[01-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[01-Jun-2009 07:37:20] <venturaville> odd ... irc used to be blocked from work and now it works ...sweet :-)
[01-Jun-2009 07:46:42] <Appiah> This is the second server I upgrade to 2.4.1 , the progressbar goes fast from 0 to 100%
[01-Jun-2009 07:46:49] <Appiah> then it just sits there for like an hour
[01-Jun-2009 09:00:44] <rocket> morning matt
[01-Jun-2009 09:07:38] <jb> yawn
[01-Jun-2009 09:09:34] <mrayzenoss> good morning
[01-Jun-2009 09:10:12] <rocket> mrayzenoss: so I started working on an ssh based zenpack .. but the unit test documentation leaves much to be desired
[01-Jun-2009 09:10:29] <rocket> mrayzenoss: and I thought there would be an example in the linux ssh zenpack but no such luch
[01-Jun-2009 09:10:32] <rocket> luck
[01-Jun-2009 09:10:36] <mrayzenoss> oh yeah? Feel free to make notes where it sucks and I'll tell Brian to fix it
[01-Jun-2009 09:10:46] <mrayzenoss> there are no tests in the Linux Monitor?
[01-Jun-2009 09:11:07] <rocket> mrayzenoss: not that I know of .. unless I accidentally have an old version
[01-Jun-2009 09:11:16] <mrayzenoss> doubt it, let me poke around
[01-Jun-2009 09:11:39] <rocket> mrayzenoss: I was looking for it in there because I figured it would be the most reasonable place to look for an example to fill in the gaps I was having
[01-Jun-2009 09:12:09] <rocket> mrayzenoss: I mean the development of the zenpack without the unittests is going fine etc .. just was trying out that unit test framework ...
[01-Jun-2009 09:18:06] <mrayzenoss> yeah, he stuck all the tests in the EnterpriseLinux ZenPack, I'm going to see about moving some into Linux Monitor
[01-Jun-2009 09:24:23] <jb> i was goign to say
[01-Jun-2009 09:24:31] <jb> the tests exist in the enterprise pack..
[01-Jun-2009 09:25:13] <rocket> ?
[01-Jun-2009 09:25:13] <rocket> oops
[01-Jun-2009 09:57:28] <jb> rocket, the ssh zenpack.. is it for AIX or something else?
[01-Jun-2009 09:57:38] <rocket> jb: gentoo first ..
[01-Jun-2009 09:57:54] <rocket> jb: getting my feet wet with an easy one first etc ..
[01-Jun-2009 10:00:28] <rocket> jb: but from what I can tell ssh zenpacks should be relatively easy to convert from snmp ones .. I just have to do more regex's etc
[01-Jun-2009 10:00:39] <jb> yeah
[01-Jun-2009 10:00:42] <jb> figure out how to parse everything
[01-Jun-2009 10:01:06] <rocket> jb: I am not having trouble with the zenpack itself .. I am still trying to figure out how these unit tests etc work
[01-Jun-2009 10:24:26] <coofamani> anyone have an idea of the context of sysUpTime in a report? tried sysUpTime, here.hw.sysUpTime, and here.os.sysUpTime.
[01-Jun-2009 10:25:07] <mrayzenoss> Appendix D of the Admin Guide, ${device/uptimeStr)
[01-Jun-2009 10:25:29] <mrayzenoss> someone just asked me about that
[01-Jun-2009 10:27:30] <coofamani> sorry, thats the string version. Im trying to get the snmp returned value in 1/100 seconds, in order to sort correctly
[01-Jun-2009 10:27:49] <coofamani> the object is sysUpTime vs uptimeStr
[01-Jun-2009 10:28:02] <coofamani> but Im having trouble referencing that in deviceQuery
[01-Jun-2009 10:29:31] <venturaville> >>> dev = devices.findDevice('mon8zen2')
[01-Jun-2009 10:29:31] <venturaville> >>> dev.sysUpTime()
[01-Jun-2009 10:29:31] <venturaville> 286156976.11000001
[01-Jun-2009 10:29:44] <venturaville> It is a call not a variable.........
[01-Jun-2009 10:29:54] <venturaville> so it would be: here.sysUpTime()
[01-Jun-2009 10:30:21] <venturaville> print '\n'.join(dir(dev))
[01-Jun-2009 10:30:23] <venturaville> is your friend............
[01-Jun-2009 10:30:37] <coofamani> nice. thanks
[01-Jun-2009 10:30:39] <coofamani> "here.sysUpTime() > 86400000"
[01-Jun-2009 10:30:53] <venturaville> do you know how to use zendmd?
[01-Jun-2009 10:30:54] <coofamani> that does what I want. I wasnt treating it as a call, had that line with the (). thanks
[01-Jun-2009 10:31:02] <coofamani> zendmd = sorta
[01-Jun-2009 10:31:07] <coofamani> need to spend more time in there
[01-Jun-2009 10:31:13] <venturaville> just start it up: and do
[01-Jun-2009 10:31:20] <venturaville> here = device.findDevice('nameofsomedevice')
[01-Jun-2009 10:31:26] <venturaville> and then you will pretty much be there......
[01-Jun-2009 10:31:48] <coofamani> then use dir to see all the available methods? nice
[01-Jun-2009 10:32:13] <venturaville> yeah ... thank python for that nicety
[01-Jun-2009 10:32:34] <mrayzenoss> feel free to send in reports as ZenPacks
[01-Jun-2009 10:35:21] <coofamani> toying with that now. thanks venturaville. this will make life better.
[01-Jun-2009 10:36:05] <venturaville> >>> callable(dev.sysUpTime)
[01-Jun-2009 10:36:05] <venturaville> True
[01-Jun-2009 10:36:05] <venturaville> >>> callable(dev.sysUpTime())
[01-Jun-2009 10:36:05] <venturaville> False
[01-Jun-2009 10:36:12] <venturaville> another nice thing to use ..........
[01-Jun-2009 11:44:13] <mrayzenoss> For those of you using the HP ProLiant Monitor ZenPack, it's been updated along with the Advanced Device Details
[01-Jun-2009 11:46:06] <mrayzenoss> and now I can finally start tackling rocket's backlog
[01-Jun-2009 11:47:07] <mrayzenoss> 7 currently in the queue
[01-Jun-2009 11:47:15] <jb>
[01-Jun-2009 11:47:31] <jb> have you heard how igor's dell pack is coming
[01-Jun-2009 11:47:31] <jb> ?
[01-Jun-2009 11:48:55] <mrayzenoss> nothing yet, he just asked me for some OIDs last week and told me he was working on one like his HP one
[01-Jun-2009 11:49:02] <jb> cool..
[01-Jun-2009 11:49:40] <mrayzenoss> 8 entries in the ZenPack contest so far
[01-Jun-2009 11:49:47] <jb> are they posted?
[01-Jun-2009 11:49:51] <mrayzenoss> not counting my backlog of 7
[01-Jun-2009 11:50:10] <mrayzenoss> I was planning on doing a post this week about progress so far
[01-Jun-2009 11:50:35] <mrayzenoss> but the new ones are the 2 Nortel ones, the updates are Perfmon, SQL Server and the 3 from Egor
[01-Jun-2009 11:50:46] <mrayzenoss> and a Barracuda update
[01-Jun-2009 11:50:51] <jb> perfmon?
[01-Jun-2009 11:51:11] <mrayzenoss> He took the Perfmon one and converted it to an egg so the SQL Server could depend on it
[01-Jun-2009 11:51:16] <jb> ahh
[01-Jun-2009 11:51:24] <mrayzenoss> if you're using Enterprise you have access to zenwinperf
[01-Jun-2009 11:51:26] <jb> it basically just allows you to pull perfmon datasources like enterprise?
[01-Jun-2009 11:51:30] <mrayzenoss> yeah
[01-Jun-2009 11:51:31] <jb> yeah
[01-Jun-2009 11:51:32] <jb> k
[01-Jun-2009 11:51:37] <mrayzenoss> well, not exactly, but similar
[01-Jun-2009 11:51:51] <mrayzenoss> it shells out to a perl call
[01-Jun-2009 11:51:57] <jb> ah ok
[01-Jun-2009 11:53:33] <mrayzenoss> rocket gave me his AIX SNMP one for release
[01-Jun-2009 11:53:48] <mrayzenoss> I think you were looking for that right?
[01-Jun-2009 11:53:49] <jb> nice
[01-Jun-2009 11:53:52] <jb> yeah
[01-Jun-2009 12:02:51] <rocket> mrayzenoss: I will probably do an update for the aix one eventually as I get time .. but I figure I might as well get it out there and solicit feedback
[01-Jun-2009 12:03:49] <rocket> mrayzenoss: and I had a thought about these tests that are missing
[01-Jun-2009 12:04:15] <rocket> mrayzenoss: couldnt the zenpacks create a template one but commented out as an example that gets copied into all new zenpacks ..
[01-Jun-2009 12:04:24] <mrayzenoss> sure
[01-Jun-2009 12:04:33] <mrayzenoss> sounds like a good enhancement if you want to open a ticket
[01-Jun-2009 12:04:57] <rocket> ok .. I just figure if you are pushing for the tests .. might as well make it easier to make them .. etc
[01-Jun-2009 12:05:15] <mrayzenoss> yeah, definitely
[01-Jun-2009 12:05:33] <mrayzenoss> I've got a todo to look into the templates at some point, just haven't found the cycles
[01-Jun-2009 12:05:42] <mrayzenoss> people keep contributing too many zenpacks
[01-Jun-2009 12:06:29] <rocket>
[01-Jun-2009 12:06:35] <rocket> want me to slow down?
[01-Jun-2009 12:06:43] <mrayzenoss> nope
[01-Jun-2009 12:06:55] <mrayzenoss> it's a good problem to have
[01-Jun-2009 12:07:25] <rocket> heh
[01-Jun-2009 12:08:27] <rocket> mrayzenoss: when do you think I might be able to get access to those sample tests?
[01-Jun-2009 12:08:37] <rocket> just so I can plan my schedule
[01-Jun-2009 12:10:24] <mrayzenoss> I'll let you know when I hear from Brian
[01-Jun-2009 12:10:31] <rocket> thanks
[01-Jun-2009 12:11:19] <mrayzenoss> possibly later this week
[01-Jun-2009 12:11:37] <mrayzenoss> I'm gonna go look at the code, maybe I'll do it
[01-Jun-2009 12:12:15] <rocket> cool .. that would be great .. especially since summer of code could really use it
[01-Jun-2009 12:29:00] <venturaville> anyone know what to do about these: Only ZenPacks that have been installed in development mode can be exported.
[01-Jun-2009 12:29:45] <venturaville> its alright
[01-Jun-2009 12:30:35] <venturaville> its not right about it not being installed
[01-Jun-2009 12:47:50] <mrayzenoss> are you trying to get a ZenPack into development mode? trac-zenpacks/wiki/DevelopmentMode
[01-Jun-2009 12:47:51] <mrayzenoss> Or some other issue?
[01-Jun-2009 12:51:18] <venturaville> I think it is something else
[01-Jun-2009 12:51:36] <venturaville> I get this when I load up zenpackmgr even though the pack is not installed anymore:
[01-Jun-2009 12:51:38] <venturaville> Type: DistributionNotFound
[01-Jun-2009 12:51:38] <venturaville> Value: ZenPacks.turner.HPBladeChassis
[01-Jun-2009 12:52:01] <venturaville> I've stopped and restarted, and done a grep across the zenoss tree, but cannot find where it is referenced
[01-Jun-2009 12:52:49] <venturaville> cd /opt/zenoss ; find . | grep -i bladechassis
[01-Jun-2009 12:52:51] <venturaville> returns nothing
[01-Jun-2009 12:55:03] <venturaville> stored in the database somehow I'm guessing......
[01-Jun-2009 12:59:21] <venturaville> >>> zp = dmd.ZenPackManager.packs._getOb('ZenPacks.turner.HPBladeChassis')
[01-Jun-2009 12:59:22] <venturaville> >>> zp
[01-Jun-2009 12:59:22] <venturaville> <ZenPack at /zport/dmd/ZenPackManager/packs/ZenPacks.turner.HPBladeChassis>
[01-Jun-2009 12:59:40] <venturaville> >>> dmd.ZenPackManager.packs._delOb('ZenPacks.turner.HPBladeChassis')
[01-Jun-2009 12:59:40] <venturaville> Traceback (most recent call last):
[01-Jun-2009 12:59:40] <venturaville> File "<stdin>", line 1, in ?
[01-Jun-2009 12:59:40] <venturaville> File "/opt/zenoss/zenoss/Products/ZenRelations/ToManyRelationshipBase.py", line 87, in _delOb
[01-Jun-2009 12:59:40] <venturaville> raise NotImplementedError
[01-Jun-2009 12:59:41] <venturaville> NotImplementedError
[01-Jun-2009 12:59:44] <venturaville> do
[01-Jun-2009 12:59:45] <venturaville> doh
[01-Jun-2009 13:05:03] <venturaville> catch 22 ...
[01-Jun-2009 13:05:12] <venturaville> how to delete an object in there?
[01-Jun-2009 13:12:11] <chudler> SNMP + zenpack dev question for gurus. I have two devices, different vendors, but supporting the same MIB tree. I have a custom modeler plugin that works for one device, but not the other.
[01-Jun-2009 13:12:48] <chudler> The reason is doesnt work for the other, is that the vendors prefix the tree with their vendor string. So, in the modeler, I have something like
[01-Jun-2009 13:13:02] <chudler> GetTableMap('subentries', '.1.3.6.1.4.1.1718.3.2.5.1', subentries)
[01-Jun-2009 13:14:16] <chudler> where 1718.3 is the vendor's prefix. Unfortunately, a different vendor will use a different prefix, in this case 2606.100.1; I don't want to make a entirely new zenpack to support a identical device, just because it has a different vendor code in the MIB.
[01-Jun-2009 13:14:52] <chudler> I thought what I might do is have alternate table maps for each known vendor, but a programatic solution would be prefered.
[01-Jun-2009 13:17:10] <venturaville> fixed it finally
[01-Jun-2009 13:17:26] <venturaville> leftovers from a previous zenpack install
[01-Jun-2009 14:09:43] <jb> matt?
[01-Jun-2009 14:16:59] <mrayzenoss> yeah?
[01-Jun-2009 14:33:34] <CapriCoRN^80> hi ! i need some information. My all hosts get IP from DHCP .. i was reading a post in which someone mentioned problem that zenoss will not detect the DHCP IP ?
[01-Jun-2009 14:41:08] <mrayzenoss> Zenoss doesn't play well with DHCP
[01-Jun-2009 14:41:20] <mrayzenoss> since it monitors based on IP addresses
[01-Jun-2009 14:41:53] <mrayzenoss> that said, you can probably configure your DHCP server to assign the same addresses to machines each time they request an address
[01-Jun-2009 14:42:11] <mrayzenoss> which allows you to keep using DHCP no problem
[01-Jun-2009 14:47:45] <mrayzenoss> CapriCoRN^80: following up from my post on LinuxQuestions I see
[01-Jun-2009 14:48:49] <mrayzenoss> CapriCoRN^80: there's a webinar tomorrow at 3pm EST for Getting Started
[01-Jun-2009 15:16:10] <CapriCoRN^80> well thats hard thing
[01-Jun-2009 15:16:28] <CapriCoRN^80> i mean if the lease time of dhcp is lease then its a problem
[01-Jun-2009 15:17:07] <mrayzenoss> do you have a lot of transient users?
[01-Jun-2009 15:17:11] <mrayzenoss> like a wireless hotspot?
[01-Jun-2009 15:17:27] <CapriCoRN^80> not wireless. i got lan users
[01-Jun-2009 15:17:55] <mrayzenoss> but they're known right? If you have their MAC addresses you can give them fixed DHCP addresses, but still use DHCP
[01-Jun-2009 15:17:55] <CapriCoRN^80> and i got dhcp server which assign ip addresses to lan users
[01-Jun-2009 15:18:46] <CapriCoRN^80> well i have to go through all the mac addresses
[01-Jun-2009 15:19:00] <CapriCoRN^80> i read that post at evening time and till that i got my server up .
[01-Jun-2009 15:19:49] <CapriCoRN^80> i am difficult situtation now
[01-Jun-2009 15:20:40] <CapriCoRN^80> in *
[01-Jun-2009 15:24:53] <CapriCoRN^80> <@mrayzenoss> since it monitors based on IP addresses .. didnt get that
[01-Jun-2009 15:25:02] <CapriCoRN^80> dhcp also assign ip addresses
[01-Jun-2009 15:25:48] <rocket> CapriCoRN^80: basically if the ip address changes for a device zenoss doesnt really like it
[01-Jun-2009 15:26:11] <rocket> CapriCoRN^80: because it keys off of the ip address
[01-Jun-2009 15:26:50] <rocket> eg if you change the ip address of a workstation eg a flipflop .. then zenoss wont know about it and the data it collects will be applied to the wrong entry in zenosses databases
[01-Jun-2009 15:27:11] <CapriCoRN^80> rocket: what if the lease time of dhcp server is 10 days ?
[01-Jun-2009 15:27:22] <CapriCoRN^80> or may be 4 days
[01-Jun-2009 15:27:23] <rocket> CapriCoRN^80: same issue exists
[01-Jun-2009 15:27:30] <CapriCoRN^80> rocket: strange
[01-Jun-2009 15:27:42] <CapriCoRN^80> 4 days is not flipflop
[01-Jun-2009 15:28:12] <rocket> CapriCoRN^80: I did not say its flapping
[01-Jun-2009 15:28:38] <rocket> CapriCoRN^80: I was just giving an example of data going to the wrong entry in the database
[01-Jun-2009 15:29:05] <CapriCoRN^80> rocket: ok
[01-Jun-2009 15:29:22] <CapriCoRN^80> rocket: i am in trouble
[01-Jun-2009 15:30:08] <rocket> CapriCoRN^80: I am assuming zenoss was really designed with a static datacenter in mind initially
[01-Jun-2009 15:30:13] <CapriCoRN^80> well dhcp method is world wide use
[01-Jun-2009 15:30:52] <CapriCoRN^80> rocket: i am not a big engineer but i guess this is big problem in zenoss
[01-Jun-2009 15:31:47] <rocket> it could be .. I havent really run into it as our datacenter is static
[01-Jun-2009 15:33:13] <CapriCoRN^80> how many users u have ?
[01-Jun-2009 15:33:18] <mrayzenoss> CapriCoRN^80: it's a known limitation, but my point was in your dhcpd.conf, you can use a host statement to get around it: some really good scenes which made other
[01-Jun-2009 15:33:32] <mrayzenoss> oops, bad copy and paste
[01-Jun-2009 15:33:33] <mrayzenoss> http://www.daemon-systems.org/man/dhcpd.conf.5.html
[01-Jun-2009 15:33:48] <mrayzenoss> CapriCoRN^80: it's a known limitation, but my point was in your dhcpd.conf, you can use a host statement to get around it: http://www.daemon-systems.org/man/dhcpd.conf.5.html
[01-Jun-2009 15:34:45] <mrayzenoss> so even after you 4 day lease is up, the server will renew and get the same IP again
[01-Jun-2009 15:34:48] <rocket> CapriCoRN^80: using zenoss .. at the moment we are still developing etc and setting up
[01-Jun-2009 15:35:09] <rocket> CapriCoRN^80: but we are not running into the issue you are because we are a static ip datacenter shop here
[01-Jun-2009 15:35:35] <mrayzenoss> My home router let's me assign static addresses in DHCP as well
[01-Jun-2009 15:36:07] <CapriCoRN^80> well routers and servers have static ip addresses
[01-Jun-2009 15:36:17] <CapriCoRN^80> i am talking about my client desktops
[01-Jun-2009 15:36:33] <rocket> CapriCoRN^80: we arent doing client desktops at the moment ..
[01-Jun-2009 15:36:50] <mrayzenoss> CapriCoRN^80: you can leave them out by turning off network modeling
[01-Jun-2009 15:37:09] <mrayzenoss> do a network discovery, delete the ones you don't care about and lock it down from there
[01-Jun-2009 15:38:15] <cgibbons> jb are you around?
[01-Jun-2009 15:39:35] <jb> yep
[01-Jun-2009 15:39:36] <CapriCoRN^80> actually we used some new versions of software and we got feed back that the performance is going down due to those softwares .. thats why we are moving to this performance thing
[01-Jun-2009 15:39:55] <jb> cgibbons: whats up?
[01-Jun-2009 15:40:03] <CapriCoRN^80> and yes my requirements are cpu usage , memory usage , process running , and performance
[01-Jun-2009 15:41:01] <mrayzenoss> CapriCoRN^80: so Zenoss will be very lightweight
[01-Jun-2009 15:41:18] <mrayzenoss> SNMP and WMI are not resource intensive
[01-Jun-2009 15:41:36] <CapriCoRN^80> i am not talking about zenoss performance
[01-Jun-2009 15:41:53] <mrayzenoss> right, I mean the effect on the monitored servers
[01-Jun-2009 15:41:56] <mrayzenoss> and devices
[01-Jun-2009 15:42:06] <CapriCoRN^80> and i am talking performance in terms of applications running on clients systems
[01-Jun-2009 15:43:58] <mrayzenoss> right.
[01-Jun-2009 15:44:23] <mrayzenoss> There's a good paper on the different techniques for process monitoring with Zenoss: http://www.skills-1st.co.uk/papers/jcurry.html
[01-Jun-2009 15:44:25] <CapriCoRN^80> i want to check cpu usage , memory usage , process running and performance of client system
[01-Jun-2009 15:44:37] <mrayzenoss> the "Methods of monitoring processes with Zenoss" one
[01-Jun-2009 15:45:09] <rocket> CapriCoRN^80: there is a lot that can go under performance ...
[01-Jun-2009 15:45:32] <mrayzenoss> well you can get overall cpu and memory and per-process cpu and memory
[01-Jun-2009 15:45:59] <CapriCoRN^80> but dhcp issue ?
[01-Jun-2009 15:50:58] <CapriCoRN^80> mrayzenoss:
[01-Jun-2009 15:57:48] <CapriCoRN^80> ?
[01-Jun-2009 15:58:43] <jb> whats the question?
[01-Jun-2009 15:58:51] <jb> you can't really monitor dhcp'd devices..
[01-Jun-2009 15:59:30] <jb> mrayzenoss was saying you could create DHCP reservations to make the device grab the same IP address everytime it requested a lease..
[01-Jun-2009 15:59:40] <jb> depending on your environment, that may or may not be plausible
[01-Jun-2009 15:59:50] <jb> if its a large amount of desktops, it probably is not..
[01-Jun-2009 16:03:31] <rocket> jb: seems he is gone ...
[01-Jun-2009 16:21:53] <jb> yep
[01-Jun-2009 16:22:03] <jb> we tried
[02-Jun-2009 00:00:42] [disconnected at Tue Jun 2 00:00:42 2009]
[02-Jun-2009 00:00:42] [connected at Tue Jun 2 00:00:42 2009]
[02-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[02-Jun-2009 09:18:56] <Dieterbe> Hi, this is not strictly a zenoss issue but maybe you could help me anyway. In device/perf for a Server/Linux on 1 hard disk I see on the "I/O" graph peaks of about 4.0kB (read or write) and about 500bytes "usually"
[02-Jun-2009 09:19:14] <Dieterbe> I was wondering if anyone knows which values i should see here, some comparison data/graphs etc
[02-Jun-2009 09:19:36] <Dieterbe> even for the random io on this box (it runs mysql) this seems quite slow
[02-Jun-2009 09:33:05] <Dieterbe> aha, according to http://wiki.postgresql.org/wiki/HP_ProLiant_DL380_G5_Tuning_Guide#random_read i should get at least 2MB/s
[02-Jun-2009 10:03:52] <afink> Hello everyone, I am having trouble getting performance information on centos 5.2 to show up in zenoss. I have been researching on the internet for some time now and haven't been able to come up with a good solution
[02-Jun-2009 10:08:07] <Dieterbe> afink: can you snmpwalk your target host?
[02-Jun-2009 10:11:10] <chudler> anyone use ZenUtils Driver classes? I have trouble understanding just how it works :-/
[02-Jun-2009 10:12:17] <rocket> chudler: what are you trying to do?
[02-Jun-2009 10:13:43] <afink> Dieterbe: Yes I can snmpwalk it, http://pastebin.com/m5bf38022
[02-Jun-2009 10:13:44] <chudler> inside a custom model, I have an AgentProxy for fetching a few values. I am using the driver for that.
[02-Jun-2009 10:15:03] <Dieterbe> afink: hmm.. check your event log (both in the web interface and the one in /usr/local/zenoss/...log/event.log)
[02-Jun-2009 10:15:11] <Dieterbe> maybe you'll see errors there
[02-Jun-2009 10:15:43] <chudler> rocket: I do yield proxy.get([self.tempOid]), and it immediately returns a defered object to the caller!
[02-Jun-2009 10:16:47] <rocket> ill look and see
[02-Jun-2009 10:16:55] <afink> Dieterbe: This is what shows up on the performance graphs: http://picpaste.com/zenoss.jpeg
[02-Jun-2009 10:17:15] <afink> So I understand that I am missing these rrd files, but how do I get them?
[02-Jun-2009 10:17:26] <chudler> I noticed that all uses of drive(blah), blah() always calls yield twice, but I can't make sense of it
[02-Jun-2009 10:18:36] <rocket> chudler: sorry cant help ya .. havent really used it yet ..
[02-Jun-2009 10:19:04] <afink> Dieterbe: I'm getting all kinds of oid is bad errors on all performance meters
[02-Jun-2009 10:19:15] <chudler> rocket: thanks anyway, I will keep hacking. I tried using plain twisted defers, but that wreaked havoc.
[02-Jun-2009 10:20:44] <capricorn-80> hi
[02-Jun-2009 10:21:03] <capricorn-80> i need some help regarding the application running on machine
[02-Jun-2009 10:21:23] <capricorn-80> i have configured zenoss .. I am monitoring my local system
[02-Jun-2009 10:21:58] <capricorn-80> i am getting cpu usage and memory related graphs
[02-Jun-2009 10:22:48] <capricorn-80> is it possible that zenoss will automatically recogize which applications/deamons are running on my server :
[02-Jun-2009 10:29:27] <capricorn-80> mrayzenoss: you there
[02-Jun-2009 10:31:00] <mrayzenoss> yeah
[02-Jun-2009 10:32:36] <mrayzenoss> Jane Curry's paper on process monitoring is really good, covers configuration and all the different techniques you can use: http://www.skills-1st.co.uk/papers/jcurry.html
[02-Jun-2009 10:34:43] <capricorn-80> ok
[02-Jun-2009 10:35:28] <Dieterbe> afink: i've had actually very similar errors. but in my case i could fix them by setting my community string correctly (to public)
[02-Jun-2009 10:38:00] <afink> Dieterbe: I think I found it..http://forums.zenoss.com/viewtopic.php?p=16513#16513
[02-Jun-2009 10:40:09] <Dieterbe> afink: well, good luck with it. i can't help you anymore
[02-Jun-2009 10:41:51] <afink> Thanks Dieterbe
[02-Jun-2009 10:47:48] <capricorn-80> in the regex. i dont understand the commnad
[02-Jun-2009 10:47:57] <capricorn-80> we have to give the pid of the process ?
[02-Jun-2009 10:51:22] <tyler79> has anyone seen caseyh? I'm considering giving up on zenoss on freebsd unless he made progress fixing the install
[02-Jun-2009 10:51:51] <capricorn-80> mrayzenoss: is it possible that zenoss automatically find different applications
[02-Jun-2009 10:52:37] <mrayzenoss> tyler79: he posted last night that he's almost done with his write-up
[02-Jun-2009 10:52:51] <mrayzenoss> tyler79: http://www.caseyhillman.com/?p=134
[02-Jun-2009 10:52:58] <tyler79> thanks!
[02-Jun-2009 10:53:54] <jb> no, you need to provide the regular expression
[02-Jun-2009 10:54:08] <jb> not the PID.. the PID always changes (if the process is restarted)
[02-Jun-2009 10:54:21] <rocket> mrayzenoss: for perf collecting using ssh am I going to be making command datasources?
[02-Jun-2009 10:54:35] <mrayzenoss> yeah
[02-Jun-2009 10:54:53] <rocket> placing items in libexec basically right?
[02-Jun-2009 10:54:55] <mrayzenoss> rocket: I've got the tests moved over, I should be getting them checked into trunk later today
[02-Jun-2009 10:55:06] <rocket> mrayzenoss: sweet .. thanks
[02-Jun-2009 10:55:07] <mrayzenoss> rocket: yeah
[02-Jun-2009 10:55:24] <mrayzenoss> if you need a script
[02-Jun-2009 10:55:41] <mrayzenoss> you can also just put the raw command in the datasource
[02-Jun-2009 10:55:51] <mrayzenoss> ie. /usr/bin/fping
[02-Jun-2009 10:55:56] <capricorn-80> regex for httpd . example
[02-Jun-2009 10:56:25] <jb> you can use something like.. " ^.*\/*httpd"
[02-Jun-2009 10:56:41] <capricorn-80> ok let me try
[02-Jun-2009 10:57:49] <capricorn-80> jb: i have use the same regex u told
[02-Jun-2009 10:57:52] <capricorn-80> is that fine ?
[02-Jun-2009 10:58:36] <jb> yes
[02-Jun-2009 10:58:52] <capricorn-80> ok thanks
[02-Jun-2009 11:02:12] <mloven> Anyone here actively (and successfully) monitoring any Dell PowerConnect 35xx switches?
[02-Jun-2009 11:04:57] <capricorn-80> jb: done but still no getting any httpd process in OS
[02-Jun-2009 11:06:14] <rmatte> I'm just finishing creating a new client base vmware image with Zenoss installed (2.4.1). All of a sudden I'm getting constant heartbeat failure alerts for each process, and they aren't going away like they normally eventually do, any ideas on what might be causing this?
[02-Jun-2009 11:10:59] <rmatte> ah, I configured NTP
[02-Jun-2009 11:11:06] <rmatte> apparently that can cause this type of issue
[02-Jun-2009 11:11:14] <rmatte> now to figure out how to fix it...
[02-Jun-2009 11:14:38] <capricorn-80> hmm
[02-Jun-2009 11:16:18] <jb> remodel the device..
[02-Jun-2009 11:16:57] <capricorn-80> jb: thats for me
[02-Jun-2009 11:19:54] <rmatte> I just went in to Event Manager and selected "Clear All Heartbeats", will see if that fixes it...
[02-Jun-2009 11:20:56] <rmatte> seems to have possibly done the trick
[02-Jun-2009 11:23:50] <capricorn-80> still not working
[02-Jun-2009 11:28:27] <capricorn-80> jb: need ur 5 mins
[02-Jun-2009 11:33:18] <jb> well, you aren't really providing any details
[02-Jun-2009 11:35:44] <capricorn-80> jb: trying to monitor my processes on linux server. Process class - name = httpd , regex = ^.*\/*httpd , monitor = true
[02-Jun-2009 11:35:52] <jb> ok
[02-Jun-2009 11:36:34] <jb> is it showing up under the Processes box in "OS" ?
[02-Jun-2009 11:37:32] <capricorn-80> zProperties Configuration: zAlertOnRestart = true , zCountProcs = true , zFailSeverity = critical , zMonitor = true
[02-Jun-2009 11:37:34] <capricorn-80> no
[02-Jun-2009 11:37:40] <jb> did you remodel the device?
[02-Jun-2009 11:38:30] <capricorn-80> what do u mean my remodel
[02-Jun-2009 11:38:37] <capricorn-80> by
[02-Jun-2009 11:38:43] <capricorn-80> restart
[02-Jun-2009 11:38:48] <jb> no
[02-Jun-2009 11:38:58] <capricorn-80> then
[02-Jun-2009 11:39:57] <jb> Browse to the device, click the down arrow, and choose Manage -> Model Device
[02-Jun-2009 11:41:15] <capricorn-80> i have Add reset features under Manage
[02-Jun-2009 11:41:55] <capricorn-80> lock
[02-Jun-2009 11:47:06] <capricorn-80> i guess i provide you information this time
[02-Jun-2009 11:52:10] <Kromel> *pulls hair* I'm trying to login a fresh install of Zenoss (VM Image) and the default root loging doesn't work.
[02-Jun-2009 11:52:24] <Kromel> the Zenoss web interface works fine
[02-Jun-2009 11:52:34] <Kromel> it's the linux OS i'm trying to log in to
[02-Jun-2009 11:53:33] <Kromel> thought it was username: root with nothing for the password.
[02-Jun-2009 11:53:48] <Kromel> then I would have to issue a passwd command to assign it a password
[02-Jun-2009 12:02:05] <capricorn-80> jb: got it
[02-Jun-2009 12:02:18] <capricorn-80> i was searching in device not in the particular device
[02-Jun-2009 12:10:35] <rocket> bah .. I need my regex book at the office .. I cant remember what [%-] will do .. :/
[02-Jun-2009 12:11:07] <rocket> nevermind .. figured it out ..
[02-Jun-2009 12:15:07] <capricorn-80> hahaha
[02-Jun-2009 12:15:35] <capricorn-80> i remove the entried of one of the processes and now i cant edit it "P
[02-Jun-2009 12:16:18] <capricorn-80> because there is no entry in the process class and the process status is monitor
[02-Jun-2009 12:16:25] <capricorn-80> how can i delete it ?
[02-Jun-2009 12:19:21] <capricorn-80> i guess i have solved it
[02-Jun-2009 12:20:02] <rmatte> Kromel: no, root is never defaulted to blank, you should have been asked to set a root password during the linux install
[02-Jun-2009 12:20:36] <rmatte> Kromel: if it's a virtual appliance, then it might be something like root/root or root/zenoss
[02-Jun-2009 12:20:38] <rmatte> no idea
[02-Jun-2009 12:32:19] <Kromel> figured it out
[02-Jun-2009 12:32:34] <Kromel> I had to use the VMware Infrastructure client
[02-Jun-2009 12:32:41] <Kromel> went in via Console
[02-Jun-2009 12:32:44] <capricorn-80> which collector is used for proccesses ?\
[02-Jun-2009 12:33:02] <capricorn-80> zcollectorplugins
[02-Jun-2009 12:34:53] <Kromel> thinking maybe I should install Zenoss on a dedicated linux
[02-Jun-2009 12:36:51] <Kromel> how can I download a zenpack using the zenoss management console?
[02-Jun-2009 12:37:00] <Kromel> i'll google that question right now
[02-Jun-2009 12:40:33] <Kromel> ahh wget!
[02-Jun-2009 13:19:58] <chudler> anyone see if google is slow for them right now?
[02-Jun-2009 13:20:16] <rocket> nope not slow for me
[02-Jun-2009 13:20:32] <rocket> anyone here work with command parsers?
[02-Jun-2009 13:20:52] <rocket> trying to understand componentScanValue .. the documentation doesnt make much sense to me at the moment
[02-Jun-2009 13:59:12] <asmith> hello there
[02-Jun-2009 13:59:23] <asmith> zenoss newbie here
[02-Jun-2009 13:59:46] <asmith> i'd like to respond to a zenoss event with some custom code
[02-Jun-2009 13:59:55] <asmith> is ZenPacks the right way to do this?
[02-Jun-2009 14:00:02] <asmith> i don't really care about a UI
[02-Jun-2009 14:01:30] <mrayzenoss> sounds like an event transform might be what you want
[02-Jun-2009 14:01:34] <mrayzenoss> depends what you're trying to do
[02-Jun-2009 14:02:15] <asmith> well
[02-Jun-2009 14:02:54] <asmith> when i get a ping status my managed host
[02-Jun-2009 14:02:58] <asmith> "might" be down
[02-Jun-2009 14:03:32] <asmith> i'd like to have a function to do a few more checks before pulling the ripcord and trying to modify some other configuration
[02-Jun-2009 14:03:54] <asmith> right now i've got this hacked up in the event manager
[02-Jun-2009 14:04:29] <asmith> to execute a shell program when this event happens and give it the stringified event as input
[02-Jun-2009 14:04:35] <asmith> but that seems like a nasty hack
[02-Jun-2009 14:04:43] <mrayzenoss> well, that would be an Event Command
[02-Jun-2009 14:04:53] <asmith> especially since the program it ultimately invokes is a python script
[02-Jun-2009 14:05:09] <asmith> which I'd like to just run inside the same interpreter that zenoss is running under
[02-Jun-2009 14:05:46] <mrayzenoss> ok, if you want to stick with Python, an Event Transform, if you want to use the shell, Event Command
[02-Jun-2009 14:05:59] <mrayzenoss> http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch07s01.html
[02-Jun-2009 14:06:19] <mrayzenoss> here are some Event Transform examples: http://www.zenoss.com/community/wiki/EventTransforms
[02-Jun-2009 14:07:02] <asmith> excellent
[02-Jun-2009 14:07:11] <asmith> a silly question
[02-Jun-2009 14:07:24] <asmith> but where do I "inject" those transforms into zenoss once I've written them?
[02-Jun-2009 14:07:51] <mrayzenoss> so if you want this on Ping Down
[02-Jun-2009 14:08:54] <mrayzenoss> /Events/Status/Ping
[02-Jun-2009 14:09:20] <mrayzenoss> More->Transform
[02-Jun-2009 14:10:15] <asmith> Can my tranform methods keep any state?
[02-Jun-2009 14:10:30] <mrayzenoss> you can look up other events
[02-Jun-2009 14:10:42] <mrayzenoss> and the state of the device
[02-Jun-2009 14:10:43] <asmith> I'd like to basically say -- wait for 5 Ping Down events
[02-Jun-2009 14:10:50] <asmith> and then do foo
[02-Jun-2009 14:11:11] <mrayzenoss> yeah
[02-Jun-2009 14:11:45] <asmith> but somebody else told me transforms don't have a count because you can only transform an event before it is committed to the db
[02-Jun-2009 14:11:55] <mrayzenoss> that's true
[02-Jun-2009 14:11:56] <asmith> and the count comes after the commit
[02-Jun-2009 14:12:21] <mrayzenoss> but you can have a transform that executes after the count
[02-Jun-2009 14:12:43] <mrayzenoss> add an EventClass mapping
[02-Jun-2009 14:12:52] <mrayzenoss> and put the transform in there instead of More->Transforms
[02-Jun-2009 14:14:01] <asmith> alright -- we've strayed outside my comfort zone, but that sounds like something to go on
[02-Jun-2009 14:14:03] <asmith> thank you
[02-Jun-2009 14:14:14] <mrayzenoss> no prob
[02-Jun-2009 14:14:17] <asmith> actually
[02-Jun-2009 14:14:19] <mrayzenoss> there's a good paper on event management
[02-Jun-2009 14:14:21] <asmith> one more question
[02-Jun-2009 14:14:30] <asmith> that transform method
[02-Jun-2009 14:14:33] <mrayzenoss> http://www.skills-1st.co.uk/papers/jcurry.html Zenoss Event Management
[02-Jun-2009 14:14:52] <asmith> if i wanted to just keep state in a file or something
[02-Jun-2009 14:14:59] <asmith> do i have anyway to do that?
[02-Jun-2009 14:15:20] <mrayzenoss> well, I think you could do that in the transform since it's Python
[02-Jun-2009 14:15:20] <asmith> so i could map some kind of host id / event type / counter myself
[02-Jun-2009 14:15:32] <asmith> ok -- so i have all the power of python at my disposal
[02-Jun-2009 14:15:34] <asmith> great
[02-Jun-2009 14:15:36] <mrayzenoss> oh yeah
[02-Jun-2009 14:15:50] <mrayzenoss> if you go back to those examples, you see import statements
[02-Jun-2009 14:15:58] <asmith> nice
[02-Jun-2009 14:16:07] <mrayzenoss> and a bit dangerous
[02-Jun-2009 14:16:14] <asmith> well - the EventClass mapping sounds like the way to go
[02-Jun-2009 14:16:26] <mrayzenoss> good luck
[02-Jun-2009 14:16:35] <asmith> thanks
[02-Jun-2009 14:35:04] <genji0714> hello all....i am doing a fresh install of zenoss core 2.4 and zenoss will only accept an alphanumeric for the mysql root password...is there a way around this because my root password for mysql has symbols in it.
[02-Jun-2009 14:39:49] <mrayzenoss> genji0714: http://dev.zenoss.org/trac/ticket/5105
[02-Jun-2009 14:40:02] <mrayzenoss> You can use the --mysql-database calls to use your DB
[02-Jun-2009 14:48:51] <mloven> Anyone here successfully monitoring some PowerConnect 35xx switches?
[02-Jun-2009 14:55:30] <mrayzenoss> I haven't seen anything for that, but I'd love a ZenPack
[02-Jun-2009 14:56:12] <rocket> bah anyone remember where the interface statistics similar to the snmp output is on a linux box? ..
[02-Jun-2009 14:56:23] <rocket> working on enterprise stuff too much lately ..
[02-Jun-2009 14:56:49] <mloven> yeah, I would too Matt....
[02-Jun-2009 14:58:38] <mrayzenoss> rocket: ifconfig
[02-Jun-2009 14:59:02] <rocket> mrayzenoss: k .. just am not seeing ucastpackets etc
[02-Jun-2009 15:06:12] <mrayzenoss> ucast? I see RX/TX
[02-Jun-2009 15:07:06] <rocket> unicast ..
[02-Jun-2009 15:20:03] <genji0714> mrayzenoss: that didn't work
[02-Jun-2009 15:28:59] bzed_ is now known as bzed
[02-Jun-2009 15:44:16] <rocket> mrayzenoss: get the examples uploaded?
[02-Jun-2009 15:46:45] <mrayzenoss> working with Brian right now
[02-Jun-2009 15:47:06] <rocket> bah .. I am still having troubles with parsers .. :/
[02-Jun-2009 15:47:17] <rocket> ImportError: No module named ZenPacks.zenoss.LinuxMonitor-1.0.0-py2.4.egg.ZenPacks.zenoss.LinuxMonitor.parsers.linux.ifconfig
[02-Jun-2009 15:47:33] <rocket> I created an ifconfig parser .. to parse the output of ifconfig -a
[02-Jun-2009 15:47:43] <rocket> hrmmm what am I doing wrong ..
[02-Jun-2009 16:30:03] * alex___ waves hello to everyone
[02-Jun-2009 16:31:01] <alex___> can anyone point me towards a replacement for informant/WMI that will let me get windows performance data over SNMP and not require me to pay money for it
[02-Jun-2009 16:31:26] <chudler> I can't. But I thought informant was free?
[02-Jun-2009 16:31:49] <chudler> oh I see, you are looking for a replacement that is *also* free
[02-Jun-2009 16:32:06] <alex___> well, informant is free for non commercial use
[02-Jun-2009 16:32:31] <chudler> ouch. always a catch.
[02-Jun-2009 16:32:37] <alex___> tell me about it
[02-Jun-2009 16:33:03] <alex___> there are a few zenpacks but they require opening port 445 on my server to the world
[02-Jun-2009 16:33:08] <alex___> which is not happening
[02-Jun-2009 16:35:54] <chudler> it is slightly outside of my area. WMI is a good alternative. Zenoss enterprise uses it very effectively. I think it would be simple to build a similar zenpack, and others probably already have.
[02-Jun-2009 16:36:19] <chudler> but that does screw your "over SNMP" requirement.
[02-Jun-2009 16:36:30] <alex___> does enterprise connect to wmi directly or does the zenpack interface wmi to snmp
[02-Jun-2009 16:36:46] <chudler> direct to wmi, using samba wmic, I believe.
[02-Jun-2009 16:36:59] <alex___> my biggest issue with wmi is that every host i monitor is behind a different firewall
[02-Jun-2009 16:37:20] <alex___> so i have to open snmp and wmi each time
[02-Jun-2009 16:37:32] <alex___> do you know what port wmic listens on
[02-Jun-2009 16:39:58] <alex___> perhaps you can help me with another smaller issue
[02-Jun-2009 16:40:16] <alex___> i dont quite get the relationship between performance templates and devices
[02-Jun-2009 16:41:29] <alex___> ive added the ethernetCsmacd to one of my servers, so my Perf page shows 4 empty graph boxes for network interface data
[02-Jun-2009 16:41:45] <alex___> the data appears fine under the OS -> eth0 tab
[02-Jun-2009 16:46:31] <alex___> is there a way to either get the graphs to populate on the perf page like they are on the OS, or not appear at all on perf and just go to OS to view them
[02-Jun-2009 17:19:07] <mrayzenoss> alex___: you want to move the OS graphs to the perf page?
[02-Jun-2009 17:19:23] <mrayzenoss> I guess you could hack up the .pt files
[02-Jun-2009 17:40:07] <alex___> what does it mean if there is no checkbox next to a template in the All Templates list
[02-Jun-2009 18:19:12] <alex___> thanks for help
[02-Jun-2009 18:33:48] <kisielk> alright, time to upgrade to 2.4
[02-Jun-2009 19:36:05] <kisielk> is zenmigrate supposed to take a *really* long time to run?
[03-Jun-2009 00:00:42] [disconnected at Wed Jun 3 00:00:42 2009]
[03-Jun-2009 00:00:42] [connected at Wed Jun 3 00:00:42 2009]
[03-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[03-Jun-2009 07:59:32] <osd_mats> I must be stupid... I'm editing a data source that use ssh agains a device, but when i use the "test against device" field it just tries agains the zenoss server. What am i missing?
[03-Jun-2009 08:06:03] <SpunkMeYeR> hi all
[03-Jun-2009 08:06:52] <SpunkMeYeR> anybody here knows how to send an alert sms from zenoss?
[03-Jun-2009 08:23:21] <chudler> spunkmeyer: two common ways are 1) send email to the sms-email-gateway of the carrier, or 2) change Page Command in Settings tab to an SMS sender app
[03-Jun-2009 08:23:28] <chudler> dang, he just left :-0
[03-Jun-2009 08:27:54] <Dieterbe> chudler: at least you tried
[03-Jun-2009 08:30:15] <rocket> bah so trying to do development with a zenoss vmware image I think is broken with the unit tests .. :/
[03-Jun-2009 09:34:39] <malbon> mrayzenoss: yo
[03-Jun-2009 09:36:19] <mrayzenoss> myellow
[03-Jun-2009 09:38:03] <malbon> mrayzenoss: have you got to the MyEventViews zenpack in your queue?
[03-Jun-2009 09:38:39] <mrayzenoss> yeah, I'm getting it out today with 5 other ZenPacks
[03-Jun-2009 09:38:51] <malbon> top banana
[03-Jun-2009 09:39:25] <mrayzenoss> actually, I'll go commit it right now if you're in a hurry
[03-Jun-2009 09:39:40] <malbon> nope, by tomorrow is fine.
[03-Jun-2009 09:39:40] <mrayzenoss> I merged it in yesterday
[03-Jun-2009 09:39:49] <mrayzenoss> ok, cool
[03-Jun-2009 09:40:13] <malbon> We can raise a change control on that basis, and it's not going to be processed until the meeting tomororw.
[03-Jun-2009 09:40:40] <malbon> this new pack corrects the refresh problem?
[03-Jun-2009 09:40:57] <mrayzenoss> yup
[03-Jun-2009 09:41:35] <malbon> cool
[03-Jun-2009 09:42:11] <mrayzenoss> trac-zenpacks/changeset/229
[03-Jun-2009 09:43:04] <malbon> heh, not much changes does it.
[03-Jun-2009 09:43:17] <mrayzenoss> yeah, you could probably patch it your self
[03-Jun-2009 09:43:30] <mrayzenoss> by hand that is
[03-Jun-2009 09:43:40] <mrayzenoss> but I'll get it out officially in a ZenPack today
[03-Jun-2009 09:43:52] <malbon> I'll wait for the package. It's a feature that our users want but they can wait for the process.
[03-Jun-2009 09:44:14] <mrayzenoss> Florian also sent in a Reports portlet that's going up today too
[03-Jun-2009 09:44:43] <malbon> what does that do?
[03-Jun-2009 09:45:31] <mrayzenoss> "I also submit a new portlet that let one create shortcuts to his favorite reports. One there are tons of reports available, one may prefere to gather the ones he reads more on his dashboard."
[03-Jun-2009 09:45:43] <mrayzenoss> you just put your favorite reports on the dashboard for easy access
[03-Jun-2009 09:46:10] <malbon> ok, I can see that.
[03-Jun-2009 09:50:06] <malbon> mrayzenoss: anyhow gotta dash. Thanks for doing those updates.
[03-Jun-2009 09:50:16] <mrayzenoss> no problem
[03-Jun-2009 09:53:40] <xulfer> Hrm, anyone managed to monitor Trango's with any success in here?
[03-Jun-2009 09:56:16] <xulfer> Ha, just as i thought.
[03-Jun-2009 10:18:09] <rocket> bah it would help if I downloaded the 32bit stack for a 32bit system ..
[03-Jun-2009 10:20:33] <chudler> ('Unable to parse arguments', ['/opt/zenoss/lib/python/Zope2/Startup/run.py', '-v', '2c', '-c', 'public', '-t', '2.5', '-r', '2', '10.128.10.25:161'])
[03-Jun-2009 10:21:26] <chudler> ARG, I get this at random when using AgentProxy (netsnmp) inside the zenhub instance
[03-Jun-2009 10:22:02] <chudler> I believe something horrible is happening with (non) thread-safe use
[03-Jun-2009 10:22:29] <rocket> chudler: figuring out the agent proxy stuff?
[03-Jun-2009 10:22:51] <chudler> yes, it works fine, except for that thing.
[03-Jun-2009 10:24:16] <chudler> I don't know which is worse. Doing cacheRRD inside a threshold with distributed collectors (locks up zenhub), or doing AgentProxy.get inside a twisted event loop (random failure, as above).
[03-Jun-2009 10:25:06] <rocket> anything twisted is evil ..
[03-Jun-2009 10:26:41] <chudler> I suspect a newer Zope might not have this problem. Lacking evidence, though, and it's a long way off yet before Zenoss can move to it.
[03-Jun-2009 10:30:03] <rocket> chudler: there is a newer zope in the tree
[03-Jun-2009 10:30:18] <rocket> if you wanted to test against the unstable tree
[03-Jun-2009 10:30:27] <rocket> but that might not prove anything ..
[03-Jun-2009 10:30:34] <mrayzenoss> yeah, Ian said he stabilized it this morning
[03-Jun-2009 10:30:51] <mrayzenoss> QA had opened several blockers on trunk with the new 2.11.2
[03-Jun-2009 10:31:17] <chudler> sweet. Not a great test though. I really need to get to the bottom of the problem. I wonder about the argument list that starts with "'/opt/zenoss/lib/python/Zope2/Startup/run.py'" I wonder why it is there.
[03-Jun-2009 10:43:00] <chudler> oh yeah, I forgot. The first argument is normally the name of the file calling. Hmm. okay, just thinking out loud :-)
[03-Jun-2009 10:53:28] <jb> hey everyone
[03-Jun-2009 10:54:39] <jb> hm, zenwin/zeneventlog cycles are like 60+ seconds now
[03-Jun-2009 11:01:40] <rocket> mrayzenoss: ping
[03-Jun-2009 11:02:17] <mrayzenoss> hello
[03-Jun-2009 11:02:35] <rocket> hey I am just getting to the point where I can run unit tests
[03-Jun-2009 11:02:45] <rocket> and its working .. and it shows me all the tests fail
[03-Jun-2009 11:03:01] <kisielk> alright, I upgraded to 2.4.1 from 2.3.3 yesterday, but it seems my LDAP auth broke
[03-Jun-2009 11:03:06] <rocket> but it doesnt show much more than that .. is there a more verbose option or someting I am missing?
[03-Jun-2009 11:03:13] <kisielk> I had to remove the product in order to even get the dashboard to work
[03-Jun-2009 11:03:24] <kisielk> but now I can't manage my users, or authenticate them against LDAP
[03-Jun-2009 11:03:30] <kisielk> this is a stack installer
[03-Jun-2009 11:10:57] <mrayzenoss> rocket: Good question, I'm asking around
[03-Jun-2009 11:11:34] <rocket> otherwise its very difficult to troubleshoot
[03-Jun-2009 11:11:34] <kisielk> how do I get it to work again ? do I have to reinstall the auth LDAP stuff?
[03-Jun-2009 11:11:37] <mrayzenoss> kisielk: sorry, not much help from me. I just posted on an LDAP thread looking for people who have it working to help people having trouble
[03-Jun-2009 11:11:52] <kisielk> it looks like the location of Python moved
[03-Jun-2009 11:11:59] <kisielk> in 2.4.1, I suspect that may be the problem
[03-Jun-2009 11:13:32] <mrayzenoss> rocket: I purposely made one of my tests fail and I got a stack trace with the AssertionError
[03-Jun-2009 11:13:35] <mrayzenoss> what are you seeing?
[03-Jun-2009 11:14:02] <rocket> an assertion that it failed .. but nothing indicating why
[03-Jun-2009 11:14:41] <rocket> expected 12 values, actual 0 .. but it doesnt show what the values were etc
[03-Jun-2009 11:15:08] <mrayzenoss> hmm... Ian said he's writing a patch to put more verbosity into runtest
[03-Jun-2009 11:15:27] <mrayzenoss> but I think that won't help us, since he already updated it substantially for the new Zope 2.11.2 stuff
[03-Jun-2009 11:16:23] <mrayzenoss> here's what mine looks like: http://pastebin.com/d1c180d1a
[03-Jun-2009 11:16:59] <rocket> hrmm .. mine doesnt look like that .. ;p
[03-Jun-2009 11:17:08] <mrayzenoss> of course, I see that it doesn't show which test it was
[03-Jun-2009 11:17:15] <mrayzenoss> just that it failed
[03-Jun-2009 11:17:25] <mrayzenoss> I mean, I know to grep for 103707
[03-Jun-2009 11:17:59] <rocket> http://pastebin.com/m2e29e41b
[03-Jun-2009 11:20:42] <mrayzenoss> trying to get Ian to look at it
[03-Jun-2009 11:21:23] <rocket> sweet thanks
[03-Jun-2009 11:21:48] <mrayzenoss> well, 2.5 now has -vvvvvv as a runtests option
[03-Jun-2009 11:21:59] <rocket> bah that doesnt help me
[03-Jun-2009 11:22:04] <mrayzenoss> but that doesn't help yet
[03-Jun-2009 11:23:52] <mrayzenoss> try zopectl test --libdir /tmp/ZenPacks.community.Gentoo -vv
[03-Jun-2009 11:23:55] <mrayzenoss> or whatever your path is
[03-Jun-2009 11:24:35] <kisielk> Gentoo zenpack?
[03-Jun-2009 11:24:46] * kisielk 's ears perk up
[03-Jun-2009 11:24:47] <mrayzenoss> kisielk: that's what rocket is writing
[03-Jun-2009 11:24:53] <kisielk> rocket: what does it do?
[03-Jun-2009 11:24:54] <mrayzenoss> an SSH one
[03-Jun-2009 11:25:09] <rocket> ssh based one .. gets the list of the software installed etc using eix
[03-Jun-2009 11:25:22] <kisielk> ah
[03-Jun-2009 11:25:25] <rocket> updating the linux one to do the interface perf stuff
[03-Jun-2009 11:25:28] <kisielk> I actually wrote an SNMP module for that
[03-Jun-2009 11:25:44] <kisielk> works similar to the Ubuntu one that was posted a few months back
[03-Jun-2009 11:25:48] <rocket> this one is ssh based though .. so you wouldnt have to install that ..
[03-Jun-2009 11:26:05] <rocket> just need the eix utility installed and its database updated every so often
[03-Jun-2009 11:26:13] <kisielk> yeah
[03-Jun-2009 11:26:29] <kisielk> although I've deprecated all of our Gentoo machines other than our cluster
[03-Jun-2009 11:26:35] <kisielk> so it's not much use to me any more
[03-Jun-2009 11:26:36] <rocket> basically I am using it as a way to learn the ssh stuff and flesh out the unit test stuff
[03-Jun-2009 11:27:01] <rocket> once I have one of these done the other linux distos should be pretty easy
[03-Jun-2009 11:27:30] <mrayzenoss> rocket: any change in output?
[03-Jun-2009 11:27:36] <capricorn-80> hi
[03-Jun-2009 11:27:48] <capricorn-80> can i run zenoss and mrtg on the same server ?
[03-Jun-2009 11:28:14] <mrayzenoss> capricorn-80: not sure why you'd need to
[03-Jun-2009 11:28:14] <rocket> mrayzenoss: nope
[03-Jun-2009 11:28:49] <mrayzenoss> rocket: anything with -vvvvv ?
[03-Jun-2009 11:28:53] <mrayzenoss> 5 vs
[03-Jun-2009 11:28:55] <rocket> nope .. the same ..
[03-Jun-2009 11:29:16] <rocket> I am sure its something wrong with my setup .. but there is very little to go on to figure out
[03-Jun-2009 11:29:52] <mrayzenoss> rocket: why don't you check it into subversion and I'll look at it
[03-Jun-2009 11:29:59] <capricorn-80> rocket: why not ?
[03-Jun-2009 11:30:25] <rocket> mrayzenoss: I would except svn access is blocked here at work :/
[03-Jun-2009 11:30:31] <mrayzenoss> nutz
[03-Jun-2009 11:30:34] <rocket> I would need to run to our lab with the ports open
[03-Jun-2009 11:30:47] <rocket> which I can do if absolutely necessary ..
[03-Jun-2009 11:31:45] <mrayzenoss> just tar it up and email it to me
[03-Jun-2009 11:31:51] <mrayzenoss> I'll see if I can debug it
[03-Jun-2009 11:32:02] <rocket> sure
[03-Jun-2009 11:32:09] <mrayzenoss> but now I'm headed to lunch
[03-Jun-2009 11:32:23] * mrayzenoss is off to lunch
[03-Jun-2009 11:32:49] <capricorn-80> rocket: ?
[03-Jun-2009 11:33:16] <rocket> capricorn-80: I was talking to mrayzenoss not answering your question
[03-Jun-2009 11:33:40] <capricorn-80> ops
[03-Jun-2009 11:33:43] <capricorn-80> can i run zenoss and mrtg on the same server ?
[03-Jun-2009 11:33:46] <capricorn-80> so answer it
[03-Jun-2009 11:34:15] <rocket> I dont know why you wouldnt be able to
[03-Jun-2009 11:34:37] <capricorn-80> rocket: what i wouldnt be able to ?
[03-Jun-2009 11:34:55] <rocket> capricorn-80: non-english native speaker?
[03-Jun-2009 11:36:02] <capricorn-80> rocket: yea . swedish
[03-Jun-2009 11:36:15] <capricorn-80> rocket: your sentence was incomplete
[03-Jun-2009 11:36:47] <rocket> I dont know why you wouldnt be able to run both at the same time
[03-Jun-2009 11:37:10] <Mousey> bleh
[03-Jun-2009 11:37:17] <Mousey> making classes is hard
[03-Jun-2009 11:37:28] <capricorn-80> rocket: ahan .. i was checking it .. its testing server
[03-Jun-2009 11:37:40] <capricorn-80> i will move to zenoss on other server
[03-Jun-2009 11:37:54] <capricorn-80> i am just testing right now coz i dont have any other server
[03-Jun-2009 11:38:02] <rocket> it would be similar to running dns and email on the same server
[03-Jun-2009 11:38:19] <capricorn-80> rocket: ok
[03-Jun-2009 11:38:48] <kisielk> mrayzenoss: alright, I got it working
[03-Jun-2009 11:38:55] <kisielk> mrayzenoss: I just had to easy_install python-ldap as the zenoss user
[03-Jun-2009 11:39:39] <capricorn-80> rocket: i add my cisco router to zenoss . its working fine and i am getting cpu and memory information
[03-Jun-2009 11:39:51] <capricorn-80> but i am not getting bandwidth graph
[03-Jun-2009 11:40:07] <rocket> check logs for errors?
[03-Jun-2009 11:40:30] <capricorn-80> rocket: log msg for me ?
[03-Jun-2009 11:46:29] <capricorn-80> anyone who can help me that how can i monitor bandwidth using zenoss and make graph similar like mrtg for bandwidth
[03-Jun-2009 12:00:33] <Mousey> why when i set the OS make, less and less OS version options bercome available?
[03-Jun-2009 12:01:45] <Mousey> i set it to debian for my first device add, and set it to unstable, for my second device add "unstable" is no longer an option for "debian" so i set it to lenny. later i went back to edit the device, and both OS fields were empty, so i set the make to "debian" and now "lenny" isn't available. =(
[03-Jun-2009 13:16:13] <bedwards> eric, you had a question about component scan value
[03-Jun-2009 13:21:32] <rocket> yeah I am trying to understand what it does
[03-Jun-2009 13:21:52] <rocket> eg whats the difference between mount and id(default)
[03-Jun-2009 13:22:04] <rocket> as my tests seem to succeed regardless of its setting
[03-Jun-2009 13:22:21] <rocket> I am building a parser for ifconfig
[03-Jun-2009 13:22:31] <rocket> for performance graphing points
[03-Jun-2009 13:22:45] <rocket> and I am assuming I need to set it to something like
[03-Jun-2009 13:22:47] <rocket> componentScanValue = 'interfaceName'
[03-Jun-2009 13:33:30] <kdub> i'm trying to install zenoss core via dpkg on a debian system. the install fails saying it cannoct connect to the local mysql socket. any help?
[03-Jun-2009 13:41:20] <mrayzenoss> kdub: are you already running mysql?
[03-Jun-2009 13:42:08] <rocket> bedwards: ?
[03-Jun-2009 13:44:40] <bedwards> don't use ComponentCommandParser, use CommandParser instead
[03-Jun-2009 13:45:32] <rocket> bedwards: I am curious as to why? eg I was looking at the df and dfi examples
[03-Jun-2009 13:46:35] <bedwards> it is harder to understand once it has been written, the framework is very limiting, and it is harder to debug
[03-Jun-2009 13:46:50] <rocket> bedwards: so its the old way of doing it basically?
[03-Jun-2009 13:47:15] <bedwards> i don't use it
[03-Jun-2009 13:48:09] <rocket> ok .. I can rewrite it to use the CommandParser .. should be pretty easy to reuse the regex's etc ..
[03-Jun-2009 13:48:33] <rocket> bedwards: I agree about the debugging I had to add print statements to the main code to figure out what was going on with it
[03-Jun-2009 13:50:18] <bedwards> python has great string processing tools outside of regexes, so ComponentCommandParser is very limiting in that way too. i use regexes as a last resort in python.
[03-Jun-2009 13:51:20] <kdub> mrayzenoss: i found some mysql processes running, killed them, and am reinstalling
[03-Jun-2009 13:51:22] <bedwards> i believe that there are some component parsers that subclass CommandParser in core
[03-Jun-2009 13:51:33] <kdub> no errors so far, this installer takes forever though...
[03-Jun-2009 13:51:39] <rocket> yeah there are ... I will take a look at them again ...
[03-Jun-2009 13:52:00] <rocket> and the linuxmonitor zenpack has a cpu one ..
[03-Jun-2009 13:53:16] <rocket> thanks bedwards
[03-Jun-2009 13:53:21] <rocket> Thats all I needed
[03-Jun-2009 13:57:18] <kdub> is this install supposed to take like an hour+ ?
[03-Jun-2009 14:02:29] <rmatte> what is the point of the list of snmp strings in zProperties? Zenoss doesn't even seem to use them.
[03-Jun-2009 14:02:57] <mrayzenoss> kdub: no
[03-Jun-2009 14:03:25] <mrayzenoss> rmatte: do you mean the zSnmpCommunities?
[03-Jun-2009 14:06:23] <kdub> mrayzenoss: i killed the apt-get, i'm hoping its fine to run
[03-Jun-2009 14:07:50] <mrayzenoss> kdub: I think it installs to /opt/zenoss on Debian/Ubuntu
[03-Jun-2009 14:08:02] <mrayzenoss> I could be mistaken, I run from source on Debian PPC
[03-Jun-2009 14:08:53] <kdub> it put something called bitrock into /opt, most of it is in /usr/local
[03-Jun-2009 14:09:39] <mrayzenoss> yeah, bitrock is the provider of the stack installs
[03-Jun-2009 14:09:55] <mrayzenoss> if you do a dpkg --list, I assume there's nothing zenoss related?
[03-Jun-2009 14:11:40] <kdub> no, zenoss-stack is there, it hung in the postinst script
[03-Jun-2009 14:13:16] <mrayzenoss> dpkg --purge zenoss-stack
[03-Jun-2009 14:15:50] <kdub> i did it, tried to reinstall, its complaining though
[03-Jun-2009 14:20:00] <rmatte> mrayzenoss: yes
[03-Jun-2009 14:20:37] <rmatte> mrayzenoss: My assumption was that Zenoss would go through that list and try each string when modelling a device until it found one that works
[03-Jun-2009 14:20:39] <mrayzenoss> rmatte: they're supposed to be used when new devices are discovered
[03-Jun-2009 14:20:42] <rmatte> but apparently it doesn't do that
[03-Jun-2009 14:20:53] <rmatte> oh, so they are only used on auto-discovery?
[03-Jun-2009 14:20:57] <rmatte> not on manual addition?
[03-Jun-2009 14:21:21] <mrayzenoss> yeah, that sounds right
[03-Jun-2009 14:21:26] <rmatte> I see
[03-Jun-2009 14:21:29] <rmatte> k, thanks
[03-Jun-2009 14:41:43] <kdub> do these installer errors make sense to anyone?
[03-Jun-2009 14:41:46] <kdub> http://pastebin.ca/1446659
[03-Jun-2009 14:43:18] <mrayzenoss> kdub: I sent that to our build engineer, he'll take a look
[03-Jun-2009 14:43:39] <mrayzenoss> kdub: btw, 2.4.1 is out so you should probably install that instead of 2.4.0
[03-Jun-2009 14:44:41] <mrayzenoss> this is 32-bit Debian 4 correct?
[03-Jun-2009 14:51:19] <rmatte> the job manager is definitely nice
[03-Jun-2009 14:51:27] <rmatte> don't have to manually kill auto-discoveries now
[03-Jun-2009 14:51:36] <mrayzenoss> more stuff will become Job Manager aware in time
[03-Jun-2009 14:51:41] <rmatte> right on
[03-Jun-2009 14:53:02] <kdub> mrayzenoss: its etch, 32bit
[03-Jun-2009 14:53:30] <kdub> i still havent figured out debians naming scheme
[03-Jun-2009 14:54:11] <mrayzenoss> kdub: it's based on characters from Toy Story
[03-Jun-2009 14:54:33] <mrayzenoss> woody, etch, lenny, sarge
[03-Jun-2009 14:54:45] <mrayzenoss> experimental is "sid" because it will break your toys
[03-Jun-2009 14:55:05] <kdub> experimental is _always_ named sid?
[03-Jun-2009 14:55:08] <mrayzenoss> yeah
[03-Jun-2009 14:55:14] <mrayzenoss> I think Bruce Perens was working for Pixar at the time
[03-Jun-2009 14:55:20] <mrayzenoss> when they started the naming scheme
[03-Jun-2009 14:55:27] <rmatte> hehe
[03-Jun-2009 14:55:30] <kdub> oh, and the stable changes from like etch to lenny
[03-Jun-2009 14:55:34] <mrayzenoss> yeah
[03-Jun-2009 14:55:40] <mrayzenoss> Lenny is current stable
[03-Jun-2009 14:55:42] <mrayzenoss> 5.0
[03-Jun-2009 14:55:43] <CapriCoRN^80> mrayzenoss: i want to monitor my bandwidth
[03-Jun-2009 14:55:45] <mrayzenoss> Etch was 4.0
[03-Jun-2009 14:55:47] <rmatte> Sid Phillips
[03-Jun-2009 14:55:50] <kdub> makes more sense, i asked in #debian one day, and got chewed out for asking
[03-Jun-2009 14:55:57] <CapriCoRN^80> mrayzenoss: please refer me some tutorials
[03-Jun-2009 14:56:07] <chudler> wow, really? well, it is in the FAQ.
[03-Jun-2009 14:56:38] <CapriCoRN^80> mrayzenoss: I have configured my cisco router and i was getting CPU and memory graphs
[03-Jun-2009 14:56:42] <mrayzenoss> CapriCoRN^80: I assume you have a router or switch that exposes bandwidth in an OID you can hit via SNMP
[03-Jun-2009 14:56:44] <mrayzenoss> right on
[03-Jun-2009 14:57:39] <CapriCoRN^80> mrayzenoss:how can i set it to monitor bandwidth ?
[03-Jun-2009 14:57:50] <mrayzenoss> http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch06s02.html
[03-Jun-2009 14:58:05] <mrayzenoss> CapriCoRN^80: you'll set up an SNMP datasource to watch the values on the OID
[03-Jun-2009 14:58:20] <mrayzenoss> and a threshold to watch for spikes
[03-Jun-2009 14:58:27] <mrayzenoss> same thing you'd do with MRTG
[03-Jun-2009 15:00:09] <CapriCoRN^80> mrayzenoss: ok thanks for this link .. i will check it tomorrow
[03-Jun-2009 15:00:26] <mrayzenoss> kdub: ok, let's start from scratch. apt-get remove zenoss-stack
[03-Jun-2009 15:00:33] <mrayzenoss> kdub: dpkg --purge zenoss-stack
[03-Jun-2009 15:00:47] <YC> Is there any kind of logging for the google maps integration? I'm trying to find out why my new install isn't placing the location dots.
[03-Jun-2009 15:00:47] <mrayzenoss> delete everything zenoss related in /usr/local/zenoss
[03-Jun-2009 15:00:55] <mrayzenoss> and /tmp/ if there's anything left over
[03-Jun-2009 15:01:07] <mrayzenoss> YC: 2.4.0 or 2.4.1?
[03-Jun-2009 15:01:30] <mrayzenoss> YC: there were Google map issues with 2.4.0
[03-Jun-2009 15:01:44] <mrayzenoss> that said, there may be some logging in z2.log I believe
[03-Jun-2009 15:01:56] <YC> 2.4.1
[03-Jun-2009 15:02:38] <mrayzenoss> YC: what are you entering in the Address fields?
[03-Jun-2009 15:03:01] <YC> Example: "668 High St, Lower Hutt, Wellington, New Zealand"
[03-Jun-2009 15:03:10] <YC> Resolves fine with maps.google.com
[03-Jun-2009 15:04:05] <mrayzenoss> oops. Could you repost that?
[03-Jun-2009 15:04:07] <YC> Welcome back?
[03-Jun-2009 15:04:10] <YC> [08:03] <YC> Example: "668 High St, Lower Hutt, Wellington, New Zealand"
[03-Jun-2009 15:04:11] <YC> [08:03] <YC> Resolves fine with maps.google.com
[03-Jun-2009 15:04:31] <YC> This is an excellent product, BTW. Much nicer than opennms so far.
[03-Jun-2009 15:05:01] <kdub> mrayzenoss: done and done
[03-Jun-2009 15:05:02] <mrayzenoss> YC: glad to hear it, let me see what happens when I enter that address
[03-Jun-2009 15:05:12] <kdub> mrayzenoss: nothing in /opt either
[03-Jun-2009 15:07:05] <mrayzenoss> YC: hmm.. maps for me, somewhere near Mitchell Park?
[03-Jun-2009 15:07:19] <mrayzenoss> kdub: ok, grab the 2.4.1 Deb
[03-Jun-2009 15:07:48] <YC> Yep.
[03-Jun-2009 15:07:54] <kdub> mrayzenoss: there is still a /home/zenoss and things in var/cache, /var/lib and /etc related to it though
[03-Jun-2009 15:08:22] <mrayzenoss> kdub: wipe it all out
[03-Jun-2009 15:08:30] <YC> For me, any maps that turn up are centred dead on the equator, on the international date line.
[03-Jun-2009 15:08:42] <mrayzenoss> YC: I assume you don't have any Javascript blocking?
[03-Jun-2009 15:08:43] <YC> Maps from the location tab are blank.
[03-Jun-2009 15:08:58] <YC> Nope. Standard firefox install.
[03-Jun-2009 15:08:59] <mrayzenoss> YC: have you entered the Google Maps API?
[03-Jun-2009 15:09:35] <mrayzenoss> on the Settings?
[03-Jun-2009 15:09:37] <YC> Yep.
[03-Jun-2009 15:09:52] <mrayzenoss> weird
[03-Jun-2009 15:09:59] <YC> On dashboard view, it shows a map, no dots.
[03-Jun-2009 15:10:22] * YC tries IE.
[03-Jun-2009 15:10:39] <YC> Heh. IE8 is not a fan of zenoss.
[03-Jun-2009 15:11:14] <kdub> mrayzenoss:running 'dpkg -i zenoss-stack_2.4.1_i386.deb'
[03-Jun-2009 15:12:16] <kdub> mrayzenoss: just a thought, could firewalls get in the way of the installation?
[03-Jun-2009 15:12:41] <mrayzenoss> kdub: not really
[03-Jun-2009 15:13:05] <kdub> alright, it looks like its doing a normal install, its working on 'setting up zenoss-stack'
[03-Jun-2009 15:13:25] <kdub> didnt make it past this step last time... :/
[03-Jun-2009 15:14:02] <mrayzenoss> YC: yeah, IE8 won't be supported until the next release
[03-Jun-2009 15:14:38] <kdub> mrayzenoss: weird mysql error http://pastebin.ca/1446692
[03-Jun-2009 15:15:09] <mrayzenoss> kdub: is another mysql running?
[03-Jun-2009 15:15:33] <kdub> ps ax says 2 are running, i didnt start them directly though
[03-Jun-2009 15:15:53] <mrayzenoss> well, the installer lays one down and runs it
[03-Jun-2009 15:16:32] <kdub> i should probably kill both and try again
[03-Jun-2009 15:16:57] <mrayzenoss> yeah, you might want to remove the other instance if you're not using it
[03-Jun-2009 15:19:44] <rmatte> zenoss looks horrible in IE8 unless you turn on IE7 compatibility mode
[03-Jun-2009 15:19:46] <rmatte> then it looks fine
[03-Jun-2009 15:20:10] <YC> I'm fine with IE8 not working.
[03-Jun-2009 15:20:18] <rmatte> firefox is where it's at
[03-Jun-2009 15:20:22] <YC> It'd be nice if I could figure out why my locations are unhappy though.
[03-Jun-2009 15:20:26] <YC> Chome 2 is very nice also..
[03-Jun-2009 15:20:28] <rmatte> unhappy?
[03-Jun-2009 15:20:50] <mrayzenoss> YC: do you have multiple Locations?
[03-Jun-2009 15:20:56] <rmatte> once chrome gets more robust linux support, then I'll start using it
[03-Jun-2009 15:21:11] <YC> Yep, do you want me to list them all?
[03-Jun-2009 15:21:17] <YC> (Only 4 or so)
[03-Jun-2009 15:21:22] <mrayzenoss> well, do any of them work?
[03-Jun-2009 15:21:29] <YC> No, none.
[03-Jun-2009 15:21:48] <rmatte> did you go under each location, click the little edit link, and specify the location string?
[03-Jun-2009 15:22:06] <rmatte> and did you then click on the map tab for each location and make sure it shows up?
[03-Jun-2009 15:23:48] <YC> That did hte trick. Apparently I'm a moron, sorry.
[03-Jun-2009 15:24:03] <mrayzenoss> Zenoss isn't completely intuitive, no worries
[03-Jun-2009 15:24:12] <YC> I was using the location name as the address.
[03-Jun-2009 15:24:19] <YC> Hey, it's better than the others I've used.
[03-Jun-2009 15:25:01] <rmatte> zenoss is nice but still has it's little problems, a few more years and it'll probably be much more powerful
[03-Jun-2009 15:25:20] <mrayzenoss> well, next month marks 1 year of having QA
[03-Jun-2009 15:26:26] <YC> QA's far less fun than writing new features, everyone knows this.
[03-Jun-2009 15:28:48] <rmatte> I'd like to see this get fixed at some point: http://dev.zenoss.com/trac/ticket/5071
[03-Jun-2009 15:29:09] <rmatte> took upwards of like 15-20 mins for me to load the zProperties pages on some servers that I was setting up today
[03-Jun-2009 15:29:15] <mrayzenoss> yeah, that's a beast of a ZenPack
[03-Jun-2009 15:29:29] <rmatte> there's something that Zenoss is doing though that's causing that lag
[03-Jun-2009 15:29:38] <rmatte> there must be a way around it
[03-Jun-2009 15:30:10] <rmatte> it was probably overlooked initially since most Zenpacks were tiny
[03-Jun-2009 15:30:32] <rocket> I would guess it a broken cache somewhere ...
[03-Jun-2009 15:30:53] <rocket> something isnt using the caching structure .. (without looking)
[03-Jun-2009 15:31:22] <mrayzenoss> supposedly the 2.11.2 upgrade is speeding up a few things
[03-Jun-2009 15:32:10] <rmatte> did anyone ever write a patch for that Linux SSH issue?
[03-Jun-2009 15:32:21] <rocket> rmatte: I am working it
[03-Jun-2009 15:32:26] <rmatte> rocket: nice
[03-Jun-2009 15:32:28] <rocket> rmatte: give me a sec
[03-Jun-2009 15:33:29] <kdub> what about trying the *.bin file?
[03-Jun-2009 15:34:15] <mrayzenoss> kdub: the only practical difference would be that it doesn't register with apt
[03-Jun-2009 15:34:32] <kdub> and i dont particularly care about that, ill give it a try
[03-Jun-2009 15:34:58] <rmatte> doesn't need to register with apt hehe, /usr/local/zenoss/uninstall
[03-Jun-2009 15:34:59] <rmatte> hehe
[03-Jun-2009 15:35:08] <rmatte> works nicely
[03-Jun-2009 15:35:43] <AngryParsley> $ZENHOME is supposed to be /usr/local/zenoss, right? and /usr/local/zenoss/zenoss is for the web app part of zenoss, correct?
[03-Jun-2009 15:36:00] <rmatte> nope
[03-Jun-2009 15:36:03] <AngryParsley> nope?
[03-Jun-2009 15:36:25] <rmatte> zenoss@gen01:/home/rmatte$ echo $ZENHOME
[03-Jun-2009 15:36:31] <rmatte> /usr/local/zenoss/zenoss
[03-Jun-2009 15:37:07] <AngryParsley> oh, but the check_http/tcp/etc scripts are all i/ /usr/local/zenoss/common/libexec, and httpmonitor wants to execute $ZENHOME/bin/check_http
[03-Jun-2009 15:37:34] <rocket> rmatte: patch applied to ticket 5407
[03-Jun-2009 15:37:55] <rocket> sorry I got delayed trying to win a contest
[03-Jun-2009 15:38:16] <rocket> rmatte: it really is a stupidly simple patch though ..
[03-Jun-2009 15:38:25] <AngryParsley> am I supposed to copy check_* to /usr/local/zenoss/zenoss/bin/ ?
[03-Jun-2009 15:38:39] <AngryParsley> because that seems kind of hack-ish
[03-Jun-2009 15:38:53] <mrayzenoss> no
[03-Jun-2009 15:39:14] <mrayzenoss> AngryParsley: see about clearing out zCommandPath on that device
[03-Jun-2009 15:39:25] <mrayzenoss> and then fix the path in the Datasource
[03-Jun-2009 15:40:52] <AngryParsley> ah, zCommandPath was set to /opt/libexec/zenoss
[03-Jun-2009 15:41:29] <AngryParsley> but the data source is an HttpMonitor. I don't see a path to set, just the regular hostname/port/ssl options
[03-Jun-2009 15:42:08] <mrayzenoss> after you cleared out that zProperty, go to the datasource and try hitting "test"
[03-Jun-2009 15:42:32] <mrayzenoss> see what it reports
[03-Jun-2009 15:45:24] <AngryParsley> I don't see a test option. this isn't a COMMAND data source type. it's an HttpMonitor data source type
[03-Jun-2009 15:45:25] <rmatte> is it normal for all devices to report as unpingable during an auto-discovery?
[03-Jun-2009 15:46:09] <AngryParsley> if I look in $ZENHOME/Products/HttpMonitor/datasources/ and edit HttpMonitorDataSource.py, I can see in the source it's adding $ZENHOME to the front of the command it's executing
[03-Jun-2009 15:46:16] <AngryParsley> cmd = '$ZENHOME/libexec/' + \
[03-Jun-2009 15:46:16] <AngryParsley> RRDDataSource.RRDDataSource.getCommand(self, context, cmd)
[03-Jun-2009 15:47:43] <AngryParsley> so that's where I got the idea that maybe $ZENHOME was set incorrectly
[03-Jun-2009 15:47:57] <AngryParsley> but it's not, so I'm guessing I'm doing something else wrong
[03-Jun-2009 15:48:32] <gemineye> Does Zenoss recommend not installing core in a VM or should that work fine?
[03-Jun-2009 15:49:37] <chudler> anyone know offhand where rrdtoolmodule.so comes from in $ZENHOME/lib/python/
[03-Jun-2009 15:50:03] <chudler> I can't find a source for it. I guess I will download zenoss source and find out
[03-Jun-2009 15:50:47] <AngryParsley> chudler: probably http://oss.oetiker.ch/rrdtool/
[03-Jun-2009 15:55:24] <chudler> yep, your right. at least in the 2.4.1 source it is in externallibs/rrdtool-1.3.5[/bindings/python/rrdtoolmodule.c], besides a small patch apparently written by Zenoss.
[03-Jun-2009 15:57:13] <chudler> I hoped this was true, since it means I could use graphv. Unfortunately, I cannot tell if python bindings are supported with graphv. The docs say ruby and perl is.
[03-Jun-2009 15:57:50] <AngryParsley> the command it's trying to run is $ZENHOME/libexec/check_http blah blah blah blah
[03-Jun-2009 15:57:55] <rmatte> gemineye: I have tons of instances of core running in different VMs, works fine
[03-Jun-2009 15:58:00] <AngryParsley> if I run the same command as zenoss on that machine, I get
[03-Jun-2009 15:58:12] <AngryParsley> bash: /usr/local/zenoss/zenoss/libexec/check_http: No such file or directory
[03-Jun-2009 15:58:26] <mrayzenoss> gemineye: lots of people use VMs
[03-Jun-2009 15:58:31] <AngryParsley> since check_http is in /usr/local/zenoss/libexec
[03-Jun-2009 15:58:46] <AngryParsley> and $ZENHOME is /usr/local/zenoss/zenoss
[03-Jun-2009 15:59:12] <mrayzenoss> AngryParsley: did you set $ZENHOME to that? That seems weird
[03-Jun-2009 15:59:49] <AngryParsley> rmatte says that is the correct setting for $ZENHOME
[03-Jun-2009 16:03:21] <gemineye> ok, thanks.
[03-Jun-2009 16:08:48] <AngryParsley> http://dev.zenoss.org/trac/ticket/3201
[03-Jun-2009 16:08:51] <AngryParsley> ah, it's a bug in zenoss
[03-Jun-2009 16:47:02] <YC> Is it possible to manually draw links on the google maps?
[04-Jun-2009 00:00:42] [disconnected at Thu Jun 4 00:00:42 2009]
[04-Jun-2009 00:00:42] [connected at Thu Jun 4 00:00:42 2009]
[04-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[04-Jun-2009 01:39:13] <torocatala> hi
[04-Jun-2009 07:27:34] <mr_claus_> hi, i monitor postfix but zenoss always disconnects before the greeting is finished, that fill my logs is there a way to do the greeting with zenoss?
[04-Jun-2009 07:56:54] <tehhobbit> is there a good way to make a perf graph for switches with the total pps and mbit ?
[04-Jun-2009 08:04:08] <coofamani> define total
[04-Jun-2009 08:06:47] <tehhobbit> ?
[04-Jun-2009 08:14:49] <coofamani> when you say total pps and mbit, do you mean total per interface or total for a switch
[04-Jun-2009 08:28:41] <tehhobbit> for a switch
[04-Jun-2009 08:36:43] <coofamani> if its just for a few switches you could do an aggregate multigraph report
[04-Jun-2009 09:28:41] <kdub> what is the difference between a 'system' and a 'group'?
[04-Jun-2009 09:32:09] <mrayzenoss> they're essentially the same thing in the code
[04-Jun-2009 09:32:21] <mrayzenoss> they're just logical ways of organizing your devices
[04-Jun-2009 09:32:27] <rocket> morning matt
[04-Jun-2009 09:32:41] <mrayzenoss> if you're sorting by hardware or OS or something, you'd probably use a System
[04-Jun-2009 09:33:07] <mrayzenoss> if you're sorting by use or function (i.e. "QA" or "Accounting") you'd use a Group
[04-Jun-2009 09:33:25] <mrayzenoss> and devices can belong to multiple Groups and multiple Systems
[04-Jun-2009 09:33:45] <mrayzenoss> good morning rocket, I'm checking in your patch to trunk
[04-Jun-2009 09:35:52] <sergeymasushko> I have an issue. I've added ifInOctets (1.3.6.1.2.1.2.2.1.10.1) and ifOutOctets (1.3.6.1.2.1.2.2.1.16.1) as datasources in the performance template. But graphs are not shown. However when I do test on the datasource edit page I get successful result... where can be the issue?
[04-Jun-2009 09:36:00] <rocket> mrayzenoss: sounds good
[04-Jun-2009 09:36:06] <rocket> mrayzenoss: working on the docs for you
[04-Jun-2009 09:36:07] <sergeymasushko> morning guys
[04-Jun-2009 09:36:19] <rocket> mrayzenoss: and there shouldnt have been a dep on mysql for the aix one
[04-Jun-2009 09:37:32] <rmatte> ummm weird, in 2.4.1, I don't see the actual threshold value on graphs
[04-Jun-2009 09:37:52] <rmatte> in the legend I just see the threshold name instead of something like CPU > 90
[04-Jun-2009 09:42:20] <kdub> thanks mrayzenoss
[04-Jun-2009 09:43:20] <rmatte> oh well, at least it draws the threshold line
[04-Jun-2009 09:44:17] <mrayzenoss> rmatte: debating your ZenPack slowness ticket right now
[04-Jun-2009 09:44:28] <mrayzenoss> rmatte: might be a blocker
[04-Jun-2009 09:44:53] <mrayzenoss> apparently it showed up on an Enterprise upgrade and took 10 hours
[04-Jun-2009 09:50:19] <rmatte> 10 hours?
[04-Jun-2009 09:50:27] <rmatte> I've never had it go that long
[04-Jun-2009 09:50:40] <rmatte> I've never seen it go more than like 15-20 minutes
[04-Jun-2009 09:52:09] <rocket> mrayzenoss: all the thresholds will need to be 2.4 I believe
[04-Jun-2009 09:52:48] <rmatte> what's a "blocker" ?
[04-Jun-2009 09:53:20] <rocket> blocker = bug that forces a release to fix it before the next version can be released
[04-Jun-2009 09:53:28] <rocket> eg the bug blocks the software release
[04-Jun-2009 09:53:54] <rmatte> ah
[04-Jun-2009 09:53:59] <rmatte> that's what I figured
[04-Jun-2009 09:54:09] <rmatte> yeh, it's a pretty crappy bug
[04-Jun-2009 09:54:15] <rmatte> I'd love to see it fixed hehe
[04-Jun-2009 09:55:17] <rmatte> oh, it's not only based on a single large Zenpack but on total amount
[04-Jun-2009 09:55:18] <rmatte> ouch
[04-Jun-2009 09:55:27] <mrayzenoss> rocket: yeah, all thresholds are marked for 2.4
[04-Jun-2009 09:55:40] <mrayzenoss> rmatte: correct, blockers all must be fixed before a release
[04-Jun-2009 09:55:55] <rmatte> cool
[04-Jun-2009 09:56:36] <mrayzenoss> all the high's have to be accounted for to do a release, either fixed, release-noted or moved out of the release
[04-Jun-2009 09:56:57] <rmatte> cool
[04-Jun-2009 09:57:11] <mrayzenoss> I'm adding the post King Crab release to Trac right now
[04-Jun-2009 09:57:21] <mrayzenoss> "Stone Crab", ETA Spring 2010
[04-Jun-2009 09:57:29] <mrayzenoss> that just sounds crazy
[04-Jun-2009 09:58:50] <cgibbons> hmmmm
[04-Jun-2009 10:08:47] <coofamani> Softshell Crab
[04-Jun-2009 10:09:12] <rmatte> Stone Crab, nice
[04-Jun-2009 10:09:28] <rmatte> maybe the next one can be Kryptonite Crab
[04-Jun-2009 10:14:50] <rmatte> waiting for a zProperties page to load right now actually...
[04-Jun-2009 10:30:22] <rmatte> ok, what do I have to do to have Zenoss auto-detect win services on a server via WMI?
[04-Jun-2009 10:30:51] <rmatte> I have wmiMonitorIgnore set to false, I have the correct username and password specified, and I have the WinServices collector plugin applied to the devices
[04-Jun-2009 10:30:57] <rmatte> I've pushed changed to the collector
[04-Jun-2009 10:31:04] <rmatte> when I model them, no Win Services show up
[04-Jun-2009 10:31:18] <rocket> mrayzenoss: sent you another email
[04-Jun-2009 10:32:56] <rocket> mrayzenoss: hopefully this aix file will not include the mysql dep
[04-Jun-2009 10:40:33] <rmatte> ah, finally got it working
[04-Jun-2009 10:48:11] <sergeymasushko> can somebody clarify mechanism of aging events...
[04-Jun-2009 10:59:08] <rmatte> it'd be nice if you could actually age critical events
[04-Jun-2009 10:59:14] <rmatte> you can age everything else except for those
[04-Jun-2009 11:01:11] <sergeymasushko> just interesting... I have events with severity 'warning' (diskspace utilization) these events updated 13 times per hour... but each hour these events aged. Is it correct behavior?
[04-Jun-2009 11:02:11] <sergeymasushko> Event Aging Threshold (hours) set to 1 and Don't Age This Severity and Above set to 'Error'
[04-Jun-2009 11:06:24] <sergeymasushko> According to the manual (http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch07s01.html) zenoss should age events only if the event counter is updated during 'Event Aging Threshold'. Am I correct?
[04-Jun-2009 11:08:19] <mrayzenoss> The aging gets reset every time the counter gets updated
[04-Jun-2009 11:10:24] <sergeymasushko> mrayzenoss: Matt, so why it clears events every hour when event counter updated 12 times per hour.
[04-Jun-2009 11:16:04] <sergeymasushko> here is an example from the event history http://pastebin.com/m76c75c7d
[04-Jun-2009 11:18:11] scfh_ is now known as scfh
[04-Jun-2009 11:19:39] <mrayzenoss> those are separate events
[04-Jun-2009 11:19:52] <mrayzenoss> or, they're not getting consolidated even though they look like the same thing
[04-Jun-2009 11:25:05] <sergeymasushko> why do you think that those events are separate?
[04-Jun-2009 11:32:19] <sergeymasushko> I've re checked events there are the same device, the same component, the same value...
[04-Jun-2009 11:35:04] <sergeymasushko> perr0: hi, have you experienced such issue on you freebsd installation?
[04-Jun-2009 11:36:57] <perr0> is that in your pastebin?
[04-Jun-2009 11:37:10] <perr0> sergeymasushko: ^
[04-Jun-2009 11:37:55] <sergeymasushko> yes
[04-Jun-2009 11:39:44] <perr0> that happens when you mount a USB? That happens when I mount .isos'
[04-Jun-2009 11:40:02] <perr0> yep
[04-Jun-2009 11:40:54] <perr0> so when the developers need to install a package I have a script that mounts 7 isos and it drives me creazy cause they never unmount them and I get those errors
[04-Jun-2009 11:41:07] <perr0> or warnings
[04-Jun-2009 11:49:38] <rmatte> tell them that they get an hour before it auto-unmounts
[04-Jun-2009 11:49:40] <rmatte> >:)
[04-Jun-2009 11:52:04] <sergeymasushko> hm... but that device is already mounted there for a long time without remount.
[04-Jun-2009 11:53:39] <perr0> rmatte: does that happen to you too?
[04-Jun-2009 11:53:57] <perr0> I haven't had much time to look into these past weeks
[04-Jun-2009 11:54:05] <perr0> it*
[04-Jun-2009 12:01:08] <sergeymasushko> ok thanks for the point... I'll try to dig more into it.
[04-Jun-2009 12:06:11] <Mark09> I see on the compare page that Pro has "WMI based modeling of disk, CPU, h/w & s/w inventory" - does that mean no nsclient on windows ?
[04-Jun-2009 12:06:35] <Mark09> do you have to install another client on the windows boxes or can it talk wmi?
[04-Jun-2009 12:30:27] <rmatte> Mark09: it can talk WMI
[04-Jun-2009 12:30:48] <rmatte> it can gather all of the info that Core can (with SNMP) via WMI
[04-Jun-2009 17:23:39] <gemineye> Anyone here?
[04-Jun-2009 17:25:35] <mrayzenoss> it's been quiet today
[04-Jun-2009 17:29:35] <gemineye> But you're here, so good
[04-Jun-2009 17:29:39] <gemineye> I have two questions for ya
[04-Jun-2009 17:30:30] <gemineye> Does Zenoss support dynamic polling frequency, either Enterprise or Core?
[04-Jun-2009 17:31:44] <gemineye> Let's say a device is polled every 5 minutes. If it detects an outage, I have my alerts setup to not notify me unless an event occurs 3 times in a row. Can Zenoss be smart and increase the polling frequency when it detects the outages so instead of waiting 15 minutes for an alert, it'll check every 1 minute when it sees the outage?
[04-Jun-2009 17:32:08] <gemineye> Possibly done through transforms?
[04-Jun-2009 17:33:55] <gemineye> mrayzenoss: WAKE UP!
[04-Jun-2009 17:34:32] <mrayzenoss> nope
[04-Jun-2009 17:35:02] <mrayzenoss> talking to a dev about it
[04-Jun-2009 17:35:09] <gemineye> ok
[04-Jun-2009 17:37:16] <gemineye> I have a second question when you're ready for it
[04-Jun-2009 17:37:33] <mrayzenoss> yeah, he said we could write that, but haven't been asked to
[04-Jun-2009 17:37:41] <gemineye> write it
[04-Jun-2009 17:37:44] <mrayzenoss> and then it could be triggered with a transform
[04-Jun-2009 17:37:47] <mrayzenoss> heh
[04-Jun-2009 17:39:02] <mrayzenoss> the priortized backlog is long, but L2 awareness has been making a big stink today
[04-Jun-2009 17:39:25] <gemineye> Is that something that could be easy to activate either as a feature of Core or Enterprise, or would it be a custom enhancement on a per user basis?
[04-Jun-2009 17:39:27] <mrayzenoss> next question
[04-Jun-2009 17:39:32] <gemineye> That's not the next question
[04-Jun-2009 17:39:37] <gemineye> What's L2?
[04-Jun-2009 17:39:49] <gemineye> That isn't the second question either
[04-Jun-2009 17:39:49] <mrayzenoss> it would not be an easy change
[04-Jun-2009 17:40:01] <mrayzenoss> since there are so many different collectors
[04-Jun-2009 17:40:20] <mrayzenoss> and how do you decide which ones to accelerate on error
[04-Jun-2009 17:40:53] <mrayzenoss> right now they're not very smart
[04-Jun-2009 17:40:56] <gemineye> Basically, Nimsoft bought Cittio which is what one of our departments has been using for monitoring and I'm trying to get them to move to Zenoss. These are features they currently enjoy with WatchTower and they'd like to see available in Zenoss.
[04-Jun-2009 17:42:09] <gemineye> There could be a zProperty for each device that enables the dynmic polling
[04-Jun-2009 17:42:19] <mrayzenoss> King Crab is going to have a lot of improvements around the durability and performance for collectors
[04-Jun-2009 17:42:31] <gemineye> What version is KC?
[04-Jun-2009 17:42:35] <mrayzenoss> next
[04-Jun-2009 17:42:42] <gemineye> #?
[04-Jun-2009 17:42:47] <mrayzenoss> it's possible collection cycles may be moving off of the collector and onto devices
[04-Jun-2009 17:42:54] <gemineye> That would be nice
[04-Jun-2009 17:43:06] <mrayzenoss> we call it King Crab since marketing gets to pick version numbers
[04-Jun-2009 17:43:12] <mrayzenoss> 2.5 if you like
[04-Jun-2009 17:43:27] <gemineye> I do like, as opposed to 2.4.2 or 3.0
[04-Jun-2009 17:43:32] <mrayzenoss> correct
[04-Jun-2009 17:43:53] <mrayzenoss> maintenance releases like 2.4.2 aren't going to get substantial new features, they're supposed to be bug fixes
[04-Jun-2009 17:43:59] <gemineye> But certainly not 2.4.1.0.0.0.0.0.1
[04-Jun-2009 17:44:10] <mrayzenoss> well, that's what zenpatch is for
[04-Jun-2009 17:44:26] <mrayzenoss> 2.4.2 will roll-up the zenpatchable things that are against the 2.4.x branch
[04-Jun-2009 17:44:31] <gemineye> When will Zenoss release numbers use imaginary numbers?
[04-Jun-2009 17:44:43] <gemineye> In my dreams?
[04-Jun-2009 17:44:45] <mrayzenoss> you mean like 2.4.kingcrab?
[04-Jun-2009 17:45:04] <mrayzenoss> and BTW, I added the post-King Crab release to Trac today... "Stone Crab"
[04-Jun-2009 17:45:06] <gemineye> no, like 2.5i
[04-Jun-2009 17:45:22] <mrayzenoss> next question
[04-Jun-2009 17:45:52] <mrayzenoss> and King Crab is tentatively scheduled for October
[04-Jun-2009 17:46:17] <gemineye> Will King Crab include report enhancements?
[04-Jun-2009 17:46:28] <mrayzenoss> there are some changes that have already gone into trunk I think
[04-Jun-2009 17:46:45] <mrayzenoss> like making straight python calls in reports
[04-Jun-2009 17:49:58] <mrayzenoss> I gotta get going, email me if you have any other questions
[04-Jun-2009 22:56:28] <zenchild> Anyone know where to find a list of xml-rpc methods available from zenoss?
[05-Jun-2009 00:00:42] [disconnected at Fri Jun 5 00:00:42 2009]
[05-Jun-2009 00:00:42] [connected at Fri Jun 5 00:00:42 2009]
[05-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[05-Jun-2009 03:41:53] <Dieterbe> I don't understand the cpu usage graph ( http://www.kangaroot.be/images/constant-cpu-usage-weird.png ). as you can see the load varies a lot, as does the mysql commands. however the cpu usage graph is almost constant o_O what's going on?
[05-Jun-2009 04:31:47] <rhettlap> load average and cpu usage are two separate things though are commonly regarding as being the same or mostly similar
[05-Jun-2009 04:31:54] <Dieterbe> indeed
[05-Jun-2009 04:35:56] <rhettlap> can you verify that the average cpu utilization is indeed variable and not static as it appears in your graph?
[05-Jun-2009 04:39:05] <Dieterbe> yes i have top open, it fluctuates
[05-Jun-2009 04:40:34] <rhettlap> well you need to know what the 5 minute average is of cpu usage
[05-Jun-2009 04:41:34] <rhettlap> and then compare that to the snmp output
[05-Jun-2009 04:42:14] <rhettlap> either buy using rrdtool fetch to the datasource or by walking or getting the output from the appropriate OID
[05-Jun-2009 04:43:03] <rhettlap> sorry to bail out on you but i have to go get some rest
[05-Jun-2009 04:43:15] <Dieterbe> rhettlap: that's ok. i'll try manual snmpwalking
[05-Jun-2009 04:43:19] <Dieterbe> i just need to find out the OID
[05-Jun-2009 04:43:38] <rhettlap> you can go into the template and see the oid
[05-Jun-2009 04:45:14] <rhettlap> snmpwalk -v 2c -c community device 1.3.6.1.4.1.2021.11
[05-Jun-2009 04:45:17] <rhettlap> that should do it for you
[05-Jun-2009 04:45:59] <rhettlap> or if you want to see the raw oid instead of the translation through the -On flag in there
[05-Jun-2009 04:46:03] <rhettlap> good luck
[05-Jun-2009 04:46:07] <Dieterbe> thanks man
[05-Jun-2009 04:46:20] <Dieterbe> and where exactly did you find this? I'm looking at /Devices /Server /Linux /Templates /Device /CPU Utilization, but don't really see it
[05-Jun-2009 04:46:48] <rhettlap> also to look at the data source you can use 'rrdtool fetch pathtodatasource AVERAGE -r 300 -s -1h'
[05-Jun-2009 04:46:55] <rhettlap> that will show the results for the past hour
[05-Jun-2009 04:47:42] <rhettlap> it should be under the 'Device' template
[05-Jun-2009 04:47:54] <Dieterbe> ah yeah, i just found it
[05-Jun-2009 04:47:58] <rhettlap> then it will be listed under data sources
[05-Jun-2009 04:47:58] <rhettlap> cool
[05-Jun-2009 04:49:58] <rhettlap> well i am off for good now. good luck again.
[05-Jun-2009 04:50:04] <Dieterbe> thanks
[05-Jun-2009 08:24:04] <cgibbons> hurm
[05-Jun-2009 08:24:29] <rocket> Hello
[05-Jun-2009 08:34:52] <etank> is it possible to change the hostname that is used when sending alerts?
[05-Jun-2009 08:35:06] <etank> i want to change whatever it is that is used in "<a href="%(eventUrl)s">Event Detail</a>"
[05-Jun-2009 08:35:41] <etank> the hostname of the server is foo.com but i have a dns alias that i use for the zenoss install that is different
[05-Jun-2009 08:35:53] <etank> i want it to use zenoss.foo.com
[05-Jun-2009 09:00:31] <perr0> etank: what?
[05-Jun-2009 09:02:34] <etank> perr0: in the properties of an alerting rule (i.e /ZenUsers /etank /Alerting Rules /Event Log Errors ) you see things like <a href="%(eventUrl)s">Event Detail</a> for the body of the message
[05-Jun-2009 09:02:52] <etank> to give you a link to acknowledge, view, etc the event
[05-Jun-2009 09:03:04] <etank> my servers hostname is rethink.lms.lexmark.com
[05-Jun-2009 09:03:10] <perr0> I am just confused as to your question is changing the hostname
[05-Jun-2009 09:03:31] <etank> and i have a cname set to zenoss.lms.lexmark.com for the same box for zenoss use
[05-Jun-2009 09:03:35] <perr0> in*
[05-Jun-2009 09:03:37] <perr0> ok
[05-Jun-2009 09:03:54] <etank> i would like the urls in the alerts to be zenoss.lms.lexmark.com instead of the hostname of the server
[05-Jun-2009 09:04:26] <perr0> gotcha
[05-Jun-2009 09:05:02] <etank> im just not sure where the variable is read from. should be easy enough to change
[05-Jun-2009 09:05:07] <rmatte> any particular reason why 2.4.1 takes more than twice as long to start than 2.3.3?
[05-Jun-2009 09:05:51] <perr0> right. Have you tried setting the host name of zenoss.xxx.lemark and restarting zenoss?
[05-Jun-2009 09:06:01] <perr0> in your server?
[05-Jun-2009 09:06:45] <chudler> I think maybe $ZENHOME/etc/zenoactions.conf has an option zopeurl
[05-Jun-2009 09:07:02] <chudler> set it like "zopeurl https://example.com"
[05-Jun-2009 09:07:16] <chudler> I mean "zenactions.conf"
[05-Jun-2009 09:07:52] <etank> perr0: yeah that would be one way to do it
[05-Jun-2009 09:09:19] <etank> chudler: my $ZENHOME/etc/zenoactions.conf is blank
[05-Jun-2009 09:10:38] <etank> ok i just set zopeurl http://zenoss.lms.lexmark.com in my zenactions.conf file
[05-Jun-2009 09:12:10] <chudler> a restart is likely to be necessary
[05-Jun-2009 09:12:23] <etank> seems i could have changed it at http://zenoss.lms.lexmark.com:8080/zport/About/showDaemonXMLConfig?daemon=zenactions
[05-Jun-2009 09:12:46] <perr0> let us know did it work?
[05-Jun-2009 09:12:53] <perr0> either way
[05-Jun-2009 09:13:56] <chudler> hm cool, I never edit the configs this way. I guess I never noticed it was even there.
[05-Jun-2009 09:15:13] <etank> chudler: i had not seen it either
[05-Jun-2009 09:15:23] <etank> i reads the file without a restart though
[05-Jun-2009 09:15:29] <etank> and it works like a champ
[05-Jun-2009 09:15:45] <etank> i did set it in the .conf though not through the web
[05-Jun-2009 09:15:51] <etank> thanks perr0 and chudler
[05-Jun-2009 09:16:07] <chudler> cool. you are welcome!
[05-Jun-2009 09:16:58] <perr0> the zenoactions.conf did it huh? Cool now we know something new
[05-Jun-2009 09:17:06] <etank> now i just need to learn how to suppress events that i dont care about at all
[05-Jun-2009 09:17:11] <etank> like DCOM junk
[05-Jun-2009 09:46:25] <kdub> is there a place where i can enter the database password? i'm trying zenrestore, and it keeps telling me the password login for mysql is rejected
[05-Jun-2009 09:49:59] <kdub> nvm, i might have just figured it out...
[05-Jun-2009 10:16:51] <venturaville> odd question for someone who might have a better way of doing this:
[05-Jun-2009 10:17:29] <venturaville> I have a device class where I need to consolidate the performance numbers from a bunch of other devices that are 'somehow' linked to it.
[05-Jun-2009 10:18:14] <venturaville> i.e. I could do this in theory by making a command that calls back into zenoss to fetch the numbers, but this seems rather clunky
[05-Jun-2009 10:18:32] <venturaville> I was hoping there was a cleaner way to do this
[05-Jun-2009 10:19:58] <zenChild> Would a multi-graph report work?
[05-Jun-2009 10:21:52] <venturaville> not in this case because the devices behind the consolidate device can drop out or be added to at will
[05-Jun-2009 10:22:07] <venturaville> which would mean the history for the consolidation would change with it
[05-Jun-2009 10:22:34] <zenChild> hmmm... that is an interesting problem.
[05-Jun-2009 10:24:02] <cgibbons> hurm
[05-Jun-2009 10:24:04] <venturaville> I looked at that as a possible solution when I first got into zenoss, but it multi-graph is still a pretty limited tool for doing these kinds of things
[05-Jun-2009 10:24:53] <zenChild> I wonder if you could use the REST API to fetch that data out and consolidate it.
[05-Jun-2009 10:25:15] <venturaville> that is the way I am looking at now, but that is pretty poor performing and really clunky
[05-Jun-2009 10:25:32] <venturaville> I was hoping for more of a solution that is within the zenoss context (like the modeler)
[05-Jun-2009 10:25:52] <venturaville> I'm looking to do this on a rather large number of devices
[05-Jun-2009 10:25:58] <zenChild> Does anyone know if there is a documented API for REST and XML-RPC?
[05-Jun-2009 10:26:07] <venturaville> there is
[05-Jun-2009 10:26:25] <chudler> a custom graph plugin can do it. Depends on the amount of work you want to do, of course.
[05-Jun-2009 10:26:26] <zenChild> I've looked at the zenoss API, but that doesn't seem to correspond 1-for-1 to the xml-rpc calls
[05-Jun-2009 10:27:45] <venturaville> that is because some of the things in the API return python objects, which don't translate well into xml-rpc
[05-Jun-2009 10:28:07] <venturaville> learn to use zendmd and then figure out how to do things without using the objects
[05-Jun-2009 10:28:22] <venturaville> and then for the really tricky things you can add pieces of code in zope to do those things
[05-Jun-2009 10:28:59] <venturaville> example: http://www.zenoss.com/community/docs/howtos/get-a-list-of-device-names/
[05-Jun-2009 10:30:26] <zenChild> Its returning objects is probably the reason I'm getting marshalling errors I presume.
[05-Jun-2009 10:30:48] <venturaville> yep
[05-Jun-2009 10:31:23] <zenChild> Thanks for your input.
[05-Jun-2009 10:33:47] <venturaville> chudler: is there any documentation on writing custom graph plugins anywher?
[05-Jun-2009 10:49:48] <rmatte> by custom graph plugin you mean a script that can be used to generate datapoints?
[05-Jun-2009 10:50:46] <venturaville> essentially, though I have a suspicion it is probably not what I am ultimately after
[05-Jun-2009 10:51:08] <venturaville> I'm hoping for more of a device centric method of doing this
[05-Jun-2009 11:15:24] <exes> whats better about zenoss than nagios or hyperic?
[05-Jun-2009 11:32:31] <chudler> exes: I can't speak for everyone, but hyperic's agent turned us off it. I think they have been bought, and now days are apparently evolving into a EE thang.
[05-Jun-2009 11:38:17] <mrayzenoss> exes: anecdotally I've been told that Nagios can be a pain to manage large numbers of devices
[05-Jun-2009 11:38:26] <mrayzenoss> and any Nagios plugins can be used by Zenoss
[05-Jun-2009 11:38:51] <mrayzenoss> Zenoss adds performance monitoring (Nagios users would use Cactii for that)
[05-Jun-2009 11:39:00] <mrayzenoss> JMX & WMI
[05-Jun-2009 11:39:07] <mrayzenoss> the network map and google map
[05-Jun-2009 11:39:19] <mrayzenoss> event and alert schedules, maintenance windows
[05-Jun-2009 11:40:03] <mrayzenoss> there's more, but is there a particular feature you're looking for?
[05-Jun-2009 11:56:37] <venturaville> the API and extendability are killer
[05-Jun-2009 12:30:09] <exes> mrayzenoss: yes... I'll have to get back to you, thanks for your response
[05-Jun-2009 12:31:07] <venturaville> mrayzenoss: I was asking this question earlier before I think you were on
[05-Jun-2009 12:31:19] <venturaville> (11:17:28 AM) venturaville: I have a device class where I need to consolidate the performance numbers from a bunch of other devices that are 'somehow' linked to it.
[05-Jun-2009 12:31:19] <venturaville> (11:18:14 AM) venturaville: i.e. I could do this in theory by making a command that calls back into zenoss to fetch the numbers, but this seems rather clunky
[05-Jun-2009 12:31:19] <venturaville> (11:18:32 AM) venturaville: I was hoping there was a cleaner way to do this
[05-Jun-2009 12:32:16] <mrayzenoss> hmmm... I'll ask around
[05-Jun-2009 12:32:19] <venturaville> and multi-graph won't work since I need to be able to handle the other devices in and out without mangling the RRD history.... also I really need to be able to tag the consolidated device with the same tags as I do the devices (i.e. /Groups/*)
[05-Jun-2009 12:32:21] <mrayzenoss> some sort of meta device?
[05-Jun-2009 12:32:30] <venturaville> nailed it in one
[05-Jun-2009 12:32:58] <mrayzenoss> the VMware Devices ZenPack kinda has that
[05-Jun-2009 12:33:07] <mrayzenoss> where there's a Guests tab
[05-Jun-2009 12:33:13] <mrayzenoss> but you want it in perf?
[05-Jun-2009 12:33:20] <venturaville> yeah I've looked at that, however it isn't reentrant
[05-Jun-2009 12:33:42] <venturaville> i.e. the performance data is collected at the device level not at the consolidated device level
[05-Jun-2009 12:33:45] <mrayzenoss> I'll ask the devs when they get back from lunch
[05-Jun-2009 12:33:50] <venturaville> thanks
[05-Jun-2009 12:34:18] <venturaville> the VMWare plugin could benefit from something like that too................
[05-Jun-2009 12:53:23] <HKhan> 18:52:57 up 1:21, 2 users, load average: 29.97, 22.83, 11.17 << I think i'm going to have to get a new box to run zenoss on >.> Lagging like hell D:
[05-Jun-2009 12:57:51] <pinball_> 5:57pm up 905 day(s), 18:52, 1 user, load average: 0.11, 0.09, 0.09
[05-Jun-2009 12:58:00] pinball_ is now known as pinball
[05-Jun-2009 13:01:23] <HKhan>
[05-Jun-2009 13:01:28] <HKhan> wonder why mine is so high :S
[05-Jun-2009 13:01:41] <HKhan> i only have a few monitored machines
[05-Jun-2009 13:08:35] <rhettlap> wow
[05-Jun-2009 13:08:44] <rhettlap> that is some serious business HKhan
[05-Jun-2009 13:08:51] <mrayzenoss> what's the top process?
[05-Jun-2009 13:10:23] <HKhan> Zope is atm, could it be caused if my MySQL box is replying slowly? >.<
[05-Jun-2009 13:11:31] <mrayzenoss> hmmm.., what are the stats on the box?
[05-Jun-2009 13:12:11] <HKhan> what do you mean by stats? hardware spec?
[05-Jun-2009 13:12:18] <mrayzenoss> yeah
[05-Jun-2009 13:12:27] <mrayzenoss> I assume the box is something reasonable?
[05-Jun-2009 13:12:36] <HKhan> 1GHz P3 with 1GB RAM
[05-Jun-2009 13:12:52] <mrayzenoss> hmm... that's kinda pokey, how many devices?
[05-Jun-2009 13:13:05] <mrayzenoss> I do 10 on a G4 1.33ghz Powerbook
[05-Jun-2009 13:13:22] <HKhan> umm around 20 i think.... >.< i think it just fell over under the load, so can't check atm heh
[05-Jun-2009 13:14:33] <venturaville> 1GB RAM is pretty low for zenoss ...period
[05-Jun-2009 13:14:43] <mrayzenoss> yeah, my powerbook has 1.25 gigs
[05-Jun-2009 13:15:00] <mrayzenoss> it sits at about .2 on load
[05-Jun-2009 13:15:23] <HKhan> hmm will see if i can upgrade it then
[05-Jun-2009 13:16:49] <venturaville> check how much swap you are using
[05-Jun-2009 13:17:11] <mrayzenoss> yeah, I never hit my swap
[05-Jun-2009 13:17:53] <venturaville> zenoss does pretty well as long as it doesn't
[05-Jun-2009 13:18:00] <HKhan> ah its using around 400MB swap
[05-Jun-2009 13:18:07] <mrayzenoss> yeah, that's gonna kill you
[05-Jun-2009 13:18:08] <venturaville> I had to do some tuning to get it to work well on my 8 year old computer at home with similar specs
[05-Jun-2009 13:18:25] <venturaville> HKhan ... you might want to see what else is running
[05-Jun-2009 13:18:30] <venturaville> are you trying to run this on your desktop?
[05-Jun-2009 13:18:34] <HKhan> nope
[05-Jun-2009 13:19:12] <HKhan> i think it'll be best to change the box its running on, this is a bit old anyway
[05-Jun-2009 13:19:16] <venturaville> do top
[05-Jun-2009 13:19:20] <venturaville> and then hit 'm'
[05-Jun-2009 13:19:28] <venturaville> err capital M
[05-Jun-2009 13:20:17] <HKhan> its only got zenoss running on the box, its a clean install of RHEL5
[05-Jun-2009 13:20:53] <HKhan> one sec, >.< box is gone down i think
[05-Jun-2009 13:21:24] <venturaville> we run RH5 ... if you do chkconfig --list I can probably tell you what to shut off ....
[05-Jun-2009 13:25:50] <HKhan> i've already disabled most of the services with chkconfig (well everything i don't need heh)
[05-Jun-2009 13:26:44] <HKhan> http://i.like.to.dre.am/services.txt
[05-Jun-2009 13:31:59] * HKhan tries upgrading ram
[05-Jun-2009 13:40:41] <venturaville> HKhan looks pretty clean
[05-Jun-2009 13:43:08] <HKhan> I think i'll change the box, P3 is quite old anyway
[05-Jun-2009 13:51:19] <cgibbons> you still have that 12" powerbook matt?
[05-Jun-2009 13:52:23] <rmatte> I'm running one box on 3GHz processing power with 4GB of memory and it's comfortably monitoring 344 devices
[05-Jun-2009 13:53:28] <rmatte> 53 of which are windows servers
[05-Jun-2009 13:57:43] <chudler> anyone know, offhand, if RenderServer is ever set do /dev/null, especially when doing getRRDValues??
[05-Jun-2009 14:01:48] <chudler> ZenModel.PerformanceConf is some really weird stuff..
[05-Jun-2009 14:03:57] <chudler> oh, here it is :-) ZenRRD.RenderServer summary(): gopts.insert(0, '/dev/null')
[05-Jun-2009 14:04:29] <rmatte> if a graph is showing 0, 200m, 400m, 600m, etc... instead of 1.0, 2.0, 3.0, 4.0, etc..., what can I do to fix it?
[05-Jun-2009 14:04:43] <rmatte> I have a bunch of graphs with identical settings, but some are displaying with the m values
[05-Jun-2009 14:04:51] <rmatte> and there's nothing that I can see that would cause that
[05-Jun-2009 14:45:18] <rocket> Hello everyone
[05-Jun-2009 14:50:21] <chudler> rrdtool python binding cannot handle unicode, it seems. Good thing I won't need it :-)
[05-Jun-2009 14:54:16] <xulfer> Hm, is there anywhere I can find documentation on how to use the core zenpacks, and maybe some of the others? I can't seem to find any solid documentation on it.
[05-Jun-2009 15:08:57] <jb> hi
[05-Jun-2009 15:09:21] <rocket> Hey jb
[05-Jun-2009 15:09:25] <jb> whats up rocket?
[05-Jun-2009 15:09:57] <rocket> not too much .. sent a fixed aix zenpack to matt today... fixed the erroneous requirement to have the mysql zenpack installed
[05-Jun-2009 15:10:10] <jb> ah cool
[05-Jun-2009 15:10:29] <rocket> trying to figure out a way to get link speed out of /proc or /sys on a linux box
[05-Jun-2009 15:10:41] <rocket> not happy with the dmesg method used in the linux zenpack ...
[05-Jun-2009 15:52:50] <jb> theres always ethtool.. although it may or may not be standard on your boxes
[05-Jun-2009 15:56:37] <chudler> probably some part of /sys/class/net/*
[05-Jun-2009 15:59:20] <jb> hm, i dont see it
[05-Jun-2009 16:01:01] <chudler> yeah, it seems to have everything but that.
[05-Jun-2009 16:11:54] <chudler> this made me so curious, I downloaded ethtool source. It seems it either interogates the eeprom of the (supported) cards, or sends them control signals.
[05-Jun-2009 16:43:43] <IoNu|z> anyone know why im getting this error on startup? /dev/hda: open failed: No medium found
[05-Jun-2009 17:07:32] <pinball> no cdrom in the drive
[06-Jun-2009 00:00:42] [disconnected at Sat Jun 6 00:00:42 2009]
[06-Jun-2009 00:00:42] [connected at Sat Jun 6 00:00:42 2009]
[06-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[06-Jun-2009 05:56:35] tehhobbit is now known as fail
[06-Jun-2009 05:57:05] fail is now known as Guest57031
[06-Jun-2009 05:57:18] Guest57031 is now known as tehhobbit
[06-Jun-2009 07:27:19] <raddy> Hello Everybody
[06-Jun-2009 07:28:01] <raddy> I am experiencing problems in Zenoss
[06-Jun-2009 07:28:08] <raddy> Can anybody help me?
[06-Jun-2009 08:45:23] <raddy> Hello Everybody
[06-Jun-2009 08:45:47] <raddy> Zenjobs and Zenactions using 100% CPU
[06-Jun-2009 08:45:55] <raddy> Can anybody help me.
[06-Jun-2009 08:46:14] <raddy> I am running Zenoss 2.4.1
[06-Jun-2009 09:01:57] <raddy> Anybody out there????
[06-Jun-2009 09:26:56] <raddy> Hello Everbody
[06-Jun-2009 09:27:17] <raddy> I am experiencing some major issues in ZenOSS
[06-Jun-2009 09:27:23] <raddy> Can anybody help me?
[06-Jun-2009 14:11:30] <rsFF> does anyone can point me a tutorial
[06-Jun-2009 14:11:38] <rsFF> wheren can i find som info regarding
[06-Jun-2009 14:11:53] <rsFF> using closeProcess in zenoss
[06-Jun-2009 14:11:55] <rsFF> thanks
[06-Jun-2009 20:45:54] <sergeymasushko> is anybody alive?
[07-Jun-2009 00:00:42] [disconnected at Sun Jun 7 00:00:42 2009]
[07-Jun-2009 00:00:42] [connected at Sun Jun 7 00:00:42 2009]
[07-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[07-Jun-2009 01:27:08] <raddy> Hello Everybody
[07-Jun-2009 01:27:18] <raddy> Anybody out there??
[07-Jun-2009 08:13:06] <raddy> Hello
[07-Jun-2009 08:13:14] <raddy> Anybody out there?
[07-Jun-2009 08:13:25] <raddy> I need you people's help.
[07-Jun-2009 08:28:33] <raddy> Anybody out theree?
[07-Jun-2009 08:36:55] <Dieterbe> yeah
[07-Jun-2009 08:37:45] <raddy> Dieterbe : I am using ZenOSS 2.4.1 in 64Bit CentOS 5.2x
[07-Jun-2009 08:38:36] <raddy> Dieterbe : ZenJobs and ZenActions occupies 100% CPU when i start it.
[07-Jun-2009 08:38:48] <raddy> Dieterbe : Can you help me?
[07-Jun-2009 08:40:30] <Dieterbe> what cpu do you have?
[07-Jun-2009 08:40:55] <raddy> 64Bit Intel Quad Core
[07-Jun-2009 08:43:26] <Dieterbe> hmm and you have many jobs/hosts?
[07-Jun-2009 08:43:43] <Dieterbe> i noticed that zenoss takes a lot of cpu when starting, but after a few minutes, it becomes normal
[07-Jun-2009 08:43:52] <Dieterbe> eg it decreases to reasonable levels
[07-Jun-2009 08:44:58] <raddy> Dieterbe : Not like that i my end, i don't even has that much jobs, more over even after sometime Zenaction and Zenjobs not decreasing the CPU usage.
[07-Jun-2009 08:47:55] <Dieterbe> raddy: can you start 'top' on command line, press '1' and then copy paste the lines at the top that say Cpu0, Cpu1 etc
[07-Jun-2009 08:50:54] <raddy> Dieterbe : What for?
[07-Jun-2009 08:51:29] <Dieterbe> to be able to help you
[07-Jun-2009 08:52:07] <raddy> Dieterbe : That is right. but i am able to tell whatever you ask, what you are seeking?
[07-Jun-2009 08:56:48] <Dieterbe> raddy: any unusual amounts of %wa, %sy/us etc
[07-Jun-2009 08:57:37] <raddy> Dieterbe : Ohh, that i didn't know,
[07-Jun-2009 08:57:51] <raddy> Dieterbe : I'll tell you now
[07-Jun-2009 19:42:15] <rbd> hey guys... I have a server with apache, mysql, etc already installed on it in use for other things. is it safe to install the zenoss core stack (e.g. using the stack installer) and not have it mess with my existing configurations, or should I install on a fresh box?
[07-Jun-2009 19:42:55] <rbd> I don't want to hose up what I already have...can see in the docs where they talk about how the installer is composed...where/how it installs on the box, etc
[07-Jun-2009 19:44:15] <rbd> s/can/can't
[07-Jun-2009 19:44:22] <sergeymasushko> I have installed zenoss on virtual FreeBSD box
[07-Jun-2009 19:53:27] scfh_ is now known as scfh
[08-Jun-2009 00:00:42] [disconnected at Mon Jun 8 00:00:42 2009]
[08-Jun-2009 00:00:42] [connected at Mon Jun 8 00:00:42 2009]
[08-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[08-Jun-2009 01:26:01] <raddy> Hello Everybody
[08-Jun-2009 01:26:08] <raddy> Anyone out there?
[08-Jun-2009 01:52:16] <raddy> Anybody out there???
[08-Jun-2009 02:08:18] <Appiah> just ask your question
[08-Jun-2009 02:08:31] <Appiah> ...
[08-Jun-2009 02:11:18] <raddy> Appiah : You always encourage people to ask their question when nobody is listening them.
[08-Jun-2009 02:11:52] <raddy> Appiah : you also won't respond to people's questions, but would poke them around.
[08-Jun-2009 02:12:07] <Appiah> always?
[08-Jun-2009 02:12:23] <raddy> Appiah : Seems so.
[08-Jun-2009 02:12:33] <Appiah> that's weird
[08-Jun-2009 02:12:53] <raddy> ok, i am asking the question.
[08-Jun-2009 02:13:20] <raddy> ZenJobs and ZenActions using 100% CPU when started.
[08-Jun-2009 02:13:55] <raddy> I am using ZenOSS 2.4.1 on 64Bit Quad Core CentOS 5.2 Server.
[08-Jun-2009 02:13:58] <Appiah> http://www.sabi.co.uk/Notes/linuxHelpAsk.html#anyone this is a "how to ask questions for linux on IRC" but it pretty much covers all tech related areas and asking forhelp on IRC
[08-Jun-2009 02:14:46] <raddy> Appiah : I am not a IRC Noob.
[08-Jun-2009 02:14:55] <Appiah> oh ok
[08-Jun-2009 02:15:04] <Appiah>
[08-Jun-2009 02:15:56] <raddy> I hope you might know get to know my proficiency in IRC from describing the question.
[08-Jun-2009 02:16:21] <Appiah> if I remeber correctly you asked for this before right?
[08-Jun-2009 02:16:28] <Appiah> 100% CPU load?
[08-Jun-2009 02:16:32] <raddy> Appiah : exactly
[08-Jun-2009 02:16:46] <raddy> Appiah : It has not been resolved yet.
[08-Jun-2009 02:16:52] <Appiah> and no ones ide sofar helped
[08-Jun-2009 02:17:02] <Appiah> you tried making a post on the forum?
[08-Jun-2009 02:17:10] <raddy> Appiah : Have to.
[08-Jun-2009 02:17:26] <Appiah> Is it a freshly installed 2.4.1 or did you upgrade to it?
[08-Jun-2009 02:17:47] <raddy> Appiah : Yesterday one person helped me, but i was unable continue the conversation at that time.
[08-Jun-2009 02:17:56] <raddy> Appiah : Nope, Upgrade.
[08-Jun-2009 02:18:08] <Appiah> Do you have the logs since the upgrade?
[08-Jun-2009 02:18:46] <raddy> Appiah : Yeah, but ZenJobs logs or empty, ZenActions log contains nothing interesting.
[08-Jun-2009 02:19:02] <raddy> *Are
[08-Jun-2009 02:19:48] <Appiah> I mean logs from the update itself
[08-Jun-2009 02:20:13] <Appiah> output from when you ran the 2.4.1 stack installer
[08-Jun-2009 02:20:13] <raddy> Appiah : I don't think so. where it exists?
[08-Jun-2009 02:20:28] <Appiah> good question ... I just dump it myself
[08-Jun-2009 02:20:41] <raddy> Appiah : Oh.
[08-Jun-2009 02:20:42] <Appiah> especially when error/warnings
[08-Jun-2009 02:20:46] <Appiah> shows up
[08-Jun-2009 02:20:57] <Appiah> you didnt get a message about doing a zenmigrate with force?
[08-Jun-2009 02:21:33] <raddy> Appiah : I got, but i didn't get any errors and so i didn't dump the message.
[08-Jun-2009 02:21:51] <raddy> Appiah : ok, see you in forums (If you have account)
[08-Jun-2009 05:38:19] scfh is now known as scfh_
[08-Jun-2009 05:38:23] scfh_ is now known as scfh
[08-Jun-2009 06:25:43] <Appiah> Zenjobs wont start after a upgrade (2.4.1), output from trying to debug in terminal -> http://pastebin.com/m1c8b1e7b
[08-Jun-2009 06:41:20] <Appiah> I ran a zenmigrate .. AGAIN and now it starts
[08-Jun-2009 13:28:01] <rocket> Hey cgibbons
[08-Jun-2009 13:28:06] <cgibbons> howdy mr rocket
[08-Jun-2009 13:28:21] <rocket> How are things going?
[08-Jun-2009 13:28:40] <cgibbons> quiet so far (my 2 week stint on L3 support), which is good
[08-Jun-2009 13:28:42] <cgibbons> enjoy the test?
[08-Jun-2009 13:28:56] <rocket> it wasnt too bad .. enjoy the answers?
[08-Jun-2009 13:29:46] <cgibbons> still looking, but it was good from my perspective
[08-Jun-2009 13:30:10] <rocket> cgibbons: linux is pissing me off at the moment .. cant seem to find a great way of getting nic speed without being root ..
[08-Jun-2009 13:30:49] <cgibbons> deja vu on that
[08-Jun-2009 13:31:08] * rocket is working on improving the linux zenpack .. eg uname -i support that maps to Intel or AMD for hardware platform
[08-Jun-2009 13:31:30] <rocket> already fixed the nic perf collection that was missing
[08-Jun-2009 13:32:32] <rocket> but to get the thresholds to work right I am trying to figure out a way besides dmesg to get nic speed
[08-Jun-2009 13:33:45] <rocket> cgibbons: hey you might know this .. is there a way to easily do a ssh query in a modeller before the main code goes?
[08-Jun-2009 13:34:26] <rocket> eg I am looking for a specific file .. if it exists run the modeller ...
[08-Jun-2009 13:35:01] <cgibbons> yeah that's a common problem in the modelers, and I don't think we have a solution for it yet (I'd like to do the same thing in the Windows ones). Brain E was looking @ that once, let me see if he had a solution.
[08-Jun-2009 13:35:41] <rocket> k sounds good .. I have a work around but it would cleanup my stuff in the future .. etc ..
[08-Jun-2009 13:46:20] <cgibbons> ok yeah, that ability doesn't exist yet
[08-Jun-2009 13:46:46] <rocket> cgibbons: thats where I was leaning to .. thanks for the followup
[08-Jun-2009 16:52:15] <_indigo> good evening everybody, can anyone help me to clarify my issue, I´m a little bit confused, is by zenoss a nagios or something like that under the hood?
[08-Jun-2009 16:53:17] <kgoedtel> zenoss is under the hood
[08-Jun-2009 16:54:30] <_indigo> ok thx, we evaluate some network management software and my work mate has installed zenoss saw during the installation some "nagios stuff". I thought I ask some guys who know what they do
[08-Jun-2009 16:55:31] <kgoedtel> you can utilize nagios plugins- but that is more or less the extent of which zenoss "uses" nagios
[08-Jun-2009 16:56:09] <kgoedtel> all the real work is done in zenoss (python)
[08-Jun-2009 16:56:41] <_indigo> ah ok I understand, for sure there are a lot of nagios plugins out there, so it make sense to support them. thx for help
[08-Jun-2009 16:56:49] <kgoedtel> np
[08-Jun-2009 17:31:50] <rmatte> If anyone is interested, I posted a detailed guide on getting LDAP working for Zenoss core:
[08-Jun-2009 17:31:51] <rmatte> http://forums.zenoss.com/viewtopic.php?p=35737#35737
[08-Jun-2009 17:32:09] <rmatte> It goes more in depth than what's in the guide that explains setting up the plugins and all that
[08-Jun-2009 19:46:29] <daMaestro> anyone know what data to enumerate when building a graph for disk space (not percent)
[08-Jun-2009 19:46:46] <daMaestro> like, raw usage (GB/TB/PB/etc)
[08-Jun-2009 19:47:46] <daMaestro> I currently use usedBlocks and then: ${here/totalBlocks},/,100,*
[08-Jun-2009 19:48:10] <daMaestro> however, that does not work when trying a multgraph for multiple devices of storage usage
[09-Jun-2009 00:00:42] [disconnected at Tue Jun 9 00:00:42 2009]
[09-Jun-2009 00:00:42] [connected at Tue Jun 9 00:00:42 2009]
[09-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[09-Jun-2009 04:07:57] <capricorn-80> hi ! i need to draw performance graph for my traffic
[09-Jun-2009 04:08:26] <capricorn-80> I want to monitor my router in/out traffic
[09-Jun-2009 04:09:15] <capricorn-80> i have created a new template with name bandwidth
[09-Jun-2009 04:09:42] <capricorn-80> zenoss has prebuil OID ?
[09-Jun-2009 04:09:49] <capricorn-80> prebuilt **
[09-Jun-2009 04:13:52] <capricorn-80> anyone there ?\
[09-Jun-2009 04:17:20] <capricorn-80> anyone home ?
[09-Jun-2009 04:54:03] <capricorn-80> anyone tell me how can i bind templates ?
[09-Jun-2009 04:55:50] <capricorn-80> ok i have bind tempates to devices
[09-Jun-2009 05:25:06] <capricorn-80> hi rocket
[09-Jun-2009 05:56:03] <capricorn-80> hello hello
[09-Jun-2009 06:40:46] <capricorn-80> hi chudler1
[09-Jun-2009 07:34:24] <capricorn-80> hello again
[09-Jun-2009 07:35:08] <capricorn-80> I am stuck in binding
[09-Jun-2009 07:35:15] <capricorn-80> need some help
[09-Jun-2009 07:39:07] <Appiah_> Does anyone know why taking a backup of the zenoss uses so much disk space and timesout?
[09-Jun-2009 07:40:29] <jb> yeah, you can change the temporary location via --temp-dir if /tmp doesn't have enough room.
[09-Jun-2009 07:41:05] <Appiah_> there's like 60gb free
[09-Jun-2009 07:41:11] <Appiah_> it used 30gb sofar
[09-Jun-2009 07:41:41] <Appiah_> previous backups have been about 1gb file when it's done , but how can it use about 30 > to make 1gb?
[09-Jun-2009 07:42:14] <jb> compression more than likely..
[09-Jun-2009 07:42:16] <jb> lots of text?
[09-Jun-2009 07:42:37] <jb> how large is your event db?
[09-Jun-2009 07:43:40] <Appiah_> good question
[09-Jun-2009 07:43:52] <Appiah_> i think i'll try from cli then the webinterface
[09-Jun-2009 07:51:11] <capricorn-80> jb: I am trying to bind ethernetCsmacd /Devices/Network/Router/Cisco Standard ethernet interface template with 75% utilization threshold
[09-Jun-2009 07:51:16] <capricorn-80> but its not working
[09-Jun-2009 07:52:18] <capricorn-80> may be i m doing something wrong in performing the steps
[09-Jun-2009 07:52:34] <jb> the ethernetCsmacd template should not be binded to a class
[09-Jun-2009 07:53:15] <jb> what is your reason for doing so?
[09-Jun-2009 07:54:09] <capricorn-80> jb: I want in/out traffic of my router .. like mrtg
[09-Jun-2009 07:54:34] <jb> the Router/Cisco class inherits that template from /Device
[09-Jun-2009 07:54:40] <jb> there is no need to do any additional binding
[09-Jun-2009 07:56:11] <capricorn-80> jb: if i see on /Devices /Network /Router /Cisco i can see Available Performance Templates and in that i can see ethernetCsmacd
[09-Jun-2009 07:56:29] <cgibbons> yeah but don't bind it
[09-Jun-2009 07:56:35] <jb> right
[09-Jun-2009 07:56:40] <capricorn-80> now how can i put this capability into my settings so that i can monitor my in/out traffic ?
[09-Jun-2009 07:56:47] <jb> its already in place..
[09-Jun-2009 07:56:51] <jb> go to the "OS" tab of the router
[09-Jun-2009 07:57:01] <jb> you should see the network interfaces there.. click on them
[09-Jun-2009 07:57:13] <jb> but, ensure that it is not binded
[09-Jun-2009 08:04:54] <capricorn-80> not binded means ?
[09-Jun-2009 08:05:05] <capricorn-80> i cant see x.x.x.254 in the list
[09-Jun-2009 08:06:44] <capricorn-80> this is my gateway ip
[09-Jun-2009 08:07:14] <capricorn-80> and i want to check its performance . and i have enable snmp on this and my zenoss is polling to this router
[09-Jun-2009 08:12:50] <rocket> morning cgibbons
[09-Jun-2009 08:13:22] <jb> sup rocket
[09-Jun-2009 08:13:55] <rocket> not much and you jb?
[09-Jun-2009 08:14:25] <rocket> jb: got a couple ssh zenpacks under my belt now ..
[09-Jun-2009 08:14:30] <jb> nice
[09-Jun-2009 08:14:32] <jb> which ones?
[09-Jun-2009 08:14:37] <jb> gentoo?
[09-Jun-2009 08:14:43] <rocket> I am just submitting them to matt now
[09-Jun-2009 08:15:02] <rocket> an extension to linux monitor that turns on network perf monitoring
[09-Jun-2009 08:15:23] <rocket> and a gentoo one that adds software installed support
[09-Jun-2009 08:16:43] <rocket> these zenpacks also include the tests etc so there is pretty decent examples with both of them
[09-Jun-2009 08:19:48] <rocket> jb: predictive threshold and aix is currently being tested by matt
[09-Jun-2009 08:23:17] <capricorn-80> jb: what should i do ?
[09-Jun-2009 08:24:41] <rocket> capricorn-80: are you seeing anything under the events tab? or the event console .. please change the dropdown to include debug events
[09-Jun-2009 08:25:28] <rocket> capricorn-80: if you arent seeing the data for just the interfaces you might need to look at the logs on the system itself to see if there is anything in there
[09-Jun-2009 08:27:53] <capricorn-80> i cant see any process for that router
[09-Jun-2009 08:27:55] <capricorn-80> its fine
[09-Jun-2009 08:28:38] <rocket> you mean you cant see anything under the events tabs?
[09-Jun-2009 08:29:47] <rocket> capricorn did you (model/add) your router after enabling all the snmp stuff .. eg making zenoss aware of it
[09-Jun-2009 08:30:07] <capricorn-80> yes
[09-Jun-2009 08:30:25] <capricorn-80> rocket: i can get all info from the router on zenoss
[09-Jun-2009 08:30:32] <rocket> capricorn-80: can you post a screenshot of what your seeing?
[09-Jun-2009 08:30:37] <capricorn-80> its that just the interface x.x.x.254 is not showing
[09-Jun-2009 08:30:52] <capricorn-80> and this is the interface which has been enabled with snmp
[09-Jun-2009 08:31:34] <rocket> capricorn-80: I will need to see a screenshot to fully understand exactly what you are seeing
[09-Jun-2009 08:45:30] <rocket> morning matt
[09-Jun-2009 08:47:09] <mrayzenoss> good morning
[09-Jun-2009 08:47:28] <rocket> matt I sent you a couple of zenpacks this morning
[09-Jun-2009 08:48:06] <mrayzenoss> yeah, looks like I've got a busy day
[09-Jun-2009 08:48:27] <rocket> these last 2 are pretty small ones ...
[09-Jun-2009 08:48:55] <rocket> did the predictive one work any better for you last night?
[09-Jun-2009 08:49:21] <mrayzenoss> haven't checked on it yet
[09-Jun-2009 08:49:26] <rocket> ok
[09-Jun-2009 08:54:22] <capricorn-80> rocket: well its very simple
[09-Jun-2009 09:08:18] rocket_ is now known as rocket
[09-Jun-2009 09:13:55] <rocket> ls
[09-Jun-2009 09:14:00] <rocket> oops sorry
[09-Jun-2009 09:14:01] <rmatte> .
[09-Jun-2009 09:14:02] <rmatte> ..
[09-Jun-2009 09:14:15] <rocket> rmatte: haa haa
[09-Jun-2009 09:14:23] <rmatte> /bin
[09-Jun-2009 09:14:26] <rmatte>
[09-Jun-2009 09:14:59] <rocket> rmatte: I did get a patch submitted for the linux zenpack not enabling those collectors by default
[09-Jun-2009 09:15:07] <rmatte> excellent
[09-Jun-2009 09:15:09] <rocket> rmatte: we will see what happens with that
[09-Jun-2009 09:15:18] <rmatte> yeh, that's awesome though, nice work
[09-Jun-2009 09:15:39] <rocket> rmatte: and I did submit a new extended linux zenpack that fixes the perf monitoring for network interfaces
[09-Jun-2009 09:15:52] <rmatte> even better hehe
[09-Jun-2009 09:15:56] <rocket> rmatte: just today ..
[09-Jun-2009 09:16:02] <rmatte> though wasn't there another thing missing?
[09-Jun-2009 09:16:14] <rmatte> there was interfaces and filesystem info missing no?
[09-Jun-2009 09:16:22] <rocket> rmatte: filesystems too?
[09-Jun-2009 09:16:34] <rmatte> let me check...
[09-Jun-2009 09:16:41] <rocket> rmatte: I guess I hadnt noticed that one .. if thats the case I can add filesystems to mine ...
[09-Jun-2009 09:17:00] <rmatte> oh no, it was memory collection missing
[09-Jun-2009 09:17:12] <rmatte> which the zenoss.cmd.linux.memory plugin does
[09-Jun-2009 09:17:28] <rocket> matt was that part of the bug report?
[09-Jun-2009 09:17:35] <rmatte> yessir
[09-Jun-2009 09:17:40] <rmatte> http://dev.zenoss.com/trac/ticket/5047
[09-Jun-2009 09:17:49] <rocket> ok then when that is fixed then it should be all good
[09-Jun-2009 09:17:52] <capricorn-80> i have enable snmp on x.x.x.254 router and i added x.x.x.254 in device
[09-Jun-2009 09:18:03] <rmatte> The "zenoss.cmd.linux.memory" and "zenoss.cmd.linux.ifconfig" zCollectorPlugins (which are included in Core) do not get automatically bound to /Linux/SSH when the Linux Monitor ZenPack is installed.
[09-Jun-2009 09:18:25] <rmatte> Only CPU and Filesystem info is pulled via SSH by default once the Linux Monitor ZenPack is installed. Collection of Interface info as well as Memory info is supported in Core by default, but not automatically activated for the /Linux/SSH device grouping.
[09-Jun-2009 09:19:16] <rocket> rmatte: hrmm so perf stuff just needed to be enabled?
[09-Jun-2009 09:19:18] <rmatte> looks like your patch does take that in to account though no?
[09-Jun-2009 09:19:26] <rmatte> pretty much
[09-Jun-2009 09:19:50] <rocket> rmatte: the templates for network perf were still snmp based it looked like to me
[09-Jun-2009 09:19:58] <rmatte> 2 plugins need to be bound to the /Linux/SSH event class when the Linux Monitor Zenpack is installed, though extending the function of the Zenpack itself might be more effective
[09-Jun-2009 09:20:05] <rmatte> you would basically just be duplicating work though
[09-Jun-2009 09:20:31] <rocket> rmatte: my zenpack doesnt include the patch to the main linux zenpack
[09-Jun-2009 09:20:37] <rmatte> your patch seems to be what's needed
[09-Jun-2009 09:20:51] <rmatte> so it's all good
[09-Jun-2009 09:21:24] <rocket> rmatte: good .. and my extended linux monitoring zenpack creates the new perf template for nics that is ssh based instead of snmp based
[09-Jun-2009 09:21:33] <rmatte> ah
[09-Jun-2009 09:21:37] <rmatte> gotcha, nice
[09-Jun-2009 09:22:11] <rocket> now you have me curious about filesystems .. if the perf ones for those are correct as well
[09-Jun-2009 09:22:34] <rmatte> hehe
[09-Jun-2009 09:23:02] <rocket> rmatte: no big deal if they arent ..
[09-Jun-2009 09:24:23] <chudler> anyone know offhand which version of twisted Zenoss uses these days?
[09-Jun-2009 09:24:41] <jb> Twisted 8.1.0
[09-Jun-2009 09:25:52] <chudler> thanks, cool. any plan to do 8.2.0? ;-) I know it needs testing, probably. Has a SSL/TLS fix for xmpp.
[09-Jun-2009 09:26:18] <jb> im not the right person to answer that one
[09-Jun-2009 09:30:16] <rocket> jb: awww why not?
[09-Jun-2009 09:32:42] <rmatte> that'd be a question for the next time the developers drop by IRC
[09-Jun-2009 09:33:00] <rocket> rmatte: just pulling your leg
[09-Jun-2009 09:33:03] * chudler adds it to the list
[09-Jun-2009 09:34:50] <capricorn-80> rocket: i have enable snmp on x.x.x.254 router and i added x.x.x.254 in device .
[09-Jun-2009 09:35:51] <rocket> capricorn-80: like I said I dont understand exactly what you are seeing ... a screenshot would help ... without that I am not sure how much help I would be at the moment
[09-Jun-2009 09:36:00] <jb> capricorn-80: you really aren't providing many details.. we already requested a screnshot..
[09-Jun-2009 09:36:36] <capricorn-80> jb: you are right but the thing is i have many ips show on that as i have configured vlan
[09-Jun-2009 09:36:51] <capricorn-80> so screenshot is not good solution thats why i repeat that again
[09-Jun-2009 09:37:19] <rocket> jb: filesystem perf is already covered in the core linux zenpack .... whew
[09-Jun-2009 09:37:50] <rocket> so do you have any perf statistics on any of the network interfaces?
[09-Jun-2009 09:38:03] <rocket> capricorn-80: or is it a problem with one specific port?
[09-Jun-2009 09:38:26] <rocket> capricorn-80: the problem is that your not exactly explaining your problem well
[09-Jun-2009 09:47:04] <capricorn-80> rocket: let me try . take like this .. my router address is 172.16.1.254 and i have enabled snmp on it so that i can monitor cpu , memory and bandwidth on it. I added it in device list and i am getting many details about it like cpu , memory performance graphs. on OS - interfaces i can see VLAN interface ip addresses
[09-Jun-2009 09:47:40] <capricorn-80> but i cant see the interface with ip address 172.16.1.254
[09-Jun-2009 09:48:00] <rocket> you see every other interface but that one?
[09-Jun-2009 09:48:16] <rocket> is that interface different than the otherones in anyway?
[09-Jun-2009 09:48:42] <rocket> capricorn-80: can you paste bin the output of this for me from the command line
[09-Jun-2009 09:48:52] <capricorn-80> no
[09-Jun-2009 09:48:57] <rocket> zenmodeler run -v 10 -d <device_name>
[09-Jun-2009 09:49:24] <jb> not in this channel though
[09-Jun-2009 09:49:24] <jb>
[09-Jun-2009 09:49:42] <rocket> jb: I did say pastebin
[09-Jun-2009 09:49:48] <jb> ah yeah
[09-Jun-2009 09:50:05] <rocket> oops the space might have made that confusing
[09-Jun-2009 09:52:35] <capricorn-80> rocket: http://pastebin.com/m13349a13
[09-Jun-2009 09:52:45] <capricorn-80> hmm
[09-Jun-2009 09:52:47] <capricorn-80> strange
[09-Jun-2009 09:53:09] <capricorn-80> the command is in /opt/zenoss/bin when i type it but when i run gave error
[09-Jun-2009 09:53:15] <capricorn-80> paths problem
[09-Jun-2009 09:53:29] <rocket> capricorn-80: run it as the zenoss user
[09-Jun-2009 09:56:34] <capricorn-80> rocket: its long long output
[09-Jun-2009 09:56:50] <rocket> capricorn-80: good .. it might show whats going on then
[09-Jun-2009 09:58:46] <capricorn-80> rocket: http://pastebin.com/m2777ed65
[09-Jun-2009 10:00:35] <mrayzenoss> chudler: http://dev.zenoss.org/trac/ticket/4991 Twisted 8.2 will probably not be in this release, unless a developer finds something really broken in 8.1
[09-Jun-2009 10:01:03] <chudler> thanks mrayzenoss. I will work around this in my code.
[09-Jun-2009 10:01:18] <rocket> capricorn-80: how many nics are in the box?
[09-Jun-2009 10:02:06] <capricorn-80> the zenoss box or the router ?
[09-Jun-2009 10:02:25] <rocket> capricorn-80: linux router? and yes how many nics in the router?
[09-Jun-2009 10:05:57] <capricorn-80> well router has vlan
[09-Jun-2009 10:10:01] <rocket> capricorn-80: I am seeing 10.x.x.x based interfaces .. not 172.x based ones ...
[09-Jun-2009 10:10:17] <capricorn-80> rocket: yea r u right
[09-Jun-2009 10:10:21] <capricorn-80> i gave u example
[09-Jun-2009 10:10:39] <rocket> capricorn-80: pastebin a netstat -rn and an ifconfig -a if possible please
[09-Jun-2009 10:13:55] <rocket> capricorn-80: I am assuming eth1 is your x.254 interface correct?
[09-Jun-2009 10:14:34] <capricorn-80> rocket: http://pastebin.com/m52f2d775
[09-Jun-2009 10:16:56] <rocket> I believe the issue is that either ucd-snmp doesnt support vlans or it is not currently configured to show them
[09-Jun-2009 10:17:20] <rocket> or is it called netsnmpd .. cant remember
[09-Jun-2009 10:18:57] <rocket> capricorn-80: how did you create the vlan?
[09-Jun-2009 10:19:40] <rocket> capricorn-80: can you pastebin the output of /proc/net/vlan/config ?
[09-Jun-2009 10:19:46] <capricorn-80> router commands
[09-Jun-2009 10:19:51] <capricorn-80> its cisco router
[09-Jun-2009 10:21:06] <rocket> capricorn-80: from the output you gave me it looks like your looking for some extra information about 169.254.0.0 correct?
[09-Jun-2009 10:21:46] <rocket> capricorn-80: the information you sent me doesnt look like it belongs to a cisco router ..
[09-Jun-2009 10:21:56] <rocket> capricorn-80: it looks very linuxy to me ..
[09-Jun-2009 10:22:51] <rocket> i guess they could be using linux now for some devices ..
[09-Jun-2009 10:23:40] <rocket> is the only thing missing information about 169.254.0.0?
[09-Jun-2009 10:24:20] <rocket> capricorn-80: have you done a snmpwalk against the router to see if snmp is showing any information about 169.254.0.0 or whatever you are thinking you are missing
[09-Jun-2009 10:24:33] <rocket> otherwise from the output I am seeing things look correct ...
[09-Jun-2009 10:26:10] <capricorn-80> rocket: right that was for linux box . i thought u want my zenoss server information using that command
[09-Jun-2009 10:26:45] <rocket> no I want the information about the router
[09-Jun-2009 10:26:56] <rocket> zenmodeler run -v 10 -d <router>
[09-Jun-2009 10:27:19] <rocket> and then any information about the interface config / vlans on the router in a pastebin
[09-Jun-2009 10:28:23] <capricorn-80> rocket: http://pastebin.com/m4ce5fcf
[09-Jun-2009 10:28:39] <capricorn-80> its long long output .. just got this one
[09-Jun-2009 10:28:43] <capricorn-80> its short one
[09-Jun-2009 10:29:32] <rocket> capricorn-80: the last pastebin is what exactly?
[09-Jun-2009 10:29:47] <rocket> capricorn-80: it looks like it isnt gathering any info from the device based on that output
[09-Jun-2009 10:30:01] <rocket> did you run with -v10?
[09-Jun-2009 10:30:05] <capricorn-80> rocket: yes
[09-Jun-2009 10:30:22] <rocket> capricorn-80: I would like the full output please
[09-Jun-2009 10:31:03] <capricorn-80> rocket: how ?
[09-Jun-2009 10:31:14] <capricorn-80> i cant insert into txt file
[09-Jun-2009 10:31:16] <capricorn-80> anyway ?
[09-Jun-2009 10:32:37] <rocket> capricorn-80: I need to step away from this for a bit ... I am getting frustrated because I am feeling that we are not communicating very well .. I am asking for very specific things and feel that I am not getting those things
[09-Jun-2009 10:32:55] <capricorn-80> i am
[09-Jun-2009 10:33:11] <capricorn-80> rocket: i am needy person
[09-Jun-2009 10:33:31] <capricorn-80> the problem is that i cannot move that output to txt file due to permission problems
[09-Jun-2009 10:34:13] <capricorn-80> rocket: http://pastebin.com/m44dacd8c
[09-Jun-2009 10:34:20] <capricorn-80> this is the whole screen now
[09-Jun-2009 10:37:32] <rocket> capricorn-80: so are you not seeing any of the ip addresses for the vlans?
[09-Jun-2009 10:38:23] <rocket> capricorn-80: a screenshot of the zenoss router information page would be useful now
[09-Jun-2009 10:38:48] <capricorn-80> wait
[09-Jun-2009 10:39:07] <rocket> capricorn-80: without it at this point I dont think i will be able to help you .. .. however i dont think zenoss supports vlans in the manner you are looking for at this point
[09-Jun-2009 10:39:40] <rocket> it will just show information about the underlying interface I believe ...
[09-Jun-2009 10:47:25] <capricorn-80> rocket:well i can see vlan traffic
[09-Jun-2009 10:48:10] <capricorn-80> i mean in/out traffic from vlan but cant see the main interface 10.254.95.254 through which i m polling ..
[09-Jun-2009 10:48:13] <capricorn-80> just wait
[09-Jun-2009 10:48:16] <capricorn-80> i talk to you
[09-Jun-2009 11:06:58] <xulfer> Is there anyway to convert from a string to an integer within a datapoint? Like when dealing with the output of extend
[09-Jun-2009 11:52:59] <lukecyca> When using COMMAND data source, is 30 second timeout hardcoded? Is there any way to extend it?
[09-Jun-2009 12:05:39] <lukecyca> I set it in zProperties for the device, but it doesn't seem to be honored
[09-Jun-2009 12:37:59] <lukecyca> Aha. zCommandCommandTimeout is honored when the command is actually run. It's just not honored when you click the "Test" button for the command. The test button seems to be hardcoded to a 30 second timeout.
[09-Jun-2009 12:38:24] <chudler> interesting. thanks
[09-Jun-2009 12:50:05] <rmatte> Question... I just had a process go down on a server then come back up, after it came back up Zenoss didn't recognize that it was back. I had to remodel the device to get it to realize it was back. This is because the process name switched From: Microsoft.Exchange.Pop3.exe -pipe:692 -stopkey:Global\ExchangeStopKey-5da6d91e-b226-4009-919f-ecbbf3831293 -resetkey:Global\ExchangeResetKey-650f2307-38fb-
[09-Jun-2009 12:50:06] <rmatte> To: Microsoft.Exchange.Pop3.exe -pipe:692 -stopkey:Global\ExchangeStopKey-c052f2b8-9ef0-4602-b2f3-7f7393a2555f -resetkey:Global\ExchangeResetKey-7e9c1ce3-672e-
[09-Jun-2009 12:50:11] <rmatte> when the process restarted
[09-Jun-2009 12:50:15] <rmatte> what's the best way around this?
[09-Jun-2009 12:50:30] <rmatte> should I edit the process and remove the extra options then lock it?
[09-Jun-2009 12:50:49] <chudler> there is a option for the process monitor to "ignore arguments". this might do what you need
[09-Jun-2009 12:51:07] <rmatte> where would I find that?
[09-Jun-2009 12:51:37] <chudler> found it. It is called "Ignore Parameters" and is on the edit tab of the process.
[09-Jun-2009 12:52:24] <rmatte> ah thanks
[09-Jun-2009 12:53:04] <chudler> unfortunately, this option doesn't help if you have things that run inside an interpreter like python, ruby, etc. I would love to be able to solve that one..
[09-Jun-2009 13:00:55] <rmatte> right
[09-Jun-2009 13:02:33] <capricorn-80> rocket: you there ?
[09-Jun-2009 13:02:57] <rocket> for a few minutes .. i have to run an errand soon
[09-Jun-2009 13:03:22] <capricorn-80> ah
[09-Jun-2009 13:04:18] <capricorn-80> you want to see pic now ?
[09-Jun-2009 13:04:25] <rocket> yes please
[09-Jun-2009 13:05:47] <capricorn-80> imagebin ?
[09-Jun-2009 13:05:55] <rocket> sure
[09-Jun-2009 13:05:57] <rmatte> chudler: One I toggle that option do I have to remodel every server that is monitoring that particular process?
[09-Jun-2009 13:06:03] <rmatte> One == Once
[09-Jun-2009 13:06:04] <capricorn-80> i forget the website address
[09-Jun-2009 13:06:17] <capricorn-80> got it
[09-Jun-2009 13:09:38] <capricorn-80> rocket: http://imagebin.ca/view/PvjbBx.html
[09-Jun-2009 13:10:40] <capricorn-80> rocket: u can see that vlan can be recognize by the zenoss
[09-Jun-2009 13:11:09] <rocket> ok
[09-Jun-2009 13:11:16] <rocket> looking
[09-Jun-2009 13:12:29] <capricorn-80> ok
[09-Jun-2009 13:15:24] <rocket> so your issue is in the routes or in the interfaces themselves?
[09-Jun-2009 13:15:40] <capricorn-80> yes
[09-Jun-2009 13:15:52] <capricorn-80> i just added my router and i did nothing
[09-Jun-2009 13:16:00] <capricorn-80> i mean added my router as device
[09-Jun-2009 13:16:10] <capricorn-80> oh sorry got it wrong
[09-Jun-2009 13:16:30] <capricorn-80> my issue is why i cannot see 10.254.95.254
[09-Jun-2009 13:16:35] <capricorn-80> in the interfaces
[09-Jun-2009 13:16:51] <capricorn-80> coz i want to monitor 10.254.95.254 and its in/out traffic
[09-Jun-2009 13:18:50] <rocket> which vlan is that interface on?
[09-Jun-2009 13:19:47] <capricorn-80> well this interface is not on any vlan .. on this router different vlans are configured
[09-Jun-2009 13:20:21] <capricorn-80> one strage thing is this IP Addresses 127.0.0.22/8
[09-Jun-2009 13:20:40] <capricorn-80> and i can see some graphs of it that it contains traffic which is very strange
[09-Jun-2009 13:22:21] <rocket> capricorn-80: I guess I would save the pictures and ask one of the main devs on thursday ..
[09-Jun-2009 13:23:36] <capricorn-80> ok
[09-Jun-2009 13:23:55] <capricorn-80> devs ?
[09-Jun-2009 13:24:10] <capricorn-80> developers ?
[09-Jun-2009 13:33:10] <drtns> Hey all I am trying to make an event mapping and it's not working I am sure it's something dumb that I am doing but I need some help. The message that comes into Zenoss is '[error] jk2_init() Can't find child 19024 in scoreboard' and my regex is [error] jk2_init() Can't find child * in scoreboard and I am telling Zenoss to send that directly to history. but it's not working. what is the best way to troubleshoot this... is my regex messe
[09-Jun-2009 13:42:54] <rocket> drtns: you probably have issues escaping some of the characters of the regex
[09-Jun-2009 13:43:18] <rocket> eg brackets need to be escaped etc
[09-Jun-2009 13:43:46] <rocket> instead of * you would need something like [0-9]+
[09-Jun-2009 13:46:10] <rmatte> drtns: check my privmsg
[09-Jun-2009 13:48:40] <rmatte> drtns: something like: [error] j2k_init() Can't find child
[09-Jun-2009 13:48:43] <rmatte> should probably work
[09-Jun-2009 13:48:53] <rmatte> I have a regex of: (ip) vrf/dest_addr=
[09-Jun-2009 13:48:56] <rmatte> and it works fine
[09-Jun-2009 13:49:14] <rmatte> or just j2k_init() Can't find child
[09-Jun-2009 13:49:29] <rmatte> you don't really need a wildcard, just need to specify part of the line as the regex
[09-Jun-2009 13:54:40] <rmatte> if it's coming in with event class of /Unknown and no event key then you need to set the event class key for the mapping as "defaultmapping" to make it work properly
[09-Jun-2009 13:54:57] <rmatte> and no event class key rather
[09-Jun-2009 14:04:56] <sergeymasushko> how can I place FastEthernet0/0 Throughput graph on the 'Perf' tab?
[09-Jun-2009 14:31:23] <rocket> Hey Matt
[09-Jun-2009 14:55:23] <rocket> mrayzenoss_: ping
[09-Jun-2009 14:55:33] mrayzenoss_ is now known as mrayzenoss
[09-Jun-2009 14:55:35] <mrayzenoss> yo
[09-Jun-2009 14:55:42] <mrayzenoss> testing the PredictiveThreshold now
[09-Jun-2009 14:55:46] <mrayzenoss> and sitting in the defect review
[09-Jun-2009 14:56:01] <rocket> rmatte's bug regarding linux zenpacks is still missing a couple modellers
[09-Jun-2009 14:56:21] <rocket> err collector plugins
[09-Jun-2009 14:56:31] <rocket> eg zenoss.cmd.linux.netstat_an
[09-Jun-2009 14:56:34] <rocket> eg zenoss.cmd.linux.netstat_rn
[09-Jun-2009 14:56:38] <rocket> eg zenoss.cmd.linux.process
[09-Jun-2009 14:56:50] <rocket> need another patch to include those?
[09-Jun-2009 14:57:13] <rocket> or can we just add those three lines to that file in __init__.py
[09-Jun-2009 14:57:28] <mrayzenoss> do those collector plugins work?
[09-Jun-2009 14:57:38] <kwame> Hi, question
[09-Jun-2009 14:57:42] <rocket> the netstat ones seem to
[09-Jun-2009 14:57:52] <rocket> they added routing info for me
[09-Jun-2009 14:58:00] <kwame> when I install zenoss and then tell it to autodiscover and give it a username and passwword for an linux user
[09-Jun-2009 14:58:04] <kwame> to be able to ssh into the hosts
[09-Jun-2009 14:58:18] <kwame> is this information stored in a text file somewhere?
[09-Jun-2009 14:58:38] <mrayzenoss> sure, update the patch. If you want to roll up the patch into your LinuxMonitorAddon as well that way it'll be out before 2.5
[09-Jun-2009 14:59:04] <mrayzenoss> kwame: it's in the zodb somewhere
[09-Jun-2009 14:59:10] <rocket> ah ok .. i can do both ..
[09-Jun-2009 14:59:25] <rocket> predictive seem to be working better now?
[09-Jun-2009 14:59:37] <mrayzenoss> rocket: not yet, blew up again, still testing
[09-Jun-2009 14:59:46] <rocket> mrayzenoss: bah!
[09-Jun-2009 15:01:09] <rocket> mrayzenoss: wonder what failed now .. :/
[09-Jun-2009 15:02:05] <mrayzenoss> kwame: you can also use keys
[09-Jun-2009 15:03:06] <kwame> mrayzenoss: what I saw in some servers was that the user that I used to log in through ssh to the boxes, was leaving a defunct (zombie) process
[09-Jun-2009 15:03:21] <kwame> so there where a lot of my_username@notty in the boxes
[09-Jun-2009 15:03:36] * rocket starts working on the suse zenpack ..
[09-Jun-2009 15:03:51] <mrayzenoss> rocket: http://pastebin.com/m49ffaa92
[09-Jun-2009 15:04:05] <mrayzenoss> kwame: there's a ticket somewhere about leftover ssh sessions
[09-Jun-2009 15:04:12] <mrayzenoss> I think it may be fixed in 2.4.2
[09-Jun-2009 15:04:30] <rocket> mrayzenoss: did you use the new zenpack for predictive?
[09-Jun-2009 15:06:21] <mrayzenoss> yeah, I think so
[09-Jun-2009 15:06:27] <mrayzenoss> I bumped the version to 1.0.1
[09-Jun-2009 15:07:35] <mrayzenoss> let me remove the thresholds and try them again
[09-Jun-2009 15:08:21] <rocket> mrayzenoss: found the bug
[09-Jun-2009 15:08:27] <rocket> new one coming your way
[09-Jun-2009 15:08:42] <mrayzenoss> ahh, cool
[09-Jun-2009 15:08:55] <mrayzenoss> I uninstalled my 1.0.1, I'll merge and update to 1.0.2
[09-Jun-2009 15:09:20] <rocket> its trying to tune an rrd that doesnt exist yet ..
[09-Jun-2009 15:10:43] <rocket> mrayzenoss: are you just doing a check_ping rrd?
[09-Jun-2009 15:11:06] <mrayzenoss> yeah
[09-Jun-2009 15:11:37] <mrayzenoss> just using the Nagios check_ping steps to add a ping, figured that would be fun for predictive thresholds
[09-Jun-2009 15:12:03] <rocket> got the steps so I can try it here?
[09-Jun-2009 15:12:53] <mrayzenoss> http://www.zenoss.com/community/projects/zenpacks/nagios-check-ping
[09-Jun-2009 15:12:55] <sergeymasushko> how can I place FastEthernet0/0 Throughput graph on the 'Perf' tab? I get the OID from the ethernetCsmacd_64 template and add .1 at the end. when I do snmpwalk I get the result... I've added the OID in the perf template, when I do test on the datasource page I also get result, but graph don't shown.
[09-Jun-2009 15:13:00] <kwame> mrayzenoss: thanks for your help
[09-Jun-2009 15:13:06] <mrayzenoss> rocket: I just follow the PDF, don't need the ZenPack anymore
[09-Jun-2009 15:13:32] <rocket> mrayzenoss: what command do I put in?
[09-Jun-2009 15:13:44] <rocket> for the command template
[09-Jun-2009 15:14:02] <rocket> just trying to go quick on this one
[09-Jun-2009 15:15:09] <rocket> nevermind ..
[09-Jun-2009 15:15:41] <mrayzenoss> libexec/check_ping -H $devname -w 180,100% -c 300,100% | sed -e 's# - Packet loss#|LOSS#;s#%,##;s#ms##;s# = #=#g'
[09-Jun-2009 15:15:58] <mrayzenoss> can't copy and paste from that PDF, saved you a step
[09-Jun-2009 15:19:19] <CamargoBP> I'm having the hardest time to get this graph to average out 1 minute. It seems like it's averaging for the entire day!
[09-Jun-2009 15:21:20] <rmatte> hey Matt
[09-Jun-2009 15:21:22] <rmatte> http://img200.yfrog.com/img200/3739/zenmanager.jpg
[09-Jun-2009 15:21:33] <rmatte> the top view there, is that part of the enterprise LDAP zenpack?
[09-Jun-2009 15:21:50] <rmatte> I'm not familliar with that view, and I'm trying to help that guy with his LDAP config
[09-Jun-2009 15:23:09] <rmatte> oh wait, nevermind, it's the view that you see when you click on the actual user in zenoss
[09-Jun-2009 15:24:05] <CamargoBP> Here is what I am trying to do, I want to see how many voicemail we send a minute. the larger spike at the right of this image (blue spike) was 20 voicemails yet the graph shows max 2
[09-Jun-2009 15:24:06] <CamargoBP> http://i43.tinypic.com/2qdo2rk.jpg
[09-Jun-2009 15:24:38] <rmatte> Camargo, if you SNMPWalk that value, do you see higher than 2?
[09-Jun-2009 15:25:11] <rmatte> do an snmpwalk and compare the value that you get to the value displayed as the current value in the graph
[09-Jun-2009 15:25:23] <rocket> mrayzenoss: ping
[09-Jun-2009 15:25:34] <rmatte> I have an idea of what's wrong, just need to see the values first to make sure
[09-Jun-2009 15:25:40] <rocket> mrayzenoss: did it fail if you went to look at the graph immediately after creating the threshold?
[09-Jun-2009 15:25:47] <mrayzenoss> rocket: yeah
[09-Jun-2009 15:25:51] <rocket> ok
[09-Jun-2009 15:25:56] <rocket> then I think I have it all fixed up
[09-Jun-2009 15:25:58] <mrayzenoss> well, after rebinding the template
[09-Jun-2009 15:26:11] <CamargoBP> rmatte: http://pastebin.com/d72c11e4a
[09-Jun-2009 15:27:00] <mrayzenoss> rmatte: that's from Zenoss Enterprise (the LDAP stuff)
[09-Jun-2009 15:27:15] <CamargoBP> When I zoom into the spike it says 20 emails
[09-Jun-2009 15:33:27] <rhettlap> do you have one minute rra's configured
[09-Jun-2009 15:34:07] <CamargoBP> probably not.
[09-Jun-2009 15:35:13] <rhettlap> easy way to check would be to rrdtool fetch the data source using -r 60 and see if the time shows 60 or 300 seconds
[09-Jun-2009 15:35:33] <CamargoBP> Here is my command: http://i40.tinypic.com/16lk9ef.png
[09-Jun-2009 15:35:34] <rhettlap> i.e. rrdtool fetch patchtodatasource.rrd AVERAGE -r 60 -s -1h
[09-Jun-2009 15:36:08] <CamargoBP> That gives the last hours 1 minute average
[09-Jun-2009 15:36:20] <rhettlap> thats just for the visual not for the data source
[09-Jun-2009 15:36:35] <rmatte> does anyone here have experience with getting the LDAP authentication working with AD? I've got mine working fine with plain LDAP, but this guy on the forums is having problems getting it working with Active Directory (which I have no experience with).
[09-Jun-2009 15:36:42] <CamargoBP> rmatte
[09-Jun-2009 15:36:49] <rmatte> More specifically with getting the role mapping working
[09-Jun-2009 15:36:51] <CamargoBP> rmatte: yes we have it working
[09-Jun-2009 15:36:56] <rhettlap> your data source shows 60 second data points?
[09-Jun-2009 15:37:02] <CamargoBP> role mapping does not work, sorry
[09-Jun-2009 15:37:20] <rmatte> role mapping just flat out doesn't work with AD?
[09-Jun-2009 15:37:22] <CamargoBP> You have to manually add a user to a role, can't map roles to AD
[09-Jun-2009 15:37:34] <CamargoBP> I spent weeks trying to get it working
[09-Jun-2009 15:37:49] <CamargoBP> Works with openldap but not AD
[09-Jun-2009 15:37:56] <CamargoBP> I don't know why
[09-Jun-2009 15:37:58] <rmatte> CamargoBP: that output you pasted in pastebin... which one of those lines represents the number of voicemails?
[09-Jun-2009 15:38:29] <CamargoBP> The ones highlited, line 10 specifically.
[09-Jun-2009 15:38:54] <rocket> hey matt ... get my updated zenpack?
[09-Jun-2009 15:39:32] <rmatte> so in theory you could setup an OpenLDAP server, configure it to sync up the logins with the AD server and use that to map the roles
[09-Jun-2009 15:39:55] <CamargoBP> So in that pastebin 68 emails were sent since I've started monitoring, which are all represented on that graph, especially if I zoom in then you can see them all
[09-Jun-2009 15:40:23] <rmatte> ok, but you don't see them as 68, you see them as some smaller value?
[09-Jun-2009 15:41:18] <rmatte> correct/
[09-Jun-2009 15:41:19] <rmatte> ?
[09-Jun-2009 15:41:31] <CamargoBP> yeah, this is a better picture of the graph, the one on our HUD
[09-Jun-2009 15:41:32] <CamargoBP> http://i43.tinypic.com/2cyr7fq.png
[09-Jun-2009 15:41:59] <CamargoBP> So that large spike was 20 emails of the value 68
[09-Jun-2009 15:42:42] <rmatte> but it's only showing 4
[09-Jun-2009 15:42:46] <rmatte> or 2
[09-Jun-2009 15:42:49] <rmatte> ok, try this...
[09-Jun-2009 15:42:58] <rmatte> (not sure if you've already done this)
[09-Jun-2009 15:43:01] <CamargoBP> Exactly, the max should be at around 20 on the left.
[09-Jun-2009 15:43:03] <rmatte> go in to the template
[09-Jun-2009 15:43:05] <rmatte> open the graph
[09-Jun-2009 15:43:10] <rmatte> click on the datapoint
[09-Jun-2009 15:43:19] <rmatte> let me know when you've done this
[09-Jun-2009 15:43:34] <CamargoBP> I'm there
[09-Jun-2009 15:43:37] <rmatte> Under format, is it currently %5.2lf%s?
[09-Jun-2009 15:43:43] <CamargoBP> yep
[09-Jun-2009 15:43:50] <rmatte> remove the %s at the end
[09-Jun-2009 15:44:03] <rmatte> that'll keep Zenoss from trying to auto-convert the value to something else
[09-Jun-2009 15:44:04] <CamargoBP> What does that standfor?
[09-Jun-2009 15:44:34] <rmatte> not sure exactly what it stands for, I just know that removing the %s forces Zenoss to interpret the value literally
[09-Jun-2009 15:45:08] <rmatte> that should solve your problem
[09-Jun-2009 15:45:11] <CamargoBP> Hmmm... no go.
[09-Jun-2009 15:45:18] <rmatte> you may or may not have to delete the .RRD file after doing this
[09-Jun-2009 15:45:22] <CamargoBP> Still at 4
[09-Jun-2009 15:45:26] <CamargoBP> K
[09-Jun-2009 15:45:27] <rmatte> or it could take a few polling cycles for it to realize
[09-Jun-2009 15:45:29] <CamargoBP> Let me do that
[09-Jun-2009 15:45:32] <rmatte> k
[09-Jun-2009 15:45:55] <rmatte> I think it hard codes the units in to the RRD file somewhere, so deleting the file and waiting 3 polling cycles for it to regenerate is your best bet
[09-Jun-2009 15:47:15] <rmatte> I need someone to explain that unit string to me sometime
[09-Jun-2009 15:48:13] <rmatte> format string rather
[09-Jun-2009 15:49:40] <CamargoBP> Dang, so I deleted it and then it took about 3 cycles to make it again. same output.
[09-Jun-2009 15:49:55] <rmatte> really strange then
[09-Jun-2009 15:50:03] <CamargoBP> WAit!
[09-Jun-2009 15:50:10] <CamargoBP> It was cached I think
[09-Jun-2009 15:50:21] <rmatte> oh, the image?
[09-Jun-2009 15:50:24] <CamargoBP> Nope.
[09-Jun-2009 15:50:27] <rmatte> shift-refresh
[09-Jun-2009 15:50:30] <rmatte> lol
[09-Jun-2009 15:50:38] <rmatte> to be sure
[09-Jun-2009 15:50:45] <CamargoBP> Shoot
[09-Jun-2009 15:50:50] <CamargoBP> yeah I did
[09-Jun-2009 15:51:14] <rmatte> hmmm
[09-Jun-2009 15:51:40] <rmatte> what's the SNMP value you're polling?
[09-Jun-2009 15:52:16] <rmatte> .1.3.6.1.4.1.2021.8.1.3.1?
[09-Jun-2009 15:52:27] <rmatte> .1.3.6.1.4.1.2021.8.1.5.1 rather?
[09-Jun-2009 15:53:20] <rmatte> if you run a test on the datapoint against the device what's the output?
[09-Jun-2009 15:54:17] <CamargoBP> 1.3.6.1.4.1.2021.8.1.101.1
[09-Jun-2009 15:54:41] <rmatte> ok, and what's the output of walking that?
[09-Jun-2009 15:54:47] <rmatte> it's 68 and not something else right?
[09-Jun-2009 15:55:14] <CamargoBP> UCD-SNMP-MIB::extOutput.1 = STRING: 78
[09-Jun-2009 15:55:15] <CamargoBP> UCD-SNMP-MIB::extOutput.1 = STRING: 78
[09-Jun-2009 15:55:15] <CamargoBP> UCD-SNMP-MIB::extOutput.1 = STRING: 78
[09-Jun-2009 15:55:15] <CamargoBP> well now 78
[09-Jun-2009 15:55:28] <CamargoBP> oops pasted 3 times... mybad
[09-Jun-2009 15:55:31] <rmatte> ok, but it outputs 3 separate lines?
[09-Jun-2009 15:55:34] <rmatte> ah ok
[09-Jun-2009 15:55:36] <rmatte> I was wondering lol
[09-Jun-2009 15:56:23] <rmatte> well, I'm out of ideas then, after removing that %s there's no reason why it shouldn't work
[09-Jun-2009 15:56:36] <rmatte> oh, you might have to push changes to the collector?
[09-Jun-2009 15:56:53] <CamargoBP> Here is the new graph after have deleted the old one and sending 10 more emails: http://i41.tinypic.com/35lraye.png
[09-Jun-2009 15:57:29] <rmatte> go to the templates section where the template is and do Manage -> Push Changes
[09-Jun-2009 15:57:35] <rmatte> then delete the RRD file and try again
[09-Jun-2009 15:57:44] <CamargoBP> k
[09-Jun-2009 15:58:07] <rmatte> what does del/min stand for?
[09-Jun-2009 15:58:14] <CamargoBP> deliveries
[09-Jun-2009 15:58:15] <CamargoBP>
[09-Jun-2009 15:58:24] <CamargoBP> The graph is too small so I did del
[09-Jun-2009 15:58:29] <CamargoBP> Scare my manager.
[09-Jun-2009 15:58:53] <rmatte> That shouldn't affect anything, but you may want to call it "voicemails per minute" or something, or just remove the units all together and see if that changes things
[09-Jun-2009 15:59:01] <rmatte> but try the push changes first
[09-Jun-2009 15:59:11] <rmatte> the collector only updates every hour
[09-Jun-2009 15:59:35] <rmatte> Zenoss does use what's in units to attempt to determine how it should convert values
[09-Jun-2009 15:59:44] <rmatte> like from bits to megabits, and so on
[09-Jun-2009 15:59:59] <rmatte> then again, with the %s removed it shouldn't even try to convert
[09-Jun-2009 16:00:58] <CamargoBP> Interesting... that it does that
[09-Jun-2009 16:01:10] <CamargoBP> Well I'm waiting for the new graph to be created
[09-Jun-2009 16:03:52] <rmatte> k
[09-Jun-2009 16:06:12] <daMaestro> <daMaestro> anyone know what data to enumerate when building a graph for disk space (not percent)
[09-Jun-2009 16:06:12] <daMaestro> <daMaestro> like, raw usage (GB/TB/PB/etc)
[09-Jun-2009 16:06:12] <daMaestro> <daMaestro> I currently use usedBlocks and then: ${here/totalBlocks},/,100,*
[09-Jun-2009 16:06:12] <daMaestro> <daMaestro> however, that does not work when trying a multgraph for multiple devices of storage usage
[09-Jun-2009 16:06:46] <CamargoBP> No go
[09-Jun-2009 16:18:47] <rmatte> daMaestro: no idea
[09-Jun-2009 16:28:13] <CamargoBP> Argh this graph thing is driving me crazy.
[09-Jun-2009 16:34:49] <rmatte> no idea why that graph is not working at this point
[09-Jun-2009 16:34:54] <rmatte> the %s should have fixed it
[09-Jun-2009 16:57:45] <cgibbons> uhoh
[09-Jun-2009 17:00:59] <mrayzenoss> yes cgibbons, that means you
[09-Jun-2009 17:02:24] <cgibbons> sure, pick on the guy that doesn't know anything
[09-Jun-2009 17:02:28] <cgibbons> JP said he wanted to do it again
[09-Jun-2009 19:00:44] <daMaestro> well... took me a little while but i figured it out
[09-Jun-2009 19:00:47] <daMaestro> usedBlocks_usedBlocks
[09-Jun-2009 19:01:01] <daMaestro> with an RPN of: ${here/blockSize},*,1073741824,/
[09-Jun-2009 19:01:06] <daMaestro> that will give you GB used
[09-Jun-2009 19:01:28] <daMaestro> and then you can stack multiple devices to get a birds eye of what is going on across storage devices
[09-Jun-2009 19:23:06] daMaestro|isBack is now known as daMaestro
[09-Jun-2009 21:09:40] timburke_ is now known as timburke
[09-Jun-2009 22:46:47] <centyx> Hi. I admit I haven't read the entire administrator's guide, but only picked through it, so forgive me if this question is answered therein. If it is, just tell me so please. Anyhow, using ZenOSS Core 2.4.1 how would I go about associating IP Services with a device class by default, so that when I create a new device under that device class it has those services monitored automatically, or something similar? Is there a way?
[10-Jun-2009 00:00:42] [disconnected at Wed Jun 10 00:00:42 2009]
[10-Jun-2009 00:00:42] [connected at Wed Jun 10 00:00:42 2009]
[10-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[10-Jun-2009 03:17:47] <gpant> hi, I have installed the HPMonitor v2.0.0 and the AdvDetail 1.1 zenpack, but when I go to the collector plugins for a device I cant see any of the HP plugins, (I do have 2 collectors called HPCPUMAP and HPDeviceMap which I think are completely different)
[10-Jun-2009 03:54:49] <blackgraz> []
[10-Jun-2009 05:27:15] <centyx> /c
[10-Jun-2009 06:57:21] <bwest> hello
[10-Jun-2009 06:57:59] <bwest> has anyone had success using winexe
[10-Jun-2009 08:31:14] <cgibbons> we used to use it internal to zenoss itself before 2.3.x, bwest. what's the question about it?
[10-Jun-2009 08:36:08] <rocket> morning cgibbons
[10-Jun-2009 08:36:16] <cgibbons> howdy
[10-Jun-2009 08:37:51] <rocket> I have been busy busy busy ... Extended the linuxmonitor zenpack .. and then have created a suse and gentoo one on top of that
[10-Jun-2009 09:01:42] <rmatte> why make zenpacks for different linux distros?
[10-Jun-2009 09:01:50] <rmatte> what additional things does it monitor?
[10-Jun-2009 09:02:12] <rocket> rmatte: the only thing it adds is the package management stuff
[10-Jun-2009 09:02:19] <rmatte> ah
[10-Jun-2009 09:02:24] <rmatte> that's cool though
[10-Jun-2009 09:02:26] <rocket> eg gentoo's is different then redhats etc
[10-Jun-2009 09:02:40] <rocket> so you can monitor just fine without them
[10-Jun-2009 09:02:41] <rmatte> yeh, portage, yum, aptitude, etc...
[10-Jun-2009 09:03:02] <rocket> but if you care at all about what packages installed you would need a specific one for the distro
[10-Jun-2009 09:03:30] <rmatte> yeh
[10-Jun-2009 09:03:51] <rmatte> if you were to make one for ubuntu, that would probably be hugely popular
[10-Jun-2009 09:03:55] <rmatte> or is there one already?
[10-Jun-2009 09:04:05] <rocket> nope thats next on my list ..
[10-Jun-2009 09:04:11] <rmatte> (and it could be used for debian as well)
[10-Jun-2009 09:04:16] <rocket> probably have that one done tonight ..
[10-Jun-2009 09:04:21] <rmatte> cool
[10-Jun-2009 09:04:51] <rocket> rmatte: I would make a debian one just because there are a few uniq things between the two ..
[10-Jun-2009 09:05:00] <rmatte> awesome, the water cooler is actually full for once and I have the only cup that was left, more water for meeee
[10-Jun-2009 09:05:04] <rocket> rmatte: its just a matter of display preference though
[10-Jun-2009 09:05:14] <rmatte> ah
[10-Jun-2009 09:05:25] <rmatte> they both use aptitude, so I figured they'd be very similar
[10-Jun-2009 09:05:33] <rocket> eg manufacturer = debian
[10-Jun-2009 09:05:38] <chudler> perhaps one zenpack that detects them as it runs?
[10-Jun-2009 09:05:39] <rmatte> ah
[10-Jun-2009 09:05:44] <rocket> eg manufacturer = ubuntu ...
[10-Jun-2009 09:06:07] <rocket> chudler: that is a possibility yes .. I probably could do that
[10-Jun-2009 09:06:10] <chudler> in fact, one zenpack per distro sucks for users + all that code duplication?
[10-Jun-2009 09:06:31] <chudler> OTOH, maybe average install would only want one or two.
[10-Jun-2009 09:06:55] <rocket> chudler: thats what I was thinking .. and in reality some of the code is the same
[10-Jun-2009 09:07:00] <rmatte> you could make a script that does a "uname -r | cut -d '-' -f 2"
[10-Jun-2009 09:07:10] <rocket> chudler: but other aspects of parsing the packager info is different
[10-Jun-2009 09:07:14] <rmatte> then configures it based on the return value
[10-Jun-2009 09:07:52] <rocket> rmatte: except I know of distros that dont always return the right info out of uname
[10-Jun-2009 09:08:06] <rocket> rmatte: or if you have a stock kernel compiled etc that wont work
[10-Jun-2009 09:08:27] <chudler> /etc/issue? LSB is strong on this. or at least so I thought.
[10-Jun-2009 09:08:35] <chudler> detecting distro would be the easy part, I suspect.
[10-Jun-2009 09:09:13] <rocket> chudler: yeah .. again I might just do them individually and if someone wants to merge them they are free to do so as well
[10-Jun-2009 09:09:24] <rocket> chudler: we are only using suse in house here etc ..
[10-Jun-2009 09:09:30] <rmatte> you could look in /etc for release files
[10-Jun-2009 09:09:35] <rmatte> but some distros have more than one
[10-Jun-2009 09:09:55] <rocket> rmatte: thats what I am looking at for suse to detect opensuse from suse enterprise for example
[10-Jun-2009 09:09:58] <rmatte> sabayon has the gentoo-release file as well as the sabayon-release
[10-Jun-2009 09:10:31] <rocket> rmatte: my main purpose is to get some examples out there so others can come up with something more complex
[10-Jun-2009 09:10:43] <rmatte> yeh
[10-Jun-2009 09:11:09] <rocket> especially since I have no desire to install each distro and figure out how it would go together .. etc
[10-Jun-2009 09:11:41] <rmatte> that's why you need to find a common ground between them
[10-Jun-2009 09:11:43] <rocket> rmatte: but my zenpacks have the __init__.py configured to setup the modeller for you, the test cases defined and custom packagers
[10-Jun-2009 09:11:59] <rmatte> test it on a few different distros and call those the "officially supported" ones
[10-Jun-2009 09:12:03] <rocket> err not custom packagers .. modellers
[10-Jun-2009 09:12:29] <rocket> rmatte: except as a volunteer i dont really want to officially support any of them
[10-Jun-2009 09:12:44] <rmatte> lol
[10-Jun-2009 09:13:10] <rmatte> ok, so say something like "This Zenpack is known to work with Suse, OpenSuse, Ubuntu, Debian, Gentoo..."
[10-Jun-2009 09:13:30] <rocket>
[10-Jun-2009 09:13:34] <rocket> morning matt
[10-Jun-2009 09:13:46] <rmatte> gmorning Matt
[10-Jun-2009 09:14:07] <mrayzenoss> morning guys
[10-Jun-2009 09:14:52] <rmatte> "cat /etc/*-rel*" seems to work for detecting the version
[10-Jun-2009 09:15:51] <mrayzenoss> today should be a good day for ZenPacks. No meetings on the calendar and 5 in the queue already
[10-Jun-2009 09:39:39] <cgibbons> I don't suppose anyone here uses the Komodo IDE for Python?
[10-Jun-2009 10:05:16] <etank> has anyone gotten the "Perfmon" and "SQL 2000/2005 Server Performance" ZenPacks to work?
[10-Jun-2009 10:05:24] <etank> i am getting just blank graphs
[10-Jun-2009 10:07:25] <etank> when i run the commands manually i get "ERROR: Failed to install service winexesvc - NT code 0x00000424"
[10-Jun-2009 10:08:29] <etank> i tried what was said on the forums about replacin the winexe 0.80 version with the 0.90 version but then i get "INTERNAL ERROR: Signal 11 in pid 7915 (4.0.0tp4-SVN-build-10231)"
[10-Jun-2009 10:11:36] <rocket> cgibbons: I dont but I am curious is that what you are using?
[10-Jun-2009 10:11:58] <cgibbons> most of the team uses eclipse, a couple of the oldies use vim
[10-Jun-2009 10:12:10] <cgibbons> everything is far from perfect sadly
[10-Jun-2009 10:12:39] <chudler> interesting. /me really needs to get on the eclipse bandwagon someday
[10-Jun-2009 10:13:20] <rocket> cgibbons: hrmm if I use vim does that make me an oldie? ;p
[10-Jun-2009 10:14:47] <cgibbons> by oldie i meant the peeps working on zenoss the longest
[10-Jun-2009 10:14:56] <rocket> ah I see ..
[10-Jun-2009 10:15:52] <etank> im using Zenoss Core 2.4.1 with Perfmon 1.2 and SQL Server Performance 1.2
[10-Jun-2009 10:16:06] <rmatte> is there a "get" function that I can use in a custom report to grab the first time a device was seen?
[10-Jun-2009 10:16:13] <rmatte> essentially the commission date
[10-Jun-2009 10:17:15] <rocket> rmatte: looking
[10-Jun-2009 10:18:10] <rocket> you looking for the "First Seen" variable on the status page?
[10-Jun-2009 10:18:52] <rmatte> essentially
[10-Jun-2009 10:18:57] <rmatte> but I need to use it in a report
[10-Jun-2009 10:19:19] <rocket> rmatte: sure just a sec let me find what is calling that from the page template
[10-Jun-2009 10:19:22] <rmatte> so far I'm using things like getId, getManageIp, getHWManufacturerName, etc...
[10-Jun-2009 10:19:29] <rmatte> thanks
[10-Jun-2009 10:20:14] <rocket> getCreatedTimeString
[10-Jun-2009 10:20:20] <rmatte> thanks
[10-Jun-2009 10:20:23] <rocket>
[10-Jun-2009 10:20:39] <rocket> for your info if you find something like that in the web interface
[10-Jun-2009 10:20:56] <rocket> the easiest way is to look for the page template
[10-Jun-2009 10:21:08] <rocket> eg ZenModel/skins/deviceStatus.pt
[10-Jun-2009 10:21:14] <rocket> grep for "First Seen"
[10-Jun-2009 10:21:28] <rocket> or look for that in the page template
[10-Jun-2009 10:21:41] <rocket> and it usually has something like here/getCreatedTimeString
[10-Jun-2009 10:21:43] <rmatte> ah, cool
[10-Jun-2009 10:21:47] <rmatte> didn't know it was that easy
[10-Jun-2009 10:21:56] <rocket> which gives you an idea of what was getting called
[10-Jun-2009 10:23:53] <rmatte> is there any easy way to pull values from RRD in to a template, but not as a graph, just as text?
[10-Jun-2009 10:24:02] <rmatte> in to a report I mean
[10-Jun-2009 10:24:17] <venturaville> here.getRRDValue('Name')
[10-Jun-2009 10:24:54] <venturaville> I don't know if that works in all of the fields though...
[10-Jun-2009 10:25:22] <rmatte> ah
[10-Jun-2009 10:25:42] <rocket> rmatte: there are a few examples .. just a sec
[10-Jun-2009 10:26:07] <venturaville> here.cacheRRDValue('Name') if you don't need it up to date.....
[10-Jun-2009 10:26:49] <rmatte> cool
[10-Jun-2009 10:27:11] <rocket> venturaville: yahtzee!!! .. that explains the one issue I was having with one of my functions .. it was always out of date!
[10-Jun-2009 10:27:21] <venturaville> indeed
[10-Jun-2009 10:27:36] <rocket> was looking for a different RRDValue function but didnt stumble across it
[10-Jun-2009 10:27:45] <venturaville> in zendmd:
[10-Jun-2009 10:27:59] <rocket> rmatte: here is an example
[10-Jun-2009 10:28:01] <rocket> def usedInodes(self, default = None):
[10-Jun-2009 10:28:01] <rocket> """
[10-Jun-2009 10:28:01] <rocket> Return the number of used inodes stored in the filesystem's rrd file
[10-Jun-2009 10:28:01] <rocket> """
[10-Jun-2009 10:28:04] <rocket> inodes = self.cacheRRDValue('usedInodes_usedInodes', default)
[10-Jun-2009 10:28:06] <rocket> if inodes is not None:
[10-Jun-2009 10:28:09] <rocket> return long(inodes)
[10-Jun-2009 10:28:11] <rocket> else:
[10-Jun-2009 10:28:12] <venturaville> d = devices.findDevice('mon8zen2')
[10-Jun-2009 10:28:14] <rocket> return None
[10-Jun-2009 10:28:33] <venturaville> d.getRRDDataPoints() ..... very long list of points
[10-Jun-2009 10:28:49] <venturaville> the name after the datapoints/ in each is what you want
[10-Jun-2009 10:28:50] <venturaville> like so:
[10-Jun-2009 10:29:04] <venturaville> >>> d.getRRDValue('ssCpuSystem')
[10-Jun-2009 10:29:04] <venturaville> 4.0
[10-Jun-2009 10:29:59] <venturaville> I'm not a big fan of the tales syntax, so I tend to fall back on the python syntax in templates.......
[10-Jun-2009 10:30:10] <venturaville> makes it easier to debug in zendmd
[10-Jun-2009 10:31:29] <rmatte> right
[10-Jun-2009 10:32:27] <rmatte> any way to pull the average value instead of the current?
[10-Jun-2009 10:32:31] <rmatte> or is that the cached value?
[10-Jun-2009 10:32:49] <rmatte> and any way to pull the max value?
[10-Jun-2009 10:34:04] <venturaville> rmatte: not certain ...let me look
[10-Jun-2009 10:34:09] <rmatte> thanks
[10-Jun-2009 10:34:25] <venturaville> cached = sticks around in zope cache for a bit I believe
[10-Jun-2009 10:34:31] <venturaville> don't believe it has anything to do with the average
[10-Jun-2009 10:34:43] <venturaville> it is a way to reduce the lookup overhead when used widely
[10-Jun-2009 10:35:17] <venturaville> getRRDContextData
[10-Jun-2009 10:35:17] <venturaville> getRRDDataPoint
[10-Jun-2009 10:35:17] <venturaville> getRRDDataPoints
[10-Jun-2009 10:35:17] <venturaville> getRRDFileName
[10-Jun-2009 10:35:17] <venturaville> getRRDNames
[10-Jun-2009 10:35:18] <venturaville> getRRDPaths
[10-Jun-2009 10:35:20] <venturaville> getRRDSum
[10-Jun-2009 10:35:22] <venturaville> getRRDTemplate
[10-Jun-2009 10:35:24] <venturaville> getRRDTemplateByName
[10-Jun-2009 10:35:26] <venturaville> getRRDTemplateName
[10-Jun-2009 10:35:28] <venturaville> getRRDTemplates
[10-Jun-2009 10:35:30] <venturaville> getRRDValue
[10-Jun-2009 10:35:32] <venturaville> getRRDValues
[10-Jun-2009 10:35:39] <venturaville> gleaned from: print '\n'.join(dir(d))
[10-Jun-2009 10:36:12] <rocket> actually you might need fetchRRDValue
[10-Jun-2009 10:36:14] <rocket> def fetchRRDValue(self, dpname, cf, resolution, start, end=""):
[10-Jun-2009 10:36:27] <rocket> where cf is AVERAGE or MAX
[10-Jun-2009 10:36:41] <venturaville> not seeing that one listed in the device objects
[10-Jun-2009 10:36:58] <rocket> ah i see its used by getRRDValues
[10-Jun-2009 10:37:12] <rocket> def getRRDValues(self, dsnames, start=None, end=None, function="LAST",
[10-Jun-2009 10:37:12] <rocket> format="%.2lf", extraRpn=""):
[10-Jun-2009 10:37:47] <rmatte> hmmm, now to figure out what get statement I can use to grab the device class
[10-Jun-2009 10:38:08] <venturaville> http://docs.huihoo.com/zenoss/api/2.1/private/ZenModel.RRDView.RRDView-class.html
[10-Jun-2009 10:38:40] <venturaville> getRRDSum(self, points, start, end, function)
[10-Jun-2009 10:38:40] <venturaville> Return a some of listed datapoints.
[10-Jun-2009 10:38:44] <venturaville> I think that might be what you want
[10-Jun-2009 10:38:49] <venturaville> (and me for that matter)
[10-Jun-2009 10:39:26] <venturaville> getRRDValue(self, dsname, start, end, function)
[10-Jun-2009 10:39:26] <venturaville> Return a single rrd value from its file using function.
[10-Jun-2009 10:39:29] <venturaville> nope that one is
[10-Jun-2009 10:39:32] <rocket> venturaville: that just calls getRRDValues and adds them together
[10-Jun-2009 10:39:34] <venturaville> it just has more optional arguments
[10-Jun-2009 10:40:32] <rocket> venturaville: getRRDValue just calls getRRDValues with a single datapoint
[10-Jun-2009 10:40:44] <venturaville> yep
[10-Jun-2009 10:40:55] <venturaville> convenience functions.....
[10-Jun-2009 10:42:23] <venturaville> something like this maybe:
[10-Jun-2009 10:42:44] <rmatte> bah
[10-Jun-2009 10:42:46] <venturaville> getRRDValue('ssCpuSystem',None,None,"AVERAGE")
[10-Jun-2009 10:42:51] <rmatte> trying to find the get value for Product Class
[10-Jun-2009 10:43:02] <rmatte> though it was getProductClass apparently not
[10-Jun-2009 10:44:16] <venturaville> >>> d.deviceClass()
[10-Jun-2009 10:44:16] <venturaville> <DeviceClass at /zport/dmd/Devices/Server/Linux/devices/mon8zen2/deviceClass/Linux>
[10-Jun-2009 10:44:18] <venturaville> this you mean?
[10-Jun-2009 10:44:34] <rmatte> yessir
[10-Jun-2009 10:44:49] <venturaville> >>> d.getDeviceClassName()
[10-Jun-2009 10:44:49] <venturaville> '/Server/Linux'
[10-Jun-2009 10:44:51] <venturaville> or in string form
[10-Jun-2009 10:44:53] <rocket> venturaville would this be more readable?
[10-Jun-2009 10:45:01] <venturaville> I searched for a while to find that one too
[10-Jun-2009 10:45:04] <rmatte> ah, was missing Name on the end when I tried it earlier
[10-Jun-2009 10:45:06] <rocket> getRRDValue('ssCpuSystem', function="AVERAGE")
[10-Jun-2009 10:45:16] <rocket> getRRDValue('ssCpuSystem', function="MAX")
[10-Jun-2009 10:45:19] <rocket> etc ..
[10-Jun-2009 10:45:31] <venturaville> you don't need the 2 start and end values?
[10-Jun-2009 10:46:07] <rocket> venturaville: shouldnt need it if you are naming the variable coming in
[10-Jun-2009 10:46:15] <venturaville> ah true
[10-Jun-2009 10:46:29] <venturaville> I'm getting my scripting languages mixed up
[10-Jun-2009 10:47:00] <rocket> give it a try
[10-Jun-2009 10:47:02] <rocket>
[10-Jun-2009 10:47:13] <rocket> >>> d = devices.findDevice('192.168.31.3')
[10-Jun-2009 10:47:13] <rocket> >>> d.getRRDValue('ssCpuSystem', function="AVERAGE")
[10-Jun-2009 10:47:13] <rocket> 2.7400000000000002
[10-Jun-2009 10:47:32] <rocket> >>> d.getRRDValue('ssCpuSystem', function="MAXIMUM")
[10-Jun-2009 10:47:32] <rocket> 32.210000000000001
[10-Jun-2009 10:47:35] <venturaville> >>> d.getRRDValue('ssCpuSystem',function="LAST")
[10-Jun-2009 10:47:36] <venturaville> 6.2199999999999998
[10-Jun-2009 10:47:36] <venturaville> >>> d.getRRDValue('ssCpuSystem',function="AVERAGE")
[10-Jun-2009 10:47:36] <venturaville> 4.7599999999999998
[10-Jun-2009 10:47:36] <venturaville> >>> d.getRRDValue('ssCpuSystem',function="MAXIMUM")
[10-Jun-2009 10:47:36] <venturaville> 6.25
[10-Jun-2009 10:47:38] <venturaville> sweet
[10-Jun-2009 10:47:46] <rmatte> so would I do getCacheRRDValue or getcacheRRDValue ?
[10-Jun-2009 10:47:46] <rocket>
[10-Jun-2009 10:47:50] <venturaville> that will help me fix a module I've been working on
[10-Jun-2009 10:48:00] <venturaville> just cacheRRDValue()
[10-Jun-2009 10:48:23] <rmatte> I'm looking for the function that I can actually use in a report
[10-Jun-2009 10:48:45] <venturaville> >>> print '\n'.join(dir(devices.findDevice('mon8zen2')))
[10-Jun-2009 10:48:54] <venturaville> print that out using some random device
[10-Jun-2009 10:49:05] <venturaville> you'll save yourself a lot of headaches......
[10-Jun-2009 10:49:31] <rocket>
[10-Jun-2009 11:21:45] <rmatte> If you set a device's model info by hand, will it write over it if you remodel the device?
[10-Jun-2009 11:21:59] <rmatte> manufacturer/model info I mean
[10-Jun-2009 11:42:41] <rhettlap> no
[10-Jun-2009 12:04:51] <rmatte> k thanks
[10-Jun-2009 12:50:14] <rmatte> .
[10-Jun-2009 13:00:36] <etank> is anyone having issues with winexe ?
[10-Jun-2009 13:01:15] <etank> ERROR: Failed to install service winexesvc - NT code 0x00000424 <-- thats what im getting
[10-Jun-2009 13:02:41] <etank> and using the 0.90 version just gives different errors
[10-Jun-2009 13:23:24] <rmatte> this page loading issue in zenoss can be a real pain sometimes
[10-Jun-2009 13:23:45] <rmatte> it's even worse in 2.4.1 than it was in 2.3.3
[10-Jun-2009 14:01:01] * Bryanstein is away:
[10-Jun-2009 14:01:45] * Bryanstein is back (gone 00:00:30)
[10-Jun-2009 14:20:44] <etank> i guess that no one uses the community Perfmon and SQL Server ZenPacks ?
[10-Jun-2009 14:32:51] <mrayzenoss> etank: I know there are at least 2 users, the original author and the one who updated them both
[10-Jun-2009 14:33:12] <mrayzenoss> that said, I don't have access to SQL Server
[10-Jun-2009 14:42:00] <etank> mrayzenoss: i wish i didn't
[10-Jun-2009 14:42:17] <etank> im just not having any luck getting those two to work at all
[10-Jun-2009 15:10:38] <javierglaz> hello
[10-Jun-2009 15:11:04] <javierglaz> is this a zenoss support channel?
[10-Jun-2009 15:12:49] <rmatte> is there any way to set a recurring maintenance window for a device?
[10-Jun-2009 15:13:13] <rmatte> we have a scheduled device reboot every week, we need to set a recurring maintenance window so that we don't get a flood of events during the reboot.
[10-Jun-2009 15:15:01] <rmatte> oh nevermind, there's a repeat option
[10-Jun-2009 15:15:40] <javierglaz> I have a problem with zensyslog
[10-Jun-2009 15:16:02] <javierglaz> I am monitoring cisco devices, and the logging host is currently pointing to the zenoss box
[10-Jun-2009 15:16:32] <rmatte> but you're not seeing any syslog events coming in?
[10-Jun-2009 15:16:39] <javierglaz> if I issue a tail -f $ZENHOME/log/zensyslog.log
[10-Jun-2009 15:16:42] <javierglaz> I see the events
[10-Jun-2009 15:16:57] <javierglaz> and I see the regex match also
[10-Jun-2009 15:17:13] <javierglaz> but I don't get any events in the web console
[10-Jun-2009 15:17:20] <rmatte> Zenoss only really translates syslog messages to an actual Zenoss event when the syslog event comes in with a certain level
[10-Jun-2009 15:17:37] <javierglaz> can I paste one event I get?
[10-Jun-2009 15:17:41] <rmatte> if it's a lower severity event, you won't see it
[10-Jun-2009 15:17:42] <rmatte> yes
[10-Jun-2009 15:18:25] <rmatte> Go to Event Manager and check the "Default Syslog Priority" setting
[10-Jun-2009 15:19:05] <rmatte> we currently have ours set at 3, with 1 being the highest severity and 7 being the lowest
[10-Jun-2009 15:19:46] <javierglaz> 2009-06-10 17:19:22 DEBUG zen.Syslog: tag match: {'eventClassKey': 'LINK-3-UPDOWN', 'component': 'LINK', 'summary': 'Interface Loopback999, changed state to up'}
[10-Jun-2009 15:19:46] <javierglaz> 2009-06-10 17:19:22 DEBUG zen.ZenSyslog: Queueing event {'firstTime': 1244665162.0504229, 'severity': 4, 'facility': 'local7', 'eventClassKey': 'LINK-3-UPDOWN', 'component': 'LINK', 'agent': 'zensyslog', 'summary': 'Interface Loopback999, changed state to up', 'priority': 3, 'manager': 'localhost', 'eventGroup': 'syslog', 'device': '172.16.100.25', 'lastTime': 1244665162.0504229, 'ipAddress': '172.16.100.25', 'monitor': 'localhost'}
[10-Jun-2009 15:19:46] <javierglaz> 2009-06-10 17:19:22 DEBUG zen.ZenSyslog: Total of 1 queued events
[10-Jun-2009 15:20:05] <rmatte> ok, so the severity is 4
[10-Jun-2009 15:20:16] <rmatte> what is your Default Syslog Priority set to in Event Manager?
[10-Jun-2009 15:20:23] <javierglaz> let me check
[10-Jun-2009 15:20:26] <rmatte> k
[10-Jun-2009 15:20:34] <javierglaz> 3
[10-Jun-2009 15:20:49] <rmatte> ok, so Zenoss won't create events for syslogs below 3 for you right now
[10-Jun-2009 15:20:58] <rmatte> if you bump it to 4 or 5 then you should see those events come in
[10-Jun-2009 15:21:21] <javierglaz> I see.... let me try
[10-Jun-2009 15:21:25] <rmatte> k
[10-Jun-2009 15:23:59] <javierglaz> I've raised it to 5 but still no luck....
[10-Jun-2009 15:24:32] <javierglaz> ah but I didn't restart the process... one sec
[10-Jun-2009 15:24:37] <rmatte> k
[10-Jun-2009 15:25:05] <rmatte> you might also have to push the changes to the collector
[10-Jun-2009 15:25:10] <rmatte> but I doubt it
[10-Jun-2009 15:26:43] <javierglaz> still no luck...
[10-Jun-2009 15:27:07] <javierglaz> how do you push the changes to the collector?
[10-Jun-2009 15:27:19] <rmatte> Manage -> Push Changes
[10-Jun-2009 15:27:40] <rmatte> do it under the Devices section
[10-Jun-2009 15:29:58] <rmatte> it looks like Zenoss is trying to create the event but queueing it instead
[10-Jun-2009 15:30:00] <rmatte> strange
[10-Jun-2009 15:30:26] <javierglaz> still not working.... I am getting some events with severity 2 now.... so it should have worked with the default severity of 3 I had before
[10-Jun-2009 15:30:46] <rmatte> hmmm
[10-Jun-2009 15:30:55] <javierglaz> so it must be something else.... the strange message is the queuing one
[10-Jun-2009 15:31:27] <rmatte> yeh
[10-Jun-2009 15:31:34] <rmatte> not sure why it would be "queuing" an event
[10-Jun-2009 15:33:07] <javierglaz> unfortunately I couldn't find nothing in google with this
[10-Jun-2009 15:34:26] <rmatte> same here
[10-Jun-2009 15:38:23] <javierglaz> I also see some heartbeat messages in that log
[10-Jun-2009 15:38:36] <javierglaz> but they don't appear in the web console
[10-Jun-2009 15:38:58] <javierglaz> 2009-06-10 17:38:53 DEBUG zen.ZenSyslog: Queueing event {'manager': 'localhost', 'timeout': 180, 'device': 'localhost', 'eventClass': '/Heartbeat', 'component': 'zensyslog', 'agent': 'zensyslog'}
[10-Jun-2009 15:38:58] <javierglaz> 2009-06-10 17:38:53 DEBUG zen.ZenSyslog: Total of 1 queued events
[10-Jun-2009 15:42:51] <javierglaz> does anyone knows what this message means? DEBUG zen.ZenSyslog: Total of 1 queued events
[10-Jun-2009 15:43:14] <chudler> what version of zenoss? I couldn't find this in the code anywhere.
[10-Jun-2009 15:43:44] <javierglaz> 2.4
[10-Jun-2009 15:44:12] <chudler> oh I did find it, I was looking in an older one :-)
[10-Jun-2009 15:44:18] <javierglaz> 2.4.1
[10-Jun-2009 15:44:50] <chudler> AFAICT, it is a normal course of events for the event to be queued before processing. But take that FWIW, I just glanced at it.
[10-Jun-2009 15:45:25] <chudler> someone somewhere once published a detailed events processing document (Jane Curry?).
[10-Jun-2009 15:45:56] <mrayzenoss> http://www.skills-1st.co.uk/papers/jcurry.html
[10-Jun-2009 15:46:06] <mrayzenoss> Zenoss Event Management
[10-Jun-2009 15:47:13] <javierglaz> downloading
[10-Jun-2009 15:49:42] <javierglaz> what should I be looking at after comfirming that the logs are hitting the zenoss box, and zenoss is parsing them succesfully?
[10-Jun-2009 15:52:03] <javierglaz> in the document says that the last line should be sending event
[10-Jun-2009 15:52:10] <javierglaz> instead of that I get queueing event
[10-Jun-2009 16:13:38] <mrayzenoss> anyone looking for a Zenoss job in the bay area?
[10-Jun-2009 16:15:09] <javierglaz> is someone using zenoss 2.4.1 as syslog server for cisco devices?
[10-Jun-2009 16:20:35] <magnachef__> is anyone around that has any experience with monitoring websites/vhosts with HttpMonitor
[10-Jun-2009 16:31:15] <magnachef__> well, if anyone can take a look at an issue I'm having, I'd appreciate it: http://forums.zenoss.com/viewtopic.php?t=10014
[11-Jun-2009 00:00:42] [disconnected at Thu Jun 11 00:00:42 2009]
[11-Jun-2009 00:00:42] [connected at Thu Jun 11 00:00:42 2009]
[11-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[11-Jun-2009 01:53:49] <frozty_sa> morning all
[11-Jun-2009 03:23:49] <sergeymasushko> how can I debug cheking of availability http/https ports on devices?
[11-Jun-2009 03:27:53] <tehhobbit> debug ?
[11-Jun-2009 04:33:52] <stefanlsd> I am trying to write a zenoss script that runs setManageIp() . It works from the zendmd interface, but not the script. In the interface, after a loop i push enter (tells it loop is finished or something), this doesnt work from script? Any ideas?
[11-Jun-2009 04:39:40] <stefanlsd> needed an extra blank line in the script... thanks for all the help
[11-Jun-2009 04:40:37] <sergeymasushko> tehhobbit: I tried to figure out lies zenoss or not
[11-Jun-2009 08:04:50] <ckrough> stefanlsd: how does your script interface with zendmd?
[11-Jun-2009 08:04:59] <ckrough> bah, hes gone.
[11-Jun-2009 08:06:33] <ckrough> why would a device have an "unknown" uptime
[11-Jun-2009 08:13:06] <rocket> Jontler: umn .. hmmm I went there
[11-Jun-2009 08:13:32] <Jontler> it's a nice school
[11-Jun-2009 08:14:16] <rocket> Yup .. its been a few years now ..
[11-Jun-2009 08:14:25] <Jontler> funny how that happens
[11-Jun-2009 08:15:45] <rocket> yeah
[11-Jun-2009 08:15:49] <Jontler> we're looking at using zenoss to replace nagios here at the math department
[11-Jun-2009 08:15:51] <rocket> what are you in for
[11-Jun-2009 08:16:04] <Jontler> ha, well, i just recently graduated with my math degree
[11-Jun-2009 08:16:17] <Jontler> I'm part of the IT staff at the math department, though
[11-Jun-2009 08:16:43] <rocket> Jontler: well zenoss can use nagios plugins so there is a bonus there
[11-Jun-2009 08:16:48] <Jontler> indeed
[11-Jun-2009 08:17:15] <Jontler> it can also do most of the things we use nagios for now natively, which is also a bonus
[11-Jun-2009 08:17:21] <rocket> Jontler: I am local if you have questions ...
[11-Jun-2009 08:17:28] <rocket> Minnetonka
[11-Jun-2009 08:17:35] <Jontler> nice
[11-Jun-2009 08:18:01] <Jontler> well, my question right now is involving the performance templates for process monitoring
[11-Jun-2009 08:18:09] <rocket> ok
[11-Jun-2009 08:18:19] <Jontler> it seems that each process being monitored defaults to using the standard OSProcess template
[11-Jun-2009 08:18:32] <Jontler> id like to change that, but can't seem to figure out how
[11-Jun-2009 08:18:53] <rocket> so basically a new OSProcess Template?
[11-Jun-2009 08:18:58] <Jontler> yep
[11-Jun-2009 08:21:10] <rocket> ok .. well I believe you will have to apply your new osprocess template per device .. its not granular to the per process leve yet
[11-Jun-2009 08:21:22] <rocket> so what you could do is create a new device organizer
[11-Jun-2009 08:21:32] <rocket> for the device thats unique
[11-Jun-2009 08:21:49] <rocket> and copy the template there and update it from there
[11-Jun-2009 08:21:56] <rocket> does that make sense?
[11-Jun-2009 08:21:58] <Jontler> yea
[11-Jun-2009 08:22:03] <Jontler> it's what i tried yesterday
[11-Jun-2009 08:22:10] <rocket> didnt work?
[11-Jun-2009 08:22:24] <Jontler> basically, ive got a lab of machines in a device organizer now
[11-Jun-2009 08:22:48] <rocket> it will have to have the same name as the original OSProcess template because its unique
[11-Jun-2009 08:22:50] <Jontler> i created a new template (MATLAB, in this case) to monitor the matlabl process
[11-Jun-2009 08:22:53] <Jontler> oh
[11-Jun-2009 08:22:58] <rocket> and you must not bind it
[11-Jun-2009 08:22:59] <Jontler> i wasnt clear
[11-Jun-2009 08:23:13] <Jontler> actually
[11-Jun-2009 08:23:15] <Jontler> hm
[11-Jun-2009 08:23:21] <Jontler> i think i understand now
[11-Jun-2009 08:23:24] <Jontler> one moment
[11-Jun-2009 08:27:00] <Jontler> so, since I cannot have a template per process, I have to define one template for a variety of processes, and then apply that to the devices that are are using those processes
[11-Jun-2009 08:27:21] <rocket> basically yes
[11-Jun-2009 08:27:26] <Jontler> and that template must have the same name, since each process always uses the OSProcess template
[11-Jun-2009 08:28:11] <rocket> yep
[11-Jun-2009 08:28:27] * Jontler has a feature request
[11-Jun-2009 08:28:30] <Jontler>
[11-Jun-2009 08:28:37] <Jontler> i think i can work with this though
[11-Jun-2009 08:28:40] <rocket> or if you want you can create a totally custom datacollector
[11-Jun-2009 08:29:15] <rocket> actually here is another way to do it
[11-Jun-2009 08:31:30] <rocket> Jontler: how good is your scripting?
[11-Jun-2009 08:31:39] <Jontler> passable
[11-Jun-2009 08:31:45] <Jontler> i can make things work
[11-Jun-2009 08:31:53] <rocket> another way to do it is to look at the dev guide
[11-Jun-2009 08:31:57] <rocket> create a new template
[11-Jun-2009 08:32:06] <rocket> add a command datapoint
[11-Jun-2009 08:32:29] <rocket> in that command datapoint you connect to the machine you need information from via a nagios type script
[11-Jun-2009 08:32:46] <rocket> save the variables as datapoints
[11-Jun-2009 08:33:00] <rocket> and then create a new graph based on these new datapoints
[11-Jun-2009 08:33:22] <rocket> final step is to bind the template to the device organizer that you need this to run against
[11-Jun-2009 08:33:41] <rocket> this will create a graph on the performance tab
[11-Jun-2009 08:33:48] <rocket> make sense?
[11-Jun-2009 08:34:01] <Jontler> yea, you're skirting the process portion of zenoss altogether
[11-Jun-2009 08:34:12] <rocket> yes
[11-Jun-2009 08:34:23] <rocket> the process portion of zenoss is semi weak
[11-Jun-2009 08:34:35] <Jontler> so ive noticed
[11-Jun-2009 08:34:44] <rocket> eg if you turn on monitoring of ssh for example it is on for all systems
[11-Jun-2009 08:34:57] <rocket> process monitoring is pretty global in zenoss
[11-Jun-2009 08:35:18] <Jontler> yea, i added sshd as a process yesterday
[11-Jun-2009 08:35:28] <Jontler> there were a lot of sshd processes
[11-Jun-2009 08:36:00] <Jontler> i assume there are plans in the future to expand on the process monitoring
[11-Jun-2009 08:36:21] <rocket> yeah from what I have heard
[11-Jun-2009 08:36:37] <Jontler> well that's neat
[11-Jun-2009 08:36:40] <rocket> I am just a community member so I couldnt say for sure .. but I know others have had issues with it
[11-Jun-2009 08:36:54] <Jontler> im just trying to get a handle on the different parts of zenoss
[11-Jun-2009 08:37:02] <Jontler> see what might be useful and what might not be
[11-Jun-2009 08:37:17] <Jontler> it already looks like it can take over for nagios without missing a beat
[11-Jun-2009 08:37:28] <rocket> take a look at the nagios section etc thats probably closer to what you need
[11-Jun-2009 08:37:46] <rocket> and/or the dev guide to add ssh based monitoring stuff
[11-Jun-2009 08:37:57] <Jontler> ive already got the ssh based monitoring rolling
[11-Jun-2009 08:38:27] <rocket> Jontler: I am talking more about sshing into a system and gathering its stats for something
[11-Jun-2009 08:38:48] <rocket> eg I have created code that gathers df output and parses it in zenoss etc
[11-Jun-2009 08:39:01] <Jontler> ive installed the zenoss-plugins on our workstations
[11-Jun-2009 08:39:11] <Jontler> that seems to do a good job of allowing data to be gathered natively
[11-Jun-2009 08:39:14] <ckrough> rocket: you can monitor storage space through snmp
[11-Jun-2009 08:39:36] <rocket> ckrough: yes you can unless you have an unstable snmp implementation for example
[11-Jun-2009 08:39:47] <rocket> ckrough: for example aix's snmp daemon really sucks
[11-Jun-2009 08:39:48] <ckrough> heh, that would be an issue
[11-Jun-2009 08:39:49] <ckrough> what os
[11-Jun-2009 08:39:50] <ckrough> ah
[11-Jun-2009 08:40:17] <rocket> ckrough: there is a patch for it I havent tried yet but at the moment snmp crashes at least once a week
[11-Jun-2009 08:40:33] <ckrough> rocket: well, nobody likes that
[11-Jun-2009 08:40:37] <rocket> ckrough: so for me zenoss provides another work around
[11-Jun-2009 08:41:06] <rocket> ckrough: df may not be the best example .. but I can also create scripts that gather information out of our backup servers cmdline client
[11-Jun-2009 08:42:34] <ckrough> I need to put together a report that shows devices (/Network) with uptimes of less than x. I built a report in the custom device reports, but it's just too slow. Is there a way to interact with zendmd from outside of zendmd?
[11-Jun-2009 08:42:53] <ckrough> I guess I could load it up in a python script.
[11-Jun-2009 08:43:14] <rocket> ckrough: yeah I think there is an example in the dev guide ..
[11-Jun-2009 08:43:44] <ckrough> that would be perfect. I have trouble finding what I need in the docs.
[11-Jun-2009 08:45:12] <rocket> ckrough: basically you will be creating a python script that is loading the modules you need for zendmd without the full zendmd
[11-Jun-2009 08:45:14] <Jontler> ckrough: me too - hence i came here
[11-Jun-2009 08:45:38] <rocket> http://www.zenoss.com/Members/netdata/use-a-zendmd-script-as-standalone-program/
[11-Jun-2009 08:45:42] <ckrough> rocket: thanks. here's hoping its faster than the built-in report
[11-Jun-2009 08:45:59] <ckrough> rocket: awesome
[11-Jun-2009 08:46:05] <rocket>
[11-Jun-2009 08:46:34] <rocket> np
[11-Jun-2009 08:49:29] <ckrough> is there a way to connect to zendmd on a remote machine?
[11-Jun-2009 08:49:52] <venturaville> btw for the custom device reports you can actually put the python code inside of a plugin and just call it from a report
[11-Jun-2009 08:50:00] <ckrough> I guess REST cals are about as close as you can get
[11-Jun-2009 08:50:18] <ckrough> nice, that might be an improvement
[11-Jun-2009 08:50:24] <venturaville> or put the code in a custom python script in zope, and call it from REST
[11-Jun-2009 08:50:35] <venturaville> again faster than REST/XML-RPC
[11-Jun-2009 08:51:48] <venturaville> look at Products/ZenReports/plugins ....
[11-Jun-2009 08:53:30] <venturaville> and: Products/ZenReports/reports/Performance_Reports/*
[11-Jun-2009 08:54:42] <ckrough> does the ZenReports/reports/ directory structure directly related to the report locations in the gui?
[11-Jun-2009 08:55:35] <venturaville> has more to do with the class names........
[11-Jun-2009 08:55:41] <venturaville> but pretty close
[11-Jun-2009 08:56:25] <ckrough> so, why are my custom device reports zope objects, but these device reports rpts on the file system
[11-Jun-2009 08:58:26] <rmatte> the native monitoring of interfaces in Zenoss is done via ping is it not?
[11-Jun-2009 08:58:37] <ckrough> availability monitoring is, yes
[11-Jun-2009 08:58:54] <rmatte> I mean for port status
[11-Jun-2009 08:59:09] <rmatte> operational & administrative status on ports
[11-Jun-2009 08:59:16] <ckrough> snmp
[11-Jun-2009 09:00:01] <ckrough> its the ethernetCsmacd template
[11-Jun-2009 09:00:24] <ckrough> or more likely ethernetCsmacd_64
[11-Jun-2009 09:01:10] <rmatte> ok, now I've implemented a "operational status" threshold and added ifOperStatus to poll .1.3.6.1.2.1.2.2.1.8, I have the threshold kick off based on the operational status of the port. A whole bunch of ports went down last night on some devices and the thresholds did kick off and notify, but when you went to the OS tab, the interfaces still showed as all green
[11-Jun-2009 09:01:23] <rmatte> any ideas as to why?
[11-Jun-2009 09:02:00] <ckrough> do you use zensyslog
[11-Jun-2009 09:02:13] <rmatte> It's running, yes
[11-Jun-2009 09:02:52] <ckrough> we ended up forwarding LINK UPDOWN messages (IOS) from our syslog servers to zensyslog, then did alerting based on that
[11-Jun-2009 09:03:03] <ckrough> for ports other than the management port on a switch
[11-Jun-2009 09:03:43] <ckrough> not sure why your threshold method didnt work.
[11-Jun-2009 09:05:23] <rmatte> my threshold method did work
[11-Jun-2009 09:05:29] <rmatte> we got alerts from the thresholds
[11-Jun-2009 09:05:42] <rmatte> I'm asking why Zenoss' default monitoring method didn't detect port status change
[11-Jun-2009 09:05:44] <venturaville> yeah ...editing reports in zope actually works in 2.4 now ...
[11-Jun-2009 09:05:44] <rmatte> and still hasn't
[11-Jun-2009 09:05:53] <rmatte> all the ports are showing green even though they are admin down now
[11-Jun-2009 09:06:29] <ckrough> rmatte: let me look at mine. not sure if it goes red for admin down or just oper down
[11-Jun-2009 09:06:39] <rmatte> k
[11-Jun-2009 09:06:56] <rmatte> you'd think it would go red for admin down
[11-Jun-2009 09:07:13] <ckrough> rmatte: check the OS tab of the device in question, there are three columns for status
[11-Jun-2009 09:07:25] <rmatte> correct
[11-Jun-2009 09:07:29] <rmatte> and they are all green
[11-Jun-2009 09:07:30] <ckrough> is it showing A as red for those interfaces, but still showing green in the status page?
[11-Jun-2009 09:07:33] <ckrough> oh
[11-Jun-2009 09:07:39] <rmatte> the third column just means whether or not the port is monitored
[11-Jun-2009 09:07:46] <rmatte> the operational and admin indicators are green
[11-Jun-2009 09:07:51] <ckrough> yeah, that's not right :
[11-Jun-2009 09:07:51] <rmatte> but the port has been admin down for hours
[11-Jun-2009 09:08:19] <ckrough> did you check an snmpget against it just to make sure its reporting correctly
[11-Jun-2009 09:08:52] <rmatte> I'd imagine that I'd use the same value that I'm using for my threshold?
[11-Jun-2009 09:09:13] <ckrough> 1.3.6.1.2.1.2.2.1.7 is ifAdminStatus
[11-Jun-2009 09:09:22] <rocket> rmatte: go to the os tab
[11-Jun-2009 09:09:43] <rocket> actually just go to the device
[11-Jun-2009 09:09:48] <rmatte> done
[11-Jun-2009 09:09:53] <rocket> on the url bar add /manage
[11-Jun-2009 09:10:06] <rmatte> in management console...
[11-Jun-2009 09:10:13] <rocket> just a sec
[11-Jun-2009 09:10:18] <rmatte> k
[11-Jun-2009 09:10:26] <rocket> click os
[11-Jun-2009 09:10:31] <venturaville> <tal:block tal:define="
[11-Jun-2009 09:10:31] <venturaville> objects python:here.getDmdRoot('Devices').getSubDevices(lambda x: x.getSnmpStatus()!=0);
[11-Jun-2009 09:10:31] <venturaville> tableName string: snmpstatusreportwithips;
[11-Jun-2009 09:10:31] <venturaville> batch python:here.ZenTableManager.getBatch(tableName,objects, sortedHeader='getSnmpStatusHeader', sortedSence='desc');
[11-Jun-2009 09:10:31] <venturaville> exportFields python:['device', 'classPath', 'hwProductName', 'productionState', 'pingStatus', 'snmpStatus'];
[11-Jun-2009 09:10:31] <rocket> click interfaces
[11-Jun-2009 09:10:32] <rmatte> k
[11-Jun-2009 09:10:33] <venturaville> ">
[11-Jun-2009 09:10:38] <rmatte> k
[11-Jun-2009 09:10:42] <rocket> choose the one that you have as down
[11-Jun-2009 09:10:47] <rmatte> got it
[11-Jun-2009 09:10:50] <rocket> click properties
[11-Jun-2009 09:10:53] <venturaville> is there a doc somewhere that explains the above....
[11-Jun-2009 09:11:11] <rmatte> I don't see properties listed?
[11-Jun-2009 09:11:41] <rocket> upper right
[11-Jun-2009 09:11:47] <rmatte> ohhh
[11-Jun-2009 09:11:50] <rocket> after you click on the interface
[11-Jun-2009 09:11:51] <rmatte> got it
[11-Jun-2009 09:12:17] <rmatte> adminStatus is set to 1, so is operStatus, even though the port is clearly down
[11-Jun-2009 09:12:18] <rocket> what is adminStatus set to?
[11-Jun-2009 09:12:23] <rocket> ok just a sec
[11-Jun-2009 09:12:26] <rmatte> k
[11-Jun-2009 09:12:38] <rmatte> (this is a 2.3.3 box btw)
[11-Jun-2009 09:12:42] <rmatte> I haven't updated it yet
[11-Jun-2009 09:16:49] <ckrough> looking at one of mine which is oper and admin down, the value is "2" for each. fyi
[11-Jun-2009 09:16:52] <rocket> rmatte: remodel the device
[11-Jun-2009 09:17:30] <rocket> rmatte: I am guessing that variable is only updated when you remodel
[11-Jun-2009 09:17:46] <rocket> rmatte: from my brief glance at the code anyway
[11-Jun-2009 09:18:17] <rocket> rmatte: I believe if I remember right admin down is managed by remodel
[11-Jun-2009 09:18:35] <rocket> rmatte: and actual interface down stuff should be set to catch a trap from the device
[11-Jun-2009 09:19:20] <rocket> rmatte: from a design perspective I wouldnt alert on admin down as that tends to mean an administrator has put the interface down .. etc ..
[11-Jun-2009 09:19:38] <rocket> rmatte: you could verify this with one of the devs on thursday though
[11-Jun-2009 09:20:35] <rmatte> yeh, but remodelling every device each time a port goes down is not a solution
[11-Jun-2009 09:20:55] <rocket> rmatte: I believe you have already applied the solution at the moment
[11-Jun-2009 09:21:10] <rmatte> it is thursday
[11-Jun-2009 09:21:16] <rmatte> I thought devs were today?
[11-Jun-2009 09:21:20] <rmatte> or is that next Thursday?
[11-Jun-2009 09:21:22] <rocket> rmatte: they are later
[11-Jun-2009 09:21:28] <rmatte> yeh, like 11:00
[11-Jun-2009 09:21:28] <rocket> later today
[11-Jun-2009 09:21:32] <rocket> rmatte: lost track of dayss
[11-Jun-2009 09:21:51] <rmatte> definitely going to ask, generally if a port is monitored you want to know what's happening to it regardless
[11-Jun-2009 09:21:57] <rmatte> otherwise you wouldn't be monitoring it
[11-Jun-2009 09:22:26] <rmatte> and the fact that it didn't even pick up on the operational status let alone the admin status is an issue
[11-Jun-2009 09:22:47] <rocket> rmatte: operational status I would agree 100% on
[11-Jun-2009 09:23:31] <rocket> rmatte: that should be obtained off of a datacollector run not a modeler run
[11-Jun-2009 09:23:48] <rocket> technically I think both should but I didnt design this up front
[11-Jun-2009 09:25:00] <rocket> in my own mind I see it as a core design flaw .. (its fixable) but it might be fixed on a core review down the road etc
[11-Jun-2009 09:25:28] <rocket> I think the team needs to look at modelling vs monitoring again and take another iteration at it
[11-Jun-2009 09:26:03] <rocket> I think that in reality modelling should probably go away .. and add the ability to monitor different things at different intervals
[11-Jun-2009 09:26:28] <rocket> and give the user the ability to define the intervals that makes sense for them
[11-Jun-2009 09:27:07] <rocket> but I dont know the whole background of why it was determined to be done this way ..
[11-Jun-2009 09:27:36] <ckrough> rocket: scalability? would be brutal on large installs
[11-Jun-2009 09:27:55] <venturaville> modeling is broken out because it is a really really heft process
[11-Jun-2009 09:28:20] <rocket> venturaville: I agree about that but did you not see the different intervals comment
[11-Jun-2009 09:28:26] <venturaville> yep I did
[11-Jun-2009 09:28:50] <venturaville> I do agree about the interfaces for sure
[11-Jun-2009 09:28:53] <rocket> eg software modelling I totally agree should be done once a week or something like that
[11-Jun-2009 09:29:12] <rocket> network modelling maybe once an hour to pick up admin down changes
[11-Jun-2009 09:29:26] <venturaville> depends on what you mean by network modelling too
[11-Jun-2009 09:29:35] <venturaville> querying a router can take a very long time
[11-Jun-2009 09:29:48] <rocket> yup .. it really does ..
[11-Jun-2009 09:30:02] <rocket> venturaville: depending on the interfaces on the router etc yeah ..
[11-Jun-2009 09:30:27] <rocket> venturaville: thats why I think that the thought was notifications of port status would be caught via a trap
[11-Jun-2009 09:30:54] <venturaville> yep ... which works in a small setup, but it is difficult to map out on a per device what you care about
[11-Jun-2009 09:31:04] <venturaville> when mapping out thousands of devices
[11-Jun-2009 09:31:36] <rocket> venturaville: yeah ...
[11-Jun-2009 09:31:43] <rmatte> I've completely disabled the zenmodel process from running as a daemon on that particular Zenoss install since it's monitoring 342, many of which are in Mexico, and Europe. Modelling some of the devices across slow links takes a ridiculous amount of time, I had to crank the SNMP timeout settings and the collector model time setting up to ridiculously high values just to model some of this stuff
[11-Jun-2009 09:31:52] <venturaville> I would hate to see the event class for that......
[11-Jun-2009 09:31:54] <rmatte> so I sure as heck don't want it modelling at intervals
[11-Jun-2009 09:32:00] <rocket> venturaville: thats why I was thinking about the granularity of being able to decide intervals ..
[11-Jun-2009 09:32:13] <cgibbons> hurm
[11-Jun-2009 09:32:30] <rocket> cgibbons: hey there
[11-Jun-2009 09:33:24] <venturaville> rocket: probably you could get away with some of it by rewriting the interface collector to gather those datapoints and use the range threshold zenpack I saw posted
[11-Jun-2009 09:33:46] <rocket> venturaville: the range threshold I wrote?
[11-Jun-2009 09:33:52] <venturaville> rocket: kudos :-)
[11-Jun-2009 09:34:28] <rmatte> even defining intervals for modelling doesn't resolve this issue, interfaces should be polled for operational and administrative status during each polling cycle
[11-Jun-2009 09:35:06] <ckrough> unless you have tens of thousands of ports
[11-Jun-2009 09:35:40] <rmatte> then you obviously go through and unmonitor anything that you don't want to monitor
[11-Jun-2009 09:36:04] <rmatte> I'm monitoring like 140 switches on this Zenoss box, but I went through and only set it to monitor things like trunk ports and such
[11-Jun-2009 09:36:05] <rocket> rmatte: again I dont disagree on the port issue ..
[11-Jun-2009 09:37:01] <rocket> rmatte: and I would tend to argue that you should be able to define if you want monitoring on or off by default before you model the device the first time etc .. depending on the type of switch it is etc
[11-Jun-2009 09:37:14] <rmatte> yeh
[11-Jun-2009 09:37:34] <rocket> eg if you are modelling a switch connected to workstations .. off by default and then turn on the interfaces you care to be notified about
[11-Jun-2009 09:37:48] <rmatte> yeh, for sure
[11-Jun-2009 09:38:08] <rocket> but if you are modelling a server switch .. set it on .. and turn off monitoring on development server ports ..
[11-Jun-2009 09:38:25] <rmatte> I had to go through and set tons of interfaces not to be monitored since they were all end user ports, or printers, or whatever
[11-Jun-2009 09:38:54] <rocket> rmatte: i can feel that pain ...
[11-Jun-2009 09:39:07] <rmatte> hehe
[11-Jun-2009 09:55:31] <bwest> hello
[11-Jun-2009 09:55:47] <venturaville> does anyone have an example of a zenpack that makes populates some performance tabs through a zenrelation it itself created?
[11-Jun-2009 09:56:35] <rocket> venturaville: I have one that does similar but doesnt create perf graphs .. :/
[11-Jun-2009 09:56:49] <venturaville> the relation I have working
[11-Jun-2009 09:56:58] <venturaville> now just trying to figure out how to populate it...
[11-Jun-2009 09:57:13] <bwest> Is it possible to use winexe as part of a plugin to doel a windows server?
[11-Jun-2009 09:57:50] <bwest> to model windows
[11-Jun-2009 10:01:09] <rocket> bwest: why are you using winexe? what version of zenoss are you using?
[11-Jun-2009 10:01:42] <rocket> bwest: I believe wmi is the new way to go for that .. with newer zenoss installations eg 3.2 and above
[11-Jun-2009 10:03:02] <cgibbons> *golfclap*
[11-Jun-2009 10:03:11] <mrayzenoss> good morning/afternoon/evening everyone, cgibbons and I will be here to answer your questions
[11-Jun-2009 10:03:21] <mrayzenoss> especially cgibbons
[11-Jun-2009 10:03:45] <rmatte> good morning Matt
[11-Jun-2009 10:03:50] <cgibbons> bwest: yeah, rocket's right. it's much better to use the built-in WMI support to do your own modeling plugins if you need.
[11-Jun-2009 10:04:51] <bwest> I am using 2.4.1
[11-Jun-2009 10:05:07] <bwest> I am using winexe to try and pull the serial number off the physical server
[11-Jun-2009 10:06:39] <cgibbons> not a good way to do it, really. winexe isn't a good tool. zenoss ships with wmic as a command-line tool that you can use, plus you can run WMI queries directly from the python code.
[11-Jun-2009 10:08:15] <bwest> So how do I get to the command line when modeling to use wmic
[11-Jun-2009 10:08:22] <rmatte> ok, so my main question for the developers: On the OS tab, port status appears to be determined during modelling. Why is no polling done on monitored interfaces each polling cycle to check status? I'm having to use a threshold to properly monitor administrative status on ports, and it doesn't update the status on the OS tab. Ports stay showing as up when they should be showing as down, and vice versa.
[11-Jun-2009 10:08:45] <cgibbons> for example I just did this here: wmic -U 'zenoss\administrator' //hostname 'select SerialNumber from Win32_SystemEnclosure'
[11-Jun-2009 10:09:27] <cgibbons> matt, isn't there a community zenpack that already does modeling using wmic, or was that just performance data?
[11-Jun-2009 10:09:42] <cgibbons> there's also a way to run WMI queries from the python code, check out zenwin.py for an example
[11-Jun-2009 10:10:13] <mrayzenoss> cgibbons: this one? http://www.zenoss.com/community/projects/zenpacks/windowswmidevicetemplate
[11-Jun-2009 10:10:25] <cgibbons> look at WMIPlugin.py in Products.ZenWin for somethinng you can extend in your own modeling plugins to do WMI queries
[11-Jun-2009 10:10:41] <bwest> ok thx
[11-Jun-2009 10:11:18] <bwest> wasn't sure if i could run the command from the plugin.
[11-Jun-2009 10:11:58] <cgibbons> the way WMIPlugin.py is done you don't have to worry about executing the query yourself, you just override the queries() method with the WMI queries you want it to execute
[11-Jun-2009 10:12:05] <cgibbons> and it'll call back your process method with the results
[11-Jun-2009 10:12:21] <cgibbons> rmatte: good question, and the last time I asked that the answer was there wasn't a good way in the architecture to do that.
[11-Jun-2009 10:13:08] <rmatte> cgibbons: ok, is it something that's being looked at for the future at all?
[11-Jun-2009 10:13:26] <rmatte> it's a fairly basic thing to have
[11-Jun-2009 10:14:07] <cgibbons> it hasn't been escalated to the point where the enterprise customers have asked for it (surprisingly). i'm surprised it doesn't come up more.
[11-Jun-2009 10:14:32] <ckrough> Not very manageable on a large scale.
[11-Jun-2009 10:15:07] <ckrough> ALready tough enough to complete a polling cycling in time with the current OIDS
[11-Jun-2009 10:18:07] <bwest> If I have an environment that has about 350 components it's monitoring, running 2.4.1 on linux what server requirments shold be large enough? The components are 200 servers and 150 switches and routers
[11-Jun-2009 10:18:24] <rmatte> I'm just looking at this from the perspective of Zenoss' campaign to win over people from the other larger monitoring companies/tools, but if you're lacking basic functionality like this, it doesn't really make a good showing.
[11-Jun-2009 10:19:19] <rmatte> bwest: I'm monitoring 342 devices, 53 are servers, the rest are cisco routers and switches, I use 3GHz processor with 4GB of RAM
[11-Jun-2009 10:19:48] <cgibbons> there's a chart we have on teh website with requirements... let me find it
[11-Jun-2009 10:20:15] <cgibbons> yeah: http://www.zenoss.com/product/deployment
[11-Jun-2009 10:20:20] <ckrough> bwest: depents on the number of things you are collecting, how long you keep the rrd data, and how frequently you collect it
[11-Jun-2009 10:21:01] <bwest> they are interested in baselining and they are monitoring this data around the clock
[11-Jun-2009 10:21:51] <bwest> so needless to say they collect alot of data and store this
[11-Jun-2009 10:22:28] <bwest> not to mention they want multi graphs running 24/7
[11-Jun-2009 10:22:57] <mrayzenoss> make sure you're on a fast filesystem
[11-Jun-2009 10:23:01] <cgibbons> mattray, do you have any stats available with some of our really big customers?
[11-Jun-2009 10:23:08] <cgibbons> i know we have some relatively large deployments
[11-Jun-2009 10:23:16] <mrayzenoss> well, the largest community install 1 box is 8500 devices
[11-Jun-2009 10:23:26] <ckrough> 1 box?
[11-Jun-2009 10:23:26] <mrayzenoss> almost all SNMP
[11-Jun-2009 10:23:37] <mrayzenoss> yeah, 8-way 32 gigs of RAM
[11-Jun-2009 10:23:54] <rocket> ckrough: devices isnt really indicitive of the amount of data they are collecting
[11-Jun-2009 10:23:55] <mrayzenoss> someone else is doing 10K
[11-Jun-2009 10:23:59] <mrayzenoss> with 4 boxes
[11-Jun-2009 10:24:03] <ckrough> I can only get 1600 devices on that spec, with 6 15k RAID10 storage config
[11-Jun-2009 10:24:08] <rocket> ckrough: need to look at the # of datapoints collected
[11-Jun-2009 10:24:19] <ckrough> 75k OIDs per box on that spec
[11-Jun-2009 10:25:01] <cgibbons> I suspect a lot of the large upcoming installations will be doing mostly Windows devices using WMI/PerfMon, so that'll change the game some
[11-Jun-2009 10:25:33] <mrayzenoss> if you go to Collectors->localhost->Performance you can see your stats
[11-Jun-2009 10:25:50] <mrayzenoss> assuming 1 collector named localhost
[11-Jun-2009 10:27:05] <bwest> one collector with 10k
[11-Jun-2009 10:27:20] <rmatte> one collector with 10k is insane
[11-Jun-2009 10:27:28] <ckrough> 10k what
[11-Jun-2009 10:27:34] <Jontler> devices
[11-Jun-2009 10:27:38] <rmatte> I assumed he meant devices
[11-Jun-2009 10:27:41] <mrayzenoss> yeah
[11-Jun-2009 10:27:48] <mrayzenoss> but it was lightweight
[11-Jun-2009 10:27:52] <rmatte> what kind of devices though?
[11-Jun-2009 10:27:59] <mrayzenoss> cable modems?
[11-Jun-2009 10:28:14] <bwest> no not devices data points
[11-Jun-2009 10:28:14] <rmatte> oh, that's not really incredibly impressive then
[11-Jun-2009 10:28:19] <rmatte> not much to poll on those things
[11-Jun-2009 10:28:25] <mrayzenoss> right
[11-Jun-2009 10:28:44] <ckrough> disk IO seems to be the killer. anyone runnign this on SSD yet?
[11-Jun-2009 10:29:41] <cgibbons> it scales nicely horizontally for that issue, since data (sans events) is kept on each collector.
[11-Jun-2009 10:29:45] <rmatte> I'm polling like 18.7k datapoints on my largest install so far, but I'm going to have much larger projects coming down the pipe soon
[11-Jun-2009 10:30:27] <bwest> but for 10K datapoints I should just look at the sizing chart?
[11-Jun-2009 10:30:54] <mrayzenoss> here's the bragging thread for Core installs: http://forums.zenoss.com/viewtopic.php?t=6230
[11-Jun-2009 10:31:25] <ckrough> >430k OID on my biggest install
[11-Jun-2009 10:31:29] <ckrough> enterprise tho
[11-Jun-2009 10:31:54] <mrayzenoss> heh, I'm doing about 110 on my home install
[11-Jun-2009 10:32:37] <mrayzenoss> Anyone going to be at SouthEast LinuxFest this weekend?
[11-Jun-2009 10:33:00] <cgibbons> hrmph only 220 on my home setup, which i haven't looked @ in months
[11-Jun-2009 10:33:25] <ckrough> Yeah, I plan to set it up at home to watch my total usage with NetFlix... b/w caps are coming...
[11-Jun-2009 10:33:30] <cgibbons> doh it's only 2.3.1, bad me
[11-Jun-2009 10:33:48] <rocket> cgibbons: heh I dont have as many datapoints .. but a fairly large number of zenpacks installed
[11-Jun-2009 10:33:59] <mrayzenoss> my home install is where the source build on PPC gets tested
[11-Jun-2009 10:34:11] <cgibbons> you still have that powerbook??
[11-Jun-2009 10:34:19] <mistich> does anyone know the status of IPV6 in zenoss
[11-Jun-2009 10:34:35] <mrayzenoss> mistich: it hasn't gotten prioritized very high
[11-Jun-2009 10:34:45] * ckrough plays sad trombone
[11-Jun-2009 10:34:46] <mrayzenoss> mistich: there is a HOWTO somewhere for doing IPV6 ping
[11-Jun-2009 10:35:04] <mrayzenoss> http://www.zenoss.com/community/wiki/user-contributed/IPv6Ping/
[11-Jun-2009 10:35:10] <mrayzenoss> which could be adapted
[11-Jun-2009 10:35:24] <bwest> I am trying to model SIP adapters. SNMP is not available on them, but they can be identified with Mac addresses, can zenoss get this information
[11-Jun-2009 10:35:25] <mistich> yes but what about snmp ssh etc
[11-Jun-2009 10:35:31] <mrayzenoss> I did get some emails from a university in China that said their class project was to port Zenoss to IPV6
[11-Jun-2009 10:35:42] <mrayzenoss> I'm guessing the class failed
[11-Jun-2009 10:35:52] <mrayzenoss> haven't heard back from them in months
[11-Jun-2009 10:35:54] <ckrough> Is there much demand for making reporting more robust? Maybe moving graph definitions to re-usable objects, etc
[11-Jun-2009 10:36:02] <mistich> yes I saw that too has anyone followed up with them
[11-Jun-2009 10:36:06] <mrayzenoss> I did
[11-Jun-2009 10:36:26] <mrayzenoss> I emailed them telling them I would answer any questions and help them any way they needed
[11-Jun-2009 10:36:38] <cgibbons> there's some good porting enhancements coming in the next release actually
[11-Jun-2009 10:36:42] <cgibbons> er reporting
[11-Jun-2009 10:36:43] <mrayzenoss> ckrough: there are already some changes going into King Crab for reporting
[11-Jun-2009 10:36:51] <rmatte> mrayzenoss: probably, I'm sure they found out that it was much more complicated than they initially thought
[11-Jun-2009 10:36:57] <mrayzenoss> rmatte: definitely
[11-Jun-2009 10:37:06] <ckrough> mrayzenoss: I will send you swag if you do
[11-Jun-2009 10:37:09] <mrayzenoss> heh
[11-Jun-2009 10:37:25] <mrayzenoss> well, bang the drum with support
[11-Jun-2009 10:37:35] <mrayzenoss> everyone shows up with their wishlist and it gets prioritized
[11-Jun-2009 10:37:36] <rmatte> you'd almost have to have some sort of layer between Zenoss and the device that takes a dummy ipv4 address and forwards the traffic to an ipv6 address and vice versa
[11-Jun-2009 10:37:39] <rmatte> as a workaround
[11-Jun-2009 10:37:51] <mrayzenoss> yeah, it's non-trivial
[11-Jun-2009 10:37:58] <mistich> next question moving away from rrds to a relational database
[11-Jun-2009 10:38:03] <ckrough> ^^ this
[11-Jun-2009 10:38:12] <sergeymasushko> can someone help me with the graph issue. I believe it should be simple, but I stack on it. http://forums.zenoss.com/viewtopic.php?t=10008
[11-Jun-2009 10:38:33] <rmatte> Matt, any progress on figuring out what the cause of that zenpack load time issue could be?
[11-Jun-2009 10:38:41] <rmatte> (just curious)
[11-Jun-2009 10:39:08] <mrayzenoss> rmatte: jplouis was looking at it, I believe nothing obvious was showing up yet
[11-Jun-2009 10:39:35] <cgibbons> meaning you want to export your raw RRD data into a relational database for custom reporting, mistich, or you want to change the data collection to store in a database directly?
[11-Jun-2009 10:39:36] <rmatte> cool
[11-Jun-2009 10:39:44] <mrayzenoss> mistich: the idea of moving off of RRD does come up, just not sure what we'd move to
[11-Jun-2009 10:40:13] <rmatte> one of our guys here came up with a way to dump an RRD to XML and then pull from there
[11-Jun-2009 10:40:16] <mistich> well for reporting relational database would be nice
[11-Jun-2009 10:40:29] <mistich> yes that is easy
[11-Jun-2009 10:40:35] <bwest> I am sorry i am so full of questions, but is it possible to send mysql queries from zenoss to read the amount of data in a table and create events off this?
[11-Jun-2009 10:40:36] <mistich> but want realtime data
[11-Jun-2009 10:40:38] <jplouis> zenpack slowness has to do with the way the relationships are loaded. The zenpack has a relationship to each packable
[11-Jun-2009 10:40:44] <mrayzenoss> bwest: yes
[11-Jun-2009 10:40:57] <mrayzenoss> mistich: understood
[11-Jun-2009 10:41:10] <cgibbons> I can tell you from experience that dumping that much raw data into a relational database in real-time is not a fun path.
[11-Jun-2009 10:41:18] <cgibbons> depending upon your install size of course.
[11-Jun-2009 10:41:54] <mistich> then once in a relational database reporting becomes very easily can use any package to do reporting
[11-Jun-2009 10:41:55] <mrayzenoss> bwest: have you tried the Core MySQL ZenPack?
[11-Jun-2009 10:42:03] <rmatte> bwest, you'd have to write a script for that and use it as a command based datapoint
[11-Jun-2009 10:42:17] <bwest> perfect
[11-Jun-2009 10:42:19] <mrayzenoss> bwest: I just got another massive set of ZenPacks for doing databases yesterday (MySQL, PostgreSQL, MS SQL, ZeoDB)
[11-Jun-2009 10:42:24] <rmatte> although that MySQL ZenPack might be able to do it, haven't looked at it
[11-Jun-2009 10:42:38] <mistich> oh I understand my install is huge but everyday getting asked for more and more reports
[11-Jun-2009 10:42:41] <mrayzenoss> it provides some stats, it may be good enough
[11-Jun-2009 10:43:03] <rmatte> MS SQL you say?
[11-Jun-2009 10:43:08] <mrayzenoss> rmatte: yeah
[11-Jun-2009 10:43:16] <mrayzenoss> MS SQL Server
[11-Jun-2009 10:43:18] <rmatte> I've been waiting for a good community Zenpack for that
[11-Jun-2009 10:43:28] <mrayzenoss> don't like the existing one?
[11-Jun-2009 10:43:38] <mrayzenoss> etank was looking for someone who had used it yesterday
[11-Jun-2009 10:43:41] <bwest> when will that one be published My SQL
[11-Jun-2009 10:43:58] <mrayzenoss> probably over the weekend
[11-Jun-2009 10:44:05] <mrayzenoss> processing new ZenPacks takes about an hour each
[11-Jun-2009 10:44:10] <mrayzenoss> I got 9 yesterday
[11-Jun-2009 10:44:19] <mrayzenoss> which is a great problem to have
[11-Jun-2009 10:44:22] <bwest> excellent, your contest is working
[11-Jun-2009 10:44:32] <mrayzenoss> yeah, we already have the same number of entries as last year
[11-Jun-2009 10:44:38] <mrayzenoss> and there are 2 more months
[11-Jun-2009 10:44:44] <mrayzenoss> we may have to bump up the prizes
[11-Jun-2009 10:44:54] <mistich> yeah we have about 6 zenpacks one day well publish
[11-Jun-2009 10:45:07] <chudler> I do have at least 4 more to send you, but I have to get time to polish them
[11-Jun-2009 10:45:11] <mistich> well I'll wait till the prizes are bumped
[11-Jun-2009 10:45:54] <mrayzenoss> the current backlog is: PredictiveThreshold, AIX SNMP, Netware, LinuxMonitorAddon, Gentoo, SuSE, Jabber XMMP, RDBMS, MySQL, MS SQL Server, PostgreSQL, ZeoDB
[11-Jun-2009 10:46:16] <mrayzenoss> no need to wait
[11-Jun-2009 10:46:28] <mistich> PredictiveThreshold is this holt winters
[11-Jun-2009 10:46:32] <mrayzenoss> yes
[11-Jun-2009 10:46:44] <mistich> already have that one
[11-Jun-2009 10:46:48] <mrayzenoss> I'm comparing the Enterprise one with the Community one now
[11-Jun-2009 10:47:25] <bwest> Are there any good zenpacks to monitor VOIP in core or enterprise?
[11-Jun-2009 10:47:27] <etank> mrayzenoss: i think that the biggest issue for me was getting the Perfmon ZenPack to work
[11-Jun-2009 10:47:43] <etank> winexe was breaking down and i think that was where the issues are
[11-Jun-2009 10:48:01] <mrayzenoss> bwest: the Asterisk one seems to have some issues, but you could probably adapt any Nagios plugin that already exists
[11-Jun-2009 10:48:05] <cgibbons> i may be biased, but if it were me, if a zenpack uses winexe I'd ditch it and never look bck.
[11-Jun-2009 10:48:06] <sergeymasushko> bwest: what do you want to monitor in the VOIP?
[11-Jun-2009 10:48:30] <sergeymasushko> bwest: I have some work around...
[11-Jun-2009 10:48:42] <bwest> Right now they use Cricket to monitor calls per minute users and the such
[11-Jun-2009 10:49:02] <etank> mrayzenoss: is there a PredictiveThreshold community Pack? or just for Enterprise?
[11-Jun-2009 10:49:13] <mrayzenoss> this will be for Core
[11-Jun-2009 10:49:36] <etank> really? how different would it be from the Enterprise one?
[11-Jun-2009 10:49:50] <mistich> what is the difference between them?
[11-Jun-2009 10:49:54] <mrayzenoss> I just set it up yesterday, they take a few days to generate data
[11-Jun-2009 10:50:08] <mrayzenoss> I've got both of them installed and testing
[11-Jun-2009 10:50:17] <mrayzenoss> my first impression is that they're pretty close
[11-Jun-2009 10:50:18] <sergeymasushko> bwest: what is the voip platform (sip server ?)
[11-Jun-2009 10:50:45] <bwest> yes
[11-Jun-2009 10:50:53] <rmatte> I have another question... when executing a command based script where you need to pass a password value to it, if the command fails the password gets spit out in plaintext in the command failed event. I logged a ticket for this and was told to "See the ZenPack creator for a fix". I've already tried fixing this ZenPack by using a python script to pull the username and password values from the Zope database, but after extensive testing I have determined that
[11-Jun-2009 10:50:59] <mistich> one thing to look for is low data in the kb holtwinters doesn't like it and sends false alerts
[11-Jun-2009 10:51:15] <rmatte> The only way to possibly fix this security issue is to have Zenoss intelligently sensor password values in command failed events.
[11-Jun-2009 10:51:39] <rmatte> Is there any particular reason why security issues like this are having a blind eye turned to them?
[11-Jun-2009 10:52:05] <mrayzenoss> well, I believe the ZenPack in question was one we don't even have access to
[11-Jun-2009 10:52:25] <cgibbons> the reason that was given as the answer rmatte is because there are two ways to deal with it: 1) not use COMMAND datatypes (which is what we mostly do), 2) a command script that needs a password gets it sent via stdin instead of on the command-line
[11-Jun-2009 10:52:25] <mrayzenoss> it's one of James Dastrup's, which we haven't looked at since he hasn't sent it to us for hosting
[11-Jun-2009 10:52:47] <sergeymasushko> bwest: for monitoring sip servers I use a shell command that run SIPP util with the different scenarios... then parse result and make an event if its needed. also I'm monitoring active calls on my cisco AS5300 GWs...
[11-Jun-2009 10:52:49] <rocket> mistich: I wrote the predictive zenpack .. and tried to make it as configurable as possible
[11-Jun-2009 10:53:12] <rocket> mistich: it should also retrofit older rrd data that didnt already have predictive trending applied
[11-Jun-2009 10:53:12] <rmatte> yeh, but it's not specific to that zenpack is my point, this issue pops up with any type of command based poll that needs a password passed to it from Zenoss
[11-Jun-2009 10:53:22] <mistich> rocket: nice
[11-Jun-2009 10:53:24] <rmatte> whether it be a custom built script, or a core Zenpack
[11-Jun-2009 10:53:29] <bwest> serg: sounds interesting, is there a way for you to share more with me
[11-Jun-2009 10:53:29] <cgibbons> there's a mechanism to do it that doesn't require it being sent on the command-line
[11-Jun-2009 10:53:37] <cgibbons> but the zenpack author has to use it
[11-Jun-2009 10:54:07] <rmatte> which would be using a python script to pull the values directly from Zope, correct?
[11-Jun-2009 10:54:11] <mistich> like that it retofits the rrd but does it also remove the holtwinters if you want to remove it
[11-Jun-2009 10:54:11] <rocket> mistich: so if you are getting false alerts my threshold will also allow you to go and reapply the beta and gamma and window etc
[11-Jun-2009 10:54:13] <sergeymasushko> bwest: a bit busy for now... lets talk about a bit later...
[11-Jun-2009 10:54:25] <bwest> ok
[11-Jun-2009 10:54:25] <rocket> mistich: it does not currently remove
[11-Jun-2009 10:54:32] <mistich> can't wait to look at it
[11-Jun-2009 10:54:54] <rocket> mistich: as I cant think of a good way to have zenoss do that on removal ..
[11-Jun-2009 10:55:18] <rocket> and as far as I can tell its not a bad performance hit to leave it enabled
[11-Jun-2009 10:55:23] <rmatte> I've already done that and modified the scripts in 3 different WMI zenpacks to use it, but we had a Zenoss server crash because the retrieval of the password and username values for each poll was way too intensive, if you can suggest a more efficient way of doing it then I'd be happy to modify the Zenpacks in that fashion.
[11-Jun-2009 10:55:47] <mistich> rocket: have run into that before thats why I bring it up
[11-Jun-2009 10:56:05] <chudler> when I need a password, I put it in a file, and read it from there. Not sure how/if that helps you, rmatte
[11-Jun-2009 10:56:15] <rmatte> It's either a little bit of code to sensor the password, or large loss of performance in the scripts
[11-Jun-2009 10:56:17] <rocket> mistich: it might be something that I could make a command line to do
[11-Jun-2009 10:56:38] <rmatte> chudler: doesn't really help when you're applying the templates against multiple devices with different usernames and passwords set for WMI in their zproperties
[11-Jun-2009 10:56:42] <rocket> mistich: eg remove_holt.py rrd_file
[11-Jun-2009 10:57:20] <mistich> have a question for everyone how are you handling spikes in graphs
[11-Jun-2009 10:57:25] <cgibbons> that's certainly why we have a daemon and WinPerf datatype in enterpise, so it can manage all that without having to spawn commands
[11-Jun-2009 10:57:27] <chudler> rmatte: good point. what might the code look like that can censor the password? wouldn't it need to know the password? or just /-p .*? -/ ?
[11-Jun-2009 10:57:43] <cgibbons> IMO, using a command script for monitoring is a short-term quick-fix work-around, and a real ZenPack should do better
[11-Jun-2009 10:57:44] <mistich> min and max I hate to use it because you have to recreate the rrd
[11-Jun-2009 10:57:54] <bwest> Another threshold questons. Right now there is the min max, but if I want ok warning critical so 3 levels how do i accomplish this
[11-Jun-2009 10:58:00] <cgibbons> (i'm also looking in our change history to find out if we hav an example of passing in the password via stdin)
[11-Jun-2009 10:58:21] <rocket> bwest: can you use better grammar your question is hard to understand
[11-Jun-2009 10:58:36] <bwest> Sorry
[11-Jun-2009 10:58:57] <bwest> Using thresholds there is a min and max
[11-Jun-2009 10:59:00] <rmatte> chudler: Zenoss should already know the password since it's passing it to the command, so when the command fails, identify that that is a password value and only display the rest of the command.
[11-Jun-2009 10:59:18] <sergeymasushko> bwest: can you add 2 different minmax thresholds to the same datapoint?
[11-Jun-2009 10:59:21] <bwest> but I want a third level so I can warn as the threshold is approachong max
[11-Jun-2009 10:59:53] <mistich> thats what we do
[11-Jun-2009 11:00:51] <Guest52519> hmm
[11-Jun-2009 11:00:55] <rmatte> cgibbons: thanks, if that can be done then that would be a viable solution
[11-Jun-2009 11:01:17] Guest52519 is now known as euphoria
[11-Jun-2009 11:01:40] <rmatte> as a side note, I'm finding myself writing quite a lot of command based scripts for monitoring all sorts of different stuff, such as the stock windows snmp agent which is a total piece of garbage and likes to move SNMP values around a lot
[11-Jun-2009 11:01:46] <rocket> bwest: there is a new community threshold zenpack
[11-Jun-2009 11:01:54] <rocket> bwest: rangethreshold
[11-Jun-2009 11:01:58] <rocket> eg you could do
[11-Jun-2009 11:02:12] <cgibbons> yeah monitoring Windows with SNMP is problematic even with the nice commerical agents
[11-Jun-2009 11:02:12] <rocket> rangethreshold = warn between 3 and 4
[11-Jun-2009 11:02:21] <jsynack> rmatte: move things around or put them in a table with different iid's ?
[11-Jun-2009 11:02:24] <mrayzenoss> http://www.zenoss.com/community/projects/zenpacks/rangethreshold
[11-Jun-2009 11:02:29] <rmatte> they have the snmp value for the RAM stored below the SNMP values for the physical drives, so the value to poll RAM is different on every server, the CPU OIDs tend to wander up and down, one day the first CPU will be .1, then it'll be .2, so I have to use scripts to deal with that
[11-Jun-2009 11:02:31] <rocket> minmax threshold critical > 4
[11-Jun-2009 11:02:32] <bwest> even better solution
[11-Jun-2009 11:03:50] <rmatte> my scripts are working quite nicely though, also allows me to easily calculate percentages and stuff
[11-Jun-2009 11:04:06] euphoria is now known as myeuphoria
[11-Jun-2009 11:05:15] <Jontler> i have a question about default performance monitoring of the cpu
[11-Jun-2009 11:05:25] <Jontler> right now, there's just one graph by default for the cpu
[11-Jun-2009 11:05:48] <rocket> bwest: make sense?
[11-Jun-2009 11:05:53] <Jontler> i have multi-core machines, though - is that cpu graph a compilation of the cpus, is it the first cpu only, or is it something else?
[11-Jun-2009 11:06:08] <bwest> yes perfect sense
[11-Jun-2009 11:06:10] <rocket> Jontler: from snmp its probably an average
[11-Jun-2009 11:06:16] <rmatte> Jontler: WMI or SNMP?
[11-Jun-2009 11:06:16] <Jontler> ssh
[11-Jun-2009 11:06:25] <rocket> bwest I thing range is <=
[11-Jun-2009 11:06:33] <rmatte> ssh would be a total I believe
[11-Jun-2009 11:06:35] <Jontler> im monitoring using ssh and the zenoss plugins on the machine side
[11-Jun-2009 11:06:41] <rocket> bwest: minmax is >=
[11-Jun-2009 11:06:53] <rocket> bwest: so you might want to have a small difference in place ..
[11-Jun-2009 11:06:56] <rocket> gotta run ..
[11-Jun-2009 11:07:40] <bwest> on one of my windows Servers the SNMP service will not stay on. It Starts and then a pop up comes up that says some sSNMP
[11-Jun-2009 11:07:50] <bwest> on local computer started and stopped
[11-Jun-2009 11:08:27] <bwest> how do I ovveride this so SNMP stays running
[11-Jun-2009 11:08:36] <rmatte> Jontler: easy way to find out, do "cat /dev/urandom > /dev/null", that should eat up all of the CPU on the first core, if it doesn't target the first core try again until it does, then check Zenoss in a bit for the current value, if it's 100% it's just monitoring that one core, if it's 25%ish, then it's monitoring everything
[11-Jun-2009 11:09:19] <Jontler> i will try that rmatte, thanks for the suggestion
[11-Jun-2009 11:09:23] <rmatte> np
[11-Jun-2009 11:09:45] <rmatte> you can use htop (you'll need to install it if you don't already have it) to monitor each core
[11-Jun-2009 11:10:04] <rmatte> htop blows top out of the water anyways, so I highly recommend it
[11-Jun-2009 11:10:10] <Jontler> ive never heard of it
[11-Jun-2009 11:10:13] <Jontler> installing now
[11-Jun-2009 11:10:19] <rmatte> you'll like it
[11-Jun-2009 11:10:21] <etank> htop is da`bomb
[11-Jun-2009 11:10:56] <etank> i leave it running in screen on this box all day long
[11-Jun-2009 11:11:03] <Jontler> oh hotness
[11-Jun-2009 11:11:10] <rmatte> hehe
[11-Jun-2009 11:11:39] <rmatte> screen is also "da`bomb"
[11-Jun-2009 11:11:43] <Jontler> yes it is
[11-Jun-2009 11:11:44] <etank> si
[11-Jun-2009 11:11:50] <mrayzenoss> wow, I didn't know top could be so sexxy
[11-Jun-2009 11:12:00] <etank> i like the screen-profiles stuff that ubuntu has
[11-Jun-2009 11:12:09] <rmatte> you can actually use mouse clicks with htop btw
[11-Jun-2009 11:12:19] <rmatte> (most people don't realize)
[11-Jun-2009 11:12:36] <Jontler> i would never have tried
[11-Jun-2009 11:13:01] <Jontler> nice, sorting by different categories with a mouse click
[11-Jun-2009 11:13:06] <rmatte> yessir
[11-Jun-2009 11:13:12] <rmatte> killing processes with mouse clicks as well
[11-Jun-2009 11:13:37] <Jontler> that's much more intuitive
[11-Jun-2009 11:14:10] <rmatte> but yeh, run that command that I told you to run and you should see the first core kick up to 100
[11-Jun-2009 11:14:16] <rmatte> then just keep an eye on Zenoss
[11-Jun-2009 11:14:21] <Jontler> it's already rolling
[11-Jun-2009 11:14:25] <rmatte> cool
[11-Jun-2009 11:14:31] <Jontler> i didnt realize that would only tank one core
[11-Jun-2009 11:14:37] <bwest> Another quick SNMP Question. on one of my windows servers, the SNMP service will not stay on. I get an message saying the SNMP service on your local machine has started and stopped. This happes when the service has nothing to do. Is there anyway I can fix this?
[11-Jun-2009 11:15:54] <rmatte> bwest: never seen that behaviour before
[11-Jun-2009 11:15:59] <sergeymasushko> heh... so no one can't help me with the graphs?
[11-Jun-2009 11:16:13] <rmatte> Jontler: yeh, it's a single thread process so it just tanks 1
[11-Jun-2009 11:16:57] <mrayzenoss> sergeymasushko: the links to the images are bad
[11-Jun-2009 11:17:09] <bwest> very strange I know, on that server I cannot get snmp to work
[11-Jun-2009 11:18:40] <rmatte> cgibbons: any luck finding that stdin method info?
[11-Jun-2009 11:20:12] <cgibbons> don't see an example yet. looks like we just stopped using command for anything like that.
[11-Jun-2009 11:20:34] <rmatte> ah
[11-Jun-2009 11:20:43] <mistich> well you all have a good day
[11-Jun-2009 11:21:01] <rmatte> I'll hunt around and see if I can find some method of doing it
[11-Jun-2009 11:21:50] <mrayzenoss> BTW, anyone looking for a Zenoss job in the bay area, let me know
[11-Jun-2009 11:22:02] <Jontler> rmatte: so the percentage is scaled from 0 to 200%, almost 200% was idle before, and zero is idle now
[11-Jun-2009 11:22:04] <Jontler> :confused:
[11-Jun-2009 11:22:24] <Jontler> something tells me it's only monitoring 1 core, and i need to tune the graphs a bit
[11-Jun-2009 11:23:08] <ckrough> if its gover over 100% I would think its adding all the cores
[11-Jun-2009 11:23:24] <ckrough> *going over*
[11-Jun-2009 11:23:34] <sergeymasushko> mrayzenoss: I've fixed the links.
[11-Jun-2009 11:23:44] <rmatte> yeh, it might be adding them up
[11-Jun-2009 11:23:48] <rmatte> 2 core or 4 core machine?
[11-Jun-2009 11:23:53] <Jontler> 2
[11-Jun-2009 11:24:03] <Jontler> and my second core - according to htop! - is only 25% in use
[11-Jun-2009 11:24:10] <Jontler> so it shouldnt be showing 0 out of 200% idle
[11-Jun-2009 11:25:01] <rmatte> yeh, then something is definitely messed
[11-Jun-2009 11:25:09] <Jontler> oh wait
[11-Jun-2009 11:25:11] <mrayzenoss> we're going to wrap this up, I'll let cgibbons get back to closing tickets. Keep the ZenPacks coming in and stop by at SouthEast LinuxFest if you're in the neighborhood
[11-Jun-2009 11:25:16] <rmatte> keep firing up commands until you see some action
[11-Jun-2009 11:25:17] <Jontler> something is messed alright, and it's called my eyes
[11-Jun-2009 11:25:25] <rmatte> run 2 or 3 in parallel
[11-Jun-2009 11:25:31] <rmatte> oh, really?
[11-Jun-2009 11:25:33] <Jontler> i zoomed in on the graph
[11-Jun-2009 11:25:34] <rmatte> need glasses do you?
[11-Jun-2009 11:25:41] <rmatte>
[11-Jun-2009 11:25:48] <Jontler> well i was all the way zoomed out and it looked like it was dropped off completely
[11-Jun-2009 11:25:59] <Jontler> zoomed in, it shows roughly 80% still available
[11-Jun-2009 11:26:02] <Jontler> so... yea
[11-Jun-2009 11:26:03] <Jontler> adding cores
[11-Jun-2009 11:26:29] <rmatte> yup
[11-Jun-2009 11:26:59] <Jontler> well, thanks for helping me solve that mystery
[11-Jun-2009 11:27:04] <rmatte> no problem
[11-Jun-2009 11:28:25] <rmatte> these Dev answer sessions are now officially marked in my calendar so no one can bug me to do stuff during them
[11-Jun-2009 11:28:52] <rmatte> Is there any pattern to them? Second Thursday of every month?
[11-Jun-2009 11:29:04] <rmatte> or is it random?
[11-Jun-2009 11:29:06] <mrayzenoss> every 2 weeks, Thursday
[11-Jun-2009 11:29:21] <rmatte> ah, I thought it only happened once a month
[11-Jun-2009 11:29:24] <mrayzenoss> http://www.zenoss.com/community/discuss/irc
[11-Jun-2009 11:29:35] <mrayzenoss> they've been pretty regular for awhile
[11-Jun-2009 11:31:05] <rmatte> What happened between February and March?
[11-Jun-2009 11:31:06] <rmatte> lol
[11-Jun-2009 11:32:02] <mrayzenoss> Spring Break
[11-Jun-2009 11:32:09] <rmatte> ah
[11-Jun-2009 11:32:11] <rmatte> right
[11-Jun-2009 11:32:25] <rmatte> ls
[11-Jun-2009 11:32:29] <rmatte> bah, wrong window
[11-Jun-2009 11:34:51] <ckrough> didnt I see some documentation at one point on how to create a report that has user input fields? Ie: Uptime report where user can specify some filters
[11-Jun-2009 11:35:17] <mrayzenoss> ckrough: like this? http://www.zenoss.com/Members/cluther/creating-methods-for-custom-reports
[11-Jun-2009 11:35:35] <mrayzenoss> King Crab will add something easier
[11-Jun-2009 11:35:48] <mrayzenoss> for making straight Python calls
[11-Jun-2009 11:37:07] <ckrough> thats close. I want to be able to build a report where the PHB can specify criteria like "show me all /Network devices with uptimes under 2hr", yada yada
[11-Jun-2009 11:37:32] <ckrough> I think this is a good start though. I can add the input fields to it
[11-Jun-2009 11:41:01] <rmatte> yeh, now if only I could make total sense of the code used in that custom method
[11-Jun-2009 11:41:02] <rmatte> lol
[11-Jun-2009 11:42:02] <rmatte> the for and if parts are obvious
[11-Jun-2009 11:42:10] <rmatte> just the first part of the code has me confuzzled
[11-Jun-2009 11:42:38] <rmatte> oh, I think I see what it's doing
[11-Jun-2009 11:42:57] <rmatte> it's essentially going through the process list and the first part is the function to join each item to make a list
[11-Jun-2009 11:43:04] <rmatte> so that it shows up as proc1, proc2, proc3, etc...
[11-Jun-2009 11:45:34] <ckrough> ya
[11-Jun-2009 11:46:14] <ckrough> where I usually get lost is in finding the data/objects I need
[11-Jun-2009 11:46:28] <ckrough> like "context.os.ipservices()" how would I know to look there for that data
[11-Jun-2009 11:46:36] <rmatte> yeh
[11-Jun-2009 11:46:45] <ckrough> can we copy Chet's brain?
[11-Jun-2009 11:46:52] <ckrough> you need to open source that dude
[11-Jun-2009 11:47:19] <rmatte> well, there is a relatively simple way to find stuff
[11-Jun-2009 11:47:28] <rmatte> go to any device, put /manage after the device name in the url
[11-Jun-2009 11:47:38] <rmatte> then it's essentially context.whatever.whatever()
[11-Jun-2009 11:48:07] <rmatte> so you could do context.os.winservices()
[11-Jun-2009 11:48:12] <ckrough> also, in zendmd: dev = devices.findDevice('whatever')
[11-Jun-2009 11:48:16] <ckrough> print '\n'.join(dir(dev))
[11-Jun-2009 11:48:25] <rmatte> or context.os.processes()
[11-Jun-2009 11:49:01] <rmatte> yeh
[11-Jun-2009 11:49:07] <ckrough> what does the context class represent
[11-Jun-2009 11:49:27] <rmatte> represents the classes under the device
[11-Jun-2009 11:49:30] <rmatte> the device is the context
[11-Jun-2009 11:49:35] <ckrough> ah
[11-Jun-2009 11:49:51] <ckrough> my problem is not zenoss, its lack of python exp
[11-Jun-2009 11:49:52] <rmatte> so if you navigate to any device then go in to the management interface you'll see the stuff you can reference
[11-Jun-2009 11:50:07] <rmatte> I'm still just learning Python, I'm no expert
[11-Jun-2009 11:50:12] <rmatte> but I do have experience with other languages
[11-Jun-2009 11:50:15] <rmatte> especially bash
[11-Jun-2009 11:50:17] <ckrough> same
[11-Jun-2009 11:50:30] <ckrough> most languages I pick up very quickly. Python is not compatible with my brain
[11-Jun-2009 11:51:21] <rmatte> hehe
[11-Jun-2009 11:51:34] <rmatte> I like the fact that python forces you to properly indent the code
[11-Jun-2009 11:51:42] <rmatte> though I do miss brackets
[11-Jun-2009 11:51:47] <ckrough> yes and yes
[11-Jun-2009 11:53:22] <rmatte> yeesh, I was wondering why my thumb was starting to burn, just noticed it still had mustard on it from my sandwich (I don't use my thumbs when typing)
[11-Jun-2009 11:53:23] <rmatte> lol
[11-Jun-2009 11:53:33] <ckrough> thats some serious mustard
[11-Jun-2009 11:53:41] <rmatte> yeh, it's hot mustard
[11-Jun-2009 11:53:57] <rmatte> one step down from mustard gas in the hazard department
[11-Jun-2009 11:57:05] <rhettlap> that stuff is pretty easy to make especially if you make you're own mustard
[11-Jun-2009 12:08:49] <ckrough> parsing every device in the dmd is painfully slow. is there some way to access a cache or index
[11-Jun-2009 13:27:54] <ckrough> anyone know where I can access a variable passed via URL in a custom method?
[11-Jun-2009 13:28:52] <mrayzenoss> you mean like a Python script attached in Zope?
[11-Jun-2009 13:28:59] <ckrough> yes sir
[11-Jun-2009 13:29:20] <mrayzenoss> request = container.REQUEST
[11-Jun-2009 13:29:37] <ckrough> lol, just found it in /zport/dmd/Reports/sql_query when you said that
[11-Jun-2009 13:29:38] <ckrough> thanks
[11-Jun-2009 13:29:38] <mrayzenoss> x = context.findDevice(request.id)
[11-Jun-2009 13:29:53] <mrayzenoss> ahh, ok
[11-Jun-2009 13:30:05] <mrayzenoss> I was just pulling up my Zenoss/MindTouch mashup for examples
[11-Jun-2009 13:30:17] <ckrough> whats this?
[11-Jun-2009 13:31:14] <mrayzenoss> https://sourceforge.net/projects/zenossdeki/
[11-Jun-2009 13:31:42] <mrayzenoss> it uses custom Python scripts installed on Zenoss to get called by MindTouch
[11-Jun-2009 13:31:52] <mrayzenoss> device.id gets passed around a lot
[11-Jun-2009 13:33:45] <CamargoBP> I'm having problems with displaying the correct units on the left side of the voice mail graph (blue graph) I pull the email server every 30 seconds an usually there are between 6 and 30 emails sent between that time. Any suggestions? http://i43.tinypic.com/2e4eyk8.png
[11-Jun-2009 13:35:11] <rhettlap> make sure the datasource and graph are setup to taken and display 30 second resolution
[11-Jun-2009 13:35:11] <CamargoBP> BTW if you use a Mac you need to get http://conceitedsoftware.com/products/linkinus The absolute BEST irc client I have ever used! Seriously... Thought I would spread the word. They have done an excellent job.
[11-Jun-2009 13:35:33] <rhettlap> but you have to pay for it :\
[11-Jun-2009 13:35:40] <CamargoBP> It's worth every penny
[11-Jun-2009 13:35:44] <rhettlap> i'll just stick w/ bitchx
[11-Jun-2009 13:35:48] <CamargoBP> You can install it on 2 computers
[11-Jun-2009 13:36:00] <mrayzenoss> I'm looking, but Colloquoy works ok for me
[11-Jun-2009 13:36:09] <mrayzenoss> and I was looking at how Adium is gonna do it in 1.4
[11-Jun-2009 13:36:24] <CamargoBP> Yeah I switched from Colloquoy last month to this version
[11-Jun-2009 13:36:33] <rhettlap> yeah colloquoy works ok. i just gave up on it since it was wigging out on me and went back to bitchx
[11-Jun-2009 13:36:47] <rhettlap> or xchat
[11-Jun-2009 13:37:01] <CamargoBP> You should at least try the 30 day trial I bet you shell out the $20 USD for it
[11-Jun-2009 13:37:23] <CamargoBP> Anyway...
[11-Jun-2009 13:37:26] <CamargoBP> Sorry tangent
[11-Jun-2009 13:37:26] <mrayzenoss> I'll try it
[11-Jun-2009 13:37:29] * ckrough heart irssi
[11-Jun-2009 13:37:50] <rhettlap> iirc the rra settings in zenoss are global correct?
[11-Jun-2009 13:38:19] <mrayzenoss> rhettlap: you can customize them on datasources
[11-Jun-2009 13:38:59] <rhettlap> ok. i never got into too much of messing with that stuff in zenoss since i am so comfortable with that stuff in cacti
[11-Jun-2009 13:39:40] <rhettlap> can you change it per template?
[11-Jun-2009 13:39:53] <mrayzenoss> yeah
[11-Jun-2009 13:39:54] <rhettlap> well the datasource is attached to a template
[11-Jun-2009 13:39:56] <rhettlap> so i guess yes
[11-Jun-2009 13:40:58] <rhettlap> wow i didnt know it was 11:40 already. i guess i should go down to the office
[11-Jun-2009 13:41:33] <CamargoBP> So I'm looking at the Graph definition tab, where do I set the 30 resolution?
[11-Jun-2009 13:47:16] <mrayzenoss> DataSource->DataPoints
[11-Jun-2009 13:47:26] <mrayzenoss> Create Cmd
[11-Jun-2009 13:47:33] <mrayzenoss> you'll need to read up on RRD stuff
[11-Jun-2009 13:50:31] <CamargoBP> Do you know if there is an example Cmd that someone has done for Zenoss, I can't seem to get that command correct. I've read up on how to do it, (the last few days) and the zenoss documentation is very vague. It would be nice if there was a screen shot of an example that actually worked.
[11-Jun-2009 13:58:58] <rmatte> mrayzenoss: that might not even be CamargoBP's issue, the snmpwalk is polling a correct value yet when he graphs it Zenoss does something weird and shrinks the value down
[11-Jun-2009 13:59:48] <rmatte> I had him remove the %s from the end of the graph point's format property, but it didn't fix it
[11-Jun-2009 13:59:51] <rmatte> so I'm totally stumped
[11-Jun-2009 14:05:08] <jb> snmptrap is UDP right?
[11-Jun-2009 14:07:00] <rmatte> The default destination port for SNMP traps is UDP port 162.
[11-Jun-2009 14:07:05] <jb> k
[11-Jun-2009 14:08:32] <rhettardo> hello
[11-Jun-2009 14:08:50] <rmatte> hello
[11-Jun-2009 14:12:31] <ckrough> hmm... whenever I loop a dictionary in zope it logs me out
[11-Jun-2009 14:13:04] <ckrough> python 1, ckrough 0. python wins again!
[11-Jun-2009 14:14:57] <rmatte> pythons have sharp teeth y'know
[11-Jun-2009 14:17:48] <rhettardo> thats true but gernally speaking boids and colubrids both have smaller catrlidge like sharp teeth!
[11-Jun-2009 14:18:11] <rmatte> yes, but they don't have their own scripting languages
[11-Jun-2009 14:18:21] <rhettardo> no :\
[11-Jun-2009 14:19:10] <rhettardo> i know more about the reptiles than the language :\
[11-Jun-2009 14:23:40] <venturaville> really confusing to go back and forth between perl/python/ruby in the same day
[11-Jun-2009 14:23:53] <venturaville> I pity the fool who looks at my code
[11-Jun-2009 14:27:00] <ckrough> venturaville: yeah, thats my main problem. perl/py/ruby/php
[11-Jun-2009 14:27:15] <ckrough> builtin methods all start converging
[11-Jun-2009 14:41:08] <cgibbons> holy smokes batman
[11-Jun-2009 14:50:26] <rocket> cgibbons: ?
[11-Jun-2009 15:01:12] <rmatte> great, we just lost a switch somewhere, all our phones are down on the second floor and none of the sales team can receive email on the first floor
[11-Jun-2009 15:01:15] <rmatte> fun fun fun
[11-Jun-2009 15:01:24] <ckrough> weee
[11-Jun-2009 15:01:29] <rocket>
[11-Jun-2009 15:01:44] <ckrough> we had an outage last night and AT&T is still catching up with SMS from zenoss
[11-Jun-2009 15:06:00] <CapriCoRN^80> hi
[11-Jun-2009 15:06:10] <CapriCoRN^80> hi rocket
[11-Jun-2009 15:06:12] <CapriCoRN^80> you around ?
[11-Jun-2009 15:06:17] <rocket> yes
[11-Jun-2009 15:06:30] <CapriCoRN^80> ok
[11-Jun-2009 15:06:42] <CapriCoRN^80> did you check that pic ?
[11-Jun-2009 15:07:06] <rocket> yeah .. but I have gotten busy here at work and not had time to really look at it ..
[11-Jun-2009 15:07:25] <CapriCoRN^80> ok
[11-Jun-2009 15:10:32] <CapriCoRN^80> see u than
[11-Jun-2009 15:10:33] <CapriCoRN^80> bye
[11-Jun-2009 15:29:24] <mloven> anyone know of a way to monitor bandwidth on all possible iterations of interface names without the mising RRD's that accompany putting lots of data sources into a graph?
[11-Jun-2009 15:30:52] <mloven> to clarify: If you were to have a Cisco router with a T1 card, that T1 card could be called "Serial 1/0" or "Serial 2/0" etc., depending on what hardware slot it's in....
[11-Jun-2009 15:32:04] <mloven> I'd like to have a graph that will monitor all of a router's T1's without putting all of the possible OID's into data sources, thereby getting "missing RRD's" when the T1 isn't in, say slot 1...
[11-Jun-2009 15:32:32] <mloven> does that make any sense at all?
[11-Jun-2009 15:35:43] <mrayzenoss> perhaps like this? http://www.zenoss.com/Members/asanabria/howto-monitor-and-add-multiple-data-points-the-easy-way-in-zenoss
[11-Jun-2009 15:36:13] <rmatte> yeh, that's the way to do it
[11-Jun-2009 15:36:30] <rmatte> either that or write your own script to do it... I'll paste an example of a script I wrote...
[11-Jun-2009 15:37:16] <rmatte> http://pastebin.com/m41034f2e
[11-Jun-2009 15:38:55] <rmatte> that script was created because windows CPU OIDs tend to wander, for each CPU that spits out a value, on a 2 CPU box the output would look like "OK|CPU1=76 CPU2=34" on a 4 CPU box it would look like "OK|CPU1=73 CPU2=48 CPU3=37 CPU4=45" and so on
[11-Jun-2009 15:39:08] <rmatte> then you just need to add the data points
[11-Jun-2009 15:40:12] <rmatte> actually I have a better example...
[11-Jun-2009 15:40:39] <rmatte> http://pastebin.com/m6aa3db28
[11-Jun-2009 15:41:05] <rmatte> that one was made because the snmp info for RAM is always a different value based on the number of physical drives in a server
[11-Jun-2009 15:41:20] <rmatte> so that script hunts down the value based on the name of the volume
[11-Jun-2009 15:41:38] <rmatte> it looks for "hrStorageRam", then polls the associated OIDs
[11-Jun-2009 15:42:47] <mloven> Thanks rmatte. So then.... and forgive me if I'm asking stupid questions... you would just create a datasource that points to this script?
[11-Jun-2009 15:43:05] <rmatte> yeh, you create a command datasource
[11-Jun-2009 15:43:26] <rmatte> and remember how I said that the output of the script should look like "OK|CPU1=73 CPU2=48 CPU3=37 CPU4=45"...
[11-Jun-2009 15:43:40] <mloven> yep
[11-Jun-2009 15:43:41] <rmatte> so you'd add CPU1 as the first datapoint, CPU2 as the second, and so on
[11-Jun-2009 15:44:20] <rmatte> that output format is Nagios syntax, but Zenoss understands it
[11-Jun-2009 15:44:36] <rmatte> let me paste you the command as it's seen in Zenoss
[11-Jun-2009 15:45:09] <mloven> excellent. We haven't really had any need to use command data sources up until now so it's still pretty new for me.
[11-Jun-2009 15:45:13] <rmatte> $$ZENHOME/Products/SNMPWindowsPerfMonitor/libexec/wincpu.sh "${dev/manageIp}" "${dev/zSnmpCommunity}"
[11-Jun-2009 15:45:14] <rmatte> $$ZENHOME/Products/SNMPWindowsPerfMonitor/libexec/winmem.sh "${dev/manageIp}" "${dev/zSnmpCommunity}"
[11-Jun-2009 15:45:35] <rmatte> I stick it in a folder in /Products, you can really put the script wherever you want on the hard drive
[11-Jun-2009 15:45:44] <rmatte> as long as the zenoss user has access to the folder
[11-Jun-2009 15:45:58] <mloven> Awesome. Thanks rmatte! That's exactly what I needed!
[11-Jun-2009 15:46:04] <rmatte> np
[11-Jun-2009 15:46:35] <rmatte> also, if snmp can't connect the script should return "Unknown"
[11-Jun-2009 15:46:46] <rmatte> as seen in my examples
[11-Jun-2009 15:47:07] <mloven> perfect, thanks!
[11-Jun-2009 15:56:38] <CamargoBP> Does anyone know how long it takes zenoss to create a rrd file? Is it the first time the data changes on the SNMP pull or does it run a process every x min that checks to see if it needs to create a rrd?
[11-Jun-2009 15:57:40] <rocket> CamargoBP: I know its after a couple of rounds of data has been collected
[11-Jun-2009 15:57:46] <rmatte> It generally does it after 3 successful polling cycles
[11-Jun-2009 15:57:59] <rmatte> so like 15 minutes give or take
[11-Jun-2009 15:58:08] <rocket> eg it might take 15 minutes if a rrd is setup for a 5 min poll cycle
[11-Jun-2009 16:00:16] <CamargoBP> So I deleted a DataPoint and saw that the rrd got deleted, then I created a new DataPoint with the same Name as the one that was deleted. My SNMP pulls every 30 seconds. I've been waiting now for 5 minutes for it to create the rrd. This is normal then?
[11-Jun-2009 16:01:02] <rmatte> I'd give it a bit longer
[11-Jun-2009 16:01:31] <rmatte> snmp pulling every 30 seconds, what, are you monitoring like 3 devices?
[11-Jun-2009 16:14:44] <CamargoBP> Currently 32 devices
[11-Jun-2009 16:15:50] <CamargoBP> So I set the "Default RRD Create Command" however Zenoss still used it's own values when creating the rrd. Are those values cached? Do I need to restart Zenoss or a certain process for the new Default take effect?
[11-Jun-2009 16:18:10] <CamargoBP> zen.zenperfsnmp: total=32 good=32 bad=0 time=1.762472 Not too bad. If it starts to take 20 + seconds I will then change that to every minute
[11-Jun-2009 16:19:37] <CamargoBP> Nope a restart of zenoss didn't take the new default RRD Create Command values. Hmmmm.... (Yes I deleted the rrd's, restarted zenoss and it created new ones but not with the new rra values)
[11-Jun-2009 16:26:07] <CamargoBP> Found the problem I accidently put a space in the command.
[11-Jun-2009 16:37:13] <rmatte> a space in which command?
[11-Jun-2009 16:47:03] <CamargoBP> After all that, when the SNMP reports 0 the rrd reports .01 http://i41.tinypic.com/x6lqgl.png
[11-Jun-2009 16:57:46] <CamargoBP> So I have to multiply my graph point by 30 to get a whole value, why would that be? ie. 30,*
[11-Jun-2009 16:58:19] <CamargoBP> http://i44.tinypic.com/23si64z.png
[11-Jun-2009 17:17:29] <CamargoBP> I shouldn't have to multiply by 30 to get a whole value, why do you think I have to do that?
[11-Jun-2009 17:28:00] <mephux_> hey anyone having issues removing the zenoss user on mac osx
[11-Jun-2009 17:31:03] <mrayzenoss> mephux_: http://dev.zenoss.org/trac/ticket/4575
[11-Jun-2009 17:31:20] <mrayzenoss> supposedly this wasn't an issue anymore, are you seeing it?
[11-Jun-2009 17:36:18] <mephux_> well i tried that..
[11-Jun-2009 17:36:47] <mephux_> the commands listed do not work.. hhmm guess I need to dig deeper.. shame
[11-Jun-2009 17:37:15] <mrayzenoss> sorry
[11-Jun-2009 17:37:20] <mrayzenoss> was this with 2.4.1?
[11-Jun-2009 17:38:42] <mephux_> not sure it was about 4 months back..
[11-Jun-2009 17:39:07] <mephux_> been lazy to fix this issue but its now bothering me.. I will be sure to update the ticket eith my steps
[11-Jun-2009 17:39:10] <mrayzenoss> so I saw the error with 2.3.3, but 2.4.0 didn't recreate it
[11-Jun-2009 17:39:42] <mrayzenoss> so knowing that timeline, you're probably on 2.3.X
[11-Jun-2009 17:40:01] <mrayzenoss> and those were the instructions I used to delete the user and group
[11-Jun-2009 17:40:25] <mephux_> I thank you for your time and information. Have a good rest of the day friend
[11-Jun-2009 17:40:28] <mephux_> late
[11-Jun-2009 17:41:58] <blocky> can anyone tell me why zenoss is appending a trailing / to whatever command i try to run through SSH
[11-Jun-2009 17:42:35] <mrayzenoss> hmm... that's a new one
[11-Jun-2009 17:43:16] <blocky> Executing command
[11-Jun-2009 17:43:16] <blocky> /usr/bin/zenplugin.py/
[11-Jun-2009 17:43:22] <blocky> /bin/sh: /usr/bin/zenplugin.py/: No such file or directory
[11-Jun-2009 17:43:41] <blocky> im pretty stumped
[11-Jun-2009 17:44:34] <blocky> it seems whatever i type in the command template box is treated like a function in python, but then it tries to execute the python script with bash
[11-Jun-2009 17:45:02] <blocky> im supposed to be using Nagios parser, correct?
[11-Jun-2009 17:46:07] <mrayzenoss> I assume so, haven't touched the zenplugins
[11-Jun-2009 17:46:46] <mrayzenoss> they're slightly deprecated in favor of the new SSH stuff
[11-Jun-2009 17:46:48] <blocky> the snmp seems to work quite a bit better, but the servers i need to monitor are behind a firewall and we can't open snmp
[11-Jun-2009 17:46:58] <blocky> er, new SSH?
[11-Jun-2009 17:47:31] <mrayzenoss> 2.4 added better SSH handling, so ZenPlugins are on their way out in favor of all remote SSH calls
[11-Jun-2009 17:47:48] <mrayzenoss> of course, the only solution right now is LinuxMonitor, and it's kinda weak
[11-Jun-2009 17:48:06] <mrayzenoss> but a lot of new ZenPacks are coming in
[11-Jun-2009 17:48:15] <mrayzenoss> and LinuxMonitor already has some patches
[11-Jun-2009 17:48:51] <blocky> would the basic idea still be to call a remote script which outputs pairs of something=value and then parse that?
[11-Jun-2009 17:49:12] <mrayzenoss> yeah, you can definitely keep doing that
[11-Jun-2009 17:49:24] <mrayzenoss> for datasources
[11-Jun-2009 17:49:35] <blocky> if I could get that working I could probaby write the scripts to gather metrics myself
[11-Jun-2009 17:49:36] <mrayzenoss> I meant instead of the zenplugins.py stuff
[11-Jun-2009 17:50:00] <blocky> how do I set up data sources independant of zenplugins.py then
[11-Jun-2009 17:50:48] <mrayzenoss> check out http://www.skills-1st.co.uk/papers/jcurry.html Methods of monitoring processes with Zenoss
[11-Jun-2009 17:51:04] <mrayzenoss> Chapter 5.2 "Zenoss and SSH" covers it
[11-Jun-2009 17:51:19] <mrayzenoss> gotta run
[11-Jun-2009 17:51:31] <blocky> cheers, thanks
[11-Jun-2009 17:54:31] <CamargoBP> So snmpwalk only returns whole numbers: UCD-SNMP-MIB::extOutput.1 = STRING: 4376 why in the world would my max be: 2.19 and current: 1.27 http://i44.tinypic.com/1zggv8.png
[11-Jun-2009 17:55:49] <CamargoBP> How in the world did it say 6 + 6 = 12.19 Shouldn't it be 12 I guess that's the real question.
[11-Jun-2009 17:56:14] <CamargoBP> My example didn't use actual numbers that zenoss reported that was an example
[11-Jun-2009 17:56:20] <rhettardo> what does the datasource say
[11-Jun-2009 17:57:27] <rhettardo> also you'll have to make a new data source if you change the rra's to 60 second. you cant add those datapoints in on a prexisting datasource (as far as i know you couldnt in the 'older' pre 1.3 version)
[11-Jun-2009 17:58:10] <CamargoBP> http://pastebin.com/d1a699c27
[11-Jun-2009 17:58:41] <CamargoBP> Yeah I delete the rrd's if I make a change
[11-Jun-2009 17:58:51] <rhettardo> good
[11-Jun-2009 17:59:52] <rhettardo> i what did you use to get this output
[11-Jun-2009 18:00:13] <rhettardo> forget the i
[11-Jun-2009 18:00:31] <CamargoBP> rrdtool info Postfix\ -Sent_Sent.rrd
[11-Jun-2009 18:01:18] <rhettardo> try 'rrdtool fetch Postfix\ -Sent_Sent.rrd AVERAGE -r 30 -s -10m
[11-Jun-2009 18:01:20] <rhettardo> '
[11-Jun-2009 18:02:18] <CamargoBP> http://pastebin.com/d614d06ac
[11-Jun-2009 18:03:34] <lukecyca> Is there a way to get a list of a host's interfaces via xml-rpc?
[11-Jun-2009 18:12:41] <blocky> is there a way to have two devices with the same ip
[11-Jun-2009 18:12:58] <rmatte> blocky: for what purpose?
[11-Jun-2009 18:13:00] <blocky> i tried creating different hosts in /etc/hosts
[11-Jun-2009 18:13:09] <blocky> because they are different machines, albeit behind a firewall
[11-Jun-2009 18:13:18] <blocky> and they are accessible through different ssh ports
[11-Jun-2009 18:13:50] <rmatte> hmmm, Zenoss checks for other devices with the same IP to avoid duplication, so there's no real easy way to do that
[11-Jun-2009 18:13:57] <blocky> oh
[11-Jun-2009 18:14:07] <blocky> wow
[11-Jun-2009 18:14:16] <blocky> well can i disable that?
[11-Jun-2009 18:15:12] <rmatte> no idea, doubtful, are you just trying to do up/down monitoring?
[11-Jun-2009 18:15:36] <blocky> no im gathering network traffic and performance data
[11-Jun-2009 18:15:39] <rmatte> ah
[11-Jun-2009 18:15:52] <rmatte> Zenoss also uses the default ssh port for monitoring
[11-Jun-2009 18:16:01] <rmatte> so that's another thing you'll have to get around somehow
[11-Jun-2009 18:16:10] <rmatte> I doubt you'll be able to do it by any conventional means
[11-Jun-2009 18:16:32] <rmatte> you might have to write a script to collect the data then feed it in to Zenoss, but that wouldn't be simple
[11-Jun-2009 18:16:56] <blocky> hmm
[11-Jun-2009 18:17:31] <blocky> i cant believe nobody thought this would be used through a NAT
[11-Jun-2009 18:17:55] <rmatte> you could always setup a vpn tunnel through the firewall and monitor through that
[11-Jun-2009 18:18:00] <rmatte> that would probably be the easiest
[11-Jun-2009 18:18:19] <rmatte> it's not designed to be used through a NAT, not many monitoring solutions are other than agent based solutions
[11-Jun-2009 18:18:32] <rmatte> but agentless monitoring is far more desirable in the majority of cases
[11-Jun-2009 18:18:37] <rhettardo> CamargoBP interesting
[11-Jun-2009 18:19:00] <rhettardo> we can see in the datasource its not getting whole numbers
[11-Jun-2009 18:19:01] <blocky> a vpn is not ideal
[11-Jun-2009 18:19:08] <blocky> but i may have found a solution
[11-Jun-2009 18:19:15] <blocky> apparently the duplicate check is in the GUI only
[11-Jun-2009 18:20:43] <rhettardo> CamargoBP is it possibel you're using a counter instead of a gauge?
[11-Jun-2009 18:20:50] <rhettardo> or the other way around?
[11-Jun-2009 18:21:19] <blocky> wow i guess i just missed all the developers this morning eh
[11-Jun-2009 18:21:34] <rhettardo> blocky yeah mray just left
[11-Jun-2009 18:21:57] <blocky> does he idle often, or just come in for the chats
[11-Jun-2009 18:22:22] <rhettardo> he's usually here during work hours
[11-Jun-2009 18:22:24] <rhettardo> or seems that way
[11-Jun-2009 18:22:39] <blocky> its 4:22PM out west
[11-Jun-2009 18:22:50] <rmatte> we use VPN tunnels to connect to all of the clients that we monitor and it works fine
[11-Jun-2009 18:23:08] <rmatte> you get the very occasional flap, but it's 99% stable
[11-Jun-2009 18:23:15] <rhettardo> i use vpn's to monitor everything
[11-Jun-2009 18:23:17] <rhettardo> openssl
[11-Jun-2009 18:23:20] <blocky> what if somebody breaks into one network
[11-Jun-2009 18:23:27] <blocky> they have access to everything?
[11-Jun-2009 18:23:39] <rmatte> nope
[11-Jun-2009 18:23:47] <rmatte> each client gets their own VM on their own VLAN
[11-Jun-2009 18:23:49] <blocky> and all the internal networks need to have exclusive ip ranges?
[11-Jun-2009 18:23:59] <rmatte> and we don't have routing between VMs on our end anyways
[11-Jun-2009 18:24:41] <blocky> i guess its one solution
[11-Jun-2009 18:25:14] <blocky> im sure there must be a way to disable that checking
[11-Jun-2009 18:25:19] <rmatte> it's really the only solution for us
[11-Jun-2009 18:25:55] <rmatte> even if you do disable that checking, you still need to get Zenoss to somehow do all of it's monitoring on a non-standard port for one of the 2 devices
[11-Jun-2009 18:25:59] <CamargoBP> rhettardo: I am using DERIVE
[11-Jun-2009 18:26:05] <rmatte> or maybe for both, depending on how you setup the ports
[11-Jun-2009 18:26:42] <blocky> hmm
[11-Jun-2009 18:26:44] <blocky> this is a real problem
[11-Jun-2009 18:26:56] <blocky> maybe ill check back tomorrow morning and see if any devs show up
[11-Jun-2009 18:26:57] <rmatte> maybe there's some option that you can set in zproperties
[11-Jun-2009 18:27:16] <rmatte> for changing the port I mean
[11-Jun-2009 18:28:52] <rmatte> you could put something between the devices in Zenoss that redirects traffic to whatever port on whatever device based on what phony IP you connect to
[11-Jun-2009 18:28:56] <rmatte> no idea how you'd do that though
[11-Jun-2009 18:29:06] <rmatte> between the devices & zenoss rather
[11-Jun-2009 18:29:32] <rhettardo> CamargoBP i see
[11-Jun-2009 18:29:38] <blocky> yea you can change the port
[11-Jun-2009 18:29:50] <blocky> i think this might work
[11-Jun-2009 18:30:12] <rmatte> I would personally write a script that runs as a cron job on those servers and dumps the info back out to the zenoss box via ssh, then have zenoss grab the data and write a custom template to graph it
[11-Jun-2009 18:32:08] <blocky> i think i figured it out
[11-Jun-2009 18:32:12] <rmatte> oh?
[11-Jun-2009 18:36:45] <blocky> it doesnt seem to check if you use the easy device add
[11-Jun-2009 18:36:52] <rmatte> ah lol
[11-Jun-2009 18:37:14] <rmatte> that's actually pretty funny, making it even easier for people to screw up
[11-Jun-2009 18:37:33] <rmatte> you could have probably done it from the commandline too
[11-Jun-2009 18:37:37] <rmatte> there's probably some option
[11-Jun-2009 18:54:10] <blocky> cool i got cpu load
[11-Jun-2009 18:54:15] <blocky> now i need network traffic
[11-Jun-2009 18:55:15] <blocky> and i need a report emailed to me every morning
[11-Jun-2009 19:44:52] <cgibbons> ah-hah!
[11-Jun-2009 19:46:27] <HKhan> Hmmm how do I make zenoss monitor TCP ports? I added a "service" but it says port number is set to "0", but i can't find where to change it :S
[11-Jun-2009 19:55:14] <HKhan> I went to Classes > Services > Add Service > typed a name for the service, but when i try to edit the service to change the port number, its not there :s
[11-Jun-2009 21:35:19] <cgibbons> h5xm2JQh
[11-Jun-2009 21:35:33] <cgibbons> h5xm2JQH
[11-Jun-2009 21:37:04] <cgibbons> whoops
[11-Jun-2009 21:48:01] <perr0> nice password
[11-Jun-2009 21:48:26] <perr0> far too complex for me thou
[11-Jun-2009 21:56:13] <rhettlap> :O
[11-Jun-2009 22:05:58] <rocket>
[11-Jun-2009 22:30:20] mephux_ is now known as mephux
[12-Jun-2009 00:00:42] [disconnected at Fri Jun 12 00:00:42 2009]
[12-Jun-2009 00:00:42] [connected at Fri Jun 12 00:00:42 2009]
[12-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[12-Jun-2009 03:27:29] tehhobbit is now known as loke
[12-Jun-2009 03:27:41] loke is now known as tehhobbit
[12-Jun-2009 08:17:27] <crow> Whenever I try to .iteritems a dict in zope, it logs me out of zenoss. no errors in the event.log. Anyone ever dealt with that?
[12-Jun-2009 09:04:56] <venturaville> first I've heard of the linuxfest
[12-Jun-2009 09:05:04] <venturaville> apparently it is almost in my backyard
[12-Jun-2009 09:07:36] crow is now known as ckrough
[12-Jun-2009 09:11:59] <jb> kells?
[12-Jun-2009 09:12:22] <rocket> jb?
[12-Jun-2009 09:32:40] <ckrough> meh. doing this: L.sort(key=operator.itemgetter('sysUpTime')) in a script logs me out of zenoss
[12-Jun-2009 09:32:51] <ckrough> same with 'for k,v in d' loops
[12-Jun-2009 09:33:32] <ckrough> ever had that happen?
[12-Jun-2009 09:33:35] <__ian__> that's pretty messed up
[12-Jun-2009 09:33:47] <__ian__> what kind of script?
[12-Jun-2009 09:33:55] <ckrough> python script in zope
[12-Jun-2009 09:34:00] <__ian__> a ZenScriptBase subclass?
[12-Jun-2009 09:34:01] <__ian__> oh
[12-Jun-2009 09:34:21] <ckrough> /zport/dmd/Devices/myscript
[12-Jun-2009 09:36:05] <ckrough> wow. its definitely the action of sorting the list. if I just return the list without sorting it works fine
[12-Jun-2009 09:36:11] * ckrough wishes there was an error message
[12-Jun-2009 09:44:14] <__ian__> What comprises the list? What kind of objects?
[12-Jun-2009 09:45:28] <__ian__> Rather, what composes the list; what does the list comprise. Sorry, I had a bad grammar moment.
[12-Jun-2009 09:46:37] <ckrough> My goal is to sort a 2D dict on the second value of each item: d['item1']['a':1, 'b':2, 'c':3}
[12-Jun-2009 09:46:52] <__ian__> right, but what are the items?
[12-Jun-2009 09:46:52] <ckrough> so there are a bunch of items and I want to return them sorted on 'b'
[12-Jun-2009 09:47:09] <__ian__> if your script doesn't have perms to access one of the items you'll get booted to the login
[12-Jun-2009 09:47:44] <ckrough> hmm. Im building the dict in the script. so the 'item' is just a key and the value is another dict full of attributes
[12-Jun-2009 09:48:08] <__ian__> so this is, at the moment, not at all Zopey?
[12-Jun-2009 09:48:13] <ckrough> Im using the List just to sort it before printing
[12-Jun-2009 09:48:13] <__ian__> other than running in Zope?
[12-Jun-2009 09:48:20] <ckrough> Ill pastebin it, sec
[12-Jun-2009 09:50:02] <ckrough> http://pastebin.com/m6e9b9696
[12-Jun-2009 09:50:15] <ckrough> remove the L.sort line (2nd from bottom) and it works
[12-Jun-2009 09:50:22] <ckrough> leave it there and you get logged out
[12-Jun-2009 09:51:20] <ckrough> its probably a logic error on my part since my python-fu is not to hot
[12-Jun-2009 09:51:29] <ckrough> but, without an error message Im a bit lost
[12-Jun-2009 09:52:13] <ckrough> x.timeframe is hundreths of seconds passed on the url as ?timeframe=n
[12-Jun-2009 09:52:53] <ckrough> Ive tried using a simple custom report in the zenoss gui, but its too slow for what we need. Im hoping I can speed it up once I get the basic layout working here
[12-Jun-2009 09:54:01] <rocket> ckrough: logged out how?
[12-Jun-2009 09:54:19] <rocket> ckrough: is this from the webinterface or cmdline?
[12-Jun-2009 09:54:22] <ckrough> just taken to the zenoss login screen as if you cleared you cache when went to there
[12-Jun-2009 09:54:26] <ckrough> webinterface
[12-Jun-2009 09:54:42] <__ian__> looks like operator.itemgetter is the culprit
[12-Jun-2009 09:54:51] <ckrough> Ive tried doing it as the zenoss admin as well and same error
[12-Jun-2009 09:55:05] <rocket> is this for a report?
[12-Jun-2009 09:55:08] <ckrough> yes
[12-Jun-2009 09:55:24] <ckrough> but I just want text on a blank page, no fancy templating
[12-Jun-2009 09:55:39] <__ian__> if you do this, it works: http://pastebin.com/m17d4e350
[12-Jun-2009 09:56:48] <rocket> yea it looks like the sort function in ckrough's code it the cause of the issue
[12-Jun-2009 09:56:52] <ckrough> nice
[12-Jun-2009 09:56:59] <ckrough> dont blame ckrough
[12-Jun-2009 09:57:08] <rocket> heh
[12-Jun-2009 09:57:10] <ckrough> am I doing something invalid, or is zope choking
[12-Jun-2009 09:57:21] <rocket> I would guess its memory starvation
[12-Jun-2009 09:57:44] <rocket> take a look at top on the server when running your code
[12-Jun-2009 09:57:47] <ckrough> would have to be an artificial limit in zope
[12-Jun-2009 09:58:00] <rocket> yes a timeout I am sure in zope
[12-Jun-2009 09:58:01] <ckrough> 32GB ram on that box and Im only sorting < 20 devices
[12-Jun-2009 09:58:06] <rocket> its taking too long to sort
[12-Jun-2009 09:58:11] <ckrough> ah
[12-Jun-2009 09:58:15] <rocket> its probably in some sort of infinite loop
[12-Jun-2009 09:58:27] <ckrough> lol. 'sort' of infinite loop
[12-Jun-2009 09:58:31] <ckrough> punny guy
[12-Jun-2009 09:58:36] <rocket>
[12-Jun-2009 09:58:52] <rocket> where as the code pasted by __ian__ creates a new compare function
[12-Jun-2009 09:58:58] <rocket> that is very specific
[12-Jun-2009 09:59:11] <rocket> is a sysuptime greater than b sysuptime
[12-Jun-2009 09:59:17] <ckrough> that will work for this report
[12-Jun-2009 10:00:04] <rocket> ah I see I think your code is always returning true to sort
[12-Jun-2009 10:00:30] <rocket> key=operator.itemgetter('sysUpTime') probably always looks like a true
[12-Jun-2009 10:00:52] <rocket> so it looks like L.sort(true) which goes on forever etc ..
[12-Jun-2009 10:00:56] <ckrough> sorry, I dont understand itemgetter well enough to know how that messes it up
[12-Jun-2009 10:01:50] <rocket> I think its the setting something to = that returns true .. its saying it succeeded in setting key = to something else
[12-Jun-2009 10:02:34] <__ian__> hmm, it should work. operator.itemgetter('mykey')({mykey:x}) == x
[12-Jun-2009 10:04:29] <__ian__> I'm leaning towards some probably arbitrary restriction in Zope's execution of Script objects.
[12-Jun-2009 10:04:41] <__ian__> They really like to do that.
[12-Jun-2009 10:04:50] <rocket> __ian__: thats a definite possibility ..
[12-Jun-2009 10:05:16] <ckrough> that makes me feel better
[12-Jun-2009 10:06:00] <venturaville> shouldn't that just be L.sort(key='sysUpTime')?
[12-Jun-2009 10:07:06] <__ian__> no. Confusingly, you pass a callable as key.
[12-Jun-2009 10:07:11] <venturaville> no never mind I see now
[12-Jun-2009 10:07:16] <venturaville> wow that is ugly
[12-Jun-2009 10:07:29] <__ian__> It is the "sort key", rather than the "dictionary key"
[12-Jun-2009 10:07:47] <__ian__> makes it more versatile, but I wish they'd named it something else.
[12-Jun-2009 10:07:47] <kisielk_home> wow, last night our Zenoss autodiscovery went totally amiss
[12-Jun-2009 10:08:05] <ckrough> kisielk_home: another faithful zenner monitoring the internet?
[12-Jun-2009 10:08:14] <kisielk_home> it discovered a VPN interface on some user's workstation, and then went on to probe a local university's entire public IP space
[12-Jun-2009 10:08:36] <kisielk_home> now I have like 1700+ devices in /Devices/Discovered that aren't even ours
[12-Jun-2009 10:08:45] <ckrough> kisielk_home: you can limit the ip ranges it will walk
[12-Jun-2009 10:08:57] <kisielk_home> yes, and somehow that limit disappeared
[12-Jun-2009 10:09:01] <kisielk_home> I think during the upgrade from 2.3.3 to 2.4
[12-Jun-2009 10:09:05] <ckrough> kisielk_home: or you could bill the university for monitoring services
[12-Jun-2009 10:09:10] <kisielk_home> because I'm fairly certain I had 10/8 in there
[12-Jun-2009 10:09:29] <ckrough> are you a single zen box or distributed collectors?
[12-Jun-2009 10:09:36] <kisielk_home> single
[12-Jun-2009 10:09:40] <kisielk_home> just using Zenoss core
[12-Jun-2009 10:10:41] <kisielk_home> I realized something was wrong when I suddenly received 7000+ emails from Zenoss
[12-Jun-2009 10:11:34] <__ian__> kisielk_home: did just your setting of the limit disappear, or the ability to set a limit?
[12-Jun-2009 10:11:52] <__ian__> because if it's the latter I'm thinking we should probably fix that
[12-Jun-2009 10:12:09] <ckrough> you can still set it, we still do
[12-Jun-2009 10:12:14] <__ian__> phew.
[12-Jun-2009 10:12:14] <ckrough> let me see where we set it
[12-Jun-2009 10:12:39] <kisielk_home> __ian__: the setting of the limit
[12-Jun-2009 10:12:51] <kisielk_home> I looked at the field and it was blank
[12-Jun-2009 10:12:57] <kisielk_home> then I cried for a while
[12-Jun-2009 10:13:24] <ckrough> in /opt/zenoss/bin/zenoss_discovery : ${ZENHOME}/bin/zendisc run --chunk 64 --net=10.5.0.0/16 ...additional_args...
[12-Jun-2009 10:13:41] <ckrough> not sure if thats 'the way' or just 'our way'. mutliple admins on these boxes
[12-Jun-2009 10:13:52] <ckrough> we run multiple discovery processes via cron
[12-Jun-2009 10:14:21] <kisielk_home> there's a field in the collector preferences
[12-Jun-2009 10:14:23] <__ian__> that's how we do it on our canary box too
[12-Jun-2009 10:14:51] <ckrough> if I remember correctly we had to move discovery back to the master box, running it on collectors was gapping graphs
[12-Jun-2009 10:15:42] <ckrough> running 2.2.4 so your mileage probably will vary
[12-Jun-2009 10:21:47] <capricorn-80> hi rocket
[12-Jun-2009 10:22:08] <rocket> Hello capricorn-80
[12-Jun-2009 10:22:22] <capricorn-80> how are you ?
[12-Jun-2009 10:22:28] <rocket> busy
[12-Jun-2009 10:22:42] <rocket> budgets time at work ..
[12-Jun-2009 10:22:52] <rocket> and you?
[12-Jun-2009 10:24:55] <capricorn-80> same here
[12-Jun-2009 10:25:04] <capricorn-80> i guess u didnt find time
[12-Jun-2009 10:25:25] <rocket> no sorry .. I would probably ask again next thursday
[12-Jun-2009 10:25:34] <rocket> thursdays the developers are usually here
[12-Jun-2009 10:25:37] <rocket> from zenoss
[12-Jun-2009 10:25:57] <rocket> unless someone like __ian__ here can help you ..
[12-Jun-2009 10:26:04] <__ian__> I can take a stab
[12-Jun-2009 10:26:26] <rocket> __ian__: he is having trouble with one interface not showing up inside zenoss
[12-Jun-2009 10:26:49] <capricorn-80> rocket: you got that pic ?
[12-Jun-2009 10:26:51] <rocket> it looks like it shows up in the modelling debug output but the web ui doesnt show it
[12-Jun-2009 10:27:00] <rocket> capricorn-80: not anymore .. forgot to save it
[12-Jun-2009 10:27:47] <rocket> __ian__: this is on a cisco router with lots of vlans ..
[12-Jun-2009 10:28:21] <__ian__> ha, and you've officially passed my "knowledge of how to actually use the code I write" threshold
[12-Jun-2009 10:28:39] <rocket> __ian__:
[12-Jun-2009 10:29:14] <__ian__> Is it just the web UI, or does it just not exist in the database?
[12-Jun-2009 10:29:28] <rocket> __ian__: I hadnt debugged that far with him ..
[12-Jun-2009 10:30:48] <ckrough> what type of interface is it?
[12-Jun-2009 10:31:01] <capricorn-80> ckrough: its router interface
[12-Jun-2009 10:31:13] <ckrough> I mean gige, te, vlan, po, etc...
[12-Jun-2009 10:31:13] <__ian__> capricorn-80: fire up zendmd and see if your interface is in the output of: find('myrouterid').os.interfaces()
[12-Jun-2009 10:34:41] <capricorn-80> __ian__: can you get the full command because zendmd is giving me some command menu
[12-Jun-2009 10:34:59] <__ian__> $ zendmd
[12-Jun-2009 10:35:07] <__ian__> >>> find('myrouterid').os.interfaces()
[12-Jun-2009 10:38:31] <capricorn-80> __ian__: http://pastebin.com/d384ba6c3
[12-Jun-2009 10:39:24] <rocket> capricorn-80: is this the router? or your zenoss machineid again?
[12-Jun-2009 10:41:22] <capricorn-80> oh sorry
[12-Jun-2009 10:41:29] <capricorn-80> rocket: sorry
[12-Jun-2009 10:41:31] <__ian__> capricorn-80: what rocket said, and also, what's the interface you're looking for?
[12-Jun-2009 10:41:34] <CamargoBP> My graph is not reporting whole numbers, however SNMP only pulls whole numbers. I am graphing MAX not AVERAGE. Would having a step value of 30 cause this? Here is the graph, the blue one http://i39.tinypic.com/rbgnlv.png
[12-Jun-2009 10:41:35] <capricorn-80> i wrote wrong ip
[12-Jun-2009 10:43:25] <capricorn-80> __ian__: http://pastebin.com/m6422f946
[12-Jun-2009 10:43:28] <capricorn-80> rocket: http://pastebin.com/m6422f946
[12-Jun-2009 10:43:44] <__ian__> so, is your interface in that list?
[12-Jun-2009 10:45:52] <capricorn-80> no
[12-Jun-2009 10:46:12] <capricorn-80> it is
[12-Jun-2009 10:46:15] <capricorn-80> IpInterface at /zport/dmd/Devices/Network/Router/Cisco/devices/10.254.95.254/os/interfaces
[12-Jun-2009 10:46:26] <capricorn-80> oh its null0
[12-Jun-2009 10:46:36] <__ian__> so it's there, the second one?
[12-Jun-2009 10:47:15] <capricorn-80> __ian__: yes
[12-Jun-2009 10:47:27] <capricorn-80> __ian__: but i cant see any trafic graphs on it
[12-Jun-2009 10:47:47] <capricorn-80> i mean the traffic is 0 on this interface
[12-Jun-2009 10:48:12] <capricorn-80> no inbound of outbound traffic
[12-Jun-2009 10:51:53] <capricorn-80> [<IpInterface at /zport/dmd/Devices/Network/Router/Cisco/devices/10.254.95.254/os/interfaces/EOBC0_6> and its IP Addresses 127.0.0.22/8
[12-Jun-2009 10:52:14] <__ian__> wait, so you can see it in the UI?
[12-Jun-2009 10:52:54] <capricorn-80> __ian__: can you tell me which i can see ?
[12-Jun-2009 10:53:10] <capricorn-80> yes i can see that
[12-Jun-2009 10:53:18] <__ian__> On the OS tab of the router, the inteface is listed?
[12-Jun-2009 10:53:24] <capricorn-80> its not showing the IP address but Null0
[12-Jun-2009 10:53:29] <capricorn-80> yes
[12-Jun-2009 10:54:22] <capricorn-80> 10.254.95.254 this is the interface as this is interface i am using for polling . i mean snmp is enabled on this interface
[12-Jun-2009 10:55:09] <__ian__> Then I am totally misunderstanding the problem. What's the issue? It isn't collecting, or something?
[12-Jun-2009 10:59:06] <capricorn-80> wait i upload the image
[12-Jun-2009 10:59:17] <capricorn-80> that will give you better idea
[12-Jun-2009 11:02:55] <capricorn-80> __ian__: http://imagebin.org/52246
[12-Jun-2009 11:03:37] <ckrough> just catching up, you are referring to the EOBC interface?
[12-Jun-2009 11:06:31] <capricorn-80> there are two issues. one is EOBC and other was Null0
[12-Jun-2009 11:06:44] <ckrough> neither will be recognized by the default templates
[12-Jun-2009 11:06:45] <capricorn-80> Null0 was issue
[12-Jun-2009 11:07:17] <__ian__> why would null0 ever have any information associated with it?
[12-Jun-2009 11:07:26] <capricorn-80> but with ian command i got it that Null0 is the router interface to which the zenoss is get information
[12-Jun-2009 11:07:36] <ckrough> not sure if they've added them or not, but they get dumped in propVirtual template
[12-Jun-2009 11:08:04] <capricorn-80> ckrough: i just added my main interface ip and that is 10.254.95.254
[12-Jun-2009 11:08:19] <capricorn-80> i didnt add another interface
[12-Jun-2009 11:11:06] <capricorn-80> Null0 is the interface which got this ip address 10.254.95.254
[12-Jun-2009 11:12:11] <capricorn-80> EOBC interface got ip address 127.0.0.22/8 which is strange
[12-Jun-2009 11:14:41] <ckrough> If I understand it correctly, which I may now, EOBC is an out of band interface between the ASIC and the supervisor, but not an interface you would poll
[12-Jun-2009 11:14:45] <ckrough> may *not*
[12-Jun-2009 11:15:24] <capricorn-80> ckrough: yes
[12-Jun-2009 11:21:53] <capricorn-80> i cant figure about this interface
[12-Jun-2009 11:25:41] <ckrough> Can you restate the problem?
[12-Jun-2009 11:33:58] <lukecyca> omg. My Zenoss is auto-discovering networks that aren't mine. Someone's IDS is lighting up like a christmas tree right now... :/
[12-Jun-2009 11:34:01] <eichenwald> where is WinUser / WinPassword stored in the zProperties of a device class
[12-Jun-2009 11:34:10] <eichenwald> is it not stored in the db?
[12-Jun-2009 11:34:22] <__ian__> it is stored in the db
[12-Jun-2009 11:34:35] <__ian__> as attribute zWinUser and zWinPassword, on the device class
[12-Jun-2009 11:36:26] <eichenwald> all I am seeming is an events db
[12-Jun-2009 11:38:13] <capricorn-80> ckrought: wait
[12-Jun-2009 11:38:18] <ckrough> lukecyca: did this just start happenning and did you recent upgrade
[12-Jun-2009 11:38:26] <ckrough> capricorn-80: Ill be here a while
[12-Jun-2009 11:38:53] <capricorn-80> ckrought: IP Addresses 127.0.0.22/8
[12-Jun-2009 11:38:58] <capricorn-80> why this ip address ?
[12-Jun-2009 11:39:04] <capricorn-80> i never saw this kind of ip address
[12-Jun-2009 11:40:02] <lukecyca> ckrough: It's my own fault. One of our computers had a VPN connection open to someone else's network. Zenoss modeled that and it showed up in the list of networks. I ran auto-discovery on some of our networks, and included that one by mistake.
[12-Jun-2009 11:40:17] <lukecyca> 4000 hosts added before I realized what was going on...
[12-Jun-2009 11:40:54] <ckrough> capricorn-80: all of my cisco routers have similar addresses for EOBC as well
[12-Jun-2009 11:41:05] <lukecyca> I'll delete them through zendmd easily enough... but my whole zenoss install is crawling. I guess it's probably trying to model them all of something
[12-Jun-2009 11:41:07] <ckrough> let me ask someone, sec
[12-Jun-2009 11:41:32] <ckrough> lukecyca: you could kill the discovery process
[12-Jun-2009 11:44:13] <lukecyca> Done. I think I've got a handle on it now. This is pretty embarrassing though!
[12-Jun-2009 11:44:45] <ckrough> capricorn-80: ok, the network 127.0.0.0/8 is used for all of the internal communications on the backplane in the catalyst series. The actual IP (22) is possibly dictated by the slot the super is in. There are probably no scenarios where you will care about monitoring it.
[12-Jun-2009 11:45:11] <ckrough> capricorn-80: its the 'internal' management interface for the chassis
[12-Jun-2009 11:45:43] <ckrough> lukecyca: someone had that same issue a couple hours ago. it happens to everyone eventually
[12-Jun-2009 11:46:31] <lukecyca> heh ok thanks. That's good to hear I guess
[12-Jun-2009 11:49:20] <capricorn-80> ckrought: ok
[12-Jun-2009 11:50:09] <capricorn-80> one more thing i figure out is that my main ip interface ip address is 10.254.95.254 but there is no inbound and outbound traffic on it
[12-Jun-2009 11:50:27] <capricorn-80> because the traffic is on vlans ip addresses right ?
[12-Jun-2009 11:50:32] <ckrough> thats your management interface?
[12-Jun-2009 11:51:01] <ckrough> the octets still get reported on the interfaces, regardless of vlan
[12-Jun-2009 11:51:16] <ckrough> when you do an snmpget against that interface does it show traffic?
[12-Jun-2009 11:51:53] <capricorn-80> wait let me check it again
[12-Jun-2009 11:54:39] <capricorn-80> ckrought can tell the switch use with snmpget ?
[12-Jun-2009 11:56:04] <eichenwald> so I made a backup and I notice the large backup file is .fs, so is this what the devices are stored in instead of the mysql db? and how do I look through it
[12-Jun-2009 11:59:11] <ckrough> capricorn-80: sec
[12-Jun-2009 12:00:57] <capricorn-80> snmpget -c communitystring -v 1 10.254.95.254
[12-Jun-2009 12:01:09] <capricorn-80> but its asking me for more switches
[12-Jun-2009 12:01:32] <rhettlap> you need the oid
[12-Jun-2009 12:01:33] <ckrough> you'll need to walk it first and figure out the snmp index of your management interface
[12-Jun-2009 12:01:42] <ckrough> then you can specify an OID for that index and get the traffic
[12-Jun-2009 12:02:03] <ckrough> there may be a better way to do the index thing, but I dont spend enough time on snmpget
[12-Jun-2009 12:02:15] <capricorn-80> ok
[12-Jun-2009 12:38:29] <capricorn-80> ckrought: no traffic on 10.254.95.254
[12-Jun-2009 12:38:57] <capricorn-80> ckrought: its not showing me the interface
[12-Jun-2009 12:39:27] <capricorn-80> ckrought: http://pastebin.com/m4848239e
[12-Jun-2009 12:41:03] <ckrough> capricorn-80: you are specifically looking for the amount of management traffic?
[12-Jun-2009 12:42:13] <capricorn-80> ckrought: i am interested in in/out traffic on interface which got ip 10.254.95.254
[12-Jun-2009 13:06:38] <FooDog> i'm having exactly this problem: http://forums.zenoss.com/viewtopic.php?p=35840 ... any ideas on troubleshooting?
[12-Jun-2009 13:29:23] willwh1 is now known as willwh
[12-Jun-2009 13:30:45] <cgibbons> hmm
[12-Jun-2009 13:45:05] <capricorn-80> I just added my windows server to zenoss its giving some errors in graph. the only graph i can see is CPU and its written Missing RRD Tool for 10.254..3.30 which is machine ip address
[12-Jun-2009 13:46:30] <capricorn-80> snmp-informant
[12-Jun-2009 13:50:04] <capricorn-80> anyone there ?
[12-Jun-2009 13:50:12] <capricorn-80> need some info about snmp informant
[12-Jun-2009 13:57:09] <capricorn-80> i m installing it on one of my running server
[12-Jun-2009 13:57:16] <capricorn-80> i hope it will not create any problem
[12-Jun-2009 13:57:24] <capricorn-80> please need you wording on it
[12-Jun-2009 14:05:38] <capricorn-80> i m getting the output from snmpwalk -v 1 -c communitystring 10.254.66.30 1.3.6.1.4.1.9600
[12-Jun-2009 14:05:52] <capricorn-80> but still rrd tool error in cpu graph
[12-Jun-2009 14:06:18] <__ian__> capricorn-80: how long has it been since you added the device with the missing rrd file?
[12-Jun-2009 14:08:03] <capricorn-80> 10 mins
[12-Jun-2009 14:08:08] <__ian__> It's a confusing error, but that's perfectly normal and expected. There simply aren't data yet. Wait half an hour or so and your graph will show up, once some data has been collected against the device.
[12-Jun-2009 14:08:28] <__ian__> There's an outstanding ticket to make that message say something like, "Wait half an hour."
[12-Jun-2009 14:08:28] <capricorn-80> i have installed snmp informant
[12-Jun-2009 14:08:43] <capricorn-80> and i can get all values using snmp
[12-Jun-2009 14:08:55] <capricorn-80> ok but its saying error
[12-Jun-2009 14:09:07] <__ian__> what error, exactly
[12-Jun-2009 14:10:05] <capricorn-80> Missing RRD File . 10.254.66.30 cpuPercentProcessorTime_cpuPercentProcessorTime
[12-Jun-2009 14:10:09] <__ian__> Right.
[12-Jun-2009 14:10:28] <__ian__> The RRD file is missing because it hasn't been created yet, because the device hasn't been around long enough to be collected against.
[12-Jun-2009 14:10:45] <capricorn-80> ok
[12-Jun-2009 14:10:56] <__ian__> Like I said, a confusing error message, but that always happens.
[12-Jun-2009 14:11:21] <capricorn-80> ok
[12-Jun-2009 14:11:29] <capricorn-80> but how much time it will take ?
[12-Jun-2009 14:11:36] <__ian__> like half an hour or so
[12-Jun-2009 14:11:43] <__ian__> you can cycle the daemons by hand if you want
[12-Jun-2009 14:12:37] <__ian__> but if you just give it a few minutes they'll all pick up the device on their own.
[12-Jun-2009 14:13:04] <capricorn-80> how about 10 mins ?
[12-Jun-2009 14:13:17] <__ian__> half an hour.
[12-Jun-2009 14:13:20] <__ian__> depends on your cycle time
[12-Jun-2009 14:13:56] <capricorn-80> ok . let me wait n see
[12-Jun-2009 14:14:29] <capricorn-80> and i also asked about snmp informant that i am running it on my production server so i hope it will not create some problem
[12-Jun-2009 14:14:38] <__ian__> beats me, but I doubt it.
[12-Jun-2009 14:15:04] <capricorn-80> ah
[12-Jun-2009 14:15:18] <capricorn-80> can you explain it ?
[12-Jun-2009 14:15:31] <__ian__> I don't know what the antecedent of "it" is
[12-Jun-2009 14:16:09] <capricorn-80> ok
[12-Jun-2009 14:16:15] <capricorn-80> lets hope for the best
[12-Jun-2009 14:16:36] <__ian__> but if it's SNMP Informant, no, I know very little about Windows.
[12-Jun-2009 14:17:10] <__ian__> corner cgibbons or cluther or kells when they show their faces, they might be able to tell you more.
[12-Jun-2009 14:17:34] <cgibbons> I probably know the most about Windows @ Zenoss but even I have very rarely touched SNMP Informant. But we recommended it as an agent solution to get performance data out of Windows systems using SNMP since the built-in agent from Microsoft doesn't do it well.
[12-Jun-2009 14:17:45] <cgibbons> but since then we always recommend the WMI/Perfmon based solution, but that's in enterprise only.
[12-Jun-2009 14:18:03] <capricorn-80> ahan
[12-Jun-2009 14:18:03] <cgibbons> there are a few community zenpacks that gather performance data via WMI as well, so you wouldn't need SNMP Informant.
[12-Jun-2009 14:18:31] <capricorn-80> can you tell me some good docs or site ?
[12-Jun-2009 14:19:06] <cgibbons> SNMP Informant is its own product, so their own site has decent docs, etc.
[12-Jun-2009 14:19:55] <capricorn-80> no i am talking about zenpacks performance data via wmi
[12-Jun-2009 14:20:37] <__ian__> http://www.zenoss.com/community/projects/zenpacks/
[12-Jun-2009 14:20:50] <cgibbons> yep, this one specifically: http://www.zenoss.com/community/projects/zenpacks/windowswmidevicetemplate
[12-Jun-2009 14:27:17] <capricorn-80> ok thank s
[12-Jun-2009 14:43:58] <capricorn-80> @cgibbons: i cant able to monitor my main interface of the router
[12-Jun-2009 14:45:18] <capricorn-80> anyways related to windows problem
[12-Jun-2009 14:45:31] <capricorn-80> its been almost 30 mins but still that RRDTool error
[12-Jun-2009 14:53:43] <cgibbons> try this: zenperfsnmp run -v10 -d yourdevicenamehere
[12-Jun-2009 14:58:45] <capricorn-80> @cgibbons: http://pastebin.com/m27dc1177
[12-Jun-2009 14:59:48] <cgibbons> yeah it's not collecting anything for the device. what happens if you go to the Event tab for that device, what's in the list?
[12-Jun-2009 15:00:50] <capricorn-80> @cgibbons: http://pastebin.com/m35e2af6c
[12-Jun-2009 15:03:34] <cgibbons> that's the modeling screen, what about Events? do you have any Events listed on that device?
[12-Jun-2009 15:05:45] <capricorn-80> SNMP agent down
[12-Jun-2009 15:05:47] <capricorn-80> but its working
[12-Jun-2009 15:05:55] <capricorn-80> i checked that using snmpwalk
[12-Jun-2009 15:06:11] <capricorn-80> i can see the hard drives
[12-Jun-2009 15:06:17] <capricorn-80> and other informations
[12-Jun-2009 15:07:03] <capricorn-80> snmp trap snmp_authenticationFailure
[12-Jun-2009 15:08:42] <capricorn-80> ahan
[12-Jun-2009 15:08:48] <capricorn-80> i can see password on snmp trap
[12-Jun-2009 15:08:53] <capricorn-80> ok will do that on monday
[12-Jun-2009 15:08:58] <capricorn-80> thanks for you help
[12-Jun-2009 15:08:59] <capricorn-80> take car e
[12-Jun-2009 15:09:00] <capricorn-80> bye
[12-Jun-2009 15:18:53] <ckrough> Anyone know of a faster way to parse objects in zope? I need to produce a report for a large number of devices (10k) and it takes around 5 min to complete
[12-Jun-2009 15:19:12] <ckrough> There isnt any kind of index in zope that I can take advantage of is there?
[12-Jun-2009 15:27:29] <ckrough> <crickets>
[12-Jun-2009 15:41:33] <ckrough> enough zenoss for the week. have a good weekend. Im out
[12-Jun-2009 16:01:07] <cmdln> Hello
[12-Jun-2009 16:02:00] <cmdln> Is there any way to manually model a topology?
[12-Jun-2009 16:02:40] <cmdln> For example if I am monitoring a bunch of virtual machines on the same host I dont want to be notified of each individual guest if the host goes down. I only need to be notified of the host.
[12-Jun-2009 16:04:27] cmdln is now known as cmdln_
[12-Jun-2009 16:04:43] cmdln_ is now known as cmdln
[12-Jun-2009 16:12:39] <FooDog> cmdln, i was wondering the same thing... ex: if the core switches are down, don't notify about edge swtiches...
[12-Jun-2009 17:44:20] <blocky> hey guys
[12-Jun-2009 17:45:01] <blocky> if i created a data source with a 300s cycle time and also a graph based on it, how hard is it to switch it to a 60s cycle time and have the graphs scale and display properly
[12-Jun-2009 17:45:09] <blocky> im not too worried about preserving the old rrd data
[12-Jun-2009 17:53:05] tehhobbit is now known as failboi
[12-Jun-2009 18:11:28] <lukecyca> cmdln, FooDog: This is a way of defining dependencies on events... http://blog.zenoss.com/2009/05/27/tip-of-the-month-event-suppression/
[12-Jun-2009 22:04:48] mephux_ is now known as mephux
[12-Jun-2009 23:22:54] <JulianMsanchez> Hi
[12-Jun-2009 23:23:33] <JulianMsanchez> Have a little problem whit my zenoss some one could help me?
[12-Jun-2009 23:27:01] <JulianMsanchez> I can't start zenjobs after upgrade...
[12-Jun-2009 23:27:13] <JulianMsanchez> it gave me AttributeError: JobManager
[13-Jun-2009 00:00:42] [disconnected at Sat Jun 13 00:00:42 2009]
[13-Jun-2009 00:00:42] [connected at Sat Jun 13 00:00:42 2009]
[13-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[13-Jun-2009 02:39:34] <JulianMsanchez> hi
[13-Jun-2009 04:33:17] <spyroux> Hi all
[13-Jun-2009 04:40:57] <JulianMsanchez> Hi spy...
[13-Jun-2009 04:41:14] <JulianMsanchez> not much movment here..
[13-Jun-2009 05:44:57] failboi is now known as loke
[13-Jun-2009 05:45:13] loke is now known as tehhobbit
[13-Jun-2009 07:04:52] <caseyh20> Hello everyone
[13-Jun-2009 14:00:58] <JulianMsanchez> hi
[13-Jun-2009 14:02:01] <rhetts> hi
[13-Jun-2009 14:05:32] <JulianMsanchez> how are you...
[13-Jun-2009 14:05:42] <JulianMsanchez> I need a zenoss guru...
[13-Jun-2009 14:05:45] <JulianMsanchez> are you?
[13-Jun-2009 14:39:56] <rhetts> im a user but i wouldnt call myself a guru
[14-Jun-2009 00:00:42] [disconnected at Sun Jun 14 00:00:42 2009]
[14-Jun-2009 00:00:42] [connected at Sun Jun 14 00:00:42 2009]
[14-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[14-Jun-2009 03:33:28] <caseyh20> so I finally have a tutorial up on how to install Zenoss 2.4.1 on FreeBSD 7.x http://www.caseyhillman.com/?p=88
[14-Jun-2009 11:23:35] <cmdln> Can I manually modify the zenping detected infrastructure? I would like to suppress alerts from hosts that depend on other hosts. For example virtual machines that are on vmhost. If vmhost goes down I do not need 5 alerts telling me about each guest that is no longer available, a single alert from vmhost is sufficient.
[14-Jun-2009 13:12:58] <spyroux> newbie question: how does zendisc choose which templates to bind to a device ? does he check every available template and add every templates that returns "something" ?
[14-Jun-2009 15:05:29] <cmdln> I cannot figure out how to suppress alerts for guest virtual machines when the host is the source of the issue
[14-Jun-2009 15:06:23] <cmdln> With alerting rules I do not seem to be able to do something like do not notify if alarm on other device
[14-Jun-2009 21:46:08] <sergeymasushko> I have in issue with the process monitoring... i' need to monitor exim process on a device. I've defined template with the ^.*\/*exim regex. zenoss detected this process with different keys e.g.
[14-Jun-2009 21:46:08] <sergeymasushko> exim-4.68-0 -bd -q30m
[14-Jun-2009 21:46:08] <sergeymasushko> exim-4.69-3
[14-Jun-2009 21:46:08] <sergeymasushko> exim-4.69-3 -q
[14-Jun-2009 21:46:08] <sergeymasushko> but actually I need to monitor only 'exim-4.68-0 -bd -q30m'. how can I prevent alerting/detecting of these 2 other processes?
[14-Jun-2009 22:22:23] <cmdln> sergeymasushko: you could adjust your regex to be more specific
[14-Jun-2009 22:28:24] <sergeymasushko> how?
[14-Jun-2009 22:29:20] <sergeymasushko> as far as I understand it collects data from different OIDs so I do not know how to adjust the regex...
[14-Jun-2009 22:30:10] <cmdln> ^.*\/*exim-4\.68-0\ -bd\ -q3m might do it
[14-Jun-2009 22:30:29] <cmdln> you said you defined the regex
[14-Jun-2009 22:32:59] <sergeymasushko> I take regex from the example.
[14-Jun-2009 22:32:59] <sergeymasushko> [zenoss@zenoss ~]$ snmpwalk -v2c -cpublic mydevice | grep exim
[14-Jun-2009 22:32:59] <sergeymasushko> HOST-RESOURCES-MIB::hrSWRunName.719 = STRING: "exim-4.69-3"
[14-Jun-2009 22:32:59] <sergeymasushko> [zenoss@zenoss ~]$ snmpwalk -v2c -cpublic mydevice | grep 'q30m'
[14-Jun-2009 22:32:59] <sergeymasushko> HOST-RESOURCES-MIB::hrSWRunParameters.719 = STRING: "-bd -q30m"
[14-Jun-2009 22:33:00] <sergeymasushko> So I guess regex that you suggested will not help me
[14-Jun-2009 22:33:55] <cmdln> just do a walk on HOST-RESOURCES-MIB::hrSWRunName.719
[14-Jun-2009 22:34:08] <cmdln> gtg
[14-Jun-2009 22:34:10] <cmdln> srry
[15-Jun-2009 00:00:42] [disconnected at Mon Jun 15 00:00:42 2009]
[15-Jun-2009 00:00:42] [connected at Mon Jun 15 00:00:42 2009]
[15-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[15-Jun-2009 09:22:30] <adr3nalin3> Hello everyone, I just installed 2.4.1 and I didn't get the initial setup screen that I usually get when installing zenoss, is there a way to invoke the initial setup wizard?
[15-Jun-2009 09:52:54] <jb> does anyone know if multiple SMTP addresses are allowed in the "Address" field of an alerting rule?
[15-Jun-2009 09:54:48] <jb> ah they do!
[15-Jun-2009 10:03:57] crow is now known as ckrouh
[15-Jun-2009 10:04:05] ckrouh is now known as ckrough
[15-Jun-2009 11:00:35] <chudler> mray?
[15-Jun-2009 11:00:40] <mrayzenoss> yeah
[15-Jun-2009 11:01:01] <chudler> I was curious, do you have anything about the criteria used to judge zenpacks?
[15-Jun-2009 11:01:47] <mrayzenoss> not really. Last year Chet and Erik and I looked at the list and picked. Perhaps this year I should try to find someone from the community without an entry
[15-Jun-2009 11:02:13] <mrayzenoss> but if we keep up this pace, we may add more prizes
[15-Jun-2009 11:02:21] <mrayzenoss> I already have more entries than last year's contest
[15-Jun-2009 11:02:26] <mrayzenoss> and we still have 2 months left
[15-Jun-2009 11:02:52] <mrayzenoss> btw, your Jabber one is in SVN, I just have to make the webpages
[15-Jun-2009 11:03:05] <chudler> thats really cool. I was just interested in what sort of methods are used.
[15-Jun-2009 11:03:41] <mrayzenoss> should we have voting?
[15-Jun-2009 11:03:47] <mrayzenoss> I'm open to suggestions
[15-Jun-2009 11:05:01] <chudler> no, but perhaps a few public descriptions of what is important to you guys when judging. Overall utlity of the zenpack? Breadth of potential installs? Coding concepts/style? Unique factor?
[15-Jun-2009 11:06:09] <chudler> whatever you are doing is working :-) and like I said, I was just curious. There is probably value in making the decision from the gut.
[15-Jun-2009 11:06:21] <mrayzenoss> yeah, that's essentially how we judged last year. Utility and popularity are important, but recognizing ZenPacks that seem ambitious in their scope
[15-Jun-2009 11:06:34] <mrayzenoss> last year only had 2 winners, this year is 1+5
[15-Jun-2009 11:06:41] <mrayzenoss> 1 grand prize and 5 runner ups
[15-Jun-2009 11:07:32] <mrayzenoss> for instance, at SELF someone described an ESX SSH solution they're working on. It models an ESX box, then adds the filesystems and guests to the host. All via SSH
[15-Jun-2009 11:08:00] <mrayzenoss> that sounds very useful and ambitious, as well as hitting SSH which gets him double credit if he gets it in as an entry
[15-Jun-2009 11:08:26] <mrayzenoss> right now there are 25 unique entries, with 4 SSH-based ones
[15-Jun-2009 11:08:33] <mrayzenoss> err, 3
[15-Jun-2009 11:09:03] <mrayzenoss> so you can go for lots of entries, which helps your odds since some of the prizes are pulled at random
[15-Jun-2009 11:09:48] <mrayzenoss> hmmm... looking at the webpage, apparently we only said 4 runner ups
[15-Jun-2009 11:10:10] <mrayzenoss> so the grand prize and 2 runner ups are picked by the judges, and 2 are randomly selected
[15-Jun-2009 11:11:50] * chudler needs to stuff the ballot with a slew of ssh modelers ;-)
[15-Jun-2009 11:12:04] <mrayzenoss> you better hurry, rocket has already sent in Gentoo and SuSE
[15-Jun-2009 11:12:23] <mrayzenoss> may I suggest the BSDs? There are a lot of them
[15-Jun-2009 11:12:51] <chudler> eheh thats a nice idea, unfortunately ssh monitoring has little value for me. I like to scratch my own itches :-)
[15-Jun-2009 11:13:04] <mrayzenoss> yeah, your Jabber bot seems pretty cool
[15-Jun-2009 11:13:25] <chudler> it is something that I use every day. Major itch that my boss wanted me to scratch, actually heh
[15-Jun-2009 11:13:30] <mrayzenoss> I'm trying to do 12 ZenPacks in parallel right now, so there should be a flood of announcements soon
[15-Jun-2009 11:13:58] <mrayzenoss> first I smoke test them, then get them into SVN, then create pages on the www and trac sites, then the announcements
[15-Jun-2009 11:14:04] <mrayzenoss> right now most are in SVN
[15-Jun-2009 11:14:11] <mrayzenoss> so next comes a lot of HTML
[15-Jun-2009 11:57:48] <rocket> Hey Matt
[15-Jun-2009 12:14:42] <FooDog> is there an official way to post bugs/suggestions against community zenpacks?
[15-Jun-2009 12:19:58] <rocket> FooDog: use trac it can reference community zenpacks I believe
[15-Jun-2009 12:20:10] <ckrough> Does "update collector" update collector files based on lastmod date, or does it push everything?
[15-Jun-2009 12:21:02] <ckrough> nm, looks like its an rsync.
[15-Jun-2009 12:21:27] <chudler> yeah rsync --times -r, or something like that.
[15-Jun-2009 12:21:52] <chudler> er rsync -tugr
[15-Jun-2009 12:28:56] <capricorn-80> hi
[15-Jun-2009 12:29:02] <capricorn-80> rocket you there ?
[15-Jun-2009 12:29:08] <rocket> yes
[15-Jun-2009 12:29:50] <capricorn-80> well i figure out some problem
[15-Jun-2009 12:29:55] <rocket> oh?
[15-Jun-2009 12:30:27] <capricorn-80> that management interface is giving RRD file missing error
[15-Jun-2009 12:32:38] <capricorn-80> but i can see CPU and memory graphs in the performance
[15-Jun-2009 12:32:58] <capricorn-80> but when i click on that Null0 it give errors mentioned on the graph
[15-Jun-2009 12:33:21] <capricorn-80> i.ie Missing RRD File
[15-Jun-2009 12:33:28] <capricorn-80> but every thing works fine on its Vlan
[15-Jun-2009 12:33:49] <capricorn-80> oh no
[15-Jun-2009 12:33:57] <capricorn-80> same error
[15-Jun-2009 12:33:59] <capricorn-80> ah
[15-Jun-2009 12:34:06] <capricorn-80> dont know what happened
[15-Jun-2009 12:36:22] <capricorn-80> rocket: http://pastebin.com/d26d25fa9
[15-Jun-2009 12:37:50] <rocket> if you can do a zenoss restart;zenperfsnmp run -v10 -d 10.254.95.254
[15-Jun-2009 12:37:57] <rocket> and see if you get anymore information
[15-Jun-2009 12:38:37] <capricorn-80> i just pasted you the out of zenprefsnmp
[15-Jun-2009 12:38:42] <capricorn-80> let me try it again
[15-Jun-2009 12:46:37] <capricorn-80> rocket: http://pastebin.com/m1b3d24e7
[15-Jun-2009 12:46:42] <capricorn-80> looks similar
[15-Jun-2009 12:48:03] <capricorn-80> some thing wrong .. graphs are too odd
[15-Jun-2009 12:52:24] <capricorn-80> need some help
[15-Jun-2009 12:52:52] <rocket> did you restart zenoss?
[15-Jun-2009 12:53:06] <rocket> have you checked the logs in the log directory?
[15-Jun-2009 12:53:35] <rocket> logs usually are in a place like /usr/local/zenoss/zenoss/logs
[15-Jun-2009 12:53:42] <rocket> once there do a
[15-Jun-2009 12:53:59] <rocket> grep Trace *;grep -i DEBUG *;grep -i WARN *
[15-Jun-2009 12:54:22] <rocket> and to find out which logs contain issues and look at the logs to figure out what is happening
[15-Jun-2009 13:01:47] <capricorn-80> rocket: http://pastebin.com/m16012e4d
[15-Jun-2009 13:03:42] <capricorn-80> really strange
[15-Jun-2009 13:05:50] <capricorn-80> rocket: http://pastebin.com/m675b93b9
[15-Jun-2009 13:05:55] <capricorn-80> this will be interesting
[15-Jun-2009 13:06:29] <rocket> thats a known response when the rrd file is missing
[15-Jun-2009 13:08:11] <capricorn-80> i was having problem with windows rrd file but for router it was working fine
[15-Jun-2009 13:08:18] <capricorn-80> and today all this happened
[15-Jun-2009 13:08:27] <capricorn-80> what should i do to solve it ?
[15-Jun-2009 13:13:05] <capricorn-80> its getting harder with zenoss
[15-Jun-2009 13:13:57] <coofamani> what is
[15-Jun-2009 13:15:58] <capricorn-80> coofamani: suddenly i got RRD missing file error
[15-Jun-2009 13:16:04] <capricorn-80> graphs are too odd
[15-Jun-2009 13:16:08] <capricorn-80> dont know what happened
[15-Jun-2009 13:16:20] <capricorn-80> help me how i can recover from it
[15-Jun-2009 13:16:36] <coofamani> Zenoss is a fairly complex application
[15-Jun-2009 13:17:13] <coofamani> Have you read through the Administrator manual and possible the Development manual. It helps to be familiar with it's foundation when you are troubleshooting issues.
[15-Jun-2009 13:18:05] <capricorn-80> coofamani: it was working fine
[15-Jun-2009 13:18:32] <capricorn-80> dont have time to switch to other nms
[15-Jun-2009 13:18:40] <capricorn-80> spent 2 weeks
[15-Jun-2009 13:20:37] <capricorn-80> kindly help me troubeshooting
[15-Jun-2009 13:20:48] <capricorn-80> in **
[15-Jun-2009 13:21:06] <rocket> capricorn-80: this is a volunteer channel .. as in some of this stuff you really need to start figuring out yourself
[15-Jun-2009 13:21:39] <rocket> if you need more support you probably should be thinking about getting a support contract with zenoss
[15-Jun-2009 13:22:38] <coofamani> You may want to consider taking their training course as well. That will save you a lot of time compared to learning it on your own.
[15-Jun-2009 13:23:30] <Jontler> checking out the weekly webex broadcast is also a good idea
[15-Jun-2009 13:23:56] <rocket> and the developers are usually here on thursdays to answer questions
[15-Jun-2009 13:24:25] <rocket> capricorn-80: but for those of us who dont have support contracts ... we have all had to figure out issues like you are having on our own
[15-Jun-2009 13:24:57] <rocket> capricorn-80: its for that reason that we understand what can go wrong and why ..
[15-Jun-2009 13:25:02] <capricorn-80> well thats what i am asking for some help right now
[15-Jun-2009 13:25:36] <capricorn-80> rocket: because it was working fine\
[15-Jun-2009 13:25:46] <capricorn-80> and suddenly it went off
[15-Jun-2009 13:26:02] <rocket> well there has to be something in the logs or event manager that would show the reason
[15-Jun-2009 13:26:17] <rocket> or you have changed something to make it no longer work ..
[15-Jun-2009 13:26:31] <rocket> have you checked disk space etc?
[15-Jun-2009 13:27:24] <capricorn-80> i did some settings in /etc/hosts
[15-Jun-2009 13:27:28] <capricorn-80> on friday
[15-Jun-2009 13:27:58] <rocket> do you have a backup of the original /etc/hosts?
[15-Jun-2009 13:28:13] <rocket> perhaps you need to remodel the device if its ip has changed etc
[15-Jun-2009 13:28:27] <capricorn-80> 127.0.0.1 sendto error Host localhost and my.machine.com are both using ip 127.0.0.1
[15-Jun-2009 13:28:36] <capricorn-80> no ip is the same
[15-Jun-2009 13:30:48] <coofamani> anyone else have no scroll bar in the gui?
[15-Jun-2009 13:30:56] <coofamani> I see some forum posts about it, but no solutions
[15-Jun-2009 13:31:01] <coofamani> firefox on fedora.
[15-Jun-2009 13:31:11] <coofamani> problem has persisted through 2 versions of FF now
[15-Jun-2009 13:31:27] <Jontler> ive got my scroll bar just fine
[15-Jun-2009 13:31:32] <Jontler> same setup
[15-Jun-2009 13:31:37] <coofamani> want to trade?
[15-Jun-2009 13:31:39] <Jontler> ha
[15-Jun-2009 13:31:55] <Jontler> what version of flash are you running?
[15-Jun-2009 13:31:59] <coofamani> its better, no matter what happens, only 30 devices are affected!
[15-Jun-2009 13:32:26] <capricorn-80> rocket: you think that can be a problem which i pasted above ?
[15-Jun-2009 13:33:16] <coofamani> Jontler: flash-plugin-10.0.22.87-release.i386
[15-Jun-2009 13:33:23] <coofamani> Jontler: you?
[15-Jun-2009 13:33:54] <coofamani> that device lists arent in flash are they?
[15-Jun-2009 13:34:34] <mrayzenoss> coofamani: the only use of flash is the network map
[15-Jun-2009 13:35:03] <Jontler> same coofamani
[15-Jun-2009 13:35:13] <Jontler> and i guess you dont need it
[15-Jun-2009 13:35:29] <rocket> capricorn-80: probably
[15-Jun-2009 13:38:17] <coofamani> heh, the onlsy results for a "Scroll bar" search on search.zenoss is other people asking about the scroll bar in IRC
[15-Jun-2009 13:38:51] <Jontler> did you try different browsers
[15-Jun-2009 13:39:01] <coofamani> yeah
[15-Jun-2009 13:39:05] <coofamani> its just on this box
[15-Jun-2009 13:39:10] <Jontler> ok
[15-Jun-2009 13:39:16] <coofamani> but its only zenoss
[15-Jun-2009 13:39:24] <coofamani> no other apps seem to have the same issue
[15-Jun-2009 13:39:25] <Jontler> i was gonna suggest you fire up epiphany, since it's based on firefox
[15-Jun-2009 13:39:42] <coofamani> regardless, Im replacing the box tomorrow. we'll see if the issue follows my profile
[15-Jun-2009 13:46:34] <coofamani> Jontler: interesting, no scroll bar in epiphany either
[15-Jun-2009 13:49:32] * Jontler shrugs
[15-Jun-2009 13:49:37] <Jontler> im interested to see if it follows your profile
[15-Jun-2009 13:50:32] <Jontler> did you try creating a new firefox profile, just for kicks?
[15-Jun-2009 13:50:50] <Jontler> that would tell you pretty quickly whether or not it's a profile issue
[15-Jun-2009 13:51:04] <coofamani> yup, not luck
[15-Jun-2009 14:12:55] <capricorn-80> rocket: need some help regarding some simple things
[15-Jun-2009 14:13:08] <capricorn-80> i just changed my server ip address and hostname
[15-Jun-2009 14:13:18] <capricorn-80> and some settings in resolve.conf
[15-Jun-2009 14:13:36] <capricorn-80> now apache is runing with but when i add 8080 with ip it just start processing
[15-Jun-2009 14:13:48] <capricorn-80> i have checked that 8080 is running
[15-Jun-2009 14:14:18] <capricorn-80> tcp 478 0 10.254.91.114:8080 10.254.91.72:4119 ESTABLISHED
[15-Jun-2009 14:14:18] <capricorn-80> tcp 479 0 10.254.91.114:8080 10.254.91.72:4114 CLOSE_WAIT
[15-Jun-2009 14:14:18] <capricorn-80> its done processing from last 10 misn
[15-Jun-2009 14:14:20] <capricorn-80> mins
[15-Jun-2009 14:14:55] <rocket> capricorn-80: which server? zenoss server?
[15-Jun-2009 14:15:04] <rocket> I am not sure what you are trying to accomplish
[15-Jun-2009 14:15:07] <capricorn-80> yes
[15-Jun-2009 14:15:29] <capricorn-80> i have changed ip address , its name and some dns related settings in resolve.conf
[15-Jun-2009 14:16:09] <capricorn-80> rocket: i was screwed and i checked that the ip address was from dhcp although i am having that ip from last 2 weeks so i change it to static
[15-Jun-2009 14:16:21] <capricorn-80> and also change my hostname
[15-Jun-2009 14:17:01] <rocket> for the server?
[15-Jun-2009 14:17:14] <capricorn-80> yes
[15-Jun-2009 14:17:27] <capricorn-80> all zenoss daemons are working fine
[15-Jun-2009 14:17:39] <rocket> ok so what is the problem then?
[15-Jun-2009 14:17:52] <jb> zenoss does not use apache..
[15-Jun-2009 14:19:05] <capricorn-80> http://ip:8080 and its processig
[15-Jun-2009 14:19:15] <capricorn-80> nothing opening the console
[15-Jun-2009 14:19:41] <rocket> check the zenoss logs and see if there are issues
[15-Jun-2009 14:20:12] <capricorn-80> rocket: i am thinking to remove zenoss and install it again
[15-Jun-2009 14:21:41] <capricorn-80> ask i messed up with things
[15-Jun-2009 14:22:08] <capricorn-80> its still a testing server the time it will take for trouble shooting .. its better that i will start new one
[15-Jun-2009 14:22:18] <rocket> start new
[15-Jun-2009 14:23:02] <capricorn-80> ok
[15-Jun-2009 14:39:28] <capricorn-80> rocket: see u tomorrow
[15-Jun-2009 14:39:29] <capricorn-80> bye
[15-Jun-2009 15:17:18] <blocky> if i delete RRD files will they be recreated automatically
[15-Jun-2009 15:32:20] <coofamani> yes
[15-Jun-2009 15:32:26] <coofamani> takes around 3 polling cycles
[15-Jun-2009 15:32:32] <coofamani> 15 minutes on a default install
[15-Jun-2009 16:11:55] <blocky> what if i change the polling cycle for the data points im gathering
[15-Jun-2009 16:11:56] <blocky> then delete
[15-Jun-2009 16:12:08] <blocky> will the newly created rrds have the correct interval
[15-Jun-2009 17:42:45] <caseyh20> how is everyone doing today?
[15-Jun-2009 17:43:16] <mrayzenoss> good, just about to check out
[15-Jun-2009 17:43:22] <mrayzenoss> but thanks for updating your blog entry
[15-Jun-2009 17:43:28] <mrayzenoss> I sent a lot of folks that way
[15-Jun-2009 17:44:31] <caseyh20> yeah, I have seen all the links, not much feedback so far though, I hope that means it is working for them lol
[15-Jun-2009 17:45:32] <mrayzenoss> I actually talked to some Gentoo users at SouthEast LinuxFest and sent them to it as well, figured they'd have similar issues
[15-Jun-2009 17:46:03] <caseyh20> it would be interesting to see if it worked for them
[15-Jun-2009 17:47:32] <caseyh20> I ran into an issue on sunday when I was doing a final test, apparently autoconf259 is no longer in the FreeBSD ports or packages so I had to go with autoconf263
[15-Jun-2009 17:47:53] <caseyh20> that is something that has changed within the past few weeks
[15-Jun-2009 17:50:12] <mrayzenoss> thanks a lot for posting that, I gotta run, but I'm looking forward to your iPhone stuff, whatever it is
[15-Jun-2009 17:53:13] <caseyh20> no problem, it is something that I enjoyed, for the most part lol, yeah the iPhone is something im looking forward to but im running a little behind. im gonna be in anaheim for a week then when I get back im gonna be working on the iPhone app as well as the FreeBSD port and a few zenpacks too
[16-Jun-2009 00:00:42] [disconnected at Tue Jun 16 00:00:42 2009]
[16-Jun-2009 00:00:42] [connected at Tue Jun 16 00:00:42 2009]
[16-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[16-Jun-2009 08:05:37] int_ is now known as int
[16-Jun-2009 09:35:49] <chudler> anyone forsee problems with puting /Perf data on NFS?
[16-Jun-2009 09:36:08] <venturaville> chudler: loads of problems
[16-Jun-2009 09:37:07] <chudler> please go on.
[16-Jun-2009 09:37:31] <venturaville> usually in RRD your are updating part of the file, but on NFS you end up forcing a cache flush on the entire file when you update it
[16-Jun-2009 09:37:57] <venturaville> which means you are downloading the entire RRD tree every x minutes
[16-Jun-2009 09:39:37] <venturaville> if your file tree is sufficiently large (read MBs or even GBs) then you are invalidating a lot of system disk cache memory as well as potentially saturating your network connection.....
[16-Jun-2009 09:39:49] <venturaville> iSCSI is an alternative.....
[16-Jun-2009 09:40:12] <chudler> I am used to those things, yeah, no big deal. Idea is to share the disk, so iscsi probably not a good fit.
[16-Jun-2009 09:40:37] <venturaville> don't know what you are trying to accomplish......
[16-Jun-2009 09:40:58] <venturaville> I might can help better if I knew that
[16-Jun-2009 09:40:58] <jb> gfs? "_
[16-Jun-2009 09:41:00] <jb>
[16-Jun-2009 09:41:10] <venturaville> good thought
[16-Jun-2009 09:41:15] <venturaville> but complicated
[16-Jun-2009 09:41:19] <jb> true
[16-Jun-2009 09:41:44] <jb> both devices need access at the same time to the data?
[16-Jun-2009 09:42:10] <chudler> its not my bag, but I will explain. Some people here move systems between collectors, and they want the rrd data to move with it. simple as that, I think.
[16-Jun-2009 09:42:45] <jb> hmm
[16-Jun-2009 09:43:28] <venturaville> ah I see
[16-Jun-2009 09:43:58] <jb> well, gfs would work..
[16-Jun-2009 09:43:58] <chudler> having it on NFS might be neato for some tangental reasons heh
[16-Jun-2009 09:44:06] <jb> but, I would not use NFS
[16-Jun-2009 09:44:13] <venturaville> seconded
[16-Jun-2009 09:44:32] <venturaville> an alternative way if you don't care so much about disk space is to use an rsync job to synchronize them around
[16-Jun-2009 09:44:38] <jb> possibly a very large performance bottleneck there..
[16-Jun-2009 09:44:45] <jb> yeah
[16-Jun-2009 09:45:19] <chudler> I think the ones pushing for a solution are considering rsync, or just leaning on Zenoss, inc for a fix.
[16-Jun-2009 09:45:54] <venturaville> I think enterprise already has stuff for stuff along that lines
[16-Jun-2009 09:47:53] <chudler> it is enterprise. perhaps the engineer is on crack
[16-Jun-2009 09:49:15] <chudler> I would love to do gfs or drbd or md+nbd 'cause those things are sexy and fun, but we already have a few stable, large (expensive) NFS clusters.
[16-Jun-2009 09:51:44] <jb> chudler: are you at uchicago?
[16-Jun-2009 09:53:14] <chudler> yes
[16-Jun-2009 10:03:25] <jb> cool.. im in chi as well.
[16-Jun-2009 11:17:31] <blocky> can anyone point me to a tutorial for creating custom reports that'
[16-Jun-2009 11:17:49] <blocky> s a bit more up to date than the one in the documentation
[16-Jun-2009 11:20:24] <mrayzenoss> so the Admin Guide has a bit: http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch12s08.html and the Developer Guide has a chapter on custom reports as well
[16-Jun-2009 11:20:31] <mrayzenoss> what are you trying to do?
[16-Jun-2009 11:21:09] <blocky> well ive got a bunch of templates now, all ssh commands
[16-Jun-2009 11:21:14] <blocky> and im going to add a few more
[16-Jun-2009 11:21:25] <blocky> but I'd like to be able to email myself and whoever else a report every morning
[16-Jun-2009 11:21:55] <blocky> ie. for all 8 boxes, here was the average 15 minute load, here was the peak mem usage, here is how many clients
[16-Jun-2009 11:22:17] <blocky> and im going to grab a bit of info from my squid as well, here is the ip of every TCP_DENIED message
[16-Jun-2009 11:22:52] <blocky> etc etc and i figured if i made it into a nice report then i could just have reportmail send it off via cron each morning before everyone gets to work
[16-Jun-2009 11:23:06] <mrayzenoss> this might prove useful too: http://www.zenoss.com/Members/cluther/creating-methods-for-custom-reports
[16-Jun-2009 11:23:35] <blocky> cool that looks great actually
[16-Jun-2009 11:24:21] <mrayzenoss> if you don't want to use the UI, you can hack up the .rpt files as well.
[16-Jun-2009 11:24:42] <blocky> does reportmail generate html email?
[16-Jun-2009 11:25:06] <blocky> okay this is where i was having problems before also
[16-Jun-2009 11:25:27] <blocky> nm
[16-Jun-2009 11:25:56] <blocky> zenoss is quite well done, btw
[16-Jun-2009 11:26:02] <blocky> once i understand it im sure it will be fantastic
[16-Jun-2009 11:26:10] <mrayzenoss> yeah, it keeps getting better
[16-Jun-2009 11:26:15] <mrayzenoss> thanks
[16-Jun-2009 11:29:04] <blocky> lets say i have two classes of server, one that is a subset of the other
[16-Jun-2009 11:29:20] <blocky> so some properties are not available and not meaningful for some of them
[16-Jun-2009 11:29:32] <blocky> is there a way i can have a device report for all servers anyway?
[16-Jun-2009 11:39:29] <mrayzenoss> if you're doing a custom report, you could probably check for a value and return "N/A" for those cases
[16-Jun-2009 11:39:46] <mrayzenoss> opened this ticket BTW: http://dev.zenoss.org/trac/ticket/5201
[16-Jun-2009 11:39:56] <mrayzenoss> we didn't have docs for reportmail
[16-Jun-2009 11:40:16] <rocket> Morning Matt
[16-Jun-2009 11:40:22] <mrayzenoss> good morning
[16-Jun-2009 11:40:39] <rocket> actually almost afternoon now ..
[16-Jun-2009 11:40:57] <mrayzenoss> yeah, doing the Getting Started Webinar in 20 minutes
[16-Jun-2009 11:41:02] <mrayzenoss> right in the middle of lunch
[16-Jun-2009 11:41:15] <rocket> things going better with the zenpacks I submitted now that most of the kinks are out of them?
[16-Jun-2009 11:41:35] <phasma> Hi! Can anybody help with graphics ?
[16-Jun-2009 11:41:37] <mrayzenoss> yeah, I've got graphs for my Interfaces
[16-Jun-2009 11:41:56] <rocket> That helps for a linux zenpack
[16-Jun-2009 11:42:09] <mrayzenoss> phasma: what's the issue?
[16-Jun-2009 11:42:33] <phasma> mrayzenoss: Where I can change encoding or font ?
[16-Jun-2009 11:42:53] <mrayzenoss> for the RRD stuff?
[16-Jun-2009 11:43:01] <mrayzenoss> or for everything?
[16-Jun-2009 11:44:08] <phasma> mrayzenoss: http://omploader.org/vMXVjMg/screen.png
[16-Jun-2009 11:44:25] <mrayzenoss> phasma: known issue. What version are you running?
[16-Jun-2009 11:44:28] <mrayzenoss> 2.4.0?
[16-Jun-2009 11:44:54] <phasma> 2.4.1
[16-Jun-2009 11:45:05] <mrayzenoss> which installer and which distro?
[16-Jun-2009 11:45:15] <phasma> mrayzenoss: FreeBSD from sources
[16-Jun-2009 11:45:32] <phasma> mrayzenoss: locale ru_RU.UTF-8
[16-Jun-2009 11:45:37] <mrayzenoss> http://dev.zenoss.org/trac/ticket/5006
[16-Jun-2009 11:46:08] <phasma> mrayzenoss: thanks
[16-Jun-2009 11:55:04] <blocky> so is there a list somewhere of other calls similar to getIpServicesListAsString
[16-Jun-2009 11:55:18] <blocky> can i retrieve my performance data source numbers?
[16-Jun-2009 11:58:20] <mrayzenoss> yeah, the dev guide has a bit, I know I've seen people do that
[16-Jun-2009 11:58:35] <mrayzenoss> there's a getRRDTemplates
[16-Jun-2009 11:58:42] <mrayzenoss> and a getRRDDataSources off of that
[16-Jun-2009 11:59:26] <mrayzenoss> I gotta go do a webinar, check the forums, I know I've seen it
[16-Jun-2009 15:04:17] <rocket> e1000g0: flags=201004843<UP,BROADCAST,RUNNING,MULTICAST,DHCP,IPv4,CoS> mtu 1500 index 2 inet 192.168.31.133 netmask ffffff00 broadcast 192.168.31.255 ether 0:c:29:22:94:aa
[16-Jun-2009 15:25:44] <zenChild> I'm trying to create a modeler that uses ssh, but I need the input from one command to send to another. Is there a away to send another command from the modeler plugin after one has run already?
[16-Jun-2009 15:26:29] <Jontler> off the top of my head, id batch it all into one python script and just call that
[16-Jun-2009 15:27:27] <zenChild> I can't it's an appliance box that uses rbash. I create or run any files on the box that aren't part of the default set. I can't even run sed or this would be a non-issue.
[16-Jun-2009 15:28:32] <Jontler> then im out of brilliant ideas
[16-Jun-2009 15:28:50] <zenChild> thanks anyway
[16-Jun-2009 18:41:44] <blocky> is there a way to reduce the size of graphs in a report/email but not on the original template?
[16-Jun-2009 19:23:26] <kgoedtel> you pretty much can't do anything useful with the reportmail
[16-Jun-2009 19:23:46] <kgoedtel> whatever your defaults are is what you get
[16-Jun-2009 19:32:03] <blocky> lol
[16-Jun-2009 21:03:59] <sergeymasushko> I have an issue with the process monitoring... i' need to monitor exim process on a device. I've defined template with the ^.*\/*exim regex. zenoss detected this process with different keys e.g.
[16-Jun-2009 21:03:59] <sergeymasushko> exim-4.68-0 -bd -q30m
[16-Jun-2009 21:03:59] <sergeymasushko> exim-4.69-3
[16-Jun-2009 21:03:59] <sergeymasushko> exim-4.69-3 -q
[16-Jun-2009 21:03:59] <sergeymasushko> but actually I need to monitor only 'exim-4.68-0 -bd -q30m'. how can I prevent alerting/detecting of these 2 other processes?
[16-Jun-2009 22:21:32] <aley__> hallow
[16-Jun-2009 22:21:43] <aley__> need help
[16-Jun-2009 22:25:20] <aley__> hellow
[16-Jun-2009 22:27:28] <sergeymasushko> hi
[16-Jun-2009 22:31:27] <sergeymasushko> aley__: ?
[16-Jun-2009 22:31:59] <aley__> i need help sergey
[16-Jun-2009 22:32:50] <aley__> when i run ( zenoss@aldebaran root]$ zendisc run
[16-Jun-2009 22:32:52] <aley__> INFO:zen.ZenDisc:Connecting to localhost:8789
[16-Jun-2009 22:32:52] <aley__> INFO:zen.ZenDisc:Connected to ZenHub
[16-Jun-2009 22:32:52] <aley__> WARNING:zen.ZenDisc:No networks configured
[16-Jun-2009 22:32:52] <aley__> INFO:zen.ZenDisc:Result: []
[16-Jun-2009 22:32:52] <aley__> INFO:zen.ZenDisc:Scan time: 0.00 seconds
[16-Jun-2009 22:32:54] <aley__> INFO:zen.ZenDisc:Daemon ZenDisc shutting down
[16-Jun-2009 22:32:56] <aley__> INFO:zen.ZenDisc:zendisc shutting down
[16-Jun-2009 22:32:58] <aley__> )
[16-Jun-2009 22:33:00] <aley__> why
[16-Jun-2009 22:36:28] <sergeymasushko> I see the same
[16-Jun-2009 22:37:42] <aley__> INFO:zen.ZenModeler:SNMP collection device Backhaul BPDE ke BAPPEDA PROV
[16-Jun-2009 22:37:43] <aley__> INFO:zen.ZenModeler:plugins: zenoss.snmp.DeviceMap, zenoss.snmp.InterfaceMap, zenoss.snmp.RouteMap
[16-Jun-2009 22:37:43] <aley__> INFO:zen.ZenModeler:No portscan plugins found for Backhaul BPDE ke BAPPEDA PROV
[16-Jun-2009 22:37:43] <aley__> INFO:zen.ZenModeler:Running 3 clients
[16-Jun-2009 22:37:43] <aley__> ERROR:zen.SshClient:Connection was refused by other side: 111: Connection refuse
[16-Jun-2009 22:38:02] <aley__> No portscan (what the meaning
[16-Jun-2009 22:38:58] <sergeymasushko> on the zProperties page of the device you should select needed portscan plugins
[16-Jun-2009 22:43:18] <aley__> how i add portscan plugins
[16-Jun-2009 22:45:29] <sergeymasushko> have you opened zProperties page?
[16-Jun-2009 22:47:37] <sergeymasushko> find 'zCollectorPlugins' and click Edit link. then click 'Add Fields' link then you can drag plugins from one box to another...
[16-Jun-2009 22:53:38] <aley__> yes
[16-Jun-2009 22:53:55] <aley__> ok i m finish to drag all plugins..
[16-Jun-2009 22:54:19] <sergeymasushko> click save and remodel the device
[16-Jun-2009 22:57:02] <aley__> ok
[16-Jun-2009 22:57:58] <aley__> wait the minit ..i reboot the my machine
[16-Jun-2009 23:09:06] <aley__> ok hai sergeymashuhko i need your help again
[16-Jun-2009 23:09:10] <aley__> [zenoss@aldebaran root]$ zendisc run
[16-Jun-2009 23:09:10] <aley__> INFO:zen.ZenDisc:Connecting to localhost:8789
[16-Jun-2009 23:09:10] <aley__> INFO:zen.ZenDisc:Connected to ZenHub
[16-Jun-2009 23:09:10] <aley__> WARNING:zen.ZenDisc:No networks configured
[16-Jun-2009 23:09:10] <aley__> INFO:zen.ZenDisc:Result: []
[16-Jun-2009 23:09:18] <aley__> INFO:zen.ZenDisc:Scan time: 0.00 seconds
[16-Jun-2009 23:09:18] <aley__> INFO:zen.ZenDisc:Daemon ZenDisc shutting down
[16-Jun-2009 23:09:18] <aley__> INFO:zen.ZenDisc:zendisc shutting down
[16-Jun-2009 23:09:18] <aley__> [zenoss@aldebaran root]$
[16-Jun-2009 23:09:37] <aley__> what the meaningn warning WARNING:zen.ZenDisc:No networks configured
[16-Jun-2009 23:14:10] <sergeymasushko> just ignore it for now
[16-Jun-2009 23:14:31] <aley__> oouh
[16-Jun-2009 23:15:16] <aley__> why the color in my device is orange ? normal is color green righ
[16-Jun-2009 23:16:00] <aley__> User timeout caused connection failure.
[16-Jun-2009 23:20:58] <sergeymasushko> open event tab and there you can read event details
[16-Jun-2009 23:35:41] <aley__> maybe zenoss 2.4 have bug ?
[17-Jun-2009 00:00:42] [disconnected at Wed Jun 17 00:00:42 2009]
[17-Jun-2009 00:00:42] [connected at Wed Jun 17 00:00:42 2009]
[17-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[17-Jun-2009 05:31:59] <phasma> Hi! I have a problem with process monitoring via net-snmp on linux host. 'snmpwalk -v2c -cprivate localhost | grep sd.fcgi' works perflectly, but Zenoss zenprocess with regex "sd.fcgi" writes to log error, what process doesn't running
[17-Jun-2009 05:33:07] <phasma> snmpwalk output: http://pastebin.com/m1bbaa835
[17-Jun-2009 06:17:23] <tripitaka> Hi, I'm getting a lot of snmp traps which are in event class /unknown, and which generate events with a severity of warning or critical. How can I tell zenoss that all events in the /Unknown class should be reclassified as severity 'info'?
[17-Jun-2009 08:21:22] <nzle> we recently had 2 dlink 1224 swtiches management/monitoring I go unresponsive. The timeframe this happened during was compatible with when we were modeling various devices on our newly installed zenoss (vmware appliance). We also monitor these devices with MRTG(snmp) and Nagios(ping). Has anyone seen this sort of thing after snmp modeling this switch with zenoss or would my time be better served looking in other directions.
[17-Jun-2009 08:21:46] <nzle> monitoring I = monitoring IP
[17-Jun-2009 08:26:17] <nzle> not sure if my message went out since my nick wasnt registered, resending...
[17-Jun-2009 08:26:43] <nzle> we recently had 2 dlink 1224 swtiches management/monitoring IP go unresponsive. The timeframe this happened during was compatible with when we were modeling various devices on our newly installed zenoss (vmware appliance). We also monitor these devices with MRTG(snmp) and Nagios(ping). Has anyone seen this sort of thing after snmp modeling this switch with zenoss or would my time be better served looking in other directions.
[17-Jun-2009 09:13:24] <phasma> it works !!!
[17-Jun-2009 09:23:32] <rocket> morning matt
[17-Jun-2009 09:26:06] <mrayzenoss> good morning
[17-Jun-2009 09:32:15] <jb> hey matt.. http://dev.zenoss.org/trac/ticket/5014
[17-Jun-2009 09:32:19] <jb> is there a patch available for this yet?
[17-Jun-2009 09:32:49] <jb> a couple are discussed in that ticket, I believe..
[17-Jun-2009 09:38:18] <tripitaka> Can I reclassify all /Unknown events as severity 'info'?
[17-Jun-2009 09:38:41] <mrayzenoss> jb: You want to zenpatch 13735
[17-Jun-2009 09:38:50] <mrayzenoss> jb: that's the one that is the 2.4.x tree
[17-Jun-2009 09:38:55] <mrayzenoss> err, branch
[17-Jun-2009 09:41:33] <mrayzenoss> 2.4.2 is getting close to release
[17-Jun-2009 09:47:29] <jb> thanks
[17-Jun-2009 10:31:47] <daMaestro> how do you add and arbitrary os process monitor?
[17-Jun-2009 10:31:54] <daMaestro> i'd like to make sure process "asdf" is running
[17-Jun-2009 10:48:08] <mrayzenoss> daMaestro: take a look at Jane Curry's "Methods of monitoring processes with Zenoss" http://www.skills-1st.co.uk/papers/jcurry.html
[17-Jun-2009 10:48:25] <mrayzenoss> she drills down into it pretty well
[17-Jun-2009 11:22:08] <kisielk> is there a way to map a particular event to an event class regardless of the component?
[17-Jun-2009 11:22:28] <kisielk> For example, I often get a message like "nss_ldap: could not search LDAP server - Server is unavailable" when and LDAP server is unreachable, but the component depends on what application was trying to use nsswitch at that time, which could be anything
[17-Jun-2009 11:28:35] <mrayzenoss> kisielk: you could use a transform to change the event to whatever you want to map it to
[17-Jun-2009 11:29:12] <kisielk> but how can I have the transform map any component?
[17-Jun-2009 11:29:29] <kisielk> I thought you could only create rules per component
[17-Jun-2009 11:29:40] <mrayzenoss> in 2.4, the event transforms are now hierarchal
[17-Jun-2009 11:30:00] <mrayzenoss> so you could place your transform in /Events to change the event before it gets to other events
[17-Jun-2009 11:30:44] <mrayzenoss> including changing the component
[17-Jun-2009 11:30:47] <kisielk> oh really
[17-Jun-2009 11:30:49] <kisielk> that's cool
[17-Jun-2009 11:30:56] <mrayzenoss> dangerous, but cool
[17-Jun-2009 11:31:27] <kisielk> I think our zenoss install is finally getting to be in a nice state
[17-Jun-2009 11:31:41] <kisielk> there's still a few things we haven't been able to replace
[17-Jun-2009 11:31:44] <kisielk> from our previous monitoring system
[17-Jun-2009 11:31:56] <mrayzenoss> examples?
[17-Jun-2009 11:32:34] <kisielk> well, we have an SNMP module that monitors quotas for one of our file servers
[17-Jun-2009 11:32:52] <jb> it outputs the quota via snmp?
[17-Jun-2009 11:32:57] <kisielk> yes
[17-Jun-2009 11:33:02] <jb> thats easy enough
[17-Jun-2009 11:33:04] <kisielk> it's in 5 arrays
[17-Jun-2009 11:33:11] <jb> just add it as a data source
[17-Jun-2009 11:33:14] <kisielk> one array is the project name, then the soft quota, hard quota
[17-Jun-2009 11:33:19] <kisielk> disk usage
[17-Jun-2009 11:33:19] <jb> and then you can set thresholds, graphs, etc on it
[17-Jun-2009 11:33:22] <kisielk> and then a message
[17-Jun-2009 11:33:42] <kisielk> yeah, but the threshold should come from the other SNMP data source
[17-Jun-2009 11:33:45] <kisielk> not be hard coded
[17-Jun-2009 11:33:49] <kisielk> because the quotas can change any time
[17-Jun-2009 11:33:56] <jb> ah
[17-Jun-2009 11:33:58] <jb> hmm
[17-Jun-2009 11:34:04] <kisielk> that was easy to do in Zabbix
[17-Jun-2009 11:34:07] <kisielk> but not so easy in Zenoss
[17-Jun-2009 11:36:12] <kisielk> anyway, I think that's the last major thing at that point
[17-Jun-2009 11:36:28] <kisielk> I guess I could write a nagios plugin for the checks
[17-Jun-2009 11:37:26] <daMaestro> mrayzenoss, thanks... reading now
[17-Jun-2009 11:41:10] <daMaestro> mrayzenoss, is there any easy way to add a simple prTable check?
[17-Jun-2009 11:41:39] <daMaestro> basically.. i'd like to simply say "send alert if there is not a process running called 'asdf'" and want this information to come from an snmp poll
[17-Jun-2009 11:45:56] <mrayzenoss> yeah, you should be able to add the process your want monitored. Section 2 of that paper covers it
[17-Jun-2009 11:49:31] <blocky> is there a way I can include the output of an external script into a report
[17-Jun-2009 11:49:53] <blocky> like generated html for example
[17-Jun-2009 11:52:41] bret is now known as CamargoBP
[17-Jun-2009 11:52:46] <mrayzenoss> blocky: I guess if you were doing something like this: http://www.zenoss.com/Members/cluther/creating-methods-for-custom-reports
[17-Jun-2009 11:52:55] <mrayzenoss> you could just return a block of HTML
[17-Jun-2009 11:56:29] <daMaestro> mrayzenoss, so it looks like i should add some proc items to our snmpd.conf and then traps will be sent?
[17-Jun-2009 11:57:26] <mrayzenoss> not traps
[17-Jun-2009 11:57:42] <mrayzenoss> snmpd will expose the process list for direct monitoring
[17-Jun-2009 11:59:02] <blocky> how would i have that call an external python script
[17-Jun-2009 11:59:39] <blocky> do i retrieve values from the rrd files manually?
[17-Jun-2009 12:00:08] <mrayzenoss> blocky: from that howto, you can put any Python you want into the script.
[17-Jun-2009 12:00:38] <mrayzenoss> this includes returning a block of HTML, as well as accessing the Zenoss system
[17-Jun-2009 12:00:47] <blocky> would it put the entire block of html into one cell of the table?
[17-Jun-2009 12:00:49] <mrayzenoss> the RRD stuff is accessible
[17-Jun-2009 12:00:58] <daMaestro> mrayzenoss, so how do i add it to the device, i guess has been my question
[17-Jun-2009 12:01:27] <daMaestro> mrayzenoss, it looks like i have to add a process class
[17-Jun-2009 12:02:37] <daMaestro> mrayzenoss, ah, d'oh
[17-Jun-2009 12:02:44] <daMaestro> mrayzenoss, just have to add it :-/;
[17-Jun-2009 12:03:37] <blocky> how do I access the rrd stuff through zenoss?
[17-Jun-2009 12:04:07] <mrayzenoss> when you create your custom report and add the custom method, you can access the device that is making the call. I'm not finding a good example
[17-Jun-2009 12:04:43] <mrayzenoss> I guess the device is the context
[17-Jun-2009 12:05:51] <blocky> hmm okay well ill play around with it a bit
[17-Jun-2009 12:06:16] <mrayzenoss> so you should be able to call context.getRRDValue("laLoadInt5_laLoadInt5") for example
[17-Jun-2009 12:06:34] <mrayzenoss> you may need to dig in the Dev Guide a bit
[17-Jun-2009 12:07:17] <mrayzenoss> some of the scripts in my Zenoss/MindTouch mashup did similar stuff http://zenossdeki.svn.sourceforge.net/viewvc/zenossdeki/zenoss/
[17-Jun-2009 12:07:32] <blocky> okay cool
[17-Jun-2009 12:07:37] <mrayzenoss> and return a block of HTML
[17-Jun-2009 12:08:00] <blocky> do you know, when a graph is generated by rrdtool with avg min and max values, how those are calculated
[17-Jun-2009 12:08:21] <blocky> cause ive got laloadint for 1 5 and 15 between 0 and 1, but my max value is around 400
[17-Jun-2009 12:08:54] <mrayzenoss> the value is whatever the datasource off of the template is pulling
[17-Jun-2009 12:09:04] <blocky> strange
[17-Jun-2009 12:09:16] <blocky> is this a good idea to try http://www.zenoss.com/community/docs/zenoss-dev-guide/2.2.4/ch09s01.html
[17-Jun-2009 12:09:18] <mrayzenoss> the max value would be whatever SNMP reported and stored in the RRD file
[17-Jun-2009 12:09:36] <mrayzenoss> yeah, I'd use the 2.4 version though
[17-Jun-2009 12:09:49] <blocky> well im using ssh command not snmp but yea
[17-Jun-2009 12:10:21] <blocky> must have got that data in there soemhow
[17-Jun-2009 12:12:45] <mrayzenoss> you can look at some of the stuff from RRDTool if you want to look at the RRD files closer
[17-Jun-2009 12:12:52] <mrayzenoss> http://oss.oetiker.ch/rrdtool/doc/index.en.html
[17-Jun-2009 12:13:47] <blocky> does he provide the python bindings?
[17-Jun-2009 12:14:00] <blocky> ah nm
[17-Jun-2009 12:17:07] <xulfer> How would I go about naming graphs in my template via snmp output?
[17-Jun-2009 12:28:17] <LW> question, I was recently asked if there is any way to identify and alarm on a % of delta change on a file system, so if overnight, a machine's disk utilization jumps 10%, then to alarm
[17-Jun-2009 12:29:56] <daMaestro> mrayzenoss, does the regex match against UCD-SNMP-MIB::prNames ?
[17-Jun-2009 12:32:07] <blocky> mrayzenoss: does a custom method return a single string always?
[17-Jun-2009 12:32:16] <venturaville> LW: not sure of the syntax but I believe you can do it based upon some time delta in getRRDValues: def getRRDValues(self, dsnames, start=None, end=None, function="LAST",
[17-Jun-2009 12:32:16] <venturaville> format="%.2lf", extraRpn=""):
[17-Jun-2009 12:32:46] <venturaville> I think the start and end ones are what you want......
[17-Jun-2009 12:33:50] <LW> intruiging
[17-Jun-2009 12:34:10] <venturaville> let me know if you get it working... I think that it might be useful for something I am doing too
[17-Jun-2009 12:35:42] * LW ponders what server to use to test
[17-Jun-2009 12:40:19] <venturaville> >>> f.getRRDValues('usedBlocks',start)
[17-Jun-2009 12:40:19] <venturaville> 2009-06-17 13:39:05 ERROR RenderServer Failed while generating summary
[17-Jun-2009 12:40:19] <venturaville> Traceback (most recent call last):
[17-Jun-2009 12:40:19] <venturaville> File "/opt/zenoss/Products/ZenRRD/RenderServer.py", line 380, in summary
[17-Jun-2009 12:40:19] <venturaville> values = rrdtool.graph(*gopts)[2]
[17-Jun-2009 12:40:20] <venturaville> error: Attempting to reuse 's_r'
[17-Jun-2009 12:40:24] <venturaville> hrm 2.4 is not very happy
[17-Jun-2009 12:43:54] <venturaville> >>> import time
[17-Jun-2009 12:43:54] <venturaville> >>> mystart = time.time() - (3600 * 8)
[17-Jun-2009 12:43:54] <venturaville> >>> args = dict(start=mystart)
[17-Jun-2009 12:43:54] <venturaville> >>> f.getRRDValue('usedBlocks',**args)
[17-Jun-2009 12:43:54] <venturaville> 10021455.52
[17-Jun-2009 12:43:55] <venturaville> >>> f.getRRDValue('usedBlocks')
[17-Jun-2009 12:43:57] <venturaville> 10019746.58
[17-Jun-2009 12:43:59] <venturaville> better
[17-Jun-2009 12:44:06] <venturaville> LW: I think that is what you want...........
[17-Jun-2009 12:44:51] <daMaestro> looks like it does
[17-Jun-2009 12:48:09] <blocky> okay ive done something wrong but i have no idea what
[17-Jun-2009 12:48:19] <blocky> followed this tutorial http://www.zenoss.com/Members/cluther/creating-methods-for-custom-reports
[17-Jun-2009 12:48:22] <blocky> and it works fine
[17-Jun-2009 12:48:29] <blocky> but when i create another method, it doesnt work
[17-Jun-2009 12:48:34] <blocky> just shows Unknown Column
[17-Jun-2009 12:48:40] <blocky> even with the exact same python code
[17-Jun-2009 12:57:42] <LW> venturaville: looks like I can do it by using a the "Create CMD" field for a datasource
[17-Jun-2009 13:02:21] <ckrough> blocky: post the method on pastebin.com
[17-Jun-2009 13:22:46] <blocky> ive fixed it
[17-Jun-2009 13:23:03] <blocky> though the only thing i changed was my method name from capital to lowercase
[17-Jun-2009 13:23:07] <blocky> so who knows
[17-Jun-2009 13:23:20] <blocky> right now im trying to figure out how to access the rest of the data from within python
[17-Jun-2009 13:23:30] <blocky> rrd data for example
[17-Jun-2009 13:23:51] <blocky> the ZenRRD interface doesnt seem to be geared towards actually reading and manipulating the data
[17-Jun-2009 13:24:47] <rmatte> anyone know if there's a way to make it so that users can reset their LDAP password in Zenoss?
[17-Jun-2009 13:24:51] <rmatte> or is it not possible?
[17-Jun-2009 13:31:29] <Matir> Hi, I just installed Zenoss 2.4.1 Core on CentOS 5, and I don't seem to be getting Page Menus anywhere. (At least, that's what I think it's called)
[17-Jun-2009 13:43:24] <blocky> anyone ever tried to get splunk results into zenoss via python
[17-Jun-2009 13:46:39] <LW> anybody know off hand how to change the file system template out on a single host?
[17-Jun-2009 13:49:53] <mloven> blocky: we're working on splunk-zenoss integration, but just via some syslog forwarding craziness. I'd be interested to see how it would work out using python...
[17-Jun-2009 14:02:59] <blocky> id like to have zenoss generate a report which includes splunk data
[17-Jun-2009 14:03:47] daMaestro|isBack is now known as daMaestro
[17-Jun-2009 14:03:48] <blocky> our splunk runs on a private proxy, so for example each morning along with the zenoss perf data id like to see how many TCP_DENIED messages were generated, sorted by IP, with a link to whois data for each ip
[17-Jun-2009 14:04:06] <blocky> im still trying to figure out the zenoss report framework
[17-Jun-2009 14:04:18] <blocky> do not really really understand the API at all yet, tbh
[17-Jun-2009 14:25:55] <blocky> is there any documentation for what i can use with the device context?
[17-Jun-2009 14:31:58] <ckrough> blocky: http://www.zenoss.com/community/docs/zenoss-dev-guide/2.4.0/apbs02.html#d4e4396
[17-Jun-2009 14:32:34] <blocky> beauty thanks
[17-Jun-2009 14:33:03] <ckrough> you can also do this in zendmd:
[17-Jun-2009 14:33:10] <ckrough> dev = devices.findDevice('_yourdevicename_')
[17-Jun-2009 14:33:10] <ckrough> print '\n'.join(dir(dev))
[17-Jun-2009 14:33:17] <ckrough> and it will return some stuff to play with
[17-Jun-2009 14:36:14] <blocky> cool
[17-Jun-2009 14:36:24] <blocky> zendmd is like a python wrapper?
[17-Jun-2009 14:44:06] <ckrough> yeah
[17-Jun-2009 14:44:21] <ckrough> its the parent class for all of zenoss, I believe
[17-Jun-2009 14:44:26] <ckrough> you can just
[17-Jun-2009 14:44:35] <ckrough> 'zendmd' as the zenoss user on your zenoss box
[17-Jun-2009 14:44:43] <ckrough> from the command line
[17-Jun-2009 14:47:28] <blocky> so do you know much about the getRRDValue func?
[17-Jun-2009 14:50:51] <blocky> it seems to return the current value just fine, but what if I want the min or max for a certain time range like what appears on the graphs
[17-Jun-2009 14:54:42] <ckrough> not sure about doing that in zenoss
[17-Jun-2009 14:55:02] <ckrough> we run a couple external perl scripts that build that data from rrdserv on the collectors
[17-Jun-2009 15:39:03] <blocky> if there was a function to iterate through each rrd file, store the max value and then use the heartbeat to show like: Load maximum was 0.75 at 2:30 this morning (click here for graph)
[17-Jun-2009 15:39:16] <blocky> that would be ideal
[17-Jun-2009 15:43:11] <daMaestro> anyone have a creative way to build a report that will break down bandwidth usage by daily total?
[17-Jun-2009 16:24:37] <daMaestro> anyone know what to look for to clear zenmodler heartbeat events?
[17-Jun-2009 16:27:51] <mrayzenoss> Event Manager->Clear All Heartbeats
[17-Jun-2009 16:30:18] <blocky> if I'm calling fetchRRDValue how do i set the resolution
[17-Jun-2009 16:33:32] <blocky> this is interesting
[17-Jun-2009 16:33:57] <blocky> if i try to 'import time' at the start of my method, i can't view the report, it redirects me to the login page and wont accept my pass
[17-Jun-2009 16:38:15] <daMaestro> mrayzenoss, i mean.. to find out why the heartbeat is alerting in the first place
[17-Jun-2009 16:39:16] <mrayzenoss> on the event is there anything interesting on the details?
[17-Jun-2009 16:39:23] <mrayzenoss> perhaps if you go to the log
[17-Jun-2009 16:39:25] <daMaestro> ah, i'll look in the details
[17-Jun-2009 16:39:26] <daMaestro> thanks.
[17-Jun-2009 16:39:50] <mrayzenoss> you can view the daemon logs from the Settings->Daemons (or from $ZENHOME/logs
[17-Jun-2009 17:24:27] <mrayzenoss> Zenoss Jobs forum is now live
[17-Jun-2009 17:24:31] <cgibbons> yay!
[17-Jun-2009 17:54:06] daMaestro|isBack is now known as daMaestro
[17-Jun-2009 17:58:19] <blocky> is zendmd strictly an interpreter
[17-Jun-2009 17:58:34] <blocky> strictly interactive i mean
[17-Jun-2009 17:58:41] <blocky> can i run .py files through it
[17-Jun-2009 18:05:51] <blocky> something is wrong with my graphs
[17-Jun-2009 18:06:09] <blocky> ive been collecting load data so all the numbers are between 0 and 1
[17-Jun-2009 18:06:13] <blocky> but the graph has scaled it somehow
[17-Jun-2009 21:30:10] <crow> anyone know the TALES syntax to run arbitrary code? I need to pull in some variables from the GET
[17-Jun-2009 21:45:06] <crow> http://zope-is-evil-666.idyll.org/
[18-Jun-2009 00:00:42] [disconnected at Thu Jun 18 00:00:42 2009]
[18-Jun-2009 00:00:42] [connected at Thu Jun 18 00:00:42 2009]
[18-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[18-Jun-2009 02:18:32] <detach-> hey guys, wtf is this zenoss-jobs mailing list about?
[18-Jun-2009 02:18:48] <detach-> can we have our e-mails removed from the list? none of us subscribed to it in the first place
[18-Jun-2009 03:00:18] <JulianMsanchez> hi
[18-Jun-2009 03:33:42] <sergeymasushko> hey guys! are you going to fix these spam messages ? it's really annoying.
[18-Jun-2009 03:38:21] <rhetts> its a bit annoying but everyone bitching isnt helping sicne there is obviously a problem
[18-Jun-2009 07:10:28] <cain22> hell all...
[18-Jun-2009 07:10:49] <cain22> hey, anyone else getting spammed from zenoss forumn?
[18-Jun-2009 07:13:05] <cain22> hello...
[18-Jun-2009 07:16:03] <cain22> anyone in here?
[18-Jun-2009 07:17:51] <cain22> list
[18-Jun-2009 07:18:59] <venturaville> zenoss forum is pretty quiet for me
[18-Jun-2009 07:19:25] <venturaville> irc is pretty quiet on here until around 9 or 10 EST
[18-Jun-2009 07:20:53] <cain22> hmm...hey, have you been spammed by the zenoss forumn?
[18-Jun-2009 07:21:58] <venturaville> zope
[18-Jun-2009 07:22:04] <venturaville> nope
[18-Jun-2009 07:22:42] <cain22> they've sent me like 30+ emials in less than a day
[18-Jun-2009 07:22:46] <cain22> its annoying
[18-Jun-2009 07:22:58] <venturaville> 30 email is nothing
[18-Jun-2009 07:23:23] <venturaville> though I don't even think the forum has sent me that much
[18-Jun-2009 07:23:47] <venturaville> users forum?
[18-Jun-2009 07:24:57] <cain22> nah
[18-Jun-2009 07:25:02] <cain22> zenoss-jobs
[18-Jun-2009 08:04:38] <detach-> hi guys YOU NEED TO FIX YOUR JOBS MAILING LIST IMMEDIATELY
[18-Jun-2009 08:11:25] <FooDog> haha, wow
[18-Jun-2009 08:15:40] <crow> ha
[18-Jun-2009 08:23:45] <crow> zenoss.com is sloooow today
[18-Jun-2009 08:37:21] <rmatte> When I'm modelling Linux servers using straight SNMP Zenoss isn't picking up anything for the OS tab
[18-Jun-2009 08:37:26] <rmatte> it's not even picking up interfaces
[18-Jun-2009 08:38:17] <rmatte> any thoughts?
[18-Jun-2009 08:38:35] <sergeymasushko> are there selected needed modeling plugins ?
[18-Jun-2009 08:39:27] <rmatte> the interface map collector plugin is bound to the device group
[18-Jun-2009 08:39:37] <rmatte> so it should be picking up on them
[18-Jun-2009 08:44:51] <sergeymasushko> rmatte: if it shouldn't maybe try to add it and remodel the device?
[18-Jun-2009 09:05:47] <dethstar> so, what's with the zenoss-jobs spam in my inbox this morning?
[18-Jun-2009 09:10:09] <sergeymasushko> dethstar: dunno, I believe Matt do not have a chance to see/fix it yet
[18-Jun-2009 09:15:05] <dethstar> sergeymasushko, that's cool... just wondering wtf was going on
[18-Jun-2009 09:15:46] <sergeymasushko> I guess it's some glitch
[18-Jun-2009 09:16:28] <LW> so my attempt to utilize the "Create cmd" field for an rrd value to create a derived value for identifing #delta over 24 hours was a failure.... I guess I may just be better with some sort command based helper script.....
[18-Jun-2009 09:22:05] <zenChild> When generating table headers in custom skins (.pt files), does anyone know how to sort numerically in the getTableHeader method?
[18-Jun-2009 09:31:38] <rmatte> What does zSnmpCommunities do again?
[18-Jun-2009 09:31:45] <rmatte> it doesn't seem to do anything from what I can see
[18-Jun-2009 09:32:31] <rmatte> oh, it's only for discovery
[18-Jun-2009 09:33:09] <crow> I think it's used for collecting performance stats as well
[18-Jun-2009 09:41:29] <LW> anybody know where I can find a list of ${here} tales expressions?
[18-Jun-2009 09:42:02] <crow> http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/apd.html
[18-Jun-2009 09:42:25] <cain22> hello all
[18-Jun-2009 09:42:56] <crow> LW: theres also the API doc, but its a bit unweildy for picking out available ecxpressions
[18-Jun-2009 09:45:43] <crow> hello cain
[18-Jun-2009 09:45:56] crow is now known as ckrough
[18-Jun-2009 09:58:20] * LW greps the codebase looking for examples
[18-Jun-2009 10:08:20] <ckrough> LW: you can also 'dir' the methods in zendmd
[18-Jun-2009 10:15:19] <ckrough> Anyone know how I can access the GET request data in a zope template
[18-Jun-2009 10:18:40] <venturaville> in tales expressions, I believe it is something like: request/blah
[18-Jun-2009 10:18:42] <venturaville> like request/URL
[18-Jun-2009 10:19:08] <ckrough> venturaville: will try, thanks
[18-Jun-2009 10:19:16] <venturaville> http://wiki.zope.org/zope2/REQUESTX
[18-Jun-2009 10:20:05] <ckrough> venturaville: yup, request/BLAH
[18-Jun-2009 10:25:19] <ckrough> http://pastebin.com/m44cab564 on line 48 I want to pass a drange in via the get, but Im no sure how to get to it.
[18-Jun-2009 10:25:53] <ckrough> request/blah doesnt work in this context, I guess because Im dealing with python and not zope templates at that level
[18-Jun-2009 10:33:22] <venturaville> <textarea rows="80" cols="20" tal:attributes="rows request/rows;cols request/cols">
[18-Jun-2009 10:33:31] <venturaville> I don't think you want to use it in a python context
[18-Jun-2009 10:33:49] <venturaville> http://www.faqs.org/docs/ZopeBook/AppendixC.html
[18-Jun-2009 10:34:52] <ckrough> venturaville: we talkin about the same thing? you lostme
[18-Jun-2009 10:35:20] <venturaville> yeah ... take a closer look at that line..............
[18-Jun-2009 10:35:46] <ckrough> the line you just pasted doesnt exist in my template
[18-Jun-2009 10:37:47] <venturaville> that was from the appendix
[18-Jun-2009 10:37:53] <venturaville> let me see if I can translate it to yours
[18-Jun-2009 10:38:09] <ckrough> ok, looking at it now
[18-Jun-2009 10:39:30] <venturaville> I'm at a loss as to how to translate it to what you are doing
[18-Jun-2009 10:39:42] <venturaville> the way listed on that AppendixC page is TAL specific
[18-Jun-2009 10:39:52] <venturaville> I am not sure how to get at that request object from the python side
[18-Jun-2009 10:40:33] <ckrough> may not be possible
[18-Jun-2009 10:41:48] <venturaville> look at Products/ZenReports/reports/Performance_Reports/CPU_Utilization.rpt
[18-Jun-2009 10:42:09] <venturaville> I think you can at least do it in another plugin and then include it in your report
[18-Jun-2009 10:44:04] <venturaville> objects python: (hasattr(request, 'doExport') and list(objects)) or objects;
[18-Jun-2009 10:44:12] <venturaville> and then referenced like:
[18-Jun-2009 10:45:37] <venturaville> request/blah
[18-Jun-2009 10:45:49] <venturaville> err no that is the tal way again
[18-Jun-2009 10:45:51] <ckrough> why does this feel so dirty
[18-Jun-2009 10:47:02] <venturaville> does here.REQUEST work?
[18-Jun-2009 10:48:17] <venturaville> >>> d = devices.findDevice('emon8snc1')
[18-Jun-2009 10:48:17] <venturaville> >>> d.REQUEST
[18-Jun-2009 10:48:17] <venturaville> <HTTPRequest, URL=http://localhost:8080>
[18-Jun-2009 10:48:19] <venturaville> ahah!
[18-Jun-2009 10:50:07] <ckrough> cool thanks
[18-Jun-2009 11:26:46] <blocky> anybody know why the scale of my load graphs doesn't match up with the rrd data
[18-Jun-2009 11:27:01] <blocky> im using the uptime parser
[18-Jun-2009 11:29:12] <blocky> all my rrd numbers are between 0 and 1 and the min/max on my graph is set correctly, but but graph is scaled up past 200m in some cases
[18-Jun-2009 11:30:55] <ckrough> m is micro
[18-Jun-2009 11:31:09] <rocket> hey matt
[18-Jun-2009 11:31:25] <ckrough> rocket: you scared
[18-Jun-2009 11:31:28] <ckrough> him off
[18-Jun-2009 11:31:34] <rocket> uh oh
[18-Jun-2009 11:31:52] <FooDog> m is milli, not micro
[18-Jun-2009 11:32:08] <ckrough> sorry, lol yeah
[18-Jun-2009 11:32:13] <blocky> i thought its mega
[18-Jun-2009 11:32:17] <ckrough> M is mega
[18-Jun-2009 11:32:27] <ckrough> case sensitive
[18-Jun-2009 11:32:28] <blocky> oh right
[18-Jun-2009 11:32:41] <blocky> so 200m is just 0.2 load?
[18-Jun-2009 11:32:48] <FooDog> correct
[18-Jun-2009 11:33:11] <blocky> is there a way I can get it to display that way?
[18-Jun-2009 11:33:26] <FooDog> (don't worry, that confused me forever also)
[18-Jun-2009 11:33:32] <ckrough> same
[18-Jun-2009 11:33:42] <rocket> blocky: its the formatting of the graph
[18-Jun-2009 11:33:42] <FooDog> i think that's an .rrd thing, not sure how to change it
[18-Jun-2009 11:33:55] <rocket> have to edit it in the template
[18-Jun-2009 11:34:10] <blocky> hmm ok
[18-Jun-2009 11:34:14] <rocket> and yes it is an rrd thing
[18-Jun-2009 11:34:29] <blocky> is there a way i can make 1 report composed of several smaller reports concatenated
[18-Jun-2009 11:35:31] <rocket> need to look at the rrd formatting options for the graph point in question
[18-Jun-2009 11:35:39] <rocket> eg look for something like %5.1lf%%
[18-Jun-2009 11:37:59] <blocky> turns out the min/max on my graph wasn't set correctly :S
[18-Jun-2009 11:38:09] <nzle> Can I restore a zenoss appliance config backup into a stack installed zenoss temporarily, and move it back to an appliance again? Some googling seemed to indicate there were some gotchas but didnt elaborate much.
[18-Jun-2009 11:38:48] <rocket> probably can do so .. I would guess that you just do the backup restore procedure etc
[18-Jun-2009 11:38:50] <nzle> also, having done a config backup on the appliance, how does one get it off the appliance. Is there a shell login?
[18-Jun-2009 11:38:57] <rocket> though I havent tried it myself ..
[18-Jun-2009 11:39:17] <rocket> nzle: there should be a root login
[18-Jun-2009 11:39:58] <nzle> One would assume, but I havent found yet where the root password is documented..
[18-Jun-2009 11:44:35] <cain22> if you bought the appliance, you could ask them...
[18-Jun-2009 11:45:16] <rocket> nzre it might be unset at the moment
[18-Jun-2009 11:47:05] <mrquick> there's no password on the appliance, just enter root
[18-Jun-2009 11:50:32] <blocky> how do i add other elements to a custom report besides a graph
[18-Jun-2009 11:50:44] <nzle> hmm, thats comforting... I asked my coworker in the other location to fire it up again for me.. Ill try that when he gets back to me. I'm trying to move the appliance to the main location that has access to all the necessary networks, but the box I wanted to run it on till we get our esxi server, is a xen box. Havent found a good way to run a vmware appliance unde xen yet so thinking in terms of restoring the config to a stack install. co worker want
[18-Jun-2009 11:50:44] <nzle> s to stay with the appliance though.
[18-Jun-2009 11:52:52] <blocky> if i set an rrd data source to a gauge, will it do any averaging of numbers
[18-Jun-2009 11:53:05] <blocky> cause my data source can only be integers but there's a decimal value in my rrd
[18-Jun-2009 11:58:17] <cain22> hey, was anyone else getting spammed by zenoss-jobs earlier today?
[18-Jun-2009 11:58:37] <mrquick> yeah, they blogged about it
[18-Jun-2009 12:03:50] <cain22> ah
[18-Jun-2009 12:10:37] <francois> Hi, is anyone aware of issues with SNMP modeling in 2.4.1 ?
[18-Jun-2009 12:11:13] <francois> the issue was described in the forum
[18-Jun-2009 12:11:16] <francois> http://forums.zenoss.com/viewtopic.php?t=10016&start=15
[18-Jun-2009 12:15:59] <ckrough> hmm. well, thats not good
[18-Jun-2009 12:16:21] <francois> Sorry, I have to leave right now
[18-Jun-2009 12:16:40] <francois> if you have any advice, please write it in the forum, thanks !
[18-Jun-2009 12:45:15] <cain22> anyone know when the next update comes out?
[18-Jun-2009 12:46:51] crow is now known as ckrouh
[18-Jun-2009 12:46:54] ckrouh is now known as ckrough
[18-Jun-2009 12:51:56] <rmatte> guys, I've got a really strange problem here that suddenly popped up at 2am this morning...
[18-Jun-2009 12:52:12] <rmatte> when I perform an snmpwalk on a firewall the requests go through looking like:
[18-Jun-2009 12:52:21] <rmatte> 13:45:27.351766 IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto UDP (17), length 87) 30.30.22.10.49304 > calibmfw01.superiorpropane.com.snmp: { SNMPv1 C=TuE892tjIt0Rg5a { GetNextRequest(35) R=1984145409 ip.ipAddrTable.ipAddrEntry.ipAdEntReasmMaxSize.209.82[|snmp] } }
[18-Jun-2009 12:52:25] <rmatte> notice the length of 87
[18-Jun-2009 12:52:31] <ckrough> do you hear a deep voice chanting "wiiiiinnnsstoooonn" from the server room?
[18-Jun-2009 12:52:47] <rmatte> lol
[18-Jun-2009 12:53:10] <rmatte> now, when Zenoss performs an SNMP poll on 4 specific OIDs it comes back with:
[18-Jun-2009 12:53:17] <rmatte> 13:48:36.229868 IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto UDP (17), length 596) 30.30.22.10.60701 > calibmfw01.superiorpropane.com.snmp: { SNMPv1 C=TuE892tjIt0Rg5a { GetRequest(540) R=2017622931 system[|snmp] } }
[18-Jun-2009 12:53:24] <rmatte> note the length of 596
[18-Jun-2009 12:53:51] <rmatte> The firewall is dropping the packets and not responding to the SNMP requests
[18-Jun-2009 12:54:03] <rmatte> it is also sending us traps with messages like:
[18-Jun-2009 12:54:04] <rmatte> calibmfw01.superiorpropane.com|SYSLOG-4|/Cisco/Syslog||4|20-Syslog Trap: 212005: incoming SNMP request (568 bytes) from IP address 30.30.22.10 Port 35294 Interface \"inside\" exceeds data buffer size, discarding this SNMP request.
[18-Jun-2009 12:54:43] <rmatte> I'm wondering why it's rejecting these packets
[18-Jun-2009 12:54:53] <rmatte> if I do snmpwalks for the exact same OIDs it works fine
[18-Jun-2009 12:55:10] <rmatte> It only happens when Zenoss runs through it's polling cycle
[18-Jun-2009 12:56:01] <ckrough> do you have a max packet size set on the firewall
[18-Jun-2009 12:56:06] <ckrough> per service
[18-Jun-2009 12:56:16] <rmatte> it's possible, I'll have to check the config
[18-Jun-2009 12:56:40] <ckrough> http://forums.zenoss.com/viewtopic.php?p=2409&sid=df1870792570ca1699a5f2225324f328
[18-Jun-2009 12:57:34] <ckrough> looks like cisco default max is 512 for that
[18-Jun-2009 12:57:48] <ckrough> reduce the max # of OIDs per request and see if it continues
[18-Jun-2009 12:58:08] <ckrough> there is a zproperty for it
[18-Jun-2009 12:58:36] <rmatte> ah
[18-Jun-2009 12:58:55] <ckrough> zMaxOIDPerRequest
[18-Jun-2009 12:59:44] <rmatte> I'll modify that, thanks for the help
[18-Jun-2009 13:00:20] <ckrough> np, let me know if it doesnt work
[18-Jun-2009 13:00:27] <rmatte> will do
[18-Jun-2009 13:01:12] <rmatte> man I can't wait until they fix this zproperties issue
[18-Jun-2009 13:01:27] <ckrough> which issue is that
[18-Jun-2009 13:01:47] <rmatte> http://dev.zenoss.com/trac/ticket/5071
[18-Jun-2009 13:01:52] <rmatte> that one
[18-Jun-2009 13:02:02] <rmatte> takes like 20 minutes to load the zProperties page sometimes
[18-Jun-2009 13:02:07] <ckrough> ouch
[18-Jun-2009 13:03:39] <rmatte> actually, I don't think "mlunt" understands the situation, breaking it in to multiple smaller Zenpacks wouldn't help, since it's also caused by having a lot of small Zenpacks installed
[18-Jun-2009 13:03:49] <rmatte> the more Zenpack data overall, the slower it goes
[18-Jun-2009 13:04:08] <rmatte> I'll have to point that note out to Matt next time I see him
[18-Jun-2009 13:04:13] <blocky> is there any way I can access the zenoss bindings from vanilla python
[18-Jun-2009 13:04:27] <ckrough> blocky: bindings?
[18-Jun-2009 13:05:19] <blocky> well, like the way i can access devices and their rrd info
[18-Jun-2009 13:05:37] <ckrough> blocky: are you aware of http://www.zenoss.com/Members/netdata/zendmd-tips-and-tricks/
[18-Jun-2009 13:06:11] <ckrough> meh, wrong page
[18-Jun-2009 13:06:30] <ckrough> but, zendmd... is that what you are looking for
[18-Jun-2009 13:06:45] <blocky> yea thats waht im using
[18-Jun-2009 13:07:00] <blocky> but i dont get how zendmd is different from python
[18-Jun-2009 13:07:07] <rmatte> so you want a python script that runs as though it were in zendmd?
[18-Jun-2009 13:07:19] <blocky> yea
[18-Jun-2009 13:07:30] <rmatte> http://www.zenoss.com/Members/netdata/use-a-zendmd-script-as-standalone-program/
[18-Jun-2009 13:07:31] <rmatte> voila
[18-Jun-2009 13:07:52] <rmatte> that's all that's different, throw that at the top of your script and you're sailing
[18-Jun-2009 13:08:33] <rmatte> the script needs to be run as the zenoss user to be able to work
[18-Jun-2009 13:09:40] <blocky> cool
[18-Jun-2009 13:09:47] <blocky> so lets say i want to access devicse
[18-Jun-2009 13:11:19] <rmatte> device = find("192.168.0.1")
[18-Jun-2009 13:11:23] <rmatte> device.whatever
[18-Jun-2009 13:12:11] <rmatte> device.getRRDValue('ssCpuSystem', default)
[18-Jun-2009 13:12:24] <rmatte> device.getRRDValue('ssCpuSystem', function="AVERAGE")
[18-Jun-2009 13:12:28] <rmatte> device.getRRDValue('ssCpuSystem', function="MAX")
[18-Jun-2009 13:12:41] <blocky> find?
[18-Jun-2009 13:13:03] <blocky> sorry im quite new at this
[18-Jun-2009 13:13:07] <rmatte> you're basically finding the path to the device with that IP or hostname, and then assigning the path to a variable
[18-Jun-2009 13:13:09] <blocky> all that tutorial gives me is dmd
[18-Jun-2009 13:13:18] <blocky> i dont have a find() function defind
[18-Jun-2009 13:13:23] <rmatte> zendmd is python
[18-Jun-2009 13:13:41] <rmatte> oh, you need to load certain functions for your script
[18-Jun-2009 13:13:48] <rmatte> hold on, I forget how I did it...
[18-Jun-2009 13:13:52] <blocky> where is find heh
[18-Jun-2009 13:14:17] <blocky> in zendmd the syntax is like dev = devices.findDevice('mag1')
[18-Jun-2009 13:14:43] <rmatte> you can do it the way I just said as well
[18-Jun-2009 13:14:47] <rmatte> have you tried it?
[18-Jun-2009 13:15:52] <rmatte> devices.finddevice would probably work better in terms of performance
[18-Jun-2009 13:15:52] <blocky> NameError: name 'find' is not defined
[18-Jun-2009 13:15:59] <blocky> i dont have devices either...
[18-Jun-2009 13:16:00] <rmatte> what are you typing exactly?
[18-Jun-2009 13:16:07] <rmatte> oh, you're just trying to run it as a script?
[18-Jun-2009 13:16:17] <blocky> im typing it into the interpreter
[18-Jun-2009 13:16:41] <rmatte> try dmd.devices.findDevice('mag1')
[18-Jun-2009 13:16:43] <blocky> oh it's findDevice
[18-Jun-2009 13:17:21] <rmatte> you might have to prefix it with dmd.
[18-Jun-2009 13:17:50] <rmatte> I use the line "for d in dmd.Devices.getSubDevices():" in one of my scripts and it works fine
[18-Jun-2009 13:18:27] <rmatte> that starts a for loop where it loops through each device in the database and sets d to the path to that device as it goes through
[18-Jun-2009 13:18:32] <blocky> cool its working
[18-Jun-2009 13:18:36] <rmatte> cool
[18-Jun-2009 13:18:41] <blocky> thanks
[18-Jun-2009 13:18:43] <rmatte> np
[18-Jun-2009 13:18:53] <blocky> now i wonder how i can get splunk involved
[18-Jun-2009 13:19:50] <rmatte> I've no knowledge of spunk, so you're on your own with that one hehe
[18-Jun-2009 13:20:13] <blocky> yea its got a python interface too
[18-Jun-2009 13:20:46] <rmatte> splunk* rather
[18-Jun-2009 13:21:11] <rmatte> I guess that's why it integrates so well with Zenoss
[18-Jun-2009 13:22:15] <blocky> well it will when im through with it
[18-Jun-2009 13:23:16] <ckrough> blocky: are you trying to go from splunk to zen or the other direction
[18-Jun-2009 13:23:34] <venturaville> splunk is good for summarizing app events for pushing up to something
[18-Jun-2009 13:23:43] <venturaville> so most likely splunk -> zen
[18-Jun-2009 13:24:00] <ckrough> yeah, thats what we do
[18-Jun-2009 13:24:12] <ckrough> devices -> syslog -> splunk
[18-Jun-2009 13:24:35] <venturaville> anyone had much experience with having splunk look directly at their logs?
[18-Jun-2009 13:24:53] <ckrough> venturaville: I have it following a file, yeah
[18-Jun-2009 13:24:56] <ckrough> no problems
[18-Jun-2009 13:25:09] <venturaville> how much data can it handle per second?
[18-Jun-2009 13:25:40] <ckrough> lemme calculate
[18-Jun-2009 13:28:11] <ckrough> Im only doing around 1mil lines per day, which is around 12 lines per secon
[18-Jun-2009 13:30:16] <ckrough> and its not dropping anything
[18-Jun-2009 13:30:16] <zenChild> I have a 1-to-many relationship established between a device and components. Is it possible to then have a 1-to-many relationship between each component and subcomponents? How does the modeler tie each subcomponent back to each component. At a high level you can think of it like so: a host can have multiple volume groups and each volume group can in turn have multiple logical volumes.
[18-Jun-2009 13:30:22] <ckrough> never hit its max
[18-Jun-2009 13:30:51] <blocky> yea its splunk to zen
[18-Jun-2009 13:31:07] <blocky> id like to get a summary in my report, stuff like how many failed logins
[18-Jun-2009 13:31:22] <blocky> need to know who's been accessing the squid proxy
[18-Jun-2009 13:32:04] <blocky> do custom zen reports only make tables with a row for each device
[18-Jun-2009 13:32:53] <ckrough> You may need to start doing custom reports: http://www.zenoss.com/Members/cluther/creating-methods-for-custom-reports/
[18-Jun-2009 13:33:11] <blocky> yea ive got a few of those methods
[18-Jun-2009 13:33:53] <blocky> but then how do i call a method from a report with something more flexible than a list of columns
[18-Jun-2009 13:34:24] <ckrough> gotta build your own report templates
[18-Jun-2009 13:34:25] <ckrough> sec
[18-Jun-2009 13:34:40] <ckrough> zenoss/Products/ZenReports
[18-Jun-2009 13:34:45] <ckrough> steal some ideas there
[18-Jun-2009 13:39:16] <ckrough> in the web interface, /zport/portal_skins/zenmodel/manage_main
[18-Jun-2009 13:39:31] <ckrough> there are a bunch of reports in there that you can copy/modify as well
[18-Jun-2009 14:05:16] <venturaville> ok so I am trying to trigger a loadDevice via zendmd
[18-Jun-2009 14:05:22] <venturaville> but it isn't showing up in the GUI
[18-Jun-2009 14:05:29] <venturaville> I can query it via zendmd though ......
[18-Jun-2009 14:05:37] <venturaville> any ideas?
[18-Jun-2009 14:08:14] <ckrough> search for it via IP in the gui?
[18-Jun-2009 14:08:23] <venturaville> it finally showed up
[18-Jun-2009 14:08:25] <venturaville> weird
[18-Jun-2009 14:08:34] <venturaville> can't seem to get this to work in the event class transform though ............
[18-Jun-2009 14:10:37] <venturaville> I am doing this in shell on a box: deviceName="myhostname" ; deviceClass="/Server/Linux" ; logger -p warn -t addtomonitoring "AddMeToMonitoring deviceName=${deviceName} deviceClass=${deviceClass}"
[18-Jun-2009 14:10:50] <venturaville> and have an event class that matches it
[18-Jun-2009 14:11:06] <venturaville> regex:
[18-Jun-2009 14:11:08] <venturaville> AddMeToMonitoring deviceName=(?P<mydeviceName>.*) deviceClass=(?P<mydeviceClass>.*)
[18-Jun-2009 14:11:20] <venturaville> transform:
[18-Jun-2009 14:11:22] <venturaville> dmd.DeviceLoader.loadDevice(evt.mydeviceName, evt.mydeviceClass)
[18-Jun-2009 14:11:46] <venturaville> seems to work in zendmd
[18-Jun-2009 14:11:53] <venturaville> but never does in events....
[18-Jun-2009 14:13:41] <venturaville> maybe I need to do a commit() in there??
[18-Jun-2009 14:19:25] <venturaville> basically trying to use dmd inside of an event.........
[18-Jun-2009 14:27:19] <venturaville> I guess what I am really asking is how to use dmd inside of an event............
[18-Jun-2009 16:00:45] <phasma> Hello. Can anybody give a link to manual, how create multi-graph report based on custom MIB OID ?
[18-Jun-2009 16:07:45] <ckrough> phasma: possibly read up on this: http://www.zenoss.com/community/docs/zenoss-dev-guide/2.4.0/ch12.html
[18-Jun-2009 18:12:51] * kisielk wishes the developer docs were better
[19-Jun-2009 00:00:42] [disconnected at Fri Jun 19 00:00:42 2009]
[19-Jun-2009 00:00:42] [connected at Fri Jun 19 00:00:42 2009]
[19-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[19-Jun-2009 04:17:11] <Dieterbe> Hi, I have a debian lenny 64bit. it's a brand new installed box. When i try to install zenoss on it (by using the .deb package for debian) I get errors
[19-Jun-2009 04:17:15] <Dieterbe> See: http://pastebin.com/m20855ac6
[19-Jun-2009 04:17:41] <Dieterbe> I tried both wgetting the deb for your site and dpkg -i, as well as adding your mirror and using aptitude. as you can see in the paste above
[19-Jun-2009 04:48:55] <Dieterbe> hmm seems like someone had the same problem on june 3rd. http://de.pastebin.ca/1446692
[19-Jun-2009 07:57:12] crow is now known as ckrough
[19-Jun-2009 08:05:07] <ckrough> morniing
[19-Jun-2009 08:09:05] <rocket> morning
[19-Jun-2009 09:49:27] <rocket> morning matt
[19-Jun-2009 09:49:46] <venturaville> mrayzenoss: ... just the person I was hoping to see
[19-Jun-2009 09:49:58] <mrayzenoss> good morning
[19-Jun-2009 09:50:23] <venturaville> mrayzenoss: I was hoping you knew how to get access to the dmd object from inside of an event class......
[19-Jun-2009 09:51:57] <mrayzenoss> let me check, I think there's an example of it somewhere
[19-Jun-2009 09:58:52] <mrayzenoss> venturaville: what are you trying to do?
[19-Jun-2009 10:03:28] <mrayzenoss> if you install the TransformReport ZenPack, there's a nice report that shows you every transform deployed on your box. Good for seeing examples
[19-Jun-2009 10:08:45] <venturaville> mrayzenoss: I'm trying to have a syslog entry from a host trigger a loadDevice .......
[19-Jun-2009 10:08:54] <venturaville> basically trying to automate addition of hosts from puppet
[19-Jun-2009 10:09:08] <mrayzenoss> how's the puppet Zenoss module?
[19-Jun-2009 10:09:18] <venturaville> eh?
[19-Jun-2009 10:09:31] <mrayzenoss> http://blog.zenoss.com/2009/01/13/new-zenoss-module-for-puppet/
[19-Jun-2009 10:09:47] <mrayzenoss> I assume you were using it
[19-Jun-2009 10:09:51] <venturaville> nope
[19-Jun-2009 10:09:57] <venturaville> though I have looked at it
[19-Jun-2009 10:10:07] <venturaville> it doesn't quite match what I need done though
[19-Jun-2009 10:10:10] <mrayzenoss> ok
[19-Jun-2009 10:11:25] <rocket> venturaville: what is it missing?
[19-Jun-2009 10:15:21] <zenChild> Anyone ever have an issue testing a SSH command for datasource where you get "command not found"? It seems as though it is trying to run the command locally even though I am testing against a remote system in which that command does indeed exist.
[19-Jun-2009 10:15:56] <rocket> zenChild: path issues?
[19-Jun-2009 10:16:10] <venturaville> rocket: it is a bit of a infrastructural issue here... not so a much a problem with the package :-)
[19-Jun-2009 10:16:14] <rocket> zenChild: eg did you have the full path to the command? /usr/sbin/foo etc
[19-Jun-2009 10:16:27] <zenChild> it's an appliance and the command is in the default path.
[19-Jun-2009 10:16:38] <zenChild> my modeler runs the same command and it succeeds.
[19-Jun-2009 10:19:33] <zenChild> is there a way i can test the datasource command on the command line so I can see the debug output?
[19-Jun-2009 10:20:21] <zenChild> I tried running...: zentestcommand -d <mydev> --datasource=lsmdiskgrp
[19-Jun-2009 10:20:39] <zenChild> but I get: No datasource lsmdiskgrp applies to device <mydev>
[19-Jun-2009 10:21:01] <zenChild> I think that's because it is a component datasource
[19-Jun-2009 10:21:37] <mrayzenoss> zenChild: http://dev.zenoss.org/trac/ticket/5175
[19-Jun-2009 10:21:54] <mrayzenoss> that's not for SSH, but the 'TEST' button seems to suck
[19-Jun-2009 10:22:02] <mrayzenoss> you can test with SSH though
[19-Jun-2009 10:22:18] <mrayzenoss> zencommand run -v10 device
[19-Jun-2009 10:24:32] <zenChild> Thanks mrayzenoss. I must be missing something somewhere because I don't see any evidence of collection for my datasource. The template is binding to the individual components though so I'm not sure where it's going wrong.
[19-Jun-2009 10:24:54] <mrayzenoss> yeah, that's what is nuts
[19-Jun-2009 10:24:58] <mrayzenoss> the test will work
[19-Jun-2009 10:25:11] <mrayzenoss> and the zencommand log won't show anything, just the path of the command it's trying to run
[19-Jun-2009 10:25:27] <mrayzenoss> which isn't exactly your issue, just a heads up
[19-Jun-2009 10:26:01] <zenChild> I should see something in the output of zencommand though shouldn't I... or do I need to peek through the logs?
[19-Jun-2009 10:27:44] <mrayzenoss> well, the zencommand with the -v10 will get you full debugging
[19-Jun-2009 10:27:48] <zenChild> ahah! I found my issue in the zenhub.log.
[19-Jun-2009 10:27:50] <mrayzenoss> which won't be in the logs
[19-Jun-2009 10:28:00] <zenChild> It's not finding my parser correctly.
[19-Jun-2009 10:30:54] <venturaville> mrayzenoss: I can't find a working example of dmd usage in transforms
[19-Jun-2009 10:31:41] <zenChild> I don't know if anyone has an IBM SVC out there, but I'm trying to write a ZenPack to monitor it. I have all the scripts to parse the performance data previously written in Perl and am in the process of cleansing them before I put them on the project repository. If anyone wants to help out or just test you can grab the code here: http://github.com/zenchild/ZenPacks.ndgov.IBMsvc/tree/master
[19-Jun-2009 10:32:05] <venturaville> zenChild: I've got SVC monitoring on my list to work on as well........
[19-Jun-2009 10:32:41] <zenChild> nice. maybe we can collaborate. Do you have anything written yet?
[19-Jun-2009 10:33:01] <venturaville> not a thing... its on my list to work on next month
[19-Jun-2009 10:33:41] <zenChild> I'd love some help. I come from a perl/ruby background so python is beating up on me a bit.
[19-Jun-2009 10:34:01] <venturaville> I'm working on a way to monitor the IBM arrays behind them as well
[19-Jun-2009 10:34:21] <venturaville> I got some docs from the engineers at LSI about how to parse out the SMcli commands for that too
[19-Jun-2009 10:36:03] <venturaville> I was looking at the CIM agents to try and get performance data out of them
[19-Jun-2009 10:36:12] <venturaville> (the SVCs)
[19-Jun-2009 10:36:18] <venturaville> but my knowledge of CIM is lacking.....
[19-Jun-2009 10:36:20] <zenChild> That would be sweet. For now I'm mostly interested in the SVC, but getting the DSx controllers would be an added bonus. We also just got in an XIV and I'd love to be able to import that data into Zenoss. Anyway, when you're ready to start working on the SVC let me know. In the meantime I'll keep plugging away.
[19-Jun-2009 10:36:47] <zenChild> CIM will be v2.0 of the ZenPack
[19-Jun-2009 10:37:27] <zenChild> I was looking at CIM for getting info out of ESXi directly. It can become a spiderweb rather quickly.
[19-Jun-2009 10:39:32] <venturaville> I've been trying to use wbemcli to talk to it
[19-Jun-2009 10:39:40] <venturaville> but the docs for it are lacking
[19-Jun-2009 10:39:47] <venturaville> and they are mostly all windows specific
[19-Jun-2009 10:40:09] <zenChild> That's what I was using too. Have you looked at yawn for browsing it?
[19-Jun-2009 10:40:16] <venturaville> got a link?
[19-Jun-2009 10:40:30] <zenChild> http://pywbem.wiki.sourceforge.net/YAWN
[19-Jun-2009 10:40:52] <zenChild> It came in handy when I was going through VMware's CIM docs.
[19-Jun-2009 10:41:02] <zenChild> I could kind of follow along...
[19-Jun-2009 10:41:25] <venturaville> much nicer
[19-Jun-2009 10:44:10] <mrayzenoss_> venturaville: zenChild: Don't know if this is useful or not, http://www.zenoss.com/community/wiki/cim-wbem/
[19-Jun-2009 10:44:27] <mrayzenoss_> Kells started on a CIM ZenPack for 2.1, haven't looked at it yet
[19-Jun-2009 10:45:24] <venturaville> looked interesting, and I played with it, but without knowing much about CIM and better docs on it I can't use it
[19-Jun-2009 10:45:31] <zenChild> Thanks. I'll have a look. Now if we can just get IBM to publish a CIMOM map of the SVC ...
[19-Jun-2009 10:45:42] <venturaville> should not need it in theory
[19-Jun-2009 10:45:51] <venturaville> CIMOM is somewhat self documenting
[19-Jun-2009 10:46:08] <venturaville> tcpdump is helpful too........................
[19-Jun-2009 10:46:15] <zenChild> still makes it nice to look at when you're coding...
[19-Jun-2009 10:46:51] <zenChild> i guess that's what YAWN attempts to do....
[19-Jun-2009 12:31:23] <ckrough> is this in seconds? "Cache Timeout - Specify the cache timeout value for the event monitor."
[19-Jun-2009 12:32:54] <ckrough> From the descriptions, I dont get a clear indication of what happens when these thresholds are reached: http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch14s02.html#d4e5211, "2.3. Changing Event Manager Cache Settings"
[19-Jun-2009 12:34:12] <blocky> can somebody point me to a tutorial for creating a custom graph template?
[19-Jun-2009 12:34:32] <ckrough> blocky: you mean fully custom, like we were discussing yesterday?
[19-Jun-2009 12:34:54] <ckrough> blocky: I've found bits and pieces that help you along, but I havent found a full blown how-to
[19-Jun-2009 12:35:09] <ckrough> blocky: but Im notoriously bad at finding things on zenoss.com
[19-Jun-2009 12:37:03] <blocky> fully custom yea
[19-Jun-2009 12:37:09] <blocky> ive got a report where I can edit the html
[19-Jun-2009 12:37:15] <blocky> but I dont know how to access content
[19-Jun-2009 12:37:31] <ckrough> TALES Expressions
[19-Jun-2009 12:37:48] <blocky> ive defined a device template with a load and a network graph
[19-Jun-2009 12:37:59] <blocky> and ive got a report for 5 devices that shows one of each for a total of 10 graphs
[19-Jun-2009 12:38:13] <ckrough> ok
[19-Jun-2009 12:38:17] <blocky> now if i could insert that entire report along with a few others...
[19-Jun-2009 12:39:17] <ckrough> Not sure
[19-Jun-2009 12:39:30] <ckrough> rocket or mray can probably point you in the right direction if you see them on
[19-Jun-2009 12:40:12] <blocky> forums.zenoss.com is failing to resolve now
[19-Jun-2009 12:40:24] <ckrough> working for me
[19-Jun-2009 12:41:08] <rocket> blocky: cant access device content? eg trying to populate datapoints?
[19-Jun-2009 12:41:42] <blocky> well no my RRDs are being collected fine
[19-Jun-2009 12:42:35] <blocky> initially what i'd like to do is agreggate a few different reports in one, so i can email it off in a nice neat package without requiring separate emails
[19-Jun-2009 12:44:23] <blocky> it seems like there's two different ways of defining a report
[19-Jun-2009 12:44:28] <blocky> pure html with tales
[19-Jun-2009 12:44:31] <rocket> blocky: so a multigraph report?
[19-Jun-2009 12:44:59] <blocky> well I'd like to have more than graphs on it, if that's possible..
[19-Jun-2009 12:45:21] <ckrough> It sounds like you want to merge a couple different types of reports
[19-Jun-2009 12:45:57] <blocky> ideally yes but i wasn't sure if that was possible, so I thought attaching them together would be almost as good
[19-Jun-2009 12:47:07] <ckrough> Can anyone explain what happens when these thresholds are reached: http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch14s02.html#d4e5211, "2.3. Changing Event Manager Cache Settings"
[19-Jun-2009 12:47:16] <rocket> blocky: I havent done too much with reports yet but I believe you would still be able to do that with a multigraph report .. you might have to do some python and html editing to get exactly what you are after
[19-Jun-2009 12:47:35] <blocky> what if not all the data im after can be accessed on every device
[19-Jun-2009 12:47:48] <blocky> like, they are all linux boxes so they all have load and network stats
[19-Jun-2009 12:47:57] <rocket> blocky: most likely it will be blank or show up as n/a
[19-Jun-2009 12:48:08] <ckrough> blocky: Make sure they (zenoss) know what you want. I am frequently pushing for improved reporting too. more voices = more priority
[19-Jun-2009 12:48:09] <blocky> but only one is a squid proxy, so only one has the list of denied IPs
[19-Jun-2009 12:48:41] <blocky> oaky who do I mass email flood
[19-Jun-2009 12:48:50] <rocket> blocky: thats where the python probably will come in .. might need to write a function that shows it if available etc .. or shows na if not
[19-Jun-2009 12:49:39] <blocky> that shouldn't be too bad
[19-Jun-2009 12:49:46] <ckrough> somewhere in the forums Mray has a feature request thread
[19-Jun-2009 12:49:57] <blocky> im not 100% sure what I want yet
[19-Jun-2009 12:50:08] <blocky> i mean, i know what i want the result to be, but not how it would fit with zenoss
[19-Jun-2009 12:50:15] <ckrough> ckrough: I know what you mean
[19-Jun-2009 12:50:50] <ckrough> I want them to object-orient the reporting system. I want to define a graph called "x", then be able to re-use "x" in any report where it makes sense
[19-Jun-2009 12:51:04] <ckrough> right now I have to redefine everything for every graph, across several zenoss installs. tedious
[19-Jun-2009 12:51:11] <blocky> i guess i could write a python script that generates all the HTML and hard code the graph URLs
[19-Jun-2009 12:52:03] <rocket> blocky: thats a possibility .. but doesnt sound very flexible but works for quick and dirty
[19-Jun-2009 12:52:16] <blocky> i dunno if reportmail would be able to use that, it might make more sense to send the email myself
[19-Jun-2009 12:52:19] <ckrough> blocky: we use that method for projectors in the NOC, etc...
[19-Jun-2009 12:53:51] <blocky> i like the way I can define data sources and threshholds and graphs at any level in the device hierarchy
[19-Jun-2009 12:53:57] <blocky> and have them be applied to each device
[19-Jun-2009 12:54:08] <rocket> blocky: did you look at the dev guide for reports?
[19-Jun-2009 12:54:19] <rocket> http://www.zenoss.com/community/docs/zenoss-dev-guide/2.4.0/ch14s01.html
[19-Jun-2009 12:55:22] <blocky> i saw that, i dont really know what it means
[19-Jun-2009 12:55:43] <blocky> can TALES markup access the other zenoss reports ive already defined, or the graphs
[19-Jun-2009 12:56:49] <rocket> blocky: examples are on the server under Products/ZenReports
[19-Jun-2009 12:57:02] <blocky> like how is a custom device report related to a custom graph report
[19-Jun-2009 12:57:47] <rocket> blocky: is there an existing report similar to what you are trying to accomplish?
[19-Jun-2009 12:57:47] <blocky> one is a table of python methods vs devices, the other is a list of which graphs
[19-Jun-2009 12:58:50] <blocky> if I could get a custom graph report and several custom device reports on the same page, and each based on a different list of devices, that would go a long way
[19-Jun-2009 12:59:02] <blocky> i could write the python myself for the device report methods
[19-Jun-2009 12:59:25] <rocket> eg you want to mix a Performance_Report with a Device_Report
[19-Jun-2009 12:59:49] <blocky> yea
[19-Jun-2009 12:59:58] <ckrough> given the way zenoss/zope generate reports based on objects within a tree, I am not sure how you could do that within a single report in zenoss
[19-Jun-2009 13:00:14] <blocky> eventually id like to be able to add content which is wholly external, like from splunk
[19-Jun-2009 13:00:20] <rocket> ckrough: your report collects groups of objects together
[19-Jun-2009 13:00:58] <ckrough> rocket: Im kinda stuck at the same point as blocky , how did you get past this stage?
[19-Jun-2009 13:01:04] <blocky> sorry where is Products/ZenReports
[19-Jun-2009 13:01:10] <rocket> ckrough: and you process them in different tables
[19-Jun-2009 13:01:34] <ckrough> blocky: in the filesystem
[19-Jun-2009 13:01:43] <ckrough> $zenhome/Products...
[19-Jun-2009 13:02:54] <ckrough> some reports are zope objects, which you can access through the /manage interface in the gui, others are on the filesystem. I believe I read that it's recommended to create file based reports over zope objects.
[19-Jun-2009 13:03:12] <ckrough> blocky: Im assuming you know about /manage in the gui, yes?
[19-Jun-2009 13:03:27] <blocky> yea ive had a look at it
[19-Jun-2009 13:03:47] <blocky> it does shed a bit of light
[19-Jun-2009 13:03:52] <ckrough> spend some time clicking around on objects in there, it helps quite a bit
[19-Jun-2009 13:03:53] <ckrough> yeah
[19-Jun-2009 13:04:03] <rocket> blocky: ckrough basically you guys need to figure out the syntax of tal expressions as shown in those sample reports
[19-Jun-2009 13:04:34] <rocket> the tal expressions create groups of objects that you loop through calling methods etc
[19-Jun-2009 13:05:04] <ckrough> blocky: /zport/portal_skins/zenmodel <-- this is a goldmine of examples
[19-Jun-2009 13:05:42] <rocket> eg within <tal:block> </tal:block> you can grab any group of objects .. display them etc
[19-Jun-2009 13:06:07] <rocket> define a new set of <tal> blocks for different objects you want to work with in the same html report
[19-Jun-2009 13:06:50] <ckrough> and within those blocks, {here} will be in reference to whatever point in the class tree you ask for the report?
[19-Jun-2009 13:07:06] <ckrough> ie: /devices/network/viewMyReport
[19-Jun-2009 13:07:47] <rocket> gotta run
[19-Jun-2009 13:07:52] <ckrough> thanks, see ya
[19-Jun-2009 13:07:53] <dberger> does anyone know why a location with devices in it might not show up on the google map portlet?
[19-Jun-2009 13:15:08] <mrayzenoss> dberger: on the location, what do you have in the "Address" field?
[19-Jun-2009 13:16:01] <dberger> 1450 Ala Moana Blvd # 3241
[19-Jun-2009 13:16:02] <dberger> Honolulu, HI
[19-Jun-2009 13:16:17] <dberger> it shows up in http://localhost:8080/zport/dmd/Locations/Hawaii/locationGeoMap
[19-Jun-2009 13:17:06] <twm1010> Have you assigned any devices to the location yet?
[19-Jun-2009 13:17:16] <twm1010>
[19-Jun-2009 13:17:27] <dberger> it even shows up in the portlet if i change it to the default location...but then the other location I've entered disappears
[19-Jun-2009 13:17:32] <dberger> yes
[19-Jun-2009 13:17:44] <twm1010> hrmm... clear browser cache and reload page?
[19-Jun-2009 13:18:45] <blocky> any good primers on tal markup?
[19-Jun-2009 13:19:15] <mrayzenoss> blocky: http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/apds02.html
[19-Jun-2009 13:20:27] <dberger> i reload, but clearing browser cache won't do me any good
[19-Jun-2009 13:21:28] <ckrough> Can anyone explain what happens when these thresholds are reached: http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch14s02.html#d4e5211, "2.3. Changing Event Manager Cache Settings"
[19-Jun-2009 13:22:30] <ckrough> heh heh: "The lower the number, the more responsible the history will be"
[19-Jun-2009 13:22:31] <mrayzenoss> dberger: ok, so the /Locations/Hawaii/locationGeoMap never works?
[19-Jun-2009 13:23:36] <mrayzenoss> and the portlet is intermittent?
[19-Jun-2009 13:23:42] <dberger> no, that works
[19-Jun-2009 13:23:52] <dberger> but the portlet never displays more than one point
[19-Jun-2009 13:23:59] <dberger> it only displays the base location
[19-Jun-2009 13:24:24] <mrayzenoss> do you have stuff assigned to /Locations/?
[19-Jun-2009 13:25:53] <dberger> the table in /zport/dmd/Locations shows 3 devices under both locations
[19-Jun-2009 13:26:15] <mrayzenoss> what's the other location?
[19-Jun-2009 13:26:34] <dberger> 200 W Madison St
[19-Jun-2009 13:26:35] <dberger> Suite 700
[19-Jun-2009 13:26:35] <dberger> Chicago, IL 60606
[19-Jun-2009 13:26:45] <dberger> named Chicago
[19-Jun-2009 13:27:57] <mrayzenoss> hmm... everything works for me
[19-Jun-2009 13:28:07] <mrayzenoss> 3 devices in 'Chicago', 2 in 'Hawaii'
[19-Jun-2009 13:28:12] <mrayzenoss> 2 in /
[19-Jun-2009 13:28:32] <mrayzenoss> dberger: which version are you running?
[19-Jun-2009 13:29:12] <dberger> 2.4.1-linux-x64
[19-Jun-2009 13:31:35] <mrayzenoss> so Hawaii doesn't render, but Chicago does in the portlet...
[19-Jun-2009 13:33:31] <dberger> yes, unless i make hawaii the default location
[19-Jun-2009 13:36:02] <dberger> in which case i have the exact opposite problem
[19-Jun-2009 13:39:35] <mrayzenoss> dberger: not sure what's wrong since I can't recreate it
[19-Jun-2009 13:40:01] <dberger> does it matter that i don't have any network links setup or anything like that?
[19-Jun-2009 13:40:42] <dberger> this is all just a test installation with several devices on the local network auto-discovered
[19-Jun-2009 13:41:43] <dberger> okay...i think i have it
[19-Jun-2009 13:42:02] <dberger> i see both now when i change the base location to /Locations
[19-Jun-2009 13:42:28] <dberger> i didn't realize "base location" meant tree root
[19-Jun-2009 13:42:50] <dberger> i thought it referred to the location that received focus by default
[19-Jun-2009 13:42:54] <dberger> thanks for your help!
[19-Jun-2009 13:42:57] <mrayzenoss> so it's working when you set it to /Locations?
[19-Jun-2009 13:43:17] <dberger> yes
[19-Jun-2009 13:43:21] <mrayzenoss> cool, you solved your problem and my printer is now located in Hawaii
[19-Jun-2009 13:43:33] <dberger> everybody wins
[19-Jun-2009 13:43:42] <dberger> it's actually located at a pizza shop in hawaii
[19-Jun-2009 13:43:50] <dberger> so started printing up fax orders!
[19-Jun-2009 14:10:11] <ckrough> maybe I could hitch a ride to hawaii with you? http://www.flickr.com/photos/zenoss/3573421972/in/photostream/
[19-Jun-2009 14:10:31] <ckrough> Is that in Annapolis Matt?
[19-Jun-2009 14:11:13] <mrayzenoss> yeah, that was in Annapolis
[19-Jun-2009 14:11:17] <ckrough> Love that place
[19-Jun-2009 14:49:59] <dberger> does anyone know of a way to create/destroy arbitrary network links?
[19-Jun-2009 14:51:13] <mrayzenoss> you can add/delete interfaces and routes on the device
[19-Jun-2009 14:51:48] <mrayzenoss> it'll get re-added on the next modeling unless you lock it
[19-Jun-2009 15:07:12] bret is now known as Bret
[19-Jun-2009 15:07:14] Bret is now known as Bret[Jive]
[19-Jun-2009 15:08:53] <dberger> could this be used to dummy a network?
[19-Jun-2009 15:09:17] <mrayzenoss> absolutely http://www.zenoss.com/Members/netdata/create-manual-device-dependencys/
[19-Jun-2009 15:12:24] <ckrough> hmmm. How does one call a zope python script from within a TAL block; "myVariable container/myScript;"?
[19-Jun-2009 19:46:02] <daMaestro> anyone doing JMX monitoring?
[19-Jun-2009 19:46:45] <daMaestro> we are getting "Error calling mbean solr:type=org.apache.solr.handler.component.SearchHandler: Exception: no MBean registered with name: solr:type=org.apache.solr.handler.component.SearchHandler" when trying to monitor a solr instance
[19-Jun-2009 20:14:52] daMaestro|isBack is now known as daMaestro
[19-Jun-2009 20:25:14] <daMaestro> ah.. it was just a mis-configured bean location
[20-Jun-2009 00:00:42] [disconnected at Sat Jun 20 00:00:42 2009]
[20-Jun-2009 00:00:42] [connected at Sat Jun 20 00:00:42 2009]
[20-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[20-Jun-2009 07:07:04] RoninX341 is now known as etank
[21-Jun-2009 00:00:42] [disconnected at Sun Jun 21 00:00:42 2009]
[21-Jun-2009 00:00:42] [connected at Sun Jun 21 00:00:42 2009]
[21-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[21-Jun-2009 09:22:33] <Appiah_> Do i need to add a networkin under /Networks before the auto-discovery to work for that network?
[21-Jun-2009 09:23:00] <Appiah_> I'm asking beacuse I can't test it myself right now >_>
[22-Jun-2009 00:00:42] [disconnected at Mon Jun 22 00:00:42 2009]
[22-Jun-2009 00:00:42] [connected at Mon Jun 22 00:00:42 2009]
[22-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[22-Jun-2009 08:50:17] <ckrough> morning
[22-Jun-2009 08:51:23] <dberger> scared him off
[22-Jun-2009 08:51:45] <ckrough> cant blame him
[22-Jun-2009 08:59:41] <Dieterbe> hehe
[22-Jun-2009 09:17:38] <ainc> hello all...
[22-Jun-2009 09:17:44] <dberger> morning
[22-Jun-2009 09:18:09] <ckrough> morning
[22-Jun-2009 09:18:28] <ckrough> morning matt
[22-Jun-2009 09:18:34] <ainc> is there a way to set a threshold to only send you a alarm when it has hit the limit like say, 6 times in a row
[22-Jun-2009 09:19:03] <rocket> Morning Matt ...
[22-Jun-2009 09:19:31] <ckrough> ainc: you can specify count in the alerting rule
[22-Jun-2009 09:20:32] <dberger> you can also make it escalate severity and set your notification rule to a higher severity
[22-Jun-2009 09:20:35] <ckrough> ainc: this will not work if your threshold alert auto-clears and come back repeatedly.
[22-Jun-2009 09:22:56] <ckrough> we are lookin for a way to alert on a bouncin threshold. no solution at the moment
[22-Jun-2009 09:27:39] <ainc> sorry, I don't want it to create an event in the viewer unless it hits a count...not an alarm
[22-Jun-2009 09:37:21] <clustermagnet> gents, i am evaluating monitoring products
[22-Jun-2009 09:38:29] <clustermagnet> i am trying to understand the difference between your free product, and the enterprise license
[22-Jun-2009 09:38:44] <clustermagnet> other than better vmware support, what are the differences?
[22-Jun-2009 09:39:01] <clustermagnet> im searching on your site, havent found a chart that would make it easy to compare
[22-Jun-2009 09:39:04] <clustermagnet> thanks!
[22-Jun-2009 09:39:18] <ckrough> clustermagnet: there may not be any official zenoss reps activein the room at the moment.
[22-Jun-2009 09:40:26] <clustermagnet> ckrough: no problem, dont care about official/nonofficial
[22-Jun-2009 09:40:42] <clustermagnet> ckrough: can you key in?
[22-Jun-2009 09:40:46] <slestak> i need to monitor a multithreaded jobscheduler in my aix based erp system
[22-Jun-2009 09:41:03] <ckrough> been so long since I used CORE... trying to remem
[22-Jun-2009 09:41:04] <ckrough> ber
[22-Jun-2009 09:41:14] <mrayzenoss> clustermagnet: I'm here
[22-Jun-2009 09:41:28] <mrayzenoss> clustermagnet: I can answer any questions you have
[22-Jun-2009 09:42:24] <slestak> i am thinking of using some of the backup-heartbeat related checks. has anyone done anything similar?
[22-Jun-2009 09:43:12] <slestak> i saw a writeup on a blog about using python xml-rpc from bash scripts to update a zenoss check. that could be interesting
[22-Jun-2009 09:43:14] <mrayzenoss> clustermagnet: Additional features in Enterprise are targeted at large Enterprise installations that need integration with commercial offerings (VMware, Oracle, commercial Unixes, Remedy, etc.) and need support. There is also additional functionality for service providers
[22-Jun-2009 09:43:33] <mrayzenoss> slestak: lots of people use XML-RPC to push events into the system
[22-Jun-2009 09:43:41] <clustermagnet> mrayzenoss: thanks!
[22-Jun-2009 09:43:48] <mrayzenoss> clustermagnet: anything specific?
[22-Jun-2009 09:43:51] <clustermagnet> mrayzenoss: sure
[22-Jun-2009 09:43:56] <clustermagnet> mrayzenoss: netapp, vmware, cisco
[22-Jun-2009 09:44:01] <dberger> slestak: have the link?
[22-Jun-2009 09:44:20] <clustermagnet> mrayzenoss: currently , we dont have a budget for zenoss
[22-Jun-2009 09:44:30] <mrayzenoss> clustermagnet: so can be monitored with Core and there are Community ZenPacks for Netapp, VMware and some Cisco stuff
[22-Jun-2009 09:44:34] <clustermagnet> i would HATE to setup nagios for now, and migrate when funds are available
[22-Jun-2009 09:44:55] <mrayzenoss> clustermagnet: understood
[22-Jun-2009 09:45:02] <slestak> dberger: im multitaskingright now. main workstation is down so i am trioubleshooting while using my netbook for now.
[22-Jun-2009 09:45:11] <clustermagnet> mrayzenoss: so it would be viable to run the Core for now... and upgrade to Enterprise when new budget is negotiated?
[22-Jun-2009 09:45:14] <mrayzenoss> we have a lot of users doing those for sure
[22-Jun-2009 09:45:21] <clustermagnet> mrayzenoss: super
[22-Jun-2009 09:45:31] <slestak> dberger: i can get if for you shortly, are you going to be aroubnd for a while?
[22-Jun-2009 09:45:37] <mrayzenoss> clustermagnet: yeah, most of our customers are upgrading from Core, they'll migrate you to it
[22-Jun-2009 09:45:49] <clustermagnet> mrayzenoss: Core has plugins for linux/sun/winblows machines i am sure...
[22-Jun-2009 09:45:54] <mrayzenoss> yeah
[22-Jun-2009 09:46:00] <clustermagnet> mrayzenoss: perfect
[22-Jun-2009 09:46:17] <clustermagnet> mrayzenoss: and zenpacks...
[22-Jun-2009 09:46:28] <clustermagnet> is that 3rd party plugins submitted by people?
[22-Jun-2009 09:46:31] <mrayzenoss> yes
[22-Jun-2009 09:46:35] <slestak> clustermagnet: ive been using the foss vm for ~2 months, it is a good way to get started
[22-Jun-2009 09:46:35] <clustermagnet> is that available on your site
[22-Jun-2009 09:46:38] <dberger> slestak: around all day
[22-Jun-2009 09:46:45] <clustermagnet> slestak: thank you
[22-Jun-2009 09:46:50] <mrayzenoss> clustermagnet: http://www.zenoss.com/community/projects/zenpacks/
[22-Jun-2009 09:47:00] <clustermagnet> mrayzenoss: slestak thanks for all your help!
[22-Jun-2009 09:47:04] <clustermagnet> decision made
[22-Jun-2009 09:47:12] <clustermagnet> talk to you guys if i have questions
[22-Jun-2009 09:47:18] <mrayzenoss> cool
[22-Jun-2009 09:47:22] <dberger> slestak: i'm lurking around to round out my knowledge of zenoss, because i am pretty clueless so far
[22-Jun-2009 09:47:37] <slestak> i am as well.
[22-Jun-2009 09:47:48] <mrayzenoss> slestak: http://www.zenoss.com/community/docs/zenoss-dev-guide/2.4.0/ch02s03.html#technique_sending_an_event
[22-Jun-2009 09:49:35] <mrayzenoss> some people use events to heartbeat their batch processes
[22-Jun-2009 09:49:52] <mrayzenoss> trigger an event with a wait at the beginning and send in a clear at the end
[22-Jun-2009 09:49:59] <mrayzenoss> no clear event received, then alarm
[22-Jun-2009 09:50:25] <slestak> i guess my specific question is: I have to track if a job is "hung" and the way i thought i would do that is send the integer pid as a "level" or the point to track. If a thread stays at one "level"toolong, it is hung, and it is time to sound alarms.
[22-Jun-2009 09:50:38] <mrayzenoss> slestak: there is process monitoring
[22-Jun-2009 09:50:56] <mrayzenoss> oh, I see
[22-Jun-2009 09:51:36] <mrayzenoss> yeah, that works
[22-Jun-2009 09:51:40] <slestak> mrayzenoss: the thread will lanuch a new pid for each job it runs, so i need to check many pids
[22-Jun-2009 09:52:05] <slestak> is therea way to check how long a check has been in a state?
[22-Jun-2009 09:53:20] <slestak> and trigeer on say 60m or sth like that?
[22-Jun-2009 09:53:27] <mrayzenoss> well, you can do a check with events to not alarm until X# of times or unless it hasn't cleared for a certain amount of time
[22-Jun-2009 09:53:46] <mrayzenoss> so if you are exposing the state somewhere, it could be monitored
[22-Jun-2009 09:54:03] <mrayzenoss> or you could send Zenoss and event when it starts
[22-Jun-2009 09:54:04] <slestak> mrayzenoss: i was envisioning a class that reports this way, with subclasses for different time allotments.
[22-Jun-2009 09:54:18] <mrayzenoss> and Zenoss would sit on that event for say, 60 minutes and then alert
[22-Jun-2009 09:54:34] <mrayzenoss> unless the Event Clear came through and cleared it
[22-Jun-2009 09:54:49] <mrayzenoss> so when the job finishes, it would send an Event Clear
[22-Jun-2009 09:54:55] <slestak> 90% of my work may be in a 15m class (a zProperty) and move other jobs into the longer classes if they require it.
[22-Jun-2009 09:55:20] <slestak> i said pids, but inactuality, it is an integer jiob number from the scheduler, so the same report is the same number every executuion
[22-Jun-2009 09:55:33] <mrayzenoss> Check out Jane Curry's Event Management paper: http://www.skills-1st.co.uk/papers/jcurry.html
[22-Jun-2009 09:55:44] <mrayzenoss> there are some good ideas in there
[22-Jun-2009 09:56:10] <mrayzenoss> she also has a Process Monitoring paper, but it sounds like handling this through Events is the way to go
[22-Jun-2009 09:56:21] <mrayzenoss> what kind of job is it?
[22-Jun-2009 09:56:31] <mrayzenoss> if it's in Java, you could also use JMX
[22-Jun-2009 09:56:37] <slestak> lol, ive been given 2 hours to come up with sth, so i may go for the hearbeat 1st, and then firm it up as i go
[22-Jun-2009 09:56:56] <slestak> no, no java, its a language youv'e prob never seen
[22-Jun-2009 09:57:09] <slestak> well, youveseen it, but you cannot believe will use it.
[22-Jun-2009 09:57:33] <mrayzenoss> I've written Delphi to wrap Fortran in DLLs, I've seen things no one should see
[22-Jun-2009 09:57:56] <dberger> is it befunge?
[22-Jun-2009 09:58:15] <mrayzenoss> Befunge is awesome
[22-Jun-2009 09:58:45] <mrayzenoss> check this out: http://www.foo.be/docs/tpj/issues/vol4_3/tpj0403-0017.html
[22-Jun-2009 09:58:54] <slestak> our db uses a dialect of Dartmouth Basic, its Unidata, an IBM database
[22-Jun-2009 09:59:03] <mrayzenoss> ugh
[22-Jun-2009 09:59:18] <mrayzenoss> the Chris Howe entry for Befunge in the Obfuscated Perl Contest: http://www.foo.be/docs/tpj/issues/vol4_3/tpj0403-0017.html
[22-Jun-2009 09:59:43] <mrayzenoss> I worked with him briefly, we sat and stared at that for hours after he won
[22-Jun-2009 10:00:33] <mrayzenoss> slestak: good luck with that
[22-Jun-2009 10:02:00] <slestak> its all ok, we can get off the ledge, i have python on the aix box
[22-Jun-2009 10:02:59] <twm1010> Morning everyone.
[22-Jun-2009 10:03:06] <mrayzenoss> twm1010: morning
[22-Jun-2009 10:04:40] <twm1010> Hey Matt... how goes it?
[22-Jun-2009 10:05:11] <mrayzenoss> ok, just putting some polish on 2.4.2
[22-Jun-2009 10:05:30] <twm1010> Excellent...
[22-Jun-2009 10:05:49] <twm1010> We had a lot of fun here this weekend
[22-Jun-2009 10:06:03] <twm1010> Our datacenter UPS is out of commission thanks to some nasty weather
[22-Jun-2009 10:08:57] <mrayzenoss> did you lose anything else?
[22-Jun-2009 10:10:08] <twm1010> Yeah, an older Windows box is having some power issues, and oddly enough, a virtual machine that was shutdown clean has a fubar filesystem that won't diskcheck.
[22-Jun-2009 10:10:48] <twm1010> What do you expect from 9 year old hardware?
[22-Jun-2009 10:19:44] <clustermagnet> Guys, im trying to setup zenoss, but my mysql box is elsewhere
[22-Jun-2009 10:19:46] <clustermagnet> in the docs:
[22-Jun-2009 10:19:47] <clustermagnet> See the section titled "Using Zenoss with a Remote MySQL Instance" in Zenoss Administration.
[22-Jun-2009 10:19:59] <clustermagnet> however, i dont see this section in the zenoss administration pDF
[22-Jun-2009 10:20:10] <clustermagnet> how do you tell the installer to go to another mysql box
[22-Jun-2009 10:20:11] <twm1010> which install method are you using?
[22-Jun-2009 10:20:32] <twm1010> Zenoss comes prepackaged with a mysql instance, so, it will run that first if you're installing from stack i believe.
[22-Jun-2009 10:20:33] <clustermagnet> twm1010: uhm, .bin
[22-Jun-2009 10:20:41] <clustermagnet> twm1010: stack
[22-Jun-2009 10:20:49] <clustermagnet> twm1010: so if i bang from source, it wont do that?
[22-Jun-2009 10:21:23] <mrayzenoss> clustermagnet: You're gonna love this: http://dev.zenoss.org/trac/ticket/5105
[22-Jun-2009 10:22:04] * mrayzenoss sighs
[22-Jun-2009 10:22:12] <clustermagnet> mrayzenoss: trac rulez
[22-Jun-2009 10:22:17] <clustermagnet> mrayzenoss: thanks man!
[22-Jun-2009 10:22:22] <mrayzenoss> clustermagnet: do you really need an external MySQL?
[22-Jun-2009 10:22:43] <clustermagnet> mrayzenoss: absolutely not
[22-Jun-2009 10:23:12] <clustermagnet> mrayzenoss: in the current environment there is a very powerful mysql setup, which we use
[22-Jun-2009 10:23:28] <twm1010> I stumbled a few weeks back when i did the 2.4.1 upgrade and didn't realize zenoss-stack was listed as a service. So I'm sitting there banging my head wondering why it tells me MySQL isn't started...
[22-Jun-2009 10:23:57] <twm1010> >service zenoss-stack start did the trick...
[22-Jun-2009 10:25:50] <clustermagnet> mrayzenoss: however, those options dont let you point to another mysql server...
[22-Jun-2009 10:25:55] <clustermagnet> they just let you change u/p and port
[22-Jun-2009 10:25:58] <clustermagnet> on a local install
[22-Jun-2009 10:26:28] <clustermagnet> anyways, im sure its not a problem if compiling from source, ya?
[22-Jun-2009 10:26:48] <twm1010> I believe there is a config file when you're building from source that you can edit and change things as you please.
[22-Jun-2009 10:27:17] <twm1010> or... hold on
[22-Jun-2009 10:28:01] <twm1010> you could just let it install, then export that DB to the server you want to put it on, and then change the settings in event manager to point over there.
[22-Jun-2009 10:28:34] <twm1010> im no MySQL guru by any means but I don't see why that wouldn't work.
[22-Jun-2009 10:28:54] <mrayzenoss> yeah, what twm1010 said
[22-Jun-2009 10:29:07] <mrayzenoss> we definitely have people not running on the stock MySQL
[22-Jun-2009 10:29:25] <mrayzenoss> I'll ask around a bit
[22-Jun-2009 10:30:06] <rmatte> you could always do remote mysql but via nfs
[22-Jun-2009 10:31:27] <rmatte> rmatte: http://dev.zenoss.org/trac/ticket/5071 apparently it's no longer a blocker?
[22-Jun-2009 10:31:37] <rmatte> erm
[22-Jun-2009 10:31:43] <rmatte> meant to say matt:
[22-Jun-2009 10:32:19] <zenChild> Does anyone know how to chain relationships in Zenoss? I have my relationships working between the device and components, but what I wanted was a relationship between a component and subcomponents and that doesn't seem to be working. I posted something to the dev forum but haven't received any responses yet. You can reference the post here: http://forums.zenoss.com/viewtopic.php?t=10086
[22-Jun-2009 10:34:09] <clustermagnet> thanks guys!
[22-Jun-2009 10:34:24] <mrayzenoss> rmatte: So development looked into ZenPack loading for obvious dumb stuff and ran it with a profiler, but nothing popped up.
[22-Jun-2009 10:34:47] <clustermagnet> mrayzenoss: so, do you know of any zenpack to monitor netapp?
[22-Jun-2009 10:35:02] <clustermagnet> i see you have good vmware feedback
[22-Jun-2009 10:35:15] <clustermagnet> we'll need netapp, cisco, and vmware
[22-Jun-2009 10:35:20] <mrayzenoss> rmatte: we're switching to Zope 2.11 for the next release, which seems to be faster. It got lowered from a blocker because it's still working
[22-Jun-2009 10:35:29] <rmatte> k
[22-Jun-2009 10:36:15] <zenChild> clustermagnet: trac-zenpacks
[22-Jun-2009 10:36:16] <mrayzenoss> rmatte: so it's not off the radar, we just don't have any solid leads on improving it other than the Zope upgrade
[22-Jun-2009 10:36:58] <mrayzenoss> clustermagnet: there are 3 NetApp ZenPacks, this one attempts to roll them all into 1: trac-zenpacks/wiki/ZenPackscommunityNetApp
[22-Jun-2009 10:37:12] <clustermagnet> mrayzenoss: thanks
[22-Jun-2009 10:37:16] <clustermagnet> zenChild: thank you
[22-Jun-2009 10:37:30] <clustermagnet> I do see that some require zenoss v 2.2 2.1... etc
[22-Jun-2009 10:37:38] <mrayzenoss> that means "or later"
[22-Jun-2009 10:37:46] <clustermagnet> mrayzenoss: sweet
[22-Jun-2009 10:37:47] <clustermagnet> thanks!
[22-Jun-2009 10:37:55] <mrayzenoss> 2.4.1 is current, 2.4.2 is probably tomorrow
[22-Jun-2009 10:38:06] <clustermagnet> im excited like crazy
[22-Jun-2009 10:38:10] <clustermagnet> no more text files
[22-Jun-2009 10:39:06] <clustermagnet> gents, thank you so much for your immediate help!
[22-Jun-2009 10:39:10] * clustermagnet owes beer
[22-Jun-2009 10:39:49] <mrayzenoss> I try to travel to different community events, collecting beers. Next stop, Community Leadership Summit next month
[22-Jun-2009 10:43:22] <clustermagnet> mrayzenoss: heh
[22-Jun-2009 10:43:29] <clustermagnet> mrayzenoss: distributed monitoring
[22-Jun-2009 10:43:41] <rocket> jb: ping
[22-Jun-2009 10:43:44] <clustermagnet> if there are 2 datacenters, how do you suggest zenoss should be implemented ?
[22-Jun-2009 10:43:48] <mrayzenoss> clustermagnet: there are 2 VMware ZenPacks, 1 is for Filesystems, 1 is for a list of Guests
[22-Jun-2009 10:43:57] <mrayzenoss> clustermagnet: are they on the same network?
[22-Jun-2009 10:44:17] <mrayzenoss> err, reachable
[22-Jun-2009 10:44:32] <clustermagnet> mrayzenoss: there is a VPN going between the 2 networks
[22-Jun-2009 10:44:40] <clustermagnet> however, they are logically different
[22-Jun-2009 10:44:44] <clustermagnet> 1 is production colo
[22-Jun-2009 10:44:51] <clustermagnet> 2nd is corporate
[22-Jun-2009 10:45:04] <rmatte> you need to come up here to Ottawa, Canada for one of our events
[22-Jun-2009 10:45:05] <clustermagnet> it would be sweet to have 2 zenoss machines, 1 at each location
[22-Jun-2009 10:45:13] <rmatte> then I can buy you a beer and score a Zenny
[22-Jun-2009 10:45:20] <mrayzenoss> clustermagnet: that would probably be optimal
[22-Jun-2009 10:45:32] <mrayzenoss> rmatte: Ohio LinuxFest is as close as I'm getting this year
[22-Jun-2009 10:46:05] <mrayzenoss> it's a big event though, 2 days, 1000 or so people
[22-Jun-2009 10:46:11] <rmatte> hmmm, I'd need to get a passport to go to that, I do know someone in Ohio though who works in the industry
[22-Jun-2009 11:08:44] <twm1010> I need to take a look at that NetApp zenpack again...
[22-Jun-2009 11:24:24] <mrayzenoss> yeah, it needs some feedback. I'm close to asking the author to go ahead and publish it
[22-Jun-2009 11:45:59] <adr3nalin3> Could someone please refresh my memory on the setting that gets google maps to draw connectors on the map?
[22-Jun-2009 11:48:55] <twm1010> you're talking about where it should be drawing them and it is not?
[22-Jun-2009 11:50:34] <adr3nalin3> I thought there was a setting to tell it to draw the connectors. Is it supposed to by default?
[22-Jun-2009 11:51:41] <dberger> adr3nalin3: you might be asking about the zDrawMapLinks property
[22-Jun-2009 11:52:09] <dberger> adr3nalin3: it is a property of the network
[22-Jun-2009 11:52:30] <adr3nalin3> dberger: that is exactly what I was looking for. thank you
[22-Jun-2009 11:52:41] <dberger> adr3nalin3: np
[22-Jun-2009 11:55:31] <adr3nalin3> What property are the map links drawn from: Is it network description?
[22-Jun-2009 11:55:39] <adr3nalin3> or location address?
[22-Jun-2009 11:56:18] <adr3nalin3> nm I got it
[22-Jun-2009 11:56:30] <dberger> what is it?
[22-Jun-2009 11:57:03] <dberger> because all I understand is that zenoss automatically scans your network topology
[22-Jun-2009 11:57:22] <dberger> i tried to build one manually using a tutorial here: http://www.zenoss.com/Members/netdata/create-manual-device-dependencys/
[22-Jun-2009 11:57:44] <dberger> but I didn't really understand what I was doing or have much success
[22-Jun-2009 11:59:11] <adr3nalin3> dberger: http://www.zenoss.com/community/docs/zenoss-guide/2.1.1/ch07s05.html
[22-Jun-2009 11:59:44] <adr3nalin3> My problem is I have 1 network with different subnets so it treats them all as different networks
[22-Jun-2009 12:00:22] <adr3nalin3> well technically they are all seperate networks but I would like to see them all connected to each other
[22-Jun-2009 12:04:02] <dberger> well, you might be able to get it up and running by building the dependency manually, but I have no idea whether that's a good idea
[22-Jun-2009 12:05:25] <adr3nalin3> ill check it out thank you
[22-Jun-2009 12:10:15] <etank> do alerting rules work like firewall rules? the first to match is the one that is used.
[22-Jun-2009 12:11:16] <twm1010> No
[22-Jun-2009 12:11:25] <twm1010> Each alert rule is independent
[22-Jun-2009 12:12:24] <etank> hmmm ... ok
[22-Jun-2009 12:13:18] <twm1010> There might be a method to make them dependent on each other, but I don't think so.
[22-Jun-2009 12:15:57] <etank> looks like the rule was not enabled
[22-Jun-2009 12:16:09] <etank> i think i may have it right now
[22-Jun-2009 12:25:53] <perr0> anybody seen casey?
[22-Jun-2009 12:26:55] <perr0> caseyh20 to be exact
[22-Jun-2009 14:40:57] <dberger> anyone know of a good way to monitor multicast? if not something pre-integrated with zenoss, something to run standalone that could trigger events on router interfaces?
[22-Jun-2009 14:47:36] <jcallaway> hey does anyone know who's the best person to reach out to regading the nmap plugin?
[22-Jun-2009 14:49:12] <jcallaway> ...the paths are pointing to the wrong place
[22-Jun-2009 15:38:18] <zenChild> does anyone have an example on how to use Device.findChild() ?
[22-Jun-2009 15:47:29] <jcallaway> i wish i could help you out! i'd like to rack up some karma so i can get some help with my issue
[22-Jun-2009 15:49:33] <mrayzenoss> zenChild: the only reference I can find is in the 2.4.1 release notes
[22-Jun-2009 15:49:46] <mrayzenoss> jcallaway: what platform and installer was that path issue?
[22-Jun-2009 15:50:14] <jcallaway> hey thanks, the nmap plugin is hardcoded to refer to the binary path to nmap via
[22-Jun-2009 15:50:38] <jcallaway> zenpath . libexec . nmap
[22-Jun-2009 15:51:08] <jcallaway> and libexec was missing in the 2.4.1 debian package, and the vmware appliance
[22-Jun-2009 15:51:20] <jcallaway> had to make a symlink
[22-Jun-2009 15:51:52] <mrayzenoss> is nmap under "Run Commands"?
[22-Jun-2009 15:51:59] <zenChild> mrayzenoss: I think I worked my way around the findChild thing anyway by creating a dictionary in my modeler code.
[22-Jun-2009 15:52:17] <jcallaway> i don't believe so, i'd have to check
[22-Jun-2009 15:52:34] <jcallaway> it's a modeler plugin
[22-Jun-2009 15:56:08] <mrayzenoss> jcallaway: which version are you running?
[22-Jun-2009 15:56:15] <jcallaway> 2.4.1
[22-Jun-2009 15:56:24] <mrayzenoss> hmmm... cause I came across this: http://dev.zenoss.org/trac/ticket/4680
[22-Jun-2009 15:56:36] <jcallaway> i'm trying to compile 2.4.x from svn so i can contribute the fix
[22-Jun-2009 15:56:42] <jcallaway> thanks, i'll check out the ticket
[22-Jun-2009 15:57:24] <jcallaway> okay, great! thanks, i'm not crazy : )
[22-Jun-2009 15:57:51] <mrayzenoss> yeah, sounds like nmap got left out
[22-Jun-2009 15:58:23] <mrayzenoss> feel free to open another ticket
[22-Jun-2009 15:58:30] <jcallaway> it would be nice to get it rolling since it can do os/application version detection if so configured
[22-Jun-2009 15:58:50] <jcallaway> ok, cool. i'll see what i can do and get as much info as possible before opening the ticket
[22-Jun-2009 15:59:22] <mrayzenoss> thanks, mark it as a "High" or "Blocker", since it sounds like missing functionality between installers
[22-Jun-2009 15:59:32] <jcallaway> gotcha
[23-Jun-2009 00:00:42] [disconnected at Tue Jun 23 00:00:42 2009]
[23-Jun-2009 00:00:42] [connected at Tue Jun 23 00:00:42 2009]
[23-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[23-Jun-2009 03:11:27] <Appiah_> All my daemons are running but i get heartbeat failure for each one
[23-Jun-2009 03:11:37] <Appiah_> this happend after upgrade from 2.3 to 2.4
[23-Jun-2009 03:11:57] <Appiah_> Had this problem on another server but I dont remember how we fixed this...
[23-Jun-2009 03:44:41] <baffle> Appiah_: Event manager -> Clear All Heartsbeats (from dropdown)
[23-Jun-2009 03:46:39] <Appiah_> but...
[23-Jun-2009 03:46:46] <Appiah_> it just comes back baffle
[23-Jun-2009 03:48:20] <baffle> Does anyone here have an event transform to support multiplatform per filesystem treshold? Preferrable as a "custom" setting. The "Tip of the month" seems to derive the treshold from the Windows FS-label, wich seems a bit.. Strange..
[23-Jun-2009 03:50:36] <baffle> Appiah_: Hmm, but are the daemons actually running?
[23-Jun-2009 03:50:53] <baffle> Appiah_: And have you checked the logs? Can they contact ZenHub?
[23-Jun-2009 03:51:45] <Appiah_> they all show up green
[23-Jun-2009 03:51:47] <Appiah_> everything works fine
[23-Jun-2009 03:57:22] <baffle> Appiah_: Strange.
[23-Jun-2009 07:01:05] ckrough_ is now known as ckrough
[23-Jun-2009 08:14:43] <klinstifen> morning all
[23-Jun-2009 08:15:08] <klinstifen> i got a quick question about datacollectors and the OS section on a device's status tab
[23-Jun-2009 08:15:16] <klinstifen> anybody have experience in that area?
[23-Jun-2009 08:15:33] <venturaville> a little
[23-Jun-2009 08:22:34] <klinstifen> ok cool
[23-Jun-2009 08:22:50] <klinstifen> so it seems you can match OIDs to that section
[23-Jun-2009 08:22:58] <klinstifen> err the values of OIDs
[23-Jun-2009 08:24:03] <klinstifen> and that's done by doing something like snmpgetmap = getmap({ oid : 'setHWProductKey' })
[23-Jun-2009 08:24:34] <klinstifen> do you know where the list of "setXYZ" attributes can be found?
[23-Jun-2009 08:24:57] <klinstifen> in that OS section there is a Tag # and Serial #
[23-Jun-2009 08:25:05] <klinstifen> i can get those values via SNMP
[23-Jun-2009 08:25:10] <klinstifen> so I want to map them there
[23-Jun-2009 08:25:19] <klinstifen> but I have no idea what those attributes are called
[23-Jun-2009 08:30:11] <rocket> klinstifen: they are attributes of the object
[23-Jun-2009 08:30:43] <rocket> klinstifen: you can either look at the source code in ZenModel
[23-Jun-2009 08:31:07] <rocket> or use zendmd to find the object in the zopedb and then do a dir(object) on it to get its methods
[23-Jun-2009 08:31:42] <rocket> eg from zendmd
[23-Jun-2009 08:31:45] <rocket> d = devices.findDevice('192.168.31.3'
[23-Jun-2009 08:31:47] <rocket> )
[23-Jun-2009 08:31:53] <rocket> dir(d)
[23-Jun-2009 08:32:42] <venturaville> setHWSerialNumber
[23-Jun-2009 08:32:55] <venturaville> setHWTag
[23-Jun-2009 08:33:21] <venturaville> grep -nr setHW on the Product directory helps here.........
[23-Jun-2009 08:33:31] <venturaville> Products
[23-Jun-2009 08:42:07] <klinstifen> ventur, grep -nr setHW on $ZENHOME/Products?
[23-Jun-2009 08:42:50] <klinstifen> rocket, thanks. is there any way to dump the output of dir(d) to a file for parsing?
[23-Jun-2009 08:42:51] <venturaville> yep
[23-Jun-2009 08:42:58] <rocket> klinstifen: yup .. actually you might just want to go to $ZENHOME/Products/ZenModel
[23-Jun-2009 08:43:20] <rocket> klinstifen: sure zendmd is just a python shell
[23-Jun-2009 08:43:31] <venturaville> rocket: DataCollector directory has some interesting references too
[23-Jun-2009 08:43:59] <rocket> venturaville: DataCollector populates teh objects defined in ZenModel
[23-Jun-2009 08:44:07] <venturaville> right......
[23-Jun-2009 08:44:09] <rocket> venturaville: but that might be easier to interpret
[23-Jun-2009 08:44:28] <rocket> venturaville: though DataCollector isnt always as complete as ZenModel
[23-Jun-2009 08:44:53] <venturaville> nope, but does provide some nice examples
[23-Jun-2009 08:44:56] <venturaville> *shrug*
[23-Jun-2009 08:45:03] <rocket> yes I cant argue with that
[23-Jun-2009 08:48:51] <klinstifen> cool I will play around with that
[23-Jun-2009 08:48:53] <klinstifen> thanks
[23-Jun-2009 08:51:34] <klinstifen> ok so another question and maybe this is what's leading me astray
[23-Jun-2009 08:51:43] <klinstifen> looking at the DellDeviceMap.py
[23-Jun-2009 08:51:56] <klinstifen> in $ZENHOME/Products/DataCollector/plugins/zenoss/snmp
[23-Jun-2009 08:52:23] <klinstifen> they use setHWProductKey
[23-Jun-2009 08:52:45] <klinstifen> and that seems to map to the HW Model on the Status tab
[23-Jun-2009 08:52:52] <klinstifen> not an exact naming match but close enough
[23-Jun-2009 08:53:28] <klinstifen> but...
[23-Jun-2009 08:54:33] <klinstifen> err..nevermind
[23-Jun-2009 08:54:40] <klinstifen> was going to say another field isn't mapped
[23-Jun-2009 08:54:46] <klinstifen> but then i noticed that they have it commented out
[23-Jun-2009 08:55:05] <venturaville> I think HWProductName is the one in the status tab
[23-Jun-2009 08:56:29] <venturaville> not 100% certain of that though
[23-Jun-2009 08:57:39] <klinstifen> yeah the serial number i dont think matches up either
[23-Jun-2009 08:57:51] <klinstifen> the oid they reference is really the tag...
[23-Jun-2009 08:58:03] <klinstifen> but it doesn't get populated either way
[23-Jun-2009 08:58:40] <rocket> klinstifen: take a look at the aix zenpack .. it sets serial numbers etc
[23-Jun-2009 08:59:00] <klinstifen> ok
[23-Jun-2009 08:59:12] <klinstifen> after you modify the datacollector code
[23-Jun-2009 08:59:19] <rocket> its in the modelers folder in that zenpack .. I know it works as I wrote it
[23-Jun-2009 08:59:28] <klinstifen> hehe
[23-Jun-2009 09:00:00] <rocket> it also set the setHWProductKey properly so the manufacturer link works etc ..
[23-Jun-2009 09:03:13] <klinstifen> do the modelers have preference over each other?
[23-Jun-2009 09:03:23] <klinstifen> i made changes to the dell modeler to grab the tag
[23-Jun-2009 09:03:29] <klinstifen> but it still isn't populating the status tab
[23-Jun-2009 09:03:39] <klinstifen> i wonder if it is being overwritten by another modeler
[23-Jun-2009 09:04:16] <rocket> klinstifen: they do .. there is order significance .. and or is the modeller enabled?
[23-Jun-2009 09:04:42] <klinstifen> top down or bottom up?
[23-Jun-2009 09:04:47] <klinstifen> which modeler runs last?
[23-Jun-2009 09:04:51] <rocket> eg doing a "zenmodel run -v 10 -d device" on the command line with show you a lot ..
[23-Jun-2009 09:04:54] <rocket> bottom last
[23-Jun-2009 09:04:58] <klinstifen> hmm
[23-Jun-2009 09:05:02] <klinstifen> my dell modeler is last
[23-Jun-2009 09:05:08] <klinstifen> i believe the modeler is enabled
[23-Jun-2009 09:05:23] <rocket> restart zenoss after updating the code?
[23-Jun-2009 09:05:26] <klinstifen> i added it to the collector plugins
[23-Jun-2009 09:05:29] <klinstifen> yeppers
[23-Jun-2009 09:05:30] <klinstifen> did that
[23-Jun-2009 09:10:50] <klinstifen> ah this could be why
[23-Jun-2009 09:10:56] <klinstifen> the modeler isn't running
[23-Jun-2009 09:11:06] <klinstifen> err the collector isn't run
[23-Jun-2009 09:11:09] <klinstifen> when i model the device
[23-Jun-2009 09:11:27] <rocket> check the logs .. most likely a bug in the code has disabled that modeller
[23-Jun-2009 09:11:34] <klinstifen> k
[23-Jun-2009 09:14:20] <klinstifen> hmm doesn't look like it
[23-Jun-2009 09:14:44] <klinstifen> i'll tinker a bit more
[23-Jun-2009 09:18:09] <rocket> klinstifen: look at adding log.debug lines to your modeller etc
[23-Jun-2009 09:18:24] <rocket> there are examples of that in the aix zenpack
[23-Jun-2009 09:18:40] <rocket> so when you do zenmodeller run -v 10 those lines show up
[23-Jun-2009 09:25:50] <klinstifen> rocket: ok, i'll give that a go
[23-Jun-2009 09:33:43] <venturaville> here is an odd one
[23-Jun-2009 09:34:13] <venturaville> barring us having any sort of 'meta-devices', is there a way to have a relationship exist that you can populate outside of a modeler
[23-Jun-2009 09:34:30] <venturaville> i.e. I have something that needs to 'push' data externally into a relationship
[23-Jun-2009 09:35:08] <venturaville> so I would define a _relations in the class for the container
[23-Jun-2009 09:35:15] <venturaville> but then not have an attached modeler?
[23-Jun-2009 09:44:09] <zenChild> venturaville: couldn't you just get a handle to the object and manipulate the relationship from there?
[23-Jun-2009 09:54:06] <rocket> venturaville: as long as you can get the object in python I dont know why you wouldnt be able to update it without a modeller
[23-Jun-2009 10:17:25] <eidolon> hey folks - anyone know if there's an easy way to customize zenoss alert messages?
[23-Jun-2009 10:17:47] <eidolon> ala, we'd like to say "HTTP alert on such and such a host" "Please escalate to Bob in engineering"
[23-Jun-2009 10:18:50] <ckrough> eidolon: the "message" tab in the alerting rule in question
[23-Jun-2009 10:19:58] <eidolon> ORLY. *looks*
[23-Jun-2009 10:20:11] <cain22> will zenoss map locations if there isn't any devives directly under that group?
[23-Jun-2009 10:20:37] <rmatte> cain22: Go to that location and click on the map tab
[23-Jun-2009 10:20:42] <rmatte> if it shows up then the answer is yes
[23-Jun-2009 10:21:02] <cain22> hmm...im trying to figure out why my map isn't drawing the network lines
[23-Jun-2009 10:21:28] <rmatte> did you go to Networks, and enable drawing map lines for the networks listed there?
[23-Jun-2009 10:21:50] <cain22> i believe yes, but i'll double check
[23-Jun-2009 10:21:54] <rmatte> if so, then Zenoss just isn't understanding your network topology
[23-Jun-2009 10:21:55] <klinstifen> zDrawMapLinks should be set to True
[23-Jun-2009 10:22:09] <rmatte> if you have sites going out over MPLS circuits or something Zenoss probably won't understand
[23-Jun-2009 10:22:18] <cain22> draw map links set to true
[23-Jun-2009 10:22:26] <rmatte> or any sites that aren't directly linked (I.E are going out over the internet)
[23-Jun-2009 10:22:58] <cain22> the weird thing is, at one point most of them awhile back were linked...
[23-Jun-2009 10:23:08] <rmatte> I have one client where all the lines were drawn fine and another client where no lines were drawn, it's just because Zenoss can't figure out what is directly linked since it isn't actually directly linked
[23-Jun-2009 10:23:27] <rmatte> it generally does better with connections like frame relay or LAN extensions
[23-Jun-2009 10:24:09] <cain22> whats "prefer snmp mapping..?
[23-Jun-2009 10:24:20] <cain22> setting for networks
[23-Jun-2009 10:24:39] <rmatte> prefer mapping via information discovered with SNMP versus info discovered with pings/traceroutes/portscans
[23-Jun-2009 10:24:40] <cain22> er, snmp naming
[23-Jun-2009 10:24:41] <cain22> nm
[23-Jun-2009 10:24:52] <rmatte> oh
[23-Jun-2009 10:24:56] <rmatte> where do you see that?
[23-Jun-2009 10:25:31] <cain22> under network properties
[23-Jun-2009 10:25:46] <rmatte> hmmm, no idea, google it
[23-Jun-2009 10:27:32] <ckrough> wonder if it has to do with path outages, choosing whether to built dependcy trees from traceroutes or snmp returned routes
[23-Jun-2009 10:27:43] <ckrough> definetely let us know what you find
[23-Jun-2009 10:28:46] <venturaville> rocket: that is what I assumed, but the documentation isn't very clear on it
[23-Jun-2009 10:29:39] <rocket> venturaville: thats one of those .. if you have programmed with it long enough you can figure out interesting ways to do the same thing
[23-Jun-2009 10:31:19] <venturaville> yeah... I'm still trying to get my head around containers/modelers... I guess for now I am just going to put _relations straight in the container class and then try and manipulate it in zendmd
[23-Jun-2009 10:31:27] <venturaville> that might give me a better feel for it
[23-Jun-2009 10:34:20] <ckrough> venturaville: the other day we were talking accessing URL variables within python in a TALEs expression. solution: "foo python: here.REQUEST.test;" will put value of test into foo, which you can use in further python in the template
[23-Jun-2009 10:35:16] <venturaville> ckrough: good to know. I need to file that away for further use........ :-)
[23-Jun-2009 10:38:33] <ckrough> venturaville: I ended up needing some sanity tests on the value, so I moved it out to a python script, but that's the syntax that work within the tempate
[23-Jun-2009 10:40:18] <ckrough> venturaville: the other day we were talking accessing URL variables within python in a TALEs expression. solution: "foo python: here.REQUEST.test;" will put value of test into foo, which you can use in further python in the template"
[23-Jun-2009 10:40:30] <ckrough> woops, too much scrollback. bad post
[23-Jun-2009 10:40:31] <klinstifen> other than zenmodeler run -v10 is there another way to test my modeler to see where its bombing out or why it isn't inserting the found data?
[23-Jun-2009 10:41:26] <ckrough> klinstifen: v10 doesnt show any issues, but the devices doesnt show up or get updated?
[23-Jun-2009 10:42:56] <klinstifen> v10 shows it is running the collector...but the device doesnt get updated
[23-Jun-2009 10:43:30] <klinstifen> i found the initial problem...the collector wasn't being run because the builtin condition wasn't true
[23-Jun-2009 10:43:41] <klinstifen> i commented that out and now the collector runs when i model the device
[23-Jun-2009 10:53:52] <eidolon> *sigh* i love messages from zenoss liket his
[23-Jun-2009 10:53:54] <eidolon> "Disk space low: 1601.6% used (-9068.24 GB free)"
[23-Jun-2009 10:53:58] <ckrough> zenmodeler log shows the collector pulling data from the device?
[23-Jun-2009 10:54:42] * ckrough thinks that disk space calculation may be inaccurate
[23-Jun-2009 10:56:24] <klinstifen> no
[23-Jun-2009 10:58:42] <ckrough> klinstifen: what type of device and is it ssh or snmp collection
[23-Jun-2009 11:07:49] <klinstifen> dell server and it is snmp collection
[23-Jun-2009 11:08:04] <klinstifen> i am working on the delldevicemap collector that is built in
[23-Jun-2009 11:34:56] <rmatte> I need to remove a zenpack without removing it's contents, anyone know how?
[23-Jun-2009 11:35:17] <rmatte> specifically, I want to remove the Cisco Mibs Pack from the ZenPack inventory but leave the Mibs behind
[23-Jun-2009 11:35:28] <rmatte> (I'm basically just seeing if it resolves load time issues)
[23-Jun-2009 11:39:18] <mrayzenoss> not sure you can do that, don't see anything like that in the zenpack command
[23-Jun-2009 11:39:42] <mrayzenoss> do you want all the MIBs or just a few? Maybe you could re-export the ones you want to another ZenPack?
[23-Jun-2009 11:39:49] <rmatte> I mean, I could go in to the zenpack and delete all of the items then remove the zenpack, only problem is that the zenpack has over 72000 items in it
[23-Jun-2009 11:40:09] <rmatte> nah, I want all the Mibs, the Cisco Mibs set has tons of dependencies it's like a web
[23-Jun-2009 11:40:17] <rmatte> most Mibs won't function without certain other Mibs
[23-Jun-2009 11:40:25] <mrayzenoss> ugh
[23-Jun-2009 11:41:09] <rmatte> plus we're an MSP so we monitor tons of different devices, we need to be prepared for anything
[23-Jun-2009 11:43:55] <rmatte> can zendmd manipulate zenpacks?
[23-Jun-2009 11:44:41] <mrayzenoss> sure
[23-Jun-2009 11:47:05] <rmatte> there must be some way to have it purge the contents of a zenpack without actually removing them from Zenoss then?
[23-Jun-2009 11:48:06] <rmatte> If removing the contents from the Zenpack fixes the response time issues, I'm thinking that there may be some way for me to have my zenpack run a script to do the same thing once it's done installing
[23-Jun-2009 11:48:42] <rmatte> or just provide the script as a workaround
[23-Jun-2009 11:51:58] <blocky> can I rewrite reportmail to send multiple reports in 1 email
[23-Jun-2009 11:57:57] <rmatte> is Chet around, or someone who would be able to figure out the zendmd command to accomplish that?
[23-Jun-2009 12:04:06] <klinstifen> ok i finally go the collector to work
[23-Jun-2009 12:04:13] <klinstifen> it's pulling the information i want
[23-Jun-2009 12:04:37] <klinstifen> just one more piece of the puzzel to finish
[23-Jun-2009 12:04:50] <klinstifen> setOSProduct...is that the same thing as OS Make
[23-Jun-2009 12:04:51] <klinstifen> ?
[23-Jun-2009 12:05:15] <klinstifen> or asking it another way: once I am in zendmd
[23-Jun-2009 12:05:24] <klinstifen> i do a d = devices.findDevice('ip')
[23-Jun-2009 12:05:27] <klinstifen> and then a dir(d)
[23-Jun-2009 12:05:48] <klinstifen> how can i output the value of a particular attribute listed by dir(d)?
[23-Jun-2009 12:06:03] <klinstifen> like i want to echo d.setOSProduct
[23-Jun-2009 12:06:09] <rmatte> print
[23-Jun-2009 12:06:29] <rmatte> instead of echo
[23-Jun-2009 12:06:41] <klinstifen> hmm
[23-Jun-2009 12:06:43] <klinstifen> it says
[23-Jun-2009 12:07:03] <klinstifen> <bound method Device.setOSProduct of <Device at 'path to device'>
[23-Jun-2009 12:07:04] <rmatte> Product would be the model
[23-Jun-2009 12:07:50] <rmatte> yeh, I guess you won't get much from that
[23-Jun-2009 12:07:58] <rmatte> are you sure that setOSProduct is even what you want?
[23-Jun-2009 12:08:23] <klinstifen> well...it's right next to setOSProductKey and none of the neighboring values seem applicable
[23-Jun-2009 12:08:23] <rmatte> try getOSProduct
[23-Jun-2009 12:08:31] <klinstifen> and there isn't anything else that looks right
[23-Jun-2009 12:09:05] <klinstifen> err....d.getOSProduct?
[23-Jun-2009 12:09:18] <rmatte> getHWManufacturerName
[23-Jun-2009 12:09:18] <rmatte> getHWProductName
[23-Jun-2009 12:09:22] <rmatte> are the two that I use
[23-Jun-2009 12:09:45] <klinstifen> d.getHWProductName
[23-Jun-2009 12:09:50] <klinstifen> gives me that same "bound" error
[23-Jun-2009 12:10:01] <rmatte> it's not an error
[23-Jun-2009 12:10:20] <rmatte> it's the content of that variable, but only zenoss can really interpret it properly
[23-Jun-2009 12:10:25] <klinstifen> gotcha
[23-Jun-2009 12:11:17] <rmatte> what if you type just d.getHWManufacturerName ?
[23-Jun-2009 12:11:20] <rmatte> instead of print
[23-Jun-2009 12:11:30] <rmatte> that should give you some output
[23-Jun-2009 12:12:10] <klinstifen> same...
[23-Jun-2009 12:12:15] <rmatte> hmmm
[23-Jun-2009 12:12:24] <rmatte> no idea then
[23-Jun-2009 12:12:59] <klinstifen> I would think that if there was a getHWManufacturer
[23-Jun-2009 12:13:05] <klinstifen> then there should be a setHWManufacturer
[23-Jun-2009 12:13:07] <klinstifen> no?
[23-Jun-2009 12:13:14] <klinstifen> it doesn't appear in dir(d)...
[23-Jun-2009 12:13:57] <rocket> setHWProduct
[23-Jun-2009 12:14:51] <rmatte> rocket, maybe you would know, is there any way to manipulate the contents of a zenpack with zendmd?
[23-Jun-2009 12:15:01] <rmatte> I have a huge zenpack and I want to empty out the contents of it
[23-Jun-2009 12:15:09] <rmatte> without actually removing them from Zenoss
[23-Jun-2009 12:15:10] <ckrough> klinstifen: thoe are callable methods
[23-Jun-2009 12:15:18] <ckrough> klinstifen: dev.getHWProductName()
[23-Jun-2009 12:15:24] <ckrough> try with ()
[23-Jun-2009 12:15:31] <rmatte> oh right, forgot about the ()
[23-Jun-2009 12:15:41] <rmatte> that should give you proper output
[23-Jun-2009 12:15:55] <klinstifen> ah ha!
[23-Jun-2009 12:16:40] <rocket> rmatte: hrmm i am not sure .. what are you trying to empty out? data out of the objects.xml?
[23-Jun-2009 12:16:54] <rmatte> yeh, here's the scenario...
[23-Jun-2009 12:17:42] <rmatte> after you install my Cisco Mibs Zenpack, the zProperties pages and other pages take ridiculous amounts of time to load, I think it's because Zenoss is trying to index the objects.xml file almost every time it tries to load one of those pages...
[23-Jun-2009 12:18:10] <rmatte> and the index.xml file in this zenpack is like 34MB in size (or more, maybe more like 64)
[23-Jun-2009 12:18:19] <rocket> rmatte: hrmm well it really is loading the objects.xml file into the zeo database ...
[23-Jun-2009 12:18:25] <rmatte> I want to empty out the zenpack without actually removing the Mibs
[23-Jun-2009 12:18:52] <rocket> rmatte: you could just move the objects.xml file out of the way but I dont think it will do anything
[23-Jun-2009 12:18:54] <rmatte> when I first installed the Mibs by hand I don't remember having any kind of response problem
[23-Jun-2009 12:19:03] <rmatte> as soon as I installed them as a zenpack the problems started
[23-Jun-2009 12:19:15] <rmatte> so I want to simulate installing them by hand, but without going through all that hassle again
[23-Jun-2009 12:19:21] <rmatte> installing the zenpack is much much easier
[23-Jun-2009 12:19:23] <rocket> rmatte: have you tried to reindex from the command line?
[23-Jun-2009 12:19:52] <rmatte> I've reindexed on boxes that have this pack installed and it's done nothing to improve performance
[23-Jun-2009 12:20:14] <rocket> rmatte: if i would guess I would say that objects installed from the zenpacks arent getting indexed at all in the zeo database
[23-Jun-2009 12:20:27] <rocket> but I havent really had time to track that theory down
[23-Jun-2009 12:20:58] <rmatte> my theory is that emptying out this zenpack will resolve the issue and leave the Mibs behind
[23-Jun-2009 12:21:10] <rmatte> but there are over 72000 items in the zenpack, so doing it by hand would take all week
[23-Jun-2009 12:21:15] <rmatte> I need a quicker way to do it
[23-Jun-2009 12:21:20] <blocky> is there a way I can add a link to another report in a custom device report
[23-Jun-2009 12:21:56] <blocky> just a little bit of html so the email reportmail sends can have a note or two in it
[23-Jun-2009 12:22:29] <rocket> rmatte: so you are thinking just get rid of the objects.xml right?
[23-Jun-2009 12:22:38] <rmatte> well, emptying it out
[23-Jun-2009 12:22:47] <rmatte> but I don't think full out deleting the file will help
[23-Jun-2009 12:23:04] <rmatte> I think the objects need to be removed properly via zenoss
[23-Jun-2009 12:23:23] <rocket> rmatte: if you remove the zenpack does the issue go away?
[23-Jun-2009 12:23:33] <rmatte> yeh, but so do the Mibs
[23-Jun-2009 12:23:53] <rmatte> I essentially want to install the Mibs, then find some way of removing the Zenpack while leaving the Mibs behind
[23-Jun-2009 12:23:57] <rocket> rmatte: have you tried moving the mibs to a new organizer?
[23-Jun-2009 12:24:00] <rmatte> that may it doesn't try to constantly reindex that zenpack
[23-Jun-2009 12:24:08] <rmatte> that way*
[23-Jun-2009 12:24:23] <rocket> and then see if the mibs are still referenced by the zenpack?
[23-Jun-2009 12:24:32] <rmatte> like rename it from cisco to cisco2, remove the zenpack and name it back?
[23-Jun-2009 12:24:36] <rmatte> I'll try it
[23-Jun-2009 12:24:41] <rocket> if not referenced delete zenpack?
[23-Jun-2009 12:24:48] <rocket> yea something like that
[23-Jun-2009 12:24:54] <rocket> make cisco2 organizer
[23-Jun-2009 12:25:07] <rocket> move mibs from cisco to cisco2
[23-Jun-2009 12:25:08] <rmatte> oh, I can't rename it, gah, it'll take like an hour to copy all those Mibs to a new container
[23-Jun-2009 12:25:30] <rocket> do this then make a new organizer
[23-Jun-2009 12:25:32] <rmatte> I'm looking for something simple
[23-Jun-2009 12:25:34] <rocket> cisco2
[23-Jun-2009 12:25:51] <rocket> click on showall for the mib table
[23-Jun-2009 12:26:09] <rocket> then click on All fro the select up by the drop down arrow
[23-Jun-2009 12:26:49] <rocket> nevermind .. just a sec
[23-Jun-2009 12:27:25] <rocket> rmatte: with the cisco mibs can you click on cisco?
[23-Jun-2009 12:27:37] <rocket> err nevermind
[23-Jun-2009 12:27:41] <rocket> gah
[23-Jun-2009 12:28:54] <klinstifen> ckrough: so getOSManufacturerName() returns the value I see on my device's status tab
[23-Jun-2009 12:28:58] <klinstifen> is there a way to set that value
[23-Jun-2009 12:28:59] <rocket> rmatte: anyway if you can figure out something like that .. maybe from the zendmd
[23-Jun-2009 12:29:11] <klinstifen> i dont see a corresponding setOSManufacturerName
[23-Jun-2009 12:33:26] <ckrough> klinstifen: print '\n'.join(dir(d))
[23-Jun-2009 12:33:36] <ckrough> that'll at least make the list a little more readable
[23-Jun-2009 12:34:20] <klinstifen> still don't see it
[23-Jun-2009 12:34:32] <klinstifen> that field can be manually set via the edit tab in the interface
[23-Jun-2009 12:34:44] <klinstifen> but i have a couple dell servers so it would be useful if i could just have the modeler do it
[23-Jun-2009 12:35:14] <ckrough> snmp can interface with Dells openmanage stuff to return that type of info
[23-Jun-2009 12:35:17] <ckrough> I think
[23-Jun-2009 12:35:20] <rocket> klinstifen: might need a dir(d.os) or a dir(d.hw)
[23-Jun-2009 12:36:14] <rmatte> rocket: you can't do a show all on this pack, trust me
[23-Jun-2009 12:36:19] <rmatte> 72000 items
[23-Jun-2009 12:36:23] <rmatte> the page won't even load that long
[23-Jun-2009 12:37:59] <rocket> klinstifen: did you ever look at the aix zenpack?
[23-Jun-2009 12:38:14] <rocket> klinstifen: it sets the manufacturer to IBM for the os
[23-Jun-2009 12:38:31] <rmatte> rocket: that's certainly the goal, using zendmd to clear the zenpack's contents
[23-Jun-2009 12:38:34] <rmatte> just wondering how
[23-Jun-2009 12:39:12] <rocket> klinstifen: om.setOSProductKey=MultiArgs(om.setOSProductKey,"IBM")
[23-Jun-2009 12:39:19] <rocket> from Products.DataCollector.plugins.DataMaps import MultiArgs
[23-Jun-2009 12:39:30] <rocket> where IBM is the manufacturer
[23-Jun-2009 12:39:45] <rocket> klinstifen: I am assuming things are showing up as Unknown for you
[23-Jun-2009 12:42:46] <ionutz> hi
[23-Jun-2009 12:42:53] <rmatte> hi
[23-Jun-2009 12:43:13] <ionutz> I remove some rights to a device from zope interface
[23-Jun-2009 12:43:32] <ionutz> is there any way to reset the permissions to default or remove this device ?
[23-Jun-2009 12:44:02] <ionutz> now devicelist is broken and when I'm trying to access that device I'm redirected to the login page
[23-Jun-2009 12:44:23] <rmatte> you should be able to login with the admin account and just delete the device
[23-Jun-2009 12:44:36] <rmatte> (from the zope management interface)
[23-Jun-2009 12:44:55] <rmatte> though there will be traces of the device elsewhere, so it's a bit tricky
[23-Jun-2009 12:44:59] <ionutz> I'm already logged in as admin
[23-Jun-2009 12:45:06] <ionutz> zport/dmd/manage
[23-Jun-2009 12:45:08] <rmatte> you really need to be careful with permissions when you're messing around in there
[23-Jun-2009 12:45:09] <ionutz> I'm as admin
[23-Jun-2009 12:45:34] <ionutz> and on Device section when I click on that device it redirects me to the login page
[23-Jun-2009 12:45:50] <ionutz> is there other ways to remove this device from zope database ?
[23-Jun-2009 12:45:55] <rmatte> ok, but can you not just put a checkmark next to that device and click Delete?
[23-Jun-2009 12:46:34] <ionutz> I've alrady tried this and I got this error
[23-Jun-2009 12:46:55] <ionutz> Type: BadRequest
[23-Jun-2009 12:46:55] <ionutz> Value: BFIDB01 does not exist
[23-Jun-2009 12:46:55] <ionutz> Traceback (innermost last):
[23-Jun-2009 12:46:55] <ionutz> * Module ZPublisher.Publish, line 114, in publish
[23-Jun-2009 12:46:55] <ionutz> * Module ZPublisher.mapply, line 88, in mapply
[23-Jun-2009 12:46:56] <ionutz> * Module ZPublisher.Publish, line 40, in call_object
[23-Jun-2009 12:46:58] <ionutz> * Module OFS.ObjectManager, line 492, in manage_delObjects
[23-Jun-2009 12:47:00] <ionutz> BadRequest: BFIDB01 does not exist
[23-Jun-2009 12:47:24] <rmatte> is that when you try to remove it from the zope management interface?
[23-Jun-2009 12:47:30] <rmatte> or when you try to remove it from Zenoss?
[23-Jun-2009 12:47:52] <ionutz> from zope management interface
[23-Jun-2009 12:48:02] <ionutz> device section
[23-Jun-2009 12:48:07] <rmatte> well, then I'd say you're pretty stuck
[23-Jun-2009 12:48:31] <rmatte> you should always backup your zope database before trying changes like that
[23-Jun-2009 12:49:31] <ionutz> I have a backup
[23-Jun-2009 12:49:38] <ionutz> I'll try to restore this
[23-Jun-2009 12:49:45] <klinstifen> rocket: yes they are showing up as "unknown"
[23-Jun-2009 12:49:47] <rmatte> how did you do the backup though?
[23-Jun-2009 12:51:08] <rocket> klinstifen: your best example is the aix zenpack and the lines I just showed you
[23-Jun-2009 12:51:25] <ionutz> rmatte from zenoss interface
[23-Jun-2009 12:52:06] <rmatte> I was talking about doing: "cp $ZENHOME/var/Data.fs to $ZENHOME/var/Data.fs.bak
[23-Jun-2009 12:52:23] <rmatte> then if you want to restore the zope DB it's just the opposite
[23-Jun-2009 12:52:48] <rmatte> cp $ZENHOME/var/Data.fs.bak $ZENHOME/var/Data.fs
[23-Jun-2009 12:52:51] <ionutz> oh ... I've made a backup for the whole zenoss
[23-Jun-2009 12:53:02] <ionutz> I'll have a look on this
[23-Jun-2009 12:53:04] <ionutz> thanks
[23-Jun-2009 12:53:09] <rmatte> np
[23-Jun-2009 12:54:01] <ionutz> rmatte there is only a Data.fs.old file from 2 days ago
[23-Jun-2009 12:54:14] <rmatte> too old?
[23-Jun-2009 12:54:20] <ionutz> should I restart zope after the restore
[23-Jun-2009 12:54:26] <rmatte> yes
[23-Jun-2009 12:54:32] <rmatte> zopectl restart
[23-Jun-2009 12:54:48] <ionutz> I think it's fine b/c I didn't made too many changes during these days
[23-Jun-2009 12:54:55] <rmatte> good
[23-Jun-2009 13:01:25] <ionutz> It's still not working ... I suppose that the change was made before this backup ... I'll try to find when the change was made
[23-Jun-2009 13:01:29] <ionutz> thanks for your help
[23-Jun-2009 13:01:38] <rmatte> no problem
[23-Jun-2009 13:04:44] adr3nalin3 is now known as afink
[23-Jun-2009 13:12:01] <zenChild> Anyone know if there is a good article anywhere on RelationshipMap and ObjectMap? I need to create some custom maps in a modeler for nested relations.
[23-Jun-2009 13:24:19] <Sparks> Can anyone tell me what license is used for Zenoss?
[23-Jun-2009 13:24:37] <chudler> GPL2, I think it is
[23-Jun-2009 13:24:48] <mrayzenoss> correct, GPLv2
[23-Jun-2009 13:26:26] <rocket> Hey Matt
[23-Jun-2009 13:26:50] <Sparks> Coolll..... Who do I talk to about possibly packaging it for Fedora? I see the Tray Applet is packaged already.
[23-Jun-2009 13:28:28] <mrayzenoss> Sparks: yeah, I've talked to Dave Nalley about it before
[23-Jun-2009 13:28:34] <Sparks> What did he say?
[23-Jun-2009 13:28:46] <mrayzenoss> well, there were objections to the use of Python 2.4 and Zope 2.8
[23-Jun-2009 13:28:58] <mrayzenoss> but the Zope issue will be fixed in the next release
[23-Jun-2009 13:29:09] <mrayzenoss> and Python upgraded to 2.6 the release after that
[23-Jun-2009 13:29:36] <Sparks> Okay. I'll talk to David and see if he wants to work it.
[23-Jun-2009 13:29:39] <mrayzenoss> we definitely want to get into the distros
[23-Jun-2009 13:30:47] <mrayzenoss> Sparks: email me at mray@zenoss.com and I'll keep in touch as we make progress
[23-Jun-2009 13:30:58] <Sparks> mrayzenoss: Will do
[23-Jun-2009 13:53:12] <venturaville> ok I am a bit lost... given a list a = ['devicename1','devicename2'] ... how do I get it to give me back the equivalent objects in tales ?
[23-Jun-2009 13:58:00] <zenChild> you have to use a "repeat" command. I am far from an expert on TALES, but I just played with it a bit the other day. This guide was a good reference: http://www.owlfish.com/software/simpleTAL/tal-guide.html
[23-Jun-2009 13:58:54] <zenChild> The fruit basket example on that page is a good example of how to traverse a list in TALES.
[23-Jun-2009 15:34:16] <rmatte> hey Matt, is there anyone around there who you could ask about using zendmd to delete the contents of a Zenpack?
[23-Jun-2009 15:34:33] <rmatte> I've been deleting by hand but I've only got maybe 2000 out of 72000 done lol
[23-Jun-2009 15:38:43] <mrayzenoss> I'll poke around a bit
[23-Jun-2009 15:42:06] <rmatte> thanks
[23-Jun-2009 15:42:28] <rmatte> if we can get something to work it might be a first step towards correcting that issue
[23-Jun-2009 15:47:48] <zenChild> How do you get access to the device object from within a CommandParser? I can get the device name via cmd.deviceConfig.device, but how do I fetch the instance?
[23-Jun-2009 15:52:58] <rmatte> zenChild: no idea, I don't have any experience with command parsers
[23-Jun-2009 15:55:33] <zenChild> k. I'll keep digging. Thanks anyway.
[23-Jun-2009 15:56:17] <rmatte> post on the zenoss forums, you might have some luck there
[23-Jun-2009 15:56:44] <zenChild> will do.
[23-Jun-2009 16:00:12] <rmatte> ok, so playing with the objects.xml doesn't work, so I obviously need to use zendmd to get in to the Zope database and remove the Mib items from the Zenpack inventory without actually removing the Mibs themselves
[23-Jun-2009 16:01:23] <tyler79> let's say that i an event. that event relates to some device or service that is managed by a 3rd party. i want to provide contact information along with the event in the alert. what's the best way to do that?
[23-Jun-2009 16:02:14] <rmatte> you'd have to add it in using a python transform for the whole /Event class
[23-Jun-2009 16:03:08] <tyler79> rmatte: that's what i thought. should i be storing the contact info somewhere in the zopedb or mysql though?
[23-Jun-2009 16:03:14] <rmatte> basically saying something like: if evt.device = 'devicename' evt.summary = '%s contact info here' % (evt.summary) or something like that
[23-Jun-2009 16:03:41] <rmatte> you could do that, but that'd be fairly complex
[23-Jun-2009 16:03:49] <rmatte> are you doing this for just 1 device or for all of them?
[23-Jun-2009 16:03:59] <tyler79> rmatte: so a plain text file would be the easiest thing to do?
[23-Jun-2009 16:04:17] <rmatte> if you're only doing it for one device you'd just include the contact info right in the transform
[23-Jun-2009 16:04:28] <rmatte> if it's for multiple devices then you'd need to use zope or something
[23-Jun-2009 16:04:28] <tyler79> rmatte: for many different devices and more complex exvents that don't necessarily relate to a single device
[23-Jun-2009 16:04:59] <rmatte> we have zenoss integrated with OTRS here and the integration script that we wrote takes care of populating info such as that from a seperate database before a ticket is created in OTRS
[23-Jun-2009 16:05:52] <tyler79> rmatte: ok, thanks. that's helpful
[23-Jun-2009 16:17:13] <zenChild> rocket: I read from one of the IRC logs that you were doing some work with multilevel relations... did you ever have any luck figuring this out? The log I am referring to is here: http://www.zenoss.com/community/discuss/irc/archives/dev-chat-04-30-2009
[23-Jun-2009 16:25:50] <mrayzenoss> New Jabber/XMPP Bot ZenPack: http://www.zenoss.com/community/projects/zenpacks/xmppbot
[23-Jun-2009 16:29:04] <rmatte> that is so cool
[23-Jun-2009 16:31:54] <mrayzenoss> rmatte: the only suggestion from devs was to move it to another ZenPack
[23-Jun-2009 16:32:03] <mrayzenoss> rmatte: which you said you were already doing
[23-Jun-2009 16:32:14] <rmatte> move it to another zenpack?
[23-Jun-2009 16:32:24] <rmatte> I'm actually just deleting the contents of the existing zenpack by hand
[23-Jun-2009 16:32:26] <mrayzenoss> yeah, that was Kells' suggstion
[23-Jun-2009 16:32:32] <rmatte> I converted it to development mode then started deleting them
[23-Jun-2009 16:32:36] <mrayzenoss> sounds like the same difference
[23-Jun-2009 16:32:43] <rmatte> but it's taking forever since I can't do all of them at once
[23-Jun-2009 16:33:05] <rmatte> I'm basically just looking for the function that Zenoss is calling when it's deleting the items
[23-Jun-2009 16:33:15] <rmatte> then maybe I can harness that function and make a script to do it quicker
[23-Jun-2009 16:34:04] <rmatte> deleting 72000 objects 40 at a time by hand with a delay of like 10-30 seconds per deletion is going to take me a whole month to delete all the objects lol
[23-Jun-2009 16:36:09] <mrayzenoss> rmatte: go ahead and post it on zenoss-dev, Kells said he'd look at it
[23-Jun-2009 16:36:16] <rmatte> k cool
[23-Jun-2009 16:36:28] <rmatte> thanks
[23-Jun-2009 16:36:35] <mrayzenoss> rmatte: or you could open a support case
[23-Jun-2009 16:36:51] <rmatte> doesn't that cost money?
[23-Jun-2009 16:38:19] <mrayzenoss> apparently I had rmatte confused with someone else, zenoss-dev it is!
[23-Jun-2009 16:38:39] <rmatte> hehe
[23-Jun-2009 16:46:38] <rmatte> http://forums.zenoss.com/viewtopic.php?p=36242#36242
[23-Jun-2009 16:46:39] <rmatte> voila
[23-Jun-2009 16:47:06] <rmatte> if you could pass that around it'd be appreciated, if this works I'm sure I won't be the only one happy about the workaround
[23-Jun-2009 17:01:42] <rmatte> did you ever ask about Zenoss not check port's actual operational status except during modelling?
[23-Jun-2009 22:13:42] <Brad_> I am having problems with the zenoss install and was hoping for help
[23-Jun-2009 22:15:50] <perr0> ...
[23-Jun-2009 22:19:34] <Brad_> My install is hanging at the setting up zenoss-stack phase
[23-Jun-2009 22:19:55] <Brad_> dpkg: error processing zenoss-stack (--configure):
[23-Jun-2009 22:20:15] <Brad_> when i hit cntrl c i to stop the process i get the message above
[23-Jun-2009 22:20:49] <perr0> what OS and what pkg are you using to install zenoss
[23-Jun-2009 22:21:26] <Brad_> ubuntu 8.04 (64bit) and zenoss 2.4.1...today i tried 2.4.2
[23-Jun-2009 22:21:56] <Brad_> i am using the apt-get method, but i also tried using dpkg -i
[23-Jun-2009 22:22:00] <Brad_> same results
[23-Jun-2009 22:22:17] <perr0> mm ya I would only use the pkg from the zenoss site
[23-Jun-2009 22:22:30] <perr0> I don't use ubuntu sorry
[23-Jun-2009 22:24:48] <perr0> night Brad, try tomorrow is lot of traffic during the day
[23-Jun-2009 22:24:57] <Brad_> take it easy
[23-Jun-2009 22:39:24] <Brad_> anyone know why a zenoss install would hang after the step Setting up zenoss-stack (2.4.1-0) ...
[24-Jun-2009 00:00:42] [disconnected at Wed Jun 24 00:00:42 2009]
[24-Jun-2009 00:00:42] [connected at Wed Jun 24 00:00:42 2009]
[24-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[24-Jun-2009 07:04:04] ckrough_ is now known as ckrough
[24-Jun-2009 08:49:38] <rocket> zenChild: ping
[24-Jun-2009 08:56:40] <ckrough> destination unreachable
[24-Jun-2009 09:00:52] <zenChild> yeah rocket?
[24-Jun-2009 09:01:20] <zenChild> ckrough: had to reroute the transmission
[24-Jun-2009 09:53:13] <rocket> zenChild: you were looking for me?
[24-Jun-2009 09:54:12] <zenChild> Yeah. I had a question for you yesterday. I saw from a irc log that you had been working on multilevel relationships with what looked like LVM. I was just wondering if you had any luck in doing so.
[24-Jun-2009 09:54:51] <rocket> zenChild: yeah its in the aix snmp zenpack
[24-Jun-2009 09:55:13] <zenChild> sweet. I'll look there then. Thanks.
[24-Jun-2009 09:55:31] <rocket> zenChild: it isnt necessarily very straight forward ..
[24-Jun-2009 09:57:18] <zenChild> rocket: I'll look it over and if you don't mind I might have a couple questions for you. I'm trying to set up a multilevel relationship like so: host(1toM ) -> diskpool ->(1toM) -> logical disks
[24-Jun-2009 09:58:12] <rocket> sure thats doable ...
[24-Jun-2009 09:58:18] <zenChild> thanks.
[24-Jun-2009 09:58:27] <rocket> zenChild: you are going to have to override some stuff from the very beginning
[24-Jun-2009 09:58:41] <rocket> eg starting with the Device object
[24-Jun-2009 09:59:07] <zenChild> eek.... not looking forward to that.
[24-Jun-2009 09:59:30] <zenChild> I'll take a look at it this afternoon. I gotta run to a meeting now. Thanks for the tips though.
[24-Jun-2009 09:59:30] <rocket> zenChild: need to do so to get a custom os or hw subobject ..
[24-Jun-2009 11:04:40] <zenChild> rocket: AIXVolumeGroupMap.py is exactly what I was looking for. It also answers my question as to whether I could split up the modelers or not. Thanks again for sending me in the right direction. I have to admit my eyes are bleeding a bit from the first run-through of the code, but I think after a few more times my understanding will hopefully improve. Cheers!
[24-Jun-2009 11:13:47] <rocket> zenChild: no problem
[24-Jun-2009 11:13:55] <rocket> zenChild: I had to figure it out blind
[24-Jun-2009 11:15:34] <zenChild> I'm impressed. I think I would have started drinking rather heavy by the time I figured it out on my own.
[24-Jun-2009 11:16:59] <rocket> zenChild: as you can see the trick is to have the modeler call special functions that reference the new subobjects
[24-Jun-2009 11:17:17] <rocket> basically build a bunch of nested dicts to go as deep as you need
[24-Jun-2009 11:18:57] <zenChild> rocket: It actually looks quite elegant compared to the hash madness that I had started in my modeler code. Hopefully I can get that part working without too much more effort. I think I know how my maps are supposed to look. Now I just need to get the code to do my bidding.
[24-Jun-2009 11:20:15] <rocket> zenChild: great ... and feel free to put in a good word for the zenpack contest It cant hurt
[24-Jun-2009 11:20:54] <zenChild> will do. I already posted kudos on the forum post I had open on the issue.
[24-Jun-2009 12:01:31] <rmatte> rocket: http://forums.zenoss.com/viewtopic.php?t=10154 <--- Any idea how to injects objects in to the zope database without actually installing as a zenpack? I essentially want to load the Mibs in without having it register as a Zenpack, that should resolve the performance issues until they can actually fix it properly.
[24-Jun-2009 12:21:56] <rocket> rmatte: could you do a workaround by scripting zenmib to load your mibs in the right order?
[24-Jun-2009 12:23:29] <rocket> rmatte: eg .. drop mibs that go in first into a folder .. run zenmib against it .. remove those files .. put next ones in .. run zenmib etc
[24-Jun-2009 12:23:56] <rocket> rmatte: zenmib will load the mibs into zope without doing zenpack stuff etc and its command line scriptable ..
[24-Jun-2009 12:26:25] <rmatte> I guess I could, it'd be a pain though, the cisco Mibs don't load very cleanly to begin with
[24-Jun-2009 12:26:31] <rmatte> it was a battle just to get them to load
[24-Jun-2009 12:26:39] <rmatte> the only sure-fire way to always have them load is the zenpack
[24-Jun-2009 12:26:51] <rmatte> which I apparently can't use due to a bug in Zenoss
[24-Jun-2009 12:29:58] <rmatte> I've had people follow the same procedure that I did with the same Mibs and not had them all load properly
[24-Jun-2009 12:31:17] <rocket> rmatte: and you are certain its a bug in the zenpack code?
[24-Jun-2009 12:33:15] <rmatte> it's not really a bug in the zenpack code...
[24-Jun-2009 12:33:58] <rmatte> when you try to load zProperties, or the ZenPacks page (as well as a couple others) Zenoss decides that it's going to reindex all ZenPacks, and you have to wait while it does this before the page will load
[24-Jun-2009 12:34:09] <rmatte> so with my ZenPack being huge, it takes upwards of 30 minutes
[24-Jun-2009 12:34:53] <rmatte> I need to find a way to inject the Mibs in to Zope so that they can be used without Zenoss deciding to reindex everything
[24-Jun-2009 12:35:37] <rmatte> That way the Mibs will be (theoretically) usable but won't show up as part of a zenpack
[24-Jun-2009 12:36:04] <rmatte> as I recall though, performance was much much better on my initial install of Zenoss where I originally installed the Mibs
[24-Jun-2009 12:36:15] <rmatte> on the second install with the Zenpack performance dropped drastically
[24-Jun-2009 12:39:03] <rocket> rmatte: I see .. so it sounds like zope is trying to index on a view when it should really be indexing on a change etc
[24-Jun-2009 12:40:37] <rmatte> exactly
[24-Jun-2009 12:40:54] <rmatte> by the way, nice work on that Linux monitoring patch, it got merged
[24-Jun-2009 12:41:08] <rocket> so in reality indexes are borked .. I messed with them a bit but dont fully understand them
[24-Jun-2009 12:41:30] <rmatte> although I think they merged the old patch, not the new one
[24-Jun-2009 12:41:32] <rocket> thats one area of code that probably is in need of a decent rewrite ..
[24-Jun-2009 12:41:44] <rmatte> yeh
[24-Jun-2009 12:42:10] <rmatte> there's no reason for it to be reindexing a zenpack with over 72000 items in it every time someone wants to access a zProperties page
[24-Jun-2009 12:42:13] <rocket> rmatte: my extended monitoring linux zenpack includes teh patch as well
[24-Jun-2009 12:42:28] <rmatte> ah, cool
[24-Jun-2009 12:42:36] <rmatte> did that get published?
[24-Jun-2009 12:42:38] <rocket> eg it will apply it if its not already done
[24-Jun-2009 12:42:41] <rocket> yes
[24-Jun-2009 12:42:44] <rmatte> nice
[24-Jun-2009 12:42:48] <rmatte> u da man
[24-Jun-2009 12:43:05] <rocket> that fixes network perf graphin and routes etc
[24-Jun-2009 12:43:49] <rocket> so if it was me I would just install both
[24-Jun-2009 12:44:43] <rmatte> hehe
[24-Jun-2009 12:45:20] <rmatte> one thing I'd love to see a patch for would be http://dev.zenoss.com/trac/ticket/5070
[24-Jun-2009 12:45:52] <rmatte> they said they won't fix it, but quite honestly, passing usernames and passwords directly is the most efficient way, I tried using python scripts to do it and it caused major performance hits
[24-Jun-2009 12:46:05] <rmatte> maybe my code isn't as good as it could be
[24-Jun-2009 12:48:32] <rocket> rmatte: what was the reason it was turned into a wont fix? I dont see that in the history of the ticket
[24-Jun-2009 12:48:52] <rmatte> "See community maintainer for a fix"
[24-Jun-2009 12:48:55] <rmatte> errr
[24-Jun-2009 12:49:00] <rmatte> See ZenPack maintained for a fix
[24-Jun-2009 12:49:01] <rmatte> rather
[24-Jun-2009 12:49:22] <rocket> so they are taking the approach its a zenpack issue ...
[24-Jun-2009 12:49:34] <rmatte> yeh, they are being lazy in my opinion
[24-Jun-2009 12:49:54] <rocket> I can see some of that I suppose .. but probably it should have been filtered before getting into the zenpack
[24-Jun-2009 12:49:56] <rmatte> they really should take security more seriously, passwords stored in Zenoss should be encrypted and sensored
[24-Jun-2009 12:50:33] <rmatte> instead of being stored in plaintext and dumped in plaintext
[24-Jun-2009 12:51:07] <chudler> would encrypting them help security?
[24-Jun-2009 12:51:46] <rmatte> chudler: yup
[24-Jun-2009 12:52:10] <rmatte> chudler: I can open up the zope DB file in a text editor, search for "password" and find all the passwords stored in it
[24-Jun-2009 12:52:11] <chudler> against?
[24-Jun-2009 12:52:28] <rmatte> pretty lax
[24-Jun-2009 12:52:30] <chudler> doesnt the encryption key have to be stored? you're only trading several secrets for one
[24-Jun-2009 12:53:13] <rocket> chudler: its one of those issues where if its displayed for a support desk person to see it then it is a security issue
[24-Jun-2009 12:53:16] <chudler> also, it seems the problem would be reading the zopeDB. can the attacker get that file?
[24-Jun-2009 12:53:22] <rmatte> so your point is that security through obscurity is better than at least attempting to make it more difficult for someone to grab sensitive information?
[24-Jun-2009 12:53:26] <rocket> eg the support desk person doesnt have full access to the system
[24-Jun-2009 12:53:29] <chudler> I agree that displaying the password is a bad thing, though
[24-Jun-2009 12:54:43] <rmatte> the whole database should require a username and password to be accessed ideally, and any password variables being passed to commands should be sensored
[24-Jun-2009 12:55:00] <rmatte> and the whole db should be encrypted, not just the passwords
[24-Jun-2009 12:55:24] <rmatte> large corporations are using this product
[24-Jun-2009 12:55:26] <chudler> it is just a file afterall... you can encrypt it
[24-Jun-2009 12:55:30] <rocket> rmatte: you might be going a bit to far with that last statement .. it might make performance royally suck ..
[24-Jun-2009 12:55:55] <rmatte> yeh, I can see performance sucking with an encrypted db, but encrypting the passwords would certainly help
[24-Jun-2009 12:56:16] <rmatte> sensoring them when commands fail is the main problem
[24-Jun-2009 12:56:29] <rocket> eg i have lots of big systems here that arent fully encrypted etc .. and if they were they would crush the hardware we can afford
[24-Jun-2009 12:56:35] <rmatte> "command failed: blah <username> <password>" isn't very secure to have kicking around
[24-Jun-2009 12:57:23] <rmatte> that's really what I logged the problem for in the first place was the command failed events that get generated and display the info in plaintext
[24-Jun-2009 12:57:54] <rocket> rmatte: can you do a transform yourself of that type of event?
[24-Jun-2009 12:58:11] <rocket> rmatte: and maybe post a howto for the zenpacks that are broken that way?
[24-Jun-2009 12:58:34] <rmatte> how do you determine which part is the command and which part is the password?
[24-Jun-2009 12:58:49] <rmatte> it's just a string of text, the sensoring needs to be done as the event is being generated
[24-Jun-2009 12:58:51] <rmatte> not afterwards
[24-Jun-2009 12:59:09] <rmatte> because ideally you still want to know what command failed
[24-Jun-2009 13:00:49] <mloven> @rmatte: couldn't you search the message for the zWinPassword (or whatever field it is for that device), and if it matches it changes the string to "blah", or something? If it doesn't find that string, it just prints the message as is?
[24-Jun-2009 13:12:14] <rmatte> mloven: I suppose you could, it'd be interesting
[24-Jun-2009 13:12:24] <rmatte> mloven: my python is not exactly prime yet
[24-Jun-2009 13:12:39] <rmatte> especially when it comes to string manipulation
[24-Jun-2009 13:16:54] <mloven> @rmatte: oh, my python isn't so good either, I just know that within the custom device reports, I can put in the zWinuser or zWinPassword fields and get the string. I assume it would be the same for zCommandPassword
[24-Jun-2009 13:21:30] <rmatte> zWinPassword is actually what most of the problematic zenpacks are using
[24-Jun-2009 13:21:35] <rmatte> since they are WMI performance packs
[24-Jun-2009 13:40:57] <rmatte> I still think it'd be more effective to have the sensoring done by Zenoss itself
[24-Jun-2009 13:41:23] <rmatte> that way you don't have to fool around with transforms or with making scripts access zope when they don't need to
[24-Jun-2009 13:42:03] <rmatte> but I'm pretty sure that'll never get done
[24-Jun-2009 13:47:40] <rocket> rmatte: its just a matter of building the appropriate api for zenpacks .. at the moment the api is fairly basic
[24-Jun-2009 13:52:01] <rmatte> too bad the google maps portlet doesn't support satelite view
[24-Jun-2009 13:52:05] <rmatte> would look awesome
[24-Jun-2009 13:52:06] <rmatte> hehe
[24-Jun-2009 13:52:37] <rmatte> rocket: agreed
[24-Jun-2009 13:53:14] <mrayzenoss> http://forums.zenoss.com/viewtopic.php?t=4231&highlight=satellite
[24-Jun-2009 13:53:39] <mrayzenoss> feel free to open an enhancement ticket to expose that in the UI (or better yet, submit a patch)
[24-Jun-2009 13:54:14] <venturaville> how do I get a zproperty to attach to a device class and not to the zenpack itself (ala ___init__.py)
[24-Jun-2009 13:55:05] <ckrough> you guys use the goole maps porlet for real, or just for fun?
[24-Jun-2009 13:58:39] <rmatte> interesting, that google maps thing started out as a core zenpack?
[24-Jun-2009 13:58:52] <rmatte> I use the google maps portlet for real
[24-Jun-2009 13:59:12] <rmatte> example: http://dmon.org/graphics/zenoss.png
[24-Jun-2009 13:59:14] <mrayzenoss> the Google maps predate ZenPacks
[24-Jun-2009 13:59:40] <mrayzenoss> I take that back, .egg ZenPacks
[24-Jun-2009 13:59:42] <rmatte> ah, cause I was reading http://morgana.sourceforge.net/
[24-Jun-2009 14:00:04] <rmatte> and yeh, it was a zip pack
[24-Jun-2009 14:01:38] <rmatte> by the way, two colour blind people at my work made the comment that the status icons for iterface status are hard to see, they suggested moving it back to colored UP/DOWN cells for accessiblity, just food for thought
[24-Jun-2009 14:02:21] <mrayzenoss> yeah, there's a ticket for that that I keep trying to get fixed
[24-Jun-2009 14:02:33] <mrayzenoss> one of our devs is color-blind
[24-Jun-2009 14:02:48] <mrayzenoss> not sure why he hasn't fixed it...
[24-Jun-2009 14:02:57] <rmatte> it's actually more common than most people think
[24-Jun-2009 14:03:13] <rmatte> especially in the computer industry surprisingly
[24-Jun-2009 14:03:25] <rmatte> I know 5 color blind people in the industry
[24-Jun-2009 14:15:50] <zenChild> Can someone explain to me exactly what Products.DataCollector.plugins.DataMaps.MultiArgs does? I think I have an idea, but it's a partly-cloudy idea.
[24-Jun-2009 14:22:10] <rmatte> it's key in determining device dependencies/relationships
[24-Jun-2009 14:22:36] <rmatte> that's about all I know about it
[24-Jun-2009 14:31:26] <rmatte> I like the comic on the front page now, very artistic
[24-Jun-2009 14:31:28] <rmatte> hehe
[24-Jun-2009 14:32:02] <mrayzenoss> venturaville: I can't find an example of a ZenPack-defined zProperty that is restricted to a single Device class
[24-Jun-2009 14:32:57] <venturaville> mrayzenoss: argh ... I have it made global for now in the absence of that
[24-Jun-2009 14:33:04] <rmatte> you could dig in and see how it's done with the zWinUser property
[24-Jun-2009 14:33:14] <venturaville> rmatte: will do
[24-Jun-2009 14:33:59] <rmatte> I actually have the developers IRC sessions as a recurring meeting in my calendar now hehe
[24-Jun-2009 14:35:44] <rmatte> by the way, why does the RSS feed only show titles? If you look at something like slashdot's it has an actual summary of each article in it. for the zenoss one you have to actually go to the site to read the summary
[24-Jun-2009 14:37:29] <mrayzenoss> which RSS feed?
[24-Jun-2009 14:37:35] <rmatte> the zenoss one
[24-Jun-2009 14:38:36] <mrayzenoss> for the blog?
[24-Jun-2009 14:38:44] <rmatte> I assume so
[24-Jun-2009 14:38:54] <rmatte> title is Zenoss - All The News
[24-Jun-2009 14:44:48] <rmatte> http://feeds.feedburner.com/zenoss/MbQB
[24-Jun-2009 14:46:08] <mrayzenoss> hmmm... that appears to be the feed generated by Zope for the "Events"
[24-Jun-2009 14:46:39] <mrayzenoss> well, the quick answer is we probably won't fix it, since it'll be gone in the next month or so
[24-Jun-2009 14:46:42] * mrayzenoss crosses fingers
[24-Jun-2009 14:47:17] <rmatte> hehe
[24-Jun-2009 14:48:45] <mrayzenoss> of course, I say the next month or so but everything takes longer than expected
[24-Jun-2009 14:49:02] <mrayzenoss> oops
[24-Jun-2009 14:52:10] <rmatte> hehe
[24-Jun-2009 14:52:37] <rmatte> anyways, I'm done my shift so time to have beer and play video games w00t
[24-Jun-2009 14:52:45] <rmatte> (worked from home today so no long bus ride home either)
[24-Jun-2009 14:58:08] <mrayzenoss> zWinUser is a zProperty on all Devices, only used by /Server/Windows
[24-Jun-2009 14:58:25] <rmatte> ah
[24-Jun-2009 14:58:30] <rmatte> maybe there is no way to specify
[24-Jun-2009 14:58:42] <rmatte> maybe it just relies on people's common sense
[24-Jun-2009 14:58:43] <rmatte> hehe
[24-Jun-2009 14:58:58] <venturaville> it is mostly just an annoyance ... no huge deal to me
[24-Jun-2009 15:00:40] <rmatte> as long as it's there I could personally care less
[24-Jun-2009 15:00:49] <rmatte> just name it something obvious
[24-Jun-2009 15:00:58] <rmatte> like zWinUser is obviously not applicable to a Linux box
[24-Jun-2009 15:01:45] <mrayzenoss> well, at one point I had "zZenPackCommunityVMwareESXlibexec" as a zProperty
[24-Jun-2009 15:01:55] <mrayzenoss> found the fix for that
[24-Jun-2009 15:02:13] <mrayzenoss> Chet has spoken
[24-Jun-2009 15:02:23] <mrayzenoss> No, zProperties are top-down.
[24-Jun-2009 15:02:23] <mrayzenoss> They must exist at / to exist anywhere.
[24-Jun-2009 15:03:58] <suprsonic> ./libxml/encoding.h:29:19: error: iconv.h: No such file or directory FreeBSD 7.2
[24-Jun-2009 15:04:05] <suprsonic> thoughts?
[24-Jun-2009 15:06:30] <mrayzenoss> suprsonic: Casey Hillman's FreeBSD howto is down, give me your email address and I'll send you a copy
[24-Jun-2009 15:36:33] <rhetts> is there a way to disable compression w/ zenbackup?
[24-Jun-2009 16:12:46] <tyler79> any pynetsnmp users here? i figured out how to do a get on an oid using a SnmpSession object, but I can't figure out how to walk the snmp tree
[24-Jun-2009 16:13:08] <tyler79> actually, what I'm really looking for is some good pynetsnmp documentation (if it exists)
[24-Jun-2009 16:29:26] <mrayzenoss> tyler79: you may have better luck tomorrow when there's a developer in the channel
[24-Jun-2009 16:30:25] <tyler79> mrayzenoss: ok, thanks
[24-Jun-2009 17:10:13] <tripitaka> hello. My device was down, but now it has recovered and I can ping it from the zenoss host. Zenoss is still showing it as down, is there any way I can force a new check?
[24-Jun-2009 17:10:28] <tripitaka> I've tried remoddeling it but no joy
[25-Jun-2009 00:00:42] [disconnected at Thu Jun 25 00:00:42 2009]
[25-Jun-2009 00:00:42] [connected at Thu Jun 25 00:00:42 2009]
[25-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[25-Jun-2009 09:06:00] <Appiah_> how do you remove mibs on zenoss 2.4.1 ?
[25-Jun-2009 09:34:01] <venturaville> for those developing zenpacks in svn ... useful code snippet:
[25-Jun-2009 09:34:03] <venturaville> #!/bin/sh
[25-Jun-2009 09:34:03] <venturaville> ZENPACKNAME="ZenPacks.community.NetApp"
[25-Jun-2009 09:34:03] <venturaville> PYVERSION="py2.4"
[25-Jun-2009 09:34:03] <venturaville> ZENPACKVERSION=`grep -i ^Version EGG-INFO/PKG-INFO | awk ' { print $2 } '`
[25-Jun-2009 09:34:03] <venturaville> echo "Building ZenPack: ${ZENPACKNAME}"
[25-Jun-2009 09:34:04] <venturaville> zip -r ${ZENPACKNAME}-${ZENPACKVERSION}-${PYVERSION}.egg EGG-INFO ZenPacks -x \*/.svn\*
[25-Jun-2009 09:34:06] <venturaville> echo "Built: ZenPack EGG file: ${ZENPACKNAME}-${ZENPACKVERSION}-${PYVERSION}.egg"
[25-Jun-2009 10:00:02] <bwest> hello
[25-Jun-2009 10:01:03] <mrayzenoss> Greetings all, mrchippy will be joining me shortly
[25-Jun-2009 10:01:29] <bwest> Quick question Is there a way to monitor the Billiant Service on windows machine?
[25-Jun-2009 10:02:16] <bwest> This service essentially send queries to Ms SQL to determine the number of placed orders, orders in queue and succesful orders
[25-Jun-2009 10:02:53] <mrayzenoss> bwest: I've got a MS SQL ZenPack I'm hoping to post later today
[25-Jun-2009 10:03:03] <mrayzenoss> or you can monitor the service
[25-Jun-2009 10:03:04] <bwest> Will that do the trick?
[25-Jun-2009 10:04:17] <atrawog> Hi Mray! Have you got my email?
[25-Jun-2009 10:04:50] <bwest> mray: Will the new Ms SQL zenpack allow me define the queries I send?
[25-Jun-2009 10:05:13] <mrayzenoss> bwest: let me check
[25-Jun-2009 10:05:39] <mrayzenoss> atrawog: yeah, sorry I haven't responded
[25-Jun-2009 10:05:56] <mrayzenoss> atrawog: I'll get the Brocade update out after the 5 DB ones from Egor
[25-Jun-2009 10:05:59] <mrayzenoss> he's on a tear
[25-Jun-2009 10:06:47] <mrayzenoss> bwest: I gotta fire up a VM to install the MS SQL one, I'll get back to you
[25-Jun-2009 10:06:55] <mrayzenoss> bwest: the docs are somewhat limited
[25-Jun-2009 10:06:58] <bwest> mrchippy: I am monitoring servers in the SSH/Linux class and I want to pull memory stats off those servers. Can I use a plugin or do I have to creat a new zenpack, and parser?
[25-Jun-2009 10:07:17] <mrayzenoss> bwest: check out the LinuxMonitorAddOn ZenPack
[25-Jun-2009 10:07:48] <zenChild> Can somewhat tell me exactly what Products.DataCollector.plugins.DataMaps.MultiArgs does? How does the actual structure look that it is creating?
[25-Jun-2009 10:08:44] <bwest> mray: are you saying as a example because I thought that only provided information on interfaces?
[25-Jun-2009 10:09:07] <mrayzenoss> bwest: no, the LinuxMonitorAddon rolls up a patch to the LinuxMonitor to add memory
[25-Jun-2009 10:09:57] <klinstifen> bwest: if that doesnt' work you could create a zenpack
[25-Jun-2009 10:10:00] <mrayzenoss> there's a patch for the LinuxMonitor ZenPack that adds memory and interfaces that will be included in the next release, but the LinuxMonitorAddOn has that patch now
[25-Jun-2009 10:10:09] <klinstifen> i did that to get print queue information from our aix box
[25-Jun-2009 10:10:10] <mrchippy> zenChild: some of the map keys are methods that can take more than one arg (like setHwProductKey, iirc). you pass a multiarg object for that
[25-Jun-2009 10:10:12] <klinstifen> worked pretty well
[25-Jun-2009 10:10:24] <mrchippy> zenChild: looking for example
[25-Jun-2009 10:10:44] <bwest> mray: I see because I have installed that zenpack but when I go to the memory reports nothing shows up
[25-Jun-2009 10:12:03] <zenChild> I think my confusion was if it was actually a key or a value in the dictionary. I'm trying to wrap my head around how a key can be more than one value.
[25-Jun-2009 10:12:07] <klinstifen> i have a generic question about threshholds
[25-Jun-2009 10:12:19] <klinstifen> i created a custom graph definition
[25-Jun-2009 10:12:31] <klinstifen> and then tried to set a threshhold against one of the datapoints
[25-Jun-2009 10:12:36] <klinstifen> but it didn't seem to work
[25-Jun-2009 10:13:00] <klinstifen> do threshholds only work against datapoints added through the datasource interface?
[25-Jun-2009 10:13:33] <mrayzenoss> klinstifen: what were you trying to use the Threshold against?
[25-Jun-2009 10:13:51] <mrayzenoss> when you add the Threshold, you have to pick one of the available data points
[25-Jun-2009 10:13:53] <rmatte> bwest: did you actually install the LinuxMonitorAddOn ZenPack, or the patch for the standard one?
[25-Jun-2009 10:14:02] <rmatte> if you haven't done either of those things then you won't get the memory info
[25-Jun-2009 10:14:21] <bwest> the LinuxMonitorAddOn Zenpack
[25-Jun-2009 10:14:34] <mrayzenoss> bwest: have you remodeled?
[25-Jun-2009 10:14:44] <rmatte> yeh, you need to remodel afterwards
[25-Jun-2009 10:14:52] <bwest> yes a number of times
[25-Jun-2009 10:15:27] <rmatte> did you locally define the collector plugins at some point?
[25-Jun-2009 10:15:31] <mrchippy> zenchild: it's the value. check out the NewDeviceMap plugin in trunk/Products/DataCollector/plugins/zenoss/snmp. It maps the hwproductkey name and manufacturer using MultiArgs (lines ~80 - 90)
[25-Jun-2009 10:15:31] <bwest> rmatte: I need to get memavail, cached etc
[25-Jun-2009 10:15:35] <klinstifen> mray: i created some custom datapoints for an exchange zenpack i made...i wanted a threshold to go against the datapoints
[25-Jun-2009 10:15:57] <klinstifen> but i couldn't select it because it wasn't a datapoint that was available through the interface...since the datapoints were made through the custom definition
[25-Jun-2009 10:16:02] <bwest> rmatte: Locally define the collector plugins?
[25-Jun-2009 10:16:06] <rmatte> bwest: that would be something under the perf tab, no?
[25-Jun-2009 10:16:11] <venturaville> btw: devices.findDevice('blah.blah') does not work because of the dots in the name..........
[25-Jun-2009 10:16:24] <bwest> rmatte: yes
[25-Jun-2009 10:16:34] <bwest> but now all I get is CPU
[25-Jun-2009 10:16:46] <zenChild> mrchippy: k. I'll check it. Thanks for the tip.
[25-Jun-2009 10:17:00] <mrchippy> venturaville: checking that now...
[25-Jun-2009 10:17:03] <rmatte> sounds like the template didn't update
[25-Jun-2009 10:17:21] <rocket> bwest: as far as I know that information isnt gathered yet
[25-Jun-2009 10:17:36] <rmatte> rocket: ah, ok
[25-Jun-2009 10:17:40] <mrayzenoss> klinstifen: so you added the datasource and a data point and when you go to add a threshold the datasource_datapoint isn't there?
[25-Jun-2009 10:18:14] <atrawog> @mray There is no hurry with my Brodade ZenPack. I just wanted to check if a Mercurial repo is ok for you.
[25-Jun-2009 10:18:15] <rmatte> bwest: rocket actually made that addon zenpack so he'd be best to answer your questions hehe
[25-Jun-2009 10:18:16] <bwest> rocket: ok will it be in the next release? if so when will that be released
[25-Jun-2009 10:18:35] <mrayzenoss> atrawog: yeah, that should be OK. I'll figure out how to pull it over
[25-Jun-2009 10:20:16] <rocket> bwest: send me the location of the commands and the data you need and I will see what I can do .. no gaurantees
[25-Jun-2009 10:20:34] <mrchippy> venturaville: i was able to do "dmd.Devices.findDevice('mybox.zenoss.loc')" and it worked. it has to be an exact device id match. is this what you did?
[25-Jun-2009 10:20:43] <rocket> bwest: it will be an extension of the advanced linux monitor zenpack then
[25-Jun-2009 10:21:49] <rocket> bwest: depending on my time I may be able to get to it .. I am working on something with zenoss that might prevent me from working on this though .. thats something I will have to discuss with matt when the time comes
[25-Jun-2009 10:22:06] <klinstifen> mray: the datapoints don't exist to be selected because they were created in the custom definition
[25-Jun-2009 10:22:22] <klinstifen> mray: i took two datapoints and divided them against each other...i wanted to set a threshold against that
[25-Jun-2009 10:22:22] <venturaville> mrchippy: yeah I am doing this: dmd.Devices.findDevice('www.myhostname.com')
[25-Jun-2009 10:22:26] <venturaville> and it returns None
[25-Jun-2009 10:22:52] <mrchippy> venturaville: also, find() and dmd.Devices.findDevice() use zope catalog glob searches so * is an option.
[25-Jun-2009 10:23:14] <venturaville> >>> dmd.Devices.Site.getSubDevices_recursive()
[25-Jun-2009 10:23:33] <venturaville> [<Device at /zport/dmd/Devices/Site/devices/www.myhostname.com>]
[25-Jun-2009 10:23:47] <venturaville> (names changed to protect the innocent)
[25-Jun-2009 10:24:01] <mrchippy> venturaville: that probably means it's not in the catalog (maybe)
[25-Jun-2009 10:24:13] <venturaville> how would that happen?
[25-Jun-2009 10:24:24] <bwest> Another question The Admin Password shows up in plain text when I run a test of the WMIC commands as well in event summary can I stop this
[25-Jun-2009 10:24:53] <mrchippy> venturaville: try this-- dmd.Devices.Site.devices._getOb('www.myhostname.com').index_object()
[25-Jun-2009 10:25:25] <venturaville> None
[25-Jun-2009 10:25:26] <mrchippy> venturaville: how that would happen-- not sure. some goof somewhere. (hopefully not one of mine )
[25-Jun-2009 10:25:35] <rocket> bwest: which zenpack are you using for that?
[25-Jun-2009 10:25:48] <mrayzenoss> klinstifen: so you've programatically added the datapoints, but they don't show up for selection in the Threshold... I guess you could add a different datapoint, and then go change it in the code to your custom one?
[25-Jun-2009 10:25:56] <venturaville> I'm guessing broken zenpack somehow...looking into clean it up now
[25-Jun-2009 10:26:02] <bwest> I have tried the zenoss one and the external one
[25-Jun-2009 10:26:11] <mrchippy> venturaville: dmd.Devices.Site.devices()[0]
[25-Jun-2009 10:26:12] <bwest> at the moment none
[25-Jun-2009 10:26:21] <mrchippy> venturaville: what does that give you?
[25-Jun-2009 10:26:33] <bwest> rocket:I have tried to run it as a non admin with no success
[25-Jun-2009 10:26:35] <venturaville> <Device at /zport/dmd/Devices/Site/devices/www.myhostname.com>
[25-Jun-2009 10:26:50] <bwest> rocket: can't monitor any services
[25-Jun-2009 10:26:53] <klinstifen> mray: hmm...could try that
[25-Jun-2009 10:27:05] <mrchippy> venturaville: then try dmd.Devices.Site.devices()[0].index_object()
[25-Jun-2009 10:27:19] <mrchippy> venturaville: it won't return anything but then try the find again
[25-Jun-2009 10:27:26] <klinstifen> mray: set that datapoint to "not drawn"?
[25-Jun-2009 10:27:29] <venturaville> None
[25-Jun-2009 10:27:37] <venturaville> voila
[25-Jun-2009 10:27:40] <venturaville> magic
[25-Jun-2009 10:28:05] <klinstifen> mray: err...if i set it to "not drawn" will the threshold still evaluate or will that setting cause the threshold to ignore the values?
[25-Jun-2009 10:28:18] <rmatte> in zendmd, is there a quicker way than doing something like mydev = dmd.Devices.findDevice('DeviceName') \ print mydev.zWinPassword to retrieve the contens of a zProperty?
[25-Jun-2009 10:28:43] <paulh> d=find('DevNme')
[25-Jun-2009 10:28:57] <rocket> bwest: I havent done anything with wmi yet .. but from what I have heard the only way to keep the password from showing would be to update the zenpack causing it ... or do an event transform that strips the information ..
[25-Jun-2009 10:29:02] <rmatte> paulh: identical speed, I meant quicker as in retrieval time
[25-Jun-2009 10:29:03] <mrchippy> rmatte: yeah, what paulh said
[25-Jun-2009 10:29:12] <paulh> got it
[25-Jun-2009 10:29:16] <paulh> don't think so
[25-Jun-2009 10:29:17] <mrchippy> rmatte: ohh
[25-Jun-2009 10:29:24] <klinstifen> bwest: yeah there are a couple forums postings about using the event transform to hide the password
[25-Jun-2009 10:29:29] <rmatte> I'm trying to amp up the performance on some scripts if possible
[25-Jun-2009 10:29:38] <mrchippy> rmatte: only if you know exactly where it lives
[25-Jun-2009 10:30:08] <rmatte> I see, hmm, doubt it's possible then
[25-Jun-2009 10:30:12] <venturaville> mrchippy: that worked for all of them thanks......
[25-Jun-2009 10:30:23] <zenChild> For the SSH modeling plugins, is it possible to get the handle of the ssh pipe to send additional commands, or am I limited to simply having "echo" delimiters for a single command.
[25-Jun-2009 10:30:24] <mrchippy> venturaville: sweet
[25-Jun-2009 10:30:25] <mrayzenoss> bwest: another problem is that the WMI ZenPacks that lots of people use aren't in the Community ZenPack repo, so nobody's updating them
[25-Jun-2009 10:30:37] <rmatte> I modified the WMI Performance zenpack a while back to retrieve the username and password values on it's own, but it's extremely intensive on the system when it's running
[25-Jun-2009 10:30:38] <bwest> rocket: good idea, any ideas about when you run the test in the template area?
[25-Jun-2009 10:31:04] <rocket> bwest: probably same solution to use a transform ..
[25-Jun-2009 10:31:55] <bwest> rocket: you can use a transform in templates when you are testing the command against a device?
[25-Jun-2009 10:32:13] <rocket> bwest: if it shows up in the event
[25-Jun-2009 10:32:58] <bwest> rocket: ok no I meant when i actually run the test in the template and it returns the result of the test the password shows up there in plain text
[25-Jun-2009 10:33:44] <rocket> bwest: probably the only solution is to update the zenpack in that case
[25-Jun-2009 10:33:59] <rocket> eg the zenpack is returning that information to the screen
[25-Jun-2009 10:34:00] <mrchippy> bwest: let me see if i'm getting this-- the prob is that testing the command shows the pword in plain txt?
[25-Jun-2009 10:34:06] <atrawog> @zenchild If you want you can send a complete shell script.
[25-Jun-2009 10:34:25] <bwest> mrchippy: yes
[25-Jun-2009 10:34:35] <rocket> mrchippy: correct .. there is a bug report filed that is closed as a wont fix because it technically is the zenpack maintainers issue
[25-Jun-2009 10:35:07] <bwest> mrchippy:so when zenoss admins are accessing this they can see admin passwords on devices they shouldn't
[25-Jun-2009 10:35:09] <zenChild> atrawog: I can't send shell scripts. It's an appliance running rbash. I'm extremely limited in what I can do at the shell.
[25-Jun-2009 10:35:18] <mrayzenoss> rocket: rmatte: bwest: and I've asked the maintainer to submit the ZenPack to the repository so it could get fixed, but he hasn't
[25-Jun-2009 10:35:45] <rmatte> mrayzenoss: yeh I know, I even submitted my changes to him as a rough start but haven't heard back from him
[25-Jun-2009 10:35:52] <rocket> mrayzenoss: should just take the zenpack and apply the maintainer mode work around and update it ourselves
[25-Jun-2009 10:36:02] <rmatte> there has to be a better way to pass the username and password info to the script
[25-Jun-2009 10:36:04] <mrchippy> bwest: actually kells is working on a mechanism that would do it or at least give you a hook that made it obvious where to do the obfuscating. hopefully, that'll be in the code soon
[25-Jun-2009 10:36:10] <mrayzenoss> rocket: well, it's his ZenPack, so we wouldn't be able to clear the licensing
[25-Jun-2009 10:36:16] <rmatte> if I can find a way I'll package it up and submit it myself
[25-Jun-2009 10:36:29] <rocket> mrayzenoss: was it gpl'd?
[25-Jun-2009 10:36:54] <rocket> mrayzenoss: my comment was based on the fact it was gpld of course
[25-Jun-2009 10:37:04] <rmatte> same here
[25-Jun-2009 10:37:06] <bwest> ok thx guys, that would be a major help
[25-Jun-2009 10:37:21] <venturaville> anyone know how I translate a list of device object to a list of objects in tales for a skin?
[25-Jun-2009 10:38:10] <rmatte> those zenpacks need to be updated badly, they are ancient
[25-Jun-2009 10:39:10] <bwest> The one I believe is Dalstrup and the other i used was from the community
[25-Jun-2009 10:39:30] <rocket> rmatte: how bad would it be to implement our own wmi zenpack?
[25-Jun-2009 10:39:48] <rocket> rmatte: I dont personally do much with windows so I havent looked at what they do etc
[25-Jun-2009 10:39:56] <rmatte> rocket: would be pretty easy from what I've seen, we should do that
[25-Jun-2009 10:40:09] <bwest> Yes please do that
[25-Jun-2009 10:40:31] <rocket> rmatte: there you go .. summer of code contest entry for ya
[25-Jun-2009 10:40:37] <mrayzenoss> So I checked, there's no license on his ZenPacks, which means we won't touch them
[25-Jun-2009 10:40:50] <rocket> mrayzenoss: makes sense to me
[25-Jun-2009 10:41:08] <venturaville> mrayzenoss: which license do you want on zenpacks anyways?
[25-Jun-2009 10:41:14] <rmatte> rocket: start with a straight performance Zenpack, then make a replacement for the Exchange and IIS Zenpacks
[25-Jun-2009 10:41:18] <zenChild> venturaville: have you looked this over? It came in handy when mucking with skins the other day. http://www.zenoss.com/community/docs/howtos/using-zentablemanager/
[25-Jun-2009 10:41:35] <mrayzenoss> GPLv2, with the Contributor Agreement signed, so we can post them and modify them as necessary
[25-Jun-2009 10:41:39] <mrchippy> venturaville: tal:define "objects here/getMyDeviceList"
[25-Jun-2009 10:41:52] <rmatte> rocket: still need to figure out how to pass the username and password to it in a more elegant fashion though
[25-Jun-2009 10:42:09] <mrchippy> venturaville: is that too obtuse? do you have a more specific thing in mind?
[25-Jun-2009 10:42:24] <mrayzenoss> more permissive licenses are OK than GPL, so you could go with BSD if you prefer
[25-Jun-2009 10:42:34] <rmatte> or AGPL
[25-Jun-2009 10:42:36] <rmatte>
[25-Jun-2009 10:42:36] <klinstifen> rmatte: i have an exchange zenpack i made...if you guys are interested i could go through the shabang and post it?
[25-Jun-2009 10:42:47] <mrayzenoss> yeah
[25-Jun-2009 10:42:51] <rocket> rmatte: just do like the command password is being done
[25-Jun-2009 10:42:51] <mrayzenoss> klinstifen: yeah
[25-Jun-2009 10:42:52] <bwest> Yes
[25-Jun-2009 10:43:03] <rocket> rmatte: in fact probably could leverage that code that exists
[25-Jun-2009 10:43:17] <rmatte> klinstifen: if it's nicer than the exchange zenpack that's there (which I'm sure it is) then please do submit it
[25-Jun-2009 10:43:29] <rmatte> klinstifen: you went through the effort of making it, people might as well use it hehe
[25-Jun-2009 10:43:32] <venturaville> mrchippy: I have a python list of references to actual devices, but it looks as though the objects that zentable manager is a little different ... the list I have is generated using a lambda and map combo from a list of device names
[25-Jun-2009 10:43:38] <klinstifen> rmatte: which one?
[25-Jun-2009 10:43:54] <rmatte> actually, sorry not that's there
[25-Jun-2009 10:44:02] <rmatte> but if it's nicer than Dastrup's
[25-Jun-2009 10:44:08] <rmatte> he sort of threw his together real quick
[25-Jun-2009 10:44:10] <rmatte> and it's outdated
[25-Jun-2009 10:44:13] <bwest> Is there an update in the works for the asterisk zenpack, I have seen mixed reviews for that one
[25-Jun-2009 10:44:28] <mrchippy> ahh, good ol' ZenTableManager. Let me find an example for you. You have to use getBatch()
[25-Jun-2009 10:44:33] <mrayzenoss> bwest: the original author said he's not using Asterisk anymore, so it's somewhat orphaned
[25-Jun-2009 10:44:36] <klinstifen> oh isn't dastrup's just in the forums?
[25-Jun-2009 10:44:43] <rmatte> asterisk is the worst VOIP system I have ever used in my life lol
[25-Jun-2009 10:45:02] <venturaville> mrchippy: yep, and getBatch() is a bit difficult to follow
[25-Jun-2009 10:45:03] <rocket> rmatte: are you using a free voip system?
[25-Jun-2009 10:45:04] <bwest> yes I agree but my company is using it and wants to monitor it
[25-Jun-2009 10:45:05] <mrayzenoss> klinstifen: yeah, and on his website. It's not hosted by Zenoss because he hasn't sent it to me
[25-Jun-2009 10:45:14] <rocket> rmatte: I was thinking of making somethign for home at one time ..
[25-Jun-2009 10:45:21] <klinstifen> ok...
[25-Jun-2009 10:45:24] <rmatte> rocket: we were at one point, we went back to the actual Cisco stuff
[25-Jun-2009 10:45:26] <klinstifen> well before i go packaging mine
[25-Jun-2009 10:45:32] <mrayzenoss> bwest: the Asterisk one is just wrapping a Nagios plugin of some sort, so you could probably do something similar
[25-Jun-2009 10:45:41] <bwest> ok
[25-Jun-2009 10:45:47] <klinstifen> rmatte: what do you think is outdated on his?
[25-Jun-2009 10:45:47] <mrchippy> venturaville: i disagree. it's impossible to follow.
[25-Jun-2009 10:45:58] <zenChild> venturaville: since you are familiar with the SVC concepts, here is a pt file that I am using to list out MDisk Groups... maybe it'll be of some help. TALES is pretty archaic though ... http://pastie.org/524340
[25-Jun-2009 10:46:06] <rmatte> klinstifen: the way it's packaged lol
[25-Jun-2009 10:46:22] <klinstifen> oh : )
[25-Jun-2009 10:46:23] <klinstifen> ok
[25-Jun-2009 10:46:25] <venturaville> I think I can write it just fine without getBatch, but I presume getBatch is used to lower memory usage
[25-Jun-2009 10:46:27] <klinstifen> i'll work on mine
[25-Jun-2009 10:46:31] <klinstifen> and try to get it posted
[25-Jun-2009 10:46:45] <rmatte> nice
[25-Jun-2009 10:46:57] <mrayzenoss> anyone else with a ZenPack, send them in. Just 7 more weeks in the contest
[25-Jun-2009 10:47:01] <klinstifen> i think i have told mray i would do this before too, but i didnt follow through ; )
[25-Jun-2009 10:47:03] <venturaville> objectValuesAll .... that is the crux of what I don't quite understand
[25-Jun-2009 10:47:12] <rmatte> I'll take a look at yours once you've posted, I can probably mod it in to a performance zenpack if it's good
[25-Jun-2009 10:47:15] <bwest> rmatte: Going back just a bit is it ossible for me to deny users access to certain pages in zenoss?
[25-Jun-2009 10:47:18] <klinstifen> mray: got a link for this contest?
[25-Jun-2009 10:47:33] <venturaville> zenChild that link hels a bit
[25-Jun-2009 10:47:33] <mrayzenoss> klinstifen: http://blog.zenoss.com/2009/05/17/summer-of-community-zenpacks-contest/
[25-Jun-2009 10:47:49] <klinstifen> rmatte: errr...whats the diff between a zenpack and a performance zenpack?
[25-Jun-2009 10:47:54] <rocket> bwest: easy answer is no for core .. but there are hacks to allow it .. I believe its supported in enterprise ...
[25-Jun-2009 10:48:06] <mrayzenoss> bwest: it's an Enterprise feature, but you can get into the Zope permissions and set it up
[25-Jun-2009 10:48:08] <klinstifen> mray: thanks
[25-Jun-2009 10:48:12] <zenChild> venturaville: that (objectValuesAll) is the python function that is being called.
[25-Jun-2009 10:48:13] <rmatte> bwest: if you go in the zope management interface, create a role, you can assign it particular permissions and then assign it to users, but I don't think it allows you to control access to particular pages
[25-Jun-2009 10:48:53] <bwest> rmatte: But I can limit a user from say using the templates area
[25-Jun-2009 10:49:01] <klinstifen> rmatte: isn't a performance zenpack just a zenpack with performance graphs?
[25-Jun-2009 10:49:01] <venturaville> zenChild: yep, but how is that different from the actual device....
[25-Jun-2009 10:49:02] <mrayzenoss> bwest: look at http://www.skills-1st.co.uk/papers/jcurry.html "Crafting Zenoss Core users for events and zProperties" for some ideas
[25-Jun-2009 10:49:03] <rmatte> plus I've only used roles with LDAP, so I'm not sure how you assign them to regular users, or if you even can
[25-Jun-2009 10:49:31] <bwest> ok
[25-Jun-2009 10:49:36] <rmatte> klinstifen: correct, but it'd be performance graphs for CPU, Memory, paging, etc...
[25-Jun-2009 10:49:36] <mrchippy> venturaville: objectValuesAll is a method on a relation that gives back all the objects at the other end. for example, deviceClass.devices.objectValuesAll(), but deviceClass.devices() does the same thing. can you post a snippet here or on pastebin?
[25-Jun-2009 10:50:03] <rmatte> klinstifen: It'd take someone very knowledgeable to actually write a full blown WMI ZenPack that matches up to the enterprise one
[25-Jun-2009 10:50:23] <bwest> Another quick question, is there a way to setup administrators of services, just like the other classes?
[25-Jun-2009 10:50:26] <venturaville> mrchippy: here is what I am trying to do (it is very short):
[25-Jun-2009 10:50:32] <venturaville> tal:define="tableName string:vipsList;
[25-Jun-2009 10:50:32] <venturaville> vipdevlist python:map(lambda x: devices.findDevice(x), here.zVIPNameList)
[25-Jun-2009 10:50:32] <venturaville> tabletitle string:VIPs Serving this Site;
[25-Jun-2009 10:50:32] <venturaville> batch python:here.ZenTableManager.getBatch(tableName,vipdevlist)">
[25-Jun-2009 10:50:39] <rmatte> and I think the sales folks over at Zenoss would cringe if they saw something like that dropped on their plate hehe
[25-Jun-2009 10:51:05] <klinstifen> rmatte: i haven't seen the enterprise one so i dont have anything to compare against
[25-Jun-2009 10:51:25] <rmatte> It basically does everything that's capable via SNMP in core, but does it via WMI
[25-Jun-2009 10:51:34] <klinstifen> ah
[25-Jun-2009 10:51:36] <venturaville> where zVIPNameList = a z property of type lines with a list of IP addresses (which correspond to device names)
[25-Jun-2009 10:51:36] <rmatte> interfaces, hardware, software, the works
[25-Jun-2009 10:51:54] <mrayzenoss> Our position is if someone writes a ZenPack and fills out the Contributor form, we'll post it
[25-Jun-2009 10:52:15] <venturaville> I'm writing a zenpack to allow for monitoring of sites/vips/hosts and have linkages between all of them.
[25-Jun-2009 10:52:19] <rmatte> that's cool, but they'd still cringe hehe
[25-Jun-2009 10:52:32] <mrayzenoss> rmatte: the new Predictive Threshold ZenPack is an example
[25-Jun-2009 10:52:43] <rmatte> yeh, that is pretty cool
[25-Jun-2009 10:52:45] <klinstifen> rmatte: hmm gotcha. nah mine is modeled off the one in the cacti forums
[25-Jun-2009 10:53:00] <klinstifen> mailq, db io, rpc requests, etc
[25-Jun-2009 10:53:08] <mrchippy> venturaville: looking at getBatch
[25-Jun-2009 10:53:12] <rmatte> klinstifen: that's all you really need though, so that's cool
[25-Jun-2009 10:53:21] <rmatte> klinstifen: if you find that it works well then it's worth contributing
[25-Jun-2009 10:53:36] <mrayzenoss> 2.4.2 was just released!
[25-Jun-2009 10:53:43] <rmatte> w00t w00t
[25-Jun-2009 10:53:46] <venturaville> got a URL for update lists?
[25-Jun-2009 10:53:49] <bwest> when
[25-Jun-2009 10:53:58] <bwest> what's new on that release
[25-Jun-2009 10:54:00] <mrayzenoss> like 2 minutes ago
[25-Jun-2009 10:54:10] <mrayzenoss> http://downloads.sourceforge.net/zenoss/Release_Notes_Core_2.4.2.pdf
[25-Jun-2009 10:54:12] <klinstifen> rmatte: mray: cool well i will try and get it packaged in time for the contest
[25-Jun-2009 10:54:23] <rmatte> "2.4.2 was JUST released!" ... "when?"
[25-Jun-2009 10:54:24] <rmatte> lol
[25-Jun-2009 10:54:27] <rmatte> too funny
[25-Jun-2009 10:54:37] <bwest> lol
[25-Jun-2009 10:54:48] <bwest> i get to excited like a kid in a candy store
[25-Jun-2009 10:55:01] <rmatte> apparently
[25-Jun-2009 10:55:03] <mrayzenoss> maintenance release, 46 tickets closed
[25-Jun-2009 10:55:16] <bwest> rmatte: Is there a way to set administrators for services like the other classes
[25-Jun-2009 10:55:35] <paulh> no way to assign admins for services
[25-Jun-2009 10:56:52] <bwest> I assume it would take some serious coding
[25-Jun-2009 10:57:04] <venturaville> 5035 was a big one
[25-Jun-2009 10:57:10] <rmatte> does 2.4.2 include rocket's SSH monitoring patch?
[25-Jun-2009 10:57:50] <mrayzenoss> rmatte: no, it wasn't backported
[25-Jun-2009 10:58:07] <mrayzenoss> and he put another patch on after I closed it, so I gotta reopen it
[25-Jun-2009 10:58:22] <mrchippy> venturaville: the object list has to be list of devices (or components, probably). also, the lambda won't work. you need to pass the object list and muck with them (get the zproperty) afterward. do you want to list all devices?
[25-Jun-2009 10:59:18] <bwest> ok thx all for all your help!!
[25-Jun-2009 10:59:31] <rmatte> yeh, I noticed that
[25-Jun-2009 10:59:34] <rmatte> k
[25-Jun-2009 11:00:05] <mrayzenoss> ran out of battery
[25-Jun-2009 11:00:42] <rmatte> hehe
[25-Jun-2009 11:01:52] <mrchippy> venturaville: hold up, i completely misread your snippet
[25-Jun-2009 11:02:25] <zenChild> rocket: you around
[25-Jun-2009 11:02:31] <rocket> yes
[25-Jun-2009 11:03:08] <rmatte> I can't wait until 4799 is fixed
[25-Jun-2009 11:03:18] <rmatte> I have a workaround for it for now
[25-Jun-2009 11:04:16] <mrchippy> venturaville: that should work as long as here.zVIPNameList actually returns as a list and the names are exact matches.
[25-Jun-2009 11:04:27] <venturaville> they are
[25-Jun-2009 11:04:30] <zenChild> rocket: in your aix snmp zenpack. there is the assignment "vgom.getLvSetup = MultiArgs(lv_hash)" in AixVolumeGroupMap.py. How does that work in relation to the method in AixVolumeGroup.py? When does the method get called?
[25-Jun-2009 11:05:05] <venturaville> mrchippy: since I don't have a way of directly linking devices, this is kind of a work around to that ... I am populating the list either manually (in some cases) or collected from load balancers otherwise
[25-Jun-2009 11:05:17] <Appiah_> How do I remove my mibs from Zenoss?
[25-Jun-2009 11:06:35] <rocket> zenChild: getLvSetup is called from AixVolumeGroup.py with the contents of lv_hash as its arguements
[25-Jun-2009 11:06:50] <mrchippy> venturaville: after that, you can do the whole <tr tal:repeat="device batch"> and in the row refer to the "device" object
[25-Jun-2009 11:06:57] <mrayzenoss> Appiah_: go to the MIBs page and select the ones you want to delete. From the page tab select "Delete Mibs..."
[25-Jun-2009 11:06:59] <rocket> zenChild: the modeller code is passing all that into the model via an object map
[25-Jun-2009 11:07:07] <Appiah_> mrayzenoss: i see nothing like that
[25-Jun-2009 11:07:08] <mrchippy> venturaville: is this what you have already?
[25-Jun-2009 11:07:23] <mrayzenoss> Appiah_: which part?
[25-Jun-2009 11:07:50] <Appiah_> there is no arrow drop menu (or whatever you call it) under /Mibs
[25-Jun-2009 11:07:54] <rocket> zenChild: lv_hash will contain getPvSetup for example that will call the special method from the logical_volume code to setup a new pv
[25-Jun-2009 11:07:58] <Appiah_> there is on under sub-folders but that's it
[25-Jun-2009 11:08:50] <Appiah_> this is weird , another zenoss server that is running got that drop menu ... same version
[25-Jun-2009 11:08:55] <rocket> zenChild: if you look at AixVolumeGroup.py AixLogicalVolume.py as examples you will see these specialized methods
[25-Jun-2009 11:09:22] <rmatte> Appiah_: try restarting zeoctl?
[25-Jun-2009 11:09:23] <zenChild> rocket: looking now.
[25-Jun-2009 11:09:32] <Appiah_> rmatte: i'll try that
[25-Jun-2009 11:09:36] <mrayzenoss> Appiah_: are you an unprivileged user?
[25-Jun-2009 11:09:49] <rocket> zenChild: but ultimately nothing happens in the modeller other than a big object map object is created
[25-Jun-2009 11:09:53] <rmatte> unprivileged users shouldn't even see the Mibs section listed
[25-Jun-2009 11:09:58] <Appiah_> mrayzenoss: oh
[25-Jun-2009 11:10:03] <rocket> zenChild: all the work is done inside the objects themselves
[25-Jun-2009 11:10:08] <Appiah_> then his user is configured wrongly
[25-Jun-2009 11:10:11] <Appiah_> this*
[25-Jun-2009 11:10:24] <zenChild> rocket: how does the method actually get called though?
[25-Jun-2009 11:10:35] <rocket> zenChild: its a special name
[25-Jun-2009 11:10:51] <rmatte> Appiah_: check with the admin account, if it works with that then that's your problem
[25-Jun-2009 11:10:57] <klinstifen> mray: are you guys offering any developer's conferences?
[25-Jun-2009 11:11:15] <rocket> eg if the object map name getPvName exists in the object it gets called
[25-Jun-2009 11:11:15] <venturaville> mrchippy: yeah the rest of the skin is basically the same as one I copied from elsewhere
[25-Jun-2009 11:11:23] <klinstifen> mray: advanced zenpack creation...zenoss and python...stuff like that?
[25-Jun-2009 11:11:35] <rocket> zenChild: and the stuff it gets set to in the objectmap is just the arguement list for it
[25-Jun-2009 11:11:38] <rmatte> if you guys eventually do zenoss development training courses I'm going to have my boss send me to one
[25-Jun-2009 11:11:42] <zenChild> rocket: That clears it up for me.
[25-Jun-2009 11:11:54] <mrayzenoss> klinstifen: well, we'll have a community day at LISA and we'll be at Ohio LinuxFest
[25-Jun-2009 11:12:00] <rocket> zenChild: great
[25-Jun-2009 11:12:07] <zenChild> rocket: I was looking for an actual method call and was a bit perplexed as to how things were happening.
[25-Jun-2009 11:12:12] <klinstifen> hmm....anything in the nyc area?
[25-Jun-2009 11:12:18] <venturaville> mrayzenoss: you know Bill Lefebvre at LISA?
[25-Jun-2009 11:12:28] <klinstifen> i need a zenoss meetup or zenossnycgroup or something
[25-Jun-2009 11:12:30] <mrayzenoss> venturaville: nope
[25-Jun-2009 11:12:44] <venturaville> mrayzenoss: just curious
[25-Jun-2009 11:12:45] <rocket> zenChild: the modeller code does a fancy trick to see if its a method or a parameter type item
[25-Jun-2009 11:12:45] <mrayzenoss> klinstifen: we do have a sales engineer in NYC
[25-Jun-2009 11:12:59] <mrayzenoss> I can look into doing a dev conference
[25-Jun-2009 11:13:08] <Appiah_> rmatte , mrayzenoss same on the Admin account , should i try restaring zeoctl ?
[25-Jun-2009 11:13:17] <mrayzenoss> Appiah_: can't hurt
[25-Jun-2009 11:13:29] <zenChild> rocket: good to know. Know I can keep moving on my project. Thanks again.
[25-Jun-2009 11:13:32] <rmatte> klinstifen: this IRC channel is about the best meetup you're going to get from what I've seen hehe
[25-Jun-2009 11:13:45] <klinstifen> mray: neat
[25-Jun-2009 11:13:51] <klinstifen> rmatte: yeah : \
[25-Jun-2009 11:13:55] <klinstifen> not fun on quiet days
[25-Jun-2009 11:14:05] <Appiah_> restarted , still the same
[25-Jun-2009 11:14:22] <klinstifen> but actually its been pretty active the past couple times i have hopped on
[25-Jun-2009 11:14:22] <rmatte> once Zenoss really takes off (which I'm sure it will) there might be something like that
[25-Jun-2009 11:14:25] <mrchippy> venturaville: if you follow all the steps in your tal:define in zendmd, does batch have things in it? that is does dmd.ZenTableManager.getBatch(tableName,vipdevlist) return empty?
[25-Jun-2009 11:15:11] <Appiah_> so what do you do when a drop down menu is missing?
[25-Jun-2009 11:15:13] <mrayzenoss> yeah, we started to plan a Zenoss User Conference, but had trouble nailing down a date
[25-Jun-2009 11:15:27] <klinstifen> rmatte: i'ven been on this boat since v.1.something. i am sold. at the time i was comparing against hyperic, but the agentlessness won me over
[25-Jun-2009 11:15:50] <mrayzenoss> Appiah_: have you logged out and logged back in, maybe something weird cached
[25-Jun-2009 11:15:51] <zenChild> mrayzenoss: what about a BOF at OSCON or something similar?
[25-Jun-2009 11:15:52] <rmatte> yeh, agentless is golden
[25-Jun-2009 11:16:18] <rmatte> and I like how easy it is to write scripts and integrate them in for data collection
[25-Jun-2009 11:16:19] <mrayzenoss> zenChild: I'll be at the Community Leadership Summit the weekend before OSCON, but I'm heading home after that
[25-Jun-2009 11:16:25] <venturaville> I can't run getBatch inside of DMD without a session...
[25-Jun-2009 11:16:27] <klinstifen> appiah: flush your cache? there've been forum postings to fix other dashboard quirkiness
[25-Jun-2009 11:16:31] <venturaville> File "/opt/zenoss/zenoss/Products/ZenWidgets/ZenTableManager.py", line 295, in getTableStates
[25-Jun-2009 11:16:31] <venturaville> session = self.REQUEST.SESSION
[25-Jun-2009 11:16:43] <Appiah_> mrayzenoss: tried both firefox and IE ( I never used it with IE before) same problem
[25-Jun-2009 11:16:44] <venturaville> File "/opt/zenoss/zenoss/lib/python/ZPublisher/HTTPRequest.py", line 1219, in __getattr__
[25-Jun-2009 11:16:44] <venturaville> raise AttributeError, key
[25-Jun-2009 11:16:44] <venturaville> AttributeError: SESSION
[25-Jun-2009 11:16:57] <mrchippy> venturaville: well, that sucks
[25-Jun-2009 11:17:08] <venturaville> yep ... makes writing templates hard if using table manager
[25-Jun-2009 11:17:09] <mrayzenoss> zenChild: OSCON is really expensive for vendors, so we dropped out
[25-Jun-2009 11:17:17] <rmatte> I'm going to be getting a company Zenoss blog up and running some time next month, I'll be posting a bunch of scripts and tricks that I've found on there
[25-Jun-2009 11:17:26] <mrayzenoss> rmatte: excellent
[25-Jun-2009 11:17:33] <klinstifen> rmatte: yeah i dig the scripting aspect. if i can think of it then it can probably be done in zenoss (or already has been ; )
[25-Jun-2009 11:17:53] <klinstifen> rmatte: pm me in the forums once that is up...i'd like to check it out
[25-Jun-2009 11:17:55] <rmatte> klinstifen: yeh, I've done some really neat stuff with scripts
[25-Jun-2009 11:18:05] <rmatte> sure thing
[25-Jun-2009 11:18:09] <mrayzenoss> I did submit Chet's name for a Guru session at LISA entitled "Try to Stump Chet, Monitor Anything With Zenoss"
[25-Jun-2009 11:18:22] <venturaville> mrchippy: I'll play with this a bit and see how far I get ... I'm booked for meetings the rest of today
[25-Jun-2009 11:18:27] <venturaville> thanks for the help
[25-Jun-2009 11:18:39] <mrchippy> venturaville: np. sorry we couldn't bang it out.
[25-Jun-2009 11:18:42] <rmatte> mrayzenoss: does he have access to Enterprise or Core for that?
[25-Jun-2009 11:19:07] <mrayzenoss> rmatte: well, it's a Q&A session so he'll just tell you how you could do it
[25-Jun-2009 11:19:14] <rmatte> ah
[25-Jun-2009 11:19:17] <mrayzenoss> so in theory, it's either
[25-Jun-2009 11:19:21] <rmatte> I'd show up with a list hehe
[25-Jun-2009 11:19:33] <rocket> how would he monitor sap
[25-Jun-2009 11:19:38] <rocket>
[25-Jun-2009 11:19:43] <rmatte> rocket: nice one
[25-Jun-2009 11:19:44] <rocket> muaaa haaa haaa
[25-Jun-2009 11:20:22] <mrayzenoss> rocket: ports, processes, web pages, the actual machines, databases
[25-Jun-2009 11:20:31] <rmatte> I'd just ask how he'd monitor AIX, since that's a bit shaky from what I understand?
[25-Jun-2009 11:20:44] <rocket> rmatte: dont get me going ... ;p
[25-Jun-2009 11:20:56] <Appiah_> i*m running 2.4.1 and these is no menu for the mibs on this server ( got other servers that actully has the dropdown menu) , is there no command line for removing?
[25-Jun-2009 11:20:56] <rocket> rmatte: did you look at my zenpack for aix?
[25-Jun-2009 11:21:03] <klinstifen> rmatte: i monitored my aix box for a while through a ssh script i wrote
[25-Jun-2009 11:21:12] <rmatte> I need to monitor some AIX boxes eventually, going to be fun, and no, I haven't taken a look at the pack yet
[25-Jun-2009 11:21:14] <klinstifen> i piped everything into zenoss from a command datasource
[25-Jun-2009 11:21:28] <klinstifen> didn't want to use snmp because we have heard bad things about snmp and aix...
[25-Jun-2009 11:21:31] <rocket> rmatte: take a look at the screenshots for it
[25-Jun-2009 11:21:41] <rocket> klinstifen: there is bad stuff for snmp ..
[25-Jun-2009 11:21:41] <atrawog> rmatte:The new AIX ZenPack is great.
[25-Jun-2009 11:21:51] <rocket> atrawog: you using it?
[25-Jun-2009 11:21:52] <rmatte> klinstifen: there is a net-snmp port for AIX though I think which is way better than the stock SNMP
[25-Jun-2009 11:22:11] <rocket> atrawog: like my improvements?
[25-Jun-2009 11:22:21] <klinstifen> rmatte: cool
[25-Jun-2009 11:22:28] <klinstifen> we'll have to check it out
[25-Jun-2009 11:22:36] <klinstifen> for the time being we just monitor that box seperatly
[25-Jun-2009 11:22:41] <atrawog> rocket: I have put on my Zenoss test server two days ago.
[25-Jun-2009 11:22:42] <rmatte> rocket: checking screenshot...
[25-Jun-2009 11:23:02] <mrayzenoss> Appiah_: hmmm... don't see a command line option for removing with zenmib
[25-Jun-2009 11:23:14] <mrayzenoss> Appiah_: still not sure why you can't get to the delete
[25-Jun-2009 11:23:17] <Appiah_> I've got 2 different servers that's missing drop down menus under mibs (wth)
[25-Jun-2009 11:23:21] <rmatte> rocket: AIX SNMP?
[25-Jun-2009 11:23:27] <rocket> rmatte: yes
[25-Jun-2009 11:24:07] <rocket> rmatte: I might work on an ssh version but it depends on time and an agreement between zenoss and myself I am working towards
[25-Jun-2009 11:24:27] <rmatte> hmmmm, does it do interfaces too?
[25-Jun-2009 11:24:31] <mrayzenoss> Appiah_: this is missing right? http://imagebin.org/53650
[25-Jun-2009 11:24:36] <rocket> rmatte: yes
[25-Jun-2009 11:24:50] <rmatte> rocket: so you know how to create a modeler plugin eh?
[25-Jun-2009 11:25:04] <rocket> of course
[25-Jun-2009 11:25:20] <atrawog> rocket: Modeling VGs + LVs is wonderfull. But I have to test how to migrate to the new version.
[25-Jun-2009 11:25:22] <Appiah_> exactly mrayzenoss
[25-Jun-2009 11:25:27] <rmatte> rocket: couldn't you just take the AIX ZenPack code and modify it for WMI?
[25-Jun-2009 11:25:29] <klinstifen> rocket: you teach yourself that or did you pick it up from one of the guides?
[25-Jun-2009 11:25:31] <Appiah_> no white arrow there, nothing to click
[25-Jun-2009 11:25:51] <mrayzenoss> Appiah_: and you're logged in as the Admin?
[25-Jun-2009 11:25:58] <Appiah_> yupp
[25-Jun-2009 11:26:07] <Appiah_> even tried another browser
[25-Jun-2009 11:26:10] <rocket> atrawog: that might be an issue .. but what I usually did was delete the old device .. keeping alerts and perf stuff .. and then adding the device back in
[25-Jun-2009 11:26:30] <mrayzenoss> Appiah_: put up a screenshot on imagebin and I'll pass it around for ideas
[25-Jun-2009 11:26:38] <Appiah_> k , just a sec
[25-Jun-2009 11:26:56] <rocket> rmatte: I am sure I could .. like I said I just dont have much time lately on windows boxes
[25-Jun-2009 11:27:19] <rocket> klinstifen: guides, self taught, a few questions here and there of chet
[25-Jun-2009 11:27:24] <rmatte> rocket: fly over here and you'll have all the time you need
[25-Jun-2009 11:27:36] <klinstifen> rocket: cool stuff
[25-Jun-2009 11:27:43] <rocket> klinstifen: actually I spent time putting debug statements in my version of zenoss to figure out how the main code worked
[25-Jun-2009 11:27:59] <klinstifen> woah intensive
[25-Jun-2009 11:28:40] <rocket> klinstifen: my zenpack is the first one to do multidepth objects .. eg the volume group -> logical volume relationship etc
[25-Jun-2009 11:28:54] <rocket> logical volume -> filesystem
[25-Jun-2009 11:29:16] <rocket> rmatte: my currently company would kill me
[25-Jun-2009 11:29:31] <rmatte> rocket: hehe
[25-Jun-2009 11:30:18] <atrawog> rocket: I have done that in the past and it did work well. But in the case of our AIX servers I have modified the filesystem templates to show absolute filesystem sizes instead of percentage usage and see how that works with the new version.
[25-Jun-2009 11:30:24] <rmatte> well, maybe I'll take a look at the bones of the AIX ZenPack and have a go at developing it myself at some point
[25-Jun-2009 11:30:40] <rmatte> though my python coding skills are still pretty bad
[25-Jun-2009 11:30:47] <Appiah_> http://imagebin.org/53653 mrayzenoss
[25-Jun-2009 11:31:02] <rmatte> if I could script it in bash it'd be done in a day
[25-Jun-2009 11:31:25] <klinstifen> rmatte: hahah
[25-Jun-2009 11:31:27] <mrayzenoss> Appiah_: bizarre
[25-Jun-2009 11:31:53] <rmatte> klinstifen: yeh, I'm a genius when it comes to bash scripting
[25-Jun-2009 11:32:11] <Appiah_> mrayzenoss: I bet it is not a common error
[25-Jun-2009 11:32:37] <Appiah_> but I have this on two different servers , i tried flush cache and 2 different browsers logged in as admin
[25-Jun-2009 11:32:49] <rmatte> man, this AIX zenpack is awesome, makes the AIX monitoring better than the Core Windows monitoring lol
[25-Jun-2009 11:33:02] <mrayzenoss> Appiah_: both servers can't delete MIBs?
[25-Jun-2009 11:33:19] <Appiah_> both servers dont have that drop down menu
[25-Jun-2009 11:33:31] <rocket> rmatte: ... cant have windows beat up on aix too much can we?
[25-Jun-2009 11:33:34] <mrayzenoss> Appiah_: is that the Cisco MIBs ZenPack?
[25-Jun-2009 11:34:00] <Appiah_> nope , no zenpacks installed
[25-Jun-2009 11:34:15] <Appiah_> I was going to remove these mibs and add the zenpack with cisco mib actully
[25-Jun-2009 11:36:03] <rocket> atrawog: is the datapoint data for filesystems the same though?
[25-Jun-2009 11:36:17] <rocket> atrawog: eg are you just displaying it differently on the graph?
[25-Jun-2009 11:38:42] <mrayzenoss> Appiah_: I assume if you restart everything you get the same results?
[25-Jun-2009 11:38:56] <klinstifen> alright all...gotta run...thanks for the help
[25-Jun-2009 11:39:01] <atrawog> rocket: I did a couple of changes in the old version.
[25-Jun-2009 11:39:23] <zenChild> I also have to run. Thanks all for your help. Cheers!
[25-Jun-2009 11:39:51] <mrayzenoss> thanks to everyone for showing up, Dev session is over but per usual, I'm here frequently
[25-Jun-2009 11:40:16] * mrayzenoss just noticed mrchippy had already left
[25-Jun-2009 11:40:28] <atrawog> rocket: I'm fetching totalBlocks used via SNMP or SSH
[25-Jun-2009 11:41:04] <rocket> atrawog: ok ..
[25-Jun-2009 11:41:17] <atrawog> atrawog: I have modified the perfs template to show totalBocksUsed
[25-Jun-2009 11:44:16] <Appiah_> mrayzenoss: logged out , restarted , logged in , same
[25-Jun-2009 11:44:19] <rmatte> rocket: that predictive threshold ZenPack is awesome too, just taking a look at it
[25-Jun-2009 11:44:29] <atrawog> rocket: And I'm running a zendmd script via cron that syncs the totalBocks values from the RRD files with ZODB.
[25-Jun-2009 11:44:51] <rocket> atrawog: why are you doing that zendmd script?
[25-Jun-2009 11:45:09] <rmatte> rocket: still no Ubuntu ZenPack?
[25-Jun-2009 11:45:35] <rocket> rmatte: I have an ubuntu virtual machine installed
[25-Jun-2009 11:45:40] <mrayzenoss> Appiah_: FYI, don't try deleting it from Zope's /manage URL, it screws up royally
[25-Jun-2009 11:45:40] <rocket> rmatte: but no zenpack yet
[25-Jun-2009 11:45:49] <rmatte> rocket: cool
[25-Jun-2009 11:46:18] <rocket> rmatte: currently working on an opensolaris one at the moment
[25-Jun-2009 11:46:27] <rmatte> nice
[25-Jun-2009 11:47:00] <rocket> rmatte: the predictive threshold gave the same results as the enterprise one according to matt
[25-Jun-2009 11:47:11] <rmatte> nice
[25-Jun-2009 11:47:18] <rocket> rmatte: however mine lets you apply the threshold to existing rrd's
[25-Jun-2009 11:47:23] <rmatte> as long as it's coded right it's all math, so it's not surprising
[25-Jun-2009 11:47:42] <rmatte> so your's is even better than the enterprise one hehe
[25-Jun-2009 11:48:00] <rocket>
[25-Jun-2009 11:48:29] <rocket> it also has a script to remove the data from rrd's that have had the predictive stuff turned on
[25-Jun-2009 11:48:35] <rocket> if necessary
[25-Jun-2009 11:48:43] <rmatte> I still need to figure out how to make some sort of install package for those Cisco Mibs so that they install much the same way as a Zenpack, but don't actually get registered as one
[25-Jun-2009 11:48:52] <atrawog> rocket: Because Threashold are based on the TotalBlock value stored in ZODB and my AIX admins complained that the threasholds still trigger after the have resized a filesystem.
[25-Jun-2009 11:48:57] <mrayzenoss> rmatte: using zenmib?
[25-Jun-2009 11:49:14] <mrayzenoss> atrawog: that was yesterday's Tip of the Month
[25-Jun-2009 11:49:16] <rmatte> mrayzenoss: can't, some people followed my guide and weren't able to get them all to load
[25-Jun-2009 11:49:25] <rmatte> only reliable way is to import them in from the objects.xl
[25-Jun-2009 11:49:27] <rmatte> xml*
[25-Jun-2009 11:51:56] <Appiah_> mrayzenoss: delete what? the mibs
[25-Jun-2009 11:52:46] <mrayzenoss> yeah, don't do that
[25-Jun-2009 11:53:23] <atrawog> rocket: Remodelling the device does fix the problem. But its hard to argue why the RRD graphs show the currect values and people still get thresholds warnings.
[25-Jun-2009 11:53:23] <Appiah_> but I cant!
[25-Jun-2009 11:53:25] <rocket> atrawog: ah I see .. but why are you using the value stored in the Zodb?
[25-Jun-2009 11:53:44] <mrayzenoss> gotta run, off to lunch
[25-Jun-2009 11:53:45] <rocket> atrawog: why arent you using the threshold based on the datapoint?
[25-Jun-2009 11:54:04] <Appiah_> then how am I supposed to remove the mibs?
[25-Jun-2009 11:55:00] <rocket> atrawog: actually I probably can code and easy fix for you in my zenpack
[25-Jun-2009 11:55:02] <rmatte> Appiah_: sounds like you have a botched install
[25-Jun-2009 11:55:10] <Appiah_> O_o
[25-Jun-2009 11:55:31] <rocket> atrawog: then you wont need to populate with an zendmd script
[25-Jun-2009 11:55:51] <atrawog> rocket: As far as a know you can only compare a RRD value with a ZODB value in a threshold and not two RRD values which are stored in two different RRD files.
[25-Jun-2009 11:56:04] <rocket> atrawog: there is a better way ..
[25-Jun-2009 11:56:41] <rocket> atrawog: extend the filesystem object class to have a method that returns the totalblock data out of the rrd
[25-Jun-2009 11:56:59] <rocket> then call here/get_totalblock_rrd method
[25-Jun-2009 11:57:47] <venturaville> I believe there is also a cache equivalent of that if you want to reduce load a bit, but I am not sure how long it keeps it
[25-Jun-2009 11:59:25] <venturaville> CACHE_TIME = 60.
[25-Jun-2009 11:59:28] <venturaville> apparently 60
[25-Jun-2009 11:59:34] <rmatte> hehe
[25-Jun-2009 12:00:20] <rocket> venturaville: yeah ...
[25-Jun-2009 12:01:06] <atrawog> rocket: I was pretty sure that there is a better way to do it. But till now I was too scared of writing my own object classes.
[25-Jun-2009 12:02:44] <atrawog> rocket: If you want to have a look you can find my AIX ZenPack version at http://code.google.com/p/atrawog-zenoss-community/source/browse/#hg/zenpacks/ZenPacks.atrawog.AIX
[25-Jun-2009 12:06:19] <rmatte> I guess I will try making a script to use zenmib to install the Cisco Mibs, I'll send it out to some of you guys for testing once it's done.
[25-Jun-2009 12:06:31] <rmatte> It'll probably install the Mibs way faster than the ZenPack anyways
[25-Jun-2009 12:07:06] <Appiah_> what happens if i add a zenpack with mibs, that includes some mibs that i already have?
[25-Jun-2009 12:07:20] <rmatte> it just won't install those Mibs
[25-Jun-2009 12:07:43] <Appiah_> so no risk for conflicts or such
[25-Jun-2009 12:07:54] <rmatte> correct
[25-Jun-2009 12:08:07] <rmatte> even if it did install duplicates it wouldn't cause issues anyways
[25-Jun-2009 12:42:01] <blocky> if my blade is locking up repeatedly, how can I enable some logging to see whats going on
[25-Jun-2009 13:01:09] <shood> Greetings. I would like to have my zenoss enterprise installation monitor permissions on a file tree, and alert on any modifications. Does this functionality exist.. or could someone point me in the direction
[25-Jun-2009 13:02:23] <mrayzenoss> you'd probably need to add a custom datasource
[25-Jun-2009 13:02:56] <mrayzenoss> perhaps a simple script that checks that the permissions for the file are what you want
[25-Jun-2009 13:05:06] * shood reads
[25-Jun-2009 13:33:56] <venturaville> can't get a menu to show up on a device in a zenpack
[25-Jun-2009 13:34:07] <venturaville> anyone have the link for the howto for that
[25-Jun-2009 13:34:11] <venturaville> something I am missing..........
[25-Jun-2009 13:44:14] <rocket> venturaville: menu examples are in the aix zenpack
[25-Jun-2009 13:44:27] <rocket> eg for print queues etc
[25-Jun-2009 13:44:47] <rocket> venturaville: check out the items in __init__.py and the new pt files
[25-Jun-2009 13:45:06] <venturaville> rocket: looks like it can't find my class :-/
[25-Jun-2009 13:45:43] <rocket> venturaville: also menus are only installed on zenpack install unless you create the menu from the gui
[25-Jun-2009 13:46:10] <rocket> venturaville: so I had to keep reinstalling my zenpack ontop of itself etc ..
[25-Jun-2009 13:46:53] <venturaville> the menus are in the more section (like the vmware one)
[25-Jun-2009 13:47:11] <venturaville> zPythonClass is set right
[25-Jun-2009 13:47:23] <venturaville> I think I'm just missing something obvious
[25-Jun-2009 13:47:35] <rocket> venturaville: did you check that the menu shows up in the gui under settings menu?
[25-Jun-2009 13:47:43] <venturaville> yep
[25-Jun-2009 13:48:00] <venturaville> when I add a new device it can't find my class though *shrug*
[25-Jun-2009 13:52:51] <venturaville> 2009-06-25T14:37:39 ERROR zen.DeviceClass Unable to import class ZenPacks.Turner.HostMap.Site
[25-Jun-2009 13:53:03] <venturaville> ImportError: No module named Site
[25-Jun-2009 13:54:06] <venturaville> I've got a zenpack if you want to see
[25-Jun-2009 13:59:51] <etank> im having an issue with setting up alerting rules
[25-Jun-2009 14:00:07] <etank> i have one rule that should show all errors with a delay of 120
[25-Jun-2009 14:00:35] <etank> and one that i want to show Perf CPU issues but only if the issue has been going on for 900 seconds
[25-Jun-2009 14:01:29] <etank> it seems that the rule for the Perf CPU is not working because i get the emails even when the CPU has been over the 90% threshold for 2 minutes or so
[25-Jun-2009 14:02:28] <etank> i may have just found the issue myself
[25-Jun-2009 14:03:48] <venturaville> what is the 'accepted' license for Zenpacks?
[25-Jun-2009 14:04:29] <mrayzenoss> venturaville: well, if someone submits it to me for posting on the Zenoss site it needs to have something GPLv2 compatible
[25-Jun-2009 14:04:58] <mrayzenoss> so the option of repackaging it exists
[25-Jun-2009 14:05:05] <venturaville> ventura@treebeard:~/svn/Zenoss/hostmap/trunk/ZenPacks/Turner/HostMap$ grep -i license ../../../EGG-INFO/PKG-INFO
[25-Jun-2009 14:05:05] <venturaville> License: GPLv2
[25-Jun-2009 14:05:06] <venturaville> like so?
[25-Jun-2009 14:05:32] <mrayzenoss> yeah, that's cool
[25-Jun-2009 14:07:02] <venturaville> now if I could just figure out why it won't import my class I am in business
[25-Jun-2009 14:11:12] <rocket> venturaville: problem loading that zenpack?
[25-Jun-2009 14:11:31] <rocket> venturaville: have you tried installing it from the commandline with a higher verbosity level?
[25-Jun-2009 14:12:33] <venturaville> doing that now ... was trying a few things
[25-Jun-2009 14:14:05] <venturaville> ObjectNotFound: Object with id SiteVIPsData not found on relation packables
[25-Jun-2009 14:14:08] <venturaville> interesting
[25-Jun-2009 14:14:28] <venturaville> ERROR:zen.ZPLoader:Error adding pack to /zport/dmd/zenMenus/More/zenMenuItems/VIPHostsData
[25-Jun-2009 14:14:45] <rocket> venturaville: got a relationship issue
[25-Jun-2009 14:14:51] <venturaville> looks like it
[25-Jun-2009 14:14:56] <rocket> venturaville: check both ends of the relationship
[25-Jun-2009 14:17:57] <venturaville> >>> menuitems
[25-Jun-2009 14:17:57] <venturaville> <ToManyContRelationship at /zport/dmd/zenMenus/More/zenMenuItems>
[25-Jun-2009 14:17:57] <venturaville> >>> menuitems.SiteVIPsData
[25-Jun-2009 14:17:57] <venturaville> 2009-06-25 15:17:08 WARNING ZODB Could not import class 'ZenPack' from module 'ZenPacks.Turner.HostMap'
[25-Jun-2009 14:24:38] <rocket> venturaville: how are you creating your menus?
[25-Jun-2009 14:24:43] <rocket> venturaville: code or gui?
[25-Jun-2009 14:24:45] <venturaville> in objects.xml
[25-Jun-2009 14:25:11] <rocket> you mean hand editing objects.xml?
[25-Jun-2009 14:25:17] <venturaville> yep
[25-Jun-2009 14:25:23] <rocket> ick!
[25-Jun-2009 14:25:27] <venturaville> indeed
[25-Jun-2009 14:25:28] <rocket> dont do that please
[25-Jun-2009 14:25:43] <venturaville> I copied it over from an existing zenpack
[25-Jun-2009 14:26:11] <venturaville> <!-- ('', 'zport', 'dmd', 'zenMenus', 'More', 'zenMenuItems', 'SiteVIPsData') -->
[25-Jun-2009 14:26:11] <venturaville> <object id='/zport/dmd/zenMenus/More/zenMenuItems/SiteVIPsData' module='Products.ZenModel.ZenMenuItem' class='ZenMenuItem'>
[25-Jun-2009 14:27:07] <rocket> venturaville: grab the aix zenpack for me and take a look at __init__.py
[25-Jun-2009 14:27:13] <venturaville> k
[25-Jun-2009 14:27:16] <rocket> look for this section
[25-Jun-2009 14:27:28] <rocket> def installMenuItems(self, dmd):
[25-Jun-2009 14:27:33] <venturaville> AIX SNMP?
[25-Jun-2009 14:27:36] <rocket> yes
[25-Jun-2009 14:27:50] <rocket> this is the code to add to the More menu as an example
[25-Jun-2009 14:27:51] <rocket> # Set an additional item in the More Menu
[25-Jun-2009 14:27:51] <rocket> #dmd.zenMenus.More.manage_addZenMenuItem("aixprintqueueDetail",
[25-Jun-2009 14:27:51] <rocket> # action="aixprintqueueDetail",
[25-Jun-2009 14:27:51] <rocket> # description="Aix Print Queue Details",
[25-Jun-2009 14:27:54] <rocket> # ordering=5.0)
[25-Jun-2009 14:28:08] <rocket> eg that would add aixprintqueueDetail to the More menu
[25-Jun-2009 14:28:23] <rocket> and would need the new zope page template aixprintqueueDetail.pt
[25-Jun-2009 14:28:53] <rocket> this is the code to remove it on zenpack uninstall
[25-Jun-2009 14:28:58] <rocket> def removeMenuItems(self,dmd):
[25-Jun-2009 14:28:58] <rocket> # Remove an additional item in the More Menu
[25-Jun-2009 14:28:58] <rocket> #id = "aixprintqueueDetail"
[25-Jun-2009 14:28:58] <rocket> #try:
[25-Jun-2009 14:28:58] <rocket> # dmd.zenMenus.More.manage_deleteZenMenuItem(("aixprintqueueDetail",))
[25-Jun-2009 14:29:01] <rocket> #except (KeyError, AttributeError):
[25-Jun-2009 14:29:03] <rocket> # pass
[25-Jun-2009 14:29:24] <rocket> obviously it needs to be uncommented
[25-Jun-2009 14:29:44] <rocket> but it is much cleaner than editing object.xml
[25-Jun-2009 14:29:57] <venturaville> +1 to that
[25-Jun-2009 14:30:46] <rocket> venturaville: you probably need to create the def install method also
[25-Jun-2009 14:31:01] <rocket> venturaville: but that should be straight forward enough from my zenpack
[25-Jun-2009 14:31:14] <venturaville> 'def install'?
[25-Jun-2009 14:31:27] <rocket> def install(self, app):
[25-Jun-2009 14:31:38] <rocket> self.installMenuItems(app.zport.dmd)
[25-Jun-2009 14:31:53] <rocket> ZenPackBase.install(self, app)
[25-Jun-2009 14:31:57] <rocket> for example ..
[25-Jun-2009 14:32:20] <venturaville> helpful
[25-Jun-2009 14:36:01] <rocket> venturaville: make sense?
[25-Jun-2009 14:36:08] <venturaville> so does that all preclude the factory_information in the class.......
[25-Jun-2009 14:36:26] <rocket> venturaville: ?
[25-Jun-2009 14:36:42] <venturaville> class Site(Device):
[25-Jun-2009 14:36:42] <venturaville> "Web Site Device"
[25-Jun-2009 14:36:42] <venturaville> factory_type_information = (
[25-Jun-2009 14:36:42] <venturaville> {
[25-Jun-2009 14:36:51] <venturaville> blah blah .... set actions ....
[25-Jun-2009 14:36:59] <venturaville> never mind I think I answered my own question
[25-Jun-2009 14:37:05] <rocket> ok
[25-Jun-2009 14:37:42] <rocket> venturaville: it has nothing to do with set actions etc
[25-Jun-2009 14:37:58] <rocket> venturaville: its all part of the Zenpack setup via its __init__.py
[25-Jun-2009 14:38:23] <rocket> eg ~/ZenPacks/ZenPacks.community.AIX/ZenPacks/community/AIX/__init__.py
[25-Jun-2009 14:45:14] <venturaville> no dice for the menu showing up though it complains less
[25-Jun-2009 14:46:14] <rocket> venturaville: complaining at all?
[25-Jun-2009 14:46:28] <venturaville> ERROR:Zope:manage_beforeDelete() threw
[25-Jun-2009 14:46:28] <venturaville> Traceback (most recent call last):
[25-Jun-2009 14:46:28] <venturaville> File "/opt/zenoss/zenoss/lib/python/OFS/ObjectManager.py", line 332, in manage_beforeDelete
[25-Jun-2009 14:46:28] <venturaville> object.manage_beforeDelete(item, container)
[25-Jun-2009 14:46:28] <venturaville> File "/opt/zenoss/zenoss/Products/ZenRelations/ToOneRelationship.py", line 145, in manage_beforeDelete
[25-Jun-2009 14:46:29] <venturaville> self._remoteRemove()
[25-Jun-2009 14:46:31] <venturaville> File "/opt/zenoss/zenoss/Products/ZenRelations/ToOneRelationship.py", line 97, in _remoteRemove
[25-Jun-2009 14:46:33] <venturaville> remoteRel._remove(self.__primary_parent__)
[25-Jun-2009 14:46:35] <venturaville> File "/opt/zenoss/zenoss/Products/ZenRelations/ToManyRelationship.py", line 113, in _remove
[25-Jun-2009 14:46:37] <venturaville> raise ObjectNotFound(
[25-Jun-2009 14:46:39] <venturaville> ObjectNotFound: Object with id SiteVIPsData not found on relation packables
[25-Jun-2009 14:46:41] <venturaville> ugly
[25-Jun-2009 14:46:54] <venturaville> that is on an install
[25-Jun-2009 14:48:13] <rocket> did you put a self.removeMenuItems(dmd)
[25-Jun-2009 14:48:18] <rocket> in def installMenuItems(self, dmd):
[25-Jun-2009 14:48:22] <rocket> if so remove it
[25-Jun-2009 14:48:37] <rocket> or wrap it in try: except: block
[25-Jun-2009 14:54:07] <venturaville> much better
[25-Jun-2009 14:54:14] <venturaville> no errors at least
[25-Jun-2009 14:54:47] <venturaville> no menu shows up though
[25-Jun-2009 14:54:49] <venturaville> frak
[25-Jun-2009 14:55:00] <rocket> ?
[25-Jun-2009 14:55:24] <venturaville> it shows up in more menu items
[25-Jun-2009 14:55:29] <venturaville> but not on the device itself now
[25-Jun-2009 14:55:34] <venturaville> no errors generated on install
[25-Jun-2009 14:56:13] <rocket> bbiab have a meeting
[25-Jun-2009 14:57:23] <rocket> ok have 5 mins
[25-Jun-2009 14:57:51] <rocket> venturaville: is that not the right location or you need it in that place and others
[25-Jun-2009 15:04:30] <venturaville> which right location are you refering to?
[25-Jun-2009 15:04:53] <venturaville> what's your email?
[25-Jun-2009 15:05:08] <venturaville> might be easier to send you the zip file then paste random snippets
[25-Jun-2009 15:22:55] <venturaville> I dropped it here for now: http://nicklay.com/public/ZenPacks.Turner.HostMap-1.18-py2.4.egg
[25-Jun-2009 16:17:42] <UNIX_USENIX> Is there a way to tell zensyslog to never start?
[25-Jun-2009 16:18:05] <UNIX_USENIX> I already have splunk running on the same host and port and I would much rather have splunk handling those issues.
[25-Jun-2009 16:40:39] <rmatte> UNIX_USENIX: yes, one second
[25-Jun-2009 16:41:07] <rmatte> http://blog.zenoss.com/2009/01/27/tip-of-the-month-running-only-some-daemons/
[25-Jun-2009 16:41:09] <rmatte> voila
[25-Jun-2009 16:41:22] <rmatte> and don't forget to put zenjobs in the list, it's not in the list there because that predates it
[25-Jun-2009 16:41:59] <UNIX_USENIX> Excellent, thanks for the link and advice
[25-Jun-2009 16:42:29] <rmatte> no problem
[25-Jun-2009 16:42:39] <rmatte> I use it to stop zenmodeler myself
[25-Jun-2009 16:42:58] <rmatte> since I don't want it remodelling large groups of servers and network devices on a regular basis
[25-Jun-2009 18:51:20] RoninX341 is now known as etank
[25-Jun-2009 19:28:47] <mjbrooks> Is there a place where there is a comparison of features? I'm looking for info on zenoss enterprise vs. zenoss community vs. nagios
[26-Jun-2009 00:00:42] [disconnected at Fri Jun 26 00:00:42 2009]
[26-Jun-2009 00:00:42] [connected at Fri Jun 26 00:00:42 2009]
[26-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[26-Jun-2009 09:32:09] <rocket> venturaville: ping
[26-Jun-2009 09:38:34] <venturaville> rocket: pong
[26-Jun-2009 09:39:05] <rocket> so I was able to get your zenpack to install
[26-Jun-2009 09:39:18] <rocket> havent figured out why the two lines I had to comment out arent working
[26-Jun-2009 09:39:35] <venturaville> yep ... then you just need to create a site in Devices/Site (use none modeler)
[26-Jun-2009 09:39:45] <venturaville> create some IPs in Devices/VIP
[26-Jun-2009 09:40:05] <venturaville> and in zVIPNameList in Site Device add in list of IPs......
[26-Jun-2009 09:40:27] <venturaville> if it was working then you should be able to get that list in the VIPs tab on the Site Device
[26-Jun-2009 09:40:34] <venturaville> (or so I am trying to get it to) :-)
[26-Jun-2009 09:41:17] <venturaville> someone else here is working on a fetcher for fetching load balancer configs to push into it
[26-Jun-2009 09:41:44] * rocket is confused on the devices/site step
[26-Jun-2009 09:41:52] <rocket> venturaville: how are you doing that?
[26-Jun-2009 09:42:08] <venturaville> add device......
[26-Jun-2009 09:42:42] <rocket> so a site is a device in this case?
[26-Jun-2009 09:42:46] <venturaville> yeah
[26-Jun-2009 09:43:03] <rocket> confusing from a ui perspective but it works ..
[26-Jun-2009 09:43:04] <venturaville> i.e. www.cnn.com
[26-Jun-2009 09:43:14] <venturaville> it makes more sense for what I am planning later
[26-Jun-2009 09:44:00] <venturaville> i.e. tag a group (like a customer) to each site device, and then do rollup reporting for traffic, power, CPU, memory, disk elsewhere.....
[26-Jun-2009 09:44:48] <venturaville> hopefully I can make this generic enough to be useful for others
[26-Jun-2009 09:45:27] <rocket> venturaville: to get it to install I had to comment out these two lines in init
[26-Jun-2009 09:45:31] <rocket> #sitedc._setProperty('zPythonClass', 'ZenPacks.Turner.HostMap.Site')
[26-Jun-2009 09:45:39] <rocket> #vipdc._setProperty('zPythonClass', 'ZenPacks.Turner.HostMap.VIP')
[26-Jun-2009 09:45:55] <venturaville> yeah I think _setProperty doesn't like that it is set already in objects.xml
[26-Jun-2009 09:45:59] <rocket> it looks like they arent needed as the settings must also be in the object.xml for it
[26-Jun-2009 09:47:23] <rocket> venturaville: when I tried to add an ip to vip i got this
[26-Jun-2009 09:47:40] <rocket> /ZenPacks.Turner.HostMap-1.18-py2.4.egg/ZenPacks/Turner/HostMap/VIP.py", line 11, in VIP _properties = Device._properties + ( NameError: name '_kw' is not defined
[26-Jun-2009 09:50:01] <venturaville> what I get for cutting and pasting.........
[26-Jun-2009 09:52:15] <rocket> replaced the lines with
[26-Jun-2009 09:52:17] <rocket> dict(id='vipSLBHealthcheck',type='string', 'mode':'w'),
[26-Jun-2009 09:52:17] <rocket> dict(id='vipExpectedResponse', type='string', 'mode':'w'),
[26-Jun-2009 09:52:42] <venturaville> _kw = dict(mode='w')
[26-Jun-2009 09:54:17] <rocket> retesting ..
[26-Jun-2009 09:56:03] <venturaville> I'm a dope
[26-Jun-2009 09:56:18] <venturaville> just realized I was dropping the link in the 'more' menu instead of making it a tab
[26-Jun-2009 09:56:22] <venturaville> no wonder I couldn't find it
[26-Jun-2009 09:57:02] <rocket> and you have an issue with the vips tab
[26-Jun-2009 09:57:07] <venturaville> yep
[26-Jun-2009 09:57:23] <venturaville> working on those now ... now that I can find them :-P
[26-Jun-2009 09:57:40] <rocket> venturaville: ok .. so you are good to go for now then?
[26-Jun-2009 09:57:53] <venturaville> I think so... I'm going to plug away at this for a bit
[26-Jun-2009 09:57:57] <venturaville> thanks for the assist
[26-Jun-2009 09:58:01] <rocket> np
[26-Jun-2009 10:28:16] <venturaville> maybe I am missing something but is the modeler the only place where a device class gets loaded from?
[26-Jun-2009 10:46:53] <venturaville> rocket: finally restart the zenoss stack ... and it just started working *sigh*
[26-Jun-2009 10:53:16] <rocket> so you are all good?
[26-Jun-2009 10:53:43] <venturaville> almost ... I'm in template debugging mode ... so at least I am making progress now :-)
[26-Jun-2009 10:54:08] <rocket> Thats great
[26-Jun-2009 11:15:16] <DarthPointer> after upgrading my Rpath Vmware image to 2.4 I'm now seeing extremely high cpu usage; top on the system just shows zenoss / python
[26-Jun-2009 11:15:34] <DarthPointer> thoughts?
[26-Jun-2009 11:17:26] <mrayzenoss> try 'su - zenoss' and do 'zenmigrate --step=JobManager'
[26-Jun-2009 11:23:12] <DarthPointer> thank you; I'll give it a shot. I read on a forum about how quickly zenoss becomes "production"; we're going to migrate it to a dedicated box
[26-Jun-2009 11:23:33] <mrayzenoss> probably a good idea
[26-Jun-2009 11:47:51] <DarthPointer> mrayzenoss; fwiw, that seems to have cleared up the issue
[26-Jun-2009 11:48:21] <mrayzenoss> not sure what the root cause is, but zenmigrate seems to clear it up for some people
[26-Jun-2009 11:49:02] <DarthPointer> The process looks like it had been launched over & over again; there were 20+ zenoss / python processes each using ~5% of the processor
[26-Jun-2009 11:51:20] <DarthPointer> thank you very much, btw
[26-Jun-2009 11:51:37] <mrayzenoss> np
[26-Jun-2009 12:07:32] [disconnected at Fri Jun 26 12:07:32 2009]
[27-Jun-2009 00:00:42] [connected at Sat Jun 27 00:00:42 2009]
[27-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[27-Jun-2009 06:34:12] HKhan_ is now known as HKhan
[27-Jun-2009 23:53:32] <mjbrooks> Does Zenoss use it's own copy of Apache & MySQL or will it use the ones already installed on the system?
[28-Jun-2009 00:00:42] [disconnected at Sun Jun 28 00:00:42 2009]
[28-Jun-2009 00:00:42] [connected at Sun Jun 28 00:00:42 2009]
[28-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[28-Jun-2009 14:00:06] <etank> mjbrooks: i know it uses zope. not sure if apache is in the picture or not
[29-Jun-2009 00:00:42] [disconnected at Mon Jun 29 00:00:42 2009]
[29-Jun-2009 00:00:42] [connected at Mon Jun 29 00:00:42 2009]
[29-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[29-Jun-2009 03:48:14] <Younger> my Zenoss does not create an event for reboot. does anyone know how to enable it?
[29-Jun-2009 04:59:28] <abdo> hi all
[29-Jun-2009 05:00:14] <abdo> I want to add a aprocess details similar to RRD graphs in OSPRocess.
[29-Jun-2009 05:03:37] <abdo> when I start apache I want to see a process details numbers upon the process like RRD graphs
[29-Jun-2009 05:06:34] <abdo> Anybody here?
[29-Jun-2009 07:15:46] <etank> is there a way to filter out an event that happens during a specific time of day without using a maintenance window?
[29-Jun-2009 07:16:45] <etank> we reindex our database at 5:30am and today I got over 4000 events from our web server trying to do an insert (thank God for deduplication)
[29-Jun-2009 08:12:19] <eidolon> hi folks - we're seeing some odd alerts coming from our zenoss monitors, particularly about disk space:
[29-Jun-2009 08:12:21] <eidolon> [zenoss] bos-aspac-tnz-app04 Disk space low: 210.4% used (-410.53 GB free)
[29-Jun-2009 08:28:16] <etank> that does look odd eidolon
[29-Jun-2009 08:28:25] <eidolon> yeah
[29-Jun-2009 08:29:47] <Jontler> id like to get 210 percent out of my drives
[29-Jun-2009 08:51:55] <abdo> hi all
[29-Jun-2009 08:52:50] <abdo> i want to retrieve all apache process and mysql process by process instead of RRD graph
[29-Jun-2009 08:53:26] <abdo> I used the zen-core package for apache and mysql monitoring
[29-Jun-2009 08:54:18] <abdo> but it list data on the RRD Graph, I want to list these data on OS process tab instead of Prof tab
[29-Jun-2009 08:56:06] <abdo> or anybody tell me about documentation part should I read.
[29-Jun-2009 08:58:02] <abdo> anybody here? anyone answer me to make sure that I am not alone!
[29-Jun-2009 08:59:52] <eidolon> so, next question. is there a way to customize a 'per-alert' message? ala, if we have an alert for low disk space, can i add something to the email generated saying something like "Contact bob in accounting to clean this up." - I don't want to edit my personal alerting rules, since that's global to all msgs i get.
[29-Jun-2009 09:09:42] <venturaville> eidolon: we do something similar to that... you can set a variable inside of an event transform (for adding a custom message for instance), and then picking up that variable in the alerting rules
[29-Jun-2009 09:10:08] <eidolon> hmmm
[29-Jun-2009 09:10:15] <eidolon> that's not a bad idea.
[29-Jun-2009 09:10:41] <eidolon> the alerts are generally being geneated by nagios checkers... hwo would i embed a var in there? somethingon the CL?
[29-Jun-2009 09:10:56] <venturaville> alerts or events?
[29-Jun-2009 09:11:02] <venturaville> there is a big distinction in zenoss
[29-Jun-2009 09:11:24] <venturaville> events = just that ... alerts = messages sent outbound...
[29-Jun-2009 09:12:24] <venturaville> if it is sending out an event, then you can write an event mapping rule to match it, and then put something in event transform to set a message
[29-Jun-2009 09:12:26] <venturaville> like:
[29-Jun-2009 09:12:37] <venturaville> evt.appendthismessage = 'message to append'
[29-Jun-2009 09:12:59] <venturaville> then in alerts rules, you would just have to check for that variable
[29-Jun-2009 09:13:05] <eidolon> well, the alerts are being generated by the events. so i'd have to add th emessage to the event, to have it embedded in the alert.
[29-Jun-2009 09:13:06] * eidolon gets the terms.
[29-Jun-2009 09:13:12] <venturaville> k
[29-Jun-2009 09:15:03] <eidolon> hmmmmm.
[29-Jun-2009 09:15:06] * eidolon makes notes.
[29-Jun-2009 09:16:02] <eidolon> thanks, i'll experiment.
[29-Jun-2009 12:07:01] <etank> will something as simple as http://dpaste.com/61196/ work as a transform?
[29-Jun-2009 12:25:53] <jb> sup everyone..
[29-Jun-2009 12:28:32] <rmatte> sup
[29-Jun-2009 12:28:48] <jb> just looking at the 2.4.2 notes
[29-Jun-2009 12:28:51] <jb> been out of town for a week..
[29-Jun-2009 12:31:12] <rmatte> cool
[29-Jun-2009 12:32:04] <jb> Process monitoring for Linux and AIX SSH hosts currently is disabled. Until re-enabled, process modeling
[29-Jun-2009 12:32:05] <jb> should be disabled (so that no processes appear when the device is modeled). (Internal Defect 1695)
[29-Jun-2009 12:32:06] <jb> hm
[29-Jun-2009 12:37:08] <rmatte> I'm trying to restore a Products directory that I stupidly accidentally deleted
[29-Jun-2009 12:37:14] <rmatte> almost done
[29-Jun-2009 13:16:25] <Schwartz> afternoon all
[29-Jun-2009 13:23:27] <Schwartz> Anyone knowledgeable with thresholds and expressions?
[29-Jun-2009 13:24:36] <rocket> jb: ping
[29-Jun-2009 13:25:52] <rocket> Schwartz: ask the question and hopefully someone will know ..
[29-Jun-2009 13:29:47] <jb> rocket: sup
[29-Jun-2009 13:29:58] <rocket> jb: my aix zenpack is posted
[29-Jun-2009 13:30:04] <rocket> jb: and the thresholds as well
[29-Jun-2009 13:30:09] <jb> yeah i saw
[29-Jun-2009 13:30:12] <jb>)
[29-Jun-2009 13:30:18] <rocket> jb: point, predictive, range ..
[29-Jun-2009 13:30:31] <jb> have you tried the apar to fix snmp?
[29-Jun-2009 13:30:42] <Schwartz> My Python's a bit thin, and I'm trying to understand what variables are available to the min/max value fields for thresholds. The documentation just refers to using a python expression. Some of the bits I've been able to find for examples are for disk percentage utilization. However, I'm unable to find what is being referred to by here.totalBlocks etc, doesn't match up with RRD datasource as far as I can tell. In that same line I'm trying to figure out
[29-Jun-2009 13:30:44] <rocket> not yet
[29-Jun-2009 13:30:59] <rocket> jb: its on my list .. but not a high priority yet
[29-Jun-2009 13:31:18] <jb> ugh aix snmpd
[29-Jun-2009 13:31:23] <jb> so much configuration
[29-Jun-2009 13:32:11] <twm1010> Afternoon all.
[29-Jun-2009 13:32:23] <rmatte> Schwartz: here.totalBlocks is collected by a collector plugin
[29-Jun-2009 13:32:58] <rocket> jb: yes that is a problem with aix snmp ;P
[29-Jun-2009 13:33:00] <jb> rocket: how did you configure snmpd?
[29-Jun-2009 13:33:09] <jb> i was reading thru that cacti thread
[29-Jun-2009 13:33:22] <rmatte> I thought the AIX pack was just drop and go
[29-Jun-2009 13:33:22] <rocket> jb: I followed that thread ..
[29-Jun-2009 13:33:32] <jb> what a pain for 50+ boxes
[29-Jun-2009 13:33:42] <rmatte> oh, you mean actually configuring each box?
[29-Jun-2009 13:33:45] <jb> i guess I could write a script...
[29-Jun-2009 13:33:46] <rocket> rmatte: it is except aix snmp suxors to setup to get the cpu info
[29-Jun-2009 13:33:48] <jb> yeah
[29-Jun-2009 13:33:57] <rmatte> I'd write a script, I never do mass changes like that by hand
[29-Jun-2009 13:34:00] <rmatte> waste of time
[29-Jun-2009 13:34:07] <jb> yeah
[29-Jun-2009 13:34:13] <jb> you havent written one yet though have you?
[29-Jun-2009 13:34:32] <rmatte> nope
[29-Jun-2009 13:34:34] <rmatte> but I could
[29-Jun-2009 13:34:41] <rocket> rmatte: I might be able to work on a ssh based aix zenpack but if I do some work with zenoss it would probably be an updated enterprise zenpack
[29-Jun-2009 13:34:56] <jb> ssh aix!
[29-Jun-2009 13:35:01] <rmatte> enterprise zenpack, ewwww
[29-Jun-2009 13:35:12] <rmatte> I vote community WMI pack
[29-Jun-2009 13:35:45] <rocket>
[29-Jun-2009 13:35:53] <jb> so there is talk of an updated aix ssh zenpack?
[29-Jun-2009 13:36:04] <jb> to incorporate the same features as your snmp?
[29-Jun-2009 13:36:07] <Schwartz> Ah. Is it possible ot have the threshold use a datasource in an rrd table? This is more for a bit of SNMP MIB that's outside of the normal zenoss world.
[29-Jun-2009 13:36:25] <rocket> well there is talk of me doing some zenpacks for them .. and if I do that I might not be able to do some community stuff
[29-Jun-2009 13:36:35] <jb> right
[29-Jun-2009 13:37:05] <rocket> and there is some talk of me working on aix but nothing nailed down at all yet
[29-Jun-2009 13:37:11] <rmatte> Schwartz: you'd have to collect the data via an snmp datapoint (figure out what OID it is and create it), then create a threshold against that value
[29-Jun-2009 13:37:30] <jb> im still thinking that SSH is the way to go with AIX
[29-Jun-2009 13:37:33] <rocket> jb: but I am sure there are some who would like an aix ssh zenpack
[29-Jun-2009 13:37:42] <jb> well enterprise has one..
[29-Jun-2009 13:37:49] <jb> just none of the features that you have in your SNMP pack
[29-Jun-2009 13:38:10] <rocket> jb: are my features what you would like in the enterprise one?
[29-Jun-2009 13:38:14] <jb> yes
[29-Jun-2009 13:38:17] <jb> definetely
[29-Jun-2009 13:38:22] <jb> the lvm stuff is huge
[29-Jun-2009 13:38:37] <rocket>
[29-Jun-2009 13:39:04] <jb> it already does the CPU and stuff (basic)
[29-Jun-2009 13:39:16] <jb> but none of your relationships
[29-Jun-2009 13:39:23] <jb> or print queues, paging
[29-Jun-2009 13:39:50] <rocket> or lpar info etc ..
[29-Jun-2009 13:39:58] <jb> right
[29-Jun-2009 13:40:01] <rocket> jb: have you loaded my zenpack?
[29-Jun-2009 13:40:03] <jb> no
[29-Jun-2009 13:40:06] <jb> not yet
[29-Jun-2009 13:40:25] <jb> i need to spin it up in dev
[29-Jun-2009 13:40:37] <rocket> I would spin it up there myself also
[29-Jun-2009 13:40:49] <rocket> jb: I wouldnt trust myself either
[29-Jun-2009 13:41:31] <jb> haha
[29-Jun-2009 13:41:33] <jb> its not that
[29-Jun-2009 13:41:53] <jb> i just don't want to conflict with any of my other templates
[29-Jun-2009 13:42:04] <jb> just a precaution
[29-Jun-2009 13:42:06] <rocket> jb: I know I was just pulling your chain
[29-Jun-2009 13:42:16] <jb> but yeah.. aix ssh would be ideal at least for me
[29-Jun-2009 13:42:21] <jb> just because no further configuration
[29-Jun-2009 13:42:23] <jb> and ssh is stable
[29-Jun-2009 13:42:26] <jb>
[29-Jun-2009 13:42:52] <rocket> jb: only issue I can see is if I run into a command that requires root priveledges to gather the info
[29-Jun-2009 13:43:01] <jb> yeah
[29-Jun-2009 13:43:05] <jb> can you think of any?
[29-Jun-2009 13:43:07] <rocket> jb: otherwise it should be a nice leisurely walk in the park ..
[29-Jun-2009 13:43:14] <rocket> jb: off the top of my head no ..
[29-Jun-2009 13:43:26] <jb> sudo
[29-Jun-2009 13:43:31] <jb> if we must
[29-Jun-2009 13:43:46] <rocket> yes but I would hate to have that etc as a setup requirement
[29-Jun-2009 13:43:52] <jb> uyeah
[29-Jun-2009 13:43:53] <jb> but
[29-Jun-2009 13:44:01] <jb> well
[29-Jun-2009 13:44:03] * rocket is thinking basic setuid scripts if we must
[29-Jun-2009 13:44:05] <jb> the snmpd configuration is even worse!
[29-Jun-2009 13:44:07] <jb> yeah
[29-Jun-2009 13:44:29] <rocket> eg it reads .. no inputs for the script and just dumps parsed info etc
[29-Jun-2009 13:44:49] <pupmini> hello
[29-Jun-2009 13:44:50] <rocket> that probably has the best chance of being accepted by adminds etc ..
[29-Jun-2009 13:44:55] <jb> yeah
[29-Jun-2009 13:44:58] <jb> that sounds better
[29-Jun-2009 13:47:50] <jb> im not sure if anything would even require root though
[29-Jun-2009 13:47:53] <jb> but i'd have to check
[29-Jun-2009 13:48:16] <rocket> yeah again it would be a minor annoyance if it needed root for something
[29-Jun-2009 13:48:32] <jb> solid CPU, memory, IO performance monitoring with LVM relationships would be an awesome start
[29-Jun-2009 13:48:37] <pupmini> has anyone here used WMI with zenoss core?
[29-Jun-2009 13:49:05] <jb> and im pretty sure all of that could be done without root
[29-Jun-2009 13:49:53] <jb> IO per hdisk, and combined IO maybe
[29-Jun-2009 13:56:05] <twm1010> pupmini: yes
[29-Jun-2009 13:56:07] <twm1010> extensively
[29-Jun-2009 13:56:40] <twm1010> I took the wmi community zenpack, and just used it as a template, did all sorts of monitoring, exchange, terminal services, a bunch of others
[29-Jun-2009 13:56:54] <pupmini> ah cool
[29-Jun-2009 13:57:01] <pupmini> i am reading the docs
[29-Jun-2009 13:57:18] <pupmini> and it appears that it does not popular any of the device information automatically...is that correct?
[29-Jun-2009 13:57:34] <pupmini> populate*
[29-Jun-2009 13:58:26] <rocket> jb: its all feasable .. just a matter of time and what my legal obligations end up being
[29-Jun-2009 14:05:36] <jb> the ACL's
[29-Jun-2009 14:12:35] <Schwartz> rmatte: Yep, I am already collecting via SNMP the various stats and even graphing them, but the problem that the maximum value of a particular host is dependent on another SNMP value. Something that would be similar would to use the file system one as an example. Assume that there's no collector variable for the number of blocks, so you're collecting both the used blocks and total blocks via snmp. I would like to have a threshold for 70% of its capacit
[29-Jun-2009 14:37:35] <rmatte> Schwartz: the way I personally do it is to write a script and use it as a datapoint
[29-Jun-2009 14:37:42] <venturaville> how do you do the variable substitution in objects.xml to make a link to your own check_* script when the zenpack directory is versioned..........?
[29-Jun-2009 14:37:44] <rmatte> the script would calculate the percentage of utilization
[29-Jun-2009 14:37:54] <rmatte> then you just set the threshold at 90
[29-Jun-2009 14:38:35] <Schwartz> how are you referring to the datapoint that you have the script populating in the threshold?
[29-Jun-2009 14:39:33] <rmatte> your script just has to output like: OK|DataPoint1=value DataPoint2=value WhateverYouWant=value
[29-Jun-2009 14:39:33] <rmatte> then you add it as a command based datapoint
[29-Jun-2009 14:39:37] <rmatte> and create datapoints underneath it with those names
[29-Jun-2009 14:39:57] <rmatte> download James Dastrup's WMI Performance Zenpack is an example and take a look at how he set that up
[29-Jun-2009 14:40:02] <rmatte> and then you'll understand
[29-Jun-2009 14:40:21] <kgoedtel> maybe one day they will let you use snmp datapoints in the threshold expression
[29-Jun-2009 14:40:30] <kgoedtel> i've been wanting that for years :[
[29-Jun-2009 14:40:42] <rmatte> yeh, that would definitely be nice
[29-Jun-2009 14:41:24] <Schwartz> lemme give that a go..
[29-Jun-2009 14:44:25] <Schwartz> Heh, it wants windows stuff.
[29-Jun-2009 14:44:35] <rmatte> wants windows stuff?
[29-Jun-2009 14:44:45] <rmatte> what does?
[29-Jun-2009 14:45:12] <Schwartz> wmiperformancemonitor.zip WARNING:zen.AddToPack:Unable to find context path /zport/dmd/Devices/Server/Windows/rrdTemplates (line 4 ?) for WMI Performance Monitors
[29-Jun-2009 14:45:22] <Schwartz> though I did delete the server/windows group
[29-Jun-2009 14:45:32] <rmatte> that's your problem then
[29-Jun-2009 14:46:02] <rmatte> it's a Window monitoring Zenpack so it expects that group to be there
[29-Jun-2009 14:46:13] <rmatte> but I'm just having you checkout the template so you get an idea of how you have to set it up
[29-Jun-2009 14:46:19] <Schwartz> right
[29-Jun-2009 14:49:43] <rmatte> ah crap, I screwed up one of my scripts this morning with a modification, just noticed, gotta fix this...
[29-Jun-2009 14:57:17] <rmatte> there, fixed
[29-Jun-2009 14:57:43] <rmatte> example of output:
[29-Jun-2009 14:57:44] <rmatte> zenoss@D1Z01:/usr/local/zenoss/zenoss/Products/SNMPWindowsPerfMonitor/libexec$ ./wincpu.sh CCADB1 community-string
[29-Jun-2009 14:57:45] <rmatte> OK|CPU1=2 CPU2=93 CPU3=4 CPU4=2 Total=25
[29-Jun-2009 14:59:38] <kgoedtel> actually hrmmm, maybe you can use the data points like this in threshold
[29-Jun-2009 14:59:42] <Schwartz> Okay, I see what you're saying now. Would be so much nicer if we could use something like (here.TotalBlocks * 100)*.9 where TotalBlocks was a data source itself.
[29-Jun-2009 14:59:52] <kgoedtel> (here.getRRDValue('data_point')) * 0.70
[29-Jun-2009 14:59:56] <Schwartz> er minus the *100
[29-Jun-2009 15:00:05] <Schwartz> hmm.
[29-Jun-2009 15:00:11] <Schwartz> lemme try that.
[29-Jun-2009 15:00:43] <kgoedtel> i'm trying it for one of my alerts
[29-Jun-2009 15:01:22] <rmatte> actually, that could definitely work
[29-Jun-2009 15:01:27] <rmatte> never thought of that
[29-Jun-2009 15:01:30] <twm1010> are you just trying to set a disk space threshold?
[29-Jun-2009 15:01:48] <rmatte> no, he's trying to thresholds against some other RRD value
[29-Jun-2009 15:01:52] <rmatte> he was using disk space as an example
[29-Jun-2009 15:01:54] <Schwartz> technically, no. just using that as an easily understandable analogy.
[29-Jun-2009 15:02:03] <Schwartz> what he said
[29-Jun-2009 15:02:13] <kgoedtel> something that doesn't use a collector (snmp dp)
[29-Jun-2009 15:02:35] <twm1010> ah, just looks up currently stored RRD values, and goes off that?
[29-Jun-2009 15:02:44] <kgoedtel> i'm hoping
[29-Jun-2009 15:02:46] <Schwartz> it is coming in via snmp
[29-Jun-2009 15:02:47] <Schwartz> yeah
[29-Jun-2009 15:03:57] <twm1010> hrmm...
[29-Jun-2009 15:04:14] <twm1010> i wonder, can you do something similar to what was in the blog post a few weeks back?
[29-Jun-2009 15:04:44] <twm1010> @ http://blog.zenoss.com/2009/06/24/tip-of-the-month-tracking-file-system-growth/
[29-Jun-2009 15:08:46] <Schwartz> I wonder what the 'built-in' datasource type is and how its used.
[29-Jun-2009 15:12:10] <venturaville> for internal zenoss statistics for one
[29-Jun-2009 15:12:26] <venturaville> things like work list queue length (zenhub) etc..
[29-Jun-2009 15:12:31] <rmatte> is there no way to have Critical tickets age out after an hour?
[29-Jun-2009 15:13:07] <rmatte> seems to be a limitation of the option
[01-Jun-2009 07:37:20] <venturaville> odd ... irc used to be blocked from work and now it works ...sweet :-)
[01-Jun-2009 07:46:42] <Appiah> This is the second server I upgrade to 2.4.1 , the progressbar goes fast from 0 to 100%
[01-Jun-2009 07:46:49] <Appiah> then it just sits there for like an hour
[01-Jun-2009 09:00:44] <rocket> morning matt
[01-Jun-2009 09:07:38] <jb> yawn
[01-Jun-2009 09:09:34] <mrayzenoss> good morning
[01-Jun-2009 09:10:12] <rocket> mrayzenoss: so I started working on an ssh based zenpack .. but the unit test documentation leaves much to be desired
[01-Jun-2009 09:10:29] <rocket> mrayzenoss: and I thought there would be an example in the linux ssh zenpack but no such luch
[01-Jun-2009 09:10:32] <rocket> luck
[01-Jun-2009 09:10:36] <mrayzenoss> oh yeah? Feel free to make notes where it sucks and I'll tell Brian to fix it
[01-Jun-2009 09:10:46] <mrayzenoss> there are no tests in the Linux Monitor?
[01-Jun-2009 09:11:07] <rocket> mrayzenoss: not that I know of .. unless I accidentally have an old version
[01-Jun-2009 09:11:16] <mrayzenoss> doubt it, let me poke around
[01-Jun-2009 09:11:39] <rocket> mrayzenoss: I was looking for it in there because I figured it would be the most reasonable place to look for an example to fill in the gaps I was having
[01-Jun-2009 09:12:09] <rocket> mrayzenoss: I mean the development of the zenpack without the unittests is going fine etc .. just was trying out that unit test framework ...
[01-Jun-2009 09:18:06] <mrayzenoss> yeah, he stuck all the tests in the EnterpriseLinux ZenPack, I'm going to see about moving some into Linux Monitor
[01-Jun-2009 09:24:23] <jb> i was goign to say
[01-Jun-2009 09:24:31] <jb> the tests exist in the enterprise pack..
[01-Jun-2009 09:25:13] <rocket> ?
[01-Jun-2009 09:25:13] <rocket> oops
[01-Jun-2009 09:57:28] <jb> rocket, the ssh zenpack.. is it for AIX or something else?
[01-Jun-2009 09:57:38] <rocket> jb: gentoo first ..
[01-Jun-2009 09:57:54] <rocket> jb: getting my feet wet with an easy one first etc ..
[01-Jun-2009 10:00:28] <rocket> jb: but from what I can tell ssh zenpacks should be relatively easy to convert from snmp ones .. I just have to do more regex's etc
[01-Jun-2009 10:00:39] <jb> yeah
[01-Jun-2009 10:00:42] <jb> figure out how to parse everything
[01-Jun-2009 10:01:06] <rocket> jb: I am not having trouble with the zenpack itself .. I am still trying to figure out how these unit tests etc work
[01-Jun-2009 10:24:26] <coofamani> anyone have an idea of the context of sysUpTime in a report? tried sysUpTime, here.hw.sysUpTime, and here.os.sysUpTime.
[01-Jun-2009 10:25:07] <mrayzenoss> Appendix D of the Admin Guide, ${device/uptimeStr)
[01-Jun-2009 10:25:29] <mrayzenoss> someone just asked me about that
[01-Jun-2009 10:27:30] <coofamani> sorry, thats the string version. Im trying to get the snmp returned value in 1/100 seconds, in order to sort correctly
[01-Jun-2009 10:27:49] <coofamani> the object is sysUpTime vs uptimeStr
[01-Jun-2009 10:28:02] <coofamani> but Im having trouble referencing that in deviceQuery
[01-Jun-2009 10:29:31] <venturaville> >>> dev = devices.findDevice('mon8zen2')
[01-Jun-2009 10:29:31] <venturaville> >>> dev.sysUpTime()
[01-Jun-2009 10:29:31] <venturaville> 286156976.11000001
[01-Jun-2009 10:29:44] <venturaville> It is a call not a variable.........
[01-Jun-2009 10:29:54] <venturaville> so it would be: here.sysUpTime()
[01-Jun-2009 10:30:21] <venturaville> print '\n'.join(dir(dev))
[01-Jun-2009 10:30:23] <venturaville> is your friend............
[01-Jun-2009 10:30:37] <coofamani> nice. thanks
[01-Jun-2009 10:30:39] <coofamani> "here.sysUpTime() > 86400000"
[01-Jun-2009 10:30:53] <venturaville> do you know how to use zendmd?
[01-Jun-2009 10:30:54] <coofamani> that does what I want. I wasnt treating it as a call, had that line with the (). thanks
[01-Jun-2009 10:31:02] <coofamani> zendmd = sorta
[01-Jun-2009 10:31:07] <coofamani> need to spend more time in there
[01-Jun-2009 10:31:13] <venturaville> just start it up: and do
[01-Jun-2009 10:31:20] <venturaville> here = device.findDevice('nameofsomedevice')
[01-Jun-2009 10:31:26] <venturaville> and then you will pretty much be there......
[01-Jun-2009 10:31:48] <coofamani> then use dir to see all the available methods? nice
[01-Jun-2009 10:32:13] <venturaville> yeah ... thank python for that nicety
[01-Jun-2009 10:32:34] <mrayzenoss> feel free to send in reports as ZenPacks
[01-Jun-2009 10:35:21] <coofamani> toying with that now. thanks venturaville. this will make life better.
[01-Jun-2009 10:36:05] <venturaville> >>> callable(dev.sysUpTime)
[01-Jun-2009 10:36:05] <venturaville> True
[01-Jun-2009 10:36:05] <venturaville> >>> callable(dev.sysUpTime())
[01-Jun-2009 10:36:05] <venturaville> False
[01-Jun-2009 10:36:12] <venturaville> another nice thing to use ..........
[01-Jun-2009 11:44:13] <mrayzenoss> For those of you using the HP ProLiant Monitor ZenPack, it's been updated along with the Advanced Device Details
[01-Jun-2009 11:46:06] <mrayzenoss> and now I can finally start tackling rocket's backlog
[01-Jun-2009 11:47:07] <mrayzenoss> 7 currently in the queue
[01-Jun-2009 11:47:15] <jb>
[01-Jun-2009 11:47:31] <jb> have you heard how igor's dell pack is coming
[01-Jun-2009 11:47:31] <jb> ?
[01-Jun-2009 11:48:55] <mrayzenoss> nothing yet, he just asked me for some OIDs last week and told me he was working on one like his HP one
[01-Jun-2009 11:49:02] <jb> cool..
[01-Jun-2009 11:49:40] <mrayzenoss> 8 entries in the ZenPack contest so far
[01-Jun-2009 11:49:47] <jb> are they posted?
[01-Jun-2009 11:49:51] <mrayzenoss> not counting my backlog of 7
[01-Jun-2009 11:50:10] <mrayzenoss> I was planning on doing a post this week about progress so far
[01-Jun-2009 11:50:35] <mrayzenoss> but the new ones are the 2 Nortel ones, the updates are Perfmon, SQL Server and the 3 from Egor
[01-Jun-2009 11:50:46] <mrayzenoss> and a Barracuda update
[01-Jun-2009 11:50:51] <jb> perfmon?
[01-Jun-2009 11:51:11] <mrayzenoss> He took the Perfmon one and converted it to an egg so the SQL Server could depend on it
[01-Jun-2009 11:51:16] <jb> ahh
[01-Jun-2009 11:51:24] <mrayzenoss> if you're using Enterprise you have access to zenwinperf
[01-Jun-2009 11:51:26] <jb> it basically just allows you to pull perfmon datasources like enterprise?
[01-Jun-2009 11:51:30] <mrayzenoss> yeah
[01-Jun-2009 11:51:31] <jb> yeah
[01-Jun-2009 11:51:32] <jb> k
[01-Jun-2009 11:51:37] <mrayzenoss> well, not exactly, but similar
[01-Jun-2009 11:51:51] <mrayzenoss> it shells out to a perl call
[01-Jun-2009 11:51:57] <jb> ah ok
[01-Jun-2009 11:53:33] <mrayzenoss> rocket gave me his AIX SNMP one for release
[01-Jun-2009 11:53:48] <mrayzenoss> I think you were looking for that right?
[01-Jun-2009 11:53:49] <jb> nice
[01-Jun-2009 11:53:52] <jb> yeah
[01-Jun-2009 12:02:51] <rocket> mrayzenoss: I will probably do an update for the aix one eventually as I get time .. but I figure I might as well get it out there and solicit feedback
[01-Jun-2009 12:03:49] <rocket> mrayzenoss: and I had a thought about these tests that are missing
[01-Jun-2009 12:04:15] <rocket> mrayzenoss: couldnt the zenpacks create a template one but commented out as an example that gets copied into all new zenpacks ..
[01-Jun-2009 12:04:24] <mrayzenoss> sure
[01-Jun-2009 12:04:33] <mrayzenoss> sounds like a good enhancement if you want to open a ticket
[01-Jun-2009 12:04:57] <rocket> ok .. I just figure if you are pushing for the tests .. might as well make it easier to make them .. etc
[01-Jun-2009 12:05:15] <mrayzenoss> yeah, definitely
[01-Jun-2009 12:05:33] <mrayzenoss> I've got a todo to look into the templates at some point, just haven't found the cycles
[01-Jun-2009 12:05:42] <mrayzenoss> people keep contributing too many zenpacks
[01-Jun-2009 12:06:29] <rocket>
[01-Jun-2009 12:06:35] <rocket> want me to slow down?
[01-Jun-2009 12:06:43] <mrayzenoss> nope
[01-Jun-2009 12:06:55] <mrayzenoss> it's a good problem to have
[01-Jun-2009 12:07:25] <rocket> heh
[01-Jun-2009 12:08:27] <rocket> mrayzenoss: when do you think I might be able to get access to those sample tests?
[01-Jun-2009 12:08:37] <rocket> just so I can plan my schedule
[01-Jun-2009 12:10:24] <mrayzenoss> I'll let you know when I hear from Brian
[01-Jun-2009 12:10:31] <rocket> thanks
[01-Jun-2009 12:11:19] <mrayzenoss> possibly later this week
[01-Jun-2009 12:11:37] <mrayzenoss> I'm gonna go look at the code, maybe I'll do it
[01-Jun-2009 12:12:15] <rocket> cool .. that would be great .. especially since summer of code could really use it
[01-Jun-2009 12:29:00] <venturaville> anyone know what to do about these: Only ZenPacks that have been installed in development mode can be exported.
[01-Jun-2009 12:29:45] <venturaville> its alright
[01-Jun-2009 12:30:35] <venturaville> its not right about it not being installed
[01-Jun-2009 12:47:50] <mrayzenoss> are you trying to get a ZenPack into development mode? trac-zenpacks/wiki/DevelopmentMode
[01-Jun-2009 12:47:51] <mrayzenoss> Or some other issue?
[01-Jun-2009 12:51:18] <venturaville> I think it is something else
[01-Jun-2009 12:51:36] <venturaville> I get this when I load up zenpackmgr even though the pack is not installed anymore:
[01-Jun-2009 12:51:38] <venturaville> Type: DistributionNotFound
[01-Jun-2009 12:51:38] <venturaville> Value: ZenPacks.turner.HPBladeChassis
[01-Jun-2009 12:52:01] <venturaville> I've stopped and restarted, and done a grep across the zenoss tree, but cannot find where it is referenced
[01-Jun-2009 12:52:49] <venturaville> cd /opt/zenoss ; find . | grep -i bladechassis
[01-Jun-2009 12:52:51] <venturaville> returns nothing
[01-Jun-2009 12:55:03] <venturaville> stored in the database somehow I'm guessing......
[01-Jun-2009 12:59:21] <venturaville> >>> zp = dmd.ZenPackManager.packs._getOb('ZenPacks.turner.HPBladeChassis')
[01-Jun-2009 12:59:22] <venturaville> >>> zp
[01-Jun-2009 12:59:22] <venturaville> <ZenPack at /zport/dmd/ZenPackManager/packs/ZenPacks.turner.HPBladeChassis>
[01-Jun-2009 12:59:40] <venturaville> >>> dmd.ZenPackManager.packs._delOb('ZenPacks.turner.HPBladeChassis')
[01-Jun-2009 12:59:40] <venturaville> Traceback (most recent call last):
[01-Jun-2009 12:59:40] <venturaville> File "<stdin>", line 1, in ?
[01-Jun-2009 12:59:40] <venturaville> File "/opt/zenoss/zenoss/Products/ZenRelations/ToManyRelationshipBase.py", line 87, in _delOb
[01-Jun-2009 12:59:40] <venturaville> raise NotImplementedError
[01-Jun-2009 12:59:41] <venturaville> NotImplementedError
[01-Jun-2009 12:59:44] <venturaville> do
[01-Jun-2009 12:59:45] <venturaville> doh
[01-Jun-2009 13:05:03] <venturaville> catch 22 ...
[01-Jun-2009 13:05:12] <venturaville> how to delete an object in there?
[01-Jun-2009 13:12:11] <chudler> SNMP + zenpack dev question for gurus. I have two devices, different vendors, but supporting the same MIB tree. I have a custom modeler plugin that works for one device, but not the other.
[01-Jun-2009 13:12:48] <chudler> The reason is doesnt work for the other, is that the vendors prefix the tree with their vendor string. So, in the modeler, I have something like
[01-Jun-2009 13:13:02] <chudler> GetTableMap('subentries', '.1.3.6.1.4.1.1718.3.2.5.1', subentries)
[01-Jun-2009 13:14:16] <chudler> where 1718.3 is the vendor's prefix. Unfortunately, a different vendor will use a different prefix, in this case 2606.100.1; I don't want to make a entirely new zenpack to support a identical device, just because it has a different vendor code in the MIB.
[01-Jun-2009 13:14:52] <chudler> I thought what I might do is have alternate table maps for each known vendor, but a programatic solution would be prefered.
[01-Jun-2009 13:17:10] <venturaville> fixed it finally
[01-Jun-2009 13:17:26] <venturaville> leftovers from a previous zenpack install
[01-Jun-2009 14:09:43] <jb> matt?
[01-Jun-2009 14:16:59] <mrayzenoss> yeah?
[01-Jun-2009 14:33:34] <CapriCoRN^80> hi ! i need some information. My all hosts get IP from DHCP .. i was reading a post in which someone mentioned problem that zenoss will not detect the DHCP IP ?
[01-Jun-2009 14:41:08] <mrayzenoss> Zenoss doesn't play well with DHCP
[01-Jun-2009 14:41:20] <mrayzenoss> since it monitors based on IP addresses
[01-Jun-2009 14:41:53] <mrayzenoss> that said, you can probably configure your DHCP server to assign the same addresses to machines each time they request an address
[01-Jun-2009 14:42:11] <mrayzenoss> which allows you to keep using DHCP no problem
[01-Jun-2009 14:47:45] <mrayzenoss> CapriCoRN^80: following up from my post on LinuxQuestions I see
[01-Jun-2009 14:48:49] <mrayzenoss> CapriCoRN^80: there's a webinar tomorrow at 3pm EST for Getting Started
[01-Jun-2009 15:16:10] <CapriCoRN^80> well thats hard thing
[01-Jun-2009 15:16:28] <CapriCoRN^80> i mean if the lease time of dhcp is lease then its a problem
[01-Jun-2009 15:17:07] <mrayzenoss> do you have a lot of transient users?
[01-Jun-2009 15:17:11] <mrayzenoss> like a wireless hotspot?
[01-Jun-2009 15:17:27] <CapriCoRN^80> not wireless. i got lan users
[01-Jun-2009 15:17:55] <mrayzenoss> but they're known right? If you have their MAC addresses you can give them fixed DHCP addresses, but still use DHCP
[01-Jun-2009 15:17:55] <CapriCoRN^80> and i got dhcp server which assign ip addresses to lan users
[01-Jun-2009 15:18:46] <CapriCoRN^80> well i have to go through all the mac addresses
[01-Jun-2009 15:19:00] <CapriCoRN^80> i read that post at evening time and till that i got my server up .
[01-Jun-2009 15:19:49] <CapriCoRN^80> i am difficult situtation now
[01-Jun-2009 15:20:40] <CapriCoRN^80> in *
[01-Jun-2009 15:24:53] <CapriCoRN^80> <@mrayzenoss> since it monitors based on IP addresses .. didnt get that
[01-Jun-2009 15:25:02] <CapriCoRN^80> dhcp also assign ip addresses
[01-Jun-2009 15:25:48] <rocket> CapriCoRN^80: basically if the ip address changes for a device zenoss doesnt really like it
[01-Jun-2009 15:26:11] <rocket> CapriCoRN^80: because it keys off of the ip address
[01-Jun-2009 15:26:50] <rocket> eg if you change the ip address of a workstation eg a flipflop .. then zenoss wont know about it and the data it collects will be applied to the wrong entry in zenosses databases
[01-Jun-2009 15:27:11] <CapriCoRN^80> rocket: what if the lease time of dhcp server is 10 days ?
[01-Jun-2009 15:27:22] <CapriCoRN^80> or may be 4 days
[01-Jun-2009 15:27:23] <rocket> CapriCoRN^80: same issue exists
[01-Jun-2009 15:27:30] <CapriCoRN^80> rocket: strange
[01-Jun-2009 15:27:42] <CapriCoRN^80> 4 days is not flipflop
[01-Jun-2009 15:28:12] <rocket> CapriCoRN^80: I did not say its flapping
[01-Jun-2009 15:28:38] <rocket> CapriCoRN^80: I was just giving an example of data going to the wrong entry in the database
[01-Jun-2009 15:29:05] <CapriCoRN^80> rocket: ok
[01-Jun-2009 15:29:22] <CapriCoRN^80> rocket: i am in trouble
[01-Jun-2009 15:30:08] <rocket> CapriCoRN^80: I am assuming zenoss was really designed with a static datacenter in mind initially
[01-Jun-2009 15:30:13] <CapriCoRN^80> well dhcp method is world wide use
[01-Jun-2009 15:30:52] <CapriCoRN^80> rocket: i am not a big engineer but i guess this is big problem in zenoss
[01-Jun-2009 15:31:47] <rocket> it could be .. I havent really run into it as our datacenter is static
[01-Jun-2009 15:33:13] <CapriCoRN^80> how many users u have ?
[01-Jun-2009 15:33:18] <mrayzenoss> CapriCoRN^80: it's a known limitation, but my point was in your dhcpd.conf, you can use a host statement to get around it: some really good scenes which made other
[01-Jun-2009 15:33:32] <mrayzenoss> oops, bad copy and paste
[01-Jun-2009 15:33:33] <mrayzenoss> http://www.daemon-systems.org/man/dhcpd.conf.5.html
[01-Jun-2009 15:33:48] <mrayzenoss> CapriCoRN^80: it's a known limitation, but my point was in your dhcpd.conf, you can use a host statement to get around it: http://www.daemon-systems.org/man/dhcpd.conf.5.html
[01-Jun-2009 15:34:45] <mrayzenoss> so even after you 4 day lease is up, the server will renew and get the same IP again
[01-Jun-2009 15:34:48] <rocket> CapriCoRN^80: using zenoss .. at the moment we are still developing etc and setting up
[01-Jun-2009 15:35:09] <rocket> CapriCoRN^80: but we are not running into the issue you are because we are a static ip datacenter shop here
[01-Jun-2009 15:35:35] <mrayzenoss> My home router let's me assign static addresses in DHCP as well
[01-Jun-2009 15:36:07] <CapriCoRN^80> well routers and servers have static ip addresses
[01-Jun-2009 15:36:17] <CapriCoRN^80> i am talking about my client desktops
[01-Jun-2009 15:36:33] <rocket> CapriCoRN^80: we arent doing client desktops at the moment ..
[01-Jun-2009 15:36:50] <mrayzenoss> CapriCoRN^80: you can leave them out by turning off network modeling
[01-Jun-2009 15:37:09] <mrayzenoss> do a network discovery, delete the ones you don't care about and lock it down from there
[01-Jun-2009 15:38:15] <cgibbons> jb are you around?
[01-Jun-2009 15:39:35] <jb> yep
[01-Jun-2009 15:39:36] <CapriCoRN^80> actually we used some new versions of software and we got feed back that the performance is going down due to those softwares .. thats why we are moving to this performance thing
[01-Jun-2009 15:39:55] <jb> cgibbons: whats up?
[01-Jun-2009 15:40:03] <CapriCoRN^80> and yes my requirements are cpu usage , memory usage , process running , and performance
[01-Jun-2009 15:41:01] <mrayzenoss> CapriCoRN^80: so Zenoss will be very lightweight
[01-Jun-2009 15:41:18] <mrayzenoss> SNMP and WMI are not resource intensive
[01-Jun-2009 15:41:36] <CapriCoRN^80> i am not talking about zenoss performance
[01-Jun-2009 15:41:53] <mrayzenoss> right, I mean the effect on the monitored servers
[01-Jun-2009 15:41:56] <mrayzenoss> and devices
[01-Jun-2009 15:42:06] <CapriCoRN^80> and i am talking performance in terms of applications running on clients systems
[01-Jun-2009 15:43:58] <mrayzenoss> right.
[01-Jun-2009 15:44:23] <mrayzenoss> There's a good paper on the different techniques for process monitoring with Zenoss: http://www.skills-1st.co.uk/papers/jcurry.html
[01-Jun-2009 15:44:25] <CapriCoRN^80> i want to check cpu usage , memory usage , process running and performance of client system
[01-Jun-2009 15:44:37] <mrayzenoss> the "Methods of monitoring processes with Zenoss" one
[01-Jun-2009 15:45:09] <rocket> CapriCoRN^80: there is a lot that can go under performance ...
[01-Jun-2009 15:45:32] <mrayzenoss> well you can get overall cpu and memory and per-process cpu and memory
[01-Jun-2009 15:45:59] <CapriCoRN^80> but dhcp issue ?
[01-Jun-2009 15:50:58] <CapriCoRN^80> mrayzenoss:
[01-Jun-2009 15:57:48] <CapriCoRN^80> ?
[01-Jun-2009 15:58:43] <jb> whats the question?
[01-Jun-2009 15:58:51] <jb> you can't really monitor dhcp'd devices..
[01-Jun-2009 15:59:30] <jb> mrayzenoss was saying you could create DHCP reservations to make the device grab the same IP address everytime it requested a lease..
[01-Jun-2009 15:59:40] <jb> depending on your environment, that may or may not be plausible
[01-Jun-2009 15:59:50] <jb> if its a large amount of desktops, it probably is not..
[01-Jun-2009 16:03:31] <rocket> jb: seems he is gone ...
[01-Jun-2009 16:21:53] <jb> yep
[01-Jun-2009 16:22:03] <jb> we tried
[02-Jun-2009 00:00:42] [disconnected at Tue Jun 2 00:00:42 2009]
[02-Jun-2009 00:00:42] [connected at Tue Jun 2 00:00:42 2009]
[02-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[02-Jun-2009 09:18:56] <Dieterbe> Hi, this is not strictly a zenoss issue but maybe you could help me anyway. In device/perf for a Server/Linux on 1 hard disk I see on the "I/O" graph peaks of about 4.0kB (read or write) and about 500bytes "usually"
[02-Jun-2009 09:19:14] <Dieterbe> I was wondering if anyone knows which values i should see here, some comparison data/graphs etc
[02-Jun-2009 09:19:36] <Dieterbe> even for the random io on this box (it runs mysql) this seems quite slow
[02-Jun-2009 09:33:05] <Dieterbe> aha, according to http://wiki.postgresql.org/wiki/HP_ProLiant_DL380_G5_Tuning_Guide#random_read i should get at least 2MB/s
[02-Jun-2009 10:03:52] <afink> Hello everyone, I am having trouble getting performance information on centos 5.2 to show up in zenoss. I have been researching on the internet for some time now and haven't been able to come up with a good solution
[02-Jun-2009 10:08:07] <Dieterbe> afink: can you snmpwalk your target host?
[02-Jun-2009 10:11:10] <chudler> anyone use ZenUtils Driver classes? I have trouble understanding just how it works :-/
[02-Jun-2009 10:12:17] <rocket> chudler: what are you trying to do?
[02-Jun-2009 10:13:43] <afink> Dieterbe: Yes I can snmpwalk it, http://pastebin.com/m5bf38022
[02-Jun-2009 10:13:44] <chudler> inside a custom model, I have an AgentProxy for fetching a few values. I am using the driver for that.
[02-Jun-2009 10:15:03] <Dieterbe> afink: hmm.. check your event log (both in the web interface and the one in /usr/local/zenoss/...log/event.log)
[02-Jun-2009 10:15:11] <Dieterbe> maybe you'll see errors there
[02-Jun-2009 10:15:43] <chudler> rocket: I do yield proxy.get([self.tempOid]), and it immediately returns a defered object to the caller!
[02-Jun-2009 10:16:47] <rocket> ill look and see
[02-Jun-2009 10:16:55] <afink> Dieterbe: This is what shows up on the performance graphs: http://picpaste.com/zenoss.jpeg
[02-Jun-2009 10:17:15] <afink> So I understand that I am missing these rrd files, but how do I get them?
[02-Jun-2009 10:17:26] <chudler> I noticed that all uses of drive(blah), blah() always calls yield twice, but I can't make sense of it
[02-Jun-2009 10:18:36] <rocket> chudler: sorry cant help ya .. havent really used it yet ..
[02-Jun-2009 10:19:04] <afink> Dieterbe: I'm getting all kinds of oid is bad errors on all performance meters
[02-Jun-2009 10:19:15] <chudler> rocket: thanks anyway, I will keep hacking. I tried using plain twisted defers, but that wreaked havoc.
[02-Jun-2009 10:20:44] <capricorn-80> hi
[02-Jun-2009 10:21:03] <capricorn-80> i need some help regarding the application running on machine
[02-Jun-2009 10:21:23] <capricorn-80> i have configured zenoss .. I am monitoring my local system
[02-Jun-2009 10:21:58] <capricorn-80> i am getting cpu usage and memory related graphs
[02-Jun-2009 10:22:48] <capricorn-80> is it possible that zenoss will automatically recogize which applications/deamons are running on my server :
[02-Jun-2009 10:29:27] <capricorn-80> mrayzenoss: you there
[02-Jun-2009 10:31:00] <mrayzenoss> yeah
[02-Jun-2009 10:32:36] <mrayzenoss> Jane Curry's paper on process monitoring is really good, covers configuration and all the different techniques you can use: http://www.skills-1st.co.uk/papers/jcurry.html
[02-Jun-2009 10:34:43] <capricorn-80> ok
[02-Jun-2009 10:35:28] <Dieterbe> afink: i've had actually very similar errors. but in my case i could fix them by setting my community string correctly (to public)
[02-Jun-2009 10:38:00] <afink> Dieterbe: I think I found it..http://forums.zenoss.com/viewtopic.php?p=16513#16513
[02-Jun-2009 10:40:09] <Dieterbe> afink: well, good luck with it. i can't help you anymore
[02-Jun-2009 10:41:51] <afink> Thanks Dieterbe
[02-Jun-2009 10:47:48] <capricorn-80> in the regex. i dont understand the commnad
[02-Jun-2009 10:47:57] <capricorn-80> we have to give the pid of the process ?
[02-Jun-2009 10:51:22] <tyler79> has anyone seen caseyh? I'm considering giving up on zenoss on freebsd unless he made progress fixing the install
[02-Jun-2009 10:51:51] <capricorn-80> mrayzenoss: is it possible that zenoss automatically find different applications
[02-Jun-2009 10:52:37] <mrayzenoss> tyler79: he posted last night that he's almost done with his write-up
[02-Jun-2009 10:52:51] <mrayzenoss> tyler79: http://www.caseyhillman.com/?p=134
[02-Jun-2009 10:52:58] <tyler79> thanks!
[02-Jun-2009 10:53:54] <jb> no, you need to provide the regular expression
[02-Jun-2009 10:54:08] <jb> not the PID.. the PID always changes (if the process is restarted)
[02-Jun-2009 10:54:21] <rocket> mrayzenoss: for perf collecting using ssh am I going to be making command datasources?
[02-Jun-2009 10:54:35] <mrayzenoss> yeah
[02-Jun-2009 10:54:53] <rocket> placing items in libexec basically right?
[02-Jun-2009 10:54:55] <mrayzenoss> rocket: I've got the tests moved over, I should be getting them checked into trunk later today
[02-Jun-2009 10:55:06] <rocket> mrayzenoss: sweet .. thanks
[02-Jun-2009 10:55:07] <mrayzenoss> rocket: yeah
[02-Jun-2009 10:55:24] <mrayzenoss> if you need a script
[02-Jun-2009 10:55:41] <mrayzenoss> you can also just put the raw command in the datasource
[02-Jun-2009 10:55:51] <mrayzenoss> ie. /usr/bin/fping
[02-Jun-2009 10:55:56] <capricorn-80> regex for httpd . example
[02-Jun-2009 10:56:25] <jb> you can use something like.. " ^.*\/*httpd"
[02-Jun-2009 10:56:41] <capricorn-80> ok let me try
[02-Jun-2009 10:57:49] <capricorn-80> jb: i have use the same regex u told
[02-Jun-2009 10:57:52] <capricorn-80> is that fine ?
[02-Jun-2009 10:58:36] <jb> yes
[02-Jun-2009 10:58:52] <capricorn-80> ok thanks
[02-Jun-2009 11:02:12] <mloven> Anyone here actively (and successfully) monitoring any Dell PowerConnect 35xx switches?
[02-Jun-2009 11:04:57] <capricorn-80> jb: done but still no getting any httpd process in OS
[02-Jun-2009 11:06:14] <rmatte> I'm just finishing creating a new client base vmware image with Zenoss installed (2.4.1). All of a sudden I'm getting constant heartbeat failure alerts for each process, and they aren't going away like they normally eventually do, any ideas on what might be causing this?
[02-Jun-2009 11:10:59] <rmatte> ah, I configured NTP
[02-Jun-2009 11:11:06] <rmatte> apparently that can cause this type of issue
[02-Jun-2009 11:11:14] <rmatte> now to figure out how to fix it...
[02-Jun-2009 11:14:38] <capricorn-80> hmm
[02-Jun-2009 11:16:18] <jb> remodel the device..
[02-Jun-2009 11:16:57] <capricorn-80> jb: thats for me
[02-Jun-2009 11:19:54] <rmatte> I just went in to Event Manager and selected "Clear All Heartbeats", will see if that fixes it...
[02-Jun-2009 11:20:56] <rmatte> seems to have possibly done the trick
[02-Jun-2009 11:23:50] <capricorn-80> still not working
[02-Jun-2009 11:28:27] <capricorn-80> jb: need ur 5 mins
[02-Jun-2009 11:33:18] <jb> well, you aren't really providing any details
[02-Jun-2009 11:35:44] <capricorn-80> jb: trying to monitor my processes on linux server. Process class - name = httpd , regex = ^.*\/*httpd , monitor = true
[02-Jun-2009 11:35:52] <jb> ok
[02-Jun-2009 11:36:34] <jb> is it showing up under the Processes box in "OS" ?
[02-Jun-2009 11:37:32] <capricorn-80> zProperties Configuration: zAlertOnRestart = true , zCountProcs = true , zFailSeverity = critical , zMonitor = true
[02-Jun-2009 11:37:34] <capricorn-80> no
[02-Jun-2009 11:37:40] <jb> did you remodel the device?
[02-Jun-2009 11:38:30] <capricorn-80> what do u mean my remodel
[02-Jun-2009 11:38:37] <capricorn-80> by
[02-Jun-2009 11:38:43] <capricorn-80> restart
[02-Jun-2009 11:38:48] <jb> no
[02-Jun-2009 11:38:58] <capricorn-80> then
[02-Jun-2009 11:39:57] <jb> Browse to the device, click the down arrow, and choose Manage -> Model Device
[02-Jun-2009 11:41:15] <capricorn-80> i have Add reset features under Manage
[02-Jun-2009 11:41:55] <capricorn-80> lock
[02-Jun-2009 11:47:06] <capricorn-80> i guess i provide you information this time
[02-Jun-2009 11:52:10] <Kromel> *pulls hair* I'm trying to login a fresh install of Zenoss (VM Image) and the default root loging doesn't work.
[02-Jun-2009 11:52:24] <Kromel> the Zenoss web interface works fine
[02-Jun-2009 11:52:34] <Kromel> it's the linux OS i'm trying to log in to
[02-Jun-2009 11:53:33] <Kromel> thought it was username: root with nothing for the password.
[02-Jun-2009 11:53:48] <Kromel> then I would have to issue a passwd command to assign it a password
[02-Jun-2009 12:02:05] <capricorn-80> jb: got it
[02-Jun-2009 12:02:18] <capricorn-80> i was searching in device not in the particular device
[02-Jun-2009 12:10:35] <rocket> bah .. I need my regex book at the office .. I cant remember what [%-] will do .. :/
[02-Jun-2009 12:11:07] <rocket> nevermind .. figured it out ..
[02-Jun-2009 12:15:07] <capricorn-80> hahaha
[02-Jun-2009 12:15:35] <capricorn-80> i remove the entried of one of the processes and now i cant edit it "P
[02-Jun-2009 12:16:18] <capricorn-80> because there is no entry in the process class and the process status is monitor
[02-Jun-2009 12:16:25] <capricorn-80> how can i delete it ?
[02-Jun-2009 12:19:21] <capricorn-80> i guess i have solved it
[02-Jun-2009 12:20:02] <rmatte> Kromel: no, root is never defaulted to blank, you should have been asked to set a root password during the linux install
[02-Jun-2009 12:20:36] <rmatte> Kromel: if it's a virtual appliance, then it might be something like root/root or root/zenoss
[02-Jun-2009 12:20:38] <rmatte> no idea
[02-Jun-2009 12:32:19] <Kromel> figured it out
[02-Jun-2009 12:32:34] <Kromel> I had to use the VMware Infrastructure client
[02-Jun-2009 12:32:41] <Kromel> went in via Console
[02-Jun-2009 12:32:44] <capricorn-80> which collector is used for proccesses ?\
[02-Jun-2009 12:33:02] <capricorn-80> zcollectorplugins
[02-Jun-2009 12:34:53] <Kromel> thinking maybe I should install Zenoss on a dedicated linux
[02-Jun-2009 12:36:51] <Kromel> how can I download a zenpack using the zenoss management console?
[02-Jun-2009 12:37:00] <Kromel> i'll google that question right now
[02-Jun-2009 12:40:33] <Kromel> ahh wget!
[02-Jun-2009 13:19:58] <chudler> anyone see if google is slow for them right now?
[02-Jun-2009 13:20:16] <rocket> nope not slow for me
[02-Jun-2009 13:20:32] <rocket> anyone here work with command parsers?
[02-Jun-2009 13:20:52] <rocket> trying to understand componentScanValue .. the documentation doesnt make much sense to me at the moment
[02-Jun-2009 13:59:12] <asmith> hello there
[02-Jun-2009 13:59:23] <asmith> zenoss newbie here
[02-Jun-2009 13:59:46] <asmith> i'd like to respond to a zenoss event with some custom code
[02-Jun-2009 13:59:55] <asmith> is ZenPacks the right way to do this?
[02-Jun-2009 14:00:02] <asmith> i don't really care about a UI
[02-Jun-2009 14:01:30] <mrayzenoss> sounds like an event transform might be what you want
[02-Jun-2009 14:01:34] <mrayzenoss> depends what you're trying to do
[02-Jun-2009 14:02:15] <asmith> well
[02-Jun-2009 14:02:54] <asmith> when i get a ping status my managed host
[02-Jun-2009 14:02:58] <asmith> "might" be down
[02-Jun-2009 14:03:32] <asmith> i'd like to have a function to do a few more checks before pulling the ripcord and trying to modify some other configuration
[02-Jun-2009 14:03:54] <asmith> right now i've got this hacked up in the event manager
[02-Jun-2009 14:04:29] <asmith> to execute a shell program when this event happens and give it the stringified event as input
[02-Jun-2009 14:04:35] <asmith> but that seems like a nasty hack
[02-Jun-2009 14:04:43] <mrayzenoss> well, that would be an Event Command
[02-Jun-2009 14:04:53] <asmith> especially since the program it ultimately invokes is a python script
[02-Jun-2009 14:05:09] <asmith> which I'd like to just run inside the same interpreter that zenoss is running under
[02-Jun-2009 14:05:46] <mrayzenoss> ok, if you want to stick with Python, an Event Transform, if you want to use the shell, Event Command
[02-Jun-2009 14:05:59] <mrayzenoss> http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch07s01.html
[02-Jun-2009 14:06:19] <mrayzenoss> here are some Event Transform examples: http://www.zenoss.com/community/wiki/EventTransforms
[02-Jun-2009 14:07:02] <asmith> excellent
[02-Jun-2009 14:07:11] <asmith> a silly question
[02-Jun-2009 14:07:24] <asmith> but where do I "inject" those transforms into zenoss once I've written them?
[02-Jun-2009 14:07:51] <mrayzenoss> so if you want this on Ping Down
[02-Jun-2009 14:08:54] <mrayzenoss> /Events/Status/Ping
[02-Jun-2009 14:09:20] <mrayzenoss> More->Transform
[02-Jun-2009 14:10:15] <asmith> Can my tranform methods keep any state?
[02-Jun-2009 14:10:30] <mrayzenoss> you can look up other events
[02-Jun-2009 14:10:42] <mrayzenoss> and the state of the device
[02-Jun-2009 14:10:43] <asmith> I'd like to basically say -- wait for 5 Ping Down events
[02-Jun-2009 14:10:50] <asmith> and then do foo
[02-Jun-2009 14:11:11] <mrayzenoss> yeah
[02-Jun-2009 14:11:45] <asmith> but somebody else told me transforms don't have a count because you can only transform an event before it is committed to the db
[02-Jun-2009 14:11:55] <mrayzenoss> that's true
[02-Jun-2009 14:11:56] <asmith> and the count comes after the commit
[02-Jun-2009 14:12:21] <mrayzenoss> but you can have a transform that executes after the count
[02-Jun-2009 14:12:43] <mrayzenoss> add an EventClass mapping
[02-Jun-2009 14:12:52] <mrayzenoss> and put the transform in there instead of More->Transforms
[02-Jun-2009 14:14:01] <asmith> alright -- we've strayed outside my comfort zone, but that sounds like something to go on
[02-Jun-2009 14:14:03] <asmith> thank you
[02-Jun-2009 14:14:14] <mrayzenoss> no prob
[02-Jun-2009 14:14:17] <asmith> actually
[02-Jun-2009 14:14:19] <mrayzenoss> there's a good paper on event management
[02-Jun-2009 14:14:21] <asmith> one more question
[02-Jun-2009 14:14:30] <asmith> that transform method
[02-Jun-2009 14:14:33] <mrayzenoss> http://www.skills-1st.co.uk/papers/jcurry.html Zenoss Event Management
[02-Jun-2009 14:14:52] <asmith> if i wanted to just keep state in a file or something
[02-Jun-2009 14:14:59] <asmith> do i have anyway to do that?
[02-Jun-2009 14:15:20] <mrayzenoss> well, I think you could do that in the transform since it's Python
[02-Jun-2009 14:15:20] <asmith> so i could map some kind of host id / event type / counter myself
[02-Jun-2009 14:15:32] <asmith> ok -- so i have all the power of python at my disposal
[02-Jun-2009 14:15:34] <asmith> great
[02-Jun-2009 14:15:36] <mrayzenoss> oh yeah
[02-Jun-2009 14:15:50] <mrayzenoss> if you go back to those examples, you see import statements
[02-Jun-2009 14:15:58] <asmith> nice
[02-Jun-2009 14:16:07] <mrayzenoss> and a bit dangerous
[02-Jun-2009 14:16:14] <asmith> well - the EventClass mapping sounds like the way to go
[02-Jun-2009 14:16:26] <mrayzenoss> good luck
[02-Jun-2009 14:16:35] <asmith> thanks
[02-Jun-2009 14:35:04] <genji0714> hello all....i am doing a fresh install of zenoss core 2.4 and zenoss will only accept an alphanumeric for the mysql root password...is there a way around this because my root password for mysql has symbols in it.
[02-Jun-2009 14:39:49] <mrayzenoss> genji0714: http://dev.zenoss.org/trac/ticket/5105
[02-Jun-2009 14:40:02] <mrayzenoss> You can use the --mysql-database calls to use your DB
[02-Jun-2009 14:48:51] <mloven> Anyone here successfully monitoring some PowerConnect 35xx switches?
[02-Jun-2009 14:55:30] <mrayzenoss> I haven't seen anything for that, but I'd love a ZenPack
[02-Jun-2009 14:56:12] <rocket> bah anyone remember where the interface statistics similar to the snmp output is on a linux box? ..
[02-Jun-2009 14:56:23] <rocket> working on enterprise stuff too much lately ..
[02-Jun-2009 14:56:49] <mloven> yeah, I would too Matt....
[02-Jun-2009 14:58:38] <mrayzenoss> rocket: ifconfig
[02-Jun-2009 14:59:02] <rocket> mrayzenoss: k .. just am not seeing ucastpackets etc
[02-Jun-2009 15:06:12] <mrayzenoss> ucast? I see RX/TX
[02-Jun-2009 15:07:06] <rocket> unicast ..
[02-Jun-2009 15:20:03] <genji0714> mrayzenoss: that didn't work
[02-Jun-2009 15:28:59] bzed_ is now known as bzed
[02-Jun-2009 15:44:16] <rocket> mrayzenoss: get the examples uploaded?
[02-Jun-2009 15:46:45] <mrayzenoss> working with Brian right now
[02-Jun-2009 15:47:06] <rocket> bah .. I am still having troubles with parsers .. :/
[02-Jun-2009 15:47:17] <rocket> ImportError: No module named ZenPacks.zenoss.LinuxMonitor-1.0.0-py2.4.egg.ZenPacks.zenoss.LinuxMonitor.parsers.linux.ifconfig
[02-Jun-2009 15:47:33] <rocket> I created an ifconfig parser .. to parse the output of ifconfig -a
[02-Jun-2009 15:47:43] <rocket> hrmmm what am I doing wrong ..
[02-Jun-2009 16:30:03] * alex___ waves hello to everyone
[02-Jun-2009 16:31:01] <alex___> can anyone point me towards a replacement for informant/WMI that will let me get windows performance data over SNMP and not require me to pay money for it
[02-Jun-2009 16:31:26] <chudler> I can't. But I thought informant was free?
[02-Jun-2009 16:31:49] <chudler> oh I see, you are looking for a replacement that is *also* free
[02-Jun-2009 16:32:06] <alex___> well, informant is free for non commercial use
[02-Jun-2009 16:32:31] <chudler> ouch. always a catch.
[02-Jun-2009 16:32:37] <alex___> tell me about it
[02-Jun-2009 16:33:03] <alex___> there are a few zenpacks but they require opening port 445 on my server to the world
[02-Jun-2009 16:33:08] <alex___> which is not happening
[02-Jun-2009 16:35:54] <chudler> it is slightly outside of my area. WMI is a good alternative. Zenoss enterprise uses it very effectively. I think it would be simple to build a similar zenpack, and others probably already have.
[02-Jun-2009 16:36:19] <chudler> but that does screw your "over SNMP" requirement.
[02-Jun-2009 16:36:30] <alex___> does enterprise connect to wmi directly or does the zenpack interface wmi to snmp
[02-Jun-2009 16:36:46] <chudler> direct to wmi, using samba wmic, I believe.
[02-Jun-2009 16:36:59] <alex___> my biggest issue with wmi is that every host i monitor is behind a different firewall
[02-Jun-2009 16:37:20] <alex___> so i have to open snmp and wmi each time
[02-Jun-2009 16:37:32] <alex___> do you know what port wmic listens on
[02-Jun-2009 16:39:58] <alex___> perhaps you can help me with another smaller issue
[02-Jun-2009 16:40:16] <alex___> i dont quite get the relationship between performance templates and devices
[02-Jun-2009 16:41:29] <alex___> ive added the ethernetCsmacd to one of my servers, so my Perf page shows 4 empty graph boxes for network interface data
[02-Jun-2009 16:41:45] <alex___> the data appears fine under the OS -> eth0 tab
[02-Jun-2009 16:46:31] <alex___> is there a way to either get the graphs to populate on the perf page like they are on the OS, or not appear at all on perf and just go to OS to view them
[02-Jun-2009 17:19:07] <mrayzenoss> alex___: you want to move the OS graphs to the perf page?
[02-Jun-2009 17:19:23] <mrayzenoss> I guess you could hack up the .pt files
[02-Jun-2009 17:40:07] <alex___> what does it mean if there is no checkbox next to a template in the All Templates list
[02-Jun-2009 18:19:12] <alex___> thanks for help
[02-Jun-2009 18:33:48] <kisielk> alright, time to upgrade to 2.4
[02-Jun-2009 19:36:05] <kisielk> is zenmigrate supposed to take a *really* long time to run?
[03-Jun-2009 00:00:42] [disconnected at Wed Jun 3 00:00:42 2009]
[03-Jun-2009 00:00:42] [connected at Wed Jun 3 00:00:42 2009]
[03-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[03-Jun-2009 07:59:32] <osd_mats> I must be stupid... I'm editing a data source that use ssh agains a device, but when i use the "test against device" field it just tries agains the zenoss server. What am i missing?
[03-Jun-2009 08:06:03] <SpunkMeYeR> hi all
[03-Jun-2009 08:06:52] <SpunkMeYeR> anybody here knows how to send an alert sms from zenoss?
[03-Jun-2009 08:23:21] <chudler> spunkmeyer: two common ways are 1) send email to the sms-email-gateway of the carrier, or 2) change Page Command in Settings tab to an SMS sender app
[03-Jun-2009 08:23:28] <chudler> dang, he just left :-0
[03-Jun-2009 08:27:54] <Dieterbe> chudler: at least you tried
[03-Jun-2009 08:30:15] <rocket> bah so trying to do development with a zenoss vmware image I think is broken with the unit tests .. :/
[03-Jun-2009 09:34:39] <malbon> mrayzenoss: yo
[03-Jun-2009 09:36:19] <mrayzenoss> myellow
[03-Jun-2009 09:38:03] <malbon> mrayzenoss: have you got to the MyEventViews zenpack in your queue?
[03-Jun-2009 09:38:39] <mrayzenoss> yeah, I'm getting it out today with 5 other ZenPacks
[03-Jun-2009 09:38:51] <malbon> top banana
[03-Jun-2009 09:39:25] <mrayzenoss> actually, I'll go commit it right now if you're in a hurry
[03-Jun-2009 09:39:40] <malbon> nope, by tomorrow is fine.
[03-Jun-2009 09:39:40] <mrayzenoss> I merged it in yesterday
[03-Jun-2009 09:39:49] <mrayzenoss> ok, cool
[03-Jun-2009 09:40:13] <malbon> We can raise a change control on that basis, and it's not going to be processed until the meeting tomororw.
[03-Jun-2009 09:40:40] <malbon> this new pack corrects the refresh problem?
[03-Jun-2009 09:40:57] <mrayzenoss> yup
[03-Jun-2009 09:41:35] <malbon> cool
[03-Jun-2009 09:42:11] <mrayzenoss> trac-zenpacks/changeset/229
[03-Jun-2009 09:43:04] <malbon> heh, not much changes does it.
[03-Jun-2009 09:43:17] <mrayzenoss> yeah, you could probably patch it your self
[03-Jun-2009 09:43:30] <mrayzenoss> by hand that is
[03-Jun-2009 09:43:40] <mrayzenoss> but I'll get it out officially in a ZenPack today
[03-Jun-2009 09:43:52] <malbon> I'll wait for the package. It's a feature that our users want but they can wait for the process.
[03-Jun-2009 09:44:14] <mrayzenoss> Florian also sent in a Reports portlet that's going up today too
[03-Jun-2009 09:44:43] <malbon> what does that do?
[03-Jun-2009 09:45:31] <mrayzenoss> "I also submit a new portlet that let one create shortcuts to his favorite reports. One there are tons of reports available, one may prefere to gather the ones he reads more on his dashboard."
[03-Jun-2009 09:45:43] <mrayzenoss> you just put your favorite reports on the dashboard for easy access
[03-Jun-2009 09:46:10] <malbon> ok, I can see that.
[03-Jun-2009 09:50:06] <malbon> mrayzenoss: anyhow gotta dash. Thanks for doing those updates.
[03-Jun-2009 09:50:16] <mrayzenoss> no problem
[03-Jun-2009 09:53:40] <xulfer> Hrm, anyone managed to monitor Trango's with any success in here?
[03-Jun-2009 09:56:16] <xulfer> Ha, just as i thought.
[03-Jun-2009 10:18:09] <rocket> bah it would help if I downloaded the 32bit stack for a 32bit system ..
[03-Jun-2009 10:20:33] <chudler> ('Unable to parse arguments', ['/opt/zenoss/lib/python/Zope2/Startup/run.py', '-v', '2c', '-c', 'public', '-t', '2.5', '-r', '2', '10.128.10.25:161'])
[03-Jun-2009 10:21:26] <chudler> ARG, I get this at random when using AgentProxy (netsnmp) inside the zenhub instance
[03-Jun-2009 10:22:02] <chudler> I believe something horrible is happening with (non) thread-safe use
[03-Jun-2009 10:22:29] <rocket> chudler: figuring out the agent proxy stuff?
[03-Jun-2009 10:22:51] <chudler> yes, it works fine, except for that thing.
[03-Jun-2009 10:24:16] <chudler> I don't know which is worse. Doing cacheRRD inside a threshold with distributed collectors (locks up zenhub), or doing AgentProxy.get inside a twisted event loop (random failure, as above).
[03-Jun-2009 10:25:06] <rocket> anything twisted is evil ..
[03-Jun-2009 10:26:41] <chudler> I suspect a newer Zope might not have this problem. Lacking evidence, though, and it's a long way off yet before Zenoss can move to it.
[03-Jun-2009 10:30:03] <rocket> chudler: there is a newer zope in the tree
[03-Jun-2009 10:30:18] <rocket> if you wanted to test against the unstable tree
[03-Jun-2009 10:30:27] <rocket> but that might not prove anything ..
[03-Jun-2009 10:30:34] <mrayzenoss> yeah, Ian said he stabilized it this morning
[03-Jun-2009 10:30:51] <mrayzenoss> QA had opened several blockers on trunk with the new 2.11.2
[03-Jun-2009 10:31:17] <chudler> sweet. Not a great test though. I really need to get to the bottom of the problem. I wonder about the argument list that starts with "'/opt/zenoss/lib/python/Zope2/Startup/run.py'" I wonder why it is there.
[03-Jun-2009 10:43:00] <chudler> oh yeah, I forgot. The first argument is normally the name of the file calling. Hmm. okay, just thinking out loud :-)
[03-Jun-2009 10:53:28] <jb> hey everyone
[03-Jun-2009 10:54:39] <jb> hm, zenwin/zeneventlog cycles are like 60+ seconds now
[03-Jun-2009 11:01:40] <rocket> mrayzenoss: ping
[03-Jun-2009 11:02:17] <mrayzenoss> hello
[03-Jun-2009 11:02:35] <rocket> hey I am just getting to the point where I can run unit tests
[03-Jun-2009 11:02:45] <rocket> and its working .. and it shows me all the tests fail
[03-Jun-2009 11:03:01] <kisielk> alright, I upgraded to 2.4.1 from 2.3.3 yesterday, but it seems my LDAP auth broke
[03-Jun-2009 11:03:06] <rocket> but it doesnt show much more than that .. is there a more verbose option or someting I am missing?
[03-Jun-2009 11:03:13] <kisielk> I had to remove the product in order to even get the dashboard to work
[03-Jun-2009 11:03:24] <kisielk> but now I can't manage my users, or authenticate them against LDAP
[03-Jun-2009 11:03:30] <kisielk> this is a stack installer
[03-Jun-2009 11:10:57] <mrayzenoss> rocket: Good question, I'm asking around
[03-Jun-2009 11:11:34] <rocket> otherwise its very difficult to troubleshoot
[03-Jun-2009 11:11:34] <kisielk> how do I get it to work again ? do I have to reinstall the auth LDAP stuff?
[03-Jun-2009 11:11:37] <mrayzenoss> kisielk: sorry, not much help from me. I just posted on an LDAP thread looking for people who have it working to help people having trouble
[03-Jun-2009 11:11:52] <kisielk> it looks like the location of Python moved
[03-Jun-2009 11:11:59] <kisielk> in 2.4.1, I suspect that may be the problem
[03-Jun-2009 11:13:32] <mrayzenoss> rocket: I purposely made one of my tests fail and I got a stack trace with the AssertionError
[03-Jun-2009 11:13:35] <mrayzenoss> what are you seeing?
[03-Jun-2009 11:14:02] <rocket> an assertion that it failed .. but nothing indicating why
[03-Jun-2009 11:14:41] <rocket> expected 12 values, actual 0 .. but it doesnt show what the values were etc
[03-Jun-2009 11:15:08] <mrayzenoss> hmm... Ian said he's writing a patch to put more verbosity into runtest
[03-Jun-2009 11:15:27] <mrayzenoss> but I think that won't help us, since he already updated it substantially for the new Zope 2.11.2 stuff
[03-Jun-2009 11:16:23] <mrayzenoss> here's what mine looks like: http://pastebin.com/d1c180d1a
[03-Jun-2009 11:16:59] <rocket> hrmm .. mine doesnt look like that .. ;p
[03-Jun-2009 11:17:08] <mrayzenoss> of course, I see that it doesn't show which test it was
[03-Jun-2009 11:17:15] <mrayzenoss> just that it failed
[03-Jun-2009 11:17:25] <mrayzenoss> I mean, I know to grep for 103707
[03-Jun-2009 11:17:59] <rocket> http://pastebin.com/m2e29e41b
[03-Jun-2009 11:20:42] <mrayzenoss> trying to get Ian to look at it
[03-Jun-2009 11:21:23] <rocket> sweet thanks
[03-Jun-2009 11:21:48] <mrayzenoss> well, 2.5 now has -vvvvvv as a runtests option
[03-Jun-2009 11:21:59] <rocket> bah that doesnt help me
[03-Jun-2009 11:22:04] <mrayzenoss> but that doesn't help yet
[03-Jun-2009 11:23:52] <mrayzenoss> try zopectl test --libdir /tmp/ZenPacks.community.Gentoo -vv
[03-Jun-2009 11:23:55] <mrayzenoss> or whatever your path is
[03-Jun-2009 11:24:35] <kisielk> Gentoo zenpack?
[03-Jun-2009 11:24:46] * kisielk 's ears perk up
[03-Jun-2009 11:24:47] <mrayzenoss> kisielk: that's what rocket is writing
[03-Jun-2009 11:24:53] <kisielk> rocket: what does it do?
[03-Jun-2009 11:24:54] <mrayzenoss> an SSH one
[03-Jun-2009 11:25:09] <rocket> ssh based one .. gets the list of the software installed etc using eix
[03-Jun-2009 11:25:22] <kisielk> ah
[03-Jun-2009 11:25:25] <rocket> updating the linux one to do the interface perf stuff
[03-Jun-2009 11:25:28] <kisielk> I actually wrote an SNMP module for that
[03-Jun-2009 11:25:44] <kisielk> works similar to the Ubuntu one that was posted a few months back
[03-Jun-2009 11:25:48] <rocket> this one is ssh based though .. so you wouldnt have to install that ..
[03-Jun-2009 11:26:05] <rocket> just need the eix utility installed and its database updated every so often
[03-Jun-2009 11:26:13] <kisielk> yeah
[03-Jun-2009 11:26:29] <kisielk> although I've deprecated all of our Gentoo machines other than our cluster
[03-Jun-2009 11:26:35] <kisielk> so it's not much use to me any more
[03-Jun-2009 11:26:36] <rocket> basically I am using it as a way to learn the ssh stuff and flesh out the unit test stuff
[03-Jun-2009 11:27:01] <rocket> once I have one of these done the other linux distos should be pretty easy
[03-Jun-2009 11:27:30] <mrayzenoss> rocket: any change in output?
[03-Jun-2009 11:27:36] <capricorn-80> hi
[03-Jun-2009 11:27:48] <capricorn-80> can i run zenoss and mrtg on the same server ?
[03-Jun-2009 11:28:14] <mrayzenoss> capricorn-80: not sure why you'd need to
[03-Jun-2009 11:28:14] <rocket> mrayzenoss: nope
[03-Jun-2009 11:28:49] <mrayzenoss> rocket: anything with -vvvvv ?
[03-Jun-2009 11:28:53] <mrayzenoss> 5 vs
[03-Jun-2009 11:28:55] <rocket> nope .. the same ..
[03-Jun-2009 11:29:16] <rocket> I am sure its something wrong with my setup .. but there is very little to go on to figure out
[03-Jun-2009 11:29:52] <mrayzenoss> rocket: why don't you check it into subversion and I'll look at it
[03-Jun-2009 11:29:59] <capricorn-80> rocket: why not ?
[03-Jun-2009 11:30:25] <rocket> mrayzenoss: I would except svn access is blocked here at work :/
[03-Jun-2009 11:30:31] <mrayzenoss> nutz
[03-Jun-2009 11:30:34] <rocket> I would need to run to our lab with the ports open
[03-Jun-2009 11:30:47] <rocket> which I can do if absolutely necessary ..
[03-Jun-2009 11:31:45] <mrayzenoss> just tar it up and email it to me
[03-Jun-2009 11:31:51] <mrayzenoss> I'll see if I can debug it
[03-Jun-2009 11:32:02] <rocket> sure
[03-Jun-2009 11:32:09] <mrayzenoss> but now I'm headed to lunch
[03-Jun-2009 11:32:23] * mrayzenoss is off to lunch
[03-Jun-2009 11:32:49] <capricorn-80> rocket: ?
[03-Jun-2009 11:33:16] <rocket> capricorn-80: I was talking to mrayzenoss not answering your question
[03-Jun-2009 11:33:40] <capricorn-80> ops
[03-Jun-2009 11:33:43] <capricorn-80> can i run zenoss and mrtg on the same server ?
[03-Jun-2009 11:33:46] <capricorn-80> so answer it
[03-Jun-2009 11:34:15] <rocket> I dont know why you wouldnt be able to
[03-Jun-2009 11:34:37] <capricorn-80> rocket: what i wouldnt be able to ?
[03-Jun-2009 11:34:55] <rocket> capricorn-80: non-english native speaker?
[03-Jun-2009 11:36:02] <capricorn-80> rocket: yea . swedish
[03-Jun-2009 11:36:15] <capricorn-80> rocket: your sentence was incomplete
[03-Jun-2009 11:36:47] <rocket> I dont know why you wouldnt be able to run both at the same time
[03-Jun-2009 11:37:10] <Mousey> bleh
[03-Jun-2009 11:37:17] <Mousey> making classes is hard
[03-Jun-2009 11:37:28] <capricorn-80> rocket: ahan .. i was checking it .. its testing server
[03-Jun-2009 11:37:40] <capricorn-80> i will move to zenoss on other server
[03-Jun-2009 11:37:54] <capricorn-80> i am just testing right now coz i dont have any other server
[03-Jun-2009 11:38:02] <rocket> it would be similar to running dns and email on the same server
[03-Jun-2009 11:38:19] <capricorn-80> rocket: ok
[03-Jun-2009 11:38:48] <kisielk> mrayzenoss: alright, I got it working
[03-Jun-2009 11:38:55] <kisielk> mrayzenoss: I just had to easy_install python-ldap as the zenoss user
[03-Jun-2009 11:39:39] <capricorn-80> rocket: i add my cisco router to zenoss . its working fine and i am getting cpu and memory information
[03-Jun-2009 11:39:51] <capricorn-80> but i am not getting bandwidth graph
[03-Jun-2009 11:40:07] <rocket> check logs for errors?
[03-Jun-2009 11:40:30] <capricorn-80> rocket: log msg for me ?
[03-Jun-2009 11:46:29] <capricorn-80> anyone who can help me that how can i monitor bandwidth using zenoss and make graph similar like mrtg for bandwidth
[03-Jun-2009 12:00:33] <Mousey> why when i set the OS make, less and less OS version options bercome available?
[03-Jun-2009 12:01:45] <Mousey> i set it to debian for my first device add, and set it to unstable, for my second device add "unstable" is no longer an option for "debian" so i set it to lenny. later i went back to edit the device, and both OS fields were empty, so i set the make to "debian" and now "lenny" isn't available. =(
[03-Jun-2009 13:16:13] <bedwards> eric, you had a question about component scan value
[03-Jun-2009 13:21:32] <rocket> yeah I am trying to understand what it does
[03-Jun-2009 13:21:52] <rocket> eg whats the difference between mount and id(default)
[03-Jun-2009 13:22:04] <rocket> as my tests seem to succeed regardless of its setting
[03-Jun-2009 13:22:21] <rocket> I am building a parser for ifconfig
[03-Jun-2009 13:22:31] <rocket> for performance graphing points
[03-Jun-2009 13:22:45] <rocket> and I am assuming I need to set it to something like
[03-Jun-2009 13:22:47] <rocket> componentScanValue = 'interfaceName'
[03-Jun-2009 13:33:30] <kdub> i'm trying to install zenoss core via dpkg on a debian system. the install fails saying it cannoct connect to the local mysql socket. any help?
[03-Jun-2009 13:41:20] <mrayzenoss> kdub: are you already running mysql?
[03-Jun-2009 13:42:08] <rocket> bedwards: ?
[03-Jun-2009 13:44:40] <bedwards> don't use ComponentCommandParser, use CommandParser instead
[03-Jun-2009 13:45:32] <rocket> bedwards: I am curious as to why? eg I was looking at the df and dfi examples
[03-Jun-2009 13:46:35] <bedwards> it is harder to understand once it has been written, the framework is very limiting, and it is harder to debug
[03-Jun-2009 13:46:50] <rocket> bedwards: so its the old way of doing it basically?
[03-Jun-2009 13:47:15] <bedwards> i don't use it
[03-Jun-2009 13:48:09] <rocket> ok .. I can rewrite it to use the CommandParser .. should be pretty easy to reuse the regex's etc ..
[03-Jun-2009 13:48:33] <rocket> bedwards: I agree about the debugging I had to add print statements to the main code to figure out what was going on with it
[03-Jun-2009 13:50:18] <bedwards> python has great string processing tools outside of regexes, so ComponentCommandParser is very limiting in that way too. i use regexes as a last resort in python.
[03-Jun-2009 13:51:20] <kdub> mrayzenoss: i found some mysql processes running, killed them, and am reinstalling
[03-Jun-2009 13:51:22] <bedwards> i believe that there are some component parsers that subclass CommandParser in core
[03-Jun-2009 13:51:33] <kdub> no errors so far, this installer takes forever though...
[03-Jun-2009 13:51:39] <rocket> yeah there are ... I will take a look at them again ...
[03-Jun-2009 13:52:00] <rocket> and the linuxmonitor zenpack has a cpu one ..
[03-Jun-2009 13:53:16] <rocket> thanks bedwards
[03-Jun-2009 13:53:21] <rocket> Thats all I needed
[03-Jun-2009 13:57:18] <kdub> is this install supposed to take like an hour+ ?
[03-Jun-2009 14:02:29] <rmatte> what is the point of the list of snmp strings in zProperties? Zenoss doesn't even seem to use them.
[03-Jun-2009 14:02:57] <mrayzenoss> kdub: no
[03-Jun-2009 14:03:25] <mrayzenoss> rmatte: do you mean the zSnmpCommunities?
[03-Jun-2009 14:06:23] <kdub> mrayzenoss: i killed the apt-get, i'm hoping its fine to run
[03-Jun-2009 14:07:50] <mrayzenoss> kdub: I think it installs to /opt/zenoss on Debian/Ubuntu
[03-Jun-2009 14:08:02] <mrayzenoss> I could be mistaken, I run from source on Debian PPC
[03-Jun-2009 14:08:53] <kdub> it put something called bitrock into /opt, most of it is in /usr/local
[03-Jun-2009 14:09:39] <mrayzenoss> yeah, bitrock is the provider of the stack installs
[03-Jun-2009 14:09:55] <mrayzenoss> if you do a dpkg --list, I assume there's nothing zenoss related?
[03-Jun-2009 14:11:40] <kdub> no, zenoss-stack is there, it hung in the postinst script
[03-Jun-2009 14:13:16] <mrayzenoss> dpkg --purge zenoss-stack
[03-Jun-2009 14:15:50] <kdub> i did it, tried to reinstall, its complaining though
[03-Jun-2009 14:20:00] <rmatte> mrayzenoss: yes
[03-Jun-2009 14:20:37] <rmatte> mrayzenoss: My assumption was that Zenoss would go through that list and try each string when modelling a device until it found one that works
[03-Jun-2009 14:20:39] <mrayzenoss> rmatte: they're supposed to be used when new devices are discovered
[03-Jun-2009 14:20:42] <rmatte> but apparently it doesn't do that
[03-Jun-2009 14:20:53] <rmatte> oh, so they are only used on auto-discovery?
[03-Jun-2009 14:20:57] <rmatte> not on manual addition?
[03-Jun-2009 14:21:21] <mrayzenoss> yeah, that sounds right
[03-Jun-2009 14:21:26] <rmatte> I see
[03-Jun-2009 14:21:29] <rmatte> k, thanks
[03-Jun-2009 14:41:43] <kdub> do these installer errors make sense to anyone?
[03-Jun-2009 14:41:46] <kdub> http://pastebin.ca/1446659
[03-Jun-2009 14:43:18] <mrayzenoss> kdub: I sent that to our build engineer, he'll take a look
[03-Jun-2009 14:43:39] <mrayzenoss> kdub: btw, 2.4.1 is out so you should probably install that instead of 2.4.0
[03-Jun-2009 14:44:41] <mrayzenoss> this is 32-bit Debian 4 correct?
[03-Jun-2009 14:51:19] <rmatte> the job manager is definitely nice
[03-Jun-2009 14:51:27] <rmatte> don't have to manually kill auto-discoveries now
[03-Jun-2009 14:51:36] <mrayzenoss> more stuff will become Job Manager aware in time
[03-Jun-2009 14:51:41] <rmatte> right on
[03-Jun-2009 14:53:02] <kdub> mrayzenoss: its etch, 32bit
[03-Jun-2009 14:53:30] <kdub> i still havent figured out debians naming scheme
[03-Jun-2009 14:54:11] <mrayzenoss> kdub: it's based on characters from Toy Story
[03-Jun-2009 14:54:33] <mrayzenoss> woody, etch, lenny, sarge
[03-Jun-2009 14:54:45] <mrayzenoss> experimental is "sid" because it will break your toys
[03-Jun-2009 14:55:05] <kdub> experimental is _always_ named sid?
[03-Jun-2009 14:55:08] <mrayzenoss> yeah
[03-Jun-2009 14:55:14] <mrayzenoss> I think Bruce Perens was working for Pixar at the time
[03-Jun-2009 14:55:20] <mrayzenoss> when they started the naming scheme
[03-Jun-2009 14:55:27] <rmatte> hehe
[03-Jun-2009 14:55:30] <kdub> oh, and the stable changes from like etch to lenny
[03-Jun-2009 14:55:34] <mrayzenoss> yeah
[03-Jun-2009 14:55:40] <mrayzenoss> Lenny is current stable
[03-Jun-2009 14:55:42] <mrayzenoss> 5.0
[03-Jun-2009 14:55:43] <CapriCoRN^80> mrayzenoss: i want to monitor my bandwidth
[03-Jun-2009 14:55:45] <mrayzenoss> Etch was 4.0
[03-Jun-2009 14:55:47] <rmatte> Sid Phillips
[03-Jun-2009 14:55:50] <kdub> makes more sense, i asked in #debian one day, and got chewed out for asking
[03-Jun-2009 14:55:57] <CapriCoRN^80> mrayzenoss: please refer me some tutorials
[03-Jun-2009 14:56:07] <chudler> wow, really? well, it is in the FAQ.
[03-Jun-2009 14:56:38] <CapriCoRN^80> mrayzenoss: I have configured my cisco router and i was getting CPU and memory graphs
[03-Jun-2009 14:56:42] <mrayzenoss> CapriCoRN^80: I assume you have a router or switch that exposes bandwidth in an OID you can hit via SNMP
[03-Jun-2009 14:56:44] <mrayzenoss> right on
[03-Jun-2009 14:57:39] <CapriCoRN^80> mrayzenoss:how can i set it to monitor bandwidth ?
[03-Jun-2009 14:57:50] <mrayzenoss> http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch06s02.html
[03-Jun-2009 14:58:05] <mrayzenoss> CapriCoRN^80: you'll set up an SNMP datasource to watch the values on the OID
[03-Jun-2009 14:58:20] <mrayzenoss> and a threshold to watch for spikes
[03-Jun-2009 14:58:27] <mrayzenoss> same thing you'd do with MRTG
[03-Jun-2009 15:00:09] <CapriCoRN^80> mrayzenoss: ok thanks for this link .. i will check it tomorrow
[03-Jun-2009 15:00:26] <mrayzenoss> kdub: ok, let's start from scratch. apt-get remove zenoss-stack
[03-Jun-2009 15:00:33] <mrayzenoss> kdub: dpkg --purge zenoss-stack
[03-Jun-2009 15:00:47] <YC> Is there any kind of logging for the google maps integration? I'm trying to find out why my new install isn't placing the location dots.
[03-Jun-2009 15:00:47] <mrayzenoss> delete everything zenoss related in /usr/local/zenoss
[03-Jun-2009 15:00:55] <mrayzenoss> and /tmp/ if there's anything left over
[03-Jun-2009 15:01:07] <mrayzenoss> YC: 2.4.0 or 2.4.1?
[03-Jun-2009 15:01:30] <mrayzenoss> YC: there were Google map issues with 2.4.0
[03-Jun-2009 15:01:44] <mrayzenoss> that said, there may be some logging in z2.log I believe
[03-Jun-2009 15:01:56] <YC> 2.4.1
[03-Jun-2009 15:02:38] <mrayzenoss> YC: what are you entering in the Address fields?
[03-Jun-2009 15:03:01] <YC> Example: "668 High St, Lower Hutt, Wellington, New Zealand"
[03-Jun-2009 15:03:10] <YC> Resolves fine with maps.google.com
[03-Jun-2009 15:04:05] <mrayzenoss> oops. Could you repost that?
[03-Jun-2009 15:04:07] <YC> Welcome back?
[03-Jun-2009 15:04:10] <YC> [08:03] <YC> Example: "668 High St, Lower Hutt, Wellington, New Zealand"
[03-Jun-2009 15:04:11] <YC> [08:03] <YC> Resolves fine with maps.google.com
[03-Jun-2009 15:04:31] <YC> This is an excellent product, BTW. Much nicer than opennms so far.
[03-Jun-2009 15:05:01] <kdub> mrayzenoss: done and done
[03-Jun-2009 15:05:02] <mrayzenoss> YC: glad to hear it, let me see what happens when I enter that address
[03-Jun-2009 15:05:12] <kdub> mrayzenoss: nothing in /opt either
[03-Jun-2009 15:07:05] <mrayzenoss> YC: hmm.. maps for me, somewhere near Mitchell Park?
[03-Jun-2009 15:07:19] <mrayzenoss> kdub: ok, grab the 2.4.1 Deb
[03-Jun-2009 15:07:48] <YC> Yep.
[03-Jun-2009 15:07:54] <kdub> mrayzenoss: there is still a /home/zenoss and things in var/cache, /var/lib and /etc related to it though
[03-Jun-2009 15:08:22] <mrayzenoss> kdub: wipe it all out
[03-Jun-2009 15:08:30] <YC> For me, any maps that turn up are centred dead on the equator, on the international date line.
[03-Jun-2009 15:08:42] <mrayzenoss> YC: I assume you don't have any Javascript blocking?
[03-Jun-2009 15:08:43] <YC> Maps from the location tab are blank.
[03-Jun-2009 15:08:58] <YC> Nope. Standard firefox install.
[03-Jun-2009 15:08:59] <mrayzenoss> YC: have you entered the Google Maps API?
[03-Jun-2009 15:09:35] <mrayzenoss> on the Settings?
[03-Jun-2009 15:09:37] <YC> Yep.
[03-Jun-2009 15:09:52] <mrayzenoss> weird
[03-Jun-2009 15:09:59] <YC> On dashboard view, it shows a map, no dots.
[03-Jun-2009 15:10:22] * YC tries IE.
[03-Jun-2009 15:10:39] <YC> Heh. IE8 is not a fan of zenoss.
[03-Jun-2009 15:11:14] <kdub> mrayzenoss:running 'dpkg -i zenoss-stack_2.4.1_i386.deb'
[03-Jun-2009 15:12:16] <kdub> mrayzenoss: just a thought, could firewalls get in the way of the installation?
[03-Jun-2009 15:12:41] <mrayzenoss> kdub: not really
[03-Jun-2009 15:13:05] <kdub> alright, it looks like its doing a normal install, its working on 'setting up zenoss-stack'
[03-Jun-2009 15:13:25] <kdub> didnt make it past this step last time... :/
[03-Jun-2009 15:14:02] <mrayzenoss> YC: yeah, IE8 won't be supported until the next release
[03-Jun-2009 15:14:38] <kdub> mrayzenoss: weird mysql error http://pastebin.ca/1446692
[03-Jun-2009 15:15:09] <mrayzenoss> kdub: is another mysql running?
[03-Jun-2009 15:15:33] <kdub> ps ax says 2 are running, i didnt start them directly though
[03-Jun-2009 15:15:53] <mrayzenoss> well, the installer lays one down and runs it
[03-Jun-2009 15:16:32] <kdub> i should probably kill both and try again
[03-Jun-2009 15:16:57] <mrayzenoss> yeah, you might want to remove the other instance if you're not using it
[03-Jun-2009 15:19:44] <rmatte> zenoss looks horrible in IE8 unless you turn on IE7 compatibility mode
[03-Jun-2009 15:19:46] <rmatte> then it looks fine
[03-Jun-2009 15:20:10] <YC> I'm fine with IE8 not working.
[03-Jun-2009 15:20:18] <rmatte> firefox is where it's at
[03-Jun-2009 15:20:22] <YC> It'd be nice if I could figure out why my locations are unhappy though.
[03-Jun-2009 15:20:26] <YC> Chome 2 is very nice also..
[03-Jun-2009 15:20:28] <rmatte> unhappy?
[03-Jun-2009 15:20:50] <mrayzenoss> YC: do you have multiple Locations?
[03-Jun-2009 15:20:56] <rmatte> once chrome gets more robust linux support, then I'll start using it
[03-Jun-2009 15:21:11] <YC> Yep, do you want me to list them all?
[03-Jun-2009 15:21:17] <YC> (Only 4 or so)
[03-Jun-2009 15:21:22] <mrayzenoss> well, do any of them work?
[03-Jun-2009 15:21:29] <YC> No, none.
[03-Jun-2009 15:21:48] <rmatte> did you go under each location, click the little edit link, and specify the location string?
[03-Jun-2009 15:22:06] <rmatte> and did you then click on the map tab for each location and make sure it shows up?
[03-Jun-2009 15:23:48] <YC> That did hte trick. Apparently I'm a moron, sorry.
[03-Jun-2009 15:24:03] <mrayzenoss> Zenoss isn't completely intuitive, no worries
[03-Jun-2009 15:24:12] <YC> I was using the location name as the address.
[03-Jun-2009 15:24:19] <YC> Hey, it's better than the others I've used.
[03-Jun-2009 15:25:01] <rmatte> zenoss is nice but still has it's little problems, a few more years and it'll probably be much more powerful
[03-Jun-2009 15:25:20] <mrayzenoss> well, next month marks 1 year of having QA
[03-Jun-2009 15:26:26] <YC> QA's far less fun than writing new features, everyone knows this.
[03-Jun-2009 15:28:48] <rmatte> I'd like to see this get fixed at some point: http://dev.zenoss.com/trac/ticket/5071
[03-Jun-2009 15:29:09] <rmatte> took upwards of like 15-20 mins for me to load the zProperties pages on some servers that I was setting up today
[03-Jun-2009 15:29:15] <mrayzenoss> yeah, that's a beast of a ZenPack
[03-Jun-2009 15:29:29] <rmatte> there's something that Zenoss is doing though that's causing that lag
[03-Jun-2009 15:29:38] <rmatte> there must be a way around it
[03-Jun-2009 15:30:10] <rmatte> it was probably overlooked initially since most Zenpacks were tiny
[03-Jun-2009 15:30:32] <rocket> I would guess it a broken cache somewhere ...
[03-Jun-2009 15:30:53] <rocket> something isnt using the caching structure .. (without looking)
[03-Jun-2009 15:31:22] <mrayzenoss> supposedly the 2.11.2 upgrade is speeding up a few things
[03-Jun-2009 15:32:10] <rmatte> did anyone ever write a patch for that Linux SSH issue?
[03-Jun-2009 15:32:21] <rocket> rmatte: I am working it
[03-Jun-2009 15:32:26] <rmatte> rocket: nice
[03-Jun-2009 15:32:28] <rocket> rmatte: give me a sec
[03-Jun-2009 15:33:29] <kdub> what about trying the *.bin file?
[03-Jun-2009 15:34:15] <mrayzenoss> kdub: the only practical difference would be that it doesn't register with apt
[03-Jun-2009 15:34:32] <kdub> and i dont particularly care about that, ill give it a try
[03-Jun-2009 15:34:58] <rmatte> doesn't need to register with apt hehe, /usr/local/zenoss/uninstall
[03-Jun-2009 15:34:59] <rmatte> hehe
[03-Jun-2009 15:35:08] <rmatte> works nicely
[03-Jun-2009 15:35:43] <AngryParsley> $ZENHOME is supposed to be /usr/local/zenoss, right? and /usr/local/zenoss/zenoss is for the web app part of zenoss, correct?
[03-Jun-2009 15:36:00] <rmatte> nope
[03-Jun-2009 15:36:03] <AngryParsley> nope?
[03-Jun-2009 15:36:25] <rmatte> zenoss@gen01:/home/rmatte$ echo $ZENHOME
[03-Jun-2009 15:36:31] <rmatte> /usr/local/zenoss/zenoss
[03-Jun-2009 15:37:07] <AngryParsley> oh, but the check_http/tcp/etc scripts are all i/ /usr/local/zenoss/common/libexec, and httpmonitor wants to execute $ZENHOME/bin/check_http
[03-Jun-2009 15:37:34] <rocket> rmatte: patch applied to ticket 5407
[03-Jun-2009 15:37:55] <rocket> sorry I got delayed trying to win a contest
[03-Jun-2009 15:38:16] <rocket> rmatte: it really is a stupidly simple patch though ..
[03-Jun-2009 15:38:25] <AngryParsley> am I supposed to copy check_* to /usr/local/zenoss/zenoss/bin/ ?
[03-Jun-2009 15:38:39] <AngryParsley> because that seems kind of hack-ish
[03-Jun-2009 15:38:53] <mrayzenoss> no
[03-Jun-2009 15:39:14] <mrayzenoss> AngryParsley: see about clearing out zCommandPath on that device
[03-Jun-2009 15:39:25] <mrayzenoss> and then fix the path in the Datasource
[03-Jun-2009 15:40:52] <AngryParsley> ah, zCommandPath was set to /opt/libexec/zenoss
[03-Jun-2009 15:41:29] <AngryParsley> but the data source is an HttpMonitor. I don't see a path to set, just the regular hostname/port/ssl options
[03-Jun-2009 15:42:08] <mrayzenoss> after you cleared out that zProperty, go to the datasource and try hitting "test"
[03-Jun-2009 15:42:32] <mrayzenoss> see what it reports
[03-Jun-2009 15:45:24] <AngryParsley> I don't see a test option. this isn't a COMMAND data source type. it's an HttpMonitor data source type
[03-Jun-2009 15:45:25] <rmatte> is it normal for all devices to report as unpingable during an auto-discovery?
[03-Jun-2009 15:46:09] <AngryParsley> if I look in $ZENHOME/Products/HttpMonitor/datasources/ and edit HttpMonitorDataSource.py, I can see in the source it's adding $ZENHOME to the front of the command it's executing
[03-Jun-2009 15:46:16] <AngryParsley> cmd = '$ZENHOME/libexec/' + \
[03-Jun-2009 15:46:16] <AngryParsley> RRDDataSource.RRDDataSource.getCommand(self, context, cmd)
[03-Jun-2009 15:47:43] <AngryParsley> so that's where I got the idea that maybe $ZENHOME was set incorrectly
[03-Jun-2009 15:47:57] <AngryParsley> but it's not, so I'm guessing I'm doing something else wrong
[03-Jun-2009 15:48:32] <gemineye> Does Zenoss recommend not installing core in a VM or should that work fine?
[03-Jun-2009 15:49:37] <chudler> anyone know offhand where rrdtoolmodule.so comes from in $ZENHOME/lib/python/
[03-Jun-2009 15:50:03] <chudler> I can't find a source for it. I guess I will download zenoss source and find out
[03-Jun-2009 15:50:47] <AngryParsley> chudler: probably http://oss.oetiker.ch/rrdtool/
[03-Jun-2009 15:55:24] <chudler> yep, your right. at least in the 2.4.1 source it is in externallibs/rrdtool-1.3.5[/bindings/python/rrdtoolmodule.c], besides a small patch apparently written by Zenoss.
[03-Jun-2009 15:57:13] <chudler> I hoped this was true, since it means I could use graphv. Unfortunately, I cannot tell if python bindings are supported with graphv. The docs say ruby and perl is.
[03-Jun-2009 15:57:50] <AngryParsley> the command it's trying to run is $ZENHOME/libexec/check_http blah blah blah blah
[03-Jun-2009 15:57:55] <rmatte> gemineye: I have tons of instances of core running in different VMs, works fine
[03-Jun-2009 15:58:00] <AngryParsley> if I run the same command as zenoss on that machine, I get
[03-Jun-2009 15:58:12] <AngryParsley> bash: /usr/local/zenoss/zenoss/libexec/check_http: No such file or directory
[03-Jun-2009 15:58:26] <mrayzenoss> gemineye: lots of people use VMs
[03-Jun-2009 15:58:31] <AngryParsley> since check_http is in /usr/local/zenoss/libexec
[03-Jun-2009 15:58:46] <AngryParsley> and $ZENHOME is /usr/local/zenoss/zenoss
[03-Jun-2009 15:59:12] <mrayzenoss> AngryParsley: did you set $ZENHOME to that? That seems weird
[03-Jun-2009 15:59:49] <AngryParsley> rmatte says that is the correct setting for $ZENHOME
[03-Jun-2009 16:03:21] <gemineye> ok, thanks.
[03-Jun-2009 16:08:48] <AngryParsley> http://dev.zenoss.org/trac/ticket/3201
[03-Jun-2009 16:08:51] <AngryParsley> ah, it's a bug in zenoss
[03-Jun-2009 16:47:02] <YC> Is it possible to manually draw links on the google maps?
[04-Jun-2009 00:00:42] [disconnected at Thu Jun 4 00:00:42 2009]
[04-Jun-2009 00:00:42] [connected at Thu Jun 4 00:00:42 2009]
[04-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[04-Jun-2009 01:39:13] <torocatala> hi
[04-Jun-2009 07:27:34] <mr_claus_> hi, i monitor postfix but zenoss always disconnects before the greeting is finished, that fill my logs
is there a way to do the greeting with zenoss?[04-Jun-2009 07:56:54] <tehhobbit> is there a good way to make a perf graph for switches with the total pps and mbit ?
[04-Jun-2009 08:04:08] <coofamani> define total
[04-Jun-2009 08:06:47] <tehhobbit> ?
[04-Jun-2009 08:14:49] <coofamani> when you say total pps and mbit, do you mean total per interface or total for a switch
[04-Jun-2009 08:28:41] <tehhobbit> for a switch
[04-Jun-2009 08:36:43] <coofamani> if its just for a few switches you could do an aggregate multigraph report
[04-Jun-2009 09:28:41] <kdub> what is the difference between a 'system' and a 'group'?
[04-Jun-2009 09:32:09] <mrayzenoss> they're essentially the same thing in the code
[04-Jun-2009 09:32:21] <mrayzenoss> they're just logical ways of organizing your devices
[04-Jun-2009 09:32:27] <rocket> morning matt
[04-Jun-2009 09:32:41] <mrayzenoss> if you're sorting by hardware or OS or something, you'd probably use a System
[04-Jun-2009 09:33:07] <mrayzenoss> if you're sorting by use or function (i.e. "QA" or "Accounting") you'd use a Group
[04-Jun-2009 09:33:25] <mrayzenoss> and devices can belong to multiple Groups and multiple Systems
[04-Jun-2009 09:33:45] <mrayzenoss> good morning rocket, I'm checking in your patch to trunk
[04-Jun-2009 09:35:52] <sergeymasushko> I have an issue. I've added ifInOctets (1.3.6.1.2.1.2.2.1.10.1) and ifOutOctets (1.3.6.1.2.1.2.2.1.16.1) as datasources in the performance template. But graphs are not shown. However when I do test on the datasource edit page I get successful result... where can be the issue?
[04-Jun-2009 09:36:00] <rocket> mrayzenoss: sounds good
[04-Jun-2009 09:36:06] <rocket> mrayzenoss: working on the docs for you
[04-Jun-2009 09:36:07] <sergeymasushko> morning guys
[04-Jun-2009 09:36:19] <rocket> mrayzenoss: and there shouldnt have been a dep on mysql for the aix one
[04-Jun-2009 09:37:32] <rmatte> ummm weird, in 2.4.1, I don't see the actual threshold value on graphs
[04-Jun-2009 09:37:52] <rmatte> in the legend I just see the threshold name instead of something like CPU > 90
[04-Jun-2009 09:42:20] <kdub> thanks mrayzenoss
[04-Jun-2009 09:43:20] <rmatte> oh well, at least it draws the threshold line
[04-Jun-2009 09:44:17] <mrayzenoss> rmatte: debating your ZenPack slowness ticket right now
[04-Jun-2009 09:44:28] <mrayzenoss> rmatte: might be a blocker
[04-Jun-2009 09:44:53] <mrayzenoss> apparently it showed up on an Enterprise upgrade and took 10 hours
[04-Jun-2009 09:50:19] <rmatte> 10 hours?
[04-Jun-2009 09:50:27] <rmatte> I've never had it go that long
[04-Jun-2009 09:50:40] <rmatte> I've never seen it go more than like 15-20 minutes
[04-Jun-2009 09:52:09] <rocket> mrayzenoss: all the thresholds will need to be 2.4 I believe
[04-Jun-2009 09:52:48] <rmatte> what's a "blocker" ?
[04-Jun-2009 09:53:20] <rocket> blocker = bug that forces a release to fix it before the next version can be released
[04-Jun-2009 09:53:28] <rocket> eg the bug blocks the software release
[04-Jun-2009 09:53:54] <rmatte> ah
[04-Jun-2009 09:53:59] <rmatte> that's what I figured
[04-Jun-2009 09:54:09] <rmatte> yeh, it's a pretty crappy bug
[04-Jun-2009 09:54:15] <rmatte> I'd love to see it fixed hehe
[04-Jun-2009 09:55:17] <rmatte> oh, it's not only based on a single large Zenpack but on total amount
[04-Jun-2009 09:55:18] <rmatte> ouch
[04-Jun-2009 09:55:27] <mrayzenoss> rocket: yeah, all thresholds are marked for 2.4
[04-Jun-2009 09:55:40] <mrayzenoss> rmatte: correct, blockers all must be fixed before a release
[04-Jun-2009 09:55:55] <rmatte> cool
[04-Jun-2009 09:56:36] <mrayzenoss> all the high's have to be accounted for to do a release, either fixed, release-noted or moved out of the release
[04-Jun-2009 09:56:57] <rmatte> cool
[04-Jun-2009 09:57:11] <mrayzenoss> I'm adding the post King Crab release to Trac right now
[04-Jun-2009 09:57:21] <mrayzenoss> "Stone Crab", ETA Spring 2010
[04-Jun-2009 09:57:29] <mrayzenoss> that just sounds crazy
[04-Jun-2009 09:58:50] <cgibbons> hmmmm
[04-Jun-2009 10:08:47] <coofamani> Softshell Crab
[04-Jun-2009 10:09:12] <rmatte> Stone Crab, nice
[04-Jun-2009 10:09:28] <rmatte> maybe the next one can be Kryptonite Crab
[04-Jun-2009 10:14:50] <rmatte> waiting for a zProperties page to load right now actually...
[04-Jun-2009 10:30:22] <rmatte> ok, what do I have to do to have Zenoss auto-detect win services on a server via WMI?
[04-Jun-2009 10:30:51] <rmatte> I have wmiMonitorIgnore set to false, I have the correct username and password specified, and I have the WinServices collector plugin applied to the devices
[04-Jun-2009 10:30:57] <rmatte> I've pushed changed to the collector
[04-Jun-2009 10:31:04] <rmatte> when I model them, no Win Services show up
[04-Jun-2009 10:31:18] <rocket> mrayzenoss: sent you another email
[04-Jun-2009 10:32:56] <rocket> mrayzenoss: hopefully this aix file will not include the mysql dep
[04-Jun-2009 10:40:33] <rmatte> ah, finally got it working
[04-Jun-2009 10:48:11] <sergeymasushko> can somebody clarify mechanism of aging events...
[04-Jun-2009 10:59:08] <rmatte> it'd be nice if you could actually age critical events
[04-Jun-2009 10:59:14] <rmatte> you can age everything else except for those
[04-Jun-2009 11:01:11] <sergeymasushko> just interesting... I have events with severity 'warning' (diskspace utilization) these events updated 13 times per hour... but each hour these events aged. Is it correct behavior?
[04-Jun-2009 11:02:11] <sergeymasushko> Event Aging Threshold (hours) set to 1 and Don't Age This Severity and Above set to 'Error'
[04-Jun-2009 11:06:24] <sergeymasushko> According to the manual (http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch07s01.html) zenoss should age events only if the event counter is updated during 'Event Aging Threshold'. Am I correct?
[04-Jun-2009 11:08:19] <mrayzenoss> The aging gets reset every time the counter gets updated
[04-Jun-2009 11:10:24] <sergeymasushko> mrayzenoss: Matt, so why it clears events every hour when event counter updated 12 times per hour.
[04-Jun-2009 11:16:04] <sergeymasushko> here is an example from the event history http://pastebin.com/m76c75c7d
[04-Jun-2009 11:18:11] scfh_ is now known as scfh
[04-Jun-2009 11:19:39] <mrayzenoss> those are separate events
[04-Jun-2009 11:19:52] <mrayzenoss> or, they're not getting consolidated even though they look like the same thing
[04-Jun-2009 11:25:05] <sergeymasushko> why do you think that those events are separate?
[04-Jun-2009 11:32:19] <sergeymasushko> I've re checked events there are the same device, the same component, the same value...
[04-Jun-2009 11:35:04] <sergeymasushko> perr0: hi, have you experienced such issue on you freebsd installation?
[04-Jun-2009 11:36:57] <perr0> is that in your pastebin?
[04-Jun-2009 11:37:10] <perr0> sergeymasushko: ^
[04-Jun-2009 11:37:55] <sergeymasushko> yes
[04-Jun-2009 11:39:44] <perr0> that happens when you mount a USB? That happens when I mount .isos'
[04-Jun-2009 11:40:02] <perr0> yep
[04-Jun-2009 11:40:54] <perr0> so when the developers need to install a package I have a script that mounts 7 isos and it drives me creazy cause they never unmount them and I get those errors
[04-Jun-2009 11:41:07] <perr0> or warnings
[04-Jun-2009 11:49:38] <rmatte> tell them that they get an hour before it auto-unmounts
[04-Jun-2009 11:49:40] <rmatte> >:)
[04-Jun-2009 11:52:04] <sergeymasushko> hm... but that device is already mounted there for a long time without remount.
[04-Jun-2009 11:53:39] <perr0> rmatte: does that happen to you too?
[04-Jun-2009 11:53:57] <perr0> I haven't had much time to look into these past weeks
[04-Jun-2009 11:54:05] <perr0> it*
[04-Jun-2009 12:01:08] <sergeymasushko> ok thanks for the point... I'll try to dig more into it.
[04-Jun-2009 12:06:11] <Mark09> I see on the compare page that Pro has "WMI based modeling of disk, CPU, h/w & s/w inventory" - does that mean no nsclient on windows ?
[04-Jun-2009 12:06:35] <Mark09> do you have to install another client on the windows boxes or can it talk wmi?
[04-Jun-2009 12:30:27] <rmatte> Mark09: it can talk WMI
[04-Jun-2009 12:30:48] <rmatte> it can gather all of the info that Core can (with SNMP) via WMI
[04-Jun-2009 17:23:39] <gemineye> Anyone here?
[04-Jun-2009 17:25:35] <mrayzenoss> it's been quiet today
[04-Jun-2009 17:29:35] <gemineye> But you're here, so good
[04-Jun-2009 17:29:39] <gemineye> I have two questions for ya
[04-Jun-2009 17:30:30] <gemineye> Does Zenoss support dynamic polling frequency, either Enterprise or Core?
[04-Jun-2009 17:31:44] <gemineye> Let's say a device is polled every 5 minutes. If it detects an outage, I have my alerts setup to not notify me unless an event occurs 3 times in a row. Can Zenoss be smart and increase the polling frequency when it detects the outages so instead of waiting 15 minutes for an alert, it'll check every 1 minute when it sees the outage?
[04-Jun-2009 17:32:08] <gemineye> Possibly done through transforms?
[04-Jun-2009 17:33:55] <gemineye> mrayzenoss: WAKE UP!
[04-Jun-2009 17:34:32] <mrayzenoss> nope
[04-Jun-2009 17:35:02] <mrayzenoss> talking to a dev about it
[04-Jun-2009 17:35:09] <gemineye> ok
[04-Jun-2009 17:37:16] <gemineye> I have a second question when you're ready for it
[04-Jun-2009 17:37:33] <mrayzenoss> yeah, he said we could write that, but haven't been asked to
[04-Jun-2009 17:37:41] <gemineye> write it
[04-Jun-2009 17:37:44] <mrayzenoss> and then it could be triggered with a transform
[04-Jun-2009 17:37:47] <mrayzenoss> heh
[04-Jun-2009 17:39:02] <mrayzenoss> the priortized backlog is long, but L2 awareness has been making a big stink today
[04-Jun-2009 17:39:25] <gemineye> Is that something that could be easy to activate either as a feature of Core or Enterprise, or would it be a custom enhancement on a per user basis?
[04-Jun-2009 17:39:27] <mrayzenoss> next question
[04-Jun-2009 17:39:32] <gemineye> That's not the next question
[04-Jun-2009 17:39:37] <gemineye> What's L2?
[04-Jun-2009 17:39:49] <gemineye> That isn't the second question either
[04-Jun-2009 17:39:49] <mrayzenoss> it would not be an easy change
[04-Jun-2009 17:40:01] <mrayzenoss> since there are so many different collectors
[04-Jun-2009 17:40:20] <mrayzenoss> and how do you decide which ones to accelerate on error
[04-Jun-2009 17:40:53] <mrayzenoss> right now they're not very smart
[04-Jun-2009 17:40:56] <gemineye> Basically, Nimsoft bought Cittio which is what one of our departments has been using for monitoring and I'm trying to get them to move to Zenoss. These are features they currently enjoy with WatchTower and they'd like to see available in Zenoss.
[04-Jun-2009 17:42:09] <gemineye> There could be a zProperty for each device that enables the dynmic polling
[04-Jun-2009 17:42:19] <mrayzenoss> King Crab is going to have a lot of improvements around the durability and performance for collectors
[04-Jun-2009 17:42:31] <gemineye> What version is KC?
[04-Jun-2009 17:42:35] <mrayzenoss> next
[04-Jun-2009 17:42:42] <gemineye> #?
[04-Jun-2009 17:42:47] <mrayzenoss> it's possible collection cycles may be moving off of the collector and onto devices
[04-Jun-2009 17:42:54] <gemineye> That would be nice
[04-Jun-2009 17:43:06] <mrayzenoss> we call it King Crab since marketing gets to pick version numbers
[04-Jun-2009 17:43:12] <mrayzenoss> 2.5 if you like
[04-Jun-2009 17:43:27] <gemineye> I do like, as opposed to 2.4.2 or 3.0
[04-Jun-2009 17:43:32] <mrayzenoss> correct
[04-Jun-2009 17:43:53] <mrayzenoss> maintenance releases like 2.4.2 aren't going to get substantial new features, they're supposed to be bug fixes
[04-Jun-2009 17:43:59] <gemineye> But certainly not 2.4.1.0.0.0.0.0.1
[04-Jun-2009 17:44:10] <mrayzenoss> well, that's what zenpatch is for
[04-Jun-2009 17:44:26] <mrayzenoss> 2.4.2 will roll-up the zenpatchable things that are against the 2.4.x branch
[04-Jun-2009 17:44:31] <gemineye> When will Zenoss release numbers use imaginary numbers?
[04-Jun-2009 17:44:43] <gemineye> In my dreams?
[04-Jun-2009 17:44:45] <mrayzenoss> you mean like 2.4.kingcrab?
[04-Jun-2009 17:45:04] <mrayzenoss> and BTW, I added the post-King Crab release to Trac today... "Stone Crab"
[04-Jun-2009 17:45:06] <gemineye> no, like 2.5i
[04-Jun-2009 17:45:22] <mrayzenoss> next question
[04-Jun-2009 17:45:52] <mrayzenoss> and King Crab is tentatively scheduled for October
[04-Jun-2009 17:46:17] <gemineye> Will King Crab include report enhancements?
[04-Jun-2009 17:46:28] <mrayzenoss> there are some changes that have already gone into trunk I think
[04-Jun-2009 17:46:45] <mrayzenoss> like making straight python calls in reports
[04-Jun-2009 17:49:58] <mrayzenoss> I gotta get going, email me if you have any other questions
[04-Jun-2009 22:56:28] <zenchild> Anyone know where to find a list of xml-rpc methods available from zenoss?
[05-Jun-2009 00:00:42] [disconnected at Fri Jun 5 00:00:42 2009]
[05-Jun-2009 00:00:42] [connected at Fri Jun 5 00:00:42 2009]
[05-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[05-Jun-2009 03:41:53] <Dieterbe> I don't understand the cpu usage graph ( http://www.kangaroot.be/images/constant-cpu-usage-weird.png ). as you can see the load varies a lot, as does the mysql commands. however the cpu usage graph is almost constant o_O what's going on?
[05-Jun-2009 04:31:47] <rhettlap> load average and cpu usage are two separate things though are commonly regarding as being the same or mostly similar
[05-Jun-2009 04:31:54] <Dieterbe> indeed
[05-Jun-2009 04:35:56] <rhettlap> can you verify that the average cpu utilization is indeed variable and not static as it appears in your graph?
[05-Jun-2009 04:39:05] <Dieterbe> yes i have top open, it fluctuates
[05-Jun-2009 04:40:34] <rhettlap> well you need to know what the 5 minute average is of cpu usage
[05-Jun-2009 04:41:34] <rhettlap> and then compare that to the snmp output
[05-Jun-2009 04:42:14] <rhettlap> either buy using rrdtool fetch to the datasource or by walking or getting the output from the appropriate OID
[05-Jun-2009 04:43:03] <rhettlap> sorry to bail out on you but i have to go get some rest
[05-Jun-2009 04:43:15] <Dieterbe> rhettlap: that's ok. i'll try manual snmpwalking
[05-Jun-2009 04:43:19] <Dieterbe> i just need to find out the OID
[05-Jun-2009 04:43:38] <rhettlap> you can go into the template and see the oid
[05-Jun-2009 04:45:14] <rhettlap> snmpwalk -v 2c -c community device 1.3.6.1.4.1.2021.11
[05-Jun-2009 04:45:17] <rhettlap> that should do it for you
[05-Jun-2009 04:45:59] <rhettlap> or if you want to see the raw oid instead of the translation through the -On flag in there
[05-Jun-2009 04:46:03] <rhettlap> good luck
[05-Jun-2009 04:46:07] <Dieterbe> thanks man
[05-Jun-2009 04:46:20] <Dieterbe> and where exactly did you find this? I'm looking at /Devices /Server /Linux /Templates /Device /CPU Utilization, but don't really see it
[05-Jun-2009 04:46:48] <rhettlap> also to look at the data source you can use 'rrdtool fetch pathtodatasource AVERAGE -r 300 -s -1h'
[05-Jun-2009 04:46:55] <rhettlap> that will show the results for the past hour
[05-Jun-2009 04:47:42] <rhettlap> it should be under the 'Device' template
[05-Jun-2009 04:47:54] <Dieterbe> ah yeah, i just found it
[05-Jun-2009 04:47:58] <rhettlap> then it will be listed under data sources
[05-Jun-2009 04:47:58] <rhettlap> cool
[05-Jun-2009 04:49:58] <rhettlap> well i am off for good now. good luck again.
[05-Jun-2009 04:50:04] <Dieterbe> thanks
[05-Jun-2009 08:24:04] <cgibbons> hurm
[05-Jun-2009 08:24:29] <rocket> Hello
[05-Jun-2009 08:34:52] <etank> is it possible to change the hostname that is used when sending alerts?
[05-Jun-2009 08:35:06] <etank> i want to change whatever it is that is used in "<a href="%(eventUrl)s">Event Detail</a>"
[05-Jun-2009 08:35:41] <etank> the hostname of the server is foo.com but i have a dns alias that i use for the zenoss install that is different
[05-Jun-2009 08:35:53] <etank> i want it to use zenoss.foo.com
[05-Jun-2009 09:00:31] <perr0> etank: what?
[05-Jun-2009 09:02:34] <etank> perr0: in the properties of an alerting rule (i.e /ZenUsers /etank /Alerting Rules /Event Log Errors ) you see things like <a href="%(eventUrl)s">Event Detail</a> for the body of the message
[05-Jun-2009 09:02:52] <etank> to give you a link to acknowledge, view, etc the event
[05-Jun-2009 09:03:04] <etank> my servers hostname is rethink.lms.lexmark.com
[05-Jun-2009 09:03:10] <perr0> I am just confused as to your question is changing the hostname
[05-Jun-2009 09:03:31] <etank> and i have a cname set to zenoss.lms.lexmark.com for the same box for zenoss use
[05-Jun-2009 09:03:35] <perr0> in*
[05-Jun-2009 09:03:37] <perr0> ok
[05-Jun-2009 09:03:54] <etank> i would like the urls in the alerts to be zenoss.lms.lexmark.com instead of the hostname of the server
[05-Jun-2009 09:04:26] <perr0> gotcha
[05-Jun-2009 09:05:02] <etank> im just not sure where the variable is read from. should be easy enough to change
[05-Jun-2009 09:05:07] <rmatte> any particular reason why 2.4.1 takes more than twice as long to start than 2.3.3?
[05-Jun-2009 09:05:51] <perr0> right. Have you tried setting the host name of zenoss.xxx.lemark and restarting zenoss?
[05-Jun-2009 09:06:01] <perr0> in your server?
[05-Jun-2009 09:06:45] <chudler> I think maybe $ZENHOME/etc/zenoactions.conf has an option zopeurl
[05-Jun-2009 09:07:02] <chudler> set it like "zopeurl https://example.com"
[05-Jun-2009 09:07:16] <chudler> I mean "zenactions.conf"
[05-Jun-2009 09:07:52] <etank> perr0: yeah that would be one way to do it
[05-Jun-2009 09:09:19] <etank> chudler: my $ZENHOME/etc/zenoactions.conf is blank
[05-Jun-2009 09:10:38] <etank> ok i just set zopeurl http://zenoss.lms.lexmark.com in my zenactions.conf file
[05-Jun-2009 09:12:10] <chudler> a restart is likely to be necessary
[05-Jun-2009 09:12:23] <etank> seems i could have changed it at http://zenoss.lms.lexmark.com:8080/zport/About/showDaemonXMLConfig?daemon=zenactions
[05-Jun-2009 09:12:46] <perr0> let us know did it work?
[05-Jun-2009 09:12:53] <perr0> either way
[05-Jun-2009 09:13:56] <chudler> hm cool, I never edit the configs this way. I guess I never noticed it was even there.
[05-Jun-2009 09:15:13] <etank> chudler: i had not seen it either
[05-Jun-2009 09:15:23] <etank> i reads the file without a restart though
[05-Jun-2009 09:15:29] <etank> and it works like a champ
[05-Jun-2009 09:15:45] <etank> i did set it in the .conf though not through the web
[05-Jun-2009 09:15:51] <etank> thanks perr0 and chudler
[05-Jun-2009 09:16:07] <chudler> cool. you are welcome!
[05-Jun-2009 09:16:58] <perr0> the zenoactions.conf did it huh? Cool now we know something new
[05-Jun-2009 09:17:06] <etank> now i just need to learn how to suppress events that i dont care about at all
[05-Jun-2009 09:17:11] <etank> like DCOM junk
[05-Jun-2009 09:46:25] <kdub> is there a place where i can enter the database password? i'm trying zenrestore, and it keeps telling me the password login for mysql is rejected
[05-Jun-2009 09:49:59] <kdub> nvm, i might have just figured it out...
[05-Jun-2009 10:16:51] <venturaville> odd question for someone who might have a better way of doing this:
[05-Jun-2009 10:17:29] <venturaville> I have a device class where I need to consolidate the performance numbers from a bunch of other devices that are 'somehow' linked to it.
[05-Jun-2009 10:18:14] <venturaville> i.e. I could do this in theory by making a command that calls back into zenoss to fetch the numbers, but this seems rather clunky
[05-Jun-2009 10:18:32] <venturaville> I was hoping there was a cleaner way to do this
[05-Jun-2009 10:19:58] <zenChild> Would a multi-graph report work?
[05-Jun-2009 10:21:52] <venturaville> not in this case because the devices behind the consolidate device can drop out or be added to at will
[05-Jun-2009 10:22:07] <venturaville> which would mean the history for the consolidation would change with it
[05-Jun-2009 10:22:34] <zenChild> hmmm... that is an interesting problem.
[05-Jun-2009 10:24:02] <cgibbons> hurm
[05-Jun-2009 10:24:04] <venturaville> I looked at that as a possible solution when I first got into zenoss, but it multi-graph is still a pretty limited tool for doing these kinds of things
[05-Jun-2009 10:24:53] <zenChild> I wonder if you could use the REST API to fetch that data out and consolidate it.
[05-Jun-2009 10:25:15] <venturaville> that is the way I am looking at now, but that is pretty poor performing and really clunky
[05-Jun-2009 10:25:32] <venturaville> I was hoping for more of a solution that is within the zenoss context (like the modeler)
[05-Jun-2009 10:25:52] <venturaville> I'm looking to do this on a rather large number of devices
[05-Jun-2009 10:25:58] <zenChild> Does anyone know if there is a documented API for REST and XML-RPC?
[05-Jun-2009 10:26:07] <venturaville> there is
[05-Jun-2009 10:26:25] <chudler> a custom graph plugin can do it. Depends on the amount of work you want to do, of course.
[05-Jun-2009 10:26:26] <zenChild> I've looked at the zenoss API, but that doesn't seem to correspond 1-for-1 to the xml-rpc calls
[05-Jun-2009 10:27:45] <venturaville> that is because some of the things in the API return python objects, which don't translate well into xml-rpc
[05-Jun-2009 10:28:07] <venturaville> learn to use zendmd and then figure out how to do things without using the objects
[05-Jun-2009 10:28:22] <venturaville> and then for the really tricky things you can add pieces of code in zope to do those things
[05-Jun-2009 10:28:59] <venturaville> example: http://www.zenoss.com/community/docs/howtos/get-a-list-of-device-names/
[05-Jun-2009 10:30:26] <zenChild> Its returning objects is probably the reason I'm getting marshalling errors I presume.
[05-Jun-2009 10:30:48] <venturaville> yep
[05-Jun-2009 10:31:23] <zenChild> Thanks for your input.
[05-Jun-2009 10:33:47] <venturaville> chudler: is there any documentation on writing custom graph plugins anywher?
[05-Jun-2009 10:49:48] <rmatte> by custom graph plugin you mean a script that can be used to generate datapoints?
[05-Jun-2009 10:50:46] <venturaville> essentially, though I have a suspicion it is probably not what I am ultimately after
[05-Jun-2009 10:51:08] <venturaville> I'm hoping for more of a device centric method of doing this
[05-Jun-2009 11:15:24] <exes> whats better about zenoss than nagios or hyperic?
[05-Jun-2009 11:32:31] <chudler> exes: I can't speak for everyone, but hyperic's agent turned us off it. I think they have been bought, and now days are apparently evolving into a EE thang.
[05-Jun-2009 11:38:17] <mrayzenoss> exes: anecdotally I've been told that Nagios can be a pain to manage large numbers of devices
[05-Jun-2009 11:38:26] <mrayzenoss> and any Nagios plugins can be used by Zenoss
[05-Jun-2009 11:38:51] <mrayzenoss> Zenoss adds performance monitoring (Nagios users would use Cactii for that)
[05-Jun-2009 11:39:00] <mrayzenoss> JMX & WMI
[05-Jun-2009 11:39:07] <mrayzenoss> the network map and google map
[05-Jun-2009 11:39:19] <mrayzenoss> event and alert schedules, maintenance windows
[05-Jun-2009 11:40:03] <mrayzenoss> there's more, but is there a particular feature you're looking for?
[05-Jun-2009 11:56:37] <venturaville> the API and extendability are killer
[05-Jun-2009 12:30:09] <exes> mrayzenoss: yes... I'll have to get back to you, thanks for your response
[05-Jun-2009 12:31:07] <venturaville> mrayzenoss: I was asking this question earlier before I think you were on
[05-Jun-2009 12:31:19] <venturaville> (11:17:28 AM) venturaville: I have a device class where I need to consolidate the performance numbers from a bunch of other devices that are 'somehow' linked to it.
[05-Jun-2009 12:31:19] <venturaville> (11:18:14 AM) venturaville: i.e. I could do this in theory by making a command that calls back into zenoss to fetch the numbers, but this seems rather clunky
[05-Jun-2009 12:31:19] <venturaville> (11:18:32 AM) venturaville: I was hoping there was a cleaner way to do this
[05-Jun-2009 12:32:16] <mrayzenoss> hmmm... I'll ask around
[05-Jun-2009 12:32:19] <venturaville> and multi-graph won't work since I need to be able to handle the other devices in and out without mangling the RRD history.... also I really need to be able to tag the consolidated device with the same tags as I do the devices (i.e. /Groups/*)
[05-Jun-2009 12:32:21] <mrayzenoss> some sort of meta device?
[05-Jun-2009 12:32:30] <venturaville> nailed it in one
[05-Jun-2009 12:32:58] <mrayzenoss> the VMware Devices ZenPack kinda has that
[05-Jun-2009 12:33:07] <mrayzenoss> where there's a Guests tab
[05-Jun-2009 12:33:13] <mrayzenoss> but you want it in perf?
[05-Jun-2009 12:33:20] <venturaville> yeah I've looked at that, however it isn't reentrant
[05-Jun-2009 12:33:42] <venturaville> i.e. the performance data is collected at the device level not at the consolidated device level
[05-Jun-2009 12:33:45] <mrayzenoss> I'll ask the devs when they get back from lunch
[05-Jun-2009 12:33:50] <venturaville> thanks
[05-Jun-2009 12:34:18] <venturaville> the VMWare plugin could benefit from something like that too................
[05-Jun-2009 12:53:23] <HKhan> 18:52:57 up 1:21, 2 users, load average: 29.97, 22.83, 11.17 << I think i'm going to have to get a new box to run zenoss on >.> Lagging like hell D:
[05-Jun-2009 12:57:51] <pinball_> 5:57pm up 905 day(s), 18:52, 1 user, load average: 0.11, 0.09, 0.09
[05-Jun-2009 12:58:00] pinball_ is now known as pinball
[05-Jun-2009 13:01:23] <HKhan>
[05-Jun-2009 13:01:28] <HKhan> wonder why mine is so high :S
[05-Jun-2009 13:01:41] <HKhan> i only have a few monitored machines
[05-Jun-2009 13:08:35] <rhettlap> wow
[05-Jun-2009 13:08:44] <rhettlap> that is some serious business HKhan
[05-Jun-2009 13:08:51] <mrayzenoss> what's the top process?
[05-Jun-2009 13:10:23] <HKhan> Zope is atm, could it be caused if my MySQL box is replying slowly? >.<
[05-Jun-2009 13:11:31] <mrayzenoss> hmmm.., what are the stats on the box?
[05-Jun-2009 13:12:11] <HKhan> what do you mean by stats? hardware spec?
[05-Jun-2009 13:12:18] <mrayzenoss> yeah
[05-Jun-2009 13:12:27] <mrayzenoss> I assume the box is something reasonable?
[05-Jun-2009 13:12:36] <HKhan> 1GHz P3 with 1GB RAM
[05-Jun-2009 13:12:52] <mrayzenoss> hmm... that's kinda pokey, how many devices?
[05-Jun-2009 13:13:05] <mrayzenoss> I do 10 on a G4 1.33ghz Powerbook
[05-Jun-2009 13:13:22] <HKhan> umm around 20 i think.... >.< i think it just fell over under the load, so can't check atm heh
[05-Jun-2009 13:14:33] <venturaville> 1GB RAM is pretty low for zenoss ...period
[05-Jun-2009 13:14:43] <mrayzenoss> yeah, my powerbook has 1.25 gigs
[05-Jun-2009 13:15:00] <mrayzenoss> it sits at about .2 on load
[05-Jun-2009 13:15:23] <HKhan> hmm will see if i can upgrade it then
[05-Jun-2009 13:16:49] <venturaville> check how much swap you are using
[05-Jun-2009 13:17:11] <mrayzenoss> yeah, I never hit my swap
[05-Jun-2009 13:17:53] <venturaville> zenoss does pretty well as long as it doesn't
[05-Jun-2009 13:18:00] <HKhan> ah its using around 400MB swap
[05-Jun-2009 13:18:07] <mrayzenoss> yeah, that's gonna kill you
[05-Jun-2009 13:18:08] <venturaville> I had to do some tuning to get it to work well on my 8 year old computer at home with similar specs
[05-Jun-2009 13:18:25] <venturaville> HKhan ... you might want to see what else is running
[05-Jun-2009 13:18:30] <venturaville> are you trying to run this on your desktop?
[05-Jun-2009 13:18:34] <HKhan> nope
[05-Jun-2009 13:19:12] <HKhan> i think it'll be best to change the box its running on, this is a bit old anyway
[05-Jun-2009 13:19:16] <venturaville> do top
[05-Jun-2009 13:19:20] <venturaville> and then hit 'm'
[05-Jun-2009 13:19:28] <venturaville> err capital M
[05-Jun-2009 13:20:17] <HKhan> its only got zenoss running on the box, its a clean install of RHEL5
[05-Jun-2009 13:20:53] <HKhan> one sec, >.< box is gone down i think
[05-Jun-2009 13:21:24] <venturaville> we run RH5 ... if you do chkconfig --list I can probably tell you what to shut off ....
[05-Jun-2009 13:25:50] <HKhan> i've already disabled most of the services with chkconfig (well everything i don't need heh)
[05-Jun-2009 13:26:44] <HKhan> http://i.like.to.dre.am/services.txt
[05-Jun-2009 13:31:59] * HKhan tries upgrading ram
[05-Jun-2009 13:40:41] <venturaville> HKhan looks pretty clean
[05-Jun-2009 13:43:08] <HKhan> I think i'll change the box, P3 is quite old anyway
[05-Jun-2009 13:51:19] <cgibbons> you still have that 12" powerbook matt?
[05-Jun-2009 13:52:23] <rmatte> I'm running one box on 3GHz processing power with 4GB of memory and it's comfortably monitoring 344 devices
[05-Jun-2009 13:53:28] <rmatte> 53 of which are windows servers
[05-Jun-2009 13:57:43] <chudler> anyone know, offhand, if RenderServer is ever set do /dev/null, especially when doing getRRDValues??
[05-Jun-2009 14:01:48] <chudler> ZenModel.PerformanceConf is some really weird stuff..
[05-Jun-2009 14:03:57] <chudler> oh, here it is :-) ZenRRD.RenderServer summary(): gopts.insert(0, '/dev/null')
[05-Jun-2009 14:04:29] <rmatte> if a graph is showing 0, 200m, 400m, 600m, etc... instead of 1.0, 2.0, 3.0, 4.0, etc..., what can I do to fix it?
[05-Jun-2009 14:04:43] <rmatte> I have a bunch of graphs with identical settings, but some are displaying with the m values
[05-Jun-2009 14:04:51] <rmatte> and there's nothing that I can see that would cause that
[05-Jun-2009 14:45:18] <rocket> Hello everyone
[05-Jun-2009 14:50:21] <chudler> rrdtool python binding cannot handle unicode, it seems. Good thing I won't need it :-)
[05-Jun-2009 14:54:16] <xulfer> Hm, is there anywhere I can find documentation on how to use the core zenpacks, and maybe some of the others? I can't seem to find any solid documentation on it.
[05-Jun-2009 15:08:57] <jb> hi
[05-Jun-2009 15:09:21] <rocket> Hey jb
[05-Jun-2009 15:09:25] <jb> whats up rocket?
[05-Jun-2009 15:09:57] <rocket> not too much .. sent a fixed aix zenpack to matt today... fixed the erroneous requirement to have the mysql zenpack installed
[05-Jun-2009 15:10:10] <jb> ah cool
[05-Jun-2009 15:10:29] <rocket> trying to figure out a way to get link speed out of /proc or /sys on a linux box
[05-Jun-2009 15:10:41] <rocket> not happy with the dmesg method used in the linux zenpack ...
[05-Jun-2009 15:52:50] <jb> theres always ethtool.. although it may or may not be standard on your boxes
[05-Jun-2009 15:56:37] <chudler> probably some part of /sys/class/net/*
[05-Jun-2009 15:59:20] <jb> hm, i dont see it
[05-Jun-2009 16:01:01] <chudler> yeah, it seems to have everything but that.
[05-Jun-2009 16:11:54] <chudler> this made me so curious, I downloaded ethtool source. It seems it either interogates the eeprom of the (supported) cards, or sends them control signals.
[05-Jun-2009 16:43:43] <IoNu|z> anyone know why im getting this error on startup? /dev/hda: open failed: No medium found
[05-Jun-2009 17:07:32] <pinball> no cdrom in the drive
[06-Jun-2009 00:00:42] [disconnected at Sat Jun 6 00:00:42 2009]
[06-Jun-2009 00:00:42] [connected at Sat Jun 6 00:00:42 2009]
[06-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[06-Jun-2009 05:56:35] tehhobbit is now known as fail
[06-Jun-2009 05:57:05] fail is now known as Guest57031
[06-Jun-2009 05:57:18] Guest57031 is now known as tehhobbit
[06-Jun-2009 07:27:19] <raddy> Hello Everybody
[06-Jun-2009 07:28:01] <raddy> I am experiencing problems in Zenoss
[06-Jun-2009 07:28:08] <raddy> Can anybody help me?
[06-Jun-2009 08:45:23] <raddy> Hello Everybody
[06-Jun-2009 08:45:47] <raddy> Zenjobs and Zenactions using 100% CPU
[06-Jun-2009 08:45:55] <raddy> Can anybody help me.
[06-Jun-2009 08:46:14] <raddy> I am running Zenoss 2.4.1
[06-Jun-2009 09:01:57] <raddy> Anybody out there????
[06-Jun-2009 09:26:56] <raddy> Hello Everbody
[06-Jun-2009 09:27:17] <raddy> I am experiencing some major issues in ZenOSS
[06-Jun-2009 09:27:23] <raddy> Can anybody help me?
[06-Jun-2009 14:11:30] <rsFF> does anyone can point me a tutorial
[06-Jun-2009 14:11:38] <rsFF> wheren can i find som info regarding
[06-Jun-2009 14:11:53] <rsFF> using closeProcess in zenoss
[06-Jun-2009 14:11:55] <rsFF> thanks
[06-Jun-2009 20:45:54] <sergeymasushko> is anybody alive?
[07-Jun-2009 00:00:42] [disconnected at Sun Jun 7 00:00:42 2009]
[07-Jun-2009 00:00:42] [connected at Sun Jun 7 00:00:42 2009]
[07-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[07-Jun-2009 01:27:08] <raddy> Hello Everybody
[07-Jun-2009 01:27:18] <raddy> Anybody out there??
[07-Jun-2009 08:13:06] <raddy> Hello
[07-Jun-2009 08:13:14] <raddy> Anybody out there?
[07-Jun-2009 08:13:25] <raddy> I need you people's help.
[07-Jun-2009 08:28:33] <raddy> Anybody out theree?
[07-Jun-2009 08:36:55] <Dieterbe> yeah
[07-Jun-2009 08:37:45] <raddy> Dieterbe : I am using ZenOSS 2.4.1 in 64Bit CentOS 5.2x
[07-Jun-2009 08:38:36] <raddy> Dieterbe : ZenJobs and ZenActions occupies 100% CPU when i start it.
[07-Jun-2009 08:38:48] <raddy> Dieterbe : Can you help me?
[07-Jun-2009 08:40:30] <Dieterbe> what cpu do you have?
[07-Jun-2009 08:40:55] <raddy> 64Bit Intel Quad Core
[07-Jun-2009 08:43:26] <Dieterbe> hmm and you have many jobs/hosts?
[07-Jun-2009 08:43:43] <Dieterbe> i noticed that zenoss takes a lot of cpu when starting, but after a few minutes, it becomes normal
[07-Jun-2009 08:43:52] <Dieterbe> eg it decreases to reasonable levels
[07-Jun-2009 08:44:58] <raddy> Dieterbe : Not like that i my end, i don't even has that much jobs, more over even after sometime Zenaction and Zenjobs not decreasing the CPU usage.
[07-Jun-2009 08:47:55] <Dieterbe> raddy: can you start 'top' on command line, press '1' and then copy paste the lines at the top that say Cpu0, Cpu1 etc
[07-Jun-2009 08:50:54] <raddy> Dieterbe : What for?
[07-Jun-2009 08:51:29] <Dieterbe> to be able to help you
[07-Jun-2009 08:52:07] <raddy> Dieterbe : That is right. but i am able to tell whatever you ask, what you are seeking?
[07-Jun-2009 08:56:48] <Dieterbe> raddy: any unusual amounts of %wa, %sy/us etc
[07-Jun-2009 08:57:37] <raddy> Dieterbe : Ohh, that i didn't know,
[07-Jun-2009 08:57:51] <raddy> Dieterbe : I'll tell you now
[07-Jun-2009 19:42:15] <rbd> hey guys... I have a server with apache, mysql, etc already installed on it in use for other things. is it safe to install the zenoss core stack (e.g. using the stack installer) and not have it mess with my existing configurations, or should I install on a fresh box?
[07-Jun-2009 19:42:55] <rbd> I don't want to hose up what I already have...can see in the docs where they talk about how the installer is composed...where/how it installs on the box, etc
[07-Jun-2009 19:44:15] <rbd> s/can/can't
[07-Jun-2009 19:44:22] <sergeymasushko> I have installed zenoss on virtual FreeBSD box
[07-Jun-2009 19:53:27] scfh_ is now known as scfh
[08-Jun-2009 00:00:42] [disconnected at Mon Jun 8 00:00:42 2009]
[08-Jun-2009 00:00:42] [connected at Mon Jun 8 00:00:42 2009]
[08-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[08-Jun-2009 01:26:01] <raddy> Hello Everybody
[08-Jun-2009 01:26:08] <raddy> Anyone out there?
[08-Jun-2009 01:52:16] <raddy> Anybody out there???
[08-Jun-2009 02:08:18] <Appiah> just ask your question
[08-Jun-2009 02:08:31] <Appiah> ...
[08-Jun-2009 02:11:18] <raddy> Appiah : You always encourage people to ask their question when nobody is listening them.
[08-Jun-2009 02:11:52] <raddy> Appiah : you also won't respond to people's questions, but would poke them around.
[08-Jun-2009 02:12:07] <Appiah> always?
[08-Jun-2009 02:12:23] <raddy> Appiah : Seems so.
[08-Jun-2009 02:12:33] <Appiah> that's weird
[08-Jun-2009 02:12:53] <raddy> ok, i am asking the question.
[08-Jun-2009 02:13:20] <raddy> ZenJobs and ZenActions using 100% CPU when started.
[08-Jun-2009 02:13:55] <raddy> I am using ZenOSS 2.4.1 on 64Bit Quad Core CentOS 5.2 Server.
[08-Jun-2009 02:13:58] <Appiah> http://www.sabi.co.uk/Notes/linuxHelpAsk.html#anyone this is a "how to ask questions for linux on IRC" but it pretty much covers all tech related areas and asking forhelp on IRC
[08-Jun-2009 02:14:46] <raddy> Appiah : I am not a IRC Noob.
[08-Jun-2009 02:14:55] <Appiah> oh ok
[08-Jun-2009 02:15:04] <Appiah>
[08-Jun-2009 02:15:56] <raddy> I hope you might know get to know my proficiency in IRC from describing the question.
[08-Jun-2009 02:16:21] <Appiah> if I remeber correctly you asked for this before right?
[08-Jun-2009 02:16:28] <Appiah> 100% CPU load?
[08-Jun-2009 02:16:32] <raddy> Appiah : exactly
[08-Jun-2009 02:16:46] <raddy> Appiah : It has not been resolved yet.
[08-Jun-2009 02:16:52] <Appiah> and no ones ide sofar helped
[08-Jun-2009 02:17:02] <Appiah> you tried making a post on the forum?
[08-Jun-2009 02:17:10] <raddy> Appiah : Have to.
[08-Jun-2009 02:17:26] <Appiah> Is it a freshly installed 2.4.1 or did you upgrade to it?
[08-Jun-2009 02:17:47] <raddy> Appiah : Yesterday one person helped me, but i was unable continue the conversation at that time.
[08-Jun-2009 02:17:56] <raddy> Appiah : Nope, Upgrade.
[08-Jun-2009 02:18:08] <Appiah> Do you have the logs since the upgrade?
[08-Jun-2009 02:18:46] <raddy> Appiah : Yeah, but ZenJobs logs or empty, ZenActions log contains nothing interesting.
[08-Jun-2009 02:19:02] <raddy> *Are
[08-Jun-2009 02:19:48] <Appiah> I mean logs from the update itself
[08-Jun-2009 02:20:13] <Appiah> output from when you ran the 2.4.1 stack installer
[08-Jun-2009 02:20:13] <raddy> Appiah : I don't think so. where it exists?
[08-Jun-2009 02:20:28] <Appiah> good question ... I just dump it myself
[08-Jun-2009 02:20:41] <raddy> Appiah : Oh.
[08-Jun-2009 02:20:42] <Appiah> especially when error/warnings
[08-Jun-2009 02:20:46] <Appiah> shows up
[08-Jun-2009 02:20:57] <Appiah> you didnt get a message about doing a zenmigrate with force?
[08-Jun-2009 02:21:33] <raddy> Appiah : I got, but i didn't get any errors and so i didn't dump the message.
[08-Jun-2009 02:21:51] <raddy> Appiah : ok, see you in forums (If you have account)
[08-Jun-2009 05:38:19] scfh is now known as scfh_
[08-Jun-2009 05:38:23] scfh_ is now known as scfh
[08-Jun-2009 06:25:43] <Appiah> Zenjobs wont start after a upgrade (2.4.1), output from trying to debug in terminal -> http://pastebin.com/m1c8b1e7b
[08-Jun-2009 06:41:20] <Appiah> I ran a zenmigrate .. AGAIN and now it starts
[08-Jun-2009 13:28:01] <rocket> Hey cgibbons
[08-Jun-2009 13:28:06] <cgibbons> howdy mr rocket
[08-Jun-2009 13:28:21] <rocket> How are things going?
[08-Jun-2009 13:28:40] <cgibbons> quiet so far (my 2 week stint on L3 support), which is good
[08-Jun-2009 13:28:42] <cgibbons> enjoy the test?
[08-Jun-2009 13:28:56] <rocket> it wasnt too bad .. enjoy the answers?
[08-Jun-2009 13:29:46] <cgibbons> still looking, but it was good from my perspective
[08-Jun-2009 13:30:10] <rocket> cgibbons: linux is pissing me off at the moment
.. cant seem to find a great way of getting nic speed without being root ..[08-Jun-2009 13:30:49] <cgibbons> deja vu on that
[08-Jun-2009 13:31:08] * rocket is working on improving the linux zenpack .. eg uname -i support that maps to Intel or AMD for hardware platform
[08-Jun-2009 13:31:30] <rocket> already fixed the nic perf collection that was missing
[08-Jun-2009 13:32:32] <rocket> but to get the thresholds to work right I am trying to figure out a way besides dmesg to get nic speed
[08-Jun-2009 13:33:45] <rocket> cgibbons: hey you might know this .. is there a way to easily do a ssh query in a modeller before the main code goes?
[08-Jun-2009 13:34:26] <rocket> eg I am looking for a specific file .. if it exists run the modeller ...
[08-Jun-2009 13:35:01] <cgibbons> yeah that's a common problem in the modelers, and I don't think we have a solution for it yet (I'd like to do the same thing in the Windows ones). Brain E was looking @ that once, let me see if he had a solution.
[08-Jun-2009 13:35:41] <rocket> k sounds good .. I have a work around but it would cleanup my stuff in the future .. etc ..
[08-Jun-2009 13:46:20] <cgibbons> ok yeah, that ability doesn't exist yet
[08-Jun-2009 13:46:46] <rocket> cgibbons: thats where I was leaning to .. thanks for the followup
[08-Jun-2009 16:52:15] <_indigo> good evening everybody, can anyone help me to clarify my issue, I´m a little bit confused, is by zenoss a nagios or something like that under the hood?
[08-Jun-2009 16:53:17] <kgoedtel> zenoss is under the hood
[08-Jun-2009 16:54:30] <_indigo> ok thx, we evaluate some network management software and my work mate has installed zenoss saw during the installation some "nagios stuff". I thought I ask some guys who know what they do
[08-Jun-2009 16:55:31] <kgoedtel> you can utilize nagios plugins- but that is more or less the extent of which zenoss "uses" nagios
[08-Jun-2009 16:56:09] <kgoedtel> all the real work is done in zenoss (python)
[08-Jun-2009 16:56:41] <_indigo> ah ok I understand, for sure there are a lot of nagios plugins out there, so it make sense to support them. thx for help
[08-Jun-2009 16:56:49] <kgoedtel> np
[08-Jun-2009 17:31:50] <rmatte> If anyone is interested, I posted a detailed guide on getting LDAP working for Zenoss core:
[08-Jun-2009 17:31:51] <rmatte> http://forums.zenoss.com/viewtopic.php?p=35737#35737
[08-Jun-2009 17:32:09] <rmatte> It goes more in depth than what's in the guide that explains setting up the plugins and all that
[08-Jun-2009 19:46:29] <daMaestro> anyone know what data to enumerate when building a graph for disk space (not percent)
[08-Jun-2009 19:46:46] <daMaestro> like, raw usage (GB/TB/PB/etc)
[08-Jun-2009 19:47:46] <daMaestro> I currently use usedBlocks and then: ${here/totalBlocks},/,100,*
[08-Jun-2009 19:48:10] <daMaestro> however, that does not work when trying a multgraph for multiple devices of storage usage
[09-Jun-2009 00:00:42] [disconnected at Tue Jun 9 00:00:42 2009]
[09-Jun-2009 00:00:42] [connected at Tue Jun 9 00:00:42 2009]
[09-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[09-Jun-2009 04:07:57] <capricorn-80> hi ! i need to draw performance graph for my traffic
[09-Jun-2009 04:08:26] <capricorn-80> I want to monitor my router in/out traffic
[09-Jun-2009 04:09:15] <capricorn-80> i have created a new template with name bandwidth
[09-Jun-2009 04:09:42] <capricorn-80> zenoss has prebuil OID ?
[09-Jun-2009 04:09:49] <capricorn-80> prebuilt **
[09-Jun-2009 04:13:52] <capricorn-80> anyone there ?\
[09-Jun-2009 04:17:20] <capricorn-80> anyone home ?
[09-Jun-2009 04:54:03] <capricorn-80> anyone tell me how can i bind templates ?
[09-Jun-2009 04:55:50] <capricorn-80> ok i have bind tempates to devices
[09-Jun-2009 05:25:06] <capricorn-80> hi rocket
[09-Jun-2009 05:56:03] <capricorn-80> hello hello
[09-Jun-2009 06:40:46] <capricorn-80> hi chudler1
[09-Jun-2009 07:34:24] <capricorn-80> hello again
[09-Jun-2009 07:35:08] <capricorn-80> I am stuck in binding
[09-Jun-2009 07:35:15] <capricorn-80> need some help
[09-Jun-2009 07:39:07] <Appiah_> Does anyone know why taking a backup of the zenoss uses so much disk space and timesout?
[09-Jun-2009 07:40:29] <jb> yeah, you can change the temporary location via --temp-dir if /tmp doesn't have enough room.
[09-Jun-2009 07:41:05] <Appiah_> there's like 60gb free
[09-Jun-2009 07:41:11] <Appiah_> it used 30gb sofar
[09-Jun-2009 07:41:41] <Appiah_> previous backups have been about 1gb file when it's done , but how can it use about 30 > to make 1gb?
[09-Jun-2009 07:42:14] <jb> compression more than likely..
[09-Jun-2009 07:42:16] <jb> lots of text?
[09-Jun-2009 07:42:37] <jb> how large is your event db?
[09-Jun-2009 07:43:40] <Appiah_> good question
[09-Jun-2009 07:43:52] <Appiah_> i think i'll try from cli then the webinterface
[09-Jun-2009 07:51:11] <capricorn-80> jb: I am trying to bind ethernetCsmacd /Devices/Network/Router/Cisco Standard ethernet interface template with 75% utilization threshold
[09-Jun-2009 07:51:16] <capricorn-80> but its not working
[09-Jun-2009 07:52:18] <capricorn-80> may be i m doing something wrong in performing the steps
[09-Jun-2009 07:52:34] <jb> the ethernetCsmacd template should not be binded to a class
[09-Jun-2009 07:53:15] <jb> what is your reason for doing so?
[09-Jun-2009 07:54:09] <capricorn-80> jb: I want in/out traffic of my router .. like mrtg
[09-Jun-2009 07:54:34] <jb> the Router/Cisco class inherits that template from /Device
[09-Jun-2009 07:54:40] <jb> there is no need to do any additional binding
[09-Jun-2009 07:56:11] <capricorn-80> jb: if i see on /Devices /Network /Router /Cisco i can see Available Performance Templates and in that i can see ethernetCsmacd
[09-Jun-2009 07:56:29] <cgibbons> yeah but don't bind it
[09-Jun-2009 07:56:35] <jb> right
[09-Jun-2009 07:56:40] <capricorn-80> now how can i put this capability into my settings so that i can monitor my in/out traffic ?
[09-Jun-2009 07:56:47] <jb> its already in place..
[09-Jun-2009 07:56:51] <jb> go to the "OS" tab of the router
[09-Jun-2009 07:57:01] <jb> you should see the network interfaces there.. click on them
[09-Jun-2009 07:57:13] <jb> but, ensure that it is not binded
[09-Jun-2009 08:04:54] <capricorn-80> not binded means ?
[09-Jun-2009 08:05:05] <capricorn-80> i cant see x.x.x.254 in the list
[09-Jun-2009 08:06:44] <capricorn-80> this is my gateway ip
[09-Jun-2009 08:07:14] <capricorn-80> and i want to check its performance . and i have enable snmp on this and my zenoss is polling to this router
[09-Jun-2009 08:12:50] <rocket> morning cgibbons
[09-Jun-2009 08:13:22] <jb> sup rocket
[09-Jun-2009 08:13:55] <rocket> not much and you jb?
[09-Jun-2009 08:14:25] <rocket> jb: got a couple ssh zenpacks under my belt now ..
[09-Jun-2009 08:14:30] <jb> nice
[09-Jun-2009 08:14:32] <jb> which ones?
[09-Jun-2009 08:14:37] <jb> gentoo?
[09-Jun-2009 08:14:43] <rocket> I am just submitting them to matt now
[09-Jun-2009 08:15:02] <rocket> an extension to linux monitor that turns on network perf monitoring
[09-Jun-2009 08:15:23] <rocket> and a gentoo one that adds software installed support
[09-Jun-2009 08:16:43] <rocket> these zenpacks also include the tests etc so there is pretty decent examples with both of them
[09-Jun-2009 08:19:48] <rocket> jb: predictive threshold and aix is currently being tested by matt
[09-Jun-2009 08:23:17] <capricorn-80> jb: what should i do ?
[09-Jun-2009 08:24:41] <rocket> capricorn-80: are you seeing anything under the events tab? or the event console .. please change the dropdown to include debug events
[09-Jun-2009 08:25:28] <rocket> capricorn-80: if you arent seeing the data for just the interfaces you might need to look at the logs on the system itself to see if there is anything in there
[09-Jun-2009 08:27:53] <capricorn-80> i cant see any process for that router
[09-Jun-2009 08:27:55] <capricorn-80> its fine
[09-Jun-2009 08:28:38] <rocket> you mean you cant see anything under the events tabs?
[09-Jun-2009 08:29:47] <rocket> capricorn did you (model/add) your router after enabling all the snmp stuff .. eg making zenoss aware of it
[09-Jun-2009 08:30:07] <capricorn-80> yes
[09-Jun-2009 08:30:25] <capricorn-80> rocket: i can get all info from the router on zenoss
[09-Jun-2009 08:30:32] <rocket> capricorn-80: can you post a screenshot of what your seeing?
[09-Jun-2009 08:30:37] <capricorn-80> its that just the interface x.x.x.254 is not showing
[09-Jun-2009 08:30:52] <capricorn-80> and this is the interface which has been enabled with snmp
[09-Jun-2009 08:31:34] <rocket> capricorn-80: I will need to see a screenshot to fully understand exactly what you are seeing
[09-Jun-2009 08:45:30] <rocket> morning matt
[09-Jun-2009 08:47:09] <mrayzenoss> good morning
[09-Jun-2009 08:47:28] <rocket> matt I sent you a couple of zenpacks this morning
[09-Jun-2009 08:48:06] <mrayzenoss> yeah, looks like I've got a busy day
[09-Jun-2009 08:48:27] <rocket> these last 2 are pretty small ones ...
[09-Jun-2009 08:48:55] <rocket> did the predictive one work any better for you last night?
[09-Jun-2009 08:49:21] <mrayzenoss> haven't checked on it yet
[09-Jun-2009 08:49:26] <rocket> ok
[09-Jun-2009 08:54:22] <capricorn-80> rocket: well its very simple
[09-Jun-2009 09:08:18] rocket_ is now known as rocket
[09-Jun-2009 09:13:55] <rocket> ls
[09-Jun-2009 09:14:00] <rocket> oops sorry
[09-Jun-2009 09:14:01] <rmatte> .
[09-Jun-2009 09:14:02] <rmatte> ..
[09-Jun-2009 09:14:15] <rocket> rmatte: haa haa
[09-Jun-2009 09:14:23] <rmatte> /bin
[09-Jun-2009 09:14:26] <rmatte>
[09-Jun-2009 09:14:59] <rocket> rmatte: I did get a patch submitted for the linux zenpack not enabling those collectors by default
[09-Jun-2009 09:15:07] <rmatte> excellent
[09-Jun-2009 09:15:09] <rocket> rmatte: we will see what happens with that
[09-Jun-2009 09:15:18] <rmatte> yeh, that's awesome though, nice work
[09-Jun-2009 09:15:39] <rocket> rmatte: and I did submit a new extended linux zenpack that fixes the perf monitoring for network interfaces
[09-Jun-2009 09:15:52] <rmatte> even better hehe
[09-Jun-2009 09:15:56] <rocket> rmatte: just today ..
[09-Jun-2009 09:16:02] <rmatte> though wasn't there another thing missing?
[09-Jun-2009 09:16:14] <rmatte> there was interfaces and filesystem info missing no?
[09-Jun-2009 09:16:22] <rocket> rmatte: filesystems too?
[09-Jun-2009 09:16:34] <rmatte> let me check...
[09-Jun-2009 09:16:41] <rocket> rmatte: I guess I hadnt noticed that one .. if thats the case I can add filesystems to mine ...
[09-Jun-2009 09:17:00] <rmatte> oh no, it was memory collection missing
[09-Jun-2009 09:17:12] <rmatte> which the zenoss.cmd.linux.memory plugin does
[09-Jun-2009 09:17:28] <rocket> matt was that part of the bug report?
[09-Jun-2009 09:17:35] <rmatte> yessir
[09-Jun-2009 09:17:40] <rmatte> http://dev.zenoss.com/trac/ticket/5047
[09-Jun-2009 09:17:49] <rocket> ok then when that is fixed then it should be all good
[09-Jun-2009 09:17:52] <capricorn-80> i have enable snmp on x.x.x.254 router and i added x.x.x.254 in device
[09-Jun-2009 09:18:03] <rmatte> The "zenoss.cmd.linux.memory" and "zenoss.cmd.linux.ifconfig" zCollectorPlugins (which are included in Core) do not get automatically bound to /Linux/SSH when the Linux Monitor ZenPack is installed.
[09-Jun-2009 09:18:25] <rmatte> Only CPU and Filesystem info is pulled via SSH by default once the Linux Monitor ZenPack is installed. Collection of Interface info as well as Memory info is supported in Core by default, but not automatically activated for the /Linux/SSH device grouping.
[09-Jun-2009 09:19:16] <rocket> rmatte: hrmm so perf stuff just needed to be enabled?
[09-Jun-2009 09:19:18] <rmatte> looks like your patch does take that in to account though no?
[09-Jun-2009 09:19:26] <rmatte> pretty much
[09-Jun-2009 09:19:50] <rocket> rmatte: the templates for network perf were still snmp based it looked like to me
[09-Jun-2009 09:19:58] <rmatte> 2 plugins need to be bound to the /Linux/SSH event class when the Linux Monitor Zenpack is installed, though extending the function of the Zenpack itself might be more effective
[09-Jun-2009 09:20:05] <rmatte> you would basically just be duplicating work though
[09-Jun-2009 09:20:31] <rocket> rmatte: my zenpack doesnt include the patch to the main linux zenpack
[09-Jun-2009 09:20:37] <rmatte> your patch seems to be what's needed
[09-Jun-2009 09:20:51] <rmatte> so it's all good
[09-Jun-2009 09:21:24] <rocket> rmatte: good .. and my extended linux monitoring zenpack creates the new perf template for nics that is ssh based instead of snmp based
[09-Jun-2009 09:21:33] <rmatte> ah
[09-Jun-2009 09:21:37] <rmatte> gotcha, nice
[09-Jun-2009 09:22:11] <rocket> now you have me curious about filesystems .. if the perf ones for those are correct as well
[09-Jun-2009 09:22:34] <rmatte> hehe
[09-Jun-2009 09:23:02] <rocket> rmatte: no big deal if they arent ..
[09-Jun-2009 09:24:23] <chudler> anyone know offhand which version of twisted Zenoss uses these days?
[09-Jun-2009 09:24:41] <jb> Twisted 8.1.0
[09-Jun-2009 09:25:52] <chudler> thanks, cool. any plan to do 8.2.0? ;-) I know it needs testing, probably. Has a SSL/TLS fix for xmpp.
[09-Jun-2009 09:26:18] <jb> im not the right person to answer that one
[09-Jun-2009 09:30:16] <rocket> jb: awww why not?
[09-Jun-2009 09:32:42] <rmatte> that'd be a question for the next time the developers drop by IRC
[09-Jun-2009 09:33:00] <rocket> rmatte: just pulling your leg
[09-Jun-2009 09:33:03] * chudler adds it to the list
[09-Jun-2009 09:34:50] <capricorn-80> rocket: i have enable snmp on x.x.x.254 router and i added x.x.x.254 in device .
[09-Jun-2009 09:35:51] <rocket> capricorn-80: like I said I dont understand exactly what you are seeing ... a screenshot would help ... without that I am not sure how much help I would be at the moment
[09-Jun-2009 09:36:00] <jb> capricorn-80: you really aren't providing many details.. we already requested a screnshot..
[09-Jun-2009 09:36:36] <capricorn-80> jb: you are right but the thing is i have many ips show on that as i have configured vlan
[09-Jun-2009 09:36:51] <capricorn-80> so screenshot is not good solution thats why i repeat that again
[09-Jun-2009 09:37:19] <rocket> jb: filesystem perf is already covered in the core linux zenpack .... whew
[09-Jun-2009 09:37:50] <rocket> so do you have any perf statistics on any of the network interfaces?
[09-Jun-2009 09:38:03] <rocket> capricorn-80: or is it a problem with one specific port?
[09-Jun-2009 09:38:26] <rocket> capricorn-80: the problem is that your not exactly explaining your problem well
[09-Jun-2009 09:47:04] <capricorn-80> rocket: let me try . take like this .. my router address is 172.16.1.254 and i have enabled snmp on it so that i can monitor cpu , memory and bandwidth on it. I added it in device list and i am getting many details about it like cpu , memory performance graphs. on OS - interfaces i can see VLAN interface ip addresses
[09-Jun-2009 09:47:40] <capricorn-80> but i cant see the interface with ip address 172.16.1.254
[09-Jun-2009 09:48:00] <rocket> you see every other interface but that one?
[09-Jun-2009 09:48:16] <rocket> is that interface different than the otherones in anyway?
[09-Jun-2009 09:48:42] <rocket> capricorn-80: can you paste bin the output of this for me from the command line
[09-Jun-2009 09:48:52] <capricorn-80> no
[09-Jun-2009 09:48:57] <rocket> zenmodeler run -v 10 -d <device_name>
[09-Jun-2009 09:49:24] <jb> not in this channel though
[09-Jun-2009 09:49:24] <jb>
[09-Jun-2009 09:49:42] <rocket> jb: I did say pastebin
[09-Jun-2009 09:49:48] <jb> ah yeah
[09-Jun-2009 09:50:05] <rocket> oops the space might have made that confusing
[09-Jun-2009 09:52:35] <capricorn-80> rocket: http://pastebin.com/m13349a13
[09-Jun-2009 09:52:45] <capricorn-80> hmm
[09-Jun-2009 09:52:47] <capricorn-80> strange
[09-Jun-2009 09:53:09] <capricorn-80> the command is in /opt/zenoss/bin when i type it but when i run gave error
[09-Jun-2009 09:53:15] <capricorn-80> paths problem
[09-Jun-2009 09:53:29] <rocket> capricorn-80: run it as the zenoss user
[09-Jun-2009 09:56:34] <capricorn-80> rocket: its long long output
[09-Jun-2009 09:56:50] <rocket> capricorn-80: good .. it might show whats going on then
[09-Jun-2009 09:58:46] <capricorn-80> rocket: http://pastebin.com/m2777ed65
[09-Jun-2009 10:00:35] <mrayzenoss> chudler: http://dev.zenoss.org/trac/ticket/4991 Twisted 8.2 will probably not be in this release, unless a developer finds something really broken in 8.1
[09-Jun-2009 10:01:03] <chudler> thanks mrayzenoss. I will work around this in my code.
[09-Jun-2009 10:01:18] <rocket> capricorn-80: how many nics are in the box?
[09-Jun-2009 10:02:06] <capricorn-80> the zenoss box or the router ?
[09-Jun-2009 10:02:25] <rocket> capricorn-80: linux router? and yes how many nics in the router?
[09-Jun-2009 10:05:57] <capricorn-80> well router has vlan
[09-Jun-2009 10:10:01] <rocket> capricorn-80: I am seeing 10.x.x.x based interfaces .. not 172.x based ones ...
[09-Jun-2009 10:10:17] <capricorn-80> rocket: yea r u right
[09-Jun-2009 10:10:21] <capricorn-80> i gave u example
[09-Jun-2009 10:10:39] <rocket> capricorn-80: pastebin a netstat -rn and an ifconfig -a if possible please
[09-Jun-2009 10:13:55] <rocket> capricorn-80: I am assuming eth1 is your x.254 interface correct?
[09-Jun-2009 10:14:34] <capricorn-80> rocket: http://pastebin.com/m52f2d775
[09-Jun-2009 10:16:56] <rocket> I believe the issue is that either ucd-snmp doesnt support vlans or it is not currently configured to show them
[09-Jun-2009 10:17:20] <rocket> or is it called netsnmpd .. cant remember
[09-Jun-2009 10:18:57] <rocket> capricorn-80: how did you create the vlan?
[09-Jun-2009 10:19:40] <rocket> capricorn-80: can you pastebin the output of /proc/net/vlan/config ?
[09-Jun-2009 10:19:46] <capricorn-80> router commands
[09-Jun-2009 10:19:51] <capricorn-80> its cisco router
[09-Jun-2009 10:21:06] <rocket> capricorn-80: from the output you gave me it looks like your looking for some extra information about 169.254.0.0 correct?
[09-Jun-2009 10:21:46] <rocket> capricorn-80: the information you sent me doesnt look like it belongs to a cisco router ..
[09-Jun-2009 10:21:56] <rocket> capricorn-80: it looks very linuxy to me ..
[09-Jun-2009 10:22:51] <rocket> i guess they could be using linux now for some devices ..
[09-Jun-2009 10:23:40] <rocket> is the only thing missing information about 169.254.0.0?
[09-Jun-2009 10:24:20] <rocket> capricorn-80: have you done a snmpwalk against the router to see if snmp is showing any information about 169.254.0.0 or whatever you are thinking you are missing
[09-Jun-2009 10:24:33] <rocket> otherwise from the output I am seeing things look correct ...
[09-Jun-2009 10:26:10] <capricorn-80> rocket: right that was for linux box . i thought u want my zenoss server information using that command
[09-Jun-2009 10:26:45] <rocket> no I want the information about the router
[09-Jun-2009 10:26:56] <rocket> zenmodeler run -v 10 -d <router>
[09-Jun-2009 10:27:19] <rocket> and then any information about the interface config / vlans on the router in a pastebin
[09-Jun-2009 10:28:23] <capricorn-80> rocket: http://pastebin.com/m4ce5fcf
[09-Jun-2009 10:28:39] <capricorn-80> its long long output .. just got this one
[09-Jun-2009 10:28:43] <capricorn-80> its short one
[09-Jun-2009 10:29:32] <rocket> capricorn-80: the last pastebin is what exactly?
[09-Jun-2009 10:29:47] <rocket> capricorn-80: it looks like it isnt gathering any info from the device based on that output
[09-Jun-2009 10:30:01] <rocket> did you run with -v10?
[09-Jun-2009 10:30:05] <capricorn-80> rocket: yes
[09-Jun-2009 10:30:22] <rocket> capricorn-80: I would like the full output please
[09-Jun-2009 10:31:03] <capricorn-80> rocket: how ?
[09-Jun-2009 10:31:14] <capricorn-80> i cant insert into txt file
[09-Jun-2009 10:31:16] <capricorn-80> anyway ?
[09-Jun-2009 10:32:37] <rocket> capricorn-80: I need to step away from this for a bit ... I am getting frustrated because I am feeling that we are not communicating very well .. I am asking for very specific things and feel that I am not getting those things
[09-Jun-2009 10:32:55] <capricorn-80> i am
[09-Jun-2009 10:33:11] <capricorn-80> rocket: i am needy person
[09-Jun-2009 10:33:31] <capricorn-80> the problem is that i cannot move that output to txt file due to permission problems
[09-Jun-2009 10:34:13] <capricorn-80> rocket: http://pastebin.com/m44dacd8c
[09-Jun-2009 10:34:20] <capricorn-80> this is the whole screen now
[09-Jun-2009 10:37:32] <rocket> capricorn-80: so are you not seeing any of the ip addresses for the vlans?
[09-Jun-2009 10:38:23] <rocket> capricorn-80: a screenshot of the zenoss router information page would be useful now
[09-Jun-2009 10:38:48] <capricorn-80> wait
[09-Jun-2009 10:39:07] <rocket> capricorn-80: without it at this point I dont think i will be able to help you .. .. however i dont think zenoss supports vlans in the manner you are looking for at this point
[09-Jun-2009 10:39:40] <rocket> it will just show information about the underlying interface I believe ...
[09-Jun-2009 10:47:25] <capricorn-80> rocket:well i can see vlan traffic
[09-Jun-2009 10:48:10] <capricorn-80> i mean in/out traffic from vlan but cant see the main interface 10.254.95.254 through which i m polling ..
[09-Jun-2009 10:48:13] <capricorn-80> just wait
[09-Jun-2009 10:48:16] <capricorn-80> i talk to you
[09-Jun-2009 11:06:58] <xulfer> Is there anyway to convert from a string to an integer within a datapoint? Like when dealing with the output of extend
[09-Jun-2009 11:52:59] <lukecyca> When using COMMAND data source, is 30 second timeout hardcoded? Is there any way to extend it?
[09-Jun-2009 12:05:39] <lukecyca> I set it in zProperties for the device, but it doesn't seem to be honored
[09-Jun-2009 12:37:59] <lukecyca> Aha. zCommandCommandTimeout is honored when the command is actually run. It's just not honored when you click the "Test" button for the command. The test button seems to be hardcoded to a 30 second timeout.
[09-Jun-2009 12:38:24] <chudler> interesting. thanks
[09-Jun-2009 12:50:05] <rmatte> Question... I just had a process go down on a server then come back up, after it came back up Zenoss didn't recognize that it was back. I had to remodel the device to get it to realize it was back. This is because the process name switched From: Microsoft.Exchange.Pop3.exe -pipe:692 -stopkey:Global\ExchangeStopKey-5da6d91e-b226-4009-919f-ecbbf3831293 -resetkey:Global\ExchangeResetKey-650f2307-38fb-
[09-Jun-2009 12:50:06] <rmatte> To: Microsoft.Exchange.Pop3.exe -pipe:692 -stopkey:Global\ExchangeStopKey-c052f2b8-9ef0-4602-b2f3-7f7393a2555f -resetkey:Global\ExchangeResetKey-7e9c1ce3-672e-
[09-Jun-2009 12:50:11] <rmatte> when the process restarted
[09-Jun-2009 12:50:15] <rmatte> what's the best way around this?
[09-Jun-2009 12:50:30] <rmatte> should I edit the process and remove the extra options then lock it?
[09-Jun-2009 12:50:49] <chudler> there is a option for the process monitor to "ignore arguments". this might do what you need
[09-Jun-2009 12:51:07] <rmatte> where would I find that?
[09-Jun-2009 12:51:37] <chudler> found it. It is called "Ignore Parameters" and is on the edit tab of the process.
[09-Jun-2009 12:52:24] <rmatte> ah thanks
[09-Jun-2009 12:53:04] <chudler> unfortunately, this option doesn't help if you have things that run inside an interpreter like python, ruby, etc. I would love to be able to solve that one..
[09-Jun-2009 13:00:55] <rmatte> right
[09-Jun-2009 13:02:33] <capricorn-80> rocket: you there ?
[09-Jun-2009 13:02:57] <rocket> for a few minutes .. i have to run an errand soon
[09-Jun-2009 13:03:22] <capricorn-80> ah
[09-Jun-2009 13:04:18] <capricorn-80> you want to see pic now ?
[09-Jun-2009 13:04:25] <rocket> yes please
[09-Jun-2009 13:05:47] <capricorn-80> imagebin ?
[09-Jun-2009 13:05:55] <rocket> sure
[09-Jun-2009 13:05:57] <rmatte> chudler: One I toggle that option do I have to remodel every server that is monitoring that particular process?
[09-Jun-2009 13:06:03] <rmatte> One == Once
[09-Jun-2009 13:06:04] <capricorn-80> i forget the website address
[09-Jun-2009 13:06:17] <capricorn-80> got it
[09-Jun-2009 13:09:38] <capricorn-80> rocket: http://imagebin.ca/view/PvjbBx.html
[09-Jun-2009 13:10:40] <capricorn-80> rocket: u can see that vlan can be recognize by the zenoss
[09-Jun-2009 13:11:09] <rocket> ok
[09-Jun-2009 13:11:16] <rocket> looking
[09-Jun-2009 13:12:29] <capricorn-80> ok
[09-Jun-2009 13:15:24] <rocket> so your issue is in the routes or in the interfaces themselves?
[09-Jun-2009 13:15:40] <capricorn-80> yes
[09-Jun-2009 13:15:52] <capricorn-80> i just added my router and i did nothing
[09-Jun-2009 13:16:00] <capricorn-80> i mean added my router as device
[09-Jun-2009 13:16:10] <capricorn-80> oh sorry got it wrong
[09-Jun-2009 13:16:30] <capricorn-80> my issue is why i cannot see 10.254.95.254
[09-Jun-2009 13:16:35] <capricorn-80> in the interfaces
[09-Jun-2009 13:16:51] <capricorn-80> coz i want to monitor 10.254.95.254 and its in/out traffic
[09-Jun-2009 13:18:50] <rocket> which vlan is that interface on?
[09-Jun-2009 13:19:47] <capricorn-80> well this interface is not on any vlan .. on this router different vlans are configured
[09-Jun-2009 13:20:21] <capricorn-80> one strage thing is this IP Addresses 127.0.0.22/8
[09-Jun-2009 13:20:40] <capricorn-80> and i can see some graphs of it that it contains traffic which is very strange
[09-Jun-2009 13:22:21] <rocket> capricorn-80: I guess I would save the pictures and ask one of the main devs on thursday ..
[09-Jun-2009 13:23:36] <capricorn-80> ok
[09-Jun-2009 13:23:55] <capricorn-80> devs ?
[09-Jun-2009 13:24:10] <capricorn-80> developers ?
[09-Jun-2009 13:33:10] <drtns> Hey all I am trying to make an event mapping and it's not working I am sure it's something dumb that I am doing but I need some help. The message that comes into Zenoss is '[error] jk2_init() Can't find child 19024 in scoreboard' and my regex is [error] jk2_init() Can't find child * in scoreboard and I am telling Zenoss to send that directly to history. but it's not working. what is the best way to troubleshoot this... is my regex messe
[09-Jun-2009 13:42:54] <rocket> drtns: you probably have issues escaping some of the characters of the regex
[09-Jun-2009 13:43:18] <rocket> eg brackets need to be escaped etc
[09-Jun-2009 13:43:46] <rocket> instead of * you would need something like [0-9]+
[09-Jun-2009 13:46:10] <rmatte> drtns: check my privmsg
[09-Jun-2009 13:48:40] <rmatte> drtns: something like: [error] j2k_init() Can't find child
[09-Jun-2009 13:48:43] <rmatte> should probably work
[09-Jun-2009 13:48:53] <rmatte> I have a regex of: (ip) vrf/dest_addr=
[09-Jun-2009 13:48:56] <rmatte> and it works fine
[09-Jun-2009 13:49:14] <rmatte> or just j2k_init() Can't find child
[09-Jun-2009 13:49:29] <rmatte> you don't really need a wildcard, just need to specify part of the line as the regex
[09-Jun-2009 13:54:40] <rmatte> if it's coming in with event class of /Unknown and no event key then you need to set the event class key for the mapping as "defaultmapping" to make it work properly
[09-Jun-2009 13:54:57] <rmatte> and no event class key rather
[09-Jun-2009 14:04:56] <sergeymasushko> how can I place FastEthernet0/0 Throughput graph on the 'Perf' tab?
[09-Jun-2009 14:31:23] <rocket> Hey Matt
[09-Jun-2009 14:55:23] <rocket> mrayzenoss_: ping
[09-Jun-2009 14:55:33] mrayzenoss_ is now known as mrayzenoss
[09-Jun-2009 14:55:35] <mrayzenoss> yo
[09-Jun-2009 14:55:42] <mrayzenoss> testing the PredictiveThreshold now
[09-Jun-2009 14:55:46] <mrayzenoss> and sitting in the defect review
[09-Jun-2009 14:56:01] <rocket> rmatte's bug regarding linux zenpacks is still missing a couple modellers
[09-Jun-2009 14:56:21] <rocket> err collector plugins
[09-Jun-2009 14:56:31] <rocket> eg zenoss.cmd.linux.netstat_an
[09-Jun-2009 14:56:34] <rocket> eg zenoss.cmd.linux.netstat_rn
[09-Jun-2009 14:56:38] <rocket> eg zenoss.cmd.linux.process
[09-Jun-2009 14:56:50] <rocket> need another patch to include those?
[09-Jun-2009 14:57:13] <rocket> or can we just add those three lines to that file in __init__.py
[09-Jun-2009 14:57:28] <mrayzenoss> do those collector plugins work?
[09-Jun-2009 14:57:38] <kwame> Hi, question
[09-Jun-2009 14:57:42] <rocket> the netstat ones seem to
[09-Jun-2009 14:57:52] <rocket> they added routing info for me
[09-Jun-2009 14:58:00] <kwame> when I install zenoss and then tell it to autodiscover and give it a username and passwword for an linux user
[09-Jun-2009 14:58:04] <kwame> to be able to ssh into the hosts
[09-Jun-2009 14:58:18] <kwame> is this information stored in a text file somewhere?
[09-Jun-2009 14:58:38] <mrayzenoss> sure, update the patch. If you want to roll up the patch into your LinuxMonitorAddon as well that way it'll be out before 2.5
[09-Jun-2009 14:59:04] <mrayzenoss> kwame: it's in the zodb somewhere
[09-Jun-2009 14:59:10] <rocket> ah ok .. i can do both ..
[09-Jun-2009 14:59:25] <rocket> predictive seem to be working better now?
[09-Jun-2009 14:59:37] <mrayzenoss> rocket: not yet, blew up again, still testing
[09-Jun-2009 14:59:46] <rocket> mrayzenoss: bah!
[09-Jun-2009 15:01:09] <rocket> mrayzenoss: wonder what failed now .. :/
[09-Jun-2009 15:02:05] <mrayzenoss> kwame: you can also use keys
[09-Jun-2009 15:03:06] <kwame> mrayzenoss: what I saw in some servers was that the user that I used to log in through ssh to the boxes, was leaving a defunct (zombie) process
[09-Jun-2009 15:03:21] <kwame> so there where a lot of my_username@notty in the boxes
[09-Jun-2009 15:03:36] * rocket starts working on the suse zenpack ..
[09-Jun-2009 15:03:51] <mrayzenoss> rocket: http://pastebin.com/m49ffaa92
[09-Jun-2009 15:04:05] <mrayzenoss> kwame: there's a ticket somewhere about leftover ssh sessions
[09-Jun-2009 15:04:12] <mrayzenoss> I think it may be fixed in 2.4.2
[09-Jun-2009 15:04:30] <rocket> mrayzenoss: did you use the new zenpack for predictive?
[09-Jun-2009 15:06:21] <mrayzenoss> yeah, I think so
[09-Jun-2009 15:06:27] <mrayzenoss> I bumped the version to 1.0.1
[09-Jun-2009 15:07:35] <mrayzenoss> let me remove the thresholds and try them again
[09-Jun-2009 15:08:21] <rocket> mrayzenoss: found the bug
[09-Jun-2009 15:08:27] <rocket> new one coming your way
[09-Jun-2009 15:08:42] <mrayzenoss> ahh, cool
[09-Jun-2009 15:08:55] <mrayzenoss> I uninstalled my 1.0.1, I'll merge and update to 1.0.2
[09-Jun-2009 15:09:20] <rocket> its trying to tune an rrd that doesnt exist yet ..
[09-Jun-2009 15:10:43] <rocket> mrayzenoss: are you just doing a check_ping rrd?
[09-Jun-2009 15:11:06] <mrayzenoss> yeah
[09-Jun-2009 15:11:37] <mrayzenoss> just using the Nagios check_ping steps to add a ping, figured that would be fun for predictive thresholds
[09-Jun-2009 15:12:03] <rocket> got the steps so I can try it here?
[09-Jun-2009 15:12:53] <mrayzenoss> http://www.zenoss.com/community/projects/zenpacks/nagios-check-ping
[09-Jun-2009 15:12:55] <sergeymasushko> how can I place FastEthernet0/0 Throughput graph on the 'Perf' tab? I get the OID from the ethernetCsmacd_64 template and add .1 at the end. when I do snmpwalk I get the result... I've added the OID in the perf template, when I do test on the datasource page I also get result, but graph don't shown.
[09-Jun-2009 15:13:00] <kwame> mrayzenoss: thanks for your help
[09-Jun-2009 15:13:06] <mrayzenoss> rocket: I just follow the PDF, don't need the ZenPack anymore
[09-Jun-2009 15:13:32] <rocket> mrayzenoss: what command do I put in?
[09-Jun-2009 15:13:44] <rocket> for the command template
[09-Jun-2009 15:14:02] <rocket> just trying to go quick on this one
[09-Jun-2009 15:15:09] <rocket> nevermind ..
[09-Jun-2009 15:15:41] <mrayzenoss> libexec/check_ping -H $devname -w 180,100% -c 300,100% | sed -e 's# - Packet loss#|LOSS#;s#%,##;s#ms##;s# = #=#g'
[09-Jun-2009 15:15:58] <mrayzenoss> can't copy and paste from that PDF, saved you a step
[09-Jun-2009 15:19:19] <CamargoBP> I'm having the hardest time to get this graph to average out 1 minute. It seems like it's averaging for the entire day!
[09-Jun-2009 15:21:20] <rmatte> hey Matt
[09-Jun-2009 15:21:22] <rmatte> http://img200.yfrog.com/img200/3739/zenmanager.jpg
[09-Jun-2009 15:21:33] <rmatte> the top view there, is that part of the enterprise LDAP zenpack?
[09-Jun-2009 15:21:50] <rmatte> I'm not familliar with that view, and I'm trying to help that guy with his LDAP config
[09-Jun-2009 15:23:09] <rmatte> oh wait, nevermind, it's the view that you see when you click on the actual user in zenoss
[09-Jun-2009 15:24:05] <CamargoBP> Here is what I am trying to do, I want to see how many voicemail we send a minute. the larger spike at the right of this image (blue spike) was 20 voicemails yet the graph shows max 2
[09-Jun-2009 15:24:06] <CamargoBP> http://i43.tinypic.com/2qdo2rk.jpg
[09-Jun-2009 15:24:38] <rmatte> Camargo, if you SNMPWalk that value, do you see higher than 2?
[09-Jun-2009 15:25:11] <rmatte> do an snmpwalk and compare the value that you get to the value displayed as the current value in the graph
[09-Jun-2009 15:25:23] <rocket> mrayzenoss: ping
[09-Jun-2009 15:25:34] <rmatte> I have an idea of what's wrong, just need to see the values first to make sure
[09-Jun-2009 15:25:40] <rocket> mrayzenoss: did it fail if you went to look at the graph immediately after creating the threshold?
[09-Jun-2009 15:25:47] <mrayzenoss> rocket: yeah
[09-Jun-2009 15:25:51] <rocket> ok
[09-Jun-2009 15:25:56] <rocket> then I think I have it all fixed up
[09-Jun-2009 15:25:58] <mrayzenoss> well, after rebinding the template
[09-Jun-2009 15:26:11] <CamargoBP> rmatte: http://pastebin.com/d72c11e4a
[09-Jun-2009 15:27:00] <mrayzenoss> rmatte: that's from Zenoss Enterprise (the LDAP stuff)
[09-Jun-2009 15:27:15] <CamargoBP> When I zoom into the spike it says 20 emails
[09-Jun-2009 15:33:27] <rhettlap> do you have one minute rra's configured
[09-Jun-2009 15:34:07] <CamargoBP> probably not.
[09-Jun-2009 15:35:13] <rhettlap> easy way to check would be to rrdtool fetch the data source using -r 60 and see if the time shows 60 or 300 seconds
[09-Jun-2009 15:35:33] <CamargoBP> Here is my command: http://i40.tinypic.com/16lk9ef.png
[09-Jun-2009 15:35:34] <rhettlap> i.e. rrdtool fetch patchtodatasource.rrd AVERAGE -r 60 -s -1h
[09-Jun-2009 15:36:08] <CamargoBP> That gives the last hours 1 minute average
[09-Jun-2009 15:36:20] <rhettlap> thats just for the visual not for the data source
[09-Jun-2009 15:36:35] <rmatte> does anyone here have experience with getting the LDAP authentication working with AD? I've got mine working fine with plain LDAP, but this guy on the forums is having problems getting it working with Active Directory (which I have no experience with).
[09-Jun-2009 15:36:42] <CamargoBP> rmatte
[09-Jun-2009 15:36:49] <rmatte> More specifically with getting the role mapping working
[09-Jun-2009 15:36:51] <CamargoBP> rmatte: yes we have it working
[09-Jun-2009 15:36:56] <rhettlap> your data source shows 60 second data points?
[09-Jun-2009 15:37:02] <CamargoBP> role mapping does not work, sorry
[09-Jun-2009 15:37:20] <rmatte> role mapping just flat out doesn't work with AD?
[09-Jun-2009 15:37:22] <CamargoBP> You have to manually add a user to a role, can't map roles to AD
[09-Jun-2009 15:37:34] <CamargoBP> I spent weeks trying to get it working
[09-Jun-2009 15:37:49] <CamargoBP> Works with openldap but not AD
[09-Jun-2009 15:37:56] <CamargoBP> I don't know why
[09-Jun-2009 15:37:58] <rmatte> CamargoBP: that output you pasted in pastebin... which one of those lines represents the number of voicemails?
[09-Jun-2009 15:38:29] <CamargoBP> The ones highlited, line 10 specifically.
[09-Jun-2009 15:38:54] <rocket> hey matt ... get my updated zenpack?
[09-Jun-2009 15:39:32] <rmatte> so in theory you could setup an OpenLDAP server, configure it to sync up the logins with the AD server and use that to map the roles
[09-Jun-2009 15:39:55] <CamargoBP> So in that pastebin 68 emails were sent since I've started monitoring, which are all represented on that graph, especially if I zoom in then you can see them all
[09-Jun-2009 15:40:23] <rmatte> ok, but you don't see them as 68, you see them as some smaller value?
[09-Jun-2009 15:41:18] <rmatte> correct/
[09-Jun-2009 15:41:19] <rmatte> ?
[09-Jun-2009 15:41:31] <CamargoBP> yeah, this is a better picture of the graph, the one on our HUD
[09-Jun-2009 15:41:32] <CamargoBP> http://i43.tinypic.com/2cyr7fq.png
[09-Jun-2009 15:41:59] <CamargoBP> So that large spike was 20 emails of the value 68
[09-Jun-2009 15:42:42] <rmatte> but it's only showing 4
[09-Jun-2009 15:42:46] <rmatte> or 2
[09-Jun-2009 15:42:49] <rmatte> ok, try this...
[09-Jun-2009 15:42:58] <rmatte> (not sure if you've already done this)
[09-Jun-2009 15:43:01] <CamargoBP> Exactly, the max should be at around 20 on the left.
[09-Jun-2009 15:43:03] <rmatte> go in to the template
[09-Jun-2009 15:43:05] <rmatte> open the graph
[09-Jun-2009 15:43:10] <rmatte> click on the datapoint
[09-Jun-2009 15:43:19] <rmatte> let me know when you've done this
[09-Jun-2009 15:43:34] <CamargoBP> I'm there
[09-Jun-2009 15:43:37] <rmatte> Under format, is it currently %5.2lf%s?
[09-Jun-2009 15:43:43] <CamargoBP> yep
[09-Jun-2009 15:43:50] <rmatte> remove the %s at the end
[09-Jun-2009 15:44:03] <rmatte> that'll keep Zenoss from trying to auto-convert the value to something else
[09-Jun-2009 15:44:04] <CamargoBP> What does that standfor?
[09-Jun-2009 15:44:34] <rmatte> not sure exactly what it stands for, I just know that removing the %s forces Zenoss to interpret the value literally
[09-Jun-2009 15:45:08] <rmatte> that should solve your problem
[09-Jun-2009 15:45:11] <CamargoBP> Hmmm... no go.
[09-Jun-2009 15:45:18] <rmatte> you may or may not have to delete the .RRD file after doing this
[09-Jun-2009 15:45:22] <CamargoBP> Still at 4
[09-Jun-2009 15:45:26] <CamargoBP> K
[09-Jun-2009 15:45:27] <rmatte> or it could take a few polling cycles for it to realize
[09-Jun-2009 15:45:29] <CamargoBP> Let me do that
[09-Jun-2009 15:45:32] <rmatte> k
[09-Jun-2009 15:45:55] <rmatte> I think it hard codes the units in to the RRD file somewhere, so deleting the file and waiting 3 polling cycles for it to regenerate is your best bet
[09-Jun-2009 15:47:15] <rmatte> I need someone to explain that unit string to me sometime
[09-Jun-2009 15:48:13] <rmatte> format string rather
[09-Jun-2009 15:49:40] <CamargoBP> Dang, so I deleted it and then it took about 3 cycles to make it again. same output.
[09-Jun-2009 15:49:55] <rmatte> really strange then
[09-Jun-2009 15:50:03] <CamargoBP> WAit!
[09-Jun-2009 15:50:10] <CamargoBP> It was cached I think
[09-Jun-2009 15:50:21] <rmatte> oh, the image?
[09-Jun-2009 15:50:24] <CamargoBP> Nope.
[09-Jun-2009 15:50:27] <rmatte> shift-refresh
[09-Jun-2009 15:50:30] <rmatte> lol
[09-Jun-2009 15:50:38] <rmatte> to be sure
[09-Jun-2009 15:50:45] <CamargoBP> Shoot
[09-Jun-2009 15:50:50] <CamargoBP> yeah I did
[09-Jun-2009 15:51:14] <rmatte> hmmm
[09-Jun-2009 15:51:40] <rmatte> what's the SNMP value you're polling?
[09-Jun-2009 15:52:16] <rmatte> .1.3.6.1.4.1.2021.8.1.3.1?
[09-Jun-2009 15:52:27] <rmatte> .1.3.6.1.4.1.2021.8.1.5.1 rather?
[09-Jun-2009 15:53:20] <rmatte> if you run a test on the datapoint against the device what's the output?
[09-Jun-2009 15:54:17] <CamargoBP> 1.3.6.1.4.1.2021.8.1.101.1
[09-Jun-2009 15:54:41] <rmatte> ok, and what's the output of walking that?
[09-Jun-2009 15:54:47] <rmatte> it's 68 and not something else right?
[09-Jun-2009 15:55:14] <CamargoBP> UCD-SNMP-MIB::extOutput.1 = STRING: 78
[09-Jun-2009 15:55:15] <CamargoBP> UCD-SNMP-MIB::extOutput.1 = STRING: 78
[09-Jun-2009 15:55:15] <CamargoBP> UCD-SNMP-MIB::extOutput.1 = STRING: 78
[09-Jun-2009 15:55:15] <CamargoBP> well now 78
[09-Jun-2009 15:55:28] <CamargoBP> oops pasted 3 times... mybad
[09-Jun-2009 15:55:31] <rmatte> ok, but it outputs 3 separate lines?
[09-Jun-2009 15:55:34] <rmatte> ah ok
[09-Jun-2009 15:55:36] <rmatte> I was wondering lol
[09-Jun-2009 15:56:23] <rmatte> well, I'm out of ideas then, after removing that %s there's no reason why it shouldn't work
[09-Jun-2009 15:56:36] <rmatte> oh, you might have to push changes to the collector?
[09-Jun-2009 15:56:53] <CamargoBP> Here is the new graph after have deleted the old one and sending 10 more emails: http://i41.tinypic.com/35lraye.png
[09-Jun-2009 15:57:29] <rmatte> go to the templates section where the template is and do Manage -> Push Changes
[09-Jun-2009 15:57:35] <rmatte> then delete the RRD file and try again
[09-Jun-2009 15:57:44] <CamargoBP> k
[09-Jun-2009 15:58:07] <rmatte> what does del/min stand for?
[09-Jun-2009 15:58:14] <CamargoBP> deliveries
[09-Jun-2009 15:58:15] <CamargoBP>
[09-Jun-2009 15:58:24] <CamargoBP> The graph is too small so I did del
[09-Jun-2009 15:58:29] <CamargoBP> Scare my manager.
[09-Jun-2009 15:58:53] <rmatte> That shouldn't affect anything, but you may want to call it "voicemails per minute" or something, or just remove the units all together and see if that changes things
[09-Jun-2009 15:59:01] <rmatte> but try the push changes first
[09-Jun-2009 15:59:11] <rmatte> the collector only updates every hour
[09-Jun-2009 15:59:35] <rmatte> Zenoss does use what's in units to attempt to determine how it should convert values
[09-Jun-2009 15:59:44] <rmatte> like from bits to megabits, and so on
[09-Jun-2009 15:59:59] <rmatte> then again, with the %s removed it shouldn't even try to convert
[09-Jun-2009 16:00:58] <CamargoBP> Interesting... that it does that
[09-Jun-2009 16:01:10] <CamargoBP> Well I'm waiting for the new graph to be created
[09-Jun-2009 16:03:52] <rmatte> k
[09-Jun-2009 16:06:12] <daMaestro> <daMaestro> anyone know what data to enumerate when building a graph for disk space (not percent)
[09-Jun-2009 16:06:12] <daMaestro> <daMaestro> like, raw usage (GB/TB/PB/etc)
[09-Jun-2009 16:06:12] <daMaestro> <daMaestro> I currently use usedBlocks and then: ${here/totalBlocks},/,100,*
[09-Jun-2009 16:06:12] <daMaestro> <daMaestro> however, that does not work when trying a multgraph for multiple devices of storage usage
[09-Jun-2009 16:06:46] <CamargoBP>
No go[09-Jun-2009 16:18:47] <rmatte> daMaestro: no idea
[09-Jun-2009 16:28:13] <CamargoBP> Argh this graph thing is driving me crazy.
[09-Jun-2009 16:34:49] <rmatte> no idea why that graph is not working at this point
[09-Jun-2009 16:34:54] <rmatte> the %s should have fixed it
[09-Jun-2009 16:57:45] <cgibbons> uhoh
[09-Jun-2009 17:00:59] <mrayzenoss> yes cgibbons, that means you
[09-Jun-2009 17:02:24] <cgibbons> sure, pick on the guy that doesn't know anything
[09-Jun-2009 17:02:28] <cgibbons> JP said he wanted to do it again
[09-Jun-2009 19:00:44] <daMaestro> well... took me a little while but i figured it out
[09-Jun-2009 19:00:47] <daMaestro> usedBlocks_usedBlocks
[09-Jun-2009 19:01:01] <daMaestro> with an RPN of: ${here/blockSize},*,1073741824,/
[09-Jun-2009 19:01:06] <daMaestro> that will give you GB used
[09-Jun-2009 19:01:28] <daMaestro> and then you can stack multiple devices to get a birds eye of what is going on across storage devices
[09-Jun-2009 19:23:06] daMaestro|isBack is now known as daMaestro
[09-Jun-2009 21:09:40] timburke_ is now known as timburke
[09-Jun-2009 22:46:47] <centyx> Hi. I admit I haven't read the entire administrator's guide, but only picked through it, so forgive me if this question is answered therein. If it is, just tell me so please. Anyhow, using ZenOSS Core 2.4.1 how would I go about associating IP Services with a device class by default, so that when I create a new device under that device class it has those services monitored automatically, or something similar? Is there a way?
[10-Jun-2009 00:00:42] [disconnected at Wed Jun 10 00:00:42 2009]
[10-Jun-2009 00:00:42] [connected at Wed Jun 10 00:00:42 2009]
[10-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[10-Jun-2009 03:17:47] <gpant> hi, I have installed the HPMonitor v2.0.0 and the AdvDetail 1.1 zenpack, but when I go to the collector plugins for a device I cant see any of the HP plugins, (I do have 2 collectors called HPCPUMAP and HPDeviceMap which I think are completely different)
[10-Jun-2009 03:54:49] <blackgraz> []
[10-Jun-2009 05:27:15] <centyx> /c
[10-Jun-2009 06:57:21] <bwest> hello
[10-Jun-2009 06:57:59] <bwest> has anyone had success using winexe
[10-Jun-2009 08:31:14] <cgibbons> we used to use it internal to zenoss itself before 2.3.x, bwest. what's the question about it?
[10-Jun-2009 08:36:08] <rocket> morning cgibbons
[10-Jun-2009 08:36:16] <cgibbons> howdy
[10-Jun-2009 08:37:51] <rocket> I have been busy busy busy
... Extended the linuxmonitor zenpack .. and then have created a suse and gentoo one on top of that [10-Jun-2009 09:01:42] <rmatte> why make zenpacks for different linux distros?
[10-Jun-2009 09:01:50] <rmatte> what additional things does it monitor?
[10-Jun-2009 09:02:12] <rocket> rmatte: the only thing it adds is the package management stuff
[10-Jun-2009 09:02:19] <rmatte> ah
[10-Jun-2009 09:02:24] <rmatte> that's cool though
[10-Jun-2009 09:02:26] <rocket> eg gentoo's is different then redhats etc
[10-Jun-2009 09:02:40] <rocket> so you can monitor just fine without them
[10-Jun-2009 09:02:41] <rmatte> yeh, portage, yum, aptitude, etc...
[10-Jun-2009 09:03:02] <rocket> but if you care at all about what packages installed you would need a specific one for the distro
[10-Jun-2009 09:03:30] <rmatte> yeh
[10-Jun-2009 09:03:51] <rmatte> if you were to make one for ubuntu, that would probably be hugely popular
[10-Jun-2009 09:03:55] <rmatte> or is there one already?
[10-Jun-2009 09:04:05] <rocket> nope thats next on my list ..
[10-Jun-2009 09:04:11] <rmatte> (and it could be used for debian as well)
[10-Jun-2009 09:04:16] <rocket> probably have that one done tonight ..
[10-Jun-2009 09:04:21] <rmatte> cool
[10-Jun-2009 09:04:51] <rocket> rmatte: I would make a debian one just because there are a few uniq things between the two ..
[10-Jun-2009 09:05:00] <rmatte> awesome, the water cooler is actually full for once and I have the only cup that was left, more water for meeee
[10-Jun-2009 09:05:04] <rocket> rmatte: its just a matter of display preference though
[10-Jun-2009 09:05:14] <rmatte> ah
[10-Jun-2009 09:05:25] <rmatte> they both use aptitude, so I figured they'd be very similar
[10-Jun-2009 09:05:33] <rocket> eg manufacturer = debian
[10-Jun-2009 09:05:38] <chudler> perhaps one zenpack that detects them as it runs?
[10-Jun-2009 09:05:39] <rmatte> ah
[10-Jun-2009 09:05:44] <rocket> eg manufacturer = ubuntu ...
[10-Jun-2009 09:06:07] <rocket> chudler: that is a possibility yes .. I probably could do that
[10-Jun-2009 09:06:10] <chudler> in fact, one zenpack per distro sucks for users + all that code duplication?
[10-Jun-2009 09:06:31] <chudler> OTOH, maybe average install would only want one or two.
[10-Jun-2009 09:06:55] <rocket> chudler: thats what I was thinking .. and in reality some of the code is the same
[10-Jun-2009 09:07:00] <rmatte> you could make a script that does a "uname -r | cut -d '-' -f 2"
[10-Jun-2009 09:07:10] <rocket> chudler: but other aspects of parsing the packager info is different
[10-Jun-2009 09:07:14] <rmatte> then configures it based on the return value
[10-Jun-2009 09:07:52] <rocket> rmatte: except I know of distros that dont always return the right info out of uname
[10-Jun-2009 09:08:06] <rocket> rmatte: or if you have a stock kernel compiled etc that wont work
[10-Jun-2009 09:08:27] <chudler> /etc/issue? LSB is strong on this. or at least so I thought.
[10-Jun-2009 09:08:35] <chudler> detecting distro would be the easy part, I suspect.
[10-Jun-2009 09:09:13] <rocket> chudler: yeah .. again I might just do them individually and if someone wants to merge them they are free to do so as well
[10-Jun-2009 09:09:24] <rocket> chudler: we are only using suse in house here etc ..
[10-Jun-2009 09:09:30] <rmatte> you could look in /etc for release files
[10-Jun-2009 09:09:35] <rmatte> but some distros have more than one
[10-Jun-2009 09:09:55] <rocket> rmatte: thats what I am looking at for suse to detect opensuse from suse enterprise for example
[10-Jun-2009 09:09:58] <rmatte> sabayon has the gentoo-release file as well as the sabayon-release
[10-Jun-2009 09:10:31] <rocket> rmatte: my main purpose is to get some examples out there so others can come up with something more complex
[10-Jun-2009 09:10:43] <rmatte> yeh
[10-Jun-2009 09:11:09] <rocket> especially since I have no desire to install each distro and figure out how it would go together .. etc
[10-Jun-2009 09:11:41] <rmatte> that's why you need to find a common ground between them
[10-Jun-2009 09:11:43] <rocket> rmatte: but my zenpacks have the __init__.py configured to setup the modeller for you, the test cases defined and custom packagers
[10-Jun-2009 09:11:59] <rmatte> test it on a few different distros and call those the "officially supported" ones
[10-Jun-2009 09:12:03] <rocket> err not custom packagers .. modellers
[10-Jun-2009 09:12:29] <rocket> rmatte: except as a volunteer i dont really want to officially support any of them
[10-Jun-2009 09:12:44] <rmatte> lol
[10-Jun-2009 09:13:10] <rmatte> ok, so say something like "This Zenpack is known to work with Suse, OpenSuse, Ubuntu, Debian, Gentoo..."
[10-Jun-2009 09:13:30] <rocket>
[10-Jun-2009 09:13:34] <rocket> morning matt
[10-Jun-2009 09:13:46] <rmatte> gmorning Matt
[10-Jun-2009 09:14:07] <mrayzenoss> morning guys
[10-Jun-2009 09:14:52] <rmatte> "cat /etc/*-rel*" seems to work for detecting the version
[10-Jun-2009 09:15:51] <mrayzenoss> today should be a good day for ZenPacks. No meetings on the calendar and 5 in the queue already
[10-Jun-2009 09:39:39] <cgibbons> I don't suppose anyone here uses the Komodo IDE for Python?
[10-Jun-2009 10:05:16] <etank> has anyone gotten the "Perfmon" and "SQL 2000/2005 Server Performance" ZenPacks to work?
[10-Jun-2009 10:05:24] <etank> i am getting just blank graphs
[10-Jun-2009 10:07:25] <etank> when i run the commands manually i get "ERROR: Failed to install service winexesvc - NT code 0x00000424"
[10-Jun-2009 10:08:29] <etank> i tried what was said on the forums about replacin the winexe 0.80 version with the 0.90 version but then i get "INTERNAL ERROR: Signal 11 in pid 7915 (4.0.0tp4-SVN-build-10231)"
[10-Jun-2009 10:11:36] <rocket> cgibbons: I dont but I am curious is that what you are using?
[10-Jun-2009 10:11:58] <cgibbons> most of the team uses eclipse, a couple of the oldies use vim
[10-Jun-2009 10:12:10] <cgibbons> everything is far from perfect sadly
[10-Jun-2009 10:12:39] <chudler> interesting. /me really needs to get on the eclipse bandwagon someday
[10-Jun-2009 10:13:20] <rocket> cgibbons: hrmm if I use vim does that make me an oldie? ;p
[10-Jun-2009 10:14:47] <cgibbons> by oldie i meant the peeps working on zenoss the longest
[10-Jun-2009 10:14:56] <rocket> ah I see ..
[10-Jun-2009 10:15:52] <etank> im using Zenoss Core 2.4.1 with Perfmon 1.2 and SQL Server Performance 1.2
[10-Jun-2009 10:16:06] <rmatte> is there a "get" function that I can use in a custom report to grab the first time a device was seen?
[10-Jun-2009 10:16:13] <rmatte> essentially the commission date
[10-Jun-2009 10:17:15] <rocket> rmatte: looking
[10-Jun-2009 10:18:10] <rocket> you looking for the "First Seen" variable on the status page?
[10-Jun-2009 10:18:52] <rmatte> essentially
[10-Jun-2009 10:18:57] <rmatte> but I need to use it in a report
[10-Jun-2009 10:19:19] <rocket> rmatte: sure just a sec let me find what is calling that from the page template
[10-Jun-2009 10:19:22] <rmatte> so far I'm using things like getId, getManageIp, getHWManufacturerName, etc...
[10-Jun-2009 10:19:29] <rmatte> thanks
[10-Jun-2009 10:20:14] <rocket> getCreatedTimeString
[10-Jun-2009 10:20:20] <rmatte> thanks
[10-Jun-2009 10:20:23] <rocket>
[10-Jun-2009 10:20:39] <rocket> for your info if you find something like that in the web interface
[10-Jun-2009 10:20:56] <rocket> the easiest way is to look for the page template
[10-Jun-2009 10:21:08] <rocket> eg ZenModel/skins/deviceStatus.pt
[10-Jun-2009 10:21:14] <rocket> grep for "First Seen"
[10-Jun-2009 10:21:28] <rocket> or look for that in the page template
[10-Jun-2009 10:21:41] <rocket> and it usually has something like here/getCreatedTimeString
[10-Jun-2009 10:21:43] <rmatte> ah, cool
[10-Jun-2009 10:21:47] <rmatte> didn't know it was that easy
[10-Jun-2009 10:21:56] <rocket> which gives you an idea of what was getting called
[10-Jun-2009 10:23:53] <rmatte> is there any easy way to pull values from RRD in to a template, but not as a graph, just as text?
[10-Jun-2009 10:24:02] <rmatte> in to a report I mean
[10-Jun-2009 10:24:17] <venturaville> here.getRRDValue('Name')
[10-Jun-2009 10:24:54] <venturaville> I don't know if that works in all of the fields though...
[10-Jun-2009 10:25:22] <rmatte> ah
[10-Jun-2009 10:25:42] <rocket> rmatte: there are a few examples .. just a sec
[10-Jun-2009 10:26:07] <venturaville> here.cacheRRDValue('Name') if you don't need it up to date.....
[10-Jun-2009 10:26:49] <rmatte> cool
[10-Jun-2009 10:27:11] <rocket> venturaville: yahtzee!!! .. that explains the one issue I was having with one of my functions .. it was always out of date!
[10-Jun-2009 10:27:21] <venturaville> indeed
[10-Jun-2009 10:27:36] <rocket> was looking for a different RRDValue function but didnt stumble across it
[10-Jun-2009 10:27:45] <venturaville> in zendmd:
[10-Jun-2009 10:27:59] <rocket> rmatte: here is an example
[10-Jun-2009 10:28:01] <rocket> def usedInodes(self, default = None):
[10-Jun-2009 10:28:01] <rocket> """
[10-Jun-2009 10:28:01] <rocket> Return the number of used inodes stored in the filesystem's rrd file
[10-Jun-2009 10:28:01] <rocket> """
[10-Jun-2009 10:28:04] <rocket> inodes = self.cacheRRDValue('usedInodes_usedInodes', default)
[10-Jun-2009 10:28:06] <rocket> if inodes is not None:
[10-Jun-2009 10:28:09] <rocket> return long(inodes)
[10-Jun-2009 10:28:11] <rocket> else:
[10-Jun-2009 10:28:12] <venturaville> d = devices.findDevice('mon8zen2')
[10-Jun-2009 10:28:14] <rocket> return None
[10-Jun-2009 10:28:33] <venturaville> d.getRRDDataPoints() ..... very long list of points
[10-Jun-2009 10:28:49] <venturaville> the name after the datapoints/ in each is what you want
[10-Jun-2009 10:28:50] <venturaville> like so:
[10-Jun-2009 10:29:04] <venturaville> >>> d.getRRDValue('ssCpuSystem')
[10-Jun-2009 10:29:04] <venturaville> 4.0
[10-Jun-2009 10:29:59] <venturaville> I'm not a big fan of the tales syntax, so I tend to fall back on the python syntax in templates.......
[10-Jun-2009 10:30:10] <venturaville> makes it easier to debug in zendmd
[10-Jun-2009 10:31:29] <rmatte> right
[10-Jun-2009 10:32:27] <rmatte> any way to pull the average value instead of the current?
[10-Jun-2009 10:32:31] <rmatte> or is that the cached value?
[10-Jun-2009 10:32:49] <rmatte> and any way to pull the max value?
[10-Jun-2009 10:34:04] <venturaville> rmatte: not certain ...let me look
[10-Jun-2009 10:34:09] <rmatte> thanks
[10-Jun-2009 10:34:25] <venturaville> cached = sticks around in zope cache for a bit I believe
[10-Jun-2009 10:34:31] <venturaville> don't believe it has anything to do with the average
[10-Jun-2009 10:34:43] <venturaville> it is a way to reduce the lookup overhead when used widely
[10-Jun-2009 10:35:17] <venturaville> getRRDContextData
[10-Jun-2009 10:35:17] <venturaville> getRRDDataPoint
[10-Jun-2009 10:35:17] <venturaville> getRRDDataPoints
[10-Jun-2009 10:35:17] <venturaville> getRRDFileName
[10-Jun-2009 10:35:17] <venturaville> getRRDNames
[10-Jun-2009 10:35:18] <venturaville> getRRDPaths
[10-Jun-2009 10:35:20] <venturaville> getRRDSum
[10-Jun-2009 10:35:22] <venturaville> getRRDTemplate
[10-Jun-2009 10:35:24] <venturaville> getRRDTemplateByName
[10-Jun-2009 10:35:26] <venturaville> getRRDTemplateName
[10-Jun-2009 10:35:28] <venturaville> getRRDTemplates
[10-Jun-2009 10:35:30] <venturaville> getRRDValue
[10-Jun-2009 10:35:32] <venturaville> getRRDValues
[10-Jun-2009 10:35:39] <venturaville> gleaned from: print '\n'.join(dir(d))
[10-Jun-2009 10:36:12] <rocket> actually you might need fetchRRDValue
[10-Jun-2009 10:36:14] <rocket> def fetchRRDValue(self, dpname, cf, resolution, start, end=""):
[10-Jun-2009 10:36:27] <rocket> where cf is AVERAGE or MAX
[10-Jun-2009 10:36:41] <venturaville> not seeing that one listed in the device objects
[10-Jun-2009 10:36:58] <rocket> ah i see its used by getRRDValues
[10-Jun-2009 10:37:12] <rocket> def getRRDValues(self, dsnames, start=None, end=None, function="LAST",
[10-Jun-2009 10:37:12] <rocket> format="%.2lf", extraRpn=""):
[10-Jun-2009 10:37:47] <rmatte> hmmm, now to figure out what get statement I can use to grab the device class
[10-Jun-2009 10:38:08] <venturaville> http://docs.huihoo.com/zenoss/api/2.1/private/ZenModel.RRDView.RRDView-class.html
[10-Jun-2009 10:38:40] <venturaville> getRRDSum(self, points, start, end, function)
[10-Jun-2009 10:38:40] <venturaville> Return a some of listed datapoints.
[10-Jun-2009 10:38:44] <venturaville> I think that might be what you want
[10-Jun-2009 10:38:49] <venturaville> (and me for that matter)
[10-Jun-2009 10:39:26] <venturaville> getRRDValue(self, dsname, start, end, function)
[10-Jun-2009 10:39:26] <venturaville> Return a single rrd value from its file using function.
[10-Jun-2009 10:39:29] <venturaville> nope that one is
[10-Jun-2009 10:39:32] <rocket> venturaville: that just calls getRRDValues and adds them together
[10-Jun-2009 10:39:34] <venturaville> it just has more optional arguments
[10-Jun-2009 10:40:32] <rocket> venturaville: getRRDValue just calls getRRDValues with a single datapoint
[10-Jun-2009 10:40:44] <venturaville> yep
[10-Jun-2009 10:40:55] <venturaville> convenience functions.....
[10-Jun-2009 10:42:23] <venturaville> something like this maybe:
[10-Jun-2009 10:42:44] <rmatte> bah
[10-Jun-2009 10:42:46] <venturaville> getRRDValue('ssCpuSystem',None,None,"AVERAGE")
[10-Jun-2009 10:42:51] <rmatte> trying to find the get value for Product Class
[10-Jun-2009 10:43:02] <rmatte> though it was getProductClass apparently not
[10-Jun-2009 10:44:16] <venturaville> >>> d.deviceClass()
[10-Jun-2009 10:44:16] <venturaville> <DeviceClass at /zport/dmd/Devices/Server/Linux/devices/mon8zen2/deviceClass/Linux>
[10-Jun-2009 10:44:18] <venturaville> this you mean?
[10-Jun-2009 10:44:34] <rmatte> yessir
[10-Jun-2009 10:44:49] <venturaville> >>> d.getDeviceClassName()
[10-Jun-2009 10:44:49] <venturaville> '/Server/Linux'
[10-Jun-2009 10:44:51] <venturaville> or in string form
[10-Jun-2009 10:44:53] <rocket> venturaville would this be more readable?
[10-Jun-2009 10:45:01] <venturaville> I searched for a while to find that one too
[10-Jun-2009 10:45:04] <rmatte> ah, was missing Name on the end when I tried it earlier
[10-Jun-2009 10:45:06] <rocket> getRRDValue('ssCpuSystem', function="AVERAGE")
[10-Jun-2009 10:45:16] <rocket> getRRDValue('ssCpuSystem', function="MAX")
[10-Jun-2009 10:45:19] <rocket> etc ..
[10-Jun-2009 10:45:31] <venturaville> you don't need the 2 start and end values?
[10-Jun-2009 10:46:07] <rocket> venturaville: shouldnt need it if you are naming the variable coming in
[10-Jun-2009 10:46:15] <venturaville> ah true
[10-Jun-2009 10:46:29] <venturaville> I'm getting my scripting languages mixed up
[10-Jun-2009 10:47:00] <rocket> give it a try
[10-Jun-2009 10:47:02] <rocket>
[10-Jun-2009 10:47:13] <rocket> >>> d = devices.findDevice('192.168.31.3')
[10-Jun-2009 10:47:13] <rocket> >>> d.getRRDValue('ssCpuSystem', function="AVERAGE")
[10-Jun-2009 10:47:13] <rocket> 2.7400000000000002
[10-Jun-2009 10:47:32] <rocket> >>> d.getRRDValue('ssCpuSystem', function="MAXIMUM")
[10-Jun-2009 10:47:32] <rocket> 32.210000000000001
[10-Jun-2009 10:47:35] <venturaville> >>> d.getRRDValue('ssCpuSystem',function="LAST")
[10-Jun-2009 10:47:36] <venturaville> 6.2199999999999998
[10-Jun-2009 10:47:36] <venturaville> >>> d.getRRDValue('ssCpuSystem',function="AVERAGE")
[10-Jun-2009 10:47:36] <venturaville> 4.7599999999999998
[10-Jun-2009 10:47:36] <venturaville> >>> d.getRRDValue('ssCpuSystem',function="MAXIMUM")
[10-Jun-2009 10:47:36] <venturaville> 6.25
[10-Jun-2009 10:47:38] <venturaville> sweet
[10-Jun-2009 10:47:46] <rmatte> so would I do getCacheRRDValue or getcacheRRDValue ?
[10-Jun-2009 10:47:46] <rocket>
[10-Jun-2009 10:47:50] <venturaville> that will help me fix a module I've been working on
[10-Jun-2009 10:48:00] <venturaville> just cacheRRDValue()
[10-Jun-2009 10:48:23] <rmatte> I'm looking for the function that I can actually use in a report
[10-Jun-2009 10:48:45] <venturaville> >>> print '\n'.join(dir(devices.findDevice('mon8zen2')))
[10-Jun-2009 10:48:54] <venturaville> print that out using some random device
[10-Jun-2009 10:49:05] <venturaville> you'll save yourself a lot of headaches......
[10-Jun-2009 10:49:31] <rocket>
[10-Jun-2009 11:21:45] <rmatte> If you set a device's model info by hand, will it write over it if you remodel the device?
[10-Jun-2009 11:21:59] <rmatte> manufacturer/model info I mean
[10-Jun-2009 11:42:41] <rhettlap> no
[10-Jun-2009 12:04:51] <rmatte> k thanks
[10-Jun-2009 12:50:14] <rmatte> .
[10-Jun-2009 13:00:36] <etank> is anyone having issues with winexe ?
[10-Jun-2009 13:01:15] <etank> ERROR: Failed to install service winexesvc - NT code 0x00000424 <-- thats what im getting
[10-Jun-2009 13:02:41] <etank> and using the 0.90 version just gives different errors
[10-Jun-2009 13:23:24] <rmatte> this page loading issue in zenoss can be a real pain sometimes
[10-Jun-2009 13:23:45] <rmatte> it's even worse in 2.4.1 than it was in 2.3.3
[10-Jun-2009 14:01:01] * Bryanstein is away:
[10-Jun-2009 14:01:45] * Bryanstein is back (gone 00:00:30)
[10-Jun-2009 14:20:44] <etank> i guess that no one uses the community Perfmon and SQL Server ZenPacks ?
[10-Jun-2009 14:32:51] <mrayzenoss> etank: I know there are at least 2 users, the original author and the one who updated them both
[10-Jun-2009 14:33:12] <mrayzenoss> that said, I don't have access to SQL Server
[10-Jun-2009 14:42:00] <etank> mrayzenoss: i wish i didn't
[10-Jun-2009 14:42:17] <etank> im just not having any luck getting those two to work at all
[10-Jun-2009 15:10:38] <javierglaz> hello
[10-Jun-2009 15:11:04] <javierglaz> is this a zenoss support channel?
[10-Jun-2009 15:12:49] <rmatte> is there any way to set a recurring maintenance window for a device?
[10-Jun-2009 15:13:13] <rmatte> we have a scheduled device reboot every week, we need to set a recurring maintenance window so that we don't get a flood of events during the reboot.
[10-Jun-2009 15:15:01] <rmatte> oh nevermind, there's a repeat option
[10-Jun-2009 15:15:40] <javierglaz> I have a problem with zensyslog
[10-Jun-2009 15:16:02] <javierglaz> I am monitoring cisco devices, and the logging host is currently pointing to the zenoss box
[10-Jun-2009 15:16:32] <rmatte> but you're not seeing any syslog events coming in?
[10-Jun-2009 15:16:39] <javierglaz> if I issue a tail -f $ZENHOME/log/zensyslog.log
[10-Jun-2009 15:16:42] <javierglaz> I see the events
[10-Jun-2009 15:16:57] <javierglaz> and I see the regex match also
[10-Jun-2009 15:17:13] <javierglaz> but I don't get any events in the web console
[10-Jun-2009 15:17:20] <rmatte> Zenoss only really translates syslog messages to an actual Zenoss event when the syslog event comes in with a certain level
[10-Jun-2009 15:17:37] <javierglaz> can I paste one event I get?
[10-Jun-2009 15:17:41] <rmatte> if it's a lower severity event, you won't see it
[10-Jun-2009 15:17:42] <rmatte> yes
[10-Jun-2009 15:18:25] <rmatte> Go to Event Manager and check the "Default Syslog Priority" setting
[10-Jun-2009 15:19:05] <rmatte> we currently have ours set at 3, with 1 being the highest severity and 7 being the lowest
[10-Jun-2009 15:19:46] <javierglaz> 2009-06-10 17:19:22 DEBUG zen.Syslog: tag match: {'eventClassKey': 'LINK-3-UPDOWN', 'component': 'LINK', 'summary': 'Interface Loopback999, changed state to up'}
[10-Jun-2009 15:19:46] <javierglaz> 2009-06-10 17:19:22 DEBUG zen.ZenSyslog: Queueing event {'firstTime': 1244665162.0504229, 'severity': 4, 'facility': 'local7', 'eventClassKey': 'LINK-3-UPDOWN', 'component': 'LINK', 'agent': 'zensyslog', 'summary': 'Interface Loopback999, changed state to up', 'priority': 3, 'manager': 'localhost', 'eventGroup': 'syslog', 'device': '172.16.100.25', 'lastTime': 1244665162.0504229, 'ipAddress': '172.16.100.25', 'monitor': 'localhost'}
[10-Jun-2009 15:19:46] <javierglaz> 2009-06-10 17:19:22 DEBUG zen.ZenSyslog: Total of 1 queued events
[10-Jun-2009 15:20:05] <rmatte> ok, so the severity is 4
[10-Jun-2009 15:20:16] <rmatte> what is your Default Syslog Priority set to in Event Manager?
[10-Jun-2009 15:20:23] <javierglaz> let me check
[10-Jun-2009 15:20:26] <rmatte> k
[10-Jun-2009 15:20:34] <javierglaz> 3
[10-Jun-2009 15:20:49] <rmatte> ok, so Zenoss won't create events for syslogs below 3 for you right now
[10-Jun-2009 15:20:58] <rmatte> if you bump it to 4 or 5 then you should see those events come in
[10-Jun-2009 15:21:21] <javierglaz> I see.... let me try
[10-Jun-2009 15:21:25] <rmatte> k
[10-Jun-2009 15:23:59] <javierglaz> I've raised it to 5 but still no luck....
[10-Jun-2009 15:24:32] <javierglaz> ah but I didn't restart the process... one sec
[10-Jun-2009 15:24:37] <rmatte> k
[10-Jun-2009 15:25:05] <rmatte> you might also have to push the changes to the collector
[10-Jun-2009 15:25:10] <rmatte> but I doubt it
[10-Jun-2009 15:26:43] <javierglaz> still no luck...
[10-Jun-2009 15:27:07] <javierglaz> how do you push the changes to the collector?
[10-Jun-2009 15:27:19] <rmatte> Manage -> Push Changes
[10-Jun-2009 15:27:40] <rmatte> do it under the Devices section
[10-Jun-2009 15:29:58] <rmatte> it looks like Zenoss is trying to create the event but queueing it instead
[10-Jun-2009 15:30:00] <rmatte> strange
[10-Jun-2009 15:30:26] <javierglaz> still not working.... I am getting some events with severity 2 now.... so it should have worked with the default severity of 3 I had before
[10-Jun-2009 15:30:46] <rmatte> hmmm
[10-Jun-2009 15:30:55] <javierglaz> so it must be something else.... the strange message is the queuing one
[10-Jun-2009 15:31:27] <rmatte> yeh
[10-Jun-2009 15:31:34] <rmatte> not sure why it would be "queuing" an event
[10-Jun-2009 15:33:07] <javierglaz> unfortunately I couldn't find nothing in google with this
[10-Jun-2009 15:34:26] <rmatte> same here
[10-Jun-2009 15:38:23] <javierglaz> I also see some heartbeat messages in that log
[10-Jun-2009 15:38:36] <javierglaz> but they don't appear in the web console
[10-Jun-2009 15:38:58] <javierglaz> 2009-06-10 17:38:53 DEBUG zen.ZenSyslog: Queueing event {'manager': 'localhost', 'timeout': 180, 'device': 'localhost', 'eventClass': '/Heartbeat', 'component': 'zensyslog', 'agent': 'zensyslog'}
[10-Jun-2009 15:38:58] <javierglaz> 2009-06-10 17:38:53 DEBUG zen.ZenSyslog: Total of 1 queued events
[10-Jun-2009 15:42:51] <javierglaz> does anyone knows what this message means? DEBUG zen.ZenSyslog: Total of 1 queued events
[10-Jun-2009 15:43:14] <chudler> what version of zenoss? I couldn't find this in the code anywhere.
[10-Jun-2009 15:43:44] <javierglaz> 2.4
[10-Jun-2009 15:44:12] <chudler> oh I did find it, I was looking in an older one :-)
[10-Jun-2009 15:44:18] <javierglaz> 2.4.1
[10-Jun-2009 15:44:50] <chudler> AFAICT, it is a normal course of events for the event to be queued before processing. But take that FWIW, I just glanced at it.
[10-Jun-2009 15:45:25] <chudler> someone somewhere once published a detailed events processing document (Jane Curry?).
[10-Jun-2009 15:45:56] <mrayzenoss> http://www.skills-1st.co.uk/papers/jcurry.html
[10-Jun-2009 15:46:06] <mrayzenoss> Zenoss Event Management
[10-Jun-2009 15:47:13] <javierglaz> downloading
[10-Jun-2009 15:49:42] <javierglaz> what should I be looking at after comfirming that the logs are hitting the zenoss box, and zenoss is parsing them succesfully?
[10-Jun-2009 15:52:03] <javierglaz> in the document says that the last line should be sending event
[10-Jun-2009 15:52:10] <javierglaz> instead of that I get queueing event
[10-Jun-2009 16:13:38] <mrayzenoss> anyone looking for a Zenoss job in the bay area?
[10-Jun-2009 16:15:09] <javierglaz> is someone using zenoss 2.4.1 as syslog server for cisco devices?
[10-Jun-2009 16:20:35] <magnachef__> is anyone around that has any experience with monitoring websites/vhosts with HttpMonitor
[10-Jun-2009 16:31:15] <magnachef__> well, if anyone can take a look at an issue I'm having, I'd appreciate it: http://forums.zenoss.com/viewtopic.php?t=10014
[11-Jun-2009 00:00:42] [disconnected at Thu Jun 11 00:00:42 2009]
[11-Jun-2009 00:00:42] [connected at Thu Jun 11 00:00:42 2009]
[11-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[11-Jun-2009 01:53:49] <frozty_sa> morning all
[11-Jun-2009 03:23:49] <sergeymasushko> how can I debug cheking of availability http/https ports on devices?
[11-Jun-2009 03:27:53] <tehhobbit> debug ?
[11-Jun-2009 04:33:52] <stefanlsd> I am trying to write a zenoss script that runs setManageIp() . It works from the zendmd interface, but not the script. In the interface, after a loop i push enter (tells it loop is finished or something), this doesnt work from script? Any ideas?
[11-Jun-2009 04:39:40] <stefanlsd> needed an extra blank line in the script... thanks for all the help
[11-Jun-2009 04:40:37] <sergeymasushko> tehhobbit: I tried to figure out lies zenoss or not
[11-Jun-2009 08:04:50] <ckrough> stefanlsd: how does your script interface with zendmd?
[11-Jun-2009 08:04:59] <ckrough> bah, hes gone.
[11-Jun-2009 08:06:33] <ckrough> why would a device have an "unknown" uptime
[11-Jun-2009 08:13:06] <rocket> Jontler: umn .. hmmm I went there
[11-Jun-2009 08:13:32] <Jontler> it's a nice school
[11-Jun-2009 08:14:16] <rocket> Yup .. its been a few years now ..
[11-Jun-2009 08:14:25] <Jontler> funny how that happens
[11-Jun-2009 08:15:45] <rocket> yeah
[11-Jun-2009 08:15:49] <Jontler> we're looking at using zenoss to replace nagios here at the math department
[11-Jun-2009 08:15:51] <rocket> what are you in for
[11-Jun-2009 08:16:04] <Jontler> ha, well, i just recently graduated with my math degree
[11-Jun-2009 08:16:17] <Jontler> I'm part of the IT staff at the math department, though
[11-Jun-2009 08:16:43] <rocket> Jontler: well zenoss can use nagios plugins so there is a bonus there
[11-Jun-2009 08:16:48] <Jontler> indeed
[11-Jun-2009 08:17:15] <Jontler> it can also do most of the things we use nagios for now natively, which is also a bonus
[11-Jun-2009 08:17:21] <rocket> Jontler: I am local if you have questions ...
[11-Jun-2009 08:17:28] <rocket> Minnetonka
[11-Jun-2009 08:17:35] <Jontler> nice
[11-Jun-2009 08:18:01] <Jontler> well, my question right now is involving the performance templates for process monitoring
[11-Jun-2009 08:18:09] <rocket> ok
[11-Jun-2009 08:18:19] <Jontler> it seems that each process being monitored defaults to using the standard OSProcess template
[11-Jun-2009 08:18:32] <Jontler> id like to change that, but can't seem to figure out how
[11-Jun-2009 08:18:53] <rocket> so basically a new OSProcess Template?
[11-Jun-2009 08:18:58] <Jontler> yep
[11-Jun-2009 08:21:10] <rocket> ok .. well I believe you will have to apply your new osprocess template per device .. its not granular to the per process leve yet
[11-Jun-2009 08:21:22] <rocket> so what you could do is create a new device organizer
[11-Jun-2009 08:21:32] <rocket> for the device thats unique
[11-Jun-2009 08:21:49] <rocket> and copy the template there and update it from there
[11-Jun-2009 08:21:56] <rocket> does that make sense?
[11-Jun-2009 08:21:58] <Jontler> yea
[11-Jun-2009 08:22:03] <Jontler> it's what i tried yesterday
[11-Jun-2009 08:22:10] <rocket> didnt work?
[11-Jun-2009 08:22:24] <Jontler> basically, ive got a lab of machines in a device organizer now
[11-Jun-2009 08:22:48] <rocket> it will have to have the same name as the original OSProcess template because its unique
[11-Jun-2009 08:22:50] <Jontler> i created a new template (MATLAB, in this case) to monitor the matlabl process
[11-Jun-2009 08:22:53] <Jontler> oh
[11-Jun-2009 08:22:58] <rocket> and you must not bind it
[11-Jun-2009 08:22:59] <Jontler> i wasnt clear
[11-Jun-2009 08:23:13] <Jontler> actually
[11-Jun-2009 08:23:15] <Jontler> hm
[11-Jun-2009 08:23:21] <Jontler> i think i understand now
[11-Jun-2009 08:23:24] <Jontler> one moment
[11-Jun-2009 08:27:00] <Jontler> so, since I cannot have a template per process, I have to define one template for a variety of processes, and then apply that to the devices that are are using those processes
[11-Jun-2009 08:27:21] <rocket> basically yes
[11-Jun-2009 08:27:26] <Jontler> and that template must have the same name, since each process always uses the OSProcess template
[11-Jun-2009 08:28:11] <rocket> yep
[11-Jun-2009 08:28:27] * Jontler has a feature request
[11-Jun-2009 08:28:30] <Jontler>
[11-Jun-2009 08:28:37] <Jontler> i think i can work with this though
[11-Jun-2009 08:28:40] <rocket> or if you want you can create a totally custom datacollector
[11-Jun-2009 08:29:15] <rocket> actually here is another way to do it
[11-Jun-2009 08:31:30] <rocket> Jontler: how good is your scripting?
[11-Jun-2009 08:31:39] <Jontler> passable
[11-Jun-2009 08:31:45] <Jontler> i can make things work
[11-Jun-2009 08:31:53] <rocket> another way to do it is to look at the dev guide
[11-Jun-2009 08:31:57] <rocket> create a new template
[11-Jun-2009 08:32:06] <rocket> add a command datapoint
[11-Jun-2009 08:32:29] <rocket> in that command datapoint you connect to the machine you need information from via a nagios type script
[11-Jun-2009 08:32:46] <rocket> save the variables as datapoints
[11-Jun-2009 08:33:00] <rocket> and then create a new graph based on these new datapoints
[11-Jun-2009 08:33:22] <rocket> final step is to bind the template to the device organizer that you need this to run against
[11-Jun-2009 08:33:41] <rocket> this will create a graph on the performance tab
[11-Jun-2009 08:33:48] <rocket> make sense?
[11-Jun-2009 08:34:01] <Jontler> yea, you're skirting the process portion of zenoss altogether
[11-Jun-2009 08:34:12] <rocket> yes
[11-Jun-2009 08:34:23] <rocket> the process portion of zenoss is semi weak
[11-Jun-2009 08:34:35] <Jontler> so ive noticed
[11-Jun-2009 08:34:44] <rocket> eg if you turn on monitoring of ssh for example it is on for all systems
[11-Jun-2009 08:34:57] <rocket> process monitoring is pretty global in zenoss
[11-Jun-2009 08:35:18] <Jontler> yea, i added sshd as a process yesterday
[11-Jun-2009 08:35:28] <Jontler> there were a lot of sshd processes
[11-Jun-2009 08:36:00] <Jontler> i assume there are plans in the future to expand on the process monitoring
[11-Jun-2009 08:36:21] <rocket> yeah from what I have heard
[11-Jun-2009 08:36:37] <Jontler> well that's neat
[11-Jun-2009 08:36:40] <rocket> I am just a community member so I couldnt say for sure .. but I know others have had issues with it
[11-Jun-2009 08:36:54] <Jontler> im just trying to get a handle on the different parts of zenoss
[11-Jun-2009 08:37:02] <Jontler> see what might be useful and what might not be
[11-Jun-2009 08:37:17] <Jontler> it already looks like it can take over for nagios without missing a beat
[11-Jun-2009 08:37:28] <rocket> take a look at the nagios section etc thats probably closer to what you need
[11-Jun-2009 08:37:46] <rocket> and/or the dev guide to add ssh based monitoring stuff
[11-Jun-2009 08:37:57] <Jontler> ive already got the ssh based monitoring rolling
[11-Jun-2009 08:38:27] <rocket> Jontler: I am talking more about sshing into a system and gathering its stats for something
[11-Jun-2009 08:38:48] <rocket> eg I have created code that gathers df output and parses it in zenoss etc
[11-Jun-2009 08:39:01] <Jontler> ive installed the zenoss-plugins on our workstations
[11-Jun-2009 08:39:11] <Jontler> that seems to do a good job of allowing data to be gathered natively
[11-Jun-2009 08:39:14] <ckrough> rocket: you can monitor storage space through snmp
[11-Jun-2009 08:39:36] <rocket> ckrough: yes you can unless you have an unstable snmp implementation for example
[11-Jun-2009 08:39:47] <rocket> ckrough: for example aix's snmp daemon really sucks
[11-Jun-2009 08:39:48] <ckrough> heh, that would be an issue
[11-Jun-2009 08:39:49] <ckrough> what os
[11-Jun-2009 08:39:50] <ckrough> ah
[11-Jun-2009 08:40:17] <rocket> ckrough: there is a patch for it I havent tried yet but at the moment snmp crashes at least once a week
[11-Jun-2009 08:40:33] <ckrough> rocket: well, nobody likes that
[11-Jun-2009 08:40:37] <rocket> ckrough: so for me zenoss provides another work around
[11-Jun-2009 08:41:06] <rocket> ckrough: df may not be the best example .. but I can also create scripts that gather information out of our backup servers cmdline client
[11-Jun-2009 08:42:34] <ckrough> I need to put together a report that shows devices (/Network) with uptimes of less than x. I built a report in the custom device reports, but it's just too slow. Is there a way to interact with zendmd from outside of zendmd?
[11-Jun-2009 08:42:53] <ckrough> I guess I could load it up in a python script.
[11-Jun-2009 08:43:14] <rocket> ckrough: yeah I think there is an example in the dev guide ..
[11-Jun-2009 08:43:44] <ckrough> that would be perfect. I have trouble finding what I need in the docs.
[11-Jun-2009 08:45:12] <rocket> ckrough: basically you will be creating a python script that is loading the modules you need for zendmd without the full zendmd
[11-Jun-2009 08:45:14] <Jontler> ckrough: me too - hence i came here
[11-Jun-2009 08:45:38] <rocket> http://www.zenoss.com/Members/netdata/use-a-zendmd-script-as-standalone-program/
[11-Jun-2009 08:45:42] <ckrough> rocket: thanks. here's hoping its faster than the built-in report
[11-Jun-2009 08:45:59] <ckrough> rocket: awesome
[11-Jun-2009 08:46:05] <rocket>
[11-Jun-2009 08:46:34] <rocket> np
[11-Jun-2009 08:49:29] <ckrough> is there a way to connect to zendmd on a remote machine?
[11-Jun-2009 08:49:52] <venturaville> btw for the custom device reports you can actually put the python code inside of a plugin and just call it from a report
[11-Jun-2009 08:50:00] <ckrough> I guess REST cals are about as close as you can get
[11-Jun-2009 08:50:18] <ckrough> nice, that might be an improvement
[11-Jun-2009 08:50:24] <venturaville> or put the code in a custom python script in zope, and call it from REST
[11-Jun-2009 08:50:35] <venturaville> again faster than REST/XML-RPC
[11-Jun-2009 08:51:48] <venturaville> look at Products/ZenReports/plugins ....
[11-Jun-2009 08:53:30] <venturaville> and: Products/ZenReports/reports/Performance_Reports/*
[11-Jun-2009 08:54:42] <ckrough> does the ZenReports/reports/ directory structure directly related to the report locations in the gui?
[11-Jun-2009 08:55:35] <venturaville> has more to do with the class names........
[11-Jun-2009 08:55:41] <venturaville> but pretty close
[11-Jun-2009 08:56:25] <ckrough> so, why are my custom device reports zope objects, but these device reports rpts on the file system
[11-Jun-2009 08:58:26] <rmatte> the native monitoring of interfaces in Zenoss is done via ping is it not?
[11-Jun-2009 08:58:37] <ckrough> availability monitoring is, yes
[11-Jun-2009 08:58:54] <rmatte> I mean for port status
[11-Jun-2009 08:59:09] <rmatte> operational & administrative status on ports
[11-Jun-2009 08:59:16] <ckrough> snmp
[11-Jun-2009 09:00:01] <ckrough> its the ethernetCsmacd template
[11-Jun-2009 09:00:24] <ckrough> or more likely ethernetCsmacd_64
[11-Jun-2009 09:01:10] <rmatte> ok, now I've implemented a "operational status" threshold and added ifOperStatus to poll .1.3.6.1.2.1.2.2.1.8, I have the threshold kick off based on the operational status of the port. A whole bunch of ports went down last night on some devices and the thresholds did kick off and notify, but when you went to the OS tab, the interfaces still showed as all green
[11-Jun-2009 09:01:23] <rmatte> any ideas as to why?
[11-Jun-2009 09:02:00] <ckrough> do you use zensyslog
[11-Jun-2009 09:02:13] <rmatte> It's running, yes
[11-Jun-2009 09:02:52] <ckrough> we ended up forwarding LINK UPDOWN messages (IOS) from our syslog servers to zensyslog, then did alerting based on that
[11-Jun-2009 09:03:03] <ckrough> for ports other than the management port on a switch
[11-Jun-2009 09:03:43] <ckrough> not sure why your threshold method didnt work.
[11-Jun-2009 09:05:23] <rmatte> my threshold method did work
[11-Jun-2009 09:05:29] <rmatte> we got alerts from the thresholds
[11-Jun-2009 09:05:42] <rmatte> I'm asking why Zenoss' default monitoring method didn't detect port status change
[11-Jun-2009 09:05:44] <venturaville> yeah ...editing reports in zope actually works in 2.4 now ...
[11-Jun-2009 09:05:44] <rmatte> and still hasn't
[11-Jun-2009 09:05:53] <rmatte> all the ports are showing green even though they are admin down now
[11-Jun-2009 09:06:29] <ckrough> rmatte: let me look at mine. not sure if it goes red for admin down or just oper down
[11-Jun-2009 09:06:39] <rmatte> k
[11-Jun-2009 09:06:56] <rmatte> you'd think it would go red for admin down
[11-Jun-2009 09:07:13] <ckrough> rmatte: check the OS tab of the device in question, there are three columns for status
[11-Jun-2009 09:07:25] <rmatte> correct
[11-Jun-2009 09:07:29] <rmatte> and they are all green
[11-Jun-2009 09:07:30] <ckrough> is it showing A as red for those interfaces, but still showing green in the status page?
[11-Jun-2009 09:07:33] <ckrough> oh
[11-Jun-2009 09:07:39] <rmatte> the third column just means whether or not the port is monitored
[11-Jun-2009 09:07:46] <rmatte> the operational and admin indicators are green
[11-Jun-2009 09:07:51] <ckrough> yeah, that's not right :
[11-Jun-2009 09:07:51] <rmatte> but the port has been admin down for hours
[11-Jun-2009 09:08:19] <ckrough> did you check an snmpget against it just to make sure its reporting correctly
[11-Jun-2009 09:08:52] <rmatte> I'd imagine that I'd use the same value that I'm using for my threshold?
[11-Jun-2009 09:09:13] <ckrough> 1.3.6.1.2.1.2.2.1.7 is ifAdminStatus
[11-Jun-2009 09:09:22] <rocket> rmatte: go to the os tab
[11-Jun-2009 09:09:43] <rocket> actually just go to the device
[11-Jun-2009 09:09:48] <rmatte> done
[11-Jun-2009 09:09:53] <rocket> on the url bar add /manage
[11-Jun-2009 09:10:06] <rmatte> in management console...
[11-Jun-2009 09:10:13] <rocket> just a sec
[11-Jun-2009 09:10:18] <rmatte> k
[11-Jun-2009 09:10:26] <rocket> click os
[11-Jun-2009 09:10:31] <venturaville> <tal:block tal:define="
[11-Jun-2009 09:10:31] <venturaville> objects python:here.getDmdRoot('Devices').getSubDevices(lambda x: x.getSnmpStatus()!=0);
[11-Jun-2009 09:10:31] <venturaville> tableName string: snmpstatusreportwithips;
[11-Jun-2009 09:10:31] <venturaville> batch python:here.ZenTableManager.getBatch(tableName,objects, sortedHeader='getSnmpStatusHeader', sortedSence='desc');
[11-Jun-2009 09:10:31] <venturaville> exportFields python:['device', 'classPath', 'hwProductName', 'productionState', 'pingStatus', 'snmpStatus'];
[11-Jun-2009 09:10:31] <rocket> click interfaces
[11-Jun-2009 09:10:32] <rmatte> k
[11-Jun-2009 09:10:33] <venturaville> ">
[11-Jun-2009 09:10:38] <rmatte> k
[11-Jun-2009 09:10:42] <rocket> choose the one that you have as down
[11-Jun-2009 09:10:47] <rmatte> got it
[11-Jun-2009 09:10:50] <rocket> click properties
[11-Jun-2009 09:10:53] <venturaville> is there a doc somewhere that explains the above....
[11-Jun-2009 09:11:11] <rmatte> I don't see properties listed?
[11-Jun-2009 09:11:41] <rocket> upper right
[11-Jun-2009 09:11:47] <rmatte> ohhh
[11-Jun-2009 09:11:50] <rocket> after you click on the interface
[11-Jun-2009 09:11:51] <rmatte> got it
[11-Jun-2009 09:12:17] <rmatte> adminStatus is set to 1, so is operStatus, even though the port is clearly down
[11-Jun-2009 09:12:18] <rocket> what is adminStatus set to?
[11-Jun-2009 09:12:23] <rocket> ok just a sec
[11-Jun-2009 09:12:26] <rmatte> k
[11-Jun-2009 09:12:38] <rmatte> (this is a 2.3.3 box btw)
[11-Jun-2009 09:12:42] <rmatte> I haven't updated it yet
[11-Jun-2009 09:16:49] <ckrough> looking at one of mine which is oper and admin down, the value is "2" for each. fyi
[11-Jun-2009 09:16:52] <rocket> rmatte: remodel the device
[11-Jun-2009 09:17:30] <rocket> rmatte: I am guessing that variable is only updated when you remodel
[11-Jun-2009 09:17:46] <rocket> rmatte: from my brief glance at the code anyway
[11-Jun-2009 09:18:17] <rocket> rmatte: I believe if I remember right admin down is managed by remodel
[11-Jun-2009 09:18:35] <rocket> rmatte: and actual interface down stuff should be set to catch a trap from the device
[11-Jun-2009 09:19:20] <rocket> rmatte: from a design perspective I wouldnt alert on admin down as that tends to mean an administrator has put the interface down .. etc ..
[11-Jun-2009 09:19:38] <rocket> rmatte: you could verify this with one of the devs on thursday though
[11-Jun-2009 09:20:35] <rmatte> yeh, but remodelling every device each time a port goes down is not a solution
[11-Jun-2009 09:20:55] <rocket> rmatte: I believe you have already applied the solution at the moment
[11-Jun-2009 09:21:10] <rmatte> it is thursday
[11-Jun-2009 09:21:16] <rmatte> I thought devs were today?
[11-Jun-2009 09:21:20] <rmatte> or is that next Thursday?
[11-Jun-2009 09:21:22] <rocket> rmatte: they are later
[11-Jun-2009 09:21:28] <rmatte> yeh, like 11:00
[11-Jun-2009 09:21:28] <rocket> later today
[11-Jun-2009 09:21:32] <rocket> rmatte: lost track of dayss
[11-Jun-2009 09:21:51] <rmatte> definitely going to ask, generally if a port is monitored you want to know what's happening to it regardless
[11-Jun-2009 09:21:57] <rmatte> otherwise you wouldn't be monitoring it
[11-Jun-2009 09:22:26] <rmatte> and the fact that it didn't even pick up on the operational status let alone the admin status is an issue
[11-Jun-2009 09:22:47] <rocket> rmatte: operational status I would agree 100% on
[11-Jun-2009 09:23:31] <rocket> rmatte: that should be obtained off of a datacollector run not a modeler run
[11-Jun-2009 09:23:48] <rocket> technically I think both should but I didnt design this up front
[11-Jun-2009 09:25:00] <rocket> in my own mind I see it as a core design flaw .. (its fixable) but it might be fixed on a core review down the road etc
[11-Jun-2009 09:25:28] <rocket> I think the team needs to look at modelling vs monitoring again and take another iteration at it
[11-Jun-2009 09:26:03] <rocket> I think that in reality modelling should probably go away .. and add the ability to monitor different things at different intervals
[11-Jun-2009 09:26:28] <rocket> and give the user the ability to define the intervals that makes sense for them
[11-Jun-2009 09:27:07] <rocket> but I dont know the whole background of why it was determined to be done this way ..
[11-Jun-2009 09:27:36] <ckrough> rocket: scalability? would be brutal on large installs
[11-Jun-2009 09:27:55] <venturaville> modeling is broken out because it is a really really heft process
[11-Jun-2009 09:28:20] <rocket> venturaville: I agree about that but did you not see the different intervals comment
[11-Jun-2009 09:28:26] <venturaville> yep I did
[11-Jun-2009 09:28:50] <venturaville> I do agree about the interfaces for sure
[11-Jun-2009 09:28:53] <rocket> eg software modelling I totally agree should be done once a week or something like that
[11-Jun-2009 09:29:12] <rocket> network modelling maybe once an hour to pick up admin down changes
[11-Jun-2009 09:29:26] <venturaville> depends on what you mean by network modelling too
[11-Jun-2009 09:29:35] <venturaville> querying a router can take a very long time
[11-Jun-2009 09:29:48] <rocket> yup .. it really does ..
[11-Jun-2009 09:30:02] <rocket> venturaville: depending on the interfaces on the router etc yeah ..
[11-Jun-2009 09:30:27] <rocket> venturaville: thats why I think that the thought was notifications of port status would be caught via a trap
[11-Jun-2009 09:30:54] <venturaville> yep ... which works in a small setup, but it is difficult to map out on a per device what you care about
[11-Jun-2009 09:31:04] <venturaville> when mapping out thousands of devices
[11-Jun-2009 09:31:36] <rocket> venturaville: yeah ...
[11-Jun-2009 09:31:43] <rmatte> I've completely disabled the zenmodel process from running as a daemon on that particular Zenoss install since it's monitoring 342, many of which are in Mexico, and Europe. Modelling some of the devices across slow links takes a ridiculous amount of time, I had to crank the SNMP timeout settings and the collector model time setting up to ridiculously high values just to model some of this stuff
[11-Jun-2009 09:31:52] <venturaville> I would hate to see the event class for that......
[11-Jun-2009 09:31:54] <rmatte> so I sure as heck don't want it modelling at intervals
[11-Jun-2009 09:32:00] <rocket> venturaville: thats why I was thinking about the granularity of being able to decide intervals ..
[11-Jun-2009 09:32:13] <cgibbons> hurm
[11-Jun-2009 09:32:30] <rocket> cgibbons: hey there
[11-Jun-2009 09:33:24] <venturaville> rocket: probably you could get away with some of it by rewriting the interface collector to gather those datapoints and use the range threshold zenpack I saw posted
[11-Jun-2009 09:33:46] <rocket> venturaville: the range threshold I wrote?
[11-Jun-2009 09:33:52] <venturaville> rocket: kudos :-)
[11-Jun-2009 09:34:28] <rmatte> even defining intervals for modelling doesn't resolve this issue, interfaces should be polled for operational and administrative status during each polling cycle
[11-Jun-2009 09:35:06] <ckrough> unless you have tens of thousands of ports
[11-Jun-2009 09:35:40] <rmatte> then you obviously go through and unmonitor anything that you don't want to monitor
[11-Jun-2009 09:36:04] <rmatte> I'm monitoring like 140 switches on this Zenoss box, but I went through and only set it to monitor things like trunk ports and such
[11-Jun-2009 09:36:05] <rocket> rmatte: again I dont disagree on the port issue ..
[11-Jun-2009 09:37:01] <rocket> rmatte: and I would tend to argue that you should be able to define if you want monitoring on or off by default before you model the device the first time etc .. depending on the type of switch it is etc
[11-Jun-2009 09:37:14] <rmatte> yeh
[11-Jun-2009 09:37:34] <rocket> eg if you are modelling a switch connected to workstations .. off by default and then turn on the interfaces you care to be notified about
[11-Jun-2009 09:37:48] <rmatte> yeh, for sure
[11-Jun-2009 09:38:08] <rocket> but if you are modelling a server switch .. set it on .. and turn off monitoring on development server ports ..
[11-Jun-2009 09:38:25] <rmatte> I had to go through and set tons of interfaces not to be monitored since they were all end user ports, or printers, or whatever
[11-Jun-2009 09:38:54] <rocket> rmatte: i can feel that pain ...
[11-Jun-2009 09:39:07] <rmatte> hehe
[11-Jun-2009 09:55:31] <bwest> hello
[11-Jun-2009 09:55:47] <venturaville> does anyone have an example of a zenpack that makes populates some performance tabs through a zenrelation it itself created?
[11-Jun-2009 09:56:35] <rocket> venturaville: I have one that does similar but doesnt create perf graphs .. :/
[11-Jun-2009 09:56:49] <venturaville> the relation I have working
[11-Jun-2009 09:56:58] <venturaville> now just trying to figure out how to populate it...
[11-Jun-2009 09:57:13] <bwest> Is it possible to use winexe as part of a plugin to doel a windows server?
[11-Jun-2009 09:57:50] <bwest> to model windows
[11-Jun-2009 10:01:09] <rocket> bwest: why are you using winexe? what version of zenoss are you using?
[11-Jun-2009 10:01:42] <rocket> bwest: I believe wmi is the new way to go for that .. with newer zenoss installations eg 3.2 and above
[11-Jun-2009 10:03:02] <cgibbons> *golfclap*
[11-Jun-2009 10:03:11] <mrayzenoss> good morning/afternoon/evening everyone, cgibbons and I will be here to answer your questions
[11-Jun-2009 10:03:21] <mrayzenoss> especially cgibbons
[11-Jun-2009 10:03:45] <rmatte> good morning Matt
[11-Jun-2009 10:03:50] <cgibbons> bwest: yeah, rocket's right. it's much better to use the built-in WMI support to do your own modeling plugins if you need.
[11-Jun-2009 10:04:51] <bwest> I am using 2.4.1
[11-Jun-2009 10:05:07] <bwest> I am using winexe to try and pull the serial number off the physical server
[11-Jun-2009 10:06:39] <cgibbons> not a good way to do it, really. winexe isn't a good tool. zenoss ships with wmic as a command-line tool that you can use, plus you can run WMI queries directly from the python code.
[11-Jun-2009 10:08:15] <bwest> So how do I get to the command line when modeling to use wmic
[11-Jun-2009 10:08:22] <rmatte> ok, so my main question for the developers: On the OS tab, port status appears to be determined during modelling. Why is no polling done on monitored interfaces each polling cycle to check status? I'm having to use a threshold to properly monitor administrative status on ports, and it doesn't update the status on the OS tab. Ports stay showing as up when they should be showing as down, and vice versa.
[11-Jun-2009 10:08:45] <cgibbons> for example I just did this here: wmic -U 'zenoss\administrator' //hostname 'select SerialNumber from Win32_SystemEnclosure'
[11-Jun-2009 10:09:27] <cgibbons> matt, isn't there a community zenpack that already does modeling using wmic, or was that just performance data?
[11-Jun-2009 10:09:42] <cgibbons> there's also a way to run WMI queries from the python code, check out zenwin.py for an example
[11-Jun-2009 10:10:13] <mrayzenoss> cgibbons: this one? http://www.zenoss.com/community/projects/zenpacks/windowswmidevicetemplate
[11-Jun-2009 10:10:25] <cgibbons> look at WMIPlugin.py in Products.ZenWin for somethinng you can extend in your own modeling plugins to do WMI queries
[11-Jun-2009 10:10:41] <bwest> ok thx
[11-Jun-2009 10:11:18] <bwest> wasn't sure if i could run the command from the plugin.
[11-Jun-2009 10:11:58] <cgibbons> the way WMIPlugin.py is done you don't have to worry about executing the query yourself, you just override the queries() method with the WMI queries you want it to execute
[11-Jun-2009 10:12:05] <cgibbons> and it'll call back your process method with the results
[11-Jun-2009 10:12:21] <cgibbons> rmatte: good question, and the last time I asked that the answer was there wasn't a good way in the architecture to do that.
[11-Jun-2009 10:13:08] <rmatte> cgibbons: ok, is it something that's being looked at for the future at all?
[11-Jun-2009 10:13:26] <rmatte> it's a fairly basic thing to have
[11-Jun-2009 10:14:07] <cgibbons> it hasn't been escalated to the point where the enterprise customers have asked for it (surprisingly). i'm surprised it doesn't come up more.
[11-Jun-2009 10:14:32] <ckrough> Not very manageable on a large scale.
[11-Jun-2009 10:15:07] <ckrough> ALready tough enough to complete a polling cycling in time with the current OIDS
[11-Jun-2009 10:18:07] <bwest> If I have an environment that has about 350 components it's monitoring, running 2.4.1 on linux what server requirments shold be large enough? The components are 200 servers and 150 switches and routers
[11-Jun-2009 10:18:24] <rmatte> I'm just looking at this from the perspective of Zenoss' campaign to win over people from the other larger monitoring companies/tools, but if you're lacking basic functionality like this, it doesn't really make a good showing.
[11-Jun-2009 10:19:19] <rmatte> bwest: I'm monitoring 342 devices, 53 are servers, the rest are cisco routers and switches, I use 3GHz processor with 4GB of RAM
[11-Jun-2009 10:19:48] <cgibbons> there's a chart we have on teh website with requirements... let me find it
[11-Jun-2009 10:20:15] <cgibbons> yeah: http://www.zenoss.com/product/deployment
[11-Jun-2009 10:20:20] <ckrough> bwest: depents on the number of things you are collecting, how long you keep the rrd data, and how frequently you collect it
[11-Jun-2009 10:21:01] <bwest> they are interested in baselining and they are monitoring this data around the clock
[11-Jun-2009 10:21:51] <bwest> so needless to say they collect alot of data and store this
[11-Jun-2009 10:22:28] <bwest> not to mention they want multi graphs running 24/7
[11-Jun-2009 10:22:57] <mrayzenoss> make sure you're on a fast filesystem
[11-Jun-2009 10:23:01] <cgibbons> mattray, do you have any stats available with some of our really big customers?
[11-Jun-2009 10:23:08] <cgibbons> i know we have some relatively large deployments
[11-Jun-2009 10:23:16] <mrayzenoss> well, the largest community install 1 box is 8500 devices
[11-Jun-2009 10:23:26] <ckrough> 1 box?
[11-Jun-2009 10:23:26] <mrayzenoss> almost all SNMP
[11-Jun-2009 10:23:37] <mrayzenoss> yeah, 8-way 32 gigs of RAM
[11-Jun-2009 10:23:54] <rocket> ckrough: devices isnt really indicitive of the amount of data they are collecting
[11-Jun-2009 10:23:55] <mrayzenoss> someone else is doing 10K
[11-Jun-2009 10:23:59] <mrayzenoss> with 4 boxes
[11-Jun-2009 10:24:03] <ckrough> I can only get 1600 devices on that spec, with 6 15k RAID10 storage config
[11-Jun-2009 10:24:08] <rocket> ckrough: need to look at the # of datapoints collected
[11-Jun-2009 10:24:19] <ckrough> 75k OIDs per box on that spec
[11-Jun-2009 10:25:01] <cgibbons> I suspect a lot of the large upcoming installations will be doing mostly Windows devices using WMI/PerfMon, so that'll change the game some
[11-Jun-2009 10:25:33] <mrayzenoss> if you go to Collectors->localhost->Performance you can see your stats
[11-Jun-2009 10:25:50] <mrayzenoss> assuming 1 collector named localhost
[11-Jun-2009 10:27:05] <bwest> one collector with 10k
[11-Jun-2009 10:27:20] <rmatte> one collector with 10k is insane
[11-Jun-2009 10:27:28] <ckrough> 10k what
[11-Jun-2009 10:27:34] <Jontler> devices
[11-Jun-2009 10:27:38] <rmatte> I assumed he meant devices
[11-Jun-2009 10:27:41] <mrayzenoss> yeah
[11-Jun-2009 10:27:48] <mrayzenoss> but it was lightweight
[11-Jun-2009 10:27:52] <rmatte> what kind of devices though?
[11-Jun-2009 10:27:59] <mrayzenoss> cable modems?
[11-Jun-2009 10:28:14] <bwest> no not devices data points
[11-Jun-2009 10:28:14] <rmatte> oh, that's not really incredibly impressive then
[11-Jun-2009 10:28:19] <rmatte> not much to poll on those things
[11-Jun-2009 10:28:25] <mrayzenoss> right
[11-Jun-2009 10:28:44] <ckrough> disk IO seems to be the killer. anyone runnign this on SSD yet?
[11-Jun-2009 10:29:41] <cgibbons> it scales nicely horizontally for that issue, since data (sans events) is kept on each collector.
[11-Jun-2009 10:29:45] <rmatte> I'm polling like 18.7k datapoints on my largest install so far, but I'm going to have much larger projects coming down the pipe soon
[11-Jun-2009 10:30:27] <bwest> but for 10K datapoints I should just look at the sizing chart?
[11-Jun-2009 10:30:54] <mrayzenoss> here's the bragging thread for Core installs: http://forums.zenoss.com/viewtopic.php?t=6230
[11-Jun-2009 10:31:25] <ckrough> >430k OID on my biggest install
[11-Jun-2009 10:31:29] <ckrough> enterprise tho
[11-Jun-2009 10:31:54] <mrayzenoss> heh, I'm doing about 110 on my home install
[11-Jun-2009 10:32:37] <mrayzenoss> Anyone going to be at SouthEast LinuxFest this weekend?
[11-Jun-2009 10:33:00] <cgibbons> hrmph only 220 on my home setup, which i haven't looked @ in months
[11-Jun-2009 10:33:25] <ckrough> Yeah, I plan to set it up at home to watch my total usage with NetFlix... b/w caps are coming...
[11-Jun-2009 10:33:30] <cgibbons> doh it's only 2.3.1, bad me
[11-Jun-2009 10:33:48] <rocket> cgibbons: heh I dont have as many datapoints .. but a fairly large number of zenpacks installed
[11-Jun-2009 10:33:59] <mrayzenoss> my home install is where the source build on PPC gets tested
[11-Jun-2009 10:34:11] <cgibbons> you still have that powerbook??
[11-Jun-2009 10:34:19] <mistich> does anyone know the status of IPV6 in zenoss
[11-Jun-2009 10:34:35] <mrayzenoss> mistich: it hasn't gotten prioritized very high
[11-Jun-2009 10:34:45] * ckrough plays sad trombone
[11-Jun-2009 10:34:46] <mrayzenoss> mistich: there is a HOWTO somewhere for doing IPV6 ping
[11-Jun-2009 10:35:04] <mrayzenoss> http://www.zenoss.com/community/wiki/user-contributed/IPv6Ping/
[11-Jun-2009 10:35:10] <mrayzenoss> which could be adapted
[11-Jun-2009 10:35:24] <bwest> I am trying to model SIP adapters. SNMP is not available on them, but they can be identified with Mac addresses, can zenoss get this information
[11-Jun-2009 10:35:25] <mistich> yes but what about snmp ssh etc
[11-Jun-2009 10:35:31] <mrayzenoss> I did get some emails from a university in China that said their class project was to port Zenoss to IPV6
[11-Jun-2009 10:35:42] <mrayzenoss> I'm guessing the class failed
[11-Jun-2009 10:35:52] <mrayzenoss> haven't heard back from them in months
[11-Jun-2009 10:35:54] <ckrough> Is there much demand for making reporting more robust? Maybe moving graph definitions to re-usable objects, etc
[11-Jun-2009 10:36:02] <mistich> yes I saw that too has anyone followed up with them
[11-Jun-2009 10:36:06] <mrayzenoss> I did
[11-Jun-2009 10:36:26] <mrayzenoss> I emailed them telling them I would answer any questions and help them any way they needed
[11-Jun-2009 10:36:38] <cgibbons> there's some good porting enhancements coming in the next release actually
[11-Jun-2009 10:36:42] <cgibbons> er reporting
[11-Jun-2009 10:36:43] <mrayzenoss> ckrough: there are already some changes going into King Crab for reporting
[11-Jun-2009 10:36:51] <rmatte> mrayzenoss: probably, I'm sure they found out that it was much more complicated than they initially thought
[11-Jun-2009 10:36:57] <mrayzenoss> rmatte: definitely
[11-Jun-2009 10:37:06] <ckrough> mrayzenoss: I will send you swag if you do
[11-Jun-2009 10:37:09] <mrayzenoss> heh
[11-Jun-2009 10:37:25] <mrayzenoss> well, bang the drum with support
[11-Jun-2009 10:37:35] <mrayzenoss> everyone shows up with their wishlist and it gets prioritized
[11-Jun-2009 10:37:36] <rmatte> you'd almost have to have some sort of layer between Zenoss and the device that takes a dummy ipv4 address and forwards the traffic to an ipv6 address and vice versa
[11-Jun-2009 10:37:39] <rmatte> as a workaround
[11-Jun-2009 10:37:51] <mrayzenoss> yeah, it's non-trivial
[11-Jun-2009 10:37:58] <mistich> next question moving away from rrds to a relational database
[11-Jun-2009 10:38:03] <ckrough> ^^ this
[11-Jun-2009 10:38:12] <sergeymasushko> can someone help me with the graph issue. I believe it should be simple, but I stack on it. http://forums.zenoss.com/viewtopic.php?t=10008
[11-Jun-2009 10:38:33] <rmatte> Matt, any progress on figuring out what the cause of that zenpack load time issue could be?
[11-Jun-2009 10:38:41] <rmatte> (just curious)
[11-Jun-2009 10:39:08] <mrayzenoss> rmatte: jplouis was looking at it, I believe nothing obvious was showing up yet
[11-Jun-2009 10:39:35] <cgibbons> meaning you want to export your raw RRD data into a relational database for custom reporting, mistich, or you want to change the data collection to store in a database directly?
[11-Jun-2009 10:39:36] <rmatte> cool
[11-Jun-2009 10:39:44] <mrayzenoss> mistich: the idea of moving off of RRD does come up, just not sure what we'd move to
[11-Jun-2009 10:40:13] <rmatte> one of our guys here came up with a way to dump an RRD to XML and then pull from there
[11-Jun-2009 10:40:16] <mistich> well for reporting relational database would be nice
[11-Jun-2009 10:40:29] <mistich> yes that is easy
[11-Jun-2009 10:40:35] <bwest> I am sorry i am so full of questions, but is it possible to send mysql queries from zenoss to read the amount of data in a table and create events off this?
[11-Jun-2009 10:40:36] <mistich> but want realtime data
[11-Jun-2009 10:40:38] <jplouis> zenpack slowness has to do with the way the relationships are loaded. The zenpack has a relationship to each packable
[11-Jun-2009 10:40:44] <mrayzenoss> bwest: yes
[11-Jun-2009 10:40:57] <mrayzenoss> mistich: understood
[11-Jun-2009 10:41:10] <cgibbons> I can tell you from experience that dumping that much raw data into a relational database in real-time is not a fun path.
[11-Jun-2009 10:41:18] <cgibbons> depending upon your install size of course.
[11-Jun-2009 10:41:54] <mistich> then once in a relational database reporting becomes very easily can use any package to do reporting
[11-Jun-2009 10:41:55] <mrayzenoss> bwest: have you tried the Core MySQL ZenPack?
[11-Jun-2009 10:42:03] <rmatte> bwest, you'd have to write a script for that and use it as a command based datapoint
[11-Jun-2009 10:42:17] <bwest> perfect
[11-Jun-2009 10:42:19] <mrayzenoss> bwest: I just got another massive set of ZenPacks for doing databases yesterday (MySQL, PostgreSQL, MS SQL, ZeoDB)
[11-Jun-2009 10:42:24] <rmatte> although that MySQL ZenPack might be able to do it, haven't looked at it
[11-Jun-2009 10:42:38] <mistich> oh I understand my install is huge but everyday getting asked for more and more reports
[11-Jun-2009 10:42:41] <mrayzenoss> it provides some stats, it may be good enough
[11-Jun-2009 10:43:03] <rmatte> MS SQL you say?
[11-Jun-2009 10:43:08] <mrayzenoss> rmatte: yeah
[11-Jun-2009 10:43:16] <mrayzenoss> MS SQL Server
[11-Jun-2009 10:43:18] <rmatte> I've been waiting for a good community Zenpack for that
[11-Jun-2009 10:43:28] <mrayzenoss> don't like the existing one?
[11-Jun-2009 10:43:38] <mrayzenoss> etank was looking for someone who had used it yesterday
[11-Jun-2009 10:43:41] <bwest> when will that one be published My SQL
[11-Jun-2009 10:43:58] <mrayzenoss> probably over the weekend
[11-Jun-2009 10:44:05] <mrayzenoss> processing new ZenPacks takes about an hour each
[11-Jun-2009 10:44:10] <mrayzenoss> I got 9 yesterday
[11-Jun-2009 10:44:19] <mrayzenoss> which is a great problem to have
[11-Jun-2009 10:44:22] <bwest> excellent, your contest is working
[11-Jun-2009 10:44:32] <mrayzenoss> yeah, we already have the same number of entries as last year
[11-Jun-2009 10:44:38] <mrayzenoss> and there are 2 more months
[11-Jun-2009 10:44:44] <mrayzenoss> we may have to bump up the prizes
[11-Jun-2009 10:44:54] <mistich> yeah we have about 6 zenpacks one day well publish
[11-Jun-2009 10:45:07] <chudler> I do have at least 4 more to send you, but I have to get time to polish them
[11-Jun-2009 10:45:11] <mistich> well I'll wait till the prizes are bumped
[11-Jun-2009 10:45:54] <mrayzenoss> the current backlog is: PredictiveThreshold, AIX SNMP, Netware, LinuxMonitorAddon, Gentoo, SuSE, Jabber XMMP, RDBMS, MySQL, MS SQL Server, PostgreSQL, ZeoDB
[11-Jun-2009 10:46:16] <mrayzenoss> no need to wait
[11-Jun-2009 10:46:28] <mistich> PredictiveThreshold is this holt winters
[11-Jun-2009 10:46:32] <mrayzenoss> yes
[11-Jun-2009 10:46:44] <mistich> already have that one
[11-Jun-2009 10:46:48] <mrayzenoss> I'm comparing the Enterprise one with the Community one now
[11-Jun-2009 10:47:25] <bwest> Are there any good zenpacks to monitor VOIP in core or enterprise?
[11-Jun-2009 10:47:27] <etank> mrayzenoss: i think that the biggest issue for me was getting the Perfmon ZenPack to work
[11-Jun-2009 10:47:43] <etank> winexe was breaking down and i think that was where the issues are
[11-Jun-2009 10:48:01] <mrayzenoss> bwest: the Asterisk one seems to have some issues, but you could probably adapt any Nagios plugin that already exists
[11-Jun-2009 10:48:05] <cgibbons> i may be biased, but if it were me, if a zenpack uses winexe I'd ditch it and never look bck.
[11-Jun-2009 10:48:06] <sergeymasushko> bwest: what do you want to monitor in the VOIP?
[11-Jun-2009 10:48:30] <sergeymasushko> bwest: I have some work around...
[11-Jun-2009 10:48:42] <bwest> Right now they use Cricket to monitor calls per minute users and the such
[11-Jun-2009 10:49:02] <etank> mrayzenoss: is there a PredictiveThreshold community Pack? or just for Enterprise?
[11-Jun-2009 10:49:13] <mrayzenoss> this will be for Core
[11-Jun-2009 10:49:36] <etank> really? how different would it be from the Enterprise one?
[11-Jun-2009 10:49:50] <mistich> what is the difference between them?
[11-Jun-2009 10:49:54] <mrayzenoss> I just set it up yesterday, they take a few days to generate data
[11-Jun-2009 10:50:08] <mrayzenoss> I've got both of them installed and testing
[11-Jun-2009 10:50:17] <mrayzenoss> my first impression is that they're pretty close
[11-Jun-2009 10:50:18] <sergeymasushko> bwest: what is the voip platform (sip server ?)
[11-Jun-2009 10:50:45] <bwest> yes
[11-Jun-2009 10:50:53] <rmatte> I have another question... when executing a command based script where you need to pass a password value to it, if the command fails the password gets spit out in plaintext in the command failed event. I logged a ticket for this and was told to "See the ZenPack creator for a fix". I've already tried fixing this ZenPack by using a python script to pull the username and password values from the Zope database, but after extensive testing I have determined that
[11-Jun-2009 10:50:59] <mistich> one thing to look for is low data in the kb holtwinters doesn't like it and sends false alerts
[11-Jun-2009 10:51:15] <rmatte> The only way to possibly fix this security issue is to have Zenoss intelligently sensor password values in command failed events.
[11-Jun-2009 10:51:39] <rmatte> Is there any particular reason why security issues like this are having a blind eye turned to them?
[11-Jun-2009 10:52:05] <mrayzenoss> well, I believe the ZenPack in question was one we don't even have access to
[11-Jun-2009 10:52:25] <cgibbons> the reason that was given as the answer rmatte is because there are two ways to deal with it: 1) not use COMMAND datatypes (which is what we mostly do), 2) a command script that needs a password gets it sent via stdin instead of on the command-line
[11-Jun-2009 10:52:25] <mrayzenoss> it's one of James Dastrup's, which we haven't looked at since he hasn't sent it to us for hosting
[11-Jun-2009 10:52:47] <sergeymasushko> bwest: for monitoring sip servers I use a shell command that run SIPP util with the different scenarios... then parse result and make an event if its needed. also I'm monitoring active calls on my cisco AS5300 GWs...
[11-Jun-2009 10:52:49] <rocket> mistich: I wrote the predictive zenpack .. and tried to make it as configurable as possible
[11-Jun-2009 10:53:12] <rocket> mistich: it should also retrofit older rrd data that didnt already have predictive trending applied
[11-Jun-2009 10:53:12] <rmatte> yeh, but it's not specific to that zenpack is my point, this issue pops up with any type of command based poll that needs a password passed to it from Zenoss
[11-Jun-2009 10:53:22] <mistich> rocket: nice
[11-Jun-2009 10:53:24] <rmatte> whether it be a custom built script, or a core Zenpack
[11-Jun-2009 10:53:29] <bwest> serg: sounds interesting, is there a way for you to share more with me
[11-Jun-2009 10:53:29] <cgibbons> there's a mechanism to do it that doesn't require it being sent on the command-line
[11-Jun-2009 10:53:37] <cgibbons> but the zenpack author has to use it
[11-Jun-2009 10:54:07] <rmatte> which would be using a python script to pull the values directly from Zope, correct?
[11-Jun-2009 10:54:11] <mistich> like that it retofits the rrd but does it also remove the holtwinters if you want to remove it
[11-Jun-2009 10:54:11] <rocket> mistich: so if you are getting false alerts my threshold will also allow you to go and reapply the beta and gamma and window etc
[11-Jun-2009 10:54:13] <sergeymasushko> bwest: a bit busy for now... lets talk about a bit later...
[11-Jun-2009 10:54:25] <bwest> ok
[11-Jun-2009 10:54:25] <rocket> mistich: it does not currently remove
[11-Jun-2009 10:54:32] <mistich> can't wait to look at it
[11-Jun-2009 10:54:54] <rocket> mistich: as I cant think of a good way to have zenoss do that on removal ..
[11-Jun-2009 10:55:18] <rocket> and as far as I can tell its not a bad performance hit to leave it enabled
[11-Jun-2009 10:55:23] <rmatte> I've already done that and modified the scripts in 3 different WMI zenpacks to use it, but we had a Zenoss server crash because the retrieval of the password and username values for each poll was way too intensive, if you can suggest a more efficient way of doing it then I'd be happy to modify the Zenpacks in that fashion.
[11-Jun-2009 10:55:47] <mistich> rocket: have run into that before thats why I bring it up
[11-Jun-2009 10:56:05] <chudler> when I need a password, I put it in a file, and read it from there. Not sure how/if that helps you, rmatte
[11-Jun-2009 10:56:15] <rmatte> It's either a little bit of code to sensor the password, or large loss of performance in the scripts
[11-Jun-2009 10:56:17] <rocket> mistich: it might be something that I could make a command line to do
[11-Jun-2009 10:56:38] <rmatte> chudler: doesn't really help when you're applying the templates against multiple devices with different usernames and passwords set for WMI in their zproperties
[11-Jun-2009 10:56:42] <rocket> mistich: eg remove_holt.py rrd_file
[11-Jun-2009 10:57:20] <mistich> have a question for everyone how are you handling spikes in graphs
[11-Jun-2009 10:57:25] <cgibbons> that's certainly why we have a daemon and WinPerf datatype in enterpise, so it can manage all that without having to spawn commands
[11-Jun-2009 10:57:27] <chudler> rmatte: good point. what might the code look like that can censor the password? wouldn't it need to know the password? or just /-p .*? -/ ?
[11-Jun-2009 10:57:43] <cgibbons> IMO, using a command script for monitoring is a short-term quick-fix work-around, and a real ZenPack should do better
[11-Jun-2009 10:57:44] <mistich> min and max I hate to use it because you have to recreate the rrd
[11-Jun-2009 10:57:54] <bwest> Another threshold questons. Right now there is the min max, but if I want ok warning critical so 3 levels how do i accomplish this
[11-Jun-2009 10:58:00] <cgibbons> (i'm also looking in our change history to find out if we hav an example of passing in the password via stdin)
[11-Jun-2009 10:58:21] <rocket> bwest: can you use better grammar your question is hard to understand
[11-Jun-2009 10:58:36] <bwest> Sorry
[11-Jun-2009 10:58:57] <bwest> Using thresholds there is a min and max
[11-Jun-2009 10:59:00] <rmatte> chudler: Zenoss should already know the password since it's passing it to the command, so when the command fails, identify that that is a password value and only display the rest of the command.
[11-Jun-2009 10:59:18] <sergeymasushko> bwest: can you add 2 different minmax thresholds to the same datapoint?
[11-Jun-2009 10:59:21] <bwest> but I want a third level so I can warn as the threshold is approachong max
[11-Jun-2009 10:59:53] <mistich> thats what we do
[11-Jun-2009 11:00:51] <Guest52519> hmm
[11-Jun-2009 11:00:55] <rmatte> cgibbons: thanks, if that can be done then that would be a viable solution
[11-Jun-2009 11:01:17] Guest52519 is now known as euphoria
[11-Jun-2009 11:01:40] <rmatte> as a side note, I'm finding myself writing quite a lot of command based scripts for monitoring all sorts of different stuff, such as the stock windows snmp agent which is a total piece of garbage and likes to move SNMP values around a lot
[11-Jun-2009 11:01:46] <rocket> bwest: there is a new community threshold zenpack
[11-Jun-2009 11:01:54] <rocket> bwest: rangethreshold
[11-Jun-2009 11:01:58] <rocket> eg you could do
[11-Jun-2009 11:02:12] <cgibbons> yeah monitoring Windows with SNMP is problematic even with the nice commerical agents
[11-Jun-2009 11:02:12] <rocket> rangethreshold = warn between 3 and 4
[11-Jun-2009 11:02:21] <jsynack> rmatte: move things around or put them in a table with different iid's ?
[11-Jun-2009 11:02:24] <mrayzenoss> http://www.zenoss.com/community/projects/zenpacks/rangethreshold
[11-Jun-2009 11:02:29] <rmatte> they have the snmp value for the RAM stored below the SNMP values for the physical drives, so the value to poll RAM is different on every server, the CPU OIDs tend to wander up and down, one day the first CPU will be .1, then it'll be .2, so I have to use scripts to deal with that
[11-Jun-2009 11:02:31] <rocket> minmax threshold critical > 4
[11-Jun-2009 11:02:32] <bwest> even better solution
[11-Jun-2009 11:03:50] <rmatte> my scripts are working quite nicely though, also allows me to easily calculate percentages and stuff
[11-Jun-2009 11:04:06] euphoria is now known as myeuphoria
[11-Jun-2009 11:05:15] <Jontler> i have a question about default performance monitoring of the cpu
[11-Jun-2009 11:05:25] <Jontler> right now, there's just one graph by default for the cpu
[11-Jun-2009 11:05:48] <rocket> bwest: make sense?
[11-Jun-2009 11:05:53] <Jontler> i have multi-core machines, though - is that cpu graph a compilation of the cpus, is it the first cpu only, or is it something else?
[11-Jun-2009 11:06:08] <bwest> yes perfect sense
[11-Jun-2009 11:06:10] <rocket> Jontler: from snmp its probably an average
[11-Jun-2009 11:06:16] <rmatte> Jontler: WMI or SNMP?
[11-Jun-2009 11:06:16] <Jontler> ssh
[11-Jun-2009 11:06:25] <rocket> bwest I thing range is <=
[11-Jun-2009 11:06:33] <rmatte> ssh would be a total I believe
[11-Jun-2009 11:06:35] <Jontler> im monitoring using ssh and the zenoss plugins on the machine side
[11-Jun-2009 11:06:41] <rocket> bwest: minmax is >=
[11-Jun-2009 11:06:53] <rocket> bwest: so you might want to have a small difference in place ..
[11-Jun-2009 11:06:56] <rocket> gotta run ..
[11-Jun-2009 11:07:40] <bwest> on one of my windows Servers the SNMP service will not stay on. It Starts and then a pop up comes up that says some sSNMP
[11-Jun-2009 11:07:50] <bwest> on local computer started and stopped
[11-Jun-2009 11:08:27] <bwest> how do I ovveride this so SNMP stays running
[11-Jun-2009 11:08:36] <rmatte> Jontler: easy way to find out, do "cat /dev/urandom > /dev/null", that should eat up all of the CPU on the first core, if it doesn't target the first core try again until it does, then check Zenoss in a bit for the current value, if it's 100% it's just monitoring that one core, if it's 25%ish, then it's monitoring everything
[11-Jun-2009 11:09:19] <Jontler> i will try that rmatte, thanks for the suggestion
[11-Jun-2009 11:09:23] <rmatte> np
[11-Jun-2009 11:09:45] <rmatte> you can use htop (you'll need to install it if you don't already have it) to monitor each core
[11-Jun-2009 11:10:04] <rmatte> htop blows top out of the water anyways, so I highly recommend it
[11-Jun-2009 11:10:10] <Jontler> ive never heard of it
[11-Jun-2009 11:10:13] <Jontler> installing now
[11-Jun-2009 11:10:19] <rmatte> you'll like it
[11-Jun-2009 11:10:21] <etank> htop is da`bomb
[11-Jun-2009 11:10:56] <etank> i leave it running in screen on this box all day long
[11-Jun-2009 11:11:03] <Jontler> oh hotness
[11-Jun-2009 11:11:10] <rmatte> hehe
[11-Jun-2009 11:11:39] <rmatte> screen is also "da`bomb"
[11-Jun-2009 11:11:43] <Jontler> yes it is
[11-Jun-2009 11:11:44] <etank> si
[11-Jun-2009 11:11:50] <mrayzenoss> wow, I didn't know top could be so sexxy
[11-Jun-2009 11:12:00] <etank> i like the screen-profiles stuff that ubuntu has
[11-Jun-2009 11:12:09] <rmatte> you can actually use mouse clicks with htop btw
[11-Jun-2009 11:12:19] <rmatte> (most people don't realize)
[11-Jun-2009 11:12:36] <Jontler> i would never have tried
[11-Jun-2009 11:13:01] <Jontler> nice, sorting by different categories with a mouse click
[11-Jun-2009 11:13:06] <rmatte> yessir
[11-Jun-2009 11:13:12] <rmatte> killing processes with mouse clicks as well
[11-Jun-2009 11:13:37] <Jontler> that's much more intuitive
[11-Jun-2009 11:14:10] <rmatte> but yeh, run that command that I told you to run and you should see the first core kick up to 100
[11-Jun-2009 11:14:16] <rmatte> then just keep an eye on Zenoss
[11-Jun-2009 11:14:21] <Jontler> it's already rolling
[11-Jun-2009 11:14:25] <rmatte> cool
[11-Jun-2009 11:14:31] <Jontler> i didnt realize that would only tank one core
[11-Jun-2009 11:14:37] <bwest> Another quick SNMP Question. on one of my windows servers, the SNMP service will not stay on. I get an message saying the SNMP service on your local machine has started and stopped. This happes when the service has nothing to do. Is there anyway I can fix this?
[11-Jun-2009 11:15:54] <rmatte> bwest: never seen that behaviour before
[11-Jun-2009 11:15:59] <sergeymasushko> heh... so no one can't help me with the graphs?
[11-Jun-2009 11:16:13] <rmatte> Jontler: yeh, it's a single thread process so it just tanks 1
[11-Jun-2009 11:16:57] <mrayzenoss> sergeymasushko: the links to the images are bad
[11-Jun-2009 11:17:09] <bwest> very strange I know, on that server I cannot get snmp to work
[11-Jun-2009 11:18:40] <rmatte> cgibbons: any luck finding that stdin method info?
[11-Jun-2009 11:20:12] <cgibbons> don't see an example yet. looks like we just stopped using command for anything like that.
[11-Jun-2009 11:20:34] <rmatte> ah
[11-Jun-2009 11:20:43] <mistich> well you all have a good day
[11-Jun-2009 11:21:01] <rmatte> I'll hunt around and see if I can find some method of doing it
[11-Jun-2009 11:21:50] <mrayzenoss> BTW, anyone looking for a Zenoss job in the bay area, let me know
[11-Jun-2009 11:22:02] <Jontler> rmatte: so the percentage is scaled from 0 to 200%, almost 200% was idle before, and zero is idle now
[11-Jun-2009 11:22:04] <Jontler> :confused:
[11-Jun-2009 11:22:24] <Jontler> something tells me it's only monitoring 1 core, and i need to tune the graphs a bit
[11-Jun-2009 11:23:08] <ckrough> if its gover over 100% I would think its adding all the cores
[11-Jun-2009 11:23:24] <ckrough> *going over*
[11-Jun-2009 11:23:34] <sergeymasushko> mrayzenoss: I've fixed the links.
[11-Jun-2009 11:23:44] <rmatte> yeh, it might be adding them up
[11-Jun-2009 11:23:48] <rmatte> 2 core or 4 core machine?
[11-Jun-2009 11:23:53] <Jontler> 2
[11-Jun-2009 11:24:03] <Jontler> and my second core - according to htop! - is only 25% in use
[11-Jun-2009 11:24:10] <Jontler> so it shouldnt be showing 0 out of 200% idle
[11-Jun-2009 11:25:01] <rmatte> yeh, then something is definitely messed
[11-Jun-2009 11:25:09] <Jontler> oh wait
[11-Jun-2009 11:25:11] <mrayzenoss> we're going to wrap this up, I'll let cgibbons get back to closing tickets. Keep the ZenPacks coming in and stop by at SouthEast LinuxFest if you're in the neighborhood
[11-Jun-2009 11:25:16] <rmatte> keep firing up commands until you see some action
[11-Jun-2009 11:25:17] <Jontler> something is messed alright, and it's called my eyes
[11-Jun-2009 11:25:25] <rmatte> run 2 or 3 in parallel
[11-Jun-2009 11:25:31] <rmatte> oh, really?
[11-Jun-2009 11:25:33] <Jontler> i zoomed in on the graph
[11-Jun-2009 11:25:34] <rmatte> need glasses do you?
[11-Jun-2009 11:25:41] <rmatte>
[11-Jun-2009 11:25:48] <Jontler> well i was all the way zoomed out and it looked like it was dropped off completely
[11-Jun-2009 11:25:59] <Jontler> zoomed in, it shows roughly 80% still available
[11-Jun-2009 11:26:02] <Jontler> so... yea
[11-Jun-2009 11:26:03] <Jontler> adding cores
[11-Jun-2009 11:26:29] <rmatte> yup
[11-Jun-2009 11:26:59] <Jontler> well, thanks for helping me solve that mystery
[11-Jun-2009 11:27:04] <rmatte> no problem
[11-Jun-2009 11:28:25] <rmatte> these Dev answer sessions are now officially marked in my calendar so no one can bug me to do stuff during them
[11-Jun-2009 11:28:52] <rmatte> Is there any pattern to them? Second Thursday of every month?
[11-Jun-2009 11:29:04] <rmatte> or is it random?
[11-Jun-2009 11:29:06] <mrayzenoss> every 2 weeks, Thursday
[11-Jun-2009 11:29:21] <rmatte> ah, I thought it only happened once a month
[11-Jun-2009 11:29:24] <mrayzenoss> http://www.zenoss.com/community/discuss/irc
[11-Jun-2009 11:29:35] <mrayzenoss> they've been pretty regular for awhile
[11-Jun-2009 11:31:05] <rmatte> What happened between February and March?
[11-Jun-2009 11:31:06] <rmatte> lol
[11-Jun-2009 11:32:02] <mrayzenoss> Spring Break
[11-Jun-2009 11:32:09] <rmatte> ah
[11-Jun-2009 11:32:11] <rmatte> right
[11-Jun-2009 11:32:25] <rmatte> ls
[11-Jun-2009 11:32:29] <rmatte> bah, wrong window
[11-Jun-2009 11:34:51] <ckrough> didnt I see some documentation at one point on how to create a report that has user input fields? Ie: Uptime report where user can specify some filters
[11-Jun-2009 11:35:17] <mrayzenoss> ckrough: like this? http://www.zenoss.com/Members/cluther/creating-methods-for-custom-reports
[11-Jun-2009 11:35:35] <mrayzenoss> King Crab will add something easier
[11-Jun-2009 11:35:48] <mrayzenoss> for making straight Python calls
[11-Jun-2009 11:37:07] <ckrough> thats close. I want to be able to build a report where the PHB can specify criteria like "show me all /Network devices with uptimes under 2hr", yada yada
[11-Jun-2009 11:37:32] <ckrough> I think this is a good start though. I can add the input fields to it
[11-Jun-2009 11:41:01] <rmatte> yeh, now if only I could make total sense of the code used in that custom method
[11-Jun-2009 11:41:02] <rmatte> lol
[11-Jun-2009 11:42:02] <rmatte> the for and if parts are obvious
[11-Jun-2009 11:42:10] <rmatte> just the first part of the code has me confuzzled
[11-Jun-2009 11:42:38] <rmatte> oh, I think I see what it's doing
[11-Jun-2009 11:42:57] <rmatte> it's essentially going through the process list and the first part is the function to join each item to make a list
[11-Jun-2009 11:43:04] <rmatte> so that it shows up as proc1, proc2, proc3, etc...
[11-Jun-2009 11:45:34] <ckrough> ya
[11-Jun-2009 11:46:14] <ckrough> where I usually get lost is in finding the data/objects I need
[11-Jun-2009 11:46:28] <ckrough> like "context.os.ipservices()" how would I know to look there for that data
[11-Jun-2009 11:46:36] <rmatte> yeh
[11-Jun-2009 11:46:45] <ckrough> can we copy Chet's brain?
[11-Jun-2009 11:46:52] <ckrough> you need to open source that dude
[11-Jun-2009 11:47:19] <rmatte> well, there is a relatively simple way to find stuff
[11-Jun-2009 11:47:28] <rmatte> go to any device, put /manage after the device name in the url
[11-Jun-2009 11:47:38] <rmatte> then it's essentially context.whatever.whatever()
[11-Jun-2009 11:48:07] <rmatte> so you could do context.os.winservices()
[11-Jun-2009 11:48:12] <ckrough> also, in zendmd: dev = devices.findDevice('whatever')
[11-Jun-2009 11:48:16] <ckrough> print '\n'.join(dir(dev))
[11-Jun-2009 11:48:25] <rmatte> or context.os.processes()
[11-Jun-2009 11:49:01] <rmatte> yeh
[11-Jun-2009 11:49:07] <ckrough> what does the context class represent
[11-Jun-2009 11:49:27] <rmatte> represents the classes under the device
[11-Jun-2009 11:49:30] <rmatte> the device is the context
[11-Jun-2009 11:49:35] <ckrough> ah
[11-Jun-2009 11:49:51] <ckrough> my problem is not zenoss, its lack of python exp
[11-Jun-2009 11:49:52] <rmatte> so if you navigate to any device then go in to the management interface you'll see the stuff you can reference
[11-Jun-2009 11:50:07] <rmatte> I'm still just learning Python, I'm no expert
[11-Jun-2009 11:50:12] <rmatte> but I do have experience with other languages
[11-Jun-2009 11:50:15] <rmatte> especially bash
[11-Jun-2009 11:50:17] <ckrough> same
[11-Jun-2009 11:50:30] <ckrough> most languages I pick up very quickly. Python is not compatible with my brain
[11-Jun-2009 11:51:21] <rmatte> hehe
[11-Jun-2009 11:51:34] <rmatte> I like the fact that python forces you to properly indent the code
[11-Jun-2009 11:51:42] <rmatte> though I do miss brackets
[11-Jun-2009 11:51:47] <ckrough> yes and yes
[11-Jun-2009 11:53:22] <rmatte> yeesh, I was wondering why my thumb was starting to burn, just noticed it still had mustard on it from my sandwich (I don't use my thumbs when typing)
[11-Jun-2009 11:53:23] <rmatte> lol
[11-Jun-2009 11:53:33] <ckrough> thats some serious mustard
[11-Jun-2009 11:53:41] <rmatte> yeh, it's hot mustard
[11-Jun-2009 11:53:57] <rmatte> one step down from mustard gas in the hazard department
[11-Jun-2009 11:57:05] <rhettlap> that stuff is pretty easy to make especially if you make you're own mustard
[11-Jun-2009 12:08:49] <ckrough> parsing every device in the dmd is painfully slow. is there some way to access a cache or index
[11-Jun-2009 13:27:54] <ckrough> anyone know where I can access a variable passed via URL in a custom method?
[11-Jun-2009 13:28:52] <mrayzenoss> you mean like a Python script attached in Zope?
[11-Jun-2009 13:28:59] <ckrough> yes sir
[11-Jun-2009 13:29:20] <mrayzenoss> request = container.REQUEST
[11-Jun-2009 13:29:37] <ckrough> lol, just found it in /zport/dmd/Reports/sql_query when you said that
[11-Jun-2009 13:29:38] <ckrough> thanks
[11-Jun-2009 13:29:38] <mrayzenoss> x = context.findDevice(request.id)
[11-Jun-2009 13:29:53] <mrayzenoss> ahh, ok
[11-Jun-2009 13:30:05] <mrayzenoss> I was just pulling up my Zenoss/MindTouch mashup for examples
[11-Jun-2009 13:30:17] <ckrough> whats this?
[11-Jun-2009 13:31:14] <mrayzenoss> https://sourceforge.net/projects/zenossdeki/
[11-Jun-2009 13:31:42] <mrayzenoss> it uses custom Python scripts installed on Zenoss to get called by MindTouch
[11-Jun-2009 13:31:52] <mrayzenoss> device.id gets passed around a lot
[11-Jun-2009 13:33:45] <CamargoBP> I'm having problems with displaying the correct units on the left side of the voice mail graph (blue graph) I pull the email server every 30 seconds an usually there are between 6 and 30 emails sent between that time. Any suggestions? http://i43.tinypic.com/2e4eyk8.png
[11-Jun-2009 13:35:11] <rhettlap> make sure the datasource and graph are setup to taken and display 30 second resolution
[11-Jun-2009 13:35:11] <CamargoBP> BTW if you use a Mac you need to get http://conceitedsoftware.com/products/linkinus The absolute BEST irc client I have ever used! Seriously... Thought I would spread the word. They have done an excellent job.
[11-Jun-2009 13:35:33] <rhettlap> but you have to pay for it :\
[11-Jun-2009 13:35:40] <CamargoBP> It's worth every penny
[11-Jun-2009 13:35:44] <rhettlap> i'll just stick w/ bitchx
[11-Jun-2009 13:35:48] <CamargoBP> You can install it on 2 computers
[11-Jun-2009 13:36:00] <mrayzenoss> I'm looking, but Colloquoy works ok for me
[11-Jun-2009 13:36:09] <mrayzenoss> and I was looking at how Adium is gonna do it in 1.4
[11-Jun-2009 13:36:24] <CamargoBP> Yeah I switched from Colloquoy last month to this version
[11-Jun-2009 13:36:33] <rhettlap> yeah colloquoy works ok. i just gave up on it since it was wigging out on me and went back to bitchx
[11-Jun-2009 13:36:47] <rhettlap> or xchat
[11-Jun-2009 13:37:01] <CamargoBP> You should at least try the 30 day trial I bet you shell out the $20 USD for it
[11-Jun-2009 13:37:23] <CamargoBP> Anyway...
[11-Jun-2009 13:37:26] <CamargoBP> Sorry tangent
[11-Jun-2009 13:37:26] <mrayzenoss> I'll try it
[11-Jun-2009 13:37:29] * ckrough heart irssi
[11-Jun-2009 13:37:50] <rhettlap> iirc the rra settings in zenoss are global correct?
[11-Jun-2009 13:38:19] <mrayzenoss> rhettlap: you can customize them on datasources
[11-Jun-2009 13:38:59] <rhettlap> ok. i never got into too much of messing with that stuff in zenoss since i am so comfortable with that stuff in cacti
[11-Jun-2009 13:39:40] <rhettlap> can you change it per template?
[11-Jun-2009 13:39:53] <mrayzenoss> yeah
[11-Jun-2009 13:39:54] <rhettlap> well the datasource is attached to a template
[11-Jun-2009 13:39:56] <rhettlap> so i guess yes
[11-Jun-2009 13:40:58] <rhettlap> wow i didnt know it was 11:40 already. i guess i should go down to the office
[11-Jun-2009 13:41:33] <CamargoBP> So I'm looking at the Graph definition tab, where do I set the 30 resolution?
[11-Jun-2009 13:47:16] <mrayzenoss> DataSource->DataPoints
[11-Jun-2009 13:47:26] <mrayzenoss> Create Cmd
[11-Jun-2009 13:47:33] <mrayzenoss> you'll need to read up on RRD stuff
[11-Jun-2009 13:50:31] <CamargoBP>
Do you know if there is an example Cmd that someone has done for Zenoss, I can't seem to get that command correct. I've read up on how to do it, (the last few days) and the zenoss documentation is very vague. It would be nice if there was a screen shot of an example that actually worked.[11-Jun-2009 13:58:58] <rmatte> mrayzenoss: that might not even be CamargoBP's issue, the snmpwalk is polling a correct value yet when he graphs it Zenoss does something weird and shrinks the value down
[11-Jun-2009 13:59:48] <rmatte> I had him remove the %s from the end of the graph point's format property, but it didn't fix it
[11-Jun-2009 13:59:51] <rmatte> so I'm totally stumped
[11-Jun-2009 14:05:08] <jb> snmptrap is UDP right?
[11-Jun-2009 14:07:00] <rmatte> The default destination port for SNMP traps is UDP port 162.
[11-Jun-2009 14:07:05] <jb> k
[11-Jun-2009 14:08:32] <rhettardo> hello
[11-Jun-2009 14:08:50] <rmatte> hello
[11-Jun-2009 14:12:31] <ckrough> hmm... whenever I loop a dictionary in zope it logs me out
[11-Jun-2009 14:13:04] <ckrough> python 1, ckrough 0. python wins again!
[11-Jun-2009 14:14:57] <rmatte> pythons have sharp teeth y'know
[11-Jun-2009 14:17:48] <rhettardo> thats true but gernally speaking boids and colubrids both have smaller catrlidge like sharp teeth!
[11-Jun-2009 14:18:11] <rmatte> yes, but they don't have their own scripting languages
[11-Jun-2009 14:18:21] <rhettardo> no :\
[11-Jun-2009 14:19:10] <rhettardo> i know more about the reptiles than the language :\
[11-Jun-2009 14:23:40] <venturaville> really confusing to go back and forth between perl/python/ruby in the same day
[11-Jun-2009 14:23:53] <venturaville> I pity the fool who looks at my code
[11-Jun-2009 14:27:00] <ckrough> venturaville: yeah, thats my main problem. perl/py/ruby/php
[11-Jun-2009 14:27:15] <ckrough> builtin methods all start converging
[11-Jun-2009 14:41:08] <cgibbons> holy smokes batman
[11-Jun-2009 14:50:26] <rocket> cgibbons: ?
[11-Jun-2009 15:01:12] <rmatte> great, we just lost a switch somewhere, all our phones are down on the second floor and none of the sales team can receive email on the first floor
[11-Jun-2009 15:01:15] <rmatte> fun fun fun
[11-Jun-2009 15:01:24] <ckrough> weee
[11-Jun-2009 15:01:29] <rocket>
[11-Jun-2009 15:01:44] <ckrough> we had an outage last night and AT&T is still catching up with SMS from zenoss
[11-Jun-2009 15:06:00] <CapriCoRN^80> hi
[11-Jun-2009 15:06:10] <CapriCoRN^80> hi rocket
[11-Jun-2009 15:06:12] <CapriCoRN^80> you around ?
[11-Jun-2009 15:06:17] <rocket> yes
[11-Jun-2009 15:06:30] <CapriCoRN^80> ok
[11-Jun-2009 15:06:42] <CapriCoRN^80> did you check that pic ?
[11-Jun-2009 15:07:06] <rocket> yeah .. but I have gotten busy here at work and not had time to really look at it ..
[11-Jun-2009 15:07:25] <CapriCoRN^80> ok
[11-Jun-2009 15:10:32] <CapriCoRN^80> see u than
[11-Jun-2009 15:10:33] <CapriCoRN^80> bye
[11-Jun-2009 15:29:24] <mloven> anyone know of a way to monitor bandwidth on all possible iterations of interface names without the mising RRD's that accompany putting lots of data sources into a graph?
[11-Jun-2009 15:30:52] <mloven> to clarify: If you were to have a Cisco router with a T1 card, that T1 card could be called "Serial 1/0" or "Serial 2/0" etc., depending on what hardware slot it's in....
[11-Jun-2009 15:32:04] <mloven> I'd like to have a graph that will monitor all of a router's T1's without putting all of the possible OID's into data sources, thereby getting "missing RRD's" when the T1 isn't in, say slot 1...
[11-Jun-2009 15:32:32] <mloven> does that make any sense at all?
[11-Jun-2009 15:35:43] <mrayzenoss> perhaps like this? http://www.zenoss.com/Members/asanabria/howto-monitor-and-add-multiple-data-points-the-easy-way-in-zenoss
[11-Jun-2009 15:36:13] <rmatte> yeh, that's the way to do it
[11-Jun-2009 15:36:30] <rmatte> either that or write your own script to do it... I'll paste an example of a script I wrote...
[11-Jun-2009 15:37:16] <rmatte> http://pastebin.com/m41034f2e
[11-Jun-2009 15:38:55] <rmatte> that script was created because windows CPU OIDs tend to wander, for each CPU that spits out a value, on a 2 CPU box the output would look like "OK|CPU1=76 CPU2=34" on a 4 CPU box it would look like "OK|CPU1=73 CPU2=48 CPU3=37 CPU4=45" and so on
[11-Jun-2009 15:39:08] <rmatte> then you just need to add the data points
[11-Jun-2009 15:40:12] <rmatte> actually I have a better example...
[11-Jun-2009 15:40:39] <rmatte> http://pastebin.com/m6aa3db28
[11-Jun-2009 15:41:05] <rmatte> that one was made because the snmp info for RAM is always a different value based on the number of physical drives in a server
[11-Jun-2009 15:41:20] <rmatte> so that script hunts down the value based on the name of the volume
[11-Jun-2009 15:41:38] <rmatte> it looks for "hrStorageRam", then polls the associated OIDs
[11-Jun-2009 15:42:47] <mloven> Thanks rmatte. So then.... and forgive me if I'm asking stupid questions... you would just create a datasource that points to this script?
[11-Jun-2009 15:43:05] <rmatte> yeh, you create a command datasource
[11-Jun-2009 15:43:26] <rmatte> and remember how I said that the output of the script should look like "OK|CPU1=73 CPU2=48 CPU3=37 CPU4=45"...
[11-Jun-2009 15:43:40] <mloven> yep
[11-Jun-2009 15:43:41] <rmatte> so you'd add CPU1 as the first datapoint, CPU2 as the second, and so on
[11-Jun-2009 15:44:20] <rmatte> that output format is Nagios syntax, but Zenoss understands it
[11-Jun-2009 15:44:36] <rmatte> let me paste you the command as it's seen in Zenoss
[11-Jun-2009 15:45:09] <mloven> excellent. We haven't really had any need to use command data sources up until now so it's still pretty new for me.
[11-Jun-2009 15:45:13] <rmatte> $$ZENHOME/Products/SNMPWindowsPerfMonitor/libexec/wincpu.sh "${dev/manageIp}" "${dev/zSnmpCommunity}"
[11-Jun-2009 15:45:14] <rmatte> $$ZENHOME/Products/SNMPWindowsPerfMonitor/libexec/winmem.sh "${dev/manageIp}" "${dev/zSnmpCommunity}"
[11-Jun-2009 15:45:35] <rmatte> I stick it in a folder in /Products, you can really put the script wherever you want on the hard drive
[11-Jun-2009 15:45:44] <rmatte> as long as the zenoss user has access to the folder
[11-Jun-2009 15:45:58] <mloven> Awesome. Thanks rmatte! That's exactly what I needed!
[11-Jun-2009 15:46:04] <rmatte> np
[11-Jun-2009 15:46:35] <rmatte> also, if snmp can't connect the script should return "Unknown"
[11-Jun-2009 15:46:46] <rmatte> as seen in my examples
[11-Jun-2009 15:47:07] <mloven> perfect, thanks!
[11-Jun-2009 15:56:38] <CamargoBP> Does anyone know how long it takes zenoss to create a rrd file? Is it the first time the data changes on the SNMP pull or does it run a process every x min that checks to see if it needs to create a rrd?
[11-Jun-2009 15:57:40] <rocket> CamargoBP: I know its after a couple of rounds of data has been collected
[11-Jun-2009 15:57:46] <rmatte> It generally does it after 3 successful polling cycles
[11-Jun-2009 15:57:59] <rmatte> so like 15 minutes give or take
[11-Jun-2009 15:58:08] <rocket> eg it might take 15 minutes if a rrd is setup for a 5 min poll cycle
[11-Jun-2009 16:00:16] <CamargoBP> So I deleted a DataPoint and saw that the rrd got deleted, then I created a new DataPoint with the same Name as the one that was deleted. My SNMP pulls every 30 seconds. I've been waiting now for 5 minutes for it to create the rrd. This is normal then?
[11-Jun-2009 16:01:02] <rmatte> I'd give it a bit longer
[11-Jun-2009 16:01:31] <rmatte> snmp pulling every 30 seconds, what, are you monitoring like 3 devices?
[11-Jun-2009 16:14:44] <CamargoBP> Currently 32 devices
[11-Jun-2009 16:15:50] <CamargoBP> So I set the "Default RRD Create Command" however Zenoss still used it's own values when creating the rrd. Are those values cached? Do I need to restart Zenoss or a certain process for the new Default take effect?
[11-Jun-2009 16:18:10] <CamargoBP> zen.zenperfsnmp: total=32 good=32 bad=0 time=1.762472 Not too bad. If it starts to take 20 + seconds I will then change that to every minute
[11-Jun-2009 16:19:37] <CamargoBP> Nope a restart of zenoss didn't take the new default RRD Create Command values. Hmmmm.... (Yes I deleted the rrd's, restarted zenoss and it created new ones but not with the new rra values)
[11-Jun-2009 16:26:07] <CamargoBP> Found the problem I accidently put a space in the command.
[11-Jun-2009 16:37:13] <rmatte> a space in which command?
[11-Jun-2009 16:47:03] <CamargoBP> After all that, when the SNMP reports 0 the rrd reports .01 http://i41.tinypic.com/x6lqgl.png
[11-Jun-2009 16:57:46] <CamargoBP> So I have to multiply my graph point by 30 to get a whole value, why would that be? ie. 30,*
[11-Jun-2009 16:58:19] <CamargoBP> http://i44.tinypic.com/23si64z.png
[11-Jun-2009 17:17:29] <CamargoBP> I shouldn't have to multiply by 30 to get a whole value, why do you think I have to do that?
[11-Jun-2009 17:28:00] <mephux_> hey anyone having issues removing the zenoss user on mac osx
[11-Jun-2009 17:31:03] <mrayzenoss> mephux_: http://dev.zenoss.org/trac/ticket/4575
[11-Jun-2009 17:31:20] <mrayzenoss> supposedly this wasn't an issue anymore, are you seeing it?
[11-Jun-2009 17:36:18] <mephux_> well i tried that..
[11-Jun-2009 17:36:47] <mephux_> the commands listed do not work.. hhmm guess I need to dig deeper.. shame
[11-Jun-2009 17:37:15] <mrayzenoss> sorry
[11-Jun-2009 17:37:20] <mrayzenoss> was this with 2.4.1?
[11-Jun-2009 17:38:42] <mephux_> not sure it was about 4 months back..
[11-Jun-2009 17:39:07] <mephux_> been lazy to fix this issue but its now bothering me.. I will be sure to update the ticket eith my steps
[11-Jun-2009 17:39:10] <mrayzenoss> so I saw the error with 2.3.3, but 2.4.0 didn't recreate it
[11-Jun-2009 17:39:42] <mrayzenoss> so knowing that timeline, you're probably on 2.3.X
[11-Jun-2009 17:40:01] <mrayzenoss> and those were the instructions I used to delete the user and group
[11-Jun-2009 17:40:25] <mephux_> I thank you for your time and information. Have a good rest of the day friend
[11-Jun-2009 17:40:28] <mephux_> late
[11-Jun-2009 17:41:58] <blocky> can anyone tell me why zenoss is appending a trailing / to whatever command i try to run through SSH
[11-Jun-2009 17:42:35] <mrayzenoss> hmm... that's a new one
[11-Jun-2009 17:43:16] <blocky> Executing command
[11-Jun-2009 17:43:16] <blocky> /usr/bin/zenplugin.py/
[11-Jun-2009 17:43:22] <blocky> /bin/sh: /usr/bin/zenplugin.py/: No such file or directory
[11-Jun-2009 17:43:41] <blocky> im pretty stumped
[11-Jun-2009 17:44:34] <blocky> it seems whatever i type in the command template box is treated like a function in python, but then it tries to execute the python script with bash
[11-Jun-2009 17:45:02] <blocky> im supposed to be using Nagios parser, correct?
[11-Jun-2009 17:46:07] <mrayzenoss> I assume so, haven't touched the zenplugins
[11-Jun-2009 17:46:46] <mrayzenoss> they're slightly deprecated in favor of the new SSH stuff
[11-Jun-2009 17:46:48] <blocky> the snmp seems to work quite a bit better, but the servers i need to monitor are behind a firewall and we can't open snmp
[11-Jun-2009 17:46:58] <blocky> er, new SSH?
[11-Jun-2009 17:47:31] <mrayzenoss> 2.4 added better SSH handling, so ZenPlugins are on their way out in favor of all remote SSH calls
[11-Jun-2009 17:47:48] <mrayzenoss> of course, the only solution right now is LinuxMonitor, and it's kinda weak
[11-Jun-2009 17:48:06] <mrayzenoss> but a lot of new ZenPacks are coming in
[11-Jun-2009 17:48:15] <mrayzenoss> and LinuxMonitor already has some patches
[11-Jun-2009 17:48:51] <blocky> would the basic idea still be to call a remote script which outputs pairs of something=value and then parse that?
[11-Jun-2009 17:49:12] <mrayzenoss> yeah, you can definitely keep doing that
[11-Jun-2009 17:49:24] <mrayzenoss> for datasources
[11-Jun-2009 17:49:35] <blocky> if I could get that working I could probaby write the scripts to gather metrics myself
[11-Jun-2009 17:49:36] <mrayzenoss> I meant instead of the zenplugins.py stuff
[11-Jun-2009 17:50:00] <blocky> how do I set up data sources independant of zenplugins.py then
[11-Jun-2009 17:50:48] <mrayzenoss> check out http://www.skills-1st.co.uk/papers/jcurry.html Methods of monitoring processes with Zenoss
[11-Jun-2009 17:51:04] <mrayzenoss> Chapter 5.2 "Zenoss and SSH" covers it
[11-Jun-2009 17:51:19] <mrayzenoss> gotta run
[11-Jun-2009 17:51:31] <blocky> cheers, thanks
[11-Jun-2009 17:54:31] <CamargoBP> So snmpwalk only returns whole numbers: UCD-SNMP-MIB::extOutput.1 = STRING: 4376 why in the world would my max be: 2.19 and current: 1.27 http://i44.tinypic.com/1zggv8.png
[11-Jun-2009 17:55:49] <CamargoBP> How in the world did it say 6 + 6 = 12.19 Shouldn't it be 12 I guess that's the real question.
[11-Jun-2009 17:56:14] <CamargoBP> My example didn't use actual numbers that zenoss reported that was an example
[11-Jun-2009 17:56:20] <rhettardo> what does the datasource say
[11-Jun-2009 17:57:27] <rhettardo> also you'll have to make a new data source if you change the rra's to 60 second. you cant add those datapoints in on a prexisting datasource (as far as i know you couldnt in the 'older' pre 1.3 version)
[11-Jun-2009 17:58:10] <CamargoBP> http://pastebin.com/d1a699c27
[11-Jun-2009 17:58:41] <CamargoBP> Yeah I delete the rrd's if I make a change
[11-Jun-2009 17:58:51] <rhettardo> good
[11-Jun-2009 17:59:52] <rhettardo> i what did you use to get this output
[11-Jun-2009 18:00:13] <rhettardo> forget the i
[11-Jun-2009 18:00:31] <CamargoBP> rrdtool info Postfix\ -Sent_Sent.rrd
[11-Jun-2009 18:01:18] <rhettardo> try 'rrdtool fetch Postfix\ -Sent_Sent.rrd AVERAGE -r 30 -s -10m
[11-Jun-2009 18:01:20] <rhettardo> '
[11-Jun-2009 18:02:18] <CamargoBP> http://pastebin.com/d614d06ac
[11-Jun-2009 18:03:34] <lukecyca> Is there a way to get a list of a host's interfaces via xml-rpc?
[11-Jun-2009 18:12:41] <blocky> is there a way to have two devices with the same ip
[11-Jun-2009 18:12:58] <rmatte> blocky: for what purpose?
[11-Jun-2009 18:13:00] <blocky> i tried creating different hosts in /etc/hosts
[11-Jun-2009 18:13:09] <blocky> because they are different machines, albeit behind a firewall
[11-Jun-2009 18:13:18] <blocky> and they are accessible through different ssh ports
[11-Jun-2009 18:13:50] <rmatte> hmmm, Zenoss checks for other devices with the same IP to avoid duplication, so there's no real easy way to do that
[11-Jun-2009 18:13:57] <blocky> oh
[11-Jun-2009 18:14:07] <blocky> wow
[11-Jun-2009 18:14:16] <blocky> well can i disable that?
[11-Jun-2009 18:15:12] <rmatte> no idea, doubtful, are you just trying to do up/down monitoring?
[11-Jun-2009 18:15:36] <blocky> no im gathering network traffic and performance data
[11-Jun-2009 18:15:39] <rmatte> ah
[11-Jun-2009 18:15:52] <rmatte> Zenoss also uses the default ssh port for monitoring
[11-Jun-2009 18:16:01] <rmatte> so that's another thing you'll have to get around somehow
[11-Jun-2009 18:16:10] <rmatte> I doubt you'll be able to do it by any conventional means
[11-Jun-2009 18:16:32] <rmatte> you might have to write a script to collect the data then feed it in to Zenoss, but that wouldn't be simple
[11-Jun-2009 18:16:56] <blocky> hmm
[11-Jun-2009 18:17:31] <blocky> i cant believe nobody thought this would be used through a NAT
[11-Jun-2009 18:17:55] <rmatte> you could always setup a vpn tunnel through the firewall and monitor through that
[11-Jun-2009 18:18:00] <rmatte> that would probably be the easiest
[11-Jun-2009 18:18:19] <rmatte> it's not designed to be used through a NAT, not many monitoring solutions are other than agent based solutions
[11-Jun-2009 18:18:32] <rmatte> but agentless monitoring is far more desirable in the majority of cases
[11-Jun-2009 18:18:37] <rhettardo> CamargoBP interesting
[11-Jun-2009 18:19:00] <rhettardo> we can see in the datasource its not getting whole numbers
[11-Jun-2009 18:19:01] <blocky> a vpn is not ideal
[11-Jun-2009 18:19:08] <blocky> but i may have found a solution
[11-Jun-2009 18:19:15] <blocky> apparently the duplicate check is in the GUI only
[11-Jun-2009 18:20:43] <rhettardo> CamargoBP is it possibel you're using a counter instead of a gauge?
[11-Jun-2009 18:20:50] <rhettardo> or the other way around?
[11-Jun-2009 18:21:19] <blocky> wow i guess i just missed all the developers this morning eh
[11-Jun-2009 18:21:34] <rhettardo> blocky yeah mray just left
[11-Jun-2009 18:21:57] <blocky> does he idle often, or just come in for the chats
[11-Jun-2009 18:22:22] <rhettardo> he's usually here during work hours
[11-Jun-2009 18:22:24] <rhettardo> or seems that way
[11-Jun-2009 18:22:39] <blocky> its 4:22PM out west
[11-Jun-2009 18:22:50] <rmatte> we use VPN tunnels to connect to all of the clients that we monitor and it works fine
[11-Jun-2009 18:23:08] <rmatte> you get the very occasional flap, but it's 99% stable
[11-Jun-2009 18:23:15] <rhettardo> i use vpn's to monitor everything
[11-Jun-2009 18:23:17] <rhettardo> openssl
[11-Jun-2009 18:23:20] <blocky> what if somebody breaks into one network
[11-Jun-2009 18:23:27] <blocky> they have access to everything?
[11-Jun-2009 18:23:39] <rmatte> nope
[11-Jun-2009 18:23:47] <rmatte> each client gets their own VM on their own VLAN
[11-Jun-2009 18:23:49] <blocky> and all the internal networks need to have exclusive ip ranges?
[11-Jun-2009 18:23:59] <rmatte> and we don't have routing between VMs on our end anyways
[11-Jun-2009 18:24:41] <blocky> i guess its one solution
[11-Jun-2009 18:25:14] <blocky> im sure there must be a way to disable that checking
[11-Jun-2009 18:25:19] <rmatte> it's really the only solution for us
[11-Jun-2009 18:25:55] <rmatte> even if you do disable that checking, you still need to get Zenoss to somehow do all of it's monitoring on a non-standard port for one of the 2 devices
[11-Jun-2009 18:25:59] <CamargoBP> rhettardo: I am using DERIVE
[11-Jun-2009 18:26:05] <rmatte> or maybe for both, depending on how you setup the ports
[11-Jun-2009 18:26:42] <blocky> hmm
[11-Jun-2009 18:26:44] <blocky> this is a real problem
[11-Jun-2009 18:26:56] <blocky> maybe ill check back tomorrow morning and see if any devs show up
[11-Jun-2009 18:26:57] <rmatte> maybe there's some option that you can set in zproperties
[11-Jun-2009 18:27:16] <rmatte> for changing the port I mean
[11-Jun-2009 18:28:52] <rmatte> you could put something between the devices in Zenoss that redirects traffic to whatever port on whatever device based on what phony IP you connect to
[11-Jun-2009 18:28:56] <rmatte> no idea how you'd do that though
[11-Jun-2009 18:29:06] <rmatte> between the devices & zenoss rather
[11-Jun-2009 18:29:32] <rhettardo> CamargoBP i see
[11-Jun-2009 18:29:38] <blocky> yea you can change the port
[11-Jun-2009 18:29:50] <blocky> i think this might work
[11-Jun-2009 18:30:12] <rmatte> I would personally write a script that runs as a cron job on those servers and dumps the info back out to the zenoss box via ssh, then have zenoss grab the data and write a custom template to graph it
[11-Jun-2009 18:32:08] <blocky> i think i figured it out
[11-Jun-2009 18:32:12] <rmatte> oh?
[11-Jun-2009 18:36:45] <blocky> it doesnt seem to check if you use the easy device add
[11-Jun-2009 18:36:52] <rmatte> ah lol
[11-Jun-2009 18:37:14] <rmatte> that's actually pretty funny, making it even easier for people to screw up
[11-Jun-2009 18:37:33] <rmatte> you could have probably done it from the commandline too
[11-Jun-2009 18:37:37] <rmatte> there's probably some option
[11-Jun-2009 18:54:10] <blocky> cool i got cpu load
[11-Jun-2009 18:54:15] <blocky> now i need network traffic
[11-Jun-2009 18:55:15] <blocky> and i need a report emailed to me every morning
[11-Jun-2009 19:44:52] <cgibbons> ah-hah!
[11-Jun-2009 19:46:27] <HKhan> Hmmm how do I make zenoss monitor TCP ports? I added a "service" but it says port number is set to "0", but i can't find where to change it :S
[11-Jun-2009 19:55:14] <HKhan> I went to Classes > Services > Add Service > typed a name for the service, but when i try to edit the service to change the port number, its not there :s
[11-Jun-2009 21:35:19] <cgibbons> h5xm2JQh
[11-Jun-2009 21:35:33] <cgibbons> h5xm2JQH
[11-Jun-2009 21:37:04] <cgibbons> whoops
[11-Jun-2009 21:48:01] <perr0> nice password
[11-Jun-2009 21:48:26] <perr0> far too complex for me thou
[11-Jun-2009 21:56:13] <rhettlap> :O
[11-Jun-2009 22:05:58] <rocket>
[11-Jun-2009 22:30:20] mephux_ is now known as mephux
[12-Jun-2009 00:00:42] [disconnected at Fri Jun 12 00:00:42 2009]
[12-Jun-2009 00:00:42] [connected at Fri Jun 12 00:00:42 2009]
[12-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[12-Jun-2009 03:27:29] tehhobbit is now known as loke
[12-Jun-2009 03:27:41] loke is now known as tehhobbit
[12-Jun-2009 08:17:27] <crow> Whenever I try to .iteritems a dict in zope, it logs me out of zenoss. no errors in the event.log. Anyone ever dealt with that?
[12-Jun-2009 09:04:56] <venturaville> first I've heard of the linuxfest
[12-Jun-2009 09:05:04] <venturaville> apparently it is almost in my backyard
[12-Jun-2009 09:07:36] crow is now known as ckrough
[12-Jun-2009 09:11:59] <jb> kells?
[12-Jun-2009 09:12:22] <rocket> jb?
[12-Jun-2009 09:32:40] <ckrough> meh. doing this: L.sort(key=operator.itemgetter('sysUpTime')) in a script logs me out of zenoss
[12-Jun-2009 09:32:51] <ckrough> same with 'for k,v in d' loops
[12-Jun-2009 09:33:32] <ckrough> ever had that happen?
[12-Jun-2009 09:33:35] <__ian__> that's pretty messed up
[12-Jun-2009 09:33:47] <__ian__> what kind of script?
[12-Jun-2009 09:33:55] <ckrough> python script in zope
[12-Jun-2009 09:34:00] <__ian__> a ZenScriptBase subclass?
[12-Jun-2009 09:34:01] <__ian__> oh
[12-Jun-2009 09:34:21] <ckrough> /zport/dmd/Devices/myscript
[12-Jun-2009 09:36:05] <ckrough> wow. its definitely the action of sorting the list. if I just return the list without sorting it works fine
[12-Jun-2009 09:36:11] * ckrough wishes there was an error message
[12-Jun-2009 09:44:14] <__ian__> What comprises the list? What kind of objects?
[12-Jun-2009 09:45:28] <__ian__> Rather, what composes the list; what does the list comprise. Sorry, I had a bad grammar moment.
[12-Jun-2009 09:46:37] <ckrough> My goal is to sort a 2D dict on the second value of each item: d['item1']['a':1, 'b':2, 'c':3}
[12-Jun-2009 09:46:52] <__ian__> right, but what are the items?
[12-Jun-2009 09:46:52] <ckrough> so there are a bunch of items and I want to return them sorted on 'b'
[12-Jun-2009 09:47:09] <__ian__> if your script doesn't have perms to access one of the items you'll get booted to the login
[12-Jun-2009 09:47:44] <ckrough> hmm. Im building the dict in the script. so the 'item' is just a key and the value is another dict full of attributes
[12-Jun-2009 09:48:08] <__ian__> so this is, at the moment, not at all Zopey?
[12-Jun-2009 09:48:13] <ckrough> Im using the List just to sort it before printing
[12-Jun-2009 09:48:13] <__ian__> other than running in Zope?
[12-Jun-2009 09:48:20] <ckrough> Ill pastebin it, sec
[12-Jun-2009 09:50:02] <ckrough> http://pastebin.com/m6e9b9696
[12-Jun-2009 09:50:15] <ckrough> remove the L.sort line (2nd from bottom) and it works
[12-Jun-2009 09:50:22] <ckrough> leave it there and you get logged out
[12-Jun-2009 09:51:20] <ckrough> its probably a logic error on my part since my python-fu is not to hot
[12-Jun-2009 09:51:29] <ckrough> but, without an error message Im a bit lost
[12-Jun-2009 09:52:13] <ckrough> x.timeframe is hundreths of seconds passed on the url as ?timeframe=n
[12-Jun-2009 09:52:53] <ckrough> Ive tried using a simple custom report in the zenoss gui, but its too slow for what we need. Im hoping I can speed it up once I get the basic layout working here
[12-Jun-2009 09:54:01] <rocket> ckrough: logged out how?
[12-Jun-2009 09:54:19] <rocket> ckrough: is this from the webinterface or cmdline?
[12-Jun-2009 09:54:22] <ckrough> just taken to the zenoss login screen as if you cleared you cache when went to there
[12-Jun-2009 09:54:26] <ckrough> webinterface
[12-Jun-2009 09:54:42] <__ian__> looks like operator.itemgetter is the culprit
[12-Jun-2009 09:54:51] <ckrough> Ive tried doing it as the zenoss admin as well and same error
[12-Jun-2009 09:55:05] <rocket> is this for a report?
[12-Jun-2009 09:55:08] <ckrough> yes
[12-Jun-2009 09:55:24] <ckrough> but I just want text on a blank page, no fancy templating
[12-Jun-2009 09:55:39] <__ian__> if you do this, it works: http://pastebin.com/m17d4e350
[12-Jun-2009 09:56:48] <rocket> yea it looks like the sort function in ckrough's code it the cause of the issue
[12-Jun-2009 09:56:52] <ckrough> nice
[12-Jun-2009 09:56:59] <ckrough> dont blame ckrough
[12-Jun-2009 09:57:08] <rocket> heh
[12-Jun-2009 09:57:10] <ckrough> am I doing something invalid, or is zope choking
[12-Jun-2009 09:57:21] <rocket> I would guess its memory starvation
[12-Jun-2009 09:57:44] <rocket> take a look at top on the server when running your code
[12-Jun-2009 09:57:47] <ckrough> would have to be an artificial limit in zope
[12-Jun-2009 09:58:00] <rocket> yes a timeout I am sure in zope
[12-Jun-2009 09:58:01] <ckrough> 32GB ram on that box and Im only sorting < 20 devices
[12-Jun-2009 09:58:06] <rocket> its taking too long to sort
[12-Jun-2009 09:58:11] <ckrough> ah
[12-Jun-2009 09:58:15] <rocket> its probably in some sort of infinite loop
[12-Jun-2009 09:58:27] <ckrough> lol. 'sort' of infinite loop
[12-Jun-2009 09:58:31] <ckrough> punny guy
[12-Jun-2009 09:58:36] <rocket>
[12-Jun-2009 09:58:52] <rocket> where as the code pasted by __ian__ creates a new compare function
[12-Jun-2009 09:58:58] <rocket> that is very specific
[12-Jun-2009 09:59:11] <rocket> is a sysuptime greater than b sysuptime
[12-Jun-2009 09:59:17] <ckrough> that will work for this report
[12-Jun-2009 10:00:04] <rocket> ah I see I think your code is always returning true to sort
[12-Jun-2009 10:00:30] <rocket> key=operator.itemgetter('sysUpTime') probably always looks like a true
[12-Jun-2009 10:00:52] <rocket> so it looks like L.sort(true) which goes on forever etc ..
[12-Jun-2009 10:00:56] <ckrough> sorry, I dont understand itemgetter well enough to know how that messes it up
[12-Jun-2009 10:01:50] <rocket> I think its the setting something to = that returns true .. its saying it succeeded in setting key = to something else
[12-Jun-2009 10:02:34] <__ian__> hmm, it should work. operator.itemgetter('mykey')({mykey:x}) == x
[12-Jun-2009 10:04:29] <__ian__> I'm leaning towards some probably arbitrary restriction in Zope's execution of Script objects.
[12-Jun-2009 10:04:41] <__ian__> They really like to do that.
[12-Jun-2009 10:04:50] <rocket> __ian__: thats a definite possibility ..
[12-Jun-2009 10:05:16] <ckrough> that makes me feel better
[12-Jun-2009 10:06:00] <venturaville> shouldn't that just be L.sort(key='sysUpTime')?
[12-Jun-2009 10:07:06] <__ian__> no. Confusingly, you pass a callable as key.
[12-Jun-2009 10:07:11] <venturaville> no never mind I see now
[12-Jun-2009 10:07:16] <venturaville> wow that is ugly
[12-Jun-2009 10:07:29] <__ian__> It is the "sort key", rather than the "dictionary key"
[12-Jun-2009 10:07:47] <__ian__> makes it more versatile, but I wish they'd named it something else.
[12-Jun-2009 10:07:47] <kisielk_home> wow, last night our Zenoss autodiscovery went totally amiss
[12-Jun-2009 10:08:05] <ckrough> kisielk_home: another faithful zenner monitoring the internet?
[12-Jun-2009 10:08:14] <kisielk_home> it discovered a VPN interface on some user's workstation, and then went on to probe a local university's entire public IP space
[12-Jun-2009 10:08:36] <kisielk_home> now I have like 1700+ devices in /Devices/Discovered that aren't even ours
[12-Jun-2009 10:08:45] <ckrough> kisielk_home: you can limit the ip ranges it will walk
[12-Jun-2009 10:08:57] <kisielk_home> yes, and somehow that limit disappeared
[12-Jun-2009 10:09:01] <kisielk_home> I think during the upgrade from 2.3.3 to 2.4
[12-Jun-2009 10:09:05] <ckrough> kisielk_home: or you could bill the university for monitoring services
[12-Jun-2009 10:09:10] <kisielk_home> because I'm fairly certain I had 10/8 in there
[12-Jun-2009 10:09:29] <ckrough> are you a single zen box or distributed collectors?
[12-Jun-2009 10:09:36] <kisielk_home> single
[12-Jun-2009 10:09:40] <kisielk_home> just using Zenoss core
[12-Jun-2009 10:10:41] <kisielk_home> I realized something was wrong when I suddenly received 7000+ emails from Zenoss
[12-Jun-2009 10:11:34] <__ian__> kisielk_home: did just your setting of the limit disappear, or the ability to set a limit?
[12-Jun-2009 10:11:52] <__ian__> because if it's the latter I'm thinking we should probably fix that
[12-Jun-2009 10:12:09] <ckrough> you can still set it, we still do
[12-Jun-2009 10:12:14] <__ian__> phew.
[12-Jun-2009 10:12:14] <ckrough> let me see where we set it
[12-Jun-2009 10:12:39] <kisielk_home> __ian__: the setting of the limit
[12-Jun-2009 10:12:51] <kisielk_home> I looked at the field and it was blank
[12-Jun-2009 10:12:57] <kisielk_home> then I cried for a while
[12-Jun-2009 10:13:24] <ckrough> in /opt/zenoss/bin/zenoss_discovery : ${ZENHOME}/bin/zendisc run --chunk 64 --net=10.5.0.0/16 ...additional_args...
[12-Jun-2009 10:13:41] <ckrough> not sure if thats 'the way' or just 'our way'. mutliple admins on these boxes
[12-Jun-2009 10:13:52] <ckrough> we run multiple discovery processes via cron
[12-Jun-2009 10:14:21] <kisielk_home> there's a field in the collector preferences
[12-Jun-2009 10:14:23] <__ian__> that's how we do it on our canary box too
[12-Jun-2009 10:14:51] <ckrough> if I remember correctly we had to move discovery back to the master box, running it on collectors was gapping graphs
[12-Jun-2009 10:15:42] <ckrough> running 2.2.4 so your mileage probably will vary
[12-Jun-2009 10:21:47] <capricorn-80> hi rocket
[12-Jun-2009 10:22:08] <rocket> Hello capricorn-80
[12-Jun-2009 10:22:22] <capricorn-80> how are you ?
[12-Jun-2009 10:22:28] <rocket> busy
[12-Jun-2009 10:22:42] <rocket> budgets time at work ..
[12-Jun-2009 10:22:52] <rocket> and you?
[12-Jun-2009 10:24:55] <capricorn-80> same here
[12-Jun-2009 10:25:04] <capricorn-80> i guess u didnt find time
[12-Jun-2009 10:25:25] <rocket> no sorry .. I would probably ask again next thursday
[12-Jun-2009 10:25:34] <rocket> thursdays the developers are usually here
[12-Jun-2009 10:25:37] <rocket> from zenoss
[12-Jun-2009 10:25:57] <rocket> unless someone like __ian__ here can help you ..
[12-Jun-2009 10:26:04] <__ian__> I can take a stab
[12-Jun-2009 10:26:26] <rocket> __ian__: he is having trouble with one interface not showing up inside zenoss
[12-Jun-2009 10:26:49] <capricorn-80> rocket: you got that pic ?
[12-Jun-2009 10:26:51] <rocket> it looks like it shows up in the modelling debug output but the web ui doesnt show it
[12-Jun-2009 10:27:00] <rocket> capricorn-80: not anymore .. forgot to save it
[12-Jun-2009 10:27:47] <rocket> __ian__: this is on a cisco router with lots of vlans ..
[12-Jun-2009 10:28:21] <__ian__> ha, and you've officially passed my "knowledge of how to actually use the code I write" threshold
[12-Jun-2009 10:28:39] <rocket> __ian__:
[12-Jun-2009 10:29:14] <__ian__> Is it just the web UI, or does it just not exist in the database?
[12-Jun-2009 10:29:28] <rocket> __ian__: I hadnt debugged that far with him ..
[12-Jun-2009 10:30:48] <ckrough> what type of interface is it?
[12-Jun-2009 10:31:01] <capricorn-80> ckrough: its router interface
[12-Jun-2009 10:31:13] <ckrough> I mean gige, te, vlan, po, etc...
[12-Jun-2009 10:31:13] <__ian__> capricorn-80: fire up zendmd and see if your interface is in the output of: find('myrouterid').os.interfaces()
[12-Jun-2009 10:34:41] <capricorn-80> __ian__: can you get the full command because zendmd is giving me some command menu
[12-Jun-2009 10:34:59] <__ian__> $ zendmd
[12-Jun-2009 10:35:07] <__ian__> >>> find('myrouterid').os.interfaces()
[12-Jun-2009 10:38:31] <capricorn-80> __ian__: http://pastebin.com/d384ba6c3
[12-Jun-2009 10:39:24] <rocket> capricorn-80: is this the router? or your zenoss machineid again?
[12-Jun-2009 10:41:22] <capricorn-80> oh sorry
[12-Jun-2009 10:41:29] <capricorn-80> rocket: sorry
[12-Jun-2009 10:41:31] <__ian__> capricorn-80: what rocket said, and also, what's the interface you're looking for?
[12-Jun-2009 10:41:34] <CamargoBP> My graph is not reporting whole numbers, however SNMP only pulls whole numbers. I am graphing MAX not AVERAGE. Would having a step value of 30 cause this? Here is the graph, the blue one http://i39.tinypic.com/rbgnlv.png
[12-Jun-2009 10:41:35] <capricorn-80> i wrote wrong ip
[12-Jun-2009 10:43:25] <capricorn-80> __ian__: http://pastebin.com/m6422f946
[12-Jun-2009 10:43:28] <capricorn-80> rocket: http://pastebin.com/m6422f946
[12-Jun-2009 10:43:44] <__ian__> so, is your interface in that list?
[12-Jun-2009 10:45:52] <capricorn-80> no
[12-Jun-2009 10:46:12] <capricorn-80> it is
[12-Jun-2009 10:46:15] <capricorn-80> IpInterface at /zport/dmd/Devices/Network/Router/Cisco/devices/10.254.95.254/os/interfaces
[12-Jun-2009 10:46:26] <capricorn-80> oh its null0
[12-Jun-2009 10:46:36] <__ian__> so it's there, the second one?
[12-Jun-2009 10:47:15] <capricorn-80> __ian__: yes
[12-Jun-2009 10:47:27] <capricorn-80> __ian__: but i cant see any trafic graphs on it
[12-Jun-2009 10:47:47] <capricorn-80> i mean the traffic is 0 on this interface
[12-Jun-2009 10:48:12] <capricorn-80> no inbound of outbound traffic
[12-Jun-2009 10:51:53] <capricorn-80> [<IpInterface at /zport/dmd/Devices/Network/Router/Cisco/devices/10.254.95.254/os/interfaces/EOBC0_6> and its IP Addresses 127.0.0.22/8
[12-Jun-2009 10:52:14] <__ian__> wait, so you can see it in the UI?
[12-Jun-2009 10:52:54] <capricorn-80> __ian__: can you tell me which i can see ?
[12-Jun-2009 10:53:10] <capricorn-80> yes i can see that
[12-Jun-2009 10:53:18] <__ian__> On the OS tab of the router, the inteface is listed?
[12-Jun-2009 10:53:24] <capricorn-80> its not showing the IP address but Null0
[12-Jun-2009 10:53:29] <capricorn-80> yes
[12-Jun-2009 10:54:22] <capricorn-80> 10.254.95.254 this is the interface as this is interface i am using for polling . i mean snmp is enabled on this interface
[12-Jun-2009 10:55:09] <__ian__> Then I am totally misunderstanding the problem. What's the issue? It isn't collecting, or something?
[12-Jun-2009 10:59:06] <capricorn-80> wait i upload the image
[12-Jun-2009 10:59:17] <capricorn-80> that will give you better idea
[12-Jun-2009 11:02:55] <capricorn-80> __ian__: http://imagebin.org/52246
[12-Jun-2009 11:03:37] <ckrough> just catching up, you are referring to the EOBC interface?
[12-Jun-2009 11:06:31] <capricorn-80> there are two issues. one is EOBC and other was Null0
[12-Jun-2009 11:06:44] <ckrough> neither will be recognized by the default templates
[12-Jun-2009 11:06:45] <capricorn-80> Null0 was issue
[12-Jun-2009 11:07:17] <__ian__> why would null0 ever have any information associated with it?
[12-Jun-2009 11:07:26] <capricorn-80> but with ian command i got it that Null0 is the router interface to which the zenoss is get information
[12-Jun-2009 11:07:36] <ckrough> not sure if they've added them or not, but they get dumped in propVirtual template
[12-Jun-2009 11:08:04] <capricorn-80> ckrough: i just added my main interface ip and that is 10.254.95.254
[12-Jun-2009 11:08:19] <capricorn-80> i didnt add another interface
[12-Jun-2009 11:11:06] <capricorn-80> Null0 is the interface which got this ip address 10.254.95.254
[12-Jun-2009 11:12:11] <capricorn-80> EOBC interface got ip address 127.0.0.22/8 which is strange
[12-Jun-2009 11:14:41] <ckrough> If I understand it correctly, which I may now, EOBC is an out of band interface between the ASIC and the supervisor, but not an interface you would poll
[12-Jun-2009 11:14:45] <ckrough> may *not*
[12-Jun-2009 11:15:24] <capricorn-80> ckrough: yes
[12-Jun-2009 11:21:53] <capricorn-80> i cant figure about this interface
[12-Jun-2009 11:25:41] <ckrough> Can you restate the problem?
[12-Jun-2009 11:33:58] <lukecyca> omg. My Zenoss is auto-discovering networks that aren't mine. Someone's IDS is lighting up like a christmas tree right now... :/
[12-Jun-2009 11:34:01] <eichenwald> where is WinUser / WinPassword stored in the zProperties of a device class
[12-Jun-2009 11:34:10] <eichenwald> is it not stored in the db?
[12-Jun-2009 11:34:22] <__ian__> it is stored in the db
[12-Jun-2009 11:34:35] <__ian__> as attribute zWinUser and zWinPassword, on the device class
[12-Jun-2009 11:36:26] <eichenwald> all I am seeming is an events db
[12-Jun-2009 11:38:13] <capricorn-80> ckrought: wait
[12-Jun-2009 11:38:18] <ckrough> lukecyca: did this just start happenning and did you recent upgrade
[12-Jun-2009 11:38:26] <ckrough> capricorn-80: Ill be here a while
[12-Jun-2009 11:38:53] <capricorn-80> ckrought: IP Addresses 127.0.0.22/8
[12-Jun-2009 11:38:58] <capricorn-80> why this ip address ?
[12-Jun-2009 11:39:04] <capricorn-80> i never saw this kind of ip address
[12-Jun-2009 11:40:02] <lukecyca> ckrough: It's my own fault. One of our computers had a VPN connection open to someone else's network. Zenoss modeled that and it showed up in the list of networks. I ran auto-discovery on some of our networks, and included that one by mistake.
[12-Jun-2009 11:40:17] <lukecyca> 4000 hosts added before I realized what was going on...
[12-Jun-2009 11:40:54] <ckrough> capricorn-80: all of my cisco routers have similar addresses for EOBC as well
[12-Jun-2009 11:41:05] <lukecyca> I'll delete them through zendmd easily enough... but my whole zenoss install is crawling. I guess it's probably trying to model them all of something
[12-Jun-2009 11:41:07] <ckrough> let me ask someone, sec
[12-Jun-2009 11:41:32] <ckrough> lukecyca: you could kill the discovery process
[12-Jun-2009 11:44:13] <lukecyca> Done. I think I've got a handle on it now. This is pretty embarrassing though!
[12-Jun-2009 11:44:45] <ckrough> capricorn-80: ok, the network 127.0.0.0/8 is used for all of the internal communications on the backplane in the catalyst series. The actual IP (22) is possibly dictated by the slot the super is in. There are probably no scenarios where you will care about monitoring it.
[12-Jun-2009 11:45:11] <ckrough> capricorn-80: its the 'internal' management interface for the chassis
[12-Jun-2009 11:45:43] <ckrough> lukecyca: someone had that same issue a couple hours ago. it happens to everyone eventually
[12-Jun-2009 11:46:31] <lukecyca> heh ok thanks. That's good to hear I guess
[12-Jun-2009 11:49:20] <capricorn-80> ckrought: ok
[12-Jun-2009 11:50:09] <capricorn-80> one more thing i figure out is that my main ip interface ip address is 10.254.95.254 but there is no inbound and outbound traffic on it
[12-Jun-2009 11:50:27] <capricorn-80> because the traffic is on vlans ip addresses right ?
[12-Jun-2009 11:50:32] <ckrough> thats your management interface?
[12-Jun-2009 11:51:01] <ckrough> the octets still get reported on the interfaces, regardless of vlan
[12-Jun-2009 11:51:16] <ckrough> when you do an snmpget against that interface does it show traffic?
[12-Jun-2009 11:51:53] <capricorn-80> wait let me check it again
[12-Jun-2009 11:54:39] <capricorn-80> ckrought can tell the switch use with snmpget ?
[12-Jun-2009 11:56:04] <eichenwald> so I made a backup and I notice the large backup file is .fs, so is this what the devices are stored in instead of the mysql db? and how do I look through it
[12-Jun-2009 11:59:11] <ckrough> capricorn-80: sec
[12-Jun-2009 12:00:57] <capricorn-80> snmpget -c communitystring -v 1 10.254.95.254
[12-Jun-2009 12:01:09] <capricorn-80> but its asking me for more switches
[12-Jun-2009 12:01:32] <rhettlap> you need the oid
[12-Jun-2009 12:01:33] <ckrough> you'll need to walk it first and figure out the snmp index of your management interface
[12-Jun-2009 12:01:42] <ckrough> then you can specify an OID for that index and get the traffic
[12-Jun-2009 12:02:03] <ckrough> there may be a better way to do the index thing, but I dont spend enough time on snmpget
[12-Jun-2009 12:02:15] <capricorn-80> ok
[12-Jun-2009 12:38:29] <capricorn-80> ckrought: no traffic on 10.254.95.254
[12-Jun-2009 12:38:57] <capricorn-80> ckrought: its not showing me the interface
[12-Jun-2009 12:39:27] <capricorn-80> ckrought: http://pastebin.com/m4848239e
[12-Jun-2009 12:41:03] <ckrough> capricorn-80: you are specifically looking for the amount of management traffic?
[12-Jun-2009 12:42:13] <capricorn-80> ckrought: i am interested in in/out traffic on interface which got ip 10.254.95.254
[12-Jun-2009 13:06:38] <FooDog> i'm having exactly this problem: http://forums.zenoss.com/viewtopic.php?p=35840 ... any ideas on troubleshooting?
[12-Jun-2009 13:29:23] willwh1 is now known as willwh
[12-Jun-2009 13:30:45] <cgibbons> hmm
[12-Jun-2009 13:45:05] <capricorn-80> I just added my windows server to zenoss its giving some errors in graph. the only graph i can see is CPU and its written Missing RRD Tool for 10.254..3.30 which is machine ip address
[12-Jun-2009 13:46:30] <capricorn-80> snmp-informant
[12-Jun-2009 13:50:04] <capricorn-80> anyone there ?
[12-Jun-2009 13:50:12] <capricorn-80> need some info about snmp informant
[12-Jun-2009 13:57:09] <capricorn-80> i m installing it on one of my running server
[12-Jun-2009 13:57:16] <capricorn-80> i hope it will not create any problem
[12-Jun-2009 13:57:24] <capricorn-80> please need you wording on it
[12-Jun-2009 14:05:38] <capricorn-80> i m getting the output from snmpwalk -v 1 -c communitystring 10.254.66.30 1.3.6.1.4.1.9600
[12-Jun-2009 14:05:52] <capricorn-80> but still rrd tool error in cpu graph
[12-Jun-2009 14:06:18] <__ian__> capricorn-80: how long has it been since you added the device with the missing rrd file?
[12-Jun-2009 14:08:03] <capricorn-80> 10 mins
[12-Jun-2009 14:08:08] <__ian__> It's a confusing error, but that's perfectly normal and expected. There simply aren't data yet. Wait half an hour or so and your graph will show up, once some data has been collected against the device.
[12-Jun-2009 14:08:28] <__ian__> There's an outstanding ticket to make that message say something like, "Wait half an hour."
[12-Jun-2009 14:08:28] <capricorn-80> i have installed snmp informant
[12-Jun-2009 14:08:43] <capricorn-80> and i can get all values using snmp
[12-Jun-2009 14:08:55] <capricorn-80> ok but its saying error
[12-Jun-2009 14:09:07] <__ian__> what error, exactly
[12-Jun-2009 14:10:05] <capricorn-80> Missing RRD File . 10.254.66.30 cpuPercentProcessorTime_cpuPercentProcessorTime
[12-Jun-2009 14:10:09] <__ian__> Right.
[12-Jun-2009 14:10:28] <__ian__> The RRD file is missing because it hasn't been created yet, because the device hasn't been around long enough to be collected against.
[12-Jun-2009 14:10:45] <capricorn-80> ok
[12-Jun-2009 14:10:56] <__ian__> Like I said, a confusing error message, but that always happens.
[12-Jun-2009 14:11:21] <capricorn-80> ok
[12-Jun-2009 14:11:29] <capricorn-80> but how much time it will take ?
[12-Jun-2009 14:11:36] <__ian__> like half an hour or so
[12-Jun-2009 14:11:43] <__ian__> you can cycle the daemons by hand if you want
[12-Jun-2009 14:12:37] <__ian__> but if you just give it a few minutes they'll all pick up the device on their own.
[12-Jun-2009 14:13:04] <capricorn-80> how about 10 mins ?
[12-Jun-2009 14:13:17] <__ian__> half an hour.
[12-Jun-2009 14:13:20] <__ian__> depends on your cycle time
[12-Jun-2009 14:13:56] <capricorn-80> ok . let me wait n see
[12-Jun-2009 14:14:29] <capricorn-80> and i also asked about snmp informant that i am running it on my production server so i hope it will not create some problem
[12-Jun-2009 14:14:38] <__ian__> beats me, but I doubt it.
[12-Jun-2009 14:15:04] <capricorn-80> ah
[12-Jun-2009 14:15:18] <capricorn-80> can you explain it ?
[12-Jun-2009 14:15:31] <__ian__> I don't know what the antecedent of "it" is
[12-Jun-2009 14:16:09] <capricorn-80> ok
[12-Jun-2009 14:16:15] <capricorn-80> lets hope for the best
[12-Jun-2009 14:16:36] <__ian__> but if it's SNMP Informant, no, I know very little about Windows.
[12-Jun-2009 14:17:10] <__ian__> corner cgibbons or cluther or kells when they show their faces, they might be able to tell you more.
[12-Jun-2009 14:17:34] <cgibbons> I probably know the most about Windows @ Zenoss but even I have very rarely touched SNMP Informant. But we recommended it as an agent solution to get performance data out of Windows systems using SNMP since the built-in agent from Microsoft doesn't do it well.
[12-Jun-2009 14:17:45] <cgibbons> but since then we always recommend the WMI/Perfmon based solution, but that's in enterprise only.
[12-Jun-2009 14:18:03] <capricorn-80> ahan
[12-Jun-2009 14:18:03] <cgibbons> there are a few community zenpacks that gather performance data via WMI as well, so you wouldn't need SNMP Informant.
[12-Jun-2009 14:18:31] <capricorn-80> can you tell me some good docs or site ?
[12-Jun-2009 14:19:06] <cgibbons> SNMP Informant is its own product, so their own site has decent docs, etc.
[12-Jun-2009 14:19:55] <capricorn-80> no i am talking about zenpacks performance data via wmi
[12-Jun-2009 14:20:37] <__ian__> http://www.zenoss.com/community/projects/zenpacks/
[12-Jun-2009 14:20:50] <cgibbons> yep, this one specifically: http://www.zenoss.com/community/projects/zenpacks/windowswmidevicetemplate
[12-Jun-2009 14:27:17] <capricorn-80> ok thank s
[12-Jun-2009 14:43:58] <capricorn-80> @cgibbons: i cant able to monitor my main interface of the router
[12-Jun-2009 14:45:18] <capricorn-80> anyways related to windows problem
[12-Jun-2009 14:45:31] <capricorn-80> its been almost 30 mins but still that RRDTool error
[12-Jun-2009 14:53:43] <cgibbons> try this: zenperfsnmp run -v10 -d yourdevicenamehere
[12-Jun-2009 14:58:45] <capricorn-80> @cgibbons: http://pastebin.com/m27dc1177
[12-Jun-2009 14:59:48] <cgibbons> yeah it's not collecting anything for the device. what happens if you go to the Event tab for that device, what's in the list?
[12-Jun-2009 15:00:50] <capricorn-80> @cgibbons: http://pastebin.com/m35e2af6c
[12-Jun-2009 15:03:34] <cgibbons> that's the modeling screen, what about Events? do you have any Events listed on that device?
[12-Jun-2009 15:05:45] <capricorn-80> SNMP agent down
[12-Jun-2009 15:05:47] <capricorn-80> but its working
[12-Jun-2009 15:05:55] <capricorn-80> i checked that using snmpwalk
[12-Jun-2009 15:06:11] <capricorn-80> i can see the hard drives
[12-Jun-2009 15:06:17] <capricorn-80> and other informations
[12-Jun-2009 15:07:03] <capricorn-80> snmp trap snmp_authenticationFailure
[12-Jun-2009 15:08:42] <capricorn-80> ahan
[12-Jun-2009 15:08:48] <capricorn-80> i can see password on snmp trap
[12-Jun-2009 15:08:53] <capricorn-80> ok will do that on monday
[12-Jun-2009 15:08:58] <capricorn-80> thanks for you help
[12-Jun-2009 15:08:59] <capricorn-80> take car e
[12-Jun-2009 15:09:00] <capricorn-80> bye
[12-Jun-2009 15:18:53] <ckrough> Anyone know of a faster way to parse objects in zope? I need to produce a report for a large number of devices (10k) and it takes around 5 min to complete
[12-Jun-2009 15:19:12] <ckrough> There isnt any kind of index in zope that I can take advantage of is there?
[12-Jun-2009 15:27:29] <ckrough> <crickets>
[12-Jun-2009 15:41:33] <ckrough> enough zenoss for the week. have a good weekend. Im out
[12-Jun-2009 16:01:07] <cmdln> Hello
[12-Jun-2009 16:02:00] <cmdln> Is there any way to manually model a topology?
[12-Jun-2009 16:02:40] <cmdln> For example if I am monitoring a bunch of virtual machines on the same host I dont want to be notified of each individual guest if the host goes down. I only need to be notified of the host.
[12-Jun-2009 16:04:27] cmdln is now known as cmdln_
[12-Jun-2009 16:04:43] cmdln_ is now known as cmdln
[12-Jun-2009 16:12:39] <FooDog> cmdln, i was wondering the same thing... ex: if the core switches are down, don't notify about edge swtiches...
[12-Jun-2009 17:44:20] <blocky> hey guys
[12-Jun-2009 17:45:01] <blocky> if i created a data source with a 300s cycle time and also a graph based on it, how hard is it to switch it to a 60s cycle time and have the graphs scale and display properly
[12-Jun-2009 17:45:09] <blocky> im not too worried about preserving the old rrd data
[12-Jun-2009 17:53:05] tehhobbit is now known as failboi
[12-Jun-2009 18:11:28] <lukecyca> cmdln, FooDog: This is a way of defining dependencies on events... http://blog.zenoss.com/2009/05/27/tip-of-the-month-event-suppression/
[12-Jun-2009 22:04:48] mephux_ is now known as mephux
[12-Jun-2009 23:22:54] <JulianMsanchez> Hi
[12-Jun-2009 23:23:33] <JulianMsanchez> Have a little problem whit my zenoss some one could help me?
[12-Jun-2009 23:27:01] <JulianMsanchez> I can't start zenjobs after upgrade...
[12-Jun-2009 23:27:13] <JulianMsanchez> it gave me AttributeError: JobManager
[13-Jun-2009 00:00:42] [disconnected at Sat Jun 13 00:00:42 2009]
[13-Jun-2009 00:00:42] [connected at Sat Jun 13 00:00:42 2009]
[13-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[13-Jun-2009 02:39:34] <JulianMsanchez> hi
[13-Jun-2009 04:33:17] <spyroux> Hi all
[13-Jun-2009 04:40:57] <JulianMsanchez> Hi spy...
[13-Jun-2009 04:41:14] <JulianMsanchez> not much movment here..
[13-Jun-2009 05:44:57] failboi is now known as loke
[13-Jun-2009 05:45:13] loke is now known as tehhobbit
[13-Jun-2009 07:04:52] <caseyh20> Hello everyone
[13-Jun-2009 14:00:58] <JulianMsanchez> hi
[13-Jun-2009 14:02:01] <rhetts> hi
[13-Jun-2009 14:05:32] <JulianMsanchez> how are you...
[13-Jun-2009 14:05:42] <JulianMsanchez> I need a zenoss guru...
[13-Jun-2009 14:05:45] <JulianMsanchez> are you?
[13-Jun-2009 14:39:56] <rhetts> im a user but i wouldnt call myself a guru
[14-Jun-2009 00:00:42] [disconnected at Sun Jun 14 00:00:42 2009]
[14-Jun-2009 00:00:42] [connected at Sun Jun 14 00:00:42 2009]
[14-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[14-Jun-2009 03:33:28] <caseyh20> so I finally have a tutorial up on how to install Zenoss 2.4.1 on FreeBSD 7.x http://www.caseyhillman.com/?p=88
[14-Jun-2009 11:23:35] <cmdln> Can I manually modify the zenping detected infrastructure? I would like to suppress alerts from hosts that depend on other hosts. For example virtual machines that are on vmhost. If vmhost goes down I do not need 5 alerts telling me about each guest that is no longer available, a single alert from vmhost is sufficient.
[14-Jun-2009 13:12:58] <spyroux> newbie question: how does zendisc choose which templates to bind to a device ? does he check every available template and add every templates that returns "something" ?
[14-Jun-2009 15:05:29] <cmdln> I cannot figure out how to suppress alerts for guest virtual machines when the host is the source of the issue
[14-Jun-2009 15:06:23] <cmdln> With alerting rules I do not seem to be able to do something like do not notify if alarm on other device
[14-Jun-2009 21:46:08] <sergeymasushko> I have in issue with the process monitoring... i' need to monitor exim process on a device. I've defined template with the ^.*\/*exim regex. zenoss detected this process with different keys e.g.
[14-Jun-2009 21:46:08] <sergeymasushko> exim-4.68-0 -bd -q30m
[14-Jun-2009 21:46:08] <sergeymasushko> exim-4.69-3
[14-Jun-2009 21:46:08] <sergeymasushko> exim-4.69-3 -q
[14-Jun-2009 21:46:08] <sergeymasushko> but actually I need to monitor only 'exim-4.68-0 -bd -q30m'. how can I prevent alerting/detecting of these 2 other processes?
[14-Jun-2009 22:22:23] <cmdln> sergeymasushko: you could adjust your regex to be more specific
[14-Jun-2009 22:28:24] <sergeymasushko> how?
[14-Jun-2009 22:29:20] <sergeymasushko> as far as I understand it collects data from different OIDs so I do not know how to adjust the regex...
[14-Jun-2009 22:30:10] <cmdln> ^.*\/*exim-4\.68-0\ -bd\ -q3m might do it
[14-Jun-2009 22:30:29] <cmdln> you said you defined the regex
[14-Jun-2009 22:32:59] <sergeymasushko> I take regex from the example.
[14-Jun-2009 22:32:59] <sergeymasushko> [zenoss@zenoss ~]$ snmpwalk -v2c -cpublic mydevice | grep exim
[14-Jun-2009 22:32:59] <sergeymasushko> HOST-RESOURCES-MIB::hrSWRunName.719 = STRING: "exim-4.69-3"
[14-Jun-2009 22:32:59] <sergeymasushko> [zenoss@zenoss ~]$ snmpwalk -v2c -cpublic mydevice | grep 'q30m'
[14-Jun-2009 22:32:59] <sergeymasushko> HOST-RESOURCES-MIB::hrSWRunParameters.719 = STRING: "-bd -q30m"
[14-Jun-2009 22:33:00] <sergeymasushko> So I guess regex that you suggested will not help me
[14-Jun-2009 22:33:55] <cmdln> just do a walk on HOST-RESOURCES-MIB::hrSWRunName.719
[14-Jun-2009 22:34:08] <cmdln> gtg
[14-Jun-2009 22:34:10] <cmdln> srry
[15-Jun-2009 00:00:42] [disconnected at Mon Jun 15 00:00:42 2009]
[15-Jun-2009 00:00:42] [connected at Mon Jun 15 00:00:42 2009]
[15-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[15-Jun-2009 09:22:30] <adr3nalin3> Hello everyone, I just installed 2.4.1 and I didn't get the initial setup screen that I usually get when installing zenoss, is there a way to invoke the initial setup wizard?
[15-Jun-2009 09:52:54] <jb> does anyone know if multiple SMTP addresses are allowed in the "Address" field of an alerting rule?
[15-Jun-2009 09:54:48] <jb> ah they do!
[15-Jun-2009 10:03:57] crow is now known as ckrouh
[15-Jun-2009 10:04:05] ckrouh is now known as ckrough
[15-Jun-2009 11:00:35] <chudler> mray?
[15-Jun-2009 11:00:40] <mrayzenoss> yeah
[15-Jun-2009 11:01:01] <chudler> I was curious, do you have anything about the criteria used to judge zenpacks?
[15-Jun-2009 11:01:47] <mrayzenoss> not really. Last year Chet and Erik and I looked at the list and picked. Perhaps this year I should try to find someone from the community without an entry
[15-Jun-2009 11:02:13] <mrayzenoss> but if we keep up this pace, we may add more prizes
[15-Jun-2009 11:02:21] <mrayzenoss> I already have more entries than last year's contest
[15-Jun-2009 11:02:26] <mrayzenoss> and we still have 2 months left
[15-Jun-2009 11:02:52] <mrayzenoss> btw, your Jabber one is in SVN, I just have to make the webpages
[15-Jun-2009 11:03:05] <chudler> thats really cool. I was just interested in what sort of methods are used.
[15-Jun-2009 11:03:41] <mrayzenoss> should we have voting?
[15-Jun-2009 11:03:47] <mrayzenoss> I'm open to suggestions
[15-Jun-2009 11:05:01] <chudler> no, but perhaps a few public descriptions of what is important to you guys when judging. Overall utlity of the zenpack? Breadth of potential installs? Coding concepts/style? Unique factor?
[15-Jun-2009 11:06:09] <chudler> whatever you are doing is working :-) and like I said, I was just curious. There is probably value in making the decision from the gut.
[15-Jun-2009 11:06:21] <mrayzenoss> yeah, that's essentially how we judged last year. Utility and popularity are important, but recognizing ZenPacks that seem ambitious in their scope
[15-Jun-2009 11:06:34] <mrayzenoss> last year only had 2 winners, this year is 1+5
[15-Jun-2009 11:06:41] <mrayzenoss> 1 grand prize and 5 runner ups
[15-Jun-2009 11:07:32] <mrayzenoss> for instance, at SELF someone described an ESX SSH solution they're working on. It models an ESX box, then adds the filesystems and guests to the host. All via SSH
[15-Jun-2009 11:08:00] <mrayzenoss> that sounds very useful and ambitious, as well as hitting SSH which gets him double credit if he gets it in as an entry
[15-Jun-2009 11:08:26] <mrayzenoss> right now there are 25 unique entries, with 4 SSH-based ones
[15-Jun-2009 11:08:33] <mrayzenoss> err, 3
[15-Jun-2009 11:09:03] <mrayzenoss> so you can go for lots of entries, which helps your odds since some of the prizes are pulled at random
[15-Jun-2009 11:09:48] <mrayzenoss> hmmm... looking at the webpage, apparently we only said 4 runner ups
[15-Jun-2009 11:10:10] <mrayzenoss> so the grand prize and 2 runner ups are picked by the judges, and 2 are randomly selected
[15-Jun-2009 11:11:50] * chudler needs to stuff the ballot with a slew of ssh modelers ;-)
[15-Jun-2009 11:12:04] <mrayzenoss> you better hurry, rocket has already sent in Gentoo and SuSE
[15-Jun-2009 11:12:23] <mrayzenoss> may I suggest the BSDs? There are a lot of them
[15-Jun-2009 11:12:51] <chudler> eheh thats a nice idea, unfortunately ssh monitoring has little value for me. I like to scratch my own itches :-)
[15-Jun-2009 11:13:04] <mrayzenoss> yeah, your Jabber bot seems pretty cool
[15-Jun-2009 11:13:25] <chudler> it is something that I use every day. Major itch that my boss wanted me to scratch, actually heh
[15-Jun-2009 11:13:30] <mrayzenoss> I'm trying to do 12 ZenPacks in parallel right now, so there should be a flood of announcements soon
[15-Jun-2009 11:13:58] <mrayzenoss> first I smoke test them, then get them into SVN, then create pages on the www and trac sites, then the announcements
[15-Jun-2009 11:14:04] <mrayzenoss> right now most are in SVN
[15-Jun-2009 11:14:11] <mrayzenoss> so next comes a lot of HTML
[15-Jun-2009 11:57:48] <rocket> Hey Matt
[15-Jun-2009 12:14:42] <FooDog> is there an official way to post bugs/suggestions against community zenpacks?
[15-Jun-2009 12:19:58] <rocket> FooDog: use trac it can reference community zenpacks I believe
[15-Jun-2009 12:20:10] <ckrough> Does "update collector" update collector files based on lastmod date, or does it push everything?
[15-Jun-2009 12:21:02] <ckrough> nm, looks like its an rsync.
[15-Jun-2009 12:21:27] <chudler> yeah rsync --times -r, or something like that.
[15-Jun-2009 12:21:52] <chudler> er rsync -tugr
[15-Jun-2009 12:28:56] <capricorn-80> hi
[15-Jun-2009 12:29:02] <capricorn-80> rocket you there ?
[15-Jun-2009 12:29:08] <rocket> yes
[15-Jun-2009 12:29:50] <capricorn-80> well i figure out some problem
[15-Jun-2009 12:29:55] <rocket> oh?
[15-Jun-2009 12:30:27] <capricorn-80> that management interface is giving RRD file missing error
[15-Jun-2009 12:32:38] <capricorn-80> but i can see CPU and memory graphs in the performance
[15-Jun-2009 12:32:58] <capricorn-80> but when i click on that Null0 it give errors mentioned on the graph
[15-Jun-2009 12:33:21] <capricorn-80> i.ie Missing RRD File
[15-Jun-2009 12:33:28] <capricorn-80> but every thing works fine on its Vlan
[15-Jun-2009 12:33:49] <capricorn-80> oh no
[15-Jun-2009 12:33:57] <capricorn-80> same error
[15-Jun-2009 12:33:59] <capricorn-80> ah
[15-Jun-2009 12:34:06] <capricorn-80> dont know what happened
[15-Jun-2009 12:36:22] <capricorn-80> rocket: http://pastebin.com/d26d25fa9
[15-Jun-2009 12:37:50] <rocket> if you can do a zenoss restart;zenperfsnmp run -v10 -d 10.254.95.254
[15-Jun-2009 12:37:57] <rocket> and see if you get anymore information
[15-Jun-2009 12:38:37] <capricorn-80> i just pasted you the out of zenprefsnmp
[15-Jun-2009 12:38:42] <capricorn-80> let me try it again
[15-Jun-2009 12:46:37] <capricorn-80> rocket: http://pastebin.com/m1b3d24e7
[15-Jun-2009 12:46:42] <capricorn-80> looks similar
[15-Jun-2009 12:48:03] <capricorn-80> some thing wrong .. graphs are too odd
[15-Jun-2009 12:52:24] <capricorn-80> need some help
[15-Jun-2009 12:52:52] <rocket> did you restart zenoss?
[15-Jun-2009 12:53:06] <rocket> have you checked the logs in the log directory?
[15-Jun-2009 12:53:35] <rocket> logs usually are in a place like /usr/local/zenoss/zenoss/logs
[15-Jun-2009 12:53:42] <rocket> once there do a
[15-Jun-2009 12:53:59] <rocket> grep Trace *;grep -i DEBUG *;grep -i WARN *
[15-Jun-2009 12:54:22] <rocket> and to find out which logs contain issues and look at the logs to figure out what is happening
[15-Jun-2009 13:01:47] <capricorn-80> rocket: http://pastebin.com/m16012e4d
[15-Jun-2009 13:03:42] <capricorn-80> really strange
[15-Jun-2009 13:05:50] <capricorn-80> rocket: http://pastebin.com/m675b93b9
[15-Jun-2009 13:05:55] <capricorn-80> this will be interesting
[15-Jun-2009 13:06:29] <rocket> thats a known response when the rrd file is missing
[15-Jun-2009 13:08:11] <capricorn-80> i was having problem with windows rrd file but for router it was working fine
[15-Jun-2009 13:08:18] <capricorn-80> and today all this happened
[15-Jun-2009 13:08:27] <capricorn-80> what should i do to solve it ?
[15-Jun-2009 13:13:05] <capricorn-80> its getting harder with zenoss
[15-Jun-2009 13:13:57] <coofamani> what is
[15-Jun-2009 13:15:58] <capricorn-80> coofamani: suddenly i got RRD missing file error
[15-Jun-2009 13:16:04] <capricorn-80> graphs are too odd
[15-Jun-2009 13:16:08] <capricorn-80> dont know what happened
[15-Jun-2009 13:16:20] <capricorn-80> help me how i can recover from it
[15-Jun-2009 13:16:36] <coofamani> Zenoss is a fairly complex application
[15-Jun-2009 13:17:13] <coofamani> Have you read through the Administrator manual and possible the Development manual. It helps to be familiar with it's foundation when you are troubleshooting issues.
[15-Jun-2009 13:18:05] <capricorn-80> coofamani: it was working fine
[15-Jun-2009 13:18:32] <capricorn-80> dont have time to switch to other nms
[15-Jun-2009 13:18:40] <capricorn-80> spent 2 weeks
[15-Jun-2009 13:20:37] <capricorn-80> kindly help me troubeshooting
[15-Jun-2009 13:20:48] <capricorn-80> in **
[15-Jun-2009 13:21:06] <rocket> capricorn-80: this is a volunteer channel .. as in some of this stuff you really need to start figuring out yourself
[15-Jun-2009 13:21:39] <rocket> if you need more support you probably should be thinking about getting a support contract with zenoss
[15-Jun-2009 13:22:38] <coofamani> You may want to consider taking their training course as well. That will save you a lot of time compared to learning it on your own.
[15-Jun-2009 13:23:30] <Jontler> checking out the weekly webex broadcast is also a good idea
[15-Jun-2009 13:23:56] <rocket> and the developers are usually here on thursdays to answer questions
[15-Jun-2009 13:24:25] <rocket> capricorn-80: but for those of us who dont have support contracts ... we have all had to figure out issues like you are having on our own
[15-Jun-2009 13:24:57] <rocket> capricorn-80: its for that reason that we understand what can go wrong and why ..
[15-Jun-2009 13:25:02] <capricorn-80> well thats what i am asking for some help right now
[15-Jun-2009 13:25:36] <capricorn-80> rocket: because it was working fine\
[15-Jun-2009 13:25:46] <capricorn-80> and suddenly it went off
[15-Jun-2009 13:26:02] <rocket> well there has to be something in the logs or event manager that would show the reason
[15-Jun-2009 13:26:17] <rocket> or you have changed something to make it no longer work ..
[15-Jun-2009 13:26:31] <rocket> have you checked disk space etc?
[15-Jun-2009 13:27:24] <capricorn-80> i did some settings in /etc/hosts
[15-Jun-2009 13:27:28] <capricorn-80> on friday
[15-Jun-2009 13:27:58] <rocket> do you have a backup of the original /etc/hosts?
[15-Jun-2009 13:28:13] <rocket> perhaps you need to remodel the device if its ip has changed etc
[15-Jun-2009 13:28:27] <capricorn-80> 127.0.0.1 sendto error Host localhost and my.machine.com are both using ip 127.0.0.1
[15-Jun-2009 13:28:36] <capricorn-80> no ip is the same
[15-Jun-2009 13:30:48] <coofamani> anyone else have no scroll bar in the gui?
[15-Jun-2009 13:30:56] <coofamani> I see some forum posts about it, but no solutions
[15-Jun-2009 13:31:01] <coofamani> firefox on fedora.
[15-Jun-2009 13:31:11] <coofamani> problem has persisted through 2 versions of FF now
[15-Jun-2009 13:31:27] <Jontler> ive got my scroll bar just fine
[15-Jun-2009 13:31:32] <Jontler> same setup
[15-Jun-2009 13:31:37] <coofamani> want to trade?
[15-Jun-2009 13:31:39] <Jontler> ha
[15-Jun-2009 13:31:55] <Jontler> what version of flash are you running?
[15-Jun-2009 13:31:59] <coofamani> its better, no matter what happens, only 30 devices are affected!
[15-Jun-2009 13:32:26] <capricorn-80> rocket: you think that can be a problem which i pasted above ?
[15-Jun-2009 13:33:16] <coofamani> Jontler: flash-plugin-10.0.22.87-release.i386
[15-Jun-2009 13:33:23] <coofamani> Jontler: you?
[15-Jun-2009 13:33:54] <coofamani> that device lists arent in flash are they?
[15-Jun-2009 13:34:34] <mrayzenoss> coofamani: the only use of flash is the network map
[15-Jun-2009 13:35:03] <Jontler> same coofamani
[15-Jun-2009 13:35:13] <Jontler> and i guess you dont need it
[15-Jun-2009 13:35:29] <rocket> capricorn-80: probably
[15-Jun-2009 13:38:17] <coofamani> heh, the onlsy results for a "Scroll bar" search on search.zenoss is other people asking about the scroll bar in IRC
[15-Jun-2009 13:38:51] <Jontler> did you try different browsers
[15-Jun-2009 13:39:01] <coofamani> yeah
[15-Jun-2009 13:39:05] <coofamani> its just on this box
[15-Jun-2009 13:39:10] <Jontler> ok
[15-Jun-2009 13:39:16] <coofamani> but its only zenoss
[15-Jun-2009 13:39:24] <coofamani> no other apps seem to have the same issue
[15-Jun-2009 13:39:25] <Jontler> i was gonna suggest you fire up epiphany, since it's based on firefox
[15-Jun-2009 13:39:42] <coofamani> regardless, Im replacing the box tomorrow. we'll see if the issue follows my profile
[15-Jun-2009 13:46:34] <coofamani> Jontler: interesting, no scroll bar in epiphany either
[15-Jun-2009 13:49:32] * Jontler shrugs
[15-Jun-2009 13:49:37] <Jontler> im interested to see if it follows your profile
[15-Jun-2009 13:50:32] <Jontler> did you try creating a new firefox profile, just for kicks?
[15-Jun-2009 13:50:50] <Jontler> that would tell you pretty quickly whether or not it's a profile issue
[15-Jun-2009 13:51:04] <coofamani> yup, not luck
[15-Jun-2009 14:12:55] <capricorn-80> rocket: need some help regarding some simple things
[15-Jun-2009 14:13:08] <capricorn-80> i just changed my server ip address and hostname
[15-Jun-2009 14:13:18] <capricorn-80> and some settings in resolve.conf
[15-Jun-2009 14:13:36] <capricorn-80> now apache is runing with but when i add 8080 with ip it just start processing
[15-Jun-2009 14:13:48] <capricorn-80> i have checked that 8080 is running
[15-Jun-2009 14:14:18] <capricorn-80> tcp 478 0 10.254.91.114:8080 10.254.91.72:4119 ESTABLISHED
[15-Jun-2009 14:14:18] <capricorn-80> tcp 479 0 10.254.91.114:8080 10.254.91.72:4114 CLOSE_WAIT
[15-Jun-2009 14:14:18] <capricorn-80> its done processing from last 10 misn
[15-Jun-2009 14:14:20] <capricorn-80> mins
[15-Jun-2009 14:14:55] <rocket> capricorn-80: which server? zenoss server?
[15-Jun-2009 14:15:04] <rocket> I am not sure what you are trying to accomplish
[15-Jun-2009 14:15:07] <capricorn-80> yes
[15-Jun-2009 14:15:29] <capricorn-80> i have changed ip address , its name and some dns related settings in resolve.conf
[15-Jun-2009 14:16:09] <capricorn-80> rocket: i was screwed and i checked that the ip address was from dhcp although i am having that ip from last 2 weeks so i change it to static
[15-Jun-2009 14:16:21] <capricorn-80> and also change my hostname
[15-Jun-2009 14:17:01] <rocket> for the server?
[15-Jun-2009 14:17:14] <capricorn-80> yes
[15-Jun-2009 14:17:27] <capricorn-80> all zenoss daemons are working fine
[15-Jun-2009 14:17:39] <rocket> ok so what is the problem then?
[15-Jun-2009 14:17:52] <jb> zenoss does not use apache..
[15-Jun-2009 14:19:05] <capricorn-80> http://ip:8080 and its processig
[15-Jun-2009 14:19:15] <capricorn-80> nothing opening the console
[15-Jun-2009 14:19:41] <rocket> check the zenoss logs and see if there are issues
[15-Jun-2009 14:20:12] <capricorn-80> rocket: i am thinking to remove zenoss and install it again
[15-Jun-2009 14:21:41] <capricorn-80> ask i messed up with things
[15-Jun-2009 14:22:08] <capricorn-80> its still a testing server the time it will take for trouble shooting .. its better that i will start new one
[15-Jun-2009 14:22:18] <rocket> start new
[15-Jun-2009 14:23:02] <capricorn-80> ok
[15-Jun-2009 14:39:28] <capricorn-80> rocket: see u tomorrow
[15-Jun-2009 14:39:29] <capricorn-80> bye
[15-Jun-2009 15:17:18] <blocky> if i delete RRD files will they be recreated automatically
[15-Jun-2009 15:32:20] <coofamani> yes
[15-Jun-2009 15:32:26] <coofamani> takes around 3 polling cycles
[15-Jun-2009 15:32:32] <coofamani> 15 minutes on a default install
[15-Jun-2009 16:11:55] <blocky> what if i change the polling cycle for the data points im gathering
[15-Jun-2009 16:11:56] <blocky> then delete
[15-Jun-2009 16:12:08] <blocky> will the newly created rrds have the correct interval
[15-Jun-2009 17:42:45] <caseyh20> how is everyone doing today?
[15-Jun-2009 17:43:16] <mrayzenoss> good, just about to check out
[15-Jun-2009 17:43:22] <mrayzenoss> but thanks for updating your blog entry
[15-Jun-2009 17:43:28] <mrayzenoss> I sent a lot of folks that way
[15-Jun-2009 17:44:31] <caseyh20> yeah, I have seen all the links, not much feedback so far though, I hope that means it is working for them lol
[15-Jun-2009 17:45:32] <mrayzenoss> I actually talked to some Gentoo users at SouthEast LinuxFest and sent them to it as well, figured they'd have similar issues
[15-Jun-2009 17:46:03] <caseyh20> it would be interesting to see if it worked for them
[15-Jun-2009 17:47:32] <caseyh20> I ran into an issue on sunday when I was doing a final test, apparently autoconf259 is no longer in the FreeBSD ports or packages so I had to go with autoconf263
[15-Jun-2009 17:47:53] <caseyh20> that is something that has changed within the past few weeks
[15-Jun-2009 17:50:12] <mrayzenoss> thanks a lot for posting that, I gotta run, but I'm looking forward to your iPhone stuff, whatever it is
[15-Jun-2009 17:53:13] <caseyh20> no problem, it is something that I enjoyed, for the most part lol, yeah the iPhone is something im looking forward to but im running a little behind. im gonna be in anaheim for a week then when I get back im gonna be working on the iPhone app as well as the FreeBSD port and a few zenpacks too
[16-Jun-2009 00:00:42] [disconnected at Tue Jun 16 00:00:42 2009]
[16-Jun-2009 00:00:42] [connected at Tue Jun 16 00:00:42 2009]
[16-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[16-Jun-2009 08:05:37] int_ is now known as int
[16-Jun-2009 09:35:49] <chudler> anyone forsee problems with puting /Perf data on NFS?
[16-Jun-2009 09:36:08] <venturaville> chudler: loads of problems
[16-Jun-2009 09:37:07] <chudler> please go on.
[16-Jun-2009 09:37:31] <venturaville> usually in RRD your are updating part of the file, but on NFS you end up forcing a cache flush on the entire file when you update it
[16-Jun-2009 09:37:57] <venturaville> which means you are downloading the entire RRD tree every x minutes
[16-Jun-2009 09:39:37] <venturaville> if your file tree is sufficiently large (read MBs or even GBs) then you are invalidating a lot of system disk cache memory as well as potentially saturating your network connection.....
[16-Jun-2009 09:39:49] <venturaville> iSCSI is an alternative.....
[16-Jun-2009 09:40:12] <chudler> I am used to those things, yeah, no big deal. Idea is to share the disk, so iscsi probably not a good fit.
[16-Jun-2009 09:40:37] <venturaville> don't know what you are trying to accomplish......
[16-Jun-2009 09:40:58] <venturaville> I might can help better if I knew that
[16-Jun-2009 09:40:58] <jb> gfs? "_
[16-Jun-2009 09:41:00] <jb>
[16-Jun-2009 09:41:10] <venturaville> good thought
[16-Jun-2009 09:41:15] <venturaville> but complicated
[16-Jun-2009 09:41:19] <jb> true
[16-Jun-2009 09:41:44] <jb> both devices need access at the same time to the data?
[16-Jun-2009 09:42:10] <chudler> its not my bag, but I will explain. Some people here move systems between collectors, and they want the rrd data to move with it. simple as that, I think.
[16-Jun-2009 09:42:45] <jb> hmm
[16-Jun-2009 09:43:28] <venturaville> ah I see
[16-Jun-2009 09:43:58] <jb> well, gfs would work..
[16-Jun-2009 09:43:58] <chudler> having it on NFS might be neato for some tangental reasons heh
[16-Jun-2009 09:44:06] <jb> but, I would not use NFS
[16-Jun-2009 09:44:13] <venturaville> seconded
[16-Jun-2009 09:44:32] <venturaville> an alternative way if you don't care so much about disk space is to use an rsync job to synchronize them around
[16-Jun-2009 09:44:38] <jb> possibly a very large performance bottleneck there..
[16-Jun-2009 09:44:45] <jb> yeah
[16-Jun-2009 09:45:19] <chudler> I think the ones pushing for a solution are considering rsync, or just leaning on Zenoss, inc for a fix.
[16-Jun-2009 09:45:54] <venturaville> I think enterprise already has stuff for stuff along that lines
[16-Jun-2009 09:47:53] <chudler> it is enterprise. perhaps the engineer is on crack
[16-Jun-2009 09:49:15] <chudler> I would love to do gfs or drbd or md+nbd 'cause those things are sexy and fun, but we already have a few stable, large (expensive) NFS clusters.
[16-Jun-2009 09:51:44] <jb> chudler: are you at uchicago?
[16-Jun-2009 09:53:14] <chudler> yes
[16-Jun-2009 10:03:25] <jb> cool.. im in chi as well.
[16-Jun-2009 11:17:31] <blocky> can anyone point me to a tutorial for creating custom reports that'
[16-Jun-2009 11:17:49] <blocky> s a bit more up to date than the one in the documentation
[16-Jun-2009 11:20:24] <mrayzenoss> so the Admin Guide has a bit: http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch12s08.html and the Developer Guide has a chapter on custom reports as well
[16-Jun-2009 11:20:31] <mrayzenoss> what are you trying to do?
[16-Jun-2009 11:21:09] <blocky> well ive got a bunch of templates now, all ssh commands
[16-Jun-2009 11:21:14] <blocky> and im going to add a few more
[16-Jun-2009 11:21:25] <blocky> but I'd like to be able to email myself and whoever else a report every morning
[16-Jun-2009 11:21:55] <blocky> ie. for all 8 boxes, here was the average 15 minute load, here was the peak mem usage, here is how many clients
[16-Jun-2009 11:22:17] <blocky> and im going to grab a bit of info from my squid as well, here is the ip of every TCP_DENIED message
[16-Jun-2009 11:22:52] <blocky> etc etc and i figured if i made it into a nice report then i could just have reportmail send it off via cron each morning before everyone gets to work
[16-Jun-2009 11:23:06] <mrayzenoss> this might prove useful too: http://www.zenoss.com/Members/cluther/creating-methods-for-custom-reports
[16-Jun-2009 11:23:35] <blocky> cool that looks great actually
[16-Jun-2009 11:24:21] <mrayzenoss> if you don't want to use the UI, you can hack up the .rpt files as well.
[16-Jun-2009 11:24:42] <blocky> does reportmail generate html email?
[16-Jun-2009 11:25:06] <blocky> okay this is where i was having problems before also
[16-Jun-2009 11:25:27] <blocky> nm
[16-Jun-2009 11:25:56] <blocky> zenoss is quite well done, btw
[16-Jun-2009 11:26:02] <blocky> once i understand it im sure it will be fantastic
[16-Jun-2009 11:26:10] <mrayzenoss> yeah, it keeps getting better
[16-Jun-2009 11:26:15] <mrayzenoss> thanks
[16-Jun-2009 11:29:04] <blocky> lets say i have two classes of server, one that is a subset of the other
[16-Jun-2009 11:29:20] <blocky> so some properties are not available and not meaningful for some of them
[16-Jun-2009 11:29:32] <blocky> is there a way i can have a device report for all servers anyway?
[16-Jun-2009 11:39:29] <mrayzenoss> if you're doing a custom report, you could probably check for a value and return "N/A" for those cases
[16-Jun-2009 11:39:46] <mrayzenoss> opened this ticket BTW: http://dev.zenoss.org/trac/ticket/5201
[16-Jun-2009 11:39:56] <mrayzenoss> we didn't have docs for reportmail
[16-Jun-2009 11:40:16] <rocket> Morning Matt
[16-Jun-2009 11:40:22] <mrayzenoss> good morning
[16-Jun-2009 11:40:39] <rocket> actually almost afternoon now ..
[16-Jun-2009 11:40:57] <mrayzenoss> yeah, doing the Getting Started Webinar in 20 minutes
[16-Jun-2009 11:41:02] <mrayzenoss> right in the middle of lunch
[16-Jun-2009 11:41:15] <rocket> things going better with the zenpacks I submitted now that most of the kinks are out of them?
[16-Jun-2009 11:41:35] <phasma> Hi! Can anybody help with graphics ?
[16-Jun-2009 11:41:37] <mrayzenoss> yeah, I've got graphs for my Interfaces
[16-Jun-2009 11:41:56] <rocket>
That helps for a linux zenpack [16-Jun-2009 11:42:09] <mrayzenoss> phasma: what's the issue?
[16-Jun-2009 11:42:33] <phasma> mrayzenoss: Where I can change encoding or font ?
[16-Jun-2009 11:42:53] <mrayzenoss> for the RRD stuff?
[16-Jun-2009 11:43:01] <mrayzenoss> or for everything?
[16-Jun-2009 11:44:08] <phasma> mrayzenoss: http://omploader.org/vMXVjMg/screen.png
[16-Jun-2009 11:44:25] <mrayzenoss> phasma: known issue. What version are you running?
[16-Jun-2009 11:44:28] <mrayzenoss> 2.4.0?
[16-Jun-2009 11:44:54] <phasma> 2.4.1
[16-Jun-2009 11:45:05] <mrayzenoss> which installer and which distro?
[16-Jun-2009 11:45:15] <phasma> mrayzenoss: FreeBSD from sources
[16-Jun-2009 11:45:32] <phasma> mrayzenoss: locale ru_RU.UTF-8
[16-Jun-2009 11:45:37] <mrayzenoss> http://dev.zenoss.org/trac/ticket/5006
[16-Jun-2009 11:46:08] <phasma> mrayzenoss: thanks
[16-Jun-2009 11:55:04] <blocky> so is there a list somewhere of other calls similar to getIpServicesListAsString
[16-Jun-2009 11:55:18] <blocky> can i retrieve my performance data source numbers?
[16-Jun-2009 11:58:20] <mrayzenoss> yeah, the dev guide has a bit, I know I've seen people do that
[16-Jun-2009 11:58:35] <mrayzenoss> there's a getRRDTemplates
[16-Jun-2009 11:58:42] <mrayzenoss> and a getRRDDataSources off of that
[16-Jun-2009 11:59:26] <mrayzenoss> I gotta go do a webinar, check the forums, I know I've seen it
[16-Jun-2009 15:04:17] <rocket> e1000g0: flags=201004843<UP,BROADCAST,RUNNING,MULTICAST,DHCP,IPv4,CoS> mtu 1500 index 2 inet 192.168.31.133 netmask ffffff00 broadcast 192.168.31.255 ether 0:c:29:22:94:aa
[16-Jun-2009 15:25:44] <zenChild> I'm trying to create a modeler that uses ssh, but I need the input from one command to send to another. Is there a away to send another command from the modeler plugin after one has run already?
[16-Jun-2009 15:26:29] <Jontler> off the top of my head, id batch it all into one python script and just call that
[16-Jun-2009 15:27:27] <zenChild> I can't it's an appliance box that uses rbash. I create or run any files on the box that aren't part of the default set. I can't even run sed or this would be a non-issue.
[16-Jun-2009 15:28:32] <Jontler> then im out of brilliant ideas
[16-Jun-2009 15:28:50] <zenChild> thanks anyway
[16-Jun-2009 18:41:44] <blocky> is there a way to reduce the size of graphs in a report/email but not on the original template?
[16-Jun-2009 19:23:26] <kgoedtel> you pretty much can't do anything useful with the reportmail
[16-Jun-2009 19:23:46] <kgoedtel> whatever your defaults are is what you get
[16-Jun-2009 19:32:03] <blocky> lol
[16-Jun-2009 21:03:59] <sergeymasushko> I have an issue with the process monitoring... i' need to monitor exim process on a device. I've defined template with the ^.*\/*exim regex. zenoss detected this process with different keys e.g.
[16-Jun-2009 21:03:59] <sergeymasushko> exim-4.68-0 -bd -q30m
[16-Jun-2009 21:03:59] <sergeymasushko> exim-4.69-3
[16-Jun-2009 21:03:59] <sergeymasushko> exim-4.69-3 -q
[16-Jun-2009 21:03:59] <sergeymasushko> but actually I need to monitor only 'exim-4.68-0 -bd -q30m'. how can I prevent alerting/detecting of these 2 other processes?
[16-Jun-2009 22:21:32] <aley__> hallow
[16-Jun-2009 22:21:43] <aley__> need help
[16-Jun-2009 22:25:20] <aley__> hellow
[16-Jun-2009 22:27:28] <sergeymasushko> hi
[16-Jun-2009 22:31:27] <sergeymasushko> aley__: ?
[16-Jun-2009 22:31:59] <aley__> i need help sergey
[16-Jun-2009 22:32:50] <aley__> when i run ( zenoss@aldebaran root]$ zendisc run
[16-Jun-2009 22:32:52] <aley__> INFO:zen.ZenDisc:Connecting to localhost:8789
[16-Jun-2009 22:32:52] <aley__> INFO:zen.ZenDisc:Connected to ZenHub
[16-Jun-2009 22:32:52] <aley__> WARNING:zen.ZenDisc:No networks configured
[16-Jun-2009 22:32:52] <aley__> INFO:zen.ZenDisc:Result: []
[16-Jun-2009 22:32:52] <aley__> INFO:zen.ZenDisc:Scan time: 0.00 seconds
[16-Jun-2009 22:32:54] <aley__> INFO:zen.ZenDisc:Daemon ZenDisc shutting down
[16-Jun-2009 22:32:56] <aley__> INFO:zen.ZenDisc:zendisc shutting down
[16-Jun-2009 22:32:58] <aley__> )
[16-Jun-2009 22:33:00] <aley__> why
[16-Jun-2009 22:36:28] <sergeymasushko> I see the same
[16-Jun-2009 22:37:42] <aley__> INFO:zen.ZenModeler:SNMP collection device Backhaul BPDE ke BAPPEDA PROV
[16-Jun-2009 22:37:43] <aley__> INFO:zen.ZenModeler:plugins: zenoss.snmp.DeviceMap, zenoss.snmp.InterfaceMap, zenoss.snmp.RouteMap
[16-Jun-2009 22:37:43] <aley__> INFO:zen.ZenModeler:No portscan plugins found for Backhaul BPDE ke BAPPEDA PROV
[16-Jun-2009 22:37:43] <aley__> INFO:zen.ZenModeler:Running 3 clients
[16-Jun-2009 22:37:43] <aley__> ERROR:zen.SshClient:Connection was refused by other side: 111: Connection refuse
[16-Jun-2009 22:38:02] <aley__> No portscan (what the meaning
[16-Jun-2009 22:38:58] <sergeymasushko> on the zProperties page of the device you should select needed portscan plugins
[16-Jun-2009 22:43:18] <aley__> how i add portscan plugins
[16-Jun-2009 22:45:29] <sergeymasushko> have you opened zProperties page?
[16-Jun-2009 22:47:37] <sergeymasushko> find 'zCollectorPlugins' and click Edit link. then click 'Add Fields' link then you can drag plugins from one box to another...
[16-Jun-2009 22:53:38] <aley__> yes
[16-Jun-2009 22:53:55] <aley__> ok i m finish to drag all plugins..
[16-Jun-2009 22:54:19] <sergeymasushko> click save and remodel the device
[16-Jun-2009 22:57:02] <aley__> ok
[16-Jun-2009 22:57:58] <aley__> wait the minit ..i reboot the my machine
[16-Jun-2009 23:09:06] <aley__> ok hai sergeymashuhko i need your help again
[16-Jun-2009 23:09:10] <aley__> [zenoss@aldebaran root]$ zendisc run
[16-Jun-2009 23:09:10] <aley__> INFO:zen.ZenDisc:Connecting to localhost:8789
[16-Jun-2009 23:09:10] <aley__> INFO:zen.ZenDisc:Connected to ZenHub
[16-Jun-2009 23:09:10] <aley__> WARNING:zen.ZenDisc:No networks configured
[16-Jun-2009 23:09:10] <aley__> INFO:zen.ZenDisc:Result: []
[16-Jun-2009 23:09:18] <aley__> INFO:zen.ZenDisc:Scan time: 0.00 seconds
[16-Jun-2009 23:09:18] <aley__> INFO:zen.ZenDisc:Daemon ZenDisc shutting down
[16-Jun-2009 23:09:18] <aley__> INFO:zen.ZenDisc:zendisc shutting down
[16-Jun-2009 23:09:18] <aley__> [zenoss@aldebaran root]$
[16-Jun-2009 23:09:37] <aley__> what the meaningn warning WARNING:zen.ZenDisc:No networks configured
[16-Jun-2009 23:14:10] <sergeymasushko> just ignore it for now
[16-Jun-2009 23:14:31] <aley__> oouh
[16-Jun-2009 23:15:16] <aley__> why the color in my device is orange ? normal is color green righ
[16-Jun-2009 23:16:00] <aley__> User timeout caused connection failure.
[16-Jun-2009 23:20:58] <sergeymasushko> open event tab and there you can read event details
[16-Jun-2009 23:35:41] <aley__> maybe zenoss 2.4 have bug ?
[17-Jun-2009 00:00:42] [disconnected at Wed Jun 17 00:00:42 2009]
[17-Jun-2009 00:00:42] [connected at Wed Jun 17 00:00:42 2009]
[17-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[17-Jun-2009 05:31:59] <phasma> Hi! I have a problem with process monitoring via net-snmp on linux host. 'snmpwalk -v2c -cprivate localhost | grep sd.fcgi' works perflectly, but Zenoss zenprocess with regex "sd.fcgi" writes to log error, what process doesn't running
[17-Jun-2009 05:33:07] <phasma> snmpwalk output: http://pastebin.com/m1bbaa835
[17-Jun-2009 06:17:23] <tripitaka> Hi, I'm getting a lot of snmp traps which are in event class /unknown, and which generate events with a severity of warning or critical. How can I tell zenoss that all events in the /Unknown class should be reclassified as severity 'info'?
[17-Jun-2009 08:21:22] <nzle> we recently had 2 dlink 1224 swtiches management/monitoring I go unresponsive. The timeframe this happened during was compatible with when we were modeling various devices on our newly installed zenoss (vmware appliance). We also monitor these devices with MRTG(snmp) and Nagios(ping). Has anyone seen this sort of thing after snmp modeling this switch with zenoss or would my time be better served looking in other directions.
[17-Jun-2009 08:21:46] <nzle> monitoring I = monitoring IP
[17-Jun-2009 08:26:17] <nzle> not sure if my message went out since my nick wasnt registered, resending...
[17-Jun-2009 08:26:43] <nzle> we recently had 2 dlink 1224 swtiches management/monitoring IP go unresponsive. The timeframe this happened during was compatible with when we were modeling various devices on our newly installed zenoss (vmware appliance). We also monitor these devices with MRTG(snmp) and Nagios(ping). Has anyone seen this sort of thing after snmp modeling this switch with zenoss or would my time be better served looking in other directions.
[17-Jun-2009 09:13:24] <phasma> it works !!!
[17-Jun-2009 09:23:32] <rocket> morning matt
[17-Jun-2009 09:26:06] <mrayzenoss> good morning
[17-Jun-2009 09:32:15] <jb> hey matt.. http://dev.zenoss.org/trac/ticket/5014
[17-Jun-2009 09:32:19] <jb> is there a patch available for this yet?
[17-Jun-2009 09:32:49] <jb> a couple are discussed in that ticket, I believe..
[17-Jun-2009 09:38:18] <tripitaka> Can I reclassify all /Unknown events as severity 'info'?
[17-Jun-2009 09:38:41] <mrayzenoss> jb: You want to zenpatch 13735
[17-Jun-2009 09:38:50] <mrayzenoss> jb: that's the one that is the 2.4.x tree
[17-Jun-2009 09:38:55] <mrayzenoss> err, branch
[17-Jun-2009 09:41:33] <mrayzenoss> 2.4.2 is getting close to release
[17-Jun-2009 09:47:29] <jb> thanks
[17-Jun-2009 10:31:47] <daMaestro> how do you add and arbitrary os process monitor?
[17-Jun-2009 10:31:54] <daMaestro> i'd like to make sure process "asdf" is running
[17-Jun-2009 10:48:08] <mrayzenoss> daMaestro: take a look at Jane Curry's "Methods of monitoring processes with Zenoss" http://www.skills-1st.co.uk/papers/jcurry.html
[17-Jun-2009 10:48:25] <mrayzenoss> she drills down into it pretty well
[17-Jun-2009 11:22:08] <kisielk> is there a way to map a particular event to an event class regardless of the component?
[17-Jun-2009 11:22:28] <kisielk> For example, I often get a message like "nss_ldap: could not search LDAP server - Server is unavailable" when and LDAP server is unreachable, but the component depends on what application was trying to use nsswitch at that time, which could be anything
[17-Jun-2009 11:28:35] <mrayzenoss> kisielk: you could use a transform to change the event to whatever you want to map it to
[17-Jun-2009 11:29:12] <kisielk> but how can I have the transform map any component?
[17-Jun-2009 11:29:29] <kisielk> I thought you could only create rules per component
[17-Jun-2009 11:29:40] <mrayzenoss> in 2.4, the event transforms are now hierarchal
[17-Jun-2009 11:30:00] <mrayzenoss> so you could place your transform in /Events to change the event before it gets to other events
[17-Jun-2009 11:30:44] <mrayzenoss> including changing the component
[17-Jun-2009 11:30:47] <kisielk> oh really
[17-Jun-2009 11:30:49] <kisielk> that's cool
[17-Jun-2009 11:30:56] <mrayzenoss> dangerous, but cool
[17-Jun-2009 11:31:27] <kisielk> I think our zenoss install is finally getting to be in a nice state
[17-Jun-2009 11:31:41] <kisielk> there's still a few things we haven't been able to replace
[17-Jun-2009 11:31:44] <kisielk> from our previous monitoring system
[17-Jun-2009 11:31:56] <mrayzenoss> examples?
[17-Jun-2009 11:32:34] <kisielk> well, we have an SNMP module that monitors quotas for one of our file servers
[17-Jun-2009 11:32:52] <jb> it outputs the quota via snmp?
[17-Jun-2009 11:32:57] <kisielk> yes
[17-Jun-2009 11:33:02] <jb> thats easy enough
[17-Jun-2009 11:33:04] <kisielk> it's in 5 arrays
[17-Jun-2009 11:33:11] <jb> just add it as a data source
[17-Jun-2009 11:33:14] <kisielk> one array is the project name, then the soft quota, hard quota
[17-Jun-2009 11:33:19] <kisielk> disk usage
[17-Jun-2009 11:33:19] <jb> and then you can set thresholds, graphs, etc on it
[17-Jun-2009 11:33:22] <kisielk> and then a message
[17-Jun-2009 11:33:42] <kisielk> yeah, but the threshold should come from the other SNMP data source
[17-Jun-2009 11:33:45] <kisielk> not be hard coded
[17-Jun-2009 11:33:49] <kisielk> because the quotas can change any time
[17-Jun-2009 11:33:56] <jb> ah
[17-Jun-2009 11:33:58] <jb> hmm
[17-Jun-2009 11:34:04] <kisielk> that was easy to do in Zabbix
[17-Jun-2009 11:34:07] <kisielk> but not so easy in Zenoss
[17-Jun-2009 11:36:12] <kisielk> anyway, I think that's the last major thing at that point
[17-Jun-2009 11:36:28] <kisielk> I guess I could write a nagios plugin for the checks
[17-Jun-2009 11:37:26] <daMaestro> mrayzenoss, thanks... reading now
[17-Jun-2009 11:41:10] <daMaestro> mrayzenoss, is there any easy way to add a simple prTable check?
[17-Jun-2009 11:41:39] <daMaestro> basically.. i'd like to simply say "send alert if there is not a process running called 'asdf'" and want this information to come from an snmp poll
[17-Jun-2009 11:45:56] <mrayzenoss> yeah, you should be able to add the process your want monitored. Section 2 of that paper covers it
[17-Jun-2009 11:49:31] <blocky> is there a way I can include the output of an external script into a report
[17-Jun-2009 11:49:53] <blocky> like generated html for example
[17-Jun-2009 11:52:41] bret is now known as CamargoBP
[17-Jun-2009 11:52:46] <mrayzenoss> blocky: I guess if you were doing something like this: http://www.zenoss.com/Members/cluther/creating-methods-for-custom-reports
[17-Jun-2009 11:52:55] <mrayzenoss> you could just return a block of HTML
[17-Jun-2009 11:56:29] <daMaestro> mrayzenoss, so it looks like i should add some proc items to our snmpd.conf and then traps will be sent?
[17-Jun-2009 11:57:26] <mrayzenoss> not traps
[17-Jun-2009 11:57:42] <mrayzenoss> snmpd will expose the process list for direct monitoring
[17-Jun-2009 11:59:02] <blocky> how would i have that call an external python script
[17-Jun-2009 11:59:39] <blocky> do i retrieve values from the rrd files manually?
[17-Jun-2009 12:00:08] <mrayzenoss> blocky: from that howto, you can put any Python you want into the script.
[17-Jun-2009 12:00:38] <mrayzenoss> this includes returning a block of HTML, as well as accessing the Zenoss system
[17-Jun-2009 12:00:47] <blocky> would it put the entire block of html into one cell of the table?
[17-Jun-2009 12:00:49] <mrayzenoss> the RRD stuff is accessible
[17-Jun-2009 12:00:58] <daMaestro> mrayzenoss, so how do i add it to the device, i guess has been my question
[17-Jun-2009 12:01:27] <daMaestro> mrayzenoss, it looks like i have to add a process class
[17-Jun-2009 12:02:37] <daMaestro> mrayzenoss, ah, d'oh
[17-Jun-2009 12:02:44] <daMaestro> mrayzenoss, just have to add it :-/;
[17-Jun-2009 12:03:37] <blocky> how do I access the rrd stuff through zenoss?
[17-Jun-2009 12:04:07] <mrayzenoss> when you create your custom report and add the custom method, you can access the device that is making the call. I'm not finding a good example
[17-Jun-2009 12:04:43] <mrayzenoss> I guess the device is the context
[17-Jun-2009 12:05:51] <blocky> hmm okay well ill play around with it a bit
[17-Jun-2009 12:06:16] <mrayzenoss> so you should be able to call context.getRRDValue("laLoadInt5_laLoadInt5") for example
[17-Jun-2009 12:06:34] <mrayzenoss> you may need to dig in the Dev Guide a bit
[17-Jun-2009 12:07:17] <mrayzenoss> some of the scripts in my Zenoss/MindTouch mashup did similar stuff http://zenossdeki.svn.sourceforge.net/viewvc/zenossdeki/zenoss/
[17-Jun-2009 12:07:32] <blocky> okay cool
[17-Jun-2009 12:07:37] <mrayzenoss> and return a block of HTML
[17-Jun-2009 12:08:00] <blocky> do you know, when a graph is generated by rrdtool with avg min and max values, how those are calculated
[17-Jun-2009 12:08:21] <blocky> cause ive got laloadint for 1 5 and 15 between 0 and 1, but my max value is around 400
[17-Jun-2009 12:08:54] <mrayzenoss> the value is whatever the datasource off of the template is pulling
[17-Jun-2009 12:09:04] <blocky> strange
[17-Jun-2009 12:09:16] <blocky> is this a good idea to try http://www.zenoss.com/community/docs/zenoss-dev-guide/2.2.4/ch09s01.html
[17-Jun-2009 12:09:18] <mrayzenoss> the max value would be whatever SNMP reported and stored in the RRD file
[17-Jun-2009 12:09:36] <mrayzenoss> yeah, I'd use the 2.4 version though
[17-Jun-2009 12:09:49] <blocky> well im using ssh command not snmp but yea
[17-Jun-2009 12:10:21] <blocky> must have got that data in there soemhow
[17-Jun-2009 12:12:45] <mrayzenoss> you can look at some of the stuff from RRDTool if you want to look at the RRD files closer
[17-Jun-2009 12:12:52] <mrayzenoss> http://oss.oetiker.ch/rrdtool/doc/index.en.html
[17-Jun-2009 12:13:47] <blocky> does he provide the python bindings?
[17-Jun-2009 12:14:00] <blocky> ah nm
[17-Jun-2009 12:17:07] <xulfer> How would I go about naming graphs in my template via snmp output?
[17-Jun-2009 12:28:17] <LW> question, I was recently asked if there is any way to identify and alarm on a % of delta change on a file system, so if overnight, a machine's disk utilization jumps 10%, then to alarm
[17-Jun-2009 12:29:56] <daMaestro> mrayzenoss, does the regex match against UCD-SNMP-MIB::prNames ?
[17-Jun-2009 12:32:07] <blocky> mrayzenoss: does a custom method return a single string always?
[17-Jun-2009 12:32:16] <venturaville> LW: not sure of the syntax but I believe you can do it based upon some time delta in getRRDValues: def getRRDValues(self, dsnames, start=None, end=None, function="LAST",
[17-Jun-2009 12:32:16] <venturaville> format="%.2lf", extraRpn=""):
[17-Jun-2009 12:32:46] <venturaville> I think the start and end ones are what you want......
[17-Jun-2009 12:33:50] <LW> intruiging
[17-Jun-2009 12:34:10] <venturaville> let me know if you get it working... I think that it might be useful for something I am doing too
[17-Jun-2009 12:35:42] * LW ponders what server to use to test
[17-Jun-2009 12:40:19] <venturaville> >>> f.getRRDValues('usedBlocks',start)
[17-Jun-2009 12:40:19] <venturaville> 2009-06-17 13:39:05 ERROR RenderServer Failed while generating summary
[17-Jun-2009 12:40:19] <venturaville> Traceback (most recent call last):
[17-Jun-2009 12:40:19] <venturaville> File "/opt/zenoss/Products/ZenRRD/RenderServer.py", line 380, in summary
[17-Jun-2009 12:40:19] <venturaville> values = rrdtool.graph(*gopts)[2]
[17-Jun-2009 12:40:20] <venturaville> error: Attempting to reuse 's_r'
[17-Jun-2009 12:40:24] <venturaville> hrm 2.4 is not very happy
[17-Jun-2009 12:43:54] <venturaville> >>> import time
[17-Jun-2009 12:43:54] <venturaville> >>> mystart = time.time() - (3600 * 8)
[17-Jun-2009 12:43:54] <venturaville> >>> args = dict(start=mystart)
[17-Jun-2009 12:43:54] <venturaville> >>> f.getRRDValue('usedBlocks',**args)
[17-Jun-2009 12:43:54] <venturaville> 10021455.52
[17-Jun-2009 12:43:55] <venturaville> >>> f.getRRDValue('usedBlocks')
[17-Jun-2009 12:43:57] <venturaville> 10019746.58
[17-Jun-2009 12:43:59] <venturaville> better
[17-Jun-2009 12:44:06] <venturaville> LW: I think that is what you want...........
[17-Jun-2009 12:44:51] <daMaestro> looks like it does
[17-Jun-2009 12:48:09] <blocky> okay ive done something wrong but i have no idea what
[17-Jun-2009 12:48:19] <blocky> followed this tutorial http://www.zenoss.com/Members/cluther/creating-methods-for-custom-reports
[17-Jun-2009 12:48:22] <blocky> and it works fine
[17-Jun-2009 12:48:29] <blocky> but when i create another method, it doesnt work
[17-Jun-2009 12:48:34] <blocky> just shows Unknown Column
[17-Jun-2009 12:48:40] <blocky> even with the exact same python code
[17-Jun-2009 12:57:42] <LW> venturaville: looks like I can do it by using a the "Create CMD" field for a datasource
[17-Jun-2009 13:02:21] <ckrough> blocky: post the method on pastebin.com
[17-Jun-2009 13:22:46] <blocky> ive fixed it
[17-Jun-2009 13:23:03] <blocky> though the only thing i changed was my method name from capital to lowercase
[17-Jun-2009 13:23:07] <blocky> so who knows
[17-Jun-2009 13:23:20] <blocky> right now im trying to figure out how to access the rest of the data from within python
[17-Jun-2009 13:23:30] <blocky> rrd data for example
[17-Jun-2009 13:23:51] <blocky> the ZenRRD interface doesnt seem to be geared towards actually reading and manipulating the data
[17-Jun-2009 13:24:47] <rmatte> anyone know if there's a way to make it so that users can reset their LDAP password in Zenoss?
[17-Jun-2009 13:24:51] <rmatte> or is it not possible?
[17-Jun-2009 13:31:29] <Matir> Hi, I just installed Zenoss 2.4.1 Core on CentOS 5, and I don't seem to be getting Page Menus anywhere. (At least, that's what I think it's called)
[17-Jun-2009 13:43:24] <blocky> anyone ever tried to get splunk results into zenoss via python
[17-Jun-2009 13:46:39] <LW> anybody know off hand how to change the file system template out on a single host?
[17-Jun-2009 13:49:53] <mloven> blocky: we're working on splunk-zenoss integration, but just via some syslog forwarding craziness. I'd be interested to see how it would work out using python...
[17-Jun-2009 14:02:59] <blocky> id like to have zenoss generate a report which includes splunk data
[17-Jun-2009 14:03:47] daMaestro|isBack is now known as daMaestro
[17-Jun-2009 14:03:48] <blocky> our splunk runs on a private proxy, so for example each morning along with the zenoss perf data id like to see how many TCP_DENIED messages were generated, sorted by IP, with a link to whois data for each ip
[17-Jun-2009 14:04:06] <blocky> im still trying to figure out the zenoss report framework
[17-Jun-2009 14:04:18] <blocky> do not really really understand the API at all yet, tbh
[17-Jun-2009 14:25:55] <blocky> is there any documentation for what i can use with the device context?
[17-Jun-2009 14:31:58] <ckrough> blocky: http://www.zenoss.com/community/docs/zenoss-dev-guide/2.4.0/apbs02.html#d4e4396
[17-Jun-2009 14:32:34] <blocky> beauty thanks
[17-Jun-2009 14:33:03] <ckrough> you can also do this in zendmd:
[17-Jun-2009 14:33:10] <ckrough> dev = devices.findDevice('_yourdevicename_')
[17-Jun-2009 14:33:10] <ckrough> print '\n'.join(dir(dev))
[17-Jun-2009 14:33:17] <ckrough> and it will return some stuff to play with
[17-Jun-2009 14:36:14] <blocky> cool
[17-Jun-2009 14:36:24] <blocky> zendmd is like a python wrapper?
[17-Jun-2009 14:44:06] <ckrough> yeah
[17-Jun-2009 14:44:21] <ckrough> its the parent class for all of zenoss, I believe
[17-Jun-2009 14:44:26] <ckrough> you can just
[17-Jun-2009 14:44:35] <ckrough> 'zendmd' as the zenoss user on your zenoss box
[17-Jun-2009 14:44:43] <ckrough> from the command line
[17-Jun-2009 14:47:28] <blocky> so do you know much about the getRRDValue func?
[17-Jun-2009 14:50:51] <blocky> it seems to return the current value just fine, but what if I want the min or max for a certain time range like what appears on the graphs
[17-Jun-2009 14:54:42] <ckrough> not sure about doing that in zenoss
[17-Jun-2009 14:55:02] <ckrough> we run a couple external perl scripts that build that data from rrdserv on the collectors
[17-Jun-2009 15:39:03] <blocky> if there was a function to iterate through each rrd file, store the max value and then use the heartbeat to show like: Load maximum was 0.75 at 2:30 this morning (click here for graph)
[17-Jun-2009 15:39:16] <blocky> that would be ideal
[17-Jun-2009 15:43:11] <daMaestro> anyone have a creative way to build a report that will break down bandwidth usage by daily total?
[17-Jun-2009 16:24:37] <daMaestro> anyone know what to look for to clear zenmodler heartbeat events?
[17-Jun-2009 16:27:51] <mrayzenoss> Event Manager->Clear All Heartbeats
[17-Jun-2009 16:30:18] <blocky> if I'm calling fetchRRDValue how do i set the resolution
[17-Jun-2009 16:33:32] <blocky> this is interesting
[17-Jun-2009 16:33:57] <blocky> if i try to 'import time' at the start of my method, i can't view the report, it redirects me to the login page and wont accept my pass
[17-Jun-2009 16:38:15] <daMaestro> mrayzenoss, i mean.. to find out why the heartbeat is alerting in the first place
[17-Jun-2009 16:39:16] <mrayzenoss> on the event is there anything interesting on the details?
[17-Jun-2009 16:39:23] <mrayzenoss> perhaps if you go to the log
[17-Jun-2009 16:39:25] <daMaestro> ah, i'll look in the details
[17-Jun-2009 16:39:26] <daMaestro> thanks.
[17-Jun-2009 16:39:50] <mrayzenoss> you can view the daemon logs from the Settings->Daemons (or from $ZENHOME/logs
[17-Jun-2009 17:24:27] <mrayzenoss> Zenoss Jobs forum is now live
[17-Jun-2009 17:24:31] <cgibbons> yay!
[17-Jun-2009 17:54:06] daMaestro|isBack is now known as daMaestro
[17-Jun-2009 17:58:19] <blocky> is zendmd strictly an interpreter
[17-Jun-2009 17:58:34] <blocky> strictly interactive i mean
[17-Jun-2009 17:58:41] <blocky> can i run .py files through it
[17-Jun-2009 18:05:51] <blocky> something is wrong with my graphs
[17-Jun-2009 18:06:09] <blocky> ive been collecting load data so all the numbers are between 0 and 1
[17-Jun-2009 18:06:13] <blocky> but the graph has scaled it somehow
[17-Jun-2009 21:30:10] <crow> anyone know the TALES syntax to run arbitrary code? I need to pull in some variables from the GET
[17-Jun-2009 21:45:06] <crow> http://zope-is-evil-666.idyll.org/
[18-Jun-2009 00:00:42] [disconnected at Thu Jun 18 00:00:42 2009]
[18-Jun-2009 00:00:42] [connected at Thu Jun 18 00:00:42 2009]
[18-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[18-Jun-2009 02:18:32] <detach-> hey guys, wtf is this zenoss-jobs mailing list about?
[18-Jun-2009 02:18:48] <detach-> can we have our e-mails removed from the list? none of us subscribed to it in the first place
[18-Jun-2009 03:00:18] <JulianMsanchez> hi
[18-Jun-2009 03:33:42] <sergeymasushko> hey guys! are you going to fix these spam messages ? it's really annoying.
[18-Jun-2009 03:38:21] <rhetts> its a bit annoying but everyone bitching isnt helping sicne there is obviously a problem
[18-Jun-2009 07:10:28] <cain22> hell all...
[18-Jun-2009 07:10:49] <cain22> hey, anyone else getting spammed from zenoss forumn?
[18-Jun-2009 07:13:05] <cain22> hello...
[18-Jun-2009 07:16:03] <cain22> anyone in here?
[18-Jun-2009 07:17:51] <cain22> list
[18-Jun-2009 07:18:59] <venturaville> zenoss forum is pretty quiet for me
[18-Jun-2009 07:19:25] <venturaville> irc is pretty quiet on here until around 9 or 10 EST
[18-Jun-2009 07:20:53] <cain22> hmm...hey, have you been spammed by the zenoss forumn?
[18-Jun-2009 07:21:58] <venturaville> zope
[18-Jun-2009 07:22:04] <venturaville> nope
[18-Jun-2009 07:22:42] <cain22> they've sent me like 30+ emials in less than a day
[18-Jun-2009 07:22:46] <cain22> its annoying
[18-Jun-2009 07:22:58] <venturaville> 30 email is nothing
[18-Jun-2009 07:23:23] <venturaville> though I don't even think the forum has sent me that much
[18-Jun-2009 07:23:47] <venturaville> users forum?
[18-Jun-2009 07:24:57] <cain22> nah
[18-Jun-2009 07:25:02] <cain22> zenoss-jobs
[18-Jun-2009 08:04:38] <detach-> hi guys YOU NEED TO FIX YOUR JOBS MAILING LIST IMMEDIATELY
[18-Jun-2009 08:11:25] <FooDog> haha, wow
[18-Jun-2009 08:15:40] <crow> ha
[18-Jun-2009 08:23:45] <crow> zenoss.com is sloooow today
[18-Jun-2009 08:37:21] <rmatte> When I'm modelling Linux servers using straight SNMP Zenoss isn't picking up anything for the OS tab
[18-Jun-2009 08:37:26] <rmatte> it's not even picking up interfaces
[18-Jun-2009 08:38:17] <rmatte> any thoughts?
[18-Jun-2009 08:38:35] <sergeymasushko> are there selected needed modeling plugins ?
[18-Jun-2009 08:39:27] <rmatte> the interface map collector plugin is bound to the device group
[18-Jun-2009 08:39:37] <rmatte> so it should be picking up on them
[18-Jun-2009 08:44:51] <sergeymasushko> rmatte: if it shouldn't
maybe try to add it and remodel the device?[18-Jun-2009 09:05:47] <dethstar> so, what's with the zenoss-jobs spam in my inbox this morning?
[18-Jun-2009 09:10:09] <sergeymasushko> dethstar: dunno, I believe Matt do not have a chance to see/fix it yet
[18-Jun-2009 09:15:05] <dethstar> sergeymasushko, that's cool... just wondering wtf was going on
[18-Jun-2009 09:15:46] <sergeymasushko> I guess it's some glitch
[18-Jun-2009 09:16:28] <LW> so my attempt to utilize the "Create cmd" field for an rrd value to create a derived value for identifing #delta over 24 hours was a failure.... I guess I may just be better with some sort command based helper script.....
[18-Jun-2009 09:22:05] <zenChild> When generating table headers in custom skins (.pt files), does anyone know how to sort numerically in the getTableHeader method?
[18-Jun-2009 09:31:38] <rmatte> What does zSnmpCommunities do again?
[18-Jun-2009 09:31:45] <rmatte> it doesn't seem to do anything from what I can see
[18-Jun-2009 09:32:31] <rmatte> oh, it's only for discovery
[18-Jun-2009 09:33:09] <crow> I think it's used for collecting performance stats as well
[18-Jun-2009 09:41:29] <LW> anybody know where I can find a list of ${here} tales expressions?
[18-Jun-2009 09:42:02] <crow> http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/apd.html
[18-Jun-2009 09:42:25] <cain22> hello all
[18-Jun-2009 09:42:56] <crow> LW: theres also the API doc, but its a bit unweildy for picking out available ecxpressions
[18-Jun-2009 09:45:43] <crow> hello cain
[18-Jun-2009 09:45:56] crow is now known as ckrough
[18-Jun-2009 09:58:20] * LW greps the codebase looking for examples
[18-Jun-2009 10:08:20] <ckrough> LW: you can also 'dir' the methods in zendmd
[18-Jun-2009 10:15:19] <ckrough> Anyone know how I can access the GET request data in a zope template
[18-Jun-2009 10:18:40] <venturaville> in tales expressions, I believe it is something like: request/blah
[18-Jun-2009 10:18:42] <venturaville> like request/URL
[18-Jun-2009 10:19:08] <ckrough> venturaville: will try, thanks
[18-Jun-2009 10:19:16] <venturaville> http://wiki.zope.org/zope2/REQUESTX
[18-Jun-2009 10:20:05] <ckrough> venturaville: yup, request/BLAH
[18-Jun-2009 10:25:19] <ckrough> http://pastebin.com/m44cab564 on line 48 I want to pass a drange in via the get, but Im no sure how to get to it.
[18-Jun-2009 10:25:53] <ckrough> request/blah doesnt work in this context, I guess because Im dealing with python and not zope templates at that level
[18-Jun-2009 10:33:22] <venturaville> <textarea rows="80" cols="20" tal:attributes="rows request/rows;cols request/cols">
[18-Jun-2009 10:33:31] <venturaville> I don't think you want to use it in a python context
[18-Jun-2009 10:33:49] <venturaville> http://www.faqs.org/docs/ZopeBook/AppendixC.html
[18-Jun-2009 10:34:52] <ckrough> venturaville: we talkin about the same thing? you lostme
[18-Jun-2009 10:35:20] <venturaville> yeah ... take a closer look at that line..............
[18-Jun-2009 10:35:46] <ckrough> the line you just pasted doesnt exist in my template
[18-Jun-2009 10:37:47] <venturaville> that was from the appendix
[18-Jun-2009 10:37:53] <venturaville> let me see if I can translate it to yours
[18-Jun-2009 10:38:09] <ckrough> ok, looking at it now
[18-Jun-2009 10:39:30] <venturaville> I'm at a loss as to how to translate it to what you are doing
[18-Jun-2009 10:39:42] <venturaville> the way listed on that AppendixC page is TAL specific
[18-Jun-2009 10:39:52] <venturaville> I am not sure how to get at that request object from the python side
[18-Jun-2009 10:40:33] <ckrough> may not be possible
[18-Jun-2009 10:41:48] <venturaville> look at Products/ZenReports/reports/Performance_Reports/CPU_Utilization.rpt
[18-Jun-2009 10:42:09] <venturaville> I think you can at least do it in another plugin and then include it in your report
[18-Jun-2009 10:44:04] <venturaville> objects python: (hasattr(request, 'doExport') and list(objects)) or objects;
[18-Jun-2009 10:44:12] <venturaville> and then referenced like:
[18-Jun-2009 10:45:37] <venturaville> request/blah
[18-Jun-2009 10:45:49] <venturaville> err no that is the tal way again
[18-Jun-2009 10:45:51] <ckrough> why does this feel so dirty
[18-Jun-2009 10:47:02] <venturaville> does here.REQUEST work?
[18-Jun-2009 10:48:17] <venturaville> >>> d = devices.findDevice('emon8snc1')
[18-Jun-2009 10:48:17] <venturaville> >>> d.REQUEST
[18-Jun-2009 10:48:17] <venturaville> <HTTPRequest, URL=http://localhost:8080>
[18-Jun-2009 10:48:19] <venturaville> ahah!
[18-Jun-2009 10:50:07] <ckrough> cool thanks
[18-Jun-2009 11:26:46] <blocky> anybody know why the scale of my load graphs doesn't match up with the rrd data
[18-Jun-2009 11:27:01] <blocky> im using the uptime parser
[18-Jun-2009 11:29:12] <blocky> all my rrd numbers are between 0 and 1 and the min/max on my graph is set correctly, but but graph is scaled up past 200m in some cases
[18-Jun-2009 11:30:55] <ckrough> m is micro
[18-Jun-2009 11:31:09] <rocket> hey matt
[18-Jun-2009 11:31:25] <ckrough> rocket: you scared
[18-Jun-2009 11:31:28] <ckrough> him off
[18-Jun-2009 11:31:34] <rocket> uh oh
[18-Jun-2009 11:31:52] <FooDog> m is milli, not micro
[18-Jun-2009 11:32:08] <ckrough> sorry, lol yeah
[18-Jun-2009 11:32:13] <blocky> i thought its mega
[18-Jun-2009 11:32:17] <ckrough> M is mega
[18-Jun-2009 11:32:27] <ckrough> case sensitive
[18-Jun-2009 11:32:28] <blocky> oh right
[18-Jun-2009 11:32:41] <blocky> so 200m is just 0.2 load?
[18-Jun-2009 11:32:48] <FooDog> correct
[18-Jun-2009 11:33:11] <blocky> is there a way I can get it to display that way?
[18-Jun-2009 11:33:26] <FooDog> (don't worry, that confused me forever also)
[18-Jun-2009 11:33:32] <ckrough> same
[18-Jun-2009 11:33:42] <rocket> blocky: its the formatting of the graph
[18-Jun-2009 11:33:42] <FooDog> i think that's an .rrd thing, not sure how to change it
[18-Jun-2009 11:33:55] <rocket> have to edit it in the template
[18-Jun-2009 11:34:10] <blocky> hmm ok
[18-Jun-2009 11:34:14] <rocket> and yes it is an rrd thing
[18-Jun-2009 11:34:29] <blocky> is there a way i can make 1 report composed of several smaller reports concatenated
[18-Jun-2009 11:35:31] <rocket> need to look at the rrd formatting options for the graph point in question
[18-Jun-2009 11:35:39] <rocket> eg look for something like %5.1lf%%
[18-Jun-2009 11:37:59] <blocky> turns out the min/max on my graph wasn't set correctly :S
[18-Jun-2009 11:38:09] <nzle> Can I restore a zenoss appliance config backup into a stack installed zenoss temporarily, and move it back to an appliance again? Some googling seemed to indicate there were some gotchas but didnt elaborate much.
[18-Jun-2009 11:38:48] <rocket> probably can do so .. I would guess that you just do the backup restore procedure etc
[18-Jun-2009 11:38:50] <nzle> also, having done a config backup on the appliance, how does one get it off the appliance. Is there a shell login?
[18-Jun-2009 11:38:57] <rocket> though I havent tried it myself ..
[18-Jun-2009 11:39:17] <rocket> nzle: there should be a root login
[18-Jun-2009 11:39:58] <nzle> One would assume, but I havent found yet where the root password is documented..
[18-Jun-2009 11:44:35] <cain22> if you bought the appliance, you could ask them...
[18-Jun-2009 11:45:16] <rocket> nzre it might be unset at the moment
[18-Jun-2009 11:47:05] <mrquick> there's no password on the appliance, just enter root
[18-Jun-2009 11:50:32] <blocky> how do i add other elements to a custom report besides a graph
[18-Jun-2009 11:50:44] <nzle> hmm, thats comforting... I asked my coworker in the other location to fire it up again for me.. Ill try that when he gets back to me. I'm trying to move the appliance to the main location that has access to all the necessary networks, but the box I wanted to run it on till we get our esxi server, is a xen box. Havent found a good way to run a vmware appliance unde xen yet so thinking in terms of restoring the config to a stack install. co worker want
[18-Jun-2009 11:50:44] <nzle> s to stay with the appliance though.
[18-Jun-2009 11:52:52] <blocky> if i set an rrd data source to a gauge, will it do any averaging of numbers
[18-Jun-2009 11:53:05] <blocky> cause my data source can only be integers but there's a decimal value in my rrd
[18-Jun-2009 11:58:17] <cain22> hey, was anyone else getting spammed by zenoss-jobs earlier today?
[18-Jun-2009 11:58:37] <mrquick> yeah, they blogged about it
[18-Jun-2009 12:03:50] <cain22> ah
[18-Jun-2009 12:10:37] <francois> Hi, is anyone aware of issues with SNMP modeling in 2.4.1 ?
[18-Jun-2009 12:11:13] <francois> the issue was described in the forum
[18-Jun-2009 12:11:16] <francois> http://forums.zenoss.com/viewtopic.php?t=10016&start=15
[18-Jun-2009 12:15:59] <ckrough> hmm. well, thats not good
[18-Jun-2009 12:16:21] <francois> Sorry, I have to leave right now
[18-Jun-2009 12:16:40] <francois> if you have any advice, please write it in the forum, thanks !
[18-Jun-2009 12:45:15] <cain22> anyone know when the next update comes out?
[18-Jun-2009 12:46:51] crow is now known as ckrouh
[18-Jun-2009 12:46:54] ckrouh is now known as ckrough
[18-Jun-2009 12:51:56] <rmatte> guys, I've got a really strange problem here that suddenly popped up at 2am this morning...
[18-Jun-2009 12:52:12] <rmatte> when I perform an snmpwalk on a firewall the requests go through looking like:
[18-Jun-2009 12:52:21] <rmatte> 13:45:27.351766 IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto UDP (17), length 87) 30.30.22.10.49304 > calibmfw01.superiorpropane.com.snmp: { SNMPv1 C=TuE892tjIt0Rg5a { GetNextRequest(35) R=1984145409 ip.ipAddrTable.ipAddrEntry.ipAdEntReasmMaxSize.209.82[|snmp] } }
[18-Jun-2009 12:52:25] <rmatte> notice the length of 87
[18-Jun-2009 12:52:31] <ckrough> do you hear a deep voice chanting "wiiiiinnnsstoooonn" from the server room?
[18-Jun-2009 12:52:47] <rmatte> lol
[18-Jun-2009 12:53:10] <rmatte> now, when Zenoss performs an SNMP poll on 4 specific OIDs it comes back with:
[18-Jun-2009 12:53:17] <rmatte> 13:48:36.229868 IP (tos 0x0, ttl 64, id 0, offset 0, flags [DF], proto UDP (17), length 596) 30.30.22.10.60701 > calibmfw01.superiorpropane.com.snmp: { SNMPv1 C=TuE892tjIt0Rg5a { GetRequest(540) R=2017622931 system[|snmp] } }
[18-Jun-2009 12:53:24] <rmatte> note the length of 596
[18-Jun-2009 12:53:51] <rmatte> The firewall is dropping the packets and not responding to the SNMP requests
[18-Jun-2009 12:54:03] <rmatte> it is also sending us traps with messages like:
[18-Jun-2009 12:54:04] <rmatte> calibmfw01.superiorpropane.com|SYSLOG-4|/Cisco/Syslog||4|20-Syslog Trap: 212005: incoming SNMP request (568 bytes) from IP address 30.30.22.10 Port 35294 Interface \"inside\" exceeds data buffer size, discarding this SNMP request.
[18-Jun-2009 12:54:43] <rmatte> I'm wondering why it's rejecting these packets
[18-Jun-2009 12:54:53] <rmatte> if I do snmpwalks for the exact same OIDs it works fine
[18-Jun-2009 12:55:10] <rmatte> It only happens when Zenoss runs through it's polling cycle
[18-Jun-2009 12:56:01] <ckrough> do you have a max packet size set on the firewall
[18-Jun-2009 12:56:06] <ckrough> per service
[18-Jun-2009 12:56:16] <rmatte> it's possible, I'll have to check the config
[18-Jun-2009 12:56:40] <ckrough> http://forums.zenoss.com/viewtopic.php?p=2409&sid=df1870792570ca1699a5f2225324f328
[18-Jun-2009 12:57:34] <ckrough> looks like cisco default max is 512 for that
[18-Jun-2009 12:57:48] <ckrough> reduce the max # of OIDs per request and see if it continues
[18-Jun-2009 12:58:08] <ckrough> there is a zproperty for it
[18-Jun-2009 12:58:36] <rmatte> ah
[18-Jun-2009 12:58:55] <ckrough> zMaxOIDPerRequest
[18-Jun-2009 12:59:44] <rmatte> I'll modify that, thanks for the help
[18-Jun-2009 13:00:20] <ckrough> np, let me know if it doesnt work
[18-Jun-2009 13:00:27] <rmatte> will do
[18-Jun-2009 13:01:12] <rmatte> man I can't wait until they fix this zproperties issue
[18-Jun-2009 13:01:27] <ckrough> which issue is that
[18-Jun-2009 13:01:47] <rmatte> http://dev.zenoss.com/trac/ticket/5071
[18-Jun-2009 13:01:52] <rmatte> that one
[18-Jun-2009 13:02:02] <rmatte> takes like 20 minutes to load the zProperties page sometimes
[18-Jun-2009 13:02:07] <ckrough> ouch
[18-Jun-2009 13:03:39] <rmatte> actually, I don't think "mlunt" understands the situation, breaking it in to multiple smaller Zenpacks wouldn't help, since it's also caused by having a lot of small Zenpacks installed
[18-Jun-2009 13:03:49] <rmatte> the more Zenpack data overall, the slower it goes
[18-Jun-2009 13:04:08] <rmatte> I'll have to point that note out to Matt next time I see him
[18-Jun-2009 13:04:13] <blocky> is there any way I can access the zenoss bindings from vanilla python
[18-Jun-2009 13:04:27] <ckrough> blocky: bindings?
[18-Jun-2009 13:05:19] <blocky> well, like the way i can access devices and their rrd info
[18-Jun-2009 13:05:37] <ckrough> blocky: are you aware of http://www.zenoss.com/Members/netdata/zendmd-tips-and-tricks/
[18-Jun-2009 13:06:11] <ckrough> meh, wrong page
[18-Jun-2009 13:06:30] <ckrough> but, zendmd... is that what you are looking for
[18-Jun-2009 13:06:45] <blocky> yea thats waht im using
[18-Jun-2009 13:07:00] <blocky> but i dont get how zendmd is different from python
[18-Jun-2009 13:07:07] <rmatte> so you want a python script that runs as though it were in zendmd?
[18-Jun-2009 13:07:19] <blocky> yea
[18-Jun-2009 13:07:30] <rmatte> http://www.zenoss.com/Members/netdata/use-a-zendmd-script-as-standalone-program/
[18-Jun-2009 13:07:31] <rmatte> voila
[18-Jun-2009 13:07:52] <rmatte> that's all that's different, throw that at the top of your script and you're sailing
[18-Jun-2009 13:08:33] <rmatte> the script needs to be run as the zenoss user to be able to work
[18-Jun-2009 13:09:40] <blocky> cool
[18-Jun-2009 13:09:47] <blocky> so lets say i want to access devicse
[18-Jun-2009 13:11:19] <rmatte> device = find("192.168.0.1")
[18-Jun-2009 13:11:23] <rmatte> device.whatever
[18-Jun-2009 13:12:11] <rmatte> device.getRRDValue('ssCpuSystem', default)
[18-Jun-2009 13:12:24] <rmatte> device.getRRDValue('ssCpuSystem', function="AVERAGE")
[18-Jun-2009 13:12:28] <rmatte> device.getRRDValue('ssCpuSystem', function="MAX")
[18-Jun-2009 13:12:41] <blocky> find?
[18-Jun-2009 13:13:03] <blocky> sorry im quite new at this
[18-Jun-2009 13:13:07] <rmatte> you're basically finding the path to the device with that IP or hostname, and then assigning the path to a variable
[18-Jun-2009 13:13:09] <blocky> all that tutorial gives me is dmd
[18-Jun-2009 13:13:18] <blocky> i dont have a find() function defind
[18-Jun-2009 13:13:23] <rmatte> zendmd is python
[18-Jun-2009 13:13:41] <rmatte> oh, you need to load certain functions for your script
[18-Jun-2009 13:13:48] <rmatte> hold on, I forget how I did it...
[18-Jun-2009 13:13:52] <blocky> where is find heh
[18-Jun-2009 13:14:17] <blocky> in zendmd the syntax is like dev = devices.findDevice('mag1')
[18-Jun-2009 13:14:43] <rmatte> you can do it the way I just said as well
[18-Jun-2009 13:14:47] <rmatte> have you tried it?
[18-Jun-2009 13:15:52] <rmatte> devices.finddevice would probably work better in terms of performance
[18-Jun-2009 13:15:52] <blocky> NameError: name 'find' is not defined
[18-Jun-2009 13:15:59] <blocky> i dont have devices either...
[18-Jun-2009 13:16:00] <rmatte> what are you typing exactly?
[18-Jun-2009 13:16:07] <rmatte> oh, you're just trying to run it as a script?
[18-Jun-2009 13:16:17] <blocky> im typing it into the interpreter
[18-Jun-2009 13:16:41] <rmatte> try dmd.devices.findDevice('mag1')
[18-Jun-2009 13:16:43] <blocky> oh it's findDevice
[18-Jun-2009 13:17:21] <rmatte> you might have to prefix it with dmd.
[18-Jun-2009 13:17:50] <rmatte> I use the line "for d in dmd.Devices.getSubDevices():" in one of my scripts and it works fine
[18-Jun-2009 13:18:27] <rmatte> that starts a for loop where it loops through each device in the database and sets d to the path to that device as it goes through
[18-Jun-2009 13:18:32] <blocky> cool its working
[18-Jun-2009 13:18:36] <rmatte> cool
[18-Jun-2009 13:18:41] <blocky> thanks
[18-Jun-2009 13:18:43] <rmatte> np
[18-Jun-2009 13:18:53] <blocky> now i wonder how i can get splunk involved
[18-Jun-2009 13:19:50] <rmatte> I've no knowledge of spunk, so you're on your own with that one hehe
[18-Jun-2009 13:20:13] <blocky> yea its got a python interface too
[18-Jun-2009 13:20:46] <rmatte> splunk* rather
[18-Jun-2009 13:21:11] <rmatte> I guess that's why it integrates so well with Zenoss
[18-Jun-2009 13:22:15] <blocky> well it will when im through with it
[18-Jun-2009 13:23:16] <ckrough> blocky: are you trying to go from splunk to zen or the other direction
[18-Jun-2009 13:23:34] <venturaville> splunk is good for summarizing app events for pushing up to something
[18-Jun-2009 13:23:43] <venturaville> so most likely splunk -> zen
[18-Jun-2009 13:24:00] <ckrough> yeah, thats what we do
[18-Jun-2009 13:24:12] <ckrough> devices -> syslog -> splunk
[18-Jun-2009 13:24:35] <venturaville> anyone had much experience with having splunk look directly at their logs?
[18-Jun-2009 13:24:53] <ckrough> venturaville: I have it following a file, yeah
[18-Jun-2009 13:24:56] <ckrough> no problems
[18-Jun-2009 13:25:09] <venturaville> how much data can it handle per second?
[18-Jun-2009 13:25:40] <ckrough> lemme calculate
[18-Jun-2009 13:28:11] <ckrough> Im only doing around 1mil lines per day, which is around 12 lines per secon
[18-Jun-2009 13:30:16] <ckrough> and its not dropping anything
[18-Jun-2009 13:30:16] <zenChild> I have a 1-to-many relationship established between a device and components. Is it possible to then have a 1-to-many relationship between each component and subcomponents? How does the modeler tie each subcomponent back to each component. At a high level you can think of it like so: a host can have multiple volume groups and each volume group can in turn have multiple logical volumes.
[18-Jun-2009 13:30:22] <ckrough> never hit its max
[18-Jun-2009 13:30:51] <blocky> yea its splunk to zen
[18-Jun-2009 13:31:07] <blocky> id like to get a summary in my report, stuff like how many failed logins
[18-Jun-2009 13:31:22] <blocky> need to know who's been accessing the squid proxy
[18-Jun-2009 13:32:04] <blocky> do custom zen reports only make tables with a row for each device
[18-Jun-2009 13:32:53] <ckrough> You may need to start doing custom reports: http://www.zenoss.com/Members/cluther/creating-methods-for-custom-reports/
[18-Jun-2009 13:33:11] <blocky> yea ive got a few of those methods
[18-Jun-2009 13:33:53] <blocky> but then how do i call a method from a report with something more flexible than a list of columns
[18-Jun-2009 13:34:24] <ckrough> gotta build your own report templates
[18-Jun-2009 13:34:25] <ckrough> sec
[18-Jun-2009 13:34:40] <ckrough> zenoss/Products/ZenReports
[18-Jun-2009 13:34:45] <ckrough> steal some ideas there
[18-Jun-2009 13:39:16] <ckrough> in the web interface, /zport/portal_skins/zenmodel/manage_main
[18-Jun-2009 13:39:31] <ckrough> there are a bunch of reports in there that you can copy/modify as well
[18-Jun-2009 14:05:16] <venturaville> ok so I am trying to trigger a loadDevice via zendmd
[18-Jun-2009 14:05:22] <venturaville> but it isn't showing up in the GUI
[18-Jun-2009 14:05:29] <venturaville> I can query it via zendmd though ......
[18-Jun-2009 14:05:37] <venturaville> any ideas?
[18-Jun-2009 14:08:14] <ckrough> search for it via IP in the gui?
[18-Jun-2009 14:08:23] <venturaville> it finally showed up
[18-Jun-2009 14:08:25] <venturaville> weird
[18-Jun-2009 14:08:34] <venturaville> can't seem to get this to work in the event class transform though ............
[18-Jun-2009 14:10:37] <venturaville> I am doing this in shell on a box: deviceName="myhostname" ; deviceClass="/Server/Linux" ; logger -p warn -t addtomonitoring "AddMeToMonitoring deviceName=${deviceName} deviceClass=${deviceClass}"
[18-Jun-2009 14:10:50] <venturaville> and have an event class that matches it
[18-Jun-2009 14:11:06] <venturaville> regex:
[18-Jun-2009 14:11:08] <venturaville> AddMeToMonitoring deviceName=(?P<mydeviceName>.*) deviceClass=(?P<mydeviceClass>.*)
[18-Jun-2009 14:11:20] <venturaville> transform:
[18-Jun-2009 14:11:22] <venturaville> dmd.DeviceLoader.loadDevice(evt.mydeviceName, evt.mydeviceClass)
[18-Jun-2009 14:11:46] <venturaville> seems to work in zendmd
[18-Jun-2009 14:11:53] <venturaville> but never does in events....
[18-Jun-2009 14:13:41] <venturaville> maybe I need to do a commit() in there??
[18-Jun-2009 14:19:25] <venturaville> basically trying to use dmd inside of an event.........
[18-Jun-2009 14:27:19] <venturaville> I guess what I am really asking is how to use dmd inside of an event............
[18-Jun-2009 16:00:45] <phasma> Hello. Can anybody give a link to manual, how create multi-graph report based on custom MIB OID ?
[18-Jun-2009 16:07:45] <ckrough> phasma: possibly read up on this: http://www.zenoss.com/community/docs/zenoss-dev-guide/2.4.0/ch12.html
[18-Jun-2009 18:12:51] * kisielk wishes the developer docs were better
[19-Jun-2009 00:00:42] [disconnected at Fri Jun 19 00:00:42 2009]
[19-Jun-2009 00:00:42] [connected at Fri Jun 19 00:00:42 2009]
[19-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[19-Jun-2009 04:17:11] <Dieterbe> Hi, I have a debian lenny 64bit. it's a brand new installed box. When i try to install zenoss on it (by using the .deb package for debian) I get errors
[19-Jun-2009 04:17:15] <Dieterbe> See: http://pastebin.com/m20855ac6
[19-Jun-2009 04:17:41] <Dieterbe> I tried both wgetting the deb for your site and dpkg -i, as well as adding your mirror and using aptitude. as you can see in the paste above
[19-Jun-2009 04:48:55] <Dieterbe> hmm seems like someone had the same problem on june 3rd. http://de.pastebin.ca/1446692
[19-Jun-2009 07:57:12] crow is now known as ckrough
[19-Jun-2009 08:05:07] <ckrough> morniing
[19-Jun-2009 08:09:05] <rocket> morning
[19-Jun-2009 09:49:27] <rocket> morning matt
[19-Jun-2009 09:49:46] <venturaville> mrayzenoss: ... just the person I was hoping to see
[19-Jun-2009 09:49:58] <mrayzenoss> good morning
[19-Jun-2009 09:50:23] <venturaville> mrayzenoss: I was hoping you knew how to get access to the dmd object from inside of an event class......
[19-Jun-2009 09:51:57] <mrayzenoss> let me check, I think there's an example of it somewhere
[19-Jun-2009 09:58:52] <mrayzenoss> venturaville: what are you trying to do?
[19-Jun-2009 10:03:28] <mrayzenoss> if you install the TransformReport ZenPack, there's a nice report that shows you every transform deployed on your box. Good for seeing examples
[19-Jun-2009 10:08:45] <venturaville> mrayzenoss: I'm trying to have a syslog entry from a host trigger a loadDevice .......
[19-Jun-2009 10:08:54] <venturaville> basically trying to automate addition of hosts from puppet
[19-Jun-2009 10:09:08] <mrayzenoss> how's the puppet Zenoss module?
[19-Jun-2009 10:09:18] <venturaville> eh?
[19-Jun-2009 10:09:31] <mrayzenoss> http://blog.zenoss.com/2009/01/13/new-zenoss-module-for-puppet/
[19-Jun-2009 10:09:47] <mrayzenoss> I assume you were using it
[19-Jun-2009 10:09:51] <venturaville> nope
[19-Jun-2009 10:09:57] <venturaville> though I have looked at it
[19-Jun-2009 10:10:07] <venturaville> it doesn't quite match what I need done though
[19-Jun-2009 10:10:10] <mrayzenoss> ok
[19-Jun-2009 10:11:25] <rocket> venturaville: what is it missing?
[19-Jun-2009 10:15:21] <zenChild> Anyone ever have an issue testing a SSH command for datasource where you get "command not found"? It seems as though it is trying to run the command locally even though I am testing against a remote system in which that command does indeed exist.
[19-Jun-2009 10:15:56] <rocket> zenChild: path issues?
[19-Jun-2009 10:16:10] <venturaville> rocket: it is a bit of a infrastructural issue here... not so a much a problem with the package :-)
[19-Jun-2009 10:16:14] <rocket> zenChild: eg did you have the full path to the command? /usr/sbin/foo etc
[19-Jun-2009 10:16:27] <zenChild> it's an appliance and the command is in the default path.
[19-Jun-2009 10:16:38] <zenChild> my modeler runs the same command and it succeeds.
[19-Jun-2009 10:19:33] <zenChild> is there a way i can test the datasource command on the command line so I can see the debug output?
[19-Jun-2009 10:20:21] <zenChild> I tried running...: zentestcommand -d <mydev> --datasource=lsmdiskgrp
[19-Jun-2009 10:20:39] <zenChild> but I get: No datasource lsmdiskgrp applies to device <mydev>
[19-Jun-2009 10:21:01] <zenChild> I think that's because it is a component datasource
[19-Jun-2009 10:21:37] <mrayzenoss> zenChild: http://dev.zenoss.org/trac/ticket/5175
[19-Jun-2009 10:21:54] <mrayzenoss> that's not for SSH, but the 'TEST' button seems to suck
[19-Jun-2009 10:22:02] <mrayzenoss> you can test with SSH though
[19-Jun-2009 10:22:18] <mrayzenoss> zencommand run -v10 device
[19-Jun-2009 10:24:32] <zenChild> Thanks mrayzenoss. I must be missing something somewhere because I don't see any evidence of collection for my datasource. The template is binding to the individual components though so I'm not sure where it's going wrong.
[19-Jun-2009 10:24:54] <mrayzenoss> yeah, that's what is nuts
[19-Jun-2009 10:24:58] <mrayzenoss> the test will work
[19-Jun-2009 10:25:11] <mrayzenoss> and the zencommand log won't show anything, just the path of the command it's trying to run
[19-Jun-2009 10:25:27] <mrayzenoss> which isn't exactly your issue, just a heads up
[19-Jun-2009 10:26:01] <zenChild> I should see something in the output of zencommand though shouldn't I... or do I need to peek through the logs?
[19-Jun-2009 10:27:44] <mrayzenoss> well, the zencommand with the -v10 will get you full debugging
[19-Jun-2009 10:27:48] <zenChild> ahah! I found my issue in the zenhub.log.
[19-Jun-2009 10:27:50] <mrayzenoss> which won't be in the logs
[19-Jun-2009 10:28:00] <zenChild> It's not finding my parser correctly.
[19-Jun-2009 10:30:54] <venturaville> mrayzenoss: I can't find a working example of dmd usage in transforms
[19-Jun-2009 10:31:41] <zenChild> I don't know if anyone has an IBM SVC out there, but I'm trying to write a ZenPack to monitor it. I have all the scripts to parse the performance data previously written in Perl and am in the process of cleansing them before I put them on the project repository. If anyone wants to help out or just test you can grab the code here: http://github.com/zenchild/ZenPacks.ndgov.IBMsvc/tree/master
[19-Jun-2009 10:32:05] <venturaville> zenChild: I've got SVC monitoring on my list to work on as well........
[19-Jun-2009 10:32:41] <zenChild> nice. maybe we can collaborate. Do you have anything written yet?
[19-Jun-2009 10:33:01] <venturaville> not a thing... its on my list to work on next month
[19-Jun-2009 10:33:41] <zenChild> I'd love some help. I come from a perl/ruby background so python is beating up on me a bit.
[19-Jun-2009 10:34:01] <venturaville> I'm working on a way to monitor the IBM arrays behind them as well
[19-Jun-2009 10:34:21] <venturaville> I got some docs from the engineers at LSI about how to parse out the SMcli commands for that too
[19-Jun-2009 10:36:03] <venturaville> I was looking at the CIM agents to try and get performance data out of them
[19-Jun-2009 10:36:12] <venturaville> (the SVCs)
[19-Jun-2009 10:36:18] <venturaville> but my knowledge of CIM is lacking.....
[19-Jun-2009 10:36:20] <zenChild> That would be sweet. For now I'm mostly interested in the SVC, but getting the DSx controllers would be an added bonus. We also just got in an XIV and I'd love to be able to import that data into Zenoss. Anyway, when you're ready to start working on the SVC let me know. In the meantime I'll keep plugging away.
[19-Jun-2009 10:36:47] <zenChild> CIM will be v2.0 of the ZenPack
[19-Jun-2009 10:37:27] <zenChild> I was looking at CIM for getting info out of ESXi directly. It can become a spiderweb rather quickly.
[19-Jun-2009 10:39:32] <venturaville> I've been trying to use wbemcli to talk to it
[19-Jun-2009 10:39:40] <venturaville> but the docs for it are lacking
[19-Jun-2009 10:39:47] <venturaville> and they are mostly all windows specific
[19-Jun-2009 10:40:09] <zenChild> That's what I was using too. Have you looked at yawn for browsing it?
[19-Jun-2009 10:40:16] <venturaville> got a link?
[19-Jun-2009 10:40:30] <zenChild> http://pywbem.wiki.sourceforge.net/YAWN
[19-Jun-2009 10:40:52] <zenChild> It came in handy when I was going through VMware's CIM docs.
[19-Jun-2009 10:41:02] <zenChild> I could kind of follow along...
[19-Jun-2009 10:41:25] <venturaville> much nicer
[19-Jun-2009 10:44:10] <mrayzenoss_> venturaville: zenChild: Don't know if this is useful or not, http://www.zenoss.com/community/wiki/cim-wbem/
[19-Jun-2009 10:44:27] <mrayzenoss_> Kells started on a CIM ZenPack for 2.1, haven't looked at it yet
[19-Jun-2009 10:45:24] <venturaville> looked interesting, and I played with it, but without knowing much about CIM and better docs on it I can't use it
[19-Jun-2009 10:45:31] <zenChild> Thanks. I'll have a look. Now if we can just get IBM to publish a CIMOM map of the SVC ...
[19-Jun-2009 10:45:42] <venturaville> should not need it in theory
[19-Jun-2009 10:45:51] <venturaville> CIMOM is somewhat self documenting
[19-Jun-2009 10:46:08] <venturaville> tcpdump is helpful too........................
[19-Jun-2009 10:46:15] <zenChild> still makes it nice to look at when you're coding...
[19-Jun-2009 10:46:51] <zenChild> i guess that's what YAWN attempts to do....
[19-Jun-2009 12:31:23] <ckrough> is this in seconds? "Cache Timeout - Specify the cache timeout value for the event monitor."
[19-Jun-2009 12:32:54] <ckrough> From the descriptions, I dont get a clear indication of what happens when these thresholds are reached: http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch14s02.html#d4e5211, "2.3. Changing Event Manager Cache Settings"
[19-Jun-2009 12:34:12] <blocky> can somebody point me to a tutorial for creating a custom graph template?
[19-Jun-2009 12:34:32] <ckrough> blocky: you mean fully custom, like we were discussing yesterday?
[19-Jun-2009 12:34:54] <ckrough> blocky: I've found bits and pieces that help you along, but I havent found a full blown how-to
[19-Jun-2009 12:35:09] <ckrough> blocky: but Im notoriously bad at finding things on zenoss.com
[19-Jun-2009 12:37:03] <blocky> fully custom yea
[19-Jun-2009 12:37:09] <blocky> ive got a report where I can edit the html
[19-Jun-2009 12:37:15] <blocky> but I dont know how to access content
[19-Jun-2009 12:37:31] <ckrough> TALES Expressions
[19-Jun-2009 12:37:48] <blocky> ive defined a device template with a load and a network graph
[19-Jun-2009 12:37:59] <blocky> and ive got a report for 5 devices that shows one of each for a total of 10 graphs
[19-Jun-2009 12:38:13] <ckrough> ok
[19-Jun-2009 12:38:17] <blocky> now if i could insert that entire report along with a few others...
[19-Jun-2009 12:39:17] <ckrough> Not sure
[19-Jun-2009 12:39:30] <ckrough> rocket or mray can probably point you in the right direction if you see them on
[19-Jun-2009 12:40:12] <blocky> forums.zenoss.com is failing to resolve now
[19-Jun-2009 12:40:24] <ckrough> working for me
[19-Jun-2009 12:41:08] <rocket> blocky: cant access device content? eg trying to populate datapoints?
[19-Jun-2009 12:41:42] <blocky> well no my RRDs are being collected fine
[19-Jun-2009 12:42:35] <blocky> initially what i'd like to do is agreggate a few different reports in one, so i can email it off in a nice neat package without requiring separate emails
[19-Jun-2009 12:44:23] <blocky> it seems like there's two different ways of defining a report
[19-Jun-2009 12:44:28] <blocky> pure html with tales
[19-Jun-2009 12:44:31] <rocket> blocky: so a multigraph report?
[19-Jun-2009 12:44:59] <blocky> well I'd like to have more than graphs on it, if that's possible..
[19-Jun-2009 12:45:21] <ckrough> It sounds like you want to merge a couple different types of reports
[19-Jun-2009 12:45:57] <blocky> ideally yes but i wasn't sure if that was possible, so I thought attaching them together would be almost as good
[19-Jun-2009 12:47:07] <ckrough> Can anyone explain what happens when these thresholds are reached: http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch14s02.html#d4e5211, "2.3. Changing Event Manager Cache Settings"
[19-Jun-2009 12:47:16] <rocket> blocky: I havent done too much with reports yet but I believe you would still be able to do that with a multigraph report .. you might have to do some python and html editing to get exactly what you are after
[19-Jun-2009 12:47:35] <blocky> what if not all the data im after can be accessed on every device
[19-Jun-2009 12:47:48] <blocky> like, they are all linux boxes so they all have load and network stats
[19-Jun-2009 12:47:57] <rocket> blocky: most likely it will be blank or show up as n/a
[19-Jun-2009 12:48:08] <ckrough> blocky: Make sure they (zenoss) know what you want. I am frequently pushing for improved reporting too. more voices = more priority
[19-Jun-2009 12:48:09] <blocky> but only one is a squid proxy, so only one has the list of denied IPs
[19-Jun-2009 12:48:41] <blocky> oaky who do I mass email flood
[19-Jun-2009 12:48:50] <rocket> blocky: thats where the python probably will come in .. might need to write a function that shows it if available etc .. or shows na if not
[19-Jun-2009 12:49:39] <blocky> that shouldn't be too bad
[19-Jun-2009 12:49:46] <ckrough> somewhere in the forums Mray has a feature request thread
[19-Jun-2009 12:49:57] <blocky> im not 100% sure what I want yet
[19-Jun-2009 12:50:08] <blocky> i mean, i know what i want the result to be, but not how it would fit with zenoss
[19-Jun-2009 12:50:15] <ckrough> ckrough: I know what you mean
[19-Jun-2009 12:50:50] <ckrough> I want them to object-orient the reporting system. I want to define a graph called "x", then be able to re-use "x" in any report where it makes sense
[19-Jun-2009 12:51:04] <ckrough> right now I have to redefine everything for every graph, across several zenoss installs. tedious
[19-Jun-2009 12:51:11] <blocky> i guess i could write a python script that generates all the HTML and hard code the graph URLs
[19-Jun-2009 12:52:03] <rocket> blocky: thats a possibility .. but doesnt sound very flexible but works for quick and dirty
[19-Jun-2009 12:52:16] <blocky> i dunno if reportmail would be able to use that, it might make more sense to send the email myself
[19-Jun-2009 12:52:19] <ckrough> blocky: we use that method for projectors in the NOC, etc...
[19-Jun-2009 12:53:51] <blocky> i like the way I can define data sources and threshholds and graphs at any level in the device hierarchy
[19-Jun-2009 12:53:57] <blocky> and have them be applied to each device
[19-Jun-2009 12:54:08] <rocket> blocky: did you look at the dev guide for reports?
[19-Jun-2009 12:54:19] <rocket> http://www.zenoss.com/community/docs/zenoss-dev-guide/2.4.0/ch14s01.html
[19-Jun-2009 12:55:22] <blocky> i saw that, i dont really know what it means
[19-Jun-2009 12:55:43] <blocky> can TALES markup access the other zenoss reports ive already defined, or the graphs
[19-Jun-2009 12:56:49] <rocket> blocky: examples are on the server under Products/ZenReports
[19-Jun-2009 12:57:02] <blocky> like how is a custom device report related to a custom graph report
[19-Jun-2009 12:57:47] <rocket> blocky: is there an existing report similar to what you are trying to accomplish?
[19-Jun-2009 12:57:47] <blocky> one is a table of python methods vs devices, the other is a list of which graphs
[19-Jun-2009 12:58:50] <blocky> if I could get a custom graph report and several custom device reports on the same page, and each based on a different list of devices, that would go a long way
[19-Jun-2009 12:59:02] <blocky> i could write the python myself for the device report methods
[19-Jun-2009 12:59:25] <rocket> eg you want to mix a Performance_Report with a Device_Report
[19-Jun-2009 12:59:49] <blocky> yea
[19-Jun-2009 12:59:58] <ckrough> given the way zenoss/zope generate reports based on objects within a tree, I am not sure how you could do that within a single report in zenoss
[19-Jun-2009 13:00:14] <blocky> eventually id like to be able to add content which is wholly external, like from splunk
[19-Jun-2009 13:00:20] <rocket> ckrough: your report collects groups of objects together
[19-Jun-2009 13:00:58] <ckrough> rocket: Im kinda stuck at the same point as blocky , how did you get past this stage?
[19-Jun-2009 13:01:04] <blocky> sorry where is Products/ZenReports
[19-Jun-2009 13:01:10] <rocket> ckrough: and you process them in different tables
[19-Jun-2009 13:01:34] <ckrough> blocky: in the filesystem
[19-Jun-2009 13:01:43] <ckrough> $zenhome/Products...
[19-Jun-2009 13:02:54] <ckrough> some reports are zope objects, which you can access through the /manage interface in the gui, others are on the filesystem. I believe I read that it's recommended to create file based reports over zope objects.
[19-Jun-2009 13:03:12] <ckrough> blocky: Im assuming you know about /manage in the gui, yes?
[19-Jun-2009 13:03:27] <blocky> yea ive had a look at it
[19-Jun-2009 13:03:47] <blocky> it does shed a bit of light
[19-Jun-2009 13:03:52] <ckrough> spend some time clicking around on objects in there, it helps quite a bit
[19-Jun-2009 13:03:53] <ckrough> yeah
[19-Jun-2009 13:04:03] <rocket> blocky: ckrough basically you guys need to figure out the syntax of tal expressions as shown in those sample reports
[19-Jun-2009 13:04:34] <rocket> the tal expressions create groups of objects that you loop through calling methods etc
[19-Jun-2009 13:05:04] <ckrough> blocky: /zport/portal_skins/zenmodel <-- this is a goldmine of examples
[19-Jun-2009 13:05:42] <rocket> eg within <tal:block> </tal:block> you can grab any group of objects .. display them etc
[19-Jun-2009 13:06:07] <rocket> define a new set of <tal> blocks for different objects you want to work with in the same html report
[19-Jun-2009 13:06:50] <ckrough> and within those blocks, {here} will be in reference to whatever point in the class tree you ask for the report?
[19-Jun-2009 13:07:06] <ckrough> ie: /devices/network/viewMyReport
[19-Jun-2009 13:07:47] <rocket> gotta run
[19-Jun-2009 13:07:52] <ckrough> thanks, see ya
[19-Jun-2009 13:07:53] <dberger> does anyone know why a location with devices in it might not show up on the google map portlet?
[19-Jun-2009 13:15:08] <mrayzenoss> dberger: on the location, what do you have in the "Address" field?
[19-Jun-2009 13:16:01] <dberger> 1450 Ala Moana Blvd # 3241
[19-Jun-2009 13:16:02] <dberger> Honolulu, HI
[19-Jun-2009 13:16:17] <dberger> it shows up in http://localhost:8080/zport/dmd/Locations/Hawaii/locationGeoMap
[19-Jun-2009 13:17:06] <twm1010> Have you assigned any devices to the location yet?
[19-Jun-2009 13:17:16] <twm1010>
[19-Jun-2009 13:17:27] <dberger> it even shows up in the portlet if i change it to the default location...but then the other location I've entered disappears
[19-Jun-2009 13:17:32] <dberger> yes
[19-Jun-2009 13:17:44] <twm1010> hrmm... clear browser cache and reload page?
[19-Jun-2009 13:18:45] <blocky> any good primers on tal markup?
[19-Jun-2009 13:19:15] <mrayzenoss> blocky: http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/apds02.html
[19-Jun-2009 13:20:27] <dberger> i reload, but clearing browser cache won't do me any good
[19-Jun-2009 13:21:28] <ckrough> Can anyone explain what happens when these thresholds are reached: http://www.zenoss.com/community/docs/zenoss-guide/2.4.0/ch14s02.html#d4e5211, "2.3. Changing Event Manager Cache Settings"
[19-Jun-2009 13:22:30] <ckrough> heh heh: "The lower the number, the more responsible the history will be"
[19-Jun-2009 13:22:31] <mrayzenoss> dberger: ok, so the /Locations/Hawaii/locationGeoMap never works?
[19-Jun-2009 13:23:36] <mrayzenoss> and the portlet is intermittent?
[19-Jun-2009 13:23:42] <dberger> no, that works
[19-Jun-2009 13:23:52] <dberger> but the portlet never displays more than one point
[19-Jun-2009 13:23:59] <dberger> it only displays the base location
[19-Jun-2009 13:24:24] <mrayzenoss> do you have stuff assigned to /Locations/?
[19-Jun-2009 13:25:53] <dberger> the table in /zport/dmd/Locations shows 3 devices under both locations
[19-Jun-2009 13:26:15] <mrayzenoss> what's the other location?
[19-Jun-2009 13:26:34] <dberger> 200 W Madison St
[19-Jun-2009 13:26:35] <dberger> Suite 700
[19-Jun-2009 13:26:35] <dberger> Chicago, IL 60606
[19-Jun-2009 13:26:45] <dberger> named Chicago
[19-Jun-2009 13:27:57] <mrayzenoss> hmm... everything works for me
[19-Jun-2009 13:28:07] <mrayzenoss> 3 devices in 'Chicago', 2 in 'Hawaii'
[19-Jun-2009 13:28:12] <mrayzenoss> 2 in /
[19-Jun-2009 13:28:32] <mrayzenoss> dberger: which version are you running?
[19-Jun-2009 13:29:12] <dberger> 2.4.1-linux-x64
[19-Jun-2009 13:31:35] <mrayzenoss> so Hawaii doesn't render, but Chicago does in the portlet...
[19-Jun-2009 13:33:31] <dberger> yes, unless i make hawaii the default location
[19-Jun-2009 13:36:02] <dberger> in which case i have the exact opposite problem
[19-Jun-2009 13:39:35] <mrayzenoss> dberger: not sure what's wrong since I can't recreate it
[19-Jun-2009 13:40:01] <dberger> does it matter that i don't have any network links setup or anything like that?
[19-Jun-2009 13:40:42] <dberger> this is all just a test installation with several devices on the local network auto-discovered
[19-Jun-2009 13:41:43] <dberger> okay...i think i have it
[19-Jun-2009 13:42:02] <dberger> i see both now when i change the base location to /Locations
[19-Jun-2009 13:42:28] <dberger> i didn't realize "base location" meant tree root
[19-Jun-2009 13:42:50] <dberger> i thought it referred to the location that received focus by default
[19-Jun-2009 13:42:54] <dberger> thanks for your help!
[19-Jun-2009 13:42:57] <mrayzenoss> so it's working when you set it to /Locations?
[19-Jun-2009 13:43:17] <dberger> yes
[19-Jun-2009 13:43:21] <mrayzenoss> cool, you solved your problem and my printer is now located in Hawaii
[19-Jun-2009 13:43:33] <dberger> everybody wins
[19-Jun-2009 13:43:42] <dberger> it's actually located at a pizza shop in hawaii
[19-Jun-2009 13:43:50] <dberger> so started printing up fax orders!
[19-Jun-2009 14:10:11] <ckrough> maybe I could hitch a ride to hawaii with you? http://www.flickr.com/photos/zenoss/3573421972/in/photostream/
[19-Jun-2009 14:10:31] <ckrough> Is that in Annapolis Matt?
[19-Jun-2009 14:11:13] <mrayzenoss> yeah, that was in Annapolis
[19-Jun-2009 14:11:17] <ckrough> Love that place
[19-Jun-2009 14:49:59] <dberger> does anyone know of a way to create/destroy arbitrary network links?
[19-Jun-2009 14:51:13] <mrayzenoss> you can add/delete interfaces and routes on the device
[19-Jun-2009 14:51:48] <mrayzenoss> it'll get re-added on the next modeling unless you lock it
[19-Jun-2009 15:07:12] bret is now known as Bret
[19-Jun-2009 15:07:14] Bret is now known as Bret[Jive]
[19-Jun-2009 15:08:53] <dberger> could this be used to dummy a network?
[19-Jun-2009 15:09:17] <mrayzenoss> absolutely http://www.zenoss.com/Members/netdata/create-manual-device-dependencys/
[19-Jun-2009 15:12:24] <ckrough> hmmm. How does one call a zope python script from within a TAL block; "myVariable container/myScript;"?
[19-Jun-2009 19:46:02] <daMaestro> anyone doing JMX monitoring?
[19-Jun-2009 19:46:45] <daMaestro> we are getting "Error calling mbean solr:type=org.apache.solr.handler.component.SearchHandler: Exception: no MBean registered with name: solr:type=org.apache.solr.handler.component.SearchHandler" when trying to monitor a solr instance
[19-Jun-2009 20:14:52] daMaestro|isBack is now known as daMaestro
[19-Jun-2009 20:25:14] <daMaestro> ah.. it was just a mis-configured bean location
[20-Jun-2009 00:00:42] [disconnected at Sat Jun 20 00:00:42 2009]
[20-Jun-2009 00:00:42] [connected at Sat Jun 20 00:00:42 2009]
[20-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[20-Jun-2009 07:07:04] RoninX341 is now known as etank
[21-Jun-2009 00:00:42] [disconnected at Sun Jun 21 00:00:42 2009]
[21-Jun-2009 00:00:42] [connected at Sun Jun 21 00:00:42 2009]
[21-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[21-Jun-2009 09:22:33] <Appiah_> Do i need to add a networkin under /Networks before the auto-discovery to work for that network?
[21-Jun-2009 09:23:00] <Appiah_> I'm asking beacuse I can't test it myself right now >_>
[22-Jun-2009 00:00:42] [disconnected at Mon Jun 22 00:00:42 2009]
[22-Jun-2009 00:00:42] [connected at Mon Jun 22 00:00:42 2009]
[22-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[22-Jun-2009 08:50:17] <ckrough> morning
[22-Jun-2009 08:51:23] <dberger> scared him off
[22-Jun-2009 08:51:45] <ckrough> cant blame him
[22-Jun-2009 08:59:41] <Dieterbe> hehe
[22-Jun-2009 09:17:38] <ainc> hello all...
[22-Jun-2009 09:17:44] <dberger> morning
[22-Jun-2009 09:18:09] <ckrough> morning
[22-Jun-2009 09:18:28] <ckrough> morning matt
[22-Jun-2009 09:18:34] <ainc> is there a way to set a threshold to only send you a alarm when it has hit the limit like say, 6 times in a row
[22-Jun-2009 09:19:03] <rocket> Morning Matt ...
[22-Jun-2009 09:19:31] <ckrough> ainc: you can specify count in the alerting rule
[22-Jun-2009 09:20:32] <dberger> you can also make it escalate severity and set your notification rule to a higher severity
[22-Jun-2009 09:20:35] <ckrough> ainc: this will not work if your threshold alert auto-clears and come back repeatedly.
[22-Jun-2009 09:22:56] <ckrough> we are lookin for a way to alert on a bouncin threshold. no solution at the moment
[22-Jun-2009 09:27:39] <ainc> sorry, I don't want it to create an event in the viewer unless it hits a count...not an alarm
[22-Jun-2009 09:37:21] <clustermagnet> gents, i am evaluating monitoring products
[22-Jun-2009 09:38:29] <clustermagnet> i am trying to understand the difference between your free product, and the enterprise license
[22-Jun-2009 09:38:44] <clustermagnet> other than better vmware support, what are the differences?
[22-Jun-2009 09:39:01] <clustermagnet> im searching on your site, havent found a chart that would make it easy to compare
[22-Jun-2009 09:39:04] <clustermagnet> thanks!
[22-Jun-2009 09:39:18] <ckrough> clustermagnet: there may not be any official zenoss reps activein the room at the moment.
[22-Jun-2009 09:40:26] <clustermagnet> ckrough: no problem, dont care about official/nonofficial
[22-Jun-2009 09:40:42] <clustermagnet> ckrough: can you key in?
[22-Jun-2009 09:40:46] <slestak> i need to monitor a multithreaded jobscheduler in my aix based erp system
[22-Jun-2009 09:41:03] <ckrough> been so long since I used CORE... trying to remem
[22-Jun-2009 09:41:04] <ckrough> ber
[22-Jun-2009 09:41:14] <mrayzenoss> clustermagnet: I'm here
[22-Jun-2009 09:41:28] <mrayzenoss> clustermagnet: I can answer any questions you have
[22-Jun-2009 09:42:24] <slestak> i am thinking of using some of the backup-heartbeat related checks. has anyone done anything similar?
[22-Jun-2009 09:43:12] <slestak> i saw a writeup on a blog about using python xml-rpc from bash scripts to update a zenoss check. that could be interesting
[22-Jun-2009 09:43:14] <mrayzenoss> clustermagnet: Additional features in Enterprise are targeted at large Enterprise installations that need integration with commercial offerings (VMware, Oracle, commercial Unixes, Remedy, etc.) and need support. There is also additional functionality for service providers
[22-Jun-2009 09:43:33] <mrayzenoss> slestak: lots of people use XML-RPC to push events into the system
[22-Jun-2009 09:43:41] <clustermagnet> mrayzenoss: thanks!
[22-Jun-2009 09:43:48] <mrayzenoss> clustermagnet: anything specific?
[22-Jun-2009 09:43:51] <clustermagnet> mrayzenoss: sure
[22-Jun-2009 09:43:56] <clustermagnet> mrayzenoss: netapp, vmware, cisco
[22-Jun-2009 09:44:01] <dberger> slestak: have the link?
[22-Jun-2009 09:44:20] <clustermagnet> mrayzenoss: currently , we dont have a budget for zenoss
[22-Jun-2009 09:44:30] <mrayzenoss> clustermagnet: so can be monitored with Core and there are Community ZenPacks for Netapp, VMware and some Cisco stuff
[22-Jun-2009 09:44:34] <clustermagnet> i would HATE to setup nagios for now, and migrate when funds are available
[22-Jun-2009 09:44:55] <mrayzenoss> clustermagnet: understood
[22-Jun-2009 09:45:02] <slestak> dberger: im multitaskingright now. main workstation is down so i am trioubleshooting while using my netbook for now.
[22-Jun-2009 09:45:11] <clustermagnet> mrayzenoss: so it would be viable to run the Core for now... and upgrade to Enterprise when new budget is negotiated?
[22-Jun-2009 09:45:14] <mrayzenoss> we have a lot of users doing those for sure
[22-Jun-2009 09:45:21] <clustermagnet> mrayzenoss: super
[22-Jun-2009 09:45:31] <slestak> dberger: i can get if for you shortly, are you going to be aroubnd for a while?
[22-Jun-2009 09:45:37] <mrayzenoss> clustermagnet: yeah, most of our customers are upgrading from Core, they'll migrate you to it
[22-Jun-2009 09:45:49] <clustermagnet> mrayzenoss: Core has plugins for linux/sun/winblows machines i am sure...
[22-Jun-2009 09:45:54] <mrayzenoss> yeah
[22-Jun-2009 09:46:00] <clustermagnet> mrayzenoss: perfect
[22-Jun-2009 09:46:17] <clustermagnet> mrayzenoss: and zenpacks...
[22-Jun-2009 09:46:28] <clustermagnet> is that 3rd party plugins submitted by people?
[22-Jun-2009 09:46:31] <mrayzenoss> yes
[22-Jun-2009 09:46:35] <slestak> clustermagnet: ive been using the foss vm for ~2 months, it is a good way to get started
[22-Jun-2009 09:46:35] <clustermagnet> is that available on your site
[22-Jun-2009 09:46:38] <dberger> slestak: around all day
[22-Jun-2009 09:46:45] <clustermagnet> slestak:
thank you[22-Jun-2009 09:46:50] <mrayzenoss> clustermagnet: http://www.zenoss.com/community/projects/zenpacks/
[22-Jun-2009 09:47:00] <clustermagnet> mrayzenoss: slestak thanks for all your help!
[22-Jun-2009 09:47:04] <clustermagnet> decision made
[22-Jun-2009 09:47:12] <clustermagnet> talk to you guys if i have questions
[22-Jun-2009 09:47:18] <mrayzenoss> cool
[22-Jun-2009 09:47:22] <dberger> slestak: i'm lurking around to round out my knowledge of zenoss, because i am pretty clueless so far
[22-Jun-2009 09:47:37] <slestak> i am as well.
[22-Jun-2009 09:47:48] <mrayzenoss> slestak: http://www.zenoss.com/community/docs/zenoss-dev-guide/2.4.0/ch02s03.html#technique_sending_an_event
[22-Jun-2009 09:49:35] <mrayzenoss> some people use events to heartbeat their batch processes
[22-Jun-2009 09:49:52] <mrayzenoss> trigger an event with a wait at the beginning and send in a clear at the end
[22-Jun-2009 09:49:59] <mrayzenoss> no clear event received, then alarm
[22-Jun-2009 09:50:25] <slestak> i guess my specific question is: I have to track if a job is "hung" and the way i thought i would do that is send the integer pid as a "level" or the point to track. If a thread stays at one "level"toolong, it is hung, and it is time to sound alarms.
[22-Jun-2009 09:50:38] <mrayzenoss> slestak: there is process monitoring
[22-Jun-2009 09:50:56] <mrayzenoss> oh, I see
[22-Jun-2009 09:51:36] <mrayzenoss> yeah, that works
[22-Jun-2009 09:51:40] <slestak> mrayzenoss: the thread will lanuch a new pid for each job it runs, so i need to check many pids
[22-Jun-2009 09:52:05] <slestak> is therea way to check how long a check has been in a state?
[22-Jun-2009 09:53:20] <slestak> and trigeer on say 60m or sth like that?
[22-Jun-2009 09:53:27] <mrayzenoss> well, you can do a check with events to not alarm until X# of times or unless it hasn't cleared for a certain amount of time
[22-Jun-2009 09:53:46] <mrayzenoss> so if you are exposing the state somewhere, it could be monitored
[22-Jun-2009 09:54:03] <mrayzenoss> or you could send Zenoss and event when it starts
[22-Jun-2009 09:54:04] <slestak> mrayzenoss: i was envisioning a class that reports this way, with subclasses for different time allotments.
[22-Jun-2009 09:54:18] <mrayzenoss> and Zenoss would sit on that event for say, 60 minutes and then alert
[22-Jun-2009 09:54:34] <mrayzenoss> unless the Event Clear came through and cleared it
[22-Jun-2009 09:54:49] <mrayzenoss> so when the job finishes, it would send an Event Clear
[22-Jun-2009 09:54:55] <slestak> 90% of my work may be in a 15m class (a zProperty) and move other jobs into the longer classes if they require it.
[22-Jun-2009 09:55:20] <slestak> i said pids, but inactuality, it is an integer jiob number from the scheduler, so the same report is the same number every executuion
[22-Jun-2009 09:55:33] <mrayzenoss> Check out Jane Curry's Event Management paper: http://www.skills-1st.co.uk/papers/jcurry.html
[22-Jun-2009 09:55:44] <mrayzenoss> there are some good ideas in there
[22-Jun-2009 09:56:10] <mrayzenoss> she also has a Process Monitoring paper, but it sounds like handling this through Events is the way to go
[22-Jun-2009 09:56:21] <mrayzenoss> what kind of job is it?
[22-Jun-2009 09:56:31] <mrayzenoss> if it's in Java, you could also use JMX
[22-Jun-2009 09:56:37] <slestak> lol, ive been given 2 hours to come up with sth, so i may go for the hearbeat 1st, and then firm it up as i go
[22-Jun-2009 09:56:56] <slestak> no, no java, its a language youv'e prob never seen
[22-Jun-2009 09:57:09] <slestak> well, youveseen it, but you cannot believe will use it.
[22-Jun-2009 09:57:33] <mrayzenoss> I've written Delphi to wrap Fortran in DLLs, I've seen things no one should see
[22-Jun-2009 09:57:56] <dberger> is it befunge?
[22-Jun-2009 09:58:15] <mrayzenoss> Befunge is awesome
[22-Jun-2009 09:58:45] <mrayzenoss> check this out: http://www.foo.be/docs/tpj/issues/vol4_3/tpj0403-0017.html
[22-Jun-2009 09:58:54] <slestak> our db uses a dialect of Dartmouth Basic, its Unidata, an IBM database
[22-Jun-2009 09:59:03] <mrayzenoss> ugh
[22-Jun-2009 09:59:18] <mrayzenoss> the Chris Howe entry for Befunge in the Obfuscated Perl Contest: http://www.foo.be/docs/tpj/issues/vol4_3/tpj0403-0017.html
[22-Jun-2009 09:59:43] <mrayzenoss> I worked with him briefly, we sat and stared at that for hours after he won
[22-Jun-2009 10:00:33] <mrayzenoss> slestak: good luck with that
[22-Jun-2009 10:02:00] <slestak> its all ok, we can get off the ledge, i have python on the aix box
[22-Jun-2009 10:02:59] <twm1010> Morning everyone.
[22-Jun-2009 10:03:06] <mrayzenoss> twm1010: morning
[22-Jun-2009 10:04:40] <twm1010> Hey Matt... how goes it?
[22-Jun-2009 10:05:11] <mrayzenoss> ok, just putting some polish on 2.4.2
[22-Jun-2009 10:05:30] <twm1010> Excellent...
[22-Jun-2009 10:05:49] <twm1010> We had a lot of fun here this weekend
[22-Jun-2009 10:06:03] <twm1010> Our datacenter UPS is out of commission thanks to some nasty weather
[22-Jun-2009 10:08:57] <mrayzenoss> did you lose anything else?
[22-Jun-2009 10:10:08] <twm1010> Yeah, an older Windows box is having some power issues, and oddly enough, a virtual machine that was shutdown clean has a fubar filesystem that won't diskcheck.
[22-Jun-2009 10:10:48] <twm1010> What do you expect from 9 year old hardware?
[22-Jun-2009 10:19:44] <clustermagnet> Guys, im trying to setup zenoss, but my mysql box is elsewhere
[22-Jun-2009 10:19:46] <clustermagnet> in the docs:
[22-Jun-2009 10:19:47] <clustermagnet> See the section titled "Using Zenoss with a Remote MySQL Instance" in Zenoss Administration.
[22-Jun-2009 10:19:59] <clustermagnet> however, i dont see this section in the zenoss administration pDF
[22-Jun-2009 10:20:10] <clustermagnet> how do you tell the installer to go to another mysql box
[22-Jun-2009 10:20:11] <twm1010> which install method are you using?
[22-Jun-2009 10:20:32] <twm1010> Zenoss comes prepackaged with a mysql instance, so, it will run that first if you're installing from stack i believe.
[22-Jun-2009 10:20:33] <clustermagnet> twm1010: uhm, .bin
[22-Jun-2009 10:20:41] <clustermagnet> twm1010: stack
[22-Jun-2009 10:20:49] <clustermagnet> twm1010: so if i bang from source, it wont do that?
[22-Jun-2009 10:21:23] <mrayzenoss> clustermagnet: You're gonna love this: http://dev.zenoss.org/trac/ticket/5105
[22-Jun-2009 10:22:04] * mrayzenoss sighs
[22-Jun-2009 10:22:12] <clustermagnet> mrayzenoss: trac rulez
[22-Jun-2009 10:22:17] <clustermagnet> mrayzenoss:
thanks man![22-Jun-2009 10:22:22] <mrayzenoss> clustermagnet: do you really need an external MySQL?
[22-Jun-2009 10:22:43] <clustermagnet> mrayzenoss: absolutely not
[22-Jun-2009 10:23:12] <clustermagnet> mrayzenoss: in the current environment there is a very powerful mysql setup, which we use
[22-Jun-2009 10:23:28] <twm1010> I stumbled a few weeks back when i did the 2.4.1 upgrade and didn't realize zenoss-stack was listed as a service. So I'm sitting there banging my head wondering why it tells me MySQL isn't started...
[22-Jun-2009 10:23:57] <twm1010> >service zenoss-stack start did the trick...
[22-Jun-2009 10:25:50] <clustermagnet> mrayzenoss: however, those options dont let you point to another mysql server...
[22-Jun-2009 10:25:55] <clustermagnet> they just let you change u/p and port
[22-Jun-2009 10:25:58] <clustermagnet> on a local install
[22-Jun-2009 10:26:28] <clustermagnet> anyways, im sure its not a problem if compiling from source, ya?
[22-Jun-2009 10:26:48] <twm1010> I believe there is a config file when you're building from source that you can edit and change things as you please.
[22-Jun-2009 10:27:17] <twm1010> or... hold on
[22-Jun-2009 10:28:01] <twm1010> you could just let it install, then export that DB to the server you want to put it on, and then change the settings in event manager to point over there.
[22-Jun-2009 10:28:34] <twm1010> im no MySQL guru by any means but I don't see why that wouldn't work.
[22-Jun-2009 10:28:54] <mrayzenoss> yeah, what twm1010 said
[22-Jun-2009 10:29:07] <mrayzenoss> we definitely have people not running on the stock MySQL
[22-Jun-2009 10:29:25] <mrayzenoss> I'll ask around a bit
[22-Jun-2009 10:30:06] <rmatte> you could always do remote mysql but via nfs
[22-Jun-2009 10:31:27] <rmatte> rmatte: http://dev.zenoss.org/trac/ticket/5071 apparently it's no longer a blocker?
[22-Jun-2009 10:31:37] <rmatte> erm
[22-Jun-2009 10:31:43] <rmatte> meant to say matt:
[22-Jun-2009 10:32:19] <zenChild> Does anyone know how to chain relationships in Zenoss? I have my relationships working between the device and components, but what I wanted was a relationship between a component and subcomponents and that doesn't seem to be working. I posted something to the dev forum but haven't received any responses yet. You can reference the post here: http://forums.zenoss.com/viewtopic.php?t=10086
[22-Jun-2009 10:34:09] <clustermagnet> thanks guys!
[22-Jun-2009 10:34:24] <mrayzenoss> rmatte: So development looked into ZenPack loading for obvious dumb stuff and ran it with a profiler, but nothing popped up.
[22-Jun-2009 10:34:47] <clustermagnet> mrayzenoss: so, do you know of any zenpack to monitor netapp?
[22-Jun-2009 10:35:02] <clustermagnet> i see you have good vmware feedback
[22-Jun-2009 10:35:15] <clustermagnet> we'll need netapp, cisco, and vmware
[22-Jun-2009 10:35:20] <mrayzenoss> rmatte: we're switching to Zope 2.11 for the next release, which seems to be faster. It got lowered from a blocker because it's still working
[22-Jun-2009 10:35:29] <rmatte> k
[22-Jun-2009 10:36:15] <zenChild> clustermagnet: trac-zenpacks
[22-Jun-2009 10:36:16] <mrayzenoss> rmatte: so it's not off the radar, we just don't have any solid leads on improving it other than the Zope upgrade
[22-Jun-2009 10:36:58] <mrayzenoss> clustermagnet: there are 3 NetApp ZenPacks, this one attempts to roll them all into 1: trac-zenpacks/wiki/ZenPackscommunityNetApp
[22-Jun-2009 10:37:12] <clustermagnet> mrayzenoss: thanks
[22-Jun-2009 10:37:16] <clustermagnet> zenChild: thank you
[22-Jun-2009 10:37:30] <clustermagnet> I do see that some require zenoss v 2.2 2.1... etc
[22-Jun-2009 10:37:38] <mrayzenoss> that means "or later"
[22-Jun-2009 10:37:46] <clustermagnet> mrayzenoss: sweet
[22-Jun-2009 10:37:47] <clustermagnet> thanks!
[22-Jun-2009 10:37:55] <mrayzenoss> 2.4.1 is current, 2.4.2 is probably tomorrow
[22-Jun-2009 10:38:06] <clustermagnet> im excited like crazy
[22-Jun-2009 10:38:10] <clustermagnet> no more text files
[22-Jun-2009 10:39:06] <clustermagnet> gents, thank you so much for your immediate help!
[22-Jun-2009 10:39:10] * clustermagnet owes beer
[22-Jun-2009 10:39:49] <mrayzenoss> I try to travel to different community events, collecting beers. Next stop, Community Leadership Summit next month
[22-Jun-2009 10:43:22] <clustermagnet> mrayzenoss: heh
[22-Jun-2009 10:43:29] <clustermagnet> mrayzenoss: distributed monitoring
[22-Jun-2009 10:43:41] <rocket> jb: ping
[22-Jun-2009 10:43:44] <clustermagnet> if there are 2 datacenters, how do you suggest zenoss should be implemented ?
[22-Jun-2009 10:43:48] <mrayzenoss> clustermagnet: there are 2 VMware ZenPacks, 1 is for Filesystems, 1 is for a list of Guests
[22-Jun-2009 10:43:57] <mrayzenoss> clustermagnet: are they on the same network?
[22-Jun-2009 10:44:17] <mrayzenoss> err, reachable
[22-Jun-2009 10:44:32] <clustermagnet> mrayzenoss: there is a VPN going between the 2 networks
[22-Jun-2009 10:44:40] <clustermagnet> however, they are logically different
[22-Jun-2009 10:44:44] <clustermagnet> 1 is production colo
[22-Jun-2009 10:44:51] <clustermagnet> 2nd is corporate
[22-Jun-2009 10:45:04] <rmatte> you need to come up here to Ottawa, Canada for one of our events
[22-Jun-2009 10:45:05] <clustermagnet> it would be sweet to have 2 zenoss machines, 1 at each location
[22-Jun-2009 10:45:13] <rmatte> then I can buy you a beer and score a Zenny
[22-Jun-2009 10:45:20] <mrayzenoss> clustermagnet: that would probably be optimal
[22-Jun-2009 10:45:32] <mrayzenoss> rmatte: Ohio LinuxFest is as close as I'm getting this year
[22-Jun-2009 10:46:05] <mrayzenoss> it's a big event though, 2 days, 1000 or so people
[22-Jun-2009 10:46:11] <rmatte> hmmm, I'd need to get a passport to go to that, I do know someone in Ohio though who works in the industry
[22-Jun-2009 11:08:44] <twm1010> I need to take a look at that NetApp zenpack again...
[22-Jun-2009 11:24:24] <mrayzenoss> yeah, it needs some feedback. I'm close to asking the author to go ahead and publish it
[22-Jun-2009 11:45:59] <adr3nalin3> Could someone please refresh my memory on the setting that gets google maps to draw connectors on the map?
[22-Jun-2009 11:48:55] <twm1010> you're talking about where it should be drawing them and it is not?
[22-Jun-2009 11:50:34] <adr3nalin3> I thought there was a setting to tell it to draw the connectors. Is it supposed to by default?
[22-Jun-2009 11:51:41] <dberger> adr3nalin3: you might be asking about the zDrawMapLinks property
[22-Jun-2009 11:52:09] <dberger> adr3nalin3: it is a property of the network
[22-Jun-2009 11:52:30] <adr3nalin3> dberger: that is exactly what I was looking for. thank you
[22-Jun-2009 11:52:41] <dberger> adr3nalin3: np
[22-Jun-2009 11:55:31] <adr3nalin3> What property are the map links drawn from: Is it network description?
[22-Jun-2009 11:55:39] <adr3nalin3> or location address?
[22-Jun-2009 11:56:18] <adr3nalin3> nm I got it
[22-Jun-2009 11:56:30] <dberger> what is it?
[22-Jun-2009 11:57:03] <dberger> because all I understand is that zenoss automatically scans your network topology
[22-Jun-2009 11:57:22] <dberger> i tried to build one manually using a tutorial here: http://www.zenoss.com/Members/netdata/create-manual-device-dependencys/
[22-Jun-2009 11:57:44] <dberger> but I didn't really understand what I was doing or have much success
[22-Jun-2009 11:59:11] <adr3nalin3> dberger: http://www.zenoss.com/community/docs/zenoss-guide/2.1.1/ch07s05.html
[22-Jun-2009 11:59:44] <adr3nalin3> My problem is I have 1 network with different subnets so it treats them all as different networks
[22-Jun-2009 12:00:22] <adr3nalin3> well technically they are all seperate networks but I would like to see them all connected to each other
[22-Jun-2009 12:04:02] <dberger> well, you might be able to get it up and running by building the dependency manually, but I have no idea whether that's a good idea
[22-Jun-2009 12:05:25] <adr3nalin3> ill check it out thank you
[22-Jun-2009 12:10:15] <etank> do alerting rules work like firewall rules? the first to match is the one that is used.
[22-Jun-2009 12:11:16] <twm1010> No
[22-Jun-2009 12:11:25] <twm1010> Each alert rule is independent
[22-Jun-2009 12:12:24] <etank> hmmm ... ok
[22-Jun-2009 12:13:18] <twm1010> There might be a method to make them dependent on each other, but I don't think so.
[22-Jun-2009 12:15:57] <etank> looks like the rule was not enabled
[22-Jun-2009 12:16:09] <etank> i think i may have it right now
[22-Jun-2009 12:25:53] <perr0> anybody seen casey?
[22-Jun-2009 12:26:55] <perr0> caseyh20 to be exact
[22-Jun-2009 14:40:57] <dberger> anyone know of a good way to monitor multicast? if not something pre-integrated with zenoss, something to run standalone that could trigger events on router interfaces?
[22-Jun-2009 14:47:36] <jcallaway> hey does anyone know who's the best person to reach out to regading the nmap plugin?
[22-Jun-2009 14:49:12] <jcallaway> ...the paths are pointing to the wrong place
[22-Jun-2009 15:38:18] <zenChild> does anyone have an example on how to use Device.findChild() ?
[22-Jun-2009 15:47:29] <jcallaway> i wish i could help you out! i'd like to rack up some karma so i can get some help with my issue
[22-Jun-2009 15:49:33] <mrayzenoss> zenChild: the only reference I can find is in the 2.4.1 release notes
[22-Jun-2009 15:49:46] <mrayzenoss> jcallaway: what platform and installer was that path issue?
[22-Jun-2009 15:50:14] <jcallaway> hey thanks, the nmap plugin is hardcoded to refer to the binary path to nmap via
[22-Jun-2009 15:50:38] <jcallaway> zenpath . libexec . nmap
[22-Jun-2009 15:51:08] <jcallaway> and libexec was missing in the 2.4.1 debian package, and the vmware appliance
[22-Jun-2009 15:51:20] <jcallaway> had to make a symlink
[22-Jun-2009 15:51:52] <mrayzenoss> is nmap under "Run Commands"?
[22-Jun-2009 15:51:59] <zenChild> mrayzenoss: I think I worked my way around the findChild thing anyway by creating a dictionary in my modeler code.
[22-Jun-2009 15:52:17] <jcallaway> i don't believe so, i'd have to check
[22-Jun-2009 15:52:34] <jcallaway> it's a modeler plugin
[22-Jun-2009 15:56:08] <mrayzenoss> jcallaway: which version are you running?
[22-Jun-2009 15:56:15] <jcallaway> 2.4.1
[22-Jun-2009 15:56:24] <mrayzenoss> hmmm... cause I came across this: http://dev.zenoss.org/trac/ticket/4680
[22-Jun-2009 15:56:36] <jcallaway> i'm trying to compile 2.4.x from svn so i can contribute the fix
[22-Jun-2009 15:56:42] <jcallaway> thanks, i'll check out the ticket
[22-Jun-2009 15:57:24] <jcallaway> okay, great! thanks, i'm not crazy : )
[22-Jun-2009 15:57:51] <mrayzenoss> yeah, sounds like nmap got left out
[22-Jun-2009 15:58:23] <mrayzenoss> feel free to open another ticket
[22-Jun-2009 15:58:30] <jcallaway> it would be nice to get it rolling since it can do os/application version detection if so configured
[22-Jun-2009 15:58:50] <jcallaway> ok, cool. i'll see what i can do and get as much info as possible before opening the ticket
[22-Jun-2009 15:59:22] <mrayzenoss> thanks, mark it as a "High" or "Blocker", since it sounds like missing functionality between installers
[22-Jun-2009 15:59:32] <jcallaway> gotcha
[23-Jun-2009 00:00:42] [disconnected at Tue Jun 23 00:00:42 2009]
[23-Jun-2009 00:00:42] [connected at Tue Jun 23 00:00:42 2009]
[23-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[23-Jun-2009 03:11:27] <Appiah_> All my daemons are running but i get heartbeat failure for each one
[23-Jun-2009 03:11:37] <Appiah_> this happend after upgrade from 2.3 to 2.4
[23-Jun-2009 03:11:57] <Appiah_> Had this problem on another server but I dont remember how we fixed this...
[23-Jun-2009 03:44:41] <baffle> Appiah_: Event manager -> Clear All Heartsbeats (from dropdown)
[23-Jun-2009 03:46:39] <Appiah_> but...
[23-Jun-2009 03:46:46] <Appiah_> it just comes back baffle
[23-Jun-2009 03:48:20] <baffle> Does anyone here have an event transform to support multiplatform per filesystem treshold? Preferrable as a "custom" setting. The "Tip of the month" seems to derive the treshold from the Windows FS-label, wich seems a bit.. Strange..
[23-Jun-2009 03:50:36] <baffle> Appiah_: Hmm, but are the daemons actually running?
[23-Jun-2009 03:50:53] <baffle> Appiah_: And have you checked the logs? Can they contact ZenHub?
[23-Jun-2009 03:51:45] <Appiah_> they all show up green
[23-Jun-2009 03:51:47] <Appiah_> everything works fine
[23-Jun-2009 03:57:22] <baffle> Appiah_: Strange.
[23-Jun-2009 07:01:05] ckrough_ is now known as ckrough
[23-Jun-2009 08:14:43] <klinstifen> morning all
[23-Jun-2009 08:15:08] <klinstifen> i got a quick question about datacollectors and the OS section on a device's status tab
[23-Jun-2009 08:15:16] <klinstifen> anybody have experience in that area?
[23-Jun-2009 08:15:33] <venturaville> a little
[23-Jun-2009 08:22:34] <klinstifen> ok cool
[23-Jun-2009 08:22:50] <klinstifen> so it seems you can match OIDs to that section
[23-Jun-2009 08:22:58] <klinstifen> err the values of OIDs
[23-Jun-2009 08:24:03] <klinstifen> and that's done by doing something like snmpgetmap = getmap({ oid : 'setHWProductKey' })
[23-Jun-2009 08:24:34] <klinstifen> do you know where the list of "setXYZ" attributes can be found?
[23-Jun-2009 08:24:57] <klinstifen> in that OS section there is a Tag # and Serial #
[23-Jun-2009 08:25:05] <klinstifen> i can get those values via SNMP
[23-Jun-2009 08:25:10] <klinstifen> so I want to map them there
[23-Jun-2009 08:25:19] <klinstifen> but I have no idea what those attributes are called
[23-Jun-2009 08:30:11] <rocket> klinstifen: they are attributes of the object
[23-Jun-2009 08:30:43] <rocket> klinstifen: you can either look at the source code in ZenModel
[23-Jun-2009 08:31:07] <rocket> or use zendmd to find the object in the zopedb and then do a dir(object) on it to get its methods
[23-Jun-2009 08:31:42] <rocket> eg from zendmd
[23-Jun-2009 08:31:45] <rocket> d = devices.findDevice('192.168.31.3'
[23-Jun-2009 08:31:47] <rocket> )
[23-Jun-2009 08:31:53] <rocket> dir(d)
[23-Jun-2009 08:32:42] <venturaville> setHWSerialNumber
[23-Jun-2009 08:32:55] <venturaville> setHWTag
[23-Jun-2009 08:33:21] <venturaville> grep -nr setHW on the Product directory helps here.........
[23-Jun-2009 08:33:31] <venturaville> Products
[23-Jun-2009 08:42:07] <klinstifen> ventur, grep -nr setHW on $ZENHOME/Products?
[23-Jun-2009 08:42:50] <klinstifen> rocket, thanks. is there any way to dump the output of dir(d) to a file for parsing?
[23-Jun-2009 08:42:51] <venturaville> yep
[23-Jun-2009 08:42:58] <rocket> klinstifen: yup .. actually you might just want to go to $ZENHOME/Products/ZenModel
[23-Jun-2009 08:43:20] <rocket> klinstifen: sure zendmd is just a python shell
[23-Jun-2009 08:43:31] <venturaville> rocket: DataCollector directory has some interesting references too
[23-Jun-2009 08:43:59] <rocket> venturaville: DataCollector populates teh objects defined in ZenModel
[23-Jun-2009 08:44:07] <venturaville> right......
[23-Jun-2009 08:44:09] <rocket> venturaville: but that might be easier to interpret
[23-Jun-2009 08:44:28] <rocket> venturaville: though DataCollector isnt always as complete as ZenModel
[23-Jun-2009 08:44:53] <venturaville> nope, but does provide some nice examples
[23-Jun-2009 08:44:56] <venturaville> *shrug*
[23-Jun-2009 08:45:03] <rocket> yes I cant argue with that
[23-Jun-2009 08:48:51] <klinstifen> cool I will play around with that
[23-Jun-2009 08:48:53] <klinstifen> thanks
[23-Jun-2009 08:51:34] <klinstifen> ok so another question and maybe this is what's leading me astray
[23-Jun-2009 08:51:43] <klinstifen> looking at the DellDeviceMap.py
[23-Jun-2009 08:51:56] <klinstifen> in $ZENHOME/Products/DataCollector/plugins/zenoss/snmp
[23-Jun-2009 08:52:23] <klinstifen> they use setHWProductKey
[23-Jun-2009 08:52:45] <klinstifen> and that seems to map to the HW Model on the Status tab
[23-Jun-2009 08:52:52] <klinstifen> not an exact naming match but close enough
[23-Jun-2009 08:53:28] <klinstifen> but...
[23-Jun-2009 08:54:33] <klinstifen> err..nevermind
[23-Jun-2009 08:54:40] <klinstifen> was going to say another field isn't mapped
[23-Jun-2009 08:54:46] <klinstifen> but then i noticed that they have it commented out
[23-Jun-2009 08:55:05] <venturaville> I think HWProductName is the one in the status tab
[23-Jun-2009 08:56:29] <venturaville> not 100% certain of that though
[23-Jun-2009 08:57:39] <klinstifen> yeah the serial number i dont think matches up either
[23-Jun-2009 08:57:51] <klinstifen> the oid they reference is really the tag...
[23-Jun-2009 08:58:03] <klinstifen> but it doesn't get populated either way
[23-Jun-2009 08:58:40] <rocket> klinstifen: take a look at the aix zenpack .. it sets serial numbers etc
[23-Jun-2009 08:59:00] <klinstifen> ok
[23-Jun-2009 08:59:12] <klinstifen> after you modify the datacollector code
[23-Jun-2009 08:59:19] <rocket> its in the modelers folder in that zenpack .. I know it works as I wrote it
[23-Jun-2009 08:59:28] <klinstifen> hehe
[23-Jun-2009 09:00:00] <rocket> it also set the setHWProductKey properly so the manufacturer link works etc ..
[23-Jun-2009 09:03:13] <klinstifen> do the modelers have preference over each other?
[23-Jun-2009 09:03:23] <klinstifen> i made changes to the dell modeler to grab the tag
[23-Jun-2009 09:03:29] <klinstifen> but it still isn't populating the status tab
[23-Jun-2009 09:03:39] <klinstifen> i wonder if it is being overwritten by another modeler
[23-Jun-2009 09:04:16] <rocket> klinstifen: they do .. there is order significance .. and or is the modeller enabled?
[23-Jun-2009 09:04:42] <klinstifen> top down or bottom up?
[23-Jun-2009 09:04:47] <klinstifen> which modeler runs last?
[23-Jun-2009 09:04:51] <rocket> eg doing a "zenmodel run -v 10 -d device" on the command line with show you a lot ..
[23-Jun-2009 09:04:54] <rocket> bottom last
[23-Jun-2009 09:04:58] <klinstifen> hmm
[23-Jun-2009 09:05:02] <klinstifen> my dell modeler is last
[23-Jun-2009 09:05:08] <klinstifen> i believe the modeler is enabled
[23-Jun-2009 09:05:23] <rocket> restart zenoss after updating the code?
[23-Jun-2009 09:05:26] <klinstifen> i added it to the collector plugins
[23-Jun-2009 09:05:29] <klinstifen> yeppers
[23-Jun-2009 09:05:30] <klinstifen> did that
[23-Jun-2009 09:10:50] <klinstifen> ah this could be why
[23-Jun-2009 09:10:56] <klinstifen> the modeler isn't running
[23-Jun-2009 09:11:06] <klinstifen> err the collector isn't run
[23-Jun-2009 09:11:09] <klinstifen> when i model the device
[23-Jun-2009 09:11:27] <rocket> check the logs .. most likely a bug in the code has disabled that modeller
[23-Jun-2009 09:11:34] <klinstifen> k
[23-Jun-2009 09:14:20] <klinstifen> hmm doesn't look like it
[23-Jun-2009 09:14:44] <klinstifen> i'll tinker a bit more
[23-Jun-2009 09:18:09] <rocket> klinstifen: look at adding log.debug lines to your modeller etc
[23-Jun-2009 09:18:24] <rocket> there are examples of that in the aix zenpack
[23-Jun-2009 09:18:40] <rocket> so when you do zenmodeller run -v 10 those lines show up
[23-Jun-2009 09:25:50] <klinstifen> rocket: ok, i'll give that a go
[23-Jun-2009 09:33:43] <venturaville> here is an odd one
[23-Jun-2009 09:34:13] <venturaville> barring us having any sort of 'meta-devices', is there a way to have a relationship exist that you can populate outside of a modeler
[23-Jun-2009 09:34:30] <venturaville> i.e. I have something that needs to 'push' data externally into a relationship
[23-Jun-2009 09:35:08] <venturaville> so I would define a _relations in the class for the container
[23-Jun-2009 09:35:15] <venturaville> but then not have an attached modeler?
[23-Jun-2009 09:44:09] <zenChild> venturaville: couldn't you just get a handle to the object and manipulate the relationship from there?
[23-Jun-2009 09:54:06] <rocket> venturaville: as long as you can get the object in python I dont know why you wouldnt be able to update it without a modeller
[23-Jun-2009 10:17:25] <eidolon> hey folks - anyone know if there's an easy way to customize zenoss alert messages?
[23-Jun-2009 10:17:47] <eidolon> ala, we'd like to say "HTTP alert on such and such a host" "Please escalate to Bob in engineering"
[23-Jun-2009 10:18:50] <ckrough> eidolon: the "message" tab in the alerting rule in question
[23-Jun-2009 10:19:58] <eidolon> ORLY. *looks*
[23-Jun-2009 10:20:11] <cain22> will zenoss map locations if there isn't any devives directly under that group?
[23-Jun-2009 10:20:37] <rmatte> cain22: Go to that location and click on the map tab
[23-Jun-2009 10:20:42] <rmatte> if it shows up then the answer is yes
[23-Jun-2009 10:21:02] <cain22> hmm...im trying to figure out why my map isn't drawing the network lines
[23-Jun-2009 10:21:28] <rmatte> did you go to Networks, and enable drawing map lines for the networks listed there?
[23-Jun-2009 10:21:50] <cain22> i believe yes, but i'll double check
[23-Jun-2009 10:21:54] <rmatte> if so, then Zenoss just isn't understanding your network topology
[23-Jun-2009 10:21:55] <klinstifen> zDrawMapLinks should be set to True
[23-Jun-2009 10:22:09] <rmatte> if you have sites going out over MPLS circuits or something Zenoss probably won't understand
[23-Jun-2009 10:22:18] <cain22> draw map links set to true
[23-Jun-2009 10:22:26] <rmatte> or any sites that aren't directly linked (I.E are going out over the internet)
[23-Jun-2009 10:22:58] <cain22> the weird thing is, at one point most of them awhile back were linked...
[23-Jun-2009 10:23:08] <rmatte> I have one client where all the lines were drawn fine and another client where no lines were drawn, it's just because Zenoss can't figure out what is directly linked since it isn't actually directly linked
[23-Jun-2009 10:23:27] <rmatte> it generally does better with connections like frame relay or LAN extensions
[23-Jun-2009 10:24:09] <cain22> whats "prefer snmp mapping..?
[23-Jun-2009 10:24:20] <cain22> setting for networks
[23-Jun-2009 10:24:39] <rmatte> prefer mapping via information discovered with SNMP versus info discovered with pings/traceroutes/portscans
[23-Jun-2009 10:24:40] <cain22> er, snmp naming
[23-Jun-2009 10:24:41] <cain22> nm
[23-Jun-2009 10:24:52] <rmatte> oh
[23-Jun-2009 10:24:56] <rmatte> where do you see that?
[23-Jun-2009 10:25:31] <cain22> under network properties
[23-Jun-2009 10:25:46] <rmatte> hmmm, no idea, google it
[23-Jun-2009 10:27:32] <ckrough> wonder if it has to do with path outages, choosing whether to built dependcy trees from traceroutes or snmp returned routes
[23-Jun-2009 10:27:43] <ckrough> definetely let us know what you find
[23-Jun-2009 10:28:46] <venturaville> rocket: that is what I assumed, but the documentation isn't very clear on it
[23-Jun-2009 10:29:39] <rocket> venturaville: thats one of those .. if you have programmed with it long enough you can figure out interesting ways to do the same thing
[23-Jun-2009 10:31:19] <venturaville> yeah... I'm still trying to get my head around containers/modelers... I guess for now I am just going to put _relations straight in the container class and then try and manipulate it in zendmd
[23-Jun-2009 10:31:27] <venturaville> that might give me a better feel for it
[23-Jun-2009 10:34:20] <ckrough> venturaville: the other day we were talking accessing URL variables within python in a TALEs expression. solution: "foo python: here.REQUEST.test;" will put value of test into foo, which you can use in further python in the template
[23-Jun-2009 10:35:16] <venturaville> ckrough: good to know. I need to file that away for further use........ :-)
[23-Jun-2009 10:38:33] <ckrough> venturaville: I ended up needing some sanity tests on the value, so I moved it out to a python script, but that's the syntax that work within the tempate
[23-Jun-2009 10:40:18] <ckrough> venturaville: the other day we were talking accessing URL variables within python in a TALEs expression. solution: "foo python: here.REQUEST.test;" will put value of test into foo, which you can use in further python in the template"
[23-Jun-2009 10:40:30] <ckrough> woops, too much scrollback. bad post
[23-Jun-2009 10:40:31] <klinstifen> other than zenmodeler run -v10 is there another way to test my modeler to see where its bombing out or why it isn't inserting the found data?
[23-Jun-2009 10:41:26] <ckrough> klinstifen: v10 doesnt show any issues, but the devices doesnt show up or get updated?
[23-Jun-2009 10:42:56] <klinstifen> v10 shows it is running the collector...but the device doesnt get updated
[23-Jun-2009 10:43:30] <klinstifen> i found the initial problem...the collector wasn't being run because the builtin condition wasn't true
[23-Jun-2009 10:43:41] <klinstifen> i commented that out and now the collector runs when i model the device
[23-Jun-2009 10:53:52] <eidolon> *sigh* i love messages from zenoss liket his
[23-Jun-2009 10:53:54] <eidolon> "Disk space low: 1601.6% used (-9068.24 GB free)"
[23-Jun-2009 10:53:58] <ckrough> zenmodeler log shows the collector pulling data from the device?
[23-Jun-2009 10:54:42] * ckrough thinks that disk space calculation may be inaccurate
[23-Jun-2009 10:56:24] <klinstifen> no
[23-Jun-2009 10:58:42] <ckrough> klinstifen: what type of device and is it ssh or snmp collection
[23-Jun-2009 11:07:49] <klinstifen> dell server and it is snmp collection
[23-Jun-2009 11:08:04] <klinstifen> i am working on the delldevicemap collector that is built in
[23-Jun-2009 11:34:56] <rmatte> I need to remove a zenpack without removing it's contents, anyone know how?
[23-Jun-2009 11:35:17] <rmatte> specifically, I want to remove the Cisco Mibs Pack from the ZenPack inventory but leave the Mibs behind
[23-Jun-2009 11:35:28] <rmatte> (I'm basically just seeing if it resolves load time issues)
[23-Jun-2009 11:39:18] <mrayzenoss> not sure you can do that, don't see anything like that in the zenpack command
[23-Jun-2009 11:39:42] <mrayzenoss> do you want all the MIBs or just a few? Maybe you could re-export the ones you want to another ZenPack?
[23-Jun-2009 11:39:49] <rmatte> I mean, I could go in to the zenpack and delete all of the items then remove the zenpack, only problem is that the zenpack has over 72000 items in it
[23-Jun-2009 11:40:09] <rmatte> nah, I want all the Mibs, the Cisco Mibs set has tons of dependencies it's like a web
[23-Jun-2009 11:40:17] <rmatte> most Mibs won't function without certain other Mibs
[23-Jun-2009 11:40:25] <mrayzenoss> ugh
[23-Jun-2009 11:41:09] <rmatte> plus we're an MSP so we monitor tons of different devices, we need to be prepared for anything
[23-Jun-2009 11:43:55] <rmatte> can zendmd manipulate zenpacks?
[23-Jun-2009 11:44:41] <mrayzenoss> sure
[23-Jun-2009 11:47:05] <rmatte> there must be some way to have it purge the contents of a zenpack without actually removing them from Zenoss then?
[23-Jun-2009 11:48:06] <rmatte> If removing the contents from the Zenpack fixes the response time issues, I'm thinking that there may be some way for me to have my zenpack run a script to do the same thing once it's done installing
[23-Jun-2009 11:48:42] <rmatte> or just provide the script as a workaround
[23-Jun-2009 11:51:58] <blocky> can I rewrite reportmail to send multiple reports in 1 email
[23-Jun-2009 11:57:57] <rmatte> is Chet around, or someone who would be able to figure out the zendmd command to accomplish that?
[23-Jun-2009 12:04:06] <klinstifen> ok i finally go the collector to work
[23-Jun-2009 12:04:13] <klinstifen> it's pulling the information i want
[23-Jun-2009 12:04:37] <klinstifen> just one more piece of the puzzel to finish
[23-Jun-2009 12:04:50] <klinstifen> setOSProduct...is that the same thing as OS Make
[23-Jun-2009 12:04:51] <klinstifen> ?
[23-Jun-2009 12:05:15] <klinstifen> or asking it another way: once I am in zendmd
[23-Jun-2009 12:05:24] <klinstifen> i do a d = devices.findDevice('ip')
[23-Jun-2009 12:05:27] <klinstifen> and then a dir(d)
[23-Jun-2009 12:05:48] <klinstifen> how can i output the value of a particular attribute listed by dir(d)?
[23-Jun-2009 12:06:03] <klinstifen> like i want to echo d.setOSProduct
[23-Jun-2009 12:06:09] <rmatte> print
[23-Jun-2009 12:06:29] <rmatte> instead of echo
[23-Jun-2009 12:06:41] <klinstifen> hmm
[23-Jun-2009 12:06:43] <klinstifen> it says
[23-Jun-2009 12:07:03] <klinstifen> <bound method Device.setOSProduct of <Device at 'path to device'>
[23-Jun-2009 12:07:04] <rmatte> Product would be the model
[23-Jun-2009 12:07:50] <rmatte> yeh, I guess you won't get much from that
[23-Jun-2009 12:07:58] <rmatte> are you sure that setOSProduct is even what you want?
[23-Jun-2009 12:08:23] <klinstifen> well...it's right next to setOSProductKey and none of the neighboring values seem applicable
[23-Jun-2009 12:08:23] <rmatte> try getOSProduct
[23-Jun-2009 12:08:31] <klinstifen> and there isn't anything else that looks right
[23-Jun-2009 12:09:05] <klinstifen> err....d.getOSProduct?
[23-Jun-2009 12:09:18] <rmatte> getHWManufacturerName
[23-Jun-2009 12:09:18] <rmatte> getHWProductName
[23-Jun-2009 12:09:22] <rmatte> are the two that I use
[23-Jun-2009 12:09:45] <klinstifen> d.getHWProductName
[23-Jun-2009 12:09:50] <klinstifen> gives me that same "bound" error
[23-Jun-2009 12:10:01] <rmatte> it's not an error
[23-Jun-2009 12:10:20] <rmatte> it's the content of that variable, but only zenoss can really interpret it properly
[23-Jun-2009 12:10:25] <klinstifen> gotcha
[23-Jun-2009 12:11:17] <rmatte> what if you type just d.getHWManufacturerName ?
[23-Jun-2009 12:11:20] <rmatte> instead of print
[23-Jun-2009 12:11:30] <rmatte> that should give you some output
[23-Jun-2009 12:12:10] <klinstifen> same...
[23-Jun-2009 12:12:15] <rmatte> hmmm
[23-Jun-2009 12:12:24] <rmatte> no idea then
[23-Jun-2009 12:12:59] <klinstifen> I would think that if there was a getHWManufacturer
[23-Jun-2009 12:13:05] <klinstifen> then there should be a setHWManufacturer
[23-Jun-2009 12:13:07] <klinstifen> no?
[23-Jun-2009 12:13:14] <klinstifen> it doesn't appear in dir(d)...
[23-Jun-2009 12:13:57] <rocket> setHWProduct
[23-Jun-2009 12:14:51] <rmatte> rocket, maybe you would know, is there any way to manipulate the contents of a zenpack with zendmd?
[23-Jun-2009 12:15:01] <rmatte> I have a huge zenpack and I want to empty out the contents of it
[23-Jun-2009 12:15:09] <rmatte> without actually removing them from Zenoss
[23-Jun-2009 12:15:10] <ckrough> klinstifen: thoe are callable methods
[23-Jun-2009 12:15:18] <ckrough> klinstifen: dev.getHWProductName()
[23-Jun-2009 12:15:24] <ckrough> try with ()
[23-Jun-2009 12:15:31] <rmatte> oh right, forgot about the ()
[23-Jun-2009 12:15:41] <rmatte> that should give you proper output
[23-Jun-2009 12:15:55] <klinstifen> ah ha!
[23-Jun-2009 12:16:40] <rocket> rmatte: hrmm i am not sure .. what are you trying to empty out? data out of the objects.xml?
[23-Jun-2009 12:16:54] <rmatte> yeh, here's the scenario...
[23-Jun-2009 12:17:42] <rmatte> after you install my Cisco Mibs Zenpack, the zProperties pages and other pages take ridiculous amounts of time to load, I think it's because Zenoss is trying to index the objects.xml file almost every time it tries to load one of those pages...
[23-Jun-2009 12:18:10] <rmatte> and the index.xml file in this zenpack is like 34MB in size (or more, maybe more like 64)
[23-Jun-2009 12:18:19] <rocket> rmatte: hrmm well it really is loading the objects.xml file into the zeo database ...
[23-Jun-2009 12:18:25] <rmatte> I want to empty out the zenpack without actually removing the Mibs
[23-Jun-2009 12:18:52] <rocket> rmatte: you could just move the objects.xml file out of the way but I dont think it will do anything
[23-Jun-2009 12:18:54] <rmatte> when I first installed the Mibs by hand I don't remember having any kind of response problem
[23-Jun-2009 12:19:03] <rmatte> as soon as I installed them as a zenpack the problems started
[23-Jun-2009 12:19:15] <rmatte> so I want to simulate installing them by hand, but without going through all that hassle again
[23-Jun-2009 12:19:21] <rmatte> installing the zenpack is much much easier
[23-Jun-2009 12:19:23] <rocket> rmatte: have you tried to reindex from the command line?
[23-Jun-2009 12:19:52] <rmatte> I've reindexed on boxes that have this pack installed and it's done nothing to improve performance
[23-Jun-2009 12:20:14] <rocket> rmatte: if i would guess I would say that objects installed from the zenpacks arent getting indexed at all in the zeo database
[23-Jun-2009 12:20:27] <rocket> but I havent really had time to track that theory down
[23-Jun-2009 12:20:58] <rmatte> my theory is that emptying out this zenpack will resolve the issue and leave the Mibs behind
[23-Jun-2009 12:21:10] <rmatte> but there are over 72000 items in the zenpack, so doing it by hand would take all week
[23-Jun-2009 12:21:15] <rmatte> I need a quicker way to do it
[23-Jun-2009 12:21:20] <blocky> is there a way I can add a link to another report in a custom device report
[23-Jun-2009 12:21:56] <blocky> just a little bit of html so the email reportmail sends can have a note or two in it
[23-Jun-2009 12:22:29] <rocket> rmatte: so you are thinking just get rid of the objects.xml right?
[23-Jun-2009 12:22:38] <rmatte> well, emptying it out
[23-Jun-2009 12:22:47] <rmatte> but I don't think full out deleting the file will help
[23-Jun-2009 12:23:04] <rmatte> I think the objects need to be removed properly via zenoss
[23-Jun-2009 12:23:23] <rocket> rmatte: if you remove the zenpack does the issue go away?
[23-Jun-2009 12:23:33] <rmatte> yeh, but so do the Mibs
[23-Jun-2009 12:23:53] <rmatte> I essentially want to install the Mibs, then find some way of removing the Zenpack while leaving the Mibs behind
[23-Jun-2009 12:23:57] <rocket> rmatte: have you tried moving the mibs to a new organizer?
[23-Jun-2009 12:24:00] <rmatte> that may it doesn't try to constantly reindex that zenpack
[23-Jun-2009 12:24:08] <rmatte> that way*
[23-Jun-2009 12:24:23] <rocket> and then see if the mibs are still referenced by the zenpack?
[23-Jun-2009 12:24:32] <rmatte> like rename it from cisco to cisco2, remove the zenpack and name it back?
[23-Jun-2009 12:24:36] <rmatte> I'll try it
[23-Jun-2009 12:24:41] <rocket> if not referenced delete zenpack?
[23-Jun-2009 12:24:48] <rocket> yea something like that
[23-Jun-2009 12:24:54] <rocket> make cisco2 organizer
[23-Jun-2009 12:25:07] <rocket> move mibs from cisco to cisco2
[23-Jun-2009 12:25:08] <rmatte> oh, I can't rename it, gah, it'll take like an hour to copy all those Mibs to a new container
[23-Jun-2009 12:25:30] <rocket> do this then make a new organizer
[23-Jun-2009 12:25:32] <rmatte> I'm looking for something simple
[23-Jun-2009 12:25:34] <rocket> cisco2
[23-Jun-2009 12:25:51] <rocket> click on showall for the mib table
[23-Jun-2009 12:26:09] <rocket> then click on All fro the select up by the drop down arrow
[23-Jun-2009 12:26:49] <rocket> nevermind .. just a sec
[23-Jun-2009 12:27:25] <rocket> rmatte: with the cisco mibs can you click on cisco?
[23-Jun-2009 12:27:37] <rocket> err nevermind
[23-Jun-2009 12:27:41] <rocket> gah
[23-Jun-2009 12:28:54] <klinstifen> ckrough: so getOSManufacturerName() returns the value I see on my device's status tab
[23-Jun-2009 12:28:58] <klinstifen> is there a way to set that value
[23-Jun-2009 12:28:59] <rocket> rmatte: anyway if you can figure out something like that .. maybe from the zendmd
[23-Jun-2009 12:29:11] <klinstifen> i dont see a corresponding setOSManufacturerName
[23-Jun-2009 12:33:26] <ckrough> klinstifen: print '\n'.join(dir(d))
[23-Jun-2009 12:33:36] <ckrough> that'll at least make the list a little more readable
[23-Jun-2009 12:34:20] <klinstifen> still don't see it
[23-Jun-2009 12:34:32] <klinstifen> that field can be manually set via the edit tab in the interface
[23-Jun-2009 12:34:44] <klinstifen> but i have a couple dell servers so it would be useful if i could just have the modeler do it
[23-Jun-2009 12:35:14] <ckrough> snmp can interface with Dells openmanage stuff to return that type of info
[23-Jun-2009 12:35:17] <ckrough> I think
[23-Jun-2009 12:35:20] <rocket> klinstifen: might need a dir(d.os) or a dir(d.hw)
[23-Jun-2009 12:36:14] <rmatte> rocket: you can't do a show all on this pack, trust me
[23-Jun-2009 12:36:19] <rmatte> 72000 items
[23-Jun-2009 12:36:23] <rmatte> the page won't even load that long
[23-Jun-2009 12:37:59] <rocket> klinstifen: did you ever look at the aix zenpack?
[23-Jun-2009 12:38:14] <rocket> klinstifen: it sets the manufacturer to IBM for the os
[23-Jun-2009 12:38:31] <rmatte> rocket: that's certainly the goal, using zendmd to clear the zenpack's contents
[23-Jun-2009 12:38:34] <rmatte> just wondering how
[23-Jun-2009 12:39:12] <rocket> klinstifen: om.setOSProductKey=MultiArgs(om.setOSProductKey,"IBM")
[23-Jun-2009 12:39:19] <rocket> from Products.DataCollector.plugins.DataMaps import MultiArgs
[23-Jun-2009 12:39:30] <rocket> where IBM is the manufacturer
[23-Jun-2009 12:39:45] <rocket> klinstifen: I am assuming things are showing up as Unknown for you
[23-Jun-2009 12:42:46] <ionutz> hi
[23-Jun-2009 12:42:53] <rmatte> hi
[23-Jun-2009 12:43:13] <ionutz> I remove some rights to a device from zope interface
[23-Jun-2009 12:43:32] <ionutz> is there any way to reset the permissions to default or remove this device ?
[23-Jun-2009 12:44:02] <ionutz> now devicelist is broken and when I'm trying to access that device I'm redirected to the login page
[23-Jun-2009 12:44:23] <rmatte> you should be able to login with the admin account and just delete the device
[23-Jun-2009 12:44:36] <rmatte> (from the zope management interface)
[23-Jun-2009 12:44:55] <rmatte> though there will be traces of the device elsewhere, so it's a bit tricky
[23-Jun-2009 12:44:59] <ionutz> I'm already logged in as admin
[23-Jun-2009 12:45:06] <ionutz> zport/dmd/manage
[23-Jun-2009 12:45:08] <rmatte> you really need to be careful with permissions when you're messing around in there
[23-Jun-2009 12:45:09] <ionutz> I'm as admin
[23-Jun-2009 12:45:34] <ionutz> and on Device section when I click on that device it redirects me to the login page
[23-Jun-2009 12:45:50] <ionutz> is there other ways to remove this device from zope database ?
[23-Jun-2009 12:45:55] <rmatte> ok, but can you not just put a checkmark next to that device and click Delete?
[23-Jun-2009 12:46:34] <ionutz> I've alrady tried this and I got this error
[23-Jun-2009 12:46:55] <ionutz> Type: BadRequest
[23-Jun-2009 12:46:55] <ionutz> Value: BFIDB01 does not exist
[23-Jun-2009 12:46:55] <ionutz> Traceback (innermost last):
[23-Jun-2009 12:46:55] <ionutz> * Module ZPublisher.Publish, line 114, in publish
[23-Jun-2009 12:46:55] <ionutz> * Module ZPublisher.mapply, line 88, in mapply
[23-Jun-2009 12:46:56] <ionutz> * Module ZPublisher.Publish, line 40, in call_object
[23-Jun-2009 12:46:58] <ionutz> * Module OFS.ObjectManager, line 492, in manage_delObjects
[23-Jun-2009 12:47:00] <ionutz> BadRequest: BFIDB01 does not exist
[23-Jun-2009 12:47:24] <rmatte> is that when you try to remove it from the zope management interface?
[23-Jun-2009 12:47:30] <rmatte> or when you try to remove it from Zenoss?
[23-Jun-2009 12:47:52] <ionutz> from zope management interface
[23-Jun-2009 12:48:02] <ionutz> device section
[23-Jun-2009 12:48:07] <rmatte> well, then I'd say you're pretty stuck
[23-Jun-2009 12:48:31] <rmatte> you should always backup your zope database before trying changes like that
[23-Jun-2009 12:49:31] <ionutz> I have a backup
[23-Jun-2009 12:49:38] <ionutz> I'll try to restore this
[23-Jun-2009 12:49:45] <klinstifen> rocket: yes they are showing up as "unknown"
[23-Jun-2009 12:49:47] <rmatte> how did you do the backup though?
[23-Jun-2009 12:51:08] <rocket> klinstifen: your best example is the aix zenpack and the lines I just showed you
[23-Jun-2009 12:51:25] <ionutz> rmatte from zenoss interface
[23-Jun-2009 12:52:06] <rmatte> I was talking about doing: "cp $ZENHOME/var/Data.fs to $ZENHOME/var/Data.fs.bak
[23-Jun-2009 12:52:23] <rmatte> then if you want to restore the zope DB it's just the opposite
[23-Jun-2009 12:52:48] <rmatte> cp $ZENHOME/var/Data.fs.bak $ZENHOME/var/Data.fs
[23-Jun-2009 12:52:51] <ionutz> oh ... I've made a backup for the whole zenoss
[23-Jun-2009 12:53:02] <ionutz> I'll have a look on this
[23-Jun-2009 12:53:04] <ionutz> thanks
[23-Jun-2009 12:53:09] <rmatte> np
[23-Jun-2009 12:54:01] <ionutz> rmatte there is only a Data.fs.old file from 2 days ago
[23-Jun-2009 12:54:14] <rmatte> too old?
[23-Jun-2009 12:54:20] <ionutz> should I restart zope after the restore
[23-Jun-2009 12:54:26] <rmatte> yes
[23-Jun-2009 12:54:32] <rmatte> zopectl restart
[23-Jun-2009 12:54:48] <ionutz> I think it's fine b/c I didn't made too many changes during these days
[23-Jun-2009 12:54:55] <rmatte> good
[23-Jun-2009 13:01:25] <ionutz> It's still not working ... I suppose that the change was made before this backup ... I'll try to find when the change was made
[23-Jun-2009 13:01:29] <ionutz> thanks for your help
[23-Jun-2009 13:01:38] <rmatte> no problem
[23-Jun-2009 13:04:44] adr3nalin3 is now known as afink
[23-Jun-2009 13:12:01] <zenChild> Anyone know if there is a good article anywhere on RelationshipMap and ObjectMap? I need to create some custom maps in a modeler for nested relations.
[23-Jun-2009 13:24:19] <Sparks> Can anyone tell me what license is used for Zenoss?
[23-Jun-2009 13:24:37] <chudler> GPL2, I think it is
[23-Jun-2009 13:24:48] <mrayzenoss> correct, GPLv2
[23-Jun-2009 13:26:26] <rocket> Hey Matt
[23-Jun-2009 13:26:50] <Sparks> Coolll..... Who do I talk to about possibly packaging it for Fedora? I see the Tray Applet is packaged already.
[23-Jun-2009 13:28:28] <mrayzenoss> Sparks: yeah, I've talked to Dave Nalley about it before
[23-Jun-2009 13:28:34] <Sparks> What did he say?
[23-Jun-2009 13:28:46] <mrayzenoss> well, there were objections to the use of Python 2.4 and Zope 2.8
[23-Jun-2009 13:28:58] <mrayzenoss> but the Zope issue will be fixed in the next release
[23-Jun-2009 13:29:09] <mrayzenoss> and Python upgraded to 2.6 the release after that
[23-Jun-2009 13:29:36] <Sparks> Okay. I'll talk to David and see if he wants to work it.
[23-Jun-2009 13:29:39] <mrayzenoss> we definitely want to get into the distros
[23-Jun-2009 13:30:47] <mrayzenoss> Sparks: email me at mray@zenoss.com and I'll keep in touch as we make progress
[23-Jun-2009 13:30:58] <Sparks> mrayzenoss: Will do
[23-Jun-2009 13:53:12] <venturaville> ok I am a bit lost... given a list a = ['devicename1','devicename2'] ... how do I get it to give me back the equivalent objects in tales ?
[23-Jun-2009 13:58:00] <zenChild> you have to use a "repeat" command. I am far from an expert on TALES, but I just played with it a bit the other day. This guide was a good reference: http://www.owlfish.com/software/simpleTAL/tal-guide.html
[23-Jun-2009 13:58:54] <zenChild> The fruit basket example on that page is a good example of how to traverse a list in TALES.
[23-Jun-2009 15:34:16] <rmatte> hey Matt, is there anyone around there who you could ask about using zendmd to delete the contents of a Zenpack?
[23-Jun-2009 15:34:33] <rmatte> I've been deleting by hand but I've only got maybe 2000 out of 72000 done lol
[23-Jun-2009 15:38:43] <mrayzenoss> I'll poke around a bit
[23-Jun-2009 15:42:06] <rmatte> thanks
[23-Jun-2009 15:42:28] <rmatte> if we can get something to work it might be a first step towards correcting that issue
[23-Jun-2009 15:47:48] <zenChild> How do you get access to the device object from within a CommandParser? I can get the device name via cmd.deviceConfig.device, but how do I fetch the instance?
[23-Jun-2009 15:52:58] <rmatte> zenChild: no idea, I don't have any experience with command parsers
[23-Jun-2009 15:55:33] <zenChild> k. I'll keep digging. Thanks anyway.
[23-Jun-2009 15:56:17] <rmatte> post on the zenoss forums, you might have some luck there
[23-Jun-2009 15:56:44] <zenChild> will do.
[23-Jun-2009 16:00:12] <rmatte> ok, so playing with the objects.xml doesn't work, so I obviously need to use zendmd to get in to the Zope database and remove the Mib items from the Zenpack inventory without actually removing the Mibs themselves
[23-Jun-2009 16:01:23] <tyler79> let's say that i an event. that event relates to some device or service that is managed by a 3rd party. i want to provide contact information along with the event in the alert. what's the best way to do that?
[23-Jun-2009 16:02:14] <rmatte> you'd have to add it in using a python transform for the whole /Event class
[23-Jun-2009 16:03:08] <tyler79> rmatte: that's what i thought. should i be storing the contact info somewhere in the zopedb or mysql though?
[23-Jun-2009 16:03:14] <rmatte> basically saying something like: if evt.device = 'devicename' evt.summary = '%s contact info here' % (evt.summary) or something like that
[23-Jun-2009 16:03:41] <rmatte> you could do that, but that'd be fairly complex
[23-Jun-2009 16:03:49] <rmatte> are you doing this for just 1 device or for all of them?
[23-Jun-2009 16:03:59] <tyler79> rmatte: so a plain text file would be the easiest thing to do?
[23-Jun-2009 16:04:17] <rmatte> if you're only doing it for one device you'd just include the contact info right in the transform
[23-Jun-2009 16:04:28] <rmatte> if it's for multiple devices then you'd need to use zope or something
[23-Jun-2009 16:04:28] <tyler79> rmatte: for many different devices and more complex exvents that don't necessarily relate to a single device
[23-Jun-2009 16:04:59] <rmatte> we have zenoss integrated with OTRS here and the integration script that we wrote takes care of populating info such as that from a seperate database before a ticket is created in OTRS
[23-Jun-2009 16:05:52] <tyler79> rmatte: ok, thanks. that's helpful
[23-Jun-2009 16:17:13] <zenChild> rocket: I read from one of the IRC logs that you were doing some work with multilevel relations... did you ever have any luck figuring this out? The log I am referring to is here: http://www.zenoss.com/community/discuss/irc/archives/dev-chat-04-30-2009
[23-Jun-2009 16:25:50] <mrayzenoss> New Jabber/XMPP Bot ZenPack: http://www.zenoss.com/community/projects/zenpacks/xmppbot
[23-Jun-2009 16:29:04] <rmatte> that is so cool
[23-Jun-2009 16:31:54] <mrayzenoss> rmatte: the only suggestion from devs was to move it to another ZenPack
[23-Jun-2009 16:32:03] <mrayzenoss> rmatte: which you said you were already doing
[23-Jun-2009 16:32:14] <rmatte> move it to another zenpack?
[23-Jun-2009 16:32:24] <rmatte> I'm actually just deleting the contents of the existing zenpack by hand
[23-Jun-2009 16:32:26] <mrayzenoss> yeah, that was Kells' suggstion
[23-Jun-2009 16:32:32] <rmatte> I converted it to development mode then started deleting them
[23-Jun-2009 16:32:36] <mrayzenoss> sounds like the same difference
[23-Jun-2009 16:32:43] <rmatte> but it's taking forever since I can't do all of them at once
[23-Jun-2009 16:33:05] <rmatte> I'm basically just looking for the function that Zenoss is calling when it's deleting the items
[23-Jun-2009 16:33:15] <rmatte> then maybe I can harness that function and make a script to do it quicker
[23-Jun-2009 16:34:04] <rmatte> deleting 72000 objects 40 at a time by hand with a delay of like 10-30 seconds per deletion is going to take me a whole month to delete all the objects lol
[23-Jun-2009 16:36:09] <mrayzenoss> rmatte: go ahead and post it on zenoss-dev, Kells said he'd look at it
[23-Jun-2009 16:36:16] <rmatte> k cool
[23-Jun-2009 16:36:28] <rmatte> thanks
[23-Jun-2009 16:36:35] <mrayzenoss> rmatte: or you could open a support case
[23-Jun-2009 16:36:51] <rmatte> doesn't that cost money?
[23-Jun-2009 16:38:19] <mrayzenoss> apparently I had rmatte confused with someone else, zenoss-dev it is!
[23-Jun-2009 16:38:39] <rmatte> hehe
[23-Jun-2009 16:46:38] <rmatte> http://forums.zenoss.com/viewtopic.php?p=36242#36242
[23-Jun-2009 16:46:39] <rmatte> voila
[23-Jun-2009 16:47:06] <rmatte> if you could pass that around it'd be appreciated, if this works I'm sure I won't be the only one happy about the workaround
[23-Jun-2009 17:01:42] <rmatte> did you ever ask about Zenoss not check port's actual operational status except during modelling?
[23-Jun-2009 22:13:42] <Brad_> I am having problems with the zenoss install and was hoping for help
[23-Jun-2009 22:15:50] <perr0> ...
[23-Jun-2009 22:19:34] <Brad_> My install is hanging at the setting up zenoss-stack phase
[23-Jun-2009 22:19:55] <Brad_> dpkg: error processing zenoss-stack (--configure):
[23-Jun-2009 22:20:15] <Brad_> when i hit cntrl c i to stop the process i get the message above
[23-Jun-2009 22:20:49] <perr0> what OS and what pkg are you using to install zenoss
[23-Jun-2009 22:21:26] <Brad_> ubuntu 8.04 (64bit) and zenoss 2.4.1...today i tried 2.4.2
[23-Jun-2009 22:21:56] <Brad_> i am using the apt-get method, but i also tried using dpkg -i
[23-Jun-2009 22:22:00] <Brad_> same results
[23-Jun-2009 22:22:17] <perr0> mm ya I would only use the pkg from the zenoss site
[23-Jun-2009 22:22:30] <perr0> I don't use ubuntu sorry
[23-Jun-2009 22:24:48] <perr0> night Brad, try tomorrow is lot of traffic during the day
[23-Jun-2009 22:24:57] <Brad_> take it easy
[23-Jun-2009 22:39:24] <Brad_> anyone know why a zenoss install would hang after the step Setting up zenoss-stack (2.4.1-0) ...
[24-Jun-2009 00:00:42] [disconnected at Wed Jun 24 00:00:42 2009]
[24-Jun-2009 00:00:42] [connected at Wed Jun 24 00:00:42 2009]
[24-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[24-Jun-2009 07:04:04] ckrough_ is now known as ckrough
[24-Jun-2009 08:49:38] <rocket> zenChild: ping
[24-Jun-2009 08:56:40] <ckrough> destination unreachable
[24-Jun-2009 09:00:52] <zenChild> yeah rocket?
[24-Jun-2009 09:01:20] <zenChild> ckrough: had to reroute the transmission
[24-Jun-2009 09:53:13] <rocket> zenChild: you were looking for me?
[24-Jun-2009 09:54:12] <zenChild> Yeah. I had a question for you yesterday. I saw from a irc log that you had been working on multilevel relationships with what looked like LVM. I was just wondering if you had any luck in doing so.
[24-Jun-2009 09:54:51] <rocket> zenChild: yeah its in the aix snmp zenpack
[24-Jun-2009 09:55:13] <zenChild> sweet. I'll look there then. Thanks.
[24-Jun-2009 09:55:31] <rocket> zenChild: it isnt necessarily very straight forward ..
[24-Jun-2009 09:57:18] <zenChild> rocket: I'll look it over and if you don't mind I might have a couple questions for you. I'm trying to set up a multilevel relationship like so: host(1toM ) -> diskpool ->(1toM) -> logical disks
[24-Jun-2009 09:58:12] <rocket> sure thats doable ...
[24-Jun-2009 09:58:18] <zenChild> thanks.
[24-Jun-2009 09:58:27] <rocket> zenChild: you are going to have to override some stuff from the very beginning
[24-Jun-2009 09:58:41] <rocket> eg starting with the Device object
[24-Jun-2009 09:59:07] <zenChild> eek.... not looking forward to that.
[24-Jun-2009 09:59:30] <zenChild> I'll take a look at it this afternoon. I gotta run to a meeting now. Thanks for the tips though.
[24-Jun-2009 09:59:30] <rocket> zenChild: need to do so to get a custom os or hw subobject ..
[24-Jun-2009 11:04:40] <zenChild> rocket: AIXVolumeGroupMap.py is exactly what I was looking for. It also answers my question as to whether I could split up the modelers or not. Thanks again for sending me in the right direction. I have to admit my eyes are bleeding a bit from the first run-through of the code, but I think after a few more times my understanding will hopefully improve. Cheers!
[24-Jun-2009 11:13:47] <rocket> zenChild: no problem
[24-Jun-2009 11:13:55] <rocket> zenChild: I had to figure it out blind
[24-Jun-2009 11:15:34] <zenChild> I'm impressed. I think I would have started drinking rather heavy by the time I figured it out on my own.
[24-Jun-2009 11:16:59] <rocket> zenChild: as you can see the trick is to have the modeler call special functions that reference the new subobjects
[24-Jun-2009 11:17:17] <rocket> basically build a bunch of nested dicts to go as deep as you need
[24-Jun-2009 11:18:57] <zenChild> rocket: It actually looks quite elegant compared to the hash madness that I had started in my modeler code. Hopefully I can get that part working without too much more effort. I think I know how my maps are supposed to look. Now I just need to get the code to do my bidding.
[24-Jun-2009 11:20:15] <rocket> zenChild: great ... and feel free to put in a good word for the zenpack contest
It cant hurt [24-Jun-2009 11:20:54] <zenChild> will do. I already posted kudos on the forum post I had open on the issue.
[24-Jun-2009 12:01:31] <rmatte> rocket: http://forums.zenoss.com/viewtopic.php?t=10154 <--- Any idea how to injects objects in to the zope database without actually installing as a zenpack? I essentially want to load the Mibs in without having it register as a Zenpack, that should resolve the performance issues until they can actually fix it properly.
[24-Jun-2009 12:21:56] <rocket> rmatte: could you do a workaround by scripting zenmib to load your mibs in the right order?
[24-Jun-2009 12:23:29] <rocket> rmatte: eg .. drop mibs that go in first into a folder .. run zenmib against it .. remove those files .. put next ones in .. run zenmib etc
[24-Jun-2009 12:23:56] <rocket> rmatte: zenmib will load the mibs into zope without doing zenpack stuff etc and its command line scriptable ..
[24-Jun-2009 12:26:25] <rmatte> I guess I could, it'd be a pain though, the cisco Mibs don't load very cleanly to begin with
[24-Jun-2009 12:26:31] <rmatte> it was a battle just to get them to load
[24-Jun-2009 12:26:39] <rmatte> the only sure-fire way to always have them load is the zenpack
[24-Jun-2009 12:26:51] <rmatte> which I apparently can't use due to a bug in Zenoss
[24-Jun-2009 12:29:58] <rmatte> I've had people follow the same procedure that I did with the same Mibs and not had them all load properly
[24-Jun-2009 12:31:17] <rocket> rmatte: and you are certain its a bug in the zenpack code?
[24-Jun-2009 12:33:15] <rmatte> it's not really a bug in the zenpack code...
[24-Jun-2009 12:33:58] <rmatte> when you try to load zProperties, or the ZenPacks page (as well as a couple others) Zenoss decides that it's going to reindex all ZenPacks, and you have to wait while it does this before the page will load
[24-Jun-2009 12:34:09] <rmatte> so with my ZenPack being huge, it takes upwards of 30 minutes
[24-Jun-2009 12:34:53] <rmatte> I need to find a way to inject the Mibs in to Zope so that they can be used without Zenoss deciding to reindex everything
[24-Jun-2009 12:35:37] <rmatte> That way the Mibs will be (theoretically) usable but won't show up as part of a zenpack
[24-Jun-2009 12:36:04] <rmatte> as I recall though, performance was much much better on my initial install of Zenoss where I originally installed the Mibs
[24-Jun-2009 12:36:15] <rmatte> on the second install with the Zenpack performance dropped drastically
[24-Jun-2009 12:39:03] <rocket> rmatte: I see .. so it sounds like zope is trying to index on a view when it should really be indexing on a change etc
[24-Jun-2009 12:40:37] <rmatte> exactly
[24-Jun-2009 12:40:54] <rmatte> by the way, nice work on that Linux monitoring patch, it got merged
[24-Jun-2009 12:41:08] <rocket> so in reality indexes are borked .. I messed with them a bit but dont fully understand them
[24-Jun-2009 12:41:30] <rmatte> although I think they merged the old patch, not the new one
[24-Jun-2009 12:41:32] <rocket> thats one area of code that probably is in need of a decent rewrite ..
[24-Jun-2009 12:41:44] <rmatte> yeh
[24-Jun-2009 12:42:10] <rmatte> there's no reason for it to be reindexing a zenpack with over 72000 items in it every time someone wants to access a zProperties page
[24-Jun-2009 12:42:13] <rocket> rmatte: my extended monitoring linux zenpack includes teh patch as well
[24-Jun-2009 12:42:28] <rmatte> ah, cool
[24-Jun-2009 12:42:36] <rmatte> did that get published?
[24-Jun-2009 12:42:38] <rocket> eg it will apply it if its not already done
[24-Jun-2009 12:42:41] <rocket> yes
[24-Jun-2009 12:42:44] <rmatte> nice
[24-Jun-2009 12:42:48] <rmatte> u da man
[24-Jun-2009 12:43:05] <rocket> that fixes network perf graphin and routes etc
[24-Jun-2009 12:43:49] <rocket> so if it was me I would just install both
[24-Jun-2009 12:44:43] <rmatte> hehe
[24-Jun-2009 12:45:20] <rmatte> one thing I'd love to see a patch for would be http://dev.zenoss.com/trac/ticket/5070
[24-Jun-2009 12:45:52] <rmatte> they said they won't fix it, but quite honestly, passing usernames and passwords directly is the most efficient way, I tried using python scripts to do it and it caused major performance hits
[24-Jun-2009 12:46:05] <rmatte> maybe my code isn't as good as it could be
[24-Jun-2009 12:48:32] <rocket> rmatte: what was the reason it was turned into a wont fix? I dont see that in the history of the ticket
[24-Jun-2009 12:48:52] <rmatte> "See community maintainer for a fix"
[24-Jun-2009 12:48:55] <rmatte> errr
[24-Jun-2009 12:49:00] <rmatte> See ZenPack maintained for a fix
[24-Jun-2009 12:49:01] <rmatte> rather
[24-Jun-2009 12:49:22] <rocket> so they are taking the approach its a zenpack issue ...
[24-Jun-2009 12:49:34] <rmatte> yeh, they are being lazy in my opinion
[24-Jun-2009 12:49:54] <rocket> I can see some of that I suppose .. but probably it should have been filtered before getting into the zenpack
[24-Jun-2009 12:49:56] <rmatte> they really should take security more seriously, passwords stored in Zenoss should be encrypted and sensored
[24-Jun-2009 12:50:33] <rmatte> instead of being stored in plaintext and dumped in plaintext
[24-Jun-2009 12:51:07] <chudler> would encrypting them help security?
[24-Jun-2009 12:51:46] <rmatte> chudler: yup
[24-Jun-2009 12:52:10] <rmatte> chudler: I can open up the zope DB file in a text editor, search for "password" and find all the passwords stored in it
[24-Jun-2009 12:52:11] <chudler> against?
[24-Jun-2009 12:52:28] <rmatte> pretty lax
[24-Jun-2009 12:52:30] <chudler> doesnt the encryption key have to be stored? you're only trading several secrets for one
[24-Jun-2009 12:53:13] <rocket> chudler: its one of those issues where if its displayed for a support desk person to see it then it is a security issue
[24-Jun-2009 12:53:16] <chudler> also, it seems the problem would be reading the zopeDB. can the attacker get that file?
[24-Jun-2009 12:53:22] <rmatte> so your point is that security through obscurity is better than at least attempting to make it more difficult for someone to grab sensitive information?
[24-Jun-2009 12:53:26] <rocket> eg the support desk person doesnt have full access to the system
[24-Jun-2009 12:53:29] <chudler> I agree that displaying the password is a bad thing, though
[24-Jun-2009 12:54:43] <rmatte> the whole database should require a username and password to be accessed ideally, and any password variables being passed to commands should be sensored
[24-Jun-2009 12:55:00] <rmatte> and the whole db should be encrypted, not just the passwords
[24-Jun-2009 12:55:24] <rmatte> large corporations are using this product
[24-Jun-2009 12:55:26] <chudler> it is just a file afterall... you can encrypt it
[24-Jun-2009 12:55:30] <rocket> rmatte: you might be going a bit to far with that last statement .. it might make performance royally suck ..
[24-Jun-2009 12:55:55] <rmatte> yeh, I can see performance sucking with an encrypted db, but encrypting the passwords would certainly help
[24-Jun-2009 12:56:16] <rmatte> sensoring them when commands fail is the main problem
[24-Jun-2009 12:56:29] <rocket> eg i have lots of big systems here that arent fully encrypted etc .. and if they were they would crush the hardware we can afford
[24-Jun-2009 12:56:35] <rmatte> "command failed: blah <username> <password>" isn't very secure to have kicking around
[24-Jun-2009 12:57:23] <rmatte> that's really what I logged the problem for in the first place was the command failed events that get generated and display the info in plaintext
[24-Jun-2009 12:57:54] <rocket> rmatte: can you do a transform yourself of that type of event?
[24-Jun-2009 12:58:11] <rocket> rmatte: and maybe post a howto for the zenpacks that are broken that way?
[24-Jun-2009 12:58:34] <rmatte> how do you determine which part is the command and which part is the password?
[24-Jun-2009 12:58:49] <rmatte> it's just a string of text, the sensoring needs to be done as the event is being generated
[24-Jun-2009 12:58:51] <rmatte> not afterwards
[24-Jun-2009 12:59:09] <rmatte> because ideally you still want to know what command failed
[24-Jun-2009 13:00:49] <mloven> @rmatte: couldn't you search the message for the zWinPassword (or whatever field it is for that device), and if it matches it changes the string to "blah", or something? If it doesn't find that string, it just prints the message as is?
[24-Jun-2009 13:12:14] <rmatte> mloven: I suppose you could, it'd be interesting
[24-Jun-2009 13:12:24] <rmatte> mloven: my python is not exactly prime yet
[24-Jun-2009 13:12:39] <rmatte> especially when it comes to string manipulation
[24-Jun-2009 13:16:54] <mloven> @rmatte: oh, my python isn't so good either, I just know that within the custom device reports, I can put in the zWinuser or zWinPassword fields and get the string. I assume it would be the same for zCommandPassword
[24-Jun-2009 13:21:30] <rmatte> zWinPassword is actually what most of the problematic zenpacks are using
[24-Jun-2009 13:21:35] <rmatte> since they are WMI performance packs
[24-Jun-2009 13:40:57] <rmatte> I still think it'd be more effective to have the sensoring done by Zenoss itself
[24-Jun-2009 13:41:23] <rmatte> that way you don't have to fool around with transforms or with making scripts access zope when they don't need to
[24-Jun-2009 13:42:03] <rmatte> but I'm pretty sure that'll never get done
[24-Jun-2009 13:47:40] <rocket> rmatte: its just a matter of building the appropriate api for zenpacks .. at the moment the api is fairly basic
[24-Jun-2009 13:52:01] <rmatte> too bad the google maps portlet doesn't support satelite view
[24-Jun-2009 13:52:05] <rmatte> would look awesome
[24-Jun-2009 13:52:06] <rmatte> hehe
[24-Jun-2009 13:52:37] <rmatte> rocket: agreed
[24-Jun-2009 13:53:14] <mrayzenoss> http://forums.zenoss.com/viewtopic.php?t=4231&highlight=satellite
[24-Jun-2009 13:53:39] <mrayzenoss> feel free to open an enhancement ticket to expose that in the UI (or better yet, submit a patch)
[24-Jun-2009 13:54:14] <venturaville> how do I get a zproperty to attach to a device class and not to the zenpack itself (ala ___init__.py)
[24-Jun-2009 13:55:05] <ckrough> you guys use the goole maps porlet for real, or just for fun?
[24-Jun-2009 13:58:39] <rmatte> interesting, that google maps thing started out as a core zenpack?
[24-Jun-2009 13:58:52] <rmatte> I use the google maps portlet for real
[24-Jun-2009 13:59:12] <rmatte> example: http://dmon.org/graphics/zenoss.png
[24-Jun-2009 13:59:14] <mrayzenoss> the Google maps predate ZenPacks
[24-Jun-2009 13:59:40] <mrayzenoss> I take that back, .egg ZenPacks
[24-Jun-2009 13:59:42] <rmatte> ah, cause I was reading http://morgana.sourceforge.net/
[24-Jun-2009 14:00:04] <rmatte> and yeh, it was a zip pack
[24-Jun-2009 14:01:38] <rmatte> by the way, two colour blind people at my work made the comment that the status icons for iterface status are hard to see, they suggested moving it back to colored UP/DOWN cells for accessiblity, just food for thought
[24-Jun-2009 14:02:21] <mrayzenoss> yeah, there's a ticket for that that I keep trying to get fixed
[24-Jun-2009 14:02:33] <mrayzenoss> one of our devs is color-blind
[24-Jun-2009 14:02:48] <mrayzenoss> not sure why he hasn't fixed it...
[24-Jun-2009 14:02:57] <rmatte> it's actually more common than most people think
[24-Jun-2009 14:03:13] <rmatte> especially in the computer industry surprisingly
[24-Jun-2009 14:03:25] <rmatte> I know 5 color blind people in the industry
[24-Jun-2009 14:15:50] <zenChild> Can someone explain to me exactly what Products.DataCollector.plugins.DataMaps.MultiArgs does? I think I have an idea, but it's a partly-cloudy idea.
[24-Jun-2009 14:22:10] <rmatte> it's key in determining device dependencies/relationships
[24-Jun-2009 14:22:36] <rmatte> that's about all I know about it
[24-Jun-2009 14:31:26] <rmatte> I like the comic on the front page now, very artistic
[24-Jun-2009 14:31:28] <rmatte> hehe
[24-Jun-2009 14:32:02] <mrayzenoss> venturaville: I can't find an example of a ZenPack-defined zProperty that is restricted to a single Device class
[24-Jun-2009 14:32:57] <venturaville> mrayzenoss: argh ... I have it made global for now in the absence of that
[24-Jun-2009 14:33:04] <rmatte> you could dig in and see how it's done with the zWinUser property
[24-Jun-2009 14:33:14] <venturaville> rmatte: will do
[24-Jun-2009 14:33:59] <rmatte> I actually have the developers IRC sessions as a recurring meeting in my calendar now hehe
[24-Jun-2009 14:35:44] <rmatte> by the way, why does the RSS feed only show titles? If you look at something like slashdot's it has an actual summary of each article in it. for the zenoss one you have to actually go to the site to read the summary
[24-Jun-2009 14:37:29] <mrayzenoss> which RSS feed?
[24-Jun-2009 14:37:35] <rmatte> the zenoss one
[24-Jun-2009 14:38:36] <mrayzenoss> for the blog?
[24-Jun-2009 14:38:44] <rmatte> I assume so
[24-Jun-2009 14:38:54] <rmatte> title is Zenoss - All The News
[24-Jun-2009 14:44:48] <rmatte> http://feeds.feedburner.com/zenoss/MbQB
[24-Jun-2009 14:46:08] <mrayzenoss> hmmm... that appears to be the feed generated by Zope for the "Events"
[24-Jun-2009 14:46:39] <mrayzenoss> well, the quick answer is we probably won't fix it, since it'll be gone in the next month or so
[24-Jun-2009 14:46:42] * mrayzenoss crosses fingers
[24-Jun-2009 14:47:17] <rmatte> hehe
[24-Jun-2009 14:48:45] <mrayzenoss> of course, I say the next month or so but everything takes longer than expected
[24-Jun-2009 14:49:02] <mrayzenoss> oops
[24-Jun-2009 14:52:10] <rmatte> hehe
[24-Jun-2009 14:52:37] <rmatte> anyways, I'm done my shift so time to have beer and play video games w00t
[24-Jun-2009 14:52:45] <rmatte> (worked from home today so no long bus ride home either)
[24-Jun-2009 14:58:08] <mrayzenoss> zWinUser is a zProperty on all Devices, only used by /Server/Windows
[24-Jun-2009 14:58:25] <rmatte> ah
[24-Jun-2009 14:58:30] <rmatte> maybe there is no way to specify
[24-Jun-2009 14:58:42] <rmatte> maybe it just relies on people's common sense
[24-Jun-2009 14:58:43] <rmatte> hehe
[24-Jun-2009 14:58:58] <venturaville> it is mostly just an annoyance ... no huge deal to me
[24-Jun-2009 15:00:40] <rmatte> as long as it's there I could personally care less
[24-Jun-2009 15:00:49] <rmatte> just name it something obvious
[24-Jun-2009 15:00:58] <rmatte> like zWinUser is obviously not applicable to a Linux box
[24-Jun-2009 15:01:45] <mrayzenoss> well, at one point I had "zZenPackCommunityVMwareESXlibexec" as a zProperty
[24-Jun-2009 15:01:55] <mrayzenoss> found the fix for that
[24-Jun-2009 15:02:13] <mrayzenoss> Chet has spoken
[24-Jun-2009 15:02:23] <mrayzenoss> No, zProperties are top-down.
[24-Jun-2009 15:02:23] <mrayzenoss> They must exist at / to exist anywhere.
[24-Jun-2009 15:03:58] <suprsonic> ./libxml/encoding.h:29:19: error: iconv.h: No such file or directory FreeBSD 7.2
[24-Jun-2009 15:04:05] <suprsonic> thoughts?
[24-Jun-2009 15:06:30] <mrayzenoss> suprsonic: Casey Hillman's FreeBSD howto is down, give me your email address and I'll send you a copy
[24-Jun-2009 15:36:33] <rhetts> is there a way to disable compression w/ zenbackup?
[24-Jun-2009 16:12:46] <tyler79> any pynetsnmp users here? i figured out how to do a get on an oid using a SnmpSession object, but I can't figure out how to walk the snmp tree
[24-Jun-2009 16:13:08] <tyler79> actually, what I'm really looking for is some good pynetsnmp documentation (if it exists)
[24-Jun-2009 16:29:26] <mrayzenoss> tyler79: you may have better luck tomorrow when there's a developer in the channel
[24-Jun-2009 16:30:25] <tyler79> mrayzenoss: ok, thanks
[24-Jun-2009 17:10:13] <tripitaka> hello. My device was down, but now it has recovered and I can ping it from the zenoss host. Zenoss is still showing it as down, is there any way I can force a new check?
[24-Jun-2009 17:10:28] <tripitaka> I've tried remoddeling it but no joy
[25-Jun-2009 00:00:42] [disconnected at Thu Jun 25 00:00:42 2009]
[25-Jun-2009 00:00:42] [connected at Thu Jun 25 00:00:42 2009]
[25-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[25-Jun-2009 09:06:00] <Appiah_> how do you remove mibs on zenoss 2.4.1 ?
[25-Jun-2009 09:34:01] <venturaville> for those developing zenpacks in svn ... useful code snippet:
[25-Jun-2009 09:34:03] <venturaville> #!/bin/sh
[25-Jun-2009 09:34:03] <venturaville> ZENPACKNAME="ZenPacks.community.NetApp"
[25-Jun-2009 09:34:03] <venturaville> PYVERSION="py2.4"
[25-Jun-2009 09:34:03] <venturaville> ZENPACKVERSION=`grep -i ^Version EGG-INFO/PKG-INFO | awk ' { print $2 } '`
[25-Jun-2009 09:34:03] <venturaville> echo "Building ZenPack: ${ZENPACKNAME}"
[25-Jun-2009 09:34:04] <venturaville> zip -r ${ZENPACKNAME}-${ZENPACKVERSION}-${PYVERSION}.egg EGG-INFO ZenPacks -x \*/.svn\*
[25-Jun-2009 09:34:06] <venturaville> echo "Built: ZenPack EGG file: ${ZENPACKNAME}-${ZENPACKVERSION}-${PYVERSION}.egg"
[25-Jun-2009 10:00:02] <bwest> hello
[25-Jun-2009 10:01:03] <mrayzenoss> Greetings all, mrchippy will be joining me shortly
[25-Jun-2009 10:01:29] <bwest> Quick question Is there a way to monitor the Billiant Service on windows machine?
[25-Jun-2009 10:02:16] <bwest> This service essentially send queries to Ms SQL to determine the number of placed orders, orders in queue and succesful orders
[25-Jun-2009 10:02:53] <mrayzenoss> bwest: I've got a MS SQL ZenPack I'm hoping to post later today
[25-Jun-2009 10:03:03] <mrayzenoss> or you can monitor the service
[25-Jun-2009 10:03:04] <bwest> Will that do the trick?
[25-Jun-2009 10:04:17] <atrawog> Hi Mray! Have you got my email?
[25-Jun-2009 10:04:50] <bwest> mray: Will the new Ms SQL zenpack allow me define the queries I send?
[25-Jun-2009 10:05:13] <mrayzenoss> bwest: let me check
[25-Jun-2009 10:05:39] <mrayzenoss> atrawog: yeah, sorry I haven't responded
[25-Jun-2009 10:05:56] <mrayzenoss> atrawog: I'll get the Brocade update out after the 5 DB ones from Egor
[25-Jun-2009 10:05:59] <mrayzenoss> he's on a tear
[25-Jun-2009 10:06:47] <mrayzenoss> bwest: I gotta fire up a VM to install the MS SQL one, I'll get back to you
[25-Jun-2009 10:06:55] <mrayzenoss> bwest: the docs are somewhat limited
[25-Jun-2009 10:06:58] <bwest> mrchippy: I am monitoring servers in the SSH/Linux class and I want to pull memory stats off those servers. Can I use a plugin or do I have to creat a new zenpack, and parser?
[25-Jun-2009 10:07:17] <mrayzenoss> bwest: check out the LinuxMonitorAddOn ZenPack
[25-Jun-2009 10:07:48] <zenChild> Can somewhat tell me exactly what Products.DataCollector.plugins.DataMaps.MultiArgs does? How does the actual structure look that it is creating?
[25-Jun-2009 10:08:44] <bwest> mray: are you saying as a example because I thought that only provided information on interfaces?
[25-Jun-2009 10:09:07] <mrayzenoss> bwest: no, the LinuxMonitorAddon rolls up a patch to the LinuxMonitor to add memory
[25-Jun-2009 10:09:57] <klinstifen> bwest: if that doesnt' work you could create a zenpack
[25-Jun-2009 10:10:00] <mrayzenoss> there's a patch for the LinuxMonitor ZenPack that adds memory and interfaces that will be included in the next release, but the LinuxMonitorAddOn has that patch now
[25-Jun-2009 10:10:09] <klinstifen> i did that to get print queue information from our aix box
[25-Jun-2009 10:10:10] <mrchippy> zenChild: some of the map keys are methods that can take more than one arg (like setHwProductKey, iirc). you pass a multiarg object for that
[25-Jun-2009 10:10:12] <klinstifen> worked pretty well
[25-Jun-2009 10:10:24] <mrchippy> zenChild: looking for example
[25-Jun-2009 10:10:44] <bwest> mray: I see because I have installed that zenpack but when I go to the memory reports nothing shows up
[25-Jun-2009 10:12:03] <zenChild> I think my confusion was if it was actually a key or a value in the dictionary. I'm trying to wrap my head around how a key can be more than one value.
[25-Jun-2009 10:12:07] <klinstifen> i have a generic question about threshholds
[25-Jun-2009 10:12:19] <klinstifen> i created a custom graph definition
[25-Jun-2009 10:12:31] <klinstifen> and then tried to set a threshhold against one of the datapoints
[25-Jun-2009 10:12:36] <klinstifen> but it didn't seem to work
[25-Jun-2009 10:13:00] <klinstifen> do threshholds only work against datapoints added through the datasource interface?
[25-Jun-2009 10:13:33] <mrayzenoss> klinstifen: what were you trying to use the Threshold against?
[25-Jun-2009 10:13:51] <mrayzenoss> when you add the Threshold, you have to pick one of the available data points
[25-Jun-2009 10:13:53] <rmatte> bwest: did you actually install the LinuxMonitorAddOn ZenPack, or the patch for the standard one?
[25-Jun-2009 10:14:02] <rmatte> if you haven't done either of those things then you won't get the memory info
[25-Jun-2009 10:14:21] <bwest> the LinuxMonitorAddOn Zenpack
[25-Jun-2009 10:14:34] <mrayzenoss> bwest: have you remodeled?
[25-Jun-2009 10:14:44] <rmatte> yeh, you need to remodel afterwards
[25-Jun-2009 10:14:52] <bwest> yes a number of times
[25-Jun-2009 10:15:27] <rmatte> did you locally define the collector plugins at some point?
[25-Jun-2009 10:15:31] <mrchippy> zenchild: it's the value. check out the NewDeviceMap plugin in trunk/Products/DataCollector/plugins/zenoss/snmp. It maps the hwproductkey name and manufacturer using MultiArgs (lines ~80 - 90)
[25-Jun-2009 10:15:31] <bwest> rmatte: I need to get memavail, cached etc
[25-Jun-2009 10:15:35] <klinstifen> mray: i created some custom datapoints for an exchange zenpack i made...i wanted a threshold to go against the datapoints
[25-Jun-2009 10:15:57] <klinstifen> but i couldn't select it because it wasn't a datapoint that was available through the interface...since the datapoints were made through the custom definition
[25-Jun-2009 10:16:02] <bwest> rmatte: Locally define the collector plugins?
[25-Jun-2009 10:16:06] <rmatte> bwest: that would be something under the perf tab, no?
[25-Jun-2009 10:16:11] <venturaville> btw: devices.findDevice('blah.blah') does not work because of the dots in the name..........
[25-Jun-2009 10:16:24] <bwest> rmatte: yes
[25-Jun-2009 10:16:34] <bwest> but now all I get is CPU
[25-Jun-2009 10:16:46] <zenChild> mrchippy: k. I'll check it. Thanks for the tip.
[25-Jun-2009 10:17:00] <mrchippy> venturaville: checking that now...
[25-Jun-2009 10:17:03] <rmatte> sounds like the template didn't update
[25-Jun-2009 10:17:21] <rocket> bwest: as far as I know that information isnt gathered yet
[25-Jun-2009 10:17:36] <rmatte> rocket: ah, ok
[25-Jun-2009 10:17:40] <mrayzenoss> klinstifen: so you added the datasource and a data point and when you go to add a threshold the datasource_datapoint isn't there?
[25-Jun-2009 10:18:14] <atrawog> @mray There is no hurry with my Brodade ZenPack. I just wanted to check if a Mercurial repo is ok for you.
[25-Jun-2009 10:18:15] <rmatte> bwest: rocket actually made that addon zenpack so he'd be best to answer your questions hehe
[25-Jun-2009 10:18:16] <bwest> rocket: ok will it be in the next release? if so when will that be released
[25-Jun-2009 10:18:35] <mrayzenoss> atrawog: yeah, that should be OK. I'll figure out how to pull it over
[25-Jun-2009 10:20:16] <rocket> bwest: send me the location of the commands and the data you need and I will see what I can do .. no gaurantees
[25-Jun-2009 10:20:34] <mrchippy> venturaville: i was able to do "dmd.Devices.findDevice('mybox.zenoss.loc')" and it worked. it has to be an exact device id match. is this what you did?
[25-Jun-2009 10:20:43] <rocket> bwest: it will be an extension of the advanced linux monitor zenpack then
[25-Jun-2009 10:21:49] <rocket> bwest: depending on my time I may be able to get to it .. I am working on something with zenoss that might prevent me from working on this though .. thats something I will have to discuss with matt when the time comes
[25-Jun-2009 10:22:06] <klinstifen> mray: the datapoints don't exist to be selected because they were created in the custom definition
[25-Jun-2009 10:22:22] <klinstifen> mray: i took two datapoints and divided them against each other...i wanted to set a threshold against that
[25-Jun-2009 10:22:22] <venturaville> mrchippy: yeah I am doing this: dmd.Devices.findDevice('www.myhostname.com')
[25-Jun-2009 10:22:26] <venturaville> and it returns None
[25-Jun-2009 10:22:52] <mrchippy> venturaville: also, find() and dmd.Devices.findDevice() use zope catalog glob searches so * is an option.
[25-Jun-2009 10:23:14] <venturaville> >>> dmd.Devices.Site.getSubDevices_recursive()
[25-Jun-2009 10:23:33] <venturaville> [<Device at /zport/dmd/Devices/Site/devices/www.myhostname.com>]
[25-Jun-2009 10:23:47] <venturaville> (names changed to protect the innocent)
[25-Jun-2009 10:24:01] <mrchippy> venturaville: that probably means it's not in the catalog (maybe)
[25-Jun-2009 10:24:13] <venturaville> how would that happen?
[25-Jun-2009 10:24:24] <bwest> Another question The Admin Password shows up in plain text when I run a test of the WMIC commands as well in event summary can I stop this
[25-Jun-2009 10:24:53] <mrchippy> venturaville: try this-- dmd.Devices.Site.devices._getOb('www.myhostname.com').index_object()
[25-Jun-2009 10:25:25] <venturaville> None
[25-Jun-2009 10:25:26] <mrchippy> venturaville: how that would happen-- not sure. some goof somewhere. (hopefully not one of mine
)[25-Jun-2009 10:25:35] <rocket> bwest: which zenpack are you using for that?
[25-Jun-2009 10:25:48] <mrayzenoss> klinstifen: so you've programatically added the datapoints, but they don't show up for selection in the Threshold... I guess you could add a different datapoint, and then go change it in the code to your custom one?
[25-Jun-2009 10:25:56] <venturaville> I'm guessing broken zenpack somehow...looking into clean it up now
[25-Jun-2009 10:26:02] <bwest> I have tried the zenoss one and the external one
[25-Jun-2009 10:26:11] <mrchippy> venturaville: dmd.Devices.Site.devices()[0]
[25-Jun-2009 10:26:12] <bwest> at the moment none
[25-Jun-2009 10:26:21] <mrchippy> venturaville: what does that give you?
[25-Jun-2009 10:26:33] <bwest> rocket:I have tried to run it as a non admin with no success
[25-Jun-2009 10:26:35] <venturaville> <Device at /zport/dmd/Devices/Site/devices/www.myhostname.com>
[25-Jun-2009 10:26:50] <bwest> rocket: can't monitor any services
[25-Jun-2009 10:26:53] <klinstifen> mray: hmm...could try that
[25-Jun-2009 10:27:05] <mrchippy> venturaville: then try dmd.Devices.Site.devices()[0].index_object()
[25-Jun-2009 10:27:19] <mrchippy> venturaville: it won't return anything but then try the find again
[25-Jun-2009 10:27:26] <klinstifen> mray: set that datapoint to "not drawn"?
[25-Jun-2009 10:27:29] <venturaville> None
[25-Jun-2009 10:27:37] <venturaville> voila
[25-Jun-2009 10:27:40] <venturaville> magic
[25-Jun-2009 10:28:05] <klinstifen> mray: err...if i set it to "not drawn" will the threshold still evaluate or will that setting cause the threshold to ignore the values?
[25-Jun-2009 10:28:18] <rmatte> in zendmd, is there a quicker way than doing something like mydev = dmd.Devices.findDevice('DeviceName') \ print mydev.zWinPassword to retrieve the contens of a zProperty?
[25-Jun-2009 10:28:43] <paulh> d=find('DevNme')
[25-Jun-2009 10:28:57] <rocket> bwest: I havent done anything with wmi yet .. but from what I have heard the only way to keep the password from showing would be to update the zenpack causing it ... or do an event transform that strips the information ..
[25-Jun-2009 10:29:02] <rmatte> paulh: identical speed, I meant quicker as in retrieval time
[25-Jun-2009 10:29:03] <mrchippy> rmatte: yeah, what paulh said
[25-Jun-2009 10:29:12] <paulh> got it
[25-Jun-2009 10:29:16] <paulh> don't think so
[25-Jun-2009 10:29:17] <mrchippy> rmatte: ohh
[25-Jun-2009 10:29:24] <klinstifen> bwest: yeah there are a couple forums postings about using the event transform to hide the password
[25-Jun-2009 10:29:29] <rmatte> I'm trying to amp up the performance on some scripts if possible
[25-Jun-2009 10:29:38] <mrchippy> rmatte: only if you know exactly where it lives
[25-Jun-2009 10:30:08] <rmatte> I see, hmm, doubt it's possible then
[25-Jun-2009 10:30:12] <venturaville> mrchippy: that worked for all of them thanks......
[25-Jun-2009 10:30:23] <zenChild> For the SSH modeling plugins, is it possible to get the handle of the ssh pipe to send additional commands, or am I limited to simply having "echo" delimiters for a single command.
[25-Jun-2009 10:30:24] <mrchippy> venturaville: sweet
[25-Jun-2009 10:30:25] <mrayzenoss> bwest: another problem is that the WMI ZenPacks that lots of people use aren't in the Community ZenPack repo, so nobody's updating them
[25-Jun-2009 10:30:37] <rmatte> I modified the WMI Performance zenpack a while back to retrieve the username and password values on it's own, but it's extremely intensive on the system when it's running
[25-Jun-2009 10:30:38] <bwest> rocket: good idea, any ideas about when you run the test in the template area?
[25-Jun-2009 10:31:04] <rocket> bwest: probably same solution to use a transform ..
[25-Jun-2009 10:31:55] <bwest> rocket: you can use a transform in templates when you are testing the command against a device?
[25-Jun-2009 10:32:13] <rocket> bwest: if it shows up in the event
[25-Jun-2009 10:32:58] <bwest> rocket: ok no I meant when i actually run the test in the template and it returns the result of the test the password shows up there in plain text
[25-Jun-2009 10:33:44] <rocket> bwest: probably the only solution is to update the zenpack in that case
[25-Jun-2009 10:33:59] <rocket> eg the zenpack is returning that information to the screen
[25-Jun-2009 10:34:00] <mrchippy> bwest: let me see if i'm getting this-- the prob is that testing the command shows the pword in plain txt?
[25-Jun-2009 10:34:06] <atrawog> @zenchild If you want you can send a complete shell script.
[25-Jun-2009 10:34:25] <bwest> mrchippy: yes
[25-Jun-2009 10:34:35] <rocket> mrchippy: correct .. there is a bug report filed that is closed as a wont fix because it technically is the zenpack maintainers issue
[25-Jun-2009 10:35:07] <bwest> mrchippy:so when zenoss admins are accessing this they can see admin passwords on devices they shouldn't
[25-Jun-2009 10:35:09] <zenChild> atrawog: I can't send shell scripts. It's an appliance running rbash. I'm extremely limited in what I can do at the shell.
[25-Jun-2009 10:35:18] <mrayzenoss> rocket: rmatte: bwest: and I've asked the maintainer to submit the ZenPack to the repository so it could get fixed, but he hasn't
[25-Jun-2009 10:35:45] <rmatte> mrayzenoss: yeh I know, I even submitted my changes to him as a rough start but haven't heard back from him
[25-Jun-2009 10:35:52] <rocket> mrayzenoss: should just take the zenpack and apply the maintainer mode work around and update it ourselves
[25-Jun-2009 10:36:02] <rmatte> there has to be a better way to pass the username and password info to the script
[25-Jun-2009 10:36:04] <mrchippy> bwest: actually kells is working on a mechanism that would do it or at least give you a hook that made it obvious where to do the obfuscating. hopefully, that'll be in the code soon
[25-Jun-2009 10:36:10] <mrayzenoss> rocket: well, it's his ZenPack, so we wouldn't be able to clear the licensing
[25-Jun-2009 10:36:16] <rmatte> if I can find a way I'll package it up and submit it myself
[25-Jun-2009 10:36:29] <rocket> mrayzenoss: was it gpl'd?
[25-Jun-2009 10:36:54] <rocket> mrayzenoss: my comment was based on the fact it was gpld of course
[25-Jun-2009 10:37:04] <rmatte> same here
[25-Jun-2009 10:37:06] <bwest> ok thx guys, that would be a major help
[25-Jun-2009 10:37:21] <venturaville> anyone know how I translate a list of device object to a list of objects in tales for a skin?
[25-Jun-2009 10:38:10] <rmatte> those zenpacks need to be updated badly, they are ancient
[25-Jun-2009 10:39:10] <bwest> The one I believe is Dalstrup and the other i used was from the community
[25-Jun-2009 10:39:30] <rocket> rmatte: how bad would it be to implement our own wmi zenpack?
[25-Jun-2009 10:39:48] <rocket> rmatte: I dont personally do much with windows so I havent looked at what they do etc
[25-Jun-2009 10:39:56] <rmatte> rocket: would be pretty easy from what I've seen, we should do that
[25-Jun-2009 10:40:09] <bwest> Yes please do that
[25-Jun-2009 10:40:31] <rocket> rmatte: there you go .. summer of code contest entry for ya
[25-Jun-2009 10:40:37] <mrayzenoss> So I checked, there's no license on his ZenPacks, which means we won't touch them
[25-Jun-2009 10:40:50] <rocket> mrayzenoss: makes sense to me
[25-Jun-2009 10:41:08] <venturaville> mrayzenoss: which license do you want on zenpacks anyways?
[25-Jun-2009 10:41:14] <rmatte> rocket: start with a straight performance Zenpack, then make a replacement for the Exchange and IIS Zenpacks
[25-Jun-2009 10:41:18] <zenChild> venturaville: have you looked this over? It came in handy when mucking with skins the other day. http://www.zenoss.com/community/docs/howtos/using-zentablemanager/
[25-Jun-2009 10:41:35] <mrayzenoss> GPLv2, with the Contributor Agreement signed, so we can post them and modify them as necessary
[25-Jun-2009 10:41:39] <mrchippy> venturaville: tal:define "objects here/getMyDeviceList"
[25-Jun-2009 10:41:52] <rmatte> rocket: still need to figure out how to pass the username and password to it in a more elegant fashion though
[25-Jun-2009 10:42:09] <mrchippy> venturaville: is that too obtuse? do you have a more specific thing in mind?
[25-Jun-2009 10:42:24] <mrayzenoss> more permissive licenses are OK than GPL, so you could go with BSD if you prefer
[25-Jun-2009 10:42:34] <rmatte> or AGPL
[25-Jun-2009 10:42:36] <rmatte>
[25-Jun-2009 10:42:36] <klinstifen> rmatte: i have an exchange zenpack i made...if you guys are interested i could go through the shabang and post it?
[25-Jun-2009 10:42:47] <mrayzenoss> yeah
[25-Jun-2009 10:42:51] <rocket> rmatte: just do like the command password is being done
[25-Jun-2009 10:42:51] <mrayzenoss> klinstifen: yeah
[25-Jun-2009 10:42:52] <bwest> Yes
[25-Jun-2009 10:43:03] <rocket> rmatte: in fact probably could leverage that code that exists
[25-Jun-2009 10:43:17] <rmatte> klinstifen: if it's nicer than the exchange zenpack that's there (which I'm sure it is) then please do submit it
[25-Jun-2009 10:43:29] <rmatte> klinstifen: you went through the effort of making it, people might as well use it hehe
[25-Jun-2009 10:43:32] <venturaville> mrchippy: I have a python list of references to actual devices, but it looks as though the objects that zentable manager is a little different ... the list I have is generated using a lambda and map combo from a list of device names
[25-Jun-2009 10:43:38] <klinstifen> rmatte: which one?
[25-Jun-2009 10:43:54] <rmatte> actually, sorry not that's there
[25-Jun-2009 10:44:02] <rmatte> but if it's nicer than Dastrup's
[25-Jun-2009 10:44:08] <rmatte> he sort of threw his together real quick
[25-Jun-2009 10:44:10] <rmatte> and it's outdated
[25-Jun-2009 10:44:13] <bwest> Is there an update in the works for the asterisk zenpack, I have seen mixed reviews for that one
[25-Jun-2009 10:44:28] <mrchippy> ahh, good ol' ZenTableManager. Let me find an example for you. You have to use getBatch()
[25-Jun-2009 10:44:33] <mrayzenoss> bwest: the original author said he's not using Asterisk anymore, so it's somewhat orphaned
[25-Jun-2009 10:44:36] <klinstifen> oh isn't dastrup's just in the forums?
[25-Jun-2009 10:44:43] <rmatte> asterisk is the worst VOIP system I have ever used in my life lol
[25-Jun-2009 10:45:02] <venturaville> mrchippy: yep, and getBatch() is a bit difficult to follow
[25-Jun-2009 10:45:03] <rocket> rmatte: are you using a free voip system?
[25-Jun-2009 10:45:04] <bwest> yes I agree but my company is using it and wants to monitor it
[25-Jun-2009 10:45:05] <mrayzenoss> klinstifen: yeah, and on his website. It's not hosted by Zenoss because he hasn't sent it to me
[25-Jun-2009 10:45:14] <rocket> rmatte: I was thinking of making somethign for home at one time ..
[25-Jun-2009 10:45:21] <klinstifen> ok...
[25-Jun-2009 10:45:24] <rmatte> rocket: we were at one point, we went back to the actual Cisco stuff
[25-Jun-2009 10:45:26] <klinstifen> well before i go packaging mine
[25-Jun-2009 10:45:32] <mrayzenoss> bwest: the Asterisk one is just wrapping a Nagios plugin of some sort, so you could probably do something similar
[25-Jun-2009 10:45:41] <bwest> ok
[25-Jun-2009 10:45:47] <klinstifen> rmatte: what do you think is outdated on his?
[25-Jun-2009 10:45:47] <mrchippy> venturaville: i disagree. it's impossible to follow.
[25-Jun-2009 10:45:58] <zenChild> venturaville: since you are familiar with the SVC concepts, here is a pt file that I am using to list out MDisk Groups... maybe it'll be of some help. TALES is pretty archaic though ... http://pastie.org/524340
[25-Jun-2009 10:46:06] <rmatte> klinstifen: the way it's packaged lol
[25-Jun-2009 10:46:22] <klinstifen> oh : )
[25-Jun-2009 10:46:23] <klinstifen> ok
[25-Jun-2009 10:46:25] <venturaville> I think I can write it just fine without getBatch, but I presume getBatch is used to lower memory usage
[25-Jun-2009 10:46:27] <klinstifen> i'll work on mine
[25-Jun-2009 10:46:31] <klinstifen> and try to get it posted
[25-Jun-2009 10:46:45] <rmatte> nice
[25-Jun-2009 10:46:57] <mrayzenoss> anyone else with a ZenPack, send them in. Just 7 more weeks in the contest
[25-Jun-2009 10:47:01] <klinstifen> i think i have told mray i would do this before too, but i didnt follow through ; )
[25-Jun-2009 10:47:03] <venturaville> objectValuesAll .... that is the crux of what I don't quite understand
[25-Jun-2009 10:47:12] <rmatte> I'll take a look at yours once you've posted, I can probably mod it in to a performance zenpack if it's good
[25-Jun-2009 10:47:15] <bwest> rmatte: Going back just a bit is it ossible for me to deny users access to certain pages in zenoss?
[25-Jun-2009 10:47:18] <klinstifen> mray: got a link for this contest?
[25-Jun-2009 10:47:33] <venturaville> zenChild that link hels a bit
[25-Jun-2009 10:47:33] <mrayzenoss> klinstifen: http://blog.zenoss.com/2009/05/17/summer-of-community-zenpacks-contest/
[25-Jun-2009 10:47:49] <klinstifen> rmatte: errr...whats the diff between a zenpack and a performance zenpack?
[25-Jun-2009 10:47:54] <rocket> bwest: easy answer is no for core .. but there are hacks to allow it .. I believe its supported in enterprise ...
[25-Jun-2009 10:48:06] <mrayzenoss> bwest: it's an Enterprise feature, but you can get into the Zope permissions and set it up
[25-Jun-2009 10:48:08] <klinstifen> mray: thanks
[25-Jun-2009 10:48:12] <zenChild> venturaville: that (objectValuesAll) is the python function that is being called.
[25-Jun-2009 10:48:13] <rmatte> bwest: if you go in the zope management interface, create a role, you can assign it particular permissions and then assign it to users, but I don't think it allows you to control access to particular pages
[25-Jun-2009 10:48:53] <bwest> rmatte: But I can limit a user from say using the templates area
[25-Jun-2009 10:49:01] <klinstifen> rmatte: isn't a performance zenpack just a zenpack with performance graphs?
[25-Jun-2009 10:49:01] <venturaville> zenChild: yep, but how is that different from the actual device....
[25-Jun-2009 10:49:02] <mrayzenoss> bwest: look at http://www.skills-1st.co.uk/papers/jcurry.html "Crafting Zenoss Core users for events and zProperties" for some ideas
[25-Jun-2009 10:49:03] <rmatte> plus I've only used roles with LDAP, so I'm not sure how you assign them to regular users, or if you even can
[25-Jun-2009 10:49:31] <bwest> ok
[25-Jun-2009 10:49:36] <rmatte> klinstifen: correct, but it'd be performance graphs for CPU, Memory, paging, etc...
[25-Jun-2009 10:49:36] <mrchippy> venturaville: objectValuesAll is a method on a relation that gives back all the objects at the other end. for example, deviceClass.devices.objectValuesAll(), but deviceClass.devices() does the same thing. can you post a snippet here or on pastebin?
[25-Jun-2009 10:50:03] <rmatte> klinstifen: It'd take someone very knowledgeable to actually write a full blown WMI ZenPack that matches up to the enterprise one
[25-Jun-2009 10:50:23] <bwest> Another quick question, is there a way to setup administrators of services, just like the other classes?
[25-Jun-2009 10:50:26] <venturaville> mrchippy: here is what I am trying to do (it is very short):
[25-Jun-2009 10:50:32] <venturaville> tal:define="tableName string:vipsList;
[25-Jun-2009 10:50:32] <venturaville> vipdevlist python:map(lambda x: devices.findDevice(x), here.zVIPNameList)
[25-Jun-2009 10:50:32] <venturaville> tabletitle string:VIPs Serving this Site;
[25-Jun-2009 10:50:32] <venturaville> batch python:here.ZenTableManager.getBatch(tableName,vipdevlist)">
[25-Jun-2009 10:50:39] <rmatte> and I think the sales folks over at Zenoss would cringe if they saw something like that dropped on their plate hehe
[25-Jun-2009 10:51:05] <klinstifen> rmatte: i haven't seen the enterprise one so i dont have anything to compare against
[25-Jun-2009 10:51:25] <rmatte> It basically does everything that's capable via SNMP in core, but does it via WMI
[25-Jun-2009 10:51:34] <klinstifen> ah
[25-Jun-2009 10:51:36] <venturaville> where zVIPNameList = a z property of type lines with a list of IP addresses (which correspond to device names)
[25-Jun-2009 10:51:36] <rmatte> interfaces, hardware, software, the works
[25-Jun-2009 10:51:54] <mrayzenoss> Our position is if someone writes a ZenPack and fills out the Contributor form, we'll post it
[25-Jun-2009 10:52:15] <venturaville> I'm writing a zenpack to allow for monitoring of sites/vips/hosts and have linkages between all of them.
[25-Jun-2009 10:52:19] <rmatte> that's cool, but they'd still cringe hehe
[25-Jun-2009 10:52:32] <mrayzenoss> rmatte: the new Predictive Threshold ZenPack is an example
[25-Jun-2009 10:52:43] <rmatte> yeh, that is pretty cool
[25-Jun-2009 10:52:45] <klinstifen> rmatte: hmm gotcha. nah mine is modeled off the one in the cacti forums
[25-Jun-2009 10:53:00] <klinstifen> mailq, db io, rpc requests, etc
[25-Jun-2009 10:53:08] <mrchippy> venturaville: looking at getBatch
[25-Jun-2009 10:53:12] <rmatte> klinstifen: that's all you really need though, so that's cool
[25-Jun-2009 10:53:21] <rmatte> klinstifen: if you find that it works well then it's worth contributing
[25-Jun-2009 10:53:36] <mrayzenoss> 2.4.2 was just released!
[25-Jun-2009 10:53:43] <rmatte> w00t w00t
[25-Jun-2009 10:53:46] <venturaville> got a URL for update lists?
[25-Jun-2009 10:53:49] <bwest> when
[25-Jun-2009 10:53:58] <bwest> what's new on that release
[25-Jun-2009 10:54:00] <mrayzenoss> like 2 minutes ago
[25-Jun-2009 10:54:10] <mrayzenoss> http://downloads.sourceforge.net/zenoss/Release_Notes_Core_2.4.2.pdf
[25-Jun-2009 10:54:12] <klinstifen> rmatte: mray: cool well i will try and get it packaged in time for the contest
[25-Jun-2009 10:54:23] <rmatte> "2.4.2 was JUST released!" ... "when?"
[25-Jun-2009 10:54:24] <rmatte> lol
[25-Jun-2009 10:54:27] <rmatte> too funny
[25-Jun-2009 10:54:37] <bwest> lol
[25-Jun-2009 10:54:48] <bwest> i get to excited like a kid in a candy store
[25-Jun-2009 10:55:01] <rmatte> apparently
[25-Jun-2009 10:55:03] <mrayzenoss> maintenance release, 46 tickets closed
[25-Jun-2009 10:55:16] <bwest> rmatte: Is there a way to set administrators for services like the other classes
[25-Jun-2009 10:55:35] <paulh> no way to assign admins for services
[25-Jun-2009 10:56:52] <bwest> I assume it would take some serious coding
[25-Jun-2009 10:57:04] <venturaville> 5035 was a big one
[25-Jun-2009 10:57:10] <rmatte> does 2.4.2 include rocket's SSH monitoring patch?
[25-Jun-2009 10:57:50] <mrayzenoss> rmatte: no, it wasn't backported
[25-Jun-2009 10:58:07] <mrayzenoss> and he put another patch on after I closed it, so I gotta reopen it
[25-Jun-2009 10:58:22] <mrchippy> venturaville: the object list has to be list of devices (or components, probably). also, the lambda won't work. you need to pass the object list and muck with them (get the zproperty) afterward. do you want to list all devices?
[25-Jun-2009 10:59:18] <bwest> ok thx all for all your help!!
[25-Jun-2009 10:59:31] <rmatte> yeh, I noticed that
[25-Jun-2009 10:59:34] <rmatte> k
[25-Jun-2009 11:00:05] <mrayzenoss> ran out of battery
[25-Jun-2009 11:00:42] <rmatte> hehe
[25-Jun-2009 11:01:52] <mrchippy> venturaville: hold up, i completely misread your snippet
[25-Jun-2009 11:02:25] <zenChild> rocket: you around
[25-Jun-2009 11:02:31] <rocket> yes
[25-Jun-2009 11:03:08] <rmatte> I can't wait until 4799 is fixed
[25-Jun-2009 11:03:18] <rmatte> I have a workaround for it for now
[25-Jun-2009 11:04:16] <mrchippy> venturaville: that should work as long as here.zVIPNameList actually returns as a list and the names are exact matches.
[25-Jun-2009 11:04:27] <venturaville> they are
[25-Jun-2009 11:04:30] <zenChild> rocket: in your aix snmp zenpack. there is the assignment "vgom.getLvSetup = MultiArgs(lv_hash)" in AixVolumeGroupMap.py. How does that work in relation to the method in AixVolumeGroup.py? When does the method get called?
[25-Jun-2009 11:05:05] <venturaville> mrchippy: since I don't have a way of directly linking devices, this is kind of a work around to that ... I am populating the list either manually (in some cases) or collected from load balancers otherwise
[25-Jun-2009 11:05:17] <Appiah_> How do I remove my mibs from Zenoss?
[25-Jun-2009 11:06:35] <rocket> zenChild: getLvSetup is called from AixVolumeGroup.py with the contents of lv_hash as its arguements
[25-Jun-2009 11:06:50] <mrchippy> venturaville: after that, you can do the whole <tr tal:repeat="device batch"> and in the row refer to the "device" object
[25-Jun-2009 11:06:57] <mrayzenoss> Appiah_: go to the MIBs page and select the ones you want to delete. From the page tab select "Delete Mibs..."
[25-Jun-2009 11:06:59] <rocket> zenChild: the modeller code is passing all that into the model via an object map
[25-Jun-2009 11:07:07] <Appiah_> mrayzenoss: i see nothing like that
[25-Jun-2009 11:07:08] <mrchippy> venturaville: is this what you have already?
[25-Jun-2009 11:07:23] <mrayzenoss> Appiah_: which part?
[25-Jun-2009 11:07:50] <Appiah_> there is no arrow drop menu (or whatever you call it) under /Mibs
[25-Jun-2009 11:07:54] <rocket> zenChild: lv_hash will contain getPvSetup for example that will call the special method from the logical_volume code to setup a new pv
[25-Jun-2009 11:07:58] <Appiah_> there is on under sub-folders but that's it
[25-Jun-2009 11:08:50] <Appiah_> this is weird , another zenoss server that is running got that drop menu ... same version
[25-Jun-2009 11:08:55] <rocket> zenChild: if you look at AixVolumeGroup.py AixLogicalVolume.py as examples you will see these specialized methods
[25-Jun-2009 11:09:22] <rmatte> Appiah_: try restarting zeoctl?
[25-Jun-2009 11:09:23] <zenChild> rocket: looking now.
[25-Jun-2009 11:09:32] <Appiah_> rmatte: i'll try that
[25-Jun-2009 11:09:36] <mrayzenoss> Appiah_: are you an unprivileged user?
[25-Jun-2009 11:09:49] <rocket> zenChild: but ultimately nothing happens in the modeller other than a big object map object is created
[25-Jun-2009 11:09:53] <rmatte> unprivileged users shouldn't even see the Mibs section listed
[25-Jun-2009 11:09:58] <Appiah_> mrayzenoss: oh
[25-Jun-2009 11:10:03] <rocket> zenChild: all the work is done inside the objects themselves
[25-Jun-2009 11:10:08] <Appiah_> then his user is configured wrongly
[25-Jun-2009 11:10:11] <Appiah_> this*
[25-Jun-2009 11:10:24] <zenChild> rocket: how does the method actually get called though?
[25-Jun-2009 11:10:35] <rocket> zenChild: its a special name
[25-Jun-2009 11:10:51] <rmatte> Appiah_: check with the admin account, if it works with that then that's your problem
[25-Jun-2009 11:10:57] <klinstifen> mray: are you guys offering any developer's conferences?
[25-Jun-2009 11:11:15] <rocket> eg if the object map name getPvName exists in the object it gets called
[25-Jun-2009 11:11:15] <venturaville> mrchippy: yeah the rest of the skin is basically the same as one I copied from elsewhere
[25-Jun-2009 11:11:23] <klinstifen> mray: advanced zenpack creation...zenoss and python...stuff like that?
[25-Jun-2009 11:11:35] <rocket> zenChild: and the stuff it gets set to in the objectmap is just the arguement list for it
[25-Jun-2009 11:11:38] <rmatte> if you guys eventually do zenoss development training courses I'm going to have my boss send me to one
[25-Jun-2009 11:11:42] <zenChild> rocket: That clears it up for me.
[25-Jun-2009 11:11:54] <mrayzenoss> klinstifen: well, we'll have a community day at LISA and we'll be at Ohio LinuxFest
[25-Jun-2009 11:12:00] <rocket> zenChild: great
[25-Jun-2009 11:12:07] <zenChild> rocket: I was looking for an actual method call and was a bit perplexed as to how things were happening.
[25-Jun-2009 11:12:12] <klinstifen> hmm....anything in the nyc area?
[25-Jun-2009 11:12:18] <venturaville> mrayzenoss: you know Bill Lefebvre at LISA?
[25-Jun-2009 11:12:28] <klinstifen> i need a zenoss meetup or zenossnycgroup or something
[25-Jun-2009 11:12:30] <mrayzenoss> venturaville: nope
[25-Jun-2009 11:12:44] <venturaville> mrayzenoss: just curious
[25-Jun-2009 11:12:45] <rocket> zenChild: the modeller code does a fancy trick to see if its a method or a parameter type item
[25-Jun-2009 11:12:45] <mrayzenoss> klinstifen: we do have a sales engineer in NYC
[25-Jun-2009 11:12:59] <mrayzenoss> I can look into doing a dev conference
[25-Jun-2009 11:13:08] <Appiah_> rmatte , mrayzenoss same on the Admin account , should i try restaring zeoctl ?
[25-Jun-2009 11:13:17] <mrayzenoss> Appiah_: can't hurt
[25-Jun-2009 11:13:29] <zenChild> rocket: good to know. Know I can keep moving on my project. Thanks again.
[25-Jun-2009 11:13:32] <rmatte> klinstifen: this IRC channel is about the best meetup you're going to get from what I've seen hehe
[25-Jun-2009 11:13:45] <klinstifen> mray: neat
[25-Jun-2009 11:13:51] <klinstifen> rmatte: yeah : \
[25-Jun-2009 11:13:55] <klinstifen> not fun on quiet days
[25-Jun-2009 11:14:05] <Appiah_> restarted , still the same
[25-Jun-2009 11:14:22] <klinstifen> but actually its been pretty active the past couple times i have hopped on
[25-Jun-2009 11:14:22] <rmatte> once Zenoss really takes off (which I'm sure it will) there might be something like that
[25-Jun-2009 11:14:25] <mrchippy> venturaville: if you follow all the steps in your tal:define in zendmd, does batch have things in it? that is does dmd.ZenTableManager.getBatch(tableName,vipdevlist) return empty?
[25-Jun-2009 11:15:11] <Appiah_> so what do you do when a drop down menu is missing?
[25-Jun-2009 11:15:13] <mrayzenoss> yeah, we started to plan a Zenoss User Conference, but had trouble nailing down a date
[25-Jun-2009 11:15:27] <klinstifen> rmatte: i'ven been on this boat since v.1.something. i am sold. at the time i was comparing against hyperic, but the agentlessness won me over
[25-Jun-2009 11:15:50] <mrayzenoss> Appiah_: have you logged out and logged back in, maybe something weird cached
[25-Jun-2009 11:15:51] <zenChild> mrayzenoss: what about a BOF at OSCON or something similar?
[25-Jun-2009 11:15:52] <rmatte> yeh, agentless is golden
[25-Jun-2009 11:16:18] <rmatte> and I like how easy it is to write scripts and integrate them in for data collection
[25-Jun-2009 11:16:19] <mrayzenoss> zenChild: I'll be at the Community Leadership Summit the weekend before OSCON, but I'm heading home after that
[25-Jun-2009 11:16:25] <venturaville> I can't run getBatch inside of DMD without a session...
[25-Jun-2009 11:16:27] <klinstifen> appiah: flush your cache? there've been forum postings to fix other dashboard quirkiness
[25-Jun-2009 11:16:31] <venturaville> File "/opt/zenoss/zenoss/Products/ZenWidgets/ZenTableManager.py", line 295, in getTableStates
[25-Jun-2009 11:16:31] <venturaville> session = self.REQUEST.SESSION
[25-Jun-2009 11:16:43] <Appiah_> mrayzenoss: tried both firefox and IE ( I never used it with IE before) same problem
[25-Jun-2009 11:16:44] <venturaville> File "/opt/zenoss/zenoss/lib/python/ZPublisher/HTTPRequest.py", line 1219, in __getattr__
[25-Jun-2009 11:16:44] <venturaville> raise AttributeError, key
[25-Jun-2009 11:16:44] <venturaville> AttributeError: SESSION
[25-Jun-2009 11:16:57] <mrchippy> venturaville: well, that sucks
[25-Jun-2009 11:17:08] <venturaville> yep ... makes writing templates hard if using table manager
[25-Jun-2009 11:17:09] <mrayzenoss> zenChild: OSCON is really expensive for vendors, so we dropped out
[25-Jun-2009 11:17:17] <rmatte> I'm going to be getting a company Zenoss blog up and running some time next month, I'll be posting a bunch of scripts and tricks that I've found on there
[25-Jun-2009 11:17:26] <mrayzenoss> rmatte: excellent
[25-Jun-2009 11:17:33] <klinstifen> rmatte: yeah i dig the scripting aspect. if i can think of it then it can probably be done in zenoss (or already has been ; )
[25-Jun-2009 11:17:53] <klinstifen> rmatte: pm me in the forums once that is up...i'd like to check it out
[25-Jun-2009 11:17:55] <rmatte> klinstifen: yeh, I've done some really neat stuff with scripts
[25-Jun-2009 11:18:05] <rmatte> sure thing
[25-Jun-2009 11:18:09] <mrayzenoss> I did submit Chet's name for a Guru session at LISA entitled "Try to Stump Chet, Monitor Anything With Zenoss"
[25-Jun-2009 11:18:22] <venturaville> mrchippy: I'll play with this a bit and see how far I get ... I'm booked for meetings the rest of today
[25-Jun-2009 11:18:27] <venturaville> thanks for the help
[25-Jun-2009 11:18:39] <mrchippy> venturaville: np. sorry we couldn't bang it out.
[25-Jun-2009 11:18:42] <rmatte> mrayzenoss: does he have access to Enterprise or Core for that?
[25-Jun-2009 11:19:07] <mrayzenoss> rmatte: well, it's a Q&A session so he'll just tell you how you could do it
[25-Jun-2009 11:19:14] <rmatte> ah
[25-Jun-2009 11:19:17] <mrayzenoss> so in theory, it's either
[25-Jun-2009 11:19:21] <rmatte> I'd show up with a list hehe
[25-Jun-2009 11:19:33] <rocket> how would he monitor sap
[25-Jun-2009 11:19:38] <rocket>
[25-Jun-2009 11:19:43] <rmatte> rocket: nice one
[25-Jun-2009 11:19:44] <rocket> muaaa haaa haaa
[25-Jun-2009 11:20:22] <mrayzenoss> rocket: ports, processes, web pages, the actual machines, databases
[25-Jun-2009 11:20:31] <rmatte> I'd just ask how he'd monitor AIX, since that's a bit shaky from what I understand?
[25-Jun-2009 11:20:44] <rocket> rmatte: dont get me going ... ;p
[25-Jun-2009 11:20:56] <Appiah_> i*m running 2.4.1 and these is no menu for the mibs on this server ( got other servers that actully has the dropdown menu) , is there no command line for removing?
[25-Jun-2009 11:20:56] <rocket> rmatte: did you look at my zenpack for aix?
[25-Jun-2009 11:21:03] <klinstifen> rmatte: i monitored my aix box for a while through a ssh script i wrote
[25-Jun-2009 11:21:12] <rmatte> I need to monitor some AIX boxes eventually, going to be fun, and no, I haven't taken a look at the pack yet
[25-Jun-2009 11:21:14] <klinstifen> i piped everything into zenoss from a command datasource
[25-Jun-2009 11:21:28] <klinstifen> didn't want to use snmp because we have heard bad things about snmp and aix...
[25-Jun-2009 11:21:31] <rocket> rmatte: take a look at the screenshots for it
[25-Jun-2009 11:21:41] <rocket> klinstifen: there is bad stuff for snmp ..
[25-Jun-2009 11:21:41] <atrawog> rmatte:The new AIX ZenPack is great.
[25-Jun-2009 11:21:51] <rocket> atrawog: you using it?
[25-Jun-2009 11:21:52] <rmatte> klinstifen: there is a net-snmp port for AIX though I think which is way better than the stock SNMP
[25-Jun-2009 11:22:11] <rocket> atrawog: like my improvements?
[25-Jun-2009 11:22:21] <klinstifen> rmatte: cool
[25-Jun-2009 11:22:28] <klinstifen> we'll have to check it out
[25-Jun-2009 11:22:36] <klinstifen> for the time being we just monitor that box seperatly
[25-Jun-2009 11:22:41] <atrawog> rocket: I have put on my Zenoss test server two days ago.
[25-Jun-2009 11:22:42] <rmatte> rocket: checking screenshot...
[25-Jun-2009 11:23:02] <mrayzenoss> Appiah_: hmmm... don't see a command line option for removing with zenmib
[25-Jun-2009 11:23:14] <mrayzenoss> Appiah_: still not sure why you can't get to the delete
[25-Jun-2009 11:23:17] <Appiah_> I've got 2 different servers that's missing drop down menus under mibs (wth)
[25-Jun-2009 11:23:21] <rmatte> rocket: AIX SNMP?
[25-Jun-2009 11:23:27] <rocket> rmatte: yes
[25-Jun-2009 11:24:07] <rocket> rmatte: I might work on an ssh version but it depends on time and an agreement between zenoss and myself I am working towards
[25-Jun-2009 11:24:27] <rmatte> hmmmm, does it do interfaces too?
[25-Jun-2009 11:24:31] <mrayzenoss> Appiah_: this is missing right? http://imagebin.org/53650
[25-Jun-2009 11:24:36] <rocket> rmatte: yes
[25-Jun-2009 11:24:50] <rmatte> rocket: so you know how to create a modeler plugin eh?
[25-Jun-2009 11:25:04] <rocket> of course
[25-Jun-2009 11:25:20] <atrawog> rocket: Modeling VGs + LVs is wonderfull. But I have to test how to migrate to the new version.
[25-Jun-2009 11:25:22] <Appiah_> exactly mrayzenoss
[25-Jun-2009 11:25:27] <rmatte> rocket: couldn't you just take the AIX ZenPack code and modify it for WMI?
[25-Jun-2009 11:25:29] <klinstifen> rocket: you teach yourself that or did you pick it up from one of the guides?
[25-Jun-2009 11:25:31] <Appiah_> no white arrow there, nothing to click
[25-Jun-2009 11:25:51] <mrayzenoss> Appiah_: and you're logged in as the Admin?
[25-Jun-2009 11:25:58] <Appiah_> yupp
[25-Jun-2009 11:26:07] <Appiah_> even tried another browser
[25-Jun-2009 11:26:10] <rocket> atrawog: that might be an issue .. but what I usually did was delete the old device .. keeping alerts and perf stuff .. and then adding the device back in
[25-Jun-2009 11:26:30] <mrayzenoss> Appiah_: put up a screenshot on imagebin and I'll pass it around for ideas
[25-Jun-2009 11:26:38] <Appiah_> k , just a sec
[25-Jun-2009 11:26:56] <rocket> rmatte: I am sure I could .. like I said I just dont have much time lately on windows boxes
[25-Jun-2009 11:27:19] <rocket> klinstifen: guides, self taught, a few questions here and there of chet
[25-Jun-2009 11:27:24] <rmatte> rocket: fly over here and you'll have all the time you need
[25-Jun-2009 11:27:36] <klinstifen> rocket: cool stuff
[25-Jun-2009 11:27:43] <rocket> klinstifen: actually I spent time putting debug statements in my version of zenoss to figure out how the main code worked
[25-Jun-2009 11:27:59] <klinstifen> woah intensive
[25-Jun-2009 11:28:40] <rocket> klinstifen: my zenpack is the first one to do multidepth objects .. eg the volume group -> logical volume relationship etc
[25-Jun-2009 11:28:54] <rocket> logical volume -> filesystem
[25-Jun-2009 11:29:16] <rocket> rmatte: my currently company would kill me
[25-Jun-2009 11:29:31] <rmatte> rocket: hehe
[25-Jun-2009 11:30:18] <atrawog> rocket: I have done that in the past and it did work well. But in the case of our AIX servers I have modified the filesystem templates to show absolute filesystem sizes instead of percentage usage and see how that works with the new version.
[25-Jun-2009 11:30:24] <rmatte> well, maybe I'll take a look at the bones of the AIX ZenPack and have a go at developing it myself at some point
[25-Jun-2009 11:30:40] <rmatte> though my python coding skills are still pretty bad
[25-Jun-2009 11:30:47] <Appiah_> http://imagebin.org/53653 mrayzenoss
[25-Jun-2009 11:31:02] <rmatte> if I could script it in bash it'd be done in a day
[25-Jun-2009 11:31:25] <klinstifen> rmatte: hahah
[25-Jun-2009 11:31:27] <mrayzenoss> Appiah_: bizarre
[25-Jun-2009 11:31:53] <rmatte> klinstifen: yeh, I'm a genius when it comes to bash scripting
[25-Jun-2009 11:32:11] <Appiah_> mrayzenoss: I bet it is not a common error
[25-Jun-2009 11:32:37] <Appiah_> but I have this on two different servers , i tried flush cache and 2 different browsers logged in as admin
[25-Jun-2009 11:32:49] <rmatte> man, this AIX zenpack is awesome, makes the AIX monitoring better than the Core Windows monitoring lol
[25-Jun-2009 11:33:02] <mrayzenoss> Appiah_: both servers can't delete MIBs?
[25-Jun-2009 11:33:19] <Appiah_> both servers dont have that drop down menu
[25-Jun-2009 11:33:31] <rocket> rmatte:
... cant have windows beat up on aix too much can we? [25-Jun-2009 11:33:34] <mrayzenoss> Appiah_: is that the Cisco MIBs ZenPack?
[25-Jun-2009 11:34:00] <Appiah_> nope , no zenpacks installed
[25-Jun-2009 11:34:15] <Appiah_> I was going to remove these mibs and add the zenpack with cisco mib actully
[25-Jun-2009 11:36:03] <rocket> atrawog: is the datapoint data for filesystems the same though?
[25-Jun-2009 11:36:17] <rocket> atrawog: eg are you just displaying it differently on the graph?
[25-Jun-2009 11:38:42] <mrayzenoss> Appiah_: I assume if you restart everything you get the same results?
[25-Jun-2009 11:38:56] <klinstifen> alright all...gotta run...thanks for the help
[25-Jun-2009 11:39:01] <atrawog> rocket: I did a couple of changes in the old version.
[25-Jun-2009 11:39:23] <zenChild> I also have to run. Thanks all for your help. Cheers!
[25-Jun-2009 11:39:51] <mrayzenoss> thanks to everyone for showing up, Dev session is over but per usual, I'm here frequently
[25-Jun-2009 11:40:16] * mrayzenoss just noticed mrchippy had already left
[25-Jun-2009 11:40:28] <atrawog> rocket: I'm fetching totalBlocks used via SNMP or SSH
[25-Jun-2009 11:41:04] <rocket> atrawog: ok ..
[25-Jun-2009 11:41:17] <atrawog> atrawog: I have modified the perfs template to show totalBocksUsed
[25-Jun-2009 11:44:16] <Appiah_> mrayzenoss: logged out , restarted , logged in , same
[25-Jun-2009 11:44:19] <rmatte> rocket: that predictive threshold ZenPack is awesome too, just taking a look at it
[25-Jun-2009 11:44:29] <atrawog> rocket: And I'm running a zendmd script via cron that syncs the totalBocks values from the RRD files with ZODB.
[25-Jun-2009 11:44:51] <rocket> atrawog: why are you doing that zendmd script?
[25-Jun-2009 11:45:09] <rmatte> rocket: still no Ubuntu ZenPack?
[25-Jun-2009 11:45:35] <rocket> rmatte: I have an ubuntu virtual machine installed
[25-Jun-2009 11:45:40] <mrayzenoss> Appiah_: FYI, don't try deleting it from Zope's /manage URL, it screws up royally
[25-Jun-2009 11:45:40] <rocket> rmatte: but no zenpack yet
[25-Jun-2009 11:45:49] <rmatte> rocket: cool
[25-Jun-2009 11:46:18] <rocket> rmatte: currently working on an opensolaris one at the moment
[25-Jun-2009 11:46:27] <rmatte> nice
[25-Jun-2009 11:47:00] <rocket> rmatte: the predictive threshold gave the same results as the enterprise one according to matt
[25-Jun-2009 11:47:11] <rmatte> nice
[25-Jun-2009 11:47:18] <rocket> rmatte: however mine lets you apply the threshold to existing rrd's
[25-Jun-2009 11:47:23] <rmatte> as long as it's coded right it's all math, so it's not surprising
[25-Jun-2009 11:47:42] <rmatte> so your's is even better than the enterprise one hehe
[25-Jun-2009 11:48:00] <rocket>
[25-Jun-2009 11:48:29] <rocket> it also has a script to remove the data from rrd's that have had the predictive stuff turned on
[25-Jun-2009 11:48:35] <rocket> if necessary
[25-Jun-2009 11:48:43] <rmatte> I still need to figure out how to make some sort of install package for those Cisco Mibs so that they install much the same way as a Zenpack, but don't actually get registered as one
[25-Jun-2009 11:48:52] <atrawog> rocket: Because Threashold are based on the TotalBlock value stored in ZODB and my AIX admins complained that the threasholds still trigger after the have resized a filesystem.
[25-Jun-2009 11:48:57] <mrayzenoss> rmatte: using zenmib?
[25-Jun-2009 11:49:14] <mrayzenoss> atrawog: that was yesterday's Tip of the Month
[25-Jun-2009 11:49:16] <rmatte> mrayzenoss: can't, some people followed my guide and weren't able to get them all to load
[25-Jun-2009 11:49:25] <rmatte> only reliable way is to import them in from the objects.xl
[25-Jun-2009 11:49:27] <rmatte> xml*
[25-Jun-2009 11:51:56] <Appiah_> mrayzenoss: delete what? the mibs
[25-Jun-2009 11:52:46] <mrayzenoss> yeah, don't do that
[25-Jun-2009 11:53:23] <atrawog> rocket: Remodelling the device does fix the problem. But its hard to argue why the RRD graphs show the currect values and people still get thresholds warnings.
[25-Jun-2009 11:53:23] <Appiah_> but I cant!
[25-Jun-2009 11:53:25] <rocket> atrawog: ah I see .. but why are you using the value stored in the Zodb?
[25-Jun-2009 11:53:44] <mrayzenoss> gotta run, off to lunch
[25-Jun-2009 11:53:45] <rocket> atrawog: why arent you using the threshold based on the datapoint?
[25-Jun-2009 11:54:04] <Appiah_> then how am I supposed to remove the mibs?
[25-Jun-2009 11:55:00] <rocket> atrawog: actually I probably can code and easy fix for you in my zenpack
[25-Jun-2009 11:55:02] <rmatte> Appiah_: sounds like you have a botched install
[25-Jun-2009 11:55:10] <Appiah_> O_o
[25-Jun-2009 11:55:31] <rocket> atrawog: then you wont need to populate with an zendmd script
[25-Jun-2009 11:55:51] <atrawog> rocket: As far as a know you can only compare a RRD value with a ZODB value in a threshold and not two RRD values which are stored in two different RRD files.
[25-Jun-2009 11:56:04] <rocket> atrawog: there is a better way ..
[25-Jun-2009 11:56:41] <rocket> atrawog: extend the filesystem object class to have a method that returns the totalblock data out of the rrd
[25-Jun-2009 11:56:59] <rocket> then call here/get_totalblock_rrd method
[25-Jun-2009 11:57:47] <venturaville> I believe there is also a cache equivalent of that if you want to reduce load a bit, but I am not sure how long it keeps it
[25-Jun-2009 11:59:25] <venturaville> CACHE_TIME = 60.
[25-Jun-2009 11:59:28] <venturaville> apparently 60
[25-Jun-2009 11:59:34] <rmatte> hehe
[25-Jun-2009 12:00:20] <rocket> venturaville: yeah ...
[25-Jun-2009 12:01:06] <atrawog> rocket: I was pretty sure that there is a better way to do it. But till now I was too scared of writing my own object classes.
[25-Jun-2009 12:02:44] <atrawog> rocket: If you want to have a look you can find my AIX ZenPack version at http://code.google.com/p/atrawog-zenoss-community/source/browse/#hg/zenpacks/ZenPacks.atrawog.AIX
[25-Jun-2009 12:06:19] <rmatte> I guess I will try making a script to use zenmib to install the Cisco Mibs, I'll send it out to some of you guys for testing once it's done.
[25-Jun-2009 12:06:31] <rmatte> It'll probably install the Mibs way faster than the ZenPack anyways
[25-Jun-2009 12:07:06] <Appiah_> what happens if i add a zenpack with mibs, that includes some mibs that i already have?
[25-Jun-2009 12:07:20] <rmatte> it just won't install those Mibs
[25-Jun-2009 12:07:43] <Appiah_> so no risk for conflicts or such
[25-Jun-2009 12:07:54] <rmatte> correct
[25-Jun-2009 12:08:07] <rmatte> even if it did install duplicates it wouldn't cause issues anyways
[25-Jun-2009 12:42:01] <blocky> if my blade is locking up repeatedly, how can I enable some logging to see whats going on
[25-Jun-2009 13:01:09] <shood> Greetings. I would like to have my zenoss enterprise installation monitor permissions on a file tree, and alert on any modifications. Does this functionality exist.. or could someone point me in the direction
[25-Jun-2009 13:02:23] <mrayzenoss> you'd probably need to add a custom datasource
[25-Jun-2009 13:02:56] <mrayzenoss> perhaps a simple script that checks that the permissions for the file are what you want
[25-Jun-2009 13:05:06] * shood reads
[25-Jun-2009 13:33:56] <venturaville> can't get a menu to show up on a device in a zenpack
[25-Jun-2009 13:34:07] <venturaville> anyone have the link for the howto for that
[25-Jun-2009 13:34:11] <venturaville> something I am missing..........
[25-Jun-2009 13:44:14] <rocket> venturaville: menu examples are in the aix zenpack
[25-Jun-2009 13:44:27] <rocket> eg for print queues etc
[25-Jun-2009 13:44:47] <rocket> venturaville: check out the items in __init__.py and the new pt files
[25-Jun-2009 13:45:06] <venturaville> rocket: looks like it can't find my class :-/
[25-Jun-2009 13:45:43] <rocket> venturaville: also menus are only installed on zenpack install unless you create the menu from the gui
[25-Jun-2009 13:46:10] <rocket> venturaville: so I had to keep reinstalling my zenpack ontop of itself etc ..
[25-Jun-2009 13:46:53] <venturaville> the menus are in the more section (like the vmware one)
[25-Jun-2009 13:47:11] <venturaville> zPythonClass is set right
[25-Jun-2009 13:47:23] <venturaville> I think I'm just missing something obvious
[25-Jun-2009 13:47:35] <rocket> venturaville: did you check that the menu shows up in the gui under settings menu?
[25-Jun-2009 13:47:43] <venturaville> yep
[25-Jun-2009 13:48:00] <venturaville> when I add a new device it can't find my class though *shrug*
[25-Jun-2009 13:52:51] <venturaville> 2009-06-25T14:37:39 ERROR zen.DeviceClass Unable to import class ZenPacks.Turner.HostMap.Site
[25-Jun-2009 13:53:03] <venturaville> ImportError: No module named Site
[25-Jun-2009 13:54:06] <venturaville> I've got a zenpack if you want to see
[25-Jun-2009 13:59:51] <etank> im having an issue with setting up alerting rules
[25-Jun-2009 14:00:07] <etank> i have one rule that should show all errors with a delay of 120
[25-Jun-2009 14:00:35] <etank> and one that i want to show Perf CPU issues but only if the issue has been going on for 900 seconds
[25-Jun-2009 14:01:29] <etank> it seems that the rule for the Perf CPU is not working because i get the emails even when the CPU has been over the 90% threshold for 2 minutes or so
[25-Jun-2009 14:02:28] <etank> i may have just found the issue myself
[25-Jun-2009 14:03:48] <venturaville> what is the 'accepted' license for Zenpacks?
[25-Jun-2009 14:04:29] <mrayzenoss> venturaville: well, if someone submits it to me for posting on the Zenoss site it needs to have something GPLv2 compatible
[25-Jun-2009 14:04:58] <mrayzenoss> so the option of repackaging it exists
[25-Jun-2009 14:05:05] <venturaville> ventura@treebeard:~/svn/Zenoss/hostmap/trunk/ZenPacks/Turner/HostMap$ grep -i license ../../../EGG-INFO/PKG-INFO
[25-Jun-2009 14:05:05] <venturaville> License: GPLv2
[25-Jun-2009 14:05:06] <venturaville> like so?
[25-Jun-2009 14:05:32] <mrayzenoss> yeah, that's cool
[25-Jun-2009 14:07:02] <venturaville> now if I could just figure out why it won't import my class I am in business
[25-Jun-2009 14:11:12] <rocket> venturaville: problem loading that zenpack?
[25-Jun-2009 14:11:31] <rocket> venturaville: have you tried installing it from the commandline with a higher verbosity level?
[25-Jun-2009 14:12:33] <venturaville> doing that now ... was trying a few things
[25-Jun-2009 14:14:05] <venturaville> ObjectNotFound: Object with id SiteVIPsData not found on relation packables
[25-Jun-2009 14:14:08] <venturaville> interesting
[25-Jun-2009 14:14:28] <venturaville> ERROR:zen.ZPLoader:Error adding pack to /zport/dmd/zenMenus/More/zenMenuItems/VIPHostsData
[25-Jun-2009 14:14:45] <rocket> venturaville: got a relationship issue
[25-Jun-2009 14:14:51] <venturaville> looks like it
[25-Jun-2009 14:14:56] <rocket> venturaville: check both ends of the relationship
[25-Jun-2009 14:17:57] <venturaville> >>> menuitems
[25-Jun-2009 14:17:57] <venturaville> <ToManyContRelationship at /zport/dmd/zenMenus/More/zenMenuItems>
[25-Jun-2009 14:17:57] <venturaville> >>> menuitems.SiteVIPsData
[25-Jun-2009 14:17:57] <venturaville> 2009-06-25 15:17:08 WARNING ZODB Could not import class 'ZenPack' from module 'ZenPacks.Turner.HostMap'
[25-Jun-2009 14:24:38] <rocket> venturaville: how are you creating your menus?
[25-Jun-2009 14:24:43] <rocket> venturaville: code or gui?
[25-Jun-2009 14:24:45] <venturaville> in objects.xml
[25-Jun-2009 14:25:11] <rocket> you mean hand editing objects.xml?
[25-Jun-2009 14:25:17] <venturaville> yep
[25-Jun-2009 14:25:23] <rocket> ick!
[25-Jun-2009 14:25:27] <venturaville> indeed
[25-Jun-2009 14:25:28] <rocket> dont do that please
[25-Jun-2009 14:25:43] <venturaville> I copied it over from an existing zenpack
[25-Jun-2009 14:26:11] <venturaville> <!-- ('', 'zport', 'dmd', 'zenMenus', 'More', 'zenMenuItems', 'SiteVIPsData') -->
[25-Jun-2009 14:26:11] <venturaville> <object id='/zport/dmd/zenMenus/More/zenMenuItems/SiteVIPsData' module='Products.ZenModel.ZenMenuItem' class='ZenMenuItem'>
[25-Jun-2009 14:27:07] <rocket> venturaville: grab the aix zenpack for me and take a look at __init__.py
[25-Jun-2009 14:27:13] <venturaville> k
[25-Jun-2009 14:27:16] <rocket> look for this section
[25-Jun-2009 14:27:28] <rocket> def installMenuItems(self, dmd):
[25-Jun-2009 14:27:33] <venturaville> AIX SNMP?
[25-Jun-2009 14:27:36] <rocket> yes
[25-Jun-2009 14:27:50] <rocket> this is the code to add to the More menu as an example
[25-Jun-2009 14:27:51] <rocket> # Set an additional item in the More Menu
[25-Jun-2009 14:27:51] <rocket> #dmd.zenMenus.More.manage_addZenMenuItem("aixprintqueueDetail",
[25-Jun-2009 14:27:51] <rocket> # action="aixprintqueueDetail",
[25-Jun-2009 14:27:51] <rocket> # description="Aix Print Queue Details",
[25-Jun-2009 14:27:54] <rocket> # ordering=5.0)
[25-Jun-2009 14:28:08] <rocket> eg that would add aixprintqueueDetail to the More menu
[25-Jun-2009 14:28:23] <rocket> and would need the new zope page template aixprintqueueDetail.pt
[25-Jun-2009 14:28:53] <rocket> this is the code to remove it on zenpack uninstall
[25-Jun-2009 14:28:58] <rocket> def removeMenuItems(self,dmd):
[25-Jun-2009 14:28:58] <rocket> # Remove an additional item in the More Menu
[25-Jun-2009 14:28:58] <rocket> #id = "aixprintqueueDetail"
[25-Jun-2009 14:28:58] <rocket> #try:
[25-Jun-2009 14:28:58] <rocket> # dmd.zenMenus.More.manage_deleteZenMenuItem(("aixprintqueueDetail",))
[25-Jun-2009 14:29:01] <rocket> #except (KeyError, AttributeError):
[25-Jun-2009 14:29:03] <rocket> # pass
[25-Jun-2009 14:29:24] <rocket> obviously it needs to be uncommented
[25-Jun-2009 14:29:44] <rocket> but it is much cleaner than editing object.xml
[25-Jun-2009 14:29:57] <venturaville> +1 to that
[25-Jun-2009 14:30:46] <rocket> venturaville: you probably need to create the def install method also
[25-Jun-2009 14:31:01] <rocket> venturaville: but that should be straight forward enough from my zenpack
[25-Jun-2009 14:31:14] <venturaville> 'def install'?
[25-Jun-2009 14:31:27] <rocket> def install(self, app):
[25-Jun-2009 14:31:38] <rocket> self.installMenuItems(app.zport.dmd)
[25-Jun-2009 14:31:53] <rocket> ZenPackBase.install(self, app)
[25-Jun-2009 14:31:57] <rocket> for example ..
[25-Jun-2009 14:32:20] <venturaville> helpful
[25-Jun-2009 14:36:01] <rocket> venturaville: make sense?
[25-Jun-2009 14:36:08] <venturaville> so does that all preclude the factory_information in the class.......
[25-Jun-2009 14:36:26] <rocket> venturaville: ?
[25-Jun-2009 14:36:42] <venturaville> class Site(Device):
[25-Jun-2009 14:36:42] <venturaville> "Web Site Device"
[25-Jun-2009 14:36:42] <venturaville> factory_type_information = (
[25-Jun-2009 14:36:42] <venturaville> {
[25-Jun-2009 14:36:51] <venturaville> blah blah .... set actions ....
[25-Jun-2009 14:36:59] <venturaville> never mind I think I answered my own question
[25-Jun-2009 14:37:05] <rocket> ok
[25-Jun-2009 14:37:42] <rocket> venturaville: it has nothing to do with set actions etc
[25-Jun-2009 14:37:58] <rocket> venturaville: its all part of the Zenpack setup via its __init__.py
[25-Jun-2009 14:38:23] <rocket> eg ~/ZenPacks/ZenPacks.community.AIX/ZenPacks/community/AIX/__init__.py
[25-Jun-2009 14:45:14] <venturaville> no dice for the menu showing up though it complains less
[25-Jun-2009 14:46:14] <rocket> venturaville: complaining at all?
[25-Jun-2009 14:46:28] <venturaville> ERROR:Zope:manage_beforeDelete() threw
[25-Jun-2009 14:46:28] <venturaville> Traceback (most recent call last):
[25-Jun-2009 14:46:28] <venturaville> File "/opt/zenoss/zenoss/lib/python/OFS/ObjectManager.py", line 332, in manage_beforeDelete
[25-Jun-2009 14:46:28] <venturaville> object.manage_beforeDelete(item, container)
[25-Jun-2009 14:46:28] <venturaville> File "/opt/zenoss/zenoss/Products/ZenRelations/ToOneRelationship.py", line 145, in manage_beforeDelete
[25-Jun-2009 14:46:29] <venturaville> self._remoteRemove()
[25-Jun-2009 14:46:31] <venturaville> File "/opt/zenoss/zenoss/Products/ZenRelations/ToOneRelationship.py", line 97, in _remoteRemove
[25-Jun-2009 14:46:33] <venturaville> remoteRel._remove(self.__primary_parent__)
[25-Jun-2009 14:46:35] <venturaville> File "/opt/zenoss/zenoss/Products/ZenRelations/ToManyRelationship.py", line 113, in _remove
[25-Jun-2009 14:46:37] <venturaville> raise ObjectNotFound(
[25-Jun-2009 14:46:39] <venturaville> ObjectNotFound: Object with id SiteVIPsData not found on relation packables
[25-Jun-2009 14:46:41] <venturaville> ugly
[25-Jun-2009 14:46:54] <venturaville> that is on an install
[25-Jun-2009 14:48:13] <rocket> did you put a self.removeMenuItems(dmd)
[25-Jun-2009 14:48:18] <rocket> in def installMenuItems(self, dmd):
[25-Jun-2009 14:48:22] <rocket> if so remove it
[25-Jun-2009 14:48:37] <rocket> or wrap it in try: except: block
[25-Jun-2009 14:54:07] <venturaville> much better
[25-Jun-2009 14:54:14] <venturaville> no errors at least
[25-Jun-2009 14:54:47] <venturaville> no menu shows up though
[25-Jun-2009 14:54:49] <venturaville> frak
[25-Jun-2009 14:55:00] <rocket> ?
[25-Jun-2009 14:55:24] <venturaville> it shows up in more menu items
[25-Jun-2009 14:55:29] <venturaville> but not on the device itself now
[25-Jun-2009 14:55:34] <venturaville> no errors generated on install
[25-Jun-2009 14:56:13] <rocket> bbiab have a meeting
[25-Jun-2009 14:57:23] <rocket> ok have 5 mins
[25-Jun-2009 14:57:51] <rocket> venturaville: is that not the right location or you need it in that place and others
[25-Jun-2009 15:04:30] <venturaville> which right location are you refering to?
[25-Jun-2009 15:04:53] <venturaville> what's your email?
[25-Jun-2009 15:05:08] <venturaville> might be easier to send you the zip file then paste random snippets
[25-Jun-2009 15:22:55] <venturaville> I dropped it here for now: http://nicklay.com/public/ZenPacks.Turner.HostMap-1.18-py2.4.egg
[25-Jun-2009 16:17:42] <UNIX_USENIX> Is there a way to tell zensyslog to never start?
[25-Jun-2009 16:18:05] <UNIX_USENIX> I already have splunk running on the same host and port and I would much rather have splunk handling those issues.
[25-Jun-2009 16:40:39] <rmatte> UNIX_USENIX: yes, one second
[25-Jun-2009 16:41:07] <rmatte> http://blog.zenoss.com/2009/01/27/tip-of-the-month-running-only-some-daemons/
[25-Jun-2009 16:41:09] <rmatte> voila
[25-Jun-2009 16:41:22] <rmatte> and don't forget to put zenjobs in the list, it's not in the list there because that predates it
[25-Jun-2009 16:41:59] <UNIX_USENIX> Excellent, thanks for the link and advice
[25-Jun-2009 16:42:29] <rmatte> no problem
[25-Jun-2009 16:42:39] <rmatte> I use it to stop zenmodeler myself
[25-Jun-2009 16:42:58] <rmatte> since I don't want it remodelling large groups of servers and network devices on a regular basis
[25-Jun-2009 18:51:20] RoninX341 is now known as etank
[25-Jun-2009 19:28:47] <mjbrooks> Is there a place where there is a comparison of features? I'm looking for info on zenoss enterprise vs. zenoss community vs. nagios
[26-Jun-2009 00:00:42] [disconnected at Fri Jun 26 00:00:42 2009]
[26-Jun-2009 00:00:42] [connected at Fri Jun 26 00:00:42 2009]
[26-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[26-Jun-2009 09:32:09] <rocket> venturaville: ping
[26-Jun-2009 09:38:34] <venturaville> rocket: pong
[26-Jun-2009 09:39:05] <rocket> so I was able to get your zenpack to install
[26-Jun-2009 09:39:18] <rocket> havent figured out why the two lines I had to comment out arent working
[26-Jun-2009 09:39:35] <venturaville> yep ... then you just need to create a site in Devices/Site (use none modeler)
[26-Jun-2009 09:39:45] <venturaville> create some IPs in Devices/VIP
[26-Jun-2009 09:40:05] <venturaville> and in zVIPNameList in Site Device add in list of IPs......
[26-Jun-2009 09:40:27] <venturaville> if it was working then you should be able to get that list in the VIPs tab on the Site Device
[26-Jun-2009 09:40:34] <venturaville> (or so I am trying to get it to) :-)
[26-Jun-2009 09:41:17] <venturaville> someone else here is working on a fetcher for fetching load balancer configs to push into it
[26-Jun-2009 09:41:44] * rocket is confused on the devices/site step
[26-Jun-2009 09:41:52] <rocket> venturaville: how are you doing that?
[26-Jun-2009 09:42:08] <venturaville> add device......
[26-Jun-2009 09:42:42] <rocket> so a site is a device in this case?
[26-Jun-2009 09:42:46] <venturaville> yeah
[26-Jun-2009 09:43:03] <rocket> confusing from a ui perspective but it works ..
[26-Jun-2009 09:43:04] <venturaville> i.e. www.cnn.com
[26-Jun-2009 09:43:14] <venturaville> it makes more sense for what I am planning later
[26-Jun-2009 09:44:00] <venturaville> i.e. tag a group (like a customer) to each site device, and then do rollup reporting for traffic, power, CPU, memory, disk elsewhere.....
[26-Jun-2009 09:44:48] <venturaville> hopefully I can make this generic enough to be useful for others
[26-Jun-2009 09:45:27] <rocket> venturaville: to get it to install I had to comment out these two lines in init
[26-Jun-2009 09:45:31] <rocket> #sitedc._setProperty('zPythonClass', 'ZenPacks.Turner.HostMap.Site')
[26-Jun-2009 09:45:39] <rocket> #vipdc._setProperty('zPythonClass', 'ZenPacks.Turner.HostMap.VIP')
[26-Jun-2009 09:45:55] <venturaville> yeah I think _setProperty doesn't like that it is set already in objects.xml
[26-Jun-2009 09:45:59] <rocket> it looks like they arent needed as the settings must also be in the object.xml for it
[26-Jun-2009 09:47:23] <rocket> venturaville: when I tried to add an ip to vip i got this
[26-Jun-2009 09:47:40] <rocket> /ZenPacks.Turner.HostMap-1.18-py2.4.egg/ZenPacks/Turner/HostMap/VIP.py", line 11, in VIP _properties = Device._properties + ( NameError: name '_kw' is not defined
[26-Jun-2009 09:50:01] <venturaville> what I get for cutting and pasting.........
[26-Jun-2009 09:52:15] <rocket> replaced the lines with
[26-Jun-2009 09:52:17] <rocket> dict(id='vipSLBHealthcheck',type='string', 'mode':'w'),
[26-Jun-2009 09:52:17] <rocket> dict(id='vipExpectedResponse', type='string', 'mode':'w'),
[26-Jun-2009 09:52:42] <venturaville> _kw = dict(mode='w')
[26-Jun-2009 09:54:17] <rocket> retesting ..
[26-Jun-2009 09:56:03] <venturaville> I'm a dope
[26-Jun-2009 09:56:18] <venturaville> just realized I was dropping the link in the 'more' menu instead of making it a tab
[26-Jun-2009 09:56:22] <venturaville> no wonder I couldn't find it
[26-Jun-2009 09:57:02] <rocket> and you have an issue with the vips tab
[26-Jun-2009 09:57:07] <venturaville> yep
[26-Jun-2009 09:57:23] <venturaville> working on those now ... now that I can find them :-P
[26-Jun-2009 09:57:40] <rocket> venturaville: ok .. so you are good to go for now then?
[26-Jun-2009 09:57:53] <venturaville> I think so... I'm going to plug away at this for a bit
[26-Jun-2009 09:57:57] <venturaville> thanks for the assist
[26-Jun-2009 09:58:01] <rocket> np
[26-Jun-2009 10:28:16] <venturaville> maybe I am missing something but is the modeler the only place where a device class gets loaded from?
[26-Jun-2009 10:46:53] <venturaville> rocket: finally restart the zenoss stack ... and it just started working *sigh*
[26-Jun-2009 10:53:16] <rocket> so you are all good?
[26-Jun-2009 10:53:43] <venturaville> almost ... I'm in template debugging mode ... so at least I am making progress now :-)
[26-Jun-2009 10:54:08] <rocket> Thats great
[26-Jun-2009 11:15:16] <DarthPointer> after upgrading my Rpath Vmware image to 2.4 I'm now seeing extremely high cpu usage; top on the system just shows zenoss / python
[26-Jun-2009 11:15:34] <DarthPointer> thoughts?
[26-Jun-2009 11:17:26] <mrayzenoss> try 'su - zenoss' and do 'zenmigrate --step=JobManager'
[26-Jun-2009 11:23:12] <DarthPointer> thank you; I'll give it a shot. I read on a forum about how quickly zenoss becomes "production"; we're going to migrate it to a dedicated box
[26-Jun-2009 11:23:33] <mrayzenoss> probably a good idea
[26-Jun-2009 11:47:51] <DarthPointer> mrayzenoss; fwiw, that seems to have cleared up the issue
[26-Jun-2009 11:48:21] <mrayzenoss> not sure what the root cause is, but zenmigrate seems to clear it up for some people
[26-Jun-2009 11:49:02] <DarthPointer> The process looks like it had been launched over & over again; there were 20+ zenoss / python processes each using ~5% of the processor
[26-Jun-2009 11:51:20] <DarthPointer> thank you very much, btw
[26-Jun-2009 11:51:37] <mrayzenoss> np
[26-Jun-2009 12:07:32] [disconnected at Fri Jun 26 12:07:32 2009]
[27-Jun-2009 00:00:42] [connected at Sat Jun 27 00:00:42 2009]
[27-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[27-Jun-2009 06:34:12] HKhan_ is now known as HKhan
[27-Jun-2009 23:53:32] <mjbrooks> Does Zenoss use it's own copy of Apache & MySQL or will it use the ones already installed on the system?
[28-Jun-2009 00:00:42] [disconnected at Sun Jun 28 00:00:42 2009]
[28-Jun-2009 00:00:42] [connected at Sun Jun 28 00:00:42 2009]
[28-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[28-Jun-2009 14:00:06] <etank> mjbrooks: i know it uses zope. not sure if apache is in the picture or not
[29-Jun-2009 00:00:42] [disconnected at Mon Jun 29 00:00:42 2009]
[29-Jun-2009 00:00:42] [connected at Mon Jun 29 00:00:42 2009]
[29-Jun-2009 00:00:52] [zenoss-logger (logger bot) has joined #zenoss]
[29-Jun-2009 03:48:14] <Younger> my Zenoss does not create an event for reboot. does anyone know how to enable it?
[29-Jun-2009 04:59:28] <abdo> hi all
[29-Jun-2009 05:00:14] <abdo> I want to add a aprocess details similar to RRD graphs in OSPRocess.
[29-Jun-2009 05:03:37] <abdo> when I start apache I want to see a process details numbers upon the process like RRD graphs
[29-Jun-2009 05:06:34] <abdo> Anybody here?
[29-Jun-2009 07:15:46] <etank> is there a way to filter out an event that happens during a specific time of day without using a maintenance window?
[29-Jun-2009 07:16:45] <etank> we reindex our database at 5:30am and today I got over 4000 events from our web server trying to do an insert (thank God for deduplication)
[29-Jun-2009 08:12:19] <eidolon> hi folks - we're seeing some odd alerts coming from our zenoss monitors, particularly about disk space:
[29-Jun-2009 08:12:21] <eidolon> [zenoss] bos-aspac-tnz-app04 Disk space low: 210.4% used (-410.53 GB free)
[29-Jun-2009 08:28:16] <etank> that does look odd eidolon
[29-Jun-2009 08:28:25] <eidolon> yeah
[29-Jun-2009 08:29:47] <Jontler> id like to get 210 percent out of my drives
[29-Jun-2009 08:51:55] <abdo> hi all
[29-Jun-2009 08:52:50] <abdo> i want to retrieve all apache process and mysql process by process instead of RRD graph
[29-Jun-2009 08:53:26] <abdo> I used the zen-core package for apache and mysql monitoring
[29-Jun-2009 08:54:18] <abdo> but it list data on the RRD Graph, I want to list these data on OS process tab instead of Prof tab
[29-Jun-2009 08:56:06] <abdo> or anybody tell me about documentation part should I read.
[29-Jun-2009 08:58:02] <abdo> anybody here? anyone answer me to make sure that I am not alone!
[29-Jun-2009 08:59:52] <eidolon> so, next question. is there a way to customize a 'per-alert' message? ala, if we have an alert for low disk space, can i add something to the email generated saying something like "Contact bob in accounting to clean this up." - I don't want to edit my personal alerting rules, since that's global to all msgs i get.
[29-Jun-2009 09:09:42] <venturaville> eidolon: we do something similar to that... you can set a variable inside of an event transform (for adding a custom message for instance), and then picking up that variable in the alerting rules
[29-Jun-2009 09:10:08] <eidolon> hmmm
[29-Jun-2009 09:10:15] <eidolon> that's not a bad idea.
[29-Jun-2009 09:10:41] <eidolon> the alerts are generally being geneated by nagios checkers... hwo would i embed a var in there? somethingon the CL?
[29-Jun-2009 09:10:56] <venturaville> alerts or events?
[29-Jun-2009 09:11:02] <venturaville> there is a big distinction in zenoss
[29-Jun-2009 09:11:24] <venturaville> events = just that ... alerts = messages sent outbound...
[29-Jun-2009 09:12:24] <venturaville> if it is sending out an event, then you can write an event mapping rule to match it, and then put something in event transform to set a message
[29-Jun-2009 09:12:26] <venturaville> like:
[29-Jun-2009 09:12:37] <venturaville> evt.appendthismessage = 'message to append'
[29-Jun-2009 09:12:59] <venturaville> then in alerts rules, you would just have to check for that variable
[29-Jun-2009 09:13:05] <eidolon> well, the alerts are being generated by the events. so i'd have to add th emessage to the event, to have it embedded in the alert.
[29-Jun-2009 09:13:06] * eidolon gets the terms.
[29-Jun-2009 09:13:12] <venturaville> k
[29-Jun-2009 09:15:03] <eidolon> hmmmmm.
[29-Jun-2009 09:15:06] * eidolon makes notes.
[29-Jun-2009 09:16:02] <eidolon> thanks, i'll experiment.
[29-Jun-2009 12:07:01] <etank> will something as simple as http://dpaste.com/61196/ work as a transform?
[29-Jun-2009 12:25:53] <jb> sup everyone..
[29-Jun-2009 12:28:32] <rmatte> sup
[29-Jun-2009 12:28:48] <jb> just looking at the 2.4.2 notes
[29-Jun-2009 12:28:51] <jb> been out of town for a week..
[29-Jun-2009 12:31:12] <rmatte> cool
[29-Jun-2009 12:32:04] <jb> Process monitoring for Linux and AIX SSH hosts currently is disabled. Until re-enabled, process modeling
[29-Jun-2009 12:32:05] <jb> should be disabled (so that no processes appear when the device is modeled). (Internal Defect 1695)
[29-Jun-2009 12:32:06] <jb> hm
[29-Jun-2009 12:37:08] <rmatte> I'm trying to restore a Products directory that I stupidly accidentally deleted
[29-Jun-2009 12:37:14] <rmatte> almost done
[29-Jun-2009 13:16:25] <Schwartz> afternoon all
[29-Jun-2009 13:23:27] <Schwartz> Anyone knowledgeable with thresholds and expressions?
[29-Jun-2009 13:24:36] <rocket> jb: ping
[29-Jun-2009 13:25:52] <rocket> Schwartz: ask the question and hopefully someone will know ..
[29-Jun-2009 13:29:47] <jb> rocket: sup
[29-Jun-2009 13:29:58] <rocket> jb: my aix zenpack is posted
[29-Jun-2009 13:30:04] <rocket> jb: and the thresholds as well
[29-Jun-2009 13:30:09] <jb> yeah i saw
[29-Jun-2009 13:30:12] <jb>
)[29-Jun-2009 13:30:18] <rocket> jb: point, predictive, range ..
[29-Jun-2009 13:30:31] <jb> have you tried the apar to fix snmp?
[29-Jun-2009 13:30:42] <Schwartz> My Python's a bit thin, and I'm trying to understand what variables are available to the min/max value fields for thresholds. The documentation just refers to using a python expression. Some of the bits I've been able to find for examples are for disk percentage utilization. However, I'm unable to find what is being referred to by here.totalBlocks etc, doesn't match up with RRD datasource as far as I can tell. In that same line I'm trying to figure out
[29-Jun-2009 13:30:44] <rocket> not yet
[29-Jun-2009 13:30:59] <rocket> jb: its on my list .. but not a high priority yet
[29-Jun-2009 13:31:18] <jb> ugh aix snmpd
[29-Jun-2009 13:31:23] <jb> so much configuration
[29-Jun-2009 13:32:11] <twm1010> Afternoon all.
[29-Jun-2009 13:32:23] <rmatte> Schwartz: here.totalBlocks is collected by a collector plugin
[29-Jun-2009 13:32:58] <rocket> jb: yes that is a problem with aix snmp ;P
[29-Jun-2009 13:33:00] <jb> rocket: how did you configure snmpd?
[29-Jun-2009 13:33:09] <jb> i was reading thru that cacti thread
[29-Jun-2009 13:33:22] <rmatte> I thought the AIX pack was just drop and go
[29-Jun-2009 13:33:22] <rocket> jb: I followed that thread ..
[29-Jun-2009 13:33:32] <jb> what a pain
for 50+ boxes[29-Jun-2009 13:33:42] <rmatte> oh, you mean actually configuring each box?
[29-Jun-2009 13:33:45] <jb> i guess I could write a script...
[29-Jun-2009 13:33:46] <rocket> rmatte: it is except aix snmp suxors to setup to get the cpu info
[29-Jun-2009 13:33:48] <jb> yeah
[29-Jun-2009 13:33:57] <rmatte> I'd write a script, I never do mass changes like that by hand
[29-Jun-2009 13:34:00] <rmatte> waste of time
[29-Jun-2009 13:34:07] <jb> yeah
[29-Jun-2009 13:34:13] <jb> you havent written one yet though have you?
[29-Jun-2009 13:34:32] <rmatte> nope
[29-Jun-2009 13:34:34] <rmatte> but I could
[29-Jun-2009 13:34:41] <rocket> rmatte: I might be able to work on a ssh based aix zenpack but if I do some work with zenoss it would probably be an updated enterprise zenpack
[29-Jun-2009 13:34:56] <jb> ssh aix!
[29-Jun-2009 13:35:01] <rmatte> enterprise zenpack, ewwww
[29-Jun-2009 13:35:12] <rmatte> I vote community WMI pack
[29-Jun-2009 13:35:45] <rocket>
[29-Jun-2009 13:35:53] <jb> so there is talk of an updated aix ssh zenpack?
[29-Jun-2009 13:36:04] <jb> to incorporate the same features as your snmp?
[29-Jun-2009 13:36:07] <Schwartz> Ah. Is it possible ot have the threshold use a datasource in an rrd table? This is more for a bit of SNMP MIB that's outside of the normal zenoss world.
[29-Jun-2009 13:36:25] <rocket> well there is talk of me doing some zenpacks for them .. and if I do that I might not be able to do some community stuff
[29-Jun-2009 13:36:35] <jb> right
[29-Jun-2009 13:37:05] <rocket> and there is some talk of me working on aix but nothing nailed down at all yet
[29-Jun-2009 13:37:11] <rmatte> Schwartz: you'd have to collect the data via an snmp datapoint (figure out what OID it is and create it), then create a threshold against that value
[29-Jun-2009 13:37:30] <jb> im still thinking that SSH is the way to go with AIX
[29-Jun-2009 13:37:33] <rocket> jb: but I am sure there are some who would like an aix ssh zenpack
[29-Jun-2009 13:37:42] <jb> well enterprise has one..
[29-Jun-2009 13:37:49] <jb> just none of the features that you have in your SNMP pack
[29-Jun-2009 13:38:10] <rocket> jb: are my features what you would like in the enterprise one?
[29-Jun-2009 13:38:14] <jb> yes
[29-Jun-2009 13:38:17] <jb> definetely
[29-Jun-2009 13:38:22] <jb> the lvm stuff is huge
[29-Jun-2009 13:38:37] <rocket>
[29-Jun-2009 13:39:04] <jb> it already does the CPU and stuff (basic)
[29-Jun-2009 13:39:16] <jb> but none of your relationships
[29-Jun-2009 13:39:23] <jb> or print queues, paging
[29-Jun-2009 13:39:50] <rocket> or lpar info etc ..
[29-Jun-2009 13:39:58] <jb> right
[29-Jun-2009 13:40:01] <rocket> jb: have you loaded my zenpack?
[29-Jun-2009 13:40:03] <jb> no
[29-Jun-2009 13:40:06] <jb> not yet
[29-Jun-2009 13:40:25] <jb> i need to spin it up in dev
[29-Jun-2009 13:40:37] <rocket> I would spin it up there myself also
[29-Jun-2009 13:40:49] <rocket> jb: I wouldnt trust myself either
[29-Jun-2009 13:41:31] <jb> haha
[29-Jun-2009 13:41:33] <jb> its not that
[29-Jun-2009 13:41:53] <jb> i just don't want to conflict with any of my other templates
[29-Jun-2009 13:42:04] <jb> just a precaution
[29-Jun-2009 13:42:06] <rocket> jb: I know I was just pulling your chain
[29-Jun-2009 13:42:16] <jb> but yeah.. aix ssh would be ideal at least for me
[29-Jun-2009 13:42:21] <jb> just because no further configuration
[29-Jun-2009 13:42:23] <jb> and ssh is stable
[29-Jun-2009 13:42:26] <jb>
[29-Jun-2009 13:42:52] <rocket> jb: only issue I can see is if I run into a command that requires root priveledges to gather the info
[29-Jun-2009 13:43:01] <jb> yeah
[29-Jun-2009 13:43:05] <jb> can you think of any?
[29-Jun-2009 13:43:07] <rocket> jb: otherwise it should be a nice leisurely walk in the park ..
[29-Jun-2009 13:43:14] <rocket> jb: off the top of my head no ..
[29-Jun-2009 13:43:26] <jb> sudo
[29-Jun-2009 13:43:31] <jb> if we must
[29-Jun-2009 13:43:46] <rocket> yes but I would hate to have that etc as a setup requirement
[29-Jun-2009 13:43:52] <jb> uyeah
[29-Jun-2009 13:43:53] <jb> but
[29-Jun-2009 13:44:01] <jb> well
[29-Jun-2009 13:44:03] * rocket is thinking basic setuid scripts if we must
[29-Jun-2009 13:44:05] <jb> the snmpd configuration is even worse!
[29-Jun-2009 13:44:07] <jb> yeah
[29-Jun-2009 13:44:29] <rocket> eg it reads .. no inputs for the script and just dumps parsed info etc
[29-Jun-2009 13:44:49] <pupmini> hello
[29-Jun-2009 13:44:50] <rocket> that probably has the best chance of being accepted by adminds etc ..
[29-Jun-2009 13:44:55] <jb> yeah
[29-Jun-2009 13:44:58] <jb> that sounds better
[29-Jun-2009 13:47:50] <jb> im not sure if anything would even require root though
[29-Jun-2009 13:47:53] <jb> but i'd have to check
[29-Jun-2009 13:48:16] <rocket> yeah again it would be a minor annoyance if it needed root for something
[29-Jun-2009 13:48:32] <jb> solid CPU, memory, IO performance monitoring with LVM relationships would be an awesome start
[29-Jun-2009 13:48:37] <pupmini> has anyone here used WMI with zenoss core?
[29-Jun-2009 13:49:05] <jb> and im pretty sure all of that could be done without root
[29-Jun-2009 13:49:53] <jb> IO per hdisk, and combined IO maybe
[29-Jun-2009 13:56:05] <twm1010> pupmini: yes
[29-Jun-2009 13:56:07] <twm1010> extensively
[29-Jun-2009 13:56:40] <twm1010> I took the wmi community zenpack, and just used it as a template, did all sorts of monitoring, exchange, terminal services, a bunch of others
[29-Jun-2009 13:56:54] <pupmini> ah cool
[29-Jun-2009 13:57:01] <pupmini> i am reading the docs
[29-Jun-2009 13:57:18] <pupmini> and it appears that it does not popular any of the device information automatically...is that correct?
[29-Jun-2009 13:57:34] <pupmini> populate*
[29-Jun-2009 13:58:26] <rocket> jb: its all feasable .. just a matter of time and what my legal obligations end up being
[29-Jun-2009 14:05:36] <jb> the ACL's
[29-Jun-2009 14:12:35] <Schwartz> rmatte: Yep, I am already collecting via SNMP the various stats and even graphing them, but the problem that the maximum value of a particular host is dependent on another SNMP value. Something that would be similar would to use the file system one as an example. Assume that there's no collector variable for the number of blocks, so you're collecting both the used blocks and total blocks via snmp. I would like to have a threshold for 70% of its capacit
[29-Jun-2009 14:37:35] <rmatte> Schwartz: the way I personally do it is to write a script and use it as a datapoint
[29-Jun-2009 14:37:42] <venturaville> how do you do the variable substitution in objects.xml to make a link to your own check_* script when the zenpack directory is versioned..........?
[29-Jun-2009 14:37:44] <rmatte> the script would calculate the percentage of utilization
[29-Jun-2009 14:37:54] <rmatte> then you just set the threshold at 90
[29-Jun-2009 14:38:35] <Schwartz> how are you referring to the datapoint that you have the script populating in the threshold?
[29-Jun-2009 14:39:33] <rmatte> your script just has to output like: OK|DataPoint1=value DataPoint2=value WhateverYouWant=value
[29-Jun-2009 14:39:33] <rmatte> then you add it as a command based datapoint
[29-Jun-2009 14:39:37] <rmatte> and create datapoints underneath it with those names
[29-Jun-2009 14:39:57] <rmatte> download James Dastrup's WMI Performance Zenpack is an example and take a look at how he set that up
[29-Jun-2009 14:40:02] <rmatte> and then you'll understand
[29-Jun-2009 14:40:21] <kgoedtel> maybe one day they will let you use snmp datapoints in the threshold expression
[29-Jun-2009 14:40:30] <kgoedtel> i've been wanting that for years :[
[29-Jun-2009 14:40:42] <rmatte> yeh, that would definitely be nice
[29-Jun-2009 14:41:24] <Schwartz> lemme give that a go..
[29-Jun-2009 14:44:25] <Schwartz> Heh, it wants windows stuff.
[29-Jun-2009 14:44:35] <rmatte> wants windows stuff?
[29-Jun-2009 14:44:45] <rmatte> what does?
[29-Jun-2009 14:45:12] <Schwartz> wmiperformancemonitor.zip WARNING:zen.AddToPack:Unable to find context path /zport/dmd/Devices/Server/Windows/rrdTemplates (line 4 ?) for WMI Performance Monitors
[29-Jun-2009 14:45:22] <Schwartz> though I did delete the server/windows group
[29-Jun-2009 14:45:32] <rmatte> that's your problem then
[29-Jun-2009 14:46:02] <rmatte> it's a Window monitoring Zenpack so it expects that group to be there
[29-Jun-2009 14:46:13] <rmatte> but I'm just having you checkout the template so you get an idea of how you have to set it up
[29-Jun-2009 14:46:19] <Schwartz> right
[29-Jun-2009 14:49:43] <rmatte> ah crap, I screwed up one of my scripts this morning with a modification, just noticed, gotta fix this...
[29-Jun-2009 14:57:17] <rmatte> there, fixed
[29-Jun-2009 14:57:43] <rmatte> example of output:
[29-Jun-2009 14:57:44] <rmatte> zenoss@D1Z01:/usr/local/zenoss/zenoss/Products/SNMPWindowsPerfMonitor/libexec$ ./wincpu.sh CCADB1 community-string
[29-Jun-2009 14:57:45] <rmatte> OK|CPU1=2 CPU2=93 CPU3=4 CPU4=2 Total=25
[29-Jun-2009 14:59:38] <kgoedtel> actually hrmmm, maybe you can use the data points like this in threshold
[29-Jun-2009 14:59:42] <Schwartz> Okay, I see what you're saying now. Would be so much nicer if we could use something like (here.TotalBlocks * 100)*.9 where TotalBlocks was a data source itself.
[29-Jun-2009 14:59:52] <kgoedtel> (here.getRRDValue('data_point')) * 0.70
[29-Jun-2009 14:59:56] <Schwartz> er minus the *100
[29-Jun-2009 15:00:05] <Schwartz> hmm.
[29-Jun-2009 15:00:11] <Schwartz> lemme try that.
[29-Jun-2009 15:00:43] <kgoedtel> i'm trying it for one of my alerts
[29-Jun-2009 15:01:22] <rmatte> actually, that could definitely work
[29-Jun-2009 15:01:27] <rmatte> never thought of that
[29-Jun-2009 15:01:30] <twm1010> are you just trying to set a disk space threshold?
[29-Jun-2009 15:01:48] <rmatte> no, he's trying to thresholds against some other RRD value
[29-Jun-2009 15:01:52] <rmatte> he was using disk space as an example
[29-Jun-2009 15:01:54] <Schwartz> technically, no. just using that as an easily understandable analogy.
[29-Jun-2009 15:02:03] <Schwartz> what he said
[29-Jun-2009 15:02:13] <kgoedtel> something that doesn't use a collector (snmp dp)
[29-Jun-2009 15:02:35] <twm1010> ah, just looks up currently stored RRD values, and goes off that?
[29-Jun-2009 15:02:44] <kgoedtel> i'm hoping
[29-Jun-2009 15:02:46] <Schwartz> it is coming in via snmp
[29-Jun-2009 15:02:47] <Schwartz> yeah
[29-Jun-2009 15:03:57] <twm1010> hrmm...
[29-Jun-2009 15:04:14] <twm1010> i wonder, can you do something similar to what was in the blog post a few weeks back?
[29-Jun-2009 15:04:44] <twm1010> @ http://blog.zenoss.com/2009/06/24/tip-of-the-month-tracking-file-system-growth/
[29-Jun-2009 15:08:46] <Schwartz> I wonder what the 'built-in' datasource type is and how its used.
[29-Jun-2009 15:12:10] <venturaville> for internal zenoss statistics for one
[29-Jun-2009 15:12:26] <venturaville> things like work list queue length (zenhub) etc..
[29-Jun-2009 15:12:31] <rmatte> is there no way to have Critical tickets age out after an hour?
[29-Jun-2009 15:13:07] <rmatte> seems to be a limitation of the option
