'm stumped
[01-Sep-2009 13:27:16] <markeriv> What exactly about 2.4.3 is broken?
[01-Sep-2009 13:35:22] <jb> event clear email alerts..
[01-Sep-2009 13:36:21] <mrayzenoss> and zencommand clears
[01-Sep-2009 13:46:20] <ckrough> is 2.4.5 out in the wild now?
[01-Sep-2009 13:47:25] <rmatte> it's been released yeh
\n[01-Sep-2009 13:47:53] <rmatte> but if you like being able to set\nhardware product type/info for each device you might want to skip this\nupgrade
[01-Sep-2009 13:54:56] <mrayzenoss> it's out
[01-Sep-2009 13:55:08] <mrayzenoss> ckrough: Enterprise will probably be tomorrow
[01-Sep-2009 13:57:13] <ckrough> thanks. was referring to CORE though
\n[01-Sep-2009 13:57:41] <ckrough> Im going to be helping someone\nget started in Zenoss, was debating on 2.4.5 or the previous stable\nrelease
[01-Sep-2009 13:58:00] <ckrough> some of the bugfixes were significant though, so we'll probably lean towards 4.5
[01-Sep-2009 13:58:20] <mrayzenoss> yeah, we've had a problem with 3 steps forward, 1 step back
[01-Sep-2009 13:59:18] <ckrough> You definitely don't get to claim exclusivity on that one
[01-Sep-2009 14:17:06] <markeriv> hey guys
[01-Sep-2009 14:17:14] <markeriv> i am getting this error when i try those zenpatches
[01-Sep-2009 14:17:21] <markeriv> Hunk #1 FAILED at 273.
[01-Sep-2009 14:17:28] <markeriv> Hunk #2 FAILED at 304.
[01-Sep-2009 14:17:35] <markeriv> Hunk #3 succeeded at 314 with fuzz 1 (offset -13 lines).
[01-Sep-2009 14:17:45] <markeriv> 2 out of 3 hunks FAILED -- saving rejects to file ZenEvents/zenactions.py.rej
[01-Sep-2009 14:20:03] <mrayzenoss> markeriv: were you patching 2.4.3?
[01-Sep-2009 14:20:37] <markeriv> yes
\n[01-Sep-2009 14:21:41] <mrayzenoss> I gotta run, but someone else\nhad a similar problem the other day, they were patching the wrong\nversion
[01-Sep-2009 14:21:48] <markeriv> ahh
[01-Sep-2009 14:21:51] <mrayzenoss> 2.4.5 is out, that rolls up the patches anyway
[01-Sep-2009 14:22:01] <mrayzenoss> be back in a bit
[01-Sep-2009 14:22:46] <markeriv> kk
[01-Sep-2009 14:50:20] <dollarbang> Just got the release notes for 2.4.5, are they skipping 2.4.4?
[01-Sep-2009 14:52:20] <rmatte> correct
[01-Sep-2009 14:52:32] <rmatte> they are basically skipping 2.4.3 too
[01-Sep-2009 14:53:16] <rmatte> 2.4.4 has patch 14744 and 2.4.5 has patch 14769
[01-Sep-2009 15:05:21] <dollarbang1> an interesting way of generating versions. different from the world I'm leaving and entering.
[01-Sep-2009 15:06:53]\n<dollarbang1> btw....how does one ftp from the zenoss site to a\nlinux server? I tried ftp ftp.zenoss.com, but that failed, so I 'm\nguessing that zenoss isn't allowing ftp except for using a browser.
[01-Sep-2009 15:07:33] <chudler> Host ftp.zenoss.com not found: 3(NXDOMAIN)
[01-Sep-2009 15:17:08] <chudler> in other news.. gmail appears to be down
[01-Sep-2009 15:17:20] <jsm> gmail been going down lately
[01-Sep-2009 15:17:49] <jsm> may have to switch my account back to in house
[01-Sep-2009 15:18:02] <jsm> gmail is more unstable than my 5 year old mail server
[01-Sep-2009 15:22:57] <cgibbons> hmmm usually it's just a portion of their cloud that's dead but this one seems systemic
[01-Sep-2009 15:23:27] <jsm> i've been trialing theme for the past few months. not impressed
[01-Sep-2009 15:23:32] <jsm> them
[01-Sep-2009 15:23:44] <jsm> and i have the pay version
[01-Sep-2009 15:28:52] <rmatte> dollarbang1: why would you need to?
[01-Sep-2009 15:29:05] <rmatte> dollarbang1: are you downloading a file, or trying to upload?
[01-Sep-2009 16:12:53] <cgibbons> so interesting, i can get to gmail just fine via imap, but not via the web
\n[01-Sep-2009 16:33:53] <gwb235> so... if i have a data source in\na performance template (nagiosCheckPing) that is a command, how/where\ndo I start tracing/debugging why the associated graph isn't getting\nupdated? (assuming the command works fine on the command line)
[01-Sep-2009 16:34:22] <gwb235> or is it too late in the day for that question?
[01-Sep-2009 16:34:46] <mrayzenoss> never too late, it depends who's around
[01-Sep-2009 16:35:07] <mrayzenoss> no graph, place holder for a graph, or graph with no data points?
[01-Sep-2009 16:35:22] <mrayzenoss> no graph means the template isn't bound
[01-Sep-2009 16:35:43] <mrayzenoss> place holder and no data points means trouble with the data source and data points
\n[01-Sep-2009 16:35:57] <gwb235> graph exists, used to gather data\n(i.e. looking over day history shows data used to be gathered), but now\ngraph shows cur: nan, avg: nan, max: nan
[01-Sep-2009 16:36:01] <mrayzenoss> does the "Test" command work?
[01-Sep-2009 16:36:51] <gwb235> yes (DONE in 4 seconds)
[01-Sep-2009 16:37:37] <mrayzenoss> absolute path or relative path on the data source?
[01-Sep-2009 16:37:55] <gwb235> absolute path /opt/zenoss/libexec/check_ping (blah blah)
[01-Sep-2009 16:38:50] <rsr> hi
[01-Sep-2009 16:38:57] <gwb235> the graph itself has 2009-09-01 16:38:31 in the output
[01-Sep-2009 16:39:13] <mrayzenoss> hmmm.. I assume it's been running for awhile?
[01-Sep-2009 16:39:23] <gwb235> few weeks
[01-Sep-2009 16:39:36] <gwb235> we're still in "deployment" (replacing nagios) stage
[01-Sep-2009 16:39:52] <mrayzenoss> yeah, I'm blowing out the install that I had that was using that ZenPack...
[01-Sep-2009 16:40:02]\n<rsr> I am getting an error when trying to access zenoo from the\nbrowser. I have just finished installing it in linux and the error I\nreceive is : "127.0.0.1" refused the connection. I have no firewall\nactive.
[01-Sep-2009 16:40:05] <mrayzenoss> and actually I didn't install it, I just followed the instructions
[01-Sep-2009 16:40:31] <gwb235> if there's a better (more supported) way of graphing ping loss/RTA i'm open to it...
[01-Sep-2009 16:40:31] <mrayzenoss> rsr: so when you try http://yourhost:8080 that comes up?
[01-Sep-2009 16:40:37] <rsr> yes
[01-Sep-2009 16:40:52] <rsr> any ideas?
[01-Sep-2009 16:41:01] <mrayzenoss> gwb235: there's a half-baked fping collector in the dev docs, but nothing concrete
[01-Sep-2009 16:41:12] <mrayzenoss> rsr: IE7?
[01-Sep-2009 16:41:16] <rsr> no
[01-Sep-2009 16:41:27] <rsr> epiphany
[01-Sep-2009 16:41:33] <rsr> a webkit based gnome browser
[01-Sep-2009 16:41:44] <rsr> Its running on linux
[01-Sep-2009 16:41:45] <mrayzenoss> rsr: yeah, try checking your /etc/hosts
[01-Sep-2009 16:41:53] <mrayzenoss> I'm guessing there's something in there
[01-Sep-2009 16:41:58] <rsr> I am trying from localhost
[01-Sep-2009 16:42:09] <mrayzenoss> rsr: try the hostname:8080 instead
[01-Sep-2009 16:42:15] <rsr> ok
[01-Sep-2009 16:42:33] <rsr> samething
[01-Sep-2009 16:42:55] <mrayzenoss> rsr: what's in your /etc/hosts for localhost and your hostname?
[01-Sep-2009 16:43:05] <mrayzenoss> gwb235: I just followed the instructions from the PDF, didn't even bother to install the ZenPack
[01-Sep-2009 16:43:41] <rsr> 127.0.0.1 localhost
[01-Sep-2009 16:43:41] <mrayzenoss> gwb235: are you using the ${here/manageIp} instead of the $devname?
[01-Sep-2009 16:43:41] <rsr> 127.0.1.1 Zioncell
[01-Sep-2009 16:44:08] <gwb235> no, just $devname
[01-Sep-2009 16:44:10] <mrayzenoss> rsr: try putting 127.0.0.1 localhost and your ip with the hostname
[01-Sep-2009 16:44:17] <mrayzenoss> gwb235: does $devname resolve?
[01-Sep-2009 16:44:38] <mrayzenoss> gwb235: if it was discovered via SNMP, it may return something not resolvable in DNS
[01-Sep-2009 16:44:49] <mrayzenoss> at least, on my home network I found that
[01-Sep-2009 16:44:58] <mrayzenoss> since I just have a hand-coded /etc/hosts
[01-Sep-2009 16:45:06] <gwb235> host <device> returns <device> has address x.x.x.x
[01-Sep-2009 16:45:11] <gwb235> (on the collector)
[01-Sep-2009 16:45:59] <mrayzenoss> gwb235: I assume you've tried "zencommand run -v10 -d yourhost" ?
[01-Sep-2009 16:46:01] <rsr> mrayzenoss, no go... sorry
[01-Sep-2009 16:46:13] <rsr> why would it be refusing the connection?
[01-Sep-2009 16:47:07] <mrayzenoss> is it running?
[01-Sep-2009 16:47:17] <gwb235> yes, i have tried zencommand run (no errors, just DEBUG and INFO)
[01-Sep-2009 16:47:33] <rsr> how can I check? ps aux | grep zenoss?
[01-Sep-2009 16:47:47] <mrayzenoss> rsr: yeah, or a bunch of python commands
[01-Sep-2009 16:47:58] <rsr> I guess not
[01-Sep-2009 16:47:58] <mrayzenoss> gwb235: is this an upgrade?
[01-Sep-2009 16:48:12] <rsr> how do I start it? zenity?
[01-Sep-2009 16:48:12] <gwb235> no, 2.4.2
[01-Sep-2009 16:48:27] <mrayzenoss> rsr: as the 'zenoss' user, do a 'zenoss status'
[01-Sep-2009 16:49:31] <rsr> zenoss command not found
\n[01-Sep-2009 16:49:35] <mrayzenoss> gwb235:\ntrac-zenpacks/wiki/CommunityFAQGeneral?version=90#Upgradingto2.4.xcausedmycommandbaseddatasourcesZenCommandtonotworkwhatdoIdo\n?
[01-Sep-2009 16:49:44] <mrayzenoss> gwb235: I'm grasping at straws
[01-Sep-2009 16:50:01] <rsr> zenoss-stack is installed though
[01-Sep-2009 16:50:04] <mrayzenoss> gwb235: is it just this 1 box that isn't working?
[01-Sep-2009 16:50:09] <mrayzenoss> rsr: which user are you?
[01-Sep-2009 16:50:22] <gwb235> multiple boxes
\n[01-Sep-2009 16:50:24] <gwb235> since we're still in deployment\nphase, is there a better suggested way to get ping RTA/loss
[01-Sep-2009 16:50:54] <rsr> mrayzenoss, ok got it
[01-Sep-2009 16:50:59] <rsr> mrayzenoss, nothings running
[01-Sep-2009 16:51:08] <rsr> sorry its my first time running zenoss
[01-Sep-2009 16:51:17] <mrayzenoss> gwb235: well, there's http://www.zenoss.com/community/projects/zenpacks/fping
[01-Sep-2009 16:51:22] <rsr> man zenoss
[01-Sep-2009 16:51:27] <mrayzenoss> heh
[01-Sep-2009 16:51:30] <rsr> that will help
[01-Sep-2009 16:51:39] <mrayzenoss> rsr: http://www.zenoss.com/community/docs/
[01-Sep-2009 16:51:51] <mrayzenoss> rsr: the Getting Started Guide and Install Guides should help
[01-Sep-2009 16:52:14] <rsr> mrayzenoss, starting the daemon for the first time isnt on the getting started guide
\n[01-Sep-2009 16:52:17] <gwb235> confirmed that zSshSession is set\nto 10 on the template for the device tree i'm working on right now
[01-Sep-2009 16:52:32] <gwb235> i mean, zproperties
[01-Sep-2009 16:52:45] <gwb235> zSshConcurrentSessions
[01-Sep-2009 16:53:10] <rsr> gotit! thanks so much for your help mrayzenoss !
[01-Sep-2009 16:54:00] <mrayzenoss> rsr: I guess the Install Guide comes before the Getting Started Guide
[01-Sep-2009 16:54:22] <rsr> is it on the install guide?
[01-Sep-2009 16:54:36] <mrayzenoss> definitely
[01-Sep-2009 16:54:41] <rsr> I missed it I guess, since I only read the relevant deb doc
[01-Sep-2009 16:54:42] <mrayzenoss> http://www.zenoss.com/community/docs/installation-guide/2.4.2/ch01s02.html
[01-Sep-2009 16:55:05] <mrayzenoss> Deb? http://www.zenoss.com/community/docs/installation-guide/2.4.2/ch04s01.html
[01-Sep-2009 16:56:33] <mrayzenoss> gwb235: so some boxes work and some don't?
[01-Sep-2009 16:56:40] <gwb235> correct
[01-Sep-2009 16:57:08] <gwb235> i'm doing a $ while true; do date; ps auxwwww | grep check_ping | grep -v grep; sleep 1; done;
[01-Sep-2009 16:57:19] <gwb235> and can see some check_pings, but not others
[01-Sep-2009 16:58:34] <mrayzenoss> gwb235: well... 2.4.5 is out :-\\
[01-Sep-2009 16:59:02]\n<gwb235> but even (some of) those that i see in the output (from\nabove) don't have a graph updated with values (although the graph is\nupdated with timestampe when I reload perf tab)
\n[01-Sep-2009 16:59:37] <gwb235> back to the original question,\nhow do i trace a command data source flow from command execution to\ninsertion into RRD?
[01-Sep-2009 16:59:48] <gwb235> or is that waaay to tricky for 4.59PM
[01-Sep-2009 17:01:25] <mrayzenoss> gwb235: usually the 'zencommand run...' is the way to do that
[01-Sep-2009 17:01:42] <mrayzenoss> gwb235: maybe check the other logs, like zenhub.log and make sure everything is ok there?
[01-Sep-2009 17:01:53] <mrayzenoss> has zenhub ever been restarted?
[01-Sep-2009 17:01:57] <mrayzenoss> and zencommand?
[01-Sep-2009 17:02:12] <gwb235> yeah... about an hour before I popped in here...
[01-Sep-2009 17:02:20]\n<mrayzenoss> or if you're overly ambitious, cgibbons wrote this:\nhttp://blog.nuclearbunny.org/2009/03/19/getting-a-native-code-stack-trace-from-a-zenoss-daemon/
[01-Sep-2009 17:02:43] <mrayzenoss> but that's probably not what we're seeing
[01-Sep-2009 17:03:42]\n<gwb235> what should I be seeing? if i run zencommand run on a\nhost that works and a host that doesn't (graph w/ data vs. graph w/out\ndata) the output appears the same
[01-Sep-2009 17:04:18] <mrayzenoss> you should see the commands getting called I believe
[01-Sep-2009 17:04:32] <mrayzenoss> go to Settings->Daemons and edit the config for zencommand, crank up the logging there
[01-Sep-2009 17:04:52] <mrayzenoss> and restart it and at least it'll be in the $ZENHOME/logs/
[01-Sep-2009 17:05:06] <mrayzenoss> change that, then restart the stack
[01-Sep-2009 17:05:17] <mrayzenoss> and start grepping for your missing host...
[01-Sep-2009 17:05:45]\n<mrayzenoss> the good thing is cgibbons put in some more\ndebugging stuff for 2.5 last week, not that that helps us today
[01-Sep-2009 17:09:50] <gwb235> if the device's performance monitor is a collector, is that the one I should be modifying ?
[01-Sep-2009 17:10:35] <mrayzenoss> yes
[01-Sep-2009 17:10:50] <mrayzenoss> zencommand runs SSH and Nagios stuff
[01-Sep-2009 17:12:43] <gwb235> is logseverity 10 what i should set to on the collector
[01-Sep-2009 17:12:50] <mrayzenoss> yeah
\n[01-Sep-2009 17:15:24] <gwb235> ok... so devices that are being\nperfmon from the hub are graphing ok, but devices w/ a perfmon\ncollector are not.
[01-Sep-2009 17:15:41] <gwb235> (looking more closely at the two zencommand logfiles)
[01-Sep-2009 17:17:28] <sergeymasushko> where I can change/remove scale on a graph?
[01-Sep-2009 17:17:59] <mrayzenoss> which scale?
[01-Sep-2009 17:18:28] <mrayzenoss> gwb235: perfmon? The Perl-based one?
[01-Sep-2009 17:18:42] <gwb235> i meant "performance monitor" (collector)
[01-Sep-2009 17:18:42] <mrayzenoss> gwb235: or the zenwinperf?
[01-Sep-2009 17:18:51] <mrayzenoss> ahh, overloaded terminology
[01-Sep-2009 17:19:02] <gwb235> i.e. those devices that have collector: localhost (hub) are graphing OK
[01-Sep-2009 17:19:11] <gwb235> those devices that have collector cornea (collector) are not
[01-Sep-2009 17:19:19] <mrayzenoss> gwb235: Core or Enterprise?
[01-Sep-2009 17:19:23] <gwb235> sorry I didn't realize that earlier
[01-Sep-2009 17:19:24] <gwb235> enterprise
[01-Sep-2009 17:19:44] <gwb235> you want me to dump a bug in and see what they say?
[01-Sep-2009 17:19:54] <mrayzenoss> yeah, that'll get you more help than me
[01-Sep-2009 17:20:07] <mrayzenoss> I don't run Enterprise myself
[01-Sep-2009 17:20:15] <mrayzenoss> even though I work for Zenoss
[01-Sep-2009 17:20:24] <mrayzenoss> I eat the Community dogfood
\n[01-Sep-2009 17:23:19] <cgibbons> hrm speaking of that, i forgot\nthat zenoss was running on that little server i had when it took a\nheader
[01-Sep-2009 17:23:26] <cgibbons> guess i wasn't using that instance much
\n[01-Sep-2009 17:29:31] <sergeymasushko> mrayzenoss: hi Matt, for\nexample I added monitoring of 'laLoadInt1' and on the graph I see wrong\nscale: http://i25.tinypic.com/28un19y.jpg
[01-Sep-2009\n17:34:20] <mrayzenoss> there's probably something in your data\nsource where that's set... you'll have to either delete your rrd file\nor migrate them by hand to get the change
[01-Sep-2009 17:34:28] <mrayzenoss> to be reflected in the graph
\n[01-Sep-2009 17:36:33] <sergeymasushko> heh... I've set the same\nsettings like on the laloadint5... however you see hardcoded 'Aliases\nloadAverage5min | 100,/' on the datasource of laloadint5
[01-Sep-2009 22:05:40] <Egyptian[Home]1> damn .. and i just downloaded the 2.4.3 release yesterday!
[01-Sep-2009 22:39:55] <davetoo> hmm
[01-Sep-2009 23:59:19] <ironpaw> hai
[01-Sep-2009 23:59:27] <ironpaw> anyone kickin?
[01-Sep-2009 23:59:58] <ironpaw> or alive
[02-Sep-2009 00:00:03] <ironpaw> with a pulse, etc?
[02-Sep-2009 00:00:46] [disconnected at Wed Sep 2 00:00:46 2009]
[02-Sep-2009 00:00:46] [connected at Wed Sep 2 00:00:46 2009]
[02-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[02-Sep-2009 02:48:13] <rhett> yo
[02-Sep-2009 06:22:51] <Apocalipse> morning
[02-Sep-2009 06:23:20] <Apocalipse> people...i started to install Cisco zenpack...yesterday
[02-Sep-2009 06:23:56] <Apocalipse> but i closed my shell session...
[02-Sep-2009 06:28:45] <Apocalipse> i am back
[02-Sep-2009 06:36:18] <Apocalipse> hello?
[02-Sep-2009 06:41:04] <Apocalipse> ???
[02-Sep-2009 06:55:51] <Appiah> ?
\n[02-Sep-2009 07:16:57] <Apocalipse> hi people...i would like to\nknow how can i check if a zenpack install process still running
[02-Sep-2009 07:17:32] <ckrough> after closing your sesstion?
\n[02-Sep-2009 07:18:10] <ckrough> some variant of 'ps -ef | grep\n-i xxxx' where xxxx is the zenpack installer process, whatever that may\nbe called. Thats my best guess.
[02-Sep-2009 07:18:31]\n<ckrough> If its been a day it either failed or finished... you\ncould check your settings page and see if it's listed, or look in the\nlogs
[02-Sep-2009 07:18:59] <Apocalipse> its cisco zenpack...
[02-Sep-2009 07:26:00] <Apocalipse> what log can i check?
[02-Sep-2009 07:28:53] <ckrough> not entirely sure. I would start with $ZENHOME/logs/event.log though
\n[02-Sep-2009 07:29:21] <ckrough> maybe grep all of the zenoss log\nfiles for the name of the zenpack and see if there is an entry that\nshows it completed or failed
[02-Sep-2009 07:30:17] <Apocalipse> hunn
\n[02-Sep-2009 07:40:23] <tripitaka_> a basic zenpack question\nhere.. I've installed the pack ZenPacks.vaibhav.brocadeswitches.\nI have some brocade devices under /Network/FCP. How can I get the\nzenpack to apply to these devices?
[02-Sep-2009 07:51:39] <Apocalipse> not found any entry...
[02-Sep-2009 07:52:22] <Apocalipse> grep -lr "Cisco*" ./log
[02-Sep-2009 07:52:32] <Apocalipse> grep -lr "ZenPack*" ./log
[02-Sep-2009 07:52:40] <Apocalipse> and other options...
[02-Sep-2009 08:03:14] <ckrough> not sure what else to recommend
[02-Sep-2009 08:03:24] <ckrough> are you familiar with 'screen'?
[02-Sep-2009 08:22:39] ke4qqq_ is now known as ke4qqq
[02-Sep-2009 08:22:57] <Apocalipse> me?
[02-Sep-2009 08:24:45] <tripitaka_> Apocalipse: what does this show you: ps ax |grep -i cisco
[02-Sep-2009 08:26:23] <ckrough> Apocalipse: yeah, not that it would help you currently, but in the future.
[02-Sep-2009 08:27:34] <Apocalipse> 58848 p1 D+ 0:00.00 grep -i cisco
[02-Sep-2009 08:27:50] <tripitaka_> that suggests that the install process is not running
\n[02-Sep-2009 08:28:35] <Apocalipse> i was talking with matt...\nand he suggested me to delete all zenpack cisco... and reinstall it
[02-Sep-2009 08:28:46] <tripitaka_> sounds good to me
[02-Sep-2009 08:30:10] <Apocalipse> i am doing that now...
[02-Sep-2009 08:30:14] <Apocalipse> 6-7 hours...
\n[02-Sep-2009 08:31:28] <tripitaka_> if you want to run it\novernight without risking losing your session again, you could either\nuse screen as ckrough suggests, or run `nohup zenpack --install $file\n&`, which will detach and background the process
[02-Sep-2009 08:32:38] <Apocalipse> hunnnnn
[02-Sep-2009 08:32:48] <Apocalipse> good suggestion...
[02-Sep-2009 08:34:32] <Apocalipse> thx
[02-Sep-2009 08:34:39] <tripitaka_> actually, you can do this for your existing process
[02-Sep-2009 08:34:50] <tripitaka_> background it with ctrl-z
[02-Sep-2009 08:35:07] <tripitaka_> and/or use 'detach' to detach the process from the parent shell
[02-Sep-2009 08:37:34] <Apocalipse> yes...
[02-Sep-2009 08:38:17] <tripitaka_> or disown
[02-Sep-2009 08:38:56] <Apocalipse> i forgot that...
[02-Sep-2009 08:39:02] <tripitaka_> http://inglorion.net/software/detach/
[02-Sep-2009 08:39:47] <jb> screen
[02-Sep-2009 08:39:47] <Apocalipse> thx again
[02-Sep-2009 08:39:59] <tripitaka_> +1 for screen
[02-Sep-2009 08:40:13] <tripitaka_> it's missing on a lot of machines, though
[02-Sep-2009 08:47:18] <tripitaka_> how do I associate a zenpack with a device class?
[02-Sep-2009 08:47:26] <tripitaka_> I'm not even sure that question makese sense
[02-Sep-2009 08:47:37] <jb> browse to the device class, click the down arrow, and choose "Add to ZenPack"
[02-Sep-2009 08:48:00] <jb> is that what you mean?
[02-Sep-2009 08:49:00]\n<tripitaka_> I've downloaded a zenpack for brocade switches. I\nhave a device class called /Devices/Network/FibreChannel. I want\nwhatever magic is in the zenpack to be applied to this class.
[02-Sep-2009 08:49:09] <jb> ah.
[02-Sep-2009 08:49:28] <jb> ok, im guessing the zenpack just provides a performance template?
[02-Sep-2009 08:49:36] <jb> for graphs/thresholds/etc?
\n[02-Sep-2009 08:50:10] <jb> browse to the device class, choose\nthe "Templates" tab, click the down arrow and choose "Bindings"
[02-Sep-2009 08:50:15] <tripitaka_> files in zenpack: http://pastebin.com/m2dbb659b
[02-Sep-2009 08:50:28] <jb> in addition to what is already selected, select the Brocade template
\n[02-Sep-2009 08:51:18] <tripitaka_> ok, I've gone to the device\nclass, and clicked templates, then 'Bind Templates..', however I do not\nsee anything new in this list
[02-Sep-2009 08:51:24] <tripitaka_> I'm wondering how this list is populated
[02-Sep-2009 08:51:36] <jb> by templates that are installed by the zenpack
[02-Sep-2009 08:51:45] <tripitaka_> does each zenpack make itself available to specific device classes only?
[02-Sep-2009 08:51:46] <jb> the zenpack maybe installed the template on another device class
[02-Sep-2009 08:52:00] <jb> yes, you would need to copy the template to other classes if it is bound to a specific class
[02-Sep-2009 08:52:03] <tripitaka_> ok, how can I find out which class it has installed itself to?
[02-Sep-2009 08:52:27] <jb> the "This ZenPack Provides" pane
[02-Sep-2009 08:52:32] <jb> should tell you if it created a device class
[02-Sep-2009 08:52:47] <tripitaka_> it says 'No database objects are included in this ZenPack '
[02-Sep-2009 08:53:02] <jb> where did you download this from?
[02-Sep-2009 08:53:32] <tripitaka_> http://www.zenoss.com/community/projects/zenpacks/brocade-switches
[02-Sep-2009 08:55:09]\n<tripitaka_> Looking at another zenpack, JuniperPerformance, the\n'Zenpack Provides' pane gives me a list of datasources and MIBs, but I\ncannot see what device class it has mapped to
\n[02-Sep-2009 08:55:18] <jb> ERROR: zenpack command failed.\nReason: Unable to create object using the following attributes:
[02-Sep-2009 08:55:18] <jb> * id: /zport/dmd/Devices/Blade Switches/rrdTemplates/BrocadeSwitch
[02-Sep-2009 08:55:18] <jb> * module: Products.ZenModel.RRDTemplate
[02-Sep-2009 08:55:24] <jb> did you get this error when you installed it?
[02-Sep-2009 08:55:37] <tripitaka_> I'll install again and check
\n[02-Sep-2009 08:56:22] <tripitaka_> ERROR: zenpack command\nfailed. Reason: [Errno 20] Not a directory:\n'/usr/local/zenoss/zenoss/Products/ZenPacks.vaibhav.brocadeswitches-1.0-py2.4.egg/skins'
[02-Sep-2009 08:56:23] <jb> the author of that zenpack didn't really package it well..
[02-Sep-2009 08:56:57] <jb> do you have a /Devices/Blade Switches/ class?
[02-Sep-2009 08:57:04] <mrayzenoss> tripitaka_: is the ZenPack an .egg or a .zip?
[02-Sep-2009 08:57:12] <jb> its an egg..
[02-Sep-2009 08:57:27] <jb> well, the egg is inside of the zip
[02-Sep-2009 08:57:30] <jb> actually.
[02-Sep-2009 08:57:36] <jb> extract the egg from the zip
[02-Sep-2009 08:58:05] <tripitaka_> hmm, unzipped, getting a different error - I need to create a 'Blade Switches' class
[02-Sep-2009 08:58:09] <jb> yes
[02-Sep-2009 08:58:23] <jb> and then, after installing, your performance template will be bound there
[02-Sep-2009 08:59:29] <Apocalipse> sorry... blue screen again
[02-Sep-2009 09:00:03] <Apocalipse> windows 7 sucks
[02-Sep-2009 09:00:05] <Apocalipse> kkkkkkkkkkkkk
[02-Sep-2009 09:00:07]\n<tripitaka_> right then, much better. The 'Zenpack Provides' pane\nnow shows: http://www.pastebin.ca/1551512. Thing is, which line shows\nme what device class the zenpack is bound to?
[02-Sep-2009 09:00:24] <jb> its bound to /Device/Blade Switches
[02-Sep-2009 09:00:40] <tripitaka_> and I can tell that from the line '/Devices/Blade Switches/rrdTemplates/BrocadeSwitch'?
[02-Sep-2009 09:00:44] <jb> yes
[02-Sep-2009 09:01:02] <tripitaka_> ok, so everything before the 'rrdTemplates' should be taken as a class path
[02-Sep-2009 09:01:30] <jb> sure..
[02-Sep-2009 09:01:50] <tripitaka_> perfect.. copy templates..
[02-Sep-2009 09:03:21] <tripitaka_> bingo
[02-Sep-2009 09:03:29] <tripitaka_> thank you jb
[02-Sep-2009 09:03:32] <jb> np
[02-Sep-2009 09:04:39] <mrayzenoss> I'll look into cleaning up the packaging on that ZenPack
[02-Sep-2009 09:04:45] <mrayzenoss> get it out of the .zip, since that's confusing
[02-Sep-2009 09:05:56] <ckrough> Is there a way to see the device class structure a zenpack requires before running the install?
[02-Sep-2009 09:06:07] <jb> yes
[02-Sep-2009 09:06:08] <ckrough> so that you can create the necessary containers
[02-Sep-2009 09:06:12] <ckrough> do share
[02-Sep-2009 09:06:17] <jb> but keep in mind, that they will be deleted when the zenpack removed..
[02-Sep-2009 09:06:22] <jb> as well as any devices underneath them
[02-Sep-2009 09:07:44] <jb> you can use manage_addDeviceClass in your __init__.py
[02-Sep-2009 09:08:45] <jb> but, typically, as long as the parent class exists, the "Add To ZenPack" on a device class will work fine
[02-Sep-2009 09:08:50] <jb> not sure why it didn't in that case..
[02-Sep-2009 09:09:04] <jb> maybe the author forgot to add the class to the pack, and just added the performance template
[02-Sep-2009 09:18:22] <mrayzenoss> 10 people already signed up for the free training at Ohio LinuxFest
[02-Sep-2009 09:18:43] <markeriv> hmm
[02-Sep-2009 09:18:45] <markeriv> ohio you say
[02-Sep-2009 09:18:48] <markeriv> when?
[02-Sep-2009 09:19:00] <mrayzenoss> http://blog.zenoss.com/2009/08/21/zenoss-community-day-september-25th/
[02-Sep-2009 09:19:04] <mrayzenoss> the Friday before OLF
[02-Sep-2009 09:19:17] <markeriv> ack
[02-Sep-2009 09:19:20] <markeriv> sept 25
[02-Sep-2009 09:19:25] <markeriv> i am getting married on sept 26
[02-Sep-2009 09:19:27] <markeriv> thats a no go
[02-Sep-2009 09:19:30] <mrayzenoss> lol
[02-Sep-2009 09:19:39] <mrayzenoss> I'm sure your fiance will understand
[02-Sep-2009 09:19:41] <jb> haha
[02-Sep-2009 09:19:46] <jb> get your priorities straight!
[02-Sep-2009 09:20:02] <mrayzenoss> we'll have free training at LISA too: http://www.usenix.org/events/lisa09/
[02-Sep-2009 09:20:15] <jb> hm you know what matt
[02-Sep-2009 09:20:24] <jb> i need to talk to someone about sending me some updated training material
[02-Sep-2009 09:20:48] bzed_ is now known as bzed
[02-Sep-2009 09:21:59] <ckrough> November in Baltimore... what were they thinking.
[02-Sep-2009 09:22:56] <mrayzenoss> yeah, last year was San Diego
[02-Sep-2009 09:23:11] <mrayzenoss> Dallas the year before that. it moves around year to year
[02-Sep-2009 09:23:35] <cgibbons> they need to pick somewhere with good weather in november, like houston
[02-Sep-2009 09:23:57] <ckrough> or just park it in San Diego and call it a wrap
[02-Sep-2009 09:24:05] <markeriv> how about seattle
[02-Sep-2009 09:24:11] <ckrough> that would rock
[02-Sep-2009 09:24:22] <markeriv> yea its nice up here
[02-Sep-2009 09:24:31] <markeriv> could do it at the everett convention center
[02-Sep-2009 09:24:36] <cgibbons> if you like wearing a lot of sweaters
[02-Sep-2009 09:24:41] <markeriv> not really
[02-Sep-2009 09:24:45] <markeriv> its 70 degrees today
[02-Sep-2009 09:24:46] <markeriv>
[02-Sep-2009 09:24:49] <Apocalipse> matt do u have a free training schedule?
[02-Sep-2009 09:25:28] <mrayzenoss> well, Ohio is the second time we've done it
[02-Sep-2009 09:25:36] <ckrough> yesterday was our first <100 degree day in over 2 months in Texas.
[02-Sep-2009 09:25:42] <mrayzenoss> and we know we're going to do it at LISA
[02-Sep-2009 09:25:47] <mrayzenoss> and we'll do it at SCALE in LA
[02-Sep-2009 09:25:49] <cgibbons> 70 degrees = sweater weather (plus it ain't november yet, brr)
[02-Sep-2009 09:26:01] <jb> well then you better stay away from chicago
[02-Sep-2009 09:27:21] <markeriv> actually, it doesnt get very cold here mostly
[02-Sep-2009 09:27:29] <markeriv> it stays rather mediumish
[02-Sep-2009 09:27:36] <markeriv> we avoid all extremes
[02-Sep-2009 09:27:43] <markeriv> which is funny, cause seattle people are like that
[02-Sep-2009 09:28:01] <cgibbons> it was sure cold when I was there last march; granted cold is relative but brr.
[02-Sep-2009 09:28:03] <ckrough> Just dont bring up coffee?
[02-Sep-2009 09:28:12] <markeriv> hey!
[02-Sep-2009 09:28:18] <markeriv> coffee was created in seattle
[02-Sep-2009 09:28:23] <markeriv> fact of life
[02-Sep-2009 09:28:30] * ckrough notes the suddent extremism
[02-Sep-2009 09:28:38] <markeriv> :-P
[02-Sep-2009 09:28:47] <markeriv> just kidding
[02-Sep-2009 09:28:52] <markeriv> we are just more addicted to it
[02-Sep-2009 09:29:00] <markeriv> and have unique tastes cause our water is so good
[02-Sep-2009 09:29:27] <markeriv> water quality makes the coffee taste different no matter where you go
[02-Sep-2009 09:29:57] <rmatte> actually, god created coffee, and beer, on the 8th day
[02-Sep-2009 09:30:27] <markeriv> yes, he did... but he did it in seattle
[02-Sep-2009 09:31:13] <ckrough> the Africans may disagree with you there
[02-Sep-2009 09:31:39] <markeriv> pfft
[02-Sep-2009 09:31:59] <Apocalipse> do u have the website of LISA?
[02-Sep-2009 09:32:00] <markeriv> they invented the bean
[02-Sep-2009 09:32:14] <markeriv> we perfected the service of coffee
[02-Sep-2009 09:32:23] <ckrough> www.usenix.org
[02-Sep-2009 09:32:25] <markeriv> just kidding in all this... coffee topic is dead
[02-Sep-2009 09:32:31] <Apocalipse> thx...
[02-Sep-2009 09:32:51] <Apocalipse> do u know the expression "God is from Brazil" ?
[02-Sep-2009 09:33:19] <rmatte>
[02-Sep-2009 09:33:45] <jb> rmatte: were you able to find anything about that aggregate report question I had yesterday?
\n[02-Sep-2009 09:36:55] <tripitaka_> hmm, I just added a number of\ncollector plugins to a device and remodelled. I now have lots of new\ninfo under the Hardware tab, but all my Perf graphs have disappeared.\nHow should I start fixing this?
[02-Sep-2009 09:37:13] <mrayzenoss> all have disappeared?
\n[02-Sep-2009 09:37:59] <tripitaka_> it is a linux server on a HP\nblade, in /Devices/Server/Linux. I did have cpu load, I/O and memory\ngraphs. Now I have no graphs under Perf
[02-Sep-2009 09:39:00]\n<tripitaka_> I added the HPMonitor zenpack, then went to the\ndevice and added community.snmp.HP*, and then remodelled the device
[02-Sep-2009 09:45:30] <tripitaka_> got it, the Device template was unbound
[02-Sep-2009 09:48:16] <tripitaka_> odd, the Device template keeps unbinding itself.
\n[02-Sep-2009 09:58:41] <tripitaka_> procedure: install the\nHPMonitor zenpack. Go to a device and add all the community.snmp.HP*\ncollector plugins, then model the device. Every time I do this, the\nDevice template unbinds and I lose my perf graphs
[02-Sep-2009 10:05:37] <tripitaka_> to get the Device template to stay bound, I have to remove all the HP* collector plugins
[02-Sep-2009 10:13:19] <mrayzenoss> tripitaka_: have you restarted zopectl and zenhub since installing the ZenPack?
[02-Sep-2009 10:14:13] <tripitaka_> yep, I'll do it again
\n[02-Sep-2009 10:35:27] <tripitaka_> rebooted.. I have the Device\ntemplate bound, as well as all the cpq* templates, and I have added all\nthe HP* collector plugins. Now I model the device, then I go back to\ntemplates and I see that the Device template is missing
[02-Sep-2009 10:36:55] <mrayzenoss> and this is using the HP ProLiant Monitor 1.4? Anyone else using this?
[02-Sep-2009 10:41:51] <rmatte> nope
[02-Sep-2009 10:42:09] <rmatte> there are so many HP monitor packs these days I'm losing track of which is which
[02-Sep-2009 10:42:43] <mrayzenoss> yeah, I imagine the 'hp' tag will get a lot of use on the new platform
[02-Sep-2009 10:43:03] <rmatte> yeh
[02-Sep-2009 10:43:15] <rmatte> the new platform will allow for categorization of ZenPacks I hope?
[02-Sep-2009 10:43:34] <rmatte> like, for grouping all MIBs ZenPacks, for instance
[02-Sep-2009 10:43:37] <mrayzenoss> yup
[02-Sep-2009 10:43:42] <tripitaka_> it's ZenPacks.community.HPMon 1.4
[02-Sep-2009 10:43:52] <rmatte> I plan on developing more MIBs ZenPacks when I have time
\n[02-Sep-2009 10:49:13] <rmatte> tripitaka_: yeh, I haven't used\nthat one, so it's dropping the local bindings each time you model?
[02-Sep-2009 10:49:45] <tripitaka_> that's right
[02-Sep-2009 10:49:55] <rmatte> have you tried it via an organizer instead of binding it locally?
[02-Sep-2009 10:49:56] <tripitaka_> I've tried it on two devices, it keeps happening
\n[02-Sep-2009 10:50:45] <tripitaka_> it seems to be related to the\ncollector plugins - if I add the community.snmp.HP* plugins and model,\nit loses Devices
[02-Sep-2009 10:51:10] <rmatte> hmmm
[02-Sep-2009 10:51:35] <rmatte> have you tried making a local copy of Devices?
[02-Sep-2009 10:51:42] <rmatte> instead of binding it?
\n[02-Sep-2009 10:54:38] <tripitaka_> ok, local copy of Devices..\nenabling HP* plugins, model device.. and the Devices template is unbound
\n[02-Sep-2009 10:57:57] <tripitaka_> add the Device template\nagain, remove the HP* plugins, model device.. and the Devices template\nis still bound
[02-Sep-2009 11:11:00] <jb> hrm hey rmatte
[02-Sep-2009 11:41:47] <jb> hrm, i wish zenbackup backed up zenpacks as well
[02-Sep-2009 11:41:52] <jb> anybody here had to deal with this yet?
[02-Sep-2009 11:48:07] <jb> i wish there was a way to export all zenpacks to egg's
[02-Sep-2009 11:48:13] <jb> that you had installed
[02-Sep-2009 11:48:18] <jb> anybody know of a way to do this?
[02-Sep-2009 11:51:11] <mrayzenoss> there's a ticket for zenbackup to include zenpacks
[02-Sep-2009 11:51:18] <jb> that would be awesome
[02-Sep-2009 11:51:29] <jb> im trying to develop a disaster recovery plan for zenoss
[02-Sep-2009 11:51:38] <jb> and the zenpacks just make everything more difficult
[02-Sep-2009 11:54:22] <jb> or even a zenpack --export option
[02-Sep-2009 11:54:57] <jb> mrayzenoss: hrm, do you have that ticket number handy?
[02-Sep-2009 11:55:32] <mrayzenoss> http://dev.zenoss.org/trac/ticket/5180
[02-Sep-2009 11:55:35] <jb> thanks
[02-Sep-2009 11:56:45] <Apocalipse> zenpack --list
\n[02-Sep-2009 11:56:45] <Apocalipse> ZenPacks.Nova.Cisco.Mibs\n(/usr/local/zenoss/ZenPacks/ZenPacks.Nova.Cisco.Mibs-1.0-py2.4.egg)
\n[02-Sep-2009 11:56:45] <Apocalipse> ZenPacks.community.mib_utils\n(/usr/local/zenoss/ZenPacks/ZenPacks.community.mib_utils-1.07-py2.4.egg)
[02-Sep-2009 11:56:49] <Apocalipse> ALELUIA!!!!
[02-Sep-2009 11:57:11] <mrayzenoss> mib-utils 1.08 is out
[02-Sep-2009 11:57:25] <Apocalipse> okok
[02-Sep-2009 11:58:16] <jb> ugh
[02-Sep-2009 11:58:17] <jb> king crab?
[02-Sep-2009 11:58:18] <jb>
[02-Sep-2009 11:58:25] <jb> oh actually
[02-Sep-2009 11:58:28] <jb> thats not too far away
[02-Sep-2009 11:59:27] <cgibbons> hurm
[02-Sep-2009 11:59:56] <jb> coo, looks like the code is already in place
[02-Sep-2009 12:12:47] <jb> ok i need to figure out how to make an aggregate report/graph
[02-Sep-2009 12:12:49] <jb> hrmm
[02-Sep-2009 14:00:18] mrayzenoss1 is now known as mrayzenoss
[02-Sep-2009 14:26:16] <Cosmin> hello
[02-Sep-2009 14:26:19] <Cosmin> so i just did an upgrade
[02-Sep-2009 14:26:39] <Cosmin> actually i did an backup and unninstalled the previous version of zenoss
[02-Sep-2009 14:26:57] <Cosmin> and now ... how can i restore from the backup all the devices and performance data?
[02-Sep-2009 14:27:50] <mrayzenoss> http://www.zenoss.com/community/docs/installation-guide/2.4.2/ch08.html
[02-Sep-2009 14:28:48] <Cosmin> yes i looked at the guide
[02-Sep-2009 14:28:53] <Cosmin> i had 2.3.2 before
[02-Sep-2009 14:29:04] <Cosmin> and i was getting error upgrading to 2.4.5
[02-Sep-2009 14:29:14] <Cosmin> so i thought that a clean install should do it
[02-Sep-2009 14:29:19] <Cosmin> but now im stuck on importing the data
[02-Sep-2009 14:29:47] <mrayzenoss> you'll use zenrestore
[02-Sep-2009 14:30:02] <mrayzenoss> make sure you have any ZenPacks installed in your new install
[02-Sep-2009 14:30:09] <mrayzenoss> before migrating
[02-Sep-2009 14:30:30] <Cosmin> after i use zenrestore and i log back in to the protal it sees the old version
[02-Sep-2009 14:30:39] <Cosmin> and also im getting erros when i start up the application
[02-Sep-2009 14:30:54] <Cosmin> about zenjobs
[02-Sep-2009 14:31:42] <Cosmin> should i restore the old zenoss and try to upgrade it differently?
[02-Sep-2009 14:31:58] <mrayzenoss> which installers are you upgrading with?
[02-Sep-2009 14:32:03] <mrayzenoss> stack to stack?
[02-Sep-2009 14:32:10] <Cosmin> the bin
[02-Sep-2009 14:32:21] <Cosmin> zenoss-stack-2.4.5-linux.bin
\n[02-Sep-2009 14:32:43] <mrayzenoss> you should be able to restore\nto your backup, then just run the .bin and let the stack installer\nupgrade itself
[02-Sep-2009 14:32:49] <mrayzenoss> http://www.zenoss.com/community/docs/installation-guide/2.4.2/ch08s02.html
[02-Sep-2009 14:32:50] <Cosmin> maybe its not possible to upgrade directly from 2.3.2 to 2.4.5?
[02-Sep-2009 14:32:58] <mrayzenoss> lemme check
[02-Sep-2009 14:32:59] <Cosmin> ok
[02-Sep-2009 14:33:22] <mrayzenoss> http://www.zenoss.com/community/docs/installation-guide/2.4.2/ch08s01.html#d4e866
[02-Sep-2009 14:33:36] <mrayzenoss> 2.3.x->2.4.x should work
[02-Sep-2009 14:33:43] <Cosmin> ok let me try again
\n[02-Sep-2009 14:34:36] <mrayzenoss> yeah, release notes say 2.3.x\n-> 2.4.5\nhttp://www.zenoss.com/news/archive/zenossannouncement.2009-09-01.5195585629#d4e118
[02-Sep-2009 14:46:47] <Cosmin> ok
[02-Sep-2009 14:46:51] <Cosmin> this is what im getting
[02-Sep-2009 14:46:53] <Cosmin> Warning: Problem running post-install step. Installation may not complete
[02-Sep-2009 14:46:53] <Cosmin> correctly
[02-Sep-2009 14:46:53] <Cosmin> Error running su zenoss -c '"/usr/local/zenoss/zenoss/bin/zenmigrate" ' :
[02-Sep-2009 14:46:53] <Cosmin> INFO:zen.migrate:The following migrate steps were not intended to run with the
[02-Sep-2009 14:46:53] <Cosmin> currently installed version of the Zenoss code. The installed version is 2.4.5.
[02-Sep-2009 14:46:54] <Cosmin> You can override this warning with the --force option.
[02-Sep-2009 14:46:56] <Cosmin> DifferentiateTemplates (2.3.1)
[02-Sep-2009 14:46:58] <Cosmin> Recovering
[02-Sep-2009 14:47:00] <Cosmin> Press [Enter] to continue :
[02-Sep-2009 14:47:45] <Cosmin> any clues?
[02-Sep-2009 14:50:10] <mrayzenoss> anything else? I'm running it by QA
[02-Sep-2009 14:51:27] <Cosmin> thats what i got in when i try to upgrade it
[02-Sep-2009 14:51:37] <Cosmin> you want me to paste the whole thing in a pastebin?
[02-Sep-2009 14:52:32] <mrayzenoss> sure
[02-Sep-2009 14:53:26] <Cosmin> here you go
[02-Sep-2009 14:53:26] <Cosmin> http://pastebin.com/m6465dd3c
[02-Sep-2009 14:55:42] <Cosmin> let me know if there is anything else that you want me to include
[02-Sep-2009 15:00:24] <mrayzenoss> QA referred me to this: http://dev.zenoss.org/trac/ticket/5164
[02-Sep-2009 15:02:31] <Cosmin> ok i was missing net-snmp-utils let me try that again
[02-Sep-2009 15:02:47] <mrayzenoss> well, the zenmigrate --force is what's needed
[02-Sep-2009 15:02:55] <mrayzenoss> to fix the upgrade
[02-Sep-2009 15:03:01] <mrayzenoss> there was a bug in 2.3.1 apparently
[02-Sep-2009 15:04:34] <Cosmin> is there a fix for that bug?
[02-Sep-2009 15:04:41] <Cosmin> maybe i8 can patch it before i try to upgrade it?
[02-Sep-2009 15:05:01] <mrayzenoss> well the zenmigrate --force should clean up the template issue
[02-Sep-2009 15:05:41] <mrayzenoss> basically there was a migrate error, so we're forcing it
[02-Sep-2009 15:05:46] <mrayzenoss> rather than let it linger
[02-Sep-2009 15:12:31] <Cosmin> allright that seems like tis doing something now
\n[02-Sep-2009 15:12:42] <Cosmin> im up to INFO:zen.migrate:This\ncan take a VERY long time to execute(eg hours). Please do not\ninterrupt.
[02-Sep-2009 15:25:55] <Cosmin> allright that completed
[02-Sep-2009 15:28:10] <daMaestro> anyone know why snmp trap 1.3.6.1.6.3.1.1.5.3 is not mapping correctly in 2.4.5?
[02-Sep-2009 15:29:31] <daMaestro> err sorry, 2.4.2
[02-Sep-2009 15:43:31] <Cosmin> anyone seen this before?
[02-Sep-2009 15:43:33] <Cosmin> zenpack --remove=ZenPacks.zenoss.ZenJMX
[02-Sep-2009 15:43:33] <Cosmin> ERROR: zenpack command failed. Reason: 'Java'
[02-Sep-2009 16:07:31] <mrayzenoss> finally getting new ZenPacks out
[02-Sep-2009 16:07:44] <mrayzenoss> WMI Terminal Server: http://www.zenoss.com/community/projects/zenpacks/wmiperf_terminalserver
[02-Sep-2009 16:08:03] <mrayzenoss> WMI Exchange 2003: http://www.zenoss.com/community/projects/zenpacks/wmiperf_exchange2003
[02-Sep-2009 16:09:19] <Cosmin> awsome
[02-Sep-2009 16:13:21] <Cosmin> allright
[02-Sep-2009 16:13:24] <Cosmin> one little problem
[02-Sep-2009 16:13:34] <Cosmin> when i ack an alert
[02-Sep-2009 16:13:44] <Cosmin> after a while it cleares and then it comes back
[02-Sep-2009 16:14:24] <mrayzenoss> events that do not re-occur get cleared after 4 hours
[02-Sep-2009 16:14:32] <Cosmin> instead of them staing acknowledged they are keep comming back
[02-Sep-2009 16:14:54] <Cosmin> they are reocrurring durring a 5 minute period
[02-Sep-2009 16:15:03] <mrayzenoss> that's odd
[02-Sep-2009 16:15:23] <Cosmin> hm.. not sure where to look to see whats wrong
[02-Sep-2009 16:15:51] <mrayzenoss> is the count increasing?
[02-Sep-2009 16:16:07] <Cosmin> nope
[02-Sep-2009 16:16:08] <mrayzenoss> it's switching them from Acknowledged back to New?
[02-Sep-2009 16:16:10] <Cosmin> its only 1
[02-Sep-2009 16:16:18] <mrayzenoss> hmmm... so it thinks they're different
[02-Sep-2009 16:16:22] <Cosmin> im not sure if its switching them
[02-Sep-2009 16:16:34] <Cosmin> but it looks like that
[02-Sep-2009 16:16:44] <Cosmin> the old ones are not there anymore
[02-Sep-2009 16:18:09] <mrayzenoss> did you 'restart' or 'start' after your migration?
[02-Sep-2009 16:18:20] <mrayzenoss> there's a bug about 'restart' skipping an upgrade step
[02-Sep-2009 16:18:54] <Cosmin> yes
[02-Sep-2009 16:19:07] <Cosmin> i also rebooted the whole box
[02-Sep-2009 16:20:19] <mrayzenoss> what kind of events? snmp, ssh...?
[02-Sep-2009 16:20:32] <Cosmin> mm..
[02-Sep-2009 16:20:43] <Cosmin> they are failed checks for http check
[02-Sep-2009 16:20:48] <mrayzenoss> or does it not matter?
[02-Sep-2009 16:21:49] <mrayzenoss> more ZenPacks
\n[02-Sep-2009 16:21:50] <mrayzenoss> WMI Terminal Services:\nhttp://www.zenoss.com/community/projects/zenpacks/wmiterminalservices
[02-Sep-2009 16:22:05] <mrayzenoss> WMI Exchange 2003 Perf http://www.zenoss.com/community/projects/zenpacks/wmiexchange2003
[02-Sep-2009 16:22:09] <Cosmin> hmm
[02-Sep-2009 16:22:17] <Cosmin> im getting the same ones over and over again
[02-Sep-2009 16:22:30] <Cosmin> these are only new ones that showed up after the upgrade
[02-Sep-2009 16:22:38] <Cosmin> the old ones that i had in there they remained acknowledged
[02-Sep-2009 16:23:04] <Cosmin> so far i get some email checks failing and the http_check failing
[02-Sep-2009 16:29:25] <Cosmin> hmm any hints?
[02-Sep-2009 16:33:07] <mrayzenoss> Cosmin: are the Acknowledged events in Event History?
[02-Sep-2009 16:34:22] <mrayzenoss> they're disappearing right? Not becoming "Unacknowledged"?
[02-Sep-2009 16:34:26] <mrayzenoss> add a note to the log
[02-Sep-2009 16:34:49] <Cosmin> hmm..
[02-Sep-2009 16:34:50] <Cosmin> they are
[02-Sep-2009 16:34:57] <Cosmin> and there is a clear after
[02-Sep-2009 16:35:07] <Cosmin> not sure how it clears do
[02-Sep-2009 16:35:22] <mrayzenoss> the clear event moves it out of the event console
[02-Sep-2009 16:35:26] <mrayzenoss> and into history
[02-Sep-2009 16:35:32] <Cosmin> right
[02-Sep-2009 16:35:40] <Cosmin> so the check is checking for ssl
[02-Sep-2009 16:35:40] <mrayzenoss> what's clearing them?
[02-Sep-2009 16:36:39] <Cosmin> im not sure
[02-Sep-2009 16:36:42] <Cosmin> i dont see anything
[02-Sep-2009 16:36:53] <Cosmin> when i click on the details i see http_content_check
[02-Sep-2009 16:36:57] <Cosmin> and thats whats making them
[02-Sep-2009 16:37:04] <Cosmin> im not sure how they become clear
[02-Sep-2009 16:37:12] <Cosmin> because i checked the website and it doesnt have ssl enabled
[02-Sep-2009 16:37:18] <mrayzenoss> what's causing the event? Command failing?
[02-Sep-2009 16:37:23] <mrayzenoss> bad credentials?
[02-Sep-2009 16:38:25] <Cosmin> connection refused
[02-Sep-2009 16:38:34] <Cosmin> because there is nothing listening on ssl
[02-Sep-2009 16:39:39] <Cosmin> maybe change the cycle time for the command?
[02-Sep-2009 16:39:44] <Cosmin> its currently set to 300
[02-Sep-2009 16:40:02] <mrayzenoss> yeah, but what's clearing it?
[02-Sep-2009 16:40:17] <mrayzenoss> are other, non-command calls generating events?
[02-Sep-2009 16:40:53] <Cosmin> eventClass /Cmd/Fail
[02-Sep-2009 16:40:53] <Cosmin> eventKey http_content_check
[02-Sep-2009 16:40:53] <Cosmin> summary Clear
[02-Sep-2009 16:41:11] <Cosmin> manager localhost
[02-Sep-2009 16:41:12] <Cosmin> agent zencommand
[02-Sep-2009 16:42:08] <Cosmin> seems like they are generated my custom made templates
[02-Sep-2009 16:42:16] <Cosmin> using nagios plugins
[02-Sep-2009 16:44:05] <Cosmin> decreasing the time didn help
[02-Sep-2009 16:51:04] <Cosmin> Available Zenoss Version 2.3.2
[02-Sep-2009 16:51:17] <Cosmin> this is what i have by version by settings
[02-Sep-2009 16:57:47] <mrayzenoss> something tells me your upgrade did not go smoothly...
[02-Sep-2009 16:58:36] <rmatte> no, he probably doesn't have the checkboxes checked off
[02-Sep-2009 16:58:42] <rmatte> and he's looking at the last available version it saw
[02-Sep-2009 16:59:02] <rmatte> Cosmin: look down the page further and you'll see the actual Zenoss version
\n[02-Sep-2009 16:59:18] <mrayzenoss> don't remember who was\nlooking at it earlier, but I fixed the Brocade Switches page:\nhttp://www.zenoss.com/community/projects/zenpacks/brocade-switches
\n[02-Sep-2009 16:59:23] <rmatte> and if you put a checkmark in the\ncheckbox to periodically check for updated version, then click update,\nyou should see the latest
[02-Sep-2009 17:00:30] <rmatte> bah, this map thing is going to drive me nuts
[02-Sep-2009 17:00:35] daMaestro|isBack is now known as daMaestro
[02-Sep-2009 17:00:37] <Cosmin> i do see the actual version by zenoss
[02-Sep-2009 17:00:49] <Cosmin> but thats the available version even if i check for updates
[02-Sep-2009 17:01:19] <Cosmin> i still have a backup of the old version
[02-Sep-2009 17:01:35] <Cosmin> if there is an another path of doing this i am willing to try it
[02-Sep-2009 17:05:05]\n<mrayzenoss> Cosmin: I guess you could try something along these\nlines:\nhttp://www.zenoss.com/community/docs/installation-guide/2.4.2/ch09s01.html
[02-Sep-2009 17:05:26] <mrayzenoss> do a zenbackup, a clean install, then zenrestore
[02-Sep-2009 17:05:39] <mrayzenoss> make sure you get your ZenPacks in place before the zenrestore step
[02-Sep-2009 17:10:09] <Cosmin> should i use the same versions of zenpacks?
[02-Sep-2009 17:10:19] <Cosmin> or i can install the latest ones?
[02-Sep-2009 17:11:34] <mrayzenoss> which ones are you using? Latest is probably fine
[02-Sep-2009 17:12:51] <Cosmin> zenjmx and some of t
[02-Sep-2009 17:13:05] <Cosmin> are you gonna be online tomorrow?
[02-Sep-2009 17:13:15] <Cosmin> its kinda of late today... id like to go home
[02-Sep-2009 17:16:09] <mrayzenoss> I'm usually around
[02-Sep-2009 17:17:34] <Cosmin> allright
[02-Sep-2009 17:17:37] <Cosmin> thanks for your help
[02-Sep-2009 17:35:35] <ironpaw> hi
[02-Sep-2009 17:37:15] <markeriv> yo ironpaw
[02-Sep-2009 17:37:16] <markeriv> how goes it?
[02-Sep-2009 17:40:20] <ironpaw> yeh good mate yerself?
[02-Sep-2009 17:40:24] <ironpaw> almost friday
[02-Sep-2009 17:40:27] <ironpaw> Thu Sep 3 08:40:27 EST 2009
[02-Sep-2009 17:40:32] <ironpaw> almost but not quite
[02-Sep-2009 17:40:34] <ironpaw> hows things with you?
\n[02-Sep-2009 20:49:24] <gekitsuu> Are there any Zenoss people \nhere in contact with someone who's at the booth at the Red Hat Summit ?
[02-Sep-2009 22:55:26] davetoo1 is now known as DaveToo
[02-Sep-2009 23:34:15] <djjay> hi. I'm having problems with the linux ssh commands. df isn't logging any data
[02-Sep-2009 23:34:29] <djjay> hopefully you aren't all asleep
[03-Sep-2009 00:00:46] [disconnected at Thu Sep 3 00:00:46 2009]
[03-Sep-2009 00:00:46] [connected at Thu Sep 3 00:00:46 2009]
[03-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[03-Sep-2009 00:37:02] <ironpaw> o.O
[03-Sep-2009 01:26:44] [disconnected at Thu Sep 3 01:26:44 2009]
[05-Sep-2009 00:00:46] [connected at Sat Sep 5 00:00:46 2009]
[05-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[05-Sep-2009 08:51:15] <Apocalipse> hello people
[05-Sep-2009 09:09:44] <Apocalipse> all quiet
[05-Sep-2009 09:09:45] <Apocalipse>
[06-Sep-2009 00:00:46] [disconnected at Sun Sep 6 00:00:46 2009]
[06-Sep-2009 00:00:46] [connected at Sun Sep 6 00:00:46 2009]
[06-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
\n[06-Sep-2009 04:54:11] <InitX> Hi all , I'm a newbie to Zenoss\nand I had read the zenoss document. In the "Getting Started with\nZenoss" there is a google map display in Zenoss Dashboard how to let my\nzenoss display that google map?
[06-Sep-2009 05:58:42] <Simon_2> hello all
[06-Sep-2009 05:59:27] <Simon_2> !rules
[06-Sep-2009 05:59:49] <Simon_2> nobody here then
[06-Sep-2009 06:27:40] <Mavrick^^> hi guys
[06-Sep-2009 06:51:45] <Mavrick^^> can some onetell me where is mib folder?
[06-Sep-2009 06:52:10] <Mavrick^^> create a "Juniper" Mibs sub-folder
[06-Sep-2009 06:52:14] <Mavrick^^> where shuld i create it
\n[06-Sep-2009 06:59:43] <idl0r> hey guys, are there any plans for\nadding rules to _not_ use all the provided external libs?
[06-Sep-2009 07:02:43] <Mavrick^^> create a "Juniper" Mibs sub-folder
[06-Sep-2009 07:02:44] <Mavrick^^> where shuld i create it
[06-Sep-2009 07:15:12] <Mavrick^^> Ok i got it
[06-Sep-2009 14:31:03] <tpdd> is it possible to access a in zenoss directly through a link so it can be embeded in a different html page?
[06-Sep-2009 14:31:22] <tpdd> access a graph directly
[06-Sep-2009 14:38:25] <Appiah> I dont see why you wanna do that but sure why not?
[06-Sep-2009 14:38:47] <Appiah> you will have to keep the session info that you are logged in ...
\n[06-Sep-2009 14:40:33] <tpdd> Right now I have mrtg setup to show\nthe traffic on a couple of interfaces and Im just trying to replicate\nthat without giving logins to the 10 people that need to see it.
[06-Sep-2009 14:42:59] <Appiah> you could probly setup some scripts providing that info
[06-Sep-2009 14:43:59] <tpdd> like to copy the graphs out to where they can be viewed?
[06-Sep-2009 20:01:53] <ironpaw> hi
[06-Sep-2009 20:01:57] <ironpaw> how are we all today?
[06-Sep-2009 22:23:46] <ironpaw> o.O
[06-Sep-2009 22:23:50] <ironpaw> hai
[06-Sep-2009 22:24:02] <ironpaw> you should get an eggdrop bot in here
[06-Sep-2009 22:24:09] <ironpaw> it'd give me someone to talk to
[06-Sep-2009 22:35:22] <Diddi> (:
[06-Sep-2009 22:38:13] <ironpaw> #drupal have an eggdrop greetings and faqs
[07-Sep-2009 00:00:46] [disconnected at Mon Sep 7 00:00:46 2009]
[07-Sep-2009 00:00:46] [connected at Mon Sep 7 00:00:46 2009]
[07-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[07-Sep-2009 00:24:29] <ironpaw> hi
[07-Sep-2009 01:00:02] <ironpaw> is it possible to get DISK IO per filesystem?
[07-Sep-2009 06:32:16] quadro_ is now known as Quadro
[07-Sep-2009 09:54:00] <malbon> ironpaw: Which OS are you monitoring?
[07-Sep-2009 13:48:59] <djack> hellow world
[07-Sep-2009 13:49:16] <djack> somebody living here ?
[07-Sep-2009 14:03:21] <djack> nobody ?
[07-Sep-2009 14:57:44] <djack> wow netslplit
[07-Sep-2009 15:17:47] [disconnected at Mon Sep 7 15:17:47 2009]
[09-Sep-2009 00:00:46] [connected at Wed Sep 9 00:00:46 2009]
[09-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
[09-Sep-2009 06:54:03] <fuzz> anyone successfully using zenoss behind an apache vhost with ssl?
\n[09-Sep-2009 06:54:14] <fuzz> i've set up stuff like this:\nhttp://www.zenoss.com/community/docs/howtos/setup-zenoss-with-apache/
[09-Sep-2009 06:54:39] <fuzz> but the login <form> still has action=http://xxxx (not https or a relative url)
[09-Sep-2009 06:55:01] <fuzz> so it POSTs to http, which redirects to https again, and in that process it seems to loose the post data
\n[09-Sep-2009 06:55:41] <fuzz> it seems possible to change the url\nof the server to https://xxx for alert mails in zenactions.conf, but\nthis does not seem to affect the login form url
[09-Sep-2009 06:55:50] <fuzz> is there another config setting to change that url?
\n[09-Sep-2009 07:42:27] <Dieterbe> Hey guys. I have a custom\ntemplate, which has a zencommand datasource which executes a script. i\nused to get notifications (and events) when this script exited with\ncode >0. but since a while not anymore (i did upgrade from 2.4.1 to\n2.4.3). i'm not even getting events. i tried setting severity to\nwarning/error/critical, parser auto/nagios and disabled all my event\ntransforms
[09-Sep-2009 07:42:32] <Dieterbe> any idea?
[09-Sep-2009 07:44:09]\n<Dieterbe> oh and i replaced my normal command with execution of\na very simple script, which i tried both exit(1) and exit(2). and it\nlogs when it's invoked, so by looking at the log i see it's executed\nevery 30s (cycle time), but no events ..
[09-Sep-2009 07:56:55] <IoNu|z> hello
[09-Sep-2009 07:57:03] <Dieterbe> hmm just reading about the 2.4.5. I'll try upgrading..
[09-Sep-2009 07:57:08] <Dieterbe> hi IoNu|z
[09-Sep-2009 07:57:16] <IoNu|z> does anybody have issues with seeing custom OS products?
[09-Sep-2009 08:05:11] <skipzoid> hello everyone
[09-Sep-2009 08:06:16] <skipzoid> no body here
[09-Sep-2009 08:08:09] <venturaville> still pretty early ...it will wake up in another hour or so
\n[09-Sep-2009 08:13:47] <skipzoid> ah, my question was - what\ndistro would give the least resistence when trying to install zenoss on\nit? - we're slackware users here - but its fighting back at every turn,\nwmi failing - etc, also theres a python error that comes up with trying\nto add a device - i'm happy to download and install another distro -\nbut what one?
[09-Sep-2009 08:16:41] <skipzoid>\nwhat are people using here and how much of a fiddle was it to get\nzenoss installed on it
[09-Sep-2009 08:17:31] <venturaville> redhat and ubuntu seem to work pretty well
[09-Sep-2009 08:17:41] <venturaville> you are on your own with pretty much everything with slackware
[09-Sep-2009 08:18:12] <venturaville> trust me... I was using slackware back when I could run it off a single floppy
[09-Sep-2009 08:18:35] <skipzoid> well we're all pretty rounded here with slackware - but this zenoss is getting us beat,
\n[09-Sep-2009 08:19:12] <venturaville> your best bet if you are\nsticking to slackware is to use one of the packaged zenoss installs
\n[09-Sep-2009 08:19:14] <skipzoid> we've got several slackware\nboxes here and dotted around the company in different offices
[09-Sep-2009 08:19:50] <skipzoid> is a packed zenoss installed different from a core?
[09-Sep-2009 08:19:54] <skipzoid> which is compiled?
[09-Sep-2009 08:20:02] <venturaville> nope same thing, but all packaged with its dependencies
[09-Sep-2009 08:20:06] <venturaville> let me see if I can find a link
[09-Sep-2009 08:20:36] <venturaville> http://www.zenoss.com/download/links?creg=null
[09-Sep-2009 08:20:41] <venturaville> I believe what you want is one of the stack installers
[09-Sep-2009 08:20:56] <skipzoid> just having a look
[09-Sep-2009 08:21:32] <skipzoid> obviosuly there isnt a slackware one
[09-Sep-2009 08:21:43] <venturaville> fraid not
[09-Sep-2009 08:21:49] <venturaville> debian 5 might be worth a try
[09-Sep-2009 08:21:56] <skipzoid> just thinking the same
[09-Sep-2009 08:22:05] <skipzoid> its pretty close in its layout
[09-Sep-2009 08:23:06] <skipzoid> will have a look and come back later - thanks very much for your help
[09-Sep-2009 08:28:54] kevin7kal_ is now known as kevin7kal
[09-Sep-2009 09:03:14] <metalklesk> Hello everyone
[09-Sep-2009 09:22:23] <metalklesk> does anyone knows if you can monitor tandem server with zenoss ?
[09-Sep-2009 09:23:02] <Appiah> does it use snmp ?
[09-Sep-2009 09:24:02] <rmatte> metalklesk: It's a web managed tool? Basically a CMS?
[09-Sep-2009 09:24:46] <rmatte> metalklesk: what aspects of it are you hoping to monitor?
[09-Sep-2009 09:31:11] <metalklesk> i'm talking about OS
[09-Sep-2009 09:31:17] <metalklesk> HP Tandem
[09-Sep-2009 09:32:04]\n<metalklesk> we need to know if we can monitor aspects of SO like\nprocess or services .. obviusly some hardware monitoring too like cpu,\nmemory and filesystems
[09-Sep-2009 09:32:14] <metalklesk> SO == OS
[09-Sep-2009 09:32:28] <metalklesk> Operative System
[09-Sep-2009 09:32:30] <metalklesk>
[09-Sep-2009 09:33:00] <ckrough> are these older boxes?
[09-Sep-2009 09:33:22] <metalklesk> i have seen in zenoss that i can monitor any process i want by adding a "process"
[09-Sep-2009 09:33:31] <metalklesk> yes, it is an old box
\n[09-Sep-2009 09:34:14] <ckrough> it mostly depends on your OS. If\nyou're OS provides decent SNMP information and/or ssh or other remote\naccess, then you can monitor just about anything. It may require\nlearning and scripting on your end to get there though.
[09-Sep-2009 09:34:33] <ckrough> "your" OS that is
[09-Sep-2009 09:36:57] <rmatte> metalklesk: yeh, but is Tandem the actual OS?
[09-Sep-2009 09:37:06] <rmatte> or is it something running on top of like Linux or something?
[09-Sep-2009 09:37:56] <mrayzenoss> Tru64?
[09-Sep-2009 09:38:48] <chudler> Tandem OS/Guardian ?
[09-Sep-2009 09:39:06] <rmatte> ah, TOS apparently
\n[09-Sep-2009 09:39:25] <rmatte> as long as it can provide\nrelatively standard SNMP output then you can monitor basically anything
\n[09-Sep-2009 09:39:57] <rmatte> you basically want to enable\nSNMP, enter the community string in to Zenoss, model the device, and\nsee what you ge
[09-Sep-2009 09:39:59] <rmatte> get*
[09-Sep-2009 09:44:09] <metalklesk> yes, i guess i should try
[09-Sep-2009 09:44:27] <rmatte> yup
[09-Sep-2009 09:45:40] <metalklesk> right now BMC Patrol cannot monitor TOS
[09-Sep-2009 09:46:53] <rmatte> Zenoss is pretty flexible, but some dev work might be required (custom scripts)
[09-Sep-2009 09:47:14] <rmatte> depending on what you can get TOS to spit out
[09-Sep-2009 09:49:43] <metalklesk> ok
\n[09-Sep-2009 09:52:49] <metalklesk> we were seeing\nhttp://www.exchangemonitoring.com.uk/Solutions/HPNonStopManagement/tabid/108/Default.aspx
[09-Sep-2009 09:53:15] <rmatte> I can't even hit that page
[09-Sep-2009 09:53:20] <rmatte> address no found
[09-Sep-2009 09:53:23] <rmatte> not*
[09-Sep-2009 09:53:29] <metalklesk> but i like zenoss and i want to try it before looking other solutions
[09-Sep-2009 09:53:54] <rmatte> I've worked with a few different monitoring tools and Zenoss is definitely my favourite
[09-Sep-2009 09:54:18] <metalklesk> i have worked with patrol an portal, some with nagios too
[09-Sep-2009 09:54:29] <metalklesk> and zenoss is my favorite too
[09-Sep-2009 09:54:44] <rmatte> well, Zenoss certainly blows Nagios out of the water
[09-Sep-2009 09:54:48] <rmatte> hehe
[09-Sep-2009 09:55:00] <metalklesk> http://www.exchangemonitoring.co.uk/Solutions/HPNonStopManagement/tabid/108/Default.aspx
[09-Sep-2009 09:55:08] <rmatte> ah, co.uk hehe
[09-Sep-2009 09:55:13] <metalklesk> yup
[09-Sep-2009 09:55:16] <metalklesk> my mistake
[09-Sep-2009 09:56:32] <rmatte> the reflex stuff is nice
[09-Sep-2009 09:57:06] <rmatte> but realistically, it's generally better to have an actual human being resolve issues
[09-Sep-2009 09:57:31] <metalklesk> why?
[09-Sep-2009 09:57:48] <rmatte> well, for simpler stuff, like restarting processes, I suppose it would be fine
\n[09-Sep-2009 09:58:58] <rmatte> because automatically fixing\nsomething can end up being just a band-aid, it's nice to have someone\nactually go in and investigate as to what caused the issue in the first\nplace when it happens, rather than just restart a process automatically\nand forget about it
[09-Sep-2009 09:59:01] <rmatte> that's my view anyways
[09-Sep-2009 09:59:38] <rmatte> I suppose from a response time aspect, automation does have it's advantages
[09-Sep-2009 09:59:45] <metalklesk> ahh yup
[09-Sep-2009 09:59:49] <rmatte> with dev work it's possible to automate stuff with Zenoss using custom scripts
[09-Sep-2009 10:04:21]\n<chudler> in my experience, automatic "healing", response,\nreflex, etc are more marketing tools than anything else. apealing to\nPHB's and sales-drones
[09-Sep-2009 10:04:30] <chudler> ok, I'll get off my high horse :-)
[09-Sep-2009 10:05:07] <rmatte> chudler: it's true though, it's ineffective for anything other that very simple issues
[09-Sep-2009 10:06:16]\n<mrayzenoss> someone told me they were using a zencommand to ssh\nto the box in question, and scrape every log and the status of the\nmachine when a certain event came in. Not autoremediation, but a good\nidea to snapshot a box in trouble
[09-Sep-2009 10:07:25] <rmatte> mrayzenoss: yeh, that's basically auto-reconnaissance
[09-Sep-2009 10:07:33] <mrayzenoss> yeah
[09-Sep-2009 10:07:34] <rmatte> which is actually a really nice idea
[09-Sep-2009 10:07:43] <rmatte> but auto break/fix is not such a great idea all of the time
[09-Sep-2009 10:07:47] <rmatte> hehe
[09-Sep-2009 10:07:58] <mrayzenoss> not unless you really nail down the error and the work-around is well known
[09-Sep-2009 10:08:07] <rmatte> yeh, exactly
[09-Sep-2009 10:08:34] <metalklesk> i agree
[09-Sep-2009 10:08:49] <rmatte> mrayzenoss: when is the 2.5 beta dropping?
[09-Sep-2009 10:09:27] <mrayzenoss> working on the beta today
[09-Sep-2009 10:09:37] <rmatte> so release is expected for tomorrow?
[09-Sep-2009 10:09:37] <mrayzenoss> announcements will be on the usual sites
[09-Sep-2009 10:09:52] <rmatte> cool
[09-Sep-2009 10:09:54] <rmatte> afk for a few
[09-Sep-2009 10:09:58] <mrayzenoss> yeah, today or tomorrow
[09-Sep-2009 10:10:04] <mrayzenoss> probably
[09-Sep-2009 10:10:08]\n<MajereDB8> hi, i haven't dropped by the dev channel in a while,\nbut i've been working on a project using the zenpop3 module and am\nhaving some issues getting it to work with openssl. i've checked TRAC\nand the boards but haven't found a solution. might anyone have a\nminute or so?
[09-Sep-2009 10:10:56] <metalklesk> one question, wich ubuntu version you can monitor with the ubuntu zenpack ?
[09-Sep-2009 10:11:31] <metalklesk> i'm trying 9.04 but zenoss cannot get the installed software list
[09-Sep-2009 10:11:51] <metalklesk> the other things like cpu and memory yes
[09-Sep-2009 10:12:24] <IoNu|z> does anybody have this issue with os tab?
[09-Sep-2009 10:12:28] <metalklesk> NOOO, nevermind, it is working
[09-Sep-2009 10:12:30] <IoNu|z> like the OS versions dont show up?
[09-Sep-2009 10:12:55] <IoNu|z> if i add another os version or hardware
[09-Sep-2009 10:13:00] <metalklesk> i have re model it and now is showing the installed software, its cool !
[09-Sep-2009 10:13:01] <IoNu|z> it doesnt show up in the fields
[09-Sep-2009 10:13:18] <IoNu|z> but it does show up by /products
[09-Sep-2009 10:13:29] <IoNu|z> is there a way to forcely poplate it?
[09-Sep-2009 10:13:47] <IoNu|z> http://forums.zenoss.com/viewtopic.php?t=10982
[09-Sep-2009 10:18:00] <IoNu|z> anyone any clues?
[09-Sep-2009 10:23:31] <Brad_K> Sorry, i'm still just a white belt.
[09-Sep-2009 10:27:25] <metalklesk> i don't understand your problem
[09-Sep-2009 10:27:27] <metalklesk>
\n[09-Sep-2009 10:35:50] <rmatte> metalklesk: the software list\njust depends on dpkg and aptitude, do I'm surprised it's not able to\npopulate it
[09-Sep-2009 10:36:18] <metalklesk> no no, it is working perfectly
[09-Sep-2009 10:36:19] <rmatte> IoNu|z: what version of Zenoss are you using?
[09-Sep-2009 10:36:41] <metalklesk> thanks rmatte
[09-Sep-2009 10:37:02] <metalklesk> zenoss is receiving the installed software list perfectly
[09-Sep-2009 10:37:10] <rmatte> cool
[09-Sep-2009 10:37:48] <metalklesk> the only thing now is that zenmodeler failed when i modeled the device
[09-Sep-2009 10:38:11] <rmatte> failed on what?
[09-Sep-2009 10:38:21] <metalklesk> Zenmodeler - /ÂCmd/ÂFail - User timeout caused connection failure
[09-Sep-2009 10:38:41] <metalklesk> i don't know what this means
[09-Sep-2009 10:38:45] <rmatte> that means that your SSH connection timed out somehow I believe
[09-Sep-2009 10:39:04] <rmatte> (You're modelling partially via SSH I assume?)
[09-Sep-2009 10:39:08] <metalklesk> i guess so
[09-Sep-2009 10:39:09] <metalklesk> yes
[09-Sep-2009 10:39:18] <metalklesk> using Ubuntu zenpack
[09-Sep-2009 10:39:23] <rmatte> make sure the username and password that are set in zProperties are correct
[09-Sep-2009 10:39:32] <metalklesk> i'm just trying that zenpack
[09-Sep-2009 10:39:39] <metalklesk> yes, credentials are ok
\n[09-Sep-2009 10:39:45] <rmatte> IoNu|z:\nhttp://dev.zenoss.org/trac/ticket/5473 <--- This is probably why (I\nassume you're using 2.4.5)
[09-Sep-2009 10:40:12] <metalklesk> in fact i can see the installed software, cpu, memory, filesystems, etc
[09-Sep-2009 10:40:23] <metalklesk> i don't see problems
\n[09-Sep-2009 10:40:25] <rmatte> IoNu|z: Also, when you add\nanother OS version or something it doesn't go to where you requested to\nadd it, it gets added to /Unknown I believe
[09-Sep-2009 10:40:40] <rmatte> It's a bug
[09-Sep-2009 10:41:09] <rmatte> metalklesk: then it's probably not an issue
[09-Sep-2009 10:41:21] <metalklesk> yes i think it is not
[09-Sep-2009 10:41:49] <rmatte> I think IoNu|z dozed off lol
[09-Sep-2009 10:42:15] dB8r is now known as MajereDB8
[09-Sep-2009 10:43:13] <MajereDB8> question, do you need python-twisted installed to get pyopenssl to work?
\n[09-Sep-2009 10:43:45] <metalklesk> a newbie question: in OS tab\n-> Interfaces, what does mean O A M Lock \n??
[09-Sep-2009 10:44:16] <metalklesk> only M has the green circle, O and A has a red one
[09-Sep-2009 10:46:27] <rmatte> metalklesk: O = Operational Status, A = Administrative Status, M = Monitored (yes/no)
\n[09-Sep-2009 10:46:50] <rmatte> metalklesk: in King Crab the O\nand the A are going to be replaced for a single "Status" indicator
\n[09-Sep-2009 10:47:34] <rmatte> metalklesk: also, the statuses\nonly get updated when you model a device right now, but in King Crab\nthey will reflect the realtime status of each interface
[09-Sep-2009 10:48:29] <metalklesk> ok, thanks
[09-Sep-2009 10:49:08] <rmatte> np
[09-Sep-2009 10:49:55] <rmatte> MajereDB8: no idea
[09-Sep-2009 10:50:03]\n<metalklesk> if you click in one interface (os tab ->\ninterfaces) like eth0, can you see some graphics ? i mean with some\nmetrics
[09-Sep-2009 10:50:27] <rmatte> metalklesk: you should see graphs indicating traffic, packets, and errors
[09-Sep-2009 10:50:56] <metalklesk> yes but in my case they have no data
[09-Sep-2009 10:51:05] <metalklesk> it say: curr nan
[09-Sep-2009 10:51:13] <rmatte> how long ago did you add the device?
[09-Sep-2009 10:51:24] <metalklesk> ten minutes ?
[09-Sep-2009 10:51:30] <rmatte> you need to wait at least 15
[09-Sep-2009 10:51:38] <rmatte> it takes 3 polling cycles for a graph to populate in Zenoss
[09-Sep-2009 10:51:46] <rmatte> each polling cycle is 5 minutes long
[09-Sep-2009 10:51:58] <metalklesk> ahh ok
\n[09-Sep-2009 10:52:37] <metalklesk> is tha polling flexible ? i\nmean, can i change th 5 minutes to lets say 10 minutes ? or 2 ?
[09-Sep-2009 10:53:29] <rmatte> metalklesk: not without actually editing Zenoss source code
[09-Sep-2009 10:53:38] <metalklesk> ok
[09-Sep-2009 10:54:11] <rmatte> generally you either want 5 minutes or 1 minute, but 1 minute is only for extreme cases
[09-Sep-2009 10:54:17] <ckrough> rmatte: thats configuratable in the gui
[09-Sep-2009 10:54:30] <rmatte> ckrough: I was under the impression that it wasn't
[09-Sep-2009 10:54:45] <rmatte> ckrough: I've always heard that code needs to be modified somewhere for that
\n[09-Sep-2009 10:54:53] <ckrough> Collectors > mycollectorname\n> edit > SNMP Performance Cycle Interval (secs) and Status Cycle\nInterval (secs)
[09-Sep-2009 10:55:51] <ckrough>\nmetalklesk: I wouldnt go higher resolution than 5min until you have an\nidea of the server load. There's a whole lot going on in the background
[09-Sep-2009 10:56:04] <metalklesk> oohh yeah, i can see the option
[09-Sep-2009 10:56:09] <metalklesk> yes, i'm agree
[09-Sep-2009 10:56:13] <ckrough> it may have to rebuild your RRDs if you change it
[09-Sep-2009 10:56:18] <ckrough> since they are built to a cycle
\n[09-Sep-2009 10:56:21] <rmatte> ckrough: yeh, but the RRD graphs\nin Zenoss are coded for 5 minutes, so there's something else that you\nneed to change for the graphinh
[09-Sep-2009 10:56:23] <metalklesk> i just wanted to know if the option is there
[09-Sep-2009 10:56:23] <rmatte> graphing*
[09-Sep-2009 10:56:31] <ckrough> "Default RRD Create Command"
[09-Sep-2009 10:56:34] <ckrough> at the bottom
[09-Sep-2009 10:57:08] <rmatte> ah
[09-Sep-2009 10:57:17] <rmatte> I don't remember seeing that before
[09-Sep-2009 10:58:18] <rmatte> cool
[09-Sep-2009 10:58:55] <metalklesk> when you add a new process, in id you put a regex right ?
[09-Sep-2009 10:59:10] <metalklesk> like mysql* ?
[09-Sep-2009 11:00:16] <IoNu|z> rmatte im still here
[09-Sep-2009 11:00:27] <IoNu|z> and yes i think i am hitting that bug
[09-Sep-2009 11:00:37] <IoNu|z> and im using 2.4.5
[09-Sep-2009 11:01:21] <metalklesk> answer to myself, yes, it can be used a regex
[09-Sep-2009 11:01:45] <rmatte> IoNu|z: yeh, no idea when that's going to get fixed, it's driving me nuts too
[09-Sep-2009 11:02:19] <rmatte> metalklesk: just put the process name
[09-Sep-2009 11:02:26] <rmatte> metalklesk: no regex formatting required
[09-Sep-2009 11:02:26] <metalklesk> yup
[09-Sep-2009 11:02:32] <metalklesk> yup
[09-Sep-2009 11:02:37] <rmatte> metalklesk: if the process is "mysqld.bin" then name it that
[09-Sep-2009 11:02:55] <rmatte> and then remodel the device and it'll pick it up along with the command options
[09-Sep-2009 11:05:34]\n<rmatte> does anyone know if the device availability value is\nstored somewhere in zope, or if it's calculated on the fly?
[09-Sep-2009 11:05:51]\n<rmatte> I want to code my own device availability report script,\nthat I can run to pull the device availability for each device
[09-Sep-2009 11:10:37] <metalklesk> why when y model a device (Ubuntu) i can't see ip services ?
[09-Sep-2009 11:10:47] <metalklesk> y == I
[09-Sep-2009 11:10:51] <metalklesk> my mistake
[09-Sep-2009 11:11:01] <rmatte> metalklesk: Have you enabled certain ones?
[09-Sep-2009 11:11:14] <rmatte> like, Click on Services on the left hand meny
[09-Sep-2009 11:11:16] <rmatte> menu*
[09-Sep-2009 11:12:09] <rmatte> annnnnd, 3 of my Zenoss instances that I've tried are frozen...
[09-Sep-2009 11:12:15] <rmatte> waiting on them to give you the other steps
[09-Sep-2009 11:12:43] <rmatte> then click IpService
[09-Sep-2009 11:12:49] <rmatte> Then click privileged
[09-Sep-2009 11:13:08] <rmatte> then find the service that you want to enable
[09-Sep-2009 11:13:11] <rmatte> click on it
[09-Sep-2009 11:13:18] <rmatte> go to the edit tab
[09-Sep-2009 11:13:22] <rmatte> change Monitor from False to True
[09-Sep-2009 11:13:25] <rmatte> then remodel the device
\n[09-Sep-2009 11:13:43] <rmatte> also, I've noticed that there's a\npotential bug with having Zenoss automatically pick up on IpServices\nduring modeling
[09-Sep-2009 11:13:52] <rmatte> it seems to only pick up on the first service, and ignores the rest
[09-Sep-2009 11:14:01] <rmatte> so you may have to add them in manually for the time being
[09-Sep-2009 11:14:25] <metalklesk> ok, thanks
\n[09-Sep-2009 11:14:57] <rmatte> If you notice the same behaviour\nas I've described let me know so that I can file a bug report
[09-Sep-2009 11:15:10] <metalklesk> ok, i will try now
[09-Sep-2009 11:15:13] <rmatte> thanks
[09-Sep-2009 11:18:17] <metalklesk> ssh can be considered as ip service in linux ?
[09-Sep-2009 11:18:24] <rmatte> correct
[09-Sep-2009 11:18:27] <rmatte> if you want
[09-Sep-2009 11:18:36] <rmatte> IpService just means it's monitoring the port
[09-Sep-2009 11:18:45] <rmatte> you could also monitor the Process as well
[09-Sep-2009 11:19:04] <rmatte> I find that for SSH monitoring the port is sufficient
[09-Sep-2009 11:19:34] <rmatte> same with stuff like ftp
[09-Sep-2009 11:19:41] <rmatte> or rdp
[09-Sep-2009 11:19:43] <metalklesk> ahh ok, great
[09-Sep-2009 11:22:49] <metalklesk> let's say i want so monitor sshd as ip service
[09-Sep-2009 11:23:30] <metalklesk> Classes -> Services -> add new ip service -> sshd
[09-Sep-2009 11:23:39] <metalklesk> then click on sshd -> edit
[09-Sep-2009 11:23:53] <metalklesk> i write port 22 and change monitored from false to true
[09-Sep-2009 11:24:01] <metalklesk> is that ok ?
[09-Sep-2009 11:26:06] <metalklesk> then i go to my device (Ubuntu) -> OS tab -> ip service -> add ip service
[09-Sep-2009 11:26:26] <metalklesk> but sshd ip service isn't there
[09-Sep-2009 11:26:37] <metalklesk> not the one i created before
[09-Sep-2009 11:27:58] <metalklesk> but it appears another one, i guess it's a default one
[09-Sep-2009 11:28:12] <metalklesk> i can use that !
[09-Sep-2009 11:37:56] <t3los> Is anyone available to answer a few questions?
[09-Sep-2009 11:38:04] <jb> just ask the questions..
[09-Sep-2009 11:38:06] <jb> don
[09-Sep-2009 11:38:09] <jb> don't ask to ask.
[09-Sep-2009 11:38:15] <t3los> Lol.
[09-Sep-2009 11:38:29] <MajereDB8> import sys
[09-Sep-2009 11:38:32] <MajereDB8> whoops sorry
\n[09-Sep-2009 11:38:45] <t3los> Alright, then. So I have this one\nsite, with less plugins loaded, and the same versions of zenpacks
\n[09-Sep-2009 11:38:49] <metalklesk> i have added to monitoring my\nnotebook with mandriva by SNMP. I have added the gnome-terminal process\n(that i have created before) and the status is green even if i don't\nhave that process running, why is that ?
[09-Sep-2009 11:38:58] <t3los> But for some reason, I get much less results here at my personal site
\n[09-Sep-2009 11:39:24] <t3los> It's kind of aggrivating because\nI've gone out of my way to make sure that things are even identical if\nnot more thorough over on my end.
[09-Sep-2009 11:39:51] <t3los> Less traffic, less machines, more attention to detail, and less results.
[09-Sep-2009 11:39:52] <t3los> ;p
\n[09-Sep-2009 11:40:47] <t3los> I read the entire documentation. I\nknow how to do a lot inside of Zenoss, but I can't find out what is\ncausing this problem.
[09-Sep-2009 11:41:09] <jb> what do you mean?
[09-Sep-2009 11:41:13] <jb> what is "less detail" ?
[09-Sep-2009 11:41:36] <t3los> I mean, as far as Status, OS, Hardware, and Performance Graphs
[09-Sep-2009 11:41:40] <t3los> I have less information
[09-Sep-2009 11:41:51] <jb> sounds like you have SNMP problems
[09-Sep-2009 11:42:00] <jb> can you walk devices from the new zenoss host?
[09-Sep-2009 11:42:19] <t3los> Well, sec
[09-Sep-2009 11:42:37] <t3los> Actually, no
[09-Sep-2009 11:42:38] <t3los> Lol.
[09-Sep-2009 11:42:46] <t3los> oh wait
[09-Sep-2009 11:43:01] <t3los> yes
[09-Sep-2009 11:43:02] <t3los> Sorry
[09-Sep-2009 11:43:41] <t3los> Yes, I can see the devices I am trying to collect for.
[09-Sep-2009 11:43:58] <jb> see them how?
[09-Sep-2009 11:45:06] <t3los> ==== hostName ====
[09-Sep-2009 11:45:06] <t3los> snmpwalk -v1 -cccissnmp 192.168.68.11 system
[09-Sep-2009 11:45:06]\n<t3los> SNMPv2-MIB::sysDescr.0 = STRING: Hardware: x86 Family 15\nModel 4 Stepping 3 AT/AT COMPATIBLE - Software: Windows Version 5.2\n(Build 3790 Multiprocessor Free)
[09-Sep-2009 11:45:06] <t3los> SNMPv2-MIB::sysObjectID.0 = OID: SNMPv2-SMI::enterprises.311.1.1.3.1.3
[09-Sep-2009 11:45:06] <t3los> DISMAN-EVENT-MIB::sysUpTimeInstance = Timeticks: (224371) 0:37:23.71
[09-Sep-2009 11:45:08] <t3los> SNMPv2-MIB::sysContact.0 = STRING: XXXXXXXXX
[09-Sep-2009 11:45:10] <t3los> SNMPv2-MIB::sysName.0 = STRING: XXXXXXX
[09-Sep-2009 11:45:12] <t3los> SNMPv2-MIB::sysLocation.0 = STRING: Server Room
[09-Sep-2009 11:45:14] <t3los> SNMPv2-MIB::sysServices.0 = INTEGER: 79
[09-Sep-2009 11:45:22] <t3los> Ignore the X's and hostName
[09-Sep-2009 11:45:30] <rmatte> metalklesk: no, ssh is already there by default
[09-Sep-2009 11:45:30] <t3los> Filtered text.
[09-Sep-2009 11:45:41] <rmatte> metalklesk: under the privileged organizer
[09-Sep-2009 11:45:50] <rmatte> metalklesk: so just find it in there and enable monitoring for it
\n[09-Sep-2009 11:46:16] <rmatte> metalklesk: Zenoss has pretty\nmuch every IP Service you can think of already there, you just need to\nenable monitoring for them
[09-Sep-2009 11:46:50] <metalklesk> what do you mean with privileged organizer ?
[09-Sep-2009 11:48:00]\n<t3los> Seriously, though. Has anyone heard of that kind of\nthing? More information at one site than the other even though you've\nmade sure to put more effort into the one?
[09-Sep-2009 11:54:44] <rmatte> metalklesk: click on Services on the left
[09-Sep-2009 11:54:58] <metalklesk> done
[09-Sep-2009 11:55:01] <rmatte> metalklesk: then look at the top, you'll see "Sub-Folders"
[09-Sep-2009 11:55:08] <rmatte> there's a subfolder called IpService
[09-Sep-2009 11:55:10] <rmatte> click on it
[09-Sep-2009 11:55:11] <metalklesk> yes
[09-Sep-2009 11:55:19] <rmatte> Then you'll see more Sub-Folders
[09-Sep-2009 11:55:23] <metalklesk> done
[09-Sep-2009 11:55:25] <rmatte> click on the privileged sub-folder
[09-Sep-2009 11:55:51] <rmatte> and voila
[09-Sep-2009 11:56:04] <rmatte> 684 of the most common ports
[09-Sep-2009 11:56:14] <rmatte> then you just need to search for what you want and enable monitoring on it
[09-Sep-2009 11:56:47] <rmatte> I generally just sort by port then scroll until I find what I want
[09-Sep-2009 11:57:29] <metalklesk> perfect
[09-Sep-2009 11:57:33] <metalklesk> many thanks
[09-Sep-2009 11:57:35] <rmatte> np
[09-Sep-2009 12:00:07] <metalklesk> gnome-terminal on mandriva is now showing a red status, that is ok
[09-Sep-2009 12:01:00]\n<metalklesk> (i created a process in zenoss for gnome-terminal,\nit must show a red status because there is no process of gnome-terminal\nin my mandriva)
[09-Sep-2009 12:04:15] <metalklesk> can i mix snmp AND ssh mointoring for one device ?
[09-Sep-2009 12:05:50] <mrayzenoss> yes
[09-Sep-2009 12:06:53]\n<metalklesk> and what if i monitor Ubuntu with snmp and Ubuntu\nzenpack ? both methods are going to catch the same parameters
[09-Sep-2009 12:07:29] <mrayzenoss> well, the Ubuntu ZenPack is monitored under /Servers/SSH/Linux/Ubuntu
[09-Sep-2009 12:07:43] <mrayzenoss> as opposed to /Servers/Linux
[09-Sep-2009 12:08:15] <mrayzenoss> you can configure them to conflict, not sure what happens
[09-Sep-2009 12:08:24] <mrayzenoss> but they don't conflict out of the box
[09-Sep-2009 12:09:11] <rmatte> yeh, you can mix and match monitoring components, but you have to be careful hehe
[09-Sep-2009 12:11:33] <rmatte> quite honestly, pure SNMP monitoring is fairly adequate for Linux servers
[09-Sep-2009 12:11:49] <rmatte> in most cases anyways
[09-Sep-2009 12:19:26] <metalklesk> ok
[09-Sep-2009 12:19:35] <Brad_K> mrayzenoss: hi! are we having fun yet? ;-)
[09-Sep-2009 12:22:35] <rmatte> he's always having fun
[09-Sep-2009 12:24:39] <mrayzenoss> yeah, every day is a picnic
[09-Sep-2009 12:25:10] <ckrough> dont you know? Community managers dont have bad days.
[09-Sep-2009 12:25:42] * rmatte whispers "That's because of the bottles of rum that they keep in their bottom drawers."
[09-Sep-2009 12:26:34] * mrayzenoss keeps it in the top drawer
[09-Sep-2009 12:27:09] <rmatte> ah, for convenience no doubt
\n[09-Sep-2009 12:30:10] <ckrough> you could rig up a zenpack that\nwatches output from a scale and alert when the bottle gets low
[09-Sep-2009 12:30:36] <ckrough> we could event graph your drinking rate
[09-Sep-2009 12:30:45] <rmatte> haha
[09-Sep-2009 12:30:55] <ckrough> "mray: my life on zenoss" .com
[09-Sep-2009 12:31:01] <rmatte> Leave it to a Zenoss user to think of something like that.
\n[09-Sep-2009 12:31:39] <ckrough> I have plans for my kegerator. I\nusually buy guiness, which is a special order, so I want to be alerted\nwhen the keg is 2 weeks to empty. I think zenoss is going to play a\nrole.
[09-Sep-2009 12:31:50] <rmatte> We might have to use predictive thresholding though, depending on the day
[09-Sep-2009 12:31:58] <mrayzenoss> I'm sure others will be interested in that ZenPack, be sure to send it in
[09-Sep-2009 12:32:55] <rmatte> ZenPacks.MattRay.Boozer.Pack-1.0-py2.4.egg
\n[09-Sep-2009 12:33:25] <venturaville> does anyone have an example\nof a zenpack which creates a /Mibs/sub-folder and/or a\n/Devices/sub-folder via script rather than objects.xml?
[09-Sep-2009 12:34:20] <rmatte> venturaville: just curious, but why use a script instead of objects.xml?
\n[09-Sep-2009 12:34:49] <venturaville> rmatte: I want to create\nit, but I don't want the zenpack to 'own it' when it gets uninstalled.
[09-Sep-2009 12:35:06] <rmatte> oh, I see
[09-Sep-2009 12:38:38] <venturaville> anyone anyone ... bueller?
[09-Sep-2009 12:38:42] <venturaville> :-)
[09-Sep-2009 12:40:33] <Brad_K> not me -- i haven't even earned my white belt yet.
[09-Sep-2009 12:40:43] <rmatte> You should probably start with how to get a script to execute in a ZenPack to begin with
[09-Sep-2009 12:40:52] <rmatte> then figure out what zendmd commands you'd need to use to make an organizer
[09-Sep-2009 12:40:58] <ckrough> can it be part of the __init__
[09-Sep-2009 12:41:06] <venturaville> ckrough: yes
[09-Sep-2009 12:41:54] <metalklesk> i'm going to launch
[09-Sep-2009 12:41:58] <metalklesk> see ya
[09-Sep-2009 12:42:15] <chudler> BLASTOFF
[09-Sep-2009 12:42:16] * ckrough thinking of a launch joke
[09-Sep-2009 12:48:08] <rmatte> "Copy that odyssey, we roger your roll."
[09-Sep-2009 13:01:18] <Brad_K> And there you have the history of how donut holes were first created....
[09-Sep-2009 13:05:15] <rmatte>
[09-Sep-2009 13:07:42] <Brad_K> I guess you really do love your bread products, eh?
[09-Sep-2009 13:09:14] <rmatte> yessir, nom nom
[09-Sep-2009 13:26:15] <chemist> evening all
[09-Sep-2009 13:26:57] <mrayzenoss> heads up ZenPack devs
[09-Sep-2009 13:27:18] <mrayzenoss> we're going to be moving community.zenoss.org to zenpacks.zenoss.org tonight
[09-Sep-2009 13:27:31] <mrayzenoss> Trac and the wiki should be unaffected
[09-Sep-2009 13:27:40] <mrayzenoss> but Subversion will need to get updated to the new URL
[09-Sep-2009 13:27:53] <mrayzenoss> there will be an announcement about it on the blog and forums soon
[09-Sep-2009 13:28:06] <mrayzenoss> (probably not a lot of people this affects)
\n[09-Sep-2009 13:38:48] <chemist> in adminstered objects, next to\nthe role type, there is a level (currently 1) can someone explain what\nthis is for?
[09-Sep-2009 13:41:09] <mrayzenoss> this is related to the use of permissions, I don't think it does much in Core unless you dig into Zope
[09-Sep-2009 13:41:26] <chemist> ok, thanks
\n[09-Sep-2009 13:41:37] <mrayzenoss> Jane Curry's got a paper on\nACLs in Core http://www.skills-1st.co.uk/papers/jcurry.html
[09-Sep-2009 13:41:51] <mrayzenoss> http://www.skills-1st.co.uk/papers/jane/users_events_zproperties_paper.pdf
[09-Sep-2009 13:42:24] <chemist> ta
[09-Sep-2009 13:53:52] <metalklesk> i'm back
[09-Sep-2009 13:54:36] <metalklesk> i have a question
[09-Sep-2009 13:55:20] <rmatte> ask
[09-Sep-2009 13:55:43] <jb> mrayzenoss: have you seen any Mitel zenpacks
[09-Sep-2009 13:55:44] <jb> ?
[09-Sep-2009 13:55:49] <mrayzenoss> nope
\n[09-Sep-2009 13:55:57] <metalklesk> with a linux box, let's say\nubuntu server 9.04 or mandriva 2009 spring with 4 GB ram CPU dual core\nand so on, how much devices i can monitor ?
[09-Sep-2009 13:56:02] <rmatte> nope
[09-Sep-2009 13:56:12] <jb> ok, going to start checking our controllers to see what kind of info I can get via snmp
[09-Sep-2009 13:56:18] <metalklesk> with devices i mean more windows and linux boxes than other type of devices
[09-Sep-2009 13:56:28] <rmatte> metalklesk: well, with 4 gigs of ram so far I've been able to monitor close to 400
[09-Sep-2009 13:56:47] <rmatte> metalklesk: you'd want 6 or 8 gigs of ram if you want to get in to thousands of devices
[09-Sep-2009 13:56:48] <jb> Mem: 32941048k total, 23419780k used, 9521268k free, 792036k buffers
[09-Sep-2009 13:56:49] <jb>
[09-Sep-2009 13:56:53] <mrayzenoss> metalklesk: http://forums.zenoss.com/viewtopic.php?t=6230
[09-Sep-2009 13:56:58] <mrayzenoss> that thread has some numbers
[09-Sep-2009 13:57:04] <metalklesk> great
[09-Sep-2009 13:57:30] <mrayzenoss> zenpacks.zenoss.org is live
[09-Sep-2009 13:57:38] <cgibbons> yippie?
[09-Sep-2009 13:57:46] <mrayzenoss> just a URL redirect
[09-Sep-2009 13:58:07] <jb> whats your backlog looking like?
[09-Sep-2009 13:59:00] <mrayzenoss> well, I'm not doing any until next week
[09-Sep-2009 13:59:07] <mrayzenoss> so rmatte has 2 in my queue
[09-Sep-2009 13:59:12] <jb> i still have one?
[09-Sep-2009 13:59:26] <mrayzenoss> ugh, yeah
[09-Sep-2009 13:59:30] <jb> hah
[09-Sep-2009 13:59:31] <mrayzenoss> never finished it
[09-Sep-2009 13:59:35] <mrayzenoss> IBM 3584
[09-Sep-2009 13:59:37] <jb> yeah
[09-Sep-2009 13:59:42] <mrayzenoss> so there are 5
\n[09-Sep-2009 13:59:58] <mrayzenoss> and we're migrating to the\nnew Community site right now, so I'm not adding new content
[09-Sep-2009 14:00:06] <jb> whens that going to launch?
[09-Sep-2009 14:00:09] <mrayzenoss> Monday
[09-Sep-2009 14:00:12] <jb> nice..
[09-Sep-2009 14:00:14] <jb> what platform again?
[09-Sep-2009 14:00:42] <mrayzenoss> Jive Social Business Software: http://www.jivesoftware.com/products
[09-Sep-2009 14:00:53] <mrayzenoss> we're migrating ZenPacks right now
[09-Sep-2009 14:00:59] <jb> k
[09-Sep-2009 14:01:02] <mrayzenoss> I'm mapping through the wiki/plone/trac content
[09-Sep-2009 14:01:06] <mrayzenoss> and redirecting URLs
[09-Sep-2009 14:01:15] <mrayzenoss> there'll be a blog post probably tomorrow
[09-Sep-2009 14:01:16] <metalklesk> zenoss is ver efficient compared to BMC Portal
[09-Sep-2009 14:01:25] <chudler> jive.. seems vaguely familiar. interesting niche "social business"...
[09-Sep-2009 14:01:39] <metalklesk> 400 devices with just one server is ver well
[09-Sep-2009 14:01:39] <mrayzenoss> JBoss.org, Communities.vmware.com
[09-Sep-2009 14:01:59] <metalklesk> *vey
[09-Sep-2009 14:02:05] <metalklesk> *very
[09-Sep-2009 14:02:38] <mrayzenoss> heh, http://communities.bmc.com uses it too
[09-Sep-2009 14:04:34] <rmatte> I can't wait until the whole new zenoss community site is up and buzzing
[09-Sep-2009 14:04:51] <rmatte> Sept 14th or 15th was the goal for the date as I recall
[09-Sep-2009 14:05:10] <mrayzenoss> yeah, we think we'll hit it
[09-Sep-2009 14:05:18] <mrayzenoss> it'll be a rolling start
[09-Sep-2009 14:05:34] <mrayzenoss> the announcement should be tomorrow, the forums will be offline over the weekend
[09-Sep-2009 14:05:44] <rmatte> cool
[09-Sep-2009 14:05:56] <mrayzenoss> and Monday we'll bring it up and kick the tires
[09-Sep-2009 14:06:09] <mrayzenoss> but for now, community.zenoss.org is free to be re-used
[09-Sep-2009 14:06:55] <mrayzenoss> and tonight we'll probably break SVN
[09-Sep-2009 14:07:11] <mrayzenoss> which was the purpose of the blog post
[09-Sep-2009 14:07:19] <rmatte> some of those ZenPacks need updating badly lol
[09-Sep-2009 14:07:26] <rmatte> particularily the .zips
[09-Sep-2009 14:07:38] <mrayzenoss> well, they're slowing getting eggifyzenpack'd
[09-Sep-2009 14:07:44] <rmatte> cool
[09-Sep-2009 14:07:50] <mrayzenoss> Chet updated the Celerra one
[09-Sep-2009 14:08:01] <jb> are you migrating forums?
[09-Sep-2009 14:08:05] <mrayzenoss> absolutely
[09-Sep-2009 14:08:06] <jb> from phpbb?
[09-Sep-2009 14:08:07] <jb> k
[09-Sep-2009 14:08:18] <mrayzenoss> and users
[09-Sep-2009 14:08:40] <mrayzenoss> and most content, we'll let it get sorted a bit more after the launch
[09-Sep-2009 14:08:46] <mrayzenoss> but it'll all be under 1 roof
[09-Sep-2009 14:09:31] <ckrough> what are you moving to from phpbb?
[09-Sep-2009 14:09:51] <mrayzenoss> Jive will hold the forums as well
[09-Sep-2009 14:10:14] <mrayzenoss> it's forums/wiki/docs/community all bundled in 1
[09-Sep-2009 14:10:24] <mrayzenoss> and the blog.
[09-Sep-2009 14:10:26] <mrayzenoss> and search
[09-Sep-2009 14:12:00] <Troubadix09> mrayzenoss: I have a question :-)
[09-Sep-2009 14:13:13] <metalklesk> one question, can i exclude a device, that i already have added, from monitoring ?
[09-Sep-2009 14:13:33] <metalklesk> i mean without delete it
[09-Sep-2009 14:13:51] <metalklesk> like monitoring on/off of bmc portal
[09-Sep-2009 14:15:32] <slip_> put it into the "discovered" item group...
[09-Sep-2009 14:15:44] <mrayzenoss> discovered is still monitored
[09-Sep-2009 14:16:28] <mrayzenoss> metalklesk: you can set the production state on the Edit tab
\n[09-Sep-2009 14:16:45] <Troubadix09> mrayzenoss: are the\nzenoss-team planning to take a place at one of the next CEbit in\nHannover, Germany?
[09-Sep-2009 14:17:25]\n<mrayzenoss> CEbit? Unlikely, though we have talked about\nheading to an European event next year, thinking FOSDEM
[09-Sep-2009 14:17:51] * rmatte reminds mrayzenoss that there's beer in Germany, good beer
[09-Sep-2009 14:18:02] * mrayzenoss has been to Germany many times
[09-Sep-2009 14:18:10] <rmatte> hehe
[09-Sep-2009 14:18:28] <rmatte> I haven't really traveled anywhere
\n[09-Sep-2009 14:18:45] <rmatte> I'd like to, just don't have the\nfinances, and I don't have the kind of job that requires me to travel\n(yet)
[09-Sep-2009 14:19:40] <venturaville> you\ncould go work for Google.... I have a friend over there that ends up in\nIreland and Australia from time to time
[09-Sep-2009 14:20:04] <rmatte> I'd love to work for google quite honestly, but I don't really know what I'd do for them hehe
[09-Sep-2009 14:20:18] <rmatte> Unless they want to use Zenoss to monitor stuff
[09-Sep-2009 14:20:36] <rmatte> (Could do Linux administrator as well I suppose)
[09-Sep-2009 14:20:43] <rmatte> administration, rather
[09-Sep-2009 14:20:48] <rhettardo> mmm spaten
[09-Sep-2009 14:21:26] <rmatte> rhettardo: hehe
[09-Sep-2009 14:21:45] <rhettardo> love me some optimator
[09-Sep-2009 14:27:53] <Troubadix09> bye all
[09-Sep-2009 14:28:42] <mrayzenoss> I guess he wanted me to go to CEbit
[09-Sep-2009 14:30:24] <rhettardo> oh well
[09-Sep-2009 14:31:20] <mrayzenoss> it's still in the planning stages... "wouldn't it be nice to go to Europe..."
[09-Sep-2009 14:32:04] <rmatte> and the answer is: "Yes, it would"
[09-Sep-2009 14:32:09]\n<Curt> Hello all, we have a predominantly windows server shop\nhere with a few linux servers doing various things. i'm trying to\nfind the best open source software for monitoring servers and log\nfiles. would zenoss make a good fit here or would there be other\nsimpler method for about 40 servers
[09-Sep-2009\n14:32:50] <rmatte> Curt: great for monitoring servers, doesn't\nhave any built in functionality for monitoring log files, other than if\nyou have syslogd sending syslog events to Zenoss
[09-Sep-2009 14:33:11] <rmatte> and in Windows case, you can monitor the event log via WMI
[09-Sep-2009 14:33:30] <metalklesk> wmi is cool for monitoring windows with zenoss
\n[09-Sep-2009 14:33:31] <rmatte> or just enable SNMP traps on the\nwindows boxes, which will result in them sending traps to Zenoss for\nmajor eventlog events
[09-Sep-2009 14:34:00]\n<rmatte> Curt: if you use the Core version of Zenoss you really\nneed a mix of WMI and SNMP to get full monitoring coverage
[09-Sep-2009 14:34:14] <rmatte> Curt: the enterprise version of Zenoss supports full WMI monitoring
[09-Sep-2009 14:34:33] <rmatte> Curt: Zenoss' Linux monitoring support via SNMP and SSH is excellent even in Core
\n[09-Sep-2009 14:34:37] <mrayzenoss> vaguely changing the\nsubject... has anyone used\nhttp://www.zenoss.com/community/projects/zenpacks/wmifilesmonitor ?
[09-Sep-2009 14:35:00] <metalklesk> rmatte and mrayzenoss: when are you planning to come visit Chile ?
[09-Sep-2009 14:35:04] <metalklesk>
[09-Sep-2009 14:35:17] <rmatte> mrayzenoss: Isn't that an older ZenPack?
[09-Sep-2009 14:35:30] <rmatte> metalklesk: as soon as you send me my plane ticket
[09-Sep-2009 14:35:35] <mrayzenoss> rmatte: yeah, hasn't been updated to the new WMI Data Source yet
[09-Sep-2009 14:35:38] <metalklesk> ejjejejje
[09-Sep-2009 14:35:53] <mrayzenoss> metalklesk: any big open source community events down there?
[09-Sep-2009 14:36:08]\n<rmatte> mrayzenoss: quite honestly, any packs not using the new\nWMI Data Source pack should be either updated to use it, or purged
[09-Sep-2009 14:36:27] <metalklesk> linuxfest
[09-Sep-2009 14:36:34] <mrayzenoss> yeah, with the new website you can add that sentiment to each one
[09-Sep-2009 14:36:51] <rmatte> mrayzenoss: and I shall, there are some that have been particularily frustrating
[09-Sep-2009 14:37:05] <rmatte> mrayzenoss: such as the old Brocade pack
\n[09-Sep-2009 14:37:30] <mrayzenoss> rmatte: I've sent 3 emails to\nthe author of that new Brocade ZenPack asking him about it
[09-Sep-2009 14:37:33] <metalklesk> the other day i was talking with a friend about making our own company
[09-Sep-2009 14:37:48] <metalklesk> a competition for the company i work now
[09-Sep-2009 14:37:57] <mrayzenoss> rmatte: he told me he was working on it last I heard
[09-Sep-2009 14:38:04] <metalklesk> using open source and for monitoring of course zenoss
[09-Sep-2009 14:38:06]\n<rmatte> mrayzenoss: well, the new Brocade pack is only for a\nspecific type of Brocade, (not the type we have), the old Brocade pack\nis for the type that we use but it's only configured to monitor port 1\non the switch
[09-Sep-2009 14:38:18] <rmatte> mrayzenoss: so it's useless since it ignores the other 24 ports lol
[09-Sep-2009 14:38:27] <mrayzenoss> metalklesk: we plan on making the Core partners much more visible in the new site
[09-Sep-2009 14:38:39] <mrayzenoss> rmatte: yeah
[09-Sep-2009 14:38:50] <metalklesk> that is what we would want, be partners
[09-Sep-2009 14:39:04] <metalklesk> here in Chile is a growing market so sell service
\n[09-Sep-2009 14:39:16] <rmatte> mrayzenoss: people need to stop\ngeneralizing their ZenPacks so much, that's like me writing a ZenPack\nfor 1 particular model of HP server and releasing it as a "HP Server\nZenPack", when it reality it doesn't work with the other gajillion\nmodels of HP servers.
[09-Sep-2009 14:39:18] <metalklesk> *to
[09-Sep-2009 14:39:35] <metalklesk> rmatte, i'm agree
[09-Sep-2009 14:39:59] <metalklesk> SuSE zenpack does not monitor SLES or SLED, only openSUSE
[09-Sep-2009 14:40:14] <metalklesk> it should be openSUSE Zenpack
[09-Sep-2009 14:40:22] <rmatte> metalklesk: you'd have to talk to Rocket about that
[09-Sep-2009 14:40:41] * rmatte points accusingly at rocket
[09-Sep-2009 14:40:42] <metalklesk> ok, but for now is just an idea
[09-Sep-2009 14:40:57] <ckrough> hmm. what would you call the * section of a zenoss path /zport/dmd/*/devices/
[09-Sep-2009 14:41:00] <rmatte> yeh, I agree with that one, would be good to rename it to openSUSE
[09-Sep-2009 14:41:04] <ckrough> container?
[09-Sep-2009 14:41:18] <metalklesk> i am looking to our clients and they are not happy with BMC Software
[09-Sep-2009 14:41:39] <rmatte> ckrough: Organizer
[09-Sep-2009 14:41:40] <metalklesk> Entel SA has dropped BMC Patrol for Nagios
[09-Sep-2009 14:42:05] <metalklesk> BCI Bank is dropping BMC Patrol for prognosis
[09-Sep-2009 14:42:25] <rmatte> ckrough: actually, isn't it just /zport/dmd/Devices?
[09-Sep-2009 14:42:38] <metalklesk> 52% of our clients does not renew contracts for Service Assurance
[09-Sep-2009 14:42:43] <ckrough> rmatte: /zport/dmd/Devices/Server/Linux/devices/
[09-Sep-2009 14:42:58] <rmatte> right
[09-Sep-2009 14:43:11] <rmatte> Organizer path
[09-Sep-2009 14:43:41] <metalklesk> and is clear, open source is the sh**, not propietary software business
[09-Sep-2009 14:43:46] <metalklesk>
[09-Sep-2009 14:43:58] <rmatte> opensource is the future of software development, it's just taking time to catch on
[09-Sep-2009 14:44:11] <rmatte> mainly because of micro$oft
[09-Sep-2009 14:44:15] <metalklesk> yes, here in Chile is even more slowly
[09-Sep-2009 14:44:40] <metalklesk> our goverment makes hidden contracts
[09-Sep-2009 14:44:45] <metalklesk> with microsoft
[09-Sep-2009 14:44:45] <rmatte> I'm glad Vista was so horrible actually, made a ton of people decide to try Linux
[09-Sep-2009 14:44:51] <metalklesk> yes
\n[09-Sep-2009 14:44:58] <ckrough> buying knowledge (people) is\nbecoming less expensive than paying insane fees for application\nextensions
[09-Sep-2009 14:45:13] <rmatte> and\nnow that they've had time to get used to it and enjoy it, they probably\nwon't bother going back to windows
[09-Sep-2009 14:45:32] <rmatte> the only real place where microsoft has the PC industry by the balls is games
\n[09-Sep-2009 14:45:36] <metalklesk> last year we knew about some\ncontracts about our goberment and microsoft, it was disgusting
[09-Sep-2009 14:46:09] <metalklesk> yes
[09-Sep-2009 14:46:39] <metalklesk> yesterday my friend quits from this company
[09-Sep-2009 14:47:04] <metalklesk> because he gets mad with the CEO
[09-Sep-2009 14:47:07]\n<rmatte> yeh, our government is heavy in to Microsoft too, it'd\nbe nice to see them take some initiative and go opensource. The\nargument is always "well who is going to support it". I'm sorry, but\nif you drop all of the licensing fees that you're paying for Windows\nyou could have a dedicated Linux support person per floor in every\nbuilding and still save money.
[09-Sep-2009 14:47:30] <metalklesk> yes
[09-Sep-2009 14:47:48] <metalklesk> so as my friend quits and he have a lot of potential
[09-Sep-2009 14:48:04] <metalklesk> we talked about build our own company
\n[09-Sep-2009 14:48:39] <rmatte> I'm tempted to do my own\nmonitoring thing on the side, I just don't know if my company would\nallow me to.
[09-Sep-2009 14:48:46] <metalklesk> as he quits, this company looses 2.5 million dolars
[09-Sep-2009 14:48:58] <rmatte> yikes
[09-Sep-2009 14:49:01] <metalklesk> now here are desperate
[09-Sep-2009 14:49:08] <metalklesk> because of that
[09-Sep-2009 14:49:25] <metalklesk> bussiness around BMC service automation is dead now
[09-Sep-2009 14:49:49] <metalklesk> my friend was the one who was making those business
[09-Sep-2009 14:49:56] <metalklesk> in Chile and Colombia
[09-Sep-2009 14:50:24] <metalklesk> so i think we can do a lot with zenoss, obviusly with partnership
[09-Sep-2009 14:50:50] <rmatte> yeh, Zenoss is pretty impressive and it's getting better and better
[09-Sep-2009 14:51:05] <metalklesk> yes
[09-Sep-2009 14:51:38] <metalklesk> the 2 open source software i think are really impressive are Zenoss and OpenBravo
[09-Sep-2009 14:52:00] <metalklesk> openbravo are an ERP Software
[09-Sep-2009 14:52:31] <metalklesk> i have seen a lot of companys interested about it here in Chile
[09-Sep-2009 14:53:06] <rmatte> openbravo eh?
[09-Sep-2009 14:53:08] <rmatte> what does it do?
\n[09-Sep-2009 14:53:23] <metalklesk> in my blog i wrote some\narticles about openbravo and i received a lot of comments about people\nthat want it in their companies instead of SAP
[09-Sep-2009 14:53:24] <rmatte> oh, inventory system
[09-Sep-2009 14:53:44] <metalklesk> inventory is just one of the modules
[09-Sep-2009 14:54:05] <metalklesk> accounts module too
[09-Sep-2009 14:54:24] <metalklesk> you can manage clients, providers, inventory, accounts
[09-Sep-2009 14:54:32] <metalklesk> multi enterprise
[09-Sep-2009 14:54:50] <metalklesk> is really impressive what you can do with open bravo
[09-Sep-2009 14:55:33] <metalklesk> my blog really jumps high here in latin america because of the open bravo tips i gave
[09-Sep-2009 14:55:54] <metalklesk> maybe it happens again because the article i wrote about zenoss
[09-Sep-2009 14:55:56] <metalklesk>
[09-Sep-2009 14:56:44] <metalklesk> http://metalklesk.blogspot.com/
[09-Sep-2009 14:57:35] <chudler> very detailed... cant read it though ;-)
[09-Sep-2009 14:57:56] <rmatte> same here hehe
[09-Sep-2009 14:58:08] <metalklesk> jejee is in spanish
[09-Sep-2009 15:01:16] <metalklesk> notice the pictures in "Mis imágenes", me with 2 friends and Ronnie Coleman
[09-Sep-2009 15:01:18] <metalklesk> joojoj
[09-Sep-2009 15:01:52] <chemist> I can not get zenoss to email me when a critical event occurrs, I can send a test mail
[09-Sep-2009 15:02:50] <chemist> I have added the locations to my administered objects
[09-Sep-2009 15:08:25] <rmatte> mrayzenoss: yeh, I can imagine
[09-Sep-2009 15:08:28] <rmatte> oops
[09-Sep-2009 15:10:22] <ckrough> "def getAllRRDTemplatesPainfully(self, rrdts=None):" ...cute
[09-Sep-2009 15:14:18] <ckrough> jsonGetDeviceNames ignores the path and always returns all matches. sad panda.
[09-Sep-2009 15:19:22] <rhettardo> chemist: sounds like an alerting rules issue
\n[09-Sep-2009 15:21:25] <chemist> rhettardo: do I have to create\nexplicit rules? I thought if I didn't create any I would get alerted\nany time, every time
[09-Sep-2009 15:21:51] <rhettardo> i dont recall what the default settings are. but i have my rules specific
[09-Sep-2009 15:22:50] <metalklesk> can i have custom threshold for device ?
[09-Sep-2009 15:23:31] <chemist> rhettardo: ok, I will give it a go
[09-Sep-2009 15:23:33] <chemist> thanks
[09-Sep-2009 15:23:45] <mrayzenoss> metalklesk: yeah, you can copy and edit the templates
[09-Sep-2009 15:23:46] <rhettardo> np
[09-Sep-2009 15:25:07] <metalklesk> is there a documentation about threshold customization ?
\n[09-Sep-2009 15:25:37] <t3los> I've been looking through the\ndocumentation on this one, and I cannot find anything about a grey\nstatus light
[09-Sep-2009 15:25:45] <t3los> I know what yellow means, and red, and green
[09-Sep-2009 15:25:55] <t3los> I'm guessing it means it can't find the events?
[09-Sep-2009 15:26:16] <t3los> I can monitor my CPU times and memory in the perf tab, but the lights won't get a color?
[09-Sep-2009 15:30:17] <t3los> Anyone?
[09-Sep-2009 15:30:34] <mrayzenoss> t3los: it means it's unmonitored
\n[09-Sep-2009 15:31:21] <t3los> Yeah that's what I figured. I\ncan't find anything in HW that will let meset monitoring true for these\ncomponents.
[09-Sep-2009 15:32:34] <jb> or that it just doesn't know the status of the component..
[09-Sep-2009 15:33:21] <chemist> is there anyway to set what time the modelling of a device takes place?
\n[09-Sep-2009 15:34:38] <t3los> Is there a way to get the\ncomponent out of the list if it's just going to stay grey? I cannot\nfind a place to map the event to it, so if it's just easier to get it\nout of the list... lol
[09-Sep-2009 15:35:51] <metalklesk> manually changing the production state ?
[09-Sep-2009 15:37:08] <rmatte> chemist: not really
\n[09-Sep-2009 15:37:26] <t3los> Well I want to keep it in\nproduction. If I could find the components property page that would\nsolve the problem.
[09-Sep-2009 15:37:28] <rmatte> chemist: you could do it with a cron job or something
[09-Sep-2009 15:38:22] <rmatte> t3los: hunh? "get it out of the list"?
[09-Sep-2009 15:38:42] <t3los> The component list, with the status lights
[09-Sep-2009 15:38:57] <t3los> cpu times, and memory pages, etc
[09-Sep-2009 15:39:02] <t3los> they won't get a status color
\n[09-Sep-2009 15:39:25] <t3los> so if I am unable to get them to\nbe monitored, I wish to just remove them from that components list.
[09-Sep-2009 15:39:36] <rmatte> t3los: oh
[09-Sep-2009 15:40:22] <t3los> Any ideas?
[09-Sep-2009 15:40:26] <rmatte> t3los: good question, I'd have to look in to that
[09-Sep-2009 15:41:29] <rmatte> t3los: I think the CPU one turns red if an event of class /Perf/CPU comes in
[09-Sep-2009 15:41:39] <rmatte> such as a threshold
\n[09-Sep-2009 15:41:58] <t3los> So they are meant to stay grey? As\nopposed to the others that are green when working right?
[09-Sep-2009 15:42:02] <t3los> That's what doesn't make sense.
[09-Sep-2009 15:43:20] <rmatte> t3los: which are grey?
[09-Sep-2009 15:44:24] <t3los> cpuPercentProcessorTime, memoryAvailableKBytes, memoryPagesPerSec
[09-Sep-2009 15:59:38] <jb> think im going to start an EqualLogic PS zenpack
[09-Sep-2009 16:01:44] <metalklesk> do it
[09-Sep-2009 16:02:40] <metalklesk> with more zenpacks, more robust zenoss will be (Yoda spoke)
[09-Sep-2009 16:04:56]\n<rmatte> nice, I'm hearing the guys talking about a call today, a\nswitch went down, they called the site, apparently someone had knocked\nthe switch off of the workbench/table or whatever that it was sitting\non, in to a pale of water
[09-Sep-2009 16:05:09] <rmatte> I love some of our clients
[09-Sep-2009 16:05:14] <rmatte>
[09-Sep-2009 16:05:36] <mrayzenoss> rmatte: that is wrong on so many levels
[09-Sep-2009 16:05:47] <rmatte> hehe
[09-Sep-2009 16:06:07] <rmatte> I've seen some pretty interesting stuff
[09-Sep-2009 16:06:49]\n<rmatte> a couple years back when I was a NOC analyst, a router\ndied at one of the sites because the coffee maker was sitting on top of\nit and someone spilled coffee on it
[09-Sep-2009 16:07:31]\n<rmatte> There was also an outage in northern canada caused by an\neagle trying to fly with a deer's head, it couldn't quite gain enough\naltitude and crashed in to a power line
[09-Sep-2009 16:08:18]\n<rmatte> There was a satellite site in northern canada that\ncouldn't be repaired because when the technician took his snowmobile up\na polarbear was blocking the door
[09-Sep-2009 16:09:35]\n<rmatte> There was a major outage in europe where someone broke\nin to one of the carrier's sites and stole 16 fiber cards which were\napparently extremely expensive, it took a day and a half to resolve the\noutage because they didn't have enough fiber cards in the country to\nreplace them with so they had to fly them in
[09-Sep-2009 16:09:41] <rmatte> I believe that was in London
[09-Sep-2009 16:09:47] <rmatte> if I remember correctly
[09-Sep-2009 16:09:55] <mrayzenoss> awesome
\n[09-Sep-2009 16:10:26] <rmatte> There was a site in Mexico that\nkept going down because people were coming and cutting down the copper\ncable running from the pole to the site and stealing it
[09-Sep-2009 16:10:31] <rmatte> it actually happened 3 times
\n[09-Sep-2009 16:10:50] <rmatte> one of the analysts was like "Why\ndon't just install a camera?" and someone was like "They'd probably\nsteal that too"
[09-Sep-2009 16:11:17]\n<rhettardo> that happened at my old job. some one came up w/ a\nladder and stole the security cameras
[09-Sep-2009 16:11:19] <mrayzenoss> lol
[09-Sep-2009 16:11:45] <rmatte> being an analyst was quite fun sometimes
[09-Sep-2009 16:11:49] <rmatte> hehe
[09-Sep-2009 16:11:54] <metalklesk> ajjajajajajajajajajaja
[09-Sep-2009 16:11:59] <metalklesk> nice stories
[09-Sep-2009 16:12:18] <metalklesk> stealing cables, sound like here in Chile
[09-Sep-2009 16:12:47] <metalklesk> some people are stealing cables in highways
[09-Sep-2009 16:13:05] <rmatte> copper can be sold for quite a bit of cash apparently
[09-Sep-2009 16:13:20] <metalklesk> shutting down systems of several companies
\n[09-Sep-2009 16:13:48] <rmatte> Oh, we were doing monitoring for\nBig Lots (the retail outlet)... during hurricane season it was always\nfun
[09-Sep-2009 16:14:13] <mrayzenoss> because of power loss or looting?
\n[09-Sep-2009 16:14:27] <rmatte> "Yeh, we've seen this site go\ndown." "Oh, yeh, that store, yeh, that store is no longer there." "What\ndo you mean no longer there?" "It was ripped apart by a hurricane."
[09-Sep-2009 16:14:33] <mrayzenoss> wow
[09-Sep-2009 16:14:48] <rmatte> yup
[09-Sep-2009 16:14:58] <metalklesk> wow
[09-Sep-2009 16:15:00] <rmatte> that was just last year, that happened to 2 of their stores
[09-Sep-2009 16:16:54] <malbon> hi all
[09-Sep-2009 16:17:03]\n<rmatte> lmao, one of my coworkers came and asked me how to\nconfigure WMI for a device in Zenoss, I just heard him say "Argh, I\nhate WMI!" from across the room
[09-Sep-2009 16:17:17] <rmatte> I'm guessing it's not going well
[09-Sep-2009 16:17:37] <rmatte> Probably to do with actually configuring the WMI access, not with Zenoss
\n[09-Sep-2009 16:17:47] <cam__> i installed zenoss from the debian\nrepository listed in the install guide. it seems to include its own\nversion of mysql--is that necessary?
[09-Sep-2009 16:18:10] <rmatte> cam__: it comes bundled with all of the necessary dependencies
\n[09-Sep-2009 16:18:37] <rmatte> cam__: If you really really want\nto, you could make it work with an existing mysql server if needed
[09-Sep-2009 16:19:05] <rmatte> cam__: the deb is basically just the Ubuntu stack installer wrapped in a deb file
\n[09-Sep-2009 16:20:53] <cam__> rmatte: ok. it's failing for some\nreason (zeoctl is not found). i was just digging around and thought\nthat it was weird that it didn't depend on mysql-server package rather\nthan bundling its own
[09-Sep-2009 16:21:07] <rmatte> zeoctl is one of the daemons that needs to run
[09-Sep-2009 16:21:29] <rmatte> cam__: it doesn't pull any dependencies down from aptitude
\n[09-Sep-2009 16:21:50] <rmatte> cam__: it's just a deb file to\nmake it so that the package can be managed with aptitude, but it's not\nfully integrated in to it
[09-Sep-2009 16:22:05] <cam__> rmatte: ok. i'll try installing with the stack installer instead
[09-Sep-2009 16:22:09]\n<rmatte> cam__: also, Zenoss required specific versions of things\nsuch as python, so it comes bundled with that as well, python2.4
\n[09-Sep-2009 16:22:30] <rmatte> cam__: the stack installer is\nwhat I use, it'll do the exact same install as the .deb did though
[09-Sep-2009 16:22:39] <rmatte> cam__: it just won't add it in to the package inventory
[09-Sep-2009 16:22:57] <cam__> rmatte: i see. i wonder why it didn't install zeoctl
[09-Sep-2009 16:23:13] <rmatte> rhettardo: updatedb && locate zeoctl as root
[09-Sep-2009 16:23:16] <rmatte> oops
[09-Sep-2009 16:23:20] <rmatte> rhettardo: updatedb && locate zeoctl as root
[09-Sep-2009 16:23:22] <rmatte> bah
[09-Sep-2009 16:23:24] <rhettardo> :O
[09-Sep-2009 16:23:35] <rmatte> I'm trying to type do but it's auto-completing as his nick
[09-Sep-2009 16:23:38] <rhettardo> ok but why?!
[09-Sep-2009 16:23:38] <rmatte> hehe
[09-Sep-2009 16:23:57] <cam__> haha, i did that but it's really not there
[09-Sep-2009 16:24:09] <rhettardo> you made need a sudo updatedb && locate zeoctl
[09-Sep-2009 16:24:11] <rmatte> cam__: weird, then I'd recommend trying the stack installer for sure
[09-Sep-2009 16:24:15] <rhettardo> i dont know. just a suggestion
[09-Sep-2009 16:24:40] <rmatte> rhettardo: I said as root, same difference lol
[09-Sep-2009 16:24:47] <rhettardo> oh i didnt see that
[09-Sep-2009 16:24:53] <rhettardo> all i saw was the highlighted lines w/ my bick
[09-Sep-2009 16:24:54] <rhettardo> nick
[09-Sep-2009 16:24:58] <rmatte> ah lol
\n[09-Sep-2009 16:25:11] <sacpinball> yo; do i need anything\ninstalled on my windows servers to collect performance stats; i have\nwmi setup and snmp, do i need snmp informant?
[09-Sep-2009 16:25:15] <rmatte> yeh, apparently my bitchx script auto-completes names if a colon is put after the word
\n[09-Sep-2009 16:25:53] <rmatte> sacpinball: not if you use my\nperformance zenpack:\nhttp://www.zenoss.com/community/projects/zenpacks/winsnmpperf
[09-Sep-2009 16:26:18] <sacpinball> sweet I will check it out
\n[09-Sep-2009 16:26:18] <rmatte> sacpinball: there's an even\nsimpler version of that ZenPack which should be publicly available next\nweek hopefully, but I can pass it to you if you want
[09-Sep-2009 16:26:30] <rmatte> sacpinball: are you looking to monitor multiple cores or just total CPU usage?
[09-Sep-2009 16:26:45] <sacpinball> probably both if possible depneing on the system
[09-Sep-2009 16:26:56] <rmatte> sacpinball: ok, then you'll want that pack
\n[09-Sep-2009 16:27:38] <rmatte> sacpinball: that's the advanced\nback, the simpler pack will just be for monitoring the total, and won't\nhave the different device groupings for each number of CPUs
\n[09-Sep-2009 16:28:38] <rmatte> sacpinball: it currently uses\nbash scripts to collect the data, make sure that you have bc and\nsnmpwalk installed
[09-Sep-2009 16:28:44] <rmatte> apt-get install snmpwalk bc
\n[09-Sep-2009 16:29:20] <rmatte> I think I can modify it to not\nuse bc at all, I'll probably do that within a couple of weeks
[09-Sep-2009 16:30:24] <sacpinball> that is sweet thanks
[09-Sep-2009 16:30:27] <sacpinball> i will try it out
[09-Sep-2009 16:30:51] <rmatte> np
[09-Sep-2009 16:50:16] <metalklesk> Zenoss Administration Guide is very educational
[09-Sep-2009 16:50:48] <rmatte> yup
[09-Sep-2009 16:50:56] <rmatte> it's definitely worth reading
\n[09-Sep-2009 16:51:14] <rmatte> I read through the whole thing\nstart to finish (minus the parts about enterprise components)
[09-Sep-2009 16:51:25] <mrayzenoss> the beta has an updated version available
[09-Sep-2009 16:51:30] <mrayzenoss> I'm writing up the blog post now
\n[09-Sep-2009 16:51:44] <rmatte> Jane Curry's Zenoss papers are\nalso worth reading: http://www.skills-1st.co.uk/papers/jcurry.html
[09-Sep-2009 16:52:11] <metalklesk> great
\n[09-Sep-2009 16:52:25] <rmatte> mrayzenoss: let me know when the\nstack installer is out for the beta and we'll setup a beta box
[09-Sep-2009 16:55:34] <jb> argh ok i need a python debugger
[09-Sep-2009 16:55:59] <jb> will they show you when your spacing is messed up?
[09-Sep-2009 16:56:09] <rmatte> yeh, they pretty much all do that
[09-Sep-2009 16:56:16] <rmatte> I'd recommend trying out IDLE
[09-Sep-2009 16:56:25] <rmatte> it's not very powerful, but it's nice for beginners
[09-Sep-2009 16:56:56] <rmatte> (I'm still a relative beginner myself)
[09-Sep-2009 16:56:58] <rmatte> hehe
[09-Sep-2009 16:57:19] <metalklesk> ok, i will leave the channel
[09-Sep-2009 16:57:19] <rmatte> my bash scripting skills are really good though
[09-Sep-2009 16:57:25] <rmatte> anyways, time for me to head out
[09-Sep-2009 16:57:33] <rmatte> have a nice evening guys
[09-Sep-2009 16:57:39] <metalklesk> in a few hours Chile vs Brasil
[09-Sep-2009 16:57:46] <rmatte> w00t
[09-Sep-2009 16:58:13] <jb> ok got it
[09-Sep-2009 16:58:14] <jb> lets try
[09-Sep-2009 17:00:09] <jb> ok..
[09-Sep-2009 17:00:18] <jb> so could you give me a quick runthrough?
[09-Sep-2009 17:00:22] <jb> i think i have some bad spacing
[09-Sep-2009 17:00:32] <metalklesk> bye !!
[09-Sep-2009 17:08:59] <Brad_K> i'm heading home. g'nite!
[09-Sep-2009 17:11:00]\n<LW> Anybody encounter cases where the snmp ipv6 services map\ndoes not return all of the ports that exist on a given host? we have a\ncase where we have java app listening on a high up port, and its never\ndetected, and when we manually run the model dumping the debug output,\nit never shows up in the list.... so we're thinking maybe its snmp and\nnot zenoss at all.... any ideas?
[09-Sep-2009\n17:19:15] <feutete> Is it possible to set up Zenoss in multiple\nlocations and have the remove installations push data to a central\nserver?
[09-Sep-2009 17:19:27] <feutete> I found this page--is this what I'm looking for?
\n[09-Sep-2009 17:19:28] <feutete>\nhttp://www.zenoss.com/community/docs/howtos/setting-up-multiple-zenperfsnmp-collectors-across-different-hosts
\n[09-Sep-2009 17:19:55] <feutete> essentially, the latency between\nour datacenters is too high for us to do all polling from a single\ninstance
[09-Sep-2009 17:20:17] <feutete> so we want to poll from a local instance, and push data to the central server
[09-Sep-2009 17:59:40] <ironpaw> morning
[09-Sep-2009 18:00:03] <ironpaw> feutete i want to set this up too
[09-Sep-2009 18:00:09] <ironpaw> but havne't had an opportunity yet
[09-Sep-2009 18:01:12]\n<feutete> a friend came across this as well:\nhttp://www.zenoss.com/Members/fdeckert/how-to-install-distributed-collectors/
[09-Sep-2009 18:01:39] <ironpaw> have seen both of those how tos but not used them yet.
[09-Sep-2009 23:31:17] <ironpaw> o.O
[10-Sep-2009 00:00:46] [disconnected at Thu Sep 10 00:00:46 2009]
[10-Sep-2009 00:00:46] [connected at Thu Sep 10 00:00:46 2009]
[10-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[10-Sep-2009 07:08:11] <ckrough> Is there an API guide for the Zope functions, like their is for the Classes?
\n[10-Sep-2009 07:37:46] <afink> hello everyone. I am having a\nproblem getting the performance graphs for centos to show up in\nzenoss. Zenoss says "Missing RRD File" laLoadInt5. Anyone know what I\ncan do to get that working?
[10-Sep-2009 07:40:39]\n<ckrough> that indicates that its not collecting what you are\ngraphing, or it doesnt have enough data yet. It takes 3 polling cycles\n(usually 5min) before there are enough datapoints in the rrd for it to\ngenerate graphs
[10-Sep-2009 07:40:43] <ckrough> how long have you waited?
[10-Sep-2009 07:55:12] <Apocalipse> morning
[10-Sep-2009 07:58:32] <ckrough> morning
\n[10-Sep-2009 08:10:57] <LW_> Anybody have an idea why zenmodler\nwould add this line to be monitored as a service on a host\nDEBUG:zen.IpV6ServiceMap:tcp new 16.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.22\n{'v6': 0L}, but not this line DEBUG:zen.IpV6ServiceMap:tcp new\n16.0.0.0.0.0.0.0.0.0.0.255.255.a.b.c.d.18580 {'v6': 0L} where a.b.c.d\nis the ipv4 addy of the machine... I see its a different address being\nbound to, but i would think it would still see it and understand "hey,
[10-Sep-2009 08:12:43] <afink> ckrough: months
[10-Sep-2009 08:13:36] <ckrough> afink: heh, unless your polling cycle is 2592000, it's probably something else
[10-Sep-2009 08:14:14] <metalklesk> hi everyone
[10-Sep-2009 08:14:54] <ckrough> afink: if you snmpget the laLoadInt5 directly, does the server return anything?
\n[10-Sep-2009 08:15:27] <afink> lol, \nhttp://picpaste.com/print1.png, I haven't tried I am not familiar with\nthat command. I'll look it up and give it a try.
[10-Sep-2009 08:16:07] <ckrough> afink: http://www.zenoss.com/community/docs/howtos/how-to-troubleshoot-snmp/
[10-Sep-2009 08:18:46] <afink> ckrough: laLoadInt5: Unknown Object Identifier (Sub-id not found: (top) -> laLoadInt5)
[10-Sep-2009 08:19:34] <afink> I have never had to set those up before for instance on suse..they're just there
[10-Sep-2009 08:28:08] <ckrough> afink: does this return anything? snmpget -v2c -cyourcommunity your.server.fqdn 1.3.6.1.4.1.2021.10.1.5.2
[10-Sep-2009 08:28:17] <ckrough> afink: "UCD-SNMP-MIB::laLoadInt.2 = INTEGER: 7" on my RHEL box
[10-Sep-2009 08:37:14] <adelbul> hello i read the dev doc about plugins but it is not very clear for me.
[10-Sep-2009 08:37:33] <adelbul> have you a good advice for a tutorial ?
[10-Sep-2009 08:38:11] <afink> ckrough: UCD-SNMP-MIB::laLoadInt.2 = No Such Object available on this agent at this OID
[10-Sep-2009 08:38:33] * afink googles
[10-Sep-2009 08:40:19] straterr1 is now known as straterra
\n[10-Sep-2009 08:46:51] <ckrough> afink: odd, on my RHEL5 boxes I\nget that data, but I have not installed anything other that net-snmp.
\n[10-Sep-2009 08:46:56] <ckrough> afink: this may help as well\nhttp://www.zenoss.com/community/docs/howtos/configuration-of-netsnmp-for-use-with-zenoss/
[10-Sep-2009 09:14:13] <rmatte> yeh, net-snmp should be all you need for a Linux box
[10-Sep-2009 09:18:20] <Apocalipse> hello people...
[10-Sep-2009 09:18:36] <Apocalipse> my zenoss box stopped working...
[10-Sep-2009 09:18:50] <Apocalipse> i restarted services and machine...
[10-Sep-2009 09:18:54] <Apocalipse> and nothing
[10-Sep-2009 09:22:12] <rmatte> stopped working how?
[10-Sep-2009 09:22:25] <rmatte> when you start zenoss do you get any errors?
[10-Sep-2009 09:22:30] <rmatte> which daemons aren't starting?
[10-Sep-2009 09:22:31] <rmatte> etc...
[10-Sep-2009 09:25:26] <Apocalipse> zenhub, zenactions
[10-Sep-2009 09:26:16] <rmatte> what version of Zenoss are you running?
[10-Sep-2009 09:26:33] <Apocalipse> 2.4.1 in freebsd
[10-Sep-2009 09:26:41] <rmatte> oh right, FreeBSD lol
[10-Sep-2009 09:26:48] <Apocalipse> kkkkkk
[10-Sep-2009 09:26:57] <rmatte> you'll want to check $ZENHOME/log/zenhub.log for starters
[10-Sep-2009 09:27:07] <rmatte> look for any errors
[10-Sep-2009 09:27:10] <rmatte> same with zenactions.log
\n[10-Sep-2009 09:29:24] <Apocalipse> MySQLConnectionError: (2002,\n"Can't connect to local MySQL server through socket '/tmp/mysql.sock'\n(2)")
[10-Sep-2009 09:29:44] <rmatte> ok, so your mysql server isn't started
[10-Sep-2009 09:29:47] <rmatte> that's your issue
[10-Sep-2009 09:30:02] <rmatte> where did you put the zenoss-stack script?
[10-Sep-2009 09:30:11] <rmatte> or the zenoss startup script in general
[10-Sep-2009 09:30:31] <rmatte> /etc/rc.d?
[10-Sep-2009 09:30:34] <Apocalipse> mysql is already started...
[10-Sep-2009 09:30:46] <rmatte> ok, do a "zenhub -v10 start"
[10-Sep-2009 09:30:48] <rmatte> see what it says
[10-Sep-2009 09:32:01] <Apocalipse> zenhub {run|start|stop|restart|status|help|genconf|genxmlconfigs} [options]
[10-Sep-2009 09:32:03] <Apocalipse>
[10-Sep-2009 09:32:09] <rmatte> sorry
[10-Sep-2009 09:32:13] <rmatte> ok, do a "zenhub -v10 run"
[10-Sep-2009 09:32:24] <rmatte> no wait
[10-Sep-2009 09:32:28] <rmatte> start should have worked
[10-Sep-2009 09:32:32] <Apocalipse> same
[10-Sep-2009 09:32:33] <rmatte> are you the zenoss user?
[10-Sep-2009 09:32:37] <Apocalipse> yes
[10-Sep-2009 09:32:43] <rmatte> try "zenhub run -v10"
[10-Sep-2009 09:32:53] <rmatte> or "zenhub start -v10"
[10-Sep-2009 09:32:55] <Apocalipse> ./zenhub start
[10-Sep-2009 09:33:00] <Apocalipse> ok
[10-Sep-2009 09:33:07] <rmatte> you shouldn't need to do ./
[10-Sep-2009 09:33:25] <rmatte> If you do, then there's a problem with your path or environment variables
[10-Sep-2009 09:33:39] <Apocalipse> zenhub status program running; pid=2188
[10-Sep-2009 09:33:51] <rmatte> do a "zenoss status"
[10-Sep-2009 09:33:55] <rmatte> what's the full output?
[10-Sep-2009 09:34:22] <Apocalipse> only this is not working...Daemon: zensyslog not running
[10-Sep-2009 09:34:24] <rmatte> paste it slowly in pieces (3 or 4 lines at a time), otherwise you'll flood out
[10-Sep-2009 09:34:35] <rmatte> ok, so do: zensyslog start
[10-Sep-2009 09:34:37] <Apocalipse> Daemon: zeoctl program running; pid=745
[10-Sep-2009 09:34:37] <Apocalipse> Daemon: zopectl program running; pid=749
[10-Sep-2009 09:34:37] <Apocalipse> Daemon: zenhub program running; pid=2188
[10-Sep-2009 09:34:45] <Apocalipse> Daemon: zenjobs program running; pid=839
[10-Sep-2009 09:34:45] <Apocalipse> Daemon: zenping program running; pid=925
[10-Sep-2009 09:34:45] <Apocalipse> Daemon: zensyslog not running
[10-Sep-2009 09:34:52] <Apocalipse> Daemon: zenstatus program running; pid=967
[10-Sep-2009 09:34:53] <Apocalipse> Daemon: zenactions program running; pid=1645
[10-Sep-2009 09:34:53] <Apocalipse> Daemon: zentrap program running; pid=1079
[10-Sep-2009 09:35:00] <Apocalipse> Daemon: zenmodeler program running; pid=1100
[10-Sep-2009 09:35:00] <Apocalipse> Daemon: zenperfsnmp program running; pid=1137
[10-Sep-2009 09:35:00] <Apocalipse> Daemon: zencommand program running; pid=1171
[10-Sep-2009 09:35:07] <Apocalipse> Daemon: zenprocess program running; pid=1207
[10-Sep-2009 09:35:07] <Apocalipse> Daemon: zenwin program running; pid=1249
[10-Sep-2009 09:35:07] <Apocalipse> Daemon: zeneventlog program running; pid=1291
[10-Sep-2009 09:35:14] <rmatte> zensyslog start
[10-Sep-2009 09:36:24] <Apocalipse> not running
[10-Sep-2009 09:36:32] <Apocalipse> but this never run before
[10-Sep-2009 09:37:01] <rmatte> well, you kind of need it
[10-Sep-2009 09:37:14] <rmatte> if you ever expect to process syslogs
[10-Sep-2009 09:37:32] <rmatte> check $ZENHOME/log/zensyslog.log
[10-Sep-2009 09:37:40] <rmatte> for errors
[10-Sep-2009 09:38:57] <Apocalipse> 0 size of file
[10-Sep-2009 09:39:14] <rmatte> so the log isn't populated at all?
[10-Sep-2009 09:39:58] <Apocalipse> yes
[10-Sep-2009 09:40:18] <rmatte> ok, well, you're running on FreeBSD so lord only knows why that daemon won't start
[10-Sep-2009 09:40:28] <Apocalipse> yes
[10-Sep-2009 09:40:29] <rmatte> it's a pretty necessary daemon to have running though
[10-Sep-2009 09:40:36] <Apocalipse> i know...
[10-Sep-2009 09:40:42] <rmatte> if you do zensyslog run -v10
[10-Sep-2009 09:40:45] <rmatte> what do you see?
[10-Sep-2009 09:40:46] <Apocalipse> but all things were working...
\n[10-Sep-2009 09:41:10] <rmatte> all the things that you were\nusing were working, but if you ever wanted to send syslog messages to\nZenoss it would never work
[10-Sep-2009 09:41:22] <Apocalipse> zensyslog run -v10 ZenSocket Error: Unable to bind to listen port (error code 48)
[10-Sep-2009 09:41:41] <rmatte> do you have some sort of firewall on that system?
[10-Sep-2009 09:41:46] <Apocalipse> none
[10-Sep-2009 09:42:30] <rmatte> you're probably already running syslogd on there which is attached to that port already
[10-Sep-2009 09:42:41] <rmatte> or syslog-ng
[10-Sep-2009 09:42:54] <rmatte> are you running either of those?
[10-Sep-2009 09:43:45] <rmatte> looks like it's quite a common issue on FreeBSD
[10-Sep-2009 09:43:48] <rmatte> http://forums.zenoss.com/viewtopic.php?p=16030
[10-Sep-2009 09:43:52] <Apocalipse> syslog is running
[10-Sep-2009 09:43:58] <rmatte> ok, kill syslog
[10-Sep-2009 09:44:02] <rmatte> and make sure it doesn't run
[10-Sep-2009 09:44:07] <rmatte> then start zensyslog
[10-Sep-2009 09:45:26] <rmatte> ah
[10-Sep-2009 09:45:29] <rmatte> here are the steps...
\n[10-Sep-2009 09:45:36] <rmatte> syslog (port 514) will probably\nalso show in your output. You don't want to disable syslog completely,\nas you do want to receive logging messages. However, you don't need to\nhave this port open to do so. In your /etc/rc.conf file, make sure\nsyslog is enabled and add a second line with some options:
[10-Sep-2009 09:45:37] <Apocalipse> zensyslog run -v10
[10-Sep-2009 09:45:42] <rmatte> syslogd_enable="YES"
[10-Sep-2009 09:45:42] <rmatte> syslogd_flags="-ss"
\n[10-Sep-2009 09:45:49] <rmatte> Those two sses (make sure you\nhave two, not just one) in the flags will disable logging from remote\nhosts and close that port, but still allow your localhost to keep its\nlogging capabilities.
[10-Sep-2009 09:46:06] <rmatte> so yeh, I forgot that FreeBSD does it differently
[10-Sep-2009 09:46:11] <rmatte> so you don't want to completely kill it
[10-Sep-2009 09:46:13] <rmatte> just use those options
[10-Sep-2009 09:46:31] <Apocalipse> DEBUG:zen.ZenSyslog:Starting PBDaemon initialization
[10-Sep-2009 09:46:32] <Apocalipse> INFO:zen.ZenSyslog:Connecting to localhost:8789
[10-Sep-2009 09:46:32] <Apocalipse> DEBUG:zen.ZenSyslog:Logging in as admin
[10-Sep-2009 09:46:32] <Apocalipse> INFO:zen.ZenSyslog:Connected to ZenHub
[10-Sep-2009 09:46:32] <Apocalipse> DEBUG:zen.ZenSyslog:Setting up initial services: EventService
[10-Sep-2009 09:46:32] <Apocalipse> DEBUG:zen.ZenSyslog:Chaining getInitialServices with d2
[10-Sep-2009 09:46:34] <Apocalipse> DEBUG:zen.ZenSyslog:Loaded service EventService from zenhub
[10-Sep-2009 09:46:55] <Apocalipse> last 2 lines
[10-Sep-2009 09:47:00] <Apocalipse> INFO:zen.ZenSyslog:zensyslog shutting down
[10-Sep-2009 09:47:00] <Apocalipse> INFO:zen.ZenSyslog:0 events processed in 0.00 seconds
[10-Sep-2009 09:47:18] <rmatte> that's when you do a "run" right?
[10-Sep-2009 09:47:24] <Apocalipse> yes
[10-Sep-2009 09:47:28] <rmatte> now do a "start"
[10-Sep-2009 09:47:31] <Apocalipse> i did start
[10-Sep-2009 09:47:32] <rmatte> and it should start properly
[10-Sep-2009 09:47:40] <Apocalipse> program running; pid=3867
[10-Sep-2009 09:48:26] <rmatte> good
[10-Sep-2009 09:48:31] <rmatte> that's fixed...
[10-Sep-2009 09:48:38] <rmatte> did you follow the steps I pasted?
[10-Sep-2009 09:48:42] <Apocalipse> yes
[10-Sep-2009 09:48:47] <rmatte> you need syslog running, so you have to use those steps specifically
[10-Sep-2009 09:48:54] <rmatte> otherwise you won't get local system log messages
[10-Sep-2009 09:49:13] <rmatte> but you need to disable it from using the external port
[10-Sep-2009 09:49:47] <rmatte> so those 2 lines are in your /etc/rc.conf and syslog is running again?
[10-Sep-2009 09:50:10] <Apocalipse> yes
[10-Sep-2009 09:50:15] <rmatte> ok good
[10-Sep-2009 09:50:27] <rmatte> do a zenoss status and paste the output again please
[10-Sep-2009 09:52:15] <Apocalipse> Daemon: zeoctl program running; pid=2728
[10-Sep-2009 09:52:15] <Apocalipse> Daemon: zopectl program running; pid=2732
[10-Sep-2009 09:52:15] <Apocalipse> Daemon: zenhub program running; pid=2785
[10-Sep-2009 09:52:15] <Apocalipse> Daemon: zenjobs program running; pid=2822
[10-Sep-2009 09:52:15] <Apocalipse> Daemon: zenping program running; pid=2907
[10-Sep-2009 09:52:16] <Apocalipse> Daemon: zensyslog program running; pid=3867
[10-Sep-2009 09:52:18] <Apocalipse> Daemon: zenstatus program running; pid=2942
[10-Sep-2009 09:52:24] <Apocalipse> Daemon: zenactions program running; pid=2980
[10-Sep-2009 09:52:25] <Apocalipse> Daemon: zentrap program running; pid=3070
[10-Sep-2009 09:52:25] <Apocalipse> Daemon: zenmodeler program running; pid=3075
[10-Sep-2009 09:52:26] <Apocalipse> Daemon: zenperfsnmp program running; pid=3108
[10-Sep-2009 09:52:28] <Apocalipse> Daemon: zencommand program running; pid=3141
[10-Sep-2009 09:52:30] <Apocalipse> Daemon: zenprocess program running; pid=3174
[10-Sep-2009 09:52:32] <Apocalipse> Daemon: zenwin program running; pid=3216
[10-Sep-2009 09:52:34] <Apocalipse> Daemon: zeneventlog program running; pid=3258
[10-Sep-2009 09:52:41] <rmatte> good stuff, ok, now what is the issue at this point?
[10-Sep-2009 09:52:47] <rmatte> are you not able to hit the web interface?
[10-Sep-2009 09:53:08] <Apocalipse> web interface is ok...
[10-Sep-2009 09:53:37] <Apocalipse> ok...
[10-Sep-2009 09:53:45] <Apocalipse> all things restarted
[10-Sep-2009 09:54:00] <Apocalipse> thx "master" ryan
[10-Sep-2009 09:54:01] <Apocalipse>
[10-Sep-2009 09:54:15] <Apocalipse> brb
[10-Sep-2009 09:54:27] <rmatte> np
\n[10-Sep-2009 09:57:42] <rmatte> If I were you I'd reboot the\nserver and make sure that it works after the reboot without having to\nstart anything manually
[10-Sep-2009 10:00:00] <rmatte> you may also want to do "zenpatch 14181" to patch a bad security bug
[10-Sep-2009 10:00:13] <rmatte> then restart zenoss
[10-Sep-2009 10:09:08] <vinzpr> hi anyone here knows how to Display CPU Threshold Events as a Percentage
[10-Sep-2009 10:09:20] <vinzpr> also in memory and interface
[10-Sep-2009 10:09:49] <vinzpr> i tried this procedure but it didnt work
[10-Sep-2009 10:09:50] <vinzpr> http://www.zenoss.com/Members/cluther/display-memory-threshold-events-as-a-percentage/
[10-Sep-2009 10:16:42] <rmatte> vinzpr: well, aren't CPU datapoints usually just a value between 1 and 100 anyways?
[10-Sep-2009 10:18:42] <vinzpr> yes but it only shows me the value of memory bytes
[10-Sep-2009 10:18:42] <vinzpr> not the percentage
[10-Sep-2009 10:18:42] <rmatte> you just said CPU Threshold
[10-Sep-2009 10:18:42] <rmatte> not Memory
[10-Sep-2009 10:18:51] <vinzpr> example: threshold of Memory Warning notmet: current value 7124.00
[10-Sep-2009 10:18:55] <vinzpr> also in cpu
[10-Sep-2009 10:19:07] <rmatte> what's CPU look like then?
[10-Sep-2009 10:19:07] <vinzpr> 617 instead of 6.17
[10-Sep-2009 10:19:17] <rmatte> ah
[10-Sep-2009 10:19:27] <rmatte> it's probably doing it out of 10000 then or something
[10-Sep-2009 10:19:38] <rmatte> what type of device is this for?
[10-Sep-2009 10:20:14] <vinzpr> linux
[10-Sep-2009 10:21:15] <rmatte> so cluther's transform doesn't work for Linux?
[10-Sep-2009 10:21:20] <vinzpr> yes
[10-Sep-2009 10:21:32] <rmatte> did you modify the match line to match what you're seeing?
[10-Sep-2009 10:21:40] <rmatte> It should be more like:
[10-Sep-2009 10:21:41] <rmatte> match = re.search('threshold of Free Memory [^:]+: current value ([\\d\\.]+)', evt.message)
[10-Sep-2009 10:21:43] <rmatte> oops
[10-Sep-2009 10:21:44] <rmatte> one sec
[10-Sep-2009 10:22:05] <rmatte> match = re.search('threshold of Memory Warning notmet [^:]+: current value ([\\d\\.]+)', evt.message)
[10-Sep-2009 10:22:12] <rmatte> in your case
[10-Sep-2009 10:22:25] <vinzpr> ah
[10-Sep-2009 10:22:35] <rmatte> otherwise the transform will never kick off
[10-Sep-2009 10:22:37] <vinzpr> ok you have to search the string of alert
[10-Sep-2009 10:22:46] <vinzpr> ok..let me check that
[10-Sep-2009 10:22:51] <rmatte> as for the CPU, that may be a bit more difficult...
[10-Sep-2009 10:22:54] <vinzpr> how about in cpu
[10-Sep-2009 10:23:48] <rmatte> just need to figure out how to extract the CPU value...
[10-Sep-2009 10:25:00] <vinzpr> ok
[10-Sep-2009 10:25:06] <rmatte> I'm not sure if avail = float(match.groups()[0]) * 1024
[10-Sep-2009 10:25:16] <rmatte> is actually pulling from the value in the threshold
[10-Sep-2009 10:25:21] <vinzpr> how about getting like this device.hw.totalMemory
[10-Sep-2009 10:25:32] <rmatte> no, you can't do that
[10-Sep-2009 10:25:36] <vinzpr> i dont know on device.hw.totalcpu?
[10-Sep-2009 10:25:43] <rmatte> that's pulling the total amount of memory on the box, which is a concrete value
[10-Sep-2009 10:25:47] <vinzpr> where can i get that list
[10-Sep-2009 10:25:51] <rmatte> the avail line is what we need to work with
[10-Sep-2009 10:25:56] <vinzpr> ok
[10-Sep-2009 10:26:18] <rmatte> I'll try to come up with something...
[10-Sep-2009 10:26:25] <vinzpr> thanks
[10-Sep-2009 10:27:01] <rmatte> give me a full example of what one of your CPU thresholds would look like as an event
[10-Sep-2009 10:27:05] <rmatte> I need to adjust the match line
[10-Sep-2009 10:27:16] <vinzpr> ok please wait
[10-Sep-2009 10:28:00]\n<adelbul> hello, il have a script on a opensolaris system which\nverify the status of an LSI Perc RAID command. I try to run this script\nwith zencommand but i believe that i don't understand the whole concept.
[10-Sep-2009 10:28:04] <vinzpr> threshold of Load Warning exceeded: current value 626.00
[10-Sep-2009 10:28:13] <vinzpr> it supposed to be 6.26
[10-Sep-2009 10:28:24] <adelbul> I create a datasource named : "LsiPerc6iCheckOpenSolaris"
[10-Sep-2009 10:28:36] <rmatte> and 626 would be 6.26%?
[10-Sep-2009 10:28:44] <rmatte> or 62.6%?
[10-Sep-2009 10:28:57] <rmatte> oh wait, is that for load average?
[10-Sep-2009 10:29:00] <adelbul> this datasource is checking the status of a Raid Controller in the server.
[10-Sep-2009 10:29:24] <adelbul> this datasource call a script named "/export/home/zenoss/scripts/check_perc.sh"
[10-Sep-2009 10:29:40] <adelbul> this script is on the client side on the opensolaris server.
[10-Sep-2009 10:29:50] <adelbul> and it is the result of the zentestcommand :
\n[10-Sep-2009 10:30:12] <adelbul> [zenoss@srvsurvey ~]$\nzentestcommand -d srvnas03\n--datasource=LsiPerc6iCheckOpenSolaris \n/bin/sh: /export/home/zenoss/scripts/check_perc.sh: no such file or\ndirectory
[10-Sep-2009 10:30:32] <vinzpr> its for load average
[10-Sep-2009 10:30:37] <rmatte> adelbul: it needs to be an SSH command based datasource
[10-Sep-2009 10:30:43] <rmatte> not a regular command based datasource
[10-Sep-2009 10:31:09] <adelbul> rmatte: yes you are right. then i enabled SSH in the datasource.
[10-Sep-2009 10:31:22] <vinzpr> rmatte: is for load average
[10-Sep-2009 10:31:25] <adelbul> rmatte: but the result is the same.
\n[10-Sep-2009 10:31:36] <rmatte> zCommandUsername and\nzCommandPassword in zProperties need to be populated with a valid\nusername and password for the device
[10-Sep-2009 10:31:49] <rmatte> vinzpr: k
[10-Sep-2009 10:33:11] <rmatte> vinzpr: try...
[10-Sep-2009 10:33:12] <rmatte> import re
\n[10-Sep-2009 10:33:12] <rmatte> match = re.search('threshold of\nLoad Warning exceeded [^:]+: current value ([\\d\\.]+)', evt.message)
[10-Sep-2009 10:33:12] <rmatte> if match and device:
[10-Sep-2009 10:33:12] <rmatte> load = float(match.groups()[0])
[10-Sep-2009 10:33:12] <rmatte> load_used = load / 100
\n[10-Sep-2009 10:33:13] <rmatte> evt.summary = "Load Average\nthreshold exceeded: current value %3.1f%%" % (percent_used)
[10-Sep-2009 10:33:26] <rmatte> hopefully it'll work
[10-Sep-2009 10:34:44] <rmatte> oops wait
[10-Sep-2009 10:34:55] <rmatte> replace (percent_used) at the end with (load_used)
[10-Sep-2009 10:34:58] <rmatte> I forgot to do that
[10-Sep-2009 10:35:47] <vinzpr> ok thank you..let me try that
[10-Sep-2009 10:36:59] <jb> anybody here user 64bit counters?
[10-Sep-2009 10:38:51] <rmatte> adelbul: the script should also output somewhat like: OK|LsiPerc6iCheckOpenSolaris=value
[10-Sep-2009 10:39:41] <rmatte> adelbul: and it should output "Unknown" if it's not able to check
[10-Sep-2009 10:39:53] <rmatte> jb: not really, why, what's up?
[10-Sep-2009 10:40:08] <adelbul> rmatte: return 0 or 1 is not good ?
[10-Sep-2009 10:40:08] <skipzoid> hello all
[10-Sep-2009 10:40:21] <jb> rmatte: well, I am trying to graph some COUNTER64 values
[10-Sep-2009 10:40:32] <jb> and just wondering if I need to do anythin special
[10-Sep-2009 10:40:36] <rmatte> nope
[10-Sep-2009 10:40:53] <rmatte> treat it the same way as you would for 32bit counters
\n[10-Sep-2009 10:41:38] <skipzoid> can someone help a newbie\nunderstand how a MIB file works inconjunction with a monitored device?\n- do I have to tell the device about the mib file or something ?
[10-Sep-2009 10:41:59] <rmatte> skipzoid: Mib files are only used to translate SNMP traps in Zenoss
\n[10-Sep-2009 10:42:20] <rmatte> skipzoid: when you're designing a\nmonitoring template to monitor a device with you need to specify the\nOIDs that you want to poll
[10-Sep-2009 10:42:36] <rmatte> skipzoid: Mibs are not required for actually polling data from devices in Zenoss
\n[10-Sep-2009 10:43:11] <skipzoid> i'm very new to this, so the\ndata is polled for? I thought the device sent it as and when an event\nhappend?
[10-Sep-2009 10:43:24] <jb> ah i think it was cause I was using V1
\n[10-Sep-2009 10:43:49] <rmatte> skipzoid: some data is polled\nfor, some data is sent from the device to Zenoss when an event happens\nwhich is called an SNMP traps
[10-Sep-2009 10:44:14]\n<rmatte> skipzoid: you don't need to actually tell Zenoss about\nMibs, provided you load the correct Mibs for the devices you are\nmonitoring they will automatically translate the traps that come in
\n[10-Sep-2009 10:44:38] <rmatte> skipzoid: examples of data that\nis polled for is performance data, CPU Usage, Memory Usage, Interface\ntraffic/errors, stuff like that
[10-Sep-2009 10:45:25] <rmatte> skipzoid: disk usage is another example of something that is polled for
\n[10-Sep-2009 10:45:34] <skipzoid> i think i understand that, i'm\ntrying to setup a monitor on a device that is duel power supply and\nseveral fans and when they fail I already know it sents a trap to a\nconfigured destination, i'm just trying to get zenoss to receive and\nact by way of sending a text message to call for help
[10-Sep-2009 10:46:07] <rmatte> what types of devices?
[10-Sep-2009 10:46:10] <rmatte> like, what brand?
\n[10-Sep-2009 10:47:32] <skipzoid> its a video server - propriety\nto harris, but they do supply a mib file for the device to use with\nyour own monitoring if you deside not to buy their monitoring system
[10-Sep-2009 10:48:47] <rmatte> ok, so just load that Mib file and you're done
[10-Sep-2009 10:48:47] <skipzoid> its that easy ?
[10-Sep-2009 10:48:47] <rmatte> it'll translate the traps
[10-Sep-2009 10:48:47] <rmatte> well, sort of
\n[10-Sep-2009 10:48:47] <rmatte> it'll translate the traps, but\nthen you need to write python scripts for each trap called transforms
[10-Sep-2009 10:48:48] <rmatte> they'll ensure that when the traps come in the message is human readable
\n[10-Sep-2009 10:49:11] <rmatte> you don't necessarily need to\nwrite transforms, it's really up to you, but you won't have to login to\nZenoss to dig for trap details since you'll be able to script Zenoss to\nput the details right in to the message
[10-Sep-2009\n10:49:32] <skipzoid> ive looked at the mib and there is already\nhuman readable errors - have i missed the point is that not what a mib\nfile does? translates the numbers into human form?
[10-Sep-2009 10:49:41] kevin7kal_ is now known as kevin7kal
[10-Sep-2009 10:49:56] <rmatte> that's correct, it does... but what will happen is this...
\n[10-Sep-2009 10:50:56] <rmatte> the trap will come in, and Zenoss\nwill create an event called something like "snmp trap\nclogMessageGenerated" or whatever
[10-Sep-2009 10:51:14] <rmatte> and if you click on the icon next to the event a details window will open
\n[10-Sep-2009 10:51:26] <rmatte> and if you click on the details\ntab in that window, you'll see what values the trap translated
\n[10-Sep-2009 10:51:41] <rmatte> now, to actually get those values\nin to actualy event messages when that sort of trap comes in you need\nto do a transform
[10-Sep-2009 10:51:56] <rmatte> I'll give you an example...
[10-Sep-2009 10:52:08]\n<adelbul> rmatte : in which section of the documentation i can\nfind the output formatting like this one :\nOK|LsiPerc6iCheckOpenSolaris=value ?
[10-Sep-2009\n10:52:17] <rmatte> in the case of the clogMessageGenerated trap,\nthat's a trap that's sent from a Cisco device for critical log\nmessages...
[10-Sep-2009 10:52:30] <rmatte> it originally came in as "snmp trap clogMessageGenerated"
\n[10-Sep-2009 10:52:44] <rmatte> when I went to the details tab I\nsaw different values (parts of the message) with the following names...
[10-Sep-2009 10:53:10] <rmatte> clogHistFacility, clogHistMsgName, clogHistMsgText, and clogHistSeverity
[10-Sep-2009 10:53:13] <vinzpr> rmatte: i tried it but the output is the same-
[10-Sep-2009 10:53:14] <vinzpr> threshold of Memory Warning not met: current value 97120.00
[10-Sep-2009 10:53:22] <rmatte> so clogHistSeverity was a number representing the severity of the event...
[10-Sep-2009 10:53:32] <rmatte> clogHisMsgText was the message
[10-Sep-2009 10:53:34] <rmatte> etc...
[10-Sep-2009 10:53:45] <skipzoid> i think i'm begining to understand now
\n[10-Sep-2009 10:53:59] <rmatte> I created a transform to make the\nmessages come in with the details in the summary instead of it just\nlooking like "snmp trap blah"
[10-Sep-2009 10:54:03] <rmatte> here's the transform I used
[10-Sep-2009 10:54:08] <vinzpr> import re
[10-Sep-2009 10:54:08] <vinzpr> match = re.search('threshold of Memory Warning not met [^:]+: current value ([\\d\\.]+)', evt.message)
[10-Sep-2009 10:54:08] <vinzpr> if match and device:
[10-Sep-2009 10:54:08] <vinzpr> avail = float(match.groups()[0]) * 1024
[10-Sep-2009 10:54:08] <vinzpr> total = device.hw.totalMemory
[10-Sep-2009 10:54:09] <vinzpr> percent_used = (1 - (avail / total)) * 100
[10-Sep-2009 10:54:11] <vinzpr> evt.summary = "Memory threshold exceeded: current value %3.1f%%" % (percent_used)
[10-Sep-2009 10:54:13] <rmatte> facility = getattr(evt, 'clogHistFacility', 'Unknown')
[10-Sep-2009 10:54:13] <rmatte> msgname = getattr(evt, 'clogHistMsgName', 'Unknown')
[10-Sep-2009 10:54:13] <rmatte> msgtext = getattr(evt, 'clogHistMsgText', 'Unknown')
[10-Sep-2009 10:54:13] <rmatte> histsev = getattr(evt, 'clogHistSeverity', 'Unknown')
[10-Sep-2009 10:54:13] <rmatte> import re
[10-Sep-2009 10:54:14] <rmatte> if re.match("decrypt: mac verify failed", msgtext):
[10-Sep-2009 10:54:16] <rmatte> evt._action = "drop"
[10-Sep-2009 10:54:18] <rmatte> elif re.match("Interface FastEthernet", msgtext):
[10-Sep-2009 10:54:20] <rmatte> evt.severity = 0
[10-Sep-2009 10:54:22] <rmatte> else:
[10-Sep-2009 10:54:24] <rmatte> evt.severity = {0:5, 1:5, 2:5, 3:4, 4:4, 5:3, 6:3, 7:0}[histsev]
[10-Sep-2009 10:54:26] <rmatte> evt.component = 'SYSLOG-%s' % (histsev)
[10-Sep-2009 10:54:28] <rmatte> evt.summary = '%s-%s: %s' % (facility,msgname,msgtext)
[10-Sep-2009 10:54:31] <rmatte> now, I know that's a bit complex, but to break it down to simpler form...
[10-Sep-2009 10:54:36] <rmatte> histsev = getattr(evt, 'clogHistSeverity', 'Unknown')
\n[10-Sep-2009 10:54:54] <rmatte> that line grabs the value that\ncorresponded to clogHistSeverity and stores it in a variable called\nhistsev
[10-Sep-2009 10:55:11] <rmatte> then if\nyou want to set the event summary to something custom with the value\nincluded you'd do something like...
[10-Sep-2009 10:55:39] <rmatte> evt.summary = 'This is the value for histsev: %s' % (histsev)
[10-Sep-2009 10:55:48] <rmatte> if you wanted to do multiple, you could do...
[10-Sep-2009 10:56:08] <rmatte> evt.summary = 'This is the value for histsev: %s and this is the value for msgtext: %s' % (histsev, msgtext)
\n[10-Sep-2009 10:56:29] <rmatte> you can also change other aspects\nof the event by setting things like evt.severity, evt.message, etc...
[10-Sep-2009 10:56:32] <rmatte> understand?
[10-Sep-2009 10:56:58] <rmatte> vinzpr: hmmm
[10-Sep-2009 10:56:59] <skipzoid> its becoming clearer - i think i need to trigger some events and then try this,
[10-Sep-2009 10:57:33] <rmatte> skipzoid: you should download the Zenoss Admin Guide and read it
[10-Sep-2009 10:57:34] <rmatte> all of it
[10-Sep-2009 10:57:41] <rmatte> you'll learn a lot of the basics from it
[10-Sep-2009 10:58:17] <skipzoid> i have and printed it for betime reading
[10-Sep-2009 10:58:17] <rmatte> cool
[10-Sep-2009 10:58:49] <Troubadix09> skipzoid: and read the community FAQ, greatful help at the beginning
[10-Sep-2009 10:59:45] <rmatte> vinzpr: what event class is the threshold coming in as?
\n[10-Sep-2009 10:59:47] <skipzoid> will do - this has been most\nhelpful too - a lot of irc channels dont like newbies and often say\nunhelpful things like - google is your friend or its in the manual...\nrefreshing not to get that here
[10-Sep-2009 11:00:04] <rmatte> skipzoid: nah, we help as best we can here
[10-Sep-2009 11:00:12] <vinzpr> rmatte:/Events /Perf /Memory
[10-Sep-2009 11:00:13] <skipzoid> appreciated
[10-Sep-2009 11:00:17] <rmatte>
[10-Sep-2009 11:00:20] <rmatte> vinzpr: ok...
[10-Sep-2009 11:00:24] <Troubadix09> skipzoid: and install the ZenPacks.community.mib_utils-Zenpack, it help to interpret or search in MIBs
[10-Sep-2009 11:01:25] <skipzoid> just looking for it now
[10-Sep-2009 11:02:15] <rmatte> vinzpr: try match = re.search('threshold of Memory Warning [^:]+: current value ([\\d\\.]+)', evt.summary)
[10-Sep-2009 11:02:18] <Apocalipse> hey...people...i cannot see any device in "network map"...
[10-Sep-2009 11:02:37] <rmatte> vinzpr: shortening the message a bit and checking evt.summary instead of evt.message just in case
[10-Sep-2009 11:03:05] <rmatte> Apocalipse: you need to enter a starting point in the top left (an IP or a hostname)
[10-Sep-2009 11:03:31] <rmatte> Apocalipse: A default starting point can be specified under the settings for your username as well.
[10-Sep-2009 11:03:51] <jb> hm
[10-Sep-2009 11:04:00] <jb> om.size = "%d" % (getattr(om, 'size', 0) / 1024
[10-Sep-2009 11:04:06] <jb> you see anythin wrong with that?
[10-Sep-2009 11:04:16] <rmatte> yeh, you're missing a close bracket
[10-Sep-2009 11:04:18] <rmatte> at the end
[10-Sep-2009 11:04:54] <Apocalipse> x.x.x.x/yy
[10-Sep-2009 11:04:55]\n<vinzpr> rmatte: like this match = re.search('threshold of Memory\nWarning not met [^:]+: current value ([\\d\\.]+)', evt.summary)
[10-Sep-2009 11:04:56] <jb> oh
[10-Sep-2009 11:04:57] <jb> oops.
[10-Sep-2009 11:05:15] <rmatte> vinzpr: take the not met part out just for the heck of it
[10-Sep-2009 11:05:29] <vinzpr> ok let me do it
[10-Sep-2009 11:05:35] <rmatte> match = re.search('threshold of Memory Warning [^:]+: current value ([\\d\\.]+)', evt.summary)
[10-Sep-2009 11:07:08]\n<rmatte> well, looks like my project for the rest of the day is\nto write a custom device availability report in python (standalone\nscript)
[10-Sep-2009 11:07:14] <vinzpr> rmatte: threshold of Memory Critical not met: current value 33356.00
[10-Sep-2009 11:07:27] <rmatte> vinzpr: oh, well there's your problem
[10-Sep-2009 11:07:36] <rmatte> Memory Critical, not Memory Warning lol
[10-Sep-2009 11:07:43] <rmatte> match = re.search('threshold of Memory [^:]+: current value ([\\d\\.]+)', evt.summary)
[10-Sep-2009 11:07:44] <rmatte> use that
[10-Sep-2009 11:08:03] <rmatte> I just assumed that you only had the one threshold
[10-Sep-2009 11:09:23] <vinzpr> ok
[10-Sep-2009 11:09:28] <rmatte> hopefully it works this time
[10-Sep-2009 11:09:36] <rmatte> if it doesn't, I'm out of ideas lol
[10-Sep-2009 11:10:18] <vinzpr> hehe
[10-Sep-2009 11:10:57] <rmatte> well not 100% out of ideas, there are some debugging steps we can use
[10-Sep-2009 11:11:04] <vinzpr> cool..its working now...Memory threshold exceeded: current value 99.1%
[10-Sep-2009 11:11:08] <rmatte> perfect
[10-Sep-2009 11:11:10] <vinzpr> how about the cpu load
[10-Sep-2009 11:11:18] <rmatte> have you tested the cpu load one?
[10-Sep-2009 11:11:24] <vinzpr> what does this means [^:]+
[10-Sep-2009 11:11:31] <vinzpr> yep its not working also
[10-Sep-2009 11:11:40] <rmatte> it's basically a regex wildcard
[10-Sep-2009 11:12:02] <rmatte> ok, paste the load average script again along with an example of a threshold
[10-Sep-2009 11:12:10] <rmatte> also..., make sure you put the transform under the right event class
[10-Sep-2009 11:13:16] <vinzpr> threshold of Load Warningexceeded: current value 710.00
[10-Sep-2009 11:13:31] <vinzpr> import re
[10-Sep-2009 11:13:31] <vinzpr> match = re.search('threshold of Load [^:]+: current value ([\\d\\.]+)', evt.message)
[10-Sep-2009 11:13:31] <vinzpr> if match and device:
[10-Sep-2009 11:13:31] <vinzpr> load = float(match.groups()[0])
[10-Sep-2009 11:13:31] <vinzpr> load_used = load / 100
[10-Sep-2009 11:13:32] <vinzpr> evt.summary = "Load Average threshold exceeded: current value %3.1f%%" (load_used)
[10-Sep-2009 11:14:00] <rmatte> vinzpr: ummm, you don't have any indentation?
[10-Sep-2009 11:14:20] <rmatte> it's not going to work without proper spacing
[10-Sep-2009 11:15:01] <rmatte> also change evt.message to evt.summary at the end of the re.search line
[10-Sep-2009 11:15:42] <vinzpr> ok let me change it
[10-Sep-2009 11:15:57] <rmatte> k
[10-Sep-2009 11:16:14] <rmatte> change the spacing to somewhat match the other transform
[10-Sep-2009 11:17:42] <rmatte> import re
[10-Sep-2009 11:17:43] <rmatte> match = re.search('threshold of Load [^:]+: current value ([\\d\\.]+)', evt.summary)
[10-Sep-2009 11:17:43] <rmatte> if match and device:
[10-Sep-2009 11:17:43] <rmatte> load = float(match.groups()[0])
[10-Sep-2009 11:17:43] <rmatte> load_used = load / 100
[10-Sep-2009 11:17:43] <rmatte> evt.summary = "Load Average threshold exceeded: current value %3.1f%%" (load_used)
[10-Sep-2009 11:17:48] <rmatte> that's how it should look
[10-Sep-2009 11:21:20] <vinzpr> ok already change it let me wait
[10-Sep-2009 11:21:25] <vinzpr> how about this
[10-Sep-2009 11:21:26] <vinzpr> Threshold of Memory Critical restored: current value: 98240.00
[10-Sep-2009 11:21:54] <rmatte> is it actually outputting with a capital T?
[10-Sep-2009 11:22:01] <rmatte> hold on, I have a fix for that
[10-Sep-2009 11:22:42] <vinzpr> ok
\n[10-Sep-2009 11:23:12] <rmatte> match = re.search('threshold of\nMemory Warning [^:]+: current value ([\\d\\.]+)',\nevt.summary,re.IGNORECASE)
[10-Sep-2009 11:23:15] <rmatte> try that
[10-Sep-2009 11:23:17] <rmatte> oops
[10-Sep-2009 11:23:18] <rmatte> wait
\n[10-Sep-2009 11:23:24] <rmatte> match = re.search('threshold of\nMemory [^:]+: current value ([\\d\\.]+)', evt.summary,re.IGNORECASE)
[10-Sep-2009 11:23:25] <rmatte> that
[10-Sep-2009 11:23:36] <vinzpr> cool
\n[10-Sep-2009 11:23:42] <rmatte> match = re.search('threshold of\nLoad [^:]+: current value ([\\d\\.]+)', evt.summary,re.IGNORECASE)
[10-Sep-2009 11:23:43] <rmatte> as well
[10-Sep-2009 11:23:46] <rmatte> for the other one
\n[10-Sep-2009 11:24:14] <rmatte> I'm actually going to log a bug\nreport for that, the case shouldn't be the same between threshold alert\nmessages and threshold recovered messages
[10-Sep-2009 11:24:26] <rmatte> or it SHOULD be the same rather
[10-Sep-2009 11:25:27] <vinzpr> rockin
[10-Sep-2009 11:25:43] <rmatte> let me know if it works
[10-Sep-2009 11:28:59] <vinzpr> ok
\n[10-Sep-2009 11:28:59] <q7joey> is there supposed to be a way to\nallow the binding of multiple templates to a specific interface? the\nbind template option at the interface level actually binds to the\ndevice, not the interface.
[10-Sep-2009 11:30:14] <rmatte> man, I love linux, so easy to make iso files: dd if=/dev/cdrom of=whatever.iso
[10-Sep-2009 11:30:35] <rmatte> q7joey: no, only 1 template per interface
\n[10-Sep-2009 11:30:47] <rmatte> q7joey: and the interface\ntemplates are bound automatically based on the name of the template
\n[10-Sep-2009 11:31:35] <q7joey> it probably would be good to\nremove the bind template option from the interface area. i know about\nthe automatic choice. for now i'll make a local copy and mangle it for\nmy testing. i'm still hunting down issues with gaps in graphs due to\ncounter vs derive issues.
[10-Sep-2009 11:31:39]\n<rmatte> q7joey: so if the interface type (you can see the type\nby clicking on the interface), is ds1, and an interface template called\n"ds1" exists under /Devices, then it'll automatically use that
[10-Sep-2009 11:31:47] <rmatte> I see
[10-Sep-2009 11:32:00] <q7joey> i actually wanted to capture two sets of data for one interface, so a temp copy is what i'll do for now.
[10-Sep-2009 11:32:00] <rmatte> I didn't even know there was a bind template option there
[10-Sep-2009 11:32:12] <rmatte> I would imagine you can probably bind multiple templates to an interface
[10-Sep-2009 11:32:17] <rmatte> no idea why you'd want to though
[10-Sep-2009 11:32:30] <rmatte> the automated method is the proper one
[10-Sep-2009 11:32:36] <q7joey> yep, that's why i was asking since the option was there, but doesn't do what i'd expect.
[10-Sep-2009 11:32:43] <rmatte> I see
[10-Sep-2009 11:33:18] <q7joey> i'm going to make two graphs to see how different counter vs derive is.
[10-Sep-2009 11:33:50] <vinzpr> rmatte: the change works but the critical values still post on the clear the same value
[10-Sep-2009 11:34:43] <rmatte> vinzpr: oh?
[10-Sep-2009 11:34:54] <vinzpr> yep
[10-Sep-2009 11:34:57] <rmatte> vinzpr: try putting the operation in to brackets
[10-Sep-2009 11:35:02] <rmatte> (load / 100)
[10-Sep-2009 11:35:16] <vinzpr> thats on memory
[10-Sep-2009 11:35:34] <rmatte> oh, on clear they don't post a percentage?
[10-Sep-2009 11:35:46] <rmatte> or what?
[10-Sep-2009 11:35:55] <rmatte> give me an example
[10-Sep-2009 11:35:59] <vinzpr> they post percentage
[10-Sep-2009 11:36:16] <vinzpr> on clear but the value percentage on critical are the same when clear
[10-Sep-2009 11:36:28] <vinzpr> it supposed to show you the current value
[10-Sep-2009 11:36:50] <rmatte> yeh, hmmm
[10-Sep-2009 11:37:06] <rmatte> paste an example
[10-Sep-2009 11:37:22] <rmatte> and the transform, please
[10-Sep-2009 11:37:49] <vinzpr> CLEAR: Memory thresholdexceeded: current value 99.1%
[10-Sep-2009 11:38:10] <vinzpr> import re
\n[10-Sep-2009 11:38:10] <vinzpr> match = re.search('threshold of\nMemory [^:]+: current value ([\\d\\.]+)', evt.summary,re.IGNORECASE)
[10-Sep-2009 11:38:10] <vinzpr> if match and device:
[10-Sep-2009 11:38:10] <vinzpr> avail = float(match.groups()[0]) * 1024
[10-Sep-2009 11:38:10] <vinzpr> total = device.hw.totalMemory
[10-Sep-2009 11:38:10] <vinzpr> percent_used = (1 - (avail / total)) * 100
[10-Sep-2009 11:38:11] <vinzpr> evt.summary = "Memory threshold exceeded: current value %3.1f%%" % (percent_used)
[10-Sep-2009 11:38:13] <vinzpr> also the load didnt work
[10-Sep-2009 11:38:21] <rmatte> does it actually look like that? "thresholdexceeded"
[10-Sep-2009 11:38:39] <vinzpr> threshold of Load Warning exceeded: current value 983.00
[10-Sep-2009 11:38:52] <rmatte> hmmm
[10-Sep-2009 11:40:01] <vinzpr> Threshold of Memory Critical restored: current value: 112652.00
\n[10-Sep-2009 11:40:31] <rmatte> vinzpr: let's say you completely\nremove the transforms, and check what memory alerts and clear events\nlook like
[10-Sep-2009 11:40:40] <rmatte> vinzpr: if you do that, are the values the same or different?
[10-Sep-2009 11:40:48] <vinzpr> its different
[10-Sep-2009 11:41:01] <vinzpr> it shows you the current value
\n[10-Sep-2009 11:41:25] <rmatte> well, looking at the transform\ncode there's really no possible reason why it would be showing the old\nvalue
[10-Sep-2009 11:42:18] <rmatte> each time an event comes in it should be going through that entire transform
[10-Sep-2009 11:42:28] <rmatte> can you paste examples of a plain alert and plain clear?
[10-Sep-2009 11:43:14] <vinzpr> Event: 'Memory threshold exceeded: current value 99.1%'
[10-Sep-2009 11:43:15] <vinzpr> Cleared by: ''
[10-Sep-2009 11:43:15] <vinzpr> At:
[10-Sep-2009 11:43:15] <vinzpr> Device:
[10-Sep-2009 11:43:15] <vinzpr> Component:
[10-Sep-2009 11:43:15] <vinzpr> Severity: Critical
[10-Sep-2009 11:43:17] <vinzpr> Message:
[10-Sep-2009 11:43:19] <vinzpr> threshold of Memory Critical not met: current value 34116.00
[10-Sep-2009 11:43:21] <vinzpr> Undelete
[10-Sep-2009 11:44:19] <vinzpr> Device:
[10-Sep-2009 11:44:19] <vinzpr> Component:
[10-Sep-2009 11:44:19] <vinzpr> Severity: Warning
[10-Sep-2009 11:44:19] <vinzpr> Time: 2009/09/10 16:37:07.000
[10-Sep-2009 11:44:19] <vinzpr> Message:
[10-Sep-2009 11:44:19] <vinzpr> threshold of Memory Warning not met: current value 47352.00
[10-Sep-2009 11:44:21] <vinzpr> Event Detail
[10-Sep-2009 11:44:23] <vinzpr> Acknowledge
[10-Sep-2009 11:44:25] <vinzpr> Delete
[10-Sep-2009 11:44:27] <vinzpr> Device Events
[10-Sep-2009 11:45:17] <rmatte> I just want to see the difference in summary messages
[10-Sep-2009 11:45:43] <vinzpr> ok..
[10-Sep-2009 11:46:23] <vinzpr> summary Memory threshold exceeded: current value 99.2%
[10-Sep-2009 11:46:23] <vinzpr> message threshold of Memory Critical not met: current value 33324.00
[10-Sep-2009 11:47:09] <rmatte> and for the clear?
[10-Sep-2009 11:49:17] <rmatte> Hey Matt
[10-Sep-2009 11:49:41] <rmatte> busy morning, I've been troubleshooting up a storm in here lol
[10-Sep-2009 11:49:46] <vinzpr> CLEAR: Memory thresholdexceeded: current value 99.2%
[10-Sep-2009 11:49:46] <vinzpr> Event: 'Memory threshold exceeded: current value 99.2%'
[10-Sep-2009 11:50:19] <rmatte> ok, I'm really curious as to why it's showing "thresholdexceeded"
[10-Sep-2009 11:50:45] <rmatte> vinzpr: what does the event message look like on the clear event?
[10-Sep-2009 11:51:13] <rmatte> vinzpr: I'm more interested in the raw message than the transformed message right now
[10-Sep-2009 11:52:20] <vinzpr> on clear message
[10-Sep-2009 11:52:22] <vinzpr> Event: 'Memory threshold exceeded: current value 99.2%'
[10-Sep-2009 11:52:22] <vinzpr> Cleared by: ''
[10-Sep-2009 11:52:22] <vinzpr> At:
[10-Sep-2009 11:52:22] <vinzpr> Device:
[10-Sep-2009 11:52:22] <vinzpr> Component:
[10-Sep-2009 11:52:23] <vinzpr> Severity: Critical
[10-Sep-2009 11:52:25] <vinzpr> Message:
[10-Sep-2009 11:52:27] <vinzpr> threshold of Memory Critical not met: current value 34116.00
[10-Sep-2009 11:52:37] <rmatte> what happened to the capital T>?
[10-Sep-2009 11:52:39] <rmatte> Threshold of Memory Critical restored: current value: 98240.00
[10-Sep-2009 11:52:43] <rmatte> that was the example you gave before
[10-Sep-2009 11:52:47] <rmatte> you don't see that at all now?
[10-Sep-2009 11:52:55] <vinzpr> why is there is a message like this Memory thresholdexceeded
[10-Sep-2009 11:53:06] <rmatte> vinzpr: no idea, we can deal with that in a bit
[10-Sep-2009 11:53:12] <rmatte> just work with me here
[10-Sep-2009 11:53:13] <rmatte> lol
[10-Sep-2009 11:53:17] <vinzpr> yep the same message
[10-Sep-2009 11:53:25] <vinzpr> as critical and clear
[10-Sep-2009 11:53:27] <rmatte> with a capital T or lowercase?
[10-Sep-2009 11:53:32] <rmatte> the case is very important
\n[10-Sep-2009 11:54:17] <rmatte> I just realized that using the\nsummary probably won't work properly, we need to use the event message
[10-Sep-2009 11:54:31] <vinzpr> on critical : Memory threshold exceeded:current value 98.8%
[10-Sep-2009 11:54:39] <rmatte> I need EXACT examples of alerts and clear events
[10-Sep-2009 11:54:51] <rmatte> messages, not summaries
[10-Sep-2009 11:55:09] <vinzpr> on clear : Event: 'Memory threshold exceeded: current value 98.8%'
[10-Sep-2009 11:55:09] <vinzpr> Cleared by: 'Threshold of Memory Critical restored: current value: 112652.00'
[10-Sep-2009 11:55:25] <rmatte> so that's the message then?
[10-Sep-2009 11:55:35] <vinzpr> did you see it the 98.8% still show on clear
[10-Sep-2009 11:55:39] <rmatte> like if you click on the details link: what does "message" show up as
[10-Sep-2009 11:55:42] <rmatte> in the details window
[10-Sep-2009 11:55:45] <rmatte> for each event
\n[10-Sep-2009 11:56:20] <rmatte> yeh, we'll deal with that in a\nsecond, I really just need to know that you are giving me 100% accurate\nevent messages here
[10-Sep-2009 11:56:36] <rmatte> I can't modify this script if I don't have accurate data to work with
[10-Sep-2009 11:56:44] <vinzpr> threshold of Memory Critical not met: current value 32784.00
[10-Sep-2009 11:56:51] <rmatte> ok, and for clear?
[10-Sep-2009 11:57:35] <vinzpr> Event: 'Memory threshold exceeded: current value 98.8%'
[10-Sep-2009 11:57:36] <vinzpr> Cleared by: 'Threshold of Memory Critical restored: current value: 112652.00'
[10-Sep-2009 11:59:49] <rmatte> ok
[10-Sep-2009 11:59:49] <rmatte> for starters, on the match = line
[10-Sep-2009 11:59:49] <rmatte> at the end of that line...
[10-Sep-2009 11:59:49] <rmatte> change evt.summary to evt.message
[10-Sep-2009 11:59:49] <rmatte> then test it
[10-Sep-2009 11:59:49] <vinzpr> ok
[10-Sep-2009 12:01:06] <rmatte> If that doesn't work then use this as a transform: http://pastebin.com/m1341a494
[10-Sep-2009 12:01:15] <rmatte> http://pastebin.com/m1341a494 is probably a better option
[10-Sep-2009 12:02:04] <rmatte> actually wait
[10-Sep-2009 12:02:30] <vinzpr> Alert : Memory threshold exceeded:current value 99.0%
[10-Sep-2009 12:02:31] <vinzpr> Clear : Threshold of MemoryCritical restored: current value: 495268.00
[10-Sep-2009 12:03:02] <rmatte> why is there no space between : and current now
[10-Sep-2009 12:03:04] <rmatte> there was before
[10-Sep-2009 12:03:35] <rmatte> paste your match line
[10-Sep-2009 12:03:45] <rmatte> and your evt.summary line
[10-Sep-2009 12:04:09] <vinzpr> import re
[10-Sep-2009 12:04:09] <vinzpr> match = re.search('threshold of Memory [^:]+: current value ([\\d\\.]+)', evt.message,re.IGNORECASE)
[10-Sep-2009 12:04:09] <vinzpr> if match and device:
[10-Sep-2009 12:04:09] <vinzpr> avail = float(match.groups()[0]) * 1024
[10-Sep-2009 12:04:09] <vinzpr> total = device.hw.totalMemory
[10-Sep-2009 12:04:09] <vinzpr> percent_used = (1 - (avail / total)) * 100
[10-Sep-2009 12:04:11] <vinzpr> evt.summary = "Memory threshold exceeded: current value %3.1f%%" % (percent_used)
[10-Sep-2009 12:04:41] <rmatte> ok, try the transform here: http://pastebin.com/m1341a494
[10-Sep-2009 12:04:58] <rmatte> I think the ignorecase is screwing it up for some reason
[10-Sep-2009 12:05:15] <vinzpr> ok
[10-Sep-2009 12:18:03] <vinzpr> is there a link how to create a transform
[10-Sep-2009 12:19:04] <rmatte> well, for pretty basic transforms
[10-Sep-2009 12:19:25] <rmatte> you basically need to know python
[10-Sep-2009 12:19:32] <rmatte> transforms are python scripts
[10-Sep-2009 12:19:51] <rmatte> http://www.zenoss.com/community/wiki/EventTransforms/?searchterm=device
[10-Sep-2009 12:20:25] <rmatte> they can get quite advanced
[10-Sep-2009 12:21:57] <vinzpr> On Alert : Memory threshold exceeded:current value 96.8%
[10-Sep-2009 12:22:25] <rmatte> I've never had issues with spaces magically dissapearing like that
[10-Sep-2009 12:22:29] <vinzpr> on Clear: Threshold of Memory Warningrestored: current value: 578820.00
[10-Sep-2009 12:23:03] <vinzpr> Event: 'Memory threshold exceeded: current value 96.8%'
[10-Sep-2009 12:23:04] <vinzpr> Cleared by: 'Threshold of Memory Warning restored: current value: 578820.00'
[10-Sep-2009 12:23:04] <vinzpr> At: 2009/09/10 17:17:07.000
[10-Sep-2009 12:23:04] <vinzpr> Device:
[10-Sep-2009 12:23:04] <vinzpr> Component:
[10-Sep-2009 12:23:04] <vinzpr> Severity: Warning
[10-Sep-2009 12:23:06] <vinzpr> Message:
[10-Sep-2009 12:23:08] <vinzpr> threshold of Memory Warning not met: current value 121556.00
[10-Sep-2009 12:23:11] <vinzpr> on Alert:
[10-Sep-2009 12:23:40] <vinzpr> Device:
[10-Sep-2009 12:23:40] <vinzpr> Component:
[10-Sep-2009 12:23:40] <vinzpr> Severity: Warning
[10-Sep-2009 12:23:40] <vinzpr> Time: 2009/09/10 17:15:27.000
[10-Sep-2009 12:23:40] <vinzpr> Message:
[10-Sep-2009 12:23:41] <vinzpr> threshold of Memory Warning not met: current value 121556.00
[10-Sep-2009 12:23:57] <rmatte> ok, you're pasting way too much stuff
[10-Sep-2009 12:23:59] <rmatte> that's not helping
[10-Sep-2009 12:24:00] <rmatte> lol
[10-Sep-2009 12:24:03] <vinzpr> hehe
[10-Sep-2009 12:24:06] <vinzpr> just showing
[10-Sep-2009 12:24:13] <vinzpr> the difference...lol
[10-Sep-2009 12:24:19] <rmatte> yeh, but I can barely make sense of what you're pasting
[10-Sep-2009 12:26:00] <rmatte> is there any particular reason that you care what the value is when it clears?
[10-Sep-2009 12:27:45] <vinzpr> actually your alert is on percentage so clear should show you the current percentage
[10-Sep-2009 12:28:03] <rmatte> ok, this is what I want you to paste, and ONLY this...
[10-Sep-2009 12:28:11] <rmatte> the summary and message for the alert
[10-Sep-2009 12:28:14] <vinzpr> shoot...
[10-Sep-2009 12:28:18] <rmatte> and then the summary and message for the clear
[10-Sep-2009 12:28:31] <rmatte> that's it, don't paste a whole wackload of stuff like you did before, it's useless
[10-Sep-2009 12:29:02] <rmatte> summary and message are 2 different things
[10-Sep-2009 12:29:06] <rmatte> so I want both of those for each
[10-Sep-2009 12:33:13] <vinzpr> CLEAR:
[10-Sep-2009 12:33:13] <vinzpr> summary Threshold of Memory Warning restored: current value: 578820.00
[10-Sep-2009 12:33:13] <vinzpr> message Threshold of Memory Warning restored: current value: 578820.00
[10-Sep-2009 12:33:13] <vinzpr> ALERT:
[10-Sep-2009 12:33:13] <vinzpr> summary Memory threshold exceeded: current value 96.8%
[10-Sep-2009 12:33:14] <vinzpr> message threshold of Memory Warning not met: current value 121556.00
[10-Sep-2009 12:34:59] <rmatte> and that's while using the pastebin version of the transform?
[10-Sep-2009 12:35:19] <rmatte> http://pastebin.com/m1341a494
[10-Sep-2009 12:35:37] <vinzpr> yes
[10-Sep-2009 12:36:43] <rmatte> that doesn't make sense, hmmm
[10-Sep-2009 12:37:26] <rmatte> ok, we'll probably have to add a couple of lines
[10-Sep-2009 12:37:33] <rmatte> I just need to finish some things I'm doing
[10-Sep-2009 12:37:39] <vinzpr> ok
[10-Sep-2009 12:40:29] <twm1010> I'm gone for a week and three new versions come out?
\n[10-Sep-2009 12:44:30] <rmatte> twm1010: well, 2.4.3 was buggy\n2.4.4 fixed 1 bug, and 2.4.5 fixed another, but it's still buggy
\n[10-Sep-2009 12:44:54] <rmatte> in 2.4.5 you can't change\nhardware of software product info on device edit tab for devices
[10-Sep-2009 12:45:01] <rmatte> so if you like being able to do that, don't upgrade
[10-Sep-2009 12:45:02] <rmatte>
[10-Sep-2009 12:46:49] <twm1010> I'm still on 2.4.2 here
[10-Sep-2009 12:46:57] <twm1010> I think I'll stick with it for the time being
[10-Sep-2009 12:49:00] <rmatte> yeh, just make sure you do "zenpatch 14181" as zenoss user
[10-Sep-2009 12:49:03] <rmatte> then restart Zenoss
[10-Sep-2009 12:49:15] <rmatte> fixes: http://dev.zenoss.org/trac/ticket/5279
[10-Sep-2009 12:52:41] <mrayzenoss> we're trying to push out the beta right now, having redirect issues with SourceForge
[10-Sep-2009 12:54:06] <rmatte> vinzpr: http://pastebin.com/m4d4dff2f
[10-Sep-2009 12:54:08] <rmatte> try that
[10-Sep-2009 12:54:32] <rmatte> mrayzenoss: cool, I'll start setting up a test box for it...
[10-Sep-2009 12:55:16] <vinzpr> ok
[10-Sep-2009 12:59:32] <vinzpr> thanks for the help rmatte..i'll inform you later regarding output of the new transform
[10-Sep-2009 12:59:38] <rmatte> k, np
[10-Sep-2009 12:59:54] <vinzpr> i have to run now..
[10-Sep-2009 13:00:14] <rmatte> k
[10-Sep-2009 13:00:23] <vinzpr> cheers
[10-Sep-2009 13:05:19] <IoNu|z> is there a way to have zenoss display the device name instead of the ip on discovery?
[10-Sep-2009 13:05:27] <IoNu|z> when i add a device by ip?
[10-Sep-2009 13:05:37] <IoNu|z> like to pull the device snmp hostname instead of the ip?
[10-Sep-2009 13:07:48] <rmatte> IoNu|z: Not that I'm aware of, you can rename it afterwards
[10-Sep-2009 13:08:04] <rmatte> or do what I do, I added the IP and hostname to the hosts file on the box
[10-Sep-2009 13:08:07] <rmatte> then add it by hostname
[10-Sep-2009 13:08:21] <mrayzenoss> 2.5 is adding a "title" as well
[10-Sep-2009 13:08:32] <rmatte> a "title" ?
[10-Sep-2009 13:08:54] <mrayzenoss> it's complicated, but basically IP address sets the hostname if it finds it in DNS
[10-Sep-2009 13:09:01] <mrayzenoss> SNMP sets hostname as well
[10-Sep-2009 13:09:14] <rmatte> ah, that's nice to see
[10-Sep-2009 13:09:18] <mrayzenoss> and title is if you want to override them
[10-Sep-2009 13:09:23] <rmatte> gotcha
[10-Sep-2009 13:09:40] <rmatte> "Super Server, DESTROYER OF WORLDS!" would be the title for instance.
[10-Sep-2009 13:09:42] <rmatte>
[10-Sep-2009 13:09:54] <mrayzenoss> so there's essentially an order of operations for how machines will show up by name
[10-Sep-2009 13:10:01] <mrayzenoss> title>hostname>ip
[10-Sep-2009 13:10:13] <rmatte> what about machines already in there?
[10-Sep-2009 13:10:24] <mrayzenoss> title won't be used unless set by the user
[10-Sep-2009 13:10:29] <mrayzenoss> so migration won't be an issue
[10-Sep-2009 13:10:32] <rmatte> ah ok
[10-Sep-2009 13:10:45] <rmatte> cool feature
[10-Sep-2009 13:10:56] <rmatte> what about when pulling info directly out of zope?
[10-Sep-2009 13:10:59] <rmatte> like if I do d.id
[10-Sep-2009 13:11:01] <rmatte> or whatever
[10-Sep-2009 13:11:14] <mrayzenoss> ahh, everywhere where I just said hostname, replace with 'id'
[10-Sep-2009 13:11:23] <mrayzenoss> title>id>ip
[10-Sep-2009 13:11:32] <rmatte> I see
[10-Sep-2009 13:11:45] <IoNu|z> hmm..
[10-Sep-2009 13:11:50] <IoNu|z> i had that in 3.2
[10-Sep-2009 13:11:50] <rmatte> I can't see myself using titles very often then
[10-Sep-2009 13:11:57] <IoNu|z> it was adding hostname instead of ip
[10-Sep-2009 13:12:03] <rmatte> is there still the option to set the ID manually as well?
[10-Sep-2009 13:12:07] <IoNu|z> but i dont remeber how i did it
[10-Sep-2009 13:12:07] <mrayzenoss> yes
[10-Sep-2009 13:12:12] <rmatte> ok, good
[10-Sep-2009 13:12:13] <IoNu|z> i think there was a patch or soemthing
[10-Sep-2009 13:13:07] <mrayzenoss> Rename Device sets the 'id'
[10-Sep-2009 13:13:20] <mrayzenoss> and in 2.5, title will be available on the 'Edit' tab
[10-Sep-2009 13:14:00] <mrayzenoss> and when you do an 'Add Device', title will be exposed there too
[10-Sep-2009 13:15:48] <rmatte> ah good
[10-Sep-2009 13:16:39] <IoNu|z> hmmm
[10-Sep-2009 13:16:48] <IoNu|z> sorry for the quetsion, but where do i edit that?
[10-Sep-2009 13:17:25] <mrayzenoss> page menu of the device->Manage
[10-Sep-2009 13:17:38] <IoNu|z> great thank you
[10-Sep-2009 13:19:10] <rmatte> afk
[10-Sep-2009 13:26:01] <twm1010> mrayzenoss: Is it the reverse DNS lookup that Zenoss uses?
[10-Sep-2009 13:26:34] <mrayzenoss> twm1010: yeah, if dns returns a hostname, it goes into 'id'
\n[10-Sep-2009 13:27:16] <twm1010> That would explain why when I\nsetup my box and had it use my new DNS servers that did not yet have\nreverse zones loaded, that nothing came back with a name
\n[10-Sep-2009 13:27:20] <mrayzenoss> I think that's right, I had a\ndeveloper walk me through the new 'title' thing for the beta
[10-Sep-2009 13:27:30] <mrayzenoss> yeah
[10-Sep-2009 13:27:38] <mrayzenoss> so in my home network, I've got no DNS
[10-Sep-2009 13:27:45] <mrayzenoss> but it's a small network
[10-Sep-2009 13:27:59] <mrayzenoss> so some boxes hostnames are set by SNMP
[10-Sep-2009 13:28:19] <mrayzenoss> some are set by /etc/hosts
[10-Sep-2009 13:28:19] <mrayzenoss> and some are blank (ip address)
\n[10-Sep-2009 13:32:43] <rmatte> yeh, I always just add the device\nto the hosts file, then add it by hostname, and then I have my scripts\nsetup on each Zenoss box which sync up the hosts file with Zenoss every\nhour
[10-Sep-2009 13:33:27] <rmatte> the nice\nthing about it is if someone doesn't have access to edit the hosts file\nbut wants to add a device, they can add it by IP, rename it, and then\nan entry gets automatically placed in to the hosts file within an hour
[10-Sep-2009 13:34:04] <rmatte> I just add them directly because when I'm adding a bunch of devices it keeps me from having to rename each one
[10-Sep-2009 13:36:09] <twm1010> as opposed to just keeping your DNS square?
[10-Sep-2009 13:37:02]\n<rmatte> We're in a different situation, we monitor devices for\nclients, so we have a ton of Zenoss boxes, and we don't want to have to\nkeep DNS entries for every client device
[10-Sep-2009 13:37:25] <twm1010> ah, gotcha
\n[10-Sep-2009 13:37:34] <rmatte> The Zenoss boxes are also used to\nconnect to client devices from, so we just need accurate hosts files
[10-Sep-2009 13:38:50] <rmatte> We may start using DNS for it down the road, but for now we're not
\n[10-Sep-2009 13:38:50] <rmatte> the only downside to this method\nis that any windows TS servers that we have don't have the entries, so\nyou have to connect via IP
[10-Sep-2009 13:39:02] <rmatte> but it's not a big deal
[10-Sep-2009 13:42:03] <rmatte> mrayzenoss: think an upgrade from 2.4.5 to the beta would work?
[10-Sep-2009 13:42:12] <mrayzenoss> not the beta 1 that's coming out
[10-Sep-2009 13:42:16] <rmatte> k
[10-Sep-2009 13:42:23] <mrayzenoss> it's known to be broken
[10-Sep-2009 13:42:28] <rmatte> any word on when stack installer will be available?
[10-Sep-2009 13:42:29] <mrayzenoss> and there's a P1
[10-Sep-2009 13:42:40] <mrayzenoss> it's on the calendar for Tuesday
[10-Sep-2009 13:42:45] <rmatte> k cool
\n[10-Sep-2009 13:43:35] <rmatte> well, as soon as the stack\ninstaller comes out I'll test it out (even if the upgrade is broken)
[10-Sep-2009 13:43:36] <twm1010> beta for.... 2.5?
[10-Sep-2009 13:43:47] <rmatte> and when the upgrade issue gets fixed I'll test the upgrade out as well
[10-Sep-2009 13:43:58] <mrayzenoss> yeah, 2.5 beta is minutes away
[10-Sep-2009 13:44:00] <rmatte> twm1010: correct, for King Crab (2.5)
\n[10-Sep-2009 13:44:18] <twm1010> Sweet.... i'll hit the tracker\nin a minute and see what's in, will be happy to beta test that
\n[10-Sep-2009 13:44:52] <rmatte> well, my beta box is setup, going\nto shut it down until I can get my hands on the stack installer
\n[10-Sep-2009 13:45:37] <rmatte> now I'm just waiting on a Windows\n2000 key so that I can test out net-snmp on Windows 2000 to see if it\nfixes the memory monitoring issues we're seeing
[10-Sep-2009 13:45:45] <mrayzenoss> http://hannahmontana.sourceforge.net
[10-Sep-2009 13:46:04] <rmatte> you can't be serious?
[10-Sep-2009 13:46:08] <rmatte> wtf
[10-Sep-2009 13:46:15] <rmatte> lol
[10-Sep-2009 13:46:22] <twm1010> I should put that on my girlfriend's daughters iMac
[10-Sep-2009 13:46:44] <rmatte> that's just scary
[10-Sep-2009 13:46:44] <twm1010> rmatte: mind if it's a not quite so legit key?
[10-Sep-2009 13:48:12] <rmatte> 2 of the items in the FAQ contradict themselves...
[10-Sep-2009 13:48:18] <rmatte> Q : How/why did you make such a great OS?
\n[10-Sep-2009 13:48:19] <rmatte> A : I thought - what would\nattract young users to Linux? So I created this idea after a lot of\nreading and work.
[10-Sep-2009 13:48:25] <rmatte> and then...
[10-Sep-2009 13:48:26] <rmatte> Q : How can I watch DVDs and other Media?
[10-Sep-2009 13:48:26] <rmatte> A : click the Hannah Montana menu, then Utilities, then Konsole, then type:
[10-Sep-2009 13:48:26] <rmatte> sudo apt-get install libdvdcss2 kubuntu-restricted-extras
[10-Sep-2009 13:48:39] <rmatte> yes, because I'm sure that all kids want to have to install apps from the commandline
[10-Sep-2009 13:48:50] <rmatte> twm1010: yeh, we need a legit key lol
\n[10-Sep-2009 13:49:15] <rmatte> twm1010: we'll be able to get\none, we're a Microsoft gold partner, I'm just waiting on the guy who\nhas access to get the key to send it to me
[10-Sep-2009 13:49:21] <jb> argh
[10-Sep-2009 13:49:32] <jb> why do vendors have to store related snmp information in like 20 different tables..
[10-Sep-2009 13:49:50] <rmatte> jb: because that's the way it's generally done hehe
[10-Sep-2009 13:49:58] <jb> no, not really
[10-Sep-2009 13:50:12] <rmatte> yeh, I know, lol
[10-Sep-2009 13:50:14] <jb> not this bad
[10-Sep-2009 13:50:14] <twm1010> ok i give, what's the URL for the zenoss milestones
[10-Sep-2009 13:50:17] <rmatte> just kidding around
[10-Sep-2009 13:50:25] <jb> it makes coding modelers difficult
[10-Sep-2009 13:50:26] <jb>
[10-Sep-2009 13:51:38] <rmatte> twm1010: Blue Crab, King Crab, Stone Crab <--- There you go lol
[10-Sep-2009 13:51:49] <rmatte> twm1010: I don't recall ever seeing a page listing anything more formal
[10-Sep-2009 13:51:55] <rmatte> though Matt would know best
\n[10-Sep-2009 13:52:51] <rmatte> jb: yeh, when I was writing my\nscript to monitor Windows Memory usage via plain SNMP the info I had to\naccess was all over the place
[10-Sep-2009 13:52:58] <rmatte> maybe not quite as bad as you're seeing
[10-Sep-2009 13:55:11] <mrayzenoss> http://blog.zenoss.com/2009/09/10/zenoss-2-5-beta-is-now-available/
[10-Sep-2009 13:55:21] <skipzoid> hey all again
[10-Sep-2009 13:57:31] <twm1010> Sweet... I'll be setting that up tomorrow
[10-Sep-2009 13:57:46] <rmatte> I'll be setting it up... Tuesday, looks like
[10-Sep-2009 13:57:47] <rmatte>
[10-Sep-2009 13:58:01] <skipzoid> i'm trying to add a data source, if I type this snmpwalk -cpublic -v2c 192.168.0.9 mib-2.16.1.1.1.7.1
[10-Sep-2009 13:58:01]\n<skipzoid> SNMPv2-SMI::mib-2.16.1.1.1.7.1 = Counter32: 17 on the\nconsole you can see it queries the device and returns the value, but is\nI put the same mib-2.16.1.1.1.7.1 in the OID box for the data source it\nsays invalid OID when I try and save it - where am i being stupid ?\nwhat am I missing
[10-Sep-2009 13:58:29] <twm1010> skipzoid: That's an abbreviated OID
[10-Sep-2009 13:58:32] <twm1010> you need the full OID
[10-Sep-2009 13:58:45] <twm1010> everything that mib- is hiding
[10-Sep-2009 13:59:02] <skipzoid> ah - how can I find that from the device ?
[10-Sep-2009 13:59:16] <twm1010> rmatte: isn't there a command flag to show full OIDs?
[10-Sep-2009 13:59:24] <rmatte> skipzoid: add -0n as an option
[10-Sep-2009 13:59:31] <twm1010> ding ding... he wins.
[10-Sep-2009 13:59:32] <rmatte> then you'll see what the OIDs really look like
[10-Sep-2009 13:59:50] <rmatte> snmpwalk tries to translate them from standard Mibs by default
[10-Sep-2009 13:59:58] <rmatte> and ends up partially translating them
[10-Sep-2009 14:00:57]\n<skipzoid> if I add -n0 to the command line it just shows me the\nsyntax of all the commands -n says CONTEXT next to it
[10-Sep-2009 14:01:03] <rmatte> no
[10-Sep-2009 14:01:05] <rmatte> -0n
[10-Sep-2009 14:01:07] <rmatte> not n0
[10-Sep-2009 14:01:09] <rmatte> lol
[10-Sep-2009 14:01:53] <skipzoid> ah - better still is a an 'o' not a zero
[10-Sep-2009 14:02:13] <rmatte> oh right
[10-Sep-2009 14:02:14] <rmatte> my bad
[10-Sep-2009 14:02:15] <rmatte>
[10-Sep-2009 14:02:25] <skipzoid> its saved it
[10-Sep-2009 14:02:37] <skipzoid> i'm trying to create a graph
[10-Sep-2009 14:02:55] <twm1010> well... you're almost there
[10-Sep-2009 14:03:04] <twm1010> getting the datapoint collecting is step 1
[10-Sep-2009 14:03:13] <twm1010> once thats square, the rest is easy
[10-Sep-2009 14:09:12]\n<skipzoid> wadaya know I have a graph - no data on it yet just\nsays 'nan' for the moment - am i to wait for it to gather some info ?
[10-Sep-2009 14:09:58] <mrayzenoss> yeah
[10-Sep-2009 14:12:26] <rmatte> skipzoid: takes 3 polling cycles (about 15 mins)
[10-Sep-2009 14:12:30] <rmatte> then you should see data
[10-Sep-2009 14:12:42] <skipzoid> ok, i feel pleased with my self now
[10-Sep-2009 14:13:32] <skipzoid> need to work out how to get emails and text messages sent from events
\n[10-Sep-2009 14:15:31] <rmatte> Settings, configure the SMTP\nsettings, then... Settings -> Users -> <Username> -> add\nEmail and pager info -> Alerting Rules tab
[10-Sep-2009 14:15:53] <rmatte> you can group users as well and apply alerting rules to groups
\n[10-Sep-2009 14:16:13] <skipzoid> question about the graphs, if I\nset it to be a guage and not a counter - is that info real time as a\nguage implys?
[10-Sep-2009 14:16:48] <twm1010>\nIs Zenoss intelligent enough not to send alerts for WinService events\nif the device is down? Or do I have to put that in my alerting criteria
\n[10-Sep-2009 14:17:13] <rmatte> twm1010: If a device is not\npingable at all Zenoss stops monitoring everything until it's pingable\nagain
[10-Sep-2009 14:17:16] <rmatte> so in short, yes
[10-Sep-2009 14:17:23] <rmatte> sometimes 1 or 2 service alerts slip through
[10-Sep-2009 14:17:41] <twm1010> Yeah, seen that before... I'm adding a whole bunch of service monitoring today.
\n[10-Sep-2009 14:17:43] <rmatte> only if it notices that the\nservices are down before it notices that the device is not pingable
[10-Sep-2009 14:17:45] <rmatte> which is rare
[10-Sep-2009 14:17:58] <rmatte> but it does happen
\n[10-Sep-2009 14:18:28] <twm1010> Don't you love when something\ndevelopement is suddenly "production" because someone else doesn't want\nto pony up 4-5k to add licenses to some "other" piece of software
[10-Sep-2009 14:21:10] <rmatte> yup
[10-Sep-2009 14:21:35] <rmatte> you're obviously not a government worker
[10-Sep-2009 14:23:20] <mrayzenoss> bbiab
[10-Sep-2009 14:23:28] <twm1010> well, at least my work so far is becoming appreciated
[10-Sep-2009 14:24:09] <rmatte> mine is about half and half for some reason
[10-Sep-2009 14:24:20] <rmatte> mostly because people expect stuff to just magically work with no dev work required
[10-Sep-2009 14:24:36] <rmatte> meaning stuff always should have been setup to work a certain way yesterday
[10-Sep-2009 14:24:44] <rmatte> regardless of whether I found out about it today or not
[10-Sep-2009 14:24:48] <rmatte>
[10-Sep-2009 14:29:23] <rmatte> hmmm, I'm trying to understand this line...
[10-Sep-2009 14:29:24] <rmatte> (((1 248 400 328 - 1 247 795 528) / 60) / 60) / 24 = 7
[10-Sep-2009 14:29:32] <rmatte> it's apparently supposed to work out to 7 days
[10-Sep-2009 14:29:41] <rmatte> just don't know what the heck all the values represent
[10-Sep-2009 14:29:58] <rmatte> oh wait
[10-Sep-2009 14:30:06] <Apocalipse> hi people
[10-Sep-2009 14:30:22] <Apocalipse> link between locations are not wroking...
[10-Sep-2009 14:30:32] <Apocalipse> how can i add then?
[10-Sep-2009 14:30:34] <rmatte> you mean on the google maps?
[10-Sep-2009 14:30:38] <Apocalipse> yes
[10-Sep-2009 14:30:57] <Apocalipse> matt send me a tutorial... but i think i made something wrong...
[10-Sep-2009 14:30:59] <rmatte> well, first step is to click on Networks on the left
[10-Sep-2009 14:31:15] <rmatte> then go to zProperties
[10-Sep-2009 14:31:15] <Apocalipse> ok
\n[10-Sep-2009 14:31:17] <dollarbang> konsa79: I saw your remark on\nthe Geomapping issue. can you provide me with how you generated that\nlist?
[10-Sep-2009 14:31:24] <rmatte> and change zDrawMapLinks to True
[10-Sep-2009 14:31:36] <rmatte> then it really depends on the types of routes that you've modeled via SNMP
[10-Sep-2009 14:31:42] <Apocalipse> already True
[10-Sep-2009 14:32:01] <Apocalipse> thats the problem...i thin
[10-Sep-2009 14:32:03] <rmatte> Ok, well then here's what else could be wrong...
[10-Sep-2009 14:32:05] <Apocalipse> i think
[10-Sep-2009 14:32:25] <rmatte> Let's say you have 7 sites, and they all connect in to a central router
[10-Sep-2009 14:32:33] <Apocalipse> ok
[10-Sep-2009 14:32:43] <Apocalipse> thats my problem...
\n[10-Sep-2009 14:32:45] <rmatte> if you have devices at those 7\nsites in Zenoss, but don't have that central router in with all router\ninfo for everything, you won't get any lines at all
[10-Sep-2009 14:33:03]\n<rmatte> I have 1 Zenoss box where I can't get the lines to draw\nno matter what just because of the types of circuits and routing they\nare using
[10-Sep-2009 14:33:16] <Apocalipse> here is a full mesh mpls network
[10-Sep-2009 14:33:31] <rmatte> yeh, mpls doesn't tend to map too well
[10-Sep-2009 14:33:39] <Apocalipse>
[10-Sep-2009 14:34:24] <Apocalipse> but is it possible to add a static link?
[10-Sep-2009 14:34:24] <rmatte> are you sure all of the devices are fully modeled?
[10-Sep-2009 14:34:29] <Apocalipse> manual added
[10-Sep-2009 14:34:29] <rmatte> and that Zenoss picked up the routes?
[10-Sep-2009 14:34:38] <rmatte> no, the links are only automatically generated
[10-Sep-2009 14:34:44] <rmatte> they are meant to represent actual links
[10-Sep-2009 14:34:45] <Apocalipse> i think they are not fully modeled
[10-Sep-2009 14:34:52] <rmatte> then make sure they are
[10-Sep-2009 14:34:58] <Apocalipse> ok
[10-Sep-2009 14:35:02] <Apocalipse> let me try
[10-Sep-2009 14:35:11] <twm1010> my MPLS graphs fine
[10-Sep-2009 14:35:53] <twm1010> its an odd picture, but its correct
[10-Sep-2009 14:36:22] <Apocalipse> i have a lot of networks listed here...
[10-Sep-2009 14:36:45] <Apocalipse> where does zenoss pic up that?
[10-Sep-2009 14:37:33] <Apocalipse> ????????
[10-Sep-2009 14:38:02] <twm1010> from the routing tables collected via SNMP from the devices you've added
[10-Sep-2009 14:38:03] <rmatte> he's probably just missing the core router
[10-Sep-2009 14:38:17] <Apocalipse> core router is here
\n[10-Sep-2009 14:38:18] <rmatte> I just checked and actually, on\nthis one that won't draw lines it's because we don't manage their core\nrouter
[10-Sep-2009 14:38:24] <rmatte> so Zenoss has no idea what they link in to
[10-Sep-2009 14:39:28] <Apocalipse> hunnn
[10-Sep-2009 14:39:38] <Apocalipse> core router is here...
[10-Sep-2009 14:39:45] <Apocalipse> but not well modeled...
[10-Sep-2009 14:39:53] <Apocalipse> :"
[10-Sep-2009 14:39:53] <rmatte> then that's your problem
[10-Sep-2009 14:39:55] <Apocalipse>
[10-Sep-2009 14:40:00] <rmatte> it needs as much route info as possible
\n[10-Sep-2009 14:40:37] <rmatte> If it's timing out when modelling\nthen go to Devices -> zProperties and change zCollectorClientTimeout\nto something like 5000
[10-Sep-2009 14:40:49] <rmatte> and when you're done modelling change it back
[10-Sep-2009 14:42:11] <Apocalipse> ok
[10-Sep-2009 14:42:17] <Apocalipse> let me do it...
[10-Sep-2009 14:42:48] <twm1010> you can do it!
[10-Sep-2009 14:44:06] <rmatte> actually this one that's not drawing lines seems to be due to an abundance of indirect bgp routes
[10-Sep-2009 14:44:24] <rmatte> we actually do monitor the core router, but it only has 4 direct routes
[10-Sep-2009 15:00:45] <chemist> good evening all
[10-Sep-2009 15:01:31]\n<chemist> as was suggested on the forums, I have installed\nsnmp-informant on some of my win boxes but I still can not get\nperformance graphs
[10-Sep-2009 15:02:24] <chemist> I keeep getting 'Missing RRD file' on the cpu graph, and nothing else
[10-Sep-2009 15:03:24] <metalklesk> mrayzenoss, can you say us what are the news about the new event console in zenoss 2.5 ?
[10-Sep-2009 15:05:22] <chemist> scrap that, it working now
[10-Sep-2009 15:12:16] <Apocalipse> hello
[10-Sep-2009 15:16:32] <Apocalipse> i modeled my core router...
[10-Sep-2009 15:16:47] <Apocalipse> and other routers
[10-Sep-2009 15:19:09] <metalklesk> now you got your routers modeled, can you export the zenpack so anyone can use it ?
[10-Sep-2009 15:19:24] <metalklesk> zenoss can do that right ?
[10-Sep-2009 15:19:55] <Apocalipse> matt can answer u... much better than me
[10-Sep-2009 15:19:56] <Apocalipse>
[10-Sep-2009 15:20:56] <mrayzenoss> what's the question?
\n[10-Sep-2009 15:22:30] <metalklesk> if Apocalipse modeled some\nrouters, can he export it as a zenpack so anyone can use it ?
[10-Sep-2009 15:23:18] <djack> hey you all
[10-Sep-2009 15:23:26] <djack> somebody awake here
[10-Sep-2009 15:23:27] <metalklesk> i'm still a newbie, i'm in a learning phase
[10-Sep-2009 15:23:37] <metalklesk> hi djack
[10-Sep-2009 15:23:46] <djack> have a strange issue
\n[10-Sep-2009 15:25:11] <djack> have java process in monitoring\n.... we changed the Xmm value and the new value is visible as a os\nprocess for the device .... only the old process generates errors ...
\n[10-Sep-2009 15:25:29] <mrayzenoss> Apocalipse: yes, once you've\ngot something you want to export, you can create a new ZenPack and\nstart marking things for inclusion in it
[10-Sep-2009 15:25:52] <djack> he mrayzenoss
[10-Sep-2009 15:26:22] <mrayzenoss> Apocalipse: http://www.zenoss.com/community/docs/zenoss-guide/2.4.3/ch13s03.html
[10-Sep-2009 15:26:27] <mrayzenoss> djack: greetings
[10-Sep-2009 15:26:55] <djack> i have strange error because os value changed
\n[10-Sep-2009 15:27:13] <Apocalipse> matt... to stop snmp\nmonitoring for a complete class... the procedure is: in Device/Server\n-> zProperties
[10-Sep-2009 15:27:36] <Apocalipse> zSnmpMonitorIgnore -> true
[10-Sep-2009 15:27:39] <Apocalipse> right?
[10-Sep-2009 15:28:54] <mrayzenoss> sounds right
\n[10-Sep-2009 15:28:54] <metalklesk> you got it Apocalipse, you\nhave modeled your routers now you can make your own zenpacks and share\nit with the community, wich i'm part of it
[10-Sep-2009 15:28:54] <djack> lol
[10-Sep-2009 15:29:32] <Apocalipse> metalklesk... its not possible... because i have some confidential info...
\n[10-Sep-2009 15:29:34] <metalklesk> when i learn more i will\nstart modeling stuffs so i can share zenpacks with the community
[10-Sep-2009 15:30:05] <Apocalipse> i think u can add Ryan Zenpack...for cisco devices...
[10-Sep-2009 15:30:15] <metalklesk> mmm then you can build a nice how to
[10-Sep-2009 15:30:26] <metalklesk> replacing the credentials
[10-Sep-2009 15:30:27] <Apocalipse> i did... and now i got a lot of MIBs to explore
[10-Sep-2009 15:30:32] <metalklesk> by xxxxxxx
[10-Sep-2009 15:30:34] <metalklesk>
[10-Sep-2009 15:31:02] <Apocalipse> matt... devices are still monitoring by snmp
[10-Sep-2009 15:31:44] <metalklesk> i was looking into the SuSE zenpack source code
[10-Sep-2009 15:32:16] <metalklesk> and found the files i should edit to adapt it to mandriva or whatever
[10-Sep-2009 15:32:34] <metalklesk> but as i don't know python (i know java)
\n[10-Sep-2009 15:32:44] <rmatte> metalklesk: modeling a router\nmeans that he collected data for it via snmp, it doesn't mean he made a\ndevice template or anything
[10-Sep-2009 15:32:51] <mrayzenoss> Apocalipse: try pushing the change
[10-Sep-2009 15:33:05] <metalklesk> what should i do to create de egg file ?
[10-Sep-2009 15:33:19] <metalklesk> ok rmatte
[10-Sep-2009 15:34:01] <mrayzenoss> from Settings->ZenPacks->your zenpack->Export
[10-Sep-2009 15:35:08] <chemist> I have a strange problem with monitoring a process
[10-Sep-2009 15:35:36] <metalklesk> mrayzenoss, i did not mean that
[10-Sep-2009 15:35:54] <metalklesk> i mean, i have downloaded the source code from the web
[10-Sep-2009 15:36:07] <jb> yo rocket.. you here?
[10-Sep-2009 15:36:08] <metalklesk> and edited with a text editor
[10-Sep-2009 15:36:27] <Apocalipse> push ok... thx matt
[10-Sep-2009 15:36:28] <metalklesk> now, how can i made a zenpack with that
\n[10-Sep-2009 15:36:29] <chemist> the process is called Meeting\nand at any one time there can be several Meeting processes running
[10-Sep-2009 15:36:35] <mrayzenoss> metalklesk: trac-zenpacks/wiki/BuildingZenPacks
[10-Sep-2009 15:37:13] <djack> hmmm
[10-Sep-2009 15:37:21] <metalklesk> wooooooowww thanks !!
\n[10-Sep-2009 15:37:42] <chemist> they are all numbered from 0 to\n18 and normally I remodel the device so it picks up the Meeting process\nthat is taking place
[10-Sep-2009 15:37:47] <djack> how do i suppress old processes ?
[10-Sep-2009 15:37:47] <chemist> this works fine
\n[10-Sep-2009 15:38:40] <chemist> but now it keeps reporting that\n2 Meeting processes are not running, despite having remodelled the\ndevice and those two processes not showing in the OS tab for thedevice
[10-Sep-2009 15:39:17] <IoNu|z> how is the HW Model getting pooled in zenoss 2.4.5
[10-Sep-2009 15:39:18] <IoNu|z> ?
[10-Sep-2009 15:39:50] <djack> rmatte awake ?
[10-Sep-2009 15:40:15] <rmatte> yeh, I'm back
[10-Sep-2009 15:40:17] <chemist> how can I get rid of the 2 phantom processes?
[10-Sep-2009 15:40:27] <rmatte> what do you mean by "old processes"?
[10-Sep-2009 15:40:40] <rmatte> ok, they aren't in the list but they're still being monitored?
[10-Sep-2009 15:40:42] <djack> well we have java process running and the Xmm value changed
[10-Sep-2009 15:40:57] <rmatte> did you push changes to the collector for that device?
[10-Sep-2009 15:41:02] <chemist> yes
[10-Sep-2009 15:41:11] <rmatte> ok, go in to zendmd and try:
[10-Sep-2009 15:41:15] <rmatte> reindex()
[10-Sep-2009 15:41:17] <rmatte> commit()
[10-Sep-2009 15:41:26] <rmatte> I find that helps to sort out ghost process issues
\n[10-Sep-2009 15:41:29] <djack> i have a process monitor java\ndefined and he sees the new and keeps bugging me about the old process\n....
[10-Sep-2009 15:41:30] <rmatte> it has worked for me in the past
[10-Sep-2009 15:41:56] <djack> same process only one parameter is changed
[10-Sep-2009 15:43:02] <rmatte> but you did remove the old process from the list right?
[10-Sep-2009 15:43:24] <rmatte> you need to make sure the old process is not in the list of processes for that device
[10-Sep-2009 15:43:27] <rmatte> if it is delete it
[10-Sep-2009 15:43:37] <rmatte> then push changes to the collector (Manage -> Push Changes)
[10-Sep-2009 15:43:43] <rmatte> then as the zenoss user:
[10-Sep-2009 15:43:43] <djack> yeah
[10-Sep-2009 15:43:44] <rmatte> zendmd
[10-Sep-2009 15:43:48] <rmatte> reindex()
[10-Sep-2009 15:43:51] <rmatte> commit()
[10-Sep-2009 15:43:56] <rmatte> the press ctrl-d to exit
[10-Sep-2009 15:44:02] <djack> and then recreate?
[10-Sep-2009 15:44:14] <rmatte> why would you have to recreate?
[10-Sep-2009 15:44:20] <rmatte> if you remodel it should find the new process
[10-Sep-2009 15:44:29] <rmatte> have you remodeled?
[10-Sep-2009 15:45:08] <djack> yeah
\n[10-Sep-2009 15:45:51] <rmatte> k, so the new process is in the\nlist, the old process isn't, and you've pushed changes to the\ncollector, but you're still getting alerts for the process
[10-Sep-2009 15:46:06] <rmatte> if that's the case, execute those commands in zendmd
[10-Sep-2009 15:46:10] <chemist> that's correct
[10-Sep-2009 15:46:44] <rmatte> the index for processes seems to get messed up more often than anything
[10-Sep-2009 15:52:20] <jb> ARGH
[10-Sep-2009 15:52:31] <jb> i always forget you need to reinstall zenpacks when you add new object classes.
[10-Sep-2009 15:52:48] <djack> grmbl
[10-Sep-2009 15:52:52] <metalklesk> i hate to be a newbie
[10-Sep-2009 15:52:57] <jb> there goes that hour.
[10-Sep-2009 15:52:59] <djack> they still pop up
[10-Sep-2009 15:54:08] <chemist> rmatte: ok, i tried that http://pastebin.ca/1561485
\n[10-Sep-2009 15:54:23] <Apocalipse> now... after i modeled core\nrouter... and other routers... what do i need to do to link locations?
\n[10-Sep-2009 15:55:11] <metalklesk> make\nZenPacks.community.Mandriva, There is no rule to build the target\n'ZenPacks.community.Mandriva'. Stop
[10-Sep-2009 15:55:19] <metalklesk>
[10-Sep-2009 15:55:49] <mrayzenoss> metalklesk: I assume the directory is ZenPacks.community.Mandriva
[10-Sep-2009 15:55:54] <mrayzenoss> and the setup.py contains that
[10-Sep-2009 15:55:59] <metalklesk> yes
[10-Sep-2009 15:56:19] <mrayzenoss> and you're running make from the directory above that?
[10-Sep-2009 15:56:25] <metalklesk> yes
[10-Sep-2009 15:56:44] <djack> going to sleep
[10-Sep-2009 15:56:54] <metalklesk> in root directory, where the setup.py is
[10-Sep-2009 15:57:07] <mrayzenoss> no, need to go up 1 more directory
[10-Sep-2009 15:57:22] <rmatte> chemist: hmmm
[10-Sep-2009 15:57:39] <metalklesk> in tags ?
\n[10-Sep-2009 15:57:42] <mrayzenoss> metalklesk: see how\ntrac-zenpacks/browser/zenpacks is setup?
[10-Sep-2009 15:57:48] <rmatte> mrayzenoss: http://pastebin.ca/1561485 <--- Are those errors trivial?
\n[10-Sep-2009 15:58:57] <mrayzenoss> if your layout is like that,\nyou can make all of them with 'make' or just 1, with 'make\nZenPacks.community.Mandriva'
[10-Sep-2009 15:59:15] <mrayzenoss> rmatte: weird
[10-Sep-2009 15:59:34] <rmatte> mrayzenoss: those aren't errors that I have, chemist has those
[10-Sep-2009 15:59:53] <rmatte> I've never seen anything like that before
[10-Sep-2009 15:59:54] <chemist> I'm also getting an error in the zenoss console now when I try to go to that device
[10-Sep-2009 16:00:23] <metalklesk> aaahh but wait, i dont have the files build-egg.sh and GNUmakefile
[10-Sep-2009 16:00:24]\n<mrayzenoss> rmatte:\nhttp://search.zenoss.com/search?q=%22ERROR+ZODB.Connection+Couldn%27t+load+state%22&btnG=Search&entqr=0&ud=1&sort=date%3AD%3AL%3Ad1&output=xml_no_dtd&oe=UTF-8&ie=UTF-8&client=default_frontend&proxystylesheet=default_frontend&site=default_collection
[10-Sep-2009 16:00:33]\n<mrayzenoss> err, chemist:\nhttp://search.zenoss.com/search?q=%22ERROR+ZODB.Connection+Couldn%27t+load+state%22&btnG=Search&entqr=0&ud=1&sort=date%3AD%3AL%3Ad1&output=xml_no_dtd&oe=UTF-8&ie=UTF-8&client=default_frontend&proxystylesheet=default_frontend&site=default_collection
[10-Sep-2009 16:00:38] <rmatte> lol
[10-Sep-2009 16:01:28] <rmatte> mrayzenoss: doesn't look like either of those links help
[10-Sep-2009 16:01:42] <rmatte> oh wait, the first one might
[10-Sep-2009 16:02:39] <mrayzenoss> metalklesk: trac-zenpacks/wiki/SVNInstructions
[10-Sep-2009 16:02:53] <rmatte> I never understand how people's Zope databases get so screwed up
[10-Sep-2009 16:03:09] <mrayzenoss> metalklesk: trac-zenpacks/wiki/BuildingZenPacks
[10-Sep-2009 16:03:50] <chemist> http://pastebin.ca/1561496 <<< that is the error on the console
[10-Sep-2009 16:03:51]\n<rmatte> mrayzenoss: actually, the solution in the first link\ndoesn't help and the second link has no solution posted
[10-Sep-2009 16:04:34] <rmatte> chemist: are you only getting that error with that one device?
[10-Sep-2009 16:04:46] <chemist> I cant browse to any device now
[10-Sep-2009 16:05:19] <rmatte> eugh, yeh, something was screwed with your Zope database and the reindex appears to have agitated it
[10-Sep-2009 16:05:41] <chemist> actually, it appears to be only the linux devices
[10-Sep-2009 16:05:49] <chemist> windows are ok
[10-Sep-2009 16:06:05] <rmatte> Yeh, looks like the problem is in /zport/dmd/Devices/Server/Linux/devices
[10-Sep-2009 16:06:07]\n<mrayzenoss> chemist: try this\nhttp://www.zenoss.com/Members/cluther/fix-a-broken-devicesearch-or-componentsearch-catalog/
[10-Sep-2009 16:06:08] <rmatte> from the looks of it
[10-Sep-2009 16:06:41] <mrayzenoss> found that in http://dev.zenoss.org/trac/ticket/5233
[10-Sep-2009 16:06:59] <mrayzenoss> I usually dig into search.zenoss.com and dev.zenoss.org for finding stuff
[10-Sep-2009 16:07:12] <rmatte> yeh, that script is worth a shot
[10-Sep-2009 16:09:11] <chemist> not in this case
[10-Sep-2009 16:10:25] <rmatte> why?
[10-Sep-2009 16:10:34] <rmatte> you obviously have some sort of conflict, possibly caused by old items
[10-Sep-2009 16:11:53] <chemist> well, I ran the script and I still cant access my linux devices
[10-Sep-2009 16:12:04] <rmatte> you did a commit at the end?
[10-Sep-2009 16:13:29] <chemist> yes
[10-Sep-2009 16:14:09] <chemist> will I have to delete the devices and rediscover the?
[10-Sep-2009 16:14:16] <chemist> *them
[10-Sep-2009 16:14:35] <rmatte> it's possible
[10-Sep-2009 16:14:55] <chemist> is this likely to happen often?
[10-Sep-2009 16:14:57] <rmatte> what if you try moving a device to a different device class
[10-Sep-2009 16:15:01] <rmatte> are you then able to view it?
[10-Sep-2009 16:15:05] <metalklesk> ok, i got it now
[10-Sep-2009 16:15:17] <metalklesk> but the problen now is python version !!
[10-Sep-2009 16:15:19] <rmatte> I've never had anything like that happen, and I manage like 8 different Zenoss boxes
[10-Sep-2009 16:15:32] <rmatte> metalklesk: python version for what?
[10-Sep-2009 16:16:57] <chemist> no, that doesn't work either
[10-Sep-2009 16:17:09] <chemist> it tells me its moved it but it hasn't
[10-Sep-2009 16:17:28] <rmatte> ah, it could be a problem with the devices or with the actual organizer itself
[10-Sep-2009 16:17:34] <metalklesk> make ZenPacks.community.Mandriva
[10-Sep-2009 16:17:34] <rmatte> it's worth a shot removing them and re-adding them
[10-Sep-2009 16:17:42] <rmatte> test it with just 1 to begin with
[10-Sep-2009 16:17:54] <metalklesk> which: no python in (/bin)
\n[10-Sep-2009 16:17:54] <metalklesk> which: no python2.4 in\n(/usr/bin:/bin:/usr/local/bin:/usr/X11R6/bin/:/usr/games:/usr/lib/qt4/bin:/home/metalklesk/bin)
\n[10-Sep-2009 16:18:23] <mrayzenoss> metalklesk: you'll need to\nupdate your .bash_profile to include the Zenoss-provided Python
[10-Sep-2009 16:18:31] <chemist> 'zenoss error has occurred' when trying to delete
[10-Sep-2009 16:18:44] <metalklesk> yes, i jus saw that in build-egg.sh jejeje
[10-Sep-2009 16:18:47] <mrayzenoss> export ZENHOME=/usr/local/zenoss
[10-Sep-2009 16:18:47] <mrayzenoss> export PYTHONPATH=${ZENHOME}/zenoss/lib/python
\n[10-Sep-2009 16:18:47] <mrayzenoss> export\nPATH=$HOME/bin:$ZENHOME/bin:$ZENHOME/python/bin:/usr/local/bin:/usr/local/sbin:/opt/local/bin:/opt/local/sbin:/bin:/usr/bin:/usr/bin/X11:/sbin:/usr/sbin:/usr/X11R6/bin/
[10-Sep-2009 16:18:55] <metalklesk> ZENHOME is not defined
[10-Sep-2009 16:19:05] <mrayzenoss> that's from my .bash_profile
[10-Sep-2009 16:19:12] <metalklesk> thanks
[10-Sep-2009 16:19:30] <rmatte> chemist: wow
[10-Sep-2009 16:19:38] <rmatte> chemist: try restarting Zenoss, then try to remove them
[10-Sep-2009 16:20:31] <chemist> i'm on it
[10-Sep-2009 16:30:52] <rmatte> I'm out of here shortly
\n[10-Sep-2009 16:31:58] <home_run_dfc> is pynetsnmp available for\ndownload as a tarball on its own somewhere? the first google result\nfor it is on sourceforge, but i don't see any files for download
[10-Sep-2009 16:32:10] <metalklesk> i guess it works !!
[10-Sep-2009 16:32:19] <metalklesk> should i do a zenpack as root user ?
[10-Sep-2009 16:32:26] <metalklesk> *build
[10-Sep-2009 16:32:35] <chemist> rmatte: ok, I can now go to each location and delete the linux devices
[10-Sep-2009 16:32:53] <chemist> this is going to be fun
[10-Sep-2009 16:33:16] <metalklesk> ZenPacks.community.Mandriva-1.1-py2.4.egg, 20 kb
[10-Sep-2009 16:33:59] <rmatte> home_run_dfc: looks like the project may have been discontinued
[10-Sep-2009 16:34:21] <home_run_dfc> rmatte: isn't it the snmp library used by zenoss?
[10-Sep-2009 16:34:27] <rmatte> home_run_dfc: have you checked your package manager?
[10-Sep-2009 16:35:29] <rmatte> zenoss seems to use netsnmp, pynetsnmp, and twistedsnmp
\n[10-Sep-2009 16:36:19] <home_run_dfc> rmatte: zenoss has\npynetsnmp in their SVN --\nhttp://dev.zenoss.com/trac/browser/trunk/pynetsnmp
[10-Sep-2009 16:36:39] <rmatte> then grab it from there, it's probably just the last version that was publicly available
[10-Sep-2009 16:36:55] <home_run_dfc> rmatte: i was hoping for a tarball
\n[10-Sep-2009 16:36:56] <rmatte> generally if files aren't\navailable on a project's actual site, it indicates that the project is\nno longer active
[10-Sep-2009 16:37:38]\n<home_run_dfc> rmatte: doesn't zenoss consider it a problem that\nthey're using a no longer maintained library for what i would consider\na core component of their product?
[10-Sep-2009 16:38:32] <rmatte> http://archive.ubuntu.com/ubuntu/pool/universe/p/pynetsnmp/pynetsnmp_0.28.8.orig.tar.gz
[10-Sep-2009 16:38:36] <rmatte> that might work for you
\n[10-Sep-2009 16:38:52] <home_run_dfc> rmatte: i'm looking to get\nstarted coding some net management tools in python & was looking\nfor an snmp module to use & i figured zenoss would probably use a\ngood one
[10-Sep-2009 16:39:01] <chemist> so it looks like my zenoss box is hosed
[10-Sep-2009 16:39:14] <home_run_dfc> but i don't want something unmaintained
[10-Sep-2009 16:39:15] <rmatte> home_run_dfc: you do realize that Zenoss development started back in 1999 right?
\n[10-Sep-2009 16:39:28] <rmatte> home_run_dfc: if they were coding\nit starting right now they'd be using a lot different stuff
[10-Sep-2009 16:39:51] <rmatte> home_run_dfc: I doubt very much that they'd be using Zope for instance
[10-Sep-2009 16:39:52] <chemist> I can no longer view linux devices and I can't discover any new devices
[10-Sep-2009 16:40:18] <rmatte> chemist: yeh, sounds hosed, no idea how your Zope db got that corrupt
\n[10-Sep-2009 16:40:35] <home_run_dfc> rmatte: well then i was\nhoping someone from zenoss could offer me a suggestion on what they\nwould use for a python snmp library if what they're currently using\nisn't it
[10-Sep-2009 16:41:01] <home_run_dfc>\nconsidering they switched to pynetsnmp for version 2.0, i don't think\nthat decision goes all the way back to 1999
[10-Sep-2009 16:41:09] <chemist> I don't know much about zope, can I 'drop' the db?
[10-Sep-2009 16:41:48] <rmatte> 2.0 is a ways back
\n[10-Sep-2009 16:42:57] <home_run_dfc> rmatte: according to what i\nfound, 2.0 was released mid 2007 (obviously development of it started\nearlier than that, but i would guess maybe late 2004/early 2005)
[10-Sep-2009 16:43:08] <rmatte> hmmm
\n[10-Sep-2009 16:43:10] <rmatte> pynetsnmp is a replacement for\nthe various Python bindings provided by PySNMP* implementations\n(available as the Debian packages python-pysnmp*).
[10-Sep-2009 16:43:43] <rmatte> maybe it got integrated with pysnmp?
[10-Sep-2009 16:45:12] <rmatte> it'd kind of make sense if it did
\n[10-Sep-2009 16:47:20] <home_run_dfc> looks like pysnmp is the\nonly full-featured & maintained python snmp library at this point
[10-Sep-2009 16:47:28] <home_run_dfc> i guess that's what i'll try out
[10-Sep-2009 16:47:32] <home_run_dfc> rmatte: thx for the help
[10-Sep-2009 16:47:35] <rmatte> np
[10-Sep-2009 16:48:13] <chemist> rmatte: is there a way to recreate the zope db?
[10-Sep-2009 16:48:32] <metalklesk> it does not work
[10-Sep-2009 16:48:52] <metalklesk> i have edited the suse zenpack code to mandriva but no software was listed
[10-Sep-2009 16:49:09] <rmatte> chemist: not that I know of other than reinstalling Zenoss
[10-Sep-2009 16:49:30] <rmatte> chemist: the zope database is stored in $ZENHOME/etc/Data.fs
[10-Sep-2009 16:49:42] <chemist> sounds like my weekend off is no more
[10-Sep-2009 16:50:13] <rmatte> sorry...
[10-Sep-2009 16:50:19] <rmatte> chemist: the zope database is stored in $ZENHOME/var/Data.fs
[10-Sep-2009 16:50:24] <rmatte> not etc
[10-Sep-2009 16:51:03] <rmatte> if you see any *.zec files in $ZENHOME/var you could try deleting them then restarting Zenoss
[10-Sep-2009 16:51:06] <rmatte> but it's a long shot
[10-Sep-2009 16:52:58] <rmatte> well, I'm done for the day, I'll be back tomorrow
[10-Sep-2009 16:53:00] <rmatte> later guys
\n[10-Sep-2009 18:16:54] <simonpt> lo... i've got the zenoss vmware\nappliance... when i try to update raa it gives me this error: Failed:\n(Unknown Exception Type) MigrateError Cannot determine which group to\nuse as the top level group
[10-Sep-2009 18:35:14] <ironpaw> morning
[10-Sep-2009 19:01:24] <simonpt> how would one start to troubleshoot why you couldn't collet WMI data from a windows server?
[10-Sep-2009 19:14:39] <ironpaw> um
[10-Sep-2009 19:14:44] <ironpaw> check your zencommand logs
[10-Sep-2009 19:15:04] <ironpaw> you can run a zenwmi command from the shell to confirm connectivity/authentication details
\n[10-Sep-2009 19:19:43] <simonpt> looks like i had to install the\nwindows components... good to know about zencommand too thanks
[10-Sep-2009 19:19:50] <ironpaw> np
[10-Sep-2009 19:19:53] <ironpaw> get it working
[10-Sep-2009 19:20:06] <ironpaw> you will also need to configure your windows box to allow remote WMI queries
[10-Sep-2009 19:20:16] <ironpaw> got the WMIPerformance zenpacks?
[10-Sep-2009 19:24:51] <simonpt> i have the wmidatasource
[10-Sep-2009 19:24:55] <simonpt> zenpack
\n[10-Sep-2009 19:36:43] <simonpt> keep getting Event code: 3005\nEvent message: An unhandled exception has occurred. on the wmi device
[10-Sep-2009 19:36:48] <simonpt> the component is ASP.net
[10-Sep-2009 19:39:40] <ironpaw> hmm
[10-Sep-2009 19:39:49] <ironpaw> from zenoss or the eventlog in windows?
[10-Sep-2009 19:39:58] <ironpaw> sorry had to do multiple reboots
[10-Sep-2009 19:40:17] <simonpt> zenoss
[10-Sep-2009 19:41:13] <ironpaw> hmm just trying to find the page with theinfo
[10-Sep-2009 19:41:16] <ironpaw> please hold
[10-Sep-2009 19:41:27] * ironpaw hums some on-hold tunes
[10-Sep-2009 19:42:37] <ironpaw> To test WMI, find the wmic command (usually in $ZENHOME/bin/wmic), and run this command:
[10-Sep-2009 19:42:37] <ironpaw> wmic -U domain/administrator%password //host "Select * from Win32_ComputerSystem"
[10-Sep-2009 19:42:54] <ironpaw> can you run that on your device and does it return stuff?
[10-Sep-2009 19:43:36] <simonpt> yea that works fine
[10-Sep-2009 19:44:53] <ironpaw> hmm
[10-Sep-2009 19:45:11] <ironpaw> zenperfwmi eventlog give you any more details?
[10-Sep-2009 19:45:34] <simonpt> no... it's not counting the one server i have set to monitor
[10-Sep-2009 19:45:51] <ironpaw> in zneoss , try pushing changes to the device
[10-Sep-2009 19:45:54] <ironpaw> not sure if that'll work
[10-Sep-2009 19:45:59] <ironpaw> or just give it some time to re-poll
[10-Sep-2009 19:46:48] <simonpt> push changes... where is that? is it same as remodel?
[10-Sep-2009 19:47:09] <simonpt> found it
\n[10-Sep-2009 19:52:32] <simonpt> looks like it's just reporting\nthe 3005 error from an asp.net message in the windows event log
[10-Sep-2009 22:17:00] <ironpaw> *yawn*
[11-Sep-2009 00:00:46] [disconnected at Fri Sep 11 00:00:46 2009]
[12-Sep-2009 00:00:46] [connected at Sat Sep 12 00:00:46 2009]
[12-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[12-Sep-2009 15:05:37] <chemist> good evening
[12-Sep-2009 15:07:01] <chemist> how can I get zenoss to run a script on a remote server and report the result of the script
[12-Sep-2009 15:07:10] <chemist> if its at allpossible
[12-Sep-2009 18:48:32] <metalklesk> hi
[12-Sep-2009 19:55:15] rhettardo_ is now known as rhettardo
[13-Sep-2009 00:00:46] [disconnected at Sun Sep 13 00:00:46 2009]
[13-Sep-2009 00:00:46] [connected at Sun Sep 13 00:00:46 2009]
[13-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[13-Sep-2009 06:35:32] kevin7kal_ is now known as kevin7kal
\n[13-Sep-2009 12:35:54] <chemist> I run several processes called\n'Meeting (tk0)', 'Meeting (tk1)' etc up to tk17, they run at different\ntimes so I model the device every time the tk numbe rchangesw
\n[13-Sep-2009 12:36:27] <chemist> but despite remodelling, zenoss\nstill reports as failed the processes that are not running
\n[13-Sep-2009 12:37:33] <chemist> I have tried 'Push Changes'\nafter remodelling but it still shows me these 'ghost' processes
[13-Sep-2009 12:37:46] <chemist> I also tried reindexing the db
[13-Sep-2009 12:55:49] etank is now known as elake
[13-Sep-2009 12:56:35] elake is now known as etank
[13-Sep-2009 13:00:42] etank is now known as elake
[13-Sep-2009 13:00:53] elake is now known as etank
[13-Sep-2009 13:01:19] etank is now known as roninx341
[13-Sep-2009 13:01:33] roninx341 is now known as etank
[13-Sep-2009 14:02:40] <chemist> how can I make zenoss run a script on a remote linux box and report the outcome?
[13-Sep-2009 14:02:49] <chemist> if it's at all possible
[13-Sep-2009 14:03:15] <elake> chemist: yes it is
[13-Sep-2009 14:03:31] <elake> the output must be in a way that zenoss expects it though
[13-Sep-2009 14:03:44] <chemist> elake: can you give me some pointers?
[13-Sep-2009 14:04:04] <elake> chemist: one sec. i need to connect to the server and i will get you something
[13-Sep-2009 14:04:09] <chemist> ta
[13-Sep-2009 14:05:12] <elake> the way i did it was in a graph
[13-Sep-2009 14:05:33] <elake> in the device class that i want to see it in i created a new template
[13-Sep-2009 14:05:59] <elake> so for me i did it in /Devices/Server/Linux
[13-Sep-2009 14:06:10] <elake> i wanted to graph the ntp stratum of my servers
[13-Sep-2009 14:06:23] <elake> i created the template and named it ntpstratum
[13-Sep-2009 14:06:51] <elake> then created a data source named ntpstratum with the command template looking like this
[13-Sep-2009 14:06:58] <elake> /usr/sbin/ntpdc -n -c sysinfo | awk '/stratum/ { printf("|stratum=%s", $$2); }'
[13-Sep-2009 14:07:17] <elake> it will use ssh of the device to run the command
[13-Sep-2009 14:07:21] <elake> the return will look like
[13-Sep-2009 14:07:30] <elake> |stratum=4
[13-Sep-2009 14:07:41] <elake> the data point should be called stratum
[13-Sep-2009 14:07:52] <elake> must match the datasource output name
[13-Sep-2009 14:08:12] <elake> then add the datapoint to a graph and add any thresholds that you desire
[13-Sep-2009 14:08:41] <elake> this of course all assumes that you are wanting to run a command on a remote linux machine
[13-Sep-2009 14:08:59] <chemist> yes, a script in fact
[13-Sep-2009 14:09:38] <elake> the steps above are what i followed to get it to work
[13-Sep-2009 14:10:12] <chemist> Am I correct in assuming that in '|stratum=4', 4 is the severity?
[13-Sep-2009 14:10:25] <elake> chemist: yes
[13-Sep-2009 14:10:39] <chemist> cool, I will give that a go
[13-Sep-2009 14:10:48] <elake> well |stratum=4 is the output of the script
[13-Sep-2009 14:11:48] <elake> make sure the datapoint matches the |foo exactly
[13-Sep-2009 14:11:51] <elake> even in case
[13-Sep-2009 14:12:04] <chemist> ok
[13-Sep-2009 14:12:09] <elake> so datapoint Foo is not the same as a script returning |foo=x
[13-Sep-2009 14:13:04] <Snyper> all i ever see is how to make graphs but is there a way to get hardware info ?
[13-Sep-2009 14:13:21] <Snyper> get i mean use hardware data
\n[13-Sep-2009 14:14:50] <Snyper> i cant use snmp so im left trying\nto figure out how to add wmi support for hardware info.. i have all the\ninfo ready togo but cant figure out how to feed zenoss
[13-Sep-2009 14:19:21] <elake> Snyper: windows servers?
[13-Sep-2009 14:19:29] <Snyper> windows period
[13-Sep-2009 14:19:55] <elake> you have the wmi stuff installed on the windows servers right
[13-Sep-2009 14:20:03] <elake> and a user set in the zProperties
[13-Sep-2009 14:20:51] <Snyper> i have no problem pulling cpu, ram ..... stats for the graphs
[13-Sep-2009 14:21:10] <Snyper> its just hardware info, like os name, harddrive info.... ect
[13-Sep-2009 14:22:12] <Snyper> which i can pull from a script with no issue
[13-Sep-2009 14:22:15] <elake> you get some hardware info on the os tab
[13-Sep-2009 14:22:31] <Snyper> os tab shows only services
[13-Sep-2009 14:22:53] <elake> should see nics and filesystems too
[13-Sep-2009 14:23:27] <Snyper> non of that shows up
[13-Sep-2009 14:24:31] <elake> are you using core?
[13-Sep-2009 14:24:37] <Snyper> yes
[13-Sep-2009 14:26:01] <elake> there was a community plugin called wmi2 or something
[13-Sep-2009 14:26:13] <elake> it may help you get more information from the boxes
[13-Sep-2009 14:26:34] <elake> http://www.zenoss.com/community/projects/zenpacks/ <-- big list
[13-Sep-2009 14:26:51] <elake> if it is an hp server then there are some packs specific to them
[13-Sep-2009 14:27:00] <Snyper> all i get is graphs
[13-Sep-2009 14:27:16] <Snyper> i could give 2 sh*t about a graph
[13-Sep-2009 14:28:19] <elake> what kind of hardware info are you wanting to get?
[13-Sep-2009 14:28:33] <Snyper> generic windows box
[13-Sep-2009 14:29:01] <Snyper> most of those things for dell servers and hp servers seems to use snmp
[13-Sep-2009 14:29:38] <elake> curious. why cant you use snmp?
[13-Sep-2009 14:30:03] <Snyper> company does not allow it
[13-Sep-2009 14:30:11] <elake> i see
[13-Sep-2009 14:30:13] <Snyper> auditor go crazy
[13-Sep-2009 14:30:27] <Snyper> and they are too stupid to understand what v3 is
[13-Sep-2009 14:30:36] <elake> if only windows supported snmp v3
[13-Sep-2009 14:30:48] <Snyper> i have an addon to allow v3
[13-Sep-2009 14:30:59] <elake> got a link?
[13-Sep-2009 14:31:12] <elake> we are interested in using v3 on our servers
[13-Sep-2009 14:31:20] <elake> we have it working on our RHEL boxes
[13-Sep-2009 14:31:38] <Snyper> http://marksw.com/snmpv3agent/windowsagent.html
[13-Sep-2009 14:31:54] <Snyper> works great on my test env.... its just i cant use it
[13-Sep-2009 14:32:21] <elake> nice
[13-Sep-2009 14:33:30] <elake> Snyper: i wish i had more help to give but i am still new to zenoss myself
[13-Sep-2009 14:33:56] <Snyper> just annoying when you have all the data and no where to feed it to zenoss
[13-Sep-2009 14:34:51] <elake> there should be admins on tomorrow morning
[13-Sep-2009 14:35:12] <elake> they could offer more help
[13-Sep-2009 14:59:44] etank is now known as elake
[13-Sep-2009 22:20:59] <ironpaw> wasabi?
[13-Sep-2009 23:24:19] <rhett> horseradish
[14-Sep-2009 00:00:46] [disconnected at Mon Sep 14 00:00:46 2009]
[14-Sep-2009 00:00:46] [connected at Mon Sep 14 00:00:46 2009]
[14-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[14-Sep-2009 04:05:28] <chemist> morning all
[14-Sep-2009 04:06:20]\n<chemist> how do I stop zenoss from telling me that every device\nbehind a router has gone down, when only the router is down?
[14-Sep-2009 05:28:43] <JMan> morning
[14-Sep-2009 05:31:00] <JMan> anyone awake yet?
[14-Sep-2009 05:33:50] <rhettardo> i was going to say yes but oh well
[14-Sep-2009 06:34:22] <n215> does anybody have experience with zenoss and asa
\n[14-Sep-2009 07:45:23] <chemist> how do I stop zenoss from\ntelling me that every device behind a router has gone down, when only\nthe router is down?
[14-Sep-2009 08:10:00] <ckrough> chemist: It sounds like your outage paths are not working.
[14-Sep-2009 08:10:00] <ckrough> chemist: do some hunting in the forums for
[14-Sep-2009 08:10:00] <ckrough> for 'path outage'
[14-Sep-2009 08:10:00] <ckrough> chemist: also http://blog.zenoss.com/2008/08/26/tip-of-the-month-layer-3-dependency-checker/
[14-Sep-2009 08:10:00] <n215> hi
[14-Sep-2009 08:10:00] <n215> i get missing RRD FILE when monitoring cisco asa
[14-Sep-2009 08:10:00] <ckrough> n215: can you snmpwalk it manually from the zen server?
[14-Sep-2009 08:10:00] <chemist> ckrough: thanks, will have a look
[14-Sep-2009 08:10:00] <n215> ckrough i get some information using snmp
[14-Sep-2009 08:10:00] <n215> snmpwalk i mean
[14-Sep-2009 08:12:00] <ckrough> n215: try doing an snmpget for the datapoint that isnt graphing
[14-Sep-2009 08:12:00] ckrough is now known as coofamani
[14-Sep-2009 08:12:00] coofamani is now known as ckrough
[14-Sep-2009 08:12:08] <n215> cpu rrd works ok now
[14-Sep-2009 08:12:13] <n215> but memory rrd still missing
\n[14-Sep-2009 08:36:45] <ckrough> Anyone know where zenoss gets\nthe values for 'build-in' data sources like those used in the\nPerformanceConf template?
[14-Sep-2009 08:38:45] <ckrough> 'built-in;
[14-Sep-2009 08:40:11] <rmatte> ckrough: collector plugins
[14-Sep-2009 08:42:11] <ckrough> thanks
\n[14-Sep-2009 08:42:11] <ckrough> trying to figure out why my web\ninterfaces sometimes bog down when zenhub gets busy. running 7\nzenhubworkers on an 8 core box, maybe have to drop to 6 workers
[14-Sep-2009 09:04:20] <ckrough> mrayzenoss: do you *have* to sign up to watch that?
[14-Sep-2009 09:04:54] <mrayzenoss> umm... I think webex emails you the invite
[14-Sep-2009 09:04:59] <mrayzenoss> with the URL
[14-Sep-2009 09:05:48] <mrayzenoss> too much webex spam?
[14-Sep-2009 09:06:33] <ckrough> heh, not that I recall. was just curious.
[14-Sep-2009 09:09:25] <mrayzenoss> if there's a good turnout, we may have Jane do another session if you can't make it
[14-Sep-2009 09:09:25] <rmatte> yeh, you need to sign up
[14-Sep-2009 09:10:07] <rmatte> I hope there's a good turnout, it's worth it
[14-Sep-2009 09:10:07] <mrayzenoss> yeah, we've got 2 more sessions already in the pipe after this one
[14-Sep-2009 09:10:07] <mrayzenoss> I'll do one next month on the 2.5 release and the beta
[14-Sep-2009 09:10:07] <rmatte> oh, I think you just sign up for the event while it's running, not in advance?
[14-Sep-2009 09:10:07] <mrayzenoss> no, it should be in advance
[14-Sep-2009 09:10:07] <rmatte> oh, I see
[14-Sep-2009 09:10:07] <rmatte> I just wasn't looking at the right spot
[14-Sep-2009 09:10:07] <rmatte> "You are now registered for the event: Zenoss Community Webinar Series: Zenoss Event Architecture"
[14-Sep-2009 09:10:07] <rmatte> woohoo
[14-Sep-2009 09:10:11] <ckrough> /win
[14-Sep-2009 09:29:24] straterr1 is now known as straterra
[14-Sep-2009 09:31:04]\n<adelbul> hello, i installed zenplugins Zenoss-Plugins-2.0.4 on a\nopensolaris system when i execute zenplugin.py the result is :
\n[14-Sep-2009 09:31:16] <adelbul> root@srvnas03:~# zenplugin.py \nTraceback (most recent call last): File "/usr/bin/zenplugin.py", line\n49, in ? from zenoss.plugins.common import ApplicationError\nImportError: No module named common
[14-Sep-2009 09:32:07] <adelbul> i don't understand.
[14-Sep-2009 09:32:25] <adelbul> common.py and common.pyc are in the good directory.
[14-Sep-2009 09:32:42] <ckrough> try it as the zenoss user
[14-Sep-2009 09:32:55] <ckrough> 'su - zenoss'
[14-Sep-2009 09:33:10] <adelbul> ckrough: same result.
\n[14-Sep-2009 09:33:25] <adelbul> zenoss@srvnas03:~$ zenplugin.py \nTraceback (most recent call last): File "/usr/bin/zenplugin.py", line\n49, in ? from zenoss.plugins.common import ApplicationError\nImportError: No module named common
[14-Sep-2009 09:33:26] <ckrough> meh
\n[14-Sep-2009 09:33:41] <mrayzenoss> adelbul: You should take a\nlook at http://www.zenoss.com/community/projects/zenpacks/opensolaris\ninstead of doing it with a zenplugin
[14-Sep-2009 09:34:06] <mrayzenoss> adelbul: zenplugins are deprecated as of 2.4
\n[14-Sep-2009 09:34:56] <adelbul> zenoss@srvnas03:~$ ls -l\n/usr/lib/python2.4/site-packages/zenoss/plugins/common.py* -rw-r--r-- 1\nroot root 6435 2007-07-09 16:49\n/usr/lib/python2.4/site-packages/zenoss/plugins/common.py -rw-r--r-- 1\nroot root 6360 2009-09-14 14:33\n/usr/lib/python2.4/site-packages/zenoss/plugins/common.pyc
[14-Sep-2009 09:35:14] <adelbul> mrayzenoss: thanks i'm going to have a look.
\n[14-Sep-2009 09:36:48] <adelbul> anyway : nobody can help me for\ndebugging this "import" issue ? (juste for my personal culture ?)
[14-Sep-2009 09:37:56] <jb> yawn
[14-Sep-2009 09:40:58] <ckrough> mrayzenoss: where should we report 2.4.7 bugs? zenoss-testing or a trac ticket?
[14-Sep-2009 09:41:54] <rmatte> 2.4.7?
[14-Sep-2009 09:42:03] <ckrough> 2.5 beta
[14-Sep-2009 09:42:07] <mrayzenoss> 2.4.70
[14-Sep-2009 09:42:08] <mrayzenoss>
[14-Sep-2009 09:42:20] <rmatte> I didn't even know there was such a thing
[14-Sep-2009 09:42:22] <rmatte>
[14-Sep-2009 09:42:23] * ckrough doesnt process trailing zeros
[14-Sep-2009 09:42:35] <ckrough> rmatte: http://forums.zenoss.com/viewtopic.php?t=11001
[14-Sep-2009 09:42:35] <mrayzenoss> remind me to hire you and pay you $1 a day
[14-Sep-2009 09:42:47] <rmatte> hehe
[14-Sep-2009 09:43:01] <mrayzenoss> yeah, hit up dev.zenoss.com since the forums are locked
[14-Sep-2009 09:43:01] <ckrough> ha. well, least significant trailing zeros
[14-Sep-2009 09:43:15] <rmatte> oh, aka 2.5 beta, yeh
[14-Sep-2009 09:43:24] <jb> did the new community site not launch?
[14-Sep-2009 09:43:28] <ckrough> mrayzenoss: I see you've put up your cool hip avatar now
[14-Sep-2009 09:43:28] <mrayzenoss> it's launching
[14-Sep-2009 09:43:48] <mrayzenoss> well, Ohio LinuxFest told me they needed a picture
[14-Sep-2009 09:43:54] <mrayzenoss> and sent me a photo of me from last year
[14-Sep-2009 09:43:56] <mrayzenoss> so I used it
[14-Sep-2009 09:44:00] <rmatte> cool
[14-Sep-2009 09:44:32] <mrayzenoss> the forums are locked as we migrate to the new site
[14-Sep-2009 09:44:41] <mrayzenoss> they're getting pushed into the right places
[14-Sep-2009 09:44:49] <mrayzenoss> and we'll be polishing content all day
[14-Sep-2009 09:44:58] <mrayzenoss> if everything works as planned, we go live tonight
[14-Sep-2009 09:45:06] <rmatte> shaweet
[14-Sep-2009 09:47:04] <Stu> hey - anyone awake yet?
[14-Sep-2009 09:47:12] <ckrough> in 2.5 is 'close' equivelant to 'move to history'?
[14-Sep-2009 09:49:23] <mrayzenoss> depends on the defect
[14-Sep-2009 09:49:30] <mrayzenoss> oh, wait
[14-Sep-2009 09:49:30] <mrayzenoss> sorry, I misread that
\n[14-Sep-2009 09:49:30] <ckrough> I dont see a way to select all\nin the event console and move to hist, other than close. is that 'the\nway'?
[14-Sep-2009 09:49:35] <mrayzenoss> yeah, 'close' sends them to history
\n[14-Sep-2009 10:15:53] <Stu> hi - does anyone have any idea if\nit's possible to pull reports into an external app through a\n(semi)static URL?
[14-Sep-2009 10:24:58]\n<rmatte> Stu: not that I'm aware of, which is why I'm coding my\nown python scripts to generate reports
[14-Sep-2009 10:25:27] <rmatte> Stu: I'll most likely release what I've coded once it's done
[14-Sep-2009 10:39:30] <mrayzenoss> rmatte: Stu left
[14-Sep-2009 10:39:37] <rmatte> ah
[14-Sep-2009 10:39:37] <rmatte> I love impatient people
[14-Sep-2009 10:41:22] <skipzoid> hello
[14-Sep-2009 11:02:40] <skipzoid> bit quiet here
[14-Sep-2009 11:03:07] <rmatte> yup
[14-Sep-2009 11:03:28] <rmatte> I'm scripting so I won't be too chatty this morning
[14-Sep-2009 11:03:33] <rmatte>
[14-Sep-2009 11:03:42] <mrayzenoss> and I'm working on the Zenoss.Org launch
[14-Sep-2009 11:05:35]\n<skipzoid> hehe - is it possible to run a command from an event?\n- my sms sender runs on the command line as - sendsms <number>\n<message>
[14-Sep-2009 11:06:13] <mrayzenoss> skipzoid: http://www.zenoss.com/community/docs/zenoss-guide/2.4.3/ch07s01.html#d4e4203
[14-Sep-2009 11:07:17]\n<mrayzenoss>\nhttp://www.zenoss.com/community/docs/zenoss-dev-guide/2.4.2/apbs03.html\nhas the variables you have access to
[14-Sep-2009 11:09:24] <mrayzenoss> so you can do sendsms <number> $device is on fire!
[14-Sep-2009 11:09:33] <rmatte> skipzoid: click on Event Manager on the left, then click on the Commands tab
[14-Sep-2009 11:09:33] <rmatte> create a new command and set filter rules as you would for an alert
[14-Sep-2009 11:09:38] <rmatte> Here's an example of a command we use to create tickets in our ticketing system:
[14-Sep-2009 11:09:48]\n<rmatte> /mspfiles/scripts/create_ticket.pl -autonotify -customer\n'gen-00025' -device '${dev/id}' -deviceIP '${dev/manageIp}' -collector\n`hostname -f` -first '${evt/firstTime}' -last '${evt/lastTime}' -count\n'${evt/count}' -summary '${evt/summary}' -noteTitle 'System Monitor\nError' -note '${evt/message}' -severity '${evt/severity}' -group\n'${evt/DeviceGroups}' -impact '${evt/DevicePriority}' -component\n'${evt/component}' -queue 'Front Line'
[14-Sep-2009 11:10:01] <rmatte> all those ${} variables can be used
[14-Sep-2009 11:10:23] <skipzoid> wow didnt realise there where so many
[14-Sep-2009 11:10:31] <rmatte> there are some others too, but I think I pretty much covered the ones you'll need there
\n[14-Sep-2009 11:11:11] <skipzoid> the fire example is as much as\nwe need - it would then make the individual go look at the equipment to\ndeduce further what is the required action
\n[14-Sep-2009 11:14:19] <skipzoid> another problem i am having -\ni'm trying to monitor a process on a linux box, so within the processes\ni have defined the process as on page 67 of the guide and in my case as\nit is in the example it is ntpd, i then added this in the os part of\nthe device and it comes up green to show its running ? - then i kill it\noff so that on the next poll its not present but it still shows as\nrunning, i'm missing something here
[14-Sep-2009 11:18:26] <rmatte> skipzoid: remodel the device
[14-Sep-2009 11:18:29] <rmatte> then it'll pick up the process properly
[14-Sep-2009 11:19:21] <skipzoid> i have done that a few times - it then in the OS processes removes the item i added
\n[14-Sep-2009 11:27:25] <skipzoid> just worked it out - the\nprocess id must match the filename of the process runing and the regex\nseems to be ignored or i dont understand the syntax properly
[14-Sep-2009 11:33:49] <rmatte> skipzoid: you just type the name of the process
[14-Sep-2009 11:33:58] <rmatte> you don't need to use any fancy regex values or anything
[14-Sep-2009 11:34:16] <rmatte> if the process name is mysqld.bin then you add the process to the processes list as that
\n[14-Sep-2009 11:35:12] <skipzoid> yeah i did figure it was going\nto be that easy - i thought the name was the 'user' name of the process\nie - ntpd I wrote ntpdeamon in the regex put ntpd -
[14-Sep-2009 11:35:24] <rmatte> nope
[14-Sep-2009 11:35:35] <rmatte> hehe
[14-Sep-2009 11:58:17] <metalklesk> hi everyone
[14-Sep-2009 12:05:18] <mrayzenoss> So I've got a community question for non-English speakers, if any are lurking
[14-Sep-2009 12:05:47] <metalklesk> i speak spanish
[14-Sep-2009 12:06:02] <mrayzenoss> The new Community site has groups, for finding and discussing stuff with people of similar interests
[14-Sep-2009 12:06:33] <mrayzenoss> we were thinking about adding them for other languages, but would countries be preferred?
[14-Sep-2009 12:06:51] <mrayzenoss> ie. Spanish or Chile?
[14-Sep-2009 12:07:02] <metalklesk> Chile
[14-Sep-2009 12:07:10] <twm1010> do any of the zenpacks add DNS monitoring?
[14-Sep-2009 12:07:54] <mrayzenoss> twm1010: http://www.zenoss.com/community/projects/zenpacks/dnsmonitor
[14-Sep-2009 12:08:00] <metalklesk> i don't know if by countries would be a better choice
[14-Sep-2009 12:08:11] <twm1010> mrayzenoss: thx
[14-Sep-2009 12:08:18] <metalklesk> i do prefer language instead of country
[14-Sep-2009 12:08:48] <twm1010> im trying to hunt down a tool to help me gather DNS stats
[14-Sep-2009 12:08:48] <jb> dns monitoring how..
[14-Sep-2009 12:08:54] <jb> BIND?
[14-Sep-2009 12:08:55] <mrayzenoss> ok, I'll probably add groups like "Spanish/Español"
[14-Sep-2009 12:08:57] <metalklesk> we can get more sinergy
[14-Sep-2009 12:09:09] <twm1010> so I can see what are the common queries... etc
[14-Sep-2009 12:09:10] <jb> like queries, etc?
[14-Sep-2009 12:09:21] <jb> i graph queries per minute, etc
[14-Sep-2009 12:09:29] <jb> and use DigMon to check resolution
[14-Sep-2009 12:09:39] <twm1010> yeah, I'm thinking it will have to be something that does packet capturing
[14-Sep-2009 12:12:00] <metalklesk> what the ...., in community, someone took my nickname
[14-Sep-2009 12:12:15] <metalklesk> i don't remember to be registered
[14-Sep-2009 12:12:21] <mrayzenoss> which community?
[14-Sep-2009 12:12:25] <mrayzenoss> zenoss.org?
[14-Sep-2009 12:12:39] <metalklesk> https://community.zenoss.org
[14-Sep-2009 12:12:57] <metalklesk> i know there is no metalklesk in the entire world
[14-Sep-2009 12:13:02] <metalklesk> except me
[14-Sep-2009 12:13:03] <rmatte> hmmm
[14-Sep-2009 12:13:04] <rmatte> File "./availability.py", line 133, in ?
[14-Sep-2009 12:13:05] <rmatte> totaldown = totaldown + (leadin - startDate) + (leadout - endDate)
[14-Sep-2009 12:13:05] <rmatte> TypeError: unsupported operand type(s) for -: 'tuple' and 'float'
[14-Sep-2009 12:13:10] <tang> e
[14-Sep-2009 12:13:20] <tang> sorry, keyboard trauma
\n[14-Sep-2009 12:13:36] <mrayzenoss> metalklesk: I don't think\nyou're supposed to be able to login yet, we're putting up\nhttp://community.zenoss.org right now
[14-Sep-2009 12:13:48] <metalklesk> aahh ok
[14-Sep-2009 12:13:58] <metalklesk> jejejej
\n[14-Sep-2009 12:14:24] <metalklesk> maybe that's why the web page\nsays: Lo sentimos, ese nombre de usuario ya está registrado. Intente\ncon otro diferente.
[14-Sep-2009 12:14:24] <metalklesk> Contraseña: *
[14-Sep-2009 12:14:38] <metalklesk>
[14-Sep-2009 12:15:53] <metalklesk> i have a question
[14-Sep-2009 12:15:56] <metalklesk> about zenoss
[14-Sep-2009 12:16:03] <mrayzenoss> ok
[14-Sep-2009 12:16:29] <twm1010> go ahead
[14-Sep-2009 12:17:08]\n<metalklesk> how many disk space we can use in about 1 year of\nevent history ? let's say monitoring 100 boxes (windows, linux, cisco,\netc)
[14-Sep-2009 12:19:21] <mrayzenoss> depends\non how active your events are... but I'm sure someone here with real\nexperience can give anecdotal evidence
[14-Sep-2009 12:19:23] <metalklesk> that would be good
[14-Sep-2009 12:19:30] <metalklesk> rmatte maybe ?
[14-Sep-2009 12:19:30] * mrayzenoss install is only 10 devices, fits on a 40 gig drive
[14-Sep-2009 12:19:30] <mrayzenoss> ckrough? rmatte? How many devices and how big?
[14-Sep-2009 12:19:44] <mrayzenoss> your performance data size is mostly fixed, so that won't grow
\n[14-Sep-2009 12:19:58] <rmatte> metalklesk: we've been using a\n20gig virtual drive to monitor 370 devices since May, and we've only\nused up 11gigs of space total on the drive
[14-Sep-2009 12:20:08] <jb> yeah i use 20GB
[14-Sep-2009 12:20:11] <rmatte> and that's 11gigs including the OS and whatever else we have on it
[14-Sep-2009 12:20:17] <twm1010> what's the default location for the events DB under stack install/
[14-Sep-2009 12:20:18] <twm1010> ?
[14-Sep-2009 12:20:23] <jb> /dev/mapper/zenossvg-zenosslv
[14-Sep-2009 12:20:23] <jb> 20G 3.7G 16G 20% /opt/zenoss
[14-Sep-2009 12:20:25] <rmatte> so I think 20gig is more than sufficient provided you're rolling over every year
[14-Sep-2009 12:20:26] <jb> 200 devices
[14-Sep-2009 12:20:29] <jb> over a year
[14-Sep-2009 12:20:56] <rmatte> twm1010: the events db is mysql
[14-Sep-2009 12:20:58] <metalklesk> ok, thanks
[14-Sep-2009 12:21:14] <rmatte> twm1010: mysql -u zenoss -p
[14-Sep-2009 12:21:21] <rmatte> password is usually zenoss by default
[14-Sep-2009 12:21:22] <twm1010> looks like after 3 months mine, is 4.7gb
[14-Sep-2009 12:21:25] <rmatte> then do "connect events"
[14-Sep-2009 12:21:27] <ckrough> hard to translate it, since it depends on how busy your environment is
[14-Sep-2009 12:21:28] <rmatte> and voila
[14-Sep-2009 12:21:36] <twm1010> at least, the data file is
[14-Sep-2009 12:21:38] <jb> and if you are doing syslog
[14-Sep-2009 12:21:40] <jb> etc
[14-Sep-2009 12:22:11] <rmatte> ckrough: well, on how busy, and on how well you've tuned Zenoss to drop garbage events
[14-Sep-2009 12:22:24] <ckrough> that too
\n[14-Sep-2009 12:23:28] <twm1010> hrmm, good thing you guys\nbrought this, the volume my zenoss is installed to is 97% full
[14-Sep-2009 12:24:12] <metalklesk> im going to launch
[14-Sep-2009 12:24:14] <metalklesk> seeya
[14-Sep-2009 12:35:07]\n<ckrough> metalklesk: For a 1300 device install: ~5mil events in\n1yr history, 8.6GB event db. That's with *some* syslog stuff.
[14-Sep-2009 12:35:34] <ckrough> metalklesk: again, totally dependent on how talkative your gear will be.
\n[14-Sep-2009 12:40:39] <rmatte> yeh, if you're monitoring\nsomething like security syslogs your events db will get large very\nquickly
[14-Sep-2009 12:41:59] <ckrough> dump\nmost of that into other systems, just use zensyslog to pick up on small\nstuff that doesnt snmptrap
[14-Sep-2009 12:42:46] <ckrough> BGP changes, Trunk failures, some OSPF and spanning-tree messages
[14-Sep-2009 12:45:57] <rmatte> yeh, that's what we do
[14-Sep-2009 12:46:12] <rmatte> we have syslog-ng running between Zenoss and the monitored devices
[14-Sep-2009 12:46:23] <ckrough> same
[14-Sep-2009 12:46:26] <ckrough> you runs plunk?
[14-Sep-2009 12:46:30] <ckrough> ^s
[14-Sep-2009 12:46:42] <rmatte> don't believe so
[14-Sep-2009 12:46:49] <ckrough> awesome app. Splunk
[14-Sep-2009 12:47:04] <ckrough> version 4 UI is amazing.
[14-Sep-2009 12:47:15] <rmatte> oh, syslog analyzer?
[14-Sep-2009 12:47:17] <ckrough> looks like they are starting to encroach on monitoring and alerting space
[14-Sep-2009 12:47:41] <ckrough> anythinglog analyzer anymore
[14-Sep-2009 12:47:41] <rmatte> ah
[14-Sep-2009 12:47:57] <rmatte> ckrough...
[14-Sep-2009 12:47:58] <rmatte> [01:03pm] <rmatte> totaldown = totaldown + (leadin - startDate) + (leadout - endDate)
[14-Sep-2009 12:47:58] <rmatte> [01:03pm] <rmatte> TypeError: unsupported operand type(s) for -: 'tuple' and 'float'
[14-Sep-2009 12:48:01] <rmatte> any ideas?
[14-Sep-2009 12:48:22] <rmatte> is it complaining that one value is tuple and the other is float?
\n[14-Sep-2009 12:49:47] <rmatte> I'm just working on finishing up\nmy availability report script, got all of the mechanics worked out,\njust need to finish coding the functions to actually figure out the\npercentage of availability during a specified period
[14-Sep-2009 12:51:49] <djack> hello world
[14-Sep-2009 12:52:00] <ckrough> rmatte: yeah, are leadin and leadout tuples?
[14-Sep-2009 12:52:04] <rmatte> actually, I think I've found the issue
\n[14-Sep-2009 12:52:37] <rmatte> ckrough: yeh, but there's also an\nissue where sometimes totaldown will be None, which won't work, I need\nto add another if statement in there to check that
[14-Sep-2009 12:55:33] <rmatte> hmmm
[14-Sep-2009 12:55:43] <djack> is there a easy way to create a service status page outside zenoss using zenoss data?
[14-Sep-2009 12:57:11] <rmatte> ckrough: check this out...
[14-Sep-2009 12:57:13] <rmatte> if totaldown:
[14-Sep-2009 12:57:13] <rmatte> float(totaldown)
[14-Sep-2009 12:59:19] <rmatte> Traceback (most recent call last):
[14-Sep-2009 12:59:19] <rmatte> File "./availability.py", line 130, in ?
[14-Sep-2009 12:59:19] <rmatte> float(totaldown)
[14-Sep-2009 12:59:19] <rmatte> TypeError: float() argument must be a string or a number
[14-Sep-2009 12:59:20] <rmatte> so it's not a string, a number, or "None"
[14-Sep-2009 12:59:29] <rmatte> then what the heck is it?
[14-Sep-2009 12:59:29] <rmatte> lol
[14-Sep-2009 12:59:29] <rmatte> it's a numerical value, so there has to be some way to work with it
[14-Sep-2009 12:59:29] <ckrough> type(totaldown)
[14-Sep-2009 12:59:29] <rmatte> actually, weird, it looks like "(None,"
[14-Sep-2009 12:59:29] <rmatte> when I print it
[14-Sep-2009 12:59:29] <ckrough> see what it thinls
[14-Sep-2009 12:59:34] <rmatte> yeh
[14-Sep-2009 13:00:06] <ckrough> you may want to do a try/except expression rather than 'if'
[14-Sep-2009 13:00:15] <ckrough> try float(totaldown): do some stuff
[14-Sep-2009 13:00:19] <ckrough> except: other stuff
[14-Sep-2009 13:00:47] <rmatte> ah
[14-Sep-2009 13:00:52] <ckrough> as a general rule I dont think it's recommend to change a variable type in a test like that
[14-Sep-2009 13:01:03] <rmatte> gotcha
[14-Sep-2009 13:01:09] <rmatte> yeh, that may work better for sure...
[14-Sep-2009 13:04:14] <rmatte> File "./availability.py", line 133
[14-Sep-2009 13:04:14] <rmatte> float(totaldown):
[14-Sep-2009 13:04:19] <rmatte> it doesn't like that
[14-Sep-2009 13:04:34] <ckrough> "try float(totaldown):"
[14-Sep-2009 13:04:35] <rmatte> oh I didn't add try
[14-Sep-2009 13:04:36] <rmatte> my bad
[14-Sep-2009 13:05:17] <rmatte> zenoss@gen01:~/reports$ ./availability.py > availtest.txt
[14-Sep-2009 13:05:17] <rmatte> File "./availability.py", line 133
[14-Sep-2009 13:05:17] <rmatte> try float(totaldown):
[14-Sep-2009 13:05:17] <rmatte> ^
[14-Sep-2009 13:05:17] <rmatte> SyntaxError: invalid syntax
[14-Sep-2009 13:05:20] <rmatte> apparently not
[14-Sep-2009 13:05:32] <ckrough> this could be a bit of the blind leading the blind
[14-Sep-2009 13:05:37] <rmatte> lol
[14-Sep-2009 13:07:03] <ckrough> it worked in my test
[14-Sep-2009 13:08:43] <ckrough> oh, "try: float(test)"
[14-Sep-2009 13:08:43] <ckrough> followed by some stuff
[14-Sep-2009 13:08:55] <rmatte> ah
[14-Sep-2009 13:08:55] <ckrough> http://docs.python.org/tutorial/errors.html#handling-exceptions
[14-Sep-2009 13:11:21] <rmatte> ah, I know what the issue is
[14-Sep-2009 13:11:31] <rmatte> totaldown: (710.48900032043002,)
[14-Sep-2009 13:11:37] <rmatte> totaldown: (None,)
[14-Sep-2009 13:12:06] <ckrough> ha
[14-Sep-2009 13:12:06] <rmatte> I'm using totaldown = cursor.fetchone() and it's outputting the data like that
[14-Sep-2009 13:12:44] <rmatte> there has to be a better way of getting the data out as an actual number
[14-Sep-2009 13:12:48] <rmatte> rather than some weird string
[14-Sep-2009 13:13:09] <rmatte> (the data is being pulled for the sql events database)
[14-Sep-2009 13:13:13] <rmatte> from*
\n[14-Sep-2009 13:13:24] <ckrough> cursor.fetchone() "Fetches the\nnext row of a query result set, returning a single sequence, or None\nwhen no more data is available."
[14-Sep-2009 13:13:40] <rmatte> yeh, but it also returns those (,)
[14-Sep-2009 13:13:42] <rmatte> apparently
[14-Sep-2009 13:14:05] <rmatte> so it's like (value,) or (None,)
[14-Sep-2009 13:14:05] <ckrough> its a sequence
[14-Sep-2009 13:14:17] <rmatte> ah, how can I just pull whatever is output?
[14-Sep-2009 13:14:30] <rmatte> cursor.fetchall?
[14-Sep-2009 13:14:55] <ckrough> maybe "try: float(totaldown[0])
[14-Sep-2009 13:15:46] <rmatte> what will the 0 do?
[14-Sep-2009 13:16:16] <ckrough> returns the value in the first position in the sequence
[14-Sep-2009 13:16:26] <ckrough> counting starts from 0
[14-Sep-2009 13:16:26] <rmatte> ah
[14-Sep-2009 13:17:10] <ckrough> >>> totaldown=("dog","cat")
[14-Sep-2009 13:17:10] <ckrough> >>> totaldown[1]
[14-Sep-2009 13:17:11] <ckrough> 'cat'
[14-Sep-2009 13:18:36] <rmatte> float(totaldown[0])
[14-Sep-2009 13:18:36] <rmatte> TypeError: float() argument must be a string or a number
[14-Sep-2009 13:20:52] <rmatte> fetchone() retrieves a row as a tuple
[14-Sep-2009 13:21:53] <rmatte> now I just need to figure out how to convert a tuple to a float
\n[14-Sep-2009 13:22:31] <ckrough> you cant convert a tuple to a\nfloat since a tuple typically contains multiple values, but you should\nbe able to turn the first value in the tuple into a float with that\nlogic
[14-Sep-2009 13:22:47] <rmatte> right, but apparently not
[14-Sep-2009 13:22:57] <ckrough> http://pastebin.com/m363c0ff7
[14-Sep-2009 13:23:07] <ckrough> looks like it should work
[14-Sep-2009 13:26:03] <rmatte> the except runs when it passes or when it fails?
\n[14-Sep-2009 13:27:13] <rmatte> oh, it probably wasn't doing it\nbecause I didn't have a statement to actually check whether it's "None"
[14-Sep-2009 13:27:40] <rmatte> I technically could just do something like: if totaldown[0]:
[14-Sep-2009 13:28:21] <rmatte> aha, it seems to have possibly worked
[14-Sep-2009 13:28:41] <rmatte> yeh, it did
[14-Sep-2009 13:28:46] <ckrough> what did you change?
[14-Sep-2009 13:29:35] <rmatte> I just did a try
[14-Sep-2009 13:29:45] <rmatte> I think it was trying to do float on None values before
[14-Sep-2009 13:29:52] <rmatte> but with try it tries and if it fails it moves on
[14-Sep-2009 13:29:59] <rmatte> do you even need the "except" part in a try?
[14-Sep-2009 13:30:13] <ckrough> dunno, never did it without
[14-Sep-2009 13:33:37] <rmatte> doesn't work without it
[14-Sep-2009 13:33:38] <rmatte> bah
[14-Sep-2009 13:35:03] <ckrough> "except: pass"
[14-Sep-2009 13:35:23] <ckrough> but you probably want something to happen if it fails, dont you?
[14-Sep-2009 13:35:59] <rmatte> yeh, I just want it to move along
[14-Sep-2009 13:36:09] <rmatte> I just need to get 3 values converted from tuple to float
[14-Sep-2009 13:36:13] <rmatte> so that I can work with them
[14-Sep-2009 13:36:46] <rmatte> actually, I could have it set them to "None" if it fails
[14-Sep-2009 13:36:55] <rmatte> I'll give that a shot
[14-Sep-2009 13:40:13] <rmatte> so far so good, no errors, and it normally would have errored by now...
[14-Sep-2009 13:42:51] <ckrough> nice
[14-Sep-2009 13:43:02] <rmatte> it eventually did error, going to try some more stuff...
[14-Sep-2009 13:43:02] <djack> what are we talking about ?
[14-Sep-2009 13:43:12] <rmatte> a script I'm writing
[14-Sep-2009 13:43:22] <djack> and that does what?
[14-Sep-2009 13:44:35] <rmatte> it's a script that's going to spit out an availability report
[14-Sep-2009 13:45:45] <djack> hmmm intresting
[14-Sep-2009 13:46:03] <djack> can i ask 2 small questions ?
[14-Sep-2009 13:46:12] <rmatte> yes
[14-Sep-2009 13:46:49] <twm1010> ask away
[14-Sep-2009 13:47:00] <djack> is there some easy way to create a service overview outside zenoss using the zenoss imput ?
[14-Sep-2009 13:47:23] <djack> i dont want to create the wheel if it already exist
[14-Sep-2009 13:47:23] <rmatte> oh, except is if it passes the test?
[14-Sep-2009 13:48:39] <rmatte> djack: you're basically talking about some sort of external portal?
[14-Sep-2009 13:48:39] <djack> yeah something for the end users to see
[14-Sep-2009 13:48:39] <twm1010> like for a kiosk?
[14-Sep-2009 13:48:39] <djack> yep
\n[14-Sep-2009 13:48:39] <rmatte> well, you can either have them\nactually log in to Zenoss, or you can spend time doing development work\nto develop one, which is basically what I'm doing right now
[14-Sep-2009 13:48:45] <rmatte> oh, kiosk eh?
[14-Sep-2009 13:48:52] <rmatte> one sec, I have something that might be of interest...
[14-Sep-2009 13:49:08] <rmatte> http://forums.zenoss.com/viewtopic.php?p=24470
[14-Sep-2009 13:49:16] <rmatte> not sure if that's exactly what you're looking for or not
[14-Sep-2009 13:49:24] <venturaville> is anyone using sun open storage?
[14-Sep-2009 13:49:31] <djack> well tv screen for our servicedesk
[14-Sep-2009 13:50:09] <rmatte> djack: yeh, you could use that method for that, or you can do dev work and design your own
\n[14-Sep-2009 13:50:25] <rmatte> djack: it's easy enough to\nextract the data from Zenoss if you're a decent Python scripter
[14-Sep-2009 13:50:43] <djack> rmatte php is my thing :-p
[14-Sep-2009 13:50:50] <rmatte> ah
[14-Sep-2009 13:50:57] <djack> can not be difficult if it is in a mysql db
\n[14-Sep-2009 13:51:27] <rmatte> well, we have an actual\nphp/python/whatever the heck else we need developer on staff who is\ngoing to do the more advanced stuff for it, I'm just writing the\nscripts that can be called on to extract the report data
\n[14-Sep-2009 13:51:32] <djack> i'll check it out have to create\none for the servicdesk and a simple protected on for the onclall iphone
[14-Sep-2009 13:52:12] <djack> rmatte generic scripts ? of specific for your company?
[14-Sep-2009 13:52:18] <twm1010> only event data is in MySQL
[14-Sep-2009 13:52:23] <twm1010> zenoss data is in ZOPE
[14-Sep-2009 13:52:47] <rmatte> djack: the scripts themselves will be quite generic
[14-Sep-2009 13:53:01] <rmatte> I'm probably going to release them publicly once we're done with this little project
[14-Sep-2009 13:53:03] <twm1010> Don't see why the kiosk solution wouldn't work.
[14-Sep-2009 13:53:33] <djack> twm1010 will chech it out
\n[14-Sep-2009 13:54:59] <djack> i don't want to give access to the\nservicedesk its not the job of a servicedesker to go check stats etc ...
[14-Sep-2009 13:55:13] <twm1010> I'd be concerned about that mode exposing information
[14-Sep-2009 13:55:40] <djack> twm1010 how do you mean?
\n[14-Sep-2009 13:56:21] <rmatte> djack: I'm assuming by\nservicedesk you mean regular helpdesk "Hi my computer is broken" type\nstaff, and not actual NOC staff?
[14-Sep-2009\n13:57:58] <djack> rmatte we do not have noc staff ... the first\nline are our nike shoes pick up call and run thrue the building
[14-Sep-2009 13:58:02] <ckrough> SneakerNet
[14-Sep-2009 13:59:06]\n<djack> they only have to notice that something is wrong that's\nit .... second the last line of defence (yeah that me and it sucks)\nshould analyse and fix issues
[14-Sep-2009 14:00:07] <rmatte> lol
[14-Sep-2009 14:00:43] <rmatte> yeh, we do things quite differently here
[14-Sep-2009 14:00:49] <djack> rmatte well everybody has his job when that is done they can do more ... but thats not
[14-Sep-2009 14:01:24] <djack> if you have 140 open tickets you should close them first and not play around
[14-Sep-2009 14:01:44] <djack> rmatte read we have bad first line staff ....
[14-Sep-2009 14:02:05] <rmatte> lol
[14-Sep-2009 14:02:19] <djack> rmatte noooooooooooooooooo :-p
[14-Sep-2009 14:03:00] <djack> rmatte is there info about getting the data out of zope ?
[14-Sep-2009 14:06:32] <rmatte> djack: read the developer's guide
[14-Sep-2009 14:06:53] <rmatte> depends on what kind of data you need
[14-Sep-2009 14:07:03] <djack> will do
[14-Sep-2009 14:07:22] <djack> basic stuff
[14-Sep-2009 14:07:38] <djack> second question is there a jmx expert that can help me out ?
[14-Sep-2009 14:07:50] <rmatte> I'm certainly not one
[14-Sep-2009 14:07:55] <rmatte> unfortunately
[14-Sep-2009 14:08:33] <djack> me neither
\n[14-Sep-2009 14:10:19] <rmatte> for some reason it's treating the\nconversion of the totaldown value differently than the other 2 values
[14-Sep-2009 14:11:09] <metalklesk> i'm back
[14-Sep-2009 14:24:02] <rmatte> eugh, there's something seriously wacky with this script
[14-Sep-2009 14:24:38] <rmatte> oh I see why
[14-Sep-2009 14:24:42] <rmatte> stupid me
[14-Sep-2009 14:26:42] <rmatte> ther we go, now I'm getting somewhere
[14-Sep-2009 14:32:56] <djack> wacky scripts are fun
[14-Sep-2009 14:33:28] <djack> rmatte i defined a java process that he does not want to graph
[14-Sep-2009 14:33:37] <cgibbons> hmmm
[14-Sep-2009 14:34:04] <rmatte> djack: I've never used the JMX ZenPack, so I can't really help with it
[14-Sep-2009 14:34:13] <rmatte> djack: are you seeing a graph at all?
[14-Sep-2009 14:34:17] <djack> no not that an OS process
[14-Sep-2009 14:34:22] <rmatte> ah
[14-Sep-2009 14:34:24] <djack> the jmx is running ;-)
[14-Sep-2009 14:34:29] <rmatte> so you added the process and remodeled the device?
[14-Sep-2009 14:34:38] <djack> yep
[14-Sep-2009 14:34:44] <rmatte> how long ago?
\n[14-Sep-2009 14:35:11] <djack> there are currently 2 java\nprocesses running the one give a nice output ... but its not relevant\nthe other that is will not graph
[14-Sep-2009 14:35:16] <rmatte> takes a minimum of 3 polling cycles (15 minutes by default) for a graph to get populated
[14-Sep-2009 14:35:18] <djack> last friday
[14-Sep-2009 14:35:23] <rmatte> ah
[14-Sep-2009 14:35:28] <djack> should be long ago not?
[14-Sep-2009 14:35:41] <rmatte> you're sure you added it with the right name?
[14-Sep-2009 14:35:48] <rmatte> it needs to be fairly exact
[14-Sep-2009 14:36:49] <djack> Regex ^.*\\/*java
[14-Sep-2009 14:37:17] <djack> and it worked before ....
[14-Sep-2009 14:38:30] <djack> i checked and no rrd files are on the system
[14-Sep-2009 14:38:50] <rmatte> no, don't actually use regex syntax
[14-Sep-2009 14:38:55] <rmatte> just literally type the process name
[14-Sep-2009 14:40:21] <djack> hmmm
[14-Sep-2009 14:40:37] <djack> the name was with a capital ....
[14-Sep-2009 14:41:39] <rmatte> it's case sensitive as well
[14-Sep-2009 14:41:41] <rmatte> it's linux
[14-Sep-2009 14:42:12] <djack> yeah silly me
[14-Sep-2009 14:42:30] <djack> I'm going to sit in the corner and cry
[14-Sep-2009 14:43:00] <rmatte> hehe
[14-Sep-2009 14:43:17] <Apocalipse> hi people...
[14-Sep-2009 14:43:29] <Apocalipse> one question about linking locations
[14-Sep-2009 14:44:27] <Apocalipse> i model my core router... and other routers linked on it
[14-Sep-2009 14:44:40] <Apocalipse> but its not showing links between then
[14-Sep-2009 14:46:27] <Apocalipse> any idea?
\n[14-Sep-2009 14:54:40] <rmatte> Apocalipse: It depends on the\ntypes of routes, whether you have modelled complete related routes,\nit's possible that you won't be able to get the links to show like I\ntold you a few days ago
[14-Sep-2009 15:03:08] <Apocalipse> rmatte... after that u told me... i re-modelled all routers...
\n[14-Sep-2009 15:10:28] <rmatte> I told you that I have a Zenoss\nbox where we can't get it to draw lines no matter what, and we actually\ndo monitor the core router as well
[14-Sep-2009 15:10:36] <rmatte> it has something to do with the way that the routing is setup
[14-Sep-2009 15:11:02] <Apocalipse> ok
[14-Sep-2009 15:11:05] <Apocalipse> thx
[14-Sep-2009 15:19:06] <rmatte> Apocalipse: also make sure that all of your locations are actually on the map
[14-Sep-2009 15:19:13] <rmatte> it's fairly picky about location info
[14-Sep-2009 15:19:25] <Apocalipse> yes they are
[14-Sep-2009 15:30:17] <djack> is there a way to auto model devices ?
[14-Sep-2009 15:30:36] <djack> lets say after the device is back up ?
\n[14-Sep-2009 15:39:24] <rmatte> npmccallum: any idea where the\nscript that works out availability for the availability report is\nstored?
[14-Sep-2009 15:39:33] <rmatte>\nnpmccallum: I've made my own but the values need to jive and they\naren't exactly right now
[14-Sep-2009 15:39:51] <npmccallum> rmatte: which availability report are you talking about?
[14-Sep-2009 15:40:14] <npmccallum> rmatte: the device availability report?
[14-Sep-2009 15:40:14] <rmatte> In Zenoss Core - Reports -> Performance Reports -> Availability
[14-Sep-2009 15:40:18] <rmatte> correct
[14-Sep-2009 15:41:25] <rmatte> I went in to the .rpt file and tried to figure it out
[14-Sep-2009 15:41:28] <rmatte> but couldn't
[14-Sep-2009 15:41:44] <rmatte> so I've been writing one from scratch, but I've run in to an interesting issue with it
[14-Sep-2009 15:51:24] <npmccallum> rmatte: there does not appear to be a separate file for that report
[14-Sep-2009 15:51:33] <metalklesk> how can i submit my zenpack of mandriva ?
[14-Sep-2009 15:52:06] <mrayzenoss> metalklesk: you can email me, mray@zenoss
[14-Sep-2009 15:52:09] <metalklesk> it's just an edited zenpack of suse (opensuse)
[14-Sep-2009 15:52:55] <rmatte> npmccallum: hmmm
[14-Sep-2009 15:53:12] <npmccallum> rmatte: it is just using the built in API, so none is required
[14-Sep-2009 15:54:10] <npmccallum> rmatte: availability is actually more complicated than it would seem at first
[14-Sep-2009 15:54:28] <npmccallum> rmatte: you have to deal with things like coalesced vs averaged, etc
\n[14-Sep-2009 15:55:12] <rmatte> basically, I just need to see how\nyou guys are actually calculating the availability over a specific\nperiod of time... so far what my script does is it collects how long\nthe device was down for during the period, but it also finds any ping\ndown events that completely overlap the window, if it completely\noverlaps it's 0% availability, and on top of that, it finds any events\nthat start before the window and end during it, then take the value and\nsubtract....
[14-Sep-2009 15:55:31] <rmatte> against the time from the start of the specified window.
[14-Sep-2009 15:56:02]\n<rmatte> It also does it for ping events starting during the\nwindow and ending afterwards, but it subtracts against the end of the\nspecified window
[14-Sep-2009 15:56:04] <npmccallum> rmatte: check out the availability() method on the Device class
\n[14-Sep-2009 15:56:32] <rmatte> now theoretically, I should come\nout with a total number that is smaller or even to the window\nspecified, but I never do
[14-Sep-2009 15:58:25] <metalklesk> mray@zenoss.com right ?
[14-Sep-2009 15:58:37] <metalklesk> .org ? .net ?. cl ? jeje
[14-Sep-2009 15:58:58] <rmatte> ah, I think I might have found what's causing the issue
[14-Sep-2009 15:59:17] <npmccallum> metalklesk: yes, mray@zenoss.com
[14-Sep-2009 16:05:19] <metalklesk> mrayzenoss, i have send you an email with the zenpack
[14-Sep-2009 16:07:23] <mrayzenoss> metalklesk: got it, I'll get back to you soon
[14-Sep-2009 16:07:46] <metalklesk> ok, thanks
[14-Sep-2009 16:21:38] <Apocalipse> hello... iam back
[14-Sep-2009 16:29:46] <metalklesk> hello Apocalipse
[14-Sep-2009 16:30:09] <metalklesk> mrayzenoss, i have send you an email with the screenshots of the mandriva zenpack
[14-Sep-2009 16:30:25] <metalklesk> i will fill out the contribution form right now
[14-Sep-2009 16:30:32] <mrayzenoss> cool, thanks
\n[14-Sep-2009 16:30:54] <mrayzenoss> like I said in the email,\nI've got a backlog of zenpacks because of this site launch, but it'll\nbe up soon
[14-Sep-2009 16:31:28] <metalklesk> ok
[14-Sep-2009 16:38:29] <Apocalipse> matt.. is the new website done?
[14-Sep-2009 16:40:13] <rmatte> Apocalipse: no, it's not done yet
[14-Sep-2009 16:40:49] <mrayzenoss> don't worry, we're working on it
[14-Sep-2009 16:41:05] <mrayzenoss> not sure it'll ever be 'done'
[14-Sep-2009 16:41:16] <rmatte> hehe
[14-Sep-2009 16:41:22] <mrayzenoss> apparently some of our spam got migrated over
[14-Sep-2009 16:41:49] <mrayzenoss> just searched for "xbox, iphone, playstation, blackberry" and found a bunch
\n[14-Sep-2009 16:42:39] <rmatte> mother of god, I think I finally\nhave this availability report working properly, found a bunch of stupid\nlittle errors in the SQL queries I had written up
[14-Sep-2009 16:50:01] <rmatte> I do! YES!
[14-Sep-2009 16:50:04] <rmatte> VICTORY!
[14-Sep-2009 16:50:23] <rmatte> now I just need to remove all of the debug lines from the script and clean up the output
[14-Sep-2009 16:50:34] <mrayzenoss> I look forward to seeing your work on the new site
[14-Sep-2009 16:50:39] <rmatte> then I need to figure out how to add options to a python script
[14-Sep-2009 16:50:46] <rmatte> oh, it'll be up there
[14-Sep-2009 16:50:57] <rmatte> this frigging script took me like 2 entire days to figure out and write
\n[14-Sep-2009 16:51:41] <rmatte> oh, this also pretty much\nconfirms that the availability report in Zenoss is fully accurate,\nwhich is good for you
[14-Sep-2009 16:51:42] <rmatte>
[14-Sep-2009 16:54:15] <mrayzenoss> finally, you find something not broken!
[14-Sep-2009 17:00:21] <rmatte> hehe
[14-Sep-2009 17:00:52] <rmatte> the exporting in the availability report actually works half the time too, but it's still semi broken
[14-Sep-2009 17:01:21] <rmatte> it'll export from the last 7 days sometimes, but most of the time it exports during the dates specified
[14-Sep-2009 17:02:07] <rmatte> well, now I just need to polish this script up and it's finally done
[14-Sep-2009 17:23:48] <rmatte> Matt, check this out: http://pastebin.com/m492ac589
[14-Sep-2009 17:23:54] <rmatte> the availability script
[14-Sep-2009 17:24:07] <rmatte> it's not 100% done yet, still need to add options for the start and end dates
[14-Sep-2009 17:24:15] <rmatte> but other than that, it's fully functional
[14-Sep-2009 17:24:19] <rmatte> outputs in csv format
[14-Sep-2009 17:26:44] <rmatte> actually, that how='AVERAGE' line doesn't need to be in there
[14-Sep-2009 17:26:47] <rmatte> going to remove that
[14-Sep-2009 17:27:01] <mrayzenoss> cool, we're very close to launching
[14-Sep-2009 17:27:06] <rmatte> nice
[14-Sep-2009 17:42:10] <Apocalipse> i am floating...on the subject
[14-Sep-2009 17:42:54] <rmatte> float(Apocalipse)
[14-Sep-2009 17:42:57] <rmatte> now you are
[14-Sep-2009 17:43:03] <Apocalipse> kkkkk
[14-Sep-2009 17:43:12] * rmatte couldn't resist a little bit of Python humour
[14-Sep-2009 17:43:19] <Apocalipse> hehehhehehe
[14-Sep-2009 17:43:33] <rmatte> well, I'm out of here, way pay time for me to leave
[14-Sep-2009 17:43:37] <rmatte> have a nice evening folks
[14-Sep-2009 17:43:47] <Apocalipse> same to u rmatte
[14-Sep-2009 17:43:51] <rmatte> way past time for me to leave* rather
[14-Sep-2009 17:44:00] <rmatte> not pay, though I get paid tomorrow
[14-Sep-2009 17:44:01] <rmatte> woot
[14-Sep-2009 19:50:08] <jb> hm
[14-Sep-2009 19:50:21] <jb> can't login to the community site.. or retreive my password
[14-Sep-2009 20:26:59] <ironpaw> hi
[15-Sep-2009 00:00:46] [disconnected at Tue Sep 15 00:00:46 2009]
[15-Sep-2009 00:00:46] [connected at Tue Sep 15 00:00:46 2009]
[15-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
[15-Sep-2009 04:20:08]\n<adelbul> hello, when i want to "Add a zenpack" to a class of\ndevice. When I click on "Add a zenpack" the loading screen never stop.\nHave you encounter a similar issue ?
[15-Sep-2009 04:23:05] <Troubadix09> adelbul: It is not "add a zenpack", it is "add TO zenpack" in the device class
[15-Sep-2009 04:23:50] <Troubadix09> adelbul: "add a zenpack" is under "settings/ZenPacks"
\n[15-Sep-2009 04:24:59] <Troubadix09> adelbul: in device class it\nis for creating an own zenpack, in settings/ZenPacks it is for adding\ndownloaded ZenPacks
[15-Sep-2009 04:28:20]\n<Troubadix09> adelbul: normaly ZenPacks itself determine to wich\nclass the included templates added or it install template that must be\nmanually bind to devices or a device class
[15-Sep-2009 04:33:18] <adelbul> Troubadix09: sorry i wanted to say "add to zenpack".
[15-Sep-2009 04:33:50] <adelbul> i added the zenpack with the command line tool zenpack --install
[15-Sep-2009 04:35:11] <adelbul> I create a device class /Server/SSH/Solaris.
[15-Sep-2009 04:35:26] <adelbul> then i installed the zenpack for opensolaris.
[15-Sep-2009 04:36:06] <adelbul> and now i try to add the opensolaris zenpack to the class /Server/SSH/Solaris
[15-Sep-2009 04:37:21] <adelbul> docs/DOC-3414
[15-Sep-2009 04:40:33] <adelbul> if i understand your commentm you say that i just have to bind the templates ?
\n[15-Sep-2009 04:51:59] <Troubadix09> adelbul: sorry, i didn't use\nsolaris, but I think, that they would install to the device-class\n"/Server/SSH/Solaris" when you create this class before installing. You\ncan control it ...
[15-Sep-2009 04:53:05] <Troubadix09> adelbul: on the "Templates"-Tab on the device-class...
\n[15-Sep-2009 04:55:33] <Troubadix09> adelbul: sometimes a ZenPack\ninstalls new collector plugins, wich must be added manually for example\nthe HP Proliant ZenPack brings new collector plugins wich replace the\noriginal zenoss core collector plugins
[15-Sep-2009\n04:58:59] <Troubadix09> adelbul: you can't add ZenPacks to device\nclasses, because ZenPacks are packages of configuration like add new\nTemplates, collector Plugins, Data-Sources to existing Templates, new\nTabs, new Dashboard-Portlets etc.
[15-Sep-2009 05:05:03]\n<Troubadix09> adelbul: In this sense it is like a recording of\nyour changes in zenoss always you use the "Add to Zenpack"-Button, wich\nthen you can give to the community
[15-Sep-2009 05:20:36] <adelbul> Troubadix09: ok thank you for this explanation.
[15-Sep-2009 06:24:30] <skipzoid> hey
\n[15-Sep-2009 06:27:21] <skipzoid> in the event manager i've\ncreated an event command to be run when an event ocours - sendsms.pl \n"$device ${evt/message}" and on the command line if this is typed as\nthe zenoss user it does send an sms - how do i attach this event\ncommand to an actual even ocourance?
[15-Sep-2009 06:48:07] <ckrough> morning
[15-Sep-2009 06:52:33] <Troubadix09> ckrough: morning is good, here it is afternoon ;-)
[15-Sep-2009 06:54:38] <ckrough> good afternoon
[15-Sep-2009 07:01:32] <skipzoid> hello
[15-Sep-2009 07:02:00] <skipzoid> what variables can be passed to an event command
[15-Sep-2009 07:07:18] <adelbull> hello i have a question.
[15-Sep-2009 07:09:17]\n<adelbull> When i want to test a based ssh zenpack. in\n'datasource', 'test against device' the zenoss server search the\ncommand on the server side and not on the client side.
\n[15-Sep-2009 07:10:21] <adelbull> it is a problem, because a\nzenoss server installed on a linux distibution doesnt have the specific\ncommand of solaris, hpux, aix...
[15-Sep-2009 07:11:57] <skipzoid> anyone know where i can find a list of varibles that can be passed to a event command
[15-Sep-2009 07:22:43] <ckrough> skipzoid: I think that is detailed in the developer guide
\n[15-Sep-2009 07:33:37] <skipzoid> I'm trying this - sendsms.pl\n"$device" - and when run on the command line a text of $device is\nreceived - when under zenoss event commands - nothing happens if I\nchange the $device so say fred I receive a text 'fred' it seems what is\nbeing passed to the sendsms.pl script is null so nothing is sent
[15-Sep-2009 07:35:23] <ckrough> does it need to be "here\\deviceId" or something more zopeish?
\n[15-Sep-2009 07:36:59] <skipzoid> i would have hoped it was going\nto be simple like the interpreter runing the command just substituted\nthe various varibles for the text required, perhaps I need to be more\ncreative, but I at least need the event id that triggered the command\nso I can extract the rest of the info I want from the database
[15-Sep-2009 07:37:37] <skipzoid> i'm still new to this so maybe it is simple once i know
[15-Sep-2009 07:40:00] <ckrough> sec
[15-Sep-2009 07:40:59] <ckrough> try "%(device)s"
[15-Sep-2009 07:41:15] <ckrough> %(summary)s
[15-Sep-2009 07:41:16] <ckrough> etc...
[15-Sep-2009 07:42:31] <skipzoid> i get a text message from the system saying %device
[15-Sep-2009 07:43:12] <skipzoid> i just tried it - but i think i knew that was going to happen
[15-Sep-2009 07:46:17] <ckrough> you entered "%(device)s" exactly? including the parens and trailing s?
[15-Sep-2009 07:46:48] <skipzoid> let me just check - i havnt changed it back yet
[15-Sep-2009 07:48:52] <skipzoid> ok - i had missed the trailing s - but exactly I receive - %(device)s
[15-Sep-2009 07:48:59] <ckrough> meh
[15-Sep-2009 07:49:02] <ckrough> worth a shot
[15-Sep-2009 07:49:15] <skipzoid> thats fine - happy to try most things
\n[15-Sep-2009 07:49:24] <ckrough> Not sure what to tell you, hang\naround for a while. there should be some more knowledgable people on in\nthe next couple of hours
[15-Sep-2009 07:50:12]\n<skipzoid> i'm still searching for the answer to this one - i\nthink once solved i'm pretty much there with my solution
[15-Sep-2009 07:52:51] <frozty_sa> hi
\n[15-Sep-2009 07:53:10] <frozty_sa> I've found the threads and\ndocumentation on how to make zenoss collect data using distributed\ncollectors
[15-Sep-2009 07:54:00] <frozty_sa>\ncan someone tell me if there's a way to have multiple zenoss core\ninstallations (these will be in physically different networks) and then\nhave a super-master installation that can control the various master\ninstallations?
[15-Sep-2009 08:26:37] <ckrough> frozty_sa: not at the moment, it's an open feature request though
\n[15-Sep-2009 08:27:18] <ckrough> frozty_sa: there is a global\nevent viewer. I believe it's enterprise only. It only allows you to\nview events, not manage the sites.
[15-Sep-2009 08:27:27] <frozty_sa> mmmm
[15-Sep-2009 08:27:28] <frozty_sa> thanks
[15-Sep-2009 08:27:35] <frozty_sa> any idea on progress of that sort of thing?
\n[15-Sep-2009 08:28:20] <ckrough> nope. but I would recommend you\nsend it to them in feedback so they can know how many people want i
[15-Sep-2009 08:28:23] <ckrough> it
[15-Sep-2009 08:28:55] <frozty_sa> guess so
[15-Sep-2009 08:35:57] <ckrough> mrayzenoss: community site looks good. What backend is that again?
[15-Sep-2009 08:44:20] <mrayzenoss> Jive SBS
[15-Sep-2009 08:45:17] <mrayzenoss> I've started up a thread to capture complaints, the site is a work in progress
\n[15-Sep-2009 08:52:51] <Troubadix09> mrayzenoss: I agree with\nckrough, nice new community, but I think it would be better that on the\nTab "All Content" the Thread-Subject is the first column ...
[15-Sep-2009 08:53:24] <mrayzenoss> yeah, I've got a ticket with Jive about that
[15-Sep-2009 08:53:41] <mrayzenoss> that's why I made the forums switch over to the Discussions tab where possible
[15-Sep-2009 08:53:56] <mrayzenoss> there's no "Threaded View" widget
[15-Sep-2009 08:53:57] <Troubadix09> mrayzenoss: because i click several times on the Author ;-)
\n[15-Sep-2009 08:57:42] <Troubadix09> mrayzenoss: I search the\nForum for a Post about Zenpack-install and where directed to an old\nPost...
[15-Sep-2009 08:58:42] <Troubadix09>\nmrayzenoss: in this Post there where a Link to another Thread, that\nwould'nt reach the Thread
[15-Sep-2009 08:59:44]\n<mrayzenoss> yeah... we mapped all the old wiki content and\npages, but we couldn't preserve the links from threads to previous\nthreads
[15-Sep-2009 09:00:13] <mrayzenoss> I'll see if there's a way to fix that
[15-Sep-2009 09:00:28] <mrayzenoss> what's your issue?
[15-Sep-2009 09:01:19] <Troubadix09> I post it in the Forum under ZenPack, I have trouble to install the WMI-DataSource-Zenpack
[15-Sep-2009 09:02:40] <mrayzenoss> responding there
[15-Sep-2009 09:02:49] <venturaville> mrayzenoss: why is the zenoss community portal spamming me with notes about thread messages?
[15-Sep-2009 09:02:50] <mrayzenoss> I believe your .egg is actually a .zip or something
[15-Sep-2009 09:03:07] <mrayzenoss> venturaville: were you previously subscribed via email?
[15-Sep-2009 09:03:35] <venturaville> mrayzenoss: subscribed to what precisely?
[15-Sep-2009 09:03:54] <Troubadix09> mrayzenoss: I rename it with ending .egg
[15-Sep-2009 09:04:05] <mrayzenoss> were you using the email bridge to use the old forums?
[15-Sep-2009 09:04:17] <venturaville> not that I am aware of
[15-Sep-2009 09:04:33] <mrayzenoss> are you getting emails based on threads you'd been on before?
[15-Sep-2009 09:04:50] <venturaville> mrayzenoss: no I am getting emails on a lot of threads I know nothing about
[15-Sep-2009 09:05:05]\n<dollarbang> Question about google maps and the enterprise\nversion of Zenoss. According to Google, their API is limited to 15000\nhits a day. How would this affect the Enterprise version of Zenoss.
[15-Sep-2009 09:05:24] <mrayzenoss> dollarbang: we pay Google for a license, no issues
[15-Sep-2009 09:05:37] <mrayzenoss> venturaville: give me an example subject?
[15-Sep-2009 09:06:26] <rmatte> mrayzenoss: just out of curiosity, what does a google maps API license cost?
[15-Sep-2009 09:06:47] <mrayzenoss> not sure, I think I heard $10K or something
[15-Sep-2009 09:06:57] <rmatte> ah, cool
[15-Sep-2009 09:06:57] <mrayzenoss> we cover it for all users, Core and Enterprise
[15-Sep-2009 09:07:25] <rmatte> ah, I was assuming it wasn't covered for Core since you actually have to manually input an API key
[15-Sep-2009 09:07:43] <metalklesk> hi
[15-Sep-2009 09:08:10] <venturaville> mrayzenoss: [zenoss-users] New message: "Re: New Site Issues Thread" [1co8L1-1tA-adH]
[15-Sep-2009 09:09:14] <chudler> ugh, I get those emails also
[15-Sep-2009 09:09:55] <rmatte> I haven't gotten any
[15-Sep-2009 09:09:57] <mrayzenoss> venturaville: with the old site... did you interact with zenoss-users via email?
[15-Sep-2009 09:10:06] <rmatte> The only email I got was someone adding me as a link
[15-Sep-2009 09:10:14] <rmatte> errr, as a connection
[15-Sep-2009 09:10:14] <mrayzenoss> venturaville: or did you post through the website?
[15-Sep-2009 09:10:15] <rmatte> rather
[15-Sep-2009 09:11:16] <mrayzenoss> chudler: that goes for you too, did you use email to respond to the old forums?
[15-Sep-2009 09:11:30] <metalklesk> i have a little tiny small question
\n[15-Sep-2009 09:11:42] <chudler> I used to get threaded email for\nevery forum post. Not sure if I ever responded via email.
[15-Sep-2009 09:11:56] <chudler> seems likely..
\n[15-Sep-2009 09:12:32] <metalklesk> in a production enviroment,\nshould zenoss be in the same network segment as the devices to be\nmonitored ?
[15-Sep-2009 09:13:45] <mrayzenoss>\nchudler: yeah, we carried that over to the new site... but obviously\nthe filters changed
[15-Sep-2009 09:13:46]\n<venturaville> mrayzenoss: I pretty much just used email (mailing\nlist), but never really did anything with the website
[15-Sep-2009 09:13:55] <mrayzenoss> venturaville: yeah
[15-Sep-2009 09:14:10] <venturaville> how do I disable it
[15-Sep-2009 09:14:26] <mrayzenoss> have you both logged into the new site?
[15-Sep-2009 09:14:44] <dollarbang> mrayzenoss: thanks. another reason to go Enterprise.
[15-Sep-2009 09:15:05] <rmatte> mrayzenoss: do I need to create a new account to login to community.zenoss.org?
[15-Sep-2009 09:15:20] <mrayzenoss> no, your existing account is there
[15-Sep-2009 09:15:26] <mrayzenoss> you may need a password reset
[15-Sep-2009 09:15:34] <chudler> I am resetting my pw now
[15-Sep-2009 09:15:37] <rmatte> hmmm, I've just tried logging in with every password that I could think of
[15-Sep-2009 09:15:45] <rmatte> do I need to reset it?
[15-Sep-2009 09:16:39] <venturaville> we are installing enterprise on our production system today...........
[15-Sep-2009 09:16:39] <mrayzenoss> yeah, a reset is required
[15-Sep-2009 09:16:43] <rmatte> k
\n[15-Sep-2009 09:19:49] <skipzoid> simple question (i hope) I've\ncreated a command in /ZenEventManager /Event Commands which is a perl\nscript sendsms.pl it takes the argument of the message and sends it to\na number preconfigured in the script its self - on the command line\nsendsms.pl "test" sends test to the handset and works - i'm trying to\nget it to send event infomation, what varibles are avable to pass to it?
\n[15-Sep-2009 09:20:52] <mrayzenoss> chudler/venturaville go to\nthe forum you're getting emails from (ie.\ncommunity/forums/zenoss-users)
[15-Sep-2009 09:21:01] <mrayzenoss> and select "Stop email notifications"
[15-Sep-2009 09:21:04] <venturaville> found the settings
[15-Sep-2009 09:21:11] <venturaville> I can take it from here thanks :-)
[15-Sep-2009 09:21:15] <mrayzenoss> sorry
[15-Sep-2009 09:24:00] <venturaville> np
[15-Sep-2009 09:26:30] crow is now known as ckrough
[15-Sep-2009 09:26:31] <metalklesk> which data should i be able to monitor from a printer (xerox 3250) ?
[15-Sep-2009 09:26:34] <jb> sup
[15-Sep-2009 09:26:38] <jb> fix the new site yet?
[15-Sep-2009 09:26:58] <mrayzenoss> it's up...
[15-Sep-2009 09:27:18] <jb> did you get my tweet?
\n[15-Sep-2009 09:27:58] <jb> my account doesn't work, and its\ntelling me my email address is not valid when I try to retreive my\npassword..
[15-Sep-2009 09:28:15] <mrayzenoss> jb: what was your old account name?
[15-Sep-2009 09:28:21] <rmatte> facebook seems to be screwed this morning, I can't even load my profile
[15-Sep-2009 09:28:22] <Apocalipse> hello people...
[15-Sep-2009 09:28:28] <jb> mrayzenoss: jbaird
[15-Sep-2009 09:28:38] <jb> facbeook has been screwed for weeks..
[15-Sep-2009 09:28:46] <Apocalipse> i think i found why my devices are not linking...
[15-Sep-2009 09:28:48] <rmatte> jb: oh?
[15-Sep-2009 09:28:53] <rmatte> jb: I hardly go on so hadn't noticed
[15-Sep-2009 09:29:33] <rmatte> I can load it fine on my iPhone though
[15-Sep-2009 09:39:12] <mrayzenoss> ok, we've updated the subjects for the forum emails
[15-Sep-2009 09:39:19] <mrayzenoss> they should be closer to previous emails
[15-Sep-2009 09:39:27] <mrayzenoss> if you were using that feature
[15-Sep-2009 09:40:01] <chudler> that's nice, thanks. It was my only other pipeline into the community :-)
[15-Sep-2009 09:40:09] <chudler> its the body of the emails that is most annoying
[15-Sep-2009 09:40:15] <mrayzenoss> yeah, we cleaned up the template
[15-Sep-2009 09:40:28] <mrayzenoss> so now it should be the body of the message, with a footer about how to respond
[15-Sep-2009 09:40:38] <mrayzenoss> dumped out the author link
[15-Sep-2009 09:41:02] <mrayzenoss> unfortunately the subjects still contain the thread ID
[15-Sep-2009 09:41:11] <mrayzenoss> which we've got a ticket open against, since that breaks threading
[15-Sep-2009 09:50:59] <mrayzenoss> grrr... the RSS feeds are kinda sucky
[15-Sep-2009 09:51:12] <mrayzenoss> for the forums, they put the date of the first post in a thread
\n[15-Sep-2009 09:53:51] <rmatte> In case anyone is interested, if\nyour facebook profile isn't showing up (if you're getting "The page you\nrequested was not found" when trying to access it), from the error page\nyou just need to click where it says English (US), or whatever language\nyou have set, then change the language to something else (English (UK)\nfor instance), then change it back to the original language, and voila,\nit should be fixed.
[15-Sep-2009 09:54:21] <mrayzenoss> what's that?
\n[15-Sep-2009 09:57:45] <npmccallum> mrayzenoss: FYI, emails from\nthe new forum are missing the list-id header, probably breaking mail\nfiltering for a lot of people
[15-Sep-2009 09:59:20]\n<ckrough> what does it mean when I get an email from community\nsaying "xxxx added you as a connection."?
[15-Sep-2009\n09:59:59] <rmatte> ckrough: I got that too, it's a community\ndriven site, so adding someone as a connection is probably similar to\nadding someone in LinkedIn
[15-Sep-2009 10:00:28] <ckrough> I assumed. I looked on the site and I see where I can 'follow' someone, but didnt see a connection
[15-Sep-2009 10:03:07] <rmatte> that's probably what that means
[15-Sep-2009 10:03:48]\n<scp2004> Hi guys .. Please , i need some help .. My zenoss stio\nto generate graphs with no explain ! In event.log i got :
[15-Sep-2009 10:03:50]\n<scp2004> 2009-09-15T11:58:34 INFO ZODB.Conflict database\nconflict error (oid 0x019a, class\nProducts.ZenUtils.PObjectCache.PObjectCache) at\n/zport/RenderServer/render (22 conflicts, of which 0 were unresolved,\nsince startup at Tue Sep 15 09:46:50 2009)
[15-Sep-2009 10:03:54] <scp2004> Any Idea ?
[15-Sep-2009 10:04:17] <ckrough> thats an ok error
[15-Sep-2009 10:04:20] <scp2004> It stop to generate graphs !
[15-Sep-2009 10:04:24] <ckrough> "0 were unresolved"
[15-Sep-2009 10:04:29] <ckrough> unrelated to that error
[15-Sep-2009 10:05:43] <scp2004> In 2009-09-15 my Zenoss stop to generate graphs with no explain ! I dont know what is happening ..
[15-Sep-2009 10:07:07] <scp2004> 2009-09-15T09:25:20 ERROR RenderServer Failed to generate a graph
[15-Sep-2009 10:07:07] <scp2004> Traceback (most recent call last):
[15-Sep-2009 10:07:07] <scp2004> File "/usr/local/zenoss/zenoss/Products/ZenRRD/RenderServer.py", line 190, in render
[15-Sep-2009 10:07:07] <scp2004> rrdtool.graph(*gopts)
[15-Sep-2009 10:07:07] <scp2004> error
[15-Sep-2009 10:07:23] <rmatte> ok, first off, calm down, secondly, are you sure it has stopped generating all graphs or just specific ones?
[15-Sep-2009 10:08:13] <scp2004> [rmatte] Are not generating any graphs
[15-Sep-2009 10:09:41] <scp2004> [rmatte] It was ok. But in 2009-09-10 all graphs stop
[15-Sep-2009 10:13:08] <ckrough> scp2004: are you out of disk space?
[15-Sep-2009 10:13:42] <scp2004> [ckrough] i thin not ..but will see
\n[15-Sep-2009 10:14:26] <skipzoid> simple question (i hope) I've\ncreated a command in /ZenEventManager /Event Commands which is a perl\nscript sendsms.pl it takes the argument of the message and sends it to\na number preconfigured in the script its self - on the command line\nsendsms.pl "test" sends test to the handset and works - i'm trying to\nget it to send event infomation, what varibles are avable to pass to it?
[15-Sep-2009 10:15:38] <scp2004> Not ... I have to much disk free ..
[15-Sep-2009 10:15:56] <scp2004> [ckrough] Im using only 4%
[15-Sep-2009 10:16:12] <ckrough> scp2004: it was a long shot:) Just a first check since all graphs stopped.
[15-Sep-2009 10:16:37] <scp2004> [ckrough] Its checked .. All graphs stoped in 2009-09-10
[15-Sep-2009 10:17:20] <metalklesk> you mustdid something in 2009-09-10
[15-Sep-2009 10:17:23] <ckrough> scp2004: so you get no graphs at all, just a traceback when you go to the graph page?
[15-Sep-2009 10:17:50] <metalklesk> things just don't stop working ... windows does but is an exception
\n[15-Sep-2009 10:18:14] <scp2004> [metalklesk] No i dont put my\nhands in it ! .. It was ok ..And i got no idea what happens
[15-Sep-2009 10:18:45] <ckrough> scp2004: do you get a traceback or empty graphs or graphs that stopped updating?
[15-Sep-2009 10:19:00] <scp2004> [ckrough] Empty graphs
[15-Sep-2009 10:19:12] <frozty_sa> ckrough: still around?
[15-Sep-2009 10:19:30] <scp2004> If i ghange Range to Daily i can see that it sopped in 2009-09-10
[15-Sep-2009 10:19:45] <scp2004> [ckrough] If i change Range to Daily i can see that it sopped in 2009-09-10
[15-Sep-2009 10:20:02] <metalklesk> sorry for ask but, did you get any new data since 2009-09-10 ?
[15-Sep-2009 10:20:09]\n<ckrough> scp2004: ah, ok. Check one of the related RRD files.\n'rrdtool last /opt/zenoss/perf/....' for one of the rrd files related\nto one of the broken graphs
[15-Sep-2009 10:20:54] <scp2004> [ckrough] ok ..
\n[15-Sep-2009 10:20:55] <ckrough> scp2004: that will give you the\ntimestamp of that last update to the rrd file. you can translate it\nhere: http://www.epochconverter.com/
[15-Sep-2009 10:21:09] <ckrough> frozty_sa: yup
\n[15-Sep-2009 10:21:55] <ckrough> scp2004: also, as the zenoss\nuser, do 'zenoss status' and make sure zenperfsnmp is running
[15-Sep-2009 10:22:17] <scp2004> [ckrough] Its running
\n[15-Sep-2009 10:22:37] <frozty_sa> ckrough: I'm seeing some\nreferences to a 'multiple hubs' setup, although the documentation's\ninternal structure seems to have changed since google's last index run\nof it
[15-Sep-2009 10:22:55] <frozty_sa>\nckrough: could you maybe point me to where I can find out more about\nthat, or give me a short summary?
[15-Sep-2009 10:22:55] <ckrough> frozty_sa: link?
\n[15-Sep-2009 10:23:11] <frozty_sa>\nhttp://www.google.co.za/search?q=zenoss+multiple+hubs&ie=utf-8&oe=utf-8&aq=t&rls=com.ubuntu:en-US:unofficial&client=firefox-a
\n[15-Sep-2009 10:24:21] <ckrough> frozty_sa: I think those are in\nrelation to the multiple collector scenario, but not a distributed\n'master>slave>collector' architecture
[15-Sep-2009 10:24:29] <scp2004> [ckrough] i cant see anything in /usr/local/zenoss/perfs/daemons/localhost
[15-Sep-2009 10:25:19] <ckrough> scp2004: they should be under $ZENHOME/perf/Devices
[15-Sep-2009 10:25:54] <ckrough> frozty_sa: docs/DOC-2683
\n[15-Sep-2009 10:26:22] <ckrough> mrayzenoss: ouch, all the google\nlinks are borked with the new urlpath. collateral damage I guess.
[15-Sep-2009 10:26:39] <mrayzenoss> ckrough: google is indexing
[15-Sep-2009 10:26:52] <mrayzenoss> or are you talking about the search.zenoss.com?
[15-Sep-2009 10:27:25] <ckrough> mrayzenoss: nope
[15-Sep-2009 10:27:29] <frozty_sa> google's index still shows old links
[15-Sep-2009 10:27:41] <ckrough> ckrough: was talking about the googles.
[15-Sep-2009 10:27:43] <frozty_sa> so anything you click on leads you to the top-level documentation index
[15-Sep-2009 10:27:59] <mrayzenoss> well, we've added redirects for most of the static content, just not the forums
[15-Sep-2009 10:28:01] <ckrough> ha, I just talked to myself. time for more coffee
\n[15-Sep-2009 10:29:43] <ckrough> anyone know if there is a way to\nsee modifications for deleted devices? someone deleted a device and I\nneed to see it's change history
[15-Sep-2009 10:30:06] <frozty_sa> ckrough: thanks, that looks like it'll do what I want it
\n[15-Sep-2009 10:30:41] <ckrough> frozty_sa: are you enterprise?\nnote that it doesnt manage remote sites... just brings together their\nevent consoles
[15-Sep-2009 10:30:52] <ckrough> frozty_sa: and its read-only. no acking
\n[15-Sep-2009 10:32:13] <ckrough> awesome, you can go through your\nZ2 log and look a the methods in the urls to see what the heck the user\ndid to break it
[15-Sep-2009 10:32:30] <frozty_sa> ckrough: not enterprise, no
\n[15-Sep-2009 10:32:47] <frozty_sa> ckrough: currently I've got a\nzabbix deployment that's supposed to do this, but it's not
[15-Sep-2009 10:32:53] <ckrough> frozty_sa: I think that zenpack only works with ent. installs
[15-Sep-2009 10:33:29] <frozty_sa> but zabbix is a horrendous piece of shit, so I'm looking at what I can move to
\n[15-Sep-2009 10:44:58] <gorkadel> hello, i try to create a new\ntemplate via the but when i click on "add datasource" the loading\nscreen appears and never stop ...
[15-Sep-2009 10:45:25] <gorkadel> is it a bug ?
[15-Sep-2009 10:45:37] <gorkadel> "via the device class"
[15-Sep-2009 10:46:29] <mrayzenoss> gorkadel: which browser?
[15-Sep-2009 10:46:52] <gorkadel> safari
[15-Sep-2009 10:47:00] <mrayzenoss> gorkadel: version?
[15-Sep-2009 10:47:36] <gorkadel> 4.0.3
[15-Sep-2009 10:47:50] <mrayzenoss> sorry, Zenoss version?
[15-Sep-2009 10:48:04] <gorkadel> zenoss 2.4.5
[15-Sep-2009 10:49:15] <gorkadel> same issue with firefox.
[15-Sep-2009 10:49:46] <gorkadel> "loading..." for ever ...
[15-Sep-2009 10:50:51] <mrayzenoss> gorkadel: any customizations?
[15-Sep-2009 10:50:55] <mrayzenoss> or ZenPacks
[15-Sep-2009 10:51:13] <gorkadel> yes i added opensolaris zenpacks
[15-Sep-2009 10:52:08] <mrayzenoss> has zope been restarted since then?
[15-Sep-2009 10:52:14] <mrayzenoss> 'zopectl restart'
[15-Sep-2009 10:54:34] <Apocalipse> hello?
[15-Sep-2009 10:54:41] <cgibbons> ¿que?
[15-Sep-2009 10:55:29] <Troubadix09> cu ... bye
[15-Sep-2009 10:55:43] <Apocalipse> ¿como estas?
[15-Sep-2009 10:55:52] <mrayzenoss> gotta go do a demo, PM me if you have any zenoss.org issues
[15-Sep-2009 10:56:08] <gorkadel> mrayzenoss: ok thanks
[15-Sep-2009 10:56:21] <gorkadel> mrayzenoss: ok try your suggestion.
[15-Sep-2009 10:57:51] <gorkadel> it is working
[15-Sep-2009 10:57:58] <gorkadel> thqnks you.
[15-Sep-2009 11:02:10]\n<markeriv> Hey guys, just a quick question for you. Trying to\nhunt down the admin guide that deals extensively with users and\npermissions
[15-Sep-2009 11:03:46] <markeriv> did you guys change your docs area recently?
[15-Sep-2009 11:03:46] <markeriv> hmm
\n[15-Sep-2009 11:26:16] <chemist> what type of checks does zenoss\ndo when you ask it to monitor a process? does it just check that it\nexists?
[15-Sep-2009 11:28:36] <chudler> it checks it exists, the cpu and mem usage, IIRC
[15-Sep-2009 11:29:14] <chudler> also counts how many processes match the name pattern
\n[15-Sep-2009 11:34:13] <scp2004> [mrayzenoss] Hi .. My Zenoss\nrunning on CentOS 5 and a HP server Proliant DL 160 are very very slow\n! Since i installed it .. Im monitoring 385 Devices.. Its normal ?
[15-Sep-2009 11:34:32] <markeriv> how much ram you got?
[15-Sep-2009 11:34:39] <markeriv> what the stats on your core?
[15-Sep-2009 11:35:19] <markeriv> also, what type of internet connection do you have and how much bandwidth?
[15-Sep-2009 11:43:28] <rmatte> also, is it specific pages that are slow, or everything?
[15-Sep-2009 11:48:10] <scp2004> I have 2GB RAM , 200 MB bandwidth , and using 12% of core
[15-Sep-2009 11:48:28] <scp2004> Especialy when i try access devices
[15-Sep-2009 11:48:53] <scp2004> When i click on a device to see status, os ..etc
[15-Sep-2009 11:49:05] <scp2004> it spend like 5 - 10 min
[15-Sep-2009 11:49:10] <markeriv> might want more ram
[15-Sep-2009 11:49:10] <scp2004> to open
[15-Sep-2009 11:49:19] <markeriv> 2 gigs is fairly low
[15-Sep-2009 11:49:20] <scp2004> Yep ..Maybe ..
[15-Sep-2009 11:49:25] <markeriv> my server is running 8
\n[15-Sep-2009 11:49:49] <scp2004> Guys ..About my graphs are\nstopped ..Look this message : localhost zenperfsnmp heartbeat failure
[15-Sep-2009 11:49:56] <scp2004> in my events ..
[15-Sep-2009 11:50:12] <markeriv> tried a restart of zenperfsnmp?
[15-Sep-2009 11:50:16] <scp2004> Its probably the cause ?
[15-Sep-2009 11:50:23] <scp2004> i will now
[15-Sep-2009 11:50:46] <markeriv> whats your current memory usage?
[15-Sep-2009 11:51:11] <scp2004> [markeriv] 99% ..
[15-Sep-2009 11:51:16] <markeriv> ahh
[15-Sep-2009 11:51:16] <scp2004> Daemon: zenperfsnmp program running; pid=3397
[15-Sep-2009 11:51:18] <markeriv> ram is your problem
[15-Sep-2009 11:51:25] <markeriv> you need more
[15-Sep-2009 11:52:36] <scp2004> It doesn´t stop ..zenperfsnmp ..
[15-Sep-2009 11:53:02] <scp2004> in /usr/local/zenoss/bin the command ./zenperfsnmp stop
[15-Sep-2009 11:53:21] <chudler> please post the output of the "free" command
[15-Sep-2009 11:55:38] <ckrough> scp2004: Linux will always use 99% ram, check your swap usage
[15-Sep-2009 11:56:03] <ckrough> scp2004: or check the second line of 'free' to see what's being used by buffering vs everything else
[15-Sep-2009 11:57:39] <scp2004> [ckrough] its using like 30% of swap
[15-Sep-2009 11:58:09] <scp2004> about graphs .. I ll stop and Start Zenoss-stack again
[15-Sep-2009 11:58:48] <ckrough> scp2004: yeah, you need ram
[15-Sep-2009 11:59:02] <ckrough> scp2004: swap usage = evil, especially for zenperfsnmp
[15-Sep-2009 12:02:49] <scp2004> ahhh ...ok ...thanks very much
[15-Sep-2009 12:06:15]\n<chudler> if you want to favor keeping pages mapped use\n/proc/sys/vm/swappiness, but it could increase load AND memory pressure.
[15-Sep-2009 12:07:33] <chudler> lower number means less apt to swap pages to disk, presumably at the expense of the cache.
\n[15-Sep-2009 13:32:56] <skipzoid> simple question (i hope) I've\ncreated a command in /ZenEventManager /Event Commands which is a perl\nscript sendsms.pl it takes the argument of the message and sends it to\na number preconfigured in the script its self - on the command line\nsendsms.pl "test" sends test to the handset and works - i'm trying to\nget it to send event infomation, what varibles are avable to pass to it?
[15-Sep-2009 13:45:57] <ckrough> ian: thanks, sloppy troubleshooting on my part
[15-Sep-2009 13:46:03]\n<skipzoid> simple question (i hope) I've created a command in\n/ZenEventManager /Event Commands which is a perl script sendsms.pl it\ntakes the argument of the message and sends it to a number\npreconfigured in the script its self - on the command line sendsms.pl\n"test" sends test to the handset and works - i'm trying to get it to\nsend event infomation, what varibles are avable to pass to it?
[15-Sep-2009 13:47:14] <djack> hello world belgium calling ;-)
[15-Sep-2009 13:48:37] <skipzoid> hello from the UK 8^)
[15-Sep-2009 13:49:06] <markeriv> Here is a question for you, I am trying to create an event view that is limited to a specific group.
[15-Sep-2009 13:49:31] <markeriv> So that I can create an alert that only affects a specific group
[15-Sep-2009 13:50:48] <mrayzenoss> skipzoid: here's rmatte's post from yesterday:
[15-Sep-2009 13:50:51] <mrayzenoss> 11:09:29 AM rmatte: Here's an example of a command we use to create tickets in our ticketing system:
[15-Sep-2009 13:50:51] <mrayzenoss> 11:09:48\nAM rmatte: /mspfiles/scripts/create_ticket.pl -autonotify -customer\n'gen-00025' -device '${dev/id}' -deviceIP '${dev/manageIp}' -collector\n`hostname -f` -first '${evt/firstTime}' -last '${evt/lastTime}' -count\n'${evt/count}' -summary '${evt/summary}' -noteTitle 'System Monitor\nError' -note '${evt/message}' -severity '${evt/severity}' -group\n'${evt/DeviceGroups}' -impact '${evt/DevicePriority}' -component\n'${evt/component}' -que
[15-Sep-2009 13:51:20] <skipzoid> yeah i was experimenting with that but with no luck
[15-Sep-2009 13:56:55] <skipzoid> got it - stupid mistake - mixed up my ' ` and " Grrrr !!
[15-Sep-2009 13:57:20] <rmatte> lol
[15-Sep-2009 14:55:04] <djack> hey you all can you use xml-rpc or rest to pull out data from zenoss ?
\n[15-Sep-2009 14:55:57] <rmatte> djack: actually, someone here\nasked me that today, so I'd be interested in knowing as well. I've\nseen xml-rpc calls to Zenoss before
[15-Sep-2009 14:57:25] <djack> rmatte i'm still looking for a way to create a easy external status page ....
[15-Sep-2009 14:57:37] <ckrough> djack: you can pull lots of data from xmlrpc
[15-Sep-2009 14:58:14] <chudler> I think you'll want to check out the dev guide, especially chapter 3
[15-Sep-2009 14:58:18] <djack> ckrough is it usable for external status pagees usage
[15-Sep-2009 14:58:35] <ckrough> djack: kinda depends on what the external status page is
[15-Sep-2009 14:58:40] <ckrough> it just gives you raw data
[15-Sep-2009 14:58:47] <chudler> docs/DOC-3136
[15-Sep-2009 14:59:01] <djack> raw data is good
\n[15-Sep-2009 14:59:21] <ckrough> what chudler said, plus browse\nthrough the API for whatever version you use.\nhttp://www.zenoss.com/community/docs/zenoss-api-docs/2.3/
[15-Sep-2009 14:59:36] <ckrough> djack: there's a good bit of detective work involved
[15-Sep-2009 15:01:32] <djack> ckrough i've seen php an that page ;-)
[15-Sep-2009 15:01:34] <ckrough> djack: you can do things like getRRDValue(ccmHistoryRunningLastSaved), getManageIp(), etc...
[15-Sep-2009 15:01:39] <djack> that's more my knowledge
[15-Sep-2009 15:03:58]\n<djack> what i basicly want is server and green orange or red in\nleft side of the page and the events in the right side of the screen ;-)
[15-Sep-2009 15:04:29] <djack> if i can set that up then I'm happy and my salary can go up :-p
[15-Sep-2009 15:05:30] <djack> ckrough i've seen but I need to see some examples before i get it ...
[15-Sep-2009 15:09:33] <djack> and all the examples are about pushing data to zenoss
[15-Sep-2009 15:09:49] <rmatte> yeh, from what I've seen
[15-Sep-2009 15:09:56] <rmatte> like calling on Zenoss to move events to history and such
[15-Sep-2009 15:11:31] <djack> i dont want to do that just query device status outside zenoss platform
[15-Sep-2009 15:11:51] <chudler> the first example on the URL I previously posted includes query for data, does it not?
[15-Sep-2009 15:11:56] <chudler> http://USERNAME:PASSWORD@MY_ZENOSS_HOST:8080/zport/dmd/Devices/Server/Linux/devices/angel/getRRDValue?dsname=laLoadInt5_laLoadInt5
[15-Sep-2009 15:12:25] <djack> chudler also all then :-p
[15-Sep-2009 15:12:52] <rmatte> djack: so you want up or down status basically?
[15-Sep-2009 15:15:00] <rmatte> Here's a python script that I wrote to retrieve status on devices, along with an example of it running...
[15-Sep-2009 15:15:02] <rmatte> http://pastebin.com/m515c3de3
[15-Sep-2009 15:15:23] <rmatte> it pulls device name, IP, status (UP/DOWN), and uptime
[15-Sep-2009 15:17:58] <djack> hmmm rmatte thanks
[15-Sep-2009 15:18:51] <rmatte> np
\n[15-Sep-2009 15:19:23] <rmatte> that's really only for ping\nstatus, but that's generally all you'd care about for basic device\nstatus, then the event console would be for any other issues
[15-Sep-2009 15:19:47] <rmatte> the event console shouldn't be too hard since the events are stored in MySQL
\n[15-Sep-2009 15:20:27] <rmatte> so if you're doing it in php you\njust need to tap in to the database, pull the rights values out, and\ndisplay them
[15-Sep-2009 15:20:36] <rmatte> you'll probably have to convert some of them too
[15-Sep-2009 15:21:05] <djack> hmmm
[15-Sep-2009 15:21:06] <djack> nice
\n[15-Sep-2009 15:21:26] <rmatte> This may be of use to you:\nhttp://dev.zenoss.org/trac/browser/branches/zenoss-2.4.x/Products/ZenEvents/db/zenevents.sql
[15-Sep-2009 15:21:51] <ckrough> If your doing it remotely you'll need to use REST calls or xmlrpc
[15-Sep-2009 15:21:59] <ckrough> xmlrpc is the preferred
\n[15-Sep-2009 15:22:38] <rmatte> ckrough: well, either that or do\nsome sort of setup where he has the php interface call on scripts\nremotely via ssh, and pass the data to it
[15-Sep-2009 15:22:42] <rmatte> but that's a bit patchwork
[15-Sep-2009 15:22:56] <rmatte> still, if it works, it works
[15-Sep-2009 15:23:04] <ckrough> yeah, reinventing the wheel at that point
\n[15-Sep-2009 15:23:42] <rmatte> yeah, but if you're like me, and\nyou're not very familliar with rest of xmlrpc, then it might be the\nsimpler solution
[15-Sep-2009 15:23:56] <rmatte> rest or* xmlrpc rather
[15-Sep-2009 15:24:38] <djack> ckrough beside the getRRDthingy is there a reference page with all calls for rest
[15-Sep-2009 15:25:02]\n<chudler> everyone knows about REST, since it is the way the web\nworks, at least in theory. xmlrpc has some learning curve, but it is\ndefinitely possible to use it without knowing anything abou tit
[15-Sep-2009 15:25:06] <ckrough> unfortunately no, I usuall browse through the API doc trying different methods. some work, some dont
[15-Sep-2009 15:25:33] <djack> well the page will be on hosted on the same machine as the zenoss
[15-Sep-2009 15:25:47] <chudler> djack: really? use zendmd!
[15-Sep-2009 15:25:48] <djack> i have memory and cpu power to spend
[15-Sep-2009 15:26:02] <djack> chudler ye[
[15-Sep-2009 15:26:02] <ckrough> then straight up python and zendmd is nice and direct
[15-Sep-2009 15:26:31] <djack> ckrough mail me the python for dummies book then :-p
[15-Sep-2009 15:26:45] <djack> however maybe it's nice to learn a new language
[15-Sep-2009 15:27:02] <rmatte> well, that script that I pasted, is pretty much what you'd use, and then you'd access mysql directly
[15-Sep-2009 15:27:09] <ckrough> djack: www.diveintopython.org
[15-Sep-2009 15:27:20] <ckrough> http://diveintopython.org/toc/index.html
\n[15-Sep-2009 15:27:21] <chudler> yeah, maybe. one of python's\nmain points is to be very easy to read and learn. Despite that, it is\nstill a great overall language
[15-Sep-2009 15:28:07] <ckrough> argh.. all these great zenoss links are broken in teh googles
[15-Sep-2009 15:28:10] <djack> chudler that is one thing that is correct the script of rmatte i can read and understand
[15-Sep-2009 15:28:27] <djack> ckrough yeah somebody needed to update the webpage :-p
[15-Sep-2009 15:28:29] <mrayzenoss> ckrough: google will catch up soon I hope
[15-Sep-2009 15:28:37] <ckrough> I know. just griping.
[15-Sep-2009 15:28:52] <ckrough> I keep finding what djack needs then having to play detective to get the link.
[15-Sep-2009 15:29:02] <rmatte> djack: yeh, it's a really basic script though, easy to work with
\n[15-Sep-2009 15:29:19] <djack> we had the same shit when we\nmigrated our website .... not nice if you are a webshop that sells cd\ndvd and games
[15-Sep-2009 15:29:22] <rmatte>\ndjack: If you saw the availability report script that I finished\nwriting yesterday you might cry (I almost did)
[15-Sep-2009 15:29:54] <djack> rmatte is it usefull to share
[15-Sep-2009 15:30:04] <djack> who know i can learn from it ;-)
[15-Sep-2009 15:30:31] <ckrough> djack: docs/DOC-3136
[15-Sep-2009 15:30:51] <ckrough> mrayzenoss: I think this room needs a bot that snarfs those urls and posts the title. ?
[15-Sep-2009 15:31:10] <mrayzenoss> there was one... I don't know what happened to it
[15-Sep-2009 15:31:23] * ckrough pours a drink for dead bot
[15-Sep-2009 15:31:56] <ckrough> djack: #3.1.2.3.3. PHP on that last link I sent
[15-Sep-2009 15:32:28] <ckrough> same methodology works for getting data back out
\n[15-Sep-2009 15:33:55] <djack> ckrough seen it and also seen how\ni can call python script from php so i will figure it out
[15-Sep-2009 15:34:12] <djack> hmmm wife does not going to like this new language :-p
[15-Sep-2009 15:34:22] <ckrough> ha
[15-Sep-2009 15:34:42] <chudler> LOL, yeah same here. Every time I bring home a new book, I have to hear about it..
\n[15-Sep-2009 15:34:53] <ckrough> I resisted python for a while.\nPreferred PHP for the most part, but now Im using python for most stuff
[15-Sep-2009 15:35:22] <djack> chudler the list is becomming long ... need to pass the new mcse and vmware
[15-Sep-2009 15:35:37] <ckrough> ..the never ending "to read" list.
[15-Sep-2009 15:36:15] <djack> idd
[15-Sep-2009 15:36:29] <djack> ckrough what is the advantage ...?
[15-Sep-2009 15:36:39] <ckrough> of?
[15-Sep-2009 15:37:00] <djack> try to pull me over to the dark side of python
[15-Sep-2009 15:37:21] <djack> maybe the force is with me
\n[15-Sep-2009 15:37:33] <ckrough> oh. it's a smart language. hard\nto put my finger on it. whenever I cant do something I find that I was\ntrying it wrong and that there is some elegant way to do it
[15-Sep-2009 15:37:50] <ckrough> and it forces you to make clean code
[15-Sep-2009 15:38:06] <ckrough> you'll spend a month trying not to use ';' tho.
[15-Sep-2009 15:38:10] <djack> ckrough not one code is made to be clean ...
[15-Sep-2009 15:38:28] <djack> haha yeah i have that with bash also ....
[15-Sep-2009 15:39:10] <djack> well i learned php in 4 months so python will work also ....
[15-Sep-2009 15:46:48] <djack> hey on the forum i lots of strange codes i staid of code how come ?
[15-Sep-2009 15:50:39] <mrayzenoss> the code got munged, we've got a ticket open
[15-Sep-2009 15:50:45] <mrayzenoss> to get them fixed
[15-Sep-2009 15:51:27] <djack> oki
[15-Sep-2009 15:51:37] <djack> going to bed
[15-Sep-2009 16:11:11] <mrayzenoss> ckrough: I've found a work-around for what you were reporting
[15-Sep-2009 16:11:29] <mrayzenoss> paste the URL into google, it'll give you the link and the topic
[15-Sep-2009 16:11:38] <mrayzenoss> copy and paste the topic into the new forums search
[15-Sep-2009 16:11:43] <mrayzenoss> profit
[15-Sep-2009 16:59:04] <rmatte> well, I'm out of here
[15-Sep-2009 16:59:06] <rmatte> later
[15-Sep-2009 17:22:05] <cgibbons> hrm
[15-Sep-2009 17:37:27] <BillBinko> hi all
[15-Sep-2009 17:37:45] <BillBinko> anyone know how to get two snmp agents on the same box modelled as different devices?
[15-Sep-2009 17:38:06] <BillBinko> they are on separate ports
[15-Sep-2009 17:45:36] <markeriv> hmm
[15-Sep-2009 17:45:44] <markeriv> when you say seperate ports
[15-Sep-2009 17:45:48] <markeriv> do you mean seperate IPs?
[15-Sep-2009 17:45:53] <markeriv> or just different port #s
\n[15-Sep-2009 17:46:29] <markeriv> cause if you are trying to pull\nstats on two different NICs, that should not be too difficult
[15-Sep-2009 18:03:25] <BillBinko> sorry was away for a second
[15-Sep-2009 18:03:31] <BillBinko> I mean separate UDP ports
[15-Sep-2009 18:03:37] <BillBinko> same nic
[15-Sep-2009 18:03:46] <BillBinko> I want to hit the system SNMP agent (port 161)
[15-Sep-2009 18:03:54] <BillBinko> and the JVM's SNMP agent (port 1161)
[15-Sep-2009 18:03:59] <BillBinko> both on the same IP
\n[15-Sep-2009 18:20:45] <BillBinko> I have to head to dinner -- if\nanyone has a simple answer, could they email me at my name at gmail?
[15-Sep-2009 18:20:50] <BillBinko> I will check back tomorrow
[15-Sep-2009 19:08:05] <scp2004> Hi Guys !
\n[15-Sep-2009 19:16:23] <scp2004> I need add a user in zenoss ,\nthat only can see the devices i configure too he see... Its possible ?
[15-Sep-2009 19:41:47] <ironpaw> i
[15-Sep-2009 19:41:49] <ironpaw> hi
[15-Sep-2009 19:42:12] <ironpaw> you could try defining the administrative servers for the usre
[15-Sep-2009 19:42:26] <ironpaw> nto sure if that would restrict visibility your after thougjh
\n[15-Sep-2009 19:54:25] <scp2004> I need add a user in zenoss ,\nthat only can see the devices i configure too he see... Its possible ?
[15-Sep-2009 19:56:32] <mrayzenoss> that's an Enterprise feature, you can create a read-only user with Zenoss Core
[15-Sep-2009 19:56:57] <mrayzenoss> you can look into the way Zope permissions work to do it by hand, but it's not pretty
\n[15-Sep-2009 19:57:50] <mrayzenoss> Jane Curry has a paper\n"Crafting Zenoss core users for events and zProperties" that may help\nhttp://www.skills-1st.co.uk/papers/jcurry.html
[15-Sep-2009 20:01:14] <scp2004> [mrayzenoss] one mor time ..Thank you Very much ..
[15-Sep-2009 22:40:33] <heckj> evening
[16-Sep-2009 00:00:46] [disconnected at Wed Sep 16 00:00:46 2009]
[16-Sep-2009 00:00:46] [connected at Wed Sep 16 00:00:46 2009]
[16-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
\n[16-Sep-2009 02:54:35] <frozty_sa> uhm, slight nitpickery, but\ndocs/DOC-2988 should be\ns/Distributed/Domain/, surely?
[16-Sep-2009 03:21:56] <salax> hai
[16-Sep-2009 03:21:58] <salax> hello
[16-Sep-2009 03:22:28] <salax> i've installed zenoss 64bit on ubuntu x64 server 8.04
[16-Sep-2009 03:22:56] <salax> i have 2 servers, 1 for vmware server, 1 for zenoss core
[16-Sep-2009 03:23:16] <salax> on top of the vmware server, i install ubuntu,freebsd and openbsd agent
[16-Sep-2009 03:23:42] <salax> everything is ok for now except for running process and list of software
[16-Sep-2009 03:24:05] <salax> on ubuntu, i manage to collect the datat thru snmp and ssh
[16-Sep-2009 03:24:19] <salax> but for the BSDs, i cant connect thru ssh
[16-Sep-2009 03:24:55] <salax> my error stated " No command plugins found for freebsd"
[16-Sep-2009 03:25:06] <salax> and also for openbsd
[16-Sep-2009 03:25:23] <salax> how can i activate or install the cmd plugin?
[16-Sep-2009 03:26:06] <salax> any idea guys? much appreciated, tq
[16-Sep-2009 06:26:55] <Apocalipse> hello people
[16-Sep-2009 06:27:16] <Apocalipse> i am still having problems with linking devices in diff locations...
[16-Sep-2009 06:27:42] <Apocalipse> does anybody knows how to static link then?
[16-Sep-2009 06:33:22] <fred_> hello
[16-Sep-2009 06:33:40] fred_ is now known as Guest84119
[16-Sep-2009 06:33:58] Guest84119 is now known as jo-go
[16-Sep-2009 06:34:05] <jo-go> anyone here ?
[16-Sep-2009 06:34:14] <Troubadix09> hello jo-go
\n[16-Sep-2009 06:34:40] <jo-go> hi,just demonstrating IRC to\nsomeone and proving that real people live here - thanks for the reply
[16-Sep-2009 06:35:29] <Troubadix09> jo-go: is this a zenoss demonstrating?
\n[16-Sep-2009 06:35:51] <jo-go> yeah, just showing off zenoss and\nshowing where to find the helpful people should a question need to be\nasked
[16-Sep-2009 06:36:13] <Apocalipse> hun?
[16-Sep-2009 06:36:51] <jo-go> off now - thanks for helping
[16-Sep-2009 06:37:00] <Troubadix09> jo-go: bye
[16-Sep-2009 06:37:10] <Apocalipse> gone
[16-Sep-2009 06:37:11] <Apocalipse>
[16-Sep-2009 06:38:05] <Troubadix09> Apocalipse: can you explain your problem a little bit more
[16-Sep-2009 06:38:07] <Troubadix09> ?
[16-Sep-2009 06:38:12] <Apocalipse> ok
[16-Sep-2009 06:38:35] <Apocalipse> i have some locations in my network
[16-Sep-2009 06:39:01] <Apocalipse> i added 3 devices to interconnect then
[16-Sep-2009 06:39:07] <Apocalipse> 3 cisco routers
[16-Sep-2009 06:39:30] <Apocalipse> 1 is core router....
[16-Sep-2009 06:40:01] <Apocalipse> the connection between then is using ATM link
[16-Sep-2009 06:40:54] <Apocalipse> i modelled then
[16-Sep-2009 06:41:41] <Apocalipse> it seen ok... i am collecting data... men... cpu... link state... link capcity... etc
[16-Sep-2009 06:41:42] <Apocalipse> but...
[16-Sep-2009 06:41:52] <Apocalipse> zenoss cannot link then...
[16-Sep-2009 06:42:01] <Apocalipse> for example...
[16-Sep-2009 06:42:11] <Apocalipse> let me try to illustrate...
[16-Sep-2009 06:43:33] <Apocalipse> location 1: core router, lan -> x.x.x.x/20 , atm net -> y.y.y.y/30
[16-Sep-2009 06:44:12] <Apocalipse> location 2: border router , lan -> z.z.z.z/20 , atm net -> y.y.y.a/30
[16-Sep-2009 06:45:01] <Apocalipse> both atm interfaces are in same network...
[16-Sep-2009 06:45:25] <Apocalipse> got it?
[16-Sep-2009 06:52:24] <Apocalipse> hey people...
[16-Sep-2009 06:52:40] <Apocalipse> what happens with zenoss website?
[16-Sep-2009 07:32:05] jabular_ is now known as jabular
[16-Sep-2009 08:07:09] <Troubadix09> bye all
\n[16-Sep-2009 09:18:16] <Brad_K> I'm still trying to figure out\nwhy the Nagios-based plugin check_email_delivery isn't storing any data\nin the rrd database. is there anyone here who might be able to help\nanswer some questions?
[16-Sep-2009 09:19:18]\n<tyler79> is there a way to embed a portlet in another page, say\non a custom company intranet site?
[16-Sep-2009\n09:27:42] <Brad_K> I'm still trying to figure out why the\nNagios-based plugin check_email_delivery isn't storing any data in the\nrrd database. is there anyone here who might be able to help answer\nsome questions?
[16-Sep-2009 09:28:32] <mrayzenoss> Brad_K: is it returning multi-line data?
[16-Sep-2009 09:28:46] <mrayzenoss> that fix just went into 2.5 last week
[16-Sep-2009 09:28:46] <rmatte> Brad_K: yeh, what does the output of the script look like?
[16-Sep-2009 09:28:47] <Brad_K> no
[16-Sep-2009 09:28:57] <Brad_K> hang on...
\n[16-Sep-2009 09:30:58] <Brad_K> When funcioning normally,\naccording to the zencommand.log file, the only ouput it generates looks\nlike this:
[16-Sep-2009 09:31:01] <Brad_K> 2009-09-16 09:22:15 DEBUG zen.zencommand: Output: 'DELIVERY OK|ELAPSED=7 DELAY=7
[16-Sep-2009 09:31:04] <Brad_K> \\n'
[16-Sep-2009 09:31:35] <rmatte> Brad_K: not sure why that \\n' is there at the end
[16-Sep-2009 09:31:39] <rmatte> it shouldn't be
[16-Sep-2009 09:31:57] <rmatte> that constitutes multiline
[16-Sep-2009 09:31:58] <Brad_K> weird line-wrap from the cut-n-paste from the zencommand.log
[16-Sep-2009 09:32:17] <rmatte> ah ok, but if you run a test on the datasource, do you see that?
[16-Sep-2009 09:32:35] <rmatte> the entire output should just be: DELIVERY OK|ELAPSED=7 DELAY=7
[16-Sep-2009 09:32:54] <Brad_K> No "\\n" at the end of the line?
\n[16-Sep-2009 09:32:58] <rmatte> and you need to have 2 datapoints\nunder the datasource called ELAPSED and DELAY (case sensitive)
[16-Sep-2009 09:33:07] <rmatte> no, you don't need the \\n at the end
[16-Sep-2009 09:33:22] <Brad_K> okay, i'll need to go fix the script.
[16-Sep-2009 09:33:27] <rmatte> Zenoss most likely won't interpret the data properly because of it
[16-Sep-2009 09:33:44] <rmatte> should be a quick change to fix that
[16-Sep-2009 09:36:23] <Brad_K> Sorry, i've got to go to a quick meeting. bbiab.
[16-Sep-2009 09:36:34] <Brad_K> it's related to this very subject, btw. ;-)
[16-Sep-2009 09:52:08] <twm1010> crickets chirping
[16-Sep-2009 09:52:20] <rmatte> tumbleweed
[16-Sep-2009 09:53:30] <mrayzenoss> fine, you want something to talk about...
[16-Sep-2009 09:53:35] <mrayzenoss> beta 2 is published
[16-Sep-2009 09:53:43] <mrayzenoss> everything except VMware
[16-Sep-2009 09:54:17] <rmatte> stack?
[16-Sep-2009 09:54:41] <rmatte> ooh, it is
[16-Sep-2009 09:54:42] <rmatte> sic
[16-Sep-2009 09:55:15] <rmatte> time to load up my test box and get cracking
[16-Sep-2009 09:57:40] <mrayzenoss> everyone coming to Jane's webinar today?
[16-Sep-2009 09:57:52] <ckrough> yup
[16-Sep-2009 09:58:07] <mrayzenoss> 40+ attendees, we may do an encore presentation
[16-Sep-2009 09:58:59] <ckrough> guess Ill go try and make webex work with linux now...
[16-Sep-2009 09:59:11] <mrayzenoss> I hope it works
[16-Sep-2009 09:59:15] <mrayzenoss> Jane's on SUSE
[16-Sep-2009 09:59:40] <ckrough> its worked as a viewer in the past with minimal issues
\n[16-Sep-2009 09:59:56] <ckrough> couldnt get it working as a\npresenter last time I tried, but I had a windows box handy so I didnt\ntry hard
[16-Sep-2009 10:01:11] <rmatte> It's fun pulling down Zenoss at over 8MB/s
[16-Sep-2009 10:01:33] <rmatte> Sourceforge have some nice mirrors
[16-Sep-2009 10:01:46] <mrayzenoss> yeah they do
[16-Sep-2009 10:01:47] <ckrough> it was rough the other day, I was in low low KBs
[16-Sep-2009 10:02:10] <jb> oops
[16-Sep-2009 10:02:12] <jb> i erased my email
[16-Sep-2009 10:02:20] <jb> mray, can you post me the link to the webinar?
[16-Sep-2009 10:02:41] * rmatte thinks jb should add it to his calendar next time
[16-Sep-2009 10:02:50] <mrayzenoss> which is one of the reasons they seem reluctant to add BitTorrent
[16-Sep-2009 10:03:06] <mrayzenoss> jb: Zenoss Community Webinar this Wednesday: Zenoss Event Architecture: http://tinyurl.com/myesxm
[16-Sep-2009 10:03:14] <mrayzenoss> it's on the blog post
[16-Sep-2009 10:03:23] * rmatte is installing King Crab, can't wait to see this thing
[16-Sep-2009 10:03:40] <jb> hmm
[16-Sep-2009 10:03:47] <jb> oh its not for another hour
[16-Sep-2009 10:03:55] <rmatte> yeh, noon eastern time
[16-Sep-2009 10:04:02] <rmatte> 11 central
[16-Sep-2009 10:04:38] <rmatte> and it'll be like 5pm Jane's time
[16-Sep-2009 10:08:52] <rmatte> http://dmon.org/graphics/girlfriend.jpg
[16-Sep-2009 10:10:01] <ckrough> ...do I want to click that
[16-Sep-2009 10:10:02] <twm1010> i just registered
[16-Sep-2009 10:10:11] <rmatte> ckrough: it's SFW
[16-Sep-2009 10:10:25] <twm1010> rmatte: i almost fell out of my chair
[16-Sep-2009 10:10:32] <rmatte> hehe
[16-Sep-2009 10:10:33] <ckrough> Paladin, pfft.. hybrid classes always suck
[16-Sep-2009 10:10:42] <ckrough>
[16-Sep-2009 10:11:09] <rmatte> hehe
[16-Sep-2009 10:11:50] <rmatte> http://dmon.org/graphics/windowsvistamarketing.jpg
[16-Sep-2009 10:12:08] <rmatte> http://dmon.org/graphics/wires.jpg
\n[16-Sep-2009 10:12:49] <rmatte> Oh, and for anyone who has seen\nthe whitest kids u know tattoo parlour skit:\nhttp://dmon.org/graphics/wkuktat.jpg
[16-Sep-2009 10:13:45] <twm1010> im looking at your zenoss screenshots :O
[16-Sep-2009 10:13:51] <rmatte> hehe
[16-Sep-2009 10:14:17] <rmatte> yeh, they are pretty decent
[16-Sep-2009 10:14:19] <twm1010> MPLS?
[16-Sep-2009 10:14:24] <twm1010> looks like totally mesh networks
[16-Sep-2009 10:14:29] <ckrough> I like schoolgirl.jpg...
[16-Sep-2009 10:14:43] <rmatte> well, the larger map is partial MPLS (they use it in Mexico)
[16-Sep-2009 10:14:50] <rmatte> the smaller map is all MPLS
\n[16-Sep-2009 10:14:57] <ckrough> bah, and using my regular nick\ntoo... comment google indexed for the rest of time. stupid internet.
[16-Sep-2009 10:15:39] <rmatte> lol
\n[16-Sep-2009 10:15:53] <twm1010> Yeah, I've been amazed when\ndoing some Zenoss related searching how many times my own name comes up\nin the google searches from captured chats from this same chatroom
[16-Sep-2009 10:16:13] <rmatte> yeh, I'm in here like all the time, so mine does as well
[16-Sep-2009 10:16:44] <rmatte> interesting, my zenoss community profile is already the top result on google for "rmatte"
[16-Sep-2009 10:16:46] <rmatte> lol
[16-Sep-2009 10:17:17] <ckrough> mrayzenoss has been spamming the google 'reindex' button.
[16-Sep-2009 10:17:33] <ckrough> yup, mine is first result for ckrough
[16-Sep-2009 10:17:41] <mrayzenoss> you guys are internet famous!
[16-Sep-2009 10:17:54] <ckrough> assuming your searching for ckrough, ya
\n[16-Sep-2009 10:17:55] <rmatte> If I search for Ryan Matte, my\nLinkedIn profile is 4th result, and the blog entry of Eric and I being\nnamed Zenoss Masters is 5th.
[16-Sep-2009 10:17:56] <ckrough>
[16-Sep-2009 10:18:40] <mrayzenoss> hmmm.. I make the first page of google just barely
[16-Sep-2009 10:18:46] <rmatte> hehe
[16-Sep-2009 10:19:03] <mrayzenoss> but if you throw in "Linux", I'm the whole page
[16-Sep-2009 10:19:09] <rmatte> oh, for the path of least resistance blog
[16-Sep-2009 10:19:15] <ckrough> lol. searching my first name returns this: http://www.myspace.com/wchrisw
[16-Sep-2009 10:19:18] <ckrough> and that is NOT me
[16-Sep-2009 10:19:23] <ckrough> poser.
[16-Sep-2009 10:19:35] <rmatte> lmao
[16-Sep-2009 10:20:04] <rmatte> Yeh, well the first result for Ryan Matte is http://www.facebook.com/people/Ryan-Matte/748530391
[16-Sep-2009 10:20:06] <rmatte> which is not me
[16-Sep-2009 10:20:08] <rmatte> lol
[16-Sep-2009 10:20:31] <rmatte> ...and this one is sooooo not me: http://www.facebook.com/people/Ryan-Matte/713617305
[16-Sep-2009 10:21:04] <rmatte> I'm on Twitter?
[16-Sep-2009 10:21:27] <rmatte> oh, right, when I posted to the Modern Warfare 2 twitter page the one time
[16-Sep-2009 10:22:02] <rmatte> and apparently I have 3 followers regardless of the fact that I've only made 1 post
\n[16-Sep-2009 10:22:38] <rmatte> oh, and a bug report that I\nlogged 2 years ago for Amarok:\nhttp://markmail.org/message/73dhm2rrcewtsrgp
[16-Sep-2009 10:22:50] <twm1010> odd, informant SNMP doens't work on one of my 2K3 boxes
[16-Sep-2009 10:22:52] <rmatte> interesting what makes the first page in Google sometimes
\n[16-Sep-2009 10:22:56] <twm1010> cpu check returns No more\nvariables left in this MIB View (It is past the end of the MIB tree)
[16-Sep-2009 10:23:15] <rmatte> twm1010: sounds like you need my ZenPack
[16-Sep-2009 10:23:30] <twm1010> which?
[16-Sep-2009 10:23:44] <twm1010> confused why only one host refuses to turn this OID
[16-Sep-2009 10:23:59] <rmatte> docs/DOC-3386
\n[16-Sep-2009 10:24:20] <rmatte> there will be a simpler version\nof that ZenPack up as soon as Matt actually has time to review all of\nthe ones sitting in his inbox
[16-Sep-2009 10:24:35] <rmatte> that one is for multicore, the other is for single core
[16-Sep-2009 10:24:41] <rmatte> errr, well, for monitoring total
[16-Sep-2009 10:24:44] <rmatte> rather than each core
[16-Sep-2009 10:25:01] <rmatte> http://dmon.org/downloads/zenoss/ZenPacks.Nova.Windows.SNMPPerfMonitor.Simple-1.2-py2.4.egg
[16-Sep-2009 10:25:05] <rmatte> there's a link to the other one
[16-Sep-2009 10:26:03]\n<rmatte> the simple one adds a template to /Server/Windows that\nmonitors total CPU usage rather than each core, the one that's up on\nthe community site created organizers for 1 to 16 CPUs/cores, that you\nthen move your devices under appropriately
[16-Sep-2009 10:27:09] <twm1010> ah, SNMP monitoring without informant?
[16-Sep-2009 10:27:10] <twm1010> very nice
[16-Sep-2009 10:27:29] <rmatte> correcto
[16-Sep-2009 10:27:40] <rmatte> you just need to have bc and snmpwalk installed on the system
[16-Sep-2009 10:27:45] <rmatte> since the scripts use those
[16-Sep-2009 10:28:08] <twm1010> hrmm
[16-Sep-2009 10:28:40] <twm1010> hrmm, i think informant DLL is tied up on that box
[16-Sep-2009 10:33:00] <ckrough> I think it's time to move my events db to a dedicated box.
[16-Sep-2009 10:33:22] <ckrough> Maybe even figure out a way to kick zenhub out of the main box... clustering some how.
[16-Sep-2009 10:35:34] <twm1010> bah, i concede defeat, the server needs a reboot
[16-Sep-2009 10:43:35] <mrayzenoss> WebEx is up for the webinar
[16-Sep-2009 10:46:48] <rmatte> I'll be in shortly
[16-Sep-2009 10:48:20] <ckrough> Is there a conference bridge for audio? WebEx isnt liking Fedora11 for audio
[16-Sep-2009 10:48:38] <rmatte> ckrough: yeh, there is
[16-Sep-2009 10:48:51] <mrayzenoss> yeah, there's a phone number
[16-Sep-2009 10:48:52] <rmatte> I always dial in since I'm on Ubuntu
[16-Sep-2009 10:48:58] <rmatte> and my speakers blow
[16-Sep-2009 10:49:17] <rmatte> I think they were manufactured in 1976
[16-Sep-2009 10:49:21] <ckrough> had that information been shared already? I dont see it
[16-Sep-2009 10:50:01] <rmatte> don't think so
[16-Sep-2009 10:50:04] <rmatte> it will be though
[16-Sep-2009 10:50:05] <ckrough> ok
[16-Sep-2009 10:51:10] <mrayzenoss> 866-699-3239
[16-Sep-2009 10:51:23] <mrayzenoss> or 1-408-792-6300
[16-Sep-2009 10:51:34] <rmatte> k
[16-Sep-2009 10:51:35] <mrayzenoss> access code 759 626 698
[16-Sep-2009 10:52:02] <ckrough> thanks
[16-Sep-2009 11:11:10] <twm1010> noise cancelling headphones
[16-Sep-2009 11:11:59] <twm1010> mrayzenoss: will this webinar be saved for download?
[16-Sep-2009 11:12:06] <mrayzenoss> yeah
[16-Sep-2009 11:12:17] <mrayzenoss> it's being recorded, there will be an encore session next week
[16-Sep-2009 11:15:44] <ckrough> my eyes
[16-Sep-2009 11:15:49] <mrayzenoss> heh
[16-Sep-2009 11:16:01] <rmatte> lol
[16-Sep-2009 11:25:14] <Apocalipse> hello
[16-Sep-2009 11:25:37] <jb> hmm
[16-Sep-2009 11:25:39] <Apocalipse> i would like to thx matt...
[16-Sep-2009 11:25:44] <Apocalipse> thx for help
[16-Sep-2009 11:25:45] <Apocalipse>
[16-Sep-2009 11:26:49] <jb> if I have a datasource returning "0.054"
[16-Sep-2009 11:27:27] <jb> and I want this to be graphed like "50.4ms"
[16-Sep-2009 11:27:48] <jb> right now im using a format of "%5.2lf%s"
[16-Sep-2009 11:27:58] <jb> and the result is something like "17.3m"
[16-Sep-2009 11:28:01] <jb> what is the "m" ?
[16-Sep-2009 11:28:08] <ckrough> micro
[16-Sep-2009 11:28:12] <ckrough> err milli
[16-Sep-2009 11:28:27] <jb> so its really like "17.3milliseconds/ms"
[16-Sep-2009 11:28:28] <ckrough> it's dividing a small datapoint out over the polling cycle
[16-Sep-2009 11:28:32] <jb> (or ms)
[16-Sep-2009 11:29:15] <ckrough> millipackets/second
[16-Sep-2009 11:30:57] <jb> ok, you have any idea on how to get that in ms (milliseconds)?
[16-Sep-2009 11:31:01] <ckrough> or milliwhatever/sec
[16-Sep-2009 11:31:28] <ckrough> what are you measuring?
[16-Sep-2009 11:31:37] <jb> disk latency
\n[16-Sep-2009 11:32:20] <ckrough> you may be seeing the right\nthing already then. Ive dealt with that with units/time measurement,\nbut not latency
[16-Sep-2009 11:32:40] <jb> well, the value returned is something like "0.054"
[16-Sep-2009 11:32:45] <jb> which is really 50.4ms
[16-Sep-2009 11:33:30] <rmatte> you could remove the %s at the end and set units to "seconds"
[16-Sep-2009 11:33:36] <rmatte> then it would graph fractions of a second
[16-Sep-2009 11:34:21] <jb> hrm yeah
[16-Sep-2009 11:34:31] <jb> has to be a way to get it to show MS tho
[16-Sep-2009 11:37:49] <rmatte> oh, I'm sure there is, I just don't fully understand the formatting
[16-Sep-2009 11:37:57] <rmatte> so I'm not a huge help hehe
[16-Sep-2009 11:38:08] <rmatte> I'll have to read up on it
[16-Sep-2009 11:38:31] <jb> CRAP
[16-Sep-2009 11:38:33] <jb> i forgot about the webinar
[16-Sep-2009 11:38:34] <jb> :/
[16-Sep-2009 11:38:42] <mrayzenoss> there will be an encore next wek
[16-Sep-2009 11:38:43] <mrayzenoss> week
[16-Sep-2009 11:38:48] <rmatte> we're about half way through
[16-Sep-2009 11:39:36] <markeriv> hey guys
[16-Sep-2009 11:39:44] <markeriv> I have a question concerning users and alerts
\n[16-Sep-2009 11:39:58] <markeriv> is there a way to tie a user to\na group so that they only recieve alerts from that group?
[16-Sep-2009 11:40:17] <markeriv> i have RTFMing and have not yet found an area specifically detailing that
[16-Sep-2009 11:40:18] <rmatte> markeriv: a device group you mean?
[16-Sep-2009 11:40:22] <mrayzenoss> you can assign administrators to specific groups
[16-Sep-2009 11:40:52] <markeriv> not a device group
\n[16-Sep-2009 11:40:58] <rmatte> markeriv: not sure I fully\nunderstand what you're asking, are you asking abour organizational\ngroups for users?
[16-Sep-2009 11:41:04] <markeriv> hmm
[16-Sep-2009 11:41:05] <rmatte> about*
[16-Sep-2009 11:41:11] <markeriv> you know how you can browse by systems
[16-Sep-2009 11:41:11] <markeriv> groups
[16-Sep-2009 11:41:13] <markeriv> locations
[16-Sep-2009 11:41:16] <markeriv> or networks
[16-Sep-2009 11:41:17] <rmatte> yes
[16-Sep-2009 11:41:22] <markeriv> i have a group
[16-Sep-2009 11:41:24] <jb> just configure the alert
[16-Sep-2009 11:41:26] <markeriv> we will call them emser tile
[16-Sep-2009 11:41:28] <jb> to only send for systems in that group
[16-Sep-2009 11:41:37] <markeriv> ok
[16-Sep-2009 11:41:43] <rmatte> yeh, you want to add a filter rule
[16-Sep-2009 11:41:55] <rmatte> something like "group is /bleh/blah"
[16-Sep-2009 11:42:01] <rmatte> to the alert
[16-Sep-2009 11:42:32] <markeriv> hmm
[16-Sep-2009 11:42:35] <markeriv> i see that area
[16-Sep-2009 11:42:40] <markeriv> but group is not an option
[16-Sep-2009 11:42:49] <rmatte> hunh?
[16-Sep-2009 11:42:52] <markeriv> ahh
[16-Sep-2009 11:42:55] <markeriv> its device groups
[16-Sep-2009 11:42:57] <markeriv> not just groups
[16-Sep-2009 11:43:01] <markeriv> sometimes i feel stupid
[16-Sep-2009 11:43:05] <markeriv> this is one of those times
[16-Sep-2009 11:43:06] <rmatte> yeh, device group
[16-Sep-2009 11:43:10] <rmatte> lol
[16-Sep-2009 11:43:48] <rmatte> (I don't have Zenoss in front of me so I was just doing it off the top of my head)
[16-Sep-2009 11:54:36] mikey__ is now known as Michael1
[16-Sep-2009 11:54:43] <Michael1> hello?
[16-Sep-2009 11:55:13] <Apocalipse> hey people... to change the google map view...
\n[16-Sep-2009 11:55:30] <Michael1> does anyone know of an app than\ncan be run on a windows machine, monitor fans/temp/smart data and then\ngenerate traps to send to zenoss ?
[16-Sep-2009 11:57:04] <chemist> Michael1: look at the snmp-informant website
[16-Sep-2009 11:57:19] <rmatte> Michael1: yeh, snmp-informant can do that, just needs to be configured for it
[16-Sep-2009 11:57:20] <chemist> they have a hw monitoring tool there
[16-Sep-2009 12:01:43] <Michael1> looks like what i'm looking for - just going to install it....
[16-Sep-2009 12:17:26] <markeriv> Hey guys
[16-Sep-2009 12:17:29] <markeriv> got another question for you
[16-Sep-2009 12:17:38] <markeriv> What is your favorite zenoss feature and why?
[16-Sep-2009 12:17:40] <ckrough> hmm. isnt there a default "Production State" report?
[16-Sep-2009 12:20:39] <rmatte> ckrough: I just created one in the custom reports section
[16-Sep-2009 12:20:48] <rmatte> you just want to see the production states of all devices?
[16-Sep-2009 12:20:51] <ckrough> thought I had one
[16-Sep-2009 12:21:07] <ckrough> yeah, just put a couple thousand into maintenance and want to see a list to be sure
[16-Sep-2009 12:21:29] <rmatte> Reports -> Custom Device Reports -> Add Device Report, name it Production States
[16-Sep-2009 12:22:15] <rmatte> set Sort Sense as asc
[16-Sep-2009 12:22:26] <rmatte> set Sort Column as getId
[16-Sep-2009 12:22:32] <rmatte> in Columns put:
[16-Sep-2009 12:22:37] <rmatte> getId
[16-Sep-2009 12:22:37] <rmatte> getManageIp
[16-Sep-2009 12:22:41] <rmatte> getProdState
[16-Sep-2009 12:22:47] <rmatte> in Column Names put:
[16-Sep-2009 12:22:51] <rmatte> Device
[16-Sep-2009 12:22:51] <rmatte> IP Address
[16-Sep-2009 12:22:59] <rmatte> Production State
[16-Sep-2009 12:23:03] <rmatte> save it, and you're done
[16-Sep-2009 12:23:07] <ckrough> thanks
[16-Sep-2009 12:23:11] <rmatte> np
[16-Sep-2009 12:24:00] <ckrough> heh, there has got to be smoke coming out of this box right now
[16-Sep-2009 12:24:50] <rmatte> lol
[16-Sep-2009 12:27:32] <malbon> ls
[16-Sep-2009 12:29:05] <rmatte> .
[16-Sep-2009 12:29:07] <rmatte> ..
[16-Sep-2009 12:29:11] <rmatte> bin
[16-Sep-2009 12:29:13] <rmatte> etc
[16-Sep-2009 12:29:15] <rmatte> lib
[16-Sep-2009 12:29:19] <rmatte> usr
[16-Sep-2009 12:29:43] <rmatte> malbon@l33tbox#:
[16-Sep-2009 12:32:11] <etank> i think that i asked this before but can zenoss do the graph zooming that opennms can do?
[16-Sep-2009 12:32:25] <etank> where you click and drag over an area that you want to zoom to
[16-Sep-2009 12:34:18] <ckrough> whats the query syntax? "here.getProdState < 1000"
[16-Sep-2009 12:34:22] <ckrough> or dev.
[16-Sep-2009 12:34:27] <ckrough> tried a couple none worked
[16-Sep-2009 12:36:50] <rmatte> etank: you click to zoom, there's no drag zooming
[16-Sep-2009 12:37:01] <rmatte> etank: you can zoom in, zoom out, and move backwards or forward
[16-Sep-2009 12:43:11] <Apocalipse> ryan... zenoss is collecting a wrong info from a device...
[16-Sep-2009 12:43:28] <Apocalipse> i enter on the device and checked...
[16-Sep-2009 12:43:37] <Apocalipse> inside it is ok...
[16-Sep-2009 12:43:45] <Apocalipse> but zenoss get wrong...
[16-Sep-2009 12:44:34] <Apocalipse> i tryed re-model the device...
[16-Sep-2009 12:44:41] <Apocalipse> but still same
[16-Sep-2009 12:44:53] <Apocalipse> can i change this value manually?
[16-Sep-2009 12:47:42] <Apocalipse> ?
[16-Sep-2009 12:48:00] <etank> rmatte: right. but drag zoom would be nice
[16-Sep-2009 12:49:13] <twm1010> Apocalipse: what value?
[16-Sep-2009 12:49:43] <Apocalipse> ip address on interfaces
[16-Sep-2009 12:49:51] <Apocalipse> os tab
[16-Sep-2009 12:50:04] <Apocalipse> look this
[16-Sep-2009 12:50:21] <Apocalipse> IP-MIB::ipAdEntNetMask.10.x.x.y = IpAddress: 255.255.255.252
[16-Sep-2009 12:50:29] <Apocalipse> but in zenoss is:
[16-Sep-2009 12:50:50] <Apocalipse> 10.x.x.y/24
[16-Sep-2009 12:51:00] <Apocalipse> not /30
[16-Sep-2009 12:51:51] <twm1010> hrmm... one sec let me look at one of mine
[16-Sep-2009 12:51:53] <Apocalipse> got it?
[16-Sep-2009 12:51:57] <Apocalipse> ok
[16-Sep-2009 12:52:31] <ckrough> that's the group that zenoss displays it under in /networks you mean?
[16-Sep-2009 12:52:36] <twm1010> ah, ok, the ip column uses the /# to display the subnet
\n[16-Sep-2009 12:53:13] <twm1010> you would figure, that when you\nchange the mask on the server, it should change in zenoss
[16-Sep-2009 12:54:58] <twm1010> hard for me to find an example here, everything in our network IS /24
[16-Sep-2009 12:55:39] <twm1010> hrmm, our MPLS links show /30
[16-Sep-2009 12:56:56] <twm1010> you could delete the interface, and then remodel
[16-Sep-2009 12:57:01] <twm1010> see where that gets you
[16-Sep-2009 12:59:38] <Apocalipse> hunnnnnnnnnnnnnnnnnnnnnn
[16-Sep-2009 12:59:43] <Apocalipse> let me do it...
[16-Sep-2009 12:59:44] <Apocalipse> hol
[16-Sep-2009 12:59:46] <Apocalipse> hold
[16-Sep-2009 12:59:47] <Apocalipse>
[16-Sep-2009 13:02:52] <Apocalipse> received same
[16-Sep-2009 13:02:55] <Apocalipse>
[16-Sep-2009 13:05:42] <Apocalipse> is it possible to change this inside the database?
[16-Sep-2009 13:05:49] <Apocalipse> directlly i mean
[16-Sep-2009 13:06:23]\n<ckrough> Im not sure exactly what you are describing, Im jumping\nbetween windows. I just want to make sure you are not mistaking Zenoss'\nnetwork organization for an interface configuration
[16-Sep-2009 13:06:59]\n<ckrough> Zenoss nests devices in organization subnets that arent\nnecessarily matched to the subnets of your interfaces
[16-Sep-2009 13:11:36] <Apocalipse> hunnn
[16-Sep-2009 13:11:50] <Apocalipse> i saw all nested subnets...
[16-Sep-2009 13:12:02] <Apocalipse> there is right...
[16-Sep-2009 13:13:38] <rmatte> lovely, can't install any ZenPacks containing Mibs in King Crab
[16-Sep-2009 13:14:01] <Apocalipse> hunnn
[16-Sep-2009 13:14:40] <rmatte> I've been using King Crab for about 30 minutes and I've already found 5 issues with it
[16-Sep-2009 13:15:10] <rmatte> I have a feeling this is going to be a painful experience
[16-Sep-2009 13:15:25] <Apocalipse> changing subject... where can i change the view of maps?
[16-Sep-2009 13:15:27] <Apocalipse> map
[16-Sep-2009 13:15:35] <rmatte> what do you mean by change the view?
[16-Sep-2009 13:16:00] <Apocalipse> map, satelite, hybrid...
[16-Sep-2009 13:16:06] <Apocalipse> google map
[16-Sep-2009 13:16:12] <rmatte> you can't
[16-Sep-2009 13:16:16] <Apocalipse> i found an article... but i am not sure...
[16-Sep-2009 13:16:19] <rmatte> it's hard coded to be a map view
[16-Sep-2009 13:16:24] <Apocalipse> message/34678#34678
[16-Sep-2009 13:16:53] <rmatte> yeh, that would be the only way
[16-Sep-2009 13:16:56] <rmatte> by editing the actual code
[16-Sep-2009 13:16:59] <Apocalipse> yes
[16-Sep-2009 13:17:41] <rmatte> You'd want to add the this.map.addControl(new GMapTypeControl()); line
\n[16-Sep-2009 13:18:43] <rmatte> the reason they don't use\nsatellite view is because a lot of the satellite images are super low\nquality
[16-Sep-2009 13:18:55] <rmatte> so it kind of ruins the look
[16-Sep-2009 13:19:25] <ckrough> and much slower
[16-Sep-2009 13:19:31] <rmatte> yeh
[16-Sep-2009 13:19:32] <rmatte> that too
[16-Sep-2009 13:19:56] <ckrough> I dont use that portlet at all. Datacenters tend not to move around.
[16-Sep-2009 13:20:08] <rmatte> hehe
[16-Sep-2009 13:20:22] <Apocalipse> heheheheh
[16-Sep-2009 13:20:27] <Apocalipse> big boss WANTS
[16-Sep-2009 13:20:28] <rmatte> we just use it because it gives a good overhead view of our client's sites
[16-Sep-2009 13:20:38] <Apocalipse> i just say..."yes boss"
[16-Sep-2009 13:20:56] <Apocalipse> where i put this line ?
[16-Sep-2009 13:21:08] <Apocalipse> end?
[16-Sep-2009 13:21:11] <rmatte> it says right in the post
[16-Sep-2009 13:21:13] <rmatte> just read
[16-Sep-2009 13:21:40] <Apocalipse> i am confused about the post...
[16-Sep-2009 13:21:51] <rmatte> it's as clear as day
[16-Sep-2009 13:21:58] <Apocalipse> for u... sorry...
[16-Sep-2009 13:22:08] <rmatte> it says to go in to a specific file and put one of those 2 lines somewhere after line 24 in the file
[16-Sep-2009 13:22:12] <Apocalipse> yes
[16-Sep-2009 13:22:18] <rmatte> so what's the issue?
[16-Sep-2009 13:24:16] <rmatte> then to do the same with a second file
[16-Sep-2009 13:31:24] <Apocalipse> here i only have geomap.js
[16-Sep-2009 13:32:05] <twm1010> rmatte: you just reminded me to download king crab
[16-Sep-2009 13:32:23] <ckrough> any zope pros in here?
\n[16-Sep-2009 13:33:20] <rmatte> twm1010: I've found\nhttp://dev.zenoss.org/trac/ticket/5546 and\nhttp://dev.zenoss.org/trac/ticket/5547 so far
[16-Sep-2009 13:33:35] <iancmcc> Apocalipse: if you're on 2.4.x, there's only one google maps js now
[16-Sep-2009 13:33:36] <rmatte> twm1010: I'll probably have quite a few bugs logged before end of day
[16-Sep-2009 13:33:48] <rmatte> I plan to test the crap out of this
[16-Sep-2009 13:34:03] <iancmcc> ckrough: whatcha got? I suppose you could call me a pro, insofar as I get paid.
[16-Sep-2009 13:34:06] <rmatte> kind of difficult to do some testing without Mibs though
[16-Sep-2009 13:34:29] <rmatte> ckrough: I'm not a Zope "pro", but I know my fair share
[16-Sep-2009 13:40:40] <ckrough> thinking about how to get some dialog going about scaling zenoss in large installs.
[16-Sep-2009 13:40:52] <ckrough> Maybe Ill start a forum thread
[16-Sep-2009 13:42:47] <rmatte> http://dev.zenoss.org/trac/ticket/5548
[16-Sep-2009 13:50:15] <jb> i think i like the background colors on the old event console :/
[16-Sep-2009 13:50:26] <rmatte> you can toggle those on
[16-Sep-2009 13:50:28] <iancmcc> The specific colors? Or just having colors?
[16-Sep-2009 13:50:32] <iancmcc> Yeah, what rmatte said.
[16-Sep-2009 13:50:34] <jb> oh yeah?
[16-Sep-2009 13:50:35] <ckrough> jb: are you looking at unack'd events?
[16-Sep-2009 13:50:40] <jb> found it
[16-Sep-2009 13:50:50] <rmatte>
[16-Sep-2009 13:51:23] <rmatte> and I agree, I like it better with the background colors too
[16-Sep-2009 13:51:54] <chudler> is there are reason, or you just talking about taste?
[16-Sep-2009 13:51:55] <jb> and "Close" is like move to history?
[16-Sep-2009 13:52:05] <jb> chudler: just easier to differentiate..
[16-Sep-2009 13:52:10] <iancmcc> Close is exactly move to history.
\n[16-Sep-2009 13:53:18] <ckrough> Some of my netEngs had initially\nbalked at the background colors because they were so intense, but the\nproblem was we were looking at a page full of unacknowledged alerts
[16-Sep-2009 13:53:34] <ckrough> iancmcc: you get my privmsg?
[16-Sep-2009 13:53:40] <rmatte> chudler: it's more blatantly obvious what's higher severity
[16-Sep-2009 13:54:15] <rmatte> ckrough: yeh, when they are acknowledged they go pastel
[16-Sep-2009 13:54:17] <chudler> interesting, I thought the default was sorted by severity?
\n[16-Sep-2009 13:54:59] <rmatte> chudler: it is, but when you're\nwoken up at 2 in the morning to investigate something you don't want to\nhave to stare at little severity icons to determine what's what
[16-Sep-2009 13:55:27] <rmatte> it can be toggled anyways
[16-Sep-2009 13:55:55] <rmatte> but I suppose it's mostly personal preference
[16-Sep-2009 13:57:05] <rmatte> ah, I guess they haven't fixed the acknowledgement toggling yet
[16-Sep-2009 13:57:16] <rmatte> you can acknowledge but you can't unacknowledge
[16-Sep-2009 13:57:24] <iancmcc> Yeah, we didn't add that in.
[16-Sep-2009 13:57:38] <iancmcc> Big deal for some? I can push it
[16-Sep-2009 13:57:43] <rmatte> Is there a trac open for it?
[16-Sep-2009 13:57:56] <jb> hah, theres a poll about this very topic in the forums now
[16-Sep-2009 13:57:58] <rmatte> Well, it's definitely something that should make the cut
[16-Sep-2009 13:58:28] <rmatte> If you accidentally acknowledge an event then you should be able to unacknowledge it
[16-Sep-2009 13:58:35] <iancmcc> We're feature-frozen, so it's got to be a big deal. I'll see what I can do.
[16-Sep-2009 13:58:45] <iancmcc> Maybe I can accidentally check that in.
[16-Sep-2009 13:59:12] <rmatte> and there's no way to move an event from history back to status?
[16-Sep-2009 13:59:25] <iancmcc> Shoot. THAT I forgot.
[16-Sep-2009 13:59:34] <iancmcc> I'll open a ticket for that one.
[16-Sep-2009 13:59:39] <rmatte>
[16-Sep-2009 13:59:44] <rmatte> my god, I am on a roll today
[16-Sep-2009 13:59:49] <rmatte> I can open it if you want
[16-Sep-2009 13:59:53] <rmatte> up to you
[16-Sep-2009 13:59:53] <jb> actually
[16-Sep-2009 13:59:57] <jb> where is the "View History" now?
[16-Sep-2009 14:00:02] <rmatte> bottom left
[16-Sep-2009 14:00:03] <iancmcc> Bottom left
[16-Sep-2009 14:00:06] <jb> oh yeah
[16-Sep-2009 14:00:06] <jb> k
[16-Sep-2009 14:00:12] <iancmcc> Hidden, I know.
[16-Sep-2009 14:00:19] <rmatte> iancmcc: by the way, I absolutely love the new popup style
[16-Sep-2009 14:00:22] <jb> yeah maybe make that a buton?
[16-Sep-2009 14:00:24] <jb> button
[16-Sep-2009 14:00:26] <rmatte> iancmcc: that's just way too sexy
[16-Sep-2009 14:00:32] <rmatte> draggable semi-transparent
[16-Sep-2009 14:00:44] <jb> yeah its way better (the popup)
[16-Sep-2009 14:00:48] <rmatte> yeh
[16-Sep-2009 14:00:48]\n<iancmcc> It's a bit hacky because this screen is our first foray\ninto the new UI, which has a different navigation scheme
[16-Sep-2009 14:00:59] <iancmcc> so we've had to shove a few things in where they don't exactly belong
[16-Sep-2009 14:00:59] <jb> yeah i wondered how you were going to do this
[16-Sep-2009 14:01:24] <jb> since the new UI is different
[16-Sep-2009 14:01:28] <rmatte> yeh
[16-Sep-2009 14:01:31] <jb> at least from the shots ive seen..
[16-Sep-2009 14:01:34] <jb> which are from months ago
[16-Sep-2009 14:01:36] <rmatte> new UI won't make it until stone crab though
[16-Sep-2009 14:01:51] <iancmcc> History can't be a button, bad usability juju. We discussed putting back the History tab.
[16-Sep-2009 14:02:16] <iancmcc> But I figured, you only have to find the link once.
[16-Sep-2009 14:02:21] <jb> yea
[16-Sep-2009 14:02:24]\n<rmatte> iancmcc: the problem with the link is that people miss\nit, but I personally think it's a dumb argument, as soon as people know\nit's there once then they ALWAYS know it's there
[16-Sep-2009 14:02:33] <rmatte> there's no point in having a big tab sitting there for it
[16-Sep-2009 14:02:36] <rmatte> but that's my opinion
[16-Sep-2009 14:03:00] <iancmcc> Tab real estate is at a premium on some contexts, so we try not to add them
[16-Sep-2009 14:03:16] <ckrough> iancmcc: agreed
[16-Sep-2009 14:03:19] <ckrough> less is more
[16-Sep-2009 14:03:48] <rmatte> minimalistic clean cut look is the way to go
[16-Sep-2009 14:03:52] <jb> i liked having the "Details" tab
[16-Sep-2009 14:03:54] <jb> for event details
[16-Sep-2009 14:04:03] <jb> now all of it is string together using plain white cells
[16-Sep-2009 14:04:07] <jb> which makes it sort of hard to read
[16-Sep-2009 14:04:08] <rmatte> jb: you just double-click
[16-Sep-2009 14:04:17] <jb> yeah, but its displayed differently
[16-Sep-2009 14:04:23] <rmatte> right, I see what you're saying
[16-Sep-2009 14:04:37] <rmatte> I agree, it would be nice to have alternating cell colors
[16-Sep-2009 14:04:40] <rmatte> for the info that's displayed
[16-Sep-2009 14:04:58] <jb> yeah for sure
[16-Sep-2009 14:05:01] <iancmcc> I'll run it by our UI designer. Might be something we can do.
[16-Sep-2009 14:05:25] <rmatte> doesn't have to be dark colors ian, just a really light shade of gray on every second column
[16-Sep-2009 14:05:39] <rmatte> erm second row
[16-Sep-2009 14:05:39] <ckrough> and a pony
[16-Sep-2009 14:05:40] <rmatte> rather
[16-Sep-2009 14:05:44] <iancmcc> I'll check it out, see how it looks
[16-Sep-2009 14:05:51] <rmatte> MOAR PONIES!
[16-Sep-2009 14:06:13] <rmatte> we should so open a trac ticket for that ckrough
[16-Sep-2009 14:06:16] <jb> its much easier to read on in current versions..
[16-Sep-2009 14:06:18] <ckrough> heh
[16-Sep-2009 14:06:31] <jb> the alignment is goofy on 2.4.7 too
[16-Sep-2009 14:06:34] <jb> its all centered
[16-Sep-2009 14:06:35] <rmatte> Title: King Crab needs MOAR PONIES!
[16-Sep-2009 14:06:36] npmccallum_ is now known as npmccallum
[16-Sep-2009 14:06:41] <rmatte> Steps to reproduce:
[16-Sep-2009 14:06:44] <rmatte> 1. Log in.
[16-Sep-2009 14:06:50] <rmatte> Results
[16-Sep-2009 14:06:57] <rmatte> NO PONIES!
[16-Sep-2009 14:07:01] <rmatte> Expected results:
[16-Sep-2009 14:07:05] <rmatte> MOAR PONIES!
[16-Sep-2009 14:07:09] <rmatte> Impact:
[16-Sep-2009 14:07:11] <rmatte> BAHHHH!
[16-Sep-2009 14:07:29] <ckrough> lol
[16-Sep-2009 14:07:32] <iancmcc> http://dev.zenoss.org/trac/ticket/5549
[16-Sep-2009 14:09:10] <jb> http://thunder.jbdesign.net/~jbaird/zenoss/event-old.jpg
[16-Sep-2009 14:09:12] <jb> http://thunder.jbdesign.net/~jbaird/zenoss/event-new.jpg
[16-Sep-2009 14:09:50] <jb> just to compare..
\n[16-Sep-2009 14:10:18] <iancmcc> It went through design. That\nwhole pane is sort of free-floating, so it's ugly to shove a big table\nin there
[16-Sep-2009 14:10:24] <iancmcc> but delicate shading might work well
[16-Sep-2009 14:10:36] <rmatte> iancmcc: ummm, help me out here...
[16-Sep-2009 14:10:49] <rmatte> iancmcc: I click on event history, and it loads...
[16-Sep-2009 14:10:57] <rmatte> then there are 4 tabs at the top
\n[16-Sep-2009 14:11:18] <rmatte> Classes, which is presumably the\nold Events section, Mappings which is a summary of all event mapping,\nEvents, and zProperties
[16-Sep-2009 14:11:20] <iancmcc> depending on context...
[16-Sep-2009 14:11:29] <iancmcc> OH you mean from the global console?
[16-Sep-2009 14:11:35] <iancmcc> bug. Missing the notabs url parameter.
\n[16-Sep-2009 14:11:36] <rmatte> If I were to go in to zProperties\nand set action to "drop" it would presumably globally drop all events,\ncorrect?
[16-Sep-2009 14:12:01] <rmatte> ah
[16-Sep-2009 14:12:08] <rmatte> ok, I thought that was a bit odd
[16-Sep-2009 14:12:12] <rmatte> I'll log that as a bug then
[16-Sep-2009 14:12:23] <iancmcc> There's no difference between the Event Console and the events tab under /Events
[16-Sep-2009 14:12:28] <iancmcc> just the tabs are hidden
[16-Sep-2009 14:12:40] <rmatte> gotcha, so it's just not hiding them in history when it should be
[16-Sep-2009 14:12:43] <iancmcc> right.
[16-Sep-2009 14:12:46] <rmatte> okie
[16-Sep-2009 14:12:53] <rmatte> I'll open a ticket and pass you the number
[16-Sep-2009 14:15:40] <rmatte> iancmcc: http://dev.zenoss.org/trac/ticket/5550
[16-Sep-2009 14:16:15] <cgibbons> today, I hate windows.
[16-Sep-2009 14:16:24] <rmatte> tomorrow, who knows?
[16-Sep-2009 14:16:27] <iancmcc> right there with you, chadwick
[16-Sep-2009 14:16:32] <markeriv> today i hate offnet product t1 circuits and edgewater routers
[16-Sep-2009 14:16:34] <markeriv> join the club
[16-Sep-2009 14:16:34] <rmatte> afk
[16-Sep-2009 14:16:46] <markeriv> err excuse me, onnet
[16-Sep-2009 14:33:05] <markeriv> I have found a bug in version 2.4.2
[16-Sep-2009 14:33:16] <rmatte> markeriv: which bug?
[16-Sep-2009 14:33:25] <markeriv> sometimes when adding a device the comments field does not save
[16-Sep-2009 14:33:36] <markeriv> and must be reinput after the fact
[16-Sep-2009 14:33:47] <rmatte> might still be a bug, you could log it
[16-Sep-2009 14:33:51] <rmatte> do you have a trac account?
[16-Sep-2009 14:33:56] <markeriv> dont think so
[16-Sep-2009 14:34:04] <markeriv> i have a forums acct
[16-Sep-2009 14:34:12] <rmatte> ask mrayzenoss if you want a trac account
[16-Sep-2009 14:34:34] <rmatte> then you can go to http://dev.zenoss.org/trac
[16-Sep-2009 14:34:38] <rmatte> and log a ticket for the bug
[16-Sep-2009 14:38:39] <rmatte> does anyone know about this?:
[16-Sep-2009 14:38:40] <rmatte> Daemon: zenperfwmi starting...
[16-Sep-2009 14:38:41] <rmatte> CRITICAL:zen.pysamba:Unable to install the WMI reactor. reactor already installed
[16-Sep-2009 14:38:58] <rmatte> does King Crab include the WMI Data Source pack by default?
[16-Sep-2009 14:40:17] <markeriv> hmm
[16-Sep-2009 14:40:58] <cgibbons> i know about that
[16-Sep-2009 14:41:09] <rmatte> ?
[16-Sep-2009 14:41:11] <cgibbons> is zenperfwmi some custom zenpack daemon?
\n[16-Sep-2009 14:41:16] <markeriv> Stupid Question, Is it a bad\nthing when the available Zenoss Version is 2.4.5 and labled in red on\nthe versions page. I am currently running 2.4.2 and have not been able\nto find any documentation on specific upgrades.
[16-Sep-2009 14:41:20] <rmatte> it's from Egor's ZenPack
[16-Sep-2009 14:41:26] <mrayzenoss> cgibbons: yeah, it's the WMI Data Source
[16-Sep-2009 14:41:27] <cgibbons> okay, it needs a chnge
[16-Sep-2009 14:41:53] <rmatte> ok, so I'll have to let Egor know to update that pack before the King Crab release then
[16-Sep-2009 14:42:14] <cgibbons> basically this in his main .py file: http://pastebin.com/m4747e457
\n[16-Sep-2009 14:42:51] <rmatte> Well, there's lots of other stuff\nthat I'd like to test but I can't do much until\nhttp://dev.zenoss.org/trac/ticket/5546 is fixed
[16-Sep-2009 14:43:07] <rmatte> I need those Cisco Mibs and I'm not in the mood to build them by hand on a beta version
[16-Sep-2009 14:44:58] <twm1010> cgibbons: are you saying that King Crab comes with a wmi datasource daemon?
[16-Sep-2009 14:45:05] <cgibbons> no.
[16-Sep-2009 14:45:28] <jb> egor's..
\n[16-Sep-2009 14:45:42] <cgibbons> but any third-party daemons\nthat were using the pysamba library (either directly or indirectly)\nrequire that change to work.
[16-Sep-2009 14:45:49]\n<rmatte> nah, he's saying that King Crab does something\ndifferently when it comes to installing custom daemons from ZenPacks
[16-Sep-2009 14:46:19] <rmatte> markeriv: no, that's not a bad thing, and I'd recommend staying with 2.4.2 for a while
[16-Sep-2009 14:46:31] <rmatte> markeriv: 2.4.5 has some annoying bugs in it
\n[16-Sep-2009 14:47:13] <rmatte> If I'd have known I would have\njust patched 2.4.2 with that security patch and stayed at that version
[16-Sep-2009 14:48:08] <rmatte> markeriv: you'll want to do "zenpatch 14181" as the Zenoss user if you haven't already
[16-Sep-2009 14:48:15] <rmatte> markeriv: fixes: http://dev.zenoss.org/trac/ticket/5279
[16-Sep-2009 14:56:27] <mrayzenoss> poll.jspa?poll=1022
[16-Sep-2009 15:02:10] <rmatte> added my 2 cents
[16-Sep-2009 15:03:24] <Troubadix09> hi all
[16-Sep-2009 15:10:58] <rmatte> weird
[16-Sep-2009 15:11:00] <metalklesk> i really hate bmc patrol and bmc portal ¬¬
[16-Sep-2009 15:11:06] <rmatte> how do you get to that polls section
[16-Sep-2009 15:11:21] <rmatte> if I go to the forum, on the right hand side it says "Latest Poll - Thre are no polls"
[16-Sep-2009 15:11:31] <rmatte> I don't see any other reference to them
[16-Sep-2009 15:11:33] <cgibbons> ditto, metalklesk
[16-Sep-2009 15:12:04] <metalklesk> what is ditto ?
[16-Sep-2009 15:12:17] <rmatte> <metalklesk> i really hate bmc patrol and bmc portal ¬¬
[16-Sep-2009 15:12:27] <rmatte> means he agrees with you
[16-Sep-2009 15:12:44] <metalklesk> aaaaaahh ok, sorry i speak spanish
[16-Sep-2009 15:12:44] <metalklesk> jeje
\n[16-Sep-2009 15:18:53] <mrayzenoss> rmatte: I'm trying to figure\nout why the Poll works on the front page, but not the Forums page
[16-Sep-2009 15:20:35] <rmatte> ah
[16-Sep-2009 15:20:52] <mrayzenoss> we've turned on Abuse reporting as well
[16-Sep-2009 15:20:59] <rmatte> cool
[16-Sep-2009 15:23:32] <metalklesk> bmc portal really sucks with snmp receiving
[16-Sep-2009 15:38:39] <rmatte> apparently there's a 100 message limit for private messages?
[16-Sep-2009 15:38:43] <rmatte> that seems pretty low
[16-Sep-2009 15:38:55] <rmatte> is the site being run on a 40gb hard drive or something?
[16-Sep-2009 15:38:56] <rmatte>
[16-Sep-2009 15:39:26] <mrayzenoss> dunno, I can look into that
[16-Sep-2009 15:39:38] <rmatte> I'd say maybe set it to 300 or 500 even
[16-Sep-2009 15:39:52] <rmatte> even 1000 probably wouldn't choke out the storage
[16-Sep-2009 15:40:22] <rmatte> up to you guys though, it's your site
[16-Sep-2009 15:40:28] <rmatte> erm, meant
[16-Sep-2009 15:40:33] <mrayzenoss> it's hosted
[16-Sep-2009 15:40:39] <rmatte> ah
[16-Sep-2009 15:41:00] <rmatte> I'm guessing it's not hosted on a dedicated server?
[16-Sep-2009 15:41:22] <mrayzenoss> nah, we're a small install for them
[16-Sep-2009 15:41:28] <rmatte> ah
\n[16-Sep-2009 15:42:26] <rmatte> mhmmm, so I disable an interface\non a test device, and King Crab doesn't even notice that it's down
[16-Sep-2009 15:42:51] <mrayzenoss> are you actually close to 100?
[16-Sep-2009 15:42:57] <mrayzenoss> I've only got 33 myself
\n[16-Sep-2009 15:42:57] <rmatte> can you find out if it's still\nnecessary to use a threshold for monitor interface status via SNMP?
[16-Sep-2009 15:43:06] <rmatte> my impression was that that was going to be native in King Crab
[16-Sep-2009 15:43:24] <rmatte> no, I'm not close to 100, but I can see myself getting there eventually
[16-Sep-2009 15:43:34] <mrayzenoss> dunno, why don't you post that to Zenoss-Testing and I'll have Brandon address it
[16-Sep-2009 15:43:43] <rmatte> ik
[16-Sep-2009 15:43:54] <mrayzenoss> I've bumped it to 300
[16-Sep-2009 15:43:54] <rmatte> k*
[16-Sep-2009 15:44:01] <rmatte> kewel
\n[16-Sep-2009 15:44:45] <rmatte> hmmm, I'm going to try adding\ninterface threshold and see if it updates interface status on the OS\ntab when it sees it as down
[16-Sep-2009 15:44:51] <rmatte> as long as that works I shall be happy
[16-Sep-2009 15:45:52] <mrayzenoss> Zenoss Developers will be here Thursday 11am EDT
[16-Sep-2009 15:46:05] <mrayzenoss> oops
[16-Sep-2009 15:46:16] <mrayzenoss> iancmcc will be here to answer your questions
[16-Sep-2009 15:46:29] <rmatte> hehe
[16-Sep-2009 15:47:03] <rmatte> well, come now Matt, he's not a "real" developer
[16-Sep-2009 15:47:04] <rmatte>
[16-Sep-2009 15:47:23] <iancmcc> I just walked in and sat down one day. They never really hired me, even.
[16-Sep-2009 15:47:26] <mrayzenoss> dunno, how much more real does it get?
[16-Sep-2009 15:47:41] <_Apocalipse_> Dam windows 7
[16-Sep-2009 15:49:46] <rmatte> hehe
\n[16-Sep-2009 15:51:18] <rmatte> interesting... so yeh, it works\nfine as always with the threshold, but the status indicator doesn't\nupdate, I wonder if the transform needs to toggle the status indicator
[16-Sep-2009 15:51:37] <rmatte> I know I could script it, I'm just blown away that it's not working the way I envisioned
[16-Sep-2009 15:53:02] <rmatte> http://dev.zenoss.org/trac/ticket/4997
[16-Sep-2009 15:53:13] <rmatte> What "real time interface events" is he talking about?
\n[16-Sep-2009 15:53:35] <rmatte> Looks like I need to take the\ntransform that I wrote in: http://dev.zenoss.org/trac/ticket/5376 and\nmodify it
[16-Sep-2009 15:53:54] <rmatte> just need to figure out what that status icon value is
[16-Sep-2009 15:54:00] <rmatte> interface.status probably?
[16-Sep-2009 15:55:27] <rmatte> weird
[16-Sep-2009 15:55:38] <rmatte> now I get that UTC error when I do a zenoss status too
[16-Sep-2009 15:55:48] <rmatte> I wonder if that relates to the WMI Data Sources pack
[16-Sep-2009 15:55:51] <rmatte> going to remove it...
[16-Sep-2009 15:56:12] <rmatte> wow, I can't even do zenpack --list anymore
[16-Sep-2009 15:56:13] <rmatte> wtf
[16-Sep-2009 15:56:38] <iancmcc> Hmm, you've got something else wrong
[16-Sep-2009 15:56:55] <iancmcc> This is a fresh install?
[16-Sep-2009 15:56:58] <rmatte> I keep getting AttributeError: 'module' object has no attribute 'UTC'
[16-Sep-2009 15:56:59] <rmatte> a lot
[16-Sep-2009 15:57:20] <rmatte> well, I had 2.4.5 on the box, I did /usr/local/zenoss/uninstall
[16-Sep-2009 15:57:27] <rmatte> then I did a fresh install using the stack installer
[16-Sep-2009 15:57:36] <rmatte> then I installed a bunch of ZenPacks
[16-Sep-2009 15:57:41] <iancmcc> that should be fine, so at least it isn't stale code
[16-Sep-2009 15:58:06] <rmatte> most of the problems I've noticed relate to ZenPacks
[16-Sep-2009 15:58:39] <rmatte> I have a feeling it's that WMI pack or something
[16-Sep-2009 15:58:45] <rmatte> I'm going to try a new fresh install
[16-Sep-2009 15:58:52] <rmatte> this is insane
[16-Sep-2009 15:59:20] <jb> it is a beta..
[16-Sep-2009 15:59:33] <iancmcc> Yeah, but I'm guessing it's just an incompatibility in a particular pack
[16-Sep-2009 16:00:09] <iancmcc> All the Zenoss packs seem to work fine, certainly no problem with the zenpack machinery
[16-Sep-2009 16:03:56] <rmatte> I'm guessing it's the WMI Data Sources pack
[16-Sep-2009 16:04:07] <rmatte> but I'll do trial and error until I figure it out
[16-Sep-2009 16:06:53] <metalklesk> some days ago i asked about nonstop tandem monitoring using zenoss
[16-Sep-2009 16:07:11] <metalklesk> does anyone did that monitoring using zenoss ?
[16-Sep-2009 16:11:04] <twm1010> tandem monitoring of what
[16-Sep-2009 16:11:21] <twm1010> you're talking like... two collectors constantly polling a device for status?
[16-Sep-2009 16:11:32] <mrayzenoss> I think he means Tandem unix systems
[16-Sep-2009 16:12:03] <rmatte> well, I managed to trash the King Crab beta in a couple of hours
[16-Sep-2009 16:12:04] <twm1010> ah... i was thinking distributed monitoring of sorts... speaking of that, does Zenoss do that?
[16-Sep-2009 16:12:05] <rmatte>
\n[16-Sep-2009 16:12:30] <rmatte> twm1010: you can do distributed\ncollectors with core, it's just a pain in the arse to setup
\n[16-Sep-2009 16:12:32] <twm1010> Say... you're doing a URL check\non a live webserver, and have zenoss collectors in 3 places across the\ncountry
[16-Sep-2009 16:12:42] <rmatte> twm1010: enterprise does it as well, much easier to setup
[16-Sep-2009 16:13:00] <twm1010> gotcha
[16-Sep-2009 16:14:05] <rmatte> docs/DOC-2522
[16-Sep-2009 16:15:39] <twm1010> interesting...
\n[16-Sep-2009 16:15:57] <twm1010> but a collector is set for a\nhost... so i'd have to have duplicate hosts to do checks from separate\ncollectors right?
[16-Sep-2009 16:16:04] <twm1010> host = device
[16-Sep-2009 16:16:39] <metalklesk> i was meaning tandem non stop from hp
\n[16-Sep-2009 16:16:55] <rmatte> well, I don't think you could\nreally have the same host monitored from multiple collectors
[16-Sep-2009 16:17:04] <rmatte> maybe if you were monitoring it from different IPs somehow
[16-Sep-2009 16:17:28] <rmatte> If you had completely seperate instances of Zenoss, then you could do it
[16-Sep-2009 16:17:38] <rmatte> and have the instances feed the data back to some central location
[16-Sep-2009 16:17:38] <metalklesk> i think if i can install net-snmp in a tandem non stop OS, zenoss should work fine
[16-Sep-2009 16:17:57] <twm1010> sure...
[16-Sep-2009 16:18:32] <twm1010> Then again, 3 separate zenoss instances would produce the same result just the same
[16-Sep-2009 16:19:25] <twm1010> thats it for me... take care
[16-Sep-2009 16:21:41] <otakup0pe> anyone in here run zenoss with a non-local db
[16-Sep-2009 16:21:50] <otakup0pe> the first time i did that, years ago, it was easy to setup
[16-Sep-2009 16:22:05] <otakup0pe> the last time, a month ago, it somehow went horribly wrong and i had to nuke the zenoss vm ^_^
[16-Sep-2009 16:26:33] <rmatte> nah, we keep all of our dbs local, I assume you're talking about the events db right?
[16-Sep-2009 16:26:54] <iancmcc> Remote ZODB and events db aren't that big a deal
[16-Sep-2009 16:28:51] <metalklesk> i can't find if tandem OS runs net-snmp
[16-Sep-2009 16:30:00] <rmatte> yeh, it's looking like it was the WMI pack that caused the issues
[16-Sep-2009 16:30:10] <rmatte> I didn't install it this time around and things are working better
[16-Sep-2009 16:30:25] <iancmcc> mrayzenoss, we'll probably have to contact the author of that pack
[16-Sep-2009 16:31:19] <mrayzenoss> bigegor's been all over the forums today
[16-Sep-2009 16:31:21] <mrayzenoss> PM him
[16-Sep-2009 16:31:25] <otakup0pe> i'm referring to any mysql db which zenoss uses
[16-Sep-2009 16:32:24] <rmatte> yup, definitely WMI packs
[16-Sep-2009 16:32:33] <rmatte> I shall, and I'll close out the trac tickets that relate to the issue
[16-Sep-2009 16:37:11] <rmatte> mrayzenoss: what's his username?
[16-Sep-2009 16:37:22] <rmatte> nevermind
[16-Sep-2009 16:37:24] <rmatte> found him
[16-Sep-2009 16:37:35] <mrayzenoss> bigegor
[16-Sep-2009 16:38:17] <rmatte> yup
[16-Sep-2009 16:40:46] <rmatte> well that's good, means I can continue with more testing tomorrow
[16-Sep-2009 16:41:06] <rmatte> I'll try to cook up a transform for interface status while I'm at it so that I can test that out
[16-Sep-2009 16:41:50] <rmatte> "Private Message: WMI Data Sources ZenPack doesn't play well with King Crab"
[16-Sep-2009 16:41:52] <rmatte>
[16-Sep-2009 16:47:44] <skipzoid> hello
[16-Sep-2009 16:51:06] <rmatte> mrayzenoss: is there any way to configure a signature for messages and forum postings?
[16-Sep-2009 16:52:04] <mrayzenoss> hmm... I haven't seen that, but for obvious reasons I would like that
[16-Sep-2009 16:52:30] <rmatte> yeh, that was missing from the old forum too
[16-Sep-2009 16:52:38] <rmatte> kind of annoying
[16-Sep-2009 16:52:43] <Brad_K> i believe that the old forum system had that functionality.
[16-Sep-2009 16:52:59] <rmatte> it had it, but it wasn't accessible to regular users
[16-Sep-2009 16:53:01] <mrayzenoss> nah, it was broken
[16-Sep-2009 16:53:07] <rmatte> only Zenoss employees had signatures
[16-Sep-2009 16:53:08] <rmatte> lol
[16-Sep-2009 16:53:09] <mrayzenoss> they had it, it just lied
\n[16-Sep-2009 16:53:10] <Brad_K> i seem to recall some fields\nrelating to signatures, etc... when i first signed up for a user\naccount.
[16-Sep-2009 16:53:23] <mrayzenoss> yeah, the ability to apply them was broken
[16-Sep-2009 16:53:45] <Brad_K> ahh, okay. i knew i saw something, but i didn't pay much attention to it at the time.
[16-Sep-2009 16:54:20] <Brad_K> the current system does have profiles, avatars, pictures, etc... right?
[16-Sep-2009 16:54:32] <rmatte> mrayzenoss: http://www.jivesoftware.com/jivespace/thread/14133
[16-Sep-2009 16:55:16] <mrayzenoss> hmm... that's from over 2 years ago...
[16-Sep-2009 16:55:20] <mrayzenoss> I'll raise the issue
[16-Sep-2009 16:55:28] <rmatte> k
\n[16-Sep-2009 16:58:40] <skipzoid> i'm already runing mysql on the\nlinux server - having installed zenoss core it has installed another\ncopy of mysql within the zenoss tree and it starts - oddly they are\nboth on the same port of 3307 and the existing mysql install doesnt\nhave the databases in it, if I run zenbackup to backup the databases\nthen stop zenoss and disable the zenoss version of the mysql and then\nrestart and do a zenrestore will it populate the datab
[16-Sep-2009 17:00:11]\n<rmatte> skipzoid: it may, but it may also have a problem with\nthe tables not already being there for it to populate the data back in\nto
[16-Sep-2009 17:00:41] <rmatte> skipzoid: http://dev.zenoss.org/trac/browser/branches/zenoss-2.4.x/Products/ZenEvents/db/zenevents.sql
[16-Sep-2009 17:00:44] <rmatte> use that
[16-Sep-2009 17:00:44] <skipzoid> am i right in thinking that I do not or shouldnt run two instances of mysql ?
[16-Sep-2009 17:00:57] <rmatte> probably best to run a single instance on one system
[16-Sep-2009 17:01:18] <rmatte> anyways, that sql file will create the database for you
[16-Sep-2009 17:01:35] <rmatte> databases, rather
[16-Sep-2009 17:01:49]\n<skipzoid> since installing zenoss the system load has gone right\nup and now I'm consistantly using arounf 200m of the swap more than\npreviosly
[16-Sep-2009 17:01:53] <skipzoid> whats the link ?
[16-Sep-2009 17:02:14] <rmatte> it's a link to the sql queries used to actually create the zenoss databases
[16-Sep-2009 17:02:29] <rmatte> so if you run that on your existing SQL server it'll create everything zenoss needs
[16-Sep-2009 17:02:51] <rmatte> I'm assuming you don't have any data that you need to import over from the other db?
[16-Sep-2009 17:03:12] <skipzoid> other?
[16-Sep-2009 17:03:21] <rmatte> the one that Zenoss installed
[16-Sep-2009 17:03:57] <skipzoid> well its got the devices and a few monitors etc, nothing that wouldnt take an hour to reset if need be
[16-Sep-2009 17:04:06] <skipzoid> i thought zenrestore would put it all back?
[16-Sep-2009 17:05:04] <skipzoid> is there a version of that file that is not encapsulated in html ?
[16-Sep-2009 17:11:37] <rmatte> zenrestore will put it back
[16-Sep-2009 17:11:47] <rmatte> but it may need the tables to already be there
[16-Sep-2009 17:11:53] <rmatte> since they are generally created during install
[16-Sep-2009 17:12:04] <rmatte> so I'd advise to run those SQL commands on the SQL server
[16-Sep-2009 17:12:07] <rmatte> then run the restore
[16-Sep-2009 17:12:55] <skipzoid> zenoss@atlantis:~$ zenrestore
[16-Sep-2009 17:12:55] <skipzoid> You must specify either --file or --dir.
[16-Sep-2009 17:12:55] <skipzoid> zenoss@atlantis:~$
\n[16-Sep-2009 17:13:27] <rmatte> mrayzenoss: I think you just\nbumped it to 300 for yourself, I'm still at 100 for message limit
[16-Sep-2009 17:13:45] <mrayzenoss> I'm actually sitting at 34
[16-Sep-2009 17:14:24] <rmatte> for the limit I mean though, some people like to archive important messages
[16-Sep-2009 17:14:31] <mrayzenoss> oh yeah
[16-Sep-2009 17:14:49] <rmatte> hehe
[16-Sep-2009 17:15:41] <rmatte> well, I think I'm going to get out of here shortly
[16-Sep-2009 17:15:50] <rmatte> there is a beer somewhere out there with my name on it
\n[16-Sep-2009 17:16:26] <rmatte> cisco Mibs will be installed by\nthe time I go in tomorrow and then I can do some real testing
[17-Sep-2009 00:00:46] [disconnected at Thu Sep 17 00:00:46 2009]
[17-Sep-2009 00:00:46] [connected at Thu Sep 17 00:00:46 2009]
[17-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
\n[17-Sep-2009 03:29:26] <qdelbul> Hello, I'm trying to make a\nzenpack. I read the developper's documentation but I don't understand\nhow i can add a selfmade plugin that generate an event inside my\nzenpack. Have you a good resource for that ?
[17-Sep-2009 04:36:09] <qdelbul> well, is somebody can help me on this :
\n[17-Sep-2009 04:36:30] <qdelbul> I'm trying to make a zenpack. I\nread the developper's documentation but I don't understand how i can\nadd a selfmade plugin that generate an event inside my zenpack. Have\nyou a good resource for that ?
[17-Sep-2009 06:11:51]\n<chemist> zenoss has stopped sending emails overnight, I can\nstill send a test email but no event notifications are coming through
[17-Sep-2009 07:05:16]\n<chemist> when is the migration of the site cgoing to be\ncomplete? none of the links in the faq work, they all end up going to\nthe same place (community/documentation)
[17-Sep-2009 09:16:39] <ckrough> <crickets>
[17-Sep-2009 09:23:10] <metalklesk> hi
[17-Sep-2009 09:30:32] <metalklesk> one question
[17-Sep-2009 09:31:08] <metalklesk> do i need to install MIB to get the jobs remaining in a printer ?
\n[17-Sep-2009 09:34:50] <chemist> zenoss has stopped sending\nemails overnight, I can still send a test email but no event\nnotifications are coming through
[17-Sep-2009 09:54:58] <Troubadix09> metalklesk: MIBs in zenoss are only use to transform snmp-traps in a readable form
\n[17-Sep-2009 09:57:27] <Troubadix09> metalklesk: try to find out\nthe right OID for your printer, maybe a general OID from the\nPRINTER.MIB or from the MIB of the Manufacturer and poll it
[17-Sep-2009 09:59:37] <Troubadix09> mray: I have a question about templates
[17-Sep-2009 10:00:54] <Troubadix09> mray: is it possible to assign a new template wich is installed with a new ZenPack
[17-Sep-2009 10:01:19] <Troubadix09> mray: and bind to a device-class
[17-Sep-2009 10:01:51] <Troubadix09> mray: that existing devices bind them automatically, such as new devices do?
[17-Sep-2009 10:01:55] <twm1010> chemist: is zenactions daemon running?
[17-Sep-2009 10:02:33]\n<twm1010> troubadix: if you bind a template to a device class,\nall devices in the class receive that binding, unless locally overridden
[17-Sep-2009 10:05:09] <Troubadix09> mray: hm... I think so, but why not in my Zenoss Install? (Virtual Appliance with Zenoss 2.4.5) :-(
[17-Sep-2009 10:05:44] <mrayzenoss> Troubadix09: I don't believe ZenPacks automatically bind to new templates
[17-Sep-2009 10:06:12] <mrayzenoss> err... device classes automatically bind to templates delivered via ZenPack
[17-Sep-2009 10:07:32]\n<rmatte> http://pastebin.com/m7445b476 <--- The parser\ncommands that I'm specifying in this script are completely getting\noverriden by one of the Zenoss imports that I'm doing, can anyone think\nof a way around this?
[17-Sep-2009 10:07:47] <iancmcc> Troubadix09: you can modify the ZenPack to change the bindings of your desired ZenPack
[17-Sep-2009 10:07:48] <rmatte> They work perfectly fine if I put them in a script by themselves
[17-Sep-2009 10:07:58] * aclark waves
[17-Sep-2009 10:08:05] <iancmcc> Troubadix09: I mean the bindings of your desired DeviceClass.
[17-Sep-2009 10:08:07] <mrayzenoss> aclark: howdy
[17-Sep-2009 10:08:46] <iancmcc> rmatte: The option parsing?
[17-Sep-2009 10:09:00]\n<Troubadix09> mray: I know, I bind a template come with the\nWMIPerf_Windows-ZenPack manually to the class /device/Server/Windows
[17-Sep-2009 10:09:06] <rmatte> iancmcc: yeh, it outputs a bunch of other zenoss specific options instead of the ones I want
[17-Sep-2009 10:10:30] <iancmcc> rmatte: Checking, I believe there's a way to stifle ZenScriptBase's option parsing
[17-Sep-2009 10:10:30] <iancmcc> rmatte: pass noopts=True into the ZenScriptBase constructor
[17-Sep-2009 10:10:32] <Troubadix09> mray: when i add new devices to this class, they have bind the template and use it
[17-Sep-2009 10:10:45] <rmatte> iancmcc: thanks, I'll try that
[17-Sep-2009 10:10:56] <Troubadix09> mray: but the existing devices in that class wouldn't
[17-Sep-2009 10:11:22] <rmatte> works like a charm, thanks so much
\n[17-Sep-2009 10:12:29] <mrayzenoss> Troubadix09: hmmm... so\nexisting devices dont' get updated, but new devices pick up the binding\nwhen they're added to the device class?
[17-Sep-2009 10:12:44] <mrayzenoss> have you remodeled the old devices?
[17-Sep-2009 10:12:58] <iancmcc> Troubadix09: Do the existing devices have their own values set?
[17-Sep-2009 10:13:25] <Troubadix09> mray: yes, i remodel
\n[17-Sep-2009 10:15:44] <rmatte> hey iancmcc, one more question,\nif you look at that code again... is there anything special that I need\nto do to set variables based on those options?
[17-Sep-2009 10:15:57] <rmatte> I was under the impression that dest= represented the variable name?
[17-Sep-2009 10:16:05]\n<Troubadix09> iancmcc: there are three templates binded to the\ndevices, two of them are "Local Copy" but not the new template
[17-Sep-2009 10:16:35] <iancmcc> rmatte: That's correct, e.g. options.startDate should have your value
[17-Sep-2009 10:16:47] <rmatte> oh, it's options.startDate
[17-Sep-2009 10:16:50] <rmatte> gotcha, thanks
\n[17-Sep-2009 10:17:38] <iancmcc> Troubadix09: Sounds like they\nhave their own values, which means they won't inherit new template\nbindings from the device class
[17-Sep-2009 10:17:50] <iancmcc> Troubadix09: You'll have to either reset the bindings or add the binding manually
[17-Sep-2009 10:20:56] <Troubadix09> iancmcc: mray: thank you, i will try it manually
[17-Sep-2009 10:22:22] <Troubadix09> iancmcc: I have another question...
[17-Sep-2009 10:25:05]\n<Troubadix09> iancmcc: is it possible to copy Data Sources in a\nTemplate? Because sometimes there are snmp-DataSources where only the\nlast OID-Value is different
[17-Sep-2009 10:25:42] <Troubadix09> iancmcc: and it would be nice to only change the deltas
\n[17-Sep-2009 10:26:14] <metalklesk> does anyone see the memory\n(ram, swap, cache) in zenoss using suse, fedora or ubuntu zenpack ?
[17-Sep-2009 10:26:25] <metalklesk> i mean total, used and free memory
[17-Sep-2009 10:27:10] <Troubadix09> iancmcc: the Menue of the Data-Sources only show "Add, Delete and Add to Graphs"
[17-Sep-2009 10:27:19] <corey> hi
\n[17-Sep-2009 10:28:19] <mrayzenoss> Troubadix09: I guess you'd\nneed to use zendmd: docs/DOC-2532
\n[17-Sep-2009 10:28:31] <ckrough> anyone have a link for\ndownloading OpenManage Mibs? trying to search their site results in\nprinters for sale.
[17-Sep-2009 10:32:37] corey is now known as kmtsun
[17-Sep-2009 10:32:40] <cgibbons> hmmm
[17-Sep-2009 10:32:49] <cgibbons> ian sure got a lot more questions than I did a couple of weeks ago
[17-Sep-2009 10:33:13] <metalklesk> cgibbons, spit it out !
[17-Sep-2009 10:33:14] <Troubadix09> mray: hmm... is this planned to implement in WebGui?
[17-Sep-2009 10:33:46] <rmatte> hehe
[17-Sep-2009 10:34:31] <mrayzenoss> Troubadix09: I'd ask iancmcc
[17-Sep-2009 10:34:44] <iancmcc> Troubadix09: No.
[17-Sep-2009 10:34:46] <ckrough> ouch
\n[17-Sep-2009 10:35:24] <mrayzenoss> Troubadix09: go ahead and\nopen a ticket as an enhancement, I know the templates are supposed to\nget some attention in the Stone Crab release
[17-Sep-2009 10:35:31] <mrayzenoss> and we can discuss it then
[17-Sep-2009 10:35:50] <aclark> Are links to 2.5 beta avail without registering on zenoss.com ?
[17-Sep-2009 10:35:55] <mrayzenoss> yeah
[17-Sep-2009 10:36:06] <mrayzenoss> http://www.zenoss.com/beta/2.5/
[17-Sep-2009 10:36:16] <mrayzenoss> "http://www.zenoss.com/beta/2.5/download"
[17-Sep-2009 10:37:01] <aclark> mrayzenoss: thanks
[17-Sep-2009 10:37:14] <mrayzenoss> aclark: thanks for checking out the beta
[17-Sep-2009 10:37:40] <aclark> mrayzenoss: no prob! excited about it (and zenoss in general as you can tell ;-)
\n[17-Sep-2009 10:39:16] <kmtsun> @aclark4life I'm have trouble\nwith my smtp monitoring, the service is up but I'm getting msgs: IP\nservice smtp is down
[17-Sep-2009 10:40:10] <aclark> kmtsun: i'm just a n00b
[17-Sep-2009 10:41:02] <metalklesk> kmtsun, did you open the port of snmp service in your firewall ?
[17-Sep-2009 10:41:04] <aclark> kmtsun: but there are experts here :-)
[17-Sep-2009 10:41:26] <metalklesk> service up means that you can do a ping to teh device
[17-Sep-2009 10:41:28] <kmtsun> ok thakns
[17-Sep-2009 10:41:37] <kmtsun> no firewall
[17-Sep-2009 10:42:01] <kmtsun> i have it monitoring 30 bsd jails and im only having trouble with 2 of them
[17-Sep-2009 10:42:08] <metalklesk> what port are you using to snmp ? 161 ? 162 ?
[17-Sep-2009 10:42:10] <kmtsun> the rest are showing smtp up
[17-Sep-2009 10:42:15] <metalklesk> mmm
[17-Sep-2009 10:42:23] <metalklesk> sorry i missupderstood
[17-Sep-2009 10:42:32] <metalklesk> *missunderstood
[17-Sep-2009 10:42:45] <metalklesk> snmp != smtp
[17-Sep-2009 10:42:48] <metalklesk>
[17-Sep-2009 10:45:13] <iancmcc> kmtsun: Try telnetting from the zenoss server to narrow it down to zenoss or network
[17-Sep-2009 10:45:23] <kmtsun> ok
[17-Sep-2009 10:46:34] <kmtsun> Trying x.x.x.x...
[17-Sep-2009 10:46:34] <kmtsun> Connected to xxx.xxx.com.
[17-Sep-2009 10:46:34] <kmtsun> Escape character is '^]'.
[17-Sep-2009 10:46:34] <kmtsun> Connection closed by foreign host.
[17-Sep-2009 10:46:55] <iancmcc> you connected to SMTP port, I assume
[17-Sep-2009 10:47:10] <kmtsun> yes telnet <host> 25
[17-Sep-2009 10:47:46] <kmtsun> i did a netstat -al and it is listening on *.* 25
[17-Sep-2009 10:47:58] <iancmcc> I don't know the answer. Try remodeling the box.
[17-Sep-2009 10:48:30] <kmtsun> that means to delete the box from the zenoss db and let it rediscover it"?
[17-Sep-2009 10:48:40] <metalklesk> nop
[17-Sep-2009 10:48:43] <iancmcc> don't need to delete it, just remodel
[17-Sep-2009 10:49:05] <metalklesk> that means -> manage -> model device
[17-Sep-2009 10:49:07] <kmtsun> ok
[17-Sep-2009 10:49:52] <metalklesk> how can i see the used, free, total ram memory ? and swap too
[17-Sep-2009 10:49:59] <kmtsun> thank you .. its a going
[17-Sep-2009 10:50:14] <metalklesk> using snmp no problem, but via ssh i can't see it
[17-Sep-2009 10:53:28] <mrayzenoss> metalklesk: which SSH ZenPacks?
[17-Sep-2009 10:53:42] <mrayzenoss> do you have installed? LinuxMonitor by itself is pretty vanilla
[17-Sep-2009 10:53:58] <metalklesk> mandriva (the one i send you), suse, ubuntu
[17-Sep-2009 10:54:15] <metalklesk> yes, i have linuxmonitor and linuxmonitoraddon installed
[17-Sep-2009 10:54:59] <kmtsun> ack events... remodeled.... smtp red still exist... i will try restarting the jail
\n[17-Sep-2009 10:55:43] <metalklesk> in zproperties i deleted ram\nfrom zFileSystemMapIgnoreTypes as i did with other device monitored by\nsnmp
[17-Sep-2009 10:56:21] <Troubadix09>\nmrayzenoss: thanks, i will open a ticket and we will seen what happend\nin Stone Crab :-)
[17-Sep-2009 10:56:28] <mrayzenoss> metalklesk: are you talking about memory graphs on the Perf tab?
\n[17-Sep-2009 10:56:47] <metalklesk> in the snmp monitored device,\nin filesystem i can see swap and ram (total, used and free) but with\nthe zenpack i can't
[17-Sep-2009 10:57:02] <metalklesk> memory graph AND in filesystem
[17-Sep-2009 10:57:48] <metalklesk> in collector plugins i have zenoss.cmd.linux.memory selected
[17-Sep-2009 10:58:33] <metalklesk> in hardware i can see the total memory and total swap
[17-Sep-2009 10:59:01] <metalklesk> but i can't see the important information of ram and swap in filesystem (total, used and free)
\n[17-Sep-2009 10:59:51] <kmtsun> i resarted the jail.. and\nremodeled... im showing this now: ip service smtp timeout waiting for\n'220'
[17-Sep-2009 11:02:06] <mrayzenoss> metalklesk: my 2.4.5 box monitoring Linux via SNMP has memory and CPU on the Perf tab
[17-Sep-2009 11:02:23] <mrayzenoss> Load Average, CPU Utilization, Memory Utilization, IO
[17-Sep-2009 11:02:28] <metalklesk> yes, i know
[17-Sep-2009 11:02:39] <mrayzenoss> sorry, I thought you were saying those were on the OS tab
[17-Sep-2009 11:02:48] <iancmcc> kmtsun: message/14824#14824
[17-Sep-2009 11:02:52] <metalklesk>
[17-Sep-2009 11:03:12] <mrayzenoss> ok, so none of those graphs are on the Perf tab using /Server/SSH/Linux/ ?
[17-Sep-2009 11:03:22] <mrayzenoss> metalklesk: are the templates all bound?
[17-Sep-2009 11:03:26] <metalklesk> just cpu
[17-Sep-2009 11:03:45] <metalklesk> how can i see that ?
[17-Sep-2009 11:03:57] <iancmcc> kmtsun: This is more informative, actually message/12865#12865
[17-Sep-2009 11:04:17] <motdman> Hello, help with filesystems freeblock Usedblocks AIX
[17-Sep-2009 11:04:37] <motdman> missing rrd files
[17-Sep-2009 11:04:50] <metalklesk> graphs are one thing, but filesystem is another
[17-Sep-2009 11:04:53] <metalklesk> i mean
[17-Sep-2009 11:05:05] <metalklesk> 2 problems i see with ssh based zenpack
[17-Sep-2009 11:05:06] <mrayzenoss> metalklesk: http:/yourserver:8080/zport/dmd/Devices/Server/SSH/perfConfig
[17-Sep-2009 11:05:59] <metalklesk> Name Definition Path Description
[17-Sep-2009 11:05:59] <metalklesk> Device /Devices/Server/SSH
[17-Sep-2009 11:05:59]\n<metalklesk> FileSystem /Devices/Server Filesystem\ntemplate that uses HOST-RESOURCES mib. Has a 90% threshold.
[17-Sep-2009 11:05:59] <metalklesk> IpService /Devices Place holder for future use.
[17-Sep-2009 11:05:59] <metalklesk> OSProcess /Devices Monitors for OSProcess object
[17-Sep-2009 11:06:00] <metalklesk> WinService /Devices Place holder for future use.
[17-Sep-2009 11:06:02] <metalklesk> ethernetCsmacd /Devices Standard ethernet interface template with 75% utilization threshold
[17-Sep-2009 11:06:04] <metalklesk> ethernetCsmacd_64
[17-Sep-2009 11:06:58] <metalklesk> bad idea to copy thos lines, sorry
[17-Sep-2009 11:08:28]\n<metalklesk> this are the templates in the page you said:\ndevice,filesystem,ipservice,osprocess,winservice,ethernetcsmacd,ethernetcsmacd_64
[17-Sep-2009 11:08:44] <mrayzenoss> yeah, I'm installing that ZenPack now to look around
[17-Sep-2009 11:09:05] <metalklesk> thanks !!!
[17-Sep-2009 11:10:24] <motdman> How do I get the filesystem usedblocks freeblocks in AIX
[17-Sep-2009 11:11:28] <motdman> If I use the linux model device on AIX, it is displayed.
[17-Sep-2009 11:11:33] <mrayzenoss> motdman: are you using docs/DOC-3454 ?
[17-Sep-2009 11:12:28] <mrayzenoss> motdman: I've heard very mixed reviews about the quality of the SNMP daemons on AIX
[17-Sep-2009 11:12:43] <mrayzenoss> jb and rocket can attest
\n[17-Sep-2009 11:13:25] <motdman> I worked with IBM to get the\nxmservd mibs to work and all looks good except for filesystem
[17-Sep-2009 11:13:40] <motdman> Actually it is quite impressive regarding the info in place already
[17-Sep-2009 11:14:16] <motdman> You don't need net-snmp
[17-Sep-2009 11:15:22] <motdman> Another thing is that it does't give you the CPU size, only the speed
\n[17-Sep-2009 11:16:40] <motdman> I would use the linux model for\nfilesystem, but I would lose the volumegroup level and hdisk info.
[17-Sep-2009 11:16:50] <Troubadix09> bye
[17-Sep-2009 11:17:31] <iancmcc> see ya troubadix
[17-Sep-2009 11:20:11] <metalklesk> mrayzenoss, did you see the problem i told you with the zenpack ?
\n[17-Sep-2009 11:20:46] <motdman> I used the DOC-3454 but that\npointed me to the cacti setup which is somewhat outdated now.
[17-Sep-2009 11:22:13] <motdman> Why can the linux model pull the aix data I need and not the aix model for filesystem?
[17-Sep-2009 11:23:39] <mrayzenoss> metalklesk: yeah, looks like the defautl LinuxMonitor is just CPU Utilization
[17-Sep-2009 11:24:11] <mrayzenoss> motdman: why not just copy the Linux Device Class and call it AIX?
[17-Sep-2009 11:24:54] <motdman> I could but I would lose the VG LV and hdisk info which is pretty cool stuff.
[17-Sep-2009 11:25:29] <motdman> the Aix model is very cool but just missing that one piece
[17-Sep-2009 11:25:56] <motdman> not sure why it wasn't included in the build, unless I am missing something
[17-Sep-2009 11:26:32] <mrayzenoss> well, it's a community ZenPack, so the author probably didn't run into your use-case
[17-Sep-2009 11:26:51] <motdman> What I do know is that is it not generating the rrd data for this particular issue
[17-Sep-2009 11:27:05] <motdman> The linux model does pull the appropriate rrd...go figure
[17-Sep-2009 11:28:09] <motdman> I would think getting alerts on filesystem free/used would be at the top of the list.
\n[17-Sep-2009 11:28:54] <mrayzenoss> motdman: feel free to dig\ninto the ZenPack, comment on the thread on the page for it or submit a\npatch to the author
[17-Sep-2009 11:30:01] <metalklesk> mrayzenoss, so what can WE do to fix the problem ?
[17-Sep-2009 11:30:04] <metalklesk>
\n[17-Sep-2009 11:30:20] <motdman> Do you know if the author built\nin this functionality or am I not configuring this correctly.
[17-Sep-2009 11:30:39] <mrayzenoss> motdman: no idea, no AIX on my side
[17-Sep-2009 11:31:01] <mrayzenoss> metalklesk: you can submit a patch to the LinuxMonitor or the LinuxMonitorAddon
\n[17-Sep-2009 11:31:11] <motdman> There is not much info on this\nout there...too bad cause it is a great product regarding AIX
[17-Sep-2009 11:31:24] <metalklesk> ok
[17-Sep-2009 11:31:34] <metalklesk> i will investigate how can i fix that
[17-Sep-2009 11:31:36] <metalklesk>
[17-Sep-2009 11:31:58] <mrayzenoss> ok, I'm gonna call that an end to the session, iancmcc is off the hook
[17-Sep-2009 11:33:15] <motdman> Thanks everyone for your help.
[17-Sep-2009 11:39:48] <twm1010> Do we have to re-register for zenoss community forums?
[17-Sep-2009 11:49:33] <cgibbons> anyone here an autoconf guru?
[17-Sep-2009 11:52:19] <rmatte> twm1010: you need to reset your password, but not re-register
[17-Sep-2009 11:52:59] <rmatte> I know what autoconf is, but I doubt that constitutes being a guru
[17-Sep-2009 11:56:24] <metalklesk> rmatte, do you have printers monitored ?
[17-Sep-2009 12:00:49] <rmatte> metalklesk: no, we don't do printers
[17-Sep-2009 12:00:51] <rmatte> we're a noc
[17-Sep-2009 12:02:12] <metalklesk> buuuuu
[17-Sep-2009 12:02:20] <rmatte> lol
[17-Sep-2009 12:02:47] <rmatte> I can't imagine it would be that different from monitoring anything else
[17-Sep-2009 12:02:54] <metalklesk> i've added a printer but i can only see interfaces, no jobs
[17-Sep-2009 12:03:19] <metalklesk> i want to see the queue
[17-Sep-2009 12:03:25] <metalklesk> if it can
[17-Sep-2009 12:03:44] <rmatte> well, are you using any kind of special pack to monitor it or something?
[17-Sep-2009 12:04:06]\n<rmatte> Zenoss isn't designed to see printer queues by default,\nyou'd have to write some sort of collector plugin and add another tab\nin there for printer jobs or whatever
[17-Sep-2009 12:04:11] <rmatte> wouldn't be the easiest thing in the world
[17-Sep-2009 12:05:29] <metalklesk> /Printer/Inkjet
[17-Sep-2009 12:05:38] <metalklesk> that class
[17-Sep-2009 12:05:47] <metalklesk> i have to leave now
[17-Sep-2009 12:05:54] <metalklesk> we talk later, bye !!
[17-Sep-2009 12:10:28] <twm1010> You'd do better just using HP webJetadmin for that stuff
[17-Sep-2009 12:17:42] <kmtsun> anyone have any idea what: HTTP CRITICAL: HTTP/1.1 502 Proxy Error is?
[17-Sep-2009 12:17:56] <kmtsun> or what it the problem may be?
[17-Sep-2009 12:18:41] <rmatte> kmtsun: where are you seeing that?
[17-Sep-2009 12:19:35] <kmtsun> that is an event on one of my jails
[17-Sep-2009 12:19:59] <kmtsun> eventClass: /Status/HTTP
[17-Sep-2009 12:20:40] <rmatte> have you tried googling at all?
\n[17-Sep-2009 12:21:38] <rmatte> Probably a cause similar to:\nhttp://mail-archives.apache.org/mod_mbox/hc-httpclient-users/200506.mbox/%3COF8C8ACECC.29497D1F-ONC125701F.00200C1F-C125701F.0020D09E@de.ibm.com%3Embox/%3COF8C8ACECC.29497D1F-ONC125701F.00200C1F-C125701F.0020D09E@de.ibm.com%3E
[17-Sep-2009 12:29:55] <davetoo> hmm
[17-Sep-2009 12:30:06] <davetoo> maybe my problem is why rc2 is out
[17-Sep-2009 12:30:45] <davetoo> I can't view/edit zProperties
[17-Sep-2009 12:31:04] <iancmcc> what error do you get?
[17-Sep-2009 12:31:04] <rmatte> seriously?
[17-Sep-2009 12:31:23] <rmatte> I was able to view and edit them fine on rc2 just yesterday
[17-Sep-2009 12:31:23] <mrayzenoss> davetoo: are you logged in as a user with those privileges?
[17-Sep-2009 12:31:42] <davetoo> admin, actaully, because I goofed up the ldap
[17-Sep-2009 12:31:44] <davetoo> but...
[17-Sep-2009 12:31:58] <davetoo> I may not have done a sufficiently clean intslall
[17-Sep-2009 12:32:10] <davetoo> I keep forgetting that when you erase the rpms,
[17-Sep-2009 12:32:31] <davetoo> opt/zenoss doesn't actually go away
[17-Sep-2009 12:32:50] <davetoo> so I'm not sure the install ran the migration code,
[17-Sep-2009 12:33:14] <davetoo> I may just clobber it all to oblivion again and start over
[17-Sep-2009 13:12:44] <rmatte> If you're testing rc2, do not install the WMI Data Sources pack
[17-Sep-2009 13:12:55] <rmatte> you'll end up having to reinstall the whole works if you do
[17-Sep-2009 13:15:44] <rmatte> hmmm, I love the new tab completion for objects in zendmd
[17-Sep-2009 13:15:47] <rmatte> quite slick
[17-Sep-2009 13:26:10] * ckrough wants syslog output from zen processes.
[17-Sep-2009 13:49:42] <kmtsun_> exit
[17-Sep-2009 13:49:44] <kmtsun_> end
[17-Sep-2009 13:50:38] <ckrough> noone leaves #zenoss. no one
[17-Sep-2009 13:50:59] <davetoo> I'm fighting ldap
[17-Sep-2009 13:52:01] * Brad_K Everybody was LDAP fighting
[17-Sep-2009 13:52:17] * Brad_K ... Those cats were slow and frightening
\n[17-Sep-2009 13:52:49] <davetoo> I'm probably too tired to be\nlooking at it, basically never slept, but I can't understand why this\none server isn't binding, but the configs and packets look the same
[17-Sep-2009 13:52:50] * ckrough awards point
[17-Sep-2009 13:53:05] <ckrough> davetoo: ldapsearch binds from CLI?
[17-Sep-2009 13:53:45] <davetoo> Haven't tried that yet... not enough brainpower until after I sleep
\n[17-Sep-2009 13:55:58] <ckrough> iancmcc: know of anyone running\ndistributed ZEO with zenoss?\n(http://www.zope.org/Products/ZEO/ZEOFactSheet)
[17-Sep-2009 13:56:31] <iancmcc> Not off the top of my head
[17-Sep-2009 13:56:59] <ckrough> that may be the next step in scaling when ZODB becomes the bottleneck
\n[17-Sep-2009 13:57:29] <ckrough> at least getting Zope Storage on\nits own box, even it its not distributed across multiple. Get it away\nfrom other daemons and mysql
[17-Sep-2009 13:57:43] <ckrough> then cache on SSD drives
[17-Sep-2009 13:58:22] * davetoo waits for enterprise zenpacs to finish installing, again
[17-Sep-2009 13:59:26] <iancmcc> No reason why it wouldn't work, remote ZEO works fine in DistributedCollector setup
\n[17-Sep-2009 13:59:41] <davetoo> I need to make a shell script or\nsomethign to change the colors in the main css sheets; I have so many\ndifferent systems I'm workign with, it's easy to b0rk the wrong one
[17-Sep-2009 13:59:58] <davetoo> maybe easier to hack the logo image
[17-Sep-2009 14:00:53] <ckrough> iancmcc: Im going to pitch it here, if we try it Im sure we'll need to get you guys involved
\n[17-Sep-2009 14:23:18] <kmtsun> in the rc.conf file...for\nsnmpd_flags= ... how do you seperate mutilple flags? with a space,\ncomma.. etc?
[17-Sep-2009 14:32:16] <jb> matt.. the timeout on the community site way too short
[17-Sep-2009 14:44:39] <rmatte> kmtsun: space I believe
[17-Sep-2009 14:45:05] <rmatte> from what I recall from my bsd days
[17-Sep-2009 14:45:31] <rmatte> you might have to enclose them in quotation marks as well
[17-Sep-2009 14:46:19] <rmatte> yeh, I was right, just found an example
[17-Sep-2009 14:46:21] <rmatte> snmpd_flags="-a -p /var/run/snmpd.pid"
[17-Sep-2009 14:46:27] <rmatte>
[17-Sep-2009 14:47:22] <rmatte> jb: timeout?
[17-Sep-2009 14:47:28] <rmatte> jb: I've yet to timeout on it
\n[17-Sep-2009 14:49:29] <rmatte> unhunh, so the WMI Data Sources\npack is totally messed for King Crab according to what Egor tells me.
[17-Sep-2009 14:49:31] <rmatte> Hey,
[17-Sep-2009 14:49:31] <rmatte>
[17-Sep-2009 14:49:31] <rmatte> we have a problem .
[17-Sep-2009 14:49:31] <rmatte>
\n[17-Sep-2009 14:49:31] <rmatte> 1. King Crab had compete new\nZenWin. WMIDataSources used ZenWin classes that no longer exist.
[17-Sep-2009 14:49:32] <rmatte> 2. I'm very busy now with my CCNP certification (until end of next month or so).
[17-Sep-2009 14:49:34] <rmatte>
[17-Sep-2009 14:49:36] <rmatte> Sorry.
[17-Sep-2009 14:49:38] <rmatte>
[17-Sep-2009 14:49:40] <rmatte> Egor
[17-Sep-2009 14:51:05] <rmatte> Figures that the best ZenPack to come out recently doesn't work in the new version
[17-Sep-2009 14:51:30] <mrayzenoss1> well, King Crab won't release for another month at least
[17-Sep-2009 14:51:45] <mrayzenoss1> jb: what was that about a timeout?
[17-Sep-2009 14:52:55] <rmatte> the timeouts might have more to do with his connection or browser than anything
[17-Sep-2009 14:53:04] <rmatte> I can leave the site open for hours, and come back to it, and it's fine
[17-Sep-2009 14:57:50] <davetoo> how odd,
[17-Sep-2009 14:58:05] <rmatte> ?
\n[17-Sep-2009 14:58:13] <davetoo> I wonder if these ldap servers\nare smart enough to lock my manager account out by individual source ips
[17-Sep-2009 14:58:38] <davetoo> talking to myself mostly, sorry
[17-Sep-2009 14:59:11] <rmatte> ah
[17-Sep-2009 14:59:30] <rmatte> I've never seen LDAP do anything like that
[17-Sep-2009 14:59:32] <davetoo> I was up all night working on a dmd script, very very tired
[17-Sep-2009 14:59:36] <davetoo> brain fuzzy
[17-Sep-2009 14:59:55] <davetoo> When it clears up I need to submit a patch for BatchDeviceLoader
[17-Sep-2009 15:00:00] <rmatte> yeh, I made the mother of all dmd scripts just put the finishing touches on it today
[17-Sep-2009 15:00:13] <davetoo> oh, you should see mine
[17-Sep-2009 15:00:17] <rmatte> hehe
[17-Sep-2009 15:00:23] <ckrough> rmatte: do share
[17-Sep-2009 15:00:54] <davetoo> evals expressions from command-line to build devfiler instances for the device searchers
[17-Sep-2009 15:01:00] <rmatte> http://pastebin.com/m4ba445b4
[17-Sep-2009 15:01:02] <rmatte> there's mine
[17-Sep-2009 15:01:07] <ckrough> Im working on a common module with reusable xmlrpc functions for building device lists, etc...
[17-Sep-2009 15:01:07] <ckrough> python that is
[17-Sep-2009 15:01:10] <rmatte> yours might be even more complex hehe
[17-Sep-2009 15:01:16] <davetoo> Basicaly what this one does is pull data useful for the new zenbatchloder
[17-Sep-2009 15:01:21] <ckrough> Ill get it to you when it's clean and documented a bit
[17-Sep-2009 15:01:25] <rmatte> ckrough: I think you've seen this one
[17-Sep-2009 15:01:30] <ckrough> it's bad right now, still in the works
[17-Sep-2009 15:01:32] <rmatte> I just added commandline options to it today
[17-Sep-2009 15:01:32] <davetoo> what's this one do?
[17-Sep-2009 15:01:35] <davetoo> oh
[17-Sep-2009 15:01:47] <davetoo> I havent' tackled xml/rpc yet
[17-Sep-2009 15:01:55] <rmatte> it generates a device availability report, but it's a standalone script, more flexible
[17-Sep-2009 15:01:58] <ckrough> ah this one
[17-Sep-2009 15:02:18] <rmatte> I also made stand-alone memory, cpu, and interface report scripts
[17-Sep-2009 15:02:25] <rmatte> we're going to use them to generate automated reports
[17-Sep-2009 15:02:39] <ckrough> whats the driving force behind this? doesnt zenoss handle most of that natively?
[17-Sep-2009 15:02:42] <rmatte> that availability script was a pain in the ass to write
[17-Sep-2009 15:02:53] <rmatte> all I had to go on was a single SQL query that Matt emailed to me
[17-Sep-2009 15:02:58] <rmatte> had to figure it out from there
[17-Sep-2009 15:03:13] <rmatte> So it was mostly a guessing game
[17-Sep-2009 15:03:27] <rmatte> but I managed to 100% duplicate the report, the results come out the exact same
[17-Sep-2009 15:03:30] <rmatte>
[17-Sep-2009 15:04:04] <rmatte> improved my python quite a bit along the way too
[17-Sep-2009 15:04:19] <ckrough> yeah
[17-Sep-2009 15:04:22] <ckrough> same with me
[17-Sep-2009 15:04:38] <davetoo> back in a minute... have to reboot
[17-Sep-2009 15:04:55] <rmatte> k
[17-Sep-2009 15:05:34]\n<rmatte> ckrough: it has native reports for this stuff, but we\nwant to be able to fully automate the reporting process
[17-Sep-2009 15:06:32]\n<rmatte> basically we'll have a daily average of availability\ndumped daily for each client, then at the end of the month we'll dump\nall of the reports for the month, then send that to the client along\nwith a list of top 10 devices/interfaces for utilization as well as a\ngraph showing overall availability from day to day during the month
[17-Sep-2009 15:06:37] <rmatte> and it'll all be automated
[17-Sep-2009 15:06:48] <rmatte> right now we're doing it by hand every month and it's extremely tedious
[17-Sep-2009 15:06:58] <rmatte> especially since the export function for the reports is broken in Zenoss
[17-Sep-2009 15:07:15]\n<rmatte> you think you're exporting for the period that you\nspecified but then it goes and exports for the last 7 days
[17-Sep-2009 15:07:33] <rmatte> so by creating external scripts we are eliminating the possibility of incorrect data
[17-Sep-2009 15:07:43] <rmatte> plus, the reports are going to stay the same regardless of Zenoss version
[17-Sep-2009 15:07:49] <rmatte> there are a lot of good reasons for these
\n[17-Sep-2009 15:13:57] <twm1010> is there any way to get zenoss\nto resolve hostnames of devices currently named by IP post-discovery?
[17-Sep-2009 15:14:14] <rmatte> twm1010: nah, you need to rename them by hand
[17-Sep-2009 15:14:26] <rmatte> twm1010: unless you whip up a zendmd script to do it
[17-Sep-2009 15:14:28] <mrayzenoss1> twm1010: snmp will rename them I believe
[17-Sep-2009 15:14:42] <mrayzenoss1> or maybe that's dns
[17-Sep-2009 15:15:00] <twm1010> ah... so i'd have to connect to every ILO card and/or RSA2 mgmt card and give it an actual hostname
[17-Sep-2009 15:15:01] <rmatte> I've never seen snmp rename them
[17-Sep-2009 15:15:02] <twm1010> RAR.
\n[17-Sep-2009 15:15:20] <rmatte> and I've never tried adding\nanything by IP when we have DNS as that defeats the purpose of having\ndns
[17-Sep-2009 15:15:21] <rmatte>
[17-Sep-2009 15:15:29] <twm1010> Ah.. I just need to do it by hand there really aren't that many.
\n[17-Sep-2009 15:15:44] <mrayzenoss1> I've got a OSX box that\nalways shows up as "elzar.local" and is monitored via SNMP. I can't\nfigure any other way it would get named that
[17-Sep-2009 15:16:02] <davetoo> So what breaks if we set the default snmp version, at the root of /devices, to v2c?
[17-Sep-2009 15:16:03] <twm1010> rmatte: see... i did my discovery against an incomplete DNS server, so alot of nodes got added by IP
[17-Sep-2009 15:16:13] <twm1010> davetoo: devices that require v1
[17-Sep-2009 15:16:22] <mrayzenoss1> twm1010: hah
[17-Sep-2009 15:16:34] <ckrough> rmatte: I see
[17-Sep-2009 15:16:50] <davetoo> and are those worse than the more modern ones that don't work with v1?
[17-Sep-2009 15:17:09] <twm1010> up to you, i change it a little further down the chain
[17-Sep-2009 15:17:20] <twm1010> my ../Windows is v2c, but others are not
[17-Sep-2009 15:17:26] <twm1010> like /Power is v1
[17-Sep-2009 15:17:35] <twm1010> because of crappy old UPS devices
\n[17-Sep-2009 15:18:20] <twm1010> but, i'm also debating not\nadding the ILO cards as separate devices, but rather manually adding\nthat IP to the server it manages
[17-Sep-2009 15:18:30] <twm1010> and then locking the interface table
\n[17-Sep-2009 15:18:35] <davetoo> I know I've run into weird\nbehavior, like the initial discovery wont' work with v2c, but continued\nmonitoring won't work withouth it...
[17-Sep-2009 15:20:48] <davetoo> What is the benefit/use of teh Default Admin Role setting for auser?
[17-Sep-2009 15:21:09] <iancmcc> Little, if you're not running Enterprise.
[17-Sep-2009 15:21:33] <davetoo> I am, so it must be an ACL thing I haven't learned yet...
\n[17-Sep-2009 15:25:49] <iancmcc> It is indeed. When you're adding\nadministered objects to a user, it's the default role they'll have
[17-Sep-2009 15:26:02] <iancmcc> Handy when you've got a bunch to add to somebody. Changing them all by hand is terrible.
\n[17-Sep-2009 15:42:45] <davetoo>\nzRouteMapCollectOnlyIndirect \n \nzRouteMapCollectOnlyLocal
[17-Sep-2009 15:42:52] <davetoo> I'm confused about these now
[17-Sep-2009 15:43:04] <davetoo> I thought it needed indirect routes to build the topology
[17-Sep-2009 15:55:32] <davetoo> hmm
[17-Sep-2009 16:05:18] <davetoo> zenbatchloader is not yet wonderful for new adds
[17-Sep-2009 16:05:28] <davetoo> but it's great for moving systems from my old core boxes
[17-Sep-2009 16:17:47] <kmtsun> anyone every seen this zenoss;snmp mesg: Unable to read processes on device
[17-Sep-2009 16:40:40] <LW> out of curiosity... has anybody created a ESX path checking plugin yet?
[17-Sep-2009 16:40:46] <LW> s/a/an
[17-Sep-2009 16:42:07]\n<caseyh2007> so who knows what is up with source build and\nFreeBSD? I am working on a port of Zenoss 2.4.5 right now, should I\nwait for 2.5 to come out?
[17-Sep-2009 16:42:37] <mrayzenoss1> caseyh2007: I see you saw my tweet
[17-Sep-2009 16:42:42] <caseyh2007> lol, yeah
[17-Sep-2009 16:42:44] mrayzenoss1 is now known as mrayzenoss
[17-Sep-2009 16:44:11] <mrayzenoss> so cgibbons is working on fixing http://dev.zenoss.org/trac/ticket/4994
[17-Sep-2009 16:44:56] <mrayzenoss> and so far he's got the pysamba stuff cleaned up, at least for Ubuntu 9.04
[17-Sep-2009 16:45:05] <mrayzenoss> http://dev.zenoss.org/svn/sandboxen/cgibbons/4994/
[17-Sep-2009 16:45:17] <mrayzenoss> but he could use testers for other platforms
[17-Sep-2009 16:45:26] <mrayzenoss> the fixes are intended for the upcoming 2.5 release
[17-Sep-2009 16:45:31] <cgibbons> tryin' it on Fedora 10 right now
[17-Sep-2009 16:45:49] <mrayzenoss> cgibbons: build completed for my Ubuntu 9.04
[17-Sep-2009 16:45:58] <cgibbons> cool
\n[17-Sep-2009 16:46:30] <mrayzenoss> caseyh2007: so if your\nFreeBSD box is ready, you can help test that part of the build if you\nwant
[17-Sep-2009 16:46:48] <mrayzenoss> actually that goes for anyone who wants to help test source builds
[17-Sep-2009 16:47:19] <mrayzenoss> just 'svn co http://dev.zenoss.org/svn/sandboxen/cgibbons/4994"
[17-Sep-2009 16:47:32] <mrayzenoss> and 'make pywmi-build'
[17-Sep-2009 16:47:44] <davetoo> w00t!
[17-Sep-2009 16:48:05] <caseyh2007> so you guys are going to make the source build work on FreeBSD with no changes in king crab?
[17-Sep-2009 16:48:15] <davetoo> I think I just fixed zenbatchloader and BatchDeviceLoader
[17-Sep-2009 16:48:25] <mrayzenoss> no, we want to fix the build before King Crab
\n[17-Sep-2009 16:48:52] <mrayzenoss> right now cgibbons is working\non a branch, but if it cleans up everything it'll go into 2.5 (King\nCrab)
[17-Sep-2009 16:49:05] <mrayzenoss> err, s/sandbox/branch/
[17-Sep-2009 16:49:14] <mrayzenoss> 2.5 is currently SVN trunk
[17-Sep-2009 16:49:35] <davetoo> ah
[17-Sep-2009 16:49:54] <mrayzenoss> 2.5 is a few weeks from branching
[17-Sep-2009 16:51:02] <mrayzenoss> the only platforms known to work right now for source builds are Debian, RHEL and OSX 10.5
[17-Sep-2009 16:51:14] <caseyh2007> k, well im gonna go ahead and make a port for 2.4.5 for now.
[17-Sep-2009 16:51:33] <mrayzenoss> caseyh2007: ok, just letting you know 2.5 is scheduled for next month
[17-Sep-2009 16:51:58] <caseyh2007> k, ill update when a new source build is out =)
[17-Sep-2009 16:54:21] <cgibbons> boom on fedora 10
[17-Sep-2009 16:54:27] <mrayzenoss> good boom?
[17-Sep-2009 16:54:32] <caseyh2007> lol
[17-Sep-2009 16:54:40] <cgibbons> nope
[17-Sep-2009 16:55:03]\n<mrayzenoss> caseyh2007: we don't have any FreeBSD boxes, so if\nyou could take a shot on that pysamba part, just building it, that\nwould help
[17-Sep-2009 16:55:21] <mrayzenoss>\nof course, if it doesn't work for Fedora 10, it's not looking good for\nother platforms
[17-Sep-2009 16:56:58] <cgibbons> yeah, lots of little changes
[17-Sep-2009 16:57:03] <cgibbons> there's some license issues i have to work out, too
[17-Sep-2009 16:57:17] <mrayzenoss> ugh
\n[17-Sep-2009 16:57:22] <cgibbons> but, hey! i just fixed\nsomething fun on the Windows stuff that's been bugging people for a\nwhile
[17-Sep-2009 16:57:31] <cgibbons> s/fixed/figured out/
[17-Sep-2009 17:32:24] <cgibbons> ah cool, Fedora 10 was easy to fix
[17-Sep-2009 17:32:41] <mrayzenoss> you want me to regression test Ubuntu?
[17-Sep-2009 17:32:53] <mrayzenoss> I'm thinking about doing my OSX 10.6 clean install tonight
[17-Sep-2009 17:32:59] <mrayzenoss> I could test there too
[17-Sep-2009 17:33:04] <cgibbons> nah already did (ubutu)
\n[17-Sep-2009 17:41:40] <cgibbons> alrighty, that branch is\nupdated. if you want to update do a 'make distclean' in the\nSamba/source directory first.
[17-Sep-2009 17:51:16] <mrayzenoss> branch or your sandbox?
[17-Sep-2009 17:52:28] <mrayzenoss> make: *** No rule to make target `gtk/config.mk', needed by `Makefile'. Stop.
[17-Sep-2009 17:53:12] <mrayzenoss> I'm building again
[17-Sep-2009 17:54:22] <cgibbons> my sandbox branch yeah
[17-Sep-2009 17:56:36] <davetoo> I tried doing a vendor branch for a bit
[17-Sep-2009 17:56:42] <davetoo> because i was doing a lot of workarounds
[17-Sep-2009 17:56:47] <davetoo> couldn't keep it up
[17-Sep-2009 18:00:42] <mrayzenoss> cgibbons: build still works
[17-Sep-2009 18:00:50] <mrayzenoss> I'm checking out for awhile
[17-Sep-2009 19:38:41] <metalklesk> hi
[17-Sep-2009 19:39:04] <davetoo> hmm
[17-Sep-2009 19:39:14] <davetoo> maybe I sholdnt have put 2.5 on my development systemyet
[17-Sep-2009 19:40:23] <davetoo> 'fact I think I'll revert
[17-Sep-2009 20:25:38] <metalklesk> one question
[17-Sep-2009 20:25:52] <metalklesk> i'm trying WBEM in Linux
[17-Sep-2009 20:26:27] <metalklesk> but in perf i can only see Load average and Memory utilization
[17-Sep-2009 20:26:36] <metalklesk> what about the other graphs ?
[17-Sep-2009 20:27:27] <metalklesk> i already have installed WBem linux monitor
[17-Sep-2009 20:27:44] <metalklesk> and bind the WBEMDevice to the device
[17-Sep-2009 20:28:56] <metalklesk> mmmm missing RRD file
[17-Sep-2009 21:31:44] <davetoo> I don't know how I managed this but now I cannot log out of zenoss
[17-Sep-2009 21:31:59] <davetoo> well, admin can't
[17-Sep-2009 21:32:42] <davetoo> oh maybe it's beucase I was in the top-level zmi
[17-Sep-2009 21:33:28] <davetoo> what
[17-Sep-2009 21:56:13] <davetoo> I jus tkilled my zope admin account
[17-Sep-2009 21:56:58] <davetoo> d'oh
[18-Sep-2009 00:00:46] [disconnected at Fri Sep 18 00:00:46 2009]
[18-Sep-2009 00:00:46] [connected at Fri Sep 18 00:00:46 2009]
[18-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[18-Sep-2009 07:21:12] <vinzpr> hi
[18-Sep-2009 07:21:17] <ckrough> hi
[18-Sep-2009 07:21:39] <vinzpr> need help on zenprocess anyone encounter this error on zenoss 2.4.5
\n[18-Sep-2009 07:22:34] <vinzpr> ERROR:zen.zenprocess:[Failure\ninstance: Traceback (failure with no frames):\nexceptions.UnboundLocalError: local variable 'update' referenced before\nassignment
[18-Sep-2009 07:22:35] <vinzpr> ]
[18-Sep-2009 07:22:35] <vinzpr> Traceback (most recent call last):
[18-Sep-2009 07:22:35] <vinzpr> File "/opt/zenoss/Products/ZenUtils/Driver.py", line 49, in _next
[18-Sep-2009 07:22:35] <vinzpr> self.iter.next().addBoth(self._finish)
[18-Sep-2009 07:22:36] <vinzpr> File "/opt/zenoss/Products/ZenRRD/zenprocess.py", line 474, in start
[18-Sep-2009 07:22:38] <vinzpr> self.updateDevices(driver.next(), devices)
[18-Sep-2009 07:22:40] <vinzpr> File "/opt/zenoss/Products/ZenRRD/zenprocess.py", line 459, in updateDevices
[18-Sep-2009 07:22:42] <vinzpr> d.updateConfig(cfg)
[18-Sep-2009 07:22:44] <vinzpr> File "/opt/zenoss/Products/ZenRRD/zenprocess.py", line 247, in updateConfig
[18-Sep-2009 07:22:46] <vinzpr> update.name)
[18-Sep-2009 07:22:48] <vinzpr> UnboundLocalError: local variable 'update' referenced before assignment
[18-Sep-2009 07:22:50] <vinzpr> INFO:zen.zenprocess:Daemon zenprocess shutting down
[18-Sep-2009 07:22:52] <vinzpr> INFO:zen.zenprocess:zenprocess shutting down
[18-Sep-2009 07:22:52] <ckrough> please use pastebin.com for that stuff
[18-Sep-2009 07:23:22] <ckrough> Does that happen everu start you start zenprocess?
[18-Sep-2009 07:23:30] <vinzpr> http://pastebin.ca/1543144
[18-Sep-2009 07:23:34] <vinzpr> yes
[18-Sep-2009 07:23:47] <vinzpr> it happens when i start zenprocess
[18-Sep-2009 07:24:12] <ckrough> looks like this has been hapenning for a while?
[18-Sep-2009 07:25:09] <vinzpr> any solution on this kind of error
[18-Sep-2009 07:25:17] <ckrough> not off the top of my head
[18-Sep-2009 07:25:29] <ckrough> did you modify anything with an 'update' variable?
[18-Sep-2009 07:25:37] <ckrough> and is it a new install?
[18-Sep-2009 07:25:39] <vinzpr> nope
[18-Sep-2009 07:26:15] <vinzpr> i update it for 2weeks already and it only appears this day
\n[18-Sep-2009 07:26:17] <cgibbons> something like that might\nhappen if zenhub hasn't been restarted after the upgrade or a zenpack\nchange, etc, but you'd likely be seeing more errors round the system too
[18-Sep-2009 07:26:55] <vinzpr> i restarted the zenhub
[18-Sep-2009 07:27:04] <vinzpr> but its still happen
[18-Sep-2009 07:28:26] <ckrough> Are the METAL templates zope objects or .pt files?
[18-Sep-2009 07:29:02] <vinzpr> im not sure when can i find that .pt files?
[18-Sep-2009 07:29:24] <ckrough> sorry vinzpr, I was asking a question for myself there
[18-Sep-2009 07:29:40] <ckrough> vinzpr: not related to your error
[18-Sep-2009 07:29:58] <vinzpr> ok
[18-Sep-2009 07:30:39] <vinzpr> anyone has an idea on this error http://pastebin.ca/1543144
[18-Sep-2009 07:32:59] <ckrough> answered my own question. templates.pt on the file system
[18-Sep-2009 07:33:08] <ckrough> vinzpr: sorry, Im not sure on that one
[18-Sep-2009 07:33:42] <vinzpr> ok..anyone encounter a problem like this?
[18-Sep-2009 07:33:50] <vinzpr> on zenprocess
[18-Sep-2009 07:33:52] <ckrough> first Ive seen it
[18-Sep-2009 07:34:08] <vinzpr> and...?
[18-Sep-2009 07:35:02] <ckrough> meaning, this is the first time Ive heard of that error, and I dont know
[18-Sep-2009 07:35:44] <vinzpr> ok
[18-Sep-2009 08:29:08]\n<ckrough> When I override \n/zport/portal_skins/zenui/img/onwhitelogo.png by customizing it, the\noriginal image still gets displayed.
[18-Sep-2009 08:30:16] <iancmcc> browser cache?
[18-Sep-2009 08:31:18] <ckrough> I tried clearing it. Ill try another machine
[18-Sep-2009 08:31:36] * ckrough doesnt trust firefox cache clearing
[18-Sep-2009 08:32:47] <ckrough> nah, not that
\n[18-Sep-2009 08:33:40] <iancmcc> I don't trust firefox cache\nclearing either, I have Web Developer Toolbar installed and just work\nwith my cache disabled.
[18-Sep-2009 08:34:03] <iancmcc> Hit up /zport/portal_skins/manage and make sure custom is at the top
[18-Sep-2009 08:34:35] <iancmcc> Sorry, go to the Properties tab and look at the layers textarea
[18-Sep-2009 08:34:36] <ckrough> nope, it's near the bottom
[18-Sep-2009 08:34:40] <ckrough> k
[18-Sep-2009 08:34:56] <ckrough> its first.
\n[18-Sep-2009 08:36:39] <ckrough> other new (non-override) images\nI put in there work, but that method would require customizing the\ntemplates, which is obviously bad mojo
[18-Sep-2009 08:36:45] <ckrough> very strange...
[18-Sep-2009 08:37:40] <iancmcc> yeah, that don't make no sense
[18-Sep-2009 08:37:54] <iancmcc> OH wait
[18-Sep-2009 08:37:56] <iancmcc> it won't work
[18-Sep-2009 08:38:05] <iancmcc> sorry, it's because it's in a folder
[18-Sep-2009 08:38:17] <ckrough> I don't follow
[18-Sep-2009 08:38:24] <iancmcc> so you have to override /zenui/img
[18-Sep-2009 08:38:30] <iancmcc> which is obviously not doable
[18-Sep-2009 08:39:18] <ckrough> so the individual components of the folder cant be overwritten
[18-Sep-2009 08:39:21] <iancmcc> Right.
[18-Sep-2009 08:39:35] <ckrough> oh, are you referring to folder as in filesystem directory?
[18-Sep-2009 08:39:38] <iancmcc> An old mistake that we have not been able to change
[18-Sep-2009 08:39:52] <iancmcc> well, yes, technically
[18-Sep-2009 08:40:11] <iancmcc> but the problem is Zopey in nature.
[18-Sep-2009 08:40:25] <iancmcc> once it resolves img, you're squarely in the zenui skin
[18-Sep-2009 08:40:37] <iancmcc> it doesn't know how to merge subfolders of different skin folders.
[18-Sep-2009 08:40:45] <ckrough> ah, img is the object, and zope is unaware of its contents?
[18-Sep-2009 08:40:56] <ckrough> err, they arent 'objects'?
\n[18-Sep-2009 08:41:28] <iancmcc> They're all "objects", including\nimg, but its url resolution isn't smart enough to straddle two\nsubfolders.
[18-Sep-2009 08:41:56] <ckrough> we've passed my comfort level with zope inheritance. I'll take your word for it
\n[18-Sep-2009 08:41:56] <iancmcc> Anyway, the solutions are\nequally silly: change the template, or symlink your image on the\nfilesystem
[18-Sep-2009 08:42:13] <ckrough> going to change the template.
[18-Sep-2009 08:42:33] <ckrough> thanks
\n[18-Sep-2009 08:42:38] <iancmcc> This will be fixed in Stone\nCrab, incidentally. Wonderful to have an opportunity to fix things.
[18-Sep-2009 08:42:53] <ckrough> which is stone crab? 2.7?
[18-Sep-2009 08:43:02] <iancmcc> the one after 2.5.
[18-Sep-2009 08:43:19] <iancmcc> Which might be 2.6, might be 3.0.
[18-Sep-2009 08:43:36] <iancmcc> Depending on how bold we feel at release planning.
[18-Sep-2009 08:43:42] <ckrough> gotcha
\n[18-Sep-2009 08:46:51] <iancmcc> Also, I forget who was asking\nfor it, rmatte or ckrough, but both row shading in the event detail\npane and the unacknowledge feature are now in trunk.
[18-Sep-2009 08:51:07] <ckrough> iancmcc: you do a lot of the UI stuff?
[18-Sep-2009 08:51:37] <iancmcc> Pretty much all of it. I've been typecast.
[18-Sep-2009 08:52:41] <ckrough> tracing the template calls and css is fun
[18-Sep-2009 08:52:46] <ckrough> onwhitelogo is elusive
[18-Sep-2009 08:52:56] <ckrough> everything pointed to head1, but no avail
[18-Sep-2009 08:54:21] <iancmcc> sadly, there are two places
[18-Sep-2009 08:54:31] <iancmcc> GraphReport.py and viewMultiGraphReportClean.pt
[18-Sep-2009 08:54:35] <ckrough> hmm, I was wrong. head1 is actually <head> info
[18-Sep-2009 08:54:37] <iancmcc> that is one old image
[18-Sep-2009 08:54:50] <ckrough> viewGraphReportClean is my starting point
[18-Sep-2009 08:55:01] <ckrough> possible in GraphReport.py, Ill look
[18-Sep-2009 08:57:35] <ckrough> bah, this requires a zopectl restart doesnt it
[18-Sep-2009 08:57:53] <iancmcc> yes
[18-Sep-2009 08:57:54] <ckrough> due to the pyc
[18-Sep-2009 08:57:56] <ckrough> kk
[18-Sep-2009 08:58:27] <ckrough> I think GraphReport.py was the one. I was searching under skins and didnt catch it
[18-Sep-2009 09:07:19] <ckrough> bah, pyo file
[18-Sep-2009 09:07:46] <ckrough> apparently pyos dont get rebuilt automatically like pyc's
[18-Sep-2009 09:13:05] <jb> iancmcc: that was me that asked about row shading..
[18-Sep-2009 09:13:25] <iancmcc> Well, it's in.
[18-Sep-2009 09:13:28] <jb> thanks
[18-Sep-2009 09:20:07] <ckrough> iancmcc: ha, its not GraphReport.py either
[18-Sep-2009 09:20:33] <iancmcc> I grepped all the code, that's the only place it appears
[18-Sep-2009 09:20:33] <ckrough> iancmcc: oh well, moving on to other things. Ill have to come back to this one
[18-Sep-2009 09:20:36] <iancmcc> lemme check enterprise
[18-Sep-2009 09:20:38] <ckrough> yeah, same
[18-Sep-2009 09:20:46] <ckrough> and I opened it on another box with clear cache
\n[18-Sep-2009 09:20:56] <etank> we are using zenoss enterprise and\nfor some reason we have stopped getting datapoints in our graphs for\nmssql. this may have happened after a failover (active/passive\nclustered systems). what can be done to start getting data again?
[18-Sep-2009 09:21:31] <jb> you are monitoring the cluster virtual IP or the individual cluster nodes?
[18-Sep-2009 09:22:24] <etank> jb in this case it is the cluster virtual ip
[18-Sep-2009 09:22:37] <jb> hrm yeah, you may need to delete those RRD's
[18-Sep-2009 09:22:50] <etank> we are monitoring the physical nodes as well though
[18-Sep-2009 09:22:55] <etank> ok
[18-Sep-2009 09:22:59] <etank> i will give that a try
[18-Sep-2009 09:23:17] <jb> just curious.. why monitor the virtual ip?
[18-Sep-2009 09:23:29] <jb> if you are already monitoring the physical nodes?
[18-Sep-2009 09:23:50] <etank> so that the graphs are specific to the clustered services
\n[18-Sep-2009 09:24:13] <etank> we dont want to have to figure out\nfirst which node is active and then go figure look at the graphs
[18-Sep-2009 09:25:52] <jb> k
\n[18-Sep-2009 09:42:25] * etank is having a hard time trying to decide\nif he should go to the Zenoss talke that mrayzenoss doing at OLF or the\nreStructured Text:
[18-Sep-2009 09:42:28] <etank> Plain Text Gets
[18-Sep-2009 09:42:31] <etank> one
[18-Sep-2009 09:43:08] <etank> (that paste did not do what i wanted it to)
[18-Sep-2009 09:47:18] <mrayzenoss> etank: there's the free training on Friday, and a demo on Saturday
[18-Sep-2009 09:47:33] <mrayzenoss> and a party Friday night
[18-Sep-2009 09:47:42] <ckrough> mrayzenoss: any news on a recast of Jane's presentation? I got pulled away...
[18-Sep-2009 09:48:01] <mrayzenoss> Tuesday
[18-Sep-2009 09:48:07] <mrayzenoss> I'll go do the announcement
[18-Sep-2009 09:48:15] <ckrough> thanks
[18-Sep-2009 09:55:45] <rmatte> ckrough: it was a pretty good presentation, did you miss it?
[18-Sep-2009 09:59:50] <ckrough> yeah, typical... 10 minutes in and I had to deal with something
[18-Sep-2009 10:00:05] <ckrough> I need to hide in a broom closet with a laptop to get through a whole webex
[18-Sep-2009 10:02:36]\n<etank> mrayzenoss: yeah we will be there for the training so i\nwill prolly do the reStructured Text talk on Saturday
[18-Sep-2009 10:03:36] <rocket> rmatte: ping
[18-Sep-2009 10:03:48] <rmatte> rocket: yus?
[18-Sep-2009 10:04:11] <rocket> mrayzenoss: tells me you run solarwinds and zenoss
[18-Sep-2009 10:04:28] <rocket> I have a solarwinds demo i am attending in a few and was wondering if you had an opinion
[18-Sep-2009 10:04:33] <rmatte> well, we are backing away from using solarwinds apparently
[18-Sep-2009 10:04:51] <rocket> oh?
[18-Sep-2009 10:04:58] <rocket> any reason in particular?
[18-Sep-2009 10:05:14]\n<rmatte> being an MSP, when we found out that we'd have to pay\nfor each collector (which wasn't communicated to us in the beginning)\nit just became way too expensive
[18-Sep-2009 10:05:39] <rmatte> so we would have ended up with using 1 collector with some monstrosity of a NATting setup
[18-Sep-2009 10:05:42] <rmatte> waste of time
[18-Sep-2009 10:06:01] <rocket> makes sense ... aside from that how was it technically?
[18-Sep-2009 10:06:22] <rocket> eg would you still recommend zenoss over it etc ..
[18-Sep-2009 10:06:47]\n<rmatte> it was alright, they are definitely going for a whole\n"drop it in and it just works approach", but then again, if there's\nsomething it doesn't happen to do it's quite difficult to get it to do\nwhatever that is
[18-Sep-2009 10:07:11] <rmatte> I would definitely recommend Zenoss over it for everything except the reporting
[18-Sep-2009 10:07:20] <rmatte> which is all we were really using solarwinds for
[18-Sep-2009 10:07:43] <rmatte> solarwinds' reports system is awesome
[18-Sep-2009 10:08:00]\n<mrayzenoss> ckrough: how's this URL:\nhttps://zenossevents.webex.com/ec0605l/eventcenter/enroll/join.do?confViewID=579483694&confId=579483694&siteurl=zenossevents&path=program_detail&theAction=detail
[18-Sep-2009 10:08:19] <mrayzenoss> I'm trying to find a direct link to the event, WebEx is annoyingly hiding everything via Javascript
[18-Sep-2009 10:08:25]\n<rmatte> but now that I'm getting the hang of writing external\npython scripts to gather data so that we can fully automate reporting\nfrom Zenoss, it's not becoming such an issue
[18-Sep-2009 10:08:56] <rmatte> with your python skills I doubt that solarwinds would give you much of an edge
[18-Sep-2009 10:11:20] <rmatte> Solarwinds is more for the lazy admin in my opinion
\n[18-Sep-2009 10:14:10] <mrayzenoss> anyone interested in the\nEvent Management Session, please test this URL for me:\nhttps://zenossevents.webex.com/ec0605l/eventcenter/enroll/join.do?confViewID=579483694&confId=579483694&siteurl=zenossevents&path=program_detail&theAction=detail
[18-Sep-2009 10:16:23] <ckrough> no good
[18-Sep-2009 10:16:26] <ckrough> the last one worked though
\n[18-Sep-2009 10:17:15] <ckrough> good link:\nhttps://zenossevents.webex.com/ec0605l/eventcenter/enroll/join.do?confViewID=579483694&confId=579483694&siteurl=zenossevents&path=program_detail&theAction=detail
[18-Sep-2009 10:17:28] <mrayzenoss> ok, I'll put that in the blog
[18-Sep-2009 10:19:08] <mrayzenoss> published
[18-Sep-2009 10:19:08] <cgibbons> hey matt, guess what?
[18-Sep-2009 10:19:12] <mrayzenoss> what's that?
[18-Sep-2009 10:20:26] <cgibbons> that latest code builds on freebsd
[18-Sep-2009 10:21:16] <mrayzenoss> woohoo!
[18-Sep-2009 10:21:34] <mrayzenoss> so... Ubuntu, FreeBSD and Fedora are all good?
[18-Sep-2009 10:23:25] <ckrough> CentOS 5.4 should be out in a week or two. have you tested on RHEL5.4?
[18-Sep-2009 10:24:40] <kmtsun> how do you turn off auto-discover?
[18-Sep-2009 10:26:35] <kmtsun> would that be: Managment -> settings -> jobs?
[18-Sep-2009 10:27:29] <mrayzenoss> ckrough: just checked with Dev mgmt, we'll QA RHEL 5.4 before shipping Enterprise 2.5
[18-Sep-2009 10:28:08] <ckrough> I was just wondering if you had done core because CentOS5.4 will release right around the same time
[18-Sep-2009 10:28:18] <ckrough> and I assume a good number of people run CentOS with zenoss core
[18-Sep-2009 10:28:31] <mrayzenoss> definitely
[18-Sep-2009 10:28:49] <ckrough> kmtsun: /networks zproperies -> zAutoDiscover = true|false
[18-Sep-2009 10:28:52] <mrayzenoss> RHEL/CentOS are 90% of Enterprise installs
[18-Sep-2009 10:29:03] <mrayzenoss> probably around 40-50% of Core
[18-Sep-2009 10:29:16] <jb> cent here..
[18-Sep-2009 10:30:07] <rmatte> I think a lot of core is Ubuntu
[18-Sep-2009 10:30:13] <mrayzenoss> yeah
[18-Sep-2009 10:31:49] <mrayzenoss> we don't know the breakdown for distros, only for platforms
[18-Sep-2009 10:31:53] <mrayzenoss> and kernel versions
[18-Sep-2009 10:40:08] <kmtsun> if you turn off autodiscovery in zenoss will it stop monitoring discovered machines via snmp wmi ssh etc?
[18-Sep-2009 10:42:30] <rmatte> kmtsun: not sure what you mean, turn off autodiscovery where?
\n[18-Sep-2009 10:42:59] <rmatte> there's no active auto-discovery\nthat I'm aware of, you have to manually specify what subnet you want\nzenoss to auto-discover on
[18-Sep-2009 10:43:32] <kmtsun> right, I don't want zenoss to find new clients... just moninitor the ones it found
[18-Sep-2009 10:43:49] <ckrough> kmtsun: right, if you turn off autodisc it will stop finding new devices
[18-Sep-2009 10:44:37] <kmtsun> but my question is .. will it stop monitoring 'discovered' devices? ie ssh wmi snmp
[18-Sep-2009 10:44:44] <rmatte> ckrough: what option is he referring to? (I don't use autodisc, so no idea)
[18-Sep-2009 10:44:45] <ckrough> no
[18-Sep-2009 10:44:51] <kmtsun> if auto is turned off
[18-Sep-2009 10:45:01] <ckrough> rmatte: check the zpropteries of blocks in the /Networks class
\n[18-Sep-2009 10:45:23] <perr0> morning all, when trying to\ninstall from source on Gentoo I am getting an error message and it\nfails building wmi-something. the forums suggest to revert back to\nautoconf-2.59 which I have but what I don't understand is which\napplication is where x listed here >\ndocs/DOC-2607
[18-Sep-2009\n10:45:24] <ckrough> kmtsun: right, autodiscovery is just for\nfinding new devices, zenoss will continue to monitor existing devices
[18-Sep-2009 10:45:56] <kmtsun> thank you
[18-Sep-2009 10:46:01]\n<rmatte> ckrough: ohhhh, yeh, but that just means that Zenoss\nwill autodiscover on that subnet when an auto-discovery is done, it\ndoesn't mean that it's going to actively go out searching that subnet\nevery so many hours
[18-Sep-2009 10:46:27] <perr0> I should say the dependancy "where x" listed in the list I posted
\n[18-Sep-2009 10:46:31] <rmatte> kmstun: Zenoss monitors anything\nin a production state based on the Device zProperties, not the network\nzProperties
[18-Sep-2009 10:48:18] <kmtsun> i\nwant things in porduction state monitored... just don't want it to find\nnew devices... i want to do that myself
[18-Sep-2009 10:49:17] <kmtsun> so i should use the zproperties of /Network class?
[18-Sep-2009 10:49:19] <ckrough> rmatte: you sure?
[18-Sep-2009 10:49:31] <ckrough> kmtsun: were not positive
[18-Sep-2009 10:49:38] <etank> i am not seeing any SQL related rrd files being built now for one of my MS SQL servers
[18-Sep-2009 10:49:49] <etank> i have even deleted the device and re-added it
[18-Sep-2009 10:49:59] <etank> it has been more than 15 minutes too
\n[18-Sep-2009 10:50:12] <ckrough> rmatte: my understanding was the\nthe autodiscover process looked at those network zProperties when\nrunning on a cycle
[18-Sep-2009 10:52:44]\n<rmatte> kmtsun: unless you explicitly tell it to go out and do a\ndiscovery though, it won't do it on it's own, is what I was saying
[18-Sep-2009 10:53:10] <kmtsun> i see thanks
\n[18-Sep-2009 10:53:33] <rmatte> so if that's what you're worried\nabout then that's not an issue, there is absolutely no automated\nauto-discovery built in to Zenoss, you have to explicitly tell it to do\nany sort of auto-discovery
[18-Sep-2009 10:55:44]\n<perr0> So does anyone at this time know which application is\n"where x>22." listed in the depandancies list >\ndocs/DOC-2607
[18-Sep-2009 10:56:23] <mrayzenoss> perr0: the MySQL version must be 5.0.22 or greater
\n[18-Sep-2009 10:57:28] <ckrough> rmatte: ha, I do it via 'zenoss'\nuser crontab... been that way so long I forgot its not standard
[18-Sep-2009 11:02:01] <perr0> mrayzenoss: thanks that is a little confusing having it on seperate lines
[18-Sep-2009 11:02:27] <mrayzenoss> yeah, we're still working out the kinks
[18-Sep-2009 11:09:51]\n<ckrough> I have this one switch that keeps coming up snmp down,\nbut it can be walked and tcpdump shows the snmpstatus check going out\nand coming back
[18-Sep-2009 11:11:21] <ckrough> tried clearing events and clearning heartbeats... everything.
\n[18-Sep-2009 11:14:48] <mrayzenoss> anyone interested in a\nZenoss-sponsored Developer training course? \nmessage/39558
[18-Sep-2009 11:14:54] <mrayzenoss> Looking for feedback
[18-Sep-2009 11:17:04] <Brad_K> mrayzenoss: is there a Zenoss-sponsored Admin training course?
[18-Sep-2009 11:18:17] <mrayzenoss> yeah, there's Zenoss training http://www.zenoss.com/support#training
[18-Sep-2009 11:18:20] <etank> mrayzenoss: i am interested in any training course
[18-Sep-2009 11:18:29] <mrayzenoss> and we'll have free training at Ohio LinuxFest next week
\n[18-Sep-2009 11:18:56] <mrayzenoss> Brad_K: assuming the Texas\nLinuxFest takes off, we hope to throw some free training into that as\nwell
[18-Sep-2009 11:19:38] <mrayzenoss> and LISA and SCALE will have free training
[18-Sep-2009 11:21:32] <etank> what can i look at to find out why my rrd files are not being built?
[18-Sep-2009 11:24:33] <mrayzenoss> bad graphs, empty graphs or no graphs?
[18-Sep-2009 11:24:46] <etank> no rrd files so no graphs at all
[18-Sep-2009 11:25:00] <etank> for the MSSQL related graphs that is
[18-Sep-2009 11:25:01] <mrayzenoss> is there a placeholder for the graphs?
[18-Sep-2009 11:25:10] <etank> yes there is a placeholder
[18-Sep-2009 11:25:33] <mrayzenoss> that usually indicates an issue between the data source and the data points
[18-Sep-2009 11:25:44] <etank> i have tried restarting zenoss and running zenocommand but nothing
[18-Sep-2009 11:26:05] <etank> and i have recently deleted and re-added the device
[18-Sep-2009 11:26:28] <mrayzenoss> does the 'Test' button work?
[18-Sep-2009 11:26:34] <etank> checking now
[18-Sep-2009 11:27:27] <etank> i get
[18-Sep-2009 11:27:28] <etank> WARNING:zen.winperf.PerfRpc:Bad counter: \\SQLServer:Latches\\Latch Waits/sec
[18-Sep-2009 11:27:28] <etank> INFO:zen.winperf:Unable to collect value for \\SQLServer:Latches\\Latch Waits/sec
[18-Sep-2009 11:28:31] <etank> it does work against a different mssql server though
\n[18-Sep-2009 11:40:35] <Brad_K> mrayzenoss: I definitely support\nmore training, if you can make it happen. i'm now on the team\ndeploying zenoss here at UT Austin/ITS, so I'd welcome any additional\ntraining that i can get.
[18-Sep-2009 11:40:36]\n<russb> i've seen perfmon counters dissapear before...can you\ncheck on that sql server (using perfmon.exe) and verify that the\ncounter is there?
[18-Sep-2009 11:41:09]\n<Brad_K> etank: we're having similar problems with zencommand\ncounters mysteriously disappearing and then reappearing hours later.
[18-Sep-2009 11:42:06]\n<Brad_K> mrayzenoss: in fact, i'm just about to file requests\nwith my management for zenoss training, books, instructional videos,\netc....
[18-Sep-2009 11:43:51] <Brad_K> I did\nfind one book on Amazon that is specifically about zenoss, namely\n"Zenoss Core Network and System Monitoring: A step-by-step guide to\nconfiguring, using, and adapting this free Open Source network\nmonitoring system". Does anyone know if this is any good?
[18-Sep-2009 11:46:06] <Brad_K> For that matter, is the book "Python for Unix and Linux System Administration" any good?
[18-Sep-2009 11:55:40] <twm1010> Brad_K: I've read that book. It is good, but beware that its for Zenoss 2.2 I believe"
[18-Sep-2009 11:55:48] <twm1010> I wrote a review on it, never got around to sending it to the author.
[18-Sep-2009 11:56:05] <twm1010> Slow reading of the admin guide will get you about the same.
\n[18-Sep-2009 11:56:52] <twm1010> Shortly here I intend to focus\non how to get the most information from each brand of server from top\nto bottom... IE.... an HP server, making sure it has an ILO and insight\nmanager correctly configured to get maximum info back to zenoss
[18-Sep-2009 12:09:06] <ckrough> mrayzenoss: yes on the course
\n[18-Sep-2009 12:11:47] <ckrough> Brad_K: just catching up, Jane\nCurry's site has some good stuff on it too\nhttp://www.skills-1st.co.uk/mgt/
[18-Sep-2009 12:12:04] <etank> is there a way to use TALES expressions in a WinPerf datasource?
\n[18-Sep-2009 12:12:29] <etank> Brad_K: i failed over the server\nand started getting data again. something on the local box was the issue
[18-Sep-2009 12:13:00] <etank> im trying to do something like this in the datasource
[18-Sep-2009 12:13:05] <etank> \\SQLServer:Databases(${here/cDatabaseName})\\Data File(s) Size (KB)
[18-Sep-2009 12:13:31] <etank> it works if i hard code the database name in the perf counter line
[18-Sep-2009 12:17:54] <Brad_K> mrayzenoss: also, is there a Zenoss user's group here in Austin?
[18-Sep-2009 12:21:20] <twm1010> on that note, is there a list of all of them?
[18-Sep-2009 12:30:14] <Brad_K> for that matter, are there any Zenoss user groups anywhere?
[18-Sep-2009 12:31:02] <twm1010> Brad_K: I would think not, I would imagine Zenoss gets discussed alot at Linux user groups though.
[18-Sep-2009 12:31:16] <mrayzenoss> sorry, I was at lunch the last 45 minutes
\n[18-Sep-2009 12:31:35] <twm1010> I find that I have trouble\ngetting Linux users interested in it for some reason. I just don't\nunderstand it.
[18-Sep-2009 12:31:56]\n<mrayzenoss> there aren't any Zenoss User Groups that I'm aware\nof, that said there are a large number of Zenoss users (and employees)\nin Austin
[18-Sep-2009 12:32:00] <twm1010> Especially anyone who has put time into Nagios or Hobbit, I can't even get them to look at it.
\n[18-Sep-2009 12:32:41] <mrayzenoss> the Zenoss book from Packt is\ngetting a bit dated, it was written around the 2.2 time frame
\n[18-Sep-2009 12:33:29] <ckrough> I havent gotten much buy-in even\non in-house groups. we have different groups using it but everyone's\nneeds are so varied there isn't much of a rallying point to get excited\nabout
[18-Sep-2009 12:33:48] <mrayzenoss>\nBrad_K: we've usually offered training in Annapolis, but we could\nobviously host it in Austin
[18-Sep-2009 12:33:58] <mrayzenoss> since our office is up on Spicewood Springs
[18-Sep-2009 12:34:03] <mrayzenoss> and there's plenty of room
\n[18-Sep-2009 12:34:12] <Brad_K> twm1010: i'm in most of the Linux\nand Unix-related groups here in Austin, and I co-founded the local\nAustin chapter of the League of Professional System Administrators. \nwe've had Matt come give a presentation on Zenoss before, but that's\nall i've ever seen at any of the groups.
[18-Sep-2009 12:34:13] <ckrough> some people here would drive up matt
[18-Sep-2009 12:34:36] <mrayzenoss> Brad_K: yeah, I did a demo for the Python User Group too
[18-Sep-2009 12:34:59] <mrayzenoss> ok, I'll see about updating the training stuff to mention Austin as an option
\n[18-Sep-2009 12:35:36] <Brad_K> that would be very helpful for me\nand my management, since we've been told that it's going to be\nvirtually impossible to get any training approved that will require\ntravel and hotel costs.
[18-Sep-2009 12:35:45] <ckrough> but I live flying back east for training. you're messing with my vacations now.
[18-Sep-2009 12:35:48] <ckrough>
[18-Sep-2009 12:35:57] <venturaville> mrayzenoss: are any of you guys going to make LISA ?
[18-Sep-2009 12:36:05] <Brad_K> of course, we've also got a support contract, but i don't know if that covers training.
[18-Sep-2009 12:36:05] <mrayzenoss> yeah, we'll have a big presence at LISA
[18-Sep-2009 12:36:28] <venturaville> I and one other fellow from here are going to be up there
[18-Sep-2009 12:36:31] <mrayzenoss> Brad_K: it's usually included in contracts
\n[18-Sep-2009 12:36:31] <Brad_K> yes, they're doing a one-day\ncommunity session, plus other stuff. See\nhttp://www.zenoss.com/news/events
[18-Sep-2009 12:36:52] <mrayzenoss> Chet has a Guru session at LISA, we'll have a day of free training
[18-Sep-2009 12:37:03] <mrayzenoss> and we'll have a booth
\n[18-Sep-2009 12:37:31] <Brad_K> i know all about the stuff that's\nlinked from that page, because i just sent an e-mail message to my\nmanagement talking about all the training and other instructional\nmaterials that i'm going to need, if i'm going to be able to help in\nthis process.
[18-Sep-2009 12:37:37] <venturaville> I'm going to try and make the puppet and zenoss guru sessions
[18-Sep-2009 12:37:57] <ckrough> Puppet is coming to Austin in November
[18-Sep-2009 12:38:02] <ckrough> I should be at that
[18-Sep-2009 12:38:21] <mrayzenoss> yeah, we may be attending the Puppet training as well
[18-Sep-2009 12:38:24] <Brad_K> we're going to have a presentation on bcfg2 at the LOPSA-Austin meeting next week.
\n[18-Sep-2009 12:38:53] <Brad_K> and if that works out, i'll see\nif i can bring the speaker back to give the same presentation at the\nnext CACTUS meeting.
[18-Sep-2009 12:39:30] <mrayzenoss> Brad_K: yeah, I've been meaning to do a presentations for CACTUS, AustinLUG and ALE
[18-Sep-2009 12:39:43] <mrayzenoss> hard to get free time at nights
\n[18-Sep-2009 12:39:46] <Brad_K> that is, if you're interested in\nthe general topic of configuration management, and not just\nspecifically Puppet.
[18-Sep-2009 12:39:47] <mrayzenoss> (2 kids)
[18-Sep-2009 12:39:56] <mrayzenoss> Brad_K: which night?
\n[18-Sep-2009 12:40:21] <Brad_K> mrayzenoss: anytime you are\nwilling or able to do a presentation for CACTUS, please let me know. \ni'm the program chair, and i'm the one responsible for organizing the\nspeakers. ;-)
[18-Sep-2009 12:41:26] <venturaville> what is CACTUS?
\n[18-Sep-2009 12:41:30] <Brad_K> CACTUS meets on the third\nthursday of the month, and the recent meetings have been at a Church\nnear 43rd and B Avenue, where one of the board members is also a member\nof the church (probably a deacon or something).
\n[18-Sep-2009 12:41:46] <etank> mrayzenoss: do you know if there\nis a way to use TALES expressions in a WinPerf datasource?
\n[18-Sep-2009 12:41:56] <Brad_K> CACTUS is the Capital Area\nCentral Texas Unix Society, and is the oldest Unix user group in Austin.
[18-Sep-2009 12:42:11] <twm1010> that's quite an acronym
[18-Sep-2009 12:42:23] <mrayzenoss> etank: I'd assume there is, I'll double check
[18-Sep-2009 12:42:27] * mrayzenoss doesn't run Enterprise
[18-Sep-2009 12:42:33] <Brad_K> they came up with it years ago.
[18-Sep-2009 12:42:45] <etank> im trying something like this --> \\SQLServer:Databases(${here/cDatabaseName})\\Data File(s) Size (KB)
[18-Sep-2009 12:43:06] <etank> but it isnt changing the cher/cDatabasename to what it should be
[18-Sep-2009 12:43:08]\n<Brad_K> LOPSA-Austin meets on the fourth Tuesday nights of the\nmonth, and right now we're trying to nail down the location.
[18-Sep-2009 12:43:21] <mrayzenoss> Brad_K: my 3rd Thursday of the month is bad the next 3 months... January?
[18-Sep-2009 12:43:36] <ckrough> Brad_K: is there a date for the bcfg2 talk? dont see it on the website
[18-Sep-2009 12:43:45] <Brad_K> We could plan for January, sure. that would give us plenty of time to get the word out.
[18-Sep-2009 12:44:07] <Brad_K> the LOPSA-Austin website needs to be updated. we've been holding off doing that, pending the location.
\n[18-Sep-2009 12:44:55] <Brad_K> but Travis Campbell (who is the\nother co-founder of LOPSA-Austin) has arranged the speaker, and we've\ngot several irons in the fire with regards to the location.
[18-Sep-2009 12:45:28] <mrayzenoss> ok, I've put January 21 on my calendar, pencil me in
[18-Sep-2009 12:46:02]\n<Brad_K> Ironically, Travis is also on the LOPSA Board, and I\nbelieve he's in the process of taking over the position of the local\nUSENIX representative on UT campus. So, they'll send him to one USENIX\nconference per year, for free.
[18-Sep-2009 12:46:28]\n<Brad_K> I'll let everyone on the CACTUS board know, but you\nmight want to send me a reminder as we get closer. Thanks!
\n[18-Sep-2009 12:49:49] <etank> mrayzenoss: trying a few different\nthings it seems that zenwinperf passes the --testCounter as is
[18-Sep-2009 12:49:58] <etank> no substitution
[18-Sep-2009 12:55:46] <daMaestro> anyone know of a nginx zenpack that utilizes stub_status output
[18-Sep-2009 12:55:47] <daMaestro> ?
[18-Sep-2009 12:56:54] <cgibbons> etank: yeah nothing in the data sources does tales expressions.
[18-Sep-2009 13:03:10] <metalklesk> hi
[18-Sep-2009 13:09:08] <daMaestro> guess not (from googling for a while)
[18-Sep-2009 13:09:28] <daMaestro> i'll be creating a zenpack based on the stub_status output
[18-Sep-2009 13:09:42] <dollarbang> Does Zenoss have the ability to push configurations to Routers/Switches?
\n[18-Sep-2009 13:11:33] <ckrough> dollarbang: you could probably\nwrite something, but I haven't seen and docs or zenpacks for that
[18-Sep-2009 13:11:42] <ckrough> dollarbang: you can *pull* them with Rancid
\n[18-Sep-2009 13:12:59] <dollarbang> ckrough: thanks. These are\nCisco Routers and Switches, and our license is coming up for renewal.\nMy company is looking for any savings they can get.
\n[18-Sep-2009 13:14:23] <Brad_K> hmm. my understanding was that\nrancid could both pull and push configurations to/from cisco\nequipment. i guess i need to go do some googling....
[18-Sep-2009 13:15:29] <etank> cgibbons: dang. ok thanks.
\n[18-Sep-2009 13:15:55] <ckrough> Very possible. I havent seen\nthat myself. There are docs for pulling configs in response to events\n(wr trap, etc..) on the zenoss site
[18-Sep-2009 13:16:01]\n<Brad_K> hmm. okay, the page at http://www.shrubbery.net/rancid/\nseems to be pretty clear -- monitoring, not management.
[18-Sep-2009 13:16:09] <etank> kinda sucks to have to create seperate datasources with hard coded stuff then to get what you need.
[18-Sep-2009 13:16:36] <ckrough> docs/DOC-2716
[18-Sep-2009 13:17:28] <Brad_K> blargh.
[18-Sep-2009 13:17:50] <Brad_K> gotta watch which window i'm typing in which command.
[18-Sep-2009 13:21:35] <rmatte> lol
\n[18-Sep-2009 13:22:44] * Brad_K also did not know that MacIrssi\ntranslates "<command>-W" into "close this irc channel and make\nthat tab go away"
[18-Sep-2009 13:36:15]\n<eidolon> hi folks - we're having a real problem here and need to\ngetthis fixed before hte weekend. we have zenoss enterprise, we've\nWMI-modelled some win2k8 (i believe) servers. The filesystems are\nshowing up (Total bytes), ut Used bytes and Free bytes and %util are\nshowing unknown.
[18-Sep-2009 13:36:30]\n<eidolon> we've waited until a few monitoring cycles have passed,\nbut the numbers have not magically appeared.
\n[18-Sep-2009 13:36:43] <eidolon> help please. these are\nproduction servers we've had outages on becasuse of disk space\nmonitoring failures
\n[18-Sep-2009 13:37:23] <ckrough> sounds like something you shoudl\nsubmit a support ticket for? assuming you have paid support as\nenterprise customer
[18-Sep-2009 13:37:42]\n<Brad_K> I'm sure the more experience people on the channel would\nneed to know what version of Zenoss Enterprise you're using, and what\nzenpacks you're using.
[18-Sep-2009 13:37:56] <Brad_K> and definitely open a ticket as well.
\n[18-Sep-2009 13:37:59] <eidolon> yes, but my login is failing -\nthis needs to betimely though. i'm going to try to recover the\npassword and get in, but i have a short timewindow.
[18-Sep-2009 13:38:29] <lsacpinbal> what template you using?
[18-Sep-2009 13:38:37] <lsacpinbal> is it WMI?
[18-Sep-2009 13:38:40] <eidolon> yes WMI.
[18-Sep-2009 13:38:41] <lsacpinbal> or just snmp
[18-Sep-2009 13:38:44] <eidolon> Zenoss 2.4.1
[18-Sep-2009 13:38:45] lsacpinbal is now known as pinball_
[18-Sep-2009 13:38:49] <eidolon> no snmp
[18-Sep-2009 13:39:28] <eidolon> Performance templates: Device WMI and SQLServer
[18-Sep-2009 13:40:07]\n<Brad_K> we're not yet in production with our Zenoss system, but\nso far the only thing we've found that works reliably is the SNMP\nstuff. I don't think we've tried to do any WMI stuff yet, and we're\nhaving major problems with anything running under zencommand.
[18-Sep-2009 13:40:25] <pinball_> yeah i agree with Brad
[18-Sep-2009 13:40:32] <pinball_> we are just working on WMI on our dev site
[18-Sep-2009 13:40:41] <eidolon> :-/
[18-Sep-2009 13:40:43] <pinball_> snmp seems to be the best opetion in this case
[18-Sep-2009 13:40:47] <pinball_> can you configure snmp service
[18-Sep-2009 13:40:50] <pinball_> on your windows host
[18-Sep-2009 13:40:58] <eidolon> i've been told several times that for windows boxes, we should be using WMI.
[18-Sep-2009 13:41:03] <pinball_> use both
[18-Sep-2009 13:41:35] <pinball_> at least with SNMP the service will send a trap when things start to break
[18-Sep-2009 13:41:36] <cgibbons> pinball_/brad are you guys enteprise uses or core with community zenpacks?
\n[18-Sep-2009 13:41:39] <Brad_K> and yes, we have a ticket open on\nthis problem. i've got 3GB of logs that i've made available to them,\nso that they can help us figure out what is going on.
[18-Sep-2009 13:41:39] <pinball_> and you can always relay on that
[18-Sep-2009 13:41:43] <pinball_> enterprise
[18-Sep-2009 13:41:56] <pinball_> but i am still on 2.3.3
[18-Sep-2009 13:41:58] <eidolon> my snmp experience is pretty limited.
[18-Sep-2009 13:42:05] <ckrough> the consensus has been that WMI is more reliable for windows hosts
[18-Sep-2009 13:42:08] <Brad_K> we're using Zenoss Enterprise 2.4.2, i think.
[18-Sep-2009 13:42:11] <pinball_> dev i have 2.4.5
[18-Sep-2009 13:42:18] <pinball_> where i am playing with WMI
\n[18-Sep-2009 13:42:18] <eidolon> thoguh, apparently my cohort\nhere has deleted, re-added, and remodelled this host, and we just got\n'used ytes' and 'free bytes' showing up
[18-Sep-2009 13:42:30] <metalklesk> one question
[18-Sep-2009 13:42:31] * eidolon loves sketchy results.
\n[18-Sep-2009 13:42:35] <Brad_K> one of my tasks is to figure out\nhow to start working on doing the upgrade to 2.4.5, next week.
[18-Sep-2009 13:42:54] <metalklesk> when i build a new command (settings -> commando -> add new command)
[18-Sep-2009 13:43:18] <metalklesk> that command is executed in server or client side ?
[18-Sep-2009 13:43:32] <metalklesk> for example, "rpm -qa"
\n[18-Sep-2009 13:43:38] <Brad_K> in theory, WMI is what you're\nsupposed to use for Windows. But check earlier archives of the logs of\nthis channel, because there were some major problems with WMI being\nreported by one of the Zenoss guys who was testing it with an upgraded\nversion of Zenoss.
[18-Sep-2009 13:44:08] <heckj> where can you find the logs for this channel?
[18-Sep-2009 13:44:18] <eidolon> :-/
[18-Sep-2009 13:44:19] * eidolon nods
[18-Sep-2009 13:44:32] <cgibbons> got any more context, brad? that would be news to me...
[18-Sep-2009 13:44:46] <eidolon> i'd like to see an official statement from zenoss stating which is suggested.
[18-Sep-2009 13:44:53] <eidolon> not that i don't trust you guys
[18-Sep-2009 13:44:58] <pinball_> haha
[18-Sep-2009 13:45:06]\n<eidolon> but i can't say "i'm going to change the mointoring on\n30 hosts because omsoen thought they saw something it he backscroll.
[18-Sep-2009 13:45:10] <pinball_> i just have had good luck with configuring both SNMP and WMI on windows boxes
[18-Sep-2009 13:45:13] <pinball_> and when i add the host
[18-Sep-2009 13:45:21] <pinball_> i don't use the add device link on the left side
\n[18-Sep-2009 13:45:24] <Brad_K> the official word is probably\nstill going to be to use WMI, because this was testing with a new\nversion of Zenoss.
[18-Sep-2009 13:45:27] <pinball_> i navigate to the device tree
[18-Sep-2009 13:45:38] <pinball_> devices->server->windows
[18-Sep-2009 13:45:42] <mrayzenoss> metalklesk: either
[18-Sep-2009 13:45:49] <pinball_> and in the drop down more add->device
[18-Sep-2009 13:45:54] <pinball_> and use snmp to discover
[18-Sep-2009 13:45:59] <ckrough> a lot of the SNMP/Windows problems I've seen reported were problems with snmp-informant
[18-Sep-2009 13:46:08] <ckrough> I dont monitor and windows hosts so not sure the specifics
[18-Sep-2009 13:46:14] <twm1010> i've got a box right now that probably need a reboot for informant to work
[18-Sep-2009 13:46:18] <twm1010> very odd.
[18-Sep-2009 13:46:22] <pinball_> we don't use informant
\n[18-Sep-2009 13:46:33] <twm1010> i consider switching to WMI, but\ni've not touched it in a while, and i'm unsure WHICH wmi zenpack to\nlook at
[18-Sep-2009 13:46:41] <ckrough> pinball_: ya. enterprise has a wmi zenpack or agent or something right?
[18-Sep-2009 13:46:42] <twm1010> also curious how the filesystems are monitoring via WMI
[18-Sep-2009 13:46:47] <pinball_> right
[18-Sep-2009 13:46:55] <pinball_> but there is a good zenpack
[18-Sep-2009 13:47:00] <pinball_> called WMIDevice
[18-Sep-2009 13:47:01]\n<Brad_K> unfortunately, all the zenoss community links that used\nto work are now broken, thanks to the new Jive system they're using.
[18-Sep-2009 13:47:03] <pinball_> in the community group
[18-Sep-2009 13:47:06] <pinball_> that works well
[18-Sep-2009 13:47:15] <metalklesk> mrayzenoss, i want to run "rpm -qa" command in the device i'm monitoring
[18-Sep-2009 13:47:33] <pinball_> ZenPacks.community.WMIDataSource community Egor Puzanov 1.1 Yes
[18-Sep-2009 13:47:41] <metalklesk> how teh command should be ? rpm -qa ? or something else ?
[18-Sep-2009 13:47:42] <pinball_> ZenPacks.community.WMIPerf_Windows community Egor Puzanov 1.0 Yes
\n[18-Sep-2009 13:48:33] <mrayzenoss> metalklesk: when you add the\nCommand data source, there's an option about using SSH, that determines\nwhether the command is local or remote
[18-Sep-2009\n13:48:48] <Brad_K> see Mark Hinkle's blog post at\nblogs/zenossblog/2009/09/03/winners-of-the-zenoss-summer-of-zenpacks-contest\nabout Egor Puzanov's WMI Data Source zenpack for Zenoss Core, which\nlooks like it provides the kind of stuff you want.
[18-Sep-2009 13:48:49] <ckrough> Brad_K: the community search at the top and search.zenoss.com might work for you
[18-Sep-2009 13:48:54] <metalklesk> aaahhh ok, thanks !!!
[18-Sep-2009 13:49:03] <ckrough> Brad_K: community searched has been working for me.
[18-Sep-2009 13:49:09] <Brad_K> i tried the community search. didn't turn up anything useful for "irc".
[18-Sep-2009 13:49:19] <ckrough> meh
\n[18-Sep-2009 13:49:31] <twm1010> mrayzenoss: i'm surprised you\nguys let Egor release that WMI datasource zenpack... what with WMI\nmonitoring a touted feature of enterprise
[18-Sep-2009 13:49:33] <mrayzenoss> Documentation->IRC
[18-Sep-2009 13:49:43] <mrayzenoss> community/documentation/irc
[18-Sep-2009 13:49:55] <mrayzenoss> twm1010: we don't block anything from the community
[18-Sep-2009 13:49:56] <cgibbons> Egor's is a long way from what Enterprise has (so far)
[18-Sep-2009 13:50:11] <twm1010> gotcha
[18-Sep-2009 13:50:34] <mrayzenoss> Predictive Thresholds is essentially the same as the Enterprise one
[18-Sep-2009 13:50:45] <mrayzenoss> (just not QA'd)
[18-Sep-2009 13:51:32] <mrayzenoss> we would get a bad reputation real fast if we tried to shut down Community people
\n[18-Sep-2009 13:52:10] <Brad_K> unfortunately, the irc archive at\ndocs/DOC-2466 doesn't include anything from\nSeptember, at least not yet.
[18-Sep-2009 13:52:22]\n<pinball_> one thing to note cgibbons is that Egor's zenpack\nactually works when collecting perf data, while the enterprise wmi\ntemplate didn't
[18-Sep-2009 13:52:27] <mrayzenoss> Brad_K: yeah, I usually get a monthly dump of docs
[18-Sep-2009 13:52:35] <cgibbons> that would also be news to me
[18-Sep-2009 13:52:39] <pinball_> at least on 2.4.5
\n[18-Sep-2009 13:52:51] <cgibbons> given most enteprrise customers\ndo actually collect perf data with the enteprise zenpack
[18-Sep-2009 13:52:58] <cgibbons> and i haven't gotten any case or tickets about that
[18-Sep-2009 13:53:02] <pinball_> we are on enterprise and in dev we had to switch over
\n[18-Sep-2009 13:53:30] <Brad_K> anyway, the problems with Egor's\nzenpack probably wouldn't have anything to do with how WMI would work\nunder Zenoss Enteroprise. D'oh!
[18-Sep-2009 13:53:47] <metalklesk> mrayzenoss, can i add a command just to be used in one monitored device ?
[18-Sep-2009 13:54:25] <mrayzenoss> metalklesk: yeah, you can add a custom template on just 1 device
[18-Sep-2009 13:54:36] <metalklesk> ok, thanks
[18-Sep-2009 13:54:52] <cgibbons> did you open up a case on that, pinball?
[18-Sep-2009 13:54:57] <pinball_> not yet
[18-Sep-2009 13:55:02] <pinball_> waiting on the new version
[18-Sep-2009 13:55:08] <pinball_> i am holding the update
[18-Sep-2009 13:55:12] <pinball_> from 2.3.3 to 2.4.5
[18-Sep-2009 13:55:17] <cgibbons> new version past 2.4.5?
[18-Sep-2009 13:55:26] <pinball_> and am going to wait for the next release
[18-Sep-2009 13:55:34] <pinball_> before proceeding with an update
[18-Sep-2009 13:55:54] <pinball_> i c 2.5 beta is out
\n[18-Sep-2009 13:55:55] <cgibbons> okay, can to share any details\nabout what happened? because right now we're in the final stages of bug\nfixes for 2.5.0 and there isn't anything i know about being broken in\n2.4.5 that would change in 2.5.0
[18-Sep-2009\n13:56:33] <pinball_> yeah the winperf / perf data was not being\npicked up on the two windows servers i have in dev
[18-Sep-2009 13:56:34] <pinball_> one vm
[18-Sep-2009 13:56:37] <mrayzenoss> cgibbons: pinball_ is on 2.3.3
[18-Sep-2009 13:56:38] <pinball_> and one dell 2850
\n[18-Sep-2009 13:56:43] <Brad_K> curiosly, the problems we were\nhaving with zencommand not working for the email_delivery_check\nsuddenly started working around midnight last night. As did the other\nhalf of a command that had been working before, but was recently\nbroken. Bleah.
[18-Sep-2009 13:56:43] <pinball_> dev. is 2.4.5
[18-Sep-2009 13:56:47] <mrayzenoss> ahh
[18-Sep-2009 13:56:54] <pinball_> prod is 2.3.3
[18-Sep-2009 13:56:57] <cgibbons> that of course won't be enough info
[18-Sep-2009 13:57:24] <pinball_> so CPU , free mem etc.. was not being collected
\n[18-Sep-2009 13:57:34] <Brad_K> the thing that annoys me about\nthis is that i had remodeled those devices many times over. ;-(
[18-Sep-2009 13:57:50] <pinball_> so i was looking around for others seeing the same thing
[18-Sep-2009 13:57:56] <pinball_> saw egors zenpacks
[18-Sep-2009 13:58:00] <pinball_> tried them and viola it worked
[18-Sep-2009 13:58:20] <pinball_> not sure if it was winperf related
\n[18-Sep-2009 13:58:40] <cgibbons> egor's zenpack collects\nperfdata via a WMI query, whereas the enterprise one does it directly\nvia PerfMon (which means the remote registry service), so it might have\nbeen something on those servers misconfigured
[18-Sep-2009 13:58:44] <cgibbons> but that's certainly been the case since 2.3.0
[18-Sep-2009 13:59:10] <cgibbons> eidolon: are you still around?
[18-Sep-2009 14:01:38] <twm1010> i wonder does remote registry respond faster than WMI?
[18-Sep-2009 14:01:52] <rmatte> twm1010: I'd imagine they'd be comparable
[18-Sep-2009 14:01:59] <cgibbons> there's less overhead with it yes
[18-Sep-2009 14:02:04] <Brad_K> and is it more reliable, both inherently and in the zenpack?
[18-Sep-2009 14:02:19] <twm1010> I remember using the perfmon zenpack which used a remotely executed perfmon query via wmiexec
[18-Sep-2009 14:02:29] <cgibbons> yeah that was the prior to 2.3
[18-Sep-2009 14:03:08] <Brad_K> and how stable is the remote registry vs. WMI on the remote Windows server side?
[18-Sep-2009 14:03:12]\n<cgibbons> i should really draw up an architectural diagram, but\nboth WMI and Remote Registry are layered on-top of Microsoft's Remote\nProcedure Call implementation, WMI is further on top of the DCOM layer\nand then is more complex inside because it translates string queries,\nand then does a local PerfMon data lookup, before returning the raw\ndata.
[18-Sep-2009 14:03:26] <rmatte> Brad_K: I think the question is what's not more stable than WMI?
[18-Sep-2009 14:03:50] <cgibbons> as of 2.3.x zenwinperf functions the same way that the PerfMon utility itself does
[18-Sep-2009 14:04:09]\n<mrayzenoss> ckrough: Brad_K: followup on the Austin training:\nAdmin training will start being offered remotely next month, Advanced\ntraining will be available in Austin and Annapolis
[18-Sep-2009 14:04:26] <twm1010> ah... so remote registry -> perfmon is simply the shortest path to the data.
[18-Sep-2009 14:04:29]\n<cgibbons> doing a WMI query to get the data is more convenient\nfor scripting users and admin because there's little parsing that's\nrequired
[18-Sep-2009 14:04:30] <Brad_K> mrayzenoss: cool!
[18-Sep-2009 14:04:32] <cgibbons> correcto
[18-Sep-2009 14:04:32] <rmatte> mrayzenoss: prices for remote training?
[18-Sep-2009 14:04:40] <mrayzenoss> rmatte: I dunno yet
[18-Sep-2009 14:04:55] <mrayzenoss> rmatte: I'd assume less, but I don't know
[18-Sep-2009 14:04:55] <rmatte> mrayzenoss: will there be an outline available of what's covered in the courses?
[18-Sep-2009 14:05:00] * ckrough vi does funny thinks when you ':mrayzenoss'...
[18-Sep-2009 14:05:04] <Brad_K> mrayzenoss: please let me know, because that may have a serious impact on my ability to get any training.
[18-Sep-2009 14:05:05] <ckrough> mrayzenoss: thanks
[18-Sep-2009 14:05:22] <mrayzenoss> Brad_K: yeah, I mentioned that when I was asking
[18-Sep-2009 14:05:35] <Brad_K> mrayzenoss: much appreciated. Thanks!
[18-Sep-2009 14:05:46] <twm1010> so... all this WMI work is nice, but probably causes lots of collector overhead... eh?
[18-Sep-2009 14:06:03] <cgibbons> not really a lot, no, but it's not as fast as perfmon data
[18-Sep-2009 14:06:06] <rmatte> E492: Not an editor command: mrayzenoss
[18-Sep-2009 14:06:09] <ckrough> rmatte: message/39572
[18-Sep-2009 14:06:37]\n<cgibbons> if you were trying to collect 1000 performance metrics\nper device every 10 seconds, then you'd probably not want to be doing\nit via WMI queries
[18-Sep-2009 14:06:47] <mrayzenoss> rmatte: sounds like you need to upgrade to Emacs
[18-Sep-2009 14:06:56] <mrayzenoss> M-x mrayzenoss
[18-Sep-2009 14:07:09] <ckrough> heh, actually I was in 'less'. talk about being disoriented.
[18-Sep-2009 14:07:24]\n<rmatte> I'm actually a nano guy myself (used to use pico when I\nwas first learning FreeBSD, so I know all of the shortcuts)
[18-Sep-2009 14:07:25] <Brad_K> no, talk is a different program from either of those.
[18-Sep-2009 14:07:31]\n<cgibbons> I know the network overhead is higher, but I haven't\nmeasured it, but for comparison a single query of data on this windows\nserver 2008 i'm testing right now is 144,542 bytes
[18-Sep-2009 14:07:33] * Brad_K ;-)
[18-Sep-2009 14:07:40] <ckrough> heh
[18-Sep-2009 14:08:44] <Brad_K> cgibbons: yowch! 144KB? for a single query?!?
[18-Sep-2009 14:09:04] <cgibbons> for all of the counters available, yeah
[18-Sep-2009 14:09:16] <rmatte> Brad_K: Welcome to Microsoft country
[18-Sep-2009 14:09:19] <twm1010> all available counters?
[18-Sep-2009 14:09:49] <twm1010> thats a ton of data
[18-Sep-2009 14:10:01] * Brad_K c|n>k
[18-Sep-2009 14:11:38] <metalklesk> mrayzenoss, i've added a new data source for a command
[18-Sep-2009 14:11:50] <metalklesk> what's next to get the output ?
[18-Sep-2009 14:11:55] <Brad_K> okay, gotta go to lunch. bbl
[18-Sep-2009 14:12:03] <mrayzenoss> use the 'test' to see how it's working
[18-Sep-2009 14:12:25] <metalklesk> ther i should grab the ip right ?
[18-Sep-2009 14:12:37] <mrayzenoss> or the device name
[18-Sep-2009 14:12:53] <mrayzenoss> the device name will grab the user/pass from the zProperties
[18-Sep-2009 14:13:04] <metalklesk> ok
[18-Sep-2009 14:13:41] <metalklesk> it works
[18-Sep-2009 14:14:06] <metalklesk> now what is next to use this command ?
[18-Sep-2009 14:15:16] <mrayzenoss> add your data points, add your threshold and graphs
[18-Sep-2009 14:15:19] <mrayzenoss> then bind it
[18-Sep-2009 14:15:58] <mrayzenoss> docs/DOC-3025
[18-Sep-2009 14:17:14] <metalklesk> thanks
[18-Sep-2009 14:20:19] <twm1010> reviewing king crab finally
\n[18-Sep-2009 14:24:33] <metalklesk> mrayzenoss, once you have the\ncommand "rpm -qa --qf\n"%{NAME}-%{VERSION}-%{RELEASE}|%{INSTALLTIME}|%{SUMMARY}\\n"" added as a\ndatasource in the performance template, how can i grab the outputed\ndata into the software list in the device ?
[18-Sep-2009 14:24:45] <twm1010> severity icons much easier on the eyes than the severity colors
[18-Sep-2009 14:26:04] <mrayzenoss> metalklesk: ahh, that's the wrong approach
[18-Sep-2009 14:26:12] <metalklesk>
[18-Sep-2009 14:26:22] <mrayzenoss> metalklesk: templates and data sources are for performance monitoring, the 'Perf' tab
[18-Sep-2009 14:26:31] <metalklesk> aaaaahh ok
[18-Sep-2009 14:26:33] <twm1010> shouldn't isntalled packages be available via SNMP?
[18-Sep-2009 14:26:42] <mrayzenoss> twm1010: if configured properly
\n[18-Sep-2009 14:26:57] <mrayzenoss> getting stuff onto the\nSoftware tab via SSH requires looking at how the other SSH Linux\nZenPacks do it
[18-Sep-2009 14:27:18] <mrayzenoss> so Fedora might be a good start docs/DOC-3382
[18-Sep-2009 14:27:24] <mrayzenoss> since it probably uses RPM too
[18-Sep-2009 14:27:27] <mrayzenoss> or OpenSUSE
[18-Sep-2009 14:27:38] <mrayzenoss> docs/DOC-3478
\n[18-Sep-2009 14:27:57] <metalklesk> the thing is that i was\ntrying opensuse edited version of zenpack but has some "problems"
[18-Sep-2009 14:28:15] <metalklesk> first, it does not show all the graphs
[18-Sep-2009 14:29:04] <metalklesk> and if you put out ram of the zignorefilesystem, it does not show used, total, free ram in OS tab
\n[18-Sep-2009 14:29:43] <metalklesk> in the other hand, using just\nsnmp monitoring, you can get everything you want except the list of\ninstalled software
[18-Sep-2009 14:30:18]\n<metalklesk> so what i'm trying to do es using snmp for all and\nbuild a command to get the installed software
[18-Sep-2009 14:30:22] <metalklesk> using ssh
[18-Sep-2009 14:31:09] <mrayzenoss> hmm... you could do it with SNMP thought
\n[18-Sep-2009 14:31:14] <mrayzenoss> take a look at\nblogs/zenossblog/2009/02/18/tip-of-the-month-snmp-software-inventory-for-debian-and-ubuntu-machines
[18-Sep-2009 14:31:57] <metalklesk> can i ??
[18-Sep-2009 14:32:00] <mrayzenoss> yeah
[18-Sep-2009 14:32:35] <mrayzenoss> wow, the formatting got totally mangled on that blog post
\n[18-Sep-2009 14:39:10] <metalklesk> one litte approach is\ncreating a simple user command in the device i want, so clicking the\nnew command in Run commands -> my command, works
\n[18-Sep-2009 14:39:46] <metalklesk> but it does not show the\noutput in software tab, but at least it shows me the software list
[18-Sep-2009 14:39:57] <metalklesk>
[18-Sep-2009 14:40:40] <metalklesk> it si really cool because i can do whatever i want remotely
[18-Sep-2009 14:40:47] <metalklesk> *it is
[18-Sep-2009 14:56:03]\n<perr0> I just installed zenoss-2.4.5 from source and get errors\nwhen starting zenoss can anyone have a look possibly point me to where\nI can begin to troubleshoot?
[18-Sep-2009 14:56:05] <perr0> http://pastebin.org/18756
[18-Sep-2009 14:59:12] <cgibbons> do you have the net-snmp package installed on that box?
[18-Sep-2009 15:00:33] <perr0> ahh probably not
[18-Sep-2009 15:00:54] <twm1010> install both net-snmp and net-snmp-utils
[18-Sep-2009 15:00:56] <twm1010>
[18-Sep-2009 15:01:08] <perr0> nice, thanks guys
[18-Sep-2009 15:08:14] <eidolon> hrm, anyone seen rrd charts fail in mid-work? event log is showing: 2009-09-18T16:06:50 ERROR RenderServer Failed to generate a graph
[18-Sep-2009 15:08:14] <eidolon> Traceback (most recent call last):
[18-Sep-2009 15:08:14] <eidolon> File "/opt/zenoss/Products/ZenRRD/RenderServer.py", line 190, in render
[18-Sep-2009 15:08:14] <eidolon> rrdtool.graph(*gopts)
[18-Sep-2009 15:11:40] <perr0> -
\n[18-Sep-2009 15:12:46] <rmatte> wow, one of our VMs was\naccidentally set to use 255GB of RAM, sucked the life out of every\nother VM on the blade
[18-Sep-2009 15:12:54] <rmatte> Fridays are never boring
[18-Sep-2009 15:12:59] <twm1010> That's awesome.
[18-Sep-2009 15:13:11] <twm1010> somebody changed an MB to a GB huh?
[18-Sep-2009 15:13:19] <rmatte> nah, we found out how it happened...
[18-Sep-2009 15:13:37] <straterra> intern
[18-Sep-2009 15:13:47] <twm1010> i've been to known to set a directory quota to 50MB instead of 50GB before
[18-Sep-2009 15:14:08]\n<rmatte> there's an input field, so they typed in the correct\namount which was 4gb, but then they clicked on a gray area of the\nwindow, far right of this vertical scale of how much memory is being\nset, he clicked somewhere near the top of the window by accident so it\nset it to that
[18-Sep-2009 15:14:46] <rmatte>\nIt's stupidly designed, I can understand if you clicked on the scale\nitself and it adjusted it, but not when you click far to the right of it
\n[18-Sep-2009 15:18:36] <Brad_K> hey, 255GB of RAM. that sounds\npretty cool! the biggest i've ever built was Sun X4440s with four\nquad-core CPUs and 64GB of RAM....
[18-Sep-2009\n15:19:51] <rmatte> well, we only actually have 32GB of RAM in the\nblade, which is also stupid that VMWare would even let us set it that\nhigh
[18-Sep-2009 15:20:53] <rmatte> I'm sitting here watching as my king crab test box takes 15 minutes to shutdown...
[18-Sep-2009 15:21:00] <rmatte> since it's on that blade
[18-Sep-2009 15:21:01] <rmatte> lol
[18-Sep-2009 15:21:18] <ckrough> rmatte: Do you use resource pools?
[18-Sep-2009 15:21:31] <ckrough> assuming you are referring to vmware
[18-Sep-2009 15:22:27] <rmatte> ckrough: we do, but this VM was created outside of the resource pools
[18-Sep-2009 15:22:48] <rmatte> so it just choked everything out
[18-Sep-2009 15:28:15] <rmatte> rebooting all of our dev boxes now...
[18-Sep-2009 15:30:24] <Brad_K> ouch....
[18-Sep-2009 15:43:13] <ckrough> have a good weekend. Im outski
[18-Sep-2009 15:43:53] <Brad_K> cya!
\n[18-Sep-2009 15:46:18] <eidolon> so is there any way to -really-\nwatch whether zenoss is using WMI to model / look at a server, and to\nforce an update?
[18-Sep-2009 15:46:48]\n<eidolon> the problem is we're diong stuff passively on a windows\nhost (like filling up a disk), and zenoss doesn't show the update. we\nwait 5 minutes, nothing. this constnat waiting and sort of 'well, it\nmight work, wait around and see' - is really frustrating.
[18-Sep-2009 15:54:03]\n<Brad_K> out of curiosity, does anyone know how to see the Zenoss\nequivalent of the Nagios event queue, so that we can tell when the next\nscheduled event is going to occur for a given template?
\n[18-Sep-2009 15:54:59] <Brad_K> actually, the template is bound\nto a device, and what we want to know is when the next SSH check is\ngoing to happen for that device.
[18-Sep-2009 17:47:45] <daMaestro> okay, i'm done with the nginx zenpack
[18-Sep-2009 17:47:50] <daMaestro> anyone know where/how to upload it?
[18-Sep-2009 17:58:58] <USENIX_UNIX> Can't wait to get my grubby hands on it
\n[18-Sep-2009 19:37:55] <Brad_K> okay, this is seriously screwed\nup. We have several Ironport e-mail security appliances here. When\nthey were modeled, they came up as "Unknown" manufacturer, even though\nZenoss does seem to know about Ironport.
[18-Sep-2009\n19:38:24] <Brad_K> I've been trying to fix these. I added the\nC300 and C600 devices to the list, and classified some older machines\ncorrectly.
[18-Sep-2009 19:38:57] <Brad_K> but\nwhen i tried to add the C660 to the list, and classify some newer\nmachines, one worked fine although C660 never showed up on the official\nlist of devices from the pull-down.
[18-Sep-2009\n19:39:47] <Brad_K> but another device I was working on is totally\nwigged out. Now there is no vendor shown at all, no device name shown\nat all, and no software vendor shown at all. I can't even go back and\nre-set them to "Ironport" and "C600", or any other value from any of\nthe pull-downs.
[18-Sep-2009 19:40:59] <Brad_K>\ni could probably reset things using zendmd, as another poster in one of\nthe forums recently did. but i don't know zendmd, and i don't see how\nit should be necessary for this sort of thing.
[18-Sep-2009 19:41:13] * Brad_K starts humming the theme from "Twilight Zone"...
[18-Sep-2009 19:41:38] <Brad_K> Anyway, i'll worry about filing a ticket on this next week.
[18-Sep-2009 19:41:42] <Brad_K> i
[18-Sep-2009 19:41:50] <Brad_K> i'm going home. g'nite!
[19-Sep-2009 00:00:46] [disconnected at Sat Sep 19 00:00:46 2009]
[19-Sep-2009 00:00:46] [connected at Sat Sep 19 00:00:46 2009]
[19-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[19-Sep-2009 09:13:44] <Apocalipse> hello people
[19-Sep-2009 12:06:03] timburke_ is now known as timburke
[20-Sep-2009 00:00:46] [disconnected at Sun Sep 20 00:00:46 2009]
[20-Sep-2009 00:00:46] [connected at Sun Sep 20 00:00:46 2009]
[20-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[21-Sep-2009 00:00:46] [disconnected at Mon Sep 21 00:00:46 2009]
[21-Sep-2009 00:00:46] [connected at Mon Sep 21 00:00:46 2009]
[21-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[21-Sep-2009 04:31:23] <Daan> o/ Good morning...
[21-Sep-2009 04:32:32] Daan is now known as danuvius
[21-Sep-2009 04:32:42] danuvius is now known as daan
[21-Sep-2009 04:34:58] daan is now known as danuvius
[21-Sep-2009 04:35:16] danuvius is now known as daan
[21-Sep-2009 04:35:44] daan is now known as danuvius
[21-Sep-2009 04:36:52] <danuvius> o/ Good morning...
[21-Sep-2009 05:46:09] <Dieterbe_> hi
[21-Sep-2009 06:37:58] <Troubadix09> hi all
[21-Sep-2009 06:45:24] <danuvius> hi
[21-Sep-2009 07:44:18] <danuvius> anyone here that is familiar with the zenmailtx zenpack?
\n[21-Sep-2009 10:24:38] <drtns> morning all I need a little regex\n101 help... we run a custom apache with the MPM-ITK module.. thus our\nprocess is called httpd.itk rather then just httpd so in zenoss I get a\nbunch of boxes that say that httpd.itk is not running. I need to mod\nmy /process/apache/httpd regex to allow either a process named httpd\nfor the stock boxes OR httpd.itk for the boxes running that. the\ncurrent regex is ^.*\\/*httpd can anyone help me out
[21-Sep-2009 10:26:55] <twm1010> play with this
[21-Sep-2009 10:26:55] <twm1010> http://gskinner.com/RegExr/
[21-Sep-2009 10:27:38] <twm1010> could you not put a * on the end?
[21-Sep-2009 10:28:08] <drtns> cool thanks for that twm1010 I tried that stock ^.*\\/*httpd*
[21-Sep-2009 10:28:15] <drtns> but it doesn't seem to do the trick
[21-Sep-2009 10:32:38] <danuvius> someone here using the zenmailtx zenpack?
[21-Sep-2009 10:33:00] <drtns> since '.' dot in a regex means any character I assume I escape the dot \\. to mean a literal dot?
[21-Sep-2009 10:35:28] <danuvius> drtns: ^.*\\/*httpd\\.itk
\n[21-Sep-2009 10:36:18] <drtns> danuvius but wouldn't that then\nkill any servers not runnin itk? think maybe I need something more\nlike ^.*\\/*(httpd|httpd\\.itk)
[21-Sep-2009 10:36:35] <drtns> to mach either the standard httpd or httpd.itk?
[21-Sep-2009 10:37:52] <danuvius> drtns: hmmm ^.*\\/*httpd(\\.itk)+
[21-Sep-2009 10:38:49] <danuvius> sorry mean ^.*\\/*httpd(\\.itk)?
[21-Sep-2009 10:39:45] <drtns> danuvius the 2nd could possibly work
[21-Sep-2009 10:40:11] <drtns> think mine might work to just using the "or" just written out longer..
[21-Sep-2009 10:40:35] <danuvius> true yours would have worked to...
[21-Sep-2009 11:06:10] <rmatte> drtns: why would you just add a completely separate process to Processes called httpd.itk ?
[21-Sep-2009 11:06:18] <rmatte> why wouldn't you
[21-Sep-2009 11:06:19] <rmatte> rather
[21-Sep-2009 11:07:17] <rmatte> instead of trying to do 2 in 1
[21-Sep-2009 11:07:25]\n<drtns> rmatte just as valid an argument to say why bother when\nyou can just "or" the regex and keep it all happy under a single process
[21-Sep-2009 11:07:48] <rmatte> drtns: because they aren't technically even the same process
[21-Sep-2009 11:08:14]\n<rmatte> drtns: When you model a device, Zenoss adds the\nprocesses based on what's enabled to be monitored in the processes\nlist...
[21-Sep-2009 11:08:39] <rmatte> if you\nhave 1 process with a regex of httpd or httpd.itk it's always going to\nmap that process to the processes list
[21-Sep-2009 11:08:57]\n<rmatte> if you have 1 process that's called httpd and 1 process\nthat's called httpd.itk, it'll pick which one it adds
[21-Sep-2009 11:09:41] <rmatte> plus it's a heck of a lot simpler
\n[21-Sep-2009 11:10:19] <rmatte> no point to make things difficult\nfor yourself over adding 1 process entry to the processes list
[21-Sep-2009 11:18:25] <Brad_K> so, zencommand is acting weirder and weirder....
\n[21-Sep-2009 11:18:43] <drtns> rmatte so I add a new service to\n/process/apache for httpd.itk and it's listing count as 0 how does\nthat get populated? can I force a refresh?
\n[21-Sep-2009 11:18:48] <Brad_K> things mysteriously start working\nwhen they weren't, and mysteriously stop working when they were.
[21-Sep-2009 11:19:23] <rmatte> drtns: you need to remodel devices for it to pick it up
[21-Sep-2009 11:19:50] <rmatte> Brad_K: you may want to adjust your zencommand timeout values in the collector settings
[21-Sep-2009 11:19:51] <drtns> ok I'll give that a whirl
\n[21-Sep-2009 11:21:35] <rmatte> actually, there aren't any\ntimeout settings for zencommand in there, they must be under\nzProperties then
[21-Sep-2009 11:22:27] <rmatte> zCommandCommandTimeout and zCommandCycleTime
\n[21-Sep-2009 11:23:30] <Brad_K> right now, i've got 15.0 and 60\nfor those values, but in the definition of the command itself, the\ncycle time is set to 300. does the zproperties over-ride the\ncommand-specific cycle time?
[21-Sep-2009 11:27:22] <rmatte> not sure
[21-Sep-2009 11:27:26] <rmatte> doubt it
[21-Sep-2009 11:27:41] <rmatte> so there's no indication in the log files of any issues?
[21-Sep-2009 11:27:55] <rmatte> I'd expect there to be at least some errors if it's not working properly
\n[21-Sep-2009 11:28:25] <Brad_K> and for a job that should really\nbe run every 300 seconds, and you want it to keep retrying/waiting\nright up to the next cycle time, what kinds of values would be\nreasonable to expect?
[21-Sep-2009 11:29:13]\n<Brad_K> we did see some timeouts in the web ui on the hub, but\ni've never seen any associated errors in the zencommand.log from the\ncollector.
[21-Sep-2009 11:33:13] <Brad_K> the\nonly relevant thing i've seen in the zencommand.log files from the\ncollector have been a failure to log certain information, but no errors\nassociated with that.
[21-Sep-2009 11:34:18] <rmatte> the default values are generally reasonable
[21-Sep-2009 11:34:24] <rmatte> what types of commands are failing?
\n[21-Sep-2009 11:35:10] <Brad_K> the same Nagios-derived\ncheck_ping and check_email_delivery that I've been fighting with all\nthis time.
[21-Sep-2009 11:35:19] <rmatte> ah
[21-Sep-2009 11:35:39] <rmatte> did you ever modify them to remove the \\n ?
[21-Sep-2009 11:36:02]\n<Brad_K> oh, and btw -- multiline output doesn't seem to be the\nissue. i hacked check_email_delivery to remove the \\n at the end of\nthe line, while check_ping still issues the \\n. Right now, check_ping\nis working, and check_email_delivery is not.
[21-Sep-2009 11:36:20] <rmatte> I see, fair enough
[21-Sep-2009 11:36:26] <Brad_K> the case of the variables being stored did seem to make a difference.
\n[21-Sep-2009 11:36:58] <rmatte> well, my suggestion would be to\nmodify the script to output some stuff to a file each time they run
[21-Sep-2009 11:37:08] <rmatte> and hopefull you can figure out what the issue is that way
\n[21-Sep-2009 11:37:37] <rmatte> and by stuff I mean: output the\ndata being passed to the command, the output, along with a timestamp if\npossible
[21-Sep-2009 11:38:02] <Brad_K> the\nproblem is not the scripts. when i run them directly from the\ncommand-line, they work perfectly. when i run them manually via\nbin/zencommand, they work perfectly. it's only when i try to integrate\nthem into being auto-run by Zenoss that they sometimes work, and\nsometimes fail.
[21-Sep-2009 11:38:48] <rmatte>\njust because they are running fine manually doesn't mean that they\naren't running fine automatically
[21-Sep-2009 11:39:07]\n<rmatte> it would just be a starting point so that you could get\ntimestamped data, to see how often the commands are failing and what's\nbeing passed to them
[21-Sep-2009 11:39:18] <rmatte> maybe some data is getting botched or something, who knows?
\n[21-Sep-2009 11:40:19] <rmatte> it's the best starting point for\ntroubleshooting this seeing as your logs are apparently useless
\n[21-Sep-2009 11:40:23] <Brad_K> you can look at the output shown\nin zencommand.log (with log severity set to 10), and see that the\noutput is exactly as expected. the problem is that zencommand\nsometimes does not process this output as it should.
\n[21-Sep-2009 11:41:26] <rmatte> well, I've personally never seen\nthat behaviour, just from scripting experience, adding some debug lines\nin to the scripts may help
[21-Sep-2009 11:41:56] <rmatte> the timestamps will probably be the most valuable thing
\n[21-Sep-2009 11:42:50] <Brad_K> i can try that, but i've been\nhacking scripts for over twenty years, and i imagine you probably have\nabout the same amount of experience. my experience tells me that this\nisn't likely to tell us much, but it's not that hard to do. it's only\nPerl after all -- it's not like i have to write any Python code or\nanything. ;-)
[21-Sep-2009 11:44:06] <Brad_K>\noh, and btw -- outside of work, i also happen to be one of the\npostmasters for python.org. i'm just not a Python programmer -- ask\nBarry Warsaw, if you don't believe me.
[21-Sep-2009 12:06:34] <rmatte> lol
[21-Sep-2009 12:06:39] <rmatte> all good
[21-Sep-2009 12:07:08]\n<rmatte> no, I have a lot of experience for my age when it comes\nto this stuff, but I'm only 24 years old, so you have me beat there
[21-Sep-2009 12:07:28]\n<rmatte> I started playing around with FreeBSD when I was 14\nthough, so I've had quite a bit of exposure to this stuff for my age
[21-Sep-2009 12:07:29] <rmatte>
[21-Sep-2009 12:08:33]\n<rmatte> I've been learning python coding because I manage Zenoss\nservers, I'm getting pretty decent at it though, I'm usually pretty\ngood at picking up languages
[21-Sep-2009 12:09:33]\n<Brad_K> well, the command is mysteriously working again, but my\nhacks to write the data to a temporary file are not yet working.
\n[21-Sep-2009 12:10:18] * Brad_K hangs his head when he realizes he\ngraduated high school and has been on the Internet longer than rmatte\nhas been alive
[21-Sep-2009 12:11:44] <rmatte> hehe
[21-Sep-2009 12:12:25] <rmatte> well, I just found yet another king crab bug
[21-Sep-2009 12:13:26] <twm1010> Brad_K: Don't feel bad, rmatte is really a machine.
[21-Sep-2009 12:13:46] <Brad_K> we're not even runnning 2.4.5, much less king crab. blech.
\n[21-Sep-2009 12:14:24] <twm1010> I've found that being on the\nbleeding edge with Zenoss is a good way to learn to how to perform\nbackups and restores
\n[21-Sep-2009 12:14:36] * Brad_K remembers that 24 years ago was about\nthe time that William Gibson published books like Mona Lisa Overdrive\nand wrote about Wintermute for the first time....
[21-Sep-2009 12:15:08] <twm1010> My approach this go around has been to install it, and leave it alone and let it do it's job.
\n[21-Sep-2009 12:15:26] <twm1010> Maximum performance seems to\ncome from watching the debug messages and getting rid of what ever\ntheir cause is
[21-Sep-2009 12:16:29] <Brad_K>\nthat's where we want to get. we don't really want to be debugging any\ncode, to be honest. after all, this is going to be a production system\n& network monitor for mission-critical systems for the entire\nUniversity, and ~70,000 LAN customers here in Austin.
[21-Sep-2009 12:17:25] <twm1010> well, i of course meant debug events
[21-Sep-2009 12:17:29] <twm1010> not debug logs
\n[21-Sep-2009 12:18:29] <twm1010> bad OID events and performance\ntemplates looking for stuff that isn't there seems to make MySQL shit a\nbrick
[21-Sep-2009 12:18:41] <twm1010> if you get that under control, its like pushing the turbo button
[21-Sep-2009 12:19:05] <twm1010> i still need to learn how to trim down syslog chatter
[21-Sep-2009 12:19:10] <twm1010> i'm getting too much information right now
[21-Sep-2009 12:22:36] <rmatte> Brad_K: we're running 2.4.5 prod (unfortunately) and I have a King Crab test box setup
[21-Sep-2009 12:23:08] <rmatte> 2.4.5 is actually quite good, if it weren't for: http://dev.zenoss.com/trac/ticket/5473
[21-Sep-2009 12:23:14] <rmatte> that is driving me absolutely up the wall
\n[21-Sep-2009 12:24:47] <Brad_K> i've seen that in 2.4.2, as\nwell. We have Ironport e-mail security devices, and I added C300 and\nC600 to the list of known devices. Now it won't let me add C660, and\nfor some machines when I set their HW product, that doesn't stick. But\nfor others, it does, but only for some variables.
[21-Sep-2009 12:24:55] <rmatte> twm1010: syslog chatter is easy...
\n[21-Sep-2009 12:26:10] <rmatte> twm1010: just create a manual\nmapping under whatever class (I created a /Suppressed class), make the\neventClassKey "defaultmapping" then specify a regex for what you want\nto block, go to the zProperties for the event mapping and select drop\nas the action
[21-Sep-2009 12:26:13] <rmatte> and voila, you're done
[21-Sep-2009 12:26:53] <rmatte> For the regex you just need to put a part of the message
[21-Sep-2009 12:27:08] <rmatte> I had messages coming in like: "(ip) vrf/dest_addr=7813781"
[21-Sep-2009 12:27:14] <rmatte> so I set my regex to: (ip) vrf/dest_addr=
[21-Sep-2009 12:27:32] <rmatte> I also had syslog messages coming in completely blank
[21-Sep-2009 12:27:37] <rmatte> so I used ^$ as the regex
[21-Sep-2009 12:28:16] <rmatte> g'day Matt
[21-Sep-2009 12:28:29] <Brad_K> mrayzenoss1: hey!
\n[21-Sep-2009 12:29:27] <Brad_K> so, shall we re-open our contest\nto see who the oldest person is on this channel, and by how much?
[21-Sep-2009 12:29:50] <rmatte> I'm definitely the youngest
[21-Sep-2009 12:29:52] <rmatte> so I lose
[21-Sep-2009 12:30:34] <Brad_K> mrayzenoss1: how about you?
[21-Sep-2009 12:31:33] <rmatte> he's so old that he doesn't like to talk about it
[21-Sep-2009 12:31:44] <rmatte>
\n[21-Sep-2009 12:32:35] <Brad_K> okay, so i'm 43. i'm old enough\nto be your dad. my beard isn't completely grey, but that's partly\nbecause the grey hairs mysteriously disappear in the morning.
[21-Sep-2009 12:33:15] <mrayzenoss1> 35
[21-Sep-2009 12:33:59] <Brad_K> twm1010: can you beat 43?
[21-Sep-2009 12:35:04] <rmatte> my dad is quite a bit older than you lol
[21-Sep-2009 12:35:29] <Brad_K> yeah, well my dad is older than your dad! :-P
[21-Sep-2009 12:35:35] <rmatte>
[21-Sep-2009 12:35:45] <rmatte> My dad could beat up your dad
[21-Sep-2009 12:35:51] <rmatte> lol
[21-Sep-2009 12:36:10] <rmatte> takes me back to kindergarten
\n[21-Sep-2009 12:36:41] <Brad_K> probably, but then my dad would\ncall in his brother who used to be in the Hells Angels, and is now a\ntruck driver -- and who has lost most of his teeth in bar fights.
[21-Sep-2009 12:36:49] <rmatte> ok, so time to log another bug
[21-Sep-2009 12:36:56] <Brad_K> heh
[21-Sep-2009 12:37:05] <rmatte> actually, I don't have my trac password here, bah, I'll have to do it tomorrow
[21-Sep-2009 12:37:07] <twm1010> I'm 27
\n[21-Sep-2009 12:37:35] <Brad_K> rmatte: if you can get them to\nfix these bugs in 2.45 and 2.4.2 as well, that would be great.
\n[21-Sep-2009 12:37:57] <rmatte> Brad_K: well, this particular bug\nrelates to the new event console, so it doesn't apply to those versions
\n[21-Sep-2009 12:38:22] <Brad_K> twm1010: so, not much older than\nrmatte. interesting. i wonder what it is about Python-derived\nprograms that seems to draw the younger crowd....
\n[21-Sep-2009 12:38:26] <rmatte> I've been trying to push to have\nthat hardware bug fixed and then have the fix backported to 2.4.5 but I\nhighly doubt that will happen
[21-Sep-2009 12:38:35] <rmatte> especially since King Crab is due out in about a month
[21-Sep-2009 12:38:51] <rmatte> they haven't even fixed that hardware bug in King Crab yet
[21-Sep-2009 12:39:06] <Brad_K> well, we are an Enterprise customer. maybe we can also help push on that one?
[21-Sep-2009 12:39:07] <twm1010> which hardware bug??
[21-Sep-2009 12:39:23] <Brad_K> http://dev.zenoss.com/trac/ticket/5473
[21-Sep-2009 12:39:26] mrayzenoss1 is now known as mrayzenoss
[21-Sep-2009 12:39:46] <rmatte> wb mrayzenoss
[21-Sep-2009 12:39:51] <twm1010> ah... that... right, the reason i stayed at 2.4.2
[21-Sep-2009 12:40:02] <mrayzenoss> yeah, new install of the OS, lost my IRC password
[21-Sep-2009 12:40:09] <Brad_K> actually, no -- it's broken in 2.4.2 as well.
\n[21-Sep-2009 12:40:12] <rmatte> mrayzenoss: any word if\nhttp://dev.zenoss.com/trac/ticket/5473 is getting fixed for King Crab\nrelease?
[21-Sep-2009 12:40:34] <Brad_K> we've seen the bug here, too.
[21-Sep-2009 12:40:43] <rmatte> lots of people are seeing the bug
[21-Sep-2009 12:40:59] <rmatte> in pretty much every 2.4.x version, except that it's more noticeable in 2.3.3 and onward
[21-Sep-2009 12:41:08] <rmatte> erm 2.4.3
[21-Sep-2009 12:41:09] <rmatte> rather
[21-Sep-2009 12:41:11] <mrayzenoss> yeah, I think someone was working on it
[21-Sep-2009 12:41:28] <rmatte> cool
[21-Sep-2009 12:41:37] <twm1010> I just set a HW Product on mine?
[21-Sep-2009 12:41:49] <Brad_K> sometimes it works, sometimes it doesn't.
[21-Sep-2009 12:41:52] <rmatte> twm1010: works for certain ones
[21-Sep-2009 12:41:53] <mrayzenoss> I like this ticket: http://dev.zenoss.com/trac/ticket/5568
[21-Sep-2009 12:42:07] <rmatte> twm1010: same with software product details, some work some don't
[21-Sep-2009 12:42:36] <Brad_K> what is the ZenAWS zenpack?
[21-Sep-2009 12:42:55] <mrayzenoss> something new
[21-Sep-2009 12:42:56] <rmatte> cool, new features
[21-Sep-2009 12:43:04] <Brad_K> cool.
\n[21-Sep-2009 12:43:11] <rmatte> though new features aren't nearly\nas cool when there are bugs preventing you from using old features
[21-Sep-2009 12:43:14] <mrayzenoss> it'll be in the beta drop next week
[21-Sep-2009 12:43:27] <Brad_K> yeah, there's definitely bugs i want fixed.
[21-Sep-2009 12:43:29] <mrayzenoss> rmatte: you're killing my moment
[21-Sep-2009 12:43:39] <rmatte> mrayzenoss: sorry, my bad
[21-Sep-2009 12:43:50] <mrayzenoss> yeah, new features means new bugs
[21-Sep-2009 12:43:52] <Brad_K> like ticket # 8722.
[21-Sep-2009 12:44:18] <mrayzenoss> not sure we have one that high... support case I assume?
\n[21-Sep-2009 12:44:29] <rmatte> as long as 5473 gets fixed, I can\nlive with the other bugs for a while, but that's the one that people\nare breathing down my neck about
[21-Sep-2009 12:44:55] <rmatte> yeh, I highly doubt there's a ticket 8722
\n[21-Sep-2009 12:44:56] <Brad_K> yes, support case. but frankly,\ni've gotten more help in this channel than through customer service --\nespecially from you and rmatte.
[21-Sep-2009 12:44:57] <rmatte>
[21-Sep-2009 12:45:23] <rmatte> Brad_K: cool, just pay us directly and we'll provide all the support you want
[21-Sep-2009 12:45:33] <rmatte> j/k
\n[21-Sep-2009 12:45:48] <Brad_K> if you could give us a written\ncontract that we could sign, we might actually think about that.
[21-Sep-2009 12:46:16] <rmatte> I'd have to quit my job to be able to do something like that lol
[21-Sep-2009 12:46:22] <rmatte> which I've no intention of doing
[21-Sep-2009 12:46:32] <rmatte> but if I ever get canned I may consider trying it out
[21-Sep-2009 12:47:04]\n<Brad_K> i definitely prefer an interactive debugging process, as\nopposed to sending an e-mail to customer service through the captive\nSugarCRM system, then wait hours or days for a response, get a\nsuggestion, and then go through that lengthy process all over again.
\n[21-Sep-2009 12:49:18] <Brad_K> i mean, we're paying money for\nthis stuff. the commercial support should be better and faster than\nthe community support, right?
[21-Sep-2009 12:49:41] <rmatte> Brad_K: in theory, yes
[21-Sep-2009 12:50:20] <rmatte> but the community support is a unique advantage of an open-source product
\n[21-Sep-2009 12:50:36] <rmatte> so you might as well take\nadvantage of it regardless as to whether or not you are paying for\nsupport
[21-Sep-2009 12:50:47] <Brad_K> i mean,\nat least with FreeBSd, the community support is from the same folks who\nwould be providing the commercial support, and it's your choice as to\nwhether or not you want to give money to PHK to have him put something\nat the top of the list, or if you want to wait for him to get around to\ndoing something about it whenever he's got time.
\n[21-Sep-2009 12:51:26] <Brad_K> and for Mailman, if you want to\npay Mark Sapiro, Tokio Kikuchi, or Barry Warsaw for some consulting,\nthey can provide that.
[21-Sep-2009 12:51:55] <Brad_K> mrayzenoss: sorry, i'm just ranting a bit....
[21-Sep-2009 12:52:21] <rmatte> this room doubles as a rant-room, he's used to it
[21-Sep-2009 12:52:31] <mrayzenoss> yeah, I've got thick skin and short memory
[21-Sep-2009 12:52:40] <mrayzenoss> look over there! shiny!
\n[21-Sep-2009 12:52:58] <rmatte> and apparently a passion for\nfarming since he was considering taking it up the day after the 2.4.3\nrelease
[21-Sep-2009 12:52:59] <Brad_K> and it helps that he's the Community Manager, even if he is a paid employee. ;-)
[21-Sep-2009 12:53:31] <mrayzenoss> my job is to keep the open source side moving, I don't really follow the support side
\n[21-Sep-2009 12:53:36] <rmatte> yeh, but he handles his job very\nprofessionally, he puts the community first and actually argues with\nthe higher ups to get stuff in to core
[21-Sep-2009 12:53:38] <mrayzenoss> I assume they're swamped, they always are
[21-Sep-2009 12:54:21] <rmatte> hire more engineers, I'm available
[21-Sep-2009 12:54:36] <mrayzenoss> if you want to relocate to Annapolis or Austin, we'd probably take you up on it
\n[21-Sep-2009 12:54:36] <Brad_K> well, this is something that\ncould be crowd-sourced, at least for some part of the problem space.
[21-Sep-2009 12:55:00] <mrayzenoss> Brad_K: are you familiar with http://zenoss.org ? that's where we crowd-source
\n[21-Sep-2009 12:55:38] <rmatte> yeh, I doubt I could do the\nrelocation, plus it's almost impossible to get a worker's visa for the\nU.S when you're canadian these days. My friend was supposed to move\ndown to L.A since he works for NBC Universal, and he could never get\nacross the border, so he's still working for them but from home here.
\n[21-Sep-2009 12:55:50] <Brad_K> yeah, i've tried the forums and\nmailing lists. i have yet to see a single response there to anything\ni've ever posted. the irc channel, however, has turned out to be very\nuseful.
[21-Sep-2009 12:56:19] <mrayzenoss> rmatte: Kells works for us from Calgary
\n[21-Sep-2009 12:56:46] <Brad_K> and the FAQs, wikis, and other\ndocumentation is frequently out-of-date, incomplete, or inaccurate.
[21-Sep-2009 12:56:56] <mrayzenoss> yeah, the new site should help.
\n[21-Sep-2009 12:56:59] <rmatte> yeh, but it'd depend if you guys\nwould be interested in having a remote engineer, which you probably\nwouldn't lol
[21-Sep-2009 12:57:01] <mrayzenoss> the old one was impossible to find stuff
[21-Sep-2009 12:57:02] <Brad_K> same with the knowledgebase, for that matter.
[21-Sep-2009 12:57:12] <rmatte> the new site is excellent
[21-Sep-2009 12:57:15] <mrayzenoss> Brad_K: there is an Enterprise KB
[21-Sep-2009 12:57:22] <rmatte> and I see that they fixed the download button on the main page
[21-Sep-2009 12:57:31] <rmatte> someone was a bit too hasty with the crop tool before apparently
[21-Sep-2009 12:58:07] <jb> enterprise knowledgebase?
[21-Sep-2009 12:58:08]\n<Brad_K> they need to get the site re-indexed, so that the search\nfunction works, and you can auto-link the old URLs to their new\nlocations.
[21-Sep-2009 12:58:25] <Brad_K> jb: yes, for paying enterprise customers.
[21-Sep-2009 12:58:31] <jb> where?
[21-Sep-2009 12:58:47] <Brad_K> so far, it has been less useful to me than the other documentation sournces.
[21-Sep-2009 12:58:55] <jb> i don't know about this KB..
[21-Sep-2009 12:59:09]\n<Brad_K> jb: if you're a paying customer, you should have access\nto their customer service portal, and from there you can access the KB.
[21-Sep-2009 12:59:13] <rmatte> Brad_K: the search works fine on the new site, I use it all the time
[21-Sep-2009 12:59:14] <jb> sec
[21-Sep-2009 12:59:45] <jb> ah I never noticed that..
[21-Sep-2009 13:00:00]\n<Brad_K> the times i've tried the search function, it didn't turn\nup much of anything useful -- not like the search function on the old\nsite.
[21-Sep-2009 13:00:36] <mrayzenoss> search.zenoss.com is working again too
[21-Sep-2009 13:00:40] <Brad_K> but then it has been a few days since i used it, so maybe they've fixed it siince then.
[21-Sep-2009 13:01:22] <mrayzenoss> today marks the 7 day anniversary of zenoss.org, so we're still ironing out issues
[21-Sep-2009 13:01:33] <mrayzenoss> but I think the email threading is going to fix today
[21-Sep-2009 13:01:39] <Brad_K> cool.
[21-Sep-2009 13:01:59] <rmatte> how can you even use the word "anniversary"?
[21-Sep-2009 13:02:01] <rmatte> lol
[21-Sep-2009 13:02:11] <rmatte> more like 7th day of survival
[21-Sep-2009 13:02:13] <rmatte> at this point
[21-Sep-2009 13:02:14] <rmatte>
[21-Sep-2009 13:02:31] <rmatte> It's definitely getting nicer day by day though
[21-Sep-2009 13:02:48] <rmatte> just need to iron out importing the code in the forum posts and fixing the wacky characters.
[21-Sep-2009 13:02:50] <Brad_K> next month, my wife and i will be coming up on our 3650th day anniversary. ;-)
[21-Sep-2009 13:03:05] <rmatte> Brad_K: haha
[21-Sep-2009 13:03:09] <rmatte> congrats
[21-Sep-2009 13:03:13] <Brad_K> thanks!
[21-Sep-2009 13:03:23] <rmatte> iancmcc: ping
[21-Sep-2009 13:03:30] <Brad_K> uhoh, we better be quiet now....
[21-Sep-2009 13:03:36] <iancmcc> that was fast
[21-Sep-2009 13:03:47] <rmatte> found an event console bug this morning...
[21-Sep-2009 13:03:51] <iancmcc> bring it
[21-Sep-2009 13:04:32]\n<rmatte> ok, so when you load the event console by default it\nshows you debug events as well, even if you haven't selected for it to\nshow them, then when you select one and click close the screen\nrefreshes and you can no longer see them, but it doesn't move the event\nto history...
[21-Sep-2009 13:04:40] <cgibbons> ian loves bugs
[21-Sep-2009 13:04:46]\n<rmatte> so to actually close the events out you need to\nspecifically click to view debug events, and then you can close them
[21-Sep-2009 13:05:01] <rmatte> oh, and another bug is if you scroll really fast from the top of the list to the bottom the list blanks out
[21-Sep-2009 13:05:09] <rmatte> those are the ones that I noticed today
[21-Sep-2009 13:05:17] <iancmcc> to the second: don't do that.
[21-Sep-2009 13:05:37] <iancmcc> hmm, ok, I'll look into both of those.
[21-Sep-2009 13:06:11]\n<iancmcc> The first seems pretty odd, dunno how it would show\ndebug events if the box isn't checked...there aren't a lot of moving\nparts between the console and the database. I shall investigate.
[21-Sep-2009 13:06:28] <iancmcc> Thanks, keep 'em coming.
[21-Sep-2009 13:08:52] <twm1010> iancmcc: I can tell you the priority icons are MUCH easier on the eyes than the colored entries
[21-Sep-2009 13:09:06] <iancmcc> Good. Did you vote?
[21-Sep-2009 13:09:10] <twm1010> I did.
[21-Sep-2009 13:09:39] <iancmcc> off-by-default is still winning, thankfully.
[21-Sep-2009 13:11:01] <twm1010> yeah, i found that i had to bump magnification up to at least 140% when the colors are on
[21-Sep-2009 13:11:15] <twm1010> the smaller bolded text doesn't work with the colors
[21-Sep-2009 13:11:33] <iancmcc> the colors are actually slightly different now
[21-Sep-2009 13:11:37] <iancmcc> less saturated
[21-Sep-2009 13:11:49] <iancmcc> better contrast. Should be in beta 3.
[21-Sep-2009 13:13:03] <mrayzenoss> beta 3 eta is Wednesday
[21-Sep-2009 13:13:23] <rmatte> yeh, I voted for off by default
[21-Sep-2009 13:13:28] <rmatte> looks cleaner
[21-Sep-2009 13:15:03]\n<rmatte> yeh, .zip zenpacks don't work well at all with King\nCrab, you should really just make it so that only .eggs can be installed
[21-Sep-2009 13:15:18] <rmatte> and all the zip zenpacks will need to be either converted or removed from the list
[21-Sep-2009 13:15:34] <venturaville> they don't work all that well in the _current_ versions
[21-Sep-2009 13:15:51] <rmatte> venturaville: I haven't had issues with them up until King Crab beta 2
[21-Sep-2009 13:16:04] <rmatte> (I didn't try King Crab beta 1, I'm sure the issues are there as well)
[21-Sep-2009 13:16:08] <rmatte>
[21-Sep-2009 13:16:22] <venturaville> your mileage may vary :-)
[21-Sep-2009 13:16:23] <djack> hellow world belgium calling
\n[21-Sep-2009 13:16:51] <rmatte> I have multiple servers running\n2.4.5 and they have several zip ZenPacks installed (which could be an\nissue for the upgrade)
[21-Sep-2009 13:17:05] <mrayzenoss> the release notes for King Crab will note that it's the last release that supports .Zip zenpacks
[21-Sep-2009 13:17:17] <djack> i installed the 2.5 but gives an error
[21-Sep-2009 13:17:36] <rmatte> mrayzenoss: you mean that 2.4.5 is the last release supporting them right?
[21-Sep-2009 13:17:46] <mrayzenoss> no, we're officially deprecating them in 2.5
[21-Sep-2009 13:17:57] <mrayzenoss> so if they're broken in 2.5, they gotta be fixed 1 last time
[21-Sep-2009 13:18:02] <rmatte> ah, so they'll still kinda sorta work but not really?
[21-Sep-2009 13:18:05] <mrayzenoss> yeah
[21-Sep-2009 13:18:18] <mrayzenoss> we'll load them, not guarantees the contents still work
[21-Sep-2009 13:18:27] <mrayzenoss> since they're Community-provided
\n[21-Sep-2009 13:18:45] <djack> anyhow beside the\ngetPingSTatus,uptimeStr,location i also need the Class where the device\nis in
[21-Sep-2009 13:18:49] <mrayzenoss> but be sure to Comment on the different ZenPacks, that's a new feature
\n[21-Sep-2009 13:19:20] <rmatte> there's a bit of an issue with\nusing the term "depricated" though. that term suggests that the\nfunction is still available and functioning properly, but that another\nmore enhanced function is now available.
[21-Sep-2009\n13:19:34] <rmatte> http://dev.zenoss.com/trac/ticket/5547 <---\nIn this case, it's not functioning properly for any zip packs
[21-Sep-2009 13:19:53] <rmatte> I'd suggest just preventing the installation of any .zip files all-together
[21-Sep-2009 13:20:02] <mrayzenoss> hmmm... I gotta go talk to bbibeault
[21-Sep-2009 13:20:05] <iancmcc> Oh, I thought we were fixing that
[21-Sep-2009 13:20:12] <jb> when zenoss queries windows machines via WMI, what port does it use?
[21-Sep-2009 13:20:19] <jb> isn't WMI RPC/DCOM based?
[21-Sep-2009 13:20:33] <cgibbons> yes
[21-Sep-2009 13:20:40] <cgibbons> so it'll start off on port 135, and then get redirected to a secondary DCOM port
[21-Sep-2009 13:20:49] <cgibbons> which you can configure on the Windows server if you want to restrict the port range
[21-Sep-2009 13:20:52] <jb> ugh
[21-Sep-2009 13:21:06] <jb> k
[21-Sep-2009 13:21:10] <cgibbons> want the link for that?
[21-Sep-2009 13:21:13] <jb> please
[21-Sep-2009 13:21:19] <cgibbons> you can do it via a Domain/Group Policy so you can deploy it consistently
\n[21-Sep-2009 13:21:28] <Brad_K> djack: so, where in Belgium are\nyou? My wife and I lived in Brussels for almost eight years.
[21-Sep-2009 13:21:38] <cgibbons> here you go: http://msdn.microsoft.com/en-us/library/ms809327.aspx
[21-Sep-2009 13:21:48] <jb> thanks
[21-Sep-2009 13:21:55] <djack> Brad_K between antwerp and brussels
\n[21-Sep-2009 13:22:31] <Brad_K> i used to drive that route four\ndays a week, commuting between Brussels and Eindhoven. where exactly\nbetween Antwerp & Brussels?
[21-Sep-2009 13:22:44] <jb> hrm, im having a sort of weird problem
[21-Sep-2009 13:22:54] <jb> zen.pysamba - Attempt to connect resulted in NT_STATUS_ACCESS_DENIED
[21-Sep-2009 13:23:03] <jb> although I think its pulling WMI perfmon counters fine
[21-Sep-2009 13:24:07] <jb> its just the services
[21-Sep-2009 13:24:22] <cgibbons> are you trying to use a non-admin user?
[21-Sep-2009 13:24:24] <djack> Brad_K a small village called Niel
[21-Sep-2009 13:24:29] <jb> nope.. same user I use for every box
[21-Sep-2009 13:24:44] <jb> every other box in this device class is fine
[21-Sep-2009 13:24:51] <cgibbons> okay, let's try somethign real quick. i can tell you where the misconfiguration is.
[21-Sep-2009 13:24:56] <jb> k
[21-Sep-2009 13:25:02] <cgibbons> do a wmic -d99 -U 'whomever' //server 'select name from win32_computersystem'
[21-Sep-2009 13:25:04] <cgibbons> that'll be noisy
\n[21-Sep-2009 13:25:45] <cgibbons> if you get a 'Failed to bind to\nuuid' message @ the bottom it's one thing, if you get a whole bunch\nmore network dump data and it keeps going before the access denied,\nit's something else
[21-Sep-2009 13:26:44] <jb> Failed to bind to uuid 4d9f4ab8-7d1c-11cf-861e-0020af6e7c57 - NT_STATUS_END_OF_FILE
[21-Sep-2009 13:26:44] <jb> ERROR: Login to remote object.
[21-Sep-2009 13:26:44] <jb> NTSTATUS: NT_STATUS_ACCESS_DENIED - Access denied
[21-Sep-2009 13:26:50] <cgibbons> okay yep
\n[21-Sep-2009 13:27:13] <cgibbons> so, when it fails that early\nyou've got a) bad username, b) bad password, c) user hasn't been\ngranted access to do remote logins to that server
[21-Sep-2009 13:27:20] <cgibbons> it's probably C in this case
[21-Sep-2009 13:27:29] <cgibbons> do you use any group policy stuff in your domain, or just local server configs?
[21-Sep-2009 13:27:29] <jb> odd
[21-Sep-2009 13:27:31] <jb> its a domain admin
[21-Sep-2009 13:27:46] <cgibbons> very odd then
\n[21-Sep-2009 13:28:47] <Brad_K> djack: ahh, yes. not too far\nfrom the A12. i took that route on occasion, when i was looking for\nsomething a bit more scenic and lower-stress than the A1/E19.
[21-Sep-2009 13:29:25] <djack> Brad_K that time as long passed
[21-Sep-2009 13:29:26] <Brad_K> djack: i definitely remember driving through Boom.
[21-Sep-2009 13:29:53] <djack> Brad_K that a mile from here
[21-Sep-2009 13:29:59] <djack> hmmm
\n[21-Sep-2009 13:30:22] <mrayzenoss> For everyone who was using\nthe email bridge, threading has been fixed. No more junk in the\nsubject lines
[21-Sep-2009 13:30:47] <Brad_K>\ndjack: anyway, you said you were having a problem with 2.5? i think\nthat iancmcc is the right guy to be talking to about that....
[21-Sep-2009 13:31:03] <djack> Brad_K
[21-Sep-2009 13:31:47] <mrayzenoss> rmatte: was 5547 a regression?
[21-Sep-2009 13:31:54] <mrayzenoss> rmatte: I'm reopening it
\n[21-Sep-2009 13:31:54] <djack> the 2.5 was for test i'm building\na external dashboard with application inteligence but i need to get the\nClass where the device is located out of dmd
[21-Sep-2009 13:32:33] <rmatte> mrayzenoss: I'm not sure if I had set it as such
[21-Sep-2009 13:33:01] <rmatte> nah, I never set it as regression
[21-Sep-2009 13:33:13] <mrayzenoss> I mean, did the author and version show up before?
[21-Sep-2009 13:33:29] <mrayzenoss> with zip zenpacks
[21-Sep-2009 13:33:37] <rmatte> yeh, they did
[21-Sep-2009 13:33:45] <rmatte> it works perfectly fine on my 2.4.5 installs
[21-Sep-2009 13:33:49] <iancmcc> djack: dmd.Devices.findDevice('mydevice').deviceClass().getPrimaryId()
[21-Sep-2009 13:34:04] <mrayzenoss> ok, reopened
[21-Sep-2009 13:34:08] <djack> iancmcc lets try
[21-Sep-2009 13:35:46] <rmatte> For example:
[21-Sep-2009 13:35:47] <rmatte> APC UPS PowerNet n/a Richard Palmer 0.5 No
[21-Sep-2009 13:35:47] <rmatte> Fortigate n/a Fabio Paracchini 1.0 No
[21-Sep-2009 13:35:47] <rmatte> HPTemperatureSensors n/a Trevor Seward 1.1 No
[21-Sep-2009 13:35:47] <rmatte> Is now:
[21-Sep-2009 13:35:47] <rmatte> APC UPS PowerNet n/a 0.1 No
[21-Sep-2009 13:35:49] <rmatte> Fortigate n/a 0.1 No
[21-Sep-2009 13:35:51] <rmatte> HPTemperatureSensors n/a 0.1 No
[21-Sep-2009 13:36:25] <mrayzenoss> rmatte: ok, just checking
[21-Sep-2009 13:36:28] <mrayzenoss> it's reopened
[21-Sep-2009 13:36:36] <mrayzenoss> might not get fixed, but it's open
[21-Sep-2009 13:36:39] <rmatte> k lol
[21-Sep-2009 13:36:52] <mrayzenoss> sev 3, there are bigger fish to fry
[21-Sep-2009 13:37:03] <rmatte> yeh, for sure
[21-Sep-2009 13:37:07] <mrayzenoss> but it may get inadvertently fixed
[21-Sep-2009 13:37:12] <mrayzenoss> if there are other .zip zenpack issues
[21-Sep-2009 13:37:25] <rmatte> yeh
[21-Sep-2009 13:37:57] <jb> cgibbons: ok think I figured it out.
\n[21-Sep-2009 13:37:58] <rmatte> I'll do some testing with that at\nsome point, just don't want to risk screwing up my install since I'm\ntesting other stuff today
[21-Sep-2009 13:38:05] <rmatte> I might just take a snapshot of the vm and try it tomorrow
[21-Sep-2009 13:40:10] <djack> iancmcc thanks it works :-) is there a list of this?
\n[21-Sep-2009 13:40:56] <iancmcc> djack: If you mean\ndocumentation, there's the dev guide; I'm not sure how comprehensive it\nis, however
[21-Sep-2009 13:41:11] <djack> iancmcc not really
\n[21-Sep-2009 13:41:37] <djack> well i have what i need for this\nmoment going to set up the mysql db and build the php interface above
\n[21-Sep-2009 13:41:40] <rmatte> the dev guide isn't very\ncomprehensive, most of the development stuff that I've learned has just\nbeen through lots of digging through other people's work and the forums\n
[21-Sep-2009 13:42:00] <mrayzenoss> djack: you can add it here: community/documentation/wiki/zendmd
[21-Sep-2009 13:42:01] <iancmcc> Yeah, might try cruising the API docs.
[21-Sep-2009 13:42:59] <djack> iancmcc i only read something when i fucked it up completely :-p
[21-Sep-2009 13:43:23] <iancmcc> Usually my MO as well.
[21-Sep-2009 13:44:23] <djack> everybody knows it when i pull up the pdf we are in real shit
[21-Sep-2009 13:44:29] <djack> hihi
[21-Sep-2009 13:44:47] <cgibbons> what was broke, jb?
\n[21-Sep-2009 13:45:25] <Brad_K> djack: isn't there an appropriate\nFlemish word that you can teach us for those terms? I know the French\nwords, but the Flemish ones would be so much more entertaining....
[21-Sep-2009 13:45:42] <rmatte> "Oh no, djack is READING! ABANDON SHIP!"
[21-Sep-2009 13:46:04] <djack> rmatte something like that ....
[21-Sep-2009 13:46:26] <djack> Brad_K Flemisch word for what ?
[21-Sep-2009 13:46:39] <Brad_K> fuck and shit?
[21-Sep-2009 13:46:56] <Brad_K> words, really. not single word, i hope.
\n[21-Sep-2009 13:47:25] <Brad_K> i'd love to be able to throw\nthose into casual conversation and completely confuse the hell out of\npeople....
[21-Sep-2009 13:47:26] <djack> Brad_K well we are very international so fuck and shit is also used here
[21-Sep-2009 13:47:41] <djack> ik zal je eens wat vertellen ;-)
[21-Sep-2009 13:48:00] <djack> nu denk ik niet dat iemand dit gaat verstaan
\n[21-Sep-2009 13:48:19] <Brad_K> well, far too many Flemish people\nare far too willing to speak English, and do so better than most\n'Merkins. They're like that, I know. But it would still be\ninteresting to learn them....
[21-Sep-2009 13:49:42] <djack> Brad_K we speak dutch french and english for the flemish part of my country
\n[21-Sep-2009 13:49:45] <Brad_K> i was especially amused when the\nwinner of the world French spelling bee was an English teacher in\nAntwerp, and he spoke five or six languages fluently, in addition to\nhis native Vlaams.
[21-Sep-2009 13:50:44]\n<Brad_K> And if you ask a Dutchman, Flemish is just a dialect of\nDutch. But if you ask a Flemish man, then Dutch and Flemish are two\ndistinct languages.
[21-Sep-2009 13:50:52] <Brad_K> yes, I know that joke.... ;-)
\n[21-Sep-2009 13:50:54] <djack> Brad_K i work in Rotterdam and\neven Dutch people and Flemish people speak the same language there is a\nworld of difference
[21-Sep-2009 13:52:04] <djack> Brad_K from where are you if i may ask?
\n[21-Sep-2009 13:52:23] <Brad_K> yes, like the fact that "worker"\nin Flemish means "slave" in Dutch. Imagine the chagrin of the Flemish\nCEO of a company in Brussels that was buying out their Dutch\ncounterpart, and he went to go give a speech at the Workers Council,\nand he *didn't* know the difference....
[21-Sep-2009 13:52:47] <rmatte> I speak french and english, no dutch though
[21-Sep-2009 13:53:29] <venturaville> I always hear about high and low dutch ... but not certain what the difference is
\n[21-Sep-2009 13:53:32] <Brad_K> ... Speaking for over an hour\nabout "... the workers ..." this, and " ... the workers ... " that, and\nall the while everyone in the room was hearing " ... the slaves ... "\nthis, and " ... the slaves ... " that.
[21-Sep-2009 13:53:33] <djack> rmatte you don't know what you are missing
\n[21-Sep-2009 13:54:33] <Brad_K> actually, there are many\ndifferent dialects of Dutch and Flemish. Not just High and Low, but\nalso North vs. South, East vs. West, and don't even get me started\nabout Frisian (the other official languages of Die Nederlands).
[21-Sep-2009 13:55:22] <djack> Fries yeah that is a completely other language
[21-Sep-2009 13:55:23] <rmatte> There are lots of different dialects of French as well
[21-Sep-2009 13:55:30] <djack> even for the dutch people
\n[21-Sep-2009 13:55:33] <Brad_K> one thing you quickly discover\nabout most Flemish people is that they are all amateur linguists,\ncompletely fluent in at least a half dozen languages, and very, very\nwilling to talk for many hours on the subject.
\n[21-Sep-2009 13:55:41] <rmatte> Canadian french is considered to\nbe "slangy" or "slum" French in comparison to parisian french
\n[21-Sep-2009 13:56:43] <djack> every town has its own version of\nflemish I'm from antwerp and my wife is from Roeselare (125 km from\nantwerp) they talk in a way that on tv they use subtitles
[21-Sep-2009 13:57:04] <rmatte> lol, that's crazy
[21-Sep-2009 13:57:20] <djack> rmatte its crazy here that a fact
[21-Sep-2009 13:57:29] <twm1010> jeebus, everytime i tab back to this window the conversation is completely different
[21-Sep-2009 13:57:43] <rmatte> twm1010: haha
[21-Sep-2009 13:59:18] <djack> lol
[21-Sep-2009 14:01:04]\n<Brad_K> And the language spoken by the Belgians is considered a\n"country" French, and they don't mean that in a nice way. It's like\nthey're the damn stupid country hicks to the north, and all the members\nof L'Academie Francaise are horrified.
[21-Sep-2009 14:02:00]\n<Brad_K> we learned this the hard way when we tried to speak\nfrench as we had learned it in Brussels when we went to visit various\nplaces in Paris and elsewhere in France -- like taking the Normandy\nInvasion tour.
[21-Sep-2009 14:02:36] <djack> Brad_K France is something else
[21-Sep-2009 14:03:03]\n<Brad_K> yes, we know. imo, no one in the world has the right to\nhate the French unless they've actually had to deal with them on\noccasion.
[21-Sep-2009 14:03:28] <venturaville> of course we have the same thing between the North and the deep south here in the US
[21-Sep-2009 14:03:34] <Brad_K> you only get the right to hate, loathe, and despise them once you've actually had dealings with them.
[21-Sep-2009 14:03:59] <Brad_K> venturaville: not like this. the French are far worse to their "country hicks" than we are to ours.
[21-Sep-2009 14:04:39]\n<Brad_K> djack: my wife and i now live in Austin, TX, which is\none of the few cities in the US where we could move back with as little\nculture shock as possible.
[21-Sep-2009 14:04:40] <djack> Brad_K I don't hate them its a nice country but the mentality is not what we are used to here in flanders
[21-Sep-2009 14:04:54]\n<venturaville> I'll take your word for it, though I don't forsee\na redneck from my home town fairing well in the blue blood states :-)
[21-Sep-2009 14:05:03] <djack> Brad_K well i lost my heart in New York
[21-Sep-2009 14:05:35]\n<Brad_K> venturaville: think about how blacks are still sometimes\ntreated in all-white areas of the south. that's closer.
[21-Sep-2009 14:06:23]\n<Brad_K> djack: and my wife and i truly loved Antwerp, Brugge,\nGhent, Liege, London, Barcelona, and various other cities there in\nEurope.
[21-Sep-2009 14:06:44] <Brad_K> oh, and Orval was another favourite of ours. Really good beer. ;-)
[21-Sep-2009 14:07:04] <venturaville> it tends to be more about affluent versus poor these days, but yeah I see your point
[21-Sep-2009 14:07:19] <djack> Brad_K beer that the one product that we can make ....
[21-Sep-2009 14:07:47] <djack> Inbev is from leuven
[21-Sep-2009 14:07:55] <Brad_K> Westvleteren is nice, but then I like all the Trappiste beers.
[21-Sep-2009 14:08:20] <djack> Westvleteren is very special
[21-Sep-2009 14:09:47] <Brad_K> I'm not so much for the more common beers. I prefer more specialist beers, like Kriek or Trappiste.
[21-Sep-2009 14:10:26] <djack> Brad_K haha
[21-Sep-2009 14:10:29] <Brad_K> In fact, most any Lambic is enough to get my attention.
[21-Sep-2009 14:10:48] <djack> Brad_K i'll mail you one ;-)
[21-Sep-2009 14:11:24] <rmatte> ooh, swag, beer swag
[21-Sep-2009 14:11:27] <rmatte>
\n[21-Sep-2009 14:12:15] <Brad_K> djack: actually, we have this\nplace here in Austin called Whole Foods, and I've been amazed at the\nkinds of beers that they stock. they even have some Belgian beers that\nI had never heard of.
[21-Sep-2009 14:12:41] <djack> Brad_K you should try them all
\n[21-Sep-2009 14:12:51] <mrayzenoss> Brad_K: check out the New\nBelgium 1554, that's my current favorite "fairly-easy-to-find beer"
[21-Sep-2009 14:13:08] <cgibbons> the rich hippy store!
[21-Sep-2009 14:13:12] <djack> well i more of a wine wiskey lover
\n[21-Sep-2009 14:13:20] <Brad_K> But please don't send any Stella\nArtois or Heineken. Those are so bad that the only thing they can do\nis export them -- and somehow, the rest of the world thinks they're\nactually good.
[21-Sep-2009 14:13:26] <mrayzenoss> cgibbons: you can get the beer other places
[21-Sep-2009 14:13:39] <djack> Heineken is NOT BELGIUM
[21-Sep-2009 14:13:54] <djack> Heineken is pissbeer
\n[21-Sep-2009 14:14:12] <Brad_K> djack: if you drink a bottle of\nBush Beer for every bottle of the ~500 so different types of Belgian\nbeer that there are, i'll take you up on that.
[21-Sep-2009 14:14:13] <djack> it's an insult
[21-Sep-2009 14:14:28] <Brad_K> yeah, well Stella isn't that much better.
[21-Sep-2009 14:15:03] <djack> Brad_K I'm jupiler
[21-Sep-2009 14:15:05] <mrayzenoss> speaking of beer... Zenoss is sponsoring the pre-party at Ohio LinuxFest, be sure to show up Friday night
[21-Sep-2009 14:15:06] <Brad_K> regretfully, i never managed to make it up to Stadsfeestzaal.
[21-Sep-2009 14:15:21] <djack> mrayzenoss you pay the tickets I'll fly over
[21-Sep-2009 14:15:25] <mrayzenoss>
[21-Sep-2009 14:15:33] <mrayzenoss> free training on Friday!
[21-Sep-2009 14:15:35] <djack> even will bring belgium beer
[21-Sep-2009 14:16:47] <djack> Brad_K its a shopping center but very nicely renovated
\n[21-Sep-2009 14:17:11] <Brad_K> One of my favourite places to\ndrink beer was in Falstaff's in Brussels. Not so much on the selection\nof beer, as the architecture. And occasionaly we could trick the\nunsuspecting tourist into drinking a few Bush Beers, and then trying to\nstand up (it's 12-14% alcohol).
[21-Sep-2009 14:17:24] <cgibbons> ouf
\n[21-Sep-2009 14:18:16] <Brad_K> I heard about the "24 Hours" at\nStadsfeestzaal, but I never managed to make it. Came close one year,\nbut not quite.
[21-Sep-2009 14:18:54] <Brad_K>\nBush Beer is kind of like Everclear -- you don't really notice how much\nalcohol you're drinking until it is much, much too late.
[21-Sep-2009 14:19:29] <kmtsun> is there a zenoss zenpack for freebsd?
[21-Sep-2009 14:19:37] <Brad_K> and to the people who are used to 3.2% horse piss, watch out!
[21-Sep-2009 14:19:38] <rmatte> everclear aka moonshine
[21-Sep-2009 14:19:57] <rmatte> I can't even imagine drinking 3.2% beer
[21-Sep-2009 14:20:13] <rmatte> our beer here rangers from 4.7 to 7
[21-Sep-2009 14:20:18] <rmatte> ranges*
[21-Sep-2009 14:20:21] <mrayzenoss> kmtsun: no, but we'd love an SSH one like the OpenSolaris and Linux ones
[21-Sep-2009 14:20:23] <rmatte> generally it's 5.5
[21-Sep-2009 14:20:35] <Brad_K> kmtsun: i don't see anything at community/zenpacks
[21-Sep-2009 14:20:48] <kmtsun> oh, kk thanks
[21-Sep-2009 14:20:52] <mrayzenoss> kmtsun: SNMP generally works
[21-Sep-2009 14:21:06]\n<Brad_K> we may have to hack on one of the Linux or OpenSolaris\nones, because we do have some FreeBSD machines we need to monitor.
[21-Sep-2009 14:21:08] <rmatte> mrayzenoss: had time to go through the pending list of ZenPacks yet?
[21-Sep-2009 14:21:13] <kmtsun> yes... generally
[21-Sep-2009 14:21:41] <mrayzenoss> rmatte: guess I'll go get to work
[21-Sep-2009 14:21:45] <rmatte> hehe
[21-Sep-2009 14:21:48] <djack> well i have snmp activated on a apc power switch ad it does not work
[21-Sep-2009 14:21:54] <Brad_K> mrayzenoss: i thought you were at work!
[21-Sep-2009 14:22:30] <djack> I'm in my sofa before the tv 9:30 in the evening here
[21-Sep-2009 14:22:42] <Brad_K> djack: did you try the zenpack at docs/DOC-3417 ?
[21-Sep-2009 14:22:45] <rmatte> Brad_K: ummm, have you tried just using the /Server/Darwin group?
[21-Sep-2009 14:22:56] <rmatte> Darwin is based on FreeBSD, so they are very similar
[21-Sep-2009 14:22:59] <rmatte> it would probably work
[21-Sep-2009 14:23:13] <djack> Brad_K let's see
\n[21-Sep-2009 14:23:35] <Brad_K> Darwin is an ancient Mach kernel,\nwith some FreeBSD userland, but also some NetBSD and OpenBSD thrown in\nfor good measure.
[21-Sep-2009 14:23:53] <djack> Brad_K well the powerswitch does not respone to an snmpwalk
[21-Sep-2009 14:23:59] <rmatte> Brad_K: yeh, but I'm talking about command structurewise
[21-Sep-2009 14:24:19] <rmatte> I'm pretty sure the Darwin device grouping collects data via SSH
[21-Sep-2009 14:24:33] <mrayzenoss> I'm at work, just haven't been working on ZenPacks
[21-Sep-2009 14:24:34] <rmatte> I'm just saying that the plugins would probably work fine for FreeBSD
[21-Sep-2009 14:24:42] <bill_no> hey all
[21-Sep-2009 14:24:54] <rmatte> hello
[21-Sep-2009 14:25:10] <djack> hey bill
\n[21-Sep-2009 14:25:54] <Brad_K> i've been a MacFanatic ever since\nDecember of 1983, when I saw an early prototype of the original Mac,\nbefore the official announcement during the SuperBowl in January of\n1984. I switched from MacOS 9 to Mac OS X as soon as it was stable\nenough to do so (right around the 10.2 days), and I'm on the latest\nrelease of Snow Leopard now. But that doesn't mean that I don't\nrecognize that this is one hell of a FrankenBeast OS that they've\ncreated. And now t
[21-Sep-2009 14:26:31] <rmatte> got cut off at And now t
[21-Sep-2009 14:26:51] <Brad_K> ... And now they're replacing gcc, of all things.
[21-Sep-2009 14:27:20] <cgibbons> you see the pumpkin ale in the fridge yet, matt?
\n[21-Sep-2009 14:27:30] <Brad_K> anyway, the Darwin stuff would\nprobably be a closer place to start than Linux or OpenSolaris. So,\nwhen the time comes, i'll try to remember that.
\n[21-Sep-2009 14:27:38] <bill_no> i found a web site that provides\na guide on how to reduce the perf graph to 8 hours - but it doesnt work\nand doesnt say which version of zenoss it applies to either (so ive\nfound) has anyone tried to reduce the perf graph time?
\n[21-Sep-2009 14:28:50] <mrayzenoss> bill_no:\ndocs/DOC-2445#HowdoIchangetheGraphintervalsfrom5minutestosomethingelse
[21-Sep-2009 14:29:00]\n<Brad_K> djack: are you sure you have the community and password\nsettings correct? and that you've got the network monitor module\ninstalled? because most APC equipment doesn't come with the network\nmonitor module by default, just an RJ-45 jack that is supposed to\nconnect to a custom USB/Rj-45 cable, and plug directly into a\nmanagement host.
[21-Sep-2009 14:29:04] <mrayzenoss> that's quite an anchor tag
[21-Sep-2009 14:29:47] <bill_no> oooo eek - I didnt find that one
[21-Sep-2009 14:30:43] <djack> Brad_K i can connect to the webinterface so i have the network module
[21-Sep-2009 14:31:06] <Brad_K> 'k
[21-Sep-2009 14:31:19] <djack> I see lots of ports that are open but not 161 ....
\n[21-Sep-2009 14:31:51] <Brad_K> ahh. sounds like someone else\nmay have reconfigured it to turn off SNMP. can you reconfigure it\nthrough the web interface?
[21-Sep-2009 14:32:06]\n<djack> first my idea was firewall issue but that is not the case\nbecause i can not do snmpwalks from the management network either
[21-Sep-2009 14:32:36] <djack> Brad_K snmp version 1 is activated
\n[21-Sep-2009 14:33:26] <djack> but i will figure it out tommorow\n.... i can not access the management network from my home so ....
[21-Sep-2009 14:34:49] <djack> the powerswitches are the last devices that are not monitored so it's the missing link
\n[21-Sep-2009 14:37:56] <Brad_K> djack: SNMP v1 may be the\nproblem. Most devices these days are V2c, if not V3. i'm not sure\nthat the zenpack will handle SNMPv1.
[21-Sep-2009 14:38:16] <djack> Brad_K i can use the v3 also
[21-Sep-2009 14:39:19] <Brad_K> check to see what the zenpack supports, then see if you can enable that on the UPS.
[21-Sep-2009 14:39:45] <twm1010> APC ups?
[21-Sep-2009 14:39:45] <djack> will do ;-)
[21-Sep-2009 14:39:57] <djack> twm1010 apc power switch
[21-Sep-2009 14:39:57] <Brad_K> twm1010: yes.
[21-Sep-2009 14:40:03] <twm1010> My two units are v1 ONLY.
[21-Sep-2009 14:40:34] <djack> the one we use are v1 and v3
[21-Sep-2009 14:41:02]\n<twm1010> In the SNMP config page for my UPS's I have to specify\nthe IP of the server that will be making SNMPGETs, or no go.
[21-Sep-2009 14:41:27] <twm1010> they're quite finicky in that respect
[21-Sep-2009 14:42:49] <djack> did anybody place a threshold on the swap usage
[21-Sep-2009 14:47:41] <Brad_K> okay, i'm off to lunch. bbl.
[21-Sep-2009 14:48:31] <djack> Lunch ?
[21-Sep-2009 14:49:21] <Brad_K> yeah, it's 2:49pm, but i got into the office late, so i'm going off to lunch late.
[21-Sep-2009 14:49:37] <djack> lol
[21-Sep-2009 14:49:48] <djack> if you get in late you can leave early
[21-Sep-2009 14:49:59] <Brad_K> i wish.
[21-Sep-2009 14:50:12] <djack> haha
[21-Sep-2009 14:50:17] <djack> have a nice lunch
\n[21-Sep-2009 14:53:44] <jb> cgibbons: so the problem was these\ntwo boxes run ISA.. and ISA and DCOM don't play together nicely
[21-Sep-2009 14:54:00] <cgibbons> crap... what's ISA?
[21-Sep-2009 14:54:14] <jb> MS internet security and accel server
[21-Sep-2009 14:54:18] <jb> basically a web proxy/firewall
[21-Sep-2009 14:54:23] <jb> we just use it to proxy
[21-Sep-2009 14:54:36] <jb> it feeds our websense policy servers (blocks certain websites, etc)
[21-Sep-2009 14:54:39] <cgibbons> well, that's convenient
[21-Sep-2009 14:54:44] <cgibbons> is it from MS or thirdparty?
[21-Sep-2009 14:54:46] <jb> MS
[21-Sep-2009 14:54:49] <cgibbons> ugh
[21-Sep-2009 14:54:52] <jb> heh yeah
[21-Sep-2009 14:54:56] <djack> jb squid squid squid :-p
\n[21-Sep-2009 14:54:57] <jb>\nhttp://blogs.technet.com/isablog/archive/2007/05/16/rpc-filter-and-enable-strict-rpc-compliance.aspx
[21-Sep-2009 14:55:05] <jb> yeah, i could use squid
[21-Sep-2009 14:55:24] <djack> haha
[21-Sep-2009 14:56:31] <djack> going to bed speak to you laters
[21-Sep-2009 15:03:26] <cmdln> hello
[21-Sep-2009 15:03:28] <twm1010> ISA is freaking....
[21-Sep-2009 15:03:39] <twm1010> particular.
[21-Sep-2009 15:05:57]\n<cmdln> I have zenoss reciving snmp traps from my APC ups. The\nUPS is also configured to send me emails. I have been getting notified\nof failed snmp logins on the UPS. the emails contain the ip of the\nclient that tried to connect but I can not seem to find the ip of the\nclient when I look throught the zenoss events
[21-Sep-2009 15:06:24] <jb> do you have the MIBs loaded?
[21-Sep-2009 15:06:31] <jb> click on the "Details" tab of the event
[21-Sep-2009 15:06:35] <jb> and see if you can see it there
[21-Sep-2009 15:06:53] <cmdln> I do have the MIBS loaded
[21-Sep-2009 15:07:13] <jb> http://thunder.jbdesign.net/~jbaird/zenoss/event-old.jpg
[21-Sep-2009 15:07:20] <jb> ^^ example
[21-Sep-2009 15:08:05] <cmdln> mmm i have fields with oids that are not mapped to names
[21-Sep-2009 15:08:17] <cmdln> but in the value column i dont see any ips anyway
[21-Sep-2009 15:08:29] <jb> so its not available via the trap..
[21-Sep-2009 15:08:34] <cmdln> ah
[21-Sep-2009 15:08:42] <cmdln> fail on apcs part
[21-Sep-2009 15:09:26] <cmdln> well that sucks
[21-Sep-2009 15:09:30] <jb> :/
[21-Sep-2009 15:09:30] <cmdln> thanks jb
[21-Sep-2009 15:09:32] <jb> np
[21-Sep-2009 15:19:03]\n<Signum> Evening... I'm currently evaluating monitoring systems\nsince Nagios and Cacti suck for our purpose. Choices left are opsview,\nzabbix and zenoss. So bear with me if I'm not 100% familiar with\nzenoss. : )
[21-Sep-2009 15:20:07] <Signum> I'm\ncurrently trying to find out how to monitor a mysql database. I'm\nalready monitoring the network interfaces and CPU load. But I'm not\nsure how I can add monitoring for th mysqld process or get further\nmysql statistics. Can someone push me in the right direction please?
\n[21-Sep-2009 15:20:40] <Signum> I suspect I'm going wrong\nsomewhere. Fancy features like the network map don't show me any\nsystems. Seems like I just forgot some stupid step.
[21-Sep-2009 15:21:55] <cmdln> Signum: you have a few different options for collecting the data
[21-Sep-2009 15:22:10] <cmdln> but pretty much any way you go you will need a script that will output the data you want
[21-Sep-2009 15:22:41] <cmdln> there is no agent that will monitor mysql automatically
[21-Sep-2009 15:23:00] <Signum> Okay, so I could use a shell/perl/whatever script to get the values I need.
[21-Sep-2009 15:23:05] <cmdln> yep
[21-Sep-2009 15:23:24] <Signum> The script would be run locally on the server that zenoss runs on, right?
[21-Sep-2009 15:23:31] <cmdln> and those values could be pulled either via ssh or as custom execs in snmpd
[21-Sep-2009 15:23:35] <Signum> ok
\n[21-Sep-2009 15:23:50] <Signum> I'm famliar with snmp/exec mess\nas that's how I've done thing with Nagios for the last 10 years.
[21-Sep-2009 15:23:54] <cmdln> no the script would run on the server thats being monitored
[21-Sep-2009 15:24:00] <cmdln> yeh
\n[21-Sep-2009 15:24:29] <Signum> That's one of the few big\nadvantages of zabbix btw. Their remote agent is pretty nifty and gets\nrid of the snmp/exec chaos. But anyway.
[21-Sep-2009 15:24:38] <cmdln> yeh
[21-Sep-2009 15:24:44] <cmdln> i used zabbix for a while
[21-Sep-2009 15:24:48] <cmdln> the agent was nice
\n[21-Sep-2009 15:25:16] <Signum> Currently no "OS Processes" or\n"IP services" are shown in the "OS" tab. Do I need to assign a template\nto get that working? I have just added that device via SNMP and get\ninformation about CPU load, network load and routes.
\n[21-Sep-2009 15:25:18] <cmdln> but then you just have to monitor\nso many things that cant run an agent so it lost its luster for me
[21-Sep-2009 15:25:39] <cmdln> for os processes i think you would need to add processes to monitor
\n[21-Sep-2009 15:25:50] <Signum> (for me zabbix big con is the\nchaotic GUI and the lot of clicking that has to be done for even simple\nmonitoring tasks)
[21-Sep-2009 15:25:59] <cmdln> not sure about the ipservices
\n[21-Sep-2009 15:26:19] <cmdln> i suppose if the device was\nrunning a very standard service like http it would pick it up\nautomatically
[21-Sep-2009 15:26:28] <Signum> If\nI try "Add processes" from the menu then I get a popup asking me for\nthe "Process Class" but the choice is empty.
[21-Sep-2009 15:26:53] <cmdln> ?
[21-Sep-2009 15:27:11] <Signum> Yes, I was surprised, too.
[21-Sep-2009 15:27:12] <cmdln> try adding from the processes class
[21-Sep-2009 15:27:20] <Signum> Even via SNMP zenoss should be able to fetch the list of running processes.
[21-Sep-2009 15:27:22] * Signum tries
[21-Sep-2009 15:29:34] <Signum> hmmm
[21-Sep-2009 15:33:01] <Signum> Cool, that worked indeed.
\n[21-Sep-2009 15:35:19] <Signum> Erm. Just if I add a process\ncalled "foo" but there is no process "foo" on the system I also get a\nstatus=UP. Very strange.
[21-Sep-2009 15:35:53] <cmdln> weird
[21-Sep-2009 15:37:17] <Signum> How would zenoss find out if a process is running. Via snmp?
[21-Sep-2009 15:37:42] <Signum> I read that zenoss is very snmp-dependent regarding its checks.
[21-Sep-2009 15:38:06] <mrayzenoss> windows boxes use WMi
[21-Sep-2009 15:38:12] <mrayzenoss> for Services
[21-Sep-2009 15:38:22] <Signum> True. This is a Debian server though.
\n[21-Sep-2009 15:38:47] <mrayzenoss> take a look at\nhttp://www.skills-1st.co.uk/papers/jcurry.html and\nhttp://www.skills-1st.co.uk/papers/jane/process_monitoring.pdf
[21-Sep-2009 15:38:59] <Signum> tnx
[21-Sep-2009 15:40:05] <baffle> Signum: But, yeah, process information is gathered via SNMP.
\n[21-Sep-2009 15:41:10] <Signum> baffle: Should I have seen the\nlist of running processes in the "OS Processes" then? SNMP read access\nuses the right community string and should return everything (not just\n"system").
[21-Sep-2009 15:43:24] <kmtsun> how do add the nmap plugin to zenoss?
[21-Sep-2009 15:45:19] <baffle> Signum: If the device (host) is in a device-class where it models this, yes.
[21-Sep-2009 15:53:21] <kmtsun> oh, so basically you make a user command using nmap commands
\n[21-Sep-2009 15:59:26] <Brad_K> hmm. does anyone know of any\nissues in upgrading from Enterprise 2.4.2 to Enterprise 2.4.5 on\nRHEL5? Is it really just as simple as pkgadd or yum update?
[21-Sep-2009 16:03:32] <mrayzenoss> this is the only issue I'm aware of in 2.4.5: http://dev.zenoss.org/trac/ticket/5494
[21-Sep-2009 16:03:37] <mrayzenoss> and the patch is available
\n[21-Sep-2009 16:23:11] <Signum> Thanks for your time. I'll\ncontinue my studies tomorrow. Although Zenoss still feels a little\nopaque it's closer to what I want to achieve.
\n[21-Sep-2009 16:23:35] <Signum> I'm one of the harder customers\nwho are not happy if something works by accident but who try to\nunderstand their tools.
[21-Sep-2009 16:29:09] <jb> Brad_K: i just went from 2.4.2 -> 2.4.5
[21-Sep-2009 16:29:13] <jb> no problems..
[21-Sep-2009 16:29:15] <jb> just update the RPM's
[21-Sep-2009 16:32:55] <Brad_K> cool. thanks!
[21-Sep-2009 17:39:10] <Brad_K> will be giving us a presentation on the bcfg2 configuration management tool.
[21-Sep-2009 17:39:34] <Brad_K> whoops, wrong window.
\n[21-Sep-2009 17:40:20] <Brad_K> Actually, I will repeat what I\njust told ckrough privately, for anyone in this channel who is in\nAustin and interested:
[21-Sep-2009 17:40:24]\n<Brad_K> LOPSA-Austin meeting Tuesday night, 22 Sep 2009 at 6:30\npm at the Mangia Pizza on Guadalupe.
[21-Sep-2009 17:40:36] <Brad_K> Sol Jerome of the Institute for Computational Engineering and Science at UT
[21-Sep-2009 17:40:37] <Brad_K> will be giving us a presentation on the bcfg2 configuration management tool.
[21-Sep-2009 22:13:37] <mloven> hey all. with the new community site, are the irc dev chats hosted anywhere?
[22-Sep-2009 00:00:46] [disconnected at Tue Sep 22 00:00:46 2009]
[22-Sep-2009 00:00:46] [connected at Tue Sep 22 00:00:46 2009]
[22-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
[22-Sep-2009 09:00:02] <ckrough> forums aren't rendering """
[22-Sep-2009 09:00:18] <mrayzenoss> new posts or old posts?
[22-Sep-2009 09:00:31] <mrayzenoss> Jive is looking at a way to fix all the old posts
[22-Sep-2009 09:00:41] <ckrough> Does anyone know if "Config Cycle Interval" pushes an entire config or just changes?
[22-Sep-2009 09:00:48] <ckrough> mrayzenoss: old post
[22-Sep-2009 09:00:53] <mrayzenoss> yeah, known issue
[22-Sep-2009 09:00:59] <ckrough> message/27767#27767
[22-Sep-2009 09:01:02] <ckrough> kk
[22-Sep-2009 09:08:39] <rmatte> ckrough: I believe it's a full config
[22-Sep-2009 09:10:11] <ckrough> bleh, brings the master to a halt
[22-Sep-2009 09:13:19] <ckrough> what is "# persistent cache file name " in zenhub.conf exactly?
[22-Sep-2009 09:23:45] <rmatte> that would be a commented line
[22-Sep-2009 09:24:13] <ckrough> heh
[22-Sep-2009 09:24:15] <rmatte> it would most likely refer to a zope cache file (.zec)
[22-Sep-2009 09:24:25] <ckrough> I mean, what exactly are we persistently caching
[22-Sep-2009 09:24:44] <ckrough> hard to tune the setting because I dont know what it's managing
[22-Sep-2009 09:28:27] <rmatte> caching Zope database transactions most likely
[22-Sep-2009 09:28:36] <rmatte> the more common transactions get cached to improve performance
[22-Sep-2009 09:29:38] <ckrough> I think this is all object cache
[22-Sep-2009 09:29:42] <ckrough> not positive
[22-Sep-2009 09:36:51] <rmatte> may be
[22-Sep-2009 10:06:24] NoxDeu is now known as NoxIn
[22-Sep-2009 10:06:50] NoxIn is now known as NoxIn-
\n[22-Sep-2009 10:17:46] <drtns> Morning all I have never had to\ndownload and install a community zenpack before but I find myself\nwanting to install the HP Procurve one. \ndocs/DOC-3486 I am totally new and use to\nRPM's so I am having a harder time even figuring out how to download\nit. is it the zip file at the bottom? how do I grap the "egg"
[22-Sep-2009 10:18:34] <mrayzenoss> weird, it's appears to have been uploaded as a .zip
[22-Sep-2009 10:18:38] <mrayzenoss> I'm looking into it
[22-Sep-2009 10:19:37] <drtns> thanks mrayzenoss
[22-Sep-2009 10:19:45] <mrayzenoss> hmmm... maybe it really is a .zip
[22-Sep-2009 10:19:52] <mrayzenoss> I guess this is an older ZenPack
[22-Sep-2009 10:20:04] <mrayzenoss> .zip style zenpacks were the precursor to .eggs
[22-Sep-2009 10:20:43] <mrayzenoss> so if you go to Settings->ZenPacks
[22-Sep-2009 10:20:59] <mrayzenoss> you can select "Install ZenPack" from the page menu
[22-Sep-2009 10:21:02] <mrayzenoss> and upload it
[22-Sep-2009 10:21:55] <rmatte> mrayzenoss: that zenpack is a zip
[22-Sep-2009 10:22:03] <rmatte> but for some reason the site has zipped up every single zenpack
[22-Sep-2009 10:22:06] <rmatte> even the egg packs
\n[22-Sep-2009 10:22:30] <drtns> I get this when I try and install\n- ERROR: zenpack command failed. Reason: [Errno 20] Not a directory:\n'/opt/zenoss/Products/HPProcurveMap.py/skins'
[22-Sep-2009 10:22:56] <rmatte> drtns: what is the whole filename of the pack?
[22-Sep-2009 10:23:31] <drtns> HPProcurveMap.py
[22-Sep-2009 10:23:51] <rmatte> hunh?
[22-Sep-2009 10:23:58] <rmatte> no, the file that you downloaded the pack as
[22-Sep-2009 10:24:04] <mrayzenoss> drtns: you don't need to unzip it
[22-Sep-2009 10:24:17] <mrayzenoss> oh wait, it's not even a ZenPack
\n[22-Sep-2009 10:24:17] <drtns> The install said to make mappings\nfor Devices and Events... but maybe it needs a Product mapping too?
[22-Sep-2009 10:24:17] <drtns> \\
[22-Sep-2009 10:24:17] <rmatte> oh
[22-Sep-2009 10:24:19] <rmatte> it's HPProcurveMap.py.zip
[22-Sep-2009 10:24:24] <rmatte> what the heck happened there
\n[22-Sep-2009 10:24:27] <etank> i am trying to run wmic from the\nzenoss server but i am getting this as a result 'NTSTATUS: NT code\n0xc002001b - NT code 0xc002001b'
[22-Sep-2009 10:24:40] <mrayzenoss> I'm looking into it
\n[22-Sep-2009 10:24:48] <rmatte> drtns: that's the site's fault\nnot yours, delete that stuff and mrayzenoss will get the proper file up
[22-Sep-2009 10:25:07] <etank> wmic -u 'AD/user%password' //host.domain.com "select name from Win32_Product"
[22-Sep-2009 10:25:13] <etank> that i the query that i am trying
\n[22-Sep-2009 10:25:27] <drtns> mrayzenoss thanks... I just\nwanted to get something that might be a little nicer to monitor my\nswitch then the stock /Network/Switch
[22-Sep-2009 10:25:29] <etank> i can get the data when running wmic locally on the windows 2003 server
[22-Sep-2009 10:26:32] <rmatte> try...
[22-Sep-2009 10:26:45] <rmatte> 'AD/user'%'password'
[22-Sep-2009 10:26:58] <rmatte> actually nevermind
[22-Sep-2009 10:27:07] <rmatte> I'm looking at something different, that won't work
[22-Sep-2009 10:27:08] <rmatte> hmmm
[22-Sep-2009 10:27:34] <rmatte> I think it's -U though
[22-Sep-2009 10:27:37] <rmatte> not -u
[22-Sep-2009 10:28:00] <etank> actually i do have a -U
[22-Sep-2009 10:28:06] <rmatte> k
[22-Sep-2009 10:28:42] <etank> ERROR: Retrieve result data.
[22-Sep-2009 10:28:42] <etank> NTSTATUS: NT code 0x80041001 - NT code 0x80041001
[22-Sep-2009 10:28:47] <etank> taht is the full error
[22-Sep-2009 10:29:09] <rmatte> ah
[22-Sep-2009 10:29:11] <rmatte> try this...
[22-Sep-2009 10:29:23] <rmatte> "Select Name from Win32_Product"
[22-Sep-2009 10:29:24] <etank> it is having a hard time with this "Problem while executing plugin zenoss.wmi.SoftwareMap"
[22-Sep-2009 10:29:32] <rmatte> Linux is quite case sensitive
[22-Sep-2009 10:30:01] <etank> that worked
[22-Sep-2009 10:30:01] <rmatte> oh, well if it's having problems with that then it's most likely some configuration issue on the server itself
[22-Sep-2009 10:30:21] <etank> the same query seemed to work on different machines
\n[22-Sep-2009 10:30:21] <rmatte> you say you can get the data\nlocally, have you tried grabbing the data from another windows box or\nsomething?
[22-Sep-2009 10:30:29] <etank> yes
[22-Sep-2009 10:30:36] <rmatte> ok, so it was your case
[22-Sep-2009 10:30:47] <rmatte> like I said, quite case sensitive
[22-Sep-2009 10:31:03] <etank> first i had to get it so that i could query locally
[22-Sep-2009 10:31:03] <rmatte> unlike windows which doesn't care about case
[22-Sep-2009 10:31:06] <etank> stupi WMI
[22-Sep-2009 10:31:16] <rmatte> lol
[22-Sep-2009 10:31:16] <etank> s/stupi/stupid/
[22-Sep-2009 10:31:27] <rmatte> we just use SNMP to monitor windows servers
[22-Sep-2009 10:31:32] <etank> i had to blow away the wmi repository to get it to work
[22-Sep-2009 10:31:50] <etank> we are using Enterprise so we were testing the functionality of wmi for monitoring
[22-Sep-2009 10:32:14] <rmatte> ah
[22-Sep-2009 10:32:28] <rmatte> we're using core so we use a mix of WMI and SNMP
[22-Sep-2009 10:32:37] <rmatte> WMI just for monitoring services when needed
[22-Sep-2009 10:32:51] <rmatte> we don't even use it for the event console monitoring since it's way too noisy
[22-Sep-2009 10:32:55] <rmatte> would take like a year to tune it
[22-Sep-2009 10:32:56] <rmatte> lol
[22-Sep-2009 10:33:13] <etank> it is very noisy for sure
[22-Sep-2009 10:33:37] <etank> but it did catch an issue that we would not have known about so that is awesome
[22-Sep-2009 10:35:51] <rmatte> what issue?
[22-Sep-2009 10:36:37] <etank> someone turned off one of our domain controllers in india
[22-Sep-2009 10:36:55] <etank> we got alerts from our dc's here in the states that they could not replicate
[22-Sep-2009 10:37:09] <etank> without zenoss monitoring the boxes here we would have missed that
[22-Sep-2009 10:37:18] <rmatte> ah
[22-Sep-2009 10:37:29] <rmatte> well, we do monitor event logs, but with traps
[22-Sep-2009 10:37:36] <rmatte> which only send out for more critical events
[22-Sep-2009 10:38:04]\n<rmatte> then again the event log severity is tunable, but the\nsnmp traps seem to be a bit more intelligent about what issues are\nactually reported
[22-Sep-2009 10:39:16] <mrayzenoss> speaking of events... Jane Curry's encore presentation is in 20 minutes
[22-Sep-2009 10:39:45] <rmatte> it's the exact same presentation as last time, correct?
[22-Sep-2009 10:40:22] <mrayzenoss> yeah, Jane's updated her slides
[22-Sep-2009 10:40:29] <etank> we are going to "borrow" from here papers tomorrow to present on Zenoss to our higher ups
[22-Sep-2009 10:40:36] <mrayzenoss> excellent
[22-Sep-2009 10:40:37] <rmatte> ok, not really any point in me attending this time around then
[22-Sep-2009 10:40:53] <mrayzenoss> all the Zenoss docs are Creative Commons, so feel free to reuse them as needed
[22-Sep-2009 10:42:21] <etank> with any luck we will get the go-ahead for a purchase of enterprise
[22-Sep-2009 10:43:24] <etank> wmi is so stupid
[22-Sep-2009 10:43:46] <etank> it seems that you have to run the queries locally sometimes to get them to work remotely
[22-Sep-2009 10:57:29] <ckrough> mrayzenoss: dial in number for conference?
[22-Sep-2009 10:57:39] <mrayzenoss> it's on the Info tab
[22-Sep-2009 10:57:49] <mrayzenoss> 866-699-3239
[22-Sep-2009 10:57:55] <ckrough> not on the tab
[22-Sep-2009 10:58:03] <ckrough> event number is, phone number not
[22-Sep-2009 10:58:53] <mrayzenoss> Call-in toll-free number (US/Canada): 866-699-3239
[22-Sep-2009 10:58:53] <mrayzenoss> Call-in toll number (US/Canada): 1-408-792-6300
[22-Sep-2009 10:58:53] <mrayzenoss> Access code: 753 961 267
[22-Sep-2009 10:59:00] <ckrough> thanks
[22-Sep-2009 10:59:03] <ckrough> im in
[22-Sep-2009 10:59:45] <mrayzenoss> anyone else joining the webex?
\n[22-Sep-2009 11:10:54] <drtns> Some advice please. I have a web\ncluster behind a load blancer IP.... I want tot use the HTTP monitor\ntemplate to monitor a few vhosts on this cluster for performance. I\nneed to add the load balancer IP as a device to do this but it's not a\n/Server/SOMETHING what is the /HTTP and /WEB classes for? and would\nthey be what I want to use?
[22-Sep-2009 11:11:41] <drtns> I also guess /Ping would be an option?
\n[22-Sep-2009 11:12:20] <rmatte> drtns: you could just create a\ncustom class as well if you wanted, and just go through the zProperties\nfor the class and disable/enable anything you like
\n[22-Sep-2009 11:12:56] <drtns> rmatte yes.. possible but If /HTTP\nand or /WEB or /PING do what I need it may be better then re inventing\nthe wheel
[22-Sep-2009 11:13:12] <drtns> trying to remember how to figure out what they do thogh
[22-Sep-2009 11:13:16] <drtns> though
[22-Sep-2009 11:13:32] <rmatte> well /Ping is for devices that you only want to monitor via ping
[22-Sep-2009 11:13:37] <rmatte> so that would most likely work fine
[22-Sep-2009 11:13:49] <rmatte> then you'd just need to apply the http_monitor template to the device
[22-Sep-2009 11:13:59] <rmatte> and customize it
[22-Sep-2009 11:14:10] <drtns> rmatte yes I was referring mostly to what the /HTTP and /Web classes do.
[22-Sep-2009 11:14:41] <rmatte> we put ours under /Application/HTTP
[22-Sep-2009 11:15:09] <rmatte> since that's the group that that zenpack creates I believe
\n[22-Sep-2009 11:16:28] <drtns> nod I am not even there though. I\nam trying to figure out the best way to get the device in there...\nwhere the device is just an IP bound to a load balancer pool and not a\nphysical server..
[22-Sep-2009 11:16:44] <drtns> if it was just one server that had the sites on it that would be a lot easier.
[22-Sep-2009 11:18:10] <rmatte> I just did this...
[22-Sep-2009 11:18:23] <rmatte> I put the device with the IP in to /Applications/HTTP
[22-Sep-2009 11:18:35] <rmatte> I then went to the device's templates tab and bound the HttpMonitor template
\n[22-Sep-2009 11:19:48] <drtns> right... I need a device to bind\nit to... that's my debate how best to create the device.. I have 8\nservers serving the same website behind a single loadblancer ip. so\nit's not good to bind it to any one of the 8 of those
\n[22-Sep-2009 11:20:35] <drtns> I need to create a device based on\nthe load balancer IP then bind the template to the new device and add\nsites to monitor to that
[22-Sep-2009 11:22:30] <rmatte> I created a local copy of the template and renamed it to whatever
[22-Sep-2009 11:22:30] <rmatte> I then configured it to monitor the specific vhost that I wanted to monitor
[22-Sep-2009 11:22:30] <rmatte> I then created another local copy of the HttpMonitor template and repeated the steps
\n[22-Sep-2009 11:22:30] <rmatte> I now have 4 different\nhttpmonitor templates named different things assigned locally to the\ndevice
[22-Sep-2009 11:22:30] <rmatte> And I have graphs for each
\n[22-Sep-2009 11:22:50] <rmatte> (sorry for the delay, I'm IRCing\nthrough my server at home and my home connection seems to be a bit flaky
[22-Sep-2009 11:25:32] <rmatte> eugh it keeps freezing
[22-Sep-2009 11:26:07] <rmatte2> there, now if it freezes at least I can still chat
\n[22-Sep-2009 11:27:36] <rmatte2> when you setup the datasource on\neach template, just make sure that you only populate the host name\nfield with the hostname you want to monitor
[22-Sep-2009 11:27:44] <rmatte2> don't bother filling out the IP Address field
[22-Sep-2009 11:28:04] <rmatte2> actually, I think I manually created each of these templates come to think of it
[22-Sep-2009 11:28:26] <rmatte2> did add template, then added an HttpMonitor type datasource
[22-Sep-2009 11:29:02]\n<rmatte2> then made the name the name of the vhost, made host\nname the name of the vhost as well, set port to 443 and checked use ssl\n(since it's https).
[22-Sep-2009 11:29:11] <rmatte2> then created a graph
[22-Sep-2009 11:29:14] <rmatte2> repeated for each host
[22-Sep-2009 11:29:34] <rmatte2> one graph for size, and one graph for time for each
[22-Sep-2009 11:30:04] <rmatte2> oh, in the URL field I put the path as well
[22-Sep-2009 11:30:17] <rmatte2> monitoring Zenoss login page so I put: /zport/acl_users/cookieAuthHelper/login_form
[22-Sep-2009 11:30:45] <rmatte2> I set redirect behaviour as follow
[22-Sep-2009 11:30:57] <rmatte2> hope that helps
[22-Sep-2009 11:43:05] <Troubadix09> hello, I have a strange behviour with two performance-graphs polling via snmp from one Windows-Server ...
[22-Sep-2009 11:43:40] <Troubadix09> ups ... behaviour i mean
[22-Sep-2009 11:44:11] <rmatte2> can you describe the issue?
\n[22-Sep-2009 11:45:19] <Troubadix09> rmatte2: always on one\nweekend of a month i have a nightly scheduled restart of some of my\nwindows dcs ...
[22-Sep-2009 11:46:09] <Troubadix09> last weekend after this restart on one windows server two performance graphs stop graphing
[22-Sep-2009 11:47:09] <Troubadix09> at the path of this device ($ZENHOME/perf/Devices/xxxxx/ the two rrd-files are missing
[22-Sep-2009 11:47:40] <Troubadix09> but they shown in Web-GUI
[22-Sep-2009 11:47:50] <Troubadix09> but stop graphing
[22-Sep-2009 11:48:32] <Troubadix09> I stop zenoss, deleting *.zec and *.pi in $ZENHOME/var and start zenoss
[22-Sep-2009 11:48:44] <Troubadix09> ups ... *.pid i mean
[22-Sep-2009 11:49:10] <rmatte2> are there any debug messages in the event console?
\n[22-Sep-2009 11:49:35] <Troubadix09> and after this the two\nperformance-graph start drawing and the missing rrd-files comes back\nagain
[22-Sep-2009 11:50:41] <rmatte2> ok, has this happened more than once?
[22-Sep-2009 11:50:54] <Troubadix09> no, nothing, all worked fine, no error messages in perfsnmp.log
[22-Sep-2009 11:51:10] <rmatte2> ok, well if it has only happened once I'd just write it off as a fluke
[22-Sep-2009 11:51:18] <rmatte2> if it happens again then I'd really look in to it
[22-Sep-2009 11:51:39] <rmatte2> the RRD files shouldn't have gotten magically deleted
\n[22-Sep-2009 11:52:50] <Troubadix09> strange, because the\nrrd-files are missing in the linux-directory and come back, but in the\ngraph there is only a gap from restart the windows-server to restart\nzenoss
[22-Sep-2009 11:53:54] <pokui> hi all,\nI'm looking through the zenoss FAQ and I can't seem to get a set of\ndifferences between the community and enterprise editions. could\nsomeone point me in the right direction?
[22-Sep-2009 11:54:43] <Troubadix09> rmatte2: it happened once as i recall
[22-Sep-2009 11:58:05] <mrayzenoss> pokui: the datasheets are here: http://www.zenoss.com/product/network-management
\n[22-Sep-2009 11:58:26] <drtns> rmatte I created a local copy of\nmy httpd monitor template. if I want to grap 3 sites on the same\nserver do you add 3 data sources to the same template and add them to\nthe graph? or do my bind multiple copies of the httpd monitor template\nand graph each site seperately?
[22-Sep-2009 11:58:44]\n<pokui> mrayzenoss: thanks. ticking off one by one where a\ndifference is mentioned at docs/DOC-2445\nwas becoming tiring.
[22-Sep-2009 11:59:47] <mrayzenoss> you can probably edit and update that doc
[22-Sep-2009 12:02:58] <rmatte2> Troubadix09: you sure you're looking at the right directory for the RRD files?
[22-Sep-2009 12:03:39] <rmatte2> drtns: you can do it with multiple data sources on one template, or multiple templates
[22-Sep-2009 12:03:55] <rmatte2> drtns: I just happened to do it with multiple templates
[22-Sep-2009 12:04:45] <rmatte> hmmm, my home connection seems to have straightened out a bit...
[22-Sep-2009 12:05:17] * rmatte smacks rmatte2 while saying "Stop hitting yourself! Stop hitting yourself!"
[22-Sep-2009 12:05:40]\n<Troubadix09> rmatte2: thats what i thought first too, not\nbelieving my eyes, control it again and ... yes, it was the right\ndirectory.
[22-Sep-2009 12:06:07] <rmatte> Troubadix09: no idea, I've been using Zenoss for almost a year and I've never seen behaviour like that
[22-Sep-2009 12:07:01] <rmatte> It'll be a year next month I believe
[22-Sep-2009 12:07:14] <rmatte> Just in time for King Crab
[22-Sep-2009 12:09:43] * rmatte sneezes
[22-Sep-2009 12:10:40] <Troubadix09> yeah, waiting for 2.5 release ;-)
[22-Sep-2009 12:11:29] <rmatte> yeh, hopefully it's relatively bug free by the time it rolls around
[22-Sep-2009 12:11:43] <rmatte> considering the last 3 releases were bugville
\n[22-Sep-2009 12:13:17] <Troubadix09> rmatte: i agree with you, i\nuse it for a year too, long time with 2.3.3, and work fine
\n[22-Sep-2009 12:13:52] <rmatte> 2.3.3 was fine, the only bug was\nthat it wouldn't properly automatically set product/model info for\ndevices when modelling
[22-Sep-2009 12:14:03] <rmatte> now 2.4.5 does that properly, but you can no longer set the info manually
[22-Sep-2009 12:14:11] <rmatte> I'd rather be able to set it manually than automatically
[22-Sep-2009 12:14:43] <rmatte> quite honestly, I wish we were still at 2.3.3, it would be a lot less headaches lol
\n[22-Sep-2009 12:15:50] <Troubadix09> rmatte: i first install on\nubuntu server with stack-installer, but haven't luck to keep it stable\nwith at least 50 Devices
[22-Sep-2009 12:16:20]\n<rmatte> we're using stack installer on ubuntu and I have 340\ndevices monitored on one of our Zenoss instances
\n[22-Sep-2009 12:17:32] <Troubadix09> rmatte: now i use the\nvmware-appliance, and it would be near by stable working with over one\nhundred devices on this instance :-)
[22-Sep-2009 12:17:58] <Troubadix09> but it cost me some time to customize it to work correctly
[22-Sep-2009 12:18:01] <rmatte> well at least they fixed the issues that I reported with the reports
\n[22-Sep-2009 12:18:42] <rmatte> Troubadix09: It doesn't matter\nwhat platform you run it on, quite honestly, the vmware-appliance is\nprobably the least efficient method of running it in my opinion
[22-Sep-2009 12:18:51] <rmatte> It's all about how you customize it
[22-Sep-2009 12:18:59] <rmatte> and clean up the monitoring as much as possible
\n[22-Sep-2009 12:19:12] <rmatte> if you do absolutely no tuning at\nall then you're obviously going to have performance issues
\n[22-Sep-2009 12:20:22] <rmatte> It also depends on how noisy the\ndevices are, if they are constantly flooding you with traps and syslogs\nthen you're not going to be able to monitor as many devices because\nZenoss is hung up processing all of that stuff
\n[22-Sep-2009 12:22:56] <Troubadix09> yes, i agree with you, i\nonly monitoring what i need, not what is possible, the devices are god\nbeware not so noisy, and the vm running on a high performance\nesx-infrastructure ;-)
[22-Sep-2009 12:23:35] <rmatte> yeh, we have all of our servers on high performance HP blades running ESX
[22-Sep-2009 12:23:58] <rmatte> 15GHz of processing with 32GB of RAM on each blade
[22-Sep-2009 12:24:00] <rmatte> we have 3 of them
[22-Sep-2009 12:26:36] <Troubadix09> rmatte: ok, on me against you its medium-performance ... lol
[22-Sep-2009 12:27:19] <rmatte> hehe
[22-Sep-2009 12:28:00] <Troubadix09> rmatte: we use only 3 hp proliant on a HP EVA4400 for virtualisation, but it seems to be enough for us
[22-Sep-2009 12:28:27] <rmatte> yeh, depends on what you want to run
[22-Sep-2009 12:28:38] <rmatte> we'll eventually have tons of Zenoss instances running off of these
[22-Sep-2009 12:29:31] <rmatte> we have about 30 vms running across all 3 servers right now
[22-Sep-2009 12:29:47] <rmatte> one of the servers only has 1 vm on it though, so the other 2 have about 15 each
[22-Sep-2009 12:30:00] <rmatte> and we're just getting started
[22-Sep-2009 12:30:01] <rmatte>
\n[22-Sep-2009 12:31:53] <rmatte> I wish Chet were around, I have\nsome questions for him about the new interface status column
[22-Sep-2009 12:32:27] <Troubadix09> rmatte: ok, you win ;-)
[22-Sep-2009 12:33:53] <Troubadix09> oh, you test the 2.5 beta?
[22-Sep-2009 12:34:01] <rmatte> yeh
[22-Sep-2009 12:34:18] <rmatte> hopefully beta 3 is a lot less buggy than beta 2
[22-Sep-2009 12:34:30] <rmatte> especially since the final release for King Crab is supposed to be around October 25th
[22-Sep-2009 12:35:23] <Troubadix09> sorry, there is not enough personal at our IT, so i have no time to test it :-(
[22-Sep-2009 12:36:04] <rmatte> I'm just testing it because I want to make sure that it's not going to be a total disaster upgrading to it
[22-Sep-2009 12:36:12] <rmatte> iancmcc: ping
[22-Sep-2009 12:42:05] <Troubadix09> uuii... worktime is long gone, i go home now, cu later
[22-Sep-2009 12:42:20] <rmatte> later
[22-Sep-2009 13:26:35] <djack> hellow world belgium calling
[22-Sep-2009 13:29:45] <rmatte> hey
[22-Sep-2009 13:37:28] <rmatte> man, that hardware model info bug is actually most likely simpler to fix than it seems
\n[22-Sep-2009 13:37:48] <rmatte> it doesn't appear to actually be\na bug with setting the info itself, since if you select stuff and click\nsave it'll apply it to the device
[22-Sep-2009 13:38:04] <rmatte> it seems to be an issue with the dropdown menus on the Edit tab actually pulling up the previous settings
\n[22-Sep-2009 13:38:28] <rmatte> so you can use the edit tab to\nset the model info once, but if you use the edit tab for anything else\nyou have to specifically select the model info again before you click\nsave
[22-Sep-2009 13:38:32] <rmatte> otherwise it'll clear it.
[22-Sep-2009 13:42:44] <rmatte> there
[22-Sep-2009 13:42:45] <rmatte> http://dev.zenoss.org/trac/ticket/5473#comment:9
[22-Sep-2009 13:43:02] <rmatte> hopefully they patch that soon
[22-Sep-2009 13:48:23] <Brad_K> djack: hey!
[22-Sep-2009 13:51:32] <djack> Brad_K: hey there
[22-Sep-2009 13:52:37] <djack> is there a eventviewer portlet?
\n[22-Sep-2009 13:53:40] <Brad_K> i was wondering if anyone had\nexperience editing the XML that is generated by TestGen4Web, so that I\ncan make my script deal with "alert" sheets that Firefox throws up that\nTestGen4Web is not capturing as part of the recording?
[22-Sep-2009 13:54:42] ideopathic_ is now known as ideopathic
[22-Sep-2009 13:58:38] <djack> huh?
[22-Sep-2009 14:12:36] <rmatte> djack: no, there isn't an event viewer portlet
[22-Sep-2009 14:12:49] <mrayzenoss> djack: docs/DOC-3457
[22-Sep-2009 14:13:03] <rmatte> aha
[22-Sep-2009 14:13:11] <rmatte> but apparently Matt's got the goods
[22-Sep-2009 14:13:24] <djack> that i installed today ;-)
\n[22-Sep-2009 14:13:45] <rmatte> I think he's talking about\nsomething more along the lines of something that mimics the event\nconsole
[22-Sep-2009 14:13:56] <rmatte> although that would look so condensed I don't know how useful it'd be
[22-Sep-2009 14:14:08] <mrayzenoss> there are Event Views
[22-Sep-2009 14:14:14] <djack> i only wonder how the group events
[22-Sep-2009 14:14:35] <djack> but not the thing i look for
\n[22-Sep-2009 14:17:13] <Brad_K> In addition to the TestGen4Web\nquestion, does anyone know the default units of measurement for the\n"size" and "time" values in the HttpMonitor? Are those KB and\nmilliseconds?
[22-Sep-2009 14:19:08] <rmatte> Brad_K: correct
\n[22-Sep-2009 14:20:20] <Brad_K> weird, because\ndocs/DOC-3234 says "seconds" and "bytes",\nbut the numbers I'm getting just don't make any sense at all with those\nunits....
[22-Sep-2009 14:21:40] <Brad_K> hmm. in fact, KB doesn't make sense, either. ;-(
[22-Sep-2009 14:22:37] <rmatte> how so?
[22-Sep-2009 14:22:40] <rmatte> makes sense for mine
\n[22-Sep-2009 14:23:25] <Brad_K> the page in question isn't\nanywhere close to 244KB in size, and the value of "size" has never\nvaried from that number. But the page is definitely larger than 244\nbytes.
[22-Sep-2009 14:23:57] <Brad_K> 16KB is more like it.
[22-Sep-2009 14:24:01]\n<rmatte> for time on one of my graphs I'm seeing cur as being\n95.11, avg as 191.49m and max as 4.40, meaning the longest it took was\n4.40 seconds, the average is 191.49 milliseconds, and the current is\n95.11 milliseconds
[22-Sep-2009 14:24:06] <rmatte> which makes perfect sense
[22-Sep-2009 14:24:52] <rmatte> size is showing as about 4 kilobytes for all of them, which sounds about right as well
[22-Sep-2009 14:26:57] <rmatte> sorry, cur for time is 95.11m not 95.11
[22-Sep-2009 14:28:56] <rmatte> afk for a few
\n[22-Sep-2009 14:31:25] <Brad_K> i was wondering if anyone had\nexperience editing the XML that is generated by TestGen4Web, so that I\ncan make my script deal with "alert" sheets that Firefox throws up that\nTestGen4Web is not capturing as part of the recording?
[22-Sep-2009 14:32:42] mrayzenoss1 is now known as mrayzenoss
[22-Sep-2009 14:45:44] <djack> you should do an event in belgium :-p
[22-Sep-2009 14:46:43] <mrayzenoss> we've talked about attending FOSDEM
[22-Sep-2009 15:05:00] <rmatte> well that was fun
[22-Sep-2009 15:05:08] <rmatte> lost our internet connection for like 25 mins
[22-Sep-2009 15:20:01] <rhettardo> did the isp blame you for the outage like they always ddo, or deny it like they always do?
[22-Sep-2009 15:20:37] <rhettardo> 'its your router', 'we dont show any issues'
[22-Sep-2009 15:21:08] <rmatte> nah, it's a circuit going from here to the datacenter that we host our equipment at
[22-Sep-2009 15:21:19] <rmatte> and one of the guys on our team's roommate works at the datacenter
[22-Sep-2009 15:21:30] <rmatte> so he just called him up and they told us that one of their switches had died
[22-Sep-2009 15:22:47] <rhettardo> ah
\n[22-Sep-2009 15:25:34] <ke4qqq> hey guys - I have a two cisco\nASAs (a 5505 and 5510) pings to the devices work, but I am getting snmp\nerrors. I tried using snmpwalk from the zenoss box to each ASA and I\nget responses - any idea what could be causing the problem?
[22-Sep-2009 15:26:17] <rmatte> snmp errors in Zenoss?
[22-Sep-2009 15:26:45] <rmatte> like, an snmpwalk works fine but Zenoss is complaining?
[22-Sep-2009 15:26:52] <ke4qqq> rmatte: yes - zenoss reporting SNMP agent down for just those devices
[22-Sep-2009 15:26:53] <ke4qqq> yes
[22-Sep-2009 15:27:01] <rmatte> great, then what you want to do is this...
[22-Sep-2009 15:27:01] <ke4qqq> snmpwalk works fine, but zenoss complains
\n[22-Sep-2009 15:27:29] <rmatte> go to the zProperties for those\ndevices (don't do this for every device in Zenoss), and change\nzMaxOIDPerRequest from 40 to 20 or 10
[22-Sep-2009 15:27:46] <rmatte> Cisco ASAs seems to be picky about the size of packets being sent to them sometimes
[22-Sep-2009 15:28:03] <rmatte> so you need to lower the amount of data that it's trying to poll per packet
[22-Sep-2009 15:28:33] <ke4qqq> awesome - I'll create an article on the wiki to document this
[22-Sep-2009 15:28:35] <ke4qqq> thanks!
[22-Sep-2009 15:28:43] <rmatte> np, make sure it works first
[22-Sep-2009 15:28:51] <rmatte> it worked for me when I was having issues in the past, so it should
[22-Sep-2009 15:28:56] <rmatte> but that may not be your exact issue
[22-Sep-2009 15:29:02] <rmatte> I'm almost certain that it is though
[22-Sep-2009 15:29:04] <ke4qqq> ok
[22-Sep-2009 15:32:23] <chemist> good evenng
[22-Sep-2009 15:32:34] <chemist> *evening even
[22-Sep-2009 15:32:43] <rmatte> hey
[22-Sep-2009 15:32:59] <ke4qqq> rmatte that fixed it - thanks!
[22-Sep-2009 15:32:59] <chemist> we are trying to model some netgear switches and we keep getting this error
\n[22-Sep-2009 15:33:10] <chemist> Unable to get data for XXXX from\neither ipAddrTable or ipNetToMediaTable -- skipping model
[22-Sep-2009 15:33:28] <rmatte> ke4qqq: np
[22-Sep-2009 15:33:29] <chemist> any ideas?
[22-Sep-2009 15:33:52] <rmatte> chemist: you sure your snmp permissions are correct on the devices?
[22-Sep-2009 15:33:57] <chemist> yes
[22-Sep-2009 15:33:58] <rmatte> sounds like they aren't
[22-Sep-2009 15:34:16] <chemist> positive
[22-Sep-2009 15:34:16] <rmatte> try doing: snmpwalk -v1 -c bleh bleh.com ipAddrTable
[22-Sep-2009 15:34:20] <rmatte> see if it returns anything
[22-Sep-2009 15:35:15] <rmatte> should be some lines like:
[22-Sep-2009 15:35:16] <rmatte> IP-MIB::ipAdEntAddr.30.30.1.25 = IpAddress: 30.30.1.25
[22-Sep-2009 15:35:16] <rmatte> IP-MIB::ipAdEntAddr.30.30.2.25 = IpAddress: 30.30.2.25
[22-Sep-2009 15:35:16] <rmatte> IP-MIB::ipAdEntAddr.30.30.3.25 = IpAddress: 30.30.3.25
\n[22-Sep-2009 15:35:35] <rmatte> If you don't see anything like\nthat returned then your permissions aren't setup correctly or you're\nusing the wrong community string in Zenoss
[22-Sep-2009 15:43:25] <chemist> rmatte: No, checked, rechecked and triple checked
[22-Sep-2009 15:43:44] <rmatte> chemist: so if you snmpwalk like I said you get output returned?>
[22-Sep-2009 15:43:46] <chemist> permissions are fine and the community string is ok
[22-Sep-2009 15:44:11] <chemist> if I snmpwalk without ipAddrTable, yes
[22-Sep-2009 15:44:40] <rmatte> no, I mean with ipAddrTable
[22-Sep-2009 15:44:58] <chemist> with ipAddrTable I get nothing
[22-Sep-2009 15:45:04] <rmatte> that's what it's complaining about, that it's not getting any output returned from walking that value
[22-Sep-2009 15:45:20] <chemist> but with any other value it works
\n[22-Sep-2009 15:45:34] <rmatte> the only time I've seen\ncomplaints about those values was on linux boxes where the snmp\npermissions weren't properly configured
[22-Sep-2009 15:45:47] <rmatte> what device group do you have the devices in?
[22-Sep-2009 15:46:13] <chemist> '/network/switch'
[22-Sep-2009 15:47:08] <rmatte> If you go to /Network/Switch then select More -> Collector Plugins
\n[22-Sep-2009 15:47:30] <rmatte> do you only see\nzenoss.snmp.NewDeviceMap, zenoss.snmp.DeviceMap, and\nzenoss.snmp.InterfaceMap?
[22-Sep-2009 15:48:27] <chemist> and RouteMap
[22-Sep-2009 15:49:12] <rmatte> try removing routemap from the list
[22-Sep-2009 15:49:15] <rmatte> then try again
[22-Sep-2009 15:50:26] <rmatte> ah
[22-Sep-2009 15:50:26] <rmatte> http://74.125.113.132/search?q=cache:DQ2LqdM5h6sJ:forums.zenoss.com/viewtopic.php%3Ft%3D10007%26sid%3D855c0e0a58d7c8a363483cd6ecbf0041+zenoss+Unable+to+get+data+for+ipAddrTable&cd=1&hl=en&ct=clnk&gl=ca
[22-Sep-2009 15:50:29] <rmatte> that is basically what you're experiencing
[22-Sep-2009 15:50:32] <rmatte> and at the end...
\n[22-Sep-2009 15:50:40] <rmatte> So the ipAddrTable or\nipNetToMediaTable are some mibs that the switch is missing. The other\nworking switch was running a new version of code, which I've downgraded\nand I'm still able to pull interface stats fine.
\n[22-Sep-2009 15:50:41] <rmatte> So now I have a new question, how\ndo I get the zen modeler to skip the ipAddrTable or ipNetToMediaTable\nwhen building the OS interfaces?
[22-Sep-2009 15:50:59] <rmatte> now I just need to figure out how to get it to ignore those values
[22-Sep-2009 15:51:21] <rmatte> although the ipAddrTable values are fairly important
[22-Sep-2009 15:54:18] <chemist> I guess I'll tell my boss we need to replace 40+ switches
[22-Sep-2009 15:54:39] <rmatte> ah, it's because it's a layer 2 switch
\n[22-Sep-2009 15:54:45] <rmatte>\nthread/10554%3Bjsessionid=1787F3F07DB545A3108BB042D9916094.node0?decorator=print&displayFullThread=true
[22-Sep-2009 15:56:48] <chemist> yeah, we are chepscapes
[22-Sep-2009 15:56:55] <chemist> *cheapscapes
\n[22-Sep-2009 15:56:57] <rmatte> From the sounds of it you may\nneed to edit\n$ZENHOME/Products/DataCollector/plugins/zenoss/snmp/InterfaceMap.py to\nhave it skip the lines that collect that data
[22-Sep-2009 15:57:09] <rmatte> I can't remember the last time I've seen a layer 2 switch
[22-Sep-2009 15:57:42] <chemist> I'll show this conversation to my boss
[22-Sep-2009 15:57:57] <rmatte> lol
[22-Sep-2009 15:58:14] <rmatte> If you have the money it would really be worth grabbing some Cisco or HP switches
[22-Sep-2009 15:58:22] <rmatte> bring yourselves in to the 21st century
[22-Sep-2009 15:58:25] <rmatte>
\n[22-Sep-2009 15:59:23] <ke4qqq> rmatte: layer 2 switchs are still\ncommon - cisco 2970s for instance. moving to 3750s for every switch\nwould be pretty heady for most
[22-Sep-2009 15:59:59] <rmatte> I suppose
[22-Sep-2009 16:00:13] <rmatte> actually, 2970s might not even have this problem
[22-Sep-2009 16:00:17] <ke4qqq> but L2 switches are including more and more L3 features
[22-Sep-2009 16:00:38] <ke4qqq> but since they don't really route it's not an issue
[22-Sep-2009 16:01:01] <ke4qqq> I have some old 2948Gs that don't exihibit that problem (along with 3coms and HPs.)
[22-Sep-2009 16:01:26] <rmatte> hmmm, no 2970s, only 297024
[22-Sep-2009 16:01:35] mrayzenoss1 is now known as mrayzenoss
[22-Sep-2009 16:02:18] <ke4qqq> yeah most of my switches are relatively ancient.....
[22-Sep-2009 16:02:38] <rmatte> hehe
[22-Sep-2009 16:02:59] <rmatte> chemist: you should log a Trac case for that
[22-Sep-2009 16:03:20] <rmatte> It is pretty dumb that it can't do it
[22-Sep-2009 16:04:56] <chemist_> or use cacti plugins
[22-Sep-2009 16:05:21] <chemist_> cacti reports them properly
[22-Sep-2009 16:05:56] <rmatte> yeh, but the whole point of a trac case is to eventually get that fixed
[22-Sep-2009 16:06:05] <rmatte> instead of just saying "meh, I'll use something else"
[22-Sep-2009 16:06:08] <rmatte>
[22-Sep-2009 16:06:17] <rmatte> up to you
[22-Sep-2009 16:06:31] <chemist_> good point
[22-Sep-2009 16:06:57] <rmatte> If you don't have a trac account ask Matt to make you one
[22-Sep-2009 16:06:58] <chemist_> and I would rather use zenoss for the whole enterprise
[22-Sep-2009 16:07:11] <rmatte> http://dev.zenoss.org/trac
[22-Sep-2009 16:07:15] <chemist_> mrayzenoss?
[22-Sep-2009 16:07:23] <pokui> chemist_: sorry I missed part of this discussion, what make/model are these switches again?
[22-Sep-2009 16:07:40] <chemist_> netgear FS726T
[22-Sep-2009 16:07:47] <mrayzenoss> yeah, I can make Trac accounts, or you can just use zenoss/zenoss
[22-Sep-2009 16:07:56] <pokui> chemist_: ah. thanks.
[22-Sep-2009 16:07:59] <rmatte> ah, didn't know there was a generic one
[22-Sep-2009 16:08:31] * pokui makes a note to test with level one swtiches.
\n[22-Sep-2009 16:10:58] <rmatte> chemist_: so yeh, go to\nhttp://dev.zenoss.org/trac and login with zenoss/zenoss (or ask Matt to\nmake you your own account) and log the issue
[22-Sep-2009 16:11:11] <chemist_> will do
\n[22-Sep-2009 16:11:20] <rmatte> I just find that it's nice to\nhave your own account since you can better track the status of your\ntickets
[22-Sep-2009 16:11:33] <chemist_> I'm surprised this hasn't come up before though
[22-Sep-2009 16:11:51] <chemist_> unless no one else buys these cheap switches
\n[22-Sep-2009 16:12:17] <Brad_K> chemist_: it has come up before,\nthat's why we can provide these URLs that quickly describe the problem.
[22-Sep-2009 16:12:44] <rmatte> ke4qqq: ah, you help out with fedora?
\n[22-Sep-2009 16:12:52] <Brad_K> the issue is that no one (so far)\nhas apparently created a trac ticket for this problem. so far, they've\napparently always decided to live with the issue, or use cacti plugins,\nor whatever.
[22-Sep-2009 16:13:04] <rmatte> chemist_: I did a search and it actually came up on 3 different occasions in the old forums
[22-Sep-2009 16:13:07] <chemist_> ouch!!
\n[22-Sep-2009 16:13:25] <rmatte> but there was never really much\nof a solution other than modifying the\n$ZENHOME/Products/DataCollector/plugins/zenoss/snmp/InterfaceMap.py file
[22-Sep-2009 16:14:12] <chemist_> but that's not really a fix
[22-Sep-2009 16:14:24] <ke4qqq> rmatte: well I at least try not to hinder Fedora
[22-Sep-2009 16:14:37] <rmatte> hehe
[22-Sep-2009 16:14:40] <chemist_> hacks like that have a tendency to come back and bite you
[22-Sep-2009 16:14:49] <rmatte> chemist_: yup, they do indeed
[22-Sep-2009 16:14:56] <Brad_K> which is why someone needs to create a trac ticket. ;-)
[22-Sep-2009 16:14:59] <ke4qqq> yes, hence the need to create a ticket and get zenoss to fix the problem upstream
[22-Sep-2009 16:14:59] <rmatte> but yeh, log a ticket and I'm sure it'll get looked at
[22-Sep-2009 16:15:04] <rmatte> because that's pretty major
[22-Sep-2009 16:15:29] * chemist_ is reading how to create a trac ticket
[22-Sep-2009 16:15:42] <rmatte> it's just wiki formatting
[22-Sep-2009 16:15:56] <rmatte> click on new ticket in the top right
[22-Sep-2009 16:16:09] <rmatte> then populate the template it provides
[22-Sep-2009 16:16:16] <rmatte> so * Build Version: ALL
[22-Sep-2009 16:16:24] <rmatte> *Operating System(s): ALL
[22-Sep-2009 16:16:29] <rmatte> Steps to Reproduce...
[22-Sep-2009 16:16:31] <rmatte> 1. bleh
[22-Sep-2009 16:16:33] <rmatte> 2. bleh
[22-Sep-2009 16:16:37] <rmatte> etc...
[22-Sep-2009 16:16:42] <rmatte> Results
[22-Sep-2009 16:16:43] <rmatte> whatever
[22-Sep-2009 16:16:47] <rmatte> expected results
[22-Sep-2009 16:16:52] <rmatte> whatever was expected
[22-Sep-2009 16:16:54] <rmatte> Impact
[22-Sep-2009 16:16:58] <rmatte> how it impacts the users
[22-Sep-2009 16:17:16] <rmatte> then change the priority if you feel it's higher priority than the default of 3
[22-Sep-2009 16:17:21] <rmatte> and click Create Ticket
[22-Sep-2009 16:17:23] <rmatte> then you're done
\n[22-Sep-2009 16:17:40] <rmatte> there are other values that can\nbe changed as well such as version, but the default should suffice for\nthis
[22-Sep-2009 16:17:43] <Brad_K> rmatte:\ncareful, you might actually get tickets with steps like "1. bleh\n2.bleh" that are actually created. ;-)
[22-Sep-2009 16:18:05] <rmatte>
[22-Sep-2009 16:18:13] <rmatte> only if people are completely retarded
[22-Sep-2009 16:18:21] <chemist_> I might add some blah's for colour
[22-Sep-2009 16:18:35] <rmatte>
[22-Sep-2009 16:18:41] <Brad_K> is there irony in being a literalist contrarian?
[22-Sep-2009 16:18:50] <rmatte> I still need to open that trac case demanding MOAR PONIES! in King Crab
[22-Sep-2009 16:19:07] <rmatte> Brad_K: I'd say so
[22-Sep-2009 16:19:09] <Brad_K> or MO4R C0WB3LL!
[22-Sep-2009 16:19:18] <rmatte> w000000h!
[22-Sep-2009 16:20:25] * Brad_K prefers more c0ffee....
[22-Sep-2009 16:20:31] <rmatte> they really need to replace the "View Event History" link with "Travel to Candy Mountain"
[22-Sep-2009 16:20:34] <rmatte> sounds more enticing
[22-Sep-2009 16:21:05] <rmatte> or Beer Volcano
[22-Sep-2009 16:21:06] <rmatte> either way
[22-Sep-2009 16:21:28] * Brad_K contemplates Beer and Mentos....
[22-Sep-2009 16:21:34] <rmatte> hehe
[22-Sep-2009 16:21:37] <mrayzenoss> mmm... skittlebrau
[22-Sep-2009 16:22:08] <Brad_K> would you eat the Mentos first, then drink the beer? And then what would happen inside you?
[22-Sep-2009 16:22:20] <rmatte> you'd die
[22-Sep-2009 16:23:01]\n<Brad_K> or maybe you'd just have fountains of beer coming out\nyour nose? kind of like an updated C|N>K type of moment?
\n[22-Sep-2009 16:24:48] <rmatte> I just like the whole\npastafarianist belief that when you die you travel to a beer volcano\nnext to a stripper factory
[22-Sep-2009 16:25:04]\n<rmatte> and if you do evil things, you still travel to a beer\nvolcano next to a stripper factory, but the beer is flat!
[22-Sep-2009 16:25:39] * Brad_K is down with the FSM
[22-Sep-2009 16:25:45] <rmatte>
[22-Sep-2009 16:26:00] <rmatte> his noodliness
[22-Sep-2009 16:26:20] <Brad_K> However, my religion is closer to Jedi than Pastafarianism.
[22-Sep-2009 16:26:37] <Brad_K> btw, you did hear that Jedi is now an official religion in the UK, right?
[22-Sep-2009 16:26:44] <rmatte> yeh
[22-Sep-2009 16:26:55] <rmatte> pretty dumb though
\n[22-Sep-2009 16:27:50] <Brad_K> well, we don't necessarily\nbelieve in all the "new trilogy" material, so perhaps that would be\nJedi Fundamentalist?
[22-Sep-2009 16:28:05] <rmatte> lol
[22-Sep-2009 16:28:10] <Brad_K> or Traditional Jeditarian?
[22-Sep-2009 16:28:26] <rmatte> Jedihova's witnesses?
[22-Sep-2009 16:28:29] <Brad_K> no, no, no -- Orthodox Jeditarianism...
[22-Sep-2009 16:28:59] <Brad_K> at least we're not the Yoda's Witnesses....
[22-Sep-2009 16:29:03] <rmatte> lol
[22-Sep-2009 16:29:23] <rmatte> wouldn't Yoda be equivalent to Jesus?
[22-Sep-2009 16:29:27] <rmatte> in that religion
\n[22-Sep-2009 16:30:12] <rmatte> Once I start seeing people\ndriving around with Yoda Fish on their car, then I'll be amazed
\n[22-Sep-2009 16:30:17] * Brad_K ponders on the concept of Yoda being\nhis own son, and the kind of whacky time-travel that would be required\nto make that happen....
[22-Sep-2009 16:30:48] <rmatte> no need for time travel, just advanced virtual reality
[22-Sep-2009 16:30:51] <rmatte> holodeck
[22-Sep-2009 16:30:53] <rmatte>
[22-Sep-2009 16:31:09] <Brad_K> with Ponoko, you can make your own Jedifish or Yodafish, etc....
[22-Sep-2009 16:31:32] <rmatte> Ok, opinion...
[22-Sep-2009 16:31:37] <rmatte> http://i571.photobucket.com/albums/ss159/KalynSS999/Paintball%20-%20Guns/EgoParts2.jpg
[22-Sep-2009 16:31:38] <rmatte> http://i420.photobucket.com/albums/pp284/SADcasper21/DSCN0448.jpg
[22-Sep-2009 16:31:42] <rmatte> which color is nicer
[22-Sep-2009 16:31:55] <rmatte> the red one has a black tip (can't see it in the photo)
[22-Sep-2009 16:32:30] <Brad_K> for what purpose? i like them both....
[22-Sep-2009 16:32:47] <rmatte> no I'm literally buying that model of paintball gun lol
[22-Sep-2009 16:33:10] <rmatte> I'm sort of leaning towards the brown/bright blue one
\n[22-Sep-2009 16:33:28] <Brad_K> i figured, but you still have to\nanswer the question of purpose. would you be a member of a team where\nthe colors in your marker would make a difference?
[22-Sep-2009 16:33:37] <rmatte> oh, no
[22-Sep-2009 16:33:40] <rmatte> it's just for style
[22-Sep-2009 16:34:15] <Brad_K> is it brown and blue, or copper and blue? it's hard to tell from the picture....
[22-Sep-2009 16:34:24] <rmatte> it's copper and blue
[22-Sep-2009 16:34:29] <rmatte> quite dark in that photo
[22-Sep-2009 16:35:19] <rmatte> http://media.photobucket.com/image/ego%20lagoon/burressfan1/IMG_0340.jpg
\n[22-Sep-2009 16:35:26] <Brad_K> i like them both. the red &\nblack would be a bit more traditional pairing of colors, while the\ncopper and blue would be a more trendy pairing. but both are good\ncombinations.
[22-Sep-2009 16:35:47] <rmatte> yeh, hmmm
[22-Sep-2009 16:36:21] <rmatte> My mask which arrives tomorrow: http://my.tbaytel.net/dementia/images/lazarus.jpg
[22-Sep-2009 16:36:41] <rmatte> can't wait
[22-Sep-2009 16:36:42] <rmatte> hehe
[22-Sep-2009 16:36:51] <Brad_K> that's pretty scary. you should wear that on halloween.
[22-Sep-2009 16:36:55] <rmatte> hehe
[22-Sep-2009 16:37:15] <rmatte> it's hand airbrushed by a company called save phace
[22-Sep-2009 16:37:21] <rmatte> they make some pretty neat looking masks
\n[22-Sep-2009 16:37:24] <chemist_> ok, I created the trac ticket\nand in the meantime my colleague hacked the InterfaceMap.py file so we\ncan model the switches
[22-Sep-2009 16:37:37] <rmatte> chemist_: good stuff
[22-Sep-2009 16:37:50] <chemist_> moving on to the next problem....
[22-Sep-2009 16:37:51] <Brad_K> chemist_: thanks!
[22-Sep-2009 16:38:13] <chemist_> zenoss has decided to stop sending emails
[22-Sep-2009 16:38:28] <chemist_> I can send a test email but I don't get any notifications
[22-Sep-2009 16:39:10] <rmatte> have you tried restarting zenoss?
[22-Sep-2009 16:39:44] <chemist_> yes
[22-Sep-2009 16:40:45] <rmatte> hmmm
[22-Sep-2009 16:42:15] <rmatte> not sure, if you check the mail logs you don't see anything?
\n[22-Sep-2009 16:54:41] <chemist_> rmatte: tbh I don't know how\nzenoss sends email, wether it uses sendmail or it's own internal\nmechanics
[22-Sep-2009 16:55:09] <rmatte> it's own internals
[22-Sep-2009 16:55:31] <chemist_> that's why I can't find anything in the logs
[22-Sep-2009 16:56:24] <rmatte> actually...
[22-Sep-2009 16:56:35] <rmatte> if you go to settings, what smtp host do you have set?
[22-Sep-2009 16:56:41] <rmatte> if you can get in to that host and check those logs
[22-Sep-2009 16:57:23] <chemist_> it's an exchange server, and I can receive email when I send a test
[22-Sep-2009 16:57:49] <rmatte> i see
[22-Sep-2009 16:58:09] <rmatte> and the alerts are setup under your account?
[22-Sep-2009 16:58:26] <chemist_> yyes
[22-Sep-2009 16:58:37] <chemist_> it did literally stop working
[22-Sep-2009 16:58:45] <rmatte> you're sure you don't have some weird filter option in there?
[22-Sep-2009 16:59:28] <chemist_> all I did was create a rule for a different account, then deleted the rule
[22-Sep-2009 16:59:36] <chemist_> since then, no emails
[22-Sep-2009 16:59:53] <rmatte> weird, I'm not sure
[22-Sep-2009 17:00:02] <rmatte> check all of the log files
[22-Sep-2009 17:00:06] <chemist_> neway, time to go home
[22-Sep-2009 17:00:17] <chemist_> thanks for the help
[22-Sep-2009 17:00:41] <rmatte> np
\n[22-Sep-2009 17:42:25] <wildcard0> hey. my disk stats are\ngetting weird replies like negative free space. i remodeled and it\nhasn't changed. can anyone point me in the right direction? i see\nvery little on the board or google
[22-Sep-2009 17:43:15] <mrayzenoss> wildcard0: really large disks?
[22-Sep-2009 17:44:05]\n<wildcard0> no, i thought it may be wrapping, but this one's a\n/boot partition and it's 100M total. zenoss reports it as 12G though
[22-Sep-2009 17:44:36] <mrayzenoss> hmm.. the bug I was thinking of is for 12TB+
[22-Sep-2009 17:44:44] <mrayzenoss> SNMP or SSH?
[22-Sep-2009 17:44:47] <mrayzenoss> or WMI?
[22-Sep-2009 17:44:50] <wildcard0> snmp
[22-Sep-2009 17:44:59] <wildcard0> it's odd that it's reporting 12GB though
[22-Sep-2009 17:45:05] <wildcard0> if there's a bug at that point
[22-Sep-2009 17:45:27] <mrayzenoss> the bug is at 12 terabytes
[22-Sep-2009 17:45:35] <wildcard0> oh TB
[22-Sep-2009 17:45:37] <wildcard0> heh i can read
[22-Sep-2009 17:45:56] <wildcard0> i see no errors in the logs at all either
\n[22-Sep-2009 17:46:23] <wildcard0> though i dont see the log that\nreturns snmp info specifically. so either i dont have logging up high\nenough or that isn't logged
[22-Sep-2009 17:46:37] <mrayzenoss> message/28849 ?
\n[22-Sep-2009 17:47:46] <wildcard0> ya i saw that. snmp looks\nlike it's giving out sane values and there's no real resolution on that\nthread
[22-Sep-2009 17:47:47] <mrayzenoss> I gotta run, but searching for "negative" pulled up a few things
[22-Sep-2009 17:48:07] <wildcard0> thanks
[22-Sep-2009 17:48:13] <mrayzenoss> maybe post to that thread? or follow those debug tips a bit and see what turns up?
[22-Sep-2009 17:49:05] <wildcard0> ya i'll keep hammering at it. im sure it'll be something dumb
[22-Sep-2009 17:49:12] <mrayzenoss> later
[22-Sep-2009 18:32:43] <wildcard0> hey. how do i restart just zenperfsnmp without restarting all of zenoss?
[22-Sep-2009 18:49:35] <ocnam> think its in settings > daemons
[22-Sep-2009 18:50:00] <wildcard0> thanks
[22-Sep-2009 21:31:03] straterr1 is now known as straterra
[23-Sep-2009 00:00:46] [disconnected at Wed Sep 23 00:00:46 2009]
[23-Sep-2009 00:00:46] [connected at Wed Sep 23 00:00:46 2009]
[23-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[23-Sep-2009 02:01:17] <Troubadix09> hi all
[23-Sep-2009 08:10:03] <Mituc> hello everyone
[23-Sep-2009 08:12:02]\n<Mituc> I just upgraded from 2.4.2 to 2.4.5 + applied the patchez\nfor zenocommand (that will be included in 2.4.6), and after I did that\nzencommand did not execute any other plug-ins, but gave an error instead
[23-Sep-2009 08:12:20] <Mituc> anybody else experiencing the same issues?
[23-Sep-2009 08:12:36] <Mituc> I had to downgrade back to 2.4.2, since the upgrade to 2.4.5 + patching did not work..
[23-Sep-2009 08:13:05]\n<Mituc> and 2.4.2 gives a lot of command timeouts, events remain\nstuck in an open state without the state being reset, and so on
[23-Sep-2009 08:13:39] <Mituc> wonder if the database was affected by the upgrade+downgrade..
\n[23-Sep-2009 08:18:51] <Mituc> also, if I take a copy of\nvar/Data.* and perf/ directories/files and restore it with a fresh copy\nof zenoss and events database, will that work ?
[23-Sep-2009 08:22:25] <Mituc> ok, so I'll just try that then
[23-Sep-2009 09:18:06] <Egyptian[web]> hi
\n[23-Sep-2009 09:18:27] <rmatte> anyone who can tell me why my\nVista install is eating up 18.2GB of space when I only have 9.79GB of\nfiles on the drive wins a cookie
[23-Sep-2009\n09:18:39] <Egyptian[web]> i just started with zenoss and did an\nautodiscover .. but now i want to change the device's "name" from ip to\nhostname... hw?
[23-Sep-2009 09:19:15] <Egyptian[web]> rmatte: have you tried #windows?
\n[23-Sep-2009 09:19:52] <rmatte> Egyptian, go to the device page,\nclick on the dropdown arrow in the top left, select Manage -> Rename\nDevice
[23-Sep-2009 09:19:58] <rmatte> and yes, thank you, #windows is quite obvious
\n[23-Sep-2009 09:20:39] <rmatte> I've googled to high heaven for a\nsolution, I doubt that #windows is going to serve any better
[23-Sep-2009 09:20:43] <rmatte>
[23-Sep-2009 09:22:30] <rmatte> People in here tend to know little tricks for this sort of thing from time to time
\n[23-Sep-2009 09:23:20] <Egyptian[web]> rmatte: then .. maybe you\nshould do what i do .. load a live linux cd and explore the drive ?
[23-Sep-2009 09:25:48] <Egyptian[web]> and thanks .. i renamed the device
\n[23-Sep-2009 09:26:49] <rmatte> Egyptian[web]: it's a VM running\nin VirtualBox on top of linux already lol, I suppose I could download a\nlivecd ISO and mount it
[23-Sep-2009 09:27:18] <Egyptian[web]> rmatte: indexing service ?
[23-Sep-2009 09:27:37] <Egyptian[web]> rmatte: system backup i think its called .. you know .. autorestore thingy
[23-Sep-2009 09:27:41] <rmatte> I've checked indexing, shadow files, etc... (all the common stuff)
[23-Sep-2009 09:27:49] <rmatte> that's why I'm stumped
\n[23-Sep-2009 09:28:15] <rmatte> system restore wouldn't use that\nmuch space, it basically just takes a snapshot like you would in vmware
[23-Sep-2009 09:28:43] * Egyptian[web] shrugs
[23-Sep-2009 09:29:14] <Egyptian[web]> live cd and find files bigger than 1 gb
\n[23-Sep-2009 09:29:17] <rmatte> meh, I hate vista anyways, if it\nstarts becoming a big deal I'll just reformat and throw XP on
\n[23-Sep-2009 09:29:43] <Egyptian[web]> hokay . time to go home ..\ni will be dropping in quite a bit now that i have zenoss running
[23-Sep-2009 09:29:58] <Egyptian[web]> salam
[23-Sep-2009 09:54:12] danuvius_ is now known as danuvius
[23-Sep-2009 10:08:08] <twm1010> where are filesystem templates kept?
[23-Sep-2009 10:37:07] <rmatte> twm1010: /Devices/Server
[23-Sep-2009 10:37:09] <rmatte> templates
[23-Sep-2009 10:37:13] <rmatte> FileSystem
[23-Sep-2009 10:37:33] <rmatte> that's the single template used for everything
[23-Sep-2009 10:37:39] <rmatte> you can make local copies of it as necessary
[23-Sep-2009 10:37:49] <rmatte> for adjusting thresholds or whatever
[23-Sep-2009 11:51:23] <twm1010> rmatte: and that has special inheritance like the interface templates?
[23-Sep-2009 11:51:30] <rmatte> no
[23-Sep-2009 11:51:35] <rmatte> general inheritance
[23-Sep-2009 11:51:46] <rmatte> the same FileSystem template is used for everything
[23-Sep-2009 11:53:52] <twm1010> but i can make copies in classes right?
[23-Sep-2009 11:59:03] <rmatte> yeh, if you want to
[23-Sep-2009 11:59:10] <rmatte> just make a local copy of it wherever
\n[23-Sep-2009 11:59:27] <rmatte> you can make local copies of it\nin classes, or make copies of it directly on hard drive objects
[23-Sep-2009 11:59:28] <rmatte> up to you
[23-Sep-2009 12:05:06] <twm1010> Just find it odd that some of my low disk space events are errors and some are warnings
[23-Sep-2009 12:12:58] <rmatte> twm1010: you might have some local copies
[23-Sep-2009 12:13:13] <rmatte> Go to /Devices/Server and do...
[23-Sep-2009 12:13:28] <rmatte> More -> All Templates
[23-Sep-2009 12:13:36] <rmatte> it'll list all of the templates under /Devices/Server
[23-Sep-2009 12:13:39] <rmatte> including local
[23-Sep-2009 12:13:54] <rmatte> also, your threshold might be set to auto-escalate events
[23-Sep-2009 12:13:58] <rmatte> another thing to check
[23-Sep-2009 12:14:16] <rmatte> oh, and if you have transforms applied for filesystem, they may be affecting it as well
\n[23-Sep-2009 12:26:10] <twm1010> if zensyslog receives message\nfrom an IP not in Zenoss as a device, what does it do with them?
[23-Sep-2009 12:32:48] <rmatte> it displays them in the event console along with that IP anyways
\n[23-Sep-2009 12:33:15] <rmatte> but if you have a filter on your\nalerting rule to only alert on devices that are in Production state\nthen it won't alert, since it doesn't consider the device to be in\nproduction state
[23-Sep-2009 12:34:40] <twm1010> Have a good link on setting up the syslog client ?
[23-Sep-2009 12:35:15] <rmatte> what do you mean? setting up syslog-ng between zensyslog and public?
[23-Sep-2009 12:35:30] <twm1010> nah, configuring syslog on a linux host
[23-Sep-2009 12:35:46] <twm1010> i did a basic config: *.* @myzenossIP
[23-Sep-2009 12:36:14] <ckrough> twm1010: docs/DOC-2499
[23-Sep-2009 12:36:15] <twm1010> but let's say i want it to only send warnings/errors/critical
[23-Sep-2009 12:37:13] <ckrough> I'd recommend doing your filtering in zenoss, not syslog
[23-Sep-2009 12:37:53] <twm1010> they recommend in the doc *.debug
[23-Sep-2009 12:41:06] <rmatte> twm1010: we just use straight traps for our stuff, no syslog
[23-Sep-2009 12:41:24] <rmatte> not for our internal stuff anyways
[23-Sep-2009 12:41:33] <rmatte> we do use a bit of syslogging for some of our client's devices
[23-Sep-2009 12:41:41] <twm1010> yeah, i'm having trouble deciding which one to do here
[23-Sep-2009 12:41:44] <rmatte> nothing linux based though
[23-Sep-2009 12:41:58] <rmatte> I prefer the traps, since it'll only trap for the more critical events
[23-Sep-2009 12:42:17] <rmatte> but it's really up to how noisy you want to be and how much tuning you want to do
[23-Sep-2009 12:42:26] <twm1010> well, i don't want to do both traps and syslog
[23-Sep-2009 12:42:33] <twm1010> rather just end up picking one or the other
\n[23-Sep-2009 12:42:38] <rmatte> the syslogs will tell you tons of\nstuff, just not sure it'll really be anything that traps can't tell you\nwhich you'd actually be interested in
[23-Sep-2009 12:42:57] <twm1010> i was thinking for servers, syslog, for devices/routers/etc, traps
[23-Sep-2009 12:43:22] <twm1010> my windows boxes are ridiculously chatty
[23-Sep-2009 12:43:38] <twm1010> can you adjust the WMI event collection to ignore informational stuff?
[23-Sep-2009 12:43:52] <rmatte> you can adjust the level, yeh
[23-Sep-2009 12:44:13] <twm1010> ah zWinEventlogMinSeverity
[23-Sep-2009 12:44:17] <rmatte> zWinEventlogMinSeverity
[23-Sep-2009 12:44:19] <rmatte> yeh
[23-Sep-2009 12:44:22] <cgibbons> don't use 5!
[23-Sep-2009 12:44:22] <rmatte> you beat me to it
[23-Sep-2009 12:44:33] <twm1010> information is lvl 2?
[23-Sep-2009 12:44:37] <rmatte> 5 is lowest
[23-Sep-2009 12:44:47] <rmatte> nah, 2 would be more like warning
[23-Sep-2009 12:44:50] <rmatte> and 1 is critical
[23-Sep-2009 12:45:08] <rmatte> you'll probably have to do some tuning with event mappings regardless of what you choose
[23-Sep-2009 12:45:29] <twm1010> well sure, i have about 100 i have to classify right now
[23-Sep-2009 12:45:39] <rmatte> hehe
[23-Sep-2009 12:45:44] <twm1010> damn terminal servers and SQL servers seem to have unique messages for EVERYTHING
[23-Sep-2009 12:45:50] <rmatte> you know the proper way to tune syslog messages?
[23-Sep-2009 12:46:08]\n<cgibbons> if you are curious about the actual levels, look for\n'EventType' on this page:\nhttp://msdn.microsoft.com/en-us/library/aa394226(VS.85).aspx
[23-Sep-2009 12:46:08] <twm1010> well, its already set to 2
[23-Sep-2009 12:47:01] <twm1010> but i guess there are a lot of things microsoft calls an "error" that are really just information
[23-Sep-2009 12:47:09]\n<rmatte> Create mappings manually for each one under some events\nclass. You can even create an event class (for instance I made one\ncalled /Suppressed), then just name them whatever, set the\neventClassKey to "defaultmapping" and set the regex value to a common\npart of the text in the messages
[23-Sep-2009 12:47:17] <twm1010> in other words, i should quit being lazy and just get these mappings reated
[23-Sep-2009 12:47:22] <rmatte> then go to zProperties and set action to drop or history (depending on what you want)
[23-Sep-2009 12:47:25] <rmatte> and voila
[23-Sep-2009 12:47:45] <twm1010> Yeah I have a /Drop and a /History
[23-Sep-2009 12:47:51] <rmatte> ok
[23-Sep-2009 12:48:09] <rmatte> so yeh, create event mapping by hand, name it whatever, make sure you set eventClassKey to defaultmapping
[23-Sep-2009 12:48:18] <rmatte> then set regex to part of the text in the syslog message
[23-Sep-2009 12:48:32] <rmatte> well, works for any events really
[23-Sep-2009 12:48:36] <rmatte> WMI, Syslog, whatever
[23-Sep-2009 12:48:47] <rmatte> they have to come in with class /Unknown for it to work
[23-Sep-2009 12:49:03] <rmatte> If they don't, you'll have to do the same but put the mapping in to the specific event class
[23-Sep-2009 12:49:28] <twm1010> Ah, I see quite a few in here already
[23-Sep-2009 12:49:33] <twm1010> under /Ignore
[23-Sep-2009 12:49:35] <rmatte> probably
[23-Sep-2009 12:50:58] <rmatte> ah, that's what rule is for
[23-Sep-2009 12:51:09] <rmatte> I've never used anything beyond transform and regex
\n[23-Sep-2009 12:51:41] <rmatte> but yeh, you can put one of the\nsyslog messages in to "Example" when you're entering your regex
[23-Sep-2009 12:51:53] <rmatte> the regex text will become red if it doesn't match properly
[23-Sep-2009 12:53:22] <twm1010> hrmm... i'll have to play with that some more
[23-Sep-2009 12:56:09] <chemist> can anyone tell me what zenoss daemon sends email notifications?
[23-Sep-2009 12:57:52] <ckrough> zenactions I believe
[23-Sep-2009 12:58:04] <chemist> thnx
[23-Sep-2009 13:00:45]\n<rmatte> eugh, frigging UPS, I order something from the U.S that\ncosts $99.99 and they charge me $51.93 in brokerage/taxes
[23-Sep-2009 13:00:49] <rmatte> extortion
[23-Sep-2009 13:00:56] <twm1010> LOL
[23-Sep-2009 13:01:12] <rmatte> last time I use them for anything from the U.S, unreal
[23-Sep-2009 13:01:21] <twm1010> where are you?
[23-Sep-2009 13:01:24] <rmatte> Canada
[23-Sep-2009 13:01:37] <twm1010> What part?
[23-Sep-2009 13:01:41] <rmatte> I'll use USPS next time, slower but much much cheaper
[23-Sep-2009 13:02:06] <ckrough> yeah but $51 US dollars is like Canadian 2 cent now right?
[23-Sep-2009 13:02:16] <rmatte> I'm in Ottawa, Ontario
[23-Sep-2009 13:02:21] <rmatte> a few hours north from Toronto
[23-Sep-2009 13:02:47] <rmatte> ckrough: incorrect lol
[23-Sep-2009 13:02:59] <ke4qqq> ckrough: I think it rose - to 10 cents
[23-Sep-2009 13:03:07] <ckrough> lol
[23-Sep-2009 13:03:23] <rmatte> 1.00 CAD
[23-Sep-2009 13:03:23] <rmatte> =
[23-Sep-2009 13:03:24] <rmatte> 0.933786 USD
[23-Sep-2009 13:03:43] <rmatte> we're back in the hole
[23-Sep-2009 13:04:27] <rmatte> basically US dollar = CAD dollar + 7 cents
[23-Sep-2009 13:06:34] <rmatte> it has to be like that otherwise you guys won't buy as much from us
[23-Sep-2009 13:06:50] <rmatte> so we get the odd time where our dollar skyrockets but it's put back in to it's place fairly quickly
[23-Sep-2009 13:06:51]\n<ke4qqq> interesting- yesterday around 1500 Eastern, my fping\nstuff stopped collecting, or at least stopped reporting on the graphs
[23-Sep-2009 13:08:13] <rmatte> that sucks
[23-Sep-2009 13:08:18] <rmatte> is it still not reporting?
[23-Sep-2009 13:08:42] <ke4qqq> well snmp is reporting, but the graphs that fping collects isn't.
[23-Sep-2009 13:08:53] <ke4qqq> but it was working til around 1500 apparently
[23-Sep-2009 13:08:57] <rmatte> check to make sure zencommand daemon is running
[23-Sep-2009 13:09:07] <rmatte> login as zenoss user
[23-Sep-2009 13:09:09] <rmatte> do zenoss status
[23-Sep-2009 13:09:13] <rmatte> if it's running try...
[23-Sep-2009 13:09:16] <rmatte> zencommand run --cycle
[23-Sep-2009 13:09:23] <rmatte> see what kind of output you get
[23-Sep-2009 13:12:29] <rmatte> might take a while for any output to show up, you can also try: zenperfsnmp run --cycle
[23-Sep-2009 13:13:06] <rmatte> actually, throw a -v10 in there
[23-Sep-2009 13:13:17] <ke4qqq> yeah it's told me it's connected to zenhub, but nothing else thus far
[23-Sep-2009 13:13:17] <rmatte> zencommand run -v10 --cycle
[23-Sep-2009 13:13:23] <rmatte> zenperfsnmp run -v10 --cycle
[23-Sep-2009 13:13:53] <rmatte> it won't exit on it's own, you'll have to ctrl-c out when you're done
[23-Sep-2009 13:14:13] <rmatte> -v10 will show you a lot more
[23-Sep-2009 13:14:29] <rmatte> and yeh, it is zencommand that you want it looks like
[23-Sep-2009 13:14:47] <rmatte> so try out: zencommand run -v10 --cycle
[23-Sep-2009 13:14:50] <rmatte> and see what happens
[23-Sep-2009 13:15:41] <ke4qqq> http://fpaste.org/jgTX/
[23-Sep-2009 13:16:22] <rmatte> hmmm, you should be seeing a lot more than that
[23-Sep-2009 13:16:37] <rmatte> try restarting zenoss
[23-Sep-2009 13:16:39] <rmatte> then run it again
[23-Sep-2009 13:17:53] <rmatte> I see stuff like:
\n[23-Sep-2009 13:17:54] <rmatte> 2009-09-23 14:17:35,420 DEBUG\nzen.zencommand: Storing MemoryTotal = 781760.0 into\nDevices/30.30.254.33/Memory_MemoryTotal
[23-Sep-2009\n13:17:54] <rmatte> 2009-09-23 14:17:35,420 DEBUG zen.RRDUtil:\n/usr/local/zenoss/zenoss/perf/Devices/30.30.254.33/Memory_MemoryTotal.rrd:\n781760.0
[23-Sep-2009 13:17:54] <rmatte> 2009-09-23 14:17:35,421 DEBUG zen.zencommand: RRD save result: 781760.0
\n[23-Sep-2009 13:17:54] <rmatte> 2009-09-23 14:17:35,421 DEBUG\nzen.zencommand: Storing MemoryUsed = 595776.0 into\nDevices/30.30.254.33/Memory_MemoryUsed
[23-Sep-2009\n13:17:57] <rmatte> 2009-09-23 14:17:35,421 DEBUG zen.RRDUtil:\n/usr/local/zenoss/zenoss/perf/Devices/30.30.254.33/Memory_MemoryUsed.rrd:\n595776.0
[23-Sep-2009 13:18:00] <rmatte> 2009-09-23 14:17:35,421 DEBUG zen.zencommand: RRD save result: 595776.0
[23-Sep-2009 13:18:06] <rmatte> when I run zencommand
[23-Sep-2009 13:18:12] <rmatte> and those are command based datapoints
[23-Sep-2009 13:19:40] <ke4qqq> weird - lots of interesting stuff showed up now, including fping statistics
[23-Sep-2009 13:20:52] <rmatte> there you go
[23-Sep-2009 13:21:01] <rmatte> so for some reason your zenactions daemon took a crap
[23-Sep-2009 13:21:04] <rmatte> lord knows why
[23-Sep-2009 13:21:20] <rmatte> anyways, should be fine now
[23-Sep-2009 13:21:25] <rmatte> if it happens again let me know
[23-Sep-2009 13:21:31] <rmatte> then we can look in to it deeper
[23-Sep-2009 13:21:47] <ke4qqq> now - if I could just get the community esx zenpack to work properly
[23-Sep-2009 13:21:48] <rmatte> afk, going to grab some lunch
[23-Sep-2009 13:21:53] <ke4qqq> have a good lunch
[23-Sep-2009 13:22:22] <rmatte> there's a better ESX pack possibly coming to Core eventually (the old Enterprise one)
[23-Sep-2009 13:22:34] <ke4qqq> when?
[23-Sep-2009 13:26:20] <rmatte> not sure
[23-Sep-2009 13:27:06] <rmatte> the new enterprise one uses ESX's API to monitor it, the old one uses SNMP
[23-Sep-2009 13:33:50] <twm1010> nice
[23-Sep-2009 13:33:56] <twm1010> thats a good tidbit of info
[23-Sep-2009 13:35:15] <ke4qqq> yes - that's a very good bit of info
[23-Sep-2009 13:35:52] <ke4qqq> does that mean the new enterprise one will work with ESXi as well
[23-Sep-2009 13:42:24] <rmatte> I believe so
[23-Sep-2009 13:42:35] <rmatte> though I'm not 100% on that
[23-Sep-2009 13:42:53] <rmatte> they may have omitted the API stuff on ESXi as well as the SNMP
[23-Sep-2009 13:42:59] <rmatte> you'd have to check with VMWare to find out
\n[23-Sep-2009 13:43:20] <ke4qqq> I think esxi has the API stuff -\nthat's how virt center communicates with it.....just no snmp
[23-Sep-2009 13:45:53] <rmatte> ah
[23-Sep-2009 13:46:05] <rmatte> yeh, makes sense
[23-Sep-2009 13:49:12] <twm1010> rmatte: Why do you set the eventclasskey to defaultmapping?
[23-Sep-2009 13:49:32] <rmatte> twm1010: because if events come in as /Unknown they have no eventclasskey set
[23-Sep-2009 13:49:41] <rmatte> so defaultmapping matches that
[23-Sep-2009 13:50:02] <rmatte> otherwise you'd never we able to map those events
[23-Sep-2009 13:50:07] <rmatte> be able to*
[23-Sep-2009 13:50:08] <rmatte> rather
[23-Sep-2009 13:51:04] <twm1010> Ah... but if they're NOT unknown, I can leave it be
[23-Sep-2009 13:51:21] <rmatte> then you would check the events to see if it has an eventClassKey
[23-Sep-2009 13:51:53] <rmatte> in the event's details
[23-Sep-2009 13:52:24] <twm1010> OK... so I see two events here, they're different, both unknown
[23-Sep-2009 13:52:31] <twm1010> with the same classkey in the details
[23-Sep-2009 13:52:38] <rmatte> eventclasskey
[23-Sep-2009 13:52:40] <rmatte> not classkey
[23-Sep-2009 13:52:42] <rmatte> 2 different things
[23-Sep-2009 13:52:55] <twm1010> yeah i abbreviated, sorry, i did mean eventclasskey when i referred to it just now
[23-Sep-2009 13:53:08] <rmatte> what's the eventclasskey for them?
[23-Sep-2009 13:53:18] <twm1010> RSR_0
[23-Sep-2009 13:53:31] <twm1010> they do have some common text I can RegEx against though.
[23-Sep-2009 13:53:33] <rmatte> ok, so make a mapping where eventClassKey is RSR_0
[23-Sep-2009 13:53:42] <rmatte> and then specify the regex and the action
[23-Sep-2009 13:56:00] <twm1010> Gotcha... I really should go back and watch that presentation on events
[23-Sep-2009 13:56:15] <rmatte> hehe
[23-Sep-2009 13:56:17] <twm1010> eventClassKey is determined by the daemon? in this case, zenwin
[23-Sep-2009 13:56:23] <rmatte> I don't think she covered this specific stuff
[23-Sep-2009 13:56:31] <rmatte> she just went more in to the inner-workings of the system
[23-Sep-2009 13:56:39] <twm1010> still i want to understand where eventclasskey comes from
[23-Sep-2009 13:56:43] <rmatte> you should read the paper though
[23-Sep-2009 13:57:13] <rmatte> yeh, I'm not even 100% sure where it gets assigned from
\n[23-Sep-2009 13:57:49] <twm1010> for instance, i have a "last\nmessage repeated" event coming from syslog from a linux host
[23-Sep-2009 13:57:54] <twm1010> it doesn't ahve an eventclasskey
[23-Sep-2009 13:59:30] <twm1010> it looks to me like eventclassKey from a windows event log is the servicename + eventID
[23-Sep-2009 13:59:33] <twm1010> if there is one
[23-Sep-2009 14:00:27] <rmatte> I think it's whatever they code it to be
[23-Sep-2009 14:03:29] <twm1010> so.. it uses both the classkey AND the regex to identify a message?
[23-Sep-2009 14:03:44] <rmatte> correct
[23-Sep-2009 14:03:57] <rmatte> because you can have events with identical eventclasskeys but different messages
[23-Sep-2009 14:05:02] <twm1010> right, so whatever you put in the one for default mapping is like a catchall
[23-Sep-2009 14:05:19] <rmatte> it'll catch any that have a blank eventclasskey
[23-Sep-2009 14:05:22] <rmatte> as I understand it
[23-Sep-2009 14:05:30] <rmatte> but I also believe they have to be /Unknown event class
[23-Sep-2009 14:06:17] <rmatte> it works great for syslogs
[23-Sep-2009 14:08:10] <twm1010> Gotcha, still don't quite get it, but I'll start reading, you know, RTFM.
[23-Sep-2009 14:08:18] <rmatte> hehe
[23-Sep-2009 14:08:21] <rmatte> ypu
[23-Sep-2009 14:08:26] <rmatte> yup*
[23-Sep-2009 14:09:04] <rmatte> I haven't done any kind of event mapping or transform in over a month, so I'm slightly fuzzy on it myself
[23-Sep-2009 14:09:30]\n<rmatte> Going to be spending the next couple of days updating\nzenpacks with new mappings and deploying them across our servers
[23-Sep-2009 14:09:32] <rmatte> fun fun
[23-Sep-2009 14:15:30] <twm1010> ahhh ok, i get it now, admin manual FTW
[23-Sep-2009 14:15:45] <rmatte> lol
[23-Sep-2009 14:18:20] <twm1010> when it doesn't have an eventClassKey, it searches for defaultmapping
[23-Sep-2009 14:18:37] <twm1010> And you can have multiple defaultmappings, with different RegEx's
[23-Sep-2009 14:19:00] <rmatte> correct
[23-Sep-2009 14:19:24] <rmatte> you can have multiple of any event key with different regexes
[23-Sep-2009 14:19:35] <rmatte> but defaultmapping applies to any events with blank event keys
[23-Sep-2009 14:19:35] <twm1010> sure, to distinguish them
[23-Sep-2009 14:19:41] <rmatte> yeh
\n[23-Sep-2009 14:20:12] <twm1010> So if you do have an\neventClassKey, but don't give it a RegEx, or Example, it then uses that\nas a default for everything with that key?
[23-Sep-2009 14:20:35] <rmatte> no, you can't leave an event mapping blank like that
[23-Sep-2009 14:20:44] <twm1010> yeah, i figured that would be so
[23-Sep-2009 14:20:48] <rmatte> it won't do anything if you do that
[23-Sep-2009 14:21:06] <rmatte> It needs either a regex, a transform, or an evaluation
[23-Sep-2009 14:21:19] <twm1010> gotcha, it needs the rule or the regex, and default mapping lets you create catch-alls
[23-Sep-2009 14:21:24] <rmatte> yeh
[23-Sep-2009 14:21:29] <rmatte> rule, regex, or transform
[23-Sep-2009 14:21:32] <rmatte> any of those work
[23-Sep-2009 14:21:51] <rmatte> leaving everything blank results in nothing
[23-Sep-2009 14:22:59] <rmatte> and even if it didn't it'd be super bad practice
[23-Sep-2009 14:23:00] <rmatte>
[23-Sep-2009 14:24:53] <twm1010> Can an event match two separate mappings?
[23-Sep-2009 14:25:17] <rmatte> it takes the first one that it matches
[23-Sep-2009 14:25:35] <twm1010> ah, that's what "sequence" is for
[23-Sep-2009 14:25:39] <rmatte> yup
[23-Sep-2009 14:25:57] <twm1010> Sweet, my event console is cleaning up rather fast.
[23-Sep-2009 14:26:03] <rmatte> hehe
[23-Sep-2009 14:26:12] <rmatte> yeh, it gets really easy once you start picking up on this stuff
[23-Sep-2009 14:26:15] <twm1010> Now if I could only fix the 25 acknowledged problems that really are issues.
[23-Sep-2009 14:26:21] <rmatte> the defaultmapping thing was the last thing that I learned
[23-Sep-2009 14:26:26] <rmatte> and it's a frigging god-send
[23-Sep-2009 14:26:30] <rmatte> hehe
[23-Sep-2009 14:27:05] <rmatte> I kept having syslog messages coming in with a completely blank message
[23-Sep-2009 14:27:19] <rmatte> so I made a defaultmapping event mapping, and made the regex ^$
[23-Sep-2009 14:27:24] <rmatte> and voila, problem solved
[23-Sep-2009 14:57:36] <rmatte> g'day Matt
[23-Sep-2009 14:57:41] <rmatte> lots of meetings today?
[23-Sep-2009 14:58:55] <mrayzenoss> yeah, lots of meetings
[23-Sep-2009 14:59:02] <rmatte> ah
\n[23-Sep-2009 14:59:39] <mrayzenoss> beta 3 artifacts are on\nSourceForge, the beta download page is being migrated from zenoss.com\nto zenoss.org, so there will be a lag in that update
[23-Sep-2009 15:06:05] <rmatte> cool
[23-Sep-2009 15:06:19] <rmatte> upgrade from beta 2 to beta 3 will work I assume?
[23-Sep-2009 15:06:46] <rmatte> and is the stack installer going to come out like a week after for this release?
[23-Sep-2009 15:14:04] <mrayzenoss> there's a stack installer
[23-Sep-2009 15:14:09] <rmatte> w00t
[23-Sep-2009 15:14:14] <mrayzenoss> upgrades are probably still broken
[23-Sep-2009 15:14:19] <rmatte> crap
[23-Sep-2009 15:14:20] <rmatte> lol
[23-Sep-2009 15:14:20] <mrayzenoss> at least from 2.4.X->beta
[23-Sep-2009 15:14:21] <rmatte> oh well
[23-Sep-2009 15:14:29] <mrayzenoss> maybe beta2->beta3 may work
[23-Sep-2009 15:14:38] <rmatte> yeh, but I'll try from beta2->beta3 to see if it works
[23-Sep-2009 15:14:42] <rmatte> if not, oh well, reinstall
[23-Sep-2009 15:14:57] <mrayzenoss> so next week's beta should be the first to be worth testing upgrades with
[23-Sep-2009 15:15:03] <rmatte> cool
[23-Sep-2009 15:17:31] <rmatte> yay, I get a new commissioning project tomorrow
[23-Sep-2009 15:17:39] <rmatte> 15 devices, but apparently most of them are stacked switches
[23-Sep-2009 15:17:52] <rmatte> where they are a bunch of switches all linked together somehow under 1 IP address
[23-Sep-2009 15:18:07] <rmatte> it's going to be fun to figure out how to monitor the other switches
[23-Sep-2009 15:32:10] <chemist> If I restart zenmodelerwill it attempt to remodel devices 12 hours after that?
[23-Sep-2009 15:32:43] <rmatte> chemist: Honestly, no idea, but I think it follows some sort of zenhub cycle time
[23-Sep-2009 15:56:57] <cgibbons> hmm
[23-Sep-2009 16:45:00]\n<ke4qqq> rmatte: the switches will be easy - they have a single\nipaddress, and act like modules of a chassis switch or router so you'll\nhave 2/0/1-48 and then 3/0/1-48 will be the next switch etc.
[23-Sep-2009 18:00:01] <davetoo> I know, everybody is probably out or asleep But does anyone know what the "Messages" dashboard portlet does?
[23-Sep-2009 18:00:05] <davetoo> Messages from what/where?
[23-Sep-2009 19:33:35] <davetoo> whee
[23-Sep-2009 19:34:36] <davetoo> ${here/ZenPackManager/packs/ZenPacks.xxorgxx.xxthingxx/path}/libexec/check_thing.py
[23-Sep-2009 20:16:02] <etank> davetoo: i think that the message portlet is for when zenoss has jobs queued up it will leave information there
[23-Sep-2009 20:16:08] <etank> i could be wrong on that though
[23-Sep-2009 20:16:22] <etank> i have never seen anything at all in messages
[23-Sep-2009 20:16:41] <etank> could be possible that it will have stuff if you are doing a discovery on a large subnet
\n[23-Sep-2009 20:17:34] <etank> davetoo:\nmessage/38671%3Bjsessionid=07F9C4F40DECB2BEC062C491E1427675.node0
[23-Sep-2009 20:18:02] <davetoo> ah, ok, thanks
[23-Sep-2009 20:18:23] <davetoo> I'd actually like to have a "wall" portlet Where admins can put an MOTD or similar
[23-Sep-2009 20:21:21] <etank> so where you could post notices and stuff
[23-Sep-2009 20:21:47] <davetoo> yes; we'll have 20-30 people looking at the dashboard around the world
[23-Sep-2009 20:22:13] <etank> so kind of a mashup of Facebook and Zenoss
[23-Sep-2009 20:22:15] <davetoo> almost a chat window
[23-Sep-2009 20:23:04] <davetoo> where we could post things like "Hey, we just lost a core router in Site X, so you might get alert spam"
[23-Sep-2009 20:24:05] <etank> that would actually be kind of nice to have
\n[23-Sep-2009 20:24:11] <LW> that would be a good idea, my end\nuser call center is getting a view into our zenoss install soon..... \nand I'd really prefer for them not to call me if I can help it for\nevery little thing
[23-Sep-2009 20:24:34] <LW> "Hey! we know about this!" would be a life saver
[23-Sep-2009 20:24:41] <davetoo> yeah, we're about to roll out Enterprise,
[23-Sep-2009 20:25:09] <etank> i wish there was a way to select multiple events at once and give them all the same "log" information
\n[23-Sep-2009 20:25:10] <davetoo> and for the most part the folks\nin Bangalore are supposed to be the human alerting system watching the\nZenoss dashboard
[23-Sep-2009 20:25:20] <etank> instead of copy paste copy paste etc
[23-Sep-2009 20:25:41] <davetoo> like, from multiple devices?
[23-Sep-2009 20:25:49] <etank> yeah
[23-Sep-2009 20:25:53] <davetoo> Or all the same device but different events
[23-Sep-2009 20:25:57] <etank> but all from the same device would be good too
[23-Sep-2009 20:26:15] <davetoo> learn zendmd/python
[23-Sep-2009 20:26:40] <etank> thats good and all but it would be nicer to be able to do it from the web interface
[23-Sep-2009 20:26:48] <davetoo> yes
[23-Sep-2009 20:27:49] <etank> i dont want to have to drop to a python console whenever something like this happens
[23-Sep-2009 20:29:13] <davetoo> event management is one area where zenoss needs work, I think
\n[23-Sep-2009 20:29:24] <etank> well im going to be at the\ncommunity day training at OLF this week so maybe i will bring it up
[23-Sep-2009 20:29:39] <davetoo> I just did my first "real" transform
[23-Sep-2009 20:29:58] <etank> zenoss does A LOT of things right but it is not perfect
[23-Sep-2009 20:30:14] <etank> i need to learn more about transforms actually
[23-Sep-2009 20:30:27] <etank> we are trying to make a purchase of enterprise at work
[23-Sep-2009 20:30:33] <davetoo> This was for an snmp trap
[23-Sep-2009 20:30:49] <etank> we made it through the evaluation and it could do most everything we need it to do
[23-Sep-2009 20:30:49] <davetoo> to change severity based on the trap value
[23-Sep-2009 20:31:18] <davetoo> yes, we used Core for a couple of years
[23-Sep-2009 20:31:29] <davetoo> needed the global view
\n[23-Sep-2009 20:31:32] <etank> i think i did one that if a\nspecific event happened between 5:30 and 6:30 in the morning it would\njust archive it
[23-Sep-2009 20:31:59] <davetoo> I have to say that I'm much happier with 2.4.x than previous releases
\n[23-Sep-2009 20:32:13] <etank> that is the time of day that we\nreindex the databases so the web servers are going to complain so we\njust ignore those
[23-Sep-2009 20:32:21] <etank> im looking forward to 2.5
[23-Sep-2009 20:32:29] <davetoo> oh we still have to go through the whole mapping/escalation thing
[23-Sep-2009 20:32:33] <etank> the SQL zenpack will have some nice new functions
[23-Sep-2009 20:32:53] <davetoo> I don't do any app monitoring(yet)
[23-Sep-2009 20:33:13] <etank> the inheritance that is in zenoss makes it so nice to use
[23-Sep-2009 20:33:27] <etank> build the device class the way you want it and then just add the devices
[23-Sep-2009 20:33:43] <davetoo> yes;
[23-Sep-2009 20:33:48] <davetoo> I modify that idea slightly.
[23-Sep-2009 20:33:51] <etank> makes life soooo easy
[23-Sep-2009 20:34:07] <davetoo> I break the big Device templates up into smaller sections
[23-Sep-2009 20:34:23] <davetoo> and put those in the Device Templates list instead
\n[23-Sep-2009 20:34:51] <etank> we have started building our own\ntemplates and device classes that are specific to a type of device
\n[23-Sep-2009 20:34:55] <davetoo> allows me to have a bunch of\ndifferent os versions that have differing levels of snmp-brokennes
[23-Sep-2009 20:35:02] <etank> same here
[23-Sep-2009 20:35:09] <etank> we do it by domains as well
[23-Sep-2009 20:35:16] <davetoo> so I have a template-let that just does very basic CPU stats
[23-Sep-2009 20:35:26] <etank> so the zenproperties will have the right authentication information
[23-Sep-2009 20:35:28] <davetoo> and then another tiny one that does stuff that only BSD systems have
[23-Sep-2009 20:36:32] <etank> whatever makes it work
[23-Sep-2009 20:37:28] <etank> well im out for now. later davetoo
[23-Sep-2009 20:37:34] <davetoo> ok bye
[24-Sep-2009 00:00:46] [disconnected at Thu Sep 24 00:00:46 2009]
[24-Sep-2009 00:00:46] [connected at Thu Sep 24 00:00:46 2009]
[24-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
[24-Sep-2009 03:51:31] <Egyptian[web]> hello all
\n[24-Sep-2009 04:16:39] <Egyptian[web]> i want to monitor my san\nstorage and i have the mib and its seen but the device class path is\nnot set correctly .. i have to manually change it from / to\n/server/scan (as per documentation on auto discovery) but i m not happy\nwith that either ..
[24-Sep-2009 04:17:01]\n<Egyptian[web]> is there something like /storage ? if not ..how\ndo i create it? and what is the best practice to name the device class?
[24-Sep-2009 04:17:06] * Egyptian[web] takes a breath
[24-Sep-2009 05:07:30]\n<sergeymasushko> Hi guys! we've updated to 2.4.5 and now when I\ntry to open 'devices' page or try to use search I got the following:\nhttp://pastebin.com/m72b50e5e
[24-Sep-2009 05:40:11] <Egyptian[web]> sergeymasushko: do you have an nfs mount on the zenoss pc ?
[24-Sep-2009 06:45:28] <Egyptian[web]> is there like a zenpack for oracle? ..
[24-Sep-2009 06:45:50] <Egyptian[web]> actually .. all i want is to monitor my oracle db .. how do i go about it?
\n[24-Sep-2009 08:36:18] <Egyptian[web]> i set up a page for myself\n( a modified command to call a script that will send me an gsm text\nmessage) and now .. i want to test it.. how?
[24-Sep-2009 08:42:28] <straterra> firewall zenoss off from a machine..and see if you get paged o.O
[24-Sep-2009 08:52:20] <Egyptian[web]> straterra: interesting .. i can bring down a test machine ..thanks
[24-Sep-2009 08:52:28] <straterra> No problem
[24-Sep-2009 08:56:29] <Egyptian[web]> is it possible to like disable monitoring of a device?
[24-Sep-2009 08:59:19] <straterra> Sure
[24-Sep-2009 08:59:28] <straterra> Put it in production state of maintenance
[24-Sep-2009 08:59:35] <straterra> (I think its called production state)
\n[24-Sep-2009 09:19:41] <rmatte> Egyptian[web]: just a tip, if you\ngo to Events, then select "Add Event" from the top dropdown menu you\ncan generate a test event with whatever values you wish
[24-Sep-2009 09:20:06] <rmatte> quite useful for testing
[24-Sep-2009 09:20:58] <kmtsun> anyone going to zenoss pres @ ohio linux fest?
[24-Sep-2009 09:21:34] <Egyptian[web]> straterra: no gsm text btw
[24-Sep-2009 09:22:19] <mrayzenoss> kmtsun: I'm there!
[24-Sep-2009 09:22:31] <mrayzenoss> kmtsun: we've got over 60 people signed up for the free training
[24-Sep-2009 09:22:35] <kmtsun> nice ..i will be there tomorrow
[24-Sep-2009 09:22:52] <kmtsun> is it too late to sign up?
[24-Sep-2009 09:22:56] <mrayzenoss> nope
[24-Sep-2009 09:27:56] <twm1010> changelog?
\n[24-Sep-2009 09:29:53] <rmatte> eugh, even MORE screwed up\nbehaviour to do with setting the Product info for devices, any server\nthat I upgraded to 2.4.x from 2.3.3 is having even more issues (can't\nsee certain product models in the dropdown box), than those that were\noriginally 2.4.x installs
[24-Sep-2009 09:30:01] <rmatte> this is driving me loopy, I need to find a workaround for this
[24-Sep-2009 09:32:40] <rmatte> how do I manually link a device to a different product class using zendmd?
[24-Sep-2009 09:32:46] <rmatte> I'll have to do that as a last resort
\n[24-Sep-2009 09:34:25] <rmatte> we use the Zenoss inventory for\nbilling, so we can't have incorrect product details in there
[24-Sep-2009 09:43:22] <twm1010> Anyone using Zenoss for IIS monitoring?
[24-Sep-2009 09:44:04] <etank> twm1010: we are (some)
\n[24-Sep-2009 09:44:40] <etank> anyone know what could be causing\n"localhost zenactions heartbeat failure" events every 2 hours?
[24-Sep-2009 09:45:41] <rmatte> etank: it's probably just getting overworked
[24-Sep-2009 09:45:47] <twm1010> does the log for zenactions daemon show anything?
\n[24-Sep-2009 09:45:54] <rmatte> I get the occasional heartbeat\nfailure for different processes on different zenoss servers
\n[24-Sep-2009 09:45:58] <etank> rmatte: i just found this\nhttp://www.mail-archive.com/zenoss-users@zenoss.org/msg24200.html//www.mail-archive.com/zenoss-users@zenoss.org/msg24200.html
[24-Sep-2009 09:45:59] <rmatte> I wouldn't be too concerned with it
[24-Sep-2009 09:46:04] <etank> i think that is the issue
[24-Sep-2009 09:46:32] <rmatte> ah
[24-Sep-2009 09:46:39] <rmatte> well, if it's Chet, it's probably right
[24-Sep-2009 09:46:41] <rmatte> lol
[24-Sep-2009 09:46:49] <etank> Last Checked 2009/09/11 00:43:38.000
[24-Sep-2009 09:47:00] <rmatte> there be your problem
[24-Sep-2009 09:47:07] <etank> looks like that is most likely the issue
[24-Sep-2009 09:47:19] <rmatte> I'll have to remember that one
[24-Sep-2009 09:47:35] <etank> thank God for google
[24-Sep-2009 09:47:38] <rmatte> I guess it checks for updates every 2 hours
[24-Sep-2009 09:47:54] <rmatte> I doubt that would have hurt anything though had you left it
[24-Sep-2009 09:48:26] <twm1010> Hrmm.. Igor's IIS template looks perfect
[24-Sep-2009 09:48:44] <rmatte> Yeh, Egor's stuff is pretty good
[24-Sep-2009 09:48:59] <rmatte> mrayzenoss: woot, was wondering when that was coming out
\n[24-Sep-2009 09:49:31] <rmatte> mrayzenoss: do you know of any\ndocs laying around that explain how to use zendmd to actually link\nitems? I need to know how to link productClass on devices to a\nspecific product class
[24-Sep-2009 09:49:38] <twm1010> hrmmm... which ports are required for WMI communication?
[24-Sep-2009 09:49:51] <cgibbons> lots
\n[24-Sep-2009 09:49:51] <rmatte> mrayzenoss: need to use it as a\nworkaround for this product bug until it gets fixed, because we can't\ngo without being able to specify that info
[24-Sep-2009 09:49:54] <cgibbons> it's dynamic by default
[24-Sep-2009 09:50:00] <cgibbons> you can configure it to be minimal if you need to restrict it
[24-Sep-2009 09:50:23] <twm1010> well, how about on the client end
\n[24-Sep-2009 09:50:32] <cgibbons> but 135 for the initial\nconnection, and then the server reassigns a dynamic port after that
[24-Sep-2009 09:50:39] <twm1010> gotcha
[24-Sep-2009 09:50:44] <cgibbons> doing firewall stuff?
\n[24-Sep-2009 09:51:11] <twm1010> yeah, im going to be monitoring\nour production webservers (windows), and there are quite a few\nfirewalls between here and there that are not very permissive
[24-Sep-2009 09:51:27] <cgibbons> k, let me get you the link... very easy to configure on the devices to be more limited
[24-Sep-2009 09:51:39] <cgibbons> http://msdn.microsoft.com/en-us/library/ms809327.aspx
\n[24-Sep-2009 09:53:38] <cgibbons> and if you use the perfmon\nmonitoring in enterprise i think you'll need 445, too, but i'll need to\ndouble chec
[24-Sep-2009 09:53:45] <Brad_K>\nrmatte: so, that product problem is driving you crazy, eh? got any\nreal fixes for it yet?
[24-Sep-2009 09:54:11] <twm1010> Does it determine whether something is "Internet" facing via it's IP class?
\n[24-Sep-2009 09:54:35] <Brad_K> rmatte: or, can you at least\npoint me at good zendmd resources, so that we can jury-rig a solution\nto the same problem here?
[24-Sep-2009 09:54:36] <rmatte> Brad_K: working on it
[24-Sep-2009 09:54:44] <rmatte> Brad_K: I need to know how to link items using zendmd
[24-Sep-2009 09:55:13] <mrayzenoss> rmatte: what was the # on that, I'll check on it
[24-Sep-2009 09:55:13] * Brad_K is going to make sure that we file a customer service case on this bug...
\n[24-Sep-2009 09:55:21] <rmatte> Brad_K: I'm more working on\nworkarounds that actually fixing the main issue, I don't have time to\ndo that at the moment
[24-Sep-2009 09:55:37] <rmatte> Brad_K: good, I already have a trac case open for it
[24-Sep-2009 09:55:47] <rmatte> Brad_K: If I find a workaround I'll let you know
[24-Sep-2009 09:55:54] <Brad_K> so you're taking the open source road, and i
\n[24-Sep-2009 09:56:21] <Brad_K> so you're taking the open source\nroad, and i'll be taking the enterprise road, and you'll probably get\nthere before me. ;-)
[24-Sep-2009 09:56:26] <rmatte> yeh, we're core users so we don't have as much push as you guys
[24-Sep-2009 09:56:32] <rmatte> you'd be surprised
[24-Sep-2009 09:56:37] <rmatte> http://dev.zenoss.com/trac/ticket/5473
[24-Sep-2009 09:56:41] <Brad_K> rmatte: if you can let me know what the trac case is, i can make sure we link to it.
[24-Sep-2009 09:56:42] <rmatte> the ticket is set as backlog too
[24-Sep-2009 09:56:49] <rmatte> meaning, they'll get to it whenever
[24-Sep-2009 09:57:01] <rmatte> been opened for 4 weeks now
\n[24-Sep-2009 09:57:19] <Brad_K> rmatte: yeah, but you have direct\naccess to the developers here, and you know a lot more about the\nproblem than i do. i'm still getting spun up on how i can properly\nmonitor a web page.
[24-Sep-2009 09:57:37] <rmatte> hehe
\n[24-Sep-2009 09:57:58] <Brad_K> the zenwebtx stuff seems to be\nmissing one critical component that i need to be able to capture,\nbefore i can actually do a proper log in/log out session.
[24-Sep-2009 09:58:19] <Brad_K> got a meeting. bbiab.
\n[24-Sep-2009 09:58:27] <rmatte> Brad_K: you can always write some\nexternal script to do the login/logout and then just tie it in to\nZenoss if worst comes to worst
[24-Sep-2009 09:58:33] <rmatte> not sure how scripty you are
[24-Sep-2009 09:58:35] <rmatte> k cool
[24-Sep-2009 10:02:08] <mrayzenoss> rmatte: Brad_K: I pinged the dev on it, it's on his schedule for this week
[24-Sep-2009 10:02:23] <mrayzenoss> of course, it gets pushed anytime he gets a P1 assigned to him
[24-Sep-2009 10:03:13]\n<rmatte> I'd almost consider this a P1 at this point since a main\ncomponent of Zenoss is *totally* not working as designed.
[24-Sep-2009 10:04:13] <mrayzenoss> yeah
[24-Sep-2009 10:04:49] <chudler> doesn't that describe all bugs? i.e. not working as designed?
[24-Sep-2009 10:05:50] <rmatte> not really
[24-Sep-2009 10:06:52]\n<twm1010> @cgibbons: so at the minimum, Zenoss will need to be\nable to reach the client on port 135, and a range of ports i designate,\nand likewise the reverse.
[24-Sep-2009 10:07:54]\n<cgibbons> the device doesn't connect back to the zenoss\ncollector, so as long as your firewall lets an outbound connection to\nthose ports from any client port work you're golden
[24-Sep-2009 10:08:43]\n<mrayzenoss> chudler: P1 indicates broken, can't ship with it. \nP2 indicates broken, possible work-around or known existing issue
[24-Sep-2009 10:08:55] <mrayzenoss> we won't ship with P1s
[24-Sep-2009 10:11:58] <twm1010> OK, thanks, I'll start plugging away at it.
\n[24-Sep-2009 10:19:52] <rmatte> mrayzenoss: I don't think you can\ntechnically ship with this defect, but it's up to you guys really
\n[24-Sep-2009 10:20:33] <rmatte> I just know that King Crab is\nsupposed to be a big deal release, so releasing with obvious bugs like\nthat wouldn't be the greatest idea
[24-Sep-2009 10:20:40] <mrayzenoss> right
[24-Sep-2009 10:20:47] <mrayzenoss> they're being worked on
[24-Sep-2009 10:23:10] <rmatte> wow, just found a solution to something crazy weird
\n[24-Sep-2009 10:23:49] <rmatte> I had a bunch of products missing\nfrom the Cisco products dropdown, I had to go in to each one's\nproperties, modify something (I removed the last number on the OID),\nsave them, then add it back and save them again, now they show up in\nthe list
[24-Sep-2009 10:24:10] <rmatte> oh well, at least I know the solution to that issue now
[24-Sep-2009 10:25:58] <twm1010> alright someone set me straight
[24-Sep-2009 10:26:05] <twm1010> should the zenpacks have .zip or.egg on the end
[24-Sep-2009 10:27:08] <rmatte> depends
[24-Sep-2009 10:27:16] <rmatte> if it says .egg.zip then it's really an egg
[24-Sep-2009 10:27:25] <rmatte> if it says just .zip, then it's a zip
[24-Sep-2009 10:27:33] <rmatte> for some reason the new site zipped up all of the egg files
[24-Sep-2009 10:27:55] <twm1010> yes, thats throwing me off again
[24-Sep-2009 10:28:02] <rmatte> .zip are older zenpacks (which will eventually need to be converted to eggs)
[24-Sep-2009 10:28:20] <rmatte> yeh, it'd be nice if all of those egg packs could get unzipped at some point
[24-Sep-2009 10:28:20] <rmatte> lol
[24-Sep-2009 10:28:51] <rmatte> not sure if that's a Jive quirk or not
[24-Sep-2009 10:29:53] <mrayzenoss> It's a Jive quirk
[24-Sep-2009 10:29:56] <mrayzenoss> I've opened a ticket
[24-Sep-2009 10:30:59] <rmatte> figured as much
[24-Sep-2009 10:31:02] <rmatte> hehe
[24-Sep-2009 10:32:07] <rmatte> ok, well now that that issue with the list is fixed the product info is semi-usable
\n[24-Sep-2009 10:32:36] <rmatte> still a pain though since you\nneed to manually set everything if you want to use the edit tab at all,\notherwise it'll clear all your info
[24-Sep-2009 10:32:53] <rmatte> (product and software info that is)
[24-Sep-2009 10:33:41] <mrayzenoss> I've closed 2 P1s in the last 30 minutes
[24-Sep-2009 10:33:53] <cgibbons> woot
[24-Sep-2009 10:33:57] <rmatte> nice
[24-Sep-2009 10:34:22] * rmatte thinks he really just set P3s to P1s right before closing them, sneaky sneaky
[24-Sep-2009 10:34:47] <mrayzenoss> nah, just glaring issues that were easy to fix
[24-Sep-2009 10:35:02] <cgibbons> we still have a bad habit of making issues P1s when they're really not anywhere close
[24-Sep-2009 10:35:37] <rmatte> hehe
\n[24-Sep-2009 10:36:52] <rmatte> cgibbons: I had a question\nrelating to http://dev.zenoss.org/trac/ticket/5376 and\nhttp://dev.zenoss.org/trac/ticket/4997, not sure if you'll be able to\nanswer...
[24-Sep-2009 10:37:40] <rmatte> I had\nopened 5376 because I was working no a transform to update interface\nstatus when an interface went down or up, Chet closed the ticket\nclaiming that 4997 resolved this issue...
[24-Sep-2009\n10:38:12] <rmatte> Chet stated in 4997 "This should provide more\nconsistency while allowing for real-time interface events to be\nrepresented as well."
[24-Sep-2009 10:39:56]\n<rmatte> My question is this, if an interface has an inactive\nstatus, will Zenoss stop monitoring it as I saw in previous versions of\nZenoss? If not, I was expecting interface status monitoring to be\nbuilt in to King Crab (which it appears to not be). To reflect actual\ninterface status, would it be possible to modify the transform in 4376\nto modify the status of interfaces as they go down/up? If so, what is\nthe zendmd property that would need to be manipulated to do
[24-Sep-2009 10:41:08] <rmatte> (sorry about the bible long question)
\n[24-Sep-2009 10:42:41] <twm1010> @cgibbons: as it turns out, the\nfirewall is already configured to allow WMI between Zenoss and the\nservers, w00t!
[24-Sep-2009 10:44:22] <rmatte>\nWhen I say "I was expecting interface status monitoring to be built in\nto King Crab" I mean not having to resort to this:\ndocs/DOC-2494
[24-Sep-2009 10:44:41] <rmatte> but I tested with King Crab and apparently it's still necessary
[24-Sep-2009 10:47:24] <twm1010> your saying, you want that to be default behavior?
[24-Sep-2009 10:47:40] <rmatte> me?
[24-Sep-2009 10:48:05]\n<rmatte> Yes, I'd like it to automatically monitor any ports that\nit sees as up when modelling, and then reflect active port status for\nthose ports based on SNMP
[24-Sep-2009 10:48:38]\n<rmatte> I constantly have NOC staff going to the OS tab and\nsaying "Well, the port is up!" and then I have to explain to them how\nthe port status shown on the OS tab is not real-time, it only gets\nupdated when modelling
[24-Sep-2009 10:48:51] <rmatte> which is something that, as I understood, was going to be changed in King Crab
[24-Sep-2009 10:49:35] <twm1010> well then, i look forward to a response
\n[24-Sep-2009 10:49:48] <rmatte> if I need to still do the\nmonitoring with a transform and a threshold then so be it, I just want\nto make sure that if the status is inactive that Zenoss continues to\nmonitor the port so that eventually a clear status can come in and the\nport status can be updated
[24-Sep-2009 10:50:33]\n<rmatte> If it does, then I can just modify my transform to\nupdate the status on ports as they go up and down
[24-Sep-2009 10:51:00] <twm1010> ah, so the perf template checks for ifOperStatus
[24-Sep-2009 10:51:12] <twm1010> and if it fails, your transform modifies that status
\n[24-Sep-2009 10:51:14] <rmatte> The issue I was having with my\nprevious attempt at doing this was that when I'd update the port status\nto down, Zenoss would completely stop monitoring the port
\n[24-Sep-2009 10:51:36] <rmatte> so no clear status would come in\nwhich would have triggered the transform to change the port status back\nto active
[24-Sep-2009 10:52:04] <rmatte> twm1010: correct
[24-Sep-2009 10:52:05] <twm1010> ah.... I see the disparity, you almost need a third indicator there
[24-Sep-2009 10:52:26] <rmatte> almost hehe
\n[24-Sep-2009 10:52:51] <rmatte> although I noticed that when you\nmodel a device now, Zenoss automatically sets monitoring status of\nports that are down to not monitored, which is perfect
[24-Sep-2009 10:53:06] <rmatte> but all I want to know is whether or not it continues to monitor those ports even if they are showing as down
[24-Sep-2009 10:53:13] <rmatte> actually, I just though of a way to test that...
[24-Sep-2009 10:53:26] <rmatte> It should be quite easy
[24-Sep-2009 10:54:33] <rmatte> well now that's just stupid
[24-Sep-2009 10:54:36] <rmatte> hmmm
[24-Sep-2009 10:56:05]\n<rmatte> If a port is detected as down during modelling it marks\nthe status of the port as gray, which is fine, but it absolutely\nrefuses to allow you to enable monitoring on it
[24-Sep-2009 10:56:16] <rmatte> so effectively you need to remodel the entire device every time a port is turned up
[24-Sep-2009 10:56:18] <rmatte> ridiculous
[24-Sep-2009 10:56:23] <etank> ZenWebTx is some cool stuff. TestGen4Web makes it fun to play with too
[24-Sep-2009 10:56:29] <rmatte> I'm going to reopen 4997
[24-Sep-2009 10:57:21] <rmatte> Actually, I'll just make a new ticket I guess
[24-Sep-2009 10:57:54] <rmatte> meh, reopen
[24-Sep-2009 11:11:10] <rmatte> http://dev.zenoss.org/trac/ticket/4997#comment:5
[24-Sep-2009 11:11:12] <rmatte> voila
[24-Sep-2009 11:12:18] <rmatte> that's a biggy
[24-Sep-2009 11:22:31] <rmatte> damn, just noticed another issue with the product model info
[24-Sep-2009 11:22:35] <rmatte> time to add it to the ticket
\n[24-Sep-2009 11:24:11] <cgibbons> i have an old email from chet\non this topic, rmatte... (I haven't read everything you've mentioned)\nand he just said customers typically handle this with trap handling but\n4997 was going to make it more obvious
[24-Sep-2009 11:24:23] <cgibbons> but i'm not going to be much help beyond that
[24-Sep-2009 11:24:43] <rmatte> all good
[24-Sep-2009 11:24:54] <rmatte> I reopened 4997 since it still has a bug
[24-Sep-2009 11:25:04] <rmatte> well, has a bug due to the change I should say
\n[24-Sep-2009 11:39:36] <rmatte> well, I finally figured out a\nreally rough way of getting product info specified for devices
\n[24-Sep-2009 11:39:56] <rmatte> I'm going to write up a quick\nblog post with the steps so that people can use them until the issue\ngets resolved
[24-Sep-2009 11:51:23] <Brad_K>\netank: one thing i've discovered about TestGenn4Web is that there is at\nleast one type of dialog box that browsers can open up that isn't seen\nduring the capture session, so there's no way to model that in your\nTwill script. At least, not so far as I can tell.
[24-Sep-2009 11:53:25] <Brad_K> etank: see http://developer.spikesource.com/forums/viewtopic.php?p=2585#2585
\n[24-Sep-2009 11:54:53] <etank> Brad_K: do you know if the\nZenWebTx pack can handle it if the site being queried returns the data\nin a popup window?
[24-Sep-2009 11:55:17] <etank> that is how our site works and im wondering if it would cause it to hang in any way
\n[24-Sep-2009 11:56:35] <Brad_K> etank: don't know. there may be\na way to get Twill to handle that situation, even if testgen4web can't\ncapture it. i have not yet gotten that far.
[24-Sep-2009 11:57:04] <Brad_K> if you take the script as generated by testgen4web and replay that, the login definitely hangs for me.
[24-Sep-2009 11:57:05] <etank> yeah me either. i just started playing with twill today
[24-Sep-2009 11:57:12] <chudler> etank: what method does the site use to popup? twill does not interpret javascript.
[24-Sep-2009 11:57:34] <etank> chudler: good question. im not the one who designed the site so im not sure.
[24-Sep-2009 11:57:39] <etank> i will look into it though
[24-Sep-2009 12:00:20]\n<rmatte> Brad_K:\nblogs/ryanmatte/2009/09/24/specifying-hardware-and-software-product-info-in-245-until-bug-is-fixed
[24-Sep-2009 12:16:17] <Brad_K> rmatte: thanks!
[24-Sep-2009 12:21:03]\n<etank> we are doing a IT Security review of Zenoss here today.\ndoes anyone have a link to the ports that it uses to gather data?
\n[24-Sep-2009 12:21:24] <etank> i know most of them i think but it\nwould be better to have product documentation to back it up
[24-Sep-2009 12:32:26] <rmatte> etank: it really depends on what kind of data you're collecting
[24-Sep-2009 12:32:38] <rmatte> it uses standard SNMP, Syslog, and mail ports
[24-Sep-2009 12:32:42] <rmatte> as well as SSH
[24-Sep-2009 12:32:54] <rmatte> It can also do WMI monitoring, which is fairly random portwise
[24-Sep-2009 12:33:22] <rmatte> It uses ports 8080 for the web interface by default
\n[24-Sep-2009 12:33:46] <rmatte> and it uses some other ports for\ncommunication between the various daemons and zenhub, don't know it off\nthe top of my head
[24-Sep-2009 12:33:49] <rmatte> plus sql ports
[24-Sep-2009 12:33:52] <rmatte> there's a wide range
[24-Sep-2009 12:34:06] <rmatte> I'm sure there's a port for zenhub to communicate with zope as well
[24-Sep-2009 12:37:07] <rmatte> I don't know of any tech sheet that lists them all
[24-Sep-2009 12:37:07] <Brad_K> other ports are like 8079 and 8091
[24-Sep-2009 12:37:29] <rmatte> yeh, that sounds about right, somewhere in that range
[24-Sep-2009 12:38:28] <Brad_K> and 8090, 6010 8081, 8100, 8789
[24-Sep-2009 12:38:42] <Brad_K> which lsof
[24-Sep-2009 12:38:52] <Brad_K> oops. wrong window. ;-)
[24-Sep-2009 12:39:20] <rmatte> hehe
\n[24-Sep-2009 12:39:28] <chudler> as rmatte mentioned, outgoing\nunsolicited monitoring traffic from zenoss can *potentially* be over\nany port, especially considering the portscan plugin :-p
[24-Sep-2009 12:39:34] <rmatte> /usr/bin/lsof
[24-Sep-2009 12:40:05] <rmatte> yeh, you really need a system that isn't too locked down outgoing to use Zenoss properly
[24-Sep-2009 12:40:17] <Brad_K> rmatte: $ which lsof
[24-Sep-2009 12:40:18] <Brad_K> lsof: Command not found.
[24-Sep-2009 12:40:24] <rmatte> that sucks
[24-Sep-2009 12:40:25] <rmatte> lol
[24-Sep-2009 12:40:52] <kisielk> is there a guide somewhere to optimizing zenoss / figuring out what part of it is slow?
\n[24-Sep-2009 12:41:51] <kisielk> some times parts of it seem to\nget dog slow. Like if I type a hostname in to the search box, it can\ntake several minutes to get me a result
[24-Sep-2009 12:42:03] <rmatte> docs/DOC-2521
[24-Sep-2009 12:42:07] <Brad_K> that's probably a DNS issue.
\n[24-Sep-2009 12:42:41] <rmatte> that explains how to tweak it\nperformancewise, but it sounds like you may have a different issue
[24-Sep-2009 12:42:55] <rmatte> what specs on the box?
[24-Sep-2009 12:42:56] <kisielk> well, it's only sometimes
[24-Sep-2009 12:42:59] <rmatte> how many devices monitored?
[24-Sep-2009 12:43:01] <chudler> yeah sounds like something up with the zope db..
[24-Sep-2009 12:43:02] <kisielk> if I just restart Zenoss, it's usually fast
[24-Sep-2009 12:43:07] <kisielk> hm
[24-Sep-2009 12:43:13] <rmatte> is it only certain pages that are slow? such as the zproperties page?
[24-Sep-2009 12:43:26] <rmatte> is it digging in to swap?
[24-Sep-2009 12:43:42] <kisielk> It's a quad core AMD 64
[24-Sep-2009 12:43:50] <kisielk> no swap usage
[24-Sep-2009 12:43:52] <rmatte> 3.0GHz?
[24-Sep-2009 12:43:58] <kisielk> naw, 2.2
[24-Sep-2009 12:44:02] <rmatte> k
[24-Sep-2009 12:44:07] <kisielk> the disks are just SATA, in RAID 1
[24-Sep-2009 12:44:07] <rmatte> how many devices?
[24-Sep-2009 12:44:13] <kisielk> about 250
[24-Sep-2009 12:44:29] <rmatte> ok, that should be relatively fine then
[24-Sep-2009 12:44:35] <rmatte> is the box doing anything other than Zenoss?
[24-Sep-2009 12:44:48] <kisielk> no, well, syslog-ng
[24-Sep-2009 12:44:51] <rmatte> that's fine
[24-Sep-2009 12:44:56] <kisielk> but that's it
[24-Sep-2009 12:44:57] <rmatte> can you pastebin your zenpack list?
[24-Sep-2009 12:45:01] <rmatte> just do a zenpack --list
[24-Sep-2009 12:45:11] <rmatte> I just want to check something
[24-Sep-2009 12:45:53] <kisielk> there's only 3
[24-Sep-2009 12:46:08] <kisielk> HPProcurve, HPMonitor, and our own which I think is actually empty
[24-Sep-2009 12:46:47] <rmatte> k
[24-Sep-2009 12:47:03] <rmatte> well, then I'd just boil it down to performance tweaks need to be made
[24-Sep-2009 12:47:12] <kisielk> alright, I will try the ones on the page
\n[24-Sep-2009 12:47:23] <rmatte> I have a Zenoss server monitoring\n340 devices, and it was fairly slow until I did some of the tweaks in\nthat doc
[24-Sep-2009 12:47:38] <rmatte> you'll at least want 3 zenhub workers and some daemons running parallel
[24-Sep-2009 12:47:52] <rmatte> and maybe some slight zopedb tweaking (I wouldn't go with values as high as in the doc)
[24-Sep-2009 12:48:25] <rmatte> Also, couldn't hurt to upgrade that processor
[24-Sep-2009 12:48:44] <rmatte> you can pick up a 2.9GHz phenom for like $80 or $90 these days
[24-Sep-2009 12:50:20] <rmatte> actually, that's the athlon 2
[24-Sep-2009 12:50:31] <rmatte> the phenoms are still quite pricy, thought they would have gone down by now
[24-Sep-2009 12:50:40] <rmatte> I have a 2.5GHz phenom in my PC at home
[24-Sep-2009 12:50:59] <Brad_K> wouldn't you want a Barthelona instead? ;-)
[24-Sep-2009 12:51:00] <rmatte> a 3.4GHz phenom costs $289.99
[24-Sep-2009 12:51:12] <rmatte> meh
[24-Sep-2009 12:51:37] <rmatte> quad-core is quad-core
[24-Sep-2009 12:51:59] <rmatte> plus barcelona is AM3, and he may only have an AM2 board
[24-Sep-2009 12:52:00] <rmatte>
[24-Sep-2009 12:52:38] <cwj> i think amd just announced some new quad core that runs on AM2 and AM3
[24-Sep-2009 12:52:45] <cwj> i forget the details though
[24-Sep-2009 12:52:48] <Brad_K> ahh, but does he have enough rain?
[24-Sep-2009 12:52:55] <cwj> my friend was yacking at me about it
[24-Sep-2009 12:53:32] <rmatte> yeh, that's the Barcelona I believe (so maybe it does support AM2)
[24-Sep-2009 12:54:01] <rmatte> Brad_K: you'd probably need a box fan to cool it
[24-Sep-2009 12:54:03] <kisielk> well, if we were upgrading, we'd swap out the whole machine
[24-Sep-2009 12:54:21] <kisielk> so that's probably about $1k worth of hardware, if not 2
[24-Sep-2009 12:54:27] <Brad_K> or move it to Spain?
[24-Sep-2009 12:54:28] <rmatte> My 2.5 phenom runs so hot compared to my old 2.5GHz Athlon II
[24-Sep-2009 12:54:35] <kisielk> I'd probably put it on SAS disks
[24-Sep-2009 12:54:46] <rmatte> the old chip idled at about 36* C, the new one idles at like 46*C
[24-Sep-2009 12:54:51] <rmatte> 10 degree difference
[24-Sep-2009 12:54:54] * Brad_K enjoys being silly sometimes
[24-Sep-2009 12:55:43] <rmatte>
[24-Sep-2009 12:56:09] <rmatte> and that's with a huge copper heatsink on it
[24-Sep-2009 12:56:30] <Brad_K> you put a huge copper heatsink on your tongue?
[24-Sep-2009 12:56:36] <rmatte>
[24-Sep-2009 12:56:49] <rmatte> http://dmon.org/graphics/nemesis/IMG_0131.JPG <--- This
\n[24-Sep-2009 12:57:10] <rmatte> (My graphics card is a lot bigger\nnow than the one that was there in that photo too, so less airflow in\nthe case)
[24-Sep-2009 12:57:42] <rmatte> http://dmon.org/graphics/nemesis/IMG_0130.JPG
[24-Sep-2009 12:57:45] <rmatte> better photo
[24-Sep-2009 13:04:16] <twm1010> Anyone using the II6 Zenpack?
[24-Sep-2009 13:06:15] <twm1010> what would -1,* do in the RPN field of a datapoint in a graph
[24-Sep-2009 13:14:35] <rmatte> multiply whatever is after it by negative 1
[24-Sep-2009 13:14:42] <rmatte> RPN is reverse
[24-Sep-2009 13:15:14] <twm1010> odd, why would the bytes received datapoint be set that way
[24-Sep-2009 13:15:15] <rmatte> start from the right and read to the left
[24-Sep-2009 13:15:20] <twm1010> its a derive type
[24-Sep-2009 13:15:44] <twm1010> the bytes sent isn't manipulated at all
[24-Sep-2009 13:15:47] * rmatte shrugs
[24-Sep-2009 13:15:58] <twm1010> hrm... no matter, zenpack works nicely
[24-Sep-2009 13:31:08] <kisielk> rmatte: hm, so even after applying all those tweaks, it's still pretty slow :/
[24-Sep-2009 13:31:19] <kisielk> I just clicked a host in the dashboard, took over 30 seconds to go to the page
\n[24-Sep-2009 13:31:36] <kisielk> weird thing is, I didn't see\nanything thrashing on the system in iostat, or a lot of activity with\ntop..
[24-Sep-2009 13:31:53] <kisielk> so I don't know where the slowness is coming from
[24-Sep-2009 13:33:11] <chudler> curious. how large is the Data.fs file?
[24-Sep-2009 13:37:54] <rmatte> $ZENHOME/var/Data.fs
[24-Sep-2009 13:58:50] <twm1010> mine is 288947716
[24-Sep-2009 13:59:40] <twm1010> that's 288MB or so, right?
[24-Sep-2009 14:29:34] <rmatte> correct
[24-Sep-2009 14:30:31] <rmatte> more like 282MB but yeh
[24-Sep-2009 14:30:39] <rmatte> you can always do an ls -alh
[24-Sep-2009 14:30:51] <rmatte> to have it convert it to kilobytes, megabytes, or gigabytes for you
[24-Sep-2009 14:30:53] <rmatte>
[24-Sep-2009 14:32:13] <kisielk> chudler: 77 MB
[24-Sep-2009 14:32:33] <chudler> nice and small... no problem there I suppose.
[24-Sep-2009 14:32:55] <twm1010> isn't there a routine you can run through to condense that?
[24-Sep-2009 14:33:35] <chudler> there is... somewhere in the zenoss site :-)
[24-Sep-2009 15:32:40] <chemist> evening
\n[24-Sep-2009 15:34:33] <chemist> rmatte:\nhttp://dev.zenoss.com/trac/ticket/5578 after the roasting I got\nthe other night it looks like I will have to find a plugin to monitor\nthe switches
[24-Sep-2009 15:37:07] <rmatte> lol
[24-Sep-2009 15:37:30] <chemist> any suggestions?
[24-Sep-2009 15:37:37] <chemist> as to what to use
[24-Sep-2009 15:37:48] <rmatte> I'd speak with Matt about that one, they are brushing any tickets that they can aside
[24-Sep-2009 15:38:06] <rmatte> since they want to work on the more critical issues before launch
[24-Sep-2009 15:38:13] <rmatte> err, before release rather
\n[24-Sep-2009 15:38:56] <rmatte> there really should be something\nin that plugin that is capable of continuing on without requiring those\nvalues
[24-Sep-2009 15:39:23] <rmatte> especially since this is fairly common (according to the forums)
[24-Sep-2009 15:40:13] <rmatte> I'll speak with him about it tomorrow if I have a second and he's not too busy
\n[24-Sep-2009 15:41:23] <chemist> my colleague hacked the script\nand it now models the devices and shows graphs but the problem lies in\nnot been able to label the switches
[24-Sep-2009 15:41:38] <rmatte> label the ports you mean?
[24-Sep-2009 15:41:42] <chemist> yes
[24-Sep-2009 15:41:47] <rmatte> do they have labels on them?
[24-Sep-2009 15:42:13] <rmatte> and if you do an snmpwalk, are you able to see the labels under a certain OID?
\n[24-Sep-2009 15:53:11] <chemist> no, but I would be happy editing\nthe names myself rather than having a name of 'Ethernet Interface' for\nall the interfaces
[24-Sep-2009 15:54:17]\n<chemist> when I then click on 'Ethernet Interface' I can see\n/Devices /Network /Switch /switch1.icarus.xxxxx.co.uk /os \n/Ethernet Interface_21
[24-Sep-2009 15:54:26]\n<rmatte> ah, well they are really meant to reflect interface\ndescriptions that are set on the device itself
[24-Sep-2009 15:54:26] <chemist> or Interface_7
[24-Sep-2009 15:54:28] <chemist> etc
[24-Sep-2009 15:54:29] <chemist> etc
[24-Sep-2009 15:54:46] <rmatte> oh
[24-Sep-2009 15:54:48] <chemist> and they are
[24-Sep-2009 15:54:54] <rmatte> like every interface is called Ethernet Interface?
[24-Sep-2009 15:55:06] <kisielk> does the switch not support naming the interfaces?
[24-Sep-2009 15:55:08] <rmatte> there's no like port 1, port 2, etc...?
[24-Sep-2009 15:55:18] <rmatte> kisielk: he has really cheap switches
[24-Sep-2009 15:55:21] <rmatte> netgear
[24-Sep-2009 15:55:25] <chemist> yes, but zenoss wont pick the names up
[24-Sep-2009 15:55:38] <chemist>
\n[24-Sep-2009 15:55:52] <rmatte> chemist: that's more of a problem\nwith the snmp on the switches than anything, Zenoss is not designed to\nlet people manually name interfaces
[24-Sep-2009 15:56:14] <rmatte> quite honestly, it's insane that you'd even have to do something like that
[24-Sep-2009 15:56:26] <rmatte> useless for monitoring too
[24-Sep-2009 15:56:31] <rmatte> "Ethernet Interface is down"
[24-Sep-2009 15:56:35] <rmatte> "Which one!?!?!?!"
[24-Sep-2009 15:56:42] <chemist> tell me about it
[24-Sep-2009 15:56:54] <rmatte> but anyways, you get what you pay for
[24-Sep-2009 15:57:02] <chemist> my boss is no hurry to replace 80+ switches bought in december 08
[24-Sep-2009 15:57:04] <rmatte> those netgears are the sort of thing you'd throw in a boardroom
[24-Sep-2009 15:57:05] <rmatte> lol
[24-Sep-2009 15:57:18] <kisielk> I wouldn't put one in my boardroom
[24-Sep-2009 15:57:22] <kisielk> it's only 10/100
[24-Sep-2009 15:57:33] <rmatte> meh, 100 is fast enough for laptops
[24-Sep-2009 15:57:43] <chemist> with gigabit uplinks I'll have you know
[24-Sep-2009 15:57:44] <rmatte> I mean, how much porn can you possibly download during a meeting?
[24-Sep-2009 15:57:56] <rmatte>
[24-Sep-2009 15:58:05] <kisielk> rmatte: you'd be surprised
[24-Sep-2009 15:58:08] <rmatte> hehe
\n[24-Sep-2009 15:58:29] <rmatte> either way, if you can't see the\ninfo in an snmpwalk then there's no way that Zenoss is going to be able\nto properly model those interfaces
[24-Sep-2009 15:58:37] <rmatte> so in short, you're pretty much screwed
[24-Sep-2009 15:58:45] <rmatte>
[24-Sep-2009 15:59:07] <chemist> but snmpwalking the device shows us at least an index number
[24-Sep-2009 15:59:17] <rmatte> you can monitor for traps, and poll for CPU, memory, and such, but that's pretty much it
[24-Sep-2009 15:59:39] <rmatte> yeh, index number is good and dandy, but Zenoss looks at the port name
[24-Sep-2009 15:59:49] <rmatte> since on every other switch that I've seen the port names actually make sense
[24-Sep-2009 15:59:51] <chemist> ok
[24-Sep-2009 16:00:02] <rmatte> when you login to a cisco device you don't want to know that port index 3 went down
[24-Sep-2009 16:00:12] <rmatte> you want to know that Fa0/3 went down
[24-Sep-2009 16:00:13] <rmatte> or whatever
[24-Sep-2009 16:00:18] <rmatte> hence why it's designed like that
[24-Sep-2009 16:00:27] <chemist> makes sense
[24-Sep-2009 16:00:53] <rmatte> or Se0/3/0
[24-Sep-2009 16:01:00] <rmatte> whatever it happens to be
[24-Sep-2009 16:02:02]\n<rmatte> If you have someone there who is good at python maybe\nthey can modify that plugin and set it up as a new collector plugin\njust for the neatgears
[24-Sep-2009 16:02:10] <rmatte> and have it use the interface number is place of the name
[24-Sep-2009 16:02:16] <rmatte> probably wouldn't be a massive change
[24-Sep-2009 16:02:34] <rmatte> in place of the name*
[24-Sep-2009 16:02:35] <rmatte> rather
[24-Sep-2009 16:03:29] <rmatte> well, I'm 1 hour and 57 minutes to being $870 poorer
[24-Sep-2009 16:03:43] * rmatte taps his foot
[24-Sep-2009 16:03:57]\n<chemist> ok, and this might be quite crude, but would it be\npossible to go into the zodb and manually enter descriptions?
[24-Sep-2009 16:04:04] <chemist> how come?
[24-Sep-2009 16:04:20] <rmatte> paintball marker auction on ebay
[24-Sep-2009 16:04:29] <rmatte> and yes, you could do that via zendmd
[24-Sep-2009 16:04:49] <chemist> hehe
[24-Sep-2009 16:04:55] <rmatte> actually, you could write a script to have it run through and increment the value each time
[24-Sep-2009 16:05:21] <rmatte> and do it on each device
[24-Sep-2009 16:05:30] <rmatte> if you can wait until tomorrow I might be able to cook it up for you
[24-Sep-2009 16:05:38] <rmatte> they are all ethernet interfaces right?
[24-Sep-2009 16:05:51] <chemist> no worries, it would be greatly appreciated
[24-Sep-2009 16:05:56] <chemist> yes
[24-Sep-2009 16:06:09] <rmatte> ports 1 to whatever on each?
[24-Sep-2009 16:06:15] <rmatte> or is there a port 0?
[24-Sep-2009 16:06:32] <chemist> no, 1 to 26
[24-Sep-2009 16:06:55] <rmatte> k
[24-Sep-2009 16:07:08] <rmatte> It'll probably only take like 7 or 8 lines of code to do
[24-Sep-2009 16:08:10]\n<rmatte> just have to have it loop through the interfaces on each\ndevice and name them whatever with the number attached and increment\nthe number each time, then reset the number each time it hits the next\ndevice
[24-Sep-2009 16:08:31] <rmatte> shouldn't be difficult
[24-Sep-2009 16:08:41] <rmatte> my brain is fried for today though, so not in the mood for scripting
\n[24-Sep-2009 16:12:34] <chemist> that would be great, the\nswitches are in around 40 different locations and they 'should' all\nhave the same devices plugged in to each port so just a number would\ngive us a clue as to what's what
[24-Sep-2009 16:13:12] <rmatte> ok, so you just want them named 1, 2, 3, 4, etc...?
[24-Sep-2009 16:13:23] <chemist> yes
[24-Sep-2009 16:22:59] <rmatte> finished
[24-Sep-2009 16:23:01] <rmatte> try this...
[24-Sep-2009 16:23:14] <rmatte> http://pastebin.com/m4f9dd4ba
[24-Sep-2009 16:23:22] <rmatte> paste that in to zendmd
[24-Sep-2009 16:24:20] <rmatte> (I was bored, decided to script it)
[24-Sep-2009 16:24:29] <rmatte> just tested it and it appears to work nicely
[24-Sep-2009 16:24:55] <rmatte> hopefully it doesn't screw up the interface graphing by changing those values
[24-Sep-2009 16:25:41] <rmatte> that'll apply to every single device
\n[24-Sep-2009 16:25:58] <rmatte> so if you have other devices in\nthere you might just want to remove them and add them back later
[24-Sep-2009 16:26:37] <rmatte> or you can do it to an individual device with http://pastebin.com/m70d39e1e
[24-Sep-2009 16:26:45] <rmatte> replace 30.30.254.33 with an ip or a devicename
[24-Sep-2009 16:27:03] <rmatte> oh crap wait
[24-Sep-2009 16:27:13] <rmatte> that'll make one of the ports port 0
[24-Sep-2009 16:27:16] <rmatte> forgot about that
[24-Sep-2009 16:27:17] <rmatte> one sec
[24-Sep-2009 16:27:47] <rmatte> there
[24-Sep-2009 16:27:48] <rmatte> http://pastebin.com/m7e56df87
[24-Sep-2009 16:27:55] <rmatte> top for single, bottom for all
[24-Sep-2009 16:29:44] <rmatte> I'd recommend going through and locking down the interfaces on all of the devices after
[24-Sep-2009 16:29:58] <rmatte> because it'll change them when it remodels if they aren't locked
\n[24-Sep-2009 16:30:22] <rmatte> the other issue you might run in\nto is that it decides to add the other interfaces in in addition to\nthese once they are renamed
[24-Sep-2009 16:30:28] <rmatte> I'd just lock all of the devices down
[24-Sep-2009 16:30:55] <rmatte> From the device list select all and click on Lock devices...
[24-Sep-2009 16:31:05] <rmatte> then select Lock from deletion and updates
[24-Sep-2009 16:32:10] <chemist> do I have to enter anything at the end?
[24-Sep-2009 16:32:18] <rmatte> just hit enter at the end
[24-Sep-2009 16:32:19] <rmatte> oh
[24-Sep-2009 16:32:26] <rmatte> in zendmd you'll have to type commit()
[24-Sep-2009 16:32:27] <rmatte> afterwards
[24-Sep-2009 16:32:31] <chemist> ok
[24-Sep-2009 16:32:37] <rmatte> since zendmd only simulates stuff until you commit it
[24-Sep-2009 16:34:15] <rmatte> 5 lines of code, shorter than I thought
\n[24-Sep-2009 16:34:30] <rmatte> (then again I was considering\nsome lines to make it a standalone script, but there's really no need)
[24-Sep-2009 16:34:53] <chemist> http://pastebin.com/mdb6d85f
[24-Sep-2009 16:34:58] <chemist> this is what I get
[24-Sep-2009 16:35:18] <rmatte> yeh, you're doing it wrong lol
[24-Sep-2009 16:35:27] <rmatte> ok, first off, you're not using the modified version of the script
[24-Sep-2009 16:35:49] <rmatte> http://pastebin.com/m7e56df87
[24-Sep-2009 16:35:55] <rmatte> use the script at the bottom there
[24-Sep-2009 16:36:03] <rmatte> otherwise you'll have interface 0 instead of interface 1
[24-Sep-2009 16:36:13] <rmatte> secondly, what you want to do is paste the script, then hit enter
[24-Sep-2009 16:36:16] <rmatte> wait for it to run through
[24-Sep-2009 16:36:24] <rmatte> once it has run through then you type commit()
[24-Sep-2009 16:36:32] <rmatte> you tried to type commit() as part of the script itself
[24-Sep-2009 16:36:48] <rmatte> when you just see ...
[24-Sep-2009 16:36:52] <rmatte> after pasting the script
[24-Sep-2009 16:36:53] <rmatte> hit enter
[24-Sep-2009 16:38:06] <rmatte> or use the script at the top of you're doing it to a specific device obviously
[24-Sep-2009 16:39:21] <rmatte> did it work?
[24-Sep-2009 16:40:21] <rmatte> (I'm leaving at a little before 6 btw)
[24-Sep-2009 16:40:58] <chemist> err, no
[24-Sep-2009 16:41:05] <rmatte> oh?
[24-Sep-2009 16:41:07] <chemist> and it's hosed tha
[24-Sep-2009 16:41:13] <chemist> that device
[24-Sep-2009 16:41:21] <rmatte> but the command did run through right?
[24-Sep-2009 16:41:46] <chemist> yes
[24-Sep-2009 16:41:57] <rmatte> oh, you know what? the device links to the interfaces by name
[24-Sep-2009 16:42:03] <rmatte> so by changing the names we screw up the links
[24-Sep-2009 16:42:06] <rmatte> so yeh, this won't work
[24-Sep-2009 16:42:20] <rmatte> and I honestly can't think of anything that will because of that, hmmm
[24-Sep-2009 16:42:25] <rmatte> I'll have to ask around tomorrow
[24-Sep-2009 16:42:37] <chemist> thanks anyway
[24-Sep-2009 16:42:39] <chemist>
[24-Sep-2009 16:42:40] <rmatte> np
[24-Sep-2009 16:43:13] <rmatte> I might just be changing the wrong value, I'll have to check it out tomorrow
[24-Sep-2009 16:43:22] <chemist> that's cool
[24-Sep-2009 16:45:09] <rmatte> anyways, have a good night
[24-Sep-2009 16:45:18] <chemist> cheers
[24-Sep-2009 18:36:18] <kisielk> this slowness is unbearable :/
[24-Sep-2009 18:48:39] <davetoo> nuh?
[24-Sep-2009 18:55:01] <kisielk> well, still trying to figure out why my Zenoss is so damn slow
[24-Sep-2009 19:20:40] <leafnunes> kisielk: i don't suppose the host is under load?
[25-Sep-2009 00:00:46] [disconnected at Fri Sep 25 00:00:46 2009]
[25-Sep-2009 00:00:46] [connected at Fri Sep 25 00:00:46 2009]
[25-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[25-Sep-2009 04:40:24] <ilejn> hello
[25-Sep-2009 06:41:39] <ilejn> hello
[25-Sep-2009 08:29:04]\n<hackeron> hey, I have 1 filesystem that's quite large and zenoss\nsends me messages like: [zenoss] 127.0.0.1 disk space threshold: 92.4%\nused (347.5GB free)
[25-Sep-2009 08:29:20] <hackeron> anyway to stop it from sending those for 1 specific partition?
[25-Sep-2009 08:29:42] <hackeron> seems kinda silly, 350GB free is not almost full
[25-Sep-2009 08:30:11] <hackeron> I know the FS performance slows down a lot, but it's non critical storage
[25-Sep-2009 08:34:47] <twm1010> morning
[25-Sep-2009 09:19:30] <rmatte> hackeron: yes
\n[25-Sep-2009 09:19:45] <rmatte> hackeron: just go to templates\nfor that drive, create a local copy, and change the threshold value for\nit
[25-Sep-2009 09:19:48] <rmatte> quite simple
[25-Sep-2009 09:20:00] <rmatte> bump it up to like 95 or 97%
[25-Sep-2009 09:49:42] <hackeron> rmatte: cool, thanks, will try
[25-Sep-2009 10:00:18] <rmatte> g'day cgibbons
[25-Sep-2009 10:00:19] <jb> cgibbons!
[25-Sep-2009 10:00:24] <jb> just the man I was looking for.
[25-Sep-2009 10:00:28] <rmatte> oh oh
[25-Sep-2009 10:00:31] <rmatte> you're in trouble now
[25-Sep-2009 10:00:32] <cgibbons> whatever it is, i didn't check it in!
[25-Sep-2009 10:00:37] <rmatte> lol
[25-Sep-2009 10:00:51] <jb> got a winperf related question again :/
[25-Sep-2009 10:01:41] <jb> i have one performance object in which every counter generaters an error
[25-Sep-2009 10:01:56] <cgibbons> ooh, which one?
[25-Sep-2009 10:02:24] <jb> http://pastebin.com/m2bfe1190
[25-Sep-2009 10:02:35] <jb> its "ISA Server Web Proxy"
[25-Sep-2009 10:02:49] <jb> #
[25-Sep-2009 10:02:50] <jb> ERROR:zen.winperf.PerfRpc:global name 'DEBUG' is not defined
[25-Sep-2009 10:02:54] <cgibbons> argh that bug
[25-Sep-2009 10:03:22] <jb> oh, so its known
[25-Sep-2009 10:03:26]\n<cgibbons> that one slipped through, if you turn on debug you get\na crash independent of whatever is going on. can you go to line 244 of\n/opt/zenoss/ZenPacks/ZenPacks.zenoss.ZenWinPerf-2.1.5-py2.4.egg/ZenPacks/zenoss/ZenWinPerf/zenwinperf.py\nand make a quick change?
[25-Sep-2009 10:03:35] <cgibbons> (it's fixed in a newer version of the zenpack, wherever that might be)
[25-Sep-2009 10:03:36] <jb> sure
[25-Sep-2009 10:03:53] <cgibbons> where it say <= DEBUG change it to <= logging.DEBUG
[25-Sep-2009 10:04:01] <ckrough> morning
[25-Sep-2009 10:04:24] ckrough is now known as Guest58427
[25-Sep-2009 10:04:25] <jb> 244 = values = driver.next()
[25-Sep-2009 10:05:00] <cgibbons> hrm maybe 398?
[25-Sep-2009 10:05:04]\n<jb>\nzenoss@fc-zenoss01:/opt/zenoss/ZenPacks/ZenPacks.zenoss.ZenWinPerf-2.1.5-py2.4.egg/ZenPacks/zenoss/ZenWinPerf$\ngrep DEBUG zenwinperf.py
[25-Sep-2009 10:05:04] <jb> from pysamba.library import DEBUGLEVEL
[25-Sep-2009 10:05:04] <jb> DEBUGLEVEL.value = 99
[25-Sep-2009 10:05:04] <jb> if self.log.isEnabledFor(logging.DEBUG):
[25-Sep-2009 10:05:05] <cgibbons> prolly 398
[25-Sep-2009 10:05:09] Guest58427 is now known as ckrough
[25-Sep-2009 10:05:23] <cgibbons> should look like: if self.logLevel <= DEBUG:
[25-Sep-2009 10:05:24] <cgibbons> somewhere
[25-Sep-2009 10:05:35] <ckrough> isnt there a core zenpack download for 2.4.5? I dont see it listed on the download page
[25-Sep-2009 10:05:43] <jb> hmm
[25-Sep-2009 10:05:52] <jb> if self.log.isEnabledFor(logging.DEBUG):
[25-Sep-2009 10:05:58] <jb> thats 267
[25-Sep-2009 10:05:59] <cgibbons> maybe just search for <= DEBUG in the file, you should come across it
[25-Sep-2009 10:06:12] <jb> nothing matching that
[25-Sep-2009 10:06:15] <cgibbons> huh
[25-Sep-2009 10:06:23] <cgibbons> lemme see if i've got that version handy
[25-Sep-2009 10:06:44] <cgibbons> you're on 2.4.5 of the main product yet ?
[25-Sep-2009 10:06:49] <jb> yep
[25-Sep-2009 10:08:22] <cgibbons> oh no wonder
[25-Sep-2009 10:08:23] <cgibbons> ahem
[25-Sep-2009 10:08:47] <cgibbons> try PerfRpc.py instead of zenwinperf.py
[25-Sep-2009 10:08:52] <cgibbons> and let me get some more coffee....
[25-Sep-2009 10:09:03] <ckrough> Anyone know where zenoss-core-zenpacks-2.4.5 lives on the website?
[25-Sep-2009 10:09:12] <rmatte> hmmm, should we be worried that the new king crab beta version code ends in -404?
[25-Sep-2009 10:09:19] <rmatte> 404 = error
[25-Sep-2009 10:09:21] <rmatte> :O
[25-Sep-2009 10:12:12] <rmatte> ckrough: is that a zenpack with all the core zenpacks bundled in to it?
[25-Sep-2009 10:12:46] <ckrough> rmatte: yeah, I found it on sourceforge
[25-Sep-2009 10:12:50] <ckrough> its missing on the community site
[25-Sep-2009 10:12:56] <ckrough> v2.4.5 that is
[25-Sep-2009 10:13:09] <rmatte> ah
[25-Sep-2009 10:13:14] <rmatte> mention it to Matt
[25-Sep-2009 10:13:18] <ckrough> ya ya
[25-Sep-2009 10:13:39] <rmatte> I'll be happy when he managed to get all of the ZenPacks unzipified on the new site
\n[25-Sep-2009 10:13:58] <rmatte> there are constant posts: "I\ncan't install this pack" ... "You have to unzip it first" ... "Oh!"
[25-Sep-2009 10:14:08] <rmatte> manages*
[25-Sep-2009 10:14:29] <rmatte> well, time to test upgrading between beta versions
[25-Sep-2009 10:15:48] <cgibbons> it should definitely be in that one file, jb
[25-Sep-2009 10:27:47] <rmatte> ooooh, I love the new event console layout
[25-Sep-2009 10:27:54] <rmatte> (the beta 3 one)
\n[25-Sep-2009 10:29:12] <cgibbons> i can't wait until the rest of\nthe UI gets the same treatment. we spent a lot of time on the screen\ndesign so it'll be nice to see it done.
[25-Sep-2009\n10:29:22] <rmatte> hmmm, might have to speak with Ian about the\nway historic events are handled though, can't select a range properly\nanymore
[25-Sep-2009 10:29:54] <rmatte> like the filters for first time and last time don't work as a range
[25-Sep-2009 10:30:02] <rmatte> so there's no way to see events older than a day
\n[25-Sep-2009 10:30:56] <rmatte> the filters for first time and\nlast time are actually pretty useless right now, since they have to\nmatch a specific time and date to pair up
[25-Sep-2009 10:31:17] <rmatte> hmmm, you can use wildcards though, but that's not really good enough
[25-Sep-2009 10:31:23] <rmatte> there need to be a place to specify a range
[25-Sep-2009 10:31:24] <cgibbons> i've passed that on to him, so he might show up
[25-Sep-2009 10:31:26] <rmatte> needs*
[25-Sep-2009 10:31:30] <rmatte> k cool
[25-Sep-2009 10:31:32] <rmatte> thanks
[25-Sep-2009 10:31:57] <rmatte> oooooh, the new severity colors are very nice
[25-Sep-2009 10:32:09] <rmatte> severity row colors
[25-Sep-2009 10:32:38] <rmatte> looks like the gun metal colors that I suggested but a bit more pastelized
[25-Sep-2009 10:36:06] <jb> cgibbons: sorry, had to step away for a minutue
[25-Sep-2009 10:42:20] <rmatte> yeh, the event console is wayyyy less buggy in the new beta
[25-Sep-2009 10:42:27] <rmatte> it's actually usable
\n[25-Sep-2009 10:44:50] <rmatte> the only thing to add which would\nmake it perfect are fields to select an actual range at the top, if Ian\npops on I can explain how I envision it
[25-Sep-2009 10:47:35] <rmatte> hmmm, just found a bug in it
[25-Sep-2009 10:49:17] <rmatte> actually weird, now it's working fine
[25-Sep-2009 10:50:12] <rmatte> ooooh, they cleaned up the details too with alternating colors for the rows
\n[25-Sep-2009 10:51:19] <rmatte> but they still haven't added a\nseparate section for trap details, they just get bunched in to the rest\nof the details
[25-Sep-2009 10:51:22] <rmatte> that needs to be fixed
[25-Sep-2009 10:52:33] <rmatte> other than those 2 things it's looking really good
\n[25-Sep-2009 10:55:21] <eidolon> hi folks - zenoss 2.4.5 new\ninstallation here. looks like it's all started - i can telnet\nlocalhost 8080 and get an answer, but remote connects are failing. i'\nmguessing zope is not bound to eth0 - how can i check this?
[25-Sep-2009 10:56:58] <jb> cgibbons: still here?
[25-Sep-2009 10:57:04] <cgibbons> yup
[25-Sep-2009 10:57:10] <jb> # first learn how big the data is
[25-Sep-2009 10:57:10] <jb> if self.logLevel <= logging.DEBUG:
[25-Sep-2009 10:57:11] <jb> self.log.debug("Learning counter object ids")
[25-Sep-2009 10:57:12] <jb> line 60
[25-Sep-2009 10:57:15] <jb> of PerfRPC.py
[25-Sep-2009 10:57:21] <jb> PerfRpc.py rather
[25-Sep-2009 10:57:36] <jb> and line 80
[25-Sep-2009 10:58:14] <cgibbons> let's also double check line 398 of winreg.py in that directory
[25-Sep-2009 10:58:29] <cgibbons> i think that might be the one that's messed up
[25-Sep-2009 10:58:38] <jb> hrm, what should I search for?
[25-Sep-2009 10:59:04] <cgibbons> yeah in fact i just double checked in that file here for that version, that one should look like <= DEBUG
[25-Sep-2009 10:59:12] <cgibbons> but needs to be logging.DEBUG
[25-Sep-2009 10:59:16] <jb> found it
[25-Sep-2009 10:59:25] <jb> 398
[25-Sep-2009 10:59:35] <jb> ok, made the change.
\n[25-Sep-2009 10:59:52] <cgibbons> okay, try rerunning zenwinperf\nin debug and now you should at least get some useful info
[25-Sep-2009 11:00:32] <jb> zenwinperf run -v 20 cool?
[25-Sep-2009 11:00:49] <cgibbons> yep. eventually we may need to do a -v 5 but we'll see if 20 workswell enough
[25-Sep-2009 11:01:22] <davetoo> oh is there actually stuff logged at < 10?
[25-Sep-2009 11:01:44] * davetoo is annoyed at the pysamba Polling: messages at 10
[25-Sep-2009 11:01:56]\n<cgibbons> yeah, rarely. i've been doing it in some of the\nlow-level windows daemons because some of that debug is useless for\nsupport but useful for me.
[25-Sep-2009 11:02:13]\n<cgibbons> me too, davetoo, in fact those are gone now. i can\ntell you where to move them if you'd like to do it before KC comes out.
[25-Sep-2009 11:02:19] <rmatte> oh, I thought the higher the number the more verbose
[25-Sep-2009 11:02:22] <rmatte> not the opposite
[25-Sep-2009 11:02:36] <cgibbons> yeah that part is weird
[25-Sep-2009 11:02:37] <davetoo> no, it's actually a threshold for the logging package
[25-Sep-2009 11:03:58] <rmatte> but yeh, 3 changes and the event console is perfect
[25-Sep-2009 11:04:07] <rmatte> (well, 4, but it's a tiny change)
[25-Sep-2009 11:04:53]\n<rmatte> they need a higher res graphic for the details window\n(the half arrow that's directly to the right of the severity icon)
[25-Sep-2009 11:05:05] <rmatte> it looks pixelated while everything else looks nice and smooth, it ruins the look
[25-Sep-2009 11:05:18] <jb> hrm, ok it worked
[25-Sep-2009 11:05:23] <jb> INFO:zen.winperf:Collected value for \\ISA Server Web Proxy\\SSL Client Bytes Received/sec: 0.0
[25-Sep-2009 11:06:18] <jb> thats odd, because nothing changed
[25-Sep-2009 11:06:23] <jb> other than that log statement
[25-Sep-2009 11:06:29] <rmatte> lol
[25-Sep-2009 11:07:09]\n<cgibbons> that area of the code is an edge case, so if you hit\nit (which this counter probably did), it would break collection for it\neven if logging wasn't enabled
[25-Sep-2009 11:07:50] <jb> so you think its fine now?
[25-Sep-2009 11:07:55]\n<davetoo> yes, I just started using the logging package; I wish\nthere was more granularity (by default) between 20 and 0
[25-Sep-2009 11:08:09] <cgibbons> if you're getting the data you want, yep. it'll be officially in the next version of the zenpack.
[25-Sep-2009 11:08:14] <jb> ok nice
[25-Sep-2009 11:08:20] <jb> thanks for the help once again
[25-Sep-2009 11:08:22] <cgibbons> you bet
[25-Sep-2009 11:15:26] <jb> hmm
[25-Sep-2009 11:16:10] <jb> Failure: exceptions.AttributeError: PerfCounterData instance has no attribute 'time'
[25-Sep-2009 11:16:12] <jb> seen that one?
[25-Sep-2009 11:17:40] <cgibbons> hurm nope, lesee
[25-Sep-2009 11:17:52] <jb> can I not redirect output of zenwinperf to a file?
[25-Sep-2009 11:17:57] <cgibbons> sure
\n[25-Sep-2009 11:18:13] <cgibbons> if you want, you can capture\nthe raw data to some file(s) and send them to me and i can run it\nthrough code here and see what it is doing
[25-Sep-2009 11:18:20] <jb> sure
[25-Sep-2009 11:18:36] <cgibbons> use the --captureFile=outputprefix argument
[25-Sep-2009 11:18:42] <jb> one sec.
[25-Sep-2009 11:18:43] <cgibbons> it should make 3 files per device
[25-Sep-2009 11:19:11] <jb> -v 10 ok?
[25-Sep-2009 12:02:13] <ckrough> meh, need to monitor RDP
[25-Sep-2009 12:50:56] <cgibbons> quiet today
[25-Sep-2009 12:52:05] <rhettardo> yes
[25-Sep-2009 12:52:54] <rmatte> I'm just hoping Ian will pop in for a second at some point
[25-Sep-2009 13:47:20] cgibbons_ is now known as cgibbons
[25-Sep-2009 13:51:28] <djack> hellow world
[25-Sep-2009 13:51:52] <djack> somebody alive and kicking here
[25-Sep-2009 13:52:58] <djack> hmmz
[25-Sep-2009 13:54:39] <cain22> hello all
[25-Sep-2009 13:54:50] <djack> hey
[25-Sep-2009 13:55:02] <djack> everybody is a sleep here
\n[25-Sep-2009 13:55:11] <cain22> is there a way to set an event on\ncpu threshold say, if it continues to be over for 5 minutes?
[25-Sep-2009 13:56:51] <djack> it can be done via the counter i thing
[25-Sep-2009 13:57:04] <djack> but how don't know exactly
[25-Sep-2009 13:57:52] <rmatte> you could do it with a transform
[25-Sep-2009 13:58:02] <cain22> counter only esculates the event severity i think
[25-Sep-2009 13:58:04] <ckrough> which WMI Terminal Services zenpack are you guys using, if at all
[25-Sep-2009 13:58:10] <ckrough> seem to be two almost identical
[25-Sep-2009 13:59:12] <cain22> ckrough: I'm not, are they any good?
[25-Sep-2009 13:59:39] <ckrough> dunno, trying to decide which to use now
[25-Sep-2009 14:01:04] <rmatte> ckrough: use Egor's packs
[25-Sep-2009 14:01:23] <ckrough> Egor wrote a terminal service pack? I just was his wmi one
[25-Sep-2009 14:01:38] <rmatte> you mean to count the number of terminal services clients connected?
[25-Sep-2009 14:01:43] <djack> hey rmatte
[25-Sep-2009 14:01:50] <ckrough> yeah
[25-Sep-2009 14:01:57] <rmatte> I was under the impression that his wmi performance template did that
[25-Sep-2009 14:02:05] <rmatte> and if not, you could add the WMI query in in like 2 seconds
[25-Sep-2009 14:02:08] <ckrough> maybe, ill look
[25-Sep-2009 14:02:13] <ckrough> it's a dependency for it anyway
[25-Sep-2009 14:02:28] <rmatte> djack: yes?
[25-Sep-2009 14:05:43] <rmatte> cain22: this is how it'd do it...
[25-Sep-2009 14:06:01] <rmatte> cain22: this is assuming that you don't send alerts on Info events
[25-Sep-2009 14:06:24] <rmatte> cain22: have the threshold come in as an info event, which won't alert
[25-Sep-2009 14:06:46] <rmatte> cain22: then write a transform to set it to whatever severity if it hits a certain count
[25-Sep-2009 14:06:49] <rmatte> so like...
[25-Sep-2009 14:06:58] <rmatte> if evt.count >= 3:
[25-Sep-2009 14:07:09]\n<djack> i hav an output CRITICAL : <datapoints>\n<datapints> i want to make an event on the CRITICAL how.... i\nread some thing on the old site but can't find it again
[25-Sep-2009 14:07:10] <rmatte> evt.severity = 5
[25-Sep-2009 14:08:48] <rmatte> you mean that's what a script is outputting?
[25-Sep-2009 14:09:44] <cain22> and info alerts are cleared if it isn't continuel?
[25-Sep-2009 14:10:07] <cain22> i'm lookin for after a certain amount of time
[25-Sep-2009 14:10:22] <rmatte> it'll eventually clear when the cpu is restored
\n[25-Sep-2009 14:10:25] <djack> i use webinject to check webflow\nwrote a wrapper that gives times but also a status report when there is\nsomething wrong ... that status is CRITICAL : All failed or Warning :\nTest 2 failed ....
[25-Sep-2009 14:10:53] <cain22> rmatte: thanks for the info, i may just have to try that
[25-Sep-2009 14:11:02]\n<rmatte> cain22: well, one polling cycle is 5 minutes, so as soon\nas the count hits 2 you know it's been like that for 5 minutes
[25-Sep-2009 14:11:26] <rmatte> so you could do if evt.count >= 2:
[25-Sep-2009 14:13:06] <cain22> hmm...thanks
[25-Sep-2009 14:13:32] <rmatte> np
[25-Sep-2009 14:22:16] <djack> rmatte do you understand?
[25-Sep-2009 14:24:03] <rmatte> ok, I'm just asking where you'd be pulling that data from
[25-Sep-2009 14:24:07] <rmatte> command based datapoints?
[25-Sep-2009 14:26:16] <djack> i run a script
[25-Sep-2009 14:26:23] <djack> and this is the output
[25-Sep-2009 14:26:35] <djack> WARNING: 1 Tests Failed | CASE 1 = 0.041 CASE 2 = 2.208 TOTAL 2.368
\n[25-Sep-2009 14:27:24] <djack> the CASE 1 and CASE 2 and TOTAL\nwill be mapped in a graph but the WARNING: should be a trigger to add\nan event to that device ....
[25-Sep-2009 14:27:52] <rmatte> ok...
[25-Sep-2009 14:28:08] <rmatte> I've never seen anything that would allow you to do that
[25-Sep-2009 14:28:21] <rmatte> the way I'd do it is instead of having warning I'd just do...
[25-Sep-2009 14:29:07] <rmatte> Test Failed|CASE1=0.041 CASE2=2.208 TOTAL=2.368 STATUS=1
[25-Sep-2009 14:29:14] <rmatte> and have status as either 0, 1, 2, etc...
[25-Sep-2009 14:29:20] <rmatte> then base the severity on that
[25-Sep-2009 14:30:41] <rmatte> datapoints shouldn't have spaces in them as a rule of thumb
[25-Sep-2009 14:30:48] <rmatte> so CASE 1 and CASE 2 isn't good practice
[25-Sep-2009 14:31:03] <rmatte> also, I find that you shouldn't have spaces between the = signs either
[25-Sep-2009 14:31:16] <djack> can be done
[25-Sep-2009 14:32:39] <djack> but there is no way to grep the first part and make an error of it...
[25-Sep-2009 14:33:05] <rmatte> not that I'm aware of, Zenoss' whole Nagios plugin support only seems to look at what's after the |
[25-Sep-2009 14:33:14] <rmatte> from my experience creating scripts for Zenoss
[25-Sep-2009 14:33:49] <rmatte> usually I just put something like OK|CPU1=75 CPU2=46 CPU3=82 CPU4=97
[25-Sep-2009 14:33:59] <rmatte> and then just put "Unknown" if the command fails
[25-Sep-2009 14:33:59] <djack> rmattte can i write event from the script ?
[25-Sep-2009 14:34:38] <rmatte> well, what would you want the event to look like?
[25-Sep-2009 14:34:52] <rmatte> (there's no reason why you can't, just depends on exactly what you expect from it)
\n[25-Sep-2009 14:37:50] <djack> Well basicly i want a warning\nplaced on a device or critical with some information like WARNING Case\n2 Failed <parameter> not found or CRITICAL All cases failed
[25-Sep-2009 14:38:16] <rmatte> and what determines whether 1 case or all cases failed?
\n[25-Sep-2009 14:38:18] <djack> i'm checking a webpage flow i need\nto know and graph the time for every step but also if the content that\ni expect is there
[25-Sep-2009 14:38:37] <djack> the script ....
[25-Sep-2009 14:38:38] <rmatte> (in the output of the script I mean)
[25-Sep-2009 14:38:46] <rmatte> how would Zenoss tell
[25-Sep-2009 14:38:50] <rmatte> you have to think of this logically
\n[25-Sep-2009 14:39:14] <rmatte> you're passing the data off to\nZenoss, how does it know when to generate an event, what type of event\nto generate, what severity to assign to it, and what info to display?
[25-Sep-2009 14:39:32] <djack> i have a value TEST_DONE TEST_OK TEST_NOK
[25-Sep-2009 14:39:54] <rmatte> ok, so that would be replaced by STATUS
[25-Sep-2009 14:40:04] <rmatte> 0, 1, 2
[25-Sep-2009 14:40:34] <djack> 0 = OK 1 = WARNING 2 = CRITICAL
[25-Sep-2009 14:40:41] <rmatte> that works
[25-Sep-2009 14:40:48] <rmatte> ok, and as for what it's displaying
[25-Sep-2009 14:40:55] <rmatte> how would you want the message to show up?
[25-Sep-2009 14:41:06] <djack> how do you mean ?
[25-Sep-2009 14:41:44] <rmatte> like: 'Bleh is really really bad, and CASE 1 is 0.041 and CASE 2 is 2.208'
[25-Sep-2009 14:41:47] <rmatte> or whatever
[25-Sep-2009 14:41:53] <rmatte> how do you want the messages formatted?
[25-Sep-2009 14:42:23] <djack> WARNING Case 1 failed or CRITICAL all caes failed
[25-Sep-2009 14:42:47] <djack> the time for that is not relevant because message here is unrelated from the time
[25-Sep-2009 14:43:01] <rmatte> ok, so basically what you want to do is this then...
[25-Sep-2009 14:43:10] <rmatte> create 2 thresholds for status
[25-Sep-2009 14:43:24] <rmatte> 1 that kicks off if status hits 2
[25-Sep-2009 14:43:28] <rmatte> and 1 that kicks off if it hits 1
[25-Sep-2009 14:43:34] <rmatte> assign appropriate severities to them
[25-Sep-2009 14:43:49] <djack> but no message is possible ?
\n[25-Sep-2009 14:43:51] <rmatte> then use transforms to modify the\nmessage for each when they are generated to whatever you wanty
[25-Sep-2009 14:43:54] <rmatte> want*
[25-Sep-2009 14:44:27] <rmatte> I'll give you an example
[25-Sep-2009 14:45:13] <djack> oki
\n[25-Sep-2009 14:45:27] <rmatte> you'd name them different things,\nso one would be "whatever Warning" the other would be "whatever\nCritical" and we'll just assume you specified event class /Perf/Memory\n(as an example)
[25-Sep-2009 14:45:32] <rmatte> you would go to Events
[25-Sep-2009 14:45:37] <rmatte> navigate to /Perf/Memory
[25-Sep-2009 14:46:07] <rmatte> and as soon as my lab box catches up with me I can continue...
[25-Sep-2009 14:46:11] <rmatte> ok, there we go
[25-Sep-2009 14:46:39] <rmatte> from the dropdown in the top left you'd select More -> Transform
[25-Sep-2009 14:46:58] <rmatte> you'd then add something like this in:
[25-Sep-2009 14:47:54] <djack> found a better way ....
[25-Sep-2009 14:48:04] <djack> via the zenevent ... i can run that from the command line
[25-Sep-2009 14:48:17] <rmatte> explain?
[25-Sep-2009 14:48:33] <djack> docs/DOC-3069
[25-Sep-2009 14:49:05] <djack> thats what i need #1.8.2. Creating Events from the Command Line
[25-Sep-2009 14:49:39] <rmatte> ummm, yes you can do that, but you'll have to set your script up to run itself in cycles
[25-Sep-2009 14:49:46] <rmatte> it's not going to follow Zenoss' monitoring cycles
[25-Sep-2009 14:50:06] <rmatte> hence why I was explaining to do it this way
[25-Sep-2009 14:50:12] <rmatte> anyways
[25-Sep-2009 14:50:13] <rmatte> if evt.summary.startswith('threshold of whatever Warning'):
[25-Sep-2009 14:50:13] <rmatte> evt.summary and evt.message = 'WARNING Case 1 failed'
[25-Sep-2009 14:50:13] <rmatte> if evt.summary.startswith('threshold of whatever Critical'):
[25-Sep-2009 14:50:13] <rmatte> evt.summary and evt.message = 'CRITICAL all Cases failed'
[25-Sep-2009 14:50:18] <rmatte> that's the transform you'd yse
[25-Sep-2009 14:50:23] <rmatte> use*
[25-Sep-2009 14:51:08] <rmatte> up to you which way you want to do it
\n[25-Sep-2009 14:52:16] <djack> not realy i will trigger the\nzenevent from a bash script that gets the time and also the status ...\nwhen a condition is there via IF statement i can run the zenevent ....
[25-Sep-2009 14:52:18] <djack> not ?
[25-Sep-2009 14:53:08] <rmatte> yes, what I'm saying is, how are you going to keep the bash script checking in a cycle?
[25-Sep-2009 14:53:13] <rmatte> are you going to cron it or something?
[25-Sep-2009 14:53:37] <djack> i execute it from a template
[25-Sep-2009 14:53:49] <djack> that i already have checking other things
[25-Sep-2009 14:53:55] <rmatte> oh, to collect performance data, and include that in it
[25-Sep-2009 14:54:00] <rmatte> actually, yeh, that could work
[25-Sep-2009 14:54:00] <djack> yep
[25-Sep-2009 14:54:07] <rmatte> worth a shot
\n[25-Sep-2009 14:54:45] <djack> zenoss is great ;-) all my\nexternal support partners want access to monitor the shit that they\nsold us ;-)
[25-Sep-2009 14:55:04] <rmatte> yeh, zenoss is quite powerful once you get good at developing it
[25-Sep-2009 14:55:19] <djack> haha
\n[25-Sep-2009 14:55:41] <rmatte> we setup IPSLA monitoring for\nsome of our clients (with custom scripts that I wrote) and they are\nsuper impressed because it works better than what they have from this\nother company which they are paying to do just IPSLA graphing
\n[25-Sep-2009 14:55:45] <djack> well i know bash some perl have a\nsmall understanding how zenoss works so it will e a success
[25-Sep-2009 14:56:00] <djack> what does ipsla doe?
\n[25-Sep-2009 14:56:30] <rmatte> basically you configure a core\nrouter (Cisco or whatever) to collect stats for all of the other\nrouters,
[25-Sep-2009 14:56:41] <ckrough> rmatte: are you using ms_wbt_server to monitor terminal services from the outside?
[25-Sep-2009 14:57:24] <djack> hmmm
\n[25-Sep-2009 14:57:29] <rmatte> basically collects round trip\ntime, packet loss dest-source, packet loss source-dest, and jitter
[25-Sep-2009 14:57:40] <rmatte> and then we pull the stats from the core router via SNMP and graph it all
[25-Sep-2009 14:58:02] <rmatte> ckrough: you mean the IpService?
\n[25-Sep-2009 14:58:36] <djack> routers are out our management but\nin the management of our telco and they will not give snmp management
[25-Sep-2009 14:59:19] <rmatte> well, if it's "managed" equipment then they are doing the management for you
\n[25-Sep-2009 14:59:39] <rmatte> the alternative would be to buy\nyour own routers and just connect the internet circuits to that
[25-Sep-2009 15:00:06] <ckrough> rmatte: yeah
[25-Sep-2009 15:01:29] <rmatte> ckrough: well, we monitor some of our terminal servers in a data center from our lab here
[25-Sep-2009 15:01:35] <rmatte> ckrough: why do you ask?
[25-Sep-2009 15:01:57] <ckrough> sec
[25-Sep-2009 15:03:02]\n<ckrough> trying to halp a buddy out, his users complain of rdp\navailability, so we are looking to monitor outside availability (3389)\nand are using wmi to monitoring session information
[25-Sep-2009 15:03:17] <ckrough> rmatte: any suggestions
[25-Sep-2009 15:04:37] <rmatte> the IpService ms-wbt-server monitors port 3389
[25-Sep-2009 15:04:48] <ckrough> rmatte: thats what I using
[25-Sep-2009 15:04:54] <rmatte> ok, so what's the issue?
[25-Sep-2009 15:05:01] <ckrough> does it just check that 3389 is open, or does it do any kind of quizzing of the service
[25-Sep-2009 15:05:16] <ckrough> I dont get what all of the options are for when you look inside that service
[25-Sep-2009 15:05:28] <rmatte> let me check, haven't looked in a while
[25-Sep-2009 15:06:10]\n<ckrough> there is an "Ip Addresses" field, and a string/regex\npair. I want to make sure the service is actually responding, so if you\nare using something in the send string/regex I'd love to steal it
[25-Sep-2009 15:07:36] <rmatte> nah, I'm using default settings, it looks like it just checks if the port is open
[25-Sep-2009 15:07:42] <rmatte> what I'd suggest is this...
[25-Sep-2009 15:08:05]\n<rmatte> continue monitoring via that IpService, monitor the\nsessions as you've been doing, but monitor the service itself via WMI\nas well
[25-Sep-2009 15:08:22] <ckrough> ok. thanks. learning how to interface with WMI now
[25-Sep-2009 15:08:29] <ckrough> for monitoring processes/services
[25-Sep-2009 15:08:40] <rmatte> the service monitoring is easy in Zenoss, read up on it, you'll see
[25-Sep-2009 15:09:01] <ckrough> thanks
[25-Sep-2009 15:09:05] <rmatte> np
[25-Sep-2009 15:23:08] <cgibbons> pouf!
[25-Sep-2009 15:24:16] <rmatte> bah, I win an auction on ebay and the guy is like "Oh sorry, I already sold that locally"
[25-Sep-2009 15:24:21] <rmatte> grrr, I hate people sometimes
[25-Sep-2009 15:24:43] <rmatte> I'd open a case with ebay about it but I'm not in a mean mood today
[25-Sep-2009 15:26:34] <cwj> kill 'em all and let dobbs sort 'em out
[25-Sep-2009 15:27:18] <ckrough> How can I access a WMI datapoing in a graph report?
[25-Sep-2009 15:27:39] <djack> rmatte zensendevent seems to work now how to clear the event ;-)
\n[25-Sep-2009 15:28:47] <rmatte> ckrough: the same way you'd\naccess an SNMP datapoint, there's no difference in the way they store\ndata
[25-Sep-2009 15:31:08] <ckrough> Not clear on what I would put for the DataPoint name. Trying to do a multigraph of active termserv sessions
[25-Sep-2009 15:31:42] <ckrough> I think I see
[25-Sep-2009 15:34:18] <ckrough> ... I love windows
[25-Sep-2009 15:36:00] <ckrough> Anyone know about templates or SNMP on Barracude load balancers?
[25-Sep-2009 15:47:08] <leafnunes> is anyone aware of a zenpack with a rudimentary example of a PythonClient used for data collection?
[25-Sep-2009 16:01:10] <rmatte> leafnunes: like a collector plugin you mean?
[25-Sep-2009 16:01:24] <leafnunes> yes, sorry this is my first crack at zenoss
[25-Sep-2009 16:01:46] <leafnunes> not quite up to date on the terminology, but i think i found what i need to inherit from
[25-Sep-2009 16:02:51]\n<rmatte> like are you basically just trying to write a script to\nimport values in to Zenoss to be graphed & thresholded against?
[25-Sep-2009 16:06:44] <leafnunes> basically. we have a python library for talking to virtuozzo hosts
[25-Sep-2009 16:08:21]\n<leafnunes> we don't want to talk to the virtual machines\ndirectly, but want to ask the host for information about the containers.
[25-Sep-2009 16:09:53] <rmatte> ok, I'm not famialliar with virtuozzo, but I assume it's just some other flavour of virtualization software
[25-Sep-2009 16:09:59] <leafnunes> it looks like i just want to subclass PythonPlugin
[25-Sep-2009 16:10:04] <leafnunes> rmatte: exactly.
[25-Sep-2009 16:10:23] <leafnunes> rmatte: it'd be easier if i just copied ex's example and use ssh
[25-Sep-2009 16:10:36] <leafnunes> (in the VirtualHostMonitor zenpack)
[25-Sep-2009 16:14:57] <rmatte> sorry our internet dropped for a bit
[25-Sep-2009 16:15:55] <leafnunes> rmatte: ah, yes. it's virtualization software
\n[25-Sep-2009 16:16:31] <rmatte> yeh, I didn't miss anything from\nthe conversation, SSHing from work to home and then to irc
[25-Sep-2009 16:16:34] <rmatte> work connection dropped
[25-Sep-2009 16:16:35] <leafnunes> there are two examples in VirtualHostMonitor, but neither is the Python
[25-Sep-2009 16:16:41] <rmatte> just had to reconnect to my screen session
[25-Sep-2009 16:16:45] <leafnunes> gotcha
[25-Sep-2009 16:16:56] <rmatte> I still don't quite understand what your end goal in terms of Zenoss is
[25-Sep-2009 16:17:02] <rmatte> what do you hope to accomplish?
\n[25-Sep-2009 16:18:34] <leafnunes> we don't want to monitor the\nvirtual machines as individual devices. we want the physical hosts as\ndevices, with the containers (or VMs) as children
[25-Sep-2009 16:19:16] <leafnunes> i'm just trying to figure out what kind of collector i'm creating.
[25-Sep-2009 16:23:28] <rmatte> I see
[25-Sep-2009 16:23:36] <rmatte> you're talking about some pretty tricky coding
[25-Sep-2009 16:23:42] <rmatte> I haven't gone that deep in to Zenoss
[25-Sep-2009 16:24:02] <leafnunes> :-)
[25-Sep-2009 16:24:07] <leafnunes> well, that makes me feel better.
[25-Sep-2009 16:28:12] <ocnam> Sounds kind of like the ESX ZenPack. Maybe you can use that as a template.
[25-Sep-2009 16:29:26] <rmatte> well, sounds sort of like that
[25-Sep-2009 16:29:49] <rmatte> but ESX more monitors specific things from each VM (Memory, CPU, etc...)
[25-Sep-2009 16:29:54] <rmatte> instead of everything on each one
[25-Sep-2009 16:30:20] <ocnam> should've said 'vmWare Virutal Machines'
\n[25-Sep-2009 16:30:22] <rmatte> ocnam: why not just add the VMs\nin and create device groups for them named after whatever host they are\non?
[25-Sep-2009 16:30:27] <rmatte> would be a lot less trouble
\n[25-Sep-2009 16:30:50] <ocnam> the one that adds a 'vitural\nmachine' tab and lists all of the guests running on that hosts
[25-Sep-2009 16:34:07] <ocnam> we have ~300 VMs
[25-Sep-2009 16:40:13] <djack> j,,,
[25-Sep-2009 16:40:17] <djack> hmmm i mean
[25-Sep-2009 16:42:09] <leafnunes> ocnam, rmatte: i'm adding a plugin to the VirtualMachinesMonitor
\n[25-Sep-2009 16:42:30] <leafnunes> but i'm not sure that will let\nme group all the containers under their individual hosts
[25-Sep-2009 16:42:43] <leafnunes> er, under their hosts
\n[25-Sep-2009 16:44:45] <djack> i wonder why my graphs keep giving\nnan and i see frs.be_front_CASE_1.rrd: 0.04299999999999999 when i see\nthis in zencommand
[25-Sep-2009 16:45:55] <rmatte> There's a patch for the issue with the edit tab in 2.5!
[25-Sep-2009 16:45:57] <rmatte> YESSSS!
\n[25-Sep-2009 16:46:12] <rmatte> just finishing up some testing on\nmy lab box and then I shall deploy it to my fleet of servers
[25-Sep-2009 16:46:14] <rmatte> muahaha
[25-Sep-2009 17:20:15] <djack> is the 2.5 already production ready?
[25-Sep-2009 17:26:43] <rmatte> it is as of today, finally got a patch to fix the one noticeable bug that was left
[25-Sep-2009 17:26:53] <rmatte> I just finished deploying it on all my systems
[25-Sep-2009 17:27:09] <rhettardo> nice
\n[25-Sep-2009 17:27:20] <rmatte> I might see if they'd be willing\nto add the patch and release a version 2.4.6 (not sure if they will\nthough)
[25-Sep-2009 17:27:30] <rmatte> I'll write a guide for patching it tomorrow anyways
[25-Sep-2009 17:27:34] <rmatte> erm, on Monday rather
[25-Sep-2009 17:33:03] <djack> lol is there an upgrade path let's say from a bin file ?
[25-Sep-2009 17:35:43] <rmatte> you'd basically install 2.4.5 then run "zenpatch 15214" and restart Zenoss
[25-Sep-2009 17:35:51] <rmatte> and you'll have a production ready version of 2.4.5
[25-Sep-2009 17:36:15] <rmatte> I am SOOOOOOO glad that bug is fixed
[25-Sep-2009 17:36:23] <rmatte> I've had engineers whining to me about that for over a month
[25-Sep-2009 17:37:28] <rmatte> that patch totally made my day (compliments of mr chippy)
[25-Sep-2009 17:43:46] <rmatte> well, done for the day, time to flee
[25-Sep-2009 17:43:49] <rmatte> cheers folks
[25-Sep-2009 17:43:59] <djack> cheers
[25-Sep-2009 17:44:04] <djack> have a nice weekend
[25-Sep-2009 17:44:18] <rmatte> you too
[25-Sep-2009 17:46:43] <daMaestro> does anyone know how to fire off the event cleanup thread/
[25-Sep-2009 17:46:58] <daMaestro> we are seeing some DB issues and would like to fire one off when we have a DBA around
[26-Sep-2009 00:00:46] [disconnected at Sat Sep 26 00:00:46 2009]
[26-Sep-2009 00:00:46] [connected at Sat Sep 26 00:00:46 2009]
[26-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[26-Sep-2009 13:34:25] davetoo1 is now known as davetoo
[26-Sep-2009 13:38:05] <metalklesk> hello
[26-Sep-2009 13:56:02] <davetoo> hi
[27-Sep-2009 00:00:46] [disconnected at Sun Sep 27 00:00:46 2009]
[27-Sep-2009 00:00:46] [connected at Sun Sep 27 00:00:46 2009]
[27-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
[27-Sep-2009 04:04:39] <Egyptian[web]_> afternoon all
[27-Sep-2009 04:05:10]\n<Egyptian[web]_> i got the page setting of alerts function to\nsend me an sms via the correct command .. but it comes blank .. any\nhelp anyone?
[27-Sep-2009 08:14:29] <Egyptian[web]_> anybody awake?
[27-Sep-2009 12:34:33] <Egyptian[Home]> evening all
\n[27-Sep-2009 12:35:44] <Egyptian[Home]> i have managed to use the\n"send page" function to put a custom script that send me text sms\nmessages. however, the messages are blank and my logs tell me there\nwere no parameters passed other than the mobile number .. help?
[27-Sep-2009 12:39:02] <Egyptian[Home]> is there a variable like $MESSAGE that i can use too ?
\n[27-Sep-2009 13:17:49] <Diddi> hello. I've just installed the\nhttp monitor zenpack, and followed the instructions to create graphs\nand thresholds... But the graphs doesn't work. Looking in event.log I\nsee
[27-Sep-2009 13:17:51] <Diddi> 2009-09-27T20:12:24 ERROR RenderServer Failed to generate a graph
[27-Sep-2009 13:17:54] <Diddi> Traceback (most recent call last):
[27-Sep-2009 13:17:54] <Diddi> rrdtool.graph(*gopts)
[27-Sep-2009 13:17:57] <Diddi> File "/usr/local/zenoss/zenoss/Products/ZenRRD/RenderServer.py", line 190, in render
\n[27-Sep-2009 13:18:28] <Diddi> I can't find the rrd-files, and I\ncan't find any solution to my problem... i've searched the forums
\n[27-Sep-2009 13:21:47] <miko_uk> I have just installed the SQL\nserver zenpack, can anyone tell me where the graphs should show up?
[27-Sep-2009 16:12:38] <etank> miko_uk: did you bind the new zenpack to the sql server device and remodel?
\n[27-Sep-2009 16:32:32] <Diddi> hm.. I've added advanced device\ndetail and dell monitor zenpacks.. but I'm not getting anything under\nthe hardware tab. what am I missing?
[27-Sep-2009\n18:12:22] <Diddi> I have problems monitoring processes and\nipservices... when I add a proc/service for monitoring, it works a\ncouple of minutes, and then I get events saying the process is not\nrunning
[27-Sep-2009 18:12:27] <Diddi> but it is..
[27-Sep-2009 18:12:58] <Diddi> currently only mysqld and ssh seems to be affected.. other processes/ipservices works
[27-Sep-2009 18:39:02] <Diddi> when I do a snmpwalk on the host, I get several results of myslqd
[27-Sep-2009 18:51:22] <Diddi> and now all of a sudden I can't monitor snmp ipservice
[27-Sep-2009 18:51:28] <Diddi> why is this?
[28-Sep-2009 00:00:46] [disconnected at Mon Sep 28 00:00:46 2009]
[28-Sep-2009 00:00:46] [connected at Mon Sep 28 00:00:46 2009]
[28-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[28-Sep-2009 00:50:15] <Snyper> question about 2.5, will it support gathering more data from wmi ?
[28-Sep-2009 00:54:00]\n<Snyper> more specific, gathering OS, hardware, software\ninformation. Right now it seems to only be used for services and\npulling data to make graphs..
[28-Sep-2009 00:54:18] <Snyper> Graphs are nice but useless to me.
\n[28-Sep-2009 00:55:55] <Snyper> For 2.4 i tried to dive in and\ncreate my own plugin to do what i needed but could not figure out where\nto place the data.. I have all the wmi queries just dont understand\nwere to put them in zenoss like snmp does
[28-Sep-2009 03:28:22] <Troubadix09> hi all
\n[28-Sep-2009 07:38:46] <Egyptian[Web]> help!! i have been\nstruggling with sending myself text sms for the last 3 days and no\nmatter what i do .. i keep getting blank messages.. help!
[28-Sep-2009 07:40:05]\n<Egyptian[Web]> i m using my own script and have replaced the\npage command with /home/zenoss/bin/sms.sh $RECIPIENT and still blank ..\nthe script works from the command line but not from zenoss
[28-Sep-2009 07:40:06] <Egyptian[Web]> help
[28-Sep-2009 07:40:56] <Diddi> the script has to read text from stdin to work iirc
[28-Sep-2009 07:44:02] <Egyptian[Web]> Diddi: it does .. it sends to $RECIPIENT ..adn then what?
\n[28-Sep-2009 07:44:24] <Egyptian[Web]> i have tried\n$NOTIFICATIONTYPE$: $HOSTALIAS$: $HOSTNAME$: $HOSTADDRESS$:\n$SERVICEDESC$ is $SERVICESTATE$ and they all come out blank
[28-Sep-2009 07:47:05]\n<Diddi> I'm currently using gammu to send sms, and it works\nperfect with zenoss... where are you trying to substitute those\nvariables?
[28-Sep-2009 07:47:23] <Egyptian[Web]> in the page command
[28-Sep-2009 07:47:58] <Egyptian[Web]> Diddi: here http://zenoss:8080/zport/dmd/editSettings
[28-Sep-2009 07:48:00]\n<Diddi> alright, the only thing you should have in the page\ncommand is $RECIPIENT. the text to be sent is set in the alerting rule
[28-Sep-2009 07:48:32] <Diddi> but it sounds like your script only reads text from command line, not stdin
[28-Sep-2009 07:48:57] <Egyptian[Web]> aaaah lerting rule
[28-Sep-2009 07:49:16] <Diddi> and btw, your local url doesn't work for me (:
[28-Sep-2009 07:49:56] <Egyptian[Web]> well of course not but .it does for me plus you can cut and paste and substitute at will
[28-Sep-2009 07:50:34] <Diddi> indeed
\n[28-Sep-2009 08:19:27] <LarsN> Hi everyone. I am running Zenoss\n2.4.5. I have an alerting rule that reads: Event State = New /\nProduction State = Production / Severity >= Error / Send Clear\nMessages False.
[28-Sep-2009 08:21:20] <LarsN>\nWhen a monitored device fails, Zenoss picks it up and sends an alert as\nexpected. The problem is in 5 minutes when it does it's next check,\nit'll send another event, and 5 minutes after that..... I updated the\nrule to send clear messages, and noticed that every 5 minutes it was\nsending a clear, followed by a failure. Our older, "migrated five or\nsix times" zenoss installation running an identical alerting rule\ndoesn't do this, it continues to add
[28-Sep-2009 08:21:27] <LarsN> what am I missing?
[28-Sep-2009 08:25:05] <ckrough> is that monitored thing bouncing?
[28-Sep-2009 08:25:16] <ckrough> ie: it's failing and correcting itself repeatedly
\n[28-Sep-2009 08:25:46] <LarsN> ckrough: pretty sure that's not\nthe case since it does it for everything, including powered off devices.
[28-Sep-2009 08:25:52] <ckrough> ah
[28-Sep-2009 08:26:18] <LarsN> ckrough: If it had just been web services I'd have thought exactly that.
[28-Sep-2009 08:53:43] <ke4qqq> LarsN: what's repeat time?
[28-Sep-2009 08:54:06] <LarsN> ke4qqq: every 5 minutes I get a clear, followed by a fail with the same timestamp.
\n[28-Sep-2009 08:54:10] <ke4qqq> LarsN: the other thing to check\nis that in the event console the count for the event in question is\nbeing incremented.
[28-Sep-2009 08:54:17] <LarsN> ke4qqq: it's not.
[28-Sep-2009 08:54:23] <ke4qqq> ahhh - so it thinks it is a new event each time
[28-Sep-2009 08:54:27] <LarsN> yes
[28-Sep-2009 08:54:48] <LarsN> our, older, and mostly broken Zenoss install, doesn't however do this
[28-Sep-2009 08:55:08] <ke4qqq> are any events incrementing??
[28-Sep-2009 08:55:27] <ke4qqq> even ones you aren't alerting on
[28-Sep-2009 08:56:08] <LarsN> ke4qqq: yes.
[28-Sep-2009 08:56:36] <ke4qqq> so what kind of events are alerting but clearing themselves?? ping? something else?
[28-Sep-2009 08:56:56] <LarsN> looks like primarily SNMP Agent Down / and Connection Refused on /status/http
[28-Sep-2009 08:57:03] <LarsN> are the two primary ones not incrimenting.
\n[28-Sep-2009 08:57:29] <LarsN> as I mentioned before, if I power\noff a device. the http monitor, and snmp monitor will still alert\nevery 5 minutes, and not incriment.
[28-Sep-2009 08:59:54] <rmatte> good morning golks
[28-Sep-2009 08:59:57] <rmatte> folks too
[28-Sep-2009 09:00:10] <rmatte> (off to a good start for typos today apparently)
[28-Sep-2009 09:00:37] <ke4qqq> weird - I can't imagine what would cause those events not to increment.
[28-Sep-2009 09:01:54] <Diddi> have you checked the logs?
[28-Sep-2009 09:02:15] <Diddi> if there's any problem with storing the events or something like that
[28-Sep-2009 09:02:27] * Diddi is just brainstorming
[28-Sep-2009 09:04:34] <rmatte> LarsN: alert every 5 minutes as in email alerts?
[28-Sep-2009 09:04:41] <LarsN> rmatte: yes.
[28-Sep-2009 09:04:51] <rmatte> LarsN: you have a repeat time set in your alert settings
[28-Sep-2009 09:04:53] <rmatte> go check
[28-Sep-2009 09:04:56] <LarsN> I'll dive into logs right after the meeting I'm in.
[28-Sep-2009 09:05:05] <rmatte> I'll bet that's the issue, if it's set to 300 change it to 0
[28-Sep-2009 09:06:46] <ke4qqq> rmatte: it's actually not incrementing events - each time it's a new instance of the event
[28-Sep-2009 09:07:40] <rmatte> yeh, but you mentioned that you powered the device off
[28-Sep-2009 09:07:53] <rmatte> so I assume zenoss has an "ip <whatever> is down" event for the device?
[28-Sep-2009 09:08:54] <rmatte> Ok, here's the scenario...
[28-Sep-2009 09:09:40]\n<rmatte> You power the device down, but since Zenoss is in the\nmiddle of a polling cycle it notices that the http monitor and snmp\nmonitor are down before it realizes that the device is unpingable, so\nit generates events for each
[28-Sep-2009 09:09:47] <rmatte> then it finally realizes the device is unreachable
[28-Sep-2009 09:10:06]\n<rmatte> it then does what it's designed to do and only checks\nthe device via ping from then on (doesn't attempt to monitor the other\nservices)
[28-Sep-2009 09:10:18] <rmatte> hence\nwhy the 2 alerts that were generated just before the ping down event\ndon't increment
[28-Sep-2009 09:10:33] <rmatte>\nbut you continue to be alerted every 5 minutes for those events because\nyou have a repeat time set in your alert settings
[28-Sep-2009 09:10:39] <rmatte> that's what's happening
[28-Sep-2009 09:15:52] <twm1010> Morning folks.
[28-Sep-2009 09:16:16] <rmatte> morning
[28-Sep-2009 09:20:15] <LarsN> rmatte: great plan. Except my Repeat time (Secs) is 0
[28-Sep-2009 09:21:56] <straterra> rmatte: what patch level are you running zenoss at?
[28-Sep-2009 09:22:26] <rmatte> you mean what version and what patches have I applied in addition?
[28-Sep-2009 09:22:31] <straterra> Yeah
\n[28-Sep-2009 09:22:42] <eidolon> hi folks - i have a disk monitor\non a windows box. in my OS pane, i see 'utilization' '100%' - but the\nstatus is green. what defines whether this is an event or not? sure\nas hell should be
\n[28-Sep-2009 09:22:53] <rmatte> I'm actually going to be writing\na blog post about the patch I have applied, I'm running 2.4.5 with 1\npatch applied
[28-Sep-2009 09:23:09] <straterra> Ok, I'll wait then
[28-Sep-2009 09:23:26] <eidolon> (there's a patch to 2.4.5?)
[28-Sep-2009 09:23:48] <straterra> I'm still running 2.4.2
\n[28-Sep-2009 09:23:50] <rmatte> eidolon: it was released on\nFriday afternoon to fix the bug where you can't properly set product\ninfo manually
[28-Sep-2009 09:24:19] <eidolon> ah. okay, not a problem for me.
[28-Sep-2009 09:24:23] <rmatte> I'm about to type up a quick guide on applying the patch and reindexing the product list
[28-Sep-2009 09:24:27] <straterra> Is updating pretty easy?
[28-Sep-2009 09:24:41] <rmatte> straterra: updating from 2.4.2 to 2.4.5 you mean?
[28-Sep-2009 09:24:44] <straterra> Yeah
[28-Sep-2009 09:24:50] <rmatte> straterra: you using stack installer?
[28-Sep-2009 09:24:59] <straterra> RPM installer for RHEL
[28-Sep-2009 09:25:12] <rmatte> ah, I'd imagine you'd just run an update with the RPM, that's about all there is to it
[28-Sep-2009 09:25:21] <rmatte> but I'd read the install guide just to be sure
\n[28-Sep-2009 09:25:50] <rmatte> with the stack installer it's\nbraindead simple, you just run the new stack installer and hit enter\ntwice, and off it goes
[28-Sep-2009 09:26:18] <eidolon> so. any ideas on my threshold problem?
[28-Sep-2009 09:27:12] <rmatte> eidolon: the filesystem template
[28-Sep-2009 09:27:20] <rmatte> eidolon: check if there is one locally applied to that drive
[28-Sep-2009 09:27:28] <eidolon> one moment
[28-Sep-2009 09:27:33] <rmatte> if not, go check the FileSystem template under /Server
[28-Sep-2009 09:27:39] <eidolon> wait wait. s ec.
[28-Sep-2009 09:27:48] <rmatte> make sure the threshold is enabled and set properly
[28-Sep-2009 09:27:55] <straterra> Release notes just say I can directly upgrade to it
[28-Sep-2009 09:28:05] <rmatte> it'll be "<something> * 0.90" by default
[28-Sep-2009 09:28:11] <rmatte> representing the percentage
[28-Sep-2009 09:28:23] <rmatte> straterra: then give'r
[28-Sep-2009 09:28:40] <eidolon> so under:
[28-Sep-2009 09:29:05] <eidolon> /Devices /Server /Windows /WMI /bos-cmds01.REDACTED.net /os /C
[28-Sep-2009 09:29:07] <eidolon> i see:
\n[28-Sep-2009 09:29:20] <eidolon>\nFileSystem \n/Devices/Server/Windows/WMI FileSystem\ntemplate that uses Windows WMI.
[28-Sep-2009 09:29:29] <straterra> rmatte: should I upgrade my core zenpacks too?
[28-Sep-2009 09:29:41] <eidolon> so the template in use is in /devices/server/windows/wmi, i should look there?
[28-Sep-2009 09:29:59] <eidolon> i see Filesystem there
[28-Sep-2009 09:30:25] <eidolon> ahh. 'low disk space' threshold
[28-Sep-2009 09:30:32] <eidolon> enabled: true
\n[28-Sep-2009 09:30:45] <Egyptian[Web]> Diddi: i fixed my script\nand it now sends sms to me .. read var instead of $* as parameters
[28-Sep-2009 09:30:51] <Egyptian[Web]> Diddi: thanks
\n[28-Sep-2009 09:33:30] <rmatte> straterra :\nblogs/ryanmatte/2009/09/28/patching-hardwaresoftware-info-bug-trac-ticket-5473
[28-Sep-2009 09:34:16] <straterra> Gotcha
[28-Sep-2009 09:36:02] <straterra> Grrr
\n[28-Sep-2009 09:36:14] <straterra> I wish Zenoss would say you\nneed zenoss to be RUNNING before you try to install the core zenpack
[28-Sep-2009 09:37:49] <rmatte> hehe
[28-Sep-2009 09:37:58] <rmatte> you need zenoss running before trying to install any ZenPack
[28-Sep-2009 09:38:09] <rmatte> otherwise it has no way of importing the objects in to Zope
[28-Sep-2009 09:42:25] <twm1010> rmatte: I'm thinking of switching all my Windows boxes to WMI perf collection
[28-Sep-2009 09:42:44] <rmatte> fun fun
[28-Sep-2009 09:43:05] <cgibbons> yay! i get a kickback anytime anyone uses WMI! woohoo
[28-Sep-2009 09:43:09] <cgibbons> er, did I just say that outloud again?!
[28-Sep-2009 09:43:35] <rmatte> haha
[28-Sep-2009 09:43:50] <rmatte> WMI is unreliable and cumbersome
[28-Sep-2009 09:44:21] <jb> but necessary..
[28-Sep-2009 09:44:42] <rmatte> yeh, but it'd be nice if they'd improve it
[28-Sep-2009 09:45:04] <rmatte> I love how WMI has a tendency to just magically break
[28-Sep-2009 09:45:09] <rmatte> never have that problem with SNMP
[28-Sep-2009 09:45:48] <cgibbons> they do seem to like creating server-side memory leaks with it for each new OS release
[28-Sep-2009 09:46:19] <rmatte> yup
[28-Sep-2009 10:00:25] <straterra> rmatte: thanks for the link..I'm all patched up
[28-Sep-2009 10:07:36] <rmatte> hmmm, this zapplet thing is pretty cool
[28-Sep-2009 10:07:39] <rmatte> straterra: no problem
[28-Sep-2009 10:08:21] <daMaestro> does anyone know how to fire off the events cleanup call?
[28-Sep-2009 10:08:45]\n<rmatte> daMaestro: like the event cleanup that fires off after\nwhatever amount of time that you set (an hour or whatever)?
[28-Sep-2009 10:08:51] <daMaestro> yeah
[28-Sep-2009 10:08:56] <rmatte> hmmm
[28-Sep-2009 10:09:12] <daMaestro> the process that deletes events older then N number of days
[28-Sep-2009 10:09:20] <rmatte> oh, that one
[28-Sep-2009 10:09:36] <daMaestro> basically, that launches around 7 pm for us.. and it almost takes down zenoss
[28-Sep-2009 10:09:37] <rmatte> not really sure, I only know how to delete all events for a particular device
[28-Sep-2009 10:09:56] <rmatte> ummmm, how many events do you guys get per day?
[28-Sep-2009 10:09:57] <daMaestro> so i need to launch it with a DBA on hand to see what we can do to make it run better
[28-Sep-2009 10:10:03] <rmatte> it shouldn't even come close to taking down Zenoss?
[28-Sep-2009 10:10:13] <daMaestro> the amount of disk i/o is absurd
[28-Sep-2009 10:10:23] <daMaestro> for about an hour it's taxed
\n[28-Sep-2009 10:11:47] <daMaestro> "Delete Historical Events\nOlder Than (days)" is the process that i'm attempting to manually run
[28-Sep-2009 10:15:03] <rmatte> yeh, I'm not sure what function that executes
[28-Sep-2009 10:15:06] <rmatte> you'd have to ask a dev
\n[28-Sep-2009 10:15:31] <rmatte> I could go digging through code\nand eventually find it, but I don't have time this morning
[28-Sep-2009 10:15:42] <daMaestro> yeah, it's just a python script that runs
\n[28-Sep-2009 10:15:56] <daMaestro> i looked into it last time it\nhappened and found the command but now i can't find it again
[28-Sep-2009 10:16:36] <daMaestro> hmm looks like it's ZenDeleteHistory.py
[28-Sep-2009 10:17:52] <rmatte> npmccallum: ping
[28-Sep-2009 10:18:07] <npmccallum> rmatte: pong
\n[28-Sep-2009 10:18:27] <rmatte> hey, I'm trying out zapplet and\nI'm loving it but it seems to have a small bug, I'm going over the code\nbut not figuring this out
[28-Sep-2009 10:19:13]\n<rmatte> when there are no events in the event log on a server a\ncaution icon is displayed and if you rollover the icon the text\ndisplays an error
[28-Sep-2009 10:19:29] <rmatte> complaining about line 56 and line 110 in events.py
[28-Sep-2009 10:19:43] <rmatte> erm, event.py
[28-Sep-2009 10:19:44] <rmatte> rather
[28-Sep-2009 10:20:12] <rmatte> 110 is: highest = filter(lambda x: x[-1] > 0, results)[-1]
[28-Sep-2009 10:20:19] <rmatte> any idea what might need to be changed?
[28-Sep-2009 10:20:39] <npmccallum> rmatte: give me a second to look at the code
[28-Sep-2009 10:20:46] <rmatte> thanks
[28-Sep-2009 10:21:23] <rmatte> let me see if I can type this error off for you (difficult since it shows up in a popup)
[28-Sep-2009 10:22:33] <npmccallum> rmatte: ok
[28-Sep-2009 10:22:37] <npmccallum> rmatte: here's the problem
[28-Sep-2009 10:22:51] <npmccallum> I've never actually coded for the scenario where there are no events
[28-Sep-2009 10:22:56] <Brad_K> rmatte: uh, so WMI is b0rken? should we be using perfmon instead?
[28-Sep-2009 10:23:05] <npmccallum> I've assumed that there will *always* be at least one event
[28-Sep-2009 10:23:11] <npmccallum> bad assumption, I know
[28-Sep-2009 10:23:12] <rmatte> Brad_K: it's not really "broken" I just find it to be tempermental
[28-Sep-2009 10:23:37] <Brad_K> rmatte: which is not what we need in a production system/network monitoring system....
\n[28-Sep-2009 10:23:40] <rmatte> oh, Brad_K:\nblogs/ryanmatte/2009/09/28/patching-hardwaresoftware-info-bug-trac-ticket-5473
[28-Sep-2009 10:23:51] <rmatte> npmccallum: ah
[28-Sep-2009 10:24:02] <rmatte> npmccallum: yeh, it should show as green or something if there are zero events
[28-Sep-2009 10:24:14] <npmccallum> rmatte: right
[28-Sep-2009 10:24:15] <npmccallum>
[28-Sep-2009 10:24:27] <Brad_K> rmatte: i greatly appreciate the patch. thanks!
[28-Sep-2009 10:24:32] <npmccallum> you obviously run the cleanest network I've ever seen
[28-Sep-2009 10:25:00]\n<rmatte> npmccallum: well, some of the zenoss boxes that I\nmonitor are more lab boxes than anything, so the event logs are often\nempty
[28-Sep-2009 10:25:28] <rmatte> Brad_K:\nthanks mrchippy if you see him around, he wrote the patch, I just\nbugged the crap out of them until they patched it and wrote the blog\nentry
[28-Sep-2009 10:25:35] <rmatte> thank* rather
[28-Sep-2009 10:26:02] <npmccallum> rmatte: you can probably just do a "if len(results) == 0" and create a suitable summary event
[28-Sep-2009 10:26:44] <npmccallum> rmatte: this function gathers returns a tuple with (Event, Event[])
[28-Sep-2009 10:26:47] <rmatte> yeh, and how do I set an alternate icon?
[28-Sep-2009 10:26:56] <rmatte> lol
\n[28-Sep-2009 10:27:20] <npmccallum> the first Event in that tuple\nis a "virtual" event that is used to generate the icon and summary of\nthe applet
[28-Sep-2009 10:27:23] <rmatte> do you plan on possibly patching it up at some point and releasing a new version?
\n[28-Sep-2009 10:27:47] <npmccallum> rmatte: I have a huge rewrite\nsitting in a backup somewhere that I just never got back to\n(unfortunately)
[28-Sep-2009 10:27:59] <npmccallum> rmatte: the rewrite will allow *all* functionality as a plugin
\n[28-Sep-2009 10:28:28] <rmatte> ah, that's pretty cool, but\ncouldn't you just throw a few lines in and do a temporary release?
[28-Sep-2009 10:28:34] <rmatte> you're the python developer, not me lol
[28-Sep-2009 10:28:42] <rmatte> my python is ok but it's very much geared towards Zenoss
[28-Sep-2009 10:28:54] <npmccallum> yes, I can
[28-Sep-2009 10:28:58] <npmccallum> I'll try to get to it today
[28-Sep-2009 10:29:05] <rmatte> thanks
[28-Sep-2009 10:29:35] <rmatte> now I just need to figure out how to use the historify xml-rpc in perl
\n[28-Sep-2009 10:29:55] <rmatte> since our ticket create script is\nperl, need the script to acknowledge events after opening tickets
\n[28-Sep-2009 10:30:23] <rmatte> might just do it as a seperate\npython script and have the perl script call on it, if worst comes to\nworst
[28-Sep-2009 10:30:46] <rmatte> g'day Matt
[28-Sep-2009 10:31:02] <mrayzenoss> morning all
[28-Sep-2009 10:31:12] <ckrough> morning
\n[28-Sep-2009 10:31:32] <rmatte> mrchippy made my day on Friday:\nblogs/ryanmatte/2009/09/28/patching-hardwaresoftware-info-bug-trac-ticket-5473
[28-Sep-2009 10:32:14] <rmatte> It'd almost be worth applying that patch to 2.4.5 and releasing it as 2.4.6
[28-Sep-2009 10:36:06]\n<exarkun> Looking around for .debs, I found a forum post that\nsuggests this source line: "deb http://dev.zenoss.org/deb main stable"
\n[28-Sep-2009 10:36:36] <exarkun> How official is that? Does\n"dev" in the domain indicate packages of potentially unstable versions?
[28-Sep-2009 10:36:42] <exarkun> (will it even work? I haven't tried yet)
[28-Sep-2009 10:36:53] <mrayzenoss> exarkun: that's in the Installation Guide
[28-Sep-2009 10:37:28] <mrayzenoss> exarkun: so fairly official
[28-Sep-2009 10:37:28] <exarkun> Okay, I'll take a look there. Thanks.
[28-Sep-2009 10:37:39] <exarkun> Oh, I see. Cool.
\n[28-Sep-2009 10:42:26] <daMaestro> rmatte, delete h,j,d from\nhistory h LEFT JOIN log j ON h.evid = j.evid LEFT JOIN detail d ON\nh.evid = d.evid WHERE StateChange < DATE_SUB(NOW(), INTERVAL 30 day)
[28-Sep-2009 10:42:39] <daMaestro> rmatte, and there are not proper indexes to make that run even slightly quickly
[28-Sep-2009 10:43:24] <rmatte> daMaestro: ah
[28-Sep-2009 10:43:54] <daMaestro> rmatte, i'm having our DBA add some indexes that should help... hopefully
[28-Sep-2009 10:43:55] <rmatte> daMaestro: I guess it really depends on the volume of events
[28-Sep-2009 10:44:12] <daMaestro> rmatte, something like 4 mil right now
[28-Sep-2009 10:44:23] <daMaestro> rmatte, and that is with only keeping 30 days
[28-Sep-2009 10:53:13] <rmatte> crazy
\n[28-Sep-2009 11:10:25] <gwb2351> anybody run into errors\n(timeout, invoalid responses) with apache as the front-end (proxy) to\nzope?
[28-Sep-2009 11:10:40] <ckrough> not normally
[28-Sep-2009 11:11:02] <gwb2351> the ssl_error log from httpd has these bits in it
\n[28-Sep-2009 11:11:25] <Brad_K> and the performance graphs aren't\ncoming up, either. and we're also having problems with some of the\nstatus pages, although not necessarily all of them.
\n[28-Sep-2009 11:11:28] <gwb2351> proxy: Error reading from\nremote server returned by /zport/dmd/getUserMessages, referer:\nhttps://eye.its.utexas.edu/zport/dmd/Devices/Server/Passive/devices/mail.utexas.edu/viewDevicePerformance
\n[28-Sep-2009 11:11:42] <gwb2351> (70007)The timeout specified has\nexpired: proxy: error reading status line from remote server 127.0.0.1,\nreferer:\nhttps://eye.its.utexas.edu/zport/dmd/Devices/Server/Passive/devices/mail.utexas.edu/viewDevicePerformance
[28-Sep-2009 11:12:50] <ckrough> this just started happenning?
[28-Sep-2009 11:13:04] <Brad_K> we've seen this before.
[28-Sep-2009 11:13:15] <Brad_K> but it has been intermittent.
\n[28-Sep-2009 11:13:56] <Brad_K> it did just occur to me that\nmaybe we're having the same database problems as daMaestro reported.
[28-Sep-2009 11:14:26] <ckrough> server load is normal?
[28-Sep-2009 11:14:43] <Brad_K> 0.08.
[28-Sep-2009 11:14:52] <ckrough> iowait
[28-Sep-2009 11:15:16] <Brad_K> not seeing any. 99.5% idle.
[28-Sep-2009 11:15:20] <ckrough> obviously only counts if the problem is currently happening
[28-Sep-2009 11:15:31] <Brad_K> currently ongoing, yes.
\n[28-Sep-2009 11:15:40] <ckrough> just wondering if zope/python is\ntied up with something and isnt returning pages to apache in a timely\nfashion
[28-Sep-2009 11:15:42] <gwb2351> problem is currently happening, iostat 2 10 shows 0.00 %iowait
[28-Sep-2009 11:15:49] <ckrough> like zenhub rebuilding configurations, etc...
[28-Sep-2009 11:16:02] <ckrough> especially since its intermittant
[28-Sep-2009 11:32:09] <gwb2351> what process in zenoss handles apache requests? the errors returned to the browser are
[28-Sep-2009 11:32:19] <gwb2351> The proxy server received an invalid response from an upstream server.
\n[28-Sep-2009 11:32:19] <gwb2351> The proxy server could not\nhandle the request GETÂ /zport/acl_users/cookieAuthHelper/login_form.
[28-Sep-2009 11:32:29] <gwb2351> (or other similiar GET /zport errors)
[28-Sep-2009 11:38:25] <ckrough> zope
[28-Sep-2009 11:41:51] <gwb2351> if I look at what zenhub is doing, I see lots of timeouts on a select
[28-Sep-2009 11:42:09] <gwb2351> [root@eye localhost]# strace -s 1024 -t -v -p 9355
[28-Sep-2009 11:42:09] <gwb2351> Process 9355 attached - interrupt to quit
[28-Sep-2009 11:42:09] <gwb2351> 11:41:23 select(112, [111], [], [111], {0, 828000}) = 0 (Timeout)
[28-Sep-2009 11:42:09] <gwb2351> 11:41:24 select(112, [111], [], [111], {1, 24000}) = 0 (Timeout)
[28-Sep-2009 11:42:38] <gwb2351> where process 9355 is the zenhub.py process
[28-Sep-2009 11:44:43] <ckrough> Do you guys run multiple zenhub workers?
[28-Sep-2009 11:44:52] <gwb2351> yes - 4 additional
\n[28-Sep-2009 11:47:24] <Brad_K> and i'm still trying to come up\nto speed on zenoss so that i can help Greg, allowing him to have more\ntime to do bigger stuff.
[28-Sep-2009 11:47:58] <ckrough> fwiw, I am getting select timehouts on zenhub on a normally functioning install
[28-Sep-2009 11:48:03] <ckrough> *timeouts*
[28-Sep-2009 11:48:19] <gwb2351> good to know, thanks
[28-Sep-2009 11:50:05] <gwb2351> i think i'll restart zope and see what happens
\n[28-Sep-2009 11:50:56] <Brad_K> and so far as i know, we have not\nyet attempted to apply the performance tuning tips mentioned in\ndocs/DOC-2521
[28-Sep-2009 11:56:34] <ckrough> are you using mod_proxy?
[28-Sep-2009 11:56:47] <ckrough> if so, whats your ProxyTimeout value?
[28-Sep-2009 11:57:13] <Brad_K> i think we are using mod_proxy, let me check...
[28-Sep-2009 11:58:52] <Brad_K> i'm not finding that string anywhere in /etc/httpd/*
[28-Sep-2009 12:01:07] <ckrough> here's mine http://pastebin.com/m4df7ad2a
[28-Sep-2009 12:02:05] <ckrough> http://httpd.apache.org/docs/2.0/mod/mod_proxy.html#proxytimeout
[28-Sep-2009 12:02:51] <ckrough> anyone on running distributed collectors?
[28-Sep-2009 12:03:15] <Brad_K> we are running distributed collectors, too.
[28-Sep-2009 12:03:54]\n<ckrough> Im experiencing a problem where zenping on a collector\ndetects a device as down, but the zenhub box never creates an event for\nit
[28-Sep-2009 12:04:10] <ckrough> Im not sure what the 'workflow' is for zenping from the collector
[28-Sep-2009 12:05:42]\n<Brad_K> ckrough: okay, so I found all instances of the\ncase-insensitive string underneath /etc/httpd/. See\nhttp://pastebin.com/m7518e26e
[28-Sep-2009 12:06:29]\n<Brad_K> we don't actually use zenping. we use the check_ping\nNagios plug-in that we hacked to get to work properly under Zenoss.
[28-Sep-2009 12:06:36]\n<twm1010> I found Zenoss performance to go up the most when I\nstarted figuring out what was generating debug "events"
[28-Sep-2009 12:06:47] <twm1010> bad OID messages, etc.
[28-Sep-2009 12:07:06]\n<ckrough> twm1010: thats good thinking, most of that stuff is\ninvisible if your arent searching for it. I'll have to do an audit on\nthat.
[28-Sep-2009 12:08:07] <ckrough> Brad_K: what are you using to forward web :80 or :443 to zope on your install? I dont see it here
[28-Sep-2009 12:08:09]\n<Brad_K> we also don't use the native Zenoss module to check ssh\nconnections, since it doesn't properly close the TCP connection. The\nNagios check_ssh module does, after we hacked it to work with zenoss.
[28-Sep-2009 12:08:28] <ckrough> Brad_K: yeah, my logs are filled with sshd complaints.
[28-Sep-2009 12:09:08] <Brad_K> ckrough: we're actually using the RewriteEngine and a RewriteRule.
[28-Sep-2009 12:09:46] <Brad_K> this is in our /etc/httpd/conf.d/ssl.conf:
[28-Sep-2009 12:09:47] <Brad_K> RewriteRule ^/(.*) http://127.0.0.1:8080/VirtualHostBase/https/%{SERVER_NAME}:44
[28-Sep-2009 12:09:47] <Brad_K> 3/$1 [P,L]
[28-Sep-2009 12:10:15] <Brad_K> sorry, that should be one long line, not two lines. didn't paste right.
[28-Sep-2009 12:17:00]\n<gwb2351> I'll peruse this information and see what happens\nblogs/zenossblog/2007/06/05/zenoss-and-apache-documentation
[28-Sep-2009 12:21:03] <jelmer> cgibbons: hi - did you get my reply wrt DCOM/WMI?
[28-Sep-2009 12:21:26] <cgibbons> jelmer: yup haven't really processed it yet but it is in my inbox
\n[28-Sep-2009 12:39:43] <Diddi> when creating a new threshold, is\nit possible to get values in certain units? I'm currently monitoring\nremaining percentage of my ups-battery, but the event will only say\n"<threshold> not met: current value #number#"
[28-Sep-2009 12:39:53] <Diddi> I'd like to add '%' at the end of that message (:
[28-Sep-2009 13:09:58]\n<Brad_K> okay, we got all our rewrite stuff resolved. what we\nhad was an incomplete cut-n-paste from the instructions that we were\nfollowing.
[28-Sep-2009 13:10:35] <purevol> ahoy zenossers
\n[28-Sep-2009 13:14:50] <daMaestro> wonderful.. so now that we\nadded the indexes it's less disruptive but now we get\n"_mysql_exceptions.OperationalError: (1206, 'The total number of locks\nexceeds the lock table size')"
[28-Sep-2009 13:14:54] <daMaestro> pos
[28-Sep-2009 13:15:56] <rmatte> daMaestro, file a bug report: dev.zenoss.com/trac login: zenoss/zenoss
[28-Sep-2009 13:18:36] <purevol> rmatte ...you made the hus Cisco MIB Zenpack?
[28-Sep-2009 13:18:42] <purevol> huge, not hus
\n[28-Sep-2009 13:20:59] <feutete> Hello, I have some hosts who\nhave never had a service running on TCP 8080, but Zenoss believes they\nshould, so it always reports http-alt down
\n[28-Sep-2009 13:21:22] <feutete> strangely, if I delete the\nIPService under the OS tab for the devices, it comes back on its own
[28-Sep-2009 13:21:35] <feutete> if I set Monitor=False, Zenoss still alerts
[28-Sep-2009 13:21:58] <feutete> any ideas on how to convince Zenoss that the devices shouldn't have anything on 8080?
[28-Sep-2009 13:22:13] <Brad_K> did you remodel?
[28-Sep-2009 13:22:19] <feutete> yes
\n[28-Sep-2009 13:22:47] <Brad_K> well, that's the limit of my\nknowlesge of zenoss -- always remodel after making any change. ;-)
[28-Sep-2009 13:23:19] <feutete> yeah, I have made a habit of doing that as well
\n[28-Sep-2009 13:23:58] <feutete> it would be nice if IPServices\nfollowed the template hierarchy...I have just a few devices that DO\nlisten on 8080
[28-Sep-2009 13:24:24] <feutete> but apparently zenoss checks ALL devices for 8080
[28-Sep-2009 13:25:12] <feutete> neet...I just remodeled the device again, and http-alt shows as Monitor=true
[28-Sep-2009 13:25:22] <feutete> even though I explicitly set it to false a few days ago
[28-Sep-2009 13:26:44] <rmatte> feutete: after you deleted the ipservice did you push the changes to the collector?
[28-Sep-2009 13:26:58] <rmatte> Manage -> Push Changes from the device page
[28-Sep-2009 13:27:10] <feutete> not explicitly, but the collector in this case is localhost
[28-Sep-2009 13:27:31] <feutete> shoudl I have to if the collector is localhost?
\n[28-Sep-2009 13:27:39] <rmatte> Also, Zenoss does a portscan, so\nif it thinks that something is listening on port 8080 then there almost\ncertainly is
[28-Sep-2009 13:27:54] <feutete> nmap shows 8080 is filtered
[28-Sep-2009 13:27:59] <feutete> can't connect to 8080 manually
\n[28-Sep-2009 13:28:19] <rmatte> feutete: you always have to\nexplicitly push changes even if the collector is localhost. The\ncollector only pulls config data for device from Zenoss once an hour
[28-Sep-2009 13:28:28] <rmatte> so you need to push changes to point out to it that something has changed
[28-Sep-2009 13:28:49] <rmatte> filtered doesn't necessarily mean closed
[28-Sep-2009 13:28:55] <feutete> ah, ok. I'll give that a spin
[28-Sep-2009 13:29:03] <rmatte> k
[28-Sep-2009 13:29:09] <feutete> thanks rmatte
[28-Sep-2009 13:29:11] <rmatte> np
[28-Sep-2009 13:29:39] <feutete> i assume the same is true if I set monitor=false on the IPService?
[28-Sep-2009 13:29:46] <rmatte> correct
[28-Sep-2009 13:29:47] <feutete> i need to push to the collector then as well?
[28-Sep-2009 13:29:48] <feutete> ok
\n[28-Sep-2009 13:30:14] <rmatte> the only thing you don't have to\npush changes on is if you were to change the production state of a\ndevice
[28-Sep-2009 13:30:36] <rmatte> basically\nall other changes need to be pushed, or you can optionally wait an hour\nfor the collector to automatically gather the monitoring config data
\n[28-Sep-2009 13:31:13] <feutete> seems like a good candidate for\na feature request...automatically push changes to the collector...
[28-Sep-2009 13:31:29] <__jd__> Hey all.. Question..\nI'm experimenting with the XMLRPC/REST APIs for injecting events into\nZenOSS. The trouble I'm having is that the clear events (severity 0)\ndon't seem to be clearing the event from the console. Is there\nanything other than setting severity to 0 than I should be doing?
[28-Sep-2009 13:31:50] <__jd__> And good to know about pushing the configs, didn't know that was necessary
\n[28-Sep-2009 13:34:25] <rmatte> feutete: it's meant to keep the\nload on the collector down, but you could request it anyways
[28-Sep-2009 13:35:02] <ckrough> ping jmp
[28-Sep-2009 13:35:50] <purevol> anyone ever get a cisco 2800 integrated services router working?
[28-Sep-2009 13:36:35] <purevol> __JD__ are you selecting the Event and moving it to history?
[28-Sep-2009 13:37:28] <rmatte> purevol: define "working"
\n[28-Sep-2009 13:38:25] <__jd__> purevol, I'm sending an XMLRPC\nevent to inject an error-level event into ZenOSS. Then, keeping all\nother info the same, I'm sending the same thing with severity 0\n(clear), in hopes that it will move the event to history
\n[28-Sep-2009 13:39:15] <feutete> rmatte: I just had another\nhttp-alt event show up, and http-alt isn't even listed under\nOS->IPServices. Any guess how I can get it to show up so I can set\nMonitor=false?
[28-Sep-2009 13:39:26] <purevol>\nyeah good point, i seem to be collecting call data, but i cannot\ncollect any data about my interfaces, environmentals, etc
\n[28-Sep-2009 13:40:31] <purevol> __jd__ : oh....im rusty with\nzenoss, havent used it in a bit, i dont recalls how to correlate and\nclear events like that
[28-Sep-2009 13:40:43] <purevol> i would imagine it needs to be deleted from the database
[28-Sep-2009 13:41:06] <rmatte> feutete: ssh in to the server, become the zenoss user, type zendmd to open zendmd
[28-Sep-2009 13:41:20] <rmatte> feutete: then type reindex() hit enter and then type commit() and hit enter
[28-Sep-2009 13:41:27] <rmatte> feutete: then type ctrl-d to exit
[28-Sep-2009 13:41:37] <rmatte> feutete: actually wait...
[28-Sep-2009 13:42:00]\n<rmatte> feutete: before you do that, stop zenoss for a minute,\nand copy $ZENHOME/var/Data.fs to $ZENHOME/var/Data.fs.bak
[28-Sep-2009 13:42:14] <feutete> ok
\n[28-Sep-2009 13:42:14] <rmatte> just in case the reindex screws\nanything up (it's really rare but I've seen it happen a couple of times)
[28-Sep-2009 13:42:20] <rmatte> then start zenoss back up
[28-Sep-2009 13:42:30] <rmatte> open zendmd, and type those commands it
[28-Sep-2009 13:42:32] <rmatte> in*
\n[28-Sep-2009 13:44:52] <rmatte> hmmm, zenprocess cycle time on\none of my boxes has been exceeded 9 times in a row now, odd
\n[28-Sep-2009 13:49:10] <purevol> so i have this 2851 that i can\nget call daya from, but i cant get any of the environmentals
[28-Sep-2009 13:49:37] <purevol> i installed your zenpack with all the cisco Mibs, but there is no 2851 mib/oid in there
[28-Sep-2009 13:51:03] <purevol> im not reawlly sure what the problem is
\n[28-Sep-2009 13:51:43] <purevol> i just built the appliance quick\nlike, so i cant do much witht the platform, maybe its that weird ssmi2\ndependency or something
[28-Sep-2009 13:52:13] <rmatte> Mibs are only for translating traps in Zenoss
[28-Sep-2009 13:52:23] <rmatte> They aren't used for polling data
[28-Sep-2009 13:52:41] <rmatte> so what specific data are you unable to poll?
[28-Sep-2009 13:53:00] <purevol> none of the environmentals
[28-Sep-2009 13:53:14] <rmatte> environmentals as in?
[28-Sep-2009 13:53:16] <purevol> cpu, interfaces, etc
[28-Sep-2009 13:53:19] <rmatte> ah
[28-Sep-2009 13:53:22] <purevol> routes
[28-Sep-2009 13:53:25] <rmatte> so you can't model the device at all essentially
[28-Sep-2009 13:53:31] <purevol> basically
[28-Sep-2009 13:53:37] <purevol> but i am receiving traps from it
[28-Sep-2009 13:53:41] <rmatte> well, as a starting point, if you snmpwalk the device do you see any of that data?
[28-Sep-2009 13:53:47] <rmatte> this is probably just an SNMP permissions issue
[28-Sep-2009 13:53:52] <purevol> i wonder if there is some hokey shit on cisco, i know the asa for example
[28-Sep-2009 13:55:25] <rmatte> asa is easy to fix in Zenoss
\n[28-Sep-2009 13:55:52] <rmatte> go to the zProperties and change\nzMaxOIDPerRequest from 40 to 20 for the specific ASA device
[28-Sep-2009 13:56:08] <purevol> no i get that weird smi error - http://pastebin.com/m537c0fc0
\n[28-Sep-2009 13:56:21] <gwb235> (potentail) resolution to our\nearlier problem (proxy timeouts) was to re-read the "using HTTPS/SSL\nfor the Zenoss Web UI" knowledge base and add some missing VirtualHost\nbits. We'll see if it comes back.
[28-Sep-2009 13:56:47] <purevol> i try v1 and i get , "End of MIB"
[28-Sep-2009 13:57:15] <purevol> rmatte: i already decreased that down to 10 actually
[28-Sep-2009 13:57:24] <purevol> my asa is working great
[28-Sep-2009 13:57:32] <purevol> its the 2851 i can get any data out of
[28-Sep-2009 13:57:52] <purevol> well, i see traps in event log, but, alas, cant poll jack
[28-Sep-2009 13:59:16] <rmatte> purevol: if you're just getting End of MIB then the SNMP isn't properly configured
\n[28-Sep-2009 13:59:28] <rmatte> you're either using the wrong\ncommunity string or the config on the device isn't setup properly
[28-Sep-2009 13:59:31] <rmatte> it's not a Zenoss issue
[28-Sep-2009 14:00:12]\n<rmatte> traps and polling are two separate things, it looks like\nyou don't have any access at all to the SNMP tree with the community\nstring that you're trying
[28-Sep-2009 14:04:16] <purevol> hrm....
[28-Sep-2009 14:07:13] <purevol> notice anything out of the ordinary? - http://pastebin.com/m2bcc5bcd
[28-Sep-2009 14:07:51] <rmatte> I see lots of trap lines but not much else, let me get you an example of what you should have
[28-Sep-2009 14:08:07] <purevol> there is not option for polling
[28-Sep-2009 14:08:19] <purevol> at least not that i can find
[28-Sep-2009 14:09:08]\n<rmatte> there's a config option that needs to be put in to allow\nSNMP access, you need to define community strings and the level of\naccess for each
[28-Sep-2009 14:09:15] <rmatte> all you've done is defined traps
[28-Sep-2009 14:09:43] <rmatte> I'm just trying to dig up an example for you... give me a second
[28-Sep-2009 14:09:52] <purevol> i think i found it
[28-Sep-2009 14:09:57] <purevol> snmp-server inform
[28-Sep-2009 14:10:34] <purevol> maybe i dunno
[28-Sep-2009 14:10:46] <purevol> yup that was it
[28-Sep-2009 14:10:49] <rmatte> ok first off you need a line like...
[28-Sep-2009 14:11:10] <purevol> my bad, gimpy figured it out
[28-Sep-2009 14:11:20] <rmatte> k
[28-Sep-2009 14:11:23] <purevol> i added the snmp communi, and snmp inform, and it worked
[28-Sep-2009 14:11:32] <rmatte> good
[28-Sep-2009 14:11:33] <rmatte> hehe
[28-Sep-2009 14:11:53] <purevol> will probably delete and re-add the device
[28-Sep-2009 14:12:04] <rmatte> you don't have to
[28-Sep-2009 14:12:09] <rmatte> just remodel, does the exact same thing
[28-Sep-2009 14:12:38] <rmatte> and rename it: Manage -> Rename Device from the device page
[28-Sep-2009 14:13:21] <purevol> well, if i delete and readd, all the device data will change
[28-Sep-2009 14:13:30] <rmatte> what do you mean?
[28-Sep-2009 14:13:34] <purevol> err performance data, evetns, that shit will get deleted
[28-Sep-2009 14:13:42] <rmatte> oh, yeh, that's correct
[28-Sep-2009 14:13:49] <rmatte> if that's what you want, then yeh
[28-Sep-2009 14:17:41] <purevol> we will see if i can get some perf data out of this cisco junk
\n[28-Sep-2009 14:18:14] <purevol> i dove into zenoss pretty hard a\ncouple years back, then i set it up last winter and handed it off to\nsomeone else
[28-Sep-2009 14:18:36] <purevol>\nnow im at this new gig, and they have cisco wroks, and i cant do\nanything with it, so ... here i am back to zenoss
[28-Sep-2009 14:18:44] <purevol> hoping the core works out alright
[28-Sep-2009 14:19:34] <rmatte> It should work fine
[28-Sep-2009 14:19:42] <rmatte> we're monitoring hundreds of cisco devices with it
[28-Sep-2009 14:20:10] <purevol> anything special, i need to setup to poll on a specificed basis?
[28-Sep-2009 14:20:40] <purevol> no graphs are popping up yet
[28-Sep-2009 14:21:21] <purevol> well some are working, but others arent, we will give it some time
[28-Sep-2009 14:24:16] <rmatte> it'll take 3 polling cycles for graphs to generate (about 15 to 20 mins)
[28-Sep-2009 14:24:40] <rmatte> the standard cisco template should be sufficient for starters
[28-Sep-2009 14:24:47] <rmatte> but you may want to customize it later on
\n[28-Sep-2009 14:25:44] <rmatte> hmmm weird, we have a director\nwhich is supposedly at 7.6G usage, but when I do a du -sh on it it\nshows up as 69 megs
[28-Sep-2009 14:25:51] <rmatte> directory*
[28-Sep-2009 14:26:10] <purevol> right on, hopefully, i can create some nifty stuff, and hook up a zenpack
[28-Sep-2009 14:26:17] <purevol> also got all this wireless tuff also
[28-Sep-2009 14:27:01]\n<rmatte> the cisco wireless stuff is quite fun, if it's the whole\nwireless controller and accesspoint setup where the controller\ngenerates the traps for every AP
[28-Sep-2009 14:27:33] <purevol> yes i have the WCS, two WLC, and a good amount of APs
[28-Sep-2009 14:27:37] <rmatte> since you have 1 device generating traps for potentially hundreds
[28-Sep-2009 14:27:59] <purevol> ill need to set it up
[28-Sep-2009 14:28:05] <rmatte> so you'll need to write a transform that displays the name of the IP which it's sending the trap for
[28-Sep-2009 14:28:14] <rmatte> I still need to write one for one of our Zenoss boxes
[28-Sep-2009 14:28:16] <purevol> its kinda dumb because the WCS is supposed to be a NMS
[28-Sep-2009 14:28:17] <rmatte> haven't gotten around to it yet
[28-Sep-2009 14:28:23] <purevol> and it has some cool stuff, but its weak
[28-Sep-2009 14:28:39] <rmatte> yeh, WCS is alright, but it's not the best
[28-Sep-2009 14:29:01] <purevol> i just updated it to 6.0, its pretty, and it has some cool stuff, but overall its weak
[28-Sep-2009 14:29:14] <purevol> because, none of it itnegrates into cisco wroks!
[28-Sep-2009 14:29:22] <purevol> and i cant type
[28-Sep-2009 14:29:29] <rmatte> lol
[28-Sep-2009 14:30:02]\n<rmatte> cisco works is great for writing and backing up device\nconfigs, but beyond that I don't really see it as a management tool
[28-Sep-2009 14:30:46] <rmatte> well, not a monitoring tool anyways
[28-Sep-2009 14:30:51] <rmatte> management, yes
[28-Sep-2009 14:31:14] <rmatte> we use an old piece of software called ZipTie to do that stuff
[28-Sep-2009 14:31:21] <rmatte> it's free and it works
[28-Sep-2009 14:32:43] <purevol> yeah you cant do any sort of monitoring with cisco works
[28-Sep-2009 14:32:52] <purevol> they have something , whats up gold
[28-Sep-2009 14:33:07] <rmatte> yeh, I've never used that but I've heard about it
[28-Sep-2009 14:33:29] <purevol> my "manager" is a micro management
[28-Sep-2009 14:33:47] <rmatte> yeh, I'm not a big fan of micro management
[28-Sep-2009 14:33:51] <rmatte> it's counter-productive
[28-Sep-2009 14:33:52] <purevol> nothing gets done
[28-Sep-2009 14:34:07] <purevol> but its cool, ill ust sit here and do nothing
[28-Sep-2009 14:34:16] <purevol> and wait for him to figure out zenoss
\n[28-Sep-2009 14:34:23] <rmatte> they spend all their time running\naround interviewing people about what they are currently doing and\nnothing gets done because of it
[28-Sep-2009 14:34:46] <rmatte> you have to put a certain level of trust in your employees
\n[28-Sep-2009 14:35:20] <purevol> actually, im taking a break from\nbeing a high level telecommnuication engineer, and took this network\nadmin job, because its like, doesnt require me to think or write\ndocumentation or manage people all day
[28-Sep-2009 14:35:36] <purevol> so, i just find this shit trite or humorous, depending on my mood
[28-Sep-2009 14:35:45] <rmatte> hehe
[28-Sep-2009 14:36:00] <purevol> i probably should set up some time with the CIO and tell him what i think
[28-Sep-2009 14:36:00] <rmatte> I took over this job from being a senior network analyst, so it's a step up for me
[28-Sep-2009 14:36:06] <purevol> but i dont want to rip everyone
[28-Sep-2009 14:36:14] <purevol> ive only been here for like 2 months
[28-Sep-2009 14:36:26] <rmatte> yeh, you might want to keep your head down then for a while
\n[28-Sep-2009 14:36:52] <purevol> i dont particularly give a shit,\nim rather good at the politics, its what you get working for the CEO of\nthe largest ISP in the US
[28-Sep-2009 14:36:58]\n<davetoo> So, who's an event/transform wizard in here? What's\nthe best way, from within a transform, to get a count of current events\nof all severities for the device that triggered the current event?
[28-Sep-2009 14:37:51] <davetoo> or at least the highest severity active event?
[28-Sep-2009 14:39:03]\n<rmatte> davetoo: well, I'm sure it's possible but transforms are\nreally designed to pertain to the current event, not to gather data\nfrom other open events
[28-Sep-2009 14:39:27] <purevol> davetoo: are you trying to correlate events?
\n[28-Sep-2009 14:39:43] <davetoo> well what I want to do is\nsupress the current event if there are already other active events at\nthe same severity
[28-Sep-2009 14:39:59] <davetoo> I'm trying to mitigate event/alert floods
[28-Sep-2009 14:40:08] <purevol> event correlation
[28-Sep-2009 14:40:35] <davetoo> My plan is to generate a new event to show that there are auto-suppressed events
[28-Sep-2009 14:40:42] <rmatte> hmmm, I don't know off the top of my head
\n[28-Sep-2009 14:42:22] <rmatte> you'd have to find whatever\nfunction is used to comb through the event console, something like:
[28-Sep-2009 14:42:29] <rmatte> for event in <function>:
[28-Sep-2009 14:42:42] <rmatte> if event.severity == evt.severity:
[28-Sep-2009 14:42:45] <rmatte> and then do whatever
[28-Sep-2009 14:43:05] <rmatte> You'd probably also have to put in:
[28-Sep-2009 14:43:05] <rmatte> #!/usr/bin/env python
[28-Sep-2009 14:43:05] <rmatte> import Globals
[28-Sep-2009 14:43:05] <rmatte> from Products.ZenUtils.ZenScriptBase import ZenScriptBase
[28-Sep-2009 14:43:05] <rmatte> from transaction import commit
[28-Sep-2009 14:43:06] <rmatte> dmd = ZenScriptBase(connect=True).dmd
[28-Sep-2009 14:43:12] <rmatte> at the top of the transform for it to work properly
[28-Sep-2009 14:43:33] <davetoo> actually I've found that I can grab the dmd from the event device entry
[28-Sep-2009 14:44:00] <rmatte> you need a global dmd in this case though
[28-Sep-2009 14:44:31] <rmatte> anyways, that's sort of how you'd do it
[28-Sep-2009 14:44:37] <rmatte> just need to find the right function
[28-Sep-2009 14:45:36] <davetoo> yeah, 'k
[28-Sep-2009 14:45:50] <davetoo> I was doing some stuff over the weekend but lost my work
[28-Sep-2009 14:50:23] <purevol> i wonder if i should remove the baffling from the exhaust on my ducati
\n[28-Sep-2009 14:54:58] <eidolon> hey folks, i'm in a weird\nsituation with a possible broken zenoss enterprise setup. I am getting\n-zero- notifications coming out of it. i set up a suer with a\n'anything in production >= error, send mail' - and I get zilcho. \ni'm looking in Z2.log for warnings or anything, nada. any suggestions?
[28-Sep-2009 14:55:07] <rmatte> no, you should ship that ducati over to me
[28-Sep-2009 14:55:11] <eidolon> test mail to the target account in settings->user does generate mail okay.
[28-Sep-2009 14:55:36] <rmatte> eidolon: for starters, as zenoss user: zenoss status
[28-Sep-2009 14:55:41] <rmatte> are all of the daemons running?
[28-Sep-2009 14:55:57] <purevol> sorry man, no shipping
\n[28-Sep-2009 14:56:21] <eidolon> rmatte: yes, all running. i did\na zenoss restart a few hours ago hoping that might be it
[28-Sep-2009 14:57:53] <eidolon> rmatte: next?
[28-Sep-2009 14:58:18] <rmatte> you mentioned Z2.log, have you checked zenhub.log?
[28-Sep-2009 14:58:36] <eidolon> zenhub.log is empty, dated May 10th.
[28-Sep-2009 14:58:57] <eidolon> doh
[28-Sep-2009 14:58:58] <eidolon> wait
[28-Sep-2009 14:58:59] <rmatte> if you perform a: zenhub run -v10
[28-Sep-2009 14:59:03] <purevol> i wonder if i can use cisco works to turn off telnet and turn on ssh on about 100 ethernet switches
[28-Sep-2009 14:59:03] <rmatte> ?
[28-Sep-2009 14:59:04] <eidolon> nonon.. hang on.
[28-Sep-2009 14:59:18] <rmatte> I can't wait to hear what the issue was lol
\n[28-Sep-2009 14:59:25] <eidolon> there's a\n/opt/zenoss/log/zenhub.log (empty). and /opt/zenoss/log/localhost/ -\nand that localhost dir has a ton of log files i didn't notice
[28-Sep-2009 14:59:28] <rmatte> purevol: yeh, you can
[28-Sep-2009 14:59:38] <rmatte> purevol: it does it all via SNMP as far as I'm aware
[28-Sep-2009 14:59:52] <eidolon> all sorts of chatter there. let me peruse. give me a few minutes.
[28-Sep-2009 14:59:57] <eidolon> (actually i need to move my laptop. brb.
[28-Sep-2009 15:00:02] <rmatte> k
[28-Sep-2009 15:00:13] <rmatte> I have to leave my desk for a few but I'll be back
[28-Sep-2009 15:00:31] <purevol> yeah, i think you can create a JOB
[28-Sep-2009 15:00:44] <purevol> but the only concern i have is generating the RSA key on every single device
[28-Sep-2009 15:02:44] <purevol> these people i work for, bought a cisco nac appliance 2 years ago, and never deployed it
[28-Sep-2009 15:11:40] <eidolon> okay, back.
[28-Sep-2009 15:23:27] <rmatte> purevol: generating the rsa key is just 1 more command
\n[28-Sep-2009 15:26:18] <rmatte> ideally what you'd want to do is\nenable ssh on each device without disabling telnet, then have some\nscript that telnets in to each device and does:
[28-Sep-2009 15:26:20] <rmatte> conf t
[28-Sep-2009 15:26:27] <rmatte> crypto key generate rsa
[28-Sep-2009 15:26:29] <rmatte> 1024
[28-Sep-2009 15:26:35] <rmatte> <enter>
[28-Sep-2009 15:26:38] * eidolon opens a ticket with zenoss enterprise support and calls it art.
[28-Sep-2009 15:26:42] <rmatte> end
[28-Sep-2009 15:26:43] <rmatte> end
[28-Sep-2009 15:26:45] <rmatte> wr mem
[28-Sep-2009 15:26:45] <rmatte> exit
[28-Sep-2009 15:27:07] <rmatte> then once that's done you disable telnet on all of the devices, and you're done
\n[28-Sep-2009 15:27:43] <rmatte> you'd obviously want to run some\nsort of check on each device to make sure that the telnet port is\nclosed and the ssh port is open afterwards
\n[28-Sep-2009 15:36:34] <mrayzenoss> I assume this will be useful:\nblogs/zenossblog/2009/09/28/tip-of-the-month-zendmd-tab-completion
[28-Sep-2009 15:38:41] <rmatte> yeh, I noticed that in King Crab, very cool
[28-Sep-2009 15:39:24] <rmatte> I might apply that during our change window tomorrow night
\n[28-Sep-2009 15:41:50] <rmatte> I also noticed that it stops\ncomplaining about 2 versions of Five in King Crab, though I suspect\nthat it might complain after doing an actual upgrade from 2.4.5 to 2.5
[28-Sep-2009 15:42:31] <mrayzenoss> that may be related to the Zope 2.11 ugprade
[28-Sep-2009 15:43:04] <rmatte> 2009-09-28 16:42:42 WARNING OFS.Application Duplicate Product name
[28-Sep-2009 15:43:04] <rmatte> After loading Product 'Five' from '/usr/local/zenoss/zenoss/Products',
[28-Sep-2009 15:43:04] <rmatte> I skipped the one in '/usr/local/zenoss/zenoss/lib/python/Products'.
[28-Sep-2009 15:43:27] <rmatte> been seeing that ever since the upgrade from 2.3 to 2.4
[28-Sep-2009 15:43:43] <rmatte> haven't tried to fix it since it doesn't appear to hurt anything
[28-Sep-2009 16:14:05] <cgibbons> hmmm
[28-Sep-2009 16:27:00] <rmatte> I love how I put a thumbs up review for something on sourceforge and it counts it as a thumbs down
[28-Sep-2009 16:27:04] <rmatte> that's really accurate
[28-Sep-2009 16:27:12] <mrayzenoss> seriously?
[28-Sep-2009 16:27:18] <mrayzenoss> their stuff is flakey
[28-Sep-2009 16:27:20] <rmatte> https://sourceforge.net/projects/zapplet/
[28-Sep-2009 16:27:22] <rmatte> check that
[28-Sep-2009 16:27:29] <rmatte> I just posted a review with a thumbs up
[28-Sep-2009 16:27:36] <rmatte> and I see it as a thumbs down
[28-Sep-2009 16:27:41] <rhettardo> semantics
[28-Sep-2009 16:27:44] <rmatte> 0% of 1 reviewer recommends this project
[28-Sep-2009 16:27:50] <rmatte> Thumbs down:
[28-Sep-2009 16:27:51] <rmatte> 1
[28-Sep-2009 16:27:57] <mrayzenoss> for me it shows 2 up
[28-Sep-2009 16:28:02] <rmatte> really?
[28-Sep-2009 16:28:02] <mrayzenoss> are you rawdmon?
[28-Sep-2009 16:28:05] <rmatte> yeh
[28-Sep-2009 16:28:10] <mrayzenoss> yeah, shows you as thumbs up
[28-Sep-2009 16:28:12] <rmatte> hmmm, maybe it's my browser then, weird
[28-Sep-2009 16:28:20] <mrayzenoss> nah, their stuff is flakey
[28-Sep-2009 16:28:30] <rhettardo> mine shows no reviews
[28-Sep-2009 16:28:33] <rmatte> hehe
[28-Sep-2009 16:28:41] <mrayzenoss> maybe you can't see yourself?
[28-Sep-2009 16:28:42] <rmatte> so we all see something different when we're looking at the same page
[28-Sep-2009 16:28:46] <rmatte> that's so excellent
[28-Sep-2009 16:28:50] <mrayzenoss> I complained to SourceForge this weekend
[28-Sep-2009 16:28:51] <rhettardo> lol
[28-Sep-2009 16:28:54] <mrayzenoss> in person
[28-Sep-2009 16:29:03] <rmatte> hehe
[28-Sep-2009 16:29:25] <rhettardo> did you tell them to stop making their website worse
[28-Sep-2009 16:29:30] <rmatte> I seem to recall when sourceforge worked a lot better
[28-Sep-2009 16:29:43] <rmatte> it seems to blow badly ever since they've been updating the look lately
[28-Sep-2009 16:29:51] <mrayzenoss> rhettardo: I did indeed
[28-Sep-2009 16:30:31] <rhettardo> it worked perfectly fine w/ the design 3 revisions ago
[28-Sep-2009 16:31:34] <rmatte> wow, I click "Show next 10 reviews" and the link just dissapears and nothing happens
[28-Sep-2009 16:31:37] <rmatte> flakey indeed
[28-Sep-2009 16:32:09] <rmatte> I miss the old red "I mean business" style design
[28-Sep-2009 16:32:16] <rmatte> with just a red top bar
[28-Sep-2009 16:32:38] <rhettardo> yeah
\n[28-Sep-2009 16:33:26] <rmatte> I remember back when I was like\n14 or 15 and started learning FreeBSD, sourceforge was mint back then,\nno bugs
[28-Sep-2009 16:33:51] <rmatte> are they using Jive or something now?
[28-Sep-2009 16:33:54] <rmatte> it looks very similar
[28-Sep-2009 16:34:07] <mrayzenoss> no, it's all custom
[28-Sep-2009 16:34:12] <mrayzenoss> they write their own stuff
[28-Sep-2009 16:34:28] <rmatte> ah
[28-Sep-2009 16:34:37] <rmatte> well the layout is definitely partially ripped off from jive
[28-Sep-2009 16:34:38] <rmatte> lol
[28-Sep-2009 16:35:15] <rmatte> If you put the Zenoss community site up beside it you can see the similarities
[28-Sep-2009 16:36:57] <rmatte> Hmmm, under "Support" on Sourceforge, you don't have the IRC channel listed
[28-Sep-2009 16:37:39] <mrayzenoss> yeah, you get 1 choice from a radio button list
[28-Sep-2009 16:37:43] <mrayzenoss> kinda weak
[28-Sep-2009 16:37:52] <mrayzenoss> I choose "URL"
[28-Sep-2009 16:37:56] <rmatte> ah
[28-Sep-2009 16:38:02] <rmatte> yeh, that is pretty weak
\n[28-Sep-2009 16:38:58] <rmatte> you guys might want to look in to\nsetting up some sort of flash IRC client on the community site which\nconnects to here
[28-Sep-2009 16:39:19] <rmatte> this place is quite honestly the most helpful support resource in my opinion
[28-Sep-2009 16:39:25] <rmatte> forums are good, but this is better
\n[28-Sep-2009 16:40:34] <rmatte> put a link to it on the main page\nthat says "Join us on IRC!" and if they click on it it opens a page\nexplaining that they can either download an IRC client and connect to\nirc.freenode.net channel #zenoss or that they can "Click here" to open\nthe flash client
[28-Sep-2009 16:40:36] <rmatte> just an idea
\n[28-Sep-2009 16:41:30] <rmatte> some people aren't used to\nworking with IRC clients, so it would make things simpler for them
[28-Sep-2009 16:41:46] <rmatte> or maybe they are just lazy and don't want to install anything
[28-Sep-2009 16:41:47] <rmatte>
[28-Sep-2009 16:42:12] <rmatte> any progress on recovering the lost code in the forums?
[28-Sep-2009 16:43:12] <davetoo> rmatte: about my earlier question
[28-Sep-2009 16:43:24] <davetoo> much easier than I thought
[28-Sep-2009 16:43:31] <rmatte> oh?
[28-Sep-2009 16:43:37] <davetoo> dev.dmd works great
[28-Sep-2009 16:43:39] <rmatte> how'd you go about doing it?
[28-Sep-2009 16:43:49] <rmatte> ah
[28-Sep-2009 16:43:57] <rmatte> cool
[28-Sep-2009 16:44:37] <rmatte> well, an hour an a half until the auction for a paintball gun that I want is over...
[28-Sep-2009 16:44:42] * rmatte taps his foot
[28-Sep-2009 16:45:06]\n<davetoo> dev.dmd.ZenEventManager.getEventSummaryME(dev) gets me\nthe count of ACKed and unACKed events at all five severities
[28-Sep-2009 16:45:06] <rmatte> I hope no one else bids because I have it set to auto-bid up, but I'd like to get it for the current bid price
[28-Sep-2009 16:45:22] <rmatte> nice
\n[28-Sep-2009 16:45:45] <rmatte> I need to figure out how to do an\nxmlrpc call in perl to Zenoss to have it move an event to history
[28-Sep-2009 16:45:48] <rmatte> should be fun
[28-Sep-2009 16:45:56] <davetoo> gotta get a bite
[28-Sep-2009 16:46:02] <rmatte> k
\n[28-Sep-2009 16:47:24] <Brad_K> anybody else getting munged rrd\ndate/time stamps on their performance graphs after changing the system\nto proxy the Collector Render URLs via SSL on the hub?
[28-Sep-2009 16:49:00] <rmatte> We haven't proxied any of ours via SSL, so I couldn't tell you
\n[28-Sep-2009 16:52:12] <mrayzenoss> davetoo: did you catch the\ntip of the month? \nblogs/zenossblog/2009/09/28/tip-of-the-month-zendmd-tab-completion
[28-Sep-2009 16:52:41] <mrayzenoss> rmatte: the lost code is still being ironed out in the staging server
[28-Sep-2009 16:53:22] <davetoo> just barely
\n[28-Sep-2009 16:53:24] <mrayzenoss> rmatte: we fixed a lot of\nlittle issues today, our sysadmins were kinda busy the last week\nrelaunching the zenoss.com site
[28-Sep-2009 16:53:37] <davetoo> should be nice; have to see how it's done
[28-Sep-2009 16:53:48] <rmatte> mrayzenoss: cool
[28-Sep-2009 16:54:26] <mrayzenoss> rmatte: and dealing with SourceForge nonsense
[28-Sep-2009 16:55:10] <rmatte> hehe, I can imagine
[28-Sep-2009 17:31:01] <davetoo> rgh
[28-Sep-2009 17:31:15] <davetoo> why is the window to edit the event transform so dinky?
[28-Sep-2009 18:11:13] <rmatte> davetoo: edit it outside that window then paste it in, not overly difficult
[28-Sep-2009 18:11:33] <rmatte> It's not meant to be a full blown editor
[28-Sep-2009 18:11:45] <rmatte> maybe it will be in the future
[29-Sep-2009 00:00:46] [disconnected at Tue Sep 29 00:00:46 2009]
[29-Sep-2009 00:00:46] [connected at Tue Sep 29 00:00:46 2009]
[29-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[29-Sep-2009 08:32:03] <cgibbons_> bouncy
[29-Sep-2009 08:59:02] <ckrough> how can I go back and see when a device was put into maintenance
[29-Sep-2009 09:02:49] <ckrough> nm, I got a log
[29-Sep-2009 09:38:59] <rmatte> good morning folks
[29-Sep-2009 09:39:22] <ckrough> hey
[29-Sep-2009 09:40:02] <ckrough> rmatte: you doing windows monitoring via wmi?
[29-Sep-2009 09:40:32] <rmatte> SNMP, the only thing we ever monitor via WMI is services
[29-Sep-2009 09:40:53] <ckrough> ah, helping a buddy with WMI only and we arent seeing system perf stats
[29-Sep-2009 09:41:00] <ckrough> gotta check my bindings again
[29-Sep-2009 09:41:22] <ckrough> got Egor's zenpacks in there and the perfmon template
[29-Sep-2009 09:41:31] <rmatte> you can try running the queries manually via wmic and see if you get an error
[29-Sep-2009 09:41:38] <rmatte> or just run a test on the datapoint
[29-Sep-2009 09:42:03] <rmatte> by the way, Egor's ZenPacks break in King Crab, so take that in to account when it comes time to upgrade
[29-Sep-2009 09:42:10] <ckrough> will do. thanks
[29-Sep-2009 09:42:20] <rmatte> I believe he's working on a new version that will work with King Crab
[29-Sep-2009 09:42:32] <ckrough> thats going to be a tough spot for the zenoss community over time.
[29-Sep-2009 09:42:38] <rmatte> yeh
\n[29-Sep-2009 09:42:58] <ckrough> when developers abandon zenpacks\nand there are incompatible upgrades. Hopefully there will be enough\ncommunity that someone picks them up
[29-Sep-2009 09:43:06] <rmatte> I'm thinking a lot of stuff will break when they do the zope and python upgrades
[29-Sep-2009 09:43:08] <ckrough> but I guess thats the nature of open source, if they are worth it someone will take them
[29-Sep-2009 09:43:14] <rmatte> I believe the zope upgrade is planned for stone crab
[29-Sep-2009 09:43:25] <rmatte> actually, python as well I believe
[29-Sep-2009 09:43:31] <rmatte> so that's going to be a fun release
[29-Sep-2009 09:43:34] <ckrough> yeah
[29-Sep-2009 09:43:44] <ckrough> we've got so many monkey patches in here now...
\n[29-Sep-2009 09:43:45] <rmatte> whole new UI should be\nimplemented on that release though, so it's going to be uber spiffy
[29-Sep-2009 09:43:48] * ckrough scared
[29-Sep-2009 09:43:51] <rmatte> hehe
[29-Sep-2009 09:44:04] <rmatte> all I can say is two words
[29-Sep-2009 09:44:05] <rmatte> lab box
\n[29-Sep-2009 09:44:45] <rmatte> After the fun that we went\nthrough with the 2.4.3, 2.4.4, and 2.4.5 upgrades we're going to be\ndoing rigorous lab testing next time around to work out any issues\nprior to upgrading our prod boxes
[29-Sep-2009 09:44:52] <ckrough> heh. got a staging config and a dev config
[29-Sep-2009 09:45:14] <rmatte> we just have a "Client base image" that we load up, it's pretty nice
\n[29-Sep-2009 09:45:31] <rmatte> we just clone the VM, boot it up,\nswitch the IPs, hostname, edit the hosts file, and we're up and running
[29-Sep-2009 09:45:38] <rmatte> takes like 15 mins to get a new server online
\n[29-Sep-2009 09:46:42] <rmatte> w00t, meeting in 15 mins to\ndiscuss how we're going to have our ticketing system integration\nauto-historify Zenoss events after it opens tickets
[29-Sep-2009 09:47:24] <rmatte> I've found the xmlrpc calls in python, just need to figure out how to do it in perl
\n[29-Sep-2009 09:47:27] <Brad_K> you know, i don't really care\nabout uber-spiffy UIs. I care more about finding and fixing bugs,\nespecially long-standing bugs that have been around since 2.3.3.
[29-Sep-2009 09:47:37] <rmatte> oh, I agree
[29-Sep-2009 09:48:04]\n<rmatte> In my opinion, I would have been good with the current\nUI for another 2 years if it meant that the actual features would get\nricher and less buggy
[29-Sep-2009 09:48:05]\n<Brad_K> i mean, that's what we're supposed to be paying for,\nright? i mean, this is supposed to be an Enterprise Monitoring system,\nright?
[29-Sep-2009 09:48:37] <rmatte> but you\nhave to remember that this is also a company's product, the new UI is\nmore of a marketing move than anything
[29-Sep-2009 09:48:42] <Brad_K> we need stability and painless backups/restores and upgrades.
\n[29-Sep-2009 09:48:50] <ckrough> We are hitting the end of\nscalability with the current archetecture too. Im firing up my 11th\ncollector for one installation now
[29-Sep-2009\n09:48:55] <rmatte> then again the new UI looks sexy as heck, and\nI'm a sucker for eye-candy, so it won't be all bad
[29-Sep-2009 09:48:55] <ckrough> thats too many uber-boxes
[29-Sep-2009 09:49:08] <ckrough> each box is 32g ram, yada yada. big hardware
[29-Sep-2009 09:49:23] * ckrough architecture
\n[29-Sep-2009 09:49:39] <Brad_K> and the more i get into this, the\nless it seems that anyone is paying attention to that critical core\ninfrastructure that isn't sexy, but without which everything else is\nworthless.
[29-Sep-2009 09:49:40] <rmatte> yeh, we have 3 boxes with 32g each, and we're just starting out
[29-Sep-2009 09:50:14] <rmatte> Brad_K: I agree, but hopefully once the UI is done they'll get back to that
\n[29-Sep-2009 09:50:18] <Brad_K> we started out with four boxes\nwith 16GB each, one as hub and the other three as collectors. so far,\nwe're really only using the one hub and one of the collectors.
\n[29-Sep-2009 09:50:21] <ckrough> I think a lot of the improvement\nare driven by the community, which is typically smaller installations\nwith only a couple users, so it drives UI flash instead of foundation\nbuilding
[29-Sep-2009 09:50:41] <Brad_K> rmatte:\ni have not been given any reason to believe that they will ever pay\nattention to anything but the sexy UI stuff.
[29-Sep-2009 09:50:43] <rmatte> Brad_K: ah, we have separate instances running on VMs on the boxes
[29-Sep-2009 09:51:10] * eidolon opens up a ticket with support at zenoss.
[29-Sep-2009 09:51:23] <rmatte> hehe
\n[29-Sep-2009 09:51:44] <rmatte> Brad_K: I doubt that's true, the\nUI is just a big project for them right now, so a lot of resources are\nbeing devoted to it
[29-Sep-2009 09:52:04]\n<Brad_K> all my research so far indicates that backups are\nproblematical, upgrades are problematical, and most of the toolkit only\nkinda-semi-sorta works. if you can do what you want/need to do with\njust SNMP, then you should be fine. otherwise, there are rough seas\nahead.
[29-Sep-2009 09:52:20] <eidolon> rmatte:\nthanks for your help yesterday. i had rusty and (nick?) on the phone\nfor a while, they helped a ton. now i'm trying to fix the\nno-alerts-going-out problem on the old install.
[29-Sep-2009 09:52:34] * eidolon is a tech support departments thorn-in-the-side.
\n[29-Sep-2009 09:52:51] <Brad_K> and that's for people who are\nPython/Zope/Zenoss system developers. if you're not a system\ndeveloper, then it seems to me that you're going to be poorly prepared\nfor those rough seas.
[29-Sep-2009 09:53:34]\n<rmatte> Brad_K: I'm not a "system developer", but I've learned\nenough along the way to be able to get Zenoss to do basically anything\nvia custom scripts
[29-Sep-2009 09:54:35]\n<rmatte> Brad_K: this is not a product that is designed to be\nmanaged by your friendly neighbourhood LAN admin. You need\nsysadmin/scripting experience as well as a knowledge of SNMP/WMI,\nnetworking, and servers.
[29-Sep-2009 09:54:37]\n<Brad_K> rmatte: so you've hacked on it enough to make it\nZenrmatte at your site, as opposed to Zenoss.
[29-Sep-2009 09:54:46] <ckrough> ^this
\n[29-Sep-2009 09:55:19] <ckrough> same boat we are in. we are in\nthe process of trying to commit codebase changes to get our non-zenpack\ncustomizations rolled in
[29-Sep-2009 09:55:23]\n<rmatte> Brad_K: It's not even "hacking on it", you design a\nscript, which is essentially a plugin, plug it in as a command based\ndatapoint (which is very simple to do), and off you go
[29-Sep-2009 09:56:03]\n<Brad_K> i've got 20 years of sysadmin experience, and SNMP is\nthe least of my worries. I don't know much about WMI, but we've got\nWindows people who do. But, most critically, i'm not a system\nprogrammer.
[29-Sep-2009 09:56:16] <Brad_K> except zencommand is b0rken.
[29-Sep-2009 09:56:25] <rmatte> how is it broken?
[29-Sep-2009 09:56:38] <rmatte> a lot of our monitoring depends on zencommand and it works perfectly fine
[29-Sep-2009 09:56:57] <rmatte> I have 5 months worth of performance graphs populated via custom scripts to prove it
[29-Sep-2009 09:57:01]\n<Brad_K> sometimes stuff runs fine for a while, then it stops\nrunning. then it might start running fine again a few days later.
\n[29-Sep-2009 09:57:22] <rmatte> That seems to be a problem that\nonly you're experiencing for whatever reason, my zencommand runs\nabsolutely flawless
[29-Sep-2009 09:57:31] <Brad_K> rmatte: and which version of Zenoss are you on?
[29-Sep-2009 09:57:51] <rmatte> 2.4.5 at the moment, but it ran flawless in 2.4.2, 2.4.1, and 2.3.3
[29-Sep-2009 09:57:58] <rmatte> I haven't had any issues with it along the way
\n[29-Sep-2009 09:59:55] <Brad_K> the Zenoss ssh stuff doesn't\nproperly close the connections, so you get massive amounts of error log\nspam on the monitored servers. so, we use the tried-and-tested Nagios\ncheck_ssh instead. the Zenoss SSL checking stuff will generate errors\nfor SSL certificates expiring on each of 30 days, 14 days, 7 days, 3\ndays, and 1 day, as opposed to just generating one error for one of\nthose timeframes. we don't need seven alerts for the same error.
[29-Sep-2009 10:00:35]\n<rmatte> it's not supposed to close the connections, the\nconnection are designed to stay open. think about it, if it logged in\nevery 5 minutes it would make your auth logging for SSH absolutely\nuseless because it would flood the log out
[29-Sep-2009 10:01:23]\n<Brad_K> the ssh checking doesn't stay logged in. it opens the\nconnection, sees that the port is being listened to, and then drops the\nconnection without properly closing it.
[29-Sep-2009 10:01:38] <rmatte> oh, you're talking about IpServices
[29-Sep-2009 10:01:40] <Brad_K> maybe this is a bug in the 2.4.2 version.
[29-Sep-2009 10:01:53] <rmatte> I've never noticed that issue, and we monitor SSH on a lot of servers
[29-Sep-2009 10:02:04] <rmatte> There was a bug with that a long while back, but it was fixed
[29-Sep-2009 10:02:31] <rmatte> anyways, I have a meeting to head off to, I'll be back in a bit
[29-Sep-2009 10:03:10]\n<rmatte> (If you are noticing these issues you might want to dig\na bit deeper, if possible, and open some trac tickets
[29-Sep-2009 10:03:36] <rmatte> You might also try a fresh 2.4.5 install and see if that solves your problems
[29-Sep-2009 10:04:23]\n<Brad_K> it seems to me that wer're getting some of the worst\naspects of open source (where people only work on the things that they\nfind sexy and the only support you get is on the IRC channel or the\nmailing lists) and the worst aspects of closed source.
[29-Sep-2009 10:05:10]\n<Brad_K> we've got a number of support cases that we've opened,\nand we're going to be doing a WebEx session in about thirty minutes on\none of those. we're going to be opening a lot more support cases as\nwell.
[29-Sep-2009 10:06:24] <eidolon> Brad_K: this is why we went with an enterprise support contract with zenoss.
[29-Sep-2009 10:06:31] <eidolon> they've been -awesome- support for us.
[29-Sep-2009 10:07:00]\n<Brad_K> eidolon: we've got an enterprise support contract with\nthem. and the best support we get is from this IRC channel, mostly\nfrom rmatte.
[29-Sep-2009 10:07:43] <eidolon>\nand, apropos of nothing, what Windows determines is a system-level\n"Warning" is really irritating. "Fine. You disabled a printer. \nThat's very nice for a production deployed coloed Win2k8 machine. \nSTFU!"
[29-Sep-2009 10:07:58] <eidolon> Brad_K: that's not been my experience, but okay
[29-Sep-2009 10:08:26]\n<eidolon> if you have an emergency / real problem, i tend to open\na case, and if there's no response within an hour or so, i call them\ndirectly.
[29-Sep-2009 10:08:26] <Brad_K> it\ntakes several hours, or days, for them to respond to support cases that\nare filed through the official SugarCRM portal. at least on irc,\nthere's a chance that you might catch one of the developers, or be\nlucky enough that rmatte may be around and have already run into the\nsame problem.
[29-Sep-2009 10:08:37] <eidolon> i've been on the phone with rusty 2-3 times in the last month, and he's -awesome-.
[29-Sep-2009 10:08:53] <eidolon> Brad_K: if you're seeing days response time on the CRM requests, then you raise high holy hell to rusty.
[29-Sep-2009 10:09:06] <eidolon> because your SLA says you should get responses a lot faster.
\n[29-Sep-2009 10:13:33] <ckrough> rmatte: you should be seeing\nlots of these: " sshd[6963]: Did not receive identification string from\nyour.zenoss.box"
[29-Sep-2009 10:13:44] <ckrough> INFO level in syslog
[29-Sep-2009 10:14:06] <ckrough> We gotten good support through the enterprise license
[29-Sep-2009 10:14:22] <ckrough> I think IRC is a better forum for the customer though, since it's interactive
[29-Sep-2009 10:14:41] <ckrough> bouncing tickets back and forth in sugarcrm is never going to be as responsive as IRC
[29-Sep-2009 10:15:00] <ckrough> but overall Im very satisfied with ent support
[29-Sep-2009 10:15:37] <eidolon> well, there's only so many questions rmatte can answer
[29-Sep-2009 10:15:43] <ckrough> I usually try to convince them to PRIVMSG me in IRC when we start working a ticket
[29-Sep-2009 10:16:23] <ckrough> too bad matt isnt here, he should be noting all of this
[29-Sep-2009 10:16:36] <ckrough> zenoss-logger: you getting all this
\n[29-Sep-2009 10:16:55] <eidolon> well, i'll throw this one out. \ni ahve a zenoss install that is catching events, and showing them in\nteh console, but though i have a group rule that should notify via\nemail when something => warning is hitting, we're not seeing any\nmail.
[29-Sep-2009 10:17:03] <Brad_K> maybe i'm just frustrated because of all the problems we've been having lately.
[29-Sep-2009 10:17:05] <eidolon> test mail works. zenoss status shows everything running.
[29-Sep-2009 10:17:13] <Brad_K> maybe you shouldn't pay too much attention to me right now.
[29-Sep-2009 10:17:15] <eidolon> i opened a case on this already
[29-Sep-2009 10:17:20] * eidolon ignores Brad_K. *nod*
\n[29-Sep-2009 10:17:53] <ckrough> Brad_K: I can relate to where\nyou are coming from, there is definitely some roughness to the product,\nand some areas that make me nervous about the future
[29-Sep-2009 10:18:05] <ckrough> but on the flipside, it's the best solution for what we need to do here
[29-Sep-2009 10:19:00]\n<exarkun> I'm looking for the part of the zenoss code that...\ndoes things. I see a bundled version of Python, a bundled version of\nZope, some tiny scripts, but I don't see anything that looks like it\ncan actually monitor anything. Can anyone give me a pointer?
[29-Sep-2009 10:19:46] <ckrough> exarkun: its a colletion of processes
[29-Sep-2009 10:19:49] <cgibbons> that's a bit open-ended
[29-Sep-2009 10:20:04] <eidolon> that's zenoss
[29-Sep-2009 10:20:12] <eidolon> zenoss is about 18 or so programs that work together
[29-Sep-2009 10:20:24] <eidolon> follow the installation process.
\n[29-Sep-2009 10:20:31] <Brad_K> and no, i've decided that i'm not\ngoing to go vent my frustrations on some sort of public blog. i guess\ni'm just going to keep beating my head against the wall for a while\nlonger....
[29-Sep-2009 10:20:33] <ckrough> exarkun: the documentation reviews the roles I believe
[29-Sep-2009 10:20:43] <exarkun> ckrough: There's a lot of documentation
[29-Sep-2009 10:20:49] <exarkun> ckrough: Any in particular?
[29-Sep-2009 10:20:53] <ckrough> sec
[29-Sep-2009 10:21:00] <eidolon> how about "Introduction to zenoss : getting started"
[29-Sep-2009 10:21:16] <eidolon> are you looking for details on how it works? I recommend setting it up and watching it.
[29-Sep-2009 10:21:16] <eidolon> otherwise the concepts won't help much.
[29-Sep-2009 10:21:17] <ckrough> start here docs/DOC-3061
[29-Sep-2009 10:21:26] <eidolon> there yago.
[29-Sep-2009 10:21:30] <eidolon> ~ckrough++
[29-Sep-2009 10:22:39] <exarkun> eidolon: I did that a bit. I'm trying to learn enough to add some features.
[29-Sep-2009 10:22:59] <eidolon> eek
[29-Sep-2009 10:23:08] <eidolon> customizing zenoss requires a really good understanding of what it does.
[29-Sep-2009 10:23:10] <eidolon> and how it does it.
[29-Sep-2009 10:23:14] <exarkun> I'm sure.
[29-Sep-2009 10:23:17] <ckrough> exarkun: this may clear things up too docs/DOC-3020
[29-Sep-2009 10:23:27] <eidolon> and, unless it's a custom application, i bet someone has done what you need already.
\n[29-Sep-2009 10:23:59] <exarkun> eidolon: Maybe on the way to\nlearning enough to extend it, I'll learn enough to recognize that what\nI'm after is already possible somehow.
[29-Sep-2009 10:24:18] <eidolon> that's why i suggest getting an installation up and running.
[29-Sep-2009 10:24:22] <eidolon> you learn a lot along the way
\n[29-Sep-2009 10:24:23] <cgibbons> there's also a developer's\nguide that continues to be enhanced, but depending upon how deep you\nwant to go it probably won't be enough
[29-Sep-2009 10:24:40] <cgibbons> so what are you after, anyway?
[29-Sep-2009 10:24:41] <exarkun> I spent a couple hours clicking around in an installation yesterday
[29-Sep-2009 10:24:50] <exarkun> Need to read some docs or source to get further
\n[29-Sep-2009 10:24:56] <eidolon> i certainly had a hard time\nunderstanding some basic concepts when i started, but once i ran,\nsetup, administered, and worked with an install for a few months, i\nknow what it can and can't do.
[29-Sep-2009 10:25:07] <eidolon> yah, what do you need to do?
[29-Sep-2009 10:25:51] <ckrough> is the API no longer available online? that was my favorite doc
\n[29-Sep-2009 10:26:21] <eidolon> oh, let me know if you find it? \ni was grumpyu yesterday looking for an object definition for evt.
\n[29-Sep-2009 10:26:59] <ckrough> not sure, but it looks like it\nis no longer 'live' in html, search for API on the community site and\nyou can download it
[29-Sep-2009 10:27:14] <eidolon> k
[29-Sep-2009 10:27:21] <cgibbons> that's on matt's todo list to fix i think
\n[29-Sep-2009 10:27:21] <exarkun> I'm not sure what all the right\nterms are yet. If "agent" is the word for the process that runs on\nnodes and communicates monitoring information back to the main zenoss\nserver (is that the "hub"?), then I need a custom agent that uses a\nspecial transport.
[29-Sep-2009 10:27:28] <purevol> ahoy there fellas
[29-Sep-2009 10:27:45] <cgibbons> that would be a collector, exarkun
[29-Sep-2009 10:27:52] <cgibbons> agent typically implies something that runs on a monitored end-point/device/system
[29-Sep-2009 10:28:02] <ckrough> I think thats what he meabt
[29-Sep-2009 10:28:03]\n<eidolon> exarkun: wht cgibbons said. but there's a ton of\ntransports already. do you need to go-to-machine-and-run-a-command?
[29-Sep-2009 10:28:07] <ckrough> node being the monitored box
[29-Sep-2009 10:28:10] <eidolon> ala, a nagios approach?
[29-Sep-2009 10:28:30] <exarkun> This needs to run on the monitored end-point/device/system
[29-Sep-2009 10:28:36] <ckrough> most zenoss monitoring is agentless, or uses something native like snmp/wmi
[29-Sep-2009 10:28:44] <ckrough> what are you monitoring
[29-Sep-2009 10:28:58] <exarkun> The monitored system can't get snmp get, and can't accept incoming connections (eg ssh)
[29-Sep-2009 10:29:09] <exarkun> Some machines in a hosted cluster
[29-Sep-2009 10:29:20] <eidolon> so it needs to publish traps back to a collector somewhere
[29-Sep-2009 10:29:22] <exarkun> s/snmp get/snmp out/
[29-Sep-2009 10:29:43] <ckrough> someone else here had that same issue, cant remember the final solution
[29-Sep-2009 10:29:47] <ckrough> deja vu
\n[29-Sep-2009 10:30:51] <cgibbons> we had a big customer @ a\ndifferent company whose entire monitoring network was that way. the\nmonitored systems could send data back to the infrastructure, but the\ninfrastructure itself couldn't pull data. thousands of remote-sites\nthat just did data collection locally.
[29-Sep-2009 10:31:05] <eidolon> yep.
[29-Sep-2009 10:31:14] <eidolon> my immediate answer is to set up an SSL VPN.
[29-Sep-2009 10:31:19] <eidolon> and monitor over that.
[29-Sep-2009 10:31:29] <eidolon> but you can push to a collector, that'll work fine.
[29-Sep-2009 10:31:45] <kmtsun> join #nagios
\n[29-Sep-2009 10:32:46] <exarkun> Not sure if it'll be possible to\nset up an SSL VPN, but I'll keep that idea in mind. Thanks for all the\npointers so far.
[29-Sep-2009 10:34:03]\n<cgibbons> there's a new, easier-to-use (for developer's)\ncollector framework in the version about to be released but there's\nstill a lot to do if you want to build an entirely new daemon\nframework, etc.
[29-Sep-2009 10:34:13] <cgibbons> i think we've only had it happen once or twice with the community
[29-Sep-2009 10:37:41] <kmtsun> ls
\n[29-Sep-2009 10:59:47] <feutete_> I am using the HTTP Monitor\nZenPack, and it appears that timeouts when running check_http never\nauto-clear. any idea why?
[29-Sep-2009 11:05:21] <rmatte> annnnnd I'm back
[29-Sep-2009 11:07:51]\n<rmatte> ckrough: if you're logging you might as well save it and\nsend it to Matt later on, these are all good points
[29-Sep-2009 11:08:32]\n<rmatte> they really need a logging bot setup in here that's here\n24/7, that way we could be like "Refer to the conversation at\napproximately 11:00-12:00 on September 29th".
[29-Sep-2009 11:13:11] <ckrough> there are some good bots that link to docmentation too, that would be helpful
[29-Sep-2009 11:13:18] <ckrough> maybe Ill play with supybot this weekend
[29-Sep-2009 11:13:42] <rmatte> yeh, it would be great to get something setup in here
[29-Sep-2009 11:13:55] <rmatte> it's almost strange to see an IRC channel without a bot these days
[29-Sep-2009 11:14:10] <rmatte> we could potentially just setup an eggdrop with some custom tcl scripts
[29-Sep-2009 11:14:12] <ckrough> matt said they used to have one, he wasnt sure where it went and why
[29-Sep-2009 11:14:44] <rmatte> I'll setup an eggdrop on my home server and see what I can do with it
[29-Sep-2009 11:14:52] <rmatte> been a while since I've done anything with one
[29-Sep-2009 11:15:24] <rmatte> woohoo, time to commission a new client
[29-Sep-2009 11:15:38] <rmatte> going to be a breeze, 45 devices, all Cisco
[29-Sep-2009 11:17:33] <eidolon> oh please don't run an eggdrop bot.
[29-Sep-2009 11:17:40] <eidolon> jeez, that's so 1997.
[29-Sep-2009 11:17:48] <eidolon> at least run an infobot that we can run factoids in.
[29-Sep-2009 11:18:06] * eidolon can probably get the javabot code in. i'm an old hat at that.
[29-Sep-2009 11:19:03] <rmatte> exarkun: all the files that "do" stuff are mostly in the Products directory ($ZENHOME/Products)
[29-Sep-2009 11:19:39] <rmatte> an eggdrop can have an infobot script thrown on it
[29-Sep-2009 11:19:40] <eidolon> but first, lunch.
[29-Sep-2009 11:19:43] * eidolon zips.
[29-Sep-2009 11:19:48] <rmatte> it's quite flexible
[29-Sep-2009 11:20:03] <rmatte> I just like them because they are stable as hell
[29-Sep-2009 11:20:56] <rmatte> and for the record, it's "1337"
[29-Sep-2009 11:22:41] * ckrough votes supybot
[29-Sep-2009 11:22:52] <ckrough> ands it python I think
[29-Sep-2009 11:23:17] <rmatte> oooh, python
[29-Sep-2009 11:56:01] <rhettardo> it is not 1337
[29-Sep-2009 11:56:03] <rhettardo> its 2009
[29-Sep-2009 12:38:15] <rmatte> lol
\n[29-Sep-2009 12:39:50] <cjoslin> not looking for a how to right\nnow.. but i am trying to find a NMS that works well in our WINDOWS\nenvironment. can zenoss alert me with error logs from eventlog? can\nit monitor disk space, ping, cpu, services? i know are decent paid\nsolutions out there. Loved PA server monitor Trial, however we have\na small budget.
[29-Sep-2009 12:40:38] <rmatte>\ncjoslin: Well, it can do all of that, the Core version needs a mix of\nSNMP and WMI to do all of that, the enterprise version can do all of\nthat via WMI
[29-Sep-2009 12:43:41] <cjoslin> is WMI just better implemented in the enterprise version?
[29-Sep-2009 12:47:51] <cjoslin> rmatte... ?
[29-Sep-2009 12:50:30] <rmatte> yes
[29-Sep-2009 12:50:39] <rmatte> sorry, had to step away for a second
[29-Sep-2009 12:50:55] <rmatte> it's better implemented in the enterprise version
\n[29-Sep-2009 12:51:26] <rmatte> you can monitor event log,\nservices, and performance data via WMI with Zenoss core, but you can't\nmonitor interfaces, and filesystem
[29-Sep-2009 12:51:43] <rmatte> so if you want to use core you can use a mix of both, or just use straight WMI
[29-Sep-2009 12:51:46] <rmatte> erm
[29-Sep-2009 12:51:49] <rmatte> straight SNMP I mean
[29-Sep-2009 12:52:21] <cjoslin> from what i have read tho monitoring event log with snmp is super tricky.
[29-Sep-2009 12:52:37] <cjoslin> but if it can do both then i would like to try it out.
[29-Sep-2009 12:52:42] <cjoslin> since its free and all
\n[29-Sep-2009 12:52:52] <rmatte> well, it depends to what degree\nyou want to monitor it, if you configure SNMP traps windows will send\nan snmp trap to Zenoss if major events occur
[29-Sep-2009 12:53:19] <rmatte> but core does support eventlog monitoring via WMI
[29-Sep-2009 12:53:23] <rmatte> so you can do it that way as well
[29-Sep-2009 12:53:27] <rmatte> it's just much chattier
[29-Sep-2009 12:55:51] <rmatte> to do the performance monitoring you'll need to install a ZenPack
\n[29-Sep-2009 12:56:32] <cjoslin> well with all that said.. do\nyou think i should go with zenoss to do those kinda things? again \nmainly windows enviroement here. aside from the asterisk pbx.
[29-Sep-2009 12:58:04]\n<cjoslin> i mean, am i looking at doing coding or going really\ndeep into config files. or is it something i can have at least working\nby the end of the day?
[29-Sep-2009 13:09:02] <rmatte> you can have it working by the end of the day
[29-Sep-2009 13:09:20] <rmatte> I use Zenoss to monitor over 100 Windows servers.
[29-Sep-2009 13:09:24] <rmatte> I do it all via SNMP though
[29-Sep-2009 13:09:41] <rmatte> but I've worked with the WMI before
[29-Sep-2009 13:11:14] <rmatte> do you want to do your actual performance monitoring (CPU/Memory) via WMI or SNMP?
\n[29-Sep-2009 13:16:48] <cjoslin> rmatte... really whatever is\neasier i am the only admin around here so its kinda just to get\nsomethign working..
[29-Sep-2009 13:17:28] <chudler> ah.. one man show, eh? Used to love those days
[29-Sep-2009 13:17:42] * chudler crawls back under rock
\n[29-Sep-2009 13:18:16] <cjoslin> lol well we have developers\nfor the service we provide and all.. but as far as sys/network admin\nthats me
[29-Sep-2009 13:18:41] <cjoslin> i prefer it to being more specialized since it helps alleviate my adhd :-)
\n[29-Sep-2009 13:18:57] <rmatte> the WMI would probably be easier\nsince Matt hasn't published the simple version of my SNMP Performance\npack yet
[29-Sep-2009 13:19:33] <rmatte> You'll\nwant to go to the ZenPacks section of the community site and download\nthe WMI Data Source and WMI Performance Template ZenPacks (both are by\nEgor)
[29-Sep-2009 13:19:43] <rmatte> then unzip them, and install them
\n[29-Sep-2009 13:19:57] <rmatte> (they aren't supposed to be\nzipped, the new community site zips everything automatically apparently)
[29-Sep-2009 13:20:11] <cjoslin> i am downloading the vmware image of zenoss core as we type
[29-Sep-2009 13:20:19] <rmatte> k
\n[29-Sep-2009 13:20:38] <rmatte> you'll want to read the Zenoss\nAdmin guide at some point to really understand how Zenoss works and how\nto configure it
[29-Sep-2009 13:20:47] <rmatte> but we can get you setup with something for the time being
[29-Sep-2009 13:22:26] <cjoslin> so no WMI out of the box eh?
[29-Sep-2009 13:22:33] <cjoslin> looks like i am going to head over to this community site.
\n[29-Sep-2009 14:48:19] <rmatte> cjoslin: it does event log and\nservice monitoring via WMI out of the box, but not performance
\n[29-Sep-2009 14:48:46] <rmatte> and you'll have to monitor\ninterfaces and disks using SNMP, unless you purchase the enterprise\nversion
[29-Sep-2009 15:04:24] <ke4qqq> I just\ninstalled dnsmonitor zenpack - and get this: http://fpaste.org/g8ZB/ ,\nwhen I try and look at the zenpack - is this a known problem??
[29-Sep-2009 15:04:44] <ke4qqq> I don't see an open ticket for it
[29-Sep-2009 15:09:47] <rmatte> restart zenoss and try again
[29-Sep-2009 15:10:07] <rmatte> some zenpacks require that zenoss be restarted after they are installed
[29-Sep-2009 15:16:40] <ke4qqq> rmatte: thanks, I'll try that
[29-Sep-2009 15:18:04] <alinux> hi all, does Zenoss require a specific client or just SNMP clients ?
[29-Sep-2009 15:18:18] <cgibbons> "client" ?
[29-Sep-2009 15:19:20] <ke4qqq> alinux: zenoss doesn't have 'clients' - it will use SNMP, WMI, ssh, etc
[29-Sep-2009 15:20:53] <ke4qqq> it's considered client-less
[29-Sep-2009 15:21:52] <gwb235> marketing lingo was "agent-less"
[29-Sep-2009 15:22:03] <rhettardo> yes
[29-Sep-2009 15:22:21] <ke4qqq> yes
[29-Sep-2009 15:24:08] <mrayzenoss> we've trained you well
[29-Sep-2009 15:28:29] <rmatte> lol
[29-Sep-2009 15:28:29] <gwb235> I, for one, am glad to welcome our new monitoring overlords
\n[29-Sep-2009 15:31:26] <rmatte> The whole, "we've trained you\nwell thing": "I am a Zedi, like my father before me!" "...sooo be it, \nZedi."
[29-Sep-2009 15:31:46] <rmatte> *** queue lightning from fingers ***
[29-Sep-2009 15:31:58] <rmatte> cue*
\n[29-Sep-2009 15:32:26] <rmatte> eughhhhh, this is so boring, I'm\nadding 45 devices in but only 15 are actually monitored, the others\njust represent stacked switches
[29-Sep-2009 15:39:29] <cgibbons> the really dumb kind?
[29-Sep-2009 15:40:33] <tung> I got another dumb question for you. Since I couldn't find the answer
[29-Sep-2009 15:40:33] <tung> in any Zenoss forum or doc I figure it must be something basic I'm
[29-Sep-2009 15:40:34] <tung> missing
[29-Sep-2009 15:40:53] <tung> I've created a new local template for a device, let's say
[29-Sep-2009 15:41:17] <tung> \\/Devices/Server/devices/fe01 (fe01 is my host name)
[29-Sep-2009 15:41:25] <tung> In this new template, I added the 3 data points along with 3
[29-Sep-2009 15:41:26] <tung> thresholds and graphs from a script I created.
[29-Sep-2009 15:41:33] <tung> I can see the graphs just find in Perf tab.
[29-Sep-2009 15:41:40] <tung> I created a new ZenPack and add this new template into that ZenPack
[29-Sep-2009 15:42:09] <tung> Now I want to replicate this template to the other 20 boxes that I
[29-Sep-2009 15:42:09] <tung> want this on.
[29-Sep-2009 15:42:17] <tung> But I when I go to the other systems and try to Bind Templates, the
[29-Sep-2009 15:42:17] <tung> new template is not listed as a choice.
[29-Sep-2009 15:42:25] <tung> What am I missing?
[29-Sep-2009 15:42:29] <tung> Thanks in advance
[29-Sep-2009 15:43:26] <ke4qqq> tung
[29-Sep-2009 15:43:37] <ke4qqq> tung: what you want to do is copy the template not make a local copy
[29-Sep-2009 15:44:15] <ke4qqq> so go to Devices> Templates> and either create a new template
[29-Sep-2009 15:44:26] <ke4qqq> or copy an existing template that is located there
[29-Sep-2009 15:44:36] <tung> How do I do that? Yes I Created the new template
[29-Sep-2009 15:44:53] <ke4qqq> Click on Devices under the "classes" menu on the left
[29-Sep-2009 15:45:03] <ke4qqq> then click on the Templates tab
[29-Sep-2009 15:45:13] <ke4qqq> you'll see Available performance templates
\n[29-Sep-2009 15:45:50] <ke4qqq> (it only shows templates\navailable at /Devices, so a Windows template might not show up there,\nand would be a bit deeper, but the same principle applies)
[29-Sep-2009 15:45:59] <tung> OK I'm there but I don't see the one I have created
\n[29-Sep-2009 15:46:23] <cgibbons> it's probably just created at a\ndeep level than you intended, so it's not available in the device class\nyou're adding your new devices to
[29-Sep-2009 15:46:51] <tung> Mine is under /Devices/Server/Linux/device/hostname
\n[29-Sep-2009 15:46:54] <ke4qqq> so - there or somehwere similar\n(/Devices/Server/Windows ) if you create it on the device, it'll only\nbe visible at that device, and you want to move at least one level up\nto create it.
[29-Sep-2009 15:47:08] <tung> HOw do I move it up one level?
[29-Sep-2009 15:47:23] <ke4qqq> so it would be at least available to all Linux servers in your example
[29-Sep-2009 15:47:49] <tung> yeah that's exactly what I need to do but I don't see an option to move a template
[29-Sep-2009 15:48:18] <tung> "Remove Local Copy" was not a good thing to do :-)
\n[29-Sep-2009 15:48:27] <ke4qqq> tung go to /Devices/Server/Linux\nclick on the templates tab and create it there. Unfortunately I am not\naware of a way to copy it
[29-Sep-2009 15:48:58] <tung> really?
[29-Sep-2009 15:49:39] <ke4qqq> others more knowledgeable than I might know, but I don't
[29-Sep-2009 15:49:59] <cgibbons> there's probably a way, let me ask... one moment
[29-Sep-2009 15:50:18] <tung> OK if that would work then I think it's good
[29-Sep-2009 15:50:23] <tung> Thanks. I think that helps.
[29-Sep-2009 15:50:38] <tung> At least I don't have to create the templates 20 times :-D
[29-Sep-2009 15:55:56] <rmatte> I might know a way to move it up...
[29-Sep-2009 15:56:34] <rmatte> hmmm, nope, won't work for local copies apparently
[29-Sep-2009 15:57:29] <cgibbons> prolly have to do it manually via zendmd
\n[29-Sep-2009 15:59:58] <tung> That's ok. Thanks everyone. I\nrecreated it in /Devices/Server/Linux and it's now visible to my other\nhosts
[29-Sep-2009 15:59:58] <tung> So that did it
[29-Sep-2009 16:00:03] <tung> didn't take too much time
[29-Sep-2009 16:01:18] <cgibbons> cool
[29-Sep-2009 16:05:28] <jb> hrm looks like all of my windows process monitoring is broken
[29-Sep-2009 16:05:30] <jb> at least the graphs
[29-Sep-2009 16:22:00]\n<chudler> tung: I know it is too late, but when viewing the\ndevice, add /perfConfig to the end of the URL. Then you should see the\ntemplates on the device, and the "Arrow menu" has a "Copy Tempaltes..."\noption to copy them anywhere in the device tree.
[29-Sep-2009 16:22:44] <ke4qqq> chudler: doesn't that only apply to non-local copies (eg non-machine specific)
[29-Sep-2009 16:23:38] <tung> That's good to know. Thanks, chudler
[29-Sep-2009 16:25:06]\n<chudler> since that option cannot be browsed to, and it doesn't\nsay what it is copying, I suspect that it is not supported or might not\nwork as expected. Try it :-)
[29-Sep-2009 16:25:26] <chudler> I mean it cannot be browsed at the device level, only organizers AFAIK
[29-Sep-2009 16:28:17] <chudler> I tried it and it didnt work, sorry. I think I did it once some other way.
[29-Sep-2009 16:28:26] <cgibbons> bbl
[29-Sep-2009 16:39:33] <ke4qqq> mrayzenoss: is there a reason this isn't listed at c.z.o? http://blog.dastrup.com/?p=26
[29-Sep-2009 16:39:40] <ke4qqq> that zenpack?
[29-Sep-2009 16:39:45] <mrayzenoss> yes
[29-Sep-2009 16:39:56] <mrayzenoss> I asked him to send it to me to post, he hasn't
[29-Sep-2009 16:40:09] <mrayzenoss> I've asked several times over the last year
[29-Sep-2009 16:40:16] <ke4qqq> he used to hang out in irc - but I haven't seen him in well over a year
[29-Sep-2009 16:40:27] <mrayzenoss> plus, the new WMI Data Source one may be better
[29-Sep-2009 16:40:42] <mrayzenoss> since it doesn't leave passwords in logs
[29-Sep-2009 16:42:06]\n<mrayzenoss> take a look at these:\ndocs/DOC-3430 and\ndocs/DOC-3387
[29-Sep-2009\n16:42:23] <mrayzenoss> they're working together on a combined\nversion of the ZenPack, I should have it up soon
[29-Sep-2009 16:45:08] <ke4qqq> mrayzenoss: interesting - I honestly had been using the other so long I didn't know about those two
\n[29-Sep-2009 16:49:26] <mrayzenoss> ke4qqq: as soon as the WMI\nData Source came out, there was a flurry of rewriting the old\nwmic-based ZenPacks
[29-Sep-2009 16:50:28]\n<mrayzenoss> yay, new docs available:\ncommunity/documentation/official_documentation/training
[29-Sep-2009 17:04:15]\n<ke4qqq> mrayzenoss: can you change the zip/egg names on those\ntwo exchange zenpacks - despite being different they have exactly the\nsame name
[29-Sep-2009 17:04:52] <mrayzenoss> yeah, I'll drop one when I post the unified one
[29-Sep-2009 17:05:08] <mrayzenoss> it's kinda funny, they were developed independently
[29-Sep-2009 17:05:15] <mrayzenoss> submitted within a day
[29-Sep-2009 17:05:23] <mrayzenoss> both authors are Swedish
[29-Sep-2009 17:05:49]\n<ke4qqq> fascinating - I just got confused when trying to install\nthem because they both had the same name, but diff md5sums
[29-Sep-2009 17:08:52]\n<mrayzenoss> and on top of that... I just had to add a disclaimer\nthat all ZenPacks downloaded from the site need to be unzipped
[29-Sep-2009 17:09:01]\n<mrayzenoss> "Please note that you will need to unzip the\nattached ZenPacks because the new platform automatically zips them for\nsecurity concerns."
[29-Sep-2009 17:09:04] <mrayzenoss> stupid IE
[29-Sep-2009 17:09:17] <ke4qqq> yeah I saw that discussion on irc last week
[29-Sep-2009 17:10:40] <mrayzenoss> I tried to get them to turn it off
\n[29-Sep-2009 17:11:22] <mrayzenoss> but they don't support that\nsince IE automatically scans filetypes, which is a known security hole
[29-Sep-2009 17:11:36] <mrayzenoss> err... scans contents of uncompressed files
[29-Sep-2009 17:11:43] <mrayzenoss> even though eggs are already zips
\n[29-Sep-2009 17:13:48] <ke4qqq> hmmmm just installed\nwmidatasource and it doesn't show up in templates at any level - is\nthat expected?
[29-Sep-2009 17:14:18] <ke4qqq> nm I can't read
[29-Sep-2009 17:15:56] <mrayzenoss> yes
[29-Sep-2009 17:16:23] <ke4qqq> yes as in you are acking my illiteracy
\n[29-Sep-2009 17:17:14] * ke4qqq wishes I had watched the 'sleeping well\nat night' presentation in its entirety - ohhh well, perhaps I'll get\nthe audio or video
[29-Sep-2009 17:20:10]\n<rmatte> eugh, after tons of screwing around I finally managed to\nget login access to the switches that I'm commissioning except for 1
[29-Sep-2009 17:20:23] <rmatte> need to find a way to break in to the damn thing and set it up properly...
\n[29-Sep-2009 17:20:50] <rmatte> otherwise we'll be waiting over a\nweek for the client to reset the password. heck, it took them 3 weeks\njust to get back to us about setting up a vpn tunnel
[29-Sep-2009 17:20:57] <mrayzenoss> ke4qqq: I got an email from Greg at OLF saying they're cleaning them up now
[29-Sep-2009 17:22:06]\n<ke4qqq> yeah I got that same email, but I wouldn't be surprised\nto see it take them a month to get all of them done. It took SELF two\nmonths for audio and 3 for video - and we actually had a company doing\nit for us.
[29-Sep-2009 17:25:09] <rmatte> I love how I can hear rain on the skylights but my weather widget is showing it as being sunny out
[29-Sep-2009 17:25:30] <straterra> If the widget used ipv6..it'd be right
[29-Sep-2009 17:25:39] <rmatte> hehe
[29-Sep-2009 17:25:57] <straterra> I look forward to an ipv6 world
[29-Sep-2009 17:26:32] <rmatte> still seems a long ways away
\n[29-Sep-2009 17:27:34] <rmatte> quite honestly, I don't know how\ngreat it's going to be... "Now everyone can have their own IP, making\nit even easier for hackers, malware, and worms to make their way\ndirectly to your PC"
[29-Sep-2009 17:28:03] <straterra> Eh, in terms of that..It'll be the same
[29-Sep-2009 17:28:07] <cwj> it seemed a lot more appealing before every developer caught on to the fact that most people are behind nat
[29-Sep-2009 17:28:17] <straterra> Everyone had their own IP too..just behind NAT
[29-Sep-2009 17:28:35] <straterra> Just means I don't have to work about VPN tunnels and crap to access machines at home
[29-Sep-2009 17:28:59] <cwj> neither do crackers and phishers
[29-Sep-2009 17:29:07] <cwj> under ipv6
\n[29-Sep-2009 17:29:17] <straterra> I have everything at home and\non my server ipv6 enabled..except for my iPod Touch. Even my Windows\nMobile phone works with ipv6
[29-Sep-2009 17:29:38] <rmatte> I don't really see the point in using it at the moment
[29-Sep-2009 17:29:48] <straterra> future proof
[29-Sep-2009 17:29:54] <straterra> and..getting around NAT
[29-Sep-2009 17:29:54] <rmatte> once it starts becoming more heavily adopted in a business sense, then I'll switch over
[29-Sep-2009 17:44:51] <rmatte> eugh, this one device is driving me crazy
\n[29-Sep-2009 17:44:56] <__jd__> Hey all.. After upgrading from\n2.4.3 to 2.4.5, I started having an issue where 'clear' messages are\nbeing sent out immediately before new alerts. This is a case where an\nalert which would normally just count is now clearing and then\nun-clearing immediately afterwards each time. \ndown->up->down->up->down->up->down.. The thing is,\nthe service has never actually come back up. The clear notifications\nare bogus
[29-Sep-2009 17:45:55] <rmatte> check your zenactions log, see if you notice anything weird
\n[29-Sep-2009 17:48:54] <__jd__> some exceptions regarding\nsendXmpp, which probably broke with the upgrade.. Other than that, I\nsee it sending the clear and then the down again 1 minute later..
[29-Sep-2009 17:51:03] <__jd__> in this case, it's just a TCP service check done by zenoss.. Not zencommand or anything
[29-Sep-2009 17:51:04] <__jd__> HTTP
[29-Sep-2009 17:52:59] <__jd__> hmm.. maybe some other alert is clearing it?
[29-Sep-2009 17:53:33] <__jd__> it uses the dedupid, right?
[29-Sep-2009 17:56:24] <__jd__> I take it back, this is zencommand
[29-Sep-2009 18:00:10] <rmatte> it's possible that something else is clearing it, though not very probable
[29-Sep-2009 18:01:13] <__jd__> yea, it doesn't look like it
[29-Sep-2009 18:01:20] <__jd__> I'm running zencommand with debug
[29-Sep-2009 18:02:13] <rmatte> is the service actually down or up?
[29-Sep-2009 18:02:19] <__jd__> it's down
[29-Sep-2009 18:08:54] <__jd__> hm
[29-Sep-2009 18:09:12] <__jd__> it clears, but there's nothing run by zencommand to clear it
[29-Sep-2009 18:09:46]\n<__jd__> in fact, the template is setup to only run the command\nevery 5 minutes, yet the clear comes 1 minute before the re-alert
[29-Sep-2009 18:10:11] <__jd__> it seems like from zencommand's perspective, the service indeed never comes back up
\n[29-Sep-2009 18:10:17] <rmatte> well, I'm out of here shortly but\nyou can try this... seems to help with wacky event stuff like that\nsometimes...
[29-Sep-2009 18:10:19] <rmatte> stop zenoss
[29-Sep-2009 18:10:29] <rmatte> copy $ZENHOME/var/Data.fs to Data.fs.bak
[29-Sep-2009 18:10:32] <rmatte> start zenoss
[29-Sep-2009 18:10:35] <rmatte> go in to zendmd
[29-Sep-2009 18:10:37] <rmatte> and type:
[29-Sep-2009 18:10:39] <rmatte> reindex()
[29-Sep-2009 18:10:41] <rmatte> commit()
[29-Sep-2009 18:10:51] <rmatte> see if you still have the issue after doing that
[29-Sep-2009 18:11:10] <__jd__> will do.. thanks
\n[29-Sep-2009 18:11:46] <rmatte> I have people backup the Data.fs\nsince I've seen someone's zope database get fried by a reindex since\nthey had a major problem in it and the reindex made it even worse, so I\nhave people back it up before doing reindexes now
[29-Sep-2009 18:11:53] <rmatte> I've never had any issues
[29-Sep-2009 18:12:17] <__jd__> k.. don e:)
[29-Sep-2009 18:12:23] <rmatte> if it runs through fine you can delete the .bak afterwards
[29-Sep-2009 18:15:58] <rmatte> anyways, I'm out of here
[29-Sep-2009 18:16:04] <rmatte> I'll be around tomorrow
[29-Sep-2009 18:50:45] <davetoo1> I wish zenoss would leave my snmpd.conf alone
[29-Sep-2009 18:50:46] <davetoo1> grumble
[29-Sep-2009 19:36:29] <cryptographrix> y0 - anyone have probs with httpd process monitoring?
[29-Sep-2009 19:36:37] <__jd__> yes
[29-Sep-2009 19:36:42] <purevol> sure
[29-Sep-2009 19:37:04] <__jd__> or wait.. process? not sure, haven't tried
\n[29-Sep-2009 19:37:33] <cryptographrix> looks like process\nmonitoring in general - basically no matter which processes I'm\nmonitoring, whichever one is first in the list is shown as being down
[29-Sep-2009 19:38:11] <cryptographrix> any ideas?
[29-Sep-2009 19:38:25] <purevol> so you have a device setup, and you modeled it?
[29-Sep-2009 19:38:31] <cryptographrix> yes
[29-Sep-2009 19:38:59] <purevol> windows or linux host?
[29-Sep-2009 19:39:05] <cryptographrix> linux host
\n[29-Sep-2009 19:39:39] <cryptographrix> just testing zenoss right\nnow, getting it to monitor another host (one of our QA boxes)\nsuccessfully before we roll it out to more
[29-Sep-2009 19:41:09] <purevol> if you goto services is httpd in there?
[29-Sep-2009 19:41:52] <cryptographrix> no
\n[29-Sep-2009 19:43:12] <purevol> im not in front of my zenoss\ninstall - but check this out\ndocs/DOC-2446#Whydoesnthttpdshowupinmylistofservices
[29-Sep-2009 19:43:42] <cryptographrix> should it work to create httpd under processes and regex? kk willdo
[29-Sep-2009 19:49:54] <purevol> you might have to add http to the collector plugins for the device
[29-Sep-2009 19:50:34] <purevol> so goto Device List, then goto the menu and select More -->Collector Plugins
[29-Sep-2009 19:50:56] <purevol> docs/DOC-3093#d4e1738
[29-Sep-2009 19:51:54] <cryptographrix> coolcool will check
\n[29-Sep-2009 19:51:59] <purevol> on the right you will see an \n"Add Fields" click that, and i think Linux.httpd should be in there
[29-Sep-2009 19:52:32] <purevol> i think
[29-Sep-2009 19:54:01] <cryptographrix> k I'll look for it - plugin's built into the core, it looks like
[29-Sep-2009 19:54:32] <cryptographrix> (not part of a non-core zenpack)
\n[29-Sep-2009 20:15:12] <cryptographrix> yet another strange\nerror: Error reading value for "eth0" on XXXXXXXX (oid\n.1.3.6.1.2.1.31.1.1.1.7.2 is bad)
[29-Sep-2009 20:15:31] <cryptographrix> for eventClass /Perf/Snmp
[29-Sep-2009 20:15:34] <purevol> you need to make sure you have MIBs installed
[29-Sep-2009 20:16:12] <cryptographrix> MIBs? and installed on zenoss host or monitored server?
[29-Sep-2009 20:16:16] <purevol> did you get you httpd monitored?
[29-Sep-2009 20:16:39] <cryptographrix> nah, still working on it - figuring stuff out slowly tho
\n[29-Sep-2009 20:18:13] <purevol> it takes a little while, but\neverything is in the admin guide, just walk through it and you will be\ngood, that oid thing, probably just needs a mib installed
[29-Sep-2009 20:18:31] <cryptographrix> kk
[29-Sep-2009 21:13:04] <davetoo> why doesn't my process monitor go away?
[29-Sep-2009 21:13:54] <purevol> what do you mean davetoo?
[29-Sep-2009 21:14:22] <davetoo> well, I had a bunch of process monitors devined and globally enabled
[29-Sep-2009 21:14:35] <davetoo> having trouble with them, so I disabled them all
[29-Sep-2009 21:14:52] <davetoo> so each device they were found on, now shows that they aren't monitored
[29-Sep-2009 21:15:03] <davetoo> and I even restarted zenprocess on all the collectors,
[29-Sep-2009 21:15:10] <davetoo> but I'm still getting events
[29-Sep-2009 21:15:23] <purevol> interesting
[29-Sep-2009 21:16:08] <purevol> maybe you need to restart the daemons
[29-Sep-2009 21:16:13] <davetoo> I did
[29-Sep-2009 21:16:23] <davetoo> going to try something next
[29-Sep-2009 21:16:48] <purevol> did you remodel the device?
[29-Sep-2009 21:16:53] <davetoo> each device still has the processes listed, but they are no longer listed as "monitored"
[29-Sep-2009 21:17:07] <davetoo> welll
[29-Sep-2009 21:17:26] <davetoo> I thought I had, but apparently not
[29-Sep-2009 21:18:19] <davetoo> I'm having a hell of a time monitoring the zenoss daemon processes themselves
[29-Sep-2009 21:18:36] <purevol> VM appliance?
\n[29-Sep-2009 21:18:52] <davetoo> it's also unfortunate that it\ndepends on what's running at the time you model the device, what\nprocesses you end up with in the model
[29-Sep-2009 21:19:01] <davetoo> Enterprise, physical remote collectors
[29-Sep-2009 21:20:08] <purevol> hrm.... im a bit rusty
[29-Sep-2009 21:20:44] <davetoo> The boxes I'm monitoring are the remote collectors themselves
[29-Sep-2009 21:21:00] <davetoo> trying to make sure that all the zenoss daemons are running
[29-Sep-2009 21:21:28] <davetoo> but the way they are started, you have to have "ignore parameters" set to False
[29-Sep-2009 21:23:27] <purevol> can you create a command to login and check the processes
[29-Sep-2009 21:24:37] <davetoo> I may decide to do that instead, yes
[29-Sep-2009 21:27:57] <davetoo> Maybe it requires a remodel to push the config out to the daemons
[29-Sep-2009 21:34:02] <davetoo> *sigh*
[29-Sep-2009 21:34:05] <davetoo> so I did remodel,
[29-Sep-2009 21:34:27] <davetoo> and the processes are back;
[29-Sep-2009 21:34:43] <davetoo> they are set not to monitor
[29-Sep-2009 21:34:49] <davetoo> but
[29-Sep-2009 21:34:55] <davetoo> I am getting events
[29-Sep-2009 21:35:03] <davetoo> baffled
[29-Sep-2009 21:35:47] <purevol> so you have half working
[29-Sep-2009 21:38:49] <davetoo> I just don't get it
[29-Sep-2009 21:41:02] <davetoo> I really don't want to erase all of the process definitions I spent so much time creating
[29-Sep-2009 21:42:28] <purevol> maybe someone tomorrow will have a better answer for you, sorry man
[29-Sep-2009 21:57:22] Traveler is now known as Guest60426
[29-Sep-2009 21:58:12] <Guest60426> Hi all - anyone provide assistance with graph rendering issue?
[29-Sep-2009 21:58:15] <Guest60426> ERROR RenderServer Failed to generate a graph
[29-Sep-2009 21:59:19] <Guest60426> happening on 2 different devices using 2 different perf templates
[29-Sep-2009 21:59:42] <purevol> either something is not setup right, or you have to wait for data to populate
[29-Sep-2009 21:59:59] <Guest60426> purevol -thanks for the reply
[29-Sep-2009 22:00:37]\n<Guest60426> these aren't new templates that i just applied - i\nhave old data already graphed - and can access it when i look at the\nweekly view
[29-Sep-2009 22:01:38] <purevol> did you make any changes?
[29-Sep-2009 22:02:00] <Guest60426> its all wmi - and as for testing out the templates i can run tests on them and they return the right data
[29-Sep-2009 22:02:42] <purevol> i dont use any WMI, but if you might want to check your colelctor plugins
[29-Sep-2009 22:02:50] <Guest60426> i recently upgraded to 2.4.4 and went up to 2.4.5 tonight
[29-Sep-2009 22:03:27]\n<purevol> weird, maybe something changed that i dont know about,\nthe best bet is to drop in tomorrow when the smart guys are around
[29-Sep-2009 22:03:39] <Guest60426> i just recently am switching over from snmp to wmi since the env is strictly windows
[29-Sep-2009 22:03:54] <purevol> maybe you try to remodel the device?
[29-Sep-2009 22:04:01] <Guest60426> yeah its got me stumped
[29-Sep-2009 22:04:37]\n<Guest60426> because it was graphing fine until something changed\n(wish i knew if it was the upgrade or zenpack additions etc)
[29-Sep-2009 22:04:52] <Guest60426> i'll try that
[29-Sep-2009 22:05:36] <Guest60426> other things i did recently were deleting, copying and binding new templates to the windows class
[29-Sep-2009 22:09:07] <Guest60426> no go with the remodel
[29-Sep-2009 22:09:25] <Guest60426> i think i may try to restore an old 2.4x db and see what happens
[29-Sep-2009 22:09:49]\n<purevol> thats a bit of a drag, did you check your collector\nplugins for that device, and maybe double check your zProperties, to\nensure they are setup with the right credentials?
\n[29-Sep-2009 22:10:27] <Guest60426> credentials are good - i have\nother perf templates on the server that are using the same ones
\n[29-Sep-2009 22:11:26] <Guest60426> i recently added all the HP\next device info snmp collector plugins so i'll try that too
[29-Sep-2009 22:15:15] <Guest60426> purevol: thanks for the suggestions- if i have time i'll try back tomorrow
[29-Sep-2009 22:15:33] <purevol> sorry im not more useful! :-)
[29-Sep-2009 22:16:10] <purevol> we have some windows at work, but i dont manage any of that, but maybe i can try it
[30-Sep-2009 00:00:46] [disconnected at Wed Sep 30 00:00:46 2009]
[30-Sep-2009 00:00:46] [connected at Wed Sep 30 00:00:46 2009]
[30-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
[30-Sep-2009 00:44:03]\n<shentonfreude> so maybe the more effective way is to put -- say\n-- $5K from Koansys credit card into a Koansys paypal account, and pay\nfrom that. But I will not be comfortable with them linking to my debit\ncard nor bank account.
[30-Sep-2009 00:44:18] <shentonfreude> Then if we send to UK from the Koansys PayPal balance it's only $5 fee.
[30-Sep-2009 00:45:01] <shentonfreude> yeah, on that page where all you do is fill out their email addr and amount.
[30-Sep-2009 00:45:08] <shentonfreude> $39 for $1000
[30-Sep-2009 00:46:21] <shentonfreude> sorry, damn, wrong window. hate that.
[30-Sep-2009 02:42:48] <MorkBork> hi im having a bit of trouble finding out if what i want to do is possible
[30-Sep-2009 02:42:55] <MorkBork> i have a small zenoss core install and im watching a few game servers
[30-Sep-2009 02:43:09] <MorkBork> i wanted to source some of the graphs on a status page
[30-Sep-2009 02:43:23] <MorkBork> for the players to see.. i.e. bandwidth, packets per second, etc
[30-Sep-2009 02:43:34] <MorkBork> is there an easy way to do this through acl/permissions
[30-Sep-2009 02:43:47] <MorkBork> like make a perf graph anonymously available
[30-Sep-2009 02:45:11] <MorkBork> or can i just write a cron job or something
[30-Sep-2009 02:45:32] <MorkBork> its not like life or death important, just wondered what the common way to do that is
[30-Sep-2009 02:48:50] <MorkBork> i found the zenreport command
[30-Sep-2009 02:49:13] <MorkBork> but thats a little overkill
\n[30-Sep-2009 02:58:44] <Troubadix09> MorkBork: I remember there\nwas a Document about "Setup a Kiosk-Zenoss" or so in the forum
[30-Sep-2009 02:58:59] <MorkBork> yea but it seems like everything i google now is moved
[30-Sep-2009 02:59:01] <MorkBork> ;(
[30-Sep-2009 02:59:12] <MorkBork> thank you google cache!
\n[30-Sep-2009 02:59:42] <Troubadix09> MorkBork: And there is a\nDocument on the website of jane curry, how to configure user rights in\nzenoss core
[30-Sep-2009 03:00:21] <Troubadix09> MorkBork: dedicated user rights only in zenoss enterprise possible
[30-Sep-2009 03:00:24] <MorkBork> ah
[30-Sep-2009 03:00:29] <MorkBork> yea its just me with an account
[30-Sep-2009 03:00:43] <MorkBork> ive been using zenoss forever but never in any sort of advanced capacity
[30-Sep-2009 03:00:55] <MorkBork> mainly just to keep track of when the gameserver dies so i can go restart it
[30-Sep-2009 03:01:38] <MorkBork> oh so zenoss enterprise has a more involved acl system?
[30-Sep-2009 03:01:44] <MorkBork> i always wondereed what the differences were
[30-Sep-2009 03:02:02] <MorkBork> i always thought the acl system in core was kind of weird
[30-Sep-2009 03:02:03] <MorkBork> makes sense
[30-Sep-2009 03:06:31]\n<Troubadix09> MorkBork: Enterprise have mor different, like\nclustering, HA-configuration, more Zenoss-instances etc.
[30-Sep-2009 03:09:20] <MorkBork> ahh so probably not much use for 1 guy and 4 servers
[30-Sep-2009 03:11:24] <MorkBork> i tried lots of different monitoring things and i keep coming back to zenoss
[30-Sep-2009 03:11:28] <MorkBork> although i love cacti graphs
[30-Sep-2009 03:32:51] <rhettardo> i use both cacti and zenoss
[30-Sep-2009 03:40:27] <MorkBork> i got sick of working with both of them
[30-Sep-2009 03:40:30] <MorkBork> so i just use zenoss
[30-Sep-2009 03:40:40] <MorkBork> the graphs arent as pretty but i dont stare at them all day anyway
[30-Sep-2009 03:40:51] <MorkBork> the only thing i miss was how easy it was to export the graphs
[30-Sep-2009 03:41:03] <MorkBork> maybe someone better than me with python will make a zenpack to export stuff easily
[30-Sep-2009 03:42:51] <rhettardo> ah. pretty much all of my cacti stuff is custom so thats why i keep it around
[30-Sep-2009 04:29:48] <MorkBork> well
[30-Sep-2009 04:29:52] <MorkBork> the rrd files are easy to export
[30-Sep-2009 04:30:06] <MorkBork> so i just figure ill use that and rrdtool and remake the graphs myself
[30-Sep-2009 07:35:28] <Quadro> bah
[30-Sep-2009 07:35:55] <Quadro> building beta 3 or svn head fails
\n[30-Sep-2009 08:31:41] <Dieterbe_> im getting 'process not\nrunning: more' events on a server, but i don't care that that process\ndoesn't run. i verified that 'more' doesn't match any of my process\nmonitoring regexes, but how can i make zenoss stop checking this\nprocess on that server?
[30-Sep-2009 08:39:53]\n<ckrough> Dieterbe_: what version are you running? My boxes have\nalso started monitoring processes I dont care about
\n[30-Sep-2009 08:40:17] <ckrough> I was under the impression it\nwould only pick up on processes that were defined in the Processes\nclass, but I guess I'm wrong.
[30-Sep-2009 08:44:42]\n<Dieterbe_> aha, so i'm not delusional.. it's 2.4.5 , it's\nlooking for a process called 'more' (the pager app must have run at\nsome point and zenoss picked it up for some reason) . i'm having the\nevents since 2009/09/24 17:13:18.000
[30-Sep-2009 08:48:17] <ckrough> yeah, mine has picked up gcc, ruby, etc...
[30-Sep-2009 09:08:14] <Dieterbe_> hah. and you don't have a regex like '.*', '.' or 'c' or something that would match stuff like that?
[30-Sep-2009 09:11:09] <ckrough> nope
[30-Sep-2009 09:11:26] <ckrough> mine are very specific. but more importantly it just started doing this out of the blue.
[30-Sep-2009 09:11:34] <ckrough> Im on 2.4.2 enterprise though.
[30-Sep-2009 09:16:35] <davetoo> There might be some bad regex syntax in processes from some of the enterprise zenpacks?
[30-Sep-2009 09:17:03] <davetoo> I found that the httpd monitor was looking in the parameters list
[30-Sep-2009 09:17:22] <davetoo> and some of my users systems had 'httpd' as an argument in some testbed command
[30-Sep-2009 09:17:39] <davetoo> and during the nightly remodeling those were added to the os processes list
\n[30-Sep-2009 09:18:52] <davetoo> We need a better way to say\n"this is the set of processes that should be running on this class of\nsystems, regardless of whether you find it running during modeling or\nnot"
[30-Sep-2009 09:20:08] <exarkun> I tried to\ninstall my first zenpack. It failed like this: ERROR: zenpack command\nfailed. Reason: [Errno 20] Not a directory:\n'/usr/local/zenoss/zenoss/Products/ZenPacks.SNMP.TCPMIB-1.0-py2.4.egg/skins'
[30-Sep-2009 09:21:05]\n<exarkun> It's true that that isn't a directory. \n/usr/local/zenoss/zenoss/Products/ZenPacks.SNMP.TCPMIB-1.0-py2.4.egg\nisn't either, it's a regular file.
[30-Sep-2009 09:21:20] <davetoo> no,
[30-Sep-2009 09:21:26] <davetoo> the .egg will be a directory
[30-Sep-2009 09:21:36] <exarkun> When? It's not now.
[30-Sep-2009 09:21:48] <davetoo> How did you install it?
[30-Sep-2009 09:22:08] <exarkun> zenpack --install .../ZenPacks.SNMP.TCPMIB-1.0-py2.4.egg.zip
[30-Sep-2009 09:22:17] <davetoo> Zenpacks don't live under .../Products
[30-Sep-2009 09:23:00] <exarkun> zenpack seems to have copied this one there.
\n[30-Sep-2009 09:23:32] <chudler> unzip the egg file first. The\nnew zenoss community website zipped all the eggs, but they are not\nsupposed to be
[30-Sep-2009 09:23:37] <ckrough> I unzip mine first
[30-Sep-2009 09:23:41] <davetoo> They are supposed to be under $ZENHOME/ZenPacks
[30-Sep-2009 09:23:41] <ckrough> yeah that
[30-Sep-2009 09:23:44] <exarkun> Ah.
[30-Sep-2009 09:24:02] <exarkun> So should I just delete this .egg from the Products directory?
[30-Sep-2009 09:24:22] <davetoo> Check what zenoss thinks, first
[30-Sep-2009 09:24:29] <davetoo> does Settings->Zenpacks show it?
[30-Sep-2009 09:24:52] <exarkun> I probably have to restart zenoss before I check?
[30-Sep-2009 09:25:03] <purevol> you are back davetoo
[30-Sep-2009 09:25:19] <davetoo> I'd just be worried about having the zodb out of sync with the actual filesystem
[30-Sep-2009 09:25:23] <davetoo> yes, too early,
[30-Sep-2009 09:25:31] <davetoo> meeting about zenoss project progress
[30-Sep-2009 09:25:42] <purevol> i hear you, i have to go and teach a class in a half hour, needless to say im thrilled
[30-Sep-2009 09:26:08] <purevol> speaking of which..i guess i should go and prepare!
[30-Sep-2009 09:26:14] <davetoo> where are you, physically?
[30-Sep-2009 09:26:49] <davetoo> I'm in California
[30-Sep-2009 09:29:17] <otakup0pe> hmm i don't seem to be able to add OS products/manufacturers
[30-Sep-2009 09:29:20] <otakup0pe> i must be doing it wrong
[30-Sep-2009 09:29:37] <otakup0pe> purevol: what's the class subject matter ? i kinda miss teaching.
\n[30-Sep-2009 09:38:53] <Dieterbe_> davetoo> We need a better\nway to say "this is the set of processes that should be running on this\nclass of systems, regardless of whether you find it running during\nmodeling or not" <-- i definitely agree with that
\n[30-Sep-2009 09:41:28] <Dieterbe_> 16:16:36 < davetoo>\nThere might be some bad regex syntax in processes from some of the\nenterprise zenpacks?
[30-Sep-2009 09:41:31] <Dieterbe_> 16:17:03 < davetoo> I found that the httpd monitor was looking in the parameters list
[30-Sep-2009 09:41:42] <Dieterbe_> aha that may be it. i use zenoss core but i use the httpmonitor zenpack
[30-Sep-2009 10:21:56] <rmatte> good morning folks
[30-Sep-2009 10:22:33] <eidolon> mornin rmatte
[30-Sep-2009 10:22:36] <eidolon> how goes the fight?
[30-Sep-2009 10:22:41] <rmatte> epic
[30-Sep-2009 10:22:44] <rmatte>
[30-Sep-2009 10:23:04] <rmatte> just got out of a meeting, been assigned my new project for the week
[30-Sep-2009 10:23:15] <rmatte> automating report dumps using scripts I wrote
\n[30-Sep-2009 10:23:28] <eidolon> so here's a setup question -\nmore 'how do folks do this'. we have 2 levels of alerting we deal\nwith. warnings and error/crits. error/crits go to pagers. pagers are\n12345@sprint.com or the like. do people make 'user' and 'user-pager'\nentries in Settings/Users and set up groups for each?
[30-Sep-2009 10:23:33] <eidolon> we don't have an SNPP gateway.
\n[30-Sep-2009 10:24:17] <rmatte> ok, so... you only want certain\nalerts to be sent out to all pager, or certain alerts to be sent to\ncertain pagers?
[30-Sep-2009 10:26:14] <eidolon>\ni want all crits and errors to go to a list of users' pagers. (i'm\ngoing to do that with a group, in the group i'll have 'dave-pager'\n'bob-pager' 'phil-pager'). warnings will go to email (dave, bob, phil)
\n[30-Sep-2009 10:26:52] <Brad_K> otakup0pe: see\nblogs/ryanmatte/2009/09/24/specifying-hardware-and-software-product-info-in-245-until-bug-is-fixed
[30-Sep-2009 10:26:53] <rmatte> ok
\n[30-Sep-2009 10:27:20] <rmatte> Brad_K: well, click on the link\nat the top of that one to be sent to the guide which explains how to\nactually patch it
[30-Sep-2009 10:27:36] <rmatte> eidolon: so what's your question then?
[30-Sep-2009 10:28:06] <rmatte> Brad_K: oh, he was asking how to actually add products hehe
[30-Sep-2009 10:28:31] <eidolon> well, 'is that the way to do it'? \nthat seems to be a good solution, but i could see some zenoss-way of\ngoing "oh no, don't do that, click here, here, here, and here, and\nyou'll see a better tool for it!"
[30-Sep-2009\n10:29:47] <rmatte> eidolon: nope, that sounds about right, the\nonly other way that I can think of to do it would be to write a script\nthat sends the SMS messages and then set it up to be executed by Zenoss\nwhen events come in (Event Manager -> Commands)
[30-Sep-2009 10:29:58] <eidolon> glrk
[30-Sep-2009 10:30:01] <eidolon> don't think i like that one.
[30-Sep-2009 10:30:03] <eidolon> okay,thanks
[30-Sep-2009 10:30:03] <rmatte> eidolon: make sure that you add a filter to the alerting rule
[30-Sep-2009 10:30:15] <rmatte> to only alert on events higher than info
[30-Sep-2009 10:30:32] <eidolon> yeah. i got it
[30-Sep-2009 10:30:33] <eidolon> thanks
[30-Sep-2009 10:31:58] <rmatte> np
\n[30-Sep-2009 10:40:26] <rmatte> hmmmm, keeps getting cycle\nexceeded on zenprocess on one of my boxes, might have to bump parallel\nup to 3
[30-Sep-2009 10:42:27] <rmatte> hmmm, zenprocess doesn't support the parallel option, weird
[30-Sep-2009 10:42:51] <kmtsun> hello everyone
[30-Sep-2009 10:42:56] <rmatte> hello
[30-Sep-2009 10:45:21] <rmatte> hmmm, cycle times in general are extremely elevated, started on tuesday
\n[30-Sep-2009 10:52:57] <chudler> cycle times could go up if just\nsome device is responding slowly. Its one thing that sucks about a\npolling architecture.
[30-Sep-2009 10:53:34] <rmatte> actually, I think it's related to the link
\n[30-Sep-2009 10:53:47] <rmatte> we monitor through tunnels, looks\nlike the same thing is happening on our other Zenoss servers too
[30-Sep-2009 10:53:48] <cgibbons> hurm
[30-Sep-2009 10:53:56] <chudler> yeah, I meant to mention that many other reasons it could go up :-)
[30-Sep-2009 10:53:57] <rmatte> though it's less noticeable on them since they have a lot less devices
\n[30-Sep-2009 10:57:22] <chudler> someday I really hope to make a\nplugin that uses check_mk http://mathias-kettner.de/check_mk
[30-Sep-2009 10:57:26] <chudler> Its probably just a dream, though
[30-Sep-2009 11:07:01] <ke4qqq> anyone have a list of decent mobile phones that work with usb and linux to send snpp/sms messages via linux?
[30-Sep-2009 11:08:17] <chudler> http://www.sierrawireless.com/product/usbmodems.aspx <-- those work good
\n[30-Sep-2009 11:11:25] <chudler> We use the Compass 885. They\nprovide a kernel module and userland tools which are very easy to use.
\n[30-Sep-2009 11:17:51] <rmatte> chudler: if check_mk is already a\nnagios plugin it should integrate fine with Zenoss (with a little\nmassaging of course)
[30-Sep-2009 11:18:25] <otakup0pe> Brad_K: thx
[30-Sep-2009 11:18:27] <rmatte> oh, I see what it does
[30-Sep-2009 11:18:29] <rmatte> interesting
[30-Sep-2009 11:27:07] <ckrough> most providers have an email address for sms, you could just send them as an email to that
\n[30-Sep-2009 11:28:54] <chudler> we did that, but it didnt work\nwhen the network was down, plus they throttled us hard. An truly\nrhobust way to do it is dialing out on a landline, but we wanted to\nreceive SMS on zenoss also.
[30-Sep-2009 11:29:14] <ckrough> yeah, AT&T is not real happy
[30-Sep-2009 11:29:19] <ckrough> we've worked some things out with them
[30-Sep-2009 11:29:34] <ckrough> pushing close to 2000 sms messages per month per user for around 20 users
[30-Sep-2009 11:29:37] <rmatte> AT&T are horrible
[30-Sep-2009 11:29:54] <ckrough> its been successful though
[30-Sep-2009 11:30:35] <chudler> for Serious Enterprises, they rent 1U rack-mount SMS gateways. not cheap ;-)
[30-Sep-2009 11:35:09] <ckrough> hey now, Im a serious enterprise
[30-Sep-2009 11:35:28] <rmatte> hehe
[30-Sep-2009 11:35:55] <rmatte> If you're a serious enterprise, why aren't you paying us to monitor your stuff?
[30-Sep-2009 11:36:23] <ckrough> you cant handle my stuff
[30-Sep-2009 11:36:35] <rmatte> probably could given enough time
[30-Sep-2009 11:36:40] <ckrough>
[30-Sep-2009 11:36:59] <rmatte> I've gotten Zenoss to do some pretty fun stuff
\n[30-Sep-2009 11:37:12] <exarkun> What other ZenPacks does "Http\nMonitor" depend on? The community page says it has no dependencies,\nbut when I try to use it, I get an error that seems to indicate there\nis a missing dependency.
[30-Sep-2009 11:37:37] <rmatte> exarkun: none
[30-Sep-2009 11:37:47] <rmatte> paste the error
[30-Sep-2009 11:38:01] <rmatte> (if it's significantly long, use pastebin.com please)
[30-Sep-2009 11:38:08] <exarkun> http://pastebin.com/m3abaf380
[30-Sep-2009 11:39:03] <rmatte> you tried restarting Zenoss as I had suggested yesterday?
[30-Sep-2009 11:40:07]\n<shentonfreude> newbie question: can Zenos be used for "asset\nmanagement" of IP-less devices like video projectors? If I try an add a\ndevice without IP the modelling fails since it can't be pinged. Or\nperhaps there's a zenpack to do this common task? thanks
[30-Sep-2009 11:40:16] <exarkun> No, I probably forgot to restart it after I installed that zenpack.
[30-Sep-2009 11:42:20] <rmatte> exarkun: that's your issue
[30-Sep-2009 11:42:48] <exarkun> waiting for it to restart now
[30-Sep-2009 11:42:52] <rmatte> shentonfreude: this is the way I'd do it...
[30-Sep-2009 11:43:06] <exarkun> is it usual for the "zopectl" step to take several minutes?
[30-Sep-2009 11:43:20] <rmatte> if it has been up for a while it could take like 30 seconds or so
[30-Sep-2009 11:43:23] <exarkun> (also, that's a pretty bad failure mode for forgetting to restart how about a "You forgot to restart" page)
[30-Sep-2009 11:43:24] <rmatte> shouldn't take much longer
[30-Sep-2009 11:43:34] <exarkun> It's been up for an hour or so.
[30-Sep-2009 11:43:50] <exarkun> The whole restart process is like a 5 minute ordeal.
\n[30-Sep-2009 11:43:55] <rmatte> exarkun: the whole restarting\nafter installing a ZenPack is something that we pretty much drill in to\npeople's heads around here
[30-Sep-2009 11:44:12] <rmatte> exarkun: is it running on a VM?
\n[30-Sep-2009 11:44:28] <exarkun> rmatte: Maybe if the UI\npresented a more understandable error you wouldn't have to do that
[30-Sep-2009 11:44:31] <exarkun> rmatte: yes
\n[30-Sep-2009 11:44:45] <rmatte> exarkun: If it's taking that long\nyou're running it on something with not enough resources, the only\nslower Zenoss box that I have is my lab box, and it doesn't have enough\nresources allocated to it
[30-Sep-2009 11:44:58] <rmatte> and it's not even as slow as you're describing
\n[30-Sep-2009 11:45:33] <rmatte> exarkun: the UI is just telling\nyou that it knows the zenpack is broken, as for actually knowing why\nit's broken, that would take quite a bit of code and it's really\nunnecessary in my opinion
[30-Sep-2009 11:45:41] <rmatte> shentonfreude: ok, sorry, getting back to your question...
[30-Sep-2009 11:45:48] <rmatte> shentonfreude: In Zenoss, go to Settings
[30-Sep-2009 11:45:57] <rmatte> you'll see a box that says State Conversions
\n[30-Sep-2009 11:46:11] <rmatte> you'll want to add a new state to\nrepresent these devices, not sure what you want to call it
\n[30-Sep-2009 11:46:21] <exarkun> rmatte: An AttributeError in the\nerror reporting process isn't the most transparent report about a\nzenpack being broken.
[30-Sep-2009 11:46:21] <rmatte> I've added states for Standby, Spare, and Stacked
\n[30-Sep-2009 11:46:46] <rmatte> so let's say you want to add one\ncalled "Stacked", you need to add it with a value below Decommissioned,\nso that Zenoss doesn't try to monitor it
[30-Sep-2009 11:46:55] <rmatte> so you'd add: Stacked:-2
[30-Sep-2009 11:46:58] <rmatte> or whatever
\n[30-Sep-2009 11:47:34] <rmatte> after you've done that, just add\nthe devices with bogus IPs and then rename them to whatever you wish,\nmake sure that when you're adding them that they are in the production\nstate that you created
[30-Sep-2009 11:48:23]\n<rmatte> exarkun: I agree that the ZenPack system could be\nbetter, but I'd rather see them focus on other things than that at the\nmoment
[30-Sep-2009 11:48:39] <rmatte> exarkun:\nit's not something that's pressing enough for them to deal with in the\nimmediate future
[30-Sep-2009 11:49:01] <ckrough> is there a list of major goals? (not trac)
[30-Sep-2009 11:49:03]\n<rmatte> if you go to dev.zenoss.com/trac you can take a look at\nthe massive number of tickets that they are currently addressing or\nhave backlogged and you'll understand why I say that
[30-Sep-2009 11:49:10] * exarkun shrugs
[30-Sep-2009 11:49:17] <ckrough> or maybe trac has last list and I didnt look well
[30-Sep-2009 11:49:20] <exarkun> I'm not trying to dictate priorities
\n[30-Sep-2009 11:49:34] <rmatte> ckrough: I've personally seen a\nplan that they have layed out during a Zenoss Masters webex session
[30-Sep-2009 11:49:37] <exarkun> Just saying buggy error reporting in the ui isn't cool.
\n[30-Sep-2009 11:50:24] <rmatte> basically, King Crab is going to\nbe where they try to resolve as many annoyances as possible, then the\nfocus for stone crab is going to be the new UI, afterwards so of the\nfocusses I've seen are reporting, and performance improvements.
\n[30-Sep-2009 11:51:11] <rmatte> exarkun: I agree, but now that\nyou know to restart Zenoss if you've installed a ZenPack and it's\ncomplaining, you'll do it next time
[30-Sep-2009 11:51:27] <rmatte> You don't have to do it for every ZenPack, just certain ones
\n[30-Sep-2009 11:52:23] <rmatte> ckrough: I wish I had saved that\nwebex session, I could have shown you, just recalling off the top of my\nhead. If you ask Matt I'm sure he could dig it up and send it to you.
[30-Sep-2009 11:55:33] <ckrough> not important
\n[30-Sep-2009 11:57:26] <rmatte> hmmm lovely, looks like one of\nthe ethernet adapters on our main blade is failing, that would account\nfor the higher config times
[30-Sep-2009 12:02:11] <rmatte> hmmm, but there are only a few errors for it and I'm not sure when they are from
\n[30-Sep-2009 12:23:43] <__jd__> rmatte, apparently zenpatch 15003\nfixes the issue I described yesterday with the spurious auto-clearing
[30-Sep-2009 12:25:49] <shentonfreude> rmatte: thanks for the pointers
[30-Sep-2009 12:26:09]\n<rmatte> ummm, __jd__, you said that you were using 2.4.5\nyesterday, if you had that problem you can't have been using 2.4.5
[30-Sep-2009 12:26:14] <rmatte> were you at 2.4.4 or something?
[30-Sep-2009 12:26:37] <rmatte> actually, I think that one got patched in 2.4.4
[30-Sep-2009 12:27:34] <rmatte> hmmm, patch was created 2 weeks ago eh?
[30-Sep-2009 12:27:48] <rmatte> maybe they did just fix that recently, thought that got patched ages ago
[30-Sep-2009 12:27:55] <rmatte> they really need to release 2.4.6
[30-Sep-2009 12:27:59] <rmatte> 2.4.5 is so broken
[30-Sep-2009 12:35:51] <ckrough> did 2.4.5 release?
[30-Sep-2009 12:35:57] <ckrough> I thought it was in beta3
[30-Sep-2009 12:37:06] <rmatte> that's 2.5
[30-Sep-2009 12:37:07] <rmatte> lol
[30-Sep-2009 12:37:13] <rmatte> 2.4.5 was the latest official release
\n[30-Sep-2009 12:37:30] <rmatte> and it still has some pretty\nobvious bugs in it, they need to patch 2.4.5 with the latest fixes and\nrelease it as 2.4.6
[30-Sep-2009 12:37:43] <rmatte> though they may not even bother since King Crab is supposed to release in like 3 weeks
[30-Sep-2009 12:38:03] <rmatte> hopefully they don't rush the release just for the sake of getting it out though
[30-Sep-2009 12:42:45] <__jd__> they had a patch for 2.5 earlier I guess, but then backported it to 2.4.5
\n[30-Sep-2009 12:42:59] <__jd__> not sure how long that took..\nactually not even sure if it fixes the bug yet.. but I'm hopeful
[30-Sep-2009 12:44:13] <rmatte> well, some patches to 2.5 will cleanly apply to 2.4.5
[30-Sep-2009 12:44:15] <rmatte> depends on the patch
[30-Sep-2009 12:45:23] <__jd__> yea
[30-Sep-2009 12:45:28] <__jd__> zenpatch is a clever hack
[30-Sep-2009 12:45:47] <__jd__> something you can do when you're not distributing binaries I suppose
\n[30-Sep-2009 12:46:49] <rmatte> well, zenpatch is basically just\ndownloads the diff file and automatically uses diff to apply the patch
[30-Sep-2009 12:46:57] <rmatte> so in reality, diff is clever
[30-Sep-2009 12:47:03] <rmatte> well "patch" rather
[30-Sep-2009 12:47:20] <rmatte> zenpatch is really nice though, big time saver
[30-Sep-2009 12:47:43] <__jd__> I know. I just meant it was a clever hack because of how simple it is
[30-Sep-2009 12:47:56] <__jd__> it's very convenient
[30-Sep-2009 12:48:17] <rmatte> yeh
[30-Sep-2009 12:48:18] * __jd__ needs to break something now to test that last patch
[30-Sep-2009 12:49:06] * rmatte breaks __jd__'s leg
[30-Sep-2009 12:49:13] <rmatte> oh, you meant software-wise
[30-Sep-2009 12:53:47] <__jd__> broken -- we'll see what happens
[30-Sep-2009 12:53:54] <purevol> who has crabs?
[30-Sep-2009 12:54:46] <__jd__> Do you happen to know how to tell zenoss that its hostname is not localhost.localdomain?
\n[30-Sep-2009 12:55:43] <__jd__> the hostname of the server is not\nlocalhost.localdomain, so I'm not sure where it's getting it from,\nunless it's just a dedfault
[30-Sep-2009 12:55:44] <__jd__> default
[30-Sep-2009 12:56:21] <__jd__> oh wait.. I think I might know what it's doing
[30-Sep-2009 13:01:25] <cgibbons> do tell
[30-Sep-2009 13:02:14] <ckrough> its looking at 127.0.0.1 in /etc/hosts would be my guess
[30-Sep-2009 13:02:31] <ckrough> you probably dont want to change that line in /etc/hosts...
[30-Sep-2009 13:02:44] <__jd__> yea, I had <myhostname> 127.0.0.1
[30-Sep-2009 13:02:59] <__jd__> or vice versa
[30-Sep-2009 13:03:23] <__jd__> so I guess it must look up the IP of `hostname` and then do a reverse lookup on that
[30-Sep-2009 13:03:30] <ckrough> lol
[30-Sep-2009 13:03:34] <rmatte> cgibbons: under collector performance, do you know what the Cycle Times are measured in?
[30-Sep-2009 13:03:35] <__jd__> kind of round-about way of getting a hostname
[30-Sep-2009 13:03:48] <rmatte> cgibbons: is it the total time it's taking to complete each cycle?
[30-Sep-2009 13:03:58] * ckrough was wondering that yesterday
[30-Sep-2009 13:03:58] <rmatte> cgibbons: or is it response time in milliseconds?
[30-Sep-2009 13:04:31] <rmatte> cgibbons: the cycle time for zenperfsnmp suddenly jumped to 62.53 when it was previously at around 15-20
[30-Sep-2009 13:04:33]\n<cgibbons> the value is the number of seconds a single\n'collection loop' takes - not sure if it gets translated for the graph\nor not
[30-Sep-2009 13:04:39] <rmatte> trying to figure out why it suddenly jumped so high...
[30-Sep-2009 13:05:01] <cgibbons> for which daemon?
[30-Sep-2009 13:05:14] <rmatte> also, it's showing nan for 4 different daemons, even though those daemons are operating fine
[30-Sep-2009 13:05:25] <rmatte> zenperfsnmp is the one that I'm seeing as much higher than it was previously
[30-Sep-2009 13:05:44] <rmatte> but I'm getting threshold alerts for zenprocess, zeneventlog, etc...
[30-Sep-2009 13:05:49] <rmatte> for cycle time thresholds
[30-Sep-2009 13:06:49]\n<rmatte> zenperfsnmp doesn't appear to be going over threshold,\nbut it's very odd that the cycle times suddenly increased so\ndrastically on tuesday morning with no increase in collected events
[30-Sep-2009 13:07:04] <cgibbons> yeah, and your CPU load on the collector server itself is okay?
[30-Sep-2009 13:07:26]\n<rmatte> we actually removed the VM from the resource pool to\ngive it unlimited resources, and we're still seeing this
[30-Sep-2009 13:07:38] <rmatte> so I'm thinking possible network issue, or something to do with Zenoss
[30-Sep-2009 13:08:41]\n<cgibbons> usually that's either a) too much load on a daemon, b)\nnetwork problems with a lot of slow to respond devices/timeouts, c)\ngeneral host overload
[30-Sep-2009 13:09:00]\n<cgibbons> zenperfsnmp prints out these cycle time stats in the\nlog if you want to look, you might see an increase in the # of slow to\nrespond devices in there?
[30-Sep-2009 13:09:32] <rmatte> hmmm, I'll check it out, just got heartbeat failures for zenping and zentrap
[30-Sep-2009 13:11:02] <rmatte> I see the odd line like: 2009-09-30 10:22:09 WARNING zen.zenprocess: Problem devices: ['XP03BMSTv01']
[30-Sep-2009 13:14:18] <rmatte> hmmm
[30-Sep-2009 13:14:35] <rmatte> the usual time that it takes for zenping is:
[30-Sep-2009 13:14:36] <rmatte> 2009-09-30 12:07:39 INFO zen.ZenPing: Finished pinging 341 jobs in 3.20 seconds
[30-Sep-2009 13:14:56] <rmatte> just before that:
[30-Sep-2009 13:14:58] <rmatte> 2009-09-30 12:06:42 INFO zen.ZenPing: Finished pinging 344 jobs in 6.23 seconds
[30-Sep-2009 13:15:18] <rmatte> wondering why it suddenly has 344 jobs instead of 341
\n[30-Sep-2009 13:15:31] <ckrough> when we get to a high number of\ndevices on a collector, a single "problem" device seems to cause major\ncycle issues
[30-Sep-2009 13:15:38] <ckrough> its actually one of our top scaling problems
[30-Sep-2009 13:15:45] <rmatte> lovely
[30-Sep-2009 13:15:53] <ckrough> but the number of jobs doesnt change
[30-Sep-2009 13:16:07] <rmatte> 2009-09-30 11:39:41 INFO zen.ZenPing: Finished pinging 349 jobs in 6.05 seconds
[30-Sep-2009 13:16:08] <ckrough> particularly with zenperfsnmp around 1600 devices.
[30-Sep-2009 13:16:12] <rmatte> seems to be fluctuating
[30-Sep-2009 13:16:24] <rmatte> Like, check this out...
[30-Sep-2009 13:16:24] <rmatte> 2009-09-30 11:35:37 INFO zen.ZenPing: Finished pinging 341 jobs in 1.99 seconds
[30-Sep-2009 13:16:25] <rmatte> 2009-09-30 11:36:39 INFO zen.ZenPing: Finished pinging 343 jobs in 3.31 seconds
[30-Sep-2009 13:16:25] <rmatte> 2009-09-30 11:37:38 INFO zen.ZenPing: Finished pinging 343 jobs in 3.26 seconds
[30-Sep-2009 13:16:25] <rmatte> 2009-09-30 11:38:37 INFO zen.ZenPing: Finished pinging 341 jobs in 1.81 seconds
\n[30-Sep-2009 13:16:26] <ckrough> it runs great when everybody is\nhappy, when one device goes down it can hang beyond 300s, and it seems\nto snowball
[30-Sep-2009 13:16:33] <rmatte> how the heck does it magically get 2 more jobs?
[30-Sep-2009 13:16:53] <__jd__> it's going for extra credit
[30-Sep-2009 13:16:57] <ckrough> that little guy? dont worry about that little guy
[30-Sep-2009 13:17:02] <rmatte> lol
[30-Sep-2009 13:17:09] <rmatte> super troopers!
[30-Sep-2009 13:17:13] <cgibbons> it can split up OID requests across tasks and repeat slow tasks; it's a complex daemon
[30-Sep-2009 13:17:13] <ckrough> win
[30-Sep-2009 13:17:33] <rmatte> well, this is zenping, so no OID requests
\n[30-Sep-2009 13:17:33] <cgibbons> the new framework doesn't have\na single collection loop anymore so deciding when it's 'slow' is going\nto be a learning experience
[30-Sep-2009 13:18:00] <rmatte> 2009-09-30 08:06:40 INFO zen.ZenPing: Finished pinging 347 jobs in 6.23 seconds
[30-Sep-2009 13:18:00] <rmatte> 2009-09-30 08:07:39 INFO zen.ZenPing: Finished pinging 358 jobs in 5.01 seconds
[30-Sep-2009 13:18:00] <rmatte> 2009-09-30 08:08:37 INFO zen.ZenPing: Finished pinging 350 jobs in 3.49 seconds
[30-Sep-2009 13:18:10] <rmatte> is it on crack?
[30-Sep-2009 13:18:11] <rmatte> lol
[30-Sep-2009 13:19:03] <ckrough> cgibbons: can you extrapolate on that?
\n[30-Sep-2009 13:19:26] <ckrough> cgibbons: when we get a problem\ndevice zenperfsnmp waits until timeout then the cycle can fail since we\nare already close to the limit
[30-Sep-2009 13:19:42] <rmatte> yeh, this client has several problem devices
[30-Sep-2009 13:19:51] <cgibbons> zenperfsnmp didn't change to the new framework for 2.5, so this doesn't affect it yet
[30-Sep-2009 13:19:54] <rmatte> where you can literally read the output of an snmpwalk line by line as it runs
[30-Sep-2009 13:19:54] <ckrough> cgibbons: is that not expected behavior in 2.3.3?
[30-Sep-2009 13:20:03] <ckrough> ah
[30-Sep-2009 13:20:09]\n<cgibbons> but in the new collector framework the way the tasks\nare scheduled for collection are different, so there's no longer a\nsingle 'everything has to run here' loop
[30-Sep-2009 13:20:22] <rmatte> but it was never this bad, it's strange that the cycle times suddenly balooned up
\n[30-Sep-2009 13:20:23] <cgibbons> it'll be a big boon for most of\nthe collectors since a slow device won't affect the rest of the world
[30-Sep-2009 13:20:35] <ckrough> awesome
[30-Sep-2009 13:20:41] <rmatte> cgibbons: is this in King Crab?
[30-Sep-2009 13:20:49] <cgibbons> yeah, for zenwin, zeneventlog, zenprocess, and zenwinperf
[30-Sep-2009 13:20:56] <rmatte> perfect, can't wait
\n[30-Sep-2009 13:21:29] <cgibbons> zenwinperf especially i expect\na big improvement on the high-end of scalability limits. i think we've\ngot one customer with about 600 devices on a single collector right\nnow, so i'd wager we could go much higher with the new version
[30-Sep-2009 13:22:17] <rmatte> yup
[30-Sep-2009 13:22:23] <rmatte> definitely looking forward to that
\n[30-Sep-2009 13:22:48] <ckrough> cgibbons: how does it get past\nslow/nonresponsive devices? moves the job to a bad device thread or\nsomething?
[30-Sep-2009 13:23:23] <cgibbons> and\ni don't know when it'll be coded, but one of the features we put into\nthe design was the ability to easily extend the framework so that tasks\ncan be scheduled via subprocesses (sorta like zenhub), so that'll give\nbetter performance on multi-core systems without the need to configure\nmore software collectors
[30-Sep-2009 13:23:59] <rmatte> well, don't you technically already have that?
[30-Sep-2009 13:24:11] <rmatte> "parallel 2"
[30-Sep-2009 13:24:11] <rmatte> or whatever
[30-Sep-2009 13:24:11] <rmatte> works on certain daemons
\n[30-Sep-2009 13:24:36] <cgibbons> right now everything is done\nvia asynchronous I/O, so there's no real waiting unless it has been\ncoded to do so - and it has. so if you can imagine a single loop that\nwaits for all 100 devices to complete before it begins the next cycle\n(zenperfsnmp is more complicated).
[30-Sep-2009 13:24:37] <jb> hey cgibbons
[30-Sep-2009 13:24:53] <rmatte> yeh
[30-Sep-2009 13:25:01]\n<cgibbons> in the new world, device A will run every 5 minutes\nand device B might run every 5 minutes, but if device C takes 7 minutes\nto respond, device A & B still keep being scheduled without having\nto wait for C to finish.
[30-Sep-2009 13:25:51] <rmatte> are there giant killer robots in this new world?
[30-Sep-2009 13:26:37] <ckrough> so what happens with 1000 devices all fall on the same cycle?
[30-Sep-2009 13:27:10] <ckrough> chaos. mass hysteria. dogs and cats living together.
\n[30-Sep-2009 13:27:23] <cgibbons> they're still independent of\none another from a collection framework, and they won't affect one\nanother
[30-Sep-2009 13:27:38] <ckrough> I was thinking more of the system resources
\n[30-Sep-2009 13:28:12] <ckrough> how many concurrent snmp\ncollections can you do before something gets io bound, and if it does,\nhow gracefully does it complete
[30-Sep-2009 13:28:15] <ckrough> just thinking out loud
\n[30-Sep-2009 13:28:34] <cgibbons> rmattte: yeah parallel is\nsimilar to the idea of independent scheduling for the daemons that\nsupport it, but it's not that robust
[30-Sep-2009 13:29:03]\n<cgibbons> SNMP is a little weird since it's a UDP-based\nprotocol, so your main limit is how many packets you can send out @\nonce without dropping them, and it's a pretty low limit
[30-Sep-2009 13:29:11] <cgibbons> lemme see, there's a constant in the zenperfsnmp daemon today for that
[30-Sep-2009 13:29:32] <cgibbons> 20-ish
[30-Sep-2009 13:29:41] <ckrough> 20 I think? but thats a limit of the target device not zenoss
[30-Sep-2009 13:29:44] <cgibbons> with "@ once" being undefined
[30-Sep-2009 13:29:57] <cgibbons> it's really a limit of the host's TCP/IP protocol stack
\n[30-Sep-2009 13:30:28] <cgibbons> you can do some experiments\neven with local echo stuff to do see how quickly UDP stuff starts to\ndrop packets
[30-Sep-2009 13:30:43] <__jd__> it's a local kernel buffer most likely
[30-Sep-2009 13:30:47] <cgibbons> yep
[30-Sep-2009 13:30:48] <__jd__> if you overflow it, packets drop
[30-Sep-2009 13:31:06] <cgibbons> in TCP world it'll figure it out, but things built on UDP.... timeout!
\n[30-Sep-2009 13:31:31] <cgibbons> so zenperfsnmp tries to be\nintelligent about that, and limits the maximum concurent packets it\nsends at once, and also how big requests are so it doesn't overflow\npacket size limitations
[30-Sep-2009 13:32:07]\n<cgibbons> i'm not sure how much, if any, scaling improvement\nwe'll see with zenperfsnmp once it moves over since it's already rather\nrobust about this and a lot of features from it were put into the new\nframework so all the daemons get them
[30-Sep-2009 13:34:08] <ckrough> I was thinking of the maxoids per request for IOS
[30-Sep-2009 13:35:01] * ckrough hears this little voice screaming c++ in the back of his head
[30-Sep-2009 13:36:37] <cgibbons> yep there's that value too... looks like ours is 40
\n[30-Sep-2009 13:36:56] <cgibbons> i think there's some logic in\nthere to split up a request into smaller in some error conditions too,\nif it is close to the edge
[30-Sep-2009 13:39:14] <rmatte> since when is ntp.nasa.gov no longer a valid ntp source?
[30-Sep-2009 13:39:59] <rmatte> oh, it just didn't work the first time I tried it
[30-Sep-2009 13:40:00] <rmatte> weird
[30-Sep-2009 13:42:08] * ckrough leaves zenoss mode and goes back to php mode for a while
[30-Sep-2009 13:42:16] <ckrough> time to relearn using semi-colons. again.
[30-Sep-2009 13:42:25] <__jd__>
[30-Sep-2009 13:42:53] <rmatte> lol
[30-Sep-2009 13:42:59] <__jd__> that was directed not so much at semicolons, but at php
[30-Sep-2009 13:43:05] <ckrough> man this is ugly. I want to go back in time and smack myself
[30-Sep-2009 13:43:22] <__jd__> oh my
[30-Sep-2009 13:43:26] <__jd__> netsplit?
[30-Sep-2009 13:43:34] <rmatte> netjoin
[30-Sep-2009 13:43:42] <ckrough> freenode does those fairly often
[30-Sep-2009 13:43:56] <ckrough> all part of the fluids of the intarwebs
[30-Sep-2009 13:43:59] <rmatte> every IRC network does those fairly often lol
[30-Sep-2009 13:44:09] <__jd__> I've been away from IRC for a while
[30-Sep-2009 13:44:15] <__jd__> brings back efnet memories
[30-Sep-2009 13:44:21] <ckrough> feels good doesnt it
[30-Sep-2009 13:44:22] <jmp242> Anyone know why a linux process wouldn't autoclear when it comes back up?
\n[30-Sep-2009 13:44:26] <rmatte> they are actually kind of fun,\nsee a bunch of people leave the room at once: "NETSPLIT!!!"
[30-Sep-2009 13:44:38] <ckrough> jmp242: something different in the dedup?
[30-Sep-2009 13:44:43] <rmatte> jmp242: did it come back up with different options?
[30-Sep-2009 13:47:05]\n<jmp242> I believe one of the processes did, though one did not -\nso I'm wondering if the different options, does that end up a wholly\nnew process?
[30-Sep-2009 13:47:25] <jmp242> I expect so...
[30-Sep-2009 13:47:42] <jmp242> Well, ckrough weirdly enough, there WAS no CLEAR event
\n[30-Sep-2009 13:47:45] <rmatte> if there are different options on\nthe process Zenoss won't recognize it as the same process
[30-Sep-2009 13:47:47] <jmp242> so I could not compare dedupID
\n[30-Sep-2009 13:48:10] <jmp242> but I guess if Zenoss thinks it's\nnot the same process, it wouldn't generate a CLEAR event
[30-Sep-2009 13:48:17] <jmp242> so that seems most likely
[30-Sep-2009 13:48:28] <rmatte> correct
[30-Sep-2009 13:48:43] <jmp242> Ok, that was my mystery
[30-Sep-2009 13:48:45] <rmatte> Go to Processes (link in the Zenoss menu)
[30-Sep-2009 13:48:48] <rmatte> then find that process
[30-Sep-2009 13:48:55] <rmatte> click on it
[30-Sep-2009 13:48:58] <rmatte> go to the edit tab
\n[30-Sep-2009 13:48:59] <jmp242> did they ever fix the display\nissue with processes showing as UP / Green in the OS tab no matter what?
[30-Sep-2009 13:49:05] <rmatte> and set Ignore parameters to true
[30-Sep-2009 13:49:10] <rmatte> and click save
[30-Sep-2009 13:49:11] <rmatte> then remodel
[30-Sep-2009 13:49:21] <jmp242> yea, rmatte that is a good idea, except these are glassfish servers, hence java
[30-Sep-2009 13:49:23] <rmatte> (won't work if you're monitoring multiple versions of the same process)
[30-Sep-2009 13:49:26] <rmatte> ah
[30-Sep-2009 13:49:28] <rmatte> gotcha
[30-Sep-2009 13:49:33] <jmp242> and so you need the parameters to use it at all
\n[30-Sep-2009 13:49:46] <rmatte> well, you might want to look in\nto why the service is coming back up with different options
[30-Sep-2009 13:49:50] <jmp242> which is fine, I just needed to understand what happened
[30-Sep-2009 13:49:58] <rmatte> k
[30-Sep-2009 13:50:01] <jmp242> the Linux guy changed the process options, and wondered why the event didn't autoclear
[30-Sep-2009 13:50:08] <jmp242> when the process came up
[30-Sep-2009 13:50:15] <rmatte> ah
[30-Sep-2009 13:50:19] <rmatte> yeh, that's why
[30-Sep-2009 13:50:28] <rmatte> any time they are changed you need to remodel
\n[30-Sep-2009 13:50:35] <jmp242> did they ever fix the display\nissue with processes showing as UP / Green in the OS tab no matter what?
[30-Sep-2009 13:50:46] <jmp242> Is that in the 2.5 beta anyone?
[30-Sep-2009 13:50:49] <rmatte> jmp242: I would imagine so
[30-Sep-2009 13:51:02] <jmp242> cause I was also getting that in 2.4.5
[30-Sep-2009 13:51:03] <rmatte> I've never noticed an issue with that
[30-Sep-2009 13:51:09] <rmatte> oh?
[30-Sep-2009 13:51:12] <rmatte> hmmm
\n[30-Sep-2009 13:51:23] <jmp242> I do wonder if it has to do with\nthe loooooong process name + options with the glassfish server
[30-Sep-2009 13:51:31] <jmp242> and that database size overflow
[30-Sep-2009 13:51:33] <rmatte> could be
[30-Sep-2009 13:51:36] <jmp242> that the fix it code got lost
[30-Sep-2009 13:51:40] <jmp242> with the forum move
[30-Sep-2009 13:51:53] <jmp242> PITA... I don't know how to recover from that - mailarchive I guess via google
\n[30-Sep-2009 13:52:32] <rmatte> I love how an entire commission\nis pending because I can't login to one frigging device, which should\nhave been configured to allow me access, gah
[30-Sep-2009 13:53:12] <jmp242> mmmm sounds fun
[30-Sep-2009 13:53:14] <rmatte> jmp242: yeh, or just use cached pages on google
[30-Sep-2009 13:56:29] <rmatte> I think I'm going to setup a website to host the old version of ZipTie
[30-Sep-2009 13:58:22] <jmp242> why?
[30-Sep-2009 13:59:24] <ckrough> mrayzenoss: docs/DOC-3563
[30-Sep-2009 13:59:32] <ckrough> mrayzenoss: that API doc cant be unzipped
[30-Sep-2009 13:59:42] <ckrough> [crow@enki Zenoss_API]$ tar xzvf ~/Download/zenoss-2.4.5-apidocs.tar.gz
[30-Sep-2009 13:59:43] <ckrough> tar: This does not look like a tar archive
[30-Sep-2009 14:00:13] <jmp242> rmatte: thanks for the google cache idea - much faster!
[30-Sep-2009 14:00:19] <jmp242> still...
[30-Sep-2009 14:00:35] <jmp242> sad I have to do that. Just spent this morning fixing links on the Community FAQ
[30-Sep-2009 14:00:52] <mrayzenoss> ckrough: bizarre, I just redownloaded and tar -xzvf'd it
[30-Sep-2009 14:01:03] <ckrough> meh. it failed for both me and Pete
[30-Sep-2009 14:01:39] <ckrough> trying it with wget
[30-Sep-2009 14:02:57] <ckrough> uh. with firefox (twice) file size is 7531240, wget snags 7600927. wget version downloads fine
[30-Sep-2009 14:03:10] <ckrough> insert WTH here<>
[30-Sep-2009 14:03:21] <mrayzenoss> agreed
[30-Sep-2009 14:03:36]\n<ckrough> someone else running FF on linux try that download?\nservlet/JiveServlet/download/3563-1-1767/zenoss-2.4.5-apidocs.tar.gz
[30-Sep-2009 14:03:54] <mrayzenoss> worked for me with FF and Chrome
[30-Sep-2009 14:04:04] <ckrough> os?
[30-Sep-2009 14:04:12] <mrayzenoss> OS X 10.6
[30-Sep-2009 14:05:02] <exarkun> Firefox on Linux, seems to work, no progress indicator though, md5sum is 21ab9f245aa77e4643d87067e90099d7
[30-Sep-2009 14:05:32] <mrayzenoss> same md5 here
[30-Sep-2009 14:06:44] <jmp242> What OS ckrough?
[30-Sep-2009 14:06:53] <jmp242> did you want me to try on Windows?
[30-Sep-2009 14:07:08] <ckrough> Fedora11, FF3.5.3. Pete is on centos5.3
[30-Sep-2009 14:07:53] <jmp242> mmm
[30-Sep-2009 14:08:34] <ckrough> wget c632f052a043583d0c267e7038407500, ff 21ab9f245aa77e4643d87067e90099d7
[30-Sep-2009 14:08:38] <ckrough> ff version wouldnt expand
[30-Sep-2009 14:08:43] <ckrough> Im running 64bit
[30-Sep-2009 14:09:37] <ckrough> oh well, looks like it's on our end. not sure wth that could be. thanks.
[30-Sep-2009 14:13:36] <exarkun> Oh. I didn't try to unpack it. Tar doesn't like it.
\n[30-Sep-2009 14:16:45] <__jd__> it could be zenoss' side too if\nit's happening all over the place.. maybe it's user-agent detection\ngone wrong
[30-Sep-2009 14:17:25] <ckrough> yeah. I thought it was just me. but exarkun had the same issue
[30-Sep-2009 14:19:25] <mrayzenoss> if I could get a couple of people to verify it, I'll raise it up with IT/Jive
\n[30-Sep-2009 14:20:13] <mrayzenoss> anyone listening, please hit\ndocs/DOC-3563 and 'tar -xzvf\nzenoss-2.4.5-apidocs.tar.gz'
[30-Sep-2009 14:21:50]\n<xamox> quick question, do I need to install zenoss on all my\nclients, or just have snmp setup and working?\\
[30-Sep-2009 14:21:58] <ckrough> the latter
[30-Sep-2009 14:22:06] <ckrough> zenoss is agentless for most monitoring
[30-Sep-2009 14:22:16] * xamox claps
[30-Sep-2009 14:22:20] <xamox> awesome
[30-Sep-2009 14:22:36] <xamox> That's all I needed to know
[30-Sep-2009 14:22:41] <xamox> Thanks
[30-Sep-2009 14:22:46] <ckrough> you'll be back
[30-Sep-2009 14:22:49] <ckrough> gl
[30-Sep-2009 14:22:50] <xamox> haha, probably
[30-Sep-2009 14:23:01] <jmp242> no, you will be xamox
[30-Sep-2009 14:23:11] * xamox hides in fear
[30-Sep-2009 14:23:18] <mrayzenoss> don't scare him away
[30-Sep-2009 14:23:19] <ckrough> its all good.
[30-Sep-2009 14:23:21] <xamox> If I can talk my boss into it, then I will be back
[30-Sep-2009 14:24:39] <ckrough> we can help with most objections
[30-Sep-2009 14:24:40] <mrayzenoss> we're recreating the issue with the tarball
[30-Sep-2009 14:24:46] <ckrough> thanks mrayzenoss
[30-Sep-2009 14:24:54] <mrayzenoss> ckrough: can I email you one to test?
[30-Sep-2009 14:24:58] <ckrough> sure
[30-Sep-2009 14:24:59] <jb> mrayzenoss: still have a zenpack backlog?
[30-Sep-2009 14:25:10] <mrayzenoss> jb: yeah, I should go do that
[30-Sep-2009 14:25:17] <mrayzenoss> you're at the top
[30-Sep-2009 14:25:18] <jb>
[30-Sep-2009 14:25:23] <jb> just checking!
[30-Sep-2009 14:25:27] <mrayzenoss> sorry, been busy with the new site and Ohio LinuxFest
[30-Sep-2009 14:25:31] <jb> yeah, not a problem.
[30-Sep-2009 14:27:30] <jmp242> Umm, question on the process monitoring
\n[30-Sep-2009 14:27:59] <jmp242> how come Zenoss is expanding out\nthe entire commandline of the process vs just the regex match?
\n[30-Sep-2009 14:28:30] <jmp242> I mean, it seems like there has\nto be a better way to deal with these java processes etc than what\nwe've currently got...
[30-Sep-2009 14:29:08] <ckrough> "expanding out the entire commandline of the process vs just the regex"
[30-Sep-2009 14:29:12] <ckrough> can you give an example?
[30-Sep-2009 14:29:36] <ckrough> I've noticed in a lot of their code they append a '.*' to searches
[30-Sep-2009 14:30:16] <mrayzenoss> ckrough: check your email
[30-Sep-2009 14:30:20] <ckrough> kk
[30-Sep-2009 14:30:44] <ckrough> krough.org or work?
[30-Sep-2009 14:30:48] <mrayzenoss> work
[30-Sep-2009 14:31:05] <ckrough> k, exchange... will be a few.
\n[30-Sep-2009 14:31:17] <rmatte> yikes, just noticed that I didn't\nhave the operational status threshold set on my 64bit ethernet template
[30-Sep-2009 14:31:22] <mrayzenoss> if that works for you, I'll replace the tarball that's up there
[30-Sep-2009 14:32:16] <jmp242> sure.
[30-Sep-2009 14:32:22] <jmp242> we have a regex like
\n[30-Sep-2009 14:32:27] <jmp242> '/opt/SUNWappserver/jdk/bin/java\n-Dcom.sun.aas.instanceRoot=/opt/SUNWappserver/domains/domain1'
\n[30-Sep-2009 14:32:59] <jmp242> but zenoss will think it's a new\nprocess as I stated above if we, for instance, change the port number,\nand the process name in the device is like
\n[30-Sep-2009 14:33:16] <jmp242> '/opt/SUNWappserver/jdk/bin/java\n-Dcom.sun.aas.instanceRoot=/opt/SUNWappserver/domains/domain1\n-Dcom.sun.aas.ClassPathPrefix= -Dcom.sun.aas.ClassPathSuffix=\n-Dcom.sun.aas.ServerClassPath=\n-Dcom.sun.aas.classloader.appserverChainJars.ee=\n-Dcom.sun.aas.classloader.appserverChainJars=admin-cli.jar,admin-cli-ee.jar,j2ee-svc.jar\n-Dcom.sun.aas.classloader.excludesList=admin-cli.jar,appserv-upgrade.jar,sun-appserv-ant.jar
\n[30-Sep-2009 14:33:16] <jmp242>\n-Dcom.sun.aas.classloader.optionalOverrideableChain.ee=\n-Dcom.sun.aas.classloader.optionalOverrideableChain=webservices-rt.jar,webservices-tools.jar
\n[30-Sep-2009 14:33:16] <jmp242>\n-Dcom.sun.aas.classloader.serverClassPath.ee=/opt/SUNWappserver/hadb/4.4.3-6/lib/hadbjdbc4.jar,/opt/SUNWappserver/lib/SUNWjdmk/5.1/lib/jdmkrt.jar,/opt/SUNWappserver/hadb/4.4.3-6/lib/dbstate.jar,/opt/SUNWappserver/hadb/4.4.3-6/lib/hadbm.jar,/opt/SUNWappserver/hadb/4.4.3-6/lib/hadbmgt.jar,/opt/SUNWmfwk/lib/mfwk_instrum_tk.jar
\n[30-Sep-2009 14:33:19] <jmp242>\n-Dcom.sun.aas.classloader.serverClassPath=/opt/SUNWappserver/lib/install/applications/jmsra/imqjmsra.jar,/opt/SUNWappserver/imq/'
[30-Sep-2009 14:34:03]\n<ckrough> mrayzenoss: 16b7fcbe530d9cac41d9283013f7fb23 , it\nworks, but there are dotfiles in there I didnt see in the other one
[30-Sep-2009 14:34:31] <mrayzenoss> ckrough: hmm... those are probably my fault
[30-Sep-2009 14:34:46] <mrayzenoss> OSX'll do that
[30-Sep-2009 14:35:31] <mrayzenoss> I'm not seeing them
[30-Sep-2009 14:35:39] <ckrough> lol
[30-Sep-2009 14:36:31] <ckrough> other than them, everything looks good
[30-Sep-2009 14:36:35] <mrayzenoss> aight
\n[30-Sep-2009 14:37:25] <jmp242> I'm assuming it must have\nsomething to do with if you have multiple versions of the process that\ncould match on specific computers
[30-Sep-2009 14:38:03]\n<jmp242> but it seems to me that it would be better for the end\nuser if they could specify how specific they want to get using the\nexisting mechanism of the regex
[30-Sep-2009 14:38:13] <mrayzenoss> ok, new version of the API uploaded if you're looking for it
[30-Sep-2009 14:38:19] <mrayzenoss> docs/DOC-3563
[30-Sep-2009 14:43:09] <ckrough> mrayzenoss: sorry, that one fails too
[30-Sep-2009 14:43:17] <ckrough> mrayzenoss: must be something Jive related
[30-Sep-2009 14:43:30] <mrayzenoss> wtf...
[30-Sep-2009 14:43:36] <ckrough> 097792cad19d0903b044c0731837769b
\n[30-Sep-2009 14:43:50] <patzer> hi, I am trying to make my own\nzenpack, and following through the tutorial I notice that if I add a\nzproperty, it doesn't seem to show up unless I export the zenpack and\nthen install that .egg
[30-Sep-2009 14:44:11] <patzer> am I missing something here to avoid having to do this?
[30-Sep-2009 14:44:47] <mrayzenoss> ckrough: confirmed
\n[30-Sep-2009 14:44:51] <ckrough> mrayzenoss: Firefox doesnt know\nthe total size during download, no progress bar... which is odd
[30-Sep-2009 14:45:04] <mrayzenoss> I get different md5s between what I uploaded and what comes back down
[30-Sep-2009 14:46:05] <cgibbons> yikes
[30-Sep-2009 14:46:39] <mrayzenoss> ckrough: email version is OK, but downloaded is messed up?
[30-Sep-2009 14:46:45] <ckrough> mrayzenoss: correct
[30-Sep-2009 14:47:05]\n<ckrough> mrayzenoss: I've only downloaded zenpacks from the new\nsite via wget, but they were fine after unzipping and --install
[30-Sep-2009 14:47:25] <mrayzenoss> ckrough: we're debugging, gonna raise the issue with Jive
[30-Sep-2009 14:47:32] <mrayzenoss> perhaps it's .tar.gz related...
[30-Sep-2009 15:04:05] <cgibbons> so how many of y'all would like to write more zenpacks if it were even easier?
[30-Sep-2009 15:06:04] <ckrough> sure
[30-Sep-2009 15:06:57] <jmp242> sure
[30-Sep-2009 15:07:12] <ckrough> is this a trick question
[30-Sep-2009 15:07:18] <cgibbons> nope
[30-Sep-2009 15:07:30]\n<jmp242> I haven't ever had the itch to write a zenpack, though\nif there was something I was going to set up I'd give it a go
[30-Sep-2009 15:08:00]\n<jmp242> My problem is my coworkers can't come up with anything\nmuch to monitor that isn't OOTB that they actually would want to spend\ntime on
[30-Sep-2009 15:08:16] <jmp242> it shows how far Zenoss is advanced from the previous monitoring solution
[30-Sep-2009 15:08:36]\n<ckrough> yeah, most of the stuff I monitor is either ootb or so\nspecific it's not relevant to enough people to writer a zenpack for
[30-Sep-2009 15:10:46] <mrayzenoss> so a little digging I just did on the stuff on zenpacks.zenoss.org
[30-Sep-2009 15:11:05] <mrayzenoss> 64 of 96 hosted ZenPacks are more than a simple template export
[30-Sep-2009 15:11:23] <ckrough> are more or are not more
[30-Sep-2009 15:11:24] <mrayzenoss> so they have some sort of coding, not just exported from the UI
[30-Sep-2009 15:11:28] <ckrough> nuce
[30-Sep-2009 15:11:28] <cgibbons> surprised its that many
[30-Sep-2009 15:11:29] <ckrough> nice
[30-Sep-2009 15:11:33] <mrayzenoss> yeah me too
[30-Sep-2009 15:12:06] <ckrough> is there much collaboration on zenpacks? seems to be one-man-one-zenpack for most
[30-Sep-2009 15:12:43] <mrayzenoss> there are just a handful with multiple authors
[30-Sep-2009 15:12:53] <mrayzenoss> but most of the ones with multiple revisions are updated based on feedback
[30-Sep-2009 15:13:04] <mrayzenoss> which includes patches
\n[30-Sep-2009 15:13:43] <mrayzenoss> so if there's more than 1\nrevision, I'd assume that someone gave them feedback and they acted on\nit, which is good
[30-Sep-2009 15:18:37] <rmatte> When is my latest ZenPack going up?
[30-Sep-2009 15:18:39] <rmatte>
[30-Sep-2009 15:18:53] <mrayzenoss> I keep trying to work in zenpacks today
[30-Sep-2009 15:19:00] <mrayzenoss> keep running into roadblocks
[30-Sep-2009 15:19:12] <mrayzenoss> this tar thing has been raised to Jive
[30-Sep-2009 15:19:31] <mrayzenoss> the fixed forum posts is supposed to go up today
[30-Sep-2009 15:19:53] <mrayzenoss> I copy-editted the newsletter that just went out
[30-Sep-2009 15:19:55] <rmatte> tar thing or zip thing?
[30-Sep-2009 15:19:59] <mrayzenoss> tar thing
[30-Sep-2009 15:20:03] <rmatte> what tar thing?
[30-Sep-2009 15:20:13] <mrayzenoss> scroll up and look for ckrough's posts
[30-Sep-2009 15:20:28] <mrayzenoss> there appears to be an issue with uploaded tarballs
[30-Sep-2009 15:20:48] <rmatte> ah
[30-Sep-2009 15:21:10] <Brad_K> aren't they getting zipped automatically?
[30-Sep-2009 15:21:18] <mrayzenoss> not the tarballs
[30-Sep-2009 15:21:27] <rmatte> the eggs are
[30-Sep-2009 15:21:31] <rmatte> which is uber annoying
[30-Sep-2009 15:21:36] <mrayzenoss> yeah, I doc'd that as much as I can
[30-Sep-2009 15:22:02]\n<Brad_K> are you sure that they're not getting zipped\nautomatically? did you actually take a look at the two different\ntarballs?
[30-Sep-2009 15:22:02] <mrayzenoss> blame IE for that one
[30-Sep-2009 15:22:04] <rmatte> I'm sure a lot of new users are getting frustrated by that
[30-Sep-2009 15:22:17] <rmatte> oh, that was an IE issue?
[30-Sep-2009 15:22:18] <Brad_K> or maybe it's a transport issue (binary vs. ascii)?
[30-Sep-2009 15:22:22] <mrayzenoss> Brad_K: yeah, I uploaded and downloaded, different MD5s
[30-Sep-2009 15:22:31] <mrayzenoss> Brad_K: yeah, I'm sure it's something dumb
[30-Sep-2009 15:22:47] <Brad_K> right, different MD5s, but maybe the same file -- one before zipping, one after.
[30-Sep-2009 15:23:11] <Brad_K> and if it's a .tgz, that would explain why the files are probably roughly the same size.
[30-Sep-2009 15:23:15] <mrayzenoss> Brad_K: just tested, no luck
[30-Sep-2009 15:23:15] <rmatte> yeh, zipping would result in a different MD5 checksum
[30-Sep-2009 15:23:41] <mrayzenoss> I can untar both versions, other people can't
[30-Sep-2009 15:24:08] <Brad_K> have you used a binary editor to look at the contents of the two different files?
[30-Sep-2009 15:24:40] <Brad_K> it might also be a DOS cr/lf conversion issue.
\n[30-Sep-2009 15:24:45] <mrayzenoss> Brad_K: I could go down that\npath... or I could hand it off to Jive and tell them to fix it
[30-Sep-2009 15:24:53] <Brad_K> 'k
[30-Sep-2009 15:25:11] <mrayzenoss> since it's hosted we can't fix much
[30-Sep-2009 15:26:05]\n<Brad_K> sorry, i just like to know more about a problem before i\ntell someone to fix it, in part because i want to make sure that they\nfix the real underlying problem, and not just whatever they might think\nit is.
[30-Sep-2009 15:27:05] <mrayzenoss> understood
[30-Sep-2009 15:27:17] <mrayzenoss> our IT guy dug into it before raising the issue with Jive
[30-Sep-2009 15:27:43] <mrayzenoss> he was recreating the issue with the tarballs and he's on Linux
[30-Sep-2009 15:27:56] <mrayzenoss> so encoding issues seems likely
[30-Sep-2009 15:28:17] <mrayzenoss> but now I gotta go push out another beta release
[30-Sep-2009 15:30:42] <rmatte> Matt, any plans to apply the latest set of patches to 2.4.5 and do a 2.4.6 release?
[30-Sep-2009 15:30:50] <mrayzenoss> no
[30-Sep-2009 15:31:03] <rmatte> release date for King Crab been set?
[30-Sep-2009 15:31:13] <rmatte> (in stone, as it were)
[30-Sep-2009 15:31:15] <mrayzenoss> not publicly
[30-Sep-2009 15:31:22] <rmatte> that's a no
[30-Sep-2009 15:31:24] <rmatte> lol
[30-Sep-2009 15:31:28] <mrayzenoss> next week is being talked about as a RC1
[30-Sep-2009 15:31:38] <cgibbons> i know the date
[30-Sep-2009 15:31:41] <Brad_K> is that "set in stone crab"?
[30-Sep-2009 15:31:44] <rmatte> yeh, but RC is very bleeding edge
[30-Sep-2009 15:31:46] <cgibbons> we don't tell matt what the date is for obvious reasons
[30-Sep-2009 15:32:07]\n<rmatte> cgibbons: it's always comforting to know that devs know\ntheir own release dates, I'd be very very concerned if you didn't.
[30-Sep-2009 15:34:03] <cgibbons> we hit our dates pretty consistently for the major releases, +- 2 weeks tops so far
\n[30-Sep-2009 15:34:24] <cgibbons> i think this one will be later\nthan our internal date just because of a marketing push coming along\nwith it
[30-Sep-2009 15:35:22] <mrayzenoss> here's what the dev's are up to: http://dev.zenoss.org/trac/report/3
[30-Sep-2009 15:36:03] <rmatte> quite honestly, I hope you guys hit +2 weeks and get this release right lol
[30-Sep-2009 15:36:18] <rmatte> after that 2.4.3 release, I can't really take much more of it
[30-Sep-2009 15:37:41] <cgibbons> we're at the stage now where we don't put in any fixes that aren't priority-1
\n[30-Sep-2009 15:38:51] <rmatte> cgibbons: well, I understand that\nthat's the approach to take in the interests of time, but the King Crab\nrelease is sort of a huge deal, it would be nice if it doesn't get\nreleased while still riddled with obvious bugs
\n[30-Sep-2009 15:39:17] <Brad_K> yes, please. make this one\nsolid. it's been very disheartening to find so many examples of people\nhaving problems with restores not working, and with upgrades not\nworking. right now, i'm definitely feeling like we're damned if we do\nupgrad, and we're damned if we don't.
[30-Sep-2009\n15:39:30] <cgibbons> yep, but there's always a line in the sand\nsomewhere. that report matt gave you is the current work list so you\ncan get an idea
[30-Sep-2009 15:40:29] <rmatte>\nyeh, I've seen the work lists and I understand that there's a lot of\nstuff to look at, but quite honestly, if it gets released without the\nissue that I reopened ticket 4997 for fixed, it's going to be hugely\nnoticeable and people will be screaming
[30-Sep-2009 15:40:54] <rmatte> It's not a show stopping issue, thus not pri 1, but it is a very noticeable bug
\n[30-Sep-2009 15:41:58] <Brad_K> it's definitely going to hurt us\ntrying to make the case to management that we chose the right\nEnterprise Monitoring software to pay thousands and thousands of\ndollars for.
[30-Sep-2009 15:42:37] <rmatte> Brad_K: yeh, always fun
\n[30-Sep-2009 15:42:58] <Brad_K> no matter how bad the other\nalternatives would have been, we didn't go down those roads, so we have\nto live by the reality of Zenoss compared to the marketing of the\nothers.
[30-Sep-2009 15:44:34] <rmatte> I'm just\nsaying that I find that the Zenoss releases are very rushed. I can't\nrecall an upgrade that I've done of Zenoss where I can truly say that\neverything was working flawlessly up until now, and even now, I have\n2.4.5 installed with patches on top of that.
[30-Sep-2009 15:44:35] <mrayzenoss> Enterprise will lag behind Core by 2 weeks for the official release
\n[30-Sep-2009 15:44:53] <chudler> someone asked me for a way to\nmake zenoss combine alerts/events when they cross some count. I wonder\nif anyone here has tread this before?
[30-Sep-2009 15:46:02]\n<Brad_K> two weeks means nothing to us. we're in the middle of\nhaving a massive internal debate as to whether or not we try to take\n2.4.2 (or 2.4.5) and maybe meet our original schedule of turning off\nthe old monitoring system by Thanksgiving, including our developing our\nown internal training for the operators.
[30-Sep-2009\n15:46:21] <Brad_K> but then we'd have to re-teach them all over\nagain once we try to upgrade to 2.5 sometime next year.
[30-Sep-2009 15:46:44] <rmatte> Brad_K: what version are you currently using?
[30-Sep-2009 15:46:47] <rmatte> 2.4.5?
[30-Sep-2009 15:47:04]\n<Brad_K> or, do we delay rolling this out until sometime next\nyear when we've had time to integrate 2.5 into what we're trying to do\nnow, and then we don't have to retrain them all over again.
\n[30-Sep-2009 15:47:30] <Brad_K> right now, we've got 2.4.2. and\ni'm not looking forward to what's going to happen when we attempt to\nupgrade to 2.4.5.
[30-Sep-2009 15:47:35] <mrayzenoss> community/beta Beta 4 is out
[30-Sep-2009 15:47:42] <mrayzenoss> writing up the blog post
[30-Sep-2009 15:47:48] <mrayzenoss> upgrades can be tested for stack installers
[30-Sep-2009 15:47:56] <mrayzenoss> LinuxMonitor is updated
[30-Sep-2009 15:47:57] <rmatte> Brad_K: I upgraded from 2.4.2 to 2.4.5 and it was fine
\n[30-Sep-2009 15:48:19] <rmatte> well, I upgraded from 2.4.2 to\n2.4.3 and it went fine, then there were 2 huge bugs, so I patched my\nsystems up to 2.4.5
[30-Sep-2009 15:48:29] <Brad_K> that is very encouraging. i just wish i knew where the upgrade process broke for others.
[30-Sep-2009 15:48:32] <rmatte> 2.4.4 fixed 1 bug, 2.4.5 fixed 1 bug
[30-Sep-2009 15:48:48] <rmatte> Well, what type of install did you do?
\n[30-Sep-2009 15:49:11] <Brad_K> dunno. i wasn't working on the\nproject back in July when they did the most recent re-install.
[30-Sep-2009 15:49:18] <rmatte> and is Zenoss running on a physical box or a VM?
[30-Sep-2009 15:49:43] <rmatte> I see, what version of Linux is it running on?
[30-Sep-2009 15:49:56] <rmatte> and what is your $ZENHOME directory?
\n[30-Sep-2009 15:49:59] <Brad_K> what we're working on right now\nis running on physical boxes. we've got a VM that we're planning to\nuse for doing development testing with in the future.
[30-Sep-2009 15:50:23] <Brad_K> RHEL 5.something. i don't know if we're up on 5.4 yet -- i doubt it, but you never know.
[30-Sep-2009 15:50:44] <Brad_K> $ZENHOME = /opt/zenoss
[30-Sep-2009 15:50:53] <rmatte> RPM install then
[30-Sep-2009 15:50:59] <Brad_K> should be.
[30-Sep-2009 15:51:07] <Brad_K> in theory.
[30-Sep-2009 15:51:09] <rmatte> I don't use the RPM install, so I can't guarantee anything in regards to how well an upgrade would go
[30-Sep-2009 15:51:15] <rmatte> I use stack installer since it's much easier
[30-Sep-2009 15:51:48] <rmatte> If it's in /opt/zenoss then guaranteed it's an RPM install
[30-Sep-2009 15:51:53] <rmatte> stack install is /usr/local/zenoss/zenoss
[30-Sep-2009 15:52:03] <rmatte> source is /usr/local/zenoss
[30-Sep-2009 15:53:00]\n<Brad_K> we've also had some local hacks made by consulting\nservices. we're hoping that they survive the upgrade process.
[30-Sep-2009 15:53:57] <Brad_K> plus there's all the stuff we've added after the install.
[30-Sep-2009 15:54:55] <tim> Hi room
[30-Sep-2009 15:55:47] <tim> Is this the correct arena to ask for support/guidance
[30-Sep-2009 15:56:46] <cgibbons> yup
[30-Sep-2009 15:57:22] <rmatte> This is the best arena for it in fact
[30-Sep-2009 15:57:36] <rmatte>
[30-Sep-2009 15:57:40] <tim> Ah cool thanks guys
\n[30-Sep-2009 15:58:32] <tim> I have installed zenoss core, and\ndone a discovery run. But it has not found all the devices. Where do I\nstart?
[30-Sep-2009 15:58:58] <rmatte> well, are they all in whatever range you specified?
[30-Sep-2009 15:59:40] <rmatte> and are they pingable from the zenoss box?
\n[30-Sep-2009 15:59:57] <tim> Yes, I put all of the network ranges\nin to the disco run but it only returned 56 out of 400 pc (for example)
[30-Sep-2009 16:00:17]\n<feutete_> Is it possible to query snmp on multiple ports on a\nsingle device? We have what amounts to an snmp daemon that listens on\n10161 to provide information specific to a webapp
[30-Sep-2009 16:00:20] <cgibbons> and, it depends on what kind of devices (i.e. all snmp?)
[30-Sep-2009 16:00:27] <tim> Yes they are pingable, not all are running snmp or WMi tho
[30-Sep-2009 16:00:37] <feutete_> but we also want to gather standard snmp data on port 161
[30-Sep-2009 16:00:50] <rmatte> cgibbons: that shouldn't matter, discovery is done partially via plain ping
[30-Sep-2009 16:01:38] <cgibbons> indeed
[30-Sep-2009 16:01:43] <tim> Ah I think I will check that port on the firewalls (there are 6)
[30-Sep-2009 16:01:43] <rmatte> you're looking to monitor workstations?
[30-Sep-2009 16:02:07] <rmatte> If I were you I'd set any firewalls to allow the Zenoss box full access
[30-Sep-2009 16:02:16] <rmatte> otherwise you're going to run in to all kinds of fun little issues
[30-Sep-2009 16:02:29] <rmatte> afk for a sec
[30-Sep-2009 16:02:33] <tim> Not as such but as a point of excorcise i proves that it is functioning
[30-Sep-2009 16:03:15] <tim> Thanks guys I will sort the firewalls first then try again
[30-Sep-2009 16:04:01]\n<rmatte> also, if you're auto-discovering workstations (which it\nsounds like you may be), they all have to be powered on during\ndiscovery (obviously)
[30-Sep-2009 16:04:14] <cgibbons> ouf that had to be hard to type on
[30-Sep-2009 16:05:06] <Brad_K> i've done ssh on my iphone. it's actually not that hard.
[30-Sep-2009 16:05:17] <rmatte> I use colloquy on my iphone all the time, not that bad
[30-Sep-2009 16:05:26] <rmatte> and I do ssh all the time too
[30-Sep-2009 16:05:27] <rmatte>
[30-Sep-2009 16:05:30] <cgibbons> i IM chat a lot on mine but it's still hard to do tech talk on it i find
[30-Sep-2009 16:05:56]\n<rmatte> yeh, tech talk isn't the simplest, especially with the\nlovely auto-correct feature which I so lovingly turned off recently
[30-Sep-2009 16:06:48]\n<Brad_K> tim must have done the same, looking at what he typed. \nthe auto-correct function would not have let some of those through.
[30-Sep-2009 16:10:06] <rmatte> yeh
[30-Sep-2009 16:10:25] <rmatte> actually, the trick I use is to put 1 letter is front of the cursor as I type
[30-Sep-2009 16:10:34] <rmatte> autocorrect ignores what you're typing if you do that
[30-Sep-2009 16:10:41] <rmatte> in front*
\n[30-Sep-2009 16:20:41] <ke4qqq> apparently I am doing something\nreally stupid with processes - I tried to step up the complexity a bit\nand instead of monitoring simple regexes for things like httpd I am\nlooking for this: http://fpaste.org/YuIL/
[30-Sep-2009\n16:20:49] <cgibbons> i'm totally addicted to my iphone. i can't\nimagine life without it anymore. sorta like life before the internet,\ntoo. scary! dark!
[30-Sep-2009 16:21:06] <rmatte> cgibbons: same
[30-Sep-2009 16:21:20] <rmatte> cgibbons: you don't think you need one until you have one
\n[30-Sep-2009 16:21:42] <cgibbons> i was actually annoyed to find\nthe NPS lodge @ big bend NP had WiFi last year (no cellular coverage)\nso I was still connected to the world even out there
[30-Sep-2009 16:22:01] <rmatte> cgibbons: take a look at what ke4qqq is trying to do, he was asking me the other day and I had no idea
[30-Sep-2009 16:22:09] <rmatte> I don't even think what he's trying to do is possible
[30-Sep-2009 16:22:16] <cgibbons> k, i shall in a moment, in a video conference ::)
[30-Sep-2009 16:22:17] <ke4qqq> rmatte: why not?
[30-Sep-2009 16:22:20] <rmatte> k
[30-Sep-2009 16:22:31] * ke4qqq thought I had only been working on this for a single day
\n[30-Sep-2009 16:22:44] <rmatte> ke4qqq: because, you're trying to\ndo the regex against the options, and I don't think that the regex even\nlooks at the options
[30-Sep-2009 16:22:51] <rmatte> it only looks at the actual process name
[30-Sep-2009 16:23:02] <rmatte> in snmp the process name and the options are displayed with different OIDs
[30-Sep-2009 16:23:09] <rmatte> they aren't stuck together
[30-Sep-2009 16:23:11] <ke4qqq> yes they are
\n[30-Sep-2009 16:23:30] <ke4qqq> but then it has the entire\n'ignore arguments' boolean - why have that if you can't use them
[30-Sep-2009 16:23:34] <rmatte> I specifically remember going through the snmp tree and seeing them displayed separately
[30-Sep-2009 16:23:50] <ke4qqq> yes they are in a separate oid iirc
[30-Sep-2009 16:23:59] <ke4qqq> I assumed concatenation
[30-Sep-2009 16:25:54] <ke4qqq> Jane Curry's paper on process monitoring
[30-Sep-2009 16:26:18] <ke4qqq> shows an example with arguments on p22
[30-Sep-2009 16:26:22] <ke4qqq> hence my attempt
[30-Sep-2009 16:26:35] <rmatte> oh, well if it's in there then Jane must have gotten it to work
[30-Sep-2009 16:26:41] <rmatte> so I guess it is possible
[30-Sep-2009 16:27:00] <ke4qqq> well it's enough for today, I'll deal with it tomorrow
[30-Sep-2009 16:35:25] <rmatte> bah, as if it's only wednesday
[30-Sep-2009 16:36:27] <mrayzenoss> I'm kinda slow
[30-Sep-2009 16:38:19] <rmatte> hehe
[30-Sep-2009 16:38:20] <rhettardo> waat
\n[30-Sep-2009 16:55:39] <mrayzenoss> here's blog post for the new\nbeta:\nblogs/zenossblog/2009/09/30/zenoss-25-beta-4-is-now-available
[30-Sep-2009 16:56:22] <rmatte> right on, I shall be installing tomorrow
[30-Sep-2009 16:56:56] <mrayzenoss> and upgrading
[30-Sep-2009 16:57:02] <mrayzenoss> stack installer upgrades can be tested
[30-Sep-2009 16:57:11] <rmatte> cool
[30-Sep-2009 16:57:24] <rmatte> I'll be sure to test it out and let you know how it goes
[30-Sep-2009 17:02:02] <mrayzenoss> if possible, try to go from you 2.4.5 install, not your previous beta
[30-Sep-2009 17:02:15] <mrayzenoss> since we're looking for testing of 2.4.x->2.5
[30-Sep-2009 18:18:00]\n<exarkun> What does the "Linux Monitor" pack do? I thought it\nwould give me a new template to bind, but it doesn't seem to have.
[30-Sep-2009 18:19:10] <exarkun> Maybe my "Device Class Path" is wrong.
[30-Sep-2009 18:21:04] <exarkun> Ah. I guess so. I changed it to "Server/SSH/Linux" and some other stuff happened.
[30-Sep-2009 18:21:09] * exarkun looks around to see if ssh keys are supported.
\n[30-Sep-2009 18:21:14] <tung> Do you know which password this\nerror refers to in this error message regarding a Windows Server:
\n[30-Sep-2009 18:21:26] <tung> Could not read the Windows event\nlog (NT_STATUS_IO_TIMEOUT). Check your username/password settings
\n[30-Sep-2009 18:32:24] <cgibbons> tung: exactly that, a timeout\nto the server. that usually means the network to the server is down or\n(more likely) a firewall is preventing communication with it.
[30-Sep-2009 18:32:58] <tung> But I can snmpwalk the server just fine
[30-Sep-2009 18:33:09] <tung> There's no firewall in between
[30-Sep-2009 18:33:36] Traveler is now known as Guest99219
\n[30-Sep-2009 18:33:59] <cgibbons> snmp doesn't come into play\nhere, reading the Windows Event Log occurs via a WMI query over DCOM,\nso that's port 135 + some other stuff
[30-Sep-2009\n18:34:16] <Guest99219> hello anyone here able to shed some light\non this: ERROR RenderServer Failed to generate a graph
\n[30-Sep-2009 18:34:26] <cgibbons> if the Windows device itself\nhas the software firewall enabled or not configured to allow DCOM then\nit'll also block it
[30-Sep-2009 18:34:54] <tung> DCOM? Really?
[30-Sep-2009 18:34:58] <cgibbons> yep
[30-Sep-2009 18:35:00] <tung> I didn't know that
[30-Sep-2009 18:35:34] <tung> OK let me get the Windows guy to check that
[30-Sep-2009 18:35:41] <tung> Thanks
[30-Sep-2009 18:35:53] <cgibbons> if you want to try something real quick from zenoss I can give him a few more hints
[30-Sep-2009 18:36:04] <cgibbons> (from the zenoss server that is, a command-line test)
[30-Sep-2009 18:36:10] <tung> BTW, is that UDP or TCP (for DCOM)
[30-Sep-2009 18:36:21] <tung> sure what test can I try?
\n[30-Sep-2009 18:36:46] <cgibbons> it's TCP. basically the\nlayering is Microsoft RPC over TCP, with DCOM layered on top of that,\nand WMI layered on top of that yet still
[30-Sep-2009\n18:37:11] <cgibbons> from the command-line: wmic -d99 -U\n'yourdomain\\youruser' //yourhostname "select name from\nwin32_computersystem"
[30-Sep-2009 18:37:28]\n<Guest99219> @tung, dcom and run over udp as well reference \n-http://en.wikipedia.org/wiki/Port_numbers
\n[30-Sep-2009 18:37:28] <cgibbons> that should produce a lot of\nnotice; if you put it in pastebin i can tell you how far it's getting
[30-Sep-2009 18:39:20] <tung> OK I don't have a login to that box :-(
[30-Sep-2009 18:39:27] <cgibbons> okie
[30-Sep-2009 18:39:45] <tung> Let me try to get him to open 135 TCP/UDP first and see
[30-Sep-2009 18:40:03] <tung> Thanks for the answer
[30-Sep-2009 18:40:03] <tung> as always :-D
[30-Sep-2009 18:40:13] <tung> This is a great forum
\n[30-Sep-2009 18:40:16] <Guest99219> cgibbons- is there a way i\ncan grab a more verbose output for "ERROR RenderServer Failed to\ngenerate a graph" such as teh wmic comand you presented tung?
[30-Sep-2009 18:40:35] <cgibbons> where are you seeing that error?
[30-Sep-2009 18:41:13] <Guest99219> in the zopectl log
\n[30-Sep-2009 18:41:41] <cgibbons> tung: you may find\nhttp://msdn.microsoft.com/en-us/library/aa389286(VS.85).aspx and\nhttp://msdn.microsoft.com/en-us/library/ms809327.aspx useful references\nfor him if it comes to that
[30-Sep-2009 18:42:09]\n<Guest99219> previous this was working - i actually have\nhistorical data going back a few weeks and can confirm the wmi query is\ncorrect and returns valid results
[30-Sep-2009\n18:42:11] <cgibbons> you can edit $ZENHOME/etc/zope.conf and look\nfor the debug-mode directive and change it to be enabled and then\nrestart zope (zopectl restart) to see more debug
[30-Sep-2009 18:43:10] <cgibbons> also look at your zenrender.log and see if it said anything useful
[30-Sep-2009 18:45:00] <Guest99219> alright - restarting zope in debug mode
\n[30-Sep-2009 18:45:40] <Guest99219> i get serveral of these but\nit seems i can safely ignore - INFO ZODB.Conflict database conflict\nerror
[30-Sep-2009 18:45:52] <cgibbons> yep
[30-Sep-2009 18:46:07] <cgibbons> iirc zenrender works by using /tmp/renderserver for a work area so double check it's got free space etc
[30-Sep-2009 18:49:47] <Guest99219> free space checks out ok, along with permissions on that directory
\n[30-Sep-2009 18:50:42] <Guest99219> can i send you a pastebin\noutput - doesn't seem like the debug mode helped with teh verbosity
[30-Sep-2009 18:52:03] <Guest99219> http://pastebin.com/d75762205
[30-Sep-2009 18:57:37] Traveler is now known as Guest49354
[30-Sep-2009 18:57:55] <Guest49354> got d/c'd.
[30-Sep-2009 18:58:06] <Guest49354> have a chance to check the pastebin?
[30-Sep-2009 19:00:34] <Guest49354> cgibbons: i do not see a zenrender log under /opt/zenoss/log - is it located somewhere else?
[30-Sep-2009 19:13:14] <Guest49354> i like the new website
[30-Sep-2009 19:25:43] <Guest49354> too bad theres no redirects for pages that were already index by search engines
\n[30-Sep-2009 19:34:15] <Guest49354> getting back to my render\ngraph issue - it does appear related to INFO ZODB.Conflict database\nconflict error
[30-Sep-2009 19:34:34] <Guest49354> i thought i could safely ignore, but that doesn't appear to be the case
[30-Sep-2009 19:36:30] Traveler is now known as Guest78799
[30-Sep-2009 20:56:52] <cryptographrix> What MIBs would I need for a basic ZenOSS install?
[30-Sep-2009 20:57:19] <cryptographrix> i.e. - is there a standard set of MIBs to install?
[30-Sep-2009 21:46:45] <cryptographrix> "Can't stop reactor that isn't running." when I model
[30-Sep-2009 21:46:47] <cryptographrix> ideas?
\n[30-Sep-2009 21:50:39] * exarkun knows what that means and why it\nhappens but not how to fix it except by editing Zenoss source.
[30-Sep-2009 21:52:00] <cryptographrix> lol ah,kk
[30-Sep-2009 21:52:12] <cryptographrix> is it a bug?
[30-Sep-2009 21:52:35] <cryptographrix> trying to figure out how ZenOss works to use it for monitoring a contract I have
[30-Sep-2009 22:01:26] <exarkun> I think it's a bug. I'm pretty new to zenoss too though.
[30-Sep-2009 22:01:33] <cryptographrix> coolcool
[30-Sep-2009 22:01:40] <cryptographrix> yah I'ma be on for the devs tomorrow lol
[01-Oct-2009 00:00:46] [disconnected at Thu Oct 1 00:00:46 2009]
[01-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[01-Sep-2009 06:52:31] kevin7kal_ is now known as kevin7kal
\n[01-Sep-2009 07:29:33] <etank> is there a way to get a graph /\ndata point of the return from SQL Transactions instead of just the\ntime it took the transaction to run?
[01-Sep-2009 08:52:25] <jb> morning..
[01-Sep-2009 08:53:05] <Apocalipse> hello
[01-Sep-2009 08:59:53] <rmatte> if he had waited another 10 minutes I would have been here (Apocalipse was messaging me)
[01-Sep-2009 09:17:34] <jb> hey rmatte
[01-Sep-2009 09:18:05] <rmatte> hey
[01-Sep-2009 09:18:14] <jb> need some report assistance
[01-Sep-2009 09:18:23] <jb>
[01-Sep-2009 09:18:51] <jb> i have a zenpack that graphs throughput of virtual servers on my F5 LTM's
[01-Sep-2009 09:19:05] <jb> one website may contain 4 virtual servers or so
[01-Sep-2009 09:19:09] <jb> which all have separate graphs
\n[01-Sep-2009 09:19:28] <jb> i want to make a combination graph\nthat takes the sum of all the virtual servers for one website's\nthroughput and graphs them on one graph
[01-Sep-2009 09:20:06] <rmatte> jb: hmmm, you're basically looking to make an aggregate report
[01-Sep-2009 09:20:24] <jb> ah ok
\n[01-Sep-2009 09:20:29] <rmatte> hold on, I'm just testing\nsomething then I'll see if I can find some reference material for you
[01-Sep-2009 09:20:36] <jb> thanks
[01-Sep-2009 09:38:48] <rmatte> jb: crap, I have a meeting that I'm late for apparently
[01-Sep-2009 09:38:51] <rmatte> I'll be back later
[01-Sep-2009 09:39:22] <jb> np
[01-Sep-2009 10:01:05] <cgibbons> hmmmm
[01-Sep-2009 10:56:42] <mrayzenoss> heading to a meeting, 2.4.5 is trickling out right now
[01-Sep-2009 11:16:33] crow is now known as ckrough
\n[01-Sep-2009 11:44:58] <jsm> so other than installing the dell\nmonitor zenpack and it's dependency, is there anything else i have to\ndo to see the newly modeled data other than re-modeling the device ? i\ndon't see any new data
[01-Sep-2009 11:45:22] <mrayzenoss> did you bind the template?
[01-Sep-2009 11:46:42] <jsm> yes, they are all bound to my /devices/server/linux class
[01-Sep-2009 11:47:03] <mrayzenoss> does it have a custom zCollectorPlugin?
[01-Sep-2009 11:48:22] <jsm> they are in the zcollectorplugins for that class
[01-Sep-2009 11:48:43] <mrayzenoss> and they're set?
[01-Sep-2009 11:48:53] <mrayzenoss> added to that device class, that is?
[01-Sep-2009 11:49:01] <mrayzenoss> you'll need to remodel after that
[01-Sep-2009 11:51:19] <jsm> if by set, you mean if i look in /Servers/Linux collector plugins, they are alll listed
[01-Sep-2009 11:51:42] <jsm> and the device has been remodelled
[01-Sep-2009 11:52:40] <mrayzenoss> I don't have any experience with that ZenPack, maybe someone else here has used it?
[01-Sep-2009 11:52:47] <mrayzenoss> or perhaps post on the forums?
[01-Sep-2009 11:52:48] <mrayzenoss> sorry
[01-Sep-2009 11:54:41] <jsm> It's somewhat disconcerting that so many of the community zenpacks have 0 documentation
[01-Sep-2009 11:54:49] <jsm> should make it a requirement from the author.
[01-Sep-2009 11:56:04]\n<jsm> Egor, while contributing a lot of code, from what i can\nfind, doesn't give much information on how to actually use the plugins.
[01-Sep-2009 11:59:58] <ckrough> Not that I disagree, but the argument usually becomes "it's open source, contribute".
[01-Sep-2009 12:01:06] <cwj> but first, use your magic divining rod to figure out how it works, so you can!
[01-Sep-2009 12:04:14]\n<jsm> the difference is in this case mray has the ability to\naccept projects.. if you accept a projects w/out documentation, you are\nbenefiting no one
[01-Sep-2009 12:05:03] <jsm> and i'd love to contribute, but that's hard when the damn thing doesn't work.
\n[01-Sep-2009 12:12:28] <aceazza> Can someone help me figure out\nhow to patch my zenoss server. I'm running 2.4.2 and I want to get to\nthe latest version. I've been searching the web but can't find much\ninstruction for applying patches. I did find something that says use\nzenpatch <patch_number> . Where can I download the patch files?\nThanks!
[01-Sep-2009 12:13:56] <ckrough> aceazza: Are you up on the issues with 2.4.3?
[01-Sep-2009 12:14:10] <aceazza> No
[01-Sep-2009 12:14:11] <ckrough> I think they are rolling out 2.4.5(?) currently
[01-Sep-2009 12:14:18] <ckrough> see the the /topic
[01-Sep-2009 12:14:25] <aceazza> Yeah, I saw 2.4.5 available for download
[01-Sep-2009 12:14:33] <ckrough> You may want to check the forums and see where everything is at before doing the upgrade
[01-Sep-2009 12:18:28] <jb> jsm.. are your dell boxes running openmanage?
[01-Sep-2009 12:18:36] <jsm> jsm, yes
[01-Sep-2009 12:18:43] <jsm> err, jb
[01-Sep-2009 12:18:49] <jb> what plugins are listed when you model the device?
[01-Sep-2009 12:19:30] <jb> i may help egor out by writing some documentation
[01-Sep-2009 12:19:40] <jb> but, theres not much to it really
[01-Sep-2009 12:20:00] <jsm> jb: http://pastebin.com/m10ca41ef
\n[01-Sep-2009 12:25:36] <tripitaka> Hi, my zenoss server is\nrunning very slow, what are the usual causes? It has 8GB of 16GB RAM\nfree, the CPU ( 2 quad core opterons) are usually around 70% idle, what\nelse should I be looking at?
[01-Sep-2009 12:26:05] <jb> jsm: and nothing in the hardware tab?
[01-Sep-2009 12:26:17] <tripitaka> (2.4.2 on rhel5.3 x86_64)
[01-Sep-2009 12:26:21] <jb> tripitaka: how many devices?
[01-Sep-2009 12:26:41] <tripitaka> just under 2,000
[01-Sep-2009 12:26:45] <jb> hmm
[01-Sep-2009 12:26:52] <jb> have you checked I/O?
[01-Sep-2009 12:27:16] <jb> check iowait..
[01-Sep-2009 12:27:24] <jb> what type of storage are you using for zenoss?
[01-Sep-2009 12:27:45] <tripitaka> internal SAS, 15k rpm
[01-Sep-2009 12:27:58] <tripitaka> top shows about 1% in wait
[01-Sep-2009 12:28:20] <jb> what type of raid?
[01-Sep-2009 12:28:33] <tripitaka> iostat shows up to 10,000 blockes written/sec
[01-Sep-2009 12:28:39] <tripitaka> raid 1+0, two mirrored disks
[01-Sep-2009 12:28:43] <jb> ah
[01-Sep-2009 12:28:45] <jb> k
[01-Sep-2009 12:29:06] <tripitaka> what's it writing to? mysql?
[01-Sep-2009 12:29:13] <jb> for events, yes
[01-Sep-2009 12:29:25] <jb> is EVERYTHING slow, or just certain things?
[01-Sep-2009 12:29:49] <aceazza> So is apt-get upgrade zenoss-stack the way to update on an Ubuntu system?
[01-Sep-2009 12:30:02] <jb> s
[01-Sep-2009 12:30:05] <tripitaka> depends, nearly all mouse clicks take a long time to respond, 30 - 90 seconds and then some timeouts
[01-Sep-2009 12:30:14] <tripitaka> but occasionally it's snappy
\n[01-Sep-2009 12:30:37] <tripitaka> modelling and device adds\noften fail with timeout connecting to zenhub, but work on retry
[01-Sep-2009 12:31:19] <jb> are you runnning core?
[01-Sep-2009 12:31:23] <tripitaka> stack
[01-Sep-2009 12:31:33] <jb> zenoss core, or enterprise?
[01-Sep-2009 12:31:38] <tripitaka> oh, core
[01-Sep-2009 12:31:42] <jb> k
[01-Sep-2009 12:32:01]\n<jb> im not sure.. i really don't have any experience with a\ncollector having that many devices.. i know that there are people on\nthe forums that do
[01-Sep-2009 12:32:03] <tripitaka> currently 2.4.2, I tried 2.4.3 last week but there were problems with processes dying
[01-Sep-2009 12:32:06] <jb> and several threads discussing performance
\n[01-Sep-2009 12:32:48] <tripitaka> about half the devices are in\nclasses I don't want to actively monitor - can I completely disable all\nmonitoring for a given class?
[01-Sep-2009 12:33:10]\n<tripitaka> I've tried removing a classes collectors, and\nunbinding all templates, hopefully that would do it
[01-Sep-2009 12:33:19] <tripitaka> but I suspect it's still pinging them
[01-Sep-2009 12:34:12] <jb> set zPingMonitorIgnore to "Yes"
[01-Sep-2009 12:34:19] <jb> on the classes that you dont want to ping
[01-Sep-2009 12:34:23] <aceazza> zenoss core
\n[01-Sep-2009 12:34:26] <tripitaka> hmm, yep, it's still snmp\npolling a device class after I've unbound the 'device' and set the\nzPingMonitorIgnore to 'yes'
[01-Sep-2009 12:34:31] <jb> as well as zSnmpMonitorIgnore
[01-Sep-2009 12:34:53] <tripitaka> I've set those in the class' zProperties, it's still walking them
[01-Sep-2009 12:35:02]\n<cgibbons> you can change hte production state of a device class\nas well; that's the first thing that the system does to see if a device\nshould be monitored by the various daemons or not
[01-Sep-2009 12:35:02] <jb> and zWmiMonitorIgnore if you need..
[01-Sep-2009 12:35:16] <jb> ah yeap
[01-Sep-2009 12:35:39] <tripitaka> oh, wait, I am a moron - I only thought I had
\n[01-Sep-2009 12:36:16] <cgibbons> what we do here is have a\nIgnored device class that has all the interesting bits disabled so the\ndevices in there are completely, well, ignored
\n[01-Sep-2009 12:36:40] <rmatte> hmmmm, am I crazy, or should\nZenoss be automatically detecting IpServices that are set to monitored\nwhen modeling?
[01-Sep-2009 12:39:10] <tripitaka> ok.. all boring systems set to a nonproduction state, and all ignores turned on..
[01-Sep-2009 12:41:10] <tripitaka> load average is dropping consistently
[01-Sep-2009 12:41:19] <tripitaka> yay
[01-Sep-2009 12:45:11] <jb>
[01-Sep-2009 12:45:15] <jb> rmatte: yeah
\n[01-Sep-2009 12:47:32] <rmatte> well, I have 1 IpService that's\nbeing picked up fine, and another 2 that aren't, during a modelling
\n[01-Sep-2009 12:47:50] <rmatte> all 3 ports are publicly\navailable on the device, and there's no difference in config on any of\nthe IpServices
[01-Sep-2009 12:47:59] <rmatte> so I'm stumped
[01-Sep-2009 13:27:16] <markeriv> What exactly about 2.4.3 is broken?
[01-Sep-2009 13:35:22] <jb> event clear email alerts..
[01-Sep-2009 13:36:21] <mrayzenoss> and zencommand clears
[01-Sep-2009 13:46:20] <ckrough> is 2.4.5 out in the wild now?
[01-Sep-2009 13:47:25] <rmatte> it's been released yeh
\n[01-Sep-2009 13:47:53] <rmatte> but if you like being able to set\nhardware product type/info for each device you might want to skip this\nupgrade
[01-Sep-2009 13:54:56] <mrayzenoss> it's out
[01-Sep-2009 13:55:08] <mrayzenoss> ckrough: Enterprise will probably be tomorrow
[01-Sep-2009 13:57:13] <ckrough> thanks. was referring to CORE though
\n[01-Sep-2009 13:57:41] <ckrough> Im going to be helping someone\nget started in Zenoss, was debating on 2.4.5 or the previous stable\nrelease
[01-Sep-2009 13:58:00] <ckrough> some of the bugfixes were significant though, so we'll probably lean towards 4.5
[01-Sep-2009 13:58:20] <mrayzenoss> yeah, we've had a problem with 3 steps forward, 1 step back
[01-Sep-2009 13:59:18] <ckrough> You definitely don't get to claim exclusivity on that one
[01-Sep-2009 14:17:06] <markeriv> hey guys
[01-Sep-2009 14:17:14] <markeriv> i am getting this error when i try those zenpatches
[01-Sep-2009 14:17:21] <markeriv> Hunk #1 FAILED at 273.
[01-Sep-2009 14:17:28] <markeriv> Hunk #2 FAILED at 304.
[01-Sep-2009 14:17:35] <markeriv> Hunk #3 succeeded at 314 with fuzz 1 (offset -13 lines).
[01-Sep-2009 14:17:45] <markeriv> 2 out of 3 hunks FAILED -- saving rejects to file ZenEvents/zenactions.py.rej
[01-Sep-2009 14:20:03] <mrayzenoss> markeriv: were you patching 2.4.3?
[01-Sep-2009 14:20:37] <markeriv> yes
\n[01-Sep-2009 14:21:41] <mrayzenoss> I gotta run, but someone else\nhad a similar problem the other day, they were patching the wrong\nversion
[01-Sep-2009 14:21:48] <markeriv> ahh
[01-Sep-2009 14:21:51] <mrayzenoss> 2.4.5 is out, that rolls up the patches anyway
[01-Sep-2009 14:22:01] <mrayzenoss> be back in a bit
[01-Sep-2009 14:22:46] <markeriv> kk
[01-Sep-2009 14:50:20] <dollarbang> Just got the release notes for 2.4.5, are they skipping 2.4.4?
[01-Sep-2009 14:52:20] <rmatte> correct
[01-Sep-2009 14:52:32] <rmatte> they are basically skipping 2.4.3 too
[01-Sep-2009 14:53:16] <rmatte> 2.4.4 has patch 14744 and 2.4.5 has patch 14769
[01-Sep-2009 15:05:21] <dollarbang1> an interesting way of generating versions. different from the world I'm leaving and entering.
[01-Sep-2009 15:06:53]\n<dollarbang1> btw....how does one ftp from the zenoss site to a\nlinux server? I tried ftp ftp.zenoss.com, but that failed, so I 'm\nguessing that zenoss isn't allowing ftp except for using a browser.
[01-Sep-2009 15:07:33] <chudler> Host ftp.zenoss.com not found: 3(NXDOMAIN)
[01-Sep-2009 15:17:08] <chudler> in other news.. gmail appears to be down
[01-Sep-2009 15:17:20] <jsm> gmail been going down lately
[01-Sep-2009 15:17:49] <jsm> may have to switch my account back to in house
[01-Sep-2009 15:18:02] <jsm> gmail is more unstable than my 5 year old mail server
[01-Sep-2009 15:22:57] <cgibbons> hmmm usually it's just a portion of their cloud that's dead but this one seems systemic
[01-Sep-2009 15:23:27] <jsm> i've been trialing theme for the past few months. not impressed
[01-Sep-2009 15:23:32] <jsm> them
[01-Sep-2009 15:23:44] <jsm> and i have the pay version
[01-Sep-2009 15:28:52] <rmatte> dollarbang1: why would you need to?
[01-Sep-2009 15:29:05] <rmatte> dollarbang1: are you downloading a file, or trying to upload?
[01-Sep-2009 16:12:53] <cgibbons> so interesting, i can get to gmail just fine via imap, but not via the web
\n[01-Sep-2009 16:33:53] <gwb235> so... if i have a data source in\na performance template (nagiosCheckPing) that is a command, how/where\ndo I start tracing/debugging why the associated graph isn't getting\nupdated? (assuming the command works fine on the command line)
[01-Sep-2009 16:34:22] <gwb235> or is it too late in the day for that question?
[01-Sep-2009 16:34:46] <mrayzenoss> never too late, it depends who's around
[01-Sep-2009 16:35:07] <mrayzenoss> no graph, place holder for a graph, or graph with no data points?
[01-Sep-2009 16:35:22] <mrayzenoss> no graph means the template isn't bound
[01-Sep-2009 16:35:43] <mrayzenoss> place holder and no data points means trouble with the data source and data points
\n[01-Sep-2009 16:35:57] <gwb235> graph exists, used to gather data\n(i.e. looking over day history shows data used to be gathered), but now\ngraph shows cur: nan, avg: nan, max: nan
[01-Sep-2009 16:36:01] <mrayzenoss> does the "Test" command work?
[01-Sep-2009 16:36:51] <gwb235> yes (DONE in 4 seconds)
[01-Sep-2009 16:37:37] <mrayzenoss> absolute path or relative path on the data source?
[01-Sep-2009 16:37:55] <gwb235> absolute path /opt/zenoss/libexec/check_ping (blah blah)
[01-Sep-2009 16:38:50] <rsr> hi
[01-Sep-2009 16:38:57] <gwb235> the graph itself has 2009-09-01 16:38:31 in the output
[01-Sep-2009 16:39:13] <mrayzenoss> hmmm.. I assume it's been running for awhile?
[01-Sep-2009 16:39:23] <gwb235> few weeks
[01-Sep-2009 16:39:36] <gwb235> we're still in "deployment" (replacing nagios) stage
[01-Sep-2009 16:39:52] <mrayzenoss> yeah, I'm blowing out the install that I had that was using that ZenPack...
[01-Sep-2009 16:40:02]\n<rsr> I am getting an error when trying to access zenoo from the\nbrowser. I have just finished installing it in linux and the error I\nreceive is : "127.0.0.1" refused the connection. I have no firewall\nactive.
[01-Sep-2009 16:40:05] <mrayzenoss> and actually I didn't install it, I just followed the instructions
[01-Sep-2009 16:40:31] <gwb235> if there's a better (more supported) way of graphing ping loss/RTA i'm open to it...
[01-Sep-2009 16:40:31] <mrayzenoss> rsr: so when you try http://yourhost:8080 that comes up?
[01-Sep-2009 16:40:37] <rsr> yes
[01-Sep-2009 16:40:52] <rsr> any ideas?
[01-Sep-2009 16:41:01] <mrayzenoss> gwb235: there's a half-baked fping collector in the dev docs, but nothing concrete
[01-Sep-2009 16:41:12] <mrayzenoss> rsr: IE7?
[01-Sep-2009 16:41:16] <rsr> no
[01-Sep-2009 16:41:27] <rsr> epiphany
[01-Sep-2009 16:41:33] <rsr> a webkit based gnome browser
[01-Sep-2009 16:41:44] <rsr> Its running on linux
[01-Sep-2009 16:41:45] <mrayzenoss> rsr: yeah, try checking your /etc/hosts
[01-Sep-2009 16:41:53] <mrayzenoss> I'm guessing there's something in there
[01-Sep-2009 16:41:58] <rsr> I am trying from localhost
[01-Sep-2009 16:42:09] <mrayzenoss> rsr: try the hostname:8080 instead
[01-Sep-2009 16:42:15] <rsr> ok
[01-Sep-2009 16:42:33] <rsr> samething
[01-Sep-2009 16:42:55] <mrayzenoss> rsr: what's in your /etc/hosts for localhost and your hostname?
[01-Sep-2009 16:43:05] <mrayzenoss> gwb235: I just followed the instructions from the PDF, didn't even bother to install the ZenPack
[01-Sep-2009 16:43:41] <rsr> 127.0.0.1 localhost
[01-Sep-2009 16:43:41] <mrayzenoss> gwb235: are you using the ${here/manageIp} instead of the $devname?
[01-Sep-2009 16:43:41] <rsr> 127.0.1.1 Zioncell
[01-Sep-2009 16:44:08] <gwb235> no, just $devname
[01-Sep-2009 16:44:10] <mrayzenoss> rsr: try putting 127.0.0.1 localhost and your ip with the hostname
[01-Sep-2009 16:44:17] <mrayzenoss> gwb235: does $devname resolve?
[01-Sep-2009 16:44:38] <mrayzenoss> gwb235: if it was discovered via SNMP, it may return something not resolvable in DNS
[01-Sep-2009 16:44:49] <mrayzenoss> at least, on my home network I found that
[01-Sep-2009 16:44:58] <mrayzenoss> since I just have a hand-coded /etc/hosts
[01-Sep-2009 16:45:06] <gwb235> host <device> returns <device> has address x.x.x.x
[01-Sep-2009 16:45:11] <gwb235> (on the collector)
[01-Sep-2009 16:45:59] <mrayzenoss> gwb235: I assume you've tried "zencommand run -v10 -d yourhost" ?
[01-Sep-2009 16:46:01] <rsr> mrayzenoss, no go... sorry
[01-Sep-2009 16:46:13] <rsr> why would it be refusing the connection?
[01-Sep-2009 16:47:07] <mrayzenoss> is it running?
[01-Sep-2009 16:47:17] <gwb235> yes, i have tried zencommand run (no errors, just DEBUG and INFO)
[01-Sep-2009 16:47:33] <rsr> how can I check? ps aux | grep zenoss?
[01-Sep-2009 16:47:47] <mrayzenoss> rsr: yeah, or a bunch of python commands
[01-Sep-2009 16:47:58] <rsr> I guess not
[01-Sep-2009 16:47:58] <mrayzenoss> gwb235: is this an upgrade?
[01-Sep-2009 16:48:12] <rsr> how do I start it? zenity?
[01-Sep-2009 16:48:12] <gwb235> no, 2.4.2
[01-Sep-2009 16:48:27] <mrayzenoss> rsr: as the 'zenoss' user, do a 'zenoss status'
[01-Sep-2009 16:49:31] <rsr> zenoss command not found
\n[01-Sep-2009 16:49:35] <mrayzenoss> gwb235:\ntrac-zenpacks/wiki/CommunityFAQGeneral?version=90#Upgradingto2.4.xcausedmycommandbaseddatasourcesZenCommandtonotworkwhatdoIdo\n?
[01-Sep-2009 16:49:44] <mrayzenoss> gwb235: I'm grasping at straws
[01-Sep-2009 16:50:01] <rsr> zenoss-stack is installed though
[01-Sep-2009 16:50:04] <mrayzenoss> gwb235: is it just this 1 box that isn't working?
[01-Sep-2009 16:50:09] <mrayzenoss> rsr: which user are you?
[01-Sep-2009 16:50:22] <gwb235> multiple boxes
\n[01-Sep-2009 16:50:24] <gwb235> since we're still in deployment\nphase, is there a better suggested way to get ping RTA/loss
[01-Sep-2009 16:50:54] <rsr> mrayzenoss, ok got it
[01-Sep-2009 16:50:59] <rsr> mrayzenoss, nothings running
[01-Sep-2009 16:51:08] <rsr> sorry its my first time running zenoss
[01-Sep-2009 16:51:17] <mrayzenoss> gwb235: well, there's http://www.zenoss.com/community/projects/zenpacks/fping
[01-Sep-2009 16:51:22] <rsr> man zenoss
[01-Sep-2009 16:51:27] <mrayzenoss> heh
[01-Sep-2009 16:51:30] <rsr> that will help
[01-Sep-2009 16:51:39] <mrayzenoss> rsr: http://www.zenoss.com/community/docs/
[01-Sep-2009 16:51:51] <mrayzenoss> rsr: the Getting Started Guide and Install Guides should help
[01-Sep-2009 16:52:14] <rsr> mrayzenoss, starting the daemon for the first time isnt on the getting started guide
\n[01-Sep-2009 16:52:17] <gwb235> confirmed that zSshSession is set\nto 10 on the template for the device tree i'm working on right now
[01-Sep-2009 16:52:32] <gwb235> i mean, zproperties
[01-Sep-2009 16:52:45] <gwb235> zSshConcurrentSessions
[01-Sep-2009 16:53:10] <rsr> gotit! thanks so much for your help mrayzenoss !
[01-Sep-2009 16:54:00] <mrayzenoss> rsr: I guess the Install Guide comes before the Getting Started Guide
[01-Sep-2009 16:54:22] <rsr> is it on the install guide?
[01-Sep-2009 16:54:36] <mrayzenoss> definitely
[01-Sep-2009 16:54:41] <rsr> I missed it I guess, since I only read the relevant deb doc
[01-Sep-2009 16:54:42] <mrayzenoss> http://www.zenoss.com/community/docs/installation-guide/2.4.2/ch01s02.html
[01-Sep-2009 16:55:05] <mrayzenoss> Deb? http://www.zenoss.com/community/docs/installation-guide/2.4.2/ch04s01.html
[01-Sep-2009 16:56:33] <mrayzenoss> gwb235: so some boxes work and some don't?
[01-Sep-2009 16:56:40] <gwb235> correct
[01-Sep-2009 16:57:08] <gwb235> i'm doing a $ while true; do date; ps auxwwww | grep check_ping | grep -v grep; sleep 1; done;
[01-Sep-2009 16:57:19] <gwb235> and can see some check_pings, but not others
[01-Sep-2009 16:58:34] <mrayzenoss> gwb235: well... 2.4.5 is out :-\\
[01-Sep-2009 16:59:02]\n<gwb235> but even (some of) those that i see in the output (from\nabove) don't have a graph updated with values (although the graph is\nupdated with timestampe when I reload perf tab)
\n[01-Sep-2009 16:59:37] <gwb235> back to the original question,\nhow do i trace a command data source flow from command execution to\ninsertion into RRD?
[01-Sep-2009 16:59:48] <gwb235> or is that waaay to tricky for 4.59PM
[01-Sep-2009 17:01:25] <mrayzenoss> gwb235: usually the 'zencommand run...' is the way to do that
[01-Sep-2009 17:01:42] <mrayzenoss> gwb235: maybe check the other logs, like zenhub.log and make sure everything is ok there?
[01-Sep-2009 17:01:53] <mrayzenoss> has zenhub ever been restarted?
[01-Sep-2009 17:01:57] <mrayzenoss> and zencommand?
[01-Sep-2009 17:02:12] <gwb235> yeah... about an hour before I popped in here...
[01-Sep-2009 17:02:20]\n<mrayzenoss> or if you're overly ambitious, cgibbons wrote this:\nhttp://blog.nuclearbunny.org/2009/03/19/getting-a-native-code-stack-trace-from-a-zenoss-daemon/
[01-Sep-2009 17:02:43] <mrayzenoss> but that's probably not what we're seeing
[01-Sep-2009 17:03:42]\n<gwb235> what should I be seeing? if i run zencommand run on a\nhost that works and a host that doesn't (graph w/ data vs. graph w/out\ndata) the output appears the same
[01-Sep-2009 17:04:18] <mrayzenoss> you should see the commands getting called I believe
[01-Sep-2009 17:04:32] <mrayzenoss> go to Settings->Daemons and edit the config for zencommand, crank up the logging there
[01-Sep-2009 17:04:52] <mrayzenoss> and restart it and at least it'll be in the $ZENHOME/logs/
[01-Sep-2009 17:05:06] <mrayzenoss> change that, then restart the stack
[01-Sep-2009 17:05:17] <mrayzenoss> and start grepping for your missing host...
[01-Sep-2009 17:05:45]\n<mrayzenoss> the good thing is cgibbons put in some more\ndebugging stuff for 2.5 last week, not that that helps us today
[01-Sep-2009 17:09:50] <gwb235> if the device's performance monitor is a collector, is that the one I should be modifying ?
[01-Sep-2009 17:10:35] <mrayzenoss> yes
[01-Sep-2009 17:10:50] <mrayzenoss> zencommand runs SSH and Nagios stuff
[01-Sep-2009 17:12:43] <gwb235> is logseverity 10 what i should set to on the collector
[01-Sep-2009 17:12:50] <mrayzenoss> yeah
\n[01-Sep-2009 17:15:24] <gwb235> ok... so devices that are being\nperfmon from the hub are graphing ok, but devices w/ a perfmon\ncollector are not.
[01-Sep-2009 17:15:41] <gwb235> (looking more closely at the two zencommand logfiles)
[01-Sep-2009 17:17:28] <sergeymasushko> where I can change/remove scale on a graph?
[01-Sep-2009 17:17:59] <mrayzenoss> which scale?
[01-Sep-2009 17:18:28] <mrayzenoss> gwb235: perfmon? The Perl-based one?
[01-Sep-2009 17:18:42] <gwb235> i meant "performance monitor" (collector)
[01-Sep-2009 17:18:42] <mrayzenoss> gwb235: or the zenwinperf?
[01-Sep-2009 17:18:51] <mrayzenoss> ahh, overloaded terminology
[01-Sep-2009 17:19:02] <gwb235> i.e. those devices that have collector: localhost (hub) are graphing OK
[01-Sep-2009 17:19:11] <gwb235> those devices that have collector cornea (collector) are not
[01-Sep-2009 17:19:19] <mrayzenoss> gwb235: Core or Enterprise?
[01-Sep-2009 17:19:23] <gwb235> sorry I didn't realize that earlier
[01-Sep-2009 17:19:24] <gwb235> enterprise
[01-Sep-2009 17:19:44] <gwb235> you want me to dump a bug in and see what they say?
[01-Sep-2009 17:19:54] <mrayzenoss> yeah, that'll get you more help than me
[01-Sep-2009 17:20:07] <mrayzenoss> I don't run Enterprise myself
[01-Sep-2009 17:20:15] <mrayzenoss> even though I work for Zenoss
[01-Sep-2009 17:20:24] <mrayzenoss> I eat the Community dogfood
\n[01-Sep-2009 17:23:19] <cgibbons> hrm speaking of that, i forgot\nthat zenoss was running on that little server i had when it took a\nheader
[01-Sep-2009 17:23:26] <cgibbons> guess i wasn't using that instance much
\n[01-Sep-2009 17:29:31] <sergeymasushko> mrayzenoss: hi Matt, for\nexample I added monitoring of 'laLoadInt1' and on the graph I see wrong\nscale: http://i25.tinypic.com/28un19y.jpg
[01-Sep-2009\n17:34:20] <mrayzenoss> there's probably something in your data\nsource where that's set... you'll have to either delete your rrd file\nor migrate them by hand to get the change
[01-Sep-2009 17:34:28] <mrayzenoss> to be reflected in the graph
\n[01-Sep-2009 17:36:33] <sergeymasushko> heh... I've set the same\nsettings like on the laloadint5... however you see hardcoded 'Aliases\nloadAverage5min | 100,/' on the datasource of laloadint5
[01-Sep-2009 22:05:40] <Egyptian[Home]1> damn .. and i just downloaded the 2.4.3 release yesterday!
[01-Sep-2009 22:39:55] <davetoo> hmm
[01-Sep-2009 23:59:19] <ironpaw> hai
[01-Sep-2009 23:59:27] <ironpaw> anyone kickin?
[01-Sep-2009 23:59:58] <ironpaw> or alive
[02-Sep-2009 00:00:03] <ironpaw> with a pulse, etc?
[02-Sep-2009 00:00:46] [disconnected at Wed Sep 2 00:00:46 2009]
[02-Sep-2009 00:00:46] [connected at Wed Sep 2 00:00:46 2009]
[02-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[02-Sep-2009 02:48:13] <rhett> yo
[02-Sep-2009 06:22:51] <Apocalipse> morning
[02-Sep-2009 06:23:20] <Apocalipse> people...i started to install Cisco zenpack...yesterday
[02-Sep-2009 06:23:56] <Apocalipse> but i closed my shell session...
[02-Sep-2009 06:28:45] <Apocalipse> i am back
[02-Sep-2009 06:36:18] <Apocalipse> hello?
[02-Sep-2009 06:41:04] <Apocalipse> ???
[02-Sep-2009 06:55:51] <Appiah> ?
\n[02-Sep-2009 07:16:57] <Apocalipse> hi people...i would like to\nknow how can i check if a zenpack install process still running
[02-Sep-2009 07:17:32] <ckrough> after closing your sesstion?
\n[02-Sep-2009 07:18:10] <ckrough> some variant of 'ps -ef | grep\n-i xxxx' where xxxx is the zenpack installer process, whatever that may\nbe called. Thats my best guess.
[02-Sep-2009 07:18:31]\n<ckrough> If its been a day it either failed or finished... you\ncould check your settings page and see if it's listed, or look in the\nlogs
[02-Sep-2009 07:18:59] <Apocalipse> its cisco zenpack...
[02-Sep-2009 07:26:00] <Apocalipse> what log can i check?
[02-Sep-2009 07:28:53] <ckrough> not entirely sure. I would start with $ZENHOME/logs/event.log though
\n[02-Sep-2009 07:29:21] <ckrough> maybe grep all of the zenoss log\nfiles for the name of the zenpack and see if there is an entry that\nshows it completed or failed
[02-Sep-2009 07:30:17] <Apocalipse> hunn
\n[02-Sep-2009 07:40:23] <tripitaka_> a basic zenpack question\nhere.. I've installed the pack ZenPacks.vaibhav.brocadeswitches.\nI have some brocade devices under /Network/FCP. How can I get the\nzenpack to apply to these devices?
[02-Sep-2009 07:51:39] <Apocalipse> not found any entry...
[02-Sep-2009 07:52:22] <Apocalipse> grep -lr "Cisco*" ./log
[02-Sep-2009 07:52:32] <Apocalipse> grep -lr "ZenPack*" ./log
[02-Sep-2009 07:52:40] <Apocalipse> and other options...
[02-Sep-2009 08:03:14] <ckrough> not sure what else to recommend
[02-Sep-2009 08:03:24] <ckrough> are you familiar with 'screen'?
[02-Sep-2009 08:22:39] ke4qqq_ is now known as ke4qqq
[02-Sep-2009 08:22:57] <Apocalipse> me?
[02-Sep-2009 08:24:45] <tripitaka_> Apocalipse: what does this show you: ps ax |grep -i cisco
[02-Sep-2009 08:26:23] <ckrough> Apocalipse: yeah, not that it would help you currently, but in the future.
[02-Sep-2009 08:27:34] <Apocalipse> 58848 p1 D+ 0:00.00 grep -i cisco
[02-Sep-2009 08:27:50] <tripitaka_> that suggests that the install process is not running
\n[02-Sep-2009 08:28:35] <Apocalipse> i was talking with matt...\nand he suggested me to delete all zenpack cisco... and reinstall it
[02-Sep-2009 08:28:46] <tripitaka_> sounds good to me
[02-Sep-2009 08:30:10] <Apocalipse> i am doing that now...
[02-Sep-2009 08:30:14] <Apocalipse> 6-7 hours...
\n[02-Sep-2009 08:31:28] <tripitaka_> if you want to run it\novernight without risking losing your session again, you could either\nuse screen as ckrough suggests, or run `nohup zenpack --install $file\n&`, which will detach and background the process
[02-Sep-2009 08:32:38] <Apocalipse> hunnnnn
[02-Sep-2009 08:32:48] <Apocalipse> good suggestion...
[02-Sep-2009 08:34:32] <Apocalipse> thx
[02-Sep-2009 08:34:39] <tripitaka_> actually, you can do this for your existing process
[02-Sep-2009 08:34:50] <tripitaka_> background it with ctrl-z
[02-Sep-2009 08:35:07] <tripitaka_> and/or use 'detach' to detach the process from the parent shell
[02-Sep-2009 08:37:34] <Apocalipse> yes...
[02-Sep-2009 08:38:17] <tripitaka_> or disown
[02-Sep-2009 08:38:56] <Apocalipse> i forgot that...
[02-Sep-2009 08:39:02] <tripitaka_> http://inglorion.net/software/detach/
[02-Sep-2009 08:39:47] <jb> screen
[02-Sep-2009 08:39:47] <Apocalipse> thx again
[02-Sep-2009 08:39:59] <tripitaka_> +1 for screen
[02-Sep-2009 08:40:13] <tripitaka_> it's missing on a lot of machines, though
[02-Sep-2009 08:47:18] <tripitaka_> how do I associate a zenpack with a device class?
[02-Sep-2009 08:47:26] <tripitaka_> I'm not even sure that question makese sense
[02-Sep-2009 08:47:37] <jb> browse to the device class, click the down arrow, and choose "Add to ZenPack"
[02-Sep-2009 08:48:00] <jb> is that what you mean?
[02-Sep-2009 08:49:00]\n<tripitaka_> I've downloaded a zenpack for brocade switches. I\nhave a device class called /Devices/Network/FibreChannel. I want\nwhatever magic is in the zenpack to be applied to this class.
[02-Sep-2009 08:49:09] <jb> ah.
[02-Sep-2009 08:49:28] <jb> ok, im guessing the zenpack just provides a performance template?
[02-Sep-2009 08:49:36] <jb> for graphs/thresholds/etc?
\n[02-Sep-2009 08:50:10] <jb> browse to the device class, choose\nthe "Templates" tab, click the down arrow and choose "Bindings"
[02-Sep-2009 08:50:15] <tripitaka_> files in zenpack: http://pastebin.com/m2dbb659b
[02-Sep-2009 08:50:28] <jb> in addition to what is already selected, select the Brocade template
\n[02-Sep-2009 08:51:18] <tripitaka_> ok, I've gone to the device\nclass, and clicked templates, then 'Bind Templates..', however I do not\nsee anything new in this list
[02-Sep-2009 08:51:24] <tripitaka_> I'm wondering how this list is populated
[02-Sep-2009 08:51:36] <jb> by templates that are installed by the zenpack
[02-Sep-2009 08:51:45] <tripitaka_> does each zenpack make itself available to specific device classes only?
[02-Sep-2009 08:51:46] <jb> the zenpack maybe installed the template on another device class
[02-Sep-2009 08:52:00] <jb> yes, you would need to copy the template to other classes if it is bound to a specific class
[02-Sep-2009 08:52:03] <tripitaka_> ok, how can I find out which class it has installed itself to?
[02-Sep-2009 08:52:27] <jb> the "This ZenPack Provides" pane
[02-Sep-2009 08:52:32] <jb> should tell you if it created a device class
[02-Sep-2009 08:52:47] <tripitaka_> it says 'No database objects are included in this ZenPack '
[02-Sep-2009 08:53:02] <jb> where did you download this from?
[02-Sep-2009 08:53:32] <tripitaka_> http://www.zenoss.com/community/projects/zenpacks/brocade-switches
[02-Sep-2009 08:55:09]\n<tripitaka_> Looking at another zenpack, JuniperPerformance, the\n'Zenpack Provides' pane gives me a list of datasources and MIBs, but I\ncannot see what device class it has mapped to
\n[02-Sep-2009 08:55:18] <jb> ERROR: zenpack command failed.\nReason: Unable to create object using the following attributes:
[02-Sep-2009 08:55:18] <jb> * id: /zport/dmd/Devices/Blade Switches/rrdTemplates/BrocadeSwitch
[02-Sep-2009 08:55:18] <jb> * module: Products.ZenModel.RRDTemplate
[02-Sep-2009 08:55:24] <jb> did you get this error when you installed it?
[02-Sep-2009 08:55:37] <tripitaka_> I'll install again and check
\n[02-Sep-2009 08:56:22] <tripitaka_> ERROR: zenpack command\nfailed. Reason: [Errno 20] Not a directory:\n'/usr/local/zenoss/zenoss/Products/ZenPacks.vaibhav.brocadeswitches-1.0-py2.4.egg/skins'
[02-Sep-2009 08:56:23] <jb> the author of that zenpack didn't really package it well..
[02-Sep-2009 08:56:57] <jb> do you have a /Devices/Blade Switches/ class?
[02-Sep-2009 08:57:04] <mrayzenoss> tripitaka_: is the ZenPack an .egg or a .zip?
[02-Sep-2009 08:57:12] <jb> its an egg..
[02-Sep-2009 08:57:27] <jb> well, the egg is inside of the zip
[02-Sep-2009 08:57:30] <jb> actually.
[02-Sep-2009 08:57:36] <jb> extract the egg from the zip
[02-Sep-2009 08:58:05] <tripitaka_> hmm, unzipped, getting a different error - I need to create a 'Blade Switches' class
[02-Sep-2009 08:58:09] <jb> yes
[02-Sep-2009 08:58:23] <jb> and then, after installing, your performance template will be bound there
[02-Sep-2009 08:59:29] <Apocalipse> sorry... blue screen again
[02-Sep-2009 09:00:03] <Apocalipse> windows 7 sucks
[02-Sep-2009 09:00:05] <Apocalipse> kkkkkkkkkkkkk
[02-Sep-2009 09:00:07]\n<tripitaka_> right then, much better. The 'Zenpack Provides' pane\nnow shows: http://www.pastebin.ca/1551512. Thing is, which line shows\nme what device class the zenpack is bound to?
[02-Sep-2009 09:00:24] <jb> its bound to /Device/Blade Switches
[02-Sep-2009 09:00:40] <tripitaka_> and I can tell that from the line '/Devices/Blade Switches/rrdTemplates/BrocadeSwitch'?
[02-Sep-2009 09:00:44] <jb> yes
[02-Sep-2009 09:01:02] <tripitaka_> ok, so everything before the 'rrdTemplates' should be taken as a class path
[02-Sep-2009 09:01:30] <jb> sure..
[02-Sep-2009 09:01:50] <tripitaka_> perfect.. copy templates..
[02-Sep-2009 09:03:21] <tripitaka_> bingo
[02-Sep-2009 09:03:29] <tripitaka_> thank you jb
[02-Sep-2009 09:03:32] <jb> np
[02-Sep-2009 09:04:39] <mrayzenoss> I'll look into cleaning up the packaging on that ZenPack
[02-Sep-2009 09:04:45] <mrayzenoss> get it out of the .zip, since that's confusing
[02-Sep-2009 09:05:56] <ckrough> Is there a way to see the device class structure a zenpack requires before running the install?
[02-Sep-2009 09:06:07] <jb> yes
[02-Sep-2009 09:06:08] <ckrough> so that you can create the necessary containers
[02-Sep-2009 09:06:12] <ckrough> do share
[02-Sep-2009 09:06:17] <jb> but keep in mind, that they will be deleted when the zenpack removed..
[02-Sep-2009 09:06:22] <jb> as well as any devices underneath them
[02-Sep-2009 09:07:44] <jb> you can use manage_addDeviceClass in your __init__.py
[02-Sep-2009 09:08:45] <jb> but, typically, as long as the parent class exists, the "Add To ZenPack" on a device class will work fine
[02-Sep-2009 09:08:50] <jb> not sure why it didn't in that case..
[02-Sep-2009 09:09:04] <jb> maybe the author forgot to add the class to the pack, and just added the performance template
[02-Sep-2009 09:18:22] <mrayzenoss> 10 people already signed up for the free training at Ohio LinuxFest
[02-Sep-2009 09:18:43] <markeriv> hmm
[02-Sep-2009 09:18:45] <markeriv> ohio you say
[02-Sep-2009 09:18:48] <markeriv> when?
[02-Sep-2009 09:19:00] <mrayzenoss> http://blog.zenoss.com/2009/08/21/zenoss-community-day-september-25th/
[02-Sep-2009 09:19:04] <mrayzenoss> the Friday before OLF
[02-Sep-2009 09:19:17] <markeriv> ack
[02-Sep-2009 09:19:20] <markeriv> sept 25
[02-Sep-2009 09:19:25] <markeriv> i am getting married on sept 26
[02-Sep-2009 09:19:27] <markeriv> thats a no go
[02-Sep-2009 09:19:30] <mrayzenoss> lol
[02-Sep-2009 09:19:39] <mrayzenoss> I'm sure your fiance will understand
[02-Sep-2009 09:19:41] <jb> haha
[02-Sep-2009 09:19:46] <jb> get your priorities straight!
[02-Sep-2009 09:20:02] <mrayzenoss> we'll have free training at LISA too: http://www.usenix.org/events/lisa09/
[02-Sep-2009 09:20:15] <jb> hm you know what matt
[02-Sep-2009 09:20:24] <jb> i need to talk to someone about sending me some updated training material
[02-Sep-2009 09:20:48] bzed_ is now known as bzed
[02-Sep-2009 09:21:59] <ckrough> November in Baltimore... what were they thinking.
[02-Sep-2009 09:22:56] <mrayzenoss> yeah, last year was San Diego
[02-Sep-2009 09:23:11] <mrayzenoss> Dallas the year before that. it moves around year to year
[02-Sep-2009 09:23:35] <cgibbons> they need to pick somewhere with good weather in november, like houston
[02-Sep-2009 09:23:57] <ckrough> or just park it in San Diego and call it a wrap
[02-Sep-2009 09:24:05] <markeriv> how about seattle
[02-Sep-2009 09:24:11] <ckrough> that would rock
[02-Sep-2009 09:24:22] <markeriv> yea its nice up here
[02-Sep-2009 09:24:31] <markeriv> could do it at the everett convention center
[02-Sep-2009 09:24:36] <cgibbons> if you like wearing a lot of sweaters
[02-Sep-2009 09:24:41] <markeriv> not really
[02-Sep-2009 09:24:45] <markeriv> its 70 degrees today
[02-Sep-2009 09:24:46] <markeriv>
[02-Sep-2009 09:24:49] <Apocalipse> matt do u have a free training schedule?
[02-Sep-2009 09:25:28] <mrayzenoss> well, Ohio is the second time we've done it
[02-Sep-2009 09:25:36] <ckrough> yesterday was our first <100 degree day in over 2 months in Texas.
[02-Sep-2009 09:25:42] <mrayzenoss> and we know we're going to do it at LISA
[02-Sep-2009 09:25:47] <mrayzenoss> and we'll do it at SCALE in LA
[02-Sep-2009 09:25:49] <cgibbons> 70 degrees = sweater weather (plus it ain't november yet, brr)
[02-Sep-2009 09:26:01] <jb> well then you better stay away from chicago
[02-Sep-2009 09:27:21] <markeriv> actually, it doesnt get very cold here mostly
[02-Sep-2009 09:27:29] <markeriv> it stays rather mediumish
[02-Sep-2009 09:27:36] <markeriv> we avoid all extremes
[02-Sep-2009 09:27:43] <markeriv> which is funny, cause seattle people are like that
[02-Sep-2009 09:28:01] <cgibbons> it was sure cold when I was there last march; granted cold is relative but brr.
[02-Sep-2009 09:28:03] <ckrough> Just dont bring up coffee?
[02-Sep-2009 09:28:12] <markeriv> hey!
[02-Sep-2009 09:28:18] <markeriv> coffee was created in seattle
[02-Sep-2009 09:28:23] <markeriv> fact of life
[02-Sep-2009 09:28:30] * ckrough notes the suddent extremism
[02-Sep-2009 09:28:38] <markeriv> :-P
[02-Sep-2009 09:28:47] <markeriv> just kidding
[02-Sep-2009 09:28:52] <markeriv> we are just more addicted to it
[02-Sep-2009 09:29:00] <markeriv> and have unique tastes cause our water is so good
[02-Sep-2009 09:29:27] <markeriv> water quality makes the coffee taste different no matter where you go
[02-Sep-2009 09:29:57] <rmatte> actually, god created coffee, and beer, on the 8th day
[02-Sep-2009 09:30:27] <markeriv> yes, he did... but he did it in seattle
[02-Sep-2009 09:31:13] <ckrough> the Africans may disagree with you there
[02-Sep-2009 09:31:39] <markeriv> pfft
[02-Sep-2009 09:31:59] <Apocalipse> do u have the website of LISA?
[02-Sep-2009 09:32:00] <markeriv> they invented the bean
[02-Sep-2009 09:32:14] <markeriv> we perfected the service of coffee
[02-Sep-2009 09:32:23] <ckrough> www.usenix.org
[02-Sep-2009 09:32:25] <markeriv> just kidding in all this... coffee topic is dead
[02-Sep-2009 09:32:31] <Apocalipse> thx...
[02-Sep-2009 09:32:51] <Apocalipse> do u know the expression "God is from Brazil" ?
[02-Sep-2009 09:33:19] <rmatte>
[02-Sep-2009 09:33:45] <jb> rmatte: were you able to find anything about that aggregate report question I had yesterday?
\n[02-Sep-2009 09:36:55] <tripitaka_> hmm, I just added a number of\ncollector plugins to a device and remodelled. I now have lots of new\ninfo under the Hardware tab, but all my Perf graphs have disappeared.\nHow should I start fixing this?
[02-Sep-2009 09:37:13] <mrayzenoss> all have disappeared?
\n[02-Sep-2009 09:37:59] <tripitaka_> it is a linux server on a HP\nblade, in /Devices/Server/Linux. I did have cpu load, I/O and memory\ngraphs. Now I have no graphs under Perf
[02-Sep-2009 09:39:00]\n<tripitaka_> I added the HPMonitor zenpack, then went to the\ndevice and added community.snmp.HP*, and then remodelled the device
[02-Sep-2009 09:45:30] <tripitaka_> got it, the Device template was unbound
[02-Sep-2009 09:48:16] <tripitaka_> odd, the Device template keeps unbinding itself.
\n[02-Sep-2009 09:58:41] <tripitaka_> procedure: install the\nHPMonitor zenpack. Go to a device and add all the community.snmp.HP*\ncollector plugins, then model the device. Every time I do this, the\nDevice template unbinds and I lose my perf graphs
[02-Sep-2009 10:05:37] <tripitaka_> to get the Device template to stay bound, I have to remove all the HP* collector plugins
[02-Sep-2009 10:13:19] <mrayzenoss> tripitaka_: have you restarted zopectl and zenhub since installing the ZenPack?
[02-Sep-2009 10:14:13] <tripitaka_> yep, I'll do it again
\n[02-Sep-2009 10:35:27] <tripitaka_> rebooted.. I have the Device\ntemplate bound, as well as all the cpq* templates, and I have added all\nthe HP* collector plugins. Now I model the device, then I go back to\ntemplates and I see that the Device template is missing
[02-Sep-2009 10:36:55] <mrayzenoss> and this is using the HP ProLiant Monitor 1.4? Anyone else using this?
[02-Sep-2009 10:41:51] <rmatte> nope
[02-Sep-2009 10:42:09] <rmatte> there are so many HP monitor packs these days I'm losing track of which is which
[02-Sep-2009 10:42:43] <mrayzenoss> yeah, I imagine the 'hp' tag will get a lot of use on the new platform
[02-Sep-2009 10:43:03] <rmatte> yeh
[02-Sep-2009 10:43:15] <rmatte> the new platform will allow for categorization of ZenPacks I hope?
[02-Sep-2009 10:43:34] <rmatte> like, for grouping all MIBs ZenPacks, for instance
[02-Sep-2009 10:43:37] <mrayzenoss> yup
[02-Sep-2009 10:43:42] <tripitaka_> it's ZenPacks.community.HPMon 1.4
[02-Sep-2009 10:43:52] <rmatte> I plan on developing more MIBs ZenPacks when I have time
\n[02-Sep-2009 10:49:13] <rmatte> tripitaka_: yeh, I haven't used\nthat one, so it's dropping the local bindings each time you model?
[02-Sep-2009 10:49:45] <tripitaka_> that's right
[02-Sep-2009 10:49:55] <rmatte> have you tried it via an organizer instead of binding it locally?
[02-Sep-2009 10:49:56] <tripitaka_> I've tried it on two devices, it keeps happening
\n[02-Sep-2009 10:50:45] <tripitaka_> it seems to be related to the\ncollector plugins - if I add the community.snmp.HP* plugins and model,\nit loses Devices
[02-Sep-2009 10:51:10] <rmatte> hmmm
[02-Sep-2009 10:51:35] <rmatte> have you tried making a local copy of Devices?
[02-Sep-2009 10:51:42] <rmatte> instead of binding it?
\n[02-Sep-2009 10:54:38] <tripitaka_> ok, local copy of Devices..\nenabling HP* plugins, model device.. and the Devices template is unbound
\n[02-Sep-2009 10:57:57] <tripitaka_> add the Device template\nagain, remove the HP* plugins, model device.. and the Devices template\nis still bound
[02-Sep-2009 11:11:00] <jb> hrm hey rmatte
[02-Sep-2009 11:41:47] <jb> hrm, i wish zenbackup backed up zenpacks as well
[02-Sep-2009 11:41:52] <jb> anybody here had to deal with this yet?
[02-Sep-2009 11:48:07] <jb> i wish there was a way to export all zenpacks to egg's
[02-Sep-2009 11:48:13] <jb> that you had installed
[02-Sep-2009 11:48:18] <jb> anybody know of a way to do this?
[02-Sep-2009 11:51:11] <mrayzenoss> there's a ticket for zenbackup to include zenpacks
[02-Sep-2009 11:51:18] <jb> that would be awesome
[02-Sep-2009 11:51:29] <jb> im trying to develop a disaster recovery plan for zenoss
[02-Sep-2009 11:51:38] <jb> and the zenpacks just make everything more difficult
[02-Sep-2009 11:54:22] <jb> or even a zenpack --export option
[02-Sep-2009 11:54:57] <jb> mrayzenoss: hrm, do you have that ticket number handy?
[02-Sep-2009 11:55:32] <mrayzenoss> http://dev.zenoss.org/trac/ticket/5180
[02-Sep-2009 11:55:35] <jb> thanks
[02-Sep-2009 11:56:45] <Apocalipse> zenpack --list
\n[02-Sep-2009 11:56:45] <Apocalipse> ZenPacks.Nova.Cisco.Mibs\n(/usr/local/zenoss/ZenPacks/ZenPacks.Nova.Cisco.Mibs-1.0-py2.4.egg)
\n[02-Sep-2009 11:56:45] <Apocalipse> ZenPacks.community.mib_utils\n(/usr/local/zenoss/ZenPacks/ZenPacks.community.mib_utils-1.07-py2.4.egg)
[02-Sep-2009 11:56:49] <Apocalipse> ALELUIA!!!!
[02-Sep-2009 11:57:11] <mrayzenoss> mib-utils 1.08 is out
[02-Sep-2009 11:57:25] <Apocalipse> okok
[02-Sep-2009 11:58:16] <jb> ugh
[02-Sep-2009 11:58:17] <jb> king crab?
[02-Sep-2009 11:58:18] <jb>
[02-Sep-2009 11:58:25] <jb> oh actually
[02-Sep-2009 11:58:28] <jb> thats not too far away
[02-Sep-2009 11:59:27] <cgibbons> hurm
[02-Sep-2009 11:59:56] <jb> coo, looks like the code is already in place
[02-Sep-2009 12:12:47] <jb> ok i need to figure out how to make an aggregate report/graph
[02-Sep-2009 12:12:49] <jb> hrmm
[02-Sep-2009 14:00:18] mrayzenoss1 is now known as mrayzenoss
[02-Sep-2009 14:26:16] <Cosmin> hello
[02-Sep-2009 14:26:19] <Cosmin> so i just did an upgrade
[02-Sep-2009 14:26:39] <Cosmin> actually i did an backup and unninstalled the previous version of zenoss
[02-Sep-2009 14:26:57] <Cosmin> and now ... how can i restore from the backup all the devices and performance data?
[02-Sep-2009 14:27:50] <mrayzenoss> http://www.zenoss.com/community/docs/installation-guide/2.4.2/ch08.html
[02-Sep-2009 14:28:48] <Cosmin> yes i looked at the guide
[02-Sep-2009 14:28:53] <Cosmin> i had 2.3.2 before
[02-Sep-2009 14:29:04] <Cosmin> and i was getting error upgrading to 2.4.5
[02-Sep-2009 14:29:14] <Cosmin> so i thought that a clean install should do it
[02-Sep-2009 14:29:19] <Cosmin> but now im stuck on importing the data
[02-Sep-2009 14:29:47] <mrayzenoss> you'll use zenrestore
[02-Sep-2009 14:30:02] <mrayzenoss> make sure you have any ZenPacks installed in your new install
[02-Sep-2009 14:30:09] <mrayzenoss> before migrating
[02-Sep-2009 14:30:30] <Cosmin> after i use zenrestore and i log back in to the protal it sees the old version
[02-Sep-2009 14:30:39] <Cosmin> and also im getting erros when i start up the application
[02-Sep-2009 14:30:54] <Cosmin> about zenjobs
[02-Sep-2009 14:31:42] <Cosmin> should i restore the old zenoss and try to upgrade it differently?
[02-Sep-2009 14:31:58] <mrayzenoss> which installers are you upgrading with?
[02-Sep-2009 14:32:03] <mrayzenoss> stack to stack?
[02-Sep-2009 14:32:10] <Cosmin> the bin
[02-Sep-2009 14:32:21] <Cosmin> zenoss-stack-2.4.5-linux.bin
\n[02-Sep-2009 14:32:43] <mrayzenoss> you should be able to restore\nto your backup, then just run the .bin and let the stack installer\nupgrade itself
[02-Sep-2009 14:32:49] <mrayzenoss> http://www.zenoss.com/community/docs/installation-guide/2.4.2/ch08s02.html
[02-Sep-2009 14:32:50] <Cosmin> maybe its not possible to upgrade directly from 2.3.2 to 2.4.5?
[02-Sep-2009 14:32:58] <mrayzenoss> lemme check
[02-Sep-2009 14:32:59] <Cosmin> ok
[02-Sep-2009 14:33:22] <mrayzenoss> http://www.zenoss.com/community/docs/installation-guide/2.4.2/ch08s01.html#d4e866
[02-Sep-2009 14:33:36] <mrayzenoss> 2.3.x->2.4.x should work
[02-Sep-2009 14:33:43] <Cosmin> ok let me try again
\n[02-Sep-2009 14:34:36] <mrayzenoss> yeah, release notes say 2.3.x\n-> 2.4.5\nhttp://www.zenoss.com/news/archive/zenossannouncement.2009-09-01.5195585629#d4e118
[02-Sep-2009 14:46:47] <Cosmin> ok
[02-Sep-2009 14:46:51] <Cosmin> this is what im getting
[02-Sep-2009 14:46:53] <Cosmin> Warning: Problem running post-install step. Installation may not complete
[02-Sep-2009 14:46:53] <Cosmin> correctly
[02-Sep-2009 14:46:53] <Cosmin> Error running su zenoss -c '"/usr/local/zenoss/zenoss/bin/zenmigrate" ' :
[02-Sep-2009 14:46:53] <Cosmin> INFO:zen.migrate:The following migrate steps were not intended to run with the
[02-Sep-2009 14:46:53] <Cosmin> currently installed version of the Zenoss code. The installed version is 2.4.5.
[02-Sep-2009 14:46:54] <Cosmin> You can override this warning with the --force option.
[02-Sep-2009 14:46:56] <Cosmin> DifferentiateTemplates (2.3.1)
[02-Sep-2009 14:46:58] <Cosmin> Recovering
[02-Sep-2009 14:47:00] <Cosmin> Press [Enter] to continue :
[02-Sep-2009 14:47:45] <Cosmin> any clues?
[02-Sep-2009 14:50:10] <mrayzenoss> anything else? I'm running it by QA
[02-Sep-2009 14:51:27] <Cosmin> thats what i got in when i try to upgrade it
[02-Sep-2009 14:51:37] <Cosmin> you want me to paste the whole thing in a pastebin?
[02-Sep-2009 14:52:32] <mrayzenoss> sure
[02-Sep-2009 14:53:26] <Cosmin> here you go
[02-Sep-2009 14:53:26] <Cosmin> http://pastebin.com/m6465dd3c
[02-Sep-2009 14:55:42] <Cosmin> let me know if there is anything else that you want me to include
[02-Sep-2009 15:00:24] <mrayzenoss> QA referred me to this: http://dev.zenoss.org/trac/ticket/5164
[02-Sep-2009 15:02:31] <Cosmin> ok i was missing net-snmp-utils let me try that again
[02-Sep-2009 15:02:47] <mrayzenoss> well, the zenmigrate --force is what's needed
[02-Sep-2009 15:02:55] <mrayzenoss> to fix the upgrade
[02-Sep-2009 15:03:01] <mrayzenoss> there was a bug in 2.3.1 apparently
[02-Sep-2009 15:04:34] <Cosmin> is there a fix for that bug?
[02-Sep-2009 15:04:41] <Cosmin> maybe i8 can patch it before i try to upgrade it?
[02-Sep-2009 15:05:01] <mrayzenoss> well the zenmigrate --force should clean up the template issue
[02-Sep-2009 15:05:41] <mrayzenoss> basically there was a migrate error, so we're forcing it
[02-Sep-2009 15:05:46] <mrayzenoss> rather than let it linger
[02-Sep-2009 15:12:31] <Cosmin> allright that seems like tis doing something now
\n[02-Sep-2009 15:12:42] <Cosmin> im up to INFO:zen.migrate:This\ncan take a VERY long time to execute(eg hours). Please do not\ninterrupt.
[02-Sep-2009 15:25:55] <Cosmin> allright that completed
[02-Sep-2009 15:28:10] <daMaestro> anyone know why snmp trap 1.3.6.1.6.3.1.1.5.3 is not mapping correctly in 2.4.5?
[02-Sep-2009 15:29:31] <daMaestro> err sorry, 2.4.2
[02-Sep-2009 15:43:31] <Cosmin> anyone seen this before?
[02-Sep-2009 15:43:33] <Cosmin> zenpack --remove=ZenPacks.zenoss.ZenJMX
[02-Sep-2009 15:43:33] <Cosmin> ERROR: zenpack command failed. Reason: 'Java'
[02-Sep-2009 16:07:31] <mrayzenoss> finally getting new ZenPacks out
[02-Sep-2009 16:07:44] <mrayzenoss> WMI Terminal Server: http://www.zenoss.com/community/projects/zenpacks/wmiperf_terminalserver
[02-Sep-2009 16:08:03] <mrayzenoss> WMI Exchange 2003: http://www.zenoss.com/community/projects/zenpacks/wmiperf_exchange2003
[02-Sep-2009 16:09:19] <Cosmin> awsome
[02-Sep-2009 16:13:21] <Cosmin> allright
[02-Sep-2009 16:13:24] <Cosmin> one little problem
[02-Sep-2009 16:13:34] <Cosmin> when i ack an alert
[02-Sep-2009 16:13:44] <Cosmin> after a while it cleares and then it comes back
[02-Sep-2009 16:14:24] <mrayzenoss> events that do not re-occur get cleared after 4 hours
[02-Sep-2009 16:14:32] <Cosmin> instead of them staing acknowledged they are keep comming back
[02-Sep-2009 16:14:54] <Cosmin> they are reocrurring durring a 5 minute period
[02-Sep-2009 16:15:03] <mrayzenoss> that's odd
[02-Sep-2009 16:15:23] <Cosmin> hm.. not sure where to look to see whats wrong
[02-Sep-2009 16:15:51] <mrayzenoss> is the count increasing?
[02-Sep-2009 16:16:07] <Cosmin> nope
[02-Sep-2009 16:16:08] <mrayzenoss> it's switching them from Acknowledged back to New?
[02-Sep-2009 16:16:10] <Cosmin> its only 1
[02-Sep-2009 16:16:18] <mrayzenoss> hmmm... so it thinks they're different
[02-Sep-2009 16:16:22] <Cosmin> im not sure if its switching them
[02-Sep-2009 16:16:34] <Cosmin> but it looks like that
[02-Sep-2009 16:16:44] <Cosmin> the old ones are not there anymore
[02-Sep-2009 16:18:09] <mrayzenoss> did you 'restart' or 'start' after your migration?
[02-Sep-2009 16:18:20] <mrayzenoss> there's a bug about 'restart' skipping an upgrade step
[02-Sep-2009 16:18:54] <Cosmin> yes
[02-Sep-2009 16:19:07] <Cosmin> i also rebooted the whole box
[02-Sep-2009 16:20:19] <mrayzenoss> what kind of events? snmp, ssh...?
[02-Sep-2009 16:20:32] <Cosmin> mm..
[02-Sep-2009 16:20:43] <Cosmin> they are failed checks for http check
[02-Sep-2009 16:20:48] <mrayzenoss> or does it not matter?
[02-Sep-2009 16:21:49] <mrayzenoss> more ZenPacks
\n[02-Sep-2009 16:21:50] <mrayzenoss> WMI Terminal Services:\nhttp://www.zenoss.com/community/projects/zenpacks/wmiterminalservices
[02-Sep-2009 16:22:05] <mrayzenoss> WMI Exchange 2003 Perf http://www.zenoss.com/community/projects/zenpacks/wmiexchange2003
[02-Sep-2009 16:22:09] <Cosmin> hmm
[02-Sep-2009 16:22:17] <Cosmin> im getting the same ones over and over again
[02-Sep-2009 16:22:30] <Cosmin> these are only new ones that showed up after the upgrade
[02-Sep-2009 16:22:38] <Cosmin> the old ones that i had in there they remained acknowledged
[02-Sep-2009 16:23:04] <Cosmin> so far i get some email checks failing and the http_check failing
[02-Sep-2009 16:29:25] <Cosmin> hmm any hints?
[02-Sep-2009 16:33:07] <mrayzenoss> Cosmin: are the Acknowledged events in Event History?
[02-Sep-2009 16:34:22] <mrayzenoss> they're disappearing right? Not becoming "Unacknowledged"?
[02-Sep-2009 16:34:26] <mrayzenoss> add a note to the log
[02-Sep-2009 16:34:49] <Cosmin> hmm..
[02-Sep-2009 16:34:50] <Cosmin> they are
[02-Sep-2009 16:34:57] <Cosmin> and there is a clear after
[02-Sep-2009 16:35:07] <Cosmin> not sure how it clears do
[02-Sep-2009 16:35:22] <mrayzenoss> the clear event moves it out of the event console
[02-Sep-2009 16:35:26] <mrayzenoss> and into history
[02-Sep-2009 16:35:32] <Cosmin> right
[02-Sep-2009 16:35:40] <Cosmin> so the check is checking for ssl
[02-Sep-2009 16:35:40] <mrayzenoss> what's clearing them?
[02-Sep-2009 16:36:39] <Cosmin> im not sure
[02-Sep-2009 16:36:42] <Cosmin> i dont see anything
[02-Sep-2009 16:36:53] <Cosmin> when i click on the details i see http_content_check
[02-Sep-2009 16:36:57] <Cosmin> and thats whats making them
[02-Sep-2009 16:37:04] <Cosmin> im not sure how they become clear
[02-Sep-2009 16:37:12] <Cosmin> because i checked the website and it doesnt have ssl enabled
[02-Sep-2009 16:37:18] <mrayzenoss> what's causing the event? Command failing?
[02-Sep-2009 16:37:23] <mrayzenoss> bad credentials?
[02-Sep-2009 16:38:25] <Cosmin> connection refused
[02-Sep-2009 16:38:34] <Cosmin> because there is nothing listening on ssl
[02-Sep-2009 16:39:39] <Cosmin> maybe change the cycle time for the command?
[02-Sep-2009 16:39:44] <Cosmin> its currently set to 300
[02-Sep-2009 16:40:02] <mrayzenoss> yeah, but what's clearing it?
[02-Sep-2009 16:40:17] <mrayzenoss> are other, non-command calls generating events?
[02-Sep-2009 16:40:53] <Cosmin> eventClass /Cmd/Fail
[02-Sep-2009 16:40:53] <Cosmin> eventKey http_content_check
[02-Sep-2009 16:40:53] <Cosmin> summary Clear
[02-Sep-2009 16:41:11] <Cosmin> manager localhost
[02-Sep-2009 16:41:12] <Cosmin> agent zencommand
[02-Sep-2009 16:42:08] <Cosmin> seems like they are generated my custom made templates
[02-Sep-2009 16:42:16] <Cosmin> using nagios plugins
[02-Sep-2009 16:44:05] <Cosmin> decreasing the time didn help
[02-Sep-2009 16:51:04] <Cosmin> Available Zenoss Version 2.3.2
[02-Sep-2009 16:51:17] <Cosmin> this is what i have by version by settings
[02-Sep-2009 16:57:47] <mrayzenoss> something tells me your upgrade did not go smoothly...
[02-Sep-2009 16:58:36] <rmatte> no, he probably doesn't have the checkboxes checked off
[02-Sep-2009 16:58:42] <rmatte> and he's looking at the last available version it saw
[02-Sep-2009 16:59:02] <rmatte> Cosmin: look down the page further and you'll see the actual Zenoss version
\n[02-Sep-2009 16:59:18] <mrayzenoss> don't remember who was\nlooking at it earlier, but I fixed the Brocade Switches page:\nhttp://www.zenoss.com/community/projects/zenpacks/brocade-switches
\n[02-Sep-2009 16:59:23] <rmatte> and if you put a checkmark in the\ncheckbox to periodically check for updated version, then click update,\nyou should see the latest
[02-Sep-2009 17:00:30] <rmatte> bah, this map thing is going to drive me nuts
[02-Sep-2009 17:00:35] daMaestro|isBack is now known as daMaestro
[02-Sep-2009 17:00:37] <Cosmin> i do see the actual version by zenoss
[02-Sep-2009 17:00:49] <Cosmin> but thats the available version even if i check for updates
[02-Sep-2009 17:01:19] <Cosmin> i still have a backup of the old version
[02-Sep-2009 17:01:35] <Cosmin> if there is an another path of doing this i am willing to try it
[02-Sep-2009 17:05:05]\n<mrayzenoss> Cosmin: I guess you could try something along these\nlines:\nhttp://www.zenoss.com/community/docs/installation-guide/2.4.2/ch09s01.html
[02-Sep-2009 17:05:26] <mrayzenoss> do a zenbackup, a clean install, then zenrestore
[02-Sep-2009 17:05:39] <mrayzenoss> make sure you get your ZenPacks in place before the zenrestore step
[02-Sep-2009 17:10:09] <Cosmin> should i use the same versions of zenpacks?
[02-Sep-2009 17:10:19] <Cosmin> or i can install the latest ones?
[02-Sep-2009 17:11:34] <mrayzenoss> which ones are you using? Latest is probably fine
[02-Sep-2009 17:12:51] <Cosmin> zenjmx and some of t
[02-Sep-2009 17:13:05] <Cosmin> are you gonna be online tomorrow?
[02-Sep-2009 17:13:15] <Cosmin> its kinda of late today... id like to go home
[02-Sep-2009 17:16:09] <mrayzenoss> I'm usually around
[02-Sep-2009 17:17:34] <Cosmin> allright
[02-Sep-2009 17:17:37] <Cosmin> thanks for your help
[02-Sep-2009 17:35:35] <ironpaw> hi
[02-Sep-2009 17:37:15] <markeriv> yo ironpaw
[02-Sep-2009 17:37:16] <markeriv> how goes it?
[02-Sep-2009 17:40:20] <ironpaw> yeh good mate yerself?
[02-Sep-2009 17:40:24] <ironpaw> almost friday
[02-Sep-2009 17:40:27] <ironpaw> Thu Sep 3 08:40:27 EST 2009
[02-Sep-2009 17:40:32] <ironpaw> almost but not quite
[02-Sep-2009 17:40:34] <ironpaw> hows things with you?
\n[02-Sep-2009 20:49:24] <gekitsuu> Are there any Zenoss people \nhere in contact with someone who's at the booth at the Red Hat Summit ?
[02-Sep-2009 22:55:26] davetoo1 is now known as DaveToo
[02-Sep-2009 23:34:15] <djjay> hi. I'm having problems with the linux ssh commands. df isn't logging any data
[02-Sep-2009 23:34:29] <djjay> hopefully you aren't all asleep
[03-Sep-2009 00:00:46] [disconnected at Thu Sep 3 00:00:46 2009]
[03-Sep-2009 00:00:46] [connected at Thu Sep 3 00:00:46 2009]
[03-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[03-Sep-2009 00:37:02] <ironpaw> o.O
[03-Sep-2009 01:26:44] [disconnected at Thu Sep 3 01:26:44 2009]
[05-Sep-2009 00:00:46] [connected at Sat Sep 5 00:00:46 2009]
[05-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[05-Sep-2009 08:51:15] <Apocalipse> hello people
[05-Sep-2009 09:09:44] <Apocalipse> all quiet
[05-Sep-2009 09:09:45] <Apocalipse>
[06-Sep-2009 00:00:46] [disconnected at Sun Sep 6 00:00:46 2009]
[06-Sep-2009 00:00:46] [connected at Sun Sep 6 00:00:46 2009]
[06-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
\n[06-Sep-2009 04:54:11] <InitX> Hi all , I'm a newbie to Zenoss\nand I had read the zenoss document. In the "Getting Started with\nZenoss" there is a google map display in Zenoss Dashboard how to let my\nzenoss display that google map?
[06-Sep-2009 05:58:42] <Simon_2> hello all
[06-Sep-2009 05:59:27] <Simon_2> !rules
[06-Sep-2009 05:59:49] <Simon_2> nobody here then
[06-Sep-2009 06:27:40] <Mavrick^^> hi guys
[06-Sep-2009 06:51:45] <Mavrick^^> can some onetell me where is mib folder?
[06-Sep-2009 06:52:10] <Mavrick^^> create a "Juniper" Mibs sub-folder
[06-Sep-2009 06:52:14] <Mavrick^^> where shuld i create it
\n[06-Sep-2009 06:59:43] <idl0r> hey guys, are there any plans for\nadding rules to _not_ use all the provided external libs?
[06-Sep-2009 07:02:43] <Mavrick^^> create a "Juniper" Mibs sub-folder
[06-Sep-2009 07:02:44] <Mavrick^^> where shuld i create it
[06-Sep-2009 07:15:12] <Mavrick^^> Ok i got it
[06-Sep-2009 14:31:03] <tpdd> is it possible to access a in zenoss directly through a link so it can be embeded in a different html page?
[06-Sep-2009 14:31:22] <tpdd> access a graph directly
[06-Sep-2009 14:38:25] <Appiah> I dont see why you wanna do that but sure why not?
[06-Sep-2009 14:38:47] <Appiah> you will have to keep the session info that you are logged in ...
\n[06-Sep-2009 14:40:33] <tpdd> Right now I have mrtg setup to show\nthe traffic on a couple of interfaces and Im just trying to replicate\nthat without giving logins to the 10 people that need to see it.
[06-Sep-2009 14:42:59] <Appiah> you could probly setup some scripts providing that info
[06-Sep-2009 14:43:59] <tpdd> like to copy the graphs out to where they can be viewed?
[06-Sep-2009 20:01:53] <ironpaw> hi
[06-Sep-2009 20:01:57] <ironpaw> how are we all today?
[06-Sep-2009 22:23:46] <ironpaw> o.O
[06-Sep-2009 22:23:50] <ironpaw> hai
[06-Sep-2009 22:24:02] <ironpaw> you should get an eggdrop bot in here
[06-Sep-2009 22:24:09] <ironpaw> it'd give me someone to talk to
[06-Sep-2009 22:35:22] <Diddi> (:
[06-Sep-2009 22:38:13] <ironpaw> #drupal have an eggdrop greetings and faqs
[07-Sep-2009 00:00:46] [disconnected at Mon Sep 7 00:00:46 2009]
[07-Sep-2009 00:00:46] [connected at Mon Sep 7 00:00:46 2009]
[07-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[07-Sep-2009 00:24:29] <ironpaw> hi
[07-Sep-2009 01:00:02] <ironpaw> is it possible to get DISK IO per filesystem?
[07-Sep-2009 06:32:16] quadro_ is now known as Quadro
[07-Sep-2009 09:54:00] <malbon> ironpaw: Which OS are you monitoring?
[07-Sep-2009 13:48:59] <djack> hellow world
[07-Sep-2009 13:49:16] <djack> somebody living here ?
[07-Sep-2009 14:03:21] <djack> nobody ?
[07-Sep-2009 14:57:44] <djack> wow netslplit
[07-Sep-2009 15:17:47] [disconnected at Mon Sep 7 15:17:47 2009]
[09-Sep-2009 00:00:46] [connected at Wed Sep 9 00:00:46 2009]
[09-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
[09-Sep-2009 06:54:03] <fuzz> anyone successfully using zenoss behind an apache vhost with ssl?
\n[09-Sep-2009 06:54:14] <fuzz> i've set up stuff like this:\nhttp://www.zenoss.com/community/docs/howtos/setup-zenoss-with-apache/
[09-Sep-2009 06:54:39] <fuzz> but the login <form> still has action=http://xxxx (not https or a relative url)
[09-Sep-2009 06:55:01] <fuzz> so it POSTs to http, which redirects to https again, and in that process it seems to loose the post data
\n[09-Sep-2009 06:55:41] <fuzz> it seems possible to change the url\nof the server to https://xxx for alert mails in zenactions.conf, but\nthis does not seem to affect the login form url
[09-Sep-2009 06:55:50] <fuzz> is there another config setting to change that url?
\n[09-Sep-2009 07:42:27] <Dieterbe> Hey guys. I have a custom\ntemplate, which has a zencommand datasource which executes a script. i\nused to get notifications (and events) when this script exited with\ncode >0. but since a while not anymore (i did upgrade from 2.4.1 to\n2.4.3). i'm not even getting events. i tried setting severity to\nwarning/error/critical, parser auto/nagios and disabled all my event\ntransforms
[09-Sep-2009 07:42:32] <Dieterbe> any idea?
[09-Sep-2009 07:44:09]\n<Dieterbe> oh and i replaced my normal command with execution of\na very simple script, which i tried both exit(1) and exit(2). and it\nlogs when it's invoked, so by looking at the log i see it's executed\nevery 30s (cycle time), but no events ..
[09-Sep-2009 07:56:55] <IoNu|z> hello
[09-Sep-2009 07:57:03] <Dieterbe> hmm just reading about the 2.4.5. I'll try upgrading..
[09-Sep-2009 07:57:08] <Dieterbe> hi IoNu|z
[09-Sep-2009 07:57:16] <IoNu|z> does anybody have issues with seeing custom OS products?
[09-Sep-2009 08:05:11] <skipzoid> hello everyone
[09-Sep-2009 08:06:16] <skipzoid> no body here
[09-Sep-2009 08:08:09] <venturaville> still pretty early ...it will wake up in another hour or so
\n[09-Sep-2009 08:13:47] <skipzoid> ah, my question was - what\ndistro would give the least resistence when trying to install zenoss on\nit? - we're slackware users here - but its fighting back at every turn,\nwmi failing - etc, also theres a python error that comes up with trying\nto add a device - i'm happy to download and install another distro -\nbut what one?
[09-Sep-2009 08:16:41] <skipzoid>\nwhat are people using here and how much of a fiddle was it to get\nzenoss installed on it
[09-Sep-2009 08:17:31] <venturaville> redhat and ubuntu seem to work pretty well
[09-Sep-2009 08:17:41] <venturaville> you are on your own with pretty much everything with slackware
[09-Sep-2009 08:18:12] <venturaville> trust me... I was using slackware back when I could run it off a single floppy
[09-Sep-2009 08:18:35] <skipzoid> well we're all pretty rounded here with slackware - but this zenoss is getting us beat,
\n[09-Sep-2009 08:19:12] <venturaville> your best bet if you are\nsticking to slackware is to use one of the packaged zenoss installs
\n[09-Sep-2009 08:19:14] <skipzoid> we've got several slackware\nboxes here and dotted around the company in different offices
[09-Sep-2009 08:19:50] <skipzoid> is a packed zenoss installed different from a core?
[09-Sep-2009 08:19:54] <skipzoid> which is compiled?
[09-Sep-2009 08:20:02] <venturaville> nope same thing, but all packaged with its dependencies
[09-Sep-2009 08:20:06] <venturaville> let me see if I can find a link
[09-Sep-2009 08:20:36] <venturaville> http://www.zenoss.com/download/links?creg=null
[09-Sep-2009 08:20:41] <venturaville> I believe what you want is one of the stack installers
[09-Sep-2009 08:20:56] <skipzoid> just having a look
[09-Sep-2009 08:21:32] <skipzoid> obviosuly there isnt a slackware one
[09-Sep-2009 08:21:43] <venturaville> fraid not
[09-Sep-2009 08:21:49] <venturaville> debian 5 might be worth a try
[09-Sep-2009 08:21:56] <skipzoid> just thinking the same
[09-Sep-2009 08:22:05] <skipzoid> its pretty close in its layout
[09-Sep-2009 08:23:06] <skipzoid> will have a look and come back later - thanks very much for your help
[09-Sep-2009 08:28:54] kevin7kal_ is now known as kevin7kal
[09-Sep-2009 09:03:14] <metalklesk> Hello everyone
[09-Sep-2009 09:22:23] <metalklesk> does anyone knows if you can monitor tandem server with zenoss ?
[09-Sep-2009 09:23:02] <Appiah> does it use snmp ?
[09-Sep-2009 09:24:02] <rmatte> metalklesk: It's a web managed tool? Basically a CMS?
[09-Sep-2009 09:24:46] <rmatte> metalklesk: what aspects of it are you hoping to monitor?
[09-Sep-2009 09:31:11] <metalklesk> i'm talking about OS
[09-Sep-2009 09:31:17] <metalklesk> HP Tandem
[09-Sep-2009 09:32:04]\n<metalklesk> we need to know if we can monitor aspects of SO like\nprocess or services .. obviusly some hardware monitoring too like cpu,\nmemory and filesystems
[09-Sep-2009 09:32:14] <metalklesk> SO == OS
[09-Sep-2009 09:32:28] <metalklesk> Operative System
[09-Sep-2009 09:32:30] <metalklesk>
[09-Sep-2009 09:33:00] <ckrough> are these older boxes?
[09-Sep-2009 09:33:22] <metalklesk> i have seen in zenoss that i can monitor any process i want by adding a "process"
[09-Sep-2009 09:33:31] <metalklesk> yes, it is an old box
\n[09-Sep-2009 09:34:14] <ckrough> it mostly depends on your OS. If\nyou're OS provides decent SNMP information and/or ssh or other remote\naccess, then you can monitor just about anything. It may require\nlearning and scripting on your end to get there though.
[09-Sep-2009 09:34:33] <ckrough> "your" OS that is
[09-Sep-2009 09:36:57] <rmatte> metalklesk: yeh, but is Tandem the actual OS?
[09-Sep-2009 09:37:06] <rmatte> or is it something running on top of like Linux or something?
[09-Sep-2009 09:37:56] <mrayzenoss> Tru64?
[09-Sep-2009 09:38:48] <chudler> Tandem OS/Guardian ?
[09-Sep-2009 09:39:06] <rmatte> ah, TOS apparently
\n[09-Sep-2009 09:39:25] <rmatte> as long as it can provide\nrelatively standard SNMP output then you can monitor basically anything
\n[09-Sep-2009 09:39:57] <rmatte> you basically want to enable\nSNMP, enter the community string in to Zenoss, model the device, and\nsee what you ge
[09-Sep-2009 09:39:59] <rmatte> get*
[09-Sep-2009 09:44:09] <metalklesk> yes, i guess i should try
[09-Sep-2009 09:44:27] <rmatte> yup
[09-Sep-2009 09:45:40] <metalklesk> right now BMC Patrol cannot monitor TOS
[09-Sep-2009 09:46:53] <rmatte> Zenoss is pretty flexible, but some dev work might be required (custom scripts)
[09-Sep-2009 09:47:14] <rmatte> depending on what you can get TOS to spit out
[09-Sep-2009 09:49:43] <metalklesk> ok
\n[09-Sep-2009 09:52:49] <metalklesk> we were seeing\nhttp://www.exchangemonitoring.com.uk/Solutions/HPNonStopManagement/tabid/108/Default.aspx
[09-Sep-2009 09:53:15] <rmatte> I can't even hit that page
[09-Sep-2009 09:53:20] <rmatte> address no found
[09-Sep-2009 09:53:23] <rmatte> not*
[09-Sep-2009 09:53:29] <metalklesk> but i like zenoss and i want to try it before looking other solutions
[09-Sep-2009 09:53:54] <rmatte> I've worked with a few different monitoring tools and Zenoss is definitely my favourite
[09-Sep-2009 09:54:18] <metalklesk> i have worked with patrol an portal, some with nagios too
[09-Sep-2009 09:54:29] <metalklesk> and zenoss is my favorite too
[09-Sep-2009 09:54:44] <rmatte> well, Zenoss certainly blows Nagios out of the water
[09-Sep-2009 09:54:48] <rmatte> hehe
[09-Sep-2009 09:55:00] <metalklesk> http://www.exchangemonitoring.co.uk/Solutions/HPNonStopManagement/tabid/108/Default.aspx
[09-Sep-2009 09:55:08] <rmatte> ah, co.uk hehe
[09-Sep-2009 09:55:13] <metalklesk> yup
[09-Sep-2009 09:55:16] <metalklesk> my mistake
[09-Sep-2009 09:56:32] <rmatte> the reflex stuff is nice
[09-Sep-2009 09:57:06] <rmatte> but realistically, it's generally better to have an actual human being resolve issues
[09-Sep-2009 09:57:31] <metalklesk> why?
[09-Sep-2009 09:57:48] <rmatte> well, for simpler stuff, like restarting processes, I suppose it would be fine
\n[09-Sep-2009 09:58:58] <rmatte> because automatically fixing\nsomething can end up being just a band-aid, it's nice to have someone\nactually go in and investigate as to what caused the issue in the first\nplace when it happens, rather than just restart a process automatically\nand forget about it
[09-Sep-2009 09:59:01] <rmatte> that's my view anyways
[09-Sep-2009 09:59:38] <rmatte> I suppose from a response time aspect, automation does have it's advantages
[09-Sep-2009 09:59:45] <metalklesk> ahh yup
[09-Sep-2009 09:59:49] <rmatte> with dev work it's possible to automate stuff with Zenoss using custom scripts
[09-Sep-2009 10:04:21]\n<chudler> in my experience, automatic "healing", response,\nreflex, etc are more marketing tools than anything else. apealing to\nPHB's and sales-drones
[09-Sep-2009 10:04:30] <chudler> ok, I'll get off my high horse :-)
[09-Sep-2009 10:05:07] <rmatte> chudler: it's true though, it's ineffective for anything other that very simple issues
[09-Sep-2009 10:06:16]\n<mrayzenoss> someone told me they were using a zencommand to ssh\nto the box in question, and scrape every log and the status of the\nmachine when a certain event came in. Not autoremediation, but a good\nidea to snapshot a box in trouble
[09-Sep-2009 10:07:25] <rmatte> mrayzenoss: yeh, that's basically auto-reconnaissance
[09-Sep-2009 10:07:33] <mrayzenoss> yeah
[09-Sep-2009 10:07:34] <rmatte> which is actually a really nice idea
[09-Sep-2009 10:07:43] <rmatte> but auto break/fix is not such a great idea all of the time
[09-Sep-2009 10:07:47] <rmatte> hehe
[09-Sep-2009 10:07:58] <mrayzenoss> not unless you really nail down the error and the work-around is well known
[09-Sep-2009 10:08:07] <rmatte> yeh, exactly
[09-Sep-2009 10:08:34] <metalklesk> i agree
[09-Sep-2009 10:08:49] <rmatte> mrayzenoss: when is the 2.5 beta dropping?
[09-Sep-2009 10:09:27] <mrayzenoss> working on the beta today
[09-Sep-2009 10:09:37] <rmatte> so release is expected for tomorrow?
[09-Sep-2009 10:09:37] <mrayzenoss> announcements will be on the usual sites
[09-Sep-2009 10:09:52] <rmatte> cool
[09-Sep-2009 10:09:54] <rmatte> afk for a few
[09-Sep-2009 10:09:58] <mrayzenoss> yeah, today or tomorrow
[09-Sep-2009 10:10:04] <mrayzenoss> probably
[09-Sep-2009 10:10:08]\n<MajereDB8> hi, i haven't dropped by the dev channel in a while,\nbut i've been working on a project using the zenpop3 module and am\nhaving some issues getting it to work with openssl. i've checked TRAC\nand the boards but haven't found a solution. might anyone have a\nminute or so?
[09-Sep-2009 10:10:56] <metalklesk> one question, wich ubuntu version you can monitor with the ubuntu zenpack ?
[09-Sep-2009 10:11:31] <metalklesk> i'm trying 9.04 but zenoss cannot get the installed software list
[09-Sep-2009 10:11:51] <metalklesk> the other things like cpu and memory yes
[09-Sep-2009 10:12:24] <IoNu|z> does anybody have this issue with os tab?
[09-Sep-2009 10:12:28] <metalklesk> NOOO, nevermind, it is working
[09-Sep-2009 10:12:30] <IoNu|z> like the OS versions dont show up?
[09-Sep-2009 10:12:55] <IoNu|z> if i add another os version or hardware
[09-Sep-2009 10:13:00] <metalklesk> i have re model it and now is showing the installed software, its cool !
[09-Sep-2009 10:13:01] <IoNu|z> it doesnt show up in the fields
[09-Sep-2009 10:13:18] <IoNu|z> but it does show up by /products
[09-Sep-2009 10:13:29] <IoNu|z> is there a way to forcely poplate it?
[09-Sep-2009 10:13:47] <IoNu|z> http://forums.zenoss.com/viewtopic.php?t=10982
[09-Sep-2009 10:18:00] <IoNu|z> anyone any clues?
[09-Sep-2009 10:23:31] <Brad_K> Sorry, i'm still just a white belt.
[09-Sep-2009 10:27:25] <metalklesk> i don't understand your problem
[09-Sep-2009 10:27:27] <metalklesk>
\n[09-Sep-2009 10:35:50] <rmatte> metalklesk: the software list\njust depends on dpkg and aptitude, do I'm surprised it's not able to\npopulate it
[09-Sep-2009 10:36:18] <metalklesk> no no, it is working perfectly
[09-Sep-2009 10:36:19] <rmatte> IoNu|z: what version of Zenoss are you using?
[09-Sep-2009 10:36:41] <metalklesk> thanks rmatte
[09-Sep-2009 10:37:02] <metalklesk> zenoss is receiving the installed software list perfectly
[09-Sep-2009 10:37:10] <rmatte> cool
[09-Sep-2009 10:37:48] <metalklesk> the only thing now is that zenmodeler failed when i modeled the device
[09-Sep-2009 10:38:11] <rmatte> failed on what?
[09-Sep-2009 10:38:21] <metalklesk> Zenmodeler - /ÂCmd/ÂFail - User timeout caused connection failure
[09-Sep-2009 10:38:41] <metalklesk> i don't know what this means
[09-Sep-2009 10:38:45] <rmatte> that means that your SSH connection timed out somehow I believe
[09-Sep-2009 10:39:04] <rmatte> (You're modelling partially via SSH I assume?)
[09-Sep-2009 10:39:08] <metalklesk> i guess so
[09-Sep-2009 10:39:09] <metalklesk> yes
[09-Sep-2009 10:39:18] <metalklesk> using Ubuntu zenpack
[09-Sep-2009 10:39:23] <rmatte> make sure the username and password that are set in zProperties are correct
[09-Sep-2009 10:39:32] <metalklesk> i'm just trying that zenpack
[09-Sep-2009 10:39:39] <metalklesk> yes, credentials are ok
\n[09-Sep-2009 10:39:45] <rmatte> IoNu|z:\nhttp://dev.zenoss.org/trac/ticket/5473 <--- This is probably why (I\nassume you're using 2.4.5)
[09-Sep-2009 10:40:12] <metalklesk> in fact i can see the installed software, cpu, memory, filesystems, etc
[09-Sep-2009 10:40:23] <metalklesk> i don't see problems
\n[09-Sep-2009 10:40:25] <rmatte> IoNu|z: Also, when you add\nanother OS version or something it doesn't go to where you requested to\nadd it, it gets added to /Unknown I believe
[09-Sep-2009 10:40:40] <rmatte> It's a bug
[09-Sep-2009 10:41:09] <rmatte> metalklesk: then it's probably not an issue
[09-Sep-2009 10:41:21] <metalklesk> yes i think it is not
[09-Sep-2009 10:41:49] <rmatte> I think IoNu|z dozed off lol
[09-Sep-2009 10:42:15] dB8r is now known as MajereDB8
[09-Sep-2009 10:43:13] <MajereDB8> question, do you need python-twisted installed to get pyopenssl to work?
\n[09-Sep-2009 10:43:45] <metalklesk> a newbie question: in OS tab\n-> Interfaces, what does mean O A M Lock \n??
[09-Sep-2009 10:44:16] <metalklesk> only M has the green circle, O and A has a red one
[09-Sep-2009 10:46:27] <rmatte> metalklesk: O = Operational Status, A = Administrative Status, M = Monitored (yes/no)
\n[09-Sep-2009 10:46:50] <rmatte> metalklesk: in King Crab the O\nand the A are going to be replaced for a single "Status" indicator
\n[09-Sep-2009 10:47:34] <rmatte> metalklesk: also, the statuses\nonly get updated when you model a device right now, but in King Crab\nthey will reflect the realtime status of each interface
[09-Sep-2009 10:48:29] <metalklesk> ok, thanks
[09-Sep-2009 10:49:08] <rmatte> np
[09-Sep-2009 10:49:55] <rmatte> MajereDB8: no idea
[09-Sep-2009 10:50:03]\n<metalklesk> if you click in one interface (os tab ->\ninterfaces) like eth0, can you see some graphics ? i mean with some\nmetrics
[09-Sep-2009 10:50:27] <rmatte> metalklesk: you should see graphs indicating traffic, packets, and errors
[09-Sep-2009 10:50:56] <metalklesk> yes but in my case they have no data
[09-Sep-2009 10:51:05] <metalklesk> it say: curr nan
[09-Sep-2009 10:51:13] <rmatte> how long ago did you add the device?
[09-Sep-2009 10:51:24] <metalklesk> ten minutes ?
[09-Sep-2009 10:51:30] <rmatte> you need to wait at least 15
[09-Sep-2009 10:51:38] <rmatte> it takes 3 polling cycles for a graph to populate in Zenoss
[09-Sep-2009 10:51:46] <rmatte> each polling cycle is 5 minutes long
[09-Sep-2009 10:51:58] <metalklesk> ahh ok
\n[09-Sep-2009 10:52:37] <metalklesk> is tha polling flexible ? i\nmean, can i change th 5 minutes to lets say 10 minutes ? or 2 ?
[09-Sep-2009 10:53:29] <rmatte> metalklesk: not without actually editing Zenoss source code
[09-Sep-2009 10:53:38] <metalklesk> ok
[09-Sep-2009 10:54:11] <rmatte> generally you either want 5 minutes or 1 minute, but 1 minute is only for extreme cases
[09-Sep-2009 10:54:17] <ckrough> rmatte: thats configuratable in the gui
[09-Sep-2009 10:54:30] <rmatte> ckrough: I was under the impression that it wasn't
[09-Sep-2009 10:54:45] <rmatte> ckrough: I've always heard that code needs to be modified somewhere for that
\n[09-Sep-2009 10:54:53] <ckrough> Collectors > mycollectorname\n> edit > SNMP Performance Cycle Interval (secs) and Status Cycle\nInterval (secs)
[09-Sep-2009 10:55:51] <ckrough>\nmetalklesk: I wouldnt go higher resolution than 5min until you have an\nidea of the server load. There's a whole lot going on in the background
[09-Sep-2009 10:56:04] <metalklesk> oohh yeah, i can see the option
[09-Sep-2009 10:56:09] <metalklesk> yes, i'm agree
[09-Sep-2009 10:56:13] <ckrough> it may have to rebuild your RRDs if you change it
[09-Sep-2009 10:56:18] <ckrough> since they are built to a cycle
\n[09-Sep-2009 10:56:21] <rmatte> ckrough: yeh, but the RRD graphs\nin Zenoss are coded for 5 minutes, so there's something else that you\nneed to change for the graphinh
[09-Sep-2009 10:56:23] <metalklesk> i just wanted to know if the option is there
[09-Sep-2009 10:56:23] <rmatte> graphing*
[09-Sep-2009 10:56:31] <ckrough> "Default RRD Create Command"
[09-Sep-2009 10:56:34] <ckrough> at the bottom
[09-Sep-2009 10:57:08] <rmatte> ah
[09-Sep-2009 10:57:17] <rmatte> I don't remember seeing that before
[09-Sep-2009 10:58:18] <rmatte> cool
[09-Sep-2009 10:58:55] <metalklesk> when you add a new process, in id you put a regex right ?
[09-Sep-2009 10:59:10] <metalklesk> like mysql* ?
[09-Sep-2009 11:00:16] <IoNu|z> rmatte im still here
[09-Sep-2009 11:00:27] <IoNu|z> and yes i think i am hitting that bug
[09-Sep-2009 11:00:37] <IoNu|z> and im using 2.4.5
[09-Sep-2009 11:01:21] <metalklesk> answer to myself, yes, it can be used a regex
[09-Sep-2009 11:01:45] <rmatte> IoNu|z: yeh, no idea when that's going to get fixed, it's driving me nuts too
[09-Sep-2009 11:02:19] <rmatte> metalklesk: just put the process name
[09-Sep-2009 11:02:26] <rmatte> metalklesk: no regex formatting required
[09-Sep-2009 11:02:26] <metalklesk> yup
[09-Sep-2009 11:02:32] <metalklesk> yup
[09-Sep-2009 11:02:37] <rmatte> metalklesk: if the process is "mysqld.bin" then name it that
[09-Sep-2009 11:02:55] <rmatte> and then remodel the device and it'll pick it up along with the command options
[09-Sep-2009 11:05:34]\n<rmatte> does anyone know if the device availability value is\nstored somewhere in zope, or if it's calculated on the fly?
[09-Sep-2009 11:05:51]\n<rmatte> I want to code my own device availability report script,\nthat I can run to pull the device availability for each device
[09-Sep-2009 11:10:37] <metalklesk> why when y model a device (Ubuntu) i can't see ip services ?
[09-Sep-2009 11:10:47] <metalklesk> y == I
[09-Sep-2009 11:10:51] <metalklesk> my mistake
[09-Sep-2009 11:11:01] <rmatte> metalklesk: Have you enabled certain ones?
[09-Sep-2009 11:11:14] <rmatte> like, Click on Services on the left hand meny
[09-Sep-2009 11:11:16] <rmatte> menu*
[09-Sep-2009 11:12:09] <rmatte> annnnnd, 3 of my Zenoss instances that I've tried are frozen...
[09-Sep-2009 11:12:15] <rmatte> waiting on them to give you the other steps
[09-Sep-2009 11:12:43] <rmatte> then click IpService
[09-Sep-2009 11:12:49] <rmatte> Then click privileged
[09-Sep-2009 11:13:08] <rmatte> then find the service that you want to enable
[09-Sep-2009 11:13:11] <rmatte> click on it
[09-Sep-2009 11:13:18] <rmatte> go to the edit tab
[09-Sep-2009 11:13:22] <rmatte> change Monitor from False to True
[09-Sep-2009 11:13:25] <rmatte> then remodel the device
\n[09-Sep-2009 11:13:43] <rmatte> also, I've noticed that there's a\npotential bug with having Zenoss automatically pick up on IpServices\nduring modeling
[09-Sep-2009 11:13:52] <rmatte> it seems to only pick up on the first service, and ignores the rest
[09-Sep-2009 11:14:01] <rmatte> so you may have to add them in manually for the time being
[09-Sep-2009 11:14:25] <metalklesk> ok, thanks
\n[09-Sep-2009 11:14:57] <rmatte> If you notice the same behaviour\nas I've described let me know so that I can file a bug report
[09-Sep-2009 11:15:10] <metalklesk> ok, i will try now
[09-Sep-2009 11:15:13] <rmatte> thanks
[09-Sep-2009 11:18:17] <metalklesk> ssh can be considered as ip service in linux ?
[09-Sep-2009 11:18:24] <rmatte> correct
[09-Sep-2009 11:18:27] <rmatte> if you want
[09-Sep-2009 11:18:36] <rmatte> IpService just means it's monitoring the port
[09-Sep-2009 11:18:45] <rmatte> you could also monitor the Process as well
[09-Sep-2009 11:19:04] <rmatte> I find that for SSH monitoring the port is sufficient
[09-Sep-2009 11:19:34] <rmatte> same with stuff like ftp
[09-Sep-2009 11:19:41] <rmatte> or rdp
[09-Sep-2009 11:19:43] <metalklesk> ahh ok, great
[09-Sep-2009 11:22:49] <metalklesk> let's say i want so monitor sshd as ip service
[09-Sep-2009 11:23:30] <metalklesk> Classes -> Services -> add new ip service -> sshd
[09-Sep-2009 11:23:39] <metalklesk> then click on sshd -> edit
[09-Sep-2009 11:23:53] <metalklesk> i write port 22 and change monitored from false to true
[09-Sep-2009 11:24:01] <metalklesk> is that ok ?
[09-Sep-2009 11:26:06] <metalklesk> then i go to my device (Ubuntu) -> OS tab -> ip service -> add ip service
[09-Sep-2009 11:26:26] <metalklesk> but sshd ip service isn't there
[09-Sep-2009 11:26:37] <metalklesk> not the one i created before
[09-Sep-2009 11:27:58] <metalklesk> but it appears another one, i guess it's a default one
[09-Sep-2009 11:28:12] <metalklesk> i can use that !
[09-Sep-2009 11:37:56] <t3los> Is anyone available to answer a few questions?
[09-Sep-2009 11:38:04] <jb> just ask the questions..
[09-Sep-2009 11:38:06] <jb> don
[09-Sep-2009 11:38:09] <jb> don't ask to ask.
[09-Sep-2009 11:38:15] <t3los> Lol.
[09-Sep-2009 11:38:29] <MajereDB8> import sys
[09-Sep-2009 11:38:32] <MajereDB8> whoops sorry
\n[09-Sep-2009 11:38:45] <t3los> Alright, then. So I have this one\nsite, with less plugins loaded, and the same versions of zenpacks
\n[09-Sep-2009 11:38:49] <metalklesk> i have added to monitoring my\nnotebook with mandriva by SNMP. I have added the gnome-terminal process\n(that i have created before) and the status is green even if i don't\nhave that process running, why is that ?
[09-Sep-2009 11:38:58] <t3los> But for some reason, I get much less results here at my personal site
\n[09-Sep-2009 11:39:24] <t3los> It's kind of aggrivating because\nI've gone out of my way to make sure that things are even identical if\nnot more thorough over on my end.
[09-Sep-2009 11:39:51] <t3los> Less traffic, less machines, more attention to detail, and less results.
[09-Sep-2009 11:39:52] <t3los> ;p
\n[09-Sep-2009 11:40:47] <t3los> I read the entire documentation. I\nknow how to do a lot inside of Zenoss, but I can't find out what is\ncausing this problem.
[09-Sep-2009 11:41:09] <jb> what do you mean?
[09-Sep-2009 11:41:13] <jb> what is "less detail" ?
[09-Sep-2009 11:41:36] <t3los> I mean, as far as Status, OS, Hardware, and Performance Graphs
[09-Sep-2009 11:41:40] <t3los> I have less information
[09-Sep-2009 11:41:51] <jb> sounds like you have SNMP problems
[09-Sep-2009 11:42:00] <jb> can you walk devices from the new zenoss host?
[09-Sep-2009 11:42:19] <t3los> Well, sec
[09-Sep-2009 11:42:37] <t3los> Actually, no
[09-Sep-2009 11:42:38] <t3los> Lol.
[09-Sep-2009 11:42:46] <t3los> oh wait
[09-Sep-2009 11:43:01] <t3los> yes
[09-Sep-2009 11:43:02] <t3los> Sorry
[09-Sep-2009 11:43:41] <t3los> Yes, I can see the devices I am trying to collect for.
[09-Sep-2009 11:43:58] <jb> see them how?
[09-Sep-2009 11:45:06] <t3los> ==== hostName ====
[09-Sep-2009 11:45:06] <t3los> snmpwalk -v1 -cccissnmp 192.168.68.11 system
[09-Sep-2009 11:45:06]\n<t3los> SNMPv2-MIB::sysDescr.0 = STRING: Hardware: x86 Family 15\nModel 4 Stepping 3 AT/AT COMPATIBLE - Software: Windows Version 5.2\n(Build 3790 Multiprocessor Free)
[09-Sep-2009 11:45:06] <t3los> SNMPv2-MIB::sysObjectID.0 = OID: SNMPv2-SMI::enterprises.311.1.1.3.1.3
[09-Sep-2009 11:45:06] <t3los> DISMAN-EVENT-MIB::sysUpTimeInstance = Timeticks: (224371) 0:37:23.71
[09-Sep-2009 11:45:08] <t3los> SNMPv2-MIB::sysContact.0 = STRING: XXXXXXXXX
[09-Sep-2009 11:45:10] <t3los> SNMPv2-MIB::sysName.0 = STRING: XXXXXXX
[09-Sep-2009 11:45:12] <t3los> SNMPv2-MIB::sysLocation.0 = STRING: Server Room
[09-Sep-2009 11:45:14] <t3los> SNMPv2-MIB::sysServices.0 = INTEGER: 79
[09-Sep-2009 11:45:22] <t3los> Ignore the X's and hostName
[09-Sep-2009 11:45:30] <rmatte> metalklesk: no, ssh is already there by default
[09-Sep-2009 11:45:30] <t3los> Filtered text.
[09-Sep-2009 11:45:41] <rmatte> metalklesk: under the privileged organizer
[09-Sep-2009 11:45:50] <rmatte> metalklesk: so just find it in there and enable monitoring for it
\n[09-Sep-2009 11:46:16] <rmatte> metalklesk: Zenoss has pretty\nmuch every IP Service you can think of already there, you just need to\nenable monitoring for them
[09-Sep-2009 11:46:50] <metalklesk> what do you mean with privileged organizer ?
[09-Sep-2009 11:48:00]\n<t3los> Seriously, though. Has anyone heard of that kind of\nthing? More information at one site than the other even though you've\nmade sure to put more effort into the one?
[09-Sep-2009 11:54:44] <rmatte> metalklesk: click on Services on the left
[09-Sep-2009 11:54:58] <metalklesk> done
[09-Sep-2009 11:55:01] <rmatte> metalklesk: then look at the top, you'll see "Sub-Folders"
[09-Sep-2009 11:55:08] <rmatte> there's a subfolder called IpService
[09-Sep-2009 11:55:10] <rmatte> click on it
[09-Sep-2009 11:55:11] <metalklesk> yes
[09-Sep-2009 11:55:19] <rmatte> Then you'll see more Sub-Folders
[09-Sep-2009 11:55:23] <metalklesk> done
[09-Sep-2009 11:55:25] <rmatte> click on the privileged sub-folder
[09-Sep-2009 11:55:51] <rmatte> and voila
[09-Sep-2009 11:56:04] <rmatte> 684 of the most common ports
[09-Sep-2009 11:56:14] <rmatte> then you just need to search for what you want and enable monitoring on it
[09-Sep-2009 11:56:47] <rmatte> I generally just sort by port then scroll until I find what I want
[09-Sep-2009 11:57:29] <metalklesk> perfect
[09-Sep-2009 11:57:33] <metalklesk> many thanks
[09-Sep-2009 11:57:35] <rmatte> np
[09-Sep-2009 12:00:07] <metalklesk> gnome-terminal on mandriva is now showing a red status, that is ok
[09-Sep-2009 12:01:00]\n<metalklesk> (i created a process in zenoss for gnome-terminal,\nit must show a red status because there is no process of gnome-terminal\nin my mandriva)
[09-Sep-2009 12:04:15] <metalklesk> can i mix snmp AND ssh mointoring for one device ?
[09-Sep-2009 12:05:50] <mrayzenoss> yes
[09-Sep-2009 12:06:53]\n<metalklesk> and what if i monitor Ubuntu with snmp and Ubuntu\nzenpack ? both methods are going to catch the same parameters
[09-Sep-2009 12:07:29] <mrayzenoss> well, the Ubuntu ZenPack is monitored under /Servers/SSH/Linux/Ubuntu
[09-Sep-2009 12:07:43] <mrayzenoss> as opposed to /Servers/Linux
[09-Sep-2009 12:08:15] <mrayzenoss> you can configure them to conflict, not sure what happens
[09-Sep-2009 12:08:24] <mrayzenoss> but they don't conflict out of the box
[09-Sep-2009 12:09:11] <rmatte> yeh, you can mix and match monitoring components, but you have to be careful hehe
[09-Sep-2009 12:11:33] <rmatte> quite honestly, pure SNMP monitoring is fairly adequate for Linux servers
[09-Sep-2009 12:11:49] <rmatte> in most cases anyways
[09-Sep-2009 12:19:26] <metalklesk> ok
[09-Sep-2009 12:19:35] <Brad_K> mrayzenoss: hi! are we having fun yet? ;-)
[09-Sep-2009 12:22:35] <rmatte> he's always having fun
[09-Sep-2009 12:24:39] <mrayzenoss> yeah, every day is a picnic
[09-Sep-2009 12:25:10] <ckrough> dont you know? Community managers dont have bad days.
[09-Sep-2009 12:25:42] * rmatte whispers "That's because of the bottles of rum that they keep in their bottom drawers."
[09-Sep-2009 12:26:34] * mrayzenoss keeps it in the top drawer
[09-Sep-2009 12:27:09] <rmatte> ah, for convenience no doubt
\n[09-Sep-2009 12:30:10] <ckrough> you could rig up a zenpack that\nwatches output from a scale and alert when the bottle gets low
[09-Sep-2009 12:30:36] <ckrough> we could event graph your drinking rate
[09-Sep-2009 12:30:45] <rmatte> haha
[09-Sep-2009 12:30:55] <ckrough> "mray: my life on zenoss" .com
[09-Sep-2009 12:31:01] <rmatte> Leave it to a Zenoss user to think of something like that.
\n[09-Sep-2009 12:31:39] <ckrough> I have plans for my kegerator. I\nusually buy guiness, which is a special order, so I want to be alerted\nwhen the keg is 2 weeks to empty. I think zenoss is going to play a\nrole.
[09-Sep-2009 12:31:50] <rmatte> We might have to use predictive thresholding though, depending on the day
[09-Sep-2009 12:31:58] <mrayzenoss> I'm sure others will be interested in that ZenPack, be sure to send it in
[09-Sep-2009 12:32:55] <rmatte> ZenPacks.MattRay.Boozer.Pack-1.0-py2.4.egg
\n[09-Sep-2009 12:33:25] <venturaville> does anyone have an example\nof a zenpack which creates a /Mibs/sub-folder and/or a\n/Devices/sub-folder via script rather than objects.xml?
[09-Sep-2009 12:34:20] <rmatte> venturaville: just curious, but why use a script instead of objects.xml?
\n[09-Sep-2009 12:34:49] <venturaville> rmatte: I want to create\nit, but I don't want the zenpack to 'own it' when it gets uninstalled.
[09-Sep-2009 12:35:06] <rmatte> oh, I see
[09-Sep-2009 12:38:38] <venturaville> anyone anyone ... bueller?
[09-Sep-2009 12:38:42] <venturaville> :-)
[09-Sep-2009 12:40:33] <Brad_K> not me -- i haven't even earned my white belt yet.
[09-Sep-2009 12:40:43] <rmatte> You should probably start with how to get a script to execute in a ZenPack to begin with
[09-Sep-2009 12:40:52] <rmatte> then figure out what zendmd commands you'd need to use to make an organizer
[09-Sep-2009 12:40:58] <ckrough> can it be part of the __init__
[09-Sep-2009 12:41:06] <venturaville> ckrough: yes
[09-Sep-2009 12:41:54] <metalklesk> i'm going to launch
[09-Sep-2009 12:41:58] <metalklesk> see ya
[09-Sep-2009 12:42:15] <chudler> BLASTOFF
[09-Sep-2009 12:42:16] * ckrough thinking of a launch joke
[09-Sep-2009 12:48:08] <rmatte> "Copy that odyssey, we roger your roll."
[09-Sep-2009 13:01:18] <Brad_K> And there you have the history of how donut holes were first created....
[09-Sep-2009 13:05:15] <rmatte>
[09-Sep-2009 13:07:42] <Brad_K> I guess you really do love your bread products, eh?
[09-Sep-2009 13:09:14] <rmatte> yessir, nom nom
[09-Sep-2009 13:26:15] <chemist> evening all
[09-Sep-2009 13:26:57] <mrayzenoss> heads up ZenPack devs
[09-Sep-2009 13:27:18] <mrayzenoss> we're going to be moving community.zenoss.org to zenpacks.zenoss.org tonight
[09-Sep-2009 13:27:31] <mrayzenoss> Trac and the wiki should be unaffected
[09-Sep-2009 13:27:40] <mrayzenoss> but Subversion will need to get updated to the new URL
[09-Sep-2009 13:27:53] <mrayzenoss> there will be an announcement about it on the blog and forums soon
[09-Sep-2009 13:28:06] <mrayzenoss> (probably not a lot of people this affects)
\n[09-Sep-2009 13:38:48] <chemist> in adminstered objects, next to\nthe role type, there is a level (currently 1) can someone explain what\nthis is for?
[09-Sep-2009 13:41:09] <mrayzenoss> this is related to the use of permissions, I don't think it does much in Core unless you dig into Zope
[09-Sep-2009 13:41:26] <chemist> ok, thanks
\n[09-Sep-2009 13:41:37] <mrayzenoss> Jane Curry's got a paper on\nACLs in Core http://www.skills-1st.co.uk/papers/jcurry.html
[09-Sep-2009 13:41:51] <mrayzenoss> http://www.skills-1st.co.uk/papers/jane/users_events_zproperties_paper.pdf
[09-Sep-2009 13:42:24] <chemist> ta
[09-Sep-2009 13:53:52] <metalklesk> i'm back
[09-Sep-2009 13:54:36] <metalklesk> i have a question
[09-Sep-2009 13:55:20] <rmatte> ask
[09-Sep-2009 13:55:43] <jb> mrayzenoss: have you seen any Mitel zenpacks
[09-Sep-2009 13:55:44] <jb> ?
[09-Sep-2009 13:55:49] <mrayzenoss> nope
\n[09-Sep-2009 13:55:57] <metalklesk> with a linux box, let's say\nubuntu server 9.04 or mandriva 2009 spring with 4 GB ram CPU dual core\nand so on, how much devices i can monitor ?
[09-Sep-2009 13:56:02] <rmatte> nope
[09-Sep-2009 13:56:12] <jb> ok, going to start checking our controllers to see what kind of info I can get via snmp
[09-Sep-2009 13:56:18] <metalklesk> with devices i mean more windows and linux boxes than other type of devices
[09-Sep-2009 13:56:28] <rmatte> metalklesk: well, with 4 gigs of ram so far I've been able to monitor close to 400
[09-Sep-2009 13:56:47] <rmatte> metalklesk: you'd want 6 or 8 gigs of ram if you want to get in to thousands of devices
[09-Sep-2009 13:56:48] <jb> Mem: 32941048k total, 23419780k used, 9521268k free, 792036k buffers
[09-Sep-2009 13:56:49] <jb>
[09-Sep-2009 13:56:53] <mrayzenoss> metalklesk: http://forums.zenoss.com/viewtopic.php?t=6230
[09-Sep-2009 13:56:58] <mrayzenoss> that thread has some numbers
[09-Sep-2009 13:57:04] <metalklesk> great
[09-Sep-2009 13:57:30] <mrayzenoss> zenpacks.zenoss.org is live
[09-Sep-2009 13:57:38] <cgibbons> yippie?
[09-Sep-2009 13:57:46] <mrayzenoss> just a URL redirect
[09-Sep-2009 13:58:07] <jb> whats your backlog looking like?
[09-Sep-2009 13:59:00] <mrayzenoss> well, I'm not doing any until next week
[09-Sep-2009 13:59:07] <mrayzenoss> so rmatte has 2 in my queue
[09-Sep-2009 13:59:12] <jb> i still have one?
[09-Sep-2009 13:59:26] <mrayzenoss> ugh, yeah
[09-Sep-2009 13:59:30] <jb> hah
[09-Sep-2009 13:59:31] <mrayzenoss> never finished it
[09-Sep-2009 13:59:35] <mrayzenoss> IBM 3584
[09-Sep-2009 13:59:37] <jb> yeah
[09-Sep-2009 13:59:42] <mrayzenoss> so there are 5
\n[09-Sep-2009 13:59:58] <mrayzenoss> and we're migrating to the\nnew Community site right now, so I'm not adding new content
[09-Sep-2009 14:00:06] <jb> whens that going to launch?
[09-Sep-2009 14:00:09] <mrayzenoss> Monday
[09-Sep-2009 14:00:12] <jb> nice..
[09-Sep-2009 14:00:14] <jb> what platform again?
[09-Sep-2009 14:00:42] <mrayzenoss> Jive Social Business Software: http://www.jivesoftware.com/products
[09-Sep-2009 14:00:53] <mrayzenoss> we're migrating ZenPacks right now
[09-Sep-2009 14:00:59] <jb> k
[09-Sep-2009 14:01:02] <mrayzenoss> I'm mapping through the wiki/plone/trac content
[09-Sep-2009 14:01:06] <mrayzenoss> and redirecting URLs
[09-Sep-2009 14:01:15] <mrayzenoss> there'll be a blog post probably tomorrow
[09-Sep-2009 14:01:16] <metalklesk> zenoss is ver efficient compared to BMC Portal
[09-Sep-2009 14:01:25] <chudler> jive.. seems vaguely familiar. interesting niche "social business"...
[09-Sep-2009 14:01:39] <metalklesk> 400 devices with just one server is ver well
[09-Sep-2009 14:01:39] <mrayzenoss> JBoss.org, Communities.vmware.com
[09-Sep-2009 14:01:59] <metalklesk> *vey
[09-Sep-2009 14:02:05] <metalklesk> *very
[09-Sep-2009 14:02:38] <mrayzenoss> heh, http://communities.bmc.com uses it too
[09-Sep-2009 14:04:34] <rmatte> I can't wait until the whole new zenoss community site is up and buzzing
[09-Sep-2009 14:04:51] <rmatte> Sept 14th or 15th was the goal for the date as I recall
[09-Sep-2009 14:05:10] <mrayzenoss> yeah, we think we'll hit it
[09-Sep-2009 14:05:18] <mrayzenoss> it'll be a rolling start
[09-Sep-2009 14:05:34] <mrayzenoss> the announcement should be tomorrow, the forums will be offline over the weekend
[09-Sep-2009 14:05:44] <rmatte> cool
[09-Sep-2009 14:05:56] <mrayzenoss> and Monday we'll bring it up and kick the tires
[09-Sep-2009 14:06:09] <mrayzenoss> but for now, community.zenoss.org is free to be re-used
[09-Sep-2009 14:06:55] <mrayzenoss> and tonight we'll probably break SVN
[09-Sep-2009 14:07:11] <mrayzenoss> which was the purpose of the blog post
[09-Sep-2009 14:07:19] <rmatte> some of those ZenPacks need updating badly lol
[09-Sep-2009 14:07:26] <rmatte> particularily the .zips
[09-Sep-2009 14:07:38] <mrayzenoss> well, they're slowing getting eggifyzenpack'd
[09-Sep-2009 14:07:44] <rmatte> cool
[09-Sep-2009 14:07:50] <mrayzenoss> Chet updated the Celerra one
[09-Sep-2009 14:08:01] <jb> are you migrating forums?
[09-Sep-2009 14:08:05] <mrayzenoss> absolutely
[09-Sep-2009 14:08:06] <jb> from phpbb?
[09-Sep-2009 14:08:07] <jb> k
[09-Sep-2009 14:08:18] <mrayzenoss> and users
[09-Sep-2009 14:08:40] <mrayzenoss> and most content, we'll let it get sorted a bit more after the launch
[09-Sep-2009 14:08:46] <mrayzenoss> but it'll all be under 1 roof
[09-Sep-2009 14:09:31] <ckrough> what are you moving to from phpbb?
[09-Sep-2009 14:09:51] <mrayzenoss> Jive will hold the forums as well
[09-Sep-2009 14:10:14] <mrayzenoss> it's forums/wiki/docs/community all bundled in 1
[09-Sep-2009 14:10:24] <mrayzenoss> and the blog.
[09-Sep-2009 14:10:26] <mrayzenoss> and search
[09-Sep-2009 14:12:00] <Troubadix09> mrayzenoss: I have a question :-)
[09-Sep-2009 14:13:13] <metalklesk> one question, can i exclude a device, that i already have added, from monitoring ?
[09-Sep-2009 14:13:33] <metalklesk> i mean without delete it
[09-Sep-2009 14:13:51] <metalklesk> like monitoring on/off of bmc portal
[09-Sep-2009 14:15:32] <slip_> put it into the "discovered" item group...
[09-Sep-2009 14:15:44] <mrayzenoss> discovered is still monitored
[09-Sep-2009 14:16:28] <mrayzenoss> metalklesk: you can set the production state on the Edit tab
\n[09-Sep-2009 14:16:45] <Troubadix09> mrayzenoss: are the\nzenoss-team planning to take a place at one of the next CEbit in\nHannover, Germany?
[09-Sep-2009 14:17:25]\n<mrayzenoss> CEbit? Unlikely, though we have talked about\nheading to an European event next year, thinking FOSDEM
[09-Sep-2009 14:17:51] * rmatte reminds mrayzenoss that there's beer in Germany, good beer
[09-Sep-2009 14:18:02] * mrayzenoss has been to Germany many times
[09-Sep-2009 14:18:10] <rmatte> hehe
[09-Sep-2009 14:18:28] <rmatte> I haven't really traveled anywhere
\n[09-Sep-2009 14:18:45] <rmatte> I'd like to, just don't have the\nfinances, and I don't have the kind of job that requires me to travel\n(yet)
[09-Sep-2009 14:19:40] <venturaville> you\ncould go work for Google.... I have a friend over there that ends up in\nIreland and Australia from time to time
[09-Sep-2009 14:20:04] <rmatte> I'd love to work for google quite honestly, but I don't really know what I'd do for them hehe
[09-Sep-2009 14:20:18] <rmatte> Unless they want to use Zenoss to monitor stuff
[09-Sep-2009 14:20:36] <rmatte> (Could do Linux administrator as well I suppose)
[09-Sep-2009 14:20:43] <rmatte> administration, rather
[09-Sep-2009 14:20:48] <rhettardo> mmm spaten
[09-Sep-2009 14:21:26] <rmatte> rhettardo: hehe
[09-Sep-2009 14:21:45] <rhettardo> love me some optimator
[09-Sep-2009 14:27:53] <Troubadix09> bye all
[09-Sep-2009 14:28:42] <mrayzenoss> I guess he wanted me to go to CEbit
[09-Sep-2009 14:30:24] <rhettardo> oh well
[09-Sep-2009 14:31:20] <mrayzenoss> it's still in the planning stages... "wouldn't it be nice to go to Europe..."
[09-Sep-2009 14:32:04] <rmatte> and the answer is: "Yes, it would"
[09-Sep-2009 14:32:09]\n<Curt> Hello all, we have a predominantly windows server shop\nhere with a few linux servers doing various things. i'm trying to\nfind the best open source software for monitoring servers and log\nfiles. would zenoss make a good fit here or would there be other\nsimpler method for about 40 servers
[09-Sep-2009\n14:32:50] <rmatte> Curt: great for monitoring servers, doesn't\nhave any built in functionality for monitoring log files, other than if\nyou have syslogd sending syslog events to Zenoss
[09-Sep-2009 14:33:11] <rmatte> and in Windows case, you can monitor the event log via WMI
[09-Sep-2009 14:33:30] <metalklesk> wmi is cool for monitoring windows with zenoss
\n[09-Sep-2009 14:33:31] <rmatte> or just enable SNMP traps on the\nwindows boxes, which will result in them sending traps to Zenoss for\nmajor eventlog events
[09-Sep-2009 14:34:00]\n<rmatte> Curt: if you use the Core version of Zenoss you really\nneed a mix of WMI and SNMP to get full monitoring coverage
[09-Sep-2009 14:34:14] <rmatte> Curt: the enterprise version of Zenoss supports full WMI monitoring
[09-Sep-2009 14:34:33] <rmatte> Curt: Zenoss' Linux monitoring support via SNMP and SSH is excellent even in Core
\n[09-Sep-2009 14:34:37] <mrayzenoss> vaguely changing the\nsubject... has anyone used\nhttp://www.zenoss.com/community/projects/zenpacks/wmifilesmonitor ?
[09-Sep-2009 14:35:00] <metalklesk> rmatte and mrayzenoss: when are you planning to come visit Chile ?
[09-Sep-2009 14:35:04] <metalklesk>
[09-Sep-2009 14:35:17] <rmatte> mrayzenoss: Isn't that an older ZenPack?
[09-Sep-2009 14:35:30] <rmatte> metalklesk: as soon as you send me my plane ticket
[09-Sep-2009 14:35:35] <mrayzenoss> rmatte: yeah, hasn't been updated to the new WMI Data Source yet
[09-Sep-2009 14:35:38] <metalklesk> ejjejejje
[09-Sep-2009 14:35:53] <mrayzenoss> metalklesk: any big open source community events down there?
[09-Sep-2009 14:36:08]\n<rmatte> mrayzenoss: quite honestly, any packs not using the new\nWMI Data Source pack should be either updated to use it, or purged
[09-Sep-2009 14:36:27] <metalklesk> linuxfest
[09-Sep-2009 14:36:34] <mrayzenoss> yeah, with the new website you can add that sentiment to each one
[09-Sep-2009 14:36:51] <rmatte> mrayzenoss: and I shall, there are some that have been particularily frustrating
[09-Sep-2009 14:37:05] <rmatte> mrayzenoss: such as the old Brocade pack
\n[09-Sep-2009 14:37:30] <mrayzenoss> rmatte: I've sent 3 emails to\nthe author of that new Brocade ZenPack asking him about it
[09-Sep-2009 14:37:33] <metalklesk> the other day i was talking with a friend about making our own company
[09-Sep-2009 14:37:48] <metalklesk> a competition for the company i work now
[09-Sep-2009 14:37:57] <mrayzenoss> rmatte: he told me he was working on it last I heard
[09-Sep-2009 14:38:04] <metalklesk> using open source and for monitoring of course zenoss
[09-Sep-2009 14:38:06]\n<rmatte> mrayzenoss: well, the new Brocade pack is only for a\nspecific type of Brocade, (not the type we have), the old Brocade pack\nis for the type that we use but it's only configured to monitor port 1\non the switch
[09-Sep-2009 14:38:18] <rmatte> mrayzenoss: so it's useless since it ignores the other 24 ports lol
[09-Sep-2009 14:38:27] <mrayzenoss> metalklesk: we plan on making the Core partners much more visible in the new site
[09-Sep-2009 14:38:39] <mrayzenoss> rmatte: yeah
[09-Sep-2009 14:38:50] <metalklesk> that is what we would want, be partners
[09-Sep-2009 14:39:04] <metalklesk> here in Chile is a growing market so sell service
\n[09-Sep-2009 14:39:16] <rmatte> mrayzenoss: people need to stop\ngeneralizing their ZenPacks so much, that's like me writing a ZenPack\nfor 1 particular model of HP server and releasing it as a "HP Server\nZenPack", when it reality it doesn't work with the other gajillion\nmodels of HP servers.
[09-Sep-2009 14:39:18] <metalklesk> *to
[09-Sep-2009 14:39:35] <metalklesk> rmatte, i'm agree
[09-Sep-2009 14:39:59] <metalklesk> SuSE zenpack does not monitor SLES or SLED, only openSUSE
[09-Sep-2009 14:40:14] <metalklesk> it should be openSUSE Zenpack
[09-Sep-2009 14:40:22] <rmatte> metalklesk: you'd have to talk to Rocket about that
[09-Sep-2009 14:40:41] * rmatte points accusingly at rocket
[09-Sep-2009 14:40:42] <metalklesk> ok, but for now is just an idea
[09-Sep-2009 14:40:57] <ckrough> hmm. what would you call the * section of a zenoss path /zport/dmd/*/devices/
[09-Sep-2009 14:41:00] <rmatte> yeh, I agree with that one, would be good to rename it to openSUSE
[09-Sep-2009 14:41:04] <ckrough> container?
[09-Sep-2009 14:41:18] <metalklesk> i am looking to our clients and they are not happy with BMC Software
[09-Sep-2009 14:41:39] <rmatte> ckrough: Organizer
[09-Sep-2009 14:41:40] <metalklesk> Entel SA has dropped BMC Patrol for Nagios
[09-Sep-2009 14:42:05] <metalklesk> BCI Bank is dropping BMC Patrol for prognosis
[09-Sep-2009 14:42:25] <rmatte> ckrough: actually, isn't it just /zport/dmd/Devices?
[09-Sep-2009 14:42:38] <metalklesk> 52% of our clients does not renew contracts for Service Assurance
[09-Sep-2009 14:42:43] <ckrough> rmatte: /zport/dmd/Devices/Server/Linux/devices/
[09-Sep-2009 14:42:58] <rmatte> right
[09-Sep-2009 14:43:11] <rmatte> Organizer path
[09-Sep-2009 14:43:41] <metalklesk> and is clear, open source is the sh**, not propietary software business
[09-Sep-2009 14:43:46] <metalklesk>
[09-Sep-2009 14:43:58] <rmatte> opensource is the future of software development, it's just taking time to catch on
[09-Sep-2009 14:44:11] <rmatte> mainly because of micro$oft
[09-Sep-2009 14:44:15] <metalklesk> yes, here in Chile is even more slowly
[09-Sep-2009 14:44:40] <metalklesk> our goverment makes hidden contracts
[09-Sep-2009 14:44:45] <metalklesk> with microsoft
[09-Sep-2009 14:44:45] <rmatte> I'm glad Vista was so horrible actually, made a ton of people decide to try Linux
[09-Sep-2009 14:44:51] <metalklesk> yes
\n[09-Sep-2009 14:44:58] <ckrough> buying knowledge (people) is\nbecoming less expensive than paying insane fees for application\nextensions
[09-Sep-2009 14:45:13] <rmatte> and\nnow that they've had time to get used to it and enjoy it, they probably\nwon't bother going back to windows
[09-Sep-2009 14:45:32] <rmatte> the only real place where microsoft has the PC industry by the balls is games
\n[09-Sep-2009 14:45:36] <metalklesk> last year we knew about some\ncontracts about our goberment and microsoft, it was disgusting
[09-Sep-2009 14:46:09] <metalklesk> yes
[09-Sep-2009 14:46:39] <metalklesk> yesterday my friend quits from this company
[09-Sep-2009 14:47:04] <metalklesk> because he gets mad with the CEO
[09-Sep-2009 14:47:07]\n<rmatte> yeh, our government is heavy in to Microsoft too, it'd\nbe nice to see them take some initiative and go opensource. The\nargument is always "well who is going to support it". I'm sorry, but\nif you drop all of the licensing fees that you're paying for Windows\nyou could have a dedicated Linux support person per floor in every\nbuilding and still save money.
[09-Sep-2009 14:47:30] <metalklesk> yes
[09-Sep-2009 14:47:48] <metalklesk> so as my friend quits and he have a lot of potential
[09-Sep-2009 14:48:04] <metalklesk> we talked about build our own company
\n[09-Sep-2009 14:48:39] <rmatte> I'm tempted to do my own\nmonitoring thing on the side, I just don't know if my company would\nallow me to.
[09-Sep-2009 14:48:46] <metalklesk> as he quits, this company looses 2.5 million dolars
[09-Sep-2009 14:48:58] <rmatte> yikes
[09-Sep-2009 14:49:01] <metalklesk> now here are desperate
[09-Sep-2009 14:49:08] <metalklesk> because of that
[09-Sep-2009 14:49:25] <metalklesk> bussiness around BMC service automation is dead now
[09-Sep-2009 14:49:49] <metalklesk> my friend was the one who was making those business
[09-Sep-2009 14:49:56] <metalklesk> in Chile and Colombia
[09-Sep-2009 14:50:24] <metalklesk> so i think we can do a lot with zenoss, obviusly with partnership
[09-Sep-2009 14:50:50] <rmatte> yeh, Zenoss is pretty impressive and it's getting better and better
[09-Sep-2009 14:51:05] <metalklesk> yes
[09-Sep-2009 14:51:38] <metalklesk> the 2 open source software i think are really impressive are Zenoss and OpenBravo
[09-Sep-2009 14:52:00] <metalklesk> openbravo are an ERP Software
[09-Sep-2009 14:52:31] <metalklesk> i have seen a lot of companys interested about it here in Chile
[09-Sep-2009 14:53:06] <rmatte> openbravo eh?
[09-Sep-2009 14:53:08] <rmatte> what does it do?
\n[09-Sep-2009 14:53:23] <metalklesk> in my blog i wrote some\narticles about openbravo and i received a lot of comments about people\nthat want it in their companies instead of SAP
[09-Sep-2009 14:53:24] <rmatte> oh, inventory system
[09-Sep-2009 14:53:44] <metalklesk> inventory is just one of the modules
[09-Sep-2009 14:54:05] <metalklesk> accounts module too
[09-Sep-2009 14:54:24] <metalklesk> you can manage clients, providers, inventory, accounts
[09-Sep-2009 14:54:32] <metalklesk> multi enterprise
[09-Sep-2009 14:54:50] <metalklesk> is really impressive what you can do with open bravo
[09-Sep-2009 14:55:33] <metalklesk> my blog really jumps high here in latin america because of the open bravo tips i gave
[09-Sep-2009 14:55:54] <metalklesk> maybe it happens again because the article i wrote about zenoss
[09-Sep-2009 14:55:56] <metalklesk>
[09-Sep-2009 14:56:44] <metalklesk> http://metalklesk.blogspot.com/
[09-Sep-2009 14:57:35] <chudler> very detailed... cant read it though ;-)
[09-Sep-2009 14:57:56] <rmatte> same here hehe
[09-Sep-2009 14:58:08] <metalklesk> jejee is in spanish
[09-Sep-2009 15:01:16] <metalklesk> notice the pictures in "Mis imágenes", me with 2 friends and Ronnie Coleman
[09-Sep-2009 15:01:18] <metalklesk> joojoj
[09-Sep-2009 15:01:52] <chemist> I can not get zenoss to email me when a critical event occurrs, I can send a test mail
[09-Sep-2009 15:02:50] <chemist> I have added the locations to my administered objects
[09-Sep-2009 15:08:25] <rmatte> mrayzenoss: yeh, I can imagine
[09-Sep-2009 15:08:28] <rmatte> oops
[09-Sep-2009 15:10:22] <ckrough> "def getAllRRDTemplatesPainfully(self, rrdts=None):" ...cute
[09-Sep-2009 15:14:18] <ckrough> jsonGetDeviceNames ignores the path and always returns all matches. sad panda.
[09-Sep-2009 15:19:22] <rhettardo> chemist: sounds like an alerting rules issue
\n[09-Sep-2009 15:21:25] <chemist> rhettardo: do I have to create\nexplicit rules? I thought if I didn't create any I would get alerted\nany time, every time
[09-Sep-2009 15:21:51] <rhettardo> i dont recall what the default settings are. but i have my rules specific
[09-Sep-2009 15:22:50] <metalklesk> can i have custom threshold for device ?
[09-Sep-2009 15:23:31] <chemist> rhettardo: ok, I will give it a go
[09-Sep-2009 15:23:33] <chemist> thanks
[09-Sep-2009 15:23:45] <mrayzenoss> metalklesk: yeah, you can copy and edit the templates
[09-Sep-2009 15:23:46] <rhettardo> np
[09-Sep-2009 15:25:07] <metalklesk> is there a documentation about threshold customization ?
\n[09-Sep-2009 15:25:37] <t3los> I've been looking through the\ndocumentation on this one, and I cannot find anything about a grey\nstatus light
[09-Sep-2009 15:25:45] <t3los> I know what yellow means, and red, and green
[09-Sep-2009 15:25:55] <t3los> I'm guessing it means it can't find the events?
[09-Sep-2009 15:26:16] <t3los> I can monitor my CPU times and memory in the perf tab, but the lights won't get a color?
[09-Sep-2009 15:30:17] <t3los> Anyone?
[09-Sep-2009 15:30:34] <mrayzenoss> t3los: it means it's unmonitored
\n[09-Sep-2009 15:31:21] <t3los> Yeah that's what I figured. I\ncan't find anything in HW that will let meset monitoring true for these\ncomponents.
[09-Sep-2009 15:32:34] <jb> or that it just doesn't know the status of the component..
[09-Sep-2009 15:33:21] <chemist> is there anyway to set what time the modelling of a device takes place?
\n[09-Sep-2009 15:34:38] <t3los> Is there a way to get the\ncomponent out of the list if it's just going to stay grey? I cannot\nfind a place to map the event to it, so if it's just easier to get it\nout of the list... lol
[09-Sep-2009 15:35:51] <metalklesk> manually changing the production state ?
[09-Sep-2009 15:37:08] <rmatte> chemist: not really
\n[09-Sep-2009 15:37:26] <t3los> Well I want to keep it in\nproduction. If I could find the components property page that would\nsolve the problem.
[09-Sep-2009 15:37:28] <rmatte> chemist: you could do it with a cron job or something
[09-Sep-2009 15:38:22] <rmatte> t3los: hunh? "get it out of the list"?
[09-Sep-2009 15:38:42] <t3los> The component list, with the status lights
[09-Sep-2009 15:38:57] <t3los> cpu times, and memory pages, etc
[09-Sep-2009 15:39:02] <t3los> they won't get a status color
\n[09-Sep-2009 15:39:25] <t3los> so if I am unable to get them to\nbe monitored, I wish to just remove them from that components list.
[09-Sep-2009 15:39:36] <rmatte> t3los: oh
[09-Sep-2009 15:40:22] <t3los> Any ideas?
[09-Sep-2009 15:40:26] <rmatte> t3los: good question, I'd have to look in to that
[09-Sep-2009 15:41:29] <rmatte> t3los: I think the CPU one turns red if an event of class /Perf/CPU comes in
[09-Sep-2009 15:41:39] <rmatte> such as a threshold
\n[09-Sep-2009 15:41:58] <t3los> So they are meant to stay grey? As\nopposed to the others that are green when working right?
[09-Sep-2009 15:42:02] <t3los> That's what doesn't make sense.
[09-Sep-2009 15:43:20] <rmatte> t3los: which are grey?
[09-Sep-2009 15:44:24] <t3los> cpuPercentProcessorTime, memoryAvailableKBytes, memoryPagesPerSec
[09-Sep-2009 15:59:38] <jb> think im going to start an EqualLogic PS zenpack
[09-Sep-2009 16:01:44] <metalklesk> do it
[09-Sep-2009 16:02:40] <metalklesk> with more zenpacks, more robust zenoss will be (Yoda spoke)
[09-Sep-2009 16:04:56]\n<rmatte> nice, I'm hearing the guys talking about a call today, a\nswitch went down, they called the site, apparently someone had knocked\nthe switch off of the workbench/table or whatever that it was sitting\non, in to a pale of water
[09-Sep-2009 16:05:09] <rmatte> I love some of our clients
[09-Sep-2009 16:05:14] <rmatte>
[09-Sep-2009 16:05:36] <mrayzenoss> rmatte: that is wrong on so many levels
[09-Sep-2009 16:05:47] <rmatte> hehe
[09-Sep-2009 16:06:07] <rmatte> I've seen some pretty interesting stuff
[09-Sep-2009 16:06:49]\n<rmatte> a couple years back when I was a NOC analyst, a router\ndied at one of the sites because the coffee maker was sitting on top of\nit and someone spilled coffee on it
[09-Sep-2009 16:07:31]\n<rmatte> There was also an outage in northern canada caused by an\neagle trying to fly with a deer's head, it couldn't quite gain enough\naltitude and crashed in to a power line
[09-Sep-2009 16:08:18]\n<rmatte> There was a satellite site in northern canada that\ncouldn't be repaired because when the technician took his snowmobile up\na polarbear was blocking the door
[09-Sep-2009 16:09:35]\n<rmatte> There was a major outage in europe where someone broke\nin to one of the carrier's sites and stole 16 fiber cards which were\napparently extremely expensive, it took a day and a half to resolve the\noutage because they didn't have enough fiber cards in the country to\nreplace them with so they had to fly them in
[09-Sep-2009 16:09:41] <rmatte> I believe that was in London
[09-Sep-2009 16:09:47] <rmatte> if I remember correctly
[09-Sep-2009 16:09:55] <mrayzenoss> awesome
\n[09-Sep-2009 16:10:26] <rmatte> There was a site in Mexico that\nkept going down because people were coming and cutting down the copper\ncable running from the pole to the site and stealing it
[09-Sep-2009 16:10:31] <rmatte> it actually happened 3 times
\n[09-Sep-2009 16:10:50] <rmatte> one of the analysts was like "Why\ndon't just install a camera?" and someone was like "They'd probably\nsteal that too"
[09-Sep-2009 16:11:17]\n<rhettardo> that happened at my old job. some one came up w/ a\nladder and stole the security cameras
[09-Sep-2009 16:11:19] <mrayzenoss> lol
[09-Sep-2009 16:11:45] <rmatte> being an analyst was quite fun sometimes
[09-Sep-2009 16:11:49] <rmatte> hehe
[09-Sep-2009 16:11:54] <metalklesk> ajjajajajajajajajajaja
[09-Sep-2009 16:11:59] <metalklesk> nice stories
[09-Sep-2009 16:12:18] <metalklesk> stealing cables, sound like here in Chile
[09-Sep-2009 16:12:47] <metalklesk> some people are stealing cables in highways
[09-Sep-2009 16:13:05] <rmatte> copper can be sold for quite a bit of cash apparently
[09-Sep-2009 16:13:20] <metalklesk> shutting down systems of several companies
\n[09-Sep-2009 16:13:48] <rmatte> Oh, we were doing monitoring for\nBig Lots (the retail outlet)... during hurricane season it was always\nfun
[09-Sep-2009 16:14:13] <mrayzenoss> because of power loss or looting?
\n[09-Sep-2009 16:14:27] <rmatte> "Yeh, we've seen this site go\ndown." "Oh, yeh, that store, yeh, that store is no longer there." "What\ndo you mean no longer there?" "It was ripped apart by a hurricane."
[09-Sep-2009 16:14:33] <mrayzenoss> wow
[09-Sep-2009 16:14:48] <rmatte> yup
[09-Sep-2009 16:14:58] <metalklesk> wow
[09-Sep-2009 16:15:00] <rmatte> that was just last year, that happened to 2 of their stores
[09-Sep-2009 16:16:54] <malbon> hi all
[09-Sep-2009 16:17:03]\n<rmatte> lmao, one of my coworkers came and asked me how to\nconfigure WMI for a device in Zenoss, I just heard him say "Argh, I\nhate WMI!" from across the room
[09-Sep-2009 16:17:17] <rmatte> I'm guessing it's not going well
[09-Sep-2009 16:17:37] <rmatte> Probably to do with actually configuring the WMI access, not with Zenoss
\n[09-Sep-2009 16:17:47] <cam__> i installed zenoss from the debian\nrepository listed in the install guide. it seems to include its own\nversion of mysql--is that necessary?
[09-Sep-2009 16:18:10] <rmatte> cam__: it comes bundled with all of the necessary dependencies
\n[09-Sep-2009 16:18:37] <rmatte> cam__: If you really really want\nto, you could make it work with an existing mysql server if needed
[09-Sep-2009 16:19:05] <rmatte> cam__: the deb is basically just the Ubuntu stack installer wrapped in a deb file
\n[09-Sep-2009 16:20:53] <cam__> rmatte: ok. it's failing for some\nreason (zeoctl is not found). i was just digging around and thought\nthat it was weird that it didn't depend on mysql-server package rather\nthan bundling its own
[09-Sep-2009 16:21:07] <rmatte> zeoctl is one of the daemons that needs to run
[09-Sep-2009 16:21:29] <rmatte> cam__: it doesn't pull any dependencies down from aptitude
\n[09-Sep-2009 16:21:50] <rmatte> cam__: it's just a deb file to\nmake it so that the package can be managed with aptitude, but it's not\nfully integrated in to it
[09-Sep-2009 16:22:05] <cam__> rmatte: ok. i'll try installing with the stack installer instead
[09-Sep-2009 16:22:09]\n<rmatte> cam__: also, Zenoss required specific versions of things\nsuch as python, so it comes bundled with that as well, python2.4
\n[09-Sep-2009 16:22:30] <rmatte> cam__: the stack installer is\nwhat I use, it'll do the exact same install as the .deb did though
[09-Sep-2009 16:22:39] <rmatte> cam__: it just won't add it in to the package inventory
[09-Sep-2009 16:22:57] <cam__> rmatte: i see. i wonder why it didn't install zeoctl
[09-Sep-2009 16:23:13] <rmatte> rhettardo: updatedb && locate zeoctl as root
[09-Sep-2009 16:23:16] <rmatte> oops
[09-Sep-2009 16:23:20] <rmatte> rhettardo: updatedb && locate zeoctl as root
[09-Sep-2009 16:23:22] <rmatte> bah
[09-Sep-2009 16:23:24] <rhettardo> :O
[09-Sep-2009 16:23:35] <rmatte> I'm trying to type do but it's auto-completing as his nick
[09-Sep-2009 16:23:38] <rhettardo> ok but why?!
[09-Sep-2009 16:23:38] <rmatte> hehe
[09-Sep-2009 16:23:57] <cam__> haha, i did that but it's really not there
[09-Sep-2009 16:24:09] <rhettardo> you made need a sudo updatedb && locate zeoctl
[09-Sep-2009 16:24:11] <rmatte> cam__: weird, then I'd recommend trying the stack installer for sure
[09-Sep-2009 16:24:15] <rhettardo> i dont know. just a suggestion
[09-Sep-2009 16:24:40] <rmatte> rhettardo: I said as root, same difference lol
[09-Sep-2009 16:24:47] <rhettardo> oh i didnt see that
[09-Sep-2009 16:24:53] <rhettardo> all i saw was the highlighted lines w/ my bick
[09-Sep-2009 16:24:54] <rhettardo> nick
[09-Sep-2009 16:24:58] <rmatte> ah lol
\n[09-Sep-2009 16:25:11] <sacpinball> yo; do i need anything\ninstalled on my windows servers to collect performance stats; i have\nwmi setup and snmp, do i need snmp informant?
[09-Sep-2009 16:25:15] <rmatte> yeh, apparently my bitchx script auto-completes names if a colon is put after the word
\n[09-Sep-2009 16:25:53] <rmatte> sacpinball: not if you use my\nperformance zenpack:\nhttp://www.zenoss.com/community/projects/zenpacks/winsnmpperf
[09-Sep-2009 16:26:18] <sacpinball> sweet I will check it out
\n[09-Sep-2009 16:26:18] <rmatte> sacpinball: there's an even\nsimpler version of that ZenPack which should be publicly available next\nweek hopefully, but I can pass it to you if you want
[09-Sep-2009 16:26:30] <rmatte> sacpinball: are you looking to monitor multiple cores or just total CPU usage?
[09-Sep-2009 16:26:45] <sacpinball> probably both if possible depneing on the system
[09-Sep-2009 16:26:56] <rmatte> sacpinball: ok, then you'll want that pack
\n[09-Sep-2009 16:27:38] <rmatte> sacpinball: that's the advanced\nback, the simpler pack will just be for monitoring the total, and won't\nhave the different device groupings for each number of CPUs
\n[09-Sep-2009 16:28:38] <rmatte> sacpinball: it currently uses\nbash scripts to collect the data, make sure that you have bc and\nsnmpwalk installed
[09-Sep-2009 16:28:44] <rmatte> apt-get install snmpwalk bc
\n[09-Sep-2009 16:29:20] <rmatte> I think I can modify it to not\nuse bc at all, I'll probably do that within a couple of weeks
[09-Sep-2009 16:30:24] <sacpinball> that is sweet thanks
[09-Sep-2009 16:30:27] <sacpinball> i will try it out
[09-Sep-2009 16:30:51] <rmatte> np
[09-Sep-2009 16:50:16] <metalklesk> Zenoss Administration Guide is very educational
[09-Sep-2009 16:50:48] <rmatte> yup
[09-Sep-2009 16:50:56] <rmatte> it's definitely worth reading
\n[09-Sep-2009 16:51:14] <rmatte> I read through the whole thing\nstart to finish (minus the parts about enterprise components)
[09-Sep-2009 16:51:25] <mrayzenoss> the beta has an updated version available
[09-Sep-2009 16:51:30] <mrayzenoss> I'm writing up the blog post now
\n[09-Sep-2009 16:51:44] <rmatte> Jane Curry's Zenoss papers are\nalso worth reading: http://www.skills-1st.co.uk/papers/jcurry.html
[09-Sep-2009 16:52:11] <metalklesk> great
\n[09-Sep-2009 16:52:25] <rmatte> mrayzenoss: let me know when the\nstack installer is out for the beta and we'll setup a beta box
[09-Sep-2009 16:55:34] <jb> argh ok i need a python debugger
[09-Sep-2009 16:55:59] <jb> will they show you when your spacing is messed up?
[09-Sep-2009 16:56:09] <rmatte> yeh, they pretty much all do that
[09-Sep-2009 16:56:16] <rmatte> I'd recommend trying out IDLE
[09-Sep-2009 16:56:25] <rmatte> it's not very powerful, but it's nice for beginners
[09-Sep-2009 16:56:56] <rmatte> (I'm still a relative beginner myself)
[09-Sep-2009 16:56:58] <rmatte> hehe
[09-Sep-2009 16:57:19] <metalklesk> ok, i will leave the channel
[09-Sep-2009 16:57:19] <rmatte> my bash scripting skills are really good though
[09-Sep-2009 16:57:25] <rmatte> anyways, time for me to head out
[09-Sep-2009 16:57:33] <rmatte> have a nice evening guys
[09-Sep-2009 16:57:39] <metalklesk> in a few hours Chile vs Brasil
[09-Sep-2009 16:57:46] <rmatte> w00t
[09-Sep-2009 16:58:13] <jb> ok got it
[09-Sep-2009 16:58:14] <jb> lets try
[09-Sep-2009 17:00:09] <jb> ok..
[09-Sep-2009 17:00:18] <jb> so could you give me a quick runthrough?
[09-Sep-2009 17:00:22] <jb> i think i have some bad spacing
[09-Sep-2009 17:00:32] <metalklesk> bye !!
[09-Sep-2009 17:08:59] <Brad_K> i'm heading home. g'nite!
[09-Sep-2009 17:11:00]\n<LW> Anybody encounter cases where the snmp ipv6 services map\ndoes not return all of the ports that exist on a given host? we have a\ncase where we have java app listening on a high up port, and its never\ndetected, and when we manually run the model dumping the debug output,\nit never shows up in the list.... so we're thinking maybe its snmp and\nnot zenoss at all.... any ideas?
[09-Sep-2009\n17:19:15] <feutete> Is it possible to set up Zenoss in multiple\nlocations and have the remove installations push data to a central\nserver?
[09-Sep-2009 17:19:27] <feutete> I found this page--is this what I'm looking for?
\n[09-Sep-2009 17:19:28] <feutete>\nhttp://www.zenoss.com/community/docs/howtos/setting-up-multiple-zenperfsnmp-collectors-across-different-hosts
\n[09-Sep-2009 17:19:55] <feutete> essentially, the latency between\nour datacenters is too high for us to do all polling from a single\ninstance
[09-Sep-2009 17:20:17] <feutete> so we want to poll from a local instance, and push data to the central server
[09-Sep-2009 17:59:40] <ironpaw> morning
[09-Sep-2009 18:00:03] <ironpaw> feutete i want to set this up too
[09-Sep-2009 18:00:09] <ironpaw> but havne't had an opportunity yet
[09-Sep-2009 18:01:12]\n<feutete> a friend came across this as well:\nhttp://www.zenoss.com/Members/fdeckert/how-to-install-distributed-collectors/
[09-Sep-2009 18:01:39] <ironpaw> have seen both of those how tos but not used them yet.
[09-Sep-2009 23:31:17] <ironpaw> o.O
[10-Sep-2009 00:00:46] [disconnected at Thu Sep 10 00:00:46 2009]
[10-Sep-2009 00:00:46] [connected at Thu Sep 10 00:00:46 2009]
[10-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[10-Sep-2009 07:08:11] <ckrough> Is there an API guide for the Zope functions, like their is for the Classes?
\n[10-Sep-2009 07:37:46] <afink> hello everyone. I am having a\nproblem getting the performance graphs for centos to show up in\nzenoss. Zenoss says "Missing RRD File" laLoadInt5. Anyone know what I\ncan do to get that working?
[10-Sep-2009 07:40:39]\n<ckrough> that indicates that its not collecting what you are\ngraphing, or it doesnt have enough data yet. It takes 3 polling cycles\n(usually 5min) before there are enough datapoints in the rrd for it to\ngenerate graphs
[10-Sep-2009 07:40:43] <ckrough> how long have you waited?
[10-Sep-2009 07:55:12] <Apocalipse> morning
[10-Sep-2009 07:58:32] <ckrough> morning
\n[10-Sep-2009 08:10:57] <LW_> Anybody have an idea why zenmodler\nwould add this line to be monitored as a service on a host\nDEBUG:zen.IpV6ServiceMap:tcp new 16.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.22\n{'v6': 0L}, but not this line DEBUG:zen.IpV6ServiceMap:tcp new\n16.0.0.0.0.0.0.0.0.0.0.255.255.a.b.c.d.18580 {'v6': 0L} where a.b.c.d\nis the ipv4 addy of the machine... I see its a different address being\nbound to, but i would think it would still see it and understand "hey,
[10-Sep-2009 08:12:43] <afink> ckrough: months
[10-Sep-2009 08:13:36] <ckrough> afink: heh, unless your polling cycle is 2592000, it's probably something else
[10-Sep-2009 08:14:14] <metalklesk> hi everyone
[10-Sep-2009 08:14:54] <ckrough> afink: if you snmpget the laLoadInt5 directly, does the server return anything?
\n[10-Sep-2009 08:15:27] <afink> lol, \nhttp://picpaste.com/print1.png, I haven't tried I am not familiar with\nthat command. I'll look it up and give it a try.
[10-Sep-2009 08:16:07] <ckrough> afink: http://www.zenoss.com/community/docs/howtos/how-to-troubleshoot-snmp/
[10-Sep-2009 08:18:46] <afink> ckrough: laLoadInt5: Unknown Object Identifier (Sub-id not found: (top) -> laLoadInt5)
[10-Sep-2009 08:19:34] <afink> I have never had to set those up before for instance on suse..they're just there
[10-Sep-2009 08:28:08] <ckrough> afink: does this return anything? snmpget -v2c -cyourcommunity your.server.fqdn 1.3.6.1.4.1.2021.10.1.5.2
[10-Sep-2009 08:28:17] <ckrough> afink: "UCD-SNMP-MIB::laLoadInt.2 = INTEGER: 7" on my RHEL box
[10-Sep-2009 08:37:14] <adelbul> hello i read the dev doc about plugins but it is not very clear for me.
[10-Sep-2009 08:37:33] <adelbul> have you a good advice for a tutorial ?
[10-Sep-2009 08:38:11] <afink> ckrough: UCD-SNMP-MIB::laLoadInt.2 = No Such Object available on this agent at this OID
[10-Sep-2009 08:38:33] * afink googles
[10-Sep-2009 08:40:19] straterr1 is now known as straterra
\n[10-Sep-2009 08:46:51] <ckrough> afink: odd, on my RHEL5 boxes I\nget that data, but I have not installed anything other that net-snmp.
\n[10-Sep-2009 08:46:56] <ckrough> afink: this may help as well\nhttp://www.zenoss.com/community/docs/howtos/configuration-of-netsnmp-for-use-with-zenoss/
[10-Sep-2009 09:14:13] <rmatte> yeh, net-snmp should be all you need for a Linux box
[10-Sep-2009 09:18:20] <Apocalipse> hello people...
[10-Sep-2009 09:18:36] <Apocalipse> my zenoss box stopped working...
[10-Sep-2009 09:18:50] <Apocalipse> i restarted services and machine...
[10-Sep-2009 09:18:54] <Apocalipse> and nothing
[10-Sep-2009 09:22:12] <rmatte> stopped working how?
[10-Sep-2009 09:22:25] <rmatte> when you start zenoss do you get any errors?
[10-Sep-2009 09:22:30] <rmatte> which daemons aren't starting?
[10-Sep-2009 09:22:31] <rmatte> etc...
[10-Sep-2009 09:25:26] <Apocalipse> zenhub, zenactions
[10-Sep-2009 09:26:16] <rmatte> what version of Zenoss are you running?
[10-Sep-2009 09:26:33] <Apocalipse> 2.4.1 in freebsd
[10-Sep-2009 09:26:41] <rmatte> oh right, FreeBSD lol
[10-Sep-2009 09:26:48] <Apocalipse> kkkkkk
[10-Sep-2009 09:26:57] <rmatte> you'll want to check $ZENHOME/log/zenhub.log for starters
[10-Sep-2009 09:27:07] <rmatte> look for any errors
[10-Sep-2009 09:27:10] <rmatte> same with zenactions.log
\n[10-Sep-2009 09:29:24] <Apocalipse> MySQLConnectionError: (2002,\n"Can't connect to local MySQL server through socket '/tmp/mysql.sock'\n(2)")
[10-Sep-2009 09:29:44] <rmatte> ok, so your mysql server isn't started
[10-Sep-2009 09:29:47] <rmatte> that's your issue
[10-Sep-2009 09:30:02] <rmatte> where did you put the zenoss-stack script?
[10-Sep-2009 09:30:11] <rmatte> or the zenoss startup script in general
[10-Sep-2009 09:30:31] <rmatte> /etc/rc.d?
[10-Sep-2009 09:30:34] <Apocalipse> mysql is already started...
[10-Sep-2009 09:30:46] <rmatte> ok, do a "zenhub -v10 start"
[10-Sep-2009 09:30:48] <rmatte> see what it says
[10-Sep-2009 09:32:01] <Apocalipse> zenhub {run|start|stop|restart|status|help|genconf|genxmlconfigs} [options]
[10-Sep-2009 09:32:03] <Apocalipse>
[10-Sep-2009 09:32:09] <rmatte> sorry
[10-Sep-2009 09:32:13] <rmatte> ok, do a "zenhub -v10 run"
[10-Sep-2009 09:32:24] <rmatte> no wait
[10-Sep-2009 09:32:28] <rmatte> start should have worked
[10-Sep-2009 09:32:32] <Apocalipse> same
[10-Sep-2009 09:32:33] <rmatte> are you the zenoss user?
[10-Sep-2009 09:32:37] <Apocalipse> yes
[10-Sep-2009 09:32:43] <rmatte> try "zenhub run -v10"
[10-Sep-2009 09:32:53] <rmatte> or "zenhub start -v10"
[10-Sep-2009 09:32:55] <Apocalipse> ./zenhub start
[10-Sep-2009 09:33:00] <Apocalipse> ok
[10-Sep-2009 09:33:07] <rmatte> you shouldn't need to do ./
[10-Sep-2009 09:33:25] <rmatte> If you do, then there's a problem with your path or environment variables
[10-Sep-2009 09:33:39] <Apocalipse> zenhub status program running; pid=2188
[10-Sep-2009 09:33:51] <rmatte> do a "zenoss status"
[10-Sep-2009 09:33:55] <rmatte> what's the full output?
[10-Sep-2009 09:34:22] <Apocalipse> only this is not working...Daemon: zensyslog not running
[10-Sep-2009 09:34:24] <rmatte> paste it slowly in pieces (3 or 4 lines at a time), otherwise you'll flood out
[10-Sep-2009 09:34:35] <rmatte> ok, so do: zensyslog start
[10-Sep-2009 09:34:37] <Apocalipse> Daemon: zeoctl program running; pid=745
[10-Sep-2009 09:34:37] <Apocalipse> Daemon: zopectl program running; pid=749
[10-Sep-2009 09:34:37] <Apocalipse> Daemon: zenhub program running; pid=2188
[10-Sep-2009 09:34:45] <Apocalipse> Daemon: zenjobs program running; pid=839
[10-Sep-2009 09:34:45] <Apocalipse> Daemon: zenping program running; pid=925
[10-Sep-2009 09:34:45] <Apocalipse> Daemon: zensyslog not running
[10-Sep-2009 09:34:52] <Apocalipse> Daemon: zenstatus program running; pid=967
[10-Sep-2009 09:34:53] <Apocalipse> Daemon: zenactions program running; pid=1645
[10-Sep-2009 09:34:53] <Apocalipse> Daemon: zentrap program running; pid=1079
[10-Sep-2009 09:35:00] <Apocalipse> Daemon: zenmodeler program running; pid=1100
[10-Sep-2009 09:35:00] <Apocalipse> Daemon: zenperfsnmp program running; pid=1137
[10-Sep-2009 09:35:00] <Apocalipse> Daemon: zencommand program running; pid=1171
[10-Sep-2009 09:35:07] <Apocalipse> Daemon: zenprocess program running; pid=1207
[10-Sep-2009 09:35:07] <Apocalipse> Daemon: zenwin program running; pid=1249
[10-Sep-2009 09:35:07] <Apocalipse> Daemon: zeneventlog program running; pid=1291
[10-Sep-2009 09:35:14] <rmatte> zensyslog start
[10-Sep-2009 09:36:24] <Apocalipse> not running
[10-Sep-2009 09:36:32] <Apocalipse> but this never run before
[10-Sep-2009 09:37:01] <rmatte> well, you kind of need it
[10-Sep-2009 09:37:14] <rmatte> if you ever expect to process syslogs
[10-Sep-2009 09:37:32] <rmatte> check $ZENHOME/log/zensyslog.log
[10-Sep-2009 09:37:40] <rmatte> for errors
[10-Sep-2009 09:38:57] <Apocalipse> 0 size of file
[10-Sep-2009 09:39:14] <rmatte> so the log isn't populated at all?
[10-Sep-2009 09:39:58] <Apocalipse> yes
[10-Sep-2009 09:40:18] <rmatte> ok, well, you're running on FreeBSD so lord only knows why that daemon won't start
[10-Sep-2009 09:40:28] <Apocalipse> yes
[10-Sep-2009 09:40:29] <rmatte> it's a pretty necessary daemon to have running though
[10-Sep-2009 09:40:36] <Apocalipse> i know...
[10-Sep-2009 09:40:42] <rmatte> if you do zensyslog run -v10
[10-Sep-2009 09:40:45] <rmatte> what do you see?
[10-Sep-2009 09:40:46] <Apocalipse> but all things were working...
\n[10-Sep-2009 09:41:10] <rmatte> all the things that you were\nusing were working, but if you ever wanted to send syslog messages to\nZenoss it would never work
[10-Sep-2009 09:41:22] <Apocalipse> zensyslog run -v10 ZenSocket Error: Unable to bind to listen port (error code 48)
[10-Sep-2009 09:41:41] <rmatte> do you have some sort of firewall on that system?
[10-Sep-2009 09:41:46] <Apocalipse> none
[10-Sep-2009 09:42:30] <rmatte> you're probably already running syslogd on there which is attached to that port already
[10-Sep-2009 09:42:41] <rmatte> or syslog-ng
[10-Sep-2009 09:42:54] <rmatte> are you running either of those?
[10-Sep-2009 09:43:45] <rmatte> looks like it's quite a common issue on FreeBSD
[10-Sep-2009 09:43:48] <rmatte> http://forums.zenoss.com/viewtopic.php?p=16030
[10-Sep-2009 09:43:52] <Apocalipse> syslog is running
[10-Sep-2009 09:43:58] <rmatte> ok, kill syslog
[10-Sep-2009 09:44:02] <rmatte> and make sure it doesn't run
[10-Sep-2009 09:44:07] <rmatte> then start zensyslog
[10-Sep-2009 09:45:26] <rmatte> ah
[10-Sep-2009 09:45:29] <rmatte> here are the steps...
\n[10-Sep-2009 09:45:36] <rmatte> syslog (port 514) will probably\nalso show in your output. You don't want to disable syslog completely,\nas you do want to receive logging messages. However, you don't need to\nhave this port open to do so. In your /etc/rc.conf file, make sure\nsyslog is enabled and add a second line with some options:
[10-Sep-2009 09:45:37] <Apocalipse> zensyslog run -v10
[10-Sep-2009 09:45:42] <rmatte> syslogd_enable="YES"
[10-Sep-2009 09:45:42] <rmatte> syslogd_flags="-ss"
\n[10-Sep-2009 09:45:49] <rmatte> Those two sses (make sure you\nhave two, not just one) in the flags will disable logging from remote\nhosts and close that port, but still allow your localhost to keep its\nlogging capabilities.
[10-Sep-2009 09:46:06] <rmatte> so yeh, I forgot that FreeBSD does it differently
[10-Sep-2009 09:46:11] <rmatte> so you don't want to completely kill it
[10-Sep-2009 09:46:13] <rmatte> just use those options
[10-Sep-2009 09:46:31] <Apocalipse> DEBUG:zen.ZenSyslog:Starting PBDaemon initialization
[10-Sep-2009 09:46:32] <Apocalipse> INFO:zen.ZenSyslog:Connecting to localhost:8789
[10-Sep-2009 09:46:32] <Apocalipse> DEBUG:zen.ZenSyslog:Logging in as admin
[10-Sep-2009 09:46:32] <Apocalipse> INFO:zen.ZenSyslog:Connected to ZenHub
[10-Sep-2009 09:46:32] <Apocalipse> DEBUG:zen.ZenSyslog:Setting up initial services: EventService
[10-Sep-2009 09:46:32] <Apocalipse> DEBUG:zen.ZenSyslog:Chaining getInitialServices with d2
[10-Sep-2009 09:46:34] <Apocalipse> DEBUG:zen.ZenSyslog:Loaded service EventService from zenhub
[10-Sep-2009 09:46:55] <Apocalipse> last 2 lines
[10-Sep-2009 09:47:00] <Apocalipse> INFO:zen.ZenSyslog:zensyslog shutting down
[10-Sep-2009 09:47:00] <Apocalipse> INFO:zen.ZenSyslog:0 events processed in 0.00 seconds
[10-Sep-2009 09:47:18] <rmatte> that's when you do a "run" right?
[10-Sep-2009 09:47:24] <Apocalipse> yes
[10-Sep-2009 09:47:28] <rmatte> now do a "start"
[10-Sep-2009 09:47:31] <Apocalipse> i did start
[10-Sep-2009 09:47:32] <rmatte> and it should start properly
[10-Sep-2009 09:47:40] <Apocalipse> program running; pid=3867
[10-Sep-2009 09:48:26] <rmatte> good
[10-Sep-2009 09:48:31] <rmatte> that's fixed...
[10-Sep-2009 09:48:38] <rmatte> did you follow the steps I pasted?
[10-Sep-2009 09:48:42] <Apocalipse> yes
[10-Sep-2009 09:48:47] <rmatte> you need syslog running, so you have to use those steps specifically
[10-Sep-2009 09:48:54] <rmatte> otherwise you won't get local system log messages
[10-Sep-2009 09:49:13] <rmatte> but you need to disable it from using the external port
[10-Sep-2009 09:49:47] <rmatte> so those 2 lines are in your /etc/rc.conf and syslog is running again?
[10-Sep-2009 09:50:10] <Apocalipse> yes
[10-Sep-2009 09:50:15] <rmatte> ok good
[10-Sep-2009 09:50:27] <rmatte> do a zenoss status and paste the output again please
[10-Sep-2009 09:52:15] <Apocalipse> Daemon: zeoctl program running; pid=2728
[10-Sep-2009 09:52:15] <Apocalipse> Daemon: zopectl program running; pid=2732
[10-Sep-2009 09:52:15] <Apocalipse> Daemon: zenhub program running; pid=2785
[10-Sep-2009 09:52:15] <Apocalipse> Daemon: zenjobs program running; pid=2822
[10-Sep-2009 09:52:15] <Apocalipse> Daemon: zenping program running; pid=2907
[10-Sep-2009 09:52:16] <Apocalipse> Daemon: zensyslog program running; pid=3867
[10-Sep-2009 09:52:18] <Apocalipse> Daemon: zenstatus program running; pid=2942
[10-Sep-2009 09:52:24] <Apocalipse> Daemon: zenactions program running; pid=2980
[10-Sep-2009 09:52:25] <Apocalipse> Daemon: zentrap program running; pid=3070
[10-Sep-2009 09:52:25] <Apocalipse> Daemon: zenmodeler program running; pid=3075
[10-Sep-2009 09:52:26] <Apocalipse> Daemon: zenperfsnmp program running; pid=3108
[10-Sep-2009 09:52:28] <Apocalipse> Daemon: zencommand program running; pid=3141
[10-Sep-2009 09:52:30] <Apocalipse> Daemon: zenprocess program running; pid=3174
[10-Sep-2009 09:52:32] <Apocalipse> Daemon: zenwin program running; pid=3216
[10-Sep-2009 09:52:34] <Apocalipse> Daemon: zeneventlog program running; pid=3258
[10-Sep-2009 09:52:41] <rmatte> good stuff, ok, now what is the issue at this point?
[10-Sep-2009 09:52:47] <rmatte> are you not able to hit the web interface?
[10-Sep-2009 09:53:08] <Apocalipse> web interface is ok...
[10-Sep-2009 09:53:37] <Apocalipse> ok...
[10-Sep-2009 09:53:45] <Apocalipse> all things restarted
[10-Sep-2009 09:54:00] <Apocalipse> thx "master" ryan
[10-Sep-2009 09:54:01] <Apocalipse>
[10-Sep-2009 09:54:15] <Apocalipse> brb
[10-Sep-2009 09:54:27] <rmatte> np
\n[10-Sep-2009 09:57:42] <rmatte> If I were you I'd reboot the\nserver and make sure that it works after the reboot without having to\nstart anything manually
[10-Sep-2009 10:00:00] <rmatte> you may also want to do "zenpatch 14181" to patch a bad security bug
[10-Sep-2009 10:00:13] <rmatte> then restart zenoss
[10-Sep-2009 10:09:08] <vinzpr> hi anyone here knows how to Display CPU Threshold Events as a Percentage
[10-Sep-2009 10:09:20] <vinzpr> also in memory and interface
[10-Sep-2009 10:09:49] <vinzpr> i tried this procedure but it didnt work
[10-Sep-2009 10:09:50] <vinzpr> http://www.zenoss.com/Members/cluther/display-memory-threshold-events-as-a-percentage/
[10-Sep-2009 10:16:42] <rmatte> vinzpr: well, aren't CPU datapoints usually just a value between 1 and 100 anyways?
[10-Sep-2009 10:18:42] <vinzpr> yes but it only shows me the value of memory bytes
[10-Sep-2009 10:18:42] <vinzpr> not the percentage
[10-Sep-2009 10:18:42] <rmatte> you just said CPU Threshold
[10-Sep-2009 10:18:42] <rmatte> not Memory
[10-Sep-2009 10:18:51] <vinzpr> example: threshold of Memory Warning notmet: current value 7124.00
[10-Sep-2009 10:18:55] <vinzpr> also in cpu
[10-Sep-2009 10:19:07] <rmatte> what's CPU look like then?
[10-Sep-2009 10:19:07] <vinzpr> 617 instead of 6.17
[10-Sep-2009 10:19:17] <rmatte> ah
[10-Sep-2009 10:19:27] <rmatte> it's probably doing it out of 10000 then or something
[10-Sep-2009 10:19:38] <rmatte> what type of device is this for?
[10-Sep-2009 10:20:14] <vinzpr> linux
[10-Sep-2009 10:21:15] <rmatte> so cluther's transform doesn't work for Linux?
[10-Sep-2009 10:21:20] <vinzpr> yes
[10-Sep-2009 10:21:32] <rmatte> did you modify the match line to match what you're seeing?
[10-Sep-2009 10:21:40] <rmatte> It should be more like:
[10-Sep-2009 10:21:41] <rmatte> match = re.search('threshold of Free Memory [^:]+: current value ([\\d\\.]+)', evt.message)
[10-Sep-2009 10:21:43] <rmatte> oops
[10-Sep-2009 10:21:44] <rmatte> one sec
[10-Sep-2009 10:22:05] <rmatte> match = re.search('threshold of Memory Warning notmet [^:]+: current value ([\\d\\.]+)', evt.message)
[10-Sep-2009 10:22:12] <rmatte> in your case
[10-Sep-2009 10:22:25] <vinzpr> ah
[10-Sep-2009 10:22:35] <rmatte> otherwise the transform will never kick off
[10-Sep-2009 10:22:37] <vinzpr> ok you have to search the string of alert
[10-Sep-2009 10:22:46] <vinzpr> ok..let me check that
[10-Sep-2009 10:22:51] <rmatte> as for the CPU, that may be a bit more difficult...
[10-Sep-2009 10:22:54] <vinzpr> how about in cpu
[10-Sep-2009 10:23:48] <rmatte> just need to figure out how to extract the CPU value...
[10-Sep-2009 10:25:00] <vinzpr> ok
[10-Sep-2009 10:25:06] <rmatte> I'm not sure if avail = float(match.groups()[0]) * 1024
[10-Sep-2009 10:25:16] <rmatte> is actually pulling from the value in the threshold
[10-Sep-2009 10:25:21] <vinzpr> how about getting like this device.hw.totalMemory
[10-Sep-2009 10:25:32] <rmatte> no, you can't do that
[10-Sep-2009 10:25:36] <vinzpr> i dont know on device.hw.totalcpu?
[10-Sep-2009 10:25:43] <rmatte> that's pulling the total amount of memory on the box, which is a concrete value
[10-Sep-2009 10:25:47] <vinzpr> where can i get that list
[10-Sep-2009 10:25:51] <rmatte> the avail line is what we need to work with
[10-Sep-2009 10:25:56] <vinzpr> ok
[10-Sep-2009 10:26:18] <rmatte> I'll try to come up with something...
[10-Sep-2009 10:26:25] <vinzpr> thanks
[10-Sep-2009 10:27:01] <rmatte> give me a full example of what one of your CPU thresholds would look like as an event
[10-Sep-2009 10:27:05] <rmatte> I need to adjust the match line
[10-Sep-2009 10:27:16] <vinzpr> ok please wait
[10-Sep-2009 10:28:00]\n<adelbul> hello, il have a script on a opensolaris system which\nverify the status of an LSI Perc RAID command. I try to run this script\nwith zencommand but i believe that i don't understand the whole concept.
[10-Sep-2009 10:28:04] <vinzpr> threshold of Load Warning exceeded: current value 626.00
[10-Sep-2009 10:28:13] <vinzpr> it supposed to be 6.26
[10-Sep-2009 10:28:24] <adelbul> I create a datasource named : "LsiPerc6iCheckOpenSolaris"
[10-Sep-2009 10:28:36] <rmatte> and 626 would be 6.26%?
[10-Sep-2009 10:28:44] <rmatte> or 62.6%?
[10-Sep-2009 10:28:57] <rmatte> oh wait, is that for load average?
[10-Sep-2009 10:29:00] <adelbul> this datasource is checking the status of a Raid Controller in the server.
[10-Sep-2009 10:29:24] <adelbul> this datasource call a script named "/export/home/zenoss/scripts/check_perc.sh"
[10-Sep-2009 10:29:40] <adelbul> this script is on the client side on the opensolaris server.
[10-Sep-2009 10:29:50] <adelbul> and it is the result of the zentestcommand :
\n[10-Sep-2009 10:30:12] <adelbul> [zenoss@srvsurvey ~]$\nzentestcommand -d srvnas03\n--datasource=LsiPerc6iCheckOpenSolaris \n/bin/sh: /export/home/zenoss/scripts/check_perc.sh: no such file or\ndirectory
[10-Sep-2009 10:30:32] <vinzpr> its for load average
[10-Sep-2009 10:30:37] <rmatte> adelbul: it needs to be an SSH command based datasource
[10-Sep-2009 10:30:43] <rmatte> not a regular command based datasource
[10-Sep-2009 10:31:09] <adelbul> rmatte: yes you are right. then i enabled SSH in the datasource.
[10-Sep-2009 10:31:22] <vinzpr> rmatte: is for load average
[10-Sep-2009 10:31:25] <adelbul> rmatte: but the result is the same.
\n[10-Sep-2009 10:31:36] <rmatte> zCommandUsername and\nzCommandPassword in zProperties need to be populated with a valid\nusername and password for the device
[10-Sep-2009 10:31:49] <rmatte> vinzpr: k
[10-Sep-2009 10:33:11] <rmatte> vinzpr: try...
[10-Sep-2009 10:33:12] <rmatte> import re
\n[10-Sep-2009 10:33:12] <rmatte> match = re.search('threshold of\nLoad Warning exceeded [^:]+: current value ([\\d\\.]+)', evt.message)
[10-Sep-2009 10:33:12] <rmatte> if match and device:
[10-Sep-2009 10:33:12] <rmatte> load = float(match.groups()[0])
[10-Sep-2009 10:33:12] <rmatte> load_used = load / 100
\n[10-Sep-2009 10:33:13] <rmatte> evt.summary = "Load Average\nthreshold exceeded: current value %3.1f%%" % (percent_used)
[10-Sep-2009 10:33:26] <rmatte> hopefully it'll work
[10-Sep-2009 10:34:44] <rmatte> oops wait
[10-Sep-2009 10:34:55] <rmatte> replace (percent_used) at the end with (load_used)
[10-Sep-2009 10:34:58] <rmatte> I forgot to do that
[10-Sep-2009 10:35:47] <vinzpr> ok thank you..let me try that
[10-Sep-2009 10:36:59] <jb> anybody here user 64bit counters?
[10-Sep-2009 10:38:51] <rmatte> adelbul: the script should also output somewhat like: OK|LsiPerc6iCheckOpenSolaris=value
[10-Sep-2009 10:39:41] <rmatte> adelbul: and it should output "Unknown" if it's not able to check
[10-Sep-2009 10:39:53] <rmatte> jb: not really, why, what's up?
[10-Sep-2009 10:40:08] <adelbul> rmatte: return 0 or 1 is not good ?
[10-Sep-2009 10:40:08] <skipzoid> hello all
[10-Sep-2009 10:40:21] <jb> rmatte: well, I am trying to graph some COUNTER64 values
[10-Sep-2009 10:40:32] <jb> and just wondering if I need to do anythin special
[10-Sep-2009 10:40:36] <rmatte> nope
[10-Sep-2009 10:40:53] <rmatte> treat it the same way as you would for 32bit counters
\n[10-Sep-2009 10:41:38] <skipzoid> can someone help a newbie\nunderstand how a MIB file works inconjunction with a monitored device?\n- do I have to tell the device about the mib file or something ?
[10-Sep-2009 10:41:59] <rmatte> skipzoid: Mib files are only used to translate SNMP traps in Zenoss
\n[10-Sep-2009 10:42:20] <rmatte> skipzoid: when you're designing a\nmonitoring template to monitor a device with you need to specify the\nOIDs that you want to poll
[10-Sep-2009 10:42:36] <rmatte> skipzoid: Mibs are not required for actually polling data from devices in Zenoss
\n[10-Sep-2009 10:43:11] <skipzoid> i'm very new to this, so the\ndata is polled for? I thought the device sent it as and when an event\nhappend?
[10-Sep-2009 10:43:24] <jb> ah i think it was cause I was using V1
\n[10-Sep-2009 10:43:49] <rmatte> skipzoid: some data is polled\nfor, some data is sent from the device to Zenoss when an event happens\nwhich is called an SNMP traps
[10-Sep-2009 10:44:14]\n<rmatte> skipzoid: you don't need to actually tell Zenoss about\nMibs, provided you load the correct Mibs for the devices you are\nmonitoring they will automatically translate the traps that come in
\n[10-Sep-2009 10:44:38] <rmatte> skipzoid: examples of data that\nis polled for is performance data, CPU Usage, Memory Usage, Interface\ntraffic/errors, stuff like that
[10-Sep-2009 10:45:25] <rmatte> skipzoid: disk usage is another example of something that is polled for
\n[10-Sep-2009 10:45:34] <skipzoid> i think i understand that, i'm\ntrying to setup a monitor on a device that is duel power supply and\nseveral fans and when they fail I already know it sents a trap to a\nconfigured destination, i'm just trying to get zenoss to receive and\nact by way of sending a text message to call for help
[10-Sep-2009 10:46:07] <rmatte> what types of devices?
[10-Sep-2009 10:46:10] <rmatte> like, what brand?
\n[10-Sep-2009 10:47:32] <skipzoid> its a video server - propriety\nto harris, but they do supply a mib file for the device to use with\nyour own monitoring if you deside not to buy their monitoring system
[10-Sep-2009 10:48:47] <rmatte> ok, so just load that Mib file and you're done
[10-Sep-2009 10:48:47] <skipzoid> its that easy ?
[10-Sep-2009 10:48:47] <rmatte> it'll translate the traps
[10-Sep-2009 10:48:47] <rmatte> well, sort of
\n[10-Sep-2009 10:48:47] <rmatte> it'll translate the traps, but\nthen you need to write python scripts for each trap called transforms
[10-Sep-2009 10:48:48] <rmatte> they'll ensure that when the traps come in the message is human readable
\n[10-Sep-2009 10:49:11] <rmatte> you don't necessarily need to\nwrite transforms, it's really up to you, but you won't have to login to\nZenoss to dig for trap details since you'll be able to script Zenoss to\nput the details right in to the message
[10-Sep-2009\n10:49:32] <skipzoid> ive looked at the mib and there is already\nhuman readable errors - have i missed the point is that not what a mib\nfile does? translates the numbers into human form?
[10-Sep-2009 10:49:41] kevin7kal_ is now known as kevin7kal
[10-Sep-2009 10:49:56] <rmatte> that's correct, it does... but what will happen is this...
\n[10-Sep-2009 10:50:56] <rmatte> the trap will come in, and Zenoss\nwill create an event called something like "snmp trap\nclogMessageGenerated" or whatever
[10-Sep-2009 10:51:14] <rmatte> and if you click on the icon next to the event a details window will open
\n[10-Sep-2009 10:51:26] <rmatte> and if you click on the details\ntab in that window, you'll see what values the trap translated
\n[10-Sep-2009 10:51:41] <rmatte> now, to actually get those values\nin to actualy event messages when that sort of trap comes in you need\nto do a transform
[10-Sep-2009 10:51:56] <rmatte> I'll give you an example...
[10-Sep-2009 10:52:08]\n<adelbul> rmatte : in which section of the documentation i can\nfind the output formatting like this one :\nOK|LsiPerc6iCheckOpenSolaris=value ?
[10-Sep-2009\n10:52:17] <rmatte> in the case of the clogMessageGenerated trap,\nthat's a trap that's sent from a Cisco device for critical log\nmessages...
[10-Sep-2009 10:52:30] <rmatte> it originally came in as "snmp trap clogMessageGenerated"
\n[10-Sep-2009 10:52:44] <rmatte> when I went to the details tab I\nsaw different values (parts of the message) with the following names...
[10-Sep-2009 10:53:10] <rmatte> clogHistFacility, clogHistMsgName, clogHistMsgText, and clogHistSeverity
[10-Sep-2009 10:53:13] <vinzpr> rmatte: i tried it but the output is the same-
[10-Sep-2009 10:53:14] <vinzpr> threshold of Memory Warning not met: current value 97120.00
[10-Sep-2009 10:53:22] <rmatte> so clogHistSeverity was a number representing the severity of the event...
[10-Sep-2009 10:53:32] <rmatte> clogHisMsgText was the message
[10-Sep-2009 10:53:34] <rmatte> etc...
[10-Sep-2009 10:53:45] <skipzoid> i think i'm begining to understand now
\n[10-Sep-2009 10:53:59] <rmatte> I created a transform to make the\nmessages come in with the details in the summary instead of it just\nlooking like "snmp trap blah"
[10-Sep-2009 10:54:03] <rmatte> here's the transform I used
[10-Sep-2009 10:54:08] <vinzpr> import re
[10-Sep-2009 10:54:08] <vinzpr> match = re.search('threshold of Memory Warning not met [^:]+: current value ([\\d\\.]+)', evt.message)
[10-Sep-2009 10:54:08] <vinzpr> if match and device:
[10-Sep-2009 10:54:08] <vinzpr> avail = float(match.groups()[0]) * 1024
[10-Sep-2009 10:54:08] <vinzpr> total = device.hw.totalMemory
[10-Sep-2009 10:54:09] <vinzpr> percent_used = (1 - (avail / total)) * 100
[10-Sep-2009 10:54:11] <vinzpr> evt.summary = "Memory threshold exceeded: current value %3.1f%%" % (percent_used)
[10-Sep-2009 10:54:13] <rmatte> facility = getattr(evt, 'clogHistFacility', 'Unknown')
[10-Sep-2009 10:54:13] <rmatte> msgname = getattr(evt, 'clogHistMsgName', 'Unknown')
[10-Sep-2009 10:54:13] <rmatte> msgtext = getattr(evt, 'clogHistMsgText', 'Unknown')
[10-Sep-2009 10:54:13] <rmatte> histsev = getattr(evt, 'clogHistSeverity', 'Unknown')
[10-Sep-2009 10:54:13] <rmatte> import re
[10-Sep-2009 10:54:14] <rmatte> if re.match("decrypt: mac verify failed", msgtext):
[10-Sep-2009 10:54:16] <rmatte> evt._action = "drop"
[10-Sep-2009 10:54:18] <rmatte> elif re.match("Interface FastEthernet", msgtext):
[10-Sep-2009 10:54:20] <rmatte> evt.severity = 0
[10-Sep-2009 10:54:22] <rmatte> else:
[10-Sep-2009 10:54:24] <rmatte> evt.severity = {0:5, 1:5, 2:5, 3:4, 4:4, 5:3, 6:3, 7:0}[histsev]
[10-Sep-2009 10:54:26] <rmatte> evt.component = 'SYSLOG-%s' % (histsev)
[10-Sep-2009 10:54:28] <rmatte> evt.summary = '%s-%s: %s' % (facility,msgname,msgtext)
[10-Sep-2009 10:54:31] <rmatte> now, I know that's a bit complex, but to break it down to simpler form...
[10-Sep-2009 10:54:36] <rmatte> histsev = getattr(evt, 'clogHistSeverity', 'Unknown')
\n[10-Sep-2009 10:54:54] <rmatte> that line grabs the value that\ncorresponded to clogHistSeverity and stores it in a variable called\nhistsev
[10-Sep-2009 10:55:11] <rmatte> then if\nyou want to set the event summary to something custom with the value\nincluded you'd do something like...
[10-Sep-2009 10:55:39] <rmatte> evt.summary = 'This is the value for histsev: %s' % (histsev)
[10-Sep-2009 10:55:48] <rmatte> if you wanted to do multiple, you could do...
[10-Sep-2009 10:56:08] <rmatte> evt.summary = 'This is the value for histsev: %s and this is the value for msgtext: %s' % (histsev, msgtext)
\n[10-Sep-2009 10:56:29] <rmatte> you can also change other aspects\nof the event by setting things like evt.severity, evt.message, etc...
[10-Sep-2009 10:56:32] <rmatte> understand?
[10-Sep-2009 10:56:58] <rmatte> vinzpr: hmmm
[10-Sep-2009 10:56:59] <skipzoid> its becoming clearer - i think i need to trigger some events and then try this,
[10-Sep-2009 10:57:33] <rmatte> skipzoid: you should download the Zenoss Admin Guide and read it
[10-Sep-2009 10:57:34] <rmatte> all of it
[10-Sep-2009 10:57:41] <rmatte> you'll learn a lot of the basics from it
[10-Sep-2009 10:58:17] <skipzoid> i have and printed it for betime reading
[10-Sep-2009 10:58:17] <rmatte> cool
[10-Sep-2009 10:58:49] <Troubadix09> skipzoid: and read the community FAQ, greatful help at the beginning
[10-Sep-2009 10:59:45] <rmatte> vinzpr: what event class is the threshold coming in as?
\n[10-Sep-2009 10:59:47] <skipzoid> will do - this has been most\nhelpful too - a lot of irc channels dont like newbies and often say\nunhelpful things like - google is your friend or its in the manual...\nrefreshing not to get that here
[10-Sep-2009 11:00:04] <rmatte> skipzoid: nah, we help as best we can here
[10-Sep-2009 11:00:12] <vinzpr> rmatte:/Events /Perf /Memory
[10-Sep-2009 11:00:13] <skipzoid> appreciated
[10-Sep-2009 11:00:17] <rmatte>
[10-Sep-2009 11:00:20] <rmatte> vinzpr: ok...
[10-Sep-2009 11:00:24] <Troubadix09> skipzoid: and install the ZenPacks.community.mib_utils-Zenpack, it help to interpret or search in MIBs
[10-Sep-2009 11:01:25] <skipzoid> just looking for it now
[10-Sep-2009 11:02:15] <rmatte> vinzpr: try match = re.search('threshold of Memory Warning [^:]+: current value ([\\d\\.]+)', evt.summary)
[10-Sep-2009 11:02:18] <Apocalipse> hey...people...i cannot see any device in "network map"...
[10-Sep-2009 11:02:37] <rmatte> vinzpr: shortening the message a bit and checking evt.summary instead of evt.message just in case
[10-Sep-2009 11:03:05] <rmatte> Apocalipse: you need to enter a starting point in the top left (an IP or a hostname)
[10-Sep-2009 11:03:31] <rmatte> Apocalipse: A default starting point can be specified under the settings for your username as well.
[10-Sep-2009 11:03:51] <jb> hm
[10-Sep-2009 11:04:00] <jb> om.size = "%d" % (getattr(om, 'size', 0) / 1024
[10-Sep-2009 11:04:06] <jb> you see anythin wrong with that?
[10-Sep-2009 11:04:16] <rmatte> yeh, you're missing a close bracket
[10-Sep-2009 11:04:18] <rmatte> at the end
[10-Sep-2009 11:04:54] <Apocalipse> x.x.x.x/yy
[10-Sep-2009 11:04:55]\n<vinzpr> rmatte: like this match = re.search('threshold of Memory\nWarning not met [^:]+: current value ([\\d\\.]+)', evt.summary)
[10-Sep-2009 11:04:56] <jb> oh
[10-Sep-2009 11:04:57] <jb> oops.
[10-Sep-2009 11:05:15] <rmatte> vinzpr: take the not met part out just for the heck of it
[10-Sep-2009 11:05:29] <vinzpr> ok let me do it
[10-Sep-2009 11:05:35] <rmatte> match = re.search('threshold of Memory Warning [^:]+: current value ([\\d\\.]+)', evt.summary)
[10-Sep-2009 11:07:08]\n<rmatte> well, looks like my project for the rest of the day is\nto write a custom device availability report in python (standalone\nscript)
[10-Sep-2009 11:07:14] <vinzpr> rmatte: threshold of Memory Critical not met: current value 33356.00
[10-Sep-2009 11:07:27] <rmatte> vinzpr: oh, well there's your problem
[10-Sep-2009 11:07:36] <rmatte> Memory Critical, not Memory Warning lol
[10-Sep-2009 11:07:43] <rmatte> match = re.search('threshold of Memory [^:]+: current value ([\\d\\.]+)', evt.summary)
[10-Sep-2009 11:07:44] <rmatte> use that
[10-Sep-2009 11:08:03] <rmatte> I just assumed that you only had the one threshold
[10-Sep-2009 11:09:23] <vinzpr> ok
[10-Sep-2009 11:09:28] <rmatte> hopefully it works this time
[10-Sep-2009 11:09:36] <rmatte> if it doesn't, I'm out of ideas lol
[10-Sep-2009 11:10:18] <vinzpr> hehe
[10-Sep-2009 11:10:57] <rmatte> well not 100% out of ideas, there are some debugging steps we can use
[10-Sep-2009 11:11:04] <vinzpr> cool..its working now...Memory threshold exceeded: current value 99.1%
[10-Sep-2009 11:11:08] <rmatte> perfect
[10-Sep-2009 11:11:10] <vinzpr> how about the cpu load
[10-Sep-2009 11:11:18] <rmatte> have you tested the cpu load one?
[10-Sep-2009 11:11:24] <vinzpr> what does this means [^:]+
[10-Sep-2009 11:11:31] <vinzpr> yep its not working also
[10-Sep-2009 11:11:40] <rmatte> it's basically a regex wildcard
[10-Sep-2009 11:12:02] <rmatte> ok, paste the load average script again along with an example of a threshold
[10-Sep-2009 11:12:10] <rmatte> also..., make sure you put the transform under the right event class
[10-Sep-2009 11:13:16] <vinzpr> threshold of Load Warningexceeded: current value 710.00
[10-Sep-2009 11:13:31] <vinzpr> import re
[10-Sep-2009 11:13:31] <vinzpr> match = re.search('threshold of Load [^:]+: current value ([\\d\\.]+)', evt.message)
[10-Sep-2009 11:13:31] <vinzpr> if match and device:
[10-Sep-2009 11:13:31] <vinzpr> load = float(match.groups()[0])
[10-Sep-2009 11:13:31] <vinzpr> load_used = load / 100
[10-Sep-2009 11:13:32] <vinzpr> evt.summary = "Load Average threshold exceeded: current value %3.1f%%" (load_used)
[10-Sep-2009 11:14:00] <rmatte> vinzpr: ummm, you don't have any indentation?
[10-Sep-2009 11:14:20] <rmatte> it's not going to work without proper spacing
[10-Sep-2009 11:15:01] <rmatte> also change evt.message to evt.summary at the end of the re.search line
[10-Sep-2009 11:15:42] <vinzpr> ok let me change it
[10-Sep-2009 11:15:57] <rmatte> k
[10-Sep-2009 11:16:14] <rmatte> change the spacing to somewhat match the other transform
[10-Sep-2009 11:17:42] <rmatte> import re
[10-Sep-2009 11:17:43] <rmatte> match = re.search('threshold of Load [^:]+: current value ([\\d\\.]+)', evt.summary)
[10-Sep-2009 11:17:43] <rmatte> if match and device:
[10-Sep-2009 11:17:43] <rmatte> load = float(match.groups()[0])
[10-Sep-2009 11:17:43] <rmatte> load_used = load / 100
[10-Sep-2009 11:17:43] <rmatte> evt.summary = "Load Average threshold exceeded: current value %3.1f%%" (load_used)
[10-Sep-2009 11:17:48] <rmatte> that's how it should look
[10-Sep-2009 11:21:20] <vinzpr> ok already change it let me wait
[10-Sep-2009 11:21:25] <vinzpr> how about this
[10-Sep-2009 11:21:26] <vinzpr> Threshold of Memory Critical restored: current value: 98240.00
[10-Sep-2009 11:21:54] <rmatte> is it actually outputting with a capital T?
[10-Sep-2009 11:22:01] <rmatte> hold on, I have a fix for that
[10-Sep-2009 11:22:42] <vinzpr> ok
\n[10-Sep-2009 11:23:12] <rmatte> match = re.search('threshold of\nMemory Warning [^:]+: current value ([\\d\\.]+)',\nevt.summary,re.IGNORECASE)
[10-Sep-2009 11:23:15] <rmatte> try that
[10-Sep-2009 11:23:17] <rmatte> oops
[10-Sep-2009 11:23:18] <rmatte> wait
\n[10-Sep-2009 11:23:24] <rmatte> match = re.search('threshold of\nMemory [^:]+: current value ([\\d\\.]+)', evt.summary,re.IGNORECASE)
[10-Sep-2009 11:23:25] <rmatte> that
[10-Sep-2009 11:23:36] <vinzpr> cool
\n[10-Sep-2009 11:23:42] <rmatte> match = re.search('threshold of\nLoad [^:]+: current value ([\\d\\.]+)', evt.summary,re.IGNORECASE)
[10-Sep-2009 11:23:43] <rmatte> as well
[10-Sep-2009 11:23:46] <rmatte> for the other one
\n[10-Sep-2009 11:24:14] <rmatte> I'm actually going to log a bug\nreport for that, the case shouldn't be the same between threshold alert\nmessages and threshold recovered messages
[10-Sep-2009 11:24:26] <rmatte> or it SHOULD be the same rather
[10-Sep-2009 11:25:27] <vinzpr> rockin
[10-Sep-2009 11:25:43] <rmatte> let me know if it works
[10-Sep-2009 11:28:59] <vinzpr> ok
\n[10-Sep-2009 11:28:59] <q7joey> is there supposed to be a way to\nallow the binding of multiple templates to a specific interface? the\nbind template option at the interface level actually binds to the\ndevice, not the interface.
[10-Sep-2009 11:30:14] <rmatte> man, I love linux, so easy to make iso files: dd if=/dev/cdrom of=whatever.iso
[10-Sep-2009 11:30:35] <rmatte> q7joey: no, only 1 template per interface
\n[10-Sep-2009 11:30:47] <rmatte> q7joey: and the interface\ntemplates are bound automatically based on the name of the template
\n[10-Sep-2009 11:31:35] <q7joey> it probably would be good to\nremove the bind template option from the interface area. i know about\nthe automatic choice. for now i'll make a local copy and mangle it for\nmy testing. i'm still hunting down issues with gaps in graphs due to\ncounter vs derive issues.
[10-Sep-2009 11:31:39]\n<rmatte> q7joey: so if the interface type (you can see the type\nby clicking on the interface), is ds1, and an interface template called\n"ds1" exists under /Devices, then it'll automatically use that
[10-Sep-2009 11:31:47] <rmatte> I see
[10-Sep-2009 11:32:00] <q7joey> i actually wanted to capture two sets of data for one interface, so a temp copy is what i'll do for now.
[10-Sep-2009 11:32:00] <rmatte> I didn't even know there was a bind template option there
[10-Sep-2009 11:32:12] <rmatte> I would imagine you can probably bind multiple templates to an interface
[10-Sep-2009 11:32:17] <rmatte> no idea why you'd want to though
[10-Sep-2009 11:32:30] <rmatte> the automated method is the proper one
[10-Sep-2009 11:32:36] <q7joey> yep, that's why i was asking since the option was there, but doesn't do what i'd expect.
[10-Sep-2009 11:32:43] <rmatte> I see
[10-Sep-2009 11:33:18] <q7joey> i'm going to make two graphs to see how different counter vs derive is.
[10-Sep-2009 11:33:50] <vinzpr> rmatte: the change works but the critical values still post on the clear the same value
[10-Sep-2009 11:34:43] <rmatte> vinzpr: oh?
[10-Sep-2009 11:34:54] <vinzpr> yep
[10-Sep-2009 11:34:57] <rmatte> vinzpr: try putting the operation in to brackets
[10-Sep-2009 11:35:02] <rmatte> (load / 100)
[10-Sep-2009 11:35:16] <vinzpr> thats on memory
[10-Sep-2009 11:35:34] <rmatte> oh, on clear they don't post a percentage?
[10-Sep-2009 11:35:46] <rmatte> or what?
[10-Sep-2009 11:35:55] <rmatte> give me an example
[10-Sep-2009 11:35:59] <vinzpr> they post percentage
[10-Sep-2009 11:36:16] <vinzpr> on clear but the value percentage on critical are the same when clear
[10-Sep-2009 11:36:28] <vinzpr> it supposed to show you the current value
[10-Sep-2009 11:36:50] <rmatte> yeh, hmmm
[10-Sep-2009 11:37:06] <rmatte> paste an example
[10-Sep-2009 11:37:22] <rmatte> and the transform, please
[10-Sep-2009 11:37:49] <vinzpr> CLEAR: Memory thresholdexceeded: current value 99.1%
[10-Sep-2009 11:38:10] <vinzpr> import re
\n[10-Sep-2009 11:38:10] <vinzpr> match = re.search('threshold of\nMemory [^:]+: current value ([\\d\\.]+)', evt.summary,re.IGNORECASE)
[10-Sep-2009 11:38:10] <vinzpr> if match and device:
[10-Sep-2009 11:38:10] <vinzpr> avail = float(match.groups()[0]) * 1024
[10-Sep-2009 11:38:10] <vinzpr> total = device.hw.totalMemory
[10-Sep-2009 11:38:10] <vinzpr> percent_used = (1 - (avail / total)) * 100
[10-Sep-2009 11:38:11] <vinzpr> evt.summary = "Memory threshold exceeded: current value %3.1f%%" % (percent_used)
[10-Sep-2009 11:38:13] <vinzpr> also the load didnt work
[10-Sep-2009 11:38:21] <rmatte> does it actually look like that? "thresholdexceeded"
[10-Sep-2009 11:38:39] <vinzpr> threshold of Load Warning exceeded: current value 983.00
[10-Sep-2009 11:38:52] <rmatte> hmmm
[10-Sep-2009 11:40:01] <vinzpr> Threshold of Memory Critical restored: current value: 112652.00
\n[10-Sep-2009 11:40:31] <rmatte> vinzpr: let's say you completely\nremove the transforms, and check what memory alerts and clear events\nlook like
[10-Sep-2009 11:40:40] <rmatte> vinzpr: if you do that, are the values the same or different?
[10-Sep-2009 11:40:48] <vinzpr> its different
[10-Sep-2009 11:41:01] <vinzpr> it shows you the current value
\n[10-Sep-2009 11:41:25] <rmatte> well, looking at the transform\ncode there's really no possible reason why it would be showing the old\nvalue
[10-Sep-2009 11:42:18] <rmatte> each time an event comes in it should be going through that entire transform
[10-Sep-2009 11:42:28] <rmatte> can you paste examples of a plain alert and plain clear?
[10-Sep-2009 11:43:14] <vinzpr> Event: 'Memory threshold exceeded: current value 99.1%'
[10-Sep-2009 11:43:15] <vinzpr> Cleared by: ''
[10-Sep-2009 11:43:15] <vinzpr> At:
[10-Sep-2009 11:43:15] <vinzpr> Device:
[10-Sep-2009 11:43:15] <vinzpr> Component:
[10-Sep-2009 11:43:15] <vinzpr> Severity: Critical
[10-Sep-2009 11:43:17] <vinzpr> Message:
[10-Sep-2009 11:43:19] <vinzpr> threshold of Memory Critical not met: current value 34116.00
[10-Sep-2009 11:43:21] <vinzpr> Undelete
[10-Sep-2009 11:44:19] <vinzpr> Device:
[10-Sep-2009 11:44:19] <vinzpr> Component:
[10-Sep-2009 11:44:19] <vinzpr> Severity: Warning
[10-Sep-2009 11:44:19] <vinzpr> Time: 2009/09/10 16:37:07.000
[10-Sep-2009 11:44:19] <vinzpr> Message:
[10-Sep-2009 11:44:19] <vinzpr> threshold of Memory Warning not met: current value 47352.00
[10-Sep-2009 11:44:21] <vinzpr> Event Detail
[10-Sep-2009 11:44:23] <vinzpr> Acknowledge
[10-Sep-2009 11:44:25] <vinzpr> Delete
[10-Sep-2009 11:44:27] <vinzpr> Device Events
[10-Sep-2009 11:45:17] <rmatte> I just want to see the difference in summary messages
[10-Sep-2009 11:45:43] <vinzpr> ok..
[10-Sep-2009 11:46:23] <vinzpr> summary Memory threshold exceeded: current value 99.2%
[10-Sep-2009 11:46:23] <vinzpr> message threshold of Memory Critical not met: current value 33324.00
[10-Sep-2009 11:47:09] <rmatte> and for the clear?
[10-Sep-2009 11:49:17] <rmatte> Hey Matt
[10-Sep-2009 11:49:41] <rmatte> busy morning, I've been troubleshooting up a storm in here lol
[10-Sep-2009 11:49:46] <vinzpr> CLEAR: Memory thresholdexceeded: current value 99.2%
[10-Sep-2009 11:49:46] <vinzpr> Event: 'Memory threshold exceeded: current value 99.2%'
[10-Sep-2009 11:50:19] <rmatte> ok, I'm really curious as to why it's showing "thresholdexceeded"
[10-Sep-2009 11:50:45] <rmatte> vinzpr: what does the event message look like on the clear event?
[10-Sep-2009 11:51:13] <rmatte> vinzpr: I'm more interested in the raw message than the transformed message right now
[10-Sep-2009 11:52:20] <vinzpr> on clear message
[10-Sep-2009 11:52:22] <vinzpr> Event: 'Memory threshold exceeded: current value 99.2%'
[10-Sep-2009 11:52:22] <vinzpr> Cleared by: ''
[10-Sep-2009 11:52:22] <vinzpr> At:
[10-Sep-2009 11:52:22] <vinzpr> Device:
[10-Sep-2009 11:52:22] <vinzpr> Component:
[10-Sep-2009 11:52:23] <vinzpr> Severity: Critical
[10-Sep-2009 11:52:25] <vinzpr> Message:
[10-Sep-2009 11:52:27] <vinzpr> threshold of Memory Critical not met: current value 34116.00
[10-Sep-2009 11:52:37] <rmatte> what happened to the capital T>?
[10-Sep-2009 11:52:39] <rmatte> Threshold of Memory Critical restored: current value: 98240.00
[10-Sep-2009 11:52:43] <rmatte> that was the example you gave before
[10-Sep-2009 11:52:47] <rmatte> you don't see that at all now?
[10-Sep-2009 11:52:55] <vinzpr> why is there is a message like this Memory thresholdexceeded
[10-Sep-2009 11:53:06] <rmatte> vinzpr: no idea, we can deal with that in a bit
[10-Sep-2009 11:53:12] <rmatte> just work with me here
[10-Sep-2009 11:53:13] <rmatte> lol
[10-Sep-2009 11:53:17] <vinzpr> yep the same message
[10-Sep-2009 11:53:25] <vinzpr> as critical and clear
[10-Sep-2009 11:53:27] <rmatte> with a capital T or lowercase?
[10-Sep-2009 11:53:32] <rmatte> the case is very important
\n[10-Sep-2009 11:54:17] <rmatte> I just realized that using the\nsummary probably won't work properly, we need to use the event message
[10-Sep-2009 11:54:31] <vinzpr> on critical : Memory threshold exceeded:current value 98.8%
[10-Sep-2009 11:54:39] <rmatte> I need EXACT examples of alerts and clear events
[10-Sep-2009 11:54:51] <rmatte> messages, not summaries
[10-Sep-2009 11:55:09] <vinzpr> on clear : Event: 'Memory threshold exceeded: current value 98.8%'
[10-Sep-2009 11:55:09] <vinzpr> Cleared by: 'Threshold of Memory Critical restored: current value: 112652.00'
[10-Sep-2009 11:55:25] <rmatte> so that's the message then?
[10-Sep-2009 11:55:35] <vinzpr> did you see it the 98.8% still show on clear
[10-Sep-2009 11:55:39] <rmatte> like if you click on the details link: what does "message" show up as
[10-Sep-2009 11:55:42] <rmatte> in the details window
[10-Sep-2009 11:55:45] <rmatte> for each event
\n[10-Sep-2009 11:56:20] <rmatte> yeh, we'll deal with that in a\nsecond, I really just need to know that you are giving me 100% accurate\nevent messages here
[10-Sep-2009 11:56:36] <rmatte> I can't modify this script if I don't have accurate data to work with
[10-Sep-2009 11:56:44] <vinzpr> threshold of Memory Critical not met: current value 32784.00
[10-Sep-2009 11:56:51] <rmatte> ok, and for clear?
[10-Sep-2009 11:57:35] <vinzpr> Event: 'Memory threshold exceeded: current value 98.8%'
[10-Sep-2009 11:57:36] <vinzpr> Cleared by: 'Threshold of Memory Critical restored: current value: 112652.00'
[10-Sep-2009 11:59:49] <rmatte> ok
[10-Sep-2009 11:59:49] <rmatte> for starters, on the match = line
[10-Sep-2009 11:59:49] <rmatte> at the end of that line...
[10-Sep-2009 11:59:49] <rmatte> change evt.summary to evt.message
[10-Sep-2009 11:59:49] <rmatte> then test it
[10-Sep-2009 11:59:49] <vinzpr> ok
[10-Sep-2009 12:01:06] <rmatte> If that doesn't work then use this as a transform: http://pastebin.com/m1341a494
[10-Sep-2009 12:01:15] <rmatte> http://pastebin.com/m1341a494 is probably a better option
[10-Sep-2009 12:02:04] <rmatte> actually wait
[10-Sep-2009 12:02:30] <vinzpr> Alert : Memory threshold exceeded:current value 99.0%
[10-Sep-2009 12:02:31] <vinzpr> Clear : Threshold of MemoryCritical restored: current value: 495268.00
[10-Sep-2009 12:03:02] <rmatte> why is there no space between : and current now
[10-Sep-2009 12:03:04] <rmatte> there was before
[10-Sep-2009 12:03:35] <rmatte> paste your match line
[10-Sep-2009 12:03:45] <rmatte> and your evt.summary line
[10-Sep-2009 12:04:09] <vinzpr> import re
[10-Sep-2009 12:04:09] <vinzpr> match = re.search('threshold of Memory [^:]+: current value ([\\d\\.]+)', evt.message,re.IGNORECASE)
[10-Sep-2009 12:04:09] <vinzpr> if match and device:
[10-Sep-2009 12:04:09] <vinzpr> avail = float(match.groups()[0]) * 1024
[10-Sep-2009 12:04:09] <vinzpr> total = device.hw.totalMemory
[10-Sep-2009 12:04:09] <vinzpr> percent_used = (1 - (avail / total)) * 100
[10-Sep-2009 12:04:11] <vinzpr> evt.summary = "Memory threshold exceeded: current value %3.1f%%" % (percent_used)
[10-Sep-2009 12:04:41] <rmatte> ok, try the transform here: http://pastebin.com/m1341a494
[10-Sep-2009 12:04:58] <rmatte> I think the ignorecase is screwing it up for some reason
[10-Sep-2009 12:05:15] <vinzpr> ok
[10-Sep-2009 12:18:03] <vinzpr> is there a link how to create a transform
[10-Sep-2009 12:19:04] <rmatte> well, for pretty basic transforms
[10-Sep-2009 12:19:25] <rmatte> you basically need to know python
[10-Sep-2009 12:19:32] <rmatte> transforms are python scripts
[10-Sep-2009 12:19:51] <rmatte> http://www.zenoss.com/community/wiki/EventTransforms/?searchterm=device
[10-Sep-2009 12:20:25] <rmatte> they can get quite advanced
[10-Sep-2009 12:21:57] <vinzpr> On Alert : Memory threshold exceeded:current value 96.8%
[10-Sep-2009 12:22:25] <rmatte> I've never had issues with spaces magically dissapearing like that
[10-Sep-2009 12:22:29] <vinzpr> on Clear: Threshold of Memory Warningrestored: current value: 578820.00
[10-Sep-2009 12:23:03] <vinzpr> Event: 'Memory threshold exceeded: current value 96.8%'
[10-Sep-2009 12:23:04] <vinzpr> Cleared by: 'Threshold of Memory Warning restored: current value: 578820.00'
[10-Sep-2009 12:23:04] <vinzpr> At: 2009/09/10 17:17:07.000
[10-Sep-2009 12:23:04] <vinzpr> Device:
[10-Sep-2009 12:23:04] <vinzpr> Component:
[10-Sep-2009 12:23:04] <vinzpr> Severity: Warning
[10-Sep-2009 12:23:06] <vinzpr> Message:
[10-Sep-2009 12:23:08] <vinzpr> threshold of Memory Warning not met: current value 121556.00
[10-Sep-2009 12:23:11] <vinzpr> on Alert:
[10-Sep-2009 12:23:40] <vinzpr> Device:
[10-Sep-2009 12:23:40] <vinzpr> Component:
[10-Sep-2009 12:23:40] <vinzpr> Severity: Warning
[10-Sep-2009 12:23:40] <vinzpr> Time: 2009/09/10 17:15:27.000
[10-Sep-2009 12:23:40] <vinzpr> Message:
[10-Sep-2009 12:23:41] <vinzpr> threshold of Memory Warning not met: current value 121556.00
[10-Sep-2009 12:23:57] <rmatte> ok, you're pasting way too much stuff
[10-Sep-2009 12:23:59] <rmatte> that's not helping
[10-Sep-2009 12:24:00] <rmatte> lol
[10-Sep-2009 12:24:03] <vinzpr> hehe
[10-Sep-2009 12:24:06] <vinzpr> just showing
[10-Sep-2009 12:24:13] <vinzpr> the difference...lol
[10-Sep-2009 12:24:19] <rmatte> yeh, but I can barely make sense of what you're pasting
[10-Sep-2009 12:26:00] <rmatte> is there any particular reason that you care what the value is when it clears?
[10-Sep-2009 12:27:45] <vinzpr> actually your alert is on percentage so clear should show you the current percentage
[10-Sep-2009 12:28:03] <rmatte> ok, this is what I want you to paste, and ONLY this...
[10-Sep-2009 12:28:11] <rmatte> the summary and message for the alert
[10-Sep-2009 12:28:14] <vinzpr> shoot...
[10-Sep-2009 12:28:18] <rmatte> and then the summary and message for the clear
[10-Sep-2009 12:28:31] <rmatte> that's it, don't paste a whole wackload of stuff like you did before, it's useless
[10-Sep-2009 12:29:02] <rmatte> summary and message are 2 different things
[10-Sep-2009 12:29:06] <rmatte> so I want both of those for each
[10-Sep-2009 12:33:13] <vinzpr> CLEAR:
[10-Sep-2009 12:33:13] <vinzpr> summary Threshold of Memory Warning restored: current value: 578820.00
[10-Sep-2009 12:33:13] <vinzpr> message Threshold of Memory Warning restored: current value: 578820.00
[10-Sep-2009 12:33:13] <vinzpr> ALERT:
[10-Sep-2009 12:33:13] <vinzpr> summary Memory threshold exceeded: current value 96.8%
[10-Sep-2009 12:33:14] <vinzpr> message threshold of Memory Warning not met: current value 121556.00
[10-Sep-2009 12:34:59] <rmatte> and that's while using the pastebin version of the transform?
[10-Sep-2009 12:35:19] <rmatte> http://pastebin.com/m1341a494
[10-Sep-2009 12:35:37] <vinzpr> yes
[10-Sep-2009 12:36:43] <rmatte> that doesn't make sense, hmmm
[10-Sep-2009 12:37:26] <rmatte> ok, we'll probably have to add a couple of lines
[10-Sep-2009 12:37:33] <rmatte> I just need to finish some things I'm doing
[10-Sep-2009 12:37:39] <vinzpr> ok
[10-Sep-2009 12:40:29] <twm1010> I'm gone for a week and three new versions come out?
\n[10-Sep-2009 12:44:30] <rmatte> twm1010: well, 2.4.3 was buggy\n2.4.4 fixed 1 bug, and 2.4.5 fixed another, but it's still buggy
\n[10-Sep-2009 12:44:54] <rmatte> in 2.4.5 you can't change\nhardware of software product info on device edit tab for devices
[10-Sep-2009 12:45:01] <rmatte> so if you like being able to do that, don't upgrade
[10-Sep-2009 12:45:02] <rmatte>
[10-Sep-2009 12:46:49] <twm1010> I'm still on 2.4.2 here
[10-Sep-2009 12:46:57] <twm1010> I think I'll stick with it for the time being
[10-Sep-2009 12:49:00] <rmatte> yeh, just make sure you do "zenpatch 14181" as zenoss user
[10-Sep-2009 12:49:03] <rmatte> then restart Zenoss
[10-Sep-2009 12:49:15] <rmatte> fixes: http://dev.zenoss.org/trac/ticket/5279
[10-Sep-2009 12:52:41] <mrayzenoss> we're trying to push out the beta right now, having redirect issues with SourceForge
[10-Sep-2009 12:54:06] <rmatte> vinzpr: http://pastebin.com/m4d4dff2f
[10-Sep-2009 12:54:08] <rmatte> try that
[10-Sep-2009 12:54:32] <rmatte> mrayzenoss: cool, I'll start setting up a test box for it...
[10-Sep-2009 12:55:16] <vinzpr> ok
[10-Sep-2009 12:59:32] <vinzpr> thanks for the help rmatte..i'll inform you later regarding output of the new transform
[10-Sep-2009 12:59:38] <rmatte> k, np
[10-Sep-2009 12:59:54] <vinzpr> i have to run now..
[10-Sep-2009 13:00:14] <rmatte> k
[10-Sep-2009 13:00:23] <vinzpr> cheers
[10-Sep-2009 13:05:19] <IoNu|z> is there a way to have zenoss display the device name instead of the ip on discovery?
[10-Sep-2009 13:05:27] <IoNu|z> when i add a device by ip?
[10-Sep-2009 13:05:37] <IoNu|z> like to pull the device snmp hostname instead of the ip?
[10-Sep-2009 13:07:48] <rmatte> IoNu|z: Not that I'm aware of, you can rename it afterwards
[10-Sep-2009 13:08:04] <rmatte> or do what I do, I added the IP and hostname to the hosts file on the box
[10-Sep-2009 13:08:07] <rmatte> then add it by hostname
[10-Sep-2009 13:08:21] <mrayzenoss> 2.5 is adding a "title" as well
[10-Sep-2009 13:08:32] <rmatte> a "title" ?
[10-Sep-2009 13:08:54] <mrayzenoss> it's complicated, but basically IP address sets the hostname if it finds it in DNS
[10-Sep-2009 13:09:01] <mrayzenoss> SNMP sets hostname as well
[10-Sep-2009 13:09:14] <rmatte> ah, that's nice to see
[10-Sep-2009 13:09:18] <mrayzenoss> and title is if you want to override them
[10-Sep-2009 13:09:23] <rmatte> gotcha
[10-Sep-2009 13:09:40] <rmatte> "Super Server, DESTROYER OF WORLDS!" would be the title for instance.
[10-Sep-2009 13:09:42] <rmatte>
[10-Sep-2009 13:09:54] <mrayzenoss> so there's essentially an order of operations for how machines will show up by name
[10-Sep-2009 13:10:01] <mrayzenoss> title>hostname>ip
[10-Sep-2009 13:10:13] <rmatte> what about machines already in there?
[10-Sep-2009 13:10:24] <mrayzenoss> title won't be used unless set by the user
[10-Sep-2009 13:10:29] <mrayzenoss> so migration won't be an issue
[10-Sep-2009 13:10:32] <rmatte> ah ok
[10-Sep-2009 13:10:45] <rmatte> cool feature
[10-Sep-2009 13:10:56] <rmatte> what about when pulling info directly out of zope?
[10-Sep-2009 13:10:59] <rmatte> like if I do d.id
[10-Sep-2009 13:11:01] <rmatte> or whatever
[10-Sep-2009 13:11:14] <mrayzenoss> ahh, everywhere where I just said hostname, replace with 'id'
[10-Sep-2009 13:11:23] <mrayzenoss> title>id>ip
[10-Sep-2009 13:11:32] <rmatte> I see
[10-Sep-2009 13:11:45] <IoNu|z> hmm..
[10-Sep-2009 13:11:50] <IoNu|z> i had that in 3.2
[10-Sep-2009 13:11:50] <rmatte> I can't see myself using titles very often then
[10-Sep-2009 13:11:57] <IoNu|z> it was adding hostname instead of ip
[10-Sep-2009 13:12:03] <rmatte> is there still the option to set the ID manually as well?
[10-Sep-2009 13:12:07] <IoNu|z> but i dont remeber how i did it
[10-Sep-2009 13:12:07] <mrayzenoss> yes
[10-Sep-2009 13:12:12] <rmatte> ok, good
[10-Sep-2009 13:12:13] <IoNu|z> i think there was a patch or soemthing
[10-Sep-2009 13:13:07] <mrayzenoss> Rename Device sets the 'id'
[10-Sep-2009 13:13:20] <mrayzenoss> and in 2.5, title will be available on the 'Edit' tab
[10-Sep-2009 13:14:00] <mrayzenoss> and when you do an 'Add Device', title will be exposed there too
[10-Sep-2009 13:15:48] <rmatte> ah good
[10-Sep-2009 13:16:39] <IoNu|z> hmmm
[10-Sep-2009 13:16:48] <IoNu|z> sorry for the quetsion, but where do i edit that?
[10-Sep-2009 13:17:25] <mrayzenoss> page menu of the device->Manage
[10-Sep-2009 13:17:38] <IoNu|z> great thank you
[10-Sep-2009 13:19:10] <rmatte> afk
[10-Sep-2009 13:26:01] <twm1010> mrayzenoss: Is it the reverse DNS lookup that Zenoss uses?
[10-Sep-2009 13:26:34] <mrayzenoss> twm1010: yeah, if dns returns a hostname, it goes into 'id'
\n[10-Sep-2009 13:27:16] <twm1010> That would explain why when I\nsetup my box and had it use my new DNS servers that did not yet have\nreverse zones loaded, that nothing came back with a name
\n[10-Sep-2009 13:27:20] <mrayzenoss> I think that's right, I had a\ndeveloper walk me through the new 'title' thing for the beta
[10-Sep-2009 13:27:30] <mrayzenoss> yeah
[10-Sep-2009 13:27:38] <mrayzenoss> so in my home network, I've got no DNS
[10-Sep-2009 13:27:45] <mrayzenoss> but it's a small network
[10-Sep-2009 13:27:59] <mrayzenoss> so some boxes hostnames are set by SNMP
[10-Sep-2009 13:28:19] <mrayzenoss> some are set by /etc/hosts
[10-Sep-2009 13:28:19] <mrayzenoss> and some are blank (ip address)
\n[10-Sep-2009 13:32:43] <rmatte> yeh, I always just add the device\nto the hosts file, then add it by hostname, and then I have my scripts\nsetup on each Zenoss box which sync up the hosts file with Zenoss every\nhour
[10-Sep-2009 13:33:27] <rmatte> the nice\nthing about it is if someone doesn't have access to edit the hosts file\nbut wants to add a device, they can add it by IP, rename it, and then\nan entry gets automatically placed in to the hosts file within an hour
[10-Sep-2009 13:34:04] <rmatte> I just add them directly because when I'm adding a bunch of devices it keeps me from having to rename each one
[10-Sep-2009 13:36:09] <twm1010> as opposed to just keeping your DNS square?
[10-Sep-2009 13:37:02]\n<rmatte> We're in a different situation, we monitor devices for\nclients, so we have a ton of Zenoss boxes, and we don't want to have to\nkeep DNS entries for every client device
[10-Sep-2009 13:37:25] <twm1010> ah, gotcha
\n[10-Sep-2009 13:37:34] <rmatte> The Zenoss boxes are also used to\nconnect to client devices from, so we just need accurate hosts files
[10-Sep-2009 13:38:50] <rmatte> We may start using DNS for it down the road, but for now we're not
\n[10-Sep-2009 13:38:50] <rmatte> the only downside to this method\nis that any windows TS servers that we have don't have the entries, so\nyou have to connect via IP
[10-Sep-2009 13:39:02] <rmatte> but it's not a big deal
[10-Sep-2009 13:42:03] <rmatte> mrayzenoss: think an upgrade from 2.4.5 to the beta would work?
[10-Sep-2009 13:42:12] <mrayzenoss> not the beta 1 that's coming out
[10-Sep-2009 13:42:16] <rmatte> k
[10-Sep-2009 13:42:23] <mrayzenoss> it's known to be broken
[10-Sep-2009 13:42:28] <rmatte> any word on when stack installer will be available?
[10-Sep-2009 13:42:29] <mrayzenoss> and there's a P1
[10-Sep-2009 13:42:40] <mrayzenoss> it's on the calendar for Tuesday
[10-Sep-2009 13:42:45] <rmatte> k cool
\n[10-Sep-2009 13:43:35] <rmatte> well, as soon as the stack\ninstaller comes out I'll test it out (even if the upgrade is broken)
[10-Sep-2009 13:43:36] <twm1010> beta for.... 2.5?
[10-Sep-2009 13:43:47] <rmatte> and when the upgrade issue gets fixed I'll test the upgrade out as well
[10-Sep-2009 13:43:58] <mrayzenoss> yeah, 2.5 beta is minutes away
[10-Sep-2009 13:44:00] <rmatte> twm1010: correct, for King Crab (2.5)
\n[10-Sep-2009 13:44:18] <twm1010> Sweet.... i'll hit the tracker\nin a minute and see what's in, will be happy to beta test that
\n[10-Sep-2009 13:44:52] <rmatte> well, my beta box is setup, going\nto shut it down until I can get my hands on the stack installer
\n[10-Sep-2009 13:45:37] <rmatte> now I'm just waiting on a Windows\n2000 key so that I can test out net-snmp on Windows 2000 to see if it\nfixes the memory monitoring issues we're seeing
[10-Sep-2009 13:45:45] <mrayzenoss> http://hannahmontana.sourceforge.net
[10-Sep-2009 13:46:04] <rmatte> you can't be serious?
[10-Sep-2009 13:46:08] <rmatte> wtf
[10-Sep-2009 13:46:15] <rmatte> lol
[10-Sep-2009 13:46:22] <twm1010> I should put that on my girlfriend's daughters iMac
[10-Sep-2009 13:46:44] <rmatte> that's just scary
[10-Sep-2009 13:46:44] <twm1010> rmatte: mind if it's a not quite so legit key?
[10-Sep-2009 13:48:12] <rmatte> 2 of the items in the FAQ contradict themselves...
[10-Sep-2009 13:48:18] <rmatte> Q : How/why did you make such a great OS?
\n[10-Sep-2009 13:48:19] <rmatte> A : I thought - what would\nattract young users to Linux? So I created this idea after a lot of\nreading and work.
[10-Sep-2009 13:48:25] <rmatte> and then...
[10-Sep-2009 13:48:26] <rmatte> Q : How can I watch DVDs and other Media?
[10-Sep-2009 13:48:26] <rmatte> A : click the Hannah Montana menu, then Utilities, then Konsole, then type:
[10-Sep-2009 13:48:26] <rmatte> sudo apt-get install libdvdcss2 kubuntu-restricted-extras
[10-Sep-2009 13:48:39] <rmatte> yes, because I'm sure that all kids want to have to install apps from the commandline
[10-Sep-2009 13:48:50] <rmatte> twm1010: yeh, we need a legit key lol
\n[10-Sep-2009 13:49:15] <rmatte> twm1010: we'll be able to get\none, we're a Microsoft gold partner, I'm just waiting on the guy who\nhas access to get the key to send it to me
[10-Sep-2009 13:49:21] <jb> argh
[10-Sep-2009 13:49:32] <jb> why do vendors have to store related snmp information in like 20 different tables..
[10-Sep-2009 13:49:50] <rmatte> jb: because that's the way it's generally done hehe
[10-Sep-2009 13:49:58] <jb> no, not really
[10-Sep-2009 13:50:12] <rmatte> yeh, I know, lol
[10-Sep-2009 13:50:14] <jb> not this bad
[10-Sep-2009 13:50:14] <twm1010> ok i give, what's the URL for the zenoss milestones
[10-Sep-2009 13:50:17] <rmatte> just kidding around
[10-Sep-2009 13:50:25] <jb> it makes coding modelers difficult
[10-Sep-2009 13:50:26] <jb>
[10-Sep-2009 13:51:38] <rmatte> twm1010: Blue Crab, King Crab, Stone Crab <--- There you go lol
[10-Sep-2009 13:51:49] <rmatte> twm1010: I don't recall ever seeing a page listing anything more formal
[10-Sep-2009 13:51:55] <rmatte> though Matt would know best
\n[10-Sep-2009 13:52:51] <rmatte> jb: yeh, when I was writing my\nscript to monitor Windows Memory usage via plain SNMP the info I had to\naccess was all over the place
[10-Sep-2009 13:52:58] <rmatte> maybe not quite as bad as you're seeing
[10-Sep-2009 13:55:11] <mrayzenoss> http://blog.zenoss.com/2009/09/10/zenoss-2-5-beta-is-now-available/
[10-Sep-2009 13:55:21] <skipzoid> hey all again
[10-Sep-2009 13:57:31] <twm1010> Sweet... I'll be setting that up tomorrow
[10-Sep-2009 13:57:46] <rmatte> I'll be setting it up... Tuesday, looks like
[10-Sep-2009 13:57:47] <rmatte>
[10-Sep-2009 13:58:01] <skipzoid> i'm trying to add a data source, if I type this snmpwalk -cpublic -v2c 192.168.0.9 mib-2.16.1.1.1.7.1
[10-Sep-2009 13:58:01]\n<skipzoid> SNMPv2-SMI::mib-2.16.1.1.1.7.1 = Counter32: 17 on the\nconsole you can see it queries the device and returns the value, but is\nI put the same mib-2.16.1.1.1.7.1 in the OID box for the data source it\nsays invalid OID when I try and save it - where am i being stupid ?\nwhat am I missing
[10-Sep-2009 13:58:29] <twm1010> skipzoid: That's an abbreviated OID
[10-Sep-2009 13:58:32] <twm1010> you need the full OID
[10-Sep-2009 13:58:45] <twm1010> everything that mib- is hiding
[10-Sep-2009 13:59:02] <skipzoid> ah - how can I find that from the device ?
[10-Sep-2009 13:59:16] <twm1010> rmatte: isn't there a command flag to show full OIDs?
[10-Sep-2009 13:59:24] <rmatte> skipzoid: add -0n as an option
[10-Sep-2009 13:59:31] <twm1010> ding ding... he wins.
[10-Sep-2009 13:59:32] <rmatte> then you'll see what the OIDs really look like
[10-Sep-2009 13:59:50] <rmatte> snmpwalk tries to translate them from standard Mibs by default
[10-Sep-2009 13:59:58] <rmatte> and ends up partially translating them
[10-Sep-2009 14:00:57]\n<skipzoid> if I add -n0 to the command line it just shows me the\nsyntax of all the commands -n says CONTEXT next to it
[10-Sep-2009 14:01:03] <rmatte> no
[10-Sep-2009 14:01:05] <rmatte> -0n
[10-Sep-2009 14:01:07] <rmatte> not n0
[10-Sep-2009 14:01:09] <rmatte> lol
[10-Sep-2009 14:01:53] <skipzoid> ah - better still is a an 'o' not a zero
[10-Sep-2009 14:02:13] <rmatte> oh right
[10-Sep-2009 14:02:14] <rmatte> my bad
[10-Sep-2009 14:02:15] <rmatte>
[10-Sep-2009 14:02:25] <skipzoid> its saved it
[10-Sep-2009 14:02:37] <skipzoid> i'm trying to create a graph
[10-Sep-2009 14:02:55] <twm1010> well... you're almost there
[10-Sep-2009 14:03:04] <twm1010> getting the datapoint collecting is step 1
[10-Sep-2009 14:03:13] <twm1010> once thats square, the rest is easy
[10-Sep-2009 14:09:12]\n<skipzoid> wadaya know I have a graph - no data on it yet just\nsays 'nan' for the moment - am i to wait for it to gather some info ?
[10-Sep-2009 14:09:58] <mrayzenoss> yeah
[10-Sep-2009 14:12:26] <rmatte> skipzoid: takes 3 polling cycles (about 15 mins)
[10-Sep-2009 14:12:30] <rmatte> then you should see data
[10-Sep-2009 14:12:42] <skipzoid> ok, i feel pleased with my self now
[10-Sep-2009 14:13:32] <skipzoid> need to work out how to get emails and text messages sent from events
\n[10-Sep-2009 14:15:31] <rmatte> Settings, configure the SMTP\nsettings, then... Settings -> Users -> <Username> -> add\nEmail and pager info -> Alerting Rules tab
[10-Sep-2009 14:15:53] <rmatte> you can group users as well and apply alerting rules to groups
\n[10-Sep-2009 14:16:13] <skipzoid> question about the graphs, if I\nset it to be a guage and not a counter - is that info real time as a\nguage implys?
[10-Sep-2009 14:16:48] <twm1010>\nIs Zenoss intelligent enough not to send alerts for WinService events\nif the device is down? Or do I have to put that in my alerting criteria
\n[10-Sep-2009 14:17:13] <rmatte> twm1010: If a device is not\npingable at all Zenoss stops monitoring everything until it's pingable\nagain
[10-Sep-2009 14:17:16] <rmatte> so in short, yes
[10-Sep-2009 14:17:23] <rmatte> sometimes 1 or 2 service alerts slip through
[10-Sep-2009 14:17:41] <twm1010> Yeah, seen that before... I'm adding a whole bunch of service monitoring today.
\n[10-Sep-2009 14:17:43] <rmatte> only if it notices that the\nservices are down before it notices that the device is not pingable
[10-Sep-2009 14:17:45] <rmatte> which is rare
[10-Sep-2009 14:17:58] <rmatte> but it does happen
\n[10-Sep-2009 14:18:28] <twm1010> Don't you love when something\ndevelopement is suddenly "production" because someone else doesn't want\nto pony up 4-5k to add licenses to some "other" piece of software
[10-Sep-2009 14:21:10] <rmatte> yup
[10-Sep-2009 14:21:35] <rmatte> you're obviously not a government worker
[10-Sep-2009 14:23:20] <mrayzenoss> bbiab
[10-Sep-2009 14:23:28] <twm1010> well, at least my work so far is becoming appreciated
[10-Sep-2009 14:24:09] <rmatte> mine is about half and half for some reason
[10-Sep-2009 14:24:20] <rmatte> mostly because people expect stuff to just magically work with no dev work required
[10-Sep-2009 14:24:36] <rmatte> meaning stuff always should have been setup to work a certain way yesterday
[10-Sep-2009 14:24:44] <rmatte> regardless of whether I found out about it today or not
[10-Sep-2009 14:24:48] <rmatte>
[10-Sep-2009 14:29:23] <rmatte> hmmm, I'm trying to understand this line...
[10-Sep-2009 14:29:24] <rmatte> (((1 248 400 328 - 1 247 795 528) / 60) / 60) / 24 = 7
[10-Sep-2009 14:29:32] <rmatte> it's apparently supposed to work out to 7 days
[10-Sep-2009 14:29:41] <rmatte> just don't know what the heck all the values represent
[10-Sep-2009 14:29:58] <rmatte> oh wait
[10-Sep-2009 14:30:06] <Apocalipse> hi people
[10-Sep-2009 14:30:22] <Apocalipse> link between locations are not wroking...
[10-Sep-2009 14:30:32] <Apocalipse> how can i add then?
[10-Sep-2009 14:30:34] <rmatte> you mean on the google maps?
[10-Sep-2009 14:30:38] <Apocalipse> yes
[10-Sep-2009 14:30:57] <Apocalipse> matt send me a tutorial... but i think i made something wrong...
[10-Sep-2009 14:30:59] <rmatte> well, first step is to click on Networks on the left
[10-Sep-2009 14:31:15] <rmatte> then go to zProperties
[10-Sep-2009 14:31:15] <Apocalipse> ok
\n[10-Sep-2009 14:31:17] <dollarbang> konsa79: I saw your remark on\nthe Geomapping issue. can you provide me with how you generated that\nlist?
[10-Sep-2009 14:31:24] <rmatte> and change zDrawMapLinks to True
[10-Sep-2009 14:31:36] <rmatte> then it really depends on the types of routes that you've modeled via SNMP
[10-Sep-2009 14:31:42] <Apocalipse> already True
[10-Sep-2009 14:32:01] <Apocalipse> thats the problem...i thin
[10-Sep-2009 14:32:03] <rmatte> Ok, well then here's what else could be wrong...
[10-Sep-2009 14:32:05] <Apocalipse> i think
[10-Sep-2009 14:32:25] <rmatte> Let's say you have 7 sites, and they all connect in to a central router
[10-Sep-2009 14:32:33] <Apocalipse> ok
[10-Sep-2009 14:32:43] <Apocalipse> thats my problem...
\n[10-Sep-2009 14:32:45] <rmatte> if you have devices at those 7\nsites in Zenoss, but don't have that central router in with all router\ninfo for everything, you won't get any lines at all
[10-Sep-2009 14:33:03]\n<rmatte> I have 1 Zenoss box where I can't get the lines to draw\nno matter what just because of the types of circuits and routing they\nare using
[10-Sep-2009 14:33:16] <Apocalipse> here is a full mesh mpls network
[10-Sep-2009 14:33:31] <rmatte> yeh, mpls doesn't tend to map too well
[10-Sep-2009 14:33:39] <Apocalipse>
[10-Sep-2009 14:34:24] <Apocalipse> but is it possible to add a static link?
[10-Sep-2009 14:34:24] <rmatte> are you sure all of the devices are fully modeled?
[10-Sep-2009 14:34:29] <Apocalipse> manual added
[10-Sep-2009 14:34:29] <rmatte> and that Zenoss picked up the routes?
[10-Sep-2009 14:34:38] <rmatte> no, the links are only automatically generated
[10-Sep-2009 14:34:44] <rmatte> they are meant to represent actual links
[10-Sep-2009 14:34:45] <Apocalipse> i think they are not fully modeled
[10-Sep-2009 14:34:52] <rmatte> then make sure they are
[10-Sep-2009 14:34:58] <Apocalipse> ok
[10-Sep-2009 14:35:02] <Apocalipse> let me try
[10-Sep-2009 14:35:11] <twm1010> my MPLS graphs fine
[10-Sep-2009 14:35:53] <twm1010> its an odd picture, but its correct
[10-Sep-2009 14:36:22] <Apocalipse> i have a lot of networks listed here...
[10-Sep-2009 14:36:45] <Apocalipse> where does zenoss pic up that?
[10-Sep-2009 14:37:33] <Apocalipse> ????????
[10-Sep-2009 14:38:02] <twm1010> from the routing tables collected via SNMP from the devices you've added
[10-Sep-2009 14:38:03] <rmatte> he's probably just missing the core router
[10-Sep-2009 14:38:17] <Apocalipse> core router is here
\n[10-Sep-2009 14:38:18] <rmatte> I just checked and actually, on\nthis one that won't draw lines it's because we don't manage their core\nrouter
[10-Sep-2009 14:38:24] <rmatte> so Zenoss has no idea what they link in to
[10-Sep-2009 14:39:28] <Apocalipse> hunnn
[10-Sep-2009 14:39:38] <Apocalipse> core router is here...
[10-Sep-2009 14:39:45] <Apocalipse> but not well modeled...
[10-Sep-2009 14:39:53] <Apocalipse> :"
[10-Sep-2009 14:39:53] <rmatte> then that's your problem
[10-Sep-2009 14:39:55] <Apocalipse>
[10-Sep-2009 14:40:00] <rmatte> it needs as much route info as possible
\n[10-Sep-2009 14:40:37] <rmatte> If it's timing out when modelling\nthen go to Devices -> zProperties and change zCollectorClientTimeout\nto something like 5000
[10-Sep-2009 14:40:49] <rmatte> and when you're done modelling change it back
[10-Sep-2009 14:42:11] <Apocalipse> ok
[10-Sep-2009 14:42:17] <Apocalipse> let me do it...
[10-Sep-2009 14:42:48] <twm1010> you can do it!
[10-Sep-2009 14:44:06] <rmatte> actually this one that's not drawing lines seems to be due to an abundance of indirect bgp routes
[10-Sep-2009 14:44:24] <rmatte> we actually do monitor the core router, but it only has 4 direct routes
[10-Sep-2009 15:00:45] <chemist> good evening all
[10-Sep-2009 15:01:31]\n<chemist> as was suggested on the forums, I have installed\nsnmp-informant on some of my win boxes but I still can not get\nperformance graphs
[10-Sep-2009 15:02:24] <chemist> I keeep getting 'Missing RRD file' on the cpu graph, and nothing else
[10-Sep-2009 15:03:24] <metalklesk> mrayzenoss, can you say us what are the news about the new event console in zenoss 2.5 ?
[10-Sep-2009 15:05:22] <chemist> scrap that, it working now
[10-Sep-2009 15:12:16] <Apocalipse> hello
[10-Sep-2009 15:16:32] <Apocalipse> i modeled my core router...
[10-Sep-2009 15:16:47] <Apocalipse> and other routers
[10-Sep-2009 15:19:09] <metalklesk> now you got your routers modeled, can you export the zenpack so anyone can use it ?
[10-Sep-2009 15:19:24] <metalklesk> zenoss can do that right ?
[10-Sep-2009 15:19:55] <Apocalipse> matt can answer u... much better than me
[10-Sep-2009 15:19:56] <Apocalipse>
[10-Sep-2009 15:20:56] <mrayzenoss> what's the question?
\n[10-Sep-2009 15:22:30] <metalklesk> if Apocalipse modeled some\nrouters, can he export it as a zenpack so anyone can use it ?
[10-Sep-2009 15:23:18] <djack> hey you all
[10-Sep-2009 15:23:26] <djack> somebody awake here
[10-Sep-2009 15:23:27] <metalklesk> i'm still a newbie, i'm in a learning phase
[10-Sep-2009 15:23:37] <metalklesk> hi djack
[10-Sep-2009 15:23:46] <djack> have a strange issue
\n[10-Sep-2009 15:25:11] <djack> have java process in monitoring\n.... we changed the Xmm value and the new value is visible as a os\nprocess for the device .... only the old process generates errors ...
\n[10-Sep-2009 15:25:29] <mrayzenoss> Apocalipse: yes, once you've\ngot something you want to export, you can create a new ZenPack and\nstart marking things for inclusion in it
[10-Sep-2009 15:25:52] <djack> he mrayzenoss
[10-Sep-2009 15:26:22] <mrayzenoss> Apocalipse: http://www.zenoss.com/community/docs/zenoss-guide/2.4.3/ch13s03.html
[10-Sep-2009 15:26:27] <mrayzenoss> djack: greetings
[10-Sep-2009 15:26:55] <djack> i have strange error because os value changed
\n[10-Sep-2009 15:27:13] <Apocalipse> matt... to stop snmp\nmonitoring for a complete class... the procedure is: in Device/Server\n-> zProperties
[10-Sep-2009 15:27:36] <Apocalipse> zSnmpMonitorIgnore -> true
[10-Sep-2009 15:27:39] <Apocalipse> right?
[10-Sep-2009 15:28:54] <mrayzenoss> sounds right
\n[10-Sep-2009 15:28:54] <metalklesk> you got it Apocalipse, you\nhave modeled your routers now you can make your own zenpacks and share\nit with the community, wich i'm part of it
[10-Sep-2009 15:28:54] <djack> lol
[10-Sep-2009 15:29:32] <Apocalipse> metalklesk... its not possible... because i have some confidential info...
\n[10-Sep-2009 15:29:34] <metalklesk> when i learn more i will\nstart modeling stuffs so i can share zenpacks with the community
[10-Sep-2009 15:30:05] <Apocalipse> i think u can add Ryan Zenpack...for cisco devices...
[10-Sep-2009 15:30:15] <metalklesk> mmm then you can build a nice how to
[10-Sep-2009 15:30:26] <metalklesk> replacing the credentials
[10-Sep-2009 15:30:27] <Apocalipse> i did... and now i got a lot of MIBs to explore
[10-Sep-2009 15:30:32] <metalklesk> by xxxxxxx
[10-Sep-2009 15:30:34] <metalklesk>
[10-Sep-2009 15:31:02] <Apocalipse> matt... devices are still monitoring by snmp
[10-Sep-2009 15:31:44] <metalklesk> i was looking into the SuSE zenpack source code
[10-Sep-2009 15:32:16] <metalklesk> and found the files i should edit to adapt it to mandriva or whatever
[10-Sep-2009 15:32:34] <metalklesk> but as i don't know python (i know java)
\n[10-Sep-2009 15:32:44] <rmatte> metalklesk: modeling a router\nmeans that he collected data for it via snmp, it doesn't mean he made a\ndevice template or anything
[10-Sep-2009 15:32:51] <mrayzenoss> Apocalipse: try pushing the change
[10-Sep-2009 15:33:05] <metalklesk> what should i do to create de egg file ?
[10-Sep-2009 15:33:19] <metalklesk> ok rmatte
[10-Sep-2009 15:34:01] <mrayzenoss> from Settings->ZenPacks->your zenpack->Export
[10-Sep-2009 15:35:08] <chemist> I have a strange problem with monitoring a process
[10-Sep-2009 15:35:36] <metalklesk> mrayzenoss, i did not mean that
[10-Sep-2009 15:35:54] <metalklesk> i mean, i have downloaded the source code from the web
[10-Sep-2009 15:36:07] <jb> yo rocket.. you here?
[10-Sep-2009 15:36:08] <metalklesk> and edited with a text editor
[10-Sep-2009 15:36:27] <Apocalipse> push ok... thx matt
[10-Sep-2009 15:36:28] <metalklesk> now, how can i made a zenpack with that
\n[10-Sep-2009 15:36:29] <chemist> the process is called Meeting\nand at any one time there can be several Meeting processes running
[10-Sep-2009 15:36:35] <mrayzenoss> metalklesk: trac-zenpacks/wiki/BuildingZenPacks
[10-Sep-2009 15:37:13] <djack> hmmm
[10-Sep-2009 15:37:21] <metalklesk> wooooooowww thanks !!
\n[10-Sep-2009 15:37:42] <chemist> they are all numbered from 0 to\n18 and normally I remodel the device so it picks up the Meeting process\nthat is taking place
[10-Sep-2009 15:37:47] <djack> how do i suppress old processes ?
[10-Sep-2009 15:37:47] <chemist> this works fine
\n[10-Sep-2009 15:38:40] <chemist> but now it keeps reporting that\n2 Meeting processes are not running, despite having remodelled the\ndevice and those two processes not showing in the OS tab for thedevice
[10-Sep-2009 15:39:17] <IoNu|z> how is the HW Model getting pooled in zenoss 2.4.5
[10-Sep-2009 15:39:18] <IoNu|z> ?
[10-Sep-2009 15:39:50] <djack> rmatte awake ?
[10-Sep-2009 15:40:15] <rmatte> yeh, I'm back
[10-Sep-2009 15:40:17] <chemist> how can I get rid of the 2 phantom processes?
[10-Sep-2009 15:40:27] <rmatte> what do you mean by "old processes"?
[10-Sep-2009 15:40:40] <rmatte> ok, they aren't in the list but they're still being monitored?
[10-Sep-2009 15:40:42] <djack> well we have java process running and the Xmm value changed
[10-Sep-2009 15:40:57] <rmatte> did you push changes to the collector for that device?
[10-Sep-2009 15:41:02] <chemist> yes
[10-Sep-2009 15:41:11] <rmatte> ok, go in to zendmd and try:
[10-Sep-2009 15:41:15] <rmatte> reindex()
[10-Sep-2009 15:41:17] <rmatte> commit()
[10-Sep-2009 15:41:26] <rmatte> I find that helps to sort out ghost process issues
\n[10-Sep-2009 15:41:29] <djack> i have a process monitor java\ndefined and he sees the new and keeps bugging me about the old process\n....
[10-Sep-2009 15:41:30] <rmatte> it has worked for me in the past
[10-Sep-2009 15:41:56] <djack> same process only one parameter is changed
[10-Sep-2009 15:43:02] <rmatte> but you did remove the old process from the list right?
[10-Sep-2009 15:43:24] <rmatte> you need to make sure the old process is not in the list of processes for that device
[10-Sep-2009 15:43:27] <rmatte> if it is delete it
[10-Sep-2009 15:43:37] <rmatte> then push changes to the collector (Manage -> Push Changes)
[10-Sep-2009 15:43:43] <rmatte> then as the zenoss user:
[10-Sep-2009 15:43:43] <djack> yeah
[10-Sep-2009 15:43:44] <rmatte> zendmd
[10-Sep-2009 15:43:48] <rmatte> reindex()
[10-Sep-2009 15:43:51] <rmatte> commit()
[10-Sep-2009 15:43:56] <rmatte> the press ctrl-d to exit
[10-Sep-2009 15:44:02] <djack> and then recreate?
[10-Sep-2009 15:44:14] <rmatte> why would you have to recreate?
[10-Sep-2009 15:44:20] <rmatte> if you remodel it should find the new process
[10-Sep-2009 15:44:29] <rmatte> have you remodeled?
[10-Sep-2009 15:45:08] <djack> yeah
\n[10-Sep-2009 15:45:51] <rmatte> k, so the new process is in the\nlist, the old process isn't, and you've pushed changes to the\ncollector, but you're still getting alerts for the process
[10-Sep-2009 15:46:06] <rmatte> if that's the case, execute those commands in zendmd
[10-Sep-2009 15:46:10] <chemist> that's correct
[10-Sep-2009 15:46:44] <rmatte> the index for processes seems to get messed up more often than anything
[10-Sep-2009 15:52:20] <jb> ARGH
[10-Sep-2009 15:52:31] <jb> i always forget you need to reinstall zenpacks when you add new object classes.
[10-Sep-2009 15:52:48] <djack> grmbl
[10-Sep-2009 15:52:52] <metalklesk> i hate to be a newbie
[10-Sep-2009 15:52:57] <jb> there goes that hour.
[10-Sep-2009 15:52:59] <djack> they still pop up
[10-Sep-2009 15:54:08] <chemist> rmatte: ok, i tried that http://pastebin.ca/1561485
\n[10-Sep-2009 15:54:23] <Apocalipse> now... after i modeled core\nrouter... and other routers... what do i need to do to link locations?
\n[10-Sep-2009 15:55:11] <metalklesk> make\nZenPacks.community.Mandriva, There is no rule to build the target\n'ZenPacks.community.Mandriva'. Stop
[10-Sep-2009 15:55:19] <metalklesk>
[10-Sep-2009 15:55:49] <mrayzenoss> metalklesk: I assume the directory is ZenPacks.community.Mandriva
[10-Sep-2009 15:55:54] <mrayzenoss> and the setup.py contains that
[10-Sep-2009 15:55:59] <metalklesk> yes
[10-Sep-2009 15:56:19] <mrayzenoss> and you're running make from the directory above that?
[10-Sep-2009 15:56:25] <metalklesk> yes
[10-Sep-2009 15:56:44] <djack> going to sleep
[10-Sep-2009 15:56:54] <metalklesk> in root directory, where the setup.py is
[10-Sep-2009 15:57:07] <mrayzenoss> no, need to go up 1 more directory
[10-Sep-2009 15:57:22] <rmatte> chemist: hmmm
[10-Sep-2009 15:57:39] <metalklesk> in tags ?
\n[10-Sep-2009 15:57:42] <mrayzenoss> metalklesk: see how\ntrac-zenpacks/browser/zenpacks is setup?
[10-Sep-2009 15:57:48] <rmatte> mrayzenoss: http://pastebin.ca/1561485 <--- Are those errors trivial?
\n[10-Sep-2009 15:58:57] <mrayzenoss> if your layout is like that,\nyou can make all of them with 'make' or just 1, with 'make\nZenPacks.community.Mandriva'
[10-Sep-2009 15:59:15] <mrayzenoss> rmatte: weird
[10-Sep-2009 15:59:34] <rmatte> mrayzenoss: those aren't errors that I have, chemist has those
[10-Sep-2009 15:59:53] <rmatte> I've never seen anything like that before
[10-Sep-2009 15:59:54] <chemist> I'm also getting an error in the zenoss console now when I try to go to that device
[10-Sep-2009 16:00:23] <metalklesk> aaahh but wait, i dont have the files build-egg.sh and GNUmakefile
[10-Sep-2009 16:00:24]\n<mrayzenoss> rmatte:\nhttp://search.zenoss.com/search?q=%22ERROR+ZODB.Connection+Couldn%27t+load+state%22&btnG=Search&entqr=0&ud=1&sort=date%3AD%3AL%3Ad1&output=xml_no_dtd&oe=UTF-8&ie=UTF-8&client=default_frontend&proxystylesheet=default_frontend&site=default_collection
[10-Sep-2009 16:00:33]\n<mrayzenoss> err, chemist:\nhttp://search.zenoss.com/search?q=%22ERROR+ZODB.Connection+Couldn%27t+load+state%22&btnG=Search&entqr=0&ud=1&sort=date%3AD%3AL%3Ad1&output=xml_no_dtd&oe=UTF-8&ie=UTF-8&client=default_frontend&proxystylesheet=default_frontend&site=default_collection
[10-Sep-2009 16:00:38] <rmatte> lol
[10-Sep-2009 16:01:28] <rmatte> mrayzenoss: doesn't look like either of those links help
[10-Sep-2009 16:01:42] <rmatte> oh wait, the first one might
[10-Sep-2009 16:02:39] <mrayzenoss> metalklesk: trac-zenpacks/wiki/SVNInstructions
[10-Sep-2009 16:02:53] <rmatte> I never understand how people's Zope databases get so screwed up
[10-Sep-2009 16:03:09] <mrayzenoss> metalklesk: trac-zenpacks/wiki/BuildingZenPacks
[10-Sep-2009 16:03:50] <chemist> http://pastebin.ca/1561496 <<< that is the error on the console
[10-Sep-2009 16:03:51]\n<rmatte> mrayzenoss: actually, the solution in the first link\ndoesn't help and the second link has no solution posted
[10-Sep-2009 16:04:34] <rmatte> chemist: are you only getting that error with that one device?
[10-Sep-2009 16:04:46] <chemist> I cant browse to any device now
[10-Sep-2009 16:05:19] <rmatte> eugh, yeh, something was screwed with your Zope database and the reindex appears to have agitated it
[10-Sep-2009 16:05:41] <chemist> actually, it appears to be only the linux devices
[10-Sep-2009 16:05:49] <chemist> windows are ok
[10-Sep-2009 16:06:05] <rmatte> Yeh, looks like the problem is in /zport/dmd/Devices/Server/Linux/devices
[10-Sep-2009 16:06:07]\n<mrayzenoss> chemist: try this\nhttp://www.zenoss.com/Members/cluther/fix-a-broken-devicesearch-or-componentsearch-catalog/
[10-Sep-2009 16:06:08] <rmatte> from the looks of it
[10-Sep-2009 16:06:41] <mrayzenoss> found that in http://dev.zenoss.org/trac/ticket/5233
[10-Sep-2009 16:06:59] <mrayzenoss> I usually dig into search.zenoss.com and dev.zenoss.org for finding stuff
[10-Sep-2009 16:07:12] <rmatte> yeh, that script is worth a shot
[10-Sep-2009 16:09:11] <chemist> not in this case
[10-Sep-2009 16:10:25] <rmatte> why?
[10-Sep-2009 16:10:34] <rmatte> you obviously have some sort of conflict, possibly caused by old items
[10-Sep-2009 16:11:53] <chemist> well, I ran the script and I still cant access my linux devices
[10-Sep-2009 16:12:04] <rmatte> you did a commit at the end?
[10-Sep-2009 16:13:29] <chemist> yes
[10-Sep-2009 16:14:09] <chemist> will I have to delete the devices and rediscover the?
[10-Sep-2009 16:14:16] <chemist> *them
[10-Sep-2009 16:14:35] <rmatte> it's possible
[10-Sep-2009 16:14:55] <chemist> is this likely to happen often?
[10-Sep-2009 16:14:57] <rmatte> what if you try moving a device to a different device class
[10-Sep-2009 16:15:01] <rmatte> are you then able to view it?
[10-Sep-2009 16:15:05] <metalklesk> ok, i got it now
[10-Sep-2009 16:15:17] <metalklesk> but the problen now is python version !!
[10-Sep-2009 16:15:19] <rmatte> I've never had anything like that happen, and I manage like 8 different Zenoss boxes
[10-Sep-2009 16:15:32] <rmatte> metalklesk: python version for what?
[10-Sep-2009 16:16:57] <chemist> no, that doesn't work either
[10-Sep-2009 16:17:09] <chemist> it tells me its moved it but it hasn't
[10-Sep-2009 16:17:28] <rmatte> ah, it could be a problem with the devices or with the actual organizer itself
[10-Sep-2009 16:17:34] <metalklesk> make ZenPacks.community.Mandriva
[10-Sep-2009 16:17:34] <rmatte> it's worth a shot removing them and re-adding them
[10-Sep-2009 16:17:42] <rmatte> test it with just 1 to begin with
[10-Sep-2009 16:17:54] <metalklesk> which: no python in (/bin)
\n[10-Sep-2009 16:17:54] <metalklesk> which: no python2.4 in\n(/usr/bin:/bin:/usr/local/bin:/usr/X11R6/bin/:/usr/games:/usr/lib/qt4/bin:/home/metalklesk/bin)
\n[10-Sep-2009 16:18:23] <mrayzenoss> metalklesk: you'll need to\nupdate your .bash_profile to include the Zenoss-provided Python
[10-Sep-2009 16:18:31] <chemist> 'zenoss error has occurred' when trying to delete
[10-Sep-2009 16:18:44] <metalklesk> yes, i jus saw that in build-egg.sh jejeje
[10-Sep-2009 16:18:47] <mrayzenoss> export ZENHOME=/usr/local/zenoss
[10-Sep-2009 16:18:47] <mrayzenoss> export PYTHONPATH=${ZENHOME}/zenoss/lib/python
\n[10-Sep-2009 16:18:47] <mrayzenoss> export\nPATH=$HOME/bin:$ZENHOME/bin:$ZENHOME/python/bin:/usr/local/bin:/usr/local/sbin:/opt/local/bin:/opt/local/sbin:/bin:/usr/bin:/usr/bin/X11:/sbin:/usr/sbin:/usr/X11R6/bin/
[10-Sep-2009 16:18:55] <metalklesk> ZENHOME is not defined
[10-Sep-2009 16:19:05] <mrayzenoss> that's from my .bash_profile
[10-Sep-2009 16:19:12] <metalklesk> thanks
[10-Sep-2009 16:19:30] <rmatte> chemist: wow
[10-Sep-2009 16:19:38] <rmatte> chemist: try restarting Zenoss, then try to remove them
[10-Sep-2009 16:20:31] <chemist> i'm on it
[10-Sep-2009 16:30:52] <rmatte> I'm out of here shortly
\n[10-Sep-2009 16:31:58] <home_run_dfc> is pynetsnmp available for\ndownload as a tarball on its own somewhere? the first google result\nfor it is on sourceforge, but i don't see any files for download
[10-Sep-2009 16:32:10] <metalklesk> i guess it works !!
[10-Sep-2009 16:32:19] <metalklesk> should i do a zenpack as root user ?
[10-Sep-2009 16:32:26] <metalklesk> *build
[10-Sep-2009 16:32:35] <chemist> rmatte: ok, I can now go to each location and delete the linux devices
[10-Sep-2009 16:32:53] <chemist> this is going to be fun
[10-Sep-2009 16:33:16] <metalklesk> ZenPacks.community.Mandriva-1.1-py2.4.egg, 20 kb
[10-Sep-2009 16:33:59] <rmatte> home_run_dfc: looks like the project may have been discontinued
[10-Sep-2009 16:34:21] <home_run_dfc> rmatte: isn't it the snmp library used by zenoss?
[10-Sep-2009 16:34:27] <rmatte> home_run_dfc: have you checked your package manager?
[10-Sep-2009 16:35:29] <rmatte> zenoss seems to use netsnmp, pynetsnmp, and twistedsnmp
\n[10-Sep-2009 16:36:19] <home_run_dfc> rmatte: zenoss has\npynetsnmp in their SVN --\nhttp://dev.zenoss.com/trac/browser/trunk/pynetsnmp
[10-Sep-2009 16:36:39] <rmatte> then grab it from there, it's probably just the last version that was publicly available
[10-Sep-2009 16:36:55] <home_run_dfc> rmatte: i was hoping for a tarball
\n[10-Sep-2009 16:36:56] <rmatte> generally if files aren't\navailable on a project's actual site, it indicates that the project is\nno longer active
[10-Sep-2009 16:37:38]\n<home_run_dfc> rmatte: doesn't zenoss consider it a problem that\nthey're using a no longer maintained library for what i would consider\na core component of their product?
[10-Sep-2009 16:38:32] <rmatte> http://archive.ubuntu.com/ubuntu/pool/universe/p/pynetsnmp/pynetsnmp_0.28.8.orig.tar.gz
[10-Sep-2009 16:38:36] <rmatte> that might work for you
\n[10-Sep-2009 16:38:52] <home_run_dfc> rmatte: i'm looking to get\nstarted coding some net management tools in python & was looking\nfor an snmp module to use & i figured zenoss would probably use a\ngood one
[10-Sep-2009 16:39:01] <chemist> so it looks like my zenoss box is hosed
[10-Sep-2009 16:39:14] <home_run_dfc> but i don't want something unmaintained
[10-Sep-2009 16:39:15] <rmatte> home_run_dfc: you do realize that Zenoss development started back in 1999 right?
\n[10-Sep-2009 16:39:28] <rmatte> home_run_dfc: if they were coding\nit starting right now they'd be using a lot different stuff
[10-Sep-2009 16:39:51] <rmatte> home_run_dfc: I doubt very much that they'd be using Zope for instance
[10-Sep-2009 16:39:52] <chemist> I can no longer view linux devices and I can't discover any new devices
[10-Sep-2009 16:40:18] <rmatte> chemist: yeh, sounds hosed, no idea how your Zope db got that corrupt
\n[10-Sep-2009 16:40:35] <home_run_dfc> rmatte: well then i was\nhoping someone from zenoss could offer me a suggestion on what they\nwould use for a python snmp library if what they're currently using\nisn't it
[10-Sep-2009 16:41:01] <home_run_dfc>\nconsidering they switched to pynetsnmp for version 2.0, i don't think\nthat decision goes all the way back to 1999
[10-Sep-2009 16:41:09] <chemist> I don't know much about zope, can I 'drop' the db?
[10-Sep-2009 16:41:48] <rmatte> 2.0 is a ways back
\n[10-Sep-2009 16:42:57] <home_run_dfc> rmatte: according to what i\nfound, 2.0 was released mid 2007 (obviously development of it started\nearlier than that, but i would guess maybe late 2004/early 2005)
[10-Sep-2009 16:43:08] <rmatte> hmmm
\n[10-Sep-2009 16:43:10] <rmatte> pynetsnmp is a replacement for\nthe various Python bindings provided by PySNMP* implementations\n(available as the Debian packages python-pysnmp*).
[10-Sep-2009 16:43:43] <rmatte> maybe it got integrated with pysnmp?
[10-Sep-2009 16:45:12] <rmatte> it'd kind of make sense if it did
\n[10-Sep-2009 16:47:20] <home_run_dfc> looks like pysnmp is the\nonly full-featured & maintained python snmp library at this point
[10-Sep-2009 16:47:28] <home_run_dfc> i guess that's what i'll try out
[10-Sep-2009 16:47:32] <home_run_dfc> rmatte: thx for the help
[10-Sep-2009 16:47:35] <rmatte> np
[10-Sep-2009 16:48:13] <chemist> rmatte: is there a way to recreate the zope db?
[10-Sep-2009 16:48:32] <metalklesk> it does not work
[10-Sep-2009 16:48:52] <metalklesk> i have edited the suse zenpack code to mandriva but no software was listed
[10-Sep-2009 16:49:09] <rmatte> chemist: not that I know of other than reinstalling Zenoss
[10-Sep-2009 16:49:30] <rmatte> chemist: the zope database is stored in $ZENHOME/etc/Data.fs
[10-Sep-2009 16:49:42] <chemist> sounds like my weekend off is no more
[10-Sep-2009 16:50:13] <rmatte> sorry...
[10-Sep-2009 16:50:19] <rmatte> chemist: the zope database is stored in $ZENHOME/var/Data.fs
[10-Sep-2009 16:50:24] <rmatte> not etc
[10-Sep-2009 16:51:03] <rmatte> if you see any *.zec files in $ZENHOME/var you could try deleting them then restarting Zenoss
[10-Sep-2009 16:51:06] <rmatte> but it's a long shot
[10-Sep-2009 16:52:58] <rmatte> well, I'm done for the day, I'll be back tomorrow
[10-Sep-2009 16:53:00] <rmatte> later guys
\n[10-Sep-2009 18:16:54] <simonpt> lo... i've got the zenoss vmware\nappliance... when i try to update raa it gives me this error: Failed:\n(Unknown Exception Type) MigrateError Cannot determine which group to\nuse as the top level group
[10-Sep-2009 18:35:14] <ironpaw> morning
[10-Sep-2009 19:01:24] <simonpt> how would one start to troubleshoot why you couldn't collet WMI data from a windows server?
[10-Sep-2009 19:14:39] <ironpaw> um
[10-Sep-2009 19:14:44] <ironpaw> check your zencommand logs
[10-Sep-2009 19:15:04] <ironpaw> you can run a zenwmi command from the shell to confirm connectivity/authentication details
\n[10-Sep-2009 19:19:43] <simonpt> looks like i had to install the\nwindows components... good to know about zencommand too thanks
[10-Sep-2009 19:19:50] <ironpaw> np
[10-Sep-2009 19:19:53] <ironpaw> get it working
[10-Sep-2009 19:20:06] <ironpaw> you will also need to configure your windows box to allow remote WMI queries
[10-Sep-2009 19:20:16] <ironpaw> got the WMIPerformance zenpacks?
[10-Sep-2009 19:24:51] <simonpt> i have the wmidatasource
[10-Sep-2009 19:24:55] <simonpt> zenpack
\n[10-Sep-2009 19:36:43] <simonpt> keep getting Event code: 3005\nEvent message: An unhandled exception has occurred. on the wmi device
[10-Sep-2009 19:36:48] <simonpt> the component is ASP.net
[10-Sep-2009 19:39:40] <ironpaw> hmm
[10-Sep-2009 19:39:49] <ironpaw> from zenoss or the eventlog in windows?
[10-Sep-2009 19:39:58] <ironpaw> sorry had to do multiple reboots
[10-Sep-2009 19:40:17] <simonpt> zenoss
[10-Sep-2009 19:41:13] <ironpaw> hmm just trying to find the page with theinfo
[10-Sep-2009 19:41:16] <ironpaw> please hold
[10-Sep-2009 19:41:27] * ironpaw hums some on-hold tunes
[10-Sep-2009 19:42:37] <ironpaw> To test WMI, find the wmic command (usually in $ZENHOME/bin/wmic), and run this command:
[10-Sep-2009 19:42:37] <ironpaw> wmic -U domain/administrator%password //host "Select * from Win32_ComputerSystem"
[10-Sep-2009 19:42:54] <ironpaw> can you run that on your device and does it return stuff?
[10-Sep-2009 19:43:36] <simonpt> yea that works fine
[10-Sep-2009 19:44:53] <ironpaw> hmm
[10-Sep-2009 19:45:11] <ironpaw> zenperfwmi eventlog give you any more details?
[10-Sep-2009 19:45:34] <simonpt> no... it's not counting the one server i have set to monitor
[10-Sep-2009 19:45:51] <ironpaw> in zneoss , try pushing changes to the device
[10-Sep-2009 19:45:54] <ironpaw> not sure if that'll work
[10-Sep-2009 19:45:59] <ironpaw> or just give it some time to re-poll
[10-Sep-2009 19:46:48] <simonpt> push changes... where is that? is it same as remodel?
[10-Sep-2009 19:47:09] <simonpt> found it
\n[10-Sep-2009 19:52:32] <simonpt> looks like it's just reporting\nthe 3005 error from an asp.net message in the windows event log
[10-Sep-2009 22:17:00] <ironpaw> *yawn*
[11-Sep-2009 00:00:46] [disconnected at Fri Sep 11 00:00:46 2009]
[12-Sep-2009 00:00:46] [connected at Sat Sep 12 00:00:46 2009]
[12-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[12-Sep-2009 15:05:37] <chemist> good evening
[12-Sep-2009 15:07:01] <chemist> how can I get zenoss to run a script on a remote server and report the result of the script
[12-Sep-2009 15:07:10] <chemist> if its at allpossible
[12-Sep-2009 18:48:32] <metalklesk> hi
[12-Sep-2009 19:55:15] rhettardo_ is now known as rhettardo
[13-Sep-2009 00:00:46] [disconnected at Sun Sep 13 00:00:46 2009]
[13-Sep-2009 00:00:46] [connected at Sun Sep 13 00:00:46 2009]
[13-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[13-Sep-2009 06:35:32] kevin7kal_ is now known as kevin7kal
\n[13-Sep-2009 12:35:54] <chemist> I run several processes called\n'Meeting (tk0)', 'Meeting (tk1)' etc up to tk17, they run at different\ntimes so I model the device every time the tk numbe rchangesw
\n[13-Sep-2009 12:36:27] <chemist> but despite remodelling, zenoss\nstill reports as failed the processes that are not running
\n[13-Sep-2009 12:37:33] <chemist> I have tried 'Push Changes'\nafter remodelling but it still shows me these 'ghost' processes
[13-Sep-2009 12:37:46] <chemist> I also tried reindexing the db
[13-Sep-2009 12:55:49] etank is now known as elake
[13-Sep-2009 12:56:35] elake is now known as etank
[13-Sep-2009 13:00:42] etank is now known as elake
[13-Sep-2009 13:00:53] elake is now known as etank
[13-Sep-2009 13:01:19] etank is now known as roninx341
[13-Sep-2009 13:01:33] roninx341 is now known as etank
[13-Sep-2009 14:02:40] <chemist> how can I make zenoss run a script on a remote linux box and report the outcome?
[13-Sep-2009 14:02:49] <chemist> if it's at all possible
[13-Sep-2009 14:03:15] <elake> chemist: yes it is
[13-Sep-2009 14:03:31] <elake> the output must be in a way that zenoss expects it though
[13-Sep-2009 14:03:44] <chemist> elake: can you give me some pointers?
[13-Sep-2009 14:04:04] <elake> chemist: one sec. i need to connect to the server and i will get you something
[13-Sep-2009 14:04:09] <chemist> ta
[13-Sep-2009 14:05:12] <elake> the way i did it was in a graph
[13-Sep-2009 14:05:33] <elake> in the device class that i want to see it in i created a new template
[13-Sep-2009 14:05:59] <elake> so for me i did it in /Devices/Server/Linux
[13-Sep-2009 14:06:10] <elake> i wanted to graph the ntp stratum of my servers
[13-Sep-2009 14:06:23] <elake> i created the template and named it ntpstratum
[13-Sep-2009 14:06:51] <elake> then created a data source named ntpstratum with the command template looking like this
[13-Sep-2009 14:06:58] <elake> /usr/sbin/ntpdc -n -c sysinfo | awk '/stratum/ { printf("|stratum=%s", $$2); }'
[13-Sep-2009 14:07:17] <elake> it will use ssh of the device to run the command
[13-Sep-2009 14:07:21] <elake> the return will look like
[13-Sep-2009 14:07:30] <elake> |stratum=4
[13-Sep-2009 14:07:41] <elake> the data point should be called stratum
[13-Sep-2009 14:07:52] <elake> must match the datasource output name
[13-Sep-2009 14:08:12] <elake> then add the datapoint to a graph and add any thresholds that you desire
[13-Sep-2009 14:08:41] <elake> this of course all assumes that you are wanting to run a command on a remote linux machine
[13-Sep-2009 14:08:59] <chemist> yes, a script in fact
[13-Sep-2009 14:09:38] <elake> the steps above are what i followed to get it to work
[13-Sep-2009 14:10:12] <chemist> Am I correct in assuming that in '|stratum=4', 4 is the severity?
[13-Sep-2009 14:10:25] <elake> chemist: yes
[13-Sep-2009 14:10:39] <chemist> cool, I will give that a go
[13-Sep-2009 14:10:48] <elake> well |stratum=4 is the output of the script
[13-Sep-2009 14:11:48] <elake> make sure the datapoint matches the |foo exactly
[13-Sep-2009 14:11:51] <elake> even in case
[13-Sep-2009 14:12:04] <chemist> ok
[13-Sep-2009 14:12:09] <elake> so datapoint Foo is not the same as a script returning |foo=x
[13-Sep-2009 14:13:04] <Snyper> all i ever see is how to make graphs but is there a way to get hardware info ?
[13-Sep-2009 14:13:21] <Snyper> get i mean use hardware data
\n[13-Sep-2009 14:14:50] <Snyper> i cant use snmp so im left trying\nto figure out how to add wmi support for hardware info.. i have all the\ninfo ready togo but cant figure out how to feed zenoss
[13-Sep-2009 14:19:21] <elake> Snyper: windows servers?
[13-Sep-2009 14:19:29] <Snyper> windows period
[13-Sep-2009 14:19:55] <elake> you have the wmi stuff installed on the windows servers right
[13-Sep-2009 14:20:03] <elake> and a user set in the zProperties
[13-Sep-2009 14:20:51] <Snyper> i have no problem pulling cpu, ram ..... stats for the graphs
[13-Sep-2009 14:21:10] <Snyper> its just hardware info, like os name, harddrive info.... ect
[13-Sep-2009 14:22:12] <Snyper> which i can pull from a script with no issue
[13-Sep-2009 14:22:15] <elake> you get some hardware info on the os tab
[13-Sep-2009 14:22:31] <Snyper> os tab shows only services
[13-Sep-2009 14:22:53] <elake> should see nics and filesystems too
[13-Sep-2009 14:23:27] <Snyper> non of that shows up
[13-Sep-2009 14:24:31] <elake> are you using core?
[13-Sep-2009 14:24:37] <Snyper> yes
[13-Sep-2009 14:26:01] <elake> there was a community plugin called wmi2 or something
[13-Sep-2009 14:26:13] <elake> it may help you get more information from the boxes
[13-Sep-2009 14:26:34] <elake> http://www.zenoss.com/community/projects/zenpacks/ <-- big list
[13-Sep-2009 14:26:51] <elake> if it is an hp server then there are some packs specific to them
[13-Sep-2009 14:27:00] <Snyper> all i get is graphs
[13-Sep-2009 14:27:16] <Snyper> i could give 2 sh*t about a graph
[13-Sep-2009 14:28:19] <elake> what kind of hardware info are you wanting to get?
[13-Sep-2009 14:28:33] <Snyper> generic windows box
[13-Sep-2009 14:29:01] <Snyper> most of those things for dell servers and hp servers seems to use snmp
[13-Sep-2009 14:29:38] <elake> curious. why cant you use snmp?
[13-Sep-2009 14:30:03] <Snyper> company does not allow it
[13-Sep-2009 14:30:11] <elake> i see
[13-Sep-2009 14:30:13] <Snyper> auditor go crazy
[13-Sep-2009 14:30:27] <Snyper> and they are too stupid to understand what v3 is
[13-Sep-2009 14:30:36] <elake> if only windows supported snmp v3
[13-Sep-2009 14:30:48] <Snyper> i have an addon to allow v3
[13-Sep-2009 14:30:59] <elake> got a link?
[13-Sep-2009 14:31:12] <elake> we are interested in using v3 on our servers
[13-Sep-2009 14:31:20] <elake> we have it working on our RHEL boxes
[13-Sep-2009 14:31:38] <Snyper> http://marksw.com/snmpv3agent/windowsagent.html
[13-Sep-2009 14:31:54] <Snyper> works great on my test env.... its just i cant use it
[13-Sep-2009 14:32:21] <elake> nice
[13-Sep-2009 14:33:30] <elake> Snyper: i wish i had more help to give but i am still new to zenoss myself
[13-Sep-2009 14:33:56] <Snyper> just annoying when you have all the data and no where to feed it to zenoss
[13-Sep-2009 14:34:51] <elake> there should be admins on tomorrow morning
[13-Sep-2009 14:35:12] <elake> they could offer more help
[13-Sep-2009 14:59:44] etank is now known as elake
[13-Sep-2009 22:20:59] <ironpaw> wasabi?
[13-Sep-2009 23:24:19] <rhett> horseradish
[14-Sep-2009 00:00:46] [disconnected at Mon Sep 14 00:00:46 2009]
[14-Sep-2009 00:00:46] [connected at Mon Sep 14 00:00:46 2009]
[14-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[14-Sep-2009 04:05:28] <chemist> morning all
[14-Sep-2009 04:06:20]\n<chemist> how do I stop zenoss from telling me that every device\nbehind a router has gone down, when only the router is down?
[14-Sep-2009 05:28:43] <JMan> morning
[14-Sep-2009 05:31:00] <JMan> anyone awake yet?
[14-Sep-2009 05:33:50] <rhettardo> i was going to say yes but oh well
[14-Sep-2009 06:34:22] <n215> does anybody have experience with zenoss and asa
\n[14-Sep-2009 07:45:23] <chemist> how do I stop zenoss from\ntelling me that every device behind a router has gone down, when only\nthe router is down?
[14-Sep-2009 08:10:00] <ckrough> chemist: It sounds like your outage paths are not working.
[14-Sep-2009 08:10:00] <ckrough> chemist: do some hunting in the forums for
[14-Sep-2009 08:10:00] <ckrough> for 'path outage'
[14-Sep-2009 08:10:00] <ckrough> chemist: also http://blog.zenoss.com/2008/08/26/tip-of-the-month-layer-3-dependency-checker/
[14-Sep-2009 08:10:00] <n215> hi
[14-Sep-2009 08:10:00] <n215> i get missing RRD FILE when monitoring cisco asa
[14-Sep-2009 08:10:00] <ckrough> n215: can you snmpwalk it manually from the zen server?
[14-Sep-2009 08:10:00] <chemist> ckrough: thanks, will have a look
[14-Sep-2009 08:10:00] <n215> ckrough i get some information using snmp
[14-Sep-2009 08:10:00] <n215> snmpwalk i mean
[14-Sep-2009 08:12:00] <ckrough> n215: try doing an snmpget for the datapoint that isnt graphing
[14-Sep-2009 08:12:00] ckrough is now known as coofamani
[14-Sep-2009 08:12:00] coofamani is now known as ckrough
[14-Sep-2009 08:12:08] <n215> cpu rrd works ok now
[14-Sep-2009 08:12:13] <n215> but memory rrd still missing
\n[14-Sep-2009 08:36:45] <ckrough> Anyone know where zenoss gets\nthe values for 'build-in' data sources like those used in the\nPerformanceConf template?
[14-Sep-2009 08:38:45] <ckrough> 'built-in;
[14-Sep-2009 08:40:11] <rmatte> ckrough: collector plugins
[14-Sep-2009 08:42:11] <ckrough> thanks
\n[14-Sep-2009 08:42:11] <ckrough> trying to figure out why my web\ninterfaces sometimes bog down when zenhub gets busy. running 7\nzenhubworkers on an 8 core box, maybe have to drop to 6 workers
[14-Sep-2009 09:04:20] <ckrough> mrayzenoss: do you *have* to sign up to watch that?
[14-Sep-2009 09:04:54] <mrayzenoss> umm... I think webex emails you the invite
[14-Sep-2009 09:04:59] <mrayzenoss> with the URL
[14-Sep-2009 09:05:48] <mrayzenoss> too much webex spam?
[14-Sep-2009 09:06:33] <ckrough> heh, not that I recall. was just curious.
[14-Sep-2009 09:09:25] <mrayzenoss> if there's a good turnout, we may have Jane do another session if you can't make it
[14-Sep-2009 09:09:25] <rmatte> yeh, you need to sign up
[14-Sep-2009 09:10:07] <rmatte> I hope there's a good turnout, it's worth it
[14-Sep-2009 09:10:07] <mrayzenoss> yeah, we've got 2 more sessions already in the pipe after this one
[14-Sep-2009 09:10:07] <mrayzenoss> I'll do one next month on the 2.5 release and the beta
[14-Sep-2009 09:10:07] <rmatte> oh, I think you just sign up for the event while it's running, not in advance?
[14-Sep-2009 09:10:07] <mrayzenoss> no, it should be in advance
[14-Sep-2009 09:10:07] <rmatte> oh, I see
[14-Sep-2009 09:10:07] <rmatte> I just wasn't looking at the right spot
[14-Sep-2009 09:10:07] <rmatte> "You are now registered for the event: Zenoss Community Webinar Series: Zenoss Event Architecture"
[14-Sep-2009 09:10:07] <rmatte> woohoo
[14-Sep-2009 09:10:11] <ckrough> /win
[14-Sep-2009 09:29:24] straterr1 is now known as straterra
[14-Sep-2009 09:31:04]\n<adelbul> hello, i installed zenplugins Zenoss-Plugins-2.0.4 on a\nopensolaris system when i execute zenplugin.py the result is :
\n[14-Sep-2009 09:31:16] <adelbul> root@srvnas03:~# zenplugin.py \nTraceback (most recent call last): File "/usr/bin/zenplugin.py", line\n49, in ? from zenoss.plugins.common import ApplicationError\nImportError: No module named common
[14-Sep-2009 09:32:07] <adelbul> i don't understand.
[14-Sep-2009 09:32:25] <adelbul> common.py and common.pyc are in the good directory.
[14-Sep-2009 09:32:42] <ckrough> try it as the zenoss user
[14-Sep-2009 09:32:55] <ckrough> 'su - zenoss'
[14-Sep-2009 09:33:10] <adelbul> ckrough: same result.
\n[14-Sep-2009 09:33:25] <adelbul> zenoss@srvnas03:~$ zenplugin.py \nTraceback (most recent call last): File "/usr/bin/zenplugin.py", line\n49, in ? from zenoss.plugins.common import ApplicationError\nImportError: No module named common
[14-Sep-2009 09:33:26] <ckrough> meh
\n[14-Sep-2009 09:33:41] <mrayzenoss> adelbul: You should take a\nlook at http://www.zenoss.com/community/projects/zenpacks/opensolaris\ninstead of doing it with a zenplugin
[14-Sep-2009 09:34:06] <mrayzenoss> adelbul: zenplugins are deprecated as of 2.4
\n[14-Sep-2009 09:34:56] <adelbul> zenoss@srvnas03:~$ ls -l\n/usr/lib/python2.4/site-packages/zenoss/plugins/common.py* -rw-r--r-- 1\nroot root 6435 2007-07-09 16:49\n/usr/lib/python2.4/site-packages/zenoss/plugins/common.py -rw-r--r-- 1\nroot root 6360 2009-09-14 14:33\n/usr/lib/python2.4/site-packages/zenoss/plugins/common.pyc
[14-Sep-2009 09:35:14] <adelbul> mrayzenoss: thanks i'm going to have a look.
\n[14-Sep-2009 09:36:48] <adelbul> anyway : nobody can help me for\ndebugging this "import" issue ? (juste for my personal culture ?)
[14-Sep-2009 09:37:56] <jb> yawn
[14-Sep-2009 09:40:58] <ckrough> mrayzenoss: where should we report 2.4.7 bugs? zenoss-testing or a trac ticket?
[14-Sep-2009 09:41:54] <rmatte> 2.4.7?
[14-Sep-2009 09:42:03] <ckrough> 2.5 beta
[14-Sep-2009 09:42:07] <mrayzenoss> 2.4.70
[14-Sep-2009 09:42:08] <mrayzenoss>
[14-Sep-2009 09:42:20] <rmatte> I didn't even know there was such a thing
[14-Sep-2009 09:42:22] <rmatte>
[14-Sep-2009 09:42:23] * ckrough doesnt process trailing zeros
[14-Sep-2009 09:42:35] <ckrough> rmatte: http://forums.zenoss.com/viewtopic.php?t=11001
[14-Sep-2009 09:42:35] <mrayzenoss> remind me to hire you and pay you $1 a day
[14-Sep-2009 09:42:47] <rmatte> hehe
[14-Sep-2009 09:43:01] <mrayzenoss> yeah, hit up dev.zenoss.com since the forums are locked
[14-Sep-2009 09:43:01] <ckrough> ha. well, least significant trailing zeros
[14-Sep-2009 09:43:15] <rmatte> oh, aka 2.5 beta, yeh
[14-Sep-2009 09:43:24] <jb> did the new community site not launch?
[14-Sep-2009 09:43:28] <ckrough> mrayzenoss: I see you've put up your cool hip avatar now
[14-Sep-2009 09:43:28] <mrayzenoss> it's launching
[14-Sep-2009 09:43:48] <mrayzenoss> well, Ohio LinuxFest told me they needed a picture
[14-Sep-2009 09:43:54] <mrayzenoss> and sent me a photo of me from last year
[14-Sep-2009 09:43:56] <mrayzenoss> so I used it
[14-Sep-2009 09:44:00] <rmatte> cool
[14-Sep-2009 09:44:32] <mrayzenoss> the forums are locked as we migrate to the new site
[14-Sep-2009 09:44:41] <mrayzenoss> they're getting pushed into the right places
[14-Sep-2009 09:44:49] <mrayzenoss> and we'll be polishing content all day
[14-Sep-2009 09:44:58] <mrayzenoss> if everything works as planned, we go live tonight
[14-Sep-2009 09:45:06] <rmatte> shaweet
[14-Sep-2009 09:47:04] <Stu> hey - anyone awake yet?
[14-Sep-2009 09:47:12] <ckrough> in 2.5 is 'close' equivelant to 'move to history'?
[14-Sep-2009 09:49:23] <mrayzenoss> depends on the defect
[14-Sep-2009 09:49:30] <mrayzenoss> oh, wait
[14-Sep-2009 09:49:30] <mrayzenoss> sorry, I misread that
\n[14-Sep-2009 09:49:30] <ckrough> I dont see a way to select all\nin the event console and move to hist, other than close. is that 'the\nway'?
[14-Sep-2009 09:49:35] <mrayzenoss> yeah, 'close' sends them to history
\n[14-Sep-2009 10:15:53] <Stu> hi - does anyone have any idea if\nit's possible to pull reports into an external app through a\n(semi)static URL?
[14-Sep-2009 10:24:58]\n<rmatte> Stu: not that I'm aware of, which is why I'm coding my\nown python scripts to generate reports
[14-Sep-2009 10:25:27] <rmatte> Stu: I'll most likely release what I've coded once it's done
[14-Sep-2009 10:39:30] <mrayzenoss> rmatte: Stu left
[14-Sep-2009 10:39:37] <rmatte> ah
[14-Sep-2009 10:39:37] <rmatte> I love impatient people
[14-Sep-2009 10:41:22] <skipzoid> hello
[14-Sep-2009 11:02:40] <skipzoid> bit quiet here
[14-Sep-2009 11:03:07] <rmatte> yup
[14-Sep-2009 11:03:28] <rmatte> I'm scripting so I won't be too chatty this morning
[14-Sep-2009 11:03:33] <rmatte>
[14-Sep-2009 11:03:42] <mrayzenoss> and I'm working on the Zenoss.Org launch
[14-Sep-2009 11:05:35]\n<skipzoid> hehe - is it possible to run a command from an event?\n- my sms sender runs on the command line as - sendsms <number>\n<message>
[14-Sep-2009 11:06:13] <mrayzenoss> skipzoid: http://www.zenoss.com/community/docs/zenoss-guide/2.4.3/ch07s01.html#d4e4203
[14-Sep-2009 11:07:17]\n<mrayzenoss>\nhttp://www.zenoss.com/community/docs/zenoss-dev-guide/2.4.2/apbs03.html\nhas the variables you have access to
[14-Sep-2009 11:09:24] <mrayzenoss> so you can do sendsms <number> $device is on fire!
[14-Sep-2009 11:09:33] <rmatte> skipzoid: click on Event Manager on the left, then click on the Commands tab
[14-Sep-2009 11:09:33] <rmatte> create a new command and set filter rules as you would for an alert
[14-Sep-2009 11:09:38] <rmatte> Here's an example of a command we use to create tickets in our ticketing system:
[14-Sep-2009 11:09:48]\n<rmatte> /mspfiles/scripts/create_ticket.pl -autonotify -customer\n'gen-00025' -device '${dev/id}' -deviceIP '${dev/manageIp}' -collector\n`hostname -f` -first '${evt/firstTime}' -last '${evt/lastTime}' -count\n'${evt/count}' -summary '${evt/summary}' -noteTitle 'System Monitor\nError' -note '${evt/message}' -severity '${evt/severity}' -group\n'${evt/DeviceGroups}' -impact '${evt/DevicePriority}' -component\n'${evt/component}' -queue 'Front Line'
[14-Sep-2009 11:10:01] <rmatte> all those ${} variables can be used
[14-Sep-2009 11:10:23] <skipzoid> wow didnt realise there where so many
[14-Sep-2009 11:10:31] <rmatte> there are some others too, but I think I pretty much covered the ones you'll need there
\n[14-Sep-2009 11:11:11] <skipzoid> the fire example is as much as\nwe need - it would then make the individual go look at the equipment to\ndeduce further what is the required action
\n[14-Sep-2009 11:14:19] <skipzoid> another problem i am having -\ni'm trying to monitor a process on a linux box, so within the processes\ni have defined the process as on page 67 of the guide and in my case as\nit is in the example it is ntpd, i then added this in the os part of\nthe device and it comes up green to show its running ? - then i kill it\noff so that on the next poll its not present but it still shows as\nrunning, i'm missing something here
[14-Sep-2009 11:18:26] <rmatte> skipzoid: remodel the device
[14-Sep-2009 11:18:29] <rmatte> then it'll pick up the process properly
[14-Sep-2009 11:19:21] <skipzoid> i have done that a few times - it then in the OS processes removes the item i added
\n[14-Sep-2009 11:27:25] <skipzoid> just worked it out - the\nprocess id must match the filename of the process runing and the regex\nseems to be ignored or i dont understand the syntax properly
[14-Sep-2009 11:33:49] <rmatte> skipzoid: you just type the name of the process
[14-Sep-2009 11:33:58] <rmatte> you don't need to use any fancy regex values or anything
[14-Sep-2009 11:34:16] <rmatte> if the process name is mysqld.bin then you add the process to the processes list as that
\n[14-Sep-2009 11:35:12] <skipzoid> yeah i did figure it was going\nto be that easy - i thought the name was the 'user' name of the process\nie - ntpd I wrote ntpdeamon in the regex put ntpd -
[14-Sep-2009 11:35:24] <rmatte> nope
[14-Sep-2009 11:35:35] <rmatte> hehe
[14-Sep-2009 11:58:17] <metalklesk> hi everyone
[14-Sep-2009 12:05:18] <mrayzenoss> So I've got a community question for non-English speakers, if any are lurking
[14-Sep-2009 12:05:47] <metalklesk> i speak spanish
[14-Sep-2009 12:06:02] <mrayzenoss> The new Community site has groups, for finding and discussing stuff with people of similar interests
[14-Sep-2009 12:06:33] <mrayzenoss> we were thinking about adding them for other languages, but would countries be preferred?
[14-Sep-2009 12:06:51] <mrayzenoss> ie. Spanish or Chile?
[14-Sep-2009 12:07:02] <metalklesk> Chile
[14-Sep-2009 12:07:10] <twm1010> do any of the zenpacks add DNS monitoring?
[14-Sep-2009 12:07:54] <mrayzenoss> twm1010: http://www.zenoss.com/community/projects/zenpacks/dnsmonitor
[14-Sep-2009 12:08:00] <metalklesk> i don't know if by countries would be a better choice
[14-Sep-2009 12:08:11] <twm1010> mrayzenoss: thx
[14-Sep-2009 12:08:18] <metalklesk> i do prefer language instead of country
[14-Sep-2009 12:08:48] <twm1010> im trying to hunt down a tool to help me gather DNS stats
[14-Sep-2009 12:08:48] <jb> dns monitoring how..
[14-Sep-2009 12:08:54] <jb> BIND?
[14-Sep-2009 12:08:55] <mrayzenoss> ok, I'll probably add groups like "Spanish/Español"
[14-Sep-2009 12:08:57] <metalklesk> we can get more sinergy
[14-Sep-2009 12:09:09] <twm1010> so I can see what are the common queries... etc
[14-Sep-2009 12:09:10] <jb> like queries, etc?
[14-Sep-2009 12:09:21] <jb> i graph queries per minute, etc
[14-Sep-2009 12:09:29] <jb> and use DigMon to check resolution
[14-Sep-2009 12:09:39] <twm1010> yeah, I'm thinking it will have to be something that does packet capturing
[14-Sep-2009 12:12:00] <metalklesk> what the ...., in community, someone took my nickname
[14-Sep-2009 12:12:15] <metalklesk> i don't remember to be registered
[14-Sep-2009 12:12:21] <mrayzenoss> which community?
[14-Sep-2009 12:12:25] <mrayzenoss> zenoss.org?
[14-Sep-2009 12:12:39] <metalklesk> https://community.zenoss.org
[14-Sep-2009 12:12:57] <metalklesk> i know there is no metalklesk in the entire world
[14-Sep-2009 12:13:02] <metalklesk> except me
[14-Sep-2009 12:13:03] <rmatte> hmmm
[14-Sep-2009 12:13:04] <rmatte> File "./availability.py", line 133, in ?
[14-Sep-2009 12:13:05] <rmatte> totaldown = totaldown + (leadin - startDate) + (leadout - endDate)
[14-Sep-2009 12:13:05] <rmatte> TypeError: unsupported operand type(s) for -: 'tuple' and 'float'
[14-Sep-2009 12:13:10] <tang> e
[14-Sep-2009 12:13:20] <tang> sorry, keyboard trauma
\n[14-Sep-2009 12:13:36] <mrayzenoss> metalklesk: I don't think\nyou're supposed to be able to login yet, we're putting up\nhttp://community.zenoss.org right now
[14-Sep-2009 12:13:48] <metalklesk> aahh ok
[14-Sep-2009 12:13:58] <metalklesk> jejejej
\n[14-Sep-2009 12:14:24] <metalklesk> maybe that's why the web page\nsays: Lo sentimos, ese nombre de usuario ya está registrado. Intente\ncon otro diferente.
[14-Sep-2009 12:14:24] <metalklesk> Contraseña: *
[14-Sep-2009 12:14:38] <metalklesk>
[14-Sep-2009 12:15:53] <metalklesk> i have a question
[14-Sep-2009 12:15:56] <metalklesk> about zenoss
[14-Sep-2009 12:16:03] <mrayzenoss> ok
[14-Sep-2009 12:16:29] <twm1010> go ahead
[14-Sep-2009 12:17:08]\n<metalklesk> how many disk space we can use in about 1 year of\nevent history ? let's say monitoring 100 boxes (windows, linux, cisco,\netc)
[14-Sep-2009 12:19:21] <mrayzenoss> depends\non how active your events are... but I'm sure someone here with real\nexperience can give anecdotal evidence
[14-Sep-2009 12:19:23] <metalklesk> that would be good
[14-Sep-2009 12:19:30] <metalklesk> rmatte maybe ?
[14-Sep-2009 12:19:30] * mrayzenoss install is only 10 devices, fits on a 40 gig drive
[14-Sep-2009 12:19:30] <mrayzenoss> ckrough? rmatte? How many devices and how big?
[14-Sep-2009 12:19:44] <mrayzenoss> your performance data size is mostly fixed, so that won't grow
\n[14-Sep-2009 12:19:58] <rmatte> metalklesk: we've been using a\n20gig virtual drive to monitor 370 devices since May, and we've only\nused up 11gigs of space total on the drive
[14-Sep-2009 12:20:08] <jb> yeah i use 20GB
[14-Sep-2009 12:20:11] <rmatte> and that's 11gigs including the OS and whatever else we have on it
[14-Sep-2009 12:20:17] <twm1010> what's the default location for the events DB under stack install/
[14-Sep-2009 12:20:18] <twm1010> ?
[14-Sep-2009 12:20:23] <jb> /dev/mapper/zenossvg-zenosslv
[14-Sep-2009 12:20:23] <jb> 20G 3.7G 16G 20% /opt/zenoss
[14-Sep-2009 12:20:25] <rmatte> so I think 20gig is more than sufficient provided you're rolling over every year
[14-Sep-2009 12:20:26] <jb> 200 devices
[14-Sep-2009 12:20:29] <jb> over a year
[14-Sep-2009 12:20:56] <rmatte> twm1010: the events db is mysql
[14-Sep-2009 12:20:58] <metalklesk> ok, thanks
[14-Sep-2009 12:21:14] <rmatte> twm1010: mysql -u zenoss -p
[14-Sep-2009 12:21:21] <rmatte> password is usually zenoss by default
[14-Sep-2009 12:21:22] <twm1010> looks like after 3 months mine, is 4.7gb
[14-Sep-2009 12:21:25] <rmatte> then do "connect events"
[14-Sep-2009 12:21:27] <ckrough> hard to translate it, since it depends on how busy your environment is
[14-Sep-2009 12:21:28] <rmatte> and voila
[14-Sep-2009 12:21:36] <twm1010> at least, the data file is
[14-Sep-2009 12:21:38] <jb> and if you are doing syslog
[14-Sep-2009 12:21:40] <jb> etc
[14-Sep-2009 12:22:11] <rmatte> ckrough: well, on how busy, and on how well you've tuned Zenoss to drop garbage events
[14-Sep-2009 12:22:24] <ckrough> that too
\n[14-Sep-2009 12:23:28] <twm1010> hrmm, good thing you guys\nbrought this, the volume my zenoss is installed to is 97% full
[14-Sep-2009 12:24:12] <metalklesk> im going to launch
[14-Sep-2009 12:24:14] <metalklesk> seeya
[14-Sep-2009 12:35:07]\n<ckrough> metalklesk: For a 1300 device install: ~5mil events in\n1yr history, 8.6GB event db. That's with *some* syslog stuff.
[14-Sep-2009 12:35:34] <ckrough> metalklesk: again, totally dependent on how talkative your gear will be.
\n[14-Sep-2009 12:40:39] <rmatte> yeh, if you're monitoring\nsomething like security syslogs your events db will get large very\nquickly
[14-Sep-2009 12:41:59] <ckrough> dump\nmost of that into other systems, just use zensyslog to pick up on small\nstuff that doesnt snmptrap
[14-Sep-2009 12:42:46] <ckrough> BGP changes, Trunk failures, some OSPF and spanning-tree messages
[14-Sep-2009 12:45:57] <rmatte> yeh, that's what we do
[14-Sep-2009 12:46:12] <rmatte> we have syslog-ng running between Zenoss and the monitored devices
[14-Sep-2009 12:46:23] <ckrough> same
[14-Sep-2009 12:46:26] <ckrough> you runs plunk?
[14-Sep-2009 12:46:30] <ckrough> ^s
[14-Sep-2009 12:46:42] <rmatte> don't believe so
[14-Sep-2009 12:46:49] <ckrough> awesome app. Splunk
[14-Sep-2009 12:47:04] <ckrough> version 4 UI is amazing.
[14-Sep-2009 12:47:15] <rmatte> oh, syslog analyzer?
[14-Sep-2009 12:47:17] <ckrough> looks like they are starting to encroach on monitoring and alerting space
[14-Sep-2009 12:47:41] <ckrough> anythinglog analyzer anymore
[14-Sep-2009 12:47:41] <rmatte> ah
[14-Sep-2009 12:47:57] <rmatte> ckrough...
[14-Sep-2009 12:47:58] <rmatte> [01:03pm] <rmatte> totaldown = totaldown + (leadin - startDate) + (leadout - endDate)
[14-Sep-2009 12:47:58] <rmatte> [01:03pm] <rmatte> TypeError: unsupported operand type(s) for -: 'tuple' and 'float'
[14-Sep-2009 12:48:01] <rmatte> any ideas?
[14-Sep-2009 12:48:22] <rmatte> is it complaining that one value is tuple and the other is float?
\n[14-Sep-2009 12:49:47] <rmatte> I'm just working on finishing up\nmy availability report script, got all of the mechanics worked out,\njust need to finish coding the functions to actually figure out the\npercentage of availability during a specified period
[14-Sep-2009 12:51:49] <djack> hello world
[14-Sep-2009 12:52:00] <ckrough> rmatte: yeah, are leadin and leadout tuples?
[14-Sep-2009 12:52:04] <rmatte> actually, I think I've found the issue
\n[14-Sep-2009 12:52:37] <rmatte> ckrough: yeh, but there's also an\nissue where sometimes totaldown will be None, which won't work, I need\nto add another if statement in there to check that
[14-Sep-2009 12:55:33] <rmatte> hmmm
[14-Sep-2009 12:55:43] <djack> is there a easy way to create a service status page outside zenoss using zenoss data?
[14-Sep-2009 12:57:11] <rmatte> ckrough: check this out...
[14-Sep-2009 12:57:13] <rmatte> if totaldown:
[14-Sep-2009 12:57:13] <rmatte> float(totaldown)
[14-Sep-2009 12:59:19] <rmatte> Traceback (most recent call last):
[14-Sep-2009 12:59:19] <rmatte> File "./availability.py", line 130, in ?
[14-Sep-2009 12:59:19] <rmatte> float(totaldown)
[14-Sep-2009 12:59:19] <rmatte> TypeError: float() argument must be a string or a number
[14-Sep-2009 12:59:20] <rmatte> so it's not a string, a number, or "None"
[14-Sep-2009 12:59:29] <rmatte> then what the heck is it?
[14-Sep-2009 12:59:29] <rmatte> lol
[14-Sep-2009 12:59:29] <rmatte> it's a numerical value, so there has to be some way to work with it
[14-Sep-2009 12:59:29] <ckrough> type(totaldown)
[14-Sep-2009 12:59:29] <rmatte> actually, weird, it looks like "(None,"
[14-Sep-2009 12:59:29] <rmatte> when I print it
[14-Sep-2009 12:59:29] <ckrough> see what it thinls
[14-Sep-2009 12:59:34] <rmatte> yeh
[14-Sep-2009 13:00:06] <ckrough> you may want to do a try/except expression rather than 'if'
[14-Sep-2009 13:00:15] <ckrough> try float(totaldown): do some stuff
[14-Sep-2009 13:00:19] <ckrough> except: other stuff
[14-Sep-2009 13:00:47] <rmatte> ah
[14-Sep-2009 13:00:52] <ckrough> as a general rule I dont think it's recommend to change a variable type in a test like that
[14-Sep-2009 13:01:03] <rmatte> gotcha
[14-Sep-2009 13:01:09] <rmatte> yeh, that may work better for sure...
[14-Sep-2009 13:04:14] <rmatte> File "./availability.py", line 133
[14-Sep-2009 13:04:14] <rmatte> float(totaldown):
[14-Sep-2009 13:04:19] <rmatte> it doesn't like that
[14-Sep-2009 13:04:34] <ckrough> "try float(totaldown):"
[14-Sep-2009 13:04:35] <rmatte> oh I didn't add try
[14-Sep-2009 13:04:36] <rmatte> my bad
[14-Sep-2009 13:05:17] <rmatte> zenoss@gen01:~/reports$ ./availability.py > availtest.txt
[14-Sep-2009 13:05:17] <rmatte> File "./availability.py", line 133
[14-Sep-2009 13:05:17] <rmatte> try float(totaldown):
[14-Sep-2009 13:05:17] <rmatte> ^
[14-Sep-2009 13:05:17] <rmatte> SyntaxError: invalid syntax
[14-Sep-2009 13:05:20] <rmatte> apparently not
[14-Sep-2009 13:05:32] <ckrough> this could be a bit of the blind leading the blind
[14-Sep-2009 13:05:37] <rmatte> lol
[14-Sep-2009 13:07:03] <ckrough> it worked in my test
[14-Sep-2009 13:08:43] <ckrough> oh, "try: float(test)"
[14-Sep-2009 13:08:43] <ckrough> followed by some stuff
[14-Sep-2009 13:08:55] <rmatte> ah
[14-Sep-2009 13:08:55] <ckrough> http://docs.python.org/tutorial/errors.html#handling-exceptions
[14-Sep-2009 13:11:21] <rmatte> ah, I know what the issue is
[14-Sep-2009 13:11:31] <rmatte> totaldown: (710.48900032043002,)
[14-Sep-2009 13:11:37] <rmatte> totaldown: (None,)
[14-Sep-2009 13:12:06] <ckrough> ha
[14-Sep-2009 13:12:06] <rmatte> I'm using totaldown = cursor.fetchone() and it's outputting the data like that
[14-Sep-2009 13:12:44] <rmatte> there has to be a better way of getting the data out as an actual number
[14-Sep-2009 13:12:48] <rmatte> rather than some weird string
[14-Sep-2009 13:13:09] <rmatte> (the data is being pulled for the sql events database)
[14-Sep-2009 13:13:13] <rmatte> from*
\n[14-Sep-2009 13:13:24] <ckrough> cursor.fetchone() "Fetches the\nnext row of a query result set, returning a single sequence, or None\nwhen no more data is available."
[14-Sep-2009 13:13:40] <rmatte> yeh, but it also returns those (,)
[14-Sep-2009 13:13:42] <rmatte> apparently
[14-Sep-2009 13:14:05] <rmatte> so it's like (value,) or (None,)
[14-Sep-2009 13:14:05] <ckrough> its a sequence
[14-Sep-2009 13:14:17] <rmatte> ah, how can I just pull whatever is output?
[14-Sep-2009 13:14:30] <rmatte> cursor.fetchall?
[14-Sep-2009 13:14:55] <ckrough> maybe "try: float(totaldown[0])
[14-Sep-2009 13:15:46] <rmatte> what will the 0 do?
[14-Sep-2009 13:16:16] <ckrough> returns the value in the first position in the sequence
[14-Sep-2009 13:16:26] <ckrough> counting starts from 0
[14-Sep-2009 13:16:26] <rmatte> ah
[14-Sep-2009 13:17:10] <ckrough> >>> totaldown=("dog","cat")
[14-Sep-2009 13:17:10] <ckrough> >>> totaldown[1]
[14-Sep-2009 13:17:11] <ckrough> 'cat'
[14-Sep-2009 13:18:36] <rmatte> float(totaldown[0])
[14-Sep-2009 13:18:36] <rmatte> TypeError: float() argument must be a string or a number
[14-Sep-2009 13:20:52] <rmatte> fetchone() retrieves a row as a tuple
[14-Sep-2009 13:21:53] <rmatte> now I just need to figure out how to convert a tuple to a float
\n[14-Sep-2009 13:22:31] <ckrough> you cant convert a tuple to a\nfloat since a tuple typically contains multiple values, but you should\nbe able to turn the first value in the tuple into a float with that\nlogic
[14-Sep-2009 13:22:47] <rmatte> right, but apparently not
[14-Sep-2009 13:22:57] <ckrough> http://pastebin.com/m363c0ff7
[14-Sep-2009 13:23:07] <ckrough> looks like it should work
[14-Sep-2009 13:26:03] <rmatte> the except runs when it passes or when it fails?
\n[14-Sep-2009 13:27:13] <rmatte> oh, it probably wasn't doing it\nbecause I didn't have a statement to actually check whether it's "None"
[14-Sep-2009 13:27:40] <rmatte> I technically could just do something like: if totaldown[0]:
[14-Sep-2009 13:28:21] <rmatte> aha, it seems to have possibly worked
[14-Sep-2009 13:28:41] <rmatte> yeh, it did
[14-Sep-2009 13:28:46] <ckrough> what did you change?
[14-Sep-2009 13:29:35] <rmatte> I just did a try
[14-Sep-2009 13:29:45] <rmatte> I think it was trying to do float on None values before
[14-Sep-2009 13:29:52] <rmatte> but with try it tries and if it fails it moves on
[14-Sep-2009 13:29:59] <rmatte> do you even need the "except" part in a try?
[14-Sep-2009 13:30:13] <ckrough> dunno, never did it without
[14-Sep-2009 13:33:37] <rmatte> doesn't work without it
[14-Sep-2009 13:33:38] <rmatte> bah
[14-Sep-2009 13:35:03] <ckrough> "except: pass"
[14-Sep-2009 13:35:23] <ckrough> but you probably want something to happen if it fails, dont you?
[14-Sep-2009 13:35:59] <rmatte> yeh, I just want it to move along
[14-Sep-2009 13:36:09] <rmatte> I just need to get 3 values converted from tuple to float
[14-Sep-2009 13:36:13] <rmatte> so that I can work with them
[14-Sep-2009 13:36:46] <rmatte> actually, I could have it set them to "None" if it fails
[14-Sep-2009 13:36:55] <rmatte> I'll give that a shot
[14-Sep-2009 13:40:13] <rmatte> so far so good, no errors, and it normally would have errored by now...
[14-Sep-2009 13:42:51] <ckrough> nice
[14-Sep-2009 13:43:02] <rmatte> it eventually did error, going to try some more stuff...
[14-Sep-2009 13:43:02] <djack> what are we talking about ?
[14-Sep-2009 13:43:12] <rmatte> a script I'm writing
[14-Sep-2009 13:43:22] <djack> and that does what?
[14-Sep-2009 13:44:35] <rmatte> it's a script that's going to spit out an availability report
[14-Sep-2009 13:45:45] <djack> hmmm intresting
[14-Sep-2009 13:46:03] <djack> can i ask 2 small questions ?
[14-Sep-2009 13:46:12] <rmatte> yes
[14-Sep-2009 13:46:49] <twm1010> ask away
[14-Sep-2009 13:47:00] <djack> is there some easy way to create a service overview outside zenoss using the zenoss imput ?
[14-Sep-2009 13:47:23] <djack> i dont want to create the wheel if it already exist
[14-Sep-2009 13:47:23] <rmatte> oh, except is if it passes the test?
[14-Sep-2009 13:48:39] <rmatte> djack: you're basically talking about some sort of external portal?
[14-Sep-2009 13:48:39] <djack> yeah something for the end users to see
[14-Sep-2009 13:48:39] <twm1010> like for a kiosk?
[14-Sep-2009 13:48:39] <djack> yep
\n[14-Sep-2009 13:48:39] <rmatte> well, you can either have them\nactually log in to Zenoss, or you can spend time doing development work\nto develop one, which is basically what I'm doing right now
[14-Sep-2009 13:48:45] <rmatte> oh, kiosk eh?
[14-Sep-2009 13:48:52] <rmatte> one sec, I have something that might be of interest...
[14-Sep-2009 13:49:08] <rmatte> http://forums.zenoss.com/viewtopic.php?p=24470
[14-Sep-2009 13:49:16] <rmatte> not sure if that's exactly what you're looking for or not
[14-Sep-2009 13:49:24] <venturaville> is anyone using sun open storage?
[14-Sep-2009 13:49:31] <djack> well tv screen for our servicedesk
[14-Sep-2009 13:50:09] <rmatte> djack: yeh, you could use that method for that, or you can do dev work and design your own
\n[14-Sep-2009 13:50:25] <rmatte> djack: it's easy enough to\nextract the data from Zenoss if you're a decent Python scripter
[14-Sep-2009 13:50:43] <djack> rmatte php is my thing :-p
[14-Sep-2009 13:50:50] <rmatte> ah
[14-Sep-2009 13:50:57] <djack> can not be difficult if it is in a mysql db
\n[14-Sep-2009 13:51:27] <rmatte> well, we have an actual\nphp/python/whatever the heck else we need developer on staff who is\ngoing to do the more advanced stuff for it, I'm just writing the\nscripts that can be called on to extract the report data
\n[14-Sep-2009 13:51:32] <djack> i'll check it out have to create\none for the servicdesk and a simple protected on for the onclall iphone
[14-Sep-2009 13:52:12] <djack> rmatte generic scripts ? of specific for your company?
[14-Sep-2009 13:52:18] <twm1010> only event data is in MySQL
[14-Sep-2009 13:52:23] <twm1010> zenoss data is in ZOPE
[14-Sep-2009 13:52:47] <rmatte> djack: the scripts themselves will be quite generic
[14-Sep-2009 13:53:01] <rmatte> I'm probably going to release them publicly once we're done with this little project
[14-Sep-2009 13:53:03] <twm1010> Don't see why the kiosk solution wouldn't work.
[14-Sep-2009 13:53:33] <djack> twm1010 will chech it out
\n[14-Sep-2009 13:54:59] <djack> i don't want to give access to the\nservicedesk its not the job of a servicedesker to go check stats etc ...
[14-Sep-2009 13:55:13] <twm1010> I'd be concerned about that mode exposing information
[14-Sep-2009 13:55:40] <djack> twm1010 how do you mean?
\n[14-Sep-2009 13:56:21] <rmatte> djack: I'm assuming by\nservicedesk you mean regular helpdesk "Hi my computer is broken" type\nstaff, and not actual NOC staff?
[14-Sep-2009\n13:57:58] <djack> rmatte we do not have noc staff ... the first\nline are our nike shoes pick up call and run thrue the building
[14-Sep-2009 13:58:02] <ckrough> SneakerNet
[14-Sep-2009 13:59:06]\n<djack> they only have to notice that something is wrong that's\nit .... second the last line of defence (yeah that me and it sucks)\nshould analyse and fix issues
[14-Sep-2009 14:00:07] <rmatte> lol
[14-Sep-2009 14:00:43] <rmatte> yeh, we do things quite differently here
[14-Sep-2009 14:00:49] <djack> rmatte well everybody has his job when that is done they can do more ... but thats not
[14-Sep-2009 14:01:24] <djack> if you have 140 open tickets you should close them first and not play around
[14-Sep-2009 14:01:44] <djack> rmatte read we have bad first line staff ....
[14-Sep-2009 14:02:05] <rmatte> lol
[14-Sep-2009 14:02:19] <djack> rmatte noooooooooooooooooo :-p
[14-Sep-2009 14:03:00] <djack> rmatte is there info about getting the data out of zope ?
[14-Sep-2009 14:06:32] <rmatte> djack: read the developer's guide
[14-Sep-2009 14:06:53] <rmatte> depends on what kind of data you need
[14-Sep-2009 14:07:03] <djack> will do
[14-Sep-2009 14:07:22] <djack> basic stuff
[14-Sep-2009 14:07:38] <djack> second question is there a jmx expert that can help me out ?
[14-Sep-2009 14:07:50] <rmatte> I'm certainly not one
[14-Sep-2009 14:07:55] <rmatte> unfortunately
[14-Sep-2009 14:08:33] <djack> me neither
\n[14-Sep-2009 14:10:19] <rmatte> for some reason it's treating the\nconversion of the totaldown value differently than the other 2 values
[14-Sep-2009 14:11:09] <metalklesk> i'm back
[14-Sep-2009 14:24:02] <rmatte> eugh, there's something seriously wacky with this script
[14-Sep-2009 14:24:38] <rmatte> oh I see why
[14-Sep-2009 14:24:42] <rmatte> stupid me
[14-Sep-2009 14:26:42] <rmatte> ther we go, now I'm getting somewhere
[14-Sep-2009 14:32:56] <djack> wacky scripts are fun
[14-Sep-2009 14:33:28] <djack> rmatte i defined a java process that he does not want to graph
[14-Sep-2009 14:33:37] <cgibbons> hmmm
[14-Sep-2009 14:34:04] <rmatte> djack: I've never used the JMX ZenPack, so I can't really help with it
[14-Sep-2009 14:34:13] <rmatte> djack: are you seeing a graph at all?
[14-Sep-2009 14:34:17] <djack> no not that an OS process
[14-Sep-2009 14:34:22] <rmatte> ah
[14-Sep-2009 14:34:24] <djack> the jmx is running ;-)
[14-Sep-2009 14:34:29] <rmatte> so you added the process and remodeled the device?
[14-Sep-2009 14:34:38] <djack> yep
[14-Sep-2009 14:34:44] <rmatte> how long ago?
\n[14-Sep-2009 14:35:11] <djack> there are currently 2 java\nprocesses running the one give a nice output ... but its not relevant\nthe other that is will not graph
[14-Sep-2009 14:35:16] <rmatte> takes a minimum of 3 polling cycles (15 minutes by default) for a graph to get populated
[14-Sep-2009 14:35:18] <djack> last friday
[14-Sep-2009 14:35:23] <rmatte> ah
[14-Sep-2009 14:35:28] <djack> should be long ago not?
[14-Sep-2009 14:35:41] <rmatte> you're sure you added it with the right name?
[14-Sep-2009 14:35:48] <rmatte> it needs to be fairly exact
[14-Sep-2009 14:36:49] <djack> Regex ^.*\\/*java
[14-Sep-2009 14:37:17] <djack> and it worked before ....
[14-Sep-2009 14:38:30] <djack> i checked and no rrd files are on the system
[14-Sep-2009 14:38:50] <rmatte> no, don't actually use regex syntax
[14-Sep-2009 14:38:55] <rmatte> just literally type the process name
[14-Sep-2009 14:40:21] <djack> hmmm
[14-Sep-2009 14:40:37] <djack> the name was with a capital ....
[14-Sep-2009 14:41:39] <rmatte> it's case sensitive as well
[14-Sep-2009 14:41:41] <rmatte> it's linux
[14-Sep-2009 14:42:12] <djack> yeah silly me
[14-Sep-2009 14:42:30] <djack> I'm going to sit in the corner and cry
[14-Sep-2009 14:43:00] <rmatte> hehe
[14-Sep-2009 14:43:17] <Apocalipse> hi people...
[14-Sep-2009 14:43:29] <Apocalipse> one question about linking locations
[14-Sep-2009 14:44:27] <Apocalipse> i model my core router... and other routers linked on it
[14-Sep-2009 14:44:40] <Apocalipse> but its not showing links between then
[14-Sep-2009 14:46:27] <Apocalipse> any idea?
\n[14-Sep-2009 14:54:40] <rmatte> Apocalipse: It depends on the\ntypes of routes, whether you have modelled complete related routes,\nit's possible that you won't be able to get the links to show like I\ntold you a few days ago
[14-Sep-2009 15:03:08] <Apocalipse> rmatte... after that u told me... i re-modelled all routers...
\n[14-Sep-2009 15:10:28] <rmatte> I told you that I have a Zenoss\nbox where we can't get it to draw lines no matter what, and we actually\ndo monitor the core router as well
[14-Sep-2009 15:10:36] <rmatte> it has something to do with the way that the routing is setup
[14-Sep-2009 15:11:02] <Apocalipse> ok
[14-Sep-2009 15:11:05] <Apocalipse> thx
[14-Sep-2009 15:19:06] <rmatte> Apocalipse: also make sure that all of your locations are actually on the map
[14-Sep-2009 15:19:13] <rmatte> it's fairly picky about location info
[14-Sep-2009 15:19:25] <Apocalipse> yes they are
[14-Sep-2009 15:30:17] <djack> is there a way to auto model devices ?
[14-Sep-2009 15:30:36] <djack> lets say after the device is back up ?
\n[14-Sep-2009 15:39:24] <rmatte> npmccallum: any idea where the\nscript that works out availability for the availability report is\nstored?
[14-Sep-2009 15:39:33] <rmatte>\nnpmccallum: I've made my own but the values need to jive and they\naren't exactly right now
[14-Sep-2009 15:39:51] <npmccallum> rmatte: which availability report are you talking about?
[14-Sep-2009 15:40:14] <npmccallum> rmatte: the device availability report?
[14-Sep-2009 15:40:14] <rmatte> In Zenoss Core - Reports -> Performance Reports -> Availability
[14-Sep-2009 15:40:18] <rmatte> correct
[14-Sep-2009 15:41:25] <rmatte> I went in to the .rpt file and tried to figure it out
[14-Sep-2009 15:41:28] <rmatte> but couldn't
[14-Sep-2009 15:41:44] <rmatte> so I've been writing one from scratch, but I've run in to an interesting issue with it
[14-Sep-2009 15:51:24] <npmccallum> rmatte: there does not appear to be a separate file for that report
[14-Sep-2009 15:51:33] <metalklesk> how can i submit my zenpack of mandriva ?
[14-Sep-2009 15:52:06] <mrayzenoss> metalklesk: you can email me, mray@zenoss
[14-Sep-2009 15:52:09] <metalklesk> it's just an edited zenpack of suse (opensuse)
[14-Sep-2009 15:52:55] <rmatte> npmccallum: hmmm
[14-Sep-2009 15:53:12] <npmccallum> rmatte: it is just using the built in API, so none is required
[14-Sep-2009 15:54:10] <npmccallum> rmatte: availability is actually more complicated than it would seem at first
[14-Sep-2009 15:54:28] <npmccallum> rmatte: you have to deal with things like coalesced vs averaged, etc
\n[14-Sep-2009 15:55:12] <rmatte> basically, I just need to see how\nyou guys are actually calculating the availability over a specific\nperiod of time... so far what my script does is it collects how long\nthe device was down for during the period, but it also finds any ping\ndown events that completely overlap the window, if it completely\noverlaps it's 0% availability, and on top of that, it finds any events\nthat start before the window and end during it, then take the value and\nsubtract....
[14-Sep-2009 15:55:31] <rmatte> against the time from the start of the specified window.
[14-Sep-2009 15:56:02]\n<rmatte> It also does it for ping events starting during the\nwindow and ending afterwards, but it subtracts against the end of the\nspecified window
[14-Sep-2009 15:56:04] <npmccallum> rmatte: check out the availability() method on the Device class
\n[14-Sep-2009 15:56:32] <rmatte> now theoretically, I should come\nout with a total number that is smaller or even to the window\nspecified, but I never do
[14-Sep-2009 15:58:25] <metalklesk> mray@zenoss.com right ?
[14-Sep-2009 15:58:37] <metalklesk> .org ? .net ?. cl ? jeje
[14-Sep-2009 15:58:58] <rmatte> ah, I think I might have found what's causing the issue
[14-Sep-2009 15:59:17] <npmccallum> metalklesk: yes, mray@zenoss.com
[14-Sep-2009 16:05:19] <metalklesk> mrayzenoss, i have send you an email with the zenpack
[14-Sep-2009 16:07:23] <mrayzenoss> metalklesk: got it, I'll get back to you soon
[14-Sep-2009 16:07:46] <metalklesk> ok, thanks
[14-Sep-2009 16:21:38] <Apocalipse> hello... iam back
[14-Sep-2009 16:29:46] <metalklesk> hello Apocalipse
[14-Sep-2009 16:30:09] <metalklesk> mrayzenoss, i have send you an email with the screenshots of the mandriva zenpack
[14-Sep-2009 16:30:25] <metalklesk> i will fill out the contribution form right now
[14-Sep-2009 16:30:32] <mrayzenoss> cool, thanks
\n[14-Sep-2009 16:30:54] <mrayzenoss> like I said in the email,\nI've got a backlog of zenpacks because of this site launch, but it'll\nbe up soon
[14-Sep-2009 16:31:28] <metalklesk> ok
[14-Sep-2009 16:38:29] <Apocalipse> matt.. is the new website done?
[14-Sep-2009 16:40:13] <rmatte> Apocalipse: no, it's not done yet
[14-Sep-2009 16:40:49] <mrayzenoss> don't worry, we're working on it
[14-Sep-2009 16:41:05] <mrayzenoss> not sure it'll ever be 'done'
[14-Sep-2009 16:41:16] <rmatte> hehe
[14-Sep-2009 16:41:22] <mrayzenoss> apparently some of our spam got migrated over
[14-Sep-2009 16:41:49] <mrayzenoss> just searched for "xbox, iphone, playstation, blackberry" and found a bunch
\n[14-Sep-2009 16:42:39] <rmatte> mother of god, I think I finally\nhave this availability report working properly, found a bunch of stupid\nlittle errors in the SQL queries I had written up
[14-Sep-2009 16:50:01] <rmatte> I do! YES!
[14-Sep-2009 16:50:04] <rmatte> VICTORY!
[14-Sep-2009 16:50:23] <rmatte> now I just need to remove all of the debug lines from the script and clean up the output
[14-Sep-2009 16:50:34] <mrayzenoss> I look forward to seeing your work on the new site
[14-Sep-2009 16:50:39] <rmatte> then I need to figure out how to add options to a python script
[14-Sep-2009 16:50:46] <rmatte> oh, it'll be up there
[14-Sep-2009 16:50:57] <rmatte> this frigging script took me like 2 entire days to figure out and write
\n[14-Sep-2009 16:51:41] <rmatte> oh, this also pretty much\nconfirms that the availability report in Zenoss is fully accurate,\nwhich is good for you
[14-Sep-2009 16:51:42] <rmatte>
[14-Sep-2009 16:54:15] <mrayzenoss> finally, you find something not broken!
[14-Sep-2009 17:00:21] <rmatte> hehe
[14-Sep-2009 17:00:52] <rmatte> the exporting in the availability report actually works half the time too, but it's still semi broken
[14-Sep-2009 17:01:21] <rmatte> it'll export from the last 7 days sometimes, but most of the time it exports during the dates specified
[14-Sep-2009 17:02:07] <rmatte> well, now I just need to polish this script up and it's finally done
[14-Sep-2009 17:23:48] <rmatte> Matt, check this out: http://pastebin.com/m492ac589
[14-Sep-2009 17:23:54] <rmatte> the availability script
[14-Sep-2009 17:24:07] <rmatte> it's not 100% done yet, still need to add options for the start and end dates
[14-Sep-2009 17:24:15] <rmatte> but other than that, it's fully functional
[14-Sep-2009 17:24:19] <rmatte> outputs in csv format
[14-Sep-2009 17:26:44] <rmatte> actually, that how='AVERAGE' line doesn't need to be in there
[14-Sep-2009 17:26:47] <rmatte> going to remove that
[14-Sep-2009 17:27:01] <mrayzenoss> cool, we're very close to launching
[14-Sep-2009 17:27:06] <rmatte> nice
[14-Sep-2009 17:42:10] <Apocalipse> i am floating...on the subject
[14-Sep-2009 17:42:54] <rmatte> float(Apocalipse)
[14-Sep-2009 17:42:57] <rmatte> now you are
[14-Sep-2009 17:43:03] <Apocalipse> kkkkk
[14-Sep-2009 17:43:12] * rmatte couldn't resist a little bit of Python humour
[14-Sep-2009 17:43:19] <Apocalipse> hehehhehehe
[14-Sep-2009 17:43:33] <rmatte> well, I'm out of here, way pay time for me to leave
[14-Sep-2009 17:43:37] <rmatte> have a nice evening folks
[14-Sep-2009 17:43:47] <Apocalipse> same to u rmatte
[14-Sep-2009 17:43:51] <rmatte> way past time for me to leave* rather
[14-Sep-2009 17:44:00] <rmatte> not pay, though I get paid tomorrow
[14-Sep-2009 17:44:01] <rmatte> woot
[14-Sep-2009 19:50:08] <jb> hm
[14-Sep-2009 19:50:21] <jb> can't login to the community site.. or retreive my password
[14-Sep-2009 20:26:59] <ironpaw> hi
[15-Sep-2009 00:00:46] [disconnected at Tue Sep 15 00:00:46 2009]
[15-Sep-2009 00:00:46] [connected at Tue Sep 15 00:00:46 2009]
[15-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
[15-Sep-2009 04:20:08]\n<adelbul> hello, when i want to "Add a zenpack" to a class of\ndevice. When I click on "Add a zenpack" the loading screen never stop.\nHave you encounter a similar issue ?
[15-Sep-2009 04:23:05] <Troubadix09> adelbul: It is not "add a zenpack", it is "add TO zenpack" in the device class
[15-Sep-2009 04:23:50] <Troubadix09> adelbul: "add a zenpack" is under "settings/ZenPacks"
\n[15-Sep-2009 04:24:59] <Troubadix09> adelbul: in device class it\nis for creating an own zenpack, in settings/ZenPacks it is for adding\ndownloaded ZenPacks
[15-Sep-2009 04:28:20]\n<Troubadix09> adelbul: normaly ZenPacks itself determine to wich\nclass the included templates added or it install template that must be\nmanually bind to devices or a device class
[15-Sep-2009 04:33:18] <adelbul> Troubadix09: sorry i wanted to say "add to zenpack".
[15-Sep-2009 04:33:50] <adelbul> i added the zenpack with the command line tool zenpack --install
[15-Sep-2009 04:35:11] <adelbul> I create a device class /Server/SSH/Solaris.
[15-Sep-2009 04:35:26] <adelbul> then i installed the zenpack for opensolaris.
[15-Sep-2009 04:36:06] <adelbul> and now i try to add the opensolaris zenpack to the class /Server/SSH/Solaris
[15-Sep-2009 04:37:21] <adelbul> docs/DOC-3414
[15-Sep-2009 04:40:33] <adelbul> if i understand your commentm you say that i just have to bind the templates ?
\n[15-Sep-2009 04:51:59] <Troubadix09> adelbul: sorry, i didn't use\nsolaris, but I think, that they would install to the device-class\n"/Server/SSH/Solaris" when you create this class before installing. You\ncan control it ...
[15-Sep-2009 04:53:05] <Troubadix09> adelbul: on the "Templates"-Tab on the device-class...
\n[15-Sep-2009 04:55:33] <Troubadix09> adelbul: sometimes a ZenPack\ninstalls new collector plugins, wich must be added manually for example\nthe HP Proliant ZenPack brings new collector plugins wich replace the\noriginal zenoss core collector plugins
[15-Sep-2009\n04:58:59] <Troubadix09> adelbul: you can't add ZenPacks to device\nclasses, because ZenPacks are packages of configuration like add new\nTemplates, collector Plugins, Data-Sources to existing Templates, new\nTabs, new Dashboard-Portlets etc.
[15-Sep-2009 05:05:03]\n<Troubadix09> adelbul: In this sense it is like a recording of\nyour changes in zenoss always you use the "Add to Zenpack"-Button, wich\nthen you can give to the community
[15-Sep-2009 05:20:36] <adelbul> Troubadix09: ok thank you for this explanation.
[15-Sep-2009 06:24:30] <skipzoid> hey
\n[15-Sep-2009 06:27:21] <skipzoid> in the event manager i've\ncreated an event command to be run when an event ocours - sendsms.pl \n"$device ${evt/message}" and on the command line if this is typed as\nthe zenoss user it does send an sms - how do i attach this event\ncommand to an actual even ocourance?
[15-Sep-2009 06:48:07] <ckrough> morning
[15-Sep-2009 06:52:33] <Troubadix09> ckrough: morning is good, here it is afternoon ;-)
[15-Sep-2009 06:54:38] <ckrough> good afternoon
[15-Sep-2009 07:01:32] <skipzoid> hello
[15-Sep-2009 07:02:00] <skipzoid> what variables can be passed to an event command
[15-Sep-2009 07:07:18] <adelbull> hello i have a question.
[15-Sep-2009 07:09:17]\n<adelbull> When i want to test a based ssh zenpack. in\n'datasource', 'test against device' the zenoss server search the\ncommand on the server side and not on the client side.
\n[15-Sep-2009 07:10:21] <adelbull> it is a problem, because a\nzenoss server installed on a linux distibution doesnt have the specific\ncommand of solaris, hpux, aix...
[15-Sep-2009 07:11:57] <skipzoid> anyone know where i can find a list of varibles that can be passed to a event command
[15-Sep-2009 07:22:43] <ckrough> skipzoid: I think that is detailed in the developer guide
\n[15-Sep-2009 07:33:37] <skipzoid> I'm trying this - sendsms.pl\n"$device" - and when run on the command line a text of $device is\nreceived - when under zenoss event commands - nothing happens if I\nchange the $device so say fred I receive a text 'fred' it seems what is\nbeing passed to the sendsms.pl script is null so nothing is sent
[15-Sep-2009 07:35:23] <ckrough> does it need to be "here\\deviceId" or something more zopeish?
\n[15-Sep-2009 07:36:59] <skipzoid> i would have hoped it was going\nto be simple like the interpreter runing the command just substituted\nthe various varibles for the text required, perhaps I need to be more\ncreative, but I at least need the event id that triggered the command\nso I can extract the rest of the info I want from the database
[15-Sep-2009 07:37:37] <skipzoid> i'm still new to this so maybe it is simple once i know
[15-Sep-2009 07:40:00] <ckrough> sec
[15-Sep-2009 07:40:59] <ckrough> try "%(device)s"
[15-Sep-2009 07:41:15] <ckrough> %(summary)s
[15-Sep-2009 07:41:16] <ckrough> etc...
[15-Sep-2009 07:42:31] <skipzoid> i get a text message from the system saying %device
[15-Sep-2009 07:43:12] <skipzoid> i just tried it - but i think i knew that was going to happen
[15-Sep-2009 07:46:17] <ckrough> you entered "%(device)s" exactly? including the parens and trailing s?
[15-Sep-2009 07:46:48] <skipzoid> let me just check - i havnt changed it back yet
[15-Sep-2009 07:48:52] <skipzoid> ok - i had missed the trailing s - but exactly I receive - %(device)s
[15-Sep-2009 07:48:59] <ckrough> meh
[15-Sep-2009 07:49:02] <ckrough> worth a shot
[15-Sep-2009 07:49:15] <skipzoid> thats fine - happy to try most things
\n[15-Sep-2009 07:49:24] <ckrough> Not sure what to tell you, hang\naround for a while. there should be some more knowledgable people on in\nthe next couple of hours
[15-Sep-2009 07:50:12]\n<skipzoid> i'm still searching for the answer to this one - i\nthink once solved i'm pretty much there with my solution
[15-Sep-2009 07:52:51] <frozty_sa> hi
\n[15-Sep-2009 07:53:10] <frozty_sa> I've found the threads and\ndocumentation on how to make zenoss collect data using distributed\ncollectors
[15-Sep-2009 07:54:00] <frozty_sa>\ncan someone tell me if there's a way to have multiple zenoss core\ninstallations (these will be in physically different networks) and then\nhave a super-master installation that can control the various master\ninstallations?
[15-Sep-2009 08:26:37] <ckrough> frozty_sa: not at the moment, it's an open feature request though
\n[15-Sep-2009 08:27:18] <ckrough> frozty_sa: there is a global\nevent viewer. I believe it's enterprise only. It only allows you to\nview events, not manage the sites.
[15-Sep-2009 08:27:27] <frozty_sa> mmmm
[15-Sep-2009 08:27:28] <frozty_sa> thanks
[15-Sep-2009 08:27:35] <frozty_sa> any idea on progress of that sort of thing?
\n[15-Sep-2009 08:28:20] <ckrough> nope. but I would recommend you\nsend it to them in feedback so they can know how many people want i
[15-Sep-2009 08:28:23] <ckrough> it
[15-Sep-2009 08:28:55] <frozty_sa> guess so
[15-Sep-2009 08:35:57] <ckrough> mrayzenoss: community site looks good. What backend is that again?
[15-Sep-2009 08:44:20] <mrayzenoss> Jive SBS
[15-Sep-2009 08:45:17] <mrayzenoss> I've started up a thread to capture complaints, the site is a work in progress
\n[15-Sep-2009 08:52:51] <Troubadix09> mrayzenoss: I agree with\nckrough, nice new community, but I think it would be better that on the\nTab "All Content" the Thread-Subject is the first column ...
[15-Sep-2009 08:53:24] <mrayzenoss> yeah, I've got a ticket with Jive about that
[15-Sep-2009 08:53:41] <mrayzenoss> that's why I made the forums switch over to the Discussions tab where possible
[15-Sep-2009 08:53:56] <mrayzenoss> there's no "Threaded View" widget
[15-Sep-2009 08:53:57] <Troubadix09> mrayzenoss: because i click several times on the Author ;-)
\n[15-Sep-2009 08:57:42] <Troubadix09> mrayzenoss: I search the\nForum for a Post about Zenpack-install and where directed to an old\nPost...
[15-Sep-2009 08:58:42] <Troubadix09>\nmrayzenoss: in this Post there where a Link to another Thread, that\nwould'nt reach the Thread
[15-Sep-2009 08:59:44]\n<mrayzenoss> yeah... we mapped all the old wiki content and\npages, but we couldn't preserve the links from threads to previous\nthreads
[15-Sep-2009 09:00:13] <mrayzenoss> I'll see if there's a way to fix that
[15-Sep-2009 09:00:28] <mrayzenoss> what's your issue?
[15-Sep-2009 09:01:19] <Troubadix09> I post it in the Forum under ZenPack, I have trouble to install the WMI-DataSource-Zenpack
[15-Sep-2009 09:02:40] <mrayzenoss> responding there
[15-Sep-2009 09:02:49] <venturaville> mrayzenoss: why is the zenoss community portal spamming me with notes about thread messages?
[15-Sep-2009 09:02:50] <mrayzenoss> I believe your .egg is actually a .zip or something
[15-Sep-2009 09:03:07] <mrayzenoss> venturaville: were you previously subscribed via email?
[15-Sep-2009 09:03:35] <venturaville> mrayzenoss: subscribed to what precisely?
[15-Sep-2009 09:03:54] <Troubadix09> mrayzenoss: I rename it with ending .egg
[15-Sep-2009 09:04:05] <mrayzenoss> were you using the email bridge to use the old forums?
[15-Sep-2009 09:04:17] <venturaville> not that I am aware of
[15-Sep-2009 09:04:33] <mrayzenoss> are you getting emails based on threads you'd been on before?
[15-Sep-2009 09:04:50] <venturaville> mrayzenoss: no I am getting emails on a lot of threads I know nothing about
[15-Sep-2009 09:05:05]\n<dollarbang> Question about google maps and the enterprise\nversion of Zenoss. According to Google, their API is limited to 15000\nhits a day. How would this affect the Enterprise version of Zenoss.
[15-Sep-2009 09:05:24] <mrayzenoss> dollarbang: we pay Google for a license, no issues
[15-Sep-2009 09:05:37] <mrayzenoss> venturaville: give me an example subject?
[15-Sep-2009 09:06:26] <rmatte> mrayzenoss: just out of curiosity, what does a google maps API license cost?
[15-Sep-2009 09:06:47] <mrayzenoss> not sure, I think I heard $10K or something
[15-Sep-2009 09:06:57] <rmatte> ah, cool
[15-Sep-2009 09:06:57] <mrayzenoss> we cover it for all users, Core and Enterprise
[15-Sep-2009 09:07:25] <rmatte> ah, I was assuming it wasn't covered for Core since you actually have to manually input an API key
[15-Sep-2009 09:07:43] <metalklesk> hi
[15-Sep-2009 09:08:10] <venturaville> mrayzenoss: [zenoss-users] New message: "Re: New Site Issues Thread" [1co8L1-1tA-adH]
[15-Sep-2009 09:09:14] <chudler> ugh, I get those emails also
[15-Sep-2009 09:09:55] <rmatte> I haven't gotten any
[15-Sep-2009 09:09:57] <mrayzenoss> venturaville: with the old site... did you interact with zenoss-users via email?
[15-Sep-2009 09:10:06] <rmatte> The only email I got was someone adding me as a link
[15-Sep-2009 09:10:14] <rmatte> errr, as a connection
[15-Sep-2009 09:10:14] <mrayzenoss> venturaville: or did you post through the website?
[15-Sep-2009 09:10:15] <rmatte> rather
[15-Sep-2009 09:11:16] <mrayzenoss> chudler: that goes for you too, did you use email to respond to the old forums?
[15-Sep-2009 09:11:30] <metalklesk> i have a little tiny small question
\n[15-Sep-2009 09:11:42] <chudler> I used to get threaded email for\nevery forum post. Not sure if I ever responded via email.
[15-Sep-2009 09:11:56] <chudler> seems likely..
\n[15-Sep-2009 09:12:32] <metalklesk> in a production enviroment,\nshould zenoss be in the same network segment as the devices to be\nmonitored ?
[15-Sep-2009 09:13:45] <mrayzenoss>\nchudler: yeah, we carried that over to the new site... but obviously\nthe filters changed
[15-Sep-2009 09:13:46]\n<venturaville> mrayzenoss: I pretty much just used email (mailing\nlist), but never really did anything with the website
[15-Sep-2009 09:13:55] <mrayzenoss> venturaville: yeah
[15-Sep-2009 09:14:10] <venturaville> how do I disable it
[15-Sep-2009 09:14:26] <mrayzenoss> have you both logged into the new site?
[15-Sep-2009 09:14:44] <dollarbang> mrayzenoss: thanks. another reason to go Enterprise.
[15-Sep-2009 09:15:05] <rmatte> mrayzenoss: do I need to create a new account to login to community.zenoss.org?
[15-Sep-2009 09:15:20] <mrayzenoss> no, your existing account is there
[15-Sep-2009 09:15:26] <mrayzenoss> you may need a password reset
[15-Sep-2009 09:15:34] <chudler> I am resetting my pw now
[15-Sep-2009 09:15:37] <rmatte> hmmm, I've just tried logging in with every password that I could think of
[15-Sep-2009 09:15:45] <rmatte> do I need to reset it?
[15-Sep-2009 09:16:39] <venturaville> we are installing enterprise on our production system today...........
[15-Sep-2009 09:16:39] <mrayzenoss> yeah, a reset is required
[15-Sep-2009 09:16:43] <rmatte> k
\n[15-Sep-2009 09:19:49] <skipzoid> simple question (i hope) I've\ncreated a command in /ZenEventManager /Event Commands which is a perl\nscript sendsms.pl it takes the argument of the message and sends it to\na number preconfigured in the script its self - on the command line\nsendsms.pl "test" sends test to the handset and works - i'm trying to\nget it to send event infomation, what varibles are avable to pass to it?
\n[15-Sep-2009 09:20:52] <mrayzenoss> chudler/venturaville go to\nthe forum you're getting emails from (ie.\ncommunity/forums/zenoss-users)
[15-Sep-2009 09:21:01] <mrayzenoss> and select "Stop email notifications"
[15-Sep-2009 09:21:04] <venturaville> found the settings
[15-Sep-2009 09:21:11] <venturaville> I can take it from here thanks :-)
[15-Sep-2009 09:21:15] <mrayzenoss> sorry
[15-Sep-2009 09:24:00] <venturaville> np
[15-Sep-2009 09:26:30] crow is now known as ckrough
[15-Sep-2009 09:26:31] <metalklesk> which data should i be able to monitor from a printer (xerox 3250) ?
[15-Sep-2009 09:26:34] <jb> sup
[15-Sep-2009 09:26:38] <jb> fix the new site yet?
[15-Sep-2009 09:26:58] <mrayzenoss> it's up...
[15-Sep-2009 09:27:18] <jb> did you get my tweet?
\n[15-Sep-2009 09:27:58] <jb> my account doesn't work, and its\ntelling me my email address is not valid when I try to retreive my\npassword..
[15-Sep-2009 09:28:15] <mrayzenoss> jb: what was your old account name?
[15-Sep-2009 09:28:21] <rmatte> facebook seems to be screwed this morning, I can't even load my profile
[15-Sep-2009 09:28:22] <Apocalipse> hello people...
[15-Sep-2009 09:28:28] <jb> mrayzenoss: jbaird
[15-Sep-2009 09:28:38] <jb> facbeook has been screwed for weeks..
[15-Sep-2009 09:28:46] <Apocalipse> i think i found why my devices are not linking...
[15-Sep-2009 09:28:48] <rmatte> jb: oh?
[15-Sep-2009 09:28:53] <rmatte> jb: I hardly go on so hadn't noticed
[15-Sep-2009 09:29:33] <rmatte> I can load it fine on my iPhone though
[15-Sep-2009 09:39:12] <mrayzenoss> ok, we've updated the subjects for the forum emails
[15-Sep-2009 09:39:19] <mrayzenoss> they should be closer to previous emails
[15-Sep-2009 09:39:27] <mrayzenoss> if you were using that feature
[15-Sep-2009 09:40:01] <chudler> that's nice, thanks. It was my only other pipeline into the community :-)
[15-Sep-2009 09:40:09] <chudler> its the body of the emails that is most annoying
[15-Sep-2009 09:40:15] <mrayzenoss> yeah, we cleaned up the template
[15-Sep-2009 09:40:28] <mrayzenoss> so now it should be the body of the message, with a footer about how to respond
[15-Sep-2009 09:40:38] <mrayzenoss> dumped out the author link
[15-Sep-2009 09:41:02] <mrayzenoss> unfortunately the subjects still contain the thread ID
[15-Sep-2009 09:41:11] <mrayzenoss> which we've got a ticket open against, since that breaks threading
[15-Sep-2009 09:50:59] <mrayzenoss> grrr... the RSS feeds are kinda sucky
[15-Sep-2009 09:51:12] <mrayzenoss> for the forums, they put the date of the first post in a thread
\n[15-Sep-2009 09:53:51] <rmatte> In case anyone is interested, if\nyour facebook profile isn't showing up (if you're getting "The page you\nrequested was not found" when trying to access it), from the error page\nyou just need to click where it says English (US), or whatever language\nyou have set, then change the language to something else (English (UK)\nfor instance), then change it back to the original language, and voila,\nit should be fixed.
[15-Sep-2009 09:54:21] <mrayzenoss> what's that?
\n[15-Sep-2009 09:57:45] <npmccallum> mrayzenoss: FYI, emails from\nthe new forum are missing the list-id header, probably breaking mail\nfiltering for a lot of people
[15-Sep-2009 09:59:20]\n<ckrough> what does it mean when I get an email from community\nsaying "xxxx added you as a connection."?
[15-Sep-2009\n09:59:59] <rmatte> ckrough: I got that too, it's a community\ndriven site, so adding someone as a connection is probably similar to\nadding someone in LinkedIn
[15-Sep-2009 10:00:28] <ckrough> I assumed. I looked on the site and I see where I can 'follow' someone, but didnt see a connection
[15-Sep-2009 10:03:07] <rmatte> that's probably what that means
[15-Sep-2009 10:03:48]\n<scp2004> Hi guys .. Please , i need some help .. My zenoss stio\nto generate graphs with no explain ! In event.log i got :
[15-Sep-2009 10:03:50]\n<scp2004> 2009-09-15T11:58:34 INFO ZODB.Conflict database\nconflict error (oid 0x019a, class\nProducts.ZenUtils.PObjectCache.PObjectCache) at\n/zport/RenderServer/render (22 conflicts, of which 0 were unresolved,\nsince startup at Tue Sep 15 09:46:50 2009)
[15-Sep-2009 10:03:54] <scp2004> Any Idea ?
[15-Sep-2009 10:04:17] <ckrough> thats an ok error
[15-Sep-2009 10:04:20] <scp2004> It stop to generate graphs !
[15-Sep-2009 10:04:24] <ckrough> "0 were unresolved"
[15-Sep-2009 10:04:29] <ckrough> unrelated to that error
[15-Sep-2009 10:05:43] <scp2004> In 2009-09-15 my Zenoss stop to generate graphs with no explain ! I dont know what is happening ..
[15-Sep-2009 10:07:07] <scp2004> 2009-09-15T09:25:20 ERROR RenderServer Failed to generate a graph
[15-Sep-2009 10:07:07] <scp2004> Traceback (most recent call last):
[15-Sep-2009 10:07:07] <scp2004> File "/usr/local/zenoss/zenoss/Products/ZenRRD/RenderServer.py", line 190, in render
[15-Sep-2009 10:07:07] <scp2004> rrdtool.graph(*gopts)
[15-Sep-2009 10:07:07] <scp2004> error
[15-Sep-2009 10:07:23] <rmatte> ok, first off, calm down
, secondly, are you sure it has stopped generating all graphs or just specific ones?[15-Sep-2009 10:08:13] <scp2004> [rmatte] Are not generating any graphs
[15-Sep-2009 10:09:41] <scp2004> [rmatte] It was ok. But in 2009-09-10 all graphs stop
[15-Sep-2009 10:13:08] <ckrough> scp2004: are you out of disk space?
[15-Sep-2009 10:13:42] <scp2004> [ckrough] i thin not ..but will see
\n[15-Sep-2009 10:14:26] <skipzoid> simple question (i hope) I've\ncreated a command in /ZenEventManager /Event Commands which is a perl\nscript sendsms.pl it takes the argument of the message and sends it to\na number preconfigured in the script its self - on the command line\nsendsms.pl "test" sends test to the handset and works - i'm trying to\nget it to send event infomation, what varibles are avable to pass to it?
[15-Sep-2009 10:15:38] <scp2004> Not ... I have to much disk free ..
[15-Sep-2009 10:15:56] <scp2004> [ckrough] Im using only 4%
[15-Sep-2009 10:16:12] <ckrough> scp2004: it was a long shot:) Just a first check since all graphs stopped.
[15-Sep-2009 10:16:37] <scp2004> [ckrough] Its checked .. All graphs stoped in 2009-09-10
[15-Sep-2009 10:17:20] <metalklesk> you mustdid something in 2009-09-10
[15-Sep-2009 10:17:23] <ckrough> scp2004: so you get no graphs at all, just a traceback when you go to the graph page?
[15-Sep-2009 10:17:50] <metalklesk> things just don't stop working ... windows does but is an exception
\n[15-Sep-2009 10:18:14] <scp2004> [metalklesk] No i dont put my\nhands in it ! .. It was ok ..And i got no idea what happens
[15-Sep-2009 10:18:45] <ckrough> scp2004: do you get a traceback or empty graphs or graphs that stopped updating?
[15-Sep-2009 10:19:00] <scp2004> [ckrough] Empty graphs
[15-Sep-2009 10:19:12] <frozty_sa> ckrough: still around?
[15-Sep-2009 10:19:30] <scp2004> If i ghange Range to Daily i can see that it sopped in 2009-09-10
[15-Sep-2009 10:19:45] <scp2004> [ckrough] If i change Range to Daily i can see that it sopped in 2009-09-10
[15-Sep-2009 10:20:02] <metalklesk> sorry for ask but, did you get any new data since 2009-09-10 ?
[15-Sep-2009 10:20:09]\n<ckrough> scp2004: ah, ok. Check one of the related RRD files.\n'rrdtool last /opt/zenoss/perf/....' for one of the rrd files related\nto one of the broken graphs
[15-Sep-2009 10:20:54] <scp2004> [ckrough] ok ..
\n[15-Sep-2009 10:20:55] <ckrough> scp2004: that will give you the\ntimestamp of that last update to the rrd file. you can translate it\nhere: http://www.epochconverter.com/
[15-Sep-2009 10:21:09] <ckrough> frozty_sa: yup
\n[15-Sep-2009 10:21:55] <ckrough> scp2004: also, as the zenoss\nuser, do 'zenoss status' and make sure zenperfsnmp is running
[15-Sep-2009 10:22:17] <scp2004> [ckrough] Its running
\n[15-Sep-2009 10:22:37] <frozty_sa> ckrough: I'm seeing some\nreferences to a 'multiple hubs' setup, although the documentation's\ninternal structure seems to have changed since google's last index run\nof it
[15-Sep-2009 10:22:55] <frozty_sa>\nckrough: could you maybe point me to where I can find out more about\nthat, or give me a short summary?
[15-Sep-2009 10:22:55] <ckrough> frozty_sa: link?
\n[15-Sep-2009 10:23:11] <frozty_sa>\nhttp://www.google.co.za/search?q=zenoss+multiple+hubs&ie=utf-8&oe=utf-8&aq=t&rls=com.ubuntu:en-US:unofficial&client=firefox-a
\n[15-Sep-2009 10:24:21] <ckrough> frozty_sa: I think those are in\nrelation to the multiple collector scenario, but not a distributed\n'master>slave>collector' architecture
[15-Sep-2009 10:24:29] <scp2004> [ckrough] i cant see anything in /usr/local/zenoss/perfs/daemons/localhost
[15-Sep-2009 10:25:19] <ckrough> scp2004: they should be under $ZENHOME/perf/Devices
[15-Sep-2009 10:25:54] <ckrough> frozty_sa: docs/DOC-2683
\n[15-Sep-2009 10:26:22] <ckrough> mrayzenoss: ouch, all the google\nlinks are borked with the new urlpath. collateral damage I guess.
[15-Sep-2009 10:26:39] <mrayzenoss> ckrough: google is indexing
[15-Sep-2009 10:26:52] <mrayzenoss> or are you talking about the search.zenoss.com?
[15-Sep-2009 10:27:25] <ckrough> mrayzenoss: nope
[15-Sep-2009 10:27:29] <frozty_sa> google's index still shows old links
[15-Sep-2009 10:27:41] <ckrough> ckrough: was talking about the googles.
[15-Sep-2009 10:27:43] <frozty_sa> so anything you click on leads you to the top-level documentation index
[15-Sep-2009 10:27:59] <mrayzenoss> well, we've added redirects for most of the static content, just not the forums
[15-Sep-2009 10:28:01] <ckrough> ha, I just talked to myself. time for more coffee
\n[15-Sep-2009 10:29:43] <ckrough> anyone know if there is a way to\nsee modifications for deleted devices? someone deleted a device and I\nneed to see it's change history
[15-Sep-2009 10:30:06] <frozty_sa> ckrough: thanks, that looks like it'll do what I want it
\n[15-Sep-2009 10:30:41] <ckrough> frozty_sa: are you enterprise?\nnote that it doesnt manage remote sites... just brings together their\nevent consoles
[15-Sep-2009 10:30:52] <ckrough> frozty_sa: and its read-only. no acking
\n[15-Sep-2009 10:32:13] <ckrough> awesome, you can go through your\nZ2 log and look a the methods in the urls to see what the heck the user\ndid to break it
[15-Sep-2009 10:32:30] <frozty_sa> ckrough: not enterprise, no
\n[15-Sep-2009 10:32:47] <frozty_sa> ckrough: currently I've got a\nzabbix deployment that's supposed to do this, but it's not
[15-Sep-2009 10:32:53] <ckrough> frozty_sa: I think that zenpack only works with ent. installs
[15-Sep-2009 10:33:29] <frozty_sa> but zabbix is a horrendous piece of shit, so I'm looking at what I can move to
\n[15-Sep-2009 10:44:58] <gorkadel> hello, i try to create a new\ntemplate via the but when i click on "add datasource" the loading\nscreen appears and never stop ...
[15-Sep-2009 10:45:25] <gorkadel> is it a bug ?
[15-Sep-2009 10:45:37] <gorkadel> "via the device class"
[15-Sep-2009 10:46:29] <mrayzenoss> gorkadel: which browser?
[15-Sep-2009 10:46:52] <gorkadel> safari
[15-Sep-2009 10:47:00] <mrayzenoss> gorkadel: version?
[15-Sep-2009 10:47:36] <gorkadel> 4.0.3
[15-Sep-2009 10:47:50] <mrayzenoss> sorry, Zenoss version?
[15-Sep-2009 10:48:04] <gorkadel> zenoss 2.4.5
[15-Sep-2009 10:49:15] <gorkadel> same issue with firefox.
[15-Sep-2009 10:49:46] <gorkadel> "loading..." for ever ...
[15-Sep-2009 10:50:51] <mrayzenoss> gorkadel: any customizations?
[15-Sep-2009 10:50:55] <mrayzenoss> or ZenPacks
[15-Sep-2009 10:51:13] <gorkadel> yes i added opensolaris zenpacks
[15-Sep-2009 10:52:08] <mrayzenoss> has zope been restarted since then?
[15-Sep-2009 10:52:14] <mrayzenoss> 'zopectl restart'
[15-Sep-2009 10:54:34] <Apocalipse> hello?
[15-Sep-2009 10:54:41] <cgibbons> ¿que?
[15-Sep-2009 10:55:29] <Troubadix09> cu ... bye
[15-Sep-2009 10:55:43] <Apocalipse> ¿como estas?
[15-Sep-2009 10:55:52] <mrayzenoss> gotta go do a demo, PM me if you have any zenoss.org issues
[15-Sep-2009 10:56:08] <gorkadel> mrayzenoss: ok thanks
[15-Sep-2009 10:56:21] <gorkadel> mrayzenoss: ok try your suggestion.
[15-Sep-2009 10:57:51] <gorkadel> it is working
[15-Sep-2009 10:57:58] <gorkadel> thqnks you.
[15-Sep-2009 11:02:10]\n<markeriv> Hey guys, just a quick question for you. Trying to\nhunt down the admin guide that deals extensively with users and\npermissions
[15-Sep-2009 11:03:46] <markeriv> did you guys change your docs area recently?
[15-Sep-2009 11:03:46] <markeriv> hmm
\n[15-Sep-2009 11:26:16] <chemist> what type of checks does zenoss\ndo when you ask it to monitor a process? does it just check that it\nexists?
[15-Sep-2009 11:28:36] <chudler> it checks it exists, the cpu and mem usage, IIRC
[15-Sep-2009 11:29:14] <chudler> also counts how many processes match the name pattern
\n[15-Sep-2009 11:34:13] <scp2004> [mrayzenoss] Hi .. My Zenoss\nrunning on CentOS 5 and a HP server Proliant DL 160 are very very slow\n! Since i installed it .. Im monitoring 385 Devices.. Its normal ?
[15-Sep-2009 11:34:32] <markeriv> how much ram you got?
[15-Sep-2009 11:34:39] <markeriv> what the stats on your core?
[15-Sep-2009 11:35:19] <markeriv> also, what type of internet connection do you have and how much bandwidth?
[15-Sep-2009 11:43:28] <rmatte> also, is it specific pages that are slow, or everything?
[15-Sep-2009 11:48:10] <scp2004> I have 2GB RAM , 200 MB bandwidth , and using 12% of core
[15-Sep-2009 11:48:28] <scp2004> Especialy when i try access devices
[15-Sep-2009 11:48:53] <scp2004> When i click on a device to see status, os ..etc
[15-Sep-2009 11:49:05] <scp2004> it spend like 5 - 10 min
[15-Sep-2009 11:49:10] <markeriv> might want more ram
[15-Sep-2009 11:49:10] <scp2004> to open
[15-Sep-2009 11:49:19] <markeriv> 2 gigs is fairly low
[15-Sep-2009 11:49:20] <scp2004> Yep ..Maybe ..
[15-Sep-2009 11:49:25] <markeriv> my server is running 8
\n[15-Sep-2009 11:49:49] <scp2004> Guys ..About my graphs are\nstopped ..Look this message : localhost zenperfsnmp heartbeat failure
[15-Sep-2009 11:49:56] <scp2004> in my events ..
[15-Sep-2009 11:50:12] <markeriv> tried a restart of zenperfsnmp?
[15-Sep-2009 11:50:16] <scp2004> Its probably the cause ?
[15-Sep-2009 11:50:23] <scp2004> i will now
[15-Sep-2009 11:50:46] <markeriv> whats your current memory usage?
[15-Sep-2009 11:51:11] <scp2004> [markeriv] 99% ..
[15-Sep-2009 11:51:16] <markeriv> ahh
[15-Sep-2009 11:51:16] <scp2004> Daemon: zenperfsnmp program running; pid=3397
[15-Sep-2009 11:51:18] <markeriv> ram is your problem
[15-Sep-2009 11:51:25] <markeriv> you need more
[15-Sep-2009 11:52:36] <scp2004> It doesn´t stop ..zenperfsnmp ..
[15-Sep-2009 11:53:02] <scp2004> in /usr/local/zenoss/bin the command ./zenperfsnmp stop
[15-Sep-2009 11:53:21] <chudler> please post the output of the "free" command
[15-Sep-2009 11:55:38] <ckrough> scp2004: Linux will always use 99% ram, check your swap usage
[15-Sep-2009 11:56:03] <ckrough> scp2004: or check the second line of 'free' to see what's being used by buffering vs everything else
[15-Sep-2009 11:57:39] <scp2004> [ckrough] its using like 30% of swap
[15-Sep-2009 11:58:09] <scp2004> about graphs .. I ll stop and Start Zenoss-stack again
[15-Sep-2009 11:58:48] <ckrough> scp2004: yeah, you need ram
[15-Sep-2009 11:59:02] <ckrough> scp2004: swap usage = evil, especially for zenperfsnmp
[15-Sep-2009 12:02:49] <scp2004> ahhh ...ok ...thanks very much
[15-Sep-2009 12:06:15]\n<chudler> if you want to favor keeping pages mapped use\n/proc/sys/vm/swappiness, but it could increase load AND memory pressure.
[15-Sep-2009 12:07:33] <chudler> lower number means less apt to swap pages to disk, presumably at the expense of the cache.
\n[15-Sep-2009 13:32:56] <skipzoid> simple question (i hope) I've\ncreated a command in /ZenEventManager /Event Commands which is a perl\nscript sendsms.pl it takes the argument of the message and sends it to\na number preconfigured in the script its self - on the command line\nsendsms.pl "test" sends test to the handset and works - i'm trying to\nget it to send event infomation, what varibles are avable to pass to it?
[15-Sep-2009 13:45:57] <ckrough> ian: thanks, sloppy troubleshooting on my part
[15-Sep-2009 13:46:03]\n<skipzoid> simple question (i hope) I've created a command in\n/ZenEventManager /Event Commands which is a perl script sendsms.pl it\ntakes the argument of the message and sends it to a number\npreconfigured in the script its self - on the command line sendsms.pl\n"test" sends test to the handset and works - i'm trying to get it to\nsend event infomation, what varibles are avable to pass to it?
[15-Sep-2009 13:47:14] <djack> hello world belgium calling ;-)
[15-Sep-2009 13:48:37] <skipzoid> hello from the UK 8^)
[15-Sep-2009 13:49:06] <markeriv> Here is a question for you, I am trying to create an event view that is limited to a specific group.
[15-Sep-2009 13:49:31] <markeriv> So that I can create an alert that only affects a specific group
[15-Sep-2009 13:50:48] <mrayzenoss> skipzoid: here's rmatte's post from yesterday:
[15-Sep-2009 13:50:51] <mrayzenoss> 11:09:29 AM rmatte: Here's an example of a command we use to create tickets in our ticketing system:
[15-Sep-2009 13:50:51] <mrayzenoss> 11:09:48\nAM rmatte: /mspfiles/scripts/create_ticket.pl -autonotify -customer\n'gen-00025' -device '${dev/id}' -deviceIP '${dev/manageIp}' -collector\n`hostname -f` -first '${evt/firstTime}' -last '${evt/lastTime}' -count\n'${evt/count}' -summary '${evt/summary}' -noteTitle 'System Monitor\nError' -note '${evt/message}' -severity '${evt/severity}' -group\n'${evt/DeviceGroups}' -impact '${evt/DevicePriority}' -component\n'${evt/component}' -que
[15-Sep-2009 13:51:20] <skipzoid> yeah i was experimenting with that but with no luck
[15-Sep-2009 13:56:55] <skipzoid> got it - stupid mistake - mixed up my ' ` and " Grrrr !!
[15-Sep-2009 13:57:20] <rmatte> lol
[15-Sep-2009 14:55:04] <djack> hey you all can you use xml-rpc or rest to pull out data from zenoss ?
\n[15-Sep-2009 14:55:57] <rmatte> djack: actually, someone here\nasked me that today, so I'd be interested in knowing as well. I've\nseen xml-rpc calls to Zenoss before
[15-Sep-2009 14:57:25] <djack> rmatte i'm still looking for a way to create a easy external status page ....
[15-Sep-2009 14:57:37] <ckrough> djack: you can pull lots of data from xmlrpc
[15-Sep-2009 14:58:14] <chudler> I think you'll want to check out the dev guide, especially chapter 3
[15-Sep-2009 14:58:18] <djack> ckrough is it usable for external status pagees usage
[15-Sep-2009 14:58:35] <ckrough> djack: kinda depends on what the external status page is
[15-Sep-2009 14:58:40] <ckrough> it just gives you raw data
[15-Sep-2009 14:58:47] <chudler> docs/DOC-3136
[15-Sep-2009 14:59:01] <djack> raw data is good
\n[15-Sep-2009 14:59:21] <ckrough> what chudler said, plus browse\nthrough the API for whatever version you use.\nhttp://www.zenoss.com/community/docs/zenoss-api-docs/2.3/
[15-Sep-2009 14:59:36] <ckrough> djack: there's a good bit of detective work involved
[15-Sep-2009 15:01:32] <djack> ckrough i've seen php an that page ;-)
[15-Sep-2009 15:01:34] <ckrough> djack: you can do things like getRRDValue(ccmHistoryRunningLastSaved), getManageIp(), etc...
[15-Sep-2009 15:01:39] <djack> that's more my knowledge
[15-Sep-2009 15:03:58]\n<djack> what i basicly want is server and green orange or red in\nleft side of the page and the events in the right side of the screen ;-)
[15-Sep-2009 15:04:29] <djack> if i can set that up then I'm happy and my salary can go up :-p
[15-Sep-2009 15:05:30] <djack> ckrough i've seen but I need to see some examples before i get it ...
[15-Sep-2009 15:09:33] <djack> and all the examples are about pushing data to zenoss
[15-Sep-2009 15:09:49] <rmatte> yeh, from what I've seen
[15-Sep-2009 15:09:56] <rmatte> like calling on Zenoss to move events to history and such
[15-Sep-2009 15:11:31] <djack> i dont want to do that just query device status outside zenoss platform
[15-Sep-2009 15:11:51] <chudler> the first example on the URL I previously posted includes query for data, does it not?
[15-Sep-2009 15:11:56] <chudler> http://USERNAME:PASSWORD@MY_ZENOSS_HOST:8080/zport/dmd/Devices/Server/Linux/devices/angel/getRRDValue?dsname=laLoadInt5_laLoadInt5
[15-Sep-2009 15:12:25] <djack> chudler also all then :-p
[15-Sep-2009 15:12:52] <rmatte> djack: so you want up or down status basically?
[15-Sep-2009 15:15:00] <rmatte> Here's a python script that I wrote to retrieve status on devices, along with an example of it running...
[15-Sep-2009 15:15:02] <rmatte> http://pastebin.com/m515c3de3
[15-Sep-2009 15:15:23] <rmatte> it pulls device name, IP, status (UP/DOWN), and uptime
[15-Sep-2009 15:17:58] <djack> hmmm rmatte thanks
[15-Sep-2009 15:18:51] <rmatte> np
\n[15-Sep-2009 15:19:23] <rmatte> that's really only for ping\nstatus, but that's generally all you'd care about for basic device\nstatus, then the event console would be for any other issues
[15-Sep-2009 15:19:47] <rmatte> the event console shouldn't be too hard since the events are stored in MySQL
\n[15-Sep-2009 15:20:27] <rmatte> so if you're doing it in php you\njust need to tap in to the database, pull the rights values out, and\ndisplay them
[15-Sep-2009 15:20:36] <rmatte> you'll probably have to convert some of them too
[15-Sep-2009 15:21:05] <djack> hmmm
[15-Sep-2009 15:21:06] <djack> nice
\n[15-Sep-2009 15:21:26] <rmatte> This may be of use to you:\nhttp://dev.zenoss.org/trac/browser/branches/zenoss-2.4.x/Products/ZenEvents/db/zenevents.sql
[15-Sep-2009 15:21:51] <ckrough> If your doing it remotely you'll need to use REST calls or xmlrpc
[15-Sep-2009 15:21:59] <ckrough> xmlrpc is the preferred
\n[15-Sep-2009 15:22:38] <rmatte> ckrough: well, either that or do\nsome sort of setup where he has the php interface call on scripts\nremotely via ssh, and pass the data to it
[15-Sep-2009 15:22:42] <rmatte> but that's a bit patchwork
[15-Sep-2009 15:22:56] <rmatte> still, if it works, it works
[15-Sep-2009 15:23:04] <ckrough> yeah, reinventing the wheel at that point
\n[15-Sep-2009 15:23:42] <rmatte> yeah, but if you're like me, and\nyou're not very familliar with rest of xmlrpc, then it might be the\nsimpler solution
[15-Sep-2009 15:23:56] <rmatte> rest or* xmlrpc rather
[15-Sep-2009 15:24:38] <djack> ckrough beside the getRRDthingy is there a reference page with all calls for rest
[15-Sep-2009 15:25:02]\n<chudler> everyone knows about REST, since it is the way the web\nworks, at least in theory. xmlrpc has some learning curve, but it is\ndefinitely possible to use it without knowing anything abou tit
[15-Sep-2009 15:25:06] <ckrough> unfortunately no, I usuall browse through the API doc trying different methods. some work, some dont
[15-Sep-2009 15:25:33] <djack> well the page will be on hosted on the same machine as the zenoss
[15-Sep-2009 15:25:47] <chudler> djack: really? use zendmd!
[15-Sep-2009 15:25:48] <djack> i have memory and cpu power to spend
[15-Sep-2009 15:26:02] <djack> chudler ye[
[15-Sep-2009 15:26:02] <ckrough> then straight up python and zendmd is nice and direct
[15-Sep-2009 15:26:31] <djack> ckrough mail me the python for dummies book then :-p
[15-Sep-2009 15:26:45] <djack> however maybe it's nice to learn a new language
[15-Sep-2009 15:27:02] <rmatte> well, that script that I pasted, is pretty much what you'd use, and then you'd access mysql directly
[15-Sep-2009 15:27:09] <ckrough> djack: www.diveintopython.org
[15-Sep-2009 15:27:20] <ckrough> http://diveintopython.org/toc/index.html
\n[15-Sep-2009 15:27:21] <chudler> yeah, maybe. one of python's\nmain points is to be very easy to read and learn. Despite that, it is\nstill a great overall language
[15-Sep-2009 15:28:07] <ckrough> argh.. all these great zenoss links are broken in teh googles
[15-Sep-2009 15:28:10] <djack> chudler that is one thing that is correct the script of rmatte i can read and understand
[15-Sep-2009 15:28:27] <djack> ckrough yeah somebody needed to update the webpage :-p
[15-Sep-2009 15:28:29] <mrayzenoss> ckrough: google will catch up soon I hope
[15-Sep-2009 15:28:37] <ckrough> I know. just griping.
[15-Sep-2009 15:28:52] <ckrough> I keep finding what djack needs then having to play detective to get the link.
[15-Sep-2009 15:29:02] <rmatte> djack: yeh, it's a really basic script though, easy to work with
\n[15-Sep-2009 15:29:19] <djack> we had the same shit when we\nmigrated our website .... not nice if you are a webshop that sells cd\ndvd and games
[15-Sep-2009 15:29:22] <rmatte>\ndjack: If you saw the availability report script that I finished\nwriting yesterday you might cry (I almost did)
[15-Sep-2009 15:29:54] <djack> rmatte is it usefull to share
[15-Sep-2009 15:30:04] <djack> who know i can learn from it ;-)
[15-Sep-2009 15:30:31] <ckrough> djack: docs/DOC-3136
[15-Sep-2009 15:30:51] <ckrough> mrayzenoss: I think this room needs a bot that snarfs those urls and posts the title. ?
[15-Sep-2009 15:31:10] <mrayzenoss> there was one... I don't know what happened to it
[15-Sep-2009 15:31:23] * ckrough pours a drink for dead bot
[15-Sep-2009 15:31:56] <ckrough> djack: #3.1.2.3.3. PHP on that last link I sent
[15-Sep-2009 15:32:28] <ckrough> same methodology works for getting data back out
\n[15-Sep-2009 15:33:55] <djack> ckrough seen it and also seen how\ni can call python script from php so i will figure it out
[15-Sep-2009 15:34:12] <djack> hmmm wife does not going to like this new language :-p
[15-Sep-2009 15:34:22] <ckrough> ha
[15-Sep-2009 15:34:42] <chudler> LOL, yeah same here. Every time I bring home a new book, I have to hear about it..
\n[15-Sep-2009 15:34:53] <ckrough> I resisted python for a while.\nPreferred PHP for the most part, but now Im using python for most stuff
[15-Sep-2009 15:35:22] <djack> chudler the list is becomming long ... need to pass the new mcse and vmware
[15-Sep-2009 15:35:37] <ckrough> ..the never ending "to read" list.
[15-Sep-2009 15:36:15] <djack> idd
[15-Sep-2009 15:36:29] <djack> ckrough what is the advantage ...?
[15-Sep-2009 15:36:39] <ckrough> of?
[15-Sep-2009 15:37:00] <djack> try to pull me over to the dark side of python
[15-Sep-2009 15:37:21] <djack> maybe the force is with me
\n[15-Sep-2009 15:37:33] <ckrough> oh. it's a smart language. hard\nto put my finger on it. whenever I cant do something I find that I was\ntrying it wrong and that there is some elegant way to do it
[15-Sep-2009 15:37:50] <ckrough> and it forces you to make clean code
[15-Sep-2009 15:38:06] <ckrough> you'll spend a month trying not to use ';' tho.
[15-Sep-2009 15:38:10] <djack> ckrough not one code is made to be clean ...
[15-Sep-2009 15:38:28] <djack> haha yeah i have that with bash also ....
[15-Sep-2009 15:39:10] <djack> well i learned php in 4 months so python will work also ....
[15-Sep-2009 15:46:48] <djack> hey on the forum i lots of strange codes i staid of code how come ?
[15-Sep-2009 15:50:39] <mrayzenoss> the code got munged, we've got a ticket open
[15-Sep-2009 15:50:45] <mrayzenoss> to get them fixed
[15-Sep-2009 15:51:27] <djack> oki
[15-Sep-2009 15:51:37] <djack> going to bed
[15-Sep-2009 16:11:11] <mrayzenoss> ckrough: I've found a work-around for what you were reporting
[15-Sep-2009 16:11:29] <mrayzenoss> paste the URL into google, it'll give you the link and the topic
[15-Sep-2009 16:11:38] <mrayzenoss> copy and paste the topic into the new forums search
[15-Sep-2009 16:11:43] <mrayzenoss> profit
[15-Sep-2009 16:59:04] <rmatte> well, I'm out of here
[15-Sep-2009 16:59:06] <rmatte> later
[15-Sep-2009 17:22:05] <cgibbons> hrm
[15-Sep-2009 17:37:27] <BillBinko> hi all
[15-Sep-2009 17:37:45] <BillBinko> anyone know how to get two snmp agents on the same box modelled as different devices?
[15-Sep-2009 17:38:06] <BillBinko> they are on separate ports
[15-Sep-2009 17:45:36] <markeriv> hmm
[15-Sep-2009 17:45:44] <markeriv> when you say seperate ports
[15-Sep-2009 17:45:48] <markeriv> do you mean seperate IPs?
[15-Sep-2009 17:45:53] <markeriv> or just different port #s
\n[15-Sep-2009 17:46:29] <markeriv> cause if you are trying to pull\nstats on two different NICs, that should not be too difficult
[15-Sep-2009 18:03:25] <BillBinko> sorry was away for a second
[15-Sep-2009 18:03:31] <BillBinko> I mean separate UDP ports
[15-Sep-2009 18:03:37] <BillBinko> same nic
[15-Sep-2009 18:03:46] <BillBinko> I want to hit the system SNMP agent (port 161)
[15-Sep-2009 18:03:54] <BillBinko> and the JVM's SNMP agent (port 1161)
[15-Sep-2009 18:03:59] <BillBinko> both on the same IP
\n[15-Sep-2009 18:20:45] <BillBinko> I have to head to dinner -- if\nanyone has a simple answer, could they email me at my name at gmail?
[15-Sep-2009 18:20:50] <BillBinko> I will check back tomorrow
[15-Sep-2009 19:08:05] <scp2004> Hi Guys !
\n[15-Sep-2009 19:16:23] <scp2004> I need add a user in zenoss ,\nthat only can see the devices i configure too he see... Its possible ?
[15-Sep-2009 19:41:47] <ironpaw> i
[15-Sep-2009 19:41:49] <ironpaw> hi
[15-Sep-2009 19:42:12] <ironpaw> you could try defining the administrative servers for the usre
[15-Sep-2009 19:42:26] <ironpaw> nto sure if that would restrict visibility your after thougjh
\n[15-Sep-2009 19:54:25] <scp2004> I need add a user in zenoss ,\nthat only can see the devices i configure too he see... Its possible ?
[15-Sep-2009 19:56:32] <mrayzenoss> that's an Enterprise feature, you can create a read-only user with Zenoss Core
[15-Sep-2009 19:56:57] <mrayzenoss> you can look into the way Zope permissions work to do it by hand, but it's not pretty
\n[15-Sep-2009 19:57:50] <mrayzenoss> Jane Curry has a paper\n"Crafting Zenoss core users for events and zProperties" that may help\nhttp://www.skills-1st.co.uk/papers/jcurry.html
[15-Sep-2009 20:01:14] <scp2004> [mrayzenoss] one mor time ..Thank you Very much ..
[15-Sep-2009 22:40:33] <heckj> evening
[16-Sep-2009 00:00:46] [disconnected at Wed Sep 16 00:00:46 2009]
[16-Sep-2009 00:00:46] [connected at Wed Sep 16 00:00:46 2009]
[16-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
\n[16-Sep-2009 02:54:35] <frozty_sa> uhm, slight nitpickery, but\ndocs/DOC-2988 should be\ns/Distributed/Domain/, surely?
[16-Sep-2009 03:21:56] <salax> hai
[16-Sep-2009 03:21:58] <salax> hello
[16-Sep-2009 03:22:28] <salax> i've installed zenoss 64bit on ubuntu x64 server 8.04
[16-Sep-2009 03:22:56] <salax> i have 2 servers, 1 for vmware server, 1 for zenoss core
[16-Sep-2009 03:23:16] <salax> on top of the vmware server, i install ubuntu,freebsd and openbsd agent
[16-Sep-2009 03:23:42] <salax> everything is ok for now except for running process and list of software
[16-Sep-2009 03:24:05] <salax> on ubuntu, i manage to collect the datat thru snmp and ssh
[16-Sep-2009 03:24:19] <salax> but for the BSDs, i cant connect thru ssh
[16-Sep-2009 03:24:55] <salax> my error stated " No command plugins found for freebsd"
[16-Sep-2009 03:25:06] <salax> and also for openbsd
[16-Sep-2009 03:25:23] <salax> how can i activate or install the cmd plugin?
[16-Sep-2009 03:26:06] <salax> any idea guys? much appreciated, tq
[16-Sep-2009 06:26:55] <Apocalipse> hello people
[16-Sep-2009 06:27:16] <Apocalipse> i am still having problems with linking devices in diff locations...
[16-Sep-2009 06:27:42] <Apocalipse> does anybody knows how to static link then?
[16-Sep-2009 06:33:22] <fred_> hello
[16-Sep-2009 06:33:40] fred_ is now known as Guest84119
[16-Sep-2009 06:33:58] Guest84119 is now known as jo-go
[16-Sep-2009 06:34:05] <jo-go> anyone here ?
[16-Sep-2009 06:34:14] <Troubadix09> hello jo-go
\n[16-Sep-2009 06:34:40] <jo-go> hi,just demonstrating IRC to\nsomeone and proving that real people live here - thanks for the reply
[16-Sep-2009 06:35:29] <Troubadix09> jo-go: is this a zenoss demonstrating?
\n[16-Sep-2009 06:35:51] <jo-go> yeah, just showing off zenoss and\nshowing where to find the helpful people should a question need to be\nasked
[16-Sep-2009 06:36:13] <Apocalipse> hun?
[16-Sep-2009 06:36:51] <jo-go> off now - thanks for helping
[16-Sep-2009 06:37:00] <Troubadix09> jo-go: bye
[16-Sep-2009 06:37:10] <Apocalipse> gone
[16-Sep-2009 06:37:11] <Apocalipse>
[16-Sep-2009 06:38:05] <Troubadix09> Apocalipse: can you explain your problem a little bit more
[16-Sep-2009 06:38:07] <Troubadix09> ?
[16-Sep-2009 06:38:12] <Apocalipse> ok
[16-Sep-2009 06:38:35] <Apocalipse> i have some locations in my network
[16-Sep-2009 06:39:01] <Apocalipse> i added 3 devices to interconnect then
[16-Sep-2009 06:39:07] <Apocalipse> 3 cisco routers
[16-Sep-2009 06:39:30] <Apocalipse> 1 is core router....
[16-Sep-2009 06:40:01] <Apocalipse> the connection between then is using ATM link
[16-Sep-2009 06:40:54] <Apocalipse> i modelled then
[16-Sep-2009 06:41:41] <Apocalipse> it seen ok... i am collecting data... men... cpu... link state... link capcity... etc
[16-Sep-2009 06:41:42] <Apocalipse> but...
[16-Sep-2009 06:41:52] <Apocalipse> zenoss cannot link then...
[16-Sep-2009 06:42:01] <Apocalipse> for example...
[16-Sep-2009 06:42:11] <Apocalipse> let me try to illustrate...
[16-Sep-2009 06:43:33] <Apocalipse> location 1: core router, lan -> x.x.x.x/20 , atm net -> y.y.y.y/30
[16-Sep-2009 06:44:12] <Apocalipse> location 2: border router , lan -> z.z.z.z/20 , atm net -> y.y.y.a/30
[16-Sep-2009 06:45:01] <Apocalipse> both atm interfaces are in same network...
[16-Sep-2009 06:45:25] <Apocalipse> got it?
[16-Sep-2009 06:52:24] <Apocalipse> hey people...
[16-Sep-2009 06:52:40] <Apocalipse> what happens with zenoss website?
[16-Sep-2009 07:32:05] jabular_ is now known as jabular
[16-Sep-2009 08:07:09] <Troubadix09> bye all
\n[16-Sep-2009 09:18:16] <Brad_K> I'm still trying to figure out\nwhy the Nagios-based plugin check_email_delivery isn't storing any data\nin the rrd database. is there anyone here who might be able to help\nanswer some questions?
[16-Sep-2009 09:19:18]\n<tyler79> is there a way to embed a portlet in another page, say\non a custom company intranet site?
[16-Sep-2009\n09:27:42] <Brad_K> I'm still trying to figure out why the\nNagios-based plugin check_email_delivery isn't storing any data in the\nrrd database. is there anyone here who might be able to help answer\nsome questions?
[16-Sep-2009 09:28:32] <mrayzenoss> Brad_K: is it returning multi-line data?
[16-Sep-2009 09:28:46] <mrayzenoss> that fix just went into 2.5 last week
[16-Sep-2009 09:28:46] <rmatte> Brad_K: yeh, what does the output of the script look like?
[16-Sep-2009 09:28:47] <Brad_K> no
[16-Sep-2009 09:28:57] <Brad_K> hang on...
\n[16-Sep-2009 09:30:58] <Brad_K> When funcioning normally,\naccording to the zencommand.log file, the only ouput it generates looks\nlike this:
[16-Sep-2009 09:31:01] <Brad_K> 2009-09-16 09:22:15 DEBUG zen.zencommand: Output: 'DELIVERY OK|ELAPSED=7 DELAY=7
[16-Sep-2009 09:31:04] <Brad_K> \\n'
[16-Sep-2009 09:31:35] <rmatte> Brad_K: not sure why that \\n' is there at the end
[16-Sep-2009 09:31:39] <rmatte> it shouldn't be
[16-Sep-2009 09:31:57] <rmatte> that constitutes multiline
[16-Sep-2009 09:31:58] <Brad_K> weird line-wrap from the cut-n-paste from the zencommand.log
[16-Sep-2009 09:32:17] <rmatte> ah ok, but if you run a test on the datasource, do you see that?
[16-Sep-2009 09:32:35] <rmatte> the entire output should just be: DELIVERY OK|ELAPSED=7 DELAY=7
[16-Sep-2009 09:32:54] <Brad_K> No "\\n" at the end of the line?
\n[16-Sep-2009 09:32:58] <rmatte> and you need to have 2 datapoints\nunder the datasource called ELAPSED and DELAY (case sensitive)
[16-Sep-2009 09:33:07] <rmatte> no, you don't need the \\n at the end
[16-Sep-2009 09:33:22] <Brad_K> okay, i'll need to go fix the script.
[16-Sep-2009 09:33:27] <rmatte> Zenoss most likely won't interpret the data properly because of it
[16-Sep-2009 09:33:44] <rmatte> should be a quick change to fix that
[16-Sep-2009 09:36:23] <Brad_K> Sorry, i've got to go to a quick meeting. bbiab.
[16-Sep-2009 09:36:34] <Brad_K> it's related to this very subject, btw. ;-)
[16-Sep-2009 09:52:08] <twm1010> crickets chirping
[16-Sep-2009 09:52:20] <rmatte> tumbleweed
[16-Sep-2009 09:53:30] <mrayzenoss> fine, you want something to talk about...
[16-Sep-2009 09:53:35] <mrayzenoss> beta 2 is published
[16-Sep-2009 09:53:43] <mrayzenoss> everything except VMware
[16-Sep-2009 09:54:17] <rmatte> stack?
[16-Sep-2009 09:54:41] <rmatte> ooh, it is
[16-Sep-2009 09:54:42] <rmatte> sic
[16-Sep-2009 09:55:15] <rmatte> time to load up my test box and get cracking
[16-Sep-2009 09:57:40] <mrayzenoss> everyone coming to Jane's webinar today?
[16-Sep-2009 09:57:52] <ckrough> yup
[16-Sep-2009 09:58:07] <mrayzenoss> 40+ attendees, we may do an encore presentation
[16-Sep-2009 09:58:59] <ckrough> guess Ill go try and make webex work with linux now...
[16-Sep-2009 09:59:11] <mrayzenoss> I hope it works
[16-Sep-2009 09:59:15] <mrayzenoss> Jane's on SUSE
[16-Sep-2009 09:59:40] <ckrough> its worked as a viewer in the past with minimal issues
\n[16-Sep-2009 09:59:56] <ckrough> couldnt get it working as a\npresenter last time I tried, but I had a windows box handy so I didnt\ntry hard
[16-Sep-2009 10:01:11] <rmatte> It's fun pulling down Zenoss at over 8MB/s
[16-Sep-2009 10:01:33] <rmatte> Sourceforge have some nice mirrors
[16-Sep-2009 10:01:46] <mrayzenoss> yeah they do
[16-Sep-2009 10:01:47] <ckrough> it was rough the other day, I was in low low KBs
[16-Sep-2009 10:02:10] <jb> oops
[16-Sep-2009 10:02:12] <jb> i erased my email
[16-Sep-2009 10:02:20] <jb> mray, can you post me the link to the webinar?
[16-Sep-2009 10:02:41] * rmatte thinks jb should add it to his calendar next time
[16-Sep-2009 10:02:50] <mrayzenoss> which is one of the reasons they seem reluctant to add BitTorrent
[16-Sep-2009 10:03:06] <mrayzenoss> jb: Zenoss Community Webinar this Wednesday: Zenoss Event Architecture: http://tinyurl.com/myesxm
[16-Sep-2009 10:03:14] <mrayzenoss> it's on the blog post
[16-Sep-2009 10:03:23] * rmatte is installing King Crab, can't wait to see this thing
[16-Sep-2009 10:03:40] <jb> hmm
[16-Sep-2009 10:03:47] <jb> oh its not for another hour
[16-Sep-2009 10:03:55] <rmatte> yeh, noon eastern time
[16-Sep-2009 10:04:02] <rmatte> 11 central
[16-Sep-2009 10:04:38] <rmatte> and it'll be like 5pm Jane's time
[16-Sep-2009 10:08:52] <rmatte> http://dmon.org/graphics/girlfriend.jpg
[16-Sep-2009 10:10:01] <ckrough> ...do I want to click that
[16-Sep-2009 10:10:02] <twm1010> i just registered
[16-Sep-2009 10:10:11] <rmatte> ckrough: it's SFW
[16-Sep-2009 10:10:25] <twm1010> rmatte: i almost fell out of my chair
[16-Sep-2009 10:10:32] <rmatte> hehe
[16-Sep-2009 10:10:33] <ckrough> Paladin, pfft.. hybrid classes always suck
[16-Sep-2009 10:10:42] <ckrough>
[16-Sep-2009 10:11:09] <rmatte> hehe
[16-Sep-2009 10:11:50] <rmatte> http://dmon.org/graphics/windowsvistamarketing.jpg
[16-Sep-2009 10:12:08] <rmatte> http://dmon.org/graphics/wires.jpg
\n[16-Sep-2009 10:12:49] <rmatte> Oh, and for anyone who has seen\nthe whitest kids u know tattoo parlour skit:\nhttp://dmon.org/graphics/wkuktat.jpg
[16-Sep-2009 10:13:45] <twm1010> im looking at your zenoss screenshots :O
[16-Sep-2009 10:13:51] <rmatte> hehe
[16-Sep-2009 10:14:17] <rmatte> yeh, they are pretty decent
[16-Sep-2009 10:14:19] <twm1010> MPLS?
[16-Sep-2009 10:14:24] <twm1010> looks like totally mesh networks
[16-Sep-2009 10:14:29] <ckrough> I like schoolgirl.jpg...
[16-Sep-2009 10:14:43] <rmatte> well, the larger map is partial MPLS (they use it in Mexico)
[16-Sep-2009 10:14:50] <rmatte> the smaller map is all MPLS
\n[16-Sep-2009 10:14:57] <ckrough> bah, and using my regular nick\ntoo... comment google indexed for the rest of time. stupid internet.
[16-Sep-2009 10:15:39] <rmatte> lol
\n[16-Sep-2009 10:15:53] <twm1010> Yeah, I've been amazed when\ndoing some Zenoss related searching how many times my own name comes up\nin the google searches from captured chats from this same chatroom
[16-Sep-2009 10:16:13] <rmatte> yeh, I'm in here like all the time, so mine does as well
[16-Sep-2009 10:16:44] <rmatte> interesting, my zenoss community profile is already the top result on google for "rmatte"
[16-Sep-2009 10:16:46] <rmatte> lol
[16-Sep-2009 10:17:17] <ckrough> mrayzenoss has been spamming the google 'reindex' button.
[16-Sep-2009 10:17:33] <ckrough> yup, mine is first result for ckrough
[16-Sep-2009 10:17:41] <mrayzenoss> you guys are internet famous!
[16-Sep-2009 10:17:54] <ckrough> assuming your searching for ckrough, ya
\n[16-Sep-2009 10:17:55] <rmatte> If I search for Ryan Matte, my\nLinkedIn profile is 4th result, and the blog entry of Eric and I being\nnamed Zenoss Masters is 5th.
[16-Sep-2009 10:17:56] <ckrough>
[16-Sep-2009 10:18:40] <mrayzenoss> hmmm.. I make the first page of google just barely
[16-Sep-2009 10:18:46] <rmatte> hehe
[16-Sep-2009 10:19:03] <mrayzenoss> but if you throw in "Linux", I'm the whole page
[16-Sep-2009 10:19:09] <rmatte> oh, for the path of least resistance blog
[16-Sep-2009 10:19:15] <ckrough> lol. searching my first name returns this: http://www.myspace.com/wchrisw
[16-Sep-2009 10:19:18] <ckrough> and that is NOT me
[16-Sep-2009 10:19:23] <ckrough> poser.
[16-Sep-2009 10:19:35] <rmatte> lmao
[16-Sep-2009 10:20:04] <rmatte> Yeh, well the first result for Ryan Matte is http://www.facebook.com/people/Ryan-Matte/748530391
[16-Sep-2009 10:20:06] <rmatte> which is not me
[16-Sep-2009 10:20:08] <rmatte> lol
[16-Sep-2009 10:20:31] <rmatte> ...and this one is sooooo not me: http://www.facebook.com/people/Ryan-Matte/713617305
[16-Sep-2009 10:21:04] <rmatte> I'm on Twitter?
[16-Sep-2009 10:21:27] <rmatte> oh, right, when I posted to the Modern Warfare 2 twitter page the one time
[16-Sep-2009 10:22:02] <rmatte> and apparently I have 3 followers regardless of the fact that I've only made 1 post
\n[16-Sep-2009 10:22:38] <rmatte> oh, and a bug report that I\nlogged 2 years ago for Amarok:\nhttp://markmail.org/message/73dhm2rrcewtsrgp
[16-Sep-2009 10:22:50] <twm1010> odd, informant SNMP doens't work on one of my 2K3 boxes
[16-Sep-2009 10:22:52] <rmatte> interesting what makes the first page in Google sometimes
\n[16-Sep-2009 10:22:56] <twm1010> cpu check returns No more\nvariables left in this MIB View (It is past the end of the MIB tree)
[16-Sep-2009 10:23:15] <rmatte> twm1010: sounds like you need my ZenPack
[16-Sep-2009 10:23:30] <twm1010> which?
[16-Sep-2009 10:23:44] <twm1010> confused why only one host refuses to turn this OID
[16-Sep-2009 10:23:59] <rmatte> docs/DOC-3386
\n[16-Sep-2009 10:24:20] <rmatte> there will be a simpler version\nof that ZenPack up as soon as Matt actually has time to review all of\nthe ones sitting in his inbox
[16-Sep-2009 10:24:35] <rmatte> that one is for multicore, the other is for single core
[16-Sep-2009 10:24:41] <rmatte> errr, well, for monitoring total
[16-Sep-2009 10:24:44] <rmatte> rather than each core
[16-Sep-2009 10:25:01] <rmatte> http://dmon.org/downloads/zenoss/ZenPacks.Nova.Windows.SNMPPerfMonitor.Simple-1.2-py2.4.egg
[16-Sep-2009 10:25:05] <rmatte> there's a link to the other one
[16-Sep-2009 10:26:03]\n<rmatte> the simple one adds a template to /Server/Windows that\nmonitors total CPU usage rather than each core, the one that's up on\nthe community site created organizers for 1 to 16 CPUs/cores, that you\nthen move your devices under appropriately
[16-Sep-2009 10:27:09] <twm1010> ah, SNMP monitoring without informant?
[16-Sep-2009 10:27:10] <twm1010> very nice
[16-Sep-2009 10:27:29] <rmatte> correcto
[16-Sep-2009 10:27:40] <rmatte> you just need to have bc and snmpwalk installed on the system
[16-Sep-2009 10:27:45] <rmatte> since the scripts use those
[16-Sep-2009 10:28:08] <twm1010> hrmm
[16-Sep-2009 10:28:40] <twm1010> hrmm, i think informant DLL is tied up on that box
[16-Sep-2009 10:33:00] <ckrough> I think it's time to move my events db to a dedicated box.
[16-Sep-2009 10:33:22] <ckrough> Maybe even figure out a way to kick zenhub out of the main box... clustering some how.
[16-Sep-2009 10:35:34] <twm1010> bah, i concede defeat, the server needs a reboot
[16-Sep-2009 10:43:35] <mrayzenoss> WebEx is up for the webinar
[16-Sep-2009 10:46:48] <rmatte> I'll be in shortly
[16-Sep-2009 10:48:20] <ckrough> Is there a conference bridge for audio? WebEx isnt liking Fedora11 for audio
[16-Sep-2009 10:48:38] <rmatte> ckrough: yeh, there is
[16-Sep-2009 10:48:51] <mrayzenoss> yeah, there's a phone number
[16-Sep-2009 10:48:52] <rmatte> I always dial in since I'm on Ubuntu
[16-Sep-2009 10:48:58] <rmatte> and my speakers blow
[16-Sep-2009 10:49:17] <rmatte> I think they were manufactured in 1976
[16-Sep-2009 10:49:21] <ckrough> had that information been shared already? I dont see it
[16-Sep-2009 10:50:01] <rmatte> don't think so
[16-Sep-2009 10:50:04] <rmatte> it will be though
[16-Sep-2009 10:50:05] <ckrough> ok
[16-Sep-2009 10:51:10] <mrayzenoss> 866-699-3239
[16-Sep-2009 10:51:23] <mrayzenoss> or 1-408-792-6300
[16-Sep-2009 10:51:34] <rmatte> k
[16-Sep-2009 10:51:35] <mrayzenoss> access code 759 626 698
[16-Sep-2009 10:52:02] <ckrough> thanks
[16-Sep-2009 11:11:10] <twm1010> noise cancelling headphones
[16-Sep-2009 11:11:59] <twm1010> mrayzenoss: will this webinar be saved for download?
[16-Sep-2009 11:12:06] <mrayzenoss> yeah
[16-Sep-2009 11:12:17] <mrayzenoss> it's being recorded, there will be an encore session next week
[16-Sep-2009 11:15:44] <ckrough> my eyes
[16-Sep-2009 11:15:49] <mrayzenoss> heh
[16-Sep-2009 11:16:01] <rmatte> lol
[16-Sep-2009 11:25:14] <Apocalipse> hello
[16-Sep-2009 11:25:37] <jb> hmm
[16-Sep-2009 11:25:39] <Apocalipse> i would like to thx matt...
[16-Sep-2009 11:25:44] <Apocalipse> thx for help
[16-Sep-2009 11:25:45] <Apocalipse>
[16-Sep-2009 11:26:49] <jb> if I have a datasource returning "0.054"
[16-Sep-2009 11:27:27] <jb> and I want this to be graphed like "50.4ms"
[16-Sep-2009 11:27:48] <jb> right now im using a format of "%5.2lf%s"
[16-Sep-2009 11:27:58] <jb> and the result is something like "17.3m"
[16-Sep-2009 11:28:01] <jb> what is the "m" ?
[16-Sep-2009 11:28:08] <ckrough> micro
[16-Sep-2009 11:28:12] <ckrough> err milli
[16-Sep-2009 11:28:27] <jb> so its really like "17.3milliseconds/ms"
[16-Sep-2009 11:28:28] <ckrough> it's dividing a small datapoint out over the polling cycle
[16-Sep-2009 11:28:32] <jb> (or ms)
[16-Sep-2009 11:29:15] <ckrough> millipackets/second
[16-Sep-2009 11:30:57] <jb> ok, you have any idea on how to get that in ms (milliseconds)?
[16-Sep-2009 11:31:01] <ckrough> or milliwhatever/sec
[16-Sep-2009 11:31:28] <ckrough> what are you measuring?
[16-Sep-2009 11:31:37] <jb> disk latency
\n[16-Sep-2009 11:32:20] <ckrough> you may be seeing the right\nthing already then. Ive dealt with that with units/time measurement,\nbut not latency
[16-Sep-2009 11:32:40] <jb> well, the value returned is something like "0.054"
[16-Sep-2009 11:32:45] <jb> which is really 50.4ms
[16-Sep-2009 11:33:30] <rmatte> you could remove the %s at the end and set units to "seconds"
[16-Sep-2009 11:33:36] <rmatte> then it would graph fractions of a second
[16-Sep-2009 11:34:21] <jb> hrm yeah
[16-Sep-2009 11:34:31] <jb> has to be a way to get it to show MS tho
[16-Sep-2009 11:37:49] <rmatte> oh, I'm sure there is, I just don't fully understand the formatting
[16-Sep-2009 11:37:57] <rmatte> so I'm not a huge help hehe
[16-Sep-2009 11:38:08] <rmatte> I'll have to read up on it
[16-Sep-2009 11:38:31] <jb> CRAP
[16-Sep-2009 11:38:33] <jb> i forgot about the webinar
[16-Sep-2009 11:38:34] <jb> :/
[16-Sep-2009 11:38:42] <mrayzenoss> there will be an encore next wek
[16-Sep-2009 11:38:43] <mrayzenoss> week
[16-Sep-2009 11:38:48] <rmatte> we're about half way through
[16-Sep-2009 11:39:36] <markeriv> hey guys
[16-Sep-2009 11:39:44] <markeriv> I have a question concerning users and alerts
\n[16-Sep-2009 11:39:58] <markeriv> is there a way to tie a user to\na group so that they only recieve alerts from that group?
[16-Sep-2009 11:40:17] <markeriv> i have RTFMing and have not yet found an area specifically detailing that
[16-Sep-2009 11:40:18] <rmatte> markeriv: a device group you mean?
[16-Sep-2009 11:40:22] <mrayzenoss> you can assign administrators to specific groups
[16-Sep-2009 11:40:52] <markeriv> not a device group
\n[16-Sep-2009 11:40:58] <rmatte> markeriv: not sure I fully\nunderstand what you're asking, are you asking abour organizational\ngroups for users?
[16-Sep-2009 11:41:04] <markeriv> hmm
[16-Sep-2009 11:41:05] <rmatte> about*
[16-Sep-2009 11:41:11] <markeriv> you know how you can browse by systems
[16-Sep-2009 11:41:11] <markeriv> groups
[16-Sep-2009 11:41:13] <markeriv> locations
[16-Sep-2009 11:41:16] <markeriv> or networks
[16-Sep-2009 11:41:17] <rmatte> yes
[16-Sep-2009 11:41:22] <markeriv> i have a group
[16-Sep-2009 11:41:24] <jb> just configure the alert
[16-Sep-2009 11:41:26] <markeriv> we will call them emser tile
[16-Sep-2009 11:41:28] <jb> to only send for systems in that group
[16-Sep-2009 11:41:37] <markeriv> ok
[16-Sep-2009 11:41:43] <rmatte> yeh, you want to add a filter rule
[16-Sep-2009 11:41:55] <rmatte> something like "group is /bleh/blah"
[16-Sep-2009 11:42:01] <rmatte> to the alert
[16-Sep-2009 11:42:32] <markeriv> hmm
[16-Sep-2009 11:42:35] <markeriv> i see that area
[16-Sep-2009 11:42:40] <markeriv> but group is not an option
[16-Sep-2009 11:42:49] <rmatte> hunh?
[16-Sep-2009 11:42:52] <markeriv> ahh
[16-Sep-2009 11:42:55] <markeriv> its device groups
[16-Sep-2009 11:42:57] <markeriv> not just groups
[16-Sep-2009 11:43:01] <markeriv> sometimes i feel stupid
[16-Sep-2009 11:43:05] <markeriv> this is one of those times
[16-Sep-2009 11:43:06] <rmatte> yeh, device group
[16-Sep-2009 11:43:10] <rmatte> lol
[16-Sep-2009 11:43:48] <rmatte> (I don't have Zenoss in front of me so I was just doing it off the top of my head)
[16-Sep-2009 11:54:36] mikey__ is now known as Michael1
[16-Sep-2009 11:54:43] <Michael1> hello?
[16-Sep-2009 11:55:13] <Apocalipse> hey people... to change the google map view...
\n[16-Sep-2009 11:55:30] <Michael1> does anyone know of an app than\ncan be run on a windows machine, monitor fans/temp/smart data and then\ngenerate traps to send to zenoss ?
[16-Sep-2009 11:57:04] <chemist> Michael1: look at the snmp-informant website
[16-Sep-2009 11:57:19] <rmatte> Michael1: yeh, snmp-informant can do that, just needs to be configured for it
[16-Sep-2009 11:57:20] <chemist> they have a hw monitoring tool there
[16-Sep-2009 12:01:43] <Michael1> looks like what i'm looking for - just going to install it....
[16-Sep-2009 12:17:26] <markeriv> Hey guys
[16-Sep-2009 12:17:29] <markeriv> got another question for you
[16-Sep-2009 12:17:38] <markeriv> What is your favorite zenoss feature and why?
[16-Sep-2009 12:17:40] <ckrough> hmm. isnt there a default "Production State" report?
[16-Sep-2009 12:20:39] <rmatte> ckrough: I just created one in the custom reports section
[16-Sep-2009 12:20:48] <rmatte> you just want to see the production states of all devices?
[16-Sep-2009 12:20:51] <ckrough> thought I had one
[16-Sep-2009 12:21:07] <ckrough> yeah, just put a couple thousand into maintenance and want to see a list to be sure
[16-Sep-2009 12:21:29] <rmatte> Reports -> Custom Device Reports -> Add Device Report, name it Production States
[16-Sep-2009 12:22:15] <rmatte> set Sort Sense as asc
[16-Sep-2009 12:22:26] <rmatte> set Sort Column as getId
[16-Sep-2009 12:22:32] <rmatte> in Columns put:
[16-Sep-2009 12:22:37] <rmatte> getId
[16-Sep-2009 12:22:37] <rmatte> getManageIp
[16-Sep-2009 12:22:41] <rmatte> getProdState
[16-Sep-2009 12:22:47] <rmatte> in Column Names put:
[16-Sep-2009 12:22:51] <rmatte> Device
[16-Sep-2009 12:22:51] <rmatte> IP Address
[16-Sep-2009 12:22:59] <rmatte> Production State
[16-Sep-2009 12:23:03] <rmatte> save it, and you're done
[16-Sep-2009 12:23:07] <ckrough> thanks
[16-Sep-2009 12:23:11] <rmatte> np
[16-Sep-2009 12:24:00] <ckrough> heh, there has got to be smoke coming out of this box right now
[16-Sep-2009 12:24:50] <rmatte> lol
[16-Sep-2009 12:27:32] <malbon> ls
[16-Sep-2009 12:29:05] <rmatte> .
[16-Sep-2009 12:29:07] <rmatte> ..
[16-Sep-2009 12:29:11] <rmatte> bin
[16-Sep-2009 12:29:13] <rmatte> etc
[16-Sep-2009 12:29:15] <rmatte> lib
[16-Sep-2009 12:29:19] <rmatte> usr
[16-Sep-2009 12:29:43] <rmatte> malbon@l33tbox#:
[16-Sep-2009 12:32:11] <etank> i think that i asked this before but can zenoss do the graph zooming that opennms can do?
[16-Sep-2009 12:32:25] <etank> where you click and drag over an area that you want to zoom to
[16-Sep-2009 12:34:18] <ckrough> whats the query syntax? "here.getProdState < 1000"
[16-Sep-2009 12:34:22] <ckrough> or dev.
[16-Sep-2009 12:34:27] <ckrough> tried a couple none worked
[16-Sep-2009 12:36:50] <rmatte> etank: you click to zoom, there's no drag zooming
[16-Sep-2009 12:37:01] <rmatte> etank: you can zoom in, zoom out, and move backwards or forward
[16-Sep-2009 12:43:11] <Apocalipse> ryan... zenoss is collecting a wrong info from a device...
[16-Sep-2009 12:43:28] <Apocalipse> i enter on the device and checked...
[16-Sep-2009 12:43:37] <Apocalipse> inside it is ok...
[16-Sep-2009 12:43:45] <Apocalipse> but zenoss get wrong...
[16-Sep-2009 12:44:34] <Apocalipse> i tryed re-model the device...
[16-Sep-2009 12:44:41] <Apocalipse> but still same
[16-Sep-2009 12:44:53] <Apocalipse> can i change this value manually?
[16-Sep-2009 12:47:42] <Apocalipse> ?
[16-Sep-2009 12:48:00] <etank> rmatte: right. but drag zoom would be nice
[16-Sep-2009 12:49:13] <twm1010> Apocalipse: what value?
[16-Sep-2009 12:49:43] <Apocalipse> ip address on interfaces
[16-Sep-2009 12:49:51] <Apocalipse> os tab
[16-Sep-2009 12:50:04] <Apocalipse> look this
[16-Sep-2009 12:50:21] <Apocalipse> IP-MIB::ipAdEntNetMask.10.x.x.y = IpAddress: 255.255.255.252
[16-Sep-2009 12:50:29] <Apocalipse> but in zenoss is:
[16-Sep-2009 12:50:50] <Apocalipse> 10.x.x.y/24
[16-Sep-2009 12:51:00] <Apocalipse> not /30
[16-Sep-2009 12:51:51] <twm1010> hrmm... one sec let me look at one of mine
[16-Sep-2009 12:51:53] <Apocalipse> got it?
[16-Sep-2009 12:51:57] <Apocalipse> ok
[16-Sep-2009 12:52:31] <ckrough> that's the group that zenoss displays it under in /networks you mean?
[16-Sep-2009 12:52:36] <twm1010> ah, ok, the ip column uses the /# to display the subnet
\n[16-Sep-2009 12:53:13] <twm1010> you would figure, that when you\nchange the mask on the server, it should change in zenoss
[16-Sep-2009 12:54:58] <twm1010> hard for me to find an example here, everything in our network IS /24
[16-Sep-2009 12:55:39] <twm1010> hrmm, our MPLS links show /30
[16-Sep-2009 12:56:56] <twm1010> you could delete the interface, and then remodel
[16-Sep-2009 12:57:01] <twm1010> see where that gets you
[16-Sep-2009 12:59:38] <Apocalipse> hunnnnnnnnnnnnnnnnnnnnnn
[16-Sep-2009 12:59:43] <Apocalipse> let me do it...
[16-Sep-2009 12:59:44] <Apocalipse> hol
[16-Sep-2009 12:59:46] <Apocalipse> hold
[16-Sep-2009 12:59:47] <Apocalipse>
[16-Sep-2009 13:02:52] <Apocalipse> received same
[16-Sep-2009 13:02:55] <Apocalipse>
[16-Sep-2009 13:05:42] <Apocalipse> is it possible to change this inside the database?
[16-Sep-2009 13:05:49] <Apocalipse> directlly i mean
[16-Sep-2009 13:06:23]\n<ckrough> Im not sure exactly what you are describing, Im jumping\nbetween windows. I just want to make sure you are not mistaking Zenoss'\nnetwork organization for an interface configuration
[16-Sep-2009 13:06:59]\n<ckrough> Zenoss nests devices in organization subnets that arent\nnecessarily matched to the subnets of your interfaces
[16-Sep-2009 13:11:36] <Apocalipse> hunnn
[16-Sep-2009 13:11:50] <Apocalipse> i saw all nested subnets...
[16-Sep-2009 13:12:02] <Apocalipse> there is right...
[16-Sep-2009 13:13:38] <rmatte> lovely, can't install any ZenPacks containing Mibs in King Crab
[16-Sep-2009 13:14:01] <Apocalipse> hunnn
[16-Sep-2009 13:14:40] <rmatte> I've been using King Crab for about 30 minutes and I've already found 5 issues with it
[16-Sep-2009 13:15:10] <rmatte> I have a feeling this is going to be a painful experience
[16-Sep-2009 13:15:25] <Apocalipse> changing subject... where can i change the view of maps?
[16-Sep-2009 13:15:27] <Apocalipse> map
[16-Sep-2009 13:15:35] <rmatte> what do you mean by change the view?
[16-Sep-2009 13:16:00] <Apocalipse> map, satelite, hybrid...
[16-Sep-2009 13:16:06] <Apocalipse> google map
[16-Sep-2009 13:16:12] <rmatte> you can't
[16-Sep-2009 13:16:16] <Apocalipse> i found an article... but i am not sure...
[16-Sep-2009 13:16:19] <rmatte> it's hard coded to be a map view
[16-Sep-2009 13:16:24] <Apocalipse> message/34678#34678
[16-Sep-2009 13:16:53] <rmatte> yeh, that would be the only way
[16-Sep-2009 13:16:56] <rmatte> by editing the actual code
[16-Sep-2009 13:16:59] <Apocalipse> yes
[16-Sep-2009 13:17:41] <rmatte> You'd want to add the this.map.addControl(new GMapTypeControl()); line
\n[16-Sep-2009 13:18:43] <rmatte> the reason they don't use\nsatellite view is because a lot of the satellite images are super low\nquality
[16-Sep-2009 13:18:55] <rmatte> so it kind of ruins the look
[16-Sep-2009 13:19:25] <ckrough> and much slower
[16-Sep-2009 13:19:31] <rmatte> yeh
[16-Sep-2009 13:19:32] <rmatte> that too
[16-Sep-2009 13:19:56] <ckrough> I dont use that portlet at all. Datacenters tend not to move around.
[16-Sep-2009 13:20:08] <rmatte> hehe
[16-Sep-2009 13:20:22] <Apocalipse> heheheheh
[16-Sep-2009 13:20:27] <Apocalipse> big boss WANTS
[16-Sep-2009 13:20:28] <rmatte> we just use it because it gives a good overhead view of our client's sites
[16-Sep-2009 13:20:38] <Apocalipse> i just say..."yes boss"
[16-Sep-2009 13:20:56] <Apocalipse> where i put this line ?
[16-Sep-2009 13:21:08] <Apocalipse> end?
[16-Sep-2009 13:21:11] <rmatte> it says right in the post
[16-Sep-2009 13:21:13] <rmatte> just read
[16-Sep-2009 13:21:40] <Apocalipse> i am confused about the post...
[16-Sep-2009 13:21:51] <rmatte> it's as clear as day
[16-Sep-2009 13:21:58] <Apocalipse> for u... sorry...
[16-Sep-2009 13:22:08] <rmatte> it says to go in to a specific file and put one of those 2 lines somewhere after line 24 in the file
[16-Sep-2009 13:22:12] <Apocalipse> yes
[16-Sep-2009 13:22:18] <rmatte> so what's the issue?
[16-Sep-2009 13:24:16] <rmatte> then to do the same with a second file
[16-Sep-2009 13:31:24] <Apocalipse> here i only have geomap.js
[16-Sep-2009 13:32:05] <twm1010> rmatte: you just reminded me to download king crab
[16-Sep-2009 13:32:23] <ckrough> any zope pros in here?
\n[16-Sep-2009 13:33:20] <rmatte> twm1010: I've found\nhttp://dev.zenoss.org/trac/ticket/5546 and\nhttp://dev.zenoss.org/trac/ticket/5547 so far
[16-Sep-2009 13:33:35] <iancmcc> Apocalipse: if you're on 2.4.x, there's only one google maps js now
[16-Sep-2009 13:33:36] <rmatte> twm1010: I'll probably have quite a few bugs logged before end of day
[16-Sep-2009 13:33:48] <rmatte> I plan to test the crap out of this
[16-Sep-2009 13:34:03] <iancmcc> ckrough: whatcha got? I suppose you could call me a pro, insofar as I get paid.
[16-Sep-2009 13:34:06] <rmatte> kind of difficult to do some testing without Mibs though
[16-Sep-2009 13:34:29] <rmatte> ckrough: I'm not a Zope "pro", but I know my fair share
[16-Sep-2009 13:40:40] <ckrough> thinking about how to get some dialog going about scaling zenoss in large installs.
[16-Sep-2009 13:40:52] <ckrough> Maybe Ill start a forum thread
[16-Sep-2009 13:42:47] <rmatte> http://dev.zenoss.org/trac/ticket/5548
[16-Sep-2009 13:50:15] <jb> i think i like the background colors on the old event console :/
[16-Sep-2009 13:50:26] <rmatte> you can toggle those on
[16-Sep-2009 13:50:28] <iancmcc> The specific colors? Or just having colors?
[16-Sep-2009 13:50:32] <iancmcc> Yeah, what rmatte said.
[16-Sep-2009 13:50:34] <jb> oh yeah?
[16-Sep-2009 13:50:35] <ckrough> jb: are you looking at unack'd events?
[16-Sep-2009 13:50:40] <jb> found it
[16-Sep-2009 13:50:50] <rmatte>
[16-Sep-2009 13:51:23] <rmatte> and I agree, I like it better with the background colors too
[16-Sep-2009 13:51:54] <chudler> is there are reason, or you just talking about taste?
[16-Sep-2009 13:51:55] <jb> and "Close" is like move to history?
[16-Sep-2009 13:52:05] <jb> chudler: just easier to differentiate..
[16-Sep-2009 13:52:10] <iancmcc> Close is exactly move to history.
\n[16-Sep-2009 13:53:18] <ckrough> Some of my netEngs had initially\nbalked at the background colors because they were so intense, but the\nproblem was we were looking at a page full of unacknowledged alerts
[16-Sep-2009 13:53:34] <ckrough> iancmcc: you get my privmsg?
[16-Sep-2009 13:53:40] <rmatte> chudler: it's more blatantly obvious what's higher severity
[16-Sep-2009 13:54:15] <rmatte> ckrough: yeh, when they are acknowledged they go pastel
[16-Sep-2009 13:54:17] <chudler> interesting, I thought the default was sorted by severity?
\n[16-Sep-2009 13:54:59] <rmatte> chudler: it is, but when you're\nwoken up at 2 in the morning to investigate something you don't want to\nhave to stare at little severity icons to determine what's what
[16-Sep-2009 13:55:27] <rmatte> it can be toggled anyways
[16-Sep-2009 13:55:55] <rmatte> but I suppose it's mostly personal preference
[16-Sep-2009 13:57:05] <rmatte> ah, I guess they haven't fixed the acknowledgement toggling yet
[16-Sep-2009 13:57:16] <rmatte> you can acknowledge but you can't unacknowledge
[16-Sep-2009 13:57:24] <iancmcc> Yeah, we didn't add that in.
[16-Sep-2009 13:57:38] <iancmcc> Big deal for some? I can push it
[16-Sep-2009 13:57:43] <rmatte> Is there a trac open for it?
[16-Sep-2009 13:57:56] <jb> hah, theres a poll about this very topic in the forums now
[16-Sep-2009 13:57:58] <rmatte> Well, it's definitely something that should make the cut
[16-Sep-2009 13:58:28] <rmatte> If you accidentally acknowledge an event then you should be able to unacknowledge it
[16-Sep-2009 13:58:35] <iancmcc> We're feature-frozen, so it's got to be a big deal. I'll see what I can do.
[16-Sep-2009 13:58:45] <iancmcc> Maybe I can accidentally check that in.
[16-Sep-2009 13:59:12] <rmatte> and there's no way to move an event from history back to status?
[16-Sep-2009 13:59:25] <iancmcc> Shoot. THAT I forgot.
[16-Sep-2009 13:59:34] <iancmcc> I'll open a ticket for that one.
[16-Sep-2009 13:59:39] <rmatte>
[16-Sep-2009 13:59:44] <rmatte> my god, I am on a roll today
[16-Sep-2009 13:59:49] <rmatte> I can open it if you want
[16-Sep-2009 13:59:53] <rmatte> up to you
[16-Sep-2009 13:59:53] <jb> actually
[16-Sep-2009 13:59:57] <jb> where is the "View History" now?
[16-Sep-2009 14:00:02] <rmatte> bottom left
[16-Sep-2009 14:00:03] <iancmcc> Bottom left
[16-Sep-2009 14:00:06] <jb> oh yeah
[16-Sep-2009 14:00:06] <jb> k
[16-Sep-2009 14:00:12] <iancmcc> Hidden, I know.
[16-Sep-2009 14:00:19] <rmatte> iancmcc: by the way, I absolutely love the new popup style
[16-Sep-2009 14:00:22] <jb> yeah maybe make that a buton?
[16-Sep-2009 14:00:24] <jb> button
[16-Sep-2009 14:00:26] <rmatte> iancmcc: that's just way too sexy
[16-Sep-2009 14:00:32] <rmatte> draggable semi-transparent
[16-Sep-2009 14:00:44] <jb> yeah its way better (the popup)
[16-Sep-2009 14:00:48] <rmatte> yeh
[16-Sep-2009 14:00:48]\n<iancmcc> It's a bit hacky because this screen is our first foray\ninto the new UI, which has a different navigation scheme
[16-Sep-2009 14:00:59] <iancmcc> so we've had to shove a few things in where they don't exactly belong
[16-Sep-2009 14:00:59] <jb> yeah i wondered how you were going to do this
[16-Sep-2009 14:01:24] <jb> since the new UI is different
[16-Sep-2009 14:01:28] <rmatte> yeh
[16-Sep-2009 14:01:31] <jb> at least from the shots ive seen..
[16-Sep-2009 14:01:34] <jb> which are from months ago
[16-Sep-2009 14:01:36] <rmatte> new UI won't make it until stone crab though
[16-Sep-2009 14:01:51] <iancmcc> History can't be a button, bad usability juju. We discussed putting back the History tab.
[16-Sep-2009 14:02:16] <iancmcc> But I figured, you only have to find the link once.
[16-Sep-2009 14:02:21] <jb> yea
[16-Sep-2009 14:02:24]\n<rmatte> iancmcc: the problem with the link is that people miss\nit, but I personally think it's a dumb argument, as soon as people know\nit's there once then they ALWAYS know it's there
[16-Sep-2009 14:02:33] <rmatte> there's no point in having a big tab sitting there for it
[16-Sep-2009 14:02:36] <rmatte> but that's my opinion
[16-Sep-2009 14:03:00] <iancmcc> Tab real estate is at a premium on some contexts, so we try not to add them
[16-Sep-2009 14:03:16] <ckrough> iancmcc: agreed
[16-Sep-2009 14:03:19] <ckrough> less is more
[16-Sep-2009 14:03:48] <rmatte> minimalistic clean cut look is the way to go
[16-Sep-2009 14:03:52] <jb> i liked having the "Details" tab
[16-Sep-2009 14:03:54] <jb> for event details
[16-Sep-2009 14:04:03] <jb> now all of it is string together using plain white cells
[16-Sep-2009 14:04:07] <jb> which makes it sort of hard to read
[16-Sep-2009 14:04:08] <rmatte> jb: you just double-click
[16-Sep-2009 14:04:17] <jb> yeah, but its displayed differently
[16-Sep-2009 14:04:23] <rmatte> right, I see what you're saying
[16-Sep-2009 14:04:37] <rmatte> I agree, it would be nice to have alternating cell colors
[16-Sep-2009 14:04:40] <rmatte> for the info that's displayed
[16-Sep-2009 14:04:58] <jb> yeah for sure
[16-Sep-2009 14:05:01] <iancmcc> I'll run it by our UI designer. Might be something we can do.
[16-Sep-2009 14:05:25] <rmatte> doesn't have to be dark colors ian, just a really light shade of gray on every second column
[16-Sep-2009 14:05:39] <rmatte> erm second row
[16-Sep-2009 14:05:39] <ckrough> and a pony
[16-Sep-2009 14:05:40] <rmatte> rather
[16-Sep-2009 14:05:44] <iancmcc> I'll check it out, see how it looks
[16-Sep-2009 14:05:51] <rmatte> MOAR PONIES!
[16-Sep-2009 14:06:13] <rmatte> we should so open a trac ticket for that ckrough
[16-Sep-2009 14:06:16] <jb> its much easier to read on in current versions..
[16-Sep-2009 14:06:18] <ckrough> heh
[16-Sep-2009 14:06:31] <jb> the alignment is goofy on 2.4.7 too
[16-Sep-2009 14:06:34] <jb> its all centered
[16-Sep-2009 14:06:35] <rmatte> Title: King Crab needs MOAR PONIES!
[16-Sep-2009 14:06:36] npmccallum_ is now known as npmccallum
[16-Sep-2009 14:06:41] <rmatte> Steps to reproduce:
[16-Sep-2009 14:06:44] <rmatte> 1. Log in.
[16-Sep-2009 14:06:50] <rmatte> Results
[16-Sep-2009 14:06:57] <rmatte> NO PONIES!
[16-Sep-2009 14:07:01] <rmatte> Expected results:
[16-Sep-2009 14:07:05] <rmatte> MOAR PONIES!
[16-Sep-2009 14:07:09] <rmatte> Impact:
[16-Sep-2009 14:07:11] <rmatte> BAHHHH!
[16-Sep-2009 14:07:29] <ckrough> lol
[16-Sep-2009 14:07:32] <iancmcc> http://dev.zenoss.org/trac/ticket/5549
[16-Sep-2009 14:09:10] <jb> http://thunder.jbdesign.net/~jbaird/zenoss/event-old.jpg
[16-Sep-2009 14:09:12] <jb> http://thunder.jbdesign.net/~jbaird/zenoss/event-new.jpg
[16-Sep-2009 14:09:50] <jb> just to compare..
\n[16-Sep-2009 14:10:18] <iancmcc> It went through design. That\nwhole pane is sort of free-floating, so it's ugly to shove a big table\nin there
[16-Sep-2009 14:10:24] <iancmcc> but delicate shading might work well
[16-Sep-2009 14:10:36] <rmatte> iancmcc: ummm, help me out here...
[16-Sep-2009 14:10:49] <rmatte> iancmcc: I click on event history, and it loads...
[16-Sep-2009 14:10:57] <rmatte> then there are 4 tabs at the top
\n[16-Sep-2009 14:11:18] <rmatte> Classes, which is presumably the\nold Events section, Mappings which is a summary of all event mapping,\nEvents, and zProperties
[16-Sep-2009 14:11:20] <iancmcc> depending on context...
[16-Sep-2009 14:11:29] <iancmcc> OH you mean from the global console?
[16-Sep-2009 14:11:35] <iancmcc> bug. Missing the notabs url parameter.
\n[16-Sep-2009 14:11:36] <rmatte> If I were to go in to zProperties\nand set action to "drop" it would presumably globally drop all events,\ncorrect?
[16-Sep-2009 14:12:01] <rmatte> ah
[16-Sep-2009 14:12:08] <rmatte> ok, I thought that was a bit odd
[16-Sep-2009 14:12:12] <rmatte> I'll log that as a bug then
[16-Sep-2009 14:12:23] <iancmcc> There's no difference between the Event Console and the events tab under /Events
[16-Sep-2009 14:12:28] <iancmcc> just the tabs are hidden
[16-Sep-2009 14:12:40] <rmatte> gotcha, so it's just not hiding them in history when it should be
[16-Sep-2009 14:12:43] <iancmcc> right.
[16-Sep-2009 14:12:46] <rmatte> okie
[16-Sep-2009 14:12:53] <rmatte> I'll open a ticket and pass you the number
[16-Sep-2009 14:15:40] <rmatte> iancmcc: http://dev.zenoss.org/trac/ticket/5550
[16-Sep-2009 14:16:15] <cgibbons> today, I hate windows.
[16-Sep-2009 14:16:24] <rmatte> tomorrow, who knows?
[16-Sep-2009 14:16:27] <iancmcc> right there with you, chadwick
[16-Sep-2009 14:16:32] <markeriv> today i hate offnet product t1 circuits and edgewater routers
[16-Sep-2009 14:16:34] <markeriv> join the club
[16-Sep-2009 14:16:34] <rmatte> afk
[16-Sep-2009 14:16:46] <markeriv> err excuse me, onnet
[16-Sep-2009 14:33:05] <markeriv> I have found a bug in version 2.4.2
[16-Sep-2009 14:33:16] <rmatte> markeriv: which bug?
[16-Sep-2009 14:33:25] <markeriv> sometimes when adding a device the comments field does not save
[16-Sep-2009 14:33:36] <markeriv> and must be reinput after the fact
[16-Sep-2009 14:33:47] <rmatte> might still be a bug, you could log it
[16-Sep-2009 14:33:51] <rmatte> do you have a trac account?
[16-Sep-2009 14:33:56] <markeriv> dont think so
[16-Sep-2009 14:34:04] <markeriv> i have a forums acct
[16-Sep-2009 14:34:12] <rmatte> ask mrayzenoss if you want a trac account
[16-Sep-2009 14:34:34] <rmatte> then you can go to http://dev.zenoss.org/trac
[16-Sep-2009 14:34:38] <rmatte> and log a ticket for the bug
[16-Sep-2009 14:38:39] <rmatte> does anyone know about this?:
[16-Sep-2009 14:38:40] <rmatte> Daemon: zenperfwmi starting...
[16-Sep-2009 14:38:41] <rmatte> CRITICAL:zen.pysamba:Unable to install the WMI reactor. reactor already installed
[16-Sep-2009 14:38:58] <rmatte> does King Crab include the WMI Data Source pack by default?
[16-Sep-2009 14:40:17] <markeriv> hmm
[16-Sep-2009 14:40:58] <cgibbons> i know about that
[16-Sep-2009 14:41:09] <rmatte> ?
[16-Sep-2009 14:41:11] <cgibbons> is zenperfwmi some custom zenpack daemon?
\n[16-Sep-2009 14:41:16] <markeriv> Stupid Question, Is it a bad\nthing when the available Zenoss Version is 2.4.5 and labled in red on\nthe versions page. I am currently running 2.4.2 and have not been able\nto find any documentation on specific upgrades.
[16-Sep-2009 14:41:20] <rmatte> it's from Egor's ZenPack
[16-Sep-2009 14:41:26] <mrayzenoss> cgibbons: yeah, it's the WMI Data Source
[16-Sep-2009 14:41:27] <cgibbons> okay, it needs a chnge
[16-Sep-2009 14:41:53] <rmatte> ok, so I'll have to let Egor know to update that pack before the King Crab release then
[16-Sep-2009 14:42:14] <cgibbons> basically this in his main .py file: http://pastebin.com/m4747e457
\n[16-Sep-2009 14:42:51] <rmatte> Well, there's lots of other stuff\nthat I'd like to test but I can't do much until\nhttp://dev.zenoss.org/trac/ticket/5546 is fixed
[16-Sep-2009 14:43:07] <rmatte> I need those Cisco Mibs and I'm not in the mood to build them by hand on a beta version
[16-Sep-2009 14:44:58] <twm1010> cgibbons: are you saying that King Crab comes with a wmi datasource daemon?
[16-Sep-2009 14:45:05] <cgibbons> no.
[16-Sep-2009 14:45:28] <jb> egor's..
\n[16-Sep-2009 14:45:42] <cgibbons> but any third-party daemons\nthat were using the pysamba library (either directly or indirectly)\nrequire that change to work.
[16-Sep-2009 14:45:49]\n<rmatte> nah, he's saying that King Crab does something\ndifferently when it comes to installing custom daemons from ZenPacks
[16-Sep-2009 14:46:19] <rmatte> markeriv: no, that's not a bad thing, and I'd recommend staying with 2.4.2 for a while
[16-Sep-2009 14:46:31] <rmatte> markeriv: 2.4.5 has some annoying bugs in it
\n[16-Sep-2009 14:47:13] <rmatte> If I'd have known I would have\njust patched 2.4.2 with that security patch and stayed at that version
[16-Sep-2009 14:48:08] <rmatte> markeriv: you'll want to do "zenpatch 14181" as the Zenoss user if you haven't already
[16-Sep-2009 14:48:15] <rmatte> markeriv: fixes: http://dev.zenoss.org/trac/ticket/5279
[16-Sep-2009 14:56:27] <mrayzenoss> poll.jspa?poll=1022
[16-Sep-2009 15:02:10] <rmatte> added my 2 cents
[16-Sep-2009 15:03:24] <Troubadix09> hi all
[16-Sep-2009 15:10:58] <rmatte> weird
[16-Sep-2009 15:11:00] <metalklesk> i really hate bmc patrol and bmc portal ¬¬
[16-Sep-2009 15:11:06] <rmatte> how do you get to that polls section
[16-Sep-2009 15:11:21] <rmatte> if I go to the forum, on the right hand side it says "Latest Poll - Thre are no polls"
[16-Sep-2009 15:11:31] <rmatte> I don't see any other reference to them
[16-Sep-2009 15:11:33] <cgibbons> ditto, metalklesk
[16-Sep-2009 15:12:04] <metalklesk> what is ditto ?
[16-Sep-2009 15:12:17] <rmatte> <metalklesk> i really hate bmc patrol and bmc portal ¬¬
[16-Sep-2009 15:12:27] <rmatte> means he agrees with you
[16-Sep-2009 15:12:44] <metalklesk> aaaaaahh ok, sorry i speak spanish
[16-Sep-2009 15:12:44] <metalklesk> jeje
\n[16-Sep-2009 15:18:53] <mrayzenoss> rmatte: I'm trying to figure\nout why the Poll works on the front page, but not the Forums page
[16-Sep-2009 15:20:35] <rmatte> ah
[16-Sep-2009 15:20:52] <mrayzenoss> we've turned on Abuse reporting as well
[16-Sep-2009 15:20:59] <rmatte> cool
[16-Sep-2009 15:23:32] <metalklesk> bmc portal really sucks with snmp receiving
[16-Sep-2009 15:38:39] <rmatte> apparently there's a 100 message limit for private messages?
[16-Sep-2009 15:38:43] <rmatte> that seems pretty low
[16-Sep-2009 15:38:55] <rmatte> is the site being run on a 40gb hard drive or something?
[16-Sep-2009 15:38:56] <rmatte>
[16-Sep-2009 15:39:26] <mrayzenoss> dunno, I can look into that
[16-Sep-2009 15:39:38] <rmatte> I'd say maybe set it to 300 or 500 even
[16-Sep-2009 15:39:52] <rmatte> even 1000 probably wouldn't choke out the storage
[16-Sep-2009 15:40:22] <rmatte> up to you guys though, it's your site
[16-Sep-2009 15:40:28] <rmatte> erm, meant
[16-Sep-2009 15:40:33] <mrayzenoss> it's hosted
[16-Sep-2009 15:40:39] <rmatte> ah
[16-Sep-2009 15:41:00] <rmatte> I'm guessing it's not hosted on a dedicated server?
[16-Sep-2009 15:41:22] <mrayzenoss> nah, we're a small install for them
[16-Sep-2009 15:41:28] <rmatte> ah
\n[16-Sep-2009 15:42:26] <rmatte> mhmmm, so I disable an interface\non a test device, and King Crab doesn't even notice that it's down
[16-Sep-2009 15:42:51] <mrayzenoss> are you actually close to 100?
[16-Sep-2009 15:42:57] <mrayzenoss> I've only got 33 myself
\n[16-Sep-2009 15:42:57] <rmatte> can you find out if it's still\nnecessary to use a threshold for monitor interface status via SNMP?
[16-Sep-2009 15:43:06] <rmatte> my impression was that that was going to be native in King Crab
[16-Sep-2009 15:43:24] <rmatte> no, I'm not close to 100, but I can see myself getting there eventually
[16-Sep-2009 15:43:34] <mrayzenoss> dunno, why don't you post that to Zenoss-Testing and I'll have Brandon address it
[16-Sep-2009 15:43:43] <rmatte> ik
[16-Sep-2009 15:43:54] <mrayzenoss> I've bumped it to 300
[16-Sep-2009 15:43:54] <rmatte> k*
[16-Sep-2009 15:44:01] <rmatte> kewel
\n[16-Sep-2009 15:44:45] <rmatte> hmmm, I'm going to try adding\ninterface threshold and see if it updates interface status on the OS\ntab when it sees it as down
[16-Sep-2009 15:44:51] <rmatte> as long as that works I shall be happy
[16-Sep-2009 15:45:52] <mrayzenoss> Zenoss Developers will be here Thursday 11am EDT
[16-Sep-2009 15:46:05] <mrayzenoss> oops
[16-Sep-2009 15:46:16] <mrayzenoss> iancmcc will be here to answer your questions
[16-Sep-2009 15:46:29] <rmatte> hehe
[16-Sep-2009 15:47:03] <rmatte> well, come now Matt, he's not a "real" developer
[16-Sep-2009 15:47:04] <rmatte>
[16-Sep-2009 15:47:23] <iancmcc> I just walked in and sat down one day. They never really hired me, even.
[16-Sep-2009 15:47:26] <mrayzenoss> dunno, how much more real does it get?
[16-Sep-2009 15:47:41] <_Apocalipse_> Dam windows 7
[16-Sep-2009 15:49:46] <rmatte> hehe
\n[16-Sep-2009 15:51:18] <rmatte> interesting... so yeh, it works\nfine as always with the threshold, but the status indicator doesn't\nupdate, I wonder if the transform needs to toggle the status indicator
[16-Sep-2009 15:51:37] <rmatte> I know I could script it, I'm just blown away that it's not working the way I envisioned
[16-Sep-2009 15:53:02] <rmatte> http://dev.zenoss.org/trac/ticket/4997
[16-Sep-2009 15:53:13] <rmatte> What "real time interface events" is he talking about?
\n[16-Sep-2009 15:53:35] <rmatte> Looks like I need to take the\ntransform that I wrote in: http://dev.zenoss.org/trac/ticket/5376 and\nmodify it
[16-Sep-2009 15:53:54] <rmatte> just need to figure out what that status icon value is
[16-Sep-2009 15:54:00] <rmatte> interface.status probably?
[16-Sep-2009 15:55:27] <rmatte> weird
[16-Sep-2009 15:55:38] <rmatte> now I get that UTC error when I do a zenoss status too
[16-Sep-2009 15:55:48] <rmatte> I wonder if that relates to the WMI Data Sources pack
[16-Sep-2009 15:55:51] <rmatte> going to remove it...
[16-Sep-2009 15:56:12] <rmatte> wow, I can't even do zenpack --list anymore
[16-Sep-2009 15:56:13] <rmatte> wtf
[16-Sep-2009 15:56:38] <iancmcc> Hmm, you've got something else wrong
[16-Sep-2009 15:56:55] <iancmcc> This is a fresh install?
[16-Sep-2009 15:56:58] <rmatte> I keep getting AttributeError: 'module' object has no attribute 'UTC'
[16-Sep-2009 15:56:59] <rmatte> a lot
[16-Sep-2009 15:57:20] <rmatte> well, I had 2.4.5 on the box, I did /usr/local/zenoss/uninstall
[16-Sep-2009 15:57:27] <rmatte> then I did a fresh install using the stack installer
[16-Sep-2009 15:57:36] <rmatte> then I installed a bunch of ZenPacks
[16-Sep-2009 15:57:41] <iancmcc> that should be fine, so at least it isn't stale code
[16-Sep-2009 15:58:06] <rmatte> most of the problems I've noticed relate to ZenPacks
[16-Sep-2009 15:58:39] <rmatte> I have a feeling it's that WMI pack or something
[16-Sep-2009 15:58:45] <rmatte> I'm going to try a new fresh install
[16-Sep-2009 15:58:52] <rmatte> this is insane
[16-Sep-2009 15:59:20] <jb> it is a beta..
[16-Sep-2009 15:59:33] <iancmcc> Yeah, but I'm guessing it's just an incompatibility in a particular pack
[16-Sep-2009 16:00:09] <iancmcc> All the Zenoss packs seem to work fine, certainly no problem with the zenpack machinery
[16-Sep-2009 16:03:56] <rmatte> I'm guessing it's the WMI Data Sources pack
[16-Sep-2009 16:04:07] <rmatte> but I'll do trial and error until I figure it out
[16-Sep-2009 16:06:53] <metalklesk> some days ago i asked about nonstop tandem monitoring using zenoss
[16-Sep-2009 16:07:11] <metalklesk> does anyone did that monitoring using zenoss ?
[16-Sep-2009 16:11:04] <twm1010> tandem monitoring of what
[16-Sep-2009 16:11:21] <twm1010> you're talking like... two collectors constantly polling a device for status?
[16-Sep-2009 16:11:32] <mrayzenoss> I think he means Tandem unix systems
[16-Sep-2009 16:12:03] <rmatte> well, I managed to trash the King Crab beta in a couple of hours
[16-Sep-2009 16:12:04] <twm1010> ah... i was thinking distributed monitoring of sorts... speaking of that, does Zenoss do that?
[16-Sep-2009 16:12:05] <rmatte>
\n[16-Sep-2009 16:12:30] <rmatte> twm1010: you can do distributed\ncollectors with core, it's just a pain in the arse to setup
\n[16-Sep-2009 16:12:32] <twm1010> Say... you're doing a URL check\non a live webserver, and have zenoss collectors in 3 places across the\ncountry
[16-Sep-2009 16:12:42] <rmatte> twm1010: enterprise does it as well, much easier to setup
[16-Sep-2009 16:13:00] <twm1010> gotcha
[16-Sep-2009 16:14:05] <rmatte> docs/DOC-2522
[16-Sep-2009 16:15:39] <twm1010> interesting...
\n[16-Sep-2009 16:15:57] <twm1010> but a collector is set for a\nhost... so i'd have to have duplicate hosts to do checks from separate\ncollectors right?
[16-Sep-2009 16:16:04] <twm1010> host = device
[16-Sep-2009 16:16:39] <metalklesk> i was meaning tandem non stop from hp
\n[16-Sep-2009 16:16:55] <rmatte> well, I don't think you could\nreally have the same host monitored from multiple collectors
[16-Sep-2009 16:17:04] <rmatte> maybe if you were monitoring it from different IPs somehow
[16-Sep-2009 16:17:28] <rmatte> If you had completely seperate instances of Zenoss, then you could do it
[16-Sep-2009 16:17:38] <rmatte> and have the instances feed the data back to some central location
[16-Sep-2009 16:17:38] <metalklesk> i think if i can install net-snmp in a tandem non stop OS, zenoss should work fine
[16-Sep-2009 16:17:57] <twm1010> sure...
[16-Sep-2009 16:18:32] <twm1010> Then again, 3 separate zenoss instances would produce the same result just the same
[16-Sep-2009 16:19:25] <twm1010> thats it for me... take care
[16-Sep-2009 16:21:41] <otakup0pe> anyone in here run zenoss with a non-local db
[16-Sep-2009 16:21:50] <otakup0pe> the first time i did that, years ago, it was easy to setup
[16-Sep-2009 16:22:05] <otakup0pe> the last time, a month ago, it somehow went horribly wrong and i had to nuke the zenoss vm ^_^
[16-Sep-2009 16:26:33] <rmatte> nah, we keep all of our dbs local, I assume you're talking about the events db right?
[16-Sep-2009 16:26:54] <iancmcc> Remote ZODB and events db aren't that big a deal
[16-Sep-2009 16:28:51] <metalklesk> i can't find if tandem OS runs net-snmp
[16-Sep-2009 16:30:00] <rmatte> yeh, it's looking like it was the WMI pack that caused the issues
[16-Sep-2009 16:30:10] <rmatte> I didn't install it this time around and things are working better
[16-Sep-2009 16:30:25] <iancmcc> mrayzenoss, we'll probably have to contact the author of that pack
[16-Sep-2009 16:31:19] <mrayzenoss> bigegor's been all over the forums today
[16-Sep-2009 16:31:21] <mrayzenoss> PM him
[16-Sep-2009 16:31:25] <otakup0pe> i'm referring to any mysql db which zenoss uses
[16-Sep-2009 16:32:24] <rmatte> yup, definitely WMI packs
[16-Sep-2009 16:32:33] <rmatte> I shall, and I'll close out the trac tickets that relate to the issue
[16-Sep-2009 16:37:11] <rmatte> mrayzenoss: what's his username?
[16-Sep-2009 16:37:22] <rmatte> nevermind
[16-Sep-2009 16:37:24] <rmatte> found him
[16-Sep-2009 16:37:35] <mrayzenoss> bigegor
[16-Sep-2009 16:38:17] <rmatte> yup
[16-Sep-2009 16:40:46] <rmatte> well that's good, means I can continue with more testing tomorrow
[16-Sep-2009 16:41:06] <rmatte> I'll try to cook up a transform for interface status while I'm at it so that I can test that out
[16-Sep-2009 16:41:50] <rmatte> "Private Message: WMI Data Sources ZenPack doesn't play well with King Crab"
[16-Sep-2009 16:41:52] <rmatte>
[16-Sep-2009 16:47:44] <skipzoid> hello
[16-Sep-2009 16:51:06] <rmatte> mrayzenoss: is there any way to configure a signature for messages and forum postings?
[16-Sep-2009 16:52:04] <mrayzenoss> hmm... I haven't seen that, but for obvious reasons I would like that
[16-Sep-2009 16:52:30] <rmatte> yeh, that was missing from the old forum too
[16-Sep-2009 16:52:38] <rmatte> kind of annoying
[16-Sep-2009 16:52:43] <Brad_K> i believe that the old forum system had that functionality.
[16-Sep-2009 16:52:59] <rmatte> it had it, but it wasn't accessible to regular users
[16-Sep-2009 16:53:01] <mrayzenoss> nah, it was broken
[16-Sep-2009 16:53:07] <rmatte> only Zenoss employees had signatures
[16-Sep-2009 16:53:08] <rmatte> lol
[16-Sep-2009 16:53:09] <mrayzenoss> they had it, it just lied
\n[16-Sep-2009 16:53:10] <Brad_K> i seem to recall some fields\nrelating to signatures, etc... when i first signed up for a user\naccount.
[16-Sep-2009 16:53:23] <mrayzenoss> yeah, the ability to apply them was broken
[16-Sep-2009 16:53:45] <Brad_K> ahh, okay. i knew i saw something, but i didn't pay much attention to it at the time.
[16-Sep-2009 16:54:20] <Brad_K> the current system does have profiles, avatars, pictures, etc... right?
[16-Sep-2009 16:54:32] <rmatte> mrayzenoss: http://www.jivesoftware.com/jivespace/thread/14133
[16-Sep-2009 16:55:16] <mrayzenoss> hmm... that's from over 2 years ago...
[16-Sep-2009 16:55:20] <mrayzenoss> I'll raise the issue
[16-Sep-2009 16:55:28] <rmatte> k
\n[16-Sep-2009 16:58:40] <skipzoid> i'm already runing mysql on the\nlinux server - having installed zenoss core it has installed another\ncopy of mysql within the zenoss tree and it starts - oddly they are\nboth on the same port of 3307 and the existing mysql install doesnt\nhave the databases in it, if I run zenbackup to backup the databases\nthen stop zenoss and disable the zenoss version of the mysql and then\nrestart and do a zenrestore will it populate the datab
[16-Sep-2009 17:00:11]\n<rmatte> skipzoid: it may, but it may also have a problem with\nthe tables not already being there for it to populate the data back in\nto
[16-Sep-2009 17:00:41] <rmatte> skipzoid: http://dev.zenoss.org/trac/browser/branches/zenoss-2.4.x/Products/ZenEvents/db/zenevents.sql
[16-Sep-2009 17:00:44] <rmatte> use that
[16-Sep-2009 17:00:44] <skipzoid> am i right in thinking that I do not or shouldnt run two instances of mysql ?
[16-Sep-2009 17:00:57] <rmatte> probably best to run a single instance on one system
[16-Sep-2009 17:01:18] <rmatte> anyways, that sql file will create the database for you
[16-Sep-2009 17:01:35] <rmatte> databases, rather
[16-Sep-2009 17:01:49]\n<skipzoid> since installing zenoss the system load has gone right\nup and now I'm consistantly using arounf 200m of the swap more than\npreviosly
[16-Sep-2009 17:01:53] <skipzoid> whats the link ?
[16-Sep-2009 17:02:14] <rmatte> it's a link to the sql queries used to actually create the zenoss databases
[16-Sep-2009 17:02:29] <rmatte> so if you run that on your existing SQL server it'll create everything zenoss needs
[16-Sep-2009 17:02:51] <rmatte> I'm assuming you don't have any data that you need to import over from the other db?
[16-Sep-2009 17:03:12] <skipzoid> other?
[16-Sep-2009 17:03:21] <rmatte> the one that Zenoss installed
[16-Sep-2009 17:03:57] <skipzoid> well its got the devices and a few monitors etc, nothing that wouldnt take an hour to reset if need be
[16-Sep-2009 17:04:06] <skipzoid> i thought zenrestore would put it all back?
[16-Sep-2009 17:05:04] <skipzoid> is there a version of that file that is not encapsulated in html ?
[16-Sep-2009 17:11:37] <rmatte> zenrestore will put it back
[16-Sep-2009 17:11:47] <rmatte> but it may need the tables to already be there
[16-Sep-2009 17:11:53] <rmatte> since they are generally created during install
[16-Sep-2009 17:12:04] <rmatte> so I'd advise to run those SQL commands on the SQL server
[16-Sep-2009 17:12:07] <rmatte> then run the restore
[16-Sep-2009 17:12:55] <skipzoid> zenoss@atlantis:~$ zenrestore
[16-Sep-2009 17:12:55] <skipzoid> You must specify either --file or --dir.
[16-Sep-2009 17:12:55] <skipzoid> zenoss@atlantis:~$
\n[16-Sep-2009 17:13:27] <rmatte> mrayzenoss: I think you just\nbumped it to 300 for yourself, I'm still at 100 for message limit
[16-Sep-2009 17:13:45] <mrayzenoss> I'm actually sitting at 34
[16-Sep-2009 17:14:24] <rmatte> for the limit I mean though, some people like to archive important messages
[16-Sep-2009 17:14:31] <mrayzenoss> oh yeah
[16-Sep-2009 17:14:49] <rmatte> hehe
[16-Sep-2009 17:15:41] <rmatte> well, I think I'm going to get out of here shortly
[16-Sep-2009 17:15:50] <rmatte> there is a beer somewhere out there with my name on it
\n[16-Sep-2009 17:16:26] <rmatte> cisco Mibs will be installed by\nthe time I go in tomorrow and then I can do some real testing
[17-Sep-2009 00:00:46] [disconnected at Thu Sep 17 00:00:46 2009]
[17-Sep-2009 00:00:46] [connected at Thu Sep 17 00:00:46 2009]
[17-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
\n[17-Sep-2009 03:29:26] <qdelbul> Hello, I'm trying to make a\nzenpack. I read the developper's documentation but I don't understand\nhow i can add a selfmade plugin that generate an event inside my\nzenpack. Have you a good resource for that ?
[17-Sep-2009 04:36:09] <qdelbul> well, is somebody can help me on this :
\n[17-Sep-2009 04:36:30] <qdelbul> I'm trying to make a zenpack. I\nread the developper's documentation but I don't understand how i can\nadd a selfmade plugin that generate an event inside my zenpack. Have\nyou a good resource for that ?
[17-Sep-2009 06:11:51]\n<chemist> zenoss has stopped sending emails overnight, I can\nstill send a test email but no event notifications are coming through
[17-Sep-2009 07:05:16]\n<chemist> when is the migration of the site cgoing to be\ncomplete? none of the links in the faq work, they all end up going to\nthe same place (community/documentation)
[17-Sep-2009 09:16:39] <ckrough> <crickets>
[17-Sep-2009 09:23:10] <metalklesk> hi
[17-Sep-2009 09:30:32] <metalklesk> one question
[17-Sep-2009 09:31:08] <metalklesk> do i need to install MIB to get the jobs remaining in a printer ?
\n[17-Sep-2009 09:34:50] <chemist> zenoss has stopped sending\nemails overnight, I can still send a test email but no event\nnotifications are coming through
[17-Sep-2009 09:54:58] <Troubadix09> metalklesk: MIBs in zenoss are only use to transform snmp-traps in a readable form
\n[17-Sep-2009 09:57:27] <Troubadix09> metalklesk: try to find out\nthe right OID for your printer, maybe a general OID from the\nPRINTER.MIB or from the MIB of the Manufacturer and poll it
[17-Sep-2009 09:59:37] <Troubadix09> mray: I have a question about templates
[17-Sep-2009 10:00:54] <Troubadix09> mray: is it possible to assign a new template wich is installed with a new ZenPack
[17-Sep-2009 10:01:19] <Troubadix09> mray: and bind to a device-class
[17-Sep-2009 10:01:51] <Troubadix09> mray: that existing devices bind them automatically, such as new devices do?
[17-Sep-2009 10:01:55] <twm1010> chemist: is zenactions daemon running?
[17-Sep-2009 10:02:33]\n<twm1010> troubadix: if you bind a template to a device class,\nall devices in the class receive that binding, unless locally overridden
[17-Sep-2009 10:05:09] <Troubadix09> mray: hm... I think so, but why not in my Zenoss Install? (Virtual Appliance with Zenoss 2.4.5) :-(
[17-Sep-2009 10:05:44] <mrayzenoss> Troubadix09: I don't believe ZenPacks automatically bind to new templates
[17-Sep-2009 10:06:12] <mrayzenoss> err... device classes automatically bind to templates delivered via ZenPack
[17-Sep-2009 10:07:32]\n<rmatte> http://pastebin.com/m7445b476 <--- The parser\ncommands that I'm specifying in this script are completely getting\noverriden by one of the Zenoss imports that I'm doing, can anyone think\nof a way around this?
[17-Sep-2009 10:07:47] <iancmcc> Troubadix09: you can modify the ZenPack to change the bindings of your desired ZenPack
[17-Sep-2009 10:07:48] <rmatte> They work perfectly fine if I put them in a script by themselves
[17-Sep-2009 10:07:58] * aclark waves
[17-Sep-2009 10:08:05] <iancmcc> Troubadix09: I mean the bindings of your desired DeviceClass.
[17-Sep-2009 10:08:07] <mrayzenoss> aclark: howdy
[17-Sep-2009 10:08:46] <iancmcc> rmatte: The option parsing?
[17-Sep-2009 10:09:00]\n<Troubadix09> mray: I know, I bind a template come with the\nWMIPerf_Windows-ZenPack manually to the class /device/Server/Windows
[17-Sep-2009 10:09:06] <rmatte> iancmcc: yeh, it outputs a bunch of other zenoss specific options instead of the ones I want
[17-Sep-2009 10:10:30] <iancmcc> rmatte: Checking, I believe there's a way to stifle ZenScriptBase's option parsing
[17-Sep-2009 10:10:30] <iancmcc> rmatte: pass noopts=True into the ZenScriptBase constructor
[17-Sep-2009 10:10:32] <Troubadix09> mray: when i add new devices to this class, they have bind the template and use it
[17-Sep-2009 10:10:45] <rmatte> iancmcc: thanks, I'll try that
[17-Sep-2009 10:10:56] <Troubadix09> mray: but the existing devices in that class wouldn't
[17-Sep-2009 10:11:22] <rmatte> works like a charm, thanks so much
\n[17-Sep-2009 10:12:29] <mrayzenoss> Troubadix09: hmmm... so\nexisting devices dont' get updated, but new devices pick up the binding\nwhen they're added to the device class?
[17-Sep-2009 10:12:44] <mrayzenoss> have you remodeled the old devices?
[17-Sep-2009 10:12:58] <iancmcc> Troubadix09: Do the existing devices have their own values set?
[17-Sep-2009 10:13:25] <Troubadix09> mray: yes, i remodel
\n[17-Sep-2009 10:15:44] <rmatte> hey iancmcc, one more question,\nif you look at that code again... is there anything special that I need\nto do to set variables based on those options?
[17-Sep-2009 10:15:57] <rmatte> I was under the impression that dest= represented the variable name?
[17-Sep-2009 10:16:05]\n<Troubadix09> iancmcc: there are three templates binded to the\ndevices, two of them are "Local Copy" but not the new template
[17-Sep-2009 10:16:35] <iancmcc> rmatte: That's correct, e.g. options.startDate should have your value
[17-Sep-2009 10:16:47] <rmatte> oh, it's options.startDate
[17-Sep-2009 10:16:50] <rmatte> gotcha, thanks
\n[17-Sep-2009 10:17:38] <iancmcc> Troubadix09: Sounds like they\nhave their own values, which means they won't inherit new template\nbindings from the device class
[17-Sep-2009 10:17:50] <iancmcc> Troubadix09: You'll have to either reset the bindings or add the binding manually
[17-Sep-2009 10:20:56] <Troubadix09> iancmcc: mray: thank you, i will try it manually
[17-Sep-2009 10:22:22] <Troubadix09> iancmcc: I have another question...
[17-Sep-2009 10:25:05]\n<Troubadix09> iancmcc: is it possible to copy Data Sources in a\nTemplate? Because sometimes there are snmp-DataSources where only the\nlast OID-Value is different
[17-Sep-2009 10:25:42] <Troubadix09> iancmcc: and it would be nice to only change the deltas
\n[17-Sep-2009 10:26:14] <metalklesk> does anyone see the memory\n(ram, swap, cache) in zenoss using suse, fedora or ubuntu zenpack ?
[17-Sep-2009 10:26:25] <metalklesk> i mean total, used and free memory
[17-Sep-2009 10:27:10] <Troubadix09> iancmcc: the Menue of the Data-Sources only show "Add, Delete and Add to Graphs"
[17-Sep-2009 10:27:19] <corey> hi
\n[17-Sep-2009 10:28:19] <mrayzenoss> Troubadix09: I guess you'd\nneed to use zendmd: docs/DOC-2532
\n[17-Sep-2009 10:28:31] <ckrough> anyone have a link for\ndownloading OpenManage Mibs? trying to search their site results in\nprinters for sale.
[17-Sep-2009 10:32:37] corey is now known as kmtsun
[17-Sep-2009 10:32:40] <cgibbons> hmmm
[17-Sep-2009 10:32:49] <cgibbons> ian sure got a lot more questions than I did a couple of weeks ago
[17-Sep-2009 10:33:13] <metalklesk> cgibbons, spit it out !
[17-Sep-2009 10:33:14] <Troubadix09> mray: hmm... is this planned to implement in WebGui?
[17-Sep-2009 10:33:46] <rmatte> hehe
[17-Sep-2009 10:34:31] <mrayzenoss> Troubadix09: I'd ask iancmcc
[17-Sep-2009 10:34:44] <iancmcc> Troubadix09: No.
[17-Sep-2009 10:34:46] <ckrough> ouch
\n[17-Sep-2009 10:35:24] <mrayzenoss> Troubadix09: go ahead and\nopen a ticket as an enhancement, I know the templates are supposed to\nget some attention in the Stone Crab release
[17-Sep-2009 10:35:31] <mrayzenoss> and we can discuss it then
[17-Sep-2009 10:35:50] <aclark> Are links to 2.5 beta avail without registering on zenoss.com ?
[17-Sep-2009 10:35:55] <mrayzenoss> yeah
[17-Sep-2009 10:36:06] <mrayzenoss> http://www.zenoss.com/beta/2.5/
[17-Sep-2009 10:36:16] <mrayzenoss> "http://www.zenoss.com/beta/2.5/download"
[17-Sep-2009 10:37:01] <aclark> mrayzenoss: thanks
[17-Sep-2009 10:37:14] <mrayzenoss> aclark: thanks for checking out the beta
[17-Sep-2009 10:37:40] <aclark> mrayzenoss: no prob! excited about it (and zenoss in general as you can tell ;-)
\n[17-Sep-2009 10:39:16] <kmtsun> @aclark4life I'm have trouble\nwith my smtp monitoring, the service is up but I'm getting msgs: IP\nservice smtp is down
[17-Sep-2009 10:40:10] <aclark> kmtsun: i'm just a n00b
[17-Sep-2009 10:41:02] <metalklesk> kmtsun, did you open the port of snmp service in your firewall ?
[17-Sep-2009 10:41:04] <aclark> kmtsun: but there are experts here :-)
[17-Sep-2009 10:41:26] <metalklesk> service up means that you can do a ping to teh device
[17-Sep-2009 10:41:28] <kmtsun> ok thakns
[17-Sep-2009 10:41:37] <kmtsun> no firewall
[17-Sep-2009 10:42:01] <kmtsun> i have it monitoring 30 bsd jails and im only having trouble with 2 of them
[17-Sep-2009 10:42:08] <metalklesk> what port are you using to snmp ? 161 ? 162 ?
[17-Sep-2009 10:42:10] <kmtsun> the rest are showing smtp up
[17-Sep-2009 10:42:15] <metalklesk> mmm
[17-Sep-2009 10:42:23] <metalklesk> sorry i missupderstood
[17-Sep-2009 10:42:32] <metalklesk> *missunderstood
[17-Sep-2009 10:42:45] <metalklesk> snmp != smtp
[17-Sep-2009 10:42:48] <metalklesk>
[17-Sep-2009 10:45:13] <iancmcc> kmtsun: Try telnetting from the zenoss server to narrow it down to zenoss or network
[17-Sep-2009 10:45:23] <kmtsun> ok
[17-Sep-2009 10:46:34] <kmtsun> Trying x.x.x.x...
[17-Sep-2009 10:46:34] <kmtsun> Connected to xxx.xxx.com.
[17-Sep-2009 10:46:34] <kmtsun> Escape character is '^]'.
[17-Sep-2009 10:46:34] <kmtsun> Connection closed by foreign host.
[17-Sep-2009 10:46:55] <iancmcc> you connected to SMTP port, I assume
[17-Sep-2009 10:47:10] <kmtsun> yes telnet <host> 25
[17-Sep-2009 10:47:46] <kmtsun> i did a netstat -al and it is listening on *.* 25
[17-Sep-2009 10:47:58] <iancmcc> I don't know the answer. Try remodeling the box.
[17-Sep-2009 10:48:30] <kmtsun> that means to delete the box from the zenoss db and let it rediscover it"?
[17-Sep-2009 10:48:40] <metalklesk> nop
[17-Sep-2009 10:48:43] <iancmcc> don't need to delete it, just remodel
[17-Sep-2009 10:49:05] <metalklesk> that means -> manage -> model device
[17-Sep-2009 10:49:07] <kmtsun> ok
[17-Sep-2009 10:49:52] <metalklesk> how can i see the used, free, total ram memory ? and swap too
[17-Sep-2009 10:49:59] <kmtsun> thank you .. its a going
[17-Sep-2009 10:50:14] <metalklesk> using snmp no problem, but via ssh i can't see it
[17-Sep-2009 10:53:28] <mrayzenoss> metalklesk: which SSH ZenPacks?
[17-Sep-2009 10:53:42] <mrayzenoss> do you have installed? LinuxMonitor by itself is pretty vanilla
[17-Sep-2009 10:53:58] <metalklesk> mandriva (the one i send you), suse, ubuntu
[17-Sep-2009 10:54:15] <metalklesk> yes, i have linuxmonitor and linuxmonitoraddon installed
[17-Sep-2009 10:54:59] <kmtsun> ack events... remodeled.... smtp red still exist... i will try restarting the jail
\n[17-Sep-2009 10:55:43] <metalklesk> in zproperties i deleted ram\nfrom zFileSystemMapIgnoreTypes as i did with other device monitored by\nsnmp
[17-Sep-2009 10:56:21] <Troubadix09>\nmrayzenoss: thanks, i will open a ticket and we will seen what happend\nin Stone Crab :-)
[17-Sep-2009 10:56:28] <mrayzenoss> metalklesk: are you talking about memory graphs on the Perf tab?
\n[17-Sep-2009 10:56:47] <metalklesk> in the snmp monitored device,\nin filesystem i can see swap and ram (total, used and free) but with\nthe zenpack i can't
[17-Sep-2009 10:57:02] <metalklesk> memory graph AND in filesystem
[17-Sep-2009 10:57:48] <metalklesk> in collector plugins i have zenoss.cmd.linux.memory selected
[17-Sep-2009 10:58:33] <metalklesk> in hardware i can see the total memory and total swap
[17-Sep-2009 10:59:01] <metalklesk> but i can't see the important information of ram and swap in filesystem (total, used and free)
\n[17-Sep-2009 10:59:51] <kmtsun> i resarted the jail.. and\nremodeled... im showing this now: ip service smtp timeout waiting for\n'220'
[17-Sep-2009 11:02:06] <mrayzenoss> metalklesk: my 2.4.5 box monitoring Linux via SNMP has memory and CPU on the Perf tab
[17-Sep-2009 11:02:23] <mrayzenoss> Load Average, CPU Utilization, Memory Utilization, IO
[17-Sep-2009 11:02:28] <metalklesk> yes, i know
[17-Sep-2009 11:02:39] <mrayzenoss> sorry, I thought you were saying those were on the OS tab
[17-Sep-2009 11:02:48] <iancmcc> kmtsun: message/14824#14824
[17-Sep-2009 11:02:52] <metalklesk>
[17-Sep-2009 11:03:12] <mrayzenoss> ok, so none of those graphs are on the Perf tab using /Server/SSH/Linux/ ?
[17-Sep-2009 11:03:22] <mrayzenoss> metalklesk: are the templates all bound?
[17-Sep-2009 11:03:26] <metalklesk> just cpu
[17-Sep-2009 11:03:45] <metalklesk> how can i see that ?
[17-Sep-2009 11:03:57] <iancmcc> kmtsun: This is more informative, actually message/12865#12865
[17-Sep-2009 11:04:17] <motdman> Hello, help with filesystems freeblock Usedblocks AIX
[17-Sep-2009 11:04:37] <motdman> missing rrd files
[17-Sep-2009 11:04:50] <metalklesk> graphs are one thing, but filesystem is another
[17-Sep-2009 11:04:53] <metalklesk> i mean
[17-Sep-2009 11:05:05] <metalklesk> 2 problems i see with ssh based zenpack
[17-Sep-2009 11:05:06] <mrayzenoss> metalklesk: http:/yourserver:8080/zport/dmd/Devices/Server/SSH/perfConfig
[17-Sep-2009 11:05:59] <metalklesk> Name Definition Path Description
[17-Sep-2009 11:05:59] <metalklesk> Device /Devices/Server/SSH
[17-Sep-2009 11:05:59]\n<metalklesk> FileSystem /Devices/Server Filesystem\ntemplate that uses HOST-RESOURCES mib. Has a 90% threshold.
[17-Sep-2009 11:05:59] <metalklesk> IpService /Devices Place holder for future use.
[17-Sep-2009 11:05:59] <metalklesk> OSProcess /Devices Monitors for OSProcess object
[17-Sep-2009 11:06:00] <metalklesk> WinService /Devices Place holder for future use.
[17-Sep-2009 11:06:02] <metalklesk> ethernetCsmacd /Devices Standard ethernet interface template with 75% utilization threshold
[17-Sep-2009 11:06:04] <metalklesk> ethernetCsmacd_64
[17-Sep-2009 11:06:58] <metalklesk> bad idea to copy thos lines, sorry
[17-Sep-2009 11:08:28]\n<metalklesk> this are the templates in the page you said:\ndevice,filesystem,ipservice,osprocess,winservice,ethernetcsmacd,ethernetcsmacd_64
[17-Sep-2009 11:08:44] <mrayzenoss> yeah, I'm installing that ZenPack now to look around
[17-Sep-2009 11:09:05] <metalklesk> thanks !!!
[17-Sep-2009 11:10:24] <motdman> How do I get the filesystem usedblocks freeblocks in AIX
[17-Sep-2009 11:11:28] <motdman> If I use the linux model device on AIX, it is displayed.
[17-Sep-2009 11:11:33] <mrayzenoss> motdman: are you using docs/DOC-3454 ?
[17-Sep-2009 11:12:28] <mrayzenoss> motdman: I've heard very mixed reviews about the quality of the SNMP daemons on AIX
[17-Sep-2009 11:12:43] <mrayzenoss> jb and rocket can attest
\n[17-Sep-2009 11:13:25] <motdman> I worked with IBM to get the\nxmservd mibs to work and all looks good except for filesystem
[17-Sep-2009 11:13:40] <motdman> Actually it is quite impressive regarding the info in place already
[17-Sep-2009 11:14:16] <motdman> You don't need net-snmp
[17-Sep-2009 11:15:22] <motdman> Another thing is that it does't give you the CPU size, only the speed
\n[17-Sep-2009 11:16:40] <motdman> I would use the linux model for\nfilesystem, but I would lose the volumegroup level and hdisk info.
[17-Sep-2009 11:16:50] <Troubadix09> bye
[17-Sep-2009 11:17:31] <iancmcc> see ya troubadix
[17-Sep-2009 11:20:11] <metalklesk> mrayzenoss, did you see the problem i told you with the zenpack ?
\n[17-Sep-2009 11:20:46] <motdman> I used the DOC-3454 but that\npointed me to the cacti setup which is somewhat outdated now.
[17-Sep-2009 11:22:13] <motdman> Why can the linux model pull the aix data I need and not the aix model for filesystem?
[17-Sep-2009 11:23:39] <mrayzenoss> metalklesk: yeah, looks like the defautl LinuxMonitor is just CPU Utilization
[17-Sep-2009 11:24:11] <mrayzenoss> motdman: why not just copy the Linux Device Class and call it AIX?
[17-Sep-2009 11:24:54] <motdman> I could but I would lose the VG LV and hdisk info which is pretty cool stuff.
[17-Sep-2009 11:25:29] <motdman> the Aix model is very cool but just missing that one piece
[17-Sep-2009 11:25:56] <motdman> not sure why it wasn't included in the build, unless I am missing something
[17-Sep-2009 11:26:32] <mrayzenoss> well, it's a community ZenPack, so the author probably didn't run into your use-case
[17-Sep-2009 11:26:51] <motdman> What I do know is that is it not generating the rrd data for this particular issue
[17-Sep-2009 11:27:05] <motdman> The linux model does pull the appropriate rrd...go figure
[17-Sep-2009 11:28:09] <motdman> I would think getting alerts on filesystem free/used would be at the top of the list.
\n[17-Sep-2009 11:28:54] <mrayzenoss> motdman: feel free to dig\ninto the ZenPack, comment on the thread on the page for it or submit a\npatch to the author
[17-Sep-2009 11:30:01] <metalklesk> mrayzenoss, so what can WE do to fix the problem ?
[17-Sep-2009 11:30:04] <metalklesk>
\n[17-Sep-2009 11:30:20] <motdman> Do you know if the author built\nin this functionality or am I not configuring this correctly.
[17-Sep-2009 11:30:39] <mrayzenoss> motdman: no idea, no AIX on my side
[17-Sep-2009 11:31:01] <mrayzenoss> metalklesk: you can submit a patch to the LinuxMonitor or the LinuxMonitorAddon
\n[17-Sep-2009 11:31:11] <motdman> There is not much info on this\nout there...too bad cause it is a great product regarding AIX
[17-Sep-2009 11:31:24] <metalklesk> ok
[17-Sep-2009 11:31:34] <metalklesk> i will investigate how can i fix that
[17-Sep-2009 11:31:36] <metalklesk>
[17-Sep-2009 11:31:58] <mrayzenoss> ok, I'm gonna call that an end to the session, iancmcc is off the hook
[17-Sep-2009 11:33:15] <motdman> Thanks everyone for your help.
[17-Sep-2009 11:39:48] <twm1010> Do we have to re-register for zenoss community forums?
[17-Sep-2009 11:49:33] <cgibbons> anyone here an autoconf guru?
[17-Sep-2009 11:52:19] <rmatte> twm1010: you need to reset your password, but not re-register
[17-Sep-2009 11:52:59] <rmatte> I know what autoconf is, but I doubt that constitutes being a guru
[17-Sep-2009 11:56:24] <metalklesk> rmatte, do you have printers monitored ?
[17-Sep-2009 12:00:49] <rmatte> metalklesk: no, we don't do printers
[17-Sep-2009 12:00:51] <rmatte> we're a noc
[17-Sep-2009 12:02:12] <metalklesk> buuuuu
[17-Sep-2009 12:02:20] <rmatte> lol
[17-Sep-2009 12:02:47] <rmatte> I can't imagine it would be that different from monitoring anything else
[17-Sep-2009 12:02:54] <metalklesk> i've added a printer but i can only see interfaces, no jobs
[17-Sep-2009 12:03:19] <metalklesk> i want to see the queue
[17-Sep-2009 12:03:25] <metalklesk> if it can
[17-Sep-2009 12:03:44] <rmatte> well, are you using any kind of special pack to monitor it or something?
[17-Sep-2009 12:04:06]\n<rmatte> Zenoss isn't designed to see printer queues by default,\nyou'd have to write some sort of collector plugin and add another tab\nin there for printer jobs or whatever
[17-Sep-2009 12:04:11] <rmatte> wouldn't be the easiest thing in the world
[17-Sep-2009 12:05:29] <metalklesk> /Printer/Inkjet
[17-Sep-2009 12:05:38] <metalklesk> that class
[17-Sep-2009 12:05:47] <metalklesk> i have to leave now
[17-Sep-2009 12:05:54] <metalklesk> we talk later, bye !!
[17-Sep-2009 12:10:28] <twm1010> You'd do better just using HP webJetadmin for that stuff
[17-Sep-2009 12:17:42] <kmtsun> anyone have any idea what: HTTP CRITICAL: HTTP/1.1 502 Proxy Error is?
[17-Sep-2009 12:17:56] <kmtsun> or what it the problem may be?
[17-Sep-2009 12:18:41] <rmatte> kmtsun: where are you seeing that?
[17-Sep-2009 12:19:35] <kmtsun> that is an event on one of my jails
[17-Sep-2009 12:19:59] <kmtsun> eventClass: /Status/HTTP
[17-Sep-2009 12:20:40] <rmatte> have you tried googling at all?
\n[17-Sep-2009 12:21:38] <rmatte> Probably a cause similar to:\nhttp://mail-archives.apache.org/mod_mbox/hc-httpclient-users/200506.mbox/%3COF8C8ACECC.29497D1F-ONC125701F.00200C1F-C125701F.0020D09E@de.ibm.com%3Embox/%3COF8C8ACECC.29497D1F-ONC125701F.00200C1F-C125701F.0020D09E@de.ibm.com%3E
[17-Sep-2009 12:29:55] <davetoo> hmm
[17-Sep-2009 12:30:06] <davetoo> maybe my problem is why rc2 is out
[17-Sep-2009 12:30:45] <davetoo> I can't view/edit zProperties
[17-Sep-2009 12:31:04] <iancmcc> what error do you get?
[17-Sep-2009 12:31:04] <rmatte> seriously?
[17-Sep-2009 12:31:23] <rmatte> I was able to view and edit them fine on rc2 just yesterday
[17-Sep-2009 12:31:23] <mrayzenoss> davetoo: are you logged in as a user with those privileges?
[17-Sep-2009 12:31:42] <davetoo> admin, actaully, because I goofed up the ldap
[17-Sep-2009 12:31:44] <davetoo> but...
[17-Sep-2009 12:31:58] <davetoo> I may not have done a sufficiently clean intslall
[17-Sep-2009 12:32:10] <davetoo> I keep forgetting that when you erase the rpms,
[17-Sep-2009 12:32:31] <davetoo> opt/zenoss doesn't actually go away
[17-Sep-2009 12:32:50] <davetoo> so I'm not sure the install ran the migration code,
[17-Sep-2009 12:33:14] <davetoo> I may just clobber it all to oblivion again and start over
[17-Sep-2009 13:12:44] <rmatte> If you're testing rc2, do not install the WMI Data Sources pack
[17-Sep-2009 13:12:55] <rmatte> you'll end up having to reinstall the whole works if you do
[17-Sep-2009 13:15:44] <rmatte> hmmm, I love the new tab completion for objects in zendmd
[17-Sep-2009 13:15:47] <rmatte> quite slick
[17-Sep-2009 13:26:10] * ckrough wants syslog output from zen processes.
[17-Sep-2009 13:49:42] <kmtsun_> exit
[17-Sep-2009 13:49:44] <kmtsun_> end
[17-Sep-2009 13:50:38] <ckrough> noone leaves #zenoss. no one
[17-Sep-2009 13:50:59] <davetoo> I'm fighting ldap
[17-Sep-2009 13:52:01] * Brad_K Everybody was LDAP fighting
[17-Sep-2009 13:52:17] * Brad_K ... Those cats were slow and frightening
\n[17-Sep-2009 13:52:49] <davetoo> I'm probably too tired to be\nlooking at it, basically never slept, but I can't understand why this\none server isn't binding, but the configs and packets look the same
[17-Sep-2009 13:52:50] * ckrough awards point
[17-Sep-2009 13:53:05] <ckrough> davetoo: ldapsearch binds from CLI?
[17-Sep-2009 13:53:45] <davetoo> Haven't tried that yet... not enough brainpower until after I sleep
\n[17-Sep-2009 13:55:58] <ckrough> iancmcc: know of anyone running\ndistributed ZEO with zenoss?\n(http://www.zope.org/Products/ZEO/ZEOFactSheet)
[17-Sep-2009 13:56:31] <iancmcc> Not off the top of my head
[17-Sep-2009 13:56:59] <ckrough> that may be the next step in scaling when ZODB becomes the bottleneck
\n[17-Sep-2009 13:57:29] <ckrough> at least getting Zope Storage on\nits own box, even it its not distributed across multiple. Get it away\nfrom other daemons and mysql
[17-Sep-2009 13:57:43] <ckrough> then cache on SSD drives
[17-Sep-2009 13:58:22] * davetoo waits for enterprise zenpacs to finish installing, again
[17-Sep-2009 13:59:26] <iancmcc> No reason why it wouldn't work, remote ZEO works fine in DistributedCollector setup
\n[17-Sep-2009 13:59:41] <davetoo> I need to make a shell script or\nsomethign to change the colors in the main css sheets; I have so many\ndifferent systems I'm workign with, it's easy to b0rk the wrong one
[17-Sep-2009 13:59:58] <davetoo> maybe easier to hack the logo image
[17-Sep-2009 14:00:53] <ckrough> iancmcc: Im going to pitch it here, if we try it Im sure we'll need to get you guys involved
\n[17-Sep-2009 14:23:18] <kmtsun> in the rc.conf file...for\nsnmpd_flags= ... how do you seperate mutilple flags? with a space,\ncomma.. etc?
[17-Sep-2009 14:32:16] <jb> matt.. the timeout on the community site way too short
[17-Sep-2009 14:44:39] <rmatte> kmtsun: space I believe
[17-Sep-2009 14:45:05] <rmatte> from what I recall from my bsd days
[17-Sep-2009 14:45:31] <rmatte> you might have to enclose them in quotation marks as well
[17-Sep-2009 14:46:19] <rmatte> yeh, I was right, just found an example
[17-Sep-2009 14:46:21] <rmatte> snmpd_flags="-a -p /var/run/snmpd.pid"
[17-Sep-2009 14:46:27] <rmatte>
[17-Sep-2009 14:47:22] <rmatte> jb: timeout?
[17-Sep-2009 14:47:28] <rmatte> jb: I've yet to timeout on it
\n[17-Sep-2009 14:49:29] <rmatte> unhunh, so the WMI Data Sources\npack is totally messed for King Crab according to what Egor tells me.
[17-Sep-2009 14:49:31] <rmatte> Hey,
[17-Sep-2009 14:49:31] <rmatte>
[17-Sep-2009 14:49:31] <rmatte> we have a problem .
[17-Sep-2009 14:49:31] <rmatte>
\n[17-Sep-2009 14:49:31] <rmatte> 1. King Crab had compete new\nZenWin. WMIDataSources used ZenWin classes that no longer exist.
[17-Sep-2009 14:49:32] <rmatte> 2. I'm very busy now with my CCNP certification (until end of next month or so).
[17-Sep-2009 14:49:34] <rmatte>
[17-Sep-2009 14:49:36] <rmatte> Sorry.
[17-Sep-2009 14:49:38] <rmatte>
[17-Sep-2009 14:49:40] <rmatte> Egor
[17-Sep-2009 14:51:05] <rmatte> Figures that the best ZenPack to come out recently doesn't work in the new version
[17-Sep-2009 14:51:30] <mrayzenoss1> well, King Crab won't release for another month at least
[17-Sep-2009 14:51:45] <mrayzenoss1> jb: what was that about a timeout?
[17-Sep-2009 14:52:55] <rmatte> the timeouts might have more to do with his connection or browser than anything
[17-Sep-2009 14:53:04] <rmatte> I can leave the site open for hours, and come back to it, and it's fine
[17-Sep-2009 14:57:50] <davetoo> how odd,
[17-Sep-2009 14:58:05] <rmatte> ?
\n[17-Sep-2009 14:58:13] <davetoo> I wonder if these ldap servers\nare smart enough to lock my manager account out by individual source ips
[17-Sep-2009 14:58:38] <davetoo> talking to myself mostly, sorry
[17-Sep-2009 14:59:11] <rmatte> ah
[17-Sep-2009 14:59:30] <rmatte> I've never seen LDAP do anything like that
[17-Sep-2009 14:59:32] <davetoo> I was up all night working on a dmd script, very very tired
[17-Sep-2009 14:59:36] <davetoo> brain fuzzy
[17-Sep-2009 14:59:55] <davetoo> When it clears up I need to submit a patch for BatchDeviceLoader
[17-Sep-2009 15:00:00] <rmatte> yeh, I made the mother of all dmd scripts just put the finishing touches on it today
[17-Sep-2009 15:00:13] <davetoo> oh, you should see mine
[17-Sep-2009 15:00:17] <rmatte> hehe
[17-Sep-2009 15:00:23] <ckrough> rmatte: do share
[17-Sep-2009 15:00:54] <davetoo> evals expressions from command-line to build devfiler instances for the device searchers
[17-Sep-2009 15:01:00] <rmatte> http://pastebin.com/m4ba445b4
[17-Sep-2009 15:01:02] <rmatte> there's mine
[17-Sep-2009 15:01:07] <ckrough> Im working on a common module with reusable xmlrpc functions for building device lists, etc...
[17-Sep-2009 15:01:07] <ckrough> python that is
[17-Sep-2009 15:01:10] <rmatte> yours might be even more complex hehe
[17-Sep-2009 15:01:16] <davetoo> Basicaly what this one does is pull data useful for the new zenbatchloder
[17-Sep-2009 15:01:21] <ckrough> Ill get it to you when it's clean and documented a bit
[17-Sep-2009 15:01:25] <rmatte> ckrough: I think you've seen this one
[17-Sep-2009 15:01:30] <ckrough> it's bad right now, still in the works
[17-Sep-2009 15:01:32] <rmatte> I just added commandline options to it today
[17-Sep-2009 15:01:32] <davetoo> what's this one do?
[17-Sep-2009 15:01:35] <davetoo> oh
[17-Sep-2009 15:01:47] <davetoo> I havent' tackled xml/rpc yet
[17-Sep-2009 15:01:55] <rmatte> it generates a device availability report, but it's a standalone script, more flexible
[17-Sep-2009 15:01:58] <ckrough> ah this one
[17-Sep-2009 15:02:18] <rmatte> I also made stand-alone memory, cpu, and interface report scripts
[17-Sep-2009 15:02:25] <rmatte> we're going to use them to generate automated reports
[17-Sep-2009 15:02:39] <ckrough> whats the driving force behind this? doesnt zenoss handle most of that natively?
[17-Sep-2009 15:02:42] <rmatte> that availability script was a pain in the ass to write
[17-Sep-2009 15:02:53] <rmatte> all I had to go on was a single SQL query that Matt emailed to me
[17-Sep-2009 15:02:58] <rmatte> had to figure it out from there
[17-Sep-2009 15:03:13] <rmatte> So it was mostly a guessing game
[17-Sep-2009 15:03:27] <rmatte> but I managed to 100% duplicate the report, the results come out the exact same
[17-Sep-2009 15:03:30] <rmatte>
[17-Sep-2009 15:04:04] <rmatte> improved my python quite a bit along the way too
[17-Sep-2009 15:04:19] <ckrough> yeah
[17-Sep-2009 15:04:22] <ckrough> same with me
[17-Sep-2009 15:04:38] <davetoo> back in a minute... have to reboot
[17-Sep-2009 15:04:55] <rmatte> k
[17-Sep-2009 15:05:34]\n<rmatte> ckrough: it has native reports for this stuff, but we\nwant to be able to fully automate the reporting process
[17-Sep-2009 15:06:32]\n<rmatte> basically we'll have a daily average of availability\ndumped daily for each client, then at the end of the month we'll dump\nall of the reports for the month, then send that to the client along\nwith a list of top 10 devices/interfaces for utilization as well as a\ngraph showing overall availability from day to day during the month
[17-Sep-2009 15:06:37] <rmatte> and it'll all be automated
[17-Sep-2009 15:06:48] <rmatte> right now we're doing it by hand every month and it's extremely tedious
[17-Sep-2009 15:06:58] <rmatte> especially since the export function for the reports is broken in Zenoss
[17-Sep-2009 15:07:15]\n<rmatte> you think you're exporting for the period that you\nspecified but then it goes and exports for the last 7 days
[17-Sep-2009 15:07:33] <rmatte> so by creating external scripts we are eliminating the possibility of incorrect data
[17-Sep-2009 15:07:43] <rmatte> plus, the reports are going to stay the same regardless of Zenoss version
[17-Sep-2009 15:07:49] <rmatte> there are a lot of good reasons for these
\n[17-Sep-2009 15:13:57] <twm1010> is there any way to get zenoss\nto resolve hostnames of devices currently named by IP post-discovery?
[17-Sep-2009 15:14:14] <rmatte> twm1010: nah, you need to rename them by hand
[17-Sep-2009 15:14:26] <rmatte> twm1010: unless you whip up a zendmd script to do it
[17-Sep-2009 15:14:28] <mrayzenoss1> twm1010: snmp will rename them I believe
[17-Sep-2009 15:14:42] <mrayzenoss1> or maybe that's dns
[17-Sep-2009 15:15:00] <twm1010> ah... so i'd have to connect to every ILO card and/or RSA2 mgmt card and give it an actual hostname
[17-Sep-2009 15:15:01] <rmatte> I've never seen snmp rename them
[17-Sep-2009 15:15:02] <twm1010> RAR.
\n[17-Sep-2009 15:15:20] <rmatte> and I've never tried adding\nanything by IP when we have DNS as that defeats the purpose of having\ndns
[17-Sep-2009 15:15:21] <rmatte>
[17-Sep-2009 15:15:29] <twm1010> Ah.. I just need to do it by hand there really aren't that many.
\n[17-Sep-2009 15:15:44] <mrayzenoss1> I've got a OSX box that\nalways shows up as "elzar.local" and is monitored via SNMP. I can't\nfigure any other way it would get named that
[17-Sep-2009 15:16:02] <davetoo> So what breaks if we set the default snmp version, at the root of /devices, to v2c?
[17-Sep-2009 15:16:03] <twm1010> rmatte: see... i did my discovery against an incomplete DNS server, so alot of nodes got added by IP
[17-Sep-2009 15:16:13] <twm1010> davetoo: devices that require v1
[17-Sep-2009 15:16:22] <mrayzenoss1> twm1010: hah
[17-Sep-2009 15:16:34] <ckrough> rmatte: I see
[17-Sep-2009 15:16:50] <davetoo> and are those worse than the more modern ones that don't work with v1?
[17-Sep-2009 15:17:09] <twm1010> up to you, i change it a little further down the chain
[17-Sep-2009 15:17:20] <twm1010> my ../Windows is v2c, but others are not
[17-Sep-2009 15:17:26] <twm1010> like /Power is v1
[17-Sep-2009 15:17:35] <twm1010> because of crappy old UPS devices
\n[17-Sep-2009 15:18:20] <twm1010> but, i'm also debating not\nadding the ILO cards as separate devices, but rather manually adding\nthat IP to the server it manages
[17-Sep-2009 15:18:30] <twm1010> and then locking the interface table
\n[17-Sep-2009 15:18:35] <davetoo> I know I've run into weird\nbehavior, like the initial discovery wont' work with v2c, but continued\nmonitoring won't work withouth it...
[17-Sep-2009 15:20:48] <davetoo> What is the benefit/use of teh Default Admin Role setting for auser?
[17-Sep-2009 15:21:09] <iancmcc> Little, if you're not running Enterprise.
[17-Sep-2009 15:21:33] <davetoo> I am, so it must be an ACL thing I haven't learned yet...
\n[17-Sep-2009 15:25:49] <iancmcc> It is indeed. When you're adding\nadministered objects to a user, it's the default role they'll have
[17-Sep-2009 15:26:02] <iancmcc> Handy when you've got a bunch to add to somebody. Changing them all by hand is terrible.
\n[17-Sep-2009 15:42:45] <davetoo>\nzRouteMapCollectOnlyIndirect \n \nzRouteMapCollectOnlyLocal
[17-Sep-2009 15:42:52] <davetoo> I'm confused about these now
[17-Sep-2009 15:43:04] <davetoo> I thought it needed indirect routes to build the topology
[17-Sep-2009 15:55:32] <davetoo> hmm
[17-Sep-2009 16:05:18] <davetoo> zenbatchloader is not yet wonderful for new adds
[17-Sep-2009 16:05:28] <davetoo> but it's great for moving systems from my old core boxes
[17-Sep-2009 16:17:47] <kmtsun> anyone every seen this zenoss;snmp mesg: Unable to read processes on device
[17-Sep-2009 16:40:40] <LW> out of curiosity... has anybody created a ESX path checking plugin yet?
[17-Sep-2009 16:40:46] <LW> s/a/an
[17-Sep-2009 16:42:07]\n<caseyh2007> so who knows what is up with source build and\nFreeBSD? I am working on a port of Zenoss 2.4.5 right now, should I\nwait for 2.5 to come out?
[17-Sep-2009 16:42:37] <mrayzenoss1> caseyh2007: I see you saw my tweet
[17-Sep-2009 16:42:42] <caseyh2007> lol, yeah
[17-Sep-2009 16:42:44] mrayzenoss1 is now known as mrayzenoss
[17-Sep-2009 16:44:11] <mrayzenoss> so cgibbons is working on fixing http://dev.zenoss.org/trac/ticket/4994
[17-Sep-2009 16:44:56] <mrayzenoss> and so far he's got the pysamba stuff cleaned up, at least for Ubuntu 9.04
[17-Sep-2009 16:45:05] <mrayzenoss> http://dev.zenoss.org/svn/sandboxen/cgibbons/4994/
[17-Sep-2009 16:45:17] <mrayzenoss> but he could use testers for other platforms
[17-Sep-2009 16:45:26] <mrayzenoss> the fixes are intended for the upcoming 2.5 release
[17-Sep-2009 16:45:31] <cgibbons> tryin' it on Fedora 10 right now
[17-Sep-2009 16:45:49] <mrayzenoss> cgibbons: build completed for my Ubuntu 9.04
[17-Sep-2009 16:45:58] <cgibbons> cool
\n[17-Sep-2009 16:46:30] <mrayzenoss> caseyh2007: so if your\nFreeBSD box is ready, you can help test that part of the build if you\nwant
[17-Sep-2009 16:46:48] <mrayzenoss> actually that goes for anyone who wants to help test source builds
[17-Sep-2009 16:47:19] <mrayzenoss> just 'svn co http://dev.zenoss.org/svn/sandboxen/cgibbons/4994"
[17-Sep-2009 16:47:32] <mrayzenoss> and 'make pywmi-build'
[17-Sep-2009 16:47:44] <davetoo> w00t!
[17-Sep-2009 16:48:05] <caseyh2007> so you guys are going to make the source build work on FreeBSD with no changes in king crab?
[17-Sep-2009 16:48:15] <davetoo> I think I just fixed zenbatchloader and BatchDeviceLoader
[17-Sep-2009 16:48:25] <mrayzenoss> no, we want to fix the build before King Crab
\n[17-Sep-2009 16:48:52] <mrayzenoss> right now cgibbons is working\non a branch, but if it cleans up everything it'll go into 2.5 (King\nCrab)
[17-Sep-2009 16:49:05] <mrayzenoss> err, s/sandbox/branch/
[17-Sep-2009 16:49:14] <mrayzenoss> 2.5 is currently SVN trunk
[17-Sep-2009 16:49:35] <davetoo> ah
[17-Sep-2009 16:49:54] <mrayzenoss> 2.5 is a few weeks from branching
[17-Sep-2009 16:51:02] <mrayzenoss> the only platforms known to work right now for source builds are Debian, RHEL and OSX 10.5
[17-Sep-2009 16:51:14] <caseyh2007> k, well im gonna go ahead and make a port for 2.4.5 for now.
[17-Sep-2009 16:51:33] <mrayzenoss> caseyh2007: ok, just letting you know 2.5 is scheduled for next month
[17-Sep-2009 16:51:58] <caseyh2007> k, ill update when a new source build is out =)
[17-Sep-2009 16:54:21] <cgibbons> boom on fedora 10
[17-Sep-2009 16:54:27] <mrayzenoss> good boom?
[17-Sep-2009 16:54:32] <caseyh2007> lol
[17-Sep-2009 16:54:40] <cgibbons> nope
[17-Sep-2009 16:55:03]\n<mrayzenoss> caseyh2007: we don't have any FreeBSD boxes, so if\nyou could take a shot on that pysamba part, just building it, that\nwould help
[17-Sep-2009 16:55:21] <mrayzenoss>\nof course, if it doesn't work for Fedora 10, it's not looking good for\nother platforms
[17-Sep-2009 16:56:58] <cgibbons> yeah, lots of little changes
[17-Sep-2009 16:57:03] <cgibbons> there's some license issues i have to work out, too
[17-Sep-2009 16:57:17] <mrayzenoss> ugh
\n[17-Sep-2009 16:57:22] <cgibbons> but, hey! i just fixed\nsomething fun on the Windows stuff that's been bugging people for a\nwhile
[17-Sep-2009 16:57:31] <cgibbons> s/fixed/figured out/
[17-Sep-2009 17:32:24] <cgibbons> ah cool, Fedora 10 was easy to fix
[17-Sep-2009 17:32:41] <mrayzenoss> you want me to regression test Ubuntu?
[17-Sep-2009 17:32:53] <mrayzenoss> I'm thinking about doing my OSX 10.6 clean install tonight
[17-Sep-2009 17:32:59] <mrayzenoss> I could test there too
[17-Sep-2009 17:33:04] <cgibbons> nah already did (ubutu)
\n[17-Sep-2009 17:41:40] <cgibbons> alrighty, that branch is\nupdated. if you want to update do a 'make distclean' in the\nSamba/source directory first.
[17-Sep-2009 17:51:16] <mrayzenoss> branch or your sandbox?
[17-Sep-2009 17:52:28] <mrayzenoss> make: *** No rule to make target `gtk/config.mk', needed by `Makefile'. Stop.
[17-Sep-2009 17:53:12] <mrayzenoss> I'm building again
[17-Sep-2009 17:54:22] <cgibbons> my sandbox branch yeah
[17-Sep-2009 17:56:36] <davetoo> I tried doing a vendor branch for a bit
[17-Sep-2009 17:56:42] <davetoo> because i was doing a lot of workarounds
[17-Sep-2009 17:56:47] <davetoo> couldn't keep it up
[17-Sep-2009 18:00:42] <mrayzenoss> cgibbons: build still works
[17-Sep-2009 18:00:50] <mrayzenoss> I'm checking out for awhile
[17-Sep-2009 19:38:41] <metalklesk> hi
[17-Sep-2009 19:39:04] <davetoo> hmm
[17-Sep-2009 19:39:14] <davetoo> maybe I sholdnt have put 2.5 on my development systemyet
[17-Sep-2009 19:40:23] <davetoo> 'fact I think I'll revert
[17-Sep-2009 20:25:38] <metalklesk> one question
[17-Sep-2009 20:25:52] <metalklesk> i'm trying WBEM in Linux
[17-Sep-2009 20:26:27] <metalklesk> but in perf i can only see Load average and Memory utilization
[17-Sep-2009 20:26:36] <metalklesk> what about the other graphs ?
[17-Sep-2009 20:27:27] <metalklesk> i already have installed WBem linux monitor
[17-Sep-2009 20:27:44] <metalklesk> and bind the WBEMDevice to the device
[17-Sep-2009 20:28:56] <metalklesk> mmmm missing RRD file
[17-Sep-2009 21:31:44] <davetoo> I don't know how I managed this but now I cannot log out of zenoss
[17-Sep-2009 21:31:59] <davetoo> well, admin can't
[17-Sep-2009 21:32:42] <davetoo> oh maybe it's beucase I was in the top-level zmi
[17-Sep-2009 21:33:28] <davetoo> what
[17-Sep-2009 21:56:13] <davetoo> I jus tkilled my zope admin account
[17-Sep-2009 21:56:58] <davetoo> d'oh
[18-Sep-2009 00:00:46] [disconnected at Fri Sep 18 00:00:46 2009]
[18-Sep-2009 00:00:46] [connected at Fri Sep 18 00:00:46 2009]
[18-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[18-Sep-2009 07:21:12] <vinzpr> hi
[18-Sep-2009 07:21:17] <ckrough> hi
[18-Sep-2009 07:21:39] <vinzpr> need help on zenprocess anyone encounter this error on zenoss 2.4.5
\n[18-Sep-2009 07:22:34] <vinzpr> ERROR:zen.zenprocess:[Failure\ninstance: Traceback (failure with no frames):\nexceptions.UnboundLocalError: local variable 'update' referenced before\nassignment
[18-Sep-2009 07:22:35] <vinzpr> ]
[18-Sep-2009 07:22:35] <vinzpr> Traceback (most recent call last):
[18-Sep-2009 07:22:35] <vinzpr> File "/opt/zenoss/Products/ZenUtils/Driver.py", line 49, in _next
[18-Sep-2009 07:22:35] <vinzpr> self.iter.next().addBoth(self._finish)
[18-Sep-2009 07:22:36] <vinzpr> File "/opt/zenoss/Products/ZenRRD/zenprocess.py", line 474, in start
[18-Sep-2009 07:22:38] <vinzpr> self.updateDevices(driver.next(), devices)
[18-Sep-2009 07:22:40] <vinzpr> File "/opt/zenoss/Products/ZenRRD/zenprocess.py", line 459, in updateDevices
[18-Sep-2009 07:22:42] <vinzpr> d.updateConfig(cfg)
[18-Sep-2009 07:22:44] <vinzpr> File "/opt/zenoss/Products/ZenRRD/zenprocess.py", line 247, in updateConfig
[18-Sep-2009 07:22:46] <vinzpr> update.name)
[18-Sep-2009 07:22:48] <vinzpr> UnboundLocalError: local variable 'update' referenced before assignment
[18-Sep-2009 07:22:50] <vinzpr> INFO:zen.zenprocess:Daemon zenprocess shutting down
[18-Sep-2009 07:22:52] <vinzpr> INFO:zen.zenprocess:zenprocess shutting down
[18-Sep-2009 07:22:52] <ckrough> please use pastebin.com for that stuff
[18-Sep-2009 07:23:22] <ckrough> Does that happen everu start you start zenprocess?
[18-Sep-2009 07:23:30] <vinzpr> http://pastebin.ca/1543144
[18-Sep-2009 07:23:34] <vinzpr> yes
[18-Sep-2009 07:23:47] <vinzpr> it happens when i start zenprocess
[18-Sep-2009 07:24:12] <ckrough> looks like this has been hapenning for a while?
[18-Sep-2009 07:25:09] <vinzpr> any solution on this kind of error
[18-Sep-2009 07:25:17] <ckrough> not off the top of my head
[18-Sep-2009 07:25:29] <ckrough> did you modify anything with an 'update' variable?
[18-Sep-2009 07:25:37] <ckrough> and is it a new install?
[18-Sep-2009 07:25:39] <vinzpr> nope
[18-Sep-2009 07:26:15] <vinzpr> i update it for 2weeks already and it only appears this day
\n[18-Sep-2009 07:26:17] <cgibbons> something like that might\nhappen if zenhub hasn't been restarted after the upgrade or a zenpack\nchange, etc, but you'd likely be seeing more errors round the system too
[18-Sep-2009 07:26:55] <vinzpr> i restarted the zenhub
[18-Sep-2009 07:27:04] <vinzpr> but its still happen
[18-Sep-2009 07:28:26] <ckrough> Are the METAL templates zope objects or .pt files?
[18-Sep-2009 07:29:02] <vinzpr> im not sure when can i find that .pt files?
[18-Sep-2009 07:29:24] <ckrough> sorry vinzpr, I was asking a question for myself there
[18-Sep-2009 07:29:40] <ckrough> vinzpr: not related to your error
[18-Sep-2009 07:29:58] <vinzpr> ok
[18-Sep-2009 07:30:39] <vinzpr> anyone has an idea on this error http://pastebin.ca/1543144
[18-Sep-2009 07:32:59] <ckrough> answered my own question. templates.pt on the file system
[18-Sep-2009 07:33:08] <ckrough> vinzpr: sorry, Im not sure on that one
[18-Sep-2009 07:33:42] <vinzpr> ok..anyone encounter a problem like this?
[18-Sep-2009 07:33:50] <vinzpr> on zenprocess
[18-Sep-2009 07:33:52] <ckrough> first Ive seen it
[18-Sep-2009 07:34:08] <vinzpr> and...?
[18-Sep-2009 07:35:02] <ckrough> meaning, this is the first time Ive heard of that error, and I dont know
[18-Sep-2009 07:35:44] <vinzpr> ok
[18-Sep-2009 08:29:08]\n<ckrough> When I override \n/zport/portal_skins/zenui/img/onwhitelogo.png by customizing it, the\noriginal image still gets displayed.
[18-Sep-2009 08:30:16] <iancmcc> browser cache?
[18-Sep-2009 08:31:18] <ckrough> I tried clearing it. Ill try another machine
[18-Sep-2009 08:31:36] * ckrough doesnt trust firefox cache clearing
[18-Sep-2009 08:32:47] <ckrough> nah, not that
\n[18-Sep-2009 08:33:40] <iancmcc> I don't trust firefox cache\nclearing either, I have Web Developer Toolbar installed and just work\nwith my cache disabled.
[18-Sep-2009 08:34:03] <iancmcc> Hit up /zport/portal_skins/manage and make sure custom is at the top
[18-Sep-2009 08:34:35] <iancmcc> Sorry, go to the Properties tab and look at the layers textarea
[18-Sep-2009 08:34:36] <ckrough> nope, it's near the bottom
[18-Sep-2009 08:34:40] <ckrough> k
[18-Sep-2009 08:34:56] <ckrough> its first.
\n[18-Sep-2009 08:36:39] <ckrough> other new (non-override) images\nI put in there work, but that method would require customizing the\ntemplates, which is obviously bad mojo
[18-Sep-2009 08:36:45] <ckrough> very strange...
[18-Sep-2009 08:37:40] <iancmcc> yeah, that don't make no sense
[18-Sep-2009 08:37:54] <iancmcc> OH wait
[18-Sep-2009 08:37:56] <iancmcc> it won't work
[18-Sep-2009 08:38:05] <iancmcc> sorry, it's because it's in a folder
[18-Sep-2009 08:38:17] <ckrough> I don't follow
[18-Sep-2009 08:38:24] <iancmcc> so you have to override /zenui/img
[18-Sep-2009 08:38:30] <iancmcc> which is obviously not doable
[18-Sep-2009 08:39:18] <ckrough> so the individual components of the folder cant be overwritten
[18-Sep-2009 08:39:21] <iancmcc> Right.
[18-Sep-2009 08:39:35] <ckrough> oh, are you referring to folder as in filesystem directory?
[18-Sep-2009 08:39:38] <iancmcc> An old mistake that we have not been able to change
[18-Sep-2009 08:39:52] <iancmcc> well, yes, technically
[18-Sep-2009 08:40:11] <iancmcc> but the problem is Zopey in nature.
[18-Sep-2009 08:40:25] <iancmcc> once it resolves img, you're squarely in the zenui skin
[18-Sep-2009 08:40:37] <iancmcc> it doesn't know how to merge subfolders of different skin folders.
[18-Sep-2009 08:40:45] <ckrough> ah, img is the object, and zope is unaware of its contents?
[18-Sep-2009 08:40:56] <ckrough> err, they arent 'objects'?
\n[18-Sep-2009 08:41:28] <iancmcc> They're all "objects", including\nimg, but its url resolution isn't smart enough to straddle two\nsubfolders.
[18-Sep-2009 08:41:56] <ckrough> we've passed my comfort level with zope inheritance. I'll take your word for it
\n[18-Sep-2009 08:41:56] <iancmcc> Anyway, the solutions are\nequally silly: change the template, or symlink your image on the\nfilesystem
[18-Sep-2009 08:42:13] <ckrough> going to change the template.
[18-Sep-2009 08:42:33] <ckrough> thanks
\n[18-Sep-2009 08:42:38] <iancmcc> This will be fixed in Stone\nCrab, incidentally. Wonderful to have an opportunity to fix things.
[18-Sep-2009 08:42:53] <ckrough> which is stone crab? 2.7?
[18-Sep-2009 08:43:02] <iancmcc> the one after 2.5.
[18-Sep-2009 08:43:19] <iancmcc> Which might be 2.6, might be 3.0.
[18-Sep-2009 08:43:36] <iancmcc> Depending on how bold we feel at release planning.
[18-Sep-2009 08:43:42] <ckrough> gotcha
\n[18-Sep-2009 08:46:51] <iancmcc> Also, I forget who was asking\nfor it, rmatte or ckrough, but both row shading in the event detail\npane and the unacknowledge feature are now in trunk.
[18-Sep-2009 08:51:07] <ckrough> iancmcc: you do a lot of the UI stuff?
[18-Sep-2009 08:51:37] <iancmcc> Pretty much all of it. I've been typecast.
[18-Sep-2009 08:52:41] <ckrough> tracing the template calls and css is fun
[18-Sep-2009 08:52:46] <ckrough> onwhitelogo is elusive
[18-Sep-2009 08:52:56] <ckrough> everything pointed to head1, but no avail
[18-Sep-2009 08:54:21] <iancmcc> sadly, there are two places
[18-Sep-2009 08:54:31] <iancmcc> GraphReport.py and viewMultiGraphReportClean.pt
[18-Sep-2009 08:54:35] <ckrough> hmm, I was wrong. head1 is actually <head> info
[18-Sep-2009 08:54:37] <iancmcc> that is one old image
[18-Sep-2009 08:54:50] <ckrough> viewGraphReportClean is my starting point
[18-Sep-2009 08:55:01] <ckrough> possible in GraphReport.py, Ill look
[18-Sep-2009 08:57:35] <ckrough> bah, this requires a zopectl restart doesnt it
[18-Sep-2009 08:57:53] <iancmcc> yes
[18-Sep-2009 08:57:54] <ckrough> due to the pyc
[18-Sep-2009 08:57:56] <ckrough> kk
[18-Sep-2009 08:58:27] <ckrough> I think GraphReport.py was the one. I was searching under skins and didnt catch it
[18-Sep-2009 09:07:19] <ckrough> bah, pyo file
[18-Sep-2009 09:07:46] <ckrough> apparently pyos dont get rebuilt automatically like pyc's
[18-Sep-2009 09:13:05] <jb> iancmcc: that was me that asked about row shading..
[18-Sep-2009 09:13:25] <iancmcc> Well, it's in.
[18-Sep-2009 09:13:28] <jb> thanks
[18-Sep-2009 09:20:07] <ckrough> iancmcc: ha, its not GraphReport.py either
[18-Sep-2009 09:20:33] <iancmcc> I grepped all the code, that's the only place it appears
[18-Sep-2009 09:20:33] <ckrough> iancmcc: oh well, moving on to other things. Ill have to come back to this one
[18-Sep-2009 09:20:36] <iancmcc> lemme check enterprise
[18-Sep-2009 09:20:38] <ckrough> yeah, same
[18-Sep-2009 09:20:46] <ckrough> and I opened it on another box with clear cache
\n[18-Sep-2009 09:20:56] <etank> we are using zenoss enterprise and\nfor some reason we have stopped getting datapoints in our graphs for\nmssql. this may have happened after a failover (active/passive\nclustered systems). what can be done to start getting data again?
[18-Sep-2009 09:21:31] <jb> you are monitoring the cluster virtual IP or the individual cluster nodes?
[18-Sep-2009 09:22:24] <etank> jb in this case it is the cluster virtual ip
[18-Sep-2009 09:22:37] <jb> hrm yeah, you may need to delete those RRD's
[18-Sep-2009 09:22:50] <etank> we are monitoring the physical nodes as well though
[18-Sep-2009 09:22:55] <etank> ok
[18-Sep-2009 09:22:59] <etank> i will give that a try
[18-Sep-2009 09:23:17] <jb> just curious.. why monitor the virtual ip?
[18-Sep-2009 09:23:29] <jb> if you are already monitoring the physical nodes?
[18-Sep-2009 09:23:50] <etank> so that the graphs are specific to the clustered services
\n[18-Sep-2009 09:24:13] <etank> we dont want to have to figure out\nfirst which node is active and then go figure look at the graphs
[18-Sep-2009 09:25:52] <jb> k
\n[18-Sep-2009 09:42:25] * etank is having a hard time trying to decide\nif he should go to the Zenoss talke that mrayzenoss doing at OLF or the\nreStructured Text:
[18-Sep-2009 09:42:28] <etank> Plain Text Gets
[18-Sep-2009 09:42:31] <etank> one
[18-Sep-2009 09:43:08] <etank> (that paste did not do what i wanted it to)
[18-Sep-2009 09:47:18] <mrayzenoss> etank: there's the free training on Friday, and a demo on Saturday
[18-Sep-2009 09:47:33] <mrayzenoss> and a party Friday night
[18-Sep-2009 09:47:42] <ckrough> mrayzenoss: any news on a recast of Jane's presentation? I got pulled away...
[18-Sep-2009 09:48:01] <mrayzenoss> Tuesday
[18-Sep-2009 09:48:07] <mrayzenoss> I'll go do the announcement
[18-Sep-2009 09:48:15] <ckrough> thanks
[18-Sep-2009 09:55:45] <rmatte> ckrough: it was a pretty good presentation, did you miss it?
[18-Sep-2009 09:59:50] <ckrough> yeah, typical... 10 minutes in and I had to deal with something
[18-Sep-2009 10:00:05] <ckrough> I need to hide in a broom closet with a laptop to get through a whole webex
[18-Sep-2009 10:02:36]\n<etank> mrayzenoss: yeah we will be there for the training so i\nwill prolly do the reStructured Text talk on Saturday
[18-Sep-2009 10:03:36] <rocket> rmatte: ping
[18-Sep-2009 10:03:48] <rmatte> rocket: yus?
[18-Sep-2009 10:04:11] <rocket> mrayzenoss: tells me you run solarwinds and zenoss
[18-Sep-2009 10:04:28] <rocket> I have a solarwinds demo i am attending in a few and was wondering if you had an opinion
[18-Sep-2009 10:04:33] <rmatte> well, we are backing away from using solarwinds apparently
[18-Sep-2009 10:04:51] <rocket> oh?
[18-Sep-2009 10:04:58] <rocket> any reason in particular?
[18-Sep-2009 10:05:14]\n<rmatte> being an MSP, when we found out that we'd have to pay\nfor each collector (which wasn't communicated to us in the beginning)\nit just became way too expensive
[18-Sep-2009 10:05:39] <rmatte> so we would have ended up with using 1 collector with some monstrosity of a NATting setup
[18-Sep-2009 10:05:42] <rmatte> waste of time
[18-Sep-2009 10:06:01] <rocket> makes sense ... aside from that how was it technically?
[18-Sep-2009 10:06:22] <rocket> eg would you still recommend zenoss over it etc ..
[18-Sep-2009 10:06:47]\n<rmatte> it was alright, they are definitely going for a whole\n"drop it in and it just works approach", but then again, if there's\nsomething it doesn't happen to do it's quite difficult to get it to do\nwhatever that is
[18-Sep-2009 10:07:11] <rmatte> I would definitely recommend Zenoss over it for everything except the reporting
[18-Sep-2009 10:07:20] <rmatte> which is all we were really using solarwinds for
[18-Sep-2009 10:07:43] <rmatte> solarwinds' reports system is awesome
[18-Sep-2009 10:08:00]\n<mrayzenoss> ckrough: how's this URL:\nhttps://zenossevents.webex.com/ec0605l/eventcenter/enroll/join.do?confViewID=579483694&confId=579483694&siteurl=zenossevents&path=program_detail&theAction=detail
[18-Sep-2009 10:08:19] <mrayzenoss> I'm trying to find a direct link to the event, WebEx is annoyingly hiding everything via Javascript
[18-Sep-2009 10:08:25]\n<rmatte> but now that I'm getting the hang of writing external\npython scripts to gather data so that we can fully automate reporting\nfrom Zenoss, it's not becoming such an issue
[18-Sep-2009 10:08:56] <rmatte> with your python skills I doubt that solarwinds would give you much of an edge
[18-Sep-2009 10:11:20] <rmatte> Solarwinds is more for the lazy admin in my opinion
\n[18-Sep-2009 10:14:10] <mrayzenoss> anyone interested in the\nEvent Management Session, please test this URL for me:\nhttps://zenossevents.webex.com/ec0605l/eventcenter/enroll/join.do?confViewID=579483694&confId=579483694&siteurl=zenossevents&path=program_detail&theAction=detail
[18-Sep-2009 10:16:23] <ckrough> no good
[18-Sep-2009 10:16:26] <ckrough> the last one worked though
\n[18-Sep-2009 10:17:15] <ckrough> good link:\nhttps://zenossevents.webex.com/ec0605l/eventcenter/enroll/join.do?confViewID=579483694&confId=579483694&siteurl=zenossevents&path=program_detail&theAction=detail
[18-Sep-2009 10:17:28] <mrayzenoss> ok, I'll put that in the blog
[18-Sep-2009 10:19:08] <mrayzenoss> published
[18-Sep-2009 10:19:08] <cgibbons> hey matt, guess what?
[18-Sep-2009 10:19:12] <mrayzenoss> what's that?
[18-Sep-2009 10:20:26] <cgibbons> that latest code builds on freebsd
[18-Sep-2009 10:21:16] <mrayzenoss> woohoo!
[18-Sep-2009 10:21:34] <mrayzenoss> so... Ubuntu, FreeBSD and Fedora are all good?
[18-Sep-2009 10:23:25] <ckrough> CentOS 5.4 should be out in a week or two. have you tested on RHEL5.4?
[18-Sep-2009 10:24:40] <kmtsun> how do you turn off auto-discover?
[18-Sep-2009 10:26:35] <kmtsun> would that be: Managment -> settings -> jobs?
[18-Sep-2009 10:27:29] <mrayzenoss> ckrough: just checked with Dev mgmt, we'll QA RHEL 5.4 before shipping Enterprise 2.5
[18-Sep-2009 10:28:08] <ckrough> I was just wondering if you had done core because CentOS5.4 will release right around the same time
[18-Sep-2009 10:28:18] <ckrough> and I assume a good number of people run CentOS with zenoss core
[18-Sep-2009 10:28:31] <mrayzenoss> definitely
[18-Sep-2009 10:28:49] <ckrough> kmtsun: /networks zproperies -> zAutoDiscover = true|false
[18-Sep-2009 10:28:52] <mrayzenoss> RHEL/CentOS are 90% of Enterprise installs
[18-Sep-2009 10:29:03] <mrayzenoss> probably around 40-50% of Core
[18-Sep-2009 10:29:16] <jb> cent here..
[18-Sep-2009 10:30:07] <rmatte> I think a lot of core is Ubuntu
[18-Sep-2009 10:30:13] <mrayzenoss> yeah
[18-Sep-2009 10:31:49] <mrayzenoss> we don't know the breakdown for distros, only for platforms
[18-Sep-2009 10:31:53] <mrayzenoss> and kernel versions
[18-Sep-2009 10:40:08] <kmtsun> if you turn off autodiscovery in zenoss will it stop monitoring discovered machines via snmp wmi ssh etc?
[18-Sep-2009 10:42:30] <rmatte> kmtsun: not sure what you mean, turn off autodiscovery where?
\n[18-Sep-2009 10:42:59] <rmatte> there's no active auto-discovery\nthat I'm aware of, you have to manually specify what subnet you want\nzenoss to auto-discover on
[18-Sep-2009 10:43:32] <kmtsun> right, I don't want zenoss to find new clients... just moninitor the ones it found
[18-Sep-2009 10:43:49] <ckrough> kmtsun: right, if you turn off autodisc it will stop finding new devices
[18-Sep-2009 10:44:37] <kmtsun> but my question is .. will it stop monitoring 'discovered' devices? ie ssh wmi snmp
[18-Sep-2009 10:44:44] <rmatte> ckrough: what option is he referring to? (I don't use autodisc, so no idea)
[18-Sep-2009 10:44:45] <ckrough> no
[18-Sep-2009 10:44:51] <kmtsun> if auto is turned off
[18-Sep-2009 10:45:01] <ckrough> rmatte: check the zpropteries of blocks in the /Networks class
\n[18-Sep-2009 10:45:23] <perr0> morning all, when trying to\ninstall from source on Gentoo I am getting an error message and it\nfails building wmi-something. the forums suggest to revert back to\nautoconf-2.59 which I have but what I don't understand is which\napplication is where x listed here >\ndocs/DOC-2607
[18-Sep-2009\n10:45:24] <ckrough> kmtsun: right, autodiscovery is just for\nfinding new devices, zenoss will continue to monitor existing devices
[18-Sep-2009 10:45:56] <kmtsun> thank you
[18-Sep-2009 10:46:01]\n<rmatte> ckrough: ohhhh, yeh, but that just means that Zenoss\nwill autodiscover on that subnet when an auto-discovery is done, it\ndoesn't mean that it's going to actively go out searching that subnet\nevery so many hours
[18-Sep-2009 10:46:27] <perr0> I should say the dependancy "where x" listed in the list I posted
\n[18-Sep-2009 10:46:31] <rmatte> kmstun: Zenoss monitors anything\nin a production state based on the Device zProperties, not the network\nzProperties
[18-Sep-2009 10:48:18] <kmtsun> i\nwant things in porduction state monitored... just don't want it to find\nnew devices... i want to do that myself
[18-Sep-2009 10:49:17] <kmtsun> so i should use the zproperties of /Network class?
[18-Sep-2009 10:49:19] <ckrough> rmatte: you sure?
[18-Sep-2009 10:49:31] <ckrough> kmtsun: were not positive
[18-Sep-2009 10:49:38] <etank> i am not seeing any SQL related rrd files being built now for one of my MS SQL servers
[18-Sep-2009 10:49:49] <etank> i have even deleted the device and re-added it
[18-Sep-2009 10:49:59] <etank> it has been more than 15 minutes too
\n[18-Sep-2009 10:50:12] <ckrough> rmatte: my understanding was the\nthe autodiscover process looked at those network zProperties when\nrunning on a cycle
[18-Sep-2009 10:52:44]\n<rmatte> kmtsun: unless you explicitly tell it to go out and do a\ndiscovery though, it won't do it on it's own, is what I was saying
[18-Sep-2009 10:53:10] <kmtsun> i see thanks
\n[18-Sep-2009 10:53:33] <rmatte> so if that's what you're worried\nabout then that's not an issue, there is absolutely no automated\nauto-discovery built in to Zenoss, you have to explicitly tell it to do\nany sort of auto-discovery
[18-Sep-2009 10:55:44]\n<perr0> So does anyone at this time know which application is\n"where x>22." listed in the depandancies list >\ndocs/DOC-2607
[18-Sep-2009 10:56:23] <mrayzenoss> perr0: the MySQL version must be 5.0.22 or greater
\n[18-Sep-2009 10:57:28] <ckrough> rmatte: ha, I do it via 'zenoss'\nuser crontab... been that way so long I forgot its not standard
[18-Sep-2009 11:02:01] <perr0> mrayzenoss: thanks that is a little confusing having it on seperate lines
[18-Sep-2009 11:02:27] <mrayzenoss> yeah, we're still working out the kinks
[18-Sep-2009 11:09:51]\n<ckrough> I have this one switch that keeps coming up snmp down,\nbut it can be walked and tcpdump shows the snmpstatus check going out\nand coming back
[18-Sep-2009 11:11:21] <ckrough> tried clearing events and clearning heartbeats... everything.
\n[18-Sep-2009 11:14:48] <mrayzenoss> anyone interested in a\nZenoss-sponsored Developer training course? \nmessage/39558
[18-Sep-2009 11:14:54] <mrayzenoss> Looking for feedback
[18-Sep-2009 11:17:04] <Brad_K> mrayzenoss: is there a Zenoss-sponsored Admin training course?
[18-Sep-2009 11:18:17] <mrayzenoss> yeah, there's Zenoss training http://www.zenoss.com/support#training
[18-Sep-2009 11:18:20] <etank> mrayzenoss: i am interested in any training course
[18-Sep-2009 11:18:29] <mrayzenoss> and we'll have free training at Ohio LinuxFest next week
\n[18-Sep-2009 11:18:56] <mrayzenoss> Brad_K: assuming the Texas\nLinuxFest takes off, we hope to throw some free training into that as\nwell
[18-Sep-2009 11:19:38] <mrayzenoss> and LISA and SCALE will have free training
[18-Sep-2009 11:21:32] <etank> what can i look at to find out why my rrd files are not being built?
[18-Sep-2009 11:24:33] <mrayzenoss> bad graphs, empty graphs or no graphs?
[18-Sep-2009 11:24:46] <etank> no rrd files so no graphs at all
[18-Sep-2009 11:25:00] <etank> for the MSSQL related graphs that is
[18-Sep-2009 11:25:01] <mrayzenoss> is there a placeholder for the graphs?
[18-Sep-2009 11:25:10] <etank> yes there is a placeholder
[18-Sep-2009 11:25:33] <mrayzenoss> that usually indicates an issue between the data source and the data points
[18-Sep-2009 11:25:44] <etank> i have tried restarting zenoss and running zenocommand but nothing
[18-Sep-2009 11:26:05] <etank> and i have recently deleted and re-added the device
[18-Sep-2009 11:26:28] <mrayzenoss> does the 'Test' button work?
[18-Sep-2009 11:26:34] <etank> checking now
[18-Sep-2009 11:27:27] <etank> i get
[18-Sep-2009 11:27:28] <etank> WARNING:zen.winperf.PerfRpc:Bad counter: \\SQLServer:Latches\\Latch Waits/sec
[18-Sep-2009 11:27:28] <etank> INFO:zen.winperf:Unable to collect value for \\SQLServer:Latches\\Latch Waits/sec
[18-Sep-2009 11:28:31] <etank> it does work against a different mssql server though
\n[18-Sep-2009 11:40:35] <Brad_K> mrayzenoss: I definitely support\nmore training, if you can make it happen. i'm now on the team\ndeploying zenoss here at UT Austin/ITS, so I'd welcome any additional\ntraining that i can get.
[18-Sep-2009 11:40:36]\n<russb> i've seen perfmon counters dissapear before...can you\ncheck on that sql server (using perfmon.exe) and verify that the\ncounter is there?
[18-Sep-2009 11:41:09]\n<Brad_K> etank: we're having similar problems with zencommand\ncounters mysteriously disappearing and then reappearing hours later.
[18-Sep-2009 11:42:06]\n<Brad_K> mrayzenoss: in fact, i'm just about to file requests\nwith my management for zenoss training, books, instructional videos,\netc....
[18-Sep-2009 11:43:51] <Brad_K> I did\nfind one book on Amazon that is specifically about zenoss, namely\n"Zenoss Core Network and System Monitoring: A step-by-step guide to\nconfiguring, using, and adapting this free Open Source network\nmonitoring system". Does anyone know if this is any good?
[18-Sep-2009 11:46:06] <Brad_K> For that matter, is the book "Python for Unix and Linux System Administration" any good?
[18-Sep-2009 11:55:40] <twm1010> Brad_K: I've read that book. It is good, but beware that its for Zenoss 2.2 I believe"
[18-Sep-2009 11:55:48] <twm1010> I wrote a review on it, never got around to sending it to the author.
[18-Sep-2009 11:56:05] <twm1010> Slow reading of the admin guide will get you about the same.
\n[18-Sep-2009 11:56:52] <twm1010> Shortly here I intend to focus\non how to get the most information from each brand of server from top\nto bottom... IE.... an HP server, making sure it has an ILO and insight\nmanager correctly configured to get maximum info back to zenoss
[18-Sep-2009 12:09:06] <ckrough> mrayzenoss: yes on the course
\n[18-Sep-2009 12:11:47] <ckrough> Brad_K: just catching up, Jane\nCurry's site has some good stuff on it too\nhttp://www.skills-1st.co.uk/mgt/
[18-Sep-2009 12:12:04] <etank> is there a way to use TALES expressions in a WinPerf datasource?
\n[18-Sep-2009 12:12:29] <etank> Brad_K: i failed over the server\nand started getting data again. something on the local box was the issue
[18-Sep-2009 12:13:00] <etank> im trying to do something like this in the datasource
[18-Sep-2009 12:13:05] <etank> \\SQLServer:Databases(${here/cDatabaseName})\\Data File(s) Size (KB)
[18-Sep-2009 12:13:31] <etank> it works if i hard code the database name in the perf counter line
[18-Sep-2009 12:17:54] <Brad_K> mrayzenoss: also, is there a Zenoss user's group here in Austin?
[18-Sep-2009 12:21:20] <twm1010> on that note, is there a list of all of them?
[18-Sep-2009 12:30:14] <Brad_K> for that matter, are there any Zenoss user groups anywhere?
[18-Sep-2009 12:31:02] <twm1010> Brad_K: I would think not, I would imagine Zenoss gets discussed alot at Linux user groups though.
[18-Sep-2009 12:31:16] <mrayzenoss> sorry, I was at lunch the last 45 minutes
\n[18-Sep-2009 12:31:35] <twm1010> I find that I have trouble\ngetting Linux users interested in it for some reason. I just don't\nunderstand it.
[18-Sep-2009 12:31:56]\n<mrayzenoss> there aren't any Zenoss User Groups that I'm aware\nof, that said there are a large number of Zenoss users (and employees)\nin Austin
[18-Sep-2009 12:32:00] <twm1010> Especially anyone who has put time into Nagios or Hobbit, I can't even get them to look at it.
\n[18-Sep-2009 12:32:41] <mrayzenoss> the Zenoss book from Packt is\ngetting a bit dated, it was written around the 2.2 time frame
\n[18-Sep-2009 12:33:29] <ckrough> I havent gotten much buy-in even\non in-house groups. we have different groups using it but everyone's\nneeds are so varied there isn't much of a rallying point to get excited\nabout
[18-Sep-2009 12:33:48] <mrayzenoss>\nBrad_K: we've usually offered training in Annapolis, but we could\nobviously host it in Austin
[18-Sep-2009 12:33:58] <mrayzenoss> since our office is up on Spicewood Springs
[18-Sep-2009 12:34:03] <mrayzenoss> and there's plenty of room
\n[18-Sep-2009 12:34:12] <Brad_K> twm1010: i'm in most of the Linux\nand Unix-related groups here in Austin, and I co-founded the local\nAustin chapter of the League of Professional System Administrators. \nwe've had Matt come give a presentation on Zenoss before, but that's\nall i've ever seen at any of the groups.
[18-Sep-2009 12:34:13] <ckrough> some people here would drive up matt
[18-Sep-2009 12:34:36] <mrayzenoss> Brad_K: yeah, I did a demo for the Python User Group too
[18-Sep-2009 12:34:59] <mrayzenoss> ok, I'll see about updating the training stuff to mention Austin as an option
\n[18-Sep-2009 12:35:36] <Brad_K> that would be very helpful for me\nand my management, since we've been told that it's going to be\nvirtually impossible to get any training approved that will require\ntravel and hotel costs.
[18-Sep-2009 12:35:45] <ckrough> but I live flying back east for training. you're messing with my vacations now.
[18-Sep-2009 12:35:48] <ckrough>
[18-Sep-2009 12:35:57] <venturaville> mrayzenoss: are any of you guys going to make LISA ?
[18-Sep-2009 12:36:05] <Brad_K> of course, we've also got a support contract, but i don't know if that covers training.
[18-Sep-2009 12:36:05] <mrayzenoss> yeah, we'll have a big presence at LISA
[18-Sep-2009 12:36:28] <venturaville> I and one other fellow from here are going to be up there
[18-Sep-2009 12:36:31] <mrayzenoss> Brad_K: it's usually included in contracts
\n[18-Sep-2009 12:36:31] <Brad_K> yes, they're doing a one-day\ncommunity session, plus other stuff. See\nhttp://www.zenoss.com/news/events
[18-Sep-2009 12:36:52] <mrayzenoss> Chet has a Guru session at LISA, we'll have a day of free training
[18-Sep-2009 12:37:03] <mrayzenoss> and we'll have a booth
\n[18-Sep-2009 12:37:31] <Brad_K> i know all about the stuff that's\nlinked from that page, because i just sent an e-mail message to my\nmanagement talking about all the training and other instructional\nmaterials that i'm going to need, if i'm going to be able to help in\nthis process.
[18-Sep-2009 12:37:37] <venturaville> I'm going to try and make the puppet and zenoss guru sessions
[18-Sep-2009 12:37:57] <ckrough> Puppet is coming to Austin in November
[18-Sep-2009 12:38:02] <ckrough> I should be at that
[18-Sep-2009 12:38:21] <mrayzenoss> yeah, we may be attending the Puppet training as well
[18-Sep-2009 12:38:24] <Brad_K> we're going to have a presentation on bcfg2 at the LOPSA-Austin meeting next week.
\n[18-Sep-2009 12:38:53] <Brad_K> and if that works out, i'll see\nif i can bring the speaker back to give the same presentation at the\nnext CACTUS meeting.
[18-Sep-2009 12:39:30] <mrayzenoss> Brad_K: yeah, I've been meaning to do a presentations for CACTUS, AustinLUG and ALE
[18-Sep-2009 12:39:43] <mrayzenoss> hard to get free time at nights
\n[18-Sep-2009 12:39:46] <Brad_K> that is, if you're interested in\nthe general topic of configuration management, and not just\nspecifically Puppet.
[18-Sep-2009 12:39:47] <mrayzenoss> (2 kids)
[18-Sep-2009 12:39:56] <mrayzenoss> Brad_K: which night?
\n[18-Sep-2009 12:40:21] <Brad_K> mrayzenoss: anytime you are\nwilling or able to do a presentation for CACTUS, please let me know. \ni'm the program chair, and i'm the one responsible for organizing the\nspeakers. ;-)
[18-Sep-2009 12:41:26] <venturaville> what is CACTUS?
\n[18-Sep-2009 12:41:30] <Brad_K> CACTUS meets on the third\nthursday of the month, and the recent meetings have been at a Church\nnear 43rd and B Avenue, where one of the board members is also a member\nof the church (probably a deacon or something).
\n[18-Sep-2009 12:41:46] <etank> mrayzenoss: do you know if there\nis a way to use TALES expressions in a WinPerf datasource?
\n[18-Sep-2009 12:41:56] <Brad_K> CACTUS is the Capital Area\nCentral Texas Unix Society, and is the oldest Unix user group in Austin.
[18-Sep-2009 12:42:11] <twm1010> that's quite an acronym
[18-Sep-2009 12:42:23] <mrayzenoss> etank: I'd assume there is, I'll double check
[18-Sep-2009 12:42:27] * mrayzenoss doesn't run Enterprise
[18-Sep-2009 12:42:33] <Brad_K> they came up with it years ago.
[18-Sep-2009 12:42:45] <etank> im trying something like this --> \\SQLServer:Databases(${here/cDatabaseName})\\Data File(s) Size (KB)
[18-Sep-2009 12:43:06] <etank> but it isnt changing the cher/cDatabasename to what it should be
[18-Sep-2009 12:43:08]\n<Brad_K> LOPSA-Austin meets on the fourth Tuesday nights of the\nmonth, and right now we're trying to nail down the location.
[18-Sep-2009 12:43:21] <mrayzenoss> Brad_K: my 3rd Thursday of the month is bad the next 3 months... January?
[18-Sep-2009 12:43:36] <ckrough> Brad_K: is there a date for the bcfg2 talk? dont see it on the website
[18-Sep-2009 12:43:45] <Brad_K> We could plan for January, sure. that would give us plenty of time to get the word out.
[18-Sep-2009 12:44:07] <Brad_K> the LOPSA-Austin website needs to be updated. we've been holding off doing that, pending the location.
\n[18-Sep-2009 12:44:55] <Brad_K> but Travis Campbell (who is the\nother co-founder of LOPSA-Austin) has arranged the speaker, and we've\ngot several irons in the fire with regards to the location.
[18-Sep-2009 12:45:28] <mrayzenoss> ok, I've put January 21 on my calendar, pencil me in
[18-Sep-2009 12:46:02]\n<Brad_K> Ironically, Travis is also on the LOPSA Board, and I\nbelieve he's in the process of taking over the position of the local\nUSENIX representative on UT campus. So, they'll send him to one USENIX\nconference per year, for free.
[18-Sep-2009 12:46:28]\n<Brad_K> I'll let everyone on the CACTUS board know, but you\nmight want to send me a reminder as we get closer. Thanks!
\n[18-Sep-2009 12:49:49] <etank> mrayzenoss: trying a few different\nthings it seems that zenwinperf passes the --testCounter as is
[18-Sep-2009 12:49:58] <etank> no substitution
[18-Sep-2009 12:55:46] <daMaestro> anyone know of a nginx zenpack that utilizes stub_status output
[18-Sep-2009 12:55:47] <daMaestro> ?
[18-Sep-2009 12:56:54] <cgibbons> etank: yeah nothing in the data sources does tales expressions.
[18-Sep-2009 13:03:10] <metalklesk> hi
[18-Sep-2009 13:09:08] <daMaestro> guess not (from googling for a while)
[18-Sep-2009 13:09:28] <daMaestro> i'll be creating a zenpack based on the stub_status output
[18-Sep-2009 13:09:42] <dollarbang> Does Zenoss have the ability to push configurations to Routers/Switches?
\n[18-Sep-2009 13:11:33] <ckrough> dollarbang: you could probably\nwrite something, but I haven't seen and docs or zenpacks for that
[18-Sep-2009 13:11:42] <ckrough> dollarbang: you can *pull* them with Rancid
\n[18-Sep-2009 13:12:59] <dollarbang> ckrough: thanks. These are\nCisco Routers and Switches, and our license is coming up for renewal.\nMy company is looking for any savings they can get.
\n[18-Sep-2009 13:14:23] <Brad_K> hmm. my understanding was that\nrancid could both pull and push configurations to/from cisco\nequipment. i guess i need to go do some googling....
[18-Sep-2009 13:15:29] <etank> cgibbons: dang. ok thanks.
\n[18-Sep-2009 13:15:55] <ckrough> Very possible. I havent seen\nthat myself. There are docs for pulling configs in response to events\n(wr trap, etc..) on the zenoss site
[18-Sep-2009 13:16:01]\n<Brad_K> hmm. okay, the page at http://www.shrubbery.net/rancid/\nseems to be pretty clear -- monitoring, not management.
[18-Sep-2009 13:16:09] <etank> kinda sucks to have to create seperate datasources with hard coded stuff then to get what you need.
[18-Sep-2009 13:16:36] <ckrough> docs/DOC-2716
[18-Sep-2009 13:17:28] <Brad_K> blargh.
[18-Sep-2009 13:17:50] <Brad_K> gotta watch which window i'm typing in which command.
[18-Sep-2009 13:21:35] <rmatte> lol
\n[18-Sep-2009 13:22:44] * Brad_K also did not know that MacIrssi\ntranslates "<command>-W" into "close this irc channel and make\nthat tab go away"
[18-Sep-2009 13:36:15]\n<eidolon> hi folks - we're having a real problem here and need to\ngetthis fixed before hte weekend. we have zenoss enterprise, we've\nWMI-modelled some win2k8 (i believe) servers. The filesystems are\nshowing up (Total bytes), ut Used bytes and Free bytes and %util are\nshowing unknown.
[18-Sep-2009 13:36:30]\n<eidolon> we've waited until a few monitoring cycles have passed,\nbut the numbers have not magically appeared.
\n[18-Sep-2009 13:36:43] <eidolon> help please. these are\nproduction servers we've had outages on becasuse of disk space\nmonitoring failures
\n[18-Sep-2009 13:37:23] <ckrough> sounds like something you shoudl\nsubmit a support ticket for? assuming you have paid support as\nenterprise customer
[18-Sep-2009 13:37:42]\n<Brad_K> I'm sure the more experience people on the channel would\nneed to know what version of Zenoss Enterprise you're using, and what\nzenpacks you're using.
[18-Sep-2009 13:37:56] <Brad_K> and definitely open a ticket as well.
\n[18-Sep-2009 13:37:59] <eidolon> yes, but my login is failing -\nthis needs to betimely though. i'm going to try to recover the\npassword and get in, but i have a short timewindow.
[18-Sep-2009 13:38:29] <lsacpinbal> what template you using?
[18-Sep-2009 13:38:37] <lsacpinbal> is it WMI?
[18-Sep-2009 13:38:40] <eidolon> yes WMI.
[18-Sep-2009 13:38:41] <lsacpinbal> or just snmp
[18-Sep-2009 13:38:44] <eidolon> Zenoss 2.4.1
[18-Sep-2009 13:38:45] lsacpinbal is now known as pinball_
[18-Sep-2009 13:38:49] <eidolon> no snmp
[18-Sep-2009 13:39:28] <eidolon> Performance templates: Device WMI and SQLServer
[18-Sep-2009 13:40:07]\n<Brad_K> we're not yet in production with our Zenoss system, but\nso far the only thing we've found that works reliably is the SNMP\nstuff. I don't think we've tried to do any WMI stuff yet, and we're\nhaving major problems with anything running under zencommand.
[18-Sep-2009 13:40:25] <pinball_> yeah i agree with Brad
[18-Sep-2009 13:40:32] <pinball_> we are just working on WMI on our dev site
[18-Sep-2009 13:40:41] <eidolon> :-/
[18-Sep-2009 13:40:43] <pinball_> snmp seems to be the best opetion in this case
[18-Sep-2009 13:40:47] <pinball_> can you configure snmp service
[18-Sep-2009 13:40:50] <pinball_> on your windows host
[18-Sep-2009 13:40:58] <eidolon> i've been told several times that for windows boxes, we should be using WMI.
[18-Sep-2009 13:41:03] <pinball_> use both
[18-Sep-2009 13:41:35] <pinball_> at least with SNMP the service will send a trap when things start to break
[18-Sep-2009 13:41:36] <cgibbons> pinball_/brad are you guys enteprise uses or core with community zenpacks?
\n[18-Sep-2009 13:41:39] <Brad_K> and yes, we have a ticket open on\nthis problem. i've got 3GB of logs that i've made available to them,\nso that they can help us figure out what is going on.
[18-Sep-2009 13:41:39] <pinball_> and you can always relay on that
[18-Sep-2009 13:41:43] <pinball_> enterprise
[18-Sep-2009 13:41:56] <pinball_> but i am still on 2.3.3
[18-Sep-2009 13:41:58] <eidolon> my snmp experience is pretty limited.
[18-Sep-2009 13:42:05] <ckrough> the consensus has been that WMI is more reliable for windows hosts
[18-Sep-2009 13:42:08] <Brad_K> we're using Zenoss Enterprise 2.4.2, i think.
[18-Sep-2009 13:42:11] <pinball_> dev i have 2.4.5
[18-Sep-2009 13:42:18] <pinball_> where i am playing with WMI
\n[18-Sep-2009 13:42:18] <eidolon> thoguh, apparently my cohort\nhere has deleted, re-added, and remodelled this host, and we just got\n'used ytes' and 'free bytes' showing up
[18-Sep-2009 13:42:30] <metalklesk> one question
[18-Sep-2009 13:42:31] * eidolon loves sketchy results.
\n[18-Sep-2009 13:42:35] <Brad_K> one of my tasks is to figure out\nhow to start working on doing the upgrade to 2.4.5, next week.
[18-Sep-2009 13:42:54] <metalklesk> when i build a new command (settings -> commando -> add new command)
[18-Sep-2009 13:43:18] <metalklesk> that command is executed in server or client side ?
[18-Sep-2009 13:43:32] <metalklesk> for example, "rpm -qa"
\n[18-Sep-2009 13:43:38] <Brad_K> in theory, WMI is what you're\nsupposed to use for Windows. But check earlier archives of the logs of\nthis channel, because there were some major problems with WMI being\nreported by one of the Zenoss guys who was testing it with an upgraded\nversion of Zenoss.
[18-Sep-2009 13:44:08] <heckj> where can you find the logs for this channel?
[18-Sep-2009 13:44:18] <eidolon> :-/
[18-Sep-2009 13:44:19] * eidolon nods
[18-Sep-2009 13:44:32] <cgibbons> got any more context, brad? that would be news to me...
[18-Sep-2009 13:44:46] <eidolon> i'd like to see an official statement from zenoss stating which is suggested.
[18-Sep-2009 13:44:53] <eidolon> not that i don't trust you guys
[18-Sep-2009 13:44:58] <pinball_> haha
[18-Sep-2009 13:45:06]\n<eidolon> but i can't say "i'm going to change the mointoring on\n30 hosts because omsoen thought they saw something it he backscroll.
[18-Sep-2009 13:45:10] <pinball_> i just have had good luck with configuring both SNMP and WMI on windows boxes
[18-Sep-2009 13:45:13] <pinball_> and when i add the host
[18-Sep-2009 13:45:21] <pinball_> i don't use the add device link on the left side
\n[18-Sep-2009 13:45:24] <Brad_K> the official word is probably\nstill going to be to use WMI, because this was testing with a new\nversion of Zenoss.
[18-Sep-2009 13:45:27] <pinball_> i navigate to the device tree
[18-Sep-2009 13:45:38] <pinball_> devices->server->windows
[18-Sep-2009 13:45:42] <mrayzenoss> metalklesk: either
[18-Sep-2009 13:45:49] <pinball_> and in the drop down more add->device
[18-Sep-2009 13:45:54] <pinball_> and use snmp to discover
[18-Sep-2009 13:45:59] <ckrough> a lot of the SNMP/Windows problems I've seen reported were problems with snmp-informant
[18-Sep-2009 13:46:08] <ckrough> I dont monitor and windows hosts so not sure the specifics
[18-Sep-2009 13:46:14] <twm1010> i've got a box right now that probably need a reboot for informant to work
[18-Sep-2009 13:46:18] <twm1010> very odd.
[18-Sep-2009 13:46:22] <pinball_> we don't use informant
\n[18-Sep-2009 13:46:33] <twm1010> i consider switching to WMI, but\ni've not touched it in a while, and i'm unsure WHICH wmi zenpack to\nlook at
[18-Sep-2009 13:46:41] <ckrough> pinball_: ya. enterprise has a wmi zenpack or agent or something right?
[18-Sep-2009 13:46:42] <twm1010> also curious how the filesystems are monitoring via WMI
[18-Sep-2009 13:46:47] <pinball_> right
[18-Sep-2009 13:46:55] <pinball_> but there is a good zenpack
[18-Sep-2009 13:47:00] <pinball_> called WMIDevice
[18-Sep-2009 13:47:01]\n<Brad_K> unfortunately, all the zenoss community links that used\nto work are now broken, thanks to the new Jive system they're using.
[18-Sep-2009 13:47:03] <pinball_> in the community group
[18-Sep-2009 13:47:06] <pinball_> that works well
[18-Sep-2009 13:47:15] <metalklesk> mrayzenoss, i want to run "rpm -qa" command in the device i'm monitoring
[18-Sep-2009 13:47:33] <pinball_> ZenPacks.community.WMIDataSource community Egor Puzanov 1.1 Yes
[18-Sep-2009 13:47:41] <metalklesk> how teh command should be ? rpm -qa ? or something else ?
[18-Sep-2009 13:47:42] <pinball_> ZenPacks.community.WMIPerf_Windows community Egor Puzanov 1.0 Yes
\n[18-Sep-2009 13:48:33] <mrayzenoss> metalklesk: when you add the\nCommand data source, there's an option about using SSH, that determines\nwhether the command is local or remote
[18-Sep-2009\n13:48:48] <Brad_K> see Mark Hinkle's blog post at\nblogs/zenossblog/2009/09/03/winners-of-the-zenoss-summer-of-zenpacks-contest\nabout Egor Puzanov's WMI Data Source zenpack for Zenoss Core, which\nlooks like it provides the kind of stuff you want.
[18-Sep-2009 13:48:49] <ckrough> Brad_K: the community search at the top and search.zenoss.com might work for you
[18-Sep-2009 13:48:54] <metalklesk> aaahhh ok, thanks !!!
[18-Sep-2009 13:49:03] <ckrough> Brad_K: community searched has been working for me.
[18-Sep-2009 13:49:09] <Brad_K> i tried the community search. didn't turn up anything useful for "irc".
[18-Sep-2009 13:49:19] <ckrough> meh
\n[18-Sep-2009 13:49:31] <twm1010> mrayzenoss: i'm surprised you\nguys let Egor release that WMI datasource zenpack... what with WMI\nmonitoring a touted feature of enterprise
[18-Sep-2009 13:49:33] <mrayzenoss> Documentation->IRC
[18-Sep-2009 13:49:43] <mrayzenoss> community/documentation/irc
[18-Sep-2009 13:49:55] <mrayzenoss> twm1010: we don't block anything from the community
[18-Sep-2009 13:49:56] <cgibbons> Egor's is a long way from what Enterprise has
(so far)[18-Sep-2009 13:50:11] <twm1010> gotcha
[18-Sep-2009 13:50:34] <mrayzenoss> Predictive Thresholds is essentially the same as the Enterprise one
[18-Sep-2009 13:50:45] <mrayzenoss> (just not QA'd)
[18-Sep-2009 13:51:32] <mrayzenoss> we would get a bad reputation real fast if we tried to shut down Community people
\n[18-Sep-2009 13:52:10] <Brad_K> unfortunately, the irc archive at\ndocs/DOC-2466 doesn't include anything from\nSeptember, at least not yet.
[18-Sep-2009 13:52:22]\n<pinball_> one thing to note cgibbons is that Egor's zenpack\nactually works when collecting perf data, while the enterprise wmi\ntemplate didn't
[18-Sep-2009 13:52:27] <mrayzenoss> Brad_K: yeah, I usually get a monthly dump of docs
[18-Sep-2009 13:52:35] <cgibbons> that would also be news to me
[18-Sep-2009 13:52:39] <pinball_> at least on 2.4.5
\n[18-Sep-2009 13:52:51] <cgibbons> given most enteprrise customers\ndo actually collect perf data with the enteprise zenpack
[18-Sep-2009 13:52:58] <cgibbons> and i haven't gotten any case or tickets about that
[18-Sep-2009 13:53:02] <pinball_> we are on enterprise and in dev we had to switch over
\n[18-Sep-2009 13:53:30] <Brad_K> anyway, the problems with Egor's\nzenpack probably wouldn't have anything to do with how WMI would work\nunder Zenoss Enteroprise. D'oh!
[18-Sep-2009 13:53:47] <metalklesk> mrayzenoss, can i add a command just to be used in one monitored device ?
[18-Sep-2009 13:54:25] <mrayzenoss> metalklesk: yeah, you can add a custom template on just 1 device
[18-Sep-2009 13:54:36] <metalklesk> ok, thanks
[18-Sep-2009 13:54:52] <cgibbons> did you open up a case on that, pinball?
[18-Sep-2009 13:54:57] <pinball_> not yet
[18-Sep-2009 13:55:02] <pinball_> waiting on the new version
[18-Sep-2009 13:55:08] <pinball_> i am holding the update
[18-Sep-2009 13:55:12] <pinball_> from 2.3.3 to 2.4.5
[18-Sep-2009 13:55:17] <cgibbons> new version past 2.4.5?
[18-Sep-2009 13:55:26] <pinball_> and am going to wait for the next release
[18-Sep-2009 13:55:34] <pinball_> before proceeding with an update
[18-Sep-2009 13:55:54] <pinball_> i c 2.5 beta is out
\n[18-Sep-2009 13:55:55] <cgibbons> okay, can to share any details\nabout what happened? because right now we're in the final stages of bug\nfixes for 2.5.0 and there isn't anything i know about being broken in\n2.4.5 that would change in 2.5.0
[18-Sep-2009\n13:56:33] <pinball_> yeah the winperf / perf data was not being\npicked up on the two windows servers i have in dev
[18-Sep-2009 13:56:34] <pinball_> one vm
[18-Sep-2009 13:56:37] <mrayzenoss> cgibbons: pinball_ is on 2.3.3
[18-Sep-2009 13:56:38] <pinball_> and one dell 2850
\n[18-Sep-2009 13:56:43] <Brad_K> curiosly, the problems we were\nhaving with zencommand not working for the email_delivery_check\nsuddenly started working around midnight last night. As did the other\nhalf of a command that had been working before, but was recently\nbroken. Bleah.
[18-Sep-2009 13:56:43] <pinball_> dev. is 2.4.5
[18-Sep-2009 13:56:47] <mrayzenoss> ahh
[18-Sep-2009 13:56:54] <pinball_> prod is 2.3.3
[18-Sep-2009 13:56:57] <cgibbons> that of course won't be enough info
[18-Sep-2009 13:57:24] <pinball_> so CPU , free mem etc.. was not being collected
\n[18-Sep-2009 13:57:34] <Brad_K> the thing that annoys me about\nthis is that i had remodeled those devices many times over. ;-(
[18-Sep-2009 13:57:50] <pinball_> so i was looking around for others seeing the same thing
[18-Sep-2009 13:57:56] <pinball_> saw egors zenpacks
[18-Sep-2009 13:58:00] <pinball_> tried them and viola it worked
[18-Sep-2009 13:58:20] <pinball_> not sure if it was winperf related
\n[18-Sep-2009 13:58:40] <cgibbons> egor's zenpack collects\nperfdata via a WMI query, whereas the enterprise one does it directly\nvia PerfMon (which means the remote registry service), so it might have\nbeen something on those servers misconfigured
[18-Sep-2009 13:58:44] <cgibbons> but that's certainly been the case since 2.3.0
[18-Sep-2009 13:59:10] <cgibbons> eidolon: are you still around?
[18-Sep-2009 14:01:38] <twm1010> i wonder does remote registry respond faster than WMI?
[18-Sep-2009 14:01:52] <rmatte> twm1010: I'd imagine they'd be comparable
[18-Sep-2009 14:01:59] <cgibbons> there's less overhead with it yes
[18-Sep-2009 14:02:04] <Brad_K> and is it more reliable, both inherently and in the zenpack?
[18-Sep-2009 14:02:19] <twm1010> I remember using the perfmon zenpack which used a remotely executed perfmon query via wmiexec
[18-Sep-2009 14:02:29] <cgibbons> yeah that was the prior to 2.3
[18-Sep-2009 14:03:08] <Brad_K> and how stable is the remote registry vs. WMI on the remote Windows server side?
[18-Sep-2009 14:03:12]\n<cgibbons> i should really draw up an architectural diagram, but\nboth WMI and Remote Registry are layered on-top of Microsoft's Remote\nProcedure Call implementation, WMI is further on top of the DCOM layer\nand then is more complex inside because it translates string queries,\nand then does a local PerfMon data lookup, before returning the raw\ndata.
[18-Sep-2009 14:03:26] <rmatte> Brad_K: I think the question is what's not more stable than WMI?
[18-Sep-2009 14:03:50] <cgibbons> as of 2.3.x zenwinperf functions the same way that the PerfMon utility itself does
[18-Sep-2009 14:04:09]\n<mrayzenoss> ckrough: Brad_K: followup on the Austin training:\nAdmin training will start being offered remotely next month, Advanced\ntraining will be available in Austin and Annapolis
[18-Sep-2009 14:04:26] <twm1010> ah... so remote registry -> perfmon is simply the shortest path to the data.
[18-Sep-2009 14:04:29]\n<cgibbons> doing a WMI query to get the data is more convenient\nfor scripting users and admin because there's little parsing that's\nrequired
[18-Sep-2009 14:04:30] <Brad_K> mrayzenoss: cool!
[18-Sep-2009 14:04:32] <cgibbons> correcto
[18-Sep-2009 14:04:32] <rmatte> mrayzenoss: prices for remote training?
[18-Sep-2009 14:04:40] <mrayzenoss> rmatte: I dunno yet
[18-Sep-2009 14:04:55] <mrayzenoss> rmatte: I'd assume less, but I don't know
[18-Sep-2009 14:04:55] <rmatte> mrayzenoss: will there be an outline available of what's covered in the courses?
[18-Sep-2009 14:05:00] * ckrough vi does funny thinks when you ':mrayzenoss'...
[18-Sep-2009 14:05:04] <Brad_K> mrayzenoss: please let me know, because that may have a serious impact on my ability to get any training.
[18-Sep-2009 14:05:05] <ckrough> mrayzenoss: thanks
[18-Sep-2009 14:05:22] <mrayzenoss> Brad_K: yeah, I mentioned that when I was asking
[18-Sep-2009 14:05:35] <Brad_K> mrayzenoss: much appreciated. Thanks!
[18-Sep-2009 14:05:46] <twm1010> so... all this WMI work is nice, but probably causes lots of collector overhead... eh?
[18-Sep-2009 14:06:03] <cgibbons> not really a lot, no, but it's not as fast as perfmon data
[18-Sep-2009 14:06:06] <rmatte> E492: Not an editor command: mrayzenoss
[18-Sep-2009 14:06:09] <ckrough> rmatte: message/39572
[18-Sep-2009 14:06:37]\n<cgibbons> if you were trying to collect 1000 performance metrics\nper device every 10 seconds, then you'd probably not want to be doing\nit via WMI queries
[18-Sep-2009 14:06:47] <mrayzenoss> rmatte: sounds like you need to upgrade to Emacs
[18-Sep-2009 14:06:56] <mrayzenoss> M-x mrayzenoss
[18-Sep-2009 14:07:09] <ckrough> heh, actually I was in 'less'. talk about being disoriented.
[18-Sep-2009 14:07:24]\n<rmatte> I'm actually a nano guy myself (used to use pico when I\nwas first learning FreeBSD, so I know all of the shortcuts)
[18-Sep-2009 14:07:25] <Brad_K> no, talk is a different program from either of those.
[18-Sep-2009 14:07:31]\n<cgibbons> I know the network overhead is higher, but I haven't\nmeasured it, but for comparison a single query of data on this windows\nserver 2008 i'm testing right now is 144,542 bytes
[18-Sep-2009 14:07:33] * Brad_K ;-)
[18-Sep-2009 14:07:40] <ckrough> heh
[18-Sep-2009 14:08:44] <Brad_K> cgibbons: yowch! 144KB? for a single query?!?
[18-Sep-2009 14:09:04] <cgibbons> for all of the counters available, yeah
[18-Sep-2009 14:09:16] <rmatte> Brad_K: Welcome to Microsoft country
[18-Sep-2009 14:09:19] <twm1010> all available counters?
[18-Sep-2009 14:09:49] <twm1010> thats a ton of data
[18-Sep-2009 14:10:01] * Brad_K c|n>k
[18-Sep-2009 14:11:38] <metalklesk> mrayzenoss, i've added a new data source for a command
[18-Sep-2009 14:11:50] <metalklesk> what's next to get the output ?
[18-Sep-2009 14:11:55] <Brad_K> okay, gotta go to lunch. bbl
[18-Sep-2009 14:12:03] <mrayzenoss> use the 'test' to see how it's working
[18-Sep-2009 14:12:25] <metalklesk> ther i should grab the ip right ?
[18-Sep-2009 14:12:37] <mrayzenoss> or the device name
[18-Sep-2009 14:12:53] <mrayzenoss> the device name will grab the user/pass from the zProperties
[18-Sep-2009 14:13:04] <metalklesk> ok
[18-Sep-2009 14:13:41] <metalklesk> it works
[18-Sep-2009 14:14:06] <metalklesk> now what is next to use this command ?
[18-Sep-2009 14:15:16] <mrayzenoss> add your data points, add your threshold and graphs
[18-Sep-2009 14:15:19] <mrayzenoss> then bind it
[18-Sep-2009 14:15:58] <mrayzenoss> docs/DOC-3025
[18-Sep-2009 14:17:14] <metalklesk> thanks
[18-Sep-2009 14:20:19] <twm1010> reviewing king crab finally
\n[18-Sep-2009 14:24:33] <metalklesk> mrayzenoss, once you have the\ncommand "rpm -qa --qf\n"%{NAME}-%{VERSION}-%{RELEASE}|%{INSTALLTIME}|%{SUMMARY}\\n"" added as a\ndatasource in the performance template, how can i grab the outputed\ndata into the software list in the device ?
[18-Sep-2009 14:24:45] <twm1010> severity icons much easier on the eyes than the severity colors
[18-Sep-2009 14:26:04] <mrayzenoss> metalklesk: ahh, that's the wrong approach
[18-Sep-2009 14:26:12] <metalklesk>
[18-Sep-2009 14:26:22] <mrayzenoss> metalklesk: templates and data sources are for performance monitoring, the 'Perf' tab
[18-Sep-2009 14:26:31] <metalklesk> aaaaahh ok
[18-Sep-2009 14:26:33] <twm1010> shouldn't isntalled packages be available via SNMP?
[18-Sep-2009 14:26:42] <mrayzenoss> twm1010: if configured properly
\n[18-Sep-2009 14:26:57] <mrayzenoss> getting stuff onto the\nSoftware tab via SSH requires looking at how the other SSH Linux\nZenPacks do it
[18-Sep-2009 14:27:18] <mrayzenoss> so Fedora might be a good start docs/DOC-3382
[18-Sep-2009 14:27:24] <mrayzenoss> since it probably uses RPM too
[18-Sep-2009 14:27:27] <mrayzenoss> or OpenSUSE
[18-Sep-2009 14:27:38] <mrayzenoss> docs/DOC-3478
\n[18-Sep-2009 14:27:57] <metalklesk> the thing is that i was\ntrying opensuse edited version of zenpack but has some "problems"
[18-Sep-2009 14:28:15] <metalklesk> first, it does not show all the graphs
[18-Sep-2009 14:29:04] <metalklesk> and if you put out ram of the zignorefilesystem, it does not show used, total, free ram in OS tab
\n[18-Sep-2009 14:29:43] <metalklesk> in the other hand, using just\nsnmp monitoring, you can get everything you want except the list of\ninstalled software
[18-Sep-2009 14:30:18]\n<metalklesk> so what i'm trying to do es using snmp for all and\nbuild a command to get the installed software
[18-Sep-2009 14:30:22] <metalklesk> using ssh
[18-Sep-2009 14:31:09] <mrayzenoss> hmm... you could do it with SNMP thought
\n[18-Sep-2009 14:31:14] <mrayzenoss> take a look at\nblogs/zenossblog/2009/02/18/tip-of-the-month-snmp-software-inventory-for-debian-and-ubuntu-machines
[18-Sep-2009 14:31:57] <metalklesk> can i ??
[18-Sep-2009 14:32:00] <mrayzenoss> yeah
[18-Sep-2009 14:32:35] <mrayzenoss> wow, the formatting got totally mangled on that blog post
\n[18-Sep-2009 14:39:10] <metalklesk> one litte approach is\ncreating a simple user command in the device i want, so clicking the\nnew command in Run commands -> my command, works
\n[18-Sep-2009 14:39:46] <metalklesk> but it does not show the\noutput in software tab, but at least it shows me the software list
[18-Sep-2009 14:39:57] <metalklesk>
[18-Sep-2009 14:40:40] <metalklesk> it si really cool because i can do whatever i want remotely
[18-Sep-2009 14:40:47] <metalklesk> *it is
[18-Sep-2009 14:56:03]\n<perr0> I just installed zenoss-2.4.5 from source and get errors\nwhen starting zenoss can anyone have a look possibly point me to where\nI can begin to troubleshoot?
[18-Sep-2009 14:56:05] <perr0> http://pastebin.org/18756
[18-Sep-2009 14:59:12] <cgibbons> do you have the net-snmp package installed on that box?
[18-Sep-2009 15:00:33] <perr0> ahh probably not
[18-Sep-2009 15:00:54] <twm1010> install both net-snmp and net-snmp-utils
[18-Sep-2009 15:00:56] <twm1010>
[18-Sep-2009 15:01:08] <perr0> nice, thanks guys
[18-Sep-2009 15:08:14] <eidolon> hrm, anyone seen rrd charts fail in mid-work? event log is showing: 2009-09-18T16:06:50 ERROR RenderServer Failed to generate a graph
[18-Sep-2009 15:08:14] <eidolon> Traceback (most recent call last):
[18-Sep-2009 15:08:14] <eidolon> File "/opt/zenoss/Products/ZenRRD/RenderServer.py", line 190, in render
[18-Sep-2009 15:08:14] <eidolon> rrdtool.graph(*gopts)
[18-Sep-2009 15:11:40] <perr0> -
\n[18-Sep-2009 15:12:46] <rmatte> wow, one of our VMs was\naccidentally set to use 255GB of RAM, sucked the life out of every\nother VM on the blade
[18-Sep-2009 15:12:54] <rmatte> Fridays are never boring
[18-Sep-2009 15:12:59] <twm1010> That's awesome.
[18-Sep-2009 15:13:11] <twm1010> somebody changed an MB to a GB huh?
[18-Sep-2009 15:13:19] <rmatte> nah, we found out how it happened...
[18-Sep-2009 15:13:37] <straterra> intern
[18-Sep-2009 15:13:47] <twm1010> i've been to known to set a directory quota to 50MB instead of 50GB before
[18-Sep-2009 15:14:08]\n<rmatte> there's an input field, so they typed in the correct\namount which was 4gb, but then they clicked on a gray area of the\nwindow, far right of this vertical scale of how much memory is being\nset, he clicked somewhere near the top of the window by accident so it\nset it to that
[18-Sep-2009 15:14:46] <rmatte>\nIt's stupidly designed, I can understand if you clicked on the scale\nitself and it adjusted it, but not when you click far to the right of it
\n[18-Sep-2009 15:18:36] <Brad_K> hey, 255GB of RAM. that sounds\npretty cool! the biggest i've ever built was Sun X4440s with four\nquad-core CPUs and 64GB of RAM....
[18-Sep-2009\n15:19:51] <rmatte> well, we only actually have 32GB of RAM in the\nblade, which is also stupid that VMWare would even let us set it that\nhigh
[18-Sep-2009 15:20:53] <rmatte> I'm sitting here watching as my king crab test box takes 15 minutes to shutdown...
[18-Sep-2009 15:21:00] <rmatte> since it's on that blade
[18-Sep-2009 15:21:01] <rmatte> lol
[18-Sep-2009 15:21:18] <ckrough> rmatte: Do you use resource pools?
[18-Sep-2009 15:21:31] <ckrough> assuming you are referring to vmware
[18-Sep-2009 15:22:27] <rmatte> ckrough: we do, but this VM was created outside of the resource pools
[18-Sep-2009 15:22:48] <rmatte> so it just choked everything out
[18-Sep-2009 15:28:15] <rmatte> rebooting all of our dev boxes now...
[18-Sep-2009 15:30:24] <Brad_K> ouch....
[18-Sep-2009 15:43:13] <ckrough> have a good weekend. Im outski
[18-Sep-2009 15:43:53] <Brad_K> cya!
\n[18-Sep-2009 15:46:18] <eidolon> so is there any way to -really-\nwatch whether zenoss is using WMI to model / look at a server, and to\nforce an update?
[18-Sep-2009 15:46:48]\n<eidolon> the problem is we're diong stuff passively on a windows\nhost (like filling up a disk), and zenoss doesn't show the update. we\nwait 5 minutes, nothing. this constnat waiting and sort of 'well, it\nmight work, wait around and see' - is really frustrating.
[18-Sep-2009 15:54:03]\n<Brad_K> out of curiosity, does anyone know how to see the Zenoss\nequivalent of the Nagios event queue, so that we can tell when the next\nscheduled event is going to occur for a given template?
\n[18-Sep-2009 15:54:59] <Brad_K> actually, the template is bound\nto a device, and what we want to know is when the next SSH check is\ngoing to happen for that device.
[18-Sep-2009 17:47:45] <daMaestro> okay, i'm done with the nginx zenpack
[18-Sep-2009 17:47:50] <daMaestro> anyone know where/how to upload it?
[18-Sep-2009 17:58:58] <USENIX_UNIX> Can't wait to get my grubby hands on it
\n[18-Sep-2009 19:37:55] <Brad_K> okay, this is seriously screwed\nup. We have several Ironport e-mail security appliances here. When\nthey were modeled, they came up as "Unknown" manufacturer, even though\nZenoss does seem to know about Ironport.
[18-Sep-2009\n19:38:24] <Brad_K> I've been trying to fix these. I added the\nC300 and C600 devices to the list, and classified some older machines\ncorrectly.
[18-Sep-2009 19:38:57] <Brad_K> but\nwhen i tried to add the C660 to the list, and classify some newer\nmachines, one worked fine although C660 never showed up on the official\nlist of devices from the pull-down.
[18-Sep-2009\n19:39:47] <Brad_K> but another device I was working on is totally\nwigged out. Now there is no vendor shown at all, no device name shown\nat all, and no software vendor shown at all. I can't even go back and\nre-set them to "Ironport" and "C600", or any other value from any of\nthe pull-downs.
[18-Sep-2009 19:40:59] <Brad_K>\ni could probably reset things using zendmd, as another poster in one of\nthe forums recently did. but i don't know zendmd, and i don't see how\nit should be necessary for this sort of thing.
[18-Sep-2009 19:41:13] * Brad_K starts humming the theme from "Twilight Zone"...
[18-Sep-2009 19:41:38] <Brad_K> Anyway, i'll worry about filing a ticket on this next week.
[18-Sep-2009 19:41:42] <Brad_K> i
[18-Sep-2009 19:41:50] <Brad_K> i'm going home. g'nite!
[19-Sep-2009 00:00:46] [disconnected at Sat Sep 19 00:00:46 2009]
[19-Sep-2009 00:00:46] [connected at Sat Sep 19 00:00:46 2009]
[19-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[19-Sep-2009 09:13:44] <Apocalipse> hello people
[19-Sep-2009 12:06:03] timburke_ is now known as timburke
[20-Sep-2009 00:00:46] [disconnected at Sun Sep 20 00:00:46 2009]
[20-Sep-2009 00:00:46] [connected at Sun Sep 20 00:00:46 2009]
[20-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[21-Sep-2009 00:00:46] [disconnected at Mon Sep 21 00:00:46 2009]
[21-Sep-2009 00:00:46] [connected at Mon Sep 21 00:00:46 2009]
[21-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[21-Sep-2009 04:31:23] <Daan> o/ Good morning...
[21-Sep-2009 04:32:32] Daan is now known as danuvius
[21-Sep-2009 04:32:42] danuvius is now known as daan
[21-Sep-2009 04:34:58] daan is now known as danuvius
[21-Sep-2009 04:35:16] danuvius is now known as daan
[21-Sep-2009 04:35:44] daan is now known as danuvius
[21-Sep-2009 04:36:52] <danuvius> o/ Good morning...
[21-Sep-2009 05:46:09] <Dieterbe_> hi
[21-Sep-2009 06:37:58] <Troubadix09> hi all
[21-Sep-2009 06:45:24] <danuvius> hi
[21-Sep-2009 07:44:18] <danuvius> anyone here that is familiar with the zenmailtx zenpack?
\n[21-Sep-2009 10:24:38] <drtns> morning all I need a little regex\n101 help... we run a custom apache with the MPM-ITK module.. thus our\nprocess is called httpd.itk rather then just httpd so in zenoss I get a\nbunch of boxes that say that httpd.itk is not running. I need to mod\nmy /process/apache/httpd regex to allow either a process named httpd\nfor the stock boxes OR httpd.itk for the boxes running that. the\ncurrent regex is ^.*\\/*httpd can anyone help me out
[21-Sep-2009 10:26:55] <twm1010> play with this
[21-Sep-2009 10:26:55] <twm1010> http://gskinner.com/RegExr/
[21-Sep-2009 10:27:38] <twm1010> could you not put a * on the end?
[21-Sep-2009 10:28:08] <drtns> cool thanks for that twm1010 I tried that stock ^.*\\/*httpd*
[21-Sep-2009 10:28:15] <drtns> but it doesn't seem to do the trick
[21-Sep-2009 10:32:38] <danuvius> someone here using the zenmailtx zenpack?
[21-Sep-2009 10:33:00] <drtns> since '.' dot in a regex means any character I assume I escape the dot \\. to mean a literal dot?
[21-Sep-2009 10:35:28] <danuvius> drtns: ^.*\\/*httpd\\.itk
\n[21-Sep-2009 10:36:18] <drtns> danuvius but wouldn't that then\nkill any servers not runnin itk? think maybe I need something more\nlike ^.*\\/*(httpd|httpd\\.itk)
[21-Sep-2009 10:36:35] <drtns> to mach either the standard httpd or httpd.itk?
[21-Sep-2009 10:37:52] <danuvius> drtns: hmmm ^.*\\/*httpd(\\.itk)+
[21-Sep-2009 10:38:49] <danuvius> sorry mean ^.*\\/*httpd(\\.itk)?
[21-Sep-2009 10:39:45] <drtns> danuvius the 2nd could possibly work
[21-Sep-2009 10:40:11] <drtns> think mine might work to just using the "or" just written out longer..
[21-Sep-2009 10:40:35] <danuvius> true yours would have worked to...
[21-Sep-2009 11:06:10] <rmatte> drtns: why would you just add a completely separate process to Processes called httpd.itk ?
[21-Sep-2009 11:06:18] <rmatte> why wouldn't you
[21-Sep-2009 11:06:19] <rmatte> rather
[21-Sep-2009 11:07:17] <rmatte> instead of trying to do 2 in 1
[21-Sep-2009 11:07:25]\n<drtns> rmatte just as valid an argument to say why bother when\nyou can just "or" the regex and keep it all happy under a single process
[21-Sep-2009 11:07:48] <rmatte> drtns: because they aren't technically even the same process
[21-Sep-2009 11:08:14]\n<rmatte> drtns: When you model a device, Zenoss adds the\nprocesses based on what's enabled to be monitored in the processes\nlist...
[21-Sep-2009 11:08:39] <rmatte> if you\nhave 1 process with a regex of httpd or httpd.itk it's always going to\nmap that process to the processes list
[21-Sep-2009 11:08:57]\n<rmatte> if you have 1 process that's called httpd and 1 process\nthat's called httpd.itk, it'll pick which one it adds
[21-Sep-2009 11:09:41] <rmatte> plus it's a heck of a lot simpler
\n[21-Sep-2009 11:10:19] <rmatte> no point to make things difficult\nfor yourself over adding 1 process entry to the processes list
[21-Sep-2009 11:18:25] <Brad_K> so, zencommand is acting weirder and weirder....
\n[21-Sep-2009 11:18:43] <drtns> rmatte so I add a new service to\n/process/apache for httpd.itk and it's listing count as 0 how does\nthat get populated? can I force a refresh?
\n[21-Sep-2009 11:18:48] <Brad_K> things mysteriously start working\nwhen they weren't, and mysteriously stop working when they were.
[21-Sep-2009 11:19:23] <rmatte> drtns: you need to remodel devices for it to pick it up
[21-Sep-2009 11:19:50] <rmatte> Brad_K: you may want to adjust your zencommand timeout values in the collector settings
[21-Sep-2009 11:19:51] <drtns> ok I'll give that a whirl
\n[21-Sep-2009 11:21:35] <rmatte> actually, there aren't any\ntimeout settings for zencommand in there, they must be under\nzProperties then
[21-Sep-2009 11:22:27] <rmatte> zCommandCommandTimeout and zCommandCycleTime
\n[21-Sep-2009 11:23:30] <Brad_K> right now, i've got 15.0 and 60\nfor those values, but in the definition of the command itself, the\ncycle time is set to 300. does the zproperties over-ride the\ncommand-specific cycle time?
[21-Sep-2009 11:27:22] <rmatte> not sure
[21-Sep-2009 11:27:26] <rmatte> doubt it
[21-Sep-2009 11:27:41] <rmatte> so there's no indication in the log files of any issues?
[21-Sep-2009 11:27:55] <rmatte> I'd expect there to be at least some errors if it's not working properly
\n[21-Sep-2009 11:28:25] <Brad_K> and for a job that should really\nbe run every 300 seconds, and you want it to keep retrying/waiting\nright up to the next cycle time, what kinds of values would be\nreasonable to expect?
[21-Sep-2009 11:29:13]\n<Brad_K> we did see some timeouts in the web ui on the hub, but\ni've never seen any associated errors in the zencommand.log from the\ncollector.
[21-Sep-2009 11:33:13] <Brad_K> the\nonly relevant thing i've seen in the zencommand.log files from the\ncollector have been a failure to log certain information, but no errors\nassociated with that.
[21-Sep-2009 11:34:18] <rmatte> the default values are generally reasonable
[21-Sep-2009 11:34:24] <rmatte> what types of commands are failing?
\n[21-Sep-2009 11:35:10] <Brad_K> the same Nagios-derived\ncheck_ping and check_email_delivery that I've been fighting with all\nthis time.
[21-Sep-2009 11:35:19] <rmatte> ah
[21-Sep-2009 11:35:39] <rmatte> did you ever modify them to remove the \\n ?
[21-Sep-2009 11:36:02]\n<Brad_K> oh, and btw -- multiline output doesn't seem to be the\nissue. i hacked check_email_delivery to remove the \\n at the end of\nthe line, while check_ping still issues the \\n. Right now, check_ping\nis working, and check_email_delivery is not.
[21-Sep-2009 11:36:20] <rmatte> I see, fair enough
[21-Sep-2009 11:36:26] <Brad_K> the case of the variables being stored did seem to make a difference.
\n[21-Sep-2009 11:36:58] <rmatte> well, my suggestion would be to\nmodify the script to output some stuff to a file each time they run
[21-Sep-2009 11:37:08] <rmatte> and hopefull you can figure out what the issue is that way
\n[21-Sep-2009 11:37:37] <rmatte> and by stuff I mean: output the\ndata being passed to the command, the output, along with a timestamp if\npossible
[21-Sep-2009 11:38:02] <Brad_K> the\nproblem is not the scripts. when i run them directly from the\ncommand-line, they work perfectly. when i run them manually via\nbin/zencommand, they work perfectly. it's only when i try to integrate\nthem into being auto-run by Zenoss that they sometimes work, and\nsometimes fail.
[21-Sep-2009 11:38:48] <rmatte>\njust because they are running fine manually doesn't mean that they\naren't running fine automatically
[21-Sep-2009 11:39:07]\n<rmatte> it would just be a starting point so that you could get\ntimestamped data, to see how often the commands are failing and what's\nbeing passed to them
[21-Sep-2009 11:39:18] <rmatte> maybe some data is getting botched or something, who knows?
\n[21-Sep-2009 11:40:19] <rmatte> it's the best starting point for\ntroubleshooting this seeing as your logs are apparently useless
\n[21-Sep-2009 11:40:23] <Brad_K> you can look at the output shown\nin zencommand.log (with log severity set to 10), and see that the\noutput is exactly as expected. the problem is that zencommand\nsometimes does not process this output as it should.
\n[21-Sep-2009 11:41:26] <rmatte> well, I've personally never seen\nthat behaviour, just from scripting experience, adding some debug lines\nin to the scripts may help
[21-Sep-2009 11:41:56] <rmatte> the timestamps will probably be the most valuable thing
\n[21-Sep-2009 11:42:50] <Brad_K> i can try that, but i've been\nhacking scripts for over twenty years, and i imagine you probably have\nabout the same amount of experience. my experience tells me that this\nisn't likely to tell us much, but it's not that hard to do. it's only\nPerl after all -- it's not like i have to write any Python code or\nanything. ;-)
[21-Sep-2009 11:44:06] <Brad_K>\noh, and btw -- outside of work, i also happen to be one of the\npostmasters for python.org. i'm just not a Python programmer -- ask\nBarry Warsaw, if you don't believe me.
[21-Sep-2009 12:06:34] <rmatte> lol
[21-Sep-2009 12:06:39] <rmatte> all good
[21-Sep-2009 12:07:08]\n<rmatte> no, I have a lot of experience for my age when it comes\nto this stuff, but I'm only 24 years old, so you have me beat there
[21-Sep-2009 12:07:28]\n<rmatte> I started playing around with FreeBSD when I was 14\nthough, so I've had quite a bit of exposure to this stuff for my age
[21-Sep-2009 12:07:29] <rmatte>
[21-Sep-2009 12:08:33]\n<rmatte> I've been learning python coding because I manage Zenoss\nservers, I'm getting pretty decent at it though, I'm usually pretty\ngood at picking up languages
[21-Sep-2009 12:09:33]\n<Brad_K> well, the command is mysteriously working again, but my\nhacks to write the data to a temporary file are not yet working.
\n[21-Sep-2009 12:10:18] * Brad_K hangs his head when he realizes he\ngraduated high school and has been on the Internet longer than rmatte\nhas been alive
[21-Sep-2009 12:11:44] <rmatte> hehe
[21-Sep-2009 12:12:25] <rmatte> well, I just found yet another king crab bug
[21-Sep-2009 12:13:26] <twm1010> Brad_K: Don't feel bad, rmatte is really a machine.
[21-Sep-2009 12:13:46] <Brad_K> we're not even runnning 2.4.5, much less king crab. blech.
\n[21-Sep-2009 12:14:24] <twm1010> I've found that being on the\nbleeding edge with Zenoss is a good way to learn to how to perform\nbackups and restores
\n[21-Sep-2009 12:14:36] * Brad_K remembers that 24 years ago was about\nthe time that William Gibson published books like Mona Lisa Overdrive\nand wrote about Wintermute for the first time....
[21-Sep-2009 12:15:08] <twm1010> My approach this go around has been to install it, and leave it alone and let it do it's job.
\n[21-Sep-2009 12:15:26] <twm1010> Maximum performance seems to\ncome from watching the debug messages and getting rid of what ever\ntheir cause is
[21-Sep-2009 12:16:29] <Brad_K>\nthat's where we want to get. we don't really want to be debugging any\ncode, to be honest. after all, this is going to be a production system\n& network monitor for mission-critical systems for the entire\nUniversity, and ~70,000 LAN customers here in Austin.
[21-Sep-2009 12:17:25] <twm1010> well, i of course meant debug events
[21-Sep-2009 12:17:29] <twm1010> not debug logs
\n[21-Sep-2009 12:18:29] <twm1010> bad OID events and performance\ntemplates looking for stuff that isn't there seems to make MySQL shit a\nbrick
[21-Sep-2009 12:18:41] <twm1010> if you get that under control, its like pushing the turbo button
[21-Sep-2009 12:19:05] <twm1010> i still need to learn how to trim down syslog chatter
[21-Sep-2009 12:19:10] <twm1010> i'm getting too much information right now
[21-Sep-2009 12:22:36] <rmatte> Brad_K: we're running 2.4.5 prod (unfortunately) and I have a King Crab test box setup
[21-Sep-2009 12:23:08] <rmatte> 2.4.5 is actually quite good, if it weren't for: http://dev.zenoss.com/trac/ticket/5473
[21-Sep-2009 12:23:14] <rmatte> that is driving me absolutely up the wall
\n[21-Sep-2009 12:24:47] <Brad_K> i've seen that in 2.4.2, as\nwell. We have Ironport e-mail security devices, and I added C300 and\nC600 to the list of known devices. Now it won't let me add C660, and\nfor some machines when I set their HW product, that doesn't stick. But\nfor others, it does, but only for some variables.
[21-Sep-2009 12:24:55] <rmatte> twm1010: syslog chatter is easy...
\n[21-Sep-2009 12:26:10] <rmatte> twm1010: just create a manual\nmapping under whatever class (I created a /Suppressed class), make the\neventClassKey "defaultmapping" then specify a regex for what you want\nto block, go to the zProperties for the event mapping and select drop\nas the action
[21-Sep-2009 12:26:13] <rmatte> and voila, you're done
[21-Sep-2009 12:26:53] <rmatte> For the regex you just need to put a part of the message
[21-Sep-2009 12:27:08] <rmatte> I had messages coming in like: "(ip) vrf/dest_addr=7813781"
[21-Sep-2009 12:27:14] <rmatte> so I set my regex to: (ip) vrf/dest_addr=
[21-Sep-2009 12:27:32] <rmatte> I also had syslog messages coming in completely blank
[21-Sep-2009 12:27:37] <rmatte> so I used ^$ as the regex
[21-Sep-2009 12:28:16] <rmatte> g'day Matt
[21-Sep-2009 12:28:29] <Brad_K> mrayzenoss1: hey!
\n[21-Sep-2009 12:29:27] <Brad_K> so, shall we re-open our contest\nto see who the oldest person is on this channel, and by how much?
[21-Sep-2009 12:29:50] <rmatte> I'm definitely the youngest
[21-Sep-2009 12:29:52] <rmatte> so I lose
[21-Sep-2009 12:30:34] <Brad_K> mrayzenoss1: how about you?
[21-Sep-2009 12:31:33] <rmatte> he's so old that he doesn't like to talk about it
[21-Sep-2009 12:31:44] <rmatte>
\n[21-Sep-2009 12:32:35] <Brad_K> okay, so i'm 43. i'm old enough\nto be your dad. my beard isn't completely grey, but that's partly\nbecause the grey hairs mysteriously disappear in the morning.
[21-Sep-2009 12:33:15] <mrayzenoss1> 35
[21-Sep-2009 12:33:59] <Brad_K> twm1010: can you beat 43?
[21-Sep-2009 12:35:04] <rmatte> my dad is quite a bit older than you lol
[21-Sep-2009 12:35:29] <Brad_K> yeah, well my dad is older than your dad! :-P
[21-Sep-2009 12:35:35] <rmatte>
[21-Sep-2009 12:35:45] <rmatte> My dad could beat up your dad
[21-Sep-2009 12:35:51] <rmatte> lol
[21-Sep-2009 12:36:10] <rmatte> takes me back to kindergarten
\n[21-Sep-2009 12:36:41] <Brad_K> probably, but then my dad would\ncall in his brother who used to be in the Hells Angels, and is now a\ntruck driver -- and who has lost most of his teeth in bar fights.
[21-Sep-2009 12:36:49] <rmatte> ok, so time to log another bug
[21-Sep-2009 12:36:56] <Brad_K> heh
[21-Sep-2009 12:37:05] <rmatte> actually, I don't have my trac password here, bah, I'll have to do it tomorrow
[21-Sep-2009 12:37:07] <twm1010> I'm 27
\n[21-Sep-2009 12:37:35] <Brad_K> rmatte: if you can get them to\nfix these bugs in 2.45 and 2.4.2 as well, that would be great.
\n[21-Sep-2009 12:37:57] <rmatte> Brad_K: well, this particular bug\nrelates to the new event console, so it doesn't apply to those versions
\n[21-Sep-2009 12:38:22] <Brad_K> twm1010: so, not much older than\nrmatte. interesting. i wonder what it is about Python-derived\nprograms that seems to draw the younger crowd....
\n[21-Sep-2009 12:38:26] <rmatte> I've been trying to push to have\nthat hardware bug fixed and then have the fix backported to 2.4.5 but I\nhighly doubt that will happen
[21-Sep-2009 12:38:35] <rmatte> especially since King Crab is due out in about a month
[21-Sep-2009 12:38:51] <rmatte> they haven't even fixed that hardware bug in King Crab yet
[21-Sep-2009 12:39:06] <Brad_K> well, we are an Enterprise customer. maybe we can also help push on that one?
[21-Sep-2009 12:39:07] <twm1010> which hardware bug??
[21-Sep-2009 12:39:23] <Brad_K> http://dev.zenoss.com/trac/ticket/5473
[21-Sep-2009 12:39:26] mrayzenoss1 is now known as mrayzenoss
[21-Sep-2009 12:39:46] <rmatte> wb mrayzenoss
[21-Sep-2009 12:39:51] <twm1010> ah... that... right, the reason i stayed at 2.4.2
[21-Sep-2009 12:40:02] <mrayzenoss> yeah, new install of the OS, lost my IRC password
[21-Sep-2009 12:40:09] <Brad_K> actually, no -- it's broken in 2.4.2 as well.
\n[21-Sep-2009 12:40:12] <rmatte> mrayzenoss: any word if\nhttp://dev.zenoss.com/trac/ticket/5473 is getting fixed for King Crab\nrelease?
[21-Sep-2009 12:40:34] <Brad_K> we've seen the bug here, too.
[21-Sep-2009 12:40:43] <rmatte> lots of people are seeing the bug
[21-Sep-2009 12:40:59] <rmatte> in pretty much every 2.4.x version, except that it's more noticeable in 2.3.3 and onward
[21-Sep-2009 12:41:08] <rmatte> erm 2.4.3
[21-Sep-2009 12:41:09] <rmatte> rather
[21-Sep-2009 12:41:11] <mrayzenoss> yeah, I think someone was working on it
[21-Sep-2009 12:41:28] <rmatte> cool
[21-Sep-2009 12:41:37] <twm1010> I just set a HW Product on mine?
[21-Sep-2009 12:41:49] <Brad_K> sometimes it works, sometimes it doesn't.
[21-Sep-2009 12:41:52] <rmatte> twm1010: works for certain ones
[21-Sep-2009 12:41:53] <mrayzenoss> I like this ticket: http://dev.zenoss.com/trac/ticket/5568
[21-Sep-2009 12:42:07] <rmatte> twm1010: same with software product details, some work some don't
[21-Sep-2009 12:42:36] <Brad_K> what is the ZenAWS zenpack?
[21-Sep-2009 12:42:55] <mrayzenoss> something new
[21-Sep-2009 12:42:56] <rmatte> cool, new features
[21-Sep-2009 12:43:04] <Brad_K> cool.
\n[21-Sep-2009 12:43:11] <rmatte> though new features aren't nearly\nas cool when there are bugs preventing you from using old features
[21-Sep-2009 12:43:14] <mrayzenoss> it'll be in the beta drop next week
[21-Sep-2009 12:43:27] <Brad_K> yeah, there's definitely bugs i want fixed.
[21-Sep-2009 12:43:29] <mrayzenoss> rmatte: you're killing my moment
[21-Sep-2009 12:43:39] <rmatte> mrayzenoss: sorry, my bad
[21-Sep-2009 12:43:50] <mrayzenoss> yeah, new features means new bugs
[21-Sep-2009 12:43:52] <Brad_K> like ticket # 8722.
[21-Sep-2009 12:44:18] <mrayzenoss> not sure we have one that high... support case I assume?
\n[21-Sep-2009 12:44:29] <rmatte> as long as 5473 gets fixed, I can\nlive with the other bugs for a while, but that's the one that people\nare breathing down my neck about
[21-Sep-2009 12:44:55] <rmatte> yeh, I highly doubt there's a ticket 8722
\n[21-Sep-2009 12:44:56] <Brad_K> yes, support case. but frankly,\ni've gotten more help in this channel than through customer service --\nespecially from you and rmatte.
[21-Sep-2009 12:44:57] <rmatte>
[21-Sep-2009 12:45:23] <rmatte> Brad_K: cool, just pay us directly and we'll provide all the support you want
[21-Sep-2009 12:45:33] <rmatte> j/k
\n[21-Sep-2009 12:45:48] <Brad_K> if you could give us a written\ncontract that we could sign, we might actually think about that.
[21-Sep-2009 12:46:16] <rmatte> I'd have to quit my job to be able to do something like that lol
[21-Sep-2009 12:46:22] <rmatte> which I've no intention of doing
[21-Sep-2009 12:46:32] <rmatte> but if I ever get canned I may consider trying it out
[21-Sep-2009 12:47:04]\n<Brad_K> i definitely prefer an interactive debugging process, as\nopposed to sending an e-mail to customer service through the captive\nSugarCRM system, then wait hours or days for a response, get a\nsuggestion, and then go through that lengthy process all over again.
\n[21-Sep-2009 12:49:18] <Brad_K> i mean, we're paying money for\nthis stuff. the commercial support should be better and faster than\nthe community support, right?
[21-Sep-2009 12:49:41] <rmatte> Brad_K: in theory, yes
[21-Sep-2009 12:50:20] <rmatte> but the community support is a unique advantage of an open-source product
\n[21-Sep-2009 12:50:36] <rmatte> so you might as well take\nadvantage of it regardless as to whether or not you are paying for\nsupport
[21-Sep-2009 12:50:47] <Brad_K> i mean,\nat least with FreeBSd, the community support is from the same folks who\nwould be providing the commercial support, and it's your choice as to\nwhether or not you want to give money to PHK to have him put something\nat the top of the list, or if you want to wait for him to get around to\ndoing something about it whenever he's got time.
\n[21-Sep-2009 12:51:26] <Brad_K> and for Mailman, if you want to\npay Mark Sapiro, Tokio Kikuchi, or Barry Warsaw for some consulting,\nthey can provide that.
[21-Sep-2009 12:51:55] <Brad_K> mrayzenoss: sorry, i'm just ranting a bit....
[21-Sep-2009 12:52:21] <rmatte> this room doubles as a rant-room, he's used to it
[21-Sep-2009 12:52:31] <mrayzenoss> yeah, I've got thick skin and short memory
[21-Sep-2009 12:52:40] <mrayzenoss> look over there! shiny!
\n[21-Sep-2009 12:52:58] <rmatte> and apparently a passion for\nfarming since he was considering taking it up the day after the 2.4.3\nrelease
[21-Sep-2009 12:52:59] <Brad_K> and it helps that he's the Community Manager, even if he is a paid employee. ;-)
[21-Sep-2009 12:53:31] <mrayzenoss> my job is to keep the open source side moving, I don't really follow the support side
\n[21-Sep-2009 12:53:36] <rmatte> yeh, but he handles his job very\nprofessionally, he puts the community first and actually argues with\nthe higher ups to get stuff in to core
[21-Sep-2009 12:53:38] <mrayzenoss> I assume they're swamped, they always are
[21-Sep-2009 12:54:21] <rmatte> hire more engineers, I'm available
[21-Sep-2009 12:54:36] <mrayzenoss> if you want to relocate to Annapolis or Austin, we'd probably take you up on it
\n[21-Sep-2009 12:54:36] <Brad_K> well, this is something that\ncould be crowd-sourced, at least for some part of the problem space.
[21-Sep-2009 12:55:00] <mrayzenoss> Brad_K: are you familiar with http://zenoss.org ? that's where we crowd-source
\n[21-Sep-2009 12:55:38] <rmatte> yeh, I doubt I could do the\nrelocation, plus it's almost impossible to get a worker's visa for the\nU.S when you're canadian these days. My friend was supposed to move\ndown to L.A since he works for NBC Universal, and he could never get\nacross the border, so he's still working for them but from home here.
\n[21-Sep-2009 12:55:50] <Brad_K> yeah, i've tried the forums and\nmailing lists. i have yet to see a single response there to anything\ni've ever posted. the irc channel, however, has turned out to be very\nuseful.
[21-Sep-2009 12:56:19] <mrayzenoss> rmatte: Kells works for us from Calgary
\n[21-Sep-2009 12:56:46] <Brad_K> and the FAQs, wikis, and other\ndocumentation is frequently out-of-date, incomplete, or inaccurate.
[21-Sep-2009 12:56:56] <mrayzenoss> yeah, the new site should help.
\n[21-Sep-2009 12:56:59] <rmatte> yeh, but it'd depend if you guys\nwould be interested in having a remote engineer, which you probably\nwouldn't lol
[21-Sep-2009 12:57:01] <mrayzenoss> the old one was impossible to find stuff
[21-Sep-2009 12:57:02] <Brad_K> same with the knowledgebase, for that matter.
[21-Sep-2009 12:57:12] <rmatte> the new site is excellent
[21-Sep-2009 12:57:15] <mrayzenoss> Brad_K: there is an Enterprise KB
[21-Sep-2009 12:57:22] <rmatte> and I see that they fixed the download button on the main page
[21-Sep-2009 12:57:31] <rmatte> someone was a bit too hasty with the crop tool before apparently
[21-Sep-2009 12:58:07] <jb> enterprise knowledgebase?
[21-Sep-2009 12:58:08]\n<Brad_K> they need to get the site re-indexed, so that the search\nfunction works, and you can auto-link the old URLs to their new\nlocations.
[21-Sep-2009 12:58:25] <Brad_K> jb: yes, for paying enterprise customers.
[21-Sep-2009 12:58:31] <jb> where?
[21-Sep-2009 12:58:47] <Brad_K> so far, it has been less useful to me than the other documentation sournces.
[21-Sep-2009 12:58:55] <jb> i don't know about this KB..
[21-Sep-2009 12:59:09]\n<Brad_K> jb: if you're a paying customer, you should have access\nto their customer service portal, and from there you can access the KB.
[21-Sep-2009 12:59:13] <rmatte> Brad_K: the search works fine on the new site, I use it all the time
[21-Sep-2009 12:59:14] <jb> sec
[21-Sep-2009 12:59:45] <jb> ah I never noticed that..
[21-Sep-2009 13:00:00]\n<Brad_K> the times i've tried the search function, it didn't turn\nup much of anything useful -- not like the search function on the old\nsite.
[21-Sep-2009 13:00:36] <mrayzenoss> search.zenoss.com is working again too
[21-Sep-2009 13:00:40] <Brad_K> but then it has been a few days since i used it, so maybe they've fixed it siince then.
[21-Sep-2009 13:01:22] <mrayzenoss> today marks the 7 day anniversary of zenoss.org, so we're still ironing out issues
[21-Sep-2009 13:01:33] <mrayzenoss> but I think the email threading is going to fix today
[21-Sep-2009 13:01:39] <Brad_K> cool.
[21-Sep-2009 13:01:59] <rmatte> how can you even use the word "anniversary"?
[21-Sep-2009 13:02:01] <rmatte> lol
[21-Sep-2009 13:02:11] <rmatte> more like 7th day of survival
[21-Sep-2009 13:02:13] <rmatte> at this point
[21-Sep-2009 13:02:14] <rmatte>
[21-Sep-2009 13:02:31] <rmatte> It's definitely getting nicer day by day though
[21-Sep-2009 13:02:48] <rmatte> just need to iron out importing the code in the forum posts and fixing the wacky characters.
[21-Sep-2009 13:02:50] <Brad_K> next month, my wife and i will be coming up on our 3650th day anniversary. ;-)
[21-Sep-2009 13:03:05] <rmatte> Brad_K: haha
[21-Sep-2009 13:03:09] <rmatte> congrats
[21-Sep-2009 13:03:13] <Brad_K> thanks!
[21-Sep-2009 13:03:23] <rmatte> iancmcc: ping
[21-Sep-2009 13:03:30] <Brad_K> uhoh, we better be quiet now....
[21-Sep-2009 13:03:36] <iancmcc> that was fast
[21-Sep-2009 13:03:47] <rmatte> found an event console bug this morning...
[21-Sep-2009 13:03:51] <iancmcc> bring it
[21-Sep-2009 13:04:32]\n<rmatte> ok, so when you load the event console by default it\nshows you debug events as well, even if you haven't selected for it to\nshow them, then when you select one and click close the screen\nrefreshes and you can no longer see them, but it doesn't move the event\nto history...
[21-Sep-2009 13:04:40] <cgibbons> ian loves bugs
[21-Sep-2009 13:04:46]\n<rmatte> so to actually close the events out you need to\nspecifically click to view debug events, and then you can close them
[21-Sep-2009 13:05:01] <rmatte> oh, and another bug is if you scroll really fast from the top of the list to the bottom the list blanks out
[21-Sep-2009 13:05:09] <rmatte> those are the ones that I noticed today
[21-Sep-2009 13:05:17] <iancmcc> to the second: don't do that.
[21-Sep-2009 13:05:37] <iancmcc> hmm, ok, I'll look into both of those.
[21-Sep-2009 13:06:11]\n<iancmcc> The first seems pretty odd, dunno how it would show\ndebug events if the box isn't checked...there aren't a lot of moving\nparts between the console and the database. I shall investigate.
[21-Sep-2009 13:06:28] <iancmcc> Thanks, keep 'em coming.
[21-Sep-2009 13:08:52] <twm1010> iancmcc: I can tell you the priority icons are MUCH easier on the eyes than the colored entries
[21-Sep-2009 13:09:06] <iancmcc> Good. Did you vote?
[21-Sep-2009 13:09:10] <twm1010> I did.
[21-Sep-2009 13:09:39] <iancmcc> off-by-default is still winning, thankfully.
[21-Sep-2009 13:11:01] <twm1010> yeah, i found that i had to bump magnification up to at least 140% when the colors are on
[21-Sep-2009 13:11:15] <twm1010> the smaller bolded text doesn't work with the colors
[21-Sep-2009 13:11:33] <iancmcc> the colors are actually slightly different now
[21-Sep-2009 13:11:37] <iancmcc> less saturated
[21-Sep-2009 13:11:49] <iancmcc> better contrast. Should be in beta 3.
[21-Sep-2009 13:13:03] <mrayzenoss> beta 3 eta is Wednesday
[21-Sep-2009 13:13:23] <rmatte> yeh, I voted for off by default
[21-Sep-2009 13:13:28] <rmatte> looks cleaner
[21-Sep-2009 13:15:03]\n<rmatte> yeh, .zip zenpacks don't work well at all with King\nCrab, you should really just make it so that only .eggs can be installed
[21-Sep-2009 13:15:18] <rmatte> and all the zip zenpacks will need to be either converted or removed from the list
[21-Sep-2009 13:15:34] <venturaville> they don't work all that well in the _current_ versions
[21-Sep-2009 13:15:51] <rmatte> venturaville: I haven't had issues with them up until King Crab beta 2
[21-Sep-2009 13:16:04] <rmatte> (I didn't try King Crab beta 1, I'm sure the issues are there as well)
[21-Sep-2009 13:16:08] <rmatte>
[21-Sep-2009 13:16:22] <venturaville> your mileage may vary :-)
[21-Sep-2009 13:16:23] <djack> hellow world belgium calling
\n[21-Sep-2009 13:16:51] <rmatte> I have multiple servers running\n2.4.5 and they have several zip ZenPacks installed (which could be an\nissue for the upgrade)
[21-Sep-2009 13:17:05] <mrayzenoss> the release notes for King Crab will note that it's the last release that supports .Zip zenpacks
[21-Sep-2009 13:17:17] <djack> i installed the 2.5 but gives an error
[21-Sep-2009 13:17:36] <rmatte> mrayzenoss: you mean that 2.4.5 is the last release supporting them right?
[21-Sep-2009 13:17:46] <mrayzenoss> no, we're officially deprecating them in 2.5
[21-Sep-2009 13:17:57] <mrayzenoss> so if they're broken in 2.5, they gotta be fixed 1 last time
[21-Sep-2009 13:18:02] <rmatte> ah, so they'll still kinda sorta work but not really?
[21-Sep-2009 13:18:05] <mrayzenoss> yeah
[21-Sep-2009 13:18:18] <mrayzenoss> we'll load them, not guarantees the contents still work
[21-Sep-2009 13:18:27] <mrayzenoss> since they're Community-provided
\n[21-Sep-2009 13:18:45] <djack> anyhow beside the\ngetPingSTatus,uptimeStr,location i also need the Class where the device\nis in
[21-Sep-2009 13:18:49] <mrayzenoss> but be sure to Comment on the different ZenPacks, that's a new feature
\n[21-Sep-2009 13:19:20] <rmatte> there's a bit of an issue with\nusing the term "depricated" though. that term suggests that the\nfunction is still available and functioning properly, but that another\nmore enhanced function is now available.
[21-Sep-2009\n13:19:34] <rmatte> http://dev.zenoss.com/trac/ticket/5547 <---\nIn this case, it's not functioning properly for any zip packs
[21-Sep-2009 13:19:53] <rmatte> I'd suggest just preventing the installation of any .zip files all-together
[21-Sep-2009 13:20:02] <mrayzenoss> hmmm... I gotta go talk to bbibeault
[21-Sep-2009 13:20:05] <iancmcc> Oh, I thought we were fixing that
[21-Sep-2009 13:20:12] <jb> when zenoss queries windows machines via WMI, what port does it use?
[21-Sep-2009 13:20:19] <jb> isn't WMI RPC/DCOM based?
[21-Sep-2009 13:20:33] <cgibbons> yes
[21-Sep-2009 13:20:40] <cgibbons> so it'll start off on port 135, and then get redirected to a secondary DCOM port
[21-Sep-2009 13:20:49] <cgibbons> which you can configure on the Windows server if you want to restrict the port range
[21-Sep-2009 13:20:52] <jb> ugh
[21-Sep-2009 13:21:06] <jb> k
[21-Sep-2009 13:21:10] <cgibbons> want the link for that?
[21-Sep-2009 13:21:13] <jb> please
[21-Sep-2009 13:21:19] <cgibbons> you can do it via a Domain/Group Policy so you can deploy it consistently
\n[21-Sep-2009 13:21:28] <Brad_K> djack: so, where in Belgium are\nyou? My wife and I lived in Brussels for almost eight years.
[21-Sep-2009 13:21:38] <cgibbons> here you go: http://msdn.microsoft.com/en-us/library/ms809327.aspx
[21-Sep-2009 13:21:48] <jb> thanks
[21-Sep-2009 13:21:55] <djack> Brad_K between antwerp and brussels
\n[21-Sep-2009 13:22:31] <Brad_K> i used to drive that route four\ndays a week, commuting between Brussels and Eindhoven. where exactly\nbetween Antwerp & Brussels?
[21-Sep-2009 13:22:44] <jb> hrm, im having a sort of weird problem
[21-Sep-2009 13:22:54] <jb> zen.pysamba - Attempt to connect resulted in NT_STATUS_ACCESS_DENIED
[21-Sep-2009 13:23:03] <jb> although I think its pulling WMI perfmon counters fine
[21-Sep-2009 13:24:07] <jb> its just the services
[21-Sep-2009 13:24:22] <cgibbons> are you trying to use a non-admin user?
[21-Sep-2009 13:24:24] <djack> Brad_K a small village called Niel
[21-Sep-2009 13:24:29] <jb> nope.. same user I use for every box
[21-Sep-2009 13:24:44] <jb> every other box in this device class is fine
[21-Sep-2009 13:24:51] <cgibbons> okay, let's try somethign real quick. i can tell you where the misconfiguration is.
[21-Sep-2009 13:24:56] <jb> k
[21-Sep-2009 13:25:02] <cgibbons> do a wmic -d99 -U 'whomever' //server 'select name from win32_computersystem'
[21-Sep-2009 13:25:04] <cgibbons> that'll be noisy
\n[21-Sep-2009 13:25:45] <cgibbons> if you get a 'Failed to bind to\nuuid' message @ the bottom it's one thing, if you get a whole bunch\nmore network dump data and it keeps going before the access denied,\nit's something else
[21-Sep-2009 13:26:44] <jb> Failed to bind to uuid 4d9f4ab8-7d1c-11cf-861e-0020af6e7c57 - NT_STATUS_END_OF_FILE
[21-Sep-2009 13:26:44] <jb> ERROR: Login to remote object.
[21-Sep-2009 13:26:44] <jb> NTSTATUS: NT_STATUS_ACCESS_DENIED - Access denied
[21-Sep-2009 13:26:50] <cgibbons> okay yep
\n[21-Sep-2009 13:27:13] <cgibbons> so, when it fails that early\nyou've got a) bad username, b) bad password, c) user hasn't been\ngranted access to do remote logins to that server
[21-Sep-2009 13:27:20] <cgibbons> it's probably C in this case
[21-Sep-2009 13:27:29] <cgibbons> do you use any group policy stuff in your domain, or just local server configs?
[21-Sep-2009 13:27:29] <jb> odd
[21-Sep-2009 13:27:31] <jb> its a domain admin
[21-Sep-2009 13:27:46] <cgibbons> very odd then
\n[21-Sep-2009 13:28:47] <Brad_K> djack: ahh, yes. not too far\nfrom the A12. i took that route on occasion, when i was looking for\nsomething a bit more scenic and lower-stress than the A1/E19.
[21-Sep-2009 13:29:25] <djack> Brad_K that time as long passed
[21-Sep-2009 13:29:26] <Brad_K> djack: i definitely remember driving through Boom.
[21-Sep-2009 13:29:53] <djack> Brad_K that a mile from here
[21-Sep-2009 13:29:59] <djack> hmmm
\n[21-Sep-2009 13:30:22] <mrayzenoss> For everyone who was using\nthe email bridge, threading has been fixed. No more junk in the\nsubject lines
[21-Sep-2009 13:30:47] <Brad_K>\ndjack: anyway, you said you were having a problem with 2.5? i think\nthat iancmcc is the right guy to be talking to about that....
[21-Sep-2009 13:31:03] <djack> Brad_K
[21-Sep-2009 13:31:47] <mrayzenoss> rmatte: was 5547 a regression?
[21-Sep-2009 13:31:54] <mrayzenoss> rmatte: I'm reopening it
\n[21-Sep-2009 13:31:54] <djack> the 2.5 was for test i'm building\na external dashboard with application inteligence but i need to get the\nClass where the device is located out of dmd
[21-Sep-2009 13:32:33] <rmatte> mrayzenoss: I'm not sure if I had set it as such
[21-Sep-2009 13:33:01] <rmatte> nah, I never set it as regression
[21-Sep-2009 13:33:13] <mrayzenoss> I mean, did the author and version show up before?
[21-Sep-2009 13:33:29] <mrayzenoss> with zip zenpacks
[21-Sep-2009 13:33:37] <rmatte> yeh, they did
[21-Sep-2009 13:33:45] <rmatte> it works perfectly fine on my 2.4.5 installs
[21-Sep-2009 13:33:49] <iancmcc> djack: dmd.Devices.findDevice('mydevice').deviceClass().getPrimaryId()
[21-Sep-2009 13:34:04] <mrayzenoss> ok, reopened
[21-Sep-2009 13:34:08] <djack> iancmcc lets try
[21-Sep-2009 13:35:46] <rmatte> For example:
[21-Sep-2009 13:35:47] <rmatte> APC UPS PowerNet n/a Richard Palmer 0.5 No
[21-Sep-2009 13:35:47] <rmatte> Fortigate n/a Fabio Paracchini 1.0 No
[21-Sep-2009 13:35:47] <rmatte> HPTemperatureSensors n/a Trevor Seward 1.1 No
[21-Sep-2009 13:35:47] <rmatte> Is now:
[21-Sep-2009 13:35:47] <rmatte> APC UPS PowerNet n/a 0.1 No
[21-Sep-2009 13:35:49] <rmatte> Fortigate n/a 0.1 No
[21-Sep-2009 13:35:51] <rmatte> HPTemperatureSensors n/a 0.1 No
[21-Sep-2009 13:36:25] <mrayzenoss> rmatte: ok, just checking
[21-Sep-2009 13:36:28] <mrayzenoss> it's reopened
[21-Sep-2009 13:36:36] <mrayzenoss> might not get fixed, but it's open
[21-Sep-2009 13:36:39] <rmatte> k lol
[21-Sep-2009 13:36:52] <mrayzenoss> sev 3, there are bigger fish to fry
[21-Sep-2009 13:37:03] <rmatte> yeh, for sure
[21-Sep-2009 13:37:07] <mrayzenoss> but it may get inadvertently fixed
[21-Sep-2009 13:37:12] <mrayzenoss> if there are other .zip zenpack issues
[21-Sep-2009 13:37:25] <rmatte> yeh
[21-Sep-2009 13:37:57] <jb> cgibbons: ok think I figured it out.
\n[21-Sep-2009 13:37:58] <rmatte> I'll do some testing with that at\nsome point, just don't want to risk screwing up my install since I'm\ntesting other stuff today
[21-Sep-2009 13:38:05] <rmatte> I might just take a snapshot of the vm and try it tomorrow
[21-Sep-2009 13:40:10] <djack> iancmcc thanks it works :-) is there a list of this?
\n[21-Sep-2009 13:40:56] <iancmcc> djack: If you mean\ndocumentation, there's the dev guide; I'm not sure how comprehensive it\nis, however
[21-Sep-2009 13:41:11] <djack> iancmcc not really
\n[21-Sep-2009 13:41:37] <djack> well i have what i need for this\nmoment going to set up the mysql db and build the php interface above
\n[21-Sep-2009 13:41:40] <rmatte> the dev guide isn't very\ncomprehensive, most of the development stuff that I've learned has just\nbeen through lots of digging through other people's work and the forums\n
[21-Sep-2009 13:42:00] <mrayzenoss> djack: you can add it here: community/documentation/wiki/zendmd
[21-Sep-2009 13:42:01] <iancmcc> Yeah, might try cruising the API docs.
[21-Sep-2009 13:42:59] <djack> iancmcc i only read something when i fucked it up completely :-p
[21-Sep-2009 13:43:23] <iancmcc> Usually my MO as well.
[21-Sep-2009 13:44:23] <djack> everybody knows it when i pull up the pdf we are in real shit
[21-Sep-2009 13:44:29] <djack> hihi
[21-Sep-2009 13:44:47] <cgibbons> what was broke, jb?
\n[21-Sep-2009 13:45:25] <Brad_K> djack: isn't there an appropriate\nFlemish word that you can teach us for those terms? I know the French\nwords, but the Flemish ones would be so much more entertaining....
[21-Sep-2009 13:45:42] <rmatte> "Oh no, djack is READING! ABANDON SHIP!"
[21-Sep-2009 13:46:04] <djack> rmatte something like that ....
[21-Sep-2009 13:46:26] <djack> Brad_K Flemisch word for what ?
[21-Sep-2009 13:46:39] <Brad_K> fuck and shit?
[21-Sep-2009 13:46:56] <Brad_K> words, really. not single word, i hope.
\n[21-Sep-2009 13:47:25] <Brad_K> i'd love to be able to throw\nthose into casual conversation and completely confuse the hell out of\npeople....
[21-Sep-2009 13:47:26] <djack> Brad_K well we are very international so fuck and shit is also used here
[21-Sep-2009 13:47:41] <djack> ik zal je eens wat vertellen ;-)
[21-Sep-2009 13:48:00] <djack> nu denk ik niet dat iemand dit gaat verstaan
\n[21-Sep-2009 13:48:19] <Brad_K> well, far too many Flemish people\nare far too willing to speak English, and do so better than most\n'Merkins. They're like that, I know. But it would still be\ninteresting to learn them....
[21-Sep-2009 13:49:42] <djack> Brad_K we speak dutch french and english for the flemish part of my country
\n[21-Sep-2009 13:49:45] <Brad_K> i was especially amused when the\nwinner of the world French spelling bee was an English teacher in\nAntwerp, and he spoke five or six languages fluently, in addition to\nhis native Vlaams.
[21-Sep-2009 13:50:44]\n<Brad_K> And if you ask a Dutchman, Flemish is just a dialect of\nDutch. But if you ask a Flemish man, then Dutch and Flemish are two\ndistinct languages.
[21-Sep-2009 13:50:52] <Brad_K> yes, I know that joke.... ;-)
\n[21-Sep-2009 13:50:54] <djack> Brad_K i work in Rotterdam and\neven Dutch people and Flemish people speak the same language there is a\nworld of difference
[21-Sep-2009 13:52:04] <djack> Brad_K from where are you if i may ask?
\n[21-Sep-2009 13:52:23] <Brad_K> yes, like the fact that "worker"\nin Flemish means "slave" in Dutch. Imagine the chagrin of the Flemish\nCEO of a company in Brussels that was buying out their Dutch\ncounterpart, and he went to go give a speech at the Workers Council,\nand he *didn't* know the difference....
[21-Sep-2009 13:52:47] <rmatte> I speak french and english, no dutch though
[21-Sep-2009 13:53:29] <venturaville> I always hear about high and low dutch ... but not certain what the difference is
\n[21-Sep-2009 13:53:32] <Brad_K> ... Speaking for over an hour\nabout "... the workers ..." this, and " ... the workers ... " that, and\nall the while everyone in the room was hearing " ... the slaves ... "\nthis, and " ... the slaves ... " that.
[21-Sep-2009 13:53:33] <djack> rmatte you don't know what you are missing
\n[21-Sep-2009 13:54:33] <Brad_K> actually, there are many\ndifferent dialects of Dutch and Flemish. Not just High and Low, but\nalso North vs. South, East vs. West, and don't even get me started\nabout Frisian (the other official languages of Die Nederlands).
[21-Sep-2009 13:55:22] <djack> Fries yeah that is a completely other language
[21-Sep-2009 13:55:23] <rmatte> There are lots of different dialects of French as well
[21-Sep-2009 13:55:30] <djack> even for the dutch people
\n[21-Sep-2009 13:55:33] <Brad_K> one thing you quickly discover\nabout most Flemish people is that they are all amateur linguists,\ncompletely fluent in at least a half dozen languages, and very, very\nwilling to talk for many hours on the subject.
\n[21-Sep-2009 13:55:41] <rmatte> Canadian french is considered to\nbe "slangy" or "slum" French in comparison to parisian french
\n[21-Sep-2009 13:56:43] <djack> every town has its own version of\nflemish I'm from antwerp and my wife is from Roeselare (125 km from\nantwerp) they talk in a way that on tv they use subtitles
[21-Sep-2009 13:57:04] <rmatte> lol, that's crazy
[21-Sep-2009 13:57:20] <djack> rmatte its crazy here that a fact
[21-Sep-2009 13:57:29] <twm1010> jeebus, everytime i tab back to this window the conversation is completely different
[21-Sep-2009 13:57:43] <rmatte> twm1010: haha
[21-Sep-2009 13:59:18] <djack> lol
[21-Sep-2009 14:01:04]\n<Brad_K> And the language spoken by the Belgians is considered a\n"country" French, and they don't mean that in a nice way. It's like\nthey're the damn stupid country hicks to the north, and all the members\nof L'Academie Francaise are horrified.
[21-Sep-2009 14:02:00]\n<Brad_K> we learned this the hard way when we tried to speak\nfrench as we had learned it in Brussels when we went to visit various\nplaces in Paris and elsewhere in France -- like taking the Normandy\nInvasion tour.
[21-Sep-2009 14:02:36] <djack> Brad_K France is something else
[21-Sep-2009 14:03:03]\n<Brad_K> yes, we know. imo, no one in the world has the right to\nhate the French unless they've actually had to deal with them on\noccasion.
[21-Sep-2009 14:03:28] <venturaville> of course we have the same thing between the North and the deep south here in the US
[21-Sep-2009 14:03:34] <Brad_K> you only get the right to hate, loathe, and despise them once you've actually had dealings with them.
[21-Sep-2009 14:03:59] <Brad_K> venturaville: not like this. the French are far worse to their "country hicks" than we are to ours.
[21-Sep-2009 14:04:39]\n<Brad_K> djack: my wife and i now live in Austin, TX, which is\none of the few cities in the US where we could move back with as little\nculture shock as possible.
[21-Sep-2009 14:04:40] <djack> Brad_K I don't hate them its a nice country but the mentality is not what we are used to here in flanders
[21-Sep-2009 14:04:54]\n<venturaville> I'll take your word for it, though I don't forsee\na redneck from my home town fairing well in the blue blood states :-)
[21-Sep-2009 14:05:03] <djack> Brad_K well i lost my heart in New York
[21-Sep-2009 14:05:35]\n<Brad_K> venturaville: think about how blacks are still sometimes\ntreated in all-white areas of the south. that's closer.
[21-Sep-2009 14:06:23]\n<Brad_K> djack: and my wife and i truly loved Antwerp, Brugge,\nGhent, Liege, London, Barcelona, and various other cities there in\nEurope.
[21-Sep-2009 14:06:44] <Brad_K> oh, and Orval was another favourite of ours. Really good beer. ;-)
[21-Sep-2009 14:07:04] <venturaville> it tends to be more about affluent versus poor these days, but yeah I see your point
[21-Sep-2009 14:07:19] <djack> Brad_K beer that the one product that we can make ....
[21-Sep-2009 14:07:47] <djack> Inbev is from leuven
[21-Sep-2009 14:07:55] <Brad_K> Westvleteren is nice, but then I like all the Trappiste beers.
[21-Sep-2009 14:08:20] <djack> Westvleteren is very special
[21-Sep-2009 14:09:47] <Brad_K> I'm not so much for the more common beers. I prefer more specialist beers, like Kriek or Trappiste.
[21-Sep-2009 14:10:26] <djack> Brad_K haha
[21-Sep-2009 14:10:29] <Brad_K> In fact, most any Lambic is enough to get my attention.
[21-Sep-2009 14:10:48] <djack> Brad_K i'll mail you one ;-)
[21-Sep-2009 14:11:24] <rmatte> ooh, swag, beer swag
[21-Sep-2009 14:11:27] <rmatte>
\n[21-Sep-2009 14:12:15] <Brad_K> djack: actually, we have this\nplace here in Austin called Whole Foods, and I've been amazed at the\nkinds of beers that they stock. they even have some Belgian beers that\nI had never heard of.
[21-Sep-2009 14:12:41] <djack> Brad_K you should try them all
\n[21-Sep-2009 14:12:51] <mrayzenoss> Brad_K: check out the New\nBelgium 1554, that's my current favorite "fairly-easy-to-find beer"
[21-Sep-2009 14:13:08] <cgibbons> the rich hippy store!
[21-Sep-2009 14:13:12] <djack> well i more of a wine wiskey lover
\n[21-Sep-2009 14:13:20] <Brad_K> But please don't send any Stella\nArtois or Heineken. Those are so bad that the only thing they can do\nis export them -- and somehow, the rest of the world thinks they're\nactually good.
[21-Sep-2009 14:13:26] <mrayzenoss> cgibbons: you can get the beer other places
[21-Sep-2009 14:13:39] <djack> Heineken is NOT BELGIUM
[21-Sep-2009 14:13:54] <djack> Heineken is pissbeer
\n[21-Sep-2009 14:14:12] <Brad_K> djack: if you drink a bottle of\nBush Beer for every bottle of the ~500 so different types of Belgian\nbeer that there are, i'll take you up on that.
[21-Sep-2009 14:14:13] <djack> it's an insult
[21-Sep-2009 14:14:28] <Brad_K> yeah, well Stella isn't that much better.
[21-Sep-2009 14:15:03] <djack> Brad_K I'm jupiler
[21-Sep-2009 14:15:05] <mrayzenoss> speaking of beer... Zenoss is sponsoring the pre-party at Ohio LinuxFest, be sure to show up Friday night
[21-Sep-2009 14:15:06] <Brad_K> regretfully, i never managed to make it up to Stadsfeestzaal.
[21-Sep-2009 14:15:21] <djack> mrayzenoss you pay the tickets I'll fly over
[21-Sep-2009 14:15:25] <mrayzenoss>
[21-Sep-2009 14:15:33] <mrayzenoss> free training on Friday!
[21-Sep-2009 14:15:35] <djack> even will bring belgium beer
[21-Sep-2009 14:16:47] <djack> Brad_K its a shopping center but very nicely renovated
\n[21-Sep-2009 14:17:11] <Brad_K> One of my favourite places to\ndrink beer was in Falstaff's in Brussels. Not so much on the selection\nof beer, as the architecture. And occasionaly we could trick the\nunsuspecting tourist into drinking a few Bush Beers, and then trying to\nstand up (it's 12-14% alcohol).
[21-Sep-2009 14:17:24] <cgibbons> ouf
\n[21-Sep-2009 14:18:16] <Brad_K> I heard about the "24 Hours" at\nStadsfeestzaal, but I never managed to make it. Came close one year,\nbut not quite.
[21-Sep-2009 14:18:54] <Brad_K>\nBush Beer is kind of like Everclear -- you don't really notice how much\nalcohol you're drinking until it is much, much too late.
[21-Sep-2009 14:19:29] <kmtsun> is there a zenoss zenpack for freebsd?
[21-Sep-2009 14:19:37] <Brad_K> and to the people who are used to 3.2% horse piss, watch out!
[21-Sep-2009 14:19:38] <rmatte> everclear aka moonshine
[21-Sep-2009 14:19:57] <rmatte> I can't even imagine drinking 3.2% beer
[21-Sep-2009 14:20:13] <rmatte> our beer here rangers from 4.7 to 7
[21-Sep-2009 14:20:18] <rmatte> ranges*
[21-Sep-2009 14:20:21] <mrayzenoss> kmtsun: no, but we'd love an SSH one like the OpenSolaris and Linux ones
[21-Sep-2009 14:20:23] <rmatte> generally it's 5.5
[21-Sep-2009 14:20:35] <Brad_K> kmtsun: i don't see anything at community/zenpacks
[21-Sep-2009 14:20:48] <kmtsun> oh, kk thanks
[21-Sep-2009 14:20:52] <mrayzenoss> kmtsun: SNMP generally works
[21-Sep-2009 14:21:06]\n<Brad_K> we may have to hack on one of the Linux or OpenSolaris\nones, because we do have some FreeBSD machines we need to monitor.
[21-Sep-2009 14:21:08] <rmatte> mrayzenoss: had time to go through the pending list of ZenPacks yet?
[21-Sep-2009 14:21:13] <kmtsun> yes... generally
[21-Sep-2009 14:21:41] <mrayzenoss> rmatte: guess I'll go get to work
[21-Sep-2009 14:21:45] <rmatte> hehe
[21-Sep-2009 14:21:48] <djack> well i have snmp activated on a apc power switch ad it does not work
[21-Sep-2009 14:21:54] <Brad_K> mrayzenoss: i thought you were at work!
[21-Sep-2009 14:22:30] <djack> I'm in my sofa before the tv 9:30 in the evening here
[21-Sep-2009 14:22:42] <Brad_K> djack: did you try the zenpack at docs/DOC-3417 ?
[21-Sep-2009 14:22:45] <rmatte> Brad_K: ummm, have you tried just using the /Server/Darwin group?
[21-Sep-2009 14:22:56] <rmatte> Darwin is based on FreeBSD, so they are very similar
[21-Sep-2009 14:22:59] <rmatte> it would probably work
[21-Sep-2009 14:23:13] <djack> Brad_K let's see
\n[21-Sep-2009 14:23:35] <Brad_K> Darwin is an ancient Mach kernel,\nwith some FreeBSD userland, but also some NetBSD and OpenBSD thrown in\nfor good measure.
[21-Sep-2009 14:23:53] <djack> Brad_K well the powerswitch does not respone to an snmpwalk
[21-Sep-2009 14:23:59] <rmatte> Brad_K: yeh, but I'm talking about command structurewise
[21-Sep-2009 14:24:19] <rmatte> I'm pretty sure the Darwin device grouping collects data via SSH
[21-Sep-2009 14:24:33] <mrayzenoss> I'm at work, just haven't been working on ZenPacks
[21-Sep-2009 14:24:34] <rmatte> I'm just saying that the plugins would probably work fine for FreeBSD
[21-Sep-2009 14:24:42] <bill_no> hey all
[21-Sep-2009 14:24:54] <rmatte> hello
[21-Sep-2009 14:25:10] <djack> hey bill
\n[21-Sep-2009 14:25:54] <Brad_K> i've been a MacFanatic ever since\nDecember of 1983, when I saw an early prototype of the original Mac,\nbefore the official announcement during the SuperBowl in January of\n1984. I switched from MacOS 9 to Mac OS X as soon as it was stable\nenough to do so (right around the 10.2 days), and I'm on the latest\nrelease of Snow Leopard now. But that doesn't mean that I don't\nrecognize that this is one hell of a FrankenBeast OS that they've\ncreated. And now t
[21-Sep-2009 14:26:31] <rmatte> got cut off at And now t
[21-Sep-2009 14:26:51] <Brad_K> ... And now they're replacing gcc, of all things.
[21-Sep-2009 14:27:20] <cgibbons> you see the pumpkin ale in the fridge yet, matt?
\n[21-Sep-2009 14:27:30] <Brad_K> anyway, the Darwin stuff would\nprobably be a closer place to start than Linux or OpenSolaris. So,\nwhen the time comes, i'll try to remember that.
\n[21-Sep-2009 14:27:38] <bill_no> i found a web site that provides\na guide on how to reduce the perf graph to 8 hours - but it doesnt work\nand doesnt say which version of zenoss it applies to either (so ive\nfound) has anyone tried to reduce the perf graph time?
\n[21-Sep-2009 14:28:50] <mrayzenoss> bill_no:\ndocs/DOC-2445#HowdoIchangetheGraphintervalsfrom5minutestosomethingelse
[21-Sep-2009 14:29:00]\n<Brad_K> djack: are you sure you have the community and password\nsettings correct? and that you've got the network monitor module\ninstalled? because most APC equipment doesn't come with the network\nmonitor module by default, just an RJ-45 jack that is supposed to\nconnect to a custom USB/Rj-45 cable, and plug directly into a\nmanagement host.
[21-Sep-2009 14:29:04] <mrayzenoss> that's quite an anchor tag
[21-Sep-2009 14:29:47] <bill_no> oooo eek - I didnt find that one
[21-Sep-2009 14:30:43] <djack> Brad_K i can connect to the webinterface so i have the network module
[21-Sep-2009 14:31:06] <Brad_K> 'k
[21-Sep-2009 14:31:19] <djack> I see lots of ports that are open but not 161 ....
\n[21-Sep-2009 14:31:51] <Brad_K> ahh. sounds like someone else\nmay have reconfigured it to turn off SNMP. can you reconfigure it\nthrough the web interface?
[21-Sep-2009 14:32:06]\n<djack> first my idea was firewall issue but that is not the case\nbecause i can not do snmpwalks from the management network either
[21-Sep-2009 14:32:36] <djack> Brad_K snmp version 1 is activated
\n[21-Sep-2009 14:33:26] <djack> but i will figure it out tommorow\n.... i can not access the management network from my home so ....
[21-Sep-2009 14:34:49] <djack> the powerswitches are the last devices that are not monitored so it's the missing link
\n[21-Sep-2009 14:37:56] <Brad_K> djack: SNMP v1 may be the\nproblem. Most devices these days are V2c, if not V3. i'm not sure\nthat the zenpack will handle SNMPv1.
[21-Sep-2009 14:38:16] <djack> Brad_K i can use the v3 also
[21-Sep-2009 14:39:19] <Brad_K> check to see what the zenpack supports, then see if you can enable that on the UPS.
[21-Sep-2009 14:39:45] <twm1010> APC ups?
[21-Sep-2009 14:39:45] <djack> will do ;-)
[21-Sep-2009 14:39:57] <djack> twm1010 apc power switch
[21-Sep-2009 14:39:57] <Brad_K> twm1010: yes.
[21-Sep-2009 14:40:03] <twm1010> My two units are v1 ONLY.
[21-Sep-2009 14:40:34] <djack> the one we use are v1 and v3
[21-Sep-2009 14:41:02]\n<twm1010> In the SNMP config page for my UPS's I have to specify\nthe IP of the server that will be making SNMPGETs, or no go.
[21-Sep-2009 14:41:27] <twm1010> they're quite finicky in that respect
[21-Sep-2009 14:42:49] <djack> did anybody place a threshold on the swap usage
[21-Sep-2009 14:47:41] <Brad_K> okay, i'm off to lunch. bbl.
[21-Sep-2009 14:48:31] <djack> Lunch ?
[21-Sep-2009 14:49:21] <Brad_K> yeah, it's 2:49pm, but i got into the office late, so i'm going off to lunch late.
[21-Sep-2009 14:49:37] <djack> lol
[21-Sep-2009 14:49:48] <djack> if you get in late you can leave early
[21-Sep-2009 14:49:59] <Brad_K> i wish.
[21-Sep-2009 14:50:12] <djack> haha
[21-Sep-2009 14:50:17] <djack> have a nice lunch
\n[21-Sep-2009 14:53:44] <jb> cgibbons: so the problem was these\ntwo boxes run ISA.. and ISA and DCOM don't play together nicely
[21-Sep-2009 14:54:00] <cgibbons> crap... what's ISA?
[21-Sep-2009 14:54:14] <jb> MS internet security and accel server
[21-Sep-2009 14:54:18] <jb> basically a web proxy/firewall
[21-Sep-2009 14:54:23] <jb> we just use it to proxy
[21-Sep-2009 14:54:36] <jb> it feeds our websense policy servers (blocks certain websites, etc)
[21-Sep-2009 14:54:39] <cgibbons> well, that's convenient
[21-Sep-2009 14:54:44] <cgibbons> is it from MS or thirdparty?
[21-Sep-2009 14:54:46] <jb> MS
[21-Sep-2009 14:54:49] <cgibbons> ugh
[21-Sep-2009 14:54:52] <jb> heh yeah
[21-Sep-2009 14:54:56] <djack> jb squid squid squid :-p
\n[21-Sep-2009 14:54:57] <jb>\nhttp://blogs.technet.com/isablog/archive/2007/05/16/rpc-filter-and-enable-strict-rpc-compliance.aspx
[21-Sep-2009 14:55:05] <jb> yeah, i could use squid
[21-Sep-2009 14:55:24] <djack> haha
[21-Sep-2009 14:56:31] <djack> going to bed speak to you laters
[21-Sep-2009 15:03:26] <cmdln> hello
[21-Sep-2009 15:03:28] <twm1010> ISA is freaking....
[21-Sep-2009 15:03:39] <twm1010> particular.
[21-Sep-2009 15:05:57]\n<cmdln> I have zenoss reciving snmp traps from my APC ups. The\nUPS is also configured to send me emails. I have been getting notified\nof failed snmp logins on the UPS. the emails contain the ip of the\nclient that tried to connect but I can not seem to find the ip of the\nclient when I look throught the zenoss events
[21-Sep-2009 15:06:24] <jb> do you have the MIBs loaded?
[21-Sep-2009 15:06:31] <jb> click on the "Details" tab of the event
[21-Sep-2009 15:06:35] <jb> and see if you can see it there
[21-Sep-2009 15:06:53] <cmdln> I do have the MIBS loaded
[21-Sep-2009 15:07:13] <jb> http://thunder.jbdesign.net/~jbaird/zenoss/event-old.jpg
[21-Sep-2009 15:07:20] <jb> ^^ example
[21-Sep-2009 15:08:05] <cmdln> mmm i have fields with oids that are not mapped to names
[21-Sep-2009 15:08:17] <cmdln> but in the value column i dont see any ips anyway
[21-Sep-2009 15:08:29] <jb> so its not available via the trap..
[21-Sep-2009 15:08:34] <cmdln> ah
[21-Sep-2009 15:08:42] <cmdln> fail on apcs part
[21-Sep-2009 15:09:26] <cmdln> well that sucks
[21-Sep-2009 15:09:30] <jb> :/
[21-Sep-2009 15:09:30] <cmdln> thanks jb
[21-Sep-2009 15:09:32] <jb> np
[21-Sep-2009 15:19:03]\n<Signum> Evening... I'm currently evaluating monitoring systems\nsince Nagios and Cacti suck for our purpose. Choices left are opsview,\nzabbix and zenoss. So bear with me if I'm not 100% familiar with\nzenoss. : )
[21-Sep-2009 15:20:07] <Signum> I'm\ncurrently trying to find out how to monitor a mysql database. I'm\nalready monitoring the network interfaces and CPU load. But I'm not\nsure how I can add monitoring for th mysqld process or get further\nmysql statistics. Can someone push me in the right direction please?
\n[21-Sep-2009 15:20:40] <Signum> I suspect I'm going wrong\nsomewhere. Fancy features like the network map don't show me any\nsystems. Seems like I just forgot some stupid step.
[21-Sep-2009 15:21:55] <cmdln> Signum: you have a few different options for collecting the data
[21-Sep-2009 15:22:10] <cmdln> but pretty much any way you go you will need a script that will output the data you want
[21-Sep-2009 15:22:41] <cmdln> there is no agent that will monitor mysql automatically
[21-Sep-2009 15:23:00] <Signum> Okay, so I could use a shell/perl/whatever script to get the values I need.
[21-Sep-2009 15:23:05] <cmdln> yep
[21-Sep-2009 15:23:24] <Signum> The script would be run locally on the server that zenoss runs on, right?
[21-Sep-2009 15:23:31] <cmdln> and those values could be pulled either via ssh or as custom execs in snmpd
[21-Sep-2009 15:23:35] <Signum> ok
\n[21-Sep-2009 15:23:50] <Signum> I'm famliar with snmp/exec mess\nas that's how I've done thing with Nagios for the last 10 years.
[21-Sep-2009 15:23:54] <cmdln> no the script would run on the server thats being monitored
[21-Sep-2009 15:24:00] <cmdln> yeh
\n[21-Sep-2009 15:24:29] <Signum> That's one of the few big\nadvantages of zabbix btw. Their remote agent is pretty nifty and gets\nrid of the snmp/exec chaos. But anyway.
[21-Sep-2009 15:24:38] <cmdln> yeh
[21-Sep-2009 15:24:44] <cmdln> i used zabbix for a while
[21-Sep-2009 15:24:48] <cmdln> the agent was nice
\n[21-Sep-2009 15:25:16] <Signum> Currently no "OS Processes" or\n"IP services" are shown in the "OS" tab. Do I need to assign a template\nto get that working? I have just added that device via SNMP and get\ninformation about CPU load, network load and routes.
\n[21-Sep-2009 15:25:18] <cmdln> but then you just have to monitor\nso many things that cant run an agent so it lost its luster for me
[21-Sep-2009 15:25:39] <cmdln> for os processes i think you would need to add processes to monitor
\n[21-Sep-2009 15:25:50] <Signum> (for me zabbix big con is the\nchaotic GUI and the lot of clicking that has to be done for even simple\nmonitoring tasks)
[21-Sep-2009 15:25:59] <cmdln> not sure about the ipservices
\n[21-Sep-2009 15:26:19] <cmdln> i suppose if the device was\nrunning a very standard service like http it would pick it up\nautomatically
[21-Sep-2009 15:26:28] <Signum> If\nI try "Add processes" from the menu then I get a popup asking me for\nthe "Process Class" but the choice is empty.
[21-Sep-2009 15:26:53] <cmdln> ?
[21-Sep-2009 15:27:11] <Signum> Yes, I was surprised, too.
[21-Sep-2009 15:27:12] <cmdln> try adding from the processes class
[21-Sep-2009 15:27:20] <Signum> Even via SNMP zenoss should be able to fetch the list of running processes.
[21-Sep-2009 15:27:22] * Signum tries
[21-Sep-2009 15:29:34] <Signum> hmmm
[21-Sep-2009 15:33:01] <Signum> Cool, that worked indeed.
\n[21-Sep-2009 15:35:19] <Signum> Erm. Just if I add a process\ncalled "foo" but there is no process "foo" on the system I also get a\nstatus=UP. Very strange.
[21-Sep-2009 15:35:53] <cmdln> weird
[21-Sep-2009 15:37:17] <Signum> How would zenoss find out if a process is running. Via snmp?
[21-Sep-2009 15:37:42] <Signum> I read that zenoss is very snmp-dependent regarding its checks.
[21-Sep-2009 15:38:06] <mrayzenoss> windows boxes use WMi
[21-Sep-2009 15:38:12] <mrayzenoss> for Services
[21-Sep-2009 15:38:22] <Signum> True. This is a Debian server though.
\n[21-Sep-2009 15:38:47] <mrayzenoss> take a look at\nhttp://www.skills-1st.co.uk/papers/jcurry.html and\nhttp://www.skills-1st.co.uk/papers/jane/process_monitoring.pdf
[21-Sep-2009 15:38:59] <Signum> tnx
[21-Sep-2009 15:40:05] <baffle> Signum: But, yeah, process information is gathered via SNMP.
\n[21-Sep-2009 15:41:10] <Signum> baffle: Should I have seen the\nlist of running processes in the "OS Processes" then? SNMP read access\nuses the right community string and should return everything (not just\n"system").
[21-Sep-2009 15:43:24] <kmtsun> how do add the nmap plugin to zenoss?
[21-Sep-2009 15:45:19] <baffle> Signum: If the device (host) is in a device-class where it models this, yes.
[21-Sep-2009 15:53:21] <kmtsun> oh, so basically you make a user command using nmap commands
\n[21-Sep-2009 15:59:26] <Brad_K> hmm. does anyone know of any\nissues in upgrading from Enterprise 2.4.2 to Enterprise 2.4.5 on\nRHEL5? Is it really just as simple as pkgadd or yum update?
[21-Sep-2009 16:03:32] <mrayzenoss> this is the only issue I'm aware of in 2.4.5: http://dev.zenoss.org/trac/ticket/5494
[21-Sep-2009 16:03:37] <mrayzenoss> and the patch is available
\n[21-Sep-2009 16:23:11] <Signum> Thanks for your time. I'll\ncontinue my studies tomorrow. Although Zenoss still feels a little\nopaque it's closer to what I want to achieve.
\n[21-Sep-2009 16:23:35] <Signum> I'm one of the harder customers\nwho are not happy if something works by accident but who try to\nunderstand their tools.
[21-Sep-2009 16:29:09] <jb> Brad_K: i just went from 2.4.2 -> 2.4.5
[21-Sep-2009 16:29:13] <jb> no problems..
[21-Sep-2009 16:29:15] <jb> just update the RPM's
[21-Sep-2009 16:32:55] <Brad_K> cool. thanks!
[21-Sep-2009 17:39:10] <Brad_K> will be giving us a presentation on the bcfg2 configuration management tool.
[21-Sep-2009 17:39:34] <Brad_K> whoops, wrong window.
\n[21-Sep-2009 17:40:20] <Brad_K> Actually, I will repeat what I\njust told ckrough privately, for anyone in this channel who is in\nAustin and interested:
[21-Sep-2009 17:40:24]\n<Brad_K> LOPSA-Austin meeting Tuesday night, 22 Sep 2009 at 6:30\npm at the Mangia Pizza on Guadalupe.
[21-Sep-2009 17:40:36] <Brad_K> Sol Jerome of the Institute for Computational Engineering and Science at UT
[21-Sep-2009 17:40:37] <Brad_K> will be giving us a presentation on the bcfg2 configuration management tool.
[21-Sep-2009 22:13:37] <mloven> hey all. with the new community site, are the irc dev chats hosted anywhere?
[22-Sep-2009 00:00:46] [disconnected at Tue Sep 22 00:00:46 2009]
[22-Sep-2009 00:00:46] [connected at Tue Sep 22 00:00:46 2009]
[22-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
[22-Sep-2009 09:00:02] <ckrough> forums aren't rendering """
[22-Sep-2009 09:00:18] <mrayzenoss> new posts or old posts?
[22-Sep-2009 09:00:31] <mrayzenoss> Jive is looking at a way to fix all the old posts
[22-Sep-2009 09:00:41] <ckrough> Does anyone know if "Config Cycle Interval" pushes an entire config or just changes?
[22-Sep-2009 09:00:48] <ckrough> mrayzenoss: old post
[22-Sep-2009 09:00:53] <mrayzenoss> yeah, known issue
[22-Sep-2009 09:00:59] <ckrough> message/27767#27767
[22-Sep-2009 09:01:02] <ckrough> kk
[22-Sep-2009 09:08:39] <rmatte> ckrough: I believe it's a full config
[22-Sep-2009 09:10:11] <ckrough> bleh, brings the master to a halt
[22-Sep-2009 09:13:19] <ckrough> what is "# persistent cache file name " in zenhub.conf exactly?
[22-Sep-2009 09:23:45] <rmatte> that would be a commented line
[22-Sep-2009 09:24:13] <ckrough> heh
[22-Sep-2009 09:24:15] <rmatte> it would most likely refer to a zope cache file (.zec)
[22-Sep-2009 09:24:25] <ckrough> I mean, what exactly are we persistently caching
[22-Sep-2009 09:24:44] <ckrough> hard to tune the setting because I dont know what it's managing
[22-Sep-2009 09:28:27] <rmatte> caching Zope database transactions most likely
[22-Sep-2009 09:28:36] <rmatte> the more common transactions get cached to improve performance
[22-Sep-2009 09:29:38] <ckrough> I think this is all object cache
[22-Sep-2009 09:29:42] <ckrough> not positive
[22-Sep-2009 09:36:51] <rmatte> may be
[22-Sep-2009 10:06:24] NoxDeu is now known as NoxIn
[22-Sep-2009 10:06:50] NoxIn is now known as NoxIn-
\n[22-Sep-2009 10:17:46] <drtns> Morning all I have never had to\ndownload and install a community zenpack before but I find myself\nwanting to install the HP Procurve one. \ndocs/DOC-3486 I am totally new and use to\nRPM's so I am having a harder time even figuring out how to download\nit. is it the zip file at the bottom? how do I grap the "egg"
[22-Sep-2009 10:18:34] <mrayzenoss> weird, it's appears to have been uploaded as a .zip
[22-Sep-2009 10:18:38] <mrayzenoss> I'm looking into it
[22-Sep-2009 10:19:37] <drtns> thanks mrayzenoss
[22-Sep-2009 10:19:45] <mrayzenoss> hmmm... maybe it really is a .zip
[22-Sep-2009 10:19:52] <mrayzenoss> I guess this is an older ZenPack
[22-Sep-2009 10:20:04] <mrayzenoss> .zip style zenpacks were the precursor to .eggs
[22-Sep-2009 10:20:43] <mrayzenoss> so if you go to Settings->ZenPacks
[22-Sep-2009 10:20:59] <mrayzenoss> you can select "Install ZenPack" from the page menu
[22-Sep-2009 10:21:02] <mrayzenoss> and upload it
[22-Sep-2009 10:21:55] <rmatte> mrayzenoss: that zenpack is a zip
[22-Sep-2009 10:22:03] <rmatte> but for some reason the site has zipped up every single zenpack
[22-Sep-2009 10:22:06] <rmatte> even the egg packs
\n[22-Sep-2009 10:22:30] <drtns> I get this when I try and install\n- ERROR: zenpack command failed. Reason: [Errno 20] Not a directory:\n'/opt/zenoss/Products/HPProcurveMap.py/skins'
[22-Sep-2009 10:22:56] <rmatte> drtns: what is the whole filename of the pack?
[22-Sep-2009 10:23:31] <drtns> HPProcurveMap.py
[22-Sep-2009 10:23:51] <rmatte> hunh?
[22-Sep-2009 10:23:58] <rmatte> no, the file that you downloaded the pack as
[22-Sep-2009 10:24:04] <mrayzenoss> drtns: you don't need to unzip it
[22-Sep-2009 10:24:17] <mrayzenoss> oh wait, it's not even a ZenPack
\n[22-Sep-2009 10:24:17] <drtns> The install said to make mappings\nfor Devices and Events... but maybe it needs a Product mapping too?
[22-Sep-2009 10:24:17] <drtns> \\
[22-Sep-2009 10:24:17] <rmatte> oh
[22-Sep-2009 10:24:19] <rmatte> it's HPProcurveMap.py.zip
[22-Sep-2009 10:24:24] <rmatte> what the heck happened there
\n[22-Sep-2009 10:24:27] <etank> i am trying to run wmic from the\nzenoss server but i am getting this as a result 'NTSTATUS: NT code\n0xc002001b - NT code 0xc002001b'
[22-Sep-2009 10:24:40] <mrayzenoss> I'm looking into it
\n[22-Sep-2009 10:24:48] <rmatte> drtns: that's the site's fault\nnot yours, delete that stuff and mrayzenoss will get the proper file up
[22-Sep-2009 10:25:07] <etank> wmic -u 'AD/user%password' //host.domain.com "select name from Win32_Product"
[22-Sep-2009 10:25:13] <etank> that i the query that i am trying
\n[22-Sep-2009 10:25:27] <drtns> mrayzenoss thanks... I just\nwanted to get something that might be a little nicer to monitor my\nswitch then the stock /Network/Switch
[22-Sep-2009 10:25:29] <etank> i can get the data when running wmic locally on the windows 2003 server
[22-Sep-2009 10:26:32] <rmatte> try...
[22-Sep-2009 10:26:45] <rmatte> 'AD/user'%'password'
[22-Sep-2009 10:26:58] <rmatte> actually nevermind
[22-Sep-2009 10:27:07] <rmatte> I'm looking at something different, that won't work
[22-Sep-2009 10:27:08] <rmatte> hmmm
[22-Sep-2009 10:27:34] <rmatte> I think it's -U though
[22-Sep-2009 10:27:37] <rmatte> not -u
[22-Sep-2009 10:28:00] <etank> actually i do have a -U
[22-Sep-2009 10:28:06] <rmatte> k
[22-Sep-2009 10:28:42] <etank> ERROR: Retrieve result data.
[22-Sep-2009 10:28:42] <etank> NTSTATUS: NT code 0x80041001 - NT code 0x80041001
[22-Sep-2009 10:28:47] <etank> taht is the full error
[22-Sep-2009 10:29:09] <rmatte> ah
[22-Sep-2009 10:29:11] <rmatte> try this...
[22-Sep-2009 10:29:23] <rmatte> "Select Name from Win32_Product"
[22-Sep-2009 10:29:24] <etank> it is having a hard time with this "Problem while executing plugin zenoss.wmi.SoftwareMap"
[22-Sep-2009 10:29:32] <rmatte> Linux is quite case sensitive
[22-Sep-2009 10:30:01] <etank> that worked
[22-Sep-2009 10:30:01] <rmatte> oh, well if it's having problems with that then it's most likely some configuration issue on the server itself
[22-Sep-2009 10:30:21] <etank> the same query seemed to work on different machines
\n[22-Sep-2009 10:30:21] <rmatte> you say you can get the data\nlocally, have you tried grabbing the data from another windows box or\nsomething?
[22-Sep-2009 10:30:29] <etank> yes
[22-Sep-2009 10:30:36] <rmatte> ok, so it was your case
[22-Sep-2009 10:30:47] <rmatte> like I said, quite case sensitive
[22-Sep-2009 10:31:03] <etank> first i had to get it so that i could query locally
[22-Sep-2009 10:31:03] <rmatte> unlike windows which doesn't care about case
[22-Sep-2009 10:31:06] <etank> stupi WMI
[22-Sep-2009 10:31:16] <rmatte> lol
[22-Sep-2009 10:31:16] <etank> s/stupi/stupid/
[22-Sep-2009 10:31:27] <rmatte> we just use SNMP to monitor windows servers
[22-Sep-2009 10:31:32] <etank> i had to blow away the wmi repository to get it to work
[22-Sep-2009 10:31:50] <etank> we are using Enterprise so we were testing the functionality of wmi for monitoring
[22-Sep-2009 10:32:14] <rmatte> ah
[22-Sep-2009 10:32:28] <rmatte> we're using core so we use a mix of WMI and SNMP
[22-Sep-2009 10:32:37] <rmatte> WMI just for monitoring services when needed
[22-Sep-2009 10:32:51] <rmatte> we don't even use it for the event console monitoring since it's way too noisy
[22-Sep-2009 10:32:55] <rmatte> would take like a year to tune it
[22-Sep-2009 10:32:56] <rmatte> lol
[22-Sep-2009 10:33:13] <etank> it is very noisy for sure
[22-Sep-2009 10:33:37] <etank> but it did catch an issue that we would not have known about so that is awesome
[22-Sep-2009 10:35:51] <rmatte> what issue?
[22-Sep-2009 10:36:37] <etank> someone turned off one of our domain controllers in india
[22-Sep-2009 10:36:55] <etank> we got alerts from our dc's here in the states that they could not replicate
[22-Sep-2009 10:37:09] <etank> without zenoss monitoring the boxes here we would have missed that
[22-Sep-2009 10:37:18] <rmatte> ah
[22-Sep-2009 10:37:29] <rmatte> well, we do monitor event logs, but with traps
[22-Sep-2009 10:37:36] <rmatte> which only send out for more critical events
[22-Sep-2009 10:38:04]\n<rmatte> then again the event log severity is tunable, but the\nsnmp traps seem to be a bit more intelligent about what issues are\nactually reported
[22-Sep-2009 10:39:16] <mrayzenoss> speaking of events... Jane Curry's encore presentation is in 20 minutes
[22-Sep-2009 10:39:45] <rmatte> it's the exact same presentation as last time, correct?
[22-Sep-2009 10:40:22] <mrayzenoss> yeah, Jane's updated her slides
[22-Sep-2009 10:40:29] <etank> we are going to "borrow" from here papers tomorrow to present on Zenoss to our higher ups
[22-Sep-2009 10:40:36] <mrayzenoss> excellent
[22-Sep-2009 10:40:37] <rmatte> ok, not really any point in me attending this time around then
[22-Sep-2009 10:40:53] <mrayzenoss> all the Zenoss docs are Creative Commons, so feel free to reuse them as needed
[22-Sep-2009 10:42:21] <etank> with any luck we will get the go-ahead for a purchase of enterprise
[22-Sep-2009 10:43:24] <etank> wmi is so stupid
[22-Sep-2009 10:43:46] <etank> it seems that you have to run the queries locally sometimes to get them to work remotely
[22-Sep-2009 10:57:29] <ckrough> mrayzenoss: dial in number for conference?
[22-Sep-2009 10:57:39] <mrayzenoss> it's on the Info tab
[22-Sep-2009 10:57:49] <mrayzenoss> 866-699-3239
[22-Sep-2009 10:57:55] <ckrough> not on the tab
[22-Sep-2009 10:58:03] <ckrough> event number is, phone number not
[22-Sep-2009 10:58:53] <mrayzenoss> Call-in toll-free number (US/Canada): 866-699-3239
[22-Sep-2009 10:58:53] <mrayzenoss> Call-in toll number (US/Canada): 1-408-792-6300
[22-Sep-2009 10:58:53] <mrayzenoss> Access code: 753 961 267
[22-Sep-2009 10:59:00] <ckrough> thanks
[22-Sep-2009 10:59:03] <ckrough> im in
[22-Sep-2009 10:59:45] <mrayzenoss> anyone else joining the webex?
\n[22-Sep-2009 11:10:54] <drtns> Some advice please. I have a web\ncluster behind a load blancer IP.... I want tot use the HTTP monitor\ntemplate to monitor a few vhosts on this cluster for performance. I\nneed to add the load balancer IP as a device to do this but it's not a\n/Server/SOMETHING what is the /HTTP and /WEB classes for? and would\nthey be what I want to use?
[22-Sep-2009 11:11:41] <drtns> I also guess /Ping would be an option?
\n[22-Sep-2009 11:12:20] <rmatte> drtns: you could just create a\ncustom class as well if you wanted, and just go through the zProperties\nfor the class and disable/enable anything you like
\n[22-Sep-2009 11:12:56] <drtns> rmatte yes.. possible but If /HTTP\nand or /WEB or /PING do what I need it may be better then re inventing\nthe wheel
[22-Sep-2009 11:13:12] <drtns> trying to remember how to figure out what they do thogh
[22-Sep-2009 11:13:16] <drtns> though
[22-Sep-2009 11:13:32] <rmatte> well /Ping is for devices that you only want to monitor via ping
[22-Sep-2009 11:13:37] <rmatte> so that would most likely work fine
[22-Sep-2009 11:13:49] <rmatte> then you'd just need to apply the http_monitor template to the device
[22-Sep-2009 11:13:59] <rmatte> and customize it
[22-Sep-2009 11:14:10] <drtns> rmatte yes I was referring mostly to what the /HTTP and /Web classes do.
[22-Sep-2009 11:14:41] <rmatte> we put ours under /Application/HTTP
[22-Sep-2009 11:15:09] <rmatte> since that's the group that that zenpack creates I believe
\n[22-Sep-2009 11:16:28] <drtns> nod I am not even there though. I\nam trying to figure out the best way to get the device in there...\nwhere the device is just an IP bound to a load balancer pool and not a\nphysical server..
[22-Sep-2009 11:16:44] <drtns> if it was just one server that had the sites on it that would be a lot easier.
[22-Sep-2009 11:18:10] <rmatte> I just did this...
[22-Sep-2009 11:18:23] <rmatte> I put the device with the IP in to /Applications/HTTP
[22-Sep-2009 11:18:35] <rmatte> I then went to the device's templates tab and bound the HttpMonitor template
\n[22-Sep-2009 11:19:48] <drtns> right... I need a device to bind\nit to... that's my debate how best to create the device.. I have 8\nservers serving the same website behind a single loadblancer ip. so\nit's not good to bind it to any one of the 8 of those
\n[22-Sep-2009 11:20:35] <drtns> I need to create a device based on\nthe load balancer IP then bind the template to the new device and add\nsites to monitor to that
[22-Sep-2009 11:22:30] <rmatte> I created a local copy of the template and renamed it to whatever
[22-Sep-2009 11:22:30] <rmatte> I then configured it to monitor the specific vhost that I wanted to monitor
[22-Sep-2009 11:22:30] <rmatte> I then created another local copy of the HttpMonitor template and repeated the steps
\n[22-Sep-2009 11:22:30] <rmatte> I now have 4 different\nhttpmonitor templates named different things assigned locally to the\ndevice
[22-Sep-2009 11:22:30] <rmatte> And I have graphs for each
\n[22-Sep-2009 11:22:50] <rmatte> (sorry for the delay, I'm IRCing\nthrough my server at home and my home connection seems to be a bit flaky
[22-Sep-2009 11:25:32] <rmatte> eugh it keeps freezing
[22-Sep-2009 11:26:07] <rmatte2> there, now if it freezes at least I can still chat
\n[22-Sep-2009 11:27:36] <rmatte2> when you setup the datasource on\neach template, just make sure that you only populate the host name\nfield with the hostname you want to monitor
[22-Sep-2009 11:27:44] <rmatte2> don't bother filling out the IP Address field
[22-Sep-2009 11:28:04] <rmatte2> actually, I think I manually created each of these templates come to think of it
[22-Sep-2009 11:28:26] <rmatte2> did add template, then added an HttpMonitor type datasource
[22-Sep-2009 11:29:02]\n<rmatte2> then made the name the name of the vhost, made host\nname the name of the vhost as well, set port to 443 and checked use ssl\n(since it's https).
[22-Sep-2009 11:29:11] <rmatte2> then created a graph
[22-Sep-2009 11:29:14] <rmatte2> repeated for each host
[22-Sep-2009 11:29:34] <rmatte2> one graph for size, and one graph for time for each
[22-Sep-2009 11:30:04] <rmatte2> oh, in the URL field I put the path as well
[22-Sep-2009 11:30:17] <rmatte2> monitoring Zenoss login page so I put: /zport/acl_users/cookieAuthHelper/login_form
[22-Sep-2009 11:30:45] <rmatte2> I set redirect behaviour as follow
[22-Sep-2009 11:30:57] <rmatte2> hope that helps
[22-Sep-2009 11:43:05] <Troubadix09> hello, I have a strange behviour with two performance-graphs polling via snmp from one Windows-Server ...
[22-Sep-2009 11:43:40] <Troubadix09> ups ... behaviour i mean
[22-Sep-2009 11:44:11] <rmatte2> can you describe the issue?
\n[22-Sep-2009 11:45:19] <Troubadix09> rmatte2: always on one\nweekend of a month i have a nightly scheduled restart of some of my\nwindows dcs ...
[22-Sep-2009 11:46:09] <Troubadix09> last weekend after this restart on one windows server two performance graphs stop graphing
[22-Sep-2009 11:47:09] <Troubadix09> at the path of this device ($ZENHOME/perf/Devices/xxxxx/ the two rrd-files are missing
[22-Sep-2009 11:47:40] <Troubadix09> but they shown in Web-GUI
[22-Sep-2009 11:47:50] <Troubadix09> but stop graphing
[22-Sep-2009 11:48:32] <Troubadix09> I stop zenoss, deleting *.zec and *.pi in $ZENHOME/var and start zenoss
[22-Sep-2009 11:48:44] <Troubadix09> ups ... *.pid i mean
[22-Sep-2009 11:49:10] <rmatte2> are there any debug messages in the event console?
\n[22-Sep-2009 11:49:35] <Troubadix09> and after this the two\nperformance-graph start drawing and the missing rrd-files comes back\nagain
[22-Sep-2009 11:50:41] <rmatte2> ok, has this happened more than once?
[22-Sep-2009 11:50:54] <Troubadix09> no, nothing, all worked fine, no error messages in perfsnmp.log
[22-Sep-2009 11:51:10] <rmatte2> ok, well if it has only happened once I'd just write it off as a fluke
[22-Sep-2009 11:51:18] <rmatte2> if it happens again then I'd really look in to it
[22-Sep-2009 11:51:39] <rmatte2> the RRD files shouldn't have gotten magically deleted
\n[22-Sep-2009 11:52:50] <Troubadix09> strange, because the\nrrd-files are missing in the linux-directory and come back, but in the\ngraph there is only a gap from restart the windows-server to restart\nzenoss
[22-Sep-2009 11:53:54] <pokui> hi all,\nI'm looking through the zenoss FAQ and I can't seem to get a set of\ndifferences between the community and enterprise editions. could\nsomeone point me in the right direction?
[22-Sep-2009 11:54:43] <Troubadix09> rmatte2: it happened once as i recall
[22-Sep-2009 11:58:05] <mrayzenoss> pokui: the datasheets are here: http://www.zenoss.com/product/network-management
\n[22-Sep-2009 11:58:26] <drtns> rmatte I created a local copy of\nmy httpd monitor template. if I want to grap 3 sites on the same\nserver do you add 3 data sources to the same template and add them to\nthe graph? or do my bind multiple copies of the httpd monitor template\nand graph each site seperately?
[22-Sep-2009 11:58:44]\n<pokui> mrayzenoss: thanks. ticking off one by one where a\ndifference is mentioned at docs/DOC-2445\nwas becoming tiring.
[22-Sep-2009 11:59:47] <mrayzenoss> you can probably edit and update that doc
[22-Sep-2009 12:02:58] <rmatte2> Troubadix09: you sure you're looking at the right directory for the RRD files?
[22-Sep-2009 12:03:39] <rmatte2> drtns: you can do it with multiple data sources on one template, or multiple templates
[22-Sep-2009 12:03:55] <rmatte2> drtns: I just happened to do it with multiple templates
[22-Sep-2009 12:04:45] <rmatte> hmmm, my home connection seems to have straightened out a bit...
[22-Sep-2009 12:05:17] * rmatte smacks rmatte2 while saying "Stop hitting yourself! Stop hitting yourself!"
[22-Sep-2009 12:05:40]\n<Troubadix09> rmatte2: thats what i thought first too, not\nbelieving my eyes, control it again and ... yes, it was the right\ndirectory.
[22-Sep-2009 12:06:07] <rmatte> Troubadix09: no idea, I've been using Zenoss for almost a year and I've never seen behaviour like that
[22-Sep-2009 12:07:01] <rmatte> It'll be a year next month I believe
[22-Sep-2009 12:07:14] <rmatte> Just in time for King Crab
[22-Sep-2009 12:09:43] * rmatte sneezes
[22-Sep-2009 12:10:40] <Troubadix09> yeah, waiting for 2.5 release ;-)
[22-Sep-2009 12:11:29] <rmatte> yeh, hopefully it's relatively bug free by the time it rolls around
[22-Sep-2009 12:11:43] <rmatte> considering the last 3 releases were bugville
\n[22-Sep-2009 12:13:17] <Troubadix09> rmatte: i agree with you, i\nuse it for a year too, long time with 2.3.3, and work fine
\n[22-Sep-2009 12:13:52] <rmatte> 2.3.3 was fine, the only bug was\nthat it wouldn't properly automatically set product/model info for\ndevices when modelling
[22-Sep-2009 12:14:03] <rmatte> now 2.4.5 does that properly, but you can no longer set the info manually
[22-Sep-2009 12:14:11] <rmatte> I'd rather be able to set it manually than automatically
[22-Sep-2009 12:14:43] <rmatte> quite honestly, I wish we were still at 2.3.3, it would be a lot less headaches lol
\n[22-Sep-2009 12:15:50] <Troubadix09> rmatte: i first install on\nubuntu server with stack-installer, but haven't luck to keep it stable\nwith at least 50 Devices
[22-Sep-2009 12:16:20]\n<rmatte> we're using stack installer on ubuntu and I have 340\ndevices monitored on one of our Zenoss instances
\n[22-Sep-2009 12:17:32] <Troubadix09> rmatte: now i use the\nvmware-appliance, and it would be near by stable working with over one\nhundred devices on this instance :-)
[22-Sep-2009 12:17:58] <Troubadix09> but it cost me some time to customize it to work correctly
[22-Sep-2009 12:18:01] <rmatte> well at least they fixed the issues that I reported with the reports
\n[22-Sep-2009 12:18:42] <rmatte> Troubadix09: It doesn't matter\nwhat platform you run it on, quite honestly, the vmware-appliance is\nprobably the least efficient method of running it in my opinion
[22-Sep-2009 12:18:51] <rmatte> It's all about how you customize it
[22-Sep-2009 12:18:59] <rmatte> and clean up the monitoring as much as possible
\n[22-Sep-2009 12:19:12] <rmatte> if you do absolutely no tuning at\nall then you're obviously going to have performance issues
\n[22-Sep-2009 12:20:22] <rmatte> It also depends on how noisy the\ndevices are, if they are constantly flooding you with traps and syslogs\nthen you're not going to be able to monitor as many devices because\nZenoss is hung up processing all of that stuff
\n[22-Sep-2009 12:22:56] <Troubadix09> yes, i agree with you, i\nonly monitoring what i need, not what is possible, the devices are god\nbeware not so noisy, and the vm running on a high performance\nesx-infrastructure ;-)
[22-Sep-2009 12:23:35] <rmatte> yeh, we have all of our servers on high performance HP blades running ESX
[22-Sep-2009 12:23:58] <rmatte> 15GHz of processing with 32GB of RAM on each blade
[22-Sep-2009 12:24:00] <rmatte> we have 3 of them
[22-Sep-2009 12:26:36] <Troubadix09> rmatte: ok, on me against you its medium-performance ... lol
[22-Sep-2009 12:27:19] <rmatte> hehe
[22-Sep-2009 12:28:00] <Troubadix09> rmatte: we use only 3 hp proliant on a HP EVA4400 for virtualisation, but it seems to be enough for us
[22-Sep-2009 12:28:27] <rmatte> yeh, depends on what you want to run
[22-Sep-2009 12:28:38] <rmatte> we'll eventually have tons of Zenoss instances running off of these
[22-Sep-2009 12:29:31] <rmatte> we have about 30 vms running across all 3 servers right now
[22-Sep-2009 12:29:47] <rmatte> one of the servers only has 1 vm on it though, so the other 2 have about 15 each
[22-Sep-2009 12:30:00] <rmatte> and we're just getting started
[22-Sep-2009 12:30:01] <rmatte>
\n[22-Sep-2009 12:31:53] <rmatte> I wish Chet were around, I have\nsome questions for him about the new interface status column
[22-Sep-2009 12:32:27] <Troubadix09> rmatte: ok, you win ;-)
[22-Sep-2009 12:33:53] <Troubadix09> oh, you test the 2.5 beta?
[22-Sep-2009 12:34:01] <rmatte> yeh
[22-Sep-2009 12:34:18] <rmatte> hopefully beta 3 is a lot less buggy than beta 2
[22-Sep-2009 12:34:30] <rmatte> especially since the final release for King Crab is supposed to be around October 25th
[22-Sep-2009 12:35:23] <Troubadix09> sorry, there is not enough personal at our IT, so i have no time to test it :-(
[22-Sep-2009 12:36:04] <rmatte> I'm just testing it because I want to make sure that it's not going to be a total disaster upgrading to it
[22-Sep-2009 12:36:12] <rmatte> iancmcc: ping
[22-Sep-2009 12:42:05] <Troubadix09> uuii... worktime is long gone, i go home now, cu later
[22-Sep-2009 12:42:20] <rmatte> later
[22-Sep-2009 13:26:35] <djack> hellow world belgium calling
[22-Sep-2009 13:29:45] <rmatte> hey
[22-Sep-2009 13:37:28] <rmatte> man, that hardware model info bug is actually most likely simpler to fix than it seems
\n[22-Sep-2009 13:37:48] <rmatte> it doesn't appear to actually be\na bug with setting the info itself, since if you select stuff and click\nsave it'll apply it to the device
[22-Sep-2009 13:38:04] <rmatte> it seems to be an issue with the dropdown menus on the Edit tab actually pulling up the previous settings
\n[22-Sep-2009 13:38:28] <rmatte> so you can use the edit tab to\nset the model info once, but if you use the edit tab for anything else\nyou have to specifically select the model info again before you click\nsave
[22-Sep-2009 13:38:32] <rmatte> otherwise it'll clear it.
[22-Sep-2009 13:42:44] <rmatte> there
[22-Sep-2009 13:42:45] <rmatte> http://dev.zenoss.org/trac/ticket/5473#comment:9
[22-Sep-2009 13:43:02] <rmatte> hopefully they patch that soon
[22-Sep-2009 13:48:23] <Brad_K> djack: hey!
[22-Sep-2009 13:51:32] <djack> Brad_K: hey there
[22-Sep-2009 13:52:37] <djack> is there a eventviewer portlet?
\n[22-Sep-2009 13:53:40] <Brad_K> i was wondering if anyone had\nexperience editing the XML that is generated by TestGen4Web, so that I\ncan make my script deal with "alert" sheets that Firefox throws up that\nTestGen4Web is not capturing as part of the recording?
[22-Sep-2009 13:54:42] ideopathic_ is now known as ideopathic
[22-Sep-2009 13:58:38] <djack> huh?
[22-Sep-2009 14:12:36] <rmatte> djack: no, there isn't an event viewer portlet
[22-Sep-2009 14:12:49] <mrayzenoss> djack: docs/DOC-3457
[22-Sep-2009 14:13:03] <rmatte> aha
[22-Sep-2009 14:13:11] <rmatte> but apparently Matt's got the goods
[22-Sep-2009 14:13:24] <djack> that i installed today ;-)
\n[22-Sep-2009 14:13:45] <rmatte> I think he's talking about\nsomething more along the lines of something that mimics the event\nconsole
[22-Sep-2009 14:13:56] <rmatte> although that would look so condensed I don't know how useful it'd be
[22-Sep-2009 14:14:08] <mrayzenoss> there are Event Views
[22-Sep-2009 14:14:14] <djack> i only wonder how the group events
[22-Sep-2009 14:14:35] <djack> but not the thing i look for
\n[22-Sep-2009 14:17:13] <Brad_K> In addition to the TestGen4Web\nquestion, does anyone know the default units of measurement for the\n"size" and "time" values in the HttpMonitor? Are those KB and\nmilliseconds?
[22-Sep-2009 14:19:08] <rmatte> Brad_K: correct
\n[22-Sep-2009 14:20:20] <Brad_K> weird, because\ndocs/DOC-3234 says "seconds" and "bytes",\nbut the numbers I'm getting just don't make any sense at all with those\nunits....
[22-Sep-2009 14:21:40] <Brad_K> hmm. in fact, KB doesn't make sense, either. ;-(
[22-Sep-2009 14:22:37] <rmatte> how so?
[22-Sep-2009 14:22:40] <rmatte> makes sense for mine
\n[22-Sep-2009 14:23:25] <Brad_K> the page in question isn't\nanywhere close to 244KB in size, and the value of "size" has never\nvaried from that number. But the page is definitely larger than 244\nbytes.
[22-Sep-2009 14:23:57] <Brad_K> 16KB is more like it.
[22-Sep-2009 14:24:01]\n<rmatte> for time on one of my graphs I'm seeing cur as being\n95.11, avg as 191.49m and max as 4.40, meaning the longest it took was\n4.40 seconds, the average is 191.49 milliseconds, and the current is\n95.11 milliseconds
[22-Sep-2009 14:24:06] <rmatte> which makes perfect sense
[22-Sep-2009 14:24:52] <rmatte> size is showing as about 4 kilobytes for all of them, which sounds about right as well
[22-Sep-2009 14:26:57] <rmatte> sorry, cur for time is 95.11m not 95.11
[22-Sep-2009 14:28:56] <rmatte> afk for a few
\n[22-Sep-2009 14:31:25] <Brad_K> i was wondering if anyone had\nexperience editing the XML that is generated by TestGen4Web, so that I\ncan make my script deal with "alert" sheets that Firefox throws up that\nTestGen4Web is not capturing as part of the recording?
[22-Sep-2009 14:32:42] mrayzenoss1 is now known as mrayzenoss
[22-Sep-2009 14:45:44] <djack> you should do an event in belgium :-p
[22-Sep-2009 14:46:43] <mrayzenoss> we've talked about attending FOSDEM
[22-Sep-2009 15:05:00] <rmatte> well that was fun
[22-Sep-2009 15:05:08] <rmatte> lost our internet connection for like 25 mins
[22-Sep-2009 15:20:01] <rhettardo> did the isp blame you for the outage like they always ddo, or deny it like they always do?
[22-Sep-2009 15:20:37] <rhettardo> 'its your router', 'we dont show any issues'
[22-Sep-2009 15:21:08] <rmatte> nah, it's a circuit going from here to the datacenter that we host our equipment at
[22-Sep-2009 15:21:19] <rmatte> and one of the guys on our team's roommate works at the datacenter
[22-Sep-2009 15:21:30] <rmatte> so he just called him up and they told us that one of their switches had died
[22-Sep-2009 15:22:47] <rhettardo> ah
\n[22-Sep-2009 15:25:34] <ke4qqq> hey guys - I have a two cisco\nASAs (a 5505 and 5510) pings to the devices work, but I am getting snmp\nerrors. I tried using snmpwalk from the zenoss box to each ASA and I\nget responses - any idea what could be causing the problem?
[22-Sep-2009 15:26:17] <rmatte> snmp errors in Zenoss?
[22-Sep-2009 15:26:45] <rmatte> like, an snmpwalk works fine but Zenoss is complaining?
[22-Sep-2009 15:26:52] <ke4qqq> rmatte: yes - zenoss reporting SNMP agent down for just those devices
[22-Sep-2009 15:26:53] <ke4qqq> yes
[22-Sep-2009 15:27:01] <rmatte> great, then what you want to do is this...
[22-Sep-2009 15:27:01] <ke4qqq> snmpwalk works fine, but zenoss complains
\n[22-Sep-2009 15:27:29] <rmatte> go to the zProperties for those\ndevices (don't do this for every device in Zenoss), and change\nzMaxOIDPerRequest from 40 to 20 or 10
[22-Sep-2009 15:27:46] <rmatte> Cisco ASAs seems to be picky about the size of packets being sent to them sometimes
[22-Sep-2009 15:28:03] <rmatte> so you need to lower the amount of data that it's trying to poll per packet
[22-Sep-2009 15:28:33] <ke4qqq> awesome - I'll create an article on the wiki to document this
[22-Sep-2009 15:28:35] <ke4qqq> thanks!
[22-Sep-2009 15:28:43] <rmatte> np, make sure it works first
[22-Sep-2009 15:28:51] <rmatte> it worked for me when I was having issues in the past, so it should
[22-Sep-2009 15:28:56] <rmatte> but that may not be your exact issue
[22-Sep-2009 15:29:02] <rmatte> I'm almost certain that it is though
[22-Sep-2009 15:29:04] <ke4qqq> ok
[22-Sep-2009 15:32:23] <chemist> good evenng
[22-Sep-2009 15:32:34] <chemist> *evening even
[22-Sep-2009 15:32:43] <rmatte> hey
[22-Sep-2009 15:32:59] <ke4qqq> rmatte that fixed it - thanks!
[22-Sep-2009 15:32:59] <chemist> we are trying to model some netgear switches and we keep getting this error
\n[22-Sep-2009 15:33:10] <chemist> Unable to get data for XXXX from\neither ipAddrTable or ipNetToMediaTable -- skipping model
[22-Sep-2009 15:33:28] <rmatte> ke4qqq: np
[22-Sep-2009 15:33:29] <chemist> any ideas?
[22-Sep-2009 15:33:52] <rmatte> chemist: you sure your snmp permissions are correct on the devices?
[22-Sep-2009 15:33:57] <chemist> yes
[22-Sep-2009 15:33:58] <rmatte> sounds like they aren't
[22-Sep-2009 15:34:16] <chemist> positive
[22-Sep-2009 15:34:16] <rmatte> try doing: snmpwalk -v1 -c bleh bleh.com ipAddrTable
[22-Sep-2009 15:34:20] <rmatte> see if it returns anything
[22-Sep-2009 15:35:15] <rmatte> should be some lines like:
[22-Sep-2009 15:35:16] <rmatte> IP-MIB::ipAdEntAddr.30.30.1.25 = IpAddress: 30.30.1.25
[22-Sep-2009 15:35:16] <rmatte> IP-MIB::ipAdEntAddr.30.30.2.25 = IpAddress: 30.30.2.25
[22-Sep-2009 15:35:16] <rmatte> IP-MIB::ipAdEntAddr.30.30.3.25 = IpAddress: 30.30.3.25
\n[22-Sep-2009 15:35:35] <rmatte> If you don't see anything like\nthat returned then your permissions aren't setup correctly or you're\nusing the wrong community string in Zenoss
[22-Sep-2009 15:43:25] <chemist> rmatte: No, checked, rechecked and triple checked
[22-Sep-2009 15:43:44] <rmatte> chemist: so if you snmpwalk like I said you get output returned?>
[22-Sep-2009 15:43:46] <chemist> permissions are fine and the community string is ok
[22-Sep-2009 15:44:11] <chemist> if I snmpwalk without ipAddrTable, yes
[22-Sep-2009 15:44:40] <rmatte> no, I mean with ipAddrTable
[22-Sep-2009 15:44:58] <chemist> with ipAddrTable I get nothing
[22-Sep-2009 15:45:04] <rmatte> that's what it's complaining about, that it's not getting any output returned from walking that value
[22-Sep-2009 15:45:20] <chemist> but with any other value it works
\n[22-Sep-2009 15:45:34] <rmatte> the only time I've seen\ncomplaints about those values was on linux boxes where the snmp\npermissions weren't properly configured
[22-Sep-2009 15:45:47] <rmatte> what device group do you have the devices in?
[22-Sep-2009 15:46:13] <chemist> '/network/switch'
[22-Sep-2009 15:47:08] <rmatte> If you go to /Network/Switch then select More -> Collector Plugins
\n[22-Sep-2009 15:47:30] <rmatte> do you only see\nzenoss.snmp.NewDeviceMap, zenoss.snmp.DeviceMap, and\nzenoss.snmp.InterfaceMap?
[22-Sep-2009 15:48:27] <chemist> and RouteMap
[22-Sep-2009 15:49:12] <rmatte> try removing routemap from the list
[22-Sep-2009 15:49:15] <rmatte> then try again
[22-Sep-2009 15:50:26] <rmatte> ah
[22-Sep-2009 15:50:26] <rmatte> http://74.125.113.132/search?q=cache:DQ2LqdM5h6sJ:forums.zenoss.com/viewtopic.php%3Ft%3D10007%26sid%3D855c0e0a58d7c8a363483cd6ecbf0041+zenoss+Unable+to+get+data+for+ipAddrTable&cd=1&hl=en&ct=clnk&gl=ca
[22-Sep-2009 15:50:29] <rmatte> that is basically what you're experiencing
[22-Sep-2009 15:50:32] <rmatte> and at the end...
\n[22-Sep-2009 15:50:40] <rmatte> So the ipAddrTable or\nipNetToMediaTable are some mibs that the switch is missing. The other\nworking switch was running a new version of code, which I've downgraded\nand I'm still able to pull interface stats fine.
\n[22-Sep-2009 15:50:41] <rmatte> So now I have a new question, how\ndo I get the zen modeler to skip the ipAddrTable or ipNetToMediaTable\nwhen building the OS interfaces?
[22-Sep-2009 15:50:59] <rmatte> now I just need to figure out how to get it to ignore those values
[22-Sep-2009 15:51:21] <rmatte> although the ipAddrTable values are fairly important
[22-Sep-2009 15:54:18] <chemist> I guess I'll tell my boss we need to replace 40+ switches
[22-Sep-2009 15:54:39] <rmatte> ah, it's because it's a layer 2 switch
\n[22-Sep-2009 15:54:45] <rmatte>\nthread/10554%3Bjsessionid=1787F3F07DB545A3108BB042D9916094.node0?decorator=print&displayFullThread=true
[22-Sep-2009 15:56:48] <chemist> yeah, we are chepscapes
[22-Sep-2009 15:56:55] <chemist> *cheapscapes
\n[22-Sep-2009 15:56:57] <rmatte> From the sounds of it you may\nneed to edit\n$ZENHOME/Products/DataCollector/plugins/zenoss/snmp/InterfaceMap.py to\nhave it skip the lines that collect that data
[22-Sep-2009 15:57:09] <rmatte> I can't remember the last time I've seen a layer 2 switch
[22-Sep-2009 15:57:42] <chemist> I'll show this conversation to my boss
[22-Sep-2009 15:57:57] <rmatte> lol
[22-Sep-2009 15:58:14] <rmatte> If you have the money it would really be worth grabbing some Cisco or HP switches
[22-Sep-2009 15:58:22] <rmatte> bring yourselves in to the 21st century
[22-Sep-2009 15:58:25] <rmatte>
\n[22-Sep-2009 15:59:23] <ke4qqq> rmatte: layer 2 switchs are still\ncommon - cisco 2970s for instance. moving to 3750s for every switch\nwould be pretty heady for most
[22-Sep-2009 15:59:59] <rmatte> I suppose
[22-Sep-2009 16:00:13] <rmatte> actually, 2970s might not even have this problem
[22-Sep-2009 16:00:17] <ke4qqq> but L2 switches are including more and more L3 features
[22-Sep-2009 16:00:38] <ke4qqq> but since they don't really route it's not an issue
[22-Sep-2009 16:01:01] <ke4qqq> I have some old 2948Gs that don't exihibit that problem (along with 3coms and HPs.)
[22-Sep-2009 16:01:26] <rmatte> hmmm, no 2970s, only 297024
[22-Sep-2009 16:01:35] mrayzenoss1 is now known as mrayzenoss
[22-Sep-2009 16:02:18] <ke4qqq> yeah most of my switches are relatively ancient.....
[22-Sep-2009 16:02:38] <rmatte> hehe
[22-Sep-2009 16:02:59] <rmatte> chemist: you should log a Trac case for that
[22-Sep-2009 16:03:20] <rmatte> It is pretty dumb that it can't do it
[22-Sep-2009 16:04:56] <chemist_> or use cacti plugins
[22-Sep-2009 16:05:21] <chemist_> cacti reports them properly
[22-Sep-2009 16:05:56] <rmatte> yeh, but the whole point of a trac case is to eventually get that fixed
[22-Sep-2009 16:06:05] <rmatte> instead of just saying "meh, I'll use something else"
[22-Sep-2009 16:06:08] <rmatte>
[22-Sep-2009 16:06:17] <rmatte> up to you
[22-Sep-2009 16:06:31] <chemist_> good point
[22-Sep-2009 16:06:57] <rmatte> If you don't have a trac account ask Matt to make you one
[22-Sep-2009 16:06:58] <chemist_> and I would rather use zenoss for the whole enterprise
[22-Sep-2009 16:07:11] <rmatte> http://dev.zenoss.org/trac
[22-Sep-2009 16:07:15] <chemist_> mrayzenoss?
[22-Sep-2009 16:07:23] <pokui> chemist_: sorry I missed part of this discussion, what make/model are these switches again?
[22-Sep-2009 16:07:40] <chemist_> netgear FS726T
[22-Sep-2009 16:07:47] <mrayzenoss> yeah, I can make Trac accounts, or you can just use zenoss/zenoss
[22-Sep-2009 16:07:56] <pokui> chemist_: ah. thanks.
[22-Sep-2009 16:07:59] <rmatte> ah, didn't know there was a generic one
[22-Sep-2009 16:08:31] * pokui makes a note to test with level one swtiches.
\n[22-Sep-2009 16:10:58] <rmatte> chemist_: so yeh, go to\nhttp://dev.zenoss.org/trac and login with zenoss/zenoss (or ask Matt to\nmake you your own account) and log the issue
[22-Sep-2009 16:11:11] <chemist_> will do
\n[22-Sep-2009 16:11:20] <rmatte> I just find that it's nice to\nhave your own account since you can better track the status of your\ntickets
[22-Sep-2009 16:11:33] <chemist_> I'm surprised this hasn't come up before though
[22-Sep-2009 16:11:51] <chemist_> unless no one else buys these cheap switches
\n[22-Sep-2009 16:12:17] <Brad_K> chemist_: it has come up before,\nthat's why we can provide these URLs that quickly describe the problem.
[22-Sep-2009 16:12:44] <rmatte> ke4qqq: ah, you help out with fedora?
\n[22-Sep-2009 16:12:52] <Brad_K> the issue is that no one (so far)\nhas apparently created a trac ticket for this problem. so far, they've\napparently always decided to live with the issue, or use cacti plugins,\nor whatever.
[22-Sep-2009 16:13:04] <rmatte> chemist_: I did a search and it actually came up on 3 different occasions in the old forums
[22-Sep-2009 16:13:07] <chemist_> ouch!!
\n[22-Sep-2009 16:13:25] <rmatte> but there was never really much\nof a solution other than modifying the\n$ZENHOME/Products/DataCollector/plugins/zenoss/snmp/InterfaceMap.py file
[22-Sep-2009 16:14:12] <chemist_> but that's not really a fix
[22-Sep-2009 16:14:24] <ke4qqq> rmatte: well I at least try not to hinder Fedora
[22-Sep-2009 16:14:37] <rmatte> hehe
[22-Sep-2009 16:14:40] <chemist_> hacks like that have a tendency to come back and bite you
[22-Sep-2009 16:14:49] <rmatte> chemist_: yup, they do indeed
[22-Sep-2009 16:14:56] <Brad_K> which is why someone needs to create a trac ticket. ;-)
[22-Sep-2009 16:14:59] <ke4qqq> yes, hence the need to create a ticket and get zenoss to fix the problem upstream
[22-Sep-2009 16:14:59] <rmatte> but yeh, log a ticket and I'm sure it'll get looked at
[22-Sep-2009 16:15:04] <rmatte> because that's pretty major
[22-Sep-2009 16:15:29] * chemist_ is reading how to create a trac ticket
[22-Sep-2009 16:15:42] <rmatte> it's just wiki formatting
[22-Sep-2009 16:15:56] <rmatte> click on new ticket in the top right
[22-Sep-2009 16:16:09] <rmatte> then populate the template it provides
[22-Sep-2009 16:16:16] <rmatte> so * Build Version: ALL
[22-Sep-2009 16:16:24] <rmatte> *Operating System(s): ALL
[22-Sep-2009 16:16:29] <rmatte> Steps to Reproduce...
[22-Sep-2009 16:16:31] <rmatte> 1. bleh
[22-Sep-2009 16:16:33] <rmatte> 2. bleh
[22-Sep-2009 16:16:37] <rmatte> etc...
[22-Sep-2009 16:16:42] <rmatte> Results
[22-Sep-2009 16:16:43] <rmatte> whatever
[22-Sep-2009 16:16:47] <rmatte> expected results
[22-Sep-2009 16:16:52] <rmatte> whatever was expected
[22-Sep-2009 16:16:54] <rmatte> Impact
[22-Sep-2009 16:16:58] <rmatte> how it impacts the users
[22-Sep-2009 16:17:16] <rmatte> then change the priority if you feel it's higher priority than the default of 3
[22-Sep-2009 16:17:21] <rmatte> and click Create Ticket
[22-Sep-2009 16:17:23] <rmatte> then you're done
\n[22-Sep-2009 16:17:40] <rmatte> there are other values that can\nbe changed as well such as version, but the default should suffice for\nthis
[22-Sep-2009 16:17:43] <Brad_K> rmatte:\ncareful, you might actually get tickets with steps like "1. bleh\n2.bleh" that are actually created. ;-)
[22-Sep-2009 16:18:05] <rmatte>
[22-Sep-2009 16:18:13] <rmatte> only if people are completely retarded
[22-Sep-2009 16:18:21] <chemist_> I might add some blah's for colour
[22-Sep-2009 16:18:35] <rmatte>
[22-Sep-2009 16:18:41] <Brad_K> is there irony in being a literalist contrarian?
[22-Sep-2009 16:18:50] <rmatte> I still need to open that trac case demanding MOAR PONIES! in King Crab
[22-Sep-2009 16:19:07] <rmatte> Brad_K: I'd say so
[22-Sep-2009 16:19:09] <Brad_K> or MO4R C0WB3LL!
[22-Sep-2009 16:19:18] <rmatte> w000000h!
[22-Sep-2009 16:20:25] * Brad_K prefers more c0ffee....
[22-Sep-2009 16:20:31] <rmatte> they really need to replace the "View Event History" link with "Travel to Candy Mountain"
[22-Sep-2009 16:20:34] <rmatte> sounds more enticing
[22-Sep-2009 16:21:05] <rmatte> or Beer Volcano
[22-Sep-2009 16:21:06] <rmatte> either way
[22-Sep-2009 16:21:28] * Brad_K contemplates Beer and Mentos....
[22-Sep-2009 16:21:34] <rmatte> hehe
[22-Sep-2009 16:21:37] <mrayzenoss> mmm... skittlebrau
[22-Sep-2009 16:22:08] <Brad_K> would you eat the Mentos first, then drink the beer? And then what would happen inside you?
[22-Sep-2009 16:22:20] <rmatte> you'd die
[22-Sep-2009 16:23:01]\n<Brad_K> or maybe you'd just have fountains of beer coming out\nyour nose? kind of like an updated C|N>K type of moment?
\n[22-Sep-2009 16:24:48] <rmatte> I just like the whole\npastafarianist belief that when you die you travel to a beer volcano\nnext to a stripper factory
[22-Sep-2009 16:25:04]\n<rmatte> and if you do evil things, you still travel to a beer\nvolcano next to a stripper factory, but the beer is flat!
[22-Sep-2009 16:25:39] * Brad_K is down with the FSM
[22-Sep-2009 16:25:45] <rmatte>
[22-Sep-2009 16:26:00] <rmatte> his noodliness
[22-Sep-2009 16:26:20] <Brad_K> However, my religion is closer to Jedi than Pastafarianism.
[22-Sep-2009 16:26:37] <Brad_K> btw, you did hear that Jedi is now an official religion in the UK, right?
[22-Sep-2009 16:26:44] <rmatte> yeh
[22-Sep-2009 16:26:55] <rmatte> pretty dumb though
\n[22-Sep-2009 16:27:50] <Brad_K> well, we don't necessarily\nbelieve in all the "new trilogy" material, so perhaps that would be\nJedi Fundamentalist?
[22-Sep-2009 16:28:05] <rmatte> lol
[22-Sep-2009 16:28:10] <Brad_K> or Traditional Jeditarian?
[22-Sep-2009 16:28:26] <rmatte> Jedihova's witnesses?
[22-Sep-2009 16:28:29] <Brad_K> no, no, no -- Orthodox Jeditarianism...
[22-Sep-2009 16:28:59] <Brad_K> at least we're not the Yoda's Witnesses....
[22-Sep-2009 16:29:03] <rmatte> lol
[22-Sep-2009 16:29:23] <rmatte> wouldn't Yoda be equivalent to Jesus?
[22-Sep-2009 16:29:27] <rmatte> in that religion
\n[22-Sep-2009 16:30:12] <rmatte> Once I start seeing people\ndriving around with Yoda Fish on their car, then I'll be amazed
\n[22-Sep-2009 16:30:17] * Brad_K ponders on the concept of Yoda being\nhis own son, and the kind of whacky time-travel that would be required\nto make that happen....
[22-Sep-2009 16:30:48] <rmatte> no need for time travel, just advanced virtual reality
[22-Sep-2009 16:30:51] <rmatte> holodeck
[22-Sep-2009 16:30:53] <rmatte>
[22-Sep-2009 16:31:09] <Brad_K> with Ponoko, you can make your own Jedifish or Yodafish, etc....
[22-Sep-2009 16:31:32] <rmatte> Ok, opinion...
[22-Sep-2009 16:31:37] <rmatte> http://i571.photobucket.com/albums/ss159/KalynSS999/Paintball%20-%20Guns/EgoParts2.jpg
[22-Sep-2009 16:31:38] <rmatte> http://i420.photobucket.com/albums/pp284/SADcasper21/DSCN0448.jpg
[22-Sep-2009 16:31:42] <rmatte> which color is nicer
[22-Sep-2009 16:31:55] <rmatte> the red one has a black tip (can't see it in the photo)
[22-Sep-2009 16:32:30] <Brad_K> for what purpose? i like them both....
[22-Sep-2009 16:32:47] <rmatte> no I'm literally buying that model of paintball gun lol
[22-Sep-2009 16:33:10] <rmatte> I'm sort of leaning towards the brown/bright blue one
\n[22-Sep-2009 16:33:28] <Brad_K> i figured, but you still have to\nanswer the question of purpose. would you be a member of a team where\nthe colors in your marker would make a difference?
[22-Sep-2009 16:33:37] <rmatte> oh, no
[22-Sep-2009 16:33:40] <rmatte> it's just for style
[22-Sep-2009 16:34:15] <Brad_K> is it brown and blue, or copper and blue? it's hard to tell from the picture....
[22-Sep-2009 16:34:24] <rmatte> it's copper and blue
[22-Sep-2009 16:34:29] <rmatte> quite dark in that photo
[22-Sep-2009 16:35:19] <rmatte> http://media.photobucket.com/image/ego%20lagoon/burressfan1/IMG_0340.jpg
\n[22-Sep-2009 16:35:26] <Brad_K> i like them both. the red &\nblack would be a bit more traditional pairing of colors, while the\ncopper and blue would be a more trendy pairing. but both are good\ncombinations.
[22-Sep-2009 16:35:47] <rmatte> yeh, hmmm
[22-Sep-2009 16:36:21] <rmatte> My mask which arrives tomorrow: http://my.tbaytel.net/dementia/images/lazarus.jpg
[22-Sep-2009 16:36:41] <rmatte> can't wait
[22-Sep-2009 16:36:42] <rmatte> hehe
[22-Sep-2009 16:36:51] <Brad_K> that's pretty scary. you should wear that on halloween.
[22-Sep-2009 16:36:55] <rmatte> hehe
[22-Sep-2009 16:37:15] <rmatte> it's hand airbrushed by a company called save phace
[22-Sep-2009 16:37:21] <rmatte> they make some pretty neat looking masks
\n[22-Sep-2009 16:37:24] <chemist_> ok, I created the trac ticket\nand in the meantime my colleague hacked the InterfaceMap.py file so we\ncan model the switches
[22-Sep-2009 16:37:37] <rmatte> chemist_: good stuff
[22-Sep-2009 16:37:50] <chemist_> moving on to the next problem....
[22-Sep-2009 16:37:51] <Brad_K> chemist_: thanks!
[22-Sep-2009 16:38:13] <chemist_> zenoss has decided to stop sending emails
[22-Sep-2009 16:38:28] <chemist_> I can send a test email but I don't get any notifications
[22-Sep-2009 16:39:10] <rmatte> have you tried restarting zenoss?
[22-Sep-2009 16:39:44] <chemist_> yes
[22-Sep-2009 16:40:45] <rmatte> hmmm
[22-Sep-2009 16:42:15] <rmatte> not sure, if you check the mail logs you don't see anything?
\n[22-Sep-2009 16:54:41] <chemist_> rmatte: tbh I don't know how\nzenoss sends email, wether it uses sendmail or it's own internal\nmechanics
[22-Sep-2009 16:55:09] <rmatte> it's own internals
[22-Sep-2009 16:55:31] <chemist_> that's why I can't find anything in the logs
[22-Sep-2009 16:56:24] <rmatte> actually...
[22-Sep-2009 16:56:35] <rmatte> if you go to settings, what smtp host do you have set?
[22-Sep-2009 16:56:41] <rmatte> if you can get in to that host and check those logs
[22-Sep-2009 16:57:23] <chemist_> it's an exchange server, and I can receive email when I send a test
[22-Sep-2009 16:57:49] <rmatte> i see
[22-Sep-2009 16:58:09] <rmatte> and the alerts are setup under your account?
[22-Sep-2009 16:58:26] <chemist_> yyes
[22-Sep-2009 16:58:37] <chemist_> it did literally stop working
[22-Sep-2009 16:58:45] <rmatte> you're sure you don't have some weird filter option in there?
[22-Sep-2009 16:59:28] <chemist_> all I did was create a rule for a different account, then deleted the rule
[22-Sep-2009 16:59:36] <chemist_> since then, no emails
[22-Sep-2009 16:59:53] <rmatte> weird, I'm not sure
[22-Sep-2009 17:00:02] <rmatte> check all of the log files
[22-Sep-2009 17:00:06] <chemist_> neway, time to go home
[22-Sep-2009 17:00:17] <chemist_> thanks for the help
[22-Sep-2009 17:00:41] <rmatte> np
\n[22-Sep-2009 17:42:25] <wildcard0> hey. my disk stats are\ngetting weird replies like negative free space. i remodeled and it\nhasn't changed. can anyone point me in the right direction? i see\nvery little on the board or google
[22-Sep-2009 17:43:15] <mrayzenoss> wildcard0: really large disks?
[22-Sep-2009 17:44:05]\n<wildcard0> no, i thought it may be wrapping, but this one's a\n/boot partition and it's 100M total. zenoss reports it as 12G though
[22-Sep-2009 17:44:36] <mrayzenoss> hmm.. the bug I was thinking of is for 12TB+
[22-Sep-2009 17:44:44] <mrayzenoss> SNMP or SSH?
[22-Sep-2009 17:44:47] <mrayzenoss> or WMI?
[22-Sep-2009 17:44:50] <wildcard0> snmp
[22-Sep-2009 17:44:59] <wildcard0> it's odd that it's reporting 12GB though
[22-Sep-2009 17:45:05] <wildcard0> if there's a bug at that point
[22-Sep-2009 17:45:27] <mrayzenoss> the bug is at 12 terabytes
[22-Sep-2009 17:45:35] <wildcard0> oh TB
[22-Sep-2009 17:45:37] <wildcard0> heh i can read
[22-Sep-2009 17:45:56] <wildcard0> i see no errors in the logs at all either
\n[22-Sep-2009 17:46:23] <wildcard0> though i dont see the log that\nreturns snmp info specifically. so either i dont have logging up high\nenough or that isn't logged
[22-Sep-2009 17:46:37] <mrayzenoss> message/28849 ?
\n[22-Sep-2009 17:47:46] <wildcard0> ya i saw that. snmp looks\nlike it's giving out sane values and there's no real resolution on that\nthread
[22-Sep-2009 17:47:47] <mrayzenoss> I gotta run, but searching for "negative" pulled up a few things
[22-Sep-2009 17:48:07] <wildcard0> thanks
[22-Sep-2009 17:48:13] <mrayzenoss> maybe post to that thread? or follow those debug tips a bit and see what turns up?
[22-Sep-2009 17:49:05] <wildcard0> ya i'll keep hammering at it. im sure it'll be something dumb
[22-Sep-2009 17:49:12] <mrayzenoss> later
[22-Sep-2009 18:32:43] <wildcard0> hey. how do i restart just zenperfsnmp without restarting all of zenoss?
[22-Sep-2009 18:49:35] <ocnam> think its in settings > daemons
[22-Sep-2009 18:50:00] <wildcard0> thanks
[22-Sep-2009 21:31:03] straterr1 is now known as straterra
[23-Sep-2009 00:00:46] [disconnected at Wed Sep 23 00:00:46 2009]
[23-Sep-2009 00:00:46] [connected at Wed Sep 23 00:00:46 2009]
[23-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[23-Sep-2009 02:01:17] <Troubadix09> hi all
[23-Sep-2009 08:10:03] <Mituc> hello everyone
[23-Sep-2009 08:12:02]\n<Mituc> I just upgraded from 2.4.2 to 2.4.5 + applied the patchez\nfor zenocommand (that will be included in 2.4.6), and after I did that\nzencommand did not execute any other plug-ins, but gave an error instead
[23-Sep-2009 08:12:20] <Mituc> anybody else experiencing the same issues?
[23-Sep-2009 08:12:36] <Mituc> I had to downgrade back to 2.4.2, since the upgrade to 2.4.5 + patching did not work..
[23-Sep-2009 08:13:05]\n<Mituc> and 2.4.2 gives a lot of command timeouts, events remain\nstuck in an open state without the state being reset, and so on
[23-Sep-2009 08:13:39] <Mituc> wonder if the database was affected by the upgrade+downgrade..
\n[23-Sep-2009 08:18:51] <Mituc> also, if I take a copy of\nvar/Data.* and perf/ directories/files and restore it with a fresh copy\nof zenoss and events database, will that work ?
[23-Sep-2009 08:22:25] <Mituc> ok, so I'll just try that then
[23-Sep-2009 09:18:06] <Egyptian[web]> hi
\n[23-Sep-2009 09:18:27] <rmatte> anyone who can tell me why my\nVista install is eating up 18.2GB of space when I only have 9.79GB of\nfiles on the drive wins a cookie
[23-Sep-2009\n09:18:39] <Egyptian[web]> i just started with zenoss and did an\nautodiscover .. but now i want to change the device's "name" from ip to\nhostname... hw?
[23-Sep-2009 09:19:15] <Egyptian[web]> rmatte: have you tried #windows?
\n[23-Sep-2009 09:19:52] <rmatte> Egyptian, go to the device page,\nclick on the dropdown arrow in the top left, select Manage -> Rename\nDevice
[23-Sep-2009 09:19:58] <rmatte> and yes, thank you, #windows is quite obvious
\n[23-Sep-2009 09:20:39] <rmatte> I've googled to high heaven for a\nsolution, I doubt that #windows is going to serve any better
[23-Sep-2009 09:20:43] <rmatte>
[23-Sep-2009 09:22:30] <rmatte> People in here tend to know little tricks for this sort of thing from time to time
\n[23-Sep-2009 09:23:20] <Egyptian[web]> rmatte: then .. maybe you\nshould do what i do .. load a live linux cd and explore the drive ?
[23-Sep-2009 09:25:48] <Egyptian[web]> and thanks .. i renamed the device
\n[23-Sep-2009 09:26:49] <rmatte> Egyptian[web]: it's a VM running\nin VirtualBox on top of linux already lol, I suppose I could download a\nlivecd ISO and mount it
[23-Sep-2009 09:27:18] <Egyptian[web]> rmatte: indexing service ?
[23-Sep-2009 09:27:37] <Egyptian[web]> rmatte: system backup i think its called .. you know .. autorestore thingy
[23-Sep-2009 09:27:41] <rmatte> I've checked indexing, shadow files, etc... (all the common stuff)
[23-Sep-2009 09:27:49] <rmatte> that's why I'm stumped
\n[23-Sep-2009 09:28:15] <rmatte> system restore wouldn't use that\nmuch space, it basically just takes a snapshot like you would in vmware
[23-Sep-2009 09:28:43] * Egyptian[web] shrugs
[23-Sep-2009 09:29:14] <Egyptian[web]> live cd and find files bigger than 1 gb
\n[23-Sep-2009 09:29:17] <rmatte> meh, I hate vista anyways, if it\nstarts becoming a big deal I'll just reformat and throw XP on
\n[23-Sep-2009 09:29:43] <Egyptian[web]> hokay . time to go home ..\ni will be dropping in quite a bit now that i have zenoss running
[23-Sep-2009 09:29:58] <Egyptian[web]> salam
[23-Sep-2009 09:54:12] danuvius_ is now known as danuvius
[23-Sep-2009 10:08:08] <twm1010> where are filesystem templates kept?
[23-Sep-2009 10:37:07] <rmatte> twm1010: /Devices/Server
[23-Sep-2009 10:37:09] <rmatte> templates
[23-Sep-2009 10:37:13] <rmatte> FileSystem
[23-Sep-2009 10:37:33] <rmatte> that's the single template used for everything
[23-Sep-2009 10:37:39] <rmatte> you can make local copies of it as necessary
[23-Sep-2009 10:37:49] <rmatte> for adjusting thresholds or whatever
[23-Sep-2009 11:51:23] <twm1010> rmatte: and that has special inheritance like the interface templates?
[23-Sep-2009 11:51:30] <rmatte> no
[23-Sep-2009 11:51:35] <rmatte> general inheritance
[23-Sep-2009 11:51:46] <rmatte> the same FileSystem template is used for everything
[23-Sep-2009 11:53:52] <twm1010> but i can make copies in classes right?
[23-Sep-2009 11:59:03] <rmatte> yeh, if you want to
[23-Sep-2009 11:59:10] <rmatte> just make a local copy of it wherever
\n[23-Sep-2009 11:59:27] <rmatte> you can make local copies of it\nin classes, or make copies of it directly on hard drive objects
[23-Sep-2009 11:59:28] <rmatte> up to you
[23-Sep-2009 12:05:06] <twm1010> Just find it odd that some of my low disk space events are errors and some are warnings
[23-Sep-2009 12:12:58] <rmatte> twm1010: you might have some local copies
[23-Sep-2009 12:13:13] <rmatte> Go to /Devices/Server and do...
[23-Sep-2009 12:13:28] <rmatte> More -> All Templates
[23-Sep-2009 12:13:36] <rmatte> it'll list all of the templates under /Devices/Server
[23-Sep-2009 12:13:39] <rmatte> including local
[23-Sep-2009 12:13:54] <rmatte> also, your threshold might be set to auto-escalate events
[23-Sep-2009 12:13:58] <rmatte> another thing to check
[23-Sep-2009 12:14:16] <rmatte> oh, and if you have transforms applied for filesystem, they may be affecting it as well
\n[23-Sep-2009 12:26:10] <twm1010> if zensyslog receives message\nfrom an IP not in Zenoss as a device, what does it do with them?
[23-Sep-2009 12:32:48] <rmatte> it displays them in the event console along with that IP anyways
\n[23-Sep-2009 12:33:15] <rmatte> but if you have a filter on your\nalerting rule to only alert on devices that are in Production state\nthen it won't alert, since it doesn't consider the device to be in\nproduction state
[23-Sep-2009 12:34:40] <twm1010> Have a good link on setting up the syslog client ?
[23-Sep-2009 12:35:15] <rmatte> what do you mean? setting up syslog-ng between zensyslog and public?
[23-Sep-2009 12:35:30] <twm1010> nah, configuring syslog on a linux host
[23-Sep-2009 12:35:46] <twm1010> i did a basic config: *.* @myzenossIP
[23-Sep-2009 12:36:14] <ckrough> twm1010: docs/DOC-2499
[23-Sep-2009 12:36:15] <twm1010> but let's say i want it to only send warnings/errors/critical
[23-Sep-2009 12:37:13] <ckrough> I'd recommend doing your filtering in zenoss, not syslog
[23-Sep-2009 12:37:53] <twm1010> they recommend in the doc *.debug
[23-Sep-2009 12:41:06] <rmatte> twm1010: we just use straight traps for our stuff, no syslog
[23-Sep-2009 12:41:24] <rmatte> not for our internal stuff anyways
[23-Sep-2009 12:41:33] <rmatte> we do use a bit of syslogging for some of our client's devices
[23-Sep-2009 12:41:41] <twm1010> yeah, i'm having trouble deciding which one to do here
[23-Sep-2009 12:41:44] <rmatte> nothing linux based though
[23-Sep-2009 12:41:58] <rmatte> I prefer the traps, since it'll only trap for the more critical events
[23-Sep-2009 12:42:17] <rmatte> but it's really up to how noisy you want to be and how much tuning you want to do
[23-Sep-2009 12:42:26] <twm1010> well, i don't want to do both traps and syslog
[23-Sep-2009 12:42:33] <twm1010> rather just end up picking one or the other
\n[23-Sep-2009 12:42:38] <rmatte> the syslogs will tell you tons of\nstuff, just not sure it'll really be anything that traps can't tell you\nwhich you'd actually be interested in
[23-Sep-2009 12:42:57] <twm1010> i was thinking for servers, syslog, for devices/routers/etc, traps
[23-Sep-2009 12:43:22] <twm1010> my windows boxes are ridiculously chatty
[23-Sep-2009 12:43:38] <twm1010> can you adjust the WMI event collection to ignore informational stuff?
[23-Sep-2009 12:43:52] <rmatte> you can adjust the level, yeh
[23-Sep-2009 12:44:13] <twm1010> ah zWinEventlogMinSeverity
[23-Sep-2009 12:44:17] <rmatte> zWinEventlogMinSeverity
[23-Sep-2009 12:44:19] <rmatte> yeh
[23-Sep-2009 12:44:22] <cgibbons> don't use 5!
[23-Sep-2009 12:44:22] <rmatte> you beat me to it
[23-Sep-2009 12:44:33] <twm1010> information is lvl 2?
[23-Sep-2009 12:44:37] <rmatte> 5 is lowest
[23-Sep-2009 12:44:47] <rmatte> nah, 2 would be more like warning
[23-Sep-2009 12:44:50] <rmatte> and 1 is critical
[23-Sep-2009 12:45:08] <rmatte> you'll probably have to do some tuning with event mappings regardless of what you choose
[23-Sep-2009 12:45:29] <twm1010> well sure, i have about 100 i have to classify right now
[23-Sep-2009 12:45:39] <rmatte> hehe
[23-Sep-2009 12:45:44] <twm1010> damn terminal servers and SQL servers seem to have unique messages for EVERYTHING
[23-Sep-2009 12:45:50] <rmatte> you know the proper way to tune syslog messages?
[23-Sep-2009 12:46:08]\n<cgibbons> if you are curious about the actual levels, look for\n'EventType' on this page:\nhttp://msdn.microsoft.com/en-us/library/aa394226(VS.85).aspx
[23-Sep-2009 12:46:08] <twm1010> well, its already set to 2
[23-Sep-2009 12:47:01] <twm1010> but i guess there are a lot of things microsoft calls an "error" that are really just information
[23-Sep-2009 12:47:09]\n<rmatte> Create mappings manually for each one under some events\nclass. You can even create an event class (for instance I made one\ncalled /Suppressed), then just name them whatever, set the\neventClassKey to "defaultmapping" and set the regex value to a common\npart of the text in the messages
[23-Sep-2009 12:47:17] <twm1010> in other words, i should quit being lazy and just get these mappings reated
[23-Sep-2009 12:47:22] <rmatte> then go to zProperties and set action to drop or history (depending on what you want)
[23-Sep-2009 12:47:25] <rmatte> and voila
[23-Sep-2009 12:47:45] <twm1010> Yeah I have a /Drop and a /History
[23-Sep-2009 12:47:51] <rmatte> ok
[23-Sep-2009 12:48:09] <rmatte> so yeh, create event mapping by hand, name it whatever, make sure you set eventClassKey to defaultmapping
[23-Sep-2009 12:48:18] <rmatte> then set regex to part of the text in the syslog message
[23-Sep-2009 12:48:32] <rmatte> well, works for any events really
[23-Sep-2009 12:48:36] <rmatte> WMI, Syslog, whatever
[23-Sep-2009 12:48:47] <rmatte> they have to come in with class /Unknown for it to work
[23-Sep-2009 12:49:03] <rmatte> If they don't, you'll have to do the same but put the mapping in to the specific event class
[23-Sep-2009 12:49:28] <twm1010> Ah, I see quite a few in here already
[23-Sep-2009 12:49:33] <twm1010> under /Ignore
[23-Sep-2009 12:49:35] <rmatte> probably
[23-Sep-2009 12:50:58] <rmatte> ah, that's what rule is for
[23-Sep-2009 12:51:09] <rmatte> I've never used anything beyond transform and regex
\n[23-Sep-2009 12:51:41] <rmatte> but yeh, you can put one of the\nsyslog messages in to "Example" when you're entering your regex
[23-Sep-2009 12:51:53] <rmatte> the regex text will become red if it doesn't match properly
[23-Sep-2009 12:53:22] <twm1010> hrmm... i'll have to play with that some more
[23-Sep-2009 12:56:09] <chemist> can anyone tell me what zenoss daemon sends email notifications?
[23-Sep-2009 12:57:52] <ckrough> zenactions I believe
[23-Sep-2009 12:58:04] <chemist> thnx
[23-Sep-2009 13:00:45]\n<rmatte> eugh, frigging UPS, I order something from the U.S that\ncosts $99.99 and they charge me $51.93 in brokerage/taxes
[23-Sep-2009 13:00:49] <rmatte> extortion
[23-Sep-2009 13:00:56] <twm1010> LOL
[23-Sep-2009 13:01:12] <rmatte> last time I use them for anything from the U.S, unreal
[23-Sep-2009 13:01:21] <twm1010> where are you?
[23-Sep-2009 13:01:24] <rmatte> Canada
[23-Sep-2009 13:01:37] <twm1010> What part?
[23-Sep-2009 13:01:41] <rmatte> I'll use USPS next time, slower but much much cheaper
[23-Sep-2009 13:02:06] <ckrough> yeah but $51 US dollars is like Canadian 2 cent now right?
[23-Sep-2009 13:02:16] <rmatte> I'm in Ottawa, Ontario
[23-Sep-2009 13:02:21] <rmatte> a few hours north from Toronto
[23-Sep-2009 13:02:47] <rmatte> ckrough: incorrect lol
[23-Sep-2009 13:02:59] <ke4qqq> ckrough: I think it rose - to 10 cents
[23-Sep-2009 13:03:07] <ckrough> lol
[23-Sep-2009 13:03:23] <rmatte> 1.00 CAD
[23-Sep-2009 13:03:23] <rmatte> =
[23-Sep-2009 13:03:24] <rmatte> 0.933786 USD
[23-Sep-2009 13:03:43] <rmatte> we're back in the hole
[23-Sep-2009 13:04:27] <rmatte> basically US dollar = CAD dollar + 7 cents
[23-Sep-2009 13:06:34] <rmatte> it has to be like that otherwise you guys won't buy as much from us
[23-Sep-2009 13:06:50] <rmatte> so we get the odd time where our dollar skyrockets but it's put back in to it's place fairly quickly
[23-Sep-2009 13:06:51]\n<ke4qqq> interesting- yesterday around 1500 Eastern, my fping\nstuff stopped collecting, or at least stopped reporting on the graphs
[23-Sep-2009 13:08:13] <rmatte> that sucks
[23-Sep-2009 13:08:18] <rmatte> is it still not reporting?
[23-Sep-2009 13:08:42] <ke4qqq> well snmp is reporting, but the graphs that fping collects isn't.
[23-Sep-2009 13:08:53] <ke4qqq> but it was working til around 1500 apparently
[23-Sep-2009 13:08:57] <rmatte> check to make sure zencommand daemon is running
[23-Sep-2009 13:09:07] <rmatte> login as zenoss user
[23-Sep-2009 13:09:09] <rmatte> do zenoss status
[23-Sep-2009 13:09:13] <rmatte> if it's running try...
[23-Sep-2009 13:09:16] <rmatte> zencommand run --cycle
[23-Sep-2009 13:09:23] <rmatte> see what kind of output you get
[23-Sep-2009 13:12:29] <rmatte> might take a while for any output to show up, you can also try: zenperfsnmp run --cycle
[23-Sep-2009 13:13:06] <rmatte> actually, throw a -v10 in there
[23-Sep-2009 13:13:17] <ke4qqq> yeah it's told me it's connected to zenhub, but nothing else thus far
[23-Sep-2009 13:13:17] <rmatte> zencommand run -v10 --cycle
[23-Sep-2009 13:13:23] <rmatte> zenperfsnmp run -v10 --cycle
[23-Sep-2009 13:13:53] <rmatte> it won't exit on it's own, you'll have to ctrl-c out when you're done
[23-Sep-2009 13:14:13] <rmatte> -v10 will show you a lot more
[23-Sep-2009 13:14:29] <rmatte> and yeh, it is zencommand that you want it looks like
[23-Sep-2009 13:14:47] <rmatte> so try out: zencommand run -v10 --cycle
[23-Sep-2009 13:14:50] <rmatte> and see what happens
[23-Sep-2009 13:15:41] <ke4qqq> http://fpaste.org/jgTX/
[23-Sep-2009 13:16:22] <rmatte> hmmm, you should be seeing a lot more than that
[23-Sep-2009 13:16:37] <rmatte> try restarting zenoss
[23-Sep-2009 13:16:39] <rmatte> then run it again
[23-Sep-2009 13:17:53] <rmatte> I see stuff like:
\n[23-Sep-2009 13:17:54] <rmatte> 2009-09-23 14:17:35,420 DEBUG\nzen.zencommand: Storing MemoryTotal = 781760.0 into\nDevices/30.30.254.33/Memory_MemoryTotal
[23-Sep-2009\n13:17:54] <rmatte> 2009-09-23 14:17:35,420 DEBUG zen.RRDUtil:\n/usr/local/zenoss/zenoss/perf/Devices/30.30.254.33/Memory_MemoryTotal.rrd:\n781760.0
[23-Sep-2009 13:17:54] <rmatte> 2009-09-23 14:17:35,421 DEBUG zen.zencommand: RRD save result: 781760.0
\n[23-Sep-2009 13:17:54] <rmatte> 2009-09-23 14:17:35,421 DEBUG\nzen.zencommand: Storing MemoryUsed = 595776.0 into\nDevices/30.30.254.33/Memory_MemoryUsed
[23-Sep-2009\n13:17:57] <rmatte> 2009-09-23 14:17:35,421 DEBUG zen.RRDUtil:\n/usr/local/zenoss/zenoss/perf/Devices/30.30.254.33/Memory_MemoryUsed.rrd:\n595776.0
[23-Sep-2009 13:18:00] <rmatte> 2009-09-23 14:17:35,421 DEBUG zen.zencommand: RRD save result: 595776.0
[23-Sep-2009 13:18:06] <rmatte> when I run zencommand
[23-Sep-2009 13:18:12] <rmatte> and those are command based datapoints
[23-Sep-2009 13:19:40] <ke4qqq> weird - lots of interesting stuff showed up now, including fping statistics
[23-Sep-2009 13:20:52] <rmatte> there you go
[23-Sep-2009 13:21:01] <rmatte> so for some reason your zenactions daemon took a crap
[23-Sep-2009 13:21:04] <rmatte> lord knows why
[23-Sep-2009 13:21:20] <rmatte> anyways, should be fine now
[23-Sep-2009 13:21:25] <rmatte> if it happens again let me know
[23-Sep-2009 13:21:31] <rmatte> then we can look in to it deeper
[23-Sep-2009 13:21:47] <ke4qqq> now - if I could just get the community esx zenpack to work properly
[23-Sep-2009 13:21:48] <rmatte> afk, going to grab some lunch
[23-Sep-2009 13:21:53] <ke4qqq> have a good lunch
[23-Sep-2009 13:22:22] <rmatte> there's a better ESX pack possibly coming to Core eventually (the old Enterprise one)
[23-Sep-2009 13:22:34] <ke4qqq> when?
[23-Sep-2009 13:26:20] <rmatte> not sure
[23-Sep-2009 13:27:06] <rmatte> the new enterprise one uses ESX's API to monitor it, the old one uses SNMP
[23-Sep-2009 13:33:50] <twm1010> nice
[23-Sep-2009 13:33:56] <twm1010> thats a good tidbit of info
[23-Sep-2009 13:35:15] <ke4qqq> yes - that's a very good bit of info
[23-Sep-2009 13:35:52] <ke4qqq> does that mean the new enterprise one will work with ESXi as well
[23-Sep-2009 13:42:24] <rmatte> I believe so
[23-Sep-2009 13:42:35] <rmatte> though I'm not 100% on that
[23-Sep-2009 13:42:53] <rmatte> they may have omitted the API stuff on ESXi as well as the SNMP
[23-Sep-2009 13:42:59] <rmatte> you'd have to check with VMWare to find out
\n[23-Sep-2009 13:43:20] <ke4qqq> I think esxi has the API stuff -\nthat's how virt center communicates with it.....just no snmp
[23-Sep-2009 13:45:53] <rmatte> ah
[23-Sep-2009 13:46:05] <rmatte> yeh, makes sense
[23-Sep-2009 13:49:12] <twm1010> rmatte: Why do you set the eventclasskey to defaultmapping?
[23-Sep-2009 13:49:32] <rmatte> twm1010: because if events come in as /Unknown they have no eventclasskey set
[23-Sep-2009 13:49:41] <rmatte> so defaultmapping matches that
[23-Sep-2009 13:50:02] <rmatte> otherwise you'd never we able to map those events
[23-Sep-2009 13:50:07] <rmatte> be able to*
[23-Sep-2009 13:50:08] <rmatte> rather
[23-Sep-2009 13:51:04] <twm1010> Ah... but if they're NOT unknown, I can leave it be
[23-Sep-2009 13:51:21] <rmatte> then you would check the events to see if it has an eventClassKey
[23-Sep-2009 13:51:53] <rmatte> in the event's details
[23-Sep-2009 13:52:24] <twm1010> OK... so I see two events here, they're different, both unknown
[23-Sep-2009 13:52:31] <twm1010> with the same classkey in the details
[23-Sep-2009 13:52:38] <rmatte> eventclasskey
[23-Sep-2009 13:52:40] <rmatte> not classkey
[23-Sep-2009 13:52:42] <rmatte> 2 different things
[23-Sep-2009 13:52:55] <twm1010> yeah i abbreviated, sorry, i did mean eventclasskey when i referred to it just now
[23-Sep-2009 13:53:08] <rmatte> what's the eventclasskey for them?
[23-Sep-2009 13:53:18] <twm1010> RSR_0
[23-Sep-2009 13:53:31] <twm1010> they do have some common text I can RegEx against though.
[23-Sep-2009 13:53:33] <rmatte> ok, so make a mapping where eventClassKey is RSR_0
[23-Sep-2009 13:53:42] <rmatte> and then specify the regex and the action
[23-Sep-2009 13:56:00] <twm1010> Gotcha... I really should go back and watch that presentation on events
[23-Sep-2009 13:56:15] <rmatte> hehe
[23-Sep-2009 13:56:17] <twm1010> eventClassKey is determined by the daemon? in this case, zenwin
[23-Sep-2009 13:56:23] <rmatte> I don't think she covered this specific stuff
[23-Sep-2009 13:56:31] <rmatte> she just went more in to the inner-workings of the system
[23-Sep-2009 13:56:39] <twm1010> still i want to understand where eventclasskey comes from
[23-Sep-2009 13:56:43] <rmatte> you should read the paper though
[23-Sep-2009 13:57:13] <rmatte> yeh, I'm not even 100% sure where it gets assigned from
\n[23-Sep-2009 13:57:49] <twm1010> for instance, i have a "last\nmessage repeated" event coming from syslog from a linux host
[23-Sep-2009 13:57:54] <twm1010> it doesn't ahve an eventclasskey
[23-Sep-2009 13:59:30] <twm1010> it looks to me like eventclassKey from a windows event log is the servicename + eventID
[23-Sep-2009 13:59:33] <twm1010> if there is one
[23-Sep-2009 14:00:27] <rmatte> I think it's whatever they code it to be
[23-Sep-2009 14:03:29] <twm1010> so.. it uses both the classkey AND the regex to identify a message?
[23-Sep-2009 14:03:44] <rmatte> correct
[23-Sep-2009 14:03:57] <rmatte> because you can have events with identical eventclasskeys but different messages
[23-Sep-2009 14:05:02] <twm1010> right, so whatever you put in the one for default mapping is like a catchall
[23-Sep-2009 14:05:19] <rmatte> it'll catch any that have a blank eventclasskey
[23-Sep-2009 14:05:22] <rmatte> as I understand it
[23-Sep-2009 14:05:30] <rmatte> but I also believe they have to be /Unknown event class
[23-Sep-2009 14:06:17] <rmatte> it works great for syslogs
[23-Sep-2009 14:08:10] <twm1010> Gotcha, still don't quite get it, but I'll start reading, you know, RTFM.
[23-Sep-2009 14:08:18] <rmatte> hehe
[23-Sep-2009 14:08:21] <rmatte> ypu
[23-Sep-2009 14:08:26] <rmatte> yup*
[23-Sep-2009 14:09:04] <rmatte> I haven't done any kind of event mapping or transform in over a month, so I'm slightly fuzzy on it myself
[23-Sep-2009 14:09:30]\n<rmatte> Going to be spending the next couple of days updating\nzenpacks with new mappings and deploying them across our servers
[23-Sep-2009 14:09:32] <rmatte> fun fun
[23-Sep-2009 14:15:30] <twm1010> ahhh ok, i get it now, admin manual FTW
[23-Sep-2009 14:15:45] <rmatte> lol
[23-Sep-2009 14:18:20] <twm1010> when it doesn't have an eventClassKey, it searches for defaultmapping
[23-Sep-2009 14:18:37] <twm1010> And you can have multiple defaultmappings, with different RegEx's
[23-Sep-2009 14:19:00] <rmatte> correct
[23-Sep-2009 14:19:24] <rmatte> you can have multiple of any event key with different regexes
[23-Sep-2009 14:19:35] <rmatte> but defaultmapping applies to any events with blank event keys
[23-Sep-2009 14:19:35] <twm1010> sure, to distinguish them
[23-Sep-2009 14:19:41] <rmatte> yeh
\n[23-Sep-2009 14:20:12] <twm1010> So if you do have an\neventClassKey, but don't give it a RegEx, or Example, it then uses that\nas a default for everything with that key?
[23-Sep-2009 14:20:35] <rmatte> no, you can't leave an event mapping blank like that
[23-Sep-2009 14:20:44] <twm1010> yeah, i figured that would be so
[23-Sep-2009 14:20:48] <rmatte> it won't do anything if you do that
[23-Sep-2009 14:21:06] <rmatte> It needs either a regex, a transform, or an evaluation
[23-Sep-2009 14:21:19] <twm1010> gotcha, it needs the rule or the regex, and default mapping lets you create catch-alls
[23-Sep-2009 14:21:24] <rmatte> yeh
[23-Sep-2009 14:21:29] <rmatte> rule, regex, or transform
[23-Sep-2009 14:21:32] <rmatte> any of those work
[23-Sep-2009 14:21:51] <rmatte> leaving everything blank results in nothing
[23-Sep-2009 14:22:59] <rmatte> and even if it didn't it'd be super bad practice
[23-Sep-2009 14:23:00] <rmatte>
[23-Sep-2009 14:24:53] <twm1010> Can an event match two separate mappings?
[23-Sep-2009 14:25:17] <rmatte> it takes the first one that it matches
[23-Sep-2009 14:25:35] <twm1010> ah, that's what "sequence" is for
[23-Sep-2009 14:25:39] <rmatte> yup
[23-Sep-2009 14:25:57] <twm1010> Sweet, my event console is cleaning up rather fast.
[23-Sep-2009 14:26:03] <rmatte> hehe
[23-Sep-2009 14:26:12] <rmatte> yeh, it gets really easy once you start picking up on this stuff
[23-Sep-2009 14:26:15] <twm1010> Now if I could only fix the 25 acknowledged problems that really are issues.
[23-Sep-2009 14:26:21] <rmatte> the defaultmapping thing was the last thing that I learned
[23-Sep-2009 14:26:26] <rmatte> and it's a frigging god-send
[23-Sep-2009 14:26:30] <rmatte> hehe
[23-Sep-2009 14:27:05] <rmatte> I kept having syslog messages coming in with a completely blank message
[23-Sep-2009 14:27:19] <rmatte> so I made a defaultmapping event mapping, and made the regex ^$
[23-Sep-2009 14:27:24] <rmatte> and voila, problem solved
[23-Sep-2009 14:57:36] <rmatte> g'day Matt
[23-Sep-2009 14:57:41] <rmatte> lots of meetings today?
[23-Sep-2009 14:58:55] <mrayzenoss> yeah, lots of meetings
[23-Sep-2009 14:59:02] <rmatte> ah
\n[23-Sep-2009 14:59:39] <mrayzenoss> beta 3 artifacts are on\nSourceForge, the beta download page is being migrated from zenoss.com\nto zenoss.org, so there will be a lag in that update
[23-Sep-2009 15:06:05] <rmatte> cool
[23-Sep-2009 15:06:19] <rmatte> upgrade from beta 2 to beta 3 will work I assume?
[23-Sep-2009 15:06:46] <rmatte> and is the stack installer going to come out like a week after for this release?
[23-Sep-2009 15:14:04] <mrayzenoss> there's a stack installer
[23-Sep-2009 15:14:09] <rmatte> w00t
[23-Sep-2009 15:14:14] <mrayzenoss> upgrades are probably still broken
[23-Sep-2009 15:14:19] <rmatte> crap
[23-Sep-2009 15:14:20] <rmatte> lol
[23-Sep-2009 15:14:20] <mrayzenoss> at least from 2.4.X->beta
[23-Sep-2009 15:14:21] <rmatte> oh well
[23-Sep-2009 15:14:29] <mrayzenoss> maybe beta2->beta3 may work
[23-Sep-2009 15:14:38] <rmatte> yeh, but I'll try from beta2->beta3 to see if it works
[23-Sep-2009 15:14:42] <rmatte> if not, oh well, reinstall
[23-Sep-2009 15:14:57] <mrayzenoss> so next week's beta should be the first to be worth testing upgrades with
[23-Sep-2009 15:15:03] <rmatte> cool
[23-Sep-2009 15:17:31] <rmatte> yay, I get a new commissioning project tomorrow
[23-Sep-2009 15:17:39] <rmatte> 15 devices, but apparently most of them are stacked switches
[23-Sep-2009 15:17:52] <rmatte> where they are a bunch of switches all linked together somehow under 1 IP address
[23-Sep-2009 15:18:07] <rmatte> it's going to be fun to figure out how to monitor the other switches
[23-Sep-2009 15:32:10] <chemist> If I restart zenmodelerwill it attempt to remodel devices 12 hours after that?
[23-Sep-2009 15:32:43] <rmatte> chemist: Honestly, no idea, but I think it follows some sort of zenhub cycle time
[23-Sep-2009 15:56:57] <cgibbons> hmm
[23-Sep-2009 16:45:00]\n<ke4qqq> rmatte: the switches will be easy - they have a single\nipaddress, and act like modules of a chassis switch or router so you'll\nhave 2/0/1-48 and then 3/0/1-48 will be the next switch etc.
[23-Sep-2009 18:00:01] <davetoo> I know, everybody is probably out or asleep
But does anyone know what the "Messages" dashboard portlet does?[23-Sep-2009 18:00:05] <davetoo> Messages from what/where?
[23-Sep-2009 19:33:35] <davetoo> whee
[23-Sep-2009 19:34:36] <davetoo> ${here/ZenPackManager/packs/ZenPacks.xxorgxx.xxthingxx/path}/libexec/check_thing.py
[23-Sep-2009 20:16:02] <etank> davetoo: i think that the message portlet is for when zenoss has jobs queued up it will leave information there
[23-Sep-2009 20:16:08] <etank> i could be wrong on that though
[23-Sep-2009 20:16:22] <etank> i have never seen anything at all in messages
[23-Sep-2009 20:16:41] <etank> could be possible that it will have stuff if you are doing a discovery on a large subnet
\n[23-Sep-2009 20:17:34] <etank> davetoo:\nmessage/38671%3Bjsessionid=07F9C4F40DECB2BEC062C491E1427675.node0
[23-Sep-2009 20:18:02] <davetoo> ah, ok, thanks
[23-Sep-2009 20:18:23] <davetoo> I'd actually like to have a "wall" portlet
Where admins can put an MOTD or similar[23-Sep-2009 20:21:21] <etank> so where you could post notices and stuff
[23-Sep-2009 20:21:47] <davetoo> yes; we'll have 20-30 people looking at the dashboard around the world
[23-Sep-2009 20:22:13] <etank> so kind of a mashup of Facebook and Zenoss
[23-Sep-2009 20:22:15] <davetoo> almost a chat window
[23-Sep-2009 20:23:04] <davetoo> where we could post things like "Hey, we just lost a core router in Site X, so you might get alert spam"
[23-Sep-2009 20:24:05] <etank> that would actually be kind of nice to have
\n[23-Sep-2009 20:24:11] <LW> that would be a good idea, my end\nuser call center is getting a view into our zenoss install soon..... \nand I'd really prefer for them not to call me if I can help it for\nevery little thing
[23-Sep-2009 20:24:34] <LW> "Hey! we know about this!" would be a life saver
[23-Sep-2009 20:24:41] <davetoo> yeah, we're about to roll out Enterprise,
[23-Sep-2009 20:25:09] <etank> i wish there was a way to select multiple events at once and give them all the same "log" information
\n[23-Sep-2009 20:25:10] <davetoo> and for the most part the folks\nin Bangalore are supposed to be the human alerting system watching the\nZenoss dashboard
[23-Sep-2009 20:25:20] <etank> instead of copy paste copy paste etc
[23-Sep-2009 20:25:41] <davetoo> like, from multiple devices?
[23-Sep-2009 20:25:49] <etank> yeah
[23-Sep-2009 20:25:53] <davetoo> Or all the same device but different events
[23-Sep-2009 20:25:57] <etank> but all from the same device would be good too
[23-Sep-2009 20:26:15] <davetoo> learn zendmd/python
[23-Sep-2009 20:26:40] <etank> thats good and all but it would be nicer to be able to do it from the web interface
[23-Sep-2009 20:26:48] <davetoo> yes
[23-Sep-2009 20:27:49] <etank> i dont want to have to drop to a python console whenever something like this happens
[23-Sep-2009 20:29:13] <davetoo> event management is one area where zenoss needs work, I think
\n[23-Sep-2009 20:29:24] <etank> well im going to be at the\ncommunity day training at OLF this week so maybe i will bring it up
[23-Sep-2009 20:29:39] <davetoo> I just did my first "real" transform
[23-Sep-2009 20:29:58] <etank> zenoss does A LOT of things right but it is not perfect
[23-Sep-2009 20:30:14] <etank> i need to learn more about transforms actually
[23-Sep-2009 20:30:27] <etank> we are trying to make a purchase of enterprise at work
[23-Sep-2009 20:30:33] <davetoo> This was for an snmp trap
[23-Sep-2009 20:30:49] <etank> we made it through the evaluation and it could do most everything we need it to do
[23-Sep-2009 20:30:49] <davetoo> to change severity based on the trap value
[23-Sep-2009 20:31:18] <davetoo> yes, we used Core for a couple of years
[23-Sep-2009 20:31:29] <davetoo> needed the global view
\n[23-Sep-2009 20:31:32] <etank> i think i did one that if a\nspecific event happened between 5:30 and 6:30 in the morning it would\njust archive it
[23-Sep-2009 20:31:59] <davetoo> I have to say that I'm much happier with 2.4.x than previous releases
\n[23-Sep-2009 20:32:13] <etank> that is the time of day that we\nreindex the databases so the web servers are going to complain so we\njust ignore those
[23-Sep-2009 20:32:21] <etank> im looking forward to 2.5
[23-Sep-2009 20:32:29] <davetoo> oh we still have to go through the whole mapping/escalation thing
[23-Sep-2009 20:32:33] <etank> the SQL zenpack will have some nice new functions
[23-Sep-2009 20:32:53] <davetoo> I don't do any app monitoring(yet)
[23-Sep-2009 20:33:13] <etank> the inheritance that is in zenoss makes it so nice to use
[23-Sep-2009 20:33:27] <etank> build the device class the way you want it and then just add the devices
[23-Sep-2009 20:33:43] <davetoo> yes;
[23-Sep-2009 20:33:48] <davetoo> I modify that idea slightly.
[23-Sep-2009 20:33:51] <etank> makes life soooo easy
[23-Sep-2009 20:34:07] <davetoo> I break the big Device templates up into smaller sections
[23-Sep-2009 20:34:23] <davetoo> and put those in the Device Templates list instead
\n[23-Sep-2009 20:34:51] <etank> we have started building our own\ntemplates and device classes that are specific to a type of device
\n[23-Sep-2009 20:34:55] <davetoo> allows me to have a bunch of\ndifferent os versions that have differing levels of snmp-brokennes
[23-Sep-2009 20:35:02] <etank> same here
[23-Sep-2009 20:35:09] <etank> we do it by domains as well
[23-Sep-2009 20:35:16] <davetoo> so I have a template-let that just does very basic CPU stats
[23-Sep-2009 20:35:26] <etank> so the zenproperties will have the right authentication information
[23-Sep-2009 20:35:28] <davetoo> and then another tiny one that does stuff that only BSD systems have
[23-Sep-2009 20:36:32] <etank> whatever makes it work
[23-Sep-2009 20:37:28] <etank> well im out for now. later davetoo
[23-Sep-2009 20:37:34] <davetoo> ok bye
[24-Sep-2009 00:00:46] [disconnected at Thu Sep 24 00:00:46 2009]
[24-Sep-2009 00:00:46] [connected at Thu Sep 24 00:00:46 2009]
[24-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
[24-Sep-2009 03:51:31] <Egyptian[web]> hello all
\n[24-Sep-2009 04:16:39] <Egyptian[web]> i want to monitor my san\nstorage and i have the mib and its seen but the device class path is\nnot set correctly .. i have to manually change it from / to\n/server/scan (as per documentation on auto discovery) but i m not happy\nwith that either ..
[24-Sep-2009 04:17:01]\n<Egyptian[web]> is there something like /storage ? if not ..how\ndo i create it? and what is the best practice to name the device class?
[24-Sep-2009 04:17:06] * Egyptian[web] takes a breath
[24-Sep-2009 05:07:30]\n<sergeymasushko> Hi guys! we've updated to 2.4.5 and now when I\ntry to open 'devices' page or try to use search I got the following:\nhttp://pastebin.com/m72b50e5e
[24-Sep-2009 05:40:11] <Egyptian[web]> sergeymasushko: do you have an nfs mount on the zenoss pc ?
[24-Sep-2009 06:45:28] <Egyptian[web]> is there like a zenpack for oracle? ..
[24-Sep-2009 06:45:50] <Egyptian[web]> actually .. all i want is to monitor my oracle db .. how do i go about it?
\n[24-Sep-2009 08:36:18] <Egyptian[web]> i set up a page for myself\n( a modified command to call a script that will send me an gsm text\nmessage) and now .. i want to test it.. how?
[24-Sep-2009 08:42:28] <straterra> firewall zenoss off from a machine..and see if you get paged o.O
[24-Sep-2009 08:52:20] <Egyptian[web]> straterra: interesting .. i can bring down a test machine ..thanks
[24-Sep-2009 08:52:28] <straterra> No problem
[24-Sep-2009 08:56:29] <Egyptian[web]> is it possible to like disable monitoring of a device?
[24-Sep-2009 08:59:19] <straterra> Sure
[24-Sep-2009 08:59:28] <straterra> Put it in production state of maintenance
[24-Sep-2009 08:59:35] <straterra> (I think its called production state)
\n[24-Sep-2009 09:19:41] <rmatte> Egyptian[web]: just a tip, if you\ngo to Events, then select "Add Event" from the top dropdown menu you\ncan generate a test event with whatever values you wish
[24-Sep-2009 09:20:06] <rmatte> quite useful for testing
[24-Sep-2009 09:20:58] <kmtsun> anyone going to zenoss pres @ ohio linux fest?
[24-Sep-2009 09:21:34] <Egyptian[web]> straterra: no gsm text btw
[24-Sep-2009 09:22:19] <mrayzenoss> kmtsun: I'm there!
[24-Sep-2009 09:22:31] <mrayzenoss> kmtsun: we've got over 60 people signed up for the free training
[24-Sep-2009 09:22:35] <kmtsun> nice ..i will be there tomorrow
[24-Sep-2009 09:22:52] <kmtsun> is it too late to sign up?
[24-Sep-2009 09:22:56] <mrayzenoss> nope
[24-Sep-2009 09:27:56] <twm1010> changelog?
\n[24-Sep-2009 09:29:53] <rmatte> eugh, even MORE screwed up\nbehaviour to do with setting the Product info for devices, any server\nthat I upgraded to 2.4.x from 2.3.3 is having even more issues (can't\nsee certain product models in the dropdown box), than those that were\noriginally 2.4.x installs
[24-Sep-2009 09:30:01] <rmatte> this is driving me loopy, I need to find a workaround for this
[24-Sep-2009 09:32:40] <rmatte> how do I manually link a device to a different product class using zendmd?
[24-Sep-2009 09:32:46] <rmatte> I'll have to do that as a last resort
\n[24-Sep-2009 09:34:25] <rmatte> we use the Zenoss inventory for\nbilling, so we can't have incorrect product details in there
[24-Sep-2009 09:43:22] <twm1010> Anyone using Zenoss for IIS monitoring?
[24-Sep-2009 09:44:04] <etank> twm1010: we are (some)
\n[24-Sep-2009 09:44:40] <etank> anyone know what could be causing\n"localhost zenactions heartbeat failure" events every 2 hours?
[24-Sep-2009 09:45:41] <rmatte> etank: it's probably just getting overworked
[24-Sep-2009 09:45:47] <twm1010> does the log for zenactions daemon show anything?
\n[24-Sep-2009 09:45:54] <rmatte> I get the occasional heartbeat\nfailure for different processes on different zenoss servers
\n[24-Sep-2009 09:45:58] <etank> rmatte: i just found this\nhttp://www.mail-archive.com/zenoss-users@zenoss.org/msg24200.html//www.mail-archive.com/zenoss-users@zenoss.org/msg24200.html
[24-Sep-2009 09:45:59] <rmatte> I wouldn't be too concerned with it
[24-Sep-2009 09:46:04] <etank> i think that is the issue
[24-Sep-2009 09:46:32] <rmatte> ah
[24-Sep-2009 09:46:39] <rmatte> well, if it's Chet, it's probably right
[24-Sep-2009 09:46:41] <rmatte> lol
[24-Sep-2009 09:46:49] <etank> Last Checked 2009/09/11 00:43:38.000
[24-Sep-2009 09:47:00] <rmatte> there be your problem
[24-Sep-2009 09:47:07] <etank> looks like that is most likely the issue
[24-Sep-2009 09:47:19] <rmatte> I'll have to remember that one
[24-Sep-2009 09:47:35] <etank> thank God for google
[24-Sep-2009 09:47:38] <rmatte> I guess it checks for updates every 2 hours
[24-Sep-2009 09:47:54] <rmatte> I doubt that would have hurt anything though had you left it
[24-Sep-2009 09:48:26] <twm1010> Hrmm.. Igor's IIS template looks perfect
[24-Sep-2009 09:48:44] <rmatte> Yeh, Egor's stuff is pretty good
[24-Sep-2009 09:48:59] <rmatte> mrayzenoss: woot, was wondering when that was coming out
\n[24-Sep-2009 09:49:31] <rmatte> mrayzenoss: do you know of any\ndocs laying around that explain how to use zendmd to actually link\nitems? I need to know how to link productClass on devices to a\nspecific product class
[24-Sep-2009 09:49:38] <twm1010> hrmmm... which ports are required for WMI communication?
[24-Sep-2009 09:49:51] <cgibbons> lots
\n[24-Sep-2009 09:49:51] <rmatte> mrayzenoss: need to use it as a\nworkaround for this product bug until it gets fixed, because we can't\ngo without being able to specify that info
[24-Sep-2009 09:49:54] <cgibbons> it's dynamic by default
[24-Sep-2009 09:50:00] <cgibbons> you can configure it to be minimal if you need to restrict it
[24-Sep-2009 09:50:23] <twm1010> well, how about on the client end
\n[24-Sep-2009 09:50:32] <cgibbons> but 135 for the initial\nconnection, and then the server reassigns a dynamic port after that
[24-Sep-2009 09:50:39] <twm1010> gotcha
[24-Sep-2009 09:50:44] <cgibbons> doing firewall stuff?
\n[24-Sep-2009 09:51:11] <twm1010> yeah, im going to be monitoring\nour production webservers (windows), and there are quite a few\nfirewalls between here and there that are not very permissive
[24-Sep-2009 09:51:27] <cgibbons> k, let me get you the link... very easy to configure on the devices to be more limited
[24-Sep-2009 09:51:39] <cgibbons> http://msdn.microsoft.com/en-us/library/ms809327.aspx
\n[24-Sep-2009 09:53:38] <cgibbons> and if you use the perfmon\nmonitoring in enterprise i think you'll need 445, too, but i'll need to\ndouble chec
[24-Sep-2009 09:53:45] <Brad_K>\nrmatte: so, that product problem is driving you crazy, eh? got any\nreal fixes for it yet?
[24-Sep-2009 09:54:11] <twm1010> Does it determine whether something is "Internet" facing via it's IP class?
\n[24-Sep-2009 09:54:35] <Brad_K> rmatte: or, can you at least\npoint me at good zendmd resources, so that we can jury-rig a solution\nto the same problem here?
[24-Sep-2009 09:54:36] <rmatte> Brad_K: working on it
[24-Sep-2009 09:54:44] <rmatte> Brad_K: I need to know how to link items using zendmd
[24-Sep-2009 09:55:13] <mrayzenoss> rmatte: what was the # on that, I'll check on it
[24-Sep-2009 09:55:13] * Brad_K is going to make sure that we file a customer service case on this bug...
\n[24-Sep-2009 09:55:21] <rmatte> Brad_K: I'm more working on\nworkarounds that actually fixing the main issue, I don't have time to\ndo that at the moment
[24-Sep-2009 09:55:37] <rmatte> Brad_K: good, I already have a trac case open for it
[24-Sep-2009 09:55:47] <rmatte> Brad_K: If I find a workaround I'll let you know
[24-Sep-2009 09:55:54] <Brad_K> so you're taking the open source road, and i
\n[24-Sep-2009 09:56:21] <Brad_K> so you're taking the open source\nroad, and i'll be taking the enterprise road, and you'll probably get\nthere before me. ;-)
[24-Sep-2009 09:56:26] <rmatte> yeh, we're core users so we don't have as much push as you guys
[24-Sep-2009 09:56:32] <rmatte> you'd be surprised
[24-Sep-2009 09:56:37] <rmatte> http://dev.zenoss.com/trac/ticket/5473
[24-Sep-2009 09:56:41] <Brad_K> rmatte: if you can let me know what the trac case is, i can make sure we link to it.
[24-Sep-2009 09:56:42] <rmatte> the ticket is set as backlog too
[24-Sep-2009 09:56:49] <rmatte> meaning, they'll get to it whenever
[24-Sep-2009 09:57:01] <rmatte> been opened for 4 weeks now
\n[24-Sep-2009 09:57:19] <Brad_K> rmatte: yeah, but you have direct\naccess to the developers here, and you know a lot more about the\nproblem than i do. i'm still getting spun up on how i can properly\nmonitor a web page.
[24-Sep-2009 09:57:37] <rmatte> hehe
\n[24-Sep-2009 09:57:58] <Brad_K> the zenwebtx stuff seems to be\nmissing one critical component that i need to be able to capture,\nbefore i can actually do a proper log in/log out session.
[24-Sep-2009 09:58:19] <Brad_K> got a meeting. bbiab.
\n[24-Sep-2009 09:58:27] <rmatte> Brad_K: you can always write some\nexternal script to do the login/logout and then just tie it in to\nZenoss if worst comes to worst
[24-Sep-2009 09:58:33] <rmatte> not sure how scripty you are
[24-Sep-2009 09:58:35] <rmatte> k cool
[24-Sep-2009 10:02:08] <mrayzenoss> rmatte: Brad_K: I pinged the dev on it, it's on his schedule for this week
[24-Sep-2009 10:02:23] <mrayzenoss> of course, it gets pushed anytime he gets a P1 assigned to him
[24-Sep-2009 10:03:13]\n<rmatte> I'd almost consider this a P1 at this point since a main\ncomponent of Zenoss is *totally* not working as designed.
[24-Sep-2009 10:04:13] <mrayzenoss> yeah
[24-Sep-2009 10:04:49] <chudler> doesn't that describe all bugs? i.e. not working as designed?
[24-Sep-2009 10:05:50] <rmatte> not really
[24-Sep-2009 10:06:52]\n<twm1010> @cgibbons: so at the minimum, Zenoss will need to be\nable to reach the client on port 135, and a range of ports i designate,\nand likewise the reverse.
[24-Sep-2009 10:07:54]\n<cgibbons> the device doesn't connect back to the zenoss\ncollector, so as long as your firewall lets an outbound connection to\nthose ports from any client port work you're golden
[24-Sep-2009 10:08:43]\n<mrayzenoss> chudler: P1 indicates broken, can't ship with it. \nP2 indicates broken, possible work-around or known existing issue
[24-Sep-2009 10:08:55] <mrayzenoss> we won't ship with P1s
[24-Sep-2009 10:11:58] <twm1010> OK, thanks, I'll start plugging away at it.
\n[24-Sep-2009 10:19:52] <rmatte> mrayzenoss: I don't think you can\ntechnically ship with this defect, but it's up to you guys really
\n[24-Sep-2009 10:20:33] <rmatte> I just know that King Crab is\nsupposed to be a big deal release, so releasing with obvious bugs like\nthat wouldn't be the greatest idea
[24-Sep-2009 10:20:40] <mrayzenoss> right
[24-Sep-2009 10:20:47] <mrayzenoss> they're being worked on
[24-Sep-2009 10:23:10] <rmatte> wow, just found a solution to something crazy weird
\n[24-Sep-2009 10:23:49] <rmatte> I had a bunch of products missing\nfrom the Cisco products dropdown, I had to go in to each one's\nproperties, modify something (I removed the last number on the OID),\nsave them, then add it back and save them again, now they show up in\nthe list
[24-Sep-2009 10:24:10] <rmatte> oh well, at least I know the solution to that issue now
[24-Sep-2009 10:25:58] <twm1010> alright someone set me straight
[24-Sep-2009 10:26:05] <twm1010> should the zenpacks have .zip or.egg on the end
[24-Sep-2009 10:27:08] <rmatte> depends
[24-Sep-2009 10:27:16] <rmatte> if it says .egg.zip then it's really an egg
[24-Sep-2009 10:27:25] <rmatte> if it says just .zip, then it's a zip
[24-Sep-2009 10:27:33] <rmatte> for some reason the new site zipped up all of the egg files
[24-Sep-2009 10:27:55] <twm1010> yes, thats throwing me off again
[24-Sep-2009 10:28:02] <rmatte> .zip are older zenpacks (which will eventually need to be converted to eggs)
[24-Sep-2009 10:28:20] <rmatte> yeh, it'd be nice if all of those egg packs could get unzipped at some point
[24-Sep-2009 10:28:20] <rmatte> lol
[24-Sep-2009 10:28:51] <rmatte> not sure if that's a Jive quirk or not
[24-Sep-2009 10:29:53] <mrayzenoss> It's a Jive quirk
[24-Sep-2009 10:29:56] <mrayzenoss> I've opened a ticket
[24-Sep-2009 10:30:59] <rmatte> figured as much
[24-Sep-2009 10:31:02] <rmatte> hehe
[24-Sep-2009 10:32:07] <rmatte> ok, well now that that issue with the list is fixed the product info is semi-usable
\n[24-Sep-2009 10:32:36] <rmatte> still a pain though since you\nneed to manually set everything if you want to use the edit tab at all,\notherwise it'll clear all your info
[24-Sep-2009 10:32:53] <rmatte> (product and software info that is)
[24-Sep-2009 10:33:41] <mrayzenoss> I've closed 2 P1s in the last 30 minutes
[24-Sep-2009 10:33:53] <cgibbons> woot
[24-Sep-2009 10:33:57] <rmatte> nice
[24-Sep-2009 10:34:22] * rmatte thinks he really just set P3s to P1s right before closing them, sneaky sneaky
[24-Sep-2009 10:34:47] <mrayzenoss> nah, just glaring issues that were easy to fix
[24-Sep-2009 10:35:02] <cgibbons> we still have a bad habit of making issues P1s when they're really not anywhere close
[24-Sep-2009 10:35:37] <rmatte> hehe
\n[24-Sep-2009 10:36:52] <rmatte> cgibbons: I had a question\nrelating to http://dev.zenoss.org/trac/ticket/5376 and\nhttp://dev.zenoss.org/trac/ticket/4997, not sure if you'll be able to\nanswer...
[24-Sep-2009 10:37:40] <rmatte> I had\nopened 5376 because I was working no a transform to update interface\nstatus when an interface went down or up, Chet closed the ticket\nclaiming that 4997 resolved this issue...
[24-Sep-2009\n10:38:12] <rmatte> Chet stated in 4997 "This should provide more\nconsistency while allowing for real-time interface events to be\nrepresented as well."
[24-Sep-2009 10:39:56]\n<rmatte> My question is this, if an interface has an inactive\nstatus, will Zenoss stop monitoring it as I saw in previous versions of\nZenoss? If not, I was expecting interface status monitoring to be\nbuilt in to King Crab (which it appears to not be). To reflect actual\ninterface status, would it be possible to modify the transform in 4376\nto modify the status of interfaces as they go down/up? If so, what is\nthe zendmd property that would need to be manipulated to do
[24-Sep-2009 10:41:08] <rmatte> (sorry about the bible long question)
\n[24-Sep-2009 10:42:41] <twm1010> @cgibbons: as it turns out, the\nfirewall is already configured to allow WMI between Zenoss and the\nservers, w00t!
[24-Sep-2009 10:44:22] <rmatte>\nWhen I say "I was expecting interface status monitoring to be built in\nto King Crab" I mean not having to resort to this:\ndocs/DOC-2494
[24-Sep-2009 10:44:41] <rmatte> but I tested with King Crab and apparently it's still necessary
[24-Sep-2009 10:47:24] <twm1010> your saying, you want that to be default behavior?
[24-Sep-2009 10:47:40] <rmatte> me?
[24-Sep-2009 10:48:05]\n<rmatte> Yes, I'd like it to automatically monitor any ports that\nit sees as up when modelling, and then reflect active port status for\nthose ports based on SNMP
[24-Sep-2009 10:48:38]\n<rmatte> I constantly have NOC staff going to the OS tab and\nsaying "Well, the port is up!" and then I have to explain to them how\nthe port status shown on the OS tab is not real-time, it only gets\nupdated when modelling
[24-Sep-2009 10:48:51] <rmatte> which is something that, as I understood, was going to be changed in King Crab
[24-Sep-2009 10:49:35] <twm1010> well then, i look forward to a response
\n[24-Sep-2009 10:49:48] <rmatte> if I need to still do the\nmonitoring with a transform and a threshold then so be it, I just want\nto make sure that if the status is inactive that Zenoss continues to\nmonitor the port so that eventually a clear status can come in and the\nport status can be updated
[24-Sep-2009 10:50:33]\n<rmatte> If it does, then I can just modify my transform to\nupdate the status on ports as they go up and down
[24-Sep-2009 10:51:00] <twm1010> ah, so the perf template checks for ifOperStatus
[24-Sep-2009 10:51:12] <twm1010> and if it fails, your transform modifies that status
\n[24-Sep-2009 10:51:14] <rmatte> The issue I was having with my\nprevious attempt at doing this was that when I'd update the port status\nto down, Zenoss would completely stop monitoring the port
\n[24-Sep-2009 10:51:36] <rmatte> so no clear status would come in\nwhich would have triggered the transform to change the port status back\nto active
[24-Sep-2009 10:52:04] <rmatte> twm1010: correct
[24-Sep-2009 10:52:05] <twm1010> ah.... I see the disparity, you almost need a third indicator there
[24-Sep-2009 10:52:26] <rmatte> almost hehe
\n[24-Sep-2009 10:52:51] <rmatte> although I noticed that when you\nmodel a device now, Zenoss automatically sets monitoring status of\nports that are down to not monitored, which is perfect
[24-Sep-2009 10:53:06] <rmatte> but all I want to know is whether or not it continues to monitor those ports even if they are showing as down
[24-Sep-2009 10:53:13] <rmatte> actually, I just though of a way to test that...
[24-Sep-2009 10:53:26] <rmatte> It should be quite easy
[24-Sep-2009 10:54:33] <rmatte> well now that's just stupid
[24-Sep-2009 10:54:36] <rmatte> hmmm
[24-Sep-2009 10:56:05]\n<rmatte> If a port is detected as down during modelling it marks\nthe status of the port as gray, which is fine, but it absolutely\nrefuses to allow you to enable monitoring on it
[24-Sep-2009 10:56:16] <rmatte> so effectively you need to remodel the entire device every time a port is turned up
[24-Sep-2009 10:56:18] <rmatte> ridiculous
[24-Sep-2009 10:56:23] <etank> ZenWebTx is some cool stuff. TestGen4Web makes it fun to play with too
[24-Sep-2009 10:56:29] <rmatte> I'm going to reopen 4997
[24-Sep-2009 10:57:21] <rmatte> Actually, I'll just make a new ticket I guess
[24-Sep-2009 10:57:54] <rmatte> meh, reopen
[24-Sep-2009 11:11:10] <rmatte> http://dev.zenoss.org/trac/ticket/4997#comment:5
[24-Sep-2009 11:11:12] <rmatte> voila
[24-Sep-2009 11:12:18] <rmatte> that's a biggy
[24-Sep-2009 11:22:31] <rmatte> damn, just noticed another issue with the product model info
[24-Sep-2009 11:22:35] <rmatte> time to add it to the ticket
\n[24-Sep-2009 11:24:11] <cgibbons> i have an old email from chet\non this topic, rmatte... (I haven't read everything you've mentioned)\nand he just said customers typically handle this with trap handling but\n4997 was going to make it more obvious
[24-Sep-2009 11:24:23] <cgibbons> but i'm not going to be much help beyond that
[24-Sep-2009 11:24:43] <rmatte> all good
[24-Sep-2009 11:24:54] <rmatte> I reopened 4997 since it still has a bug
[24-Sep-2009 11:25:04] <rmatte> well, has a bug due to the change I should say
\n[24-Sep-2009 11:39:36] <rmatte> well, I finally figured out a\nreally rough way of getting product info specified for devices
\n[24-Sep-2009 11:39:56] <rmatte> I'm going to write up a quick\nblog post with the steps so that people can use them until the issue\ngets resolved
[24-Sep-2009 11:51:23] <Brad_K>\netank: one thing i've discovered about TestGenn4Web is that there is at\nleast one type of dialog box that browsers can open up that isn't seen\nduring the capture session, so there's no way to model that in your\nTwill script. At least, not so far as I can tell.
[24-Sep-2009 11:53:25] <Brad_K> etank: see http://developer.spikesource.com/forums/viewtopic.php?p=2585#2585
\n[24-Sep-2009 11:54:53] <etank> Brad_K: do you know if the\nZenWebTx pack can handle it if the site being queried returns the data\nin a popup window?
[24-Sep-2009 11:55:17] <etank> that is how our site works and im wondering if it would cause it to hang in any way
\n[24-Sep-2009 11:56:35] <Brad_K> etank: don't know. there may be\na way to get Twill to handle that situation, even if testgen4web can't\ncapture it. i have not yet gotten that far.
[24-Sep-2009 11:57:04] <Brad_K> if you take the script as generated by testgen4web and replay that, the login definitely hangs for me.
[24-Sep-2009 11:57:05] <etank> yeah me either. i just started playing with twill today
[24-Sep-2009 11:57:12] <chudler> etank: what method does the site use to popup? twill does not interpret javascript.
[24-Sep-2009 11:57:34] <etank> chudler: good question. im not the one who designed the site so im not sure.
[24-Sep-2009 11:57:39] <etank> i will look into it though
[24-Sep-2009 12:00:20]\n<rmatte> Brad_K:\nblogs/ryanmatte/2009/09/24/specifying-hardware-and-software-product-info-in-245-until-bug-is-fixed
[24-Sep-2009 12:16:17] <Brad_K> rmatte: thanks!
[24-Sep-2009 12:21:03]\n<etank> we are doing a IT Security review of Zenoss here today.\ndoes anyone have a link to the ports that it uses to gather data?
\n[24-Sep-2009 12:21:24] <etank> i know most of them i think but it\nwould be better to have product documentation to back it up
[24-Sep-2009 12:32:26] <rmatte> etank: it really depends on what kind of data you're collecting
[24-Sep-2009 12:32:38] <rmatte> it uses standard SNMP, Syslog, and mail ports
[24-Sep-2009 12:32:42] <rmatte> as well as SSH
[24-Sep-2009 12:32:54] <rmatte> It can also do WMI monitoring, which is fairly random portwise
[24-Sep-2009 12:33:22] <rmatte> It uses ports 8080 for the web interface by default
\n[24-Sep-2009 12:33:46] <rmatte> and it uses some other ports for\ncommunication between the various daemons and zenhub, don't know it off\nthe top of my head
[24-Sep-2009 12:33:49] <rmatte> plus sql ports
[24-Sep-2009 12:33:52] <rmatte> there's a wide range
[24-Sep-2009 12:34:06] <rmatte> I'm sure there's a port for zenhub to communicate with zope as well
[24-Sep-2009 12:37:07] <rmatte> I don't know of any tech sheet that lists them all
[24-Sep-2009 12:37:07] <Brad_K> other ports are like 8079 and 8091
[24-Sep-2009 12:37:29] <rmatte> yeh, that sounds about right, somewhere in that range
[24-Sep-2009 12:38:28] <Brad_K> and 8090, 6010 8081, 8100, 8789
[24-Sep-2009 12:38:42] <Brad_K> which lsof
[24-Sep-2009 12:38:52] <Brad_K> oops. wrong window. ;-)
[24-Sep-2009 12:39:20] <rmatte> hehe
\n[24-Sep-2009 12:39:28] <chudler> as rmatte mentioned, outgoing\nunsolicited monitoring traffic from zenoss can *potentially* be over\nany port, especially considering the portscan plugin :-p
[24-Sep-2009 12:39:34] <rmatte> /usr/bin/lsof
[24-Sep-2009 12:40:05] <rmatte> yeh, you really need a system that isn't too locked down outgoing to use Zenoss properly
[24-Sep-2009 12:40:17] <Brad_K> rmatte: $ which lsof
[24-Sep-2009 12:40:18] <Brad_K> lsof: Command not found.
[24-Sep-2009 12:40:24] <rmatte> that sucks
[24-Sep-2009 12:40:25] <rmatte> lol
[24-Sep-2009 12:40:52] <kisielk> is there a guide somewhere to optimizing zenoss / figuring out what part of it is slow?
\n[24-Sep-2009 12:41:51] <kisielk> some times parts of it seem to\nget dog slow. Like if I type a hostname in to the search box, it can\ntake several minutes to get me a result
[24-Sep-2009 12:42:03] <rmatte> docs/DOC-2521
[24-Sep-2009 12:42:07] <Brad_K> that's probably a DNS issue.
\n[24-Sep-2009 12:42:41] <rmatte> that explains how to tweak it\nperformancewise, but it sounds like you may have a different issue
[24-Sep-2009 12:42:55] <rmatte> what specs on the box?
[24-Sep-2009 12:42:56] <kisielk> well, it's only sometimes
[24-Sep-2009 12:42:59] <rmatte> how many devices monitored?
[24-Sep-2009 12:43:01] <chudler> yeah sounds like something up with the zope db..
[24-Sep-2009 12:43:02] <kisielk> if I just restart Zenoss, it's usually fast
[24-Sep-2009 12:43:07] <kisielk> hm
[24-Sep-2009 12:43:13] <rmatte> is it only certain pages that are slow? such as the zproperties page?
[24-Sep-2009 12:43:26] <rmatte> is it digging in to swap?
[24-Sep-2009 12:43:42] <kisielk> It's a quad core AMD 64
[24-Sep-2009 12:43:50] <kisielk> no swap usage
[24-Sep-2009 12:43:52] <rmatte> 3.0GHz?
[24-Sep-2009 12:43:58] <kisielk> naw, 2.2
[24-Sep-2009 12:44:02] <rmatte> k
[24-Sep-2009 12:44:07] <kisielk> the disks are just SATA, in RAID 1
[24-Sep-2009 12:44:07] <rmatte> how many devices?
[24-Sep-2009 12:44:13] <kisielk> about 250
[24-Sep-2009 12:44:29] <rmatte> ok, that should be relatively fine then
[24-Sep-2009 12:44:35] <rmatte> is the box doing anything other than Zenoss?
[24-Sep-2009 12:44:48] <kisielk> no, well, syslog-ng
[24-Sep-2009 12:44:51] <rmatte> that's fine
[24-Sep-2009 12:44:56] <kisielk> but that's it
[24-Sep-2009 12:44:57] <rmatte> can you pastebin your zenpack list?
[24-Sep-2009 12:45:01] <rmatte> just do a zenpack --list
[24-Sep-2009 12:45:11] <rmatte> I just want to check something
[24-Sep-2009 12:45:53] <kisielk> there's only 3
[24-Sep-2009 12:46:08] <kisielk> HPProcurve, HPMonitor, and our own which I think is actually empty
[24-Sep-2009 12:46:47] <rmatte> k
[24-Sep-2009 12:47:03] <rmatte> well, then I'd just boil it down to performance tweaks need to be made
[24-Sep-2009 12:47:12] <kisielk> alright, I will try the ones on the page
\n[24-Sep-2009 12:47:23] <rmatte> I have a Zenoss server monitoring\n340 devices, and it was fairly slow until I did some of the tweaks in\nthat doc
[24-Sep-2009 12:47:38] <rmatte> you'll at least want 3 zenhub workers and some daemons running parallel
[24-Sep-2009 12:47:52] <rmatte> and maybe some slight zopedb tweaking (I wouldn't go with values as high as in the doc)
[24-Sep-2009 12:48:25] <rmatte> Also, couldn't hurt to upgrade that processor
[24-Sep-2009 12:48:44] <rmatte> you can pick up a 2.9GHz phenom for like $80 or $90 these days
[24-Sep-2009 12:50:20] <rmatte> actually, that's the athlon 2
[24-Sep-2009 12:50:31] <rmatte> the phenoms are still quite pricy, thought they would have gone down by now
[24-Sep-2009 12:50:40] <rmatte> I have a 2.5GHz phenom in my PC at home
[24-Sep-2009 12:50:59] <Brad_K> wouldn't you want a Barthelona instead? ;-)
[24-Sep-2009 12:51:00] <rmatte> a 3.4GHz phenom costs $289.99
[24-Sep-2009 12:51:12] <rmatte> meh
[24-Sep-2009 12:51:37] <rmatte> quad-core is quad-core
[24-Sep-2009 12:51:59] <rmatte> plus barcelona is AM3, and he may only have an AM2 board
[24-Sep-2009 12:52:00] <rmatte>
[24-Sep-2009 12:52:38] <cwj> i think amd just announced some new quad core that runs on AM2 and AM3
[24-Sep-2009 12:52:45] <cwj> i forget the details though
[24-Sep-2009 12:52:48] <Brad_K> ahh, but does he have enough rain?
[24-Sep-2009 12:52:55] <cwj> my friend was yacking at me about it
[24-Sep-2009 12:53:32] <rmatte> yeh, that's the Barcelona I believe (so maybe it does support AM2)
[24-Sep-2009 12:54:01] <rmatte> Brad_K: you'd probably need a box fan to cool it
[24-Sep-2009 12:54:03] <kisielk> well, if we were upgrading, we'd swap out the whole machine
[24-Sep-2009 12:54:21] <kisielk> so that's probably about $1k worth of hardware, if not 2
[24-Sep-2009 12:54:27] <Brad_K> or move it to Spain?
[24-Sep-2009 12:54:28] <rmatte> My 2.5 phenom runs so hot compared to my old 2.5GHz Athlon II
[24-Sep-2009 12:54:35] <kisielk> I'd probably put it on SAS disks
[24-Sep-2009 12:54:46] <rmatte> the old chip idled at about 36* C, the new one idles at like 46*C
[24-Sep-2009 12:54:51] <rmatte> 10 degree difference
[24-Sep-2009 12:54:54] * Brad_K enjoys being silly sometimes
[24-Sep-2009 12:55:43] <rmatte>
[24-Sep-2009 12:56:09] <rmatte> and that's with a huge copper heatsink on it
[24-Sep-2009 12:56:30] <Brad_K> you put a huge copper heatsink on your tongue?
[24-Sep-2009 12:56:36] <rmatte>
[24-Sep-2009 12:56:49] <rmatte> http://dmon.org/graphics/nemesis/IMG_0131.JPG <--- This
\n[24-Sep-2009 12:57:10] <rmatte> (My graphics card is a lot bigger\nnow than the one that was there in that photo too, so less airflow in\nthe case)
[24-Sep-2009 12:57:42] <rmatte> http://dmon.org/graphics/nemesis/IMG_0130.JPG
[24-Sep-2009 12:57:45] <rmatte> better photo
[24-Sep-2009 13:04:16] <twm1010> Anyone using the II6 Zenpack?
[24-Sep-2009 13:06:15] <twm1010> what would -1,* do in the RPN field of a datapoint in a graph
[24-Sep-2009 13:14:35] <rmatte> multiply whatever is after it by negative 1
[24-Sep-2009 13:14:42] <rmatte> RPN is reverse
[24-Sep-2009 13:15:14] <twm1010> odd, why would the bytes received datapoint be set that way
[24-Sep-2009 13:15:15] <rmatte> start from the right and read to the left
[24-Sep-2009 13:15:20] <twm1010> its a derive type
[24-Sep-2009 13:15:44] <twm1010> the bytes sent isn't manipulated at all
[24-Sep-2009 13:15:47] * rmatte shrugs
[24-Sep-2009 13:15:58] <twm1010> hrm... no matter, zenpack works nicely
[24-Sep-2009 13:31:08] <kisielk> rmatte: hm, so even after applying all those tweaks, it's still pretty slow :/
[24-Sep-2009 13:31:19] <kisielk> I just clicked a host in the dashboard, took over 30 seconds to go to the page
\n[24-Sep-2009 13:31:36] <kisielk> weird thing is, I didn't see\nanything thrashing on the system in iostat, or a lot of activity with\ntop..
[24-Sep-2009 13:31:53] <kisielk> so I don't know where the slowness is coming from
[24-Sep-2009 13:33:11] <chudler> curious. how large is the Data.fs file?
[24-Sep-2009 13:37:54] <rmatte> $ZENHOME/var/Data.fs
[24-Sep-2009 13:58:50] <twm1010> mine is 288947716
[24-Sep-2009 13:59:40] <twm1010> that's 288MB or so, right?
[24-Sep-2009 14:29:34] <rmatte> correct
[24-Sep-2009 14:30:31] <rmatte> more like 282MB but yeh
[24-Sep-2009 14:30:39] <rmatte> you can always do an ls -alh
[24-Sep-2009 14:30:51] <rmatte> to have it convert it to kilobytes, megabytes, or gigabytes for you
[24-Sep-2009 14:30:53] <rmatte>
[24-Sep-2009 14:32:13] <kisielk> chudler: 77 MB
[24-Sep-2009 14:32:33] <chudler> nice and small... no problem there I suppose.
[24-Sep-2009 14:32:55] <twm1010> isn't there a routine you can run through to condense that?
[24-Sep-2009 14:33:35] <chudler> there is... somewhere in the zenoss site :-)
[24-Sep-2009 15:32:40] <chemist> evening
\n[24-Sep-2009 15:34:33] <chemist> rmatte:\nhttp://dev.zenoss.com/trac/ticket/5578 after the roasting I got\nthe other night it looks like I will have to find a plugin to monitor\nthe switches
[24-Sep-2009 15:37:07] <rmatte> lol
[24-Sep-2009 15:37:30] <chemist> any suggestions?
[24-Sep-2009 15:37:37] <chemist> as to what to use
[24-Sep-2009 15:37:48] <rmatte> I'd speak with Matt about that one, they are brushing any tickets that they can aside
[24-Sep-2009 15:38:06] <rmatte> since they want to work on the more critical issues before launch
[24-Sep-2009 15:38:13] <rmatte> err, before release rather
\n[24-Sep-2009 15:38:56] <rmatte> there really should be something\nin that plugin that is capable of continuing on without requiring those\nvalues
[24-Sep-2009 15:39:23] <rmatte> especially since this is fairly common (according to the forums)
[24-Sep-2009 15:40:13] <rmatte> I'll speak with him about it tomorrow if I have a second and he's not too busy
\n[24-Sep-2009 15:41:23] <chemist> my colleague hacked the script\nand it now models the devices and shows graphs but the problem lies in\nnot been able to label the switches
[24-Sep-2009 15:41:38] <rmatte> label the ports you mean?
[24-Sep-2009 15:41:42] <chemist> yes
[24-Sep-2009 15:41:47] <rmatte> do they have labels on them?
[24-Sep-2009 15:42:13] <rmatte> and if you do an snmpwalk, are you able to see the labels under a certain OID?
\n[24-Sep-2009 15:53:11] <chemist> no, but I would be happy editing\nthe names myself rather than having a name of 'Ethernet Interface' for\nall the interfaces
[24-Sep-2009 15:54:17]\n<chemist> when I then click on 'Ethernet Interface' I can see\n/Devices /Network /Switch /switch1.icarus.xxxxx.co.uk /os \n/Ethernet Interface_21
[24-Sep-2009 15:54:26]\n<rmatte> ah, well they are really meant to reflect interface\ndescriptions that are set on the device itself
[24-Sep-2009 15:54:26] <chemist> or Interface_7
[24-Sep-2009 15:54:28] <chemist> etc
[24-Sep-2009 15:54:29] <chemist> etc
[24-Sep-2009 15:54:46] <rmatte> oh
[24-Sep-2009 15:54:48] <chemist> and they are
[24-Sep-2009 15:54:54] <rmatte> like every interface is called Ethernet Interface?
[24-Sep-2009 15:55:06] <kisielk> does the switch not support naming the interfaces?
[24-Sep-2009 15:55:08] <rmatte> there's no like port 1, port 2, etc...?
[24-Sep-2009 15:55:18] <rmatte> kisielk: he has really cheap switches
[24-Sep-2009 15:55:21] <rmatte> netgear
[24-Sep-2009 15:55:25] <chemist> yes, but zenoss wont pick the names up
[24-Sep-2009 15:55:38] <chemist>
\n[24-Sep-2009 15:55:52] <rmatte> chemist: that's more of a problem\nwith the snmp on the switches than anything, Zenoss is not designed to\nlet people manually name interfaces
[24-Sep-2009 15:56:14] <rmatte> quite honestly, it's insane that you'd even have to do something like that
[24-Sep-2009 15:56:26] <rmatte> useless for monitoring too
[24-Sep-2009 15:56:31] <rmatte> "Ethernet Interface is down"
[24-Sep-2009 15:56:35] <rmatte> "Which one!?!?!?!"
[24-Sep-2009 15:56:42] <chemist> tell me about it
[24-Sep-2009 15:56:54] <rmatte> but anyways, you get what you pay for
[24-Sep-2009 15:57:02] <chemist> my boss is no hurry to replace 80+ switches bought in december 08
[24-Sep-2009 15:57:04] <rmatte> those netgears are the sort of thing you'd throw in a boardroom
[24-Sep-2009 15:57:05] <rmatte> lol
[24-Sep-2009 15:57:18] <kisielk> I wouldn't put one in my boardroom
[24-Sep-2009 15:57:22] <kisielk> it's only 10/100
[24-Sep-2009 15:57:33] <rmatte> meh, 100 is fast enough for laptops
[24-Sep-2009 15:57:43] <chemist> with gigabit uplinks I'll have you know
[24-Sep-2009 15:57:44] <rmatte> I mean, how much porn can you possibly download during a meeting?
[24-Sep-2009 15:57:56] <rmatte>
[24-Sep-2009 15:58:05] <kisielk> rmatte: you'd be surprised
[24-Sep-2009 15:58:08] <rmatte> hehe
\n[24-Sep-2009 15:58:29] <rmatte> either way, if you can't see the\ninfo in an snmpwalk then there's no way that Zenoss is going to be able\nto properly model those interfaces
[24-Sep-2009 15:58:37] <rmatte> so in short, you're pretty much screwed
[24-Sep-2009 15:58:45] <rmatte>
[24-Sep-2009 15:59:07] <chemist> but snmpwalking the device shows us at least an index number
[24-Sep-2009 15:59:17] <rmatte> you can monitor for traps, and poll for CPU, memory, and such, but that's pretty much it
[24-Sep-2009 15:59:39] <rmatte> yeh, index number is good and dandy, but Zenoss looks at the port name
[24-Sep-2009 15:59:49] <rmatte> since on every other switch that I've seen the port names actually make sense
[24-Sep-2009 15:59:51] <chemist> ok
[24-Sep-2009 16:00:02] <rmatte> when you login to a cisco device you don't want to know that port index 3 went down
[24-Sep-2009 16:00:12] <rmatte> you want to know that Fa0/3 went down
[24-Sep-2009 16:00:13] <rmatte> or whatever
[24-Sep-2009 16:00:18] <rmatte> hence why it's designed like that
[24-Sep-2009 16:00:27] <chemist> makes sense
[24-Sep-2009 16:00:53] <rmatte> or Se0/3/0
[24-Sep-2009 16:01:00] <rmatte> whatever it happens to be
[24-Sep-2009 16:02:02]\n<rmatte> If you have someone there who is good at python maybe\nthey can modify that plugin and set it up as a new collector plugin\njust for the neatgears
[24-Sep-2009 16:02:10] <rmatte> and have it use the interface number is place of the name
[24-Sep-2009 16:02:16] <rmatte> probably wouldn't be a massive change
[24-Sep-2009 16:02:34] <rmatte> in place of the name*
[24-Sep-2009 16:02:35] <rmatte> rather
[24-Sep-2009 16:03:29] <rmatte> well, I'm 1 hour and 57 minutes to being $870 poorer
[24-Sep-2009 16:03:43] * rmatte taps his foot
[24-Sep-2009 16:03:57]\n<chemist> ok, and this might be quite crude, but would it be\npossible to go into the zodb and manually enter descriptions?
[24-Sep-2009 16:04:04] <chemist> how come?
[24-Sep-2009 16:04:20] <rmatte> paintball marker auction on ebay
[24-Sep-2009 16:04:29] <rmatte> and yes, you could do that via zendmd
[24-Sep-2009 16:04:49] <chemist> hehe
[24-Sep-2009 16:04:55] <rmatte> actually, you could write a script to have it run through and increment the value each time
[24-Sep-2009 16:05:21] <rmatte> and do it on each device
[24-Sep-2009 16:05:30] <rmatte> if you can wait until tomorrow I might be able to cook it up for you
[24-Sep-2009 16:05:38] <rmatte> they are all ethernet interfaces right?
[24-Sep-2009 16:05:51] <chemist> no worries, it would be greatly appreciated
[24-Sep-2009 16:05:56] <chemist> yes
[24-Sep-2009 16:06:09] <rmatte> ports 1 to whatever on each?
[24-Sep-2009 16:06:15] <rmatte> or is there a port 0?
[24-Sep-2009 16:06:32] <chemist> no, 1 to 26
[24-Sep-2009 16:06:55] <rmatte> k
[24-Sep-2009 16:07:08] <rmatte> It'll probably only take like 7 or 8 lines of code to do
[24-Sep-2009 16:08:10]\n<rmatte> just have to have it loop through the interfaces on each\ndevice and name them whatever with the number attached and increment\nthe number each time, then reset the number each time it hits the next\ndevice
[24-Sep-2009 16:08:31] <rmatte> shouldn't be difficult
[24-Sep-2009 16:08:41] <rmatte> my brain is fried for today though, so not in the mood for scripting
\n[24-Sep-2009 16:12:34] <chemist> that would be great, the\nswitches are in around 40 different locations and they 'should' all\nhave the same devices plugged in to each port so just a number would\ngive us a clue as to what's what
[24-Sep-2009 16:13:12] <rmatte> ok, so you just want them named 1, 2, 3, 4, etc...?
[24-Sep-2009 16:13:23] <chemist> yes
[24-Sep-2009 16:22:59] <rmatte> finished
[24-Sep-2009 16:23:01] <rmatte> try this...
[24-Sep-2009 16:23:14] <rmatte> http://pastebin.com/m4f9dd4ba
[24-Sep-2009 16:23:22] <rmatte> paste that in to zendmd
[24-Sep-2009 16:24:20] <rmatte> (I was bored, decided to script it)
[24-Sep-2009 16:24:29] <rmatte> just tested it and it appears to work nicely
[24-Sep-2009 16:24:55] <rmatte> hopefully it doesn't screw up the interface graphing by changing those values
[24-Sep-2009 16:25:41] <rmatte> that'll apply to every single device
\n[24-Sep-2009 16:25:58] <rmatte> so if you have other devices in\nthere you might just want to remove them and add them back later
[24-Sep-2009 16:26:37] <rmatte> or you can do it to an individual device with http://pastebin.com/m70d39e1e
[24-Sep-2009 16:26:45] <rmatte> replace 30.30.254.33 with an ip or a devicename
[24-Sep-2009 16:27:03] <rmatte> oh crap wait
[24-Sep-2009 16:27:13] <rmatte> that'll make one of the ports port 0
[24-Sep-2009 16:27:16] <rmatte> forgot about that
[24-Sep-2009 16:27:17] <rmatte> one sec
[24-Sep-2009 16:27:47] <rmatte> there
[24-Sep-2009 16:27:48] <rmatte> http://pastebin.com/m7e56df87
[24-Sep-2009 16:27:55] <rmatte> top for single, bottom for all
[24-Sep-2009 16:29:44] <rmatte> I'd recommend going through and locking down the interfaces on all of the devices after
[24-Sep-2009 16:29:58] <rmatte> because it'll change them when it remodels if they aren't locked
\n[24-Sep-2009 16:30:22] <rmatte> the other issue you might run in\nto is that it decides to add the other interfaces in in addition to\nthese once they are renamed
[24-Sep-2009 16:30:28] <rmatte> I'd just lock all of the devices down
[24-Sep-2009 16:30:55] <rmatte> From the device list select all and click on Lock devices...
[24-Sep-2009 16:31:05] <rmatte> then select Lock from deletion and updates
[24-Sep-2009 16:32:10] <chemist> do I have to enter anything at the end?
[24-Sep-2009 16:32:18] <rmatte> just hit enter at the end
[24-Sep-2009 16:32:19] <rmatte> oh
[24-Sep-2009 16:32:26] <rmatte> in zendmd you'll have to type commit()
[24-Sep-2009 16:32:27] <rmatte> afterwards
[24-Sep-2009 16:32:31] <chemist> ok
[24-Sep-2009 16:32:37] <rmatte> since zendmd only simulates stuff until you commit it
[24-Sep-2009 16:34:15] <rmatte> 5 lines of code, shorter than I thought
\n[24-Sep-2009 16:34:30] <rmatte> (then again I was considering\nsome lines to make it a standalone script, but there's really no need)
[24-Sep-2009 16:34:53] <chemist> http://pastebin.com/mdb6d85f
[24-Sep-2009 16:34:58] <chemist> this is what I get
[24-Sep-2009 16:35:18] <rmatte> yeh, you're doing it wrong lol
[24-Sep-2009 16:35:27] <rmatte> ok, first off, you're not using the modified version of the script
[24-Sep-2009 16:35:49] <rmatte> http://pastebin.com/m7e56df87
[24-Sep-2009 16:35:55] <rmatte> use the script at the bottom there
[24-Sep-2009 16:36:03] <rmatte> otherwise you'll have interface 0 instead of interface 1
[24-Sep-2009 16:36:13] <rmatte> secondly, what you want to do is paste the script, then hit enter
[24-Sep-2009 16:36:16] <rmatte> wait for it to run through
[24-Sep-2009 16:36:24] <rmatte> once it has run through then you type commit()
[24-Sep-2009 16:36:32] <rmatte> you tried to type commit() as part of the script itself
[24-Sep-2009 16:36:48] <rmatte> when you just see ...
[24-Sep-2009 16:36:52] <rmatte> after pasting the script
[24-Sep-2009 16:36:53] <rmatte> hit enter
[24-Sep-2009 16:38:06] <rmatte> or use the script at the top of you're doing it to a specific device obviously
[24-Sep-2009 16:39:21] <rmatte> did it work?
[24-Sep-2009 16:40:21] <rmatte> (I'm leaving at a little before 6 btw)
[24-Sep-2009 16:40:58] <chemist> err, no
[24-Sep-2009 16:41:05] <rmatte> oh?
[24-Sep-2009 16:41:07] <chemist> and it's hosed tha
[24-Sep-2009 16:41:13] <chemist> that device
[24-Sep-2009 16:41:21] <rmatte> but the command did run through right?
[24-Sep-2009 16:41:46] <chemist> yes
[24-Sep-2009 16:41:57] <rmatte> oh, you know what? the device links to the interfaces by name
[24-Sep-2009 16:42:03] <rmatte> so by changing the names we screw up the links
[24-Sep-2009 16:42:06] <rmatte> so yeh, this won't work
[24-Sep-2009 16:42:20] <rmatte> and I honestly can't think of anything that will because of that, hmmm
[24-Sep-2009 16:42:25] <rmatte> I'll have to ask around tomorrow
[24-Sep-2009 16:42:37] <chemist> thanks anyway
[24-Sep-2009 16:42:39] <chemist>
[24-Sep-2009 16:42:40] <rmatte> np
[24-Sep-2009 16:43:13] <rmatte> I might just be changing the wrong value, I'll have to check it out tomorrow
[24-Sep-2009 16:43:22] <chemist> that's cool
[24-Sep-2009 16:45:09] <rmatte> anyways, have a good night
[24-Sep-2009 16:45:18] <chemist> cheers
[24-Sep-2009 18:36:18] <kisielk> this slowness is unbearable :/
[24-Sep-2009 18:48:39] <davetoo> nuh?
[24-Sep-2009 18:55:01] <kisielk> well, still trying to figure out why my Zenoss is so damn slow
[24-Sep-2009 19:20:40] <leafnunes> kisielk: i don't suppose the host is under load?
[25-Sep-2009 00:00:46] [disconnected at Fri Sep 25 00:00:46 2009]
[25-Sep-2009 00:00:46] [connected at Fri Sep 25 00:00:46 2009]
[25-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[25-Sep-2009 04:40:24] <ilejn> hello
[25-Sep-2009 06:41:39] <ilejn> hello
[25-Sep-2009 08:29:04]\n<hackeron> hey, I have 1 filesystem that's quite large and zenoss\nsends me messages like: [zenoss] 127.0.0.1 disk space threshold: 92.4%\nused (347.5GB free)
[25-Sep-2009 08:29:20] <hackeron> anyway to stop it from sending those for 1 specific partition?
[25-Sep-2009 08:29:42] <hackeron> seems kinda silly, 350GB free is not almost full
[25-Sep-2009 08:30:11] <hackeron> I know the FS performance slows down a lot, but it's non critical storage
[25-Sep-2009 08:34:47] <twm1010> morning
[25-Sep-2009 09:19:30] <rmatte> hackeron: yes
\n[25-Sep-2009 09:19:45] <rmatte> hackeron: just go to templates\nfor that drive, create a local copy, and change the threshold value for\nit
[25-Sep-2009 09:19:48] <rmatte> quite simple
[25-Sep-2009 09:20:00] <rmatte> bump it up to like 95 or 97%
[25-Sep-2009 09:49:42] <hackeron> rmatte: cool, thanks, will try
[25-Sep-2009 10:00:18] <rmatte> g'day cgibbons
[25-Sep-2009 10:00:19] <jb> cgibbons!
[25-Sep-2009 10:00:24] <jb> just the man I was looking for.
[25-Sep-2009 10:00:28] <rmatte> oh oh
[25-Sep-2009 10:00:31] <rmatte> you're in trouble now
[25-Sep-2009 10:00:32] <cgibbons> whatever it is, i didn't check it in!
[25-Sep-2009 10:00:37] <rmatte> lol
[25-Sep-2009 10:00:51] <jb> got a winperf related question again :/
[25-Sep-2009 10:01:41] <jb> i have one performance object in which every counter generaters an error
[25-Sep-2009 10:01:56] <cgibbons> ooh, which one?
[25-Sep-2009 10:02:24] <jb> http://pastebin.com/m2bfe1190
[25-Sep-2009 10:02:35] <jb> its "ISA Server Web Proxy"
[25-Sep-2009 10:02:49] <jb> #
[25-Sep-2009 10:02:50] <jb> ERROR:zen.winperf.PerfRpc:global name 'DEBUG' is not defined
[25-Sep-2009 10:02:54] <cgibbons> argh that bug
[25-Sep-2009 10:03:22] <jb> oh, so its known
[25-Sep-2009 10:03:26]\n<cgibbons> that one slipped through, if you turn on debug you get\na crash independent of whatever is going on. can you go to line 244 of\n/opt/zenoss/ZenPacks/ZenPacks.zenoss.ZenWinPerf-2.1.5-py2.4.egg/ZenPacks/zenoss/ZenWinPerf/zenwinperf.py\nand make a quick change?
[25-Sep-2009 10:03:35] <cgibbons> (it's fixed in a newer version of the zenpack, wherever that might be)
[25-Sep-2009 10:03:36] <jb> sure
[25-Sep-2009 10:03:53] <cgibbons> where it say <= DEBUG change it to <= logging.DEBUG
[25-Sep-2009 10:04:01] <ckrough> morning
[25-Sep-2009 10:04:24] ckrough is now known as Guest58427
[25-Sep-2009 10:04:25] <jb> 244 = values = driver.next()
[25-Sep-2009 10:05:00] <cgibbons> hrm maybe 398?
[25-Sep-2009 10:05:04]\n<jb>\nzenoss@fc-zenoss01:/opt/zenoss/ZenPacks/ZenPacks.zenoss.ZenWinPerf-2.1.5-py2.4.egg/ZenPacks/zenoss/ZenWinPerf$\ngrep DEBUG zenwinperf.py
[25-Sep-2009 10:05:04] <jb> from pysamba.library import DEBUGLEVEL
[25-Sep-2009 10:05:04] <jb> DEBUGLEVEL.value = 99
[25-Sep-2009 10:05:04] <jb> if self.log.isEnabledFor(logging.DEBUG):
[25-Sep-2009 10:05:05] <cgibbons> prolly 398
[25-Sep-2009 10:05:09] Guest58427 is now known as ckrough
[25-Sep-2009 10:05:23] <cgibbons> should look like: if self.logLevel <= DEBUG:
[25-Sep-2009 10:05:24] <cgibbons> somewhere
[25-Sep-2009 10:05:35] <ckrough> isnt there a core zenpack download for 2.4.5? I dont see it listed on the download page
[25-Sep-2009 10:05:43] <jb> hmm
[25-Sep-2009 10:05:52] <jb> if self.log.isEnabledFor(logging.DEBUG):
[25-Sep-2009 10:05:58] <jb> thats 267
[25-Sep-2009 10:05:59] <cgibbons> maybe just search for <= DEBUG in the file, you should come across it
[25-Sep-2009 10:06:12] <jb> nothing matching that
[25-Sep-2009 10:06:15] <cgibbons> huh
[25-Sep-2009 10:06:23] <cgibbons> lemme see if i've got that version handy
[25-Sep-2009 10:06:44] <cgibbons> you're on 2.4.5 of the main product yet ?
[25-Sep-2009 10:06:49] <jb> yep
[25-Sep-2009 10:08:22] <cgibbons> oh no wonder
[25-Sep-2009 10:08:23] <cgibbons> ahem
[25-Sep-2009 10:08:47] <cgibbons> try PerfRpc.py instead of zenwinperf.py
[25-Sep-2009 10:08:52] <cgibbons> and let me get some more coffee....
[25-Sep-2009 10:09:03] <ckrough> Anyone know where zenoss-core-zenpacks-2.4.5 lives on the website?
[25-Sep-2009 10:09:12] <rmatte> hmmm, should we be worried that the new king crab beta version code ends in -404?
[25-Sep-2009 10:09:19] <rmatte> 404 = error
[25-Sep-2009 10:09:21] <rmatte> :O
[25-Sep-2009 10:12:12] <rmatte> ckrough: is that a zenpack with all the core zenpacks bundled in to it?
[25-Sep-2009 10:12:46] <ckrough> rmatte: yeah, I found it on sourceforge
[25-Sep-2009 10:12:50] <ckrough> its missing on the community site
[25-Sep-2009 10:12:56] <ckrough> v2.4.5 that is
[25-Sep-2009 10:13:09] <rmatte> ah
[25-Sep-2009 10:13:14] <rmatte> mention it to Matt
[25-Sep-2009 10:13:18] <ckrough> ya ya
[25-Sep-2009 10:13:39] <rmatte> I'll be happy when he managed to get all of the ZenPacks unzipified on the new site
\n[25-Sep-2009 10:13:58] <rmatte> there are constant posts: "I\ncan't install this pack" ... "You have to unzip it first" ... "Oh!"
[25-Sep-2009 10:14:08] <rmatte> manages*
[25-Sep-2009 10:14:29] <rmatte> well, time to test upgrading between beta versions
[25-Sep-2009 10:15:48] <cgibbons> it should definitely be in that one file, jb
[25-Sep-2009 10:27:47] <rmatte> ooooh, I love the new event console layout
[25-Sep-2009 10:27:54] <rmatte> (the beta 3 one)
\n[25-Sep-2009 10:29:12] <cgibbons> i can't wait until the rest of\nthe UI gets the same treatment. we spent a lot of time on the screen\ndesign so it'll be nice to see it done.
[25-Sep-2009\n10:29:22] <rmatte> hmmm, might have to speak with Ian about the\nway historic events are handled though, can't select a range properly\nanymore
[25-Sep-2009 10:29:54] <rmatte> like the filters for first time and last time don't work as a range
[25-Sep-2009 10:30:02] <rmatte> so there's no way to see events older than a day
\n[25-Sep-2009 10:30:56] <rmatte> the filters for first time and\nlast time are actually pretty useless right now, since they have to\nmatch a specific time and date to pair up
[25-Sep-2009 10:31:17] <rmatte> hmmm, you can use wildcards though, but that's not really good enough
[25-Sep-2009 10:31:23] <rmatte> there need to be a place to specify a range
[25-Sep-2009 10:31:24] <cgibbons> i've passed that on to him, so he might show up
[25-Sep-2009 10:31:26] <rmatte> needs*
[25-Sep-2009 10:31:30] <rmatte> k cool
[25-Sep-2009 10:31:32] <rmatte> thanks
[25-Sep-2009 10:31:57] <rmatte> oooooh, the new severity colors are very nice
[25-Sep-2009 10:32:09] <rmatte> severity row colors
[25-Sep-2009 10:32:38] <rmatte> looks like the gun metal colors that I suggested but a bit more pastelized
[25-Sep-2009 10:36:06] <jb> cgibbons: sorry, had to step away for a minutue
[25-Sep-2009 10:42:20] <rmatte> yeh, the event console is wayyyy less buggy in the new beta
[25-Sep-2009 10:42:27] <rmatte> it's actually usable
\n[25-Sep-2009 10:44:50] <rmatte> the only thing to add which would\nmake it perfect are fields to select an actual range at the top, if Ian\npops on I can explain how I envision it
[25-Sep-2009 10:47:35] <rmatte> hmmm, just found a bug in it
[25-Sep-2009 10:49:17] <rmatte> actually weird, now it's working fine
[25-Sep-2009 10:50:12] <rmatte> ooooh, they cleaned up the details too with alternating colors for the rows
\n[25-Sep-2009 10:51:19] <rmatte> but they still haven't added a\nseparate section for trap details, they just get bunched in to the rest\nof the details
[25-Sep-2009 10:51:22] <rmatte> that needs to be fixed
[25-Sep-2009 10:52:33] <rmatte> other than those 2 things it's looking really good
\n[25-Sep-2009 10:55:21] <eidolon> hi folks - zenoss 2.4.5 new\ninstallation here. looks like it's all started - i can telnet\nlocalhost 8080 and get an answer, but remote connects are failing. i'\nmguessing zope is not bound to eth0 - how can i check this?
[25-Sep-2009 10:56:58] <jb> cgibbons: still here?
[25-Sep-2009 10:57:04] <cgibbons> yup
[25-Sep-2009 10:57:10] <jb> # first learn how big the data is
[25-Sep-2009 10:57:10] <jb> if self.logLevel <= logging.DEBUG:
[25-Sep-2009 10:57:11] <jb> self.log.debug("Learning counter object ids")
[25-Sep-2009 10:57:12] <jb> line 60
[25-Sep-2009 10:57:15] <jb> of PerfRPC.py
[25-Sep-2009 10:57:21] <jb> PerfRpc.py rather
[25-Sep-2009 10:57:36] <jb> and line 80
[25-Sep-2009 10:58:14] <cgibbons> let's also double check line 398 of winreg.py in that directory
[25-Sep-2009 10:58:29] <cgibbons> i think that might be the one that's messed up
[25-Sep-2009 10:58:38] <jb> hrm, what should I search for?
[25-Sep-2009 10:59:04] <cgibbons> yeah in fact i just double checked in that file here for that version, that one should look like <= DEBUG
[25-Sep-2009 10:59:12] <cgibbons> but needs to be logging.DEBUG
[25-Sep-2009 10:59:16] <jb> found it
[25-Sep-2009 10:59:25] <jb> 398
[25-Sep-2009 10:59:35] <jb> ok, made the change.
\n[25-Sep-2009 10:59:52] <cgibbons> okay, try rerunning zenwinperf\nin debug and now you should at least get some useful info
[25-Sep-2009 11:00:32] <jb> zenwinperf run -v 20 cool?
[25-Sep-2009 11:00:49] <cgibbons> yep. eventually we may need to do a -v 5 but we'll see if 20 workswell enough
[25-Sep-2009 11:01:22] <davetoo> oh is there actually stuff logged at < 10?
[25-Sep-2009 11:01:44] * davetoo is annoyed at the pysamba Polling: messages at 10
[25-Sep-2009 11:01:56]\n<cgibbons> yeah, rarely. i've been doing it in some of the\nlow-level windows daemons because some of that debug is useless for\nsupport but useful for me.
[25-Sep-2009 11:02:13]\n<cgibbons> me too, davetoo, in fact those are gone now. i can\ntell you where to move them if you'd like to do it before KC comes out.
[25-Sep-2009 11:02:19] <rmatte> oh, I thought the higher the number the more verbose
[25-Sep-2009 11:02:22] <rmatte> not the opposite
[25-Sep-2009 11:02:36] <cgibbons> yeah that part is weird
[25-Sep-2009 11:02:37] <davetoo> no, it's actually a threshold for the logging package
[25-Sep-2009 11:03:58] <rmatte> but yeh, 3 changes and the event console is perfect
[25-Sep-2009 11:04:07] <rmatte> (well, 4, but it's a tiny change)
[25-Sep-2009 11:04:53]\n<rmatte> they need a higher res graphic for the details window\n(the half arrow that's directly to the right of the severity icon)
[25-Sep-2009 11:05:05] <rmatte> it looks pixelated while everything else looks nice and smooth, it ruins the look
[25-Sep-2009 11:05:18] <jb> hrm, ok it worked
[25-Sep-2009 11:05:23] <jb> INFO:zen.winperf:Collected value for \\ISA Server Web Proxy\\SSL Client Bytes Received/sec: 0.0
[25-Sep-2009 11:06:18] <jb> thats odd, because nothing changed
[25-Sep-2009 11:06:23] <jb> other than that log statement
[25-Sep-2009 11:06:29] <rmatte> lol
[25-Sep-2009 11:07:09]\n<cgibbons> that area of the code is an edge case, so if you hit\nit (which this counter probably did), it would break collection for it\neven if logging wasn't enabled
[25-Sep-2009 11:07:50] <jb> so you think its fine now?
[25-Sep-2009 11:07:55]\n<davetoo> yes, I just started using the logging package; I wish\nthere was more granularity (by default) between 20 and 0
[25-Sep-2009 11:08:09] <cgibbons> if you're getting the data you want, yep. it'll be officially in the next version of the zenpack.
[25-Sep-2009 11:08:14] <jb> ok nice
[25-Sep-2009 11:08:20] <jb> thanks for the help once again
[25-Sep-2009 11:08:22] <cgibbons> you bet
[25-Sep-2009 11:15:26] <jb> hmm
[25-Sep-2009 11:16:10] <jb> Failure: exceptions.AttributeError: PerfCounterData instance has no attribute 'time'
[25-Sep-2009 11:16:12] <jb> seen that one?
[25-Sep-2009 11:17:40] <cgibbons> hurm nope, lesee
[25-Sep-2009 11:17:52] <jb> can I not redirect output of zenwinperf to a file?
[25-Sep-2009 11:17:57] <cgibbons> sure
\n[25-Sep-2009 11:18:13] <cgibbons> if you want, you can capture\nthe raw data to some file(s) and send them to me and i can run it\nthrough code here and see what it is doing
[25-Sep-2009 11:18:20] <jb> sure
[25-Sep-2009 11:18:36] <cgibbons> use the --captureFile=outputprefix argument
[25-Sep-2009 11:18:42] <jb> one sec.
[25-Sep-2009 11:18:43] <cgibbons> it should make 3 files per device
[25-Sep-2009 11:19:11] <jb> -v 10 ok?
[25-Sep-2009 12:02:13] <ckrough> meh, need to monitor RDP
[25-Sep-2009 12:50:56] <cgibbons> quiet today
[25-Sep-2009 12:52:05] <rhettardo> yes
[25-Sep-2009 12:52:54] <rmatte> I'm just hoping Ian will pop in for a second at some point
[25-Sep-2009 13:47:20] cgibbons_ is now known as cgibbons
[25-Sep-2009 13:51:28] <djack> hellow world
[25-Sep-2009 13:51:52] <djack> somebody alive and kicking here
[25-Sep-2009 13:52:58] <djack> hmmz
[25-Sep-2009 13:54:39] <cain22> hello all
[25-Sep-2009 13:54:50] <djack> hey
[25-Sep-2009 13:55:02] <djack> everybody is a sleep here
\n[25-Sep-2009 13:55:11] <cain22> is there a way to set an event on\ncpu threshold say, if it continues to be over for 5 minutes?
[25-Sep-2009 13:56:51] <djack> it can be done via the counter i thing
[25-Sep-2009 13:57:04] <djack> but how don't know exactly
[25-Sep-2009 13:57:52] <rmatte> you could do it with a transform
[25-Sep-2009 13:58:02] <cain22> counter only esculates the event severity i think
[25-Sep-2009 13:58:04] <ckrough> which WMI Terminal Services zenpack are you guys using, if at all
[25-Sep-2009 13:58:10] <ckrough> seem to be two almost identical
[25-Sep-2009 13:59:12] <cain22> ckrough: I'm not, are they any good?
[25-Sep-2009 13:59:39] <ckrough> dunno, trying to decide which to use now
[25-Sep-2009 14:01:04] <rmatte> ckrough: use Egor's packs
[25-Sep-2009 14:01:23] <ckrough> Egor wrote a terminal service pack? I just was his wmi one
[25-Sep-2009 14:01:38] <rmatte> you mean to count the number of terminal services clients connected?
[25-Sep-2009 14:01:43] <djack> hey rmatte
[25-Sep-2009 14:01:50] <ckrough> yeah
[25-Sep-2009 14:01:57] <rmatte> I was under the impression that his wmi performance template did that
[25-Sep-2009 14:02:05] <rmatte> and if not, you could add the WMI query in in like 2 seconds
[25-Sep-2009 14:02:08] <ckrough> maybe, ill look
[25-Sep-2009 14:02:13] <ckrough> it's a dependency for it anyway
[25-Sep-2009 14:02:28] <rmatte> djack: yes?
[25-Sep-2009 14:05:43] <rmatte> cain22: this is how it'd do it...
[25-Sep-2009 14:06:01] <rmatte> cain22: this is assuming that you don't send alerts on Info events
[25-Sep-2009 14:06:24] <rmatte> cain22: have the threshold come in as an info event, which won't alert
[25-Sep-2009 14:06:46] <rmatte> cain22: then write a transform to set it to whatever severity if it hits a certain count
[25-Sep-2009 14:06:49] <rmatte> so like...
[25-Sep-2009 14:06:58] <rmatte> if evt.count >= 3:
[25-Sep-2009 14:07:09]\n<djack> i hav an output CRITICAL : <datapoints>\n<datapints> i want to make an event on the CRITICAL how.... i\nread some thing on the old site but can't find it again
[25-Sep-2009 14:07:10] <rmatte> evt.severity = 5
[25-Sep-2009 14:08:48] <rmatte> you mean that's what a script is outputting?
[25-Sep-2009 14:09:44] <cain22> and info alerts are cleared if it isn't continuel?
[25-Sep-2009 14:10:07] <cain22> i'm lookin for after a certain amount of time
[25-Sep-2009 14:10:22] <rmatte> it'll eventually clear when the cpu is restored
\n[25-Sep-2009 14:10:25] <djack> i use webinject to check webflow\nwrote a wrapper that gives times but also a status report when there is\nsomething wrong ... that status is CRITICAL : All failed or Warning :\nTest 2 failed ....
[25-Sep-2009 14:10:53] <cain22> rmatte: thanks for the info, i may just have to try that
[25-Sep-2009 14:11:02]\n<rmatte> cain22: well, one polling cycle is 5 minutes, so as soon\nas the count hits 2 you know it's been like that for 5 minutes
[25-Sep-2009 14:11:26] <rmatte> so you could do if evt.count >= 2:
[25-Sep-2009 14:13:06] <cain22> hmm...thanks
[25-Sep-2009 14:13:32] <rmatte> np
[25-Sep-2009 14:22:16] <djack> rmatte do you understand?
[25-Sep-2009 14:24:03] <rmatte> ok, I'm just asking where you'd be pulling that data from
[25-Sep-2009 14:24:07] <rmatte> command based datapoints?
[25-Sep-2009 14:26:16] <djack> i run a script
[25-Sep-2009 14:26:23] <djack> and this is the output
[25-Sep-2009 14:26:35] <djack> WARNING: 1 Tests Failed | CASE 1 = 0.041 CASE 2 = 2.208 TOTAL 2.368
\n[25-Sep-2009 14:27:24] <djack> the CASE 1 and CASE 2 and TOTAL\nwill be mapped in a graph but the WARNING: should be a trigger to add\nan event to that device ....
[25-Sep-2009 14:27:52] <rmatte> ok...
[25-Sep-2009 14:28:08] <rmatte> I've never seen anything that would allow you to do that
[25-Sep-2009 14:28:21] <rmatte> the way I'd do it is instead of having warning I'd just do...
[25-Sep-2009 14:29:07] <rmatte> Test Failed|CASE1=0.041 CASE2=2.208 TOTAL=2.368 STATUS=1
[25-Sep-2009 14:29:14] <rmatte> and have status as either 0, 1, 2, etc...
[25-Sep-2009 14:29:20] <rmatte> then base the severity on that
[25-Sep-2009 14:30:41] <rmatte> datapoints shouldn't have spaces in them as a rule of thumb
[25-Sep-2009 14:30:48] <rmatte> so CASE 1 and CASE 2 isn't good practice
[25-Sep-2009 14:31:03] <rmatte> also, I find that you shouldn't have spaces between the = signs either
[25-Sep-2009 14:31:16] <djack> can be done
[25-Sep-2009 14:32:39] <djack> but there is no way to grep the first part and make an error of it...
[25-Sep-2009 14:33:05] <rmatte> not that I'm aware of, Zenoss' whole Nagios plugin support only seems to look at what's after the |
[25-Sep-2009 14:33:14] <rmatte> from my experience creating scripts for Zenoss
[25-Sep-2009 14:33:49] <rmatte> usually I just put something like OK|CPU1=75 CPU2=46 CPU3=82 CPU4=97
[25-Sep-2009 14:33:59] <rmatte> and then just put "Unknown" if the command fails
[25-Sep-2009 14:33:59] <djack> rmattte can i write event from the script ?
[25-Sep-2009 14:34:38] <rmatte> well, what would you want the event to look like?
[25-Sep-2009 14:34:52] <rmatte> (there's no reason why you can't, just depends on exactly what you expect from it)
\n[25-Sep-2009 14:37:50] <djack> Well basicly i want a warning\nplaced on a device or critical with some information like WARNING Case\n2 Failed <parameter> not found or CRITICAL All cases failed
[25-Sep-2009 14:38:16] <rmatte> and what determines whether 1 case or all cases failed?
\n[25-Sep-2009 14:38:18] <djack> i'm checking a webpage flow i need\nto know and graph the time for every step but also if the content that\ni expect is there
[25-Sep-2009 14:38:37] <djack> the script ....
[25-Sep-2009 14:38:38] <rmatte> (in the output of the script I mean)
[25-Sep-2009 14:38:46] <rmatte> how would Zenoss tell
[25-Sep-2009 14:38:50] <rmatte> you have to think of this logically
\n[25-Sep-2009 14:39:14] <rmatte> you're passing the data off to\nZenoss, how does it know when to generate an event, what type of event\nto generate, what severity to assign to it, and what info to display?
[25-Sep-2009 14:39:32] <djack> i have a value TEST_DONE TEST_OK TEST_NOK
[25-Sep-2009 14:39:54] <rmatte> ok, so that would be replaced by STATUS
[25-Sep-2009 14:40:04] <rmatte> 0, 1, 2
[25-Sep-2009 14:40:34] <djack> 0 = OK 1 = WARNING 2 = CRITICAL
[25-Sep-2009 14:40:41] <rmatte> that works
[25-Sep-2009 14:40:48] <rmatte> ok, and as for what it's displaying
[25-Sep-2009 14:40:55] <rmatte> how would you want the message to show up?
[25-Sep-2009 14:41:06] <djack> how do you mean ?
[25-Sep-2009 14:41:44] <rmatte> like: 'Bleh is really really bad, and CASE 1 is 0.041 and CASE 2 is 2.208'
[25-Sep-2009 14:41:47] <rmatte> or whatever
[25-Sep-2009 14:41:53] <rmatte> how do you want the messages formatted?
[25-Sep-2009 14:42:23] <djack> WARNING Case 1 failed or CRITICAL all caes failed
[25-Sep-2009 14:42:47] <djack> the time for that is not relevant because message here is unrelated from the time
[25-Sep-2009 14:43:01] <rmatte> ok, so basically what you want to do is this then...
[25-Sep-2009 14:43:10] <rmatte> create 2 thresholds for status
[25-Sep-2009 14:43:24] <rmatte> 1 that kicks off if status hits 2
[25-Sep-2009 14:43:28] <rmatte> and 1 that kicks off if it hits 1
[25-Sep-2009 14:43:34] <rmatte> assign appropriate severities to them
[25-Sep-2009 14:43:49] <djack> but no message is possible ?
\n[25-Sep-2009 14:43:51] <rmatte> then use transforms to modify the\nmessage for each when they are generated to whatever you wanty
[25-Sep-2009 14:43:54] <rmatte> want*
[25-Sep-2009 14:44:27] <rmatte> I'll give you an example
[25-Sep-2009 14:45:13] <djack> oki
\n[25-Sep-2009 14:45:27] <rmatte> you'd name them different things,\nso one would be "whatever Warning" the other would be "whatever\nCritical" and we'll just assume you specified event class /Perf/Memory\n(as an example)
[25-Sep-2009 14:45:32] <rmatte> you would go to Events
[25-Sep-2009 14:45:37] <rmatte> navigate to /Perf/Memory
[25-Sep-2009 14:46:07] <rmatte> and as soon as my lab box catches up with me I can continue...
[25-Sep-2009 14:46:11] <rmatte> ok, there we go
[25-Sep-2009 14:46:39] <rmatte> from the dropdown in the top left you'd select More -> Transform
[25-Sep-2009 14:46:58] <rmatte> you'd then add something like this in:
[25-Sep-2009 14:47:54] <djack> found a better way ....
[25-Sep-2009 14:48:04] <djack> via the zenevent ... i can run that from the command line
[25-Sep-2009 14:48:17] <rmatte> explain?
[25-Sep-2009 14:48:33] <djack> docs/DOC-3069
[25-Sep-2009 14:49:05] <djack> thats what i need #1.8.2. Creating Events from the Command Line
[25-Sep-2009 14:49:39] <rmatte> ummm, yes you can do that, but you'll have to set your script up to run itself in cycles
[25-Sep-2009 14:49:46] <rmatte> it's not going to follow Zenoss' monitoring cycles
[25-Sep-2009 14:50:06] <rmatte> hence why I was explaining to do it this way
[25-Sep-2009 14:50:12] <rmatte> anyways
[25-Sep-2009 14:50:13] <rmatte> if evt.summary.startswith('threshold of whatever Warning'):
[25-Sep-2009 14:50:13] <rmatte> evt.summary and evt.message = 'WARNING Case 1 failed'
[25-Sep-2009 14:50:13] <rmatte> if evt.summary.startswith('threshold of whatever Critical'):
[25-Sep-2009 14:50:13] <rmatte> evt.summary and evt.message = 'CRITICAL all Cases failed'
[25-Sep-2009 14:50:18] <rmatte> that's the transform you'd yse
[25-Sep-2009 14:50:23] <rmatte> use*
[25-Sep-2009 14:51:08] <rmatte> up to you which way you want to do it
\n[25-Sep-2009 14:52:16] <djack> not realy i will trigger the\nzenevent from a bash script that gets the time and also the status ...\nwhen a condition is there via IF statement i can run the zenevent ....
[25-Sep-2009 14:52:18] <djack> not ?
[25-Sep-2009 14:53:08] <rmatte> yes, what I'm saying is, how are you going to keep the bash script checking in a cycle?
[25-Sep-2009 14:53:13] <rmatte> are you going to cron it or something?
[25-Sep-2009 14:53:37] <djack> i execute it from a template
[25-Sep-2009 14:53:49] <djack> that i already have checking other things
[25-Sep-2009 14:53:55] <rmatte> oh, to collect performance data, and include that in it
[25-Sep-2009 14:54:00] <rmatte> actually, yeh, that could work
[25-Sep-2009 14:54:00] <djack> yep
[25-Sep-2009 14:54:07] <rmatte> worth a shot
\n[25-Sep-2009 14:54:45] <djack> zenoss is great ;-) all my\nexternal support partners want access to monitor the shit that they\nsold us ;-)
[25-Sep-2009 14:55:04] <rmatte> yeh, zenoss is quite powerful once you get good at developing it
[25-Sep-2009 14:55:19] <djack> haha
\n[25-Sep-2009 14:55:41] <rmatte> we setup IPSLA monitoring for\nsome of our clients (with custom scripts that I wrote) and they are\nsuper impressed because it works better than what they have from this\nother company which they are paying to do just IPSLA graphing
\n[25-Sep-2009 14:55:45] <djack> well i know bash some perl have a\nsmall understanding how zenoss works so it will e a success
[25-Sep-2009 14:56:00] <djack> what does ipsla doe?
\n[25-Sep-2009 14:56:30] <rmatte> basically you configure a core\nrouter (Cisco or whatever) to collect stats for all of the other\nrouters,
[25-Sep-2009 14:56:41] <ckrough> rmatte: are you using ms_wbt_server to monitor terminal services from the outside?
[25-Sep-2009 14:57:24] <djack> hmmm
\n[25-Sep-2009 14:57:29] <rmatte> basically collects round trip\ntime, packet loss dest-source, packet loss source-dest, and jitter
[25-Sep-2009 14:57:40] <rmatte> and then we pull the stats from the core router via SNMP and graph it all
[25-Sep-2009 14:58:02] <rmatte> ckrough: you mean the IpService?
\n[25-Sep-2009 14:58:36] <djack> routers are out our management but\nin the management of our telco and they will not give snmp management
[25-Sep-2009 14:59:19] <rmatte> well, if it's "managed" equipment then they are doing the management for you
\n[25-Sep-2009 14:59:39] <rmatte> the alternative would be to buy\nyour own routers and just connect the internet circuits to that
[25-Sep-2009 15:00:06] <ckrough> rmatte: yeah
[25-Sep-2009 15:01:29] <rmatte> ckrough: well, we monitor some of our terminal servers in a data center from our lab here
[25-Sep-2009 15:01:35] <rmatte> ckrough: why do you ask?
[25-Sep-2009 15:01:57] <ckrough> sec
[25-Sep-2009 15:03:02]\n<ckrough> trying to halp a buddy out, his users complain of rdp\navailability, so we are looking to monitor outside availability (3389)\nand are using wmi to monitoring session information
[25-Sep-2009 15:03:17] <ckrough> rmatte: any suggestions
[25-Sep-2009 15:04:37] <rmatte> the IpService ms-wbt-server monitors port 3389
[25-Sep-2009 15:04:48] <ckrough> rmatte: thats what I using
[25-Sep-2009 15:04:54] <rmatte> ok, so what's the issue?
[25-Sep-2009 15:05:01] <ckrough> does it just check that 3389 is open, or does it do any kind of quizzing of the service
[25-Sep-2009 15:05:16] <ckrough> I dont get what all of the options are for when you look inside that service
[25-Sep-2009 15:05:28] <rmatte> let me check, haven't looked in a while
[25-Sep-2009 15:06:10]\n<ckrough> there is an "Ip Addresses" field, and a string/regex\npair. I want to make sure the service is actually responding, so if you\nare using something in the send string/regex I'd love to steal it
[25-Sep-2009 15:07:36] <rmatte> nah, I'm using default settings, it looks like it just checks if the port is open
[25-Sep-2009 15:07:42] <rmatte> what I'd suggest is this...
[25-Sep-2009 15:08:05]\n<rmatte> continue monitoring via that IpService, monitor the\nsessions as you've been doing, but monitor the service itself via WMI\nas well
[25-Sep-2009 15:08:22] <ckrough> ok. thanks. learning how to interface with WMI now
[25-Sep-2009 15:08:29] <ckrough> for monitoring processes/services
[25-Sep-2009 15:08:40] <rmatte> the service monitoring is easy in Zenoss, read up on it, you'll see
[25-Sep-2009 15:09:01] <ckrough> thanks
[25-Sep-2009 15:09:05] <rmatte> np
[25-Sep-2009 15:23:08] <cgibbons> pouf!
[25-Sep-2009 15:24:16] <rmatte> bah, I win an auction on ebay and the guy is like "Oh sorry, I already sold that locally"
[25-Sep-2009 15:24:21] <rmatte> grrr, I hate people sometimes
[25-Sep-2009 15:24:43] <rmatte> I'd open a case with ebay about it but I'm not in a mean mood today
[25-Sep-2009 15:26:34] <cwj> kill 'em all and let dobbs sort 'em out
[25-Sep-2009 15:27:18] <ckrough> How can I access a WMI datapoing in a graph report?
[25-Sep-2009 15:27:39] <djack> rmatte zensendevent seems to work now how to clear the event ;-)
\n[25-Sep-2009 15:28:47] <rmatte> ckrough: the same way you'd\naccess an SNMP datapoint, there's no difference in the way they store\ndata
[25-Sep-2009 15:31:08] <ckrough> Not clear on what I would put for the DataPoint name. Trying to do a multigraph of active termserv sessions
[25-Sep-2009 15:31:42] <ckrough> I think I see
[25-Sep-2009 15:34:18] <ckrough> ... I love windows
[25-Sep-2009 15:36:00] <ckrough> Anyone know about templates or SNMP on Barracude load balancers?
[25-Sep-2009 15:47:08] <leafnunes> is anyone aware of a zenpack with a rudimentary example of a PythonClient used for data collection?
[25-Sep-2009 16:01:10] <rmatte> leafnunes: like a collector plugin you mean?
[25-Sep-2009 16:01:24] <leafnunes> yes, sorry this is my first crack at zenoss
[25-Sep-2009 16:01:46] <leafnunes> not quite up to date on the terminology, but i think i found what i need to inherit from
[25-Sep-2009 16:02:51]\n<rmatte> like are you basically just trying to write a script to\nimport values in to Zenoss to be graphed & thresholded against?
[25-Sep-2009 16:06:44] <leafnunes> basically. we have a python library for talking to virtuozzo hosts
[25-Sep-2009 16:08:21]\n<leafnunes> we don't want to talk to the virtual machines\ndirectly, but want to ask the host for information about the containers.
[25-Sep-2009 16:09:53] <rmatte> ok, I'm not famialliar with virtuozzo, but I assume it's just some other flavour of virtualization software
[25-Sep-2009 16:09:59] <leafnunes> it looks like i just want to subclass PythonPlugin
[25-Sep-2009 16:10:04] <leafnunes> rmatte: exactly.
[25-Sep-2009 16:10:23] <leafnunes> rmatte: it'd be easier if i just copied ex's example and use ssh
[25-Sep-2009 16:10:36] <leafnunes> (in the VirtualHostMonitor zenpack)
[25-Sep-2009 16:14:57] <rmatte> sorry our internet dropped for a bit
[25-Sep-2009 16:15:55] <leafnunes> rmatte: ah, yes. it's virtualization software
\n[25-Sep-2009 16:16:31] <rmatte> yeh, I didn't miss anything from\nthe conversation, SSHing from work to home and then to irc
[25-Sep-2009 16:16:34] <rmatte> work connection dropped
[25-Sep-2009 16:16:35] <leafnunes> there are two examples in VirtualHostMonitor, but neither is the Python
[25-Sep-2009 16:16:41] <rmatte> just had to reconnect to my screen session
[25-Sep-2009 16:16:45] <leafnunes> gotcha
[25-Sep-2009 16:16:56] <rmatte> I still don't quite understand what your end goal in terms of Zenoss is
[25-Sep-2009 16:17:02] <rmatte> what do you hope to accomplish?
\n[25-Sep-2009 16:18:34] <leafnunes> we don't want to monitor the\nvirtual machines as individual devices. we want the physical hosts as\ndevices, with the containers (or VMs) as children
[25-Sep-2009 16:19:16] <leafnunes> i'm just trying to figure out what kind of collector i'm creating.
[25-Sep-2009 16:23:28] <rmatte> I see
[25-Sep-2009 16:23:36] <rmatte> you're talking about some pretty tricky coding
[25-Sep-2009 16:23:42] <rmatte> I haven't gone that deep in to Zenoss
[25-Sep-2009 16:24:02] <leafnunes> :-)
[25-Sep-2009 16:24:07] <leafnunes> well, that makes me feel better.
[25-Sep-2009 16:28:12] <ocnam> Sounds kind of like the ESX ZenPack. Maybe you can use that as a template.
[25-Sep-2009 16:29:26] <rmatte> well, sounds sort of like that
[25-Sep-2009 16:29:49] <rmatte> but ESX more monitors specific things from each VM (Memory, CPU, etc...)
[25-Sep-2009 16:29:54] <rmatte> instead of everything on each one
[25-Sep-2009 16:30:20] <ocnam> should've said 'vmWare Virutal Machines'
\n[25-Sep-2009 16:30:22] <rmatte> ocnam: why not just add the VMs\nin and create device groups for them named after whatever host they are\non?
[25-Sep-2009 16:30:27] <rmatte> would be a lot less trouble
\n[25-Sep-2009 16:30:50] <ocnam> the one that adds a 'vitural\nmachine' tab and lists all of the guests running on that hosts
[25-Sep-2009 16:34:07] <ocnam> we have ~300 VMs
[25-Sep-2009 16:40:13] <djack> j,,,
[25-Sep-2009 16:40:17] <djack> hmmm i mean
[25-Sep-2009 16:42:09] <leafnunes> ocnam, rmatte: i'm adding a plugin to the VirtualMachinesMonitor
\n[25-Sep-2009 16:42:30] <leafnunes> but i'm not sure that will let\nme group all the containers under their individual hosts
[25-Sep-2009 16:42:43] <leafnunes> er, under their hosts
\n[25-Sep-2009 16:44:45] <djack> i wonder why my graphs keep giving\nnan and i see frs.be_front_CASE_1.rrd: 0.04299999999999999 when i see\nthis in zencommand
[25-Sep-2009 16:45:55] <rmatte> There's a patch for the issue with the edit tab in 2.5!
[25-Sep-2009 16:45:57] <rmatte> YESSSS!
\n[25-Sep-2009 16:46:12] <rmatte> just finishing up some testing on\nmy lab box and then I shall deploy it to my fleet of servers
[25-Sep-2009 16:46:14] <rmatte> muahaha
[25-Sep-2009 17:20:15] <djack> is the 2.5 already production ready?
[25-Sep-2009 17:26:43] <rmatte> it is as of today, finally got a patch to fix the one noticeable bug that was left
[25-Sep-2009 17:26:53] <rmatte> I just finished deploying it on all my systems
[25-Sep-2009 17:27:09] <rhettardo> nice
\n[25-Sep-2009 17:27:20] <rmatte> I might see if they'd be willing\nto add the patch and release a version 2.4.6 (not sure if they will\nthough)
[25-Sep-2009 17:27:30] <rmatte> I'll write a guide for patching it tomorrow anyways
[25-Sep-2009 17:27:34] <rmatte> erm, on Monday rather
[25-Sep-2009 17:33:03] <djack> lol is there an upgrade path let's say from a bin file ?
[25-Sep-2009 17:35:43] <rmatte> you'd basically install 2.4.5 then run "zenpatch 15214" and restart Zenoss
[25-Sep-2009 17:35:51] <rmatte> and you'll have a production ready version of 2.4.5
[25-Sep-2009 17:36:15] <rmatte> I am SOOOOOOO glad that bug is fixed
[25-Sep-2009 17:36:23] <rmatte> I've had engineers whining to me about that for over a month
[25-Sep-2009 17:37:28] <rmatte> that patch totally made my day (compliments of mr chippy)
[25-Sep-2009 17:43:46] <rmatte> well, done for the day, time to flee
[25-Sep-2009 17:43:49] <rmatte> cheers folks
[25-Sep-2009 17:43:59] <djack> cheers
[25-Sep-2009 17:44:04] <djack> have a nice weekend
[25-Sep-2009 17:44:18] <rmatte> you too
[25-Sep-2009 17:46:43] <daMaestro> does anyone know how to fire off the event cleanup thread/
[25-Sep-2009 17:46:58] <daMaestro> we are seeing some DB issues and would like to fire one off when we have a DBA around
[26-Sep-2009 00:00:46] [disconnected at Sat Sep 26 00:00:46 2009]
[26-Sep-2009 00:00:46] [connected at Sat Sep 26 00:00:46 2009]
[26-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[26-Sep-2009 13:34:25] davetoo1 is now known as davetoo
[26-Sep-2009 13:38:05] <metalklesk> hello
[26-Sep-2009 13:56:02] <davetoo> hi
[27-Sep-2009 00:00:46] [disconnected at Sun Sep 27 00:00:46 2009]
[27-Sep-2009 00:00:46] [connected at Sun Sep 27 00:00:46 2009]
[27-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
[27-Sep-2009 04:04:39] <Egyptian[web]_> afternoon all
[27-Sep-2009 04:05:10]\n<Egyptian[web]_> i got the page setting of alerts function to\nsend me an sms via the correct command .. but it comes blank .. any\nhelp anyone?
[27-Sep-2009 08:14:29] <Egyptian[web]_> anybody awake?
[27-Sep-2009 12:34:33] <Egyptian[Home]> evening all
\n[27-Sep-2009 12:35:44] <Egyptian[Home]> i have managed to use the\n"send page" function to put a custom script that send me text sms\nmessages. however, the messages are blank and my logs tell me there\nwere no parameters passed other than the mobile number .. help?
[27-Sep-2009 12:39:02] <Egyptian[Home]> is there a variable like $MESSAGE that i can use too ?
\n[27-Sep-2009 13:17:49] <Diddi> hello. I've just installed the\nhttp monitor zenpack, and followed the instructions to create graphs\nand thresholds... But the graphs doesn't work. Looking in event.log I\nsee
[27-Sep-2009 13:17:51] <Diddi> 2009-09-27T20:12:24 ERROR RenderServer Failed to generate a graph
[27-Sep-2009 13:17:54] <Diddi> Traceback (most recent call last):
[27-Sep-2009 13:17:54] <Diddi> rrdtool.graph(*gopts)
[27-Sep-2009 13:17:57] <Diddi> File "/usr/local/zenoss/zenoss/Products/ZenRRD/RenderServer.py", line 190, in render
\n[27-Sep-2009 13:18:28] <Diddi> I can't find the rrd-files, and I\ncan't find any solution to my problem... i've searched the forums
\n[27-Sep-2009 13:21:47] <miko_uk> I have just installed the SQL\nserver zenpack, can anyone tell me where the graphs should show up?
[27-Sep-2009 16:12:38] <etank> miko_uk: did you bind the new zenpack to the sql server device and remodel?
\n[27-Sep-2009 16:32:32] <Diddi> hm.. I've added advanced device\ndetail and dell monitor zenpacks.. but I'm not getting anything under\nthe hardware tab. what am I missing?
[27-Sep-2009\n18:12:22] <Diddi> I have problems monitoring processes and\nipservices... when I add a proc/service for monitoring, it works a\ncouple of minutes, and then I get events saying the process is not\nrunning
[27-Sep-2009 18:12:27] <Diddi> but it is..
[27-Sep-2009 18:12:58] <Diddi> currently only mysqld and ssh seems to be affected.. other processes/ipservices works
[27-Sep-2009 18:39:02] <Diddi> when I do a snmpwalk on the host, I get several results of myslqd
[27-Sep-2009 18:51:22] <Diddi> and now all of a sudden I can't monitor snmp ipservice
[27-Sep-2009 18:51:28] <Diddi> why is this?
[28-Sep-2009 00:00:46] [disconnected at Mon Sep 28 00:00:46 2009]
[28-Sep-2009 00:00:46] [connected at Mon Sep 28 00:00:46 2009]
[28-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[28-Sep-2009 00:50:15] <Snyper> question about 2.5, will it support gathering more data from wmi ?
[28-Sep-2009 00:54:00]\n<Snyper> more specific, gathering OS, hardware, software\ninformation. Right now it seems to only be used for services and\npulling data to make graphs..
[28-Sep-2009 00:54:18] <Snyper> Graphs are nice but useless to me.
\n[28-Sep-2009 00:55:55] <Snyper> For 2.4 i tried to dive in and\ncreate my own plugin to do what i needed but could not figure out where\nto place the data.. I have all the wmi queries just dont understand\nwere to put them in zenoss like snmp does
[28-Sep-2009 03:28:22] <Troubadix09> hi all
\n[28-Sep-2009 07:38:46] <Egyptian[Web]> help!! i have been\nstruggling with sending myself text sms for the last 3 days and no\nmatter what i do .. i keep getting blank messages.. help!
[28-Sep-2009 07:40:05]\n<Egyptian[Web]> i m using my own script and have replaced the\npage command with /home/zenoss/bin/sms.sh $RECIPIENT and still blank ..\nthe script works from the command line but not from zenoss
[28-Sep-2009 07:40:06] <Egyptian[Web]> help
[28-Sep-2009 07:40:56] <Diddi> the script has to read text from stdin to work iirc
[28-Sep-2009 07:44:02] <Egyptian[Web]> Diddi: it does .. it sends to $RECIPIENT ..adn then what?
\n[28-Sep-2009 07:44:24] <Egyptian[Web]> i have tried\n$NOTIFICATIONTYPE$: $HOSTALIAS$: $HOSTNAME$: $HOSTADDRESS$:\n$SERVICEDESC$ is $SERVICESTATE$ and they all come out blank
[28-Sep-2009 07:47:05]\n<Diddi> I'm currently using gammu to send sms, and it works\nperfect with zenoss... where are you trying to substitute those\nvariables?
[28-Sep-2009 07:47:23] <Egyptian[Web]> in the page command
[28-Sep-2009 07:47:58] <Egyptian[Web]> Diddi: here http://zenoss:8080/zport/dmd/editSettings
[28-Sep-2009 07:48:00]\n<Diddi> alright, the only thing you should have in the page\ncommand is $RECIPIENT. the text to be sent is set in the alerting rule
[28-Sep-2009 07:48:32] <Diddi> but it sounds like your script only reads text from command line, not stdin
[28-Sep-2009 07:48:57] <Egyptian[Web]> aaaah lerting rule
[28-Sep-2009 07:49:16] <Diddi> and btw, your local url doesn't work for me (:
[28-Sep-2009 07:49:56] <Egyptian[Web]> well of course not but .it does for me
plus you can cut and paste and substitute at will [28-Sep-2009 07:50:34] <Diddi>
indeed\n[28-Sep-2009 08:19:27] <LarsN> Hi everyone. I am running Zenoss\n2.4.5. I have an alerting rule that reads: Event State = New /\nProduction State = Production / Severity >= Error / Send Clear\nMessages False.
[28-Sep-2009 08:21:20] <LarsN>\nWhen a monitored device fails, Zenoss picks it up and sends an alert as\nexpected. The problem is in 5 minutes when it does it's next check,\nit'll send another event, and 5 minutes after that..... I updated the\nrule to send clear messages, and noticed that every 5 minutes it was\nsending a clear, followed by a failure. Our older, "migrated five or\nsix times" zenoss installation running an identical alerting rule\ndoesn't do this, it continues to add
[28-Sep-2009 08:21:27] <LarsN> what am I missing?
[28-Sep-2009 08:25:05] <ckrough> is that monitored thing bouncing?
[28-Sep-2009 08:25:16] <ckrough> ie: it's failing and correcting itself repeatedly
\n[28-Sep-2009 08:25:46] <LarsN> ckrough: pretty sure that's not\nthe case since it does it for everything, including powered off devices.
[28-Sep-2009 08:25:52] <ckrough> ah
[28-Sep-2009 08:26:18] <LarsN> ckrough: If it had just been web services I'd have thought exactly that.
[28-Sep-2009 08:53:43] <ke4qqq> LarsN: what's repeat time?
[28-Sep-2009 08:54:06] <LarsN> ke4qqq: every 5 minutes I get a clear, followed by a fail with the same timestamp.
\n[28-Sep-2009 08:54:10] <ke4qqq> LarsN: the other thing to check\nis that in the event console the count for the event in question is\nbeing incremented.
[28-Sep-2009 08:54:17] <LarsN> ke4qqq: it's not.
[28-Sep-2009 08:54:23] <ke4qqq> ahhh - so it thinks it is a new event each time
[28-Sep-2009 08:54:27] <LarsN> yes
[28-Sep-2009 08:54:48] <LarsN> our, older, and mostly broken Zenoss install, doesn't however do this
[28-Sep-2009 08:55:08] <ke4qqq> are any events incrementing??
[28-Sep-2009 08:55:27] <ke4qqq> even ones you aren't alerting on
[28-Sep-2009 08:56:08] <LarsN> ke4qqq: yes.
[28-Sep-2009 08:56:36] <ke4qqq> so what kind of events are alerting but clearing themselves?? ping? something else?
[28-Sep-2009 08:56:56] <LarsN> looks like primarily SNMP Agent Down / and Connection Refused on /status/http
[28-Sep-2009 08:57:03] <LarsN> are the two primary ones not incrimenting.
\n[28-Sep-2009 08:57:29] <LarsN> as I mentioned before, if I power\noff a device. the http monitor, and snmp monitor will still alert\nevery 5 minutes, and not incriment.
[28-Sep-2009 08:59:54] <rmatte> good morning golks
[28-Sep-2009 08:59:57] <rmatte> folks too
[28-Sep-2009 09:00:10] <rmatte> (off to a good start for typos today apparently)
[28-Sep-2009 09:00:37] <ke4qqq> weird - I can't imagine what would cause those events not to increment.
[28-Sep-2009 09:01:54] <Diddi> have you checked the logs?
[28-Sep-2009 09:02:15] <Diddi> if there's any problem with storing the events or something like that
[28-Sep-2009 09:02:27] * Diddi is just brainstorming
[28-Sep-2009 09:04:34] <rmatte> LarsN: alert every 5 minutes as in email alerts?
[28-Sep-2009 09:04:41] <LarsN> rmatte: yes.
[28-Sep-2009 09:04:51] <rmatte> LarsN: you have a repeat time set in your alert settings
[28-Sep-2009 09:04:53] <rmatte> go check
[28-Sep-2009 09:04:56] <LarsN> I'll dive into logs right after the meeting I'm in.
[28-Sep-2009 09:05:05] <rmatte> I'll bet that's the issue, if it's set to 300 change it to 0
[28-Sep-2009 09:06:46] <ke4qqq> rmatte: it's actually not incrementing events - each time it's a new instance of the event
[28-Sep-2009 09:07:40] <rmatte> yeh, but you mentioned that you powered the device off
[28-Sep-2009 09:07:53] <rmatte> so I assume zenoss has an "ip <whatever> is down" event for the device?
[28-Sep-2009 09:08:54] <rmatte> Ok, here's the scenario...
[28-Sep-2009 09:09:40]\n<rmatte> You power the device down, but since Zenoss is in the\nmiddle of a polling cycle it notices that the http monitor and snmp\nmonitor are down before it realizes that the device is unpingable, so\nit generates events for each
[28-Sep-2009 09:09:47] <rmatte> then it finally realizes the device is unreachable
[28-Sep-2009 09:10:06]\n<rmatte> it then does what it's designed to do and only checks\nthe device via ping from then on (doesn't attempt to monitor the other\nservices)
[28-Sep-2009 09:10:18] <rmatte> hence\nwhy the 2 alerts that were generated just before the ping down event\ndon't increment
[28-Sep-2009 09:10:33] <rmatte>\nbut you continue to be alerted every 5 minutes for those events because\nyou have a repeat time set in your alert settings
[28-Sep-2009 09:10:39] <rmatte> that's what's happening
[28-Sep-2009 09:15:52] <twm1010> Morning folks.
[28-Sep-2009 09:16:16] <rmatte> morning
[28-Sep-2009 09:20:15] <LarsN> rmatte: great plan. Except my Repeat time (Secs) is 0
[28-Sep-2009 09:21:56] <straterra> rmatte: what patch level are you running zenoss at?
[28-Sep-2009 09:22:26] <rmatte> you mean what version and what patches have I applied in addition?
[28-Sep-2009 09:22:31] <straterra> Yeah
\n[28-Sep-2009 09:22:42] <eidolon> hi folks - i have a disk monitor\non a windows box. in my OS pane, i see 'utilization' '100%' - but the\nstatus is green. what defines whether this is an event or not? sure\nas hell should be
\n[28-Sep-2009 09:22:53] <rmatte> I'm actually going to be writing\na blog post about the patch I have applied, I'm running 2.4.5 with 1\npatch applied
[28-Sep-2009 09:23:09] <straterra> Ok, I'll wait then
[28-Sep-2009 09:23:26] <eidolon> (there's a patch to 2.4.5?)
[28-Sep-2009 09:23:48] <straterra> I'm still running 2.4.2
\n[28-Sep-2009 09:23:50] <rmatte> eidolon: it was released on\nFriday afternoon to fix the bug where you can't properly set product\ninfo manually
[28-Sep-2009 09:24:19] <eidolon> ah. okay, not a problem for me.
[28-Sep-2009 09:24:23] <rmatte> I'm about to type up a quick guide on applying the patch and reindexing the product list
[28-Sep-2009 09:24:27] <straterra> Is updating pretty easy?
[28-Sep-2009 09:24:41] <rmatte> straterra: updating from 2.4.2 to 2.4.5 you mean?
[28-Sep-2009 09:24:44] <straterra> Yeah
[28-Sep-2009 09:24:50] <rmatte> straterra: you using stack installer?
[28-Sep-2009 09:24:59] <straterra> RPM installer for RHEL
[28-Sep-2009 09:25:12] <rmatte> ah, I'd imagine you'd just run an update with the RPM, that's about all there is to it
[28-Sep-2009 09:25:21] <rmatte> but I'd read the install guide just to be sure
\n[28-Sep-2009 09:25:50] <rmatte> with the stack installer it's\nbraindead simple, you just run the new stack installer and hit enter\ntwice, and off it goes
[28-Sep-2009 09:26:18] <eidolon> so. any ideas on my threshold problem?
[28-Sep-2009 09:27:12] <rmatte> eidolon: the filesystem template
[28-Sep-2009 09:27:20] <rmatte> eidolon: check if there is one locally applied to that drive
[28-Sep-2009 09:27:28] <eidolon> one moment
[28-Sep-2009 09:27:33] <rmatte> if not, go check the FileSystem template under /Server
[28-Sep-2009 09:27:39] <eidolon> wait wait. s ec.
[28-Sep-2009 09:27:48] <rmatte> make sure the threshold is enabled and set properly
[28-Sep-2009 09:27:55] <straterra> Release notes just say I can directly upgrade to it
[28-Sep-2009 09:28:05] <rmatte> it'll be "<something> * 0.90" by default
[28-Sep-2009 09:28:11] <rmatte> representing the percentage
[28-Sep-2009 09:28:23] <rmatte> straterra: then give'r
[28-Sep-2009 09:28:40] <eidolon> so under:
[28-Sep-2009 09:29:05] <eidolon> /Devices /Server /Windows /WMI /bos-cmds01.REDACTED.net /os /C
[28-Sep-2009 09:29:07] <eidolon> i see:
\n[28-Sep-2009 09:29:20] <eidolon>\nFileSystem \n/Devices/Server/Windows/WMI FileSystem\ntemplate that uses Windows WMI.
[28-Sep-2009 09:29:29] <straterra> rmatte: should I upgrade my core zenpacks too?
[28-Sep-2009 09:29:41] <eidolon> so the template in use is in /devices/server/windows/wmi, i should look there?
[28-Sep-2009 09:29:59] <eidolon> i see Filesystem there
[28-Sep-2009 09:30:25] <eidolon> ahh. 'low disk space' threshold
[28-Sep-2009 09:30:32] <eidolon> enabled: true
\n[28-Sep-2009 09:30:45] <Egyptian[Web]> Diddi: i fixed my script\nand it now sends sms to me .. read var instead of $* as parameters
[28-Sep-2009 09:30:51] <Egyptian[Web]> Diddi: thanks
\n[28-Sep-2009 09:33:30] <rmatte> straterra :\nblogs/ryanmatte/2009/09/28/patching-hardwaresoftware-info-bug-trac-ticket-5473
[28-Sep-2009 09:34:16] <straterra> Gotcha
[28-Sep-2009 09:36:02] <straterra> Grrr
\n[28-Sep-2009 09:36:14] <straterra> I wish Zenoss would say you\nneed zenoss to be RUNNING before you try to install the core zenpack
[28-Sep-2009 09:37:49] <rmatte> hehe
[28-Sep-2009 09:37:58] <rmatte> you need zenoss running before trying to install any ZenPack
[28-Sep-2009 09:38:09] <rmatte> otherwise it has no way of importing the objects in to Zope
[28-Sep-2009 09:42:25] <twm1010> rmatte: I'm thinking of switching all my Windows boxes to WMI perf collection
[28-Sep-2009 09:42:44] <rmatte> fun fun
[28-Sep-2009 09:43:05] <cgibbons> yay! i get a kickback anytime anyone uses WMI! woohoo
[28-Sep-2009 09:43:09] <cgibbons> er, did I just say that outloud again?!
[28-Sep-2009 09:43:35] <rmatte> haha
[28-Sep-2009 09:43:50] <rmatte> WMI is unreliable and cumbersome
[28-Sep-2009 09:44:21] <jb> but necessary..
[28-Sep-2009 09:44:42] <rmatte> yeh, but it'd be nice if they'd improve it
[28-Sep-2009 09:45:04] <rmatte> I love how WMI has a tendency to just magically break
[28-Sep-2009 09:45:09] <rmatte> never have that problem with SNMP
[28-Sep-2009 09:45:48] <cgibbons> they do seem to like creating server-side memory leaks with it for each new OS release
[28-Sep-2009 09:46:19] <rmatte> yup
[28-Sep-2009 10:00:25] <straterra> rmatte: thanks for the link..I'm all patched up
[28-Sep-2009 10:07:36] <rmatte> hmmm, this zapplet thing is pretty cool
[28-Sep-2009 10:07:39] <rmatte> straterra: no problem
[28-Sep-2009 10:08:21] <daMaestro> does anyone know how to fire off the events cleanup call?
[28-Sep-2009 10:08:45]\n<rmatte> daMaestro: like the event cleanup that fires off after\nwhatever amount of time that you set (an hour or whatever)?
[28-Sep-2009 10:08:51] <daMaestro> yeah
[28-Sep-2009 10:08:56] <rmatte> hmmm
[28-Sep-2009 10:09:12] <daMaestro> the process that deletes events older then N number of days
[28-Sep-2009 10:09:20] <rmatte> oh, that one
[28-Sep-2009 10:09:36] <daMaestro> basically, that launches around 7 pm for us.. and it almost takes down zenoss
[28-Sep-2009 10:09:37] <rmatte> not really sure, I only know how to delete all events for a particular device
[28-Sep-2009 10:09:56] <rmatte> ummmm, how many events do you guys get per day?
[28-Sep-2009 10:09:57] <daMaestro> so i need to launch it with a DBA on hand to see what we can do to make it run better
[28-Sep-2009 10:10:03] <rmatte> it shouldn't even come close to taking down Zenoss?
[28-Sep-2009 10:10:13] <daMaestro> the amount of disk i/o is absurd
[28-Sep-2009 10:10:23] <daMaestro> for about an hour it's taxed
\n[28-Sep-2009 10:11:47] <daMaestro> "Delete Historical Events\nOlder Than (days)" is the process that i'm attempting to manually run
[28-Sep-2009 10:15:03] <rmatte> yeh, I'm not sure what function that executes
[28-Sep-2009 10:15:06] <rmatte> you'd have to ask a dev
\n[28-Sep-2009 10:15:31] <rmatte> I could go digging through code\nand eventually find it, but I don't have time this morning
[28-Sep-2009 10:15:42] <daMaestro> yeah, it's just a python script that runs
\n[28-Sep-2009 10:15:56] <daMaestro> i looked into it last time it\nhappened and found the command but now i can't find it again
[28-Sep-2009 10:16:36] <daMaestro> hmm looks like it's ZenDeleteHistory.py
[28-Sep-2009 10:17:52] <rmatte> npmccallum: ping
[28-Sep-2009 10:18:07] <npmccallum> rmatte: pong
\n[28-Sep-2009 10:18:27] <rmatte> hey, I'm trying out zapplet and\nI'm loving it but it seems to have a small bug, I'm going over the code\nbut not figuring this out
[28-Sep-2009 10:19:13]\n<rmatte> when there are no events in the event log on a server a\ncaution icon is displayed and if you rollover the icon the text\ndisplays an error
[28-Sep-2009 10:19:29] <rmatte> complaining about line 56 and line 110 in events.py
[28-Sep-2009 10:19:43] <rmatte> erm, event.py
[28-Sep-2009 10:19:44] <rmatte> rather
[28-Sep-2009 10:20:12] <rmatte> 110 is: highest = filter(lambda x: x[-1] > 0, results)[-1]
[28-Sep-2009 10:20:19] <rmatte> any idea what might need to be changed?
[28-Sep-2009 10:20:39] <npmccallum> rmatte: give me a second to look at the code
[28-Sep-2009 10:20:46] <rmatte> thanks
[28-Sep-2009 10:21:23] <rmatte> let me see if I can type this error off for you (difficult since it shows up in a popup)
[28-Sep-2009 10:22:33] <npmccallum> rmatte: ok
[28-Sep-2009 10:22:37] <npmccallum> rmatte: here's the problem
[28-Sep-2009 10:22:51] <npmccallum> I've never actually coded for the scenario where there are no events
[28-Sep-2009 10:22:56] <Brad_K> rmatte: uh, so WMI is b0rken? should we be using perfmon instead?
[28-Sep-2009 10:23:05] <npmccallum> I've assumed that there will *always* be at least one event
[28-Sep-2009 10:23:11] <npmccallum> bad assumption, I know
[28-Sep-2009 10:23:12] <rmatte> Brad_K: it's not really "broken" I just find it to be tempermental
[28-Sep-2009 10:23:37] <Brad_K> rmatte: which is not what we need in a production system/network monitoring system....
\n[28-Sep-2009 10:23:40] <rmatte> oh, Brad_K:\nblogs/ryanmatte/2009/09/28/patching-hardwaresoftware-info-bug-trac-ticket-5473
[28-Sep-2009 10:23:51] <rmatte> npmccallum: ah
[28-Sep-2009 10:24:02] <rmatte> npmccallum: yeh, it should show as green or something if there are zero events
[28-Sep-2009 10:24:14] <npmccallum> rmatte: right
[28-Sep-2009 10:24:15] <npmccallum>
[28-Sep-2009 10:24:27] <Brad_K> rmatte: i greatly appreciate the patch. thanks!
[28-Sep-2009 10:24:32] <npmccallum> you obviously run the cleanest network I've ever seen
[28-Sep-2009 10:25:00]\n<rmatte> npmccallum: well, some of the zenoss boxes that I\nmonitor are more lab boxes than anything, so the event logs are often\nempty
[28-Sep-2009 10:25:28] <rmatte> Brad_K:\nthanks mrchippy if you see him around, he wrote the patch, I just\nbugged the crap out of them until they patched it and wrote the blog\nentry
[28-Sep-2009 10:25:35] <rmatte> thank* rather
[28-Sep-2009 10:26:02] <npmccallum> rmatte: you can probably just do a "if len(results) == 0" and create a suitable summary event
[28-Sep-2009 10:26:44] <npmccallum> rmatte: this function gathers returns a tuple with (Event, Event[])
[28-Sep-2009 10:26:47] <rmatte> yeh, and how do I set an alternate icon?
[28-Sep-2009 10:26:56] <rmatte> lol
\n[28-Sep-2009 10:27:20] <npmccallum> the first Event in that tuple\nis a "virtual" event that is used to generate the icon and summary of\nthe applet
[28-Sep-2009 10:27:23] <rmatte> do you plan on possibly patching it up at some point and releasing a new version?
\n[28-Sep-2009 10:27:47] <npmccallum> rmatte: I have a huge rewrite\nsitting in a backup somewhere that I just never got back to\n(unfortunately)
[28-Sep-2009 10:27:59] <npmccallum> rmatte: the rewrite will allow *all* functionality as a plugin
\n[28-Sep-2009 10:28:28] <rmatte> ah, that's pretty cool, but\ncouldn't you just throw a few lines in and do a temporary release?
[28-Sep-2009 10:28:34] <rmatte> you're the python developer, not me lol
[28-Sep-2009 10:28:42] <rmatte> my python is ok but it's very much geared towards Zenoss
[28-Sep-2009 10:28:54] <npmccallum> yes, I can
[28-Sep-2009 10:28:58] <npmccallum> I'll try to get to it today
[28-Sep-2009 10:29:05] <rmatte> thanks
[28-Sep-2009 10:29:35] <rmatte> now I just need to figure out how to use the historify xml-rpc in perl
\n[28-Sep-2009 10:29:55] <rmatte> since our ticket create script is\nperl, need the script to acknowledge events after opening tickets
\n[28-Sep-2009 10:30:23] <rmatte> might just do it as a seperate\npython script and have the perl script call on it, if worst comes to\nworst
[28-Sep-2009 10:30:46] <rmatte> g'day Matt
[28-Sep-2009 10:31:02] <mrayzenoss> morning all
[28-Sep-2009 10:31:12] <ckrough> morning
\n[28-Sep-2009 10:31:32] <rmatte> mrchippy made my day on Friday:\nblogs/ryanmatte/2009/09/28/patching-hardwaresoftware-info-bug-trac-ticket-5473
[28-Sep-2009 10:32:14] <rmatte> It'd almost be worth applying that patch to 2.4.5 and releasing it as 2.4.6
[28-Sep-2009 10:36:06]\n<exarkun> Looking around for .debs, I found a forum post that\nsuggests this source line: "deb http://dev.zenoss.org/deb main stable"
\n[28-Sep-2009 10:36:36] <exarkun> How official is that? Does\n"dev" in the domain indicate packages of potentially unstable versions?
[28-Sep-2009 10:36:42] <exarkun> (will it even work? I haven't tried yet)
[28-Sep-2009 10:36:53] <mrayzenoss> exarkun: that's in the Installation Guide
[28-Sep-2009 10:37:28] <mrayzenoss> exarkun: so fairly official
[28-Sep-2009 10:37:28] <exarkun> Okay, I'll take a look there. Thanks.
[28-Sep-2009 10:37:39] <exarkun> Oh, I see. Cool.
\n[28-Sep-2009 10:42:26] <daMaestro> rmatte, delete h,j,d from\nhistory h LEFT JOIN log j ON h.evid = j.evid LEFT JOIN detail d ON\nh.evid = d.evid WHERE StateChange < DATE_SUB(NOW(), INTERVAL 30 day)
[28-Sep-2009 10:42:39] <daMaestro> rmatte, and there are not proper indexes to make that run even slightly quickly
[28-Sep-2009 10:43:24] <rmatte> daMaestro: ah
[28-Sep-2009 10:43:54] <daMaestro> rmatte, i'm having our DBA add some indexes that should help... hopefully
[28-Sep-2009 10:43:55] <rmatte> daMaestro: I guess it really depends on the volume of events
[28-Sep-2009 10:44:12] <daMaestro> rmatte, something like 4 mil right now
[28-Sep-2009 10:44:23] <daMaestro> rmatte, and that is with only keeping 30 days
[28-Sep-2009 10:53:13] <rmatte> crazy
\n[28-Sep-2009 11:10:25] <gwb2351> anybody run into errors\n(timeout, invoalid responses) with apache as the front-end (proxy) to\nzope?
[28-Sep-2009 11:10:40] <ckrough> not normally
[28-Sep-2009 11:11:02] <gwb2351> the ssl_error log from httpd has these bits in it
\n[28-Sep-2009 11:11:25] <Brad_K> and the performance graphs aren't\ncoming up, either. and we're also having problems with some of the\nstatus pages, although not necessarily all of them.
\n[28-Sep-2009 11:11:28] <gwb2351> proxy: Error reading from\nremote server returned by /zport/dmd/getUserMessages, referer:\nhttps://eye.its.utexas.edu/zport/dmd/Devices/Server/Passive/devices/mail.utexas.edu/viewDevicePerformance
\n[28-Sep-2009 11:11:42] <gwb2351> (70007)The timeout specified has\nexpired: proxy: error reading status line from remote server 127.0.0.1,\nreferer:\nhttps://eye.its.utexas.edu/zport/dmd/Devices/Server/Passive/devices/mail.utexas.edu/viewDevicePerformance
[28-Sep-2009 11:12:50] <ckrough> this just started happenning?
[28-Sep-2009 11:13:04] <Brad_K> we've seen this before.
[28-Sep-2009 11:13:15] <Brad_K> but it has been intermittent.
\n[28-Sep-2009 11:13:56] <Brad_K> it did just occur to me that\nmaybe we're having the same database problems as daMaestro reported.
[28-Sep-2009 11:14:26] <ckrough> server load is normal?
[28-Sep-2009 11:14:43] <Brad_K> 0.08.
[28-Sep-2009 11:14:52] <ckrough> iowait
[28-Sep-2009 11:15:16] <Brad_K> not seeing any. 99.5% idle.
[28-Sep-2009 11:15:20] <ckrough> obviously only counts if the problem is currently happening
[28-Sep-2009 11:15:31] <Brad_K> currently ongoing, yes.
\n[28-Sep-2009 11:15:40] <ckrough> just wondering if zope/python is\ntied up with something and isnt returning pages to apache in a timely\nfashion
[28-Sep-2009 11:15:42] <gwb2351> problem is currently happening, iostat 2 10 shows 0.00 %iowait
[28-Sep-2009 11:15:49] <ckrough> like zenhub rebuilding configurations, etc...
[28-Sep-2009 11:16:02] <ckrough> especially since its intermittant
[28-Sep-2009 11:32:09] <gwb2351> what process in zenoss handles apache requests? the errors returned to the browser are
[28-Sep-2009 11:32:19] <gwb2351> The proxy server received an invalid response from an upstream server.
\n[28-Sep-2009 11:32:19] <gwb2351> The proxy server could not\nhandle the request GETÂ /zport/acl_users/cookieAuthHelper/login_form.
[28-Sep-2009 11:32:29] <gwb2351> (or other similiar GET /zport errors)
[28-Sep-2009 11:38:25] <ckrough> zope
[28-Sep-2009 11:41:51] <gwb2351> if I look at what zenhub is doing, I see lots of timeouts on a select
[28-Sep-2009 11:42:09] <gwb2351> [root@eye localhost]# strace -s 1024 -t -v -p 9355
[28-Sep-2009 11:42:09] <gwb2351> Process 9355 attached - interrupt to quit
[28-Sep-2009 11:42:09] <gwb2351> 11:41:23 select(112, [111], [], [111], {0, 828000}) = 0 (Timeout)
[28-Sep-2009 11:42:09] <gwb2351> 11:41:24 select(112, [111], [], [111], {1, 24000}) = 0 (Timeout)
[28-Sep-2009 11:42:38] <gwb2351> where process 9355 is the zenhub.py process
[28-Sep-2009 11:44:43] <ckrough> Do you guys run multiple zenhub workers?
[28-Sep-2009 11:44:52] <gwb2351> yes - 4 additional
\n[28-Sep-2009 11:47:24] <Brad_K> and i'm still trying to come up\nto speed on zenoss so that i can help Greg, allowing him to have more\ntime to do bigger stuff.
[28-Sep-2009 11:47:58] <ckrough> fwiw, I am getting select timehouts on zenhub on a normally functioning install
[28-Sep-2009 11:48:03] <ckrough> *timeouts*
[28-Sep-2009 11:48:19] <gwb2351> good to know, thanks
[28-Sep-2009 11:50:05] <gwb2351> i think i'll restart zope and see what happens
\n[28-Sep-2009 11:50:56] <Brad_K> and so far as i know, we have not\nyet attempted to apply the performance tuning tips mentioned in\ndocs/DOC-2521
[28-Sep-2009 11:56:34] <ckrough> are you using mod_proxy?
[28-Sep-2009 11:56:47] <ckrough> if so, whats your ProxyTimeout value?
[28-Sep-2009 11:57:13] <Brad_K> i think we are using mod_proxy, let me check...
[28-Sep-2009 11:58:52] <Brad_K> i'm not finding that string anywhere in /etc/httpd/*
[28-Sep-2009 12:01:07] <ckrough> here's mine http://pastebin.com/m4df7ad2a
[28-Sep-2009 12:02:05] <ckrough> http://httpd.apache.org/docs/2.0/mod/mod_proxy.html#proxytimeout
[28-Sep-2009 12:02:51] <ckrough> anyone on running distributed collectors?
[28-Sep-2009 12:03:15] <Brad_K> we are running distributed collectors, too.
[28-Sep-2009 12:03:54]\n<ckrough> Im experiencing a problem where zenping on a collector\ndetects a device as down, but the zenhub box never creates an event for\nit
[28-Sep-2009 12:04:10] <ckrough> Im not sure what the 'workflow' is for zenping from the collector
[28-Sep-2009 12:05:42]\n<Brad_K> ckrough: okay, so I found all instances of the\ncase-insensitive string underneath /etc/httpd/. See\nhttp://pastebin.com/m7518e26e
[28-Sep-2009 12:06:29]\n<Brad_K> we don't actually use zenping. we use the check_ping\nNagios plug-in that we hacked to get to work properly under Zenoss.
[28-Sep-2009 12:06:36]\n<twm1010> I found Zenoss performance to go up the most when I\nstarted figuring out what was generating debug "events"
[28-Sep-2009 12:06:47] <twm1010> bad OID messages, etc.
[28-Sep-2009 12:07:06]\n<ckrough> twm1010: thats good thinking, most of that stuff is\ninvisible if your arent searching for it. I'll have to do an audit on\nthat.
[28-Sep-2009 12:08:07] <ckrough> Brad_K: what are you using to forward web :80 or :443 to zope on your install? I dont see it here
[28-Sep-2009 12:08:09]\n<Brad_K> we also don't use the native Zenoss module to check ssh\nconnections, since it doesn't properly close the TCP connection. The\nNagios check_ssh module does, after we hacked it to work with zenoss.
[28-Sep-2009 12:08:28] <ckrough> Brad_K: yeah, my logs are filled with sshd complaints.
[28-Sep-2009 12:09:08] <Brad_K> ckrough: we're actually using the RewriteEngine and a RewriteRule.
[28-Sep-2009 12:09:46] <Brad_K> this is in our /etc/httpd/conf.d/ssl.conf:
[28-Sep-2009 12:09:47] <Brad_K> RewriteRule ^/(.*) http://127.0.0.1:8080/VirtualHostBase/https/%{SERVER_NAME}:44
[28-Sep-2009 12:09:47] <Brad_K> 3/$1 [P,L]
[28-Sep-2009 12:10:15] <Brad_K> sorry, that should be one long line, not two lines. didn't paste right.
[28-Sep-2009 12:17:00]\n<gwb2351> I'll peruse this information and see what happens\nblogs/zenossblog/2007/06/05/zenoss-and-apache-documentation
[28-Sep-2009 12:21:03] <jelmer> cgibbons: hi - did you get my reply wrt DCOM/WMI?
[28-Sep-2009 12:21:26] <cgibbons> jelmer: yup
haven't really processed it yet but it is in my inbox\n[28-Sep-2009 12:39:43] <Diddi> when creating a new threshold, is\nit possible to get values in certain units? I'm currently monitoring\nremaining percentage of my ups-battery, but the event will only say\n"<threshold> not met: current value #number#"
[28-Sep-2009 12:39:53] <Diddi> I'd like to add '%' at the end of that message (:
[28-Sep-2009 13:09:58]\n<Brad_K> okay, we got all our rewrite stuff resolved. what we\nhad was an incomplete cut-n-paste from the instructions that we were\nfollowing.
[28-Sep-2009 13:10:35] <purevol> ahoy zenossers
\n[28-Sep-2009 13:14:50] <daMaestro> wonderful.. so now that we\nadded the indexes it's less disruptive but now we get\n"_mysql_exceptions.OperationalError: (1206, 'The total number of locks\nexceeds the lock table size')"
[28-Sep-2009 13:14:54] <daMaestro> pos
[28-Sep-2009 13:15:56] <rmatte> daMaestro, file a bug report: dev.zenoss.com/trac login: zenoss/zenoss
[28-Sep-2009 13:18:36] <purevol> rmatte ...you made the hus Cisco MIB Zenpack?
[28-Sep-2009 13:18:42] <purevol> huge, not hus
\n[28-Sep-2009 13:20:59] <feutete> Hello, I have some hosts who\nhave never had a service running on TCP 8080, but Zenoss believes they\nshould, so it always reports http-alt down
\n[28-Sep-2009 13:21:22] <feutete> strangely, if I delete the\nIPService under the OS tab for the devices, it comes back on its own
[28-Sep-2009 13:21:35] <feutete> if I set Monitor=False, Zenoss still alerts
[28-Sep-2009 13:21:58] <feutete> any ideas on how to convince Zenoss that the devices shouldn't have anything on 8080?
[28-Sep-2009 13:22:13] <Brad_K> did you remodel?
[28-Sep-2009 13:22:19] <feutete> yes
\n[28-Sep-2009 13:22:47] <Brad_K> well, that's the limit of my\nknowlesge of zenoss -- always remodel after making any change. ;-)
[28-Sep-2009 13:23:19] <feutete> yeah, I have made a habit of doing that as well
\n[28-Sep-2009 13:23:58] <feutete> it would be nice if IPServices\nfollowed the template hierarchy...I have just a few devices that DO\nlisten on 8080
[28-Sep-2009 13:24:24] <feutete> but apparently zenoss checks ALL devices for 8080
[28-Sep-2009 13:25:12] <feutete> neet...I just remodeled the device again, and http-alt shows as Monitor=true
[28-Sep-2009 13:25:22] <feutete> even though I explicitly set it to false a few days ago
[28-Sep-2009 13:26:44] <rmatte> feutete: after you deleted the ipservice did you push the changes to the collector?
[28-Sep-2009 13:26:58] <rmatte> Manage -> Push Changes from the device page
[28-Sep-2009 13:27:10] <feutete> not explicitly, but the collector in this case is localhost
[28-Sep-2009 13:27:31] <feutete> shoudl I have to if the collector is localhost?
\n[28-Sep-2009 13:27:39] <rmatte> Also, Zenoss does a portscan, so\nif it thinks that something is listening on port 8080 then there almost\ncertainly is
[28-Sep-2009 13:27:54] <feutete> nmap shows 8080 is filtered
[28-Sep-2009 13:27:59] <feutete> can't connect to 8080 manually
\n[28-Sep-2009 13:28:19] <rmatte> feutete: you always have to\nexplicitly push changes even if the collector is localhost. The\ncollector only pulls config data for device from Zenoss once an hour
[28-Sep-2009 13:28:28] <rmatte> so you need to push changes to point out to it that something has changed
[28-Sep-2009 13:28:49] <rmatte> filtered doesn't necessarily mean closed
[28-Sep-2009 13:28:55] <feutete> ah, ok. I'll give that a spin
[28-Sep-2009 13:29:03] <rmatte> k
[28-Sep-2009 13:29:09] <feutete> thanks rmatte
[28-Sep-2009 13:29:11] <rmatte> np
[28-Sep-2009 13:29:39] <feutete> i assume the same is true if I set monitor=false on the IPService?
[28-Sep-2009 13:29:46] <rmatte> correct
[28-Sep-2009 13:29:47] <feutete> i need to push to the collector then as well?
[28-Sep-2009 13:29:48] <feutete> ok
\n[28-Sep-2009 13:30:14] <rmatte> the only thing you don't have to\npush changes on is if you were to change the production state of a\ndevice
[28-Sep-2009 13:30:36] <rmatte> basically\nall other changes need to be pushed, or you can optionally wait an hour\nfor the collector to automatically gather the monitoring config data
\n[28-Sep-2009 13:31:13] <feutete> seems like a good candidate for\na feature request...automatically push changes to the collector...
[28-Sep-2009 13:31:29] <__jd__> Hey all.. Question
..\nI'm experimenting with the XMLRPC/REST APIs for injecting events into\nZenOSS. The trouble I'm having is that the clear events (severity 0)\ndon't seem to be clearing the event from the console. Is there\nanything other than setting severity to 0 than I should be doing?[28-Sep-2009 13:31:50] <__jd__> And good to know about pushing the configs, didn't know that was necessary
\n[28-Sep-2009 13:34:25] <rmatte> feutete: it's meant to keep the\nload on the collector down, but you could request it anyways
[28-Sep-2009 13:35:02] <ckrough> ping jmp
[28-Sep-2009 13:35:50] <purevol> anyone ever get a cisco 2800 integrated services router working?
[28-Sep-2009 13:36:35] <purevol> __JD__ are you selecting the Event and moving it to history?
[28-Sep-2009 13:37:28] <rmatte> purevol: define "working"
\n[28-Sep-2009 13:38:25] <__jd__> purevol, I'm sending an XMLRPC\nevent to inject an error-level event into ZenOSS. Then, keeping all\nother info the same, I'm sending the same thing with severity 0\n(clear), in hopes that it will move the event to history
\n[28-Sep-2009 13:39:15] <feutete> rmatte: I just had another\nhttp-alt event show up, and http-alt isn't even listed under\nOS->IPServices. Any guess how I can get it to show up so I can set\nMonitor=false?
[28-Sep-2009 13:39:26] <purevol>\nyeah good point, i seem to be collecting call data, but i cannot\ncollect any data about my interfaces, environmentals, etc
\n[28-Sep-2009 13:40:31] <purevol> __jd__ : oh....im rusty with\nzenoss, havent used it in a bit, i dont recalls how to correlate and\nclear events like that
[28-Sep-2009 13:40:43] <purevol> i would imagine it needs to be deleted from the database
[28-Sep-2009 13:41:06] <rmatte> feutete: ssh in to the server, become the zenoss user, type zendmd to open zendmd
[28-Sep-2009 13:41:20] <rmatte> feutete: then type reindex() hit enter and then type commit() and hit enter
[28-Sep-2009 13:41:27] <rmatte> feutete: then type ctrl-d to exit
[28-Sep-2009 13:41:37] <rmatte> feutete: actually wait...
[28-Sep-2009 13:42:00]\n<rmatte> feutete: before you do that, stop zenoss for a minute,\nand copy $ZENHOME/var/Data.fs to $ZENHOME/var/Data.fs.bak
[28-Sep-2009 13:42:14] <feutete> ok
\n[28-Sep-2009 13:42:14] <rmatte> just in case the reindex screws\nanything up (it's really rare but I've seen it happen a couple of times)
[28-Sep-2009 13:42:20] <rmatte> then start zenoss back up
[28-Sep-2009 13:42:30] <rmatte> open zendmd, and type those commands it
[28-Sep-2009 13:42:32] <rmatte> in*
\n[28-Sep-2009 13:44:52] <rmatte> hmmm, zenprocess cycle time on\none of my boxes has been exceeded 9 times in a row now, odd
\n[28-Sep-2009 13:49:10] <purevol> so i have this 2851 that i can\nget call daya from, but i cant get any of the environmentals
[28-Sep-2009 13:49:37] <purevol> i installed your zenpack with all the cisco Mibs, but there is no 2851 mib/oid in there
[28-Sep-2009 13:51:03] <purevol> im not reawlly sure what the problem is
\n[28-Sep-2009 13:51:43] <purevol> i just built the appliance quick\nlike, so i cant do much witht the platform, maybe its that weird ssmi2\ndependency or something
[28-Sep-2009 13:52:13] <rmatte> Mibs are only for translating traps in Zenoss
[28-Sep-2009 13:52:23] <rmatte> They aren't used for polling data
[28-Sep-2009 13:52:41] <rmatte> so what specific data are you unable to poll?
[28-Sep-2009 13:53:00] <purevol> none of the environmentals
[28-Sep-2009 13:53:14] <rmatte> environmentals as in?
[28-Sep-2009 13:53:16] <purevol> cpu, interfaces, etc
[28-Sep-2009 13:53:19] <rmatte> ah
[28-Sep-2009 13:53:22] <purevol> routes
[28-Sep-2009 13:53:25] <rmatte> so you can't model the device at all essentially
[28-Sep-2009 13:53:31] <purevol> basically
[28-Sep-2009 13:53:37] <purevol> but i am receiving traps from it
[28-Sep-2009 13:53:41] <rmatte> well, as a starting point, if you snmpwalk the device do you see any of that data?
[28-Sep-2009 13:53:47] <rmatte> this is probably just an SNMP permissions issue
[28-Sep-2009 13:53:52] <purevol> i wonder if there is some hokey shit on cisco, i know the asa for example
[28-Sep-2009 13:55:25] <rmatte> asa is easy to fix in Zenoss
\n[28-Sep-2009 13:55:52] <rmatte> go to the zProperties and change\nzMaxOIDPerRequest from 40 to 20 for the specific ASA device
[28-Sep-2009 13:56:08] <purevol> no i get that weird smi error - http://pastebin.com/m537c0fc0
\n[28-Sep-2009 13:56:21] <gwb235> (potentail) resolution to our\nearlier problem (proxy timeouts) was to re-read the "using HTTPS/SSL\nfor the Zenoss Web UI" knowledge base and add some missing VirtualHost\nbits. We'll see if it comes back.
[28-Sep-2009 13:56:47] <purevol> i try v1 and i get , "End of MIB"
[28-Sep-2009 13:57:15] <purevol> rmatte: i already decreased that down to 10 actually
[28-Sep-2009 13:57:24] <purevol> my asa is working great
[28-Sep-2009 13:57:32] <purevol> its the 2851 i can get any data out of
[28-Sep-2009 13:57:52] <purevol> well, i see traps in event log, but, alas, cant poll jack
[28-Sep-2009 13:59:16] <rmatte> purevol: if you're just getting End of MIB then the SNMP isn't properly configured
\n[28-Sep-2009 13:59:28] <rmatte> you're either using the wrong\ncommunity string or the config on the device isn't setup properly
[28-Sep-2009 13:59:31] <rmatte> it's not a Zenoss issue
[28-Sep-2009 14:00:12]\n<rmatte> traps and polling are two separate things, it looks like\nyou don't have any access at all to the SNMP tree with the community\nstring that you're trying
[28-Sep-2009 14:04:16] <purevol> hrm....
[28-Sep-2009 14:07:13] <purevol> notice anything out of the ordinary? - http://pastebin.com/m2bcc5bcd
[28-Sep-2009 14:07:51] <rmatte> I see lots of trap lines but not much else, let me get you an example of what you should have
[28-Sep-2009 14:08:07] <purevol> there is not option for polling
[28-Sep-2009 14:08:19] <purevol> at least not that i can find
[28-Sep-2009 14:09:08]\n<rmatte> there's a config option that needs to be put in to allow\nSNMP access, you need to define community strings and the level of\naccess for each
[28-Sep-2009 14:09:15] <rmatte> all you've done is defined traps
[28-Sep-2009 14:09:43] <rmatte> I'm just trying to dig up an example for you... give me a second
[28-Sep-2009 14:09:52] <purevol> i think i found it
[28-Sep-2009 14:09:57] <purevol> snmp-server inform
[28-Sep-2009 14:10:34] <purevol> maybe i dunno
[28-Sep-2009 14:10:46] <purevol> yup that was it
[28-Sep-2009 14:10:49] <rmatte> ok first off you need a line like...
[28-Sep-2009 14:11:10] <purevol> my bad, gimpy figured it out
[28-Sep-2009 14:11:20] <rmatte> k
[28-Sep-2009 14:11:23] <purevol> i added the snmp communi, and snmp inform, and it worked
[28-Sep-2009 14:11:32] <rmatte> good
[28-Sep-2009 14:11:33] <rmatte> hehe
[28-Sep-2009 14:11:53] <purevol> will probably delete and re-add the device
[28-Sep-2009 14:12:04] <rmatte> you don't have to
[28-Sep-2009 14:12:09] <rmatte> just remodel, does the exact same thing
[28-Sep-2009 14:12:38] <rmatte> and rename it: Manage -> Rename Device from the device page
[28-Sep-2009 14:13:21] <purevol> well, if i delete and readd, all the device data will change
[28-Sep-2009 14:13:30] <rmatte> what do you mean?
[28-Sep-2009 14:13:34] <purevol> err performance data, evetns, that shit will get deleted
[28-Sep-2009 14:13:42] <rmatte> oh, yeh, that's correct
[28-Sep-2009 14:13:49] <rmatte> if that's what you want, then yeh
[28-Sep-2009 14:17:41] <purevol> we will see if i can get some perf data out of this cisco junk
\n[28-Sep-2009 14:18:14] <purevol> i dove into zenoss pretty hard a\ncouple years back, then i set it up last winter and handed it off to\nsomeone else
[28-Sep-2009 14:18:36] <purevol>\nnow im at this new gig, and they have cisco wroks, and i cant do\nanything with it, so ... here i am back to zenoss
[28-Sep-2009 14:18:44] <purevol> hoping the core works out alright
[28-Sep-2009 14:19:34] <rmatte> It should work fine
[28-Sep-2009 14:19:42] <rmatte> we're monitoring hundreds of cisco devices with it
[28-Sep-2009 14:20:10] <purevol> anything special, i need to setup to poll on a specificed basis?
[28-Sep-2009 14:20:40] <purevol> no graphs are popping up yet
[28-Sep-2009 14:21:21] <purevol> well some are working, but others arent, we will give it some time
[28-Sep-2009 14:24:16] <rmatte> it'll take 3 polling cycles for graphs to generate (about 15 to 20 mins)
[28-Sep-2009 14:24:40] <rmatte> the standard cisco template should be sufficient for starters
[28-Sep-2009 14:24:47] <rmatte> but you may want to customize it later on
\n[28-Sep-2009 14:25:44] <rmatte> hmmm weird, we have a director\nwhich is supposedly at 7.6G usage, but when I do a du -sh on it it\nshows up as 69 megs
[28-Sep-2009 14:25:51] <rmatte> directory*
[28-Sep-2009 14:26:10] <purevol> right on, hopefully, i can create some nifty stuff, and hook up a zenpack
[28-Sep-2009 14:26:17] <purevol> also got all this wireless tuff also
[28-Sep-2009 14:27:01]\n<rmatte> the cisco wireless stuff is quite fun, if it's the whole\nwireless controller and accesspoint setup where the controller\ngenerates the traps for every AP
[28-Sep-2009 14:27:33] <purevol> yes i have the WCS, two WLC, and a good amount of APs
[28-Sep-2009 14:27:37] <rmatte> since you have 1 device generating traps for potentially hundreds
[28-Sep-2009 14:27:59] <purevol> ill need to set it up
[28-Sep-2009 14:28:05] <rmatte> so you'll need to write a transform that displays the name of the IP which it's sending the trap for
[28-Sep-2009 14:28:14] <rmatte> I still need to write one for one of our Zenoss boxes
[28-Sep-2009 14:28:16] <purevol> its kinda dumb because the WCS is supposed to be a NMS
[28-Sep-2009 14:28:17] <rmatte> haven't gotten around to it yet
[28-Sep-2009 14:28:23] <purevol> and it has some cool stuff, but its weak
[28-Sep-2009 14:28:39] <rmatte> yeh, WCS is alright, but it's not the best
[28-Sep-2009 14:29:01] <purevol> i just updated it to 6.0, its pretty, and it has some cool stuff, but overall its weak
[28-Sep-2009 14:29:14] <purevol> because, none of it itnegrates into cisco wroks!
[28-Sep-2009 14:29:22] <purevol> and i cant type
[28-Sep-2009 14:29:29] <rmatte> lol
[28-Sep-2009 14:30:02]\n<rmatte> cisco works is great for writing and backing up device\nconfigs, but beyond that I don't really see it as a management tool
[28-Sep-2009 14:30:46] <rmatte> well, not a monitoring tool anyways
[28-Sep-2009 14:30:51] <rmatte> management, yes
[28-Sep-2009 14:31:14] <rmatte> we use an old piece of software called ZipTie to do that stuff
[28-Sep-2009 14:31:21] <rmatte> it's free and it works
[28-Sep-2009 14:32:43] <purevol> yeah you cant do any sort of monitoring with cisco works
[28-Sep-2009 14:32:52] <purevol> they have something , whats up gold
[28-Sep-2009 14:33:07] <rmatte> yeh, I've never used that but I've heard about it
[28-Sep-2009 14:33:29] <purevol> my "manager" is a micro management
[28-Sep-2009 14:33:47] <rmatte> yeh, I'm not a big fan of micro management
[28-Sep-2009 14:33:51] <rmatte> it's counter-productive
[28-Sep-2009 14:33:52] <purevol> nothing gets done
[28-Sep-2009 14:34:07] <purevol> but its cool, ill ust sit here and do nothing
[28-Sep-2009 14:34:16] <purevol> and wait for him to figure out zenoss
\n[28-Sep-2009 14:34:23] <rmatte> they spend all their time running\naround interviewing people about what they are currently doing and\nnothing gets done because of it
[28-Sep-2009 14:34:46] <rmatte> you have to put a certain level of trust in your employees
\n[28-Sep-2009 14:35:20] <purevol> actually, im taking a break from\nbeing a high level telecommnuication engineer, and took this network\nadmin job, because its like, doesnt require me to think or write\ndocumentation or manage people all day
[28-Sep-2009 14:35:36] <purevol> so, i just find this shit trite or humorous, depending on my mood
[28-Sep-2009 14:35:45] <rmatte> hehe
[28-Sep-2009 14:36:00] <purevol> i probably should set up some time with the CIO and tell him what i think
[28-Sep-2009 14:36:00] <rmatte> I took over this job from being a senior network analyst, so it's a step up for me
[28-Sep-2009 14:36:06] <purevol> but i dont want to rip everyone
[28-Sep-2009 14:36:14] <purevol> ive only been here for like 2 months
[28-Sep-2009 14:36:26] <rmatte> yeh, you might want to keep your head down then for a while
\n[28-Sep-2009 14:36:52] <purevol> i dont particularly give a shit,\nim rather good at the politics, its what you get working for the CEO of\nthe largest ISP in the US
[28-Sep-2009 14:36:58]\n<davetoo> So, who's an event/transform wizard in here? What's\nthe best way, from within a transform, to get a count of current events\nof all severities for the device that triggered the current event?
[28-Sep-2009 14:37:51] <davetoo> or at least the highest severity active event?
[28-Sep-2009 14:39:03]\n<rmatte> davetoo: well, I'm sure it's possible but transforms are\nreally designed to pertain to the current event, not to gather data\nfrom other open events
[28-Sep-2009 14:39:27] <purevol> davetoo: are you trying to correlate events?
\n[28-Sep-2009 14:39:43] <davetoo> well what I want to do is\nsupress the current event if there are already other active events at\nthe same severity
[28-Sep-2009 14:39:59] <davetoo> I'm trying to mitigate event/alert floods
[28-Sep-2009 14:40:08] <purevol> event correlation
[28-Sep-2009 14:40:35] <davetoo> My plan is to generate a new event to show that there are auto-suppressed events
[28-Sep-2009 14:40:42] <rmatte> hmmm, I don't know off the top of my head
\n[28-Sep-2009 14:42:22] <rmatte> you'd have to find whatever\nfunction is used to comb through the event console, something like:
[28-Sep-2009 14:42:29] <rmatte> for event in <function>:
[28-Sep-2009 14:42:42] <rmatte> if event.severity == evt.severity:
[28-Sep-2009 14:42:45] <rmatte> and then do whatever
[28-Sep-2009 14:43:05] <rmatte> You'd probably also have to put in:
[28-Sep-2009 14:43:05] <rmatte> #!/usr/bin/env python
[28-Sep-2009 14:43:05] <rmatte> import Globals
[28-Sep-2009 14:43:05] <rmatte> from Products.ZenUtils.ZenScriptBase import ZenScriptBase
[28-Sep-2009 14:43:05] <rmatte> from transaction import commit
[28-Sep-2009 14:43:06] <rmatte> dmd = ZenScriptBase(connect=True).dmd
[28-Sep-2009 14:43:12] <rmatte> at the top of the transform for it to work properly
[28-Sep-2009 14:43:33] <davetoo> actually I've found that I can grab the dmd from the event device entry
[28-Sep-2009 14:44:00] <rmatte> you need a global dmd in this case though
[28-Sep-2009 14:44:31] <rmatte> anyways, that's sort of how you'd do it
[28-Sep-2009 14:44:37] <rmatte> just need to find the right function
[28-Sep-2009 14:45:36] <davetoo> yeah, 'k
[28-Sep-2009 14:45:50] <davetoo> I was doing some stuff over the weekend but lost my work
[28-Sep-2009 14:50:23] <purevol> i wonder if i should remove the baffling from the exhaust on my ducati
\n[28-Sep-2009 14:54:58] <eidolon> hey folks, i'm in a weird\nsituation with a possible broken zenoss enterprise setup. I am getting\n-zero- notifications coming out of it. i set up a suer with a\n'anything in production >= error, send mail' - and I get zilcho. \ni'm looking in Z2.log for warnings or anything, nada. any suggestions?
[28-Sep-2009 14:55:07] <rmatte> no, you should ship that ducati over to me
[28-Sep-2009 14:55:11] <eidolon> test mail to the target account in settings->user does generate mail okay.
[28-Sep-2009 14:55:36] <rmatte> eidolon: for starters, as zenoss user: zenoss status
[28-Sep-2009 14:55:41] <rmatte> are all of the daemons running?
[28-Sep-2009 14:55:57] <purevol> sorry man, no shipping
\n[28-Sep-2009 14:56:21] <eidolon> rmatte: yes, all running. i did\na zenoss restart a few hours ago hoping that might be it
[28-Sep-2009 14:57:53] <eidolon> rmatte: next?
[28-Sep-2009 14:58:18] <rmatte> you mentioned Z2.log, have you checked zenhub.log?
[28-Sep-2009 14:58:36] <eidolon> zenhub.log is empty, dated May 10th.
[28-Sep-2009 14:58:57] <eidolon> doh
[28-Sep-2009 14:58:58] <eidolon> wait
[28-Sep-2009 14:58:59] <rmatte> if you perform a: zenhub run -v10
[28-Sep-2009 14:59:03] <purevol> i wonder if i can use cisco works to turn off telnet and turn on ssh on about 100 ethernet switches
[28-Sep-2009 14:59:03] <rmatte> ?
[28-Sep-2009 14:59:04] <eidolon> nonon.. hang on.
[28-Sep-2009 14:59:18] <rmatte> I can't wait to hear what the issue was lol
\n[28-Sep-2009 14:59:25] <eidolon> there's a\n/opt/zenoss/log/zenhub.log (empty). and /opt/zenoss/log/localhost/ -\nand that localhost dir has a ton of log files i didn't notice
[28-Sep-2009 14:59:28] <rmatte> purevol: yeh, you can
[28-Sep-2009 14:59:38] <rmatte> purevol: it does it all via SNMP as far as I'm aware
[28-Sep-2009 14:59:52] <eidolon> all sorts of chatter there. let me peruse. give me a few minutes.
[28-Sep-2009 14:59:57] <eidolon> (actually i need to move my laptop. brb.
[28-Sep-2009 15:00:02] <rmatte> k
[28-Sep-2009 15:00:13] <rmatte> I have to leave my desk for a few but I'll be back
[28-Sep-2009 15:00:31] <purevol> yeah, i think you can create a JOB
[28-Sep-2009 15:00:44] <purevol> but the only concern i have is generating the RSA key on every single device
[28-Sep-2009 15:02:44] <purevol> these people i work for, bought a cisco nac appliance 2 years ago, and never deployed it
[28-Sep-2009 15:11:40] <eidolon> okay, back.
[28-Sep-2009 15:23:27] <rmatte> purevol: generating the rsa key is just 1 more command
\n[28-Sep-2009 15:26:18] <rmatte> ideally what you'd want to do is\nenable ssh on each device without disabling telnet, then have some\nscript that telnets in to each device and does:
[28-Sep-2009 15:26:20] <rmatte> conf t
[28-Sep-2009 15:26:27] <rmatte> crypto key generate rsa
[28-Sep-2009 15:26:29] <rmatte> 1024
[28-Sep-2009 15:26:35] <rmatte> <enter>
[28-Sep-2009 15:26:38] * eidolon opens a ticket with zenoss enterprise support and calls it art.
[28-Sep-2009 15:26:42] <rmatte> end
[28-Sep-2009 15:26:43] <rmatte> end
[28-Sep-2009 15:26:45] <rmatte> wr mem
[28-Sep-2009 15:26:45] <rmatte> exit
[28-Sep-2009 15:27:07] <rmatte> then once that's done you disable telnet on all of the devices, and you're done
\n[28-Sep-2009 15:27:43] <rmatte> you'd obviously want to run some\nsort of check on each device to make sure that the telnet port is\nclosed and the ssh port is open afterwards
\n[28-Sep-2009 15:36:34] <mrayzenoss> I assume this will be useful:\nblogs/zenossblog/2009/09/28/tip-of-the-month-zendmd-tab-completion
[28-Sep-2009 15:38:41] <rmatte> yeh, I noticed that in King Crab, very cool
[28-Sep-2009 15:39:24] <rmatte> I might apply that during our change window tomorrow night
\n[28-Sep-2009 15:41:50] <rmatte> I also noticed that it stops\ncomplaining about 2 versions of Five in King Crab, though I suspect\nthat it might complain after doing an actual upgrade from 2.4.5 to 2.5
[28-Sep-2009 15:42:31] <mrayzenoss> that may be related to the Zope 2.11 ugprade
[28-Sep-2009 15:43:04] <rmatte> 2009-09-28 16:42:42 WARNING OFS.Application Duplicate Product name
[28-Sep-2009 15:43:04] <rmatte> After loading Product 'Five' from '/usr/local/zenoss/zenoss/Products',
[28-Sep-2009 15:43:04] <rmatte> I skipped the one in '/usr/local/zenoss/zenoss/lib/python/Products'.
[28-Sep-2009 15:43:27] <rmatte> been seeing that ever since the upgrade from 2.3 to 2.4
[28-Sep-2009 15:43:43] <rmatte> haven't tried to fix it since it doesn't appear to hurt anything
[28-Sep-2009 16:14:05] <cgibbons> hmmm
[28-Sep-2009 16:27:00] <rmatte> I love how I put a thumbs up review for something on sourceforge and it counts it as a thumbs down
[28-Sep-2009 16:27:04] <rmatte> that's really accurate
[28-Sep-2009 16:27:12] <mrayzenoss> seriously?
[28-Sep-2009 16:27:18] <mrayzenoss> their stuff is flakey
[28-Sep-2009 16:27:20] <rmatte> https://sourceforge.net/projects/zapplet/
[28-Sep-2009 16:27:22] <rmatte> check that
[28-Sep-2009 16:27:29] <rmatte> I just posted a review with a thumbs up
[28-Sep-2009 16:27:36] <rmatte> and I see it as a thumbs down
[28-Sep-2009 16:27:41] <rhettardo> semantics
[28-Sep-2009 16:27:44] <rmatte> 0% of 1 reviewer recommends this project
[28-Sep-2009 16:27:50] <rmatte> Thumbs down:
[28-Sep-2009 16:27:51] <rmatte> 1
[28-Sep-2009 16:27:57] <mrayzenoss> for me it shows 2 up
[28-Sep-2009 16:28:02] <rmatte> really?
[28-Sep-2009 16:28:02] <mrayzenoss> are you rawdmon?
[28-Sep-2009 16:28:05] <rmatte> yeh
[28-Sep-2009 16:28:10] <mrayzenoss> yeah, shows you as thumbs up
[28-Sep-2009 16:28:12] <rmatte> hmmm, maybe it's my browser then, weird
[28-Sep-2009 16:28:20] <mrayzenoss> nah, their stuff is flakey
[28-Sep-2009 16:28:30] <rhettardo> mine shows no reviews
[28-Sep-2009 16:28:33] <rmatte> hehe
[28-Sep-2009 16:28:41] <mrayzenoss> maybe you can't see yourself?
[28-Sep-2009 16:28:42] <rmatte> so we all see something different when we're looking at the same page
[28-Sep-2009 16:28:46] <rmatte> that's so excellent
[28-Sep-2009 16:28:50] <mrayzenoss> I complained to SourceForge this weekend
[28-Sep-2009 16:28:51] <rhettardo> lol
[28-Sep-2009 16:28:54] <mrayzenoss> in person
[28-Sep-2009 16:29:03] <rmatte> hehe
[28-Sep-2009 16:29:25] <rhettardo> did you tell them to stop making their website worse
[28-Sep-2009 16:29:30] <rmatte> I seem to recall when sourceforge worked a lot better
[28-Sep-2009 16:29:43] <rmatte> it seems to blow badly ever since they've been updating the look lately
[28-Sep-2009 16:29:51] <mrayzenoss> rhettardo: I did indeed
[28-Sep-2009 16:30:31] <rhettardo> it worked perfectly fine w/ the design 3 revisions ago
[28-Sep-2009 16:31:34] <rmatte> wow, I click "Show next 10 reviews" and the link just dissapears and nothing happens
[28-Sep-2009 16:31:37] <rmatte> flakey indeed
[28-Sep-2009 16:32:09] <rmatte> I miss the old red "I mean business" style design
[28-Sep-2009 16:32:16] <rmatte> with just a red top bar
[28-Sep-2009 16:32:38] <rhettardo> yeah
\n[28-Sep-2009 16:33:26] <rmatte> I remember back when I was like\n14 or 15 and started learning FreeBSD, sourceforge was mint back then,\nno bugs
[28-Sep-2009 16:33:51] <rmatte> are they using Jive or something now?
[28-Sep-2009 16:33:54] <rmatte> it looks very similar
[28-Sep-2009 16:34:07] <mrayzenoss> no, it's all custom
[28-Sep-2009 16:34:12] <mrayzenoss> they write their own stuff
[28-Sep-2009 16:34:28] <rmatte> ah
[28-Sep-2009 16:34:37] <rmatte> well the layout is definitely partially ripped off from jive
[28-Sep-2009 16:34:38] <rmatte> lol
[28-Sep-2009 16:35:15] <rmatte> If you put the Zenoss community site up beside it you can see the similarities
[28-Sep-2009 16:36:57] <rmatte> Hmmm, under "Support" on Sourceforge, you don't have the IRC channel listed
[28-Sep-2009 16:37:39] <mrayzenoss> yeah, you get 1 choice from a radio button list
[28-Sep-2009 16:37:43] <mrayzenoss> kinda weak
[28-Sep-2009 16:37:52] <mrayzenoss> I choose "URL"
[28-Sep-2009 16:37:56] <rmatte> ah
[28-Sep-2009 16:38:02] <rmatte> yeh, that is pretty weak
\n[28-Sep-2009 16:38:58] <rmatte> you guys might want to look in to\nsetting up some sort of flash IRC client on the community site which\nconnects to here
[28-Sep-2009 16:39:19] <rmatte> this place is quite honestly the most helpful support resource in my opinion
[28-Sep-2009 16:39:25] <rmatte> forums are good, but this is better
\n[28-Sep-2009 16:40:34] <rmatte> put a link to it on the main page\nthat says "Join us on IRC!" and if they click on it it opens a page\nexplaining that they can either download an IRC client and connect to\nirc.freenode.net channel #zenoss or that they can "Click here" to open\nthe flash client
[28-Sep-2009 16:40:36] <rmatte> just an idea
\n[28-Sep-2009 16:41:30] <rmatte> some people aren't used to\nworking with IRC clients, so it would make things simpler for them
[28-Sep-2009 16:41:46] <rmatte> or maybe they are just lazy and don't want to install anything
[28-Sep-2009 16:41:47] <rmatte>
[28-Sep-2009 16:42:12] <rmatte> any progress on recovering the lost code in the forums?
[28-Sep-2009 16:43:12] <davetoo> rmatte: about my earlier question
[28-Sep-2009 16:43:24] <davetoo> much easier than I thought
[28-Sep-2009 16:43:31] <rmatte> oh?
[28-Sep-2009 16:43:37] <davetoo> dev.dmd works great
[28-Sep-2009 16:43:39] <rmatte> how'd you go about doing it?
[28-Sep-2009 16:43:49] <rmatte> ah
[28-Sep-2009 16:43:57] <rmatte> cool
[28-Sep-2009 16:44:37] <rmatte> well, an hour an a half until the auction for a paintball gun that I want is over...
[28-Sep-2009 16:44:42] * rmatte taps his foot
[28-Sep-2009 16:45:06]\n<davetoo> dev.dmd.ZenEventManager.getEventSummaryME(dev) gets me\nthe count of ACKed and unACKed events at all five severities
[28-Sep-2009 16:45:06] <rmatte> I hope no one else bids because I have it set to auto-bid up, but I'd like to get it for the current bid price
[28-Sep-2009 16:45:22] <rmatte> nice
\n[28-Sep-2009 16:45:45] <rmatte> I need to figure out how to do an\nxmlrpc call in perl to Zenoss to have it move an event to history
[28-Sep-2009 16:45:48] <rmatte> should be fun
[28-Sep-2009 16:45:56] <davetoo> gotta get a bite
[28-Sep-2009 16:46:02] <rmatte> k
\n[28-Sep-2009 16:47:24] <Brad_K> anybody else getting munged rrd\ndate/time stamps on their performance graphs after changing the system\nto proxy the Collector Render URLs via SSL on the hub?
[28-Sep-2009 16:49:00] <rmatte> We haven't proxied any of ours via SSL, so I couldn't tell you
\n[28-Sep-2009 16:52:12] <mrayzenoss> davetoo: did you catch the\ntip of the month? \nblogs/zenossblog/2009/09/28/tip-of-the-month-zendmd-tab-completion
[28-Sep-2009 16:52:41] <mrayzenoss> rmatte: the lost code is still being ironed out in the staging server
[28-Sep-2009 16:53:22] <davetoo> just barely
\n[28-Sep-2009 16:53:24] <mrayzenoss> rmatte: we fixed a lot of\nlittle issues today, our sysadmins were kinda busy the last week\nrelaunching the zenoss.com site
[28-Sep-2009 16:53:37] <davetoo> should be nice; have to see how it's done
[28-Sep-2009 16:53:48] <rmatte> mrayzenoss: cool
[28-Sep-2009 16:54:26] <mrayzenoss> rmatte: and dealing with SourceForge nonsense
[28-Sep-2009 16:55:10] <rmatte> hehe, I can imagine
[28-Sep-2009 17:31:01] <davetoo> rgh
[28-Sep-2009 17:31:15] <davetoo> why is the window to edit the event transform so dinky?
[28-Sep-2009 18:11:13] <rmatte> davetoo: edit it outside that window then paste it in, not overly difficult
[28-Sep-2009 18:11:33] <rmatte> It's not meant to be a full blown editor
[28-Sep-2009 18:11:45] <rmatte> maybe it will be in the future
[29-Sep-2009 00:00:46] [disconnected at Tue Sep 29 00:00:46 2009]
[29-Sep-2009 00:00:46] [connected at Tue Sep 29 00:00:46 2009]
[29-Sep-2009 00:00:56] [zenoss-logger (logger bot) has joined #zenoss]
[29-Sep-2009 08:32:03] <cgibbons_> bouncy
[29-Sep-2009 08:59:02] <ckrough> how can I go back and see when a device was put into maintenance
[29-Sep-2009 09:02:49] <ckrough> nm, I got a log
[29-Sep-2009 09:38:59] <rmatte> good morning folks
[29-Sep-2009 09:39:22] <ckrough> hey
[29-Sep-2009 09:40:02] <ckrough> rmatte: you doing windows monitoring via wmi?
[29-Sep-2009 09:40:32] <rmatte> SNMP, the only thing we ever monitor via WMI is services
[29-Sep-2009 09:40:53] <ckrough> ah, helping a buddy with WMI only and we arent seeing system perf stats
[29-Sep-2009 09:41:00] <ckrough> gotta check my bindings again
[29-Sep-2009 09:41:22] <ckrough> got Egor's zenpacks in there and the perfmon template
[29-Sep-2009 09:41:31] <rmatte> you can try running the queries manually via wmic and see if you get an error
[29-Sep-2009 09:41:38] <rmatte> or just run a test on the datapoint
[29-Sep-2009 09:42:03] <rmatte> by the way, Egor's ZenPacks break in King Crab, so take that in to account when it comes time to upgrade
[29-Sep-2009 09:42:10] <ckrough> will do. thanks
[29-Sep-2009 09:42:20] <rmatte> I believe he's working on a new version that will work with King Crab
[29-Sep-2009 09:42:32] <ckrough> thats going to be a tough spot for the zenoss community over time.
[29-Sep-2009 09:42:38] <rmatte> yeh
\n[29-Sep-2009 09:42:58] <ckrough> when developers abandon zenpacks\nand there are incompatible upgrades. Hopefully there will be enough\ncommunity that someone picks them up
[29-Sep-2009 09:43:06] <rmatte> I'm thinking a lot of stuff will break when they do the zope and python upgrades
[29-Sep-2009 09:43:08] <ckrough> but I guess thats the nature of open source, if they are worth it someone will take them
[29-Sep-2009 09:43:14] <rmatte> I believe the zope upgrade is planned for stone crab
[29-Sep-2009 09:43:25] <rmatte> actually, python as well I believe
[29-Sep-2009 09:43:31] <rmatte> so that's going to be a fun release
[29-Sep-2009 09:43:34] <ckrough> yeah
[29-Sep-2009 09:43:44] <ckrough> we've got so many monkey patches in here now...
\n[29-Sep-2009 09:43:45] <rmatte> whole new UI should be\nimplemented on that release though, so it's going to be uber spiffy
[29-Sep-2009 09:43:48] * ckrough scared
[29-Sep-2009 09:43:51] <rmatte> hehe
[29-Sep-2009 09:44:04] <rmatte> all I can say is two words
[29-Sep-2009 09:44:05] <rmatte> lab box
\n[29-Sep-2009 09:44:45] <rmatte> After the fun that we went\nthrough with the 2.4.3, 2.4.4, and 2.4.5 upgrades we're going to be\ndoing rigorous lab testing next time around to work out any issues\nprior to upgrading our prod boxes
[29-Sep-2009 09:44:52] <ckrough> heh. got a staging config and a dev config
[29-Sep-2009 09:45:14] <rmatte> we just have a "Client base image" that we load up, it's pretty nice
\n[29-Sep-2009 09:45:31] <rmatte> we just clone the VM, boot it up,\nswitch the IPs, hostname, edit the hosts file, and we're up and running
[29-Sep-2009 09:45:38] <rmatte> takes like 15 mins to get a new server online
\n[29-Sep-2009 09:46:42] <rmatte> w00t, meeting in 15 mins to\ndiscuss how we're going to have our ticketing system integration\nauto-historify Zenoss events after it opens tickets
[29-Sep-2009 09:47:24] <rmatte> I've found the xmlrpc calls in python, just need to figure out how to do it in perl
\n[29-Sep-2009 09:47:27] <Brad_K> you know, i don't really care\nabout uber-spiffy UIs. I care more about finding and fixing bugs,\nespecially long-standing bugs that have been around since 2.3.3.
[29-Sep-2009 09:47:37] <rmatte> oh, I agree
[29-Sep-2009 09:48:04]\n<rmatte> In my opinion, I would have been good with the current\nUI for another 2 years if it meant that the actual features would get\nricher and less buggy
[29-Sep-2009 09:48:05]\n<Brad_K> i mean, that's what we're supposed to be paying for,\nright? i mean, this is supposed to be an Enterprise Monitoring system,\nright?
[29-Sep-2009 09:48:37] <rmatte> but you\nhave to remember that this is also a company's product, the new UI is\nmore of a marketing move than anything
[29-Sep-2009 09:48:42] <Brad_K> we need stability and painless backups/restores and upgrades.
\n[29-Sep-2009 09:48:50] <ckrough> We are hitting the end of\nscalability with the current archetecture too. Im firing up my 11th\ncollector for one installation now
[29-Sep-2009\n09:48:55] <rmatte> then again the new UI looks sexy as heck, and\nI'm a sucker for eye-candy, so it won't be all bad
[29-Sep-2009 09:48:55] <ckrough> thats too many uber-boxes
[29-Sep-2009 09:49:08] <ckrough> each box is 32g ram, yada yada. big hardware
[29-Sep-2009 09:49:23] * ckrough architecture
\n[29-Sep-2009 09:49:39] <Brad_K> and the more i get into this, the\nless it seems that anyone is paying attention to that critical core\ninfrastructure that isn't sexy, but without which everything else is\nworthless.
[29-Sep-2009 09:49:40] <rmatte> yeh, we have 3 boxes with 32g each, and we're just starting out
[29-Sep-2009 09:50:14] <rmatte> Brad_K: I agree, but hopefully once the UI is done they'll get back to that
\n[29-Sep-2009 09:50:18] <Brad_K> we started out with four boxes\nwith 16GB each, one as hub and the other three as collectors. so far,\nwe're really only using the one hub and one of the collectors.
\n[29-Sep-2009 09:50:21] <ckrough> I think a lot of the improvement\nare driven by the community, which is typically smaller installations\nwith only a couple users, so it drives UI flash instead of foundation\nbuilding
[29-Sep-2009 09:50:41] <Brad_K> rmatte:\ni have not been given any reason to believe that they will ever pay\nattention to anything but the sexy UI stuff.
[29-Sep-2009 09:50:43] <rmatte> Brad_K: ah, we have separate instances running on VMs on the boxes
[29-Sep-2009 09:51:10] * eidolon opens up a ticket with support at zenoss.
[29-Sep-2009 09:51:23] <rmatte> hehe
\n[29-Sep-2009 09:51:44] <rmatte> Brad_K: I doubt that's true, the\nUI is just a big project for them right now, so a lot of resources are\nbeing devoted to it
[29-Sep-2009 09:52:04]\n<Brad_K> all my research so far indicates that backups are\nproblematical, upgrades are problematical, and most of the toolkit only\nkinda-semi-sorta works. if you can do what you want/need to do with\njust SNMP, then you should be fine. otherwise, there are rough seas\nahead.
[29-Sep-2009 09:52:20] <eidolon> rmatte:\nthanks for your help yesterday. i had rusty and (nick?) on the phone\nfor a while, they helped a ton. now i'm trying to fix the\nno-alerts-going-out problem on the old install.
[29-Sep-2009 09:52:34] * eidolon is a tech support departments thorn-in-the-side.
\n[29-Sep-2009 09:52:51] <Brad_K> and that's for people who are\nPython/Zope/Zenoss system developers. if you're not a system\ndeveloper, then it seems to me that you're going to be poorly prepared\nfor those rough seas.
[29-Sep-2009 09:53:34]\n<rmatte> Brad_K: I'm not a "system developer", but I've learned\nenough along the way to be able to get Zenoss to do basically anything\nvia custom scripts
[29-Sep-2009 09:54:35]\n<rmatte> Brad_K: this is not a product that is designed to be\nmanaged by your friendly neighbourhood LAN admin. You need\nsysadmin/scripting experience as well as a knowledge of SNMP/WMI,\nnetworking, and servers.
[29-Sep-2009 09:54:37]\n<Brad_K> rmatte: so you've hacked on it enough to make it\nZenrmatte at your site, as opposed to Zenoss.
[29-Sep-2009 09:54:46] <ckrough> ^this
\n[29-Sep-2009 09:55:19] <ckrough> same boat we are in. we are in\nthe process of trying to commit codebase changes to get our non-zenpack\ncustomizations rolled in
[29-Sep-2009 09:55:23]\n<rmatte> Brad_K: It's not even "hacking on it", you design a\nscript, which is essentially a plugin, plug it in as a command based\ndatapoint (which is very simple to do), and off you go
[29-Sep-2009 09:56:03]\n<Brad_K> i've got 20 years of sysadmin experience, and SNMP is\nthe least of my worries. I don't know much about WMI, but we've got\nWindows people who do. But, most critically, i'm not a system\nprogrammer.
[29-Sep-2009 09:56:16] <Brad_K> except zencommand is b0rken.
[29-Sep-2009 09:56:25] <rmatte> how is it broken?
[29-Sep-2009 09:56:38] <rmatte> a lot of our monitoring depends on zencommand and it works perfectly fine
[29-Sep-2009 09:56:57] <rmatte> I have 5 months worth of performance graphs populated via custom scripts to prove it
[29-Sep-2009 09:57:01]\n<Brad_K> sometimes stuff runs fine for a while, then it stops\nrunning. then it might start running fine again a few days later.
\n[29-Sep-2009 09:57:22] <rmatte> That seems to be a problem that\nonly you're experiencing for whatever reason, my zencommand runs\nabsolutely flawless
[29-Sep-2009 09:57:31] <Brad_K> rmatte: and which version of Zenoss are you on?
[29-Sep-2009 09:57:51] <rmatte> 2.4.5 at the moment, but it ran flawless in 2.4.2, 2.4.1, and 2.3.3
[29-Sep-2009 09:57:58] <rmatte> I haven't had any issues with it along the way
\n[29-Sep-2009 09:59:55] <Brad_K> the Zenoss ssh stuff doesn't\nproperly close the connections, so you get massive amounts of error log\nspam on the monitored servers. so, we use the tried-and-tested Nagios\ncheck_ssh instead. the Zenoss SSL checking stuff will generate errors\nfor SSL certificates expiring on each of 30 days, 14 days, 7 days, 3\ndays, and 1 day, as opposed to just generating one error for one of\nthose timeframes. we don't need seven alerts for the same error.
[29-Sep-2009 10:00:35]\n<rmatte> it's not supposed to close the connections, the\nconnection are designed to stay open. think about it, if it logged in\nevery 5 minutes it would make your auth logging for SSH absolutely\nuseless because it would flood the log out
[29-Sep-2009 10:01:23]\n<Brad_K> the ssh checking doesn't stay logged in. it opens the\nconnection, sees that the port is being listened to, and then drops the\nconnection without properly closing it.
[29-Sep-2009 10:01:38] <rmatte> oh, you're talking about IpServices
[29-Sep-2009 10:01:40] <Brad_K> maybe this is a bug in the 2.4.2 version.
[29-Sep-2009 10:01:53] <rmatte> I've never noticed that issue, and we monitor SSH on a lot of servers
[29-Sep-2009 10:02:04] <rmatte> There was a bug with that a long while back, but it was fixed
[29-Sep-2009 10:02:31] <rmatte> anyways, I have a meeting to head off to, I'll be back in a bit
[29-Sep-2009 10:03:10]\n<rmatte> (If you are noticing these issues you might want to dig\na bit deeper, if possible, and open some trac tickets
[29-Sep-2009 10:03:36] <rmatte> You might also try a fresh 2.4.5 install and see if that solves your problems
[29-Sep-2009 10:04:23]\n<Brad_K> it seems to me that wer're getting some of the worst\naspects of open source (where people only work on the things that they\nfind sexy and the only support you get is on the IRC channel or the\nmailing lists) and the worst aspects of closed source.
[29-Sep-2009 10:05:10]\n<Brad_K> we've got a number of support cases that we've opened,\nand we're going to be doing a WebEx session in about thirty minutes on\none of those. we're going to be opening a lot more support cases as\nwell.
[29-Sep-2009 10:06:24] <eidolon> Brad_K: this is why we went with an enterprise support contract with zenoss.
[29-Sep-2009 10:06:31] <eidolon> they've been -awesome- support for us.
[29-Sep-2009 10:07:00]\n<Brad_K> eidolon: we've got an enterprise support contract with\nthem. and the best support we get is from this IRC channel, mostly\nfrom rmatte.
[29-Sep-2009 10:07:43] <eidolon>\nand, apropos of nothing, what Windows determines is a system-level\n"Warning" is really irritating. "Fine. You disabled a printer. \nThat's very nice for a production deployed coloed Win2k8 machine. \nSTFU!"
[29-Sep-2009 10:07:58] <eidolon> Brad_K: that's not been my experience, but okay
[29-Sep-2009 10:08:26]\n<eidolon> if you have an emergency / real problem, i tend to open\na case, and if there's no response within an hour or so, i call them\ndirectly.
[29-Sep-2009 10:08:26] <Brad_K> it\ntakes several hours, or days, for them to respond to support cases that\nare filed through the official SugarCRM portal. at least on irc,\nthere's a chance that you might catch one of the developers, or be\nlucky enough that rmatte may be around and have already run into the\nsame problem.
[29-Sep-2009 10:08:37] <eidolon> i've been on the phone with rusty 2-3 times in the last month, and he's -awesome-.
[29-Sep-2009 10:08:53] <eidolon> Brad_K: if you're seeing days response time on the CRM requests, then you raise high holy hell to rusty.
[29-Sep-2009 10:09:06] <eidolon> because your SLA says you should get responses a lot faster.
\n[29-Sep-2009 10:13:33] <ckrough> rmatte: you should be seeing\nlots of these: " sshd[6963]: Did not receive identification string from\nyour.zenoss.box"
[29-Sep-2009 10:13:44] <ckrough> INFO level in syslog
[29-Sep-2009 10:14:06] <ckrough> We gotten good support through the enterprise license
[29-Sep-2009 10:14:22] <ckrough> I think IRC is a better forum for the customer though, since it's interactive
[29-Sep-2009 10:14:41] <ckrough> bouncing tickets back and forth in sugarcrm is never going to be as responsive as IRC
[29-Sep-2009 10:15:00] <ckrough> but overall Im very satisfied with ent support
[29-Sep-2009 10:15:37] <eidolon> well, there's only so many questions rmatte can answer
[29-Sep-2009 10:15:43] <ckrough> I usually try to convince them to PRIVMSG me in IRC when we start working a ticket
[29-Sep-2009 10:16:23] <ckrough> too bad matt isnt here, he should be noting all of this
[29-Sep-2009 10:16:36] <ckrough> zenoss-logger: you getting all this
\n[29-Sep-2009 10:16:55] <eidolon> well, i'll throw this one out. \ni ahve a zenoss install that is catching events, and showing them in\nteh console, but though i have a group rule that should notify via\nemail when something => warning is hitting, we're not seeing any\nmail.
[29-Sep-2009 10:17:03] <Brad_K> maybe i'm just frustrated because of all the problems we've been having lately.
[29-Sep-2009 10:17:05] <eidolon> test mail works. zenoss status shows everything running.
[29-Sep-2009 10:17:13] <Brad_K> maybe you shouldn't pay too much attention to me right now.
[29-Sep-2009 10:17:15] <eidolon> i opened a case on this already
[29-Sep-2009 10:17:20] * eidolon ignores Brad_K. *nod*
\n[29-Sep-2009 10:17:53] <ckrough> Brad_K: I can relate to where\nyou are coming from, there is definitely some roughness to the product,\nand some areas that make me nervous about the future
[29-Sep-2009 10:18:05] <ckrough> but on the flipside, it's the best solution for what we need to do here
[29-Sep-2009 10:19:00]\n<exarkun> I'm looking for the part of the zenoss code that...\ndoes things. I see a bundled version of Python, a bundled version of\nZope, some tiny scripts, but I don't see anything that looks like it\ncan actually monitor anything. Can anyone give me a pointer?
[29-Sep-2009 10:19:46] <ckrough> exarkun: its a colletion of processes
[29-Sep-2009 10:19:49] <cgibbons> that's a bit open-ended
[29-Sep-2009 10:20:04] <eidolon> that's zenoss
[29-Sep-2009 10:20:12] <eidolon> zenoss is about 18 or so programs that work together
[29-Sep-2009 10:20:24] <eidolon> follow the installation process.
\n[29-Sep-2009 10:20:31] <Brad_K> and no, i've decided that i'm not\ngoing to go vent my frustrations on some sort of public blog. i guess\ni'm just going to keep beating my head against the wall for a while\nlonger....
[29-Sep-2009 10:20:33] <ckrough> exarkun: the documentation reviews the roles I believe
[29-Sep-2009 10:20:43] <exarkun> ckrough: There's a lot of documentation
[29-Sep-2009 10:20:49] <exarkun> ckrough: Any in particular?
[29-Sep-2009 10:20:53] <ckrough> sec
[29-Sep-2009 10:21:00] <eidolon> how about "Introduction to zenoss : getting started"
[29-Sep-2009 10:21:16] <eidolon> are you looking for details on how it works? I recommend setting it up and watching it.
[29-Sep-2009 10:21:16] <eidolon> otherwise the concepts won't help much.
[29-Sep-2009 10:21:17] <ckrough> start here docs/DOC-3061
[29-Sep-2009 10:21:26] <eidolon> there yago.
[29-Sep-2009 10:21:30] <eidolon> ~ckrough++
[29-Sep-2009 10:22:39] <exarkun> eidolon: I did that a bit. I'm trying to learn enough to add some features.
[29-Sep-2009 10:22:59] <eidolon> eek
[29-Sep-2009 10:23:08] <eidolon> customizing zenoss requires a really good understanding of what it does.
[29-Sep-2009 10:23:10] <eidolon> and how it does it.
[29-Sep-2009 10:23:14] <exarkun> I'm sure.
[29-Sep-2009 10:23:17] <ckrough> exarkun: this may clear things up too docs/DOC-3020
[29-Sep-2009 10:23:27] <eidolon> and, unless it's a custom application, i bet someone has done what you need already.
\n[29-Sep-2009 10:23:59] <exarkun> eidolon: Maybe on the way to\nlearning enough to extend it, I'll learn enough to recognize that what\nI'm after is already possible somehow.
[29-Sep-2009 10:24:18] <eidolon> that's why i suggest getting an installation up and running.
[29-Sep-2009 10:24:22] <eidolon> you learn a lot along the way
\n[29-Sep-2009 10:24:23] <cgibbons> there's also a developer's\nguide that continues to be enhanced, but depending upon how deep you\nwant to go it probably won't be enough
[29-Sep-2009 10:24:40] <cgibbons> so what are you after, anyway?
[29-Sep-2009 10:24:41] <exarkun> I spent a couple hours clicking around in an installation yesterday
[29-Sep-2009 10:24:50] <exarkun> Need to read some docs or source to get further
\n[29-Sep-2009 10:24:56] <eidolon> i certainly had a hard time\nunderstanding some basic concepts when i started, but once i ran,\nsetup, administered, and worked with an install for a few months, i\nknow what it can and can't do.
[29-Sep-2009 10:25:07] <eidolon> yah, what do you need to do?
[29-Sep-2009 10:25:51] <ckrough> is the API no longer available online? that was my favorite doc
\n[29-Sep-2009 10:26:21] <eidolon> oh, let me know if you find it? \ni was grumpyu yesterday looking for an object definition for evt.
\n[29-Sep-2009 10:26:59] <ckrough> not sure, but it looks like it\nis no longer 'live' in html, search for API on the community site and\nyou can download it
[29-Sep-2009 10:27:14] <eidolon> k
[29-Sep-2009 10:27:21] <cgibbons> that's on matt's todo list to fix i think
\n[29-Sep-2009 10:27:21] <exarkun> I'm not sure what all the right\nterms are yet. If "agent" is the word for the process that runs on\nnodes and communicates monitoring information back to the main zenoss\nserver (is that the "hub"?), then I need a custom agent that uses a\nspecial transport.
[29-Sep-2009 10:27:28] <purevol> ahoy there fellas
[29-Sep-2009 10:27:45] <cgibbons> that would be a collector, exarkun
[29-Sep-2009 10:27:52] <cgibbons> agent typically implies something that runs on a monitored end-point/device/system
[29-Sep-2009 10:28:02] <ckrough> I think thats what he meabt
[29-Sep-2009 10:28:03]\n<eidolon> exarkun: wht cgibbons said. but there's a ton of\ntransports already. do you need to go-to-machine-and-run-a-command?
[29-Sep-2009 10:28:07] <ckrough> node being the monitored box
[29-Sep-2009 10:28:10] <eidolon> ala, a nagios approach?
[29-Sep-2009 10:28:30] <exarkun> This needs to run on the monitored end-point/device/system
[29-Sep-2009 10:28:36] <ckrough> most zenoss monitoring is agentless, or uses something native like snmp/wmi
[29-Sep-2009 10:28:44] <ckrough> what are you monitoring
[29-Sep-2009 10:28:58] <exarkun> The monitored system can't get snmp get, and can't accept incoming connections (eg ssh)
[29-Sep-2009 10:29:09] <exarkun> Some machines in a hosted cluster
[29-Sep-2009 10:29:20] <eidolon> so it needs to publish traps back to a collector somewhere
[29-Sep-2009 10:29:22] <exarkun> s/snmp get/snmp out/
[29-Sep-2009 10:29:43] <ckrough> someone else here had that same issue, cant remember the final solution
[29-Sep-2009 10:29:47] <ckrough> deja vu
\n[29-Sep-2009 10:30:51] <cgibbons> we had a big customer @ a\ndifferent company whose entire monitoring network was that way. the\nmonitored systems could send data back to the infrastructure, but the\ninfrastructure itself couldn't pull data. thousands of remote-sites\nthat just did data collection locally.
[29-Sep-2009 10:31:05] <eidolon> yep.
[29-Sep-2009 10:31:14] <eidolon> my immediate answer is to set up an SSL VPN.
[29-Sep-2009 10:31:19] <eidolon> and monitor over that.
[29-Sep-2009 10:31:29] <eidolon> but you can push to a collector, that'll work fine.
[29-Sep-2009 10:31:45] <kmtsun> join #nagios
\n[29-Sep-2009 10:32:46] <exarkun> Not sure if it'll be possible to\nset up an SSL VPN, but I'll keep that idea in mind. Thanks for all the\npointers so far.
[29-Sep-2009 10:34:03]\n<cgibbons> there's a new, easier-to-use (for developer's)\ncollector framework in the version about to be released but there's\nstill a lot to do if you want to build an entirely new daemon\nframework, etc.
[29-Sep-2009 10:34:13] <cgibbons> i think we've only had it happen once or twice with the community
[29-Sep-2009 10:37:41] <kmtsun> ls
\n[29-Sep-2009 10:59:47] <feutete_> I am using the HTTP Monitor\nZenPack, and it appears that timeouts when running check_http never\nauto-clear. any idea why?
[29-Sep-2009 11:05:21] <rmatte> annnnnd I'm back
[29-Sep-2009 11:07:51]\n<rmatte> ckrough: if you're logging you might as well save it and\nsend it to Matt later on, these are all good points
[29-Sep-2009 11:08:32]\n<rmatte> they really need a logging bot setup in here that's here\n24/7, that way we could be like "Refer to the conversation at\napproximately 11:00-12:00 on September 29th".
[29-Sep-2009 11:13:11] <ckrough> there are some good bots that link to docmentation too, that would be helpful
[29-Sep-2009 11:13:18] <ckrough> maybe Ill play with supybot this weekend
[29-Sep-2009 11:13:42] <rmatte> yeh, it would be great to get something setup in here
[29-Sep-2009 11:13:55] <rmatte> it's almost strange to see an IRC channel without a bot these days
[29-Sep-2009 11:14:10] <rmatte> we could potentially just setup an eggdrop with some custom tcl scripts
[29-Sep-2009 11:14:12] <ckrough> matt said they used to have one, he wasnt sure where it went and why
[29-Sep-2009 11:14:44] <rmatte> I'll setup an eggdrop on my home server and see what I can do with it
[29-Sep-2009 11:14:52] <rmatte> been a while since I've done anything with one
[29-Sep-2009 11:15:24] <rmatte> woohoo, time to commission a new client
[29-Sep-2009 11:15:38] <rmatte> going to be a breeze, 45 devices, all Cisco
[29-Sep-2009 11:17:33] <eidolon> oh please don't run an eggdrop bot.
[29-Sep-2009 11:17:40] <eidolon> jeez, that's so 1997.
[29-Sep-2009 11:17:48] <eidolon> at least run an infobot that we can run factoids in.
[29-Sep-2009 11:18:06] * eidolon can probably get the javabot code in. i'm an old hat at that.
[29-Sep-2009 11:19:03] <rmatte> exarkun: all the files that "do" stuff are mostly in the Products directory ($ZENHOME/Products)
[29-Sep-2009 11:19:39] <rmatte> an eggdrop can have an infobot script thrown on it
[29-Sep-2009 11:19:40] <eidolon> but first, lunch.
[29-Sep-2009 11:19:43] * eidolon zips.
[29-Sep-2009 11:19:48] <rmatte> it's quite flexible
[29-Sep-2009 11:20:03] <rmatte> I just like them because they are stable as hell
[29-Sep-2009 11:20:56] <rmatte> and for the record, it's "1337"
[29-Sep-2009 11:22:41] * ckrough votes supybot
[29-Sep-2009 11:22:52] <ckrough> ands it python I think
[29-Sep-2009 11:23:17] <rmatte> oooh, python
[29-Sep-2009 11:56:01] <rhettardo> it is not 1337
[29-Sep-2009 11:56:03] <rhettardo> its 2009
[29-Sep-2009 12:38:15] <rmatte> lol
\n[29-Sep-2009 12:39:50] <cjoslin> not looking for a how to right\nnow.. but i am trying to find a NMS that works well in our WINDOWS\nenvironment. can zenoss alert me with error logs from eventlog? can\nit monitor disk space, ping, cpu, services? i know are decent paid\nsolutions out there. Loved PA server monitor Trial, however we have\na small budget.
[29-Sep-2009 12:40:38] <rmatte>\ncjoslin: Well, it can do all of that, the Core version needs a mix of\nSNMP and WMI to do all of that, the enterprise version can do all of\nthat via WMI
[29-Sep-2009 12:43:41] <cjoslin> is WMI just better implemented in the enterprise version?
[29-Sep-2009 12:47:51] <cjoslin> rmatte... ?
[29-Sep-2009 12:50:30] <rmatte> yes
[29-Sep-2009 12:50:39] <rmatte> sorry, had to step away for a second
[29-Sep-2009 12:50:55] <rmatte> it's better implemented in the enterprise version
\n[29-Sep-2009 12:51:26] <rmatte> you can monitor event log,\nservices, and performance data via WMI with Zenoss core, but you can't\nmonitor interfaces, and filesystem
[29-Sep-2009 12:51:43] <rmatte> so if you want to use core you can use a mix of both, or just use straight WMI
[29-Sep-2009 12:51:46] <rmatte> erm
[29-Sep-2009 12:51:49] <rmatte> straight SNMP I mean
[29-Sep-2009 12:52:21] <cjoslin> from what i have read tho monitoring event log with snmp is super tricky.
[29-Sep-2009 12:52:37] <cjoslin> but if it can do both then i would like to try it out.
[29-Sep-2009 12:52:42] <cjoslin> since its free and all
\n[29-Sep-2009 12:52:52] <rmatte> well, it depends to what degree\nyou want to monitor it, if you configure SNMP traps windows will send\nan snmp trap to Zenoss if major events occur
[29-Sep-2009 12:53:19] <rmatte> but core does support eventlog monitoring via WMI
[29-Sep-2009 12:53:23] <rmatte> so you can do it that way as well
[29-Sep-2009 12:53:27] <rmatte> it's just much chattier
[29-Sep-2009 12:55:51] <rmatte> to do the performance monitoring you'll need to install a ZenPack
\n[29-Sep-2009 12:56:32] <cjoslin> well with all that said.. do\nyou think i should go with zenoss to do those kinda things? again \nmainly windows enviroement here. aside from the asterisk pbx.
[29-Sep-2009 12:58:04]\n<cjoslin> i mean, am i looking at doing coding or going really\ndeep into config files. or is it something i can have at least working\nby the end of the day?
[29-Sep-2009 13:09:02] <rmatte> you can have it working by the end of the day
[29-Sep-2009 13:09:20] <rmatte> I use Zenoss to monitor over 100 Windows servers.
[29-Sep-2009 13:09:24] <rmatte> I do it all via SNMP though
[29-Sep-2009 13:09:41] <rmatte> but I've worked with the WMI before
[29-Sep-2009 13:11:14] <rmatte> do you want to do your actual performance monitoring (CPU/Memory) via WMI or SNMP?
\n[29-Sep-2009 13:16:48] <cjoslin> rmatte... really whatever is\neasier i am the only admin around here so its kinda just to get\nsomethign working..
[29-Sep-2009 13:17:28] <chudler> ah.. one man show, eh? Used to love those days
[29-Sep-2009 13:17:42] * chudler crawls back under rock
\n[29-Sep-2009 13:18:16] <cjoslin> lol well we have developers\nfor the service we provide and all.. but as far as sys/network admin\nthats me
[29-Sep-2009 13:18:41] <cjoslin> i prefer it to being more specialized since it helps alleviate my adhd :-)
\n[29-Sep-2009 13:18:57] <rmatte> the WMI would probably be easier\nsince Matt hasn't published the simple version of my SNMP Performance\npack yet
[29-Sep-2009 13:19:33] <rmatte> You'll\nwant to go to the ZenPacks section of the community site and download\nthe WMI Data Source and WMI Performance Template ZenPacks (both are by\nEgor)
[29-Sep-2009 13:19:43] <rmatte> then unzip them, and install them
\n[29-Sep-2009 13:19:57] <rmatte> (they aren't supposed to be\nzipped, the new community site zips everything automatically apparently)
[29-Sep-2009 13:20:11] <cjoslin> i am downloading the vmware image of zenoss core as we type
[29-Sep-2009 13:20:19] <rmatte> k
\n[29-Sep-2009 13:20:38] <rmatte> you'll want to read the Zenoss\nAdmin guide at some point to really understand how Zenoss works and how\nto configure it
[29-Sep-2009 13:20:47] <rmatte> but we can get you setup with something for the time being
[29-Sep-2009 13:22:26] <cjoslin> so no WMI out of the box eh?
[29-Sep-2009 13:22:33] <cjoslin> looks like i am going to head over to this community site.
\n[29-Sep-2009 14:48:19] <rmatte> cjoslin: it does event log and\nservice monitoring via WMI out of the box, but not performance
\n[29-Sep-2009 14:48:46] <rmatte> and you'll have to monitor\ninterfaces and disks using SNMP, unless you purchase the enterprise\nversion
[29-Sep-2009 15:04:24] <ke4qqq> I just\ninstalled dnsmonitor zenpack - and get this: http://fpaste.org/g8ZB/ ,\nwhen I try and look at the zenpack - is this a known problem??
[29-Sep-2009 15:04:44] <ke4qqq> I don't see an open ticket for it
[29-Sep-2009 15:09:47] <rmatte> restart zenoss and try again
[29-Sep-2009 15:10:07] <rmatte> some zenpacks require that zenoss be restarted after they are installed
[29-Sep-2009 15:16:40] <ke4qqq> rmatte: thanks, I'll try that
[29-Sep-2009 15:18:04] <alinux> hi all, does Zenoss require a specific client or just SNMP clients ?
[29-Sep-2009 15:18:18] <cgibbons> "client" ?
[29-Sep-2009 15:19:20] <ke4qqq> alinux: zenoss doesn't have 'clients' - it will use SNMP, WMI, ssh, etc
[29-Sep-2009 15:20:53] <ke4qqq> it's considered client-less
[29-Sep-2009 15:21:52] <gwb235> marketing lingo was "agent-less"
[29-Sep-2009 15:22:03] <rhettardo> yes
[29-Sep-2009 15:22:21] <ke4qqq>
yes[29-Sep-2009 15:24:08] <mrayzenoss> we've trained you well
[29-Sep-2009 15:28:29] <rmatte> lol
[29-Sep-2009 15:28:29] <gwb235> I, for one, am glad to welcome our new monitoring overlords
\n[29-Sep-2009 15:31:26] <rmatte> The whole, "we've trained you\nwell thing": "I am a Zedi, like my father before me!" "...sooo be it, \nZedi."
[29-Sep-2009 15:31:46] <rmatte> *** queue lightning from fingers ***
[29-Sep-2009 15:31:58] <rmatte> cue*
\n[29-Sep-2009 15:32:26] <rmatte> eughhhhh, this is so boring, I'm\nadding 45 devices in but only 15 are actually monitored, the others\njust represent stacked switches
[29-Sep-2009 15:39:29] <cgibbons> the really dumb kind?
[29-Sep-2009 15:40:33] <tung> I got another dumb question for you. Since I couldn't find the answer
[29-Sep-2009 15:40:33] <tung> in any Zenoss forum or doc I figure it must be something basic I'm
[29-Sep-2009 15:40:34] <tung> missing
[29-Sep-2009 15:40:53] <tung> I've created a new local template for a device, let's say
[29-Sep-2009 15:41:17] <tung> \\/Devices/Server/devices/fe01 (fe01 is my host name)
[29-Sep-2009 15:41:25] <tung> In this new template, I added the 3 data points along with 3
[29-Sep-2009 15:41:26] <tung> thresholds and graphs from a script I created.
[29-Sep-2009 15:41:33] <tung> I can see the graphs just find in Perf tab.
[29-Sep-2009 15:41:40] <tung> I created a new ZenPack and add this new template into that ZenPack
[29-Sep-2009 15:42:09] <tung> Now I want to replicate this template to the other 20 boxes that I
[29-Sep-2009 15:42:09] <tung> want this on.
[29-Sep-2009 15:42:17] <tung> But I when I go to the other systems and try to Bind Templates, the
[29-Sep-2009 15:42:17] <tung> new template is not listed as a choice.
[29-Sep-2009 15:42:25] <tung> What am I missing?
[29-Sep-2009 15:42:29] <tung> Thanks in advance
[29-Sep-2009 15:43:26] <ke4qqq> tung
[29-Sep-2009 15:43:37] <ke4qqq> tung: what you want to do is copy the template not make a local copy
[29-Sep-2009 15:44:15] <ke4qqq> so go to Devices> Templates> and either create a new template
[29-Sep-2009 15:44:26] <ke4qqq> or copy an existing template that is located there
[29-Sep-2009 15:44:36] <tung> How do I do that? Yes I Created the new template
[29-Sep-2009 15:44:53] <ke4qqq> Click on Devices under the "classes" menu on the left
[29-Sep-2009 15:45:03] <ke4qqq> then click on the Templates tab
[29-Sep-2009 15:45:13] <ke4qqq> you'll see Available performance templates
\n[29-Sep-2009 15:45:50] <ke4qqq> (it only shows templates\navailable at /Devices, so a Windows template might not show up there,\nand would be a bit deeper, but the same principle applies)
[29-Sep-2009 15:45:59] <tung> OK I'm there but I don't see the one I have created
\n[29-Sep-2009 15:46:23] <cgibbons> it's probably just created at a\ndeep level than you intended, so it's not available in the device class\nyou're adding your new devices to
[29-Sep-2009 15:46:51] <tung> Mine is under /Devices/Server/Linux/device/hostname
\n[29-Sep-2009 15:46:54] <ke4qqq> so - there or somehwere similar\n(/Devices/Server/Windows ) if you create it on the device, it'll only\nbe visible at that device, and you want to move at least one level up\nto create it.
[29-Sep-2009 15:47:08] <tung> HOw do I move it up one level?
[29-Sep-2009 15:47:23] <ke4qqq> so it would be at least available to all Linux servers in your example
[29-Sep-2009 15:47:49] <tung> yeah that's exactly what I need to do but I don't see an option to move a template
[29-Sep-2009 15:48:18] <tung> "Remove Local Copy" was not a good thing to do :-)
\n[29-Sep-2009 15:48:27] <ke4qqq> tung go to /Devices/Server/Linux\nclick on the templates tab and create it there. Unfortunately I am not\naware of a way to copy it
[29-Sep-2009 15:48:58] <tung> really?
[29-Sep-2009 15:49:39] <ke4qqq> others more knowledgeable than I might know, but I don't
[29-Sep-2009 15:49:59] <cgibbons> there's probably a way, let me ask... one moment
[29-Sep-2009 15:50:18] <tung> OK if that would work then I think it's good
[29-Sep-2009 15:50:23] <tung> Thanks. I think that helps.
[29-Sep-2009 15:50:38] <tung> At least I don't have to create the templates 20 times :-D
[29-Sep-2009 15:55:56] <rmatte> I might know a way to move it up...
[29-Sep-2009 15:56:34] <rmatte> hmmm, nope, won't work for local copies apparently
[29-Sep-2009 15:57:29] <cgibbons> prolly have to do it manually via zendmd
\n[29-Sep-2009 15:59:58] <tung> That's ok. Thanks everyone. I\nrecreated it in /Devices/Server/Linux and it's now visible to my other\nhosts
[29-Sep-2009 15:59:58] <tung> So that did it
[29-Sep-2009 16:00:03] <tung> didn't take too much time
[29-Sep-2009 16:01:18] <cgibbons> cool
[29-Sep-2009 16:05:28] <jb> hrm looks like all of my windows process monitoring is broken
[29-Sep-2009 16:05:30] <jb> at least the graphs
[29-Sep-2009 16:22:00]\n<chudler> tung: I know it is too late, but when viewing the\ndevice, add /perfConfig to the end of the URL. Then you should see the\ntemplates on the device, and the "Arrow menu" has a "Copy Tempaltes..."\noption to copy them anywhere in the device tree.
[29-Sep-2009 16:22:44] <ke4qqq> chudler: doesn't that only apply to non-local copies (eg non-machine specific)
[29-Sep-2009 16:23:38] <tung> That's good to know. Thanks, chudler
[29-Sep-2009 16:25:06]\n<chudler> since that option cannot be browsed to, and it doesn't\nsay what it is copying, I suspect that it is not supported or might not\nwork as expected. Try it :-)
[29-Sep-2009 16:25:26] <chudler> I mean it cannot be browsed at the device level, only organizers AFAIK
[29-Sep-2009 16:28:17] <chudler> I tried it and it didnt work, sorry. I think I did it once some other way.
[29-Sep-2009 16:28:26] <cgibbons> bbl
[29-Sep-2009 16:39:33] <ke4qqq> mrayzenoss: is there a reason this isn't listed at c.z.o? http://blog.dastrup.com/?p=26
[29-Sep-2009 16:39:40] <ke4qqq> that zenpack?
[29-Sep-2009 16:39:45] <mrayzenoss> yes
[29-Sep-2009 16:39:56] <mrayzenoss> I asked him to send it to me to post, he hasn't
[29-Sep-2009 16:40:09] <mrayzenoss> I've asked several times over the last year
[29-Sep-2009 16:40:16] <ke4qqq> he used to hang out in irc - but I haven't seen him in well over a year
[29-Sep-2009 16:40:27] <mrayzenoss> plus, the new WMI Data Source one may be better
[29-Sep-2009 16:40:42] <mrayzenoss> since it doesn't leave passwords in logs
[29-Sep-2009 16:42:06]\n<mrayzenoss> take a look at these:\ndocs/DOC-3430 and\ndocs/DOC-3387
[29-Sep-2009\n16:42:23] <mrayzenoss> they're working together on a combined\nversion of the ZenPack, I should have it up soon
[29-Sep-2009 16:45:08] <ke4qqq> mrayzenoss: interesting - I honestly had been using the other so long I didn't know about those two
\n[29-Sep-2009 16:49:26] <mrayzenoss> ke4qqq: as soon as the WMI\nData Source came out, there was a flurry of rewriting the old\nwmic-based ZenPacks
[29-Sep-2009 16:50:28]\n<mrayzenoss> yay, new docs available:\ncommunity/documentation/official_documentation/training
[29-Sep-2009 17:04:15]\n<ke4qqq> mrayzenoss: can you change the zip/egg names on those\ntwo exchange zenpacks - despite being different they have exactly the\nsame name
[29-Sep-2009 17:04:52] <mrayzenoss> yeah, I'll drop one when I post the unified one
[29-Sep-2009 17:05:08] <mrayzenoss> it's kinda funny, they were developed independently
[29-Sep-2009 17:05:15] <mrayzenoss> submitted within a day
[29-Sep-2009 17:05:23] <mrayzenoss> both authors are Swedish
[29-Sep-2009 17:05:49]\n<ke4qqq> fascinating - I just got confused when trying to install\nthem because they both had the same name, but diff md5sums
[29-Sep-2009 17:08:52]\n<mrayzenoss> and on top of that... I just had to add a disclaimer\nthat all ZenPacks downloaded from the site need to be unzipped
[29-Sep-2009 17:09:01]\n<mrayzenoss> "Please note that you will need to unzip the\nattached ZenPacks because the new platform automatically zips them for\nsecurity concerns."
[29-Sep-2009 17:09:04] <mrayzenoss> stupid IE
[29-Sep-2009 17:09:17] <ke4qqq> yeah I saw that discussion on irc last week
[29-Sep-2009 17:10:40] <mrayzenoss> I tried to get them to turn it off
\n[29-Sep-2009 17:11:22] <mrayzenoss> but they don't support that\nsince IE automatically scans filetypes, which is a known security hole
[29-Sep-2009 17:11:36] <mrayzenoss> err... scans contents of uncompressed files
[29-Sep-2009 17:11:43] <mrayzenoss> even though eggs are already zips
\n[29-Sep-2009 17:13:48] <ke4qqq> hmmmm just installed\nwmidatasource and it doesn't show up in templates at any level - is\nthat expected?
[29-Sep-2009 17:14:18] <ke4qqq> nm I can't read
[29-Sep-2009 17:15:56] <mrayzenoss> yes
[29-Sep-2009 17:16:23] <ke4qqq> yes as in you are acking my illiteracy
\n[29-Sep-2009 17:17:14] * ke4qqq wishes I had watched the 'sleeping well\nat night' presentation in its entirety - ohhh well, perhaps I'll get\nthe audio or video
[29-Sep-2009 17:20:10]\n<rmatte> eugh, after tons of screwing around I finally managed to\nget login access to the switches that I'm commissioning except for 1
[29-Sep-2009 17:20:23] <rmatte> need to find a way to break in to the damn thing and set it up properly...
\n[29-Sep-2009 17:20:50] <rmatte> otherwise we'll be waiting over a\nweek for the client to reset the password. heck, it took them 3 weeks\njust to get back to us about setting up a vpn tunnel
[29-Sep-2009 17:20:57] <mrayzenoss> ke4qqq: I got an email from Greg at OLF saying they're cleaning them up now
[29-Sep-2009 17:22:06]\n<ke4qqq> yeah I got that same email, but I wouldn't be surprised\nto see it take them a month to get all of them done. It took SELF two\nmonths for audio and 3 for video - and we actually had a company doing\nit for us.
[29-Sep-2009 17:25:09] <rmatte> I love how I can hear rain on the skylights but my weather widget is showing it as being sunny out
[29-Sep-2009 17:25:30] <straterra> If the widget used ipv6..it'd be right
[29-Sep-2009 17:25:39] <rmatte> hehe
[29-Sep-2009 17:25:57] <straterra> I look forward to an ipv6 world
[29-Sep-2009 17:26:32] <rmatte> still seems a long ways away
\n[29-Sep-2009 17:27:34] <rmatte> quite honestly, I don't know how\ngreat it's going to be... "Now everyone can have their own IP, making\nit even easier for hackers, malware, and worms to make their way\ndirectly to your PC
"[29-Sep-2009 17:28:03] <straterra> Eh, in terms of that..It'll be the same
[29-Sep-2009 17:28:07] <cwj> it seemed a lot more appealing before every developer caught on to the fact that most people are behind nat
[29-Sep-2009 17:28:17] <straterra> Everyone had their own IP too..just behind NAT
[29-Sep-2009 17:28:35] <straterra> Just means I don't have to work about VPN tunnels and crap to access machines at home
[29-Sep-2009 17:28:59] <cwj> neither do crackers and phishers
[29-Sep-2009 17:29:07] <cwj> under ipv6
\n[29-Sep-2009 17:29:17] <straterra> I have everything at home and\non my server ipv6 enabled..except for my iPod Touch. Even my Windows\nMobile phone works with ipv6
[29-Sep-2009 17:29:38] <rmatte> I don't really see the point in using it at the moment
[29-Sep-2009 17:29:48] <straterra> future proof
[29-Sep-2009 17:29:54] <straterra> and..getting around NAT
[29-Sep-2009 17:29:54] <rmatte> once it starts becoming more heavily adopted in a business sense, then I'll switch over
[29-Sep-2009 17:44:51] <rmatte> eugh, this one device is driving me crazy
\n[29-Sep-2009 17:44:56] <__jd__> Hey all.. After upgrading from\n2.4.3 to 2.4.5, I started having an issue where 'clear' messages are\nbeing sent out immediately before new alerts. This is a case where an\nalert which would normally just count is now clearing and then\nun-clearing immediately afterwards each time. \ndown->up->down->up->down->up->down.. The thing is,\nthe service has never actually come back up. The clear notifications\nare bogus
[29-Sep-2009 17:45:55] <rmatte> check your zenactions log, see if you notice anything weird
\n[29-Sep-2009 17:48:54] <__jd__> some exceptions regarding\nsendXmpp, which probably broke with the upgrade.. Other than that, I\nsee it sending the clear and then the down again 1 minute later..
[29-Sep-2009 17:51:03] <__jd__> in this case, it's just a TCP service check done by zenoss.. Not zencommand or anything
[29-Sep-2009 17:51:04] <__jd__> HTTP
[29-Sep-2009 17:52:59] <__jd__> hmm.. maybe some other alert is clearing it?
[29-Sep-2009 17:53:33] <__jd__> it uses the dedupid, right?
[29-Sep-2009 17:56:24] <__jd__> I take it back, this is zencommand
[29-Sep-2009 18:00:10] <rmatte> it's possible that something else is clearing it, though not very probable
[29-Sep-2009 18:01:13] <__jd__> yea, it doesn't look like it
[29-Sep-2009 18:01:20] <__jd__> I'm running zencommand with debug
[29-Sep-2009 18:02:13] <rmatte> is the service actually down or up?
[29-Sep-2009 18:02:19] <__jd__> it's down
[29-Sep-2009 18:08:54] <__jd__> hm
[29-Sep-2009 18:09:12] <__jd__> it clears, but there's nothing run by zencommand to clear it
[29-Sep-2009 18:09:46]\n<__jd__> in fact, the template is setup to only run the command\nevery 5 minutes, yet the clear comes 1 minute before the re-alert
[29-Sep-2009 18:10:11] <__jd__> it seems like from zencommand's perspective, the service indeed never comes back up
\n[29-Sep-2009 18:10:17] <rmatte> well, I'm out of here shortly but\nyou can try this... seems to help with wacky event stuff like that\nsometimes...
[29-Sep-2009 18:10:19] <rmatte> stop zenoss
[29-Sep-2009 18:10:29] <rmatte> copy $ZENHOME/var/Data.fs to Data.fs.bak
[29-Sep-2009 18:10:32] <rmatte> start zenoss
[29-Sep-2009 18:10:35] <rmatte> go in to zendmd
[29-Sep-2009 18:10:37] <rmatte> and type:
[29-Sep-2009 18:10:39] <rmatte> reindex()
[29-Sep-2009 18:10:41] <rmatte> commit()
[29-Sep-2009 18:10:51] <rmatte> see if you still have the issue after doing that
[29-Sep-2009 18:11:10] <__jd__> will do.. thanks
\n[29-Sep-2009 18:11:46] <rmatte> I have people backup the Data.fs\nsince I've seen someone's zope database get fried by a reindex since\nthey had a major problem in it and the reindex made it even worse, so I\nhave people back it up before doing reindexes now
[29-Sep-2009 18:11:53] <rmatte> I've never had any issues
[29-Sep-2009 18:12:17] <__jd__> k.. don e:)
[29-Sep-2009 18:12:23] <rmatte> if it runs through fine you can delete the .bak afterwards
[29-Sep-2009 18:15:58] <rmatte> anyways, I'm out of here
[29-Sep-2009 18:16:04] <rmatte> I'll be around tomorrow
[29-Sep-2009 18:50:45] <davetoo1> I wish zenoss would leave my snmpd.conf alone
[29-Sep-2009 18:50:46] <davetoo1> grumble
[29-Sep-2009 19:36:29] <cryptographrix> y0 - anyone have probs with httpd process monitoring?
[29-Sep-2009 19:36:37] <__jd__> yes
[29-Sep-2009 19:36:42] <purevol> sure
[29-Sep-2009 19:37:04] <__jd__> or wait.. process? not sure, haven't tried
\n[29-Sep-2009 19:37:33] <cryptographrix> looks like process\nmonitoring in general - basically no matter which processes I'm\nmonitoring, whichever one is first in the list is shown as being down
[29-Sep-2009 19:38:11] <cryptographrix> any ideas?
[29-Sep-2009 19:38:25] <purevol> so you have a device setup, and you modeled it?
[29-Sep-2009 19:38:31] <cryptographrix> yes
[29-Sep-2009 19:38:59] <purevol> windows or linux host?
[29-Sep-2009 19:39:05] <cryptographrix> linux host
\n[29-Sep-2009 19:39:39] <cryptographrix> just testing zenoss right\nnow, getting it to monitor another host (one of our QA boxes)\nsuccessfully before we roll it out to more
[29-Sep-2009 19:41:09] <purevol> if you goto services is httpd in there?
[29-Sep-2009 19:41:52] <cryptographrix> no
\n[29-Sep-2009 19:43:12] <purevol> im not in front of my zenoss\ninstall - but check this out\ndocs/DOC-2446#Whydoesnthttpdshowupinmylistofservices
[29-Sep-2009 19:43:42] <cryptographrix> should it work to create httpd under processes and regex? kk willdo
[29-Sep-2009 19:49:54] <purevol> you might have to add http to the collector plugins for the device
[29-Sep-2009 19:50:34] <purevol> so goto Device List, then goto the menu and select More -->Collector Plugins
[29-Sep-2009 19:50:56] <purevol> docs/DOC-3093#d4e1738
[29-Sep-2009 19:51:54] <cryptographrix> coolcool will check
\n[29-Sep-2009 19:51:59] <purevol> on the right you will see an \n"Add Fields" click that, and i think Linux.httpd should be in there
[29-Sep-2009 19:52:32] <purevol> i think
[29-Sep-2009 19:54:01] <cryptographrix> k I'll look for it - plugin's built into the core, it looks like
[29-Sep-2009 19:54:32] <cryptographrix> (not part of a non-core zenpack)
\n[29-Sep-2009 20:15:12] <cryptographrix> yet another strange\nerror: Error reading value for "eth0" on XXXXXXXX (oid\n.1.3.6.1.2.1.31.1.1.1.7.2 is bad)
[29-Sep-2009 20:15:31] <cryptographrix> for eventClass /Perf/Snmp
[29-Sep-2009 20:15:34] <purevol> you need to make sure you have MIBs installed
[29-Sep-2009 20:16:12] <cryptographrix> MIBs? and installed on zenoss host or monitored server?
[29-Sep-2009 20:16:16] <purevol> did you get you httpd monitored?
[29-Sep-2009 20:16:39] <cryptographrix> nah, still working on it - figuring stuff out slowly tho
\n[29-Sep-2009 20:18:13] <purevol> it takes a little while, but\neverything is in the admin guide, just walk through it and you will be\ngood, that oid thing, probably just needs a mib installed
[29-Sep-2009 20:18:31] <cryptographrix> kk
[29-Sep-2009 21:13:04] <davetoo> why doesn't my process monitor go away?
[29-Sep-2009 21:13:54] <purevol> what do you mean davetoo?
[29-Sep-2009 21:14:22] <davetoo> well, I had a bunch of process monitors devined and globally enabled
[29-Sep-2009 21:14:35] <davetoo> having trouble with them, so I disabled them all
[29-Sep-2009 21:14:52] <davetoo> so each device they were found on, now shows that they aren't monitored
[29-Sep-2009 21:15:03] <davetoo> and I even restarted zenprocess on all the collectors,
[29-Sep-2009 21:15:10] <davetoo> but I'm still getting events
[29-Sep-2009 21:15:23] <purevol> interesting
[29-Sep-2009 21:16:08] <purevol> maybe you need to restart the daemons
[29-Sep-2009 21:16:13] <davetoo> I did
[29-Sep-2009 21:16:23] <davetoo> going to try something next
[29-Sep-2009 21:16:48] <purevol> did you remodel the device?
[29-Sep-2009 21:16:53] <davetoo> each device still has the processes listed, but they are no longer listed as "monitored"
[29-Sep-2009 21:17:07] <davetoo> welll
[29-Sep-2009 21:17:26] <davetoo> I thought I had, but apparently not
[29-Sep-2009 21:18:19] <davetoo> I'm having a hell of a time monitoring the zenoss daemon processes themselves
[29-Sep-2009 21:18:36] <purevol> VM appliance?
\n[29-Sep-2009 21:18:52] <davetoo> it's also unfortunate that it\ndepends on what's running at the time you model the device, what\nprocesses you end up with in the model
[29-Sep-2009 21:19:01] <davetoo> Enterprise, physical remote collectors
[29-Sep-2009 21:20:08] <purevol> hrm.... im a bit rusty
[29-Sep-2009 21:20:44] <davetoo> The boxes I'm monitoring are the remote collectors themselves
[29-Sep-2009 21:21:00] <davetoo> trying to make sure that all the zenoss daemons are running
[29-Sep-2009 21:21:28] <davetoo> but the way they are started, you have to have "ignore parameters" set to False
[29-Sep-2009 21:23:27] <purevol> can you create a command to login and check the processes
[29-Sep-2009 21:24:37] <davetoo> I may decide to do that instead, yes
[29-Sep-2009 21:27:57] <davetoo> Maybe it requires a remodel to push the config out to the daemons
[29-Sep-2009 21:34:02] <davetoo> *sigh*
[29-Sep-2009 21:34:05] <davetoo> so I did remodel,
[29-Sep-2009 21:34:27] <davetoo> and the processes are back;
[29-Sep-2009 21:34:43] <davetoo> they are set not to monitor
[29-Sep-2009 21:34:49] <davetoo> but
[29-Sep-2009 21:34:55] <davetoo> I am getting events
[29-Sep-2009 21:35:03] <davetoo> baffled
[29-Sep-2009 21:35:47] <purevol> so you have half working
[29-Sep-2009 21:38:49] <davetoo> I just don't get it
[29-Sep-2009 21:41:02] <davetoo> I really don't want to erase all of the process definitions I spent so much time creating
[29-Sep-2009 21:42:28] <purevol> maybe someone tomorrow will have a better answer for you, sorry man
[29-Sep-2009 21:57:22] Traveler is now known as Guest60426
[29-Sep-2009 21:58:12] <Guest60426> Hi all - anyone provide assistance with graph rendering issue?
[29-Sep-2009 21:58:15] <Guest60426> ERROR RenderServer Failed to generate a graph
[29-Sep-2009 21:59:19] <Guest60426> happening on 2 different devices using 2 different perf templates
[29-Sep-2009 21:59:42] <purevol> either something is not setup right, or you have to wait for data to populate
[29-Sep-2009 21:59:59] <Guest60426> purevol -thanks for the reply
[29-Sep-2009 22:00:37]\n<Guest60426> these aren't new templates that i just applied - i\nhave old data already graphed - and can access it when i look at the\nweekly view
[29-Sep-2009 22:01:38] <purevol> did you make any changes?
[29-Sep-2009 22:02:00] <Guest60426> its all wmi - and as for testing out the templates i can run tests on them and they return the right data
[29-Sep-2009 22:02:42] <purevol> i dont use any WMI, but if you might want to check your colelctor plugins
[29-Sep-2009 22:02:50] <Guest60426> i recently upgraded to 2.4.4 and went up to 2.4.5 tonight
[29-Sep-2009 22:03:27]\n<purevol> weird, maybe something changed that i dont know about,\nthe best bet is to drop in tomorrow when the smart guys are around
[29-Sep-2009 22:03:39] <Guest60426> i just recently am switching over from snmp to wmi since the env is strictly windows
[29-Sep-2009 22:03:54] <purevol> maybe you try to remodel the device?
[29-Sep-2009 22:04:01] <Guest60426> yeah its got me stumped
[29-Sep-2009 22:04:37]\n<Guest60426> because it was graphing fine until something changed\n(wish i knew if it was the upgrade or zenpack additions etc)
[29-Sep-2009 22:04:52] <Guest60426> i'll try that
[29-Sep-2009 22:05:36] <Guest60426> other things i did recently were deleting, copying and binding new templates to the windows class
[29-Sep-2009 22:09:07] <Guest60426> no go with the remodel
[29-Sep-2009 22:09:25] <Guest60426> i think i may try to restore an old 2.4x db and see what happens
[29-Sep-2009 22:09:49]\n<purevol> thats a bit of a drag, did you check your collector\nplugins for that device, and maybe double check your zProperties, to\nensure they are setup with the right credentials?
\n[29-Sep-2009 22:10:27] <Guest60426> credentials are good - i have\nother perf templates on the server that are using the same ones
\n[29-Sep-2009 22:11:26] <Guest60426> i recently added all the HP\next device info snmp collector plugins so i'll try that too
[29-Sep-2009 22:15:15] <Guest60426> purevol: thanks for the suggestions- if i have time i'll try back tomorrow
[29-Sep-2009 22:15:33] <purevol> sorry im not more useful! :-)
[29-Sep-2009 22:16:10] <purevol> we have some windows at work, but i dont manage any of that, but maybe i can try it
[30-Sep-2009 00:00:46] [disconnected at Wed Sep 30 00:00:46 2009]
[30-Sep-2009 00:00:46] [connected at Wed Sep 30 00:00:46 2009]
[30-Sep-2009 00:00:55] [zenoss-logger (logger bot) has joined #zenoss]
[30-Sep-2009 00:44:03]\n<shentonfreude> so maybe the more effective way is to put -- say\n-- $5K from Koansys credit card into a Koansys paypal account, and pay\nfrom that. But I will not be comfortable with them linking to my debit\ncard nor bank account.
[30-Sep-2009 00:44:18] <shentonfreude> Then if we send to UK from the Koansys PayPal balance it's only $5 fee.
[30-Sep-2009 00:45:01] <shentonfreude> yeah, on that page where all you do is fill out their email addr and amount.
[30-Sep-2009 00:45:08] <shentonfreude> $39 for $1000
[30-Sep-2009 00:46:21] <shentonfreude> sorry, damn, wrong window. hate that.
[30-Sep-2009 02:42:48] <MorkBork> hi im having a bit of trouble finding out if what i want to do is possible
[30-Sep-2009 02:42:55] <MorkBork> i have a small zenoss core install and im watching a few game servers
[30-Sep-2009 02:43:09] <MorkBork> i wanted to source some of the graphs on a status page
[30-Sep-2009 02:43:23] <MorkBork> for the players to see.. i.e. bandwidth, packets per second, etc
[30-Sep-2009 02:43:34] <MorkBork> is there an easy way to do this through acl/permissions
[30-Sep-2009 02:43:47] <MorkBork> like make a perf graph anonymously available
[30-Sep-2009 02:45:11] <MorkBork> or can i just write a cron job or something
[30-Sep-2009 02:45:32] <MorkBork> its not like life or death important, just wondered what the common way to do that is
[30-Sep-2009 02:48:50] <MorkBork> i found the zenreport command
[30-Sep-2009 02:49:13] <MorkBork> but thats a little overkill
\n[30-Sep-2009 02:58:44] <Troubadix09> MorkBork: I remember there\nwas a Document about "Setup a Kiosk-Zenoss" or so in the forum
[30-Sep-2009 02:58:59] <MorkBork> yea but it seems like everything i google now is moved
[30-Sep-2009 02:59:01] <MorkBork> ;(
[30-Sep-2009 02:59:12] <MorkBork> thank you google cache!
\n[30-Sep-2009 02:59:42] <Troubadix09> MorkBork: And there is a\nDocument on the website of jane curry, how to configure user rights in\nzenoss core
[30-Sep-2009 03:00:21] <Troubadix09> MorkBork: dedicated user rights only in zenoss enterprise possible
[30-Sep-2009 03:00:24] <MorkBork> ah
[30-Sep-2009 03:00:29] <MorkBork> yea its just me with an account
[30-Sep-2009 03:00:43] <MorkBork> ive been using zenoss forever but never in any sort of advanced capacity
[30-Sep-2009 03:00:55] <MorkBork> mainly just to keep track of when the gameserver dies so i can go restart it
[30-Sep-2009 03:01:38] <MorkBork> oh so zenoss enterprise has a more involved acl system?
[30-Sep-2009 03:01:44] <MorkBork> i always wondereed what the differences were
[30-Sep-2009 03:02:02] <MorkBork> i always thought the acl system in core was kind of weird
[30-Sep-2009 03:02:03] <MorkBork> makes sense
[30-Sep-2009 03:06:31]\n<Troubadix09> MorkBork: Enterprise have mor different, like\nclustering, HA-configuration, more Zenoss-instances etc.
[30-Sep-2009 03:09:20] <MorkBork> ahh so probably not much use for 1 guy and 4 servers
[30-Sep-2009 03:11:24] <MorkBork> i tried lots of different monitoring things and i keep coming back to zenoss
[30-Sep-2009 03:11:28] <MorkBork> although i love cacti graphs
[30-Sep-2009 03:32:51] <rhettardo> i use both cacti and zenoss
[30-Sep-2009 03:40:27] <MorkBork> i got sick of working with both of them
[30-Sep-2009 03:40:30] <MorkBork> so i just use zenoss
[30-Sep-2009 03:40:40] <MorkBork> the graphs arent as pretty but i dont stare at them all day anyway
[30-Sep-2009 03:40:51] <MorkBork> the only thing i miss was how easy it was to export the graphs
[30-Sep-2009 03:41:03] <MorkBork> maybe someone better than me with python will make a zenpack to export stuff easily
[30-Sep-2009 03:42:51] <rhettardo> ah. pretty much all of my cacti stuff is custom so thats why i keep it around
[30-Sep-2009 04:29:48] <MorkBork> well
[30-Sep-2009 04:29:52] <MorkBork> the rrd files are easy to export
[30-Sep-2009 04:30:06] <MorkBork> so i just figure ill use that and rrdtool and remake the graphs myself
[30-Sep-2009 07:35:28] <Quadro> bah
[30-Sep-2009 07:35:55] <Quadro> building beta 3 or svn head fails
\n[30-Sep-2009 08:31:41] <Dieterbe_> im getting 'process not\nrunning: more' events on a server, but i don't care that that process\ndoesn't run. i verified that 'more' doesn't match any of my process\nmonitoring regexes, but how can i make zenoss stop checking this\nprocess on that server?
[30-Sep-2009 08:39:53]\n<ckrough> Dieterbe_: what version are you running? My boxes have\nalso started monitoring processes I dont care about
\n[30-Sep-2009 08:40:17] <ckrough> I was under the impression it\nwould only pick up on processes that were defined in the Processes\nclass, but I guess I'm wrong.
[30-Sep-2009 08:44:42]\n<Dieterbe_> aha, so i'm not delusional.. it's 2.4.5 , it's\nlooking for a process called 'more' (the pager app must have run at\nsome point and zenoss picked it up for some reason) . i'm having the\nevents since 2009/09/24 17:13:18.000
[30-Sep-2009 08:48:17] <ckrough> yeah, mine has picked up gcc, ruby, etc...
[30-Sep-2009 09:08:14] <Dieterbe_> hah. and you don't have a regex like '.*', '.' or 'c' or something that would match stuff like that?
[30-Sep-2009 09:11:09] <ckrough> nope
[30-Sep-2009 09:11:26] <ckrough> mine are very specific. but more importantly it just started doing this out of the blue.
[30-Sep-2009 09:11:34] <ckrough> Im on 2.4.2 enterprise though.
[30-Sep-2009 09:16:35] <davetoo> There might be some bad regex syntax in processes from some of the enterprise zenpacks?
[30-Sep-2009 09:17:03] <davetoo> I found that the httpd monitor was looking in the parameters list
[30-Sep-2009 09:17:22] <davetoo> and some of my users systems had 'httpd' as an argument in some testbed command
[30-Sep-2009 09:17:39] <davetoo> and during the nightly remodeling those were added to the os processes list
\n[30-Sep-2009 09:18:52] <davetoo> We need a better way to say\n"this is the set of processes that should be running on this class of\nsystems, regardless of whether you find it running during modeling or\nnot"
[30-Sep-2009 09:20:08] <exarkun> I tried to\ninstall my first zenpack. It failed like this: ERROR: zenpack command\nfailed. Reason: [Errno 20] Not a directory:\n'/usr/local/zenoss/zenoss/Products/ZenPacks.SNMP.TCPMIB-1.0-py2.4.egg/skins'
[30-Sep-2009 09:21:05]\n<exarkun> It's true that that isn't a directory. \n/usr/local/zenoss/zenoss/Products/ZenPacks.SNMP.TCPMIB-1.0-py2.4.egg\nisn't either, it's a regular file.
[30-Sep-2009 09:21:20] <davetoo> no,
[30-Sep-2009 09:21:26] <davetoo> the .egg will be a directory
[30-Sep-2009 09:21:36] <exarkun> When? It's not now.
[30-Sep-2009 09:21:48] <davetoo> How did you install it?
[30-Sep-2009 09:22:08] <exarkun> zenpack --install .../ZenPacks.SNMP.TCPMIB-1.0-py2.4.egg.zip
[30-Sep-2009 09:22:17] <davetoo> Zenpacks don't live under .../Products
[30-Sep-2009 09:23:00] <exarkun> zenpack seems to have copied this one there.
\n[30-Sep-2009 09:23:32] <chudler> unzip the egg file first. The\nnew zenoss community website zipped all the eggs, but they are not\nsupposed to be
[30-Sep-2009 09:23:37] <ckrough> I unzip mine first
[30-Sep-2009 09:23:41] <davetoo> They are supposed to be under $ZENHOME/ZenPacks
[30-Sep-2009 09:23:41] <ckrough> yeah that
[30-Sep-2009 09:23:44] <exarkun> Ah.
[30-Sep-2009 09:24:02] <exarkun> So should I just delete this .egg from the Products directory?
[30-Sep-2009 09:24:22] <davetoo> Check what zenoss thinks, first
[30-Sep-2009 09:24:29] <davetoo> does Settings->Zenpacks show it?
[30-Sep-2009 09:24:52] <exarkun> I probably have to restart zenoss before I check?
[30-Sep-2009 09:25:03] <purevol> you are back davetoo
[30-Sep-2009 09:25:19] <davetoo> I'd just be worried about having the zodb out of sync with the actual filesystem
[30-Sep-2009 09:25:23] <davetoo> yes, too early,
[30-Sep-2009 09:25:31] <davetoo> meeting about zenoss project progress
[30-Sep-2009 09:25:42] <purevol> i hear you, i have to go and teach a class in a half hour, needless to say im thrilled
[30-Sep-2009 09:26:08] <purevol> speaking of which..i guess i should go and prepare!
[30-Sep-2009 09:26:14] <davetoo> where are you, physically?
[30-Sep-2009 09:26:49] <davetoo> I'm in California
[30-Sep-2009 09:29:17] <otakup0pe> hmm i don't seem to be able to add OS products/manufacturers
[30-Sep-2009 09:29:20] <otakup0pe> i must be doing it wrong
[30-Sep-2009 09:29:37] <otakup0pe> purevol: what's the class subject matter ? i kinda miss teaching.
\n[30-Sep-2009 09:38:53] <Dieterbe_> davetoo> We need a better\nway to say "this is the set of processes that should be running on this\nclass of systems, regardless of whether you find it running during\nmodeling or not" <-- i definitely agree with that
\n[30-Sep-2009 09:41:28] <Dieterbe_> 16:16:36 < davetoo>\nThere might be some bad regex syntax in processes from some of the\nenterprise zenpacks?
[30-Sep-2009 09:41:31] <Dieterbe_> 16:17:03 < davetoo> I found that the httpd monitor was looking in the parameters list
[30-Sep-2009 09:41:42] <Dieterbe_> aha that may be it. i use zenoss core but i use the httpmonitor zenpack
[30-Sep-2009 10:21:56] <rmatte> good morning folks
[30-Sep-2009 10:22:33] <eidolon> mornin rmatte
[30-Sep-2009 10:22:36] <eidolon> how goes the fight?
[30-Sep-2009 10:22:41] <rmatte> epic
[30-Sep-2009 10:22:44] <rmatte>
[30-Sep-2009 10:23:04] <rmatte> just got out of a meeting, been assigned my new project for the week
[30-Sep-2009 10:23:15] <rmatte> automating report dumps using scripts I wrote
\n[30-Sep-2009 10:23:28] <eidolon> so here's a setup question -\nmore 'how do folks do this'. we have 2 levels of alerting we deal\nwith. warnings and error/crits. error/crits go to pagers. pagers are\n12345@sprint.com or the like. do people make 'user' and 'user-pager'\nentries in Settings/Users and set up groups for each?
[30-Sep-2009 10:23:33] <eidolon> we don't have an SNPP gateway.
\n[30-Sep-2009 10:24:17] <rmatte> ok, so... you only want certain\nalerts to be sent out to all pager, or certain alerts to be sent to\ncertain pagers?
[30-Sep-2009 10:26:14] <eidolon>\ni want all crits and errors to go to a list of users' pagers. (i'm\ngoing to do that with a group, in the group i'll have 'dave-pager'\n'bob-pager' 'phil-pager'). warnings will go to email (dave, bob, phil)
\n[30-Sep-2009 10:26:52] <Brad_K> otakup0pe: see\nblogs/ryanmatte/2009/09/24/specifying-hardware-and-software-product-info-in-245-until-bug-is-fixed
[30-Sep-2009 10:26:53] <rmatte> ok
\n[30-Sep-2009 10:27:20] <rmatte> Brad_K: well, click on the link\nat the top of that one to be sent to the guide which explains how to\nactually patch it
[30-Sep-2009 10:27:36] <rmatte> eidolon: so what's your question then?
[30-Sep-2009 10:28:06] <rmatte> Brad_K: oh, he was asking how to actually add products hehe
[30-Sep-2009 10:28:31] <eidolon> well, 'is that the way to do it'?
\nthat seems to be a good solution, but i could see some zenoss-way of\ngoing "oh no, don't do that, click here, here, here, and here, and\nyou'll see a better tool for it!"[30-Sep-2009\n10:29:47] <rmatte> eidolon: nope, that sounds about right, the\nonly other way that I can think of to do it would be to write a script\nthat sends the SMS messages and then set it up to be executed by Zenoss\nwhen events come in (Event Manager -> Commands)
[30-Sep-2009 10:29:58] <eidolon> glrk
[30-Sep-2009 10:30:01] <eidolon> don't think i like that one.
[30-Sep-2009 10:30:03] <eidolon> okay,thanks
[30-Sep-2009 10:30:03] <rmatte> eidolon: make sure that you add a filter to the alerting rule
[30-Sep-2009 10:30:15] <rmatte> to only alert on events higher than info
[30-Sep-2009 10:30:32] <eidolon> yeah. i got it
[30-Sep-2009 10:30:33] <eidolon> thanks
[30-Sep-2009 10:31:58] <rmatte> np
\n[30-Sep-2009 10:40:26] <rmatte> hmmmm, keeps getting cycle\nexceeded on zenprocess on one of my boxes, might have to bump parallel\nup to 3
[30-Sep-2009 10:42:27] <rmatte> hmmm, zenprocess doesn't support the parallel option, weird
[30-Sep-2009 10:42:51] <kmtsun> hello everyone
[30-Sep-2009 10:42:56] <rmatte> hello
[30-Sep-2009 10:45:21] <rmatte> hmmm, cycle times in general are extremely elevated, started on tuesday
\n[30-Sep-2009 10:52:57] <chudler> cycle times could go up if just\nsome device is responding slowly. Its one thing that sucks about a\npolling architecture.
[30-Sep-2009 10:53:34] <rmatte> actually, I think it's related to the link
\n[30-Sep-2009 10:53:47] <rmatte> we monitor through tunnels, looks\nlike the same thing is happening on our other Zenoss servers too
[30-Sep-2009 10:53:48] <cgibbons> hurm
[30-Sep-2009 10:53:56] <chudler> yeah, I meant to mention that many other reasons it could go up :-)
[30-Sep-2009 10:53:57] <rmatte> though it's less noticeable on them since they have a lot less devices
\n[30-Sep-2009 10:57:22] <chudler> someday I really hope to make a\nplugin that uses check_mk http://mathias-kettner.de/check_mk
[30-Sep-2009 10:57:26] <chudler> Its probably just a dream, though
[30-Sep-2009 11:07:01] <ke4qqq> anyone have a list of decent mobile phones that work with usb and linux to send snpp/sms messages via linux?
[30-Sep-2009 11:08:17] <chudler> http://www.sierrawireless.com/product/usbmodems.aspx <-- those work good
\n[30-Sep-2009 11:11:25] <chudler> We use the Compass 885. They\nprovide a kernel module and userland tools which are very easy to use.
\n[30-Sep-2009 11:17:51] <rmatte> chudler: if check_mk is already a\nnagios plugin it should integrate fine with Zenoss (with a little\nmassaging of course)
[30-Sep-2009 11:18:25] <otakup0pe> Brad_K: thx
[30-Sep-2009 11:18:27] <rmatte> oh, I see what it does
[30-Sep-2009 11:18:29] <rmatte> interesting
[30-Sep-2009 11:27:07] <ckrough> most providers have an email address for sms, you could just send them as an email to that
\n[30-Sep-2009 11:28:54] <chudler> we did that, but it didnt work\nwhen the network was down, plus they throttled us hard. An truly\nrhobust way to do it is dialing out on a landline, but we wanted to\nreceive SMS on zenoss also.
[30-Sep-2009 11:29:14] <ckrough> yeah, AT&T is not real happy
[30-Sep-2009 11:29:19] <ckrough> we've worked some things out with them
[30-Sep-2009 11:29:34] <ckrough> pushing close to 2000 sms messages per month per user for around 20 users
[30-Sep-2009 11:29:37] <rmatte> AT&T are horrible
[30-Sep-2009 11:29:54] <ckrough> its been successful though
[30-Sep-2009 11:30:35] <chudler> for Serious Enterprises, they rent 1U rack-mount SMS gateways. not cheap ;-)
[30-Sep-2009 11:35:09] <ckrough> hey now, Im a serious enterprise
[30-Sep-2009 11:35:28] <rmatte> hehe
[30-Sep-2009 11:35:55] <rmatte> If you're a serious enterprise, why aren't you paying us to monitor your stuff?
[30-Sep-2009 11:36:23] <ckrough> you cant handle my stuff
[30-Sep-2009 11:36:35] <rmatte> probably could given enough time
[30-Sep-2009 11:36:40] <ckrough>
[30-Sep-2009 11:36:59] <rmatte> I've gotten Zenoss to do some pretty fun stuff
\n[30-Sep-2009 11:37:12] <exarkun> What other ZenPacks does "Http\nMonitor" depend on? The community page says it has no dependencies,\nbut when I try to use it, I get an error that seems to indicate there\nis a missing dependency.
[30-Sep-2009 11:37:37] <rmatte> exarkun: none
[30-Sep-2009 11:37:47] <rmatte> paste the error
[30-Sep-2009 11:38:01] <rmatte> (if it's significantly long, use pastebin.com please)
[30-Sep-2009 11:38:08] <exarkun> http://pastebin.com/m3abaf380
[30-Sep-2009 11:39:03] <rmatte> you tried restarting Zenoss as I had suggested yesterday?
[30-Sep-2009 11:40:07]\n<shentonfreude> newbie question: can Zenos be used for "asset\nmanagement" of IP-less devices like video projectors? If I try an add a\ndevice without IP the modelling fails since it can't be pinged. Or\nperhaps there's a zenpack to do this common task? thanks
[30-Sep-2009 11:40:16] <exarkun> No, I probably forgot to restart it after I installed that zenpack.
[30-Sep-2009 11:42:20] <rmatte> exarkun: that's your issue
[30-Sep-2009 11:42:48] <exarkun> waiting for it to restart now
[30-Sep-2009 11:42:52] <rmatte> shentonfreude: this is the way I'd do it...
[30-Sep-2009 11:43:06] <exarkun> is it usual for the "zopectl" step to take several minutes?
[30-Sep-2009 11:43:20] <rmatte> if it has been up for a while it could take like 30 seconds or so
[30-Sep-2009 11:43:23] <exarkun> (also, that's a pretty bad failure mode for forgetting to restart
how about a "You forgot to restart" page)[30-Sep-2009 11:43:24] <rmatte> shouldn't take much longer
[30-Sep-2009 11:43:34] <exarkun> It's been up for an hour or so.
[30-Sep-2009 11:43:50] <exarkun> The whole restart process is like a 5 minute ordeal.
\n[30-Sep-2009 11:43:55] <rmatte> exarkun: the whole restarting\nafter installing a ZenPack is something that we pretty much drill in to\npeople's heads around here
[30-Sep-2009 11:44:12] <rmatte> exarkun: is it running on a VM?
\n[30-Sep-2009 11:44:28] <exarkun> rmatte: Maybe if the UI\npresented a more understandable error you wouldn't have to do that
[30-Sep-2009 11:44:31] <exarkun> rmatte: yes
\n[30-Sep-2009 11:44:45] <rmatte> exarkun: If it's taking that long\nyou're running it on something with not enough resources, the only\nslower Zenoss box that I have is my lab box, and it doesn't have enough\nresources allocated to it
[30-Sep-2009 11:44:58] <rmatte> and it's not even as slow as you're describing
\n[30-Sep-2009 11:45:33] <rmatte> exarkun: the UI is just telling\nyou that it knows the zenpack is broken, as for actually knowing why\nit's broken, that would take quite a bit of code and it's really\nunnecessary in my opinion
[30-Sep-2009 11:45:41] <rmatte> shentonfreude: ok, sorry, getting back to your question...
[30-Sep-2009 11:45:48] <rmatte> shentonfreude: In Zenoss, go to Settings
[30-Sep-2009 11:45:57] <rmatte> you'll see a box that says State Conversions
\n[30-Sep-2009 11:46:11] <rmatte> you'll want to add a new state to\nrepresent these devices, not sure what you want to call it
\n[30-Sep-2009 11:46:21] <exarkun> rmatte: An AttributeError in the\nerror reporting process isn't the most transparent report about a\nzenpack being broken.
[30-Sep-2009 11:46:21] <rmatte> I've added states for Standby, Spare, and Stacked
\n[30-Sep-2009 11:46:46] <rmatte> so let's say you want to add one\ncalled "Stacked", you need to add it with a value below Decommissioned,\nso that Zenoss doesn't try to monitor it
[30-Sep-2009 11:46:55] <rmatte> so you'd add: Stacked:-2
[30-Sep-2009 11:46:58] <rmatte> or whatever
\n[30-Sep-2009 11:47:34] <rmatte> after you've done that, just add\nthe devices with bogus IPs and then rename them to whatever you wish,\nmake sure that when you're adding them that they are in the production\nstate that you created
[30-Sep-2009 11:48:23]\n<rmatte> exarkun: I agree that the ZenPack system could be\nbetter, but I'd rather see them focus on other things than that at the\nmoment
[30-Sep-2009 11:48:39] <rmatte> exarkun:\nit's not something that's pressing enough for them to deal with in the\nimmediate future
[30-Sep-2009 11:49:01] <ckrough> is there a list of major goals? (not trac)
[30-Sep-2009 11:49:03]\n<rmatte> if you go to dev.zenoss.com/trac you can take a look at\nthe massive number of tickets that they are currently addressing or\nhave backlogged and you'll understand why I say that
[30-Sep-2009 11:49:10] * exarkun shrugs
[30-Sep-2009 11:49:17] <ckrough> or maybe trac has last list and I didnt look well
[30-Sep-2009 11:49:20] <exarkun> I'm not trying to dictate priorities
\n[30-Sep-2009 11:49:34] <rmatte> ckrough: I've personally seen a\nplan that they have layed out during a Zenoss Masters webex session
[30-Sep-2009 11:49:37] <exarkun> Just saying buggy error reporting in the ui isn't cool.
\n[30-Sep-2009 11:50:24] <rmatte> basically, King Crab is going to\nbe where they try to resolve as many annoyances as possible, then the\nfocus for stone crab is going to be the new UI, afterwards so of the\nfocusses I've seen are reporting, and performance improvements.
\n[30-Sep-2009 11:51:11] <rmatte> exarkun: I agree, but now that\nyou know to restart Zenoss if you've installed a ZenPack and it's\ncomplaining, you'll do it next time
[30-Sep-2009 11:51:27] <rmatte> You don't have to do it for every ZenPack, just certain ones
\n[30-Sep-2009 11:52:23] <rmatte> ckrough: I wish I had saved that\nwebex session, I could have shown you, just recalling off the top of my\nhead. If you ask Matt I'm sure he could dig it up and send it to you.
[30-Sep-2009 11:55:33] <ckrough> not important
\n[30-Sep-2009 11:57:26] <rmatte> hmmm lovely, looks like one of\nthe ethernet adapters on our main blade is failing, that would account\nfor the higher config times
[30-Sep-2009 12:02:11] <rmatte> hmmm, but there are only a few errors for it and I'm not sure when they are from
\n[30-Sep-2009 12:23:43] <__jd__> rmatte, apparently zenpatch 15003\nfixes the issue I described yesterday with the spurious auto-clearing
[30-Sep-2009 12:25:49] <shentonfreude> rmatte: thanks for the pointers
[30-Sep-2009 12:26:09]\n<rmatte> ummm, __jd__, you said that you were using 2.4.5\nyesterday, if you had that problem you can't have been using 2.4.5
[30-Sep-2009 12:26:14] <rmatte> were you at 2.4.4 or something?
[30-Sep-2009 12:26:37] <rmatte> actually, I think that one got patched in 2.4.4
[30-Sep-2009 12:27:34] <rmatte> hmmm, patch was created 2 weeks ago eh?
[30-Sep-2009 12:27:48] <rmatte> maybe they did just fix that recently, thought that got patched ages ago
[30-Sep-2009 12:27:55] <rmatte> they really need to release 2.4.6
[30-Sep-2009 12:27:59] <rmatte> 2.4.5 is so broken
[30-Sep-2009 12:35:51] <ckrough> did 2.4.5 release?
[30-Sep-2009 12:35:57] <ckrough> I thought it was in beta3
[30-Sep-2009 12:37:06] <rmatte> that's 2.5
[30-Sep-2009 12:37:07] <rmatte> lol
[30-Sep-2009 12:37:13] <rmatte> 2.4.5 was the latest official release
\n[30-Sep-2009 12:37:30] <rmatte> and it still has some pretty\nobvious bugs in it, they need to patch 2.4.5 with the latest fixes and\nrelease it as 2.4.6
[30-Sep-2009 12:37:43] <rmatte> though they may not even bother since King Crab is supposed to release in like 3 weeks
[30-Sep-2009 12:38:03] <rmatte> hopefully they don't rush the release just for the sake of getting it out though
[30-Sep-2009 12:42:45] <__jd__> they had a patch for 2.5 earlier I guess, but then backported it to 2.4.5
\n[30-Sep-2009 12:42:59] <__jd__> not sure how long that took..\nactually not even sure if it fixes the bug yet.. but I'm hopeful
[30-Sep-2009 12:44:13] <rmatte> well, some patches to 2.5 will cleanly apply to 2.4.5
[30-Sep-2009 12:44:15] <rmatte> depends on the patch
[30-Sep-2009 12:45:23] <__jd__> yea
[30-Sep-2009 12:45:28] <__jd__> zenpatch is a clever hack
[30-Sep-2009 12:45:47] <__jd__> something you can do when you're not distributing binaries I suppose
\n[30-Sep-2009 12:46:49] <rmatte> well, zenpatch is basically just\ndownloads the diff file and automatically uses diff to apply the patch
[30-Sep-2009 12:46:57] <rmatte> so in reality, diff is clever
[30-Sep-2009 12:47:03] <rmatte> well "patch" rather
[30-Sep-2009 12:47:20] <rmatte> zenpatch is really nice though, big time saver
[30-Sep-2009 12:47:43] <__jd__> I know. I just meant it was a clever hack because of how simple it is
[30-Sep-2009 12:47:56] <__jd__> it's very convenient
[30-Sep-2009 12:48:17] <rmatte> yeh
[30-Sep-2009 12:48:18] * __jd__ needs to break something now to test that last patch
[30-Sep-2009 12:49:06] * rmatte breaks __jd__'s leg
[30-Sep-2009 12:49:13] <rmatte> oh, you meant software-wise
[30-Sep-2009 12:53:47] <__jd__> broken -- we'll see what happens
[30-Sep-2009 12:53:54] <purevol> who has crabs?
[30-Sep-2009 12:54:46] <__jd__> Do you happen to know how to tell zenoss that its hostname is not localhost.localdomain?
\n[30-Sep-2009 12:55:43] <__jd__> the hostname of the server is not\nlocalhost.localdomain, so I'm not sure where it's getting it from,\nunless it's just a dedfault
[30-Sep-2009 12:55:44] <__jd__> default
[30-Sep-2009 12:56:21] <__jd__> oh wait.. I think I might know what it's doing
[30-Sep-2009 13:01:25] <cgibbons> do tell
[30-Sep-2009 13:02:14] <ckrough> its looking at 127.0.0.1 in /etc/hosts would be my guess
[30-Sep-2009 13:02:31] <ckrough> you probably dont want to change that line in /etc/hosts...
[30-Sep-2009 13:02:44] <__jd__> yea, I had <myhostname> 127.0.0.1
[30-Sep-2009 13:02:59] <__jd__> or vice versa
[30-Sep-2009 13:03:23] <__jd__> so I guess it must look up the IP of `hostname` and then do a reverse lookup on that
[30-Sep-2009 13:03:30] <ckrough> lol
[30-Sep-2009 13:03:34] <rmatte> cgibbons: under collector performance, do you know what the Cycle Times are measured in?
[30-Sep-2009 13:03:35] <__jd__> kind of round-about way of getting a hostname
[30-Sep-2009 13:03:48] <rmatte> cgibbons: is it the total time it's taking to complete each cycle?
[30-Sep-2009 13:03:58] * ckrough was wondering that yesterday
[30-Sep-2009 13:03:58] <rmatte> cgibbons: or is it response time in milliseconds?
[30-Sep-2009 13:04:31] <rmatte> cgibbons: the cycle time for zenperfsnmp suddenly jumped to 62.53 when it was previously at around 15-20
[30-Sep-2009 13:04:33]\n<cgibbons> the value is the number of seconds a single\n'collection loop' takes - not sure if it gets translated for the graph\nor not
[30-Sep-2009 13:04:39] <rmatte> trying to figure out why it suddenly jumped so high...
[30-Sep-2009 13:05:01] <cgibbons> for which daemon?
[30-Sep-2009 13:05:14] <rmatte> also, it's showing nan for 4 different daemons, even though those daemons are operating fine
[30-Sep-2009 13:05:25] <rmatte> zenperfsnmp is the one that I'm seeing as much higher than it was previously
[30-Sep-2009 13:05:44] <rmatte> but I'm getting threshold alerts for zenprocess, zeneventlog, etc...
[30-Sep-2009 13:05:49] <rmatte> for cycle time thresholds
[30-Sep-2009 13:06:49]\n<rmatte> zenperfsnmp doesn't appear to be going over threshold,\nbut it's very odd that the cycle times suddenly increased so\ndrastically on tuesday morning with no increase in collected events
[30-Sep-2009 13:07:04] <cgibbons> yeah, and your CPU load on the collector server itself is okay?
[30-Sep-2009 13:07:26]\n<rmatte> we actually removed the VM from the resource pool to\ngive it unlimited resources, and we're still seeing this
[30-Sep-2009 13:07:38] <rmatte> so I'm thinking possible network issue, or something to do with Zenoss
[30-Sep-2009 13:08:41]\n<cgibbons> usually that's either a) too much load on a daemon, b)\nnetwork problems with a lot of slow to respond devices/timeouts, c)\ngeneral host overload
[30-Sep-2009 13:09:00]\n<cgibbons> zenperfsnmp prints out these cycle time stats in the\nlog if you want to look, you might see an increase in the # of slow to\nrespond devices in there?
[30-Sep-2009 13:09:32] <rmatte> hmmm, I'll check it out, just got heartbeat failures for zenping and zentrap
[30-Sep-2009 13:11:02] <rmatte> I see the odd line like: 2009-09-30 10:22:09 WARNING zen.zenprocess: Problem devices: ['XP03BMSTv01']
[30-Sep-2009 13:14:18] <rmatte> hmmm
[30-Sep-2009 13:14:35] <rmatte> the usual time that it takes for zenping is:
[30-Sep-2009 13:14:36] <rmatte> 2009-09-30 12:07:39 INFO zen.ZenPing: Finished pinging 341 jobs in 3.20 seconds
[30-Sep-2009 13:14:56] <rmatte> just before that:
[30-Sep-2009 13:14:58] <rmatte> 2009-09-30 12:06:42 INFO zen.ZenPing: Finished pinging 344 jobs in 6.23 seconds
[30-Sep-2009 13:15:18] <rmatte> wondering why it suddenly has 344 jobs instead of 341
\n[30-Sep-2009 13:15:31] <ckrough> when we get to a high number of\ndevices on a collector, a single "problem" device seems to cause major\ncycle issues
[30-Sep-2009 13:15:38] <ckrough> its actually one of our top scaling problems
[30-Sep-2009 13:15:45] <rmatte> lovely
[30-Sep-2009 13:15:53] <ckrough> but the number of jobs doesnt change
[30-Sep-2009 13:16:07] <rmatte> 2009-09-30 11:39:41 INFO zen.ZenPing: Finished pinging 349 jobs in 6.05 seconds
[30-Sep-2009 13:16:08] <ckrough> particularly with zenperfsnmp around 1600 devices.
[30-Sep-2009 13:16:12] <rmatte> seems to be fluctuating
[30-Sep-2009 13:16:24] <rmatte> Like, check this out...
[30-Sep-2009 13:16:24] <rmatte> 2009-09-30 11:35:37 INFO zen.ZenPing: Finished pinging 341 jobs in 1.99 seconds
[30-Sep-2009 13:16:25] <rmatte> 2009-09-30 11:36:39 INFO zen.ZenPing: Finished pinging 343 jobs in 3.31 seconds
[30-Sep-2009 13:16:25] <rmatte> 2009-09-30 11:37:38 INFO zen.ZenPing: Finished pinging 343 jobs in 3.26 seconds
[30-Sep-2009 13:16:25] <rmatte> 2009-09-30 11:38:37 INFO zen.ZenPing: Finished pinging 341 jobs in 1.81 seconds
\n[30-Sep-2009 13:16:26] <ckrough> it runs great when everybody is\nhappy, when one device goes down it can hang beyond 300s, and it seems\nto snowball
[30-Sep-2009 13:16:33] <rmatte> how the heck does it magically get 2 more jobs?
[30-Sep-2009 13:16:53] <__jd__> it's going for extra credit
[30-Sep-2009 13:16:57] <ckrough> that little guy? dont worry about that little guy
[30-Sep-2009 13:17:02] <rmatte> lol
[30-Sep-2009 13:17:09] <rmatte> super troopers!
[30-Sep-2009 13:17:13] <cgibbons> it can split up OID requests across tasks and repeat slow tasks; it's a complex daemon
[30-Sep-2009 13:17:13] <ckrough> win
[30-Sep-2009 13:17:33] <rmatte> well, this is zenping, so no OID requests
\n[30-Sep-2009 13:17:33] <cgibbons> the new framework doesn't have\na single collection loop anymore so deciding when it's 'slow' is going\nto be a learning experience
[30-Sep-2009 13:18:00] <rmatte> 2009-09-30 08:06:40 INFO zen.ZenPing: Finished pinging 347 jobs in 6.23 seconds
[30-Sep-2009 13:18:00] <rmatte> 2009-09-30 08:07:39 INFO zen.ZenPing: Finished pinging 358 jobs in 5.01 seconds
[30-Sep-2009 13:18:00] <rmatte> 2009-09-30 08:08:37 INFO zen.ZenPing: Finished pinging 350 jobs in 3.49 seconds
[30-Sep-2009 13:18:10] <rmatte> is it on crack?
[30-Sep-2009 13:18:11] <rmatte> lol
[30-Sep-2009 13:19:03] <ckrough> cgibbons: can you extrapolate on that?
\n[30-Sep-2009 13:19:26] <ckrough> cgibbons: when we get a problem\ndevice zenperfsnmp waits until timeout then the cycle can fail since we\nare already close to the limit
[30-Sep-2009 13:19:42] <rmatte> yeh, this client has several problem devices
[30-Sep-2009 13:19:51] <cgibbons> zenperfsnmp didn't change to the new framework for 2.5, so this doesn't affect it yet
[30-Sep-2009 13:19:54] <rmatte> where you can literally read the output of an snmpwalk line by line as it runs
[30-Sep-2009 13:19:54] <ckrough> cgibbons: is that not expected behavior in 2.3.3?
[30-Sep-2009 13:20:03] <ckrough> ah
[30-Sep-2009 13:20:09]\n<cgibbons> but in the new collector framework the way the tasks\nare scheduled for collection are different, so there's no longer a\nsingle 'everything has to run here' loop
[30-Sep-2009 13:20:22] <rmatte> but it was never this bad, it's strange that the cycle times suddenly balooned up
\n[30-Sep-2009 13:20:23] <cgibbons> it'll be a big boon for most of\nthe collectors since a slow device won't affect the rest of the world
[30-Sep-2009 13:20:35] <ckrough> awesome
[30-Sep-2009 13:20:41] <rmatte> cgibbons: is this in King Crab?
[30-Sep-2009 13:20:49] <cgibbons> yeah, for zenwin, zeneventlog, zenprocess, and zenwinperf
[30-Sep-2009 13:20:56] <rmatte> perfect, can't wait
\n[30-Sep-2009 13:21:29] <cgibbons> zenwinperf especially i expect\na big improvement on the high-end of scalability limits. i think we've\ngot one customer with about 600 devices on a single collector right\nnow, so i'd wager we could go much higher with the new version
[30-Sep-2009 13:22:17] <rmatte> yup
[30-Sep-2009 13:22:23] <rmatte> definitely looking forward to that
\n[30-Sep-2009 13:22:48] <ckrough> cgibbons: how does it get past\nslow/nonresponsive devices? moves the job to a bad device thread or\nsomething?
[30-Sep-2009 13:23:23] <cgibbons> and\ni don't know when it'll be coded, but one of the features we put into\nthe design was the ability to easily extend the framework so that tasks\ncan be scheduled via subprocesses (sorta like zenhub), so that'll give\nbetter performance on multi-core systems without the need to configure\nmore software collectors
[30-Sep-2009 13:23:59] <rmatte> well, don't you technically already have that?
[30-Sep-2009 13:24:11] <rmatte> "parallel 2"
[30-Sep-2009 13:24:11] <rmatte> or whatever
[30-Sep-2009 13:24:11] <rmatte> works on certain daemons
\n[30-Sep-2009 13:24:36] <cgibbons> right now everything is done\nvia asynchronous I/O, so there's no real waiting unless it has been\ncoded to do so - and it has. so if you can imagine a single loop that\nwaits for all 100 devices to complete before it begins the next cycle\n(zenperfsnmp is more complicated).
[30-Sep-2009 13:24:37] <jb> hey cgibbons
[30-Sep-2009 13:24:53] <rmatte> yeh
[30-Sep-2009 13:25:01]\n<cgibbons> in the new world, device A will run every 5 minutes\nand device B might run every 5 minutes, but if device C takes 7 minutes\nto respond, device A & B still keep being scheduled without having\nto wait for C to finish.
[30-Sep-2009 13:25:51] <rmatte> are there giant killer robots in this new world?
[30-Sep-2009 13:26:37] <ckrough> so what happens with 1000 devices all fall on the same cycle?
[30-Sep-2009 13:27:10] <ckrough> chaos. mass hysteria. dogs and cats living together.
\n[30-Sep-2009 13:27:23] <cgibbons> they're still independent of\none another from a collection framework, and they won't affect one\nanother
[30-Sep-2009 13:27:38] <ckrough> I was thinking more of the system resources
\n[30-Sep-2009 13:28:12] <ckrough> how many concurrent snmp\ncollections can you do before something gets io bound, and if it does,\nhow gracefully does it complete
[30-Sep-2009 13:28:15] <ckrough> just thinking out loud
\n[30-Sep-2009 13:28:34] <cgibbons> rmattte: yeah parallel is\nsimilar to the idea of independent scheduling for the daemons that\nsupport it, but it's not that robust
[30-Sep-2009 13:29:03]\n<cgibbons> SNMP is a little weird since it's a UDP-based\nprotocol, so your main limit is how many packets you can send out @\nonce without dropping them, and it's a pretty low limit
[30-Sep-2009 13:29:11] <cgibbons> lemme see, there's a constant in the zenperfsnmp daemon today for that
[30-Sep-2009 13:29:32] <cgibbons> 20-ish
[30-Sep-2009 13:29:41] <ckrough> 20 I think? but thats a limit of the target device not zenoss
[30-Sep-2009 13:29:44] <cgibbons> with "@ once" being undefined
[30-Sep-2009 13:29:57] <cgibbons> it's really a limit of the host's TCP/IP protocol stack
\n[30-Sep-2009 13:30:28] <cgibbons> you can do some experiments\neven with local echo stuff to do see how quickly UDP stuff starts to\ndrop packets
[30-Sep-2009 13:30:43] <__jd__> it's a local kernel buffer most likely
[30-Sep-2009 13:30:47] <cgibbons> yep
[30-Sep-2009 13:30:48] <__jd__> if you overflow it, packets drop
[30-Sep-2009 13:31:06] <cgibbons> in TCP world it'll figure it out, but things built on UDP.... timeout!
\n[30-Sep-2009 13:31:31] <cgibbons> so zenperfsnmp tries to be\nintelligent about that, and limits the maximum concurent packets it\nsends at once, and also how big requests are so it doesn't overflow\npacket size limitations
[30-Sep-2009 13:32:07]\n<cgibbons> i'm not sure how much, if any, scaling improvement\nwe'll see with zenperfsnmp once it moves over since it's already rather\nrobust about this and a lot of features from it were put into the new\nframework so all the daemons get them
[30-Sep-2009 13:34:08] <ckrough> I was thinking of the maxoids per request for IOS
[30-Sep-2009 13:35:01] * ckrough hears this little voice screaming c++ in the back of his head
[30-Sep-2009 13:36:37] <cgibbons> yep there's that value too... looks like ours is 40
\n[30-Sep-2009 13:36:56] <cgibbons> i think there's some logic in\nthere to split up a request into smaller in some error conditions too,\nif it is close to the edge
[30-Sep-2009 13:39:14] <rmatte> since when is ntp.nasa.gov no longer a valid ntp source?
[30-Sep-2009 13:39:59] <rmatte> oh, it just didn't work the first time I tried it
[30-Sep-2009 13:40:00] <rmatte> weird
[30-Sep-2009 13:42:08] * ckrough leaves zenoss mode and goes back to php mode for a while
[30-Sep-2009 13:42:16] <ckrough> time to relearn using semi-colons. again.
[30-Sep-2009 13:42:25] <__jd__>
[30-Sep-2009 13:42:53] <rmatte> lol
[30-Sep-2009 13:42:59] <__jd__> that was directed not so much at semicolons, but at php
[30-Sep-2009 13:43:05] <ckrough> man this is ugly. I want to go back in time and smack myself
[30-Sep-2009 13:43:22] <__jd__> oh my
[30-Sep-2009 13:43:26] <__jd__> netsplit?
[30-Sep-2009 13:43:34] <rmatte> netjoin
[30-Sep-2009 13:43:42] <ckrough> freenode does those fairly often
[30-Sep-2009 13:43:56] <ckrough> all part of the fluids of the intarwebs
[30-Sep-2009 13:43:59] <rmatte> every IRC network does those fairly often lol
[30-Sep-2009 13:44:09] <__jd__> I've been away from IRC for a while
[30-Sep-2009 13:44:15] <__jd__> brings back efnet memories
[30-Sep-2009 13:44:21] <ckrough> feels good doesnt it
[30-Sep-2009 13:44:22] <jmp242> Anyone know why a linux process wouldn't autoclear when it comes back up?
\n[30-Sep-2009 13:44:26] <rmatte> they are actually kind of fun,\nsee a bunch of people leave the room at once: "NETSPLIT!!!"
[30-Sep-2009 13:44:38] <ckrough> jmp242: something different in the dedup?
[30-Sep-2009 13:44:43] <rmatte> jmp242: did it come back up with different options?
[30-Sep-2009 13:47:05]\n<jmp242> I believe one of the processes did, though one did not -\nso I'm wondering if the different options, does that end up a wholly\nnew process?
[30-Sep-2009 13:47:25] <jmp242> I expect so...
[30-Sep-2009 13:47:42] <jmp242> Well, ckrough weirdly enough, there WAS no CLEAR event
\n[30-Sep-2009 13:47:45] <rmatte> if there are different options on\nthe process Zenoss won't recognize it as the same process
[30-Sep-2009 13:47:47] <jmp242> so I could not compare dedupID
\n[30-Sep-2009 13:48:10] <jmp242> but I guess if Zenoss thinks it's\nnot the same process, it wouldn't generate a CLEAR event
[30-Sep-2009 13:48:17] <jmp242> so that seems most likely
[30-Sep-2009 13:48:28] <rmatte> correct
[30-Sep-2009 13:48:43] <jmp242> Ok, that was my mystery
[30-Sep-2009 13:48:45] <rmatte> Go to Processes (link in the Zenoss menu)
[30-Sep-2009 13:48:48] <rmatte> then find that process
[30-Sep-2009 13:48:55] <rmatte> click on it
[30-Sep-2009 13:48:58] <rmatte> go to the edit tab
\n[30-Sep-2009 13:48:59] <jmp242> did they ever fix the display\nissue with processes showing as UP / Green in the OS tab no matter what?
[30-Sep-2009 13:49:05] <rmatte> and set Ignore parameters to true
[30-Sep-2009 13:49:10] <rmatte> and click save
[30-Sep-2009 13:49:11] <rmatte> then remodel
[30-Sep-2009 13:49:21] <jmp242> yea, rmatte that is a good idea, except these are glassfish servers, hence java
[30-Sep-2009 13:49:23] <rmatte> (won't work if you're monitoring multiple versions of the same process)
[30-Sep-2009 13:49:26] <rmatte> ah
[30-Sep-2009 13:49:28] <rmatte> gotcha
[30-Sep-2009 13:49:33] <jmp242> and so you need the parameters to use it at all
\n[30-Sep-2009 13:49:46] <rmatte> well, you might want to look in\nto why the service is coming back up with different options
[30-Sep-2009 13:49:50] <jmp242> which is fine, I just needed to understand what happened
[30-Sep-2009 13:49:58] <rmatte> k
[30-Sep-2009 13:50:01] <jmp242> the Linux guy changed the process options, and wondered why the event didn't autoclear
[30-Sep-2009 13:50:08] <jmp242> when the process came up
[30-Sep-2009 13:50:15] <rmatte> ah
[30-Sep-2009 13:50:19] <rmatte> yeh, that's why
[30-Sep-2009 13:50:28] <rmatte> any time they are changed you need to remodel
\n[30-Sep-2009 13:50:35] <jmp242> did they ever fix the display\nissue with processes showing as UP / Green in the OS tab no matter what?
[30-Sep-2009 13:50:46] <jmp242> Is that in the 2.5 beta anyone?
[30-Sep-2009 13:50:49] <rmatte> jmp242: I would imagine so
[30-Sep-2009 13:51:02] <jmp242> cause I was also getting that in 2.4.5
[30-Sep-2009 13:51:03] <rmatte> I've never noticed an issue with that
[30-Sep-2009 13:51:09] <rmatte> oh?
[30-Sep-2009 13:51:12] <rmatte> hmmm
\n[30-Sep-2009 13:51:23] <jmp242> I do wonder if it has to do with\nthe loooooong process name + options with the glassfish server
[30-Sep-2009 13:51:31] <jmp242> and that database size overflow
[30-Sep-2009 13:51:33] <rmatte> could be
[30-Sep-2009 13:51:36] <jmp242> that the fix it code got lost
[30-Sep-2009 13:51:40] <jmp242> with the forum move
[30-Sep-2009 13:51:53] <jmp242> PITA... I don't know how to recover from that - mailarchive I guess via google
\n[30-Sep-2009 13:52:32] <rmatte> I love how an entire commission\nis pending because I can't login to one frigging device, which should\nhave been configured to allow me access, gah
[30-Sep-2009 13:53:12] <jmp242> mmmm sounds fun
[30-Sep-2009 13:53:14] <rmatte> jmp242: yeh, or just use cached pages on google
[30-Sep-2009 13:56:29] <rmatte> I think I'm going to setup a website to host the old version of ZipTie
[30-Sep-2009 13:58:22] <jmp242> why?
[30-Sep-2009 13:59:24] <ckrough> mrayzenoss: docs/DOC-3563
[30-Sep-2009 13:59:32] <ckrough> mrayzenoss: that API doc cant be unzipped
[30-Sep-2009 13:59:42] <ckrough> [crow@enki Zenoss_API]$ tar xzvf ~/Download/zenoss-2.4.5-apidocs.tar.gz
[30-Sep-2009 13:59:43] <ckrough> tar: This does not look like a tar archive
[30-Sep-2009 14:00:13] <jmp242> rmatte: thanks for the google cache idea - much faster!
[30-Sep-2009 14:00:19] <jmp242> still...
[30-Sep-2009 14:00:35] <jmp242> sad I have to do that. Just spent this morning fixing links on the Community FAQ
[30-Sep-2009 14:00:52] <mrayzenoss> ckrough: bizarre, I just redownloaded and tar -xzvf'd it
[30-Sep-2009 14:01:03] <ckrough> meh. it failed for both me and Pete
[30-Sep-2009 14:01:39] <ckrough> trying it with wget
[30-Sep-2009 14:02:57] <ckrough> uh. with firefox (twice) file size is 7531240, wget snags 7600927. wget version downloads fine
[30-Sep-2009 14:03:10] <ckrough> insert WTH here<>
[30-Sep-2009 14:03:21] <mrayzenoss> agreed
[30-Sep-2009 14:03:36]\n<ckrough> someone else running FF on linux try that download?\nservlet/JiveServlet/download/3563-1-1767/zenoss-2.4.5-apidocs.tar.gz
[30-Sep-2009 14:03:54] <mrayzenoss> worked for me with FF and Chrome
[30-Sep-2009 14:04:04] <ckrough> os?
[30-Sep-2009 14:04:12] <mrayzenoss> OS X 10.6
[30-Sep-2009 14:05:02] <exarkun> Firefox on Linux, seems to work, no progress indicator though, md5sum is 21ab9f245aa77e4643d87067e90099d7
[30-Sep-2009 14:05:32] <mrayzenoss> same md5 here
[30-Sep-2009 14:06:44] <jmp242> What OS ckrough?
[30-Sep-2009 14:06:53] <jmp242> did you want me to try on Windows?
[30-Sep-2009 14:07:08] <ckrough> Fedora11, FF3.5.3. Pete is on centos5.3
[30-Sep-2009 14:07:53] <jmp242> mmm
[30-Sep-2009 14:08:34] <ckrough> wget c632f052a043583d0c267e7038407500, ff 21ab9f245aa77e4643d87067e90099d7
[30-Sep-2009 14:08:38] <ckrough> ff version wouldnt expand
[30-Sep-2009 14:08:43] <ckrough> Im running 64bit
[30-Sep-2009 14:09:37] <ckrough> oh well, looks like it's on our end. not sure wth that could be. thanks.
[30-Sep-2009 14:13:36] <exarkun> Oh. I didn't try to unpack it. Tar doesn't like it.
\n[30-Sep-2009 14:16:45] <__jd__> it could be zenoss' side too if\nit's happening all over the place.. maybe it's user-agent detection\ngone wrong
[30-Sep-2009 14:17:25] <ckrough> yeah. I thought it was just me. but exarkun had the same issue
[30-Sep-2009 14:19:25] <mrayzenoss> if I could get a couple of people to verify it, I'll raise it up with IT/Jive
\n[30-Sep-2009 14:20:13] <mrayzenoss> anyone listening, please hit\ndocs/DOC-3563 and 'tar -xzvf\nzenoss-2.4.5-apidocs.tar.gz'
[30-Sep-2009 14:21:50]\n<xamox> quick question, do I need to install zenoss on all my\nclients, or just have snmp setup and working?\\
[30-Sep-2009 14:21:58] <ckrough> the latter
[30-Sep-2009 14:22:06] <ckrough> zenoss is agentless for most monitoring
[30-Sep-2009 14:22:16] * xamox claps
[30-Sep-2009 14:22:20] <xamox> awesome
[30-Sep-2009 14:22:36] <xamox> That's all I needed to know
[30-Sep-2009 14:22:41] <xamox> Thanks
[30-Sep-2009 14:22:46] <ckrough> you'll be back
[30-Sep-2009 14:22:49] <ckrough> gl
[30-Sep-2009 14:22:50] <xamox> haha, probably
[30-Sep-2009 14:23:01] <jmp242> no, you will be xamox
[30-Sep-2009 14:23:11] * xamox hides in fear
[30-Sep-2009 14:23:18] <mrayzenoss> don't scare him away
[30-Sep-2009 14:23:19] <ckrough> its all good.
[30-Sep-2009 14:23:21] <xamox> If I can talk my boss into it, then I will be back
[30-Sep-2009 14:24:39] <ckrough> we can help with most objections
[30-Sep-2009 14:24:40] <mrayzenoss> we're recreating the issue with the tarball
[30-Sep-2009 14:24:46] <ckrough> thanks mrayzenoss
[30-Sep-2009 14:24:54] <mrayzenoss> ckrough: can I email you one to test?
[30-Sep-2009 14:24:58] <ckrough> sure
[30-Sep-2009 14:24:59] <jb> mrayzenoss: still have a zenpack backlog?
[30-Sep-2009 14:25:10] <mrayzenoss> jb: yeah, I should go do that
[30-Sep-2009 14:25:17] <mrayzenoss> you're at the top
[30-Sep-2009 14:25:18] <jb>
[30-Sep-2009 14:25:23] <jb> just checking!
[30-Sep-2009 14:25:27] <mrayzenoss> sorry, been busy with the new site and Ohio LinuxFest
[30-Sep-2009 14:25:31] <jb> yeah, not a problem.
[30-Sep-2009 14:27:30] <jmp242> Umm, question on the process monitoring
\n[30-Sep-2009 14:27:59] <jmp242> how come Zenoss is expanding out\nthe entire commandline of the process vs just the regex match?
\n[30-Sep-2009 14:28:30] <jmp242> I mean, it seems like there has\nto be a better way to deal with these java processes etc than what\nwe've currently got...
[30-Sep-2009 14:29:08] <ckrough> "expanding out the entire commandline of the process vs just the regex"
[30-Sep-2009 14:29:12] <ckrough> can you give an example?
[30-Sep-2009 14:29:36] <ckrough> I've noticed in a lot of their code they append a '.*' to searches
[30-Sep-2009 14:30:16] <mrayzenoss> ckrough: check your email
[30-Sep-2009 14:30:20] <ckrough> kk
[30-Sep-2009 14:30:44] <ckrough> krough.org or work?
[30-Sep-2009 14:30:48] <mrayzenoss> work
[30-Sep-2009 14:31:05] <ckrough> k, exchange... will be a few.
\n[30-Sep-2009 14:31:17] <rmatte> yikes, just noticed that I didn't\nhave the operational status threshold set on my 64bit ethernet template
[30-Sep-2009 14:31:22] <mrayzenoss> if that works for you, I'll replace the tarball that's up there
[30-Sep-2009 14:32:16] <jmp242> sure.
[30-Sep-2009 14:32:22] <jmp242> we have a regex like
\n[30-Sep-2009 14:32:27] <jmp242> '/opt/SUNWappserver/jdk/bin/java\n-Dcom.sun.aas.instanceRoot=/opt/SUNWappserver/domains/domain1'
\n[30-Sep-2009 14:32:59] <jmp242> but zenoss will think it's a new\nprocess as I stated above if we, for instance, change the port number,\nand the process name in the device is like
\n[30-Sep-2009 14:33:16] <jmp242> '/opt/SUNWappserver/jdk/bin/java\n-Dcom.sun.aas.instanceRoot=/opt/SUNWappserver/domains/domain1\n-Dcom.sun.aas.ClassPathPrefix= -Dcom.sun.aas.ClassPathSuffix=\n-Dcom.sun.aas.ServerClassPath=\n-Dcom.sun.aas.classloader.appserverChainJars.ee=\n-Dcom.sun.aas.classloader.appserverChainJars=admin-cli.jar,admin-cli-ee.jar,j2ee-svc.jar\n-Dcom.sun.aas.classloader.excludesList=admin-cli.jar,appserv-upgrade.jar,sun-appserv-ant.jar
\n[30-Sep-2009 14:33:16] <jmp242>\n-Dcom.sun.aas.classloader.optionalOverrideableChain.ee=\n-Dcom.sun.aas.classloader.optionalOverrideableChain=webservices-rt.jar,webservices-tools.jar
\n[30-Sep-2009 14:33:16] <jmp242>\n-Dcom.sun.aas.classloader.serverClassPath.ee=/opt/SUNWappserver/hadb/4.4.3-6/lib/hadbjdbc4.jar,/opt/SUNWappserver/lib/SUNWjdmk/5.1/lib/jdmkrt.jar,/opt/SUNWappserver/hadb/4.4.3-6/lib/dbstate.jar,/opt/SUNWappserver/hadb/4.4.3-6/lib/hadbm.jar,/opt/SUNWappserver/hadb/4.4.3-6/lib/hadbmgt.jar,/opt/SUNWmfwk/lib/mfwk_instrum_tk.jar
\n[30-Sep-2009 14:33:19] <jmp242>\n-Dcom.sun.aas.classloader.serverClassPath=/opt/SUNWappserver/lib/install/applications/jmsra/imqjmsra.jar,/opt/SUNWappserver/imq/'
[30-Sep-2009 14:34:03]\n<ckrough> mrayzenoss: 16b7fcbe530d9cac41d9283013f7fb23 , it\nworks, but there are dotfiles in there I didnt see in the other one
[30-Sep-2009 14:34:31] <mrayzenoss> ckrough: hmm... those are probably my fault
[30-Sep-2009 14:34:46] <mrayzenoss> OSX'll do that
[30-Sep-2009 14:35:31] <mrayzenoss> I'm not seeing them
[30-Sep-2009 14:35:39] <ckrough> lol
[30-Sep-2009 14:36:31] <ckrough> other than them, everything looks good
[30-Sep-2009 14:36:35] <mrayzenoss> aight
\n[30-Sep-2009 14:37:25] <jmp242> I'm assuming it must have\nsomething to do with if you have multiple versions of the process that\ncould match on specific computers
[30-Sep-2009 14:38:03]\n<jmp242> but it seems to me that it would be better for the end\nuser if they could specify how specific they want to get using the\nexisting mechanism of the regex
[30-Sep-2009 14:38:13] <mrayzenoss> ok, new version of the API uploaded if you're looking for it
[30-Sep-2009 14:38:19] <mrayzenoss> docs/DOC-3563
[30-Sep-2009 14:43:09] <ckrough> mrayzenoss: sorry, that one fails too
[30-Sep-2009 14:43:17] <ckrough> mrayzenoss: must be something Jive related
[30-Sep-2009 14:43:30] <mrayzenoss> wtf...
[30-Sep-2009 14:43:36] <ckrough> 097792cad19d0903b044c0731837769b
\n[30-Sep-2009 14:43:50] <patzer> hi, I am trying to make my own\nzenpack, and following through the tutorial I notice that if I add a\nzproperty, it doesn't seem to show up unless I export the zenpack and\nthen install that .egg
[30-Sep-2009 14:44:11] <patzer> am I missing something here to avoid having to do this?
[30-Sep-2009 14:44:47] <mrayzenoss> ckrough: confirmed
\n[30-Sep-2009 14:44:51] <ckrough> mrayzenoss: Firefox doesnt know\nthe total size during download, no progress bar... which is odd
[30-Sep-2009 14:45:04] <mrayzenoss> I get different md5s between what I uploaded and what comes back down
[30-Sep-2009 14:46:05] <cgibbons> yikes
[30-Sep-2009 14:46:39] <mrayzenoss> ckrough: email version is OK, but downloaded is messed up?
[30-Sep-2009 14:46:45] <ckrough> mrayzenoss: correct
[30-Sep-2009 14:47:05]\n<ckrough> mrayzenoss: I've only downloaded zenpacks from the new\nsite via wget, but they were fine after unzipping and --install
[30-Sep-2009 14:47:25] <mrayzenoss> ckrough: we're debugging, gonna raise the issue with Jive
[30-Sep-2009 14:47:32] <mrayzenoss> perhaps it's .tar.gz related...
[30-Sep-2009 15:04:05] <cgibbons> so how many of y'all would like to write more zenpacks if it were even easier?
[30-Sep-2009 15:06:04] <ckrough> sure
[30-Sep-2009 15:06:57] <jmp242> sure
[30-Sep-2009 15:07:12] <ckrough> is this a trick question
[30-Sep-2009 15:07:18] <cgibbons> nope
[30-Sep-2009 15:07:30]\n<jmp242> I haven't ever had the itch to write a zenpack, though\nif there was something I was going to set up I'd give it a go
[30-Sep-2009 15:08:00]\n<jmp242> My problem is my coworkers can't come up with anything\nmuch to monitor that isn't OOTB that they actually would want to spend\ntime on
[30-Sep-2009 15:08:16] <jmp242> it shows how far Zenoss is advanced from the previous monitoring solution
[30-Sep-2009 15:08:36]\n<ckrough> yeah, most of the stuff I monitor is either ootb or so\nspecific it's not relevant to enough people to writer a zenpack for
[30-Sep-2009 15:10:46] <mrayzenoss> so a little digging I just did on the stuff on zenpacks.zenoss.org
[30-Sep-2009 15:11:05] <mrayzenoss> 64 of 96 hosted ZenPacks are more than a simple template export
[30-Sep-2009 15:11:23] <ckrough> are more or are not more
[30-Sep-2009 15:11:24] <mrayzenoss> so they have some sort of coding, not just exported from the UI
[30-Sep-2009 15:11:28] <ckrough> nuce
[30-Sep-2009 15:11:28] <cgibbons> surprised its that many
[30-Sep-2009 15:11:29] <ckrough> nice
[30-Sep-2009 15:11:33] <mrayzenoss> yeah me too
[30-Sep-2009 15:12:06] <ckrough> is there much collaboration on zenpacks? seems to be one-man-one-zenpack for most
[30-Sep-2009 15:12:43] <mrayzenoss> there are just a handful with multiple authors
[30-Sep-2009 15:12:53] <mrayzenoss> but most of the ones with multiple revisions are updated based on feedback
[30-Sep-2009 15:13:04] <mrayzenoss> which includes patches
\n[30-Sep-2009 15:13:43] <mrayzenoss> so if there's more than 1\nrevision, I'd assume that someone gave them feedback and they acted on\nit, which is good
[30-Sep-2009 15:18:37] <rmatte> When is my latest ZenPack going up?
[30-Sep-2009 15:18:39] <rmatte>
[30-Sep-2009 15:18:53] <mrayzenoss> I keep trying to work in zenpacks today
[30-Sep-2009 15:19:00] <mrayzenoss> keep running into roadblocks
[30-Sep-2009 15:19:12] <mrayzenoss> this tar thing has been raised to Jive
[30-Sep-2009 15:19:31] <mrayzenoss> the fixed forum posts is supposed to go up today
[30-Sep-2009 15:19:53] <mrayzenoss> I copy-editted the newsletter that just went out
[30-Sep-2009 15:19:55] <rmatte> tar thing or zip thing?
[30-Sep-2009 15:19:59] <mrayzenoss> tar thing
[30-Sep-2009 15:20:03] <rmatte> what tar thing?
[30-Sep-2009 15:20:13] <mrayzenoss> scroll up and look for ckrough's posts
[30-Sep-2009 15:20:28] <mrayzenoss> there appears to be an issue with uploaded tarballs
[30-Sep-2009 15:20:48] <rmatte> ah
[30-Sep-2009 15:21:10] <Brad_K> aren't they getting zipped automatically?
[30-Sep-2009 15:21:18] <mrayzenoss> not the tarballs
[30-Sep-2009 15:21:27] <rmatte> the eggs are
[30-Sep-2009 15:21:31] <rmatte> which is uber annoying
[30-Sep-2009 15:21:36] <mrayzenoss> yeah, I doc'd that as much as I can
[30-Sep-2009 15:22:02]\n<Brad_K> are you sure that they're not getting zipped\nautomatically? did you actually take a look at the two different\ntarballs?
[30-Sep-2009 15:22:02] <mrayzenoss> blame IE for that one
[30-Sep-2009 15:22:04] <rmatte> I'm sure a lot of new users are getting frustrated by that
[30-Sep-2009 15:22:17] <rmatte> oh, that was an IE issue?
[30-Sep-2009 15:22:18] <Brad_K> or maybe it's a transport issue (binary vs. ascii)?
[30-Sep-2009 15:22:22] <mrayzenoss> Brad_K: yeah, I uploaded and downloaded, different MD5s
[30-Sep-2009 15:22:31] <mrayzenoss> Brad_K: yeah, I'm sure it's something dumb
[30-Sep-2009 15:22:47] <Brad_K> right, different MD5s, but maybe the same file -- one before zipping, one after.
[30-Sep-2009 15:23:11] <Brad_K> and if it's a .tgz, that would explain why the files are probably roughly the same size.
[30-Sep-2009 15:23:15] <mrayzenoss> Brad_K: just tested, no luck
[30-Sep-2009 15:23:15] <rmatte> yeh, zipping would result in a different MD5 checksum
[30-Sep-2009 15:23:41] <mrayzenoss> I can untar both versions, other people can't
[30-Sep-2009 15:24:08] <Brad_K> have you used a binary editor to look at the contents of the two different files?
[30-Sep-2009 15:24:40] <Brad_K> it might also be a DOS cr/lf conversion issue.
\n[30-Sep-2009 15:24:45] <mrayzenoss> Brad_K: I could go down that\npath... or I could hand it off to Jive and tell them to fix it
[30-Sep-2009 15:24:53] <Brad_K> 'k
[30-Sep-2009 15:25:11] <mrayzenoss> since it's hosted we can't fix much
[30-Sep-2009 15:26:05]\n<Brad_K> sorry, i just like to know more about a problem before i\ntell someone to fix it, in part because i want to make sure that they\nfix the real underlying problem, and not just whatever they might think\nit is.
[30-Sep-2009 15:27:05] <mrayzenoss> understood
[30-Sep-2009 15:27:17] <mrayzenoss> our IT guy dug into it before raising the issue with Jive
[30-Sep-2009 15:27:43] <mrayzenoss> he was recreating the issue with the tarballs and he's on Linux
[30-Sep-2009 15:27:56] <mrayzenoss> so encoding issues seems likely
[30-Sep-2009 15:28:17] <mrayzenoss> but now I gotta go push out another beta release
[30-Sep-2009 15:30:42] <rmatte> Matt, any plans to apply the latest set of patches to 2.4.5 and do a 2.4.6 release?
[30-Sep-2009 15:30:50] <mrayzenoss> no
[30-Sep-2009 15:31:03] <rmatte> release date for King Crab been set?
[30-Sep-2009 15:31:13] <rmatte> (in stone, as it were)
[30-Sep-2009 15:31:15] <mrayzenoss> not publicly
[30-Sep-2009 15:31:22] <rmatte> that's a no
[30-Sep-2009 15:31:24] <rmatte> lol
[30-Sep-2009 15:31:28] <mrayzenoss> next week is being talked about as a RC1
[30-Sep-2009 15:31:38] <cgibbons> i know the date
[30-Sep-2009 15:31:41] <Brad_K> is that "set in stone crab"?
[30-Sep-2009 15:31:44] <rmatte> yeh, but RC is very bleeding edge
[30-Sep-2009 15:31:46] <cgibbons> we don't tell matt what the date is for obvious reasons
[30-Sep-2009 15:32:07]\n<rmatte> cgibbons: it's always comforting to know that devs know\ntheir own release dates, I'd be very very concerned if you didn't.
[30-Sep-2009 15:34:03] <cgibbons> we hit our dates pretty consistently for the major releases, +- 2 weeks tops so far
\n[30-Sep-2009 15:34:24] <cgibbons> i think this one will be later\nthan our internal date just because of a marketing push coming along\nwith it
[30-Sep-2009 15:35:22] <mrayzenoss> here's what the dev's are up to: http://dev.zenoss.org/trac/report/3
[30-Sep-2009 15:36:03] <rmatte> quite honestly, I hope you guys hit +2 weeks and get this release right lol
[30-Sep-2009 15:36:18] <rmatte> after that 2.4.3 release, I can't really take much more of it
[30-Sep-2009 15:37:41] <cgibbons> we're at the stage now where we don't put in any fixes that aren't priority-1
\n[30-Sep-2009 15:38:51] <rmatte> cgibbons: well, I understand that\nthat's the approach to take in the interests of time, but the King Crab\nrelease is sort of a huge deal, it would be nice if it doesn't get\nreleased while still riddled with obvious bugs
\n[30-Sep-2009 15:39:17] <Brad_K> yes, please. make this one\nsolid. it's been very disheartening to find so many examples of people\nhaving problems with restores not working, and with upgrades not\nworking. right now, i'm definitely feeling like we're damned if we do\nupgrad, and we're damned if we don't.
[30-Sep-2009\n15:39:30] <cgibbons> yep, but there's always a line in the sand\nsomewhere. that report matt gave you is the current work list so you\ncan get an idea
[30-Sep-2009 15:40:29] <rmatte>\nyeh, I've seen the work lists and I understand that there's a lot of\nstuff to look at, but quite honestly, if it gets released without the\nissue that I reopened ticket 4997 for fixed, it's going to be hugely\nnoticeable and people will be screaming
[30-Sep-2009 15:40:54] <rmatte> It's not a show stopping issue, thus not pri 1, but it is a very noticeable bug
\n[30-Sep-2009 15:41:58] <Brad_K> it's definitely going to hurt us\ntrying to make the case to management that we chose the right\nEnterprise Monitoring software to pay thousands and thousands of\ndollars for.
[30-Sep-2009 15:42:37] <rmatte> Brad_K: yeh, always fun
\n[30-Sep-2009 15:42:58] <Brad_K> no matter how bad the other\nalternatives would have been, we didn't go down those roads, so we have\nto live by the reality of Zenoss compared to the marketing of the\nothers.
[30-Sep-2009 15:44:34] <rmatte> I'm just\nsaying that I find that the Zenoss releases are very rushed. I can't\nrecall an upgrade that I've done of Zenoss where I can truly say that\neverything was working flawlessly up until now, and even now, I have\n2.4.5 installed with patches on top of that.
[30-Sep-2009 15:44:35] <mrayzenoss> Enterprise will lag behind Core by 2 weeks for the official release
\n[30-Sep-2009 15:44:53] <chudler> someone asked me for a way to\nmake zenoss combine alerts/events when they cross some count. I wonder\nif anyone here has tread this before?
[30-Sep-2009 15:46:02]\n<Brad_K> two weeks means nothing to us. we're in the middle of\nhaving a massive internal debate as to whether or not we try to take\n2.4.2 (or 2.4.5) and maybe meet our original schedule of turning off\nthe old monitoring system by Thanksgiving, including our developing our\nown internal training for the operators.
[30-Sep-2009\n15:46:21] <Brad_K> but then we'd have to re-teach them all over\nagain once we try to upgrade to 2.5 sometime next year.
[30-Sep-2009 15:46:44] <rmatte> Brad_K: what version are you currently using?
[30-Sep-2009 15:46:47] <rmatte> 2.4.5?
[30-Sep-2009 15:47:04]\n<Brad_K> or, do we delay rolling this out until sometime next\nyear when we've had time to integrate 2.5 into what we're trying to do\nnow, and then we don't have to retrain them all over again.
\n[30-Sep-2009 15:47:30] <Brad_K> right now, we've got 2.4.2. and\ni'm not looking forward to what's going to happen when we attempt to\nupgrade to 2.4.5.
[30-Sep-2009 15:47:35] <mrayzenoss> community/beta Beta 4 is out
[30-Sep-2009 15:47:42] <mrayzenoss> writing up the blog post
[30-Sep-2009 15:47:48] <mrayzenoss> upgrades can be tested for stack installers
[30-Sep-2009 15:47:56] <mrayzenoss> LinuxMonitor is updated
[30-Sep-2009 15:47:57] <rmatte> Brad_K: I upgraded from 2.4.2 to 2.4.5 and it was fine
\n[30-Sep-2009 15:48:19] <rmatte> well, I upgraded from 2.4.2 to\n2.4.3 and it went fine, then there were 2 huge bugs, so I patched my\nsystems up to 2.4.5
[30-Sep-2009 15:48:29] <Brad_K> that is very encouraging. i just wish i knew where the upgrade process broke for others.
[30-Sep-2009 15:48:32] <rmatte> 2.4.4 fixed 1 bug, 2.4.5 fixed 1 bug
[30-Sep-2009 15:48:48] <rmatte> Well, what type of install did you do?
\n[30-Sep-2009 15:49:11] <Brad_K> dunno. i wasn't working on the\nproject back in July when they did the most recent re-install.
[30-Sep-2009 15:49:18] <rmatte> and is Zenoss running on a physical box or a VM?
[30-Sep-2009 15:49:43] <rmatte> I see, what version of Linux is it running on?
[30-Sep-2009 15:49:56] <rmatte> and what is your $ZENHOME directory?
\n[30-Sep-2009 15:49:59] <Brad_K> what we're working on right now\nis running on physical boxes. we've got a VM that we're planning to\nuse for doing development testing with in the future.
[30-Sep-2009 15:50:23] <Brad_K> RHEL 5.something. i don't know if we're up on 5.4 yet -- i doubt it, but you never know.
[30-Sep-2009 15:50:44] <Brad_K> $ZENHOME = /opt/zenoss
[30-Sep-2009 15:50:53] <rmatte> RPM install then
[30-Sep-2009 15:50:59] <Brad_K> should be.
[30-Sep-2009 15:51:07] <Brad_K> in theory.
[30-Sep-2009 15:51:09] <rmatte> I don't use the RPM install, so I can't guarantee anything in regards to how well an upgrade would go
[30-Sep-2009 15:51:15] <rmatte> I use stack installer since it's much easier
[30-Sep-2009 15:51:48] <rmatte> If it's in /opt/zenoss then guaranteed it's an RPM install
[30-Sep-2009 15:51:53] <rmatte> stack install is /usr/local/zenoss/zenoss
[30-Sep-2009 15:52:03] <rmatte> source is /usr/local/zenoss
[30-Sep-2009 15:53:00]\n<Brad_K> we've also had some local hacks made by consulting\nservices. we're hoping that they survive the upgrade process.
[30-Sep-2009 15:53:57] <Brad_K> plus there's all the stuff we've added after the install.
[30-Sep-2009 15:54:55] <tim> Hi room
[30-Sep-2009 15:55:47] <tim> Is this the correct arena to ask for support/guidance
[30-Sep-2009 15:56:46] <cgibbons> yup
[30-Sep-2009 15:57:22] <rmatte> This is the best arena for it in fact
[30-Sep-2009 15:57:36] <rmatte>
[30-Sep-2009 15:57:40] <tim> Ah cool thanks guys
\n[30-Sep-2009 15:58:32] <tim> I have installed zenoss core, and\ndone a discovery run. But it has not found all the devices. Where do I\nstart?
[30-Sep-2009 15:58:58] <rmatte> well, are they all in whatever range you specified?
[30-Sep-2009 15:59:40] <rmatte> and are they pingable from the zenoss box?
\n[30-Sep-2009 15:59:57] <tim> Yes, I put all of the network ranges\nin to the disco run but it only returned 56 out of 400 pc (for example)
[30-Sep-2009 16:00:17]\n<feutete_> Is it possible to query snmp on multiple ports on a\nsingle device? We have what amounts to an snmp daemon that listens on\n10161 to provide information specific to a webapp
[30-Sep-2009 16:00:20] <cgibbons> and, it depends on what kind of devices (i.e. all snmp?)
[30-Sep-2009 16:00:27] <tim> Yes they are pingable, not all are running snmp or WMi tho
[30-Sep-2009 16:00:37] <feutete_> but we also want to gather standard snmp data on port 161
[30-Sep-2009 16:00:50] <rmatte> cgibbons: that shouldn't matter, discovery is done partially via plain ping
[30-Sep-2009 16:01:38] <cgibbons> indeed
[30-Sep-2009 16:01:43] <tim> Ah I think I will check that port on the firewalls (there are 6)
[30-Sep-2009 16:01:43] <rmatte> you're looking to monitor workstations?
[30-Sep-2009 16:02:07] <rmatte> If I were you I'd set any firewalls to allow the Zenoss box full access
[30-Sep-2009 16:02:16] <rmatte> otherwise you're going to run in to all kinds of fun little issues
[30-Sep-2009 16:02:29] <rmatte> afk for a sec
[30-Sep-2009 16:02:33] <tim> Not as such but as a point of excorcise i proves that it is functioning
[30-Sep-2009 16:03:15] <tim> Thanks guys I will sort the firewalls first then try again
[30-Sep-2009 16:04:01]\n<rmatte> also, if you're auto-discovering workstations (which it\nsounds like you may be), they all have to be powered on during\ndiscovery (obviously)
[30-Sep-2009 16:04:14] <cgibbons> ouf that had to be hard to type on
[30-Sep-2009 16:05:06] <Brad_K> i've done ssh on my iphone. it's actually not that hard.
[30-Sep-2009 16:05:17] <rmatte> I use colloquy on my iphone all the time, not that bad
[30-Sep-2009 16:05:26] <rmatte> and I do ssh all the time too
[30-Sep-2009 16:05:27] <rmatte>
[30-Sep-2009 16:05:30] <cgibbons> i IM chat a lot on mine but it's still hard to do tech talk on it i find
[30-Sep-2009 16:05:56]\n<rmatte> yeh, tech talk isn't the simplest, especially with the\nlovely auto-correct feature which I so lovingly turned off recently
[30-Sep-2009 16:06:48]\n<Brad_K> tim must have done the same, looking at what he typed. \nthe auto-correct function would not have let some of those through.
[30-Sep-2009 16:10:06] <rmatte> yeh
[30-Sep-2009 16:10:25] <rmatte> actually, the trick I use is to put 1 letter is front of the cursor as I type
[30-Sep-2009 16:10:34] <rmatte> autocorrect ignores what you're typing if you do that
[30-Sep-2009 16:10:41] <rmatte> in front*
\n[30-Sep-2009 16:20:41] <ke4qqq> apparently I am doing something\nreally stupid with processes - I tried to step up the complexity a bit\nand instead of monitoring simple regexes for things like httpd I am\nlooking for this: http://fpaste.org/YuIL/
[30-Sep-2009\n16:20:49] <cgibbons> i'm totally addicted to my iphone. i can't\nimagine life without it anymore. sorta like life before the internet,\ntoo. scary! dark!
[30-Sep-2009 16:21:06] <rmatte> cgibbons: same
[30-Sep-2009 16:21:20] <rmatte> cgibbons: you don't think you need one until you have one
\n[30-Sep-2009 16:21:42] <cgibbons> i was actually annoyed to find\nthe NPS lodge @ big bend NP had WiFi last year (no cellular coverage)\nso I was still connected to the world even out there
[30-Sep-2009 16:22:01] <rmatte> cgibbons: take a look at what ke4qqq is trying to do, he was asking me the other day and I had no idea
[30-Sep-2009 16:22:09] <rmatte> I don't even think what he's trying to do is possible
[30-Sep-2009 16:22:16] <cgibbons> k, i shall in a moment, in a video conference ::)
[30-Sep-2009 16:22:17] <ke4qqq> rmatte: why not?
[30-Sep-2009 16:22:20] <rmatte> k
[30-Sep-2009 16:22:31] * ke4qqq thought I had only been working on this for a single day
\n[30-Sep-2009 16:22:44] <rmatte> ke4qqq: because, you're trying to\ndo the regex against the options, and I don't think that the regex even\nlooks at the options
[30-Sep-2009 16:22:51] <rmatte> it only looks at the actual process name
[30-Sep-2009 16:23:02] <rmatte> in snmp the process name and the options are displayed with different OIDs
[30-Sep-2009 16:23:09] <rmatte> they aren't stuck together
[30-Sep-2009 16:23:11] <ke4qqq> yes they are
\n[30-Sep-2009 16:23:30] <ke4qqq> but then it has the entire\n'ignore arguments' boolean - why have that if you can't use them
[30-Sep-2009 16:23:34] <rmatte> I specifically remember going through the snmp tree and seeing them displayed separately
[30-Sep-2009 16:23:50] <ke4qqq> yes they are in a separate oid iirc
[30-Sep-2009 16:23:59] <ke4qqq> I assumed concatenation
[30-Sep-2009 16:25:54] <ke4qqq> Jane Curry's paper on process monitoring
[30-Sep-2009 16:26:18] <ke4qqq> shows an example with arguments on p22
[30-Sep-2009 16:26:22] <ke4qqq> hence my attempt
[30-Sep-2009 16:26:35] <rmatte> oh, well if it's in there then Jane must have gotten it to work
[30-Sep-2009 16:26:41] <rmatte> so I guess it is possible
[30-Sep-2009 16:27:00] <ke4qqq> well it's enough for today, I'll deal with it tomorrow
[30-Sep-2009 16:35:25] <rmatte> bah, as if it's only wednesday
[30-Sep-2009 16:36:27] <mrayzenoss> I'm kinda slow
[30-Sep-2009 16:38:19] <rmatte> hehe
[30-Sep-2009 16:38:20] <rhettardo> waat
\n[30-Sep-2009 16:55:39] <mrayzenoss> here's blog post for the new\nbeta:\nblogs/zenossblog/2009/09/30/zenoss-25-beta-4-is-now-available
[30-Sep-2009 16:56:22] <rmatte> right on, I shall be installing tomorrow
[30-Sep-2009 16:56:56] <mrayzenoss> and upgrading
[30-Sep-2009 16:57:02] <mrayzenoss> stack installer upgrades can be tested
[30-Sep-2009 16:57:11] <rmatte> cool
[30-Sep-2009 16:57:24] <rmatte> I'll be sure to test it out and let you know how it goes
[30-Sep-2009 17:02:02] <mrayzenoss> if possible, try to go from you 2.4.5 install, not your previous beta
[30-Sep-2009 17:02:15] <mrayzenoss> since we're looking for testing of 2.4.x->2.5
[30-Sep-2009 18:18:00]\n<exarkun> What does the "Linux Monitor" pack do? I thought it\nwould give me a new template to bind, but it doesn't seem to have.
[30-Sep-2009 18:19:10] <exarkun> Maybe my "Device Class Path" is wrong.
[30-Sep-2009 18:21:04] <exarkun> Ah. I guess so. I changed it to "Server/SSH/Linux" and some other stuff happened.
[30-Sep-2009 18:21:09] * exarkun looks around to see if ssh keys are supported.
\n[30-Sep-2009 18:21:14] <tung> Do you know which password this\nerror refers to in this error message regarding a Windows Server:
\n[30-Sep-2009 18:21:26] <tung> Could not read the Windows event\nlog (NT_STATUS_IO_TIMEOUT). Check your username/password settings
\n[30-Sep-2009 18:32:24] <cgibbons> tung: exactly that, a timeout\nto the server. that usually means the network to the server is down or\n(more likely) a firewall is preventing communication with it.
[30-Sep-2009 18:32:58] <tung> But I can snmpwalk the server just fine
[30-Sep-2009 18:33:09] <tung> There's no firewall in between
[30-Sep-2009 18:33:36] Traveler is now known as Guest99219
\n[30-Sep-2009 18:33:59] <cgibbons> snmp doesn't come into play\nhere, reading the Windows Event Log occurs via a WMI query over DCOM,\nso that's port 135 + some other stuff
[30-Sep-2009\n18:34:16] <Guest99219> hello anyone here able to shed some light\non this: ERROR RenderServer Failed to generate a graph
\n[30-Sep-2009 18:34:26] <cgibbons> if the Windows device itself\nhas the software firewall enabled or not configured to allow DCOM then\nit'll also block it
[30-Sep-2009 18:34:54] <tung> DCOM? Really?
[30-Sep-2009 18:34:58] <cgibbons> yep
[30-Sep-2009 18:35:00] <tung> I didn't know that
[30-Sep-2009 18:35:34] <tung> OK let me get the Windows guy to check that
[30-Sep-2009 18:35:41] <tung> Thanks
[30-Sep-2009 18:35:53] <cgibbons> if you want to try something real quick from zenoss I can give him a few more hints
[30-Sep-2009 18:36:04] <cgibbons> (from the zenoss server that is, a command-line test)
[30-Sep-2009 18:36:10] <tung> BTW, is that UDP or TCP (for DCOM)
[30-Sep-2009 18:36:21] <tung> sure what test can I try?
\n[30-Sep-2009 18:36:46] <cgibbons> it's TCP. basically the\nlayering is Microsoft RPC over TCP, with DCOM layered on top of that,\nand WMI layered on top of that yet still
[30-Sep-2009\n18:37:11] <cgibbons> from the command-line: wmic -d99 -U\n'yourdomain\\youruser' //yourhostname "select name from\nwin32_computersystem"
[30-Sep-2009 18:37:28]\n<Guest99219> @tung, dcom and run over udp as well reference \n-http://en.wikipedia.org/wiki/Port_numbers
\n[30-Sep-2009 18:37:28] <cgibbons> that should produce a lot of\nnotice; if you put it in pastebin i can tell you how far it's getting
[30-Sep-2009 18:39:20] <tung> OK I don't have a login to that box :-(
[30-Sep-2009 18:39:27] <cgibbons> okie
[30-Sep-2009 18:39:45] <tung> Let me try to get him to open 135 TCP/UDP first and see
[30-Sep-2009 18:40:03] <tung> Thanks for the answer
[30-Sep-2009 18:40:03] <tung> as always :-D
[30-Sep-2009 18:40:13] <tung> This is a great forum
\n[30-Sep-2009 18:40:16] <Guest99219> cgibbons- is there a way i\ncan grab a more verbose output for "ERROR RenderServer Failed to\ngenerate a graph" such as teh wmic comand you presented tung?
[30-Sep-2009 18:40:35] <cgibbons> where are you seeing that error?
[30-Sep-2009 18:41:13] <Guest99219> in the zopectl log
\n[30-Sep-2009 18:41:41] <cgibbons> tung: you may find\nhttp://msdn.microsoft.com/en-us/library/aa389286(VS.85).aspx and\nhttp://msdn.microsoft.com/en-us/library/ms809327.aspx useful references\nfor him if it comes to that
[30-Sep-2009 18:42:09]\n<Guest99219> previous this was working - i actually have\nhistorical data going back a few weeks and can confirm the wmi query is\ncorrect and returns valid results
[30-Sep-2009\n18:42:11] <cgibbons> you can edit $ZENHOME/etc/zope.conf and look\nfor the debug-mode directive and change it to be enabled and then\nrestart zope (zopectl restart) to see more debug
[30-Sep-2009 18:43:10] <cgibbons> also look at your zenrender.log and see if it said anything useful
[30-Sep-2009 18:45:00] <Guest99219> alright - restarting zope in debug mode
\n[30-Sep-2009 18:45:40] <Guest99219> i get serveral of these but\nit seems i can safely ignore - INFO ZODB.Conflict database conflict\nerror
[30-Sep-2009 18:45:52] <cgibbons> yep
[30-Sep-2009 18:46:07] <cgibbons> iirc zenrender works by using /tmp/renderserver for a work area so double check it's got free space etc
[30-Sep-2009 18:49:47] <Guest99219> free space checks out ok, along with permissions on that directory
\n[30-Sep-2009 18:50:42] <Guest99219> can i send you a pastebin\noutput - doesn't seem like the debug mode helped with teh verbosity
[30-Sep-2009 18:52:03] <Guest99219> http://pastebin.com/d75762205
[30-Sep-2009 18:57:37] Traveler is now known as Guest49354
[30-Sep-2009 18:57:55] <Guest49354> got d/c'd.
[30-Sep-2009 18:58:06] <Guest49354> have a chance to check the pastebin?
[30-Sep-2009 19:00:34] <Guest49354> cgibbons: i do not see a zenrender log under /opt/zenoss/log - is it located somewhere else?
[30-Sep-2009 19:13:14] <Guest49354> i like the new website
[30-Sep-2009 19:25:43] <Guest49354> too bad theres no redirects for pages that were already index by search engines
\n[30-Sep-2009 19:34:15] <Guest49354> getting back to my render\ngraph issue - it does appear related to INFO ZODB.Conflict database\nconflict error
[30-Sep-2009 19:34:34] <Guest49354> i thought i could safely ignore, but that doesn't appear to be the case
[30-Sep-2009 19:36:30] Traveler is now known as Guest78799
[30-Sep-2009 20:56:52] <cryptographrix> What MIBs would I need for a basic ZenOSS install?
[30-Sep-2009 20:57:19] <cryptographrix> i.e. - is there a standard set of MIBs to install?
[30-Sep-2009 21:46:45] <cryptographrix> "Can't stop reactor that isn't running." when I model
[30-Sep-2009 21:46:47] <cryptographrix> ideas?
\n[30-Sep-2009 21:50:39] * exarkun knows what that means and why it\nhappens but not how to fix it except by editing Zenoss source.
[30-Sep-2009 21:52:00] <cryptographrix> lol ah,kk
[30-Sep-2009 21:52:12] <cryptographrix> is it a bug?
[30-Sep-2009 21:52:35] <cryptographrix> trying to figure out how ZenOss works to use it for monitoring a contract I have
[30-Sep-2009 22:01:26] <exarkun> I think it's a bug. I'm pretty new to zenoss too though.
[30-Sep-2009 22:01:33] <cryptographrix> coolcool
[30-Sep-2009 22:01:40] <cryptographrix> yah I'ma be on for the devs tomorrow lol
[01-Oct-2009 00:00:46] [disconnected at Thu Oct 1 00:00:46 2009]
