Skip navigation
Currently Being Moderated

IRC May 2011

Created on: Jun 2, 2011 12:55 PM by Nick Yeates - Last Modified:  Jun 2, 2011 1:00 PM by Nick Yeates

[01-May-2011 00:00:02] [connected at Sun May  1 00:00:02 2011]
[01-May-2011 00:00:20] [zenoss-logger (logger bot) has joined #zenoss]
[01-May-2011 10:51:30] <sloof3> Is there a way to mass un-monitor any interfaces that are "unrouted vlan XXX" ?
[01-May-2011 10:59:11] <Sam-I-Am> i usually use the properties to ignore interface name or type, then remodel.
[01-May-2011 10:59:32] <Sam-I-Am> you can use regular expressions for matching
[01-May-2011 12:00:01] [disconnected at Sun May  1 12:00:01 2011]
[01-May-2011 12:00:01] [connected at Sun May  1 12:00:01 2011]
[01-May-2011 12:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[01-May-2011 23:16:09] kocolosk is now known as kocolosk|away
[02-May-2011 00:00:01] [disconnected at Mon May  2 00:00:01 2011]
[02-May-2011 00:00:02] [connected at Mon May  2 00:00:02 2011]
[02-May-2011 00:00:21] [zenoss-logger (logger bot) has joined #zenoss]
[02-May-2011 05:45:48] fOrsberg is now known as forsberg
[02-May-2011 08:06:02] forsberg is now known as fOrsberg
[02-May-2011 09:27:59] kocolosk|away is now known as kocolosk
[02-May-2011 09:38:33] kocolosk is now known as kocolosk|away
[02-May-2011 09:49:19] kocolosk|away is now known as kocolosk
[02-May-2011 10:43:12] <davetoo>
[02-May-2011 10:44:23] <Sam-I-Am> me.too.
[02-May-2011 11:15:53] <Sam-I-Am> moo.
[02-May-2011 11:20:53] <joko_> boooooooo
[02-May-2011 12:00:01] [disconnected at Mon May  2 12:00:01 2011]
[02-May-2011 12:00:01] [connected at Mon May  2 12:00:01 2011]
[02-May-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[02-May-2011 13:25:22] * rmatte yawns
[02-May-2011 13:25:35] <rmatte> it's sure quiet in here today for a Monday
[02-May-2011 13:38:51] <jmp242> Anyone know why the fping zenpack would timeout when doing the same at the command line works?
[02-May-2011 14:09:42] <davetoo> socket permissions?
[02-May-2011 14:31:46] <JohnnyNoc> we're all shooting off guns celebrating the demise of OBL
[02-May-2011 14:32:06] <JohnnyNoc> at least those of us in America
[02-May-2011 14:32:07] <JohnnyNoc>
[02-May-2011 14:39:00] <jmp242> ??
[02-May-2011 14:39:54] kocolosk is now known as kocolosk|away
[02-May-2011 14:49:26] <macpablo> Hi, i have a problem with a ZenPack i got it installed wrong, when i try to remove it i got an error that it doesnt exists, and when i try to install again it says that es a zenpack with the same same, any idea how to remove it or fix it?
[02-May-2011 14:49:55] <rmatte> macpabola: can you that actual output when you try to install?
[02-May-2011 14:51:16] <macpablo>
[02-May-2011 14:56:49] <rmatte> oh no no, you don't do --remove against the name of the file
[02-May-2011 14:56:54] <rmatte> do a zenpack --list
[02-May-2011 14:57:00] <rmatte> you'll see the actual name of the pack
[02-May-2011 14:57:04] <rmatte> that's what you do the remove against
[02-May-2011 14:57:25] <rmatte> IE: zenpack --remove ZenPacks.zenoss.HttpMonitor
[02-May-2011 14:58:21] <rmatte> also, when it says that a ZenPack exists with the same name, that doesn't keep it from installing again
[02-May-2011 14:58:24] <rmatte> it's just informational
[02-May-2011 14:58:27] <rmatte> it still installs
[02-May-2011 14:58:51] <rmatte> also, if the pack is marked as broken in your ZenPack list after you install it, that just means that you need to restart Zenoss
[02-May-2011 14:58:57] <rmatte> it doesn't mean that the pack didn't install properly
[02-May-2011 15:02:52] <macpablo> i c so now i was able to remove and to isntall it again. but its failing to start 3 deamons
[02-May-2011 15:03:53] <macpablo> i get this when i try to start the zentrap service:
[02-May-2011 15:08:55] <rmatte> hmmm
[02-May-2011 15:09:15] <rmatte> try removing it then restart Zenoss and see if those daemons start
[02-May-2011 15:12:52] <macpablo> mmm i still get the same error after restarting and its not in the zenpack --list
[02-May-2011 15:13:26] <macpablo> oh wait the log is old...
[02-May-2011 15:16:39] <macpablo> its not logging anything now
[02-May-2011 15:16:59] <rmatte> all you should care about is whether or not the daemons are actually in a running state
[02-May-2011 15:17:10] <rmatte> are they?
[02-May-2011 15:17:14] <macpablo> yes the status is stopped
[02-May-2011 15:17:25] <rmatte> what happens when you try to start them?
[02-May-2011 15:17:33] <macpablo> for zentrap, zenping and zensyslog
[02-May-2011 15:17:35] <macpablo> they wont start
[02-May-2011 15:17:40] <rmatte> try this...
[02-May-2011 15:17:43] <rmatte> zenping run -v10
[02-May-2011 15:17:49] <rmatte> fpaste the output
[02-May-2011 15:18:19] <macpablo> ZenSocket Error: zensocket needs to be run as root or setuid
[02-May-2011 15:19:49] <rmatte> It sounds like you messed up your file permissions
[02-May-2011 15:19:54] <rmatte> did you play around with them?
[02-May-2011 15:20:31] <rmatte> if you did a chown -R zenoss or something on the Zenoss directory then that would cause that
[02-May-2011 15:20:58] <macpablo> i did it but to fix it i did exactly that and broke mysql
[02-May-2011 15:21:17] <rmatte> yeh, don't touch those permissions, they are more complex than you think...
[02-May-2011 15:21:31] <rmatte>
[02-May-2011 15:21:33] <rmatte> do that as root
[02-May-2011 15:21:35] <rmatte> and it'll fix it
[02-May-2011 15:22:24] <rmatte> change the paths if it's in /opt/zenoss for you
[02-May-2011 15:23:09] <rmatte> ummm wow
[02-May-2011 15:23:14] <rmatte> you have it installed to your home directory?
[02-May-2011 15:23:25] <rmatte> did you do that with the stack installer?
[02-May-2011 15:23:49] <macpablo> i installed it in my home dir from the stack installer
[02-May-2011 15:23:52] <rmatte> (first time I've ever seen someone do that, the two usual spots are either /usr/local/zenoss or /opt/zenoss
[02-May-2011 15:23:55] <rmatte> ok...
[02-May-2011 15:24:05] <macpablo> the thing is that is running inside a linux container
[02-May-2011 15:24:22] <macpablo> so i dont have read access to the /usr dir
[02-May-2011 15:24:31] <macpablo> because its shared with the host
[02-May-2011 15:24:50] <macpablo> great! now its working
[02-May-2011 15:24:57] <macpablo> but i screwed the permissions later
[02-May-2011 15:24:59] <rmatte> This is what you want
[02-May-2011 15:24:59] <rmatte>
[02-May-2011 15:25:03] <rmatte> ah ok, you already got it
[02-May-2011 15:25:38] <rmatte> so now just install the pack, restart Zenoss, and you should be set
[02-May-2011 15:25:52] <macpablo> yes its working now
[02-May-2011 15:26:16] <macpablo> i have a lot of questions, i used the 2.1 version a lot. this is very different
[02-May-2011 15:26:41] <jmp242> Hey, a little OT: what do people use in the free / open source tools for monitoring netflow?
[02-May-2011 15:26:43] <rmatte> yeh, it is
[02-May-2011 15:26:51] <jmp242> and sflow
[02-May-2011 15:27:13] <rmatte> jmp242: the only one that I've found that even remotely works well is ntop
[02-May-2011 15:27:28] <rmatte> but it's unstable as all hell (crashed a lot when I was using it, and hogged resources)
[02-May-2011 15:27:35] <rmatte> we ended up buying software to do it
[02-May-2011 15:27:41] <macpablo> how is the network map built? because i manually added a route in my firewalls because i have vpn tunnels and they dont show linked in the map
[02-May-2011 15:28:01] <rmatte> macpablo: the google map or the flash network map?
[02-May-2011 15:28:13] <macpablo> flash network map
[02-May-2011 15:28:19] <rmatte> the flash network map is borderline useless
[02-May-2011 15:28:26] <rmatte> it connects stuff based on discovered routes
[02-May-2011 15:28:29] <rmatte> but it's pretty horrible
[02-May-2011 15:28:35] <rmatte> I never even look at it these days
[02-May-2011 15:29:24] <rmatte> They'll probably come out with some replacement for it eventually, but it'll probably be made an enterprise only feature since that seems to be the trend lately lol
[02-May-2011 15:29:26] <macpablo> yes i know but i want the dependencies to work properly. i stopped using the old version because of that
[02-May-2011 15:29:50] <rmatte> the dependencies are still screwy, they are no better than they were back in 2.1 as far as I'm aware
[02-May-2011 15:30:00] <rmatte> I've never been able to get them working properly
[02-May-2011 15:30:05] <jmp242> rmatte: what software did you end up getting
[02-May-2011 15:30:11] <rmatte> I wish they'd just add the ability for manual dependencies
[02-May-2011 15:30:35] <rmatte> jmp242: ManageEngine Netflow Analyzer
[02-May-2011 16:00:47] <froztbyte> rmatte: in about 3 weeks I might be able to do some actual dev
[02-May-2011 16:00:58] <froztbyte> my minion is finally starting tomorrow! \o/
[02-May-2011 16:01:15] <froztbyte> and the first two things I want to add are L2 and L3 discovery/mapping/dependency code
[02-May-2011 16:35:14] <rmatte> lol nice
[02-May-2011 16:41:41] <rmatte> alright guys, I'm done for the day, ttyl
[02-May-2011 16:43:19] <macpablo> bye rmatte and thanks!
[02-May-2011 16:45:24] [disconnected at Mon May  2 16:45:24 2011]
[03-May-2011 00:00:02] [connected at Tue May  3 00:00:02 2011]
[03-May-2011 00:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[03-May-2011 01:04:06] salahudin_mycert is now known as slawank
[03-May-2011 01:04:10] slawank is now known as salawank
[03-May-2011 03:52:23] <Rickson-> i tried to install a zenpack for zenoss 2.2 in 3.0 (juniper j series pack) i get this error is this any basic way to get older zenpacks to work?
[03-May-2011 07:18:45] <kokey> what's a good way to do SOAP based checks with zenoss?
[03-May-2011 07:25:40] <froztbyte> I put together a zencommand with suds sometime last year
[03-May-2011 08:38:57] kocolosk is now known as kocolosk|away
[03-May-2011 09:06:56] kocolosk|away is now known as kocolosk
[03-May-2011 09:26:10] <Sam-I-Am> .
[03-May-2011 09:26:57] <Simon4> moo
[03-May-2011 09:28:40] <Sam-I-Am> yup
[03-May-2011 09:55:21] <JohnnyNoc> morning guys
[03-May-2011 09:55:28] <JohnnyNoc> i've got a question about alert emails
[03-May-2011 09:55:43] <JohnnyNoc> i've set up an alert to look for interface errors/drops, and can't for the life of me determine why these alerts are being twice
[03-May-2011 09:55:53] <JohnnyNoc> was hoping someone here may have an idea?
[03-May-2011 09:56:12] <JohnnyNoc> meaning, duplicates of an alert are sent..  same alert, same time, etc.
[03-May-2011 09:56:56] <Sam-I-Am> are there duplicate rules?
[03-May-2011 09:56:59] <Simon4> two alert rules, one to a group, one to a user?
[03-May-2011 09:57:27] <Sam-I-Am> alternatively, check the log for zenactions and/or restart zenactions.
[03-May-2011 09:57:34] <JohnnyNoc> hmm, no i don't think so.  i went in and renamed some of the rules to be more specific regarding what user gets them
[03-May-2011 09:57:38] <Sam-I-Am> i had a case a few weeks ago where significant changes to alerting rules confused zenactions
[03-May-2011 09:57:48] <JohnnyNoc> i'm tailing zenactions and thats whre i see the dupes
[03-May-2011 09:58:01] <Sam-I-Am> if you renamed via zope you should restart zenactions
[03-May-2011 09:58:14] <JohnnyNoc> hrm, no id idn't
[03-May-2011 09:58:24] <JohnnyNoc> but i had this problem yesterday and thought bouncing zenoss last night would fix the problem
[03-May-2011 09:58:28] <Sam-I-Am> in 2.5.2 i dont think theres a normal way to rename alerting rules
[03-May-2011 09:58:37] <JohnnyNoc> no there isn't
[03-May-2011 09:58:38] <Sam-I-Am> bouncing would restart zenactions
[03-May-2011 09:58:40] <JohnnyNoc> i removed and re-added
[03-May-2011 09:59:00] <JohnnyNoc> because i had rules like "interface errors" for multiple users, i made it more specific like "noc interface errors"
[03-May-2011 09:59:32] <JohnnyNoc> i thought maybe there were 2 zenactions running
[03-May-2011 10:00:15] <JohnnyNoc> NOC Interface Errors0TrueNow0 days 21:43:28Every Weekday
[03-May-2011 10:00:24] <JohnnyNoc> (prodState = 1000) and (eventState = 0) and (severity >= 3) and (summary like '%interface drops%' or summary like '%interface errors%')
[03-May-2011 10:00:27] <JohnnyNoc> thats the alert rule
[03-May-2011 10:00:58] <JohnnyNoc> and there's only 1, and in zenactions log I see NOC Interface errors twice, one sent a little more than a second after the first
[03-May-2011 10:02:10] <jmp242> I don't suppose you have two events?
[03-May-2011 10:02:46] <Simon4> JohnnyNoc: if you're seeing the errors twice, you need to debug further upstream
[03-May-2011 10:02:49] <JohnnyNoc> I suppose that's possible, but the values in the event are the same
[03-May-2011 10:02:56] <Simon4> how are the alerts being sent to zenoss?
[03-May-2011 10:03:26] <JohnnyNoc> not sure exactly, but I've setup a threshold for interface drops/errors
[03-May-2011 10:04:34] <JohnnyNoc> ah
[03-May-2011 10:04:36] <JohnnyNoc> think i found the problem
[03-May-2011 10:04:37] <JohnnyNoc>
[03-May-2011 10:04:48] <JohnnyNoc> i've got 2 thresholds setup..  1 for interface drops and 1 for interface errors
[03-May-2011 10:05:03] <JohnnyNoc> but in my interface drops threshold i can see one of my datapoints are ifinerrors
[03-May-2011 10:05:06] <JohnnyNoc> i think it must be related
[03-May-2011 10:05:35] <Simon4> sounds like both thresholds are being triggered, which will def send two events
[03-May-2011 10:05:53] <JohnnyNoc> yea, that's what i'm thinking at the moment
[03-May-2011 10:08:09] * JohnnyNoc taps his fingers waiting for more alerts
[03-May-2011 10:21:03] <_tete_> re
[03-May-2011 10:21:19] <_tete_> The Topic URL gives a 404
[03-May-2011 10:22:27] <JohnnyNoc> works fine
[03-May-2011 10:22:33] <JohnnyNoc> are you including the trailing " ?
[03-May-2011 10:22:34] <JohnnyNoc>
[03-May-2011 10:35:08] <_tete_> that's true.... sorry and thanks
[03-May-2011 10:35:09] <_tete_>
[03-May-2011 11:18:54] <JohnnyNoc> i'm an idiot
[03-May-2011 11:19:02] <JohnnyNoc> those alerts seem doubled up because they're for different interfaces
[03-May-2011 11:19:08] <JohnnyNoc> ie, bond0 and eth0
[03-May-2011 11:19:36] <Simon4> ahh
[03-May-2011 11:21:04] <Sam-I-Am> hehe
[03-May-2011 11:21:15] <JohnnyNoc> would help if i read the damn alerts
[03-May-2011 11:21:15] <JohnnyNoc>
[03-May-2011 11:21:17] <JohnnyNoc> but that being said
[03-May-2011 11:21:37] <JohnnyNoc> seems like i should probably create a transform to suppress the eth0 event if there's a bond interface complaining
[03-May-2011 11:22:56] <Simon4> I would suppress the bond drops/etc alerts
[03-May-2011 11:23:03] <Simon4> they're more useful on the phys interface anyway
[03-May-2011 11:23:11] <JohnnyNoc> good point
[03-May-2011 11:24:25] <Simon4> lame, Linux "bond" interfaces don't report as a different interface type
[03-May-2011 11:24:43] <Simon4> in ciscoland the portchannel interfaces report as "link aggregation xxxxx" so yo ucan create a separate template just for them
[03-May-2011 11:37:35] <JohnnyNoc> fortunately it's not a big deal
[03-May-2011 12:00:01] [disconnected at Tue May  3 12:00:01 2011]
[03-May-2011 12:00:01] [connected at Tue May  3 12:00:01 2011]
[03-May-2011 12:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[03-May-2011 12:51:01] <citrusfizz> whats the default ssh login for the zenoss virtual machine?
[03-May-2011 12:53:16] <citrusfizz> nevermind
[03-May-2011 13:13:15] <jmp242> whooohoooo predictive threshold zenpack seems working for 3.x...
[03-May-2011 14:38:52] <moath> anybody set up thresholds on ZenJMX stuff before?  I'm having a hard time referencing the variables ( or whatever) and would be interested in an example if anyone has one they can share
[03-May-2011 15:06:26] <joko_> hmm
[03-May-2011 15:07:22] <joko_> why can't i find out how to change a devices group membership in this new 3.1 UI!
[03-May-2011 15:08:56] <jmp242> It's probably drag and drop , but you have to click on the white space to the right of the device name
[03-May-2011 15:12:02] <Hackman238> Can't say I like v3 interface very much myself
[03-May-2011 15:14:15] <joko_> hmm
[03-May-2011 15:20:06] <jmp242> mmm Hackman238 it's starting to grow on me a bit
[03-May-2011 15:20:14] <jmp242> but it is far less discoverable than the v2 one was
[03-May-2011 15:20:23] <joko_> yeah i dont mind it, but i seriously still can't find out how to change the group or system
[03-May-2011 15:20:45] <joko_> maybe my perms..
[03-May-2011 15:21:41] <joko_> no...
[03-May-2011 15:25:54] <jmp242> I just figured it out?
[03-May-2011 15:26:03] <jmp242> you just drag the item to the group you want it in.
[03-May-2011 15:26:06] <joko_> enlighten me
[03-May-2011 15:26:10] <joko_> oh snazzy
[03-May-2011 15:26:12] <jmp242> then you need to go back to the old group, say delete
[03-May-2011 15:26:13] <joko_> mac style
[03-May-2011 15:26:20] <jmp242> and say, just remove from "groupname"
[03-May-2011 15:26:33] <joko_> thats not bad
[03-May-2011 15:26:51] <jmp242> The trick is not to click the underlined name
[03-May-2011 15:27:04] <jmp242> but grab the whitespace to the right of it in the row
[03-May-2011 15:28:09] <joko_> yeah, i didn't realize you could drag that stuff
[03-May-2011 15:28:12] <joko_> ty
[03-May-2011 15:28:17] <joko_> that was such a dumb problem
[03-May-2011 15:53:08] <moath> is there a way to get names of the variables exposed through a monitoring template?
[03-May-2011 15:53:29] <moath> so if i've got a zenjmx template that monitors stuff how can i get the variable names i'll need to reference within a threshold?
[03-May-2011 17:30:09] kocolosk is now known as kocolosk|away
[03-May-2011 20:03:17] <moath> anybody using zenoss with snmp traps?  can't seem to get zentrap to listen for network traffic (it doesn't show up in netstat)
[03-May-2011 20:17:23] kocolosk|away is now known as kocolosk
[03-May-2011 20:24:36] kocolosk is now known as kocolosk|away
[03-May-2011 20:24:43] kocolosk|away is now known as kocolosk
[03-May-2011 21:37:37] <wjimenez5271> hello
[04-May-2011 00:00:01] [disconnected at Wed May  4 00:00:01 2011]
[04-May-2011 00:00:02] [connected at Wed May  4 00:00:02 2011]
[04-May-2011 00:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[04-May-2011 03:49:05] <Milad> loading data from zenoss's collector  take forever, I check tcpdump and main and collector  pass some data but loading json request (http://monitoring:8080/zport/dmd/Devices/Server/Linux/devices/collector.monitoring/device_router) take forever
[04-May-2011 03:49:19] <Milad> anyone has clue ?
[04-May-2011 08:48:56] <Gat0rvean> can anyone explain to me how often the Modeler Plugins run?
[04-May-2011 08:49:08] <kokey> once every 6 hours i think
[04-May-2011 08:49:54] <Gat0rvean> Kokey: do you know if there is any reason to have the NewDeviceMap plugin run any more than the first time?
[04-May-2011 08:50:42] <Gat0rvean> My issue is, that randomly Zenoss will cause one of my network devices to have very high CPU usage, and I need to figure out why because it's throwing false positives because of this
[04-May-2011 08:51:09] <froztbyte> "randomly"?
[04-May-2011 08:51:14] <froztbyte> it's all logged and such
[04-May-2011 08:51:17] <froztbyte> no black magic
[04-May-2011 08:51:19] <froztbyte> you can go find out
[04-May-2011 08:52:18] <kokey> Gat0rvean: many reasons, one could be that an interface speed can change in between, or the interface label, etc. etc.
[04-May-2011 08:52:29] <Gat0rvean> well, I guess what I mean by "randomly" is I have 150 or so of these same blades, and only 1 or 2 of them go off, but it's random as to which ones throw errors
[04-May-2011 08:52:38] <kokey> Gat0rvean: sounds like you've done some pretty dodgy equipment if some snmp walks can hammer the CPU
[04-May-2011 08:52:50] <kokey> done=got
[04-May-2011 08:53:42] <kokey> I have seen it happen before tho, on dodgy equipment
[04-May-2011 08:54:14] <Gat0rvean> Well, they are Occam BLC's, when they get walked with the default zenoss plugins, they will return HUNDREDS of snmp statistics from POTS lines DSL lines, Gig Ethernet ports etc
[04-May-2011 08:54:35] <Gat0rvean> and it's not necessarily "hammering" the CPU, but raising the average enough to be annoying
[04-May-2011 08:54:40] <kokey> Gat0rvean: ah, sounds like their smtp implementation doesn't cache like most others
[04-May-2011 08:54:53] <froztbyte> kokey: pull a full internet table from any cisco, that also gives it a snotklap
[04-May-2011 08:54:56] <kokey> normally you can walk an entire tree in under 10 seconds
[04-May-2011 08:55:00] <froztbyte> (via SNMP)
[04-May-2011 08:55:47] <kokey> Gat0rvean: from my experience the CPU should only be high during the walk time, and that should complete in well under a minute, depending on the device
[04-May-2011 08:56:08] <kokey> Gat0rvean: and your snmp polls for CPU usage should not really see it show up in the average
[04-May-2011 08:56:13] <kokey> but yeah, depends on the equipment
[04-May-2011 08:56:46] <Gat0rvean> kokey: ok, thanks for your help
[04-May-2011 08:58:58] <kokey> Gat0rvean: you can change the modeler cycle interval, mine's default was every 12 hours
[04-May-2011 08:59:06] <kokey> Gat0rvean: but it's a global setting
[04-May-2011 09:00:26] <Gat0rvean> kokey:  That would still cause it to throw the errors, I need to just figure out a way to edit the modeler to only poll the Gig E interfaces so it's not getting back hundreds upon hundreds of DSL and POTS statistics
[04-May-2011 09:02:42] <froztbyte> you don't need to edit the modeler, there are zProperties for that
[04-May-2011 09:02:47] <froztbyte> read the documentation to see what to do
[04-May-2011 09:04:55] <sergeymasushko> from time to time zenoss sends alerts that <ipservice ...> is down for several devices and almost immediately sends recovery, what can be the issue?
[04-May-2011 09:04:58] <Gat0rvean> froztbyte:  I've added the interfaces I want it _not_ to poll to zInterfaceMapIgnoreTypes, is there another zProperty that would be applicable in this situation?
[04-May-2011 09:05:53] <kokey> sergeymasushko: they don't have anything in common, e.g. a location or a switch?
[04-May-2011 09:07:10] <sergeymasushko> no... all the devices in the same data center and the alerts are related only to ip services... not ping or snmp... it happens once in a day... don't know why...
[04-May-2011 09:19:01] <kokey> sergeymasushko: the same devices each time or roughly different ones every time?
[04-May-2011 09:19:27] <Sam-I-Am> moo.
[04-May-2011 09:22:59] <sergeymasushko> different devices, different ip services...
[04-May-2011 09:24:01] kocolosk is now known as kocolosk|away
[04-May-2011 09:38:13] kocolosk|away is now known as kocolosk
[04-May-2011 09:54:07] <kokey> blah
[04-May-2011 09:54:25] <kokey> i need to find a job where i can do a greenfields zenoss implementation
[04-May-2011 09:54:41] <kokey> instead of legacy crap migration along with the guy who made the legacy stuff
[04-May-2011 09:56:25] <Sashness> kokey: LOL
[04-May-2011 10:55:04] <joko_> stupid
[04-May-2011 11:30:14] <davetoo> is as stupid does
[04-May-2011 12:00:01] [disconnected at Wed May  4 12:00:01 2011]
[04-May-2011 12:00:02] [connected at Wed May  4 12:00:02 2011]
[04-May-2011 12:00:21] [zenoss-logger (logger bot) has joined #zenoss]
[04-May-2011 12:01:11] <Hackman238> Any zenoss employees in right now?
[04-May-2011 13:25:18] <joko_> zenoss 3.1 gives me a headache somedays
[04-May-2011 13:28:30] <Sam-I-Am> heh
[04-May-2011 13:28:39] <Sam-I-Am> 2.5.2 is headache-free lol
[04-May-2011 13:28:41] <Sam-I-Am> (not)
[04-May-2011 13:28:56] <joko_> well, just getting use to the UI for things that were simple before
[04-May-2011 13:30:07] <joko_> simply want to move a local template into a different location
[04-May-2011 13:37:23] <joko_> looks like rmattes zope interface trick will work
[04-May-2011 13:37:28] <joko_> altho how dumb
[04-May-2011 14:39:08] * joko_ wonders
[04-May-2011 15:26:15] <St3v3o> Is there a Datacenter view in the core version of zenoss or this is something strictly with the enterprise version
[04-May-2011 15:26:43] <joko_> unless someone hacked together something, pretty sure it's only enterprise
[04-May-2011 15:26:53] <St3v3o> bummer...
[04-May-2011 15:27:32] <St3v3o> currently we use a single file to document our racks and power…we have been looking at some additional tools…anyone have anything cool they use ?
[04-May-2011 15:28:02] <St3v3o> I've gotten a lot of suggestions for just visio and or excel…but looking for something that might be a little more robust
[04-May-2011 15:30:36] <joko_> yeah no idea here, we wanted to use the DC view, but enterprise only killed it
[04-May-2011 15:30:41] <joko_> any now it's on the back burner
[04-May-2011 17:32:15] <joko_> HMMMMMMMMMMmmmm
[04-May-2011 17:34:01] kocolosk is now known as kocolosk|away
[04-May-2011 19:46:40] kocolosk|away is now known as kocolosk
[04-May-2011 22:33:41] kocolosk is now known as kocolosk|away
[05-May-2011 00:00:01] [disconnected at Thu May  5 00:00:01 2011]
[05-May-2011 00:00:02] [connected at Thu May  5 00:00:02 2011]
[05-May-2011 00:00:17] [zenoss-logger (logger bot) has joined #zenoss]
[05-May-2011 04:44:05] <guest_user156> just testing the irc
[05-May-2011 06:06:03] bzed_ is now known as bzed
[05-May-2011 08:42:28] <jmp242> So I'm trying to set up a percentage threshold on memory use on Windows servers using the WMIPerf zenpack
[05-May-2011 08:43:20] <jmp242> is there something analogous to the here.totalBlocks for filesystem use?
[05-May-2011 08:43:32] <jmp242> that applies to total memory available?
[05-May-2011 09:28:44] kocolosk|away is now known as kocolosk
[05-May-2011 10:06:53] <joko_> hmm
[05-May-2011 10:07:34] <joko_> can i aggregate interfaces on a report? i assume so... our lacp trunks would be nice to aggregate
[05-May-2011 10:31:45] <Sam-I-Am> mooooo
[05-May-2011 10:31:56] <Hackman238> Sam-I-Am: ?
[05-May-2011 10:33:43] <Sam-I-Am> cows are trendy
[05-May-2011 10:36:08] <jmp242> So I'm trying to set up a percentage threshold on memory use on Windows servers using the WMIPerf zenpack
[05-May-2011 10:36:08] <jmp242> 08:43    jmp242    is there something analogous to the here.totalBlocks for filesystem use?
[05-May-2011 10:36:09] <jmp242>     jmp242    that applies to total memory available?
[05-May-2011 10:42:24] <Sam-I-Am> probably
[05-May-2011 10:42:32] <Sam-I-Am> just need to find the variable :/
[05-May-2011 10:43:33] <jmp242> any idea how to find the variable? I'm not at all sure how to do that.
[05-May-2011 10:44:52] <Sam-I-Am> should be in the code for the zenpack
[05-May-2011 10:44:56] <Sam-I-Am> at least thats my guess
[05-May-2011 10:45:12] <jmp242> ahh
[05-May-2011 10:45:15] <jmp242> ok
[05-May-2011 10:51:43] <jmp242> I'm going to post a forum thread to see if Egor responds
[05-May-2011 10:57:56] <joko_> any cool portlets people use? seeing if im missing any goodies
[05-May-2011 12:00:01] [disconnected at Thu May  5 12:00:01 2011]
[05-May-2011 12:00:02] [connected at Thu May  5 12:00:02 2011]
[05-May-2011 12:00:20] [zenoss-logger (logger bot) has joined #zenoss]
[05-May-2011 12:10:03] <kotique> hi. how do I get a list of devices that match certain criteria via zendmd?
[05-May-2011 12:10:20] <kotique> i've found _findDevice(self, devicename, useTitle=True), but want to get by group/system, not by title
[05-May-2011 12:30:31] <kotique> dmd.Groups['Shared servers']
[05-May-2011 12:32:20] <salax> no idea
[05-May-2011 12:34:51] <moath> anybody set up thresholds on ZenJMX stuff before?  I'm having a hard time referencing the variables ( or whatever) and would be interested in an example if anyone has one they can share
[05-May-2011 12:56:12] <twm1010> can you upgrade from 2.5.2 directly to 3.1?
[05-May-2011 12:57:45] <jmp242> Check the upgrade docs twm1010
[05-May-2011 12:59:14] <twm1010> it seems out of date, no reference to 3.1 in dependencies
[05-May-2011 12:59:29] <twm1010> i see 2.5.2 to 3.0.3 looks good
[05-May-2011 12:59:50] <Sam-I-Am> no you cant
[05-May-2011 12:59:51] <Sam-I-Am> best i can tell
[05-May-2011 13:00:04] <Sam-I-Am> 3.1 is bug fixes for 3, so you do the upgrade to 3, then 3.1
[05-May-2011 13:00:40] <twm1010> k, thx
[05-May-2011 13:00:43] <jmp242> The upgrade to 3.0.3 is a huge one anywa
[05-May-2011 13:00:44] <jmp242> uy
[05-May-2011 13:00:56] <jmp242> and you have to be careful in doing it - what with pre-upgrade zenpack etc
[05-May-2011 13:01:13] <jmp242> make sure sure sure you have working backups of the entire server before attempting
[05-May-2011 13:01:17] <jmp242> IMO anyway
[05-May-2011 13:02:07] <twm1010>
[05-May-2011 13:02:24] <Sam-I-Am> if you look carefully at the docs you'll see redactions of text involving sacrifice of animals
[05-May-2011 13:03:52] <twm1010> i noticed that, was havint trouble locating a guide for 2.41 - 2.5.2
[05-May-2011 13:04:06] <Sam-I-Am> i would probably avoid 2.4->2.5->3
[05-May-2011 13:05:04] <twm1010> in the thick of it already
[05-May-2011 13:05:21] <twm1010> 2.5.2 seems to be behaving, some of the portlet content not showing up
[05-May-2011 13:05:23] <Sam-I-Am> pause and grab yourself a case
[05-May-2011 13:06:23] <twm1010> vmware snapshots are nice
[05-May-2011 13:06:38] <twm1010> decided just to start at v3.1
[05-May-2011 13:06:43] <twm1010> recreate stuff
[05-May-2011 13:06:47] <twm1010> thx guys
[05-May-2011 14:56:58] <joko> interesting
[05-May-2011 14:57:04] <joko> "You can configure IP SLA on the BGP speaking router for route reachability out of that link and along with EEM automatically shutdown the interface so BGP can failover to the other MPLS provider."
[05-May-2011 15:01:25] <St3v3o> afternoon al
[05-May-2011 15:01:51] <Rickson-> is it possible to customize the network map, like change backgroundcolour?
[05-May-2011 15:03:47] <Sam-I-Am> dont think so
[05-May-2011 15:03:54] <Sam-I-Am> well, you can do whatever, but it might take some code
[05-May-2011 15:07:19] <Rickson-> yeah ofcourse
[05-May-2011 15:14:59] <davetoo> I thought somebody had made an OpenStreetMap zenpack some time ago?
[05-May-2011 15:15:20] <davetoo> Which looks a little different but more interestingly is probably more hackable in that respect.
[05-May-2011 15:18:18] <St3v3o> so I'm trying to build a brocade ADX template…but can't seem to figure out how to graph the connections for VIPS (Virtual Server).  The OID give a table listing
[05-May-2011 15:18:36] <St3v3o> has anyone come across something like this before
[05-May-2011 15:34:38] <joko> hmm
[05-May-2011 15:35:42] <joko> i use to use a zenpack that would ping a host, in my case i was using it to ping external connections, and it would chart the RTT for me
[05-May-2011 15:36:03] <joko> anyone know what thats called or something similar?
[05-May-2011 15:36:20] <St3v3o> fping ?
[05-May-2011 15:37:44] <St3v3o>
[05-May-2011 15:38:20] <joko> nice
[05-May-2011 15:38:24] <joko> that looks better than what i was using
[05-May-2011 15:38:25] <joko> ty
[05-May-2011 15:38:42] <St3v3o> np I use it for "meeting" routers…works great
[05-May-2011 15:40:09] <joko> cool stuff
[05-May-2011 15:41:59] joko is now known as joko_
[05-May-2011 16:11:26] <otakup0pe> ugh to zenoss reports really not auto refresh
[05-May-2011 17:35:11] kocolosk is now known as kocolosk|away
[05-May-2011 18:08:06] <ericenns> hey
[05-May-2011 18:09:27] <ericenns> anybody here done some gui dev in 2.5
[05-May-2011 18:11:35] <citrusfizz> how i can i start/stop zen daemons from the command line?
[05-May-2011 18:11:55] <ericenns> zenoss restart (will restasrt all daemons and also has stop or start options
[05-May-2011 18:13:51] <citrusfizz> i can tell zenoss is working.. cuz i am getting alerts.. but the webserver isn't running anymore
[05-May-2011 18:14:00] <citrusfizz> zenoss restart didn't get that back up
[05-May-2011 18:14:28] <ericenns> you could do a zenoss status to see if a particular daemon is down
[05-May-2011 18:14:47] <citrusfizz> Daemon: zopectl not running
[05-May-2011 18:15:12] <ericenns> you perform the same options as zenoss
[05-May-2011 18:19:32] <citrusfizz> wont stay running
[05-May-2011 18:19:38] <citrusfizz> Daemon: zopectl not running
[05-May-2011 18:19:58] <citrusfizz> zopectl start doesn't keep it running
[05-May-2011 18:20:03] <citrusfizz> is there a log file i can look at
[05-May-2011 18:20:33] <ericenns> all the logs are under $ZENHOME/logs not sure which one though
[05-May-2011 21:02:01] <otakup0pe> bleh is there like a max process arg lenght for zenoss
[05-May-2011 21:02:09] <otakup0pe> trying to add a new process on a host but it refuses to pick it up
[05-May-2011 22:02:42] kocolosk|away is now known as kocolosk
[05-May-2011 23:42:06] <MrCerulean> How can I have alert emails show hostname instead of ip address?
[05-May-2011 23:43:53] <MrCerulean> In the subject line, that is.
[05-May-2011 23:44:24] <MrCerulean> Currently, it says [zenoss] Process not running
[05-May-2011 23:44:38] <MrCerulean> I would like it to say [zenoss] hostname Process not running
[05-May-2011 23:44:57] <MrCerulean> Because the DNS server in my head is overloaded.
[06-May-2011 00:00:01] [disconnected at Fri May  6 00:00:01 2011]
[06-May-2011 00:00:01] [connected at Fri May  6 00:00:01 2011]
[06-May-2011 00:00:20] [zenoss-logger (logger bot) has joined #zenoss]
[06-May-2011 08:34:34] kocolosk is now known as kocolosk|away
[06-May-2011 08:49:47] kocolosk|away is now known as kocolosk
[06-May-2011 09:24:16] <St3v3o> morning folks
[06-May-2011 09:25:37] <Sam-I-Am> moo
[06-May-2011 09:28:18] <St3v3o> Sam you wouldn't have any experience with using snmp tables with zenoss would you.  I know Cacti has functionality to be able to read from an OID that outputs a table and generate multiple graphs .
[06-May-2011 09:28:49] <Sam-I-Am> the interface modeler component knows how to deal with indices
[06-May-2011 09:29:11] <St3v3o> So I would need to write a modeler then ?
[06-May-2011 09:29:18] <Simon4> snmp modelers tend to walk snmp tables
[06-May-2011 09:29:18] <Sam-I-Am> in cacti you wind up writing an xml file of some sort
[06-May-2011 09:29:25] <Sam-I-Am> its nasty
[06-May-2011 09:29:32] <St3v3o> yes cacti needs xml
[06-May-2011 09:29:34] <Simon4> and then create "components" for each table entry
[06-May-2011 09:29:43] <Sam-I-Am> i'd take a look at the code behind the interface modeler
[06-May-2011 09:30:15] <Simon4> and also Jane Curry's "zenpack" documentation, since her example zenpack enumerates a new snmp table
[06-May-2011 09:30:25] <Simon4> and she has step by step instructions for it
[06-May-2011 09:30:35] <Sam-I-Am> or that
[06-May-2011 09:31:47] <St3v3o> sweet, thanks guys…looks like I got come reading to do
[06-May-2011 09:47:16] <fragfutter> St3v3o: the funny part starts when you configure the columns of the components table in zenoss3. Javascript and multiple indirection interfaces
[06-May-2011 09:47:53] * Simon4 shudders
[06-May-2011 09:51:54] <Sam-I-Am> its good for you
[06-May-2011 09:52:31] <fragfutter> i would have preferred writing metal page templates
[06-May-2011 09:53:46] * Simon4 has cut his teeth on the v3 stuff now
[06-May-2011 09:53:51] <Simon4> it makes my brain bleed
[06-May-2011 09:54:01] * Sam-I-Am hasnt gotten there yet
[06-May-2011 09:54:09] <Simon4> I was tempted to write a code generator for all the copy/paste that I ended up doing
[06-May-2011 10:00:22] <froztbyte> haha
[06-May-2011 10:00:23] <froztbyte> how so?
[06-May-2011 10:01:05] <Simon4> you just end up repeating the same data in about four different files, just in slightly different formats
[06-May-2011 10:01:28] <froztbyte> on an upgrade from v2 or what?
[06-May-2011 10:01:31] <Simon4> potential for typo error is high
[06-May-2011 10:01:49] <Simon4> froztbyte: converting from a metal page template style zenpack to a V3 javascript-foo one
[06-May-2011 10:02:35] <froztbyte> oh, right
[06-May-2011 10:02:45] <froztbyte> yeah, that's gonna be razed's problem
[06-May-2011 10:02:59] <froztbyte> (a coworker of mine)
[06-May-2011 12:00:01] [disconnected at Fri May  6 12:00:01 2011]
[06-May-2011 12:00:02] [connected at Fri May  6 12:00:02 2011]
[06-May-2011 12:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[06-May-2011 12:05:32] <Sam-I-Am> so, if i'm going to crontab zenmodeler with collectors, should i run it from the collectors, or call the command on the hub which contacts the collectors to run it?
[06-May-2011 12:52:45] <davetoo> I would favor centralized management, but probably local execution if the management is strong
[06-May-2011 12:52:52] <davetoo> erm,
[06-May-2011 12:52:56] <davetoo> I dunno, maybe not
[06-May-2011 12:54:58] <xuru> hey guys, is there a good site on performance tuning zenoss?  one that is up-to-date would be great
[06-May-2011 14:46:25] mray is now known as mattray
[06-May-2011 14:52:08] mattray is now known as mray
[06-May-2011 15:28:29] <Hackman238> xuru: Still there?
[06-May-2011 15:28:37] <xuru> yep
[06-May-2011 15:28:53] <Hackman238> What type of performance problem are you having?
[06-May-2011 15:29:22] <xuru> I'm setting up a new box for zenoss
[06-May-2011 15:29:53] <xuru> our current one is very very slow in responsiveness in the web browser
[06-May-2011 15:29:56] <Hackman238> Ah. V3.0.3?
[06-May-2011 15:30:09] <xuru> so I wanted to get the filesystem and stuff just right
[06-May-2011 15:30:11] <xuru> yep
[06-May-2011 15:30:25] <Hackman238> About how many devices and web UI users?
[06-May-2011 15:31:36] <xuru> only a handful of web uses (maybe 4 or 5)
[06-May-2011 15:32:06] <xuru> still waiting for the "infrastructure" page to load
[06-May-2011 15:32:20] <Hackman238> Yeah, that's not atypical.
[06-May-2011 15:32:49] <xuru> ok, it's up now
[06-May-2011 15:32:53] <xuru> 123 devices
[06-May-2011 15:33:02] <Hackman238> There's a patch to help that problem, but I'd also reccomend modifying your /etc/my.cnf so your events table is *mostly* cached in RAM
[06-May-2011 15:33:40] <xuru> yeah, I've looked at
[06-May-2011 15:33:49] <xuru> and
[06-May-2011 15:33:56] <xuru> but they are a bit dated
[06-May-2011 15:34:15] <Hackman238> Most are. I don't think there's an updated Core tuning page.
[06-May-2011 15:34:20] <xuru> not too bad, but I was wondering if there was a more official page
[06-May-2011 15:34:26] <xuru> yeah
[06-May-2011 15:34:32] <Hackman238> Hows Zope performance?
[06-May-2011 15:35:17] <xuru> how do I quantify that?
[06-May-2011 15:35:39] <Hackman238> If you have to ask then its working plenty well for you :-)
[06-May-2011 15:35:59] <xuru> I think it's more of a function of the database
[06-May-2011 15:36:04] <Hackman238> Zope can be a huge performance limiter due to the fact its a GIL limited python app.
[06-May-2011 15:36:14] <xuru> we have a ton of events being sent to it
[06-May-2011 15:36:44] <nyeates> xuru: what points you to the bottleneck being the database?
[06-May-2011 15:37:16] <xuru> the page loads, then takes several minutes to load the list of events
[06-May-2011 15:37:24] <xuru> and the shear number of them
[06-May-2011 15:37:28] <Hackman238> Which page?
[06-May-2011 15:37:38] <Hackman238> Zenoss has two databases.
[06-May-2011 15:37:43] <xuru> ~140k yellow
[06-May-2011 15:37:53] <xuru> good question
[06-May-2011 15:38:26] <Hackman238> The best ways to help mysql slowness is caching the events table and patching code for sane queries.
[06-May-2011 15:38:58] <Hackman238> Zeo generally doens't have a problem- it's usually Zope/Zenhub causing anything percieved as a Zeo problem.
[06-May-2011 15:39:33] <Hackman238> s/percieved/perceived/
[06-May-2011 15:39:46] <xuru> I've modified the my.cnf file with what they recommend here
[06-May-2011 15:40:01] <nyeates> so the main evens page is very slow....are there any other pages that are also very slow...or most other pages come back in a few seconds?
[06-May-2011 15:40:40] <Hackman238> Whats the size of your entire events table verse your total RAM?
[06-May-2011 15:40:48] <xuru> devices and dashboard are both slow
[06-May-2011 15:41:26] <xuru> seems like any window is slow loading
[06-May-2011 15:41:40] <nyeates> might not necessarily be yer events mysql db then....
[06-May-2011 15:41:43] <nyeates> but it could be too
[06-May-2011 15:42:00] <xuru> like infrastructure > IP Services
[06-May-2011 15:42:04] <Hackman238> I agree, mysql could bne contributing, but it looks like your Zope is the problem
[06-May-2011 15:42:19] <xuru> the table on the left bottom is still trying to load
[06-May-2011 15:43:07] <xuru> so how do I improve that?
[06-May-2011 15:43:40] <nyeates> that tuning doc u gave is still up to date
[06-May-2011 15:43:41] <Hackman238> I'd set your zope cache-size to 125% the size of (Data.fs + Data.fs.old) / 2
[06-May-2011 15:44:01] <xuru> nyeates: ok, cool
[06-May-2011 15:44:18] <xuru> Hackman238: ah, I'll have a look
[06-May-2011 15:44:45] <xuru> currently it's cache-size 20MB
[06-May-2011 15:44:48] <Hackman238> Also, the zope object cache to 125% the total number of objects in the database
[06-May-2011 15:45:17] <nyeates> yeah make sure u llookup num of objects
[06-May-2011 15:45:28] <xuru> yeah, I figured that one out already (55k objects or so) so I set that to 100000
[06-May-2011 15:45:29] <Hackman238> Can check the number of Zope Objects in http://yoururl.local:8080/manage
[06-May-2011 15:46:03] <xuru> -rw-r--r--  1 zenoss zenoss  18G May  6 14:28 Data.fs
[06-May-2011 15:46:03] <xuru> -rw-r--r--  1 zenoss zenoss 5.3G Nov  8 10:31 Data.fs.bak
[06-May-2011 15:46:06] <Hackman238> Also, the python check interval is insanely important on processors with large pipelines.
[06-May-2011 15:46:16] <nyeates> If you log in as the admin user and  visit the following URL: http://YOUR_SERVER:8080/Control_Panel/Database/main/manage_cacheParameters you will be able to see how many objects are in your database.
[06-May-2011 15:47:13] <Hackman238> you need to pack your database more often
[06-May-2011 15:47:55] <xuru> nyeates: a admin, or "admin"?
[06-May-2011 15:47:56] <nyeates> whoa yeah those are large files for so few devices
[06-May-2011 15:48:05] <Hackman238> Zeopack line is in /etc/cron.weekely/zenoss; dont forget the environmental parameters. Create a new cron job to pack every couple days
[06-May-2011 15:48:10] <nyeates> a zope admin uesr
[06-May-2011 15:48:40] <xuru> I have this in the zenoss users crontab:
[06-May-2011 15:48:43] <xuru> # m h  dom mon dow   command
[06-May-2011 15:48:43] <xuru> #run pack every monday morning at 2am
[06-May-2011 15:48:43] <xuru> 0 2 * * 1 bash -lc “$ZENHOME/bin/ -h localhost -p 8100 >> usr/local/zenoss/zenoss/logs/zeopack.log 2>&
[06-May-2011 15:48:47] <Hackman238> After repack, recalculate the right amount to assign to zope.conf zeo cache
[06-May-2011 15:49:24] <Hackman238> Make sure that bash runs as user zenoss
[06-May-2011 15:49:35] <xuru> interesting
[06-May-2011 15:49:38] <Hackman238> Helps with permissions and evnironemntal params
[06-May-2011 15:49:38] <xuru> [02:32 PM]zenoss@ks-mgmt-monitor-01:etc$ bash -lc $ZENHOME/bin/ -h localhost -p 8100
[06-May-2011 15:49:38] <xuru> -h: /usr/local/zenoss/zenoss/bin/ No such file or directory
[06-May-2011 15:49:55] <Hackman238> Really?
[06-May-2011 15:50:09] <xuru> yeah, it's not there
[06-May-2011 15:50:12] <Hackman238> do a updatedb as root then a locate
[06-May-2011 15:50:28] <Hackman238> It should be. Its' very important
[06-May-2011 15:51:13] <xuru> yeah, you would think
[06-May-2011 15:51:32] <nyeates> are u on the stack install?
[06-May-2011 15:51:34] <xuru> [02:33 PM]zenoss@ks-mgmt-monitor-01:etc$ aptitude show zenoss-stack
[06-May-2011 15:51:34] <xuru> Package: zenoss-stack
[06-May-2011 15:51:38] <Hackman238> Yes, hes stack
[06-May-2011 15:51:39] <xuru> Version: 3.1.0-0
[06-May-2011 15:51:46] <nyeates> huh weird that its not ther
[06-May-2011 15:52:24] <xuru> [02:35 PM]root@ks-mgmt-monitor-01:eplaster# locate
[06-May-2011 15:52:25] <xuru> /usr/local/zenoss/python/lib/python2.6/site-packages/ZODB3-3.9.3-py2.6-linux-x86_64.egg/ZEO/scripts/
[06-May-2011 15:52:25] <xuru> /usr/local/zenoss/python/lib/python2.6/site-packages/ZODB3-3.9.3-py2.6-linux-x86_64.egg/ZEO/scripts/manual_tests/
[06-May-2011 15:52:25] <xuru> /usr/local/zenoss/python/lib/python2.6/site-packages/ZODB3-3.9.3-py2.6-linux-x86_64.egg/ZODB/scripts/manual_tests/
[06-May-2011 15:52:28] <nyeates> i havent used stack in a while, so maybe it puts it elsewhere? dunno
[06-May-2011 15:52:31] <xuru> sorry for the spam
[06-May-2011 15:53:02] <Hackman238> cat /usr/local/zenoss/python/lib/python2.6/site-packages/ZODB3-3.9.3-py2.6-linux-x86_64.egg/ZEO/scripts/ | wc -l for me
[06-May-2011 15:53:34] <xuru> 181
[06-May-2011 15:53:50] <Hackman238> Okay yeah, thats the sub script.
[06-May-2011 15:54:04] <Hackman238> your zeopack in ./bin should be there.
[06-May-2011 15:54:32] <xuru> should I just link it?
[06-May-2011 15:54:37] <Hackman238> negative
[06-May-2011 15:54:41] <Hackman238> They are differnt.
[06-May-2011 15:54:49] <xuru> ok
[06-May-2011 15:55:06] <Hackman238> Download the tarball or rpm and decompress it and fetch the file manually
[06-May-2011 15:55:19] <xuru> k
[06-May-2011 15:55:26] <Hackman238> I'll be about 125 lines
[06-May-2011 15:56:22] <Hackman238> Wait, my bad
[06-May-2011 15:57:03] <Hackman238> *Face palm for Zenoss folks*
[06-May-2011 15:57:14] <xuru> ?
[06-May-2011 15:57:15] <Hackman238> It's been moved in V3 but the cron job isnt updated
[06-May-2011 15:57:26] <xuru> doh
[06-May-2011 15:57:40] <Hackman238> I just installed a V3.1 client in a VM for testing. It's not there and the cron is wrong.
[06-May-2011 15:58:10] <xuru> interesting
[06-May-2011 15:58:15] <xuru> bug?
[06-May-2011 15:58:24] <Hackman238> Yep.
[06-May-2011 15:58:37] <Hackman238> Is your platform backed up?
[06-May-2011 15:59:07] <xuru> Yes, but I better do a backup of the database
[06-May-2011 15:59:10] <Hackman238> If not, move Data.fs.bak to Data.fs.1
[06-May-2011 15:59:40] <Hackman238> then you can su zenoss; python /usr/local/zenoss/python/lib/python2.6/site-packages/ZODB3-3.9.3-py2.6-linux-x86_64.egg/ZEO/scripts/ -h localhost -p 8100
[06-May-2011 16:00:13] <Hackman238> I mean, at your own risk. Though it does match my file and it worked here in my test.
[06-May-2011 16:00:29] <xuru> ok, let me take a look here
[06-May-2011 16:00:41] <xuru> I have a backup from last night, so I'll try that now
[06-May-2011 16:00:52] <xuru> do I need to stop zenoss-stack first?
[06-May-2011 16:00:59] <Hackman238> No, keep it running
[06-May-2011 16:01:10] <Hackman238> The web UI will be unusably slow while this is running.
[06-May-2011 16:01:14] <Hackman238> It'll take a few mintues.
[06-May-2011 16:01:35] <Hackman238> So if you have users on right now, inform them or hold off.
[06-May-2011 16:02:35] <Hackman238> A good friend of mine used to say "I don't always test my code, but when I do, I do it in production!" - Suffice to say he had a bad reputation with management :-P
[06-May-2011 16:02:53] <xuru> hehe
[06-May-2011 16:03:06] <xuru> I get a permission denied
[06-May-2011 16:03:15] <Hackman238> Are you user zenoss?
[06-May-2011 16:03:21] <xuru> yeah
[06-May-2011 16:03:28] <Hackman238> Wow thats odd
[06-May-2011 16:03:35] <xuru>  #!/usr/bin/env python2.3
[06-May-2011 16:03:46] <xuru> don't have python2.3 on my box I don't think
[06-May-2011 16:04:04] <xuru> 2.6.2
[06-May-2011 16:04:07] <Hackman238> Ugh... second face palm for Zenoss folks!
[06-May-2011 16:04:13] <xuru>
[06-May-2011 16:04:44] <Hackman238> set it to #!/usr/bin/env python
[06-May-2011 16:05:57] <xuru> didn't have execute permissions either, so I ran it with python out in front
[06-May-2011 16:06:40] <xuru> nyeates: can I get an account in Trac?
[06-May-2011 16:07:05] <Hackman238> chmod a+x it
[06-May-2011 16:07:22] <Hackman238> and ls -lh it to make sure owned by zenoss:zenoss
[06-May-2011 16:07:46] <xuru> done
[06-May-2011 16:08:13] <xuru> change the cron job to run this file instead?  or link it now?
[06-May-2011 16:08:42] <Hackman238> I'd change the cron directly
[06-May-2011 16:08:53] <Hackman238> Did it run successfully?
[06-May-2011 16:08:55] <xuru> k
[06-May-2011 16:08:59] <xuru> still running
[06-May-2011 16:09:03] <Hackman238> Okay
[06-May-2011 16:10:27] <Hackman238> Grr...This problem has been noted in V3 far back as Sept 9 2010 and it hasn't been fixed in sub releases.
[06-May-2011 16:12:26] <xuru> what ticket number?
[06-May-2011 16:12:46] <Hackman238> Just in the forums I mean.
[06-May-2011 16:12:59] <xuru> oh
[06-May-2011 16:13:09] <Hackman238> Core users are the BEST code testers, however. This problem shouldn'
[06-May-2011 16:13:14] <Hackman238> t have been over looked so long
[06-May-2011 16:13:24] * xuru nods
[06-May-2011 16:13:46] <xuru> I put a lot of time into tracking down possible issues
[06-May-2011 16:14:04] <xuru> not saying it wasn't worth it, but...
[06-May-2011 16:14:12] <Hackman238> Yep and yep.
[06-May-2011 16:14:59] <xuru> can't find a ticket for it, so I'll go ahead and make one
[06-May-2011 16:16:35] <Hackman238> Ok
[06-May-2011 16:16:37] <Hackman238> TY
[06-May-2011 16:16:45] <xuru> np
[06-May-2011 16:17:47] <xuru> hey...  if I'm in the pulldown for the "assign to", I must have a login
[06-May-2011 16:17:53] <xuru> just got to remember the password now
[06-May-2011 16:18:32] <Hackman238> LOL
[06-May-2011 16:22:12] <xuru> woot! found my email from matt
[06-May-2011 16:22:48] <xuru> heh, it's still running
[06-May-2011 16:23:03] <mattray> xuru: ?
[06-May-2011 16:23:16] <mattray> trac?
[06-May-2011 16:23:23] <xuru> mattray: sorry to wake you   yeah
[06-May-2011 16:23:56] <mattray> carry on
[06-May-2011 16:24:20] <Hackman238> xuru: it will take a while
[06-May-2011 16:24:47] <xuru> hehe, you're like me and have trigger words that pops up a libnotify don't you?
[06-May-2011 16:25:05] <mattray> yeah, I've got a few
[06-May-2011 16:30:18] <Hackman238> BRB- coffee time
[06-May-2011 16:30:34] <xuru> Hackman238: cool, thanks for your help
[06-May-2011 16:31:00] <xuru> it finished, and Data.fs is now 257M  >.<
[06-May-2011 16:33:52] <Hackman238> Great
[06-May-2011 16:34:10] <Hackman238> Make your zope zeo cache 500 MB
[06-May-2011 16:34:37] <Hackman238> Once changed issue a zopectl restart.
[06-May-2011 16:35:05] <Hackman238> You should see a speed differnce and the bottleneck should be elsewhere now.
[06-May-2011 16:35:39] <xuru> sweet
[06-May-2011 16:37:06] <xuru> ok, I'll go ahead and assign this ticket to matt then
[06-May-2011 16:37:25] <Hackman238> Yep, should be pretty good. In comparrison, one of my largest zenoss installs has a packed Data.fs pushing 25GB
[06-May-2011 16:38:18] <Hackman238> ...which at this point I'm at all ends to optimize it XD
[06-May-2011 16:38:41] <xuru> heh, yeah, that's pretty big
[06-May-2011 16:39:06] <xuru> where do I put that cache size?
[06-May-2011 16:39:15] <xuru> don't see it in zeo.conf
[06-May-2011 16:39:22] <Hackman238> Do yourself a favor...if it looks like it's going to break 8GB packed- figure a way to split up your instances.
[06-May-2011 16:39:28] <Hackman238> zope.conf
[06-May-2011 16:39:55] <xuru> oh the zeoclient section?
[06-May-2011 16:40:18] <Hackman238> Correct
[06-May-2011 16:40:49] <Hackman238> Above that, in the zodb section, cache-size is a int of objects
[06-May-2011 16:41:21] <xuru> so I have:
[06-May-2011 16:41:23] <xuru> <zodb_db main>
[06-May-2011 16:41:29] <xuru>   cache-size 100000
[06-May-2011 16:41:34] <xuru>   <zeoclient>
[06-May-2011 16:41:39] <xuru>     cache-size 500MB
[06-May-2011 16:41:44] <xuru> is that correct?
[06-May-2011 16:41:49] <Hackman238> yep
[06-May-2011 16:41:55] <xuru> ok, cool
[06-May-2011 16:42:08] <xuru> hey, I just noticed something else
[06-May-2011 16:42:20] <xuru>  /usr/local/zenoss/python/bin/zeopack
[06-May-2011 16:42:22] <Hackman238> Also, in /opt/zenoss/etc/zeo.conf put invalidation-queue-size 200 under zeo section
[06-May-2011 16:42:41] <Hackman238> Oh, guess they renamed it
[06-May-2011 16:42:59] <Hackman238> Still didnt update the cron job though
[06-May-2011 16:43:06] <xuru> yep
[06-May-2011 16:44:26] <Hackman238> For the zeo change to take affect, you'll need to restart all of zenoss. I'd wait until your next scheduled down time for that.
[06-May-2011 16:44:39] <Hackman238> That tweak will only help under load.
[06-May-2011 16:44:48] <otakup0pe> ugh zenoss is awful on mobile
[06-May-2011 16:44:56] <otakup0pe> impossible to navigate. sigh.
[06-May-2011 16:45:03] <Hackman238> Blame Ajax
[06-May-2011 16:45:11] <Hackman238> Personally....I hate ajax
[06-May-2011 16:45:12] <otakup0pe> won't do
[06-May-2011 16:45:16] <otakup0pe> splunk works quite well on mobile
[06-May-2011 16:45:33] <otakup0pe> parts of it anyway
[06-May-2011 16:45:37] <Hackman238> heh
[06-May-2011 16:45:47] <Hackman238> Have you tried opera mini?
[06-May-2011 16:45:55] <Hackman238> more funcationality than opera mobile.
[06-May-2011 16:45:55] <otakup0pe> nope
[06-May-2011 16:46:17] <Hackman238> Works very well. Though I run Windows Mobile 6.5
[06-May-2011 16:46:38] <Hackman238> Cant vouch for Android performance, etc.
[06-May-2011 16:47:50] <xuru> haven't had a reason to try, but it would be nice to have a mobile version of the site for top critical events etc
[06-May-2011 16:48:44] <Hackman238> I won't lie. I gave up working Zenoss work on mobile...I just decided to make a Joomla plugin to fetch info from Zenoss
[06-May-2011 16:49:04] <xuru> heh
[06-May-2011 16:51:09] <Hackman238> nyeates: You around?
[06-May-2011 16:53:57] <otakup0pe> yeah i've seen other apps which tap the zenoss "api" to provide mobile read-only info
[06-May-2011 16:54:00] <otakup0pe> which is generally fine
[06-May-2011 16:55:13] <Hackman238> otakup0pe: My plugin is R/W ;-) handy for lots of little things
[06-May-2011 16:55:52] <Hackman238> when I find time to clean up the code I'll release it.
[06-May-2011 16:56:37] <otakup0pe> cool
[06-May-2011 16:58:43] <Hackman238> Well I'm headed out. Have a good weekend all.
[06-May-2011 16:58:43] <xuru> Hackman238: zenoss feels much more responsive.  Ty
[06-May-2011 16:58:53] <Hackman238> xuru: NP, anytime.
[06-May-2011 16:58:54] <xuru> you too Hackman238
[06-May-2011 16:59:07] <Hackman238> xuru: That's why I have so many Zenoss customers!
[06-May-2011 16:59:16] <xuru>
[06-May-2011 17:42:44] kocolosk is now known as kocolosk|away
[06-May-2011 19:37:02] kocolosk|away is now known as kocolosk
[06-May-2011 21:34:10] <davetoo> yadda
[06-May-2011 21:36:28] <davetoo> has anyone tried FireFox 4.0 yet?
[07-May-2011 00:00:01] [disconnected at Sat May  7 00:00:01 2011]
[07-May-2011 00:00:01] [connected at Sat May  7 00:00:01 2011]
[07-May-2011 00:00:17] [zenoss-logger (logger bot) has joined #zenoss]
[07-May-2011 01:46:24] kocolosk is now known as kocolosk|away
[07-May-2011 05:44:53] citrusfizz_ is now known as citrusfizz
[07-May-2011 06:27:42] <kotique> do you guys have a bugtracker? I'm getting  host "does not publish HOST-RESOURCES-MIB" after snmpd restart
[07-May-2011 12:00:01] [disconnected at Sat May  7 12:00:01 2011]
[07-May-2011 12:00:02] [connected at Sat May  7 12:00:02 2011]
[07-May-2011 12:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[07-May-2011 15:23:45] kocolosk|away is now known as kocolosk
[07-May-2011 16:59:41] kocolosk is now known as kocolosk|away
[08-May-2011 00:00:01] [disconnected at Sun May  8 00:00:01 2011]
[08-May-2011 00:00:02] [connected at Sun May  8 00:00:02 2011]
[08-May-2011 00:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[08-May-2011 02:51:57] kocolosk|away is now known as kocolosk
[08-May-2011 09:10:19] <alinux_> hi I am getting Remote command execution failed: Host key verification failed. I am using nrpe ...the remote user must be way I can run this remote command as root ?
[08-May-2011 11:08:22] kocolosk is now known as kocolosk|away
[08-May-2011 12:00:01] [disconnected at Sun May  8 12:00:01 2011]
[08-May-2011 12:00:02] [connected at Sun May  8 12:00:02 2011]
[08-May-2011 12:00:21] [zenoss-logger (logger bot) has joined #zenoss]
[08-May-2011 17:31:40] <mkunze> hey little question i got snmp data source but it seems to give me integes in following format: dbInfoTransactions_dbInfoTransactions.rrd: 3968927L   << so rrdtool says me its nan, anyway to simple fix that?
[08-May-2011 17:36:32] <mkunze> snmpwald in zenoss shows me: SNMPv2-SMI::mib- = Gauge32: 3969004
[08-May-2011 21:34:54] kocolosk|away is now known as kocolosk
[08-May-2011 22:51:59] kocolosk is now known as kocolosk|away
[09-May-2011 00:00:01] [disconnected at Mon May  9 00:00:01 2011]
[09-May-2011 00:00:02] [connected at Mon May  9 00:00:02 2011]
[09-May-2011 00:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[09-May-2011 09:09:30] kocolosk|away is now known as kocolosk
[09-May-2011 09:41:17] <Gat0rvean> Is there a way to tell zenoss not to ask SNMP statistics for anything except certain interfaces?
[09-May-2011 09:41:50] <Simon4> there's a configuration property you can set to tell it to ignore certain interfaces when monitoring, or you can set "monitored" to false on the interfaces you don't care about
[09-May-2011 09:41:59] <Simon4> the config property takes a regex from memory
[09-May-2011 09:43:37] <Gat0rvean> Simon4: are you talking about the zInterfaceMapIgnoreNames?
[09-May-2011 09:45:32] <Simon4> Gat0rvean: yup
[09-May-2011 09:46:43] <Gat0rvean> Simon4:  I've got that setup to ignore certain interfaces (I'm monitoring an Occam BLC which has 10 GigE ports, 30 DSL ports and 30 POTS ports) and it's getting HAMMERED with SNMP queries from Zenoss still, even with the DSL and POTS ports ignored, and I can't figure out why
[09-May-2011 09:47:12] <Simon4> have you remodelled since changing the setting?
[09-May-2011 09:47:30] <Gat0rvean> Simon4:  Yes, and only the GigE ports show up under "Interfaces" now
[09-May-2011 09:47:52] <Simon4> my only guess is to restart zenoss, the collector process may not have picked up the changes
[09-May-2011 12:00:01] [disconnected at Mon May  9 12:00:01 2011]
[09-May-2011 12:00:02] [connected at Mon May  9 12:00:02 2011]
[09-May-2011 12:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[09-May-2011 12:08:51] <kokey> my zenoss interface has gotten faster
[09-May-2011 12:08:57] <kokey> i'm worried as to why this could happen
[09-May-2011 12:09:10] <Hackman238> kokey: XD
[09-May-2011 12:09:18] <kokey> XD?
[09-May-2011 12:10:13] <Hackman238> kokey: its laughing with eyes closed
[09-May-2011 12:10:29] <kokey> ooh
[09-May-2011 12:10:31] <kokey> heh
[09-May-2011 12:10:33] <Hackman238> kokey: what version are you running?
[09-May-2011 12:10:38] <kokey> 3.1
[09-May-2011 12:10:51] <Hackman238> kokey: faster since 3.0.3 -> 3.1?
[09-May-2011 12:11:01] <Hackman238> kokey: or just out of nowheres?
[09-May-2011 12:11:08] <kokey> nope, just faster by itself, magic
[09-May-2011 12:11:15] <kokey> actually, i've done one real change
[09-May-2011 12:11:37] <kokey> and that was to put some ignore file system names for some netapp filesystems we had
[09-May-2011 12:11:54] <kokey> anyway
[09-May-2011 12:11:56] <kokey> i talk too fast
[09-May-2011 12:12:02] <kokey> suddenly the whole thing is very slow
[09-May-2011 12:12:12] <Hackman238> kokey: Thats more like it :-D
[09-May-2011 12:12:38] <Hackman238> kokey: Have you tuned your setup for performance?
[09-May-2011 12:14:02] <kokey> yeah I have
[09-May-2011 12:14:08] <kokey> ok, I have figured out the root of the slowness
[09-May-2011 12:14:15] <kokey> it's my mysql performance
[09-May-2011 12:14:20] <kokey> that comes and goes at the moment
[09-May-2011 12:14:36] <kokey> otherwise I believe filtering out the netapp snapshots sped things up a lot
[09-May-2011 12:14:43] <Hackman238> kokey: That'll hurt you everywhere in Zenoss 3. Do you cache the entire table?
[09-May-2011 12:14:53] <kokey> the netapp zencommand was eating a lot of CPU generally
[09-May-2011 12:15:11] <Hackman238> kokey: Absolutely. Avoid using zencommand whenever possible.
[09-May-2011 12:15:26] <kokey> Hackman238: the zenpack uses that zencommand so I have no choice
[09-May-2011 12:16:00] <kokey> it doesn't run it that often, it's just that each run uses a lot of CPU for some reason, but I suspect it may have been because of the netapp snmp bug on snapshots
[09-May-2011 12:16:17] <kokey> actually there are a few more things I can do to tune the mysql server
[09-May-2011 12:16:30] <kokey> like let it do dirty reads
[09-May-2011 12:16:32] <Hackman238> kokey: I'm unfamilliar with that pack. It's not possible to convert it to use snmp?
[09-May-2011 12:16:38] <kokey> not sure where I can set that tho
[09-May-2011 12:16:51] <kokey> Hackman238: it's part of enterprise, they seem to be a bit lazy about it
[09-May-2011 12:17:37] <Hackman238> kokey: Yeah the team has been busy with Avalon.
[09-May-2011 12:18:16] <Hackman238> kokey: Never the les,s if the pack is hurt performance too much, it might be worth your while to convert it to snmp or write your own zenoss daemon.
[09-May-2011 12:19:13] <kokey> Hackman238: I figure if we pay for enterprise we should just nag for it to be improved
[09-May-2011 12:19:33] <kokey> Hackman238: the CPU usage is actually one of the lesser worries with this zenpack really
[09-May-2011 12:20:57] <Hackman238> kokey: Good luck. I have ent. and I can't get ahold of staff timely very often.
[09-May-2011 12:21:44] <Hackman238> kokey: Its my understanding they're just too busy. Yeah with zencommand you really have to worry about runaway and zombie operations piling up.
[09-May-2011 12:24:38] <kokey> I wonder if set global transaction isolation level read uncommitted is safe with zenoss
[09-May-2011 12:24:42] <kokey> on mysql that is
[09-May-2011 12:26:25] <Hackman238> kokey: Hum. I cant think of a situation where it would hurt anything.
[09-May-2011 12:26:42] <kokey> Hackman238: I've just set it and will watch the situation for a bit
[09-May-2011 12:26:47] <kokey> it's looking good tho
[09-May-2011 12:27:32] <Hackman238> kokey: Let me know how it goes with that.
[09-May-2011 12:28:22] <Hackman238> kokey: personally I cache the entire table in RAM, but my boxes all have 64GB RAM+.
[09-May-2011 12:28:55] <kokey> Hackman238: you force that?
[09-May-2011 12:29:07] <kokey> I think mine doesn't grow to more than 2GB so I could possibly do that
[09-May-2011 12:29:13] <kokey> that said, it's on a VM
[09-May-2011 12:29:19] <kokey> this install is only monitoring 500 devices
[09-May-2011 12:29:38] <Hackman238> kokey: yep. Cache about 25GB of tables.
[09-May-2011 12:29:51] <kokey> actually the read uncommited thing works so well I'm not going to bother with throwing RAM at it for a while
[09-May-2011 12:30:19] <Hackman238> kokey: absoltely- no need to waste VM RAM if the uncommitted tweak is helping
[09-May-2011 12:30:27] <Hackman238> kokey: *absolutely
[09-May-2011 12:31:23] <kokey> you with RS in SA?
[09-May-2011 12:31:52] <Hackman238> kokey: Yes.
[09-May-2011 12:32:13] <kokey> i worked on their uk side for a while, in 2003-2004
[09-May-2011 12:32:27] <kokey> some of my old friends and colleagues moved over to SA recently
[09-May-2011 12:32:42] <Hackman238> kokey: Nice. Why such a short tour?
[09-May-2011 12:33:22] <Hackman238> kokey: Yeah, I just moved to SA myself. I'm from NY...needless to say the weather here is one item requiring adaptation.
[09-May-2011 12:33:28] <kokey> it was early days, we were only 8 techs and did everything from taking delivering to 3rd level support
[09-May-2011 12:34:07] <Hackman238> kokey: Gotcha. That can be a grind.
[09-May-2011 12:34:09] <kokey> Hackman238: the guys who I know who moved there are all south africans like me, they prefer the weather over there a lot compared to the UK
[09-May-2011 12:34:57] <kokey> Hackman238: yeah it was insane, the busiest I've ever been at a job until my current contract
[09-May-2011 12:35:27] <kokey> things did get better as I left, hence why most of the people who stayed behind remained there for years longer
[09-May-2011 12:37:26] <Hackman238> kokey: I've heard some depts. talk of being so busy like that. I can't say that here in the NOC I or the others are over whelmed. I couldn't blame anyone for leaving if everyday is rush rush, fire fire.
[09-May-2011 12:38:14] <kokey> Hackman238: it helped when the office grew so they could split out the 1st level, DC and enterprise people
[09-May-2011 12:38:52] <Hackman238> kokey: Yep.
[09-May-2011 12:39:02] <Hackman238> kokey: Where are you at now?
[09-May-2011 12:39:16] <kokey> I remember having to be the only person on a night shift, having some lonely customer try help him with his email settings over the phone while trying to do some scheduled upgrades and the pager going off because of some servers crashing
[09-May-2011 12:39:24] <rmatte> Hackman238: I get a 403 forbidden error on the link you sent me
[09-May-2011 12:39:54] <Hackman238> rmatte: It should work in the browser. I'm not sure why wget 403's
[09-May-2011 12:40:00] <rmatte> ah
[09-May-2011 12:40:07] <kokey> Hackman238: some financial company called markitserv, they're co owned by the DTCC from NY, but i'm out of here in 3 weeks time
[09-May-2011 12:40:16] <Hackman238> rmatte: Any idea why so I can fix it? :-)
[09-May-2011 12:40:20] <rmatte> you're right, weird
[09-May-2011 12:40:24] <rmatte> hmmm
[09-May-2011 12:40:58] <kokey> rmatte: they probably don't like someone mirroring them, perhaps change your user-agent ;-)
[09-May-2011 12:41:00] <Hackman238> kokey: Sounds cool
[09-May-2011 12:41:24] <rmatte> Hackman238: well, I know what's causing it
[09-May-2011 12:41:32] <rmatte> if I spoof the user-agent with wget it works
[09-May-2011 12:41:51] <rmatte> wget -U 'Mozilla/5.0 (X11; U; Linux i686; en-US; rv: Gecko/20070802 SeaMonkey/1.1.4'
[09-May-2011 12:41:52] <rmatte> works
[09-May-2011 12:41:52] <Hackman238> rmatte: hum. I'll fix that.
[09-May-2011 12:42:10] <rmatte>
[09-May-2011 12:43:00] <Hackman238> rmatte: Let me know how that pack tests in V3. Please don't dist. yet.
[09-May-2011 12:43:08] <rmatte> will do
[09-May-2011 12:43:24] <Hackman238> rmatte: Also, don't install it in'll eat zope!
[09-May-2011 12:43:29] <rmatte> I have some v2 deadlines so I have to make a custom mod of the v2 pack first
[09-May-2011 12:43:38] <rmatte> k
[09-May-2011 12:43:47] <Hackman238> rmatte: I have a 98% done V2 one for you later
[09-May-2011 12:44:33] <rmatte> awesome
[09-May-2011 12:44:43] <Hackman238> rmatte: If you make code changes, just email them to me
[09-May-2011 12:44:53] <rmatte> will do
[09-May-2011 12:45:11] <Hackman238> rmatte: I'll get them all intergrated pretty quick. The V3 version *almost* works with Avalon
[09-May-2011 12:45:28] <rmatte> nice, I'll give it a shot on my v3 lab box later
[09-May-2011 12:45:38] <Hackman238> rmatte: TY
[09-May-2011 12:45:56] <Hackman238> kokey: You ise IPSLA?
[09-May-2011 12:46:57] <Hackman238> kokey: *use
[09-May-2011 12:48:21] <kokey> Hackman238: not here, nope
[09-May-2011 12:53:47] <kokey> haha
[09-May-2011 12:53:51] <kokey> i just realised now
[09-May-2011 12:54:10] <kokey> zenoss implemented their alerting rules thing almost exactly like i built one for nagios in my last job
[09-May-2011 12:54:32] <kokey> except I couldn't find a nice query builder to use, so I let it use perl expressions instead
[09-May-2011 12:55:24] <Hackman238> LOL
[09-May-2011 12:57:38] <kokey> we have a guy here who thought putting in 48 different alerting rules is the way to go
[09-May-2011 12:57:56] <kokey> 6 groups, 8 rules per group
[09-May-2011 12:58:15] <Hackman238>  *face palm* 48 rules  == unmanagable
[09-May-2011 12:58:38] <kokey> well this guy is an expert and desinging and implementing unmanagable stuff
[09-May-2011 12:58:52] <kokey> I can run my only daily wtf site from his creations
[09-May-2011 12:59:01] <kokey> only=own
[09-May-2011 12:59:17] <Hackman238> Ha. Whys he still ahev ajob?
[09-May-2011 12:59:39] <kokey> well they told him his job is finished at the end of the month
[09-May-2011 13:02:25] <Hackman238> Yeah. Its bad for any business to have a guy design and implement unmanagable and convoluted works.
[09-May-2011 13:05:58] <kokey> it's quite remarkable to see him in action
[09-May-2011 13:06:02] <kokey> he can't help it
[09-May-2011 13:06:06] <kokey> and he confuses himself a lot of the time
[09-May-2011 13:08:55] <Hackman238> LOL
[09-May-2011 13:19:57] <rmatte> Hackman238: to answer your question of why he still has a job:
[09-May-2011 13:21:11] <rmatte> I support several scripts that I'm convinced were written according to that guide
[09-May-2011 13:21:31] <rmatte> but I'm good at decyphering crappy code, just takes a bit of time
[09-May-2011 13:22:31] <Hackman238> rmatte: Oh should see my personal code. I normalize variable names before giving code away usually
[09-May-2011 13:22:54] <rmatte> hehe
[09-May-2011 13:23:10] <rmatte> Yeh, I'm not the cleanest coder in the world either, but I'm alright
[09-May-2011 13:23:32] <rmatte> Generally my code is pretty clean unless I'm in the process of learning a new language, then it's horrid
[09-May-2011 13:23:45] <Hackman238> rmatte: I abbriviate really badly, but its 100% consistant. -> for pSs_inter in range (int(dFzT), int(pSsS)):
[09-May-2011 13:24:31] <Hackman238> rmatte: And -> lG('\n processing ' + str(hNSgr) + ' ' + str(gPu) + ' file: ' + str(fqfL), vBs)
[09-May-2011 13:25:38] <Hackman238> rmatte: LOL my super at RackSpace saw some of my scripts and was like 'M`ere. I want to test you tos ee if you can read your code.' Was no sweat, but definately needed normalization.
[09-May-2011 13:29:56] <rmatte> hehe
[09-May-2011 13:30:37] <rmatte> I don't really abbreviate, I just use underscores
[09-May-2011 13:30:43] <rmatte> and keep it short and sweet
[09-May-2011 13:31:04] <rmatte> line_count, last_space, etc...
[09-May-2011 13:32:27] <Hackman238> rmatte: Yep. Much better way. I dont remember the representation of words well, but I can associate and remember long numbers and seemingly random lines of characters to a variable, function, etc.
[09-May-2011 13:32:42] <rmatte> lol
[09-May-2011 13:33:04] <Hackman238> Like PsSs is always the number of iterations assosiated as a pass
[09-May-2011 13:33:36] <Hackman238> And pSsS os the current pass
[09-May-2011 13:34:07] <rmatte> I so badly want to start my own MSP business and use nothing but open source tools... This is such a windows shop here and I'm sore tired of having to try to integrate closed solutions with open solutions
[09-May-2011 13:34:10] <rmatte> nightmare
[09-May-2011 13:34:20] <rmatte> so tired*
[09-May-2011 13:34:39] <Hackman238> rmatte: I know how you feel.
[09-May-2011 13:35:13] <Hackman238> rmatte: Problem is all our closed source tools are developed by another department. XD
[09-May-2011 13:35:25] <rmatte> hehe
[09-May-2011 13:35:49] <Hackman238> rmatte: So I cant just say 'your app sucks because {logical reason for sucakge} or people get upset LOL
[09-May-2011 13:35:57] <rmatte> yeh
[09-May-2011 13:36:30] <rmatte> honestly though, I have enough knowledge that it would probably take me about a month to setup a full NOC toolset
[09-May-2011 13:36:37] <rmatte> made up of all opensource stuff
[09-May-2011 13:36:41] <rmatte> so tempting
[09-May-2011 13:36:42] <rmatte> hehe
[09-May-2011 13:36:50] <Hackman238> rmatte: You a webportal that gets thousands of hits a second written in Python (essentially 1 thread due to GIL)
[09-May-2011 13:37:05] <rmatte> ah, yeh
[09-May-2011 13:37:21] <Hackman238> rmatte: *cough*..thats our CRM!
[09-May-2011 13:37:39] <Hackman238> rmatte: You should go for it assuming available resources
[09-May-2011 13:37:46] <rmatte> we use microsoft CRM here, worst software ever
[09-May-2011 13:38:00] <Hackman238>  *double face palm*
[09-May-2011 13:38:03] <Hackman238> Very sorry.
[09-May-2011 13:38:21] <rmatte> if I had a choice of CRM software to use I'd use vtiger crm
[09-May-2011 13:38:59] <Hackman238> yeah
[09-May-2011 13:43:41] <Hackman238> Wow 1 already. Days almost over.
[09-May-2011 13:54:02] <rmatte> hehe
[09-May-2011 14:20:09] <rmatte> It's 2:20 here and I'm probably here until around 6:30
[09-May-2011 14:21:37] <Hackman238> rmatte: Wow. Work late. I'm out by 4
[09-May-2011 14:30:16] <rmatte> well, I started late
[09-May-2011 14:30:32] <rmatte> started at 10:30
[09-May-2011 14:30:56] <rmatte> Hackman238: you know perl yeh?
[09-May-2011 14:38:41] <Hackman238> rmatte: Yes
[09-May-2011 14:41:12] <rmatte> ok, quick question
[09-May-2011 14:41:46] <rmatte> if I want to break up a line in to multiple lines, can I just do \ at the end of the line and space the rest down to the next line like I do in bash and python?
[09-May-2011 14:41:51] <rmatte> or is there some other way of doing it?
[09-May-2011 14:43:33] <rmatte> just to keep it from running over the end of the page
[09-May-2011 14:43:47] <Hackman238> rmatte: after \ a normal return and tab / spaces to align should work
[09-May-2011 14:44:05] <rmatte> ok, excellent
[09-May-2011 14:44:06] <rmatte> thanks
[09-May-2011 14:46:42] <Hackman238> np
[09-May-2011 16:11:56] <rmatte> Hackman238: it doesn't seem to like the \
[09-May-2011 16:12:09] <rmatte> I'm trying to break up an if statement though, so that might have something to do with it...
[09-May-2011 16:12:18] <rmatte> if ((($note =~ m/^threshold of Memory exceeded/i) || \
[09-May-2011 16:12:18] <rmatte> ($note =~ m/^Unable to read processes on device/i) || \
[09-May-2011 16:12:18] <rmatte> ($note =~ m/^threshold of process/i) || \
[09-May-2011 16:12:18] <rmatte> ($note =~ m/^threshold of CPU/i) || \
[09-May-2011 16:12:20] <rmatte> etc...
[09-May-2011 16:13:23] <rmatte> I'm going to fool around, see if I can get it to work
[09-May-2011 16:14:27] <rmatte> actually you know what?  perl uses brackets and semicolons
[09-May-2011 16:14:31] <rmatte> it probably doesn't even need the \
[09-May-2011 16:14:44] <rmatte> I'll try without it
[09-May-2011 16:16:02] <rmatte> yup, that works
[09-May-2011 16:18:25] <Hackman238> rmatte: Yeah since its inside a if statement like that I can see that confusing the interpreter
[09-May-2011 16:19:35] <rmatte> yeh
[09-May-2011 16:19:57] <rmatte> well, it doesn't matter if you spread anything out across multiple lines since lines have to end in a semicolon in perl
[09-May-2011 16:20:08] <rmatte> unlike bash or python
[09-May-2011 16:20:17] <Hackman238> rmatte: true true
[09-May-2011 16:22:47] <Hackman238> rmatte: ugh...I've been fighting Avalon all day now. Looks like I might finally have successful migration from V3.1 migrated from a highly customized V2.5.5 install.
[09-May-2011 16:23:10] <rmatte> cool
[09-May-2011 16:23:26] <rmatte> is 2.5.5 some enterprise release?
[09-May-2011 16:23:30] <rmatte> highest core goes to is 2.5.2
[09-May-2011 16:24:53] <Hackman238> rmatte: Nope, typo. Meant 2.5.2. Went from 2.5.s > 3.1 > 3.7.1 (Avalon). WebUI works...objects gone.
[09-May-2011 16:25:00] <Hackman238> faffi: 'tis a fail
[09-May-2011 16:25:03] <rmatte> ah cool
[09-May-2011 16:25:12] <rmatte> so far I've gone from 2.3 -> 2.4 -> 2.5...
[09-May-2011 16:25:21] <rmatte> working towards an upgrade to 3.x
[09-May-2011 16:25:45] <rmatte> The upgrades from 2.4 to 2.5 took me 7 hours to do
[09-May-2011 16:25:58] <rmatte> so I'm anticipating it to be even longer to go from 2.5 to 3
[09-May-2011 16:26:06] <Hackman238> rmatte: yeah, 2.5.X -> 3.X can go very very bad very easy
[09-May-2011 16:26:13] <Hackman238> rmatte: way longer
[09-May-2011 16:26:35] <rmatte> well, I had some 2.4 to 2.5 upgrades go really bad, but I eventually managed to sort them out
[09-May-2011 16:26:49] <rmatte> the beauty of having vms with snapshots is that if it screws up I just roll back and give it another go
[09-May-2011 16:26:58] <Hackman238> rmatte: Absolautely :-)
[09-May-2011 16:27:08] <rmatte> honestly though, 2.5.2 is running so damn smooth I'm reluctant to upgrade
[09-May-2011 16:27:32] <rmatte> knowing that I'll be jumping right back in to the fire
[09-May-2011 16:27:35] <Hackman238> rmatte: The big problem is going from py 2.4.X to py 2.6.x....when the upgrade happens 'extra' stuff some times gets left behind that blows everything up
[09-May-2011 16:27:54] <rmatte> 2.3 was kind of ok, 2.4 was a nightmare, 2.5.0 was a horrid release, but 2.5.2 is great (after applying a ton of patches to it)
[09-May-2011 16:27:55] <Hackman238> rmatte: I hear that.
[09-May-2011 16:28:04] <rmatte> ah
[09-May-2011 16:28:21] <Hackman238> rmatte: When I say sometimes I mean always
[09-May-2011 16:28:28] <Hackman238> LOL
[09-May-2011 16:28:29] <rmatte> lol
[09-May-2011 16:29:06] <rmatte> well, I did do a test upgrade of a copy of our Zenoss VM template from 2.5.2 to 3.0, and it worked fine
[09-May-2011 16:29:17] <rmatte> and all of my VMs are based off that image
[09-May-2011 16:29:23] <rmatte> so I expect it to go fairly smoothly
[09-May-2011 16:29:48] <rmatte> the main issue that I have which I haven't had time to finish dealing with is the LDAP auth in 3.x
[09-May-2011 16:30:03] <rmatte> haven't quite gotten it working
[09-May-2011 16:30:14] <rmatte> was pretty close but had to drop it to work on other stuff and haven't been back to it since
[09-May-2011 16:31:23] <Hackman238> rmatte: Do you plan to migrate from 2.5.2 to a clean 3.1 or upgrade the 2.5.2?
[09-May-2011 16:31:36] <rmatte> upgrade
[09-May-2011 16:31:46] <rmatte> I can't do migration, I have 16 Zenoss servers
[09-May-2011 16:32:07] <Hackman238> rmatte: Just be sure to plan plenty of downtime
[09-May-2011 16:32:28] <rmatte> well, any specifics that you can point out?
[09-May-2011 16:32:38] <Hackman238> rmatte: Two words: chef || puppet
[09-May-2011 16:32:54] <rmatte> nah, I'm comfortable doing it by hand
[09-May-2011 16:32:55] <faffi> Hackman238: what
[09-May-2011 16:33:28] <Hackman238> faffi: Sorry, typo + tab + autofill = faffi
[09-May-2011 16:33:32] <faffi> np
[09-May-2011 16:33:38] <Hackman238> rmatte: Yes. Uninstall all your packs including the core.
[09-May-2011 16:33:59] <rmatte> Hackman238: oh, I did the upgrade with them all still installed and it worked fine
[09-May-2011 16:34:00] <Hackman238> rmatte: then mv $ZENHOME/lib -> lib.old
[09-May-2011 16:34:12] <rmatte> The only pack that completely hosed the upgrade was Egor's WMI pack
[09-May-2011 16:34:15] <rmatte> but I know how to fix that
[09-May-2011 16:34:37] <rmatte> ok, moving the lib directory is noted
[09-May-2011 16:34:58] <rmatte> why uninstall all packs?
[09-May-2011 16:34:58] <Hackman238> then mv $ZENHOME/{a} -> $ZENHOME/{a}.old a = {lib, include, share, Prodcuts}
[09-May-2011 16:35:02] <rmatte> I have too many packs to do that
[09-May-2011 16:35:18] <rmatte> plus uninstalling packs removes device classes and any devices in those classes
[09-May-2011 16:35:23] <rmatte> so it's not a preferred thing to do on a prod box
[09-May-2011 16:35:33] <Hackman238> rmatte: Its the move from py 2.4.X -> 2.6.X that'll get you if you haven't
[09-May-2011 16:35:46] <Hackman238> rmatte: Absolutely understand.
[09-May-2011 16:35:48] <rmatte> get me how?
[09-May-2011 16:35:57] <rmatte> I didn't notice any issues in my test upgrade
[09-May-2011 16:36:21] <rmatte> If you can point out specific things to test
[09-May-2011 16:36:24] <Hackman238> rmatte: A copy of your production 2.5.2 upgraded to v3 no sweat?
[09-May-2011 16:36:26] <rmatte> then I will
[09-May-2011 16:36:30] <rmatte> correct
[09-May-2011 16:37:11] <Hackman238> rmatte: Seriously? I've never seen any production 2.5.X upgrade to 3.X without complications.
[09-May-2011 16:37:12] <rmatte> I removed a couple of ZenPacks which I knew absolutely would not work in 3.x because of them havng just 2.5 skins, but other than that, 95% of my packs were left in and they worked fine.
[09-May-2011 16:37:48] <Hackman238> rmatte: Core or Ent?
[09-May-2011 16:37:51] <rmatte> Core
[09-May-2011 16:38:08] <Hackman238> rmatte: Maybe thats the differnce. I should test that
[09-May-2011 16:38:10] <rmatte> We're in talks with Zenoss right now to get Ent, but I don't know if we will or not
[09-May-2011 16:38:18] <rmatte> depends on whether or not they are willing to come down in price
[09-May-2011 16:38:25] <rmatte> since the price that they quoted us is pretty unrealistic
[09-May-2011 16:39:40] <Hackman238> rmatte: Yeah, the pricing is pretty nuts
[09-May-2011 16:44:57] <rmatte> I wish there was something like spacewalk for ubuntu (puppet and chef are a bit more than I need at the moment)
[09-May-2011 16:46:55] <Hackman238> rmatte: Im unfamilliar with spacewalk
[09-May-2011 16:48:24] <rmatte> It's a web GUI that allows you to manage packages across redhat hosts
[09-May-2011 16:48:40] <Hackman238> rmatte: I'll have to check that out
[09-May-2011 16:48:51] <Hackman238> ls
[09-May-2011 16:48:51] <rmatte> it's pretty cool
[09-May-2011 16:49:06] <rmatte> If they had a plugin for aptitude I'd jump right at it
[09-May-2011 16:49:22] <rmatte> someone was supposedly writing in aptitude support a few years back but nothing has come from it
[09-May-2011 16:50:04] <rmatte> ooooh, apt-dater looks interesting
[09-May-2011 16:52:28] <Hackman238> rmatte: I hate it when that happens.
[09-May-2011 16:52:55] <rmatte> yeh, don't hype it unless you plan to complete i
[09-May-2011 16:52:57] <rmatte> it*
[09-May-2011 16:52:58] <rmatte> lol
[09-May-2011 16:56:14] <Hackman238> LOL
[09-May-2011 17:06:10] kocolosk is now known as kocolosk|away
[09-May-2011 17:08:13] <Hackman238> Later all.
[09-May-2011 17:12:05] <rmatte> later
[09-May-2011 18:07:23] kocolosk|away is now known as kocolosk
[09-May-2011 19:13:52] kocolosk is now known as kocolosk|away
[09-May-2011 20:05:06] kocolosk|away is now known as kocolosk
[10-May-2011 00:00:01] [disconnected at Tue May 10 00:00:01 2011]
[10-May-2011 09:24:07] [connected at Tue May 10 09:24:07 2011]
[10-May-2011 09:24:23] [zenoss-logger (logger bot) has joined #zenoss]
[10-May-2011 09:24:39] <Hackman238> Gat0rvean: And is it high latency from the perspective of the zenoss collector?
[10-May-2011 09:25:02] <Gat0rvean> Hackman238: it is a RAD RiCi-4T1 device (converts a t1 to ethernet basically)
[10-May-2011 09:25:23] <Gat0rvean> Hackman238: no, not high latency
[10-May-2011 09:25:53] <Hackman238> Gat0rvean: Try lowering the zMaxOIDPerRequest to 20 and rerun the zenperf... test
[10-May-2011 09:27:14] <Hackman238> Gat0rvean: I know that sounds rediculous, but I've seen this problem before
[10-May-2011 09:27:39] <Hackman238> Gat0rvean: Rather, I've seen it fixed by lowering the OID request rate.
[10-May-2011 09:27:54] <Gat0rvean> Hackman238:  Change the zMaxOIDPerRequest via the web interface and re-run test via command line?
[10-May-2011 09:28:41] <Hackman238> Gat0rvean: Correct. If this device has its own device class, change the zProperty there. If its a one of a kind, change the zProperty on the device
[10-May-2011 09:29:42] <Gat0rvean> Hackman238:  Worked. awesome.  thanks a bunch
[10-May-2011 09:30:45] <Hackman238> Gat0rvean: No problem.
[10-May-2011 09:31:25] <Hackman238> Gat0rvean: If you need Zenoss consulting, Web services or hosting, feel free to contact me!
[10-May-2011 09:31:45] <Gat0rvean> Hackman238:  Do you have a site, or just your email?
[10-May-2011 09:32:06] <Hackman238> Hackman238: and Http://
[10-May-2011 09:32:39] <Gat0rvean> Hackman238:  Gotcha.  thanks.
[10-May-2011 09:33:10] <Hackman238> Gat0rvean: Anytime.
[10-May-2011 09:59:24] <jmp242> Hackman238: hey, do you happen to know what the MaxOIDs that the command line snmpwalk sends?
[10-May-2011 10:07:48] <Hackman238> jmp242: I dont, sorry
[10-May-2011 10:10:32] <Hackman238> jmp242: It's definately nothing like zenperfsnmp, however.
[10-May-2011 10:13:51] <Hackman238> jmp242: twisted breaks the tests up into blocks of maxoids where I think snmpwalk is serial stream and waits for reply rather than fetching.
[10-May-2011 10:30:33] <rmatte> zenperfsnmp basically sends MaxOIDs = 1 I'm pretty sure
[10-May-2011 10:30:41] <rmatte> It's not designed for performance
[10-May-2011 10:30:53] <rmatte> twisted's efficiency is insane
[10-May-2011 10:31:17] <Hackman238> rmatte: Zenperfsnmp chunks the requests as the value of maxoids
[10-May-2011 10:31:24] <rmatte> yeh
[10-May-2011 10:31:30] <rmatte> It also doesn't wait for replies
[10-May-2011 10:31:40] <rmatte> it just launches them out and collects them in whatever order that they return
[10-May-2011 10:31:52] <Hackman238> rmatte: yep.
[10-May-2011 10:32:12] <rmatte> and sorry, I meant to say "snmpwalk basically sends MaxOIDs = 1 I'm pretty sure"
[10-May-2011 10:32:48] <Hackman238> rmatte: gotcha. did you have a chance to test that pack?
[10-May-2011 10:33:07] <rmatte> one of my Zenoss instances collects 22800 OIDs in under 20 seconds
[10-May-2011 10:33:10] <rmatte> pretty decent
[10-May-2011 10:33:39] <uifjlh> I'm running the v. 3.1.0 virtual appliance, I'm unable to add a MIB, the system just jumps back to the "login page" any ideas ?
[10-May-2011 10:34:21] <rmatte> uifjlh: install the MIB from commandline
[10-May-2011 10:34:34] <rmatte> installing MIBs from the UI is not recommended
[10-May-2011 10:34:46] <rmatte> zenmib run -v10 WHATEVER.MIB
[10-May-2011 10:34:54] <rmatte> or to do a whole directory of MIBs...
[10-May-2011 10:34:54] <uifjlh> ah... that helps
[10-May-2011 10:34:57] <rmatte> cd /dir
[10-May-2011 10:35:04] <rmatte> zenmib run -v10 *
[10-May-2011 10:35:09] <Hackman238> rmatte: Thats pretty good, especially for virt. Twisted is highly efficient.
[10-May-2011 10:35:15] <uifjlh> let me give that a try ...
[10-May-2011 10:35:32] <rmatte> uifjlh: make sure you restart zenoss after installing the MIB, it won't start using newly installed MIBs until you do
[10-May-2011 10:36:05] <uifjlh> I'll try that & "report back"
[10-May-2011 10:36:54] <rmatte> k
[10-May-2011 10:40:54] <rmatte> Hackman238: any idea when you'll have the 2.5 version of v3 of the IPSLA pack ready for testing?
[10-May-2011 10:41:11] <Hackman238> rmatte: Tonight I'll put it up.
[10-May-2011 10:41:19] <rmatte> Hackman238: Awesome, thanks.
[10-May-2011 10:41:25] <Hackman238> rmatte: Np.
[10-May-2011 10:41:38] <rmatte> I'll be able to do some Major testing with that one
[10-May-2011 10:42:03] <rmatte> Which reminds me, I'll install the 3.x version right now and give it a whirl
[10-May-2011 10:43:07] <Hackman238> rmatte: Thanks. Be sure to remove the old one, if installed, zopectl restart, zenhub restart, zenmodeler restart, zensla start
[10-May-2011 10:43:28] <rmatte> Will do
[10-May-2011 10:43:37] <rmatte> I just restart all of Zenoss hehe
[10-May-2011 10:43:48] <Hackman238> rmatte: LOL That works.
[10-May-2011 10:44:01] <rmatte> Usually when I restart just zope everything else divebombs and I have to restart it all anyways
[10-May-2011 10:44:06] <rmatte> so I just do the whole shebang
[10-May-2011 10:45:19] <Hackman238> rmatte: I hear that.
[10-May-2011 10:50:13] <rmatte> I'm probably going to use your daemon code as a guideline for creating a daemon for my formula datasource pack
[10-May-2011 10:50:21] <rmatte> since it needs one desperately
[10-May-2011 10:50:40] <rmatte> I tried using the WMI pack as a guideline, but that daemon code is nuts
[10-May-2011 10:50:50] <Hackman238> rmatte: Whats your pack do?
[10-May-2011 10:51:29] <rmatte>
[10-May-2011 10:51:39] <rmatte> "The Formula Data Source ZenPack allows you to create a new data source in a template, that contains multiple other existing data sources added up, multiplied, divided. This functionality is not provided in Zenoss by default. Now, you might for example add up two types of memory and create a threshold on the new formula datapoint. The threshold would go off when both of the added memory types go over a certain amount."
[10-May-2011 10:51:54] <rmatte> Examples:
[10-May-2011 10:51:54] <rmatte> 
[10-May-2011 10:51:54] <rmatte> $laLoadInt1 + $laLoadInt15 / $laLoadInt5 + sqrt(1000)
[10-May-2011 10:51:54] <rmatte> 
[10-May-2011 10:51:54] <rmatte> $laLoadInt1_laLoadInt1 + $laLoadInt15_laLoadInt15 / $laLoadInt5_laLoadInt5 + sqrt(1000)
[10-May-2011 10:51:54] <rmatte> 
[10-May-2011 10:51:55] <rmatte> $laLoadInt1_laLoadInt1+$laLoadInt15_laLoadInt15/$laLoadInt5_laLoadInt5+sqrt(1000)
[10-May-2011 10:51:55] <rmatte> 
[10-May-2011 10:51:56] <rmatte> $laLoadInt1 + ($laLoadInt15 / $laLoadInt5) + sqrt(1000)
[10-May-2011 10:52:05] <Hackman238> rmatte: Ah. For only snmp?
[10-May-2011 10:52:13] <rmatte> no, works for anything
[10-May-2011 10:52:21] <rmatte> reads from RRD files
[10-May-2011 10:52:28] <Hackman238> rmatte: Very cool
[10-May-2011 10:52:40] <Hackman238> rmatte: But very demanding. Does need a daemon LOL
[10-May-2011 10:52:49] <rmatte> yeh, the only issue is that it's executing a script right now to do the calculation...
[10-May-2011 10:52:59] <rmatte> so you end up with the multiple threads choking out the system issue
[10-May-2011 10:53:16] <rmatte> making a daemon that could queue up requests then perform them all under one thread would be much better
[10-May-2011 10:53:25] <Hackman238> rmatte: Yep. Competely agreem
[10-May-2011 10:53:32] <Hackman238> rmatte: Let me know if you need help.
[10-May-2011 10:53:38] <rmatte> Thanks, I appreciate it.
[10-May-2011 10:54:07] <rmatte> Make sure you release the new version of your pack before the spring into bling contest is over so that you get some free stuff lol
[10-May-2011 10:54:34] <Hackman238> rmatte: Pro Tip: Zenoss changes stuff way often- resuse existing routines, methods, etc. Don't reinvent the wheel because you'll need to do it again when the next sub release / release comes out and changes two lines.
[10-May-2011 10:55:08] <rmatte> yeh, I try to use existing routines as much as possible
[10-May-2011 10:55:13] <Hackman238> rmatte: LOL I'll try to release it in the next few weeks,
[10-May-2011 10:55:33] <rmatte> yeh that should be fine, the contest should run until Junish
[10-May-2011 10:55:33] <Hackman238> rmatte: I only say that because so many patches end up being issued.
[10-May-2011 10:56:10] <rmatte> Yeh, I'm the patch king, I have text files with patches to apply after install of 2.4 and 2.5
[10-May-2011 10:56:11] <rmatte>
[10-May-2011 10:56:50] <Hackman238> rmatte: Ex: if a patch modifies SnmpPerfConf, say to tweak caching or fix some obsure problem, it won't break my code. However, a custom written Perf fetcher might not work without tweaking.
[10-May-2011 10:57:07] <rmatte> yeh
[10-May-2011 10:57:44] <Hackman238> rmatte: well thats the idea anyway. We'll see how it works in practice with zenoss :-)
[10-May-2011 10:57:46] <rmatte> The only daemon that I wrote for Zenoss doesn't actually follow any of the standard Zenoss daemon code at all...
[10-May-2011 10:57:50] <rmatte> It just emulates parts of it
[10-May-2011 10:57:54] <rmatte>
[10-May-2011 10:58:03] <rmatte> The code is also super messy since I was just learning python...
[10-May-2011 10:58:12] <rmatte> and was piecing it together as I went along (deadlines)
[10-May-2011 10:58:25] <rmatte> I need to re-code it and break it all up in to functions
[10-May-2011 10:59:48] <Hackman238> rmatte: Looks good, but yes, should be broken up a bit :-)
[10-May-2011 11:00:47] <Hackman238> rmatte: The way we ticket from zenoss is to have the CRM directly look in the events database, condition and ticket.
[10-May-2011 11:01:31] <Hackman238> rmatte: Has it a pack?
[10-May-2011 11:01:48] <Hackman238> rmatte: Rather, has it been made in to a zenpack?
[10-May-2011 11:02:14] <rmatte> Yeh, I deploy it as a ZenPack
[10-May-2011 11:02:23] <Hackman238> rmatte: Nice
[10-May-2011 11:02:24] <rmatte> it's very custom though, so useless for public consumption
[10-May-2011 11:04:39] <Hackman238> rmatte: it would be useful to anyone trying to replicate the same concept with no where to start
[10-May-2011 11:07:11] <Hackman238> rmatte: did my pack install without event?
[10-May-2011 11:07:26] <rmatte> yessir
[10-May-2011 11:07:36] <rmatte> just adding a device to the lab box right now to test with
[10-May-2011 11:08:38] <Hackman238> rmatte: I ended up having the GNS3 emulate a box and bridge it to my loop to test the pack LOL
[10-May-2011 11:08:48] <rmatte> lol nice
[10-May-2011 11:11:26] <davetoo> Wha'ts a GNS3?
[10-May-2011 11:11:41] <rmatte> graphical network simulator
[10-May-2011 11:11:53] <davetoo> ??
[10-May-2011 11:11:54] <Hackman238> davetoo: GNS3 is a app using qemu and dynamips to enumate Cisco gear.
[10-May-2011 11:11:55] <davetoo> WANT
[10-May-2011 11:11:59] <rmatte>
[10-May-2011 11:12:00] <davetoo>
[10-May-2011 11:12:35] <Hackman238> davetoo: For a brief few hours I had two cisco 6509's running in my living room...then dynamips crashed. LOL
[10-May-2011 11:12:45] <rmatte>
[10-May-2011 11:12:52] <davetoo> Will it run NX-OS images?
[10-May-2011 11:13:25] <Hackman238> davetoo: I dont know to be honest. Download it and give it a shot.
[10-May-2011 11:13:47] <davetoo> Bookmarked.  When I have time I'll try it.
[10-May-2011 11:15:06] <rmatte> Hackman238: just a bit of feedback... I see Type: 1, but I don't see any translation of that
[10-May-2011 11:15:21] <rmatte> also, your table referencing types in the v2 of the pack was wrong
[10-May-2011 11:16:43] <Hackman238> rmatte: when I wrote that part I couldnt find good documentation. Do you have a correct table?
[10-May-2011 11:16:52] <rmatte> Yeh, one sec
[10-May-2011 11:17:11] <Hackman238> rmatte: If you tail -n 200 -f $ZENHOME/log/zensla.log, is it collecting?
[10-May-2011 11:17:25] <rmatte>
[10-May-2011 11:17:37] <rmatte> That's based on the Cisco MIB
[10-May-2011 11:17:57] <Hackman238> rmatte: I'll roll that in.
[10-May-2011 11:18:04] <rmatte>
[10-May-2011 11:18:07] <Hackman238> rmatte: TY
[10-May-2011 11:18:12] <rmatte> np
[10-May-2011 11:18:27] <rmatte> but yeh, is there a spot in 3.x where you can actually see that?
[10-May-2011 11:18:38] <rmatte> it doesn't show up as a column in the actual list
[10-May-2011 11:18:46] <rmatte> might be an idea to add it in if possible
[10-May-2011 11:18:51] <Hackman238> rmatte: Not yet, I haven't finshed all that,
[10-May-2011 11:18:56] <rmatte> ah ok
[10-May-2011 11:19:18] <rmatte> I might take a stab at it myself and pass the code on to you, I can't imagine it would be too difficult
[10-May-2011 11:19:23] <Hackman238> rmatte: All the infos provided in the past will be rolled back in.
[10-May-2011 11:19:34] <Hackman238> rmatte: Go for it, can't hurt.
[10-May-2011 11:19:50] <rmatte> Your VoIP template is going to need to be renamed to Jitter after you add the official Cisco definitions in
[10-May-2011 11:21:54] <Hackman238> rmatte: Yeah. The GUI stuff I haven't reimplemented in this version yet. I have another version of the GUI in my old V3, but its not portable to V4.
[10-May-2011 11:22:23] <Hackman238> rmatte: Trying to code everything compatible for V2,3 and 4 all at once :-)
[10-May-2011 11:23:09] <rmatte> nice
[10-May-2011 11:23:52] <Hackman238> rmatte: When done with reimplementation should have options to remotely set SLA's and all sorts of jaz
[10-May-2011 11:24:02] <rmatte> nice
[10-May-2011 11:24:08] <rmatte> so actual management you mean?
[10-May-2011 11:24:17] <Hackman238> rmatte: Yep.
[10-May-2011 11:24:21] <rmatte> very cool
[10-May-2011 11:24:34] <rmatte> that's one thing that isn't done very often with Zenoss, management stuff
[10-May-2011 11:24:58] <Hackman238> rmatte: Though management usually doesnt like device  management like so. I tell them that if I can snmpwalk it- I can break it. It matters not if I can snmp write to it
[10-May-2011 11:25:09] <rmatte>
[10-May-2011 11:25:43] <rmatte> I don't understand writing via SNMP enough to do any management like that
[10-May-2011 11:27:02] <Hackman238> rmatte: The very concept is not understood by managers far above, however. Them:'Why would we DoS our own device?'  Me:'The same reason we would snmp write malicious items- to break things. But we don't break our own stuff- thats for Anon. to do!'
[10-May-2011 11:27:43] <rmatte>
[10-May-2011 11:28:10] <rmatte> does your zensla daemon use twisted?
[10-May-2011 11:28:36] <Hackman238> rmatte: I meant that excluding RS. Fortunately the managers, virtually all the way up the chain, are on the ball here. OFT is particularly mismanaged.
[10-May-2011 11:28:41] <Hackman238> rmatte: It does.
[10-May-2011 11:28:45] <rmatte> nice
[10-May-2011 11:30:07] <Hackman238> rmatte: Is the daemon collecting correctly?
[10-May-2011 11:30:21] <rmatte> seems to be working fine as far as I can see
[10-May-2011 11:30:41] <Hackman238> rmatte: It also implements all the jaz the other zenoss daemons do- you can zensla help and zensla run now -d yourdevice
[10-May-2011 11:41:59] <rmatte> cool
[10-May-2011 12:00:01] [disconnected at Tue May 10 12:00:01 2011]
[10-May-2011 12:00:02] [connected at Tue May 10 12:00:02 2011]
[10-May-2011 12:00:22] [zenoss-logger (logger bot) has joined #zenoss]
[10-May-2011 12:16:08] <diatonic_> I just upgraded a 3.01 32-bit CentOS stack installed zenoss box to 3.1 and got an error at the end about zenperfwmi missing. Any ideas where to grab it?
[10-May-2011 12:16:41] <rmatte> zenperfwmi is part of Egor's WMI Data Source pack
[10-May-2011 12:16:49] <rmatte> It's available on the ZenPacks page
[10-May-2011 12:16:49] <Hackman238> diatonic_: Reinstall the pack
[10-May-2011 12:16:59] <diatonic_> Hackman238, thanks
[10-May-2011 12:17:24] <Hackman238> Hackman238: Make sure you restart the entire stack afterwards.
[10-May-2011 12:17:43] <Hackman238> rmatte: I forgot to ask you- do you use the rpm or stack?
[10-May-2011 12:17:56] <rmatte> stack
[10-May-2011 12:18:29] <Hackman238> rmatte: Interesting. This makes me wonder if the upgrade nightmare is caused by RPM not cleaning up
[10-May-2011 12:18:37] <rmatte> could be
[10-May-2011 12:18:39] <Hackman238> rmatte: Since I only use RPM.
[10-May-2011 12:18:43] <rmatte> I find stack to be incredibly problem free
[10-May-2011 12:18:57] <rmatte> plus since it's stand-alone, other system packages don't affect it
[10-May-2011 12:19:11] <Hackman238> rmatte: I've heard that, but it's hard to sell customers on.
[10-May-2011 12:19:28] <rmatte> I can imagine
[10-May-2011 12:19:47] <Hackman238> rmatte: thanks for the insight.
[10-May-2011 12:21:23] <rmatte> no prob
[10-May-2011 12:26:24] <diatonic_> rmatte, Thanks to you too
[10-May-2011 12:44:00] <ReGiStRaS> hi all....I have a sidetrack question to ask...
[10-May-2011 12:44:05] <ReGiStRaS> does the Acer eRecovery Management support dual layer DVDs (DVD +R DL)?
[10-May-2011 12:44:51] <Hackman238> ReGiStRaS: not a clue. It'll write to them for sure, but I can't say if it'll create a DVD9 image to use them fully
[10-May-2011 12:45:24] <Hackman238> ReGiStRaS: I'd hit up Google or Acer support
[10-May-2011 12:46:10] <ReGiStRaS> they take a long time to respond
[10-May-2011 12:47:15] <Hackman238> ReGiStRaS: If you don't mine the potiential waste of one DVD DL, I'd just try it
[10-May-2011 12:47:24] <Hackman238> ReGiStRaS: *mind
[10-May-2011 12:50:27] <rmatte> yeh, might as well just try it
[10-May-2011 13:35:22] diatonic_ is now known as diatonic
[10-May-2011 15:36:18] <Hackman238> Been so quiet in here today. LOL
[10-May-2011 15:36:43] <rmatte> no kidding
[10-May-2011 15:38:49] <uifjlh> let me ask a question ...
[10-May-2011 15:39:57] <uifjlh> I did a bad (duh) thing ... I deleted the definetation for the ping command & forgot it was a global thing, not a object specific thing
[10-May-2011 15:40:37] <uifjlh> could you please let me know what that command line is ?
[10-May-2011 15:41:18] <uifjlh> I was trying to disable ping on one specific server that's firewalled...
[10-May-2011 15:41:46] <rmatte> you're supposed to disable ping by changing the zProperty zPingMonitorIgnore to True
[10-May-2011 15:42:14] <uifjlh> I know, I had someone lurking behind me asking an Exchange question ... uggg
[10-May-2011 15:42:31] <rmatte> one second, I'll find the command
[10-May-2011 15:42:35] <uifjlh> ty
[10-May-2011 15:42:44] <rmatte> Name: ping
[10-May-2011 15:42:50] <rmatte> Description: Is the device responding to ping?
[10-May-2011 15:42:56] <rmatte> Command: ping -c2 ${device/manageIp}
[10-May-2011 15:43:06] <rmatte> That's what you're looking for I assume?
[10-May-2011 15:43:15] <uifjlh> ty very much ... exactly
[10-May-2011 15:43:19] <rmatte> np
[10-May-2011 15:43:55] <uifjlh> I'm in a prototype environment & am still learning the way around this thing....
[10-May-2011 15:44:00] <rmatte>
[10-May-2011 15:44:12] <rmatte> We're mostly veterans in here, so it's a good spot to ask questions
[10-May-2011 15:44:46] <uifjlh> I like it, I'm getting good data... I gave a demo to my CIO, he likes it, now on to the team next ! 
[10-May-2011 15:45:18] <uifjlh> just watching ~ a dozen devices, running the VM in Virtual Box, on my laptop ... LOL
[10-May-2011 15:45:19] <Hackman238> uifjlh: demo to CIO before your team?
[10-May-2011 15:45:25] <Hackman238> LOL
[10-May-2011 15:45:32] <uifjlh> yes, get the buy in from top ...
[10-May-2011 15:46:02] <Hackman238> Gotcha. Large environment?
[10-May-2011 15:46:22] <uifjlh> this is the 1st place I've ever been that's been w/o a monitoring system, (besides the phone) ...
[10-May-2011 15:47:03] <uifjlh> but most of the IT folks here are from an OS/400 world... ooooeeeeeeeooooooooo
[10-May-2011 15:48:22] <Hackman238> Oh wow.
[10-May-2011 15:48:50] <Hackman238> Do you have a large environment to monitor?
[10-May-2011 15:49:38] <uifjlh> ~ 80 servers, 98% vm, a couple sans, 5 big vm hoses, ~dozen switches, ~50 printers, ~dozen HVAC devices ...
[10-May-2011 15:49:42] <uifjlh> vm hosts
[10-May-2011 15:50:09] <Hackman238> I'd suggest sticking with Zenoss core in that case
[10-May-2011 15:50:17] <uifjlh> that's the plan
[10-May-2011 15:50:28] <uifjlh> one location...
[10-May-2011 15:50:45] <uifjlh> 2 w/ the DR site, one big flat lan
[10-May-2011 15:51:03] <Hackman238> Yep.
[10-May-2011 15:51:26] <Hackman238> I think my smallest DC has 35K devices in zenoss.
[10-May-2011 15:51:31] <Hackman238> LOL
[10-May-2011 15:54:22] <rmatte> yeh, I wish I worked with you guys lol
[10-May-2011 15:54:29] <rmatte> would be a lot more fun
[10-May-2011 15:55:14] <Hackman238> rmatte: Where do you live? I have an opening.
[10-May-2011 15:55:23] <rmatte> I live in Canada
[10-May-2011 15:55:47] <Hackman238> rmatte: Oh. Well I don't know if you really want to come here then. LOL
[10-May-2011 15:55:50] <rmatte> I'm guessing I'd have to relocate? lol
[10-May-2011 15:56:21] <Hackman238> rmatte: Yeah, to San Antonio, Tx. We have relocation assistance, however, so the move is painless.
[10-May-2011 15:56:41] <rmatte> ah, I'd have to reallllly think that one over hehe
[10-May-2011 15:56:46] <Hackman238> rmatte: Except for the fact Texans refuse to use directionals on the highway, it's not bad here.
[10-May-2011 15:56:47] <rmatte> I like this city
[10-May-2011 15:57:15] <rmatte> You haven't seen bad driving until you've seen Quebec drivers
[10-May-2011 15:57:16] <Hackman238> rmatte: I just moved to SA from New York.
[10-May-2011 15:57:27] <rmatte> Yeh, I saw on your blog
[10-May-2011 15:57:36] <rmatte> looks like a nice spot
[10-May-2011 15:57:44] <Hackman238> rmatte: LOL Never been there before. Just never have the time for travel seems like.
[10-May-2011 15:58:02] <rmatte> I'm also in the process of saving up for a house... so moving there and having to rent would put a bit of a wrench in my savings
[10-May-2011 15:58:45] <Hackman238> rmatte: Yeah it's pretty nice. Inexpensive comparred to NY too. About $600/month cheaper than my office in Delmar.
[10-May-2011 15:58:47] <rmatte> the paperwork for a Canadian to work in the states is pretty crazy too, I'd have to prove that I've been working in IT for the past 5 years and have written proof... some of the companies that I've worked for in the past don't even exist anymore.
[10-May-2011 15:59:09] <rmatte> Yeh, I can imagine NY is really expensive
[10-May-2011 16:05:25] <uifjlh> you guys should come to Iowa ...
[10-May-2011 16:06:16] <uifjlh> good opportunities for IT, low cost of living, and plenty of political intrigue ...
[10-May-2011 16:06:23] <rmatte>
[10-May-2011 16:07:49] <uifjlh> not quite as interesting as Wisconsin, but close ...
[10-May-2011 16:08:11] <Hackman238> LOL
[10-May-2011 16:08:15] <uifjlh> YW
[10-May-2011 16:08:50] <Hackman238> I've driven through Iowa twice on my way to and from CA, but I didnt have an oppertunity to stop really.
[10-May-2011 16:10:17] <uifjlh> IDK, it's a nice place, no mountains, few earth quakes, hurricanes=0, tornadoes - we watch those from the porch!
[10-May-2011 16:11:08] <Hackman238> Front row seat to a tornado show? Count me in. LOL
[10-May-2011 16:11:27] <uifjlh> yea we are always in the top 10
[10-May-2011 16:11:55] <rmatte> That's one nice thing about living in Ottawa Canada, no tornadoes, no hurricanes, no volcanoes, no floods
[10-May-2011 16:12:10] <rmatte> The worst we get here are ice storms and they are really rare
[10-May-2011 16:12:13] <uifjlh> just metre's of snow
[10-May-2011 16:12:32] <rmatte> yeh, but we have an army of snow removal equipment, so it's not a big deal
[10-May-2011 16:13:12] <Hackman238> Ottawa sounds like NY with less political BS.
[10-May-2011 16:13:42] <rmatte> Ottawa is the political capital of Canada
[10-May-2011 16:13:46] <rmatte> trust me, tons of politics
[10-May-2011 16:13:57] <uifjlh> not far from Toronto ?
[10-May-2011 16:14:02] <Hackman238> NY is the politcial sh!t heap of the US.
[10-May-2011 16:14:07] <rmatte> It's about a 5 to 6 hour drive from Toronto
[10-May-2011 16:14:13] <Hackman238> It's terrible.
[10-May-2011 16:14:18] <rmatte> lol
[10-May-2011 16:14:37] <uifjlh> oh my ok, my brother lives in way up north NY state, ~30 minutes from Toronto ...
[10-May-2011 16:14:39] <Hackman238> Its not all bad, but NY has some terrible features.
[10-May-2011 16:14:41] <rmatte> Our parliament is in Ottawa, meaning the prime minister and other cabinet ministers (similar to senators)
[10-May-2011 16:14:54] <uifjlh> dorks, dorks, dorks
[10-May-2011 16:14:58] <rmatte> and we're right next to Quebec, so there's more politics
[10-May-2011 16:15:59] <rmatte> Then again being next to Quebec has it's advantages, they allow drinking at the age of 18 which was great when I turned that age, and they sell alcohol at every corner store... we're limited to beer stores and liquor stores here in Ontario.
[10-May-2011 16:16:02] <Hackman238> Do the cabinet ministers collect pay whilst instituting one temporary budget after another since they have no time to work out an actul budget?
[10-May-2011 16:16:22] <Hackman238> rmatte: Interesting.
[10-May-2011 16:16:39] <rmatte> Hackman238: no, they collect pay whilst plotting the next tax to drop on us
[10-May-2011 16:16:49] <Hackman238> LOL
[10-May-2011 16:17:03] <rmatte> they dropped a new "harmonized sales tax" on us last year which is absolutely killer
[10-May-2011 16:17:13] <rmatte> the pay it worked before is we had 2 taxes, federal, and provincial
[10-May-2011 16:17:17] <rmatte> they decided to combine them
[10-May-2011 16:17:31] <rmatte> the problem with that is that there is LOTS of stuff that we never paid provincial tax on before
[10-May-2011 16:17:40] <rmatte> and now after they've combined them, we do
[10-May-2011 16:17:50] <rmatte> so heating and electricity bills have gone up a lot
[10-May-2011 16:17:53] <rmatte> gasoline has gone up
[10-May-2011 16:17:55] <Hackman238> Oh wow.
[10-May-2011 16:17:56] <rmatte> etc...
[10-May-2011 16:18:16] <rmatte> We're the only province that has the tax right now, but they are going to be pushing it in to Quebec within the next few years
[10-May-2011 16:18:27] <rmatte> They tried it in British Columbia and they revolted big time.
[10-May-2011 16:18:37] <rmatte> But they have a law that allows them to vote stuff like that down.
[10-May-2011 16:18:39] <rmatte> We don't.
[10-May-2011 16:18:49] <Hackman238> Yeah I'm waiting for the US to lose reserve currency status. We'll see some crazy hikes in utility, gas and food costs pretty soon here as well.
[10-May-2011 16:19:15] <rmatte> yeh
[10-May-2011 16:19:24] <rmatte> we get taxed to shit here in Canada
[10-May-2011 16:19:25] <Hackman238> 450-500% increase at least
[10-May-2011 16:19:36] <rmatte> I'd say about 43% of my income goes towards taxes
[10-May-2011 16:19:48] <Hackman238> OMG.
[10-May-2011 16:19:56] <Hackman238> Thats crazy!
[10-May-2011 16:20:26] <rmatte> Well, for the free healthcare I can understand it... but I hate knowing that a lot of my tax money is going towards funding wellfare cases
[10-May-2011 16:21:17] <rmatte> I don't mind it going towards people who can't work... but when someone is just lazy and claims that they can't work, then gets a $700 check each month, plus gets to live in a 3 bedroom townhouse for $70/month, that annoys the crap out of me.
[10-May-2011 16:22:01] <Hackman238> rmatte: Yep.
[10-May-2011 16:36:13] <rmatte> cool
[10-May-2011 16:36:24] <rmatte> ignore, meant for privmsg
[10-May-2011 16:53:49] <rmatte> Unfortunately not, I have a little bit of Solaris experience, but nothing to write home about.
[10-May-2011 16:53:52] <rmatte> oops
[10-May-2011 17:04:42] <davetoo> heh
[10-May-2011 17:06:17] <davetoo> I hate when I do that
[10-May-2011 17:07:17] <rmatte>
[10-May-2011 17:07:21] <rmatte> I just forget to hit tab
[10-May-2011 17:11:11] <davetoo> Perhaps I should not have gotten up at 04:30 this morning
[10-May-2011 17:14:14] <ashp> Guys, I am trying to monitor a box that zenoss says already exists but I can't find it anywhere in the interface
[10-May-2011 17:14:25] <ashp> is there a way to link directly to a host by name so i can find out where it is
[10-May-2011 17:17:09] kocolosk is now known as kocolosk|away
[10-May-2011 17:38:10] <rmatte> ashp: yes...
[10-May-2011 17:38:47] <ashp> I couldn't figure out how, I can't find it with search or anything
[10-May-2011 17:38:49] <rmatte> http://your-zenoss:8080/zport/dmd/deviceSearchResults?query=YourDeviceName
[10-May-2011 17:38:53] <rmatte> try that
[10-May-2011 17:39:09] <ashp> will do, i'm on a train so if i vanish the signal dropped
[10-May-2011 17:39:22] <rmatte> k
[10-May-2011 17:40:01] <ashp> oh THERE it is
[10-May-2011 17:40:05] <rmatte>
[10-May-2011 17:40:13] <ashp> called something totally different in a different group
[10-May-2011 17:40:18] <ashp> i added it to Email and it showed up in Web
[10-May-2011 17:40:18] <rmatte> hehe
[10-May-2011 17:40:22] <ashp> which is something that has happened to me before
[10-May-2011 17:40:23] <rmatte> ah
[10-May-2011 17:40:24] <ashp> i don't know why :/
[10-May-2011 17:40:27] <rmatte> weird
[10-May-2011 17:40:35] <ashp> things end up in places i don't create them, but i suppose it's me
[10-May-2011 17:40:47] <rmatte> could be, could be buggy new interface too
[10-May-2011 17:40:50] <rmatte> you never know
[10-May-2011 17:40:53] <ashp>
[10-May-2011 17:41:05] <rmatte> I still don't 100% trust that new UI
[10-May-2011 17:41:06] <ashp> is 3.2 close? I've sort of lost track as I've been working on other stuff
[10-May-2011 17:41:18] <rmatte> It's probably going to be 3.5 or 4.0
[10-May-2011 17:41:20] <ashp> I am kind of bummed out with zenoss after my last ticket
[10-May-2011 17:41:23] <davetoo> was it in /Discovered?
[10-May-2011 17:41:24] <rmatte> Avalon
[10-May-2011 17:41:28] <ashp> the twill stuff sucks. :/
[10-May-2011 17:41:33] <ashp> davetoo: no, it was in /Web for some reason
[10-May-2011 17:41:43] <rmatte> what sucks about twill?
[10-May-2011 17:41:49] <ashp> twill can't understand outlook 2010's web access stuff
[10-May-2011 17:41:55] <ashp> it just fails when you try to hit the homepage
[10-May-2011 17:42:06] <ashp> I love the feature but wish you had something better in the product to do it
[10-May-2011 17:42:19] <rmatte> That's really not twill's fault, that's Microsoft's for making all of their crap dependant on IE proprietary features
[10-May-2011 17:42:35] <ashp> well, true
[10-May-2011 17:42:39] <ashp> but it works fine in chrome and firefox
[10-May-2011 17:42:45] <ashp> it's the javascript that seems to fuck with twill
[10-May-2011 17:42:53] <rmatte> maybe you just need to spoof the user-agent with twill?
[10-May-2011 17:43:00] <rmatte> oh
[10-May-2011 17:43:08] <ashp> I tried, didn't seem to help
[10-May-2011 17:43:21] <rmatte> lame
[10-May-2011 17:43:26] <ashp> All I know is they wanted me to monitor our outsourced exchange
[10-May-2011 17:43:36] <ashp> maybe there's a better way than just monitoring the webpage and smtp etc
[10-May-2011 17:44:29] <rmatte> you'd have to hunt around
[10-May-2011 17:45:06] <ashp> I should see if there's a zenpack for doing this because I've been putting it off due to laziness
[10-May-2011 17:45:59] <rmatte> well, besides monitoring via WMI I doubt it
[10-May-2011 17:46:27] <ashp> yeah, and as it's outsourced I don't think I can
[10-May-2011 17:46:46] <ashp> i'm not even sure monitoring smtp is very legitimate outside of ensuring the port answers
[10-May-2011 17:46:52] <ashp> pop and imap is probably easier
[10-May-2011 17:47:52] <rmatte> do you have Zenoss Enterprise?
[10-May-2011 17:48:11] <ashp> yes
[10-May-2011 17:49:16] <rmatte>
[10-May-2011 17:49:19] <rmatte> that's what you want
[10-May-2011 17:49:55] <ashp> Aha, I will have a play when I get home
[10-May-2011 17:50:05] <ashp> i'm going to run away as I'm tired of about 1 minute of latency on this connection!
[10-May-2011 17:50:06] * rmatte nods
[10-May-2011 17:50:12] <rmatte> ttyl
[10-May-2011 18:46:27] diatonic is now known as diatonic|away
[10-May-2011 20:03:44] kocolosk|away is now known as kocolosk
[10-May-2011 21:47:07] kocolosk is now known as kocolosk|away
[11-May-2011 00:00:01] [disconnected at Wed May 11 00:00:01 2011]
[11-May-2011 00:00:01] [connected at Wed May 11 00:00:01 2011]
[11-May-2011 00:00:20] [zenoss-logger (logger bot) has joined #zenoss]
[11-May-2011 06:24:42] <fragfutter> is there a timeframe for availability of native rpm for rhel6?
[11-May-2011 08:00:16] <kotique> what are groups& systems divisions are for if I can't bind a monitoring template to servers belonging to them?
[11-May-2011 08:01:00] <fragfutter> kotique: display purposes
[11-May-2011 08:01:06] <kotique> damn
[11-May-2011 08:01:11] <jmp242> and for alerting rules
[11-May-2011 08:01:40] <jmp242> Monitoring templates only bind to Device Classes (unless they're component templates and then it's confusing)
[11-May-2011 08:01:46] <kotique> i've got cpanel servers and at the same time servers that run specific software. i need to distinguish between them
[11-May-2011 08:01:58] <kotique> bind different templates.
[11-May-2011 08:02:08] <jmp242> so put them in different device classes
[11-May-2011 08:02:19] <kotique> 1 server can belong to 2 classes
[11-May-2011 08:02:23] <kotique> some may belong to 1 only
[11-May-2011 08:02:30] <jmp242> well
[11-May-2011 08:02:34] <kotique> it's achieved with groups
[11-May-2011 08:02:39] <jmp242> a device can only belong to one device class
[11-May-2011 08:02:42] <fragfutter> kotique: inherit. make a folder cpanel and inside this one a folder webserver.
[11-May-2011 08:03:15] <kotique> what if a server runs webserver, but not cpanel? and i have specific templates for webserver?
[11-May-2011 08:03:24] <fragfutter> kotique: then you have lost
[11-May-2011 08:03:41] <fragfutter> kotique: restructure your setup to fit your monitoring tools
[11-May-2011 08:03:49] <jmp242> Well, what I do is this
[11-May-2011 08:04:19] <jmp242> it's a bit clunky, but I would have /Devices/Server/cpanel
[11-May-2011 08:04:33] <jmp242> and /Devices/Server/webserver
[11-May-2011 08:04:40] <jmp242> and /Devices/Server/cpanel/webserver
[11-May-2011 08:04:49] <jmp242> for instance
[11-May-2011 08:04:56] <jmp242> and bind the appropriate templates in each
[11-May-2011 08:05:41] <kotique> yeah, i just have to make deep trees there
[11-May-2011 08:07:06] <kotique> and then come the cpanel xen domUs and pure cpanel machines
[11-May-2011 08:07:15] <kotique> and then plain xen domUs
[11-May-2011 08:09:49] <fragfutter> or you write a modeller that detects if it is a xen domU and starts monitoring accordingly.
[11-May-2011 08:15:21] <z0> Hi all
[11-May-2011 09:31:06] <dec3pti0n> hello
[11-May-2011 09:52:40] <subbu> Hi All i am unable to find the graph for one of the queries and in the logs i found the belwo error can any one please help me out in this
[11-May-2011 09:52:41] <subbu> 2011-05-11T09:45:47 ERROR RenderServer Failed to generate a graph
[11-May-2011 09:52:58] <subbu> Traceback:  File "/opt/zenoss/Products/ZenRRD/", line 151, in render
[11-May-2011 09:53:18] <subbu> rrdtool.graph(*gopts)
[11-May-2011 09:54:46] <kokey> it's not saying any more in debug mode?
[11-May-2011 09:58:12] <subbu> @kokey:i will paste in and will send please help me out i guess i cannot paste the entire thing here
[11-May-2011 09:59:16] <subbu> @Kokey:
[11-May-2011 09:59:20] <subbu> please find the error
[11-May-2011 10:01:36] <jmp242> It sounds like your data or RRD file is bad
[11-May-2011 10:01:51] <dec3pti0n> is there a way to stop zenoss from picking up snmp traps ?
[11-May-2011 10:02:01] <jmp242> I believe it's saying that you have a value 'none' - possibly from the datasource
[11-May-2011 10:02:09] <jmp242> that is a string, rather than an integer
[11-May-2011 10:02:22] <Simon4> dec3pti0n: stop zentrap ?
[11-May-2011 10:02:31] <dec3pti0n> hmm
[11-May-2011 10:02:35] <jmp242> I think this usually happens because the SNMP agent is sending a string 'none' instead of, say, '0'
[11-May-2011 10:06:02] <JohnnyNoc> anyone here using bigegor's DellMon zenpack?  had a question about monitoring RAID arrays
[11-May-2011 10:06:53] <JohnnyNoc> didn't know if i was missing something but, we had some disks go bad in an array and zenoss failed to report any warnings/alerts.  upon further inspection it doesn't look like Zenoss knows much of anything about this perc 5/i controller besides the fact it's an expansion card
[11-May-2011 10:29:29] <subbu> @Jmp242:Can u please tell me how to fix this error
[11-May-2011 10:48:13] kocolosk|away is now known as kocolosk
[11-May-2011 10:49:56] <sergeymasushko> hello. from time-to-time zenoss reports about IP Service check failure (e.g. 15-20 checks) on different devices for different ip services... what can cause this issue?
[11-May-2011 10:58:03] <JohnnyNoc> flakey network
[11-May-2011 10:58:08] <JohnnyNoc> :X
[11-May-2011 11:27:06] <sergeymasushko> JohnnyNoc: all the devices are in the same gray network in the same cabinet in the data center
[11-May-2011 11:27:38] <fragfutter> broken switch
[11-May-2011 11:27:42] <sergeymasushko> zenoss is also there...
[11-May-2011 11:28:12] <sergeymasushko> I don't think so... it is working perfectly for the rest
[11-May-2011 11:28:53] <sergeymasushko> the switch is cisco 3745
[11-May-2011 12:00:01] [disconnected at Wed May 11 12:00:01 2011]
[11-May-2011 12:00:02] [connected at Wed May 11 12:00:02 2011]
[11-May-2011 12:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[11-May-2011 12:12:19] diatonic|away is now known as diatonic
[11-May-2011 12:39:56] <dec3pti0n> this is wierd I have a device on the same network as 3 other new ones that have been picked up by running a discovery on the network. All devices are clones with all having the same snmpd configs and everything else
[11-May-2011 12:40:34] <dec3pti0n> but zenoss does not pick up this 4th one in the discovery and trying to add the device by itself seems to be hanging
[11-May-2011 12:40:50] <dec3pti0n> I can do a snmpwalk from the zenoss box to the device just fine
[11-May-2011 14:12:46] <subbu> Hi All I installed the WMI pkages and when added the device in /CIM/WMI/Win2000 when i am modeling  i am getting the error
[11-May-2011 14:13:02] <subbu> can anyone please help me out in this
[11-May-2011 14:15:07] <jmp242> Is the computer a Win2000 computer?
[11-May-2011 14:15:12] <jmp242> what is the error
[11-May-2011 14:21:40] <subbu> @Jmp242:the computer is windows2000
[11-May-2011 14:23:10] <fus10nx> Hi all - does anyone here use LogMeIn ?
[11-May-2011 14:23:10] <subbu> @Jmp242:I am getting this error
[11-May-2011 14:23:24] <subbu> can u please help me out in this
[11-May-2011 14:26:30] <jmp242> I use LogMeIn
[11-May-2011 14:27:08] <jmp242> subbu: looks like you've already got that IP address in Zenoss
[11-May-2011 14:27:19] <jmp242> it may be stuck ... search the forums on getting rid of ghost IPs
[11-May-2011 14:27:25] <fus10nx> jmp242: do you use the LogMeIn feature to send syslogs to a syslog server?
[11-May-2011 14:27:33] <jmp242> no, sorry I don't
[11-May-2011 14:27:41] <fus10nx> hmmmm ok
[11-May-2011 14:27:42] <fus10nx> thanks
[11-May-2011 14:27:50] <fus10nx> trying to figure out how to send syslog stuff to zenoss from logmein
[11-May-2011 14:28:14] <fus10nx> but not sure how to troubleshoot when it doesnt work
[11-May-2011 14:28:48] <fus10nx> i.e. does Zenoss work as a syslog server?
[11-May-2011 14:29:22] <subbu> @Jmp242:no i gave that ip address only for that device . even i searched with the IP address but it shows only this device
[11-May-2011 14:29:29] <subbu> Can you please help me out i nthis
[11-May-2011 14:30:24] <jmp242> zensyslog will work to recieve syslog messages
[11-May-2011 14:30:39] <fus10nx> really im trying to find a way to look at the windows event logs of machine but all through zenoss, is this even possible?
[11-May-2011 14:30:50] <jmp242> subbu, is this a clean zenoss install, or could you have used that IP address before?
[11-May-2011 14:31:00] <jmp242> fus10nx: sure, you can use Zenoss to watch the windows event logs
[11-May-2011 14:31:06] <jmp242> as long as it can keep up with the events
[11-May-2011 14:31:16] <jmp242> look into the zenwineventlog daemon
[11-May-2011 14:31:17] <subbu> yes its a clean zenoss install i never used it before
[11-May-2011 14:31:27] <jmp242> it will use WMI to query the event log
[11-May-2011 14:31:35] <jmp242> to a set minimum severituy
[11-May-2011 14:31:39] <fus10nx> but will it show you the historial event log as well ?
[11-May-2011 14:31:40] <jmp242> set by a zProperty
[11-May-2011 14:31:51] <jmp242> It's not going to read in existing event log entries that I know of
[11-May-2011 14:31:57] <fus10nx> gotcha
[11-May-2011 14:32:02] <jmp242> but it will keep events it sees as events in Zenoss History
[11-May-2011 14:32:26] <fus10nx> yea i think it does that already by default
[11-May-2011 14:32:29] <jmp242> subbu: Hmmm, no idea
[11-May-2011 14:32:43] <subbu> @Jmp242:Its happening only to the devices for Windows 2000 server only
[11-May-2011 14:34:03] <jmp242> Did you redact the IP in the log you posted?
[11-May-2011 14:38:55] <subbu> I didnt put the Ip in the Post
[11-May-2011 14:39:05] <subbu> do u want the Ip add?
[11-May-2011 14:39:28] <jmp242> no I just wondered if the IP was there
[11-May-2011 14:39:36] <jmp242> that then matches the error discussed in this post:
[11-May-2011 14:39:42] <jmp242>;jsessionid=D9A7161F7C3412354E4C5A62D98758E4.node0?tstart=-38
[11-May-2011 14:39:55] <jmp242> which shows that it is somehow a duplicated IP
[11-May-2011 14:42:13] <subbu> @Jmp242:I will try this
[11-May-2011 14:46:17] <subbu> @Jmp242:after following the Link can u please tell me where i need to check the logs , i checked in the zenping and zenmodeler as well and didnt found any error
[11-May-2011 14:48:10] <jmp242> It's right in the log you posted to pastebin
[11-May-2011 14:48:20] <jmp242> so just check for that IP address
[11-May-2011 14:49:04] <subbu> ok i will delete the device first and will search with the IP address
[11-May-2011 14:49:11] <subbu> then i can come to know the error
[11-May-2011 14:56:40] <subbu> @Jmp242:I had the Ip address Subnet but not that ip address
[11-May-2011 15:04:17] <jmp242> Hmmm, I don't know then
[11-May-2011 15:04:20] <jmp242> sorry
[11-May-2011 15:16:08] <subbu> @Hmp242:I tried in all the ways but no luck with this ]
[11-May-2011 15:16:32] <subbu> i have one query i have another zenoss services which is monitoring it through WMI
[11-May-2011 15:17:11] <subbu> both the zenoss servers are in the same Sub-net will it make any difference for this IP error ?
[11-May-2011 15:41:34] <jmp242> Are the zenoss servers set up as distributed collectors?
[11-May-2011 15:50:59] <subbu> no
[11-May-2011 15:52:18] <jmp242> Then it shouldn't matter as far as I know
[11-May-2011 15:54:46] <subbu> oh ok
[11-May-2011 15:55:59] <subbu> i even posted it on the forum i guess Bigegor will respond because he was teh one who created the zenpacks for WMI monitoring
[11-May-2011 15:56:01] <subbu>
[11-May-2011 17:20:13] kocolosk is now known as kocolosk|away
[11-May-2011 18:29:16] diatonic is now known as diatonic|away
[11-May-2011 20:01:52] kocolosk|away is now known as kocolosk
[11-May-2011 21:03:56] <dhopp> When monitoring linux servers what do people prefer?  SSH or SNMP?
[11-May-2011 21:20:18] <dhopp> Matt you actually there?
[11-May-2011 21:22:34] <mattray> dhopp: leaving in a minute
[11-May-2011 21:23:18] <dhopp> what's your thoughts on ssh vs snmp for linux monitoring?
[11-May-2011 21:23:33] <dhopp> is one better from a performance standpoint?
[11-May-2011 21:47:26] <kocolosk> Is it possible to apply zFileSystemSizeOffset to a subset of filesystems on a device?
[11-May-2011 21:48:30] <kocolosk> I'd like to set a value less than 1 only for '/' (all other FS on the server have no reserved blocks)
[11-May-2011 21:53:33] <mattray> dhopp: snmp is higher performance
[12-May-2011 00:00:01] [disconnected at Thu May 12 00:00:01 2011]
[12-May-2011 00:00:01] [connected at Thu May 12 00:00:01 2011]
[12-May-2011 00:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[12-May-2011 04:10:07] <zykes-> rmatte: around ?
[12-May-2011 08:27:04] <vahi> HI all can any one please help me how to trigger the alerts for CPU utilization , as its collecting the graph , i gave the right template but i am not getting any triggers for the cpu utilization . can any one please help me out in this
[12-May-2011 08:27:23] <JohnnyNoc> did you setup a threshold
[12-May-2011 08:29:02] <vahi007> yeah i setup the threshold as well
[12-May-2011 08:29:20] <JohnnyNoc> ok
[12-May-2011 08:29:28] <JohnnyNoc> well, is it hitting the minimum of what your threshold is set to?
[12-May-2011 08:29:45] <JohnnyNoc> otherwise set it to an artificially low level
[12-May-2011 08:30:35] <vahi007> no, currently the cpu utilization is 30 i gave the threshold as 10 so that i should get the trigger but i dont know why its not triggering the alerts
[12-May-2011 08:30:52] <vahi007> @johnnynoc:I even tried this as well
[12-May-2011 08:31:37] <vahi007> is there any thing wrong with this
[12-May-2011 08:31:58] <vahi007> the device class is /perf/cpu
[12-May-2011 08:32:13] <JohnnyNoc> sounds like an event class
[12-May-2011 08:32:43] <JohnnyNoc> in which case that's probably correct
[12-May-2011 08:32:51] <JohnnyNoc> are you sure you chose your datapoints correectly when setting up the threshold?
[12-May-2011 08:33:06] <vahi007> generally for the cpu utilization we should give the event class as /perf/cpu rght
[12-May-2011 08:33:11] <JohnnyNoc> yes
[12-May-2011 08:33:18] <vahi007> i chose the dataponts correctly
[12-May-2011 08:34:52] <vahi007> even i selected the datapoint as "cpuPetcentProcessorTime_cpuPercentProcessortime"
[12-May-2011 08:35:16] <JohnnyNoc> that sounds right
[12-May-2011 08:35:29] <vahi007> please find the datasource for the same as well
[12-May-2011 08:35:29] <JohnnyNoc> have you run zenperfsnmp manually to see what happens?
[12-May-2011 08:35:30] <vahi007> Preparing Command...
[12-May-2011 08:35:38] <JohnnyNoc> zenperfsnmp run -d <devicename> -v 10
[12-May-2011 08:35:41] <JohnnyNoc> for example
[12-May-2011 08:36:15] <vahi007> Executing command snmpwalk  -v1 against ATL631 (Encorre 1H)
[12-May-2011 08:36:24] <vahi007> SNMPv2-SMI::enterprises.9600. = Gauge32: 3
[12-May-2011 08:36:39] <JohnnyNoc> well
[12-May-2011 08:36:43] <vahi007> i will try to run zenperfsnmp now
[12-May-2011 08:36:46] <JohnnyNoc> if that's the OID question it's not hitting your threshold
[12-May-2011 08:36:48] <JohnnyNoc> which you said is 10
[12-May-2011 08:36:49] <JohnnyNoc>
[12-May-2011 08:36:51] <JohnnyNoc> set it to 0
[12-May-2011 08:37:01] <vahi007> now it decreased
[12-May-2011 08:37:11] <vahi007> before it was 30
[12-May-2011 08:39:31] <dubs554> i ran the snmpwalk
[12-May-2011 08:39:58] <dubs554> now can i put the threshold to 1 so that it should trigger the alert rght
[12-May-2011 08:41:25] <dubs554> ok i will try
[12-May-2011 08:41:30] kocolosk is now known as kocolosk|away
[12-May-2011 08:42:20] <dubs554> @jOHHNYnoc:I put it 0
[12-May-2011 08:42:58] <dubs554> even the graph is showing is getting critical and i didnt get the alert as well
[12-May-2011 08:43:12] <Simon4> dubs554: do you have an alerting rule set up?
[12-May-2011 08:43:23] <dubs554> yeah i have it
[12-May-2011 08:43:26] kocolosk|away is now known as kocolosk
[12-May-2011 08:43:37] <dubs554> atleast it should show on the dashboard rgth
[12-May-2011 08:43:49] <dubs554> even its not showing on the dashboard
[12-May-2011 08:45:31] <dubs554> can any one please help me out in this
[12-May-2011 08:46:05] <lorin> Hi all i am not getting the graphs for the devices which is monitoring through WMI .. Can any one please help me in this
[12-May-2011 08:47:07] <lorin> I am getting the below error as "User-supplied Python expression ((here.speed or 1e9) / 8 * .95) for maximum value caused error: ['ifHCInOctets_ifHCInOctets', 'ifHCOutOctets_ifHCOutOctets']"
[12-May-2011 08:47:25] <lorin> any help will be much appreciated'
[12-May-2011 09:59:50] <dawn2020> good morning everyone....we are using zenoss core 3.1 and it was working fine till now...but recently we noticed that CPU events are triggered for some reason...can anyone please help me on this one
[12-May-2011 10:00:24] <dawn2020> sorry...i mean CPU events are not triggered
[12-May-2011 10:02:05] <dubs554> @dawn2020:I am also facing the same issue
[12-May-2011 10:04:28] <dhopp> dawn2020: what do you mean?  Do you mean that server CPU is going above the set threshold but no event is happening?
[12-May-2011 10:05:34] <dawn2020> @dhopp:yes...memory alerts and other alerts seems to be working fine
[12-May-2011 10:05:46] <subbu> @Jmp242:as disscussed with u yesterday i when i ran Zenoss fixit i found the below error
[12-May-2011 10:05:47] <subbu> The /opt/zenoss/ZenPacks/easy-install.pth doesn't know about these ZenPack directories:
[12-May-2011 10:06:59] <dhopp> dawn2020: does the CPU graph for that server show a threshold?
[12-May-2011 10:07:15] <dawn2020> dhopp:yes it does
[12-May-2011 10:07:43] <dawn2020> i also tried chaging the thresholds values...still no luck
[12-May-2011 10:07:57] <dhopp> hrm
[12-May-2011 10:10:25] <kokey> dawn2020: you have ticked the 'enabled' box? ;-)
[12-May-2011 10:10:31] kocolosk is now known as kocolosk|away
[12-May-2011 10:11:26] <dawn2020> yes...they are enabled
[12-May-2011 10:11:45] <dawn2020> they were working two weeks back
[12-May-2011 10:11:55] <dhopp> what do you have your threshold set to?
[12-May-2011 10:12:06] <dhopp> is this via snmp, wmi, ssh?
[12-May-2011 10:12:11] <dawn2020> snmp
[12-May-2011 10:12:23] <dhopp> how do you have it configured?
[12-May-2011 10:14:05] <dawn2020> sorry i didnt get you
[12-May-2011 10:16:28] <dhopp> How is the threshold configured?  On what datapoint, are you looking for idle cpu dropping below a certain threshold, or for actual usage to go above?
[12-May-2011 10:18:26] <dawn2020> ohh...for windows boxes we are using the cpuPercentProcessorTime datapoint and for Linux boxes ssCPUSystem
[12-May-2011 10:19:09] <dhopp> what's the event class set to?  Should be /Perf/CPU I believe
[12-May-2011 10:19:29] <dawn2020> yes thats the one we are using
[12-May-2011 10:19:42] <dhopp> hrmph
[12-May-2011 10:21:12] kocolosk|away is now known as kocolosk
[12-May-2011 10:21:16] <dawn2020> i dont see any errors in the zenperfsnmp logs either
[12-May-2011 10:22:05] <nyeates> Hi all
[12-May-2011 10:22:20] <dhopp> hey nyeates
[12-May-2011 10:22:56] <dhopp> dawn2020 has a weird problem…CPU threshold alerts are not triggers…threshold is enabled and from what I can tell seems like it's setup correctly…zenoss 3.1.0..ideas?
[12-May-2011 10:23:10] <dhopp> err..threshold alerts are not triggering
[12-May-2011 10:24:07] <nyeates> glad you clarified, had a hard time reading that
[12-May-2011 10:25:03] davetoo is now known as dcarmean
[12-May-2011 10:25:14] <nyeates> dawn2020: so yoiur graph shows a threshold being broken, but no event happens?
[12-May-2011 10:25:28] <dawn2020> yes
[12-May-2011 10:25:31] <nyeates> does the graphs visibly show the thresh on it? as a horizontal line....
[12-May-2011 10:25:39] <dawn2020> yes it does
[12-May-2011 10:25:44] <nyeates> can u post to us'
[12-May-2011 10:25:54] <dawn2020> i varied the values too...but still no luck
[12-May-2011 10:25:55] <dawn2020> sure
[12-May-2011 10:25:58] <dpetzel> anyone bored enough to try and help a first time zenpack developer (i use the word developer loosely)
[12-May-2011 10:26:26] <nyeates> dpetzel: in 35 mins it is developer hr, and we will have some zenoss devs in here
[12-May-2011 10:26:33] <nyeates> keep those Qs for sure!
[12-May-2011 10:26:41] <dpetzel> ah sweet
[12-May-2011 10:26:45] <nyeates> dpetzel: what are you working on
[12-May-2011 10:26:59] <dpetzel> trying to get a super basic f5 pack for core
[12-May-2011 10:27:17] <dpetzel> is my dilema
[12-May-2011 10:27:25] <nyeates> im not familiar with what f5 is
[12-May-2011 10:27:31] <dpetzel> bigip load balancer
[12-May-2011 10:27:40] <dpetzel> there is an enterprise pack already, but nothing for core
[12-May-2011 10:27:52] <nyeates> ahh yes
[12-May-2011 10:28:35] <dpetzel> Im pretty sure Im only 1 or 2 "change this you dummy" and I'll have it working
[12-May-2011 10:30:01] <nyeates> gotcha, that is a long post....good detail....hopefully it can be digested in the dev hr
[12-May-2011 10:30:18] <dawn2020> Hi nyeates....please see the link
[12-May-2011 10:30:26] <lorin> @Nick:I am monitoring the devices through WMI i am not able to get the graphs for the same . can u please help me out in this
[12-May-2011 10:30:37] <dhopp> The distributed collector zenpack that Egor wrote…how does that work if the main zenoss (zenhub) goes down…if there is an event that happens is that lost?
[12-May-2011 10:31:23] diatonic|away is now known as diatonic
[12-May-2011 10:31:34] <lorin> when i click the grah getting the error as User-supplied Python expression ((here.speed or 1e9) / 8 * .95) for maximum value caused error: ['ifHCInOctets_ifHCInOctets', 'ifHCOutOctets_ifHCOutOctets']
[12-May-2011 10:35:11] <nyeates> others feel free to chime in, im helping dawn2020 right now
[12-May-2011 10:36:43] <nyeates> dawn2020: the x-axis.... 05, 06 07, etc...are those minutes? days?
[12-May-2011 10:36:51] <dawn2020> days
[12-May-2011 10:37:44] <nyeates> oh hmmm, can you soon in to show just the overage? like so the overage part, if its just 5 or 20 mins, zoom in to show just that
[12-May-2011 10:37:54] <Hackman238> dawn2020: Are you running zenhubs on the remote collectors?
[12-May-2011 10:38:09] <nyeates> i want to assure that it was sustained at that high amt, not jumping up an ddown
[12-May-2011 10:38:25] <dawn2020> @nyeates:sure
[12-May-2011 10:38:49] <dawn2020> @hackman23
[12-May-2011 10:38:54] <dawn2020> nope
[12-May-2011 10:39:44] <Hackman238> If there is an event and zenhub is down, there is no route from the daemons to mysql.
[12-May-2011 10:40:10] <dhopp> Hackman238: Was that question for dawn2020 meant for me?
[12-May-2011 10:40:51] <Hackman238> dhopp: Meant for you. Typo + {TAB} = wrong name :-)
[12-May-2011 10:40:58] <dhopp> hehe...
[12-May-2011 10:41:19] <Hackman238> dhopp: Do you run zenhub(s) on the remote collectors?
[12-May-2011 10:42:04] <dhopp> I haven't built it out yet…but we are building out a new zenoss infrastructure…my thought was to 3 zenoss machines.  1 "main" zenoss (primarily just the web interface) and then 2 remote collectors (1 in each of our DCs)
[12-May-2011 10:42:42] <dhopp> if we run zenhubs on the remote collector then all the collectors can't share a common device list right?
[12-May-2011 10:43:01] <fragfutter> i'm a bit lost. I want to write a modeler for a device, that detects hardware components. Should i write a new DeviceClass with it's own relations or do i only write my ComponentClass and extend the HWComponent._relations?
[12-May-2011 10:43:06] <Hackman238> dhopp: I'd reccomend dediated zenhubs on remote collectors
[12-May-2011 10:43:15] <dawn2020> @nyeates:
[12-May-2011 10:43:40] <Hackman238> dhopp: Sure they can. You don't want to run a seperate instance, only a dedicated zenhub.
[12-May-2011 10:43:40] <dhopp> Hackman238: ok…is that accomplished through Egor's zenpack or is that custom config?
[12-May-2011 10:44:33] <dhopp> also, how do you manage the device lists?  Is there a central interface? (I'm going to be using core unfortunately…budget for enterprise got shot down)
[12-May-2011 10:44:53] <Hackman238> dhopp: I generally do it all custom, you'll need to check with someone about that pack.
[12-May-2011 10:45:37] <Hackman238> dhopp: Zope is what manages the device objects, zenhub is a daemon which handles dolling out configs, scheduling, and event stuff.
[12-May-2011 10:45:52] <dhopp> ah ok
[12-May-2011 10:46:59] <Hackman238> dhopp: Personally I can't tell you what Egors pack can and cannot do. I'd PM him and ask.
[12-May-2011 10:47:20] <dhopp> so users can go to one interface to see all events that happen (assuming all zenhubs use the same db).  And then add templates through the same interface regardless of zenhub...
[12-May-2011 10:47:57] <dhopp> Hackman238: no problem…I can always install it too..just haven't gotten there...
[12-May-2011 10:48:00] <Hackman238> dhopp: Yes, you'd point each of your zenhubs at the same db.
[12-May-2011 10:48:26] <Hackman238> dhopp: Let me know if you need any help.
[12-May-2011 10:48:34] <dhopp> I'm sure I will :-)
[12-May-2011 10:49:23] <dhopp> Going to have ~750 servers to start and then will hopefully add ~300 network devices eventually
[12-May-2011 10:50:27] <dhopp> well in reality probably to start will be fewer servers then that…but that's what it should grow to
[12-May-2011 10:51:13] <Hackman238> dhopp: Always start small and work your way up. Jumping at something too large is almost certain failure.
[12-May-2011 10:51:25] <Hackman238> dhopp: Least thats how I look at stuff. :-)
[12-May-2011 10:51:29] <nyeates> dawn2020: let me get back to you....or ping me later....trying to get a developer lined up to come in here
[12-May-2011 10:52:07] <dawn2020> oh ok
[12-May-2011 10:53:23] <dhopp> Hackman238: So lets take my scenario..1 web interface for users to interact with, 2 "remote collectors" (each running their own zenhub).  What daemons would have to be running on each of the servers then?, would the web interface have everything just nothing really be utilized?
[12-May-2011 10:55:10] <Hackman238> dhopp: On the zope hostign box you could disable the perf daemons, but I would leave them on for self monitoring. Daemons on remote collectors would include the perf daemons you need, zenhub, zenmodeler, and zenrender.
[12-May-2011 11:00:39] <nyeates> I want to repeat a specific dev-heavy line of questioning from dpetzel....i told him to wait til other had come in
[12-May-2011 11:00:57] <nyeates> Dpetzel, you want to iterate a summary here? he posted a long version at
[12-May-2011 11:01:14] <Hackman238> nyeates: That reminds me, I've a few people testing the new IPSLA pack for V2.5.2/3.X,4 ;-)
[12-May-2011 11:01:36] <nyeates> hackman238: custs or core ppl?
[12-May-2011 11:01:55] <Hackman238> nyeates: Both.
[12-May-2011 11:02:10] <Hackman238> nyeates: Core compatible. Ent version includes remote management.
[12-May-2011 11:02:46] <dpetzel> sure thing, so basiacally Im compelety new to writing zenpacks (aside from creating from the gui)
[12-May-2011 11:02:51] <dcarmean> Egor's distributed collector pack (nominally) takes care of deciding which daemons run where.
[12-May-2011 11:03:06] <dpetzel> I'm trying to write a pack to list BigIP virtual servers as components of the BigIP device
[12-May-2011 11:03:37] <dpetzel> I've followed Janes Doc and the Developers guide, and things seem to be modeling correclty, but the Virtual Server to device relationships never seem to happen
[12-May-2011 11:03:39] <nyeates> dawn2020: i saw your 2nd graph. Very odd how the CPU % is obviously over the threshold and you got no events. Have you checked history and/or made ABSOLUTELY sure there is no transform or settings in the event class properties that would be dropping the event? Make sure there is not some rule way up the tree that is applying to many events
[12-May-2011 11:04:40] <nyeates> Hackman238: Very nice! Youll have to send me a copy to check it out
[12-May-2011 11:05:11] <dhopp> dcaremean: do you know if it starts zenhub? or does it rely on zenhub from the "main" zenoss?
[12-May-2011 11:05:15] <Hackman238> nyeates: Only RC3 right this moment. I'll PM you in a bit.
[12-May-2011 11:05:23] <dhopp> I'm working on setting up a second collector to test it with
[12-May-2011 11:05:37] <dhopp> if that doesn't work, I might need more help from Hackman238 :-)
[12-May-2011 11:05:57] <dawn2020> @nyeates: i did check that and we are not using any transforms for CPU
[12-May-2011 11:06:25] <dawn2020> the issue is not only for that device...its for all the devices
[12-May-2011 11:06:38] <dawn2020> windows as well as Linux
[12-May-2011 11:06:53] <nyeates> dpetzel: so im not familiar with how f5 big ip and its virtual servers relate.... but basically you are trying to make a new zenpack with components, and the components are not relating correctly back to the root device?
[12-May-2011 11:08:29] <nyeates> dawn2020: make a new device class from the root /Devices. Put a server in there. Make it go over threshold. See if event comes in
[12-May-2011 11:09:06] <nyeates> allows you to determine for sure that there isnt something blocking the events...unless of course its applying at root device lvl :-p
[12-May-2011 11:09:36] <dawn2020> sure will try it right away
[12-May-2011 11:14:31] <jplouis> dpetzel: is the zPythonClass set to BigipLtm on your device class?
[12-May-2011 11:15:13] <rocket> dpetzel: also an example of adding components is the community aix zenpack ..
[12-May-2011 11:15:32] <rocket> but thats more from the 2.5.2 timeframe
[12-May-2011 11:15:48] <nyeates> do components change big time for 3.x?
[12-May-2011 11:16:21] <rocket> no .. they themselves havent changed
[12-May-2011 11:16:36] <rocket> the ui elements for displaying them have somewhat .. eg the page templates etc
[12-May-2011 11:16:47] <rocket> the relationship code itself is still the same
[12-May-2011 11:17:00] <jplouis> dpetzel: when a device is added to a device class it looks at the zPythonClass zProperty to determine what type to instantiate.  By default the value is Products.ZenModel.Device.
[12-May-2011 11:17:13] <fragfutter> is it the right way to extend the DeviceHW._relations, or is it better to have a new deviceclass?
[12-May-2011 11:18:06] <rocket> fragfutter: how are you extending the relations?
[12-May-2011 11:18:26] <Dpetzel-mobile> Yes python class is set
[12-May-2011 11:18:29] <rocket> fragfutter: It would be better to have a new class such that you are not impacting other devices ..
[12-May-2011 11:18:53] <fragfutter> the zenpacks __init__ has something like DeviceHW._relations += (( "myrelation", ...))
[12-May-2011 11:19:15] <rocket> especially if you are an enterprise customer .. as we dont really like it when you modify core features etc
[12-May-2011 11:19:33] <rocket> yea .. while you can do that .. please dont ..
[12-May-2011 11:19:43] <fragfutter> rocket: ok. thats new to me.
[12-May-2011 11:20:30] <rocket> fragfutter: though its been about a year since I have done anything with it specifically.. I could be remembering wrongly
[12-May-2011 11:21:58] <Jane_Curry> Hiya....
[12-May-2011 11:22:08] <jplouis> dpetzel: the error in the forum, ApplyDataMap: no relationship:LtmVs found on:MyDeviceName (<class 'Products.ZenModel.Device.Device'>, makes it seem like the device is the wrong class.  Have you verified via zendmd that your device is an instance of the correct class
[12-May-2011 11:22:12] <fragfutter> rocket: i have a question of understanding then. If DeviceClass A adds hardware components hw_a. And DeviceClass B (written by someone else) has hw_b. How can i have a device that has both hw_a and hw_b? would i need to write a new DeviceClass C that inherits from both?
[12-May-2011 11:22:15] <Jane_Curry> I would suggest that the UI display has changed LOTS in V3
[12-May-2011 11:22:54] <dpetzel> jpluis, I agree it lead me to believe it was wrong class, I'll check zendmd now
[12-May-2011 11:22:56] <nyeates> Hi Jane. Cant wait to find out what yer many new ZPs do
[12-May-2011 11:22:58] <rocket> fragfutter: yes
[12-May-2011 11:23:17] <Jane_Curry> That would be telling.......
[12-May-2011 11:23:39] <fragfutter> rocket: then i would need stick to extending _relations. That way i can slap my modeller on ony device i see.
[12-May-2011 11:24:34] <rocket> fragfutter: you can do it .. its possible .. it just adds to object complexity and possibly performance issues with those larger objects down the road
[12-May-2011 11:24:35] <dpetzel> deviceClass <ToOneRelationship at deviceClass>
[12-May-2011 11:25:52] <dpetzel> when I look at the configuration properties for the oranizer "/network/f5" zPythonClass = ""
[12-May-2011 11:26:27] <fragfutter> rocket: but for example the HP Hardware monitoring by bigegor will detect Fans and add them to the device. If i now add a modeller to detect attached Monitors (stupid example, can't think of anything else). I would end up with a HP-Device, Monitor-Device, HP+Monitor-device
[12-May-2011 11:27:07] <Jane_Curry> OK.  I have done a fair-sized piece of work for ZenSystems in Denmark and they have graciously agreed to make them available to the community
[12-May-2011 11:27:15] <rocket> fragfutter: you can extend his class .. thats fine .. eg a suborganizer that extends his class
[12-May-2011 11:27:33] <rocket> however I wouldnt recommend modifying the top level one
[12-May-2011 11:28:02] pmcguire is now known as ptmcg
[12-May-2011 11:28:06] <rocket> dpetzel: did you remodel?
[12-May-2011 11:28:12] <fragfutter> rocket: with class i mean Products.ZenModel.DeviceClass
[12-May-2011 11:28:29] <rocket> is that what Egor is modifying?
[12-May-2011 11:28:29] <fragfutter> rocket: not the hierachy in the web/gui
[12-May-2011 11:28:40] <dpetzel> rocket: I have a few times, tried deleteing re-addding with no luck as well
[12-May-2011 11:29:01] <fragfutter> rocket:
[12-May-2011 11:29:12] <rocket> fragfutter: right .. I understand that .. but egor really should have Products.ZenModel.HpDeviceClass or something like that
[12-May-2011 11:29:19] <rocket> and then extended that device class
[12-May-2011 11:29:56] <rocket> otherwise you are modifying core objects for everything in zenoss ..
[12-May-2011 11:30:07] <fragfutter> rocket: and jane has BridgeDeviceClass. Whenever i have a server that is from HP and has a bridgetable i would need to decide what to do.
[12-May-2011 11:31:26] <rocket> fragfutter: I see your point .. my main issue is trying to keep you more upgrade safe etc ..
[12-May-2011 11:31:38] <Jane_Curry> Jolly good point fragfutter - what IS the recommended way forward when you want to use attributes from 2 different device classes??
[12-May-2011 11:32:35] <Hackman238> fragfutter:You can find, in the IPSLAv2 zenpack, a version 2.5.2 example of creating a new device class
[12-May-2011 11:33:15] <fragfutter> Hackman238: or in janes bridge example (for zenoss 3).
[12-May-2011 11:33:37] <rocket> Jane_Curry: its a tricky issue, as one view recommends adding more code via a new zenpack, vs less code but object complexity
[12-May-2011 11:33:53] <fragfutter> they questions is (as jane said) how to combine attributes from multiple classes
[12-May-2011 11:33:58] <Hackman238> fragfutter: yep, your right LOL
[12-May-2011 11:34:00] <Jane_Curry> But I don't want to have to create a whole new device class to use functionality that is already provided by 2 other ZenPacks,
[12-May-2011 11:35:25] <Jane_Curry> Has it suddenly gone quiet????  
[12-May-2011 11:35:31] <Hackman238> Thinking.
[12-May-2011 11:35:32] <Hackman238> LOL
[12-May-2011 11:35:33] <kokey> yes
[12-May-2011 11:35:37] <rocket> Jane_Curry: the flexibility is there to do as you suggest .. some of it can be accomplished via smart use of organizer nesting and zope acquisition .. I would believe.  jplouis probably has a better perspective on it than I
[12-May-2011 11:35:39] <Jane_Curry> ... hard....
[12-May-2011 11:36:00] <rocket> honestly I havent done engineering level zenpack work for a while, with the position I am in now.
[12-May-2011 11:36:06] <Hackman238> I agree with rocket, its just a matter of smart design.
[12-May-2011 11:36:13] <Jane_Curry> How about a sample posted to the wiki for us??????
[12-May-2011 11:37:14] <rocket> nick?  do you want to take that down as something to check into
[12-May-2011 11:37:31] <rocket> I am not sure if Chets example zenpack has anything like that in it
[12-May-2011 11:37:48] <rocket> Jane it gets tricky with different authors of zenpacks doing different things
[12-May-2011 11:38:08] <nyeates> Jane: I continue to work on convincing zenoss mgmt to up the amt of dev documentation. I have sent clear communications of this to our managment and #1 on the list was to get some cycles to put toward more detailed dev docs, so the community can fend for itself
[12-May-2011 11:38:18] <rocket> again I am not necessarily saying what you are doing is wrong.  I am just noting some of the other side effects that you may not have thought about
[12-May-2011 11:39:28] <dhopp> is there going to be an RH6 rpm of the next release?
[12-May-2011 11:40:38] <kokey> yeah
[12-May-2011 11:40:44] <kokey> somewhere in the next 6 and a half years
[12-May-2011 11:40:57] <nyeates> Jane, also, I want to have a side convo with you about Zenoss possibly taking in and vetting and officiating your dev docs. Right now, it is your stuff and we respect that, and we want to see if we can both take it to the next level. We can discuss offline.
[12-May-2011 11:41:07] <jplouis> There is no clear recommended way to use attributes and relationships from multiple devices.   It isn't problem zenoss has addressed
[12-May-2011 11:41:22] <rocket> dhopp: as far as I know no there will not be a RH6 rpm in our next immediate release ..
[12-May-2011 11:41:41] <dhopp> doh
[12-May-2011 11:41:46] <Jane_Curry> I guess what we are saying is that this stuff isn't easy!  But if we want the community to go forward, it's going to be snail-pace without input from dev
[12-May-2011 11:41:53] <rocket> dhopp: I believe our current strategy is to support RH6 when Centos6 is out ..
[12-May-2011 11:42:16] <rocket> 90% of our customers are using Centos vs Redhat
[12-May-2011 11:42:19] <fragfutter> rocket: you might want to start building your rpms using scientific linux
[12-May-2011 11:42:23] <Hackman238> I dont understand why not just create a device class, like SLADevice (in my pack), create a new device class organizer under say /Networks called whatever and set whatevers class to ZenPacks.yourpack.yourclass.
[12-May-2011 11:42:28] <rocket> so that is more of the driving force
[12-May-2011 11:42:36] <Jane_Curry> Thanks jp - that's what I thought but I could always hope that I had missed something
[12-May-2011 11:43:10] <dhopp> understood (we might use centos for this, we just have a patching process for rh)
[12-May-2011 11:43:39] <rocket> dhopp: we just dont have enough build resources to track that one yet
[12-May-2011 11:43:52] <dhopp> understood
[12-May-2011 11:43:55] <rocket> dhopp: its been discussed.  We just are not ready for it yet
[12-May-2011 11:44:33] <Jane_Curry> One thing that helps us all build ZenPack experience is lots of good examples - Egor has been a real trailblazer on this
[12-May-2011 11:44:36] <dawn2020> @nyeates: still the same....i created a new class 'test' with same
[12-May-2011 11:44:53] <dawn2020> datasources
[12-May-2011 11:45:10] <Jane_Curry> One of the ZenPacks that I have created for ZenSystems is for Juniper devices
[12-May-2011 11:45:26] <jplouis> Jane: unfortunately most of our good examples are in Enterprise
[12-May-2011 11:45:44] <rocket> jane did you look at chets example zenpack?
[12-May-2011 11:45:54] <rocket> chet was putting some of those examples in there
[12-May-2011 11:46:39] <kokey> I think better dev documentation is also important to give people guidance on quality
[12-May-2011 11:46:42] <Jane_Curry> I looked at it early on, but I haven't looked at it recently and I think it has been updated???  I'll go and grab it
[12-May-2011 11:47:00] <kokey> you want to avoid the nagios situation where there are a lot of plugins but most of them are terrible
[12-May-2011 11:47:06] <dawn2020> @nyeates: i tried putting in different event class and it worked
[12-May-2011 11:47:07] <rocket> I dont know when he has had time to update it ..
[12-May-2011 11:47:18] <Hackman238> kokey: Could not be more true
[12-May-2011 11:47:23] <dawn2020> any ideas why it is like tat
[12-May-2011 11:47:35] <Jane_Curry> Totally agreed kokey.  I build ZenPacks by copying other people's stuff but perhaps we are all passing on bad habits"???
[12-May-2011 11:48:13] <rocket> Jane_Curry: I think ultimately zenpacks will be addressed as we get more of the json api etc fleshed out ..
[12-May-2011 11:48:40] <rocket> as we start building more of these apis and consistent ways of doing things progress will be made in this area
[12-May-2011 11:48:48] <fragfutter> i would like to have an incremental tutorial along the lines of. Adding an SNMP Modller to discover new components. Adding a Javascript so it will display in the gui. Adding Stuff so it will collect perfdata. ...
[12-May-2011 11:49:48] <rocket> fragfutter: honestly so would I .. .. I havent had a chance to catch up on all of the new ui elements etc
[12-May-2011 11:50:28] <nyeates> dawn2020: i dont get yer last statement.... did you mean device class? if not, how did u go about doing what u are saying?
[12-May-2011 11:50:59] <rocket> but I have no idea when any of this will be done .. as a support engineer, I have little to no say in the direction this will take ... other than describe to management customer frustrations
[12-May-2011 11:51:20] <Jane_Curry> fragfutter - do you have a copy of my Creating ZenPacks doc?
[12-May-2011 11:51:46] <rocket> as you can tell I deal more with upgrades and stability as of late, vs new features and device support
[12-May-2011 11:51:49] <Jane_Curry> It isn't a tutorial but it does try to lay out the steps
[12-May-2011 11:52:13] <fragfutter> Jane_Curry: yes. but for me it's lacking the tutorial part.
[12-May-2011 11:52:19] <rocket> so you can see different perspectives for the same problem
[12-May-2011 11:52:59] <nyeates> kokey: interesting comment about nagios plugins
[12-May-2011 11:53:05] <Jane_Curry> <rocket> I hear what you say about hopefully things will get better but the V3 GUI has been out for nearly 12 months now and many of us are still struggling in the dark....
[12-May-2011 11:54:07] <nyeates> fragfutter: ive noted the tutorial idea for when we ever get dev documentation more serious; tutorials or step by step are great initial learning toosl
[12-May-2011 11:54:11] <Jane_Curry> Just looked at the page for Chet's example ZenPack - not very happy to see that Zenoss 2.3!!!! is a pre-req
[12-May-2011 11:54:36] <Hackman238> Jane_Curry: Why not just unzip it and view the code?
[12-May-2011 11:54:59] <Hackman238> Jane_Curry: I'm sure it'll all work in v3 XD LOL
[12-May-2011 11:55:03] <fragfutter> Hackman238: because anything related to gui has changed
[12-May-2011 11:55:14] <dawn2020>  @nyeates: i mean i created a new device class and created the same windows datasource.... to collect the CPU datapoints...but put  it in CPU event class it didnt work.....but when i put it in /perf/interface or /perf/memory event class ...event was triggered
[12-May-2011 11:55:23] <Jane_Curry> Not got that far yet - will do - but there probably isn't much that helps us with the V3 UI if the prereq is Zenoss 2.3
[12-May-2011 11:55:38] <Hackman238> fragfutter: Was supposed to be a joke about it supporting only an arcane version of zenoss
[12-May-2011 11:56:08] <Hackman238> GUI work in V3 is simply madness with the docs available.
[12-May-2011 11:56:16] <Jane_Curry> XD?????  X-Directory???  eXcept Dafties????
[12-May-2011 11:56:18] <rocket> I dont think there was as much ui stuff in there
[12-May-2011 11:56:32] <Hackman238> Jane_Curry: Its laughing with eyes closed.
[12-May-2011 11:56:34] <nyeates> dawn2020: can we work this in the forums from here? Can you paste our back and forth convo - or what you have, and your screenshots, and the descript above of what worked....then email url to
[12-May-2011 11:56:54] <rocket> Jane_Curry: I believe he was targeting more base class stuff than ui elements persay with that zenpack .. though I am sure there are some in there
[12-May-2011 11:57:31] <dpetzel> does anyeone know which logging I need to crank up to see why my device is being detected as deviceClass, rather than the class I have defiend in zPythonClass?
[12-May-2011 11:57:40] <dawn2020> sure....but how to resolve the issue with /perf/cpu event class
[12-May-2011 11:57:43] <Jane_Curry> OK. I have had other folk ask if I have training for ZenPacks.  At the moment I haven't....
[12-May-2011 11:58:05] <jplouis> dpetzel:  try running zenhub with -v10
[12-May-2011 11:58:11] <Jane_Curry> I don't wnat to start work on a ZenPack workshop is Zenoss themselves are already doing that?????????????????
[12-May-2011 11:58:14] <Hackman238> rocket: Maybe Zenoss should audit a handful of popular packs for best practice compliance and the community can then build docs from said known proper implemented code.
[12-May-2011 11:58:19] <dawn2020> @nyeates: can i delete it and create a new event class with same name
[12-May-2011 11:58:45] <Jane_Curry> Hackman238 - great suggestion
[12-May-2011 11:59:54] <Hackman238> *cough*
[12-May-2011 12:00:01] [disconnected at Thu May 12 12:00:01 2011]
[12-May-2011 12:00:01] [connected at Thu May 12 12:00:01 2011]
[12-May-2011 12:00:16] [zenoss-logger (logger bot) has joined #zenoss]
[12-May-2011 12:00:31] <nyeates> Jane_Curry: I do not know of any active ZenPack or dev workshops right now. Kells had worked on some stuff for one in the past, and he has mentioned that he would like to see it happen, but it has not had any backing from management at Zenoss Inc and I know of no plans.
[12-May-2011 12:00:54] <dpetzel> jpluis: thanks, wading through that output now
[12-May-2011 12:01:05] <Jane_Curry> Suppose I create a workshop, with input mebbe from Kells...............
[12-May-2011 12:01:56] <Jane_Curry> I can deliver it.  Youe guys can be licensed to deliver it.  We run it in classrooms if folk will come.  We run it as a remote class if there's no travel budget
[12-May-2011 12:02:23] <Jane_Curry> Should it be more than 1-day???  I suspect you need at least 2 days
[12-May-2011 12:02:25] <kells> Jane_Curry: That would be cool -- just need to get an okey-doke from my management.
[12-May-2011 12:02:48] <Jane_Curry> Hmmmm - that always seems to be where things stop
[12-May-2011 12:03:00] <kells> w/ JS stuff etc, probably three days is comfortable.
[12-May-2011 12:03:06] <Jane_Curry> What can we do to make it stick this time???
[12-May-2011 12:03:30] * nyeates thinks
[12-May-2011 12:03:33] <Jane_Curry> **************************
[12-May-2011 12:03:33] <kells> nyeates: See if it's okay to pass along the outline to Jane
[12-May-2011 12:03:39] <dpetzel> jpluis, that uncovered 'ImportError: No module named Prodcuts.ZenRelations.RelSchema'
[12-May-2011 12:04:11] <nyeates> kells: sure... Jane, ill see if my boss is ok with us sending you the outline kells has but together
[12-May-2011 12:04:18] <dpetzel> so my ttypo is now very obvious
[12-May-2011 12:04:25] <Jane_Curry> Of the customers amongst us here, who would be likely to "attend" a 3-day ZenPack workshop?????
[12-May-2011 12:04:56] <nyeates> or a virtual one
[12-May-2011 12:05:02] <dhopp> Jane: I would like to, depends on where and cost of course
[12-May-2011 12:05:05] <nyeates> which i think is more likely to get people to attend
[12-May-2011 12:05:13] <Jane_Curry> Thats why "attend" is in quotes
[12-May-2011 12:05:13] <Simon4> Jane_Curry: I have workmates who could well be sent on such a thing
[12-May-2011 12:06:02] <Hackman238> Jane_Curry: Would if virtual.
[12-May-2011 12:06:20] <Jane_Curry> I suspect Zenoss is the sort of company that now needs a "business case" to do anything.  We probably need some "guesstimates" for numbers of attendees....
[12-May-2011 12:06:23] <nyeates> curious why you said customers Jane.... did you mean Zenoss users in general?
[12-May-2011 12:06:42] <nyeates> yes true
[12-May-2011 12:07:04] <Jane_Curry> Sorry - yes - I meant Zenoss users.  Definitely NOT implying just Enterprise customers
[12-May-2011 12:07:13] <Jane_Curry> probably more likely to get Core folk
[12-May-2011 12:07:26] <dhopp> Hackman238: so I was looking at Egors distributed monitor zenpack and it looks like he sets the remote monitor up to run everything but ['zeoctl','zopectl','zenhub','zenjobs','zenactions','zenmodeler'].  Of those, is zenhub the only one that would need to be started on a remote collector to make the collector use that zenhub rather then the main one? (assuming zenhub is also configured for the right db)
[12-May-2011 12:07:51] <nyeates> Also, as I said jane: There has been some headway made on getting your zenpack docs looked over and made official....possibility of zenoss inc dev time spent on yer doc is high :-)
[12-May-2011 12:07:54] <Jane_Curry> but they do need to be "customers" in the sense that the workshop would NOT be "open source" as in FREE
[12-May-2011 12:08:18] <Hackman238> dhopp: Zenhub would need to be configured, set running, and daemons pointed to it rather than the Zope hosting zenhub.
[12-May-2011 12:08:58] <dhopp> but zeoctl would only run on the main server right?
[12-May-2011 12:09:28] <Hackman238> dhopp: zeo and zope run only on the main box.
[12-May-2011 12:09:44] <dhopp> ok
[12-May-2011 12:09:48] <kells> Uh oh.  The rabble rouser has joined...
[12-May-2011 12:10:07] <Hackman238> kells: rabble rouser?
[12-May-2011 12:10:11] <Jane_Curry> Let the rabble commence.....
[12-May-2011 12:10:38] <kells> Welcoming matt into IRC
[12-May-2011 12:10:52] <Hackman238> kells: Oh.
[12-May-2011 12:11:00] <Jane_Curry> mattray - thanks for the lead the other week.  Didn't come to anything for me but I appreciate the pointer
[12-May-2011 12:11:51] <mattray> Jane_Curry: that's too bad, but I'll keep referring people
[12-May-2011 12:12:17] <Jane_Curry> Nick, Kells - can we get together outside the IRC sometime to try and make this ZenPack workshop happen?
[12-May-2011 12:12:24] <Jane_Curry> Thanks Matt
[12-May-2011 12:12:40] <jplouis>     dpetzel: can you do the following in zendmd? find("<yourdevicename>")
[12-May-2011 12:13:55] <dpetzel> jpluis, yeap I can
[12-May-2011 12:14:14] <nyeates> Jane, Kells: how is this coming Tues at 11pm EST (same time of day as this dev meeting)
[12-May-2011 12:14:14] <kells> Jane: You've got my e-mail, so yeah, shoot me a note
[12-May-2011 12:14:19] <dpetzel> its working!!!
[12-May-2011 12:14:36] <dpetzel> jpluis, so you gave me the info I needed with the zenhb verbosity
[12-May-2011 12:14:45] <dpetzel> it poitned out the typo I had causing the import error for the device class
[12-May-2011 12:15:00] <dpetzel> I fixed that (and then the subsequent typos I had elsewhere, and I can see the components now
[12-May-2011 12:15:01] <nyeates> death to typos
[12-May-2011 12:15:03] <dpetzel> thanks a ton!
[12-May-2011 12:15:46] <kells> nyeates: I'd feel better if management had a heads up before anything got started.
[12-May-2011 12:16:34] <kells> At the very least, an "Unless otherwise disallowed, we are going to do X"  note
[12-May-2011 12:16:38] <Jane_Curry> I can make Tuesday May 17th at 11 (I hope that is am, not pm) EST - but only if we get some node from management, as Kells says
[12-May-2011 12:16:58] <nyeates> kells is right. We are employees spending our time. We will follow up by email and see what works.
[12-May-2011 12:16:59] <Jane_Curry> sorry s/node///nod/
[12-May-2011 12:18:55] <nyeates> Thanks all for attending and taking part in our dev meeting
[12-May-2011 12:19:48] <nyeates> i hope it was helpful and we can continue to increase the dev and zenpack knowledge
[12-May-2011 12:58:28] <nyeates> [12:16]  <dawn2020> Hi Nick...thanks a lot ....deleting the /perf/cpu event class and adding it back worked ...the cpu events has now triggered....
[12-May-2011 12:58:28] <nyeates> [12:16]  <dawn2020> thanks a lot
[12-May-2011 12:58:45] <nyeates> ^ wanted to note that this was solved, for anyone who reads these logs after the fact
[12-May-2011 13:35:14] <lorin> Hi Nick
[12-May-2011 13:36:24] <lorin> I am monitoring the WINDOWS server 2000 through WMI . I put the device in /CIM/WMI/2000 . I gave the WMI username and Passwd as well . When i am modeling the device i am getting the Ip address error . Even i deleted the device and checked the IP address if i am using the same ip for the other device, but i am not using the same ip . I am using the IP address only for the devices which i am modeling . This is happening for only Win2000 devices . I am using Zeno
[12-May-2011 15:01:41] <dhopp> Hackman238: you around?
[12-May-2011 16:12:40] <d-rock> Getting a strange error with a custom report. I made a report to get me device name and serial number, using "getId" and "getHWSerialNumber" as the columns. The first time I run the report it works great, but the second (and subsequent) time, I get the error: cannot concatenate 'str' and 'int' objects . If I wait 30 minutes or so and re-run it works again once, and then fails again with the...
[12-May-2011 16:12:41] <d-rock> ...same error
[12-May-2011 16:16:43] <d-rock> Is there any way to get more details as to what's going on here?
[12-May-2011 16:44:15] <hmp> anyone used community.DistrubutedCollectors to set up dis collectors?
[12-May-2011 16:45:05] <dhopp> hmp what's your question?
[12-May-2011 16:45:20] <hmp> my graphs dont show
[12-May-2011 16:45:38] <hmp> and as soon as i switch back to local collector they do
[12-May-2011 16:46:11] <hmp> no fw issue, checked with coontrack/tcpdump
[12-May-2011 16:46:59] <dhopp> if you go to Advanced->Collectors->(select remote collector)->Edit..what is the Render URL set to?
[12-May-2011 16:47:12] <hmp> tried both
[12-May-2011 16:47:20] <hmp> the locationA:8090/locationB
[12-May-2011 16:47:21] <dhopp> what do you mean tried both?
[12-May-2011 16:47:25] <hmp> and default
[12-May-2011 16:47:30] <dhopp> oh hrm
[12-May-2011 16:48:06] <dhopp> that's what I thought the problem would be…I'm looking at using that zenpack but might have to do it manual (I probably want multiple zenhubs)
[12-May-2011 16:50:56] <hmp> think ill go for manual too, if this doesnt work after some more debugging
[12-May-2011 17:20:13] kocolosk is now known as kocolosk|away
[12-May-2011 18:31:41] kocolosk|away is now known as kocolosk
[12-May-2011 18:36:35] kocolosk is now known as kocolosk|away
[12-May-2011 19:53:42] kocolosk|away is now known as kocolosk
[12-May-2011 20:06:40] <joko_> 
[12-May-2011 21:07:03] kocolosk is now known as kocolosk|away
[12-May-2011 21:10:26] kocolosk|away is now known as kocolosk
[13-May-2011 00:00:01] [disconnected at Fri May 13 00:00:01 2011]
[13-May-2011 00:00:01] [connected at Fri May 13 00:00:01 2011]
[13-May-2011 00:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[13-May-2011 06:23:43] <kotique> guys, am I the only one with ?
[13-May-2011 06:24:20] <kotique> when you restart net-snmp daemon, what alerts are you usually getting?
[13-May-2011 07:20:36] <fragfutter> kotique: none. but i use snmp V2 and no traps.
[13-May-2011 07:23:36] <kotique> anyone on v3 here?
[13-May-2011 07:54:12] kocolosk is now known as kocolosk|away
[13-May-2011 08:48:23] <Morthez> Any one got a clue why the webinterface is so sluggish over here? We are running zenoss on a 8 core/24Gb memory (Mem:         26179       4119      22060 ) Any suggestions on what i can do to make it a bit more snappy?
[13-May-2011 08:49:23] <fragfutter> Morthez: check mysql events history table size
[13-May-2011 08:49:34] <Morthez> 8gb
[13-May-2011 08:49:41] <fragfutter> Morthez: number of events?
[13-May-2011 08:49:48] <Morthez> no, sorry.. That's the total size of the DB
[13-May-2011 08:49:57] <Simon4> select count(*) from history;
[13-May-2011 08:50:09] <fragfutter> and 8GB sounds way to big
[13-May-2011 08:51:54] <Morthez> +----------+
[13-May-2011 08:51:54] <Morthez> | count(*) |
[13-May-2011 08:51:54] <Morthez> +----------+
[13-May-2011 08:51:54] <Morthez> |  2676709 |
[13-May-2011 08:52:03] <Morthez> hum, that's quite a lot
[13-May-2011 08:52:36] <fragfutter> start the printer, print out every event to file it, and then clean up.
[13-May-2011 08:52:38] <Simon4> that won't be helping interface speed, especially viewing devices
[13-May-2011 08:53:01] <fragfutter> delete from history where unix_timestamp()-lastTime > 30*24*60*60;
[13-May-2011 08:53:28] <Morthez> only to keep last month?
[13-May-2011 08:53:41] <fragfutter> Morthez: congratulations.
[13-May-2011 08:54:08] <Simon4> Morthez: there's a setting under event manager that lets zenoss automatically do that
[13-May-2011 08:54:43] <fragfutter> i need to add a calculated horizontal line to a graph. Currently i add a MinMaxThreshold, enter the value i want into the Maximum Value "here.getMyFunkValue". Is there a nicer way?
[13-May-2011 08:55:19] <Simon4> fragfutter: add a custom datapoint to the graph definition of HRULE ?
[13-May-2011 08:55:43] <fragfutter> Simon4: how can i specify the value there?
[13-May-2011 08:55:55] <Simon4> ahhhhh
[13-May-2011 08:56:04] * Simon4 quickly researches his answer some mroe
[13-May-2011 08:56:52] <Simon4> datapoint -> add custom
[13-May-2011 08:56:58] <Simon4> pick hrule, give it a name, then hit ok
[13-May-2011 08:57:06] <Simon4> you then get to set "value" in the next screen
[13-May-2011 08:57:31] <fragfutter> Simon4: zenoss3. and the value can only be an RRD value (at least as far as i see)
[13-May-2011 08:57:40] <Simon4> gah
[13-May-2011 08:57:44] <Simon4> where is my v3 install
[13-May-2011 08:59:33] <fragfutter> using "here.supplyMax" as a value breaks the graph. it can only be a rrd datapoint
[13-May-2011 09:00:45] <Morthez> well, that poor server is really banging it's IO now
[13-May-2011 09:02:28] <Simon4> and works for me
[13-May-2011 09:03:33] <fragfutter> Simon4: and now enter something non static and not part of RRD into the value field
[13-May-2011 09:03:56] <fragfutter> fragfutter: for example if you graph a interface, try here.speed
[13-May-2011 09:04:28] <Morthez> Simon4: any suggestions to make the web UI snappier?
[13-May-2011 09:05:12] <Simon4> fragfutter: ahh, I just tried with an integer
[13-May-2011 09:05:26] <fragfutter> Morthez: cleanup mysql.
[13-May-2011 09:05:42] <Morthez> That is in progress
[13-May-2011 09:05:57] <fragfutter> then wait for it to finish and check the ui again.
[13-May-2011 09:06:43] <Simon4> Morthez: cleanup mysql, get it's data files off the same spindles as /opt/zenoss/var/* and also the rrd files away from /opt/zenoss/var, if you need more speed than that there are ways to run multiple web interface processes in parallel
[13-May-2011 09:07:09] <Simon4> the current web interface is multi-threaded, but due to the python global interpreter lock it's restricted to a single core in total CPU, so  your 8-way box isn't being used a lot for the UI
[13-May-2011 09:08:08] <Simon4> fragfutter: you might need to go have a read of or similar to make sure that if you have a HRULE defined it actualy proces tales expressions on the value field
[13-May-2011 09:08:16] <Simon4> processes even
[13-May-2011 09:08:47] <fragfutter> Simon4: maybe later, for now i can live with my max threshhold workaround
[13-May-2011 09:08:57] <Simon4>
[13-May-2011 09:09:42] <fragfutter> writing zenpacks is like poking into mud and see what you find. Documentation lacks a lot.
[13-May-2011 09:10:10] <Simon4> yeah, Jane's doc is fairly awesome, but she's only learned the same way so you're left wondering what else is possible
[13-May-2011 09:11:14] <fragfutter> Nothing against Janes doc (it's the best that is out there) but i simply can't grasp how she structures it. I would need a reference book.
[13-May-2011 09:12:15] <Simon4> yeah
[13-May-2011 09:12:20] <Simon4> so from
[13-May-2011 09:12:49] <Simon4> self.value is not tales eval'ed
[13-May-2011 09:12:57] <Simon4> it's just returned raw as the HRULE command
[13-May-2011 09:13:09] <Simon4> it would be trivial to make it eval'ed, but it's a code patch
[13-May-2011 09:15:59] <fragfutter> my problem is that i monitor a supply level. Which is dropping. So i have a min-threshold. Graphing current and min into a graph is easy. But i would like to have the max value in there also. I could poll the snmp max value on every cycle and simply also graph it. But the max is static, so it would be enough to pull it during modelling.
[13-May-2011 09:19:42] <fragfutter> Simon4: are you simonh on the forums?
[13-May-2011 09:19:47] <Morthez> Simon4: Would it be possible to host zenoss using Apache's Python modules?
[13-May-2011 09:19:55] <fragfutter> Morthez: no
[13-May-2011 09:20:04] <Simon4> Morthez: not that I know of, I mod_proxy_balance it via apache
[13-May-2011 09:20:28] <fragfutter> Morthez: for zope you would use mod_proxy. Not sure how zenoss handles it (with regards to virtualhostmonster)
[13-May-2011 09:20:57] <Morthez> Simon4: Does it give you any better performance?
[13-May-2011 09:21:20] <Simon4> Morthez: only because we run 8 zope UI processes and balance across them
[13-May-2011 09:21:25] <Simon4> let me find the doc
[13-May-2011 09:21:30] <Simon4> fragfutter: from memory I am, yes
[13-May-2011 09:21:33] <Simon4> (simonh)
[13-May-2011 09:22:29] <Simon4> Morthez: <-- that was worked out here where I work and we run it successfully
[13-May-2011 09:22:48] <Simon4> you want to be sure it's zopectl eating a whole core of CPU that's making things slow before going down that path though
[13-May-2011 09:22:54] <Simon4> so do some analysis first
[13-May-2011 09:24:00] <Simon4> /opt/zenoss/bin/runzope are things to match on
[13-May-2011 09:35:09] <dhopp1> Hackman238: you awake?
[13-May-2011 09:35:24] dhopp1 is now known as dhopp
[13-May-2011 09:40:08] kocolosk|away is now known as kocolosk
[13-May-2011 09:49:38] <fragfutter> i model a snmp device that use a two dot table index. Now i want to poll performance data for the discovered components. These would be SomeBaseOID.snmpindexpart1.snmpindexpart2. can i do this?
[13-May-2011 09:52:15] <fragfutter> nevermind. i can just use a string in Componten.snmpindex (didn't know, tried an integer all the time)
[13-May-2011 09:53:25] <dhopp> does anybody know that when using the stack installer if it keeps changes to config files?  What I mean is if I install zenoss with a stack installer and then customize zope.conf or zenhub.conf and then upgrade are my changes kept or they blown away?
[13-May-2011 09:54:46] <Simon4> fragfutter: heh, I found that 'feature" by accident too
[13-May-2011 09:57:49] <fragfutter> next question. how can i influence the name of the rrd file zenoss creates? is it the name of the component?
[13-May-2011 09:58:02] <Simon4> yeah
[13-May-2011 10:05:35] <fragfutter> there was some utility function to generate an id (normalize the string and make sure it is unique). Anyone got the name in the head?
[13-May-2011 10:05:58] <Simon4> um um um prepId
[13-May-2011 10:06:20] <fragfutter> which is where?
[13-May-2011 10:06:39] <Simon4> you didn't ask that before
[13-May-2011 10:06:46] <fragfutter> i'll ask grep
[13-May-2011 10:06:57] <fragfutter> Products.ZenUtils.Util
[13-May-2011 10:07:00] <Simon4> it's available in the scope of a modeller plugin though
[13-May-2011 10:07:01] <fragfutter> (for the log)
[13-May-2011 10:08:06] <Simon4> = self.prepId("%s slot %2d" % (, om.bsPosition))
[13-May-2011 10:08:10] <Simon4> = self.prepId("%s slot %2d" % (, om.bsPosition)) even
[13-May-2011 10:08:19] <Simon4> is how I call it in the modeller code for one of my zenpacks
[13-May-2011 10:08:40] <fragfutter> = self.prepId("printer_supply_%s" % id)
[13-May-2011 10:08:42] <fragfutter>
[13-May-2011 10:11:03] joko is now known as joko_
[13-May-2011 10:41:15] <fragfutter> my custom component has some weird entries in the display drop down box (Events, Details, OS, Software, Graphs, Edit).
[13-May-2011 10:55:16] <fragfutter> Where do those entries come from?
[13-May-2011 10:55:55] <Simon4> fragfutter: they're some factory or other thing from somewhere
[13-May-2011 10:56:05] * Simon4 doesn't know, basically
[13-May-2011 10:56:18] <fragfutter> did i already say it is like mud?
[13-May-2011 10:56:56] <Simon4> yeah
[13-May-2011 10:57:16] <Simon4> maybe we need to come up with a wiki page of questions like that one, and hurl it at a dev next dev day
[13-May-2011 11:02:22] <Hackman238> Anyone know how I might XMLRPC call /zport/dmd/Monitors/Hub/localhost/manage_installCollector from Egors Distributed Collector pack?
[13-May-2011 11:24:19] <kotique> guys, am I the only one with ?
[13-May-2011 11:27:48] kocolosk is now known as kocolosk|away
[13-May-2011 11:28:58] <dhopp> Hackman238: can I pick your brain about the distributed zenhubs we briefly talked about yesterday?
[13-May-2011 11:30:45] <Hackman238> dhopp: Sure
[13-May-2011 11:30:56] <Hackman238> kotique: I've never seen that problem
[13-May-2011 11:31:12] <Simon4> kotique: I haven't seen that either
[13-May-2011 11:31:36] <Hackman238> Anyone very good with XMLRPC?
[13-May-2011 11:32:26] <kotique> hmm, that's weird. because it happens right when you use snmpv3 and reload net-snmp daemon
[13-May-2011 11:32:31] <dhopp> Hackman238: let's say I have 4 machines, A,B,C,D.  Machine A is running zope and zeoctl.  Machine's B and C are remote collectors are running every daemon except zope and zeoctl.  Machine D is mysql for the zenoss infrastructure.
[13-May-2011 11:32:45] <Hackman238> dhopp: Ok
[13-May-2011 11:33:02] <dhopp> Machine A goes down (say for a reboot, or some other reason).  Do B and C continue to monitor and I just lose visibility?
[13-May-2011 11:33:39] <Hackman238> Sort of. The perf tests will continue, but alerting will not occur.
[13-May-2011 11:34:16] <Hackman238> Essentially tests collecting points stored in RRD's will continue uninterrupted until they go to fetch a new config
[13-May-2011 11:34:17] <dhopp> but as soon as Machine A comes back up, the web interface will see whatever events happened?
[13-May-2011 11:34:25] <Hackman238> Sort of.
[13-May-2011 11:35:12] <Hackman238> Machine A, handeling zenactions, will play catch up, but depending on the downtime of A, some events may never be acted on
[13-May-2011 11:38:27] <Hackman238> Its my understanding Egors pack will disable zenactions, zenhub and zenjobs also on the collectors
[13-May-2011 11:38:38] <dhopp> yeah that's what I have seen
[13-May-2011 11:38:44] <dhopp> so I was looking at doing it manually
[13-May-2011 11:39:34] <Hackman238> Well you chould enable the local hub and point items to it, etc. I cant say I've ever had more than one zenactions running- that seems like it could be trouble.
[13-May-2011 11:40:08] <dhopp> so what do you normally have running on remote collectors?  all daemons except zope,zeoctl and zenactions?
[13-May-2011 11:41:18] <Hackman238> zenperfsnmp, zenmodeler, zencommand, zensyslog, zenprocess, zenrender
[13-May-2011 11:41:26] <Hackman238> For you you'll also want zenping
[13-May-2011 11:41:57] <dhopp> what happens if zeoctl is unavailable when zenperfsnmp goes to re-read it's config?…does zenperfsnmp shit the bed or does it just keep running with it's current config?
[13-May-2011 11:42:00] <Hackman238> My mistake, it's zenperfsnmp, zenmodeler, zencommand, zensyslog, zenprocess, zenrender, zenstatus, zentrap
[13-May-2011 11:42:31] <Hackman238> Its supposed to keep going
[13-May-2011 11:42:36] <dhopp> ok
[13-May-2011 11:42:53] <dhopp> just trying to get a grasp at where my points of failure are
[13-May-2011 11:43:53] <Hackman238> Np
[13-May-2011 11:44:35] kocolosk|away is now known as kocolosk
[13-May-2011 11:46:43] <dhopp> Hackman238: do you know if the stack installer keeps customized config files intact during upgrade? for example with zenhub.conf or zope.conf?
[13-May-2011 11:48:28] <Hackman238> dhopp: Unfortunately I dont know, I only use RPM.
[13-May-2011 11:48:39] <Hackman238> dhopp: rmatte can tell you, however.
[13-May-2011 11:49:02] <dhopp> rmatte: you watching?
[13-May-2011 11:50:50] <dhopp> one reason I'm looking at going with stack is getting ldap auth to work in core on RHEL/CentOS 5 and Zenoss 3.1 isn't fun because of the different python versions…and we were looking at using RHEL6
[13-May-2011 11:55:36] <Hackman238> dhopp: Gotcha. rmatte tells me stack runs very well.
[13-May-2011 12:00:01] [disconnected at Fri May 13 12:00:01 2011]
[13-May-2011 12:00:02] [connected at Fri May 13 12:00:02 2011]
[13-May-2011 12:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[13-May-2011 13:35:28] <Hackman238> Anyone in?
[13-May-2011 13:36:03] <Hackman238> Definately Friday.
[13-May-2011 13:40:56] <joko_> frickin
[13-May-2011 13:40:57] <joko_> tired
[13-May-2011 13:42:11] <Hackman238> *remembers joke from American Dad about tiredness*
[13-May-2011 13:42:33] <joko_> <3 that show, don't remember the joke tho
[13-May-2011 13:43:23] <Hackman238> joko_: Its the one where Stan tells steve that if he doesn't get enough sleep he'll smell like sweet berries and attract pedophiles. LOL
[13-May-2011 13:43:36] <joko_> haha
[13-May-2011 13:43:40] <joko_> yeah that sounds about right
[13-May-2011 13:43:46] <Hackman238> joko_: How are you with XMLRPC?
[13-May-2011 13:43:58] <joko_> non-existant
[13-May-2011 13:44:28] <Hackman238> joko_: damn. I've a call I need to make but I cant figure it out.
[13-May-2011 13:44:54] <joko_> yeah, not a clue :\
[13-May-2011 13:45:27] <Hackman238> joko_: No biggie.
[13-May-2011 13:59:06] <joko_> donkey poop
[13-May-2011 14:34:58] <dhopp> is there an easy way in zenoss 3.x (in particular 3.1) to allow non manager users to be able to acknowledge and close events?
[13-May-2011 15:03:01] <jmp242> dhopp: I don't think so, there's a bit of a convoluted way that Jane wrote up IIRC
[13-May-2011 15:56:39] <nlloyds> Hi, I'm new to zenoss. I'm having trouble on the Event Console for devices: If a device has more than just a few events I get the Loading. Please wait... indefinitely.
[13-May-2011 15:56:51] <nlloyds> zenoss 3.1.0
[13-May-2011 15:57:18] <jmp242> mmm
[13-May-2011 15:57:38] <jmp242> Do you mean if you go to a device in infrastructure
[13-May-2011 15:57:45] <jmp242> and select events in the left hand column
[13-May-2011 15:57:46] <jmp242> ?
[13-May-2011 15:57:51] <nlloyds> right
[13-May-2011 15:58:10] <jmp242> Have you tried the various tuning Zenoss docs?
[13-May-2011 15:58:41] <nlloyds> I have not. Just googled it. Here?
[13-May-2011 15:59:19] <nlloyds> What is considered "an enormous amount of events"?
[13-May-2011 16:00:44] <jmp242> That's a decent one
[13-May-2011 16:02:30] <jmp242> mmm
[13-May-2011 16:02:37] <jmp242> I can't say on enormous amount of events
[13-May-2011 16:02:47] <jmp242> I mostly have had ~30 events on one device at the most
[13-May-2011 16:03:03] <jmp242> no problem for me in Firefox
[13-May-2011 16:04:45] <nlloyds> I've got syslog logging set up, so it's probably going to be more than that. I was using before, but needed more monitoring and graph goodness.
[13-May-2011 16:07:12] <nlloyds> lol I think my innodb_buffer_pool_size is 8M
[13-May-2011 16:13:48] <nlloyds> Also I just learned that the my.cnf used is the one in /usr/local/zenoss/mysql *not* the one in /usr/local/zenoss/etc
[13-May-2011 16:14:01] <jmp242> Hmmm, we don't dump all syslog into Zenoss. Check the threads on the forums
[13-May-2011 16:14:20] <jmp242> there's some info on max number of syslogs Zenoss Core can realistically process per collector
[13-May-2011 16:14:36] <jmp242> I think it maxes out around 200 per second or some such
[13-May-2011 16:15:52] <jmp242> It depends on the hardware realluy
[13-May-2011 16:16:03] <nlloyds> That sounds like it could work. I'm tweaking the mysql settings which should help a lot. I'm using the zenoss-stack ubuntu package, and the default mysql settings are pretty weak.
[13-May-2011 16:16:40] <jmp242> Some people who get slammed with syslog
[13-May-2011 16:17:00] <jmp242> use a front end collector that forwards to Zenoss like syslogng or there were others
[13-May-2011 16:17:45] <jmp242> What are you looking to do with the syslogs? If you want to do log parsing /collation there may be better syslog processors
[13-May-2011 16:17:59] <jmp242> Of course I can't rember what they are right now
[13-May-2011 16:18:31] <nlloyds> Yeah I'm using syslog-ng on the devices. Probably just want to be alerted if there are major error messages.
[13-May-2011 16:19:15] <nlloyds> A little analysis stuff might happen, but it's mostly for monitoring the health of all processes.
[13-May-2011 16:19:26] <jmp242> What you may want to do is look into event mappings and set up ones that drop the events you don't care about
[13-May-2011 16:19:43] <jmp242> But yes, you want to increase your zope cache and mysql cache if you have the RAM for it
[13-May-2011 16:26:15] <nlloyds> Thanks for the help.
[13-May-2011 17:16:06] <JohnnyNoc> big ups to jmp242
[13-May-2011 17:16:38] kocolosk is now known as kocolosk|away
[13-May-2011 18:36:15] kocolosk|away is now known as kocolosk
[14-May-2011 00:00:01] [disconnected at Sat May 14 00:00:01 2011]
[14-May-2011 12:00:02] [connected at Sat May 14 12:00:02 2011]
[14-May-2011 12:00:20] [zenoss-logger (logger bot) has joined #zenoss]
[14-May-2011 22:46:23] kocolosk is now known as kocolosk|away
[15-May-2011 00:00:01] [disconnected at Sun May 15 00:00:01 2011]
[15-May-2011 00:00:02] [connected at Sun May 15 00:00:02 2011]
[15-May-2011 00:00:20] [zenoss-logger (logger bot) has joined #zenoss]
[15-May-2011 08:23:53] kocolosk|away is now known as kocolosk
[15-May-2011 12:00:01] [disconnected at Sun May 15 12:00:01 2011]
[15-May-2011 12:00:02] [connected at Sun May 15 12:00:02 2011]
[15-May-2011 12:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[15-May-2011 21:29:40] <lronpaw> high
[15-May-2011 21:29:41] <lronpaw> hi
[15-May-2011 21:29:46] <lronpaw> anyone alive?
[15-May-2011 21:42:30] <dhopp> sort of
[15-May-2011 21:42:31] <dhopp> heh
[15-May-2011 21:45:28] <lronpaw> hehe
[15-May-2011 21:45:34] <lronpaw> its ok i found what i was after
[15-May-2011 21:45:49] <lronpaw> http://zenoss:8080/zport/manage/
[16-May-2011 00:00:01] [disconnected at Mon May 16 00:00:01 2011]
[16-May-2011 00:00:02] [connected at Mon May 16 00:00:02 2011]
[16-May-2011 00:00:17] [zenoss-logger (logger bot) has joined #zenoss]
[16-May-2011 08:32:10] <micheal> Hi All I followed the zenoss doc even i was unable to set the threshold for swap . Can any one please help me in this
[16-May-2011 08:38:36] <micheal> this is for a linux box
[16-May-2011 09:13:13] <fragfutter> micheal: insufficient date doesn't compute. Or with other words, what are you trying to do, what failed (and which zenoss version are you on)
[16-May-2011 09:24:50] <micheal> @fragfutter:I am using zenoss 3.10 I am using the Datapoint as memoAvailSwap_memAvailSwap and gave the theshold as minimun "here.os.totalSwap / 1000 * 0.5"
[16-May-2011 09:25:46] <micheal> I want to trigger the alert if the free swap is below 50%
[16-May-2011 09:29:57] <fragfutter> and?
[16-May-2011 09:31:29] kocolosk is now known as kocolosk|away
[16-May-2011 09:34:12] <micheal> can u give me the threshold expression for the sam e
[16-May-2011 09:34:32] <micheal> is the expression which i have given is correct ?
[16-May-2011 09:35:46] <fragfutter> yes its correct
[16-May-2011 09:42:26] kocolosk|away is now known as kocolosk
[16-May-2011 09:47:29] <dhopp> fragfutter: I'm coming into this late.  Is your threshold not triggering? or is the problem something else?
[16-May-2011 09:47:34] <dhopp> oop
[16-May-2011 09:47:40] <dhopp> that was meant for michael
[16-May-2011 10:47:05] <micheal> @fragfutter:Can i use this expression  for realmemory used above 95% here.os.totalRealmem / 1000 * 0.95
[16-May-2011 10:48:35] <fragfutter> micheal: it makes no sense to monitor memory usage on a linux server. it _will_ use all memory at some point
[16-May-2011 10:49:04] <micheal> can u please give me the correct expression to monitor the real memory
[16-May-2011 10:49:24] <fragfutter> micheal: there is none. just monitor swap.
[16-May-2011 10:51:32] <micheal> i would like to monitor the memory as well
[16-May-2011 10:51:53] <micheal> zenoss will allow me to monitor as well rght ?
[16-May-2011 10:53:57] <fragfutter> you can monitor it the same way like your swap threshold. But it makes no sense. You could also monitor the current date. at some point it will be the year 2012.
[16-May-2011 10:56:01] <micheal> for this i have given the datapoint as "memAvailReal_memAvailReal" and expresssion as here.os.totalRealmem / 1000 * 0.25
[16-May-2011 10:56:29] <micheal> can u help me out in the expression
[16-May-2011 11:00:56] <micheal> @fragfutter:this is for realmemory available is more than 25%
[16-May-2011 11:10:43] <micheal> can anyone please help me on this ?
[16-May-2011 11:16:49] <astrostl> 2011-05-16 10:11:18,071 WARNING zen.WMIClient: Unable to collect WMI data from [FQDN]: NT_STATUS_IO_TIMEOUT
[16-May-2011 11:17:12] <astrostl> i've tried completely disabling the firewall, and connecting using wmi query from an external windows host works.  any ideas?
[16-May-2011 11:27:01] <micheal> @astrostl:try to restart WMI , RPC if not reboot the system
[16-May-2011 11:44:55] <astrostl> i figured it out.  another monitored host was entered with the same ip.  strange error msg, reporting as a bug.  thx!
[16-May-2011 12:00:01] [disconnected at Mon May 16 12:00:01 2011]
[16-May-2011 12:00:01] [connected at Mon May 16 12:00:01 2011]
[16-May-2011 12:00:21] [zenoss-logger (logger bot) has joined #zenoss]
[16-May-2011 12:25:23] <kokey> anyone here monitoring HP hardware on ESXi?
[16-May-2011 13:14:00] er1k757_ is now known as erik757
[16-May-2011 13:15:04] erik757 is now known as er1k757
[16-May-2011 13:42:18] <kokey> dmd.ZenEventManager.getEventList doesn't seem to list all events
[16-May-2011 13:48:45] <xuru> anyone use ipython with zenoss dmd?
[16-May-2011 13:50:12] <Sam-I-Am> yes
[16-May-2011 13:50:23] <Sam-I-Am> i believe 3.x automatically detects its existence
[16-May-2011 13:50:51] <xuru> you don't have to install it from a source?
[16-May-2011 13:52:18] <xuru> hmm... looks that way
[16-May-2011 13:52:31] <xuru> I get a ZopeXMLConfigurationError: File "/usr/local/zenoss/zenoss/etc/site.zcml", line 1.0
[16-May-2011 13:52:39] <xuru> when I try:
[16-May-2011 13:52:45] <xuru> dmd = ZenScriptBase(connect=True).dmd
[16-May-2011 13:53:18] <xuru> does it have to be in a particular directory?
[16-May-2011 13:55:38] <Sam-I-Am> i just installed the rpm for it
[16-May-2011 13:55:57] <Sam-I-Am> (for ipython)
[16-May-2011 13:56:19] <xuru> k
[16-May-2011 14:45:03] <xuru> lol, zendmd is ipython
[16-May-2011 15:17:56] <JohnnyNoc> anyone use winexe with a command check to run a powershell script?
[16-May-2011 15:18:01] <JohnnyNoc> having a timeout issue I don't undersatnd
[16-May-2011 16:27:43] <JohnnyNoc> id10t error
[16-May-2011 16:27:48] <JohnnyNoc> PEBKAC
[16-May-2011 16:35:30] <Hackman238> JohnnyNoc: User error?
[16-May-2011 16:54:20] <JohnnyNoc> yea pmuch
[16-May-2011 16:54:20] <JohnnyNoc>
[16-May-2011 16:55:17] <JohnnyNoc> i was trying to execute these checks on a server running nsclient with winexe, after doing it the right way using check_nrpe i didn't have any more of these timeout issues when trying to run my powershell script
[16-May-2011 16:56:12] <JohnnyNoc> oh yea, luser error
[16-May-2011 16:56:16] <JohnnyNoc> forgot about that one
[16-May-2011 17:29:01] kocolosk is now known as kocolosk|away
[16-May-2011 17:31:55] uifjlh_ is now known as uifjlh
[16-May-2011 20:06:53] kocolosk|away is now known as kocolosk
[16-May-2011 23:18:28] <davetoo> The then block is executed if the expression is truthy; otherwise, the optional
[16-May-2011 23:18:28] <davetoo> else branch is taken
[16-May-2011 23:18:37] <davetoo> </quote>
[17-May-2011 00:00:01] [disconnected at Tue May 17 00:00:01 2011]
[17-May-2011 00:00:02] [connected at Tue May 17 00:00:02 2011]
[17-May-2011 00:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[17-May-2011 09:02:08] kocolosk is now known as kocolosk|away
[17-May-2011 09:13:23] <JohnnyNoc> anyone know how i could search for devices with a particular string in their hostname in a Zenoss report?
[17-May-2011 09:15:17] <JohnnyNoc> can i somehow import re and use match?
[17-May-2011 09:25:38] kocolosk|away is now known as kocolosk
[17-May-2011 10:48:20] <kokey> ok argh argh
[17-May-2011 10:48:47] <kokey> so i write a python script
[17-May-2011 10:48:49] <kokey> that uses
[17-May-2011 10:48:50] <kokey> from Products.ZenUtils.ZenScriptBase import ZenScriptBase
[17-May-2011 10:48:51] <kokey> from transaction import commit
[17-May-2011 10:48:59] <kokey> so it runs fine from the shell as the zenoss user
[17-May-2011 10:49:11] <kokey> but if i put it into the crontab for the zenoss user, it doesn't run
[17-May-2011 10:49:34] <kokey> i also make sure the following envs are set ZENHOME PATH PYTHONPATH LD_LIBRARY_PATH LD_LIBRARY_PATH_64
[17-May-2011 10:49:44] <kokey> problem is it doesn't give an error either
[17-May-2011 10:53:54] <kokey> yay
[17-May-2011 10:53:57] <kokey> I got an error message
[17-May-2011 10:54:06] <kokey>     from Products.ZenUtils.ZenScriptBase import ZenScriptBase
[17-May-2011 10:54:07] <kokey> ImportError: No module named ZenUtils.ZenScriptBase
[17-May-2011 10:54:22] <kokey> sounds like i am missing something from the path
[17-May-2011 10:56:58] <kokey> ah, I needed INSTANCE_HOME
[17-May-2011 10:57:20] <SEJeff_work> When is the next deveoper's IRC meeting?
[17-May-2011 11:25:54] <fragfutter> SEJeff_work: every second thursday if i'm not misstaken. That would be max.26
[17-May-2011 11:25:59] <fragfutter> may
[17-May-2011 11:26:32] <SEJeff_work> fragfutter, thanks
[17-May-2011 12:00:01] [disconnected at Tue May 17 12:00:01 2011]
[17-May-2011 12:00:02] [connected at Tue May 17 12:00:02 2011]
[17-May-2011 12:00:21] [zenoss-logger (logger bot) has joined #zenoss]
[17-May-2011 13:25:06] <uifjlh> Why aren't defined Windows components given a specific "event class" ... ie.  "Windows SharePoint Services 3" "
[17-May-2011 13:25:21] <uifjlh> given "/unknown" event class ...
[17-May-2011 13:34:02] <uifjlh> I guess I answered my own ??... TY for listening, I'm in the 15+ listings in "the community" ...
[17-May-2011 14:13:11] <mike4353> Hi ther....i recently upgraded Zenoss core from 2.4 to 2.4 version i was using the event transform nterface High Utilization event enhancement using if, then, elif from...
[17-May-2011 14:13:45] <mike4353> but after the upgrade now i'm getting error n the zenhub.log I am seeing the following when the event is triggered.
[17-May-2011 14:14:16] <mike4353> has anyone faced this same problem or can anyone help me on this
[17-May-2011 14:15:21] <mike4353> i donot understand please help how resolve this issue
[17-May-2011 14:49:36] <mike4353> Hi guys...can anyone please help me with this error ...In the zenhub.log I am seeing the following when the event is triggered.
[17-May-2011 14:49:48] <mike4353> Problem on line 28: NameError: name 'evtNewKey' is not defined
[17-May-2011 14:50:05] <mike4353> evt.summary = "High " + evtNewKey + " Utilization: Currently (%3.2f Mbps) or %3.2f%%  is being used." %  (Usage, p)
[17-May-2011 14:59:45] <dhopp> mike4353: Did you go from 2.4 directly to 3.1?
[17-May-2011 15:05:22] <mike4353> @dhopp: nope...made the upgrade 2.5 and then to 3.1 using the preupgrade package
[17-May-2011 15:10:30] <dhopp> Have you tried to remove the transform and recreate it?
[17-May-2011 15:11:28] <mike4353> yes i did...but still the same error
[17-May-2011 15:14:17] <mike4353> i removed the evtNewKey from transform ....and then the interface events started coming up with wrong values
[17-May-2011 15:14:34] <mike4353> but the error in zenhub log stopped
[17-May-2011 15:14:39] <dhopp> Can you start zenhub with verbose logging?  (as the zenoss user run zenhub -v10)
[17-May-2011 15:14:59] <mike4353> sure
[17-May-2011 15:15:17] <dhopp> Looking at the Doc my guess is that for some reason evtNewKey isn't being set to any value before it's being used
[17-May-2011 15:15:28] <dhopp> I'm hoping that verbose logging will give some more info
[17-May-2011 15:15:43] <mike4353> oh ok
[17-May-2011 15:19:52] <mike4353> please see the output
[17-May-2011 15:22:19] <dhopp> you need to stop zenhub first
[17-May-2011 15:22:26] <dhopp> so as zenoss user
[17-May-2011 15:22:28] <dhopp> zenhub stop
[17-May-2011 15:22:30] <dhopp> zenhub -v10
[17-May-2011 15:23:05] <mike4353> ah ok
[17-May-2011 15:34:22] <mike4353>
[17-May-2011 15:36:13] <dhopp> I don't see that error anywhere in that log?  Has it an event happened yet?
[17-May-2011 15:37:36] <mike4353> after removing the keyword evtNewKey  from the transform the error has gone
[17-May-2011 15:38:01] <mike4353> however the interface are shoring wrong values
[17-May-2011 15:38:08] <mike4353> showinr*
[17-May-2011 15:39:18] <dhopp> Can you put the transform back in there so we can see if zenhub logs more?
[17-May-2011 15:39:50] <mike4353> sure
[17-May-2011 15:51:31] <uifjlh> has anyone tried to monitor an OS/400 environment ?
[17-May-2011 15:52:05] <mike4353> sorry dhopp! i couldnt find anything related to the transform in using zenhub run -v10...however i hav found this in the zenhub log...
[17-May-2011 15:55:02] <dhopp> I'm not sure what's happening then…you might need to modify the transform a little to try to debug it
[17-May-2011 15:56:00] <mike4353> oh ok....can please help me with i have no clue with python
[17-May-2011 15:56:55] <dhopp> in your transform under the line evtNewKey = "Output" try adding the following:
[17-May-2011 15:56:59] <dhopp> elif
[17-May-2011 15:57:07] <mike4353> ok
[17-May-2011 15:57:09] <dhopp> evtNewKey = evtKey
[17-May-2011 15:57:16] <dhopp> white space matters in python
[17-May-2011 15:57:22] <mike4353> ohh
[17-May-2011 15:57:26] <dhopp> so make it indented the same way as the line above
[17-May-2011 15:57:33] <mike4353> oh ok
[17-May-2011 15:57:59] <dhopp> if this works the evt.summary is going to be REALLY ugly, but it might tell us what's happening
[17-May-2011 15:58:11] <mike4353> oh ok
[17-May-2011 15:59:09] <dhopp> oops
[17-May-2011 15:59:15] <dhopp> it shouldn't be elif
[17-May-2011 15:59:25] <dhopp> it should be "else:" (without quotes)
[17-May-2011 15:59:50] <dhopp> the "else:" and "evtNewKey = evtKey" should be a separate lines
[17-May-2011 16:01:05] <mike4353> else syntax worked
[17-May-2011 16:01:19] <dhopp> so the transform is working?
[17-May-2011 16:01:29] <dhopp> what is the event summary?
[17-May-2011 16:02:15] <mike4353> not sure...waiting for it to reflect in the event console
[17-May-2011 16:16:56] <Slapx_> Hi guys i hope you're fine.
[17-May-2011 16:17:16] <Slapx_> Im having some troubls with DNS and LDAP Zenpacks in Zenoss 3.1
[17-May-2011 16:18:08] <Slapx_> After all the basic configuration, the event window shows a warning: Misuse of shell Builtins
[17-May-2011 16:18:16] <locohost> been a while since I have used zenoss, nearly fesh install, I model a device I see pop3 green for this host hooray,  I change "Monitored" to true it statys green.  I go to the system and stop the service and it is still green..
[17-May-2011 16:18:27] <locohost> 20m later now, with 5m polling interval
[17-May-2011 16:18:34] <locohost> am I missing something here?
[17-May-2011 16:21:04] <Slapx__> i'd read about it  in some forums and apparently the solution may be to modify the output to replace the colons for dashes
[17-May-2011 16:21:35] <Slapx__> The problem is that, i don't really know where to change those options
[17-May-2011 16:22:23] <Slapx__> Also, i dont have the command option which set Auto to Nagios
[17-May-2011 16:23:26] <mike4353> @dhopp: no luck! still the interface events are showing up wrong values
[17-May-2011 16:23:52] <dhopp> mike4343: but what is the event summary?
[17-May-2011 16:25:11] <mike4353> High Utilization: Currently (5.66 Mbps) or 25.79% is being used.
[17-May-2011 16:25:23] <mike4353> which is the inbound traffic usage
[17-May-2011 16:25:36] <mike4353> it shud show outbound traffic usage
[17-May-2011 16:27:05] <xSLAP> ?
[17-May-2011 16:28:20] <locohost> anybody? strange that the check mark for the service stays green after I stop the service.
[17-May-2011 16:34:09] <dhopp> locohost: is this a windows service?
[17-May-2011 16:46:27] <locohost> it is a windows machine., yes
[17-May-2011 16:46:44] <locohost> im referencing something under "ip services" though, and not "windows services"
[17-May-2011 16:47:45] <locohost> im guessing its getting this info from the netstat info thats in..i think the host resources mib
[17-May-2011 16:48:44] <locohost> TCP-MIB::tcpConnectionState.ipv4."".110.ipv4."".0 = INTEGER: listen(2)  (maybe out of that from an snmp walk)
[17-May-2011 16:50:52] <locohost> ive confirmed that if i stop and start the service, that oid in hte tcp-mib reflect the service stopping/starting seeminly instantly
[17-May-2011 16:56:00] <locohost> well, tomorrow is another day
[17-May-2011 17:11:17] <dhopp> locohost: is the service set to manual or automatic start?
[17-May-2011 17:15:12] <dhopp> I seem to remember that if the service was set to manual start the detection of whether it was running or not wasn't accurate.  Although according to it was supposed to be fixed in 3.0.2
[17-May-2011 17:27:58] kocolosk is now known as kocolosk|away
[17-May-2011 17:49:33] <locohost> it is automatic, ty though
[17-May-2011 20:10:17] kocolosk|away is now known as kocolosk
[17-May-2011 20:49:24] <locohost> anyone around? cant seem to get servicing polling the way i remember...used to use zenoss extensivly like 3 years ago and am back anew like 2 major revs later
[17-May-2011 20:49:43] <locohost> anyway, i add  and model a device and undder ip services see my services
[17-May-2011 20:49:54] <locohost> say, pop3 or 25,
[17-May-2011 20:50:07] <locohost> it shows green and i take that service down and it never goes red
[17-May-2011 20:50:46] <locohost> i give a valid or invalid command and change the response regex to garbage and it will not fail no matter what i do
[17-May-2011 20:54:23] <locohost> ahh, dhopp, im back at it, lol u just logged in after i started asking questions agoin
[17-May-2011 20:54:27] <locohost> again
[17-May-2011 20:55:37] <dhopp> locohost: I'm not sure the answer to be honest.  Might be a good idea to post specifics to the forums
[17-May-2011 20:56:47] <locohost> do you monitor based on service polls? ie, telnet into 25, say helo zenossbox, check to see that you get 220*?
[17-May-2011 21:00:18] <dhopp> I'm actually just starting to build out our zenoss infrastructure at my current company.  At my past job where we used zenoss we didn't do much of that.
[17-May-2011 21:01:36] <locohost> ahh, i did all this like 4 years ago for a different company and much has changed..i dont remember this not working heh
[17-May-2011 23:28:21] kocolosk is now known as kocolosk|away
[18-May-2011 00:00:02] [disconnected at Wed May 18 00:00:02 2011]
[18-May-2011 00:00:02] [connected at Wed May 18 00:00:02 2011]
[18-May-2011 00:00:21] [zenoss-logger (logger bot) has joined #zenoss]
[18-May-2011 06:04:51] <squig_> im trying to find a way to change the interface template from ethernetCsmacd to the _64 version
[18-May-2011 06:05:18] <squig_> but im at a loss to how you work the ui to do this, (i am trying to change all my network devices to use this template_
[18-May-2011 06:05:43] <squig_> I dont really understand the docs in regards to this, can any one help me or point me at good explanation?
[18-May-2011 06:06:03] <fragfutter> squig_: zenoss 2 or 3
[18-May-2011 06:06:07] <squig_> 3
[18-May-2011 06:06:29] <fragfutter> advanced -> monitoring templates
[18-May-2011 06:06:38] <squig_> yep
[18-May-2011 06:06:41] <fragfutter> then find the ethernetCsmacd template.
[18-May-2011 06:06:44] <fragfutter> modify it
[18-May-2011 06:06:45] <fragfutter> done
[18-May-2011 06:07:28] <fragfutter> its a component template
[18-May-2011 06:07:43] <squig_> I dont understand how to modify it
[18-May-2011 06:07:52] <fragfutter> what do you want to change?
[18-May-2011 06:08:19] <squig_> there is a ethernetCsmacd_64 template, I want that to be used for my network devices rather than the ethernetcsmacd
[18-May-2011 06:09:11] <fragfutter> your device is modelled. The modeller detects the type of component. It either detects ethernetCsmacd_64 (64 bit counters) or ethernetCsmacd.
[18-May-2011 06:09:31] <fragfutter> Then it will search for a template matching this component.
[18-May-2011 06:09:39] <fragfutter> Once found it will use this template.
[18-May-2011 06:10:04] <squig_> does it prefer ethercsmacd_64 the documents say that it defaults to ethernetCsmacd
[18-May-2011 06:10:06] <fragfutter> You can modify the ethernetCsmacd template, but you can't use the same template for ethernetCsmacd_64 and ethernetCsmacd
[18-May-2011 06:10:37] <fragfutter> squig_: if it can't find 64bit counters it will use the ethernetCsmacd
[18-May-2011 06:11:54] <squig_> ok let me look for those 64 bit values on my devices
[18-May-2011 06:15:21] <squig_> aah I think i understand now, its querying against snmpv1 rather than2c so those counters don't exist
[18-May-2011 06:18:06] <squig_> aah remodeled the device and now i see the 64 bit monitoring template
[18-May-2011 06:20:24] <Sonne> greetings!
[18-May-2011 06:21:04] <Sonne> i have this "localhost zenjmx heartbeat failure" spamming my event list every 5 minutes
[18-May-2011 06:21:44] <Sonne> i've checked the logs, and it seems zenjmx can't start because there's no JVM installed
[18-May-2011 06:23:05] <Sonne> thing is, i don't really use java at all, so i'd like to just disable the zenjmx daemon or suppress the warnings
[18-May-2011 06:23:10] <Sonne> is there any way to do that?
[18-May-2011 06:27:36] fOrsberg is now known as forsberg
[18-May-2011 09:06:03] kocolosk|away is now known as kocolosk
[18-May-2011 09:07:08] <Hackman238> Hello all
[18-May-2011 09:10:43] forsberg is now known as fOrsberg
[18-May-2011 09:29:24] <JohnnyNoc> morning Hackman238
[18-May-2011 09:29:35] <JohnnyNoc> your nick reminds me of an old Misfits song..  We Are 138
[18-May-2011 09:29:39] <JohnnyNoc> +100 for your
[18-May-2011 09:29:42] <JohnnyNoc> you
[18-May-2011 09:30:40] <Hackman238> LOL
[18-May-2011 09:31:05] <JohnnyNoc> are you familiar with it?
[18-May-2011 09:31:15] <Hackman238> Im not
[18-May-2011 09:31:30] <JohnnyNoc>
[18-May-2011 09:31:38] <JohnnyNoc> nice studio version
[18-May-2011 09:32:46] <Hackman238> Oh yes, I've heard this before :-)
[18-May-2011 09:34:37] <locohost> hey guys, I cannot seem to get ip service polling working correctly.  I model a device and it shows smtp under ip services, but if i stop the service or even iptable that host off from the zenoss server, It stays green
[18-May-2011 09:35:16] <locohost> ive also tried manually adding smtp ip service to a machine not listening for smtp and it says its green...but just does not show an interface its listening on
[18-May-2011 09:36:18] <davetoo>   I think I need one of these for my laptop
[18-May-2011 09:36:41] <JohnnyNoc> duuuude
[18-May-2011 09:36:45] <davetoo> I've heard one the last couple of days and it's ... amazing
[18-May-2011 09:36:46] <JohnnyNoc> only if it's made by Dr Dre
[18-May-2011 09:36:59] <davetoo> que?
[18-May-2011 09:37:16] <JohnnyNoc> i'm just kinda poking fun at all the "Beats By Dre" audio stuff out now
[18-May-2011 09:37:43] <JohnnyNoc> you haven't seen em?
[18-May-2011 09:37:46] <davetoo> I'm not buying headphones unless they start with Sony or Sennheiser
[18-May-2011 09:37:56] <davetoo> or Grado
[18-May-2011 09:37:58] <Simon4> s/Sony//
[18-May-2011 09:37:58] <JohnnyNoc> haha good strategy
[18-May-2011 09:38:08] <Simon4> <3 my HD-25's
[18-May-2011 09:38:16] <davetoo> Sony MDR-7506
[18-May-2011 09:38:19] <Simon4> 11 years old and still amazing
[18-May-2011 09:38:27] <JohnnyNoc> davetoo are you a djtoo
[18-May-2011 09:38:42] <davetoo> no, but I do audio recording
[18-May-2011 09:38:57] <davetoo> used to do a bit of concert taping, now I just do field/ambient stuff.
[18-May-2011 09:39:01] <locohost> what happened to the OS tab?
[18-May-2011 09:39:23] <davetoo> version?
[18-May-2011 09:40:11] <locohost> new: zenoss-3.1.0-1031.el5
[18-May-2011 09:40:40] <davetoo> it's in the left nav bar
[18-May-2011 09:40:49] <JohnnyNoc> so i have some new machines i've added that is sending out high utilization alerts..  the nics were speed unkown, i've since manually set them..  but alerts keep hitting the dashboard
[18-May-2011 09:40:58] <JohnnyNoc> is there a daemon i should bounce?  or am i being impatient?
[18-May-2011 09:42:30] <locohost> dont see os there, overview, event, components, software, graph, admin, config properties, modeler plugins, custom, modifications nad monitoring templates in left nav batr
[18-May-2011 09:42:46] <davetoo> sorry, still sleepy
[18-May-2011 09:42:57] <davetoo> it's rough equivalent is "Components"
[18-May-2011 09:43:18] <davetoo> in other words you don't need it
[18-May-2011 09:43:48] <Hackman238> They continue to alert whilst having the speed set?
[18-May-2011 09:43:54] <locohost> ty
[18-May-2011 09:44:25] <davetoo> I'm still finding my way around 3.x but unless you're writing a zenpack, Components seems to be the equivalent.
[18-May-2011 09:48:35] <davetoo> locohost: looks like the Hardware components (cpu, memory, disks) also appear under Components.
[18-May-2011 10:14:31] <miheal> When i am modeling a devices in Zenoss I am getting the error as File "/opt/zenoss/Products/ZenHub/", line 75, in inner"
[18-May-2011 10:24:06] <locohost> I am beginning to wonder if my issue is not a configuration issue but a problem with one of hte running services.  I have added OS process  monitors now and when i stop the service on the polled machine, said service name no longer shows in an snmpwalk, but still shows green on the zenoss console
[18-May-2011 10:25:53] <locohost> oh hrm, maybe it dissapeared rather then alarming, grr,
[18-May-2011 10:37:09] <locohost>
[18-May-2011 10:37:23] <locohost> any idea why these continue to be green when I stop these services?
[18-May-2011 10:39:53] <locohost> zenprocess.log last lines say
[18-May-2011 10:39:54] <locohost> 2011-05-18 10:38:12,069 INFO zen.zenprocess: 0 devices processed (0 datapoints)
[18-May-2011 10:39:54] <locohost> 2011-05-18 10:38:12,070 INFO zen.collector.scheduler: Tasks: 0 Successful_Runs: 0 Failed_Runs: 0 Missed_Runs: 0
[18-May-2011 10:40:14] <locohost> does it sound like it not knowing to check these hosts for running processes on the back end?>
[18-May-2011 10:46:35] <Slapx_> Hi Guys! im having some troubles with Zenpack DNS and LDAP in Zenoss 3.1
[18-May-2011 10:46:47] <Slapx_> Is this the right place to find a solution?
[18-May-2011 10:56:53] <locohost> possibly, its really quiet in here
[18-May-2011 11:01:05] <Slapx_> i see =/
[18-May-2011 11:01:27] <Hackman238> Slapx, what seems to be the problem?
[18-May-2011 11:01:30] <Slapx_> I'm having an event warning that says: Misuse of shell builtins
[18-May-2011 11:01:52] <Slapx_> After making the basic configuration of Zenpacks, DNS and LDAP
[18-May-2011 11:02:34] <Hackman238> Slapx_: Ah- you'll need to amek sure your command returns in Nagios format, and returns (0) OK or (1) Error
[18-May-2011 11:02:50] <Slapx_> Cmd: -H server -b dc=zenoss,dc=com -D "" -P "" --ver2 -p 389 -t 60 - Code: 2 - Msg: Misuse of shell builtins
[18-May-2011 11:03:14] <Hackman238> Slapx_: Try chaning the parser from Auto to Nagios
[18-May-2011 11:03:16] <fragfutter> Slapx_: its missing the command
[18-May-2011 11:03:21] <Slapx_> i'd read about that but, i can't find the option which allows to change from auto to Nagios
[18-May-2011 11:03:30] <Hackman238> Slapx_: V3?
[18-May-2011 11:03:35] <Slapx_> Yes
[18-May-2011 11:03:40] <Slapx_> 3.1
[18-May-2011 11:04:24] <Hackman238> Slapx_: Do you know how to edit the tempaltes?
[18-May-2011 11:04:27] <Slapx_> Also, i don't know the right place where to change the output conf
[18-May-2011 11:04:33] <fragfutter> Slapx_: there has to be something before the -H, like check_...
[18-May-2011 11:04:39] <Slapx_> Sort of
[18-May-2011 11:05:03] <Hackman238> This is from memory since I haven't V3 in front of me.
[18-May-2011 11:05:37] <Hackman238> click advanced, monitoring templates
[18-May-2011 11:05:56] <Hackman238> Pick the template on the left
[18-May-2011 11:06:02] <Slapx_> Ok, is there a way to make it directly in /usr/local/zenoss ?
[18-May-2011 11:06:14] <Hackman238> No, its all GUI
[18-May-2011 11:06:41] <Hackman238> It could be done via zendmd, but it wouldn't be as easy.
[18-May-2011 11:07:10] <fragfutter> It could be done by locating the zenpack and editing object.xml and reinstalling it
[18-May-2011 11:07:48] <Hackman238> Yes, but he'd have to zip the packs folder and reinstall from that.
[18-May-2011 11:08:14] <Sonne> i might as well try asking again my earlier question:
[18-May-2011 11:08:16] <fragfutter> i was just kidding, it's only theoritcal.
[18-May-2011 11:08:37] <Hackman238> fragfutter: You're absolytely right, he could do that.
[18-May-2011 11:08:55] <Sonne> zenjmx won't start because there's no java installed, and this is flooding my events with zenjmx heartbeat failure - since i don't use java, how can i get rid of zenjmx overall - or suppress the warnings?
[18-May-2011 11:09:13] <Slapx_> Excelent, ill try the xml option =)
[18-May-2011 11:09:48] <Hackman238> Disable the daemon and uninstall the jmx zenpack
[18-May-2011 11:10:32] <locohost> what could be broken that would cause  no alarms to alarm?
[18-May-2011 11:10:53] <Sonne> Hackman238, are you sure it will work?
[18-May-2011 11:11:09] <Simon4> you'll also want to "clear all heartbeats"
[18-May-2011 11:11:20] <Simon4> which hides under event manager somewhere from memory
[18-May-2011 11:12:29] <Hackman238> Slapx_: You'll need to modify the xml as suggested then compress the pack back into an egg and re zenpack --install it. Optionally you could use zenobjectcopier to import the objects.xml
[18-May-2011 11:12:35] <Hackman238> Simon4: Agreed.
[18-May-2011 11:12:41] <Sonne> aight, thanks
[18-May-2011 11:12:47] <Hackman238> Sonne: If you're not using that Zenpack, you shouldnt have it installed.
[18-May-2011 11:13:04] <Slapx_> Hackman238: Im in the file right now, what should i modify?
[18-May-2011 11:13:07] <Hackman238> Sonne: Hopefully whatever broke that pack hasnt broken anything else.
[18-May-2011 11:13:08] <Sonne> Hackman238, it comes by default IIRC - can i reinstall it separately should i one unlucky day need it again?
[18-May-2011 11:13:27] <Hackman238> Sonne: Yes
[18-May-2011 11:13:36] <Sonne> Hackman238, the absence of a JVM is what's breaking that package, because the daemon isn't starting
[18-May-2011 11:13:56] <locohost> bahm, maybe my zen pack screwed something up, i gues sill reinstall
[18-May-2011 11:14:58] <fragfutter> on a device page, there is the commands menu (bottom left), clicking any command in there (ping for example) produces a javascript error. Is this menu working for anyone?
[18-May-2011 11:15:28] <locohost> yes, those work for me frag
[18-May-2011 11:15:41] <locohost> using latest rpm
[18-May-2011 11:15:59] <fragfutter> ok. it's a chrome problem. in firefox they work.
[18-May-2011 11:16:15] <locohost> chrome is listed on the site as a supported browser
[18-May-2011 11:16:35] <locohost> not using chrome over here though, so cant really help
[18-May-2011 11:17:01] <Sonne> hm
[18-May-2011 11:17:16] <Sonne> Hackman238, Simon4, i uninstalled the zenpack but i'm still having heartbeat events
[18-May-2011 11:17:33] <locohost> well, i guess ill grab a smoke whle my reinstall is building the zope db
[18-May-2011 11:17:39] <Simon4> Sonne: so this is where you have to go to the Event Manager, and find the "clear all heartbeats" option, probably in the gear menu somewhere
[18-May-2011 11:18:21] <Sonne> aha Simon4, didn't check the gear menu...
[18-May-2011 11:18:22] <Simon4> a "heartbeat" is just a row in the events database, if the row isn't continuously updated zenoss creates the alerts you see. Clearing down the heartbeats clears this database
[18-May-2011 11:19:26] <Sonne> well all is working now, thanks
[18-May-2011 11:22:13] <fragfutter> created new ticket 7884 (Device Context Commands Javascript Error in Chrome), if anyone using chrome can confirm this?
[18-May-2011 11:22:31] <fragfutter>
[18-May-2011 11:24:24] <Slapx_> Hackman238: Can the problem be that the script check_ldap isn't in the zenoss/libexec folder?
[18-May-2011 11:24:49] <fragfutter> Slapx_: from your logfile it isn't in the template
[18-May-2011 11:26:43] <Slapx_> fragfutter: Ok but, the objects.xml doesnt show nothing about the command of the logfile
[18-May-2011 11:27:02] <fragfutter> Slapx_: please check in the gui, by opening the templates datasource.
[18-May-2011 11:27:45] <Slapx_> fragfutter: Ok
[18-May-2011 11:41:27] <locohost> when you add a os process monitor, do you use the name the process says in an snmp walk or the name as displayed in windows services?
[18-May-2011 11:53:56] <Hackman238> Sorry all, had to run for a second.
[18-May-2011 11:54:07] <Hackman238> Sonne: heartbeats cleared up?
[18-May-2011 11:54:23] <Sonne> Hackman238, aye, all's good now
[18-May-2011 11:54:25] <Sonne>
[18-May-2011 11:54:27] <Hackman238> Slapx_: I'd make the changes in the GUI then add them back to a new Zenpack
[18-May-2011 11:54:38] <Hackman238> Sonne: Glad it is.
[18-May-2011 11:55:15] <Hackman238> Slapx_: As for check_ldap- the command the script is trying to run must exist where zenoss can find it in its paths
[18-May-2011 11:56:42] <Slapx_> Hackman238: I move it to the /libexec path
[18-May-2011 11:58:20] <Hackman238> Slapx_: Okay. Please make sure it has the correct permissions
[18-May-2011 11:58:36] <Slapx_> Hackman238: ok =)
[18-May-2011 12:00:01] [disconnected at Wed May 18 12:00:01 2011]
[18-May-2011 12:00:01] [connected at Wed May 18 12:00:01 2011]
[18-May-2011 12:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[18-May-2011 12:09:00] <locohost> anyone have issues with wmi not disabling? I have set the rpc and wmi ignore option in config to true nad remodeled and still showing me wmi access errors
[18-May-2011 12:09:11] <locohost> ive also removed the zenoss.wmi plugin from modeler plugins
[18-May-2011 12:14:52] <Hackman238> locohost: Are you seeing this in logs?
[18-May-2011 12:16:42] <locohost> i am seeing an event popup in the console
[18-May-2011 12:16:56] <locohost> Could not read the Windows event log (NT_STATUS_ACCESS_DENIED). Check your username/password settings a
[18-May-2011 12:17:33] <locohost> ive set zWmiMonitorIgnore
[18-May-2011 12:17:34] <locohost> zXmlRpcMonitorIgnore both to true
[18-May-2011 12:17:42] <locohost> and zwinEventlog to false
[18-May-2011 12:17:46] <Hackman238> locohost: The device in question doesnt have overridden zproperties, does it?
[18-May-2011 12:18:13] <locohost> hmm, maybe, where is t hat?
[18-May-2011 12:18:26] <Hackman238> Goto device screen
[18-May-2011 12:18:55] <locohost> there, i see a override template option
[18-May-2011 12:19:01] <Hackman238> Configuration properties
[18-May-2011 12:19:59] <locohost> that is where I am settings this
[18-May-2011 12:20:05] <Hackman238> Just look through to see if any oare overridden.
[18-May-2011 12:20:12] <Hackman238> Oh ok. Hum.
[18-May-2011 12:20:18] <Hackman238> Have you restarted zenoss?
[18-May-2011 12:20:22] <locohost> no
[18-May-2011 12:20:31] <Hackman238> try this
[18-May-2011 12:20:34] <locohost> i was thinking about doing that then you resonded, heh
[18-May-2011 12:20:41] <Hackman238> zopectlrestart, zenhub restart, zenmodeler restart
[18-May-2011 12:21:08] <Hackman238> Once done, remodel the device
[18-May-2011 12:21:27] <Hackman238> If it still doesnt stop zenoss stop, then zenoss start
[18-May-2011 12:21:42] <Hackman238> Avoid zenoss restart- zenhub hangs often
[18-May-2011 12:21:53] <locohost> done, thanks, we will see
[18-May-2011 12:22:01] <Hackman238> Np.
[18-May-2011 12:23:03] <locohost> ugh, its back
[18-May-2011 12:23:21] <Hackman238> I'd restart all of zenoss next
[18-May-2011 12:28:47] <locohost> that may have worked, kinda quirky, popped up immidiatly after a restart then i restarted a 2nd time and not seeing it yet
[18-May-2011 12:28:50] <locohost> its been almost 5m
[18-May-2011 12:30:48] <locohost> cool, and i restarted pop3 on the exchange server (only server in here atm) and it alerted me that the service restarted
[18-May-2011 12:30:52] <locohost> and still no wmi alert
[18-May-2011 12:31:07] <Hackman238> Nice
[18-May-2011 12:31:55] <locohost> now where im changing that, thats for the device only, to change it globally to not use wmi when i add new devices?
[18-May-2011 12:32:24] <Hackman238> configuration properties on the device classes
[18-May-2011 12:33:51] <locohost> ahh, had to click details
[18-May-2011 12:34:07] <Hackman238> yep
[18-May-2011 12:34:14] <Hackman238> Same zprops
[18-May-2011 14:43:01] <locohost> okay, nother issue now...have setup both tcp and process checks for pop3 and smtp, im pretty sure i have them configured exactly the same way
[18-May-2011 14:43:17] <locohost> except, when i take the pop service down, i get alarms for both service not running and tcp port unresponsive..good
[18-May-2011 14:43:44] <locohost> when it take smpt down, i only get an alert saying the service is not running, says port 25 is green still on ip services
[18-May-2011 14:44:23] <Hackman238> Hum.
[18-May-2011 14:44:42] <locohost> i have restarted zenoss 3x just to be sure, heh
[18-May-2011 14:44:46] <locohost> and remodeled those devices
[18-May-2011 14:46:11] <Hackman238> take smtp down and try to hit it with a client.
[18-May-2011 14:46:18] <Hackman238> is it refused or does it timeout?
[18-May-2011 14:47:58] <locohost> timeout
[18-May-2011 14:48:17] <Hackman238> Hum. Not sure, that what should be happening. Anyone else know what might be up?
[18-May-2011 14:48:35] <locohost> [root@d01mtrending01 log]# telnet 25
[18-May-2011 14:48:35] <locohost> Trying
[18-May-2011 14:48:35] <locohost> telnet: connect to address Connection timed ou
[18-May-2011 14:49:09] <Hackman238> locohost: nein nein, try telnet 25
[18-May-2011 14:49:30] <Hackman238> locohost: my mistake, I missed that line above.
[18-May-2011 14:49:39] <Hackman238> locohost: been a long day LOL
[18-May-2011 14:50:12] * Simon4 isn't sure what is happening there, it almost sounds like a UI bug
[18-May-2011 14:50:23] <Simon4> if you're getting the alert it means the collector daemon is behaving
[18-May-2011 14:50:38] <Simon4> just that the getStatus() call isn't returning the appropriate thing
[18-May-2011 14:50:40] <Hackman238> locohost: Are you on V2.5 or V3?
[18-May-2011 14:51:08] <locohost> 3, latest rpm available
[18-May-2011 14:51:12] <locohost> downloaded it yesterday
[18-May-2011 14:51:17] <Hackman238> locohost: If V2.5, the bubbles are almost certainly not working.
[18-May-2011 14:51:21] <Hackman238> locohost: Hum.
[18-May-2011 14:51:34] <locohost> im looking at the status table in the events database and only show the os process fail there too
[18-May-2011 14:52:05] <locohost> well, let me emulate a failure for pop3 and see if the tcp even shows in that table
[18-May-2011 14:53:07] <locohost> I could see if it worked that way, like if hte service is down no need to try telnetting into the port..but there is nothing mapping my service name to the actual smtp port...
[18-May-2011 14:53:19] <locohost> i know it does that for it it doesnt ping, dont both ssmp
[18-May-2011 14:53:21] <locohost> snmp
[18-May-2011 14:54:10] <locohost> yeah, when i take pop down, i see both the service alarm and tcp alarm in the database
[18-May-2011 14:56:26] <locohost> whicih log file do the tcp check diag info go to?
[18-May-2011 14:58:47] <Simon4> zenstatus.log for ipservices
[18-May-2011 14:58:54] <Simon4> and zenprocess.log for the process check
[18-May-2011 14:59:24] <locohost> thks, while i am looking at that i will start the service back up and see what happens if i iptables -A OUTPUT -o eth0 -p tcp --dport 25 -j REJECT
[18-May-2011 15:00:45] <locohost> thanks simon, the pop3 errors show there but nothing with the smtp port being down
[18-May-2011 15:02:04] * Simon4 is just afk for 10 min, back soon
[18-May-2011 15:06:24] <locohost> wonder if something naughty is cached somewhere or something
[18-May-2011 15:20:40] <locohost> what does clear heartbeats do?
[18-May-2011 15:21:02] <Simon4> clear heartbeats isn't related to this, it's used to work out whether collection daemons are up
[18-May-2011 15:21:23] <Simon4> locohost: when you stop smtp on the host, do you get an alert/event from zenoss about it?
[18-May-2011 15:21:47] <locohost> if i restart it, i get a process alert saying it was restarted, because i selected that box
[18-May-2011 15:21:57] <locohost> and if i stop it, i get a service alert saying the process is stopped
[18-May-2011 15:22:07] <locohost> but, it still shows port 25 green under ip services
[18-May-2011 15:22:38] <locohost> on my pop servers, if i stop the pop service, i get 2 alerts
[18-May-2011 15:22:45] <locohost> one saying the port is unreachable, and one saying the service is stopped
[18-May-2011 15:22:48] <locohost> which is what i want
[18-May-2011 15:23:16] <locohost> u know, incase the service gets wedged, but still shows in ps list
[18-May-2011 15:24:29] <Simon4> okay, so you're only seeing a process alert, not an ipservice one
[18-May-2011 15:24:41] <locohost> yes
[18-May-2011 15:25:02] <locohost> and, i currently have the services running with iptables on my zenoss box blocking all outbound 25 traffic
[18-May-2011 15:25:12] <locohost> and its still all green on the ipservice check
[18-May-2011 15:25:42] <Simon4> let me fire up my 3.1 install, I need to refresh my memory
[18-May-2011 15:27:23] <locohost> cool, ide be happy to send you a screen shot or something, but it sounds like we are on the same page...
[18-May-2011 15:28:35] <Simon4> if you navigate to infrastructure-> ip services, then type smtp on the left side to filter the list and click on smtp
[18-May-2011 15:28:56] <Simon4> do you see your host under the "service instances" in the bottom-right pane?
[18-May-2011 15:29:25] <locohost> yes
[18-May-2011 15:29:28] <locohost> and they all show green
[18-May-2011 15:29:49] <locohost> and monitored is true
[18-May-2011 15:32:04] <locohost>
[18-May-2011 15:33:34] * Simon4 is just modelling a box
[18-May-2011 15:34:15] <Simon4> right
[18-May-2011 15:34:20] <Simon4> I have smtp as an ip service
[18-May-2011 15:36:56] <Simon4> when you tail your zenstatus.log, what is the last line? something like "finished n jobs" ?
[18-May-2011 15:37:13] <Simon4> I've just added and modelled a device, and zenstatus isn't even trying to check it
[18-May-2011 15:38:17] <locohost> yeah, finished jobs, ill grab it
[18-May-2011 15:38:35] <locohost> 2011-05-18 15:38:18,170 INFO zen.ZenStatus: Finished 30 jobs (30 good, 0 bad) in 0.02 seconds
[18-May-2011 15:38:55] <locohost> if i stop pop, on my pop server, I do see it say stuff about pop in there
[18-May-2011 15:39:06] <locohost> 2011-05-18 14:41:12,432 INFO zen.ZenStatus: Finished 30 jobs (30 good, 0 bad) in 15.01 seconds
[18-May-2011 15:39:06] <locohost> 2011-05-18 14:42:12,429 WARNING zen.ZenTcpClient: IP Service pop3 is down
[18-May-2011 15:39:06] <locohost> 2011-05-18 14:42:12,430 WARNING zen.ZenTcpClient: IP Service pop3s is down
[18-May-2011 15:39:27] <Simon4> try running "zenstatus run -v 10" as the zenoss user on the command-line
[18-May-2011 15:43:39] <locohost> it appears its only checking my pop servers
[18-May-2011 15:43:54] <locohost> and it is checking eveyr tcp port i have it watching fine
[18-May-2011 15:44:02] <locohost> (there more then just pop servers, ms exchange junk)
[18-May-2011 15:44:15] <locohost> its not doing anything to my hub servers there at all...hmm
[18-May-2011 15:44:37] <Simon4> you've tried bouncing zenoss?
[18-May-2011 15:45:48] <Simon4> I've just added a device in v3 with smtp, and I'm failing to get zenstatus to monitor anything on it, which is wierd
[18-May-2011 15:45:53] * Simon4 is trying to find out why
[18-May-2011 15:46:42] <locohost> ive restarted the service like 10x, not the box
[18-May-2011 15:50:32] <locohost> ive just added my Mstores and they dont appear to be scanning in zenstatus run -v 10...0only my client access servers
[18-May-2011 15:50:40] <locohost> hrm, i have not restarted since i added those though
[18-May-2011 15:52:53] <locohost> ive just stopped zenoss again and made sure no processes were running by zenoss before starting
[18-May-2011 15:53:26] <locohost> we will see if they are included when i run that status command this time
[18-May-2011 15:53:47] <locohost> unfortunatly, i cannot stop any srevices to test on the mailbox servers to see if they are having the same issue
[18-May-2011 15:53:53] <locohost> got migrations going to them
[18-May-2011 15:57:14] <locohost> wt crap, its only doing tcp polls on all my client access servers and nothing else
[18-May-2011 15:57:26] <locohost> (all thats in here are client access/ hub and mailbox servers atm)
[18-May-2011 15:57:50] <locohost> i added those servers that are working first
[18-May-2011 15:58:01] <locohost> got pop to work the way i wanted, then added the hub servers
[18-May-2011 15:58:27] <locohost> so, its like something happened after i added the cas/pop servers and stuff added since that point are not getting tcp polled
[18-May-2011 15:58:42] <Simon4> yeah
[18-May-2011 15:59:00] <Simon4> I seem to have hit a similar thing on my lab box, which just isn't polling anything Ijust added
[18-May-2011 15:59:05] * Simon4 is trying to debug it
[18-May-2011 15:59:42] <locohost> sweet
[18-May-2011 15:59:46] <locohost> appreciate the help
[18-May-2011 15:59:53] <locohost> im ganna reboot the box just for hte hell of it
[18-May-2011 16:00:38] <locohost> what os is your lab machine?
[18-May-2011 16:00:46] <Simon4> centos
[18-May-2011 16:00:58] <locohost> yeah, this is centos 5.6 and zenoss-3.1.0-1031.el5
[18-May-2011 16:13:11] * Simon4 has to run away for a bit, but I'll look at this again a bit later
[18-May-2011 16:14:08] <Hackman238> locohost: Sorry I cant be more help with this problem
[18-May-2011 16:14:14] <locohost> no problem
[18-May-2011 16:14:16] <locohost> thanks simon
[18-May-2011 16:14:31] <locohost> im probably ganna head home an an hour or so but ill jump back in irc
[18-May-2011 16:14:46] <locohost> ive googled my eyes out to no avail, heh
[18-May-2011 16:15:41] <locohost> thanks both of you guys
[18-May-2011 16:17:52] <locohost> ill start looking at /opt/zenoss/Products/ZenStatus/ and see if i can figure out how its looping through devices...
[18-May-2011 16:26:57] <locohost> can i comma or ; seperate email addresses for an account or will i have to use a distro
[18-May-2011 17:43:38] kocolosk is now known as kocolosk|away
[18-May-2011 17:54:07] kocolosk|away is now known as kocolosk
[18-May-2011 18:54:23] <akafritz> hello folks.  i have a quick question about monitoring template inheritance.  If I bind a template to a device class, but one of the devices in that class has a template bound to it directly, does it STOP inheriting from its event class?
[18-May-2011 19:29:24] <locohost> I beleive so aka, but cant say for usre
[18-May-2011 22:20:43] <locohost> burned all my notebooks, what good are notebooks, they wont help me survive
[18-May-2011 23:12:28] kocolosk is now known as kocolosk|away
[18-May-2011 23:12:41] kocolosk|away is now known as kocolosk
[18-May-2011 23:51:46] <davetoo> I got some records, some peanut butter,
[18-May-2011 23:51:59] <davetoo> groceries
[18-May-2011 23:52:02] <davetoo> dangit
[18-May-2011 23:52:15] <davetoo> now I have to hear that
[19-May-2011 00:00:01] [disconnected at Thu May 19 00:00:01 2011]
[19-May-2011 00:00:01] [connected at Thu May 19 00:00:01 2011]
[19-May-2011 00:00:01] [disconnected at Thu May 19 00:00:01 2011]
[19-May-2011 00:00:02] [connected at Thu May 19 00:00:02 2011]
[19-May-2011 00:00:20] [zenoss-logger (logger bot) has joined #zenoss]
[19-May-2011 00:18:23] <locohost> "To last a couple of days"
[19-May-2011 02:05:53] <amorphic> Hello all - anyone here with a clue about process monitoring under zenoss?
[19-May-2011 02:06:11] <amorphic> I've been delving rather deeply into this and I've hit a bti of a snag
[19-May-2011 03:24:55] fOrsberg is now known as forsberg
[19-May-2011 03:55:01] <alib0b0> hello all
[19-May-2011 03:55:17] <alib0b0> I'm just new in zenoss
[19-May-2011 03:55:49] <alib0b0> and I wonder are there any way using zenoss to monitor a specific folder (or files) ?
[19-May-2011 03:56:26] <forsberg> since you can do a command, you can monitor _anything_ you want
[19-May-2011 03:57:03] <forsberg> you can make a small shell/perl/whatever script that returns number of files and you can graph it
[19-May-2011 03:57:03] <forsberg> and make alerts bases on threshold
[19-May-2011 03:57:08] <alib0b0> I mean, I have a sever centralized all log I have
[19-May-2011 03:57:29] <alib0b0> and now I want zenoss monitor that directory (contain log)
[19-May-2011 03:57:36] <alib0b0> is it ok ?
[19-May-2011 03:57:46] <hmp> what do you want to monitor? changes to the log file?
[19-May-2011 03:58:18] <alib0b0> I want to monitor all the event in the log files
[19-May-2011 03:58:27] <alib0b0> just like splunk do
[19-May-2011 03:58:31] <forsberg> how so monitor
[19-May-2011 03:58:50] <forsberg> tail it and look for something ?
[19-May-2011 03:58:55] <alib0b0> yes
[19-May-2011 03:59:07] <hmp> thats not efective, try inotify for monitoring
[19-May-2011 03:59:39] <forsberg> sure, anything you can do in a prompt you can make output a way that you would be able to monitor in zenoss
[19-May-2011 04:00:03] <forsberg> but there is no out of the box solution for this i think in zenoss
[19-May-2011 04:00:04] <forsberg> only windows event log
[19-May-2011 04:00:36] <alib0b0>
[19-May-2011 04:01:15] <forsberg> its a good thing man zenoss rocks
[19-May-2011 04:01:54] <forsberg> if you figured out how to do this, you would pretty much be able to monitor anything you would ever want
[19-May-2011 04:03:19] <hmp> also, depends of what you are trying to monitor, have a look at zensyslog also
[19-May-2011 04:04:00] <alib0b0> ok, thanks hmp & forsberg, I 'll try
[19-May-2011 04:04:53] <Jane_Curry> amorphic - what's your snag
[19-May-2011 04:05:14] <forsberg> yeah, i think you would have to be much more specific, to get a more precise help to there the solution may end
[19-May-2011 04:06:36] <Jane_Curry> I know you already have this log - whast architecture is it and what is in it?
[19-May-2011 04:07:08] <Jane_Curry> Could you arrange for these log messages to go instead / as well to either a Windows event log or a unix syslog?
[19-May-2011 04:07:20] <Jane_Curry> Then you get your events into zenoss for free
[19-May-2011 04:10:18] <alib0b0> We have some server to manage
[19-May-2011 04:10:49] <alib0b0> now, we have centralized all log from those server to a log server (use rsyslog)
[19-May-2011 04:11:17] <fragfutter> alib0b0: now you configure a forwaring of your central syslog server to the zenoss syslog receiver
[19-May-2011 04:11:42] <alib0b0> I know
[19-May-2011 04:12:08] <alib0b0> but currently, I want to monitoring & get alert from the log files I have centralized
[19-May-2011 04:12:40] <alib0b0> I have used splunk, It 's good but the licence quite expensive
[19-May-2011 04:12:59] <fragfutter> alib0b0: that's why you configure an _additional_ forward in your central syslog server. All logfiles will stay there, splunk can do anything that it wants to do with it. But also zenoss will receive them.
[19-May-2011 04:15:15] <alib0b0> fragfutter: ok, I have done that. But I want to know is there any way to set zenoss monitoring a specific folder
[19-May-2011 04:15:55] <fragfutter> as mister spock would say: Insuccificent data does not compute
[19-May-2011 04:16:45] <fragfutter> what do you mean by monitoring a specific folder. And as multiple persons in this channel told you, yes you can monitor anything you can write a comman for.
[19-May-2011 04:18:29] <alib0b0> ok, thanks
[19-May-2011 05:07:20] <Rickson->  im trying to monitor a esx host ( with this, i have added it but nothing appers in the graphs, and when i try model device, i get this error
[19-May-2011 05:21:44] <forsberg> if you install both dependencies correctly and type in the username/password correct both places it works
[19-May-2011 05:25:10] <Rickson-> hmm, whats the different between vsphareusername and zcommandusername?
[19-May-2011 05:33:16] <forsberg> i dont remember, but i remember something in the documentation stating you need it two places
[19-May-2011 05:33:26] <forsberg> and it worked flawlesly for me
[19-May-2011 05:33:39] <forsberg> after eric fixed something with the esxi 4.1 release
[19-May-2011 05:35:28] <Rickson-> well just to get it work, i use the root and root password on both of them
[19-May-2011 05:36:14] <Rickson-> but still problem
[19-May-2011 06:23:09] <mabuta> hi all
[19-May-2011 06:23:20] <mabuta> any bot to check for help?
[19-May-2011 06:30:23] <mabuta> any clue on distributed monitoring with 3.1 core?
[19-May-2011 06:58:25] forsberg is now known as fOrsberg
[19-May-2011 07:03:36] <mabuta> anyone using distributed collectors
[19-May-2011 07:03:38] <mabuta> ?
[19-May-2011 07:11:41] <Simon4> mabuta: egor's zenpack is v3 compatible I believe, I haven't used it personally but have heard pretty good things
[19-May-2011 07:12:11] <mabuta> yes indeed
[19-May-2011 07:12:58] <mabuta> on the zenpack page claims as 3.0 compatible, but i'm trying now without success to use it
[19-May-2011 07:13:50] <mabuta> i'm looking for some experiences to dig out the real state
[19-May-2011 07:14:05] <mabuta> egor's instructions are quite simple
[19-May-2011 07:15:03] <mabuta> i think i will try again to use it and then troubleshoot it as the thang was done manually
[19-May-2011 07:16:38] <mabuta> first of all: may i try to exclude loclahost from the collecotrs?
[19-May-2011 08:26:56] fOrsberg is now known as forsberg
[19-May-2011 08:28:33] <forsberg> hmm i have 21162 oids in 272 devices completed in 10-35 seconds
[19-May-2011 08:29:11] <forsberg> is there any way i can make it go slower, like turn down the number of threads doing it or whatever, so it takes longer time
[19-May-2011 08:29:12] <forsberg> anyone knows something ?
[19-May-2011 08:30:33] <hmp> forsberg: yes, you can lower the number of threads
[19-May-2011 08:31:24] <forsberg> know where? i looked through all daemons in settings
[19-May-2011 08:31:24] <forsberg> didnt seem to find anything
[19-May-2011 08:32:08] <hmp> what deamon?
[19-May-2011 08:32:35] <forsberg> hmm, im asking you know where i lower the number of threads?
[19-May-2011 08:32:38] <Simon4> forsberg: why do you want it to go slower?
[19-May-2011 08:33:01] <hmp> forsberg: many of the zenoss daemons are threaded, which one? zenmodeler?
[19-May-2011 08:34:10] <forsberg> long story short, we are getting some "false" alarms, stuff that answers whenever you test it, but sometimes zenoss makes alert, stupid for the 24/7 on call duty team, cause it fucks their sleep
[19-May-2011 08:34:20] <forsberg> aaaaand my boss thinks its because the zenoss box is to busy
[19-May-2011 08:34:25] <forsberg> i watch it, and its only busy those 10-35 seconds
[19-May-2011 08:34:35] <forsberg> figured i could test it by doing it slower
[19-May-2011 08:34:35] <forsberg> it just cant take more than 5 minutes
[19-May-2011 08:34:36] <forsberg> obviously
[19-May-2011 08:34:49] <mabuta> forsberg: did you have a look at the collector's properties?
[19-May-2011 08:35:03] <forsberg> no
[19-May-2011 08:35:07] <Simon4> forsberg: you could drop the maxOIDsPerRequest value
[19-May-2011 08:35:12] <forsberg> yes
[19-May-2011 08:35:14] <Simon4> that'll make it do more requests with less oids per time
[19-May-2011 08:35:14] <forsberg> where
[19-May-2011 08:35:14] <forsberg>
[19-May-2011 08:35:21] <mabuta> advanced
[19-May-2011 08:35:26] <mabuta> collectors
[19-May-2011 08:35:27] <Simon4> Configuration Properties, either for a device or a device class
[19-May-2011 08:35:34] <forsberg> oh btw im on 2.5.2
[19-May-2011 08:36:03] <forsberg> ah yes ok
[19-May-2011 08:36:14] <Simon4> or zProperties then
[19-May-2011 08:36:54] <forsberg> hm
[19-May-2011 08:37:14] <forsberg> found the collectors edit settings
[19-May-2011 08:37:29] <forsberg> n1, but not the maxOIDsPerRequest
[19-May-2011 08:37:42] <mabuta> my clue was different than the simon4 tip
[19-May-2011 08:37:47] <forsberg> yes i know
[19-May-2011 08:37:50] <mabuta> simon4 surely knows better than me
[19-May-2011 08:37:56] <forsberg> i dont mind 2 things to test ;>
[19-May-2011 08:38:22] <forsberg> yours was fine, i found the collector process, gonna try tinker a little with that
[19-May-2011 08:38:22] <mabuta> try first simon4...
[19-May-2011 08:38:22] <forsberg> but i also wanna try simons hehe ;>
[19-May-2011 08:38:57] <forsberg> yours was actually the one i asked for i think, Process Parallel Jobs
[19-May-2011 08:39:02] <forsberg> its 20
[19-May-2011 08:39:02] <forsberg> gonna try with 10 ;>
[19-May-2011 08:39:08] <Simon4> forsberg: zproperties, it's zMaxOIDPerRequest
[19-May-2011 08:39:23] <forsberg> yeah but zproperties on what ?
[19-May-2011 08:39:38] <Simon4> either the device or just set it on /Devices/ the device class
[19-May-2011 08:39:50] <Simon4> it inherits, so you could drop it globally, or just for a group of devs
[19-May-2011 08:40:26] <forsberg> ah yes
[19-May-2011 08:40:31] <forsberg> ok thanks!
[19-May-2011 08:40:46] * Simon4 votes for "get more disk IO" as a better solution
[19-May-2011 08:40:46] <forsberg> worst thing is, its our asterisk voip servers
[19-May-2011 08:41:23] <forsberg> everyone hates that part of our business, and now the technicians with the outside office hours duty get false alarms
[19-May-2011 08:41:23] <forsberg> hehe yes
[19-May-2011 08:41:33] <forsberg> im just gonna check it out so i can present the best solution to ze boss
[19-May-2011 08:41:38] <forsberg> gonna get some ssd's i guess
[19-May-2011 08:42:07] <Simon4> I run 6x15k drives in raid 0+1, speeds along okay, but I think SSD's are probably better value nowdays
[19-May-2011 08:42:19] <forsberg> i think so too
[19-May-2011 08:42:36] <forsberg> there are more than one benefit from doing it
[19-May-2011 08:42:39] <Simon4> I ran into an infrastructure team scared to put 3rd party ssd's in hp kit, so had to go down the drive route
[19-May-2011 08:42:49] <forsberg> heh
[19-May-2011 08:43:17] <forsberg> i can also understand that i guess, but we are already using ssd some places
[19-May-2011 08:43:18] <forsberg> so its not gonna be a problem
[19-May-2011 08:47:47] <forsberg> wo0aaa
[19-May-2011 08:48:02] <forsberg> process parallel jobs didnt work so well ^^
[19-May-2011 08:50:58] <mabuta>
[19-May-2011 08:51:53] <Hackman238> Disk I/O is the way to go. My smaller customers build large RAID 10's and larger customers use Fusion I/O cards
[19-May-2011 08:55:39] <forsberg> yeah
[19-May-2011 08:55:58] <forsberg> it writes in 21000 files during 12 seconds
[19-May-2011 08:55:58] <forsberg> how much io can i need
[19-May-2011 08:56:19] <forsberg> doesnt make sense, since the only limit is 5 minutes
[19-May-2011 08:56:59] <forsberg> i should/will/can/okok get ssd discs prolly anyways, but!
[19-May-2011 08:57:52] <locohost> mounting the fs no atime may help bring iodown a bit
[19-May-2011 08:58:33] * Simon4 wants to test one of those Fusion I/O cards
[19-May-2011 09:01:01] <locohost> forsberg, is iowait pegged at 100% durring the poll cycle?
[19-May-2011 09:01:23] <locohost> shoudl be able to tell with sar if you have sysstat installed on the box
[19-May-2011 09:03:21] <locohost> me want one
[19-May-2011 09:03:25] <forsberg> no
[19-May-2011 09:05:07] <locohost> it would be nice to read rainbow tables off of
[19-May-2011 09:06:54] <locohost> is it swapping, are you running out of memory, cpu load like 50 or some crap?
[19-May-2011 09:12:29] <forsberg> nopes
[19-May-2011 09:12:35] <forsberg> cpu load is at 50
[19-May-2011 09:12:36] kocolosk is now known as kocolosk|away
[19-May-2011 09:12:46] <forsberg> for the duration on all 4 cores
[19-May-2011 09:12:58] <forsberg> no swapping or anything else that makes the box seems slow
[19-May-2011 09:13:24] <Simon4> IOwait up? (top will show)
[19-May-2011 09:13:24] <forsberg> sometimes the gui can stall for a while also (10ish secs)
[19-May-2011 09:13:24] <forsberg> rarelyt
[19-May-2011 09:13:39] <Simon4> 21k files sounds about right for toasting a raid-1 set
[19-May-2011 09:15:41] <locohost> well, load does not factor in the number of cores, so that is about equivelent to a load of 12 on a single core box
[19-May-2011 09:15:44] <locohost> still high
[19-May-2011 09:15:49] <locohost> physical or virtual?
[19-May-2011 09:16:10] <forsberg> no iowait
[19-May-2011 09:16:14] <forsberg> stays at 0
[19-May-2011 09:16:28] <locohost> im surprized iowait is not high, io contention will make your cpu load go up, the cpu is busy waiting for io..hrm
[19-May-2011 09:16:33] <locohost> thats really sruprizing
[19-May-2011 09:16:38] <Simon4> forsberg: that seems wierd
[19-May-2011 09:16:46] <Simon4> you should see some spike in iowait right at collection time
[19-May-2011 09:16:56] <Simon4> even if it's only 20-30% (but I would think a bit more)
[19-May-2011 09:17:09] <forsberg> i already got an ibm 3650 im gonna mirror ssd and copy the installation over
[19-May-2011 09:17:26] <locohost> in my vm that is doing nothing but serving my shell session, iowait hits 1% at least
[19-May-2011 09:18:00] <forsberg> im lying
[19-May-2011 09:18:07] <forsberg> Cpu3  :  4.6%us,  1.3%sy,  0.0%ni, 13.2%id, 80.1%wa,  0.0%hi,  0.7%si,  0.0%st
[19-May-2011 09:18:11] <forsberg> thats not during the cycle tho
[19-May-2011 09:18:16] <locohost> okay
[19-May-2011 09:18:22] <forsberg> thats constant
[19-May-2011 09:18:39] <locohost> so, 80% of that load of 50, the cpu is not really doing anything, just waiting for disks
[19-May-2011 09:19:12] <locohost> is this raid 5?
[19-May-2011 09:19:55] <forsberg> i think its actually just a mirror
[19-May-2011 09:19:56] <forsberg> 10k discs
[19-May-2011 09:20:00] <forsberg> so yeah
[19-May-2011 09:20:15] <locohost> well, if its just a mirrior, as in raid 1, thats faster then raid 5
[19-May-2011 09:20:20] <forsberg> not gonna use much more time on it, gonna get the ibm up next week and copy the shit over and see
[19-May-2011 09:20:31] <forsberg> 2 weeks before i get the oncall duty so ;>
[19-May-2011 09:20:31] <forsberg> raid 1 yeah
[19-May-2011 09:20:36] <locohost> 15k disks sound like they would make a huge difference
[19-May-2011 09:20:37] <Simon4> forsberg: right, that's your reason for slow
[19-May-2011 09:20:43] <Sam-I-Am> moo.
[19-May-2011 09:20:44] <forsberg> no, i know its slow
[19-May-2011 09:20:45] <locohost> i would also try mounding the disks noatime
[19-May-2011 09:20:49] <forsberg> i always knew that
[19-May-2011 09:20:53] <forsberg> i tried that now locohost
[19-May-2011 09:20:56] <forsberg> just rebooted it
[19-May-2011 09:20:59] <locohost> that may buy you 2 or 3% w/o having to do anything
[19-May-2011 09:21:22] <Simon4> forsberg: is your mysql db on the same spindles as your rrd's as your /opt/zenoss/var ?
[19-May-2011 09:21:31] <locohost> w/o that, when you read files and dont write to them, its still wrighting to the disk..just saying the timestamp of when it was read
[19-May-2011 09:22:13] <forsberg> but still, i hope you can see reason in my logic, if its very loaded for 12 seconds and gives false alarms, AND if its then not doing anything at all for 4 mins and 48 seconds
[19-May-2011 09:22:13] <forsberg> yes
[19-May-2011 09:22:13] <forsberg> simon
[19-May-2011 09:22:13] <forsberg> that too
[19-May-2011 09:22:33] <Simon4> separating those three will make a huge difference also
[19-May-2011 09:22:42] <Sam-I-Am> sup nerds
[19-May-2011 09:23:11] <forsberg> cool thanks, gonna check it out
[19-May-2011 09:30:43] <kokey> my mysql and rrd and zodb are all on the same partition
[19-May-2011 09:30:45] <kokey> mounted over NFS
[19-May-2011 09:30:55] <kokey> from a virtual machine
[19-May-2011 09:31:07] <kokey> we've only got 500 devices so it's not that hardcore
[19-May-2011 09:36:08] <dhopp> kokey: how many data points are you getting roughly?
[19-May-2011 09:37:46] <kokey> dhopp: not sure, 51,000 RRD files tho
[19-May-2011 09:38:54] <dhopp> well it's one datapoint per RRD file typically so I guess that's the answer :-)
[19-May-2011 09:39:16] <kokey> before zenoss i had something do my own RRDs
[19-May-2011 09:39:16] <dhopp> so you have just one zenoss server or multiple collectors?
[19-May-2011 09:39:32] <kokey> and i had about 4 data points per RRD on average
[19-May-2011 09:39:43] <kokey> dhopp: just one zenoss server
[19-May-2011 09:40:04] <dhopp> how many people access the web interface typically?
[19-May-2011 09:40:13] <kokey> about 8 people at least
[19-May-2011 09:40:20] <kokey> i'm running on about 6GB of ram
[19-May-2011 09:40:29] <kokey> i've pushed the zope cache up a lot
[19-May-2011 09:40:36] <kokey> so the interface is fine
[19-May-2011 09:40:46] <kokey> mysql is probably tuned quite well, and I only keep 14 days of history
[19-May-2011 09:41:01] <kokey> mysql is all innodbed and i have set dirty reads
[19-May-2011 09:41:26] <Simon4> that'll def help
[19-May-2011 09:41:42] * Simon4 had to tune the web interface for ~200 simul users
[19-May-2011 09:41:44] <Simon4> that hurt
[19-May-2011 09:42:13] <kokey> it would have been nice to have the interface run a bit more snappy
[19-May-2011 09:42:41] <kokey> but i figure in that case i'll be throwing a ton of RAM at it and let zope cache like a mofo
[19-May-2011 09:42:52] <kokey> and plenty CPU cores
[19-May-2011 09:43:08] <kokey> but i'm out of here next week
[19-May-2011 09:43:20] <kokey> don't think i'll be doing much zenoss at my next gig
[19-May-2011 09:43:28] <kokey> i'll be building dashboards and things though
[19-May-2011 09:43:41] <kokey> so i might be tempted to make a zenoss friendly RRD viewer
[19-May-2011 09:44:09] <kokey> i've done some work at my last job to makae a nice graph selecting interface, better than most i've seen out there
[19-May-2011 09:44:37] <kokey> only beaten by circonus really
[19-May-2011 09:45:14] kocolosk|away is now known as kocolosk
[19-May-2011 09:48:33] <ashp> monitoring: good, total site power outage taking down the monitoring system too: bad
[19-May-2011 09:48:50] <Sam-I-Am> well, at least you werent getting paged a ton
[19-May-2011 09:49:40] <dhopp> I'm looking at probably 60-70 simultaneous users at peak..and average about 30-35
[19-May-2011 09:49:54] <dhopp> we will have to keep a lot of history though do to some of our contracts with clients
[19-May-2011 09:51:24] <kokey> ashp: had that a lot with data centres that have everything set up with backup power except not properly for aircon
[19-May-2011 10:00:28] forsberg is now known as fOrsberg
[19-May-2011 10:07:02] <Sam-I-Am> i just love how search doesnt work for limited users in zenoss
[19-May-2011 10:07:09] <Sam-I-Am> it just blows up and throws an error
[19-May-2011 10:09:33] <Hackman238> For the best web experience, I reccomend tens of Zope's session balanced behind apache.
[19-May-2011 10:27:33] <dhopp> anybody know what the heartbeat timer is?
[19-May-2011 10:27:55] <dhopp> so how long after a zenoss daemon crashes or is stopped will a hearbeat notification show up?
[19-May-2011 10:28:13] <Simon4> three collection cycles
[19-May-2011 10:28:24] <dhopp> so 15 minutes?
[19-May-2011 10:28:32] <dhopp> in the case of zenperfsnmp
[19-May-2011 10:28:32] <Simon4> yah
[19-May-2011 10:28:36] <dhopp> ack
[19-May-2011 10:30:19] <kokey> enough to make gaps in your graphs
[19-May-2011 10:30:20] <kokey> hehe
[19-May-2011 10:30:27] <dhopp> Hackman238: are you saying you have 10 zope instances running behind apache?
[19-May-2011 10:30:32] <Simon4> not the dreaded gaps!
[19-May-2011 10:30:43] * Simon4 has 8 zopes behind apache
[19-May-2011 10:30:52] <dhopp> I'm not worried about the graphs as I am worried about something bad happening in that 15 minutes
[19-May-2011 10:30:54] <kokey> how do you create more zope instances?
[19-May-2011 10:31:01] <Simon4> kokey: there's a patch
[19-May-2011 10:31:16] <kokey> Simon4: ah ok, it didn't look like there was a straightforward way
[19-May-2011 10:31:24] <Simon4>
[19-May-2011 10:31:50] <dhopp> Simon4: how much RAM do you have in your web machine?
[19-May-2011 10:32:10] <Simon4> um, 24GB
[19-May-2011 10:33:00] <kokey> does zope talk to zeo over tcp?  as in can you run more zopes on another server?
[19-May-2011 10:34:18] <dhopp> Simon4: is it only the web interface or is it running all the other daemons?  Do you have remote collectors too?
[19-May-2011 10:34:20] <Sam-I-Am> hmm, guess i'll need to give at least zenuser access to everyone for everything
[19-May-2011 10:34:27] <Sam-I-Am> not having search is going to confuse the hell out of my users
[19-May-2011 10:34:41] <Simon4> dhopp: it's web interface + zenhub + zeo
[19-May-2011 10:34:45] <Simon4> we have remote collectors
[19-May-2011 10:35:07] <dhopp> ok..sounds like you are having it similar to what I'm going to do...
[19-May-2011 10:35:21] <dhopp> are you using egor's distributed collectors zenpack?
[19-May-2011 10:35:33] <Sam-I-Am> i have apache in front of zop
[19-May-2011 10:35:35] <Sam-I-Am> zope
[19-May-2011 10:36:51] <kokey> i have apache in front of my single zope at the moment
[19-May-2011 10:40:49] <dhopp> Simon4: are you using egor's distributed collectors zenpack?
[19-May-2011 10:43:43] <Sam-I-Am> i'm using enterprise
[19-May-2011 10:52:08] <kokey> is enterprise one of egor's zenpacks?
[19-May-2011 10:52:15] <kokey> ;-)
[19-May-2011 10:52:17] <Sam-I-Am> hah
[19-May-2011 10:52:51] <Simon4> nahh, enterprise is slightly different, and more polished in some ways, less so in others
[19-May-2011 10:53:29] <Sam-I-Am> Simon4: good description
[19-May-2011 10:53:43] * Simon4 tells it like it is
[19-May-2011 10:59:13] <Sam-I-Am> permissions are a joke in enterprise
[19-May-2011 10:59:21] <Sam-I-Am> needs more egorpack for it
[19-May-2011 11:01:38] <kokey> hehe
[19-May-2011 11:16:47] <mabuta> still need some help on egor zenpack on 3.1 stack. anybody had experience?
[19-May-2011 11:37:00] <dhopp> mabuta: which egor zenpack?  he has a few :-)
[19-May-2011 11:37:22] <mabuta> the distributed collector
[19-May-2011 11:37:36] <dhopp> what problem are you having?  I just installed it this morning
[19-May-2011 11:37:38] <mabuta> is there a dumbproof way?
[19-May-2011 11:37:40] <dhopp> on stack
[19-May-2011 11:38:15] <mabuta> which stack version? are you able to see events of distributing monitored devices and their graph'
[19-May-2011 11:38:29] <mabuta> i'm on ubuntu 10.04, stack 3.1.0
[19-May-2011 11:38:34] <dhopp> I haven't gotten that far :-P
[19-May-2011 11:38:38] <dhopp> I'm using stack 3.1.0
[19-May-2011 11:38:48] <dhopp> I did verify that I get heartbeats from the remote collectors
[19-May-2011 11:38:53] <dhopp> let me throw a device or two up there
[19-May-2011 11:41:09] <mabuta> did you have experience with the same scenario, using the manual setup @
[19-May-2011 11:41:44] <mabuta> i can use a previous version too since, proven it works
[19-May-2011 11:41:53] <dhopp> I haven't actually done the manual configuration in a while so I don't remember
[19-May-2011 11:42:14] <mabuta> here and there it's told to use dns instead of ip
[19-May-2011 11:42:36] <mabuta> may i use just ip and zvoid dns?
[19-May-2011 11:47:35] <dhopp> I think so
[19-May-2011 11:47:56] <mabuta> thanks
[19-May-2011 11:55:32] <rmatte> good day all
[19-May-2011 11:55:38] <mabuta> hi
[19-May-2011 12:00:01] [disconnected at Thu May 19 12:00:01 2011]
[19-May-2011 12:00:02] [connected at Thu May 19 12:00:02 2011]
[19-May-2011 12:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[19-May-2011 12:17:21] <dhopp> mabuta: I threw a device in and assigned it to the distributed collector and I can see the graphs.  I'm using DNS
[19-May-2011 12:30:51] <locohost> darn walkups, its lunch time already
[19-May-2011 17:27:18] kocolosk is now known as kocolosk|away
[19-May-2011 18:40:29] kocolosk|away is now known as kocolosk
[19-May-2011 22:36:39] <davetoo> anybody awake?
[19-May-2011 22:51:01] kocolosk is now known as kocolosk|away
[19-May-2011 22:55:15] <Sam-I-Am> maybe
[19-May-2011 23:03:53] <Hackman238b> Maybe? LOL
[19-May-2011 23:04:54] <Sam-I-Am> trying to clean up this vista machine
[19-May-2011 23:05:02] <Sam-I-Am> making my head hurt
[19-May-2011 23:08:20] <Hackman238b> Vista!
[19-May-2011 23:08:26] <Sam-I-Am> yes
[19-May-2011 23:08:55] <Hackman238b> Blasphemy! LOL Should be using Linux :-)
[19-May-2011 23:08:56] <Sam-I-Am> nothing like a store bought HP consumer-grade laptop to rake in the consulting cash
[19-May-2011 23:09:05] <Hackman238b> LOL!
[19-May-2011 23:09:17] <Sam-I-Am> it would be quicker just to format this thing and install 7
[19-May-2011 23:09:54] <Hackman238b> Win7 performs pretty well. Lots of kernel and GUI schedule tweaks.
[19-May-2011 23:09:59] <Sam-I-Am> yeah it does
[19-May-2011 23:10:06] <Sam-I-Am> the UAC on this thing is killing me
[19-May-2011 23:10:12] <Sam-I-Am> or just making things take that much longer
[19-May-2011 23:10:19] <Hackman238b> Win7 UAC isnt a tenth as bad either LOL
[19-May-2011 23:10:26] <Sam-I-Am> the laptop was to the point where the hdd would grind from power up to eternity
[19-May-2011 23:10:48] <Sam-I-Am> spyware, malware, a handful of virus scanners
[19-May-2011 23:10:53] <Hackman238b> Yikes! Have enough RAM?
[19-May-2011 23:11:01] <Sam-I-Am> enough to run all the spyware
[19-May-2011 23:11:15] <Hackman238b> LOL! Oh my. Is a customer laptop?
[19-May-2011 23:11:22] <Sam-I-Am> consulting gig
[19-May-2011 23:11:30] <Sam-I-Am> i dont run windows myself
[19-May-2011 23:12:00] <Hackman238b> I'd wipe it and reload with Win7 if they'll buy a license. Tell them data backup/restore is extra
[19-May-2011 23:12:10] <Sam-I-Am> heh
[19-May-2011 23:12:30] <Hackman238b> What I do for my two retail PC shops is offer a contract.
[19-May-2011 23:13:41] <Hackman238b> A customer can 1) pay for the wipe and reload + data(optional) or 2) pay for the wipe + reload + data(optional) + contract and henceforth during the contract period can have their PC/Laptop/Mac reimaged back to its last reinstall state for $50, unlimited times.
[19-May-2011 23:14:06] <Hackman238b> Customers with kids or careless employees spring for the contract after having to wipe + reload a few times.
[19-May-2011 23:14:21] <Sam-I-Am> hmmm interesting, 'data execution prevention' killed the windows installer
[19-May-2011 23:14:35] <Hackman238b> Wow- that doesnt sound good.
[19-May-2011 23:14:39] <Sam-I-Am> yeh
[19-May-2011 23:14:51] <Sam-I-Am> thing is still grinding strong even with the easy cruft removed
[19-May-2011 23:14:52] <Hackman238b> Must be a modified trustedinstaller
[19-May-2011 23:14:59] <Hackman238b> Hence no longer trusted :-)
[19-May-2011 23:15:12] <Sam-I-Am> i dont do enough consulting work... should do more
[19-May-2011 23:15:26] <Sam-I-Am> in the market for another gig anyway
[19-May-2011 23:15:53] <Sam-I-Am> this laptop also warms my legs
[19-May-2011 23:17:49] <Hackman238b> Consulting is where the big money is at, but you can compete as a serious business in the local area with a small retail front doing wipe/reload and contracts. Personally we try to target small/medium businesses which always have no inhouse IT solution. Retail operations are essentially a means of advertising the system to the local population inexpensively.
[19-May-2011 23:18:23] <Sam-I-Am> yep
[19-May-2011 23:18:36] <Hackman238b> :-) Then again, I'm always up to my eye balls in work
[19-May-2011 23:18:48] <Sam-I-Am> but it pays the bills
[19-May-2011 23:18:49] <Hackman238b> Where are you at Sam?
[19-May-2011 23:18:58] <Sam-I-Am> kansas
[19-May-2011 23:20:02] <Hackman238b> I only have ops in NY, CA and TX, nothing in Kansas. Is there a fairly large demand for IT work in the region?
[19-May-2011 23:20:13] <Sam-I-Am> i have no plans to stay in kansas
[19-May-2011 23:20:43] <Hackman238b> Ah gotcha
[19-May-2011 23:21:00] <Sam-I-Am> where in CA?
[19-May-2011 23:21:06] <Hackman238b> Should come to Texas :-)
[19-May-2011 23:21:12] <Hackman238b> San Marcos, CA
[19-May-2011 23:21:28] <Sam-I-Am> where in texas?
[19-May-2011 23:21:30] <Hackman238b> Watervliet, NY
[19-May-2011 23:21:37] <Hackman238b> San Antonio, TX
[19-May-2011 23:22:27] <Sam-I-Am> hmm san marcos is near san diego... nice weather
[19-May-2011 23:22:38] <Sam-I-Am> you looking to do any zenoss consulting?
[19-May-2011 23:22:52] <Hackman238b> Its fantastic there...but the economy is failing hard in CA and NY. Do not move there.
[19-May-2011 23:23:06] <Hackman238b> I do Zenoss consulting, yes.
[19-May-2011 23:23:22] <Sam-I-Am> hmm, figured CA would be the place for IT jobs
[19-May-2011 23:23:29] <Sam-I-Am> that or NY and chicago
[19-May-2011 23:24:14] <Hackman238b> You're absolutely correct, but at this time (since late 2007) credit is tight with small/medium businesses which limits growth
[19-May-2011 23:24:54] <Hackman238b> NY on the otherhand is a disaster of redtape, taxes, and negative growth in many areas. I'd avoid it as well for the time.
[19-May-2011 23:24:55] <Sam-I-Am> that could be a problem
[19-May-2011 23:25:00] <Sam-I-Am> not quite looking to run my own biz
[19-May-2011 23:25:09] <Sam-I-Am> kinda looking to get back to colorado
[19-May-2011 23:25:46] <Hackman238b> If you're looking for work, I reccomend RackSpace. I do work for them and they are an excellent company
[19-May-2011 23:26:54] <Sam-I-Am> so i've heard
[19-May-2011 23:27:02] <Hackman238b> I've only just driven through Colorado- very nice state.
[19-May-2011 23:27:29] <Sam-I-Am> i lived there for 4 years
[19-May-2011 23:27:35] <Sam-I-Am> only left when the economy went poo
[19-May-2011 23:27:46] <Hackman238b> Has it improved since?
[19-May-2011 23:27:57] <Sam-I-Am> yeah, but not in the place i lived
[19-May-2011 23:28:02] <Sam-I-Am> more in the denver area
[19-May-2011 23:28:09] <Hackman238b> Frankly there are few places in the US with posative growth right now.
[19-May-2011 23:28:16] <Hackman238b> *positive
[19-May-2011 23:28:22] <Sam-I-Am> so whats the deal in san antonio/austin?
[19-May-2011 23:28:26] <Sam-I-Am> i know zenoss is down there
[19-May-2011 23:28:45] <Hackman238b> Zenoss and RackSpace
[19-May-2011 23:29:23] <Hackman238b> The job market is good here in San Antonio, cost of living is low, and the weather isn't *too* bad
[19-May-2011 23:29:58] <Sam-I-Am> i have a bit of trouble with sticky hot weather... lived in florida for 17 years and hated it.  but... lack of winter would be a nice touch.
[19-May-2011 23:30:26] <Sam-I-Am> as long as october-may or so is nice, i can sit in the a/c for the other few months
[19-May-2011 23:30:30] <Hackman238b> Its definately hot here. Its muggy sometimes, but nothing like Flordia.
[19-May-2011 23:30:55] <Sam-I-Am> colorado's summer made up for the winter... and in kansas i get about 3 weeks on either side of summer and winter to enjoy
[19-May-2011 23:31:07] <Hackman238b> Few places compare to San Diego county though :-)
[19-May-2011 23:31:12] <Sam-I-Am> well, yeah
[19-May-2011 23:31:22] <Sam-I-Am> i'm a cyclist and a pilot, so i like nice weather
[19-May-2011 23:31:44] <Hackman238b> Very cool- what do you fly?
[19-May-2011 23:31:46] <Sam-I-Am> however, i'm also a storm chaser and having nice weather year round may drive me nuts... but i might be cycling and flying too much to care
[19-May-2011 23:32:11] <Sam-I-Am> mostly just small cessna stuff... 152, 172
[19-May-2011 23:32:28] <Hackman238b> Hahaha- one does miss the weather occasionally, but not enough to be a serious problem
[19-May-2011 23:32:35] <Hackman238b> Cool
[19-May-2011 23:32:56] <Sam-I-Am> is it windy a lot down there?
[19-May-2011 23:33:15] <Hackman238b> Its often windy, yes.
[19-May-2011 23:33:19] <Sam-I-Am> doh
[19-May-2011 23:33:30] <Hackman238b> Very annoying, but helpful in the heat.
[19-May-2011 23:33:37] <Sam-I-Am> not so good for flying
[19-May-2011 23:33:55] <Hackman238b> Not so good for flying or driving a light car LOL
[19-May-2011 23:34:00] <Sam-I-Am> what sort of consulting work do you do for rackspace?
[19-May-2011 23:34:39] <Hackman238b> Zenoss, but I'm on with RackSpace as a full time employee...frankly I work too much, but I like to work.
[19-May-2011 23:34:45] <davetoo> I'm in Irvine right now, and..
[19-May-2011 23:35:12] <davetoo> the place I'm at has more exotic cars in their parking lot than anyplace around home other than Sand Hill Road (where all the VCs are)
[19-May-2011 23:36:10] <Hackman238b> Anything good?
[19-May-2011 23:36:21] <Sam-I-Am> Hackman238b: oh, i thought you had a consulting biz... think we've talked about this before.
[19-May-2011 23:36:34] <Sam-I-Am> rackspace and their stupidly large zenoss install or something
[19-May-2011 23:36:35] <Hackman238b> I do. :-)
[19-May-2011 23:36:56] <davetoo> Lotus Esprit,
[19-May-2011 23:37:08] <Hackman238b> I have several businesses and my fulltime with RackSpace.
[19-May-2011 23:37:09] <davetoo> an M3 with many carbon fibre panels,
[19-May-2011 23:37:21] <Hackman238b> M3's always nice
[19-May-2011 23:37:26] <davetoo> saw a Bentley in the Albertson's parking lot yesterday
[19-May-2011 23:37:45] <davetoo> some of the new Jags
[19-May-2011 23:37:51] <Hackman238b> Ugh...drives a Bently but goes to none other than Albersons! LOL
[19-May-2011 23:38:07] <Hackman238b> I always disliked Albersons comparred to Vons
[19-May-2011 23:38:07] <davetoo> which I think look like they tried to hammer Jaguar cars into Mazda body molds
[19-May-2011 23:38:17] <Hackman238b> Hahaha
[19-May-2011 23:38:30] <davetoo> Ford owned them both at one time, not sure if they still do?
[19-May-2011 23:38:53] <davetoo> Porsches are a dime a dozen,
[19-May-2011 23:39:02] <Hackman238b> Ford owns both yet, I believe
[19-May-2011 23:39:09] <davetoo> and I've even seen a couple of the new 4-door 911s, whatever those are called
[19-May-2011 23:40:01] <Sam-I-Am> Hackman238b: heh, no wonder you're busy
[19-May-2011 23:40:27] <davetoo> ok, I'm ... sleeeepy
[19-May-2011 23:40:32] <Sam-I-Am> sleep is for the weak
[19-May-2011 23:40:39] <Hackman238b> LOL!
[19-May-2011 23:40:39] <davetoo> not much last night, and food coma now.
[19-May-2011 23:40:59] <davetoo> l8r
[19-May-2011 23:40:59] <Hackman238b> Yes, always busy busy. I'm on a call with a customer in IL right now rolling out Zenoss with chef
[19-May-2011 23:41:07] <Hackman238b> davetoo: Later
[19-May-2011 23:41:11] <davetoo> Illinois or Israel?
[19-May-2011 23:41:13] <davetoo> bye
[19-May-2011 23:41:20] <Hackman238b> Illinois.
[19-May-2011 23:42:11] <Sam-I-Am> seems the chef thing is popular
[19-May-2011 23:42:49] <Hackman238b> Yeah, its a fork of puppet.
[19-May-2011 23:43:00] <Sam-I-Am> heh, i find it amusing how an HD camcorder can be $250 and a neutral density filter run $190
[19-May-2011 23:43:04] <Hackman238b> Personally...I like puppet- but thats just because I'm used to it.
[19-May-2011 23:44:10] <Sam-I-Am> hey look, the laptop is behaving
[19-May-2011 23:44:17] <Hackman238b> Well they sell enough of the cam's to everyday people that they can work volume. Special accessories need to be expensive to justify stocking since less population will buy them regularly.
[19-May-2011 23:44:24] <Sam-I-Am> yep
[19-May-2011 23:44:29] <Sam-I-Am> stupid expensive hobbies :/
[19-May-2011 23:44:54] <Hackman238b> I hear that- I've got my own money ditches LOL
[19-May-2011 23:45:46] <Hackman238b> Personally I'm big in to developing hardware and high end audio. Kills me Fujitsu-Tens Eclipse is no longer making subs- they were so clean sounding.
[19-May-2011 23:45:49] <Sam-I-Am> cycling, flying, photography... but particularly flying.
[19-May-2011 23:45:59] <Sam-I-Am> mmm audio
[19-May-2011 23:46:35] <Hackman238b> 7.2 DTS surround in my Civic makes traffic no issue :-)
[19-May-2011 23:48:46] <Sam-I-Am> i still have 2 speakers
[19-May-2011 23:48:53] <Sam-I-Am> but they're good speakers
[19-May-2011 23:49:11] <Sam-I-Am> some time ago i bought a surround sound amp and just never expanded to more speakers... mostly because they're hideously expensive
[19-May-2011 23:49:50] <Hackman238b> They are, yes.
[19-May-2011 23:50:11] <Hackman238b> You can hit up ebay for discontinued Eclipse subs and interiors
[19-May-2011 23:50:57] <Hackman238b> Long before my days of regular IT madness I did car audio systems LOL
[19-May-2011 23:51:56] <Hackman238b>
[19-May-2011 23:52:02] <Sam-I-Am> i'm upgrading the car stereo this weekend... $80
[19-May-2011 23:52:27] <Hackman238b> Oh my goodness
[19-May-2011 23:53:56] <Sam-I-Am> yeah its bad
[19-May-2011 23:54:28] <Sam-I-Am> the most painful part is trying to take the dash apart without cracking 12 year old plastic
[19-May-2011 23:54:51] <Hackman238b> What are you driving?
[19-May-2011 23:55:08] <Sam-I-Am> toyota solara
[19-May-2011 23:55:35] <Hackman238b> If the clips are very tight hit the area with a low temp heat gun to make it more pliable
[19-May-2011 23:56:11] <Hackman238b> Not too hot, just enough to increase plasticity for the working time. It'll help prevent cracking if the plastic hasnt already cracked
[19-May-2011 23:56:36] <Sam-I-Am> that might work
[19-May-2011 23:56:51] <Sam-I-Am> might also just grab a replacement bezel
[19-May-2011 23:57:29] <Hackman238b> Can always do that, but is extra expense.
[19-May-2011 23:57:49] <Sam-I-Am> yep.  looks a lot nicer than some cracked mess though
[19-May-2011 23:57:56] <Sam-I-Am> trying to keep the interior somewhat decent
[19-May-2011 23:58:06] <Hackman238b> True true.
[19-May-2011 23:58:37] <Sam-I-Am> did the yearly detailing job on the other car last weekend... need to do this one soon plus an interior job.
[20-May-2011 00:00:01] [disconnected at Fri May 20 00:00:01 2011]
[20-May-2011 00:00:02] [connected at Fri May 20 00:00:02 2011]
[20-May-2011 00:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[20-May-2011 00:00:20] <Hackman238b> Yep.
[20-May-2011 00:00:47] <Hackman238b> People around SA are always cracking on me since I have NY plates and my car still has salt on it from this past winter :-)
[20-May-2011 00:01:59] <Hackman238b> I figure it this way- the Civic is my work car, IDC how it looks. I keep the 3000GT VR-4 nice and clean for when I need to go out and show off :-)
[20-May-2011 00:03:36] <Sam-I-Am> mmm 3000 gt
[20-May-2011 00:03:41] <Sam-I-Am> dont see too many of those
[20-May-2011 00:05:00] <Hackman238b> Is the VR-4 model- very nice.
[20-May-2011 00:05:19] <Hackman238b> '95, so its a pain in the nuts to work on the interior. LOL
[20-May-2011 00:07:49] <Sam-I-Am> well, i should get to bed.
[20-May-2011 00:08:07] <Sam-I-Am> or at least go stare at a dark ceiling for a while
[20-May-2011 00:11:37] <Hackman238b> LOL. Later man
[20-May-2011 00:13:15] <Sam-I-Am> heh, this network engineering job at rackspace 'requires' a ccna
[20-May-2011 00:13:32] * Sam-I-Am was around a few mins longer
[20-May-2011 00:16:46] <Sam-I-Am> zzzzz
[20-May-2011 00:25:26] <Hackman238b> Sam-I-Am: If you're interested in a position, forward your resume to and I'll get you interviewed
[20-May-2011 04:06:02] <zykes-> you work at rackspace Sam-I-Am ?
[20-May-2011 04:47:19] <xpf> Hi. Where can I get OID wich zenoss use to retreive SNMP-agent status?
[20-May-2011 05:35:57] <xpf> "Unable to send password reset email: - timed out"
[20-May-2011 05:36:01] <xpf> wtf?
[20-May-2011 09:07:13] <kokey> is there a way to change something to make it show more than just the production state devices when you navigate to infrastructure
[20-May-2011 09:07:17] <kokey> ?
[20-May-2011 09:09:11] <fragfutter> checking the checkbox?
[20-May-2011 09:10:21] <kokey> yeah but when you navigate away and come back it defaults to production only
[20-May-2011 09:24:27] <Sam-I-Am> .
[20-May-2011 09:26:04] <fragfutter> .
[20-May-2011 09:27:03] <Simon4> moo
[20-May-2011 09:27:49] <Sam-I-Am> sup
[20-May-2011 09:28:23] <Simon4> nommuch
[20-May-2011 09:39:40] <Sam-I-Am> sotired
[20-May-2011 09:39:45] <Sam-I-Am> ====C coffee
[20-May-2011 10:31:58] <kokey> where is setHWProductKey defined?
[20-May-2011 10:32:58] <fragfutter> kokey: grep -r -H $ZENHOME/Products -e 'def setHWProductKey'
[20-May-2011 10:33:51] <kokey> fragfutter: haha, i grepped lib and the zenpacks
[20-May-2011 10:33:56] <kokey> thanks, found it
[20-May-2011 10:34:03] <Simon4> Products is where all the good stuff is
[20-May-2011 10:38:24] <fragfutter> i have a broken device in my zodb. i can't delete it through the GUI. is there a way to trash it from zendmd?
[20-May-2011 10:43:21] <kokey> can one run 32 bit python libraries with a 64 bit python?
[20-May-2011 10:43:24] <kokey> we have a 32 bit python API for an application, but 64 bit python on the box
[20-May-2011 10:43:56] <kokey> fragfutter: ah, we're trying to write a modeler that modifies a custom property
[20-May-2011 10:44:08] <fragfutter> kokey: assumin cpython and that your 32bit python library wraps a c-lib. no.
[20-May-2011 10:44:23] <kokey> so I'm assuming we can use getmap, and that points to a def for it that executes a d.dmd.setZenProperty
[20-May-2011 10:44:56] <kokey> fragfutter: yeah it's looking like we'll have to keep a 32 bit python on the box for this purpose
[20-May-2011 10:45:06] <kokey> blah
[20-May-2011 10:49:18] <Sam-I-Am> ywp
[20-May-2011 10:49:22] <Sam-I-Am> er, yep
[20-May-2011 10:55:20] <kokey> what's a good way to modify a custom property from a modeler?
[20-May-2011 11:01:37] <Simon4> I think you can just do om.customProperty = "foo" before returning om
[20-May-2011 11:10:10] <fragfutter> anyone an idea how to throw a broken object out of the zopedb?
[20-May-2011 11:12:03] <fragfutter> i try do delete it but get a "value does not exist"
[20-May-2011 11:34:25] <kokey> i think you can do it if edit the zopedb file with vim
[20-May-2011 11:42:28] kocolosk|away is now known as kocolosk
[20-May-2011 11:43:01] [disconnected at Fri May 20 11:43:01 2011]
[20-May-2011 12:00:01] [connected at Fri May 20 12:00:01 2011]
[20-May-2011 12:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[20-May-2011 15:37:39] <Hackman238> Anyone ever had all severity 5 events disappear from mysql?
[20-May-2011 16:35:03] <Sam-I-Am> Hackman238: nope
[20-May-2011 17:48:02] kocolosk is now known as kocolosk|away
[20-May-2011 19:59:18] kocolosk|away is now known as kocolosk
[20-May-2011 20:09:08] kocolosk is now known as kocolosk|away
[21-May-2011 00:00:01] [disconnected at Sat May 21 00:00:01 2011]
[21-May-2011 00:00:01] [connected at Sat May 21 00:00:01 2011]
[21-May-2011 00:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[21-May-2011 12:00:01] [disconnected at Sat May 21 12:00:01 2011]
[21-May-2011 12:00:02] [connected at Sat May 21 12:00:02 2011]
[21-May-2011 12:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[21-May-2011 18:16:50] <akafritz> hello.  anyone know how to create a datapoint that is the sum of other datapoints?  not a graph, an actual datapoint that I can create a threshold for
[21-May-2011 18:17:38] <akafritz> basically i'm grabbing values returned from a database query, and want to create a threshold when the sum of those results reaches a value
[22-May-2011 00:00:01] [disconnected at Sun May 22 00:00:01 2011]
[22-May-2011 00:00:02] [connected at Sun May 22 00:00:02 2011]
[22-May-2011 00:00:21] [zenoss-logger (logger bot) has joined #zenoss]
[22-May-2011 11:24:08] kocolosk|away is now known as kocolosk
[22-May-2011 12:00:01] [disconnected at Sun May 22 12:00:01 2011]
[22-May-2011 12:00:02] [connected at Sun May 22 12:00:02 2011]
[22-May-2011 12:00:20] [zenoss-logger (logger bot) has joined #zenoss]
[22-May-2011 12:00:20] <> [freenode-info] please register your nickname...don't forget to auto-identify!
[23-May-2011 00:00:01] [disconnected at Mon May 23 00:00:01 2011]
[23-May-2011 00:00:02] [connected at Mon May 23 00:00:02 2011]
[23-May-2011 00:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[23-May-2011 03:00:51] <silentg> hi all, I have just deleted and added the same monitoring template command with a response time guage and now the time rrd is not being updated..any ideas why this would be the case?
[23-May-2011 04:02:50] <fragfutter> silentg: did you modify the template? If true, you might need to recreate the rrds (just delete them)
[23-May-2011 04:04:29] <silentg> ok, would like to keep the old data in the rrd though
[23-May-2011 04:06:59] <fragfutter> so what did you do? switched from derive to counter or what?
[23-May-2011 04:08:14] <silentg> no just deleted the monitoring template command for a device thinking it was just for that device but it removes the template from all devices, so I just added it again with the same command
[23-May-2011 04:08:50] <fragfutter> then the rrd should still match.
[23-May-2011 04:08:57] <silentg> trying to debug using zencommand -v 10 --device hostname gives an error: "zen.zencommand: Received exit code: 2"
[23-May-2011 04:10:56] <fragfutter> zencommand run -d somehostname -v 2
[23-May-2011 04:11:02] <silentg> doesn't seem to indicate the reason why though, the check_ldap command its running completes successfully with the response time I want it to update the rrd with
[23-May-2011 04:21:28] <silentg> ok that runs a bit differently, I get an ok result with response time but it doesn't proceed to store the time, I have another similar command check_ldaps that is run fine and when done zen.RRDUtil is called and saves the result ok - strange because I deleted it too.
[23-May-2011 04:26:17] <silentg> I'll try deleting the rrd and restoring from a backup copy..
[23-May-2011 04:31:40] <hmp> i'm getting heartbeat failures for a remote collector that is removed, anybody know why?
[23-May-2011 04:32:14] <hmp> zenoss 3.1 on both sides, no zenpacks
[23-May-2011 04:35:33] <hmp> solved: (
[23-May-2011 05:12:43] <rockhopper_> hey guys. i'm trying to install zenoss. i;m following the manual. on "Tuning Zope" it is written to add zserver-threads and python-check-interval before <zodb_db main>
[23-May-2011 05:13:10] <rockhopper_> it's under some othrer tag?
[23-May-2011 05:13:12] <rockhopper_> other*
[23-May-2011 05:13:42] <rockhopper_> and another question, I can find zendmd
[23-May-2011 05:15:07] <rockhopper_> where can i*
[23-May-2011 05:28:33] <rockhopper_> I succeed finding it in the end.
[23-May-2011 08:45:58] kocolosk|away is now known as kocolosk
[23-May-2011 10:42:15] <kokey> blah
[23-May-2011 10:42:26] <Sam-I-Am> yep
[23-May-2011 10:42:52] <Simon4> freakin mondays
[23-May-2011 10:43:13] <Sam-I-Am> yep
[23-May-2011 10:48:10] <Hackman238> Hey alll
[23-May-2011 10:49:48] <Hackman238> Unrelated, but important
[23-May-2011 10:51:12] <Hackman238> rockhopper_: vi $ZENHOM/etc/zope.conf, do /zserver-threads {enter} and make the change there. Do the same for python-check-interval. Zendmd is in $ZENHOME/bin/zendmd
[23-May-2011 10:58:02] <Sam-I-Am> Hackman238: lol
[23-May-2011 10:59:26] * dhopp wants to rewind and start this week over
[23-May-2011 11:02:26] <Hackman238> Sam-I-Am: LOL to what?
[23-May-2011 11:03:16] <Hackman238> dhopp: Did you get your distributed collectors working?
[23-May-2011 11:03:42] <dhopp> On my test setup I'm using Egor's zenpack
[23-May-2011 11:05:12] <Hackman238> dhopp: Working? I've recently tried out his pack- gave me issues with more complex configurations
[23-May-2011 11:09:54] <Sam-I-Am> Hackman238: the link
[23-May-2011 11:19:56] <dhopp> Hackman238: it's working right now but I don't have a super complex setup in test
[23-May-2011 11:19:59] <dhopp> just 2 servers
[23-May-2011 11:21:18] <dhopp> Hackman238: I had a discussion on the thread about his zenpack about how to handle multiple zenhubs (I was going to modify the zenpack code slightly to start zenhub and configure the daemons appropriately) and people were saying they weren't sure if core could handle that
[23-May-2011 11:21:37] <dhopp> Hackman238: What problems were you having with it?
[23-May-2011 11:25:19] <Hackman238> dhopp: His pack wont do multiple hubs, but multiple hubs are possible by hand.
[23-May-2011 11:25:48] <Hackman238> dhopp: Mostly broken relationships on collector removal -> readdition.
[23-May-2011 11:26:09] <Hackman238> dhopp: Been testing an extensive automation project I'm working on using chef
[23-May-2011 11:27:05] <Hackman238> dhopp: Should make zenoss as easy to install as installing chef, the cookbook, filling in some blanks and pressign the easy button- multiple collectors, multi hubs and all
[23-May-2011 11:27:23] <Hackman238> Sam-I-Am: Oh gotcha
[23-May-2011 11:27:46] <Hackman238> Sam-I-Am: Step 7 always best LOL
[23-May-2011 11:29:51] <dhopp> Hackman238: Theoretically collectors shouldn't be removed :-)
[23-May-2011 11:30:15] <Simon4> um
[23-May-2011 11:30:15] <Hackman238> dhopp: Ideally yes, but this is for a fully automated cloud
[23-May-2011 11:30:22] <dhopp> ahh
[23-May-2011 11:31:35] <Hackman238> dhopp: Will let us refocus on more imporant dev by reducing administration whilst also allowing for significant automated scalability.
[23-May-2011 11:32:43] <kokey> ok i think my mysql db is hosed again
[23-May-2011 11:32:50] <kokey> i wonder if it'll self heal this time or not
[23-May-2011 11:33:36] <Hackman238> kokey: whats it doing?
[23-May-2011 11:33:54] <kokey> being very slow
[23-May-2011 11:34:17] <kokey> it's been fuller than it is now so i'm not sure what the cause is
[23-May-2011 11:34:31] <Hackman238> kokey: dump it to a file and take a peek
[23-May-2011 11:34:43] <dhopp> kokey: the mysql machine isn't swapping memory is it?
[23-May-2011 11:35:06] <kokey> it's all innodbed, have lots of RAM so everything can be in cache, using sloppy reads, etc.
[23-May-2011 11:35:15] <kokey> it's not using much CPU either
[23-May-2011 11:35:28] <Hackman238> kokey: have you killed and restarted the procesS?
[23-May-2011 11:36:11] <kokey> Hackman238: yeah, actually i rebooted while adding more CPU and RAM
[23-May-2011 11:37:10] <Hackman238> kokey: hum. do the triggers and stored procedures look ok?
[23-May-2011 11:38:57] <kokey> i'm going to switch query logging on
[23-May-2011 11:39:05] <Hackman238> kokey: ok
[23-May-2011 11:39:12] <kokey> Hackman238: hmmm, i think there is only one, that hasn't been modified, but it could be misbehaving
[23-May-2011 11:39:51] <Sam-I-Am> hmmmmmm
[23-May-2011 11:40:11] <Hackman238> kokey: /opt/zenoss/Products/ZenEvents/db/zenevents.sql has trigger code and /opt/zenoss/Products/ZenEvents/db/zenprocs.sql stored procedures if you need to fix them
[23-May-2011 11:40:33] <Sam-I-Am> so i have some alerting rules where i delay alerts.  if something self-clears in that window, no one hears about it
[23-May-2011 11:41:06] <Sam-I-Am> however, if someone's watching zenoss and they acknowledge it, the event hits my acknowledgement alert
[23-May-2011 11:41:07] <Hackman238> kokey: just vi them and look for the applicable block of sql and paste into mysql -u root -p -d events
[23-May-2011 11:41:17] <Sam-I-Am> so... people see that an event was ack'd, but they dont get the original event
[23-May-2011 11:41:41] <kokey> Hackman238: actually the only procedure it hits is the ageevents() one
[23-May-2011 11:41:44] <Sam-I-Am> i suspect adding a delay to the ack alerting rule won't fix this issue since acks don't have a 'correcting' state
[23-May-2011 11:42:09] <Sam-I-Am> or rather, acks dont clear
[23-May-2011 11:43:08] <Hackman238> Sam-I-Am: Hum. I'd have the ack test to see if the alert was sent to begin with, if not, dont alert
[23-May-2011 11:43:35] <Sam-I-Am> how'd that work?
[23-May-2011 11:43:44] <Sam-I-Am> the alert was sent to some groups, but not others
[23-May-2011 11:44:04] <Sam-I-Am> the operations folks see everything, but it takes a while for stuff to get up to the engineering level
[23-May-2011 11:44:18] <Hackman238> Sam-I-Am: Oh.
[23-May-2011 11:44:31] <Hackman238> Sam-I-Am: Hum.
[23-May-2011 11:44:37] <Hackman238> Sam-I-Am: Im not really sure
[23-May-2011 11:44:58] <Sam-I-Am> yeah its weird
[23-May-2011 11:45:03] <kokey> the origins of this query baffles me a bit... SELECT h.device,h.component,h.message,h.firstTime,h.summary,h.severity,h.summary,h.ownerid,h.stateChange, h.evid FROM history h, alert_state a  WHERE h.evid=a.evid AND a.userid='FILER' AND a.rule='FILER_Email';
[23-May-2011 11:45:10] <Sam-I-Am> might just make all the rules consistent on times to reduce confusion
[23-May-2011 11:45:18] <kokey> what would query history and use alert rules?
[23-May-2011 11:46:05] <Hackman238> kokey: Good question. Im not sure.
[23-May-2011 11:46:28] <Sam-I-Am> magic
[23-May-2011 11:46:38] <Simon4> kokey: get your grep on
[23-May-2011 11:46:51] <Sam-I-Am> thats what she said
[23-May-2011 11:46:53] <Hackman238> Sam-I-Am: grep * -R
[23-May-2011 11:47:41] <Sam-I-Am> hmm, another interesting thing... on some of my collectors, the performance graphs for the collector itself show numeric values, but no points on the graph
[23-May-2011 11:48:32] <Sam-I-Am> 2 collectors are fine, 2 arent
[23-May-2011 11:51:06] <kokey> ok so it seems like mysqld runs fine until me or someone clicks on History
[23-May-2011 11:51:20] <kokey> history only have like 700k entries
[23-May-2011 11:54:34] <kokey> i'm a bit worried
[23-May-2011 11:54:38] <kokey> the last time i had this problem
[23-May-2011 11:54:46] <kokey> i dumped the mysql db, and reimported it
[23-May-2011 11:54:49] <kokey> and all was fine
[23-May-2011 11:54:53] <Sam-I-Am> lol
[23-May-2011 11:54:55] <Sam-I-Am> btdt :/
[23-May-2011 11:55:08] <kokey> i am supposed to finish handover documentation, not fight with this
[23-May-2011 11:55:46] <Sam-I-Am> how long do you keep history for?
[23-May-2011 11:55:56] <Sam-I-Am> i wound up doing some interesting db tweaks
[23-May-2011 11:55:59] <Sam-I-Am> some helped, some did not
[23-May-2011 11:58:18] <kokey> 14 days
[23-May-2011 11:58:27] <kokey> well i switched mine to innodb
[23-May-2011 11:58:44] <kokey> and i've done transaction isolation level read uncommitted
[23-May-2011 11:58:49] <Sam-I-Am> mine is 100 days
[23-May-2011 11:58:58] <kokey> things were working really fast until now
[23-May-2011 11:59:11] <Sam-I-Am> whats the slow query log look like?
[23-May-2011 11:59:17] <kokey> > select count(*) from history;
[23-May-2011 11:59:17] <kokey> +----------+
[23-May-2011 11:59:17] <kokey> | count(*) |
[23-May-2011 11:59:22] <kokey> 1 row in set (4 min 5.93 sec)
[23-May-2011 11:59:25] <Sam-I-Am> ugh, that one
[23-May-2011 11:59:42] <kokey> last time i had this happen was when the partition ran out of space
[23-May-2011 11:59:44] <Sam-I-Am> i told the operators not to do history searches with dates way in the past... that helped a bit.
[23-May-2011 11:59:50] <kokey> it never recovered, until i reimported
[23-May-2011 11:59:53] <kokey> but that didn't happen this time
[23-May-2011 12:00:01] [disconnected at Mon May 23 12:00:01 2011]
[23-May-2011 12:00:02] [connected at Mon May 23 12:00:02 2011]
[23-May-2011 12:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[23-May-2011 12:00:19] <kokey> Sam-I-Am: yeah
[23-May-2011 12:00:19] <Sam-I-Am> aka, via direct sql
[23-May-2011 12:00:27] <Sam-I-Am> how about when zenoss is not running?
[23-May-2011 12:00:28] <Hackman238> kokey: strange
[23-May-2011 12:00:35] <Sam-I-Am> in case theres some stuck processes
[23-May-2011 12:00:50] <Sam-I-Am> what happened here was someone wouldnt get results in 5 seconds, then hit load history again, repeat, repeat
[23-May-2011 12:00:51] <kokey> i rebooted and restarted a number of times
[23-May-2011 12:01:00] <Sam-I-Am> eventually some threads were used up and broke zenoss
[23-May-2011 12:01:15] <kokey> i've got slow query logging on so i thought i would see those situations
[23-May-2011 12:01:35] <kokey> ok now zenoss is hosed again
[23-May-2011 12:01:52] <kokey> one thing i do see...
[23-May-2011 12:01:53] <kokey> type=1400 audit(1306166342.130:79): avc:  denied  { write } for  pid=16974 comm="mysqld" path="/opt/lib/mysql/ib_logfile1" dev=dm-4 ino=6651919 scontext=user_u:system_r:mysqld_t:s0 tcontext=user_u:object_r:usr_t:s0 tclass=file
[23-May-2011 12:02:11] <kokey> i'm told i shouldn't worry about it since it's set to permissive or something
[23-May-2011 12:02:25] <Sam-I-Am> make sure it actually thinks its permissive
[23-May-2011 12:03:15] <kokey> hmmm someone else might be trying to view history
[23-May-2011 12:03:22] <kokey> but all that said
[23-May-2011 12:03:28] <kokey> the DB fits comfortably into RAM#
[23-May-2011 12:03:36] <kokey> and the mysql process takes like 10% CPU
[23-May-2011 12:04:41] <Sam-I-Am> how many mysql threads are running?
[23-May-2011 12:06:50] <kokey> hmmm
[23-May-2011 12:06:58] <kokey> i wonder if my innodb files are big enough
[23-May-2011 12:08:06] <kokey> actually naw, my second one autoextends
[23-May-2011 12:08:15] <Sam-I-Am> did you use the zenoss suggested tweaks for mysql?
[23-May-2011 12:08:34] <kokey> Sam-I-Am: yup, as is
[23-May-2011 12:08:57] <kokey> something is weird tho
[23-May-2011 12:09:05] <kokey> because it was running pretty fast until this afternoon
[23-May-2011 12:09:52] <kokey> i'm going to leave it like this for a while
[23-May-2011 12:10:06] <kokey> as long as no one tries to do history
[23-May-2011 12:10:27] <Sam-I-Am> some of those tweaks apparently dont work well in SAN environments
[23-May-2011 12:10:54] <kokey> well i have way more RAM available than the DB
[23-May-2011 12:11:05] <kokey> so i would assume the system's disk cache would handle a fair chunk of it for me
[23-May-2011 12:11:36] <Sam-I-Am> i'm in the same boat...
[23-May-2011 12:11:39] <Sam-I-Am> #innodb_flush_method = O_DIRECT
[23-May-2011 12:11:46] <Sam-I-Am> that thing caused me a headache
[23-May-2011 12:11:54] <kokey> ah, i got that set
[23-May-2011 12:12:04] <kokey> i'll comment it out, i feel like restart mysqld again for fun
[23-May-2011 12:13:01] <Sam-I-Am> i've got some other stuff in there... some of it based on my environment
[23-May-2011 12:14:26] <Sam-I-Am>
[23-May-2011 12:16:49] <kokey> Sam-I-Am: so you didn't actually like the buffer pool size stuff?
[23-May-2011 12:16:58] <kokey> the innodb ones that is
[23-May-2011 12:17:08] <kokey> oh wait i see you modified them
[23-May-2011 12:24:13] <kokey> ok, so without zenoss running, doing a count on history takes about 30 seconds at most
[23-May-2011 12:24:22] <kokey> while after it's running, it's still fine too
[23-May-2011 12:24:28] <kokey> until you try select history from the interface
[23-May-2011 12:25:25] <kokey> the query from the interface is a lot more complex than from the command line
[23-May-2011 12:25:32] <kokey> i mean, than a count
[23-May-2011 12:25:43] <kokey> i suspect things will get better once some stuff has been aged away
[23-May-2011 12:27:07] <Sam-I-Am> yep
[23-May-2011 12:27:18] <Hackman238> kokey: problem was innodb_flush_method = O_DIRECT?
[23-May-2011 12:33:30] <kokey> nope that wasn't the problem
[23-May-2011 12:33:32] <kokey> it's still there...
[23-May-2011 12:33:33] <kokey> # Query_time: 541  Lock_time: 0  Rows_sent: 100  Rows_examined: 67671
[23-May-2011 12:33:33] <kokey> select  SQL_CALC_FOUND_ROWS eventState,severity,device,component,eventClass,summary,firstTime,lastTime,count,eventState,severity,evid from history where eventClass like '/%' and eventState <= 2 and lastTime >= 1306081292  and (severity=5 or severity=4 or severity=3 or severity=2 or severity=0)  and (eventState=0 or eventState=1)  order by lasttime DESC limit 0, 100;
[23-May-2011 12:36:08] <Hackman238> kokey: yikes!
[23-May-2011 12:36:27] <Hackman238> kokey: like 0.01 seconds on one of my largest DC's
[23-May-2011 12:38:40] <kokey> Hackman238: that was very quick on mine until, uhm, today
[23-May-2011 12:39:19] <Hackman238> kokey: Was happening before the addition/swap of RAM/CPUs?
[23-May-2011 12:42:26] <kokey> Hackman238: yup, it was
[23-May-2011 12:43:05] <dhopp> Hackman238: Do you have a complete zenoss install per DC?  Did you do some custom interface to tie them all together?
[23-May-2011 12:46:44] <Hackman238> dhopp: One instance per DC, each with dedicated zope server, deciated  mysql server and several collectors. We use global event console on another box to tie events together- this box hosts its own mysql, which recieves events from all DC's.
[23-May-2011 12:48:14] <dhopp> Ahh…not using core I'm assuming :-)
[23-May-2011 12:49:39] <Hackman238> dhopp: Negative, Ent here at RackSpace- though its very custom. Some of my other large clients using Core have similar rollouts without the global event console and they make use of multiple hubs and distributed collectors configured manually.
[23-May-2011 12:49:59] <dhopp> gotcha
[23-May-2011 12:50:06] <Hackman238> dhopp: There is no decernable performance differnce between core and ent.
[23-May-2011 12:50:26] <Hackman238> dhopp: Correction, Ent is core with a handful of extra code and extra zenpacks.
[23-May-2011 12:50:33] <dhopp> right….but doing all the remote collector stuff is easier in ent
[23-May-2011 12:50:44] <Hackman238> dhopp: Meh, not really.
[23-May-2011 12:51:01] <Hackman238> dhopp: All the items which cause headaches happen either way.
[23-May-2011 12:51:14] <dhopp> and those items are?
[23-May-2011 12:53:07] <Hackman238> dhopp: Complications at upgrade, heartbeat issues, poor zenrender performance, device load balancing between collectors on an automatic basis whilst saving history, managability when installing packs which modify classes or modelers...theres more
[23-May-2011 12:54:10] <Hackman238> dhopp: For distributed collectors the advantages outwheigh the disadvantages, though there are many problems with having distributed collectors
[23-May-2011 12:55:04] <Hackman238> dhopp: Same goes for Zenoss in general- Zenoss comes with unique headaches, but theres no other product that offers the same advantages.
[23-May-2011 12:55:29] <dhopp> Hackman238: I would imagine upgrades would be difficult (which to do first, take the whole system down, etc.).  the lb between collectors and saving history is going to be big for me,
[23-May-2011 12:55:51] <dhopp> Hackman238: what problems do you have when installing packs that modify/add modelers?
[23-May-2011 12:56:13] <Hackman238> dhopp: When I say all of this I mean it with one consideration: zeno downtime.
[23-May-2011 12:56:17] <Hackman238> *zero
[23-May-2011 12:56:51] <Hackman238> dhopp: When adding packs or code that modifies classes or modelers the change much be replicated to colelctor installations and the related daemons restarted.
[23-May-2011 12:57:24] <Hackman238> dhopp: to maintain zero downtime one must move devices from collector to collector, restarting daemons and swapping the load back once complete.
[23-May-2011 12:57:41] <dhopp> Hackman238: I'm with you on that one…when our existing monitoring has issues and we lose visibility to what's happening you would think the world was coming to an end..heh
[23-May-2011 12:58:05] <Hackman238> dhopp: Really the complication lies with the fact one must restart non-concurrent daemons at all.
[23-May-2011 12:59:22] <Hackman238> dhopp: A good example of the contrary is Zope- we can run a donzen instances of zope off the same zeo. If we make changes that require a zope reload, we can simply reload the zopes one after another and users fail over to the next available instance. Eventually everyone is bummped to a session that has been restarted whilst having no knowledge of downtime.
[23-May-2011 13:00:05] <Hackman238> dhopp: You cant do that with the daemons unless you move the load to differnt collectors, whether they be duplicate instances on the same box or an instance on another box.
[23-May-2011 13:01:11] <Hackman238> dhopp: hence I'm going to cloudify all of it.
[23-May-2011 13:01:15] <Hackman238> dhopp: LOL
[23-May-2011 13:01:34] <dhopp> Hackman238: But during a daemon restart you are talking about approx ~20-30 seconds that that particular daemon is "blind" ?
[23-May-2011 13:01:57] <dhopp> Hackman238: or does the daemon restart get much worse when it's a remote collector?
[23-May-2011 13:02:03] <Hackman238> dhopp: Its much longer than that with large instances. Could be 5 to 7 minutes
[23-May-2011 13:02:15] <dhopp> oh
[23-May-2011 13:03:28] <Hackman238> dhopp: Its the same when its a remote, however, the change of error increases since you're dealing with X number more boxes and the remotes can often take very long to collect configs at very very large scale instances- hence the only situation you'd need distributed collectors.
[23-May-2011 13:04:07] <Hackman238> *chance
[23-May-2011 13:04:30] <dhopp> Hackman238: it seems to me that pretty much any solution that offers distributed monitoring would have similar issues…it's not like distributed monitoring is simple
[23-May-2011 13:04:45] <Hackman238> dhopp: Thats correct.
[23-May-2011 13:05:10] <Hackman238> dhopp: Thats why we're going to cloud the remote colelctors.
[23-May-2011 13:05:13] <Hackman238> *colelctors
[23-May-2011 13:05:20] <dhopp> lol
[23-May-2011 13:05:27] <dhopp> third time's a charm?
[23-May-2011 13:06:19] <Hackman238> dhopp: We'll be able to bring them up with new copies of code/updates, threaded rsync the rrd's and tranfer the load. Once done we deconstruct the old vm and the process restarts
[23-May-2011 13:06:33] <Hackman238> dhopp: Bah. Sorry, busy busy today. :-)
[23-May-2011 13:07:45] <Hackman238> dhopp: one might note that vm performance sucks- far too much for zenoss. This is true, but with the whoel thing automated we can have tens of dozens of collectors instead of just a few.
[23-May-2011 13:07:58] <Hackman238> *whole
[23-May-2011 13:10:38] <dhopp> I'm still in the process of trying to get the hardware for the infrastructure….We have 2 DCs (well really 3 but the third does not have a whole lot of important stuff in it).  The 2 DCs are connected via fiber…Still trying to decide what the best setup is
[23-May-2011 13:14:06] <Hackman238> Plan well.
[23-May-2011 13:19:59] <dhopp> Hackman238: I think I'm going to be able to get 2 16 Core servers with 16-24 GB of RAM and probably 10-12 disks.  As well as 2 8 Core servers with 12 GB of RAM and 8 disks.  The 16 Core machines I was thinking of having them be the web interface as well as MySQL (with probably a 10 Disk RAID 10 and mysql replication between the 2).  The 8 Core would be remote collectors with a 6 Disk RAID 10.
[23-May-2011 13:20:34] <dhopp> In the end will be 700-1000 devices
[23-May-2011 13:20:37] <dhopp> monitoried
[23-May-2011 13:20:40] <dhopp> monitored
[23-May-2011 13:21:58] <jmp242> Hey, odd issue with graphs. I'm watching memory use of a VM instance as a standard device in Zenoss. This memory graph was fine, until I increased memory on the VM from 512MB to 2GB. It doesn't seem to ever update the graph, even with a manual model of the device and a push changes. The data in the overview page is correct, but the graph is still showing like it hasn't updated the memory. Do I need to just kill the rrd file?
[23-May-2011 13:24:03] <Hackman238> dhopp: seems like a good start
[23-May-2011 13:24:18] <jmp242>
[23-May-2011 13:24:22] <jmp242> is what I'm getting
[23-May-2011 13:25:08] <Hackman238> jmp242: whoa. I'd try renaming the rrd and see if the new one works right.
[23-May-2011 13:27:09] <jmp242> should I stop zenoss first? or can I do it live do you think?
[23-May-2011 13:27:19] <Hackman238> jmp242: I'd try it live.
[23-May-2011 13:27:23] <Sam-I-Am> .
[23-May-2011 13:27:33] <Hackman238> jmp242: the new rrd should be created automatically
[23-May-2011 13:29:15] <Sam-I-Am> was there some fixed scale for it?
[23-May-2011 13:29:18] <Sam-I-Am> rather than auto
[23-May-2011 13:29:41] <jmp242> well . . . not that I know of
[23-May-2011 13:30:05] <jmp242> This graph is a bit of an oddity
[23-May-2011 13:30:25] <jmp242> it's from the old Advanced Linux zenpack, from around v 2.0 era
[23-May-2011 13:30:54] <jmp242> at some point it was removed from the download site (and I lost the installer), but it's been working more or less for us and the upgrades all worked so far
[23-May-2011 13:33:19] <jmp242> Interesting
[23-May-2011 13:33:32] <jmp242> there are a bunch of rrds that are individual datapoints that are all graphed in one graph there
[23-May-2011 13:33:38] <jmp242> I assume I ought to rename all of them
[23-May-2011 13:33:44] <Hackman238> jmp242: Yes please
[23-May-2011 13:36:07] <jmp242> well, that did clear up the graph
[23-May-2011 13:36:13] <jmp242> now to wait and see if the data looks right
[23-May-2011 13:36:52] <Hackman238> jmp242: Let us know! :-)
[23-May-2011 13:36:58] <jmp242> of course
[23-May-2011 13:40:08] <dhopp> Hackman238: what are you using for historical trend analysis since Zenoss doesn't give a good way to do date range searching?
[23-May-2011 13:40:59] <CrummyGummy> Hi all, I've installed from the ubuntu stack and from src an I keep getting problems with zenhub logging into mysql. Is there a cache somewhere I need to clear to stop this from happening?
[23-May-2011 13:41:49] <dhopp> GrummyGummy: what's the error message specifically?
[23-May-2011 13:41:50] <Hackman238> dhopp: what items are you lookig to trend?
[23-May-2011 13:42:06] <CrummyGummy> MySQLConnectionError: (1045, "Access denied for user 'zenoss'@'localhost' (using password: YES)")
[23-May-2011 13:42:28] <dhopp> Hackman238: CPU/Memory/Various JMX stats out of Tomcat/SQL Server stats
[23-May-2011 13:43:02] <dhopp> CrummyGummy: can you login to mysql through the command line: mysql -u zenoss -p -h localhost ?
[23-May-2011 13:43:12] <dhopp> I think the default password is 'zenoss' (without quotes)
[23-May-2011 13:44:35] <CrummyGummy> No, not using port 3307 which is where the stack has installed the db.
[23-May-2011 13:45:53] <Hackman238> dhopp: Most of our trending is done in zenoss using linear expressions and highly complex report graph definitions.
[23-May-2011 13:46:47] <CrummyGummy> dhopp: I can though using my stock mysql install that I used for the src install with the password provided.
[23-May-2011 13:47:07] <CrummyGummy> Any ideas where this username and password etc are stored so I can verify this?
[23-May-2011 13:47:26] <Hackman238> CrummyGummy: mysql -u root -p -P 3307 -d events
[23-May-2011 13:47:35] <Hackman238> CrummyGummy: Then run 'mysql.user SET Password=PASSWORD('zenoss');'
[23-May-2011 13:48:42] <Hackman238> CrummyGummy: Correction, 'mysql.user SET Password=PASSWORD('zenoss') WHERE User='zenoss';'
[23-May-2011 13:48:51] <dhopp> Hackman238: But that doesn't give you away of saying "give me cpu and tomcat db connections for May 2 at 6AM through May 5 at 6 AM" or something similar?
[23-May-2011 13:50:59] <CrummyGummy> Hackman238: Any idea what that root password is?
[23-May-2011 13:51:24] <dhopp> CrummyGummy: it's either blank or it's whatever you set it to be during install
[23-May-2011 13:51:50] <Hackman238> dhopp: Sure it does- we build graph definitions to compare metrics for whatever is requested
[23-May-2011 13:51:56] <CrummyGummy> There was no option in the stack install
[23-May-2011 13:52:04] <CrummyGummy> and blank doesn't work.
[23-May-2011 13:52:08] <Hackman238> CrummyGummy: Try blank or zenoss
[23-May-2011 13:54:38] <Hackman238> dhopp: For trending based on events, we've a custom app for that.
[23-May-2011 13:56:07] <dhopp> Hackman238: so I assume you pull the data from RRDs into some db?
[23-May-2011 13:57:14] <Hackman238> dhopp: Not at all, we almost always write trending reports which use RRD data to run in zenoss- though they mostly use custom reporting code
[23-May-2011 13:57:38] <Hackman238> dhopp: I think the only rrd manipulation we do outside of zenoss is spike skimming.
[23-May-2011 13:57:55] <dhopp> Hackman238: spike skimming?
[23-May-2011 13:58:12] <Hackman238> dhopp: E.x. if an vlan blows up and reports 1 PB of traffic, we need to skim off that invalid value.
[23-May-2011 13:58:26] <dhopp> gotcha
[23-May-2011 13:58:52] <Hackman238> dhopp: Not that I wouldnt love to bill for that 1 PB of traffic, but the customers would be very angry :-P
[23-May-2011 13:58:54] <CrummyGummy> Ah, it works ok if you use the correct client binary.
[23-May-2011 13:59:04] <Hackman238> CrummyGummy: Really?
[23-May-2011 13:59:20] <CrummyGummy> the default one ignores the port and just hits my default mysql installation
[23-May-2011 13:59:31] <CrummyGummy> wierd
[23-May-2011 13:59:53] <Hackman238> CrummyGummy: Indeed
[23-May-2011 14:00:32] <dhopp> Hackman238: Although a picture of their face when they opened the bill would be priceless :-P
[23-May-2011 14:01:26] <Hackman238> dhopp: Priceless indeed
[23-May-2011 14:04:11] <CrummyGummy> Oh well. I think the only way to get this to play nice is to install it on its own vm.  Tomorrows problem...
[23-May-2011 14:05:13] <Hackman238> CrummyGummy: Hum.
[23-May-2011 14:09:03] <dhopp> Ok..this is weird.  Did a test graph report and included just one graph (CPU Utilization from one of the servers).  If I choose "yearly" the date range on the graph shows "1991-09-05 13:07:17 to 2011-05-23 13:07:17"
[23-May-2011 14:09:12] <dhopp> I'm willing to bet I don't have data from 1991 :-P
[23-May-2011 14:09:27] <Hackman238> dhopp: Almost certainly not LOL
[23-May-2011 14:09:39] <Simon4> that's some impressive extrapolation
[23-May-2011 14:10:22] <dhopp> Monthly shows from 1998-04-01
[23-May-2011 14:10:23] <dhopp> wtf
[23-May-2011 14:31:19] <jmp242> well
[23-May-2011 14:31:27] <jmp242> strangely enough it did re-create the graph
[23-May-2011 14:31:41] <jmp242> but still wrong data . . . I guess it might be the zenpack is broken for this somehow
[23-May-2011 14:32:01] <Hackman238> jmp242: can you post the graph? Does indeed sound like broken zenpack
[23-May-2011 14:32:58] <jmp242>
[23-May-2011 14:33:24] <jmp242> What's odd is it seems to just not be picking up the new max memory. And seems to work for other devices lol
[23-May-2011 14:34:08] <jmp242> Oh, well, there were supposed known problems with this zenpack. I just am not entirely sure how I would make the graph better - I could re-define it I guess, but I'd end up at the same place I would think
[23-May-2011 14:34:31] <Hackman238> Whats the pack basing the total memory off of?
[23-May-2011 14:36:16] <jmp242> I have to admit, it's not obvious just glancing at it
[23-May-2011 14:36:36] <Hackman238> jmp242: check the hw page of the device- does it have the correct memory amount?
[23-May-2011 14:36:55] <Hackman238> Can you zip up the pack and email it to me?
[23-May-2011 14:42:09] <jmp242> Well, I zipped the folder
[23-May-2011 14:42:11] <jmp242> here it comes
[23-May-2011 14:43:04] <jmp242> Note I tried to hack this very simply to an .egg and I think I failed - but again it kept working in place so I've mostly left it alone lol
[23-May-2011 14:46:54] <Hackman238> jmp242: Far as I can tell its just adding values it's getting from snmp.
[23-May-2011 14:47:20] <Hackman238> jmp242: What amount of RAM did you have to start and how much after the upgrade?
[23-May-2011 14:49:30] <jmp242> we had 512MB to start
[23-May-2011 14:49:36] <jmp242> and 2048MB to finish
[23-May-2011 14:49:55] <Hackman238> jmp242: Hum.
[23-May-2011 14:49:57] <jmp242> Though we might have done this live
[23-May-2011 14:50:06] <jmp242> without rebooting the server. I'll check that next
[23-May-2011 14:50:11] <Hackman238> Oh, is it a VM?
[23-May-2011 14:50:14] <jmp242> yes
[23-May-2011 14:50:16] <Hackman238> restart snmpd
[23-May-2011 14:50:33] <Hackman238> LOL lets start simple in such case
[23-May-2011 14:51:52] <jmp242> Done, now to see if it's fixed
[23-May-2011 14:52:02] <Hackman238> Alrighty
[23-May-2011 14:52:05] <jmp242> I should have thought of that, but the overview data was right in Zenoss
[23-May-2011 14:53:44] <Hackman238> It happens- the other day I couldnt get my garage door to open; turns out I grabbed the wrong garage remote (have two garages on differnt areas of the property.) LOL
[23-May-2011 14:56:36] <Sam-I-Am> big property heh
[23-May-2011 14:57:00] <Hackman238> Sam-I-Am: Its Texas LOL
[23-May-2011 14:57:30] <Sam-I-Am> everything is bigger in texas
[23-May-2011 14:58:54] <Hackman238> Sam-I-Am: LOL- insanely huge trucks and bad investment sports cars are very in around here- therefore I suspect there is something which is not bigger LOL
[23-May-2011 14:59:38] <Sam-I-Am> lots of money floating around?
[23-May-2011 14:59:46] <Sam-I-Am> if i moved there i would need an F350 dually lol
[23-May-2011 14:59:51] <Sam-I-Am> with horns on the front
[23-May-2011 14:59:55] <Hackman238> Sam-I-Am: No, but the local debt average is very very high.
[23-May-2011 15:00:02] <jmp242> Do you think I need to reset the rrd files again Hackman238?
[23-May-2011 15:00:24] <Hackman238> jmp242: You shouldnt have to, but it cant hurt
[23-May-2011 15:02:06] <dhopp> If Zenoss can't ping a device (because of a firewall restriction) but it can get to port 161 (SNMP) will it still gather data or not bother since it can't ping it?
[23-May-2011 15:02:10] <Hackman238> Sam-I-Am: My first comment about the trucks, etc was sorta of reffering to the most recent South Park episode about gizmo size vs security, and sorta based on how common 'excessive' transportation is here.
[23-May-2011 15:02:21] <dhopp> I'm going to have the firewall restriction opened…but just wondering how it behaves in this case
[23-May-2011 15:02:24] <jmp242> dhopp: generally no, but you can disable ping monitoring
[23-May-2011 15:02:33] <jmp242> well, no that may not be true
[23-May-2011 15:02:45] <jmp242> it probably tries both SNMP and ping as long as it's in production
[23-May-2011 15:02:49] <Hackman238> dhopp: What ping problem are you having?
[23-May-2011 15:03:00] <dhopp> Firewall doesn't allow ICMP :-)
[23-May-2011 15:03:13] <Hackman238> dhopp: oh man.
[23-May-2011 15:03:27] <dhopp> Going to have networking open it up..but just wondering what zenoss would do
[23-May-2011 15:04:08] <dhopp> It polled via SNMP and got all of the relevant info (processes, interfaces, routes, etc.)..but just didn't know if it wouldn't bother collecting perf data if it thinks it can't ping it
[23-May-2011 15:04:09] <Hackman238> dhopp: I'd say, if you cant get them to change the policy, disable ping on that device and add a custom test to check the system uptime every 1 minute and create a device down critical event if it fails
[23-May-2011 15:04:25] <Hackman238> dhopp: That way you have 1 minute resolution on up/down testing.
[23-May-2011 15:04:35] <dhopp> They will change the policy..not worried about that...
[23-May-2011 15:05:08] <dhopp> but it came up and was just wondering what would happen in the event that something like that did happen
[23-May-2011 15:07:52] <Hackman238> dhopp: Gotcha. I'd do a 1 minute test on a service on that device that qualifies if the device is up or not. E.x. If its DNS, hit DNS every minute. If DNS is down, the device is as good as down anyway.
[23-May-2011 15:08:21] <Hackman238> dhopp: For switches, etc, do a 1 minute smnp test on system uptime.
[23-May-2011 15:11:04] <dhopp> Hackman238: if said device is a webserver that hosts 6 tomcat instances and zenoss is unable to ping the device I should only get "Device down" event and not "Device Down" and "Tomcat 01 Down" and "Tomcat 02 Down" etc.?
[23-May-2011 15:13:59] <Hackman238> dhopp: Are they 6 VM's or jsut 6 tomcat instances?
[23-May-2011 15:14:09] <dhopp> Hackman238: tomcat instances
[23-May-2011 15:14:58] <Hackman238> dhopp: You'll want to do 6 tests and have a transform on the recieving event class to qualify if an existing failure is at large, and if so, increment the count, but done send another alert
[23-May-2011 15:15:44] kocolosk is now known as kocolosk|away
[23-May-2011 15:17:16] <Hackman238> jmp242: How'd it turn out?
[23-May-2011 15:17:30] <Hackman238> All, I'll BRB. Need  coffee :-)
[23-May-2011 15:18:33] kocolosk|away is now known as kocolosk
[23-May-2011 15:25:10] <jmp242> LOL
[23-May-2011 15:25:18] <jmp242> damn, such a waste of time too
[23-May-2011 15:25:42] <Hackman238> jmp242: Looks like 2GB to me :-)
[23-May-2011 15:26:08] <jmp242> Oh well. Do I bother to restore the old rrds for historical data? nah
[23-May-2011 15:26:31] <Sam-I-Am> i've been known to munge RRDs together
[23-May-2011 15:26:46] <Hackman238> jmp242: what Sam-I-Am said.
[23-May-2011 15:29:03] <Sam-I-Am> it aint pretty, but someone's gotta do it
[23-May-2011 15:29:46] <Hackman238> Is very scriptable
[23-May-2011 15:32:21] kocolosk is now known as kocolosk|away
[23-May-2011 15:33:34] kocolosk|away is now known as kocolosk
[23-May-2011 15:34:27] kocolosk is now known as kocolosk|away
[23-May-2011 15:35:07] kocolosk|away is now known as kocolosk
[23-May-2011 15:54:20] <jmp242> Is there a HOWTO or FAQ on the munging anywhere for my own edification
[23-May-2011 15:57:53] <Hackman238> jmp242: Unfortunately I have none to provide.
[23-May-2011 15:58:15] <Hackman238> jmp242: And my script isnt really 'readable' by anyone but me
[23-May-2011 15:59:18] <Sam-I-Am> yay encryption
[23-May-2011 16:00:43] <jmp242> Ahh, ok
[23-May-2011 16:00:52] <jmp242> thanks for the help Hackman238
[23-May-2011 16:01:23] <jmp242> I really thought they would have rebooted the VM by now and of course didn't think SNMPd would get messed up that way. live an learn
[23-May-2011 16:02:50] <dhopp> If were to say I'm going to keep 2 years worth of event data, would you think I'm crazy?
[23-May-2011 16:03:00] <Sam-I-Am> yes
[23-May-2011 16:03:26] <Sam-I-Am> either you only hve a handful of events, or you have a supercomputer housing your database
[23-May-2011 16:04:01] <dhopp> That's what I thought…I might have to come up with a script to keep 2 years worth of "important" events…(i.e. only keep critical events)
[23-May-2011 16:04:01] <Sam-I-Am> lots of tweaking also helps
[23-May-2011 16:04:13] <Sam-I-Am> i keep 100 days and even that bogs things down
[23-May-2011 16:05:45] <dhopp> Is removing events as easy as sql statement "delete from events where end_date < date" ?  Or is it more advanced then that?
[23-May-2011 16:06:01] <dhopp> i.e. do I need to do something with the zeodb
[23-May-2011 16:06:17] <Sam-I-Am> i think the two are separate... at least for now
[23-May-2011 16:06:38] <Hackman238> We keep 2 years here
[23-May-2011 16:06:40] <Sam-I-Am> other thing you could do is copy older events to another database which doesnt get access very often... or at least not by zenoss
[23-May-2011 16:06:52] <Hackman238> But we drop most of the junk events.
[23-May-2011 16:07:28] <Hackman238> dhopp: whats up with your zeodb?
[23-May-2011 16:07:43] <dhopp> Hackman238: what do you mean?
[23-May-2011 16:09:29] <Hackman238> dhopp: You asked if you need to do something to zeodb
[23-May-2011 16:10:29] <dhopp> Hackman238: I meant if I had a script that just did a "delete from events" would I need to make an update to zeodb or anything
[23-May-2011 16:10:47] <Hackman238> dhopp: Oh. not at all, thats all mysql
[23-May-2011 16:10:53] <Hackman238> dhopp: For now...
[23-May-2011 16:12:02] <dhopp> Hackman238: We want to be able to paint a "full" picture of a client's farm (i.e. performance and events) that occurred for up to 2 years.  Which is why I asked about events.  I think the only "info" event I would need to keep are device added or device removed.  And I think the client really is only going to care about critical events, but I have to convince management of that
[23-May-2011 16:12:39] <Hackman238> dhopp: What I would suggest is to simply drop these extra events before they are addded to the db.
[23-May-2011 16:12:59] <Hackman238> dhopp: E.x. we drop all change/set info events because there are tens of thousands a day.
[23-May-2011 16:13:24] <dhopp> is there an easy way to do that?
[23-May-2011 16:15:21] <Hackman238> dhopp: Yes, on the event class which recieved the event you wish to drop, set drop on the action.
[23-May-2011 16:15:55] <Sam-I-Am> i do a lot of dropping to keep things cleanish
[23-May-2011 16:16:00] <Sam-I-Am> rocket: !!!
[23-May-2011 16:16:34] <Hackman238> dhopp: For example, go to /zport/dmd/Events/Change/Set/eventClassStatus
[23-May-2011 16:16:43] <Hackman238> dhopp: click confi props.
[23-May-2011 16:16:56] <Hackman238> dhopp: set zaction to drop
[23-May-2011 16:17:12] <Hackman238> dhopp: All events which are and will be assigned to this class will be dropped
[23-May-2011 16:18:08] <rocket> Sam
[23-May-2011 16:18:22] <Sam-I-Am> long time no speak
[23-May-2011 16:18:31] <rocket> very swamped ..
[23-May-2011 16:19:40] <dhopp> Hackman238: holy crap…you aint lieing about the number of events.  In this test system I have like 3 devices right now and 807 events in that class in just a couple of days
[23-May-2011 16:20:49] <Sam-I-Am> rocket: i can see that
[23-May-2011 16:20:57] <Sam-I-Am> dhopp: lots of stuff gets tossed in there
[23-May-2011 16:20:58] <Hackman238> dhopp: Yep. Try to drop junk syslog messages too if you cant filter them before they get to zenoss
[23-May-2011 16:21:32] <rocket> Sam-I-Am: things are starting to calm down a bit though ..
[23-May-2011 16:21:45] <dhopp> Hackman238: Not doing syslog yet
[23-May-2011 16:21:53] <rocket> between work and my personal life I dont know how I can keep anything straight.
[23-May-2011 16:22:11] <Sam-I-Am> whats this personal life you speak of
[23-May-2011 16:22:27] <rocket> but I have a nice 4 day weekend coming this weekend so that will help .. camping away from computers and technology for a bit ..
[23-May-2011 16:22:36] <rocket> 7 month old ..
[23-May-2011 16:23:09] <Sam-I-Am> i think i'm doing a week away from technology in a few weeks
[23-May-2011 16:23:13] <Sam-I-Am> no internets
[23-May-2011 16:23:25] <Sam-I-Am> although... that might change depending on how job things go
[23-May-2011 16:23:26] <Hackman238> rocket: How does one go camping with such a young child?
[23-May-2011 16:23:45] <axelilly> I'm having trouble finding good documentation on how to create a zenpack for Zenoss 3    Any recommendations?
[23-May-2011 16:24:12] <rocket> if your a purist its not camping
[23-May-2011 16:24:29] <Hackman238> axelilly: There is no comprehensive docs, unfortunately
[23-May-2011 16:24:30] <rocket> but since its a tent and outdoors with a campfire .. I call it good enough
[23-May-2011 16:24:45] <Sam-I-Am> awww, you're not taking the 50 foot RV?
[23-May-2011 16:24:54] <Sam-I-Am> with air conditioning
[23-May-2011 16:25:14] <axelilly> Hackman238: bummer,  I'm finding the most confusion with how to create componenets.
[23-May-2011 16:25:17] <rocket> no .. its in the shop .. getting a spoiler for speed
[23-May-2011 16:25:19] <Hackman238> Sam-I-Am: Dont forget the dish and 4g reciever
[23-May-2011 16:25:29] <Hackman238> rocket: LOL
[23-May-2011 16:25:43] <dhopp> I had some friends that wanted to do a "ManCamping" weekend…they were taking a camper that had A/C.  I told them that's not ManCamping
[23-May-2011 16:26:12] <Hackman238> dhopp: Agreed.
[23-May-2011 16:26:31] <axelilly> I'm confused as to why Zenoss corp doesn't publish documentation on how to extend the product.
[23-May-2011 16:26:32] <Hackman238> dhopp: Real camping makes you appreciate regular life more
[23-May-2011 16:26:37] <rocket> Well we are just going to a campsite .. not really a hiking expedition in etc with one pack etc ..
[23-May-2011 16:26:37] <Sam-I-Am> last time i went camping it involved a tent, began raining, and i woke up surrounded by water.  awesome.
[23-May-2011 16:26:47] <Hackman238> axelilly: they claim they're too busy.
[23-May-2011 16:27:03] <axelilly> Hackman238: bummerz
[23-May-2011 16:27:21] <axelilly> dhopp: That's like grilling on gas.
[23-May-2011 16:27:26] <Hackman238> axelilly: The real extent of it is that the community is not hight enough priority, at least right now.
[23-May-2011 16:28:07] <dhopp> axelilly: I perfer the taste of charcoal…but the problem is I always wait until I'm really hungry and then it's "shit..this is going to take 30 minutes before I can even start to cook"
[23-May-2011 16:28:08] <dhopp> heh
[23-May-2011 16:28:09] <axelilly> Hackman238: yea, it seems that way to me too.
[23-May-2011 16:28:13] <rocket> axelilly: to be honest we really are very very busy.  There are some changes in the works I am not at liberty to discuss but it sounds positive for the community
[23-May-2011 16:28:37] <axelilly> dhopp: Yea, that's true.
[23-May-2011 16:28:40] <Sam-I-Am> just need to hire more peeps
[23-May-2011 16:28:57] <rocket> we have in support .. it takes time to get them up to speed
[23-May-2011 16:29:05] <axelilly> rocket: That sounds good.
[23-May-2011 16:29:09] <rocket> until then its like doing your job plus half of someone elses
[23-May-2011 16:29:26] <axelilly> Seems to be a good doc, but not complete.
[23-May-2011 16:29:46] <Hackman238> axelilly: I like to think of it like a puzzle. See if you can figure the depths and mystery that is zenoss!
[23-May-2011 16:30:03] <axelilly> Hackman238: yea, it is like that for sure.
[23-May-2011 16:30:08] <Sam-I-Am> rocket: doh, no more support positions open :/
[23-May-2011 16:30:12] <Hackman238> axelilly: Frankly I'm not positive Zenoss completely understands zenoss
[23-May-2011 16:30:13] <Sam-I-Am> rocket: but there is client services
[23-May-2011 16:30:14] <rocket> axelilly: if you can make sure there is a notation with that paper on what is missing perhaps we can help jane fill in the details
[23-May-2011 16:30:27] <axelilly> Hackman238: I think...oh I'll extend zenoss to do that, it'll save us time...DOH, this is confusing.
[23-May-2011 16:30:53] <rocket> Hackman238: thems fighting words!
[23-May-2011 16:31:07] <rocket> Sam-I-Am: you want me at client services?
[23-May-2011 16:31:10] <Hackman238> axelilly: It can be. Just wait until you have to write a complex DMD script to rework broken relationships in zope!
[23-May-2011 16:31:26] <axelilly> rocket: yea, I'm re-reading it again.
[23-May-2011 16:31:26] <Hackman238> rocket: I'm only kidding :-)
[23-May-2011 16:31:47] <rocket> zenchkrels -r -x1
[23-May-2011 16:31:49] <Sam-I-Am> rocket: parse error?
[23-May-2011 16:31:54] <Sam-I-Am> rocket: i was just looking at openings
[23-May-2011 16:31:56] <rocket> that will fix most relationship issues
[23-May-2011 16:32:13] <Hackman238> rocket: nein, chkrels fails to find items not enumerated as relationships.
[23-May-2011 16:32:21] <Hackman238> rocket: Exactly.
[23-May-2011 16:32:42] <Hackman238> rocket: Leaves all the really hard to reach plaque for me to figure how to get rid of LOL
[23-May-2011 16:33:18] <axelilly> Arnold Schwarzenegger should have run zenchkrels
[23-May-2011 16:33:20] <Sam-I-Am> Hackman238: dental picks work well
[23-May-2011 16:33:23] <Hackman238> rocket: Dont get me wrong, I love the product.
[23-May-2011 16:33:37] <Sam-I-Am> do you floss your zenoss daily?
[23-May-2011 16:34:06] <Hackman238> Sam-I-Am: Acctually yes :-) We've a series of scripts we have to run to keep it tidy...mostly LOL
[23-May-2011 16:34:14] <Sam-I-Am> zenoss isnt quite FLOSS
[23-May-2011 16:34:16] <rocket> and for the most part I believe we have the relationships that are broken resolved as of 3.1
[23-May-2011 16:34:39] <Hackman238> rocket: Agreed. My bigest customers use 2.5.2 though
[23-May-2011 16:34:49] <Sam-I-Am> i'm still on 2.5.2
[23-May-2011 16:34:58] <Sam-I-Am> it mostly behaves
[23-May-2011 16:35:18] <Sam-I-Am> after some official patches, on-a-napkin patches, and other tweaks
[23-May-2011 16:35:22] <Hackman238> I dont really like the new interface- I've been experimenting in v3 to bring back the old interface ;-)
[23-May-2011 16:35:52] <Hackman238> Believe it or not most of it still works in v4 too!
[23-May-2011 16:36:15] <Sam-I-Am> v... 4?
[23-May-2011 16:36:56] <axelilly> zenoss has to get Amazon to use it to monitor AWS
[23-May-2011 16:36:59] <rocket> Hackman238: the old interface is not coming back .. there are too many advantages to the new interface
[23-May-2011 16:37:11] <kokey> Hackman238: you know that mysql issue I had earlier?
[23-May-2011 16:37:17] <Sam-I-Am> kokey: yes
[23-May-2011 16:37:17] <kokey> it went away by itself
[23-May-2011 16:37:19] <Hackman238> kokey: yes
[23-May-2011 16:37:24] <Hackman238> kokey: Reall?
[23-May-2011 16:37:30] <Hackman238> kokey: check dmesg
[23-May-2011 16:37:30] <kokey> > select count(*) from history;
[23-May-2011 16:37:30] <kokey> +----------+
[23-May-2011 16:37:30] <kokey> | count(*) |
[23-May-2011 16:37:35] <kokey> 1 row in set (0.22 sec)
[23-May-2011 16:37:39] <kokey> i promise you no caching either
[23-May-2011 16:37:57] <Hackman238> kokey: LOL sounds like hw weirdness
[23-May-2011 16:38:06] <kokey> yeah except it's on a vm cluster
[23-May-2011 16:38:10] <Hackman238> rocket: Agreed. I'm just stubborn :-)
[23-May-2011 16:38:15] <Sam-I-Am> kokey: format c:
[23-May-2011 16:38:22] <kokey> my suspicion is some kind of indexing weirdness
[23-May-2011 16:39:06] <Hackman238> kokey: Very possible
[23-May-2011 16:39:18] <Sam-I-Am> i've found after making some db tweaks, mysql goes to lunch while it rebuilds stuff
[23-May-2011 16:39:41] <kokey> yeah it's akin to rebuilding itself
[23-May-2011 16:39:52] <kokey> it seems to self-heal in the end
[23-May-2011 16:40:01] <kokey> perhaps just after a big history purge or something
[23-May-2011 16:40:22] <kokey> i wasted an hour trying to debug it earlier
[23-May-2011 16:40:40] <kokey> which i am now spending back at the office after 9pm to complete the handover documentation i wanted to complete
[23-May-2011 16:40:51] <Sam-I-Am> great timing
[23-May-2011 16:46:03] <rocket> gotta run .. see ya guys
[23-May-2011 16:46:18] <Sam-I-Am> we'll be here when you get back
[23-May-2011 16:46:25] <Hackman238> That we will.
[23-May-2011 16:59:44] <Hackman238> Quiet in in here all of a sudden LOL
[23-May-2011 17:00:02] <outkast_> you need some noise?
[23-May-2011 17:00:46] <Hackman238> Noise is ok, ordered noise is better
[23-May-2011 17:02:22] <Sam-I-Am> it gets busier in here during zenpack development months
[23-May-2011 17:02:50] <Hackman238> Sam-I-Am: yeah I bet
[23-May-2011 17:02:57] <Hackman238> Anyway I'm out. Later all.
[23-May-2011 17:03:46] <Sam-I-Am> seeya
[23-May-2011 17:08:18] kocolosk is now known as kocolosk|away
[23-May-2011 17:11:34] <kokey> ok
[23-May-2011 17:11:37] <kokey> 25 pages done
[23-May-2011 17:11:38] <kokey> going home
[23-May-2011 17:11:39] <kokey> later
[23-May-2011 18:11:44] <brandonleach> it is possible to make zenbatchload skip devices that already exist (not model them) and only model devies that do not already exist in zenoss?
[23-May-2011 18:29:11] kocolosk|away is now known as kocolosk
[23-May-2011 20:39:42] <ryker> Can anyone help me with a problem i'm having using zenbatchload?  I get the following error: 2011-05-23 17:16:56,187 INFO zen.DeviceLoader: Ip exists on
[23-May-2011 20:40:35] <ryker> The ip .16 IP belongs to a Linux vserver host and the IP .101 belongs to a guest on that host
[23-May-2011 20:42:27] <ryker> I don't see the .101 IP anywhere in the components, graphs, or properties for the .16 IP server that's already in zenoss
[24-May-2011 00:00:01] [disconnected at Tue May 24 00:00:01 2011]
[24-May-2011 00:00:01] [connected at Tue May 24 00:00:01 2011]
[24-May-2011 00:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[24-May-2011 00:35:15] kocolosk is now known as kocolosk|away
[24-May-2011 02:08:29] <rockhopper_> Hey guys, how do i make commands to run on a specific machine?
[24-May-2011 08:44:47] <Hackman238> Hey all
[24-May-2011 08:45:01] <Hackman238> Fedora 15 out today
[24-May-2011 09:14:43] kocolosk|away is now known as kocolosk
[24-May-2011 09:23:17] <chachster> Hi, having trouble getting zenjmx working for my zenoss installation.  When trying to monitor a remote box on the same subnet, i get a noroutetohost error.  I'm using the IP which I'm thinking might not be helping.  Any suggestions?
[24-May-2011 09:28:39] <Hackman238> chachster: Can you ping the target from zenoss?
[24-May-2011 09:29:36] <chachster> @Hackman238:  Yes, i can ping the IP
[24-May-2011 09:32:00] <locohost> im trying to add disk io threaholds, when i navigate to infrastructure, I click Server/Exchange and advanced then under monitoring templates I see  several thresholds that I have working, cpu, and available memory
[24-May-2011 09:32:18] <locohost> under advanced, monitoring templates...
[24-May-2011 09:33:07] <locohost> how do I move this into /servers/windows/exchange so the threhhold applies to these machines?
[24-May-2011 09:35:55] <Hackman238> chachster: Thats a strange error to get if the zenoss host can route packets to the target. Do you use distributed collectors?
[24-May-2011 09:36:00] <dhopp> chachster: jmx typically will use a different hostname (commonly hostname-jmx).  Do you have that in your DNS or HOST file?
[24-May-2011 09:37:04] <chachster> @Hackman238:  Not sure what distributed collectors are, installed Zenoss not long ago and still learning about it.
[24-May-2011 09:37:22] <chachster> @dhopp:  How would i find out what hostname jmx is trying to use?
[24-May-2011 09:37:26] <dhopp> chachster:  what appserver are you using?  Tomcat?
[24-May-2011 09:37:55] <chachster> @dhopp:  I'm using Tomcat that comes bundles in Liferay
[24-May-2011 09:38:46] <dhopp> chachster: can you do a ps aux | grep tomcat and see if you see something like java.rmi.server.hostname= ?
[24-May-2011 09:41:00] <locohost> guess i cant just apply a copy of the built in...i have to add one and copy all the oids and such in..
[24-May-2011 09:41:14] <chachster> @dhopp:  I don't see it.  The following parameters I've set myself as per recommendations from several articles I've googled:  Maybe it's just a matter of having one more parameter?
[24-May-2011 09:46:01] <dhopp> chachster: is your hostname resolvable?
[24-May-2011 09:48:09] <chachster> @dhopp:  Yes, my hostname is resolvable.  Zenoss is currently using the IP Address and I've seen some articles mentionning this might possibly be an issue, not sure how i'd change that in Zenoss and before I do, would like to know that it's a problem.
[24-May-2011 09:49:39] <dhopp> chachster: according to you need to set the -Djava.rmi.server.hostname variable
[24-May-2011 09:50:03] <dhopp> chachster: probably can set it to the hostname of the machine, or set it to hostname-jmx and add that to your zenoss server HOST file
[24-May-2011 09:51:07] <chachster> @dhopp:  Thanks for the article!  I'll try that out, it would make sense that it's something like that that's missing.
[24-May-2011 09:57:56] <locohost> after I add my new threshold, do I have to do anything other then set zmonitor to true?
[24-May-2011 09:58:05] <locohost> i dont need to remodel or anything, rihgt?
[24-May-2011 10:07:55] <rockhopper_> hey guys, I want to notify by sms when there are errors, but via an sms service
[24-May-2011 10:08:28] <rockhopper_> like twilio
[24-May-2011 10:08:29] <froztbyte> I replaced the actual pager command with a script that does exactly that
[24-May-2011 10:08:53] <rmatte> rockhopper_: as you as you have some way of sending the message via the commandline you just change the pager command in the settings
[24-May-2011 10:08:59] <rmatte> as long as you have*
[24-May-2011 10:09:37] <rockhopper_> rmatte: i'll try that
[24-May-2011 10:09:38] <rockhopper_> thanks
[24-May-2011 10:10:40] <rockhopper_> rmatte: hm.. any chance you'll help me more "step by step" to get to there?
[24-May-2011 10:10:51] <rockhopper_> I can't find the pager command
[24-May-2011 10:12:03] <froztbyte> settings
[24-May-2011 10:12:10] <froztbyte> general/first tab
[24-May-2011 10:13:46] <rockhopper_> froztbyte: huh? on zenoss-core?
[24-May-2011 10:13:55] <froztbyte> yes
[24-May-2011 10:13:59] <froztbyte> one 2.5 anyway
[24-May-2011 10:16:04] <froztbyte> on*
[24-May-2011 10:16:10] <rmatte> rockhopper_: I don't have time to do that, I'm swamped today
[24-May-2011 10:17:01] <rmatte> rockhopper_: Just go to Settings, it's right there
[24-May-2011 10:17:08] <rmatte> "Page Command"
[24-May-2011 10:17:38] <rockhopper_> rmatte: oh, i see. many thanks
[24-May-2011 10:18:13] <rmatte> np
[24-May-2011 10:19:42] <rmatte> Enter a Page Command as necessary if you are using Zenoss to send pages. The pageCommand variable enables Zenoss to execute the pageCommand when a page is sent, and writes the message to the standard input of the subshell. The command prints any error messages to standard output. This enables a wider ranging of paging customization.
[24-May-2011 10:19:42] <rmatte> The standard Zenoss Page Command is:
[24-May-2011 10:19:42] <rmatte> $ZENHOME/bin/zensnpp localhost 444 $RECIPIENT
[24-May-2011 10:19:42] <rmatte> This uses ZENHOME to send the page from the localhost; however, you can use any page command customizations you want. In this case $RECIPIENT is actually the paging address for the user, as set in the settings for each user.
[24-May-2011 10:20:27] <rmatte> so as long as the page command can accept the message on stdin it'll work
[24-May-2011 10:20:41] <rmatte> you may have to write a little wrapper script to receive the message and pass it to the command
[24-May-2011 10:23:28] <fragfutter> can i add additional actions (beside page, email)?
[24-May-2011 10:24:16] <rmatte> fragfutter: explain the scenario
[24-May-2011 10:24:26] <rmatte> I assume you mean for alerting rules?
[24-May-2011 10:24:32] <fragfutter> correct.
[24-May-2011 10:24:45] <rmatte> what types of additional actions are you looking for?
[24-May-2011 10:25:34] <fragfutter> click your username, click alerting rules, edit a rule, see the dropdown box for action. It contains two values page and email. i would like to have, something like page, email, jabber, sms, ...
[24-May-2011 10:25:57] <rmatte> you can't do that... but you could use the event manager for it
[24-May-2011 10:26:05] <rmatte> Events -> Event Manager -> Commands
[24-May-2011 10:26:37] <fragfutter> *hm* but then i have two different places to define alert rules.
[24-May-2011 10:27:10] <rmatte> The only way you're going to get what you want is to write some sort of custom script to do all 4 things and execute it via the commands section
[24-May-2011 10:27:21] <rmatte> you can't add new actions in to the alerting system, they are hardcoded
[24-May-2011 10:27:33] <fragfutter> rmatte: ok. hardcoded was the keyword.
[24-May-2011 10:33:13] <fragfutter> i have a cluster that provides a service on a floating IP Address. i would like to monitor each cluster node and an additional device which is the floating ip. but modelling the cluster nodes will discover the IP and zenoss refuses to allow me adding it to a different device. Any idea?
[24-May-2011 10:35:55] <rmatte> fragfutter: delete the interface that has that IP, add the device with the IP, then remodel the device to rediscover the interface with the ip
[24-May-2011 10:39:39] <fragfutter> rmatte: and it will not complain that two devices have this IP?
[24-May-2011 10:40:59] <fragfutter> just tested it doesn't. strange.
[24-May-2011 10:56:02] <fragfutter> assume i have a command-check that only produces a text message (and the right exit code). There is nothing i want to graph, but it should still be polled every five minutes. can i integrate this into zenoss so that from the device it is visible that it is monitoring and displays me the last output somewhere?
[24-May-2011 10:58:12] <fragfutter> in case of an error, that is easy. it produces an event. but during normal operation i don't want to produce an event and i still need to see to output somewhere
[24-May-2011 11:02:36] <Simon4> fragfutter: you can probably still make a command template without any datapoints, and have the command return "all happy" when it's fine and raise an event when it fails?
[24-May-2011 11:02:53] <Simon4> unsure what you mean by "need to see output"
[24-May-2011 11:03:05] <fragfutter> Simon4: yes. the template without graph part is fine.
[24-May-2011 11:04:13] <fragfutter> Simon4: something like: "the user count is fine and currently the following users are logged in"
[24-May-2011 11:05:11] <fragfutter> so i a) know that there are not to many users. b) can simply click into zenoss to look who is logged in (at least five minutes ago) and c) see in the interface that it is monitored without looking which templates are bound to this device.
[24-May-2011 11:06:02] <Simon4> hmm
[24-May-2011 11:21:49] <eric256> to zenoss, got it setup in a virutal appliance, got SNMP working on my server, snmpwalk works to get data from the server...but the graphs for the server don't seem to be populating. i'm not sure how to check if zenoss is actualy getting data from the server or not...and at what interval
[24-May-2011 11:22:56] <jmp242> It checks every 60 seconds or 5 minutes depending on the perf daemon
[24-May-2011 11:23:10] <jmp242> you have to wait 15 minutes or 3 full cycles to see graphs
[24-May-2011 11:23:57] <eric256> thanks..thought it might be something like that... is there anywhere that it reports the last value it got?  i've used nagios in the past so i was expecting something similar
[24-May-2011 11:23:57] <lthrasher> for a counter, anyway.  gauges should take only two cycles in most cases.
[24-May-2011 11:24:33] <fragfutter> eric256: the last value is normaly displayed below the graph
[24-May-2011 11:26:57] <fragfutter> eric256: and to check which values are polled you can run (from a shell) "zenperfsnmp run -v 2 -d"
[24-May-2011 11:27:00] <eric256> lol i was just too impatient, while you said that the graph page reloaded and shows data
[24-May-2011 11:27:02] <eric256> awsome
[24-May-2011 11:27:14] <lthrasher> hmm.  josh's latest e-mail campaign urges me to "take the next step and request a free evaluation of Zenoss."  not Zenoss Enterprise, just Zenoss.  hope i'm just being hypersensitive about that.
[24-May-2011 11:28:06] <rmatte> lthrasher: I assume that email as because you did the Zenoss Enterprise test drive
[24-May-2011 11:28:12] <rmatte> so it's kind of implied
[24-May-2011 11:28:19] <rmatte> is because*
[24-May-2011 11:32:45] <lthrasher> rmatte: i hope that's it; been a bit on edge about the prospects of core ever since the relstorage kerfuffle.
[24-May-2011 11:33:10] <rmatte> they're not going to kill core
[24-May-2011 11:33:40] <lthrasher> but the decision seems taken that they will cripple it relative to enterprise
[24-May-2011 11:33:54] <rmatte> well, that's always technically been the case
[24-May-2011 11:34:02] <lthrasher> <-- freetard
[24-May-2011 11:34:37] <rmatte> I'm not a fan of the removal of the relstorage code either, but we can do without
[24-May-2011 11:35:19] <lthrasher> when you come to a fork in the code...
[24-May-2011 11:36:03] <rmatte> that's not too much of a fork... honestly, relstorage is a Zope plugin, we could probably develop a community pack for it
[24-May-2011 11:36:15] <fragfutter> ... you need to find enough people to maintain it
[24-May-2011 11:37:06] <rmatte> the code that they tore out is floating around as GPL code (some people got their hands on it)
[24-May-2011 11:37:17] <rmatte> so it probably wouldn't be overly difficult to put the hooks back in with a ZenPack
[24-May-2011 11:37:21] <jmp242> Wait, I thought the main change for the next Zenoss version was relstorage?
[24-May-2011 11:37:27] <jmp242> so what is going on with Core then?
[24-May-2011 11:37:29] <lthrasher> seems a waste to duplicate the effort, that's all.
[24-May-2011 11:37:43] <rmatte> jmp242: Core won't have relstorage
[24-May-2011 11:37:56] <rmatte> jmp242: though there are also significant changes to the daemons, which Core will get
[24-May-2011 11:38:15] <rmatte> they are pushing more of the event filtering downstream to the daemons themselves, rather than having zenhub do all the work
[24-May-2011 11:38:29] <jmp242> Hmmm - - - so what's Core using, same ZODB?
[24-May-2011 11:38:34] <rmatte> correct
[24-May-2011 11:38:50] <jmp242> seems odd to make that level different between enterprise and Core to me
[24-May-2011 11:38:58] <rmatte> though like I said, if a few people put their minds together I'm sure we could get relstorage working in Core via a ZenPack
[24-May-2011 11:39:20] <jmp242> won't that make Zenpacks for core not work for enterprise etc? if the whole method for storing properties is different?
[24-May-2011 11:39:20] <rmatte> They are clawing for unique enterprise features right now...
[24-May-2011 11:39:29] <rmatte> though in my opinion their pricing is completely unrealistic
[24-May-2011 11:39:37] <rmatte> which is what's hurting their sales more than anything
[24-May-2011 11:39:55] <rmatte> We managed to talk the price down with them lately, and it was still really high compared to most other tools
[24-May-2011 11:40:10] <jmp242> well, I suppose it depends though
[24-May-2011 11:40:21] <jmp242> but I certainly could not ever afford their enterprise product
[24-May-2011 11:40:31] <fragfutter> rmatte: thats it. the pricing for enterprise is so unrealistic that people prefer to bother rewriting enterprise packs and features
[24-May-2011 11:40:33] <lthrasher> lowering the ceiling...
[24-May-2011 11:40:43] <jmp242> But compared to Tivoli etal I suppose
[24-May-2011 11:40:46] <jmp242> it may be cheaper
[24-May-2011 11:40:48] <rmatte> The first quote they came back to us with for Zenoss Enterprise + Datacenter Insight, for 1009 devices... $80,000
[24-May-2011 11:40:54] <rmatte> which is ridiculous
[24-May-2011 11:40:58] <jmp242> A year right
[24-May-2011 11:41:01] <rmatte> a year, yes
[24-May-2011 11:41:02] <lthrasher> but tivoli et al keep working if you don't renew your maintenance
[24-May-2011 11:41:04] <jmp242> not even buy in with 20% maintenance
[24-May-2011 11:41:14] <jmp242> that's the other big thing
[24-May-2011 11:41:22] <rmatte> and that $80,000 was after a supposed 20% discount
[24-May-2011 11:41:32] <jmp242> we *might* pay a "large" (for us, which is much less than that) buy in
[24-May-2011 11:41:39] <rmatte> though they did eventually lower the price to the point where we will be considering it again in about a year
[24-May-2011 11:41:49] <jmp242> with the standard sort of maintenance costs
[24-May-2011 11:42:15] <rmatte> I wish they'd be more flexible in the sense that... I don't want nor need support from them, yet it's a forced inclusion in the price
[24-May-2011 11:42:17] <jmp242> The main issue for us is that we went with Core because it's free
[24-May-2011 11:42:27] <rmatte> same with the large number of enterprise Zenpacks which I'd never use
[24-May-2011 11:42:39] <rmatte> they don't offer any form of modular pricing
[24-May-2011 11:42:42] <rmatte> it's all or nothing
[24-May-2011 11:42:45] <jmp242> That's true
[24-May-2011 11:42:55] <jmp242> I've posted before about the pricing being kind of crazy
[24-May-2011 11:42:56] <lthrasher> rmatte: they're at the five-year mark with the investors, so flexibility isn't something they can really afford.
[24-May-2011 11:43:18] <rmatte> lthrasher: well, it would be nice to see them get there eventually
[24-May-2011 11:43:20] <jmp242> I thought they had a LOT of customers even at that price though
[24-May-2011 11:43:53] <rmatte> They claim to be a cheaper alternative to the big 4... but with pricing like that it doesn't cost much more to just go with one of the big 4 lol
[24-May-2011 11:43:56] <lthrasher> show of hands: enterprise customers who have renewed at least one time
[24-May-2011 11:44:34] <rmatte> lthrasher: I'm sure the majority do
[24-May-2011 11:44:41] <jmp242> But I do think that if they want customers, they need to make the minimum buy in around 5% of that, and have it not be per monitored device I think. It doesn't cost a lot to run a webstore for software licenses w/out the support
[24-May-2011 11:44:54] <rmatte> lthrasher: once you get over the hurdle of justifying the pricing then it's not a big deal
[24-May-2011 11:45:28] <rmatte> jmp242: that per device thing is killer for smaller to medium sized businesses...
[24-May-2011 11:45:47] <rmatte> it's great for huge companies like rackspace with devices in the tens of thousands
[24-May-2011 11:46:21] <Hackman238> Zenoss does indeed need to rework pricing for small and medium businesses.
[24-May-2011 11:46:59] <jmp242> Well - I would expect per device would be a killer all around myself
[24-May-2011 11:46:59] <rmatte> In contrast, we're monitoring those same 1009 devices in Solarwinds and we're paying them $12,000/year
[24-May-2011 11:47:04] <jmp242> I much prefer site licenses
[24-May-2011 11:47:22] <jmp242> Soo much easier to deal with the licensing / renewal
[24-May-2011 11:47:31] <jmp242> even if it cost more than say the exact amount I have now
[24-May-2011 11:48:14] <lthrasher> free software without a community / enterprise divide and available renewable support is even easier to deal with
[24-May-2011 11:48:24] <jmp242> Well, yes
[24-May-2011 11:48:44] <jmp242> I think the divide as it *was* where enterprise got you zenpacks
[24-May-2011 11:48:47] <jmp242> works fine for me
[24-May-2011 11:49:00] <jmp242> I.e. you're buying plugins seems fine for both sides
[24-May-2011 11:49:14] <jmp242> but changing out the guts seems to lose a lot of benefit to me for both sides
[24-May-2011 11:49:35] <jmp242> i.e. we can hire someone to code a zenpack on Core
[24-May-2011 11:49:43] <jmp242> but that can benefit enterprise customers
[24-May-2011 11:49:58] <jmp242> third parties could probably even sell their own zenpacks
[24-May-2011 11:49:59] <lthrasher> i suspect zenoss eng feels the same but is getting overruled by the bottom-line watchers.
[24-May-2011 11:51:00] <rmatte> lthrasher: I don't see how they are doing anything other than making lots of money right now.  They have some big companies as customers.
[24-May-2011 11:52:16] <lthrasher> rmatte: unless / until the IPO, all we can do here is speculate
[24-May-2011 11:52:17] <eric256> i feel stupid sometimes. lol. down at the bootm it mentions in stalling informant (which i did on the server, and ran the test walk and it works)....but how do i get that data into graphs etc?  it almost makes it sound automatic, but there is some good data there in the SNMP walk i would like to report on
[24-May-2011 11:54:06] <lthrasher> rmatte: to me, the tightening of the grip on feature parity suggests desperation.  but i'm a skeptic.
[24-May-2011 11:54:22] <jmp242> One of the great things about zenoss is / was the community and extensibility etc
[24-May-2011 11:54:34] <jmp242> but yes, you can get some great monitoring tools that are closed source for a lot less
[24-May-2011 11:55:05] <lthrasher> a skeptic who has a meeting to get to... ta!
[24-May-2011 11:55:07] <rmatte> lthrasher: The thing is, like I said, this is nothing new... they've done stuff like this before
[24-May-2011 11:55:23] <lthrasher> and yet we all keep coming back for more
[24-May-2011 11:55:33] <rmatte> lthrasher: I'm not sure how long you've been using Zenoss, but I'm coming up on about 3 years soon.
[24-May-2011 11:56:18] <jmp242> PRTG is for instance  4700 for a unlimited node monitored system
[24-May-2011 11:56:38] <jmp242> and I think it does a lot more OOTB
[24-May-2011 11:56:54] <lthrasher> rmatte: about two years, but i'm not nearly as dedicated to keeping up with happenings as you are.
[24-May-2011 11:57:01] <lthrasher> gotta run for real now
[24-May-2011 11:57:28] <rmatte> One thing that's changed recently is Matt Ray leaving... and Nick Yeates taking his position.  Matt used to fight tooth and nail to try to keep as many features in Core as possible.  I'm not sure if Nick is doing the same or not.
[24-May-2011 11:57:32] <rmatte> He may be
[24-May-2011 11:57:35] * rmatte shrugs
[24-May-2011 11:58:12] <jmp242> yea
[24-May-2011 11:58:57] <jmp242> Zenoss should be careful, the community does bring in a lot of customers, builds a good third party ecosystem (look at Jane Curry + Skills 1st for instance)
[24-May-2011 11:59:11] <rmatte> jmp242: There's also the community ZenPacks
[24-May-2011 11:59:22] <jmp242> but if you kill that off with making many differences between CORE and Enterprise .  . .
[24-May-2011 11:59:28] <rmatte> some of which are horribly authored, but there are some gems
[24-May-2011 11:59:52] <jmp242> I mean, Zenoss isn't going to the UK, or providing random zenpack authoring services that I know of
[24-May-2011 12:00:02] [disconnected at Tue May 24 12:00:02 2011]
[24-May-2011 12:00:02] [connected at Tue May 24 12:00:02 2011]
[24-May-2011 12:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[24-May-2011 12:00:45] <rmatte> I'm going to devote some time this week to coding a daemon for my formula datasource pack
[24-May-2011 12:00:49] <rmatte> I want to see it finished.
[24-May-2011 12:01:23] <jmp242> yea
[24-May-2011 12:01:37] <rmatte> Once it's finished it's going to rock
[24-May-2011 12:01:38] <rmatte>
[24-May-2011 12:02:30] <rmatte> It's a shame that most of the more polished packs are just packs to duplicate functionality present in enterprise
[24-May-2011 12:02:45] <rmatte> wastes community development time that could be spent elsewhere
[24-May-2011 12:03:43] <rmatte> one void that we still need filled is vmware monitoring (which Eric Enns was working on)
[24-May-2011 12:04:12] <jmp242> Mmmm...
[24-May-2011 12:04:46] <jmp242> What is relstorage supposed to do (i.e. why would the community want to do a Zenpack to add that in)?
[24-May-2011 12:05:00] <rmatte> It allows the zope database to reside within MySQL
[24-May-2011 12:05:05] <fragfutter> jmp242: move objects from the zodb to a "normal" relational database.
[24-May-2011 12:05:08] <rmatte> which in turn provides better performance than using a flat file
[24-May-2011 12:05:36] <rmatte> Major performance gain
[24-May-2011 12:05:40] <jmp242> Ahhh
[24-May-2011 12:05:48] <Hackman238> jmp242: Very significant performance gain
[24-May-2011 12:05:52] <jmp242> I thought that was the main point
[24-May-2011 12:06:07] <jmp242> though I don't do enough on the hardware I have to have performance issues so . . . meh from me lol
[24-May-2011 12:06:23] <rmatte> ...but like I said... it's just a Zope plugin... if people in the community put their minds together, we could duplicate support for it
[24-May-2011 12:06:44] <rmatte> the problem is, the support would need to be maintained between versions
[24-May-2011 12:06:48] <jmp242> Zenoss might also want to consider something like OPSI does for funding some community development
[24-May-2011 12:06:58] <jmp242> or for the community developers or something
[24-May-2011 12:07:09] <rmatte> jmp242: wishful thinking
[24-May-2011 12:07:29] <jmp242> i.e. work out how much it costs to do X, have people pay for the code till X is paid off, then OSS it (or pledge in advance before coding)
[24-May-2011 12:07:35] <jmp242> yea
[24-May-2011 12:07:36] <jmp242> I know
[24-May-2011 12:10:33] <jmp242> If the community had something to pledge amounts or whatever to fund community developers - I could see the Zope plugin getting done and maintained
[24-May-2011 12:10:50] <jmp242> but I don't know really how to get that done
[24-May-2011 12:11:20] <fragfutter> *shhh* everbody quiet. community support manager entered the channel
[24-May-2011 12:11:37] <Simon4> how about that local sports team?
[24-May-2011 12:11:41] <nyeates> hehe
[24-May-2011 12:11:50] <jmp242> lol
[24-May-2011 12:12:21] <rmatte>
[24-May-2011 12:12:24] <Simon4> fragfutter: pity about that zenoss-logger thing, we really need to kill it
[24-May-2011 12:12:41] <fragfutter> we could start the zenoss-foss channel
[24-May-2011 12:12:55] <rmatte> fragfutter: #zenoss-fork
[24-May-2011 12:12:56] <rmatte> >:)
[24-May-2011 12:13:32] <nyeates> welp, its ok, i cant stop you all from anything - only communicate back to our people what is going on
[24-May-2011 12:13:48] <rmatte> hehe
[24-May-2011 12:14:24] <jmp242> I'm sure communities can be a pain
[24-May-2011 12:14:27] <rmatte> "They are rallying in the streets with torches and pitchforks."
[24-May-2011 12:14:30] <nyeates> brainstorm - what, minimally, would make the community healthy?
[24-May-2011 12:14:42] <rmatte> relstorage
[24-May-2011 12:14:47] <nyeates> im working on a number of things and want to see if im down the right rd
[24-May-2011 12:14:54] <nyeates> makes sense :-)
[24-May-2011 12:14:58] <rmatte> lol
[24-May-2011 12:15:15] <rmatte> I guess it's all based on your definition of "healthy"
[24-May-2011 12:15:21] <nyeates> this is true
[24-May-2011 12:15:43] <jmp242> I think community members like myself find it hard to contribute in some ways
[24-May-2011 12:16:24] <jmp242> I don't want to be a "freeloader", and certainly enjoy when I can make contributions (like helping get the Predictive Threshold zenpack updated), but I often find the financial buy in beyond my budgets.
[24-May-2011 12:16:30] <eric256> hmmph, it keeps hanging on "Loading. Please wait..." , in the devices listing, under chrome, explorer and FF
[24-May-2011 12:16:58] <rmatte> eric256: sounds like your Zope database may be overworked and is hence hanging
[24-May-2011 12:17:25] <fragfutter> nyeates: relstorage, documentation, tell us the strategy (where is zenoss going, how do we distinguish core-enterprise)
[24-May-2011 12:17:29] <nyeates> jmp242: I think you contribute quite a bit to the support side of the community - you tirelessly keep up the FAQs
[24-May-2011 12:17:30] <eric256> only have 2 hosts lol
[24-May-2011 12:18:34] <jmp242> nyeates: I think the main thing is we as a community don't want to find that Zenoss Core is removed from relevance or so different that what we do wouldn't translate at all to enterprise customers (such as incompatible zenpacks because of relstorage etc)
[24-May-2011 12:19:07] <rmatte> nyeates: You've probably been hearing "relstorage" a lot from people in here lately... it was a long anticipated feature for Core... what they did by tearing it out of Core has soured quite a few people in the community.  I've noticed a change in atmosphere in here since that happened.
[24-May-2011 12:19:44] <nyeates> fragfutter: points taken. I know nothing of what is occuring with relstorage - i will pass the msg for sure. Documentation has made headway.... we are going to work with Jane to vet and QA her docs and bring them under our wing. Strategy from CEO is going to come soon. He might make a blog post or come here or some kind of wide communication.
[24-May-2011 12:19:47] <rmatte> I don't know if there's anything you can actually do about that... but it would be appreciated if that could be voiced to the higher ups.
[24-May-2011 12:20:21] <Hackman238> nyeates: Mabye some of the higher ups should take 5 and join this thread.
[24-May-2011 12:20:21] <jmp242> I think the community wants to make sure that Core is actually still the core of Enterprise
[24-May-2011 12:20:26] <rmatte> I'm glad to hear about the documentation.
[24-May-2011 12:20:47] <jmp242> especially for those of us who might someday want to upgrade and know that they can just plug in zenpacks etc but their customizations or whatever is going to work
[24-May-2011 12:21:02] <rmatte> Yeh, that's really the main issue here.  What is the criteria for something being a "Core" feature?
[24-May-2011 12:21:29] <nyeates> How about we see if I can get someone with a managerial voice, to come to this weeks Thur dev meeting? No guarentee, but how is that idea?
[24-May-2011 12:21:31] <rmatte> If you start using different database storage methods between core and enterprise, is core really still a core release by definition?
[24-May-2011 12:21:41] <rmatte> That's a good idea
[24-May-2011 12:22:18] <fragfutter> rmatte: afaik the storage backend is transparent from zopes point of view.
[24-May-2011 12:22:24] <Hackman238> nyeates: Thats what I'm talking about.
[24-May-2011 12:22:28] <jmp242> And for the worry about enterprise features being enticing - the datacenter view stuff both makes sense as an enterprise feature (not that I'd turn it down in Core, but it is in the same flavor as the original enterprise divide) and would be enticing to people with budgets to upgrade
[24-May-2011 12:22:45] <jmp242> the database storage does seem pretty core IMO
[24-May-2011 12:22:48] <eric256> just curious, is the data storage backend not abstracted away enough that plugins use the abstraction and don't care where data goes?
[24-May-2011 12:22:56] <rmatte> fragfutter: I understand that, but I'd assume there's some sort of migration process when going from standard zeodb to relstorage
[24-May-2011 12:23:51] <nyeates> eric256: i think there are tie ins that dont abstract it....i cant be sure though
[24-May-2011 12:24:31] <nyeates> u all might want to read simons blog
[24-May-2011 12:24:33] <jmp242> nyeates: I think the other point in this kerfluffle as was said - is that if you advertise something as a core feature, follow through with that. I know it can be hard to do sometimes when you're not sure as you're coding it, but up till now Zenoss has pulled it off that I know of
[24-May-2011 12:24:36] <rmatte> Yeh agreed, relstorage should be put back in to Core, Datacenter Insight is more than enticing enough to attract customers
[24-May-2011 12:24:42] <nyeates> have u  seen simons blog on relstorage??
[24-May-2011 12:24:48] <rmatte> (though once they see the price of datacenter insight they may change their minds :))
[24-May-2011 12:25:01] <jmp242> link?
[24-May-2011 12:25:23] <nyeates>
[24-May-2011 12:25:51] <Hackman238> Relstorage is something the community could handle, but I think we would all rather have the feature from Zenoss as previously offered rather than hack in a solution
[24-May-2011 12:26:16] <nyeates> understood hackman
[24-May-2011 12:26:39] <rmatte> nyeates: It seems like we're always hacking in solutions and Core just keeps getting harder and harder to support between releases, so no one upgrades
[24-May-2011 12:26:45] <Hackman238> If the feature is so easy for the community to handle- why can Zenoss drop it in core?
[24-May-2011 12:28:01] <rmatte> I agree with Hackman on that
[24-May-2011 12:28:44] <rmatte> The problem with not having relstorage in core, but having community solutions for implementing it is that proper QA won't be done on it for future releases
[24-May-2011 12:28:54] <nyeates> I dont know that I can answer that - as in, I dont have the answer...Keep it on queue for when a line of comunication is opened to mgmt
[24-May-2011 12:28:55] <rmatte> which delays community members upgrading their installs
[24-May-2011 12:29:03] <rmatte> which in turns delays proper testing of new releases
[24-May-2011 12:29:18] <nyeates> this is true
[24-May-2011 12:30:27] <nyeates> jmp242: btw, we are very clear on the sentiment of the code pull - we (me, managers, etc) have heard about it from many angles. Thanks for establishing all this from many angles. We definetely want to hear your voice, so its good.
[24-May-2011 12:31:01] <jmp242> I would also think that if you've got a major internal change like relstorage, you're not going to get the second beta for Enterprise you have gotten with the final Core releases. Occasionally they have found big issues that if I were an enterprise customer would make me quite angry to not have been caught before I did the install.
[24-May-2011 12:31:46] <jmp242> I'm recalling around the 2.4ish timeframe there were some good fixes before Enterprise went out due (I think) to the community hitting some snags and giving some bug reports etc.
[24-May-2011 12:32:16] <rmatte> Yeh, there's a reason why Core users get releases before Enterprise users... because we catch bad bugs
[24-May-2011 12:32:39] <rmatte> If the core codebase isn't using the same storage backend, that'll affect results
[24-May-2011 12:32:46] <nyeates> So here is my plan - I am going to go to my manager and show some of this convo and suggest we try to get someone in here for this Thur at 11am EST during the dev chat.
[24-May-2011 12:32:51] <nyeates> also aim for a blog post
[24-May-2011 12:32:59] <rmatte> awesome
[24-May-2011 12:33:03] <rmatte> you da man
[24-May-2011 12:33:07] <Hackman238> Sounds good.
[24-May-2011 12:33:14] <jmp242> thanks nyeates I think that's a great idea
[24-May-2011 12:35:11] <nyeates> I will say this - Me personally, over the last week or so, I have been increasingly optimistic about where Zenoss Inc is going with community. Not only is it my interest, but my job - so I think stuff is trending positive.
[24-May-2011 12:35:44] <rmatte> Always nice to hear
[24-May-2011 12:37:13] <Hackman238> Thanks Nick.
[24-May-2011 12:37:41] <nyeates> Np. Ill see you all gonna go move on this now
[24-May-2011 12:38:02] <rmatte> nyeates: sounds good, I'll be here
[24-May-2011 12:38:23] <jmp242> 11AM thursday - on my calendar
[24-May-2011 12:41:53] <Hackman238> Anyone know of a quick way to add two datapoints and threshold against that value?
[24-May-2011 12:42:24] <rmatte> Hackman238: That's what my formula datasource ZenPack does
[24-May-2011 12:42:43] <jmp242> Yea, when rmatte finishes his formula datasource?
[24-May-2011 12:42:49] <rmatte> but it's not 100% done, it uses zencommand right now.  I'm going to be writing a daemon for it soon enough.
[24-May-2011 12:43:03] <rmatte> It works fine if you don't overuse it
[24-May-2011 12:43:11] <rmatte> I wouldn't recommend using it on filesystem or interface templates
[24-May-2011 12:43:19] <rmatte> but for performance templates it's not bad, on smaller numbers of devices
[24-May-2011 12:43:23] <Hackman238> Unfortuantely I'll need to do this over many many devices.
[24-May-2011 12:43:25] <rmatte> it'll scale much better when the daemon is coded
[24-May-2011 12:43:33] <rmatte> how many is many many?
[24-May-2011 12:43:34] <rmatte> lol
[24-May-2011 12:43:41] <Hackman238> RackSpace many
[24-May-2011 12:43:43] <Hackman238> LOL
[24-May-2011 12:43:47] <rmatte> oh right
[24-May-2011 12:44:03] <rmatte> yeh, you'll have to wait on my upgraded pack unless you can figure something else out
[24-May-2011 12:44:49] <Hackman238> Alrighty. Oh, ahve you had a chance to test out that new ipsla pack?
[24-May-2011 12:45:37] <rmatte> so far the 3.x version works fine
[24-May-2011 12:45:47] <rmatte> you haven't sent me the 2.5 version
[24-May-2011 12:47:12] <Hackman238> rmatte: Oh damn. Sorry- I'll send that ASAP
[24-May-2011 12:47:59] <nyeates> Hackman238: Yeah Ryans zenpack has been the only solution in that space. I *think* there was some hackish workaround I found in a prior enterprise support case. Open a support case and send it at me if you want to look into it further
[24-May-2011 12:50:30] <rmatte> Hackman238: cool, I'll work on getting a daemon coded for my pack sooner rather than later.
[24-May-2011 12:51:18] <Hackman238> nyeates: Okay
[24-May-2011 12:51:30] <Hackman238> rmatte: Sounds good.
[24-May-2011 12:52:02] <rmatte> Interesting... just got a modelling error with the 3.x version of your pack
[24-May-2011 12:52:03] <rmatte> 2011-05-24 12:51:24,937 ERROR zen.ZenModeler: Problem while executing plugin SLADevice
[24-May-2011 12:52:03] <rmatte> 2011-05-24 12:51:24,938 ERROR zen.ZenModeler: Traceback (most recent call last):
[24-May-2011 12:52:03] <rmatte> File "/usr/local/zenoss/zenoss/Products/DataCollector/", line 613, in processClient
[24-May-2011 12:52:03] <rmatte> datamaps = plugin.process(device, results, self.log)
[24-May-2011 12:52:03] <rmatte> File "/usr/local/zenoss/zenoss/ZenPacks/ZenPacks.ShaneScott.ipSLA-3.0-py2.6.egg/ZenPacks/ShaneScott/ipSLA/modeler/plugins/", line 114, in process
[24-May-2011 12:52:03] <rmatte> if omA.rttMonCtrlAdminRttType == 1:
[24-May-2011 12:52:04] <rmatte> NameError: global name 'omA' is not defined
[24-May-2011 12:53:01] <rmatte> didn't happen on the first device that I tried it with but it did happen on the second
[24-May-2011 12:53:14] <Hackman238> rmatte: check to see if omA hasnt been interchanged with om in that python file.
[24-May-2011 12:53:22] <rmatte> k
[24-May-2011 12:54:40] <Hackman238> rmatte: Theres an updated copy up on my site if you want to grab it.
[24-May-2011 12:54:41] <rmatte> oh, that must be it... you probably pasted the stuff that I sent you, but that was for omA in the old pack
[24-May-2011 12:55:17] <rmatte> nah, I'll just make the adjustment myself for now, all good... fire me the link along with the 2.5 link and I'll grab it later.
[24-May-2011 12:55:20] <rmatte>
[24-May-2011 12:55:30] <Hackman238> rmatte: Sounds good. Thanks!
[24-May-2011 12:56:04] <rmatte> (you had already found this problem and fixed it in the updated version on your site right?
[24-May-2011 12:56:56] <rmatte> by the way, did you rename the VOiP template to Jitter?
[24-May-2011 12:57:30] <rmatte>                         if om.rttMonCtrlAdminRttType == 9:
[24-May-2011 12:57:30] <rmatte>                                 om.rttMonCtrlAdminRttType = "Jitter"
[24-May-2011 12:57:46] <rmatte> you used to have that defined as VOiP, but voip is actually 13, jitter is 9
[24-May-2011 12:58:27] <Hackman238> rmatte: Its updated, yes.
[24-May-2011 12:58:33] <rmatte> cool
[24-May-2011 12:58:45] <Hackman238> rmatte: Also all the gui is implemented.
[24-May-2011 12:59:27] <rmatte> ?
[24-May-2011 12:59:35] <rmatte> not sure what you mean by that
[24-May-2011 12:59:58] <Hackman238> rmatte: the v3 version didnt list most collected details
[24-May-2011 13:00:16] <rmatte> It did, just on a different selection in the dropdown
[24-May-2011 13:00:29] <Hackman238> rmatte: Theres more now.
[24-May-2011 13:00:39] <Hackman238> rmatte: much much more
[24-May-2011 13:01:02] <rmatte> under "Graphs" or under "Perf"?
[24-May-2011 13:01:35] <rmatte> I guess I can just download it and see for myself
[24-May-2011 13:01:46] <Hackman238> rmatte: Display: Details
[24-May-2011 13:02:05] <rmatte> oh
[24-May-2011 13:02:08] <rmatte> cool
[24-May-2011 13:02:18] <rmatte> yeh, it wouldn't show the type there or anything
[24-May-2011 13:02:51] <rmatte> you going to message me the links through the community site?
[24-May-2011 13:03:24] <rmatte> eugh, time for me to do some php coding
[24-May-2011 13:03:34] <rmatte> which I am not exceptionally good at
[24-May-2011 13:03:35] <rmatte> lol
[24-May-2011 13:04:51] <Hackman238> I'll get you a link via forum pm
[24-May-2011 13:05:05] <rmatte> thanks
[24-May-2011 13:05:05] <Hackman238> Good luck!
[24-May-2011 13:05:13] <rmatte>
[24-May-2011 13:08:57] <Sam-I-Am> moo.
[24-May-2011 13:09:05] <jmp242> bbl
[24-May-2011 13:09:14] <Sam-I-Am> rmatte: sup!
[24-May-2011 13:50:12] <rmatte> not much, just doing some php coding
[24-May-2011 13:50:39] <Sam-I-Am> havent seen you in here much
[24-May-2011 13:50:43] <Sam-I-Am> must be buried
[24-May-2011 13:56:43] <rmatte> yeh I am
[24-May-2011 13:56:48] <rmatte> non-stop coding over the last week
[24-May-2011 13:56:54] <rmatte> just finally finishing the project up today
[24-May-2011 13:57:50] <Sam-I-Am> sounds exciting (in the same way as hitting your head repeatedly)
[24-May-2011 13:59:21] <Sam-I-Am> awesome, latest release of code for my brocade devices broke temperature sensors via snmp
[24-May-2011 13:59:28] <Sam-I-Am> everything returns 0
[24-May-2011 13:59:43] <Sam-I-Am> luckily low values dont trigger threshold alerts
[24-May-2011 14:46:32] <rmatte> booyeh, finally done this coding project
[24-May-2011 14:46:39] <rmatte> now on to the next one...
[24-May-2011 14:57:41] <Hackman238> rmatte: There is always more...
[24-May-2011 15:16:25] <dhopp> ugh…I was just reading through the room chat and about core/enterprise…We are implementing core because we could not get the budget approved to get enterprise…but if they are going to split it and have a "core" difference (enterprise use relstorage, core not) I might have to recommend we look at other open source.  I could see this getting really really messy with upgrades
[24-May-2011 15:17:30] <jmp242> dhopp: come to the meeting when it's scheduled then and voice your concerns
[24-May-2011 15:18:46] <dhopp> jmp242: I will be here, but if we don't have somebody that can really make a decision I have a feeling it will fall on deaf ears.  That's not a knock on Nick, it's just a matter of fact.
[24-May-2011 15:19:38] <dhopp> I don't have a problem with Zenoss having features in Enterprise that aren't in Core….but if they are going to do something like this then they can no longer call it "core".  As has been stated already
[24-May-2011 15:21:58] <Hackman238> All- I've made it abundantly clear to Nick and others over at Zenoss that definate clarification is a requirement for the community.
[24-May-2011 15:22:56] <Hackman238> Hopefully they'll listen this time.
[24-May-2011 15:25:42] <dhopp> I can't recommend nagios because every web interface out there for it sucks.  OpenNMS has an ok web interface but relies way too much on auto discovery.  Zabbix has an OK interface but requires an agent to do any real monitoring
[24-May-2011 15:26:35] <Hackman238> dhopp: There aren't any monitoring platforms with the advantages Zenoss has as far as I've seen.
[24-May-2011 15:27:16] <dhopp> Hackman238: I agree.
[24-May-2011 15:27:31] <dhopp> Hackman238: but then again you are an Enterprise customer :-P
[24-May-2011 15:28:37] <Hackman238> dhopp: I have many enterprise and core customers. I jumped on Zenoss and gave them an analysis from h3ll what will happen if they leave core behind.
[24-May-2011 15:30:00] <Hackman238> dhopp: If core were left behind- Zenoss leaves itself in a position where it's begging for the community to rise up and fork zenoss. Zenoss would create competition and hurt their product since they would no longer have the defacto solution offering the many advantages Zenoss has over over solutions.
[24-May-2011 15:30:54] <Hackman238> dhopp: I promise that Nick and the others are on our side, it's really Zenoss upper management who makes the calls,
[24-May-2011 15:33:13] <dhopp> Hackman238: Our existing monitoring system is home grown.  And it's ok for what it does..but it was written in the late 90s/early 00s and since then has just had crap bolted on the side and not a lot of real work to the underlying engine.  It relies heavily on screen scraping and so any real performance metrics that you really want to get are a pain.  The other problem is all of the original developers are no longer with the company.  If we roll out Zenoss and
[24-May-2011 15:34:29] <dhopp> Hackman238: I didn't mean an disrespect to Nick…I understand he's on our side and I understand the upper management doesn't want to give away their entire product…but I don't want to be the guy that rolls this out and then in 2 years say "well we are stuck again"
[24-May-2011 15:35:41] <Hackman238> dhopp: Dont worry- I've so many customers using this product that I wont let Zenoss make such a dumb mistake :-)
[24-May-2011 15:36:52] <Hackman238> dhopp: No worries!
[24-May-2011 17:04:07] <Hackman238> Later all
[24-May-2011 17:35:49] kocolosk is now known as kocolosk|away
[24-May-2011 19:19:04] <Hackman238b> exit
[24-May-2011 20:10:29] kocolosk|away is now known as kocolosk
[24-May-2011 20:48:42] <locohost> just modeled a cisco IBM bladecenter switch and it picked up only the interface and descriptions, thats cool, thats really mostly only what i want off t his.  Just have a 15m of data but traffic seems low on the interfaces and I cant seem to find what mib/oid the default model is pulling
[24-May-2011 20:48:52] <locohost> and want to make sure its 64 bit, cant seem to find it in the interface
[24-May-2011 20:49:08] <locohost> doing a tcpdump now waiting for next pulling cycle
[24-May-2011 20:50:16] <locohost> ewwww
[24-May-2011 20:50:19] <locohost> it is grabbing interfaces.ifTable.ifEntry.ifInUcastPkts.3
[24-May-2011 20:50:27] <locohost> yeah, thats 32 bit, huh
[24-May-2011 20:51:43] <locohost> err, thats just ucast, thats not what i want
[24-May-2011 20:55:05] <Sam-I-Am> usually theres some stuff in the logs about what the modeler is pulling
[24-May-2011 20:55:13] <Sam-I-Am> but yeah, tcpdump works in a pinch
[24-May-2011 20:56:38] <locohost> im willing to just find the 32-bit oid its pulling for throughput and replace it with the corrisponding 64bit in the collection method
[24-May-2011 20:57:01] <locohost> yeah, i grepped for hostname in /opt/zenoss/log *, lemmie try ip
[24-May-2011 20:59:08] <locohost> actually, there is something in there that spun my regex, weird
[24-May-2011 21:02:18] <locohost> bah, i guess im not seeing it cus its udp, lol
[24-May-2011 21:05:23] <Sam-I-Am> yep, it would be udp
[24-May-2011 21:06:05] <Sam-I-Am> it might be basing it on what interface type it thinks it is
[24-May-2011 21:06:29] <Sam-I-Am> i have a handful of odd devices that claim 64-bit, but return 32-bit oids... or the other way around
[24-May-2011 21:06:36] <Sam-I-Am> i usually make special templates for them
[24-May-2011 21:14:41] <locohost> i did an snmpwalk of it and see both were in there
[24-May-2011 21:15:09] <locohost> but, i always use v2 and i kinda assumed it would be too, but dont see where to change or confirm in the config
[24-May-2011 21:15:23] <locohost> i may be making progress, i have added a few scan pluggins and remodeled through the interface
[24-May-2011 21:15:49] <locohost> im hoping zenoss.snmnp.ciscomap does something cool
[24-May-2011 21:16:10] <locohost> and my graphs for interfaces appear to have been deleteted, all showing nana, so mabe it found a better counter on its own
[24-May-2011 21:16:31] <locohost> ahh, the 15m wait after editing something in that uses rrd
[24-May-2011 21:18:38] <Sam-I-Am> yeah, gotta wait a few cycles
[24-May-2011 21:28:14] <locohost> ive found where the monitoring template appears to be associated, yeah, its using ethernetCsmacd
[24-May-2011 21:28:44] <locohost> in advancted then monitoring templates, im ganna try to rename ethernetCsmacd to ethernetCsmacd.original and then rename ethernetCsmacd64 to ethernetCsmacd
[24-May-2011 21:28:47] <locohost> see what that does
[24-May-2011 21:28:57] <davetoo> wait
[24-May-2011 21:29:00] <davetoo> what version are you using?
[24-May-2011 21:31:05] <locohost> [root@d01mtrending01 log]# cat /etc/issue
[24-May-2011 21:31:06] <locohost> CentOS release 5.6 (Final)
[24-May-2011 21:31:06] <locohost> Kernel \r on an \m
[24-May-2011 21:31:06] <locohost> [root@d01mtrending01 log]# rpm -qa zenoss
[24-May-2011 21:31:06] <locohost> zenoss-3.1.0-1031.el5
[24-May-2011 21:31:21] <davetoo> I came in late, but you shouldn't have to do that.
[24-May-2011 21:31:37] <locohost> who uses 32bit now adays
[24-May-2011 21:32:00] <locohost> its a cisco bladecenter chassis switch for ibm bladecenter
[24-May-2011 21:32:18] <locohost> but, it should have pretty standard cisco data, as far as interface stats are concerned anyway
[24-May-2011 21:32:24] <davetoo> it's binding the wrong templates?
[24-May-2011 21:33:33] <locohost> well, not hte one i want, i guess it probably thinks its the right one
[24-May-2011 21:33:38] <locohost> well, now im getting snowflakes
[24-May-2011 21:33:48] <locohost> yeah, errors that it cant find the oid
[24-May-2011 21:33:53] <locohost> it must be using snmp V1
[24-May-2011 21:33:56] <davetoo> What SNMP version is the modeler using?
[24-May-2011 21:33:57] <davetoo> yeah,
[24-May-2011 21:34:12] <locohost> u know where to change that, i ddint see it in device config
[24-May-2011 21:34:16] <davetoo> I used to have to copy the template as you mentioned above, but many many verions
[24-May-2011 21:34:20] <davetoo> versions ago
[24-May-2011 21:34:33] <davetoo> It's in the Settings in the /Devices tree
[24-May-2011 21:34:39] <Sam-I-Am> i think i set snmpv2 way high in the tree so it would default for most everything
[24-May-2011 21:34:51] <davetoo> I thought that was the default these days?
[24-May-2011 21:34:53] <locohost> cool, i was wondering that but have not tried yet
[24-May-2011 21:34:59] <locohost> im also flashing 20 blackberrys
[24-May-2011 21:35:05] <Sam-I-Am> davetoo: not sure, i'm still on 2.5 :/
[24-May-2011 21:35:18] <davetoo> why? P)
[24-May-2011 21:35:58] <Sam-I-Am> it what was stable back in august
[24-May-2011 21:36:04] <davetoo> I should make a spreadsheet/table of changes between versions
[24-May-2011 21:36:12] <davetoo> s/should/would like to find/
[24-May-2011 21:36:18] <Sam-I-Am> think i compared against 3.0.0 or 3.0.1
[24-May-2011 21:36:48] <locohost> ahh, 2c, there my hommie
[24-May-2011 21:37:03] <locohost> me too dave, who uses 1
[24-May-2011 21:37:43] <Sam-I-Am> i have a couple of devices which speak 2 and only have 32-bit counters
[24-May-2011 21:37:45] <locohost> i didnt realize i was using 1, just deployed this in a rush last week to get stats on a new cluster for an email migration
[24-May-2011 21:37:45] <davetoo> Sometimes on older devices, with older versions of net-snmp or router firmware, there are bugs that make it better to use one or the other, IIRC
[24-May-2011 21:38:22] <locohost> yeah, well, seems like having it as an optoin in lower trees is usefull
[24-May-2011 21:38:42] <locohost> wasnt under the device, maybe i can set it on every class
[24-May-2011 21:38:53] <davetoo> well you don't need to
[24-May-2011 21:39:39] <davetoo> It's heirarchical
[24-May-2011 21:40:02] <davetoo> but once you add a device, it's set for that device.
[24-May-2011 21:41:22] <locohost> i guess i just missed it, i do see it not i can set it anywhere in the heirarc
[24-May-2011 21:41:46] <locohost> seems to me to make more sence to set it to 2c and you can change it for a legacy device
[24-May-2011 21:41:53] <locohost> or legacy class
[24-May-2011 21:42:18] <locohost> maybe that was causing issues in upgrades, but seems like in an upgrade you could import the setting and override
[24-May-2011 21:42:21] <locohost> this was a clean install
[24-May-2011 21:42:33] <locohost> with latest rpm on site last week
[24-May-2011 21:43:43] <davetoo> so go to INFRASTRUCTURE->Device Classes
[24-May-2011 21:43:51] <davetoo> select Details
[24-May-2011 21:44:01] <davetoo> and then Configuration Properties
[24-May-2011 21:44:05] <davetoo> zSnmpVer is what you're looking for;
[24-May-2011 21:44:15] <davetoo> it does look like V1 is still the default;
[24-May-2011 21:44:27] <davetoo> you can also look at the Overridden Objects finder thingy just below that
[24-May-2011 21:44:56] <davetoo> and in the zProperty pulldown in the upper right, select zSnmpVer and see what's assigned where.
[24-May-2011 21:45:34] <locohost> should i restart for zenoss for good measure?
[24-May-2011 21:45:42] <locohost> just cleared all snowflakes, may be worth just waiting
[24-May-2011 21:46:02] <davetoo> How many devices have the zsnmpver overridden?
[24-May-2011 21:46:10] <locohost> err, i should remodel
[24-May-2011 21:46:22] <davetoo> Try just a few devices to see what happens
[24-May-2011 21:46:24] <locohost> just like 20 windows 2k8 servers
[24-May-2011 21:46:29] <davetoo> How many devices are loaded?
[24-May-2011 21:46:34] <locohost> 22
[24-May-2011 21:46:49] <locohost> 20wk servers, the centos 5.7 host and shit swich
[24-May-2011 21:46:49] <davetoo> Worth trying.
[24-May-2011 21:47:05] <locohost> may have had snowflakes for all the servers for other things
[24-May-2011 21:47:16] <locohost> kinda dont want it to screw up my snmpinformant stats
[24-May-2011 21:47:19] <locohost> but, could just do it for 1
[24-May-2011 21:47:53] <davetoo> as the zenoss user, do this:
[24-May-2011 21:47:54] <locohost> more concerned about alerts on the servers then data, i have cacti too
[24-May-2011 21:48:12] <davetoo> echo "print dmd.Devices.zSnmpVer" | zendmd
[24-May-2011 21:48:21] <davetoo> and tell me what that returns
[24-May-2011 21:48:27] <locohost> i assume remodeling wont screw up a custom threshold thats in tthe device class /device/windows/server/exchange
[24-May-2011 21:49:01] <davetoo> Remodeling won't change anythign you've set in a template
[24-May-2011 21:49:51] <locohost> [root@d01mtrending01 log]# su - zenoss -c 'echo "print dmd.Devices.zSnmpVer" | zendmd'
[24-May-2011 21:49:51] <locohost> Welcome to the Zenoss dmd command shell!
[24-May-2011 21:49:51] <locohost> 'dmd' is bound to the DataRoot. 'zhelp()' to get a list of commands.
[24-May-2011 21:49:51] <locohost> Use TAB-TAB to see a list of zendmd related commands.
[24-May-2011 21:49:51] <locohost> Tab completion also works for objects -- hit tab after an object name and '.'
[24-May-2011 21:49:51] <locohost>  (eg dmd. + tab-key).
[24-May-2011 21:49:53] <locohost> >>> v2c
[24-May-2011 21:50:05] <davetoo> ok, that's what I wanted to see
[24-May-2011 21:50:27] <locohost> might remodeling change the windows systems to a 64bit counter for iftraffic?
[24-May-2011 21:50:39] <locohost> i wasnt too worried about them, but they do have gigabit interfaces
[24-May-2011 21:50:45] <locohost> maybe some are doing more then i though
[24-May-2011 21:50:55] <locohost> if i recall, the rollover is around 140mbps?
[24-May-2011 21:51:02] <davetoo> I wish I could answer that off the top of my head; I would think it should.
[24-May-2011 21:51:20] <davetoo> but.. it's been more than a year since I've used that snmpInformant agent
[24-May-2011 21:53:08] <locohost> well, the interface stats would not be from snmp informant, i think..ide hae to look at the oid its pulling but im guessing its just hostresournces..probably 32 bit
[24-May-2011 21:54:21] <davetoo> Try an snmp walk on a windows box for each version?
[24-May-2011 21:57:12] <locohost> now when i click on inferstructure, device, interfaces, GI/12 i see 40megabits and if i go to graphs to details on the bottom rather t hen graph i see ethernetCsmacd     /Devices     Template for 64-bit interface counters. Must use SNMP v2c for it to work.
[24-May-2011 21:57:24] <locohost> still pulling the one i renamed, but now getting values
[24-May-2011 21:57:40] <locohost> thats okay, im not ganna pull any legacy devices with this
[24-May-2011 21:57:56] <locohost> maybe these switches, a few big cats and a checkpoint firewall
[24-May-2011 21:58:14] <davetoo> You should find a way to revert that copying hack; it should not be required and could be confusing later.
[24-May-2011 21:59:47] <davetoo> I need to build some windows VMs for myself to test this, I suppose.
[24-May-2011 22:04:08] <locohost> so hard to beleive this interfface, descriptoin blade9, Gi/12 is only doing 12k
[24-May-2011 22:04:11] <locohost> bps
[24-May-2011 22:04:19] <locohost> its the storage interface for the entire blade
[24-May-2011 22:04:54] <locohost> each esx server has 2 interfaces on sw5 and sw6 for there storage, so this all ip traffic to nfs mounts
[24-May-2011 22:05:10] <locohost> maybe its all going throuhg sw5, im only looking at sw6
[24-May-2011 22:05:47] <locohost> does now s how Type:
[24-May-2011 22:05:47] <locohost> ethernetCsmacd_64 under details for the interface
[24-May-2011 22:07:32] <locohost> (modeling sw5)
[24-May-2011 22:07:47] <locohost> not sure how to revert my hack, ug, name had parenthis and crap in it
[24-May-2011 22:08:22] <davetoo> well,
[24-May-2011 22:08:39] <davetoo> if you remodel and it uses _64 then at least it's not intefering
[24-May-2011 22:10:56] <locohost> guess it could make sence t hat the traffic is on sw5 and not sw6, they are esx serers with bonded interfaaces i guess
[24-May-2011 22:14:22] <locohost> ill see what surpirzed it does not appear to be attempting to call the 64bit one and not finding it
[24-May-2011 22:14:36] <locohost> it looks to me like its still calling the 32 bit, but finding the 64 bit
[24-May-2011 22:15:26] <locohost> interface stats for windows machines were blanked out somehow, guessing from changing to snmpv2
[24-May-2011 22:15:50] <locohost> not to worried, have their stats in cacti with 64bit oid, just want an alert if an interface goes link down or something i guess
[24-May-2011 22:19:58] <locohost> nan nan nan nan, nan nan nan nan nan, nan nan nan, nan nan
[24-May-2011 22:20:17] <locohost>  /me taps foot
[24-May-2011 22:24:24] <locohost> balls, 490mbps+ on multiple interfaces
[24-May-2011 22:24:29] <locohost> on sw5
[24-May-2011 22:30:27] <locohost> tried adding 1-4, see what i get
[24-May-2011 22:30:45] <locohost> then on too all the switches in bc-2 and bc-3
[24-May-2011 22:31:06] <locohost> esx inferstructure, 40+ blades, hoping to find some kinda io bottleneck
[24-May-2011 22:31:35] <locohost> something is going on somewhere and as of yet i have not been able to determine where the bottle neck is
[24-May-2011 22:32:36] <locohost> i did find a few sources of large of the dbas is doing a dump on a 170G mssql db 2x a nice...i asked him and he said the 2nd was for good measure..../slap
[24-May-2011 22:32:48] <locohost> "because  he had problems where they didnt always complete in the past"
[24-May-2011 22:33:05] <locohost> okay, so saturate the network for 6 hrs rather then 3, great
[24-May-2011 22:36:54] <locohost> nother querky think i noticed in latest, (been almost 2 years since i ran it before last week) when i add several devices at once, it puts them in wrong catagories, doesnt do as good of a job as if i do em 1 at a time, i even started in the correct class and they are all the same class
[24-May-2011 22:37:08] <locohost> its putting them in linux when t hey are switches and im in switches, but works 1 at a time
[24-May-2011 22:37:57] <locohost> it pust the first entrey in the form in the correct spot, that i selected
[24-May-2011 22:38:04] <locohost> and took a guess for the rest, and was wrong
[24-May-2011 22:38:55] <locohost> what i mean is, i was on switches, went to add multiple devices, and put sw1, sw2, sw3 and it only does the full model for sw1
[24-May-2011 22:41:02] <locohost> well, just going 1 by one, ahh zenoss, does some amazing things awesomely but then glitchy and unexpected in weirrd ways at every turn
[25-May-2011 00:00:01] [disconnected at Wed May 25 00:00:01 2011]
[25-May-2011 00:00:01] [connected at Wed May 25 00:00:01 2011]
[25-May-2011 00:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[25-May-2011 00:00:36] kocolosk is now known as kocolosk|away
[25-May-2011 05:15:40] <kokey> anyone monitoring esxi hosts?
[25-May-2011 05:16:53] <Simon4> not I
[25-May-2011 05:18:12] <Rickson-> question about alert/ipservices, why can i select "Monitor these components" under "Monitoring" on some ipservice, and some not?
[25-May-2011 06:29:58] <kokey> anyone monitoring esxi hosts?
[25-May-2011 06:30:12] <kokey> or rather
[25-May-2011 06:30:23] <kokey> anyone monitoring esxi hosts on HP hardware?
[25-May-2011 06:30:30] <kokey> i'm going to keep trying until I find someone
[25-May-2011 06:51:16] <froztbyte>, sorry
[25-May-2011 06:51:25] <froztbyte> the one ESXi box I have around is !HP
[25-May-2011 06:51:35] <froztbyte> and I think I only track the basics on it
[25-May-2011 06:51:46] <froztbyte> "is it up? is the CPU screwed?" etc
[25-May-2011 06:52:18] <Simon4> kokey: I was going to try it once my microproliant turns up
[25-May-2011 06:55:49] <dhopp> kokey: I have an esxi on HP hardware…unfortunately haven't tried to monitor it yet
[25-May-2011 06:55:51] <dhopp> lol
[25-May-2011 06:56:58] <dhopp> kokey: what's your problem?
[25-May-2011 07:02:12] <froztbyte> Simon4: the microservers are so awesome
[25-May-2011 07:02:15] * froztbyte has one
[25-May-2011 07:02:22] <Simon4> froztbyte: sweet
[25-May-2011 07:02:31] <Simon4> it's going to be my new dev vm host
[25-May-2011 07:02:40] <froztbyte> I was raging about it in the glug channel, and a friend was just like "....why?"
[25-May-2011 07:02:41] <Simon4> 8GB ram and four 1-2TB drives
[25-May-2011 07:03:45] <froztbyte> it seems like the entire niceness of dealing with HP-level kit doesn't really rub off on some people
[25-May-2011 07:04:08] <froztbyte> but really, once you've used things in the way they design it...why would you want to ever use something "less"?
[25-May-2011 07:05:16] <Simon4> it's £99, why buy anything else?
[25-May-2011 07:05:37] <squig> froztbyte, coz we cantz affords it
[25-May-2011 07:06:14] <froztbyte> squig: the microserver pricing beats the socks off of *anything* else I found
[25-May-2011 07:06:19] <froztbyte> even stuff I build myself
[25-May-2011 07:06:56] <froztbyte> and you can pick up G2~G5 for relatively sane prices too
[25-May-2011 07:07:08] <squig> but thats why people use less because they cant afford more, im not familiat with what a microserver is exactly
[25-May-2011 07:21:48] <dhopp> Anybody here running zenoss with 600+ devices?
[25-May-2011 07:22:00] <dhopp> Zenoss core particularly
[25-May-2011 07:26:54] <squig> chances are some one does but they may not be looking at the screen right now
[25-May-2011 07:28:37] <dhopp> yeah I knew there people..was just hoping one of them was here
[25-May-2011 07:31:45] <squig> if you ask the question one of them will probably answer it
[25-May-2011 07:31:58] <squig> ive ran zenoss on systems that big just not at the moment
[25-May-2011 07:40:43] <dhopp> I'm just trying to figure out if it's best to have a completely separate mysql server or if I should have the webinterface/hub/mysql machine that is like a 16 core server and have remote collectors for all of the rrd collection
[25-May-2011 07:56:09] <rocket> dhopp: where is your current bottleneck?
[25-May-2011 07:56:30] <Sam-I-Am> sup rocket
[25-May-2011 07:56:45] <rocket> keeping the dream alive
[25-May-2011 07:57:01] <rocket> dhopp: you probably need more collectors from a rrd io perspective
[25-May-2011 07:57:18] <Sam-I-Am> dhopp: i have about 600 devices and use separate collectors to keep RRDs happy
[25-May-2011 07:58:07] <Sam-I-Am> i have 4 collectors... devices get broken up by organization and importance
[25-May-2011 07:58:29] <Sam-I-Am> rocket: apparently not dreaming if you're up this early
[25-May-2011 07:58:58] <rocket> dhopp: for your reference, with some patches that will be coming out in 3.1.1 we are running 6000 devices at a client .. a 16 core mysql/hub x 3/zope central system with 12 remote collectors at various geographic locations
[25-May-2011 07:59:57] <rocket> jack up the cachesize in the hub.conf etc
[25-May-2011 08:00:19] <rocket> tune the heck out of mysql .. etc
[25-May-2011 08:00:47] <rocket> the biggest issue at the moment is modelling performance, and that should be addressed in 3.1.1
[25-May-2011 08:01:46] <rocket> dhopp: .. however there isnt an upgrade path with this one at the moment
[25-May-2011 08:02:05] <Sam-I-Am> rocket: what is this ?
[25-May-2011 08:02:30] <rocket> wget >; ./ ..
[25-May-2011 08:03:21] <rocket> its a little perl script that analyzes the performance of a mysql instance and makes recommendations
[25-May-2011 08:03:48] <rocket> froztbyte: which microservers?
[25-May-2011 08:03:57] <Sam-I-Am> interesting
[25-May-2011 08:04:26] <rocket> it just lets you know things about your query cache performance etc
[25-May-2011 08:05:09] <froztbyte> rocket: proliant
[25-May-2011 08:05:27] <froztbyte>
[25-May-2011 08:05:31] <dhopp> rocket: don't have a bottleneck…still building the infrastructure out
[25-May-2011 08:05:43] <froztbyte> it's the bare cheapest HP-design style thing you can get, afaik
[25-May-2011 08:05:58] <froztbyte> of course you can get some neat rackmount kit from supermicro
[25-May-2011 08:06:10] <dhopp> rocket: I'm thinking 16 Core, 16GB with 8 disk RAID10 for web/hub/mysql.  8 Core, 12-16GB RAM, 6 DIsk RAID10 for collectors
[25-May-2011 08:06:34] <dhopp> rocket: OS would be installed on separate RAID1
[25-May-2011 08:06:48] <rocket> froztbyte: I am a little spoiled still with the hardware from my last job .. nice big ibm pseries hardware etc ..
[25-May-2011 08:06:55] <Sam-I-Am> mmm pseries
[25-May-2011 08:07:02] <Sam-I-Am> comes with a suit and tie
[25-May-2011 08:07:17] <froztbyte> rocket: yeah, that's the "or better" part
[25-May-2011 08:07:28] <froztbyte> rocket: I've only had a few moments with that line of's pretty
[25-May-2011 08:07:43] <rocket> even the p6 blades where wicked fast ..
[25-May-2011 08:08:36] <rocket> heh .. I even got to play with a 4 cec p5-570 ... it was fun carving that one up ..
[25-May-2011 08:08:43] <Sam-I-Am> lol
[25-May-2011 08:08:49] <Sam-I-Am> i love ibm's virtualization
[25-May-2011 08:09:12] <Sam-I-Am> dont care much for aix, but its become a lot better since 4
[25-May-2011 08:14:07] <dhopp> rocket: yeah I read blog a week or so ago.  But I can't not have an upgrade path
[25-May-2011 08:15:03] <jmp242> Hey does anyone know if gets posted on the wiki or whatever more recently than the 2.5x one from this thread?
[25-May-2011 08:16:48] <dhopp> anyway…guess I should think about getting to the office…be back in an hour or so
[25-May-2011 08:18:02] <jmp242> I only get to play with SystemX servers ... nothing that cool
[25-May-2011 08:18:13] <Sam-I-Am> the one i have is version 92320101
[25-May-2011 08:19:05] <jmp242> hmmm - I have no idea, I don't have a version (well I could DL that one). I was just responding to a forum post and realized that zenfixit is often used for phantom devices, but not actually in the Zenoss install that I see
[25-May-2011 08:19:47] <rocket> its not really a released script.  We use it internally for fixing issues in specific situations
[25-May-2011 08:19:50] <Sam-I-Am> if you read the code theres a version in it
[25-May-2011 08:19:58] <Sam-I-Am> but yeah, what rocket said
[25-May-2011 08:20:03] <jmp242> Oh, I see. lol
[25-May-2011 08:20:07] <rocket> Sam-I-Am: his version is too old to have the version numbers
[25-May-2011 08:20:31] <rocket> there is an internal debate in some cases zenfixit might be causing an issue
[25-May-2011 08:20:36] <jmp242> So - is there a best practices for dealing with the phantom devices that are popping up on the forums?
[25-May-2011 08:20:38] <Sam-I-Am> lol
[25-May-2011 08:20:40] <Sam-I-Am> fight!
[25-May-2011 08:20:41] <rocket> we have really tried to stop using it as much
[25-May-2011 08:20:49] <rocket> zenchkrels -r -x1
[25-May-2011 08:20:56] <rocket> is what we typically recommend first
[25-May-2011 08:20:57] <jmp242> I haven't had that because I don't delete much and I've been lucky I think
[25-May-2011 08:21:23] <Sam-I-Am> rocket: btw, i have someone slaving away at looking at each and every address in those 400+ locations
[25-May-2011 08:21:32] <Sam-I-Am> rocket: someone else e-mailed me about the case asking for an update
[25-May-2011 08:22:07] <rocket> Sam-I-Am: I did make a stink about this particular case ..
[25-May-2011 08:22:07] <jmp242> Off to the FAQ Part 2 . . .
[25-May-2011 08:22:35] <Sam-I-Am> rocket: howso?
[25-May-2011 08:23:07] <rocket> in that it shouldnt be this difficult to figure out the map problem
[25-May-2011 08:23:20] <Sam-I-Am> yep
[25-May-2011 08:23:57] <Sam-I-Am> well, hopefully soonish i'll have every address verified on google maps, and a list of those which google can't pin down exactly
[25-May-2011 08:24:05] <Sam-I-Am> its just... time consuming
[25-May-2011 08:24:31] <Sam-I-Am> once thats done, i can become the poster child for maps not working lol
[25-May-2011 08:25:10] <rocket> I know ..
[25-May-2011 08:26:43] <Sam-I-Am> i really need to get a 3 box going with all those locations
[25-May-2011 08:26:46] <Sam-I-Am> since 2.5.2 is... old
[25-May-2011 08:27:04] <jmp242> An oldie but a goodie Sam-I-Am
[25-May-2011 08:27:14] <Sam-I-Am> at the time it was mo bettah than 3
[25-May-2011 08:27:50] <jmp242> Yea, talking about 3, how's the Spring into bling thing going for updating zenpacks?
[25-May-2011 08:28:34] <jmp242> eventually the non 3.x ones ought to be moved to a legacy page or something
[25-May-2011 08:30:50] <Sam-I-Am> havent heard much
[25-May-2011 08:31:06] <Sam-I-Am> last zenpackfest made for a fairly busy channel
[25-May-2011 08:31:36] <Sam-I-Am> if i had resources/time to run 3 somewhere i'd be working on it
[25-May-2011 08:31:42] <jmp242> Yea. Is the process still to go through nyeats to get an update posted?
[25-May-2011 08:31:54] <Sam-I-Am> probably
[25-May-2011 08:32:02] <Sam-I-Am> as in i havent heard anything different
[25-May-2011 08:32:05] <jmp242> yea, so there might be a backlog then too
[25-May-2011 08:32:16] <jmp242> waiting to be updated.
[25-May-2011 08:35:42] <Hackman238> jmp242: In the next few days I'll post up a handful of community packs I've reworked for v3
[25-May-2011 08:36:24] <Hackman238> jmp242: ...and a bunch I've reworked to eliminate the need to precreate classes, custom schema, etc.
[25-May-2011 08:49:43] kocolosk|away is now known as kocolosk
[25-May-2011 08:58:42] <Hackman238> Always gets quiet when I say something.
[25-May-2011 09:02:40] <jmp242> not always
[25-May-2011 09:23:54] <Sam-I-Am> moo.
[25-May-2011 09:24:03] <rmatte> moo indeed
[25-May-2011 09:25:31] <Sam-I-Am>
[25-May-2011 09:25:37] <Sam-I-Am> from saturday night here
[25-May-2011 09:25:50] <Sam-I-Am> was a little close for comfort, but thats how i get the shots :/
[25-May-2011 09:26:10] <rmatte> nice
[25-May-2011 09:26:15] <Hackman238> Sam-I-Am: Holy cow
[25-May-2011 09:26:19] <dhopp> for 600+ big (disk space wise) could the mysql db grow to?
[25-May-2011 09:26:28] <Sam-I-Am> i'm a photographer in my spare time
[25-May-2011 09:26:31] <Sam-I-Am> and a storm chaser
[25-May-2011 09:26:40] <dhopp> Sam-I-Am: impressive picture
[25-May-2011 09:26:46] <Sam-I-Am> dhopp: depends on your event history
[25-May-2011 09:26:49] <Hackman238> dhopp: It depends on the amount of events your devices generate. I forget, do you use syslog?
[25-May-2011 09:26:52] <rmatte> dhopp: it's not dependant on number of devices, it's dependendant on how noisy they are and on how well you do tuning
[25-May-2011 09:26:57] <Sam-I-Am> syslog is chatty
[25-May-2011 09:27:11] <dhopp> not doing syslog right now, but could possibly in the future
[25-May-2011 09:27:43] <rmatte> either way, the highest any of my Zenoss boxes is for disk space is 20GB, and I haven't had any problems
[25-May-2011 09:27:54] <Sam-I-Am> my db is about 4.6 gigs with 100 days worth of events on 600 or so devices
[25-May-2011 09:27:56] <rmatte> speaking of which, I should zeopack them today
[25-May-2011 09:28:02] <dhopp> so you think ~450GB of space is good enough? :-)
[25-May-2011 09:28:07] <Sam-I-Am> however, not many of those devices are super chatty
[25-May-2011 09:28:08] <rmatte> dhopp: more than
[25-May-2011 09:28:14] <Hackman238> dhopp: plenty
[25-May-2011 09:28:31] <Sam-I-Am> you need 1 million terabytes *pinky on mouth*
[25-May-2011 09:28:43] <rmatte> dhopp: one of the Zenoss boxes with 20GB has around 430 devices on it
[25-May-2011 09:29:06] <Sam-I-Am> for me, mysql wasnt the problem... i/o for rrd files was
[25-May-2011 09:29:08] <rmatte> though I did do a lot of tuning to make sure that only events that I care about make it in to the db
[25-May-2011 09:29:21] <Hackman238> Sam-I-Am: Agreed.
[25-May-2011 09:29:21] <Sam-I-Am> mysql was an issue before i tuned it
[25-May-2011 09:29:30] <dhopp> It's looking like I might be able to get a couple HP 580s and a couple of 380s for the infrastructure.  The 580s I might be able to do a 10 disk RAID10, so now I might put the collectors on them and put mysql/web on the 380s
[25-May-2011 09:29:35] <rmatte> same, mysql blew with the default settings
[25-May-2011 09:29:40] <rmatte> after I tuned it it was great
[25-May-2011 09:29:58] <Sam-I-Am> dhopp: wow, you mean your employer supports your projects? lol
[25-May-2011 09:30:31] <Hackman238> lol
[25-May-2011 09:30:48] <dhopp> Sam-I-Am: well sort of…I tried to get enterprise but that budget got shot down…Luckily we have some 580s and 380s not in use so there isn't a huge up front hardware cost
[25-May-2011 09:30:56] <Sam-I-Am> oh, that helps
[25-May-2011 09:31:08] <kokey> i'm monitoring 580 devices from a virtual machine
[25-May-2011 09:31:11] <kokey> the VM host is a DL380
[25-May-2011 09:31:34] <Sam-I-Am> i have 5 VMs... one for the hub and 4 collectors
[25-May-2011 09:31:52] <dhopp> Sam-I-Am: still have to get plenty of RAM and some extra processors so there is some cost but not the huge outlay of an entire server and shit load of disks
[25-May-2011 09:31:56] <dhopp> kokey: really?
[25-May-2011 09:32:04] <dhopp> kokey: how many data points?
[25-May-2011 09:32:14] <kokey> dhopp: 50,000 RRD files, so roughly that
[25-May-2011 09:32:25] <Sam-I-Am> i find disk i/o is #1, followed by ram on the hub, followed by cpu
[25-May-2011 09:32:28] <kokey> until two days ago, it was set to use 4 virtual CPUs and 6GB of RAM
[25-May-2011 09:32:46] <Sam-I-Am> kokey: you fix your db problem?
[25-May-2011 09:33:04] <kokey> Sam-I-Am: yeah, it fixed itself, i suspect it just neede some reindexing
[25-May-2011 09:33:19] <dhopp> kokey: wow….is the web interface on the same vm?  How many users do you have accessing it?
[25-May-2011 09:33:35] <Sam-I-Am> my web interface is on the hub with the mysql db
[25-May-2011 09:33:39] <Sam-I-Am> and zope db
[25-May-2011 09:33:47] <Sam-I-Am> i have apache reverse proxying it
[25-May-2011 09:34:14] <kokey> dhopp: yup, on the same VM, and mysql and zodb on the same partition
[25-May-2011 09:34:50] <dhopp> I'm confused…everywhere I read makes me think that couldn't be possible
[25-May-2011 09:35:01] <dhopp> that the interface would be exteremely slow
[25-May-2011 09:35:04] <Sam-I-Am> its magically delicious
[25-May-2011 09:35:07] <dhopp> lol
[25-May-2011 09:35:15] <Sam-I-Am> theres not a LOT of i/o on the hub
[25-May-2011 09:35:19] <Sam-I-Am> just ram
[25-May-2011 09:35:39] <Sam-I-Am> having any amount of rrd files on the hub kills the box
[25-May-2011 09:35:45] <Hackman238> dhopp: It can be virtualized. The hardest part is disk i/o i a vm for collectors
[25-May-2011 09:36:02] <Hackman238> *in a vm
[25-May-2011 09:36:23] <kokey> dhopp: i think vmware is using storage over NFS to netapp filers
[25-May-2011 09:36:26] <Sam-I-Am> my place cheaped out on san storage, so i wound up having to use local storage on each VM box
[25-May-2011 09:36:26] <dhopp> Hackman238: I didn't mean the virtualization..but having hub/mysql/and 50,000 rrds on one machine
[25-May-2011 09:36:32] <kokey> dhopp: that might be what's helping
[25-May-2011 09:36:44] <Sam-I-Am> i killed the san storage
[25-May-2011 09:37:01] <Hackman238> dhopp: Oh- you should seperate collection from the other processes. Youll need it for scalability
[25-May-2011 09:37:03] <kokey> dhopp: i only have about 15 people max accessing the interface, it gets slow, i found throwing more RAM and increasing the zope cache helps
[25-May-2011 09:37:27] <dhopp> kokey: I will probably average 25 people and peak out at ~50-60
[25-May-2011 09:37:36] <kokey> dhopp: but if interface speed is important i figure getting a ton of RAM and doing the multiple zope thing behind a load balancer is the way to go
[25-May-2011 09:37:52] <Sam-I-Am> hmmm for that many folks i would consider what kokey said
[25-May-2011 09:37:53] <Hackman238> dhopp: I agree with kokey
[25-May-2011 09:37:56] <Sam-I-Am> i have about 10-15 usually
[25-May-2011 09:38:02] <dhopp> kokey: you mean a hardware load balancer..or just apache load balancing?
[25-May-2011 09:38:04] <Sam-I-Am> and one instance seems to hold up
[25-May-2011 09:38:12] <Hackman238> dhopp: apache loadbalacing
[25-May-2011 09:38:28] <kokey> dhopp: apache load balancing should do fine
[25-May-2011 09:38:43] <kokey> not much point in getting all fancy with hw load balancing
[25-May-2011 09:38:59] * Sam-I-Am needs a better test bed for systems stuff
[25-May-2011 09:39:04] <Sam-I-Am> i have all the network gear i can ever want
[25-May-2011 09:39:06] <Sam-I-Am> but no servers
[25-May-2011 09:39:20] <dhopp> Hackman: whew…if I asked for hardware load balancing for a monitoring system I probably would have been looked at like I had 2 heads
[25-May-2011 09:39:33] <kokey> Sam-I-Am: haha, happens often in companies, network and sysadmins are different teams
[25-May-2011 09:39:35] <Hackman238>
[25-May-2011 09:39:39] <kokey> kind of why i like appliances
[25-May-2011 09:39:56] <Hackman238> Video related. :-)
[25-May-2011 09:39:57] <Sam-I-Am> Hackman238: do you guys use things like mpls or openflow?
[25-May-2011 09:40:07] <kokey> hmmmm
[25-May-2011 09:40:09] <kokey> that gives me an idea
[25-May-2011 09:40:21] <kokey> zenoss network monitoring is pretty good
[25-May-2011 09:40:32] <Sam-I-Am> yeah, once you get all the snmp stuff right
[25-May-2011 09:40:33] <kokey> would be nice to make an appliance for it, for network teams in mind in particular
[25-May-2011 09:40:45] <Sam-I-Am> i have custom templates for each kind of device here
[25-May-2011 09:40:51] <Hackman238> Sam-I-Am: mpls, yes
[25-May-2011 09:41:03] <Sam-I-Am> Hackman238: mmm... vpls, l2vpns, and stuff?
[25-May-2011 09:41:21] <kokey> yeah, perhaps focus on the network device stuff, make sure the models from the various vendors are well supported
[25-May-2011 09:41:37] <kokey> and stick it on an appliance box, with some glue for 'firmware upgrades' and that sort of thing
[25-May-2011 09:41:45] <kokey> and focus on selling it to network teams
[25-May-2011 09:42:07] <Sam-I-Am> its difficult to make templates which support everything... especially with vendors that dont streamline snmp (like cisco and brocade... well, anyone but juniper)
[25-May-2011 09:42:12] <Hackman238> Sam-I-Am: yes
[25-May-2011 09:42:53] <Hackman238> kokey: It'll become your full time job to keep it up to date and compatible :-)
[25-May-2011 09:43:14] <Sam-I-Am> i collect the basics for each model of device i have... which is maybe about 15 different things... and it took me a month or two
[25-May-2011 09:43:27] <kokey> Hackman238: yup, for sure, throw some PS in there while you're at it
[25-May-2011 09:44:02] <Sam-I-Am> kokey: lets go make a company
[25-May-2011 09:44:07] <kokey> heh
[25-May-2011 09:44:15] <kokey> actually there's certainly business to be had here in the UK
[25-May-2011 09:44:17] <Sam-I-Am> kinda hoping my next job is zenossy
[25-May-2011 09:44:28] <Sam-I-Am> oh thats right, you're over the pond
[25-May-2011 09:44:47] <dhopp> Sam-I-Am: International company baby!
[25-May-2011 09:45:03] <Sam-I-Am> yeah, maybe we all need to get together
[25-May-2011 09:45:14] <Sam-I-Am> although i'm sure theres a number of zenoss consulting companies
[25-May-2011 09:45:23] <Hackman238> Come to SA....its only 101 degrees out.
[25-May-2011 09:45:25] <Hackman238> LOL
[25-May-2011 09:45:33] <Sam-I-Am> whats the humidity though?
[25-May-2011 09:45:43] <dhopp> It's like 50 and raining in Chicago :-)
[25-May-2011 09:45:54] <Hackman238> Not high enough to rain, but high enough to be awful
[25-May-2011 09:46:12] <dhopp> Although I love SA…was there a few years ago…need to get back thre
[25-May-2011 09:46:13] <dhopp> there
[25-May-2011 09:46:17] <Sam-I-Am> wow, 78/72 now... going to be a hot one when it warms up
[25-May-2011 09:46:33] <Sam-I-Am> the dry air is only about 75 miles west of you
[25-May-2011 09:46:47] <Hackman238> Sam-I-Am: Yep. Thats where the cheap land is too.
[25-May-2011 09:46:54] <Hackman238> Sam-I-Am: I should move out that way LOL
[25-May-2011 09:47:12] <Sam-I-Am> i just need a mobile home and i can drive to the dryline
[25-May-2011 09:47:22] <Sam-I-Am> go east during the day, back west at night
[25-May-2011 09:47:28] <Sam-I-Am> and... happen to chase thunderstorms along it
[25-May-2011 09:47:33] <Hackman238> Sam-I-Am: LOL!
[25-May-2011 09:48:02] <dhopp> Hackman238: For 50 simultaneous users, how many zopes do you think are needed?
[25-May-2011 09:48:04] <Sam-I-Am> what a job
[25-May-2011 09:48:06] <dhopp> 5?
[25-May-2011 09:48:07] <rmatte> eugh, is it payday yet?
[25-May-2011 09:48:07] <dhopp> 10?
[25-May-2011 09:48:13] <Sam-I-Am> rmatte: sooooon
[25-May-2011 09:48:17] <Sam-I-Am> rmatte: gotta buy booze?
[25-May-2011 09:48:18] <Hackman238> dhopp: 3 or 4
[25-May-2011 09:48:28] <rmatte> no, I just need money, digging slightly in to my overdraft lol
[25-May-2011 09:48:47] <Hackman238> rmatte: Oh man. tisk tisk.
[25-May-2011 09:48:53] <rmatte> I've been putting $2000/month in to RRSPs lately, so I don't have the kind of cash that I used to
[25-May-2011 09:48:54] <rmatte>
[25-May-2011 09:49:20] <Hackman238> rmatte: Reminds me- you forgot to snd your resume
[25-May-2011 09:49:30] <Sam-I-Am> uh ohs
[25-May-2011 09:49:33] <rmatte> I haven't forgotten, just haven't had time lol
[25-May-2011 09:49:40] * Sam-I-Am gets his in before rmatte
[25-May-2011 09:49:42] <Hackman238> rmatte: Oh LOL np man
[25-May-2011 09:49:58] <rmatte> I have an email sitting in my personal mailbox that says "Send updated resume to..."
[25-May-2011 09:49:59] <rmatte>
[25-May-2011 09:50:08] <rmatte> (the way I remind myself of things)
[25-May-2011 09:50:13] <Sam-I-Am> rmatte: whatchoo applying for?
[25-May-2011 09:50:46] <Hackman238> rmatte: Thats how I do it. I get more mail from myself than anyone other than Zenoss :-)
[25-May-2011 09:51:14] <Hackman238> Sam-I-Am: Lesser overlord of RackSpace Zenoss
[25-May-2011 09:51:21] <Hackman238> Sam-I-Am: LOL J/K
[25-May-2011 09:52:00] <Sam-I-Am> Hackman238: if i sent you something, could you get it to the right place?
[25-May-2011 09:52:12] <Hackman238> Sam-I-Am: Yes.
[25-May-2011 09:52:19] <rmatte> We really just spoke about it in passing... I won't be serious about it unless the pay knocks my socks off lol.
[25-May-2011 09:53:06] <rmatte> The whole having to move thing would suck
[25-May-2011 09:53:18] <dhopp> moving sucks
[25-May-2011 09:53:21] <dhopp> I hate it
[25-May-2011 09:53:33] <dhopp> last time I did it though..I paid a moving company…best $900 I ever spent
[25-May-2011 09:53:45] <rmatte> dhopp: well, in my case it's not just moving between states, it's moving between countries... lots of red tape involved
[25-May-2011 09:53:58] <rmatte> and I hate paperwork
[25-May-2011 09:53:59] <dhopp> rmatte: that will probably cost more then $900 :-P
[25-May-2011 09:54:10] <dhopp> rmatte: where are you at?
[25-May-2011 09:54:13] <rmatte> well, rackspace have relocation services
[25-May-2011 09:54:17] <Hackman238> If RackSpace wants you, we'll move you here.
[25-May-2011 09:54:20] <rmatte> so it wouldn't cost me anything in that regard
[25-May-2011 09:54:37] <rmatte> dhopp: I'm currentling living in Ottawa, Canada
[25-May-2011 09:54:42] <rmatte> currently*
[25-May-2011 09:55:24] <rmatte> RackSpace should setup shop here, then they can hire me, lol
[25-May-2011 09:55:30] <Hackman238> rmatte: LOL
[25-May-2011 09:55:42] <Sam-I-Am> Hackman238: what format do you like resumes in?
[25-May-2011 09:55:58] <Hackman238> Sam-I-Am: Digital ;-)
[25-May-2011 09:56:10] <rmatte> Sam-I-Am: encrypt it to give them a challenge
[25-May-2011 09:56:18] <Hackman238> Sam-I-Am: Whatever format is good. OpenOffice opens all.
[25-May-2011 09:56:32] <rmatte> except for office xml docs
[25-May-2011 09:56:38] <rmatte> unless they recently added support
[25-May-2011 09:56:55] <Hackman238> rmatte: Its in there :-)
[25-May-2011 09:57:01] <rmatte> ah, must be new
[25-May-2011 09:57:16] <Hackman238> rmatte: yep.
[25-May-2011 09:57:21] <rmatte> that drove me nuts
[25-May-2011 09:57:22] <Sam-I-Am> rmatte: or they'll just say "whatever" and move  on:/
[25-May-2011 09:57:28] <rmatte> "Could you please resend your docx as a doc?"
[25-May-2011 09:57:37] <dhopp> sometimes I think openoffice opens MS docs better the MS Office
[25-May-2011 09:57:47] <rmatte> hehe
[25-May-2011 09:57:53] <Hackman238> dhopp: true true- except for certain types of tables
[25-May-2011 09:57:58] <jmp242> Ehh, PDF for me if I want it to be readable
[25-May-2011 09:58:03] <jmp242> or plain text
[25-May-2011 09:58:15] <jmp242> MS Office bit us so hard between versions recently
[25-May-2011 09:58:18] <dhopp> I work on a Mac and even MS Office 2011…if you use Excel to save a csv file they still save it with the stupid windows CRLF characters
[25-May-2011 09:58:21] <jmp242> major reports were moved to LaTeX
[25-May-2011 09:58:25] <rmatte> so it turns out that my coding project from yesterday isn't done yet... feature requests...
[25-May-2011 09:58:28] <rmatte> curse them
[25-May-2011 09:58:30] <Sam-I-Am> Hackman238: do your conform to rfc2549?
[25-May-2011 09:58:42] * Sam-I-Am loves latex
[25-May-2011 09:58:47] <Hackman238> I do resume this way:
[25-May-2011 09:58:52] <dhopp> You wrote the damn product for OS X which is a *nix OS..please save the files in a *nix friendly format..not windows...
[25-May-2011 09:58:56] <dhopp> stupid MS
[25-May-2011 09:59:00] <Sam-I-Am> thats how i originally did my resume since word kept scrwing up the formatting
[25-May-2011 09:59:33] <jmp242> I even find that by faking it with LyX it's very like using a Wiki, but the print output is better
[25-May-2011 10:00:00] <Sam-I-Am> maybe i should html mine
[25-May-2011 10:00:29] <Hackman238> Sam-I-Am: I'm not familliar with rfc2549- what is it?
[25-May-2011 10:00:38] <rmatte> I should make a web based resume for myself
[25-May-2011 10:00:55] <jmp242> I don't really want my resume that public
[25-May-2011 10:01:15] <jmp242> I find plain text into the system used by various people works OK for me
[25-May-2011 10:01:37] <rmatte> public is better
[25-May-2011 10:01:37] <Hackman238> jmp242: I do a lot of contract business so I advertise that way
[25-May-2011 10:01:43] <jmp242> I think one great thing IMO about digital is much of the formatting bunk went out the window for resumes
[25-May-2011 10:01:54] <jmp242> to the great appreciation of techies like myself anyway
[25-May-2011 10:01:54] <xuru> Does anyone know a good site/channel/anything to help setup fiber channel multipath on ubuntu?
[25-May-2011 10:02:56] <Sam-I-Am> Hackman238: ip over carrier pigeon
[25-May-2011 10:03:23] <Hackman238> jmp242: Whats your reasoning?
[25-May-2011 10:03:23] <Sam-I-Am> i'm debating whether a public resume is a good thing
[25-May-2011 10:03:29] <Sam-I-Am> might be good for recruiters and contract jobs
[25-May-2011 10:03:33] <Hackman238> Sam-I-Am: Ah LOL
[25-May-2011 10:03:39] <Sam-I-Am> also, posting it on someplace like dice... but i suspect i'd get a lot of junk
[25-May-2011 10:03:42] <jmp242> my reasoning on what exactly?
[25-May-2011 10:03:52] <Hackman238> jmp242: Non-public resume, sorry.
[25-May-2011 10:04:00] <jmp242> oh, I lost the thread sorry
[25-May-2011 10:04:11] <jmp242> I don't necessarily want recruiters calling me or the SPAM
[25-May-2011 10:04:42] <kokey> Sam-I-Am: maybe just fill in linked in, many places hunt on there
[25-May-2011 10:04:49] <jmp242> If I'm not looking for work, it ranges from a short waste of time to a long waste of time
[25-May-2011 10:04:50] <Hackman238> jmp242: I hear that reasoning often, but I dont get such calls, and rarely spam.
[25-May-2011 10:04:57] <kokey> i get calls daily from recruiters
[25-May-2011 10:05:01] <kokey> but i'm a contractor
[25-May-2011 10:05:03] <Hackman238> jmp242: true
[25-May-2011 10:05:03] <kokey> so it's a good thing
[25-May-2011 10:05:20] <kokey> i tell them when my next contract is finishing so their database will flag me when the time comes
[25-May-2011 10:05:21] <Hackman238> kokey: daily? wow.
[25-May-2011 10:05:31] <Simon4> Hackman238: they're keen over here
[25-May-2011 10:05:34] <kokey> Hackman238: at least once per day
[25-May-2011 10:05:56] <Sam-I-Am> grrr, resume has one line leaking over to another page
[25-May-2011 10:05:59] <Sam-I-Am> (in the word version)
[25-May-2011 10:06:16] <Simon4> Sam-I-Am: txt-speak
[25-May-2011 10:06:29] <Hackman238> Sam-I-Am: No sweat
[25-May-2011 10:06:46] <Hackman238> Simon4: I dont understand.
[25-May-2011 10:07:13] <Simon4> Hackman238: sorry, was talking about recruiters here in ze UK
[25-May-2011 10:07:36] <Simon4> they may not have anything useful/interesting, but they call anyway
[25-May-2011 10:07:46] <Hackman238> Simon4: Oh gotcha. I've had a few that bugged me a bunch- I just add them to the call blacklist.
[25-May-2011 10:07:55] <Sam-I-Am> Simon4: yeah, that'd be one awesome resume
[25-May-2011 10:08:05] <Simon4> Sam-I-Am:
[25-May-2011 10:10:24] <Hackman238> Sam-I-Am: email it to
[25-May-2011 10:10:38] <Hackman238> Sam-I-Am: Rather,
[25-May-2011 10:11:05] * Sam-I-Am shudders at the idea of sending a resume using word
[25-May-2011 10:11:21] <Sam-I-Am> but... its the most updated since the last recruiter wanted it
[25-May-2011 10:11:26] <Sam-I-Am> i had to take a few showers
[25-May-2011 10:11:34] <Hackman238> Sam-I-Am: Export as PDF if you want.
[25-May-2011 10:12:18] <Hackman238> Sam-I-Am: What? Should always shower. LOL
[25-May-2011 10:14:48] <jmp242> My biggest problem isn't the quick ignores
[25-May-2011 10:15:08] <jmp242> it's the "I should seriously consider this", but I'm not a contractor really
[25-May-2011 10:15:14] <Sam-I-Am> Hackman238: geeks shower?
[25-May-2011 10:15:16] <jmp242> that usually takes a couple weeks to play out
[25-May-2011 10:15:45] <Hackman238> jmp242: Oh gotcha.
[25-May-2011 10:16:13] <Hackman238> Sam-I-Am: Well, I cant speak for everyone, but I do. LOL
[25-May-2011 10:16:26] <Sam-I-Am> Hackman238: on its way
[25-May-2011 10:16:41] <Sam-I-Am> Hackman238: using certain apps or operating systems make me feel more obligated to shower
[25-May-2011 10:17:03] <Hackman238> Sam-I-Am: Oh gotcha. LOL
[25-May-2011 10:34:38] <dhopp> does zenbackup interrupt zenoss operation?
[25-May-2011 10:35:02] <Sam-I-Am> dhopp: it will lock the db while it does a mysqldump
[25-May-2011 10:35:11] <Sam-I-Am> dhopp: but hopefully not long enough to fry zenoss
[25-May-2011 10:35:30] <Sam-I-Am> i had a problem a few months ago with mysqldump taking 15 minutes and it hosed zenoss in weird ways
[25-May-2011 10:35:54] <dhopp> Sam-I-Am: if I backup mysql separately, what kind of implications does that have during a restore?
[25-May-2011 10:36:30] <Sam-I-Am> you;d just manually need to restore mysql. maybe comment out the jazz in zenbackup
[25-May-2011 10:36:36] <dhopp> Sam-I-Am: I'm thinking to backup mysql on a replicated server…so I lose replication during the backup, but that shouldn't affect zenoss operations…so then I just tell zenbackup to not do mysql
[25-May-2011 10:36:42] <Sam-I-Am> someone here recommended replicating to another DB and backing that one up
[25-May-2011 10:37:19] <dhopp> well I don't "lose" the will catchup once it's done
[25-May-2011 11:00:25] <Sam-I-Am> yeh
[25-May-2011 11:07:10] <mloven> Morning all.   I've got one of those questions that I think I _should_ know the answer to, and yet I still seem to be having issues...  ...
[25-May-2011 11:07:26] <Sam-I-Am> it happens
[25-May-2011 11:07:29] <Sam-I-Am> try more caffeine
[25-May-2011 11:07:31] <Hackman238> Is it 42?
[25-May-2011 11:08:34] <mloven> Caffeine only goes in so fast...     I'm trying to create a threshold that calculates a percentage of a particular data point (that I created), but I'm getting an error.   Here's the particulars:
[25-May-2011 11:10:33] <mloven> I have a data point called fgVdNumber that shows the number of licenses in use on a device.  I also have a data point called fgVdMaxVdoms that returns the maximum number of licenses available on a device.   I'd like to have a threshold that trips when fgVdNumber reaches, say, 90% of fgVdMaxVdoms
[25-May-2011 11:11:17] <Sam-I-Am> i used custom CFs to standardize a lot of things as percentages
[25-May-2011 11:11:23] <Sam-I-Am> then base the thresholds on those
[25-May-2011 11:11:45] <Sam-I-Am> typically for routers which return memory information in various ways, i wanted to standardize
[25-May-2011 11:13:02] <mloven> When I put fgVdMaxVdoms_fgVdMaxVdoms * .90 into the "max" field of the threshold against the fgVdNumber_fgVdNumber data point, I get an error stating "User-supplied Python expression (fgVdMaxVdoms_fgVdMaxVdoms * .90) for maximum value caused error: ['fgVdNumber_fgVdNumber']
[25-May-2011 11:13:20] <mloven> and no graphs show on the perf tab for the device.
[25-May-2011 11:13:46] <mloven> If I change the "max" value to a number, say "90", it works fine and all graphs show as expected.
[25-May-2011 11:14:03] <Hackman238> mloven: try 0.90
[25-May-2011 11:15:20] <Hackman238> mloven: or try fgVdMaxVdoms_fgVdMaxVdoms * 9 / 10
[25-May-2011 11:15:57] <mloven> Hackman238: same result in each case
[25-May-2011 11:16:21] <mloven> Hackman238: with the error message changing to reflect the new equation, of course.
[25-May-2011 11:16:22] <Hackman238> mloven: fgVdMaxVdoms_fgVdMaxVdoms * 90 works, however?
[25-May-2011 11:17:06] <Hackman238> mloven: or do you mean just a 90
[25-May-2011 11:17:11] <mloven> Hackman238: no, what I meant was, if I just put a number in the "max" field, with no other calculation, it works
[25-May-2011 11:17:17] <Hackman238> Hum.
[25-May-2011 11:18:51] <Hackman238> Can you post the graph definition?
[25-May-2011 11:20:22] <mloven> sure.
[25-May-2011 11:24:58] <nyeates> Hi Mike! long time no see
[25-May-2011 11:25:07] <mloven> Hi Nick!
[25-May-2011 11:26:26] <nyeates> You guys still plugging away with zenoss I see... everything going good?
[25-May-2011 11:26:40] <Hackman238> mloven: Have you printed fgVdNumber to be sure its being populated?
[25-May-2011 11:27:11] <mloven> Hackman238: yeah, if I remove the threshold, both data points populate just fine.
[25-May-2011 11:28:23] <mloven> nyeates: absolutely.  we're still waiting on some zenpacks to be 3.0 friendly before we do our big move to 3.x, but our 2.x systems keep chugging along.
[25-May-2011 11:29:50] <Hackman238> mloven: I'm not sure, to be honest.
[25-May-2011 11:30:34] <Hackman238> mloven: I'd very much like to know. Anyone else have any ideas?
[25-May-2011 11:30:38] <nyeates> mloven: maybe you could get more detail on the error from a log file
[25-May-2011 11:30:48] <mloven> Hackman238:  well, there's some comfort in knowing it wasn't just some obvious user error on my part. 
[25-May-2011 11:31:06] <Sam-I-Am> mloven: did you try making a cdef which calculates percentage for you, then checking the threshold against it?
[25-May-2011 11:33:00] <Sam-I-Am> err, nevermind thresholds dont look at graphs :/
[25-May-2011 11:33:11] <nyeates> its weird that the error references the actual datapoint you are setting the threshold on.... or does it always do that? I cant recall
[25-May-2011 11:33:40] <Sam-I-Am> i'd love for thresholds to work on graph points
[25-May-2011 11:34:47] <mloven> nyeates: a "grep 'fgVdMaxdoms' *.log shows multiple entries in the zenhub.log file, but they all say the same thing as the error I posted....
[25-May-2011 11:35:17] <Sam-I-Am> i'm doing stuff like this...
[25-May-2011 11:35:17] <Sam-I-Am>
[25-May-2011 11:39:11] <Hackman238> Sam-I-Am: Can you setup a mock threshold?
[25-May-2011 11:42:43] <Sam-I-Am> howso?
[25-May-2011 11:42:54] <Sam-I-Am> er, in what sense
[25-May-2011 11:43:58] <Hackman238> Sam-I-Am: add a thresh and set max field to Slot_1_Used_Percent
[25-May-2011 11:44:14] <Hackman238> Sam-I-Am: Just to see if it'll kick about it
[25-May-2011 11:44:22] <Sam-I-Am> doesnt it only let you use one from the list?
[25-May-2011 11:45:17] <Hackman238> Sam-I-Am: I'm not positive- I have no thresholds that refer to graph points using min/max
[25-May-2011 11:45:50] <Sam-I-Am> awesome, history is timing out.
[25-May-2011 11:47:50] <Sam-I-Am> looks like someone has some nasty sql query going
[25-May-2011 11:47:58] <Sam-I-Am> theres a handful which still grind things to a halt
[25-May-2011 11:47:59] <Hackman238> Sam-I-Am: I hate it when that happens.
[25-May-2011 11:48:25] <Sam-I-Am> you guys still have that with all those tweaks?
[25-May-2011 11:49:16] <Hackman238> Sam-I-Am: We use multi zope, so in the case of a huge sql query, only one instace is tied up. Users are diverted by apache to a less busy session.
[25-May-2011 11:49:35] <Hackman238> Sam-I-Am: Besides that we cache the entire events database
[25-May-2011 11:49:58] <Sam-I-Am> going on 1-2 minutes for whatevers going on
[25-May-2011 11:50:34] <Sam-I-Am> must be nice to have the hardware resources
[25-May-2011 11:51:06] <Sam-I-Am> around these parts i'm given a list of required features and performance and no budget for hardware to make it 100% successful
[25-May-2011 11:51:39] <rmatte> Sam-I-Am: my problem isn't hardware, my problem is lack of time
[25-May-2011 11:52:04] <rmatte> I'm one person managing more than half of the tools for a NOC
[25-May-2011 11:52:13] <rmatte> managing/developing I should say
[25-May-2011 11:52:18] <Sam-I-Am> well, theres always that
[25-May-2011 11:52:22] <rmatte> lol
[25-May-2011 11:52:22] <Sam-I-Am> zenoss is not my main job
[25-May-2011 11:52:29] <Sam-I-Am> and i'm the only guy who works on it
[25-May-2011 11:53:26] <rmatte> my main job is zenoss/otrs/ziptie/wiki/dns/ldap/web portal/web proxy/random script development & maintenance/linux professional services work for clients/whatever else
[25-May-2011 11:53:44] <Sam-I-Am> just add -bitch to the end of that
[25-May-2011 11:53:49] <rmatte> pretty much
[25-May-2011 11:53:50] <Sam-I-Am> but you're getting paid the big bucks
[25-May-2011 11:53:51] <Hackman238> Sam-I-Am: LOL
[25-May-2011 11:54:14] <rmatte> yeh, I'm making decent money for what I do, though I could probably be making the same money with less of a workload somewhere else
[25-May-2011 11:54:27] <Hackman238> *cough* yep.
[25-May-2011 11:54:28] <fragfutter> rmatte: is it fun?
[25-May-2011 11:54:32] <rmatte> well, maybe not less of a workload, but less of a backlog
[25-May-2011 11:54:41] <rmatte> fragfutter: I enjoy it for the most part
[25-May-2011 11:54:43] <fragfutter> rmatte: does it pay the rent?
[25-May-2011 11:54:49] <rmatte> fragfutter: more than
[25-May-2011 11:54:50] <rmatte> lol
[25-May-2011 11:54:55] <Hackman238> lol
[25-May-2011 11:55:09] <Sam-I-Am> its interesting how some event history searches break while others work... same time frame, just a different event class
[25-May-2011 11:55:10] <fragfutter> rmatte: stick with it. pastures are always greener somewhere else.
[25-May-2011 11:55:30] <rmatte> Yeh, I don't plan to leave unless I get some explosive offer
[25-May-2011 11:55:34] <Hackman238> Sam-I-Am: theres a few items at work there
[25-May-2011 11:56:06] <mloven> you hear that Zenoss?  rmatte needs an explosive offer.....
[25-May-2011 11:56:08] <mloven>
[25-May-2011 11:56:17] <Sam-I-Am> we all need explosive offers
[25-May-2011 11:56:32] <Hackman238> Detonation is a requirement
[25-May-2011 11:56:34] <Hackman238> LOL
[25-May-2011 11:56:55] <rmatte> extreme detonation
[25-May-2011 11:57:37] <subbu005> Hi All I want to monitor the swap on the linux box and it should trigger the alert if swap is used above 75% .For that i gave the threhsold as "here.os.totalSwap / 1000 * 0.75"
[25-May-2011 11:57:38] <Hackman238> rmatte: Get me that resume and I'll see what I can do.
[25-May-2011 11:57:45] <rmatte> god I hate excel... let's open a CSV file but not auto-expand the columns... it's not like you want to actually see that data or anything
[25-May-2011 11:58:01] <subbu005> is it the correct expression which i have given
[25-May-2011 11:58:06] <Hackman238> rmatte: pfff...all those numbers are hard to look at. LOL
[25-May-2011 11:58:28] <rmatte> well, not number in this case, switch names, port names, port descriptions, etc...
[25-May-2011 11:59:12] <rmatte> openoffice on the other hand handles it beautifully
[25-May-2011 11:59:18] <Hackman238> subbu005: Why 1000 and not 1024?
[25-May-2011 11:59:32] <Hackman238> rmatte: Yep. Business app. :-)
[25-May-2011 11:59:40] <rmatte>
[25-May-2011 12:00:01] [disconnected at Wed May 25 12:00:01 2011]
[25-May-2011 12:00:02] [connected at Wed May 25 12:00:02 2011]
[25-May-2011 12:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[25-May-2011 12:00:57] <Hackman238> fragfutter: for the filesystem though?
[25-May-2011 12:02:19] <Hackman238> fragfutter: I mean I know the disks are measured in capacity by the manu as base 1000, but aretn most filesystems base 1024?
[25-May-2011 12:03:18] <fragfutter> Hackman238: i was just kidding. Manufactures use the base 10 numbers so disks sound bigger. it uses base 2.
[25-May-2011 12:03:31] <subbu005> @Hackman23:I tought to change it to 1024 but in some of the zenoss forum its says 1000
[25-May-2011 12:04:08] <Sam-I-Am> subbu005: if you can figure out thresholds with calculations, i think that'll also help mloven
[25-May-2011 12:05:41] <Hackman238> fragfutter: Oh gotcha. Thx for the right bases- I wasnt thinking :-)
[25-May-2011 12:05:45] <fragfutter> i have low swap threshold with minumum value "here.os.totalSwap /1000 * 0.80"
[25-May-2011 12:06:06] <fragfutter> against memAvailSwap_memAvailSwap
[25-May-2011 12:06:21] <kokey> ugh
[25-May-2011 12:06:35] <kokey> so to move/add a device to a group, addDeviceGroup is right?
[25-May-2011 12:06:42] <kokey> but now I'm wondering how to remove it from a group
[25-May-2011 12:07:46] <subbu005> @fragfutter:Can i use the above expression and make change as .75
[25-May-2011 12:07:51] <fragfutter> sure
[25-May-2011 12:08:39] <fragfutter> it means that you will trip it, as soon as 25% of the swap are in use (-> less then 75% available)
[25-May-2011 12:08:39] <Hackman238> kokey: Can use setGroups()
[25-May-2011 12:08:59] <kokey> Hackman238: ah, so setGroups is better than addDeviceGroup?
[25-May-2011 12:09:04] <kokey> or setGroups is good for removing?
[25-May-2011 12:09:16] <Hackman238> kokey: get the groups, figure what you want to remove, use setGroups method on device object to set new list
[25-May-2011 12:09:37] <subbu005> @fragfutter:Just wondering Can u please send me the transform which you are using for the same
[25-May-2011 12:09:47] <Hackman238> kokey: it sets a supplied list, so you can add or remove
[25-May-2011 12:10:17] <kokey> Hackman238: ok, so if it's in /GroupA and /GroupB and I want to remove /GroupB it's something like d.setGroups('/GroupA')
[25-May-2011 12:10:35] <fragfutter> subbu005: which transform? its a threshold. Threshold will trigger.
[25-May-2011 12:10:39] <Hackman238> kokey: yep
[25-May-2011 12:10:40] <subbu005> @Fragfutter:I want the zenoss to trgigger the swap alerts in tems of Percentage but its triggering in terms of value
[25-May-2011 12:10:50] <kokey> a longer list would be d.setGroups('/GroupA', '/GroupB')?
[25-May-2011 12:11:28] <rmatte> subbu005: that's because the swap threshold configured in the default snmp linux template is garbage
[25-May-2011 12:11:32] <Hackman238> kokey: if its more than one group, create a list obj with all the groups and setGroups(list)
[25-May-2011 12:11:45] <subbu005> @Fragfuuter:The transform which displays in terms of Percntages
[25-May-2011 12:12:01] <kokey> Hackman238: ah ok
[25-May-2011 12:12:37] <fragfutter> subbu005: it triggers with 80% for me. That the message contains only bytes doesn't matter so i have no transformation rule.
[25-May-2011 12:12:37] <subbu005> @rmatte:Yeah Excatly the NOC is Confused with this I request to send the transform so that We will be happy
[25-May-2011 12:12:42] <kokey> Hackman238: I guess that means d.setGroups(['/GroupA', '/GroupB'])
[25-May-2011 12:13:00] <Hackman238> kokey: Can do that way also
[25-May-2011 12:13:01] <subbu005> @gragfuuter:ohh ok
[25-May-2011 12:13:05] <kokey> cool, seems to work well
[25-May-2011 12:13:09] <kokey> Hackman238: thanks
[25-May-2011 12:13:14] <Hackman238> kokey: no sweat.
[25-May-2011 12:13:20] <rmatte> subbu005: do you mean it's triggering on the percentage but not displaying as a percentage?
[25-May-2011 12:13:31] <rmatte> subbu005: or that it's triggering on the byte value?
[25-May-2011 12:13:36] <subbu005> @rmatte:Yeah
[25-May-2011 12:14:00] <rmatte> yeah which>?
[25-May-2011 12:14:13] <fragfutter> rmatte: that formula will trigger it on a percentage value of the total available swap. But the event will say something like more then 12345bytes used.
[25-May-2011 12:14:19] <subbu005> theshold is in Percentage and in the description we are getting in bytes . we want to change in the description in terms of %
[25-May-2011 12:15:14] <subbu005> @fragfutter:As per you if it triggers in terms of valeus it confuses the people
[25-May-2011 12:15:44] <subbu005> I know Rmatt will figure out this
[25-May-2011 12:15:54] <rmatte> subbu005: you need to write a transform to do the conversion
[25-May-2011 12:16:10] <rmatte> I'll give you an example
[25-May-2011 12:16:28] <subbu005> sure
[25-May-2011 12:16:41] <rmatte>
[25-May-2011 12:16:47] <rmatte> look at the transform on that page
[25-May-2011 12:16:58] <rmatte> it does that for the threshold in that pack
[25-May-2011 12:17:02] <rmatte> you need to write something similar
[25-May-2011 12:17:56] <subbu005> I will try it now if its not working Please help me out
[25-May-2011 12:18:13] <rmatte> well, I'm working on coding something right now, so I don't have time to guide you step by step
[25-May-2011 12:19:37] <subbu005> @rmatte:Sure i will not distrub you will get back to u some other tine
[25-May-2011 12:19:39] <subbu005>
[25-May-2011 12:22:18] <rmatte> to get the swap value you'd want to do dev.os.totalSwap
[25-May-2011 12:22:27] <rmatte> for the total
[25-May-2011 12:22:38] <rmatte> then the actual usage would be dependant on what the datasource is called
[25-May-2011 12:22:48] <rmatte> dev.getRRDValue('Whatever')
[25-May-2011 12:25:38] <dhopp> For Linux servers did you guys switch to raw snmp counters or still using the stock ones?  I can see that Wait is a raw counter and the other ones are not could make reading the graph very confusing…
[25-May-2011 12:25:55] <dhopp> however, to really get use out of raw counters you would have to take into consideration the number of CPUs
[25-May-2011 12:27:52] <rmatte> I just use whatever the stock template has
[25-May-2011 12:28:31] <kokey> I guess I should do a reindex after doing d.setGroups or d.addDeviceGroup
[25-May-2011 12:28:49] <rmatte> you could but I doubt it's necessary
[25-May-2011 12:29:11] <rmatte> you need to do a commit though
[25-May-2011 12:29:14] <kokey> well under the bonnet the change is made, just not showing in the interface
[25-May-2011 12:29:20] <kokey> did a commit
[25-May-2011 12:29:34] <kokey> running a reindex now
[25-May-2011 12:29:40] <rmatte> the commit should have made it visible in the interface, it should never require a full reindex
[25-May-2011 12:29:59] <rmatte> a reindex is really just a maintenance step when indexes get out of wack
[25-May-2011 12:30:23] <dhopp> rmatte: how often do you typically need to do a reindex?
[25-May-2011 12:30:32] <rmatte> I only do it when I notice issues
[25-May-2011 12:30:39] <rmatte> the last time I did a reindex was months ago
[25-May-2011 12:30:44] <Hackman238> kokey: you might need todmd.Devices.deviceSearch.catalog_object(d)
[25-May-2011 12:31:00] <kokey> Hackman238: what does that do?
[25-May-2011 12:31:22] <Hackman238> kokey: Adds the device to the catalog. If its already there, nothing.
[25-May-2011 12:31:46] <kokey> Hackman238: ah ok
[25-May-2011 12:31:48] <Hackman238> kokey: We take those steps here as a precaution so we dont have to reindex...because reindex takes forever
[25-May-2011 12:32:06] <Hackman238> kokey: v2.5.x or v3?
[25-May-2011 12:32:08] <kokey> Hackman238: excellent, just what I was after, will have to test if it works
[25-May-2011 12:32:09] <dhopp> does reindex hose up zenoss while it's running?
[25-May-2011 12:32:10] <kokey> it's v3
[25-May-2011 12:32:19] <kokey> reindex just takes _ages_
[25-May-2011 12:32:48] <Hackman238> dhopp: takes a while. should not hose zenoss unless you're running at the limit of your hardware
[25-May-2011 12:33:30] <kokey> i forgot reindex takes like a week or something
[25-May-2011 12:33:34] <dhopp> opinion:  How hard is it to write a script that will create a New Device class and then add a few sub classes under that and then apply base modelers/templates to those device classes
[25-May-2011 12:33:56] <Hackman238> dhopp: dont rebuild deviceSearch while users are working on zenoss either. you'll hear 'OMG I cant find my device!" when i reality they dont know how to use anything in the UI except the search.
[25-May-2011 12:33:59] <fragfutter> dhopp: you click it together and export it as a zenpack?
[25-May-2011 12:34:40] <dhopp> fragfutter: yeah I could do that, but then I end up with 100s of zenpacks
[25-May-2011 12:34:45] <rmatte> kokey: reindex doesn't take a week, just a few minutes on most installs
[25-May-2011 12:35:03] <Hackman238> kokey: If it takes a week- yikes!
[25-May-2011 12:35:10] <kokey> ok
[25-May-2011 12:35:14] <kokey> more like over 5 minutes
[25-May-2011 12:35:41] <Hackman238> kokey: worst DC's I have take 26 to 30 hours
[25-May-2011 12:35:42] <kokey> make that 6
[25-May-2011 12:35:44] <dhopp> reason I'm asking is we have clients…those clients have various windows/linux servers.  Some of the configuration for all of those servers will be client rather then mixing clients under "Servers/Linux", seems to make more sense to break them out
[25-May-2011 12:36:00] <Hackman238> dhopp: Ah
[25-May-2011 12:36:10] <Hackman238> dhopp: chef?
[25-May-2011 12:36:21] <dhopp> Hackman238: ?
[25-May-2011 12:36:34] <fragfutter> can i somehow unbind/block a template lower in the device class tree?
[25-May-2011 12:36:46] <Hackman238> dhopp:
[25-May-2011 12:37:01] <fragfutter> dhopp: chef as in the automatation tool chef
[25-May-2011 12:37:31] <Hackman238> dhopp: Yep. If you have that many customers, it'll be madness to keep it all accurate.
[25-May-2011 12:37:57] <Hackman238> dhopp: using chef the servers can be clients assigned roles. chef can then automagically put them in zenoss where they should be.
[25-May-2011 12:38:20] <Hackman238> dhopp: And keep them there until you make changes.
[25-May-2011 12:38:26] <rmatte> fragfutter: any lower templates take priority over higher ones no matter what... you could rename it to something else and unbind it, then bind the template higher up
[25-May-2011 12:38:29] <kokey> or puppet, or cfengine
[25-May-2011 12:38:33] <dhopp> hrm..chef looks interesting…is this "stable"
[25-May-2011 12:38:45] <Hackman238> dhopp: As stable as ruby can be
[25-May-2011 12:38:48] <Hackman238> dhopp: LOL
[25-May-2011 12:38:54] <rmatte> chef is widely used, quite stable
[25-May-2011 12:39:22] <Hackman238> dhopp: I'm currently using chef to cloudify a massive zenoss rollout for another customer.
[25-May-2011 12:39:32] <Hackman238> dhopp: Dont worry, its stable. :-)
[25-May-2011 12:39:57] <fragfutter> rmatte: i have a template ntp defined for all servers. now there are two which don't run ntp. So i need to block it for those two servers. i can bin a local (empty) ntp template to them. but i would like to just manually unbind it for them.
[25-May-2011 12:39:58] <dhopp> Hackman238: hrm…we have another system for managing our configs...
[25-May-2011 12:40:24] <Hackman238> dhopp: Does it work well?
[25-May-2011 12:40:35] <Hackman238> kokey: did your stuff show up?
[25-May-2011 12:40:36] <dhopp> Hackman238: define "well"
[25-May-2011 12:40:57] <Hackman238> dhopp: Is it managable, supported, reliable, etc?
[25-May-2011 12:41:45] <dhopp> Hackman238: the Nix team is looking at replacing it..however it supports Nix and Windows and is commercial so there is support
[25-May-2011 12:42:43] <Hackman238> dhopp: lobby for chef then ask for a pay raise ;-)
[25-May-2011 12:42:58] <dhopp> Hackman238: you hiring :-P
[25-May-2011 12:43:17] <kokey> Hackman238: still waiting for my last reindex to finish before i try again ;-)
[25-May-2011 12:43:21] <kokey> it's been 13 minutes
[25-May-2011 12:43:38] <dhopp> Hackman238: I'm part of the Nix team so I wouldn't be sad to move to a new configuration management
[25-May-2011 12:43:40] <Hackman238> dhopp: I am, but not for my businesses. LOL We're looking for a NOC superadmin/developer like rmatte.
[25-May-2011 12:44:20] <Hackman238> dhopp: email me your resume if you'd like consideration :-)
[25-May-2011 12:44:33] <dhopp> Hackman238: ha!
[25-May-2011 12:45:15] <dhopp> oh and when I asked about it being stable..I meant the zenoss integration
[25-May-2011 12:45:16] <Hackman238> dhopp: There's four of us on the team, looking for a fifth.
[25-May-2011 12:45:16] <fragfutter> rmatte: ah i can edit zDeviceTemplates. Not nice, but working
[25-May-2011 12:45:40] <Hackman238> dhopp: oh yeah, works well.
[25-May-2011 12:46:14] <dhopp> Hackman238: No offense…I don't want to live in Texas…I get sad when it's 90 degrees with 80% humidity….I will be a big sweaty cranky mess at 110 degrees and 95% humidity
[25-May-2011 12:46:53] <Hackman238> dhopp: Can't blame you. I'm not a fan of the humidity myself- I'm from NY
[25-May-2011 12:48:30] <dhopp> Hackman238: Although I have a buddy from Houston who now lives in Chicago (where I am)…he went back home during December two years ago and was complaining it was only 55 degrees there…I called him and told him to shove it since it was so cold outside my dog didn't even want to go outside (it was −35 with windchill)
[25-May-2011 12:48:59] <dhopp> If San Diego wasn't in California that would be my ideal spot…75-85 pretty much year round
[25-May-2011 12:49:37] <Hackman238> dhopp: San Diego is great. I lived in San Marcos CA for some time- always the perfect temp.
[25-May-2011 12:50:02] <Hackman238> dhopp: apocalyptic economy in CA right now though.
[25-May-2011 12:50:13] <dhopp> Hackman238: yeah except it's in California
[25-May-2011 12:50:14] <dhopp> lol
[25-May-2011 12:50:25] <Hackman238> LOL
[25-May-2011 12:51:07] <kokey> 22 minutes for a reindex
[25-May-2011 12:51:10] <kokey> so far
[25-May-2011 12:51:17] <Hackman238> kokey: how many devices?
[25-May-2011 12:51:22] <kokey> 560
[25-May-2011 12:51:35] <Hackman238> kokey: mostly servers or big switches?
[25-May-2011 12:52:02] <kokey> Hackman238: half servers, half cisco kit
[25-May-2011 12:52:04] <dhopp> But really…all I want the script to do is Take a client name and client code (i.e. abbreviation) and then Setup a client Device Class with Windows/Linux and Dev/Prod sub classes…shouldn't be that hard right :-)
[25-May-2011 12:52:07] <kokey> roughly
[25-May-2011 12:52:11] <kokey> and some vmware infrastructure
[25-May-2011 12:52:39] <Hackman238> dhopp: Its not ahrd at all.
[25-May-2011 12:52:44] <dhopp> oh and maybe "network" sub class as well
[25-May-2011 12:53:10] <dhopp> I thought I saw a sample forum post…will have to try to find it again
[25-May-2011 12:53:22] <Hackman238> kokey: in top- whats busy?
[25-May-2011 12:53:35] <dhopp> top! :-O
[25-May-2011 12:53:53] <dhopp> top can be busy in top :-)
[25-May-2011 12:54:31] <Hackman238> dhopp: not with zenoss on the box. LOL. top is just easy for everyone.
[25-May-2011 12:54:44] <dhopp> Hackman238: hehe
[25-May-2011 12:54:54] <dhopp> anyway stomach is telling me it's time to eat
[25-May-2011 12:54:57] <dhopp> be back
[25-May-2011 12:55:00] <dhopp> in a bit
[25-May-2011 12:55:41] <kokey> Hackman238: nothing unusual, a spread of the python daemons
[25-May-2011 12:56:07] <Hackman238> dhopp: For the script all you need are the following two lines in a py script: from Products.ZenUtils.ZenScriptBase import ZenScriptBase
[25-May-2011 12:56:15] <Hackman238> dhopp: and: dmd = ZenScriptBase(connect=True, noopts=True).dmd
[25-May-2011 12:56:25] <kokey> Hackman238: haha, i just realised i ran top in the same terminal as i did the reindex, so it completed
[25-May-2011 12:56:40] <Hackman238> dhopp: The rest of it is calling dmd methods off of dmd
[25-May-2011 12:56:52] <Hackman238> kokey: LOL
[25-May-2011 12:57:58] <Hackman238> kokey: If you find zeo is busy and zope is not, some one is hitting a lot of objects fast or zope is caching. If zope and zeo are busy, some massive object operation is occurring- reindex will do this.
[25-May-2011 12:58:45] <Hackman238> kokey: if zope is busy and zeo isnt showing activity during object operations somethings broke in zope.
[25-May-2011 12:59:24] <Hackman238> kokey: mind you zope can be busy for no reason, often in such case a user has tied up all the threads by holding F5
[25-May-2011 13:00:57] <kokey> i haven't had much zeo and zope CPU issues really before
[25-May-2011 13:01:18] <Hackman238> kokey: Lucky :-)
[25-May-2011 13:18:25] <Rickson-> if you monitor a linux host, should you see stuff under the "software" tab then?
[25-May-2011 13:18:54] <Rickson-> if you dont, i guess there may be something with my snmpd.conf?
[25-May-2011 13:29:45] <rmatte> Rickson-: no, it doesn't support that
[25-May-2011 13:29:55] <rmatte> net-snmp doesn't expose that kind of info
[25-May-2011 13:30:05] <rmatte> since each Linux distro uses a different package management system
[25-May-2011 13:30:26] <rmatte> There are some ZenPacks that provide that info, take a look
[25-May-2011 13:30:42] <rmatte> (they may not have been updated for use with Zenoss 3.x, I haven't looked at them in a while)
[25-May-2011 13:31:47] <Rickson-> ah
[25-May-2011 13:32:35] <chachster> Hi, I asked about liferay zenoss jmx monitoring yesterday and got this artcile as a reference: .  I'm not having much luck getting this going and don't understand why.  I've even tried just using jconsole locally and remotely and locally jconsole jmx connectivity worked fine for my liferay process but remotely it won't work.  Any ideas?
[25-May-2011 13:33:14] <Hackman238> chachster: firewall rules?
[25-May-2011 13:33:27] <chachster> I realize this is nothing specifically with Zenoss but if i can figure out why that's not working i imagine my zenoss jmx monitoring will be working afterwards.
[25-May-2011 13:33:38] <chachster> @Hackman238:  nope, i can telnet to the port both remotely and locally
[25-May-2011 13:36:25] <Rickson-> rmatte so you need a zenpack to monitor if a application is running on the host aswell?
[25-May-2011 13:47:44] <Sam-I-Am> moo.
[25-May-2011 13:52:28] <rmatte> Rickson-: no, that's built in to Zenoss
[25-May-2011 13:53:06] <rmatte> Rickson-: Checking if a process is running and checking if an application is installed are two completely different things...
[25-May-2011 13:53:20] <rmatte> for example, to check if an application is running, I can just do: ps aux | grep whatever
[25-May-2011 13:53:32] <rmatte> now let's say I decide to compile an application and install it by hand
[25-May-2011 13:53:44] <rmatte> there's no simple way for me to query and see that that package is installed
[25-May-2011 13:54:12] <rmatte> now I could use something like "dpkg" to query what packages are installed
[25-May-2011 13:54:23] <rmatte> but dpkg only applies to debian, what about the other distros out there?
[25-May-2011 13:54:36] <rmatte> hence why there are ZenPacks for checking software on certain types of Linux
[25-May-2011 13:54:43] <rmatte> but monitoring processes is totally universal
[25-May-2011 13:55:46] <rmatte> There's no difference between monitoring processes via SNMP on a Linux host and opposed to a Windows host
[25-May-2011 13:56:52] <rmatte> Monitoring a process is simple... Infrastructure -> Processes -> Select Add Process in the bottom left hand menu
[25-May-2011 13:57:06] <rmatte> then add it with a name to identify it
[25-May-2011 13:57:20] <rmatte> could be something like /usr/local/snmpd or whatever
[25-May-2011 13:57:34] <rmatte> once it's added, verify the settings for it, then remodel the device
[25-May-2011 13:57:41] <rmatte> Zenoss will pick up the process and begin monitoring it
[25-May-2011 13:58:25] <rmatte> I recommend a restart of the zenprocess daemon after discovering any processes... it's not very quick to understand that new processes have been discovered without a restart
[25-May-2011 14:11:36] <jackery> Any idea why zenjmx is continuing to poll a device after the zenjmx-containing template is unbound?
[25-May-2011 14:11:49] <jackery> I'm seeing this in my zenjmx log as well as when running "zenjmx run -v10"
[25-May-2011 14:15:38] <jmp242> push changes?
[25-May-2011 14:20:00] <jackery> jmp242 unbinding a template should be immediately effictive, right?
[25-May-2011 14:21:49] <jmp242> mmm
[25-May-2011 14:22:01] <jmp242> I would guess so, I can't say I've ever watched that closely
[25-May-2011 14:22:18] <jmp242> but it sounds like somehow the perf daemon hasn't noticed the template isn't bound
[25-May-2011 14:22:52] <jackery> Is the perf daemon involved in jmx? Excuse my ignorance, I am only beginning to use zenjmx.
[25-May-2011 14:23:30] <jackery> It looks like zenjmx polls independently of other daemons (like zenperfsnmp).
[25-May-2011 14:23:59] <jmp242> right
[25-May-2011 14:24:05] <jmp242> it's a different daemon
[25-May-2011 14:24:25] <jmp242> you might try doing the push changes (to push out changes to collectors) or you could try bouncing the daemon...
[25-May-2011 14:24:42] <jmp242> I've never used zenjmx so I have no experience with it
[25-May-2011 14:25:23] <Rickson-> rmatte thx!
[25-May-2011 14:26:01] <rmatte> np
[25-May-2011 14:27:08] <rmatte> jackery: unbinding a template is instant, yes
[25-May-2011 14:27:22] <jackery> jmp242 Thanks for the suggestions. I'll try a full stack restart. The machine in question is running on the local collector.
[25-May-2011 14:27:26] <rmatte> it may continue to collect data for a bit, but it won't actually display it or use it for anything, so it doesn't matter
[25-May-2011 14:27:53] <jackery> rmatte: thanks for the info
[25-May-2011 14:28:53] <jackery> In an earlier version of ZenOss, I had a problem with the zenperfsnmp daemon throwing an exception when it retrieved an unexpected data type from an SNMP query. When it encountered a problem poll, it would cease to continue to collect for other devices.
[25-May-2011 14:28:56] <jackery> This has since been resolved.
[25-May-2011 14:29:09] <jackery> however, the zenjmx daemon seems to be exhibiting this behavior.
[25-May-2011 14:29:31] <jackery> Has anyone successfully used the zenjmx daemon to pull machines in a wild-west type environment where continuing to poll after a failure is acceptable?
[25-May-2011 14:32:48] <jackery> Stack restart failed.
[25-May-2011 14:33:33] <jackery> Looks like a template unbind leaves cruft in the zenjmx configs:
[25-May-2011 14:33:33] <jackery> grep -l replyask002 *jmx*
[25-May-2011 14:33:33] <jackery> zenjmx-configs.db.bak
[25-May-2011 14:33:33] <jackery> zenjmx-configs.db.dat
[25-May-2011 14:33:33] <jackery> zenjmx-configs.db.dir
[25-May-2011 14:38:18] <rmatte> I don't have any experience with the zenjmx pack so I'm not sure
[25-May-2011 14:38:22] <rmatte> it could possibly be a bug
[25-May-2011 14:39:27] <jackery> rmatte: Thanks. I'll tinker some more and submit an oficial bug ticket if I feel that is necessary.
[25-May-2011 14:39:45] <jackery> I'm guessing that, by the amount of documentation about zenjmx, it isn't widely-used.
[25-May-2011 14:40:08] <rmatte> yeh, I rarely hear about it
[25-May-2011 14:56:00] <dhopp> rmatte: are you saying not a lot of people monitor jmx with zenoss?
[25-May-2011 14:56:06] <dhopp> find that hard to believe
[25-May-2011 14:56:18] <rmatte> well, more that jmx isn't an extremely common thing
[25-May-2011 14:56:36] <rmatte> in my experience anyways
[25-May-2011 14:57:46] <dhopp> better work….we have A LOT of tomcats
[25-May-2011 14:58:07] <rmatte> every time I hear the word tomcat it makes me cringer
[25-May-2011 14:58:09] <rmatte> cringe*
[25-May-2011 14:58:11] * rmatte grins
[25-May-2011 14:59:56] <rmatte> either way, if it is a bug, submit a bug report
[25-May-2011 15:04:54] f00f is now known as f00fSteR
[25-May-2011 15:05:25] <f00fSteR> what is a class in zenoss ?
[25-May-2011 15:05:32] <f00fSteR> or what are classes ?
[25-May-2011 15:11:11] <dhopp> rmatte: you don't like tomcat?
[25-May-2011 15:18:27] <jmp242> f00fSteR: you mean Device Classes and Event Classes?
[25-May-2011 15:18:40] <f00fSteR> jmp242: yessir
[25-May-2011 15:18:55] <jmp242> They are kind of like OUs, but for templates instead of GPOs
[25-May-2011 15:19:32] <jmp242> you bind perf templates to them so that a bunch of devices that are the same can get the same monitoring
[25-May-2011 15:19:48] <jmp242> it also lets you use them to filter alerting rules
[25-May-2011 15:20:10] <f00fSteR> ahh
[25-May-2011 15:20:21] <f00fSteR> complete noob at this
[25-May-2011 15:20:37] <f00fSteR> but ive been told to shut off a certain alert we've been getting from zenoss
[25-May-2011 15:20:45] <f00fSteR> so i took the class and moved it to /archive
[25-May-2011 15:20:50] <f00fSteR> which is cool no ?
[25-May-2011 15:21:14] <f00fSteR> is there a a smarter way ? or can i just turn off e-mails for certain type of alerts
[25-May-2011 15:21:18] <jmp242> umm?
[25-May-2011 15:21:40] <jmp242> You could change the event mapping or the event class zproperty to be ignore rather than status
[25-May-2011 15:21:55] <jmp242> or you could change the alerting rule that's matching the event so it doesn't match the event anymore and send e-mails
[25-May-2011 15:22:24] <jmp242> zproperty might actually be drop, pre set event class is /Ignore sorry
[25-May-2011 15:23:13] <f00fSteR> ahh
[25-May-2011 15:23:22] <f00fSteR> so change it from /archive to /ignore ?
[25-May-2011 15:41:25] <rmatte> ignore will completely drop it from the database, /archive will move it to history, so it won't alert but you'll still have it in the database
[25-May-2011 15:59:23] <f00fSteR> rmatte: ty
[25-May-2011 15:59:50] <f00fSteR> jmp242:  you almost pulled a fast one on me there bud
[25-May-2011 16:00:00] <f00fSteR> jmp242: lol j/k thanks for the help
[25-May-2011 16:04:12] <jackery> Any idea what this indicates when running zenjmx:
[25-May-2011 16:04:13] <jackery> 2011-05-25 13:03:45,602 DEBUG zen.zenjmx: Logging in as root
[25-May-2011 16:04:13] <jackery> Traceback from remote host -- Traceback unavailable
[25-May-2011 16:10:05] <rmatte> a traceback would indicate a code execution error of some sort I would imagine
[25-May-2011 16:45:10] <ben6> s there a way to import rrd file into zenoss ? i have created some with other tools and wish to compare
[25-May-2011 17:30:01] <xuru> is there a generator for dmd.ZenEventManager.getEventList?
[25-May-2011 17:30:21] <xuru> takes forever when there's 100,000+
[25-May-2011 17:34:36] <nyeates> ben6: rrdtools
[25-May-2011 17:37:08] <ben6> nyeats: ok , thx.  any docs on how to convert my rrd files so that zenoss can import them ?
[25-May-2011 17:41:44] kocolosk is now known as kocolosk|away
[25-May-2011 17:51:11] <nyeates> not that i know of off hand
[25-May-2011 17:51:29] <nyeates> you would not import them really
[25-May-2011 17:52:25] <nyeates> you would get it in rrd format with the same exact time stepping, file size, etc and then bring it into zenoss with the right file name and in the right directory
[25-May-2011 17:52:31] <nyeates> its not easy i will say that
[25-May-2011 17:52:37] <nyeates> probly why its not doced
[25-May-2011 17:52:44] <nyeates> documented*
[25-May-2011 17:55:17] kocolosk|away is now known as kocolosk
[25-May-2011 18:37:21] <jackery> Is there an easy way to check the credentials of the ZenHub? IE is there an easy way for me to test a login/password pair.
[25-May-2011 19:03:30] <locohost> anymore monitor/trend 6509 cat's with zenoss?
[25-May-2011 19:21:48] <locohost> be back in a bit
[25-May-2011 19:22:14] <locohost> just interested to know how successful the build in modeling will be of CatOS. of at all
[25-May-2011 19:22:31] <locohost> was hoping to just get cpu/mem and most importantly interface stats on all meh 10G cards
[25-May-2011 19:22:46] <Simon4> locohost: so long as you config snmp on the 6509's you should be fine
[25-May-2011 19:22:59] * Simon4 has 650something's with IOS/snmp running and gets all the stats
[25-May-2011 19:23:22] <Simon4> cpu/mem you may need to dig out the OID's and make a custom template, but that's easy enough
[25-May-2011 19:23:38] * Simon4 has to go to bed - is past midnight here, but I'm happy to chat more tomorrow
[25-May-2011 19:23:44] <locohost> yeah, thats what i think i was reading
[25-May-2011 19:23:56] <locohost> but, use the router one, and then change the oids for cpu becaue the router will be different
[25-May-2011 19:24:13] <locohost> as long as it names my interfaces, thats the most important
[25-May-2011 19:24:27] <Simon4> it should do that happily by default
[25-May-2011 19:24:37] <locohost> thks
[25-May-2011 19:24:51] <locohost> everyone is touchy about the core, need a maintanance window just to do the initial discovery
[25-May-2011 19:25:10] <Simon4> does this carry a full table?
[25-May-2011 19:25:58] <Simon4> there's a property for only collecting direct routes, which will reduce modelling load on devices with huge routing tables (like internet facing routers/switches)
[25-May-2011 19:26:20] <locohost> routing table shouldnt be that huge
[25-May-2011 19:28:17] <locohost> wait, are you suggesting modeling may try to do something to every device connected to it?
[25-May-2011 19:28:39] <Simon4> no, it just struggles on routers/switches with huge routing tables
[25-May-2011 19:28:57] <Simon4> (in the thousands)
[25-May-2011 19:29:03] <locohost> im thining it should only see like 50 or so..
[25-May-2011 19:29:12] <Simon4> just scanning interfaces etc will be easy
[25-May-2011 19:29:19] <Simon4> that will be fine
[25-May-2011 19:29:53] <Simon4> if you set up snmp and run snmpwalk against it you'll see fairly quickly whether the interface data is coming back
[25-May-2011 19:30:20] <locohost> yeah, i typically snmpwalk before a model
[25-May-2011 19:31:01] <locohost> write it to a file then start grepping through that
[25-May-2011 19:31:08] <Simon4> right, must head to bed - catchya tomorrow
[25-May-2011 19:31:14] <locohost> take care
[25-May-2011 22:26:46] kocolosk is now known as kocolosk|away
[26-May-2011 00:00:01] [disconnected at Thu May 26 00:00:01 2011]
[26-May-2011 00:00:01] [connected at Thu May 26 00:00:01 2011]
[26-May-2011 00:00:17] [zenoss-logger (logger bot) has joined #zenoss]
[26-May-2011 00:09:25] kocolosk|away is now known as kocolosk
[26-May-2011 00:45:03] bobrog is now known as 36DABOJJR
[26-May-2011 03:14:45] <Rickson-> why can i select "Monitor these components" under "Monitoring" on some ipservice, and some not?
[26-May-2011 03:52:37] <fragfutter> Rickson-: you can only monitor IP services that listen on an interface that zenoss can reach
[26-May-2011 07:37:19] <hmp> I get this while installing the zenoss-stack pacakage -
[26-May-2011 07:37:48] <hmp> anyone got this worked out?
[26-May-2011 08:28:13] kocolosk is now known as kocolosk|away
[26-May-2011 08:35:01] <jmp242> hmp you may want to post on the forum
[26-May-2011 08:42:16] kocolosk|away is now known as kocolosk
[26-May-2011 08:43:20] <hmp> jmp242: i just added the LSB header myself and ran update-rc.d
[26-May-2011 08:43:34] <hmp> jmp242: i would report that as a bug, couse it should be solved in debian package
[26-May-2011 08:46:40] <jmp242> Feel free to report the bug
[26-May-2011 08:47:11] <hmp> was just searching for that, thx
[26-May-2011 09:07:00] <brice> How well can Zenoss scale?  What sort of resources would it need to monitor ~6,000 nodes?
[26-May-2011 09:08:15] <rocket> multiple hubs and collectors .. it can do it ..
[26-May-2011 09:08:29] <rocket> you will need to tune mysql and zenoss caches
[26-May-2011 09:09:21] <rocket> typically you will have a large master and a couple smaller collector machines
[26-May-2011 09:09:31] <brice> ahh okay
[26-May-2011 09:09:55] <rocket> there is a community zenpack for distributed collectors
[26-May-2011 09:10:11] <rocket> I havent used it myself but others in here have
[26-May-2011 09:10:11] <brice> cool, I'll take a look
[26-May-2011 09:10:42] <rocket> as the enterprise side has its own collector zenpack
[26-May-2011 09:22:03] <fragfutter> nyeates: anyone from management coming?
[26-May-2011 09:22:52] <stevenhirsch> hi just was wondering if anyone has zenoss configured to send an alert via voice or sms?
[26-May-2011 09:25:21] <nyeates> fragfutter: possibly not. I talked to the ceo, and though he has added it to his calendar, he is not sure about the sec he is pingin me now actually heh
[26-May-2011 09:39:21] <rmatte> nice, we have 20+% packet loss to the internet this morning
[26-May-2011 09:39:46] <rmatte> packet loss to our tools as well apparently
[26-May-2011 09:39:54] <fragfutter> i only have 20% brainloss since debugging kerberos this morning
[26-May-2011 09:39:59] <rmatte> lol
[26-May-2011 09:40:11] <rmatte> yeh, kerberos is a pain
[26-May-2011 09:40:58] <fragfutter> how can anyone write a (rather) modern protocol that gives close to no debugging output and if it does it gives you messages like "Unspecified GSS failure"
[26-May-2011 09:41:43] <rmatte> Mi-cro-soft
[26-May-2011 09:57:05] <rmatte> --- ping statistics ---
[26-May-2011 09:57:05] <rmatte> 247 packets transmitted, 194 received, 21% packet loss, time 246445ms
[26-May-2011 09:57:05] <rmatte> rtt min/avg/max/mdev = 25.451/45.676/61.027/4.017 ms
[26-May-2011 09:57:08] <rmatte> eugh
[26-May-2011 09:57:14] <rmatte> impossible to get any work done this morning
[26-May-2011 09:57:59] <fragfutter> latency doesn't look too good either
[26-May-2011 09:58:41] <rmatte> lol
[26-May-2011 10:00:05] <fragfutter> to adapt userfriendly: how would you work a decent session of quake with this latency.
[26-May-2011 10:00:40] <rmatte> unreal tournament here, but I see your point
[26-May-2011 10:00:41] <rmatte>
[26-May-2011 10:00:59] <rocket> fragfutter: you would have to see the future so you can preselect which packets are important and send those appropriately
[26-May-2011 10:01:18] <rmatte> time warp routing
[26-May-2011 10:01:29] <Simon4> quantum routing would be great
[26-May-2011 10:01:34] <rmatte> lol
[26-May-2011 10:01:39] <Simon4> nfi how it got there, but hell, it's there!
[26-May-2011 10:02:01] <rmatte> "How did they get the data before I sent it!?"
[26-May-2011 10:02:44] <rocket> imagine a world where one packet shows up at a router with no wires connecting them ... "no not wifi, you idiot...."
[26-May-2011 10:03:37] <rmatte> lol
[26-May-2011 10:04:31] <dhopp> rmatte: are you saying impossible to get any work done because you have network issues, or just because you can't get to google? lol
[26-May-2011 10:04:45] <rocket> whats the difference?
[26-May-2011 10:04:46] <rocket> 
[26-May-2011 10:04:51] <dhopp> rocket: true
[26-May-2011 10:04:54] <rmatte> dhopp: it's affecting our connectivity in general, including to our tools
[26-May-2011 10:05:03] <rmatte> rocket: or take it one step further, schodinger's router
[26-May-2011 10:05:14] <rmatte> The packets exist and don't exist simultaneously
[26-May-2011 10:05:17] <dhopp> rocket: as I like to say…I'm a sophisticated interface to google :-P
[26-May-2011 10:05:28] <rocket> rmatte: which tool?  "the bow and arrow?"
[26-May-2011 10:05:44] <rmatte> the hatchet, that is flying towards you
[26-May-2011 10:05:45] <rmatte>
[26-May-2011 10:06:31] * rocket holds up a large stump .. ;p
[26-May-2011 10:06:41] <rocket> I knew I needed to keep one of these handy ..
[26-May-2011 10:07:14] <Simon4> o/~ log, log, everyone get your log o/~
[26-May-2011 10:10:03] <rmatte> lol
[26-May-2011 10:10:30] <rmatte> there, the packet loss is down to 0
[26-May-2011 10:16:11] <Hackman238> hello all
[26-May-2011 10:16:42] <dhopp> hola como estas
[26-May-2011 10:17:44] <Hackman238> dhopp: Not too bad. Yourself?
[26-May-2011 10:18:06] <dhopp> Hackman238: alright…but don't expect any more spanish out of me
[26-May-2011 10:18:11] <rocket> me grande ..
[26-May-2011 10:18:42] <Hackman238> dhopp: No worries- I don't know any spanish. English, French and German- but only english well
[26-May-2011 10:18:53] <dhopp> Hackman238: I know that and una cerveza, por favor
[26-May-2011 10:20:35] <Hackman238> dhopp: LOL
[26-May-2011 10:25:06] <Jane_Curry> stevenhirsch - yup - I've configured Zenoss for SMS alerts
[26-May-2011 10:33:15] <kokey> there isn't much documentation about zenoss handling syslog messages
[26-May-2011 10:33:39] <Jane_Curry> nyeates - be good to have some big guns to talk to
[26-May-2011 10:33:43] <kokey> how does it map syslog entries to event classes?
[26-May-2011 10:34:46] <Jane_Curry> get my Event Management paper from
[26-May-2011 10:35:08] <chachster> Hi, figured I'd come give the solution to a problem i had during the last couple of days in case it could help somebody else.   I had issues with zenoss monitoring liferay through zenjmx.  My problem was that it uses RMI and while my RMI port on my server was open, RMI was opening a new socket to send the communication back to zenoss which was getting blocked by my firewall
[26-May-2011 10:35:14] <Jane_Curry> there's a section (4) in there on syslog processing
[26-May-2011 10:35:28] <chachster> solution:  allow all traffic from the zenoss server through on your server firewall and it will work fine.
[26-May-2011 10:35:34] <kokey> Jane_Curry: cool, thanks, I'll have a look right now
[26-May-2011 10:35:54] <dhopp> chachster: easy solution :-)…sometimes tough to trackdown
[26-May-2011 10:36:24] <kokey> it's weird how I seem to care about zenoss stuff when tomorrow is my last day at this contract
[26-May-2011 10:36:24] <Jane_Curry> chachster - might be hard to justify to security geeks too
[26-May-2011 10:36:32] <kokey> but I suspect zenoss will cross my path in the future
[26-May-2011 10:36:48] <chachster> dhopp:  Yeah it was a doozy to track down, I assumed the whole way that all communication was going through my open port which was a bad assumption from my part.
[26-May-2011 10:40:08] <chachster> Jane_Curry:  Yeah it could be, i think it wouldn't be too bad if the zenoss server is on the same subnet as the servers it is monitoring though.  Then it's just inside traffic
[26-May-2011 10:46:03] <dpetzel> if I have a ConflictError: database conflict error, is there a way to find the oid its referencing?
[26-May-2011 10:47:16] <rmatte> fragfutter: from the support ticket regarding our latency "identifed a massive download that was consuming almost all of
[26-May-2011 10:47:16] <rmatte> the Nova bandwidth
[26-May-2011 10:47:38] <rmatte> figures
[26-May-2011 10:47:54] <rmatte> dpetzel: look carefully at that message...
[26-May-2011 10:48:08] <rmatte> It should say something like "Unresolved: 0"?
[26-May-2011 10:48:14] <rmatte> as long as it's 0, it's good.
[26-May-2011 10:48:30] <rmatte> that means it self-resolved the issue (which is normal)
[26-May-2011 10:48:37] <nyeates> Jane_Curry: your wish will be granted today! I got your reply on contributions - we will hash it out. I think this is finally moving with more speed.
[26-May-2011 10:49:49] <nyeates> For those who do not know - Zenoss Inc will likely be QA'ing and reviewing and officiating Janes zenpack dev docs.
[26-May-2011 10:50:03] <jmp242> That sounds good to me
[26-May-2011 10:50:10] <nyeates> meeting will start in ~10 mins
[26-May-2011 10:50:16] <dpetzel> rmatte: thanks, but I'm not seeing that. I should probably provide a little more context Im running the appProfiler pack, and when kicking off the following via zendmd I get the message
[26-May-2011 10:50:17] <dpetzel> dmd.Profiles.rulesets.manage_runAllMatches()
[26-May-2011 10:50:21] <dhopp> nyeates: that's awesome
[26-May-2011 10:50:30] <dpetzel> yields ConflictError: database conflict error (oid 0xa2df91, class
[26-May-2011 10:50:31] <jmp242> !!
[26-May-2011 10:50:35] <jmp242>
[26-May-2011 10:50:38] <rmatte> dpetzel: perform a reindex
[26-May-2011 10:50:41] <rmatte> reindex()
[26-May-2011 10:50:43] <rmatte> commit()
[26-May-2011 10:50:47] <rmatte> then try again
[26-May-2011 10:51:04] <dpetzel> rmatte, actually just did that based on some forum posts, still no luck
[26-May-2011 10:51:17] <Hackman238> dpetzel: try sync()
[26-May-2011 10:51:23] <Hackman238> dpetzel: then commit()
[26-May-2011 10:51:31] <dpetzel> k tryin sync now
[26-May-2011 10:52:32] <kokey> Jane_Curry: thanks for that, a good document, goes some way towards explaining why there's no Zenoss documentation on this
[26-May-2011 10:52:53] <kokey> Jane_Curry: it looks like the syslog stuff is not really made to be user editable at the moment
[26-May-2011 10:53:38] <Hackman238> dpetzel: Work for you?
[26-May-2011 10:53:48] <kokey> this also looks like a good place to wedge in something like an event correlator, like SEC but in python
[26-May-2011 10:54:22] <dpetzel> sync ran without error, re-running my autoProfile dmd script
[26-May-2011 10:54:27] <dpetzel> it takes a while before it barfs
[26-May-2011 10:55:00] <Hackman238> dpetzel: If the app runs for a long time, consider running sync() during the script occasionally to avoid conflicts
[26-May-2011 10:55:25] <nyeates> meeting will commence in ~5 mins
[26-May-2011 10:55:31] <dpetzel> k, what is the functional difference between sync and commit? is one preferred over another?
[26-May-2011 10:55:32] <rmatte> T-minus!
[26-May-2011 10:56:04] <rmatte> dpetzel: they have different purposes alltogether, and I think you mean between sync and reindex
[26-May-2011 10:56:04] <Hackman238> dpetzel: sync() syncs dmd with zeodb current state, commit commits the changes.
[26-May-2011 10:56:23] <Hackman238> dpetzel: you cant replace commit with sync, you'll want to sync before you commit.
[26-May-2011 10:56:37] <rmatte> Hackman238: I think he meant between sync and reindex
[26-May-2011 10:56:44] * rmatte shrugs
[26-May-2011 10:57:12] <dpetzel> so it bombed again ConflictError: database conflict error (oid 0xa2df91, class
[26-May-2011 10:57:18] <Hackman238> dpetzel: Oh- reindex reindex objects, sync just clears up conflicts
[26-May-2011 10:57:29] <Hackman238> dpetzel: Edit your script and sync() before you commit()
[26-May-2011 10:57:38] <dpetzel> the challenge I have is that I'm only making the single call (which takes a long time)
[26-May-2011 10:57:48] <dpetzel> this is what Ihave right now
[26-May-2011 10:57:48] <dpetzel> dmd.Profiles.rulesets.manage_runAllMatches()
[26-May-2011 10:57:49] <dpetzel> commit()
[26-May-2011 10:57:58] <rmatte> ok, so add sync() right before the commit
[26-May-2011 10:58:01] <rmatte> in the script itself
[26-May-2011 10:58:02] <Hackman238> dpetzel: correct
[26-May-2011 10:58:17] <Hackman238> dpetzel: it non fatal cases it'll clear up conflicts
[26-May-2011 10:58:21] <dpetzel> K, adding the sync() before the commit
[26-May-2011 10:58:46] <Hackman238> dpetzel: also you could try: commit() exception: sync commit()
[26-May-2011 10:59:15] <Hackman238> dpetzel: not on one line, but broken out as needed for python
[26-May-2011 10:59:35] <rmatte> Hackman238: general except lines are a no no
[26-May-2011 10:59:48] <dpetzel> Hackman238: yeah I see what your after there with the try block
[26-May-2011 11:00:19] <Hackman238> rmatte: Just proof of concept :-)
[26-May-2011 11:00:23] <rmatte> lol
[26-May-2011 11:00:24] <dpetzel> going back tot he original question a bit, is that OID something that can be looked up to find the offending object?
[26-May-2011 11:00:59] <rocket> dpetzel change the session-resolution-seconds to 1200 in zope.conf and restart zenoss
[26-May-2011 11:01:04] <Hackman238> dpetzel: No, the object has nothing to do with the oid except in the case of a mib object
[26-May-2011 11:01:26] <dpetzel> Hackman238: K thanks
[26-May-2011 11:01:50] <Hackman238> dpetzel: Also, rockets suggestion will help prevent conflicts on long running dmd operations.
[26-May-2011 11:02:15] <rmatte>
[26-May-2011 11:02:25] <rmatte> there's a working except statement for you
[26-May-2011 11:02:34] <dpetzel> Hackman238 & rocket: thanks will look at the session option
[26-May-2011 11:02:53] <dpetzel> rmatte: thanks, exactly what I had written myself
[26-May-2011 11:02:59] <Hackman238> dpetzel: Use rmatte's pasted code- its a little safer than just an except, which will catch everything
[26-May-2011 11:03:04] <nyeates> Everyone! Meeting is adjourning. If we can get some silence for the next few minutes, that would be great
[26-May-2011 11:04:10] <nyeates> We want to properly explain our commitment to the community and to do that, we have Zenoss CEO, Bill Karpovich here today.
[26-May-2011 11:04:44] <nyeates> We will hear from him for a little and open to a few comments. Bill has limited time, so will have to leave us in a bit.
[26-May-2011 11:04:54] <bedwards_zenoss> Hi Nick.
[26-May-2011 11:04:54] <nyeates> Take it away Bill.
[26-May-2011 11:05:14] <bkarpovich> hey, everybody
[26-May-2011 11:05:31] <bkarpovich> hope everyone is doing well.
[26-May-2011 11:05:31] <Jane_Curry> Greetings - good to have you along!
[26-May-2011 11:05:39] <jmp242> hi
[26-May-2011 11:05:54] <Hackman238> Glad to have you here today. Look forward to soem infos.
[26-May-2011 11:06:03] <bkarpovich> my pleasure... sorry that we have been a bit invisible...
[26-May-2011 11:07:01] <bkarpovich> as Nick suggested.... the key thing that i would like to make clear is that Core and community continue to be essential to Zenoss overall...
[26-May-2011 11:07:08] <Jane_Curry> focused elsewhere, perhaps
[26-May-2011 11:07:23] <bkarpovich> we have had several changes over the last 6 months
[26-May-2011 11:07:46] <bkarpovich> with our community team, product evolution, repo evolution, etc...
[26-May-2011 11:08:48] <bkarpovich> our communications during this have not been as clear as they certainly could have been and we have been working through internally the details of whats next
[26-May-2011 11:08:59] pmcguire is now known as ptmcg
[26-May-2011 11:09:25] <bkarpovich> in terms of how we are supporting the community from a team perspective (in addition the good work that Nick is doing)
[26-May-2011 11:09:49] <bkarpovich> and how the repos evolve and the future releases of Core
[26-May-2011 11:10:28] <bkarpovich> our plan is to spell out the details on each of these in the coming weeks
[26-May-2011 11:10:50] <bkarpovich> and more opps to collaborate on this (IRC, etc...)
[26-May-2011 11:11:48] <bkarpovich> the one thing is not a question is that Core and the community are essential and we are going get them both moving with the energy that we have had in the past and more
[26-May-2011 11:12:51] <bkarpovich> Nick has pulled together some of the key questions that are on people's minds...
[26-May-2011 11:13:40] <bkarpovich> it would be great in this session and over the next couple of days to get those shared and appreciated on our end
[26-May-2011 11:14:28] <bkarpovich> we can certainly address any specific questions/concerns directly... or put them in the queue for the coming weeks
[26-May-2011 11:14:43] <nyeates> Here is what we have so far: 1) Where is core/community going? What is the strategy? 2) What is happening with relstorage and future features? 3) Documentation improvement for developers 4) How will Core be released? 5) Pricing 6) Communication needs to happen now
[26-May-2011 11:14:51] Hugh is now known as Hugh_
[26-May-2011 11:15:24] bedwards_zenoss_ is now known as bedwards_zenoss
[26-May-2011 11:15:53] <nyeates> Any questions / topics anyone can add to these?
[26-May-2011 11:15:57] <Hackman238> nyeates: Can you qualify how core features are differentiated from enterprise features?
[26-May-2011 11:16:01] <Hackman238> nyeates: Can you qualify Zenoss's prioritization in moving core releases in sync with enterprise releases?
[26-May-2011 11:16:13] <Jane_Curry> Suggest also some specifics as to what Core DON'T get as opposed to Enterprise (and some of this may be under the covers...)
[26-May-2011 11:16:56] <jmp242> I'd also like to see some idea if Zenoss inc is OK with, wants to promote, or would prefer this didn't happen wrt
[26-May-2011 11:17:28] <jmp242> Core / community bug bounties, or pledges / whatever to pay third party developers to extend Core etc
[26-May-2011 11:17:51] <bkarpovich> jane -- please help me understand the distinction you're looking for (thanks, btw, for all that you do in the community. i am continually in awe! :-)
[26-May-2011 11:18:13] <jmp242> so that people who can contribute percentages of the funding for a feature can join together to get it done by the talented Community developers etc
[26-May-2011 11:18:29] <Jane_Curry> Some things are specific to Enterprise and are documented as such
[26-May-2011 11:19:21] <rmatte> Question:  Zenoss Core is labeled as such because it is supposed to represent the core functionality of the product.  A decision was made some time ago to move from standard zopedb to relstorage to improve performance.  Recently, a decision was made to remove the relstorage code from Zenoss Core.  My question is, how can you continue to consider Zenoss Core as a core version when you are omitting core features from it?  The database is the foundation of the produ
[26-May-2011 11:19:22] <rmatte> By using standard zopedb on one and relstorage on the other you are essentially creating a fork between Enterprise and Core rather than having Core be the foundation of Enterprise.  This is going to affect future development as well as testing.
[26-May-2011 11:19:27] <Jane_Curry> but I suspect there is some "enabling stuff" that has gone on for some enterprise features that is there under the covers in Core
[26-May-2011 11:20:02] <Hackman238> nyeates: What is Zenoss's strategy for capitalizing on small and medium businesses instead of turning them away with per device licensing? I have tens of dozens of customers who would buy Zenoss if the pricing model fit the business size.
[26-May-2011 11:20:39] <Jane_Curry> rmatte - thanks - this is the sort of thing I suspected and was trying to get at - didn't realise relstorage had become Enterprise only
[26-May-2011 11:21:21] <rmatte> Jane_Curry: yeh, they removed the relstorage code from the Core subversion a month or two ago
[26-May-2011 11:21:54] <Jane_Curry> As you say, this effectively means there are 2 different products
[26-May-2011 11:22:04] <Hackman238> nyeates: What is Zenoss's strategy for handling Zenpack comptibility and best practice? I might suggest random Zenpack audits for bad code. The best packs can then be pointed to as ideal examples of best practice.
[26-May-2011 11:22:16] <bkarpovich> all good questions and suggestions.... please keep them coming...
[26-May-2011 11:22:32] <rmatte> Jane_Curry: exactly
[26-May-2011 11:22:52] <Jane_Curry> I thought a lot of the value in the Core / Enterprise model is that you can "upgrade" to Enterprise but you simply get extra stuff
[26-May-2011 11:23:03] <jmp242> Jane_Curry: and of course that split does have implications for users and I think affects their calculus in what product to use
[26-May-2011 11:23:04] <bkarpovich> re relstore.... here what our thinking was....
[26-May-2011 11:23:27] <Jane_Curry> .. plus the support contract which was what most people REALLY wanted - as Hackman238 referred to
[26-May-2011 11:23:27] <Hackman238> The idea was to limit the scalability of Core.
[26-May-2011 11:23:42] <rmatte> Hackman238: ZenPack audits is a phenominal idea, there are some baaaad baaaad ZenPacks out there
[26-May-2011 11:23:48] <bkarpovich> part of the value of "Enterprise" is easier large-scale deployments
[26-May-2011 11:24:48] <rmatte> I don't see relstorage as increasing the scalability of Zenoss, you still have the RRD bottleneck to deal with, which you need to turn to collectors for, which Enterprise does much better than Core.
[26-May-2011 11:24:55] <Jane_Curry> rmatte - and some might be mine.  We need good documentation of good / bad practise
[26-May-2011 11:25:05] <jmp242> rmatte: I think that zenpack audits should be enabled for the community to step up on also, assuming we get the docs or best practices so we can do it appropriately
[26-May-2011 11:25:07] <bkarpovich> while the DB is certainly base plumbing, mysql here was really being used a persistence layer under the same DB interface
[26-May-2011 11:25:29] <rmatte> Jane_Curry: I don't think any of the ones that I saw from you were really that bad
[26-May-2011 11:25:40] <Jane_Curry> hmmmmmmmm
[26-May-2011 11:26:04] <rmatte> jmp242: agreed
[26-May-2011 11:26:27] <Hackman238> bkarpovich: The simple fact is Zenoss will not capture a large lost market of small and medium business by limiting Core scalability.
[26-May-2011 11:26:33] <rmatte> jmp242: they would also allow us to obsolete ZenPacks, because lord knows there are a lot of them in the list that need obsoleting.
[26-May-2011 11:26:35] <bkarpovich> so the DB layer was not being changed perse... just a faster persistence layer and Core was not being crippled in any way (kind of like supporting Oracle in an Enterprise edition)
[26-May-2011 11:27:53] <Jane_Curry> but if you have a load of different base infrastructure code then you have different products, different support groups,
[26-May-2011 11:28:26] <bkarpovich> a mistake we made in the process was that product management had it slated for an enterprise feature but we began development in the core trunk.  for better or worse, our decision was to pull it back. we are continuing to look this as part of the plan forward.
[26-May-2011 11:28:35] <jmp242> I guess the confusion here is that much of the community thinks that zenoss hasn't really been selling enterprise as a product so much as a turnkey solution. I.e., most of the value hasn't been in the Zenpacks - and in fact selling the software more commercially like some competitiors can be done much more cheaply (we think anyway per the above competitiors), and that limiting core scalibility doesn't get you purchases
[26-May-2011 11:28:42] <Jane_Curry> What about folk who have both Core and Enterprise (perhaps through company mergers....) - they need two different sets of knowledge
[26-May-2011 11:29:00] <nyeates> Jane_Curry: my understanding is that relstorage still stores everything in object stores as before - just puts it in  a different storage mechanism
[26-May-2011 11:29:05] <rmatte> bkarpovich: Not to mention that the community then has to once again hack together something and try to support it for themselves.  I read a one page document explaining how to integrate relstorage in to core, and for such a relatively simple process, I'm amazed that it's not being officially supported in Core when it is in Enterprise.  You're opening the door for all sorts of conflicts and oversights when testing.
[26-May-2011 11:29:42] <Hackman238> bkarpovich: By making accurate documentation, community bennifits and enforcing Zenpack best practice Zenoss Inc could reduce the costs of Enterprise for small and medium businesses and at the same time promote Core development whilst reducing the number of support calls for cryptic, time consuming Zenoss repairs due to broken Zenpacks or the simple lack of documentation.
[26-May-2011 11:30:18] <Jane_Curry> hear , hear!
[26-May-2011 11:30:21] <Hackman238> It'll help everyone- you, me, the community and new customers.
[26-May-2011 11:30:48] <Hackman238> I have three businesses in three states across the country. I assure you this will help.
[26-May-2011 11:30:59] <bkarpovich> rmatte: all good points.  to be clear, the Core/Enterprise feature decisions are challenging and involve tough trade offs.
[26-May-2011 11:31:11] <rmatte> Hackman238: I have found the support documentation to be slightly lacking... I always feel like I'm doing more reverse engineering than development when creating more complex ZenPacks
[26-May-2011 11:31:24] <rmatte> sorry, I meant development documentation, not support
[26-May-2011 11:31:48] <Jane_Curry> Yup - especially around the "new" (nearly 12 months old) UI
[26-May-2011 11:31:49] <Hackman238> rmatte: Exactly- if this was not the case, many enterprise tickets would never happen. This reduces Zenoss's costs for supporting lower paying customers.
[26-May-2011 11:32:20] <bkarpovich> with the service impact management and analytics capabilities in Enterprise, our goal is to be more "generous"
[26-May-2011 11:32:40] <bkarpovich> on the platform stuff
[26-May-2011 11:32:56] <Jane_Curry> generous??
[26-May-2011 11:32:58] <rmatte> bkarpovich: You don't need to take relstorage away from Core to accomplish that
[26-May-2011 11:33:07] <jmp242> Are you interested in selling the features a la prtg, or the quick up and running solution? I.e. selling the code ought to be cheap - far less than what you're currently charging. If you're selling the whole integrated package, then I understand the costs a lot better - but in that case, having less / unsupported core versions doesn't hurt Enterprise at all
[26-May-2011 11:34:04] <bkarpovich> rmatte: as I said.... tough decisions... we don
[26-May-2011 11:34:35] <bkarpovich> 't always get them right and we are continuing to discuss
[26-May-2011 11:34:59] <jmp242> I.e. you can buy A.D. etc from MS for cheap, but getting it going take s a lot of work. You can also engage MS or a solution provider to do it for you and or help you a long and that costs a lot more, but that's also the value. Where is Zenoss Inc wanting to be on that continuum?
[26-May-2011 11:35:00] <Jane_Curry> What's happening with the DataCenter Insight stuff?  My understanding is that is extra charge to Enterprise and a no-no to Core
[26-May-2011 11:35:16] <dhopp> I got a quote for Enterprise but could not secure a budget.  I have since been able to convince them to go with Core and that could lead to an eventual Enterprise purchase (primarily for support) but taking a key "core" component out of Zenoss Core is making me think twice about this recommendation
[26-May-2011 11:35:59] <jmp242> and note dhopp as a good example. He's not necessarily sold on the Enterprise features, but on the support for the tricky traceback errors etc
[26-May-2011 11:36:08] <Jane_Curry> dhopp - exactly my point.  Eventually, you don't have an "upgrade", you have an "install the new product".  Why do it?
[26-May-2011 11:36:31] <rmatte> bkarpovich: also, while I have you here I just wanted to mention that we recently did some pricing negotiations with you guys for Enterprise, and even after talking the price down quite a bit it was still very high.  I'm sure that your current pricing model works great for large companies like RackSpace for instance... but for small to medium sized companies, and managed services providers that are in their infancy (such as ourselves), it's pretty unreasonable...
[26-May-2011 11:36:38] <Hackman238> bkarpovich: I might add that some large customers, like RackSpace, choose Zenoss for the fact its open source. We buy enterprise Zenoss for the mere fact we must have a supported product. If Zenoss were to close source or branch core- it would limit inernal development and open us to consider other products.
[26-May-2011 11:36:38] <rmatte> I think that there's a lot of room for improvement with that.
[26-May-2011 11:37:45] <Hackman238> rmatte: Yeah I like the way bkarpovich hasn't commented on my questions or remarks.
[26-May-2011 11:38:10] <jmp242> Per what rmatte is saying - where I work we have a budget that would be pretty easy to get to buy a zenpack or two at the $500-$2k mark
[26-May-2011 11:38:45] <jmp242> or to purchase a one off support call at the $300 range or so much per hour, but not at an 8hr minimum commitment etc
[26-May-2011 11:38:58] <bkarpovich> Hackman238:  will definitely comment on your points....
[26-May-2011 11:39:02] <rmatte> One of the big things with the pricing is that it includes full support and all ZenPacks, whether you like it or not.  Beyond logging enterprise bug tickets, we wouldn't need any type of support as I already know the product inside and out.  We'd also only need select ZenPacks.  With the current pricing model we'd be paying to support code and features that we don't use.
[26-May-2011 11:39:02] <jmp242> I want to get across that I'm not looking to be a leach or looking for handouts
[26-May-2011 11:39:04] <Jane_Curry> jmp242 - but who "suppports" your chargeable zenpack?
[26-May-2011 11:39:12] <jmp242> the community
[26-May-2011 11:39:20] <jmp242> or we buy a support pack or something
[26-May-2011 11:39:34] <jmp242> the idea here is that we can buy LOTS of software without a support commitment from lots of vendors
[26-May-2011 11:39:41] <jmp242> i.e. the support is extra
[26-May-2011 11:39:56] <jmp242> we can also get, from MS for instance, packs of 2 or 4 support calls for a price we could afford
[26-May-2011 11:39:58] <rmatte> Yeh, bundling of support is what kills the price
[26-May-2011 11:40:05] <Jane_Curry> and of course one option is to support it yourself
[26-May-2011 11:40:09] <jmp242> right
[26-May-2011 11:40:15] <Hackman238> bkarpovich: I want you to know I look at this problem from the perspective of a rep for a large customer, a business man, a community developer and a consultant with many Zenoss contracts- please don't take anything I say as malice, it's just my perspective.
[26-May-2011 11:40:32] <Jane_Curry> and the ebenfit of a strong community is that you get lots of help
[26-May-2011 11:40:45] <bkarpovich> all -- unfortunately, I need to jump out.   thanks for the initial discussion.  please keep it going.  Nick will bring this all back and we'll directly address in the coming weeks.
[26-May-2011 11:40:47] <rmatte> The price of the code itself should cover the maintenance and development of the code... but if I'm not going to be calling the support team there, then why am I chipping in for their salaries?
[26-May-2011 11:40:52] <jmp242> We've been doing great with Core and no support. Hell, even on the stuff we buy, we generally don't need lots of support time
[26-May-2011 11:41:01] <jmp242> that's why I have a job for instance
[26-May-2011 11:41:03] <Hackman238> bkarpovich: Saved by the bell, huh?
[26-May-2011 11:41:18] <jmp242> so I don't think that changes for buying a zenpack say
[26-May-2011 11:41:22] <bkarpovich> Hackman238: :-)
[26-May-2011 11:41:27] <Hackman238> bkarpovich: We appreciate you're participation in the meeting. Thank you very much!
[26-May-2011 11:41:27] <rocket> Hackman238: fortunately for us he is a very busy guy
[26-May-2011 11:41:32] <rmatte> jmp242: yeh, I provide more support in here (sometimes to Enterprise users) than I ask for, and we've been doing perfectly fine.
[26-May-2011 11:41:38] <jmp242> Thanks for listening
[26-May-2011 11:41:47] <Jane_Curry> Be seein' ya
[26-May-2011 11:41:51] <nyeates> I did state that there would be limited time here, and that this was a question collection phase
[26-May-2011 11:41:57] <jmp242> I know, I've noticed enterprise users on the forums too
[26-May-2011 11:42:09] <jmp242> I assume they sometimes get what they need from there
[26-May-2011 11:42:14] <davetoo> I'm surprised it's actually been 45 minutes already
[26-May-2011 11:42:14] <nyeates> I am definetley going to parse over everything and get it all summarized
[26-May-2011 11:42:23] <rmatte> nyeates: make sure that you make it clear to Bill that we appreciate him showing up... I wasn't expecting that.
[26-May-2011 11:42:27] <rocket> How many CEO's do you find on irc any more?
[26-May-2011 11:42:35] <jmp242> not too many that I know of
[26-May-2011 11:42:42] <jmp242> I'm very glad he at least took the questions
[26-May-2011 11:42:47] <Hackman238> nyeates: Try to reword my stuff so it doesnt sound cold and souless- I just wanted to be clear.
[26-May-2011 11:43:01] <Hackman238> :-)
[26-May-2011 11:43:04] <rmatte> Hackman238: don't worry yourself, my stuff always sounds cold and souless
[26-May-2011 11:43:06] <nyeates> Hackman238: Heh No worries.
[26-May-2011 11:43:33] <rmatte> Hackman238: as long as the point is made, no sense in dressing it up, in my opinion
[26-May-2011 11:43:34] <rmatte>
[26-May-2011 11:43:46] <jmp242> I want it to be clear, there are a lot of people, myself included, who aren't looking for "free" with Zenoss, but who can't pay $xx,xxx + a year either
[26-May-2011 11:43:46] <Jane_Curry> Could we pick up athread about the example ZenPack from the last IRC......
[26-May-2011 11:44:10] <rmatte> jmp242: that's our exact position right now
[26-May-2011 11:44:20] <Hackman238> jmp242: You and many.
[26-May-2011 11:44:20] <Jane_Curry> jmp242 - totally agree.  I think many of us have been trying to get this message through for some time now
[26-May-2011 11:44:27] <jmp242> I don't wat this to be "waaaahhhh!!! you owe us the free toy and work always"
[26-May-2011 11:44:30] <rmatte> jmp242: we'd grab enterprise in a heartbeat if we didn't have to pay for the stuff that we won't use (mostly support)
[26-May-2011 11:44:49] <Jane_Curry> If you want free then it is free but there is NO support for free
[26-May-2011 11:44:52] <dhopp> I should also say that the ability to have Zenoss develop complicated zenpacks (especially since zenpack documentation blows) was a big selling point to us, but just getting the enterprise product licensed was way too much for us.  In the quoting phase we asked for some custom development and the cost of that was even 10% of the overall quote
[26-May-2011 11:44:56] <Hackman238> jmp242: Zenoss turns down a large segment of business by being ignorant to the impact of enterprise costs to small and medium businesses.
[26-May-2011 11:44:58] <jmp242> rmatte: and we'd grab a 1 or 2 call support pack pretty quickly
[26-May-2011 11:45:09] <jmp242> if it was in the $2k range yearly
[26-May-2011 11:45:14] <jmp242> and that's on Core
[26-May-2011 11:45:19] <nyeates> There are so many threads going on in here over the last 45 mins. I think that I cant begin to handle them. I hope this was a useful bit of your alls time and a sign of good things to come.
[26-May-2011 11:45:21] <dhopp> wasn't even..not was even
[26-May-2011 11:45:41] <jmp242> Thanks nyeates...
[26-May-2011 11:45:43] <rmatte> jmp242: when they came back to us with an 80k pricetag for Zenoss Enterprise and Datacenter Insight for 1009 devices, we did a double take to make sure we weren't dreaming lol
[26-May-2011 11:46:03] <jmp242> Yea
[26-May-2011 11:46:10] <jmp242> I wonder what support they're looking at in that
[26-May-2011 11:46:16] <nyeates> How abouts we let the convo's continue, and see where stuff goes. Also if you have TECHNICAL questions, bring them now as there are some devs in room
[26-May-2011 11:46:17] <jmp242> I mean, are they configuring it for you?
[26-May-2011 11:46:45] <Jane_Curry> Could we pick up athread about the example ZenPack from the last IRC......
[26-May-2011 11:46:51] <rmatte> jmp242: and that was for the SILVER package, where you get 4 phonecalls and 8 hour response time on emails
[26-May-2011 11:46:59] <nyeates> Jane: go for it
[26-May-2011 11:47:30] <nyeates> Jane: give background....not sure if the person who was here last time is here
[26-May-2011 11:47:44] <rmatte> nyeates: that's what IRC logs are for -> "I think that I cant begin to handle them"
[26-May-2011 11:47:48] <rmatte>
[26-May-2011 11:47:53] <dpetzel> I just wanted to circle back on the issue folks were helping with before. Adding the sync() to the line before my commit() has resolved the conflict errors in my dmd script. I still want to explore the session and try block, but for now that seemed to do the trick
[26-May-2011 11:47:54] <Jane_Curry> Finally pulled the example ZenPack as it was suggested last IRC that it might be a start towards good practice examples....
[26-May-2011 11:48:10] <rmatte> dpetzel: excellent
[26-May-2011 11:48:21] <Jane_Curry> It shows as "not tested" on the website
[26-May-2011 11:48:45] <rmatte> Jane_Curry: I don't think that pack has been touched for a while... It probably hasn't even been updated for 3.x
[26-May-2011 11:49:06] <nyeates> my savior, zenoss-logger :-)
[26-May-2011 11:49:19] <Jane_Curry> It only has a py2.4 version...
[26-May-2011 11:49:42] <Jane_Curry> .. and the install fails (even after renaming the egg)
[26-May-2011 11:50:06] <Jane_Curry> this ain't gonna help much with 3.0 UI good practices and examples, I fear
[26-May-2011 11:50:55] <nyeates> which pack referiung to?
[26-May-2011 11:51:09] <Hackman238> Jane_Curry: agreed. We need rocket or another Zenoss master to audit some packs and point out good examples of pristine code.
[26-May-2011 11:51:24] <Jane_Curry> ZenPacks.example.Techniques-1.4.1
[26-May-2011 11:51:49] <Jane_Curry> there was a debate last time that it might provide 3.0 UI help.......
[26-May-2011 11:51:50] <nyeates> Rocket will be helping on this general problem, as we QA and review through Janes doc
[26-May-2011 11:52:12] <nyeates> Hopefully we will see holes too, and suggest places where new documentation can be added
[26-May-2011 11:52:43] <magman> Have you guys looked at the latest ZenPack
[26-May-2011 11:53:06] <Jane_Curry> Another suggestion last time was to build a ZenPacks tutorial offering
[26-May-2011 11:53:13] <magman>
[26-May-2011 11:53:34] <Jane_Curry> There has been a little interest shown, but not enough, per se, to justify the load of work it would take
[26-May-2011 11:53:37] <magman> it pretty much follows best practice all the way
[26-May-2011 11:53:54] <magman> Facades, Routers, 3.0 UI, etc.
[26-May-2011 11:54:13] <rmatte> Jane_Curry: it probably has outdated skin files and such
[26-May-2011 11:55:57] <nyeates> Intersting about the OpenStack ZP Chet is making, i didnt know of this
[26-May-2011 11:56:08] <nyeates> Just a few days old :-)
[26-May-2011 11:56:42] <nyeates> cluther: any comments on if that ZenPack is a good example of 3.0 UI element interaction?
[26-May-2011 11:57:28] <cluther> nyeates: Yes, it and the CloudFoundry packs are good examples of best practices for ZenPacks compatible with Zenoss 3.0 or 3,1,
[26-May-2011 11:57:53] <rmatte> cluther: awesome, I'll have to take a look then.
[26-May-2011 11:58:37] <nyeates> cluther: is its skeleton or basic logic based off any prior zenpack, or made from scratch?
[26-May-2011 11:58:43] <cluther> The OpenStack pack even shows never before seen examples on how to tie your custom components into the Enterprise dynamic view stuff if you're running Zenoss Enterprise.
[26-May-2011 11:59:20] <Jane_Curry> I don't even see it on the ZenPacks site - give me a clue - not under "O" for OpenStack
[26-May-2011 11:59:47] <rocket>
[26-May-2011 11:59:52] <nyeates> No its not on the main site Jane_Curry, I just found out about it myself
[26-May-2011 11:59:54] <cluther> Jane_Curry: It's a spare-time project of mine that hasn't really been released yet.
[26-May-2011 12:00:01] [disconnected at Thu May 26 12:00:01 2011]
[26-May-2011 12:00:02] [connected at Thu May 26 12:00:02 2011]
[26-May-2011 12:00:20] [zenoss-logger (logger bot) has joined #zenoss]
[26-May-2011 12:00:48] <nyeates> Also
[26-May-2011 12:01:10] <Jane_Curry> OK - can see it there - thought I was missing something on the ZenPacks page - I'll grab it if I may?
[26-May-2011 12:01:24] <nyeates> I am behind on sync'ing up our ZenPack listing with what has been submited....feel free to yell at me :-)
[26-May-2011 12:01:27] <rocket> Jane_Curry: its fair game there ..
[26-May-2011 12:02:06] <cluther> Some of the patterns used in those packs have been around for a while, but only implemented in Enterprise ZenPacks.
[26-May-2011 12:03:00] <Jane_Curry> but presumably this stuff will work on Core??
[26-May-2011 12:03:05] <cluther> Jane_Curry: Absolutely.
[26-May-2011 12:04:00] <Jane_Curry> Many thanks, Chet
[26-May-2011 12:04:17] <Hackman238> cluther: I'd be interested in testing or assisting with your openstack Zenpack
[26-May-2011 12:04:33] <Hackman238> cluther: I say so since I have open in development for a customer now
[26-May-2011 12:04:45] <Hackman238> *one
[26-May-2011 12:05:17] <Hackman238> cluther: Idea being to cloudify and fully automate Zenoss
[26-May-2011 12:05:28] <cluther> Hackman238: I had a feeling.. if you look closely you'll see that it currently is only doing consumer perspective monitoring of cloud servers (nova).
[26-May-2011 12:05:38] <cluther> I did test it against Rackspace's Cloud Servers though.
[26-May-2011 12:05:56] <nyeates> Matt Ray had done some neat interactions between chef and openstack....wondering if zenoss can be wound into that stack
[26-May-2011 12:06:12] <Hackman238> nyeates: Thats what Im doing ;-)
[26-May-2011 12:06:41] <nyeates> cluther: so its using the "public facing" http rest API?
[26-May-2011 12:06:46] <nyeates> of openstack
[26-May-2011 12:06:57] <Hackman238> nyeates: Chef goes to add clients, sees if a collector is free, if no collector spool up a new collector, automatically configure it, add the client, and reverse the process when collectors are free'd up
[26-May-2011 12:07:02] <cluther> nyeates: yes, that's right
[26-May-2011 12:07:21] <cluther> Hackman238: Awesome.
[26-May-2011 12:07:34] <Hackman238> nyeates: Result is a cloud of collectors for huge clients- ;-)
[26-May-2011 12:08:14] <nyeates> wow, spin up a new collector each!
[26-May-2011 12:08:17] <Hackman238> nyeates: This is why Zenoss will die without supporting the community. Many of my enterprise size customers use core and pay me to build solutions because its cheaper than ent!
[26-May-2011 12:08:36] <squig> some one will fork
[26-May-2011 12:09:09] <nyeates> I am going to deem the dev meeting officially ended. Thanks all for participating, including our CEO, and see you next time. Continue convos if you wish.
[26-May-2011 12:09:15] <Hackman238> nyeates: With an ideal pricing model, I'd have little or no business in Zenoss consulting because there would be no market
[26-May-2011 12:09:22] <Jane_Curry> thanks, Nick
[26-May-2011 12:09:27] <Hackman238> Thanks again Nick
[26-May-2011 12:10:06] <rmatte> Hackman238: I'm sure there would still be some interest
[26-May-2011 12:10:22] <Hackman238> rmatte: Some yes, but not enough for it to be profitable.
[26-May-2011 12:10:42] <nyeates> I for one am interested in a chef openstack zenoss tie up
[26-May-2011 12:10:43] <Hackman238> rmatte: Zenoss's failure to capitalize on all the market makes my niche market.
[26-May-2011 12:10:52] <rmatte> true
[26-May-2011 12:11:00] <Hackman238> nyeates/cluther: I'll get you infos soon.
[26-May-2011 12:11:25] <rmatte> nyeates: then matt ray and chet are going to need to get together and make one hehe
[26-May-2011 12:11:28] <Hackman238> rmatte: And I'm telling them how to fix it so the product doesnt die. It'll take money out of my pocket, but I can't work like this forever.
[26-May-2011 12:11:43] <rmatte> Hackman238: yeh, for sure
[26-May-2011 12:12:07] <rmatte> Their pricing model blows my mind right now, they really really need to do something about it
[26-May-2011 12:12:12] <rmatte> I mean, it's worse than it was before
[26-May-2011 12:12:12] <dhopp> Hackman238: if you are giving money away I can give you my address :-P
[26-May-2011 12:12:13] <rmatte> lol
[26-May-2011 12:12:16] <Hackman238> rmatte: I'll get you some info on my openstack/chef setup too
[26-May-2011 12:12:26] <rmatte> Hackman238: cool
[26-May-2011 12:12:30] <nyeates> allright all, im out...later
[26-May-2011 12:12:31] <mattray> Hackman238: feel free to send it to me
[26-May-2011 12:12:35] <Hackman238> dhopp: Well not giving away money, I just dont want Zenoss to die as a product.
[26-May-2011 12:12:35] <rmatte> Hackman238: did you send me those IPSLA pack links?
[26-May-2011 12:12:45] <dhopp> Hackman238: I was kidding
[26-May-2011 12:12:47] <rmatte> I didn't see an email indicating a message
[26-May-2011 12:12:55] <Hackman238> rmatte: Doh! Sorry, I forgot. PM me a reminder please :-)
[26-May-2011 12:13:03] <rmatte> Hackman238: ok, will do
[26-May-2011 12:13:16] <Hackman238> rmatte: Busy busy!
[26-May-2011 12:13:22] <rmatte> I know the drill
[26-May-2011 12:14:35] <rmatte> sent
[26-May-2011 12:14:44] <Hackman238> rmatte: Thanks much
[26-May-2011 12:14:50] <rmatte> no no, thank you
[26-May-2011 12:15:03] <rmatte> It's going to make this upcoming IPSLA project of ours much easier
[26-May-2011 12:15:16] <rmatte> well, much easier on me anyways
[26-May-2011 12:15:18] <rmatte> less coding
[26-May-2011 12:15:28] <rmatte>
[26-May-2011 12:15:28] <Hackman238> rmatte: thats my job :-) The only thing one can really do is contribute to the world.
[26-May-2011 12:15:35] <rmatte> agreed
[26-May-2011 12:15:51] <rmatte> management here was originally against contributing ZenPacks that I developed while working here...
[26-May-2011 12:15:57] <rmatte> until I explained that you give and you get
[26-May-2011 12:16:00] <rmatte> it's a 2 way street
[26-May-2011 12:16:27] <rmatte> We're mostly a windows shop here, so not much open source mentality floating around
[26-May-2011 12:16:36] <rmatte> They were an all windows shop until our group came along
[26-May-2011 12:17:18] <rmatte> I'm one of three people that know Linux in the building.  One is a pure UNIX admin by trade (AIX, HP-UX, Solaris), and the other is a web developer.
[26-May-2011 12:17:40] <rmatte> I end up getting all of the really fun professional services jobs though
[26-May-2011 12:17:49] <Hackman238> rmatte: Thats exactly right. Additionally no one person can know everything.
[26-May-2011 12:17:55] <rmatte> My most recent project is migrating a CVS server from a windows server to Linux
[26-May-2011 12:18:07] <Hackman238> rmatte: I can bet you my openstack zenpack can borrow code from chets and chets from mine ;-)
[26-May-2011 12:18:14] <rmatte> setting up AD auth on the linux box, also moving a bunch of things like bugzilla, wiki, etc... off of their old linux box to the new one
[26-May-2011 12:18:22] <Hackman238> rmatte: nice
[26-May-2011 12:18:33] <rmatte> then duplicating horrid integration that they had between bugzilla and cvs
[26-May-2011 12:18:58] <rmatte> they had the windows server sending an email to a gmail account, then the linux server scraping the emails from there and generating an entry in bugzilla
[26-May-2011 12:19:06] <rmatte> I modified the process so that it's all internal to the box
[26-May-2011 12:19:15] <rmatte> I had to use the existing scripts though so it's still done via email
[26-May-2011 12:19:16] <rmatte> but whatever
[26-May-2011 12:19:52] <rmatte> Hackman238: probably, you should talk to Chet about that
[26-May-2011 12:19:57] <rmatte> so that you're not reinventing the wheel
[26-May-2011 12:20:57] <rmatte> My next project for that same customer is doing an in-house installation of Zenoss for them
[26-May-2011 12:21:02] <rmatte> and then training them how to use it
[26-May-2011 12:21:11] <rmatte> should be fun
[26-May-2011 12:25:11] <Hackman238> rmatte: Nice
[26-May-2011 12:34:05] <kokey> rmatte: should charge zenoss PS rates for it ;-)
[26-May-2011 12:37:36] <Hackman238> nyeates: whats the dead line?
[26-May-2011 12:38:02] <Hackman238> nyeates: What code level changes will need to be in place to ensure 3.1 pack compatability with 3.1.1?
[26-May-2011 12:38:50] <kokey> what's a good way to make a report that shows the current critical events?
[26-May-2011 12:38:57] <kokey> to e-mail out
[26-May-2011 12:40:50] <Sam-I-Am> moo.
[26-May-2011 12:41:53] <jmp242> Sam-I-Am: do you have some special bovine affinity?
[26-May-2011 12:42:09] <Sam-I-Am> cows are trendy
[26-May-2011 12:42:19] <Sam-I-Am> well, until they killed off gateway computer
[26-May-2011 12:42:20] <jmp242> I always think gateway circa 2000 ...
[26-May-2011 12:42:28] <jmp242> lol
[26-May-2011 12:43:09] <Hackman238> LOL
[26-May-2011 12:43:16] <Sam-I-Am> Hackman238: sup
[26-May-2011 12:43:52] stevenhirsch_ is now known as stevenhirsch
[26-May-2011 12:43:55] <Hackman238> Sam-I-Am: Samething as everyday- trying to take over the world!
[26-May-2011 12:44:14] <Sam-I-Am> uh oh, we're in competition
[26-May-2011 12:44:22] <Hackman238> Sam-I-Am: *cough* Rather take over Zenoss Inc's entire  market
[26-May-2011 12:44:30] <Hackman238> :-)
[26-May-2011 12:44:41] <Sam-I-Am> with hackmanoss?
[26-May-2011 12:44:48] <Sam-I-Am> zenoss fork?
[26-May-2011 12:45:11] <Hackman238> Sam-I-Am: Na, dont need a fork when they ignore all small and medium business needs with poor pricing structure.
[26-May-2011 12:45:26] <Hackman238> ;-)
[26-May-2011 12:46:24] <jmp242> Hackman238: what would be ok that niche though would be if you either sold or made available w/ support costs zenpacks etc you do. I hate to see the wheel reinvented just because people didn't know it had already been done
[26-May-2011 12:46:45] <Hackman238> Sam-I-Am: I told them in the CEO/DEV meeting today how to fix the problem. Its up to them to do the math and look at market logic.
[26-May-2011 12:47:06] <jmp242> I.e. try and not do them so you can't re-use or re-sell or contribute to the Community ... IMO anyway
[26-May-2011 12:47:16] <Hackman238> jmp242: I do both.
[26-May-2011 12:47:19] <Sam-I-Am> the pricing is a bit weird
[26-May-2011 12:47:25] <Sam-I-Am> i only slip in because i'm .edu
[26-May-2011 12:47:32] <Sam-I-Am> otherwise its wayyyy out of reach
[26-May-2011 12:47:57] <Hackman238> jmp242: Some customers just dont like the idea a product I develop on their dime might help others. On the other hand, some dont mind at all.
[26-May-2011 12:48:17] <jmp242> Do you list the publicly available (i.e. for sale) ones anywhere?
[26-May-2011 12:48:17] <Hackman238> jmp242: I dont sell packs, I only sell time.
[26-May-2011 12:48:59] <jmp242> oh, see when I contract for zenpack developlemt (lol I only have done one) I make it part of the contract to put it in the community
[26-May-2011 12:48:59] <Hackman238> jmp242: If you need something, I'll make it. If one exists and you just want help or need tweaking, I'll do that.
[26-May-2011 12:49:40] <Hackman238> jmp242: I leave it up to the customer. Most are okay with recontributing to the community.
[26-May-2011 12:49:40] <jmp242> Zenoss should consider (if they don't have one I've missed) a partner page like other projects have
[26-May-2011 12:49:40] <Hackman238> jmp242: Agreed.
[26-May-2011 12:50:08] <jmp242> I only found devs for this because I know the community really well
[26-May-2011 12:50:20] <jmp242> I did try using the elance etc but no one on there knows Zenoss, and I expect it must be hard for a newbie with Zenoss who needs help
[26-May-2011 12:50:23] <Hackman238> jmp242: Yep. Sorry I didnt get back to you more quickly- I was hosed for time that month :-)
[26-May-2011 12:50:25] <Sam-I-Am> some corporations dont like people developing thing which get into the public
[26-May-2011 12:50:29] <Sam-I-Am> at least not on their dime
[26-May-2011 12:50:41] <jmp242> That's ok Hackman238
[26-May-2011 12:50:45] <jmp242> I got it done, and done well
[26-May-2011 12:50:53] <kokey> yeah but from a support perspective it's better if something has been contributed back to the community
[26-May-2011 12:50:55] <Hackman238> Sam-I-Am: Very true- I leave it up to them and the decision listed in the SOW
[26-May-2011 12:50:58] <jmp242> and I have you on my list for the next time I get some approval for stuff like that
[26-May-2011 12:51:05] <kokey> better value than having to keep some people on a retainer or something to maintain it
[26-May-2011 12:51:27] <jmp242> I would suggest the corporation may be missing the point of how a good OSS community works, but hey
[26-May-2011 12:51:30] <jmp242> it's their money
[26-May-2011 12:51:38] <Sam-I-Am> back to the conf i'm at...
[26-May-2011 12:51:39] <jmp242> I'm ok with them wasting it
[26-May-2011 12:51:50] <Hackman238> jmp242: I explained their model failure and how they can fix it to bennifit everyone.
[26-May-2011 12:51:59] <jmp242> Oh yea
[26-May-2011 12:52:00] <jmp242> I was here
[26-May-2011 12:52:29] <Hackman238> jmp242: More moeny for them, less stress for me, better support, more growth, investment oppertunity, reduced zenoss support operations cost, capitalization on small markets.
[26-May-2011 12:52:48] <jmp242> yea, I'm not an MBA, but I can see them leaving money on the table
[26-May-2011 12:53:09] <Hackman238> jmp242: No MBA here either- I'm just an experienced business man.
[26-May-2011 12:53:12] <jmp242> Not that they can't fill that via partners if they don't want to deal - Microsoft does that a lot
[26-May-2011 12:53:24] <jmp242> and that still builds value up and down the whole chain
[26-May-2011 12:53:32] <jmp242> but just leaving a void is a bad idea IMO
[26-May-2011 12:53:40] <Hackman238> jmp242: Very bad.
[26-May-2011 12:54:17] <Hackman238> jmp242: Especially with OSS- they are asking for mutiny by leaving a gap.
[26-May-2011 12:54:27] <jmp242> If zenoss forks, I'd be tempted to look at the fully OSS fork - or if it dies for non enterprise, PRTG looks very promising in my price range for instance
[26-May-2011 12:54:36] <Hackman238> jmp242: With closed source people are trapped. OSS people will turn on you and fork.
[26-May-2011 12:54:39] <jmp242> I hate to through out the 3+ years of knowledge though
[26-May-2011 12:55:12] <Hackman238> jmp242: no worries- if a fork happens I imagine cross compatability wont be an immediate issue.
[26-May-2011 12:55:52] <Hackman238> jmp242: I blew up at Zenoss management about all of this which is why we had this meeting today.
[26-May-2011 12:56:44] <squig> when i said fork the community manager ran away
[26-May-2011 12:57:01] <Hackman238> jmp242: Its like a kid who refuses to acknowledge speeding 30 over the limit increases the chances of terminal sumamry. Zenoss does this by ignoring this gap.
[26-May-2011 12:57:15] <rmatte> There are enough experienced developers that a fork probably would be possible.  The release cycle would be much slower than it is now but it would work.
[26-May-2011 12:57:58] <rmatte> I doubt that there would be a shortage in funding for it either
[26-May-2011 12:58:16] <jmp242> Naw
[26-May-2011 12:58:29] <rmatte> but hopefully it never comes to that
[26-May-2011 12:58:31] <jmp242> especially if they set up a donation / pledge / whatever I'm thinking of
[26-May-2011 12:58:34] <Hackman238> rmatte: Would be so very easy. This is why its all so very dangerous for Zenoss
[26-May-2011 12:58:35] <rmatte> yeh
[26-May-2011 12:58:40] <jmp242> Zenoss is very important for us
[26-May-2011 12:59:02] <jmp242> but it's Core. Enterprise is too expensive, as well as mostly stuff we won't use
[26-May-2011 12:59:08] <rmatte> yeh
[26-May-2011 12:59:25] <rmatte> That removal of relstorage from Core really irked me (and others), I hope they reconsider that.
[26-May-2011 12:59:25] <Hackman238> I'm not trying to scare Zenoss or anyone here- I look at problems and try to fix them.
[26-May-2011 12:59:56] <Hackman238> rmatte: agreed
[26-May-2011 12:59:59] <rmatte> They kind of did it incognito too hoping no one would notice.
[26-May-2011 13:00:14] <squig> I dont imagine if they just charged for support and extra zen packs and charged the enterprise price then they would loose any revenue
[26-May-2011 13:00:23] <Hackman238> rmatte: The CEO accidently validated my statement about it. The idea was to limit Core scalability.
[26-May-2011 13:00:28] <squig> big customers can pay and dont care
[26-May-2011 13:00:35] <dhopp> I'm not sure PRTG is really that "cheap"'s priced per sensor..
[26-May-2011 13:00:37] <rmatte> Hackman238: yeh, I noticed that hehe
[26-May-2011 13:00:42] <dhopp> and "One sensor monitors one network service, one URL, one network connection, one port of a switch, one NetFlow export stream, one CPU load, one disk drive, etc."
[26-May-2011 13:00:48] <dhopp> a 24 port switch is 24 sensors!
[26-May-2011 13:01:04] <Hackman238> rmatte: He ignored my questions since they'll trap him with a comitting reply
[26-May-2011 13:01:34] <kokey> yeah i figure it would help if the pricing covers device per zenpack rather
[26-May-2011 13:01:34] <rmatte> Hackman238: yeh
[26-May-2011 13:01:34] <Hackman238> He came in here thinking we would be fools.
[26-May-2011 13:01:38] <kokey> helps with focus too
[26-May-2011 13:01:47] <squig> Hackman238, i cant imagine that is the case
[26-May-2011 13:02:11] <Hackman238> squig: How so?
[26-May-2011 13:02:40] <squig> most ceos of small enterprises understand their business
[26-May-2011 13:03:14] <Hackman238> squig: Ill agree- very true. This seems to be an exception.
[26-May-2011 13:03:39] <rmatte> The way I feel lately is that they would disable the ability to login to core if it would give them an edge for Enterprise.  Obviously a large exageration, but it's starting to get out of hand.  They are pricing the product incredibly badly and then trying everything except for adjusting the price to attract new customers.
[26-May-2011 13:03:52] <squig> i was pretty impressed to see him in here, I was just lurking wasnt aware that that was on
[26-May-2011 13:04:13] <Hackman238> squig: Agreed- but his attendance was forced.
[26-May-2011 13:04:16] <rmatte> Even if it means pulling things away from core that shouldn't be.
[26-May-2011 13:04:23] <squig> they may not want to attract small customers
[26-May-2011 13:04:25] <Hackman238> squig: Very forced.
[26-May-2011 13:04:29] <rmatte> squig: agreed, I was very happy that he took the time to talk to us.
[26-May-2011 13:04:38] <squig> supporting small customers is harder than supporting big ones
[26-May-2011 13:04:39] <Hackman238> rmatte: Absolutely
[26-May-2011 13:04:49] <squig> with a profit ratio
[26-May-2011 13:04:59] <Hackman238> squig: not at all- its a matter of excellent community
[26-May-2011 13:05:01] <squig> with a lower profit ratio
[26-May-2011 13:05:22] <rmatte> squig: a lot of the smaller customers don't want support at all
[26-May-2011 13:05:34] <rmatte> squig: we'd do fine with zero support here, I already know Zenoss inside and out
[26-May-2011 13:05:47] <rmatte> I've been successfully managing 15 Zenoss instances for over 2 years now
[26-May-2011 13:06:03] <Hackman238> squig: For example, most of RS's tickets are for minor problems- things we couldnt find utilities or docs for.
[26-May-2011 13:06:08] <rmatte> ...and then eventually if we got much larger we'd consider paying for support at that point
[26-May-2011 13:06:22] <Hackman238> squig: We'd rather fix it ourself than wait for a ticket- ticket is last resort.
[26-May-2011 13:06:22] <rmatte> the fact that support is absolutely mandatory in the pricing is annoying
[26-May-2011 13:06:33] <squig> its very common with software
[26-May-2011 13:06:44] <Hackman238> squig: Agreed.
[26-May-2011 13:07:03] <rmatte> I've actually heard several Enterprise users say that they often get better support in this IRC channel than they do from Zenoss
[26-May-2011 13:07:15] <Hackman238> squig: the model is unsustainable in my perspective. I fear it'll spawn OSS competition to the product.
[26-May-2011 13:07:36] <Hackman238> rmatte: *cough* :-)
[26-May-2011 13:07:37] <squig> there already is in one way or another
[26-May-2011 13:07:50] <rmatte>
[26-May-2011 13:08:19] <Hackman238> squig: Agreed, but I mean with the feature set and code set to grab away customers and market from zenoss directly
[26-May-2011 13:09:06] <squig> the last big install i worked with >20,000 nodes went with ganglia
[26-May-2011 13:09:14] <rmatte> ah crap, I'm 10 minutes late for a meeting
[26-May-2011 13:09:17] <rmatte> talk to you guys later
[26-May-2011 13:09:28] <dhopp> rmatte: your fired
[26-May-2011 13:09:29] <dhopp> :-P
[26-May-2011 13:09:35] <Hackman238> squig: Two of my businesses are retail computer repair offices. We service tens of thousands of devices a year with a very high turn over rate and satisfaction rate.
[26-May-2011 13:09:39] <rmatte> they can fire me, but good luck to them
[26-May-2011 13:09:47] <Hackman238> rmatte: Later
[26-May-2011 13:10:04] <Hackman238> squig: Computer service is a hard industry to make profitable.
[26-May-2011 13:10:44] <Hackman238> squig: We do it by offering a soluition which reduces our support costs 50 fold, customer expense far below competitors, and ensures QC.
[26-May-2011 13:11:20] <Hackman238> squig: All in house with smart pricing model, repair model and volume market left out to dry by competition.
[26-May-2011 13:11:53] <Hackman238> Zenoss needs to optimize to reduce supprot costs to take on the smaller, volume, market
[26-May-2011 13:12:05] <Hackman238> ;-)
[26-May-2011 13:12:22] <squig> or find partners
[26-May-2011 13:12:30] <Hackman238> squig: Yep.
[26-May-2011 13:12:32] <squig> that want to do that for them
[26-May-2011 13:12:40] <Hackman238> squig: Exactly.
[26-May-2011 13:13:26] * mattray sighs
[26-May-2011 13:13:38] <Hackman238> mattray: Whats the sigh for?
[26-May-2011 13:14:10] <Hackman238> BTW, sorry all if my business jargen is poor, I have no degree in that field.
[26-May-2011 13:15:08] <mattray> I'd had similar observations to some of the comments
[26-May-2011 13:15:45] <mattray> open source works well when partners fill in the gaps
[26-May-2011 13:15:52] <Hackman238> squig: Whats ganglia?
[26-May-2011 13:15:56] * mattray goes back to #Chef
[26-May-2011 13:16:01] <Hackman238> mattray: Yep.
[26-May-2011 13:16:29] <mattray> Hackman238: it's a monitoring application for clusters and large scale installs. Twitter and Facebook both use it
[26-May-2011 13:16:46] <mattray> more focused on aggregate results
[26-May-2011 13:16:51] <Hackman238> mattray: Very cool.
[26-May-2011 13:17:32] <Hackman238> mattray: I'm hoping this new cloudification direction with zenoss will help me take on clients in that size catagory
[26-May-2011 13:17:32] <mattray> it's an old tool, been around for a long time
[26-May-2011 13:17:56] <mattray> yeah, I'm happy to help since that's a huge focus of Chef users
[26-May-2011 13:18:05] <mattray> but not exclusively
[26-May-2011 13:18:14] <Hackman238> mattray: no sweat
[26-May-2011 13:18:32] <mattray> and we work with our partners a lot
[26-May-2011 13:18:53] <mattray> but enough trolling
[26-May-2011 13:21:35] <jmp242> dhopp PRTG has an unlimited sensor price though for under $10k (I was afk a bit
[26-May-2011 13:29:07] <dhopp> Hackman238: ganglia is a monitoring tool but it was build with large clusters and grids in mind
[26-May-2011 13:29:26] <dhopp> sorry..didn't see mattray responded
[26-May-2011 13:29:39] <dhopp> having trouble multi tasking today
[26-May-2011 13:33:49] <Hackman238> dhopp: No sweat :-)
[26-May-2011 13:47:52] <xuru> howdy
[26-May-2011 13:48:51] <xuru> When I try to delete a bunch of events, I get the following error:
[26-May-2011 13:48:54] <xuru> "The server reported the following error:
[26-May-2011 13:48:54] <xuru> OperationalError (1153, "Got a packet bigger than 'max_allowed_packet' bytes")"
[26-May-2011 13:49:55] <Hackman238> xuru: ah had this the other day
[26-May-2011 13:50:13] <Hackman238> xuru: add a line in /etc/my.cnf to increase the max packet size to 64MB
[26-May-2011 13:50:23] <Hackman238> xuru: then restart mysqld
[26-May-2011 13:50:34] <xuru> ah, ok, thanks... thought that might be the case
[26-May-2011 13:50:41] <Hackman238> xuru: no sweat
[26-May-2011 13:50:54] <Hackman238> xuru: v3.1.7 or v3.1?
[26-May-2011 13:51:42] <xuru> Version: 3.1.0-0
[26-May-2011 13:51:57] <Hackman238> xuru: Gotcha. Should work. Let me know :-)
[26-May-2011 13:52:05] <xuru> thanks
[26-May-2011 13:55:26] <xuru> hmm... still getting it.  probably has something to do with it being 40k events
[26-May-2011 13:55:48] <xuru> any easy way to delete a lot of events?  perhaps with a zendmd script
[26-May-2011 13:57:05] <xuru> I have some ESX servers spewing out crap (I think I solved it) but I have some 326k events that are seriously slowing down the system
[26-May-2011 14:02:35] <Hackman238> xuru: events in history or events creating alerts flooding zenactions?
[26-May-2011 14:02:42] <Hackman238> xuru: you can truncat tables
[26-May-2011 14:02:51] <Hackman238> *truncate
[26-May-2011 14:03:14] <Hackman238> xuru: did yo reload mysqld and restart zenhub?
[26-May-2011 14:20:05] <xuru> sorry, had a call
[26-May-2011 14:20:21] <xuru> Hackman238: yeah, I restarted the whole stack
[26-May-2011 14:20:56] <Hackman238> xuru: Hum.
[26-May-2011 14:21:04] <xuru> I have 326k live events and because it's got the date/time in the message, it's a unique event
[26-May-2011 14:21:24] <Hackman238> I'd truncate tables if I were you. Otherwise write a sql statement to delete those events.
[26-May-2011 14:21:39] <Hackman238> Also, put in a transform to drop that event in the future.
[26-May-2011 14:30:36] <xuru> yeah, already put in the transform
[26-May-2011 14:32:37] <Hackman238> xuru: Gotcha.
[26-May-2011 14:35:33] <dhopp> Is there an easy way to have zenoss suppress an event from the dashboard until it has happened so many times?  I know you can do this with the alert e-mail..but I'm thinking about how our NOC works and the "big board" idea
[26-May-2011 14:37:11] <Hackman238> dhopp: You could use event severity escalation to make the severity increase over x number of occurances.
[26-May-2011 14:37:44] <Hackman238> dhopp: But you would need to filter the event console on severity. For the dash you'd need to find a way to filter by that key severity
[26-May-2011 14:38:11] <Hackman238> dhopp: Off the top of my head I cant think of a way to do it otherwise- Im sure its doable
[26-May-2011 14:44:28] <dhopp> Hackman238: thanks…I just a had a meeting about the zenoss implementation and it came up
[26-May-2011 14:45:06] <Hackman238> dhopp: NP
[26-May-2011 14:45:24] <dhopp> Hackman238: our existing monitoring has "suspect" and "error".  An alert goes suspect for 3 polls before it's an error.  The suspect still shows up, but the NOC doesn't usually act on it until it's an error
[26-May-2011 14:45:49] <Hackman238> dhopp: Sounds like the NOC is what needs fixing, not zenoss
[26-May-2011 14:45:58] <dhopp> Hackman238: it's a way to not have the NOC unnecessarily taking action for something that is just a hiccup
[26-May-2011 14:46:02] <Hackman238> dhopp: *cough administrative solution*
[26-May-2011 14:46:11] <Hackman238> dhopp: Ah gotcha
[26-May-2011 14:47:00] <dhopp> Hackman238: or having me woken up at 2:30 AM because the monitoring system decided to be ornary
[26-May-2011 14:47:21] <xuru> Anyone know how to do filtering using dmd.ZenEventManager.getEventList?
[26-May-2011 14:47:35] <locohost> i attempted to add a custom report that aggregated the bandwidth of about 200 interfaces
[26-May-2011 14:47:38] <Hackman238> xuru: I wouldnt dmd that process
[26-May-2011 14:47:45] <Hackman238> xuru: It'll take forever
[26-May-2011 14:47:51] <locohost> anyway, zenoss was not a big fan and now every time i click on it, zeoctrl locks up
[26-May-2011 14:47:55] <locohost> i cant even delete it, any idea?
[26-May-2011 14:48:00] <dhopp> Hackman238: I believe a lot of our "temporary" alerts (i.e. alerts that go suspect and but don't actually error) are because the backend database is locked (it's using mysql and myisam so there is table level locking)
[26-May-2011 14:48:11] <Hackman238> xuru: You're better off mysql -u root -p -d events and deleting it out of there with a select
[26-May-2011 14:48:18] <xuru> Hackman238: that's why I was hoping to filter it to grab just the events I want
[26-May-2011 14:48:25] <dhopp> Hackman238: so I think some of those will get cleared up hopefully but moving to a monitoring system that doesn't suck :-)
[26-May-2011 14:48:47] <xuru> Hackman238: not sure how to do that though
[26-May-2011 14:48:53] <xuru> not much of a sql guy
[26-May-2011 14:48:54] <Hackman238> dhopp: Gotcha. Moving away from Zenoss?
[26-May-2011 14:49:01] <Hackman238> xuru: Ah gotcha.
[26-May-2011 14:49:11] <Hackman238> xuru: pm me event details
[26-May-2011 14:49:14] <dhopp> Hackman238: me moving away from zenoss?
[26-May-2011 14:49:26] <dhopp> Hackman238: I'm trying to move TO zenoss
[26-May-2011 14:49:48] <Hackman238> dhopp: Oh gotcha. forgive me, Im sure you told me that in the past but my memory wraps fast.
[26-May-2011 14:50:09] <xuru>
[26-May-2011 14:52:07] <dhopp> Hackman238: no problem…our existing monitoring is home grown and relies heavily on screen scraping (i.e. retrieve a web page and look for key words).  Our trending is extremely lacking because we get so little data.  On most of our servers I couldn't even begin to tell you what the disk i/o does over time
[26-May-2011 14:53:07] <Hackman238> dhopp: Ah, i recall that now.
[26-May-2011 14:53:20] <dhopp> we get quite a bit of data from the application level (through jmx) and we get cpu/memory utilization through snmp…but that's really about it
[26-May-2011 14:53:40] <dhopp> networking has their own trying to correlate events that could be related to networking is a PITA
[26-May-2011 14:57:03] <Hackman238> dhopp: yikes!
[26-May-2011 14:57:46] <Hackman238> dhopp: you should use zenoss and ipsla to help correlate network:server:app evetns
[26-May-2011 15:00:13] <dhopp> Hackman238: all I need is an ipsla zenpack that works in 3.1.0 :-P
[26-May-2011 15:01:08] <Hackman238> dhopp: Its comming
[26-May-2011 15:01:11] <Hackman238> :-)
[26-May-2011 15:03:06] <jmp242> How does the ipsla help you correlate network:server:app events?
[26-May-2011 15:03:15] <locohost> any idea about deleting a custom report that cannot be deleted via gui?>
[26-May-2011 15:03:59] <locohost> hey,, look what i found http://d01mtrending01:8080/manage
[26-May-2011 15:05:33] <jmp242> you found the Zope interface?
[26-May-2011 15:05:37] <jmp242> not terribly amazing
[26-May-2011 15:06:13] <Hackman238> jmp242: LOL
[26-May-2011 15:06:23] <jmp242> sorry for the snark
[26-May-2011 15:06:36] <Hackman238> jmp242: Its doesnt, but using ipsla can help you figure if its your app that sucks or if its the network.
[26-May-2011 15:06:46] <Hackman238> jmp242: plus it'll all be in zenoss.
[26-May-2011 15:07:01] <jmp242> I'm fighting a recalcitrant laptop, a MS webcam and tryinf to find if I can get decent video from EVO video conferencing at a decent resolution from a PC
[26-May-2011 15:08:30] <Hackman238> jmp242: Never fun
[26-May-2011 15:15:52] <jmp242> And now I have a user who needs Acrobat to edit whitespace out of an Inventor print and can't use the Terminal service one because they can't upload the file to the file server . . .
[26-May-2011 15:16:50] <Hackman238> jmp242: Thats a double face palm
[26-May-2011 15:28:41] <rmatte> hmmm, swollen lymph node, not good
[26-May-2011 15:28:49] <rmatte> I anticipate becoming ill in the near future
[26-May-2011 15:32:23] <dhopp> rmatte: vodka
[26-May-2011 15:33:15] <Hackman238> rmatte: Never fun- especially when you're busy LOL
[26-May-2011 15:33:22] <rmatte> yeh
[26-May-2011 15:34:13] <Hackman238> xuru: How'd that rm / -Rf go?
[26-May-2011 15:34:26] <Hackman238> xuru: J/K, did that delete statement clear up the issue?
[26-May-2011 15:34:51] <locohost> cant seem to see where to set thredholds by the interface, seems like i can only do it by deice class
[26-May-2011 15:34:54] <locohost> device class
[26-May-2011 15:34:59] <xuru> Hackman238: it's still going
[26-May-2011 15:35:11] <Hackman238> xuru: Wow. Okay
[26-May-2011 15:35:31] <locohost> for example, I may want to be alerted if any interface on the switch is over 150megabits but i dont want to know anything about hte uplink port unless its over 500mb or something
[26-May-2011 15:35:32] <Hackman238> locohost: Navigate to an interface and you'll see the template there
[26-May-2011 15:36:16] <Hackman238> locohost: Or maybe I misunderstand what you're looking for.
[26-May-2011 15:37:22] <locohost> hmm, maybe
[26-May-2011 15:37:25] <locohost> so, create a local copy?
[26-May-2011 15:37:30] <locohost> and then edit it and its only for that device?>
[26-May-2011 15:37:47] <locohost> balls, y eah, think ur right
[26-May-2011 15:37:48] <locohost> ethernetCsmacd     /Devices/Network/Switch/devices/bc-375-sw1/os/interfaces/GigabitEthernet0_20     Template for 64-bit interface counters. Must use SNMP v2c for it to work.
[26-May-2011 15:38:03] <Hackman238> locohost: if its a one off, yes
[26-May-2011 15:38:17] <locohost> perfect, that does it
[26-May-2011 15:38:19] <Hackman238> locohost: if its not then...hum. Let me think
[26-May-2011 15:38:59] <Hackman238> locohost: You could edit to apply a template based on the speed
[26-May-2011 15:38:59] <locohost> so, ill just set a theshold of .15% utilization xross the board (let me know if any gig E is running at 150+mbps and then hit up the port channels and interfaces I expect more traffic on at higher failes
[26-May-2011 15:40:01] <locohost> yeah, the regext in hte default does use the speed, but not always helpfull, i mean, the defailt is .75% utilization, and im sure we all have gigI interface that we do not want 750mbps on
[26-May-2011 15:40:49] <Hackman238> locohost: You'd be lucky if you can even hit 75% on gig copper
[26-May-2011 15:40:57] <Hackman238> locohost: And sustain it, I mean
[26-May-2011 15:43:10] <locohost> yeah, well, if its a 500,000$ cisco switch, to 500,000$ cisco switch, you will probably get close to that, heh
[26-May-2011 15:43:26] <locohost> but, yeah, im setting threshold to 150mbps xross the board, then ill stup up interfaces that bark
[26-May-2011 15:43:42] <locohost> for syure dont want to set .75 on my 10G interfaces, lol
[26-May-2011 15:43:59] <Sam-I-Am> moooo.
[26-May-2011 15:44:09] <nyeates> woof
[26-May-2011 15:44:10] <Hackman238> locohost: LOL yup
[26-May-2011 15:44:26] <Hackman238> old MacOS 7 "bark"
[26-May-2011 15:45:10] <Hackman238> nyeates: Whats the deadline for the zenpack contest?
[26-May-2011 15:45:25] <nyeates> ooh that reminds me to push it out
[26-May-2011 15:45:35] <Hackman238> nyeates: Also, anything comming up in 3.1.1 that'll make 3.1 zenpacks buckle?
[26-May-2011 15:45:40] <nyeates> it will gget pushed out a few weeks, maybe to end of june
[26-May-2011 15:45:54] <Hackman238> nyeates: Gotcha.
[26-May-2011 15:46:41] <nyeates> 3.1.1  should be same as 3,1 for zenpacks, as far as i know
[26-May-2011 15:46:48] <JohnnyNoc> has anyone figured out how (if possible at all) to get RRD to show the actual MAX value in the graphs versus an average?  I've changed the consolidation from AVERAGE to MAX and it's still showing an average for the max
[26-May-2011 15:46:49] <Hackman238> nyeates: Alrighty
[26-May-2011 15:47:05] <nyeates> 3.1.1 is pushed back b/c of ever-shifting prioritites
[26-May-2011 15:47:06] <Hackman238> JohnnyNoc: It's because its a rate
[26-May-2011 15:47:07] <nyeates> not sure on a due date
[26-May-2011 15:47:29] <JohnnyNoc> Hackman238 ok so the asnwer is no?
[26-May-2011 15:47:31] <JohnnyNoc> :/
[26-May-2011 15:48:02] <Hackman238> nyeates: Understandable. Also, can you give rachael a heads up I need a reply about a list of bugs I sent
[26-May-2011 15:48:36] <Hackman238> JohnnyNoc: Well if you use MAX in consolidation it will be the MAX if you print the value. However, to graph it, the value will fall between points often- this is because its a rate.
[26-May-2011 15:49:04] <nyeates> Hackman238: I....suppose....though Im not really sure if its my place since im not in the middle of that
[26-May-2011 15:49:13] <Hackman238> JohnnyNoc: Should be anyway. Can anyone else confirm that?
[26-May-2011 15:49:43] <JohnnyNoc> Hackman238 what do you mean by 'if i print the value'?
[26-May-2011 15:50:03] <Hackman238> nyeates: Can you pm me the name of someone so I can go around her? I just need to move along quickly is all.
[26-May-2011 15:50:11] <JohnnyNoc> i've got a COMMAND datasource setup that receives Nagios style output like STATUS=WARN| DATASOURCE=X
[26-May-2011 15:50:40] <JohnnyNoc> my expectations, especailly after making sure the datasource was a gauage, and the CF is MAX, that the MAX would in fact be a max, and not some sort of average or rate as you say
[26-May-2011 15:50:47] <Hackman238> JohnnyNoc: Meaning if you add a GPRINT of that value to the graph commands it'l be what you expect
[26-May-2011 15:51:43] <JohnnyNoc> Hackman238 here's the line when i look at 'graph commands'
[26-May-2011 15:51:45] <JohnnyNoc> GPRINT:DataRate-raw-max:MAX:max\:%5.2lf%s\j
[26-May-2011 15:52:09] <JohnnyNoc> i'm expecting to see the MAX value, a whole number
[26-May-2011 15:53:14] <Hackman238> JohnnyNoc: Hum. Im thinking whats happening is its value1 + value2 / 2 in all cases except when you're dead on the exact collection time.
[26-May-2011 15:54:36] <Hackman238> JohnnyNoc: Can you try GPRINT:DataRate-raw-max:last:LAST:last\:%5.2lf%s\j
[26-May-2011 15:55:16] <JohnnyNoc> when i changed the CF To LAST my graph was never generated
[26-May-2011 15:55:16] <Hackman238> JohnnyNoc: Not sure if that'll work or not- last is the last colelcted value, if I recall right. Let me consult the goole.
[26-May-2011 15:55:31] <Hackman238> JohnnyNoc: Really? Hum.
[26-May-2011 15:55:38] <Hackman238> JohnnyNoc: Let me google it
[26-May-2011 15:55:41] <JohnnyNoc> yes, but i'm not convinced that's what i need
[26-May-2011 15:55:54] <JohnnyNoc> i mean, everything i've read points to CF of MAX and that i'll never actually see my peaks
[26-May-2011 15:57:46] <Hackman238> JohnnyNoc: I think best case with max is you'll have value1 + value2 / 2 unless the rendered view falls exactly on a collection point
[26-May-2011 16:09:19] <JohnnyNoc> i'm sure toby is much smarter than muyself, but it doesn't seem that hard to be able to keep a max value
[26-May-2011 16:09:20] <JohnnyNoc>
[26-May-2011 16:09:31] <JohnnyNoc> thanks for the help nonetheless
[26-May-2011 16:10:13] <Hackman238> JohnnyNoc: Anytime.
[26-May-2011 16:11:09] <Hackman238> JohnnyNoc: Idea- get the max and get the last. take (max * 2) - last = now
[26-May-2011 16:11:25] <Hackman238> JohnnyNoc: No idea if that'll work :-)
[26-May-2011 16:12:21] <JohnnyNoc> haha
[26-May-2011 16:13:29] <JohnnyNoc> i'm really trying to avoid digging into the custom graph command, but i may not have a choice
[26-May-2011 16:13:42] <JohnnyNoc> but it's really discouraging knowing other people have wanted/tried the same thing apparently with no success
[26-May-2011 16:13:55] <JohnnyNoc> so, would i be wasting my time?  all signs point to yes
[26-May-2011 16:14:46] <Hackman238> JohnnyNoc: Unfortuantely. I cant escape the thought that I solved this once for someone.
[26-May-2011 16:15:27] <Hackman238> JohnnyNoc: Are you on the forums? I'll pm you a solution if I can find my notes on the problem.
[26-May-2011 16:15:37] <JohnnyNoc> Hackman238 yes, johnnynoc
[26-May-2011 16:15:48] <JohnnyNoc> i've looked on the forums and it doesn't look like anyone has a real answer
[26-May-2011 16:15:48] <Hackman238> JohnnyNoc: Alrighty.
[26-May-2011 16:15:49] <JohnnyNoc>
[26-May-2011 16:15:55] <JohnnyNoc> but it's much appreciated, srsly
[26-May-2011 16:16:31] <JohnnyNoc> Hackman238 do you know where i can find/change the RRA config (maybe it's not a config)
[26-May-2011 16:16:53] <JohnnyNoc> i see mention of diff RRA's being used, but not seeing anything about it in the GUI anywhere..  and admittedly haven't looked very hard on the server
[26-May-2011 16:18:11] <dhopp> JohnnyNoc: what version of Zenoss?
[26-May-2011 16:18:15] <JohnnyNoc> 2.5.3
[26-May-2011 16:19:06] <xuru> Hackman238: should it be down by now?
[26-May-2011 16:19:15] <dhopp> JohnnyNoc: I'm not sure the exact steps (I haven't seen 2.x interface in a while).  But It's defined in the collector…In 3.x it's under Advanced -> Collectors, select the collector and click edit
[26-May-2011 16:19:37] <f00fSteR> in order to upgrade 2.4
[26-May-2011 16:19:43] <f00fSteR> i have to upgrade selectively correct ?
[26-May-2011 16:19:53] <Hackman238> JohnnyNoc: Ah! got it
[26-May-2011 16:20:04] <JohnnyNoc> Hackman238 you better not be playing with my emotions!
[26-May-2011 16:20:07] <f00fSteR> anyone have a guide in order to copy an existising installation of zenoss to a test system
[26-May-2011 16:20:11] <f00fSteR> so i can run the upgrade
[26-May-2011 16:20:13] <f00fSteR> and push across
[26-May-2011 16:20:18] <Hackman238> xuru: It depends. Can you ssh to a fresh term and top it?
[26-May-2011 16:20:24] <JohnnyNoc> dhopp thanks thats it
[26-May-2011 16:20:29] <Hackman238> JohnnyNoc: Hereit is:
[26-May-2011 16:20:30] <Hackman238> CDEF:CurrentIveNCHits=iveNCHits-raw
[26-May-2011 16:20:30] <Hackman238> CDEF:LastIveNCHits=PREV(CurrentIveNCHits)
[26-May-2011 16:20:30] <Hackman238> CDEF:PrevIveNCHits=PREV(LastIveNCHits)
[26-May-2011 16:20:30] <Hackman238> CDEF:DiffIveNCHits=CurrentIveNCHits,PrevIveNCHits,-,0.001,+
[26-May-2011 16:20:48] <Hackman238> JohnnyNoc: Was from my own post on the subject
[26-May-2011 16:21:21] <JohnnyNoc> ok, so could i take what exists for the current graph commands and add that line?
[26-May-2011 16:21:30] <f00fSteR> ?
[26-May-2011 16:21:39] <Hackman238> JohnnyNoc: Mind you the + 0.001 is to prevent a negative value. For you, if its not a counter, you wont need that
[26-May-2011 16:21:49] <JohnnyNoc> it's a gauge
[26-May-2011 16:22:02] <Hackman238> JohnnyNoc: Yeah, just tweak it for your data source
[26-May-2011 16:22:16] <Hackman238> JohnnyNoc: Names too- that was for netscreen stuff
[26-May-2011 16:22:31] <JohnnyNoc> yea
[26-May-2011 16:22:48] <Hackman238> f00fSteR: I dont know what you mean.
[26-May-2011 16:23:16] <Hackman238> JohnnyNoc: Here's the post
[26-May-2011 16:23:27] <f00fSteR> Hackman238, can i copy an existing installation of zenoss to a new server
[26-May-2011 16:23:33] <f00fSteR> if so how ?
[26-May-2011 16:23:41] <Hackman238> f00fSteR: Ah. Yes
[26-May-2011 16:23:50] <f00fSteR> any guides ?
[26-May-2011 16:23:52] <f00fSteR>
[26-May-2011 16:24:03] <Hackman238> f00fSteR: Install the same version of zenoss on the target box, ideally same platform.
[26-May-2011 16:24:08] <JohnnyNoc> oh nice
[26-May-2011 16:24:09] <JohnnyNoc> shane scott
[26-May-2011 16:24:12] <JohnnyNoc> you're a helpful dude
[26-May-2011 16:24:41] <Hackman238> f00fSteR: Next stop zenoss on the old box.
[26-May-2011 16:24:57] <Hackman238> f00fSteR: tar czvf bak.tar.gz /opt/zenoss
[26-May-2011 16:25:08] <Hackman238> f00fSteR: scp bak.tar.gz to target server
[26-May-2011 16:25:14] <f00fSteR> following you so far
[26-May-2011 16:25:31] <Hackman238> f00fSteR: mysqldump your events db to a file and scp to target server
[26-May-2011 16:25:41] <f00fSteR> k
[26-May-2011 16:25:55] <Hackman238> f00fSteR: on target server stop zenoss and untar your tarball over it. replace all files.
[26-May-2011 16:26:05] <JohnnyNoc> Hackman238 does this look right?
[26-May-2011 16:26:09] <Hackman238> f00fSteR: dump your bak.sql into your mysql
[26-May-2011 16:26:27] <Hackman238> f00fSteR: check permissions, if good, start it up.
[26-May-2011 16:26:40] <f00fSteR> just like that huh ?
[26-May-2011 16:26:56] <f00fSteR> thats fairly easy
[26-May-2011 16:27:08] <Hackman238> JohnnyNoc: Line 12, remove the ,+
[26-May-2011 16:27:11] <JohnnyNoc> i simply removed the 0.001 but left the ,
[26-May-2011 16:27:19] <Hackman238> JohnnyNoc: rather ,,+
[26-May-2011 16:27:39] <JohnnyNoc> alrighty
[26-May-2011 16:27:40] <JohnnyNoc> done
[26-May-2011 16:27:43] <Hackman238> f00fSteR: Let me know if you ahve trouble.
[26-May-2011 16:27:44] <JohnnyNoc> blasted away my existing rrd
[26-May-2011 16:28:08] <f00fSteR> Hackman238, thanks. I'm going to try this now unfortunately i need up the ass approval for stopping zenoss even for a second
[26-May-2011 16:28:34] <Hackman238> JohnnyNoc: f00fSteR You can tar zenoss live. jsut ignore the errors
[26-May-2011 16:28:45] <Hackman238> f00fSteR: Rather ^^
[26-May-2011 16:28:58] <Hackman238> f00fSteR: Pro Tip: zeopack twice to save space
[26-May-2011 16:29:02] <JohnnyNoc> ok
[26-May-2011 16:29:09] <Hackman238> JohnnyNoc: 1 second
[26-May-2011 16:29:28] <f00fSteR> whats zeopack
[26-May-2011 16:29:28] <f00fSteR> ?
[26-May-2011 16:29:35] <f00fSteR> i'm not a zenoss pro
[26-May-2011 16:29:48] <Hackman238> f00fSteR: It packs zodb. zeodb is like MSJet, its poop.
[26-May-2011 16:29:55] <f00fSteR> sys admin at current job and just sort of got handed a zenoss 2.4 setup
[26-May-2011 16:30:05] <Hackman238> f00fSteR: All changes are added to the end of the db, so it grow infinately
[26-May-2011 16:30:05] <JohnnyNoc> f00fSteR it packs your zope database so it's smaller
[26-May-2011 16:30:10] <JohnnyNoc> and faster
[26-May-2011 16:30:30] <dhopp> Hackman238: how often do you zeopack?
[26-May-2011 16:30:41] <Hackman238> dhopp: its customer dependant.
[26-May-2011 16:30:47] <f00fSteR> zeopack twice huh... how do i use this zeopack
[26-May-2011 16:30:50] <f00fSteR> it's a python script ?
[26-May-2011 16:30:54] <JohnnyNoc> i zeopack once a day
[26-May-2011 16:31:15] <Hackman238> f00fSteR: $ZENHOME/bin/zeopack local:8100 for v3
[26-May-2011 16:31:16] <JohnnyNoc> right before i get to work
[26-May-2011 16:31:24] <f00fSteR> Hackman238, 2.4.5 ?
[26-May-2011 16:31:30] <JohnnyNoc> 40 6 * *  1-5 bash -lc "/usr/local/zenoss/zenoss/bin/ -h -p 8100
[26-May-2011 16:31:36] <Hackman238> f00fSteR: $ZENHOME/bin/zeopack --host=localhost --port=8100
[26-May-2011 16:31:39] <JohnnyNoc> f00fSteR i'm using 2.5.3 but that's it's location
[26-May-2011 16:32:44] <Hackman238> JohnnyNoc: Did the graph draw?
[26-May-2011 16:33:07] <JohnnyNoc> Hackman238 no sorry, still waiting
[26-May-2011 16:33:12] <JohnnyNoc> it cycles every 60s
[26-May-2011 16:33:29] <JohnnyNoc> and it seems to need to cycle x number of times before the graph draws
[26-May-2011 16:33:37] <Hackman238> JohnnyNoc: Okay, in 5 minutes of so check and let me know.
[26-May-2011 16:35:17] <f00fSteR> hey i tried to run zeopack
[26-May-2011 16:35:21] <f00fSteR> i get the following:
[26-May-2011 16:35:30] <f00fSteR> Traceback (most recent call last):
[26-May-2011 16:35:30] <f00fSteR>   File "./", line 33, in ?
[26-May-2011 16:35:30] <f00fSteR>     from ZEO.ClientStorage import ClientStorage
[26-May-2011 16:35:30] <f00fSteR> ImportError: No module named ZEO.ClientStorage
[26-May-2011 16:35:39] <dhopp> JohnnyNoc: you are collecting perf data every 60 seconds?
[26-May-2011 16:35:56] <dhopp> foofster: run it as the zenoss user
[26-May-2011 16:35:57] <Hackman238> f00fSteR: You must run as zneoss
[26-May-2011 16:36:09] <dhopp> Hackman238: HA! I win
[26-May-2011 16:36:15] <Hackman238> dhopp: LOL
[26-May-2011 16:36:23] <f00fSteR> bash: zeopack: command not found
[26-May-2011 16:36:30] <f00fSteR> as user zenoss
[26-May-2011 16:36:30] <Hackman238> dhopp: I'm coding patches for avalon man, come on! :-P
[26-May-2011 16:36:36] <dhopp> lol
[26-May-2011 16:36:51] <Hackman238> f00fSteR: $ZENHOME/bin/zeopack
[26-May-2011 16:37:32] <f00fSteR> got it
[26-May-2011 16:37:48] <f00fSteR> ok did it twice
[26-May-2011 16:37:52] <f00fSteR> now tar the dir
[26-May-2011 16:37:53] <f00fSteR> move it ?
[26-May-2011 16:38:12] <Hackman238> f00fSteR: cd /opt
[26-May-2011 16:38:32] <Hackman238> f00fSteR: as root
[26-May-2011 16:38:46] <Hackman238> f00fSteR: Whats your $ZENHOME?
[26-May-2011 16:39:04] <f00fSteR> /opt/zenoss
[26-May-2011 16:39:11] <Hackman238> f00fSteR: Okay. cd /opt as root
[26-May-2011 16:39:21] <Hackman238> f00fSteR: tar czvf bak.tar.gz zenoss/
[26-May-2011 16:39:27] <f00fSteR> while it's running ?
[26-May-2011 16:39:35] <Hackman238> f00fSteR: Yep
[26-May-2011 16:39:36] <dhopp> f00fSteR: is your zenoss user id the same on both servers?
[26-May-2011 16:39:58] <Hackman238> dhopp: I hope so- that user is hard coded everywhere!
[26-May-2011 16:40:11] <f00fSteR> zenoss ?
[26-May-2011 16:40:13] <f00fSteR> yeah
[26-May-2011 16:40:30] <dhopp> Hackman238: I mean the UID (for permissions)...
[26-May-2011 16:40:43] <Hackman238> dhopp: Zenoss sets UID 1337
[26-May-2011 16:40:53] <dhopp> Hackman238: ahh didn't know that
[26-May-2011 16:41:04] <Hackman238> dhopp: Its more of a pain than a help in most cases...
[26-May-2011 16:41:18] <Hackman238> dhopp: results in rpmrebuild of the rpm's often LOL
[26-May-2011 16:41:38] <f00fSteR> ok so i have a 220 tar.gz file
[26-May-2011 16:41:39] <Hackman238> f00fSteR: once that tar is done, scp it as root to the target server
[26-May-2011 16:41:42] <f00fSteR> move it over to the server
[26-May-2011 16:41:46] <f00fSteR> get a mysql dump of the tables
[26-May-2011 16:41:53] <Hackman238> f00fSteR: after installing zenoss on the target, ofcoarse
[26-May-2011 16:41:58] <f00fSteR> FUCK!
[26-May-2011 16:42:05] <f00fSteR> my new server is an x64 install
[26-May-2011 16:42:11] <f00fSteR> compared to the x86
[26-May-2011 16:42:15] <Hackman238> f00fSteR: fuck indeed
[26-May-2011 16:42:18] <f00fSteR> uname -a
[26-May-2011 16:42:20] <f00fSteR> sorry
[26-May-2011 16:42:28] <Hackman238> f00fSteR: Alrighty. This is much harder, but doable
[26-May-2011 16:42:51] <Hackman238> f00fSteR: On the target install the same version of zenoss, but x86-64 version. Install ALL the same packs and patches
[26-May-2011 16:42:52] <f00fSteR> doable ? i was gonna throw up a new install of the same server architecture
[26-May-2011 16:43:12] <Hackman238> f00fSteR: You shouldnt run zenoss in 32-bit mode.
[26-May-2011 16:43:15] <f00fSteR> i wouldnt even know where to begin to look for the exact packs/patches
[26-May-2011 16:43:19] <Hackman238> f00fSteR: You're going to hit a RAM barrier quick
[26-May-2011 16:43:36] <Hackman238> f00fSteR: How much are you monitoring?
[26-May-2011 16:43:52] <f00fSteR> Hackman238, dude. ty man... you know zenoss like the back of your hand huh ?
[26-May-2011 16:43:58] <f00fSteR> 12 servers
[26-May-2011 16:44:01] <Hackman238> f00fSteR: And fo you plan on it growing much?
[26-May-2011 16:44:03] <f00fSteR> all application instances
[26-May-2011 16:44:06] <f00fSteR> lamp stacks all of them
[26-May-2011 16:44:21] <Hackman238> f00fSteR: Is no sweat. Yeah I do zenoss consulting all over the place.
[26-May-2011 16:44:30] <Hackman238> f00fSteR: How many devices?
[26-May-2011 16:44:40] <f00fSteR> Hackman238, i do, adding ec2 instances as well as xenserver stacks through dell open manage console
[26-May-2011 16:44:51] <Hackman238> f00fSteR: Can you give me a free -m on the zenoss box?
[26-May-2011 16:45:14] <f00fSteR>              total       used       free     shared    buffers     cached
[26-May-2011 16:45:14] <f00fSteR> Mem:          4053       3883        169          0         33       1261
[26-May-2011 16:45:14] <f00fSteR> -/+ buffers/cache:       2588       1465
[26-May-2011 16:45:14] <f00fSteR> Swap:         4102        354       3748
[26-May-2011 16:45:58] <Hackman238> f00fSteR: Okay, I'm going to be honest- you should move to 64-bit before you hit a RAM limit.
[26-May-2011 16:46:06] <f00fSteR> ok
[26-May-2011 16:46:27] <Hackman238> f00fSteR: My concervative customers ahve 32 and 64GB of RAM on their single box zenoss rollouts
[26-May-2011 16:46:38] <Hackman238> *conservative
[26-May-2011 16:47:25] <__virtual> we run on 8GB of ram, 64bit system with over 500 devices
[26-May-2011 16:47:27] <__virtual> runs fine
[26-May-2011 16:47:36] __virtual is now known as virtual
[26-May-2011 16:47:46] virtual is now known as pinball__
[26-May-2011 16:49:36] <Hackman238> virtual: Will highly depend on rate, types of devices, modelers, rate of model, event console usage, number of zopes, etc.
[26-May-2011 16:49:44] <JohnnyNoc> Hackman238 my graph no works
[26-May-2011 16:49:45] <JohnnyNoc> :[
[26-May-2011 16:50:13] <Hackman238> JohnnyNoc: Alrighty, can you send me the code zenoss is creating for the graph?
[26-May-2011 16:51:11] <JohnnyNoc> i can paste the log entry
[26-May-2011 16:51:12] <f00fSteR> Hackman238, just sent an e-mail for it
[26-May-2011 16:51:18] <JohnnyNoc> not sure where i just get the code for creating the gbraph
[26-May-2011 16:51:22] <JohnnyNoc> rrdtool dump/info or something?
[26-May-2011 16:51:39] <f00fSteR> Hackman238, so maybe doing the push to a new server on x64 would be better
[26-May-2011 16:51:39] <JohnnyNoc> Hackman238
[26-May-2011 16:52:05] <Hackman238> f00fSteR: I would go to x64 now before you're dealing with RAM bottle neck.
[26-May-2011 16:52:06] <JohnnyNoc> Hackman238 this is the custom graph command
[26-May-2011 16:52:08] <froztbyte> f00fSteR / Hackman238: I've got ~90 devices on ~2GB, running pretty okay
[26-May-2011 16:52:09] <JohnnyNoc>
[26-May-2011 16:52:20] <froztbyte> could probably fit another 30 on
[26-May-2011 16:53:21] <Hackman238> Admins who do not plan ahead get fired
[26-May-2011 16:53:26] <Hackman238> Thats all I have to say
[26-May-2011 16:53:27] <Hackman238> LOL
[26-May-2011 16:53:54] <Hackman238> JohnnyNoc: Problem found. Give me a bit
[26-May-2011 16:54:07] <JohnnyNoc> Hackman238 i think i see my problem
[26-May-2011 16:54:14] <froztbyte> I'll...rather stfu
[26-May-2011 16:54:42] <Hackman238> JohnnyNoc: its the names
[26-May-2011 16:54:43] <jduque> anyone knows where in the code I have to change to show the interface alias instead of the device component when creating a new report
[26-May-2011 16:54:54] <Hackman238> froztbyte: I dont understand. I mean no offense.
[26-May-2011 16:55:16] <rmatte> Anyone who isn't running 64bit these days (other than servers that were purchased prior to 64bit becoming available) should be shot.
[26-May-2011 16:55:57] <Hackman238> JohnnyNoc: CDEF:CurrentDataRate=DataRate-raw
[26-May-2011 16:56:02] <rmatte> "You know, that 64bit CPU looks too fast, I think I'll buy the slower 32bit"
[26-May-2011 16:56:41] <JohnnyNoc> Hackman238 i got rid of the line above that
[26-May-2011 16:57:12] <JohnnyNoc> looks like i needed CurrentDataRate set and it was complaining about reuse
[26-May-2011 16:57:32] <Hackman238> JohnnyNoc: Correct. Also, you'll want to graph the diffrate
[26-May-2011 16:58:24] <Hackman238> JohnnyNoc: Repaste your code, if you dont mind
[26-May-2011 16:59:45] <Hackman238> rmatte: LOL brutal rmatte  :-)
[26-May-2011 16:59:48] kocolosk is now known as kocolosk|away
[26-May-2011 17:00:07] <JohnnyNoc> but what if my server only has 2GB of RAMZ!
[26-May-2011 17:01:35] <dhopp> I had a sysadmin at a previous job (2 years ago) that built a new server that had 2 drives but didn't bother to setup a mirror…put the OS on drive 1 and the database for a business accounting package on disk 2
[26-May-2011 17:01:40] <JohnnyNoc> dhopp yes, every 60s
[26-May-2011 17:01:42] <dhopp> I then didn't understand why I was pissed
[26-May-2011 17:02:15] <dhopp> I then = And then
[26-May-2011 17:02:20] <f00fSteR> dhopp, lol we have hardware raid, also backups running all across the week
[26-May-2011 17:03:23] <dhopp> f00fSteR: wasn't implying you did…was just making a comment to Hackman238 points…some admins just don't think ahead at the consequences
[26-May-2011 17:03:27] <Hackman238> All, have to run. I'll be back later. Have to do some engine work on my car.
[26-May-2011 17:03:39] <JohnnyNoc> gl with that
[26-May-2011 17:03:45] <f00fSteR> Hackman238, i'm gonna follow as per your instructions
[26-May-2011 17:03:59] <f00fSteR> Hackman238, at least to the sense of the x64 bit
[26-May-2011 17:04:03] <dhopp> Hackman238: I don't think your engine understands python :-P
[26-May-2011 17:04:18] <f00fSteR> godd... i wish they just used nagios
[26-May-2011 17:04:21] <Hackman238> f00fSteR: Alrighty. I'll be back later to give you the list of instr. to do that.
[26-May-2011 17:04:23] <f00fSteR> or is that a nono in this channel ?
[26-May-2011 17:04:35] <JohnnyNoc> shouldn't be
[26-May-2011 17:04:38] <Hackman238> dhopp: Oh I do automotive and industrial automation too
[26-May-2011 17:04:38] <JohnnyNoc> we use nagios and zenoss
[26-May-2011 17:04:48] <Hackman238> Later all!
[26-May-2011 17:04:53] <f00fSteR> Hackman238, tc dude
[26-May-2011 17:05:01] <xuru> see ya
[26-May-2011 17:05:29] <f00fSteR> JohnnyNoc, honestly how good is zenoss compared to nagios
[26-May-2011 17:05:48] <JohnnyNoc> they both have their pros and cons, depends what you want/need it for
[26-May-2011 17:05:58] <f00fSteR> ike i said only about 12 servers
[26-May-2011 17:06:01] <JohnnyNoc> here, zenoss shines because of the historical and modeling data
[26-May-2011 17:06:26] <JohnnyNoc> nagios shines because it's what we've always used, it's fast, the checks are super quick/easy to write
[26-May-2011 17:06:32] <JohnnyNoc> so support people prefer nagios interface
[26-May-2011 17:06:52] <f00fSteR> and sysadmins like me prefer what axactly ?
[26-May-2011 17:07:07] <JohnnyNoc> again, it depends on what your needs are
[26-May-2011 17:07:12] <JohnnyNoc> but speaking by numbers, i think they prefer nagios
[26-May-2011 17:07:12] <JohnnyNoc>
[26-May-2011 17:07:38] <f00fSteR> 12 servers production... with a mail server sending out over 4 million mails a day just the regular snmp zenpacks and tomcat logging
[26-May-2011 17:08:05] <rmatte> Zenoss is nice if you don't want to bang your head in the wall setting up a monitoring too
[26-May-2011 17:08:06] <f00fSteR> can i set up nagios without any scripting experience or is that needed in across the board in these monitoring enviorments
[26-May-2011 17:08:07] <rmatte> tool*
[26-May-2011 17:08:17] <rmatte> against the wall*
[26-May-2011 17:08:22] <jmp242> f00fSteR: you're going to want some scripting experiance I would thinkl
[26-May-2011 17:08:27] * f00fSteR goes and bangs his head against the wall
[26-May-2011 17:08:29] <xuru> in mysql, how do you do a count of rows in a table?
[26-May-2011 17:08:39] <jmp242> I mean, nothing does everything out of the box
[26-May-2011 17:08:51] <f00fSteR> so tomcat logging
[26-May-2011 17:08:53] <jmp242> though I suppose a commercial tool might be a bit more plug an play near the bottom - spiceworks?
[26-May-2011 17:08:55] <f00fSteR> how does it work in zenoss
[26-May-2011 17:09:00] <f00fSteR> how does it tail logs ?
[26-May-2011 17:09:05] <rmatte> xuru: select count(*) from blah
[26-May-2011 17:09:10] <JohnnyNoc> rmatte i find myself banging my head a lot with zenoss
[26-May-2011 17:09:14] <xuru> rmatte: sweet, ty
[26-May-2011 17:09:17] <rmatte> np
[26-May-2011 17:09:19] <jmp242> Well, I've not used tomcat stuff
[26-May-2011 17:09:20] <dhopp> nagios has been around a long time…it is proven.  However every interface for it sucks.  When you get into large installations configuration management is a nightmare.  Most every check it spawns a seperate process so the overhead can get REALLY bad.  For 12 servers it really doesn't matter which way you go
[26-May-2011 17:09:42] <f00fSteR> to be 100% fair
[26-May-2011 17:09:42] <JohnnyNoc> dhopp people here seem to enjoy check_mk
[26-May-2011 17:09:46] <f00fSteR> can i set up both
[26-May-2011 17:09:55] <jmp242> but mostly Zenoss uses syslog . . . or SNMP Traps
[26-May-2011 17:09:56] <f00fSteR> and have them pull checks on each other
[26-May-2011 17:10:03] <JohnnyNoc> f00fSteR yes
[26-May-2011 17:10:04] <rmatte> JohnnyNoc: then you haven't used it long enough
[26-May-2011 17:10:04] <jmp242> Zenoss can use most nagios plugins
[26-May-2011 17:10:05] <JohnnyNoc> we do this
[26-May-2011 17:10:09] <jmp242> as a command datasource
[26-May-2011 17:10:10] <dhopp> JohnnyNoc: I tried it…still wasn't a fan…although it probably is the best out there
[26-May-2011 17:10:13] <jmp242> but you can do both
[26-May-2011 17:10:27] <JohnnyNoc> rmatte perhaps, but i really don't feel alone
[26-May-2011 17:10:46] <JohnnyNoc> but i understand it's a learning curve, and i'm not bashing zenoss for it
[26-May-2011 17:10:46] <JohnnyNoc>
[26-May-2011 17:10:50] <f00fSteR> well disaster recovery would be good... so i'm doing what you guys are saying in par of thinking ahead
[26-May-2011 17:10:52] <f00fSteR> ...planning ahead
[26-May-2011 17:11:18] <rmatte> JohnnyNoc: I know the feeling, I didn't just start out knowing Zenoss this well, I clawed my way up over a couple years
[26-May-2011 17:11:22] <rmatte>
[26-May-2011 17:11:26] <jmp242> Well, obvioudly we like Zenoss as we're using it
[26-May-2011 17:11:26] <jmp242> lol
[26-May-2011 17:11:41] <JohnnyNoc> jmp242 do you use enterprise?
[26-May-2011 17:11:43] <rmatte> yeh, it's like going in #linux and asking them if linux is any good
[26-May-2011 17:11:46] <jmp242> Starting from scratch, 3+ years ago
[26-May-2011 17:11:47] <f00fSteR> rmatte, i think almost everyone in my dept wants to get ridd of zenoss
[26-May-2011 17:11:56] <jmp242> I chose Zenoss over Nagios
[26-May-2011 17:11:57] <rmatte> f00fSteR: why?
[26-May-2011 17:11:58] <JohnnyNoc> rmatte a lot of my headaches i'm sure are due to the fact that i picked up an existing installation and the guy before me didn't have much clue either
[26-May-2011 17:12:10] <jmp242> because I thought it was much easier to get going than Nagios was
[26-May-2011 17:12:12] <f00fSteR> rmatte, honestly, noone here knows how to use it
[26-May-2011 17:12:13] <jmp242> I use Core
[26-May-2011 17:12:15] <rmatte> JohnnyNoc: yeh, those are the worst situations
[26-May-2011 17:12:16] <f00fSteR> or tell it from head to toe
[26-May-2011 17:12:24] <JohnnyNoc> f00fSteR same situation here
[26-May-2011 17:12:29] <rmatte> f00fSteR: It's really not rocket science
[26-May-2011 17:12:32] <jmp242> And it's stayed easy for me to gradually add functionality
[26-May-2011 17:12:35] <f00fSteR> JohnnyNoc, yeah i see
[26-May-2011 17:12:37] <rmatte> Not from a general user perspective anyways
[26-May-2011 17:12:38] <jmp242> so no reason to switch
[26-May-2011 17:12:48] <f00fSteR> anyways the 4gb cap isnt a big deal tho is it ?
[26-May-2011 17:12:58] <jmp242> But I guess it depends on where you're coming from
[26-May-2011 17:13:00] <rmatte> 4gb for how many devices?
[26-May-2011 17:13:04] <JohnnyNoc> yea, i'm dying for the opportunity to start a 3.x install from scratch
[26-May-2011 17:13:15] <jmp242> if you don't know Linux at all, you're going to hate all the *nix monitoring tools
[26-May-2011 17:13:17] <JohnnyNoc> boss said something about an SSD for my perf graphs
[26-May-2011 17:13:23] <JohnnyNoc> that made me
[26-May-2011 17:13:25] <f00fSteR> is there a zenoss command for listing devices ?
[26-May-2011 17:13:29] <rmatte> I absolutely hate all of the windows monitoring tools
[26-May-2011 17:13:35] <rmatte> talk about useless overhead
[26-May-2011 17:13:35] <f00fSteR> heck i learned so much just idling in this chan
[26-May-2011 17:13:37] <jmp242> zendmd
[26-May-2011 17:13:42] <dhopp> JohnnyNoc: rrd is not real friendly to SSDs
[26-May-2011 17:13:46] <rmatte> and impossible to integrate with anything
[26-May-2011 17:13:47] <JohnnyNoc> f00fSteR i think most people use a script to query zend/dmd
[26-May-2011 17:13:47] <jmp242> will get you the console
[26-May-2011 17:13:54] <jmp242> but do you want to do this at the command line?
[26-May-2011 17:13:57] <jmp242> or in the web GUI/
[26-May-2011 17:13:58] <jmp242> ?
[26-May-2011 17:14:04] <f00fSteR> command line
[26-May-2011 17:14:06] <f00fSteR> i understand
[26-May-2011 17:14:12] <f00fSteR> webgui... wtf is that ?
[26-May-2011 17:14:37] <jmp242> there's a script here somewhere - forums etc
[26-May-2011 17:14:42] <f00fSteR> kk
[26-May-2011 17:14:56] <f00fSteR> lets do this... giyf
[26-May-2011 17:15:01] <JohnnyNoc> dhopp yea i read something about that too
[26-May-2011 17:15:02] <dhopp> jmp242: is that like "there's an app for that"?
[26-May-2011 17:15:05] <rmatte> to get a list of devices in zendmd you'd do...
[26-May-2011 17:15:06] <rmatte> for d in dmd.Devices.getSubDevices():
[26-May-2011 17:15:06] <rmatte>  print '%s - %s' % (, d.ipAddress)
[26-May-2011 17:15:46] <dhopp> JohnnyNoc: don't get me wrong, you probably won't have performance problems due to disk i/o..but you will have a lot of very small rrds that are constantly written over
[26-May-2011 17:15:46] <jmp242> it's like I don't use zendmd a lot, nor python, but I'm sure I could hack it up in 10 minutes + google
[26-May-2011 17:15:47] <rmatte> zendmd is python
[26-May-2011 17:16:08] <rmatte> actually hmmm, it's not ipAddress... hold on lol
[26-May-2011 17:16:12] <JohnnyNoc> dhopp you're basically saying it's going to die sooner than i expect?
[26-May-2011 17:16:12] <JohnnyNoc>
[26-May-2011 17:16:12] <rmatte> trying this from the top of my head
[26-May-2011 17:16:17] <f00fSteR> flucks
[26-May-2011 17:16:19] <jmp242>
[26-May-2011 17:16:26] <dhopp> JohnnyNoc: and SSDs cells still has a limited number of writes (granted much higher then it used to be)
[26-May-2011 17:16:32] <jmp242> gives me the doc on it, and then some work, but rmatte knows from his head lol
[26-May-2011 17:16:46] <dhopp> JohnnyNoc: I don't know the answer to that :-)
[26-May-2011 17:16:47] <jmp242> I just look at the gui for what devices there are lol
[26-May-2011 17:17:15] <jmp242> the limited number of writes for modern SSDs is like 2x the average spinning disk lifetime now
[26-May-2011 17:17:33] <jmp242> I don't worry about that anymore - I worry more about the cost premium
[26-May-2011 17:17:38] <f00fSteR> ok lets see
[26-May-2011 17:17:46] <rmatte> ah, here we go...
[26-May-2011 17:17:47] <f00fSteR> how do i do it from gui note: i use the 2.4.5 interface
[26-May-2011 17:17:58] <f00fSteR> haha
[26-May-2011 17:18:02] <f00fSteR> 1 + 2 ?
[26-May-2011 17:18:03] <jmp242> s.4.5?
[26-May-2011 17:18:04] <rmatte> for d in dmd.Devices.getSubDevices():
[26-May-2011 17:18:05] <rmatte>  print '%s - %s' % (d.getId(), d.getDeviceIp())
[26-May-2011 17:18:06] <jmp242> 2.4.5
[26-May-2011 17:18:06] <rmatte> there
[26-May-2011 17:18:11] <jmp242> that's ollllldddd
[26-May-2011 17:18:17] <rmatte> the space before the second line is necessary
[26-May-2011 17:18:25] <jmp242> yes, python whitespace matters
[26-May-2011 17:18:41] <dhopp> jmp242: which I hate
[26-May-2011 17:18:58] <dhopp> jmp242: yes it makes for more readable code..but a tab is not the same as 5 spaces..but it looks the same
[26-May-2011 17:19:08] <rmatte> f00fSteR: to do it from the gui you need to create a custom report, which can be done from the GUI itself
[26-May-2011 17:19:17] <f00fSteR> rmatte, thanks
[26-May-2011 17:19:19] <rmatte> go to the reports section
[26-May-2011 17:19:22] <rmatte> click on custom reports
[26-May-2011 17:19:30] <rmatte> add one
[26-May-2011 17:19:31] <rmatte> then edit it
[26-May-2011 17:19:33] <jmp242> that dmd works nice
[26-May-2011 17:19:47] <jmp242> dhopp: just always use spaces per policy
[26-May-2011 17:20:05] <f00fSteR> rmatte, i tried it in command line nothting happened
[26-May-2011 17:20:08] <f00fSteR> >>> for d in dmd.Devices.getSubDevices():
[26-May-2011 17:20:09] <f00fSteR> ...  print '%s - %s' % (d.getId(), d.getDeviceIp())
[26-May-2011 17:20:09] <f00fSteR> ...
[26-May-2011 17:20:13] <dhopp> jmp242: yeah if I'm the writing the code :-)…doesn't help me if somebody else wrote the code
[26-May-2011 17:20:15] <jmp242> hit enter one mroe time
[26-May-2011 17:20:38] <f00fSteR> lol
[26-May-2011 17:20:42] <f00fSteR> it;s the same 12 servers
[26-May-2011 17:20:45] <rmatte> f00fSteR: here are the settings for the custom report:
[26-May-2011 17:20:45] <f00fSteR> i was talking about
[26-May-2011 17:20:49] <f00fSteR> so you guys meant this
[26-May-2011 17:20:52] <f00fSteR> yeah i have 12 devices
[26-May-2011 17:20:55] <rmatte> f00fSteR: it's a bit more complex than you'd think at first
[26-May-2011 17:21:05] <rmatte> but you'll get used to it
[26-May-2011 17:21:33] <rmatte> you basically specify column names and get functions
[26-May-2011 17:21:35] <jmp242> why would you get more devices?
[26-May-2011 17:21:40] <rmatte> a trick to view the get functions is as follows...
[26-May-2011 17:21:46] <rmatte> login to zendmd, then do...
[26-May-2011 17:21:56] <rmatte> d = dmd.Devices.findDevice('devicename')
[26-May-2011 17:22:02] <rmatte> then you can type d. and hit tab
[26-May-2011 17:22:08] <rmatte> or d.get and hit tab
[26-May-2011 17:22:13] <rmatte> and it'll show you all the possibilities
[26-May-2011 17:22:18] <JohnnyNoc> you knwo something funny
[26-May-2011 17:22:19] <rmatte> you can experiment until you find what you want
[26-May-2011 17:22:26] <JohnnyNoc> i've found that i have to mistype a command before i can tab out
[26-May-2011 17:22:44] <JohnnyNoc> it's weird, i don't know why
[26-May-2011 17:22:46] <rmatte> JohnnyNoc: I never have that problem
[26-May-2011 17:22:53] <rmatte> quite weird
[26-May-2011 17:23:06] <JohnnyNoc> maybe not necessarily mistyping
[26-May-2011 17:23:18] <JohnnyNoc> but i'll have to enter some lines first, then options will show up in tab
[26-May-2011 17:23:45] <rmatte> JohnnyNoc: you do know that tab doesn't work for for loops or whatever right?
[26-May-2011 17:23:46] <f00fSteR> tab doesnt work for me
[26-May-2011 17:23:47] <f00fSteR> with that
[26-May-2011 17:23:53] <rmatte> you need to be doing it on a variable that's already set
[26-May-2011 17:24:01] <JohnnyNoc> rmatte, hrm, maybe that would explain it
[26-May-2011 17:24:02] <JohnnyNoc>
[26-May-2011 17:24:14] <rmatte> if I do: for d in dmd.Devices.getSubDevices(): I can't then do d.<tab>
[26-May-2011 17:24:29] <rmatte> but if I were to type some other garbage it would run through the for loop and set d to something at the end
[26-May-2011 17:24:31] <JohnnyNoc> yea that probably explains the issue
[26-May-2011 17:24:33] <rmatte> so I'd then be able to do it
[26-May-2011 17:24:37] <rmatte> that's probably what you're seeing
[26-May-2011 17:24:49] <JohnnyNoc> yea, i agree
[26-May-2011 17:24:55] <rmatte> use d = dmd.Devices.findDevice('') or even just d = find('')
[26-May-2011 17:25:56] <rmatte> f00fSteR: you need to replace 'devicename' with the name of an actual device
[26-May-2011 17:26:00] <rmatte> that was just an example
[26-May-2011 17:26:04] <f00fSteR> haha
[26-May-2011 17:26:16] <rmatte> d = dmd.Devices.findDevice('nameofanactualdevice')
[26-May-2011 17:26:19] * f00fSteR slaps rmatte with a large fuckin trout
[26-May-2011 17:26:19] <rmatte> then d.<tab>
[26-May-2011 17:26:21] <f00fSteR> silly goose
[26-May-2011 17:26:40] <rmatte>
[26-May-2011 17:26:52] <JohnnyNoc> rmatte are you RRD clued?
[26-May-2011 17:27:01] <rmatte> as in learn-ed?
[26-May-2011 17:27:05] <JohnnyNoc> i was having an RRD issue earlier that Hackman238 was helping with
[26-May-2011 17:27:06] <JohnnyNoc> rmatte yes
[26-May-2011 17:27:21] <rmatte> well, I have a fair bit of experience... but more with dealing with the files themselves
[26-May-2011 17:27:35] <rmatte> as far as actual RRD statements go, not much experience
[26-May-2011 17:27:39] <JohnnyNoc> gotcha
[26-May-2011 17:27:47] <f00fSteR> let me ask you something when we get alertts for null pointer exceptions from like zenoss
[26-May-2011 17:27:51] <f00fSteR> what is it using to find those
[26-May-2011 17:27:53] <f00fSteR> like zenpacks
[26-May-2011 17:27:55] <f00fSteR> snmp
[26-May-2011 17:27:56] <JohnnyNoc> i was banging my head earlier (see i told you!) trying to figure out how to get actual max values in the graph
[26-May-2011 17:27:58] <rmatte> I wrote a script not too long ago that went through and removed any negative values from RRD files
[26-May-2011 17:27:59] <f00fSteR> or like some custom python script
[26-May-2011 17:28:10] <rmatte> for some reason there was no min value of 0 set on my 64bit ethernet template
[26-May-2011 17:28:21] <rmatte> so I was getting crazy wonky negative values from time to time
[26-May-2011 17:28:50] <rmatte> f00fSteR: alerts for null pointer exceptions?
[26-May-2011 17:28:56] <JohnnyNoc> rmatte yea i'm going to have to do the same
[26-May-2011 17:28:58] <JohnnyNoc> and set MAX too
[26-May-2011 17:29:11] <JohnnyNoc> still have those scripts laying around?
[26-May-2011 17:29:12] <JohnnyNoc> :]
[26-May-2011 17:29:15] <rmatte> JohnnyNoc: well, you still want to set MAX liberally high
[26-May-2011 17:29:30] <rmatte> maybe like 10gbps
[26-May-2011 17:29:47] <rmatte> JohnnyNoc: I probably do, let me check
[26-May-2011 17:29:55] <JohnnyNoc> nice
[26-May-2011 17:30:02] <f00fSteR> rmatte, i wanna know how it works
[26-May-2011 17:30:04] <f00fSteR> so for example
[26-May-2011 17:30:10] <rmatte> they are quick hacky scripts, not very polished
[26-May-2011 17:30:11] <f00fSteR> we get an alert like the following
[26-May-2011 17:30:15] <f00fSteR> Message:
[26-May-2011 17:30:15] <f00fSteR> Read timed out
[26-May-2011 17:30:22] <f00fSteR> where is that coming from ?
[26-May-2011 17:30:23] <JohnnyNoc> rmatte no worries, that's about as good as a script i would have written would be
[26-May-2011 17:30:29] <f00fSteR> log files... snmp... plugins ?
[26-May-2011 17:31:04] <JohnnyNoc> i'll brb, going for a smoke, maybe shoot me a link?  or send pm on the forums?
[26-May-2011 17:31:09] <rmatte> f00fSteR: you're seeing this as an event in the event console?
[26-May-2011 17:31:18] <dhopp> f00fSteR: that looks like a jmx thing
[26-May-2011 17:31:29] <rmatte> double click on it, expand the properties, and look for agent
[26-May-2011 17:31:39] <rmatte> that'll tell you what daemon is generating it
[26-May-2011 17:31:48] <f00fSteR> dhopp, ahh so jmx is whats pulling tomcat data ?
[26-May-2011 17:32:40] <f00fSteR> rmatte, agent zensyslog
[26-May-2011 17:33:04] <f00fSteR> that does what pulls the syslogs and sends them out as alerts based on regex ?
[26-May-2011 17:33:11] <rmatte> JohnnyNoc: ok, I have a few scripts here, let me walk you through these so you understand what they are...
[26-May-2011 17:33:20] <rmatte> f00fSteR: zensyslog is purely a syslog server
[26-May-2011 17:33:26] <rmatte> so it receives syslogs from devices
[26-May-2011 17:33:28] <rmatte> that's it
[26-May-2011 17:33:43] <rmatte> so you have some device configured to send syslogs to Zenoss which is sending that error
[26-May-2011 17:34:17] <f00fSteR> so pretty much anything the syslog spews out will come in as alerts on zenoss ?
[26-May-2011 17:34:35] <rmatte> unless you tune them, yes
[26-May-2011 17:34:42] <JohnnyNoc> or transform them
[26-May-2011 17:34:49] <rmatte> Zenoss has a powerful event management system that allows you to tune/transform events
[26-May-2011 17:34:50] <f00fSteR> any lace to start on where i can go about learning more of this ?
[26-May-2011 17:34:56] <f00fSteR> place*
[26-May-2011 17:35:04] <JohnnyNoc> zenoss admin guide
[26-May-2011 17:35:05] <JohnnyNoc>
[26-May-2011 17:35:10] <rmatte> f00fSteR: have you read the Zenoss Admin guide top to bottom?
[26-May-2011 17:35:15] <rmatte> If not, you should
[26-May-2011 17:35:20] <rmatte> that's where I started
[26-May-2011 17:35:22] <f00fSteR> i work for a small company... as the new sys admin ive pretty much been hired to take care of monitoring... they said either use zenoss or whatever you prefer
[26-May-2011 17:35:37] <f00fSteR> i read the zenoss core book by packt publishing... 2008 edition
[26-May-2011 17:35:42] <f00fSteR> it didnt teach me anything on syslogd
[26-May-2011 17:35:51] <rmatte> Yeh, the admin guide is much more up to date than that
[26-May-2011 17:35:55] <f00fSteR> just install devices snmp and zenpacks
[26-May-2011 17:35:58] <rmatte> and contains better info in my opinion
[26-May-2011 17:36:06] <rmatte> I read that book too, and the admin guide was honestly more helpful
[26-May-2011 17:36:07] <rmatte> lol
[26-May-2011 17:36:07] <JohnnyNoc> i've never paid mych thought to zensyslogd
[26-May-2011 17:36:20] <JohnnyNoc> i just know shit in the logs ends up in my dashboard
[26-May-2011 17:36:23] <JohnnyNoc> automagically
[26-May-2011 17:36:25] <rmatte> it's zensyslog, not zensyslogd
[26-May-2011 17:36:26] <rmatte> lol
[26-May-2011 17:36:29] <f00fSteR> JohnnyNoc, it seems like everything in alerts is generated off the agent syslogd
[26-May-2011 17:36:35] <JohnnyNoc> f00fSteR not everything
[26-May-2011 17:36:41] <JohnnyNoc> depends on the datasource
[26-May-2011 17:36:42] <JohnnyNoc> SNMP
[26-May-2011 17:36:44] <JohnnyNoc> COMMAND
[26-May-2011 17:36:53] <JohnnyNoc> rmatte no... ?
[26-May-2011 17:36:59] <JohnnyNoc> i mean, there are def snmp alerts that get sent
[26-May-2011 17:37:02] <JohnnyNoc> or snmp traps that area caught
[26-May-2011 17:37:09] <rmatte> JohnnyNoc: I'm saying there's no d at the end of it lol
[26-May-2011 17:37:14] <rmatte> anyways, back to these scripts...
[26-May-2011 17:37:20] <f00fSteR> lol
[26-May-2011 17:37:22] <f00fSteR> kk
[26-May-2011 17:37:26] <f00fSteR> admin guide it is
[26-May-2011 17:37:32] <f00fSteR> surprised noone yelled RTFM yet
[26-May-2011 17:37:41] <f00fSteR> rmatte, will you do the honor ?
[26-May-2011 17:37:58] * JohnnyNoc trouts f00fSteR with a slap
[26-May-2011 17:38:01] <rmatte> JohnnyNoc: Save this one as a .sh, chmod +x it... to use it, cd to $ZENHOME/Perf/Devices, then execute it...
[26-May-2011 17:38:17] <rmatte> JohnnyNoc: what it does is tune the RRD files themselves so that the min is set to 0
[26-May-2011 17:38:25] <JohnnyNoc> *nod*
[26-May-2011 17:38:27] <JohnnyNoc> seems simple enough
[26-May-2011 17:38:35] <rmatte> because even after you change the setting in Zenoss, it won't change it in the files.
[26-May-2011 17:38:46] <rmatte> obviously change the name of the RRD to the one you want to tune
[26-May-2011 17:38:49] <rmatte> alright, next script...
[26-May-2011 17:38:55] kocolosk|away is now known as kocolosk
[26-May-2011 17:39:50] <rmatte>
[26-May-2011 17:40:02] <rmatte> That one normalizes the RRDs by removing negative values from them.
[26-May-2011 17:40:17] <JohnnyNoc> nice
[26-May-2011 17:40:21] <rmatte> It dumps the values to an RRD file, then replaces any negative values with 'NaN', then restores the RRD file
[26-May-2011 17:40:24] <JohnnyNoc> we have some disks whose writes are written as negative values
[26-May-2011 17:40:27] <rmatte> It's hacky, but it worked nicely
[26-May-2011 17:40:29] <JohnnyNoc> and others that don't
[26-May-2011 17:40:48] <rmatte> obviously backup your RRD directory beforehand if the worst should happen, but I ran it and had no issues
[26-May-2011 17:40:59] <rmatte> just make sure you're actually in the $ZENHOME/perf directory before running it
[26-May-2011 17:41:14] <rmatte> since it uses the find command, which will take forever from elsewhere
[26-May-2011 17:41:18] <JohnnyNoc> ok
[26-May-2011 17:41:21] <JohnnyNoc> you could justr do
[26-May-2011 17:41:29] <JohnnyNoc> find $ZENHOME/perf
[26-May-2011 17:41:32] <rmatte> I know, I could replace the . with $ZENHOME/perf
[26-May-2011 17:41:38] <rmatte> but I was lazy when I wrote it
[26-May-2011 17:41:40] <JohnnyNoc> hah
[26-May-2011 17:41:40] <rmatte>
[26-May-2011 17:41:41] <JohnnyNoc> no big deal
[26-May-2011 17:44:18] <rmatte> hmmm
[26-May-2011 17:44:26] <rmatte> I just noticed something about that code...
[26-May-2011 17:44:49] <JohnnyNoc> no big deal i'll figure it out
[26-May-2011 17:44:50] <JohnnyNoc>
[26-May-2011 17:44:55] <JohnnyNoc> i'm more worried about my RRD problem right now
[26-May-2011 17:45:08] <rmatte> the rm and the rrdtool restore should be outside that while loop I believe
[26-May-2011 17:46:09] <rmatte> yeh they should
[26-May-2011 17:47:09] <rmatte> I think I noticed that at some point but never bothered updating the script that I had backed up
[26-May-2011 17:47:22] <rmatte>
[26-May-2011 17:47:25] <rmatte> should be that
[26-May-2011 17:51:44] <rmatte> alright, beer time
[26-May-2011 17:51:46] <rmatte> later all
[26-May-2011 17:51:48] <JohnnyNoc> thanks rmatte
[26-May-2011 17:51:49] <JohnnyNoc> enjoy
[26-May-2011 17:51:51] <rmatte> np
[26-May-2011 18:57:52] <Hackman238b> JohnnyNoc: How'd it turn out?
[26-May-2011 18:58:06] <Hackman238b> f00fSteR: I'll get you full instructions shortly.
[26-May-2011 18:58:15] <Hackman238b> xuru: How's it turn out?
[26-May-2011 19:15:30] <locohost> hack, you got any idea how to delete a custom report that cannot be deleted via ui
[26-May-2011 19:15:50] <locohost> its a really big report..and just locked up zenoss. till i restart zeocrtl
[26-May-2011 19:15:56] <locohost> both when i try to view or delete it
[26-May-2011 19:21:11] <xuru> Hackman238b: hey, it finally finished after a few hours
[26-May-2011 19:21:38] <xuru> Hackman238b: haven't had a chance to see if it made a difference yet
[26-May-2011 19:22:01] <locohost> whats slow?
[26-May-2011 19:22:30] <xuru> oh, I was trying to delete some 360k events
[26-May-2011 19:26:38] <locohost> ugh, thats a lot of events
[26-May-2011 19:26:48] <locohost> i wonder what happens if you delete them from mysql?
[26-May-2011 19:26:52] <locohost> might screw up zope
[26-May-2011 19:27:42] <xuru> seems to be just fine
[26-May-2011 19:29:24] <locohost> whats fine? truncating the mysql events database status table?
[26-May-2011 19:29:47] <locohost> i have no diea what kinda bad side effects that would have, but guessing it would at least make current alarms dissapear
[26-May-2011 19:29:51] <xuru> zope
[26-May-2011 19:30:00] <locohost> could always try it in a test box or something
[26-May-2011 19:30:17] <Hackman238b> xuru: Worked?
[26-May-2011 19:30:27] <xuru> All the events that I deleted where bogus anyway
[26-May-2011 19:30:33] <locohost> i just mean, there could be stuff in zope, tied to events and druncating the table may confuse it
[26-May-2011 19:30:41] <xuru> Hackman238b: if finished, yes
[26-May-2011 19:30:54] <Hackman238b> locohost: sure. Whats the full path?
[26-May-2011 19:31:04] <Hackman238b> xuru: Glad it worked.
[26-May-2011 19:31:17] <locohost> hack, the path to the report?
[26-May-2011 19:31:24] <Hackman238b> locohost: correct
[26-May-2011 19:31:34] <locohost> 1 sec
[26-May-2011 19:31:42] <Hackman238b> locohost: Can just paste the URL for me
[26-May-2011 19:31:49] <xuru> it's having trouble updating the event console though
[26-May-2011 19:32:04] <Hackman238b> xuru: mysql is almost certainly indexing
[26-May-2011 19:32:11] <locohost> http://d01mtrending01:8080/zport/dmd/reports#reporttree:.zport.dmd.Reports.Multi-Graph%20Reports.Total_BW
[26-May-2011 19:32:18] <xuru> ah, ok
[26-May-2011 19:32:19] <Hackman238b> xuru: It might be all night to be honest.
[26-May-2011 19:32:24] <locohost> clicking t hat means i have to restart zeoctrl, lol
[26-May-2011 19:32:29] <Hackman238b> xuru: Thats alot of events in a short time :-)
[26-May-2011 19:32:50] <xuru> Hackman238b: hehe, yeah, I'm leaving soon anyway
[26-May-2011 19:33:37] <locohost> polking around in zope /manage interface but not finding anything
[26-May-2011 19:33:46] <locohost> i remember i did a bucn of cool stuff here a few years ago
[26-May-2011 19:34:17] <Hackman238b> locohost: goto http://d01mtrending01:8080/zport/dmd/Reports/Multi-Graph Reports/Total_BW/manage
[26-May-2011 19:34:21] <locohost> and replaced the olduseless hardware tab with a html/dtml wrapper that brought up an preexisting mysql cmdb
[26-May-2011 19:34:30] <Hackman238b> locohost: You can fix the report there.
[26-May-2011 19:34:47] <Hackman238b> locohost: goto http://d01mtrending01:8080/zport/dmd/Reports/Multi-Graph Reports/manage
[26-May-2011 19:34:56] <Hackman238b> locohost: If you just want to pick it and delete it
[26-May-2011 19:35:30] <locohost> yeah, maybe delete "elements"
[26-May-2011 19:35:47] <Hackman238b> locohost: No no, you'll want to go into elements and tweak the contents
[26-May-2011 19:36:05] <Hackman238b> locohost: Dont delete any organizers under the report unless you want to just get rid of the report entirely
[26-May-2011 19:38:10] <xuru> Hackman238b: thanks again for your help
[26-May-2011 19:40:10] <locohost> good grief trying to delete the entire report is spinning zopecontrol again
[26-May-2011 19:40:29] <locohost> or "runzope
[26-May-2011 19:40:43] <locohost> yeah, i do want to just drop it for now
[26-May-2011 19:41:04] <Hackman238b> goto http://d01mtrending01:8080/zport/dmd/Reports/Multi-Graph Reports/manage
[26-May-2011 19:41:14] <Hackman238b> And check the box next to it > Delte
[26-May-2011 19:41:29] <Hackman238b> locohost: Then as zenoss zopectl restart
[26-May-2011 19:41:37] <Hackman238b> locohost: It almost certainly needs it :-)
[26-May-2011 19:44:36] <locohost> thats where i clicked delete and runzope is taking an entire cpu
[26-May-2011 19:44:41] <locohost> maybe ill just let it for for a bit
[26-May-2011 19:44:57] <Hackman238b> locohost: Oh gotcha- it didnt return an error did it?
[26-May-2011 19:45:13] <Hackman238b> locohost: If not- just wait.
[26-May-2011 19:45:46] <locohost> no, its still going, but my click on elements in a nother window loaded after like 10m
[26-May-2011 19:45:52] <locohost> and says too many relations
[26-May-2011 19:45:58] <locohost> at least im getting somewhere
[26-May-2011 19:46:01] <locohost> man, there are alot
[26-May-2011 19:46:30] <Hackman238b> locohost: Yeah dont mess with stuff in there unless you've backed everything up and its not production.
[26-May-2011 19:46:43] <Hackman238b> locohost: Zope can be...unpredictable
[26-May-2011 19:49:21] <Hackman238b> locohost: Has it finished?
[26-May-2011 19:50:20] <locohost> sweet, it worked
[26-May-2011 19:50:25] <locohost> but it took some friends down with it
[26-May-2011 19:50:38] <locohost> zopectrl, zeoctrl and firefox all crashed, lol
[26-May-2011 19:50:50] <locohost> zenoss restart and it was gone though
[26-May-2011 19:50:50] <Hackman238b> locohost: wow. Ok, do this.
[26-May-2011 19:51:03] <Hackman238b> locohost: su zenoss
[26-May-2011 19:51:07] <Hackman238b> locohost: zendmd
[26-May-2011 19:51:28] <Hackman238b> locohost: rather, that zendmd, we'll use zenoss's tool
[26-May-2011 19:51:40] <Hackman238b> locohost: So su zenoss, zenchkrels -x 1 -r
[26-May-2011 19:51:52] <Hackman238b> locohost: Once done, zendmd
[26-May-2011 19:51:56] <Hackman238b> locohost: then reindex()
[26-May-2011 19:52:00] <Hackman238b> locohost: then sync()
[26-May-2011 19:52:03] <Hackman238b> locohost: then commit()
[26-May-2011 19:52:12] <Hackman238b> locohost: If anything bad did happen, that'll fix it
[26-May-2011 19:53:32] <locohost> cool, it is working, this is cleaning up junk i left behind i assume
[26-May-2011 19:53:35] <locohost> reindex is running
[26-May-2011 19:54:32] <locohost> just hope i dont regret not taking a vmware snapshot
[26-May-2011 19:54:47] <Hackman238b> locohost: Once its done open up the webui and test it out
[26-May-2011 19:55:39] <locohost> yep, i kicked zenoss for godo measure, working fine
[26-May-2011 19:55:43] <locohost> thanks
[26-May-2011 19:55:53] <Hackman238b> locohost: no sweat
[26-May-2011 19:55:54] <locohost> and now my boss wont click on that report and take down my server, lol
[26-May-2011 19:56:22] <locohost> the built in interface canned report is awesome
[26-May-2011 19:56:34] <Hackman238b> locohost: Yeah, its easy to lock up zope. If you have the resources you should multi-zope and reverse proxy load balance using httpd
[26-May-2011 19:56:46] <locohost> i can do expand and sort every interface on every switch i have by inbound traffic, that is so nice
[26-May-2011 19:57:27] <locohost> i am getting some pretty sick new F5's next month
[26-May-2011 19:57:30] kocolosk is now known as kocolosk|away
[26-May-2011 19:57:42] <locohost> and i mgiht move this to a physical machine, if im going to be trending the big catalysts
[26-May-2011 19:57:43] <Hackman238b> locohost: F5?
[26-May-2011 19:57:53] <locohost> we have 4 catalyst 6509;s
[26-May-2011 19:57:59] <locohost> F5 load balancer
[26-May-2011 19:58:02] <Hackman238b> locohost: Oh gotcha.
[26-May-2011 19:58:07] <JohnnyNoc> hey Hackman238
[26-May-2011 19:58:10] <JohnnyNoc> no luck on my graph
[26-May-2011 19:58:12] <JohnnyNoc>
[26-May-2011 19:58:35] <Hackman238b> JohnnyNoc: darn. PM me what you have and let me try to rework it.
[26-May-2011 19:58:36] <JohnnyNoc> all that stuff i added into the 'graph custom definition' tab seems to occur after all the original stuff
[26-May-2011 19:58:48] <JohnnyNoc> which would explain why it thought i was tryign to reuse 'datarate'
[26-May-2011 19:59:09] <JohnnyNoc> well i'm back to square 1, playing with this --step option
[26-May-2011 19:59:12] <Hackman238b> JohnnyNoc: Where is it showing that?
[26-May-2011 19:59:28] <JohnnyNoc> well, mind you, i'm using 2.5.3 so it may be dif thatn what you're looking at but
[26-May-2011 19:59:37] <JohnnyNoc> when i look at the template, and the graph definition
[26-May-2011 19:59:47] <Hackman238b> JohnnyNoc: Oh you are? Awesome- makes it easier :-)
[26-May-2011 19:59:59] <Hackman238b> JohnnyNoc: Copy what the graph commands page has
[26-May-2011 20:00:07] <JohnnyNoc> there are then other tabs for 'graph custom definition' and 'graph commands'
[26-May-2011 20:00:07] <JohnnyNoc> ah
[26-May-2011 20:00:21] <JohnnyNoc> well, as i said, i'm back to square one.  but after trying what you had suggested i noticed it was right there underneath all the original stuff
[26-May-2011 20:00:22] <JohnnyNoc> :/
[26-May-2011 20:00:39] <Hackman238b> JohnnyNoc: Its doable- I just forget how to do it.
[26-May-2011 20:00:40] <JohnnyNoc>
[26-May-2011 20:00:45] <JohnnyNoc> that's whats currently there
[26-May-2011 20:01:08] <locohost> hack, do you bother to set no atime on your zenoss filesystem?
[26-May-2011 20:01:31] <Hackman238b> JohnnyNoc: can you paste graph commands page too?
[26-May-2011 20:01:47] <JohnnyNoc> i didn't want to start mucking with the graph definitions for the collector, like, the default rrd create command
[26-May-2011 20:01:53] <JohnnyNoc> but i'm confident it's part of the puzzle
[26-May-2011 20:01:54] <JohnnyNoc> hacksure
[26-May-2011 20:02:06] <Hackman238b> locohost: I do
[26-May-2011 20:02:22] <JohnnyNoc> oh wait, that's what i already pasted
[26-May-2011 20:02:26] <JohnnyNoc> with the --step 1 on the bottom
[26-May-2011 20:02:33] <JohnnyNoc> that was added in the 'graph custom defintion' tab
[26-May-2011 20:02:43] <Hackman238b> JohnnyNoc: send that tab
[26-May-2011 20:02:51] <Hackman238b> JohnnyNoc: I forget which is which. LOL
[26-May-2011 20:02:55] <JohnnyNoc>
[26-May-2011 20:03:01] <JohnnyNoc> thats what i already pasted tho
[26-May-2011 20:03:02] <JohnnyNoc>
[26-May-2011 20:03:16] <Hackman238b> JohnnyNoc: Whats the other tab say?
[26-May-2011 20:03:17] <JohnnyNoc> the graph custom definition only has --step 1
[26-May-2011 20:03:27] <Hackman238b> JohnnyNoc: Oh gotcha
[26-May-2011 20:03:39] <JohnnyNoc> graph definition, graph custom definition, graph commands
[26-May-2011 20:03:50] <JohnnyNoc> my CF is max
[26-May-2011 20:03:54] <JohnnyNoc> my data point is a Gauge
[26-May-2011 20:03:56] <Hackman238b> JohnnyNoc: Gotcha
[26-May-2011 20:04:18] <Hackman238b> locohost: noatime is important if disk i/o is limited
[26-May-2011 20:05:15] <JohnnyNoc> so now my output in the script is minute.minute so it continually increases
[26-May-2011 20:05:25] <JohnnyNoc> like, 5.5
[26-May-2011 20:05:26] <JohnnyNoc> 6.6
[26-May-2011 20:05:27] <JohnnyNoc> and so on
[26-May-2011 20:05:32] <JohnnyNoc> but anyway
[26-May-2011 20:05:46] <JohnnyNoc> how do i only use the custom graph definition stuff instead of having it be appended to the original
[26-May-2011 20:05:55] <JohnnyNoc> thats what i need to know to figure out if what you were suggesting works
[26-May-2011 20:05:55] <JohnnyNoc>
[26-May-2011 20:06:40] <Hackman238b> JohnnyNoc: You cant- it'll append always
[26-May-2011 20:07:01] <locohost> yeah, cuts down on a lot of writes
[26-May-2011 20:07:45] <Hackman238b> JohnnyNoc: to your custom graph page add CDEF:PREVDataRate=PREV(DataRate)
[26-May-2011 20:08:05] <Hackman238b> JohnnyNoc: And refresh the graph- does it draw?
[26-May-2011 20:08:10] <locohost> know of any good netapp templates?
[26-May-2011 20:08:27] <locohost> debating if i want to do netapps in zenoss or cacti sdk plugin
[26-May-2011 20:08:39] <Hackman238b> locohost: The Ent one is good, but there is a NetApp template in the community zenpack section
[26-May-2011 20:09:00] <JohnnyNoc> Hackman238b i'll let you know..  but it will be a few minutes
[26-May-2011 20:09:08] <JohnnyNoc> need the graph to rrdcreate itself again
[26-May-2011 20:09:36] <Hackman238b> JohnnyNoc: Alrighty.
[26-May-2011 20:09:56] <Hackman238b> JohnnyNoc: I'll be around- I basically work perpetually. LOL
[26-May-2011 20:10:16] <locohost> cool,hoping to get the cpu's, iuops and latency on all aggregates/volumes and probably more importatnly traffic in e0 ...
[26-May-2011 20:10:34] <Hackman238b> locohost: I believe it does that, yes
[26-May-2011 20:10:54] <Hackman238b> The community pack is slightly broken in the sense you'll need to create the classes manually.
[26-May-2011 20:11:00] <locohost> the cacti sdk one uses https...which is kinda heavy
[26-May-2011 20:11:10] <Hackman238b> locohost: Let me send you a copy I fixed myself
[26-May-2011 20:11:12] <locohost> also, its really ineffeciant with the way it collects data
[26-May-2011 20:11:30] <locohost> and requires a root account
[26-May-2011 20:11:40] <locohost> i did it a few years ago, maybe its gotten better, but the data it got was really really nice
[26-May-2011 20:12:10] <locohost> cool, ur hae ur own netapp stuff?
[26-May-2011 20:12:35] <Hackman238b> locohost: I do.
[26-May-2011 20:13:49] <locohost> cool, i have 4 N5600 and 5 FAS6240
[26-May-2011 20:13:54] <locohost> s/5/4
[26-May-2011 20:14:23] <Hackman238b> locohost: v3.x, right?
[26-May-2011 20:14:30] <Hackman238b> locohost: Zenoss v3.x, rather
[26-May-2011 20:14:50] <locohost> zenoss-3.1.0-1031.el5
[26-May-2011 20:15:00] <JohnnyNoc> Hackman238b I don't see anything additional in my graph
[26-May-2011 20:15:01] <locohost> yep, pretty new
[26-May-2011 20:15:05] <JohnnyNoc> and the max is 11.21
[26-May-2011 20:15:14] <JohnnyNoc> which is wrong
[26-May-2011 20:15:15] <JohnnyNoc> heh
[26-May-2011 20:16:32] <Hackman238b> JohnnyNoc: The graph did draw though, right?
[26-May-2011 20:16:34] <JohnnyNoc> values have been like, 11.11, 12.12, 13.13
[26-May-2011 20:16:35] <JohnnyNoc> yes
[26-May-2011 20:16:36] <JohnnyNoc> it drew
[26-May-2011 20:16:46] <JohnnyNoc> but i don't see anything about prevdatarate
[26-May-2011 20:16:47] <Hackman238b> JohnnyNoc: Alrighty, one sec
[26-May-2011 20:18:14] <Hackman238b> JohnnyNoc: Add LINE1:PREVDataRate#00cc00ff:PREVDataRate
[26-May-2011 20:18:34] <JohnnyNoc> ok
[26-May-2011 20:18:35] <Hackman238b> JohnnyNoc: then redraw the graph. It shouldnt erase your rrd's
[26-May-2011 20:18:36] <JohnnyNoc> done and rm'd my rrd file
[26-May-2011 20:18:39] <JohnnyNoc> oh
[26-May-2011 20:18:41] <JohnnyNoc> i rm'd it
[26-May-2011 20:18:44] <JohnnyNoc> i thought i was supposed to
[26-May-2011 20:18:52] <Hackman238b> JohnnyNoc: You wont need to do that other than for consolitation function changes
[26-May-2011 20:19:17] <JohnnyNoc> hrm ok, hopefully i can remember that
[26-May-2011 20:19:18] <JohnnyNoc> heh
[26-May-2011 20:20:54] <locohost> hack, i sent you a spot to put it ot make it easy for you
[26-May-2011 20:21:12] <Hackman238b> locohost: Sent you a URL :-)
[26-May-2011 20:21:33] <JohnnyNoc> i still have no data in my graph, but i now see PREVDataRAte as a datasource
[26-May-2011 20:21:58] <Hackman238b> JohnnyNoc: Once it has data tell me if you get a line for PREV data
[26-May-2011 20:22:17] <JohnnyNoc> will do, looks like it will though
[26-May-2011 20:22:32] <JohnnyNoc> just by virtue of the fact it's there, and i don't see cur/avg/max
[26-May-2011 20:24:13] <JohnnyNoc> elmhurt in da houz
[26-May-2011 20:24:16] <JohnnyNoc> elmhurst too
[26-May-2011 20:24:23] <Hackman238b> JohnnyNoc: ?
[26-May-2011 20:24:30] <JohnnyNoc> i'm still waiting
[26-May-2011 20:24:37] <JohnnyNoc> i was saying that in reference to dhopp's hostname
[26-May-2011 20:24:49] <locohost> yep, ty much, ill check it out tomorrow
[26-May-2011 20:25:12] <Hackman238b> JohnnyNoc: Oh LOL
[26-May-2011 20:25:26] <Hackman238b> locohost: NP.
[26-May-2011 20:25:51] <JohnnyNoc> hackman, i don't see a line yet
[26-May-2011 20:26:00] <Hackman238b> God I hate Gnome 3....I'm going to have to roll this back. Ugh.
[26-May-2011 20:26:01] <JohnnyNoc> but even if i did, i wouldn't see it cuz it's the same color as my data point
[26-May-2011 20:26:01] <JohnnyNoc> heh
[26-May-2011 20:26:07] <JohnnyNoc> but, it's max is currently 24.24
[26-May-2011 20:26:21] <JohnnyNoc> so is cur, and avg tho
[26-May-2011 20:26:21] <JohnnyNoc>
[26-May-2011 20:26:45] <Hackman238b> JohnnyNoc: Let me think about this. LOL
[26-May-2011 20:26:58] <JohnnyNoc> haha
[26-May-2011 20:27:06] <JohnnyNoc> well, that too me sounds like it only got that output thus far
[26-May-2011 20:27:18] <JohnnyNoc> just 1 data point/source/whatever
[26-May-2011 20:27:25] <JohnnyNoc> i wanna see what it does when it updates
[26-May-2011 20:28:17] <dhopp> except I'm not in elmhurst
[26-May-2011 20:28:18] <dhopp> lol
[26-May-2011 20:28:24] <dhopp> about 10 miles from there
[26-May-2011 20:28:36] <Hackman238b> dhopp: LOL
[26-May-2011 20:29:48] <JohnnyNoc> it should have updated by now
[26-May-2011 20:29:53] <JohnnyNoc> still at 24.24 across the baord
[26-May-2011 20:30:07] <JohnnyNoc> the command has output as high as 29.29 now
[26-May-2011 20:30:10] <JohnnyNoc> it's supposed to run every 60s
[26-May-2011 20:30:11] <JohnnyNoc> ugh
[26-May-2011 20:30:52] <JohnnyNoc> finally updated but it's still doing some averaging
[26-May-2011 20:30:58] <JohnnyNoc> cur and max are 30.22
[26-May-2011 20:31:35] <locohost> i have not tried gnome 3 yet, but i hate grub3
[26-May-2011 20:32:03] <Hackman238b> JohnnyNoc: I dont think you can do it. I cant figure a way.
[26-May-2011 20:32:19] <Hackman238b> JohnnyNoc: I'll add it to my list of challenges
[26-May-2011 20:32:27] <Hackman238b> locohost: LOL
[26-May-2011 20:32:29] <JohnnyNoc> i think, based upon what i read previously, my problem here is i'm looking for --step 1 type accuracy
[26-May-2011 20:32:34] <JohnnyNoc> where it runs every second
[26-May-2011 20:32:41] <JohnnyNoc> like, RRD is just going to average that shit either way
[26-May-2011 20:33:54] <Hackman238b> JohnnyNoc: I think its being consolidated before its added to the rrd file
[26-May-2011 20:34:28] <Hackman238b> JohnnyNoc: I'm going to have to email Tobi about this shit. Seems very implementable.
[26-May-2011 20:34:36] <Hackman238b> LOL
[26-May-2011 20:34:36] <JohnnyNoc> man
[26-May-2011 20:34:39] <JohnnyNoc> good luck with that
[26-May-2011 20:34:39] <JohnnyNoc> heh
[26-May-2011 20:34:52] <JohnnyNoc> i think toby says true min/max is not possible, again, based on something i read earlier
[26-May-2011 20:41:04] <dhopp> god this vpn connection is painful
[26-May-2011 20:42:00] <Hackman238b> JohnnyNoc: I'll try to squeeze a reply out of him.
[26-May-2011 20:42:11] <Hackman238b> dhopp: Sporting a 33.6?
[26-May-2011 20:42:13] <locohost> not sure i follow, i think if you use MAX when it consolidates, it will use the highest point from that time frame
[26-May-2011 20:42:23] <JohnnyNoc> locohost it doesn't
[26-May-2011 20:42:24] <JohnnyNoc>
[26-May-2011 20:42:25] <dhopp> Hackman238b: you would think lol
[26-May-2011 20:42:42] <dhopp> Hackman238b: but it's VPN to Italy…so I have to cut it some slack
[26-May-2011 20:42:42] <Hackman238b> locohost: Yeah it isnt for some reason- its madness!
[26-May-2011 20:43:00] <locohost> ive seen it work before
[26-May-2011 20:43:08] <locohost> are you sure you doint have a 32bit counter or something
[26-May-2011 20:43:11] <Hackman238b> dhopp: Bah! Thats just across town for electrons :-P
[26-May-2011 20:43:24] <dhopp> Hackman238b: Ha
[26-May-2011 20:43:24] <Hackman238b> locohost: If you can get it to work, fill me in.
[26-May-2011 20:43:55] <dhopp> Hackman238b: what I've been doing has made me REALLY appreciate our automation for our application
[26-May-2011 20:43:56] <Hackman238b> dhopp: Must be all the fancy routing and packet inspection byt the government slowing it down
[26-May-2011 20:44:11] <dhopp> been setting up many instances all manual...
[26-May-2011 20:44:15] <locohost> a few years back i changed our graphs to MAX rather then average because we billed clients off of them and it made the trends a little higher
[26-May-2011 20:44:17] <JohnnyNoc> locohost this is a custom command
[26-May-2011 20:44:23] <locohost> assumed it was working..but didnt stop and do all the math
[26-May-2011 20:44:24] <dhopp> If I have to do this again I will cry
[26-May-2011 20:44:24] <Hackman238b> dhopp: Oh yeah, it can be painful.
[26-May-2011 20:44:38] <JohnnyNoc> i dont see/know of any 32/64 bit counters involved here
[26-May-2011 20:44:47] <dhopp> Hackman238b: this isn't for zenoss :-)..but our java app with a sql backend
[26-May-2011 20:44:49] <Hackman238b> dhopp: Automation is the only way. :-)
[26-May-2011 20:45:03] <dhopp> I haven't done a manual setup in so long that I have been making just stupid mistakes left and right
[26-May-2011 20:45:24] <locohost> maybe not johnny, sounds like you troubleshoot past that, i was thinking maybe bacause the value was not has high as you expected it to be it could be rollover or osmething
[26-May-2011 20:45:36] <Hackman238b> dhopp: That'll happen. Automation does make one a tad lazy. *cough* tab-complete :-)
[26-May-2011 20:46:02] <JohnnyNoc> But what if you really would like to know the highest (or lowest) rate seen? You will need to store those as well. Just create another RRA in your RRD, with the CF you desire. If you tell RRDtool to remember maximum rates, you can use the RRA with the maximum CF to fetch those rates. Similarly for minimum, average and last.
[26-May-2011 20:46:10] <JohnnyNoc> i think that's what i need to attack here
[26-May-2011 20:46:15] <JohnnyNoc> and create another RRA
[26-May-2011 20:46:22] <locohost> come to think if it...
[26-May-2011 20:46:23] <JohnnyNoc> i noticed there's no RRA for what look like 1 minute intervals
[26-May-2011 20:46:34] <JohnnyNoc> when looking at the default settings for this collector
[26-May-2011 20:46:35] <dhopp> Hackman238b: Ha!…I remember showing somebody how to do something on a cisco router and he was like "geez you type like 200 words per minute"…my response "tab is your friend"
[26-May-2011 20:46:45] <locohost> i may have made the RRD file really big and store unconolidated data for a long time
[26-May-2011 20:46:59] <JohnnyNoc> hopefully it was an old job?
[26-May-2011 20:47:00] <JohnnyNoc>
[26-May-2011 20:47:17] <dhopp> Hackman238b: and the he asked how I remembered all the response "tab is your friend"
[26-May-2011 20:47:18] <dhopp> hehe
[26-May-2011 20:48:07] <Hackman238b> dhopp: Yep. I've interviewed a few people recently who included 'tab' in IOS literacy tests LOL
[26-May-2011 20:51:00] <dhopp> Hackman238b: lol…but in all honesty, I'd rather hire someone that clearly knows the steps and can use that knowlege to lookup what he/she needs then somebody that is an encyclopedia of commands.  Sometimes they know the commands, but don't know how to use them
[26-May-2011 20:52:28] <Hackman238b> dhopp: Absolutely. The most useful people are the people who know how to properly learn
[26-May-2011 20:52:42] <Hackman238b> dhopp: I never hire for any cert or degree- even a masters.
[26-May-2011 20:53:16] <Hackman238b> dhopp: Only exception would be CCIE- you cant get the IE being a memorization chode
[26-May-2011 20:55:18] <dhopp> sometimes I wonder how windows guys get by without sed
[26-May-2011 20:55:36] <Hackman238b>  *reminds self of something to watch while working- tripping the rift*
[26-May-2011 20:55:44] <Hackman238b> dhopp: Robitic madness is how!
[26-May-2011 20:55:46] <Hackman238b> :-)
[26-May-2011 20:55:54] <Hackman238b> *robotic
[26-May-2011 21:26:30] <Hackman238b> Have to bounce. Later all!
[27-May-2011 00:00:01] [disconnected at Fri May 27 00:00:01 2011]
[27-May-2011 00:00:02] [connected at Fri May 27 00:00:02 2011]
[27-May-2011 00:00:20] [zenoss-logger (logger bot) has joined #zenoss]
[27-May-2011 00:00:20] <> [freenode-info] please register your nickname...don't forget to auto-identify!
[27-May-2011 07:41:30] <hmp> zenperfsnmp daemon is not monitored by heartbeat when using distributed collectors?
[27-May-2011 07:42:03] fOrsberg is now known as forsberg
[27-May-2011 08:19:21] forsberg is now known as fOrsberg
[27-May-2011 09:23:16] <Sam-I-Am> moo.
[27-May-2011 09:25:28] <dhopp> quack
[27-May-2011 09:27:32] <Sam-I-Am> sup
[27-May-2011 09:30:32] kocolosk|away is now known as kocolosk
[27-May-2011 09:33:58] <Hackman238> Hey all
[27-May-2011 09:59:40] <kokey> hey Hackman238
[27-May-2011 09:59:50] <Sam-I-Am> yo yo
[27-May-2011 09:59:56] <kokey> i'm wondering, what's a good way to create a report of all the current critical alerts?
[27-May-2011 10:01:55] <Sam-I-Am> custom reports
[27-May-2011 10:01:59] <Sam-I-Am> probably not too difficult
[27-May-2011 10:05:41] <kokey> not very well documented it seems
[27-May-2011 10:05:45] <kokey> but i'll give it a go
[27-May-2011 10:05:50] <kokey> examples would have been nice
[27-May-2011 10:08:18] <kokey> is there a way to see the source of some of the built in reports?
[27-May-2011 10:08:43] <kokey> i might have to write a script that gets the events from the mysql db
[27-May-2011 10:08:55] <kokey> it's my last day at this gig and i have 2 hours to deliver something
[27-May-2011 10:11:27] <Sam-I-Am> you're leaving, so why care? lol
[27-May-2011 10:14:59] <kokey> bad habit
[27-May-2011 10:15:47] <kokey> ok bugger it, i'm going to do a bodge
[27-May-2011 10:16:02] <kokey> perl script, doing mysql get, sending an e-mail
[27-May-2011 10:19:52] <Sam-I-Am> that works
[27-May-2011 10:25:13] <Hackman238> kokey: Last day?
[27-May-2011 10:25:24] <Hackman238> kokey: No more Zenoss?
[27-May-2011 10:28:18] <kokey> no more zenoss for a while
[27-May-2011 10:28:24] <kokey> next gig, subject to background checks
[27-May-2011 10:28:39] <kokey> will involve doing custom dashboards, but looks like in a mostly tivoli environment
[27-May-2011 10:28:42] <kokey> and doing cfengine work
[27-May-2011 10:29:09] <kokey> but i suspect zenoss will come back my way
[27-May-2011 10:38:02] <f00fSteR> hey guys
[27-May-2011 10:38:10] <f00fSteR> how do i clear entire events history ?
[27-May-2011 10:39:36] <dhopp> f00fSter: you want to remove all event history, or just a particular event?
[27-May-2011 10:39:43] <f00fSteR> all event history
[27-May-2011 10:39:48] <f00fSteR> i just wanna clear this shit
[27-May-2011 10:42:41] <dhopp> f00fSteR: you want it completely gone (i.e. out of the mysql db completely)?
[27-May-2011 10:42:44] <kokey> drop the events table
[27-May-2011 10:42:51] <kokey> works for me, YMMV
[27-May-2011 10:43:15] <kokey> i'm joking
[27-May-2011 10:43:29] <dhopp> kokey: although that would work :-P
[27-May-2011 10:43:39] <kokey> what you can do, tho, is just do a 'delete from tablename' on all the events tables
[27-May-2011 10:43:47] <kokey> i mean't the events DB
[27-May-2011 10:44:09] <kokey> you can recreate it too,  zeneventbuild localhost zenoss zenossuserpassword events
[27-May-2011 10:44:12] <kokey> meant
[27-May-2011 10:44:17] <kokey> i've done that a few times
[27-May-2011 10:44:37] <dhopp> kokey: that's where I was going to go…I just wanted to make sure I understood that that is what f00fSteR wanted
[27-May-2011 10:54:28] <kokey> i had three pints of beer over lunch
[27-May-2011 10:54:33] <kokey> so take what I say with a pinch of salt
[27-May-2011 10:54:43] <kokey> and some chicken wings
[27-May-2011 10:56:28] <dhopp> kokey: I don't think there is a way in the GUI to say "remove history from events right now"…probably could do it through the command line...
[27-May-2011 10:57:07] <dhopp> kokey: so I think getting into mysql is the only way….
[27-May-2011 10:57:19] <dhopp> Hackman238b: can you verify what I'm saying?  Or am I crazy
[27-May-2011 10:59:31] <kokey> perhaps just go to the events console, select all, and delete events
[27-May-2011 10:59:40] <Hackman238> kokey: Im so sorry...tivoli :-(
[27-May-2011 11:00:01] <kokey> Hackman238: well, fortunately there's a whole team doing all the tivoli stuff
[27-May-2011 11:00:11] <kokey> they've got the collection part covered
[27-May-2011 11:00:22] <kokey> all I got to do is extract data and display it
[27-May-2011 11:01:15] <Hackman238> f00fSteR: drop the events table and rebuild the tables
[27-May-2011 11:01:27] <Hackman238> or truncate the events tables
[27-May-2011 11:27:38] <f00fSteR> i know how to drop tables... how does one rebuild
[27-May-2011 11:28:00] <Hackman238> f00fSteR: v2.5?
[27-May-2011 11:28:16] <f00fSteR> Hackman238, ahoyy yee remembers
[27-May-2011 11:30:28] <Hackman238> /opt/zenoss/Products/ZenEvents/db/
[27-May-2011 11:30:51] <f00fSteR> hrmm
[27-May-2011 11:30:59] <f00fSteR> i see's a zenevents .sql table
[27-May-2011 11:31:20] <Hackman238> f00fSteR: No worries, it'll wort. use the zenevents.sql or run loadZenEvents
[27-May-2011 11:32:35] <f00fSteR> wort
[27-May-2011 11:32:35] <f00fSteR> argg
[27-May-2011 11:32:42] <f00fSteR> i'm getting internal server errors from our monitoring solution stating: WebApplicationImpl : Internal Server Error... anyone have any diagnosis for this ? or an idea of what it could be.
[27-May-2011 11:32:57] <Hackman238> f00fSteR: Where are you seeing this?
[27-May-2011 11:33:05] <f00fSteR> just a zenoss event alert
[27-May-2011 11:33:16] <f00fSteR> was just asking... wonder if anyone knew
[27-May-2011 11:33:19] <f00fSteR> googling now
[27-May-2011 11:33:22] <Hackman238> f00fSteR: I would consult sir google
[27-May-2011 11:33:28] <Hackman238> f00fSteR: Yep. LOL
[27-May-2011 11:33:55] <f00fSteR> networking hiccup
[27-May-2011 11:33:57] <f00fSteR>
[27-May-2011 11:34:04] <f00fSteR> ok so these hiccups
[27-May-2011 11:34:08] <f00fSteR> ... is there a fix
[27-May-2011 11:34:31] <f00fSteR> to not get notified
[27-May-2011 11:46:27] <Hackman238> f00fSteR: To prevent the event or prevent alert?
[27-May-2011 11:53:14] <locohost> hrm, trying to add a few zenpacks through gui and i dont seem them under advanced, zenpackes but they did appear in /opt/zenoss/ZenPacks
[27-May-2011 11:54:07] <locohost> and if i do zenpack --remove ZenPacks.Nova.Cisco.Catalyst-1.0.egg on the folder, it says its not installed
[27-May-2011 11:57:35] <Sam-I-Am> i usually add them via command line
[27-May-2011 11:58:00] <Sam-I-Am> locohost: might have half-installed
[27-May-2011 11:58:13] <Hackman238> locohost: just beacuse its in $ZENHOME/ZenPacks does not mean its installed.
[27-May-2011 11:58:40] <Hackman238> locohost: I highly reccomend you do this work at the CLI all the time.
[27-May-2011 11:58:59] <Hackman238> locohost: For community packs, I mean
[27-May-2011 12:00:01] [disconnected at Fri May 27 12:00:01 2011]
[27-May-2011 12:00:02] [connected at Fri May 27 12:00:02 2011]
[27-May-2011 12:00:20] [zenoss-logger (logger bot) has joined #zenoss]
[27-May-2011 12:05:15] <locohost> okay, so should I rm -rf the sub dir under zenpack and do it from cli?
[27-May-2011 12:05:24] <f00fSteR> anyone know what this error is under ssl:
[27-May-2011 12:05:24] <f00fSteR> (Error code: sec_error_revoked_certificate)
[27-May-2011 12:05:27] <locohost> or is there a cleaner way to back the half install out
[27-May-2011 12:15:44] <locohost>     Unable to install ZenPacks.Nova.Cisco.Catalyst-1.0.egg
[27-May-2011 12:15:49] <locohost> which the error was more verbose
[27-May-2011 12:16:12] <locohost> thats  hits in the console as an event from my server when using zenpack --install
[27-May-2011 12:16:54] <Hackman238> locohost: Reinstall the pack
[27-May-2011 12:16:58] <Hackman238> locohost: then remove it
[27-May-2011 12:17:45] <locohost> its not installed now, not sure i follow, install and remove?
[27-May-2011 12:20:21] <locohost> tried zipping it back up and not running the install on hte egg, but the zip (not sure which way it wants it) but it appears to get futher,  tells me
[27-May-2011 12:20:22] <locohost> OSError: [Errno 20] Not a directory: '/opt/zenoss/Products/ZenPacks.Nova.Cisco.Catalyst-1.0.egg/skins'
[27-May-2011 12:21:29] <locohost> indeed the skins dir does not exist there...zenoss ownes everything under /opt/zenoss so shouldnt be a parm issue
[27-May-2011 12:22:33] <locohost> okay, no, its not a dir, just the .egg file put in there from the zenpack --install
[27-May-2011 12:26:08] <locohost> may just be i need to create the devic eclass first on the cisco one, lets see
[27-May-2011 12:30:50] <Hackman238> locohost: /opt/zenoss/ZenPacks/* should be all directories
[27-May-2011 12:31:01] <Hackman238> locohost: Dont put .egg files in there- it'll confise bash.
[27-May-2011 12:31:07] <Hackman238> *confuse
[27-May-2011 12:31:36] <Hackman238> locohost: put your packs in /opt/zenoss/packs/ as aregular practice
[27-May-2011 12:32:16] <Hackman238> locohost: Never chown zenoss:zenoss /opt/zenoss -R, this will break permissions.
[27-May-2011 12:32:26] <Hackman238> locohost: Same for chmod
[27-May-2011 12:32:51] <Hackman238> locohost: If the pack is not installing, be sure it isnt double zipped.
[27-May-2011 12:33:04] <Hackman238> locohost: wget it to /opt/zenoss/packs
[27-May-2011 12:33:10] <Hackman238> locohost: cd /opt/zenoss/packs
[27-May-2011 12:33:22] <Hackman238> locohost: mv file
[27-May-2011 12:33:29] <locohost> [zenoss@d01mtrending01 ~]$ zenpack -v 1 --install /opt/zenoss/packs/
[27-May-2011 12:33:29] <Hackman238> locohost: unzip
[27-May-2011 12:33:29] <locohost> DEBUG:zen.ZenPackCmd:Extracting ZenPack "ZenPacks.Nova.Cisco.Catalyst-1.0.egg"
[27-May-2011 12:33:29] <locohost> DEBUG:zen.ZenPackCmd:Extracting ZenPacks.Nova.Cisco.Catalyst-1.0.egg
[27-May-2011 12:33:29] <locohost> ERROR:zen.ZenPackCmd:zenpack command failed
[27-May-2011 12:34:01] <locohost> ahh
[27-May-2011 12:34:12] <locohost> okay, yeah, that looks like it works
[27-May-2011 12:34:15] <Hackman238> locohost: is dumb, but the packs get double zipped sometimes
[27-May-2011 12:34:47] <Hackman238> locohost:  if the resulting file is another single file, rename it to .egg, if its not already, and zenpack --install file
[27-May-2011 12:35:57] <f00fSteR> Hackman238, I just got an alert for an event 2 hours after it happened... even the timestamp is fro 2 hours ago ... any idea why this is happening ?
[27-May-2011 12:36:12] <Hackman238> locohost: f00fSteR Is it sms?
[27-May-2011 12:36:23] <Hackman238> f00fSteR: Is it SMS or email?
[27-May-2011 12:36:30] <Hackman238> locohost: did it work?
[27-May-2011 12:36:39] <f00fSteR> Hackman238, e-mail
[27-May-2011 12:36:41] <locohost> yeah, catalsyst one worked, ganna do the netapp one now
[27-May-2011 12:37:06] <f00fSteR> Hackman238, on top of that it was for a warning ... i mean i have a delay set for 120 MS that shouldnt delay it for 120 minutes tho
[27-May-2011 12:37:10] <Hackman238> locohost: Alrighty. my packs are not double zipped- the one I sent you a link too should be good
[27-May-2011 12:37:39] <Hackman238> f00fSteR: is sendmail, postfix or your mail server backed up/slow?
[27-May-2011 12:38:10] <Hackman238> f00fSteR: is zenactions busy? zenactions -f /opt/zenoss/log/zenactions.log
[27-May-2011 12:38:20] <Hackman238> f00fSteR: If its busy it'll look so
[27-May-2011 12:42:09] <f00fSteR> i have 2.5
[27-May-2011 12:42:15] <f00fSteR> that command doesnt do the same
[27-May-2011 12:42:27] <f00fSteR> $ zenactions -f /opt/zenoss/log/zenactions.log
[27-May-2011 12:42:28] <f00fSteR> Usage: /opt/zenoss/bin/zenactions {run|start|stop|restart|status|help|genconf|genxmlconfigs} [options]
[27-May-2011 12:42:31] <Hackman238> f00fSteR: Im so sorry, my fail- Im so busy here
[27-May-2011 12:42:40] <Hackman238> f00fSteR: I mean tail -f /opt/zenoss/log/zenactions.log
[27-May-2011 12:42:40] <f00fSteR> np dude take your time man
[27-May-2011 12:44:08] <Hackman238> f00fSteR: Active dev, Avalon testing/debug/dev, cloud fire at my personal DC, firewall doom...its madness today!
[27-May-2011 12:44:40] <f00fSteR> sounds f'in exciting man
[27-May-2011 12:44:48] <f00fSteR> i love the probvlems that come up before a long weekend
[27-May-2011 12:44:54] <Hackman238> f00fSteR: Its all under control, but its very busy here
[27-May-2011 12:45:06] <f00fSteR> SSL certs revoked... tomcat failing... log files paring up the kazuu... it's a good day to die
[27-May-2011 12:45:07] <Hackman238> f00fSteR: You and me both.
[27-May-2011 12:45:17] <Hackman238> f00fSteR: LOL!
[27-May-2011 12:45:29] <Hackman238> f00fSteR: Hows that tail look?
[27-May-2011 12:45:54] <f00fSteR> all of it seems to be processing in less than 4 seconds
[27-May-2011 12:46:48] <f00fSteR> i have a few of these
[27-May-2011 12:46:49] <f00fSteR> tail -f /opt/zenoss/log/zenactions.log
[27-May-2011 12:46:55] <f00fSteR> OperationalError: (1205, 'Lock wait timeout exceeded; try restarting transaction')
[27-May-2011 12:46:58] <f00fSteR> ^^
[27-May-2011 12:47:08] <Hackman238> f00fSteR: timesatme?
[27-May-2011 12:47:16] <Hackman238> *timestamp
[27-May-2011 12:47:32] <f00fSteR> ahh --> 2011-05-27 12:27:14 ERROR zen.ZenActions: unexpected exception
[27-May-2011 12:47:49] <Hackman238> f00fSteR: do zenactions stop
[27-May-2011 12:47:55] <Hackman238> f00fSteR: then ps aux | grep zenactions
[27-May-2011 12:48:03] <Hackman238> f00fSteR: if it didnt stop, kill -9 it
[27-May-2011 12:48:17] <Hackman238> f00fSteR: then do zenactions start; zenactions debug
[27-May-2011 12:48:29] <Hackman238> f00fSteR: next tail -f that log file
[27-May-2011 12:51:35] <f00fSteR> i dont have debug command
[27-May-2011 12:51:39] <f00fSteR> run ?
[27-May-2011 12:51:45] <f00fSteR> /opt/zenoss/bin/zenactions {run|start|stop|restart|status|help|genconf|genxmlconfigs} [options]
[27-May-2011 12:51:50] <Hackman238> zenactions debug will work
[27-May-2011 12:52:07] <f00fSteR> negative
[27-May-2011 12:52:24] <Hackman238> f00fSteR: stack? what version of 2.5?
[27-May-2011 12:52:32] <Hackman238> f00fSteR: LOL it sounds really old
[27-May-2011 12:53:22] <f00fSteR> lol 2.5 core
[27-May-2011 12:53:54] <f00fSteR> man i cant even change ti bc some oft he plugins dont work with the newer versions
[27-May-2011 12:54:36] <Hackman238> f00fSteR: not 2.5.1 or 2.5.2 or ... 2.4.5?
[27-May-2011 13:03:25] <locohost> foo, if you were to send me the headers of hte message, I would be happy to take a look and tell you where the delay is
[27-May-2011 13:03:35] <f00fSteR> Version 2.4.2
[27-May-2011 13:03:47] <locohost> of your delayed message, that is
[27-May-2011 13:03:49] <Hackman238> f00fSteR: Oh my
[27-May-2011 13:03:54] <f00fSteR>
[27-May-2011 13:03:59] <f00fSteR> my o my
[27-May-2011 13:03:59] <Hackman238> f00fSteR: You need to upgrade that
[27-May-2011 13:04:07] <f00fSteR> sigh
[27-May-2011 13:04:09] <f00fSteR> i cant in production
[27-May-2011 13:04:14] <f00fSteR> i need to copy the instance
[27-May-2011 13:04:17] <f00fSteR> into a test instance
[27-May-2011 13:04:19] <f00fSteR> upgrade it
[27-May-2011 13:04:24] <Hackman238> f00fSteR: I cant even begin to tell you what isnt working/what will work on that puppy
[27-May-2011 13:04:25] <f00fSteR> and then push it
[27-May-2011 13:04:36] <f00fSteR> sigh
[27-May-2011 13:04:36] <Hackman238> f00fSteR: Lets try this
[27-May-2011 13:04:54] <f00fSteR> Hackman238, this is depressing
[27-May-2011 13:05:05] <f00fSteR> i'm about to go wget nagios and fire it up
[27-May-2011 13:05:06] <Hackman238> f00fSteR: On a centos/redhat5 64-bit box install zenoss 2.5.2
[27-May-2011 13:05:12] <f00fSteR> ok
[27-May-2011 13:05:18] <f00fSteR> that'll take 2 mins
[27-May-2011 13:05:20] <f00fSteR>
[27-May-2011 13:05:21] <f00fSteR> BRB
[27-May-2011 13:05:22] <Hackman238> f00fSteR: do native packages
[27-May-2011 13:05:32] <f00fSteR> ok
[27-May-2011 13:05:38] <f00fSteR> starting
[27-May-2011 13:05:41] <Hackman238> f00fSteR: install the core packs if your current install has them
[27-May-2011 13:06:03] <Hackman238> f00fSteR: let it start up and test the GUI. Once its working ping me and we'll continue
[27-May-2011 13:06:15] <f00fSteR> you got it
[27-May-2011 13:06:23] <f00fSteR> i'm gonna install it get the gui up and running
[27-May-2011 13:06:35] <f00fSteR> and then go pray ... it;s friday my sabbath... i'll see you guys in an hour
[27-May-2011 13:06:40] <f00fSteR> but i'll make sure to get it up and running first
[27-May-2011 13:06:43] <f00fSteR> 5 mins max
[27-May-2011 13:09:17] <f00fSteR> is there a diff between el4 el5  ?
[27-May-2011 13:10:35] <f00fSteR> 5 it is...
[27-May-2011 13:10:57] <Hackman238> f00fSteR: yes, use 5. 4 is very old.
[27-May-2011 13:11:24] <Hackman238> f00fSteR: almost requires knowledge of sourcery to get newer versions of zenoss to run on it.
[27-May-2011 13:11:50] <f00fSteR> haha... in the IT world.. we are masters of sourcery
[27-May-2011 13:11:58] <f00fSteR> when i can fix an entire server with the add of a #
[27-May-2011 13:12:03] <f00fSteR> i know notthing else
[27-May-2011 13:12:26] <Hackman238> f00fSteR: comment the kernel ine in grub.conf, reboot- fixed!
[27-May-2011 13:12:32] <f00fSteR> just be glad i dont yell out sim sim sala bim every time i do it
[27-May-2011 13:12:46] <Hackman238> f00fSteR: lol
[27-May-2011 13:16:14] <f00fSteR> FML... dependencies
[27-May-2011 13:16:15] <f00fSteR> lol
[27-May-2011 13:17:33] <Hackman238> f00fSteR: chill
[27-May-2011 13:17:41] <Hackman238> f00fSteR: yum localinstall zenosspackage -y
[27-May-2011 13:17:49] <Hackman238> f00fSteR: dont use RPM
[27-May-2011 13:18:04] <Hackman238> f00fSteR: it might work sometimes, but you'll wish you didnt in the future
[27-May-2011 13:18:16] <f00fSteR> wait what ?
[27-May-2011 13:18:20] <f00fSteR> it's done already
[27-May-2011 13:18:26] <f00fSteR> -e ?
[27-May-2011 13:18:36] <Hackman238> f00fSteR: oh ok good
[27-May-2011 13:18:48] <Hackman238> f00fSteR: Not everyone can handle rpm, if you got it, your fine
[27-May-2011 13:19:07] <f00fSteR> hrmm i didnt add any database information
[27-May-2011 13:19:13] <f00fSteR> how is it loading database ?
[27-May-2011 13:19:15] <f00fSteR> daemon process started, pid=874
[27-May-2011 13:19:15] <f00fSteR> Loading initial Zenoss objects into the Zeo database
[27-May-2011 13:19:15] <f00fSteR> (this can take a few minutes)
[27-May-2011 13:19:26] <Hackman238> f00fSteR: Energy from the void
[27-May-2011 13:19:35] <f00fSteR> ahh
[27-May-2011 13:19:36] <Hackman238> f00fSteR: Rather it has a little template.
[27-May-2011 13:19:37] <f00fSteR> zopedb ?
[27-May-2011 13:19:51] <Hackman238> f00fSteR: Yep. It'll take a bit depending on the speed of the box
[27-May-2011 13:19:57] <f00fSteR> soo many frriggin ZZz's
[27-May-2011 13:20:13] <f00fSteR> zope zeo zen zenoss zero on my xen server ahhh
[27-May-2011 13:20:25] <Hackman238> f00fSteR: LOL
[27-May-2011 13:21:01] <Hackman238> f00fSteR: Once its done, check the gui. if it works and has a little getting started thing, service zenoss stop
[27-May-2011 13:21:55] <Hackman238> f00fSteR: Next copy /opt/zenoss/var/Data.* to the /opt/zenoss/var/ on the target server. replace everything
[27-May-2011 13:21:56] <f00fSteR> zenoss setup ?
[27-May-2011 13:21:57] <f00fSteR> yepp
[27-May-2011 13:22:20] <Hackman238> f00fSteR: once the files are copied, double check with ls -lh that its zenoss:zenoss
[27-May-2011 13:22:34] <Hackman238> f00fSteR: next, as zenoss user, issue zeoctl start
[27-May-2011 13:23:23] <Hackman238> f00fSteR: f00fSteR next, as zenoss user, issue zenmigrate
[27-May-2011 13:23:50] <Hackman238> f00fSteR: it'll do a bunch of stuff. you;ll almost certainly have some problems, but we'll work through them
[27-May-2011 13:24:40] <f00fSteR> mann you ARE an expert
[27-May-2011 13:24:50] <Hackman238> f00fSteR: :-)
[27-May-2011 13:24:54] <f00fSteR> tell me you live in NY ?
[27-May-2011 13:24:59] * Sam-I-Am tries to figure out inter-as bgp vpls stuff
[27-May-2011 13:24:59] <f00fSteR> you deserve dinner and some drinks
[27-May-2011 13:25:00] <Hackman238> f00fSteR:
[27-May-2011 13:25:13] <Sam-I-Am> theres a blackboard in here where we mark how much beer we owe people
[27-May-2011 13:25:16] <Sam-I-Am> rmatte is on the top
[27-May-2011 13:25:18] <Hackman238> f00fSteR: I have an office in upstate NY, but I live in San Antonio Tx now
[27-May-2011 13:25:52] <Hackman238> f00fSteR: If you ever need zenpack dev or any of that, contact me and we'll set something up.
[27-May-2011 13:26:35] <f00fSteR> Hackman238, dude, i will assuredly be needing some dev plugins... i'mm be sure to consult your services.
[27-May-2011 13:26:47] <f00fSteR> you are skillful ninja of the python tekniques ?
[27-May-2011 13:30:27] <locohost> Cmd: /opt/zenoss/ZenPacks/ -H -C xxxx -l filesystem -f 77 - Code: 2 - Msg: Misuse of shell builtins
[27-May-2011 13:31:52] <locohost> hrm, its possible that it worked...
[27-May-2011 13:33:02] <locohost> hows my voliumes, but no free or used bytes, maybe that will populate after a few polls..
[27-May-2011 13:33:32] <locohost> if the nfsv3calls graph populates, I will be 1 happy camper, this is pretty
[27-May-2011 13:39:33] <Hackman238> f00fSteR: Yep
[27-May-2011 13:39:49] <Hackman238> locohost: your missing the Net::SNMP module
[27-May-2011 13:39:53] <Hackman238> locohost: yum install it
[27-May-2011 13:41:41] <Hackman238> f00fSteR: Just let me know and I'll get back to you ASAP. BTW, excuse my slow site
[27-May-2011 13:44:21] <locohost> cool
[27-May-2011 13:45:58] <locohost> how bout that, ro filesystem on my repo server, lol
[27-May-2011 13:46:15] <Hackman238> lol
[27-May-2011 13:46:36] <locohost> network issue last week, lost its disks,
[27-May-2011 13:49:37] <Hackman238> locohost: Never good
[27-May-2011 13:49:42] <Hackman238> locohost: pack working?
[27-May-2011 13:49:54] <Hackman238> f00fSteR: How goes the zenmigrate?
[27-May-2011 14:07:12] <locohost> damn walkups
[27-May-2011 14:07:25] <locohost> got distracted, its not in my repo, got to find the rpm, dont need help with that though
[27-May-2011 14:09:20] <Hackman238> locohost: cpna install Net::SNMP
[27-May-2011 14:09:25] <Hackman238> *cpan
[27-May-2011 14:09:33] <Hackman238> locohost: ...if you dare :-)
[27-May-2011 14:10:02] <locohost> python-net-snmp, right
[27-May-2011 14:10:10] <Hackman238> locohost: its perl
[27-May-2011 14:10:15] <locohost> ahhhh
[27-May-2011 14:11:08] <locohost> that makes things easier
[27-May-2011 14:12:44] <locohost> nfsv3 graph was working by the way, nice, ganna delete and readd though now that i got the perl moduel
[27-May-2011 14:14:32] <Hackman238> no worries, you wont need to reinstall it
[27-May-2011 14:21:49] <locohost> still getting ..should i perl <script> that as zenoss and see what it does?    Cmd: /opt/zenoss/ZenPacks/ -H -xxx -l filesystem -f 27 - Code: 2 - Msg: Misuse of shell builtins
[27-May-2011 14:23:01] <locohost> this doesnt require the netapp sdk, does it?
[27-May-2011 14:24:24] <dhopp> Hackman238: out of curiosity…what's the largest zenoss installation you've worked on (Device wise)
[27-May-2011 14:24:47] <Hackman238> dhopp: ~40,000 devices, single instance, 1 minute rate
[27-May-2011 14:25:07] <Hackman238> locohost: I didnt tihnk so, 1 sec
[27-May-2011 14:25:36] <dhopp> Hackman238: you had one zenoss server checking ~40,000 devices every 1 minute?  Including snmp?
[27-May-2011 14:25:55] <Hackman238> dhopp: single zodb, many many many collectors
[27-May-2011 14:26:06] <dhopp> Hackman238: gotcha
[27-May-2011 14:26:54] <Hackman238> dhopp: :-) Big
[27-May-2011 14:27:18] <Hackman238> locohost: doesnt req. it. Do as zenoss zencommand debug
[27-May-2011 14:27:26] <locohost> thats a lot of devices, lemmie guess, cable modems?
[27-May-2011 14:27:48] <dhopp> Hackman238: just a little
[27-May-2011 14:28:03] <Hackman238> locohost: tail -n 1000 -f /opt/zenoss/log/zencommand.log | grep check_netapp
[27-May-2011 14:29:31] <locohost> zencommand debug /path/to/ntap_check ?
[27-May-2011 14:29:51] <Hackman238> locohost: locohost no, zencommadn debug
[27-May-2011 14:29:56] <Hackman238> locohost: itll go to debug mode
[27-May-2011 14:30:19] <locohost> [zenoss@d01mtrending01 root]$ zencommand debug
[27-May-2011 14:30:19] <locohost> Sending SIGUSR1 to 22989
[27-May-2011 14:30:53] <locohost> nothing yet in /opt/zenoss/log/zencommand.log for check_netapp
[27-May-2011 14:30:57] <Hackman238> locohost: no, 20% servers, 70% cisco switches, 5% juniper gss/ive, etc, 5% cisco 650X's
[27-May-2011 14:31:34] <Hackman238> locohost: okay, watch it. When the next test using that commadn runs give me the output
[27-May-2011 14:33:07] <Sam-I-Am> Hackman238: rackspace more or less a cisco shop eh?
[27-May-2011 14:34:56] <locohost> no reference to check_netapp but there is this:"
[27-May-2011 14:34:58] <locohost>
[27-May-2011 14:35:13] <Hackman238> Sam-I-Am: Govt.
[27-May-2011 14:36:20] <Hackman238> locohost: that is an error I've never seen.
[27-May-2011 14:36:27] <Hackman238> locohost: stack or native?
[27-May-2011 14:37:49] <Hackman238> locohost: lets try this. get a name of a device in zenoss that you can run this test against. do as zenoss, zencommand run now -v10 -d device and paste all the output for me
[27-May-2011 14:40:43] <locohost> appears to be suck in aloop or something
[27-May-2011 14:40:50] <locohost> well, getting an error nad retrying
[27-May-2011 14:41:46] <locohost> maybe not, after looking closer they are different
[27-May-2011 14:42:57] <Hackman238> locohost: su zenoss
[27-May-2011 14:43:03] <Hackman238> locohost: run the following
[27-May-2011 14:43:07] <Hackman238> locohost: /opt/zenoss/ZenPacks/ -H -C C10wn -l filesystem -f 83
[27-May-2011 14:43:14] <Hackman238> locohost: Whats the result?
[27-May-2011 14:43:44] <locohost> Can't locate Net/ in @INC (@INC contains: /usr/lib64/perl5/site_perl/5.8.8/x86_64-linux-thread-multi /usr/lib/perl5/site_perl/5.8.8 /usr/lib/perl5/site_perl /usr/lib64/perl5/vendor_perl/5.8.8/x86_64-linux-thread-multi /usr/lib/perl5/vendor_perl/5.8.8 /usr/lib/perl5/vendor_perl /usr/lib64/perl5/5.8.8/x86_64-linux-thread-multi /usr/lib/perl5/5.8.8 .) at /opt/zenoss/ZenPacks/
[27-May-2011 14:43:44] <locohost> ity/NetApp/libexec/check_netapp line 8.
[27-May-2011 14:43:44] <locohost> BEGIN failed--compilation aborted at /opt/zenoss/ZenPacks/ line 8.
[27-May-2011 14:43:49] <locohost> ugh, doesnt know where hte perl moduel is
[27-May-2011 14:43:57] <Hackman238> locohost: your perl is fuxored
[27-May-2011 14:44:05] <Hackman238> locohost: su zenoss
[27-May-2011 14:44:16] <Hackman238> cpan install Net::SNMP
[27-May-2011 14:44:20] <Sam-I-Am> Hackman238: i thought govt liked foundry stuff
[27-May-2011 14:44:31] <Hackman238> locohost: after yum remove the rpm of the module
[27-May-2011 14:44:49] <locohost> see if that does it, i do have /usr/lib64/perl5/vendor_perl/5.8.8/x86_64-linux-thread-multi/
[27-May-2011 14:45:55] <Hackman238> locohost: retest that command after tweaking perl
[27-May-2011 14:46:20] <Hackman238> Sam-I-Am: We acctually didnt have any foundry
[27-May-2011 14:47:08] <Sam-I-Am> i've got a lot of foundry... good stuff
[27-May-2011 14:50:30] <f00fSteR> heyy guys
[27-May-2011 14:50:40] <Hackman238> f00fSteR: how goes the war?
[27-May-2011 14:51:51] <f00fSteR> the war has been lost
[27-May-2011 14:51:54] <f00fSteR> ok so zenmigrate
[27-May-2011 14:51:58] <f00fSteR> = FAIL
[27-May-2011 14:52:09] <Hackman238> f00fSteR: please paste the trace
[27-May-2011 14:53:19] <f00fSteR>
[27-May-2011 14:53:58] <Hackman238> f00fSteR: on your old box, as zenoss, do zenpack --list and paste the result
[27-May-2011 14:56:31] <f00fSteR>
[27-May-2011 14:57:11] <Hackman238> f00fSteR: before you started zenmigrate, did you install the zenoss core packs for zenoss 2.5.2?
[27-May-2011 14:57:18] <f00fSteR> i forgots
[27-May-2011 14:57:21] <f00fSteR> should i do it now ?
[27-May-2011 14:57:27] <f00fSteR> rpm -ivh zenoss-core-zenpacks-2.5.2.el5.x86_64.rpm
[27-May-2011 14:57:30] <f00fSteR> eek
[27-May-2011 14:57:33] <f00fSteR> wrong windows
[27-May-2011 14:57:34] <Hackman238> f00fSteR: wait wait
[27-May-2011 14:57:40] <f00fSteR> whoa
[27-May-2011 14:57:42] <f00fSteR> kk
[27-May-2011 14:57:46] <Hackman238> f00fSteR: service zenoss stop
[27-May-2011 14:57:58] <Hackman238> f00fSteR: ps aux | grep zen
[27-May-2011 14:58:08] <Hackman238> f00fSteR: kill it if it hasnt stopped
[27-May-2011 14:58:18] <Hackman238> f00fSteR: as zenoss, zeoctl start
[27-May-2011 14:58:36] <f00fSteR> done
[27-May-2011 14:58:41] <Hackman238> f00fSteR: as root, rpm -Uvh --nodeps zenoss-core-zenpacks-2.5.2.el5.x86_64.rpm
[27-May-2011 14:58:42] <f00fSteR> daemon process started, pid=6221
[27-May-2011 14:59:04] <f00fSteR> ...
[27-May-2011 14:59:23] <f00fSteR> lots of errors : Reason: KeyError: 'titleOrId
[27-May-2011 15:00:10] <Hackman238> f00fSteR: okay, you're going to need to rm /opt/zenoss/ -Rf, reinstall zenoss. when its done, dont start it.
[27-May-2011 15:00:41] <f00fSteR>
[27-May-2011 15:00:46] <f00fSteR> that was the paste of the erros
[27-May-2011 15:00:49] <f00fSteR> doing rm -rf /
[27-May-2011 15:00:50] <f00fSteR> now
[27-May-2011 15:00:50] <Hackman238> f00fSteR: when install is done, copy the Data.* over src->trgt /opt/zenoss/var/
[27-May-2011 15:01:24] <Hackman238> f00fSteR: double check perms on Data.*
[27-May-2011 15:01:54] <Hackman238> f00fSteR: once done, as zenoss, zeoctl start, zeoctl status
[27-May-2011 15:02:10] <Hackman238> f00fSteR: if daemon is running, rpm in the packs
[27-May-2011 15:02:27] <f00fSteR> Header V3 DSA signature: NOKEY, key ID aa5a1ad7
[27-May-2011 15:02:27] <f00fSteR>  does that matter when install starts ?
[27-May-2011 15:02:37] <Hackman238> f00fSteR: nope
[27-May-2011 15:02:59] <Hackman238> f00fSteR: errors this time?
[27-May-2011 15:03:16] <dhopp> is it safe to copy Data.* while zenoss is runnign?
[27-May-2011 15:03:47] <Hackman238> dhopp: copying Data.fs from a source is safe
[27-May-2011 15:03:53] <Hackman238> dhopp: never overwrite one in use
[27-May-2011 15:03:57] <f00fSteR> there's a Data.fs.old thats over 300 mb
[27-May-2011 15:03:59] <f00fSteR> is that needed ?
[27-May-2011 15:04:04] <Hackman238> f00fSteR: No
[27-May-2011 15:04:30] <f00fSteR> there was no /opt/zenoss/var/ i had to mkdir and then move data over
[27-May-2011 15:04:33] <f00fSteR> it's almost done
[27-May-2011 15:04:36] <f00fSteR> checking permissions
[27-May-2011 15:04:48] <Hackman238> make sure that var dir is zenoss:zenoss
[27-May-2011 15:05:24] <dhopp> Hackman238: I figured overwriting while in use was bad :-P
[27-May-2011 15:05:31] <Hackman238> dhopp:  :-)
[27-May-2011 15:05:39] <dhopp> Hackman238: I just didn't know how it would handle if during the copy the file was updated
[27-May-2011 15:05:47] <Hackman238> dhopp: if you must be anal about safety, zeopack and copy data.fs.old
[27-May-2011 15:06:00] <Hackman238> dhopp: that file is committed only occasionally
[27-May-2011 15:06:00] <dhopp> kind of like trying to backup a db with making sure it's in a safe state
[27-May-2011 15:06:21] <Hackman238> dhopp: transactions sit in limbo for some time
[27-May-2011 15:06:42] <Hackman238> f00fSteR: how's it going?
[27-May-2011 15:06:42] <f00fSteR> how do i change dir ?
[27-May-2011 15:06:43] <f00fSteR> lol
[27-May-2011 15:06:48] <f00fSteR> chmod zenoss.zenoss var
[27-May-2011 15:06:51] <Hackman238> f00fSteR: LOL
[27-May-2011 15:07:03] <Hackman238> chmod zenoss:zenoss var
[27-May-2011 15:07:47] <f00fSteR> invalid
[27-May-2011 15:08:03] <dhopp> Hackman238: I didn't mean to imply you were suggesting something unsafe…I was more curious myself
[27-May-2011 15:08:51] <Hackman238> dhopp: Oh its no problem, I just want to be clear so everyone's in sync
[27-May-2011 15:09:08] <Hackman238> f00fSteR: groupadd zenoss
[27-May-2011 15:09:10] <locohost> sweet, looks like its working,dont know what i had to go through, fw rules were prevending me from getting out to ftp
[27-May-2011 15:09:15] <Hackman238> f00fSteR: useradd zenoss -g zenoss
[27-May-2011 15:09:20] <locohost> had move to another netowrk temporarly
[27-May-2011 15:09:43] <Hackman238> locohost: LOL
[27-May-2011 15:09:48] <Hackman238> f00fSteR: wait wait
[27-May-2011 15:10:13] <Hackman238> f00fSteR: your on centos or redhat 5, just finished installing the zenoss rpm and the zenoss user is invalid?
[27-May-2011 15:11:13] <Hackman238> f00fSteR: I've lost what step you're on.
[27-May-2011 15:19:39] <locohost> sick, filesystem/disk usage stuff all working and nfs/cifs stats
[27-May-2011 15:19:48] <locohost> ganna have to play with the interface stats, those oids do not respond
[27-May-2011 15:20:12] <Hackman238> locohost: glad its all working :-)
[27-May-2011 15:20:45] <f00fSteR> Hackman238, sorry i was talking to management to explain SSL cert errors
[27-May-2011 15:20:55] <f00fSteR> Hackman238, i should be heading home now... like the last person here
[27-May-2011 15:20:58] <f00fSteR> Hackman238,  lets finish this do hickey
[27-May-2011 15:21:12] <Hackman238> f00fSteR: what step are you at?
[27-May-2011 15:21:21] <f00fSteR> chmod step
[27-May-2011 15:21:23] <f00fSteR> i have user zenoss
[27-May-2011 15:21:29] <Hackman238> f00fSteR: su zenoss
[27-May-2011 15:21:52] <Hackman238> f00fSteR: mkdir var2
[27-May-2011 15:22:15] <Hackman238> f00fSteR: mv the contents of var to var2
[27-May-2011 15:22:23] <Hackman238> f00fSteR: su root
[27-May-2011 15:22:32] <Hackman238> f00fSteR: remove var; mv var2 var
[27-May-2011 15:22:53] <Hackman238> f00fSteR: or as root chown zenoss:zenoss var
[27-May-2011 15:23:41] <f00fSteR> ok thats weird
[27-May-2011 15:23:43] <f00fSteR> it worked
[27-May-2011 15:23:47] <f00fSteR> but why didnt it work before
[27-May-2011 15:24:19] <f00fSteR> -rw-r--r-- 1 zenoss zenoss  19347961 May 27 09:01 Data.fs
[27-May-2011 15:24:20] <f00fSteR> -rw-r--r-- 1 zenoss zenoss    311336 May 27 09:01 Data.fs.index
[27-May-2011 15:24:20] <f00fSteR> -rw-r--r-- 1 zenoss zenoss         6 May 27 09:01 Data.fs.lock
[27-May-2011 15:24:20] <f00fSteR> -rwxr-xr-x 1 zenoss zenoss 197629527 May 27 09:04 Data.fs.old
[27-May-2011 15:24:20] <f00fSteR> -rwxr-xr-x 1 zenoss zenoss     47803 May 27 09:04 Data.fs.tmp
[27-May-2011 15:24:33] <Hackman238> f00fSteR: in /opt/zenoss/var, right?
[27-May-2011 15:24:54] <Hackman238> f00fSteR: su zenoss
[27-May-2011 15:24:56] <f00fSteR> pwd
[27-May-2011 15:24:56] <f00fSteR> /opt/zenoss/var
[27-May-2011 15:25:03] <Hackman238> f00fSteR: zeoctl status
[27-May-2011 15:25:17] <f00fSteR> zeoctl status
[27-May-2011 15:25:17] <f00fSteR> bash: zeoctl: command not found
[27-May-2011 15:25:19] <f00fSteR> lmfao
[27-May-2011 15:25:34] <Hackman238> f00fSteR: okay, as root, service zenoss start
[27-May-2011 15:25:39] <Hackman238> f00fSteR: No, thats correct
[27-May-2011 15:25:57] <Hackman238> f00fSteR: I wanted to make sure we didnt have part of the old install or anything
[27-May-2011 15:25:59] <f00fSteR> # service zenoss start
[27-May-2011 15:25:59] <f00fSteR> sh: /opt/zenoss/bin/zenoss: No such file or directory
[27-May-2011 15:26:18] <f00fSteR> it didnt install
[27-May-2011 15:26:19] <f00fSteR> FUCK
[27-May-2011 15:26:22] <f00fSteR> im so sorry man
[27-May-2011 15:26:26] <Hackman238> this is the rpm problem
[27-May-2011 15:26:28] <Hackman238> LOL
[27-May-2011 15:26:32] <f00fSteR> rpm -ivh zenoss-2.5.2.el5.x86_64.rpm warning: zenoss-2.5.2.el5.x86_64.rpm: Header V3 DSA signature: NOKEY, key ID aa5a1ad7
[27-May-2011 15:26:32] <f00fSteR> Preparing...                ########################################### [100%]
[27-May-2011 15:26:32] <f00fSteR>     package zenoss-2.5.2-590.el5.x86_64 is already installed
[27-May-2011 15:26:48] <Hackman238> okay, do this
[27-May-2011 15:27:01] <Hackman238> mv /opt/zenoss/var /opt/
[27-May-2011 15:27:17] <f00fSteR> done
[27-May-2011 15:27:19] <Hackman238> rpm -ivh --nodeps --replacefiles --replacepkgs zenoss.rpm
[27-May-2011 15:27:52] <Hackman238> f00fSteR: its no sweat, I work perpetually LOL
[27-May-2011 15:28:11] <locohost> trying to think it it will work (be able to figure out w hich interface is which if i just make the interface template local and use SNMPv2-SMI::enterprises.789.
[27-May-2011 15:28:41] <Hackman238> locohost: you must use numerical oids
[27-May-2011 15:28:50] <locohost> yeah, i know that, thks though
[27-May-2011 15:28:59] <f00fSteR> done
[27-May-2011 15:29:01] <locohost> pasted it before i wen tto resolve that to a number
[27-May-2011 15:29:23] <Hackman238> locohost: I dont know what youre trying to do. can you elaborate?
[27-May-2011 15:29:41] <Hackman238> f00fSteR: mv /opt/var /opt/zenoss/
[27-May-2011 15:29:52] <Hackman238> f00fSteR: replace all, if any
[27-May-2011 15:30:01] <Sam-I-Am> my head is spinning heh
[27-May-2011 15:30:05] <Hackman238> f00fSteR: su zenoss
[27-May-2011 15:30:12] <Hackman238> f00fSteR: try zeoctl status
[27-May-2011 15:30:22] <Hackman238> Sam-I-Am: Drinking at work? :-)
[27-May-2011 15:31:11] <locohost> bit ethernetCsmacd is not responding)
[27-May-2011 15:31:14] <f00fSteR> Sam-I-Am, my head is spinning trying to follow Hackman238
[27-May-2011 15:31:18] <dhopp> Sam-I-Am: if so, where do you work? :-)
[27-May-2011 15:31:19] <Hackman238> Sam-I-Am: Or is the list of instructions? I'm doing this, chatting politics and debugging avalon code right now LOL
[27-May-2011 15:31:22] <f00fSteR> in a good way of course
[27-May-2011 15:31:32] <locohost> so, i was going to try and create a local copy of the template and enter an oid that does give data back
[27-May-2011 15:31:52] <Hackman238> locohost: Oh. for each device?
[27-May-2011 15:31:58] <locohost> yeah
[27-May-2011 15:31:59] <Hackman238> locohost: not a good solution.
[27-May-2011 15:32:09] <locohost> there are only 8, okay, what would you suggest?
[27-May-2011 15:32:35] <Hackman238> locohost: is the .32 in that past OID the index?
[27-May-2011 15:32:52] <locohost> yes
[27-May-2011 15:33:23] <locohost> kicks back .32.1 to .32.48 or so
[27-May-2011 15:33:25] <Hackman238> locohost: create a snmp datasource with the full oid, omit the index. Zenoss knows to add the index of the tested interface
[27-May-2011 15:33:36] <f00fSteR> $ zeoctl status
[27-May-2011 15:33:36] <f00fSteR> bash: zeoctl: command not found
[27-May-2011 15:33:44] <Hackman238> f00fSteR: okay, su root
[27-May-2011 15:33:48] <locohost> as zenoss
[27-May-2011 15:34:24] <Hackman238> f00fSteR: service zenoss start
[27-May-2011 15:34:34] <Sam-I-Am> Hackman238: wish i was at this point
[27-May-2011 15:34:45] <Hackman238> Sam-I-Am: Not sure what you mean.
[27-May-2011 15:34:52] <f00fSteR> Daemon: zeoctl /opt/zenoss/bin/zenoss: line 105: /opt/zenoss/bin/zeoctl: No such file or directory
[27-May-2011 15:34:52] <f00fSteR> Daemon: zopectl /opt/zenoss/bin/zenoss: line 105: /opt/zenoss/bin/zopectl: No such file or directory
[27-May-2011 15:34:53] <Sam-I-Am> Hackman238: trying to figure out how to hack inter-as bgp-signaled vpls on low-end juniper gear
[27-May-2011 15:35:13] <Sam-I-Am> Hackman238: so i've been watching this, and in configuration hell... so my head is spinning
[27-May-2011 15:35:21] <Sam-I-Am> f00fSteR: your path correct?
[27-May-2011 15:35:25] <Sam-I-Am> and are you the zenoss user?
[27-May-2011 15:35:28] <f00fSteR> yeah
[27-May-2011 15:35:40] <f00fSteR> service zenoss start
[27-May-2011 15:35:40] <f00fSteR> Zenoss upgraded.  Running data migration scripts...
[27-May-2011 15:35:40] <f00fSteR> Upgrade pre step
[27-May-2011 15:35:40] <f00fSteR> /opt/zenoss/bin/ converted
[27-May-2011 15:35:40] <f00fSteR> /opt/zenoss/bin/zenoss_upgrade_pre converted
[27-May-2011 15:35:40] <f00fSteR> /usr/bin/replace: File '/opt/zenoss/etc/zeo.conf' not found (Errcode: 2)
[27-May-2011 15:35:43] <f00fSteR> /usr/bin/replace: File '/opt/zenoss/etc/zeo.conf' not found (Errcode: 2)
[27-May-2011 15:35:44] <f00fSteR> Starting Zope Object Database
[27-May-2011 15:35:46] <f00fSteR> /opt/zenoss/bin/ line 138: /opt/zenoss/bin/zeoctl: No such file or directory
[27-May-2011 15:35:48] <f00fSteR> Unable to start the Zeo object database (ZODB)
[27-May-2011 15:35:50] <f00fSteR> sorry!
[27-May-2011 15:35:57] <f00fSteR> but ^^
[27-May-2011 15:35:59] <f00fSteR> -.-
[27-May-2011 15:36:12] <f00fSteR> ok
[27-May-2011 15:36:12] <Hackman238> f00fSteR: no sweat- relax, not critical errors
[27-May-2011 15:36:16] <f00fSteR> these are xen servers
[27-May-2011 15:36:22] <f00fSteR> i can throw up a new box
[27-May-2011 15:36:24] <f00fSteR> and start over
[27-May-2011 15:36:26] <f00fSteR>  if its easier for you
[27-May-2011 15:36:32] <Hackman238> f00fSteR: nein nein
[27-May-2011 15:36:42] <Hackman238> f00fSteR: one sec, have to look at code
[27-May-2011 15:36:55] <f00fSteR> ich bin nicht einverstanden
[27-May-2011 15:38:26] <Hackman238> f00fSteR: okay we can work around it, but we shouldnt
[27-May-2011 15:38:35] <f00fSteR> ok
[27-May-2011 15:38:37] <f00fSteR> new box
[27-May-2011 15:38:41] <f00fSteR> firing up
[27-May-2011 15:38:57] <Hackman238> f00fSteR: yum this time please :-)
[27-May-2011 15:40:25] <Hackman238> f00fSteR: Also, found a possible prob, so we need to add some steps
[27-May-2011 15:40:46] <Hackman238> f00fSteR: on source box, as root, tar czvf zenoss.bak zenoss
[27-May-2011 15:41:34] <Hackman238> f00fSteR: rather just grab the Data.fs
[27-May-2011 15:41:49] <Hackman238> f00fSteR: rather just grab the Data.*; its automatic for me to tell you to tar it all
[27-May-2011 15:41:52] <Hackman238> LOL
[27-May-2011 15:42:21] <Hackman238> f00fSteR: on the new target wget zenoss 2.4.? (whatever version is on the source) but the x64 version
[27-May-2011 15:42:42] <Hackman238> f00fSteR: cd to its dir, yum localinstall zenoss.rpm --nogpg -y
[27-May-2011 15:43:21] <Hackman238> f00fSteR: as root, service zenoss start
[27-May-2011 15:43:42] <Hackman238> f00fSteR: when done with starting, service zenoss stop
[27-May-2011 15:43:55] <Hackman238> f00fSteR: copy the Data.* /opt/zenoss/var, replace all
[27-May-2011 15:44:12] <Hackman238> f00fSteR: as zenoss, zeoctl start
[27-May-2011 15:44:28] <Hackman238> f00fSteR: as root, rpm -Uvi --nodeps zenpacks-core.rpm
[27-May-2011 15:44:46] <Hackman238> f00fSteR: when done, service zenoss stop
[27-May-2011 15:44:56] <Hackman238> f00fSteR: ps aux | grep zen
[27-May-2011 15:45:08] <Hackman238> f00fSteR: kill it if it didnt stop
[27-May-2011 15:45:16] <Hackman238> f00fSteR: service zenoss start
[27-May-2011 15:45:30] <Hackman238> lets see what that gets up
[27-May-2011 15:45:32] <Hackman238> *us
[27-May-2011 15:45:53] <Hackman238> Sam-I-Am: ah gotcha
[27-May-2011 15:46:04] <Hackman238> Sam-I-Am: Sorry for the reply delay, I'm short on threads :-)
[27-May-2011 15:46:13] * Sam-I-Am opens Hackman238 and adds ram
[27-May-2011 15:46:33] <dhopp> Sam-I-Am: He's short on threads..he needs more cores
[27-May-2011 15:47:10] <dhopp> sed -i 's|ActiveThreads=\(.*\)|ActiveThreads=1000|' Hackman.conf
[27-May-2011 15:47:16] <Sam-I-Am> MOAR COREZ
[27-May-2011 15:47:23] <Hackman238> Sam-I-Am: LOL
[27-May-2011 15:47:34] <Hackman238> dhopp: LOL
[27-May-2011 15:47:56] <Hackman238> f00fSteR: Still here? :-)
[27-May-2011 15:47:59] <f00fSteR> yeah
[27-May-2011 15:48:00] <f00fSteR> sigh
[27-May-2011 15:48:05] <f00fSteR> i installed the 2.5 package
[27-May-2011 15:48:12] <locohost> bah, thats not the right oid anyway, thats bytes used per disk
[27-May-2011 15:48:28] <Hackman238> f00fSteR: locohost sorry. What is it your trying to get?
[27-May-2011 15:48:45] <Hackman238> f00fSteR: yeah, we need 2.4.?? (whatever production version is)
[27-May-2011 15:48:51] <f00fSteR> kk
[27-May-2011 15:48:55] <Hackman238> f00fSteR: Sorry
[27-May-2011 15:48:57] <dhopp> I hope Hackman238 isn't running Windows…we'll see a blue screen soon
[27-May-2011 15:49:23] <locohost> interface status
[27-May-2011 15:49:27] <Hackman238> dhopp: Fedora and Solaris :-)
[27-May-2011 15:49:31] <locohost> it only likes the 32bit conter
[27-May-2011 15:49:39] <Hackman238> locohost: oh- up down?
[27-May-2011 15:49:42] <locohost> which, is going to roll over...well, maybe not in 1 minute
[27-May-2011 15:49:49] <locohost> no, throughput
[27-May-2011 15:49:54] <f00fSteR> Hackman238,
[27-May-2011 15:49:57] <f00fSteR> take a look at that
[27-May-2011 15:49:59] <Hackman238> locohost: oh gotcha.
[27-May-2011 15:50:00] <f00fSteR> i dont think it installed
[27-May-2011 15:50:04] <f00fSteR> but dependencies did
[27-May-2011 15:50:23] <Hackman238> f00fSteR: you forgot --nogpg
[27-May-2011 15:50:33] <f00fSteR> ahh
[27-May-2011 15:50:35] <f00fSteR> gotit
[27-May-2011 15:50:38] <Hackman238> f00fSteR: live dangerous- we dont need signed pkgs
[27-May-2011 15:50:41] <Hackman238> ;-)
[27-May-2011 15:50:42] <f00fSteR> i'll do it with the 2.4 package this time
[27-May-2011 15:51:17] <Hackman238> locohost: what type of device? netapp ports?
[27-May-2011 15:51:51] <Hackman238> Sam-I-Am: thinking it over- I'd consult sir google
[27-May-2011 15:51:58] <locohost> yeah, like e0a e0b etc
[27-May-2011 15:52:09] <Hackman238> locohost: 1 sec
[27-May-2011 15:52:31] <locohost> 32bit counter is probably going to roll over, maybe even at 1m pulls
[27-May-2011 15:53:37] <Hackman238> locohost: definately
[27-May-2011 15:53:42] <Hackman238> locohost: but zenoss should compensate
[27-May-2011 15:53:44] <Sam-I-Am> Hackman238: on my vpls issue? or your threads
[27-May-2011 15:54:02] <locohost> it works for you with the 32bit counter?
[27-May-2011 15:54:21] <locohost> im not sure how it would know, but if its working for you at multiple gigabits/s
[27-May-2011 15:54:31] <f00fSteR> i cant friggin find 2.4.2
[27-May-2011 15:54:36] <f00fSteR> it's not even on source forge
[27-May-2011 15:54:44] <f00fSteR> would it be smart t get it from this russian search engine
[27-May-2011 15:54:46] <f00fSteR>
[27-May-2011 15:54:58] <f00fSteR> Hackman238, what were you saying about living dangerously ?
[27-May-2011 15:55:10] <Hackman238> Sam-I-Am: vpls
[27-May-2011 15:55:17] <Hackman238> f00fSteR: meaning --nogpg
[27-May-2011 15:55:31] <Sam-I-Am> Hackman238: it hurts
[27-May-2011 15:55:47] <locohost> data/netinstall/centos/5.3/localrepo/RPMS/zenoss-2.4.2.el5.i386.rpm
[27-May-2011 15:56:00] <f00fSteR> x64  ?
[27-May-2011 15:56:01] <locohost> foof, i hvae it in my local repo at home, from like 3 years ago
[27-May-2011 15:56:08] <f00fSteR>
[27-May-2011 15:56:09] <locohost> err, no, only 32bit
[27-May-2011 15:56:23] <Hackman238> locohost: not sure about this oid. you're going to need to hunt for it
[27-May-2011 15:56:38] <Hackman238> f00fSteR: 1 sec
[27-May-2011 15:56:45] <dhopp> f00fSteR: do you have it on your existing serve?
[27-May-2011 15:56:48] <dhopp> server
[27-May-2011 15:56:53] <f00fSteR> Hackman238, locohost : it says sourceforge on it... but i mean it's a french server i found on google listed on a russian search engine
[27-May-2011 15:57:27] <dhopp> f00fSteR:
[27-May-2011 15:57:27] <f00fSteR> even if i did it'd be 64 bits
[27-May-2011 15:57:45] <Hackman238> f00fSteR: 2.4.5 is ok
[27-May-2011 15:57:53] <locohost> well, the easiest thing for me to try is the 32bit oid and see if zenoss can indeed compensate as you say it does, youve been right every time so far, heh
[27-May-2011 15:57:59] <f00fSteR> i got 2.4.2 tho
[27-May-2011 15:58:02] <f00fSteR> dhopp, thanks
[27-May-2011 15:58:08] <Hackman238> f00fSteR: please install 64-bit version
[27-May-2011 15:58:29] <Hackman238> locohost: its 'supposed to'
[27-May-2011 15:58:46] <Hackman238> locohost: most of the time it will as long as you set the type right. make sure its set a counter
[27-May-2011 15:58:47] <f00fSteR> Hackman238, lol you've been screaming at me about that since yesterday
[27-May-2011 15:58:58] <f00fSteR> also i went to my manager asked him about a 32 gb box for zenoss... he laughed in my face
[27-May-2011 15:59:07] <f00fSteR> -x-
[27-May-2011 15:59:15] <Hackman238> f00fSteR: Sorry
[27-May-2011 15:59:57] <f00fSteR> haha it's cool
[27-May-2011 16:00:07] <f00fSteR> convinced him to get me a new x64 box with 8 gb's in it
[27-May-2011 16:00:14] <f00fSteR> worked like a charm
[27-May-2011 16:00:20] <Hackman238> locohost: if the data source type isnt counter, it wont auto compensate
[27-May-2011 16:00:24] <f00fSteR> if you want something... always exxagerate
[27-May-2011 16:00:38] <f00fSteR> Installed:
[27-May-2011 16:00:38] <f00fSteR>   zenoss.x86_64 0:2.4.2-329.el5
[27-May-2011 16:00:44] <Hackman238> f00fSteR: Always double and add 1/3
[27-May-2011 16:00:47] <Hackman238> ;-)
[27-May-2011 16:01:06] <Hackman238> okay, copy over the Data.* /opt/zenoss/var replace all
[27-May-2011 16:01:17] <Hackman238> wait
[27-May-2011 16:01:22] <Hackman238> my bad, did you start it?
[27-May-2011 16:01:54] <Hackman238> f00fSteR: I just realized the ambiguity of its installed with installed & started
[27-May-2011 16:02:36] <f00fSteR> i is lost
[27-May-2011 16:02:44] * f00fSteR looks at *you* funny
[27-May-2011 16:02:46] <Hackman238> f00fSteR: zenoss is installed, did you start it
[27-May-2011 16:02:47] <Hackman238> ?
[27-May-2011 16:02:50] <f00fSteR> nope
[27-May-2011 16:02:57] <Hackman238> f00fSteR: service zenoss start before Data.* copy
[27-May-2011 16:03:20] <Hackman238> f00fSteR: once its done with it initial startup, service zenoss stop and copy the Data.* /opt/zenoss/var
[27-May-2011 16:03:30] <f00fSteR> gotcha
[27-May-2011 16:04:39] <f00fSteR> 2011-05-27 10:02:36 (291 MB/s) - `/dev/null' saved [4582/4582]
[27-May-2011 16:04:40] <f00fSteR> Zenoss installation completed.
[27-May-2011 16:04:46] <f00fSteR> copying data
[27-May-2011 16:05:10] <Hackman238> f00fSteR: alrighty. once copied, verify perms ls -lhR /opt/zenoss/var
[27-May-2011 16:06:08] <Hackman238> Sam-I-Am: I wish I could help, but I know little about those items
[27-May-2011 16:06:33] <Hackman238> Sam-I-Am: I know programming, electronics, chemistry, math- but not vpls
[27-May-2011 16:06:56] * dhopp notices Hackman238 left out zenoss
[27-May-2011 16:07:02] <dhopp> :-P
[27-May-2011 16:07:39] <Hackman238> dhopp: No one can truely understand Zenoss because no one has yet plumbed the depths of Zope ;-)
[27-May-2011 16:08:14] <Hackman238> dhopp: Zope is an object of mystery, iteration and madness :-)
[27-May-2011 16:09:10] <Hackman238> BRB, need coffee
[27-May-2011 16:09:19] <dhopp> Hackman238: I know the feeling…I wanted to add permissions so ZenUsers could create new Graph reports…it said that in needed DMD Modify or something like I added that permission on Reports and it didn't work...
[27-May-2011 16:09:33] <dhopp> Manage DMD rather
[27-May-2011 16:10:33] <Sam-I-Am> Hackman238: i know it... at this point its more like trying to hack around some limitations in low-end devices
[27-May-2011 16:12:38] <locohost> thanks, yeah, these are counters
[27-May-2011 16:12:50] <locohost> now, to wait for the next polling cycle
[27-May-2011 16:14:03] <f00fSteR> when do the zencore plugins go in ?
[27-May-2011 16:14:56] <Hackman238> back
[27-May-2011 16:17:28] <Hackman238> f00fSteR: okay, you copied the Data.* in to zenoss?
[27-May-2011 16:17:44] <Hackman238> f00fSteR: and ls -lhR /opt/zenoss/var looks ok?
[27-May-2011 16:18:04] <f00fSteR> ok
[27-May-2011 16:18:11] <f00fSteR> jus tyhe data.old is by root root
[27-May-2011 16:18:15] <Hackman238> dhopp: 1 sec, your prob has a solution, but its complex
[27-May-2011 16:18:27] <Hackman238> Sam-I-Am: Yeah, I hate it when that happens
[27-May-2011 16:18:41] <Hackman238> f00fSteR: get rid of the .old
[27-May-2011 16:18:52] <dhopp> Hackman238: Yeah I know :-P…I did get it work..but it was far less then I'd be interested in the "real" solution
[27-May-2011 16:18:53] <Hackman238> f00fSteR: next, as zenoss, zeoctl start
[27-May-2011 16:19:17] <Hackman238> dhopp: What method did you use?
[27-May-2011 16:19:30] <Hackman238> f00fSteR: next, zeoctl status
[27-May-2011 16:19:36] <f00fSteR> zeoctl start
[27-May-2011 16:19:36] <f00fSteR> daemon process already running; pid=25659
[27-May-2011 16:19:47] <f00fSteR> program running; pid=25659
[27-May-2011 16:19:59] <Hackman238> f00fSteR: as root, rpm -Uvh --nodeps zenoss-core-packs.rpm
[27-May-2011 16:20:06] <dhopp> Hackman238: Honestly I don't remember exactly…it was a trial and error thing
[27-May-2011 16:20:14] <Hackman238> dhopp: All in the ZMI?
[27-May-2011 16:20:22] <dhopp> Hackman238: yes
[27-May-2011 16:20:36] <Hackman238> dhopp: Thats the core solution unless you code a pack for it
[27-May-2011 16:20:47] <Hackman238> dhopp: Ent has a UI for it
[27-May-2011 16:21:22] <Hackman238> dhopp: I'd be interested in making a UI for it after a few projects finish
[27-May-2011 16:21:40] <Hackman238> f00fSteR: is installing?
[27-May-2011 16:22:12] <f00fSteR> Hackman238, nien nien
[27-May-2011 16:22:14] <f00fSteR>
[27-May-2011 16:23:20] <dhopp> Hackman238: Yeah I know the Enterprise has a UI..but I don't have Enterprise...
[27-May-2011 16:23:31] <dhopp> Hackman238: And that is one feature that I understand why it's Enterprise
[27-May-2011 16:24:41] <Hackman238> f00fSteR: thats ok, ignore that
[27-May-2011 16:24:50] <Hackman238> f00fSteR: do service zenoss stop
[27-May-2011 16:25:00] <Hackman238> f00fSteR: then su zenoss
[27-May-2011 16:25:13] <Hackman238> f00fSteR: zenmigrate
[27-May-2011 16:26:13] <Hackman238> dhopp: Zenoss Inc tries to limit Core scalability/managability to push Ent- it will be the end of their business if they continue to be like that.
[27-May-2011 16:27:52] <f00fSteR> [zenoss@localhost root]$ zenmigrate
[27-May-2011 16:27:52] <f00fSteR> WARNING:ZEO.zrpc:(28836) CW: error connecting to ('localhost', 8100): ECONNREFUSED
[27-May-2011 16:29:16] <Hackman238> f00fSteR: my bad, su zenoss, zeoctl start, zopectl start, zenmigrate
[27-May-2011 16:29:30] <Hackman238> f00fSteR: This is all from memory, sorry :-)
[27-May-2011 16:31:37] <f00fSteR> cntil z first
[27-May-2011 16:31:44] <f00fSteR> WARNING:ZEO.zrpc:(28836) CW: error connecting to ('localhost', 8100): ECONNREFUSED
[27-May-2011 16:31:45] <f00fSteR> WARNING:ZEO.zrpc:(28836) CW: error connecting to ('localhost', 8100): ECONNREFUSED
[27-May-2011 16:31:45] <f00fSteR> WARNING:ZEO.zrpc:(28836) CW: error connecting to ('localhost', 8100): ECONNREFUSED
[27-May-2011 16:32:17] <Hackman238> f00fSteR: do ps aux | grep zen
[27-May-2011 16:32:24] <dhopp> Hackman238:  Yeah I we saw with relStorage…I don't get why they want relStorage only in makes the "core" of the product different…but Giving more granular control over ACLs I'm ok with being a Enterprise feature
[27-May-2011 16:32:52] <dhopp> Hackman238: Even if I want to use it...
[27-May-2011 16:33:19] kocolosk is now known as kocolosk|away
[27-May-2011 16:34:30] <Hackman238> f00fSteR: zeo running?
[27-May-2011 16:35:49] <f00fSteR>  INFO:zen.migrate:Could not retrieve datasource at Server/Linux/rrdTemplates/Device/datasources/laLoadInt1. The template or datasource is not present.
[27-May-2011 16:35:49] <f00fSteR> INFO:zen.migrate:Could not retrieve datasource at Server/Linux/rrdTemplates/Device/datasources/laLoadInt15. The template or datasource is not present.
[27-May-2011 16:35:49] <f00fSteR> INFO:zen.migrate:Could not retrieve datasource at Server/Solaris/rrdTemplates/Device/datasources/laLoadInt1. The template or datasource is not present.
[27-May-2011 16:35:49] <f00fSteR> INFO:zen.migrate:Could not retrieve datasource at Server/Solaris/rrdTemplates/Device/datasources/laLoadInt15. The template or datasource is not present.
[27-May-2011 16:36:02] <Hackman238> f00fSteR: oh good, its working
[27-May-2011 16:36:05] <f00fSteR> INFO:zen.migrate:Installing FixBadJobs (2.4.2)
[27-May-2011 16:36:05] <f00fSteR> INFO:zen.migrate:Installing UseIntLoadAverages (2.4.2)
[27-May-2011 16:36:10] <f00fSteR> INFO:zen.migrate:Installing zFileSystemSizeOffset (2.4.2)
[27-May-2011 16:36:10] <f00fSteR> INFO:zen.migrate:Loading Reports
[27-May-2011 16:36:12] <f00fSteR> done
[27-May-2011 16:36:13] <Hackman238> f00fSteR: let me know when its done
[27-May-2011 16:36:16] <Hackman238> f00fSteR: nice
[27-May-2011 16:36:17] <f00fSteR> ^^
[27-May-2011 16:36:38] <Hackman238> f00fSteR: su zenoss, zeoctl restart; zopectl restart
[27-May-2011 16:36:54] <Hackman238> f00fSteR: su zenoss, zenchkrels -x 1 -r
[27-May-2011 16:37:17] <Hackman238> f00fSteR: while its going, try to hit the web UI to see if it comes up or has trace
[27-May-2011 16:37:39] <Hackman238> dhopp: Granular, yes, but Core has none.
[27-May-2011 16:38:09] <f00fSteR> everything else went good
[27-May-2011 16:38:12] <f00fSteR> lots of jumble
[27-May-2011 16:38:18] <f00fSteR> Traceback (most recent call last):
[27-May-2011 16:38:18] <f00fSteR>   File "/opt/zenoss/Products/ZenUtils/", line 72, in ?
[27-May-2011 16:38:18] <f00fSteR>     tmbk.rebuild()
[27-May-2011 16:38:18] <f00fSteR>   File "/opt/zenoss/Products/ZenUtils/", line 36, in rebuild
[27-May-2011 16:38:18] <f00fSteR>     for object in getAllConfmonObjects(self.dmd):
[27-May-2011 16:38:19] <f00fSteR>   File "/opt/zenoss/Products/ZenUtils/", line 440, in getSubObjectsMemo
[27-May-2011 16:38:21] <f00fSteR>     for x in getSubObjectsMemo(obj, filter, descend, memo):
[27-May-2011 16:38:23] <f00fSteR>   File "/opt/zenoss/Products/ZenUtils/", line 440, in getSubObjectsMemo
[27-May-2011 16:38:25] <f00fSteR>     for x in getSubObjectsMemo(obj, filter, descend, memo):
[27-May-2011 16:38:27] <f00fSteR>   File "/opt/zenoss/Products/ZenUtils/", line 440, in getSubObjectsMemo
[27-May-2011 16:38:29] <f00fSteR>     for x in getSubObjectsMemo(obj, filter, descend, memo):
[27-May-2011 16:38:31] <f00fSteR>   File "/opt/zenoss/Products/ZenUtils/", line 440, in getSubObjectsMemo
[27-May-2011 16:38:33] <f00fSteR>     for x in getSubObjectsMemo(obj, filter, descend, memo):
[27-May-2011 16:38:35] <f00fSteR>   File "/opt/zenoss/Products/ZenUtils/", line 432, in getSubObjectsMemo
[27-May-2011 16:38:39] <f00fSteR>     objs = base.objectValues()
[27-May-2011 16:38:41] <f00fSteR>   File "/opt/zenoss/Products/ZenRelations/", line 222, in objectValues
[27-May-2011 16:38:43] <f00fSteR>     return [ob.__of__(self) for ob in self._objects.values()]
[27-May-2011 16:38:45] <f00fSteR> AttributeError: 'DnsMonitorDataSource' object has no attribute '__of__'
[27-May-2011 16:38:47] <f00fSteR> end of bits
[27-May-2011 16:38:55] <Hackman238> f00fSteR: ok; gui work?
[27-May-2011 16:40:43] <dhopp> Hackman238: well there is manager and zenuser :-)…I would like one more that is a little more in the middle..but I understand their decision with that one
[27-May-2011 16:41:29] <dhopp> Hackman238: however, I wish it was documented what each of the permissions mean "Manage DMD" etc.
[27-May-2011 16:43:41] <Hackman238> f00fSteR: if the gui works, do su zenoss, zendmd, reindex()
[27-May-2011 16:43:46] <Hackman238> dhopp: agreed
[27-May-2011 16:44:32] <f00fSteR> ok gui works
[27-May-2011 16:44:56] <f00fSteR> wtf
[27-May-2011 16:45:26] <Hackman238> f00fSteR: ok, once that commadn returns a prompt >>, type sync() {enter} then commit() {enter}.
[27-May-2011 16:45:33] <f00fSteR>
[27-May-2011 16:45:38] <f00fSteR> what was that when i started zendmd
[27-May-2011 16:45:59] <Hackman238> f00fSteR: ignore it
[27-May-2011 16:46:06] <f00fSteR> 2011-05-27 10:43:51 WARNING ZODB Could not import class 'ZenPack' from module 'ZenPacks.zenoss.MySqlMonitor'
[27-May-2011 16:46:07] <f00fSteR> 2011-05-27 10:43:54 WARNING ZODB Could not import class 'ZenPack' from module 'ZenPacks.zenoss.LDAPMonitor'
[27-May-2011 16:46:07] <f00fSteR> 2011-05-27 10:43:54 WARNING ZODB Could not import class 'ZenPack' from module 'ZenPacks.zenoss.RPCMonitor'
[27-May-2011 16:46:09] <f00fSteR> 2011-05-27 10:43:54 WARNING ZODB Could not import class 'ZenPack' from module 'ZenPacks.zenoss.ZenJMX'
[27-May-2011 16:46:11] <f00fSteR> 2011-05-27 10:43:57 WARNING ZODB Could not import class 'ZenPack' from module 'ZenPacks.zenoss.DellMonitor'
[27-May-2011 16:46:13] <f00fSteR> 2011-05-27 10:43:58 WARNING ZODB Could not import class 'ZenPack' from module 'ZenPacks.zenoss.HPMonitor'
[27-May-2011 16:46:18] <Hackman238> f00fSteR: ignore all zenpack related trace
[27-May-2011 16:46:38] <f00fSteR> k
[27-May-2011 16:46:39] <f00fSteR> done
[27-May-2011 16:46:41] <f00fSteR> done and done
[27-May-2011 16:46:47] <Hackman238> f00fSteR: we will fix, but zenoss is very stupid- has no concept of migration 32 -> 64 bit, so this is patch & fix migration
[27-May-2011 16:46:55] <Hackman238> f00fSteR: crtl + d to exit
[27-May-2011 16:47:19] <Hackman238> f00fSteR: su root, rpm -Uvh --nodeps zenoss-core-packs.rpm
[27-May-2011 16:47:33] <Hackman238> f00fSteR: We're going to install them a second time to try to fix the problem
[27-May-2011 16:47:46] <Hackman238> f00fSteR: let me know what it says
[27-May-2011 16:49:35] <f00fSteR> warning: zenoss-core-zenpacks-2.4.2.el5.x86_64.rpm: Header V3 DSA signature: NOKEY, key ID aa5a1ad7
[27-May-2011 16:49:35] <f00fSteR> Preparing...                ########################################### [100%]
[27-May-2011 16:49:35] <f00fSteR>     package zenoss-core-zenpacks-2.4.2-329.x86_64 is already installed
[27-May-2011 16:49:59] <Hackman238> f00fSteR: su root, rpm -Uvh --nodeps zenoss-core-packs.rpm --replacepkgs --replacefiles --force
[27-May-2011 16:50:05] <Hackman238> f00fSteR: Sorry :-)
[27-May-2011 16:50:57] <f00fSteR> ERROR: zenpack command failed. Reason: (1045, "Access denied for user 'zenoss'@'localhost' (using password: YES)")
[27-May-2011 16:51:09] <f00fSteR> just a whole bunch of those for each plugin
[27-May-2011 16:51:37] <Hackman238> f00fSteR: did you set mysql passwords?
[27-May-2011 16:51:43] <f00fSteR> nope
[27-May-2011 16:51:56] <Hackman238> f00fSteR: okay, 1 sec
[27-May-2011 16:52:07] <f00fSteR> kk
[27-May-2011 16:54:14] <f00fSteR> by the way
[27-May-2011 16:54:16] <f00fSteR> when i log into gui
[27-May-2011 16:54:17] <f00fSteR> There is a serious problem with the MySQL event database
[27-May-2011 16:54:17] <f00fSteR> Below is the detailed information provided:
[27-May-2011 16:54:17] <f00fSteR> (1045, "Access denied for user 'zenoss'@'localhost' (using password: YES)")
[27-May-2011 16:54:25] <f00fSteR> im looking into events manager
[27-May-2011 16:54:34] <f00fSteR> the password is prolly different than the one set on mysql
[27-May-2011 16:54:35] <f00fSteR> ok
[27-May-2011 16:54:39] <f00fSteR> this is looking grim for today
[27-May-2011 16:54:42] <f00fSteR> it's 5 pm
[27-May-2011 16:54:47] <f00fSteR> i'm gonna head out
[27-May-2011 16:54:56] <f00fSteR> can we continue where we left off on say tuesday  ?
[27-May-2011 16:54:57] <f00fSteR>
[27-May-2011 16:55:17] <Hackman238> f00fSteR: sigh- is mysql installed and running?
[27-May-2011 16:56:29] <f00fSteR> yes it's installed and running
[27-May-2011 16:56:33] <Hackman238> f00fSteR: if yes, just run /opt/zenoss/bin/zenoss_init_pre
[27-May-2011 16:56:39] <Hackman238> f00fSteR: should be quick
[27-May-2011 16:56:43] <f00fSteR> root
[27-May-2011 16:56:46] <f00fSteR> of course
[27-May-2011 16:56:51] <f00fSteR> dude if you wat me to stay for a bit np
[27-May-2011 16:57:07] <Hackman238> f00fSteR: yes run as root
[27-May-2011 16:57:10] <f00fSteR> but i have a hot lady waiting for me in this 88 degree ny weather
[27-May-2011 16:57:32] <Hackman238> f00fSteR: LOL- lets try this final thing for today
[27-May-2011 16:57:36] <f00fSteR> zport portal object exits; exiting.
[27-May-2011 16:57:36] <f00fSteR> Stopping Zope Object Database
[27-May-2011 16:57:36] <f00fSteR> . daemon process stopped
[27-May-2011 16:58:19] <Hackman238> f00fSteR: then install the packs again
[27-May-2011 16:58:29] <Hackman238> f00fSteR: should be good. We can finish tuesday
[27-May-2011 16:59:12] <f00fSteR>    1:zenoss-core-zenpacks   ########################################### [100%]
[27-May-2011 16:59:13] <f00fSteR> Installing /opt/zenoss/packs/ZenPacks.zenoss.ApacheMonitor-2.0.2-py2.4.egg...
[27-May-2011 16:59:13] <f00fSteR> WARNING:ZEO.zrpc:(32267) CW: error connecting to ('localhost', 8100): ECONNREFUSED
[27-May-2011 16:59:13] <f00fSteR> WARNING:ZEO.zrpc:(32267) CW: error connecting to ('localhost', 8100): ECONNREFUSED
[27-May-2011 16:59:16] <f00fSteR> what is going on
[27-May-2011 16:59:20] <f00fSteR> maybe i shoul;d set a mysql password ?
[27-May-2011 16:59:27] <f00fSteR> i remember reading the book
[27-May-2011 16:59:31] <Hackman238> f00fSteR: its starting yes, let it run.
[27-May-2011 16:59:34] <f00fSteR> it says zenoss requires mysql password to be set
[27-May-2011 16:59:51] <f00fSteR> WARNING:ZEO.zrpc:(32267) CW: error connecting to ('localhost', 8100): ECONNREFUSED
[27-May-2011 16:59:51] <f00fSteR> WARNING:ZEO.zrpc:(32267) CW: error connecting to ('localhost', 8100): ECONNREFUSED
[27-May-2011 16:59:51] <f00fSteR> WARNING:ZEO.zrpc:(32267) CW: error connecting to ('localhost', 8100): ECONNREFUSED
[27-May-2011 16:59:51] <f00fSteR> WARNING:ZEO.zrpc:(32267) CW: error connecting to ('localhost', 8100): ECONNREFUSED
[27-May-2011 16:59:52] <f00fSteR> ^^
[27-May-2011 16:59:53] <f00fSteR>
[27-May-2011 17:00:10] <Hackman238> hum. Monday I'll ssh in.
[27-May-2011 17:00:16] <Hackman238> *tuesday
[27-May-2011 17:00:21] <Hackman238> TTY Then
[27-May-2011 17:00:30] <f00fSteR> dude
[27-May-2011 17:00:52] <f00fSteR> it's not doing anything
[27-May-2011 17:00:53] <Hackman238> ?
[27-May-2011 17:00:56] <f00fSteR> does that mean it's doing something
[27-May-2011 17:00:57] <f00fSteR>
[27-May-2011 17:01:06] <Hackman238> f00fSteR: it might, just leave it
[27-May-2011 17:01:11] <f00fSteR> i will actually
[27-May-2011 17:01:16] <Hackman238> f00fSteR: 2.4 is unpredictable
[27-May-2011 17:01:21] <f00fSteR> i see
[27-May-2011 17:01:21] <Hackman238> tty tuesday
[27-May-2011 17:01:25] <f00fSteR> 2.5 is that much better ?
[27-May-2011 17:01:29] <Hackman238> Later All!
[27-May-2011 17:01:35] <Hackman238> 10000% better
[27-May-2011 17:04:04] <f00fSteR> haha
[27-May-2011 17:04:05] <f00fSteR> l8r guys
[27-May-2011 17:04:11] <f00fSteR> TYTYty Hackman238
[27-May-2011 17:16:53] <willwh> howdy folks
[27-May-2011 17:17:05] <willwh> anyone using the HTTPMonitor zenpack?
[27-May-2011 17:17:25] <willwh> I am having a little issue - I've set a template for it, etc, added data points (and to a graph)
[27-May-2011 17:17:31] <willwh> although - it seems my regex is failing
[27-May-2011 17:19:15] <willwh> I followed this -
[27-May-2011 17:19:24] <willwh> although - I want to check for specific regex
[27-May-2011 17:19:30] <willwh> and I guess that guide was written for 2.5*
[27-May-2011 17:20:39] <willwh> I just have a string in my "Regular expression" field
[27-May-2011 17:20:43] <willwh> I wonder if that is a problem
[27-May-2011 17:20:47] <willwh> back shortly.
[27-May-2011 17:28:47] <willwh> rmatte: are you around?
[27-May-2011 17:32:45] <dhopp> willwh: what problem are you having?
[27-May-2011 17:41:12] <willwh> dhopp: I'm getting "HTTP Critical" pattern not found
[27-May-2011 17:41:26] <dhopp> and what's your pattern look like?
[27-May-2011 17:42:30] <willwh> just a string
[27-May-2011 17:43:46] <dhopp> what is the string?
[27-May-2011 17:43:54] <willwh> it's a guid
[27-May-2011 17:44:01] <willwh> in the regular expression field
[27-May-2011 17:44:14] <willwh> I wonder where I can see what ./check_http full cmd looks like
[27-May-2011 17:44:17] <willwh> from what I've built
[27-May-2011 17:44:22] <dhopp> Can you copy and paste it?
[27-May-2011 17:46:25] <willwh> the guid?
[27-May-2011 17:46:26] <willwh> sure
[27-May-2011 17:46:42] <dhopp> you can probably run zencommand in debug mode.  As zenoss use zencommand stop, zencommand run debug
[27-May-2011 17:46:47] <willwh> bfb8a6b0-425f-44ca-8338-07ccdaaa6dc9
[27-May-2011 17:47:23] <dhopp> err…debug doesn't seem to work
[27-May-2011 17:49:49] <willwh> hrm
[27-May-2011 17:49:58] <willwh> the thing is, I have a custom command that does what I need
[27-May-2011 17:50:05] <willwh> but I can't graph that
[27-May-2011 17:50:45] <dhopp> one sec…I'm setting up a httpmonitor
[27-May-2011 17:52:09] <willwh> alright
[27-May-2011 17:53:19] <willwh> did you install the zenpack?
[27-May-2011 17:53:23] <willwh> I assume so
[27-May-2011 17:54:15] <dhopp> http://chi-zen-dev-01:8080/zport/dmd/ZenPackManager/packs/ZenPacks.zenoss.HttpMonitor
[27-May-2011 17:54:15] <dhopp> zenoss
[27-May-2011 17:54:15] <dhopp> Zenoss
[27-May-2011 17:54:15] <dhopp> 2.0.3
[27-May-2011 18:01:27] <dhopp> Oh might nee to escape the '-'
[27-May-2011 18:01:54] <dhopp> bfb8a6b0\-425f\-44ca\-8338\-07ccdaaa6dc9
[27-May-2011 18:04:46] <willwh> really?
[27-May-2011 18:04:55] <willwh> let me see
[27-May-2011 18:05:31] <dhopp> the hyphen in a regular expression means a range…so for it could be interpreting 0-4 as 1 character that is a number from 0 to 4
[27-May-2011 18:06:01] <dhopp> normally you would do that inside brackets [0-4] but maybe zenoss is interpreting it differently
[27-May-2011 18:07:18] <willwh> I wonder where I can see the build out for what the plugin is passing out
[27-May-2011 18:07:32] <willwh> i..e ./check_http bla bla -flags etc
[27-May-2011 18:08:15] <dhopp> if you changed the regex you might need to push changes to collectors…in the icon that looks like a gear in the bottom left choose "Push Changes"
[27-May-2011 18:08:34] <dhopp> otherwise zencommand might not reread your change
[27-May-2011 18:08:50] <dhopp> at least that's what I had to do when I just changed my regex
[27-May-2011 18:09:39] <willwh> omh
[27-May-2011 18:09:41] <willwh> ok
[27-May-2011 18:09:41] <willwh> so
[27-May-2011 18:09:44] <willwh> very very very odd
[27-May-2011 18:09:46] <willwh> one sec
[27-May-2011 18:09:50] <willwh> ahhhhhh
[27-May-2011 18:10:07] <dhopp> willwh:  working?
[27-May-2011 18:14:13] <dhopp> I have to take off…I'll probably be back on later
[27-May-2011 18:26:36] <willwh> ah - he left
[27-May-2011 18:26:54] <willwh> it is indeed working though - had to escape to the -'s in the guid, push changes, and all working nicely now
[27-May-2011 20:02:30] kocolosk|away is now known as kocolosk
[27-May-2011 20:49:29] kocolosk is now known as kocolosk|away
[27-May-2011 21:12:45] kocolosk|away is now known as kocolosk
[27-May-2011 21:18:25] kocolosk is now known as kocolosk|away
[28-May-2011 00:00:01] [disconnected at Sat May 28 00:00:01 2011]
[28-May-2011 00:00:02] [connected at Sat May 28 00:00:02 2011]
[28-May-2011 00:00:17] [zenoss-logger (logger bot) has joined #zenoss]
[28-May-2011 12:00:01] [disconnected at Sat May 28 12:00:01 2011]
[28-May-2011 12:00:02] [connected at Sat May 28 12:00:02 2011]
[28-May-2011 12:00:19] [zenoss-logger (logger bot) has joined #zenoss]
[28-May-2011 12:01:22] <froztbyte> right
[28-May-2011 12:01:36] <froztbyte> let's see how badly this breaks
[28-May-2011 12:02:02] <froztbyte> (migrating a zenoss instance by just installing the .deb, then rsync-ing the old one over)
[28-May-2011 12:31:22] kocolosk|away is now known as kocolosk
[28-May-2011 12:32:03] kocolosk is now known as kocolosk|away
[28-May-2011 13:08:51] <locohost> my guess is really badly
[28-May-2011 14:44:26] <locohost> how can i force the creation of a non existant rrd? ive even tried remodeling, all graphs for this host are showing except for one and i see the graphing definitions graphing command is looking for cpu5min_cpu5min.rrd which does not exist on my /opt/zenoss/perf/Devices/hostnmae
[28-May-2011 14:56:11] <Hackman238b> locohost: whats happening?
[28-May-2011 15:02:21] <locohost> trying to delete it and remodle the device, this is in a different instance, the cpu graph for my pix 5m cpu appears to have been modled corrrectly, except for the rrdtool create command was never run
[28-May-2011 20:19:21] <locohost_> anywah, was the the wirdest thing, I ended up copying the memory rrd overtop of what it was calling in the graphing statement, and it started to work, the cpu graph had memory data in it that was off the i deleted the device with the plan of readding it and quickly copying in the cpu rrd before the polling cycle put numbers over 100 in it
[28-May-2011 20:19:29] <locohost_> but, afterdoing that, zenoss did create it
[28-May-2011 20:19:47] <locohost_> ..on its own finally, weird
[29-May-2011 00:00:01] [disconnected at Sun May 29 00:00:01 2011]
[29-May-2011 00:00:02] [connected at Sun May 29 00:00:02 2011]
[29-May-2011 00:00:20] [zenoss-logger (logger bot) has joined #zenoss]
[29-May-2011 12:00:01] [disconnected at Sun May 29 12:00:01 2011]
[29-May-2011 12:00:02] [connected at Sun May 29 12:00:02 2011]
[29-May-2011 12:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[29-May-2011 14:07:40] kocolosk|away is now known as kocolosk
[30-May-2011 00:00:01] [disconnected at Mon May 30 00:00:01 2011]
[30-May-2011 00:00:02] [connected at Mon May 30 00:00:02 2011]
[30-May-2011 00:00:20] [zenoss-logger (logger bot) has joined #zenoss]
[30-May-2011 09:45:54] <subbu> HI All,, Can any one please help me how to monitor the port 9988 in zenoss
[30-May-2011 09:48:32] <rockhopper_> hey guys
[30-May-2011 09:48:52] <rockhopper_> I'm trying to run snmp in my server, but it doesn't work!
[30-May-2011 09:49:14] <rockhopper_> sudo service snmpd start runs well.
[30-May-2011 09:49:45] <rockhopper_> but i still get errors when trying snmpwalk it
[30-May-2011 09:50:07] <hmp> did you check your /etc/snmp/snmpd.conf?
[30-May-2011 09:50:37] <rockhopper_> /etc/snmp/snmpd.conf = rocommunity public
[30-May-2011 09:51:41] <rockhopper_> when doing service snmpd status i get:
[30-May-2011 09:51:54] <rockhopper_>  * snmpd is running  * snmptrapd is not running
[30-May-2011 09:52:31] <hmp> and snmpwalk -c public localhost system is not working?
[30-May-2011 09:53:03] <rockhopper_> system: Unknown Object Identifier (Sub-id not found: (top) -> system)
[30-May-2011 09:53:42] <hmp> and snmpwalk -c public localhost .
[30-May-2011 09:55:02] <rockhopper_> snmpwalk: No securityName specified
[30-May-2011 09:55:31] <rockhopper_> if i use the ip instead of localhost i get the same error
[30-May-2011 09:55:53] <hmp> try adding -v1
[30-May-2011 09:56:07] <hmp> or -v2c
[30-May-2011 09:56:44] <rockhopper_> system: Unknown Object Identifier (Sub-id not found: (top) -> system)
[30-May-2011 09:56:47] <rockhopper_> for both
[30-May-2011 09:57:00] <hmp> did you use system or the oid number?
[30-May-2011 09:57:28] <rockhopper_> system
[30-May-2011 09:57:45] <hmp> try with the oid number
[30-May-2011 09:58:12] <rockhopper_> got answer
[30-May-2011 09:58:16] <rockhopper_>
[30-May-2011 09:58:17] <hmp> now do
[30-May-2011 09:58:40] <hmp> snmpwalk -Dparse-mibs -v2c -c public localhost system
[30-May-2011 09:58:50] <hmp> and look where it's looking for mibs
[30-May-2011 09:59:04] <hmp> should say something like: parse-mibs: Scanning directory /usr/local/zenoss/common/share/snmp/mibs/
[30-May-2011 09:59:35] <hmp> to stderr, not stdout
[30-May-2011 10:02:11] <rockhopper_> parse-mibs: Checking file: /usr/share/mibs/netsnmp/NET-SNMP-MONITOR-MIB...
[30-May-2011 10:02:38] <rockhopper_> hmp: it looks in many places
[30-May-2011 10:02:50] <rockhopper_> but in the end it gives system: Unknown Object Identifier (Sub-id not found: (top) -> system)
[30-May-2011 10:02:52] <rockhopper_> (last line)
[30-May-2011 10:03:17] <rockhopper_> oh, my fault, sec
[30-May-2011 10:03:56] <rmatte> rockhopper_: you running ubuntu?
[30-May-2011 10:04:12] <rmatte> (I ran in to this exact same issue with ubuntu the other day and I have a solution for you if that's the case)
[30-May-2011 10:05:03] <rmatte> do "apt-get install snmp-mibs-downloader" then edit /etc/snmp/snmp.conf and comment out the "mibs :" line
[30-May-2011 10:05:07] <rockhopper_> indeed
[30-May-2011 10:05:16] <rmatte> They don't bundle the Mibs with Ubuntu anymore for copyright reasons
[30-May-2011 10:05:23] <rmatte> or licensing reasons
[30-May-2011 10:05:33] <rockhopper_> ... =\ stupid copyright shit
[30-May-2011 10:05:37] <rmatte> hehe
[30-May-2011 10:05:52] <rmatte> but oh well, simple enough fix for it
[30-May-2011 10:06:00] <rockhopper_> that's the part when i understand why FOSS is really so awesome
[30-May-2011 10:06:17] <rockhopper_> rmatte: it's for 11.04 right?
[30-May-2011 10:06:21] <rmatte> correct
[30-May-2011 10:06:37] <rockhopper_> yea, cause on 10.10 it worked like charm
[30-May-2011 10:06:41] <rmatte> yup
[30-May-2011 10:06:46] <rmatte> they just did it last release
[30-May-2011 10:06:53] <rockhopper_> yea
[30-May-2011 10:07:07] <rockhopper_> you see, that's what goo in using something popular
[30-May-2011 10:07:24] <rockhopper_> if i have any trouble, in most cases someone found an answer for that
[30-May-2011 10:07:26] <rockhopper_>
[30-May-2011 10:07:31] <hmp> rockhopper_: you could run it with the zenoss user, it has all the mibs by default
[30-May-2011 10:07:33] <rmatte> well, it also helps being in an IRC channel with a bunch of Linux admins
[30-May-2011 10:08:01] <rmatte> hmp: snmpwalk is only configured to look in certain places, it won't look where Zenoss stores the Mibs
[30-May-2011 10:08:13] <rmatte> Zenoss only has the Mibs for use with zenmib
[30-May-2011 10:08:34] <hmp> rmatte: sure it will, check out the MIBDIRS env
[30-May-2011 10:09:02] <subbu> @rmatte:Good Morning
[30-May-2011 10:09:03] <rmatte> hmp: well, that was overridden anyways by the config option in /etc/snmp/snmp.conf
[30-May-2011 10:09:17] <rmatte> hmp: better to just install the system Mibs anyways
[30-May-2011 10:09:23] <subbu> Can you please help me how to monitor the Port no 9988 in zenoss
[30-May-2011 10:09:26] <rmatte> you don't want to be relying on Zenoss' Mibs system-wide
[30-May-2011 10:09:32] <hmp> rmatte: am I missing something or the zenoss user uses its own snmpwalk?
[30-May-2011 10:09:43] <rmatte> hmp: no, it doesn't
[30-May-2011 10:09:50] <hmp> from /usr/local/zenoss/common/bin/snmpwalk?
[30-May-2011 10:10:05] <subbu> i even added this is in IPservices once i modlede the device its going away
[30-May-2011 10:10:11] <subbu> can u please help em in this
[30-May-2011 10:10:22] <rmatte> hmp: Zenoss executes snmpwalk from PATH
[30-May-2011 10:10:25] <hmp> rmatte: i dont even have system-wide snmpwalk installed
[30-May-2011 10:10:32] <hmp> rmatte: and that is in its PATH
[30-May-2011 10:10:49] <rmatte> I always install system snmpwalk, which probably takes priority in the path
[30-May-2011 10:10:56] <rockhopper_> rmatte: i just did what you said and restarted snmpd, but still it doesn't work
[30-May-2011 10:10:57] <rockhopper_> :X
[30-May-2011 10:11:23] <rmatte> rockhopper_: I'm saying do that on the system that you're doing the snmpwalk from
[30-May-2011 10:11:28] <rmatte> not on the system that you're querying
[30-May-2011 10:11:37] <rockhopper_> oh, oops
[30-May-2011 10:11:38] <rockhopper_> sec
[30-May-2011 10:11:38] <rmatte> you're missing the Mibs locally on the system that you're doing the snmpwalk from
[30-May-2011 10:11:46] <rmatte> the server itself is fine
[30-May-2011 10:11:48] <rockhopper_> that's weird, it worked so far
[30-May-2011 10:12:02] <hmp> rmatte: that could be the case, just saying that zenoss has its own snmpwalk and its own mibs
[30-May-2011 10:12:15] <rmatte> rockhopper_: it's because you're doing <snmpwalk command> system
[30-May-2011 10:12:23] <rmatte> and it doesn't understand what OID "system" references
[30-May-2011 10:12:27] <rmatte> since it doesn't have the Mibs
[30-May-2011 10:14:03] <rmatte> hmp: noted...
[30-May-2011 10:14:03] <rmatte> zenoss@lab01:~$ which snmpwalk
[30-May-2011 10:14:14] <rmatte> /usr/local/zenoss/common/bin/snmpwalk
[30-May-2011 10:17:52] <Sam-I-Am> sup rmatte
[30-May-2011 10:18:27] <rmatte> not much
[30-May-2011 10:18:29] <rockhopper_> rmatte: hm.. i've changed the files
[30-May-2011 10:18:31] <rockhopper_> the file*
[30-May-2011 10:18:36] <rockhopper_> and installed thre packages
[30-May-2011 10:18:39] <rmatte> which file did you change?
[30-May-2011 10:18:40] <rockhopper_> still.. not working
[30-May-2011 10:18:54] <rockhopper_>  /etc/snmp/snmp.conf
[30-May-2011 10:19:00] <rockhopper_> commented out the last line
[30-May-2011 10:19:16] <rmatte> right ok... when you installed the package, did you see it installing the Mibs?
[30-May-2011 10:19:34] <hmp> rockhopper_: you shoudn't comment it out
[30-May-2011 10:19:37] <hmp> you should comment it
[30-May-2011 10:19:40] <rockhopper_> rmatte: yea, it filled the screen with patching file ......-MIB
[30-May-2011 10:19:55] <hmp> (nvm lol :))
[30-May-2011 10:19:56] <rockhopper_> hmp: yea, that's what i meant
[30-May-2011 10:20:07] <hmp> should probably stop drinking
[30-May-2011 10:20:14] <hmp> (not gonna happen)
[30-May-2011 10:20:17] <rockhopper_> hmp: lol
[30-May-2011 10:20:43] <rockhopper_> we have a rule here in the office, if someone makes someone else to work too much he owe him a beer
[30-May-2011 10:20:44] <rockhopper_>
[30-May-2011 10:20:59] <hmp> we shold do that here too
[30-May-2011 10:21:05] <hmp> i would be drunk all the time
[30-May-2011 10:21:05] <rockhopper_>
[30-May-2011 10:21:39] <rmatte> rockhopper_: at the prompt that you're trying to do the snmpwalk from type: which snmpwalk
[30-May-2011 10:21:43] <rmatte> and tell me what the output is
[30-May-2011 10:21:53] <rockhopper_> nada
[30-May-2011 10:22:57] <Hackman238b> rmatte: I'll get that pkg to you tonigt- been having trouble getting remote admin GUI to work in the 2.5.X pack
[30-May-2011 10:23:29] <rmatte> Hackman238b: cool, appreciate it
[30-May-2011 10:24:02] <rockhopper_> rmatte: ^
[30-May-2011 10:24:35] <rmatte> rockhopper_: there has to be some output
[30-May-2011 10:24:53] <rmatte> the which command tells you where it's finding the snmpwalk command to execute it when you type it
[30-May-2011 10:25:20] <rmatte> rockhopper_: provide me with some background here...
[30-May-2011 10:25:29] <rockhopper_> weird. i was sure it is installed. i'll make sure it is. sec
[30-May-2011 10:25:31] <rmatte> when you're seeing that error, are you running the snmpwalk via the Zenoss GUI?
[30-May-2011 10:25:38] <rmatte> or are you actually running it from the commandline?
[30-May-2011 10:25:39] <rockhopper_> yes
[30-May-2011 10:25:43] <rockhopper_> 1
[30-May-2011 10:25:52] <rmatte> ok, so login to that server and become the Zenoss user
[30-May-2011 10:27:08] <rmatte> "sudo su - zenoss"
[30-May-2011 10:28:17] <rmatte> actually, can you try installing system snmpwalk too?
[30-May-2011 10:28:25] <rmatte> apt-get install snmpwalk
[30-May-2011 10:29:22] <rockhopper_> hm.. is it possible that this package is missing?
[30-May-2011 10:29:51] <rockhopper_> rmatte: E: Unable to locate package snmpwalk
[30-May-2011 10:30:03] <rockhopper_> weird. i'm quite sure i installed it before :X
[30-May-2011 10:31:37] <rmatte> I installed it on both of my 11.04 systems
[30-May-2011 10:31:50] <rmatte> make sure you're trying to install it as root
[30-May-2011 10:37:46] <Sam-I-Am> net-snmp should come with it
[30-May-2011 10:38:54] <rockhopper_> $ sudo apt-get install snmpwalk ...... E: Unable to locate package snmpwalk
[30-May-2011 10:38:58] <rockhopper_> odd =\
[30-May-2011 10:39:03] <Sam-I-Am> apt-cache search snmpwalk
[30-May-2011 10:39:10] <Sam-I-Am> i bet apt-get install net-snmp would fix itr
[30-May-2011 10:39:17] <rockhopper_> done and couldn't find anything
[30-May-2011 10:39:47] <rockhopper_> i'm trying update first
[30-May-2011 10:39:50] <fragfutter> apt-get install snmp
[30-May-2011 10:39:54] <Sam-I-Am> anything under snmp?  apt-cache search snmp | grep -i snmp
[30-May-2011 10:40:03] <Sam-I-Am> sometimes they rename packages
[30-May-2011 10:40:08] <fragfutter> its snmp.
[30-May-2011 10:40:20] <Sam-I-Am> awesome :/
[30-May-2011 10:40:46] <rockhopper_> snmp does work
[30-May-2011 10:40:54] <fragfutter> and you could use a package index. here is your hint
[30-May-2011 10:40:58] <rockhopper_> and i thought i installed that one too
[30-May-2011 10:41:30] <rockhopper_> i must close the access to the other guys here, i don't know what's going on in my machine anymore
[30-May-2011 10:41:31] <rockhopper_>
[30-May-2011 10:41:53] <rockhopper_> $ which snmpwalk  /usr/bin/snmpwalk
[30-May-2011 10:49:26] <rmatte> did you do that as the zenoss user though?
[30-May-2011 10:54:55] <rockhopper_> rmatte: no
[30-May-2011 10:59:30] <rockhopper_> rmatte: i mean, i did it threw the zenoss gui
[30-May-2011 10:59:38] <rockhopper_> so i guess it is from the zenoss user
[30-May-2011 11:09:19] <rmatte> don't do the which through the interface
[30-May-2011 11:09:22] <rmatte> do it from the commandline
[30-May-2011 11:09:44] <rmatte> you really need to configure the server with sudo so that you can do "sudo su - zenoss" to become the zenoss user
[30-May-2011 11:09:55] <rmatte> you're going to need it eventually
[30-May-2011 11:10:09] <rmatte> It's invaluable for debugging/testing
[30-May-2011 11:12:15] <rockhopper_> rmatte: i did the which from the cli
[30-May-2011 11:12:40] <rmatte> right, but did you do it as the zenoss user?
[30-May-2011 11:13:04] <rmatte> the zenoss user has completely different environment variables than a standard user on the system
[30-May-2011 11:13:08] <rmatte> so it's very different
[30-May-2011 11:14:06] <rockhopper_> $ which snmpwalk   /usr/local/zenoss/common/bin/snmpwalk
[30-May-2011 11:14:39] <rmatte> ok... so the zenoss snmpwalk must not be looking in the same place as the system snmpwalk
[30-May-2011 11:14:56] <rmatte> so you did manage to get snmpwalk installed on the system right?
[30-May-2011 11:15:08] <rmatte> seeing as you have it at /usr/bin/snmpwalk
[30-May-2011 11:15:20] <rockhopper_> rmatte: indeed
[30-May-2011 11:15:33] <rmatte> try the same snmpwalk with the system snmpwalk from the commandline
[30-May-2011 11:15:35] <rmatte> and see if it works
[30-May-2011 11:15:42] <rmatte> (literally copy and paste the command from Zenoss)
[30-May-2011 11:16:00] <rmatte> do it as a regular user on the system not as zenoss
[30-May-2011 11:18:20] <rockhopper_> $ snmpwalk -v1 -cpublic system
[30-May-2011 11:18:20] <rockhopper_> Timeout: No Response from
[30-May-2011 11:18:29] <rockhopper_> same as in zenoss GUI
[30-May-2011 11:18:40] <rmatte> that's a network problem obviously
[30-May-2011 11:18:50] <rmatte> try a different device that you can actually hit?
[30-May-2011 11:19:22] <rockhopper_> the only device i can hit is in the same network
[30-May-2011 11:19:32] <rockhopper_> okay, g2g seriously. thanks!
[30-May-2011 11:19:34] <rockhopper_> ttyl
[30-May-2011 11:19:39] <rmatte> later
[30-May-2011 12:00:02] [disconnected at Mon May 30 12:00:02 2011]
[30-May-2011 12:00:02] [connected at Mon May 30 12:00:02 2011]
[30-May-2011 12:00:18] [zenoss-logger (logger bot) has joined #zenoss]
[30-May-2011 12:16:50] * ReGiStRaS says Happy Memorial Day to all!!!
[30-May-2011 16:29:52] bzed_ is now known as bzed
[30-May-2011 16:29:53] HKhan is now known as Hamzah
[30-May-2011 17:37:32] c00p_ is now known as c00p
[30-May-2011 22:28:49] <amorphic> Hello all... anybody around who has some knowledge on process monitoring via snmp?
[31-May-2011 00:00:01] [disconnected at Tue May 31 00:00:01 2011]
[31-May-2011 00:00:01] [connected at Tue May 31 00:00:01 2011]
[31-May-2011 00:00:16] [zenoss-logger (logger bot) has joined #zenoss]
[31-May-2011 00:00:16] <> [freenode-info] please register your nickname...don't forget to auto-identify!
[31-May-2011 05:27:42] fOrsberg is now known as forsberg
[31-May-2011 08:12:42] <forsberg> hmmm a few questions here, anyone remember where the mysql user/pass is located ?
[31-May-2011 08:13:15] <Hackman248> forsberg: advanced, event manager
[31-May-2011 08:13:35] <forsberg> yes cool thanks, event manager it was, im on 2.5.2
[31-May-2011 08:14:01] <Hackman248> forsberg: ah, just event manager on the left menu then :-)
[31-May-2011 08:14:35] <forsberg> hmm that wasnt it, after a reinstall/move of old installation
[31-May-2011 08:14:41] <forsberg> the even tab is always totally emtpy
[31-May-2011 08:18:21] <hmp> forsberg: you can get/set it from zendmd console
[31-May-2011 08:19:54] <Hackman248> forsberg: It should be. Hmp, how to set using dmd?
[31-May-2011 08:20:12] <hmp> Hackman248: you can get it there and set it through mysql console
[31-May-2011 08:20:17] <forsberg> ok yes, but found it and was corret, it was a matter of googling harder, had to clear browser cache..
[31-May-2011 08:21:33] <forsberg> so i have a pretty customized 2.5.2 32bit with 300ish devices, and i want 3.1.0 (newest) 64bit, any thoughts ? im on rpm isntalled on centos 5.6
[31-May-2011 08:21:38] <Hackman248> hmp: oh I thought forsberg was looking for Zenoss's list of credts to auth to mysql, not how to set the creds in mysqld
[31-May-2011 08:21:38] <forsberg> customized = lots of templates and bindings and shit
[31-May-2011 08:22:07] <forsberg> it was both, i wanted to change since i didnt remember what it was
[31-May-2011 08:22:36] <forsberg> but was zenoss/zenoss from localhost, and it was the browser cache clearing that did it
[31-May-2011 08:22:37] <Hackman248> forsberg: Gotcha
[31-May-2011 08:23:44] Hackman248 is now known as Hackman238
[31-May-2011 08:59:35] forsberg is now known as fOrsberg
[31-May-2011 09:14:45] <froztbyte> I really should work through the dev guide to see how to properly create my zenpacks one day
[31-May-2011 09:17:48] <fragfutter> you mean other then click it, export it, uninstall, restart, reinstall with link , restart?
[31-May-2011 09:18:10] <Hackman238> froztbyte: What do you mean?
[31-May-2011 09:19:33] <froztbyte> so that they don't include relations to system devices and some other minor things like that
[31-May-2011 09:24:20] <Sam-I-Am> sup nerds
[31-May-2011 09:32:06] <rmatte> sup
[31-May-2011 09:34:45] <Hackman238> Sam-I-Am: Ha1!
[31-May-2011 09:34:53] <Hackman238> Hello all.
[31-May-2011 09:35:00] <rmatte> hello
[31-May-2011 09:35:24] <Hackman238> rmatte: Sorry, I didnt get back from Austin until like 3AM yesterday- promise today. Official release this 4th :-)
[31-May-2011 09:37:05] <Sam-I-Am> yo yo yo
[31-May-2011 09:46:40] <rmatte> Hackman238: It's cool, but the sooner the better, I need to get rolling on some IPSLA projects and I've been holding off
[31-May-2011 09:47:08] <Sam-I-Am> meh, history searching failing on a number of event classes
[31-May-2011 09:47:17] <Hackman238> rmatte: Sorry 'bout that. It's been madness the last couple months work load wise.
[31-May-2011 09:47:18] <Sam-I-Am> somethings slowly going south again
[31-May-2011 09:47:34] <rmatte> Hackman238: I know, we're in the same boat you and I lol
[31-May-2011 09:47:56] <Sam-I-Am> pfft, you guys dont do any work
[31-May-2011 09:47:59] <Hackman238> rmatte: Thats for understanding.
[31-May-2011 09:48:05] <rmatte>
[31-May-2011 09:48:56] <Hackman238> Sam-I-Am: According to my girlfriend all I do is work LOL- I told her if she wants to continue to live at such high standards she'll need to put up with it or get a better job to cover the diff, herself
[31-May-2011 09:49:29] <Sam-I-Am> wow
[31-May-2011 09:49:42] <Sam-I-Am> wait, nerds have girlfriends?
[31-May-2011 09:49:46] <dec3pti0n> is there a reason zenoss keeps on picking up OS processes that I have not asked it to monitor ? like rsync or miniperl .. I think when it remodels all the devices at night it picks those up if they are running !
[31-May-2011 09:50:09] <Hackman238> dec3pti0n: That'd be right
[31-May-2011 09:50:40] <Hackman238> dec3pti0n: Let it pick them up- once they are there set each you dont want to monitor to 'not monitored'
[31-May-2011 09:50:47] <rmatte> Hackman238: Amen to that lol, if she's not working she can't comment
[31-May-2011 09:51:25] <rmatte> oh wait, I misread, so she is working, just not making that much
[31-May-2011 09:51:27] <Hackman238> rmatte: It kills me too where we live- we're like 15 minutes from all sorts of designer outlets so she can waste my fortune! LOL
[31-May-2011 09:51:34] <rmatte> haha
[31-May-2011 09:51:35] <Hackman238> rmatte: Correct LOL
[31-May-2011 09:52:03] <rmatte> It'd be cheaper just to buy a girlfriend
[31-May-2011 09:52:26] <Sam-I-Am> lol
[31-May-2011 09:52:30] <Sam-I-Am> rmatte: now you know my secret
[31-May-2011 09:52:37] <Sam-I-Am> and i only need her for a few hours
[31-May-2011 09:52:38] <rmatte> lol
[31-May-2011 09:53:03] <rmatte> Well, I was thinking for mail order bride than escort, but whatever floats your boat
[31-May-2011 09:53:06] <rmatte> lol
[31-May-2011 09:53:12] <rmatte> thinking more*
[31-May-2011 09:53:15] <Sam-I-Am> lol
[31-May-2011 09:53:27] <Hackman238> Sam-I-Am: I've heard that same advice from a business partner of mine in San Marocs, CA. LOL
[31-May-2011 09:53:34] <Hackman238> rmatte: LOL
[31-May-2011 09:54:01] <rmatte> muahaha
[31-May-2011 09:54:30] <Sam-I-Am> its popular
[31-May-2011 09:54:33] * Simon4 can't see a mail order bride coming in cheaper than a regular girlfriend
[31-May-2011 09:54:43] <Sam-I-Am> Simon4: but hourly does
[31-May-2011 09:54:48] <Simon4> Sam-I-Am: indeed
[31-May-2011 09:54:53] <Sam-I-Am> rather, at least i know i'll get what i'm paying for
[31-May-2011 09:55:57] <rmatte> Sam-I-Am often uses the phrase "Shhhhh, daddy's working!"
[31-May-2011 09:56:04] <rmatte>
[31-May-2011 09:56:10] <Hackman238> rmatte: Hahahaha!
[31-May-2011 09:56:43] <rmatte> lol
[31-May-2011 09:56:47] <Sam-I-Am> oh wow...
[31-May-2011 09:56:51] <dhopp> Sam-I-Am: is there a zenpack to monitor the hours?  Make sure I'm not over billed :-P
[31-May-2011 09:56:52] <Sam-I-Am> i'm not that bad
[31-May-2011 09:57:08] <Sam-I-Am> i tried polling her with snmp but it didnt go over so well
[31-May-2011 09:57:12] <rmatte> dhopp: wouldn't be difficult to write
[31-May-2011 09:57:20] <dhopp> rmatte: lol'
[31-May-2011 09:57:21] <Sam-I-Am> damn oid was too long to fit
[31-May-2011 09:57:28] <rmatte>
[31-May-2011 09:57:50] <rmatte> It's like a new age floppy joke
[31-May-2011 10:02:11] <rmatte> woohoo, I got paid today
[31-May-2011 10:02:55] * Sam-I-Am read that as laid
[31-May-2011 10:03:01] <dhopp> lol
[31-May-2011 10:03:11] * Sam-I-Am wonders where that zenoss logger is
[31-May-2011 10:03:16] <rmatte>
[31-May-2011 10:03:18] <rocket_> rmatte: another unemployment check? .. oh wait thats almost everyone else I know ..
[31-May-2011 10:03:35] <rmatte> rocket_: haha
[31-May-2011 10:03:50] rocket_ is now known as rocket
[31-May-2011 10:04:11] <Hackman238> Sam-I-Am: LOL!
[31-May-2011 10:04:31] <Hackman238> rocket: Oh man LOL
[31-May-2011 10:04:56] <dhopp> this channel is in rare form today
[31-May-2011 10:05:05] * fragfutter wonders if this channel is no longer safe for work
[31-May-2011 10:05:12] <rmatte> lol
[31-May-2011 10:05:12] <Hackman238> rocket: I'm waiting for the runaway inflation to collapse the economy at this point.
[31-May-2011 10:05:36] <Hackman238> dhopp: Agreed- feels like a mix of zenoss and 4chan
[31-May-2011 10:05:39] <dhopp> Hackman238: we are more likely have stagflation…which isn't much better
[31-May-2011 10:05:54] <rocket> what economy?
[31-May-2011 10:05:59] <Sam-I-Am> well, you get a bunch of angry nerds in a channel together...
[31-May-2011 10:06:08] <rocket> oh wait are you in the military? .. I see ...
[31-May-2011 10:06:08] <Hackman238> dhopp: Very possibly as well, and agreed.
[31-May-2011 10:06:09] <Sam-I-Am> its either going to be super nerdy or super raumchy
[31-May-2011 10:06:11] <Sam-I-Am> raunchy
[31-May-2011 10:06:26] <Sam-I-Am> or some scary combination of both i'd rather not think about
[31-May-2011 10:06:29] <Hackman238> Sam-I-Am: LOL
[31-May-2011 10:06:41] <jmp242> we all could try for professional . . .
[31-May-2011 10:06:47] <Sam-I-Am> Hackman238: i think high fuel prices will kill whatever recovery we sorta had </politics>
[31-May-2011 10:06:50] * rocket waits for the nerdy fart jokes ..
[31-May-2011 10:07:03] <Sam-I-Am> rocket: we're not that immature
[31-May-2011 10:07:16] <dhopp> Sam-I-Am:  speak for yourself
[31-May-2011 10:07:28] <Hackman238> Sam-I-Am: Rule 34 on Zenoss
[31-May-2011 10:07:48] <fragfutter> everyone check their iphone for some stupid app rocket is looking for
[31-May-2011 10:07:56] <Hackman238> rocket: Is who in the mil?
[31-May-2011 10:09:08] <Hackman238> Sam-I-Am: Whatever the case- build up that portfolio of gold/platinium
[31-May-2011 10:09:10] <rocket> Hackman238: I just meant our only economy is gov spending of which the military is a large part
[31-May-2011 10:09:40] <rocket> we dont make much anymore .. we sent all of that out of country with things like Nafta etc ..
[31-May-2011 10:09:50] <Hackman238> rocket: Oh gotcha- definately consists of most of it outside trading debts.
[31-May-2011 10:10:00] <rocket> but thats all I really want to say about it as I dont want to get into politics in here etc ..
[31-May-2011 10:10:26] <Hackman238> rocket: No sweat.
[31-May-2011 10:10:27] <dhopp> rocket: agreed
[31-May-2011 10:11:07] * dhopp senses a switch in topic back to mail order brides
[31-May-2011 10:11:29] <Hackman238> dhopp: Still waiting for rule 34 for Zenoss
[31-May-2011 10:11:38] <Hackman238> LOL
[31-May-2011 10:11:40] <rocket> I hear they are hot if they are from fukishima ..
[31-May-2011 10:11:59] <Hackman238> rocket: LOL
[31-May-2011 10:12:01] <dhopp> so..on to a more on topic subject…wtf is going on with 3.1.1 ?
[31-May-2011 10:12:18] <dhopp> seems like I keep hearing about it..but don't see anything about a release
[31-May-2011 10:12:38] <rocket> dhopp: its a patch fix release for 3.1
[31-May-2011 10:12:46] <fragfutter> dhopp: they are still removing relstore from it
[31-May-2011 10:13:15] <dhopp> rocket: yeah..but I thought it was supposed to have a bunch of performance tweaks (minus relstore of course)
[31-May-2011 10:13:21] <rocket> fragfutter: actually its a large performance fix release etc based on results from customers running 3.1
[31-May-2011 10:13:47] <Hackman238> fragfutter: Someone will make a relstorage pack- if not me, someone else for sure.
[31-May-2011 10:13:53] <Sam-I-Am> Hackman238: no gold here... only the latest in trendy bitcoins
[31-May-2011 10:14:10] <rocket> well the notes for relstore are already on the intertubes ...
[31-May-2011 10:14:40] <rocket> you can convert a 3.X system to relstore today ..
[31-May-2011 10:15:03] * Simon4 was tempted to lab that up
[31-May-2011 10:15:10] <dhopp> rocket: until 3.1.1 comes out and the api changes and you have a broken db
[31-May-2011 10:15:12] <rocket> we just dont have an upgrade strategy for that in place that we will support for enterprise
[31-May-2011 10:15:24] <rocket> dhopp: I dont think that should happen
[31-May-2011 10:15:55] <Hackman238> Sam-I-Am: LOL, might suggest you invest in some real coins :-)
[31-May-2011 10:15:58] <rocket> dhopp: that part doesnt change that much ..
[31-May-2011 10:16:25] <rocket> dhopp: most of what we change are based on object relations which a migration strategy is already in place etc
[31-May-2011 10:17:05] <Hackman238> rocket: True, but it's always nice to have a pack. I might throw it out there that I've considered writting one with an export feature so one can acctually spit out a Data.fs to upgrade
[31-May-2011 10:17:12] <Sam-I-Am> Hackman238: i cant afford real coins
[31-May-2011 10:17:29] <Sam-I-Am> but i can afford to spend $10k a year on electricity to "make" coins
[31-May-2011 10:17:55] <Hackman238> Sam-I-Am: Make'em and call them money. When the dollar is useless you're the new central bank. Brilliant!
[31-May-2011 10:18:01] <Sam-I-Am> the real winnars here are the utility companies
[31-May-2011 10:18:14] <davetoo> quarter-crushers?
[31-May-2011 10:18:17] <rocket> Hackman238: that is definately a possibility
[31-May-2011 10:18:20] <Hackman238> Sam-I-Am: Utility and Bankers.
[31-May-2011 10:18:20] <Sam-I-Am> i bet they're behind bitcoins
[31-May-2011 10:18:39] <davetoo> wtf... are bitcoins?
[31-May-2011 10:18:49] <Simon4> davetoo: don't ask, just run away
[31-May-2011 10:18:50] <Hackman238> davetoo: sir google knows.
[31-May-2011 10:19:53] <fragfutter> davetoo: numbers
[31-May-2011 10:20:10] <davetoo> Note w.r.t. the RelStorage blog entry on the community site: the bits about where to get mysql are totally screwy   Just get 5.1.57 from SunaCle directly
[31-May-2011 10:20:35] <davetoo> Forget about that French repo
[31-May-2011 10:21:23] <rocket> really though unless you are in the 4000+ node range you probably wont need relstorage and most of the time you can setup multiple zenoss masters instead
[31-May-2011 10:22:06] <Hackman238> rocket: Many core users fall in that catagory
[31-May-2011 10:22:24] <dhopp> rocket: but with multiple zenoss masters, you can't move devices between those "masters" can you?  as in the masters have their own device db..right?
[31-May-2011 10:22:24] <davetoo> Oeh Noez: the house is out of coffee and I have to be awake in six minutes (it's 07:24 here)
[31-May-2011 10:22:47] <rocket> Hackman238: and I *think* that there is a patch for 3.1 that will be in 3.1.1 that will minimize the need for relstorage anyway
[31-May-2011 10:22:53] <Hackman238> rocket: Most of my core customers do acctually. They wont buy Zenoss because it costs so much comparred to having a full service contract with me or others.
[31-May-2011 10:23:10] <jmp242> Good to know - it would be useful to have a stock Core expected scaleability thing
[31-May-2011 10:23:23] <Hackman238> rocket: Interesting to know. Is that the one concerning mitigrating hub floods?
[31-May-2011 10:23:24] <rocket> since the core issue is the invalidation blocking stuff going on
[31-May-2011 10:23:35] <rocket> Hackman238: yes
[31-May-2011 10:23:36] <Sam-I-Am> davetoo: uh... jump out the window
[31-May-2011 10:23:56] * dhopp doesn't like coffee
[31-May-2011 10:24:21] <Sam-I-Am> well, someone here is not a nerd
[31-May-2011 10:24:33] <Sam-I-Am> (unless you're slamming back energy drinks)
[31-May-2011 10:24:43] <dhopp> Sam-I-Am:  hey…I happen to have a half drank mt. dew sitting next to me..
[31-May-2011 10:24:44] <fragfutter> Sam-I-Am: there is this stuff called tea
[31-May-2011 10:24:47] <dhopp> and it's only 9:24 AM here
[31-May-2011 10:24:53] <Hackman238> rocket: Gotcha. Very much needed in many places, though I've found that 1:1 zenhub to daemon using zenhub and a huge invalidation queue entry  seems to fix really bad cases
[31-May-2011 10:25:31] <Hackman238> dhopp / Sam-I-Am : Mt. Dew vs Coffee - GO!
[31-May-2011 10:25:49] <dhopp> Hackman238:  that's tastes good, the other like shit
[31-May-2011 10:25:50] <rocket> Hackman238: "The zenhub performance improvements are in 3.1.x"
[31-May-2011 10:26:17] <rocket> Hackman238: so there is a queue of changes and the daemons are notified instead of the current blocking loop
[31-May-2011 10:26:33] <davetoo> I *require* caffienne, and prefer Peets
[31-May-2011 10:26:38] <Hackman238> rocket: Ah thats much better than me throwing more hubs at the problem
[31-May-2011 10:26:45] <jmp242> I hate coffee too
[31-May-2011 10:26:51] <jmp242> Diet Coke for me in the morning lol
[31-May-2011 10:26:52] <rocket> Hackman238: that change alone will probably resolve almost all needs for relstorage
[31-May-2011 10:26:56] <Hackman238> Whats with all the coffee haters?
[31-May-2011 10:27:06] <JohnnyNoc> dr oz says 2-3 cups is good for your health!
[31-May-2011 10:27:07] <jmp242> I'm just weird
[31-May-2011 10:27:13] <JohnnyNoc> 2-3 cups/day
[31-May-2011 10:27:16] <Hackman238> rocket: Interesting. Is it in the SVN?
[31-May-2011 10:27:24] <rocket> Hackman238: and after that .. you probably should be thinking about enterprise anyway
[31-May-2011 10:27:32] <dhopp> I don't like steak either...
[31-May-2011 10:27:41] <JohnnyNoc> dhopp blashphemy
[31-May-2011 10:28:07] <dhopp> JohnnyNoc: nope..I don't like it…but I'll eat hamburger…now that's weird
[31-May-2011 10:28:17] <JohnnyNoc> no, it's delicious!
[31-May-2011 10:28:20] <rocket> Hackman238: it is in the internal svn repo for 3.1.1 I dont know how the bits for core 3.1.1 will be pushed out yet .. that not my job to worry about ..
[31-May-2011 10:28:21] <JohnnyNoc> :P~
[31-May-2011 10:28:23] <jmp242> mmm, I only like some steaks
[31-May-2011 10:28:31] <Hackman238> rocket: I'll be honest, most customers wont touch Ent wearing neoprene gloves.
[31-May-2011 10:28:37] <jmp242> sirloin is boring
[31-May-2011 10:28:42] <Hackman238> rocket: It never gets past budget.
[31-May-2011 10:29:06] <Hackman238> dhopp: Steak!?
[31-May-2011 10:29:13] <JohnnyNoc> i tend to gravitate towards ribeyes :B
[31-May-2011 10:29:15] <Hackman238> dhopp: Its like the best food in the world?
[31-May-2011 10:29:36] <dhopp> Hackman238 / rocket: I tried to get Enterprise…I REALLY tried…but the budget just wasn't going to happen
[31-May-2011 10:29:55] <JohnnyNoc> i need per device cost to be like $20/each before my boss says we can use enterprise
[31-May-2011 10:29:56] <JohnnyNoc>
[31-May-2011 10:30:05] <Hackman238> dhopp: Thats my point to Zenoss Inc.
[31-May-2011 10:30:22] <jmp242> Strip steak for me - well done ,and I know how blasphemous that is lol
[31-May-2011 10:30:30] <dhopp> Hackman238: that's what everybody tells me…but I just don't like it…every once in a great while I can eat some prime rib..but that's rare
[31-May-2011 10:30:41] <dhopp> no pun intended
[31-May-2011 10:30:42] <Hackman238> dhopp: Wow.
[31-May-2011 10:31:00] <Sam-I-Am> JohnnyNoc: mmm ribeyes
[31-May-2011 10:31:07] <rocket> what is licensing like for the other large vendors?  Its been a long time since I have checked... and where I am at now I havent had time to look into that etc
[31-May-2011 10:31:23] <Sam-I-Am> and it has to be rare
[31-May-2011 10:31:37] <Sam-I-Am> like... sliced off the steer, then put over a cigarette lighter for a few minutes
[31-May-2011 10:31:41] <rocket> and I have no pull in that side of our organization other than to mention what is being said etc
[31-May-2011 10:31:45] <Hackman238> rocket: Alrighty, gotcha. Is it possible to point me to the rep which might point me to the blocking code so I can tweak RackSpaces 2.5.2 instance?
[31-May-2011 10:32:10] <Hackman238> rocket: We know and we appreciate you listening.
[31-May-2011 10:32:11] <Sam-I-Am> i wonder if licensing for large vendors is customized
[31-May-2011 10:32:31] <Sam-I-Am> er, large customers
[31-May-2011 10:32:33] <jmp242> I'm sure there's negotiated license costs
[31-May-2011 10:32:43] <jmp242> every major vendor does it
[31-May-2011 10:33:34] <JohnnyNoc> rocket tell them to find a way to offer reduced cost licenses for reduced services/support
[31-May-2011 10:33:36] <fragfutter> licensing oracle and negotiating prices is for example quite some "fun"
[31-May-2011 10:33:51] <JohnnyNoc> that's my .02 on that subject
[31-May-2011 10:34:25] <rocket> JohnnyNoc: some of this is a critical mass situation .. I think .. again this is my personal opinion .. and not that of managements
[31-May-2011 10:35:22] <froztbyte> uh
[31-May-2011 10:35:23] <JohnnyNoc> rocket totally understood
[31-May-2011 10:35:36] <froztbyte> can I backup on a 32bit zenoss, and install on a 64bit zenoss?
[31-May-2011 10:35:42] <f00fSteR> Hackman238, wb
[31-May-2011 10:35:43] <froztbyte> s/install/recover/
[31-May-2011 10:35:56] <rocket> JohnnyNoc: we need the dollars to get enough engineers to build up the teams to get to a point to offer other options etc
[31-May-2011 10:36:00] <jmp242> not really froztbyte
[31-May-2011 10:36:06] <jmp242> you have to convert or drop the rrd files
[31-May-2011 10:36:15] <froztbyte> CorruptedDataError: Error reading oid 0x4611. Found '\x02.}q\x03(U\x08isdialogq\x04I01\nU\x0bdescriptionq\x05U\x0bAdd' at 11203662
[31-May-2011 10:36:19] <froztbyte> ^ and if I get that?
[31-May-2011 10:36:37] <froztbyte> almost sounds like the backup is bad
[31-May-2011 10:36:45] <jmp242> I ahve no idea
[31-May-2011 10:36:51] <froztbyte> lemme paste the whole stacktrack
[31-May-2011 10:36:52] <jmp242> never seen anything like that
[31-May-2011 10:37:09] <jmp242> I always just backup the whole $ZENHOME
[31-May-2011 10:37:10] <dhopp> froztbyte:  rrd doesn't migrate easily between 32bit and 64bit if I recall right
[31-May-2011 10:37:13] <JohnnyNoc> rocket again, i understand, but from our perspective, as a relatively large organization with 1000+ devices being monitord
[31-May-2011 10:37:14] <froztbyte> <- trying to open devices
[31-May-2011 10:37:15] <jmp242> and paste that in
[31-May-2011 10:37:17] <Hackman238> f00fSteR: Hows ir going
[31-May-2011 10:37:25] <JohnnyNoc> for $125k/year we could hire 1, maybe 2 FT guys to take care of Zenoss
[31-May-2011 10:37:31] <froztbyte> dhopp: yeah, that's a minimal problem
[31-May-2011 10:37:32] <JohnnyNoc> or, we could contract it out to some 3rd party and be done with it
[31-May-2011 10:37:39] <Sam-I-Am> JohnnyNoc: feel free to hire me for 125k
[31-May-2011 10:37:51] <jmp242> looks bad froztbyte
[31-May-2011 10:37:52] <JohnnyNoc> Sam-I-Am hahaha you'd have to be a sr sysadmin
[31-May-2011 10:37:58] <JohnnyNoc> but let me know, send me a resume
[31-May-2011 10:38:00] <JohnnyNoc>
[31-May-2011 10:38:03] <dhopp> JohnnyNoc: I'm only 124K :-P
[31-May-2011 10:38:05] <JohnnyNoc> we're in chicago
[31-May-2011 10:38:06] <JohnnyNoc> hahaha
[31-May-2011 10:38:08] <froztbyte> 2.5.1 on both sides
[31-May-2011 10:38:16] <froztbyte> 32bit on the one, 64bit on the other
[31-May-2011 10:38:21] <froztbyte> (64bit is the new one)
[31-May-2011 10:38:22] <Hackman238> JohnnyNoc: What comapny?
[31-May-2011 10:38:31] <Sam-I-Am> ha
[31-May-2011 10:38:32] <froztbyte> otherwise I just say "fuckit" and drop all events/perf data
[31-May-2011 10:38:38] <Hackman238> Alright f00fSteR, where did we leave off?
[31-May-2011 10:38:43] <rocket> 2.5.1 ... should really get to 2.5.2 and run zenchkrels -r -x1
[31-May-2011 10:38:50] <f00fSteR> Hackman238, it's goin good. I've had a most excellent weekend. broke up with my gf... life is good birds chirping, wind whistling.
[31-May-2011 10:38:53] <rocket> 2.5.1 has serious relationship issues
[31-May-2011 10:39:00] <Sam-I-Am> f00fSteR: that bad?
[31-May-2011 10:39:05] <JohnnyNoc> anyone here use HPMon zenpack and notice timeout problems with the hpexpansioncardmap collector plygin?
[31-May-2011 10:39:08] <JohnnyNoc> plugin too
[31-May-2011 10:39:10] <froztbyte> rocket: yeah, that's the next step
[31-May-2011 10:39:18] <Sam-I-Am> rocket: does that mean i'm running 2.5.1?  i have serious relationship issues...
[31-May-2011 10:39:29] <dhopp> JohnnyNoc: I'm using it…but I don't have that many devices on it yet
[31-May-2011 10:39:36] <froztbyte> rocket: otherwise I should upgrade the 32bit system to 2.5.2, export there and then swing over to new system
[31-May-2011 10:39:36] <dhopp> JohnnyNoc: I haven't seen any problems though
[31-May-2011 10:39:40] <Hackman238> f00fSteR: Alrighty.
[31-May-2011 10:39:47] <rocket> froztbyte: thats the first step yes
[31-May-2011 10:39:58] <froztbyte> rocket: okay, lemme do it that way around
[31-May-2011 10:40:15] <f00fSteR> Sam-I-Am, life is good always no woman or man adds to my happiness... they simply compliment...
[31-May-2011 10:40:25] <Sam-I-Am> JohnnyNoc: if i moved to chicago i'd die from chicago-style hotdog and hot beef overload
[31-May-2011 10:40:34] <f00fSteR> Hackman238, we left off i think after we installed the zenpacks and the gui itself
[31-May-2011 10:40:36] <JohnnyNoc> dhopp thanks, not sure why but when i try to model/add a device
[31-May-2011 10:40:37] <Sam-I-Am> f00fSteR: good way to think of it
[31-May-2011 10:40:40] <f00fSteR> we were having sql connection errors
[31-May-2011 10:40:55] <Sam-I-Am> f00fSteR: my girlfriend and i were going strong then she just wandered off one day
[31-May-2011 10:42:14] <Hackman238> f00fSteR: Oh right
[31-May-2011 10:42:18] <f00fSteR> Sam-I-Am, same... is it the warm spring weather... i'd like to think it's the changing of her BC's ... just couldnt troubleshoot this one... she was a fine young lassie tho
[31-May-2011 10:42:20] <Hackman238> f00fSteR: So the gui runs, right?
[31-May-2011 10:42:26] <f00fSteR> Hackman238, positive
[31-May-2011 10:42:49] <f00fSteR> BRB coffee. Hackman238, you remember now ?
[31-May-2011 10:43:13] <Hackman238> f00fSteR: Okay, lets su root and rpm -Uvh --nodeps zenoss-core.packs.rpm again
[31-May-2011 10:43:20] <f00fSteR> Sam-I-Am, to be honest... your happiness never changes it's always the same in life... you can wlays choose to smile and be happy... fight club, sufism, power of now, zen buddism, yoga
[31-May-2011 10:43:26] <f00fSteR> kk BRB
[31-May-2011 10:43:27] <f00fSteR> 2 mins
[31-May-2011 10:43:33] <Hackman238> f00fSteR: I'm lucky I remember my addresses some days LOL. Alrighty
[31-May-2011 10:44:09] <Hackman238> Sam-I-Am: Lost her in the parking garage and never saw her again?
[31-May-2011 10:44:26] <Sam-I-Am> Hackman238: more or less.. wait, were you the one that stole her?
[31-May-2011 10:44:55] <Hackman238> Sam-I-Am: No...I dont drive a black van *look innocent*
[31-May-2011 10:44:57] <Sam-I-Am> f00fSteR: all women are crazy... it just depends to what extent of crazy you can tolerate
[31-May-2011 10:45:37] <Sam-I-Am> f00fSteR: this was by far the best relationship i've had and it ended like all the others... for no apparent reason and i'll never know why
[31-May-2011 10:48:36] <f00fSteR> Sam-I-Am, relationships tend to be broken to start off with... people think they need each other to be whole ... not necessarily true
[31-May-2011 10:48:58] <Hackman238> f00fSteR: Agreed.
[31-May-2011 10:49:02] <Hackman238> f00fSteR: BRB
[31-May-2011 10:49:13] <Sam-I-Am> f00fSteR: single life isnt all that bad
[31-May-2011 10:49:32] <f00fSteR> complimenting is nicwer... you dance to each others tune... it's a play of words... but time heals all... and sooner or later... a better relationship will pop up... it's inevitible... 70% of the world consists of women... and well the 30% of us gotta make end meet
[31-May-2011 10:50:16] <f00fSteR> how do i set a disk usage warning at 80% on a 2.4 installation
[31-May-2011 10:50:18] <f00fSteR> anyone know ?
[31-May-2011 10:50:30] <f00fSteR> madd nub qestion i know... kill me not
[31-May-2011 10:51:32] <dhopp> f00fSteR:  you should be able to modify the template…either at the device / fs level..or globally
[31-May-2011 10:53:50] <f00fSteR> Hackman238, remember this? since ast friday: rpm -Uvh --nodeps zenoss-core-zenpacks-2.4.2.el5.x86_64.rpm --replacepkgs --replacefiles --force &
[31-May-2011 10:55:52] <Hackman238> f00fSteR: Ah, right
[31-May-2011 10:56:08] <Hackman238> f00fSteR: cancel it, ps aux | grep zeo
[31-May-2011 10:57:03] <Hackman238> f00fSteR: Is it possible to get me a vnc or let me ssh in?
[31-May-2011 11:06:17] <f00fSteR> Hackman238, cancelled, done
[31-May-2011 11:07:03] <f00fSteR> Hackman238, we're behind a firewall, I'd have to get them to nat or open a port... requires VP signature and parents company yadi yada
[31-May-2011 11:07:40] <fragfutter> fOrsberg: make a reverse vnc. if you can do the outbound connection
[31-May-2011 11:08:40] <fragfutter> f00fSteR: sorry. wrong person
[31-May-2011 11:09:23] <fragfutter> *argh* tab completion. guess you'll figure out who was meant with which message.
[31-May-2011 11:11:17] <f00fSteR> think we can do this on here... or maybe try some tunnling hack as with fragfutter
[31-May-2011 11:11:18] <dhopp> fragfutter:  it could be worse…
[31-May-2011 11:16:29] <Hackman238> f00fSteR: It would be much easier If I could jump on
[31-May-2011 11:16:48] <Sam-I-Am> Hackman238: just hack your way in, hackman
[31-May-2011 11:16:50] <f00fSteR> Hackman238, what do you suggest... it has a 10.* ip address and is no way opened from the outside
[31-May-2011 11:16:59] <Sam-I-Am> ssh tunnel!
[31-May-2011 11:17:05] <Sam-I-Am> (reverse)
[31-May-2011 11:17:07] <f00fSteR> Sam-I-Am, link me!
[31-May-2011 11:17:09] <fragfutter> reverse vnc
[31-May-2011 11:17:24] <f00fSteR> there is no gui
[31-May-2011 11:17:27] <f00fSteR> it's only ssh
[31-May-2011 11:17:41] <Sam-I-Am> reverse ssh tunnel GO
[31-May-2011 11:17:42] <f00fSteR> Hackman238, i presume you know links and lynx from the terminal days
[31-May-2011 11:17:57] <Sam-I-Am> wget
[31-May-2011 11:18:00] <dhopp> Sam-I-Am:  that's assuming he can ssh out
[31-May-2011 11:18:01] <Hackman238> f00fSteR:
[31-May-2011 11:18:04] <f00fSteR>
[31-May-2011 11:18:06] <f00fSteR> ahh fack
[31-May-2011 11:18:15] <f00fSteR> Hackman238, youre already on my box huh ?
[31-May-2011 11:18:22] <Hackman238> f00fSteR: I'll do from my cell. is no sweat
[31-May-2011 11:18:29] <Hackman238> f00fSteR: Negative LOL
[31-May-2011 11:18:39] <Hackman238> f00fSteR: Lynx=best browser
[31-May-2011 11:18:47] <f00fSteR> uh huh ... i see wireshark running
[31-May-2011 11:18:49] <Sam-I-Am> elinks = bettar
[31-May-2011 11:18:53] <Sam-I-Am> lets take this outside
[31-May-2011 11:18:57] <Hackman238> Sam-I-Am: LOL
[31-May-2011 11:19:14] <f00fSteR> bitchX > Xchat
[31-May-2011 11:19:21] <Sam-I-Am> irssi!
[31-May-2011 11:19:29] <Sam-I-Am> wait for it..
[31-May-2011 11:19:31] <Sam-I-Am> vi > emacs
[31-May-2011 11:19:33] * Sam-I-Am runs out
[31-May-2011 11:19:43] <Sam-I-Am> you knew it was coming
[31-May-2011 11:19:54] <Hackman238> Sam-I-Am: irssi here too
[31-May-2011 11:19:55] <dhopp> Sam-I-Am: no need to run when you are stating the truth
[31-May-2011 11:20:05] <fragfutter> Sam-I-Am: and you are totaly wrong. it must be vim > emacs
[31-May-2011 11:20:40] <Sam-I-Am> lol
[31-May-2011 11:20:53] <Hackman238> fragfutter: wine notepad > emacs; vi > vim, vim > emacs
[31-May-2011 11:20:53] * f00fSteR slaps emacs operating system users with a whale of a trout
[31-May-2011 11:21:17] <f00fSteR> whyy why would you use a 140 text based editor... with an email program browser... sigh
[31-May-2011 11:21:27] <f00fSteR> emacs OS FTL!
[31-May-2011 11:21:28] <mattray> question.. since I've been away from the state of the art… did either rmatte or bigegor's Windows stuff ever cover everything 100%?
[31-May-2011 11:21:36] <mattray> ie. pure SNMP or pure WMI?
[31-May-2011 11:21:44] <dhopp> Hackman238: wine?  Really…you think running a windows application, on linux makes it any better
[31-May-2011 11:22:00] <Hackman238> mattray: Not positive, sorry.
[31-May-2011 11:22:06] <fragfutter> mattray: AFAIK bigegors WMI can monitor any perfmon value
[31-May-2011 11:22:09] <Hackman238> dhopp: Not at all, just for the lulz
[31-May-2011 11:22:18] <mattray> yeah, I wanted modeling + monitoring
[31-May-2011 11:22:45] <Hackman238> mattray: Woudlnt you demolish the box with monitoring if you  hit all WMI counters?
[31-May-2011 11:23:16] <f00fSteR> Hackman238, what ip an username an i rev-sshing into ?
[31-May-2011 11:23:44] <f00fSteR> ssh -R 19999:localhost:22 Hackman238 isnt working ?
[31-May-2011 11:25:07] <mattray> Hackman238: was supposed to be all-in-one modeling and monitoring via WMI
[31-May-2011 11:25:17] <mattray> but I may need to do SNMP
[31-May-2011 11:26:11] <mattray> so will get some looks
[31-May-2011 11:26:28] <mattray> can use Chef's Ohai agent to decide how to monitor
[31-May-2011 11:28:00] <Hackman238> mattray: Gotcha
[31-May-2011 11:30:27] <dhopp> is there good docs on the dmd and all of the functions it provides?
[31-May-2011 11:30:35] <dhopp> command line I mean
[31-May-2011 11:32:01] <Hackman238> f00fSteR: Whats the output of ps aux | grep zeo?
[31-May-2011 11:32:17] <f00fSteR> ohh it's not running i did a killall hup on the pid
[31-May-2011 11:32:51] <Hackman238> f00fSteR: su zenoss, zeoctl start; zeoctl status
[31-May-2011 11:33:07] <Hackman238> f00fSteR: if running, ls -lh /opt/zenoss/packs/
[31-May-2011 11:33:20] <Hackman238> f00fSteR: Doe exist?
[31-May-2011 11:33:42] <f00fSteR> it exists
[31-May-2011 11:33:46] <f00fSteR> beautifully at that
[31-May-2011 11:34:01] <Hackman238> f00fSteR: cd /opt/zenoss/packs
[31-May-2011 11:34:10] <f00fSteR> k
[31-May-2011 11:34:15] <f00fSteR> in packs
[31-May-2011 11:34:22] <Hackman238> f00fSteR: su zenoss; zenpack --install={pick a pack} {enter}
[31-May-2011 11:35:47] <f00fSteR> yeach one ?
[31-May-2011 11:35:57] <f00fSteR> meaning one by one of the following ?
[31-May-2011 11:35:59] <Hackman238> f00fSteR: Any- first one is fine
[31-May-2011 11:36:16] <f00fSteR> ZenPacks.zenoss.ApacheMonitor-2.0.2-py2.4.egg  ZenPacks.zenoss.FtpMonitor-1.0.1-py2.4.egg   ZenPacks.zenoss.JabberMonitor-1.0.1-py2.4.egg  ZenPacks.zenoss.NNTPMonitor-1.0.1-py2.4.egg
[31-May-2011 11:36:16] <f00fSteR> ZenPacks.zenoss.DellMonitor-2.0.0-py2.4.egg    ZenPacks.zenoss.HPMonitor-2.0.0-py2.4.egg    ZenPacks.zenoss.LDAPMonitor-1.0.1-py2.4.egg    ZenPacks.zenoss.NtpMonitor-2.0.2-py2.4.egg
[31-May-2011 11:36:16] <f00fSteR> ZenPacks.zenoss.DigMonitor-1.0.1-py2.4.egg     ZenPacks.zenoss.HttpMonitor-2.0.1-py2.4.egg  ZenPacks.zenoss.LinuxMonitor-1.0.0-py2.4.egg   ZenPacks.zenoss.RPCMonitor-1.0.1-py2.4.egg
[31-May-2011 11:36:16] <f00fSteR> ZenPacks.zenoss.DnsMonitor-2.0.1-py2.4.egg     ZenPacks.zenoss.IRCDMonitor-1.0.1-py2.4.egg  ZenPacks.zenoss.MySqlMonitor-2.0.2-py2.4.egg   ZenPacks.zenoss.ZenJMX-3.1.7-py2.4.egg
[31-May-2011 11:36:32] <Hackman238> f00fSteR: Any one is good. is a test
[31-May-2011 11:36:48] <f00fSteR> zenpack --install=ZenPacks.zenoss.ApacheMonitor-2.0.2-py2.4.egg
[31-May-2011 11:36:48] <f00fSteR> INFO:zen.ZPLoader:Loading /opt/zenoss/ZenPacks/ZenPacks.zenoss.ApacheMonitor-2.0.2-py2.4.egg/ZenPacks/zenoss/ApacheMonitor/objects/objects.xml
[31-May-2011 11:36:48] <f00fSteR> INFO:zen.AddToPack:End loading objects
[31-May-2011 11:36:48] <f00fSteR> INFO:zen.AddToPack:Processing links
[31-May-2011 11:36:48] <f00fSteR> INFO:zen.AddToPack:Loaded 0 objects into the ZODB database
[31-May-2011 11:36:50] <f00fSteR> INFO:zen.HookReportLoader:loading reports from:/opt/zenoss/ZenPacks/ZenPacks.zenoss.ApacheMonitor-2.0.2-py2.4.egg/ZenPacks/zenoss/ApacheMonitor/reports
[31-May-2011 11:36:52] <f00fSteR> INFO:zen.migrate:Converting Apache templates to use totalAccesses/totalKBytes
[31-May-2011 11:36:55] <f00fSteR> ERROR: zenpack command failed. Reason: (1045, "Access denied for user 'zenoss'@'localhost' (using password: YES)")
[31-May-2011 11:36:59] <f00fSteR> same mysql error
[31-May-2011 11:37:08] <Hackman238> f00fSteR: Ah right, the sql error
[31-May-2011 11:37:15] <f00fSteR> mysql> show databases;
[31-May-2011 11:37:16] <f00fSteR> +--------------------+
[31-May-2011 11:37:16] <f00fSteR> | Database           |
[31-May-2011 11:37:16] <f00fSteR> +--------------------+
[31-May-2011 11:37:16] <f00fSteR> | information_schema |
[31-May-2011 11:37:16] <f00fSteR> | events             |
[31-May-2011 11:37:19] <f00fSteR> | mysql              |
[31-May-2011 11:37:20] <f00fSteR> | test               |
[31-May-2011 11:37:22] <f00fSteR> +--------------------+
[31-May-2011 11:37:23] <fragfutter> f00fSteR: FLOOOD!
[31-May-2011 11:37:25] <f00fSteR> 4 rows in set (0.00 sec)
[31-May-2011 11:37:28] <f00fSteR> events i presume is zenoss
[31-May-2011 11:37:31] <f00fSteR> fragfutter, sorrieee!
[31-May-2011 11:37:37] <Hackman238> f00fSteR: as root or zenoss you logged in to mysql?
[31-May-2011 11:37:43] <f00fSteR> root
[31-May-2011 11:37:46] <Hackman238> ok
[31-May-2011 11:37:51] <Hackman238> f00fSteR: 1 sec
[31-May-2011 11:37:54] <f00fSteR> exit -> root -> mysql (no passwd)
[31-May-2011 11:38:23] <f00fSteR> iptables are off fyi
[31-May-2011 11:38:50] <Hackman238> f00fSteR: as root run /opt/zenoss/bin/zenoss_init_pre
[31-May-2011 11:39:16] <Hackman238> It'll recreate the permissions and fix other problems. Should bunk on building zope- which is ok
[31-May-2011 11:41:40] <Hackman238> f00fSteR: Alternatively you can:
[31-May-2011 11:41:45] <Hackman238> grant all on events.* to 'zenoss'@'localhost' identified by 'zenoss';
[31-May-2011 11:41:46] <Hackman238> grant super, file on *.* to 'zenoss'@'localhost';
[31-May-2011 11:41:46] <Hackman238> grant SELECT on mysql.user to 'zenoss'@'localhost';
[31-May-2011 11:41:46] <Hackman238> grant all on events.* to 'zenoss'@'%' identified by 'zenoss';
[31-May-2011 11:41:46] <Hackman238> grant SELECT on mysql.user to 'zenoss'@'%';
[31-May-2011 11:41:57] <Hackman238> flush privileges;
[31-May-2011 11:42:27] <Hackman238> f00fSteR: Let me know when ops done
[31-May-2011 11:44:41] <dhopp> hrm...
[31-May-2011 11:44:45] <dhopp> this doesn't look right..
[31-May-2011 11:44:46] <dhopp> >>> print dmd
[31-May-2011 11:44:46] <dhopp> <DataRoot at dmd>
[31-May-2011 11:44:46] <dhopp> >>>
[31-May-2011 11:44:51] <dhopp> shouldn't that be /zport/dmd ?
[31-May-2011 11:45:12] <Hackman238> dhopp: Nope
[31-May-2011 11:45:51] <dhopp> Hackman238:  >>> find('localhost.localdomain')
[31-May-2011 11:45:52] <dhopp> >>>
[31-May-2011 11:45:56] <dhopp> shouldn't that find something?
[31-May-2011 11:46:05] <Hackman238> dhopp: only if that device is in zenoss
[31-May-2011 11:46:23] <Hackman238> dhopp: try dmd.Devices.findDevice('known existing device')
[31-May-2011 11:46:30] <dhopp> Hackman238:  ahh…it's in there as 'localhost'
[31-May-2011 11:46:51] <Hackman238> dhopp:  :-)
[31-May-2011 11:46:59] <dhopp> Hackman238:  only one devices…new zenoss installation so I can get familiar with zendmd for writing some scripts
[31-May-2011 11:47:10] <Hackman238> dhopp: Ah gotcha
[31-May-2011 11:47:39] <Hackman238> dhopp: zendmd is pretty slick.
[31-May-2011 11:47:39] <dhopp> Hackman238:  figured I didn't want to screw up a real installation (although I don't have zenoss in production yet)
[31-May-2011 11:47:58] <Hackman238> dhopp: Good idea- dmd is dangerous
[31-May-2011 11:53:21] <dhopp> Hackman238: if I use zendmd to create a new device class..should the script call sync() and commit() at the end?
[31-May-2011 11:54:22] <Hackman238> dhopp: You'll want to sync() before commit().
[31-May-2011 11:54:40] <dhopp> Hackman238: that's what I thought..I just wanted to verify
[31-May-2011 11:54:42] <Hackman238> dhopp: extra syncs will slow DMD especially if they are in an iterative routine
[31-May-2011 11:55:20] <Hackman238> dhopp: Also, set a if count > x to commit during iterative ops.
[31-May-2011 11:55:41] <dhopp> Hackman238: I'm not sure I understand
[31-May-2011 11:55:47] <Hackman238> dhopp: If you drop huge list of transactions on dmd suddenly it increases the likelyhood of conflict and problems.
[31-May-2011 11:55:59] <Hackman238> counter = 0
[31-May-2011 11:56:13] <Hackman238> for i in dmd.Devices.getsubDevices():
[31-May-2011 11:56:25] <Hackman238>    i.setProdState(300)
[31-May-2011 11:56:40] <Hackman238>    i.setPerformanceMonitor('localhost')
[31-May-2011 11:56:48] <Hackman238>    counter = counter + 1
[31-May-2011 11:56:54] <Hackman238>    if counter > 10:
[31-May-2011 11:56:57] <Hackman238>       sync()
[31-May-2011 11:56:59] <Hackman238>       commit()
[31-May-2011 11:57:03] <dhopp> gotcha
[31-May-2011 11:57:14] <f00fSteR> Loading initial Zenoss objects into the Zeo database
[31-May-2011 11:57:14] <f00fSteR> (this can take a few minutes)
[31-May-2011 11:57:14] <f00fSteR> zport portal object exits; exiting.
[31-May-2011 11:57:14] <f00fSteR> Stopping Zope Object Database
[31-May-2011 11:57:14] <f00fSteR> . daemon process stopped
[31-May-2011 11:57:34] <Hackman238> dhopp: that way it balances safety and reliability against slowing up to commit and sync
[31-May-2011 11:57:52] <Hackman238> f00fSteR: ps aux | grep zeo
[31-May-2011 11:57:57] <Hackman238> f00fSteR: is running?
[31-May-2011 11:58:38] <dhopp> However..what if dmd.Devices.getSubDevices returned only 10 devices?  Then your sync/commit never happen
[31-May-2011 11:59:04] <Hackman238> dhopp: just have a sync() and commit() at the end for good measure
[31-May-2011 11:59:13] <dhopp> gotcha
[31-May-2011 11:59:22] <Hackman238> dhopp: Rather just outside the for loop is what I meant
[31-May-2011 11:59:35] <dhopp> Hackman238:  I knew what you meant
[31-May-2011 11:59:55] <Hackman238> dhopp: I also reccomend you use try: except: to catch problems.
[31-May-2011 12:00:01] [disconnected at Tue May 31 12:00:01 2011]
[31-May-2011 12:00:01] [connected at Tue May 31 12:00:01 2011]
[31-May-2011 12:00:17] [zenoss-logger (logger bot) has joined #zenoss]
[31-May-2011 12:00:25] <dhopp> Hackman238: I will…but right now I'm just trying to make it work…don't care if I blow it up
[31-May-2011 12:01:13] <Hackman238> dhopp: Gtocha.
[31-May-2011 12:04:30] <f00fSteR> Hackman238, it stops after saying zport portal exists
[31-May-2011 12:04:35] <f00fSteR> and then exits
[31-May-2011 12:05:48] <f00fSteR> Hackman238, i ran all the sql queries
[31-May-2011 12:06:30] <Hackman238> f00fSteR: su zenoss; zeoctl start
[31-May-2011 12:07:08] <Hackman238> f00fSteR: if running, then su zenoss, cd /opt/zenoss/packs; zenpack --install={pick one} {enter}
[31-May-2011 12:07:45] <f00fSteR> INFO:zen.ZenPackCMD:Previous ZenPack exists with same name ZenPacks.zenoss.ApacheMonitor
[31-May-2011 12:07:46] <f00fSteR> ERROR: zenpack command failed. Reason: (1045, "Access denied for user 'zenoss'@'localhost' (using password: YES)")
[31-May-2011 12:08:15] <rocket> f00fSteR: your database access is wrong .. the grants need to be fixed
[31-May-2011 12:08:32] <Hackman238> f00fSteR: open the gui
[31-May-2011 12:08:48] <f00fSteR> Hackman238, rocket, i've done the whole batch of grant privilages ending with the flush on mysql for zenoss at % identified by zenoss
[31-May-2011 12:08:48] <Hackman238> f00fSteR: click event manager on left menu
[31-May-2011 12:09:19] <Hackman238> on that page fill in zenoss for the mysql user and zenoss for the password
[31-May-2011 12:09:47] <f00fSteR> kk
[31-May-2011 12:09:49] <Hackman238> f00fSteR: then click save
[31-May-2011 12:10:03] <Hackman238> f00fSteR: at the cli try the zenpack again
[31-May-2011 12:10:34] <f00fSteR> w00t
[31-May-2011 12:10:36] <f00fSteR> it worked
[31-May-2011 12:10:45] <Hackman238> f00fSteR: alrighty, install each of the packs
[31-May-2011 12:11:20] <Hackman238> f00fSteR: you could su root and rpm -Uvh zenoss-vore-packs.rpm --nodeps
[31-May-2011 12:11:32] <Hackman238> f00fSteR: of su zenoss; zenpack --install={each} {enter}
[31-May-2011 12:11:45] <Hackman238> f00fSteR: *or su...
[31-May-2011 12:11:56] <Hackman238> f00fSteR: While thats going, look at the gui
[31-May-2011 12:12:03] <Hackman238> f00fSteR: click device list on the left menu
[31-May-2011 12:12:08] <Hackman238> f00fSteR: Does it have your devices?
[31-May-2011 12:12:33] <f00fSteR> do i have to service zenoss stop
[31-May-2011 12:12:36] <f00fSteR> or ctl stop ?
[31-May-2011 12:12:59] <Hackman238> f00fSteR: f00fSteR no.
[31-May-2011 12:13:23] <f00fSteR> rpm -Uvh --nodeps zenoss-core-zenpacks-2.4.2.el5.x86_64.rpm --replacepkgs --replacefiles --force
[31-May-2011 12:13:23] <f00fSteR> warning: zenoss-core-zenpacks-2.4.2.el5.x86_64.rpm: Header V3 DSA signature: NOKEY, key ID aa5a1ad7
[31-May-2011 12:13:24] <f00fSteR> warning: waiting for transaction lock on /var/lib/rpm/__db.000
[31-May-2011 12:14:05] <f00fSteR> Hackman238, my devices are not listed.. i need to change the ip addresses i think
[31-May-2011 12:14:15] <Hackman238> f00fSteR: cancel it if it doesnt go right away
[31-May-2011 12:14:28] <Hackman238> f00fSteR: Your devices are not listed at all?
[31-May-2011 12:14:39] <f00fSteR> they are no events or anything
[31-May-2011 12:14:45] <f00fSteR> this is running internally on our servers
[31-May-2011 12:15:05] <Hackman238> f00fSteR: the events will be last, we didnt do that yet.
[31-May-2011 12:15:14] <Hackman238> f00fSteR: did rpm resume or is it trapped?
[31-May-2011 12:15:16] <f00fSteR> i will have to open the port eventually and then change the ip addresses to the web addresses as opposed to the 10.* addresses listed on here... everything is green and no events
[31-May-2011 12:16:29] <Hackman238> f00fSteR: thats not what I mean. I'm trying to figure if the database wasnt overwritten. On the device list, is there more than one device?
[31-May-2011 12:16:48] <f00fSteR> yeah there are 13 servers
[31-May-2011 12:16:49] <f00fSteR> thats a;;
[31-May-2011 12:16:51] <f00fSteR> umm
[31-May-2011 12:16:53] <f00fSteR> it's installing
[31-May-2011 12:16:54] <Hackman238> f00fSteR: okay good
[31-May-2011 12:17:00] <f00fSteR> it found device issues
[31-May-2011 12:17:03] <f00fSteR> while it;s installing
[31-May-2011 12:17:08] <f00fSteR> madd daemons starring
[31-May-2011 12:17:09] <Hackman238> f00fSteR: alrighty
[31-May-2011 12:17:12] <f00fSteR> in good will hunting
[31-May-2011 12:17:28] <Hackman238> f00fSteR: once its done installing, su zenoss then zenoss status
[31-May-2011 12:17:36] <Hackman238> Tell me which daemons are stopped, if any
[31-May-2011 12:18:06] <f00fSteR>    
[31-May-2011 12:18:06] <f00fSteR>    
[31-May-2011 12:18:06] <f00fSteR> ip is down
[31-May-2011 12:18:06] <f00fSteR>    
[31-May-2011 12:18:13] <f00fSteR> 2011/05/31 06:13:53.000
[31-May-2011 12:18:14] <f00fSteR>    
[31-May-2011 12:18:16] <f00fSteR> 2011/05/31 06:14:47.000
[31-May-2011 12:18:18] <f00fSteR>    
[31-May-2011 12:18:20] <f00fSteR> 2
[31-May-2011 12:18:22] <f00fSteR>    
[31-May-2011 12:18:24] <Hackman238> f00fSteR:  thats all ok
[31-May-2011 12:18:24] <f00fSteR>    
[31-May-2011 12:18:27] <f00fSteR> process
[31-May-2011 12:18:30] <f00fSteR>    
[31-May-2011 12:18:32] <f00fSteR>    
[31-May-2011 12:18:35] <f00fSteR> Unable to read processes on device app3   
[31-May-2011 12:18:36] <f00fSteR>    
[31-May-2011 12:18:37] <Hackman238> f00fSteR:  if ports arent open that doesnt matter, we'll fix later
[31-May-2011 12:18:38] <f00fSteR> ip is down
[31-May-2011 12:18:40] <f00fSteR>    
[31-May-2011 12:18:42] <f00fSteR> 2011/05/31 06:13:53.000
[31-May-2011 12:18:45] <f00fSteR>    
[31-May-2011 12:18:47] <f00fSteR> 2011/05/31 06:14:47.000
[31-May-2011 12:18:49] <f00fSteR>    
[31-May-2011 12:18:51] <f00fSteR> 2
[31-May-2011 12:18:52] <Hackman238> f00fSteR: Thats all output from su zenoss; zenoss status?
[31-May-2011 12:18:53] <f00fSteR>    
[31-May-2011 12:18:55] <f00fSteR>    
[31-May-2011 12:18:57] <f00fSteR> process
[31-May-2011 12:19:01] <f00fSteR>    
[31-May-2011 12:19:03] <f00fSteR>    
[31-May-2011 12:19:05] <f00fSteR> Unable to read processes on device app3
[31-May-2011 12:19:07] <f00fSteR> whoa
[31-May-2011 12:19:09] <f00fSteR> sorry!
[31-May-2011 12:19:11] <f00fSteR> it looked like 2 lines
[31-May-2011 12:19:13] <f00fSteR> Daemon: zenjmx not running
[31-May-2011 12:19:15] <f00fSteR> no
[31-May-2011 12:19:17] <f00fSteR> output is on private window
[31-May-2011 12:19:29] <Hackman238> f00fSteR: okay, only zenjmx isnt running?
[31-May-2011 12:20:20] <Hackman238> f00fSteR: Are you doign jmx tests?
[31-May-2011 12:22:38] <Hackman238> f00fSteR: If no, ignore zenjmx
[31-May-2011 12:23:05] <Hackman238> f00fSteR: Next, on source server, mysqldump -u zenoss -p 'zenoss' events > events.sql
[31-May-2011 12:23:20] <Hackman238> f00fSteR: copy events.sql to new box
[31-May-2011 12:23:33] <Hackman238> f00fSteR: and dump it into your events table
[31-May-2011 12:23:54] <Hackman238> f00fSteR: All you'll need to do to finish is make sure all ports, etc are open for your tests.
[31-May-2011 12:25:38] <dhopp> ok…apparently I don't know the dmd as well as I thought...
[31-May-2011 12:25:50] <dhopp> >>> dmd.Devices.createOrganizer('/Clients')
[31-May-2011 12:25:50] <dhopp> <DeviceClass at /zport/dmd/Devices/Clients>
[31-May-2011 12:25:56] <dhopp> but that doesn't show up in the interface
[31-May-2011 12:26:12] <dhopp> I did a sync and commit
[31-May-2011 12:27:23] <Hackman238> dhopp: should have
[31-May-2011 12:27:34] <Hackman238> dhopp: goto devices root in the web ui
[31-May-2011 12:27:40] <Hackman238> dhopp: press f5
[31-May-2011 12:28:14] <dhopp> Hackman238: I tried that…Under Device Classes I still only see the defaults
[31-May-2011 12:28:42] <Hackman238> dhopp: my bad, take off the /
[31-May-2011 12:29:04] <Hackman238> dhopp: dmd.Devices.createOrganizer('Clients')
[31-May-2011 12:29:08] <Hackman238> dhopp: then commit()
[31-May-2011 12:29:14] <dhopp> >>> dmd.Devices.createOrganizer('Clients')
[31-May-2011 12:29:14] <dhopp> <DeviceClass at /zport/dmd/Devices/Clients>
[31-May-2011 12:29:19] <dhopp> looks the same
[31-May-2011 12:29:30] <Hackman238> dhopp: Yep. commit it and refresh your screen.
[31-May-2011 12:29:35] <Hackman238> dhopp: what version?
[31-May-2011 12:29:40] <dhopp> 3.1.0
[31-May-2011 12:29:53] <Hackman238> dhopp: Should be working.
[31-May-2011 12:30:09] <Hackman238> dhopp: ctrl + d your dmd session
[31-May-2011 12:30:23] <Hackman238> dhopp: su zenoss; zopectl restart
[31-May-2011 12:30:53] <Hackman238> dhopp: then refresh the page. note with v3.x you have to deal with java script funness and browser caching.
[31-May-2011 12:31:09] <Hackman238> dhopp: you might want to clear your cache if things dont show up automagically
[31-May-2011 12:31:12] <dhopp> Hackman238:  I tried that too (zopectl restart)
[31-May-2011 12:31:31] <dhopp> Hackman238: I tried a different browser
[31-May-2011 12:31:38] <dhopp> Hackman238: still doesn't show up
[31-May-2011 12:32:06] <Hackman238> dhopp: hum. broke install sounds like.
[31-May-2011 12:32:16] <Hackman238> dhopp: Just tested it right here LOL
[31-May-2011 12:32:45] <Hackman238> dhopp: dmd.Devices.Server.createOrganizer('Clients')
[31-May-2011 12:32:49] <Hackman238> dhopp: then commit()
[31-May-2011 12:33:02] <Hackman238> then refresh and look under /Devices/Server
[31-May-2011 12:34:03] <f00fSteR> Hackman238, finding out the old zenoss password from the old install... which config file ?
[31-May-2011 12:34:21] <dhopp> Hackman238:  this doesn't look right...
[31-May-2011 12:34:21] <Hackman238> f00fSteR: zenoss user for the zenoss webui?
[31-May-2011 12:34:22] <dhopp> >>> dmd.Devices.Server.createOrganizer('Clients')
[31-May-2011 12:34:22] <dhopp> <DeviceClass at /zport/dmd/Devices/Clients>
[31-May-2011 12:34:42] <f00fSteR> Hackman238, the config file that has the zenoss db password in it
[31-May-2011 12:34:56] <Hackman238> f00fSteR: which db? zodb or mysql?
[31-May-2011 12:34:56] <f00fSteR> so i can grep the zenoss db user password
[31-May-2011 12:34:59] <f00fSteR> mysql
[31-May-2011 12:35:08] <f00fSteR> on the old system
[31-May-2011 12:35:09] <Hackman238> f00fSteR: we set it to zenoss
[31-May-2011 12:35:13] <Hackman238> f00fSteR: Oh
[31-May-2011 12:35:17] <Hackman238> f00fSteR: hum
[31-May-2011 12:35:28] <Hackman238> f00fSteR: Its in zope, you cant grep it
[31-May-2011 12:35:34] <dhopp> Although…I see Clients under the main Device tree now
[31-May-2011 12:35:45] <Hackman238> Use TAB-TAB to see a list of zendmd related commands.
[31-May-2011 12:35:45] <Hackman238> Tab completion also works for objects -- hit tab after an object name and '.' (eg dmd. + tab-key).
[31-May-2011 12:35:47] <dhopp> (not under Server)
[31-May-2011 12:35:48] <Hackman238> >>> dmd.Devices.Server.createOrganizer('Clients')
[31-May-2011 12:35:51] <Hackman238> <DeviceClass at /zport/dmd/Devices/Clients>
[31-May-2011 12:35:53] <Hackman238> >>>
[31-May-2011 12:36:11] <Hackman238> dhopp: should be working. you'll see it under Server's ajax and js caching funness
[31-May-2011 12:36:21] <dhopp> Hackman238: ok
[31-May-2011 12:36:33] <Hackman238> dhopp: Personally I hate ajax and js
[31-May-2011 12:36:52] <Hackman238> f00fSteR: Let me think about that
[31-May-2011 12:37:06] <Hackman238> f00fSteR: your almost better off trying to get mysql to tell you what its set to
[31-May-2011 12:38:03] <f00fSteR> lol
[31-May-2011 12:38:05] <f00fSteR> lets see
[31-May-2011 12:38:14] <f00fSteR> >>> dmd.Devices.Server.createOrganizer('Clients')
[31-May-2011 12:38:14] <f00fSteR> <DeviceClass at /zport/dmd/Devices/Clients>
[31-May-2011 12:38:35] <Hackman238> f00fSteR: dont forget to commit()
[31-May-2011 12:38:47] <Hackman238> f00fSteR: you could do this:
[31-May-2011 12:38:59] <Hackman238> f00fSteR: but it WILL interrupt your production box
[31-May-2011 12:39:10] <f00fSteR> cant have that
[31-May-2011 12:39:11] <f00fSteR> sigh
[31-May-2011 12:39:31] <Hackman238> f00fSteR: no idea then. Tell boss to keep important passwords safe next time
[31-May-2011 12:39:35] <Hackman238> f00fSteR: :-)
[31-May-2011 12:39:50] * f00fSteR goes to flush mysql root password.
[31-May-2011 12:40:01] <f00fSteR> noone uses that in a production box correct ?
[31-May-2011 12:40:12] <f00fSteR> fuck... sys admin the most paradoxical jobs on the planet
[31-May-2011 12:40:36] <Hackman238> f00fSteR: what??
[31-May-2011 12:40:54] <Hackman238> f00fSteR: IDK- depends on how it is used
[31-May-2011 12:41:20] <Hackman238> f00fSteR: dont change the production zenoss mysql password- it'll break zenhub
[31-May-2011 12:41:47] <dhopp> Hackman238: I'm an idiot
[31-May-2011 12:41:48] <f00fSteR> yeah
[31-May-2011 12:41:49] <f00fSteR> hrmm
[31-May-2011 12:41:57] <Hackman238> dhopp: Sorry to hear that
[31-May-2011 12:42:11] <dhopp> Hackman238: I looked at dmd.Devices.Server and was thinking that meant we were working on Devices/Server
[31-May-2011 12:42:21] <dhopp> The organizer Devices/Server
[31-May-2011 12:42:26] <dhopp> that's why I was confused
[31-May-2011 12:42:29] <Hackman238> dhopp: gotcha
[31-May-2011 12:42:58] <Hackman238> f00fSteR: my suggestion is this:
[31-May-2011 12:43:08] <Hackman238> f00fSteR: get the new box monitoring all the old stuff
[31-May-2011 12:43:20] <dhopp> Hackman238: Although…what is dmd.Devices.createOrganizer then?  If I have to call Devices.Server.createOrganizer?
[31-May-2011 12:43:31] <Hackman238> f00fSteR: then cut the old one off and reset the mysql password, dump the events and them move the evetns to the new box.
[31-May-2011 12:44:03] <Hackman238> dhopp: dmd.Devices.createOrganizer() is a method on 'DeviceClasses'
[31-May-2011 12:44:17] <Hackman238> dhopp: do it'll be callable at any leven of the device classes.
[31-May-2011 12:44:40] <Hackman238> dmd.Devices.createOrganizer('any') will create in /Devices/
[31-May-2011 12:44:50] <Hackman238> dmd.Devices.Server.createOrganizer('any') will create in /Devices/Server/
[31-May-2011 12:44:59] <Hackman238> dmd.Devices.Server.Linux.createOrganizer('any') will create in /Devices/Server/Linux/
[31-May-2011 12:45:08] <dhopp> Hackman238: hrm...
[31-May-2011 12:45:24] <f00fSteR> Hackman238, thats an awesome idea... now where do i get started if i want to change the ip addresses manually ?
[31-May-2011 12:45:31] <f00fSteR> admin huide ?
[31-May-2011 12:45:34] <f00fSteR> guide*
[31-May-2011 12:45:51] <Hackman238> f00fSteR: change the devices your going to monitors IPs?
[31-May-2011 12:46:05] <f00fSteR> yeah
[31-May-2011 12:46:16] <Hackman238> f00fSteR: this new box is not on the samer segment as the old one?
[31-May-2011 12:46:22] <f00fSteR> it is not
[31-May-2011 12:46:26] <f00fSteR> interbally
[31-May-2011 12:46:33] <Hackman238> f00fSteR: open each device then pick the drop down -> reset ip address
[31-May-2011 12:46:53] <dhopp> Hackman238: wait..that means my original thinking was right…I must be just getting confused with browser cache
[31-May-2011 12:47:12] <Hackman238> dhopp: I must misunderstand your misunderstanding
[31-May-2011 12:47:33] <Hackman238> dhopp: LOL. Anyways, you call the method on the parent to create a child.
[31-May-2011 12:47:58] <dhopp> Hackman238: dmd.Devices.createOrganizer('Clients') did nothing for me..regardless of browser
[31-May-2011 12:48:07] <Hackman238> dhopp: to create a child of a child, the first child is the parent of the second child, so it would be childA.method(childB)
[31-May-2011 12:48:34] <Hackman238> dhopp: Thats odd- it should.
[31-May-2011 12:48:35] <dhopp> Hackman238:  dmd.Devices.Server.createOrganizer('Clients') put a Clients orgainizer under the main Devices tree NOT under Server...
[31-May-2011 12:48:48] <dhopp> Hackman238: but I think what I was seeing was the cache finally timing out
[31-May-2011 12:49:26] <Hackman238> dmd.Devices.Server.createOrganizer('Clients') should end up /Devices/Server/Client
[31-May-2011 12:49:34] <dhopp> Hackman238: I'm going to try it on another installation...
[31-May-2011 12:49:41] <dhopp> Hackman238: I'm clearly confused
[31-May-2011 12:49:52] <Hackman238> dhopp: Good idea- I think that ones borked
[31-May-2011 12:50:34] <dhopp> Hackman238: it shouldn't be borked…I litterally just installed the rpms!
[31-May-2011 12:50:45] <dhopp> Hackman238: I think it's a matter of ajax/js fucking with me
[31-May-2011 12:51:18] <dhopp> Hackman238: and things happening delayed…i.e. when I did Devices.Server, it just HAPPENED that the other one showed up
[31-May-2011 12:51:34] <Hackman238> dhopp: Oh ok
[31-May-2011 12:52:11] <Hackman238> dhopp: if you want to look for orgs, can do dmd.Devices.getOrganizerNames()
[31-May-2011 12:52:39] <Hackman238> dhopp: for servers, dmd.Devices.Server.getOrganizerNames()
[31-May-2011 12:56:47] <Hackman238> dhopp: yeah ajax will do that to you.
[31-May-2011 12:57:28] <dhopp> ok
[31-May-2011 12:57:31] <dhopp> Hackman238:
[31-May-2011 12:57:39] <dhopp> Hackman238: this doesn't make any sense
[31-May-2011 12:58:14] <dhopp> Hackman238: different installation … dmd.Devices.Server.createOrganizer('TestClient') shows up under man Device tree
[31-May-2011 12:58:18] <dhopp> *main
[31-May-2011 12:59:23] <f00fSteR> Hackman238, I've been connecting them one by one
[31-May-2011 12:59:33] <f00fSteR> Hackman238, seems like we're getting somewhere
[31-May-2011 12:59:44] <dhopp> >>> dmd.Devices.Server.createOrganizer('TestClient')
[31-May-2011 12:59:44] <dhopp> <DeviceClass at /zport/dmd/Devices/TestClient> <—————— under main tree
[31-May-2011 12:59:56] <dhopp> >>> dmd.Devices.Server.createOrganizer('Server/TestClient2')
[31-May-2011 12:59:56] <dhopp> <DeviceClass at /zport/dmd/Devices/Server/TestClient2> <————— under Server tree
[31-May-2011 13:00:10] <dhopp> and it shows up in the UI that way instantly if I don't do a sync() before commit()
[31-May-2011 13:02:31] <Hackman238> dhopp: Thats odd.
[31-May-2011 13:03:35] <dhopp> Hackman238: going to revert my VM to a pre-zenoss stat and reinstall
[31-May-2011 13:03:42] <dhopp> *state
[31-May-2011 13:03:52] <Hackman238> >>> dmd.Devices.Server.createOrganizer('test')
[31-May-2011 13:04:03] <Hackman238> <DeviceClass at /zport/dmd/Devices/Server/test>
[31-May-2011 13:04:31] <Hackman238> dhopp: Let me try on v3...I'm testing with 2.5.
[31-May-2011 13:04:49] <Hackman238> dhopp: I wouldnt expect this to change though.
[31-May-2011 13:05:09] <Hackman238> >>> dmd.Devices.Server.createOrganizer('Server/test')
[31-May-2011 13:05:09] <Hackman238> <DeviceClass at /zport/dmd/Devices/Server/test>
[31-May-2011 13:05:17] <Hackman238> Also works. Is a valuid method.
[31-May-2011 13:05:38] <dhopp> Hackman238: I would think it should work the same (and how your's is working is how I would expect it to)…
[31-May-2011 13:05:51] <Hackman238> dhopp: Oh wow, in the beta the method doesnt refer to the parent who calls it.
[31-May-2011 13:06:03] <Hackman238> dhopp: Must be in v3 its broken too.
[31-May-2011 13:06:26] <dhopp> Hackman238: wouldn't that completely screw the UI though?
[31-May-2011 13:06:33] <dhopp> Hackman238: the UI seems to do it fine
[31-May-2011 13:06:36] <Hackman238> dhopp: In that case just use the full path like before
[31-May-2011 13:07:20] <dhopp> Hackman238: ah…UI is probably using full paths anyway which is why it works
[31-May-2011 13:07:24] <Hackman238> dhopp: No, they probably dont call the method that way in the newer gui
[31-May-2011 13:07:32] <dhopp> Hackman238:  oh
[31-May-2011 13:08:03] <Hackman238> dhopp: by that way I mean refering to parents, but instead using the full path as you just stated
[31-May-2011 13:08:49] <rocket> dhopp: the json api is the only real supported way of doing this stuff by the way .. *of course its not feature complete yet*
[31-May-2011 13:09:19] <rocket> dhopp: we are starting to change more and more of the backend and that is the solution provided so that customers dont have to rework scripts
[31-May-2011 13:09:29] <dhopp> Hackman238: ok…if I do a sync() before commit() changes don't show up in the UI right away…sync should be before commit right?
[31-May-2011 13:09:36] <rocket> dhopp: doing things from zendmd may change from release to release
[31-May-2011 13:09:44] <Hackman238> dhopp: generally I would
[31-May-2011 13:09:54] <rocket> dhopp: you dont need to sync right before a commit
[31-May-2011 13:10:02] <rocket> a sync throws away changes so
[31-May-2011 13:10:10] <rocket> sync -> do changes -> commit()
[31-May-2011 13:10:12] <Hackman238> dhopp: its generally after an operation runs for a while though
[31-May-2011 13:10:37] <rocket> a commit also includes a builtin sync so you dont need to sync right after a commit
[31-May-2011 13:11:20] <dhopp> rocket: I know that…and I would prefer a gui way (at some point will make a zenpack for it) but I need a quick way to setup a device tree for my clients…we have clients that will have 4 or 5 linux servers, 2 or 3 windows servers, 2 or 3 switches and a firewall…some zproperties will be custom per client and using the default device tree would be madness
[31-May-2011 13:11:46] <dhopp> rocket: so I was going to write a "Add Client" script that sets up the device tree with the appropriate templates and modelers
[31-May-2011 13:11:55] <rocket> dhopp: zenbatchload may work for you
[31-May-2011 13:12:07] <rocket> it will create the organizers if they dont exist for the device etc ..
[31-May-2011 13:12:14] <Hackman238> dhopp: I'd reccomend using chef, matt rays zenoss cookbook and zenbatchloader
[31-May-2011 13:12:28] <rocket> but it wont setup the templates etc...
[31-May-2011 13:12:51] <mattray> rocket: it won't create new ones, but it can bind to existing ones
[31-May-2011 13:13:09] <Hackman238> dhopp: That way you can create custom roles for customersw ho need differnt stuff and you can sumbit dmd code using chef straight to zenoss to make the changes as chef runs on clients
[31-May-2011 13:13:15] <dhopp> Hackman238: that may be the end result…but I had a hard enough time getting them to buy into replacing the existing monitoring system…getting them to replace the configuration/deployment system will probably be hell
[31-May-2011 13:13:29] <rocket> mattray: I think 3.1.1 and Avalon will now make them
[31-May-2011 13:13:46] <Hackman238> dhopp: 'I've had good results with ether'
[31-May-2011 13:13:47] <rocket> mattray: there have been some additional features added for QA etc ..
[31-May-2011 13:13:52] <mattray> cool
[31-May-2011 13:14:01] <mattray> I'll read the Release Notes attentively
[31-May-2011 13:14:20] <rocket> mattray: assuming they get release noted ..
[31-May-2011 13:14:27] <Hackman238> rocket: create new organizers?
[31-May-2011 13:15:11] <rocket> Hackman238: zenbatchloader will create new organizers if they dont exist when specifying them in the config file if I recall correctly
[31-May-2011 13:15:16] <dhopp> Hackman238: you've had good results with what?
[31-May-2011 13:15:34] <Hackman238> rocket: I was going to say in v3 it creates if they dont exist
[31-May-2011 13:16:43] <Hackman238> dhopp: With chef and puppet? Yes. Ether? Was a Archer joke.
[31-May-2011 13:19:15] <Hackman238> mattray: Also I didnt forget about getting you info on the zenoss cloud project utilizing chef, openstack and zenoss. Just been madness work wise lately.
[31-May-2011 13:19:42] <mattray> Hackman238: understood, my schedule is similarly hectic.
[31-May-2011 13:19:54] <rocket> Hackman238: good story! .. you ignored him ..
[31-May-2011 13:21:17] <Hackman238> rocket: not at all, its just been busy since theres so many simultanious projects. I'm out of threads and we're context switching hard- cache misses galore.
[31-May-2011 13:22:16] <Hackman238> Like a P4 trying to do 'real' work. Dump that pipeline after starting and switch context only to have to redo the previously partially completed work.
[31-May-2011 13:22:59] <dhopp> Hackman238: you are outdated..time for a new model
[31-May-2011 13:23:07] <Hackman238> dhopp: Indeed.
[31-May-2011 13:23:32] <Sam-I-Am> lol
[31-May-2011 13:23:58] <Sam-I-Am> elevendyseven stages ought to be good for anything
[31-May-2011 13:24:30] <Sam-I-Am> then marketing can crank up your clock speed and invest in energy companies
[31-May-2011 13:25:39] <Hackman238> LOL - extra supply power and extra cooling - double win!
[31-May-2011 13:26:39] <Hackman238> f00fSteR: Let me know if you need any more help.
[31-May-2011 13:28:27] <dhopp> one problem I see with zenbatchload, it looks like it either A) expects device1 to be DNS resolvable or B) a management IP address…problem with A is in our environment and how we monitor the devices through the firewalls this might not always be the case (I'm trying to change that) and the problem with B is then I have a boat load of devices in Zenoss that just have the IP address and have to change the name manually
[31-May-2011 13:33:15] * Sam-I-Am is in diagram hell today
[31-May-2011 13:33:46] <dhopp> Sam-I-Am: just reminded me I have to do a visio diagram
[31-May-2011 13:33:53] <dhopp> (and no that's not a good thing)
[31-May-2011 13:34:00] <Hackman238> dhopp: BAH! No visio! use Dia
[31-May-2011 13:34:01] <dhopp> I would rather not doing the diagram
[31-May-2011 13:34:26] <dhopp> Hackman238: Yeah…I can't though….it's how we do our diagrams
[31-May-2011 13:34:49] <Hackman238> dhopp: madnesss!
[31-May-2011 13:34:55] <dhopp> Hackman238: agreed
[31-May-2011 13:36:04] <dhopp> Hackman238: we also have a bunch of spreadsheets that keep track of ip addresses for different clients…I REALLY want that in a database..but have to take this in baby steps
[31-May-2011 13:38:06] <Hackman238> dhopp: Oh my
[31-May-2011 13:38:10] <dhopp> I guess for a while there was a joke about how some guy (who is no longer with the company) got $1 everytime he made us use Excel and people were giving him grief how the IP stuff wasn't in a database…so he named the spreadsheet "IP Database"
[31-May-2011 13:38:20] <dhopp> which if funny..but extremely annoying
[31-May-2011 13:38:23] <dhopp> *is
[31-May-2011 13:39:45] <dhopp> Everytime I hear "it's in the buildsheet" I want to scream
[31-May-2011 13:40:04] <Sam-I-Am> dhopp: this is omnigraffle... lots more tolerable
[31-May-2011 13:40:45] <dhopp> Sam-I-Am: luckily I basically just have to take a template and modify it as it's not that bad
[31-May-2011 13:45:16] <Hackman238> LOL
[31-May-2011 13:46:20] frozty_sa is now known as froztbyte
[31-May-2011 13:47:17] <uifjlh> I'm working on the same type of "newtork map" with Visio.  I"ve hooked some of my drawings to an Access dB, but I really want to have a SQL space & keep all the device information in one place.  Too Bad, there isn't a SharePoint plugin for that!
[31-May-2011 13:47:53] <Hackman238> uifjlh: Oh my
[31-May-2011 13:47:53] <jmp242> Why not use a network mapping program?
[31-May-2011 13:49:26] <squig> is there a good free network mapping program? jmp242
[31-May-2011 13:49:38] <jmp242> mmmm
[31-May-2011 13:49:49] <jmp242> I use NetDisco, which may work ok
[31-May-2011 13:50:02] <jmp242> I've also used LanTopolog which is ~$10 or so
[31-May-2011 13:50:14] <squig> i know netdisco
[31-May-2011 13:50:16] <jmp242> I suppose it depends on your switches mostly
[31-May-2011 13:50:29] <dhopp> hrm…looking at zenbatchload more closely..I could probably script around that to do what I want…
[31-May-2011 13:50:36] <squig> sadly not all my switches talk any of the network discovery protocols
[31-May-2011 13:50:42] <jmp242> mmm
[31-May-2011 13:50:50] <jmp242> Well, yes, that makes it a pain
[31-May-2011 13:51:02] <squig> i think i can look at the tables on each of the switches and build my own graph
[31-May-2011 13:51:04] <jmp242> LanTopolog seems to work better in that environment using a MAC address matching
[31-May-2011 13:52:07] <squig> i have to write a nice node graph program for some thing else, maybe i just write one
[31-May-2011 13:52:45] <jmp242>
[31-May-2011 13:53:12] <squig> aah then I would need windows, we are happily linux only
[31-May-2011 13:53:23] <uifjlh> netdisco looks very nice!
[31-May-2011 13:54:04] <squig> i just ran into this looking for the netdisco webpage
[31-May-2011 13:55:10] <jmp242> Yea
[31-May-2011 13:55:16] <jmp242> I know about that - it seems in perpetual alpha
[31-May-2011 13:55:26] <jmp242> I like NetDisco, when it works
[31-May-2011 13:55:36] <jmp242> but if you don't have all switches that do LLDP, you're a bit screwed
[31-May-2011 14:00:50] <mattray> squig: interesting
[31-May-2011 14:11:29] <froztbyte> Preparing to replace zenoss-stack 2.5.1-0 (using zenoss-stack_2.5.2_i386.deb) ...
[31-May-2011 14:11:32] <froztbyte> Preparing to upgrade zenoss.
[31-May-2011 14:11:36] <froztbyte> this might have an entertaining end..
[31-May-2011 14:11:48] <Hackman238> froztbyte: unemployment?
[31-May-2011 14:12:23] <froztbyte> Hackman238: you seem to work in a ridiculously high-stress environment if you worry about being fired that often
[31-May-2011 14:13:04] <Hackman238> froztbyte: Na, RackSpace is really comfortable. I just try to plan ahead. No worries :-)
[31-May-2011 14:13:32] <froztbyte> ah, frustrating, not high-stress
[31-May-2011 14:13:35] <froztbyte> which team are you on?
[31-May-2011 14:13:41] <froztbyte> and what office, I suppose
[31-May-2011 14:22:19] <Hackman238> froztbyte: NOC Systems
[31-May-2011 14:22:26] <Hackman238> froztbyte: Datapoint
[31-May-2011 14:25:19] <Sam-I-Am> my place likes to make people feel uncomfortable
[31-May-2011 14:25:22] <Sam-I-Am> its pretty awesome
[31-May-2011 14:25:46] <Sam-I-Am> 10% is getting the work done, 90% is making sure someones not trying to set you up for failure
[31-May-2011 14:25:53] <Sam-I-Am> or throw you under a bus
[31-May-2011 14:26:32] <Hackman238> Sam-I-Am: Wow- sorry to ehar that
[31-May-2011 14:27:05] <jmp242> Academia
[31-May-2011 14:27:11] <Hackman238> froztbyte: Admittadly I over think and over plan since I'm also a business owner LOL
[31-May-2011 14:27:11] <jmp242> little you can do to get fired
[31-May-2011 14:27:36] <Hackman238> jmp242: Agreed- a win
[31-May-2011 14:27:48] <Sam-I-Am> jmp242: this is semi-academia
[31-May-2011 14:27:54] <Hackman238> jmp242: Plus nice environment and sometimes education bennifits
[31-May-2011 14:27:58] <jmp242> Yup
[31-May-2011 14:28:02] <Sam-I-Am> it has all the politics of academia with all the politics of a small business
[31-May-2011 14:28:16] <jmp242> I'm debating whether to take on the stress of trying to get a masters here at Cornell
[31-May-2011 14:28:18] <Sam-I-Am> and other small business crap
[31-May-2011 14:28:41] <Sam-I-Am> Hackman238: i about gave up on education... worked in it for almost 10 years
[31-May-2011 14:28:44] <jmp242> Problem is in what - they don't seem to offer what I'd like to do, but I don't want to move /change employment to go somewhere else
[31-May-2011 14:29:03] <Sam-I-Am> theres never enough money to do anything and management has no clue
[31-May-2011 14:29:06] <Hackman238> jmp242: Thats a ton of extra stress depending on the area
[31-May-2011 14:29:14] <jmp242> yea
[31-May-2011 14:29:16] <Hackman238> Sam-I-Am: Public school???
[31-May-2011 14:29:17] <Sam-I-Am> so if you want to sit around and be bored... but have a stable job... education is good
[31-May-2011 14:29:23] <Sam-I-Am> Hackman238: higher ed
[31-May-2011 14:29:29] <jmp242> but I figure eventually I need a masters to get into management ever
[31-May-2011 14:29:41] <jmp242> So maybe the MBA track
[31-May-2011 14:29:47] <Hackman238> Sam-I-Am: ah
[31-May-2011 14:29:47] <Sam-I-Am> i'm sorta aiming for a medium-size company
[31-May-2011 14:29:59] <Sam-I-Am> not so big nothing ever happens (e.g., HP or IBM)
[31-May-2011 14:30:01] <Hackman238> jmp242: An ideal area
[31-May-2011 14:30:11] <jmp242> I'm in a research lab in Cornell
[31-May-2011 14:30:11] <Hackman238> Sam-I-Am: RackSpace or google
[31-May-2011 14:30:16] <jmp242> we're almost never bored
[31-May-2011 14:30:19] <Hackman238> jmp242: I'm envious of that
[31-May-2011 14:30:30] <jmp242> and when when we are
[31-May-2011 14:30:33] <Hackman238> jmp242: I love physics and math
[31-May-2011 14:30:39] <Sam-I-Am> Hackman238: i tried google... their interview process is really unorganized and otherwise not fun
[31-May-2011 14:30:39] <jmp242> we have a party or BBQ or something
[31-May-2011 14:30:49] <jmp242> Watch daily show lol
[31-May-2011 14:30:54] <Hackman238> jmp242: Nice!
[31-May-2011 14:31:05] <Sam-I-Am> Hackman238: i also dont want to work 24 hours/day
[31-May-2011 14:31:08] <jmp242> But lots happens - I can pick out some new toy, get it in, and see if it helps
[31-May-2011 14:31:12] <jmp242> or not
[31-May-2011 14:31:15] <Sam-I-Am> as nice as their facilities are, i have other interests
[31-May-2011 14:31:50] <Sam-I-Am> i was a workaholic for several years and look back wondering what all i missed... so not doing that again
[31-May-2011 14:32:07] <jmp242> Yea, I used to work here at the lab and at best buy for a while
[31-May-2011 14:32:13] <jmp242> to get extra money from BB
[31-May-2011 14:32:22] <jmp242> glad I stopped doing that - never a day off!
[31-May-2011 14:34:18] <dhopp> rocket: you around?
[31-May-2011 14:34:47] <rocket> nope
[31-May-2011 14:34:52] <Sam-I-Am> heh
[31-May-2011 14:34:54] <dhopp> rocket:  damn
[31-May-2011 14:34:55] <dhopp> heh
[31-May-2011 14:34:56] <Sam-I-Am> he lurks
[31-May-2011 14:35:11] <dhopp> rocket: did you say that zenbatchload does or does not create device classes if they don't exist?
[31-May-2011 14:35:31] <rocket> I *think* it does ... if I recall correctly
[31-May-2011 14:36:20] <dhopp> rocket: thanks…I'll try it out….I looked at the config options more closely and I can probably script around it…i.e. write a script that generates the config for zenbatchload
[31-May-2011 14:37:19] <Sam-I-Am> the docs for zenbatchload are a little thin :/
[31-May-2011 14:37:42] <dhopp> Sam-I-Am:  yeah..but zenbatchload —sample_configs is good enough for me to figure it out I think
[31-May-2011 14:41:25] <Sam-I-Am> yeah thats not too bad
[31-May-2011 14:41:35] <Sam-I-Am> for some reason it took me a while to find that after looking through the normal docs
[31-May-2011 14:42:10] <Hackman238> dhopp: It should. yes
[31-May-2011 15:18:25] <f00fSteR> Hackman238, wb man
[31-May-2011 15:19:04] <f00fSteR> so how do i connect to my servers when they're not open with a public ip address for zenoss which is on a different network... any work around you guys can think of ?
[31-May-2011 15:19:12] <Hackman238> f00fSteR: wb?
[31-May-2011 15:19:18] <f00fSteR> welcome back
[31-May-2011 15:19:39] <Hackman238> f00fSteR: Oh LOL.
[31-May-2011 15:20:12] <Sam-I-Am> f00fSteR: remote collectors or some firewall magic
[31-May-2011 15:20:54] <Hackman238> rocket: Whats the supported way of adding actions to the OS/components gear menu on the device status page?
[31-May-2011 15:21:10] <Hackman238> f00fSteR: What Sam-I-Am said
[31-May-2011 15:22:01] <f00fSteR> Hackman238, i know the firewall route... management wont let me open ports on all our systems... we dont even have sshd running on those machines... or is it fairly safe ?
[31-May-2011 15:22:10] <f00fSteR> explain remote collectors to me please ?
[31-May-2011 15:23:28] <Hackman238> f00fSteR: remote collectors are seperate machiens on the remote segment which run the perf daemons to collect from its local devices. It communicated with the main zenoss instance, or the master, via a handful of ports
[31-May-2011 15:23:58] <f00fSteR> ahh
[31-May-2011 15:24:01] <f00fSteR> this we can do
[31-May-2011 15:24:27] <Hackman238> f00fSteR: I'd suggest egors distributed collectors pack
[31-May-2011 15:24:30] <f00fSteR> so set up one system to have something like prefmon running collecting all the data and then pushing it all to my zenoss here at the office ?
[31-May-2011 15:24:40] <Hackman238> f00fSteR: what version of zenoss is the newly migrates system?
[31-May-2011 15:25:03] <Hackman238> f00fSteR: If I recall right we've just moved you to 64 bit.
[31-May-2011 15:25:40] <f00fSteR> you recall correctly
[31-May-2011 15:25:43] <f00fSteR>
[31-May-2011 15:25:46] * f00fSteR mumbles
[31-May-2011 15:25:59] <Hackman238> f00fSteR: okay, first, lets backup
[31-May-2011 15:26:05] <f00fSteR> this coming from the man who said "if his head wasnt attached to his body then he's forget it too"
[31-May-2011 15:26:09] <Hackman238> tar czvf backup.tar.gz /opt/zenoss/
[31-May-2011 15:26:21] <Sam-I-Am> f00fSteR: time to go 128 bits
[31-May-2011 15:26:22] <Hackman238> f00fSteR: Its true
[31-May-2011 15:26:32] <Hackman238> f00fSteR: LOL
[31-May-2011 15:26:43] <jmp242> Hey, does Zope have any sort of IP based ACL for the web interface?
[31-May-2011 15:26:48] <Hackman238> f00fSteR: once that tar is done we have a easy backup
[31-May-2011 15:26:56] <jmp242> I don't see anything in zopectl.conf for instance
[31-May-2011 15:27:36] <Hackman238> jmp242: to my knowledge no, but you can 1) bind it to a switch port which has ACL's applied or ACL behind httpd
[31-May-2011 15:27:36] <Sam-I-Am> jmp242: no... i use apache in front of it
[31-May-2011 15:27:57] <Sam-I-Am> and i guess a firewall in front of that
[31-May-2011 15:29:29] <Hackman238> f00fSteR: Once thats done, tell me what version of zenoss it is. I think its 2.4 something
[31-May-2011 15:29:36] <Hackman238> f00fSteR: We'll upgrade it to 2.5.2
[31-May-2011 15:29:52] <Hackman238> f00fSteR: You'll definately want to do this.
[31-May-2011 15:40:38] <f00fSteR> 2.4.2
[31-May-2011 15:40:44] <f00fSteR> 2.5.2 w00t!
[31-May-2011 15:40:49] <f00fSteR> schweeet!
[31-May-2011 15:41:31] <Hackman238> f00fSteR: you ahve 2.4.2 or 2.5.2?
[31-May-2011 15:54:19] <froztbyte> rocket: looks like I still make it blow up if I backup from 2.5.2 on the one side and restore on another 2.5.2
[31-May-2011 15:55:09] <Hackman238> froztbyte: whats the problem?
[31-May-2011 15:56:55] <froztbyte> I seem to have found a surefire way to blow up zope or something
[31-May-2011 15:58:15] <rocket> froztbyte: run a zenchkrels -r -x1 before you do the backup
[31-May-2011 15:58:17] <froztbyte> it even possible at all to do zenbackup on 32bit and then restore on a 64bit? now that I think about how zodb stores stuff, it's pickled python, and I don't think that's really got any portability at all whatsovever
[31-May-2011 15:58:26] <froztbyte> rocket: mmm..okay
[31-May-2011 15:58:35] <Sam-I-Am> froztbyte: fail?
[31-May-2011 15:58:47] <Hackman238> froztbyte: sure, jsut did it for f00fSteR
[31-May-2011 15:58:52] <Sam-I-Am> i know rrdtool will crap out
[31-May-2011 15:58:56] <Sam-I-Am> er, the rrds
[31-May-2011 15:58:58] <rocket> froztbyte: we have done it ..
[31-May-2011 15:59:19] <froztbyte> I'm gonna try the zenchkrels thing now and see what's up
[31-May-2011 15:59:31] <froztbyte> will just take a while
[31-May-2011 15:59:56] <Hackman238> froztbyte: do a fresh install of the 64-bit same version on a seperate box. Install all the packs. Copy over your Data.*, restart, works
[31-May-2011 16:01:08] <froztbyte> Hackman238: so you just entirely skip using zenbackup/zenrestore?
[31-May-2011 16:01:23] <Hamzah> is it possible to "import" a objects XML file from a ZenPack into Zenoss? For some reason I can't import the whole zenpack so I'm having to do it manually..
[31-May-2011 16:01:33] <Hamzah> I don't think the zenpack is compatible with 3.x
[31-May-2011 16:01:58] <froztbyte> I'm doing pretty much the rest of those things, but s%copy Data.*%zenbackup/zenrestore%
[31-May-2011 16:02:03] <Hackman238> froztbyte: yep.
[31-May-2011 16:02:18] <Hackman238> froztbyte: Zenbackup has always been a fail.
[31-May-2011 16:02:45] <Hackman238> froztbyte: backup that install with tar, restore it with install and untar. LOL
[31-May-2011 16:03:14] <Hackman238> froztbyte: to be honest, zenbackup does work, but I've not had great luck wit it.
[31-May-2011 16:03:42] <Hackman238> froztbyte: In your case, back up the production by taring the whole zenoss dir for good measure.
[31-May-2011 16:03:48] <froztbyte> in two separate instances trying to use it now, neither have I
[31-May-2011 16:03:56] <froztbyte> Hackman238: already done, loooooooong ago
[31-May-2011 16:04:06] <Hackman238> froztbyte: oh alrighty
[31-May-2011 16:04:15] <froztbyte> I'm not gonna fuck around with my mon instances without a fallback
[31-May-2011 16:04:18] <rocket> froztbyte: 1. stop zenoss 2. copy Data.fs from $ZENHOME/var to the new system
[31-May-2011 16:04:22] <froztbyte> (since it'll be me who has to rebuild them)
[31-May-2011 16:04:26] <Hackman238> froztbyte: install the 64-bit version of zenoss of the same revision on a second box.
[31-May-2011 16:04:42] <Hackman238> froztbyte: install all the packs and make sure it works.
[31-May-2011 16:04:44] <rocket> froztbyte: make sure the destination system has the same zenpacks installed prior to replacing the Data.fs files
[31-May-2011 16:04:52] <Hackman238> froztbyte: stop it and copy the Data.fs over to the new install
[31-May-2011 16:04:56] <Hackman238> froztbyte: start it up
[31-May-2011 16:05:22] <Hackman238> froztbyte: If you need your rrd's they can be converted, but its painful. I'd reccomend not.
[31-May-2011 16:05:29] <froztbyte> may I suggest putting this in a well-labelled section of the documentation/wiki/forum (to those with edit privs)?
[31-May-2011 16:05:45] <froztbyte> Hackman238: eh, that's a mere moment's work
[31-May-2011 16:06:01] <froztbyte> Hackman238: I've already got the shell loops for migrating RRDs in my history from when I used it on cacti
[31-May-2011 16:06:17] <Hackman238> froztbyte: nice. Can share?
[31-May-2011 16:06:22] <froztbyte> just takes bloody forever to run since dumping to xml is not the fastest thing to possibly do
[31-May-2011 16:06:30] <Hackman238> froztbyte: agreed. LOL
[31-May-2011 16:06:41] <froztbyte> but, that said, thanks to you both for the advice about Data.fs
[31-May-2011 16:06:48] <froztbyte> Hackman238: sure
[31-May-2011 16:07:27] <Hackman238> froztbyte: anytime..I've dont it hundreds of times today alone...trying to test migrate to debug development versions LOL
[31-May-2011 16:07:33] <Hackman238> *done
[31-May-2011 16:08:10] <uifjlh> so does Zenoss run in native 64 bit mode ?
[31-May-2011 16:08:15] <froztbyte> Hackman238:
[31-May-2011 16:08:26] <froztbyte> Hackman238: that's for all RRDs in one folder
[31-May-2011 16:08:42] <Hackman238>  *relstorage migration do you fail me differntly every try? LOL*
[31-May-2011 16:08:46] <froztbyte> Hackman238: I modified it slightly to use find with -execdir for subdirectory stuff
[31-May-2011 16:08:50] <Hackman238> froztbyte: Nice, thanks!
[31-May-2011 16:08:59] <jmp242>
[31-May-2011 16:09:08] <jmp242> froztbyte: that touches on it
[31-May-2011 16:09:22] <jmp242> Unfortunately, I recall Jive makes it so I can't edit that particuar page
[31-May-2011 16:09:41] <jmp242> and I'm not sure there's enough difference to be worth writing up a new FAQ2 entry
[31-May-2011 16:09:54] <froztbyte> the forums are really a bit difficult to use with Jive :/
[31-May-2011 16:10:06] <jmp242> Yea, the old Wiki on TRAC didn't look as nice I guess
[31-May-2011 16:10:11] <jmp242> but it was WAY WAY WAY easier to use
[31-May-2011 16:10:12] <froztbyte> things can get quite "hidden"
[31-May-2011 16:10:25] <jmp242> I don't update near as often as I did with the old easy WIKI style FAQ entry
[31-May-2011 16:10:32] <froztbyte> Trac's wiki software is not exactly fantastic, sure
[31-May-2011 16:10:33] <jmp242> cause it's a PITA to use the Jive editor
[31-May-2011 16:10:34] <Hackman238> froztbyte: let me know how the migration goes or if you need help
[31-May-2011 16:10:42] <froztbyte> dokuwiki or something though, that's pretty cool
[31-May-2011 16:10:57] <froztbyte> Hackman238: just waiting for the zenchkrel thing to finish running then I'm poking some more
[31-May-2011 16:11:04] <jmp242> The Jive editor just is close enough to not working for me in Firefox and it breaks entirely in Opera
[31-May-2011 16:11:20] <jmp242> I complained a LOT when they went to Jive
[31-May-2011 16:11:26] <Hackman238> jmp242: I hate it when stuff doesnt work i opera
[31-May-2011 16:11:40] <jmp242> Well, it actually barely works in Firefox
[31-May-2011 16:11:48] <Hackman238> froztbyte: Alrighty. did zenchkrels -r?
[31-May-2011 16:12:02] <jmp242> I'm close to seeing if I can convince my management to let me host a Zenoss FAQ on our TWIKI server
[31-May-2011 16:12:10] <froztbyte> Hackman238: yeah, still running
[31-May-2011 16:12:12] <froztbyte> -r -x1
[31-May-2011 16:12:20] <Hackman238> froztbyte: Alrighty.
[31-May-2011 16:12:25] <jmp242> that at least doesn't randomly shift me down a line when I set a style, it doesn't randomly delete half a long page when I try and add to it etc
[31-May-2011 16:12:40] <jmp242> Not that anyone would find it though lol
[31-May-2011 16:12:54] <jmp242> many don't find the zenoss hosted one
[31-May-2011 16:13:07] <froztbyte> Hackman238: oh, the find version: find . -iname '*rrd' -exec rrdtool dump {} > {}.xml \;
[31-May-2011 16:13:57] <froztbyte> Hackman238: and then on the recovery side: find . -iname '*rrd' -exec rrdtool restore -f $(whatever string manip magic you like here) {} \;
[31-May-2011 16:14:17] <froztbyte> in the top-level dir for RRDs
[31-May-2011 16:15:14] <Hackman238> froztbyte: ty
[31-May-2011 16:15:43] <Hackman238> Anyone know about this: (1129, "Host '' is blocked because of many connection errors; unblock with 'mysqladmin flush-hosts'")
[31-May-2011 16:17:59] <rocket> Hackman238: are you doing mysql checks of the mysql database
[31-May-2011 16:18:16] <rocket> it usually happens when you have too many connections open that dont close properly
[31-May-2011 16:18:20] <Hackman238> rocket: Its a fresh install- does it while I'm trying to populate the table LOL
[31-May-2011 16:18:54] <Hackman238> rocket: happens as I try to back install packs.
[31-May-2011 16:19:04] <rocket> that I have not seen ... maybe the database is tuned too small?
[31-May-2011 16:19:11] <Hackman238> rocket: trying turning logging on and max_connect_errors=99999
[31-May-2011 16:19:46] <Hackman238> rocket: Hum. Shouldnt be- its the default db built by 3.1.7
[31-May-2011 16:20:15] <Hackman238> rocket: eradicated and reinstalled each migration attempt
[31-May-2011 16:21:45] <rocket> strange
[31-May-2011 16:22:33] <Hackman238> rocket: Very. I've just been flushing the hosts like a mad man to compenate up to now- time for a solution LOL
[31-May-2011 16:23:26] <Hackman238> rocket: Also, while I have you, whats the 'supposed' method of adding actions to the OS/Component action/gear menu in 3.X on a device status page?
[31-May-2011 16:23:47] <Hackman238> rocket: *supported
[31-May-2011 16:26:22] <rocket> Hackman238: I will have to look that up ..
[31-May-2011 16:28:13] <froztbyte> yay, that's finally finished running
[31-May-2011 16:30:02] <rocket> Hackman238: I believe the example zenpack that chet wrote up has an example of that against 2.5.2
[31-May-2011 16:30:07] <rocket> it shouldnt have changed
[31-May-2011 16:30:24] <Hackman238> rocket: Version 3, 2.5.2 is easy
[31-May-2011 16:30:32] <rocket> I am glancing to see if you can still do it from the ui
[31-May-2011 16:31:10] <Hackman238> rocket: Problem is if I manipulate the menus it can break the instance if the pack is removed.
[31-May-2011 16:31:23] <Hackman238> rocket: only happens v3+
[31-May-2011 16:32:51] <rocket> yea you need to remove the menu when you remove the pack
[31-May-2011 16:33:15] <Hackman238> rocket: Right, but when the menu is removed so is its parent
[31-May-2011 16:33:25] <Hackman238> rocket: No idea why. Seemingly no reason
[31-May-2011 16:37:25] <rocket> you need to remove the menu element that you added ... not the whole menu itself
[31-May-2011 16:37:25] <Hackman238> froztbyte: Howd that go?
[31-May-2011 16:37:58] <Hackman238> rocket: I know that, but what seems to happen is the entire menu ends up missing.
[31-May-2011 16:39:15] <rocket>         menu = getattr(dmd.Devices.FOO, 'zenMenus', None)
[31-May-2011 16:39:15] <rocket>         if menu:
[31-May-2011 16:39:16] <rocket>             manageMenu = getattr(menu, 'Manage', None)
[31-May-2011 16:39:16] <rocket>             if manageMenu:
[31-May-2011 16:39:16] <rocket>                 if getattr(manageMenu.zenMenuItems, 'addFoo', None):
[31-May-2011 16:39:18] <rocket>           'Removing old add foo menu')
[31-May-2011 16:39:20] <rocket>                     manageMenu.manage_deleteZenMenuItem('addFoo')
[31-May-2011 16:39:32] <rocket> is that how you are removing the element?
[31-May-2011 16:40:01] <Hackman238> rocket: No I was adding it to the objects.xml
[31-May-2011 16:40:11] <Hackman238> rocket: just the single entry.
[31-May-2011 16:40:45] <rocket> Hackman238: yea take a look at the zenpack example chet put out .. do it from the file to add or remove the menu
[31-May-2011 16:41:03] <rocket> changing it via the objects.xml will most likely have it broken as you discovered
[31-May-2011 16:41:36] <Hackman238> rocket: That explains it. I'll look at the example. Thanks very much!
[31-May-2011 16:41:48] <Hackman238> rocket: Sorry to be such a pest :-)
[31-May-2011 16:43:02] <rocket>
[31-May-2011 16:43:16] <rocket> see the install and remove methods in the class ZenPack
[31-May-2011 16:43:44] <Hackman238> rocket: yep.
[31-May-2011 16:44:05] <Hackman238> rocket: any other objects I shouldnt populate using objects.xml?
[31-May-2011 16:44:31] <rocket> i guess I dont know what you are populating that way ..
[31-May-2011 16:44:50] <rocket> I prefer to use the zenpack install/remove methods for most things
[31-May-2011 16:45:08] <Hackman238> rocket: The menu. If you go to editMenus and create the menus manually, export them to xml, walla.
[31-May-2011 16:45:11] <rocket> as long as you know the zendmd to make it work .. you can be more flexible that way
[31-May-2011 16:45:36] <Hackman238> rocket: Good point.
[31-May-2011 16:45:36] <rocket> in this case I believe it is exporting the whole menu
[31-May-2011 16:45:47] <rocket> not just the menu element you are trying to tweak
[31-May-2011 16:46:06] <Hackman238> rocket: Yep it does, but once I prune out everything but my entry the problem persists.
[31-May-2011 16:47:58] <Hackman238> rocket: you're right though, I should use the dmd methods
[31-May-2011 16:48:07] <Hackman238> rocket: I appreciate it
[31-May-2011 16:49:18] <Apocalipse> hello people
[31-May-2011 16:49:42] <Hackman238> Helllo
[31-May-2011 16:51:27] <Apocalipse> hello Hackman238
[31-May-2011 16:55:48] <JohnnyNoc> hey Hackman238an238
[31-May-2011 16:55:50] <Sam-I-Am> .
[31-May-2011 16:55:56] <JohnnyNoc> or anyone
[31-May-2011 16:56:02] <JohnnyNoc> can Zenoss Core do syslog forwarding?
[31-May-2011 16:56:05] <Hackman238> JohnnyNoc: 'ello
[31-May-2011 16:56:20] <JohnnyNoc> i was under the impression it was a Core feature, but now i'm seeing something that leads me to believe it's enterprise only
[31-May-2011 16:56:20] <Hackman238> JohnnyNoc: Can use tee?
[31-May-2011 16:56:22] <JohnnyNoc> am i crazy?
[31-May-2011 16:57:45] <froztbyte> JohnnyNoc: I found it easier just to make syslog-ng do that, and set up zensyslog on another port (which syslog-ng then forwards to)
[31-May-2011 16:58:38] <JohnnyNoc> but why
[31-May-2011 16:58:41] <Apocalipse> any people here has zenoss making voice alert calls?
[31-May-2011 16:59:05] <JohnnyNoc> whats wrong with the default syslog?
[31-May-2011 16:59:58] <froztbyte> "default syslog"?
[31-May-2011 17:00:08] <JohnnyNoc> standard syslog
[31-May-2011 17:00:10] <JohnnyNoc> non syslog-ng
[31-May-2011 17:00:15] <JohnnyNoc> the syslog that comes by default in my OS
[31-May-2011 17:00:24] <froztbyte> you do know that that means basically nothing, right?
[31-May-2011 17:00:41] <froztbyte> some distros have syslog, others have ksyslog, others have rsyslog, others...
[31-May-2011 17:00:44] <froztbyte> you get the picture
[31-May-2011 17:00:50] <JohnnyNoc> well
[31-May-2011 17:00:52] <JohnnyNoc> mine is called syslog
[31-May-2011 17:00:56] <JohnnyNoc> :>
[31-May-2011 17:01:47] <JohnnyNoc> ok how about sysklogd-1.4.1-46.el5 on CentOs 5.5
[31-May-2011 17:01:49] <froztbyte> I like syslog-ng because the conf syntax is non-shit and it's easy to do awesome stuff like writing to /var/log/netlogs/$ip/logfile-of-sorts(-by-another-macro-if-you-so-choose).log as well as mirroring to a network sink
[31-May-2011 17:02:16] <froztbyte> I tried figuring out how to do it with rsyslog (since it's the debian default) but the documentation was a bloody nightmare, so I just gave up
[31-May-2011 17:02:25] <Hackman238> froztbyte: Well stated :-)
[31-May-2011 17:02:44] <Hackman238> Have to run everyone. TTYT
[31-May-2011 17:04:20] <Sam-I-Am> lates
[31-May-2011 17:05:07] <froztbyte> in syslog-ng I define a new destination of type udp and port 5514 (which is what I use for zenoss), and then add both destinations to stuff coming from the network
[31-May-2011 17:05:40] <froztbyte> additionally I can filter stuff at the syslog daemon level, as I need to do in lesotho for hits coming from the one ASA-pair
[31-May-2011 17:05:45] <froztbyte> works wonderfully
[31-May-2011 17:06:11] <froztbyte> and nearly anyone can open the config and understand what's going on, as opposed to rsyslog's gibberish
[31-May-2011 17:07:18] <froztbyte> holy crap they've finally gotten something useful on their wiki:
[31-May-2011 17:07:54] <rmatte> JohnnyNoc: syslog-ng is what I use and it works great for that
[31-May-2011 17:08:14] <rmatte> It's high performance, does source spoofing well, and is fairly simple to configure
[31-May-2011 17:08:48] <rmatte> You just run it on the standard syslog port, change the port that Zenoss is listening for syslog on, then forward the syslogs that come to syslog-ng to Zenoss while doing whatever else you want to do with them.
[31-May-2011 17:08:58] <JohnnyNoc> i understand
[31-May-2011 17:09:07] <JohnnyNoc> i was just surprised to read a post on the forums saying thiw as a non-core feature
[31-May-2011 17:09:20] <froztbyte> destination recvRemote { file("/var/log/recvRemote/$HOST/$FACILITY.log"); };
[31-May-2011 17:09:21] <froztbyte> destination nZenoss { udp("" port(5514)); };
[31-May-2011 17:09:21] <JohnnyNoc> so i was hping someone could confirm it was
[31-May-2011 17:09:36] <froztbyte> log { source(s_network);  destination(recvRemote); destination(nZenoss); };
[31-May-2011 17:09:42] <rmatte> JohnnyNoc: zensyslog is a really low performance syslog server compared to rsyslog/syslog-ng/ksyslog, so it's not really ideal to use for that anyways
[31-May-2011 17:09:48] <froztbyte> ^ how to make this happen on syslog-ng
[31-May-2011 17:10:10] <froztbyte> see documentation for more funky steps
[31-May-2011 17:13:18] <rmatte> What annoys me is that most distros don't provide syslog-ng with the spoofing functionality compiled in... I had to make my own custom .debs that have it compiled in
[31-May-2011 17:13:49] <froztbyte> yar, I had to pull 3.0 into our source repo too
[31-May-2011 17:17:24] kocolosk is now known as kocolosk|away
[31-May-2011 17:18:45] <JohnnyNoc> so, if zensyslog is low performance and sucks
[31-May-2011 17:18:56] <JohnnyNoc> what good does having syslog-ng forward to it?
[31-May-2011 17:20:28] <rmatte> I'm just saying that even if zensyslog could forward the syslogs, then you'd have it doing 2 things
[31-May-2011 17:20:43] <JohnnyNoc> but it wouldn't have to forward in that case, no?
[31-May-2011 17:20:52] <rmatte> either way, we've explained how it needs to be done with core, we'll leave it at that and not make a debate out of this
[31-May-2011 17:20:55] <rmatte> use what you want to use
[31-May-2011 17:20:58] <JohnnyNoc> i'm not trying to be difficult, but you say that zensyslog sucks and is low performa
[31-May-2011 17:21:00] <JohnnyNoc> ok
[31-May-2011 17:21:49] <rmatte> I didn't say "it sucks", I said it's lower performance meaning it has the capacity to process about 5000 syslogs per second (I think is what I heard it quoted at)
[31-May-2011 17:22:03] <rmatte> whereas syslog-ng can do like 5 times that much
[31-May-2011 17:22:14] <Sam-I-Am> gehhhhh @ organizer graph report
[31-May-2011 17:22:23] * Sam-I-Am watches his VM catch fire
[31-May-2011 17:22:55] <JohnnyNoc> yes, i was paraphrasing but i'll stfu
[31-May-2011 17:24:20] <rmatte> lol
[31-May-2011 17:43:31] <hmp> rmatte: ever hit the 5k limit?
[31-May-2011 17:44:44] <hmp> since syslog-ng can do over 150k logs per sec, that sounds waaay too low
[31-May-2011 17:45:29] <f00fSteR> 30 more mins ... then then FREEEEEEEeeee---ddoooommmmmmmmmUUUUUUU!!!
[31-May-2011 17:46:28] <hmp> wait, you get a free xen child?
[31-May-2011 17:50:05] <f00fSteR> Hackman238, dude... the events table is over 2.5 gb's
[31-May-2011 17:50:11] <f00fSteR> i've been waiting for the dump all day
[31-May-2011 17:50:16] <f00fSteR> accidently cancelled it a few times
[31-May-2011 17:50:21] <f00fSteR> it's all sorts of crazy over hea
[31-May-2011 17:50:22] <f00fSteR> !
[31-May-2011 17:52:39] <rmatte> hmp: no I haven't hit that limit, but that's what I was told the limit was when I asked one of the developers one time
[31-May-2011 17:53:08] <f00fSteR> hmp, get with it man... i said freedom!
[31-May-2011 17:53:34] <rmatte> hmp: I use syslog-ng to filter out security event floods on some of my Zenoss servers so that they don't make it to Zenoss at all.
[31-May-2011 17:55:45] <f00fSteR> rmatte, how can i pack/compress my events table ?
[31-May-2011 17:59:16] <JohnnyNoc> ok
[31-May-2011 17:59:31] <JohnnyNoc> so i was a rebel and decided to follow a doc on the forums which is similar to what you had suggested
[31-May-2011 17:59:32] <f00fSteR> johnf1911, wb
[31-May-2011 17:59:40] <f00fSteR> JohnnyNoc, err. wb
[31-May-2011 18:02:29] <JohnnyNoc> thanks
[31-May-2011 18:02:38] <JohnnyNoc> f00fSteR to answer your question, you don't really pack/compress it (i don't think)
[31-May-2011 18:02:45] <JohnnyNoc> you do that with zope but this is mysql
[31-May-2011 18:02:52] <JohnnyNoc> but you can reduce the amount of history you keep
[31-May-2011 18:03:26] <lthrasher> :q
[31-May-2011 18:03:40] <lthrasher> hah, go me
[31-May-2011 18:03:41] <f00fSteR> how do i do that
[31-May-2011 18:03:45] <hmp> rmatte: i dont have that much syslogs anyway
[31-May-2011 18:04:02] <hmp> rmatte: just curious about it beeing that low
[31-May-2011 18:04:21] <JohnnyNoc> f00fSteR i use 2.5.3 and you find it by clicking 'event manager' in the lower left hand corner
[31-May-2011 18:04:34] <JohnnyNoc> from there you should find it
[31-May-2011 18:04:40] <JohnnyNoc> <lthrasher> :q
[31-May-2011 18:04:41] <JohnnyNoc> er
[31-May-2011 18:04:46] <JohnnyNoc> Delete Historical Events Older Than (days)
[31-May-2011 18:05:28] <locohost_> just put my cable modem in zenoss, pretty happy with the results, heh, graphing upstream powerlevel, downstream power level, signal/noise , bit errors, up QAM, Down QAM, upstream frequency, downstream frequency and channel ID
[31-May-2011 18:05:55] <rmatte> lol
[31-May-2011 18:06:13] <rmatte> well, beer time, later all
[31-May-2011 18:06:28] <f00fSteR> locohost_, thats pretty awesome... you should write up a tiny howto on how to do it... i'd like to do that for our switches and maybe my cable modem at home... i have a ir33t surfboard modem
[31-May-2011 18:06:35] <f00fSteR> rmatte, tc dude
[31-May-2011 18:07:59] <locohost_> the problem with cable modems is your isp will not give you access to snmp on it, it only listentens for t hat on the outside interface. almost all of them though, have a web interface at and some sort of diag tab
[31-May-2011 18:08:25] <f00fSteR> what is the zen pcak used for it ?
[31-May-2011 18:08:26] <locohost_> im just curling the html of the diag page and parsing it with sed/awk
[31-May-2011 18:09:28] <locohost_> i know surfboards have a diag page at that address that im sure is parsable, i have an Thompson modem which are not too popular
[31-May-2011 18:10:00] <f00fSteR> sed awk ?
[31-May-2011 18:11:37] <hmp> f00fSteR: line/text manipulation tools on *nix systems
[31-May-2011 18:12:02] <locohost_> foofSter: bash
[31-May-2011 18:12:11] <f00fSteR> ahh
[31-May-2011 18:12:13] <f00fSteR> shit i dunno
[31-May-2011 18:12:15] <f00fSteR> i dunno
[31-May-2011 18:12:27] <f00fSteR> how'm i a sys admin wo kowing any scripting/programming
[31-May-2011 18:12:36] <JohnnyNoc> windows admin
[31-May-2011 18:12:39] <JohnnyNoc> >:X
[31-May-2011 18:12:40] <hmp> :F
[31-May-2011 18:13:08] <locohost_> [zenoss@healie log]$ curl -s|grep value|awk -Fvalue= '{print $2}'|awk -F"/>" '{print $1}'|sed s/\'//g
[31-May-2011 18:13:08] <locohost_> 705.000 MHz
[31-May-2011 18:13:08] <locohost_> 37.6 dB
[31-May-2011 18:13:08] <locohost_> -2.1 dBmV
[31-May-2011 18:13:08] <locohost_> 0.000 %
[31-May-2011 18:13:10] <locohost_> 256 QAM
[31-May-2011 18:13:12] <locohost_> Acquired
[31-May-2011 18:13:14] <locohost_> 38.0 MHz
[31-May-2011 18:13:16] <locohost_> 50.0 dBmV
[31-May-2011 18:13:18] <locohost_> 12
[31-May-2011 18:13:22] <locohost_> 16 QAM
[31-May-2011 18:14:04] <locohost_> lol, i have vb
[31-May-2011 18:14:12] <locohost_> s/have/hate/
[31-May-2011 18:15:35] <locohost_> so, exactly that is not going to work unless you have the exact same modem as me, probably the same version even, but its easy enough to parse
[31-May-2011 18:16:14] <locohost_> was pretty easy except i was scratching my head and eventually banging my head against my keyboard on the way zenoss expects input from custom commands, i was just giving it a float
[31-May-2011 18:22:18] <froztbyte> oooookay
[31-May-2011 18:22:19] <froztbyte> finally
[31-May-2011 18:22:31] <froztbyte> Achievement Unlocked
[31-May-2011 18:22:39] <froztbyte> (2.5.2, migrated, 64bit)
[31-May-2011 18:22:51] <froztbyte> rocket / Hackman238: ta for the Data.fs tip
[31-May-2011 18:23:03] <froztbyte> soon to come, zenoss v3!
[31-May-2011 18:23:10] <hmp> froztbyte: too moch WoW :))
[31-May-2011 18:23:19] <froztbyte> hmp: ....WoW?
[31-May-2011 18:23:46] <hmp> "Achievement Unlocked" - I think thay have something like that in World of Warcraft
[31-May-2011 18:24:11] <froztbyte> I don't think so, but it's definitely in XBox/Steam world
[31-May-2011 18:24:19] <froztbyte> gamer expression in general
[31-May-2011 18:28:12] <locohost_>
[31-May-2011 18:28:41] <locohost_> if it was wow, he would get some kind of "drake" for doing it
[31-May-2011 18:36:01] locohost_ is now known as locohost
[31-May-2011 19:01:08] <froztbyte> man it's nice to not be running that in an IO-constrained VM anymore
[31-May-2011 19:01:13] <froztbyte> so snappy
[31-May-2011 19:11:35] <locohost> yeah, io contention sucks
[31-May-2011 20:26:07] kocolosk|away is now known as kocolosk
[31-May-2011 21:17:23] <locohost> under advanced and monitoring templates and click on deviceCSMacd and go to override template and select /Network/Switch/Cisco/Catalyst, im creating a local copy of devices under catalysists, correct?
[31-May-2011 21:17:36] <locohost> so, i can apply threaholds there, non-globally, but for catalysts
[31-May-2011 21:17:47] <locohost> thats not overiding the device template with whats in catalysts, correct?
[31-May-2011 21:18:00] <johnf1911> correct
[31-May-2011 21:19:36] <locohost> thks
[31-May-2011 21:19:40] <locohost> anyone around monitor netapp sans?
[31-May-2011 21:20:21] <locohost> may be giving up monitoring interfaces on the device, ill just get it at the switch
[31-May-2011 21:20:46] <locohost> only problem with that is dont see errors that the switch thinks are legit packets and the san doesnt like
[31-May-2011 21:24:52] <locohost> im trying to think of the best way to move about 15 devices to a sub class of the class they are in, keep all existing data/bindings and then add some unique things
[31-May-2011 21:30:48] <rmatte2> has everyone cleared out for the evening?
[31-May-2011 21:33:20] <rmatte2> guess so
[31-May-2011 21:58:11] <locohost> gah, it appears to me that threholds set higer in the tree, still trickle down to lower calsses even if they have that same template applied localty
[31-May-2011 21:59:18] <davetoo> Should not.
[31-May-2011 21:59:24] <Sam-I-Am> yea
[31-May-2011 21:59:25] <locohost> guess ill just take ethernetCsmcd /device templates out and put them in each class
[31-May-2011 21:59:44] <locohost> u sure? i have a threhold of 400mbps set for /devices in the ethernetcsmcd template
[31-May-2011 22:00:03] <locohost> and applied it locally to /device/switch/cisco/catalyst
[31-May-2011 22:00:32] <davetoo> Did you un-rename your templates?  Didn't you copy _64 to the base template name last week?  I wonder if that's left some crumbs.
[31-May-2011 22:00:33] <locohost> i then set the throughput threaholds higher in this localized template for this class
[31-May-2011 22:02:27] <locohost> hehe, didnt rename anything doing this, and its pulling the vars fine, maybe i didnt want long enough and the poller didnt catch my change, ill delete the 400mbps alerts for the device in a sub class with a 800mbps threshold(alarming at 600mbps) and see if it doesnt bark next iteration
[01-Jun-2011 00:00:01] [disconnected at Wed Jun  1 00:00:01 2011]

Comments (0)