[09:56] * Now talking on #zenoss
[09:56] * Topic for #zenoss is: Zenoss developers will be here Tuesday, July 22 at 11am EST
[09:56] * Topic for #zenoss set by mrayzenoss at Tue Jul 15 17:16:55 2008
[09:56] you can set zCommandCycleTime in the zproperties for the device tree..
[09:56] i beleive that should do it.
[09:56] its defauletd to 60
[09:56] 2.1 or 2.2?
[09:56] 2.2
[09:57] * ericnewton (n=ericnewt@74.93.194.82) has joined #zenoss
[09:57] be careful doing 1 min polling cycle if you are pulling alot of oid your going to need a big box
[09:58] you can set it under /Monitors/Performance/localhost/editPerformanceConf
[09:58] 1 minute cycle eats a lot of disk IO if you have a high number of OIDS
[09:58] replace localhost with the collector name
[09:59] Hi everybody, Matt Ray from Zenoss here. Looks like discussions are already under way.
[10:00] * ian__ (n=__ian__@bas1-kitchener06-1096648702.dsl.bell.ca) has joined #zenoss
[10:00] There are a couple of Zenoss folks in the channel, they can introduce themselves if they like, or lurk anonymously
[10:00] Hello... this is Eric (not Erik) from Zenoss
[10:00] I'd prefer to lurk anonymously.
[10:01] wow... a lot more people than when I used to come here 2 years ago
[10:01] I will be the first to kiss a$$. Thanks Zenoss Devs for making such a great product!
[10:02] Now can you please do all my features requests?
[10:02] and mine too
[10:02] yes... +1 for all kiss-ass requests
[10:02] ericnewton would be happy to.
[10:02] so well have those next week
[10:02] so features... 2.2.3 is in the polishing stages
[10:03] in keeping with our normally timely schedules, yes of course
[10:03] good.
[10:03] sept for 2.2.3? or oct?
[10:03] * ckeyes (n=ckeyes@c-71-206-90-150.hsd1.fl.comcast.net) has joined #Zenoss
[10:03] not a lot of new functionality, lots of installer issues for this one
[10:03] i'm having to use 2.1.3 because of some bugs we have run into
[10:03] 2.2.3 is this week (hopefully). 2.3 is October
[10:03] ah
[10:03] it's just some TALES issues
[10:04] daMaestro davetoo davewinter
[10:04] * Shylock wishes for support of a debian based distro for Zenoss Enterprise
[10:04] * Johnny312 (n=Johnny31@c-98-206-10-126.hsd1.il.comcast.net) has joined #zenoss
[10:04] if you run Core, then you can probably pull a source release from the branch and be running 2.2.3 today
[10:04] daMaestro:have you opened tickets?
[10:04] mrayzenoss, yes
[10:04] daMaestro davetoo davewinter
[10:04] I have a question is there a place we can look at what is going to be in 2.3
[10:04] mrayzenoss, have not had time to send a patch
[10:04] daMaestro: have they been addressed?
[10:04] or any version
[10:04] mrayzenoss, dunno
[10:04] any upgrades to the docs with these upcomming releases?
[10:04] Shylock: have you tried the .deb stack installer?
[10:05] http://dev.zenoss.com/trac/ticket/3425
[10:05] is it true that there is no easy way to enable HTTPS in the web GUI?
[10:05] Title: #3425 (AttributeError when trying to render viewDevicePerformance) - Zenoss - Trac (at dev.zenoss.com)
[10:05] magnachef: minor documentation refreshes
[10:05] I can install core, but was informed that enterprise is not supported on Debian
[10:05] anyone running zenoss-enterprise with an offbox mysql? or mysql cluster?
[10:05] specifically.. our 64TB mount was reporting something like 5TB
[10:05] magnachef: dev doc stuff for adding tabs didn't make it in (2.2.4)
[10:05] davewinter: are you findin
[10:05] g mysql as a bottleneck?
[10:05] Shylock: this will be remedied with 2.2.3
[10:05] but there were many other graphing issues, which sucked
[10:06] Shylock: we have to make choices on how many platforms we can support officially. Debian may be covered by BitRock installers for Enterprise eventually
[10:06] ckrough, no, just need to have a common platform or other apps to use mysql. dont need a lot of mysql servers to manage.
[10:07] ericnewton: Ok... I just installed Cenos5 to get 2.2 up and going here, but I've been debian guy for the last 8 years or so... My redhat-fu has become weak.
[10:07] so, i'd really like to get interface monitoring working again
[10:07] we had a 32 million row history table that i had to deal with recently, so i need a more robust mysql solution rather than keeping it on the zneoss server
[10:07] daMaestro: do you know if netsnmp reports the correct value for your 64 TB mount?
[10:07] What is the long term direction for reporting? Will there be major changes in the way reports are created, or will you be sticking with the current method?
[10:07] ericnewton, it does, yet
[10:07] mistich: http://community.zenoss.com/forums/viewtopic.php?t=6651
[10:07] ericnewton, it also works correctly with 2.1.3
[10:07] Title: Forums :: View topic - The 2.3 Release Plan (at community.zenoss.com)
[10:08] what i'd really like help with is getting interface perf monitoring to work correctly
[10:08] yes I've seen the post but something more detailed
[10:08] daMaestro: please post a bug for the 64 TB mount and attach an SNMP walk of the device (edit out any bits you don't want public)
[10:08] hi, i've been testing the zenpop3 module for a while and have been having some trouble getting the program to generate events. pjp13 posted some fixes that have eliminated one of my problems (unable to parse the from: line), but i'm still getting a bunch of errors
[10:08] ericnewton, ok
[10:09] ckrough: I haven't seen much about changing reporting. I'd like to put out more documentation and examples so people can do more
[10:09] daMaestro: what issues are you having with interface graphs?
[10:09] i've documented my problems here - http://community.zenoss.com/forums/viewtopic.php?t=6677 - problem 2 is solved by pjp13's patch
[10:09] Throughput, Packets, Errors (via the perf tab) are not rendering with the shipped ethernetCsmacd_64 template
[10:09] yeah, maybe get a wiki going where people can post examples of their reports
[10:09] I also get:
[10:09] Title: Forums :: View topic - Zenpop3 errors - unable to generate e-mail event (at community.zenoss.com)
[10:10] Error reading value for "ifOutErrors" on server (oid .1.3.6.1.2.1.2.2.1.20 is bad)
[10:10] magnachef: We are down to 2 finalists for replacing our wiki and community infrastructure
[10:10] Error reading value for "ifHCInOctets" on server (oid .1.3.6.1.2.1.31.1.1.1.6 is bad)
[10:10] i get the "is bad" snmp issue sometimes as well.
[10:10] Error reading value for "ifHCOutOctets" on server (oid .1.3.6.1.2.1.31.1.1.1.10 is bad)
[10:10] etc
[10:10] MajereDB8: zenpop3 is unfinished... I expect it to take some elbo-grease to get it working
[10:10] magnachef: we're looking for something much easier to add content to and collaborate on
[10:11] So, is there a model diagram or documentation for all of the dmd classes? All of the zenoss classes inherit from such method heavy superclasses that introspection isn't very helpful in figuring out how to navigate the dmd.
[10:11] any chance ticket 363, acknowledging and clearing via smtp will make it in soon? I see that being bery useful
[10:11] I'm trying to create a catch-all event for NTBackup errors. The event class key is in the form of NTBackup_#### and component is NTBackup.. Is there a way to create an event that will map to any of these events without creating an event for each possible error??
[10:11] hi guys.
[10:11] what was the .bmp that you tried to dcc?
[10:12] i missed it..
[10:12] Shylock: we use a couple of techniques while working in zendmd
[10:12] I agree would be nice to have docs on all of the dmd classes
[10:12] http://fpaste.org/paste/3785 is what snmpwalk shows
[10:12] Shylock: we use "grepdir" to search for properties
[10:12] ericnewton: bummer. i understand how that goes. the zenackevents script posted by edahl is pretty slick and would work great in combination with zenpop3, but i guess all things come with time
[10:12] Title: Fedora Pastebin - Viewing paste #3785 (at fpaste.org)
[10:12] Shylock: I find it's easier to print obj.__dic__
[10:12] guys, is 2.2.1 out yet?
[10:12] jshadow: Nothing specific is being done on SMS/SMTP control that I'm aware of, but we are going to start cleaning out zenoss-3000 to get better traction on tickets
[10:12] so, obviously 1.3.6.1.2.1.31.1.1.1.10 is not bad
[10:13] mrayzenss: Ok, thanks
[10:13] jb: 2.2.1 and 2.2.2 didn't come out officially, 2.2.3 is about to come out
[10:13] daMaestro: is snmp version set to v2c on the device?
[10:13] mrayzenoss: can I expect a -CLEAN- upgrade from 2.2.0?
[10:13] ericnewton: Yeah... I generally load the rlcompleter tab completion stuff. I haven't used grepdir before, I'll try that.
[10:14] jb: we have spent most of the time/effort for 2.2.3 on upgrades
[10:14] awesome!
[10:14] jb: and testing
[10:14] my 2.1.2 -> 2.2 experience was horrific
[10:14] ericnewton, yes
[10:14] ericnewton, if i go to an individual interface.. the graph is properly generated
[10:14] so when can we expect 2.2.3? this week?
[10:15] jb: yes, if things keep going well
[10:15] ok
[10:15] ericnewton, i've just tried deleting lo and sit0 and then locked the ip interfaces to see if that gets rid of the snmp warnings and makes the "perf" tab interface graphs work
[10:15] 2.2.3 is in a RC right now
[10:15] so far, the perf tab graphs are still not rendering
[10:15] is it still the case that if you rename your device, you lose all of your RRD data associated with that device? Because I noticed that with 2.1.3
[10:15] I had read that there was an official BigIP zenpack, but I haven't seen any talk of it since the announcement..
[10:16] is this still in progress?
[10:16] On the note of the feature requests, it would be nice to have a voting system sort of like the ubuntu brainstorm
[10:16] daMaestro: you can run zenperfsnmp against one device in debug mode:
[10:16] jb: Enterprise has a BigIP ZenPack I believe
[10:16] ah.
[10:16] i have started one for core
[10:16] yes a voting system would be nice
[10:16] depends on how many votes I get
[10:16] i wonder what the enterprise pack does exactly
[10:16] daMastro: zenperfsnmp -v 10 -d myproblemdevice
[10:17] i also seem to have an issue with http and https ip service randomly stop monitoring
[10:17] jb: I'll see if we have a PDF
[10:17] one for you ckrough
[10:17] thanks
[10:17] daMaestro: I believe a bug was fixed: the services would toggle monitoring w/each modeling run
[10:18] * warwick (n=warwickp@ool-182ec7cf.dyn.optonline.net) has joined #zenoss
[10:18] deMaestro: fixed in 2.2.3
[10:18] * wonderhamster_ has quit (Read error: 104 (Connection reset by peer))
[10:18] hey guys, has there been any progress on documentation for making custom modelers?
[10:18] Will multi-graphing adding reports to zenpacks be fixed?
[10:18] ie, populating fields such as HW Make, OS Make, OS Version (and other custom fields maybe?) with data via snmp?
[10:18] * wonderhamster_ (n=wonderha@adsl-68-78-3-5.dsl.ipltin.ameritech.net) has joined #zenoss
[10:19] ptothedo: could you be more specific (and post a bug report if you haven't)?
[10:19] ericnewton, good news
[10:19] ericnewton, is there an upgrade path?
[10:20] jb: checking with the doc guy
[10:20] I'd like to talk about zenhub in 2.2. I'm seeing a lot of timeouts now.
[10:20] What changed?
[10:20] * daMaestro wonders if there are 2.2.3rc rpms somewhere?
[10:20] davetoo... in 2.2 we switched
[10:20] or even *gulp* a repo?
[10:20] I haven't created a bug report yet, but if I try to add sub report organizers to a zenpack they don't get add. Also, resequencing items in reporting isn't working for me.
[10:20] ;-/
[10:21] (sorry, I haven't looked at the changelists for it yet)
[10:21] davetoo: nearly every daemon to get data from zenhub
[10:21] http://community.zenoss.com/forums/viewtopic.php?t=3928&highlight=https <-- Is this really the only way to get the web GUI to support HTTPS?
[10:21] I think custom modelers are very important.. and we haven't seen any documentation/easy way to create them..
[10:21] daMaestro davetoo davewinter
[10:21] Title: Forums :: View topic - Modifying Zenoss for SSL encryption (at community.zenoss.com)
[10:21] davetoo: for simpler distributed collection
[10:21] ericnewton: I was speculating about that
[10:21] ericnewton, i assume that a workaround for now is to lock ip services?
[10:21] daMaestro: there is a yum repo, but it doesn't have rc rpms in it
[10:21] daMaestro: we haven't put together infrastructure for providing betas very well
[10:21] daMaestro davetoo davewinter
[10:21] mrayzenoss, ok
[10:21] daMaestro: yes for now
[10:22] daMaestro: betas tend to get out in the wild, we still see people running the pre-2.2. betas
[10:22] ian__, i'll look around and sync the tree into our cobbler infrastructure
[10:22] * cgibbons (n=cgibbons@75.148.215.110) has joined #zenoss
[10:22] davetoo: tell me a bit about your setup and where you are seeing timeouts
[10:23] jb: when we get the new wiki up, I plan on seeding discussion areas so people can get such stuff started if we don't provide it yet
[10:23] I think I have about 33,000 rrd files
[10:23] k
[10:23] * rdb (n=rdb@65.164.102.100) has joined #zenoss
[10:23] I'm looking for system specs right now...
[10:23] jb: Have you opened a ticket for it? That'll get it some love
[10:23] we are fighting zenperf and zenprocess timeouts as well
[10:23] davetoo: most of the collection via snmp?
[10:23] dual-Xeon 3G
[10:24] yeah theres a ticket
[10:24] yes
[10:24] eric dahl opened it I beileve..
[10:24] I'm trying to create a catch-all event for NTBackup errors. The event class key is in the form of NTBackup_#### and component is NTBackup.. Is there a way to create an event that will map to any of these events without creating an event for each possible error??
[10:24] jb: number? I'll get it to Doc and out of Erik's queue
[10:24] Error reading value for "eth0" on server (oid .1.3.6.1.2.1.2.2.1.10.2 is bad)
[10:24] Not much by commands, none by WMI
[10:24] whats the ticket URL again?
[10:24] trac
[10:24] blah, well after i removed the lo interface, i'm now getting these events
[10:24] davetoo: that's good... the timeouts: are you missing your cycle time?
[10:25] I began to notice the timeouts when adding new devices
[10:25] jb: dev.zenoss.com
[10:25] Not missing poll cycles currently
[10:25] eric: thanks
[10:26] * wonderhamster_ has quit (Read error: 60 (Operation timed out))
[10:26] davetoo: where do you see timeouts?
[10:26] http://fpaste.org/paste/3786 is what the shipped template requests, and i can snmpwalk those OIDs
[10:26] Title: Fedora Pastebin - Viewing paste #3786 (at fpaste.org)
[10:26] ericnewton: zendeviceloader timeouts
[10:27] http://pastebin.com/m686d5e5d
[10:27] I'm also surprised that I'm seeing zenstatus log messages in the zenhub logfile now....
[10:28] davetoo: that actually looks ok... have you investigated why those 10 devices have timedout?
[10:28] arg, I can't find the ticket number or the forum thread.
[10:29] davetoo: when you have bad devices zenpref retries the devices and it takes longer
[10:29] here it is
[10:29] mrayzenoss: http://dev.zenoss.org/trac/ticket/3292
[10:29] Title: #3292 (document modeler plugins) - Zenoss - Trac (at dev.zenoss.org)
[10:29] so it wasn't edahl
[10:29] if you remove or fix the bad devices your polling cycle will finish on time
[10:30] yes,
[10:30] Hey ericnewton, #3292 is in your box.
[10:30] but my problem is not currently with the snmp polling finishing on time
[10:30] mrayzenoss: oh good.. I was running out
[10:30]
[10:30] the problem is that while adding devices the connection to zenhub times out
[10:31] I get a message asking if zenhub is running
[10:31] speaking of numerous and interesting ZenPacks.... we've had 15 entries in the Submit a ZenPack contest
[10:31] oh, that I can believe
[10:31] from what you pasted all I see is the bad devices
[10:31] mrayzenoss: oh yeah? are they published anywhere?
[10:31] there's a big backlog to be published
[10:31] I believe 7 are up so far
[10:31] davetoo: zenhub is trying to push the changes (the new device) to the collectors
[10:31] I'm not very well prepared for this discussion, sorry; I'm at a remote site and just learned about this yesterday.
[10:31] http://www.zenoss.com/community/projects/zenpacks/
[10:32] yeah I need to post some of mine and quit hording them for myself
[10:32] Title: ZenPack Project Site - Commercial Open Source Network, Systems and Application Monitoring - Zenoss (at www.zenoss.com)
[10:32] umm... well when i re-modeled the device... it seems to have ignored my locks
[10:33] daMaestro: that's unfortunate
[10:33] new ZenPacks up so far are: ASSP Anti-Spam, Brocade Switches, Improved Linux Monitoring, MIB Browser, OpenLayers Map.
[10:33] * bootay (n=bootay@rrcs-97-77-9-2.sw.biz.rr.com) has joined #zenoss
[10:34] daMaestro: you can probably find the changesets that fix the issue if the need is urgent
[10:34] anyway what is your strategy for preventing zenhub from becoming a bottleneck?
[10:34] the only thing would be nice to have a link on the zenpacks page of stuff people are currently working-on or would like to see
[10:34] davetoo: you can have multiple zenhubs
[10:35] oh yes we have 8 on one box
[10:35] it helps alot
[10:35] mistich is a crazy man
[10:35] in the queue: updated Funkwerks, Alvarion, NetApp, Netasq, WMIFilesMonitor, EMC Celerra NAS, SQL Server
[10:35] * mistich has quit (Remote closed the connection)
[10:35] Is this similar to the old multiple monitor trick that Chet figured out?
[10:35] * mistich (n=mistich@nat-vlan0202.sat4.rackspace.com) has joined #zenoss
[10:35] is there a wiki page/faq entry?
[10:35] davetoo: probably
[10:35] oh yeah and Colubris, MRV and Raytalk
[10:36] davetoo: not that I know of... enterprise 2.2.3 will have a GUI to set it up
[10:36] nice
[10:37] ok...
[10:37] I have a script that we create multpile zenhubs
[10:37] are there any plans to ship an SSL-enabled Zope?
[10:37] ericnewton, http://fpaste.org/paste/3789 is what i get if i try to use the ethernetCsmacd template, which seems to be properly graphing interfaces
[10:37] we will have to try to talk to Gary again about that
[10:37] cccharles: not that I know of
[10:37] Title: Fedora Pastebin - Viewing paste #3789 (at fpaste.org)
[10:37] the last phone call didn't go so well :|
[10:37] modified from something zenoss sent me
[10:37] cccharles: most people just front the thing with Apache
[10:37] lol
[10:38] ericnewton: I've seen some discussion about that... it seems heavy when you have Zope already running.
[10:38] cccharles ccesario
[10:39] ericnewton: thanks for the tip, though. at least I know now :-).
[10:39] we front ours with apache works great
[10:40] daMaestro: perhaps this bug/and patches will help: http://dev.zenoss.org/trac/ticket/2960
[10:40] Title: #2960 (OS Process modeling/monitoring exhibiting strange behavior) - Zenoss - Trac (at dev.zenoss.org)
[10:40] that's for processes, not services
[10:40] so perhaps it is not fixed after all
[10:41] * wonderhamster_ (n=wonderha@adsl-68-78-3-5.dsl.ipltin.ameritech.net) has joined #zenoss
[10:41] ericnewton, looks like this will be in 2.2.3?
[10:42] daMaestro davetoo davewinter
[10:42] daMaestro: http://fpaste.org/paste/3789 from a system that has been upgraded from 2.0?
[10:42] ericnewton, we tried 2.2.0 and that liked to epic fail a lot
[10:42] Another topic: the loss of zendisc --routers functionality
[10:42] Title: Fedora Pastebin - Viewing paste #3789 (at fpaste.org)
[10:42] ericnewton, no, that is just 2.1.3
[10:42] had nobody been using that to bring up new systems?
[10:42] well, we're trying to cut down on Epic Fail... that's why you haven't seen 2.2.1 or 2.2.2
[10:43] yeah, 2.2.1 and 2.2.2 had upgrade issues, like you couldn't
[10:43] we kept those Fails to ourselves
[10:43] we plan on having a more vigorous beta program for 2.3
[10:43] good
[10:43] can you use 2.2.2 collectors to speak to a 2.1.3 master
[10:43] crosses fingers
[10:44] I'd like to be on the beta
[10:44] we'll be soliciting 2.3 beta testers in a controlled fashion, so not too many people blowing up their junk
[10:44] lol... too bad MS doesn't follow the "Don't release epic fails" philosophy sometimes...
[10:44] mistich: probably not a good idea
[10:44] mistich: it'll be for Enterprise and Core customers
[10:44] err... users
[10:44] * natepm has quit (Remote closed the connection)
[10:44] mistich: you gonna work double shifts?
[10:45] soon I'll have some test boxes and once there up I'd be glad to help not on my production systems
[10:45] * davetoo has quit ("Leaving.")
[10:45] darnit
[10:45] * davetoo (n=dlc@216.240.27.4) has joined #zenoss
[10:46] what data points should i be using to be able to utilized the data i can see in the os/ ethX perf reports?
[10:46] anyone have any luck with holt-winters?
[10:46] daMaestro: what are you trying to do?
[10:46] Is this even possible? If not, please say so... I'm trying to create a catch-all event for NTBackup errors. The event class key is in the form of NTBackup_#### and component is NTBackup.. Is there a way to create an event that will map to any of these events without creating an event for each possible error??
[10:47] ericnewton, add a report that graphs all our web nodes interface traffic
[10:49] daMastro: there's already a graph that sort of does that: /zport/dmd/Reports/Performance Reports/Aggregate Reports
[10:49] the last one: just click the graph and add a different pattern for the devices
[10:50] you might be able to trace how that graph is generated to create your own
[10:50] but to answer your question: ifInOctets and ifOutOctets
[10:51] if you are using the v1 32-bit counters
[10:51] has anyone done aggregated multi-graph reports?
[10:52] and ifHCInOctets and ifHCOutOctets if you are using the high-performance counter config
[10:53] * MAHerron (n=Administ@199.232.234.2) has joined #zenoss
[10:53] TBKdan: the defaultmapping event mapper runs if the event class key doesn't match: try modifying that
[10:54] ericnewton: Ok, I'll take a look. Thanks.
[10:55] Thanks to everybody for coming to #zenoss to chat for a bit, I'm gonna let Eric get back to fixing your tickets
[10:55] Just ran into a bug I believe... trying to view multiple pages of eventClassInstSequence and trying to view any page other than the default just brings it back to the event.
[10:56] Thanks guys/gals... having tickets entered with simple reproducible test cases helps a lot... and we do try to chase those down
[10:56] Thanks for your time
[10:56] I'll be chatting w/you all soon again!
[10:56] * ericnewton (n=ericnewt@74.93.194.82) has left #zenoss
[10:56] I'll try to get the transcript for this session put up somewhere soon
[10:56] in addition to our standard IRC logging
[10:56] and we'll try to do these with a little more frequency
[10:57] gah, even with ifHCInOctets and ifHCOutOctets the report graphs don't seem to be rendering :-/
[10:57] daMaestro davetoo davewinter
[10:57] * MajereDB8 has quit ("Leaving")
[10:57] it would be very nice if we could inspect more about what the shipped reports use to generate data
[10:57] * cccharles (n=user@strongbad.cs.uoguelph.ca) has left #zenoss ("ERC Version 5.3 (IRC client for Emacs)")
[10:58] hey is it possible to run an 'SNMP GET' from within a transform?
[10:58] such /Reports/Performance Reports/Interface Utilization shows data, but i'd like to be able to graph that data on a custom multi-graph report
[10:58] * mrayzenoss has changed the topic to: Zenoss is Open Source Systems Monitoring and Management
[10:58] * mrayzenoss has changed the topic to: Zenoss is Open Source Systems Monitoring and Management http://zenoss.com
[09:56] * Topic for #zenoss is: Zenoss developers will be here Tuesday, July 22 at 11am EST
[09:56] * Topic for #zenoss set by mrayzenoss at Tue Jul 15 17:16:55 2008
[09:56] you can set zCommandCycleTime in the zproperties for the device tree..
[09:56] i beleive that should do it.
[09:56] its defauletd to 60
[09:56] 2.1 or 2.2?
[09:56] 2.2
[09:57] * ericnewton (n=ericnewt@74.93.194.82) has joined #zenoss
[09:57] be careful doing 1 min polling cycle if you are pulling alot of oid your going to need a big box
[09:58] you can set it under /Monitors/Performance/localhost/editPerformanceConf
[09:58] 1 minute cycle eats a lot of disk IO if you have a high number of OIDS
[09:58] replace localhost with the collector name
[09:59] Hi everybody, Matt Ray from Zenoss here. Looks like discussions are already under way.
[10:00] * ian__ (n=__ian__@bas1-kitchener06-1096648702.dsl.bell.ca) has joined #zenoss
[10:00] There are a couple of Zenoss folks in the channel, they can introduce themselves if they like, or lurk anonymously
[10:00] Hello... this is Eric (not Erik) from Zenoss
[10:00] I'd prefer to lurk anonymously.
[10:01] wow... a lot more people than when I used to come here 2 years ago
[10:01] I will be the first to kiss a$$. Thanks Zenoss Devs for making such a great product!
[10:02] Now can you please do all my features requests?
[10:02] and mine too
[10:02] yes... +1 for all kiss-ass requests
[10:02] ericnewton would be happy to.
[10:02] so well have those next week
[10:02] so features... 2.2.3 is in the polishing stages
[10:03] in keeping with our normally timely schedules, yes of course
[10:03] good.
[10:03] sept for 2.2.3? or oct?
[10:03] * ckeyes (n=ckeyes@c-71-206-90-150.hsd1.fl.comcast.net) has joined #Zenoss
[10:03] not a lot of new functionality, lots of installer issues for this one
[10:03] i'm having to use 2.1.3 because of some bugs we have run into
[10:03] 2.2.3 is this week (hopefully). 2.3 is October
[10:03] ah
[10:03] it's just some TALES issues
[10:04] daMaestro davetoo davewinter
[10:04] * Shylock wishes for support of a debian based distro for Zenoss Enterprise
[10:04] * Johnny312 (n=Johnny31@c-98-206-10-126.hsd1.il.comcast.net) has joined #zenoss
[10:04] if you run Core, then you can probably pull a source release from the branch and be running 2.2.3 today
[10:04] daMaestro:have you opened tickets?
[10:04] mrayzenoss, yes
[10:04] daMaestro davetoo davewinter
[10:04] I have a question is there a place we can look at what is going to be in 2.3
[10:04] mrayzenoss, have not had time to send a patch
[10:04] daMaestro: have they been addressed?
[10:04] or any version
[10:04] mrayzenoss, dunno
[10:04] any upgrades to the docs with these upcomming releases?
[10:04] Shylock: have you tried the .deb stack installer?
[10:05] http://dev.zenoss.com/trac/ticket/3425
[10:05] is it true that there is no easy way to enable HTTPS in the web GUI?
[10:05] Title: #3425 (AttributeError when trying to render viewDevicePerformance) - Zenoss - Trac (at dev.zenoss.com)
[10:05] magnachef: minor documentation refreshes
[10:05] I can install core, but was informed that enterprise is not supported on Debian
[10:05] anyone running zenoss-enterprise with an offbox mysql? or mysql cluster?
[10:05] specifically.. our 64TB mount was reporting something like 5TB
[10:05] magnachef: dev doc stuff for adding tabs didn't make it in (2.2.4)
[10:05] davewinter: are you findin
[10:05] g mysql as a bottleneck?
[10:05] Shylock: this will be remedied with 2.2.3
[10:05] but there were many other graphing issues, which sucked
[10:06] Shylock: we have to make choices on how many platforms we can support officially. Debian may be covered by BitRock installers for Enterprise eventually
[10:06] ckrough, no, just need to have a common platform or other apps to use mysql. dont need a lot of mysql servers to manage.
[10:07] ericnewton: Ok... I just installed Cenos5 to get 2.2 up and going here, but I've been debian guy for the last 8 years or so... My redhat-fu has become weak.
[10:07] so, i'd really like to get interface monitoring working again
[10:07] we had a 32 million row history table that i had to deal with recently, so i need a more robust mysql solution rather than keeping it on the zneoss server
[10:07] daMaestro: do you know if netsnmp reports the correct value for your 64 TB mount?
[10:07] What is the long term direction for reporting? Will there be major changes in the way reports are created, or will you be sticking with the current method?
[10:07] ericnewton, it does, yet
[10:07] mistich: http://community.zenoss.com/forums/viewtopic.php?t=6651
[10:07] ericnewton, it also works correctly with 2.1.3
[10:07] Title: Forums :: View topic - The 2.3 Release Plan (at community.zenoss.com)
[10:08] what i'd really like help with is getting interface perf monitoring to work correctly
[10:08] yes I've seen the post but something more detailed
[10:08] daMaestro: please post a bug for the 64 TB mount and attach an SNMP walk of the device (edit out any bits you don't want public)
[10:08] hi, i've been testing the zenpop3 module for a while and have been having some trouble getting the program to generate events. pjp13 posted some fixes that have eliminated one of my problems (unable to parse the from: line), but i'm still getting a bunch of errors
[10:08] ericnewton, ok
[10:09] ckrough: I haven't seen much about changing reporting. I'd like to put out more documentation and examples so people can do more
[10:09] daMaestro: what issues are you having with interface graphs?
[10:09] i've documented my problems here - http://community.zenoss.com/forums/viewtopic.php?t=6677 - problem 2 is solved by pjp13's patch
[10:09] Throughput, Packets, Errors (via the perf tab) are not rendering with the shipped ethernetCsmacd_64 template
[10:09] yeah, maybe get a wiki going where people can post examples of their reports
[10:09] I also get:
[10:09] Title: Forums :: View topic - Zenpop3 errors - unable to generate e-mail event (at community.zenoss.com)
[10:10] Error reading value for "ifOutErrors" on server (oid .1.3.6.1.2.1.2.2.1.20 is bad)
[10:10] magnachef: We are down to 2 finalists for replacing our wiki and community infrastructure
[10:10] Error reading value for "ifHCInOctets" on server (oid .1.3.6.1.2.1.31.1.1.1.6 is bad)
[10:10] i get the "is bad" snmp issue sometimes as well.
[10:10] Error reading value for "ifHCOutOctets" on server (oid .1.3.6.1.2.1.31.1.1.1.10 is bad)
[10:10] etc
[10:10] MajereDB8: zenpop3 is unfinished... I expect it to take some elbo-grease to get it working
[10:10] magnachef: we're looking for something much easier to add content to and collaborate on
[10:11] So, is there a model diagram or documentation for all of the dmd classes? All of the zenoss classes inherit from such method heavy superclasses that introspection isn't very helpful in figuring out how to navigate the dmd.
[10:11] any chance ticket 363, acknowledging and clearing via smtp will make it in soon? I see that being bery useful
[10:11] I'm trying to create a catch-all event for NTBackup errors. The event class key is in the form of NTBackup_#### and component is NTBackup.. Is there a way to create an event that will map to any of these events without creating an event for each possible error??
[10:11] hi guys.
[10:11] what was the .bmp that you tried to dcc?
[10:12] i missed it..
[10:12] Shylock: we use a couple of techniques while working in zendmd
[10:12] I agree would be nice to have docs on all of the dmd classes
[10:12] http://fpaste.org/paste/3785 is what snmpwalk shows
[10:12] Shylock: we use "grepdir" to search for properties
[10:12] ericnewton: bummer. i understand how that goes. the zenackevents script posted by edahl is pretty slick and would work great in combination with zenpop3, but i guess all things come with time
[10:12] Title: Fedora Pastebin - Viewing paste #3785 (at fpaste.org)
[10:12] Shylock: I find it's easier to print obj.__dic__
[10:12] guys, is 2.2.1 out yet?
[10:12] jshadow: Nothing specific is being done on SMS/SMTP control that I'm aware of, but we are going to start cleaning out zenoss-3000 to get better traction on tickets
[10:12] so, obviously 1.3.6.1.2.1.31.1.1.1.10 is not bad
[10:13] mrayzenss: Ok, thanks
[10:13] jb: 2.2.1 and 2.2.2 didn't come out officially, 2.2.3 is about to come out
[10:13] daMaestro: is snmp version set to v2c on the device?
[10:13] mrayzenoss: can I expect a -CLEAN- upgrade from 2.2.0?
[10:13] ericnewton: Yeah... I generally load the rlcompleter tab completion stuff. I haven't used grepdir before, I'll try that.
[10:14] jb: we have spent most of the time/effort for 2.2.3 on upgrades
[10:14] awesome!
[10:14] jb: and testing
[10:14] my 2.1.2 -> 2.2 experience was horrific
[10:14] ericnewton, yes
[10:14] ericnewton, if i go to an individual interface.. the graph is properly generated
[10:14] so when can we expect 2.2.3? this week?
[10:15] jb: yes, if things keep going well
[10:15] ok
[10:15] ericnewton, i've just tried deleting lo and sit0 and then locked the ip interfaces to see if that gets rid of the snmp warnings and makes the "perf" tab interface graphs work
[10:15] 2.2.3 is in a RC right now
[10:15] so far, the perf tab graphs are still not rendering
[10:15] is it still the case that if you rename your device, you lose all of your RRD data associated with that device? Because I noticed that with 2.1.3
[10:15] I had read that there was an official BigIP zenpack, but I haven't seen any talk of it since the announcement..
[10:16] is this still in progress?
[10:16] On the note of the feature requests, it would be nice to have a voting system sort of like the ubuntu brainstorm
[10:16] daMaestro: you can run zenperfsnmp against one device in debug mode:
[10:16] jb: Enterprise has a BigIP ZenPack I believe
[10:16] ah.
[10:16] i have started one for core
[10:16] yes a voting system would be nice
[10:16] depends on how many votes I get
[10:16] i wonder what the enterprise pack does exactly
[10:16] daMastro: zenperfsnmp -v 10 -d myproblemdevice
[10:17] i also seem to have an issue with http and https ip service randomly stop monitoring
[10:17] jb: I'll see if we have a PDF
[10:17] one for you ckrough
[10:17] thanks
[10:17] daMaestro: I believe a bug was fixed: the services would toggle monitoring w/each modeling run
[10:18] * warwick (n=warwickp@ool-182ec7cf.dyn.optonline.net) has joined #zenoss
[10:18] deMaestro: fixed in 2.2.3
[10:18] * wonderhamster_ has quit (Read error: 104 (Connection reset by peer))
[10:18] hey guys, has there been any progress on documentation for making custom modelers?
[10:18] Will multi-graphing adding reports to zenpacks be fixed?
[10:18] ie, populating fields such as HW Make, OS Make, OS Version (and other custom fields maybe?) with data via snmp?
[10:18] * wonderhamster_ (n=wonderha@adsl-68-78-3-5.dsl.ipltin.ameritech.net) has joined #zenoss
[10:19] ptothedo: could you be more specific (and post a bug report if you haven't)?
[10:19] ericnewton, good news
[10:19] ericnewton, is there an upgrade path?
[10:20] jb: checking with the doc guy
[10:20] I'd like to talk about zenhub in 2.2. I'm seeing a lot of timeouts now.
[10:20] What changed?
[10:20] * daMaestro wonders if there are 2.2.3rc rpms somewhere?
[10:20] davetoo... in 2.2 we switched
[10:20] or even *gulp* a repo?
[10:20] I haven't created a bug report yet, but if I try to add sub report organizers to a zenpack they don't get add. Also, resequencing items in reporting isn't working for me.
[10:20] ;-/
[10:21] (sorry, I haven't looked at the changelists for it yet)
[10:21] davetoo: nearly every daemon to get data from zenhub
[10:21] http://community.zenoss.com/forums/viewtopic.php?t=3928&highlight=https <-- Is this really the only way to get the web GUI to support HTTPS?
[10:21] I think custom modelers are very important.. and we haven't seen any documentation/easy way to create them..
[10:21] daMaestro davetoo davewinter
[10:21] Title: Forums :: View topic - Modifying Zenoss for SSL encryption (at community.zenoss.com)
[10:21] davetoo: for simpler distributed collection
[10:21] ericnewton: I was speculating about that
[10:21] ericnewton, i assume that a workaround for now is to lock ip services?
[10:21] daMaestro: there is a yum repo, but it doesn't have rc rpms in it
[10:21] daMaestro: we haven't put together infrastructure for providing betas very well
[10:21] daMaestro davetoo davewinter
[10:21] mrayzenoss, ok
[10:21] daMaestro: yes for now
[10:22] daMaestro: betas tend to get out in the wild, we still see people running the pre-2.2. betas
[10:22] ian__, i'll look around and sync the tree into our cobbler infrastructure
[10:22] * cgibbons (n=cgibbons@75.148.215.110) has joined #zenoss
[10:22] davetoo: tell me a bit about your setup and where you are seeing timeouts
[10:23] jb: when we get the new wiki up, I plan on seeding discussion areas so people can get such stuff started if we don't provide it yet
[10:23] I think I have about 33,000 rrd files
[10:23] k
[10:23] * rdb (n=rdb@65.164.102.100) has joined #zenoss
[10:23] I'm looking for system specs right now...
[10:23] jb: Have you opened a ticket for it? That'll get it some love
[10:23] we are fighting zenperf and zenprocess timeouts as well
[10:23] davetoo: most of the collection via snmp?
[10:23] dual-Xeon 3G
[10:24] yeah theres a ticket
[10:24] yes
[10:24] eric dahl opened it I beileve..
[10:24] I'm trying to create a catch-all event for NTBackup errors. The event class key is in the form of NTBackup_#### and component is NTBackup.. Is there a way to create an event that will map to any of these events without creating an event for each possible error??
[10:24] jb: number? I'll get it to Doc and out of Erik's queue
[10:24] Error reading value for "eth0" on server (oid .1.3.6.1.2.1.2.2.1.10.2 is bad)
[10:24] Not much by commands, none by WMI
[10:24] whats the ticket URL again?
[10:24] trac
[10:24] blah, well after i removed the lo interface, i'm now getting these events
[10:24] davetoo: that's good... the timeouts: are you missing your cycle time?
[10:25] I began to notice the timeouts when adding new devices
[10:25] jb: dev.zenoss.com
[10:25] Not missing poll cycles currently
[10:25] eric: thanks
[10:26] * wonderhamster_ has quit (Read error: 60 (Operation timed out))
[10:26] davetoo: where do you see timeouts?
[10:26] http://fpaste.org/paste/3786 is what the shipped template requests, and i can snmpwalk those OIDs
[10:26] Title: Fedora Pastebin - Viewing paste #3786 (at fpaste.org)
[10:26] ericnewton: zendeviceloader timeouts
[10:27] http://pastebin.com/m686d5e5d
[10:27] I'm also surprised that I'm seeing zenstatus log messages in the zenhub logfile now....
[10:28] davetoo: that actually looks ok... have you investigated why those 10 devices have timedout?
[10:28] arg, I can't find the ticket number or the forum thread.
[10:29] davetoo: when you have bad devices zenpref retries the devices and it takes longer
[10:29] here it is
[10:29] mrayzenoss: http://dev.zenoss.org/trac/ticket/3292
[10:29] Title: #3292 (document modeler plugins) - Zenoss - Trac (at dev.zenoss.org)
[10:29] so it wasn't edahl
[10:29] if you remove or fix the bad devices your polling cycle will finish on time
[10:30] yes,
[10:30] Hey ericnewton, #3292 is in your box.
[10:30] but my problem is not currently with the snmp polling finishing on time
[10:30] mrayzenoss: oh good.. I was running out
[10:30]
[10:30] the problem is that while adding devices the connection to zenhub times out
[10:31] I get a message asking if zenhub is running
[10:31] speaking of numerous and interesting ZenPacks.... we've had 15 entries in the Submit a ZenPack contest
[10:31] oh, that I can believe
[10:31] from what you pasted all I see is the bad devices
[10:31] mrayzenoss: oh yeah? are they published anywhere?
[10:31] there's a big backlog to be published
[10:31] I believe 7 are up so far
[10:31] davetoo: zenhub is trying to push the changes (the new device) to the collectors
[10:31] I'm not very well prepared for this discussion, sorry; I'm at a remote site and just learned about this yesterday.
[10:31] http://www.zenoss.com/community/projects/zenpacks/
[10:32] yeah I need to post some of mine and quit hording them for myself
[10:32] Title: ZenPack Project Site - Commercial Open Source Network, Systems and Application Monitoring - Zenoss (at www.zenoss.com)
[10:32] umm... well when i re-modeled the device... it seems to have ignored my locks
[10:33] daMaestro: that's unfortunate
[10:33] new ZenPacks up so far are: ASSP Anti-Spam, Brocade Switches, Improved Linux Monitoring, MIB Browser, OpenLayers Map.
[10:33] * bootay (n=bootay@rrcs-97-77-9-2.sw.biz.rr.com) has joined #zenoss
[10:34] daMaestro: you can probably find the changesets that fix the issue if the need is urgent
[10:34] anyway what is your strategy for preventing zenhub from becoming a bottleneck?
[10:34] the only thing would be nice to have a link on the zenpacks page of stuff people are currently working-on or would like to see
[10:34] davetoo: you can have multiple zenhubs
[10:35] oh yes we have 8 on one box
[10:35] it helps alot
[10:35] mistich is a crazy man
[10:35] in the queue: updated Funkwerks, Alvarion, NetApp, Netasq, WMIFilesMonitor, EMC Celerra NAS, SQL Server
[10:35] * mistich has quit (Remote closed the connection)
[10:35] Is this similar to the old multiple monitor trick that Chet figured out?
[10:35] * mistich (n=mistich@nat-vlan0202.sat4.rackspace.com) has joined #zenoss
[10:35] is there a wiki page/faq entry?
[10:35] davetoo: probably
[10:35] oh yeah and Colubris, MRV and Raytalk
[10:36] davetoo: not that I know of... enterprise 2.2.3 will have a GUI to set it up
[10:36] nice
[10:37] ok...
[10:37] I have a script that we create multpile zenhubs
[10:37] are there any plans to ship an SSL-enabled Zope?
[10:37] ericnewton, http://fpaste.org/paste/3789 is what i get if i try to use the ethernetCsmacd template, which seems to be properly graphing interfaces
[10:37] we will have to try to talk to Gary again about that
[10:37] cccharles: not that I know of
[10:37] Title: Fedora Pastebin - Viewing paste #3789 (at fpaste.org)
[10:37] the last phone call didn't go so well :|
[10:37] modified from something zenoss sent me
[10:37] cccharles: most people just front the thing with Apache
[10:37] lol
[10:38] ericnewton: I've seen some discussion about that... it seems heavy when you have Zope already running.
[10:38] cccharles ccesario
[10:39] ericnewton: thanks for the tip, though. at least I know now :-).
[10:39] we front ours with apache works great
[10:40] daMaestro: perhaps this bug/and patches will help: http://dev.zenoss.org/trac/ticket/2960
[10:40] Title: #2960 (OS Process modeling/monitoring exhibiting strange behavior) - Zenoss - Trac (at dev.zenoss.org)
[10:40] that's for processes, not services
[10:40] so perhaps it is not fixed after all
[10:41] * wonderhamster_ (n=wonderha@adsl-68-78-3-5.dsl.ipltin.ameritech.net) has joined #zenoss
[10:41] ericnewton, looks like this will be in 2.2.3?
[10:42] daMaestro davetoo davewinter
[10:42] daMaestro: http://fpaste.org/paste/3789 from a system that has been upgraded from 2.0?
[10:42] ericnewton, we tried 2.2.0 and that liked to epic fail a lot
[10:42] Another topic: the loss of zendisc --routers functionality
[10:42] Title: Fedora Pastebin - Viewing paste #3789 (at fpaste.org)
[10:42] ericnewton, no, that is just 2.1.3
[10:42] had nobody been using that to bring up new systems?
[10:42] well, we're trying to cut down on Epic Fail... that's why you haven't seen 2.2.1 or 2.2.2
[10:43] yeah, 2.2.1 and 2.2.2 had upgrade issues, like you couldn't
[10:43] we kept those Fails to ourselves
[10:43] we plan on having a more vigorous beta program for 2.3
[10:43] good
[10:43] can you use 2.2.2 collectors to speak to a 2.1.3 master
[10:43] crosses fingers
[10:44] I'd like to be on the beta
[10:44] we'll be soliciting 2.3 beta testers in a controlled fashion, so not too many people blowing up their junk
[10:44] lol... too bad MS doesn't follow the "Don't release epic fails" philosophy sometimes...
[10:44] mistich: probably not a good idea
[10:44] mistich: it'll be for Enterprise and Core customers
[10:44] err... users
[10:44] * natepm has quit (Remote closed the connection)
[10:44] mistich: you gonna work double shifts?
[10:45] soon I'll have some test boxes and once there up I'd be glad to help not on my production systems
[10:45] * davetoo has quit ("Leaving.")
[10:45] darnit
[10:45] * davetoo (n=dlc@216.240.27.4) has joined #zenoss
[10:46] what data points should i be using to be able to utilized the data i can see in the os/ ethX perf reports?
[10:46] anyone have any luck with holt-winters?
[10:46] daMaestro: what are you trying to do?
[10:46] Is this even possible? If not, please say so... I'm trying to create a catch-all event for NTBackup errors. The event class key is in the form of NTBackup_#### and component is NTBackup.. Is there a way to create an event that will map to any of these events without creating an event for each possible error??
[10:47] ericnewton, add a report that graphs all our web nodes interface traffic
[10:49] daMastro: there's already a graph that sort of does that: /zport/dmd/Reports/Performance Reports/Aggregate Reports
[10:49] the last one: just click the graph and add a different pattern for the devices
[10:50] you might be able to trace how that graph is generated to create your own
[10:50] but to answer your question: ifInOctets and ifOutOctets
[10:51] if you are using the v1 32-bit counters
[10:51] has anyone done aggregated multi-graph reports?
[10:52] and ifHCInOctets and ifHCOutOctets if you are using the high-performance counter config
[10:53] * MAHerron (n=Administ@199.232.234.2) has joined #zenoss
[10:53] TBKdan: the defaultmapping event mapper runs if the event class key doesn't match: try modifying that
[10:54] ericnewton: Ok, I'll take a look. Thanks.
[10:55] Thanks to everybody for coming to #zenoss to chat for a bit, I'm gonna let Eric get back to fixing your tickets
[10:55] Just ran into a bug I believe... trying to view multiple pages of eventClassInstSequence and trying to view any page other than the default just brings it back to the event.
[10:56] Thanks guys/gals... having tickets entered with simple reproducible test cases helps a lot... and we do try to chase those down
[10:56] Thanks for your time
[10:56] I'll be chatting w/you all soon again!
[10:56] * ericnewton (n=ericnewt@74.93.194.82) has left #zenoss
[10:56] I'll try to get the transcript for this session put up somewhere soon
[10:56] in addition to our standard IRC logging
[10:56] and we'll try to do these with a little more frequency
[10:57] gah, even with ifHCInOctets and ifHCOutOctets the report graphs don't seem to be rendering :-/
[10:57] daMaestro davetoo davewinter
[10:57] * MajereDB8 has quit ("Leaving")
[10:57] it would be very nice if we could inspect more about what the shipped reports use to generate data
[10:57] * cccharles (n=user@strongbad.cs.uoguelph.ca) has left #zenoss ("ERC Version 5.3 (IRC client for Emacs)")
[10:58] hey is it possible to run an 'SNMP GET' from within a transform?
[10:58] such /Reports/Performance Reports/Interface Utilization shows data, but i'd like to be able to graph that data on a custom multi-graph report
[10:58] * mrayzenoss has changed the topic to: Zenoss is Open Source Systems Monitoring and Management
[10:58] * mrayzenoss has changed the topic to: Zenoss is Open Source Systems Monitoring and Management http://zenoss.com
