<html<pre>[01-Oct-2011 00:00:01] [connected at Sat Oct 1 00:00:01 2011]
[01-Oct-2011 00:00:13] [zenoss-logger (logger bot) has joined #zenoss]
[01-Oct-2011 01:02:49] kocolosk|away is now known as kocolosk
[01-Oct-2011 01:15:11] kocolosk is now known as kocolosk|away
[01-Oct-2011 10:16:53] kocolosk|away is now known as kocolosk
[01-Oct-2011 11:22:16] kocolosk is now known as kocolosk|away
[01-Oct-2011 12:00:01] [disconnected at Sat Oct 1 12:00:01 2011]
[01-Oct-2011 12:00:08] [connected at Sat Oct 1 12:00:08 2011]
[01-Oct-2011 12:00:21] [zenoss-logger (logger bot) has joined #zenoss]
[02-Oct-2011 00:00:01] [disconnected at Sun Oct 2 00:00:01 2011]
[02-Oct-2011 00:00:02] [connected at Sun Oct 2 00:00:02 2011]
[02-Oct-2011 00:00:12] [zenoss-logger (logger bot) has joined #zenoss]
[02-Oct-2011 12:00:01] [disconnected at Sun Oct 2 12:00:01 2011]
[02-Oct-2011 12:00:02] [connected at Sun Oct 2 12:00:02 2011]
[02-Oct-2011 12:00:16] [zenoss-logger (logger bot) has joined #zenoss]
[03-Oct-2011 00:00:01] [disconnected at Mon Oct 3 00:00:01 2011]
[03-Oct-2011 00:00:01] [connected at Mon Oct 3 00:00:01 2011]
[03-Oct-2011 00:00:12] [zenoss-logger (logger bot) has joined #zenoss]
[03-Oct-2011 01:13:18] <rcngeoff> Hi. After an upgrade to 3.2.0 we see some weird false alets like: "unabel to reach proces so device XXX: err: usr_bin_YYY: b4873e7d11...
[03-Oct-2011 01:13:58] <rcngeoff> so.. too early for me, that should be: : "unable to read proces on device XXX: err: usr_bin_YYY: b4873e7d11..
[03-Oct-2011 01:14:28] <rcngeoff> Has anyone else seen such false alerts?
[03-Oct-2011 01:21:06] <rcngeoff> a ha... I just restarted zenprocess and these alerts are clearing. Still, I am curious if anyone has any insight into what causes this?
[03-Oct-2011 05:59:14] <Yanto_cpt> morning every1!
[03-Oct-2011 05:59:31] <Yanto_cpt> ooops that was insensitive - and good afternoon and good evening to the rest
[03-Oct-2011 07:17:41] <squig> I was looking at my linux cpu graphs and see that there is a problem with wait time on my cpu graphs, I have multi proc/core machines. user cpu is limited to 100% but wait can go to 800%
[03-Oct-2011 07:17:48] <squig> and ideas on what I need to do to resolve it
[03-Oct-2011 08:17:10] <Gat0rvean> Morning all
[03-Oct-2011 09:02:12] <jb> morning.
[03-Oct-2011 09:51:06] <humanix> hi there guys
[03-Oct-2011 09:52:28] <SDuensin> Morning.
[03-Oct-2011 10:13:19] <JohnnyNOC> is there a better way to re-ip a machine that was already being used without having to delete the device first?
[03-Oct-2011 10:13:44] <JohnnyNOC> i have 2 devices that need their ips flipped, but i can't temporarily assign an ip to one of the servers and then assign it's old ip to the OTHER server..
[03-Oct-2011 10:13:54] <JohnnyNOC> bounced zopectl, then zenoss, but it still says failed to reset ip address
[03-Oct-2011 10:13:55] <JohnnyNOC> :[
[03-Oct-2011 10:24:36] <JohnnyNOC> looks like it's just better to do it in/through zend
[03-Oct-2011 10:24:41] <JohnnyNOC> at least the device was actually deleted
[03-Oct-2011 10:24:42] <JohnnyNOC> :/
[03-Oct-2011 11:41:09] <pplask> hello!
[03-Oct-2011 11:42:52] <pplask> im new to zenoss, and i think its the best monitor system i ever see. But i read some complaints about zenoss is not able to monitor hosts that uses dynamic DNS services. Is it true?
[03-Oct-2011 11:46:36] <rocket> pplask: out of the box that is correct
[03-Oct-2011 11:46:56] <rocket> pplask: we cache the initial ip address of the device as its discovered the first time
[03-Oct-2011 11:47:09] <pplask> mmmm
[03-Oct-2011 11:47:16] <pplask> is there any solution for that?
[03-Oct-2011 11:47:17] <rocket> pplask: you could write a script that scanned for changes and updated the ips
[03-Oct-2011 11:47:49] <pplask> mmmm
[03-Oct-2011 11:48:07] <rocket> pplask: not out of the box. From a monitoring perspective we have taken the approach that if it is something worth monitoring its ip probably shouldnt change that much
[03-Oct-2011 11:49:07] <pplask> i would like to monitor the status of some remote offices of my company
[03-Oct-2011 11:49:23] <pplask> check if the cameras are recording
[03-Oct-2011 11:49:25] <pplask> that stuff
[03-Oct-2011 11:49:37] <pplask> they are small offices with normal ISPs
[03-Oct-2011 11:50:43] <rocket> I guess I would have a script that compared the original ip to the new ip
[03-Oct-2011 11:50:48] <pplask> mmm
[03-Oct-2011 11:50:50] <rocket> if the ip changed then update zenoss
[03-Oct-2011 11:50:51] <pplask> i understand
[03-Oct-2011 11:51:02] <pplask> and if i wanted to ue SNMP
[03-Oct-2011 11:51:04] <rocket> run that from a cron job every 5 minutes etc
[03-Oct-2011 11:51:13] <pplask> i could just configure the office to send traps, right?
[03-Oct-2011 11:51:44] <rocket> yes .. there is no reason zenoss will not accept the trap .. the only issue might be that zenoss doesnt match the correct device to the source ip
[03-Oct-2011 11:52:16] <pplask> mmm so it could mark as UNKNOWN right?
[03-Oct-2011 11:53:28] <rocket> it would show up in the event console .. if it has never seen that ip before there would not be the possibility to click on the device ip and go to a zenoss monitored device
[03-Oct-2011 11:53:48] <rocket> if it matches an existing ip but the mapping is wrong it might link to the wrong device
[03-Oct-2011 11:54:12] <rocket> so this is another thought too
[03-Oct-2011 11:54:20] <rocket> zenoss will ping the remote device
[03-Oct-2011 11:54:31] <rocket> if it cannot ping the remote device because the ip has changed
[03-Oct-2011 11:54:59] <pplask> mmm, but when my script correct the ip of the device, will zenoss attribute those unknown traps to the right server? dont know if i was clear
[03-Oct-2011 11:55:02] <rocket> you can have zenoss do a command and that command queries dns and reconfigures zenoss etc
[03-Oct-2011 11:55:22] <rocket> no .. it would just be new events coming in after that point
[03-Oct-2011 11:55:35] <pplask> mmm k, so traps are not a good solution
[03-Oct-2011 11:56:09] <rocket> search the forums .. others may have a solution as well
[03-Oct-2011 11:58:26] <pplask> mmm... ok. My second and last point is about NAT'ed servers. I saw a tutorial explaining how to install a collector behind the firewall using zenoss core. My question is: if i install a normal zenoss core (with all features enabled) inside a local network, to monitor all devices inside this network, can i ALSO send all info collected to a remote zenoss out of the box?
[03-Oct-2011 12:00:01] [disconnected at Mon Oct 3 12:00:01 2011]
[03-Oct-2011 12:00:02] [connected at Mon Oct 3 12:00:02 2011]
[03-Oct-2011 12:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[03-Oct-2011 12:05:54] <dhopp> If nothing has changed in zenoss (no zenpack install/removal, no device addition/removal, etc.) then I should be able to stop zenoss remove Data.fs.index and restore Data.fs to a backed up version and start Zenoss and everything should be as it was when that file was backed up right?
[03-Oct-2011 12:13:45] <jrh0090> dhopp: as long as the same python files are present you should be able to just copy the Data.fs file (I think you don't need the index but it won't to copy it just in case)
[03-Oct-2011 12:14:08] <jrh0090> I know I have gotten other peoples backedup Data.fs files in the past to test things
[03-Oct-2011 12:15:28] <dhopp> my understanding is if you are restoring an older Data.fs then you need to remove the index since the index would be out of sync (I think it gets recreated)…I'm asking because I think I'm going to be doing a zeopack every 4 hours or so and in the event of something going haywire could just restore that rather than doing a full restore (again assuming nothing major has changed)
[03-Oct-2011 12:17:00] <jrh0090> Yes, this would restore the Model data only, not rrd or events, but yeah it is much quicker than doing a full zenbackup and zenrestore
[03-Oct-2011 12:17:41] <dhopp> jrh0090: right..I'm assuming in this case the reason I'm restoring Data.fs is because some relationship is just completely borked…and I still have the RRD and events db
[03-Oct-2011 12:17:59] <jrh0090> cool
[03-Oct-2011 12:18:28] <dhopp> If the box completely dies then I would need to do a full restore to get the RRD data and events data (although I'm replicating the DB so hopefully that isn't even required)
[03-Oct-2011 12:19:14] <jrh0090> correct 
[03-Oct-2011 12:19:21] <jrh0090> hopefully that won't happen though
[03-Oct-2011 12:19:26] <dhopp> jrh0090: I'm starting down the path of my first zenpack…I'm assuming in dev I'll need to do this :-)
[03-Oct-2011 12:20:12] <dhopp> i.e. while I'm learning the ropes I'll probably bork a relationship or two and want to be able to backup Data.fs prior to installing my zenpack and restore if it goes horribly wrong
[03-Oct-2011 12:20:53] <jrh0090> hah yeah i definitely have. Unfortunately when you get something wrong you don't know about it until later and then the ZODB is all jacked up
[03-Oct-2011 12:21:56] <dhopp> jrh0090: how much later? as in when you install another zenpack and it bombs? or just one day all of a sudden no monitoring is working?
[03-Oct-2011 12:22:17] <dhopp> jrh0090: and how did you fix that?
[03-Oct-2011 12:27:08] <dhopp> :q!
[03-Oct-2011 12:27:11] <dhopp> oops
[03-Oct-2011 12:39:06] <jb> daemons and packages that don't provide a native stats mechanism are very annoying :/
[03-Oct-2011 12:46:45] <Sashness> hello. i'm trying to create a report that lists every device with a certain event class. I'm using the Custom Device Report
[03-Oct-2011 12:46:46] <Sashness> but I'm not sure what the query would be to query events
[03-Oct-2011 12:46:53] <Sashness> or if i could even query events from there
[03-Oct-2011 12:59:13] <jdross> what options do I need to set in globals.conf to have a remote mysql server in 4.0.2
[03-Oct-2011 13:15:24] <nyeates> jdross: your talking enterprise right... i mean we dont have core in 4.* yet
[03-Oct-2011 13:16:04] <nyeates> I do not know the answer off hand, and I am thinking that it is laid out in our install or admin guide specifically for 4.*
[03-Oct-2011 13:18:48] <nyeates> here are the instructions: servicedynamics/resource_manager_installation/4.0-v03/ch02s02.html
[03-Oct-2011 13:18:57] <nyeates> 2.3 (Optional)
[03-Oct-2011 14:13:05] <jdross> nyeates, that guide and the install guide only has how to set it via the install or upgrade scripts. I am actually moving zends on an already installed instance
[03-Oct-2011 14:23:01] <tvincent> anyone know how to clone a existing dashboard to another user?
[03-Oct-2011 14:35:07] <rocket> tvincent: I believe this will do what you want
[03-Oct-2011 14:35:10] <rocket> thread/14840;jsessionid=EBE927B34098BEA33E85446003471FCB.node0
[03-Oct-2011 14:44:40] <humanix> someone there? :d
[03-Oct-2011 14:49:08] <tvincent> its a quiet monday for this channel.
[03-Oct-2011 14:49:13] <tvincent> Hackman must be on vacation. =)
[03-Oct-2011 14:49:30] <jmp242> Hackman is actually travelling
[03-Oct-2011 14:49:37] <jmp242> but I don't know he's on vacation
[03-Oct-2011 15:34:07] kocolosk|away is now known as kocolosk
[03-Oct-2011 16:11:44] <jb> anybody around that has some experience with nagios plugins/zencommand?
[03-Oct-2011 16:16:21] <jmp242> Not too much experiance, but what's the question jb
[03-Oct-2011 16:17:22] <jb> might have got it.. one sec
[03-Oct-2011 16:17:37] <jb> I think I was over-complicating something that zenoss already is capable of doing natievly..
[03-Oct-2011 16:21:37] <jb> yeah, it works as expected.. just making a check_smtp template.
[03-Oct-2011 16:25:43] <jb> hrm, how often does zencommand run?
[03-Oct-2011 16:30:59] <jb> zencommand noticed that my nagios plugin was CRITICAL, but it didn't clear the event until I manually re-ran zencommand again.
[03-Oct-2011 16:36:06] <rocket> jb: your cycle time should be 5 minutes by default and it should rerun every 5 minutes
[03-Oct-2011 16:36:27] <rocket> without seeing logs I wouldnt have a clue why it didnt clear on its own
[03-Oct-2011 16:36:32] <jb> hrm, yeah, 5 minutes is too long for things like check_ftp/check_ntp/check_smtp
[03-Oct-2011 16:36:54] <jb> that means I could possibly not know that a service was unavailable for 5 minutes.
[03-Oct-2011 16:37:20] <jb> rocket: I probably didn't wait 5 minutes.. when I manually ran zencommand, it cleared.
[03-Oct-2011 16:38:06] <rocket> jb: makes sense to me
[03-Oct-2011 16:38:28] <jb> so, even with check_ftp/other build in zenpacks that use zencommand, the interval is 5 minutes?
[03-Oct-2011 16:38:38] <rocket> yes
[03-Oct-2011 16:38:50] <jb> and that a collector-wide setting, right?
[03-Oct-2011 16:39:10] <rocket> zencommand might let you choose a shorter cycle
[03-Oct-2011 16:39:23] <rocket> double checking
[03-Oct-2011 16:40:29] <jb> hrm yeah
[03-Oct-2011 16:40:37] <rocket> cycle time 300
[03-Oct-2011 16:40:37] <jb> there is a zprop for zcommandcycletime
[03-Oct-2011 16:40:52] <rocket> well when you configure the datasource there is an option
[03-Oct-2011 16:40:56] <jb> and it's set to 60 on this device.
[03-Oct-2011 16:40:59] <jb> oh!
[03-Oct-2011 16:40:59] <jb> sec.
[03-Oct-2011 16:41:17] <rocket> you will need to remove the rrd files because they have the cycle times set in them
[03-Oct-2011 16:41:52] <jb> ok lets try this again.
[03-Oct-2011 16:54:46] <jb> rocket: cool, that worked. thanks
[03-Oct-2011 17:45:52] <humanix> hi there
[03-Oct-2011 17:46:03] <humanix> rmatte,
[03-Oct-2011 17:46:15] <humanix> have you restored zenoss from a backup using --zenpack option under 2.5?
[03-Oct-2011 21:44:58] kocolosk is now known as kocolosk|away
[04-Oct-2011 00:00:01] [disconnected at Tue Oct 4 00:00:01 2011]
[04-Oct-2011 00:00:02] [connected at Tue Oct 4 00:00:02 2011]
[04-Oct-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[04-Oct-2011 09:15:54] <Gat0rvean> Could anyone help me with some Event mgmt? I'm trying to get these traps to clear eachother, but they don't seem to want to
[04-Oct-2011 09:17:43] <fragfutter> define an event class, assign the trigger event to it with a severity. assign the clear trap to it with a severity of clear.
[04-Oct-2011 09:18:29] <Gat0rvean> fragfutter: I did exactly that, and it basically just doesn't come in any more, I tried setting it back to "Error" (as opposed to clear) and see it plain as day when I trigger the event
[04-Oct-2011 09:18:32] <fragfutter> fragfutter: can't walk you through it. at the moment i have no zenoss available
[04-Oct-2011 09:19:37] <fragfutter> Gat0rvean: only the second trap (the one that says everything is fine) is a clear event. The first one has to be an error/warn ...
[04-Oct-2011 09:19:52] <Gat0rvean> fragfutter: correct
[04-Oct-2011 09:21:15] <Gat0rvean> fragfutter: the "SetAIS..." is error and the "ClrAIS..." is clear, and under config properties of the "SetAIS..." zEventClearClasses is set to /Net/Link/Adtran - MUX/adMX2800xClrT1E1AISCarrier, which is where the ClrAIS... is located
[04-Oct-2011 09:28:06] <dhopp> ZenPack development question…what exactly the difference between DeviceComponent and OSComponent?
[04-Oct-2011 09:35:17] <jb> morning.
[04-Oct-2011 09:35:42] <dhopp> jb: morning
[04-Oct-2011 09:39:21] <SDuensin> Good morning.
[04-Oct-2011 09:50:05] <fragfutter> dhopp: in zenoss2 they where on different tabs/menu-paths
[04-Oct-2011 09:50:16] <fragfutter> -h
[04-Oct-2011 09:54:28] <dhopp> fragfutter: I'm trying to figure out what is best to build a subclass from. Looking at other zenpacks things that I thought would be a subclass of OSComponent are a subclass of DeviceComponent and vice versa so I'm trying to get an understanding of why. I *thought* the way to think about it is if the component was tangible (i.e a CPU, Memory, Disk, etc.) then it was a DeviceComponent, if it was that only existed once an OS was running (a process, filesystem,
[04-Oct-2011 10:01:24] kocolosk|away is now known as kocolosk
[04-Oct-2011 10:05:51] <fragfutter> dhopp: i would think the same. But i never checked on the code level.
[04-Oct-2011 10:20:10] <SDuensin> OK, I've confused myself again. I've managed to write a whole suite of modeling plugins that I've put into a ZenPack for querying my Windows machines. But I'm having problems finding information on how to get "data points" for performance monitoring out of them.
[04-Oct-2011 10:39:53] <arfon> Hi, can someone tell me the format of the device line when using zenbatchload? I need to pass the IP and the device name...
[04-Oct-2011 10:40:24] <arfon> I know it starts with the IP but, how do you specify the device name?
[04-Oct-2011 10:44:13] <SDuensin> Sorry, arfon, I've got no idea. Arguing with collecting data here.
[04-Oct-2011 10:44:40] <arfon> 
Thanks for the reply though.
[04-Oct-2011 10:44:56] <dhopp> arfon: I'm trying to figure it out...
[04-Oct-2011 10:45:23] <arfon> ty
[04-Oct-2011 10:46:27] <arfon> I have 500 routers to add to /ping and writing a script to wget them isn't something I really want to do...
[04-Oct-2011 10:46:48] <SDuensin> arfon, YW. Sitting in silence can be maddening. :-)
[04-Oct-2011 10:46:56] <fragfutter> for i in $(cat somefile.txt); do wget ...; done
[04-Oct-2011 10:47:11] * SDuensin loves him some wget.
[04-Oct-2011 10:47:27] <arfon> ty fragfutter....
[04-Oct-2011 10:49:27] <arfon> It would be nice if: /Ping; 192.268.1.1 devicename="router1"; 192.168.1.2 devicename="router2"; zenbatchload ....
[04-Oct-2011 10:49:54] <fragfutter> arfon: i think it's the reverse DNS lookup
[04-Oct-2011 10:50:46] <arfon> OOoo, that sux. these routers aren't in our DNS....
[04-Oct-2011 10:52:00] <arfon> (our IT guys that controll these router are a bunch of lazy bastages)
[04-Oct-2011 10:52:05] <dhopp> fragfutter: that is the default way that it will set you should be able to set it explicitly to
[04-Oct-2011 10:52:26] <arfon> I thought maybe there was a zProperty?
[04-Oct-2011 10:52:52] <arfon> "192.168.1.1 zDevicename=router1" ??
[04-Oct-2011 10:53:08] <arfon> <---painfully new to zenoss
[04-Oct-2011 10:53:24] <dhopp> arfon: you can do 192.168.1.1 setTitle="router1" and that will atleast make it so the device list shows "router1"
[04-Oct-2011 10:53:41] <arfon> THAT is what I want.... TY
[04-Oct-2011 10:53:54] <dhopp> but the devicename will still be 192.168.1.1 (i.e. all of the internal zenoss stuff will use 192.168.1.1 and the Perf directory will be 192.168.1.1)
[04-Oct-2011 10:54:04] <arfon>
[04-Oct-2011 10:54:10] <fragfutter> arfon: you can look into zenbatchload, i just can't find the source at the moment
[04-Oct-2011 10:54:14] <arfon> That is not what I want....
[04-Oct-2011 10:55:44] <arfon> Is zenbatchload a script or compiled proggie?
[04-Oct-2011 10:55:59] <fragfutter> arfon: welcome to python, everything is source code
[04-Oct-2011 10:56:06] <fragfutter> http://dev.zenoss.org/svn/trunk/Products/ZenModel/BatchDeviceLoader.py
[04-Oct-2011 10:56:07] <arfon> HA HA!
[04-Oct-2011 10:56:16] <arfon> sweet!
[04-Oct-2011 10:57:32] <arfon> IF I am reading python right, it looks like deviceName
[04-Oct-2011 10:57:41] <arfon> <--perl guy
[04-Oct-2011 10:58:38] <SDuensin> <-- Not enjoying Python guy
[04-Oct-2011 10:58:49] <arfon>
[04-Oct-2011 10:58:58] <arfon> At least it's not java
[04-Oct-2011 10:59:04] <SDuensin> !!!!!
[04-Oct-2011 10:59:11] <SDuensin> I could deal with that!
[04-Oct-2011 10:59:20] * SDuensin has been doing some psycho Java hackery lately.
[04-Oct-2011 11:00:01] <arfon> I'm gonna try a batchload, wish me luck.
[04-Oct-2011 11:00:03] <dhopp> arfon: ok I got it to work...
[04-Oct-2011 11:00:07] <dhopp> do this:
[04-Oct-2011 11:00:23] <dhopp> router1 setManageIp='192.168.1.1'
[04-Oct-2011 11:00:24] <arfon> [holding his breath]
[04-Oct-2011 11:00:44] <dhopp> so reverse the thinking
[04-Oct-2011 11:00:56] <dhopp> add the device as the devicename and then set it's manageip
[04-Oct-2011 11:01:02] <arfon> Ah.... let me try
[04-Oct-2011 11:01:48] <jiveturkey> Getting "Lost connection to the server" on the Zenoss Dashboard on 2.5.2. MySQL is running on another box and is up and accessible. The Event Console mostly works. Thoughts?
[04-Oct-2011 11:02:08] <jiveturkey> It's also stopping my Production States and Device Issues portlets from working.
[04-Oct-2011 11:02:57] <dhopp> jiveturkey: did you move the mysql database recently?
[04-Oct-2011 11:03:11] <dhopp> did it use to be local and then you moved it remotely?
[04-Oct-2011 11:03:41] <jiveturkey> It's been remote for some time. We did however recently change the connection so that it goes through the private net rather than the public (which it really should have been all along. Whoops).
[04-Oct-2011 11:04:17] <arfon> Weird... No Such Option -i
[04-Oct-2011 11:04:47] <jiveturkey> On the box with zope and zeo and zenactions and yadda yadda--the head unit, lets call it--I can connect to MySQL on the remote box through the private net without issue.
[04-Oct-2011 11:04:58] <dhopp> jiveturkey: this sounds like a routing problem. from the command line of the zenoss server as the zenoss user can you do mysql -h remotemysqlserver -u zenoss —database=events -p
[04-Oct-2011 11:05:15] <dhopp> jiveturkey: ok…did you change the IP in Zenoss?
[04-Oct-2011 11:05:29] <jiveturkey> Through the GUI in Event Manager, you mean? Yes.
[04-Oct-2011 11:05:33] <dhopp> hrm
[04-Oct-2011 11:05:56] <dhopp> arfon: you get that no such option when doing SetManageIp?
[04-Oct-2011 11:06:07] <jiveturkey> It's a weird one... Much easier to solve problems where it doesn't work at all... The Event Console mostly works (meaning we can get to MySQL) but some times gives me a Server connection failure.
[04-Oct-2011 11:07:24] <arfon> dhopp, it failed: "WARNING zen.BatchDeviceLoader: The function 'setManageIP' for device router1 is not found."
[04-Oct-2011 11:08:03] <davetoo> hmm
[04-Oct-2011 11:08:11] <dhopp> arfon: the 'p' needs to be lower case in ManageIp
[04-Oct-2011 11:08:20] <arfon> doh
[04-Oct-2011 11:08:20] <davetoo> oh, that!
[04-Oct-2011 11:09:11] <jiveturkey> I'm getting a bunch of "Got timeout reading communication packets" in my MySQL logs.
[04-Oct-2011 11:09:13] <arfon> Looking good so far!
[04-Oct-2011 11:10:00] <arfon> dhopp, If you were a chick and I were a chick, I'd kiss you!
[04-Oct-2011 11:10:12] <dhopp> jiveturkey: you have a network problem somewhere..is there any firewalls between? I would check to make sure some kind NAT isn't happening somewhere that you aren't expecting
[04-Oct-2011 11:10:39] <dhopp> arfon: glad I'm not a chick :-
[04-Oct-2011 11:10:41] <dhopp> :-P
[04-Oct-2011 11:11:10] <arfon> Anyway, thanks, I'm gonna lurk now
[04-Oct-2011 11:12:05] <SDuensin> What? Two chicks? What'd I miss?
[04-Oct-2011 11:17:08] <jiveturkey> dhopp: Thanks for the help, by the way. The setup is pretty simple. They're both connected to a switch on a 172 network. The switch doesn't route.
[04-Oct-2011 11:17:22] <SDuensin> So anyone here good with collectors? I'm lost trying to make some custom ones.
[04-Oct-2011 11:17:34] <jiveturkey> Previously, they were connecting to each other on the public side. I'm going to switch them back to the public side and see if that fixes it. Maybe a bad network cable or something?
[04-Oct-2011 11:18:09] <dhopp> SDuensin: you mean adding more collectors to zenoss or coding a deamon?
[04-Oct-2011 11:18:13] <dhopp> *daemon
[04-Oct-2011 11:18:58] <SDuensin> dhopp, I'm not sure. I wrote a ton of modeler plugins for my network and need to be able to graph and such now. I know modelers aren't used for that but I can't find any info on writing collectors.
[04-Oct-2011 11:20:20] <dhopp> SDuensin: what did you model? Did you create custom components?
[04-Oct-2011 11:21:21] <SDuensin> No, nothing custom. I've done all I can find examples for. :-) CPU, general info, memory, disks, filesystems, routes, software, Windows services, IP services, processes, maybe a couple more.
[04-Oct-2011 11:21:49] <dhopp> SDuensin: and when looking at your device do you see those components?
[04-Oct-2011 11:22:09] <SDuensin> Oh yea. I get the modeled data back just fine.
[04-Oct-2011 11:22:22] <SDuensin> Well, most of it.
[04-Oct-2011 11:22:37] <dhopp> SDuensin: if you are collecting standard datapoints (SNMP/WMI) you don't need to write a collector, you just need templates that tell zenoss what datapoints to get
[04-Oct-2011 11:22:45] <SDuensin> Some, like in filesystems, I'm not seeing things like used disk space.
[04-Oct-2011 11:23:05] <dhopp> but this is where I'm a little fuzzy because components need components templates
[04-Oct-2011 11:23:13] <SDuensin> Ah, yea, well, that's the problem, isn't it? I can't make WMI work on this #%#%&! network.
[04-Oct-2011 11:23:29] <dhopp> and I'm not 100% how to create a component template
[04-Oct-2011 11:23:43] <dhopp> so what are you trying to monitor with, snmp?
[04-Oct-2011 11:24:05] <SDuensin> Ideally, we'd like to only have ssh open.
[04-Oct-2011 11:24:08] <dhopp> have you looked into WBEM? Egor has a zenpack that can handle that and from what I understand is way easier then dealing with the security mess that is DCOM
[04-Oct-2011 11:24:39] <dhopp> SDuensin: oh so you are going to ssh into Windows boxes?
[04-Oct-2011 11:24:47] <dhopp> are they Windows 2003 and higher?
[04-Oct-2011 11:24:51] <SDuensin> I currently am. :-)
[04-Oct-2011 11:24:55] <SDuensin> XP and up.
[04-Oct-2011 11:25:12] <fragfutter> ssh monitoring has a high cpu and network overhead
[04-Oct-2011 11:25:21] <SDuensin> Once I get on the box, I can do WMI queries and such. Getting the data isn't a problem. What to do with it is.
[04-Oct-2011 11:25:36] <SDuensin> fragfutter, yea, I know. :-( But it works and direct WMI doesn't.
[04-Oct-2011 11:25:47] <dhopp> SDuensin: at least for your Windows 2003 and 2008 boxes I would look at WBEM docs/DOC-3409
[04-Oct-2011 11:27:08] <SDuensin> dhopp, can I put something on the clients to enable that on OSs "down" to XP?
[04-Oct-2011 11:27:24] <dhopp> SDuensin: I don't know..you would need to google it
[04-Oct-2011 11:27:34] <dhopp> and what is on XP that you need to monitor?
[04-Oct-2011 11:27:44] <dhopp> you have services running on XP machines?!?
[04-Oct-2011 11:28:09] <SDuensin> dhopp, anything I can get. Disk space, interface stats, services, processes, installed software, etc.
[04-Oct-2011 11:28:25] <SDuensin> Not public services, but services for licence management and such.
[04-Oct-2011 11:28:51] <SDuensin> We're trying to defend ourselves against an army of moron users. :-)
[04-Oct-2011 11:29:14] <dhopp> SDuensin: are you wanting to know what gets installed on a desktop machine?
[04-Oct-2011 11:29:31] <SDuensin> dhopp, yes. Among other items.
[04-Oct-2011 11:29:33] <fragfutter> minesweeper
[04-Oct-2011 11:30:23] <SDuensin> You guys deal with "users". Just imagine the worst.
[04-Oct-2011 11:30:45] <dhopp> SDuensin: I'm having trouble grasping the concept of having Zenoss monitor your desktop machines…if you just want to know what is going on with them I would look at inventory software..something like http://www.ocsinventory-ng.org/en/
[04-Oct-2011 11:31:34] <SDuensin> dhopp, fewer tools the better.
[04-Oct-2011 11:33:13] <dhopp> SDuensin: agreed..but desktops change so frequently that I'm not sure a monitoring package is the best solution (PCs get shutdown and rebooted often, you are going to have a crapload of events caused by that)..but then again I don't know your environment
[04-Oct-2011 11:34:13] <SDuensin> dhopp, we realize that. We want to track as much as we can. We may ignore most of it, but we want it.
[04-Oct-2011 11:36:27] <pplask> hello
[04-Oct-2011 11:39:19] <SDuensin> Hi pplask
[04-Oct-2011 11:39:33] <pplask> im new to zenoss. Im using the template /Server/Linux to monitor a server for testing purposes, to learn how to use zenoss
[04-Oct-2011 11:39:53] <pplask> i see that he generates events of high utilization of a network interface
[04-Oct-2011 11:40:12] <pplask> i followed some tutorials to make it more human readable (using % values)
[04-Oct-2011 11:40:48] <pplask> but i couldnt find a way to set the default value of the interface
[04-Oct-2011 11:41:29] <pplask> the speed, i mean
[04-Oct-2011 11:41:38] <pplask> because this interface is used for backups for example
[04-Oct-2011 11:42:16] <pplask> so it keeps generating events saying that the utilization is 800% of the interface
[04-Oct-2011 11:42:29] <dhopp> pplask: first thing you should do is switch zenoss to using snmp v2c..it defaults to v1
[04-Oct-2011 11:43:13] <pplask> and how do i do that?
[04-Oct-2011 11:43:54] <dhopp> pplask: to do that click on Infrastructure->Device Classes->Details->ConfigurationProperties->zSnmpVer
[04-Oct-2011 11:43:57] <dhopp> change it to v2c
[04-Oct-2011 11:44:04] <pplask> i see in configuration properties, zsnmpver
[04-Oct-2011 11:44:08] <pplask> mmm
[04-Oct-2011 11:44:46] <dhopp> after you change that you need to do a couple of things
[04-Oct-2011 11:44:52] <pplask> ok but then i will need to change something in the monitored server right?
[04-Oct-2011 11:45:10] <pplask> ok, what do i need to do?
[04-Oct-2011 11:45:32] <dhopp> zsnmpver by default is inherited from the top level. Unless you overrode it anywhere else setting it at the Device Class level means all devices will use it
[04-Oct-2011 11:45:46] <dhopp> so now log into your zenoss server with ssh
[04-Oct-2011 11:46:10] <dhopp> and as the zenoss user rm -rf $ZENHOME/perf/Devices/YOURDEVICENAME
[04-Oct-2011 11:46:41] <pplask> ok
[04-Oct-2011 11:46:43] <pplask> i did it
[04-Oct-2011 11:46:57] <dhopp> then back in the UI you need to remodel the device…go to the device and click the gear menu in the bottom left and select "Model Device"
[04-Oct-2011 11:47:49] <pplask> ok done
[04-Oct-2011 11:48:12] <dhopp> this should allow you to use 64bit counters which will make monitoring gigabit cleaner
[04-Oct-2011 11:48:34] <dhopp> now clear all events for this device and wait and see if you get the same events
[04-Oct-2011 11:49:36] <pplask> ok, but to model i will have to delete always the device from this path?
[04-Oct-2011 11:49:48] <dhopp> no
[04-Oct-2011 11:50:14] <pplask> ok
[04-Oct-2011 11:50:21] <dhopp> we had to do it this time because we changed from v1 to v2c
[04-Oct-2011 11:50:47] <dhopp> I'm not 100% if it is necessary but since you only have 1 device it's better to make sure it doesn't cause any other strange issues
[04-Oct-2011 11:50:51] <pplask> oh i see, well i have another device here, but i cant find it in .../perf/Devices/
[04-Oct-2011 11:51:01] <dhopp> pplask: what device?
[04-Oct-2011 11:51:07] <dhopp> err
[04-Oct-2011 11:51:09] <dhopp> what kind of device I mean
[04-Oct-2011 11:51:15] <pplask> linux server too
[04-Oct-2011 11:51:28] <dhopp> pplask: go to the device in the UI and look for any events
[04-Oct-2011 11:51:38] <pplask> no events
[04-Oct-2011 11:51:46] <dhopp> and it shows as up?
[04-Oct-2011 11:52:01] <dhopp> pplask: what happens if you model the device? Any errors?
[04-Oct-2011 11:52:10] <pplask> yes up. no, any erros
[04-Oct-2011 11:52:12] <pplask> errors
[04-Oct-2011 11:52:15] <pplask> all seems ok
[04-Oct-2011 11:53:02] <pplask> is there a way to manually set the interface speed?
[04-Oct-2011 11:53:19] <dhopp> pplask: so you clicked the gear menu and chose "Model Device" and it doesn't say anything like "SNMP timeout on device"?
[04-Oct-2011 11:53:28] <dhopp> what do you mean set the interface speed?
[04-Oct-2011 11:53:29] <pplask> cos if i want to monitor the bandwidth of internet connection, and be warned if it goes to 75%
[04-Oct-2011 11:53:44] <pplask> dhopp, no timeouts, it worked fine
[04-Oct-2011 11:54:10] <dhopp> pplask: when did you add the device?
[04-Oct-2011 11:54:19] <dhopp> in the last few minutes?
[04-Oct-2011 11:54:23] <pplask> yesterday, i added 2 linux server
[04-Oct-2011 11:54:28] <dhopp> hrm
[04-Oct-2011 11:54:45] <pplask> it seems fine
[04-Oct-2011 11:55:12] <dhopp> crap..I got to run to a meeting
[04-Oct-2011 11:55:46] <pplask> well, thanks for your help
[04-Oct-2011 11:56:09] <pplask> if someone knows the answer for my question to monitor bandwitdh
[04-Oct-2011 12:00:01] [disconnected at Tue Oct 4 12:00:01 2011]
[04-Oct-2011 12:00:01] [connected at Tue Oct 4 12:00:01 2011]
[04-Oct-2011 12:00:13] [zenoss-logger (logger bot) has joined #zenoss]
[04-Oct-2011 13:10:58] <jb> grr, hey rocket, you around?
[04-Oct-2011 13:13:34] <jb> i'm having a stupid issue with this DERIVE/COUNTER datasource.
[04-Oct-2011 13:13:58] <jb> the change between poles is over 200+, and the graph is only showing like 1. can't seem to figure out what the issue is.
[04-Oct-2011 16:53:24] <luizzmizz> hi there
[04-Oct-2011 16:53:38] <luizzmizz> i was wondering if this happened to anyone...
[04-Oct-2011 16:53:51] <luizzmizz> the thing is: i did a fresh install of zenoss 3.2.0
[04-Oct-2011 16:54:21] <luizzmizz> and did zenbatchload of a big bunch of devices
[04-Oct-2011 16:54:49] <luizzmizz> and the thing is... they are correctly modeled and that stuff
[04-Oct-2011 16:55:10] <luizzmizz> the interfaces have the correct addresses, including the /22 netmask
[04-Oct-2011 16:55:12] <luizzmizz> BUT
[04-Oct-2011 16:55:31] <luizzmizz> no interface is listed on any IP on the Infrastructure/network page
[04-Oct-2011 16:56:30] <luizzmizz> anyone met the same problem? it's not a bug of 3.2.0, as my previous install was also this version and this issue didn't happen
[04-Oct-2011 16:59:29] <cgibbons> allo!
[04-Oct-2011 17:10:17] <cgibbons> anyone awake? I have questions
[04-Oct-2011 17:12:30] <dhopp> cgibbons: what's your questions?
[04-Oct-2011 17:25:22] <cgibbons> Is there a idiomatic way in zenoss to receive monitoring data out-of-band, and then import it into as performance data? Like say I've got an external processing script that's collecting data from some remote sites that zenoss cant get to... and it saves some data on the server for later processing, in nagios format.
[04-Oct-2011 17:29:54] <tvincent_> Just write a nagios check and graph it.
[04-Oct-2011 17:31:59] kocolosk is now known as kocolosk|away
[04-Oct-2011 17:32:03] <tvincent_> is zenoss able to import Cacti templates?
[04-Oct-2011 17:34:24] <dhopp> it understands cacti commands…I don't think it can import templates…
[04-Oct-2011 20:15:08] <tvincent_> is the zenhub daemon required to be running on a collector?
[05-Oct-2011 00:00:01] [disconnected at Wed Oct 5 00:00:01 2011]
[05-Oct-2011 12:00:02] [connected at Wed Oct 5 12:00:02 2011]
[05-Oct-2011 12:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[05-Oct-2011 12:17:52] <pipeline> perl doens't have a good/easy soap library
[05-Oct-2011 12:18:07] <pipeline> when I need soap in perl I use apache axis2 to generate proxy objects / stubs, then use Inline::Java to actually do the heavy lifting
[05-Oct-2011 12:18:16] <pipeline> setting up the initial build script sucks but from there on out it's ridiculously easy
[05-Oct-2011 12:18:20] <pipeline> inline::java is amazing
[05-Oct-2011 14:33:39] <bseklecki> to manually snmp fetch perf data: % zenperfsnmp run -d [ip]
[05-Oct-2011 14:33:43] <bseklecki> optionally hostname
[05-Oct-2011 14:34:25] <bseklecki> INFO zen.zenperfsnmp: success:0 fail:0 pending:0 todo:0
[05-Oct-2011 14:34:30] <bseklecki> wait, whaaat
[05-Oct-2011 14:35:41] <Praxi> My zenoss server has been neglected for the last month or so (everything was working great). Last night I had something that zenoss monitors go down, and zenoss didn't say a peep about it. I login to the server today, and the dashboard portlet for devices showed "No Records Found". A reboot of the server fixed this. Anyone have any ideas for me to check as to why it did this?
[05-Oct-2011 14:36:09] <bseklecki> did you check if your database was up?
[05-Oct-2011 14:36:57] <bseklecki> tail your log files (tail -1000 -F *.log) >:}
[05-Oct-2011 14:36:58] <Praxi> negative
[05-Oct-2011 14:37:30] <Praxi> I know little about linux or zenoss, but I will start reading to figure out how to check my database log lol
[05-Oct-2011 14:37:51] <bseklecki> future versions of MySQL will supposedly have useful logging
[05-Oct-2011 14:37:57] <bseklecki> version 9x and version 10x
[05-Oct-2011 14:38:03] <bseklecki> 2021 2022
[05-Oct-2011 14:46:28] <Praxi> wow 4 months of mysqld.log, maybe 30 lines
[05-Oct-2011 14:46:34] <Praxi> so not helpful lol
[05-Oct-2011 14:53:31] <jmp242> well, I assume it has to do with the log level, IDK
[05-Oct-2011 15:09:48] <Praxi> so to the best of my ability, it looks like mysql server was up, anything else I can check for?
[05-Oct-2011 15:10:16] <Praxi> oh interesting, my dashboard went back to "No Records Found"
[05-Oct-2011 15:14:00] <jmp242> Did you click on the events tab?
[05-Oct-2011 15:14:31] <Praxi> ya looking through it now, looks like a lot of heartbeat failures for hostname & localhost.localdomain
[05-Oct-2011 15:15:18] <Praxi> hostname being the zenserver
[05-Oct-2011 15:16:14] <jmp242> It sounds like daemons aren't running
[05-Oct-2011 15:16:21] <jmp242> do a zenoss status as the zenoss user
[05-Oct-2011 15:16:24] <jmp242> what does it say?
[05-Oct-2011 15:17:19] <Praxi> a lot of running daemons, does that show ALL daemons or just running daemons?
[05-Oct-2011 15:19:21] <jmp242> shows all of them that I know
[05-Oct-2011 15:19:35] <jmp242> If they're all running
[05-Oct-2011 15:19:49] <jmp242> then IDK - sounds like your Zenoss install is borked in a new and unusual way
[05-Oct-2011 15:20:18] <Praxi> ruhroh thats not good hehe
[05-Oct-2011 15:20:58] <Praxi> thats super interesting, that after a reboot, it worked for a bit.
[05-Oct-2011 15:21:01] <jmp242> If you connect to MySQL
[05-Oct-2011 15:21:09] <jmp242> and show the status table
[05-Oct-2011 15:21:12] <jmp242> is there anything in there/
[05-Oct-2011 15:21:13] <jmp242> ?
[05-Oct-2011 15:22:41] <Praxi> checking
[05-Oct-2011 15:22:45] <Praxi> or trying to hehe
[05-Oct-2011 15:24:35] <jmp242> mysql -uzenoss -pzenoss
[05-Oct-2011 15:24:43] <jmp242> if you use defaults
[05-Oct-2011 15:25:50] <jmp242> this I haven't tried but I think
[05-Oct-2011 15:25:54] <jmp242> use status
[05-Oct-2011 15:26:03] <jmp242> err no
[05-Oct-2011 15:26:05] <jmp242> use events
[05-Oct-2011 15:26:14] <jmp242> select * from status;
[05-Oct-2011 15:26:27] <jmp242> see if you get anything assuming my cribbing from notes isn't wrong
[05-Oct-2011 15:26:58] <Praxi> that did kick out a lot
[05-Oct-2011 15:27:06] <Praxi> looking at it now
[05-Oct-2011 15:29:14] <Praxi> looks like all the stuff I see in the event viewer about heartbeat failures
[05-Oct-2011 15:31:19] <Praxi> hmm theres also a message about snmp being down on the zenhost
[05-Oct-2011 15:32:20] <Praxi> oh interesting
[05-Oct-2011 15:32:44] <Praxi> so the heartbeats are all for those daemons that you had me list earlier
[05-Oct-2011 15:40:52] <dhopp> doh…apparently zenbatchload doesn't like device classes with spaces in the names
[05-Oct-2011 15:41:57] <dhopp> oh wait
[05-Oct-2011 15:42:00] <dhopp> I didn't put quotes around it
[05-Oct-2011 15:42:01] <dhopp> heh
[05-Oct-2011 15:52:51] <jmp242> if you're getting heartbeat failures there's an issue
[05-Oct-2011 15:53:00] <jmp242> but I can't say what it is...
[05-Oct-2011 15:56:27] kocolosk is now known as kocolosk|away
[05-Oct-2011 16:04:50] <dhopp> sweet..zenoss seems to think that one of my lo interfaces on a server is currently doing 10 Gbits / second
[05-Oct-2011 16:15:53] <jiveturkey> dhopp, counter issue on the interface you're polling?
[05-Oct-2011 16:19:56] <dhopp> it was a newly added device that I set the manage IP wrong when I imported it (imported through zenbatchload). so I think zenoss might have just been confused…it cleared on it's own
[05-Oct-2011 16:19:58] <Praxi> do heartbeats work over snmp?
[05-Oct-2011 16:20:06] <bseklecki> why is the SNMP OID 'test' function trying to run snmpv1 >:} that doesn't seem consistent
[05-Oct-2011 16:20:28] <bseklecki> its also mixing up the OID and Hostname argument order
[05-Oct-2011 16:24:57] <Praxi> If I have a device I'm monitoring over SNMP (temp sensor that has multiple probes), I set threshold for datapoints, on some it says (missing) any clues?
[05-Oct-2011 16:25:09] <jiveturkey> dhopp, cool. I've seen plenty of times where a counter rolls over to zero, and there's a super ginormous spike.
[05-Oct-2011 16:26:09] <jiveturkey> Can you snmpget the SNMP values that are showing as missing?
[05-Oct-2011 16:27:07] <Praxi> Was that directed at me jiveturkey ? If so yes, and I still see the graphs built on the datapoints, and they are updating.
[05-Oct-2011 16:28:30] <Praxi> Just the threshold on the ones flagged as 'missing' isn't working. so I get no events.
[05-Oct-2011 16:28:50] <Praxi> I've removed the missing from the threshold, saved it, then reloaded it, and added them back.
[05-Oct-2011 16:31:03] <bseklecki> hmmm
[05-Oct-2011 16:31:32] <Praxi> googling I am seeing others with similar issues, but not a lot of solutions yet.
[05-Oct-2011 16:33:30] <Praxi> This thread seems the most promising thread/15536
[05-Oct-2011 16:33:43] <Praxi> no fix, but sounds like a reasonable explanation of why it keeps breaking
[05-Oct-2011 16:41:31] <bseklecki> enough to make you old
[05-Oct-2011 17:00:52] <bseklecki> Executing command snmpwalk -cFOO -v1 1.3.6.1.4.1.10133.200.200.100.225 against hostname.tld
[05-Oct-2011 17:00:55] <bseklecki> getaddrinfo: 1.3.6.1.4.1.10133.200.200.100.225 No address associated with hostname snmpwalk: Unknown host (1.3.6.1.4.1.10133.200.200.100.225) <--- You're using it wrong. Put the otherside in. Don't force it(R)
[05-Oct-2011 17:12:56] <dhopp> wtf…so we have snmp informant advanced (not the free version) and apparently the oids are all different in advanced..
[05-Oct-2011 17:12:57] <dhopp> argh
[05-Oct-2011 17:13:04] <dhopp> now I need to go hunt down the oids
[05-Oct-2011 18:05:41] <cgibbons> hmmm
[06-Oct-2011 00:00:01] [disconnected at Thu Oct 6 00:00:01 2011]
[06-Oct-2011 00:00:02] [connected at Thu Oct 6 00:00:02 2011]
[06-Oct-2011 00:00:13] [zenoss-logger (logger bot) has joined #zenoss]
[06-Oct-2011 00:02:14] tvincent_ is now known as tvincent
[06-Oct-2011 08:41:32] <konvert> hello
[06-Oct-2011 08:44:09] <konvert> i have this problem, i have a printer in my organisation, i can do an snmp walk so snmp works, i've configured the communities and... correct
[06-Oct-2011 08:44:21] <konvert> but the zenoss doesn't catch any traps
[06-Oct-2011 08:44:41] <konvert> i have another trap receiver and that is receiving traps, what do i do wrong
[06-Oct-2011 08:47:48] <fragfutter> konvert: you need to configure the trap destionation on the printer itself
[06-Oct-2011 08:50:19] <konvert> it is correctly set,
[06-Oct-2011 08:50:39] <konvert> i can set two destinations, i've set one to the zenoss machine and one to the other trap receiver
[06-Oct-2011 08:51:07] <jb> the update.zenoss.org query on 4.0.2 makes things slow :/
[06-Oct-2011 09:45:20] <SDuensin> Greetings.
[06-Oct-2011 09:46:52] <Sashness> hello
[06-Oct-2011 10:03:55] kocolosk|away is now known as kocolosk
[06-Oct-2011 10:07:44] <dhopp> argh…I can't get any multi graph reports to go to the "printable" format
[06-Oct-2011 10:08:09] <dhopp> no matter what I do I get:
[06-Oct-2011 10:08:09] <dhopp> Type:
[06-Oct-2011 10:08:09] <dhopp> Value: invalid literal for int() with base 10: 'undefined'
[06-Oct-2011 10:11:05] <rocket> anyone know what Hackman238 is up to? Is he travelling to the openstack conference?
[06-Oct-2011 10:11:23] <dhopp> He's in NY I believe..but I don't know what for
[06-Oct-2011 10:11:32] <dhopp> rocket: do you have any idea on my error?
[06-Oct-2011 10:11:47] <rocket> dhopp: look at the event.log
[06-Oct-2011 10:11:56] <rocket> the relevant traceback is most likely there
[06-Oct-2011 10:12:18] <rocket> but my guess is one datapoint is not coming through as an int
[06-Oct-2011 10:12:54] <rocket> eg zenrender is probably spitting out undefined for that value and the report is expecting a numerical value
[06-Oct-2011 10:13:12] <dhopp> http://pastebin.com/FqJLuCZ9
[06-Oct-2011 10:13:44] <dhopp> that looks like it's bombing trying to get the number of graphs
[06-Oct-2011 10:13:49] <dhopp> which would be a bug
[06-Oct-2011 10:15:04] <rocket> dhopp: there is a good chance that is a bug with the report
[06-Oct-2011 10:16:29] <rocket> zenoss version?
[06-Oct-2011 10:16:47] <dhopp> rocket: All I did was create a collection that is a device class with 2 servers…then added a Graph Definition and added the datapoint of "cpuPercentProcessorTime_cpuPercenProcessorTime" and then defined a Graph Group and set it to all servers on 1 graph
[06-Oct-2011 10:16:49] <dhopp> 3.2.0
[06-Oct-2011 10:18:22] <rocket> well the source file pt hasnt been touched in a year
[06-Oct-2011 10:19:18] <dhopp> ok..then what did I do wrong? I can see the graph report
[06-Oct-2011 10:19:26] <dhopp> it only happens when I click the "printable" button
[06-Oct-2011 10:19:40] <rocket> dhopp: most likely you did nothing wrong
[06-Oct-2011 10:19:56] <rocket> dhopp: its probably something that has not been tested recently etc
[06-Oct-2011 10:20:04] <dhopp> rocket: ugh
[06-Oct-2011 10:20:13] <rocket> if you can create a defect report with the steps
[06-Oct-2011 10:20:24] <rocket> you used we can try and reproduce it
[06-Oct-2011 10:21:02] <dhopp> I know there is a bug with Graph Reports (not multi) where printable will always show you hourly graphs..regardless of your zoom level…I was trying to see if multigraphs suffered the same problem
[06-Oct-2011 10:21:36] <konvert> i have this problem, i have a printer in my organisation, i can do an snmp walk so snmp works, i've configured the communities and... correct
[06-Oct-2011 10:21:53] <konvert> but the zenoss doesn't catch any traps
[06-Oct-2011 10:21:59] <konvert> i have another trap receiver and that is receiving traps, what do i do wrong
[06-Oct-2011 10:22:44] <rocket> konvert: you will need to start a tcpdump etc to see if the server is recieving traps at all from that interface
[06-Oct-2011 10:23:11] <rocket> konvert: just because a device is listening to snmp requests is not enough to troubleshoot that issue
[06-Oct-2011 10:23:16] <konvert> but i have an other trap receiver that is receiving the traps
[06-Oct-2011 10:23:21] <rocket> konvert: it really tells you basically nothing
[06-Oct-2011 10:23:39] <psyklone> konvert - if you're on 3.2 and haven't installed the patches, that might be it
[06-Oct-2011 10:23:50] <konvert> have installed the patche
[06-Oct-2011 10:23:51] <konvert> s
[06-Oct-2011 10:23:53] <rocket> konvert: so? there could be a firewall issue. your trap sender may not be configured properly, zentrap may not be running
[06-Oct-2011 10:23:55] <psyklone> one of them specifically addresses a trap problem that jane found
[06-Oct-2011 10:25:03] <rocket> konvert: I would do this
[06-Oct-2011 10:25:06] <konvert> rocket: i receive other traps, there is no firewall issue everything is open, trap sender is correct
[06-Oct-2011 10:25:19] <rocket> zentrap stop && zentrap run -v 10
[06-Oct-2011 10:25:54] <rocket> sounds like you need to determine if the trap packet is even getting to the zenoss server first though
[06-Oct-2011 10:26:01] <rocket> thus the use of tcpdump etc
[06-Oct-2011 10:27:09] <konvert> and how do i do tcpdump? i'm noob in that kind of stuff
[06-Oct-2011 10:27:27] <rocket> as root
[06-Oct-2011 10:27:59] <rocket> tcpdump -i <interface_you_would_expect_the_packet_to_come_in_on>
[06-Oct-2011 10:28:34] <rocket> konvert: I would recommend a web search for filter options for tcpdump to limit the data to certain ips .. eg src and dst == the trap sender
[06-Oct-2011 10:30:59] <konvert> thx rocket will look into that
[06-Oct-2011 10:32:11] <dhopp> rocket: ticket #7875
[06-Oct-2011 10:40:59] <rocket> it should get evaluated as part of todays defect review meeting
[06-Oct-2011 10:41:06] <dhopp> thanks
[06-Oct-2011 10:42:47] <humanix> hi there guys
[06-Oct-2011 10:42:56] <humanix> hi rmatte Hackman238
[06-Oct-2011 10:48:15] <jiveturkey> Howdy. I asked this a day or two ago, but I'll give it another shot.
[06-Oct-2011 10:48:38] <jiveturkey> Having a weird issue with Zenoss 2.5.2. Showing "Lost connection to the server" on the Dashboard and not loading anything in the Device Issues portlet.
[06-Oct-2011 10:49:52] <rocket> jiveturkey: inspect $ZENHOME/log/event.log
[06-Oct-2011 10:50:03] <rocket> jiveturkey: there should be a traceback there most likely
[06-Oct-2011 10:50:41] <rocket> jiveturkey: you probably could try running zenchkrels -r -x1 to try and repair relationship issues that I am guessing is causing this
[06-Oct-2011 10:52:05] <dhopp> rocket: I think what might be happening (and this would be similar to the printable graph reports only showing hourly) is numGraphs isn't getting set. Is there a good way to debug so I can see what is happening? doing zopectl fg doesn't seem to show me anything of significance
[06-Oct-2011 10:52:30] <jiveturkey> rocket: I'll have a look at event.log. By the way, the Event Console *is* working, and I can access mysql (which is on another box) from the main Zenoss box.
[06-Oct-2011 10:55:34] <jiveturkey> RE zenchkrels, I'm not sure I dare at this point. I know Zeo has a lot of weird relationship issues on this box that we've already spent man weeks working on...
[06-Oct-2011 11:13:08] <konvert> rocket: tcpdump done and i see the traps on the machine
[06-Oct-2011 11:13:09] <rocket> jiveturkey: are you an enterprise customer?
[06-Oct-2011 11:14:00] <rocket> konvert: zentrap stop && zentrap run -v 10
[06-Oct-2011 11:14:11] <rocket> that will run zentrap in the foreground with debugging turned on
[06-Oct-2011 11:14:33] <rocket> konvert: it will give you an idea if zentrap is actually recieving the trap at that point which it should
[06-Oct-2011 11:15:20] <konvert> what does zentrap run -v 10 do?
[06-Oct-2011 11:15:43] <rocket> debug + foreground as I just said
[06-Oct-2011 11:17:08] <konvert> ok done it and now?
[06-Oct-2011 11:17:55] <rocket> wait for a trap. you should watch your other trap reciever to see if you have recieved one there
[06-Oct-2011 11:18:15] <rocket> you should see zentrap logging the traps it recieved at this point
[06-Oct-2011 11:19:07] <konvert> that 10 in the command is that 10 seconds?
[06-Oct-2011 11:19:19] <wayne___> verbosity
[06-Oct-2011 11:19:56] <konvert> because i can see some lines comming up but then it stops
[06-Oct-2011 11:20:43] <rocket> you may need to add --cycle
[06-Oct-2011 11:21:04] <rocket> to the command line .. I forgot that zentrap needs that option to keep running in the foreground
[06-Oct-2011 11:21:06] <cgibbons> hmmm
[06-Oct-2011 11:21:27] <rocket> cgibbons: How are you?
[06-Oct-2011 11:21:50] <cgibbons> tired
[06-Oct-2011 11:22:09] <cgibbons> i keep forgetting i'm logged in here. i even had a zenoss thing to figure out @ some point
[06-Oct-2011 11:22:22] <rocket> well if you are leading a full life that is a good thing 
[06-Oct-2011 11:22:52] <rocket> cgibbons: what were you trying to figure out?
[06-Oct-2011 11:25:04] <konvert> ok i see the trap in there
[06-Oct-2011 11:25:41] <konvert> i get an error : unhandled error in deferred
[06-Oct-2011 11:25:41] <rocket> so if you see a trap recieved ... then most likely you have a configuration on the event class that is transforming or dropping the event
[06-Oct-2011 11:26:08] <rocket> unhandled error in deferred sounds like you have NOT applied the patches that psyklone mentioned for that
[06-Oct-2011 11:26:11] <konvert> and then a traceback with some files and line numbers
[06-Oct-2011 11:26:29] <konvert> i have applied two patches after upgrade
[06-Oct-2011 11:27:01] <konvert> can someone give me the patch command?
[06-Oct-2011 11:27:37] <psyklone> konvert, you're probably missing the one that fixes this issue
[06-Oct-2011 11:27:43] <rocket> zenpatch 26337
[06-Oct-2011 11:27:46] <psyklone> zenpatch 26337
[06-Oct-2011 11:27:53] <rocket> then restart zentrap
[06-Oct-2011 11:27:54] <psyklone> yeah, what rocket said
[06-Oct-2011 11:28:09] <psyklone> dhopp, are you around?
[06-Oct-2011 11:28:13] <dhopp> I am
[06-Oct-2011 11:28:56] <psyklone> can you pop open $ZENHOME/Products/ZenModel/skins/zenmodel/viewMultiGraphReport.pt ...
[06-Oct-2011 11:29:15] <dhopp> ok
[06-Oct-2011 11:30:10] <dhopp> and now what?
[06-Oct-2011 11:30:19] <psyklone> depending on what version you're ... look around line 33 you'll have a line that ends with ZenGraphs.length
[06-Oct-2011 11:30:34] <dhopp> CreateFormField(printableForm, 'numGraphs', ZenGraphs.length);
[06-Oct-2011 11:30:38] <psyklone> yep, that one
[06-Oct-2011 11:30:44] <psyklone> put // in front to comment it
[06-Oct-2011 11:30:52] <psyklone> then save, and restart zopectl
[06-Oct-2011 11:31:02] <psyklone> then try your print again
[06-Oct-2011 11:31:18] <dhopp> woot!
[06-Oct-2011 11:31:19] <dhopp> it worked
[06-Oct-2011 11:31:37] <dhopp> now just to get it to print the zoom level I want
[06-Oct-2011 11:31:40] <dhopp> rather than hourly
[06-Oct-2011 11:32:30] <psyklone> i had a look, and you can actually path (from the report) directly to the viewMultiGraphReportClean page template and it works, so that one isn't the problem ... it's the viewMultiGraphReport template and the way it's trying to determine the number of graphs contained
[06-Oct-2011 11:32:44] <psyklone> that it doesn't look like it needs, so ... this should do it
[06-Oct-2011 11:32:53] <dhopp> sweet thanks
[06-Oct-2011 11:33:02] <dhopp> any idea on why drange doesn't get passed?
[06-Oct-2011 11:33:11] <cgibbons> rocket: we've got some stuff here that collects nagios data remotely, and then forwards it back to a central place. push versus pull. i was wondering if there was a way to import that data into zenoss via an api, or if I'll need to write a little collector that eats the data from a file or something.
[06-Oct-2011 11:33:36] <psyklone> dhopp - hmm, i'm not familiar with that
[06-Oct-2011 11:33:38] <konvert> sweet guys that did the trick
[06-Oct-2011 11:33:47] <psyklone> *high five*
[06-Oct-2011 11:33:59] <dhopp> psyklone: 7873
[06-Oct-2011 11:34:42] <dhopp> when you end up at viewGraphReportClean you can append ?drange=blah to the url and then it will graph the range you asked for
[06-Oct-2011 11:35:10] <dhopp> so it seems like drange is always getting passed with the value that signifies hourly
[06-Oct-2011 11:35:24] <psyklone> dhopp, i might be able to have a look at that later. i'm out tomorrow so i've got some stuff i have to get knocked out.
[06-Oct-2011 11:35:33] <dhopp> no problem
[06-Oct-2011 11:35:53] <dhopp> I've been trying to look through the code..but I'm trying to figure out how all the pieces tie together
[06-Oct-2011 11:36:09] <psyklone> i hear that's dangerous
[06-Oct-2011 11:36:11] <psyklone> 
[06-Oct-2011 11:36:15] <rocket> cgibbons: you could write a quick nagios parser script and use a command template pretty easily
[06-Oct-2011 11:47:53] <cgibbons> yeah
[06-Oct-2011 12:00:01] [disconnected at Thu Oct 6 12:00:01 2011]
[06-Oct-2011 12:00:02] [connected at Thu Oct 6 12:00:02 2011]
[06-Oct-2011 12:00:13] [zenoss-logger (logger bot) has joined #zenoss]
[06-Oct-2011 12:26:17] <billc_> If the mysql event data base fills up, can I simply truncate the tables if I do not care about historical events?
[06-Oct-2011 12:26:49] <psyklone> yes
[06-Oct-2011 12:38:40] <Praxi> has anyone played with sending a zenoss SMS via google voice?
[06-Oct-2011 12:39:23] <Praxi> was reading some other zenoss SMS stuff off google, and most of them seem to involve hardware
[06-Oct-2011 12:41:45] <tvincent> good morning
[06-Oct-2011 12:41:51] <tvincent> playdom01 is the password to the gui
[06-Oct-2011 12:44:02] <psyklone> you might want to change that password now, though.
[06-Oct-2011 12:44:14] <Praxi> lol
[06-Oct-2011 12:44:16] <tvincent> i just saw that. shit. god damn i need coffee
[06-Oct-2011 12:44:33] <psyklone> something more like: *******
[06-Oct-2011 13:52:42] <jb> rocket: hey
[06-Oct-2011 14:12:31] <nexex> Praxi: http://code.google.com/p/pygooglevoice/
[06-Oct-2011 14:13:29] <Praxi> ahh nice thank you nexex
[06-Oct-2011 14:14:03] <Praxi> have you tried it before? with zenoss?
[06-Oct-2011 14:16:16] <Praxi> I'm thinking I want to use one of those phone company usb modems, in case the building loses power or what not.
[06-Oct-2011 14:20:18] <Praxi> doh just sec, reboot
[06-Oct-2011 14:23:24] <rocket> Hey jb
[06-Oct-2011 14:24:05] <Sashness> uhh ohh
[06-Oct-2011 14:29:40] <rmatte> does anyone know if there's a way to identify what VLAN an interface is on via SNMP? I can't find much on the subject.
[06-Oct-2011 14:30:05] <tvincent> rmatte: if you are querying the switch you might.
[06-Oct-2011 14:30:27] <rmatte> what do you mean?
[06-Oct-2011 14:30:31] <tvincent> but you wont be able to do it from the host unless you have VLAN tagging set on the interface itself.
[06-Oct-2011 14:31:09] <dhopp> rmatte: http://www.cisco.com/en/US/tech/tk648/tk362/technologies_configuration_example09186a008015773e.shtml
[06-Oct-2011 14:31:27] <dhopp> oh wait
[06-Oct-2011 14:31:30] <rmatte> dhopp: I was just looking at that, but that isn't what I want
[06-Oct-2011 14:31:32] <dhopp> that just finds vlans
[06-Oct-2011 14:31:38] <jbk> rmatte: cisco switch?
[06-Oct-2011 14:31:49] <rmatte> I'm hoping there's some OID that I can poll for specific interfaces and find out what vlan they are on
[06-Oct-2011 14:31:53] <rmatte> cisco switches, yes
[06-Oct-2011 14:32:05] <jbk> there is
[06-Oct-2011 14:32:07] <jbk> lemme dig it up
[06-Oct-2011 14:32:11] <rmatte> thanks
[06-Oct-2011 14:33:02] <rmatte> It's for an interface report that I'm putting together for one of our engineers, I know how to grab the rest of the data, just not that.
[06-Oct-2011 14:33:34] <jbk> the table is .1.3.6.1.4.1.9.9.46.1.3.1
[06-Oct-2011 14:33:47] <jbk> vlan name is 1.14
[06-Oct-2011 14:34:03] <jbk> err
[06-Oct-2011 14:34:03] <jbk> sorry
[06-Oct-2011 14:34:07] <jbk> that's the table of configured vlans
[06-Oct-2011 14:34:22] <rmatte> yeh
[06-Oct-2011 14:35:10] <dhopp> rmatte: http://ciscofaq.blogspot.com/2005/07/how-do-i-set-ports-vlan-membership.html
[06-Oct-2011 14:35:19] <rmatte> dhopp: that's how to set, not to get
[06-Oct-2011 14:35:25] <dhopp> doh!
[06-Oct-2011 14:35:49] <dhopp> I need to look more closely at the pages I'm finding
[06-Oct-2011 14:35:50] <dhopp> heh
[06-Oct-2011 14:35:54] <rmatte>
[06-Oct-2011 14:36:03] <rmatte> trust me, I've googled the crap out of this already
[06-Oct-2011 14:36:06] <dhopp> although..if you can set it there, why can't you read it?
[06-Oct-2011 14:37:04] <rmatte> hmmmm, you may have something there...
[06-Oct-2011 14:37:30] <rmatte> I just walked 1.3.6.1.4.1.9.9.68.1.2.2.1.2 and it did output data for each of the physical interfaces
[06-Oct-2011 14:38:45] <rmatte> I don't know if the numbers directly correspond to the vlan numbers though, checking that now...
[06-Oct-2011 14:40:38] <jbk> they should
[06-Oct-2011 14:40:46] <jbk> ftp://ftp.cisco.com/pub/mibs/v2/CISCO-VLAN-MEMBERSHIP-MIB.my
[06-Oct-2011 14:40:47] <jbk> that's the MIB
[06-Oct-2011 14:41:39] <rmatte> the real question is, are the numbers literally the vlan numbers or are they snmp indexes?
[06-Oct-2011 14:41:54] <jbk> the index is the index of the port
[06-Oct-2011 14:41:59] <rmatte> i.e: .1.3.6.1.4.1.9.9.68.1.2.2.1.2.29 = INTEGER: 5
[06-Oct-2011 14:42:02] <rmatte> right
[06-Oct-2011 14:42:07] <rmatte> but does that mean that's vlan 5
[06-Oct-2011 14:42:11] <rmatte> or that that's an index of the vlan
[06-Oct-2011 14:42:12] <jbk> yes
[06-Oct-2011 14:42:32] <jbk> it's the actual vlan value
[06-Oct-2011 14:42:38] <jbk> it's not an index into another table
[06-Oct-2011 14:42:38] <rmatte> awesome, thanks
[06-Oct-2011 14:42:57] <jbk> note, it appears (at least on some switches) that trunk ports don't show up
[06-Oct-2011 14:43:08] <rmatte> that's fine
[06-Oct-2011 14:43:10] <jbk> (looking on one of my 6509s)
[06-Oct-2011 14:43:14] <rmatte> more concerned with user ports in this case
[06-Oct-2011 14:44:22] <jbk> are you doing cisco voip phones?
[06-Oct-2011 14:45:03] <rmatte> not sure what the engineer wants it for
[06-Oct-2011 14:45:57] <rmatte> he just requested a report with specific values in it for all interfaces on a certain set of switches (switch, port, vlan, speed, duplex, label)
[06-Oct-2011 14:46:18] <rmatte> I'm just going to throw together a quick little bash script to do it
[06-Oct-2011 14:46:23] <rmatte> should only take a few minutes to code
[06-Oct-2011 14:47:04] <dhopp> rmatte: why use bash? Is there a reason you wouldn't do it in Zenoss?
[06-Oct-2011 14:47:45] <rmatte> dhopp: because Zenoss doesn't have the duplex/vlan
[06-Oct-2011 14:48:08] <rmatte> otherwise I would
[06-Oct-2011 14:48:10] <billc_> Is there a way to shrink the size of the Zenoss database... /usr/lkocal/zenoss/zenoss/var/Data.fs is 39G
[06-Oct-2011 14:48:22] <rmatte> billc_: yes, you need to run zeopack
[06-Oct-2011 14:48:39] <billc_> thank you !
[06-Oct-2011 14:48:52] <dhopp> billc_: look in /etc/cron.weekly/zenoss for the command
[06-Oct-2011 14:49:16] <dhopp> not sure why they don't have that at least daily
[06-Oct-2011 14:49:53] <rmatte> it never seems to run automatically for me even though I do have the cron file
[06-Oct-2011 14:49:57] <rmatte> I do it by hand
[06-Oct-2011 14:50:39] <dhopp> rmatte: I scheduled in in zenoss crontab…I had to write a bash script that sources the zenoss environment then runs the command
[06-Oct-2011 14:50:43] <dhopp> so far it seems to work
[06-Oct-2011 14:51:19] <rmatte> dhopp: this script is probably failing for some reason
[06-Oct-2011 14:51:39] <dhopp> you mean the one in /etc/cron.weekly?
[06-Oct-2011 14:51:43] <rmatte> yeh
[06-Oct-2011 14:51:53] <rmatte> root@gen01:/etc/cron.weekly# ./zenoss
[06-Oct-2011 14:52:01] <rmatte> eugh, IRC
[06-Oct-2011 14:52:06] <rmatte> home/zenoss/.bashrc: 15: shopt: not found
[06-Oct-2011 14:52:10] <rmatte> etc/bash_completion: 32: [[: not found
[06-Oct-2011 14:52:14] <rmatte> etc/bash_completion: 38: [[: not found
[06-Oct-2011 14:52:18] <rmatte> etc/bash_completion: 50: Bad substitution
[06-Oct-2011 14:52:20] <rmatte> lol
[06-Oct-2011 14:52:26] <dhopp> well there we go...
[06-Oct-2011 14:52:40] <dhopp> I just fixed it by writing a bash script myself :-)
[06-Oct-2011 14:52:49] <rmatte> yeh, I'll do the same when I have a second
[06-Oct-2011 14:53:16] <dhopp> #!/bin/bash
[06-Oct-2011 14:53:16] <dhopp> . ~zenoss/.bashrc
[06-Oct-2011 14:53:17] <dhopp> $ZENHOME/bin/zeopack -h localhost -p 8100; cp $ZENHOME/var/Data.fs.old $ZENHOME/backups/Data.fs.`date +%m%d%H%M`
[06-Oct-2011 14:53:19] <dhopp> there you go :-)
[06-Oct-2011 14:53:44] <dhopp> then I have a cron that removes files older then 7 days in backups
[06-Oct-2011 14:54:10] <billc_> is port 8100 default for ZopeDB
[06-Oct-2011 14:54:24] <dhopp> billc_: yes
[06-Oct-2011 14:54:29] <rmatte> ah, I don't keep backups, I just delete the .old, if something goes wonky we have backups of the entire image, so we just restore from those
[06-Oct-2011 14:54:56] <dhopp> I do whole backups every night and zeopack every 6 or 8 hours
[06-Oct-2011 14:55:13] <dhopp> I'm paranoid about corruption
[06-Oct-2011 14:55:24] <rmatte> hehe
[06-Oct-2011 14:55:36] <rmatte> I've only had one corruption issue so far
[06-Oct-2011 14:56:02] <dhopp> I read about people in the forums having all sorts of issues so I'm paranoid
[06-Oct-2011 14:56:03] <rmatte> it sucks when it happens because a lot of the time it's impossible to fix
[06-Oct-2011 14:56:26] <rmatte> when zope's broke it's broke
[06-Oct-2011 15:00:49] <rmatte> jbk: thanks for the help with finding that OID
[06-Oct-2011 15:00:59] <jbk> np
[06-Oct-2011 15:01:14] <rmatte> I won this last night in a draw at a pub: http://a1.sphotos.ak.fbcdn.net/hphotos-ak-snc7/293533_10150344127249547_513149546_8000609_790963885_n.jpg
[06-Oct-2011 15:01:24] <rmatte> Good timing since my old one died on me (fridge)
[06-Oct-2011 15:04:44] <jbk> nice
[06-Oct-2011 15:05:03] * jbk got screwed out of part of his lunch today
[06-Oct-2011 15:05:44] <jbk> went to get up and get more ketchup, and when i got back to my table
[06-Oct-2011 15:05:53] <jbk> tehy had taken my lunch away thinking i had left
[06-Oct-2011 15:06:11] <rmatte> holy crap, I'd lose it
[06-Oct-2011 15:07:08] <dhopp> I would be a very unhappy camper
[06-Oct-2011 15:07:29] <jbk> yeah, so avoid the fuddruckers on greens rd if you're in the houston area
[06-Oct-2011 15:07:40] <rmatte> haha
[06-Oct-2011 15:07:42] <rmatte> noted
[06-Oct-2011 15:07:54] <dhopp> that's a shame…I'm a fan of fuddruckers
[06-Oct-2011 15:08:29] <rmatte> fuddruckers looks good, we don't have any of those around here
[06-Oct-2011 15:09:08] <dhopp> rmatte: it has pretty decent hamburgers for a fast food place…and the milk shakes are delicious
[06-Oct-2011 15:09:30] <jbk> they were bigger several years ago
[06-Oct-2011 15:09:33] <dhopp> as you can see I'm very health conscious
[06-Oct-2011 15:09:34] <jbk> then a bunch of them closed
[06-Oct-2011 15:09:41] <jbk> this one is always packed though
[06-Oct-2011 15:09:48] <rmatte> yeh, most fast food places have really crappy hamburgers
[06-Oct-2011 15:10:09] <rmatte> all this talk about food is making me hungry lol
[06-Oct-2011 15:10:33] <dhopp> me too and I just ate lunch
[06-Oct-2011 15:10:39] <rmatte> lol
[06-Oct-2011 15:10:49] <rmatte> I haven't had anything to eat yet today, think I'll order something
[06-Oct-2011 15:10:49] <dhopp> although really now I want a milk shake
[06-Oct-2011 15:11:30] <rmatte> lol
[06-Oct-2011 15:14:03] <jbk> i had to run and get some ear plugs @ gc (closest place), so figured i'd go out for lunch for once
[06-Oct-2011 15:14:51] <rmatte> there, just ordered some spaghetti
[06-Oct-2011 15:15:47] <rmatte> "Expected deliverytime:
[06-Oct-2011 15:15:49] <rmatte> 16:10
[06-Oct-2011 15:15:55] <rmatte> are they flying it in from italy?
[06-Oct-2011 15:16:03] <jbk> i hope you're on the east coast
[06-Oct-2011 15:16:14] <jbk> though even then that's a bit much
[06-Oct-2011 15:16:28] <rmatte> (It's 15:17 right now)
[06-Oct-2011 15:16:38] <rmatte> I am on the east coast
[06-Oct-2011 15:16:39] <rmatte> lol
[06-Oct-2011 15:16:48] <dhopp> jbk: if I'm waiting an hour for spaghetti it better be REALLY good spaghetti
[06-Oct-2011 15:16:55] <rmatte> tell me about it
[06-Oct-2011 15:17:02] <rmatte> guess I'll wait and see lol
[06-Oct-2011 15:30:48] <dhopp> ugh..I really need more then 24 hours in a day to get Zenoss up and running for us
[06-Oct-2011 15:34:36] <tvincent> dhopp: just think how nice it will be when it gets done.
[06-Oct-2011 15:36:11] <dhopp> tvincent: that's what I'm hoping..but I haven't opened our dev environment up to very many people so when I do that I'm sure I'm going to get "why can't it do this, or that" "our existing monitoring system does this, we need to do this"…nevermind that fact that our existing monitoring system is like 10 years old (no joke, there is comments in the code with the year 2001) and almost everything it can do that Zenoss can't was a custom thing
[06-Oct-2011 15:37:15] <dhopp> and I'm going to get skewered on Zenoss lack of ability to give the graph reports a date range
[06-Oct-2011 15:43:00] <dhopp> hmm and apparently you can't copy reports
[06-Oct-2011 15:43:15] <jiveturkey> You can copy them through the ZMI...
[06-Oct-2011 15:43:21] <dhopp> jiveturkey: ahh
[06-Oct-2011 15:43:23] <jiveturkey> I wouldn't spread that around to your users, however...
[06-Oct-2011 15:43:39] <dhopp> jiveturkey: I think me and about 1 or 2 other people will have Manager access anyway :-)
[06-Oct-2011 15:43:48] <jiveturkey> Good, good.
[06-Oct-2011 15:44:39] <dhopp> jiveturkey: although in our existing monitoring everybody in our MS dept can create checks and I'm nervous about that because I can so see somebody editing the template a /server/Linux and then borking every linux server
[06-Oct-2011 15:44:59] <dhopp> and then my ass getting woken up to fix it
[06-Oct-2011 15:45:00] <dhopp> heh
[06-Oct-2011 15:46:18] <dhopp> although I wonder why copying isn't part of the main UI
[06-Oct-2011 15:58:54] <rmatte> dhopp: copying is part of the main UI, it's just called Override
[06-Oct-2011 15:58:56] <rmatte> and it sucks
[06-Oct-2011 15:59:13] <rmatte> They're renaming it to Copy / Override in future versions, but you can only copy 1 template at a time with it
[06-Oct-2011 15:59:22] <rmatte> they need to make it so that you can copy multiple objects at a time
[06-Oct-2011 15:59:43] <rmatte> oh you meant copying reports
[06-Oct-2011 15:59:48] <rmatte> yeh, that'd be cool if you could
[06-Oct-2011 16:11:51] <dhopp> rmatte: not for reports…there is no copy
[06-Oct-2011 16:12:32] <dhopp> rmatte: how do you guys get around the no date range ability when looking at graphs?
[06-Oct-2011 16:12:36] <rmatte> yeh I know, thought you were talking about something else
[06-Oct-2011 16:12:48] <rmatte> dhopp: we don't currently
[06-Oct-2011 16:12:57] <dhopp> rmatte: damn
[06-Oct-2011 16:13:07] <rmatte> dhopp: Hackman has a ZenPack that does it, but it's only compatible with 2.5
[06-Oct-2011 16:13:24] <dhopp> rmatte: I'm looking at doing something with jasper (which is essentially what SD does) but that is a whole different project in itself
[06-Oct-2011 16:13:34] <rmatte> we'll be migrating it to 3.x
[06-Oct-2011 16:13:50] <dhopp> rmatte: yeah but he needs to send it to somebody to look at it :-)
[06-Oct-2011 16:14:09] <rmatte> yeh, he'll be sending it to me eventually
[06-Oct-2011 16:38:52] <rmatte> dhopp: we're looking at jasper too... we need some sort of community fueled project for integrating jasper with Zenoss
[06-Oct-2011 16:39:07] <rmatte> the ZCA will most likely get something rolling for that eventually
[06-Oct-2011 16:39:12] <rmatte> it was mentioned in a previous meeting
[06-Oct-2011 16:54:02] <dhopp> rmatte: yeah I sort of said I would take a look at it for the ZCA but I just haven't had much time until now
[06-Oct-2011 16:54:15] <dhopp> and even then I still don't have much tie
[06-Oct-2011 16:54:15] <dhopp> time
[06-Oct-2011 17:11:13] <bseklecki> hmmmm non-numeric SNMP OIDs do not seem to work
[06-Oct-2011 17:11:40] <bseklecki> HOST-RESOURCES-MIB::hrSystemProcesses.0
[06-Oct-2011 17:12:34] <hmp> install the mib
[06-Oct-2011 17:17:49] <bseklecki> yea i guess; when i run the test it just exec()'s snmpget(1) which has $MIBDIR set right
[06-Oct-2011 17:18:16] <bseklecki> does the background daemon not use snmplib?
[06-Oct-2011 17:21:45] <hmp> zenoss uses it's own userland snmpget/walk, and has it's own mibs
[06-Oct-2011 17:22:14] <hmp> you dont even have to have snmp installed on the system
[06-Oct-2011 17:27:49] <bseklecki> i figured 'test' was exec()'ing the net-snmp bunblded w/ zenoss
[06-Oct-2011 17:28:31] <hmp> oh, you meant test in the gui
[06-Oct-2011 17:28:34] <hmp> yeah, does that
[06-Oct-2011 17:28:49] <bseklecki> yea, I'm looking now, i don't see any snmpget(1) in /opt/zenoss so must be using my system's one
[06-Oct-2011 17:29:43] <bseklecki> I know net-SNMP's mibs have depdencies; probably need HOST-RESOURCES-TYPES.txt in there, too
[06-Oct-2011 17:30:14] <bseklecki> hoepfully zenmib(8) will speak up if dependencies aren't met
[06-Oct-2011 17:37:06] kocolosk is now known as kocolosk|away
[06-Oct-2011 17:42:12] <dhopp> I thought MIBs were only used for snmptrap resolution
[06-Oct-2011 17:42:20] <dhopp> in zenoss
[06-Oct-2011 17:44:55] <psyklone> dhopp - that's correct
[06-Oct-2011 17:59:31] <jbk> yes, however to map a trap name to an oid
[06-Oct-2011 17:59:36] <jbk> you might need multiple mibs
[06-Oct-2011 18:00:09] <jbk> for example, the 'enterprises' oid is defined in SNMPv2-SMI (IIRC)
[06-Oct-2011 18:00:27] <jbk> so your oid is 'enterprises.foo', your MIB probably imports the value of enterprises from that other MIB
[06-Oct-2011 18:18:45] kocolosk|away is now known as kocolosk
[06-Oct-2011 18:28:28] <jdross> I have modelers running as a nightly cron job rather than a deamon, and so i constantly get heartbeat failures for the zenmodeler daemons. is there a way to fix this?
[06-Oct-2011 21:44:29] dookie is now known as dookie_
[06-Oct-2011 21:45:08] <dookie_> i searched all over for an answer ... i put an asterisk * in a search box, and now it says "Loading..." and is stuck there
[06-Oct-2011 21:45:29] <dookie_> i did this yesterday and re-installed, but now i'm much further along and don't want to have to do that ;-) tried restarting zenoss-init, to no avail
[06-Oct-2011 21:45:44] <dookie_> any ideas? has anyone seen this before? i'm on the latest zenoss core version ...
[06-Oct-2011 21:50:49] meddle- is now known as bender[a]
[06-Oct-2011 22:02:20] <dookie_> looks like a complete browser cache clearing fix it
[06-Oct-2011 22:02:24] <dookie_> sorry to be a bother, guys
[06-Oct-2011 22:03:05] <dookie_> if anyone's interested, the error was: Exception Cannot find "". KeyError: ''
[06-Oct-2011 23:07:50] <rocket> jdross: yes
[06-Oct-2011 23:11:00] <rocket> jdross: there is an option to clear all heartbeats in the ui. keep in mind that you will want to make sure the other daemons you do want running are running before pushing that button.
[07-Oct-2011 00:00:01] [disconnected at Fri Oct 7 00:00:01 2011]
[07-Oct-2011 00:00:02] [connected at Fri Oct 7 00:00:02 2011]
[07-Oct-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[07-Oct-2011 00:09:20] tvincent_ is now known as tvincent
[07-Oct-2011 00:10:48] kocolosk is now known as kocolosk|away
[07-Oct-2011 01:04:07] <mahimahi1> Is there a command line to add a user to the zenoss system? thanks
[07-Oct-2011 03:43:24] <Diddi> hi! I'm trying to get the fping zenpack working. I've installed the pack, and bound the template to /server/Linux. When I check graphs for a device, they're there (packet loss, latency, count etc..) but they're empty
[07-Oct-2011 03:43:51] <Diddi> just as if nothing is running to fill the graphs up. The template is enabled, and testing it shows everything is working
[07-Oct-2011 03:44:02] <Diddi> any ideas?
[07-Oct-2011 03:44:22] <fragfutter> how long did you wait
[07-Oct-2011 03:44:54] <Diddi> a couple of days
[07-Oct-2011 03:44:55] <Diddi>
[07-Oct-2011 03:45:04] <fragfutter> ok.
[07-Oct-2011 03:45:08] <fragfutter> anything in the zencommand log?
[07-Oct-2011 03:45:16] <fragfutter> can you execute fping as the user zenoss?
[07-Oct-2011 03:46:18] <Diddi> yes, when editing the template, I can "test against a device", and that test shows that fping is executed and I get the expected output
[07-Oct-2011 03:47:56] <fragfutter> no more ideas. I would check the logs next.
[07-Oct-2011 03:48:01] <Diddi> all I see in zencommand logfile is a lot of "INFO zen.zencommand: config: <Products.ZenRRD.zencommand.DeviceConfig instance at 0xa96432c>"
[07-Oct-2011 03:48:26] <Diddi> no idea what it means, but RRD sound relevant.
[07-Oct-2011 04:05:24] <Diddi> hm, I also have "Deleting command z.skynet.cb from /usr/sbin/fping [...]". No idea what that means, but the command seems to be what the template runs
[07-Oct-2011 04:06:25] <Diddi> have my devices been removed somehow so the template won't work on them? =)
[07-Oct-2011 05:51:21] <konvert> hellow does anyone have mib for exchange 2010
[07-Oct-2011 08:23:24] <jmp242> Morning all
[07-Oct-2011 08:25:55] <jmp242> Anyone (Jane? Hackman? rmatte?) seen zenoss or rrdtool cut off history at exactly 3 months before?
[07-Oct-2011 08:26:15] <jmp242> I recall the 30 day issue with timestamps
[07-Oct-2011 08:26:18] <jmp242> but not a 90 day one
[07-Oct-2011 09:27:59] kocolosk|away is now known as kocolosk
[07-Oct-2011 09:46:59] <konvert> can someone help me? i don't seem to succeed in installing a particular zenpack
[07-Oct-2011 09:47:32] <Sashness> what error do you get?
[07-Oct-2011 09:48:52] <konvert> Unable to create object using the following attributes: * id: /zport/dmd/Devices/Power/UPS/ApcUps * module: Products.ZenModel.DeviceClass * class: DeviceClass
[07-Oct-2011 09:49:03] <konvert> it's the zenpack for apc ups
[07-Oct-2011 09:49:25] <fragfutter> konvert: does it alread exist?
[07-Oct-2011 09:49:31] <fragfutter> +y
[07-Oct-2011 09:50:17] <konvert> i think i installed it once but deleted it again
[07-Oct-2011 09:51:07] <fragfutter> just check.
[07-Oct-2011 09:51:31] <konvert> ok i'm sure i installed it once but deleted it afterwards
[07-Oct-2011 10:20:56] <konvert> fragfutter ?
[07-Oct-2011 10:21:10] <fragfutter> yes?
[07-Oct-2011 10:25:47] <fragfutter> just check for the class within ZMI or zenoss.
[07-Oct-2011 10:25:53] <fragfutter> if it's there remove it
[07-Oct-2011 10:25:58] <fragfutter> and i'm off for the weekend
[07-Oct-2011 10:38:23] <Omeganon> Good morning, afternoon, evening and night everyone!
[07-Oct-2011 10:38:24] <Omeganon> heh
[07-Oct-2011 10:39:33] <Omeganon> I have what I hope is a quick-to-resolve problem. When I go to infrastructure -> processes, I get the following error. It appears to be directly related to loading the process listing frame as that part never loads (lower left) --
[07-Oct-2011 10:40:04] <Omeganon> The server reported the following error: "Exception Cannot find "". KeyError: '' The system has encountered an error. Please reload the page.
[07-Oct-2011 10:40:14] <Omeganon> any tips/pointers to resolve this?
[07-Oct-2011 10:41:39] <wayne___> Omeganon: somebody reported that last night and said clearing their browser cache did the trick
[07-Oct-2011 10:42:06] <wayne___> (they were in a different context, but it was the same error)
[07-Oct-2011 10:42:29] <Omeganon> yeah, I've tried that but I'll see if I can be more aggressive about it.
[07-Oct-2011 10:46:36] <rocket> Omeganon: also check the event.log for more information re that traceback
[07-Oct-2011 10:47:37] <Omeganon> mmm
[07-Oct-2011 10:47:37] <Omeganon> File "/usr/local/zenoss/zenoss/Products/Zuul/facades/__init__.py", line 66, in _getObject
[07-Oct-2011 10:47:37] <Omeganon> raise Exception('Cannot find "%s". %s: %s' % args)
[07-Oct-2011 10:47:37] <Omeganon> Exception: Cannot find "<DataRoot at dmd>". KeyError: '<DataRoot at dmd>'
[07-Oct-2011 10:48:01] <Omeganon> I _think_ it might be firefox trying to be smart about remembering form fields for something that I've since removed
[07-Oct-2011 10:50:47] <rocket> Omeganon: firebug may assist with that as well .. you can inspect what you are sending to the form
[07-Oct-2011 10:51:22] <Omeganon> got it.
[07-Oct-2011 10:52:09] <Omeganon> clear cache (not enough to fix it). close all zenoss related tabs (app tabs and other), force quit firefox (probably not needed). restart firefox.
[07-Oct-2011 10:52:53] <Omeganon> thanks for the tips and listening =)
[07-Oct-2011 10:54:03] <bseklecki> after installing a MIB, do I need to restart everyhing?
[07-Oct-2011 10:55:00] <bseklecki> HOST-RESOURCES-MIB::hrSystemProcesses.0 definitely not reading as a data souce
[07-Oct-2011 11:00:55] <wayne___> bseklecki: unless it's changed recently, you can't use symbolic names for data sources
[07-Oct-2011 11:04:40] <rocket> wayne___: is correct .. the mibs are really only used for snmp trap oid to name resolution
[07-Oct-2011 11:06:00] <wayne___> bseklecki: if you've got the net-snmp tools installed somewhere you can get the full OID using: snmptranslate -Ofn HOST-RESOURCES-MIB::hrSystemProcesses.0
[07-Oct-2011 11:08:13] <konvert> rocket wat can i do if i have installed a zenpack in the past and uninstalled it and i want to install it again and i get an error?
[07-Oct-2011 11:08:18] <konvert> Unable to create object using the following attributes: * id: /zport/dmd/Devices/Power/UPS/ApcUps * module: Products.ZenModel.DeviceClass * class: DeviceClass
[07-Oct-2011 11:18:00] <SDuensin> FrIdAy!
[07-Oct-2011 11:23:08] <rocket> konvert: most likely ApcUps exists currently?
[07-Oct-2011 11:23:50] <konvert> not as zenpack
[07-Oct-2011 11:24:01] <rocket> as a device class organizer
[07-Oct-2011 11:24:16] <konvert> and where can i delete that
[07-Oct-2011 11:24:51] <rocket> go to the UPS organizer and select that ApcUps organizer and there should be an option to delete it
[07-Oct-2011 11:25:08] <rocket> probably a gear menu in the lower left corner
[07-Oct-2011 11:26:38] <konvert> no there isn't i have an organizer power
[07-Oct-2011 11:26:50] <rocket> no organizer UPS?
[07-Oct-2011 11:26:54] <konvert> no
[07-Oct-2011 11:27:16] <rocket> then try running zenchkrels -r -x1 from the command line
[07-Oct-2011 11:28:16] <konvert> what does that do?
[07-Oct-2011 11:28:32] <rocket> repair corrupt broken relationships
[07-Oct-2011 11:28:38] <konvert> ok is running
[07-Oct-2011 11:32:18] <konvert> done, still same error
[07-Oct-2011 11:41:19] <rocket> doh .. make the UPS organizer
[07-Oct-2011 11:41:34] <rocket> its expecting it to be there first
[07-Oct-2011 12:00:01] [disconnected at Fri Oct 7 12:00:01 2011]
[07-Oct-2011 12:00:02] [connected at Fri Oct 7 12:00:02 2011]
[07-Oct-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]
[07-Oct-2011 12:17:54] <rocket> I must assume that resolve the issue ...
[07-Oct-2011 13:16:13] <mahimahi> Didn't see a response yesterday, so asking again: Is there a way to add a zenoss user via command line like zencommand?
[07-Oct-2011 13:29:38] <dpetzel> mahimahi: its not exactly like zencommand, but I think something like this would work: echo "dmd.ZenUsers.manage_addUser(userid='User1', password='pwd1'); commit()" | zendmd
[07-Oct-2011 13:33:03] <mahimahi> thanks, I'll try that out
[07-Oct-2011 13:37:38] <humanix> hi there Hackman238 rmatte
[07-Oct-2011 13:37:43] <humanix> ive been idle for a while
[07-Oct-2011 14:02:23] <bseklecki> zenoss.snmp.UCDHardDiskMap should get me disk statistics on usage from hrDiskIOTable ?
[07-Oct-2011 14:03:53] <bseklecki> UCDHardDiskMap.py
[07-Oct-2011 15:13:17] <dhopp> ugh..Why does Microsoft insist on using protocols that make traversing firewalls impossible?
[07-Oct-2011 15:27:52] <jmp242> darned if I know
[07-Oct-2011 15:33:13] <wayne___> dhopp: MSRPC issues?
[07-Oct-2011 15:40:35] <dhopp> wayne___: yeah…in our environment each of our client servers are behind their own firewall and are natted..so dcom doesn't work
[07-Oct-2011 15:45:43] <wayne___> dhopp: yeah, I hate that. Some firewalls have ALGs to detect the MSRPC traffic and open pinholes, but those tend to be spendy.
[07-Oct-2011 16:13:30] <dhopp> most of our firewalls are cisco asas
[07-Oct-2011 16:13:52] <dhopp> so they aren't cheap..but doesn't seem to be working…maybe could figure it out with a configuration change to the firewall
[07-Oct-2011 16:14:00] <dhopp> but microsoft should stop being stupid
[07-Oct-2011 16:31:46] <wayne___> dhopp: good luck there (getting MS to do anything different, that is 
I believe the ASA needs explicit configuration (look for docs on inspect dcerpc), but most of my recent firewall experience is on Junos so I'm not 100% certain there
[07-Oct-2011 16:45:30] <dhopp> wayne___: I'll probably just find a different way to do it :-)..I hate WMI anyway
[07-Oct-2011 17:24:49] <mahimahi> Having problems with dev zenpack. Created zenpack, but cannot add monitoring template data point to my zenpack via the GUI. I've skimmed thru Jane Curry write up but cannot figure out how to add this data point via command line (source mode?) ? Any help is much appreciated. Cudos to dpetzel for user add solution in my earlier post.
[07-Oct-2011 18:46:07] kocolosk is now known as kocolosk|away
[07-Oct-2011 20:05:22] kocolosk|away is now known as kocolosk
[07-Oct-2011 20:33:18] tvincent_ is now known as tvincent
[07-Oct-2011 20:35:33] tvincent_ is now known as tvincent
[07-Oct-2011 20:42:55] kocolosk is now known as kocolosk|away
[08-Oct-2011 00:00:01] [disconnected at Sat Oct 8 00:00:01 2011]
[08-Oct-2011 00:00:02] [connected at Sat Oct 8 00:00:02 2011]
[08-Oct-2011 00:00:16] [zenoss-logger (logger bot) has joined #zenoss]
[08-Oct-2011 12:00:01] [disconnected at Sat Oct 8 12:00:01 2011]
[08-Oct-2011 12:00:01] [connected at Sat Oct 8 12:00:01 2011]
[08-Oct-2011 12:00:13] [zenoss-logger (logger bot) has joined #zenoss]
[09-Oct-2011 00:00:01] [disconnected at Sun Oct 9 00:00:01 2011]
[09-Oct-2011 00:00:01] [connected at Sun Oct 9 00:00:01 2011]
[09-Oct-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[09-Oct-2011 12:00:01] [disconnected at Sun Oct 9 12:00:01 2011]
[09-Oct-2011 12:00:02] [connected at Sun Oct 9 12:00:02 2011]
[09-Oct-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[09-Oct-2011 18:52:38] <Hamzah> Hmm does zenoss always compile external libraries regardless of if they are available on the system already? I'm trying to build the RHEL5 SRPM on RHEL6. I managed to get it to build, but I noticed there are a lot of libs packaged with it (for example rrdtool, and libsmi and a lot of others)
[10-Oct-2011 00:00:01] [disconnected at Mon Oct 10 00:00:01 2011]
[10-Oct-2011 00:00:02] [connected at Mon Oct 10 00:00:02 2011]
[10-Oct-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[10-Oct-2011 06:57:49] <rcngeoff> Which process is responsible for processing ip service availability? I am getting alerts that appear and then clear quickly about a specific IP service on a server. I'd like to restart the process, but I'm not sure which one it is.
[10-Oct-2011 07:05:46] <froztbyte> meh
[10-Oct-2011 07:05:53] <froztbyte> I should look into Flap Detection again
[10-Oct-2011 07:24:15] <Hackman238> Morning all!
[10-Oct-2011 07:24:18] <whyzgeek> hi guys
[10-Oct-2011 07:24:40] <whyzgeek> any one has any experience with zopedb larger than 20GB
[10-Oct-2011 07:24:47] <whyzgeek> after zeopack
[10-Oct-2011 07:25:18] <whyzgeek> I have modeling problems which each modeling sometimes takes more than 1hr
[10-Oct-2011 07:25:20] <Hackman238> whyzgeek: Yes, did your zeopack fail?
[10-Oct-2011 07:25:28] <whyzgeek> nope
[10-Oct-2011 07:25:31] <whyzgeek> it is happy
[10-Oct-2011 07:25:37] <Hackman238> whyzgeek: Okay and the packed db is 20GB?
[10-Oct-2011 07:25:38] <whyzgeek> but my problem is with modeling
[10-Oct-2011 07:25:42] <whyzgeek> yes
[10-Oct-2011 07:25:53] <whyzgeek> so everything is slow
[10-Oct-2011 07:26:06] <whyzgeek> and sometimes the modeling get stuck at applyDataMap
[10-Oct-2011 07:26:11] <Hackman238> whyzgeek: I'd increase the object cache very substantially and increase the zeocache amount in zope.conf
[10-Oct-2011 07:26:11] <whyzgeek> which runs on the zenhub
[10-Oct-2011 07:26:38] <whyzgeek> does the modeling use the cach?
[10-Oct-2011 07:26:47] <Hackman238> whyzgeek: Yeah, sounds like its taking too long to commit transactions
[10-Oct-2011 07:26:49] <whyzgeek> I thought cache is for ready
[10-Oct-2011 07:26:52] <whyzgeek> reading
[10-Oct-2011 07:27:07] <whyzgeek> but I couldn't figure out
[10-Oct-2011 07:27:15] <whyzgeek> why it only happens sometimes
[10-Oct-2011 07:27:22] <whyzgeek> for example a single device
[10-Oct-2011 07:27:27] <whyzgeek> out of 5 modeling
[10-Oct-2011 07:27:34] <whyzgeek> 3 success and 2 fails
[10-Oct-2011 07:27:46] <whyzgeek> and basically it get into infinite loop
[10-Oct-2011 07:27:58] <whyzgeek> means the zCollectorTimeout expires
[10-Oct-2011 07:28:03] <whyzgeek> and it tries to kill the reactor
[10-Oct-2011 07:28:04] <Hackman238> sounds like race condition. After so many transaction attempts fail it just stops
[10-Oct-2011 07:28:31] <whyzgeek> ya sometime it takes a couple of hours to quit
[10-Oct-2011 07:28:32] <Hackman238> whyzgeek: You'll also want to increase the invalidation-queue-size in $ZENHOME/etc/zeo.conf
[10-Oct-2011 07:28:36] <whyzgeek> and sometime it never quites
[10-Oct-2011 07:28:53] <whyzgeek> good I will check it out
[10-Oct-2011 07:29:01] <whyzgeek> thanks for the advice
[10-Oct-2011 07:30:00] <Hackman238> whyzgeek: No problem. Here at rackspace we have a couple similarly sizes zodb's and we use 10000 for the invalidation queue size
[10-Oct-2011 07:30:08] <Hackman238> *sized
[10-Oct-2011 07:30:32] <whyzgeek> ic how can we measure that how much of it is used now?
[10-Oct-2011 07:30:55] <whyzgeek> just want to double check
[10-Oct-2011 07:35:18] <Hackman238> whyzgeek: Heres a good page on Zeo http://wiki.zope.org/ZODB/howto.html
[10-Oct-2011 07:35:30] <Hackman238> whyzgeek: Off the top of my head I dont know how to check it
[10-Oct-2011 07:35:57] <whyzgeek> Thank you, I will have a read
[10-Oct-2011 07:37:27] <Hackman238> whyzgeek: NP
[10-Oct-2011 07:38:21] meddle- is now known as bender[a]
[10-Oct-2011 07:39:09] <whyzgeek> ah I c now where it is bleeding, we have huge datamaps, I assume they get committed in one transaction
[10-Oct-2011 07:39:36] <Hackman238> whyzgeek: They often are, yes
[10-Oct-2011 07:39:51] <whyzgeek> yep basically that blocks everything else
[10-Oct-2011 07:39:58] <Hackman238> whyzgeek: yep
[10-Oct-2011 07:40:33] <whyzgeek> horrible, it seems to match it because I don't see much activity on the zeo process either so certainly there is no resource issue
[10-Oct-2011 07:43:47] <whyzgeek> Hackman238: what will be your recommendation on zenhub and zope cache?
[10-Oct-2011 07:44:46] <Hackman238> whyzgeek: pcache 2000, cachesize 100000
[10-Oct-2011 07:45:18] <whyzgeek> thanks
[10-Oct-2011 07:45:24] <Hackman238> whyzgeek: NP
[10-Oct-2011 07:50:21] <rcngeoff> Hi again... which process is responsible for processing ip service availability? I am getting alerts that appear and then clear quickly about a specific IP service on a server. I'd like to restart the process, but I'm not sure which one it is.
[10-Oct-2011 07:54:46] <whyzgeek> zenstatus I guess
[10-Oct-2011 07:54:54] <whyzgeek> but it should in your event
[10-Oct-2011 07:55:00] <whyzgeek> check agent feild
[10-Oct-2011 07:58:20] <rcngeoff> ok.. thanks
[10-Oct-2011 08:12:16] <jmp242> Morning
[10-Oct-2011 08:12:24] <jmp242> Hey Hackman238 - back home yet?
[10-Oct-2011 08:16:15] <Hackman238> jmp242: Hey!
[10-Oct-2011 08:16:23] <Hackman238> jmp242: I am, just in yesterday.
[10-Oct-2011 08:17:01] <Hackman238> jmp242: Chopping down my to do list like a college student double-fisting beers
[10-Oct-2011 08:17:12] <Hackman238> jmp242: How's by you?
[10-Oct-2011 08:50:20] <jmp242> ehh
[10-Oct-2011 08:50:21] <jmp242> ok
[10-Oct-2011 08:50:42] <jmp242> Still working on nTop - got everything in place, now I need to figure out what I'm doing with monitoring the crashy thing
[10-Oct-2011 08:55:27] <Hackman238> jmp242: nTop?
[10-Oct-2011 08:55:46] <jmp242> www.ntop.org
[10-Oct-2011 08:55:48] <Hackman238> jmp242: What in the blazes are you torturing yourself with that mess for?
[10-Oct-2011 08:56:03] <jmp242> Umm, it's pretty nice when it doesn't crash
[10-Oct-2011 08:56:55] <Hackman238> jmp242: Tried using is back at OFT, never could get it to run for any reasonable amount of time
[10-Oct-2011 08:57:00] <jmp242> although I don't know why the web interface crashes as often as it does
[10-Oct-2011 08:57:03] <Hackman238> s/is/it/
[10-Oct-2011 08:57:10] <jmp242> the nprobe runs fine
[10-Oct-2011 08:57:28] <jmp242> I just wish it would save more data somewhere other than memory lol
[10-Oct-2011 08:57:29] <Hackman238> jmp242: yeah the UI is the problem we had
[10-Oct-2011 08:57:43] <jmp242> Well, it's OSS, and better than what we had
[10-Oct-2011 08:57:50] <Hackman238> jmp242: Ended up using flows instead.
[10-Oct-2011 08:58:04] <Hackman238> jmp242: Good luck!
[10-Oct-2011 08:58:25] <jmp242> Yea, I plan on having Zenoss restart it when it dies
[10-Oct-2011 08:58:35] <Hackman238> jmp242: LOL!
[10-Oct-2011 08:58:46] <jmp242> I just need to see how I want to check for "dead"
[10-Oct-2011 08:59:08] <jmp242> I'm thinking just doing a
[10-Oct-2011 08:59:10] <jmp242> service ntop status
[10-Oct-2011 08:59:18] <jmp242> and if it says zombie, event command to restart it
[10-Oct-2011 09:02:00] <Hackman238> jmp242: Should work long as its a slayable zombie
[10-Oct-2011 09:03:17] <jmp242> It seems to be - every time I do it manually it just works
[10-Oct-2011 09:04:44] <Hackman238> nice.
[10-Oct-2011 09:19:29] <ion-fan> @hackman238: can you have a look on my forums topic ? message/61776#61776
[10-Oct-2011 09:21:46] <Hackman238> ion-fan: Yep, its on my to do for this afternoon. I appologize, I was on vacation all last week.
[10-Oct-2011 09:25:42] <ion-fan> oh sorry for bugging you.. hope you had a great vacation
[10-Oct-2011 09:31:39] <Hackman238> ion-fan: No reason to be sorry at all Was very good!
[10-Oct-2011 09:37:39] <pplask> hello!
[10-Oct-2011 09:38:24] <pplask> im using lastest zenoss core. Is there a way to view my monitored devices in a map like nagios status map? I mean, with host dependencies
[10-Oct-2011 09:38:52] <fragfutter> no host dependencies in zenoss
[10-Oct-2011 09:39:42] <rocket> fragfutter: in zenoss core that is
[10-Oct-2011 09:39:53] <rocket> fragfutter: there is a way to do that with an add on to enterprise
[10-Oct-2011 09:40:37] <pplask> so, theres no way to have an IT Management view of my monitored devices?
[10-Oct-2011 09:41:01] <pplask> like nagios status map or zabbix IT view screen
[10-Oct-2011 09:41:17] kocolosk|away is now known as kocolosk
[10-Oct-2011 09:41:24] <rocket> there is no way to do a dependency map sort of thing with core out of the box
[10-Oct-2011 09:41:32] <rocket> from a graphical perspective
[10-Oct-2011 09:42:08] <rocket> you can always create dependancy rules via transforms etc but that is a clunky way to do it
[10-Oct-2011 09:45:55] <rocket> pplask: there is a google maps view of locations etc
[10-Oct-2011 09:46:29] <rocket> pplask: and you could create a group of like resources and related events would show up in the device view
[10-Oct-2011 09:49:36] <rocket> whyzgeek: are you there?
[10-Oct-2011 09:51:53] <fragfutter> nope he isn't
[10-Oct-2011 09:52:03] <rocket> lol
[10-Oct-2011 09:52:11] <rocket> I guess we will see if he comes back
[10-Oct-2011 09:52:24] <fragfutter> cue terminator music
[10-Oct-2011 09:52:26] <pplask> hey im here
[10-Oct-2011 09:52:32] <pplask> sry
[10-Oct-2011 09:52:33] <pplask> let me read
[10-Oct-2011 09:53:08] <pplask> mm, ok
[10-Oct-2011 09:53:16] <pplask> i will play with that groups thing
[10-Oct-2011 09:53:40] <pplask> but whats the use for "systems" in zenoss core if you cant have that dynamic view?
[10-Oct-2011 09:53:50] <rocket> pplask: you can filter by group in the event console etc
[10-Oct-2011 09:53:56] <rocket> manually grouping
[10-Oct-2011 09:54:04] <rocket> its for filtering in the event console etc
[10-Oct-2011 09:54:09] <pplask> mmm
[10-Oct-2011 09:54:19] <pplask> but i can use "groups" for it, no?
[10-Oct-2011 09:54:27] <rocket> systems and groups are basically the same thing
[10-Oct-2011 09:54:40] <rocket> just more ways to categorize
[10-Oct-2011 09:54:47] <pplask> mmm ok
[10-Oct-2011 09:55:43] <rocket> honestly most people use that just fine and are able to figure out the issues etc
[10-Oct-2011 10:28:36] <rocket> whyzgeek: are you there?
[10-Oct-2011 10:33:13] <jmp242> Hey, if you were going to run a command like /sbin/service ntop restart from an event command - this runs as the Zenoss user... What method would you use to run as root on EL5?
[10-Oct-2011 10:33:37] <rocket> jmp use sudo without a password setup for that command only?
[10-Oct-2011 10:33:49] <rocket> err meant to be jmp242
[10-Oct-2011 10:33:58] <rocket> sorry didnt meant to truncate your nickname ..
[10-Oct-2011 10:34:02] <jmp242> that's ok
[10-Oct-2011 10:34:16] <jmp242> ok, sounds good to me, thanks
[10-Oct-2011 10:34:24] <rocket> or create a setuid wrapper script?
[10-Oct-2011 10:34:40] <rocket> that just calls that command
[10-Oct-2011 10:34:52] <rocket> whyzgeek: are you there?
[10-Oct-2011 10:47:32] <whyzgeek> rocket: yes I am back
[10-Oct-2011 10:52:18] <rocket> whyzgeek: do you have a custom modeller?
[10-Oct-2011 10:52:43] <whyzgeek> nope
[10-Oct-2011 10:53:01] <whyzgeek> it is Zenoss Enterprise 2.5.2
[10-Oct-2011 10:53:10] <rocket> you mentioned the data for apply datamap was huge
[10-Oct-2011 10:53:22] <whyzgeek> yep that's because of the zenpack we have
[10-Oct-2011 10:53:30] <whyzgeek> a plugin which has these maps
[10-Oct-2011 10:53:30] <rocket> whyzgeek: I would upgrade to 3.2 + patches if possible
[10-Oct-2011 10:53:40] <whyzgeek> well it is not that easy
[10-Oct-2011 10:53:48] <whyzgeek> this is massive setup here
[10-Oct-2011 10:53:51] <rocket> there is a huge invalidation queue patch for 3.0+
[10-Oct-2011 10:53:58] <whyzgeek> probably we will go straigh to Avalon
[10-Oct-2011 10:54:06] <whyzgeek> really
[10-Oct-2011 10:54:15] <rocket> that patch is in avalon as well
[10-Oct-2011 10:54:23] <whyzgeek> that's good to know, then I have to check the commit logs
[10-Oct-2011 10:54:36] <whyzgeek> probably we have to backport it
[10-Oct-2011 10:54:42] <whyzgeek> until we do upgrade
[10-Oct-2011 10:55:47] <whyzgeek> what does that patch do?
[10-Oct-2011 10:59:40] <rocket> whyzgeek: it basically batches changes
[10-Oct-2011 11:01:13] <Hackman238> rocket: It it a change in the transaction code?
[10-Oct-2011 11:02:39] * whyzgeek I guess he is looking for url to it
[10-Oct-2011 11:06:51] <rocket> Hackman238: you already knew about this patch .. I think you wrote a backport for 2.5.2 didnt you?
[10-Oct-2011 11:09:55] <Hackman238> rocket: Oh that patch, yes
[10-Oct-2011 11:10:28] <Hackman238> rocket: I'd forgot about that. Huge performance gains
[10-Oct-2011 11:14:04] <rmatte> Hackman238: right, at that point it's the actual creation of the RRD that's the problem
[10-Oct-2011 11:14:07] <rmatte> oops
[10-Oct-2011 11:14:08] <rmatte> privmsg
[10-Oct-2011 11:34:01] <rmatte> What patch?
[10-Oct-2011 11:34:25] <rmatte> I
[10-Oct-2011 11:34:35] <rmatte> I'm always up for huge performance gains lol
[10-Oct-2011 11:34:58] <dhopp> me too
[10-Oct-2011 11:35:29] <dhopp> rmatte: I was looking for something else and stumbled across your pastebin for your zenticket daemon...lol
[10-Oct-2011 11:35:49] <Hackman238> rmatte / dhopp: Its a patch that fixes blocking code in v3
[10-Oct-2011 11:35:55] <Hackman238> ...backported to v2.5.2
[10-Oct-2011 11:36:03] <rmatte> blocking code?
[10-Oct-2011 11:36:10] <dhopp> rmatte: for zenhub I believe
[10-Oct-2011 11:36:14] <rmatte> ah
[10-Oct-2011 11:36:17] <Hackman238> rmatte: Takes case of situations where zenhub gets plugged up
[10-Oct-2011 11:36:18] <dhopp> and I think that is in 3.2
[10-Oct-2011 11:36:24] <rmatte> nice
[10-Oct-2011 12:00:01] [disconnected at Mon Oct 10 12:00:01 2011]
[10-Oct-2011 12:00:01] [connected at Mon Oct 10 12:00:01 2011]
[10-Oct-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]
[10-Oct-2011 12:05:50] <Hackman238> dhopp: correct
[10-Oct-2011 12:57:17] <nyeates> any tested 3.2.1 beta upgrade from 3.2.0??
[10-Oct-2011 12:57:44] <nyeates> anyone*
[10-Oct-2011 13:22:02] <dhopp> nyeates: I've only done a clean install…I can do a quick test but it would be on a system with very few devices
[10-Oct-2011 13:22:48] <dhopp> is there any issues that you know of that you are wanting verification on?
[10-Oct-2011 15:28:04] <nyeates> dhopp: yeah someone here just tried a 3.2 -> 3.2.1 upgrade and it barfed
[10-Oct-2011 15:28:23] <rmatte> that's lame
[10-Oct-2011 15:28:29] <rmatte> will it be fixed by release?
[10-Oct-2011 15:29:02] <nyeates> yeah well, i have no idea if it is a one-of thing with his setup or is repeatable....was looking to see if anyone here could disprove
[10-Oct-2011 15:29:13] <nyeates> one-off*
[10-Oct-2011 15:29:34] <rmatte> ah, well I have a 3.2 install that was originall upgraded from 3.1
[10-Oct-2011 15:29:41] <rmatte> let me give it a shot
[10-Oct-2011 15:29:48] <nyeates> it will be fixed, of course -
[10-Oct-2011 15:29:52] <nyeates> that wouldbe helpful ryan
[10-Oct-2011 15:30:28] <rmatte> I'm not actually working today since it's a holiday so I actually have time to do stuff like this lol
[10-Oct-2011 15:30:37] <nyeates> it was a 3.2 enterprise machine to 3.2.1 core....so that could be it
[10-Oct-2011 15:30:44] <nyeates> hah
[10-Oct-2011 15:30:49] <rmatte> ah, I'd say that's probably it
[10-Oct-2011 15:30:55] <rmatte> enterprise and core have quite a few differences
[10-Oct-2011 15:31:37] <rmatte> oh actually, I think I may still be running a beta of 3.2 too lol, should install the full version first.
[10-Oct-2011 15:31:46] <nyeates> yeah but enterprise is base of core
[10-Oct-2011 15:31:54] <nyeates> errr, other way around
[10-Oct-2011 15:32:32] <rmatte> right, you can't go from enterprise to core
[10-Oct-2011 15:32:35] <rmatte> won't work
[10-Oct-2011 15:33:21] <rmatte> and technically what you just said is only true until version 4.x
[10-Oct-2011 15:33:25] <nyeates> yeah but to upgrade enterprise , you upgrade the core parts too
[10-Oct-2011 15:33:31] <nyeates> i think this is what he was testing
[10-Oct-2011 15:33:36] <rmatte> ah
[10-Oct-2011 15:33:53] <nyeates> either way, itd be good to get verification
[10-Oct-2011 15:34:41] <rmatte> upgrading from 3.2 beta to 3.2 might not work but I'm trying it anyways, worst case scenario I'll just install 3.1.0 and then do the upgrade path from there.
[10-Oct-2011 15:34:55] <rmatte> just a lab box anyways
[10-Oct-2011 15:35:20] <nyeates> beta to GA is unsupported yeah.... in fact i think it is discouraged b/c it can run into issues....i dont know the technical reason tho
[10-Oct-2011 15:35:56] <rmatte> probably in terms of database upgrade sequences, but it should technically work
[10-Oct-2011 15:40:09] <rmatte> 3.2 install done, installing 3.2.1...
[10-Oct-2011 15:40:29] <nyeates> speedy gon-rmatte-ez
[10-Oct-2011 15:40:35] <rmatte> lol
[10-Oct-2011 15:41:00] <dhopp> nyeates: sorry I was up all night performing a load test for one client and another client's site is exploding
[10-Oct-2011 15:41:10] <rmatte> kaboom
[10-Oct-2011 15:41:51] <dhopp> rmatte: yeah initial traffic graphs are showing about 20x normal traffic
[10-Oct-2011 15:42:05] <rmatte> wow
[10-Oct-2011 15:43:02] <rmatte> DoS, backups gone ary, torrents, packet storm (which shouldn't technically be possible these days)
[10-Oct-2011 15:43:05] <rmatte> could be anything
[10-Oct-2011 15:43:43] <nyeates> rmatte, latest beta may not be up on http://alpha.zenoss.com/files/
[10-Oct-2011 15:43:49] <nyeates> hold a sec, unless you started
[10-Oct-2011 15:43:59] <rmatte> already started
[10-Oct-2011 15:44:04] <nyeates> ok go ahead with it
[10-Oct-2011 15:44:05] <rmatte> install 3.2.1-1284
[10-Oct-2011 15:44:10] <rmatte> installing*
[10-Oct-2011 15:44:36] <nyeates> will be interesting to see if the issue was introduced with a newer build - which is where we saw it
[10-Oct-2011 15:45:02] <dhopp> rmatte: yeah we are investigating a possible DDos but the site is behind akamai so from a web server perspective all of the traffic is coming from akamai so we have them engaged to try and see if they are seeing anything unusual
[10-Oct-2011 15:45:03] <rmatte> I'm using stack installer by the way, not RPM
[10-Oct-2011 15:45:18] <rmatte> dhopp: ah
[10-Oct-2011 15:46:35] <Hackman238> nyeates: 3.2.1 bug?
[10-Oct-2011 15:47:06] <rmatte> well, so far the install worked and I'm starting Zenoss up...
[10-Oct-2011 15:47:21] <nyeates> dhopp: spam possibly involved?
[10-Oct-2011 15:47:53] <rmatte> well 3.2.1 fixed the stupid bug I was having with my browser
[10-Oct-2011 15:48:01] <dhopp> nyeates: the client did send a mailer out advertising a sale..but the traffic spike is still really abnormal...
[10-Oct-2011 15:48:05] <rmatte> the loading screen would pop up and stay on the infrastructure page
[10-Oct-2011 15:48:18] <nyeates> dhopp: its insane the amount of troubles we have had to go through for user registration lately.....they have entire products and tool-chains for creating spam accounts on various web platforms
[10-Oct-2011 15:48:50] <nyeates> rmatte: so 3.2.1 beta (1284) install was errorless
[10-Oct-2011 15:48:59] <rmatte> correct, with stack installer
[10-Oct-2011 15:49:06] <nyeates> ok cool - thanks
[10-Oct-2011 15:49:10] <rmatte> np
[10-Oct-2011 15:50:32] <nyeates> just relayed this to our QA team - they said this help a lot - thanks rmatte
[10-Oct-2011 15:51:03] <dhopp> nyeates: what's the latest build of 3.2.1?
[10-Oct-2011 15:51:48] <nyeates> dhopp: latest public build that we have up is 1284.... apparently we have a 1302 that is internal for the time being
[10-Oct-2011 15:51:56] <dhopp> gotcha
[10-Oct-2011 15:52:03] <nyeates> takes resources and spinning to get it out in consumable format
[10-Oct-2011 15:52:51] <rmatte> nyeates: when you go to delete a device in 3.2.1 it gives you 2 options, one to just remove it from the device classes and one to remove it complete, but removing it completely is the only selectable option. Why is the other option even there at all?
[10-Oct-2011 15:54:01] <nyeates> hmmm, not sure....lemme see if there is a ticket associated
[10-Oct-2011 15:54:06] <rmatte> k
[10-Oct-2011 15:55:50] <hmp> rmatte: maybe its a multichoice if there can be multiple groupings (like Groups)
[10-Oct-2011 15:56:30] <hmp> so if a device is a member of only one group, you can just delete a device
[10-Oct-2011 15:56:30] <rmatte> hmp: no that's not it
[10-Oct-2011 15:56:54] <rmatte> no, it's not groups, it's device class
[10-Oct-2011 15:57:03] <rmatte> and a device can only be a member of one device class
[10-Oct-2011 15:57:24] <hmp> yea, but maybe it's a generic form for deleting
[10-Oct-2011 15:57:27] <rmatte> "Just remove from Device Classes" and "Delete completely"
[10-Oct-2011 15:57:36] <hmp> but you probably know more about it than I do
[10-Oct-2011 15:57:39] <rmatte> but Delete completely is the only option that's selectable
[10-Oct-2011 15:58:23] <nyeates> yeah im wondering when this came into effect....like if its beenn there for 3.2 or if its new in 321
[10-Oct-2011 15:58:40] <rmatte> not sure, first time I noticed it
[10-Oct-2011 15:58:45] <rmatte> but it's weird that it's there if it's not usable
[10-Oct-2011 15:58:54] <rmatte> and even if it was, it sounds dangerous
[10-Oct-2011 15:59:37] <nyeates> i know we had discussed the problem of users not knowing what they are truly doing and to bring in solutions for it....kinda sounds like that
[10-Oct-2011 15:59:57] <nyeates> but i am not aware of it or how it works
[10-Oct-2011 16:00:35] <rmatte> "Zenoss, the mystery interface. Every version is a new adventure!"
[10-Oct-2011 16:00:40] <rmatte> good slogan
[10-Oct-2011 16:01:50] <nyeates> I dunno, it sounds like a good feature to me....if it works right
[10-Oct-2011 16:02:24] <Hackman238> rmatte: Its annoying, but one has to be on the deviceStatus page and delete from there to be able to uncheck any delete options
[10-Oct-2011 16:03:42] <rmatte> Hackman238: those options used to be available directly from the infrastructure page
[10-Oct-2011 16:03:52] <rmatte> and now it has a new option that's not even available
[10-Oct-2011 16:04:16] <rmatte> never ceases to amaze
[10-Oct-2011 16:05:50] <jmp242> rmatte: no worse than Microsoft programs lol
[10-Oct-2011 16:05:53] <Hackman238> rmatte: Yep, only accessible at the device [age now.
[10-Oct-2011 16:05:58] <Hackman238> Very annoying
[10-Oct-2011 16:06:00] <Hackman238> LOL
[10-Oct-2011 16:06:57] <rmatte> Is there someone at Zenoss whose job it is to say things like "Hey, this option is working just a bit too well... let's move it to someplace that they'll never find it! Muahaha!"
[10-Oct-2011 16:07:28] <rmatte> Because that's what I suspect at this point lol
[10-Oct-2011 16:07:59] <Hackman238> rmatte: Its the UI specializt they hired. "More drag and drop" he says
[10-Oct-2011 16:08:05] <Hackman238> LOL
[10-Oct-2011 16:08:17] <rmatte> what does that have to do with removing options from dialogs?
[10-Oct-2011 16:08:45] <rmatte> and more drag and drop is fine, but I should be able to drag and drop *multiple* objects including templates
[10-Oct-2011 16:08:55] <rmatte> in order to copy or move them (which I should be prompted for)
[10-Oct-2011 16:09:09] <rmatte> this type of interface design is not rocket science, and they are failing miserably so far.
[10-Oct-2011 16:09:40] <Hackman238> rmatte: Nothing, just trying to break-balls.
[10-Oct-2011 16:09:45] <rmatte> lol
[10-Oct-2011 16:09:48] <Hackman238> LOL
[10-Oct-2011 16:10:25] <Hackman238> rmatte: http://www.youtube.com/watch?v=4iSXtO9z6VU
[10-Oct-2011 16:10:34] <rmatte> hmmmm, day off, beer sounds like a plan right about now, I am pub bound shortly
[10-Oct-2011 16:10:40] <Hackman238> (sorry for youtube/moron quality)
[10-Oct-2011 16:10:53] <rmatte> haha
[10-Oct-2011 16:11:38] <Hackman238>
[10-Oct-2011 16:12:08] <rmatte> Hackman238: did you see the bar fridge I won the other day?
[10-Oct-2011 16:13:11] <Hackman238> rmatte: I did- very nice
[10-Oct-2011 16:13:28] <rmatte> I've had people offer me $1000 for it
[10-Oct-2011 16:13:32] <rmatte> lol
[10-Oct-2011 16:13:37] <Hackman238> rmatte: Damn!
[10-Oct-2011 16:13:44] <rmatte> but I like it too much
[10-Oct-2011 16:13:46] <Hackman238> rmatte: I'd try to build one before paying that.
[10-Oct-2011 16:14:02] <rmatte> lol yeh, it's because of the Keith's labelling on it
[10-Oct-2011 16:14:05] <jmp242> UI specialist things more drag and drop?
[10-Oct-2011 16:14:06] <rmatte> really popular beer around here
[10-Oct-2011 16:14:11] <jmp242> god I hate that model
[10-Oct-2011 16:14:15] <jmp242> first, no one expects it on the net
[10-Oct-2011 16:14:23] <jmp242> second, it often breaks across browsers
[10-Oct-2011 16:14:32] <rmatte> jmp242: agreed on both points
[10-Oct-2011 16:14:34] <jmp242> and it's the only model that almost encourages the accidental
[10-Oct-2011 16:14:41] <jmp242> whoops my finger slipped
[10-Oct-2011 16:14:41] <rmatte> though the second point is the most valid
[10-Oct-2011 16:14:47] <jmp242> where the F**K did that go?
[10-Oct-2011 16:14:49] <rmatte> yeh, that's the big one
[10-Oct-2011 16:14:55] <rmatte> accidental copies/moves
[10-Oct-2011 16:15:04] <rmatte> not the greatest thing when you're managing a monitoring tool
[10-Oct-2011 16:15:15] <jmp242> It's never great IMO
[10-Oct-2011 16:15:16] <rmatte> "Why is printers listed under Network/Router/Cisco!?"
[10-Oct-2011 16:15:22] <rmatte> "oops"
[10-Oct-2011 16:15:35] <jmp242> I read an article yesterday in Time
[10-Oct-2011 16:15:47] <rmatte> Whatever UI specialist they've hired is missing the point in many ways
[10-Oct-2011 16:15:47] <Hackman238> jmp242: Relax guy. You sound like you need a beer too.
[10-Oct-2011 16:15:51] <jmp242> regarding Steve Jobs understanding that form was more important in sales
[10-Oct-2011 16:15:55] <jmp242> than function
[10-Oct-2011 16:16:09] <jmp242> That may sell, but the working thing is actually important IMO
[10-Oct-2011 16:16:18] <Hackman238> jmp242: Definately
[10-Oct-2011 16:16:21] <jmp242> i.e. Old jaguars had awesome form
[10-Oct-2011 16:16:29] <jmp242> but I didn't buy it for a lawn ornament
[10-Oct-2011 16:16:38] <jmp242> and the cheapest Honda actually got me places lol
[10-Oct-2011 16:16:45] <rmatte> agreed
[10-Oct-2011 16:17:01] <rmatte> I wouldn't still be running 2.5.2 on all my production boxes if I cared about form
[10-Oct-2011 16:17:11] <jmp242> IDK, I get that for personal users, design and form is important
[10-Oct-2011 16:17:15] <rmatte> When it comes to a monitoring tool function very much comes first
[10-Oct-2011 16:17:16] <jmp242> but this is a Monitoring Product
[10-Oct-2011 16:17:19] <rmatte> form is a distant second
[10-Oct-2011 16:17:36] <jmp242> If they need "form" for sales, make it an optional "dashboard" zenpack or something
[10-Oct-2011 16:17:57] <jmp242> i.e. separate out the extraneous unusuable "pretty" from the workable admin interface
[10-Oct-2011 16:18:09] <rmatte> "I'm sorry that the product didn't detect your outage, but didn't it look damn good not doing it?"
[10-Oct-2011 16:18:57] <jmp242> Indeed
[10-Oct-2011 16:18:57] <jmp242> lol
[10-Oct-2011 16:19:26] <jmp242> oh well. See you all tomorrow
[10-Oct-2011 16:19:47] <rmatte> "You can't mass copy objects, but look! Shiny icons!"
[10-Oct-2011 16:19:49] <rmatte> cya lol
[10-Oct-2011 16:19:52] <rmatte> I'm taking off too
[10-Oct-2011 16:19:53] <rmatte> later all
[10-Oct-2011 16:20:11] <Hackman238> rmatte: Later
[10-Oct-2011 16:20:15] <Hackman238> jmp242: Later man!
[10-Oct-2011 16:22:20] <cgibbons> kinda cranky
[10-Oct-2011 16:23:01] <Hackman238> cgibbons: LOL
[10-Oct-2011 16:26:08] <dhopp> jmp242: the problem is you are thinking like an admin…and when it comes to sales most of the time the people making the decisions of what to spend money on only sees the eye candy..they don't have to worry about it being functional…that's your job to figure out how to make what they bought actually work
[10-Oct-2011 16:27:51] <dhopp> and with Zenoss the under the cover workings is very good..and the cover looks great but still has some usability issues...
[10-Oct-2011 16:28:12] <dhopp> with that said I actually prefer the 3.x UI but there is definitely room for approvement
[10-Oct-2011 16:28:23] <dhopp> improvement even
[10-Oct-2011 16:43:33] <nyeates> dhopp: I think your above comments are true, from what I have observed
[10-Oct-2011 16:45:44] <nyeates> I think its a little unfortunate that mgrs or decision makers in purchasing arent relying more on the opinions and feedback of those taht will be using what they buy
[10-Oct-2011 17:05:21] <pplask> hello!
[10-Oct-2011 17:06:42] <pplask> is there a way to pass the reason of an event via pager script? the sintax is pagerscript.sh $RECIPIENT, but is there a way to pass a message too?
[10-Oct-2011 18:09:57] kocolosk is now known as kocolosk|away
[10-Oct-2011 19:35:04] kocolosk|away is now known as kocolosk
[10-Oct-2011 20:22:46] <syoma> hi
[10-Oct-2011 20:23:07] <syoma> I've installed distributed collector for zenoss core
[10-Oct-2011 20:23:22] <syoma> but the problem is when I need view graph
[10-Oct-2011 20:23:36] <syoma> for example, fping zenpack
[10-Oct-2011 20:23:56] <syoma> I've zenoss 3.2.0
[10-Oct-2011 20:24:30] <syoma> the integration of the collector work for me
[10-Oct-2011 20:25:23] <syoma> and another problem is when I can't view any performance graph of the collector...
[10-Oct-2011 20:25:40] <syoma> (sorry for my english guys)
[10-Oct-2011 20:26:44] <syoma> somebody alive?
[10-Oct-2011 20:26:52] <dopey__> yes, but i don't understand the question
[10-Oct-2011 20:27:02] <dopey__> plus, i don't know zenoss that well ... also just starting out with it :-)
[10-Oct-2011 20:28:39] <syoma> I can't see rrd graph of a zenpack when installing distributed collectors
[10-Oct-2011 20:28:40] <syoma> :/
[10-Oct-2011 20:29:43] <syoma> when the device model and associate it to the collector I can not see the chart rrd being that was built in the collector
[10-Oct-2011 20:32:14] <dopey__> after installing your ZenPack, did you (1) restart the zenoss stack, (2) associate the collector/model stuff with the device, and (3) remodel the device?
[10-Oct-2011 20:32:36] <dopey__> i think devices remodel them selves every 12 hours, but you can force it with a `zenmodeler` command
[10-Oct-2011 20:33:01] <dopey__> i.e.: zenmodeler run --now -d device_you_want_to_remodel_here
[10-Oct-2011 20:33:36] <syoma> the distributed collector not work with a zenoss stack 3.2.0
[10-Oct-2011 20:34:49] <syoma> dopey__: I try remodel the device
[10-Oct-2011 20:37:26] <syoma> also commands like snmpwalk, ping, do not work for the collector device
[10-Oct-2011 20:38:31] <syoma> but if I run these commands from the linux box running without any problems
[10-Oct-2011 20:40:15] <syoma> any ideas?
[10-Oct-2011 20:43:52] <syoma> These are created in the collector rrd: fping_avg.rrd fping_loss.rrd fping_max.rrd fping_min.rrd fping_rcv.rrd fping_xmt.rrd
[10-Oct-2011 20:44:10] <syoma> and I can view
[10-Oct-2011 20:44:46] <syoma> rrdtool xport --start now-600 --end now --step 300 DEF:xx=../192.168.255.1/fping_max.rrd:ds0:AVERAGE XPORT:xx |grep "<row>" <row><t>1318293300</t><v>2.8688814823e+00</v></row> <row><t>1318293600</t><v>NaN</v></row>
[10-Oct-2011 20:44:52] <syoma> ?!?!?1
[10-Oct-2011 20:44:55] <syoma> but
[10-Oct-2011 20:45:06] <syoma> I can't see the rrd graph !!!
[10-Oct-2011 20:46:21] <syoma> the graphic performance of the collector nor I can see
[10-Oct-2011 20:46:56] <syoma> but if I can add devices, refresh zenpacks..
[10-Oct-2011 20:47:02] <syoma> on the collector
[10-Oct-2011 20:50:05] <dopey__> hmm, not sure about that :-)
[10-Oct-2011 20:50:11] <dopey__> i haven't used the fping pack, myself
[10-Oct-2011 20:50:21] <dopey__> semms like it's collecting data ... that is odd
[10-Oct-2011 20:50:30] <dopey__> sorry, i'm relatively new to ZenOss
[10-Oct-2011 20:52:00] <syoma> ok thanks man
[10-Oct-2011 21:02:05] <syoma> I believe the problem is zenrender
[10-Oct-2011 21:02:09] <syoma> <type 'exceptions.Exception'>: Renderer 192.168.255.200:8091 unavailable
[10-Oct-2011 21:12:14] <dopey__> so ... i've come to the conclusion that making ZenPacks would be a lot easier if i knew Python ;-)
[10-Oct-2011 21:12:31] <dopey__> all i want to do is model a new indexed OID ... is that too much to ask? ;-) LOL
[10-Oct-2011 21:12:42] <dopey__> sheesh ... this is a big learning curve for a n00b
[10-Oct-2011 21:20:54] <bender183> dopey__, this is a big learning curve for anyone
[10-Oct-2011 21:33:23] <nyeates> .,/0;p;p
[10-Oct-2011 21:35:35] <syoma> nyeates: hola
[10-Oct-2011 21:37:23] <syoma> as I can get into the zenrender on zenoss 3.2.0 version for a remote collector?
[10-Oct-2011 21:37:37] <syoma> what's the url?
[10-Oct-2011 21:37:37] <syoma> XD
[10-Oct-2011 22:21:40] kocolosk is now known as kocolosk|away
[11-Oct-2011 00:00:01] [disconnected at Tue Oct 11 00:00:01 2011]
[11-Oct-2011 00:00:01] [connected at Tue Oct 11 00:00:01 2011]
[11-Oct-2011 00:00:01] [disconnected at Tue Oct 11 00:00:01 2011]
[11-Oct-2011 00:00:02] [connected at Tue Oct 11 00:00:02 2011]
[11-Oct-2011 00:00:13] [zenoss-logger (logger bot) has joined #zenoss]
[11-Oct-2011 05:03:32] ShaneB_ is now known as ShaneB
[11-Oct-2011 06:06:02] <khem_> hey
[11-Oct-2011 06:06:35] <khem_> can ZenOSS analyze logs by specified local files within the filesystems or only by pulling logs from syslog nodes?
[11-Oct-2011 07:49:38] <zaoul> sup
[11-Oct-2011 08:01:29] <zaoul> doin any monitoring lately?
[11-Oct-2011 08:31:11] <Hackman238> Morning all!
[11-Oct-2011 08:31:23] <Hackman238> zaoul: Thats what Zenoss does
[11-Oct-2011 08:56:23] <whyzgeek> good morning
[11-Oct-2011 08:58:03] <whyzgeek> Hackman238 I did the change you suggested, it improved the situation alot
[11-Oct-2011 08:58:12] <whyzgeek> thanks\
[11-Oct-2011 08:58:22] <whyzgeek> however it is not completely resolved
[11-Oct-2011 08:58:30] <Hackman238> whyzgeek: Glad it helped
[11-Oct-2011 08:58:43] <whyzgeek> probably rocket is going to point me to the patch
[11-Oct-2011 08:58:46] <Hackman238> whyzgeek: What are the new symptoms?
[11-Oct-2011 08:58:46] <whyzgeek> today
[11-Oct-2011 08:58:57] <Hackman238> whyzgeek: what version of Zenoss?
[11-Oct-2011 08:59:05] <whyzgeek> well, overall the modeling times dropped
[11-Oct-2011 08:59:12] <whyzgeek> signioficantly
[11-Oct-2011 08:59:19] <Hackman238> whyzgeek: Nice
[11-Oct-2011 08:59:25] <whyzgeek> but not all of the devices are like that
[11-Oct-2011 08:59:46] <whyzgeek> some take still more than half anout
[11-Oct-2011 08:59:50] <whyzgeek> hour
[11-Oct-2011 09:00:00] <Hackman238> whyzgeek: Wow.
[11-Oct-2011 09:00:23] <whyzgeek> how much ram your instance had?
[11-Oct-2011 09:00:36] <whyzgeek> I think we have memory issue as well
[11-Oct-2011 09:00:48] <whyzgeek> so I can't increase the caches that much
[11-Oct-2011 09:00:48] <Hackman238> whyzgeek: 256GB
[11-Oct-2011 09:00:56] <whyzgeek> wow man that's why!
[11-Oct-2011 09:01:01] <whyzgeek> I am running on 32gb
[11-Oct-2011 09:01:12] <Hackman238> whyzgeek: Oh yeah you need way more ram
[11-Oct-2011 09:01:22] <whyzgeek> good to know
[11-Oct-2011 09:01:49] <jmp242> how many things are you monitoring where 32GB isn't enough?
[11-Oct-2011 09:01:54] <whyzgeek> I think our size of zeodb is roughly the same as yours
[11-Oct-2011 09:02:23] <Hackman238> whyzgeek: Whats your object count?
[11-Oct-2011 09:02:52] <whyzgeek> well never calculated I think that it is aroughly around 1mil
[11-Oct-2011 09:03:14] <Hackman238> whyzgeek: Can you su zenoss; zendmd
[11-Oct-2011 09:03:22] <whyzgeek> yep
[11-Oct-2011 09:03:26] <Hackman238> whyzgeek: and paste 'len(zport.global_catalog)' without quotes?
[11-Oct-2011 09:04:55] <whyzgeek> AttributeError
[11-Oct-2011 09:04:56] <Hackman238> whyzgeek: or go to http://yourzenoss:8080/Control_Panel/Database/main/manage_cacheParameters
[11-Oct-2011 09:04:58] <whyzgeek> we run 2.5.2
[11-Oct-2011 09:05:08] <Hackman238> whyzgeek: Gotcha. The link method will work then
[11-Oct-2011 09:05:31] <Hackman238> whyzgeek: Total count should be first variable
[11-Oct-2011 09:05:57] <whyzgeek> 1698409
[11-Oct-2011 09:06:45] <Hackman238> whyzgeek: And the Data.fs packed is ~20GB?
[11-Oct-2011 09:06:53] <whyzgeek> yep
[11-Oct-2011 09:07:09] <Hackman238> Interesting. I expected a larger object count.
[11-Oct-2011 09:07:26] <Hackman238> Do you have a backup?
[11-Oct-2011 09:07:30] <whyzgeek> yep
[11-Oct-2011 09:07:50] <Hackman238> And it this critical?
[11-Oct-2011 09:07:57] <Hackman238> *is
[11-Oct-2011 09:08:14] <Hackman238> Rather, is web UI uptime critical?
[11-Oct-2011 09:08:15] <humanix> hi thjere
[11-Oct-2011 09:08:21] <Hackman238> humanix: Hey!
[11-Oct-2011 09:08:22] <whyzgeek> yes basically I had to turn off modelling for it
[11-Oct-2011 09:08:29] <whyzgeek> yes
[11-Oct-2011 09:08:37] <whyzgeek> eveyrbody is using it
[11-Oct-2011 09:08:37] <Hackman238> whyzgeek: Okay. Do you have zenfixit.py?
[11-Oct-2011 09:08:41] <whyzgeek> yes
[11-Oct-2011 09:08:45] <whyzgeek> shall i run it
[11-Oct-2011 09:08:49] <whyzgeek> I did that though
[11-Oct-2011 09:08:55] <whyzgeek> I didn't see that much effect
[11-Oct-2011 09:08:56] <humanix> how are you Hackman238
[11-Oct-2011 09:09:02] <whyzgeek> it found some IP address problems
[11-Oct-2011 09:09:06] <whyzgeek> but nothing major
[11-Oct-2011 09:09:08] <Hackman238> whyzgeek: did it bunch of orphans?
[11-Oct-2011 09:09:12] <Hackman238> whyzgeek: Gotcha.
[11-Oct-2011 09:09:20] <Hackman238> humanix: Well, and yourself?
[11-Oct-2011 09:09:29] <humanix> well
[11-Oct-2011 09:09:43] <humanix> wanting to sleep
[11-Oct-2011 09:09:56] <humanix> lol, i just have worked like 1h
[11-Oct-2011 09:10:03] <whyzgeek> so what figure do you have for teh number of objects?
[11-Oct-2011 09:10:52] <Hackman238> whyzgeek: 9808493
[11-Oct-2011 09:11:06] <whyzgeek> really! that's massive
[11-Oct-2011 09:11:20] <Hackman238> whyzgeek: Very tricky to fix LOL
[11-Oct-2011 09:11:33] <whyzgeek> so why the DB sizes are roughly the same
[11-Oct-2011 09:11:44] <whyzgeek> although I have to tell you most of these are
[11-Oct-2011 09:11:48] <whyzgeek> apps
[11-Oct-2011 09:11:54] <whyzgeek> so we have a zenpack
[11-Oct-2011 09:11:58] <Hackman238> humanix: Oh man. I know how you mean, I was woke up at 4AM CDT by a customer of mine!
[11-Oct-2011 09:12:06] <whyzgeek> which creates apps as compont of the devices
[11-Oct-2011 09:12:24] <Hackman238> whyzgeek: Gotcha, collecting software inventory, right?
[11-Oct-2011 09:12:27] <humanix> uhhh
[11-Oct-2011 09:12:34] <humanix> it hurts
[11-Oct-2011 09:12:45] <whyzgeek> not just that all apps and their KPIs
[11-Oct-2011 09:13:00] <whyzgeek> so something like Service Dynamics
[11-Oct-2011 09:13:07] <Hackman238> whyzgeek: Ah. I bet these objects are just very large
[11-Oct-2011 09:13:12] <whyzgeek> and their dependencies
[11-Oct-2011 09:13:13] <Hackman238> whyzgeek: That accounts for the size.
[11-Oct-2011 09:13:16] <whyzgeek> yes
[11-Oct-2011 09:13:21] <whyzgeek> they should be
[11-Oct-2011 09:13:44] <Hackman238> whyzgeek: You'll definately need to apply the zenhub patches
[11-Oct-2011 09:13:51] <whyzgeek> also they have very complex relationship map
[11-Oct-2011 09:14:02] <Hackman238> whyzgeek: Have you more than one zenhub running?
[11-Oct-2011 09:14:06] <whyzgeek> IC ya rocket promised to provide it
[11-Oct-2011 09:14:41] <Hackman238> whyzgeek: I can imagine. What means do you use to figure the dependancies?
[11-Oct-2011 09:15:02] <whyzgeek> well it is very long story
[11-Oct-2011 09:15:09] <whyzgeek> we had to use snmp is a different way
[11-Oct-2011 09:15:16] <whyzgeek> but we decided to open source the solution
[11-Oct-2011 09:15:29] <humanix> there is still no native rpm for EL6 of zenoss 3.2
[11-Oct-2011 09:15:31] <whyzgeek> so soon will put everything on teh github
[11-Oct-2011 09:15:51] <Hackman238> humanix: not yet
[11-Oct-2011 09:16:25] <Hackman238> whyzgeek: Very cool. I'm interested to look at it if you could take my email address and let me know when its up on github
[11-Oct-2011 09:16:38] <whyzgeek> sure by all means
[11-Oct-2011 09:16:51] <whyzgeek> we are just in the process of cleaning up and adding docs
[11-Oct-2011 09:17:02] <humanix> is this on roadmap or something Hackman238 ?
[11-Oct-2011 09:17:03] <whyzgeek> so i guess in a couple of weeks
[11-Oct-2011 09:17:25] <Hackman238> whyzgeek: TY. hackman238@gmail.com or shane.scott@rackspace.com
[11-Oct-2011 09:17:53] <Hackman238> humanix: I haven't heard anything official, but I suspect it will be about the time of v4.x Core beta release
[11-Oct-2011 09:18:06] <whyzgeek> thanks sure will do
[11-Oct-2011 09:18:42] <Hackman238> Darn...meeting. BRB
[11-Oct-2011 09:19:09] <humanix> i see
[11-Oct-2011 09:19:15] <humanix> it can take time
[11-Oct-2011 09:24:01] <humanix> so i see zenoss core is not supported under el6
[11-Oct-2011 09:24:06] <humanix> not even the stack
[11-Oct-2011 09:28:34] <humanix>
[11-Oct-2011 09:32:54] <jmp242> EL6 is pretty new
[11-Oct-2011 09:33:15] <jmp242> I would expect maybe around 4.x EL6 support, but I don't know anything official
[11-Oct-2011 09:34:39] <humanix> but for example
[11-Oct-2011 09:34:50] <humanix> ubuntu 10.04 is almost that new
[11-Oct-2011 09:35:00] <humanix> and it has deb packages
[11-Oct-2011 09:35:12] <fragfutter> jmp242: rhel6 is close to a year
[11-Oct-2011 09:35:49] <fragfutter> jmp242: i would guess that the guys at zenoss had to wait for centos6 before starting to work....
[11-Oct-2011 09:35:58] <fragfutter> and centos6 was late
[11-Oct-2011 09:37:01] <jmp242> yea but no major Zenoss release
[11-Oct-2011 09:37:04] <jmp242> for Core
[11-Oct-2011 09:37:16] <jmp242> so, I wouldn't expect a 3.2 to focus on a new OS for instance
[11-Oct-2011 09:37:23] <jmp242> plus, they could have used SL
[11-Oct-2011 09:48:10] <SDuensin> Morning.
[11-Oct-2011 09:49:47] <ali3n0> hi folks. I'm in the process of evaluating a monitoring tool for this new venture and I'm new to zenoss. My first question is: is there a way to automate device adding / removing from ec2 pools into zenoss?
[11-Oct-2011 09:51:24] <Hackman238> ali3n0: Absolutely
[11-Oct-2011 09:51:31] <Hackman238> ali3n0: It's tested and works very well.
[11-Oct-2011 09:51:42] <Hackman238> ali3n0: See the Amazon EC2 Zenpack
[11-Oct-2011 09:51:51] <ali3n0> I'm running through the admin guide Hackman238 so far nothing about that
[11-Oct-2011 09:52:09] <ali3n0> Hackman238, must instances be running cloudwatch?
[11-Oct-2011 09:52:12] <Hackman238> ali3n0: It wont be in that guide since its an extension
[11-Oct-2011 09:52:19] <ali3n0> got it
[11-Oct-2011 09:52:48] <Hackman238> ali3n0: Not certain.
[11-Oct-2011 09:53:05] <ali3n0> I'll have a look at that, thanks Hackman238
[11-Oct-2011 09:53:16] <Hackman238> ali3n0: NP
[11-Oct-2011 09:55:06] kocolosk|away is now known as kocolosk
[11-Oct-2011 10:42:27] <bseklecki> 'Component templates should not be manually bound.'
[11-Oct-2011 10:42:36] <bseklecki> exept in case that discovery is fucking broken?
[11-Oct-2011 10:43:09] <Hackman238> bseklecki: component tempaltes are assigned by modellers
[11-Oct-2011 10:43:28] <Hackman238> bseklecki: You'll need to fix the modeller that isn't assigning for the broken case
[11-Oct-2011 10:43:33] <Hackman238> bseklecki: Whats not working?
[11-Oct-2011 10:43:38] <bseklecki> i need to fighure out how modelers apply index-based OIDs
[11-Oct-2011 10:43:48] <bseklecki> and now that is stored in the database, and how to troubleshoot it
[11-Oct-2011 10:45:07] <bseklecki> zenoss.snmp.UCDHardDiskMap
[11-Oct-2011 10:45:35] <bseklecki> it discovered entries by crawling HR-SYSTEM-INFO::diskInfoTable apparently
[11-Oct-2011 10:46:05] <Hackman238> bseklecki: Okay. Whats the problem?
[11-Oct-2011 10:46:09] <bseklecki> but it'snot reading the respective SNMP mibs for IO operations, 'No graph Data'
[11-Oct-2011 10:46:37] <bseklecki> message/61770#61770
[11-Oct-2011 10:46:42] <Hackman238> bsesounds like the template isnt hitting the right oid's for the target platform
[11-Oct-2011 10:46:59] <Hackman238> bseklecki: Zenoss only uses loaded mibs for trap translation
[11-Oct-2011 10:47:28] <bseklecki> oh, well okay that sorta makes sense
[11-Oct-2011 10:47:45] <bseklecki> i mean the .py file for zenoss.snmp.UCDHardDiskMap has a numeric OID to the table's root
[11-Oct-2011 10:49:07] <Hackman238> bseklecki: Yeah the modeller just creates the objects and should set the .index property of created objects to the snmp index. When the objects are polled by snmp templates the .index will be appended to the oid to be polled automatically
[11-Oct-2011 10:49:35] <Hackman238> bseklecki: I would double check the templates.
[11-Oct-2011 10:49:37] <bseklecki> yea I was justa about to go find a technote/design guide ont hat feature
[11-Oct-2011 10:50:42] <bseklecki> Its iether /opt/zenoss/Products/ZenModel/HardDisk.py or /opt/zenoss/Products/DataCollector/plugins/zenoss/snmp/UCDHardDiskMap.py
[11-Oct-2011 10:51:06] <bseklecki> GetTableMap('diskIOTable', '.1.3.6.1.4.1.2021.13.15.1.1', {
[11-Oct-2011 10:53:20] <bseklecki> snmptable -v2c -c foo FQDN 1.3.6.1.4.1.2021.13.15.1
[11-Oct-2011 10:53:43] <bseklecki> 15.1 is the root of the table, 15.1.1 is....i'd have to walk it
[11-Oct-2011 10:55:03] <bseklecki> FYI I'm using the worlds most generic SNMP agent: The Net-SNMP that ships with RHEL5 and its generic HOST-RESOURCES MI
[11-Oct-2011 11:00:39] <Hackman238> bseklecki: Are all the volumes you expect listed on the device in Zenoss?
[11-Oct-2011 11:01:28] <Hackman238> bseklecki: If they are I would check the templates found under 'Advanced' -> 'Monitoring Temnplates' -> 'Filesystems'
[11-Oct-2011 11:10:02] <bseklecki> quick note: disks v.s. filesystems
[11-Oct-2011 11:10:10] <bseklecki> there's a different table for filesystems
[11-Oct-2011 11:11:09] <bseklecki> drives are...UCD-DISKIO-MIB::diskIOTable
[11-Oct-2011 11:11:30] <bseklecki> filesystems are: HOST-RESOURCES-MIB::hrStorageTable
[11-Oct-2011 11:14:13] <Hackman238> bseklecki: yeah
[11-Oct-2011 11:15:50] <Hackman238> bseklecki: OID .1.3.6.1.4.1.2021.13.15.1.1 is appended with each {.2, .3, .4, .5, and .6} in the modeller
[11-Oct-2011 11:16:07] <bseklecki> yea i figured it was doing some magic
[11-Oct-2011 11:16:34] <bseklecki> I can explore the tree and see what index values it has associated from discovery
[11-Oct-2011 11:16:38] <Hackman238> bseklecki: modelling does not create graphs
[11-Oct-2011 11:16:49] <bseklecki> err, My next step would be to * ^^
[11-Oct-2011 11:16:58] <Hackman238> bseklecki: only objects which are assigned templates which are polled to create graphs.
[11-Oct-2011 11:17:15] <bseklecki> ahh it creates objects
[11-Oct-2011 11:18:56] <bseklecki> I'm gonna have to run to a meeting soon
[11-Oct-2011 11:19:42] <bseklecki> so there's an automatic binding based on the discovered object from the discovery plugin
[11-Oct-2011 11:20:01] <Hackman238> bseklecki: Yep
[11-Oct-2011 11:20:42] <bseklecki> so like in ethernetCsmacd_64 the OID listed is ifHCInOctets\
[11-Oct-2011 11:20:51] <bseklecki> 1.3.6.1.2.1.31.1.1.1.6
[11-Oct-2011 11:20:55] <bseklecki> let me snmptranslate that
[11-Oct-2011 11:21:18] <bseklecki> $ snmptranslate -Oa 1.3.6.1.2.1.31.1.1.1.6
[11-Oct-2011 11:21:18] <bseklecki> IF-MIB::ifHCInOctets
[11-Oct-2011 11:21:45] <bseklecki> ahhh; now to take the 5 dried grams of mushrooms that came with my copy of ZenOSS
[11-Oct-2011 11:21:52] <Hackman238> bseklecki: yep. and then the object its applied to is polled the objects .index is appeneded
[11-Oct-2011 11:22:17] <Hackman238> bseklecki: One note, Zenoss does not like translated oids. stick with numerical format.
[11-Oct-2011 11:22:24] <cluther> It would have probably made more sense to set the OID in those datasources to something like 1.3.6.1.2.1.31.1.1.1.6.${here/snmpindex}
[11-Oct-2011 11:22:38] <cluther> It would have been less magical.
[11-Oct-2011 11:22:52] <Hackman238> bseklecki: LOL
[11-Oct-2011 11:23:06] <bseklecki> well this is a Fresh Install(R) and the 'HardDisk' template for /Server/Linux only has 'reads' with no OID filed
[11-Oct-2011 11:23:43] <Hackman238> bseklecki: Thats weird
[11-Oct-2011 11:24:00] <bseklecki> 1.3.6.1.4.1.2021.13.15.1.1.3 is the diskIOTable prefix
[11-Oct-2011 11:24:27] <bseklecki> well no
[11-Oct-2011 11:24:28] <bseklecki> $ snmptranslate -On UCD-DISKIO-MIB::diskIOTable
[11-Oct-2011 11:24:29] <bseklecki> .1.3.6.1.4.1.2021.13.15.1
[11-Oct-2011 11:24:39] <bseklecki> Hackman238: hey man, I just work here.
[11-Oct-2011 11:25:17] <Hackman238> bseklecki: LOL
[11-Oct-2011 11:25:33] <Hackman238> bseklecki: NYS?
[11-Oct-2011 11:35:51] <cgibbons> hmmm
[11-Oct-2011 11:38:53] <bseklecki> gotta get on a conference call w/ HP, I'll try adjusting the OID prefixes in a moment
[11-Oct-2011 11:49:27] <bseklecki> yea what do you kids see under 'HardDisk --> /Server/Linux'
[11-Oct-2011 11:51:26] kocolosk is now known as kocolosk|away
[11-Oct-2011 11:51:48] kocolosk|away is now known as kocolosk
[11-Oct-2011 12:00:01] [disconnected at Tue Oct 11 12:00:01 2011]
[11-Oct-2011 12:00:02] [connected at Tue Oct 11 12:00:02 2011]
[11-Oct-2011 12:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[11-Oct-2011 12:29:22] <bseklecki> hmmm getting somewhere
[11-Oct-2011 12:30:13] <bseklecki> did I hear that Counter64 isn't supported yet?
[11-Oct-2011 12:31:00] <rocket> bseklecki: where would you hear that Counter64 isnt supported?
[11-Oct-2011 12:31:16] <bseklecki> probably a 6 year old forum post
[11-Oct-2011 12:31:35] <bseklecki> but the internal SNMP definitely isn't linked to or based on Net-SNMP libnetsnmp?
[11-Oct-2011 12:31:45] <rocket> 6 year old forum posts are an eternity
[11-Oct-2011 12:32:05] <bseklecki> [rubs face]
[11-Oct-2011 12:33:39] <rocket> we can handle any 64bit oid we poll
[11-Oct-2011 12:34:03] <bseklecki> you guys were getting started on ZenOSS around the time i wrote this wiki page apparently: http://oss.oetiker.ch/mrtg-trac/wiki/Net-SNMP#MonitoringDiskIOActivity
[11-Oct-2011 12:34:04] <rocket> we leverage twisted-snmp implementation
[11-Oct-2011 12:35:11] <bseklecki> So is it safe to assume that If I buy commercial ZenOSS I get a whole fuck-load more discovery plugins >:} ?
[11-Oct-2011 12:35:42] <bseklecki> because the SNMP agent that comes with AIX 7 and HP-UX 11iV3 doesn't support _shit_
[11-Oct-2011 12:36:09] <rocket> we ssh model those for that reason
[11-Oct-2011 12:36:23] <bseklecki> no doubt
[11-Oct-2011 12:36:47] <bseklecki> i was about to start reading about OpenView Performance Agent to see how it tries to glean data
[11-Oct-2011 12:43:08] <bseklecki> I cooked up a dozen snmpd passthru OIDs for getting advanced data out of the vmstat(1)/VM subsystem on linux/bsd/aix/hfux
[11-Oct-2011 12:43:40] <bseklecki> context switches, interrups, swap ins, swap outs, run queue, block queue, scan rate, page fault rate
[11-Oct-2011 12:44:26] <bseklecki> it was 'obvious' to me how to setup new OID collections in a new Template, with graphs and thresholds
[11-Oct-2011 12:45:13] <bseklecki> but the per-Device dicovery plugins and per-device index-based OIDS weren't so obvious; the only breadcrum I can see now is that you can browse the associated template from the Component's drop down menu
[11-Oct-2011 12:47:55] <bseklecki> but Monitoring Templates for Component's aren't diferentiated in any way so you would never know that there's auto-append magic waiting for you without trying to compare to ethernetCsmacd and snmptranslate(1) the OID
[11-Oct-2011 12:49:49] <whyzgeek> hi rocket
[11-Oct-2011 12:50:27] <rocket> hi whyzgeek
[11-Oct-2011 13:38:57] <bseklecki> rocket/Hackman238: the graphs are drawing....I'm not sure that they're honoring the COUNTER type though, or maybe I need to use the Counter64
[11-Oct-2011 13:39:23] <bseklecki> (BTW if its not already apparent, that's what I meant to say earlier s/32/64/g)
[11-Oct-2011 13:57:01] <Fr_mon> Hi Everyone, I'm looking for the 3.1 stack x86 installer on the downloads page at zenoss. All i see is the 3.2 version...
[11-Oct-2011 13:57:13] <Fr_mon> Can someone please send me a link?
[11-Oct-2011 13:58:36] <Hackman238> bseklecki: Nice
[11-Oct-2011 13:59:34] <Hackman238> Fr_mon: http://sourceforge.net/projects/zenoss/files/zenoss-3.1/zenoss-3.1.0/zenoss-stack-3.1.0-linux-x64.bin
[11-Oct-2011 14:01:32] <Fr_mon> Thanks Hackman. Is there a 32bit version?
[11-Oct-2011 14:01:47] <Fr_mon> ill take a look at sourceforge
[11-Oct-2011 14:02:25] <Hackman238> Fr_mon: http://downloads.sourceforge.net/zenoss/zenoss-stack-3.1.0-linux.bin
[11-Oct-2011 14:02:38] <Fr_mon> thanks
[11-Oct-2011 14:03:44] <Fr_mon> im all set. thanks again hack!
[11-Oct-2011 14:04:09] <Hackman238> Fr_mon: NP
[11-Oct-2011 14:26:21] kocolosk is now known as kocolosk|away
[11-Oct-2011 15:08:32] kocolosk|away is now known as kocolosk
[11-Oct-2011 16:07:33] <jmp242> Hackman238: you ever see anyone use syslog as a datasource per this thread? message/61789#61789
[11-Oct-2011 16:07:56] <jmp242> I can't imagine doing anything easy, maybe something that parses the event database and generates RRDs that way?
[11-Oct-2011 16:10:16] <Hackman238> jmp242: Definately not an easy task
[11-Oct-2011 16:11:48] <jmp242> OT, I got Backtrack 5R1 working finally on a W510 with an Alpha1 adapter YAY
[11-Oct-2011 16:13:02] <Hackman238> Nice
[11-Oct-2011 16:13:58] <jmp242> yea, wireshark against wifi
[11-Oct-2011 16:14:25] <davetoo> Wireshark is a pig. A very nice pig, though
[11-Oct-2011 16:14:44] <davetoo> it takes *forever* to build from source because it's gotten so large.
[11-Oct-2011 16:15:08] <jmp242> well, I don't build from source
[11-Oct-2011 16:15:28] <davetoo> but it and/or tshark are lifesavers
[11-Oct-2011 16:16:01] * davetoo used to be a BSD-ite
[11-Oct-2011 16:16:24] <jmp242> I see
[11-Oct-2011 16:16:27] <davetoo> freebsd.. and tended to build from ports instead of using packages.
[11-Oct-2011 16:17:18] <jmp242> I'm mostly RHEL (SL actually due to work), but every random pre-built is now seemingly debian based
[11-Oct-2011 16:28:53] <humanix> Im monitoring a switch
[11-Oct-2011 16:29:05] <humanix> with ethernetCsmacd_64 template
[11-Oct-2011 16:29:13] <humanix> and i see 13.46m errors
[11-Oct-2011 16:29:20] <humanix> on a physical port
[11-Oct-2011 16:29:29] <humanix> it is equal to 13 or to 13.000?
[11-Oct-2011 16:34:01] <dhopp> humanix: that's 13.46m. m is an SI unit for 10^-3 so I think that would be 0.0013
[11-Oct-2011 16:36:53] <bseklecki> damn none of the Amazon EC2 NetBSD images will run in Micro or Small size
[11-Oct-2011 16:37:00] <bseklecki> errr wrong chan
[11-Oct-2011 16:51:58] <pplask> guys
[11-Oct-2011 16:52:15] <pplask> i can pass the variable $RECIPIENT for the pager command
[11-Oct-2011 16:52:28] <pplask> is there a way to pass the message to the pager command too?
[11-Oct-2011 16:52:43] <pplask> something like $PROBLEM
[11-Oct-2011 17:14:55] <humanix> thanks dhopp
[11-Oct-2011 19:30:51] meddle- is now known as bender183
[11-Oct-2011 20:35:06] dec3pti01 is now known as dec3pti0n
[11-Oct-2011 21:14:56] kocolosk is now known as kocolosk|away
[11-Oct-2011 23:40:13] <c00p> Sam-I-Am: Once I have imported all the mibs and I can see them in advanced MIBS is there anything else I need to do for traps to be recongnised now?
[11-Oct-2011 23:40:33] <Sam-I-Am> restart everything
[12-Oct-2011 00:00:01] [disconnected at Wed Oct 12 00:00:01 2011]
[12-Oct-2011 00:00:02] [connected at Wed Oct 12 00:00:02 2011]
[12-Oct-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[12-Oct-2011 00:35:16] tvincent_ is now known as tvincent
[12-Oct-2011 01:22:16] <c00p> Sam-I-Am: ta - tho even grepping through the mib files I can't find the OSPF mib it seems I am getting from an SRX240
[12-Oct-2011 01:22:19] <c00p> 1.3.6.1.2.1.14.16.2.16
[12-Oct-2011 01:22:31] <c00p> will get the standard mib from a website and import tant
[12-Oct-2011 01:22:41] <c00p> that seems to have that OID
[12-Oct-2011 03:58:53] <DesiJat> hi all, is the Zenoss Enterprise release based on the same base version as the latest open source release?
[12-Oct-2011 07:12:42] gareth__ is now known as NetworksAreMadeO
[12-Oct-2011 07:12:48] NetworksAreMadeO is now known as NetworkString
[12-Oct-2011 07:42:50] <enkrypt> I'm having a problem importing a Cisco MIB via download, it has a dependency on another Cisco mib (that I imported succesfully already) but it keeps complaining that it can't find the dependency... I'm talking about ftp://ftp.cisco.com/pub/mibs/v2/CISCO-FIREWALL-MIB.my
[12-Oct-2011 08:27:23] <enkrypt> hi, is anyone here familiar with the Zenoss JSON API ? I'm writing some python to access it, but having a trouble retrieving my device list. getDevices results in {u'message': u"AttributeError 'NoneType' object has no attribute 'iteritems'", u'type': u'exception'}
[12-Oct-2011 08:27:44] <enkrypt> other functions work, like getEvents and getCollectors
[12-Oct-2011 08:52:42] <ali3n0> hi folks. I'm new to zenoss, can't find out if I should get graphs for ssh monitored boxes or it's just an snmp feature
[12-Oct-2011 08:54:49] <jmp242> depends on the templates - I've not tried the SSH ones...
[12-Oct-2011 09:00:04] <ali3n0> should I install a plugin or something on the monitored device?
[12-Oct-2011 09:05:39] <jmp242> I wouldn't think so
[12-Oct-2011 09:10:55] <wayne____> enkrypt: try clearing your browser cache and restarting (the browser)
[12-Oct-2011 09:16:26] <Sam-I-Am> c00p: there was a lot of seeking out oid numbers when fine-tuning MIBS
[12-Oct-2011 09:17:16] <Sam-I-Am> c00p: i also threw the same mibs into snmpwalk/snmptrapd so i could see what was "missing" ... net-snmp detects dependency problems
[12-Oct-2011 09:18:21] <Sashness> has anyone seen an issue where the filesystem free bytes are wrong causing zenoss to think usage is above %100?
[12-Oct-2011 09:19:04] <enkrypt> wayne____: I'm talking about python scripts here, not using a browser
[12-Oct-2011 09:21:32] kocolosk|away is now known as kocolosk
[12-Oct-2011 09:22:37] <Sashness> modeling the device seems to fix the issue however thats pretty annoying
[12-Oct-2011 09:28:43] <enkrypt> so nobody here has worked with the Zenoss JSON API in a Python context before? Simply requesting a list of all devices fails for me.
[12-Oct-2011 09:40:23] <Hackman238> Morning all!
[12-Oct-2011 09:40:44] <humanix> hi Hackman238
[12-Oct-2011 09:40:51] <humanix> morning
[12-Oct-2011 09:41:16] <Hackman238> Unfortuantely I'll only be available on a spare basis today. Zenoss v4.1 migrations
[12-Oct-2011 09:41:28] <Hackman238> *sparse
[12-Oct-2011 09:43:28] <wayne____> enkrypt: oops, missed that part (lots of people lately with an identical looking error from the browser). I haven't tried programming against it through the web - I find it easier to go direct to dmd from the server
[12-Oct-2011 09:45:05] <Sam-I-Am> Hackman238: moo.
[12-Oct-2011 09:45:22] <dhopp> so should I be worried that a vendor gave us a 32 bit library and when we asked for a 64 bit the guy e-mails us the source with a makefile and asks us to compile it?
[12-Oct-2011 09:45:41] <Sam-I-Am> at least its open source now?
[12-Oct-2011 09:48:05] <enkrypt> wayne____: yeah I guess dmd would be easier, but I read somewhere the JSON/REST api is the "official way to go" now
[12-Oct-2011 09:50:27] <kerick> any idea why when my script updates the id on a custom property nothing shows up in its' text field in the web interface?
[12-Oct-2011 09:50:33] <kerick> the id shows up in dmd
[12-Oct-2011 09:53:10] <Hackman238> wayne____: REST is depricated in v4, JSON/XMLRPC the officiated API
[12-Oct-2011 09:53:41] <Hackman238> wayne____: DMD more for scripting changes, maintenance, etc.
[12-Oct-2011 09:54:27] <Hackman238> wayne____: Be cautioned about remote DMD operations...DMD connections are automatically admin. Anyone on the network with a copy of Zenoss could zendmd -h {your zeo server} and tear it all down
[12-Oct-2011 09:55:27] <wayne____> Hackman238: I only use it locally. My last experience with programming against a web API was unpleasant.
[12-Oct-2011 09:55:48] <Hackman238> wayne____: Wayne from NTI, right?
[12-Oct-2011 09:55:56] <wayne____> Hackman238: nope
[12-Oct-2011 09:56:01] <Hackman238> wayne____: The API has improved significatnly
[12-Oct-2011 09:56:07] <Hackman238> wayne____: My bad.
[12-Oct-2011 09:56:12] <enkrypt> Hackman238: where would I go to if I'm having problems with the JSON/XMLRPC api ? I'm trying to do getDevices in python, using the example API code, and that fails. other stuff (like getEvents or getCollectors) works fine.
[12-Oct-2011 09:56:27] <wayne____> Hackman238: np. The API I'm referring to wasn't Zenoss
[12-Oct-2011 09:57:01] <Hackman238> enkrypt: I would grep the Products forlder for the 'def getDevices' and 'def getEvents' and compare how they return, thats probably the problem
[12-Oct-2011 09:57:10] <Hackman238> wayne____: Oh gotcha.
[12-Oct-2011 09:57:40] <Hackman238> wayne____: Zenoss definately gets a gold star for most imporved in the inconsistant API area LOL
[12-Oct-2011 10:02:01] <Hackman238> nyeates: Hey, how goes it?
[12-Oct-2011 10:02:17] <Sam-I-Am> yo
[12-Oct-2011 10:02:26] <Hackman238> Sam-I-Am: Woof
[12-Oct-2011 10:02:53] <Sam-I-Am> yep
[12-Oct-2011 10:03:05] <Sam-I-Am> trying to caffeinate
[12-Oct-2011 10:03:39] <Hackman238> Sam-I-Am: Same here....just ran in to a SegFault during a migration and I'm trying to quench the rage with coffee
[12-Oct-2011 10:04:36] <Sam-I-Am> ewww
[12-Oct-2011 10:04:41] <Sam-I-Am> migrating from/to what?
[12-Oct-2011 10:05:00] <nyeates> Hackman238: hey there
[12-Oct-2011 10:05:26] <Hackman238> Sam-I-Am: Zenoss v4.0.2->v4.1
[12-Oct-2011 10:05:31] <Hackman238> nyeates: How goes it?
[12-Oct-2011 10:05:46] <nyeates> In Tejas
[12-Oct-2011 10:05:59] <Hackman238> Sam-I-Am: It's not a test either, live migration. Mgmt will be raging soon too.
[12-Oct-2011 10:06:09] <nyeates> All-hands fly in meeting - entire company
[12-Oct-2011 10:06:10] <Hackman238> nyeates: Oh cool. You're in luck its been cooler lately
[12-Oct-2011 10:06:36] <nyeates> We are in training all day tomorrow - so no IRC dev meeting
[12-Oct-2011 10:06:44] <Hackman238> nyeates: Alrighty
[12-Oct-2011 10:06:48] <Sam-I-Am> Hackman238: and i'm guessing it didnt segfault in practice?
[12-Oct-2011 10:07:04] <Hackman238> nyeates: If you bump in to any support guys I just put in a sev=high ticket for a migration blow up
[12-Oct-2011 10:07:16] <Hackman238> Sam-I-Am: You'd be correct
[12-Oct-2011 10:07:17] <Hackman238> LOL
[12-Oct-2011 10:07:23] <Sam-I-Am> figures
[12-Oct-2011 10:07:39] * Sam-I-Am hasnt even messed with zenoss 4
[12-Oct-2011 10:07:53] * Sam-I-Am puts on his old skool hat
[12-Oct-2011 10:07:56] <Hackman238> Sam-I-Am: Its coming like a train with no brakes!
[12-Oct-2011 10:08:08] <Sam-I-Am> fascinating
[12-Oct-2011 10:08:21] <Sam-I-Am> i could never get my company to let me roll zenoss for customers
[12-Oct-2011 10:08:30] <Hackman238> Sam-I-Am: As much as its tricky to migrate sometimes, its a huge jump in awesomeness.
[12-Oct-2011 10:08:37] <Hackman238> nyeates: TY
[12-Oct-2011 10:09:46] <Sam-I-Am> Hackman238: so you couldnt get around the segfault?
[12-Oct-2011 10:10:34] <Hackman238> Sam-I-Am: A successive zenoss restart gave me 5 pages of trace
[12-Oct-2011 10:10:43] <dhopp> Hackman238: don't you have backups you can restore to? ;-)
[12-Oct-2011 10:10:56] <Hackman238> Sam-I-Am: If it ate my preupgraded instance I'll be pissed since it'll put me behind schedule.
[12-Oct-2011 10:11:06] <Hackman238> dhopp: Oh yeah, its just the preupgrade box
[12-Oct-2011 10:11:38] <Hackman238> dhopp: But our DC's are way to large for me to try again before talking with support while in the maint window
[12-Oct-2011 10:12:17] <Hackman238> dhopp: Even HongKong DC takes an entire work day to migrate from v2.5.2->v3.2->v4.0.2
[12-Oct-2011 10:12:42] <Sam-I-Am> thats a big jump
[12-Oct-2011 10:13:01] <Hackman238> Sam-I-Am: Aye it is
[12-Oct-2011 10:13:18] <Hackman238> Sam-I-Am: And with great migrations comes great pain
[12-Oct-2011 10:13:19] <Hackman238> LOL
[12-Oct-2011 10:17:00] <dhopp> Hackman238: maybe you shouldn't wait so long between upgrades :-P
[12-Oct-2011 10:17:27] <Hackman238> dhopp: We like v2.5.2 that much
[12-Oct-2011 10:42:27] <enkrypt> an error that has been showing up in the GUI of Zenoss 3.2.0 enterprise, when switching between event history and events: AttributeError 'NoneType' object has no attribute 'replace'
[12-Oct-2011 10:42:29] <ion-fan> hackman238: did you get a look on my problems with these transforms and global search?
[12-Oct-2011 10:47:52] lederka is now known as bender183
[12-Oct-2011 11:15:29] <enkrypt> Hackman238: remember my problem with JSON API earlier on? The thing is, getDevices() takes an optional argument params which is a dict with search parameters. the problem is, if you don't define that variable, the code simply breaks.
[12-Oct-2011 12:00:01] [disconnected at Wed Oct 12 12:00:01 2011]
[12-Oct-2011 12:00:02] [connected at Wed Oct 12 12:00:02 2011]
[12-Oct-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[12-Oct-2011 12:00:15] <sendak.freenode.net> [freenode-info] why register and identify? your IRC nick is how people know you. http://freenode.net/faq.shtml#nicksetup
[12-Oct-2011 12:35:09] <dhopp> hmm..setting zenhub cache size to a value that causes zenhub to use 2.7 GB of memory on a server with only 3 GB RAM (dev box) probably wasn't the smartest idea
[12-Oct-2011 12:37:01] <nyeates> IRC dev tomorrow @ 11am EDT will be held.
[12-Oct-2011 12:37:49] <nyeates> I had formerly cancelled it, due to company-wide events here....We will still hold it though
[12-Oct-2011 12:38:11] <nyeates> and Chet, a zenoss uber master, will be in the house
[12-Oct-2011 14:32:17] <nexex> bigegor: is there a way to do a cmdline test of an SMI query, I get 401 errors when try to model, but I am able to connect using cimcli command
[12-Oct-2011 14:33:01] <bigegor> nexex: wbemcli
[12-Oct-2011 14:58:50] <mahimahi> Zenoss 3.2.0 - having problem with DERIVE graphs. Graph look like GAUGE. I have other older DERIVE graphs that work, but not my newly created ones. Anyone else running into this issue? Is there a patch? What would be the steps to debug this issue? Thanks in advance
[12-Oct-2011 15:19:23] <hmp> mahimahi: did you patch zenoss
[12-Oct-2011 15:48:16] <enkrypt> this afternoon I added a very big amount of SNMP MIB's to zenoss (like thousands of them, downloaded from the Cisco FTP), could this have a negative impact on performance?
[12-Oct-2011 15:52:28] <jmp242> enkrypt: shouldn't - those are only used for SNMP trap processing
[12-Oct-2011 15:52:40] <jmp242> How many traps do you get normally?
[12-Oct-2011 15:53:36] <enkrypt> jmp242: I get very few traps atm
[12-Oct-2011 15:54:22] <enkrypt> jmp242: it's just that I'm seeing performance problems this afternoon, and I have only 2 things to relate it to: I added a big amount of SNMP MIB's to my system (zenoss 3.2.0 enterprise), and I see in the graphs that I am getting many syslogs (like thousands currently)
[12-Oct-2011 15:57:08] <jmp242> syslogs
[12-Oct-2011 15:57:10] <jmp242> are far more likely
[12-Oct-2011 15:57:19] <jmp242> Zenoss isn't a "high performance" syslog reciever
[12-Oct-2011 15:57:30] <enkrypt> ok thanks, I was thinking in that direction, just wanted to make sure
[12-Oct-2011 15:57:36] <jmp242> often users front end zenoss with the standard syslog server to filter it out
[12-Oct-2011 15:57:41] <enkrypt> I'll have to place an rsyslog or syslog-ng in front of zenoss
[12-Oct-2011 15:57:53] <enkrypt> yes exactly, that's what was recommended by Zenoss professional services
[12-Oct-2011 15:58:02] <rocket> jmp242: I would say pre-4.X that is very true .. we are definalty making huge strides in the event system
[12-Oct-2011 15:58:31] <enkrypt> thanks for verifying this info
[12-Oct-2011 15:59:06] <rocket> enkrypt: we typically have customers filter syslogs if possible before going to zenoss.
[12-Oct-2011 15:59:14] <rocket> enkrypt: pre-4.X
[12-Oct-2011 15:59:36] <enkrypt> rocket: exactly that's what was recommended to me by David Carmean. I'll set up rsyslog or syslog-ng tomorrow (10pm over here atm)
[12-Oct-2011 15:59:37] <rocket> enkrypt: there is an option to filter items out of zensyslog etc
[12-Oct-2011 15:59:48] <zaoul> ive been trying to use sagan for syslog monitoring, I like the rule based approach
[12-Oct-2011 15:59:51] <enkrypt> rocket: very curious about 4.x by the way
[12-Oct-2011 15:59:54] <enkrypt> brb
[12-Oct-2011 16:05:02] <mahimahi> hmp: yes I did all 3 patches
[12-Oct-2011 16:05:23] <jmp242> So 4.x might handle more events / sec?
[12-Oct-2011 17:27:12] kocolosk is now known as kocolosk|away
[12-Oct-2011 17:51:03] <c00p> mibs should be easy but never are.
[12-Oct-2011 19:29:48] meddle- is now known as bender183
[12-Oct-2011 20:07:50] DesiJat is now known as LouisLin
[12-Oct-2011 20:08:11] LouisLin is now known as DesiJat
[12-Oct-2011 20:42:38] kocolosk|away is now known as kocolosk
[12-Oct-2011 22:11:10] kocolosk is now known as kocolosk|away
[12-Oct-2011 22:29:33] kocolosk|away is now known as kocolosk
[13-Oct-2011 00:00:01] [disconnected at Thu Oct 13 00:00:01 2011]
[13-Oct-2011 00:00:01] [connected at Thu Oct 13 00:00:01 2011]
[13-Oct-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[13-Oct-2011 00:05:11] kocolosk is now known as kocolosk|away
[13-Oct-2011 08:30:43] <jmp242> Morning all
[13-Oct-2011 08:31:10] <jmp242> I've just noticed something unexpected (for me anyway) with Event Commands and trying to run sudo ___SOMECOMMAND___
[13-Oct-2011 08:31:30] <jmp242> I get 2011-10-12 22:23:04,160 ERROR zen.ZenActions: Command NtopRestart says sudo: sorry, you must have a tty to run sudo
[13-Oct-2011 08:31:37] <jmp242> Is anyone doing this successfully?
[13-Oct-2011 09:02:01] <jmp242> Still too early on a Thursday?
[13-Oct-2011 09:08:40] <froztbyte> it's 15h00 here
[13-Oct-2011 09:08:42] * froztbyte reads
[13-Oct-2011 09:09:11] <jmp242> lol
[13-Oct-2011 09:09:18] <froztbyte> jmp242: what's your command, and what does your sudo config look like?
[13-Oct-2011 09:09:58] <jmp242> My command is pretty simple:
[13-Oct-2011 09:09:59] <jmp242> sudo /sbin/service ntop restart
[13-Oct-2011 09:10:08] <jmp242> this is on SL5.7
[13-Oct-2011 09:10:21] <froztbyte> jmp242: mm
[13-Oct-2011 09:10:31] <jmp242> My sudo config is set by the Linux Sysadmin, I'll have to see if I can figure it out or wait for him to come in
[13-Oct-2011 09:10:37] <froztbyte> jmp242: do `grep requiretty /etc/sudoers`
[13-Oct-2011 09:11:08] <froztbyte> otherwise if you have reasonable sudo access on there, write a script to do that, then use sudo to set it suid root
[13-Oct-2011 09:11:09] <jmp242> I get
[13-Oct-2011 09:11:12] <jmp242> Defaults requiretty
[13-Oct-2011 09:11:12] <jmp242> # changed in order to be able to use sudo without a tty. See requiretty above.
[13-Oct-2011 09:11:20] <froztbyte> jmp242: yup, there's your problem
[13-Oct-2011 09:11:27] <jmp242> Ahh, simple config issue
[13-Oct-2011 09:11:28] <jmp242> thanks
[13-Oct-2011 09:11:37] <froztbyte> sure
[13-Oct-2011 09:11:39] <jmp242> I wasn't sure if it was a OS issue (which it is)
[13-Oct-2011 09:11:52] <jmp242> so I wanted to rule out Zenoss things before going to the *nix sysadmin
[13-Oct-2011 09:12:07] <froztbyte> hehe
[13-Oct-2011 09:12:18] * froztbyte tries this upgrade-to-v3 thing again
[13-Oct-2011 09:12:35] <froztbyte> this time I actually have a clue wtf I'd be doing though
[13-Oct-2011 09:12:44] <froztbyte> the last time I did this I just hit it with a large hammer
[13-Oct-2011 09:13:12] <jmp242> lol
[13-Oct-2011 09:13:32] <jmp242> I'm worried about the 3.x -> 4.x myself
[13-Oct-2011 09:15:41] <froztbyte> jmp242: I'm not...or at least, not yet
[13-Oct-2011 09:15:54] <froztbyte> can probably worry about that in 3 months
[13-Oct-2011 09:16:10] <froztbyte> but I'd rather get the upgrade to 3 sorted out now
[13-Oct-2011 09:17:04] <jmp242> yea
[13-Oct-2011 09:17:08] <jmp242> good idea
[13-Oct-2011 09:44:14] <dhopp> anybody know a way (or if it's possible) to have Zenoss pass the username that was used to login to a site window portlet?
[13-Oct-2011 10:04:28] <ali3n0> hi folks. I'm trying to define a custom datasource + datapoint, any help about how to debug? the "test" button for datasource returns me the expected value, but I can't graph it
[13-Oct-2011 10:21:32] kocolosk|away is now known as kocolosk
[13-Oct-2011 10:23:17] <Jane_Curry> ali3n0: You need to be a bit careful with the test button - sometimes it can hide instance problems
[13-Oct-2011 10:23:50] <ali3n0> Jane_Curry I see. Now I'm reading zencommand.log
[13-Oct-2011 10:23:56] <Jane_Curry> Does your command return output in "Nagios format"? It needs to to get values into data points
[13-Oct-2011 10:24:10] <ali3n0> it just returns a 0.00 like value
[13-Oct-2011 10:24:20] <ali3n0> with auto parser
[13-Oct-2011 10:29:10] <Jane_Curry> To get values into the data point the script needs to echo a line in the format:
[13-Oct-2011 10:31:37] <Jane_Curry> <message, typically including status> | <var1>=<value1> <var2>=<value> etc
[13-Oct-2011 10:43:02] <cmek> hi, does anyone here know of a good document describing the internals of zenperfsnmp and the performance collection process in general?
[13-Oct-2011 10:43:15] <ali3n0> Jane_Curry something like "loadtest | load=0.12" ? with the pipe too?
[13-Oct-2011 10:46:13] <cmek> I'm trying to find a good way of implementing VACM extensions and I'm a bit stuck with performance collection
[13-Oct-2011 10:55:46] <Jane_Curry> ali3n0 - yup the pipe separates the message from the var=value section
[13-Oct-2011 10:56:04] <ali3n0> Jane_Curry what parser should I use?
[13-Oct-2011 10:56:32] <Jane_Curry> The other trick you need to know is that the <var> name(s) echoed must match EXACTLY with the name of the datapoint(s) in your template
[13-Oct-2011 10:57:03] <Jane_Curry> Never worried about the parser - just use the auto
[13-Oct-2011 10:57:27] <ali3n0> cool
[13-Oct-2011 10:59:03] <ali3n0> Jane_Curry, you saved my day
[13-Oct-2011 10:59:12] <ali3n0> just out of curiosity: how did you find out?
[13-Oct-2011 11:03:26] <Jane_Curry> it's there if you dig hard enough but not very obvious - have a look at section 6.3 of the Admin guide about Monitoring using ZenCommand
[13-Oct-2011 11:05:19] <Jane_Curry> You might also pull my process monitoring doc which talks about this and gives examples - docs/DOC-3537
[13-Oct-2011 11:05:40] <Jane_Curry> Is it dev chat time???????
[13-Oct-2011 11:09:21] <ion-fan> does anyone know how to sum graphs together in multigraph view? where can i put in the Graph definition that calculates the other graphs together?
[13-Oct-2011 11:09:59] <ali3n0> Jane_Curry yep I'm going through Admin guide but this data point stuff you told me isn't there, trust me thanks btw for the document
[13-Oct-2011 11:11:36] <cmek> ion-fan: do you mean to have summaries at the bottom?
[13-Oct-2011 11:13:24] <ion-fan> cmek: no i have to cisco switch interfaces. and now i want to have a third graph in which the other two graphs are summed up together in one graph...
[13-Oct-2011 11:13:57] <cmek> ion-fan: /zport/dmd/Reports/Multi-Graph%20Reports
[13-Oct-2011 11:15:12] <cmek> ion-fan: have a look there
[13-Oct-2011 11:15:59] <ion-fan> cmek: yes i'm already in there... i already added the two datasources i want to sum up... they were displayed in my multigraph report... but i dont know how to define a graph that has the other two graphs as datasource and sums them up
[13-Oct-2011 11:17:01] <cmek> ion-fan: i'm not sure you can have graphs as datasources. but you can sum all your datasources and then graph them
[13-Oct-2011 11:17:20] <ion-fan> cmek: how?
[13-Oct-2011 11:17:51] <cmek> ion-fan: give me a sec, need to open one of mine
[13-Oct-2011 11:18:23] <cmek> ion-fan: have you ever played with rrdtools before?
[13-Oct-2011 11:19:08] <ion-fan> in "graph definitions" when i click "add custom" there is no field to enter the formula or some tales expressions...
[13-Oct-2011 11:20:02] <ion-fan> cmek: no i havent played ever with rrdtools... i hoped i can avoid cryptic graph definitions when using zenoss
[13-Oct-2011 11:31:42] <Jane_Curry> Hi cluther - do we have a dev chat or not today?? No sign of Nick...
[13-Oct-2011 11:31:59] <Jane_Curry> .. or JP
[13-Oct-2011 11:33:11] <cluther> Jane_Curry: Great question. I'm trying to figure that out myself.
[13-Oct-2011 11:33:41] <cluther> Yesterday and today is our yearly company get-together, so things are a little chaotic.
[13-Oct-2011 11:36:55] <cluther> Did you see the update I made to that permissions overriding forum thread?
[13-Oct-2011 11:39:07] <dhopp> one of the last things Nick said yesterday was that there would be a dev chat and that chet would be in the house (not quite word for word but he did say "in the house")
[13-Oct-2011 11:39:49] <Jane_Curry> cluther - yup saw that. Trying to test it as we speak. Did Nick forward my MenuExample stuff?
[13-Oct-2011 11:40:14] <Jane_Curry> I thought we were also expecting JPL today....
[13-Oct-2011 11:40:19] <cluther> I haven't seen it yet. I'll ask him about it as soon as I track him down.
[13-Oct-2011 11:41:36] <Jane_Curry> You can get the tarball at message/61820#61820
[13-Oct-2011 11:54:06] <cluther> Very cool, Jane. Are the undone items a matter of your time, or do you have unanswered questions?
[13-Oct-2011 11:58:38] <Jane_Curry> 3 main questions at this stage......
[13-Oct-2011 12:00:01] [disconnected at Thu Oct 13 12:00:01 2011]
[13-Oct-2011 12:00:02] [connected at Thu Oct 13 12:00:02 2011]
[13-Oct-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[13-Oct-2011 12:01:04] <Jane_Curry> The code to provide a footer menu in myFooterMenu.js came from the last dev chat, along with the associated configure.zcml
[13-Oct-2011 12:01:52] <Jane_Curry> .. but it breaks my Zenoss 3.1 completely - just get a white screen when refreshing any Zenoss panel after the change and bouncing zopectl
[13-Oct-2011 12:02:05] <Jane_Curry> No error messages when running zopectl fg though
[13-Oct-2011 12:02:09] <cluther> I can take a look at that.
[13-Oct-2011 12:02:55] <cluther> Did it work any any version for you? 3.2?
[13-Oct-2011 12:03:21] <Jane_Curry> Next one is in dialog_myDropDownMenu2 under browser/templates...
[13-Oct-2011 12:04:11] <Jane_Curry> I would like to put up a box prompting the user for input into the comments attriibute, take that input, and modify dev/comments accordingly
[13-Oct-2011 12:06:22] <Jane_Curry> Third question is - I have provided myDropDownMenu1 and 2 as objects in objects.xml - I suspect there is a better way of doing this in Zenoss 3??????
[13-Oct-2011 12:07:07] <Jane_Curry> Haven't tried the footer menu with 3.2 - hope it's not that as I suspect a lot of folk will stick with 3.1 until 4.2 is out and stable....
[13-Oct-2011 12:07:41] <cluther> We can probably come up with a way to make sure its cross-compatible.
[13-Oct-2011 12:08:13] <Jane_Curry> I guess you are saying I need to run up my 3.2 and try it there
[13-Oct-2011 12:08:26] <cluther> Apparently I have a meeting to attend right now. I'll get back to you on those on the topic you started.
[13-Oct-2011 12:09:38] <Jane_Curry> Sorry about that - input would be much appreciated
[13-Oct-2011 12:09:41] <Jane_Curry> Cheers!
[13-Oct-2011 12:10:17] <cluther> Sorry about the IRC session mix-up today. It probably should have been cancelled in advance.
[13-Oct-2011 12:10:44] <Jane_Curry> Hmmm - not your fault - I'll beat up Nick
[13-Oct-2011 15:50:27] <twm1010> anyone good at troubleshooting a mysql that won't start as part of a stack install?
[13-Oct-2011 15:54:03] <Hackman238> twm1010: Whats the message?
[13-Oct-2011 15:55:19] <twm1010> Number of processes running now: 0
[13-Oct-2011 15:55:19] <twm1010> 111013 14:46:55 mysqld restarted
[13-Oct-2011 15:55:21] <twm1010> over and over
[13-Oct-2011 15:55:35] <Hackman238> twm1010: did you modify the my.cnf?
[13-Oct-2011 15:55:47] <twm1010> I don't believe so
[13-Oct-2011 15:56:15] <twm1010> date modified on it is the same as the rest of the files from the install
[13-Oct-2011 15:56:19] <Hackman238> twm1010: if you do a free -m, are you out of ram?
[13-Oct-2011 15:56:53] <Hackman238> twm1010: If mysql cant allocate itself in free ram, it may throw a memalloc fail and bail
[13-Oct-2011 15:57:18] <twm1010> looks alright
[13-Oct-2011 15:57:24] <twm1010> total 1972, used 1239, free 733
[13-Oct-2011 15:57:36] <Hackman238> twm1010: Can you pastebin your my.cnf?
[13-Oct-2011 15:57:47] <twm1010> aye
[13-Oct-2011 15:58:21] <twm1010> http://pastebin.com/FSxQtAcb
[13-Oct-2011 15:58:31] <twm1010> should be defaulted, stack install of 3.1.0 on SLES10
[13-Oct-2011 15:59:02] <Hackman238> twm1010: can you pastebing your mysql.log too?
[13-Oct-2011 15:59:08] <Hackman238> twm1010: yeah looks default
[13-Oct-2011 16:00:15] <twm1010> sure, one sec
[13-Oct-2011 16:00:44] <twm1010> gotta find it first
[13-Oct-2011 16:02:33] <twm1010> hehe, giggle
[13-Oct-2011 16:02:42] <twm1010> it's 48G in size
[13-Oct-2011 16:03:09] <Hackman238> twm1010: oh damn. Well the last 1000 lines is okay LOL
[13-Oct-2011 16:03:51] <twm1010> http://pastebin.com/8dkP1NxU
[13-Oct-2011 16:03:58] <twm1010> that's not encouraging, haha
[13-Oct-2011 16:04:43] <Hackman238> twm1010: Oh yeah thats not good
[13-Oct-2011 16:04:48] <Hackman238> lets try something
[13-Oct-2011 16:05:07] <twm1010> ok
[13-Oct-2011 16:05:11] <Hackman238> twm1010: in /usr/local/zenoss/mysql/data
[13-Oct-2011 16:05:30] <Hackman238> twm1010: delete ib_logfile1 and ib_logfile0 or rename them.
[13-Oct-2011 16:05:43] <Hackman238> twm1010: then start mysql
[13-Oct-2011 16:06:09] <Hackman238> twm1010: if you have space best to rename
[13-Oct-2011 16:06:16] <twm1010> plenty of space
[13-Oct-2011 16:06:33] <twm1010> renamed the log file too, here goes
[13-Oct-2011 16:06:36] <Hackman238> twm1010: sometimes uncommited data is corrupted and therefore the logs need to be purged.
[13-Oct-2011 16:07:12] <twm1010> hrmm... seems to be doing its cycling "mysqld restarted" messages
[13-Oct-2011 16:08:00] <Hackman238> twm1010: tail the log
[13-Oct-2011 16:08:12] <Hackman238> is it doing its rehash?
[13-Oct-2011 16:09:10] <twm1010> yeah, a whole lot of "your database may be corrupt"
[13-Oct-2011 16:09:41] <Hackman238> twm1010: okay, if you mysql -u root -p, can you get in?
[13-Oct-2011 16:10:30] <twm1010> what's the default stack password?
[13-Oct-2011 16:10:46] <Hackman238> twm1010: Not sure, try blank, zenoss, password
[13-Oct-2011 16:10:59] <twm1010> got it, apparently i did set one, i'm at a mysql prompt
[13-Oct-2011 16:11:00] <Hackman238> twm1010: worst case can try -u zenoss -p zenoss
[13-Oct-2011 16:11:10] <Hackman238> twm1010: Alrighty and it hasnt kicked you out?
[13-Oct-2011 16:11:20] <twm1010> nope, still there
[13-Oct-2011 16:11:36] <Hackman238> twm1010: use events;
[13-Oct-2011 16:11:47] <Hackman238> twm1010: then check table {each table};
[13-Oct-2011 16:12:24] <Hackman238> twm1010: check table heartbeats; check table status; etc.
[13-Oct-2011 16:12:58] <twm1010> getting connection error, mysql has gone away, unable to connect, etc
[13-Oct-2011 16:13:07] <Hackman238> twm1010: Ah poop
[13-Oct-2011 16:13:40] <Hackman238> twm1010: Okay in the my.cnf in /usr/local/zenoss/mysql/etc add innodb_force_recovery=1 under [mysqld]
[13-Oct-2011 16:14:00] <Hackman238> then stop mysql, make sure its dead and if not kill it, then start
[13-Oct-2011 16:14:28] <Hackman238> twm1010: Do you have a backup and or is the event data important?
[13-Oct-2011 16:14:43] <twm1010> no backup, event data unimportant
[13-Oct-2011 16:15:34] <Hackman238> twm1010: might be easier to recreate this puppt
[13-Oct-2011 16:15:37] <Hackman238> *puppy
[13-Oct-2011 16:15:48] <Hackman238> twm1010: Especially if the data isnt important
[13-Oct-2011 16:16:02] <twm1010> i figured that might be the case, the customer was nice enough to provide me a desktop class device with no UPS and likes to unplug it
[13-Oct-2011 16:16:13] <twm1010> they were askin for it, no worries
[13-Oct-2011 16:16:20] <Hackman238> twm1010: LOL!
[13-Oct-2011 16:16:57] <twm1010> guess i'll brush on my zenbackup from the CLI
[13-Oct-2011 16:16:59] <twm1010> sigh..
[13-Oct-2011 16:17:04] <Hackman238> twm1010: I'd drop the events db, then look in the $ZENHOME/bin/zenoss_init_pre script for the events db creation stuff
[13-Oct-2011 16:17:23] <twm1010> just whack the data files?
[13-Oct-2011 16:17:39] <Hackman238> twm1010: yeah mysql in zenoss 2/3 is just event data
[13-Oct-2011 16:18:04] <Hackman238> twm1010: long as the tables exist and zenoss can hit it everything will work
[13-Oct-2011 16:18:59] <Hackman238> twm1010: Gotta run. PM me or hit me up tomorrow if you need help
[13-Oct-2011 16:19:10] <twm1010> ok
[13-Oct-2011 16:19:50] <Hackman238> twm1010: message/60348#60348
[13-Oct-2011 17:26:11] kocolosk is now known as kocolosk|away
[13-Oct-2011 17:35:01] <pplask> hello!
[13-Oct-2011 17:35:19] <pplask> how do i generate an event when a switch port goes down?
[13-Oct-2011 17:46:11] <straterra> pplask: You could use an snmp trap..or the standard way (monitor the NIC in Zenoss)
[13-Oct-2011 17:46:41] <pplask> straterra, i monitor it automatically
[13-Oct-2011 17:47:00] <pplask> straterra, if i unplug a cable from switch, which event would it generate?
[13-Oct-2011 17:47:22] <straterra> No idea, likely a downed NIC event
[13-Oct-2011 18:10:49] <nexex> will zenoss send replys for snmp informs?
[13-Oct-2011 20:32:17] <illsci> Can you guys please move the uid's and gid's used by zenoss-pack to something less than 1000
[13-Oct-2011 20:32:36] <illsci> Puppet will try and remove those users by default
[13-Oct-2011 20:32:52] <illsci> most system users are < 500
[13-Oct-2011 20:48:03] <Sam-I-Am> file a bug report?
[13-Oct-2011 21:23:56] <illsci> Ok
[14-Oct-2011 00:00:01] [disconnected at Fri Oct 14 00:00:01 2011]
[14-Oct-2011 00:00:02] [connected at Fri Oct 14 00:00:02 2011]
[14-Oct-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[14-Oct-2011 02:35:22] AB is now known as Guest93325
[14-Oct-2011 02:43:31] <MoobvdaZ> morning guys
[14-Oct-2011 03:32:20] <MoobvdaZ> When I get a high CPU alert message, is there a way to see which proccess causes the high CPU alert ?
[14-Oct-2011 04:11:07] <ion-fan> the command top ?
[14-Oct-2011 04:12:52] <MoobvdaZ> on linux that would work but in this case its a windows machine. Zenoss reports the high CPU usage, so I wondered if it was possible that zenoss also could tell me which proccess caused the high cpu alert
[14-Oct-2011 04:15:15] <ali3n0> hi folks. I've just installed Amazon zenpack, added aws credentials in the Ec2Manager, but I can't see any instance (they should be in EU region). Any help about how to debug this?
[14-Oct-2011 04:33:18] <ali3n0> I think I spotted a minor bug, where should I report it?
[14-Oct-2011 05:27:21] <hmp> i would like to delete zLinks attribute on all classes which have it "localy defined"
[14-Oct-2011 05:27:28] <hmp> through zendmd ofc
[14-Oct-2011 05:27:38] <hmp> locally?*
[14-Oct-2011 05:30:34] <whyzgeek> hi does any body know how to access rrd values from zendmd?
[14-Oct-2011 05:54:11] <ali3n0> hi folks, is there a way to change modeling cycle time for Ec2Manager ?
[14-Oct-2011 06:00:21] <froztbyte> ali3n0: are there any options under the classpath for it?
[14-Oct-2011 06:00:43] <ali3n0> froztbyte can't find anything obvious like cycle* or something
[14-Oct-2011 06:02:26] <froztbyte> whyzgeek: I haven't really tried, but you can dig through the codepath to see how it's accessed
[14-Oct-2011 06:02:44] <froztbyte> whyzgeek: open something that renders RRDs, look at the URL, and then start tracing based on that
[14-Oct-2011 06:04:49] <cmek> whyzgeek: yes, via performance server (get.PerformanceServer())
[14-Oct-2011 06:05:03] <cmek> erm. getPerformanceServer()
[14-Oct-2011 06:06:14] <cmek> end then performanceCustomSummary on that
[14-Oct-2011 06:09:35] <whyzgeek> hmm
[14-Oct-2011 06:09:47] <whyzgeek> I am actually trying getRRDValue on device
[14-Oct-2011 06:10:03] <cmek> that's probably even easier
[14-Oct-2011 06:11:13] <cmek> i can't remember why I had to do it with customsummary
[14-Oct-2011 06:14:22] <ali3n0> it would be nice if I can setup a cronjob forcing remodeling just one device instead of the global modeling cycle process
[14-Oct-2011 06:14:27] <cmek> btw, does anyone know how to make multi-graph reports work on data spread between multiple collectors?
[14-Oct-2011 06:14:43] <cmek> ali3n0: you can run modeling from cli
[14-Oct-2011 06:14:44] <whyzgeek> I am trying to see what it needs as an argument
[14-Oct-2011 06:14:58] <ali3n0> cmek ah cool
[14-Oct-2011 06:15:18] <cmek> ali3n0: zenmodeler run -d YOURDEVICEHERE
[14-Oct-2011 06:15:28] <ali3n0> cmek great, thanks
[14-Oct-2011 06:15:35] <whyzgeek>
[14-Oct-2011 06:15:37] <whyzgeek> I found it
[14-Oct-2011 06:15:52] <whyzgeek> dev.getRRDValue('sysUpTime_sysUpTime')
[14-Oct-2011 06:15:57] <cmek>
[14-Oct-2011 06:16:33] <whyzgeek> now I have to figure out a way to get max and min in a period!
[14-Oct-2011 06:16:52] <cmek> whyzgeek: you can pass start, end and function to that method
[14-Oct-2011 06:17:04] <cmek> if it's the method I'm looking at
[14-Oct-2011 06:17:33] <whyzgeek> ya probably it is the same arguments that it available in graphs
[14-Oct-2011 06:17:37] <cmek> whyzgeek: see what happens if you pass function="MAX"
[14-Oct-2011 06:17:44] <whyzgeek> like drange
[14-Oct-2011 06:32:32] <whyzgeek> dev.getRRDValue('sysUpTime_sysUpTime', function="LAST")
[14-Oct-2011 06:32:36] <whyzgeek> worked but not MAX
[14-Oct-2011 06:34:46] <whyzgeek>
[14-Oct-2011 06:34:58] <whyzgeek> dev.getRRDValue(dsname='sysUpTime_sysUpTime', function="MAXIMUM")
[14-Oct-2011 06:35:01] <whyzgeek> this works
[14-Oct-2011 06:35:26] <whyzgeek> and MINIMUM
[14-Oct-2011 06:35:32] <whyzgeek> plus start and end
[14-Oct-2011 06:36:29] <cmek> is LAST, MAXIMUN/MINIMUM what you use with VDEFs?
[14-Oct-2011 06:36:48] <cmek> it looks like it just applies that function to the end of the VDEF statement
[14-Oct-2011 06:38:39] <cmek> hm, yeap, VARIABLES section in rrdgraph_rpn
[14-Oct-2011 06:39:03] <whyzgeek> yep
[14-Oct-2011 06:39:07] <whyzgeek> it seems like i
[14-Oct-2011 06:39:10] <whyzgeek> it
[14-Oct-2011 08:47:18] <ali3n0> hi guys. this is weird, seems that EC2 instances managed via aws zenpack keep being monitored even after manual termination, ie they disappear from device list but events keep coming
[14-Oct-2011 08:55:36] <Hackman238> ali3n0: You'll need to delete the devices created by the pack in zenoss
[14-Oct-2011 08:56:10] <ali3n0> Hackman238 I can't see them anymore in the GUI, how can I remove them?
[14-Oct-2011 08:56:57] <Hackman238> ali3n0: They aren't in /Devices/Server/Linux or /Devices/Server/Windows?
[14-Oct-2011 08:57:17] <ali3n0> nope, I see only running instances there
[14-Oct-2011 08:57:57] <ali3n0> if I click on a device link from an event, it gives me a "An error was encountered while publishing this resource"
[14-Oct-2011 08:58:01] <Hackman238> ali3n0: Thats strange. Sounds like zenhub might not have notified collector daemons to stop checking those devices
[14-Oct-2011 08:58:09] <ali3n0> The requested resource does not exist
[14-Oct-2011 08:58:17] <ali3n0> exactly
[14-Oct-2011 08:58:21] <Hackman238> ali3n0: Have you tried restarting zenoss?
[14-Oct-2011 08:58:24] <ali3n0> yes
[14-Oct-2011 08:58:32] <Hackman238> ali3n0: 1 second
[14-Oct-2011 08:58:34] <ali3n0> sure
[14-Oct-2011 08:59:47] <Hackman238> ali3n0: Run this script as zenoss user on the master http://pastebin.com/3natsJpS
[14-Oct-2011 08:59:56] <Hackman238> ali3n0: It'll take a long while depending on the device count
[14-Oct-2011 09:00:11] <ali3n0> Hackman238 ok I'll get back to you
[14-Oct-2011 09:00:25] <Hackman238> ali3n0: But it's pretty good at fiding orphaned devices, orphaned componenets and bad catalog entries
[14-Oct-2011 09:00:29] <Hackman238> ali3n0: Alrighty
[14-Oct-2011 09:02:20] <jmp242> Hackman238: is that something you wrote?
[14-Oct-2011 09:02:49] <Hackman238> jmp242: Negative. Work of the infamous rocket.
[14-Oct-2011 09:03:01] <ali3n0> Hackman238 it finished right now
[14-Oct-2011 09:03:15] <jmp242> mmm
[14-Oct-2011 09:03:44] <Hackman238> ali3n0: Alrighty. Once done do a zenchkrels -x1 -r to clean up any hlaf rels left from orphan removal
[14-Oct-2011 09:03:53] <jmp242> I wonder if we can put it on the wiki somewhere... It's an issue that comes up occasionally, and should be more permanently available
[14-Oct-2011 09:04:06] <jmp242> rocket: what's the license on that code?
[14-Oct-2011 09:04:27] <ali3n0> Hackman238 this is going straight away into a fabric file!
[14-Oct-2011 09:04:42] <Hackman238> jmp242: I think rockets away today
[14-Oct-2011 09:04:48] <Hackman238> jmp242: travel if I recall right
[14-Oct-2011 09:04:56] <Hackman238> (which I may or may not LOL)
[14-Oct-2011 09:05:18] <Hackman238> jmp242: Unknown, but he just gave it to me so I suspect GPL.
[14-Oct-2011 09:05:32] <jmp242> I just wanted to make sure if I repost it I'm not treading on any toes
[14-Oct-2011 09:05:58] <Hackman238> jmp242: I think your safe.
[14-Oct-2011 09:06:06] <Hackman238> jmp242: How's by you?
[14-Oct-2011 09:06:21] <jmp242> Well, not too ba
[14-Oct-2011 09:06:21] <jmp242> d
[14-Oct-2011 09:06:29] <jmp242> Getting in my morning Diet Cokes
[14-Oct-2011 09:06:35] <jmp242>
[14-Oct-2011 09:06:36] <Hackman238> jmp242: hahaha
[14-Oct-2011 09:06:49] <ali3n0> Hackman238 is that script going to be linked/documented somewere on zenoss.com?
[14-Oct-2011 09:06:56] <Hackman238> jmp242: I'm on my 4th or 5th coffee
[14-Oct-2011 09:07:02] <jmp242> ali3n0: I'll be posting it to the ZCAportal wiki
[14-Oct-2011 09:07:09] <Hackman238> ali3n0: jmp will get it somewhere
[14-Oct-2011 09:07:19] <ali3n0> nice!
[14-Oct-2011 09:07:23] <ali3n0> thanks a lot guys
[14-Oct-2011 09:07:54] <Hackman238> ali3n0: NP
[14-Oct-2011 09:08:01] <ali3n0> I just discovered the existence of that wiki
[14-Oct-2011 09:08:07] <Hackman238> ali3n0: Did it clean up the problem?
[14-Oct-2011 09:08:49] <Hackman238> jmp242: Tha reminds me, we should vote on a logo so I can print tshirts and other stuff as swag for ZCA promotion
[14-Oct-2011 09:08:56] <ali3n0> Hackman238 to be honest I think it already did the restart... I didn't notice last seen wasn't growing
[14-Oct-2011 09:09:06] <jmp242> Yes
[14-Oct-2011 09:09:10] <Hackman238> jmp242: $2,400 vouched by a client for the ZCA
[14-Oct-2011 09:09:25] <jmp242> great - we just need the accounts so we can move things along
[14-Oct-2011 09:09:29] <Hackman238> ali3n0: Alrighty, keep an eye on it.
[14-Oct-2011 09:09:41] <Hackman238> jmp242: Yes, its mostly set. I'm sorry I've been behind.
[14-Oct-2011 09:09:47] <ali3n0> Hackman238 sure thing
[14-Oct-2011 09:10:43] <Hackman238> jmp242: Been fighting NY state, on vacation, handeling a malfunctioning scientific grade industrial heat exchanger, reworking the image and business model of my two stores, Zenoss 4 migrations and my regular contracts.
[14-Oct-2011 09:10:50] <Hackman238> jmp242: Been madness!
[14-Oct-2011 09:10:51] <jmp242> Hackman238: on that script
[14-Oct-2011 09:11:06] <jmp242> do they need to restart Zenoss before the zencheckrels, after or at all?
[14-Oct-2011 09:11:23] <Hackman238> jmp242: its not important to restart.
[14-Oct-2011 09:11:26] <jmp242> ok
[14-Oct-2011 09:11:35] <Hackman238> jmp242: Just helps clear out bad pcache in zenhubs
[14-Oct-2011 09:11:57] <jmp242> Ok, well I'll recommend it at the end then
[14-Oct-2011 09:12:07] <Hackman238> jmp242: OH! I wanted to mention this- we've a DC on Z4.1 now. It's amazing...stuff is acctually incredibly painless
[14-Oct-2011 09:12:13] <Hackman238> jmp242: Alrighty
[14-Oct-2011 09:13:54] <jmp242> Now posted here: https://zcaportal.org/wiki/bin/view/ZCA/ZCABestPractices
[14-Oct-2011 09:14:01] <jmp242> under troubleshooting
[14-Oct-2011 09:14:29] <Hackman238> jmp242: Very cool. Thanks very much!
[14-Oct-2011 09:21:38] kocolosk|away is now known as kocolosk
[14-Oct-2011 09:35:26] <Sam-I-Am> moo
[14-Oct-2011 09:35:51] <Hackman238> Sam-I-Am: Hey, how goes it?
[14-Oct-2011 09:36:00] <Hackman238> Sam-I-Am: You using Zenoss at your new place yet?
[14-Oct-2011 09:39:39] <Sam-I-Am> naw
[14-Oct-2011 09:39:45] <Sam-I-Am> it aint gonna happen :/
[14-Oct-2011 09:40:02] <Sam-I-Am> which is getting me way out of practice
[14-Oct-2011 09:40:17] <Hackman238> Sam-I-Am: Man. v4.1 is serisouly boss
[14-Oct-2011 09:41:18] <Sam-I-Am> woot
[14-Oct-2011 09:41:33] <Sam-I-Am> guess theres no free ver yet tho
[14-Oct-2011 09:41:39] <Sam-I-Am> or something i could try somewhere
[14-Oct-2011 09:47:40] <Hackman238> Sam-I-Am: Not yet, coming soo though
[14-Oct-2011 09:51:14] <Sam-I-Am> yep
[14-Oct-2011 09:55:06] <ali3n0> bye people
[14-Oct-2011 10:20:05] <ion-fan> hackman238: did you get further?
[14-Oct-2011 10:21:45] <Hackman238> ion-fan: I did, I'm acctually working on it now
[14-Oct-2011 10:22:20] <Hackman238> ion-fan: I'm going to intergrate it in to the pack as a default option along with a cataloging change for 4.x compatability
[14-Oct-2011 10:22:36] <Hackman238> ion-fan: I'll get back to you soon. Sorry for the wait!
[14-Oct-2011 10:23:58] <ion-fan> np i'm happy to hear those good news!!
[14-Oct-2011 10:24:28] <ion-fan> when will 4.x be out?
[14-Oct-2011 10:27:27] <Hackman238> ion-fan: Near the end of the year
[14-Oct-2011 10:27:35] <ion-fan> ah ok..
[14-Oct-2011 10:27:40] <Hackman238> ion-fan: That pack will work with both 3.x and 4.x
[14-Oct-2011 10:27:52] <ion-fan> thats good
[14-Oct-2011 10:28:03] <Hackman238> ion-fan: Turns out the way I did something in the pack breaks 4.x compatability and could complicate upgrades
[14-Oct-2011 10:36:03] <Hackman238> ion-fan: i *should* have this one fixed and tested by the end of the day
[14-Oct-2011 10:37:01] <ion-fan> anyway on thursday next week im in vacation...
[14-Oct-2011 10:37:56] <Hackman238> ion-fan: Very cool. Where to?
[14-Oct-2011 10:38:54] <ion-fan> turkey...
[14-Oct-2011 10:39:51] <ion-fan> 2 weeks... i cant await it
[14-Oct-2011 10:44:24] <Hackman238> ion-fan: Very cool. I hope you have a good time!
[14-Oct-2011 11:53:45] <willwh> hi guys.... got a bit of an odd one
[14-Oct-2011 11:53:57] <willwh> I tried renaming an IP service yesterday - and did a full restart
[14-Oct-2011 11:54:15] <willwh> it's still using the old detail
[14-Oct-2011 11:54:24] <willwh> anyone been successful in renaming IP services?
[14-Oct-2011 11:54:33] <willwh> (specifically "tincan" - which is :1935)
[14-Oct-2011 12:00:01] [disconnected at Fri Oct 14 12:00:01 2011]
[14-Oct-2011 12:00:01] [connected at Fri Oct 14 12:00:01 2011]
[14-Oct-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[14-Oct-2011 12:09:49] <whyzgeek> hi guys does any body know any way to aggregate a bunch of datasources and set threshold on it and graph it?
[14-Oct-2011 12:11:51] <jmp242> mulitgraph reports maybe?
[14-Oct-2011 12:12:04] <jmp242> There was also some things with graph definitions
[14-Oct-2011 12:12:11] <jmp242> but I think those are only for datapoints
[14-Oct-2011 12:13:04] <whyzgeek> in multigraph reports you can stack them
[14-Oct-2011 12:13:07] <whyzgeek> but thresholds
[14-Oct-2011 12:13:10] <whyzgeek> are just lines
[14-Oct-2011 12:13:16] <whyzgeek> they don't raise events
[14-Oct-2011 12:13:50] <whyzgeek> I saw something like alias in some datapoint, I have no idea what it does
[14-Oct-2011 12:14:01] <whyzgeek> for example for cpu load
[14-Oct-2011 12:14:25] <whyzgeek> it has a formula to divide it per core
[14-Oct-2011 12:14:34] <whyzgeek> but has no effect
[14-Oct-2011 12:14:58] <jmp242> Hmm
[14-Oct-2011 12:15:10] <jmp242> may be that thresholds don't work on multigraph reports
[14-Oct-2011 12:15:24] <jmp242> Hackman238: any idea on whyzgeek question?
[14-Oct-2011 12:15:31] <jmp242> afk for a bit
[14-Oct-2011 12:30:10] <pplask> hello! im trying to create a threshold for errors on network interfaces... what should i put in MAX value? just the number of errors or any kind of formula?
[14-Oct-2011 13:14:56] <pplask> guys?
[14-Oct-2011 13:20:05] <jmp242> pplask: I think you need to figure out what the SNMP datasource is providing
[14-Oct-2011 13:20:22] <jmp242> For errors
[14-Oct-2011 13:20:25] <pplask> mmm
[14-Oct-2011 13:20:30] <jmp242> I think it's just the number of errors
[14-Oct-2011 13:20:37] <pplask> yeah it is
[14-Oct-2011 13:20:41] <pplask> i just saw it
[14-Oct-2011 13:20:55] <jmp242> but then I expect you'd want to use the appropriate graph type
[14-Oct-2011 13:21:01] <jmp242> err, datasource type I mean
[14-Oct-2011 13:21:10] <jmp242> Derive comes to mind, but check the RRD stuff
[14-Oct-2011 13:21:28] <jmp242> then I'd expect you'd set errors to whatever you wanted them to be under in a 5 minute period
[14-Oct-2011 13:23:07] <tvincent> is jake around? I need to get information about Darwin from him.
[14-Oct-2011 13:23:08] <pplask> well, i dont expect interface errors hehe
[14-Oct-2011 13:23:33] <jmp242> Yea, so set it to 1 maybe
[14-Oct-2011 13:38:05] <pplask> jmp242, its funny cos zenoss seems to be graphing interface erros correctly. but now that i created a threshold, i can see events generated by interface errors, but this errors are not shown on graphs
[14-Oct-2011 13:39:09] <jmp242> Hmmm - I don't know, I haven't tried this - partially because I almost never see interface errors on my LAN
[14-Oct-2011 13:39:34] <pplask> jmp242, maybe if you setup a threshold u will see a lot!
[14-Oct-2011 13:40:52] <jmp242> That could be. I'd need to talk to my network guys and they are out for the day right now
[14-Oct-2011 13:40:57] <jmp242> maybe on Monday if I remember
[14-Oct-2011 13:41:31] <pplask> mmm weird, zenoss events says that there is an 120.72 value on error, but snmpwalk shows me zero.
[14-Oct-2011 13:42:32] <jmp242> I wonder if it was transient
[14-Oct-2011 13:42:47] <pplask> im using watch -n1 snmpwalk...
[14-Oct-2011 13:42:55] <pplask> and watching zenoss events
[14-Oct-2011 13:43:07] <pplask> just cleared it now
[14-Oct-2011 13:43:15] <pplask> the event
[14-Oct-2011 13:43:17] <pplask> lets see
[14-Oct-2011 13:43:43] <pplask> zenoss got a new value: 340.70
[14-Oct-2011 13:43:46] <pplask> of erros
[14-Oct-2011 13:43:52] <pplask> and nothing on snmpwalk
[14-Oct-2011 13:44:37] <jmp242> That's strange
[14-Oct-2011 13:45:44] <jmp242> Can you check the logs - and make sure they're at debug?
[14-Oct-2011 13:46:40] <pplask> jmp242, mmm how do i check that? im new to zenoss
[14-Oct-2011 13:47:00] <jmp242> oh ok
[14-Oct-2011 13:47:04] <jmp242> go to Advanced
[14-Oct-2011 13:47:07] <jmp242> Daemons
[14-Oct-2011 13:47:12] <jmp242> and check the config
[14-Oct-2011 13:47:20] <jmp242> they probably aren't in verbose mode
[14-Oct-2011 13:47:32] <jmp242> so you will have to set the zenperfsnmp
[14-Oct-2011 13:47:45] <jmp242> to debug or logseverity 10
[14-Oct-2011 13:47:55] <jmp242> depending on whether there's a gui or a text conf file
[14-Oct-2011 13:48:04] <jmp242> save, restart the daemon
[14-Oct-2011 13:48:08] <jmp242> wait for the wrong value to show up
[14-Oct-2011 13:48:11] <jmp242> click on view log
[14-Oct-2011 13:48:13] <jmp242> and see what it says
[14-Oct-2011 13:48:24] <pplask> ok
[14-Oct-2011 13:48:25] <jmp242> you may need to also look at other daemons, but I have to go right now,
[14-Oct-2011 13:48:26] <pplask> w8 a sec
[14-Oct-2011 13:48:37] <pplask> ok i will try this
[14-Oct-2011 13:48:40] <pplask> thanks for you help!
[14-Oct-2011 13:50:46] <jmp242> yum
[14-Oct-2011 13:50:48] <jmp242> yup
[14-Oct-2011 13:50:58] <jmp242> I hope to bbs - got to go see a user ~shudder
[14-Oct-2011 14:11:10] <jmp242> ok, back for now
[14-Oct-2011 14:23:11] cgibbons_ is now known as cgibbons
[14-Oct-2011 14:53:45] <kenhen93> Hi, can anyone help with a custom parser?
[14-Oct-2011 15:36:44] <rmatte> quiet in here today
[14-Oct-2011 15:43:58] <jmp242> yea\
[14-Oct-2011 15:44:01] <jmp242> Friday I guess
[14-Oct-2011 15:48:29] <rmatte> yeh
[14-Oct-2011 16:35:50] <Hackman238> from AccessControl.PermissionRole import _what_not_even_god_should_do
[14-Oct-2011 16:36:01] <Hackman238> mod = __import__(mname, *_import_chickens)
[14-Oct-2011 16:36:09] <Hackman238> ^Real things I found in Zenoss today
[14-Oct-2011 16:40:57] <rmatte> haha
[14-Oct-2011 16:41:06] <humanix> hi Hackman238 rmatte
[14-Oct-2011 16:41:07] <rmatte> I've seen stuff like that in the code before
[14-Oct-2011 16:41:07] <humanix>
[14-Oct-2011 16:41:12] <rmatte> can't remember any specific lines
[14-Oct-2011 16:41:14] <rmatte> humanix: hi
[14-Oct-2011 16:41:36] <Hackman238> Hey all
[14-Oct-2011 16:41:44] <Hackman238> Sorry been short on resources the last couple days
[14-Oct-2011 16:41:58] <rmatte> I've been unrealistically busy so I wasn't around either
[14-Oct-2011 16:42:04] <rmatte> finally dying down
[14-Oct-2011 16:42:25] <rmatte> I submitted my request for vacation today since I've been putting it off for months because of deadlines
[14-Oct-2011 16:42:48] <rmatte> trying to take November 7th to 18th.
[14-Oct-2011 16:43:15] <Hackman238> rmatte: Nice
[14-Oct-2011 16:43:20] <Hackman238> rmatte: question for ya
[14-Oct-2011 16:43:29] <rmatte> yarr?
[14-Oct-2011 16:43:30] <Hackman238> rmatte: Have you been able to <configure xmlns="http://namespaces.zope.org/browser"> in zenoss v3.2?
[14-Oct-2011 16:43:45] <Hackman238> rmatte: I'm getting no global browser
[14-Oct-2011 16:43:50] <rmatte> no idea, never tried
[14-Oct-2011 16:44:14] <Hackman238> rmatte: Damn. Alrighty
[14-Oct-2011 16:44:24] <Hackman238> rmatte: Where you vacationing to?
[14-Oct-2011 16:44:33] <rmatte> it's hunting season, going to be at my hunt camp
[14-Oct-2011 16:44:50] <rmatte> won't actually be hunting since I don't have a gun license and it's gun season
[14-Oct-2011 16:44:58] <rmatte> but I invited a couple of my friends up who gun hunt
[14-Oct-2011 16:45:07] <rmatte> I'll just be drinking beer and ATVing around
[14-Oct-2011 16:45:07] <rmatte> lol
[14-Oct-2011 16:45:13] <Hackman238> LOL nice
[14-Oct-2011 16:45:52] <rmatte> we have a cabin there with wood stoves, oil lamps, etc...
[14-Oct-2011 16:45:56] <rmatte> hookups for a generator
[14-Oct-2011 16:51:54] <Hackman238> rmatte: Sounds like a nice place
[14-Oct-2011 16:54:15] <rmatte> I have photos of it in my facebook gallery
[14-Oct-2011 16:56:03] <rmatte> too bad KDE sucks now
[14-Oct-2011 16:56:15] <rmatte> KDE 3 was excellent, KDE 4 is meh.
[14-Oct-2011 16:57:55] <Hackman238> rmatte: rmatte I'll ahve to take a look
[14-Oct-2011 16:58:23] <rmatte>
[14-Oct-2011 17:00:03] <cgibbons> oooh ahh
[14-Oct-2011 17:00:11] <rmatte> ?
[14-Oct-2011 17:00:34] <cgibbons> managed to get something to work
[14-Oct-2011 17:00:38] <rmatte> haha
[14-Oct-2011 17:00:47] <cgibbons> (writing a new zenoss plugin). it didn't all leave the brain afterall.
[14-Oct-2011 17:00:50] <rmatte> glad stuff is working for someone
[14-Oct-2011 17:00:56] <rmatte> nice
[14-Oct-2011 17:01:18] <rmatte> so you're still doing Zenoss related work in your day job?
[14-Oct-2011 17:01:21] <rmatte> or just on the side?
[14-Oct-2011 17:02:19] <cgibbons> unexpectedly. my team has a new product in limited availability right now (beta) and we're responsible for support until it goes production, and it needed some monitoring. they originally had planned on using shinken, and, well, it was a huge POS. so I got zenoss up and running in the typical amount of short time.
[14-Oct-2011 17:02:38] <rmatte> nice
[14-Oct-2011 17:02:39] <cgibbons> wanted to add a custom henpeck that queries the REST API we have in this system and model some of the data as components
[14-Oct-2011 17:04:30] <Hackman238> Later guys
[14-Oct-2011 17:04:35] <Hackman238> Have a good weekend
[14-Oct-2011 17:07:16] <rmatte> u 2
[14-Oct-2011 17:22:15] kocolosk is now known as kocolosk|away
[14-Oct-2011 17:26:01] <rmatte> finding SNMP OIDs for Nortel devices is the most infuriating thing ever
[14-Oct-2011 18:18:00] <rmatte> time to take off, later all
[14-Oct-2011 19:34:45] kocolosk|away is now known as kocolosk
[14-Oct-2011 19:35:20] kocolosk is now known as kocolosk|away
[15-Oct-2011 00:00:01] [disconnected at Sat Oct 15 00:00:01 2011]
[15-Oct-2011 00:00:01] [connected at Sat Oct 15 00:00:01 2011]
[15-Oct-2011 00:00:11] [zenoss-logger (logger bot) has joined #zenoss]
[15-Oct-2011 12:00:01] [disconnected at Sat Oct 15 12:00:01 2011]
[15-Oct-2011 12:00:01] [connected at Sat Oct 15 12:00:01 2011]
[15-Oct-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]
[16-Oct-2011 00:00:01] [disconnected at Sun Oct 16 00:00:01 2011]
[16-Oct-2011 00:00:01] [connected at Sun Oct 16 00:00:01 2011]
[16-Oct-2011 00:00:13] [zenoss-logger (logger bot) has joined #zenoss]
[16-Oct-2011 12:00:01] [disconnected at Sun Oct 16 12:00:01 2011]
[16-Oct-2011 12:00:02] [connected at Sun Oct 16 12:00:02 2011]
[16-Oct-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[17-Oct-2011 00:00:01] [disconnected at Mon Oct 17 00:00:01 2011]
[17-Oct-2011 00:00:01] [connected at Mon Oct 17 00:00:01 2011]
[17-Oct-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[17-Oct-2011 06:05:22] <QubeZ> hello all
[17-Oct-2011 06:05:43] <QubeZ> I am trying to add a new Google Maps API key to zenoss but when I go to configure the portlet for locations, i see no where to enter the key
[17-Oct-2011 06:20:30] <enkrypt> I'm running 3.2.0 enterprise with the 3 patches applied, I'm getting these error messages in the log files of zenprocess, zentrap, zenwin, zeneventlog etc: CRITICAL zen.zenprocess: Unrecoverable Error: Task configLoader already exists
[17-Oct-2011 06:22:39] <enkrypt> which results in all of those daemons not working
[17-Oct-2011 06:31:33] <enkrypt> Hackman238 / jmp242 / rmatte any ideas for the above issue?
[17-Oct-2011 08:25:46] <enkrypt> lots of my daemons or dying with the following error message: CRITICAL zen.zenwinperf: Unrecoverable Error: Task configLoader already exists
[17-Oct-2011 08:25:51] <enkrypt> any idea what might be the problem?
[17-Oct-2011 08:34:33] <Hackman238> enkrypt: Have you restarted zenhub?
[17-Oct-2011 08:34:47] <enkrypt> Hackman238: restarted about everything
[17-Oct-2011 08:34:53] <Hackman238> enkrypt: sounds like the deamon is asking zenhub to load a config service thats already loaded
[17-Oct-2011 08:35:14] <enkrypt> even after restarting zenoss entirely on both the hub and the collector, right away I've got some daemons that die with that error messages
[17-Oct-2011 08:35:34] <enkrypt> I manage to fix it eventually bu starting the failing daemons manually a few times, until they stayed alive
[17-Oct-2011 08:35:36] <Hackman238> enkrypt: Anything in zenhub when it happens?
[17-Oct-2011 08:35:49] <enkrypt> the ones having this problem are zentrap, zenwinperf, zenprocess, zeneventlog and some others
[17-Oct-2011 08:36:09] <enkrypt> Hackman238: not really. I do have one stack trace in zenhub however, but not necessarily at the same moment
[17-Oct-2011 08:36:36] <enkrypt> exceptions.AttributeError: 'NoneType' object has no attribute 'monitorDevice'
[17-Oct-2011 08:44:46] <Hackman238> enkrypt: Ah
[17-Oct-2011 08:45:05] <Hackman238> enkrypt: could you try a 'update remote' collectors?
[17-Oct-2011 08:45:21] <Hackman238> enkrypt: Sounds like monitor association is borked
[17-Oct-2011 08:47:31] <rocket> or do a "zenchkrels -r -x1"
[17-Oct-2011 08:48:26] <Hackman238> rocket: Morning! How goes it?
[17-Oct-2011 08:51:41] <enkrypt> Hackman238: I'll try update remote
[17-Oct-2011 08:51:45] <rocket> busy ... very very busy ... some guy "Shane" has a billion open cases in my queue this morning ..
[17-Oct-2011 08:52:36] <Hackman238> enkrypt: Alrighty
[17-Oct-2011 08:52:40] <Hackman238> rocket: Sorry LOL
[17-Oct-2011 08:54:23] <jmp242> morning all
[17-Oct-2011 08:54:39] <Hackman238> Morning jmp242 $ cluther
[17-Oct-2011 08:54:44] <cluther> jmp242: No question about it.
[17-Oct-2011 08:55:11] <Hackman238> *&
[17-Oct-2011 08:55:44] <Hackman238> (thought I had an amperstand rather than a dollar sign in my paste buffer)
[17-Oct-2011 08:55:57] <jmp242> I have two forum threads that stump me a bit - anyone see: message/61904#61904
[17-Oct-2011 08:56:20] <Hackman238> jmp242: Ill take a look
[17-Oct-2011 08:56:41] <jmp242> I think this one, being a zenpack issue, is for bigegor
[17-Oct-2011 08:56:53] <jmp242> but before I e-mail it around, maybe someone has an idea here: message/61810#61810
[17-Oct-2011 08:57:30] <jmp242> I've been using it with no issues (the zenpack) so I would figure it's something the user did, but I have no idea what they might have edited to break it lol
[17-Oct-2011 09:07:41] <jb> morning.
[17-Oct-2011 09:08:06] kocolosk|away is now known as kocolosk
[17-Oct-2011 09:31:35] <St3v3o> good morning
[17-Oct-2011 09:33:43] <St3v3o> we had an odd issue today with a bunch of our boxes where zenoss was alarming that the file system was WAY over its threshold. An example is disk space threshold: 385.6% used (-281.7MB free). This was definitely calculated incorrectly, I restarted zenoss and remodeled a bunch of the nodes that were not reporting correctly and it looks to have cleared up. Has anyone seen anything similar…or might be able to point me to a log
[17-Oct-2011 09:35:50] <jmp242> I have for disk space
[17-Oct-2011 09:36:01] <jmp242> if the disk space changed, say by expanding a partition
[17-Oct-2011 09:36:09] <jmp242> Zenoss may not know about it till it is remodeldd
[17-Oct-2011 09:36:24] <jmp242> and zenmodeler seems to zombieify itself
[17-Oct-2011 09:36:33] <jmp242> so you may need to restart it even if it's heartbeating correctly
[17-Oct-2011 09:37:18] <St3v3o> the partition wasn't modified…and the really odd part is that it was reporting the /boot partition was over by 200+%
[17-Oct-2011 09:37:55] <St3v3o> this happened to the monitoring of the /boot partition for about 20 servers
[17-Oct-2011 09:39:00] <cluther> St3v3o: Did anything else happen to these 20 servers that might have caused it? I've seen this sort of thing before, but never on so many servers at the same time.
[17-Oct-2011 09:39:27] <St3v3o> nope
[17-Oct-2011 09:39:32] <cluther> Usually the cause is the SNMP indexes of the file systems changing. So Zenoss is monitoring what was the SNMP index of the /boot filesystem, but that SNMP index is now the / filesystem.
[17-Oct-2011 09:39:33] <St3v3o> no other alarms
[17-Oct-2011 09:39:52] <cluther> I'm thinking if they all got an updated net-snmp package and restarted snmpd daemon this kind of thing could happen.
[17-Oct-2011 09:39:54] <St3v3o> and it was spread across multiple data centers
[17-Oct-2011 09:40:09] <St3v3o> hmm….thats a good theory
[17-Oct-2011 09:40:15] <St3v3o> let me check my log watch
[17-Oct-2011 09:42:07] <enkrypt> Hackman238: this is the traceback I keep getting in my zenhub.log: http://dpaste.com/635896/
[17-Oct-2011 09:42:24] <St3v3o> cluther: you are correct it looks like net-snmp was updated on these boxes
[17-Oct-2011 09:42:27] <St3v3o> I should have checked that
[17-Oct-2011 09:42:29] <Hackman238> enkrypt: try zenchkrels -x1 -r
[17-Oct-2011 09:43:22] <cluther> St3v3o: Makes sense. The next zenmodeler run realigns the SNMP indexes and should have put things back to normal on its own.
[17-Oct-2011 09:44:05] <St3v3o> yeah just sucks it sent a bunch of alarms before doing so
[17-Oct-2011 09:47:18] <cluther> enkrypt: Somehow this device is not assigned to a collector. You paste the snippet at the following URL into zendmd to find and fix all of these.
[17-Oct-2011 09:47:20] <cluther> enkrypt: https://gist.github.com/1292636
[17-Oct-2011 09:48:12] <Hackman238> enkrypt: ^ that should work too
[17-Oct-2011 09:48:30] <enkrypt> what do I do, the zenchkrels command, or the zendmd snippet? or both?
[17-Oct-2011 09:48:38] <Hackman238> enkrypt: the snippet
[17-Oct-2011 09:48:43] <cluther> It would be a good idea to do that zenchkrels too because this isn't something that should happen naturally and might have other symptoms.
[17-Oct-2011 09:48:52] <Hackman238> enkrypt: then chkrels after
[17-Oct-2011 09:51:05] <enkrypt> Hackman238: the snippet didn't do anything
[17-Oct-2011 09:51:35] <Hackman238> enkrypt: there shouldnt be any output from it
[17-Oct-2011 09:52:12] <Hackman238> enkrypt: correction, yes there should
[17-Oct-2011 09:52:20] <Hackman238> haus1: just reread the code. Thats odd.
[17-Oct-2011 09:52:47] <Hackman238> enkrypt: just reread the code. Thats odd.
[17-Oct-2011 09:52:56] <enkrypt> indeed, it should print smth
[17-Oct-2011 09:53:07] <enkrypt> now doing the zenchkrels command
[17-Oct-2011 09:53:36] <Hackman238> enkrypt: can you:
[17-Oct-2011 09:53:38] <Hackman238> for d in dmd.Devices.getSubDevicesGen(): print str(d.getPerformanceServer())
[17-Oct-2011 09:53:49] <Hackman238> enkrypt: after zenchkrels
[17-Oct-2011 09:54:01] <enkrypt> ok will do
[17-Oct-2011 09:54:05] <enkrypt> takes some time, big installation
[17-Oct-2011 09:54:34] <Hackman238> enkrypt: np. that for loop will fail on whatever device(s) have a bad collector set and we'll fix it
[17-Oct-2011 09:55:09] <enkrypt> Hackman238: that backtrace shows up in zenhub.log 5 times in a row every few minutes
[17-Oct-2011 09:55:34] <enkrypt> so probably as you said there are a few orphan devices that haven't got a collector assigned
[17-Oct-2011 09:56:15] <Hackman238> enkrypt: they probably have a bad value or something. if they were None chets code would have caught them
[17-Oct-2011 10:09:37] <enkrypt> Hackman238: & cluther: still checking relations
[17-Oct-2011 10:10:39] <enkrypt> Hackman238: the second zendmd snippet shows no errors by the way
[17-Oct-2011 10:11:09] <Hackman238> enkrypt: ok
[17-Oct-2011 10:36:46] <enkrypt> Hackman238: so I did the first zendmd snippet (no results), then the zenchkrels command (no error) then the second zendmd snippet (no problem)
[17-Oct-2011 10:37:15] <Hackman238> enkrypt: Hum.
[17-Oct-2011 10:37:19] <enkrypt> Hackman238: but even so, 1 minute ago the tracebacks showed up in zenhub.log
[17-Oct-2011 10:37:27] <Hackman238> Thats very strange
[17-Oct-2011 10:37:33] <Hackman238> enkrypt: can sed the trace again?
[17-Oct-2011 10:38:38] <cluther> Looking at the snippet again, I think I led you astray with the device-not-assigned-to-collector statement.
[17-Oct-2011 10:39:29] <enkrypt> here are the traces: http://dpaste.com/635927/
[17-Oct-2011 10:51:20] <St3v3o> cluther: With the snmp indexes…where are they located in the file system? I'd like to see if I can verify indexes were "out of wack" when we were having our issues this morning
[17-Oct-2011 10:51:35] <St3v3o> and to figure out how to stop it from happening again
[17-Oct-2011 10:53:48] <cluther> St3v3o: You can see them as the last .# if you snmpwalk the following OID on your devices.
[17-Oct-2011 10:53:50] <cluther> .1.3.6.1.2.1.25.2.3.1
[17-Oct-2011 10:54:14] <Hackman238> enkrypt: if you zenhub stop does zenhub stop?
[17-Oct-2011 10:54:27] <Hackman238> enkrypt: check ps aux to make sure
[17-Oct-2011 10:54:59] <cluther> St3v3o: If you look for /Change events in your event history you should see a bunch of informational events about the "snmpindex" attribute changing.
[17-Oct-2011 10:59:17] <St3v3o> i see….what would cause the index to change via net-snmp ?
[17-Oct-2011 11:03:45] <enkrypt> Hackman238: checking.
[17-Oct-2011 11:04:18] <enkrypt> Hackman238: yes it did stop, verified with ps
[17-Oct-2011 11:06:12] <whyzgeek> hmm, modeling pushes back the changes to the collector. When it does that it sends down a list of devices which one of them is None. I don't think it is related to filesystem map.
[17-Oct-2011 11:06:13] <Hackman238> enkrypt: hum. ok start it back up if you havent already
[17-Oct-2011 11:10:57] <St3v3o> looks like I may have found a bug that caused my issue ...https://bugzilla.redhat.com/show_bug.cgi?id=595322
[17-Oct-2011 11:18:09] <enkrypt> Hackman238: yeah, zenhub started again.
[17-Oct-2011 11:36:18] <jmp242> Hey, Hackman238 - have any input on message/61910#61910
[17-Oct-2011 11:36:43] <jmp242> Is anyone even using commands really in Zenoss Web UI? I always assumed people just opened a terminal ...
[17-Oct-2011 11:40:57] <enkrypt> jmp242: I use the commands to do a quick snmpwalk or ping once and a while yeah, I find it perfect for that
[17-Oct-2011 12:00:01] [disconnected at Mon Oct 17 12:00:01 2011]
[17-Oct-2011 12:00:02] [connected at Mon Oct 17 12:00:02 2011]
[17-Oct-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]
[17-Oct-2011 12:02:03] <rmatte> Is anyone in here monitoring a fair bit of Nortel gear with Zenoss by any chance? I'm having a hell of a time tracking down OIDs, figured I'd check if someone already has some on hand.
[17-Oct-2011 12:02:10] <rmatte> Really just looking for CPU/Memory OIDs
[17-Oct-2011 12:02:30] <rmatte> Nortel's MIBs are a pain to go through
[17-Oct-2011 12:22:45] <dhopp> rmatte: I did a few years ago but I don't remember if I got cpu/memory...
[17-Oct-2011 12:22:50] <dhopp> what model nortel?
[17-Oct-2011 12:35:13] <jmp242> rmatte: thread/16904?start=0&tstart=0
[17-Oct-2011 12:35:17] <jmp242> does that help any
[17-Oct-2011 12:37:30] <rmatte> it definitely does if it'll run in 2.5
[17-Oct-2011 12:37:36] <rmatte> which it should
[17-Oct-2011 12:37:47] <rmatte> actually hmmm, it might not
[17-Oct-2011 12:37:54] <rmatte> looks like he's added all sorts of custom components
[17-Oct-2011 12:38:55] <rmatte> also, why is that pack not on the ZenPacks page? lol
[17-Oct-2011 12:39:00] <jmp242> lol
[17-Oct-2011 12:39:06] <jmp242> I think you know why
[17-Oct-2011 12:39:33] <jmp242> he's beta testing it now, but I'm not sure he knows how to get it on the zenpacks page
[17-Oct-2011 12:39:52] <rmatte> oh well, I'll install it on my 3.x test box and grab the OIDs from it
[17-Oct-2011 12:39:53] <rmatte> thanks
[17-Oct-2011 12:39:57] <jmp242> at the least some of the people on that thread seem to know the OIDs
[17-Oct-2011 12:40:03] <dhopp> I wonder how many of those zenpacks are out there though (could be quite useful but not on the zenpack page)
[17-Oct-2011 12:40:11] <jmp242> you really need to get off of 2.x lol
[17-Oct-2011 12:40:50] <jmp242> dhopp: there are a number that I've seen
[17-Oct-2011 12:40:55] <rmatte> I can't while the 3.x releases all blow
[17-Oct-2011 12:41:15] <jmp242> 3.1 is nice enough I think . . .
[17-Oct-2011 12:41:32] <rmatte> I haven't seen a single release yet in the 3.x stream that I would consider production worthy.
[17-Oct-2011 12:41:38] <jmp242> oh
[17-Oct-2011 12:41:39] <jmp242> ok
[17-Oct-2011 12:41:44] <pipeline> how are you feeling about 4.x
[17-Oct-2011 12:41:51] <rmatte> I haven't seen 4.x
[17-Oct-2011 12:42:28] <rmatte> but I really hope that it's a vast improvement
[17-Oct-2011 12:43:21] <dhopp> jmp242: rmatte really dislikes the UI…I personally think 3.1 and 3.2 (with patches) is pretty good…there are still some annoying bugs but nothing that I personally think are show stoppers
[17-Oct-2011 12:43:22] <jmp242> We'll know in a few months when 4.2 is supposed to be out
[17-Oct-2011 12:43:35] <jmp242> I hated the new UI
[17-Oct-2011 12:43:40] <jmp242> but now I'm sort of used to it
[17-Oct-2011 12:43:41] <rmatte> I hate the limitations on copying objects
[17-Oct-2011 12:43:49] <rmatte> I do so much object copying that it makes it a total no go for me
[17-Oct-2011 12:43:50] <jmp242> it's kind of like the ribbon lol
[17-Oct-2011 12:44:08] <jmp242> can't you use dmd to copy objects?
[17-Oct-2011 12:44:09] <rmatte> I'm not going to copy one template at a time when I have 20 to copy
[17-Oct-2011 12:44:28] <rmatte> I probably could, but other people have to use this other than myself
[17-Oct-2011 12:44:36] <jmp242> Oh
[17-Oct-2011 12:44:40] <rmatte> and they haven't worked with the back end at all
[17-Oct-2011 12:44:48] <jmp242> I thought if you were doing it regularily, scripting might be the way to go
[17-Oct-2011 12:45:09] <rmatte> the thing is, why are we still having to deal with this 3 versions in to the 3.x stream?
[17-Oct-2011 12:45:22] <jmp242> Might be hard to make work with the new UI
[17-Oct-2011 12:45:25] <rmatte> It's a regression from the old UI functionality
[17-Oct-2011 12:45:30] <rmatte> and yet no one seems to care
[17-Oct-2011 12:45:44] <jmp242> mmm, why do you have to copy templates all the time, many at a time?
[17-Oct-2011 12:45:55] <jmp242> shouldn't inheritance make this work easier?
[17-Oct-2011 12:45:59] <rmatte> interface templates
[17-Oct-2011 12:46:02] <jmp242> i.e. use heirarchies of device classes
[17-Oct-2011 12:46:11] <jmp242> oh
[17-Oct-2011 12:46:16] <rmatte> when I update them from a ZenPack across all my instances, the ZenPack won't install them in /
[17-Oct-2011 12:46:21] <rmatte> so I install them in /Interfaces
[17-Oct-2011 12:46:24] <rmatte> then copy them over to /
[17-Oct-2011 12:46:31] <jmp242> Oh
[17-Oct-2011 12:46:34] <jmp242> tricky
[17-Oct-2011 12:46:35] <rmatte> and I have like 25 interface templates in my interface ZenPack
[17-Oct-2011 12:46:54] <jmp242> but that still seems like it ought to be scriptable
[17-Oct-2011 12:46:58] <rmatte> or sometimes when I install a ZenPack I want to create new classes for it and copy all the templates from the default class over
[17-Oct-2011 12:47:12] <rmatte> copying templates one at a time is impractical, bottom line
[17-Oct-2011 12:47:28] <dhopp> jmp242: it can be but he does have a pretty good point…the 2.x UI had this functionality and it's very useful, why remove a very useful feature?
[17-Oct-2011 12:47:32] <rmatte> and then there's the fact that you can't even move device classes
[17-Oct-2011 12:47:43] <rmatte> which was possible previously
[17-Oct-2011 12:47:45] <jmp242> Yea
[17-Oct-2011 12:47:50] <rmatte> the same goes for most other objects, groups, locations
[17-Oct-2011 12:47:55] <rmatte> I do a lot of work with groups and locations
[17-Oct-2011 12:47:58] <jmp242> I didn't notice lol
[17-Oct-2011 12:48:06] <jmp242> Yea, the new UI isn't awesome
[17-Oct-2011 12:48:06] <rmatte> the whole thing is just one big obstacle that I don't want to have to deal with
[17-Oct-2011 12:48:23] <jmp242> mmm
[17-Oct-2011 12:48:35] <rmatte> I really wish they'd focus on fixing these issues rather than pressing onward with new features
[17-Oct-2011 12:48:42] <dhopp> rmatte: I'm going to guess that most of that will still be a problem in 4.x…isn't the 4.x basically the same UI but with substantial backend improvements?
[17-Oct-2011 12:48:43] <rmatte> they revamp the UI and then leave it half broken
[17-Oct-2011 12:49:05] <rmatte> dhopp: according to Hackman, it is still a problem in 4.x
[17-Oct-2011 12:49:11] <rmatte> which makes me even more upset
[17-Oct-2011 12:49:13] <rmatte> lol
[17-Oct-2011 12:49:18] <jmp242> great, you'll be on 2.5.2 forever
[17-Oct-2011 12:49:30] <rmatte> At least until they fix the object copying
[17-Oct-2011 12:50:05] <rmatte> I'm managing 15 separate instances, I can't deal with stuff like this
[17-Oct-2011 12:51:01] <rmatte> ...and I feel that I can't even really contribute to the community with finding/logging bugs and creating packs because I've a bunch of versions behind.
[17-Oct-2011 12:51:07] <rmatte> it's frustrating
[17-Oct-2011 12:51:16] <rmatte> I'm*
[17-Oct-2011 12:51:29] <jmp242> I imagine
[17-Oct-2011 12:51:32] <dhopp> rmatte: maybe you could contribute a zenpack that fixes all of your annoyances ;-)
[17-Oct-2011 12:51:35] <jmp242> I feel for you
[17-Oct-2011 12:51:50] <rmatte> dhopp: It would involve monkey patching which I'd rather not do lol
[17-Oct-2011 12:51:56] <jmp242> Otherwise I don't have any constructive comments - I figure there are bugs logged so it's up to Zenoss
[17-Oct-2011 12:52:08] <rmatte> Exactly
[17-Oct-2011 12:52:23] <rocket> there are serious issues with the 2.5.2 ui from an extensibility standpoint ..
[17-Oct-2011 12:52:26] <rmatte> It's technically up to the sales guys at Zenoss
[17-Oct-2011 12:52:30] <rmatte> it's about what they want to see
[17-Oct-2011 12:52:57] <rmatte> rocket: The 2.5.2 UI is more functional than the new one, and until that changes I'm stuck on 2.5.2
[17-Oct-2011 12:53:04] <rmatte> and I've been patiently waiting for over a year now
[17-Oct-2011 12:53:20] <rmatte> the issues just never get addressed
[17-Oct-2011 12:54:30] <jmp242> While we're ranting
[17-Oct-2011 12:54:52] <jmp242> on the Zenpacks page - what is the current method to get these things on the list? Is it still nyeats?
[17-Oct-2011 12:55:15] <rmatte> jmp242: as far as I'm aware yes
[17-Oct-2011 12:55:26] <jmp242> I saw a bunch moved to some git instance, but I think I'd love to suggest the Nortel poster put them somewhere people other than me can find them
[17-Oct-2011 12:56:49] <rmatte> The current nortel pack that's up there is garbage. We need to start obsoleting old packs when replacements come along
[17-Oct-2011 12:57:19] <jmp242> I just asked the poster if I can poke Nick re that updated pack
[17-Oct-2011 12:57:37] <rmatte> well the real question is, does he have to sign the contrib agreement?
[17-Oct-2011 12:57:38] <rmatte> lol
[17-Oct-2011 12:57:55] <rmatte> I haven't been keeping up to date with the ZCA as much as I should lately since I've been swamped
[17-Oct-2011 12:57:57] <jmp242> Well, yea
[17-Oct-2011 12:58:01] <rmatte> not sure where we're at with that?
[17-Oct-2011 12:58:11] <jmp242> Well, we're waiting to hear from Zenoss
[17-Oct-2011 12:58:38] <jmp242> I haven't set up a ZCA git because I don't want to fragment, but we need to figure it out I guess
[17-Oct-2011 12:58:49] * rmatte nods
[17-Oct-2011 12:59:13] <jmp242> or hell, we could use a public git I guess, but as I said, Zenoss has started to, so if they'd drop the agreement, we'd have a single place to look
[17-Oct-2011 12:59:14] <rmatte> I need to learn how to actually use git since I'm a subversion guy
[17-Oct-2011 12:59:34] <jmp242> that said, if there needs to be a GPL git and a Zenoss Contrib Agreement Git, that can happen I guess
[17-Oct-2011 12:59:43] * rmatte nods
[17-Oct-2011 13:00:15] <jmp242> I'm more interested in cleaning up the Zenpacks page
[17-Oct-2011 13:00:29] <rmatte> absolutely
[17-Oct-2011 13:00:33] <jmp242> but I don't think that was possible with Jive I guess
[17-Oct-2011 13:00:36] <rmatte> it's a mess
[17-Oct-2011 13:00:38] <jmp242> some limitation of the platform
[17-Oct-2011 13:00:55] <jmp242> Anyway, we've all been swamped
[17-Oct-2011 13:00:58] <jmp242> so don't feel bad rmatte
[17-Oct-2011 13:01:12] <rmatte> It would be nice if there was a separate sort of "ZenPacks Forge" page or something
[17-Oct-2011 13:01:19] <rmatte> would give more control of the format
[17-Oct-2011 13:01:36] <jmp242> Something
[17-Oct-2011 13:01:56] <rocket> rmatte: honestly there are reasons some of the UI changes have happened re organizers moving etc .. the way zenoss is architected forces us to keep some customers from totally shooting thier foot
[17-Oct-2011 13:02:35] <rmatte> rocket: the big thing for me is not being able to copy multiple templates at once
[17-Oct-2011 13:02:43] <rocket> the multiple template select I can see .. but the moving organizers issue is much less of an issue in my mind. We had more support calls from organizers changing than from customers complaining we couldnt change them
[17-Oct-2011 13:02:47] <rmatte> that's a total show stopper for me
[17-Oct-2011 13:03:41] <rmatte> rocket: That's like microsoft saying "Well, people were accidentally copying files in to the wrong directory, so we disabled the ability to copy files"
[17-Oct-2011 13:03:56] <rmatte> User education
[17-Oct-2011 13:04:28] <rmatte> or at least there should be some "Advanced" back-end section where you can actually do it
[17-Oct-2011 13:04:28] <jmp242> rmatte: user education is expensive
[17-Oct-2011 13:04:33] <jmp242> and often doesn't actually work
[17-Oct-2011 13:04:35] <rmatte> just don't make it exposed by default
[17-Oct-2011 13:04:37] <jmp242> vs disabling stuff
[17-Oct-2011 13:04:40] <rocket> rmatte: we have enough other things to deal with first .. before resolving that issue
[17-Oct-2011 13:04:48] <jmp242> look at Apple's business model lol
[17-Oct-2011 13:04:51] <rmatte> make them actually have to go to a section with a warning like "If you don't know what you're doing, turn back now!"
[17-Oct-2011 13:05:20] <rocket> you guys are not supporting it though .. I hear what your saying, but you really havent had to deal with the customers ..
[17-Oct-2011 13:05:47] <rmatte> rocket: I'm sure you do, but it's like new work is constantly being created for you guys and no one cares about actually having to fix these issues that have been around for well over a year now
[17-Oct-2011 13:06:21] <rmatte> ...and as such I'm stuck twiddling my thumbs waiting for a version that can
[17-Oct-2011 13:06:25] <jmp242> mmm, rmatte I think rocket is saying it was more of a design decision
[17-Oct-2011 13:06:31] <jmp242> you might be stuck
[17-Oct-2011 13:06:52] <rmatte> jmp242: I understand what he was saying... but there's no reason why they can't add some sort of "Advanced Organizer Management" section
[17-Oct-2011 13:07:15] <rmatte> and have a warning dialog when entering it
[17-Oct-2011 13:07:27] <rmatte> crippling power users because of typical users is lame
[17-Oct-2011 13:08:59] <rocket> rmatte: the template copy is an issue I just havent looked at
[17-Oct-2011 13:09:55] <rmatte> If it could be looked at within the next couple of months it would be hugely appreciated.
[17-Oct-2011 13:10:06] <rmatte> but I realize that you don't make the schedule.
[17-Oct-2011 13:11:42] <rmatte> It's getting to the point where I'm going to have to upgrade, but the template copying issue is going to be really painful for me
[17-Oct-2011 13:11:52] <rmatte> I just don't want to get too far behind in terms of versions.
[17-Oct-2011 13:27:17] <rmatte> hmmm, this nortel pack doesn't really help me...
[17-Oct-2011 13:27:30] <rmatte> no CPU/Memory OIDs for anything other than passport devices, which I already had
[17-Oct-2011 13:27:36] <rmatte> and the collector plugins will only work with 3.x
[17-Oct-2011 13:28:05] <tvincent> any timeline for 3.2.1? it seems have been next week for a couple of weeks.
[17-Oct-2011 13:28:28] <rmatte> tvincent: I'd honestly rather see it take longer and actually get a solid release out
[17-Oct-2011 13:28:46] <rmatte> They rushed 3.2 which is why they need a 3.2.1 in the first place
[17-Oct-2011 13:28:59] <tvincent> @rmatte agreed
[17-Oct-2011 13:29:17] <tvincent> but when i tell my boss next week two weeks ago. I like to give a more accurate answer the next time he asks.
[17-Oct-2011 13:29:44] <rmatte> Just say that they are following a "When it's done" release on this one and that you're not sure.
[17-Oct-2011 13:45:00] <rmatte> Nice, the datacenter that we host our stuff at has an outage causing major packet loss to our stuff
[17-Oct-2011 14:22:04] <rocket> rmatte why cant you copy a root level template?
[17-Oct-2011 14:22:22] <rocket> rmatte: that seems to me to be your main issue in this .. because you have to do the moving etc
[17-Oct-2011 14:23:22] <dhopp> rmatte: I'm asking because I'm not sure…but could you use the ZMI as the "advanced" way of moving objects around?
[17-Oct-2011 15:18:40] <rmatte> dhopp: you probably could but it's not really suited for that. Often when I try to copy objects with it I'm met with errors
[17-Oct-2011 15:19:01] <rocket> I would NOT recommend anyone really use the ZMI
[17-Oct-2011 15:19:10] <rocket> you are just asking for hidden problems down the road
[17-Oct-2011 15:19:12] <rmatte> plus you can do a heck of a lot more damage with ZMI than you could with a UI specifically coded to do it
[17-Oct-2011 15:19:31] <rmatte> no safeguards in place
[17-Oct-2011 15:19:33] <rocket> I understand there are exceptions but do try and avoid it
[17-Oct-2011 15:19:53] <dhopp> rocket: he was saying he wanted an advanced page…does it get any more advanced then the ZMI ;-)
[17-Oct-2011 15:20:29] <rocket> well if I understand the problem .. he really wnats to be able to copy templates from one system to another
[17-Oct-2011 15:21:05] <rocket> I cannot find a restriction that would keep you from copying them from/to the top level .. I just tested that
[17-Oct-2011 15:38:25] <rmatte> no no
[17-Oct-2011 15:38:29] <rmatte> the problem is the following...
[17-Oct-2011 15:38:50] <rmatte> It won't allow me to add templates at the root level to a ZenPack, so when I want to update them across all my servers I do the following...
[17-Oct-2011 15:38:58] <rmatte> I edit them in /Interfaces, load them in to a ZenPack
[17-Oct-2011 15:39:00] <rmatte> deploy the ZenPacks
[17-Oct-2011 15:39:31] <rmatte> and when it's deployed to all of the servers I go on to each one and copy the 25+ interface templates from /Interfaces to the root (/)
[17-Oct-2011 15:39:45] <rmatte> instead of being able to copy 25 at once I'd be stuck doing one at a time
[17-Oct-2011 15:39:48] <rmatte> which is just ridiculous
[17-Oct-2011 15:40:30] <rmatte> Any time someone needs to copy a large amount of templates it's a massive waste of time with the new UI
[17-Oct-2011 15:46:18] <rocket> rmatte: at what root level? I just tried doing a copy into a zenpack and it worked just fine for me
[17-Oct-2011 15:51:55] <rmatte> I meant at /Devices
[17-Oct-2011 15:52:21] <rmatte> Either way, that's only one of many possible scenarios for wanting to copy more than a single template at once
[17-Oct-2011 15:52:50] <rmatte> the other thing is that a ZenPack won't auto-update the templates
[17-Oct-2011 15:53:02] <rmatte> it'll see that templates are already there and won't replace them
[17-Oct-2011 15:53:19] <rmatte> so I have to manually do it, and I much prefer knowing that the templates installed properly to /Interfaces before doing that
[17-Oct-2011 15:53:38] <rmatte> At the end of the day it's a regression from the old UI
[17-Oct-2011 17:16:13] kocolosk is now known as kocolosk|away
[17-Oct-2011 20:52:50] kocolosk|away is now known as kocolosk
[17-Oct-2011 22:49:24] kocolosk is now known as kocolosk|away
[18-Oct-2011 00:00:01] [disconnected at Tue Oct 18 00:00:01 2011]
[18-Oct-2011 00:00:02] [connected at Tue Oct 18 00:00:02 2011]
[18-Oct-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[18-Oct-2011 07:53:56] <jmp242> Mon3y752
[18-Oct-2011 09:44:49] kocolosk|away is now known as kocolosk
[18-Oct-2011 10:45:58] <twm1010> Jane_Curry: testing out your JunOS zenpack today
[18-Oct-2011 10:46:28] <rmatte> You made a JunOS pack? Will it work in 2.5 or is it 3.x specific?
[18-Oct-2011 10:46:47] <twm1010> says 3.1
[18-Oct-2011 10:47:03] <twm1010> docs/DOC-10328
[18-Oct-2011 10:47:59] <kerick> has anyone used tales expressions in an alert email?
[18-Oct-2011 10:48:27] <kerick> i am having an issue trying to push a device's custom property
[18-Oct-2011 10:50:13] <twm1010> wow, this is one nice zenpack, thank you jane!
[18-Oct-2011 10:50:19] <rmatte> kerick: Pretty sure that you're limited to event properties
[18-Oct-2011 10:51:16] <rmatte> kerick: you'd have to use a transform to set one of the event properties to the value of that cProperty
[18-Oct-2011 10:51:19] <rmatte> then reference that
[18-Oct-2011 10:51:48] <kerick> rmatte: that works, i'll look into it, thanks
[18-Oct-2011 10:51:52] <rmatte> np
[18-Oct-2011 10:55:24] <rmatte> definitely 3.1 only since she used components
[18-Oct-2011 10:55:40] <rmatte> eugh, I need to get us upgraded soon and just work around the UI problems
[18-Oct-2011 10:56:09] <twm1010> isn't 4.x going to change the UI up a bit more?
[18-Oct-2011 10:56:13] <rmatte> need to work on getting the LDAP integration working properly in 3.x first
[18-Oct-2011 10:56:20] <rmatte> twm1010: not from what I hear
[18-Oct-2011 10:56:28] <rmatte> maybe from a coding standpoint
[18-Oct-2011 10:56:32] <dhopp> twm1010: my understanding is that the UI is basically the same…it's the guts that is different
[18-Oct-2011 10:56:37] <twm1010> ok
[18-Oct-2011 10:57:05] <twm1010> interesting, everytime i change the class of a device, it changes the name of the device to it's management IP
[18-Oct-2011 10:58:01] <jmp242> Jane_Curry: you on? You've delved far deeper into process monitoring than I, but I suspect that this is a bug or something in this thread: message/61896#61896
[18-Oct-2011 10:58:22] <rmatte> twm1010: that's odd, though it might not be actually renaming the device, just changing the title of the device.
[18-Oct-2011 10:58:32] <twm1010> well thats what i'm referring to
[18-Oct-2011 10:59:01] <rmatte> that's a pretty serious bug, is that in 3.2?
[18-Oct-2011 10:59:29] <twm1010> 3.1.0
[18-Oct-2011 10:59:35] <rmatte> ah
[18-Oct-2011 10:59:40] <twm1010> with maybe one patch installed, if i remember
[18-Oct-2011 10:59:49] <twm1010> i was holding off on 3.2 for prod
[18-Oct-2011 11:00:08] <rmatte> might as well wait for 3.2.1 yeh
[18-Oct-2011 11:01:01] <rmatte> I just moved a device between classes and didn't see anything like what you're describing
[18-Oct-2011 11:01:35] <twm1010> i should observe it from the cli and see if it changes the name of the directory for the perf data
[18-Oct-2011 11:02:29] <rmatte> What happens if you click on the device
[18-Oct-2011 11:02:34] <rmatte> and then clear the Device Name: field
[18-Oct-2011 11:02:36] <rmatte> and click save
[18-Oct-2011 11:02:48] <rmatte> that should make it display the actual (collection) device name
[18-Oct-2011 11:04:43] <rmatte> that's too funny...
[18-Oct-2011 11:04:50] <rmatte> so the Device Name: field is for a title...
[18-Oct-2011 11:05:07] <rmatte> but if you do an actual rename, it doesn't maintain the title that was previously set, it changes it to the new device name
[18-Oct-2011 11:05:10] <rmatte> that's confusing...
[18-Oct-2011 11:05:25] <rmatte> so people are going to change the Device Name: field thinking they are renaming the device when they actually aren't
[18-Oct-2011 11:05:40] <rmatte> yet when they actually do the rename properly it removes whatever they had set in the title field previously
[18-Oct-2011 11:05:47] <rmatte> what an absolutely ridiculous system
[18-Oct-2011 11:06:51] <rmatte> I guess the name of the game with this new UI is confusion
[18-Oct-2011 11:08:07] <rmatte> I need to find a way to make the Device Name: field not editable or I'm going to have a huge mess on my hands
[18-Oct-2011 11:08:24] <rmatte> because half my users will rename devices one way, the other half will do it the other way, I can just see it now.
[18-Oct-2011 11:19:25] <Sashness> rmatte: heh
[18-Oct-2011 11:30:06] <rmatte> hmmm, so I found the file responsible for the field, edited it, but no change... they must compile the code for performance or something
[18-Oct-2011 11:30:53] <davetoo> que?
[18-Oct-2011 11:31:12] <rmatte> I want to disable the "Device Name" input field on device overview pages
[18-Oct-2011 11:31:25] <rmatte> since I don't want any of my users confusing it with "Rename Device" in the menu
[18-Oct-2011 11:31:30] <rmatte> which can and will happen
[18-Oct-2011 11:31:31] <davetoo> .js or .pt file?
[18-Oct-2011 11:31:35] <rmatte> .js
[18-Oct-2011 11:31:58] <rmatte> disabled: true,
[18-Oct-2011 11:31:58] <rmatte> fieldLabel: _t('Device Name')
[18-Oct-2011 11:32:07] <rmatte> I assumed that would do it
[18-Oct-2011 11:32:11] <davetoo> yeah, you need to run buildjs.sh, which is somewhere under the inst directory in the sources
[18-Oct-2011 11:32:18] <rmatte> k I'll find it
[18-Oct-2011 11:33:01] <davetoo> it all ends up in one file (I forget exactly which/where). Actually, two files, one is a "debug" version with human-readable formatting.
[18-Oct-2011 11:33:04] <rmatte> oh, so I actually have to grab the sources, it doesn't come with the default install
[18-Oct-2011 11:33:09] <davetoo> correct
[18-Oct-2011 11:33:20] <rmatte> it ends up in some file called zenoss-compile.js I believe
[18-Oct-2011 11:33:43] <rmatte> zenoss-compiled.js
[18-Oct-2011 11:33:44] <rmatte> rather
[18-Oct-2011 11:34:03] <davetoo> You might be able to just grab the builder file by itself. it's been a few months since I did this.
[18-Oct-2011 11:34:24] <davetoo> version?
[18-Oct-2011 11:34:58] <davetoo> http://dev.zenoss.com/trac/browser/trunk/inst/buildjs.sh
[18-Oct-2011 11:35:06] <davetoo> modulo version
[18-Oct-2011 11:36:51] <rmatte> 3.2.1 beta
[18-Oct-2011 11:37:01] <rmatte> I just directly edited the compiled file, should work just as well
[18-Oct-2011 11:37:10] <davetoo> I would think so.
[18-Oct-2011 11:37:17] <rmatte> yup, that did it
[18-Oct-2011 11:37:32] <rmatte> perfect, now I just need to do that on all of my installs when I upgrade
[18-Oct-2011 11:38:19] <davetoo> mmmm... Peet's
[18-Oct-2011 11:38:49] <rmatte> I assume Peet's is an establishment that sells food?
[18-Oct-2011 11:38:54] <davetoo> coffee
[18-Oct-2011 11:39:07] <rmatte> ah
[18-Oct-2011 11:39:20] <davetoo> Peet's >> Charbuck$
[18-Oct-2011 11:39:35] <rmatte> hehe
[18-Oct-2011 11:39:59] <davetoo> If you want to keep quality engineers in my area (SF Bay), you need free Peet's instead of Starbucks
[18-Oct-2011 11:40:25] <rmatte> our most popular coffee place up here is Tim Horton's
[18-Oct-2011 11:40:45] <davetoo> I've heard that name, I think.
[18-Oct-2011 11:40:55] <rmatte> it's just regular coffee though, not much of these fancy shmancy types
[18-Oct-2011 11:41:32] <rmatte> Tim Horton's has more locations than McDonald's does in Canada
[18-Oct-2011 11:41:36] <rmatte> which is fairly impressive
[18-Oct-2011 11:42:21] <davetoo> oh, right.. he was a hockey player?
[18-Oct-2011 11:42:26] <rmatte> yeh he was
[18-Oct-2011 11:43:04] <davetoo> I have a couple of Canadian teammates.
[18-Oct-2011 11:43:27] <rmatte> cool
[18-Oct-2011 11:48:00] <twm1010> heh, its amusing watching this thing model the vlans on my switch used for vcloud
[18-Oct-2011 11:53:15] <twm1010> hey rmatte, do you have a checklist for building a zenoss VM?
[18-Oct-2011 11:53:28] <twm1010> like a "things you change from stock" immediately
[18-Oct-2011 11:53:44] <twm1010> for me, its filesystem and interface map ignore names|types
[18-Oct-2011 11:53:49] <twm1010> and adding some templates and zenpacks
[18-Oct-2011 11:54:10] <rmatte> I maintain a template VM image, so when I need a new Zenoss VM I just clone a new image from that, spin it up, edit the interface/hosts/hostname files and I'm good to go
[18-Oct-2011 11:54:21] <rmatte> takes me like 5 minutes to setup a new server
[18-Oct-2011 11:54:38] <twm1010> well that's what im working towards, actually
[18-Oct-2011 11:54:45] <twm1010> you use ubuntu base right?
[18-Oct-2011 11:54:50] <rmatte> My image was designed over years though
[18-Oct-2011 11:54:54] <rmatte> yeh, Ubuntu Server
[18-Oct-2011 11:55:52] <twm1010> right, i have two images but each is a little different, little tweaks done differently, trying to make my playbook cookie cutter install and pull together the best of both
[18-Oct-2011 11:57:45] <rmatte> cool
[18-Oct-2011 12:00:01] [disconnected at Tue Oct 18 12:00:01 2011]
[18-Oct-2011 12:00:02] [connected at Tue Oct 18 12:00:02 2011]
[18-Oct-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[18-Oct-2011 12:16:37] <kyle__> I'm on zenoss 3, and I temporarily brought up an smtp service on one server. Then I got rid of it, since I don't want it. Now zenoss is monitoring the service, and even when I delete it, it still monitors it, and still throws a service down event.
[18-Oct-2011 12:17:38] <kyle__> Anyone know a solution?
[18-Oct-2011 12:18:37] <twm1010> i can't seem to remember that
[18-Oct-2011 12:18:52] <twm1010> you're deleting the event?
[18-Oct-2011 12:19:48] <kyle__> I'm deleting the ip service from the IP Services list on that server. In older versions of zenoss that's what you did to stop monitoring that service.
[18-Oct-2011 12:20:03] <twm1010> might need to "push changes to the collector" after you do so
[18-Oct-2011 12:20:07] <kyle__> ... I think.
[18-Oct-2011 12:20:37] <twm1010> can't remember which daemon monitors those
[18-Oct-2011 12:20:55] <kyle__> Humm. OK. Tried that. Now to wait a few minutes and see if it complains again.
[18-Oct-2011 12:20:59] <kyle__> Thanks.
[18-Oct-2011 12:31:15] <rmatte> kyle__: restart zenprocess
[18-Oct-2011 12:31:24] <rmatte> that forces it to pick up the new config from zenhub
[18-Oct-2011 12:31:30] <rmatte> that's what I always do
[18-Oct-2011 12:32:19] <rmatte> pushing changes doesn't always work with zenprocess
[18-Oct-2011 12:56:00] <cgibbons> tada
[18-Oct-2011 12:58:06] <rmatte> ?
[18-Oct-2011 12:58:13] <rmatte> got your pack working?
[18-Oct-2011 12:59:26] <cgibbons> some, i haven't had time to work on it since friday
[18-Oct-2011 13:03:45] <cgibbons> got the framework for collecting a bunch of stuff at once via a REST API, just need to turn that into modeler objects now
[18-Oct-2011 13:04:28] <rmatte> cool
[18-Oct-2011 13:19:47] <twm1010> anyone using the vcloud zenpack?
[18-Oct-2011 13:29:40] <pplask> hello!
[18-Oct-2011 13:30:14] <pplask> how do i send an alert when there is a ping packet loss to a device?
[18-Oct-2011 13:33:17] <pplask> ?
[18-Oct-2011 13:33:48] <rmatte> pplask: Read the Zenoss Administration guide, this is basic stuff that it covers in detail
[18-Oct-2011 13:34:03] <rmatte> http://downloads.sourceforge.net/zenoss/Zenoss_Administration_06-102010-3.0-v03.pdf
[18-Oct-2011 13:34:56] <pplask> rmatte, you mean, specific about monitoring packet loss?
[18-Oct-2011 13:35:27] <rmatte> when a ping fails you get an ip 192.168.0.1 is down event, or whatever the IP is
[18-Oct-2011 13:35:34] <rmatte> It also has a class of /Status/Ping
[18-Oct-2011 13:35:49] <rmatte> you can setup alert filters to only send an alert when it matches that type of event
[18-Oct-2011 13:35:59] <rmatte> and setting the filters is covered in the admin guide
[18-Oct-2011 13:36:02] <pplask> rmatte, yes, i know, but i ment, packet loss.. sometimes ping doesnt fails, but there is packet loss
[18-Oct-2011 13:36:16] <rmatte> there is no build in packet loss measurement in Zenoss
[18-Oct-2011 13:36:30] <rmatte> you'd need to use some package like the fping ZenPack
[18-Oct-2011 13:37:10] <rmatte> It's fairly intensive to properly calculate packet loss each ping cycle which is probably why they don't bother doing it
[18-Oct-2011 13:37:17] <pplask> rmatte, this is bad. cos im trying to replace our nagios by zenoss, but nagios just detected packet loss to a server and zenoss didnt say a word, and we had a problem that was solved thanks to nagios
[18-Oct-2011 13:37:59] <rmatte> then try the fping zenpack
[18-Oct-2011 13:38:29] <pplask> im looking at it
[18-Oct-2011 13:38:52] <kyle__> quit
[18-Oct-2011 13:38:55] <kyle__> wrong window
[18-Oct-2011 13:38:57] <rmatte> haha
[18-Oct-2011 13:49:49] <jmp242> I second fping
[18-Oct-2011 14:37:16] <Sashness> wondering if someone could help me with an event transition: http://pastebin.com/8ZLw6x2J
[18-Oct-2011 14:37:37] <Sashness> the event is the default high load in Zenoss: threshold of high load exceeded: current value 2323
[18-Oct-2011 14:37:49] <Sashness> and all i want to do is divide the number by 10
[18-Oct-2011 14:38:16] <Sashness> http://pastebin.com/8ZLw6x2J <— is the event transition i wrote but zenoss doesn't even match it
[18-Oct-2011 14:45:07] <nyeates> hi
[18-Oct-2011 14:45:12] <Sashness> hey nyeates
[18-Oct-2011 14:45:30] <Sashness> you've always been helpful, maybe you know the answer to this
[18-Oct-2011 14:45:33] <nyeates> ooh brownies just finished...
[18-Oct-2011 14:45:36] <Sashness> wondering if someone could help me with an event transition: http://pastebin.com/8ZLw6x2J
[18-Oct-2011 14:45:36] <Sashness> 2:37
[18-Oct-2011 14:45:36] <Sashness> the event is the default high load in Zenoss: threshold of high load exceeded: current value 2323
[18-Oct-2011 14:45:36] <Sashness> 2:37
[18-Oct-2011 14:45:37] <Sashness> and all i want to do is divide the number by 10
[18-Oct-2011 14:45:38] <Sashness> 2:38
[18-Oct-2011 14:45:39] <Sashness> http://pastebin.com/8ZLw6x2J <— is the event transition i wrote but zenoss doesn't even match it
[18-Oct-2011 14:46:21] <nyeates> reading
[18-Oct-2011 14:47:33] <nyeates> ummmm, without even looking at the transform
[18-Oct-2011 14:47:55] <nyeates> we might consider instead, changing the datasource
[18-Oct-2011 14:48:04] <nyeates> high load....like CPU high load?
[18-Oct-2011 14:49:05] <jmp242> Yea, I was going to ask - why not do that in the graph definition
[18-Oct-2011 14:49:23] <nyeates> RPN maybe...
[18-Oct-2011 14:50:21] <jmp242> exactly
[18-Oct-2011 14:50:34] <nyeates> Sashness: give us url of the datasource that is going over threshold
[18-Oct-2011 14:51:08] <nyeates> jmp242: would RPN change it in the event too tho? cant recall if that is just in the graph, or changes it at the datasource level
[18-Oct-2011 14:57:24] <jmp242> IDK
[18-Oct-2011 14:57:29] <jmp242> that is a concern
[18-Oct-2011 14:57:46] <nyeates> RPN is different in 3.x.... trying to find now
[18-Oct-2011 14:57:54] <jmp242> then it would have to be in the transform
[18-Oct-2011 14:58:27] <Sashness> nyeats: its the default alert for high load in Zenoss
[18-Oct-2011 14:58:43] <Sashness> its just a threshold alert
[18-Oct-2011 14:59:41] <nyeates> jmp242: can you check on a 3.x system this: go into a template on any device class - some standard one - double click on one of the graph definitions (i used Load Average on Device template)
[18-Oct-2011 14:59:56] <nyeates> jmp242: do you get RPN in the pop up?
[18-Oct-2011 15:00:04] <nyeates> I do not....and our docs say i should
[18-Oct-2011 15:00:20] <jmp242> one min
[18-Oct-2011 15:00:23] <jmp242> logging in now
[18-Oct-2011 15:01:10] <nyeates> Hmm, that is a "Graph Definition" not a "Graph Point"....still not seeing RPN in graph point either
[18-Oct-2011 15:01:30] <jmp242> I see one
[18-Oct-2011 15:01:39] <nyeates> Ahhh I found it now....hidden heh
[18-Oct-2011 15:01:40] <jmp242> if you open the graph def with the little arrow thing
[18-Oct-2011 15:01:42] <jmp242> yea
[18-Oct-2011 15:01:46] <jmp242> I was confused for a sec too
[18-Oct-2011 15:02:02] <jmp242> prolly he could try 10,/
[18-Oct-2011 15:02:12] <nyeates> exactly
[18-Oct-2011 15:02:20] <jmp242> but I don't know that changes the actual data either
[18-Oct-2011 15:02:23] <jmp242> but I think it might
[18-Oct-2011 15:02:55] <jmp242> Otherwise, for his transform I'd go about it differently
[18-Oct-2011 15:03:27] <jmp242> well, maybe - I'd have to know more
[18-Oct-2011 15:03:36] <jmp242> but Sashness - can you test using the change to the RPN?
[18-Oct-2011 15:04:00] <Sashness> sorry got distracted. let me read the past msgs
[18-Oct-2011 15:04:32] <Sashness> k so what should i change to the RPN?
[18-Oct-2011 15:05:07] <nyeates> Sashness: go into the template, look for "Graph Definitions" section
[18-Oct-2011 15:05:17] <nyeates> click the gear menu under "Graph Definitions"
[18-Oct-2011 15:05:27] <nyeates> click Manage Graph Points
[18-Oct-2011 15:06:00] <Sashness> yep there
[18-Oct-2011 15:06:12] <nyeates> heh wait...cant find it again
[18-Oct-2011 15:06:57] <nyeates> click on one of the graph points, not a transform
[18-Oct-2011 15:07:03] <nyeates> double click rather
[18-Oct-2011 15:07:09] <nyeates> rpn is in there
[18-Oct-2011 15:07:26] <Sashness> ahh so your saying change it there instead of the transform
[18-Oct-2011 15:07:27] <Sashness> ?
[18-Oct-2011 15:07:45] <nyeates> jmp242: I am starting to think that this wont work at all - cause we are changing how it is shown on the graph, nothing else
[18-Oct-2011 15:08:14] <jmp242> Yea
[18-Oct-2011 15:08:21] <jmp242> I was wondering, we'd have to check the actual RRD
[18-Oct-2011 15:08:26] <jmp242> to see what got created
[18-Oct-2011 15:08:37] <jmp242> I think you are right, this may be the garden path lol
[18-Oct-2011 15:08:48] <jmp242> cause we use transforms to fix filesystem stuff to be readable
[18-Oct-2011 15:09:06] <nyeates> u could make an alias on a datapoint
[18-Oct-2011 15:09:28] <jmp242> I don't know how to do that
[18-Oct-2011 15:09:35] <nyeates> double click the IaLoadInt5 datapoint
[18-Oct-2011 15:09:53] <nyeates> at the bottom it says "Alias:"
[18-Oct-2011 15:10:07] <nyeates> first box is the new name you want, second box is RPN formula
[18-Oct-2011 15:10:52] <nyeates> so do like "IaLoadInt5_DividedBy10" and "10,/"
[18-Oct-2011 15:11:35] <nyeates> oh i just noticed on mine that by default there is already one in there called "loadAverage5min"
[18-Oct-2011 15:11:44] <nyeates> so use that as an example...but make a your own
[18-Oct-2011 15:12:35] <nyeates> Sashness: btw, 10,/ means "Ok, take the datapoint at hand, now take the next variable 10, now divide those two"
[18-Oct-2011 15:12:42] <jmp242> and why would this help Sashness? I'm not getting it
[18-Oct-2011 15:12:55] <jmp242> you think the alias is going to be what gets passed then?
[18-Oct-2011 15:13:07] <nyeates> Welp, now he goes to the Threshold definition
[18-Oct-2011 15:13:15] <jmp242> OOOOOHHH!!!
[18-Oct-2011 15:13:16] <nyeates> and changes it to look at the new variable called IaLoadInt5_DividedBy10
[18-Oct-2011 15:13:19] <jmp242> the lightbulb goes on
[18-Oct-2011 15:13:23] <nyeates> ding
[18-Oct-2011 15:13:56] <jmp242> that sounds awesomenessfull to me (a perfectly cromulent term) for this use case
[18-Oct-2011 15:18:02] <nyeates> hmmm, except that its looking like the new alias doesnt show in the threshold
[18-Oct-2011 15:18:09] <rmatte> nyeates: I'm just reading back here about that high load transform...
[18-Oct-2011 15:18:16] <nyeates> ....jmp242: help on the transform ? :-(
[18-Oct-2011 15:18:18] <rmatte> that's high system load, it's not a percentage of CPU
[18-Oct-2011 15:18:34] <rmatte> basically it's the total number of processes currently using or waiting on the CPU
[18-Oct-2011 15:18:39] <rmatte> so let's say you have 4 cores
[18-Oct-2011 15:18:41] <rmatte> and load is 12
[18-Oct-2011 15:18:45] <jmp242> mmm grumpy
[18-Oct-2011 15:18:51] <jmp242> it sounded so cool to me
[18-Oct-2011 15:19:00] <rmatte> that means that 4 out of the 12 are using the CPU and the other 8 are waiting on it (which is bad)
[18-Oct-2011 15:19:00] <nyeates> me too
[18-Oct-2011 15:19:05] <rmatte> ideally you want a load average under 3
[18-Oct-2011 15:19:43] <nyeates> rmatte: thanks...im not familiar with that
[18-Oct-2011 15:19:59] <nyeates> rmatte: where would i see that? what template
[18-Oct-2011 15:20:02] <rmatte> what he wanted to do was take the raw load value that he was seeing, which was 2323 and divide it by 10 to get the actual load value of 23.23
[18-Oct-2011 15:20:12] <rmatte> 23.23 is aweful
[18-Oct-2011 15:20:13] <rmatte> lol
[18-Oct-2011 15:20:25] <rmatte> awful*
[18-Oct-2011 15:20:52] <jmp242> sure
[18-Oct-2011 15:20:54] <Sashness> lol. if you think thats awful, you should see one of the systems we have running java with custom code thats not that multithreaded
[18-Oct-2011 15:21:01] <Sashness> load avg is over 40-50 at peak
[18-Oct-2011 15:21:13] <rmatte> I don't even understand how a system can function with a load that high
[18-Oct-2011 15:21:15] <rmatte> it's ridiculous
[18-Oct-2011 15:21:31] <rmatte> the highest I get up to is just over 10 on some of mine, but I quickly make adjustments to bring it back down
[18-Oct-2011 15:21:33] <jmp242> so only way is transform
[18-Oct-2011 15:21:37] <jmp242> ?
[18-Oct-2011 15:21:43] <Sashness> the developers decided to have multiple threads but only a single thread actually do work and the rest of threads are just network listeners that feed info to the main thread
[18-Oct-2011 15:21:49] <rmatte> jmp242: for displaying it in the event yes
[18-Oct-2011 15:22:01] <rmatte> jmp242: the RPN expression is already in there by default
[18-Oct-2011 15:22:07] <rmatte> but the event still comes in with the raw value
[18-Oct-2011 15:23:05] <jmp242> Mmm
[18-Oct-2011 15:23:22] <jmp242> Sashness: can you map that event or is it mapped already? So it matches without the searching?
[18-Oct-2011 15:23:59] <Sashness> its mapped already to /Perf/CPU with default zenoss-core install
[18-Oct-2011 15:25:05] <jmp242> mmm
[18-Oct-2011 15:25:17] <jmp242> Well, it gets into python then and I don't know off the top of my head
[18-Oct-2011 15:25:35] <rmatte> give me a second, I've done similar transforms, let me grab one
[18-Oct-2011 15:26:25] <Sashness> k.. interested to see why my transform failed. comparing it should answer that question
[18-Oct-2011 15:27:05] <rmatte> nyeates: just noticed you asked what template... it's the Devices template in /Server/Linux
[18-Oct-2011 15:27:17] <nyeates> which datapoint
[18-Oct-2011 15:27:29] <rmatte> laLoadInt5_laLoadInt5
[18-Oct-2011 15:27:44] <nyeates> oh ok, yeah that is the same datapoint i was giving as example
[18-Oct-2011 15:27:50] <rmatte> k
[18-Oct-2011 15:27:56] <nyeates> i just didnt understand what the datapoint reprsetned. thanks
[18-Oct-2011 15:28:13] <rmatte> np
[18-Oct-2011 15:28:48] <rmatte> looking at that transform that was pastebinned it should would
[18-Oct-2011 15:28:50] <rmatte> work*
[18-Oct-2011 15:28:59] <rmatte> where did you actually place the transform?
[18-Oct-2011 15:30:05] <rmatte> actually I think I know what's wrong, you've created a mapping and you've put the transform in the actual mapping?
[18-Oct-2011 15:31:07] <Sashness> no no.. its already mapped to /Perf/CPU
[18-Oct-2011 15:31:15] <rmatte> directly to the class?
[18-Oct-2011 15:31:15] <Sashness> i added the transform on /Perf/CPU
[18-Oct-2011 15:31:18] <Sashness> yeah
[18-Oct-2011 15:31:20] <rmatte> k
[18-Oct-2011 15:31:33] <rmatte> that's the proper way to do it...
[18-Oct-2011 15:31:42] <rmatte> let me check the transform, gimme a min
[18-Oct-2011 15:35:10] <Sashness> k. thnks
[18-Oct-2011 15:40:29] <rmatte> grrrr, trying to find a transform for reference, though I knew which one it was but apparently not
[18-Oct-2011 15:40:48] <rmatte> few more mins...
[18-Oct-2011 15:44:22] <Sashness> np
[18-Oct-2011 15:48:54] <rmatte> your regex is wrong for starters
[18-Oct-2011 15:48:55] <rmatte> >>> evtsummary = "threshold of high load exceeded: current value 2323"
[18-Oct-2011 15:48:55] <rmatte> >>> m = re.search("threshold of high load exceeded: current value ([\d\.]+)", evtsummary)
[18-Oct-2011 15:48:55] <rmatte> >>> currentload = (float(m.group()[0])) / 10
[18-Oct-2011 15:48:55] <rmatte> Traceback (most recent call last):
[18-Oct-2011 15:48:55] <rmatte> File "<stdin>", line 1, in <module>
[18-Oct-2011 15:48:56] <rmatte> ValueError: could not convert string to float: t
[18-Oct-2011 15:49:07] <rmatte> a little test I did
[18-Oct-2011 15:49:51] <rmatte> so it's picking up "t" instead of the value
[18-Oct-2011 15:51:40] <Sashness> hmm. i modified a transform for interface utilization to get that:
[18-Oct-2011 15:51:41] <Sashness> # Extract the percent and utilization from the summary
[18-Oct-2011 15:51:41] <Sashness> m = re.search("threshold of [^:]+: current value ([\d\.]+)", evt.message)
[18-Oct-2011 15:52:16] <rmatte> where did you get the m.group() from though?
[18-Oct-2011 15:52:24] <rmatte> >>> m.group()
[18-Oct-2011 15:52:25] <rmatte> 'threshold of high load exceeded: current value 2323'
[18-Oct-2011 15:52:25] <rmatte> >>> m.group()[3]
[18-Oct-2011 15:52:25] <rmatte> 'e'
[18-Oct-2011 15:52:25] <rmatte> >>> m.group()[4]
[18-Oct-2011 15:52:25] <rmatte> 's'
[18-Oct-2011 15:52:25] <rmatte> >>> m.group()[5]
[18-Oct-2011 15:52:26] <rmatte> 'h'
[18-Oct-2011 15:52:26] <rmatte> >>> m.group()[10]
[18-Oct-2011 15:52:27] <rmatte> 'o'
[18-Oct-2011 15:52:36] <rmatte> It breaks it up by letter when you do what you're doing
[18-Oct-2011 15:53:06] <rmatte> I've done a transform to pull values before, but I can't find it, was going to use it as quick reference
[18-Oct-2011 15:53:08] <rmatte> time to google
[18-Oct-2011 15:53:32] <Sashness> yeah i was using this for reference: docs/DOC-2554#Interface_High_Utilization_event_enhancement_using_if_then_elif
[18-Oct-2011 15:56:59] <nyeates> later guys, thanks for helping rmatte
[18-Oct-2011 15:57:04] <rmatte> np
[18-Oct-2011 15:57:16] <Sashness> yeah.. thanks a ton!
[18-Oct-2011 16:00:10] <rmatte> k, think I got it
[18-Oct-2011 16:00:16] <rmatte> just give me a sec to write it up
[18-Oct-2011 16:03:45] <rmatte> here: http://fpaste.org/7eN2/raw/
[18-Oct-2011 16:03:54] <rmatte> and you wanted / 100, not /10
[18-Oct-2011 16:05:35] <rmatte> oh crap
[18-Oct-2011 16:05:36] <rmatte> one sec
[18-Oct-2011 16:06:10] <rmatte> http://fpaste.org/wMQ8/raw/
[18-Oct-2011 16:06:13] <rmatte> made a small adjustment
[18-Oct-2011 16:06:24] <rmatte> accidentally left it as evtsummary instead of evt.summary, fixed
[18-Oct-2011 16:06:55] <Sashness> i added the second one you pasted and zenoss has it in red after saving
[18-Oct-2011 16:07:09] <rmatte> oh, let me review it
[18-Oct-2011 16:07:54] <Sashness> nm. was something with the formating
[18-Oct-2011 16:07:55] <rmatte> I see another error I made, one sec
[18-Oct-2011 16:07:59] <Sashness> ohh okay
[18-Oct-2011 16:08:06] <billc> Need help with shrinking a huge 50G zopeDB. Running zeopack.py -d 1 -p 8100 right now. got an error... is "No handlers could be fournd for logger ZEO.zrpc" ok.
[18-Oct-2011 16:09:06] <rmatte> http://fpaste.org/eR4J/raw/
[18-Oct-2011 16:09:21] <rmatte> billc: stack install?
[18-Oct-2011 16:10:04] <rmatte> billc: for reference I run the following command as root to shrink mine...
[18-Oct-2011 16:10:09] <rmatte> /usr/local/zenoss/python/bin/python /usr/local/zenoss/zenoss/bin/zeopack.py -h localhost -p 8100 && rm /usr/local/zenoss/zenoss/var/Data.fs.old
[18-Oct-2011 16:10:40] <billc> 2.x; most likely from the installer rather than via an rpm or zypper (where a SUSE shop)
[18-Oct-2011 16:10:58] <rmatte> what's your $ZENHOME?
[18-Oct-2011 16:11:37] <Sashness> rmatte: so i added it in and removed the old event but the new event came in with the same format
[18-Oct-2011 16:12:13] <billc> /usr/local/zenoss/zenoss
[18-Oct-2011 16:12:17] <rmatte> Sashness: paste the exact summary of the new event
[18-Oct-2011 16:12:42] <rmatte> oh crap
[18-Oct-2011 16:12:43] <billc> Space is supper tight, is there a way to tell it to NOT make a backup before packing.
[18-Oct-2011 16:12:48] <rmatte> I forgot to change something in the transform
[18-Oct-2011 16:12:49] <rmatte> hang on
[18-Oct-2011 16:13:03] ShaneB_ is now known as ShaneB
[18-Oct-2011 16:13:06] <Sashness> threshold of high load exceeded: current value 3124.00
[18-Oct-2011 16:13:12] <rmatte> it's m.group(0), not m.group()[0]
[18-Oct-2011 16:13:19] <Sashness> k.. i'll change it
[18-Oct-2011 16:13:21] <rmatte> but wait
[18-Oct-2011 16:13:28] <rmatte> I don't think it was actually 0, might have been 1
[18-Oct-2011 16:13:30] <rmatte> let me test
[18-Oct-2011 16:13:44] <Sashness> how are you testing? so i can troubleshoot this when it happens again?
[18-Oct-2011 16:14:46] <rmatte> http://fpaste.org/qXse/raw/
[18-Oct-2011 16:14:49] <rmatte> there
[18-Oct-2011 16:14:58] <rmatte> I just type python at the commandline
[18-Oct-2011 16:15:02] <rmatte> then I do something like...
[18-Oct-2011 16:15:17] <rmatte> evtsummary = "threshold of high load exceeded: current value 2323"
[18-Oct-2011 16:15:25] <rmatte> then try the lines out in the transform
[18-Oct-2011 16:15:35] <rmatte> just need to change evt.summary to evtsummary when testing
[18-Oct-2011 16:15:58] <rmatte> python has an interactive shell when you can test stuff, quite handy
[18-Oct-2011 16:16:11] <rmatte> zendmd is basically just a python shell that loads a bunch of Zenoss libraries and connects to zope
[18-Oct-2011 16:16:27] <rmatte> where you can test stuff* rather
[18-Oct-2011 16:16:36] <billc> Other question w/ zeopack... I have zenoss stopped, I assume it doesn't need to be running to shrink the zopeDB?
[18-Oct-2011 16:16:58] <rmatte> billc: I'm not sure about not having it make a backup, if it were to screw up while packing you'd be done
[18-Oct-2011 16:17:14] <rmatte> you need zope running
[18-Oct-2011 16:17:27] <rmatte> all the zenoss daemons can be stopped but zopectl and zeoctl should be running
[18-Oct-2011 16:17:27] <billc> Go it covered by the power of XEN.... (grin), I have a copy of the VM.
[18-Oct-2011 16:19:55] <rmatte> no option to disable the .old file as far as I can see from the docs
[18-Oct-2011 16:20:35] <rmatte> oh wait
[18-Oct-2011 16:20:38] <rmatte> maybe there is...
[18-Oct-2011 16:20:57] <rmatte> pack-keep-old
[18-Oct-2011 16:20:58] <rmatte> Can be set to false to disable the creation of *.fs.old files before the pack is run. Defaults to true.
[18-Oct-2011 16:21:11] <rmatte> let me find where that needs to go
[18-Oct-2011 16:22:18] <billc> where is the doc... that your checking
[18-Oct-2011 16:22:27] <rmatte> http://pypi.python.org/pypi/plone.recipe.zeoserver
[18-Oct-2011 16:22:29] <rmatte> this one right now
[18-Oct-2011 16:24:04] <rmatte> ahhh
[18-Oct-2011 16:24:05] <rmatte> The pack process will still create a new file during packing, so you do need to have enough free disk space to hold
[18-Oct-2011 16:24:11] <rmatte> no way around it
[18-Oct-2011 16:24:25] <Sashness> rmatte: i'm sorry to report it still didn't work
[18-Oct-2011 16:24:27] <rmatte> all that pack-keep-old does is removes the file automatically after packing is complete
[18-Oct-2011 16:24:42] <rmatte> Sashness: I really don't see how that's possible at this point
[18-Oct-2011 16:25:05] <Sashness> yeah me either
[18-Oct-2011 16:25:12] <Sashness> its okay, i'm going to play around with it with zendmd
[18-Oct-2011 16:25:15] <Sashness> thanks for all your help
[18-Oct-2011 16:25:20] <Sashness> i'll let you know if i find anything
[18-Oct-2011 16:25:24] <rmatte> k. np
[18-Oct-2011 16:26:33] <rmatte> you used the one from this link right?: http://fpaste.org/qXse/raw/
[18-Oct-2011 16:27:26] <rmatte> oh damnit
[18-Oct-2011 16:27:33] <rmatte> when I revised it I left it as evtsummary again
[18-Oct-2011 16:27:36] <rmatte> that's probably why
[18-Oct-2011 16:28:05] <rmatte> Sashness: http://fpaste.org/uar7/raw/
[18-Oct-2011 16:28:21] <rmatte> make sure that on the m = line you have evt.summary at the end, not evtsummary
[18-Oct-2011 16:28:26] <rmatte> I accidentally left it
[18-Oct-2011 16:29:00] <Sashness> haha. that worked perfectly!
[18-Oct-2011 16:29:00] <rmatte> any error in execution will cause the transform to be ignored
[18-Oct-2011 16:29:04] <rmatte> good
[18-Oct-2011 16:29:09] <Sashness> yeah did a diff against it
[18-Oct-2011 16:29:19] <Sashness> thank you soo much. i actually learned how this stuff works now too
[18-Oct-2011 16:29:24] <rmatte> no problem
[18-Oct-2011 16:29:30] <Sashness> and with that zendmd tip i'll be able to troubleshoot it myself now
[18-Oct-2011 16:29:50] <rmatte> that's actually a transform that I've been meaning to implement on my servers but that I'd been too lazy to write
[18-Oct-2011 16:29:52] <rmatte> lol
[18-Oct-2011 16:30:08] <rmatte> going to toss it on them right now
[18-Oct-2011 16:32:27] <jmp242> rmatte: want to post it somewhere too? What is this doing - I've lost the thread
[18-Oct-2011 16:34:51] <rmatte> when you poll a load average for a unix box via SNMP it returns the load average in a raw non decimal format
[18-Oct-2011 16:35:04] <rmatte> i.e 1013 instead of 10.13
[18-Oct-2011 16:35:14] <rmatte> so to display the actual load average you need to divide it by 100
[18-Oct-2011 16:35:19] <jmp242> I see
[18-Oct-2011 16:35:21] <jmp242> great
[18-Oct-2011 16:35:31] <jmp242> It seems like these sorts of things ought to be OOTB lol
[18-Oct-2011 16:35:56] <rmatte> yeh, especially since they put a decent amount of time in to the stock linux template but then sort of left it half finished
[18-Oct-2011 16:36:11] <jmp242> Can I put this on the zca wiki?
[18-Oct-2011 16:36:16] <rmatte> go for it
[18-Oct-2011 16:36:31] <jmp242> it should go in which event class?
[18-Oct-2011 16:36:31] <rmatte> anything I post in here is up for grabs
[18-Oct-2011 16:36:50] <rmatte> /Perf/CPU is the default event class for the alerts
[18-Oct-2011 16:36:52] <rmatte> so that'd be the one
[18-Oct-2011 16:38:38] <jmp242> https://zcaportal.org/wiki/bin/view/ZCA/ZCABestPractices#Useful_Changes_to_Out_Of_The_Box_Zenoss_config
[18-Oct-2011 16:38:47] <jmp242> see what you think - feel free to add more or suggest more
[18-Oct-2011 16:46:01] <rmatte> looks good
[18-Oct-2011 16:47:55] <rmatte> jmp242: you might be interested in the second half that I recently added on to this: docs/DOC-2494
[18-Oct-2011 16:48:05] <rmatte> "Setting up Active Monitoring of Interface Last Change"
[18-Oct-2011 16:59:19] <rmatte> hmmm, I can't for the life of me get the newer zope ldap packages to install properly
[18-Oct-2011 16:59:30] <rmatte> tons of errors then gcc craps out
[18-Oct-2011 17:00:46] <rmatte> ah, it's crapping out when installing python-ldap under zenoss
[18-Oct-2011 17:01:15] <billc> I have the zeopack running for an hour. How can I tell if it is doing anything. Disc space not changing.
[18-Oct-2011 17:02:47] <rmatte> billc: if it hasn't crashed yet then it's doing something, though I thought you said you didn't have enough disk space to make the .old?
[18-Oct-2011 17:04:47] <jb> hrm ok
[18-Oct-2011 17:04:49] <jb> have a situation
[18-Oct-2011 17:05:09] <jb> i have lots of devices that use the same performance template, but some that need to have different thresholds set
[18-Oct-2011 17:05:11] <jb> is that even possible?
[18-Oct-2011 17:05:38] <billc> Might make it... I am killing megs. Worst case, I will mount a big disk and get it done.
[18-Oct-2011 17:06:03] <billc> I don't see it creating an "old" file.
[18-Oct-2011 17:06:52] <billc> Expect it to be in /usr/local/zenos/zenos/var
[18-Oct-2011 17:07:35] <Sashness> are mac addresses stored in the zencatalog?
[18-Oct-2011 17:07:39] <Sashness> how do i see what's stored there?
[18-Oct-2011 17:19:25] <rmatte> YES!
[18-Oct-2011 17:19:32] <rmatte> I finally got these damn plugins to install
[18-Oct-2011 17:20:18] <rmatte> well, one of them
[18-Oct-2011 17:21:00] <rmatte> aha, there we go
[18-Oct-2011 17:21:18] <rmatte> I'm going to write up a guide for this after I'm done because it's a huge pain in the ass
[18-Oct-2011 17:27:47] kocolosk is now known as kocolosk|away
[18-Oct-2011 17:28:39] <rmatte> bingo, that worked
[18-Oct-2011 17:50:54] <rmatte> huzzah! just got the LDAP plugin working with 3.2.1 beta
[18-Oct-2011 18:05:19] <humanix> bye rmatte Hackman238
[18-Oct-2011 18:05:21] <humanix> bye all
[18-Oct-2011 18:05:24] <humanix> going home
[18-Oct-2011 18:05:25] <humanix>
[18-Oct-2011 18:07:32] <rmatte> later
[18-Oct-2011 18:07:37] <rmatte> same here, heading out, later all
[18-Oct-2011 20:05:50] kocolosk|away is now known as kocolosk
[18-Oct-2011 21:32:13] <billc> .
[18-Oct-2011 23:23:53] kocolosk is now known as kocolosk|away
[19-Oct-2011 00:00:01] [disconnected at Wed Oct 19 00:00:01 2011]
[19-Oct-2011 00:00:02] [connected at Wed Oct 19 00:00:02 2011]
[19-Oct-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[19-Oct-2011 03:44:16] <ali3n0> hi folks
[19-Oct-2011 03:45:01] <ali3n0> what if I want to keep collected data after an ec2 instance termination?
[19-Oct-2011 06:14:30] <enkrypt> hello! Hackman238 et al, did you remember my problem from yesterday? I keep getting backtraces in zenhub.log (http://dpaste.com/637268/) and I get a few dozen heartbeat failures from my daemons each hour, on several remote collectors. my workListLength is growing like crazy, and my zenhub.log keeps telling me that all workers are busy. any idea what is going on? I've got a support ticket running, I'm on enterprise 3.2.0. the system is becoming unusable, my
[19-Oct-2011 06:15:13] <enkrypt> I performed a collector update on all remote collectors. I was recommended by Jake Lampe to clean all .pickle and config-db files, so I did that, and restarting zenoss. but didn't help.
[19-Oct-2011 06:25:47] <ali3n0> hi folks, is it possible to run ssh commands when a certain event occours?
[19-Oct-2011 06:49:14] <ali3n0> apologies,didn't see chapter 7.1.12.1
[19-Oct-2011 06:50:01] <ali3n0> anyway, I'm puzzled: what about collected data after a device is not monitored anymore? is data still there somewhere in the rrd dbs or does it get removed?
[19-Oct-2011 06:56:07] enkrypt_ is now known as enkrypt
[19-Oct-2011 07:15:26] <enkrypt> what is the correct way to add two or more times the same template to an asset? for example if I need to have 2 url's monitored for a specific machine, I want to add twice the HttpMonitor template, each time with different parameters. the documentation about that doesn't seem to be correct?
[19-Oct-2011 07:59:45] <enkrypt> same question for Oracle enterprise zenpack: the procedure listed at docs/DOC-7864 for monitoring multiple SID's simply doesn't work
[19-Oct-2011 08:07:58] <jmp242> mmm
[19-Oct-2011 08:08:09] <jmp242> I don't think you can just add the same template twice
[19-Oct-2011 08:08:24] <jmp242> For the HTTP one, you could copy it and create 2 templates perhaps
[19-Oct-2011 08:09:00] <enkrypt> jmp242: the problem is I'm in a migration period, and my legacy system has like thousands of HTTP URL check's
[19-Oct-2011 08:09:45] <enkrypt> the same for Oracle, I've got oracle servers that have dozens of instances running, I need a flexible way to perform oracle monitoring on each instance, but the documentation about that doesn't work on zenoss enterprise 3.2.0
[19-Oct-2011 08:23:43] <wayne___> enkrypt: I've struggled with that same issue with similar cases (individual switches in a stack, etc.). I think the cleanest solution is to define a new component type and modeler (like how interfaces are handled) - but I haven't figured out how to do that yet
[19-Oct-2011 08:25:36] <wayne___> enkrypt: the zenpack howto for 2.x had an example using something bridging related (it's been a while so I don't remember exactly what) - but it required hacking up page templates in order to work. I'm hoping that isn't required in 3.x, but I haven't read the updated howto yet
[19-Oct-2011 08:29:48] <jmp242> Those sorts of needs probably ought to be part of a bigger programming effort - I doubt it's plug and play
[19-Oct-2011 08:37:07] <enkrypt> damn that is... depressing
[19-Oct-2011 08:37:20] <enkrypt> for now I'm solving this for HttpMonitor template by adding extra datasources
[19-Oct-2011 08:37:45] <enkrypt> but this means that the Oracle enterprise zenpack documentation is wrong, you can't easily monitor an extra instance
[19-Oct-2011 08:38:06] <enkrypt> also, what is the recommended way to set a low memory threshold for linux systems?
[19-Oct-2011 08:49:26] <Hackman238> Morning all!
[19-Oct-2011 08:49:43] <Hackman238> enkrypt: Whats depressing?
[19-Oct-2011 08:50:05] <rocket> enkrypt: you can monitor more than one instance pretty easily. You copy the template and then change variables inside the copied template to point to the new instance
[19-Oct-2011 08:50:21] <Hackman238> Morning rocket
[19-Oct-2011 08:51:16] <rocket> Good Morning
[19-Oct-2011 09:01:40] <enkrypt> Hackman238: I'm looking at a gargantuan task of implementing a few hundred HTTP url monitoring checks
[19-Oct-2011 09:01:50] <enkrypt> sometimes a simple url check for code 200, sometimes doing a regex etc
[19-Oct-2011 09:02:17] <enkrypt> also I started implementing the Oracle enterprise zenpack, which works fine in itself, but I need to monitor dozens of oracle instances on each oracle server
[19-Oct-2011 09:02:26] <enkrypt> and the documentation for that simply doesn't work as described
[19-Oct-2011 09:02:54] <enkrypt> for the httpmonitor story I could in theory literally COPY the template for each URL, but that seems awfully unproductive?
[19-Oct-2011 09:03:21] <enkrypt> right now I'm planning to simply add a new datasource for each URL to the locally copied template
[19-Oct-2011 09:05:28] <enkrypt> Hackman238: I've also opened a support ticket for the issue I'm having with the CRITICAL error backtrace in my zenhub.log, talking about exceptions.AttributeError: 'NoneType' object has no attribute 'monitorDevice'
[19-Oct-2011 09:05:52] <enkrypt> Hackman238: and since a few days I'm having big troubles with heartbeat failures, like hundreds per day, causing big gaps in my graphs
[19-Oct-2011 09:06:19] <enkrypt> but apparently those heartbeat failures are caused by zensyslog not being able to handle my syslog traffic, and using up all zenhub workers
[19-Oct-2011 09:08:51] <enkrypt> I temporarily disabled zensyslog (well, I moved it to a different port) and now I get much less heartbeat failures
[19-Oct-2011 09:09:22] <enkrypt> I'm in the process of installing rsyslog on my servers to listen on 514, do some filtering, and then pass a subset of the syslog traffic onto zensyslog which I've configured to listen on 10514
[19-Oct-2011 09:09:26] <enkrypt> I hope this will help
[19-Oct-2011 09:09:40] <enkrypt> but still the errors and tracebacks I keep seeing in zenhub.log worry me
[19-Oct-2011 09:10:24] <enkrypt> and the fact that any syslog sending server can easily take down my (rather beefy) zenoss installation with a few thousand syslog triggers per minute is a bit scary
[19-Oct-2011 09:10:42] <Hackman238> enkrypt: Can you post your zenhub.log to pastebin?
[19-Oct-2011 09:11:25] <Hackman238> enkrypt: Yeah the oracle monitor I've had less than complete success with. I'm afraid I can'
[19-Oct-2011 09:11:30] <Hackman238> t help with that much
[19-Oct-2011 09:11:40] <Hackman238> enkrypt: Which httpmonitor pack are you using?
[19-Oct-2011 09:11:58] <enkrypt> Hackman238: not sure, just the regular one? I'm using 3.2.0 enterprise, the template is called HttpMonitor
[19-Oct-2011 09:13:08] <Hackman238> enkrypt: Gotcha. Let me look at that real quick.
[19-Oct-2011 09:14:37] <ali3n0> anyone here using zenoss + amazon rds ?
[19-Oct-2011 09:16:17] <jb> mornin.
[19-Oct-2011 09:27:12] kocolosk|away is now known as kocolosk
[19-Oct-2011 09:28:11] <Hackman238> enkrypt: Looks like the best way to do those datapoints is by copying them or using a dmd script.
[19-Oct-2011 09:28:48] <Hackman238> enkrypt: Sicne you ahve enterprise I might suggest you put in a ticket to have support craft you some dmd magic to do this, if its possible to do by dmd.
[19-Oct-2011 09:35:11] <bseklecki> enkrypt: oracle provides a whole free fancy web interface thing for that
[19-Oct-2011 09:35:19] <bseklecki> monitoring instance performance
[19-Oct-2011 09:35:44] <Hackman238> bseklecki: Aye, but it is nice to have the metrics in Zenoss as well.
[19-Oct-2011 09:36:05] <Hackman238> bseklecki: Fewer tools = happier and more effective staff
[19-Oct-2011 09:36:13] <bseklecki> if memory serves, there's an Oracle SNMP agent\
[19-Oct-2011 09:36:17] <bseklecki> or AgentX sub-agent
[19-Oct-2011 09:36:41] <bseklecki> because for years I was trying to get the PostgreSQL people to cook one up
[19-Oct-2011 09:44:39] <enkrypt> it's a shame because the oracle zenpack seems to work OK
[19-Oct-2011 09:44:49] <enkrypt> but it's practically useless if I can only monitor 1 oracle instance per server
[19-Oct-2011 09:57:45] <rocket> enkrypt: you should be able to copy the templates for the oracle monitor
[19-Oct-2011 10:00:54] <enkrypt> rocket: how/where would I do that in the GUI ? I guess it's not the same as "override template here" and creating a local copy?
[19-Oct-2011 10:03:08] <bseklecki> so what, you gotta hack the python to make it aware of multiple instances?
[19-Oct-2011 10:07:03] <bseklecki> the discovery plugin that is
[19-Oct-2011 10:19:26] <jb> question.. I have several devices that share the same performance type, but reside in different groups/systems/etc. I need to be able to set different thresholds on various data sources of devices that share the same performance template.. any idea if this is possible?
[19-Oct-2011 10:19:47] <jb> for example, I have a perf template for "Motorola Canopy Access Points"
[19-Oct-2011 10:20:09] <jb> but depending if the AP is a 900mhz or 2.4GHZ ap, I need to set different thresholds on various data sources..
[19-Oct-2011 10:21:08] <billc> Need help recovering a disc space. Both mysql and zopeDB are huge (50G). attempt to clear mysql has failed. Tried dropping/creating the events database but after, i can not connect to mysql. yes I did the grant.
[19-Oct-2011 10:21:37] lederka is now known as bender183
[19-Oct-2011 10:21:48] bender183 is now known as bender[a]
[19-Oct-2011 10:22:35] <billc> The mysql instance on port 3307 (zenoss's) doesn't have a mysql.user table.
[19-Oct-2011 10:24:03] <Hackman238> billc: su zenoss and $ZENHOME/bin/zeopack -h localhost -p 8100
[19-Oct-2011 10:24:05] <rocket> enkrypt: go to the advanced page and find the main template you wish to copy
[19-Oct-2011 10:25:01] <Hackman238> billc: That will pack your ZODB. in $ZENHOME/var/ you'll find Data.fs and Data.fs.old. Once you zeopack once, compare the files. The Data.fs should be much smaller. Set the Data.fs.old aside or delete it. Zeopack once more to recreate it.
[19-Oct-2011 10:25:15] <jb> .. or is the only way to do this is create a duplicate performance template with different threshold definitions?
[19-Oct-2011 10:25:35] <Hackman238> billc: as for mysql, did you drop only events or also the schema?
[19-Oct-2011 10:26:27] <billc> drop database events.... run the create scripts.
[19-Oct-2011 10:27:03] <billc> System has two instances of mysql running. 3306 is the classic and zenoss is on 3307.
[19-Oct-2011 10:27:33] <billc> drop database events; create database events; use events; .... run the create scripts.
[19-Oct-2011 10:27:47] <billc> did this as user zenoss.
[19-Oct-2011 10:29:01] <billc> I do not have enough room on the server to run the zenpack. Need to clear mysql first.
[19-Oct-2011 10:29:41] <Hackman238> billc: in that case delete $ZENHOME/var/Data.fs.old. Please please be sure you have a backup if this is an important instance
[19-Oct-2011 10:30:08] <Hackman238> billc: are you using the stack install?
[19-Oct-2011 10:30:15] <Hackman238> billc: (zenoss stack)
[19-Oct-2011 10:30:44] <enkrypt> rocket & Hackman238: I am getting lots of heartbeat failures from my remote collectors again. in the graphs I see my zenhub workListLength increasing rapidly to 100+. my zenhub.log shows I'm running out of workers. how can I discover what is using so many workers?
[19-Oct-2011 10:31:25] <rocket> enkrypt: what version of zenoss again?
[19-Oct-2011 10:31:33] <Hackman238> rocket: v3.2 right?
[19-Oct-2011 10:31:48] <Hackman238> enkrypt: v3.2 right?
[19-Oct-2011 10:31:57] <Hackman238> rocket: Sorry
[19-Oct-2011 10:33:21] <rocket> enkrypt: go to advanced -> monitoring templates
[19-Oct-2011 10:33:29] <rocket> enkrypt: find the template you wish to copy
[19-Oct-2011 10:33:50] <rocket> enkrypt: choose the gear copy/Override template option
[19-Oct-2011 10:34:02] <enkrypt> Enterprise v3.2.0 + 3 patches
[19-Oct-2011 10:34:04] <rocket> enkrypt: choose the same organizer to copy it to
[19-Oct-2011 10:34:20] <rocket> enkrypt: then you should see "copy_of_<template>"
[19-Oct-2011 10:34:30] <enkrypt> rocket: ah ok chosing the same organizer, didn't do that last time. ok will keep that in mind
[19-Oct-2011 10:34:52] <rocket> you can move it once its been copied at the point
[19-Oct-2011 10:35:02] <enkrypt> ok I'll do that for now
[19-Oct-2011 10:35:32] <Hackman238> Off topic, but important- http://www.baconaustin.com/
[19-Oct-2011 10:35:36] <enkrypt> about my problem with heartbeat failures and zenhub workers getting exhausted, any easy way of finding out what causes this? I've got a zenoss support ticket open, but it's kind of urgent, getting worse each day
[19-Oct-2011 10:37:30] <rocket> enkrypt: how many collectors per hub are you running?
[19-Oct-2011 10:37:57] <enkrypt> rocket: not a lot, 3
[19-Oct-2011 10:38:37] <enkrypt> rocket: thing is, I didn't use to have this problem. I thought it was related to zensyslog getting too much traffic, but I stopped that (changed the port) and it improved for some hours, but now I'm back to the same mess, heartbeat failures all over the place
[19-Oct-2011 11:19:54] <humanix> hi there
[19-Oct-2011 11:19:55] <humanix> hi Hackman238 rmatte
[19-Oct-2011 11:20:04] <humanix> how are you guys?
[19-Oct-2011 11:20:48] <Hackman238> humanix: Hey. Well, and yourself?
[19-Oct-2011 11:21:26] <humanix> fine, thanks guys
[19-Oct-2011 11:21:47] <Hackman238> humanix: Glad to hear.
[19-Oct-2011 11:28:12] <humanix> working
[19-Oct-2011 11:28:21] <humanix> improving our f5 templates
[19-Oct-2011 11:28:35] <humanix> xD
[19-Oct-2011 11:30:48] <Hackman238> humanix: Nice
[19-Oct-2011 11:34:19] <humanix> trying to learn about how tmm work and what do i look
[19-Oct-2011 11:34:37] <humanix> cause we're taking bad memory samples
[19-Oct-2011 11:34:42] <humanix> cause of tmm
[19-Oct-2011 11:44:57] <Hackman238> humanix: ttm?
[19-Oct-2011 11:46:08] <fragfutter> i would guess f5 bigip
[19-Oct-2011 11:47:20] <fragfutter> traffic management microkernel
[19-Oct-2011 11:47:45] <Hackman238> fragfutter: Oh gotcha
[19-Oct-2011 11:47:53] <Hackman238> humanix: Good luck.
[19-Oct-2011 11:53:43] <jmp242> bbl
[19-Oct-2011 11:55:52] <jmp242> Opera 11.52 now installed
[19-Oct-2011 11:59:42] <Hackman238> jmp242: How's it work?
[19-Oct-2011 11:59:49] <jmp242> Same as 11.51
[19-Oct-2011 11:59:52] <jmp242> but a security patch
[19-Oct-2011 11:59:57] <Hackman238> jmp242: Ah gotcha
[19-Oct-2011 12:00:01] [disconnected at Wed Oct 19 12:00:01 2011]
[19-Oct-2011 12:00:02] [connected at Wed Oct 19 12:00:02 2011]
[19-Oct-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[19-Oct-2011 12:00:19] <jmp242> Whole thing about a 0day
[19-Oct-2011 12:00:20] <jmp242> http://my.opera.com/securitygroup/blog/2011/10/19/about-the-svg-font-manipulation-vulnerability-that-was-fixed-in-11-52
[19-Oct-2011 12:02:43] <Hackman238> jmp242: Ah gotcha
[19-Oct-2011 12:02:49] <Hackman238> jmp242: Not a serious exploit.
[19-Oct-2011 12:07:48] <jmp242> Well, might be IDK
[19-Oct-2011 12:07:56] <jmp242> but better to be safe than sorry
[19-Oct-2011 12:08:02] <Hackman238> jmp242: True true
[19-Oct-2011 12:11:11] fragfutter is now known as fragfutter1
[19-Oct-2011 12:11:40] fragfutter1 is now known as fragfutter
[19-Oct-2011 12:34:24] <kyle__> Yesterday I asked for help removing the monitoring of a service from a device, suggesting I had to push changes after removing the service. Didn't work. Zenoss still threw an error overnight, because the service (which I don't want to monitor) isn't there.
[19-Oct-2011 12:42:52] <kyle__> so, any other ideas how to keep zenoss 3 from monitoring services I don't want it to monitor?
[19-Oct-2011 12:44:03] <Hackman238> kyle__: Try stopping zenhub, doing a ps aux | grep zenhub to be sure it stopped, starting zenhub and repushing the changes
[19-Oct-2011 12:44:28] <kyle__> Should that be necessary with single-server installs of zenoss?
[19-Oct-2011 12:45:12] <Hackman238> kyle__: Not generally.
[19-Oct-2011 12:45:18] <Hackman238> kyle__: What version are you running?
[19-Oct-2011 12:46:04] <kyle__> Three. Err, found it. 3.2.0
[19-Oct-2011 12:46:38] <Hackman238> kyle__: Generally shouldn't be needed. Not to say it can't happen.
[19-Oct-2011 12:46:52] <kyle__> Gotcha.
[19-Oct-2011 12:47:59] <kyle__> The problem with the problem i'm having is, it can take 12+ hours for the symptoms to occur again. Makes quick troubleshooting a PITA
[19-Oct-2011 12:48:55] <Hackman238> kyle__: Are the processes being readded?
[19-Oct-2011 12:49:01] * kyle__ nods
[19-Oct-2011 12:49:12] <kyle__> That's exactly the problem.
[19-Oct-2011 12:49:30] <Hackman238> kyle__: Gotcha. You'll want to model the device, find the processes being monitored and set them to no monitored.
[19-Oct-2011 12:49:45] <Hackman238> kyle__: That way they aren't removed, but zenoss knows to ignore their status
[19-Oct-2011 12:50:08] <kyle__> Ahh. For some reason I remember just removing them did that in older versions. Thanks!
[19-Oct-2011 12:50:14] <Hackman238> kyle__: No problem.
[19-Oct-2011 12:50:26] <Hackman238> kyle__: another way would be to remove them but to lock the device from updates.
[19-Oct-2011 12:51:10] <kyle__> So the master profile or linux servers was updating this instance, pushing in what was missing, because it wasn't locked?
[19-Oct-2011 12:51:13] <Hackman238> kyle__: Also you can force a model by going to the device status screen and picking the gear icon in lower left -> model device
[19-Oct-2011 12:51:28] <Hackman238> kyle__: Correct.
[19-Oct-2011 12:51:39] <kyle__> This makes more sense now.
[19-Oct-2011 12:51:45] <Hackman238> kyle__: Well zenmodeler 'finds' the processes and add them back in.
[19-Oct-2011 12:52:50] <Hackman238> kyle__: So if its locked, nothing will be added or removed. If they are set to not monitored, items can be added/removed by the modeller, but Zenoss tells the daemon responsible for testing that object to ignore it.
[19-Oct-2011 13:07:44] <humanix> right Hackman238
[19-Oct-2011 13:07:48] <humanix> i was on lunchtime
[19-Oct-2011 13:07:54] <humanix> traffic management microkernel
[19-Oct-2011 13:14:49] <Hackman238> humanix: Np. Gotcha.
[19-Oct-2011 13:15:12] <kenhen93> hi, can anyone help me trouble shoot why my command will not run
[19-Oct-2011 13:15:20] <kenhen93> i run it manually via zencommand run
[19-Oct-2011 13:15:22] <kenhen93> but it will not run
[19-Oct-2011 13:17:10] <kenhen93> please help me :-), I am trying to run the command via ssh
[19-Oct-2011 13:17:16] <videoconferencin> Is there an agent that can be installed on machines that do not have public IP addresses so that they can push their stats to a server?
[19-Oct-2011 13:17:38] <Hackman238> kenhen93: sure, whats the commadn your using in the datasource?
[19-Oct-2011 13:17:47] <Hackman238> videoconferencin: Negative.
[19-Oct-2011 13:17:53] <videoconferencin> Ouch
[19-Oct-2011 13:17:54] <rmatte> kenhen93: did you actually configure your zproperties properly? and did you set the command datasource type to ssh?
[19-Oct-2011 13:18:20] <kenhen93> i am trying to run a pearl script remotely with a config file in the synatx
[19-Oct-2011 13:18:31] <kenhen93> yes, i checked the box for SSH
[19-Oct-2011 13:18:33] <Hackman238> videoconferencin: Why do you need push?
[19-Oct-2011 13:18:39] <rmatte> kenhen93: are you prefixing the command with perl?
[19-Oct-2011 13:18:40] <kenhen93> i added a username and password in zproperties
[19-Oct-2011 13:18:48] <rmatte> i.e: perl script.pl blah.conf
[19-Oct-2011 13:19:19] <kenhen93> just /dir/dir/pearl.pl
[19-Oct-2011 13:19:23] <videoconferencin> I need my monitoring server to be in a different datacenter than my servers, and 99% of my servers do not have public IP addresses… and those that do are mapped IPs, not dedicated to the machine. Terremark dedicated cloud.
[19-Oct-2011 13:19:24] <kenhen93> it works when i run it from the command line
[19-Oct-2011 13:19:36] <kenhen93> if i run the command via ssh from the zenoss server
[19-Oct-2011 13:19:43] <kenhen93> it returns the right values
[19-Oct-2011 13:19:51] <rmatte> kenhen93: is there a valid shebang at the start of the script?
[19-Oct-2011 13:19:57] <rmatte> i.e: #!/bin/perl
[19-Oct-2011 13:19:58] <rmatte> or whatever
[19-Oct-2011 13:20:14] <rmatte> because if there isn't, and you're trying to just execute the script directly, that's most likely the issue
[19-Oct-2011 13:20:27] <Hackman238> videoconferencin: Ah. You don't have a vpn in?
[19-Oct-2011 13:20:38] <kenhen93> yes
[19-Oct-2011 13:20:39] <kenhen93> #!/usr/bin/perl -w
[19-Oct-2011 13:20:40] <rmatte> also, this may be a stupid question, but you did chmod +x the script correct?
[19-Oct-2011 13:20:47] <kenhen93> yes
[19-Oct-2011 13:20:56] <kenhen93> it does excute
[19-Oct-2011 13:21:01] <kenhen93> i can run it remotely and locally
[19-Oct-2011 13:21:01] <rmatte> what output do you see when doing: zencommand run -v10 -d <devicename>
[19-Oct-2011 13:21:02] <rmatte> ?
[19-Oct-2011 13:21:05] <kenhen93> i made a parser
[19-Oct-2011 13:21:08] <videoconferencin> Hackman238: I could do that, but there is a cost associated with that which I do not want to incur
[19-Oct-2011 13:21:11] <kenhen93> and that seems to made it stop working almost
[19-Oct-2011 13:21:20] <rmatte> It should tell you what's wrong when you run that command
[19-Oct-2011 13:21:24] <rmatte> since -v10 enables debug output
[19-Oct-2011 13:21:27] <Hackman238> videoconferencin: In the case of OpenStack and EC2, one can use the provided API in conjunction with a Zenpack to model and monitor otherwise unaccessible devices.
[19-Oct-2011 13:21:31] <kenhen93> nothing really
[19-Oct-2011 13:21:43] <rmatte> kenhen93: that's impossible, it has to output some error
[19-Oct-2011 13:21:52] <rmatte> read through the output carefully
[19-Oct-2011 13:22:01] <rmatte> and do -d <devicename>
[19-Oct-2011 13:22:02] <kenhen93> 2011-10-19 13:21:54,926 DEBUG zen.zencommand: Starting PBDaemon initialization 2011-10-19 13:21:54,930 INFO zen.zencommand: Connecting to localhost:8789 2011-10-19 13:21:54,931 DEBUG zen.zencommand: Logging in as admin
[19-Oct-2011 13:22:05] <rmatte> so that it only runs for the one device
[19-Oct-2011 13:22:17] <kenhen93> 2011-10-19 13:21:54,938 INFO zen.zencommand: Connected to ZenHub 2011-10-19 13:21:54,939 DEBUG zen.zencommand: Setting up initial services: EventService, CommandConfig 2011-10-19 13:21:54,939 DEBUG zen.zencommand: Chaining getInitialServices with d2
[19-Oct-2011 13:22:26] <rmatte> those lines have nothing to do with anything
[19-Oct-2011 13:22:27] <videoconferencin> Hackman238: We have some very specific requirements (i.e. PCI / HIPPA / etc) that require a specific type of infrastructure
[19-Oct-2011 13:22:28] <kenhen93> 2011-10-19 13:21:54,941 DEBUG zen.zencommand: Loaded service EventService from zenhub 2011-10-19 13:21:54,941 DEBUG zen.zencommand: Loaded service CommandConfig from zenhub
[19-Oct-2011 13:22:32] <rmatte> k stop
[19-Oct-2011 13:22:39] <rmatte> just pate the output to fpaste.org
[19-Oct-2011 13:22:42] <rmatte> paste*
[19-Oct-2011 13:22:44] <kenhen93> sorry, i was trying to paste everything
[19-Oct-2011 13:22:47] <rmatte> the whole output
[19-Oct-2011 13:22:52] <kenhen93> ok
[19-Oct-2011 13:22:56] <rmatte> yeh don't paste in a channel, disruptive
[19-Oct-2011 13:23:00] <rmatte> fpate.org or pastebin.com
[19-Oct-2011 13:23:04] <rmatte> fpaste.org*
[19-Oct-2011 13:23:12] <Hackman238> videoconferencin: I can't really think of any product that has real push for performance data. You could setup your collector in the datacenter then open a handful of ports back to the zenoss master.
[19-Oct-2011 13:23:22] <rmatte> but honestly, if you'd just give it a careful read I'm sure you'd spot the error
[19-Oct-2011 13:23:46] <rmatte> also, if that's output from anything other than: zencommand run -v10 -d devicename
[19-Oct-2011 13:23:47] <kenhen93> http://fpaste.org/DjfB/
[19-Oct-2011 13:23:49] <rmatte> then it's useless
[19-Oct-2011 13:23:58] <videoconferencin> Hackman238: Interesting… set up a collector inside the datacenter and have it push to the outside server.
[19-Oct-2011 13:24:30] <rmatte> can you put the -v10 before the -d?
[19-Oct-2011 13:24:44] <rmatte> I think putting it at the end is making it not recognize it
[19-Oct-2011 13:24:48] <rmatte> because that's too little output
[19-Oct-2011 13:24:52] <Hackman238> videoconferencin: Well, it doesnt push the perf data. That remains on the collector in RRD's. Event data and object data bidirectionally move between the master and collectors. There isn't a push.
[19-Oct-2011 13:25:15] <rmatte> run the command exactly like I said to run it: zencommand run -v10 -d devicename
[19-Oct-2011 13:25:26] <rmatte> then re-paste the output
[19-Oct-2011 13:25:31] <videoconferencin> Hackman238: Gotcha. Thanks. I may re-investigate the VPN route.
[19-Oct-2011 13:25:48] <kenhen93> http://fpaste.org/zD2e/
[19-Oct-2011 13:26:07] <kenhen93> oops
[19-Oct-2011 13:26:08] <kenhen93> sorry
[19-Oct-2011 13:26:42] <rmatte> kenhen93: you're sure that you actually have the template that you created bound to the device?
[19-Oct-2011 13:26:54] <rmatte> because based on the output either the template isn't bound, or the datasource is disabled
[19-Oct-2011 13:27:01] <kenhen93> yes it is locally defined
[19-Oct-2011 13:27:08] <Hackman238> videoconferencin: I'm not sure what you mean by specific infrastructure. We're PCI and HIPPA compliant here. I guess I'm unsure how that's relevant to the location of Zenoss.
[19-Oct-2011 13:27:18] <kenhen93> enabled is checked
[19-Oct-2011 13:27:46] <rmatte> kenhen93: oh wait a second, are you running version 3.2?
[19-Oct-2011 13:28:03] <rmatte> who removed the 3.2 patches from the topic?
[19-Oct-2011 13:28:07] * rmatte glares at Hackman
[19-Oct-2011 13:28:10] <rmatte> hehe
[19-Oct-2011 13:28:18] <kenhen93> no, we are still running 3.1
[19-Oct-2011 13:28:36] <kenhen93> was about to upgrade to 3.2 and saw not to, till 3.2.1 comes out
[19-Oct-2011 13:28:38] <rmatte> you're positive? Because there was a zencommand bug that caused this exact type of issue in 3.2
[19-Oct-2011 13:28:44] <rmatte> k
[19-Oct-2011 13:28:46] <rmatte> hmmm
[19-Oct-2011 13:29:01] <kenhen93> let me explain a little more
[19-Oct-2011 13:29:09] <kenhen93> i created a custom parser for this command
[19-Oct-2011 13:29:17] <kenhen93> and it seems since then it will not even run
[19-Oct-2011 13:29:18] <Hackman238> videoconferencin a bit rude....some guy from Augme Technologies in NYC
[19-Oct-2011 13:29:31] <rmatte> kenhen93: there's a chance that there's a problem with your parser
[19-Oct-2011 13:29:51] <rmatte> kenhen93: though I'm surprised that no errors are coming from zencommand if there is
[19-Oct-2011 13:29:56] <kenhen93> there is a chance since this is the first one I have created :-)
[19-Oct-2011 13:29:59] <kenhen93> exactly
[19-Oct-2011 13:30:02] <rmatte> kenhen93: how about you try changing the parser just for testing purposes?
[19-Oct-2011 13:30:05] <kenhen93> i expected errors
[19-Oct-2011 13:30:09] <rmatte> change the parser, then do zencommand run -v10 -d again
[19-Oct-2011 13:30:16] <kenhen93> i used to see errors in zenhub
[19-Oct-2011 13:30:16] <rmatte> and see if it runs with a stock parser
[19-Oct-2011 13:30:21] <kenhen93> ok, i can try that
[19-Oct-2011 13:30:50] <kenhen93> not it tuns
[19-Oct-2011 13:30:52] <kenhen93> runs!
[19-Oct-2011 13:30:59] <kenhen93> but i need to use my parser
[19-Oct-2011 13:31:13] <rmatte> then you need to fix your parser lol
[19-Oct-2011 13:31:21] <rmatte> also, why exactly do you need to use your parser?
[19-Oct-2011 13:31:24] <rmatte> what's so special about it?
[19-Oct-2011 13:31:36] <kenhen93> because the values are not in nagios or cacti format
[19-Oct-2011 13:31:44] <pplask> guys i think i found an annoying bug. Zenoss insists that "INFO zen.ZenModeler: SNMP monitoring off for device xxxxxxx"
[19-Oct-2011 13:31:46] <pplask> while its not
[19-Oct-2011 13:31:51] <kenhen93> i have a community discussion outlining it
[19-Oct-2011 13:31:59] <rmatte> it would have just been easier to change the script to make the values in that format than to write a new parser
[19-Oct-2011 13:32:16] <kenhen93> message/61922#61922
[19-Oct-2011 13:32:22] <rmatte> or to pipe the output of the script through a few commands
[19-Oct-2011 13:32:46] <rmatte> what does the output from the script look like?
[19-Oct-2011 13:32:50] <Hackman238> pplask: Have you tried going to the device in question and pushing the config?
[19-Oct-2011 13:33:11] <Hackman238> pplask: You shouldn't need to most of the time, but sometimes the auto push gets lost in the zenhub jumble
[19-Oct-2011 13:33:11] <pplask> Hackman238, yes
[19-Oct-2011 13:33:13] <kenhen93> i can paste it here
[19-Oct-2011 13:33:22] <kenhen93> or it might be easier to look at my post
[19-Oct-2011 13:33:35] <rmatte> I'm looking at the post, I don't see perference to output
[19-Oct-2011 13:33:36] <rmatte> oggg
[19-Oct-2011 13:33:37] <kenhen93> its basically value=0|value=1|
[19-Oct-2011 13:33:38] <rmatte> ohh*
[19-Oct-2011 13:33:44] <Hackman238> pplask: Hum. Have you restarted the zenhub for that collector?
[19-Oct-2011 13:33:45] <rmatte> Performance_1=0|Performance_2=224|
[19-Oct-2011 13:33:48] <rmatte> and so on?
[19-Oct-2011 13:33:51] <kenhen93> right
[19-Oct-2011 13:33:53] <rmatte> is it always 1 and 2 or can it be more?
[19-Oct-2011 13:33:55] <kenhen93> multiple that by 40
[19-Oct-2011 13:33:59] <rmatte> ah
[19-Oct-2011 13:34:02] <pplask> Hackman238, i restarted zenoss and tried to re-model the device
[19-Oct-2011 13:34:04] <kenhen93> the names are all different
[19-Oct-2011 13:34:05] <rmatte> give me a sec...
[19-Oct-2011 13:34:17] <kenhen93> like one could be QMI_CONNECTIONS
[19-Oct-2011 13:34:26] <kenhen93> the other could be QMI_CK_TOT_CONNECTION
[19-Oct-2011 13:34:52] <Hackman238> pplask: Huh. That's really weird.
[19-Oct-2011 13:35:07] <Hackman238> pplask: So you're getting 'snmp down' events, right?
[19-Oct-2011 13:35:08] <pplask> yeah
[19-Oct-2011 13:35:14] <pplask> no im not
[19-Oct-2011 13:35:22] <rmatte> kenhen93: show me the full command from your datasource
[19-Oct-2011 13:35:37] <pplask> Hackman238, zenoss thinks that my new device has zsnmpignore TRUE, when its false
[19-Oct-2011 13:35:38] <rmatte> kenhen93: fixing it to run with nagios parser will be a piece of cake, I just need to see the command
[19-Oct-2011 13:35:46] <Hackman238> pplask: Ah gotcha.
[19-Oct-2011 13:36:15] <Hackman238> pplask: Have you toggled it back to true, saved, pusehd, toggled to false, pushed?
[19-Oct-2011 13:36:38] <pplask> Hackman238, im trying this now, hold on a sec
[19-Oct-2011 13:36:47] <Hackman238> pplask: Alrighty
[19-Oct-2011 13:37:05] <rmatte> kenhen93: I understand, I just need to see the command
[19-Oct-2011 13:37:07] <kenhen93> http://fpaste.org/Q4Tw/
[19-Oct-2011 13:37:22] <kenhen93> thanks for you help rmatte!
[19-Oct-2011 13:37:30] <pplask> Hackman238, still the same problem
[19-Oct-2011 13:37:51] <Hackman238> pplask: Thats really bizzare
[19-Oct-2011 13:38:07] <rmatte> kenhen93: try this: output=$(/usr/local/blah/monitoring/bin/blah.pl /usr/local/blah/cs-monitoring/config/blah.prod.cfg | sed 's/|/ /g') && echo "|$output"
[19-Oct-2011 13:38:28] <Hackman238> pplask: If you don't mind, could you go to the device status page, gear icon -> delete device, uncheck both boxes, readd device?
[19-Oct-2011 13:38:30] <kenhen93> one more thing in the out :-/
[19-Oct-2011 13:38:49] <rmatte> kenhen93: if it doesn't work directly you can create a wrapper script on the server to do that and just execute that script instead
[19-Oct-2011 13:39:12] <kenhen93> the first line is something like this Wed, 19 Oct 2011 13:38:49 EDT: instance1
[19-Oct-2011 13:39:17] <kenhen93> then all the perf=1|perf=2
[19-Oct-2011 13:39:22] <rmatte> oh
[19-Oct-2011 13:39:27] <rmatte> one sec...
[19-Oct-2011 13:39:37] <kenhen93> i was using the instance 1 as my component
[19-Oct-2011 13:39:44] <rmatte> output=$(/usr/local/blah/monitoring/bin/blah.pl /usr/local/blah/cs-monitoring/config/blah.prod.cfg | tail -1 | sed 's/|/ /g') && echo "|$output"
[19-Oct-2011 13:39:46] <rmatte> there
[19-Oct-2011 13:39:46] <kenhen93> i need to know what instance of the application it is
[19-Oct-2011 13:40:04] <rmatte> oh I see
[19-Oct-2011 13:40:10] <rmatte> yeh, can't do that with this
[19-Oct-2011 13:40:38] <rmatte> all my way is able to do is actually get the data out
[19-Oct-2011 13:40:56] <rmatte> let me see your parser code?
[19-Oct-2011 13:41:02] <kenhen93> ok
[19-Oct-2011 13:43:05] <kenhen93> http://fpaste.org/T7nC/
[19-Oct-2011 13:45:14] <pplask> Hackman238, i did it. Still the same problem
[19-Oct-2011 13:45:31] <Hackman238> pplask: Hum.
[19-Oct-2011 13:45:42] <Hackman238> pplask: No other devices are exhibiting this issue?
[19-Oct-2011 13:46:16] <humanix> Hackman238, you know any template for F5?
[19-Oct-2011 13:46:35] <humanix> im researching, i found some cacti ones but tmm memory usage is inconsistant with dashboard statics
[19-Oct-2011 13:46:49] <rmatte> kenhen93: have you thoroughly tested the regex?
[19-Oct-2011 13:47:15] <rmatte> kenhen93: the fact that it's not throwing any errors tells me that the parser is running without erroring out, but that it's not actually pulling any data
[19-Oct-2011 13:47:15] <Hackman238> humanix: Big IP device?
[19-Oct-2011 13:47:24] <rmatte> and the biggest culprit for that would be your regex strings
[19-Oct-2011 13:47:30] <kenhen93> yes, at http://gskinner.com/RegExr/
[19-Oct-2011 13:47:33] <kenhen93> the problem i have though
[19-Oct-2011 13:47:46] <kenhen93> is the question in my discussion
[19-Oct-2011 13:48:03] <kenhen93> for the 'scanners' plugin, what is the purpose of the '\d+' in (?P<availableBlocks>\d+). I understand what \d+ means in terms of python regex but is that \d+ the regex of the variable <availableBlocks> or is that saying it has trailing digits after the variable?
[19-Oct-2011 13:50:40] <rmatte> If you compare the regex...
[19-Oct-2011 13:50:42] <rmatte> ( r' (?P<availableBlocks>\d+) +(?P<percentUsed>\d+)%' )
[19-Oct-2011 13:50:48] <rmatte> to the output of the actually df command
[19-Oct-2011 13:50:52] <rmatte> Filesystem 1K-blocks Used Available Use% Mounted on
[19-Oct-2011 13:51:00] <rmatte> /dev/sda1 472765484 227198296 221741192 51% /
[19-Oct-2011 13:51:17] <rmatte> the regex basically starts at the % symbol
[19-Oct-2011 13:51:20] <rmatte> and works it's way back
[19-Oct-2011 13:51:28] <kenhen93> right
[19-Oct-2011 13:51:31] <kenhen93> i understand that
[19-Oct-2011 13:51:44] <kenhen93> the % is a place marker
[19-Oct-2011 13:52:42] <kenhen93> my starts by searching for the 'value=1|
[19-Oct-2011 13:52:54] <kenhen93> or the value name
[19-Oct-2011 13:53:01] <kenhen93> is that not the same thing?
[19-Oct-2011 13:53:31] <pplask> Hackman238, no, ony this device (its new)
[19-Oct-2011 13:53:43] <pplask> Hackman238, i will try another
[19-Oct-2011 13:54:14] <Hackman238> pplask: Thats really really strange.
[19-Oct-2011 13:54:26] <enkrypt> any people here know their way around rsyslog? I can't get the dynamic filenames to work. I've got 2 identical servers, identical rsyslog.conf, on one server it works fine, but the other keeps complaining he can't write to the files. google doesn't help me out.
[19-Oct-2011 13:54:43] <pplask> Hackman238, maybe its some kind of cache somewhere...
[19-Oct-2011 13:55:21] <rmatte> kenhen93: I see what it does
[19-Oct-2011 13:55:59] <kenhen93> ok, any idea whats wrong? :-/
[19-Oct-2011 13:56:07] <Hackman238> pplask: we could try removing the device and restarting zenoss altogether.
[19-Oct-2011 13:56:14] <pplask> Hackman238, ok
[19-Oct-2011 13:56:18] <humanix> yes Hackman238
[19-Oct-2011 13:56:18] <Hackman238> pplask: that would clear all caches
[19-Oct-2011 13:56:37] <Hackman238> humanix: I have a guy here working on a pack for that, let me see what he has
[19-Oct-2011 13:56:48] <pplask> Hackman238, i deleted and restarted, now im gonna add again
[19-Oct-2011 13:57:04] <rmatte> kenhen93: nope, I just know what the + is doing
[19-Oct-2011 13:57:24] <Hackman238> pplask: Alrighty
[19-Oct-2011 13:57:44] <rmatte> "The plus sign tells the Regular Expression to match the character, group, or character class that immediately precedes it one or more times. This means that the character, group, or character class must be found at least once. After it is found once it will be matched again if it follows the first match."
[19-Oct-2011 13:58:20] <pplask> Hackman238, added, same problem
[19-Oct-2011 13:58:27] <kenhen93> hmm
[19-Oct-2011 13:58:50] <Hackman238> pplask: Sorry to have you trying a ton of stuff, but can you run a zenchkrels -x1 -r as user zenoss?
[19-Oct-2011 13:58:57] <Hackman238> pplask: Something really weird going on LOL
[19-Oct-2011 13:59:04] <rmatte> kenhen93: http://fpaste.org/YbZ9/raw/
[19-Oct-2011 13:59:35] <pplask> Hackman238, its checking all relations...
[19-Oct-2011 13:59:48] <rmatte> basically it tells it to match it as long as it's numerical characters
[19-Oct-2011 13:59:48] <Hackman238> humanix: He gave me this MIB link: http://www.oidview.com/mibs/3375/F5-BIGIP-COMMON-MIB.html
[19-Oct-2011 13:59:58] <rmatte> so once it hits anything other than them it stops
[19-Oct-2011 14:00:24] <rmatte> i.e, it'd hit the =
[19-Oct-2011 14:00:33] <pplask> Hackman238, ops, error!
[19-Oct-2011 14:00:53] <pplask> Hackman238, omg, i found the problem i think. Disk full!
[19-Oct-2011 14:00:58] <humanix> im right there
[19-Oct-2011 14:01:01] <Hackman238> pplask: df -h?
[19-Oct-2011 14:01:05] <rmatte> or the | rather
[19-Oct-2011 14:01:05] <Hackman238> pplask:
[19-Oct-2011 14:01:09] <kenhen93> ok, but how do i specify each value, just by the order its outputed?
[19-Oct-2011 14:01:10] <humanix> i have some inconsistency with some information
[19-Oct-2011 14:01:22] <Hackman238> humanix: Or this site if you don't mind a 'pain in the ass site' http://support.ipmonitor.com/mibs/F5-BIGIP-SYSTEM-MIB/tree.aspx
[19-Oct-2011 14:01:22] <pplask> Hackman238, 2% free... but wtf??? 200gb???
[19-Oct-2011 14:01:32] <Hackman238> pplask: Have you run Zeopack altely?
[19-Oct-2011 14:01:43] <Hackman238> pplask: do a `ls -lh /$ZENHOME/var`
[19-Oct-2011 14:01:52] <pplask> Hackman238, oh no, im so sry man, i read it wrong, the disk isnt full, its 2% used.
[19-Oct-2011 14:01:53] <Hackman238> pplask: I bet its Zope
[19-Oct-2011 14:02:04] <Hackman238> pplask: Ah ok LOL
[19-Oct-2011 14:02:16] <pplask> Hackman238, i will pastebin the error
[19-Oct-2011 14:03:19] <Hackman238> pplask: TY
[19-Oct-2011 14:03:22] <kenhen93> right, because the | is the end of the value
[19-Oct-2011 14:03:31] <pplask> Hackman238, http://pastebin.com/u6qqJED1
[19-Oct-2011 14:03:57] <rmatte> kenhen93: I honestly don't know at this point.
[19-Oct-2011 14:04:30] <kenhen93> ok, thanks for your help
[19-Oct-2011 14:04:33] <rmatte> np
[19-Oct-2011 14:05:25] <Hackman238> pplask: please `su zenoss`; `zendmd`; then `for user in dmd.ZenUsers.getUsers(): user.messages.removeRelation()` {double enter} `commit()` {enter}
[19-Oct-2011 14:06:25] <pplask> Hackman238, done
[19-Oct-2011 14:06:42] <Hackman238> pplask: CTRL +D then rerun zenchkrels -x1 -r
[19-Oct-2011 14:06:58] <pplask> Hackman238, running...
[19-Oct-2011 14:08:30] <pplask> Hackman238, same error
[19-Oct-2011 14:08:59] <Hackman238> pplask: Huh, that shouldnt be.
[19-Oct-2011 14:10:58] <Hackman238> pplask: `su zenoss`; `zendmd`; `u=dmd.ZenUsers._getOb('admin')`; `u.removeRelation('messages')`; `commit()`
[19-Oct-2011 14:11:50] <pplask> Hackman238, done and running that command again
[19-Oct-2011 14:12:08] <Hackman238> pplask: Alrighty
[19-Oct-2011 14:13:35] <pplask> Hackman238, no errors now
[19-Oct-2011 14:14:42] <pplask> Hackman238, i will try to readd the device
[19-Oct-2011 14:15:16] <Hackman238> pplask: Nice
[19-Oct-2011 14:17:17] <pplask> Hackman238, same problem
[19-Oct-2011 14:17:23] <Hackman238> pplask: Damn.
[19-Oct-2011 14:17:48] <Hackman238> pplask: When you add the device are you adding by ip or hostname?
[19-Oct-2011 14:17:48] <pplask> Hackman238, i will delete one device here and readd it to see if it happens
[19-Oct-2011 14:17:53] <pplask> Hackman238, IP
[19-Oct-2011 14:18:18] <Hackman238> pplask: try this...add it again as the wrong ip.
[19-Oct-2011 14:18:35] <Hackman238> pplask: then find the device and click the gear icon - > reset ip -> correct ip
[19-Oct-2011 14:18:45] <Hackman238> pplask: gear icon -> model
[19-Oct-2011 14:18:55] <Hackman238> pplask: Lets see if that tricks Zenoss in to creating a new object
[19-Oct-2011 14:18:57] <Hackman238> LOL
[19-Oct-2011 14:19:01] <pplask> Hackman238, ok lemme try
[19-Oct-2011 14:19:33] <Hackman238> pplask: TY
[19-Oct-2011 14:21:21] <pplask> Hackman238, snmp was ON for the new device with wrong ip, now im changing the ip
[19-Oct-2011 14:21:36] <Hackman238> pplask: Thats a start
[19-Oct-2011 14:22:45] <Hackman238> rmatte: How's your formula pack going?
[19-Oct-2011 14:24:54] <pplask> Hackman238, hey man, i think we really found a bug... the wrong ip i used is unused. i changed the ip to the right one, and when i click model device, everything works fine, it models my device with the right IP, but in the windows "Model Device", where you can follow the progress of the modelling it says: "Collecting for device $oldIP"
[19-Oct-2011 14:25:29] <Hackman238> pplask: Afraid it looks that way. Could you submit a bug report?
[19-Oct-2011 14:25:30] <pplask> Hackman238, i dont know if i was clear. It worked, but in the database of zenoss, my device is named with the wrong ip
[19-Oct-2011 14:25:39] <Hackman238> pplask: Oh- ok
[19-Oct-2011 14:25:41] <Hackman238> pplask: thats ok
[19-Oct-2011 14:25:53] <Hackman238> pplask: Go to the device and gear icon -> rename
[19-Oct-2011 14:25:53] <pplask> Hackman238, yes
[19-Oct-2011 14:26:30] <Hackman238> pplask: Thats just because zenoss set the title to the initial ip Rename sets it to whatever you want
[19-Oct-2011 14:27:38] <pplask> Hackman238, mmm ok. So i should submit a bug saying that zenoss was insisting that my device has snmpmonitorignore TRUE until i trick it adding with a new ip and then changing it?
[19-Oct-2011 14:28:39] <Hackman238> pplask: I would submit a bug report on it descibing what we ended up having to do to get around it. It sort of sounds like it was keeping bits of the origonal device object.
[19-Oct-2011 14:28:48] <Hackman238> pplask: If you could please.
[19-Oct-2011 14:28:56] <pplask> Hackman238, i will
[19-Oct-2011 14:29:51] <Hackman238> pplask: Thanks man. Glad we got it working, at least in a workaround fashion
[19-Oct-2011 14:33:12] <pplask> Hackman238, thanks for you help
[19-Oct-2011 14:34:48] <pplask> Hackman238, didnt find a way to create an account to open a ticket
[19-Oct-2011 14:34:57] <pplask> http://dev.zenoss.com/trac/
[19-Oct-2011 14:36:39] <Hackman238> pplask: user:zenoss password:zenoss
[19-Oct-2011 14:36:44] <Hackman238> pplask: No idea why
[19-Oct-2011 14:36:49] <Hackman238> LOL
[19-Oct-2011 15:00:26] <Hackman238> All - Please vote in the ZCA ZenPack poll - polls/1041 - Help drive the direction of CORE!
[19-Oct-2011 16:31:06] <Hackman238> All - Please vote in the ZCA ZenPack poll - polls/1041 - Help drive the direction of CORE!
[19-Oct-2011 17:10:04] kocolosk is now known as kocolosk|away
[19-Oct-2011 19:13:22] kocolosk|away is now known as kocolosk
[19-Oct-2011 21:58:15] kocolosk is now known as kocolosk|away
[20-Oct-2011 00:00:01] [disconnected at Thu Oct 20 00:00:01 2011]
[20-Oct-2011 00:00:01] [connected at Thu Oct 20 00:00:01 2011]
[20-Oct-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[20-Oct-2011 05:01:18] <enkrypt> I suddenly noticed that my /Ping device class (including around 30 assets in there) has simply disappeared
[20-Oct-2011 05:01:28] <enkrypt> any idea how that might be? any way to check which user did that?
[20-Oct-2011 05:03:19] <enkrypt> I checked /opt/zenoss/log/Z2.log and it shows no evidence of a user deleting it manually
[20-Oct-2011 05:15:18] <aberrios> How can I debug an install of a zenpack that's erroring.
[20-Oct-2011 05:15:21] <aberrios> ?
[20-Oct-2011 06:48:11] <aberrios> I have a zenpack "UPS-MIB based UPSs", when trying to install on 3.2.0 it just dies with no error message except for "Error installing zenpack UPS-MIB based UPSs egg etc..."
[20-Oct-2011 06:48:22] <aberrios> It is marked as compatible with 3.0
[20-Oct-2011 06:48:54] <aberrios> I'm trying to find exactly what the error is while installing
[20-Oct-2011 07:35:26] <luizzmizz> my zenoss 3.2.0 catalogs don't get updated after changes, and updating them manually works alright.... did someone live the same situation?
[20-Oct-2011 07:35:49] <luizzmizz> (hi)
[20-Oct-2011 08:00:58] <luizzmizz> could someone please help me with this? message/61748#61748
[20-Oct-2011 08:28:57] <hwinkel> Hello
[20-Oct-2011 08:30:17] <hwinkel> Hi, I'll evaluate a few NMS systems the next days. We run Zenoss 3.x for quite some time now in an lab environment
[20-Oct-2011 08:30:46] <jmp242> Hi hwinkel
[20-Oct-2011 08:30:50] <jmp242> Morning all
[20-Oct-2011 08:30:56] <hwinkel> because it will be a provider centric system there is a strong fucus an IPv6
[20-Oct-2011 08:31:40] <hwinkel> Dioes anybody know the state of affairs for IPv6 and what is comingin 4.0. I found some empty docs but no roadmap etc...
[20-Oct-2011 08:31:51] <hwinkel> But some Zenpack already mention a 4.0 number ?!?
[20-Oct-2011 08:33:45] <jmp242> hwinkel: 4.1 is actually the current paid release
[20-Oct-2011 08:33:59] <jmp242> if you buy Service Dynamics (Enterprise is our shorthand)
[20-Oct-2011 08:34:12] <jmp242> 3.2.1 is out tomorrow or next week for Core
[20-Oct-2011 08:34:18] <jmp242> When 4.2 comes out
[20-Oct-2011 08:34:28] <jmp242> which is supposed to be a core release by early next year
[20-Oct-2011 08:34:37] <jmp242> it may support IPv6
[20-Oct-2011 08:34:54] <jmp242> It's supposed to be the core of what SD is
[20-Oct-2011 08:35:08] <jmp242> so I wouldn't be surprised, but it depends on what's released as Core
[20-Oct-2011 08:35:52] <jmp242> 4.2 is supposed to be a big improvement over 3.2.x when it comes out according to many using the paid version of 4
[20-Oct-2011 08:36:02] <hwinkel> jmp242: thanks a lot... its a bit confusing after the years of similar versions for core and paid
[20-Oct-2011 08:36:19] <jmp242> yes
[20-Oct-2011 08:36:27] <jmp242> I think this is a confused period
[20-Oct-2011 08:36:42] <jmp242> my understanding is once 4.2 of Core is released, the versions will sync up again
[20-Oct-2011 08:37:11] <jmp242> At least, that's what the blogs etc say
[20-Oct-2011 08:37:29] <hwinkel> I need to do proper consultant job for a SP, how can I evaluate the IPv6 and other new features If they have such a huge difference in versioning
[20-Oct-2011 08:37:34] <jmp242> Plus, with the ZCA we hope to get more good community stuff going
[20-Oct-2011 08:37:43] <jmp242> well
[20-Oct-2011 08:37:56] <hwinkel> but thanks for this clarification
[20-Oct-2011 08:37:57] <jmp242> you could get a trial of the paid product (if you were ever thinking about buying it
[20-Oct-2011 08:38:25] <jmp242> or if you can let people know there's a new release shortly you could always wait (lol)
[20-Oct-2011 08:38:28] <hwinkel> The SP thinks about buying, if the License model starts to reflect Service Provider needs
[20-Oct-2011 08:38:46] <jmp242> I think you would have to talk to sales
[20-Oct-2011 08:38:50] <hwinkel> sure, management by anoncement
[20-Oct-2011 08:39:01] <jmp242> I don't think there's actually a set license model per se
[20-Oct-2011 08:39:11] <hwinkel> I'll ask sales ... would be the best way
[20-Oct-2011 08:39:20] <jmp242> many enterprisy things are you negotiate the model that works for both sides
[20-Oct-2011 08:39:45] <jmp242> I know, even at mid size where I work, we rarely actually pay a list price exactly
[20-Oct-2011 08:39:55] <jmp242> afk for a bit
[20-Oct-2011 08:40:07] <hwinkel> Zenoss has proven to be sort of stable across the upgrading pass. We spend not much effort to keep the system going after patch releases
[20-Oct-2011 08:40:19] <hwinkel> This is already a good sign for us
[20-Oct-2011 08:40:21] <hwinkel> ;-)
[20-Oct-2011 08:40:40] <hwinkel> I mad other bad expierence with other systems
[20-Oct-2011 09:01:52] <cek> Server Exception The server reported the following error: ValueError Supplied vocabulary values resulted in duplicate term tokens
[20-Oct-2011 09:02:16] <cek> kek. when tried to add a second "domain" ipservice. Can't zenoss monitor udp?
[20-Oct-2011 09:03:00] <jmp242> cek no
[20-Oct-2011 09:03:09] <jmp242> zenoss doesn't monitor UDP
[20-Oct-2011 09:04:09] <cek> is zenoss monitoring by ip or by hostname? does it do resolution during probing?
[20-Oct-2011 09:04:29] <cek> or is taht done only during discovery phase?
[20-Oct-2011 09:04:41] <jmp242> uses IP
[20-Oct-2011 09:04:49] <jmp242> DNS lookup is done when it's created only
[20-Oct-2011 09:05:02] <cek> how do i change that ip?
[20-Oct-2011 09:05:30] <cek> oh, found mnaageIP property in zope
[20-Oct-2011 09:05:44] <jmp242> don't mess in zope unless you're an expert
[20-Oct-2011 09:05:48] <jmp242> you can break Zenoss
[20-Oct-2011 09:06:04] <jmp242> there should be something in the Zenoss GUI for that, though I don't recall exactly
[20-Oct-2011 09:06:47] <luizzmizz> hi there... could someone please help me with this? message/61748#61748
[20-Oct-2011 09:06:55] <cek> yeah, where is that "something" ?:)
[20-Oct-2011 09:12:08] <jmp242> mmm
[20-Oct-2011 09:12:14] <jmp242> I don't know
[20-Oct-2011 09:12:20] <jmp242> There were forum threads about this
[20-Oct-2011 09:12:26] <jmp242> it may be ok to change ManageIp
[20-Oct-2011 09:12:32] <cek> i've got alternating events all over logs: threshold of zenperfsnmp cycle time exceeded: current value 300.00
[20-Oct-2011 09:12:32] <jmp242> though most people were doing it via dmd
[20-Oct-2011 09:12:38] <cek> threshold of zenperfsnmp cycle time restored: current value 0.01
[20-Oct-2011 09:14:19] <jmp242> Sounds like you're going over the snmp cycle time
[20-Oct-2011 09:41:01] <dhopp> hmmm…I seem to have a number of devices where zenprocess is saying it can't find http and then the next polling cycle it clears
[20-Oct-2011 09:43:55] <jmp242> timeouts on the process list via SNMP?
[20-Oct-2011 09:46:51] <dhopp> I don't think so because I only have like 10 devices on this installation
[20-Oct-2011 09:50:45] <jmp242> no
[20-Oct-2011 09:50:52] <jmp242> I mean on the device itself
[20-Oct-2011 09:50:57] <jmp242> where you're getting the error
[20-Oct-2011 09:51:03] <jmp242> Zenoss lists running processes via SNMP
[20-Oct-2011 09:51:15] <jmp242> if the device is intermittantly timing out, you'll get what you're saying
[20-Oct-2011 09:51:19] <jmp242> or if it's a really long process list
[20-Oct-2011 09:51:57] <jmp242> I think if it's truncating the process list
[20-Oct-2011 09:52:09] <dhopp> restarting zenprocess seems to have fixed it
[20-Oct-2011 09:52:09] <jmp242> you might be able to help by playing with the maxoidsperrequest
[20-Oct-2011 09:52:13] <jmp242> or that
[20-Oct-2011 09:52:25] <jmp242> zenprocess also sometimes needs to be booted
[20-Oct-2011 09:52:30] <jmp242> a lot of zenoss is like windows
[20-Oct-2011 09:52:32] <jmp242> IME
[20-Oct-2011 09:52:36] <jmp242> try a restart
[20-Oct-2011 09:52:47] <dhopp> yeah…that's not going to fly
[20-Oct-2011 09:52:55] <dhopp> I'm going to have to find a solution
[20-Oct-2011 09:54:16] <dhopp> I'm going to have ~1200 devices being monitored…having to restart zenoss services regularly is not an option
[20-Oct-2011 09:59:04] <Hackman238> Morning all
[20-Oct-2011 09:59:51] <Hackman238> dhopp: What's up with your install?
[20-Oct-2011 10:00:33] <dhopp> Hackman238: I'm not sure…but I just noticed that 'httpd' process alerts for missing and then next polling cycle clears
[20-Oct-2011 10:00:39] <dhopp> apparently been doing it all night
[20-Oct-2011 10:00:56] <dhopp> restart of zenprocess seems to have fixed it…but I'm trying to figure out why it even started
[20-Oct-2011 10:01:35] <dhopp> Unable to read processes on device server1; error: httpd
[20-Oct-2011 10:01:37] <Hackman238> dhopp: Sounds like zenhub was flooded
[20-Oct-2011 10:01:39] <dhopp> that's the exact error it gave me
[20-Oct-2011 10:02:09] <dhopp> with only 36 devices on this particular install?
[20-Oct-2011 10:02:15] <Hackman238> dhopp: How many processes are you collecting for?
[20-Oct-2011 10:02:22] <dhopp> currently httpd
[20-Oct-2011 10:02:24] <dhopp> that's it
[20-Oct-2011 10:02:33] <Hackman238> dhopp: Just 36 devices?
[20-Oct-2011 10:02:37] <dhopp> currently
[20-Oct-2011 10:02:40] <Hackman238> dhopp: Hum. Not zenhub then.
[20-Oct-2011 10:02:40] <dhopp> it's a dev install
[20-Oct-2011 10:02:46] <Hackman238> dhopp: That weird.
[20-Oct-2011 10:03:02] <dhopp> but my production will have ~1200 devices..of which ~700 or so will monitor for httpd among others
[20-Oct-2011 10:03:06] <Hackman238> dhopp: Whats your zenhub.conf and zenprocess.conf look like?
[20-Oct-2011 10:03:20] <Hackman238> dhopp: can you email them to me or pastebin them?
[20-Oct-2011 10:03:54] <dhopp> Hackman238: zenprocess.conf has nothing special in it..just specify monitor, hubhost and logpath
[20-Oct-2011 10:03:57] <dhopp> (remote collector)
[20-Oct-2011 10:04:44] <dhopp> zenhub is just:
[20-Oct-2011 10:04:45] <dhopp> #PARAMETER VALUE
[20-Oct-2011 10:04:45] <dhopp> workers 6
[20-Oct-2011 10:04:45] <dhopp> cachesize 100000
[20-Oct-2011 10:04:45] <dhopp> pcachesize 2000
[20-Oct-2011 10:04:50] <Hackman238> dhopp: Do you mind if I reccomend some tweaks?
[20-Oct-2011 10:04:54] <dhopp> sure
[20-Oct-2011 10:05:03] <dhopp> sure as in I don't mind :-)
[20-Oct-2011 10:06:16] <dhopp> oh actually I just realized this is my production install doing it…and I only have 14 devices on it right now
[20-Oct-2011 10:07:12] <Hackman238> dhopp: in zenprocess.conf, maxparallel 100 eventflushchunksize 200
[20-Oct-2011 10:07:22] <Hackman238> dhopp: Then stop it and start it
[20-Oct-2011 10:07:28] <dhopp> ok…and what do those do?
[20-Oct-2011 10:07:38] <dhopp> and should I be adding something similar to other daemons?
[20-Oct-2011 10:07:47] <Hackman238> dhopp: maxparallel reduces the number of processes checked in bulk
[20-Oct-2011 10:08:05] <Hackman238> dhopp: eventChunk is how many events to throw at zenhub at a time
[20-Oct-2011 10:08:23] <dhopp> ok but I only have 1 process and 14 devices..I'm no where near even 100?
[20-Oct-2011 10:08:51] <Hackman238> dhopp: yes, but it ensures if the queue gets heavy for some reason it doesnt plow zenhub
[20-Oct-2011 10:09:01] <dhopp> ok
[20-Oct-2011 10:09:35] <Hackman238> dhopp: zenhub.conf looks good. How many cores does this box have?
[20-Oct-2011 10:10:05] <dhopp> 16
[20-Oct-2011 10:10:19] <dhopp> but it also will run multiple zopes and mysql
[20-Oct-2011 10:10:27] <Hackman238> dhopp: Lets try an experiement
[20-Oct-2011 10:11:01] <dhopp> ok
[20-Oct-2011 10:12:53] <dhopp> Hackman238: basically I did the configuration such that zenhub has 6 threads, mysql has ~6 for Innodb, and then 4 zopes on the zenhub machine (which isn't monitoring any other devices besides the zenoss devices and a few core devices)
[20-Oct-2011 10:13:00] <Hackman238> dhopp: time how long a zenperfsnmp run now takes to get to the line "Configured XXX of XXX devices" where XXX is all the devices monitored by zenperfsnmp[
[20-Oct-2011 10:13:58] <Hackman238> dhopp: If its more than a couple minutes, there's a zenhub bottleneck we need to locate. If it doesn't, how many devices total?
[20-Oct-2011 10:16:52] <dhopp> Hackman238: considering that a zenperfsnmp run -v10 takes less then 3 seconds to run completely through my guess is there isn't a bottleneck
[20-Oct-2011 10:17:10] <Hackman238> dhopp: I'm just wondering about the configuration time
[20-Oct-2011 10:17:28] <Hackman238> dhopp: Long configuration times are a tell of hampered zenhub.
[20-Oct-2011 10:17:49] <dhopp> Hackman238: no that's what I mean..from the time I did zenperfsnmp run -v10 and hit enter and the time the shell returned was less than 3 seconds
[20-Oct-2011 10:18:01] <Hackman238> dhopp: Oh...thats config and all
[20-Oct-2011 10:18:04] <Hackman238> dhopp: LOL
[20-Oct-2011 10:18:05] <dhopp> yes
[20-Oct-2011 10:18:40] <Hackman238> dhopp: I'd try those zenprocess entries. I thinnk your zenhub config is working right like it should be
[20-Oct-2011 10:18:49] <dhopp> but like I said..this particular zenperfsnmp is only currently responsible for 7 devices
[20-Oct-2011 10:18:54] <dhopp> so I wouldn't expect it to take long
[20-Oct-2011 10:19:38] <dhopp> Hackman238: and I have cacheconfigs true
[20-Oct-2011 10:19:44] <Hackman238> dhopp: Oh gotcha. Not a good test then
[20-Oct-2011 10:19:52] <dhopp> cacheconfigs true
[20-Oct-2011 10:19:53] <dhopp> maxqueuelen 500000
[20-Oct-2011 10:19:53] <dhopp> eventflushchunksize 200
[20-Oct-2011 10:20:04] <dhopp> that's in my zenperfsnmp
[20-Oct-2011 10:20:04] <Hackman238> dhopp: Ah ok.
[20-Oct-2011 10:20:30] <Hackman238> dhopp: v3.2 + patches, right?
[20-Oct-2011 10:20:33] <dhopp> yes
[20-Oct-2011 10:20:52] <Hackman238> dhopp: Okay. Yeah just tweak the zenprocess.conf and lets let it ride for a littlle while.
[20-Oct-2011 10:21:17] <Hackman238> dhopp: Its definiately not that your setup is hampered, thats for sure
[20-Oct-2011 10:21:21] <Hackman238>
[20-Oct-2011 10:26:31] <dhopp> Hackman238: that brings up a question..with cacheconfigs true how long does it cache for?
[20-Oct-2011 10:27:10] <Hackman238> dhopp: long as the persistant cache hasnt been flagged dirty and the config hasnt changed
[20-Oct-2011 10:27:55] <Hackman238> dhopp: pcache is usually in $ZENHOME/var
[20-Oct-2011 10:28:15] <dhopp> I didn't configure a pcache though
[20-Oct-2011 10:28:25] <dhopp> or is it automatic with cacheconfigs true
[20-Oct-2011 10:28:28] <Hackman238> dhopp: Zenhubs using pcache
[20-Oct-2011 10:29:15] <Hackman238> dhopp: idk where config caching happens
[20-Oct-2011 10:29:21] <Hackman238> dhopp: let me look and find out
[20-Oct-2011 10:30:00] <dhopp> Hackman238: and do you if zenperfsnmp is the only one that has cacheconfig option? I can look at the code but figured you might now off the top of your head
[20-Oct-2011 10:30:30] <Hackman238> dhopp: Stored in a pickle looks like
[20-Oct-2011 10:30:56] <dhopp> ah yes..I see those in the perf/Device tree
[20-Oct-2011 10:31:42] <Hackman238> dhopp: I think only zenperf does
[20-Oct-2011 10:31:48] <Hackman238> not 100% on that
[20-Oct-2011 10:32:29] <Hackman238> dhopp: zenperfsnmp*
[20-Oct-2011 10:32:50] <dhopp> Hackman238: I knew what you meant
[20-Oct-2011 10:35:51] <Hackman238> dhopp: Let me know if the zenprocess problem acts up or stays working.
[20-Oct-2011 10:36:06] <Hackman238> dhopp: Aldo, did you get the data range pack?
[20-Oct-2011 10:36:12] <Hackman238> *Also
[20-Oct-2011 10:42:03] <dhopp> Hackman238: I haven't seen it..did you e-mail it to me?
[20-Oct-2011 10:42:12] <dhopp> might have gotten caught in the good ol' spam filter
[20-Oct-2011 10:43:26] <Hackman238> dhopp: Yep
[20-Oct-2011 10:43:37] <Hackman238> dhopp: To all ZCA
[20-Oct-2011 10:45:42] <dhopp> when did you send it?
[20-Oct-2011 10:45:53] <jmp242> I don't remember it either
[20-Oct-2011 10:45:57] <jmp242> can you re-send
[20-Oct-2011 10:47:25] <Hackman238> Resent to ZCA
[20-Oct-2011 10:47:41] <fragfutter> any known issues with 3.2.0?
[20-Oct-2011 10:48:04] <fragfutter> or should i start with 3.2.1
[20-Oct-2011 10:48:21] <jmp242> start with 3.2.1
[20-Oct-2011 10:48:22] <Hackman238> fragfutter: More than a few. I'd wait for 3.2.1 if you can hold off a week.
[20-Oct-2011 10:48:37] <jmp242> several known issues with 3.2
[20-Oct-2011 10:49:08] <fragfutter> ah, 21.oct would be tomorrow...
[20-Oct-2011 10:51:00] <fragfutter> then i will prepare kickstart and puppet today and wait...
[20-Oct-2011 10:51:44] <Hackman238> jmp242 / dhopp: get it this time?
[20-Oct-2011 10:52:51] <jmp242> yes
[20-Oct-2011 10:52:55] <jmp242> I saw that before
[20-Oct-2011 10:52:59] <jmp242> but didn't know what it was lol
[20-Oct-2011 10:53:17] <Hackman238> jmp242: Ah ok. Date range for graphs
[20-Oct-2011 10:53:36] <Hackman238> jmp242: ..and more. It's the v2.5 copy
[20-Oct-2011 10:54:53] <aberrios> I'm not getting any events pop up from SNMP traps sent from my UPS.. any ideas?
[20-Oct-2011 10:55:33] <jmp242> What, so I can specify a range for a graph display?
[20-Oct-2011 10:55:45] <Hackman238> aberrios: Is the UPS sending to the right community? Is zentrap running?
[20-Oct-2011 10:56:17] <aberrios> zentrap is running, its receiving traps from other devices. The UPS is setup with the correct community as far as I can see.
[20-Oct-2011 10:56:43] <aberrios> I have two different trap types in the UPS setup
[20-Oct-2011 10:56:56] <aberrios> PPC and RFC1628
[20-Oct-2011 10:57:08] <aberrios> I'm not sure which to use, but I've imported the MIBs from the UPS cd into Zenoss
[20-Oct-2011 10:59:20] <Hackman238> aberrios: UPS sending to port 162 and ports are open?
[20-Oct-2011 10:59:39] <aberrios> it is sending to port 162, let me check if its open
[20-Oct-2011 10:59:59] <aberrios> ah doesnt seem to be open., where do i configure that?
[20-Oct-2011 11:00:52] <Hackman238> aberrios: it could be blocked along the path
[20-Oct-2011 11:01:20] <aberrios> i cant telnet to it from my machine
[20-Oct-2011 11:01:48] <aberrios> nor can I telnet to it from the zenoss machine
[20-Oct-2011 11:02:20] <dhopp> aberrios: you won't be able to telnet to it because it's a udp port
[20-Oct-2011 11:02:27] <dhopp> aberrios: do netstat -an | grep 162
[20-Oct-2011 11:02:35] <dhopp> on the zenoss server
[20-Oct-2011 11:02:42] <aberrios> yes its open
[20-Oct-2011 11:02:43] <aberrios> udp 0 0 0.0.0.0:162 0.0.0.0:*
[20-Oct-2011 11:03:27] <dhopp> aberrios: do you have the linux firewall on?
[20-Oct-2011 11:03:31] <aberrios> nope
[20-Oct-2011 11:03:33] <aberrios> check it
[20-Oct-2011 11:03:36] <aberrios> checked*
[20-Oct-2011 11:03:42] <Hackman238> aberrios: Oh gotcha. IDK if thats a good test.
[20-Oct-2011 11:04:09] <fragfutter> dhopp: go advanced... netstat -nlu | grep 162
[20-Oct-2011 11:04:35] <dhopp> fragfutter: yeah I could do that…but I have a habbit of just doing -an..usually has what I need :-)
[20-Oct-2011 11:04:54] <fragfutter> dhopp: until you do it on a realy busy server
[20-Oct-2011 11:05:04] <dhopp> fragfutter: touche
[20-Oct-2011 11:05:16] <dhopp> aberrios: is the ups and zenoss on the same subnet?
[20-Oct-2011 11:05:23] <aberrios> yes
[20-Oct-2011 11:05:31] <dhopp> so there is no gateway involved?
[20-Oct-2011 11:05:42] <fragfutter> aberrios: leave a tcpdump running on the zenoss to see what is coming in?
[20-Oct-2011 11:07:09] <aberrios> got this: 16:06:52.847003 IP 192.168.1.157.simba-cs > sylvanas.sheffieldcitytaxis.com.snmptrap: Trap(26) E:935 192.168.1.157 coldStart 0
[20-Oct-2011 11:07:20] <aberrios> when testing traps from the UPS
[20-Oct-2011 11:07:26] <aberrios> UPS is 192.168.1.157
[20-Oct-2011 11:07:46] <fragfutter> what is "netstat -nlup | grep 162" saying?
[20-Oct-2011 11:08:04] <aberrios> udp 0 0 0.0.0.0:162 0.0.0.0:* 3401/.python.bin
[20-Oct-2011 11:08:29] <fragfutter> aberrios: ok. justed wanted to make sure it's not net-snmp.
[20-Oct-2011 11:12:13] <aberrios> So the traps are coming in but nothing happening in Zenoss
[20-Oct-2011 11:48:02] <aberrios> bingo was fixed with a patch
[20-Oct-2011 11:48:06] <aberrios> this patch
[20-Oct-2011 11:48:20] <aberrios> 26337
[20-Oct-2011 11:58:45] * aclark waves
[20-Oct-2011 11:58:56] <enkrypt> after updating some collector parameters (namely hostname and renderurl) of a collector via zendmd, do I need to restart something on the hub? because after doing that, my zenoss web interface acts funky, not showing anymore the details of the assets, neither graphs
[20-Oct-2011 12:00:01] [disconnected at Thu Oct 20 12:00:01 2011]
[20-Oct-2011 12:00:01] [connected at Thu Oct 20 12:00:01 2011]
[20-Oct-2011 12:00:13] [zenoss-logger (logger bot) has joined #zenoss]
[20-Oct-2011 12:07:02] <luizzmizz> sorry for asking for third time, but... anyone suspecting which is my problem in: message/61748#61748
[20-Oct-2011 12:10:23] <luizzmizz> (that's what i hear: http://bit.ly/am52iK :S)
[20-Oct-2011 12:10:46] <Hackman238> luizzmizz: su zenoss, zendmd, reindex(), commit()
[20-Oct-2011 12:10:53] <Hackman238> luizzmizz: Let me know if that works
[20-Oct-2011 12:11:02] <luizzmizz> did it lots of times
[20-Oct-2011 12:11:23] <luizzmizz> it gets reindexed perfectly
[20-Oct-2011 12:11:49] <luizzmizz> but when making a change (adding a ipinterface, modifying a hwtag)
[20-Oct-2011 12:12:10] <luizzmizz> it doesn't get reflected on the deviceSearch/ipSearch/whateverCatalog...
[20-Oct-2011 12:12:15] <Hackman238> luizzmizz: So the top search doesnt register the new ipinterface when you seach?
[20-Oct-2011 12:12:23] <luizzmizz> nope
[20-Oct-2011 12:12:32] <Hackman238> luizzmizz: 1 sec
[20-Oct-2011 12:13:27] <luizzmizz> Hackman238: and when listing the ipadressess at infrastructure/networks... it shows the ip, but not the device and interface connected... unless you do another reindex()
[20-Oct-2011 12:14:01] <luizzmizz> Hackman238: it's just like for some reason the internal calls to updateCatalogs didn't work...
[20-Oct-2011 12:14:24] <Hackman238> luizzmizz: vi /opt/zenoss/Products/Zuul/catalog/global_catalog.py, line 274, add import pdb; pdb.set_trace(), save, zopectl restart, zenhub restart, zendmd, reindex()
[20-Oct-2011 12:14:28] <luizzmizz> Hackman238: my install begun with a zenbatchload, that's the only different thing i did
[20-Oct-2011 12:14:37] <luizzmizz> i'm on it
[20-Oct-2011 12:15:06] <Hackman238> luizzmizz: on line 274, make sure the line is the same indent as the 'adapts(IpInterface) above it
[20-Oct-2011 12:15:49] <luizzmizz> :O
[20-Oct-2011 12:17:26] <luizzmizz> when doing zenhub restart...
[20-Oct-2011 12:17:36] <luizzmizz> it gets into a (Pdb) prompt
[20-Oct-2011 12:17:50] <luizzmizz> (Hackman238)
[20-Oct-2011 12:17:57] <Hackman238> luizzmizz: Okay good
[20-Oct-2011 12:18:05] <Hackman238> luizzmizz: that means its being called.
[20-Oct-2011 12:18:13] <luizzmizz> ok
[20-Oct-2011 12:18:22] <Hackman238> luizzmizz: break out of it, undo that line and restart zope and zenhub
[20-Oct-2011 12:18:52] <luizzmizz> done
[20-Oct-2011 12:20:25] <Hackman238> zendmd
[20-Oct-2011 12:20:26] <Hackman238> from Products.ZCatalog.ProgressHandler import StdoutHandler
[20-Oct-2011 12:20:27] <Hackman238> dmd.Networks.ipSearch.refreshCatalog(clear=1,pghandler=StdoutHandler())
[20-Oct-2011 12:20:36] <Hackman238> luizzmizz: once done, commit()
[20-Oct-2011 12:21:46] <luizzmizz> its on 69%...
[20-Oct-2011 12:21:51] <luizzmizz> its on 100%
[20-Oct-2011 12:21:55] <luizzmizz> it did alright
[20-Oct-2011 12:22:08] <Hackman238> okay, commit() and CTRL + D
[20-Oct-2011 12:22:09] <luizzmizz> hackman238: commited!
[20-Oct-2011 12:22:17] <luizzmizz> (controldied)
[20-Oct-2011 12:22:31] <Hackman238> luizzmizz: see if ipSearch is updated now
[20-Oct-2011 12:22:59] <luizzmizz> it is
[20-Oct-2011 12:23:26] <luizzmizz> hackman238: the problem here is: updating or regenerating the catalog works alright
[20-Oct-2011 12:24:00] <luizzmizz> hackman238: the thing which does not happen is... those catalogs being updated after the creation of a device with an ipinterface, in example...
[20-Oct-2011 12:24:01] <Hackman238> luizzmizz: Gotcha. So when when you batchload items the catalog doesnt get updated?
[20-Oct-2011 12:24:09] <luizzmizz> nope
[20-Oct-2011 12:24:26] <Hackman238> luizzmizz: If you add a device in the UI does it get cataloged?
[20-Oct-2011 12:24:28] <luizzmizz> i batchloaded some devices at the beginning
[20-Oct-2011 12:25:09] <luizzmizz> if i discover/add a device via UI, the ipsearch doesn't show the entry for its ipinterface
[20-Oct-2011 12:25:39] <luizzmizz> and via zendmd i can see that dev.os.interfaces()[0] and its ipaddress
[20-Oct-2011 12:25:58] <luizzmizz> it's just like there's no call to cata
[20-Oct-2011 12:26:18] <Hackman238> luizzmizz: Thats odd
[20-Oct-2011 12:26:25] <luizzmizz> to catalog update once it's needed (after a change that must trigger it)
[20-Oct-2011 12:26:26] <Hackman238> luizzmizz: lets check something
[20-Oct-2011 12:26:34] <Hackman238> luizzmizz: as zenoss, zenmigrate
[20-Oct-2011 12:26:52] <luizzmizz> zenmigrate like that
[20-Oct-2011 12:26:56] <luizzmizz> no other options...?
[20-Oct-2011 12:26:56] <Hackman238> luizzmizz: Yep
[20-Oct-2011 12:27:02] <Hackman238> luizzmizz: yep
[20-Oct-2011 12:27:11] <Hackman238> luizzmizz: what steps does it run?
[20-Oct-2011 12:27:13] <luizzmizz> should i backup? (sorry, the demo is next week)
[20-Oct-2011 12:27:28] <Hackman238> luizzmizz: Yes please, backup your Data.fs
[20-Oct-2011 12:27:47] <Hackman238> luizzmizz: Its not dangerous operation, but backup never hurts
[20-Oct-2011 12:30:32] <luizzmizz> hackman238: ok done
[20-Oct-2011 12:30:36] <luizzmizz> hackman238: zenmigrate done
[20-Oct-2011 12:31:00] <luizzmizz> (INFO:zen.migrate:Loading Reports)
[20-Oct-2011 12:31:01] <Hackman238> luizzmizz: what steps did it run?
[20-Oct-2011 12:31:12] <luizzmizz> hackman238: that's the only step
[20-Oct-2011 12:31:23] <luizzmizz> "INFO:zen.migrate:Loading Reports"
[20-Oct-2011 12:31:26] <Hackman238> luizzmizz: Hum. ok. I was thinking maybe the catalog updates didnt get installed
[20-Oct-2011 12:31:41] <Hackman238> >>>
[20-Oct-2011 12:31:41] <Hackman238> [zenoss@localhost root]$ zenmigrate
[20-Oct-2011 12:31:41] <Hackman238> INFO:zen.migrate:Database going to version Zenoss 3.1.70
[20-Oct-2011 12:31:41] <Hackman238> INFO:zen.migrate:Installing DeviceSearchCatalogUpdate (3.1.1)
[20-Oct-2011 12:31:41] <Hackman238> INFO:zen.migrate:Installing IpSearchCatalogUpdate (3.1.1)
[20-Oct-2011 12:31:44] <Hackman238> INFO:zen.migrate:Updating IpAddress Catalog
[20-Oct-2011 12:31:46] <Hackman238> INFO:zen.migrate:Finished total time=0.02 rate=120.44 count=2
[20-Oct-2011 12:31:48] <Hackman238> ^ like so
[20-Oct-2011 12:32:09] <luizzmizz>
[20-Oct-2011 12:32:16] <luizzmizz> i only have one line...
[20-Oct-2011 12:32:16] <Hackman238> luizzmizz: what version?
[20-Oct-2011 12:32:22] <luizzmizz> 3.2.1
[20-Oct-2011 12:32:26] <Hackman238> luizzmizz: yeah that means all the other steps are done
[20-Oct-2011 12:32:31] <luizzmizz> nono
[20-Oct-2011 12:32:33] <luizzmizz> 3.2.0
[20-Oct-2011 12:34:34] <luizzmizz> hackman238: i've been looking at it for days... and never found any error after creating a device
[20-Oct-2011 12:35:26] <luizzmizz> hackman238: the install is pretty nice, 1236 devices (the Cisco Phones moved via changeDeviceClass and not moveObject :D)
[20-Oct-2011 12:35:40] <Hackman238> luizzmizz: zenmigrate --level=3.0.0
[20-Oct-2011 12:35:44] <luizzmizz> hackman238: but it happens THIS and i cannot explain it...
[20-Oct-2011 12:35:45] <luizzmizz> ok
[20-Oct-2011 12:35:46] <Hackman238> luizzmizz: It will take a while
[20-Oct-2011 12:36:10] <Hackman238> luizzmizz: Nice
[20-Oct-2011 12:36:50] <luizzmizz> AddWinServiceClass(3.0.0),AddWinServiceProperties(3.0.0)...
[20-Oct-2011 12:37:58] <luizzmizz> hackman238: i'm on speedupglobalcatalog(3.0.3)...
[20-Oct-2011 12:39:00] <luizzmizz> updateing IpAddressCatalog...
[20-Oct-2011 12:42:32] <luizzmizz> it died on installing rebuiildpathindex(3.1.1) by a conflicterror...
[20-Oct-2011 12:42:55] <luizzmizz> i'm gonna shutdown some daemons and repeat the zenmigrate from --level 3.1.1, ok hackman238?
[20-Oct-2011 12:59:05] <aclark> anyone seen devicedetail throwing KeyError 'Devices' on 3.2.0?
[20-Oct-2011 12:59:24] <Hackman238> luizzmizz: What was the message?
[20-Oct-2011 12:59:39] <luizzmizz> sorry i lost it
[20-Oct-2011 12:59:42] <Hackman238> aclark: Can where are you seeing that?
[20-Oct-2011 13:00:05] <Hackman238> luizzmizz: zenmigrate --level=3.1.1
[20-Oct-2011 13:00:09] <luizzmizz> but was a conflict, i have a cron hourly for modeling some phones :S...
[20-Oct-2011 13:00:14] <luizzmizz> did it already
[20-Oct-2011 13:00:16] <luizzmizz> just finished
[20-Oct-2011 13:00:22] <Hackman238> luizzmizz: Ah ok
[20-Oct-2011 13:00:25] <luizzmizz> i was about to restart and try
[20-Oct-2011 13:00:32] <Hackman238> luizzmizz: gotcha
[20-Oct-2011 13:01:38] <aclark> Hackman238: oh weird. it only happens on a vhost. if i ssh -L to the host and check http://localhost:8080/zport/dmd/Devices/Discovered/devices/device0/devicedetail#deviceDetailNav:device_overview
[20-Oct-2011 13:01:42] <aclark> it works
[20-Oct-2011 13:02:13] <aclark> but if i go in through the vhost e.g. myhost.com/Devices… i get the error
[20-Oct-2011 13:02:30] <aclark> i wonder if this is a bug or i screwed up the vhosting ;-)
[20-Oct-2011 13:03:04] <aclark> at least it's not "real" breakage
[20-Oct-2011 13:04:12] <luizzmizz> hackman238: it's strange...
[20-Oct-2011 13:07:22] <luizzmizz> http://prbrpresidl07:8080/zport/dmd/Devices/deviceSearch/manage_objectInformation?rid=884334638 is the entry of a device in the deviceSearch catalog...
[20-Oct-2011 13:07:49] <luizzmizz> and when I modify the hwtag of the device, it gets correctly reflected on the device catalog entry ...
[20-Oct-2011 13:08:04] <luizzmizz> but the device search box doesn't work for the new string on hwtag...
[20-Oct-2011 13:08:11] <luizzmizz> (hackman238)
[20-Oct-2011 13:08:40] kocolosk|away is now known as kocolosk
[20-Oct-2011 13:08:42] <luizzmizz> the server is internal, i should have "sed -e"-it
[20-Oct-2011 13:09:37] <aclark> Hackman238: anyway, it happens on any infrastructure -> device on the vhosted app; don't see anything obvious as to why: https://gist.github.com/1301692
[20-Oct-2011 13:14:12] <nexex> There are 100+ mibs I don't need anymore, can I just delete the directory the are in /opt/zenoss/share/mibs/hpn -- do I need to run zenmib or anything after that?
[20-Oct-2011 13:18:13] <Hackman238> luizzmizz: Hum
[20-Oct-2011 13:18:18] <Hackman238> luizzmizz: still no good?
[20-Oct-2011 13:18:25] <Hackman238> aclark: Honestly not sure
[20-Oct-2011 13:18:26] <luizzmizz> nope
[20-Oct-2011 13:18:43] <Hackman238> nexex: You'll need to remove them from zope
[20-Oct-2011 13:18:46] <luizzmizz> hackman238: and the ipsearch catalog doesn't get updated either...
[20-Oct-2011 13:19:30] <Hackman238> luizzmizz: pop back on tommorrow, I'll check with a developer on that issue
[20-Oct-2011 13:19:51] <luizzmizz> details {"device": null, "interface": null, "interfaceDescription": null, "macAddress": null}
[20-Oct-2011 13:20:03] <luizzmizz> ok thanks for your time, hackman
[20-Oct-2011 13:20:29] <Hackman238> luizzmizz: anytime
[20-Oct-2011 13:20:42] <luizzmizz> bye!
[20-Oct-2011 13:20:53] <Hackman238> Later
[20-Oct-2011 13:56:49] <aclark> mattray!
[20-Oct-2011 13:57:05] <mattray> yup
[20-Oct-2011 14:16:59] <dhopp> anybody know of a zenpack that will show the event console as a portlet? I know I could do a site window but then I get the header in the site window as well
[20-Oct-2011 14:17:29] <dhopp> is mattray the "Norm" of #zenoss ?
[20-Oct-2011 14:17:31] <dhopp> :-
[20-Oct-2011 14:17:33] <dhopp> :-)
[20-Oct-2011 14:18:34] * mattray resolves to drink less beer
[20-Oct-2011 14:24:57] <dhopp> mattray: what fun is that/
[20-Oct-2011 14:24:57] <dhopp> ?
[20-Oct-2011 14:25:06] <dhopp> God I can't type (and no I haven't been drinking)
[20-Oct-2011 14:35:43] <tvincent> Anyone know of a membase zenpack?
[20-Oct-2011 14:38:24] <cgibbons> mmm beer
[20-Oct-2011 14:55:01] <Hackman238> ZCA ZenPack wishlist poll- polls/1041 - Please vote and post comments!
[20-Oct-2011 14:55:24] bender[a] is now known as bender[w]
[20-Oct-2011 15:22:02] <aclark> so i have some hosts being monitored with public snmp, this typically does not produce any interesting data for graphing correct?
[20-Oct-2011 15:23:15] <pipeline> depends on how dumb your snmp config is
[20-Oct-2011 15:23:44] <Hackman238> aclark: networking info, hardware info, cpu and mem are generally open
[20-Oct-2011 15:23:54] <aclark> snmpd on ubuntu with the defaults
[20-Oct-2011 15:24:04] <Hackman238> aclark: depends on the device, agent implementation and the config as pipline said
[20-Oct-2011 15:24:04] <aclark> ah, hmmm
[20-Oct-2011 15:24:20] <Hackman238> aclark: What type of device?
[20-Oct-2011 15:25:17] <aclark> Hackman238: all linux boxes on ec2
[20-Oct-2011 15:26:48] <Hackman238> aclark: Lots of interesting data is available
[20-Oct-2011 15:27:08] <Hackman238> aclark: filesystems, volumes, cpu use, memory use, interface list, interface usages, disk io
[20-Oct-2011 15:27:15] <Hackman238> aclark: and more.
[20-Oct-2011 15:27:43] <dhopp> Hackman238: but wait there's more!
[20-Oct-2011 15:28:03] <dhopp> oh sorry thought you were an infomerical ;-)
[20-Oct-2011 15:28:51] <Hackman238> dhopp: LOL
[20-Oct-2011 15:29:00] <Hackman238>
[20-Oct-2011 15:29:44] <aclark> heh
[20-Oct-2011 15:30:05] <aclark> well right now i just batch loaded some devices… waiting to see what happens :-)
[20-Oct-2011 15:30:30] <Hackman238> aclark: make sure snmp enabled on your vms
[20-Oct-2011 15:30:35] <aclark> it is
[20-Oct-2011 15:30:38] <aclark> i just puppetized it
[20-Oct-2011 15:32:59] <aclark> Hackman238: how long does a "first data" run take? i can recall seeing graphs with nothing in them before vs. "no graph data"
[20-Oct-2011 15:36:29] <dhopp> aclark: no graph data implies that no template that has any graphs defined is assigned to the device/component
[20-Oct-2011 15:37:23] <aclark> dhopp: ah! i should put them in the right class
[20-Oct-2011 15:37:30] <aclark> dhopp: instead of /Discovered
[20-Oct-2011 15:37:48] <dhopp> aclark: that will do it
[20-Oct-2011 15:38:13] <aclark> dhopp: thanks
[20-Oct-2011 15:45:48] <aclark> ok now i have empty graphs :-)
[20-Oct-2011 15:47:14] <aclark> hmmm i must be missing some mibs
[20-Oct-2011 15:53:01] <aclark> ah, using v2 helps
[20-Oct-2011 18:01:03] kocolosk is now known as kocolosk|away
[20-Oct-2011 18:19:09] kocolosk|away is now known as kocolosk
[21-Oct-2011 00:00:01] [disconnected at Fri Oct 21 00:00:01 2011]
[21-Oct-2011 00:00:01] [connected at Fri Oct 21 00:00:01 2011]
[21-Oct-2011 00:00:01] [disconnected at Fri Oct 21 00:00:01 2011]
[21-Oct-2011 00:00:02] [connected at Fri Oct 21 00:00:02 2011]
[21-Oct-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[21-Oct-2011 03:12:26] <ali3n0> hi guys. I'm hitting a quite strange problem: if I use the IP address instead of the FQDN of my zenoss instance, I can't see graph displayed
[21-Oct-2011 03:12:48] <ali3n0> sorry, viceversa, if I use IP it does work, otherwise it doesn't
[21-Oct-2011 03:39:01] <fragfutter> log into the zenoss server, check output of "host fqdn". check that it resolves to the IP.
[21-Oct-2011 05:01:48] <aberrios> I've created a graph with a data point that is a command. The command runs fine when testing against the device but no data seems to be being gathered/graphed.
[21-Oct-2011 05:48:09] <ali3n0> fragfutter tnx for your reply. doesn't work if I've got it right into /etc/hosts ? it has to be a DNS respose?
[21-Oct-2011 05:54:47] <fragfutter> ali3n0: i have no idea how zenoss handles it, but i would make sure that dns and host match.
[21-Oct-2011 05:55:05] <ali3n0> fragfutter ok thanks for the hint
[21-Oct-2011 05:55:25] <ali3n0> I've posted on forum too in the meanwhile: thread/16929
[21-Oct-2011 07:43:10] <enkrypt> Hackman238 jmp242 etc: I often see in my zenhub.log that there are problems with the network topology. I get such error messages: 2011-10-21 12:01:09,061 INFO zen.ZenHub: Worker reports 2011-10-21 12:01:09,061 INFO zen.hub: ZenPing 'zenoss1.cegeka.be' not found, ignoring network topology.
[21-Oct-2011 07:43:27] <enkrypt> and I get those for zenoss1.cegeka.be until zenoss6.cegeka.be, which are in fact all my zenoss servers
[21-Oct-2011 07:44:03] <enkrypt> which is strange because all of them are pingable, and all of them are reachable by DNS and reverse DNS. also, they are simply registered in /etc/hosts to not depend on DNS servers at all. but still they get ignored for the topology.
[21-Oct-2011 08:32:06] kocolosk is now known as kocolosk|away
[21-Oct-2011 08:42:28] kocolosk|away is now known as kocolosk
[21-Oct-2011 09:09:09] <Hackman238> Morning all!
[21-Oct-2011 09:40:07] <JohnnyNOC> Shane Scott
[21-Oct-2011 09:40:09] <JohnnyNOC> doing big things
[21-Oct-2011 09:43:02] <JohnnyNOC> knock, knock
[21-Oct-2011 09:43:27] <Hackman238> JohnnyNOC: Hey man
[21-Oct-2011 09:43:36] <Hackman238> JohnnyNOC: Web cient work?
[21-Oct-2011 09:43:49] <JohnnyNOC> yea!
[21-Oct-2011 09:43:53] <JohnnyNOC> i don't know why I didn't even consider it
[21-Oct-2011 09:47:06] <Hackman238> JohnnyNOC: LOL
[21-Oct-2011 09:47:20] <Hackman238> JohnnyNOC: How's it been going man?
[21-Oct-2011 09:47:51] <JohnnyNOC> going well, can't complain really.. still getting more familiar with zenoss
[21-Oct-2011 09:47:56] <JohnnyNOC> lots of questions still though
[21-Oct-2011 09:48:05] <JohnnyNOC> we hired a datacenter tech to do a lot of the hard work in the DCs
[21-Oct-2011 09:48:10] <JohnnyNOC> he's become my minion, i enjoy it
[21-Oct-2011 09:48:12] <JohnnyNOC>
[21-Oct-2011 09:49:03] <jb> erm
[21-Oct-2011 09:49:43] <jb> man, I have this odd traceback on all of my devices for zenoss.snmp.HRFileSystemMap now. http://pastebin.com/X5wPKRtA
[21-Oct-2011 09:49:50] <jb> look familiar to anyone?
[21-Oct-2011 09:50:01] <jb> "Problem while executing plugin zenoss.snmp.HRFileSystemMap" is the event
[21-Oct-2011 09:50:53] <Sashness> has anyone seen an issue where when zencommand runs the rrd file isn't updated even though the debug logs says it saved it to the file. Running the command by hand actually stores the data
[21-Oct-2011 09:53:02] <JohnnyNOC> Hackman238: whats your halloween costume this year
[21-Oct-2011 09:53:23] <psyklone> jb - you have an invalid regex in one of the filesystem zproperties
[21-Oct-2011 09:53:51] <jb> psyklone: bingo.
[21-Oct-2011 09:54:03] <psyklone> what do i win?
[21-Oct-2011 09:54:26] <jb> hah
[21-Oct-2011 09:54:27] <jb> (/Admin|/Admin-Out|/Admin-DBA|/Oracle|/oracle|/cdrom|/mnt/cdrom|/www/repos*)
[21-Oct-2011 09:54:35] <jb> i was trying to ignore "/www/repos/*"
[21-Oct-2011 09:54:40] <jb> obviously, thats not the correct regex for that.
[21-Oct-2011 10:01:59] <Hackman238> JohnnyNOC: Nice! always fun to have aminon to do your bidding
[21-Oct-2011 10:02:48] <Hackman238> JohnnyNOC: Not sure I'll have time for halloween this year. Just too busy. Zombie Steve Jobs would be good though
[21-Oct-2011 10:22:37] <jb> any idea what regexp to use for that?
[21-Oct-2011 10:24:41] <fragfutter> jb: |/www/repos/.*|
[21-Oct-2011 10:34:38] <dhopp> hmmm interesting…I have a server with 2 OS processes…one is just httpd and one is /usr/sbin/httpd
[21-Oct-2011 10:34:43] <dhopp> now to figure out why
[21-Oct-2011 10:45:10] <jb> fragfutter: thanks!
[21-Oct-2011 10:59:38] <kerick> does an event transform require more than code in the transform field?
[21-Oct-2011 10:59:48] <jmp242> no
[21-Oct-2011 11:01:11] <kerick> hmmm, i have a simple if statement that checks device.customproperty, if true it sets evt.customproperty
[21-Oct-2011 11:01:34] <kerick> then i am trying to print it in the alert message....which fails
[21-Oct-2011 11:02:08] <tightwork> How could I install the new version of Predictive Threshold docs/DOC-3408.. There is no egg?
[21-Oct-2011 11:02:54] <tightwork> Would I place the files somewhere and run a zendoit command?
[21-Oct-2011 11:06:52] <jmp242> oh, you use a zenpack --install --link or something like that
[21-Oct-2011 11:06:59] <jmp242> Give me a bit I'll find the syntax
[21-Oct-2011 11:08:08] <rmatte> Or you could just ask Jane Curry for the egg as I think it's her who updated it.
[21-Oct-2011 11:08:16] <rmatte> I'm surprised the egg isn't up there
[21-Oct-2011 11:08:41] <humanix> hi rmatte Hackman238
[21-Oct-2011 11:08:42] <humanix>
[21-Oct-2011 11:09:02] <jmp242> cd $ZENHOME/local
[21-Oct-2011 11:09:02] <jmp242> copy the tar bundle to this directory
[21-Oct-2011 11:09:02] <jmp242> tar -xvf <tar bundle file> - this should create a subdirectory
[21-Oct-2011 11:09:02] <jmp242> called ZenPacks.community.PredictiveThreshold .
[21-Oct-2011 11:09:02] <jmp242> zenpack --link --install ZenPacks.community.PredictiveThreshold
[21-Oct-2011 11:09:03] <jmp242> zenhub restart
[21-Oct-2011 11:09:03] <jmp242> zopectl restart
[21-Oct-2011 11:09:16] <jmp242> Jane seemed to not want to use the .egg for some reason
[21-Oct-2011 11:09:20] <jmp242> and prefered this
[21-Oct-2011 11:13:38] <tightwork> thats cool thanks a lot... cant wait to get my network under control
[21-Oct-2011 11:14:16] <tightwork> so many false positives using ipswitch, its pretty much worthless.
[21-Oct-2011 11:14:23] <humanix> how are you rmatte ?
[21-Oct-2011 11:16:37] <cgibbons> hmmm i can't remember how to get path working in a zenpack
[21-Oct-2011 11:18:51] <cgibbons> xpath even
[21-Oct-2011 11:32:08] <JohnnyNOC> anyone have experience creating custom device reports?
[21-Oct-2011 11:32:25] <JohnnyNOC> i'd like to create one that has the max power drawn by our APCs, but I'm unsure how I would capture that value and bring it into the report
[21-Oct-2011 11:32:27] <JohnnyNOC> :/
[21-Oct-2011 11:35:31] <rmatte> JohnnyNOC: I do, yes
[21-Oct-2011 11:35:51] <rmatte> JohnnyNOC: though you could code one just as easily
[21-Oct-2011 11:36:01] <rmatte> just use one of the existing reports as an example
[21-Oct-2011 11:36:06] <rmatte> the code is actually really simple
[21-Oct-2011 11:36:47] <rmatte> Hackman238: speaking of reports and such, planning to release that updated IPSLA pack soon?
[21-Oct-2011 11:36:58] <JohnnyNOC> rmatte is this not possible by simply editing a report in the GUI?
[21-Oct-2011 11:37:44] <rmatte> JohnnyNOC: you can try to use the custom reports option in the GUI, but in my experience it's more of a pain in the ass to work with than just to actually code one
[21-Oct-2011 11:37:52] <JohnnyNOC> *nod*
[21-Oct-2011 11:38:06] <rmatte> It's fine for very simple reports, but when you start getting in to RRD values it gets tougher
[21-Oct-2011 11:38:28] <Hackman238> rmatte: Very soon. Your reports are in there and work well!
[21-Oct-2011 11:38:41] <rmatte>
[21-Oct-2011 11:38:46] <Hackman238> JohnnyNOC: Zombie Steve Jobs constume
[21-Oct-2011 11:39:00] <Hackman238> *costume
[21-Oct-2011 11:39:28] <JohnnyNOC> Hackman238: haha are those turtlenecks still in stock?
[21-Oct-2011 11:39:56] <Hackman238> JohnnyNOC: I'll have to find one...cant mock steve jobs without one!
[21-Oct-2011 11:40:50] <JohnnyNOC> rmatte: what are you doing for halloween
[21-Oct-2011 11:42:45] <rmatte> not sure yet
[21-Oct-2011 11:43:01] <rmatte> I have a v for vendetta costume laying around that I haven't used yet
[21-Oct-2011 11:43:03] <rmatte> might just do that
[21-Oct-2011 11:43:16] <jmp242> I don't do halloween myself but
[21-Oct-2011 11:43:46] <Hackman238> jmp242: Please dont say you're the guy witht he super soaker filled with pee
[21-Oct-2011 11:43:54] <Hackman238> LOL
[21-Oct-2011 11:46:50] <Hackman238> (guess that wasn't funny)
[21-Oct-2011 11:48:22] <jmp242> no
[21-Oct-2011 11:48:26] <jmp242> I just stay home
[21-Oct-2011 11:48:35] <jmp242> I get enough candy from WalMart lol
[21-Oct-2011 11:48:41] <Hackman238> jmp242: LOL
[21-Oct-2011 11:49:47] <JohnnyNOC> ahahaha
[21-Oct-2011 11:50:06] <JohnnyNOC> my gf is getting really into the season.. starting to see pumpkin themed everythin gin the kitchen
[21-Oct-2011 11:50:49] <Hackman238> JohnnyNOC: ...its like summer in San Antonio comparred to the North-Eastern Fall.
[21-Oct-2011 11:51:19] <Hackman238> JohnnyNOC: I don't think I've even seen a real pumpkin in San Antonio that wasn't part of a pie
[21-Oct-2011 11:51:24] <Hackman238> LOL
[21-Oct-2011 11:52:38] <JohnnyNOC> hahaha
[21-Oct-2011 11:52:41] <JohnnyNOC> i hope she makes a pumpkin pie
[21-Oct-2011 11:52:58] <JohnnyNOC> she saw some post on reddit and put these crazy cardboard cutouts in our front windows as decorations
[21-Oct-2011 11:53:03] <JohnnyNOC> she's crazy
[21-Oct-2011 11:53:06] <JohnnyNOC> women are crazy
[21-Oct-2011 11:53:36] <Hackman238> JohnnyNOC: Hard to beat pumpkin pie
[21-Oct-2011 11:53:45] <Hackman238> JohnnyNOC: You said it not I. XD
[21-Oct-2011 11:57:57] <rmatte> 3.2.1 officially released
[21-Oct-2011 11:58:29] <psyklone> i halloweened out a co-worker's cube when he was out of the office a couple of weeks back. so many cobwebs he couldn't get into it
[21-Oct-2011 12:00:01] [disconnected at Fri Oct 21 12:00:01 2011]
[21-Oct-2011 12:00:02] [connected at Fri Oct 21 12:00:02 2011]
[21-Oct-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[21-Oct-2011 12:01:10] <JohnnyNOC> so i know manually creating the device report is preferred, but while i'm poking around in zendmd i'm wondering how to get an rrd value using getRRDValue
[21-Oct-2011 12:01:19] <JohnnyNOC> no one has suggestions? evewrything i'm trying is failing miserably
[21-Oct-2011 12:02:52] <Hackman238> psyklone: LOL! Nice
[21-Oct-2011 12:03:08] <Hackman238> JohnnyNOC: 1 second
[21-Oct-2011 12:03:13] <JohnnyNOC> i've tried datapoint names, full paths to the rrd
[21-Oct-2011 12:05:54] <at0m1sk> Hello, can I pick someone's brain about an issue I have been having after upgrading from 2.5.2 -> 3.2.0?
[21-Oct-2011 12:06:08] <psyklone> at0m1sk, shoot
[21-Oct-2011 12:07:23] <Hackman238> JohnnyNOC: I can't recall how to do it.
[21-Oct-2011 12:07:28] <Hackman238> JohnnyNOC: rmatte knows
[21-Oct-2011 12:07:41] <at0m1sk> Mostly everything went pretty smooth , but... most of my rrd graphs , example proc count/mem usage/cpu usage, are showing NaN. Is there anyway I can debug the RRD graph building?
[21-Oct-2011 12:08:14] <at0m1sk> I am not an export on the matter and I am not sure how zenoss converts data from it's RRD files to the graphs I see through the web interface
[21-Oct-2011 12:08:21] <JohnnyNOC> Hackman238: he suggested just write everything custom
[21-Oct-2011 12:08:22] <JohnnyNOC>
[21-Oct-2011 12:08:24] <JohnnyNOC> i'll keep poking around
[21-Oct-2011 12:09:21] <psyklone> at0m1sk, have you patched 3.2 at all or are you at a base install of 3.2?
[21-Oct-2011 12:10:07] <at0m1sk> I read about some patches and I applied some but let me check really quickly
[21-Oct-2011 12:10:09] <JohnnyNOC> Hackman238: i'm making progress.. getting the value out of a datapoint now, however i want to know the MAX
[21-Oct-2011 12:10:20] <Hackman238> JohnnyNOC: Looks like if d=dmd.Devices.findDevice('yourDevice') you can data=d.getRRDValue('nameOfDataPoint)
[21-Oct-2011 12:10:34] <JohnnyNOC> *nod*
[21-Oct-2011 12:10:40] <JohnnyNOC> now how do i get the MAX is my next question
[21-Oct-2011 12:10:40] <JohnnyNOC>
[21-Oct-2011 12:10:46] <at0m1sk> psyklone, I have 3 patches
[21-Oct-2011 12:11:36] <Hackman238> JohnnyNOC: data=d.getRRDValue('nameOfDataPoint','MAXIMUM')
[21-Oct-2011 12:11:38] <psyklone> at0m1sk, ok, cool. what kind of device?
[21-Oct-2011 12:12:15] <Hackman238> JohnnyNOC: Maybe. Try it
[21-Oct-2011 12:12:27] <at0m1sk> psyklone, i have patch 26334,26335,26337
[21-Oct-2011 12:12:31] <JohnnyNOC> didn't work
[21-Oct-2011 12:12:31] <JohnnyNOC>
[21-Oct-2011 12:12:43] <JohnnyNOC> i tried max, Max, Maximum, MAXIMUM
[21-Oct-2011 12:12:44] <JohnnyNOC> :[
[21-Oct-2011 12:13:03] <at0m1sk> are those all of the ones I need, or are there others I need apply?
[21-Oct-2011 12:13:05] <Hackman238> JohnnyNOC: Damn. grep 'def getRRDValue' %ZENHOME/Products/* -R
[21-Oct-2011 12:13:19] <Hackman238> JohnnyNOC: Look at the method to see what it takes
[21-Oct-2011 12:14:35] <JohnnyNOC> not sure i udnerstand it though
[21-Oct-2011 12:14:39] <JohnnyNOC> def getRRDValue(self, dsname, start=None, end=None, function="LAST",
[21-Oct-2011 12:15:20] <JohnnyNOC> the man
[21-Oct-2011 12:15:24] <JohnnyNOC> the myth
[21-Oct-2011 12:15:26] <JohnnyNOC> Shane William Scott
[21-Oct-2011 12:15:30] <JohnnyNOC> comes through again
[21-Oct-2011 12:15:32] <JohnnyNOC> >>> print d.getRRDValue('Aggregate', start=None, end=None, function="MAXIMUM")
[21-Oct-2011 12:15:35] <JohnnyNOC> that's what i needed
[21-Oct-2011 12:15:52] <JohnnyNOC> >>> print d.getRRDValue('Aggregate', function="MAXIMUM")
[21-Oct-2011 12:15:56] <JohnnyNOC> will do
[21-Oct-2011 12:16:48] <Hackman238> JohnnyNOC: Win
[21-Oct-2011 12:17:43] <psyklone> at0m1sk, what kind of device?
[21-Oct-2011 12:18:13] <at0m1sk> A linux server, centos 5.6
[21-Oct-2011 12:21:43] <JohnnyNOC> Hackman238: darn
[21-Oct-2011 12:21:54] <JohnnyNOC> name 'getRRDValue' is not defined
[21-Oct-2011 12:21:56] <JohnnyNOC> in my report
[21-Oct-2011 12:21:56] <JohnnyNOC> :/
[21-Oct-2011 12:22:32] <JohnnyNOC> woot
[21-Oct-2011 12:22:40] <JohnnyNOC> i needed to prefix it with dev.
[21-Oct-2011 12:23:00] <Hackman238> JohnnyNOC: Yep LOL
[21-Oct-2011 12:23:12] <Hackman238> Be back in a bit
[21-Oct-2011 12:33:18] <Sashness> So i can't figure this out and none of the logs show any issues. zencommand seems to work but none of the RRD files are updated. Running the same command zencommand says it runs from the logs works and i see the rrd file updated. zensnmpperf works just fine. Any ideas?
[21-Oct-2011 12:49:43] <rmatte> Sashness: what version of Zenoss?
[21-Oct-2011 12:50:12] <Sashness> 3.1 core
[21-Oct-2011 12:50:45] <rmatte> was it you coding the custom parser?
[21-Oct-2011 12:51:41] <Sashness> no. i was setting up a transition to convert an alert text
[21-Oct-2011 12:52:19] <rmatte> you mean a transform
[21-Oct-2011 12:52:27] <Sashness> yeah sorry. transform
[21-Oct-2011 12:52:30] <rmatte> hehe
[21-Oct-2011 12:52:47] <rmatte> ok, what does that have to do with your RRD problems?
[21-Oct-2011 12:53:03] <Sashness> heh hopefully nothing.
[21-Oct-2011 12:53:30] <Sashness> zencommand broke after i verified the the transform
[21-Oct-2011 12:53:41] <Sashness> but now ic an't figure out wtf is going on with zencommand
[21-Oct-2011 12:53:49] <Sashness> the best type of problem. silent error
[21-Oct-2011 12:54:28] <rmatte> you've tried running it with -v10?
[21-Oct-2011 12:54:47] <rmatte> zencommand run -v10
[21-Oct-2011 12:55:06] <Sashness> yeah. doesnt show a problem it actually shows the commands running and says its updating the rrd file but if you look at the rrd file its not updated
[21-Oct-2011 12:55:14] <Sashness> running the same command it output in the logs works
[21-Oct-2011 12:55:59] <rmatte> how can you tell it's not updated?
[21-Oct-2011 12:56:07] <rmatte> how are you looking at it?
[21-Oct-2011 12:59:58] <Sashness> rrdtool dump
[21-Oct-2011 13:10:42] <tightwork> The advanced device detail pack, this seems only available with dell hardware? there is no way to pull in the hardware info from generic Linux devices with just snmp?
[21-Oct-2011 13:11:27] <pipeline> tightwork: there is and there isn't
[21-Oct-2011 13:11:44] <pipeline> snmp is hella extensible. You could stuff the output of e.g. dmidecode into an snmp node.
[21-Oct-2011 13:11:55] <pipeline> but the level of integration dell provides isn't going to be fun/easy to do on your own
[21-Oct-2011 13:12:14] <tightwork> yeah dmidecode, although im trying to avoid as many cruft scripts (extends) as possible
[21-Oct-2011 13:12:23] <pipeline> HP has the same sorta thing in the uhh PSPs ?
[21-Oct-2011 13:12:33] <pipeline> I don't know what it's called this week but it used to the Proliant Support Pack
[21-Oct-2011 13:24:35] <Sashness> funny enough when i connect to zencommand process with strace it freaking works. LOL
[21-Oct-2011 13:24:40] <Sashness> argh
[21-Oct-2011 13:25:01] <rmatte> strace prevents some crashes
[21-Oct-2011 13:25:14] <rmatte> I've seen that when using strace before
[21-Oct-2011 13:29:56] <Sashness> when starting or just attaching to an existing process
[21-Oct-2011 13:29:58] <Sashness> ?
[21-Oct-2011 13:46:39] <Sashness> so i quite the strace process i had going and ran zencommand debug
[21-Oct-2011 13:46:45] <Sashness> and watching the output now
[21-Oct-2011 13:47:34] <Sashness> should I grep -v DEBUG so i can see more important things?
[21-Oct-2011 13:59:44] <tightwork> What should I be aware of if I intent to monitor over 500 nodes? both config and hardware wise? snmp timeouts and polling perhaps? Current box is a quad xeon raid 10, 4GB Ram.
[21-Oct-2011 14:01:54] <tightwork> would distributed monitors also bring load balancing? That is each monitor should scan different nodes.
[21-Oct-2011 14:03:08] <rmatte> tightwork: the most important thing is disk IO
[21-Oct-2011 14:03:15] <rmatte> that's the biggest bottleneck when it comes to monitoring
[21-Oct-2011 14:03:31] <rmatte> also, what kinds of nodes are they?
[21-Oct-2011 14:03:37] <psyklone> ^ +1
[21-Oct-2011 14:04:02] <rmatte> tightwork: distributed monitoring does not do dynamic load balancing
[21-Oct-2011 14:04:06] <rmatte> you have to balance it out by hand
[21-Oct-2011 14:04:42] <tightwork> nodes are linux or windows nodes, plus the associated cisco switches.
[21-Oct-2011 14:05:37] <rmatte> tight work, you don't really need distributed collectors for that numbers...
[21-Oct-2011 14:05:47] <rmatte> I'm monitoring 455 nodes on a single instance just fine
[21-Oct-2011 14:05:53] <rmatte> and it's a mix of server and cisco gear
[21-Oct-2011 14:06:28] <rmatte> I'd recommend more RAM and the fastest disks that you can get for it though
[21-Oct-2011 14:06:28] <tightwork> thats good to know
[21-Oct-2011 14:06:40] <rmatte> my 455 device box has 6GB RAM
[21-Oct-2011 14:06:44] <rmatte> and is running off a fast SAN
[21-Oct-2011 14:07:21] <Hackman238> DFW DC 14,000 devices, 256GB RAM and fusion IO disks
[21-Oct-2011 14:07:36] <tightwork> woah
[21-Oct-2011 14:07:43] <rmatte> Hackman238: show-off
[21-Oct-2011 14:07:48] <Hackman238> Though v4.2 will halve those requirements
[21-Oct-2011 14:07:55] <Hackman238> It's that good.
[21-Oct-2011 14:08:02] <Hackman238> rmatte: Sorry, I couldnt resist
[21-Oct-2011 14:08:05] <Hackman238> rmatte: LOL
[21-Oct-2011 14:08:06] <rmatte> lol
[21-Oct-2011 14:08:08] <tightwork>
[21-Oct-2011 14:08:43] <Hackman238> http://www.youtube.com/watch?v=gl2NKZtl07c <- Song to live by
[21-Oct-2011 14:08:43] <rmatte> tightwork: my 455 devices work out to 25,000 SNMP requests per cycle
[21-Oct-2011 14:08:43] <tightwork> the san itself isnt a bottleneck, or you have a sas/infiniband link?
[21-Oct-2011 14:09:09] <rmatte> and it takes it about 40 seconds to complete a cycle
[21-Oct-2011 14:09:17] <tightwork> thats pretty nice
[21-Oct-2011 14:09:48] <rmatte> the san is connected via fiber and it's fast enough
[21-Oct-2011 14:10:02] <tightwork> true, so zen does not stagger the polling?
[21-Oct-2011 14:10:20] <rmatte> each daemon has it's own cycle
[21-Oct-2011 14:10:26] <rmatte> and they kick off at intervals
[21-Oct-2011 14:10:47] <rmatte> you can't tell it to poll one set of device's SNMP at 1 minute and one set at 5 unless you actually have different collectores
[21-Oct-2011 14:10:50] <rmatte> collectors*
[21-Oct-2011 14:11:19] <rmatte> not a big deal in my experience
[21-Oct-2011 14:11:28] <tightwork> I see, well I did see an option that it would only grab so many mibs at once?
[21-Oct-2011 14:11:45] <tightwork> My expierence is I have such a crappy system now I get nothing but false positives because of snmp loads and timeouts
[21-Oct-2011 14:12:33] <tightwork> so thats where my head is at
[21-Oct-2011 14:14:02] <rmatte> that's zMaxOIDPerRequest that you're referring to I assume?
[21-Oct-2011 14:14:40] <rmatte> It defaults to sending 40 OIDs per request, but some devices such as certain Fortigates and Cisco ASA son't like that many OIDs per request and choke
[21-Oct-2011 14:14:45] <rmatte> so I usually drop it to 10 for those
[21-Oct-2011 14:14:56] <rmatte> but you should leave it at 40 for the majority of devices
[21-Oct-2011 14:15:03] <rmatte> it's not harder on performance, it actually helps performance
[21-Oct-2011 14:15:39] <rmatte> also, MIBs are just files used to translate OIDs
[21-Oct-2011 14:15:46] <rmatte> in Zenoss they are only used to translate SNMP traps
[21-Oct-2011 14:15:59] <rmatte> Performance templates are designed with actual OID values
[21-Oct-2011 14:16:57] <tightwork> ah yes that makes sense since I had to use the OID for nsExtend
[21-Oct-2011 14:17:39] * rmatte nods
[21-Oct-2011 14:18:05] <rmatte> hmmm, time to give 3.2.1 a whirl
[21-Oct-2011 14:20:06] <humanix> guys
[21-Oct-2011 14:20:11] <humanix> how can i exclude a syslog event
[21-Oct-2011 14:20:42] <rmatte> humanix: with a transform... but it depends on how the syslog is structured when it comes in
[21-Oct-2011 14:20:48] <rmatte> does it have an eventClassKey?
[21-Oct-2011 14:21:16] <rmatte> (you don't necessarily need a transform, but you need an event mapping)
[21-Oct-2011 14:24:45] <humanix> i see
[21-Oct-2011 14:25:06] <humanix> it says eventclass /HW/Network
[21-Oct-2011 14:26:28] <rmatte> oh, it's already mapped then
[21-Oct-2011 14:26:33] <rmatte> if it's not /Unknown then it's mapped
[21-Oct-2011 14:26:39] <rmatte> so go find the mapping in /HW/Network
[21-Oct-2011 14:26:57] <rmatte> once you do you can go to zProperties for the mapping and set action to 'drop' or 'history'
[21-Oct-2011 14:27:07] <humanix> i see
[21-Oct-2011 14:27:07] <rmatte> drop will completely ignore it and history will keep it in history
[21-Oct-2011 14:27:39] <rmatte> the thing is, if it's a syslog type that can come in with multiple messages you may want to do it with a transform instead
[21-Oct-2011 14:27:42] <rmatte> for example...
[21-Oct-2011 14:27:54] <rmatte> if "some message you want to suppress" in evt.summary:
[21-Oct-2011 14:28:01] <rmatte> evt._action = 'drop'
[21-Oct-2011 14:33:49] <humanix> i see
[21-Oct-2011 14:46:02] <Sashness> rmatte: so i think i found the issue
[21-Oct-2011 14:48:12] <dhopp> rmatte: quick question…We monitor a lot of JMX values and the problem is if tomcat timesout (say it's been shutdown for whatever reason) we will get an event for every JMX datapoint which is ideal. So I'm going to do a transform that will just enter one event…got any pointers on how to go about that?
[21-Oct-2011 14:48:26] <dhopp> err isn't ideal I mean
[21-Oct-2011 14:52:59] <Hackman238> dhopp: Tomcat monitor?
[21-Oct-2011 14:53:32] <rmatte> dhopp: consolidating events like that is near impossible with Zenoss
[21-Oct-2011 14:56:40] <dhopp> Hackman238: well I'm core so I don't have the enterprise pack
[21-Oct-2011 14:57:01] <rmatte> Sashness: did you restart Zenoss after doing the renaming?
[21-Oct-2011 14:57:14] <Hackman238> dhopp: Email me a reminder, I have a tomcat monitor
[21-Oct-2011 14:57:24] <dhopp> Hackman238: you do?
[21-Oct-2011 14:57:32] <dhopp> does it handle multiple tomcat's per server?
[21-Oct-2011 14:57:34] <Hackman238> dhopp: yeah
[21-Oct-2011 14:57:40] <dhopp> you would make me a very happy camper
[21-Oct-2011 14:57:44] <Hackman238> dhopp: yes
[21-Oct-2011 14:57:51] <dhopp> I might fly down to Texas and kiss you
[21-Oct-2011 14:57:54] <dhopp> or buy you beer
[21-Oct-2011 14:57:57] <dhopp> whichever
[21-Oct-2011 14:58:04] <Hackman238> dhopp: I haven't tested it well. I wrote it for a customer
[21-Oct-2011 14:58:14] <dhopp> Hackman238: I can definitely be your test :-)
[21-Oct-2011 14:58:18] <Hackman238> dhopp: Hum. Wild Turkey 101 and its a deal
[21-Oct-2011 14:58:21] <dhopp> only have about 700 instances of tomcat
[21-Oct-2011 14:58:27] <dhopp> Wild Turkey, really?
[21-Oct-2011 14:58:28] <Hackman238> LOL!
[21-Oct-2011 14:58:37] <Hackman238> dhopp: Yeah, its good
[21-Oct-2011 14:58:49] <jmp242> well, zcaportal redirect in place now so you don't have to add /wiki anymore
[21-Oct-2011 14:58:56] <dhopp> Hackman238: actually does your tomcat monitor use JMX or use the /manager webapp?
[21-Oct-2011 14:58:58] <jmp242> kinosearch installed and working too I think
[21-Oct-2011 14:59:00] <Hackman238> jmp242: Appreciate it
[21-Oct-2011 14:59:04] <dhopp> jmp242: sweet
[21-Oct-2011 14:59:19] <Hackman238> dhopp: standalone test
[21-Oct-2011 14:59:34] <dhopp> Hackman238: hmm..I'm curious how it works
[21-Oct-2011 15:00:06] <Hackman238> dhopp: I forget honestly, I have few customers with that type of stuff to monitor. I'll send it over for you to pick apart at the very least
[21-Oct-2011 15:01:09] <dhopp> Hackman238: I would very much appreciate it
[21-Oct-2011 15:01:12] <dhopp> does it work with 3.x?
[21-Oct-2011 15:03:01] <Hackman238> dhopp: Yep,w ritten for v3.1
[21-Oct-2011 15:17:51] <dhopp> ok..riddle me this….a model of the device is succesful (no snmp errors and even detects interfaces). However snmp monitoring gives "snmp agent down"
[21-Oct-2011 15:21:38] <Hackman238> dhopp: is its grapsh populating?
[21-Oct-2011 15:21:57] <rmatte> dhopp: what kind of device?
[21-Oct-2011 15:22:05] <dhopp> Hackman238: no
[21-Oct-2011 15:22:07] <dhopp> rmatte: asas
[21-Oct-2011 15:22:09] <dhopp> err
[21-Oct-2011 15:22:10] <dhopp> asa
[21-Oct-2011 15:22:16] <rmatte> yeh, common thing with ASAs
[21-Oct-2011 15:22:17] <dhopp> oh crap
[21-Oct-2011 15:22:20] <dhopp> oids per request
[21-Oct-2011 15:22:23] <dhopp> forgot to change that
[21-Oct-2011 15:22:25] <rmatte> lower zMaxOIDPerRequest from 40 to 10
[21-Oct-2011 15:22:28] <rmatte> that'll do the trick
[21-Oct-2011 15:22:31] <Hackman238> ^thats it
[21-Oct-2011 15:22:35] <rmatte> I just set it on the whole ASA class so I don't forget
[21-Oct-2011 15:22:37] <dhopp> rmatte: I did that on my dev instance
[21-Oct-2011 15:22:39] <Hackman238> Just like PIX
[21-Oct-2011 15:22:42] <dhopp> but forgot to do it on my production
[21-Oct-2011 15:22:42] <rmatte> since it's always an issue with ASAs
[21-Oct-2011 15:22:46] <rmatte>
[21-Oct-2011 15:22:58] <dhopp> rmatte: thanks
[21-Oct-2011 15:23:06] <rmatte> np
[21-Oct-2011 15:27:04] <jmp242> I think car repair shops are a rip off
[21-Oct-2011 15:27:08] <jmp242> $45 to change a lightbulb?
[21-Oct-2011 15:27:10] <rmatte> I'm getting a lab box setup right now to attempt an upgrade from 2.5.2 to 3.2.1
[21-Oct-2011 15:27:19] <rmatte> should be interesting to see if I can pull it off smoothly
[21-Oct-2011 15:27:34] <rmatte> jmp242: they go by a book price rather than the actual time it'll take
[21-Oct-2011 15:27:37] <jmp242> it's always the 2.5.2 to 3.0 that sucks
[21-Oct-2011 15:27:41] <rmatte> and in the book it says something like a half hour
[21-Oct-2011 15:27:46] <jmp242> Yea, they say it takes 30 min
[21-Oct-2011 15:27:48] <rmatte> when in reality it's like 2 minutes if they bust their ass
[21-Oct-2011 15:27:53] <jmp242> I've seen people do it in like 5
[21-Oct-2011 15:28:00] <jmp242> just on the side of the road
[21-Oct-2011 15:28:09] <rmatte> I have two friends who are mechanics so I hear about all the stuff that happens behind the scenes lol
[21-Oct-2011 15:28:21] <jmp242> I wish I had a mechanic friend
[21-Oct-2011 15:28:34] <rmatte> They can generally bust their asses and finish any job in half the time that it's listed as in the book
[21-Oct-2011 15:28:42] <jmp242> Yea
[21-Oct-2011 15:28:44] <rmatte> and then they just sit around relaxing
[21-Oct-2011 15:28:45] <rmatte> lol
[21-Oct-2011 15:28:57] <rmatte> or squeeze more work in
[21-Oct-2011 15:29:15] <jmp242> What I hate though is the jobs like the light that I know you could just walk through it in 10 minutes, work diligently and do it in 5
[21-Oct-2011 15:29:19] <jmp242> and they charge 30 min
[21-Oct-2011 15:29:46] <jmp242> OTOH, at least this place doesn't charge the $100/hr that's normal around here
[21-Oct-2011 15:30:30] <rmatte> yeh
[21-Oct-2011 16:07:16] <rmatte> brb, shutting my computer down to toss more ram in it
[21-Oct-2011 16:08:00] <Exdaix> Hey guys, I just updated to 3.2.1 and now my CPU/load is WAY up. looks like a ton of python processes referring to ZenHub and stuff
[21-Oct-2011 16:12:43] <Exdaix> Nevermind it went back down after like 15 minutes... I guess it was just updating stuff?
[21-Oct-2011 16:14:34] <Hackman238> Exdaix: could have been recataloging or zenhub processing configs
[21-Oct-2011 16:14:51] <Exdaix> Just scared me for a minute... I'm ok now
[21-Oct-2011 16:14:52] <Exdaix> lol
[21-Oct-2011 16:17:14] <rmatte> there, 4GB of RAM in my workstation now instead of 2
[21-Oct-2011 16:17:28] <rmatte> grabbed the extra RAM from a system that's just been laying around here doing nothing
[21-Oct-2011 16:17:28] <Exdaix> Hey Ryan
[21-Oct-2011 16:17:33] <rmatte> hey
[21-Oct-2011 16:18:06] <rmatte> well, time for my to swap my old RAM back in to that system, afk for a bit
[21-Oct-2011 16:18:09] <Exdaix> I've been meaning to ask you... Any forthcoming update to FormulaDataSource as a daemon so it doesn't eat my CPU?
[21-Oct-2011 16:18:26] <rmatte> I'm planning one but it's very complex to do
[21-Oct-2011 16:18:32] <rmatte> it won't be for a while
[21-Oct-2011 16:18:57] <Exdaix> What if I found a waytto have my company pay you. lol
[21-Oct-2011 16:21:08] <rmatte> Then it would probably give me more enthusiam to actually complete the project lol
[21-Oct-2011 16:21:22] <rmatte> since right now I'm just doing it on my own time
[21-Oct-2011 16:21:25] <Exdaix> I shall investigate that avenue then
[21-Oct-2011 16:21:30] <rmatte> and I have other coding projects on the go besides that one
[21-Oct-2011 16:21:58] <rmatte> but I was looking at it very recently...
[21-Oct-2011 16:22:01] <rmatte> it's just really complex
[21-Oct-2011 16:22:07] <rmatte> lots of factors to consider
[21-Oct-2011 16:24:49] <rmatte> my thumb is so sore after just installing 8 sticks of RAM
[21-Oct-2011 16:25:02] <rmatte> they make it so that you practically have to jump on top of the sticks to get them to go in
[21-Oct-2011 16:25:26] <Exdaix> Yeah I miss the old EDO that just had to be pushed sideways
[21-Oct-2011 16:25:30] <rmatte> on the up side no more swap space usage and my system is performing remarkably better
[21-Oct-2011 16:25:41] <rmatte> yeh, the way EDO RAM did it was great
[21-Oct-2011 16:27:55] <tightwork> Would custom properties show up anywhere but the custom properties page? I'd like to define fields like 'customer', 'phone number' or 'plan
[21-Oct-2011 16:28:04] <tightwork> Ideally for them to show up in overview
[21-Oct-2011 16:49:53] <rmatte> ok... test upgrade from 2.5.2 (with ZenPacks still installed) to 3.2.1 commencing
[21-Oct-2011 16:50:14] <rmatte> ah, forgot about the extra ZenPack that I need
[21-Oct-2011 16:51:18] <rmatte> hmmm, if I can find the link to it...
[21-Oct-2011 16:57:49] kocolosk is now known as kocolosk|away
[21-Oct-2011 16:58:47] <rmatte> they removed the PreUpgrade ZenPack from the ZenPacks page
[21-Oct-2011 16:58:48] <rmatte> that's lame
[21-Oct-2011 16:59:43] <rmatte> now I have to hunt around until I find it
[21-Oct-2011 17:01:31] <rmatte> found it through some sourceforge link
[21-Oct-2011 17:04:24] <tightwork> Is anyone using the holt-winters threshold prediction pack? I have it installed and the predicted line is pretty close, and the confidence bands are also reasonable. Overall my goal is to avoid hard thresholds that may bounce at them regularly and be extremely annoying. Any thoughts?
[21-Oct-2011 17:05:15] <tightwork> Especially for load average, which its great if the system is loaded for good reason but almost impossible to be constantly changing a baseline.
[21-Oct-2011 17:05:28] <tightwork> a baseline threshold that is
[21-Oct-2011 17:12:06] * rmatte waits for zencatalog to finish running
[21-Oct-2011 17:12:18] <rmatte> tightwork: can't say that I've used that pack
[21-Oct-2011 17:12:53] <tightwork> well, just tell me you get what im sayin
[21-Oct-2011 17:13:00] <rmatte> yeh I do
[21-Oct-2011 17:13:45] <rmatte> that's really the only way to get what you're wanting
[21-Oct-2011 17:14:31] <tightwork> cool, especially for bandwidth on main uplinks. hard thresholds are useless sometimes, especially when running a marketing campaign.
[21-Oct-2011 17:15:19] <tightwork> good stuff
[21-Oct-2011 17:36:50] <rmatte> hmmm, I've had this Zenny for so long that it's legs are turning yellow
[21-Oct-2011 17:48:58] <densone> Hey Zenoss friends. Need some help installing the httpmonitor zenpack. Getting an error. I am running the command: zenpack --install ZenPacks.zenoss.HttpMonitor-2.0.3-py2.6.egg , and am getting the error.
[21-Oct-2011 17:49:08] <tvincent> whats the error? =)
[21-Oct-2011 17:50:40] <densone> error: Not a URL, existing file, or requirement
[21-Oct-2011 17:50:59] <densone> Awesomely the error shows up in Zenoss.
[21-Oct-2011 17:51:08] <densone> spec: '/home/zenoss/ZenPacks.zenoss.HttpMonitor-2.0.3-py2.6.egg'
[21-Oct-2011 17:51:41] <densone> I am running Zenoss Stack from : zenoss-stack-3.2.0-linux-x64.bin
[21-Oct-2011 18:04:07] <rmatte> 3.2.0 is buggy as all hell, recommend you upgrade to 3.2.1 asap
[21-Oct-2011 18:04:42] <rmatte> also, it's telling you that the file doesn't exist
[21-Oct-2011 18:05:00] <rmatte> you're absolutely positive there's a file in your current directory by that name?
[21-Oct-2011 18:05:19] <rmatte> If there's a .zip at the end you have to unzip it before installing
[21-Oct-2011 18:08:45] <densone> rmatte: Will it upgrade in place?
[21-Oct-2011 18:08:51] <densone> Yes. the file exists.
[21-Oct-2011 18:09:29] <rmatte> the file can't exist, it's telling you that it doesn't, do me a favour...
[21-Oct-2011 18:09:31] <rmatte> do an "ls"
[21-Oct-2011 18:09:34] <rmatte> and paste the output
[21-Oct-2011 18:09:50] <rmatte> densone: yes it'll upgrade in place
[21-Oct-2011 18:10:20] <densone> rmatte: shit. I deleted the egg file instead of the zip.
[21-Oct-2011 18:10:25] <densone> Sorry for hassling.
[21-Oct-2011 18:10:44] <rmatte> haha
[21-Oct-2011 18:10:47] <rmatte> I figured
[21-Oct-2011 18:10:54] <densone> Also, there are lots of articles out there that say you can zenpack --install a zip file.
[21-Oct-2011 18:11:12] <rmatte> ZenPacks used to be zip files wayyyyy back in the day
[21-Oct-2011 18:11:25] <rmatte> no longer the case, .egg now always
[21-Oct-2011 18:11:31] <rmatte> the old .zip packs are no longer supported as of Zenoss 2.5
[21-Oct-2011 18:12:07] <rmatte> the reason the files come down zipped from the Zenoss site is because Jive (the web software they use) auto-zips everything for whatever stupid reason
[21-Oct-2011 18:12:36] <rmatte> a .egg file is already essentially a zip file with a different extension
[21-Oct-2011 18:12:43] <rmatte> so it ends up basically double zipping
[21-Oct-2011 18:15:43] <densone> ahh.. rmatte thanks for the info my friend.
[21-Oct-2011 18:20:03] <rmatte> np
[21-Oct-2011 19:41:00] kocolosk|away is now known as kocolosk
[21-Oct-2011 22:24:56] <tightwork> I have a node that is under high load. I set collectorclienttimeout to 360, and MaxOIDPerRequest to 20 although I still get constant alerts for "Unable to read processes on device timeout on device" .. just keep cranking up the timeout?
[21-Oct-2011 22:28:02] <tightwork> oh wait, I should set zSnmpTimeout not collectorclienttimeout?
[22-Oct-2011 00:00:01] [disconnected at Sat Oct 22 00:00:01 2011]
[22-Oct-2011 00:00:02] [connected at Sat Oct 22 00:00:02 2011]
[22-Oct-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[22-Oct-2011 02:37:36] <raddy> Hello Everybody
[22-Oct-2011 02:38:01] <raddy> Does ZenOSS Core migrated to Java from python?
[22-Oct-2011 02:48:03] <raddy> anybody there?
[22-Oct-2011 03:03:32] <raddy> Anybody there ?
[22-Oct-2011 03:37:43] <raddy> Anybody there?
[22-Oct-2011 06:38:19] <raddy> Hello Everybody
[22-Oct-2011 06:38:44] <raddy> Is it possible to change zenhub's 8081 port?
[22-Oct-2011 07:13:26] <hwinkel1> hi
[22-Oct-2011 07:14:09] <raddy> I fed up with zenoss and uninstalled it
[22-Oct-2011 07:14:42] <raddy> zenoss seems very adament in requiring port 8081 for xml rpc based communication.
[22-Oct-2011 07:15:18] <raddy> Hard-coding is always a bad practice, that too in a matured software
[22-Oct-2011 12:00:01] [disconnected at Sat Oct 22 12:00:01 2011]
[22-Oct-2011 12:00:01] [connected at Sat Oct 22 12:00:01 2011]
[22-Oct-2011 12:00:13] [zenoss-logger (logger bot) has joined #zenoss]
[22-Oct-2011 16:31:24] tvincent_ is now known as tvincent
[23-Oct-2011 00:00:01] [disconnected at Sun Oct 23 00:00:01 2011]
[23-Oct-2011 00:00:02] [connected at Sun Oct 23 00:00:02 2011]
[23-Oct-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[23-Oct-2011 04:08:36] <nozen> so can i ask zenpack dev questions here?
[23-Oct-2011 09:27:58] kocolosk is now known as kocolosk|away
[23-Oct-2011 12:00:01] [disconnected at Sun Oct 23 12:00:01 2011]
[23-Oct-2011 12:00:02] [connected at Sun Oct 23 12:00:02 2011]
[23-Oct-2011 12:00:13] [zenoss-logger (logger bot) has joined #zenoss]
[23-Oct-2011 12:51:01] kocolosk|away is now known as kocolosk
[24-Oct-2011 00:00:01] [disconnected at Mon Oct 24 00:00:01 2011]
[24-Oct-2011 00:00:01] [connected at Mon Oct 24 00:00:01 2011]
[24-Oct-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[24-Oct-2011 05:00:28] <tsener> hello
[24-Oct-2011 05:20:07] <ali3n0> hi folks. I'm wondering if I can set thresholds on partitions usage from data already gathered buy components plugins or if I have to define a new data source
[24-Oct-2011 06:40:44] <ali3n0> found...thanks anyway
[24-Oct-2011 07:55:58] <fragfutter> is there a known issue with zenoss 3.2.1 and graph rendering?
[24-Oct-2011 08:14:20] <fragfutter> its an issue with chrome...
[24-Oct-2011 08:33:45] kocolosk is now known as kocolosk|away
[24-Oct-2011 08:38:46] <Hackman238> Morning all!
[24-Oct-2011 08:46:37] <enkrypt> hiya Hackman238
[24-Oct-2011 08:49:25] <Hackman238> enkrypt: How goes it?
[24-Oct-2011 08:49:30] <jmp242> morning
[24-Oct-2011 08:49:40] <Hackman238> jmp242: Morning
[24-Oct-2011 08:49:47] kocolosk|away is now known as kocolosk
[24-Oct-2011 08:50:27] <jb> morning.
[24-Oct-2011 08:50:48] <fragfutter> ah, channel is waking up...
[24-Oct-2011 08:51:04] <psyklone> *yawn*
[24-Oct-2011 08:51:26] <jmp242> don't do that psyklone
[24-Oct-2011 08:51:28] <jmp242>
[24-Oct-2011 08:51:39] <Hackman238> ...speaking of that, need coffee. BRB
[24-Oct-2011 08:51:49] <fragfutter> anyone tried to use zenoss 3.2.1 with chrome/chromium?
[24-Oct-2011 08:52:04] <psyklone> jmp242, doh! alright, coffee run ...
[24-Oct-2011 08:52:14] <fragfutter> rum?
[24-Oct-2011 08:52:48] <enkrypt> Hackman238: updated to 3.2.1 enterprise, just finished updating my collectors. but I guess there isn't much difference between 3.2.0 + 3 patches and 3.2.1
[24-Oct-2011 08:54:03] <Sashness> fragfutter: i have. works for the most part but some things don't work like the command list at the bottom of the device page
[24-Oct-2011 08:54:39] <fragfutter> Sashness: did you try to view graphs (like interfaces)? click on first interface, admire graph, click on second interface, admire graph, click on first again, Render error.
[24-Oct-2011 08:54:44] <ali3n0> hi folks, anyone here using zenoss 3.2.0 + ssl + json apis?
[24-Oct-2011 08:54:52] <fragfutter> Sashness: i know it work in 3.0
[24-Oct-2011 08:55:27] <fragfutter> Sashness: i try find a confirmation for the bug before reporting it
[24-Oct-2011 08:55:51] <Sashness> fragfutter: hehe.. always a good idea
[24-Oct-2011 08:56:10] <Sashness> i've never had it work with 3.x before
[24-Oct-2011 08:56:10] <Hackman238> enkrypt: That's my understanding. Run in to any problems?
[24-Oct-2011 08:56:13] <Sashness> but i don't try much either
[24-Oct-2011 08:56:27] <Hackman238> fragfutter: Rum goes in the coffee
[24-Oct-2011 09:01:45] <fragfutter> can somoene else check chrome with zenoss 3.2.1?
[24-Oct-2011 09:02:46] <Hackman238> fragfutter: I'll have to upgrade to it, but I'll give it a shot for you
[24-Oct-2011 09:03:49] <Hamzah> hmmm how do you guys usually deal with graphing histogram values? I want to graph the threads value from /proc/net/rpc/nfsd, but I can't think of how to do it with zenoss >.<
[24-Oct-2011 09:07:45] <fragfutter> Hamzah: i would write a script to expose the values through snmp. Then add a new datasource which polls 11 values and graphs them all in one graph
[24-Oct-2011 09:07:51] <enkrypt> Hackman238: so far the upgrade went smooth. I'm having some problems in general with zenoss, but those pre-date the upgrade
[24-Oct-2011 09:11:11] <Hackman238> enkrypt: Glad it went smooth. Lets see if we can get those other problems cleared up!
[24-Oct-2011 09:13:10] <Hamzah> fragfutter i've already exported the values using nrpe, but I'm unsure what the best way to create the graph for the values would be. Should I just use GUAGE as the data source type, and put those onto a graph? I guess that would kind of work?
[24-Oct-2011 09:13:36] <fragfutter> Hamzah: yes that should work (besides i hate nrpe)
[24-Oct-2011 09:13:42] <Hamzah> hehe
[24-Oct-2011 09:32:22] <Hackman238> fragfutter: Yeah theres some sort of graphing problem with Chrome and Opera in v3.2.1, at least when you upgrade to it anyway
[24-Oct-2011 09:32:57] <fragfutter> Hackman238: same with a fresh install
[24-Oct-2011 09:35:30] <Hackman238> fragfutter: Damn.
[24-Oct-2011 09:38:11] <jmp242> hey, Hackman238 / other coders - can you take a look at this thread? message/62086#62086
[24-Oct-2011 09:41:59] <Hackman238> jmp242: Replied TY for the heads up
[24-Oct-2011 09:42:17] <Hackman238> fragfutter: I ahte to say it, but you might need to put in a bug report for it
[24-Oct-2011 09:42:39] <fragfutter> Hackman238: yep. i need to find my login...
[24-Oct-2011 09:51:07] <fragfutter> Hackman238: ok. created as http://dev.zenoss.org/trac/ticket/7881
[24-Oct-2011 09:51:30] <ali3n0> does anyone know how frequent is the ping check?
[24-Oct-2011 09:51:31] <fragfutter> Hackman238: as opera user, can you also confirm the bug http://dev.zenoss.org/trac/ticket/7879 ?
[24-Oct-2011 09:53:41] <Hackman238> ali3n0: 1 minute
[24-Oct-2011 09:53:46] <ali3n0> tnx
[24-Oct-2011 09:53:57] <ali3n0> just 1 icmp packet?
[24-Oct-2011 09:55:14] <Hackman238> ali3n0: Yes, one out, one back.
[24-Oct-2011 09:55:20] <ali3n0> nice
[24-Oct-2011 09:55:35] <Hackman238> ali3n0: by default it'll retry 4 times, 1.5s timeout
[24-Oct-2011 09:55:46] <ali3n0> great, thanks
[24-Oct-2011 09:56:16] <Hackman238> ali3n0: If a bottleneck like a ISG is being nailed down by the ICMP traffic, reduce the 'ping packets in flight' number on the collectors page
[24-Oct-2011 09:57:05] <Hackman238> fragfutter: Double checking with a clean profile. I was messing with Opera to try to ghet the dashboard working.
[24-Oct-2011 09:58:47] <jmp242> dashboard never works in Opera
[24-Oct-2011 10:00:19] <Hackman238> fragfutter: Yep, still broke. I updated the trac
[24-Oct-2011 10:00:34] <Hackman238> jmp242: I put in a ticket for opera support
[24-Oct-2011 10:01:32] <jmp242> I don't know that I would expect much work for Opera, for me as long as Firefox Works I can't get to incensed
[24-Oct-2011 10:03:05] <Hackman238> jmp242: We're finding that the js engine in FF is way too slow.
[24-Oct-2011 10:03:21] <jmp242> Really
[24-Oct-2011 10:03:23] <fragfutter> Hackman238: thanks for confirming it.
[24-Oct-2011 10:03:32] <Hackman238> jmp242: Zenoss is very fast in Opera. Only issue is the dashboard and the trace looks insignificant. They can fix it.
[24-Oct-2011 10:03:36] <Hackman238> fragfutter: NP
[24-Oct-2011 10:03:42] <jmp242> well I knew that Opera and Chrome both have better JS than FF
[24-Oct-2011 10:04:03] <jmp242> actually, FF for me seems to be the IE of alternative browsers lol
[24-Oct-2011 10:04:08] <Hackman238> jmp242: Its especially ovbious in Linux for some reason.
[24-Oct-2011 10:04:10] <jmp242> but it's got the network effects
[24-Oct-2011 10:04:27] <Hackman238> jmp242: LOL yep
[24-Oct-2011 10:10:06] <jb> erm, any idea whats wrong with this regex?
[24-Oct-2011 10:10:07] <jb> (/Admin|/Admin-Out|/Admin-DBA|/Oracle|/oracle|/cdrom|/mnt/cdrom|/www/repos/.*)
[24-Oct-2011 10:10:15] <jb> i think hte problem is from the "/www/repos/*" part.
[24-Oct-2011 10:10:41] <rmatte> just do: ^/www/repos/
[24-Oct-2011 10:10:47] <rmatte> it'll do the same thing
[24-Oct-2011 10:11:06] <jb> k, i'll give that a shot. thanks.
[24-Oct-2011 10:11:09] <rmatte> np
[24-Oct-2011 10:11:49] <Sashness> hey rmatte!
[24-Oct-2011 10:12:40] <Sashness> that problem i had last week? the info you gave me about checking the relationships fixed the issue!
[24-Oct-2011 10:12:42] <jb> i have a device class that uses password based SSH auth, but it's also attempting to use key based. how do I stop it from trying to use keys? delete the contents of zKeyPath maybe?
[24-Oct-2011 10:12:57] <Sashness> thanks a bunch. i owe you a beer if your ever in the Miami area!!
[24-Oct-2011 10:14:35] <rmatte> Sashness: good to hear man
[24-Oct-2011 10:14:37] <rmatte>
[24-Oct-2011 10:20:37] <jb> ugh
[24-Oct-2011 10:20:58] <jb> File "/opt/zenoss/lib/python2.6/re.py", line 245, in _compile
[24-Oct-2011 10:20:59] <jb> raise error, v # invalid expression
[24-Oct-2011 10:21:03] <jb> still not working.
[24-Oct-2011 10:24:41] <Hackman238> kill -9 21865
[24-Oct-2011 10:25:00] <Hackman238> Whoops. Sorry
[24-Oct-2011 10:28:19] <jb> wow, this can't be that hard
[24-Oct-2011 10:28:27] <jb> no tracebacks now, but the filesystems are getting modeled.
[24-Oct-2011 10:30:14] <jb> all im trying to do is ignore filesystems that begin with /www/repos
[24-Oct-2011 10:30:28] <jb> and '^/www/repos' doesn't appear to be doing the trick in zFileSystemIgnore.
[24-Oct-2011 10:31:08] <Hackman238> jb: Here's what we use: ^[hs]d[a-z]\d+$|c\d+t\d+d\d+s\d+$|^cciss\/c\dd\dp\d$|^dm\-\d$
[24-Oct-2011 10:31:31] <jb> hmm
[24-Oct-2011 10:31:37] <Hackman238> jb: I'm wondering what the () are for in yours
[24-Oct-2011 10:31:47] <jb> yeah, I removed those.
[24-Oct-2011 10:32:03] <jb> i think the problem is actually with inheritence.
[24-Oct-2011 10:33:05] <Hackman238> jb: Hum
[24-Oct-2011 10:33:32] <rmatte> The () are originally there because the default is something like ^(/filesys1|/filesys2)
[24-Oct-2011 10:33:37] <rmatte> so it applies the ^ to each one
[24-Oct-2011 10:33:43] <rmatte> the ( shouldn't be required otherwise
[24-Oct-2011 10:33:59] <jb> ok, got it.
[24-Oct-2011 10:34:01] <jb> it was single quotes.
[24-Oct-2011 10:34:26] <jb> yay
[24-Oct-2011 10:34:31] <rmatte> what do you mean?
[24-Oct-2011 10:34:36] <jb> i had it wrapped in single quotes
[24-Oct-2011 10:34:37] <rmatte> you had to single quote everything?
[24-Oct-2011 10:34:40] <rmatte> ohhh
[24-Oct-2011 10:34:41] <jb> ie, '^/www/repos'
[24-Oct-2011 10:34:45] <rmatte> yeh, you can't do that hehe
[24-Oct-2011 10:34:48] <jb> thanks
[24-Oct-2011 10:34:52] <jb> ok, one more annoying event.
[24-Oct-2011 10:35:15] <jb> Error processing transform/mapping on Event Class /Change/Set/Status
[24-Oct-2011 10:35:34] <ali3n0> uhm, I'm not sure but seems that api_example.py shipped with api 3.0 is broken, or at least for me. there's a method that doesn't work
[24-Oct-2011 10:35:53] <jb> hm, error with line 474. i think this is code that shipped with zenoss though.
[24-Oct-2011 10:36:28] <rmatte> jb, you probably want ^/www/repos/ or else something like /www/reposition would get blocked too
[24-Oct-2011 10:37:35] <jb> ah yes.
[24-Oct-2011 10:43:59] <rmatte> hmmm, so I did a test upgrade for 2.5.2, it went well enough but zenmigrate fails
[24-Oct-2011 10:44:09] <rmatte> File "/usr/local/zenoss/zenoss/Products/ZenModel/IpAddress.py", line 232, in index_object
[24-Oct-2011 10:44:09] <rmatte> super(IpAddress, self).index_object(idxs)
[24-Oct-2011 10:44:09] <rmatte> TypeError: new_index_object() takes exactly 1 argument (2 given)
[24-Oct-2011 10:44:42] <rmatte> I think Chet gave me some options to use with it to get it working last time, trying to dig them up
[24-Oct-2011 10:44:52] <psyklone> rmatte - that means the pre-upgrade zenpack didn't get removed like it should have
[24-Oct-2011 10:45:16] <psyklone> rmatte, remove that zenpack and then re-run zenmigrate
[24-Oct-2011 10:45:47] <psyklone> you might want to have a close look at the install.log, though, to see if it offers any clues about what happened when it should have removed that zenpack in the course of the upgrade
[24-Oct-2011 10:48:17] <rmatte> another thing that I didn't try that Hackman suggested was deleting the contents of $ZENHOME/Products prior to the upgrade
[24-Oct-2011 10:48:26] <rmatte> I'll probably repeat the process and do that for testing purposes
[24-Oct-2011 10:48:38] <rmatte> I'll check if that pack is still installed
[24-Oct-2011 10:49:47] <rmatte> you're right, it didn't get removed
[24-Oct-2011 10:49:48] <psyklone> rmatte - that's the only code where 'new_index_object()' is called
[24-Oct-2011 10:49:50] <rmatte> I'll give that a go
[24-Oct-2011 10:49:58] <rmatte> thanks
[24-Oct-2011 11:17:05] <nexex> If you set the apache collector to use ssl, will it try port 443, or does that need to be changed as well?
[24-Oct-2011 12:00:01] [disconnected at Mon Oct 24 12:00:01 2011]
[24-Oct-2011 12:00:01] [connected at Mon Oct 24 12:00:01 2011]
[24-Oct-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[24-Oct-2011 12:00:15] <sendak.freenode.net> [freenode-info] why register and identify? your IRC nick is how people know you. http://freenode.net/faq.shtml#nicksetup
[24-Oct-2011 12:09:00] <rmatte> nexex: the port needs to be changed to 443
[24-Oct-2011 12:09:32] <rmatte> the act of selecting SSL just ensure that it'll use the SSL protocol
[24-Oct-2011 12:09:40] <rmatte> you still need to tell it what port you want it to hit
[24-Oct-2011 12:10:37] <nexex> rmatte: thanks, I think thats the problem I'm having, apache redirects to https automatically, I'll try having it go to 443 directly
[24-Oct-2011 12:14:52] * rmatte nods
[24-Oct-2011 12:18:51] <Hackman238> rmatte: What browser/plat are you on?
[24-Oct-2011 12:19:04] <rmatte> Linux/Firefox
[24-Oct-2011 12:19:10] <rmatte> why?
[24-Oct-2011 12:19:11] <Hackman238> rmatte: Can you run this test? http://v8.googlecode.com/svn/data/benchmarks/v6/run.html
[24-Oct-2011 12:19:17] <rmatte> k
[24-Oct-2011 12:19:28] <Hackman238> rmatte: I need complete confirmation FF js engine blow nuts.
[24-Oct-2011 12:19:43] <Hackman238> rmatte: Probably going to push hard for alternative browser support.
[24-Oct-2011 12:21:34] <jmp242> On the 3.6 line, FF keeps thinking the scripts have hung for me on Windows
[24-Oct-2011 12:23:27] <Hackman238> jmp242: It did that to me also. I clicked continue like 6 times and it finshed
[24-Oct-2011 12:24:43] <Hackman238> Opera we're seeing averages >9,500, Chrome >11,000, FF 3.6 64 ~350, FF 7.0.1 ~850
[24-Oct-2011 12:24:43] <jmp242> It's not exactly fair though
[24-Oct-2011 12:25:02] <jmp242> because I'm using latest stable Opera
[24-Oct-2011 12:25:18] <jmp242> but older Firefox line, except I can update Opera *because* we don't support it lol
[24-Oct-2011 12:25:34] <jmp242> whereas FF is 3.6.x until it dies cause we can't do the testing for major updates every 6 weeks
[24-Oct-2011 12:26:03] <Hackman238> jmp242: Even FF 7.0.1 only scored just barely 850 average
[24-Oct-2011 12:26:09] <Hackman238> (on linux)
[24-Oct-2011 12:26:14] <Hackman238> jmp242: On windows its faster.
[24-Oct-2011 12:26:29] <jmp242> strange
[24-Oct-2011 12:26:36] <jmp242> isn't it the same code?
[24-Oct-2011 12:27:12] <jmp242> Though if Zenoss did support Opera I'd be thrilled
[24-Oct-2011 12:27:16] <Hackman238> jmp242: it is, but there must be some serious compile differnce.
[24-Oct-2011 12:28:10] <Hackman238> rmatte: Did it run?
[24-Oct-2011 12:28:15] <nexex> on linux for, firefox 7.0.1: 2817, chromium 16.0.912.0: 5793
[24-Oct-2011 12:29:46] <Hackman238> nexex: Interesting. What version of linux and arch?
[24-Oct-2011 12:31:34] <jmp242> Trying via XForwarding from our Linux app server
[24-Oct-2011 12:31:42] <nexex> gentoo x86, kernel version 3.0.4, its an old machine (core2 duo)
[24-Oct-2011 12:31:59] <Hackman238> nexex: ah not fair. LOL gentoo
[24-Oct-2011 12:32:19] <Hackman238> nexex: TY for testing.
[24-Oct-2011 12:32:26] <nexex> Hackman238: only cflag is -march=core2
[24-Oct-2011 12:32:30] <nexex> Hackman238: yw
[24-Oct-2011 12:33:09] <Hackman238> nexex: ja, I recompiled a bunch of various versions with said flag and -O2, all faster than stock bin, but still slow
[24-Oct-2011 12:33:54] <jmp242> Well, I only got 203 on SL
[24-Oct-2011 12:33:59] <jmp242> but it's an older computer
[24-Oct-2011 12:34:32] <Hackman238> jmp242: Wow LOL
[24-Oct-2011 13:01:26] <Hackman238> jmp242: What do you think about a ZCA poll on what browser Zenoss should add support for?
[24-Oct-2011 13:01:54] <Hackman238> jmp242: Options: Opera and Chrome. (Safari not an option because not cross platform)
[24-Oct-2011 13:07:20] <jmp242> mmm
[24-Oct-2011 13:07:31] <jmp242> Well, I expect you'll have equal groups of users
[24-Oct-2011 13:20:53] <jmp242> Anyone who knows the upgrade issues between 3.1 and 3.2.1, can you look at: message/62103#62103 ?
[24-Oct-2011 13:21:05] <jmp242> Also, how many people have had 3.2.1 work OK / well / bad ??
[24-Oct-2011 13:31:48] <nexex> jmp242: changing graph range doesn't work for me, a couple apache webservers need their config redone, and larger than default graphs are cutoff -- other than that, I have had no issues
[24-Oct-2011 13:37:04] <Hackman238> jmp242: Opera/Chrome range select confirmed here
[24-Oct-2011 13:38:08] <Hackman238> FireFox 7.0.1 prebuilt by some nameless internet someone runs okay, but its definately not what I would qualify as fast at a scrore of 3800 vs Opera >10000, and Chrome >11000
[24-Oct-2011 13:38:41] <jmp242> yea
[24-Oct-2011 13:38:54] <jmp242> Hey Hackman238 - can you look at that thread re the upgrade?
[24-Oct-2011 13:40:14] <Hackman238> http://lifehacker.com/5844150/browser-speed-tests-firefox-7-chrome-14-internet-explorer-9-and-more
[24-Oct-2011 14:15:13] <hikenboot> hello I am wondering how well frequented this group is in here
[24-Oct-2011 14:15:41] <hikenboot> looking for a good community support solution cacti just aint it
[24-Oct-2011 14:17:23] <Hackman238> hikenboot: Pretty good
[24-Oct-2011 14:17:42] <Hackman238> hikenboot: ZCA has been working to revitalize the community
[24-Oct-2011 14:18:10] <hikenboot> cause 3 weeks of cacti just aint going to cut it with the problems it seems to have that are mysterious and no solutions
[24-Oct-2011 14:18:18] <hikenboot> thats great glad to here it
[24-Oct-2011 14:18:26] <hikenboot> s/here/hear
[24-Oct-2011 14:19:19] <hikenboot> zenoss appliance 3.2.1 downloading now
[24-Oct-2011 14:19:39] <hikenboot> how come they chose fedora instead of centos to build it on
[24-Oct-2011 14:19:51] <hikenboot> fedora has never been a real stable solution
[24-Oct-2011 14:20:09] <hikenboot> soluton/platform
[24-Oct-2011 14:23:02] <Hackman238> hikenboot: Not sure. To be honest I'd try to avoid using the appliance
[24-Oct-2011 14:24:19] <Hackman238> hikenboot: also a significantly revamped version of Zenoss, zenoss 4.2 will be out in the coming months
[24-Oct-2011 14:28:37] <jmp242> Hackman238: is 4.2 going to require a pre-upgrade zenpack etc?
[24-Oct-2011 14:28:57] <Hackman238> jmp242: No
[24-Oct-2011 14:29:14] <jmp242> Did you do the survey on ZCA?
[24-Oct-2011 14:29:22] <jmp242> for the browser(s)?
[24-Oct-2011 14:29:31] <Hackman238> jmp242: Not yet, but I will
[24-Oct-2011 14:29:47] <jmp242> Ah ok... I think you do have a good case
[24-Oct-2011 14:29:53] <Hackman238> jmp242: Trying to bring the Zenoss CSS to Opera compliance myself
[24-Oct-2011 14:30:02] <jmp242> Oh fun
[24-Oct-2011 14:30:12] <jmp242> Did you see message/62103#62103
[24-Oct-2011 14:30:12] <jmp242> ?
[24-Oct-2011 14:30:14] <Hackman238> jmp242: We'll get that browser support one way or another
[24-Oct-2011 14:31:30] <Hackman238> jmp242: Opera is particularly unfond of {zoom:###}
[24-Oct-2011 14:53:08] <rmatte> There's a point where browser standards compliance becomes ludicrous
[24-Oct-2011 14:53:12] <rmatte> Opera crosses the line
[24-Oct-2011 14:53:40] <rmatte> one of the reasons why I refuse to use it
[24-Oct-2011 14:54:02] <rmatte> if a tag works in another browser and it's not called Internet Explorer, then why doesn't it work in your browser?
[24-Oct-2011 14:54:03] <rmatte> lol
[24-Oct-2011 14:54:15] <rmatte> IE is just a lost cause
[24-Oct-2011 14:54:32] <humanix> hi there guys
[24-Oct-2011 14:54:37] <rmatte> aloha
[24-Oct-2011 14:54:47] <humanix> hi rmatte
[24-Oct-2011 14:55:00] <humanix> how are you?
[24-Oct-2011 14:55:34] <rmatte> good, retardedly busy
[24-Oct-2011 14:55:54] <rmatte> 3 projects on the go all at once
[24-Oct-2011 14:55:57] <humanix> ohh
[24-Oct-2011 14:55:58] <rmatte> trying to juggle as best I can
[24-Oct-2011 14:56:02] <humanix> haha
[24-Oct-2011 14:56:04] <humanix> it happends
[24-Oct-2011 14:56:24] <rmatte> people are getting pissed off, but I keep asking them "What do you want me to do? I'm stuck here."
[24-Oct-2011 14:56:29] <rmatte> lol
[24-Oct-2011 14:56:40] <humanix> hahahaha
[24-Oct-2011 14:56:44] <humanix> it also happends
[24-Oct-2011 14:56:59] <Hackman238> LOL
[24-Oct-2011 14:57:21] <Hackman238> rmatte: Relax guy, I'm about 30% done hacking Zenoss to be CSS compliant
[24-Oct-2011 15:09:02] <rmatte> so I just logged in to Tivoli Live for the first time... as I suspected lots of super colorful graphs that managers love... other than that, bleh
[24-Oct-2011 15:09:20] <rmatte> but I have to learn how to use this thing unfortunately
[24-Oct-2011 15:22:32] <mahimahi> Is it just me or does 3.2.1 stylesheet fail in Chrome?
[24-Oct-2011 15:23:13] <mahimahi> or is that what rmatte is working on?
[24-Oct-2011 15:25:58] <rmatte> Hackman238 is working on it actually
[24-Oct-2011 15:29:48] <Hackman238> AAHH!!! /opt/zenoss/lib/python/zenoss/extjs/src/resources/css/ext-all.css ful of ^M out of the boX!
[24-Oct-2011 15:29:57] <rmatte> lol
[24-Oct-2011 15:39:52] <rmatte> File "/usr/local/zenoss/python/lib/python2.6/site-packages/zope.interface-3.5.2-py2.6-linux-x86_64.egg/zope/interface/adapter.py", line 579, in subscribers
[24-Oct-2011 15:39:52] <rmatte> subscription(*objects)
[24-Oct-2011 15:39:52] <rmatte> File "/usr/local/zenoss/zenoss/Products/Zuul/catalog/events.py", line 71, in onObjectRemoved
[24-Oct-2011 15:39:52] <rmatte> catalog.uncatalog_object(uid)
[24-Oct-2011 15:39:52] <rmatte> AttributeError: 'GlobalCatalog' object has no attribute 'uncatalog_object'
[24-Oct-2011 15:40:02] <rmatte> hmmm, maybe a reindex will fix that
[24-Oct-2011 15:41:46] <bseklecki> i dunno guys, HP OpenView Performance Manager has a nice data graph browser
[24-Oct-2011 15:42:20] <rmatte> yeh, and it's closed source software so good luck integrating it with other systems
[24-Oct-2011 15:42:31] <rmatte> every tool has their pros and cons
[24-Oct-2011 15:43:22] <rmatte> Zenoss happens to use RRDTool for it's graphs which may not be as pretty as some but very detailed.
[24-Oct-2011 15:44:03] <rmatte> Zenoss Enterprise also has Datacenter Insight which generates pretty graph reports for managers to look at
[24-Oct-2011 15:46:43] <rmatte> hmm, so I'm stuck on that error
[24-Oct-2011 15:46:58] <rmatte> seems like every time I do this upgrade I get a different error from zenmigrate
[24-Oct-2011 15:47:11] <rmatte> hmmm, I have an idea...
[24-Oct-2011 15:47:39] <rmatte> going to revert from snapshot and this time I'll remove the Pre-Upgrade pack prior to doing the upgrade
[24-Oct-2011 15:48:08] <rmatte> hopefully it checks if the database has been converted instead of whether or not the Pre-Upgrade pack is installed (I assume it does)
[24-Oct-2011 15:49:03] <Hackman238> rmatte: try rebuild globalcatalog
[24-Oct-2011 15:49:13] <rmatte> how is that done?
[24-Oct-2011 15:49:24] <Hackman238> rmatte: Also check to make sure zenmigrate ran completely
[24-Oct-2011 15:49:39] <rmatte> I ran it by hand and it never completes, always errors
[24-Oct-2011 15:50:15] <Hackman238> rmatte: Which step?
[24-Oct-2011 15:50:25] <rmatte> Restoring... or whatever
[24-Oct-2011 15:50:28] <rmatte> errors right away
[24-Oct-2011 15:51:06] <Hackman238> Whoever did this CSS in the ZenUI3 needs a serious, relentless flogging.
[24-Oct-2011 15:51:10] <rmatte> I'm re-running zencatalog right now, then going to snapshot so that I don't have to do this each time
[24-Oct-2011 15:51:16] <rmatte> lol
[24-Oct-2011 15:51:19] <Hackman238> rmatte: Ah gotcha
[24-Oct-2011 15:51:22] <rmatte> It was probably Ian
[24-Oct-2011 15:51:28] <rmatte> I think he did most of the UI
[24-Oct-2011 15:51:42] <rmatte> I know cgibbons did a bit too
[24-Oct-2011 15:52:19] <Hackman238> Can't be. Not...this.
[24-Oct-2011 15:52:23] <rmatte> lol
[24-Oct-2011 15:52:35] <Hackman238> I mean CSS is always ungly
[24-Oct-2011 15:52:37] <Hackman238> *ugly
[24-Oct-2011 15:52:38] <rmatte> maybe they hired a monkey off the street
[24-Oct-2011 15:52:48] <Hackman238> But this is way inconsistant
[24-Oct-2011 15:53:09] <rmatte> full of ^M is bad in general
[24-Oct-2011 15:53:14] <rmatte> means it was done in wordpad or something
[24-Oct-2011 15:53:16] <Hackman238> rmatte: Like the kind that collect change while another turns a handle on a record player?
[24-Oct-2011 15:53:28] <rmatte> exactly
[24-Oct-2011 15:54:29] <rmatte> "Your minky is breaking ze lawwww." "Minky!?" "What?" "You said minky!" "Yes, chimpanzee minky!"
[24-Oct-2011 15:54:44] <rmatte> (Pink Panther)
[24-Oct-2011 15:55:05] <rmatte> http://www.youtube.com/watch?v=mQAMvmi1Zwk
[24-Oct-2011 15:55:38] <Hackman238> rmatte: LOL
[24-Oct-2011 15:57:37] <humanix> a question
[24-Oct-2011 15:57:47] <humanix> i already did it before, but i lost the answer
[24-Oct-2011 15:58:10] <humanix> how and where do you ignore dev and proc fs statics on a monitoring
[24-Oct-2011 15:58:32] <humanix> i know is something like zFileSystemMapIgnoreNames or zFileSystemMapIgnoreTypes
[24-Oct-2011 15:58:42] <rmatte> zFileSystemMapIgnoreNames
[24-Oct-2011 15:58:55] <rmatte> ^(/dev/|/proc/)
[24-Oct-2011 15:58:58] <rmatte> something like that
[24-Oct-2011 15:59:05] <rmatte> or...
[24-Oct-2011 15:59:19] <rmatte> /dev/.*|/proc/.*
[24-Oct-2011 15:59:23] <bseklecki> rmatte: its the ability to generate graphs on-demand from SQL-backed historical data
[24-Oct-2011 15:59:41] <bseklecki> i _thought_ Z was doing that, but these seem to be traditional RRD data stores
[24-Oct-2011 15:59:44] <rmatte> bseklecki: which is what Datacenter Insight does for Zenoss
[24-Oct-2011 15:59:55] <rmatte> and yes, they are traditional RRD files
[24-Oct-2011 16:00:37] <rmatte> You're comparing an enterprise product to the free open source version of Zenoss. The pay version of Zenoss supports that sort of thing.
[24-Oct-2011 16:01:19] <rmatte> Core may eventually support it through community development
[24-Oct-2011 16:01:39] <rmatte> but we're not there yet
[24-Oct-2011 16:01:43] <humanix> thanks rmatte
[24-Oct-2011 16:01:48] <rmatte> np
[24-Oct-2011 16:16:07] <humanix> it doesnt work
[24-Oct-2011 16:16:15] <humanix> can i do it to the class
[24-Oct-2011 16:16:22] <humanix> or i have to apply to each device?
[24-Oct-2011 16:21:02] <rmatte> to the class
[24-Oct-2011 16:22:27] <rmatte> unless the devices already have it locally applied, which will overrule anything below it
[24-Oct-2011 16:22:29] <rmatte> make sure they don't
[24-Oct-2011 16:23:21] <rmatte> The thing is, you should already have a bunch of stuff in zFileSystemMapIgnoreTypes by default that should prevent /proc and /dev stuff from being modeled
[24-Oct-2011 16:24:01] <rmatte> http://fpaste.org/0POG/raw/ <--- This is the default list for zFileSystemMapIgnoreTypes
[24-Oct-2011 16:24:23] <rmatte> that's all I have set
[24-Oct-2011 16:24:50] <rmatte> but the ^(/dev/|/proc/) for zFileSystemMapIgnoreNames should work, no reason for it not to
[24-Oct-2011 16:24:54] <rmatte> it's valid regex
[24-Oct-2011 16:25:18] <rmatte> I helped someone with the same sort of thing earlier today and it worked fine.
[24-Oct-2011 16:31:02] <Hackman238> I give on the CSS madness for the day
[24-Oct-2011 16:31:05] <Hackman238> LOL
[24-Oct-2011 16:37:15] <Sashness> rmatte: so i thought the issue was fixed but apparently it came back
[24-Oct-2011 16:41:32] <bseklecki> rmatte: noted.
[24-Oct-2011 16:42:04] <bseklecki> now, lets talka bout why ethernetCsmacd's ifOutErrors data point is showing values in millsions
[24-Oct-2011 16:42:45] <bseklecki> wtf is a DERIVE type [looks]
[24-Oct-2011 16:44:13] <bseklecki> a variation on counter, possibly, w/o overflow
[24-Oct-2011 17:01:06] <rmatte> it's not showing in million
[24-Oct-2011 17:01:10] <rmatte> it's showing in milli
[24-Oct-2011 17:01:14] <rmatte> as in a fraction of a value
[24-Oct-2011 17:01:17] <rmatte> m is milli
[24-Oct-2011 17:01:20] <rmatte> M is mega
[24-Oct-2011 17:01:21] <bseklecki> m
[24-Oct-2011 17:01:26] <bseklecki> mmmmmm
[24-Oct-2011 17:01:41] <bseklecki> mmmmmyeammmeh? mm.
[24-Oct-2011 17:02:14] <rmatte> DERIVE shows the difference between the previous value and the current value
[24-Oct-2011 17:02:40] <rmatte> http://oss.oetiker.ch/rrdtool/doc/rrdcreate.en.html
[24-Oct-2011 17:02:52] <rmatte> That explains all the different datapoint types
[24-Oct-2011 17:02:53] <bseklecki> yep was just reading
[24-Oct-2011 17:05:52] <rmatte> but to answer your question about the errors in more detail, those values represent errors per second over the 5 minute polling period
[24-Oct-2011 17:06:07] <rmatte> so it essentially counts the number of error during that period and divides by 300 seconds (5 minutes)
[24-Oct-2011 17:06:26] <rmatte> so if you had 30 errors during that period your end value is 0.1
[24-Oct-2011 17:06:31] <rmatte> which is 100m
[24-Oct-2011 17:06:41] <bseklecki> right, averaged over the sample time its a fraction
[24-Oct-2011 17:07:24] <rmatte> yup
[24-Oct-2011 17:07:25] <bseklecki> i think in the past i've done interface errors as a counter; and the hill should never climb
[24-Oct-2011 17:07:49] <bseklecki> since Cisco has a 'clear interface coutners err'
[24-Oct-2011 17:08:04] <rmatte> that doesn't affect the actual snmp counters though
[24-Oct-2011 17:08:10] <rmatte> just the ones you see in show interface
[24-Oct-2011 17:08:56] * bseklecki ponders that
[24-Oct-2011 17:09:05] <bseklecki> RMON MIB or....IF-MIB?
[24-Oct-2011 17:09:20] <bseklecki> *shrug*
[24-Oct-2011 17:09:38] <rmatte> IF-MIB which is what Zenoss does, I'm not sure about the RMON-MIB stuff, but it makes absolutely zero sense to allow a command to reset an SNMP counter
[24-Oct-2011 17:09:47] <rmatte> by doing so you throw off anything that is monitoring that counter
[24-Oct-2011 17:10:13] <rmatte> I'd be blown away if they did actually do that
[24-Oct-2011 17:12:25] <bseklecki> maybe program it for gauge? I think freebsd had '-z' to netstat(8), and a bump of Net-SNMP would reset it
[24-Oct-2011 17:13:25] <rmatte> gauge would just have a hill effect of constantly increasing
[24-Oct-2011 17:13:40] <rmatte> and then a sudden drop off if the counter were reset
[24-Oct-2011 17:14:35] <rmatte> If the counter automatically reset every time it was queried then you'd want to use an ABSOLUTE type
[24-Oct-2011 17:16:19] kocolosk is now known as kocolosk|away
[24-Oct-2011 17:16:28] <humanix> I SEE
[24-Oct-2011 17:16:32] <humanix> now its working rmatte
[24-Oct-2011 17:16:40] <rmatte> what was the problem?"
[24-Oct-2011 17:17:11] <humanix> i put the ^(/dev.... one, remodel and it works fine
[24-Oct-2011 17:17:20] <rmatte> good
[24-Oct-2011 17:17:25] <humanix> when i used the first regex, it didnt seem to match at remodel
[24-Oct-2011 17:17:28] <humanix> this matches, and logs says it
[24-Oct-2011 17:17:32] <bseklecki> <humanix> i put the ^(/dev.... one, remodel and it works fine
[24-Oct-2011 17:17:35] <rmatte> cool
[24-Oct-2011 17:17:38] <bseklecki> errr sorry
[24-Oct-2011 17:17:42] <rmatte> lol
[24-Oct-2011 17:43:22] <dhopp> in the dmd..if I wanted to show all devices that are part of a Group what would be the syntax?
[24-Oct-2011 17:53:34] <rmatte> dhopp: one second, it's simple...
[24-Oct-2011 17:54:18] <rmatte> Example for a group called Network...
[24-Oct-2011 17:54:20] <rmatte> >>> for d in dmd.Groups.Network.getSubDevices():
[24-Oct-2011 17:54:20] <rmatte> ... d.id
[24-Oct-2011 17:56:14] <rmatte> or...
[24-Oct-2011 17:56:16] <rmatte> >>> for d in dmd.Groups.getObjByPath('Network').getSubDevices():
[24-Oct-2011 17:56:16] <rmatte> ... d.id
[24-Oct-2011 18:06:32] <rmatte> You can also do...
[24-Oct-2011 18:06:32] <rmatte> for d in dmd.Groups.getObjByPath('Network').getObjByPath('Test Sub Group').getSubDevices():
[24-Oct-2011 18:06:41] <rmatte> helps with groups that have spaces in the names
[24-Oct-2011 20:26:59] kocolosk|away is now known as kocolosk
[25-Oct-2011 00:00:01] [disconnected at Tue Oct 25 00:00:01 2011]
[25-Oct-2011 00:00:02] [connected at Tue Oct 25 00:00:02 2011]
[25-Oct-2011 00:00:16] [zenoss-logger (logger bot) has joined #zenoss]
[25-Oct-2011 05:27:51] <ali3n0> hi folks... wondering, what happens to graph + old events when a production state is set to decommissioned?
[25-Oct-2011 05:35:32] <ali3n0> and what does the device status "up" mean if it is not supposed to be monitored anymore?
[25-Oct-2011 08:30:18] <froztbyte> ali3n0: decommissioned means it won't form part of alerts or such
[25-Oct-2011 08:30:28] <froztbyte> essentially that it's basically ignored
[25-Oct-2011 08:31:12] <ali3n0> frozbyte: yes, but I found that you need to push changes too, cfr: message/61470
[25-Oct-2011 08:40:56] kocolosk is now known as kocolosk|away
[25-Oct-2011 09:21:02] <froztbyte> ali3n0: ah, you didn't indicate you were poking at things from !webinterface
[25-Oct-2011 09:21:10] <froztbyte> I don't suppose anyone from Zenoss upstream is here?
[25-Oct-2011 09:21:25] <ali3n0> forztbyte I'm not, I'm pushing it via api
[25-Oct-2011 09:21:35] <ali3n0> ah sorr, form !
[25-Oct-2011 09:21:44] <ali3n0> got it
[25-Oct-2011 09:31:19] kocolosk|away is now known as kocolosk
[25-Oct-2011 09:45:37] <kenhen93> hi, can anyone PLEASE help with a zenhub error partaining to a parser, componentscanValue
[25-Oct-2011 09:45:57] <kenhen93> I know my regex is right now and I have hit a wall
[25-Oct-2011 09:47:40] <kenhen93> i cannot find any relevant information on the forum about this particular error
[25-Oct-2011 09:57:07] <kenhen93> return dict(componentScanValue = getattr(context, self.componentScanValue)) AttributeError: ck_instance
[25-Oct-2011 10:34:38] <syoma> hello
[25-Oct-2011 10:34:51] <syoma> I've problem with upgrade to 3.2.1
[25-Oct-2011 10:35:18] <syoma> warning: zenoss-3.2.1.el4.i386.rpm: V3 DSA signature: NOKEY, key ID aa5a1ad7
[25-Oct-2011 10:35:18] <syoma> Preparing... ########################################### [100%]
[25-Oct-2011 10:35:18] <syoma> ====================================================================
[25-Oct-2011 10:35:18] <syoma> Your system is not yet ready for upgrade.
[25-Oct-2011 10:35:20] <syoma> Please contact Zenoss support for help upgrading your
[25-Oct-2011 10:35:22] <syoma> RelStorage-enabled ZODB [TxnHistoryFree schema].
[25-Oct-2011 10:35:25] <syoma> ====================================================================
[25-Oct-2011 10:35:27] <syoma> error: %pre(zenoss-3.2.1-1326.el4.i386) scriptlet failed, exit status 1
[25-Oct-2011 10:35:30] <syoma> error: install: %pre scriptlet failed (2), skipping zenoss-3.2.1-1326.el4
[25-Oct-2011 10:35:40] <syoma> anybody can help me? XD
[25-Oct-2011 10:36:04] <bseklecki> hmmmm
[25-Oct-2011 10:38:18] <syoma> lelola, juas !
[25-Oct-2011 10:39:12] <syoma> bseklecki: the os is centos 4.9 :/
[25-Oct-2011 10:42:07] <syoma> why reason my system is not yet ready for the upgrade?
[25-Oct-2011 10:43:54] <syoma> is possible to do a rollback to zenoss 3.2.0 :¬
[25-Oct-2011 10:45:58] <dhopp> syoma: it is if you have a backup
[25-Oct-2011 10:46:15] <syoma> dhopp: I've problem with the collectors
[25-Oct-2011 10:46:26] <syoma> only the collector can not be upgrade
[25-Oct-2011 10:46:31] <syoma> to zenoss 3.2.1
[25-Oct-2011 10:47:22] <syoma> sorry for the flood
[25-Oct-2011 10:47:25] <syoma> # rpm -Uvh --nodeps zenoss-3.2.1.el5.i386.rpm
[25-Oct-2011 10:47:25] <syoma> warning: zenoss-3.2.1.el5.i386.rpm: Header V3 DSA signature: NOKEY, key ID aa5a1ad7
[25-Oct-2011 10:47:28] <syoma> Preparing... ########################################### [100%]
[25-Oct-2011 10:47:31] <syoma> ====================================================================
[25-Oct-2011 10:47:33] <syoma> Your system is not yet ready for upgrade.
[25-Oct-2011 10:47:36] <syoma> Please contact Zenoss support for help upgrading your
[25-Oct-2011 10:47:38] <syoma> RelStorage-enabled ZODB [TxnHistoryFree schema].
[25-Oct-2011 10:47:41] <syoma> ====================================================================
[25-Oct-2011 10:47:42] <syoma> the same behavior
[25-Oct-2011 10:47:44] <syoma> error: %pre(zenoss-3.2.1-1326.el5.i386) scriptlet failed, exit status 1
[25-Oct-2011 10:47:47] <syoma> error: install: %pre scriptlet failed (2), skipping zenoss-3.2.1-1326.el5
[25-Oct-2011 10:47:53] <syoma> with centos 5
[25-Oct-2011 10:48:37] <dhopp> is this enterprise or core?
[25-Oct-2011 10:48:40] <syoma> core
[25-Oct-2011 10:48:43] <syoma>
[25-Oct-2011 10:48:47] <syoma> I don't have money
[25-Oct-2011 10:48:49] <syoma> jajaj
[25-Oct-2011 10:50:09] <syoma> do you have some idea?
[25-Oct-2011 10:53:17] <syoma> oops
[25-Oct-2011 10:53:19] <syoma> chown: cannot access `Monitor)-QoS': No such file or directory
[25-Oct-2011 10:53:20] <syoma> chown: cannot access `Packet': No such file or directory
[25-Oct-2011 10:53:20] <syoma> chown: cannot access `Scheduler-0000/ifOutOctets_ifOutOctets.rrd': No such file or directory
[25-Oct-2011 10:53:22] <syoma> chown: cannot access `/opt/zenoss/perf/Devices/10.1.9.22/os/interfaces/WAN': No such file or directory
[25-Oct-2011 10:53:26] <syoma> chown: cannot access `Miniport': No such file or directory
[25-Oct-2011 10:53:28] <syoma> chown: cannot access `(Network': No such file or directory
[25-Oct-2011 10:53:31] <syoma> chown: cannot access `Monitor)-QoS': No such file or directory
[25-Oct-2011 10:53:33] <syoma> chown: cannot access `Packet': No such file or directory
[25-Oct-2011 10:53:36] <syoma> !
[25-Oct-2011 10:53:38] <syoma> FUCK
[25-Oct-2011 10:53:41] <syoma>
[25-Oct-2011 10:53:44] <syoma> zenoss upgrade DESTROY MY PLATFORM
[25-Oct-2011 10:53:57] <syoma> juajuajua
[25-Oct-2011 10:56:03] <dhopp> syoma: the reason why I asked if it was core or enterprise was because you had "RelStorage-enabled ZODB [TxnHistoryFree schema]." and I don't remember seeing that before
[25-Oct-2011 10:56:09] <dhopp> and RelStorage is an Enterprise thing
[25-Oct-2011 10:56:33] <syoma> ;/
[25-Oct-2011 11:17:42] <w0lv3> (this can take a few minutes) maybe ?
[25-Oct-2011 11:18:22] <w0lv3> Loading initial Zenoss objects into the Zeo database when install zenoss 3.2 in centos
[25-Oct-2011 11:18:26] <davetoo> how big/fast is your machine?
[25-Oct-2011 11:19:04] <davetoo> How long has it been?
[25-Oct-2011 11:19:29] <w0lv3> five min.
[25-Oct-2011 11:24:25] <Uchiha_Itachi> i'm having a problem with zenoss, i added devices that work just fine, if i check then on
[25-Oct-2011 11:24:42] <Uchiha_Itachi> Infrastructure i see them fine
[25-Oct-2011 11:25:00] <Uchiha_Itachi> but every time i add an even to check on them, they go off
[25-Oct-2011 11:26:27] <davetoo> w0lv3: finished yet?
[25-Oct-2011 11:26:42] <davetoo> Is this on a VM?
[25-Oct-2011 11:29:50] <Uchiha_Itachi> davetoo, asking me?
[25-Oct-2011 11:39:27] <syoma> zenperfsnmp show me the following error
[25-Oct-2011 11:39:30] <syoma> File "/opt/zenoss/Products/ZenUtils/Utils.py", line 244, in getObjByPath
[25-Oct-2011 11:39:30] <syoma> next = bobo_traverse(REQUEST, name)
[25-Oct-2011 11:39:30] <syoma> File "/opt/zenoss/lib/python/OFS/Application.py", line 148, in __bobo_traverse__
[25-Oct-2011 11:39:33] <syoma> raise KeyError, name
[25-Oct-2011 11:39:36] <syoma> KeyError: '10.3.0.57'
[25-Oct-2011 11:39:41] <w0lv3> thanks davetoo, finished install.
[25-Oct-2011 11:39:46] <syoma> 2011-10-25 12:37:45,404 INFO zen.zenperfsnmp: Daemon zenperfsnmp shutting down
[25-Oct-2011 11:40:35] <w0lv3> ...and yes, i running in Vmachine
[25-Oct-2011 11:44:36] <syoma> when I enable distributed collector
[25-Oct-2011 11:46:36] <w0lv3> syoma: See if this helps docs/DOC-2535
[25-Oct-2011 11:50:35] <syoma> thank you
[25-Oct-2011 11:54:27] <aclark> hi all
[25-Oct-2011 11:55:47] <aclark> i've just set zMonitor to False for /IpService/Privileged/serviceclasses/smtp
[25-Oct-2011 11:56:06] <aclark> but Zenoss is still alerting me about smtp…
[25-Oct-2011 11:56:27] <aclark> do i need to restart something?
[25-Oct-2011 11:58:09] <kenhen93> is there a log that will tell me why my zenpack will not install
[25-Oct-2011 11:58:19] <kenhen93> when i run -v10 option it just shows me the event creation
[25-Oct-2011 11:58:53] <dhopp> aclark: you probably need to restart zenstatus and maybe zenhub
[25-Oct-2011 11:59:01] <aclark> dhopp: thx
[25-Oct-2011 11:59:15] <dhopp> kenhen93: are you installing via the gui or command line?
[25-Oct-2011 11:59:22] <kenhen93> command
[25-Oct-2011 11:59:49] <dhopp> if it isn't installing it should give you an error
[25-Oct-2011 11:59:59] <dhopp> which zenpack is it?
[25-Oct-2011 12:00:01] [disconnected at Tue Oct 25 12:00:01 2011]
[25-Oct-2011 12:00:02] [connected at Tue Oct 25 12:00:02 2011]
[25-Oct-2011 12:00:16] [zenoss-logger (logger bot) has joined #zenoss]
[25-Oct-2011 12:00:28] <kenhen93> a custom one with a few monitoring templates from another zenoss core
[25-Oct-2011 12:00:36] <kenhen93> basically moving it from one DC to another
[25-Oct-2011 12:01:14] <kenhen93> this is the first thing that comes from the command line install
[25-Oct-2011 12:01:16] <kenhen93> DEBUG:zen.Events:=============== incoming event ===============
[25-Oct-2011 12:03:01] <dhopp> can you pastebin the output of just doing zenpack —install (without -v10)
[25-Oct-2011 12:04:04] <kenhen93> when I do it without -v10 it shows no output
[25-Oct-2011 12:04:35] <dhopp> it shows nothing?
[25-Oct-2011 12:04:42] <dhopp> it just returns to a prompt?
[25-Oct-2011 12:04:55] <kenhen93> right
[25-Oct-2011 12:05:01] <dhopp> what is the exact command you are using? and are you doing it as the zenoss user?
[25-Oct-2011 12:05:06] <kenhen93> but i do not see the pack installed in the GUI
[25-Oct-2011 12:05:19] <dhopp> it should give you some output ..even in a succesful install
[25-Oct-2011 12:05:23] <kenhen93> zenpack --install=ZenPacks.blah.blah
[25-Oct-2011 12:05:26] <dhopp> *successful
[25-Oct-2011 12:05:54] <dhopp> and you do it as the zenoss user?
[25-Oct-2011 12:06:39] <kenhen93> yes
[25-Oct-2011 12:08:52] <kenhen93> zenoss hates me today
[25-Oct-2011 12:11:16] <kenhen93> does anyone know how to move zenoss default tmp directory?
[25-Oct-2011 12:11:30] <kenhen93> i need it to be /var/tmp, not /tmp
[25-Oct-2011 12:11:41] <dhopp> kenhen93: try taking the '=' out
[25-Oct-2011 12:11:44] <dhopp> so
[25-Oct-2011 12:11:51] <dhopp> zenpack —install ZenPacks.blah.blah
[25-Oct-2011 12:11:57] <kenhen93> ok
[25-Oct-2011 12:13:20] <kenhen93> same output
[25-Oct-2011 12:13:28] <kenhen93> nothing
[25-Oct-2011 12:13:54] <dhopp> something isn't right….how big is the egg file?
[25-Oct-2011 12:14:09] <dhopp> I'm wondering if for some reason you have a blank egg..i.e. nothing in it to add
[25-Oct-2011 12:15:08] <kenhen93> -rw-r--r-- 1 zenoss zenoss 11439 Oct 25 11:38 ZenPacks.blah.blah-1.0-py2.6.egg
[25-Oct-2011 12:16:26] <dhopp> kenhen93: what happens if you do zenpack —install 1234 ?
[25-Oct-2011 12:16:35] <dhopp> I'm hoping you get an error message
[25-Oct-2011 12:17:40] <kenhen93> i appreciate your help, give a few minutes to try that, I have a really chatty server, so I stopped zenoss and I am deleting all the events from status
[25-Oct-2011 12:27:54] <JohnnyNOC_> hello everybody
[25-Oct-2011 12:27:59] <JohnnyNOC_> it's Dr. Nick?
[25-Oct-2011 12:32:12] <kenhen93> dhopp: I do get an error specifying a zenpack that does not exist
[25-Oct-2011 12:32:31] <bseklecki> syoma: did someone help you?
[25-Oct-2011 12:32:46] <dhopp> kenhen93: hmm..ok I'm stumped…my initial thinking is something is not right with your zenpack but I'm not sure how to verify that
[25-Oct-2011 12:33:25] <kenhen93> ok, np, thanks for your help
[25-Oct-2011 12:48:30] <Sashness> anyone see the following error before: http://pastebin.com/6GwAfSuq
[25-Oct-2011 12:48:41] <Sashness> ConflictError: database conflict error (oid 0x017d8d, class BTrees.IIBTree.IITreeSet, serial this txn started with 0x0391f7ce6a9fb6aa 2011-10-25 16:46:24.989972, serial currently committed 0x0391f7cec267a455 2011-10-25 16:46:45.563637)
[25-Oct-2011 12:55:46] <JohnnyNOC_> yes
[25-Oct-2011 12:56:16] <JohnnyNOC_> not sure what you were doing to see that but in my experience if i shut down a bunch of the zenoss daemons and restarted that transaction it would complete without errors
[25-Oct-2011 12:57:24] <JohnnyNOC_> i would see errors like that when reindexing zope, or zeopacking
[25-Oct-2011 13:16:31] <aclark> dhopp: hmmm, i restarted. still getting events :-\
[25-Oct-2011 14:02:13] <jmp242> Hey all
[25-Oct-2011 14:02:27] <jmp242> Anyone used the Nagios parser for multiple values?
[25-Oct-2011 14:14:52] <kenhen93> can anyone explain componentScanValue to me? My parser keeps erroring out on this value
[25-Oct-2011 14:21:23] <JohnnyNOC_> jmp242: how do you mean?
[25-Oct-2011 14:21:26] <JohnnyNOC_> are you talking multiple lines?
[25-Oct-2011 14:22:01] <JohnnyNOC_> normally if i have multiple values i like output like: |x=1, y=2, z=3
[25-Oct-2011 14:22:08] <JohnnyNOC_> just as an example
[25-Oct-2011 14:22:09] <JohnnyNOC_>
[25-Oct-2011 14:24:57] <jmp242> I expected it to work as in my FAQ, but I haven't tried it in a while, and this thread claims it may have a bug : message/62122#62122
[25-Oct-2011 14:25:20] <jmp242> maybe I misunderstood
[25-Oct-2011 14:35:08] <cek> Hi. How do you make zenoss shutup on large events, for ex when host goes down, it triggers smpt,http down and all monitored ipservices events.
[25-Oct-2011 14:36:31] <jmp242> cek - search the forums
[25-Oct-2011 14:36:38] <jmp242> there are many discussions on this
[25-Oct-2011 14:37:04] <jmp242> there isn't a way to guarantee that you won't get multiple alerts - it depends on whether Zenoss knows the host is down before the check for the services happens
[25-Oct-2011 14:37:25] <jmp242> there are event transforms available though that sometimes wor
[25-Oct-2011 14:37:26] <jmp242> k
[25-Oct-2011 14:39:03] <cek> it's getting even worse when there's waterfall of events. because of failed router, for ex.
[25-Oct-2011 14:43:31] <jmp242> Hmm, Check the pingtree
[25-Oct-2011 14:43:40] <jmp242> Zenoss should be able to tell Layer 3 dependencies
[25-Oct-2011 14:44:12] <jmp242> assuming, of course again that Zenoss checked the router *before* it tries to reach the devices behind it
[25-Oct-2011 14:45:58] <JohnnyNOC_> isn't there a transform out there for exactly this purpose?
[25-Oct-2011 14:46:02] <JohnnyNOC_> can't seem to find it for some reason
[25-Oct-2011 14:46:43] <cek> next question is how do you monitor router properly so that it won't trigger false alerts
[25-Oct-2011 14:47:19] <jmp242> There is JohnnyNOC_ , I think it may even be in the FAQ
[25-Oct-2011 14:47:41] <cek> didn't know you can write python in web UI in zenoss
[25-Oct-2011 14:47:59] <jmp242> # Change any non-ping, non-clear events from devices that have outstanding
[25-Oct-2011 14:47:59] <jmp242> # ping down events to Severity Info.
[25-Oct-2011 14:47:59] <jmp242> #
[25-Oct-2011 14:47:59] <jmp242> # This transform is sensitive to the event cache timeout. If the ping down
[25-Oct-2011 14:47:59] <jmp242> # event comes in less than "timeout" seconds before the subsequent events, the
[25-Oct-2011 14:48:00] <jmp242> # subsequent events will not be suppressed. You can find the event cache
[25-Oct-2011 14:48:00] <jmp242> # timeout setting on the Event Manager page in the web interface.
[25-Oct-2011 14:48:01] <jmp242> if getattr(evt, 'severity', 0) > 0 \
[25-Oct-2011 14:48:01] <jmp242> and getattr(evt, 'eventClass', '/Unknown') != '/Status/Ping' \
[25-Oct-2011 14:48:02] <jmp242> and device and device.getPingStatus() > 0:
[25-Oct-2011 14:48:02] <jmp242> evt.severity = 2
[25-Oct-2011 14:48:09] <jmp242> though, of course the spacing is killed there
[25-Oct-2011 14:48:26] <jmp242> http://pastebin.com/Vj8KNV5M
[25-Oct-2011 14:48:47] <jmp242> Put that the the top level event class transform
[25-Oct-2011 14:53:20] <JohnnyNOC_> jmp242: i thought there was one specific for routers
[25-Oct-2011 14:53:32] <JohnnyNOC_> if i understand correctly that's only for a specific device
[25-Oct-2011 14:54:06] <JohnnyNOC_> let me have a look
[25-Oct-2011 14:54:06] <JohnnyNOC_>
[25-Oct-2011 14:57:49] <jmp242> oh yea
[25-Oct-2011 14:59:50] <JohnnyNOC_> i think the reason why i never implemented it is because i haven't put much thought into our layer 3 topology
[25-Oct-2011 15:00:01] <jmp242> Yea
[25-Oct-2011 15:00:08] <JohnnyNOC_> on our previous zenoss install there were always log messages about topology errors
[25-Oct-2011 15:00:27] <JohnnyNOC_> but i would like to make it work eventually, god forbid we actually lose a router
[25-Oct-2011 15:00:27] <JohnnyNOC_> :X
[25-Oct-2011 15:06:35] <cek> hmm, no alerts yet while ping is down
[25-Oct-2011 15:07:24] <cek> 1 0 0 0 Down DEVICE STATUS
[25-Oct-2011 15:12:04] <cek> erm... guys, is that transform doing its work properly? i mean, i got the behaviour, but somehow, all those IP services that supposed to go down, they are seen as UP
[25-Oct-2011 15:12:32] <cek> i just brought a host down to test by firewalling it
[25-Oct-2011 15:23:48] <cek> will zenoss support newlines in "send string" on ip services? want to monitor http reponse
[25-Oct-2011 15:25:39] <cek> alright, host absolutely down. Yet still all ip services are green.
[25-Oct-2011 15:25:49] <aclark> ok i *really* want to make these events stop. each device reports False for obj.monitor…
[25-Oct-2011 15:36:54] <jmp242> cek: do you have events for the IP services?
[25-Oct-2011 15:37:02] <jmp242> I'm not sure that it works like you're saying
[25-Oct-2011 15:37:30] <cek> sure i have. they used to alert like crazy.
[25-Oct-2011 15:37:41] <cek> Isn't that automatic host down detection kicked in?
[25-Oct-2011 15:41:37] <dhopp> cek: if the host is down you *should* only get one ping down event. All other services won't alert or change status because zenoss didn't actually check the state of the services because it already knows it can't get to the device
[25-Oct-2011 15:41:40] <cek> yeah, definitely something's going on udner the hood. just firewalled http only, it triggeret an alert
[25-Oct-2011 15:41:53] <cek> dhopp, that explains it
[25-Oct-2011 15:42:44] <cek> thanks for help! btw, is there any "events flapping" module available in zenoss? similar to the one in nagios?...
[25-Oct-2011 15:43:11] <dhopp> cek: the thinking is if I can't get to device then the services must be down. If you have a device that you can't ping test there is a zproperty to tell zenoss to ignore ping
[25-Oct-2011 15:43:18] <dhopp> cek: not that I know of..
[25-Oct-2011 16:14:32] <dhopp> who wants to do my work for me?
[25-Oct-2011 16:14:37] <dhopp> show of hands?
[25-Oct-2011 16:15:01] <jmp242> umm not it
[25-Oct-2011 16:43:08] <JohnnyNOC_> has anyone had disk space threshold alerts show up for events that are below your threshold?
[25-Oct-2011 16:43:28] <JohnnyNOC_> for example, my threshold is 95% but occasionally I'll notice events kick off for filesystems even if they're way under the threshold
[25-Oct-2011 16:47:30] <bseklecki> Jonny: that's a feature, HP sponored it, to try to get you to proactively buy more SAN
[25-Oct-2011 16:53:30] <JohnnyNOC_> hah
[25-Oct-2011 16:53:41] <JohnnyNOC_> Dell must be in on the scam too then
[25-Oct-2011 16:54:16] <JohnnyNOC_> but in all seriousness it's kind of annoying, especially to my users who receive these alerts for filesystem thresholds less than 95%
[25-Oct-2011 16:54:20] <JohnnyNOC_> there's a lot of 'wtf is going on?'
[25-Oct-2011 16:54:21] <JohnnyNOC_>
[25-Oct-2011 16:56:41] kocolosk is now known as kocolosk|away
[25-Oct-2011 17:01:39] <dhopp> JohnnyNOC_: how big is the filesystem?
[25-Oct-2011 17:02:16] <JohnnyNOC_> these are multiple filesystems across multiple servers, i don't think the size of them is the issue
[25-Oct-2011 17:02:28] <JohnnyNOC_> but some as small as 5GB, others as large as a couple hundred GB with this issue
[25-Oct-2011 17:02:38] <JohnnyNOC_> and it seems to be only occasional, and very likely to happen if i bounce zenoss
[25-Oct-2011 17:02:39] <dhopp> JohnnyNOC_: the reason I ask is there was a known problem with very large filesystems
[25-Oct-2011 17:02:40] <dhopp> and snmp
[25-Oct-2011 17:02:58] <JohnnyNOC_> dhopp i seem to have that problem too.. but that's another issue unfortunately
[25-Oct-2011 17:02:59] <JohnnyNOC_>
[25-Oct-2011 17:03:05] <dhopp> windows or linux?
[25-Oct-2011 17:03:09] <dhopp> or does it matter?
[25-Oct-2011 17:03:13] <dhopp> wmi or snmp?
[25-Oct-2011 17:03:18] <JohnnyNOC_> both
[25-Oct-2011 17:03:34] <dhopp> what version of zenoss?
[25-Oct-2011 17:04:11] <JohnnyNOC_> what i'm noticing is it will send all these alerts and eventually they will "catch up" and stop sending those alerts
[25-Oct-2011 17:04:15] <JohnnyNOC_> 3.1.0
[25-Oct-2011 17:05:11] <JohnnyNOC_> but i'm just at a loss for why they're sent in the first place, my assumption is somehow/someway the usedBlocks aren't being calculated properly
[25-Oct-2011 17:10:32] <bseklecki> yea if you're running 2 PB ZFS store then you might run into some SNMP data abstraction problems
[25-Oct-2011 17:10:53] <bseklecki> you'll have to ditch oracle at that point and call Sun
[25-Oct-2011 17:11:20] <JohnnyNOC_> yea that would be sweet, except i see this on 2TB filesystems let alone 2PB
[25-Oct-2011 17:38:40] kocolosk|away is now known as kocolosk
[25-Oct-2011 17:45:04] <nexex> bseklecki: ditch oracle at that point and call Sun -- whaaat?
[25-Oct-2011 17:55:52] kocolosk is now known as kocolosk|away
[25-Oct-2011 18:34:57] <densone> Hey Zenoss Friends. Wondering if it's possible to disable ping in the zenoss http plugin?
[25-Oct-2011 18:40:07] <densone> zPingMonitorIgnore?
[25-Oct-2011 18:40:12] <densone> I think I found it.
[25-Oct-2011 18:40:58] <densone> So on to my next http monitor question. Is it possible to test a specific uri.
[25-Oct-2011 18:41:12] <densone> I can't find this in the documentation.
[25-Oct-2011 18:42:33] kocolosk|away is now known as kocolosk
[25-Oct-2011 18:52:17] <densone> Found the datasource. \
[25-Oct-2011 18:52:25] <densone> This is all starting to make sense now.
[25-Oct-2011 18:53:15] kocolosk is now known as kocolosk|away
[25-Oct-2011 19:28:24] densone_ is now known as densone
[25-Oct-2011 20:24:20] <densone> I am on a Mac and Am logged in as Admin to Zenoss. I am unable to add users. Anyone seen this behavior before?
[25-Oct-2011 20:25:50] <densone> Under Advanced --> Users. there is no sprocket to add a new user.
[25-Oct-2011 20:28:42] <densone> Using zenoss 3.2.1
[25-Oct-2011 20:46:48] densone_ is now known as densone
[25-Oct-2011 21:12:05] kocolosk|away is now known as kocolosk
[25-Oct-2011 22:57:41] <zaoul> Hi. I see that zenoss is populating networks.. even internal private ones. I have two machines that have the same logical lan address but are actually different physical private networks. When I look at the map it shows they are connected. What can I do?
[26-Oct-2011 00:00:01] [disconnected at Wed Oct 26 00:00:01 2011]
[26-Oct-2011 00:00:01] [connected at Wed Oct 26 00:00:01 2011]
[26-Oct-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[26-Oct-2011 05:00:01] <ali3n0> hi folks. I'm trying to understand how to different thresholds for disk usage, do I have to create a custom template than add data sources copying from existing ones of is there a better way to do it?
[26-Oct-2011 05:13:30] <enkrypt> anyone awake to ask highly complex zenoss enterprise debugging questions?
[26-Oct-2011 05:20:53] <ali3n0> uh
[26-Oct-2011 05:21:08] <ali3n0> I'm awake but not yet ready for that kind of questions
[26-Oct-2011 05:21:44] <enkrypt> ah
[26-Oct-2011 05:48:58] <ali3n0> I'm getting troubles with setInfo() api call
[26-Oct-2011 05:49:28] <ali3n0> I get a success response but I can't see any change on propriety
[26-Oct-2011 05:58:27] <enkrypt> in the zenoss collector performance graphs, what does the Config time one mean? my config times are HUGE the past few days
[26-Oct-2011 08:41:00] <jb> morning.
[26-Oct-2011 08:44:19] <jmp242> morning
[26-Oct-2011 10:02:37] kocolosk is now known as kocolosk|away
[26-Oct-2011 10:15:19] kocolosk|away is now known as kocolosk
[26-Oct-2011 10:18:16] <jmp242> hey
[26-Oct-2011 10:18:29] <jmp242> Does 3.2.x change the zproperty screen?
[26-Oct-2011 10:18:29] <davetoo> ho
[26-Oct-2011 10:18:44] <jmp242> such that you can't type in values, but have a drop down?
[26-Oct-2011 10:18:56] <davetoo> I .. haven't actually run it yet :/
[26-Oct-2011 10:19:15] <rocket> jmp242: you mean the filtered zproperty screen?
[26-Oct-2011 10:20:30] <jmp242> message/62182#62182
[26-Oct-2011 10:20:38] <jmp242> this thread is showing a UI I don't see at all in 3.1
[26-Oct-2011 10:20:44] <jmp242> so I can't really help
[26-Oct-2011 10:23:06] <cooper6581> Hello all, sorry for the noobish question, but I couldn't find this anywhere. Do you know if it's possible to access the SNMP data via the device context? For example, I would like to do here.snmp.memCached.memCached
[26-Oct-2011 10:23:35] <cooper6581> I'm basically trying to setup a threshold for mem available, that takes cache into account
[26-Oct-2011 10:25:16] <rocket> jmp242: there are only 5 severities
[26-Oct-2011 10:25:29] <rocket> thread/4896
[26-Oct-2011 10:25:42] <rocket> The #### is the zWinEventlogMinSeverity. You can reference the MSDN
[26-Oct-2011 10:25:42] <rocket> site at http://msdn2.microsoft.com/en-us/library/aa394226.aspx to see
[26-Oct-2011 10:25:42] <rocket> what the different EventType fields mean, but there are only five of
[26-Oct-2011 10:25:42] <rocket> them:
[26-Oct-2011 10:25:42] <rocket> 1 = Error
[26-Oct-2011 10:25:44] <rocket> 2 = Warning
[26-Oct-2011 10:25:47] <rocket> 3 = Information
[26-Oct-2011 10:25:49] <rocket> 4 = Security Audit Success
[26-Oct-2011 10:25:52] <rocket> 5 = Security Audit Failure
[26-Oct-2011 10:26:26] <rocket> and yes that ui did change in 3.2 to a searchable grid etc .. its not only raw text anymore .. we are trying to enforce some input validation where it makes sense
[26-Oct-2011 10:26:39] <jmp242> Yes I recall there only being 5, but for some reason back in the day they were set by
[26-Oct-2011 10:26:41] <jmp242> 1 = Error
[26-Oct-2011 10:26:42] <jmp242> 2 = Warning
[26-Oct-2011 10:26:42] <jmp242> 4 = Information
[26-Oct-2011 10:26:42] <jmp242> 8 = Security audit success
[26-Oct-2011 10:26:42] <jmp242> 16 = Security audit failure
[26-Oct-2011 10:26:54] <jmp242> so that's changed, though I don't think I can change the FAQ there
[26-Oct-2011 10:27:04] <jmp242> it starts eating later entries last time I tried
[26-Oct-2011 10:28:30] <Jane_Curry> cooper6851: I doubt that you can do that. If I am understanding you correctly you are trying to access rrd data with your suggested syntax??
[26-Oct-2011 10:33:36] <cooper6581> Jane_Curry: Yeah, either that, or to make a new data point which is the addition of two seperate SNMP values (memAvailReal + memCached)
[26-Oct-2011 10:34:43] <cooper6581> or, if it's better that I use a command which runs a script and gets the value
[26-Oct-2011 10:35:59] <cooper6581> ultimately I'm trying to set a threashold for (memoryFree + cached) * 100 / total
[26-Oct-2011 10:36:35] <cooper6581> because almost all of my devices are throwing memory alerts since the cache is taking up most the physical memory
[26-Oct-2011 10:36:52] <cooper6581> sorry for the long winded explination
[26-Oct-2011 10:37:41] <cooper6581> This guy explains the exact same problem message/5546
[26-Oct-2011 10:56:03] <dhopp> cooper6581: docs/DOC-10224
[26-Oct-2011 10:56:17] <dhopp> that will make a data source that is a combination of other ones
[26-Oct-2011 10:56:32] <dhopp> however, be warned that it uses command datasources and can have quite a bit of overhead
[26-Oct-2011 10:57:37] <cooper6581> dhopp: awesome, thanks so much!
[26-Oct-2011 10:58:02] <fragfutter> cooper6581: linux will always use all memory available. start monitoring your swap. But do not monitor a swap of zero, because even with free memory linux may decide to page out memory.
[26-Oct-2011 10:58:24] <cooper6581> fragfutter: Normally, I would definitely do that, but all our machines are headless, so no swap partition
[26-Oct-2011 10:58:39] <dhopp> how does headless mean no swap?
[26-Oct-2011 10:58:41] <cooper6581> the root fs is just on tempfs in ram
[26-Oct-2011 10:58:51] <cooper6581> there are no hds
[26-Oct-2011 10:58:57] <dhopp> oh gotcha
[26-Oct-2011 10:59:14] <fragfutter> dhopp: no.
[26-Oct-2011 10:59:27] <dhopp> fragfutter: ?
[26-Oct-2011 10:59:27] * fragfutter must search for a server with a head...
[26-Oct-2011 10:59:46] <fragfutter> dhopp: headless does not necessarily mean no swap.
[26-Oct-2011 11:00:11] <dhopp> fragfutter: I understand..hence why I was posing the quest
[26-Oct-2011 11:00:15] <cooper6581> sorry, I should have been more specific
[26-Oct-2011 11:00:19] <dhopp> *question
[26-Oct-2011 11:00:49] <fragfutter> dhopp: yes. i missed the "how" from your sentence.
[26-Oct-2011 11:01:28] <dhopp> fragfutter: well it was more of a question rather than a sentence :-P
[26-Oct-2011 11:01:38] <fragfutter> *argh*
[26-Oct-2011 11:02:36] <fragfutter> on diskless systems i normaly have only one non-system process. So just monitor this one.
[26-Oct-2011 11:09:27] <dhopp> is it possible to change the font color in the events console? I know I can change it to highlight the rows in severity color but wondering if I could then change the text color too
[26-Oct-2011 11:13:26] <psyklone> dhopp - what version?
[26-Oct-2011 11:16:31] <psyklone> dhopp, go check out $ZENHOME/Products/ZenUI3/browser/resources/css/zenoss.css
[26-Oct-2011 11:16:38] <psyklone> line 5
[26-Oct-2011 11:19:35] <dhopp> psyklone: I was hoping to not have to modify the css since I would then have to change it
[26-Oct-2011 11:19:54] <dhopp> and I really only want this for one particular user that will be used to show the event console on some plasmas in the NOC
[26-Oct-2011 11:20:27] <dhopp> I mean I would then have to change it everytime I upgrade zenoss
[26-Oct-2011 11:20:48] <dhopp> it's not the end of the world..but was something I was asked
[26-Oct-2011 11:21:07] <ve2dmn> I have a "small" problem.
[26-Oct-2011 11:21:58] <ve2dmn> I physically moved a device 3 days ago, and the graphs still shows NaN...
[26-Oct-2011 11:22:29] <ve2dmn> They were working great before.
[26-Oct-2011 11:23:19] <ve2dmn> The device in question is a 24ports D-link switch
[26-Oct-2011 11:23:21] <psyklone> dhopp - ah, yeah. the CSS is where that's set, not sure if a user-specific way to have that changed
[26-Oct-2011 11:23:59] <ve2dmn> I can do the whole snmpwalk without any problems....
[26-Oct-2011 11:24:23] <dhopp> ve2dmn: you don't have a ping down event active for the device do you?
[26-Oct-2011 11:24:43] <ve2dmn> but most of the files are missing in $ZENHOME/perf/
[26-Oct-2011 11:25:16] <ve2dmn> dhopp: Actually, I deletd the old device and re-added it, which may be part of the problem
[26-Oct-2011 11:25:30] <dhopp> ve2dmn: yeah that could do it
[26-Oct-2011 11:26:05] <ve2dmn> So now I have 2 folders in perf/Devices/
[26-Oct-2011 11:26:22] <ve2dmn> And I'm not 100% sure how to correct this problem....
[26-Oct-2011 11:27:00] <dhopp> you have 2 folders? what are they?
[26-Oct-2011 11:28:07] <ve2dmn> I still have perf/Devices/Switch1 and perf/Devices/192.168.1.250 (the new IP)
[26-Oct-2011 11:28:33] <dhopp> is Switch1 the old name?
[26-Oct-2011 11:28:49] <dhopp> and are there any files in 192.168.1.250?
[26-Oct-2011 11:28:59] <ve2dmn> "some"
[26-Oct-2011 11:30:10] <ve2dmn> For exemple, " sysUpTime_sysUpTime.rrd" is missing, but 192.168.1.250/os/interfaces/GbE_23/ifOperStatus_ifOperStatus.rrd is there
[26-Oct-2011 11:30:25] <dhopp> can you try to remodel the device?
[26-Oct-2011 11:30:27] <dhopp> any errors?
[26-Oct-2011 11:30:47] <ve2dmn> I tried re-modeling the devices 3 or 4 times...
[26-Oct-2011 11:30:56] <dhopp> were there any errors?
[26-Oct-2011 11:31:19] <dhopp> is this the only device in your zenoss install?
[26-Oct-2011 11:32:09] <ve2dmn> I have 4 linux servers, 3 switches and 1 router. The rest works fine
[26-Oct-2011 11:32:42] <dhopp> can you pastebin the output of running the following as the zenoss user:
[26-Oct-2011 11:32:55] <dhopp> zenperfsnmp run -v10 -d 192.168.1.250
[26-Oct-2011 11:37:31] <dhopp> ok I have kind of an interesting problem….we have linux servers and the lo interface is being detected and the speed is set to 10 Mbit. We have tomcat behind apache and apache talks to tomcat via localhost so there is quite a bit of comunication on this interface…right now I have a server reporting 20 mbits/second on the interface that zenoss thinks is 10 Mbit…anyway to fix this? I don't want to override the interface template for every lo interface of
[26-Oct-2011 11:37:58] <dhopp> I guess I could just add lo as an interface to ignore
[26-Oct-2011 11:38:08] <ve2dmn> ran... just want to remove the name of the server and company I work for
[26-Oct-2011 11:39:13] <ve2dmn> http://pastebin.com/5zDrKyW5
[26-Oct-2011 11:39:36] <ve2dmn> (the IPs are correct btw)
[26-Oct-2011 11:40:19] <dhopp> 2011-10-26 11:34:40,224 DEBUG zen.zenperfsnmp: Device 192.168.2.230 timed out
[26-Oct-2011 11:40:43] <ve2dmn> i know... why?
[26-Oct-2011 11:40:54] <dhopp> can you ping 192.168.2.230 from the zenoss server?
[26-Oct-2011 11:41:01] <dhopp> and wait...
[26-Oct-2011 11:41:07] <dhopp> you originally asked about 192.168.1.250
[26-Oct-2011 11:41:10] <dhopp> what is 2.230?
[26-Oct-2011 11:41:24] <ve2dmn> --- 192.168.2.230 ping statistics ---
[26-Oct-2011 11:41:24] <ve2dmn> 8 packets transmitted, 8 received, 0% packet loss, time 7002ms
[26-Oct-2011 11:41:25] <ve2dmn> rtt min/avg/max/mdev = 1.500/23.332/50.041/20.129 ms
[26-Oct-2011 11:41:49] <ve2dmn> I'm sorry, I was going from memory before. the correct IP is 2.230
[26-Oct-2011 11:41:55] <dhopp> ok
[26-Oct-2011 11:42:18] <dhopp> what happens if you do this from the zenoss server:
[26-Oct-2011 11:42:40] <dhopp> snmpwalk -v2c -c(whatever your snmp community is) 192.168.2.230
[26-Oct-2011 11:43:32] <dhopp> oh actually…your zenperfsnmp was successful..however it took 5 tries to connect
[26-Oct-2011 11:44:09] <dhopp> is your zenoss server in the same subnet (192.168.2.xxx)?
[26-Oct-2011 11:44:46] <dhopp> are you using snmp version 1 or 2?
[26-Oct-2011 11:44:51] <dhopp> in zenoss
[26-Oct-2011 11:44:52] <ve2dmn> only 1 error:
[26-Oct-2011 11:44:56] <ve2dmn> Error: OID not increasing: UDP-MIB::udpLocalAddress.192.168.2.230.64518
[26-Oct-2011 11:44:56] <ve2dmn> >= UDP-MIB::udpLocalAddress.192.168.2.230.64515
[26-Oct-2011 11:45:00] <ve2dmn> using v2
[26-Oct-2011 11:45:40] <ve2dmn> Everything is in the same subnet. We use the 192.168.1.X for tests
[26-Oct-2011 11:45:57] <dhopp> can you verify that zSnmpVer on this device is in fact set to v2c ?
[26-Oct-2011 11:47:04] <ve2dmn> give me 1 min, I have to connect to the VPN for that... (working from home)
[26-Oct-2011 11:47:32] <dhopp> ve2dmn: wait..how are you looking at zenoss then?
[26-Oct-2011 11:47:43] <ve2dmn> via a ssh tunnel
[26-Oct-2011 11:47:43] <dhopp> zSnmpVer is a zproperty in zenoss
[26-Oct-2011 11:47:50] <dhopp> oh gotcha
[26-Oct-2011 11:49:30] <bigegor> hi all, i've updated SQLDataSource ZenPack, please test it. people/bigegor/blog/2011/10/26/sqldatasource-zenpack-20-beta5
[26-Oct-2011 11:49:39] <ve2dmn> same way I'm connected to the server
[26-Oct-2011 11:58:21] <ve2dmn> Everything seems to be set to v1 by default, both in Zenoss and the switch
[26-Oct-2011 12:00:01] [disconnected at Wed Oct 26 12:00:01 2011]
[26-Oct-2011 12:00:02] [connected at Wed Oct 26 12:00:02 2011]
[26-Oct-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]
[26-Oct-2011 12:04:54] <ve2dmn> I searched the logs yesterday over and over again and I couldn't find anything obvious
[26-Oct-2011 12:06:06] <ve2dmn> Of course, I'll be the first one to admit that I'm don't know Zenoss well.
[26-Oct-2011 12:28:45] <nexex> bigegor: any specific install/update order needed?
[26-Oct-2011 12:34:39] <bigegor> nexex: SQLDataSource can be simply updated, if you want WMIDataSource too, stop zenperfwmi daemon befor updating.
[26-Oct-2011 12:35:42] <bigegor> nexex: ZenODBC must be updated too servlet/JiveServlet/download/38-3604/ZenPacks.community.ZenODBC-3.90.egg.zip
[26-Oct-2011 13:05:57] <nexex> bigegor: thanks
[26-Oct-2011 13:06:05] <bigegor> np
[26-Oct-2011 14:00:37] <jmp242> hey all
[26-Oct-2011 14:00:45] <bigegor> hi
[26-Oct-2011 14:02:38] <zaoul1> Hi. I see that zenoss is populating networks.. even internal private ones. I have two machines that have the same logical lan address but are actually different physical private networks. When I look at the map it shows they are connected. What can I do?
[26-Oct-2011 14:03:27] <jmp242> Anyone ever see a Zenoss install keep looping with the setup wizard?
[26-Oct-2011 14:12:13] <kenhen93> how do I register a component within Zenoss?
[26-Oct-2011 14:23:28] <zaoul1> Anyone recommend a SAS/RAID card? I ask here because I need to know more of which cards have the most robust SNMP for monitoring. Currently I have been using LSI and its total junk, cant even get machines to post sometimes.
[26-Oct-2011 14:23:41] <zaoul1> The adaptecs should be a defacto in this area eh?
[26-Oct-2011 14:24:15] <zaoul1> Ideally I would like to have a MIB for monitoring than write a cruft script to an extend, which tells me if the array is degraded.
[26-Oct-2011 14:27:10] <jmp242> LSI is pretty defacto zaoul1 as far as I know
[26-Oct-2011 14:27:20] <jmp242> IBM uses them for instance in many System X servers
[26-Oct-2011 14:27:30] <jmp242> but we generally use software RAID now
[26-Oct-2011 14:34:46] <zaoul1> werid, I've tried gigabyte, msi, supermicro boards with lsi 9420s and its an 80/20 sucess rate, from no post to not showing disks. to smashing your head flashing and praying.
[26-Oct-2011 14:39:35] <kenhen93> buy a dell or hp and use the extended zenpacks to monitor the RAID card
[26-Oct-2011 14:41:24] <zaoul2> Thad be too easy. Plus im just a lowly tech... saying to 'b' word to management is a waste of time.
[26-Oct-2011 14:41:49] <zaoul2> I've tried 'acquire' recently though.. seems to not give such a brain hemmorage
[26-Oct-2011 14:42:03] <zaoul2> ugh, net...
[26-Oct-2011 14:44:06] <kenhen93> can someone please help me with my parser? I have opened a topic in the community and searched and I cannot find an answer to my problem
[26-Oct-2011 14:47:00] <kenhen93> I know the Regex is right
[26-Oct-2011 14:50:27] <JohnnyNOC> if i make a change to a monitoring template, say, disable a threshold, what daemon(s) do i have to restart so the alerts stop?
[26-Oct-2011 14:50:34] <JohnnyNOC> bounced zenhub, and zopectl, but they're still coming in
[26-Oct-2011 14:50:41] <JohnnyNOC> :[
[26-Oct-2011 15:24:06] <jmp242> MM
[26-Oct-2011 15:32:10] <dhopp> johnnyNOC: you probably need to restart which ever daemon would be responsible for that data source…so zenperfsnmp or zenjmx, etc.
[26-Oct-2011 15:32:49] <dhopp> JohnnyNOC: if this template is only on one device you might be able to just do a "push changes" on the device
[26-Oct-2011 17:44:17] kocolosk is now known as kocolosk|away
[26-Oct-2011 19:41:35] kocolosk|away is now known as kocolosk
[27-Oct-2011 00:00:01] [disconnected at Thu Oct 27 00:00:01 2011]
[27-Oct-2011 00:00:01] [connected at Thu Oct 27 00:00:01 2011]
[27-Oct-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]
[27-Oct-2011 00:13:31] <Tabmow> Is there an easy way to roll up alerts/errors? For example, if a HTTPD process is down, but the server itself is down also, I just want the 1 alert saying the server is down. Also when it comes back up I would like the reverse to also occur.
[27-Oct-2011 00:17:38] <tvincent> Google zenoss event transformation
[27-Oct-2011 00:50:04] <nyeates> An easy way to do that? buy Zenoss enterprise and use our new service-based monitoring capabilities :-p Lets you define your services and then down into your infrastructure and components, etc. You set dependancies and definitions for failure or at risk or up
[27-Oct-2011 00:51:10] <nyeates> in Core - yes, event transforms can do this - though, you are basically writting python code to tell specific events when to show and how to bubble up
[27-Oct-2011 00:51:51] <nyeates> There are many examples in our forums and wiki docs
[27-Oct-2011 01:05:50] <Tabmow> We are using zenoss enterprise
[27-Oct-2011 01:21:52] <tvincent> nyeates this is all zenoss 4?
[27-Oct-2011 01:46:37] <zaoul> Hi. I had graphs on existing nodes, then I discovered nodes by network and my existing graphs are gone. Just a broken image icon?
[27-Oct-2011 02:37:46] <zaoul> broken graphs? whole point is graphs
[27-Oct-2011 02:49:38] <zaoul> ERROR ZServer uncaptured python exception, closing channel <ZServer.HTTPServer.zhttp_channel HTTP headers invalid (too long) (got: 8215 bytes, allowed 8192 bytes ... google tells me this is an ancient problem?
[27-Oct-2011 03:11:52] <zaoul> zope http-header-max-length 16384 ?
[27-Oct-2011 04:48:49] <enkrypt> hi all! I've got a strange issue where I am getting ping failure events being sent for an asset, but they are not sent by the remote collector that is responsible for monitoring said asset.
[27-Oct-2011 05:00:31] <enkrypt> also: I've got enterprise and doing vmware esx infrastructure monitoring, I would like to DELETE my entire infrastructure, how do I do that?
[27-Oct-2011 05:02:04] <enkrypt> disregard that last question, deleting it from the UI works fine, but slow
[27-Oct-2011 05:11:20] <enkrypt> but my original Q still stands: I get alerts coming from a distributed collector, that is NOT the designated collector for a specific host
[27-Oct-2011 05:45:26] <DW-10297> Is there a way to graph the result of this command in Zenoss: cat /proc/sys/net/ipv4/netfilter/ip_conntrack_count
[27-Oct-2011 08:02:37] <Hackman238> Morning all!
[27-Oct-2011 08:04:54] <enkrypt> hiya Hackman238
[27-Oct-2011 08:05:05] <Hackman238> enkrypt: How goes the war?
[27-Oct-2011 08:06:38] <enkrypt> Hackman238: one battle at a time!
[27-Oct-2011 08:06:58] <Hackman238> enkrypt: That's the way to do it.
[27-Oct-2011 08:07:20] <enkrypt> I had been having big zenhub performance issues, but those seem to be conquered. it was probably due to wrong ownership of my mysql directory
[27-Oct-2011 08:07:30] <Hackman238> enkrypt: Oh wow.
[27-Oct-2011 08:07:43] <enkrypt> but the thing is, I'm not 100% certain that was the thing that caused it
[27-Oct-2011 08:08:08] <enkrypt> you see my /var/lib/mysql is a symbolic link to /opt/zenoss/var/mysql
[27-Oct-2011 08:08:15] <enkrypt> and at one time I needed to fix some zenoss:zenoss ownership
[27-Oct-2011 08:08:24] <enkrypt> so stupidly I did chown -R zenoss:zenoss /opt/zenoss
[27-Oct-2011 08:08:33] <Hackman238> enkrypt: I do know that if the zencleanuphistory runs it pretty much locks mysqld up until the operation is finished. This causes zenhub to wait in line forever.
[27-Oct-2011 08:08:37] <enkrypt> which is all OK, except for mysql, which needs mysql:mysql of course
[27-Oct-2011 08:08:48] <Hackman238> enkrypt: LOL It happens
[27-Oct-2011 08:08:59] <enkrypt> one strange issue that I'm having now still, is something new
[27-Oct-2011 08:09:08] <enkrypt> because you see, my topology never used to work until yesterday
[27-Oct-2011 08:09:22] <enkrypt> because my zenoss server names didn't resolve correctly, small error in /etc/hosts
[27-Oct-2011 08:09:35] <enkrypt> so all good, I fixed that, now my topology actually gets built
[27-Oct-2011 08:10:11] <enkrypt> but what happens now: I have 2 hosts, being monitored by remote collector X, and they are pingable by collector X. but I receive events for those 2 hosts, saying they are not pingable, and the events are coming from collector Y
[27-Oct-2011 08:10:13] <Hackman238> enkrypt: Nice
[27-Oct-2011 08:10:23] <enkrypt> and they are INDEED not pingable from collector Y, but that is perfectly normal
[27-Oct-2011 08:10:33] <enkrypt> but collector Y is not the designated collector
[27-Oct-2011 08:10:56] <enkrypt> I tried moving said 2 hosts to collector "Z" so to speak, and doing a collector update afterwards and everything, but doesn't help
[27-Oct-2011 08:11:04] <Hackman238> enkrypt: Is zenping running on collector x?
[27-Oct-2011 08:11:06] <enkrypt> collector Y keeps telling me those 2 hosts are down
[27-Oct-2011 08:11:12] <enkrypt> zenping is running on collector x, y and z
[27-Oct-2011 08:11:41] <Hackman238> enkrypt: Y shouldn't be trying at all.
[27-Oct-2011 08:11:43] <Hackman238> Hum.
[27-Oct-2011 08:11:44] <enkrypt> when I open up the hosts in the UI they are clearly stated as monitored by collector X. but still they are pinged by collector Y
[27-Oct-2011 08:11:55] <enkrypt> this started happening when I fixed the topology feature
[27-Oct-2011 08:12:10] <enkrypt> you see, collector Y is like a customer specific collector, to overcome network overlaps etc
[27-Oct-2011 08:12:26] <enkrypt> so i'm not sure why this happens
[27-Oct-2011 08:12:50] <Hackman238> enkrypt: Odd. Did you run it by rocket?
[27-Oct-2011 08:12:51] <enkrypt> I've tried reconfiguring on different collectors, but doesn't help. maybe I can just remove the asset completely, and re add it cleanly
[27-Oct-2011 08:13:15] <enkrypt> Hackman238: the past few days I've been doing webex debug sessions with Zack / psyklone
[27-Oct-2011 08:13:33] <enkrypt> haven't ran it by rocket
[27-Oct-2011 08:16:26] <Hackman238> enkrypt: Ah gotcha
[27-Oct-2011 08:18:25] <hmp> enkrypt: stupid question, but did you try restarting the zenping daemon on collector Y?
[27-Oct-2011 08:19:21] <enkrypt> hmp: restarted literally everything, everywhere
[27-Oct-2011 08:20:25] <enkrypt> for now I'll try simply deleting the assets, and adding them anew
[27-Oct-2011 08:36:31] <enkrypt> Hackman238 & hmp: ok I deleted the devices, and added them again. but after a few minutes, again a totally irrelevant collector starts sending me events saying the 2 machines are down
[27-Oct-2011 08:36:48] <enkrypt> what is strange, when I check zenping.log on the CORRECT collector, it states clearly that the pings are succesful
[27-Oct-2011 08:37:09] <enkrypt> as it should be
[27-Oct-2011 08:52:29] kocolosk is now known as kocolosk|away
[27-Oct-2011 09:35:01] <JohnnyNOC> i have some thresholds for a custom performance template i've disabled but the alerts keep hitting the dashboard
[27-Oct-2011 09:35:13] <JohnnyNOC> what daemon(s) do i need to bounce so Zenoss figures out that you know, i disabled the thresholds?
[27-Oct-2011 09:36:15] <dhopp> JohnnyNOC: zenhub and zenactions I *think*
[27-Oct-2011 09:40:07] <JohnnyNOC> i doubt it
[27-Oct-2011 09:40:18] <JohnnyNOC> zenactions for sure i doubt it, as far as i know that's only for responsible efro emailing the alerts
[27-Oct-2011 09:40:21] <JohnnyNOC> i'm still seeing them on the dashboard
[27-Oct-2011 09:40:24] <JohnnyNOC> and that makes me angry!
[27-Oct-2011 09:40:26] * JohnnyNOC raises fists
[27-Oct-2011 09:44:52] <JohnnyNOC> dhopp: they're still coming
[27-Oct-2011 09:47:43] <dhopp> JohnnyNOC: take a hammer to it: zenoss restart :-)
[27-Oct-2011 09:50:25] <JohnnyNOC> dhopp you know, i'm pretty sure i did that yesterday
[27-Oct-2011 09:50:28] <JohnnyNOC> and unfortunately i'm still seeing this problem today
[27-Oct-2011 09:50:29] <JohnnyNOC> :/
[27-Oct-2011 09:51:19] <JohnnyNOC> and it's super frustrating because there are probably 8-10 dif thresholds that were setup, they were all disabled, yet i keep getting alerted on 2 of them
[27-Oct-2011 09:51:20] <JohnnyNOC> ;/
[27-Oct-2011 09:51:37] <JohnnyNOC> i just wish i knew why why why
[27-Oct-2011 09:52:10] <dpetzel> JohnnyNOC: Any chance you could SNMPWALK | grep 1600 one of your BigIPs? I don't have access to any older devices, and I'd really like to know what additional OID I can check to avoid the 'deprecated' model deal
[27-Oct-2011 09:53:29] <JohnnyNOC> dpetzel: sure
[27-Oct-2011 09:53:45] <dpetzel> Cool thanks, its really annoying me that it reports deprecated
[27-Oct-2011 09:55:18] <JohnnyNOC> dpetzel: i don't see anything interesting
[27-Oct-2011 09:55:25] <dpetzel> JohnnyNOC:
[27-Oct-2011 09:59:39] <dpetzel> JohnnyNOC: so based on your SNMP data you're unable to determine the model? Might explain the deprecated value I suppose
[27-Oct-2011 10:01:59] <JohnnyNOC> dpetzel: i wouldn't say that, but i didn't find '1600' in the output apparently
[27-Oct-2011 10:02:02] <JohnnyNOC> let me take a closer look
[27-Oct-2011 10:02:35] <dpetzel> JohnnyNOC: Cool, thanks for taking the time to look.
[27-Oct-2011 10:04:22] <JohnnyNOC> OK
[27-Oct-2011 10:04:22] <JohnnyNOC> so
[27-Oct-2011 10:04:24] <JohnnyNOC> this is what you want
[27-Oct-2011 10:04:26] <JohnnyNOC> bigip1600 1.3.6.1.4.1.3375.2.1.3.4.20
[27-Oct-2011 10:04:31] <JohnnyNOC> SNMPv2-MIB::sysObjectID.0 = OID: SNMPv2-SMI::enterprises.3375.2.1.3.4.20
[27-Oct-2011 10:04:45] <JohnnyNOC> so no, i do not explicitly see the bigip1600 but the OID at that value is what it's supposed to mean
[27-Oct-2011 10:04:56] <JohnnyNOC> according to F50-BIGIP-SYSTEM-MIB.txt
[27-Oct-2011 10:04:59] <JohnnyNOC> er
[27-Oct-2011 10:05:00] <JohnnyNOC> F5
[27-Oct-2011 10:09:19] <dpetzel> Cool thanks
[27-Oct-2011 10:11:51] <dpetzel> Actually I'm not sure thats what I'm after. That OID will be bigip1600 on ALL systems. Its simply a LIST of model numbers. Its not really an OID that says "I AM a 1600"
[27-Oct-2011 10:12:12] <dpetzel> 4.1 - 4.23 list the various models
[27-Oct-2011 10:22:53] <JohnnyNOC> ok
[27-Oct-2011 10:23:03] <JohnnyNOC> that was the only OID that I believed indicated what ht emodel type was
[27-Oct-2011 10:25:11] <JohnnyNOC> 1600
[27-Oct-2011 10:25:11] <JohnnyNOC> so in my opinion, it seems the zenpack should be checking the device for sysObjectID and when it's given that OID, to say it's a bigip260
[27-Oct-2011 10:28:39] <bseklecki> which MIBs does the InterfaceMap plugin iterate?
[27-Oct-2011 10:28:41] <JohnnyNOC> pretty sure that OID is saying "I AM a 1600"
[27-Oct-2011 10:28:44] <JohnnyNOC>
[27-Oct-2011 10:29:37] <JohnnyNOC> bseklecki: if i had to guess i'd say IF-MIB.txt
[27-Oct-2011 10:30:26] <JohnnyNOC> and judging by this comment in the code i think i'm correct
[27-Oct-2011 10:30:29] <JohnnyNOC> # handle 10GB interfaces using IF-MIB::ifHighSpeed
[27-Oct-2011 10:30:41] <bseklecki> checking IpInterface.py now
[27-Oct-2011 10:30:44] <JohnnyNOC> dpetzel: am i making sense?
[27-Oct-2011 10:31:01] <dpetzel> sorry missed your last commends
[27-Oct-2011 10:31:04] <zaoul> Does zenoss delete old rrd data?
[27-Oct-2011 10:31:05] <dpetzel> comments
[27-Oct-2011 10:31:14] <zaoul> ..automatically
[27-Oct-2011 10:31:34] <JohnnyNOC> dpetzel: you are really going to check the f5 for sysObjectID
[27-Oct-2011 10:31:39] <JohnnyNOC> and you'll be returned that OID i pasted earlier
[27-Oct-2011 10:31:40] <dpetzel> JohnnyNOC: Your making sense, but I'm not sure that I agree, what do you get if your SNMPGET .1.3.6.1.4.1.3375.2.1.3.4.21 instead
[27-Oct-2011 10:31:42] <zaoul> Delete old data if device does not exist. Specifically going for virtual interfaces.
[27-Oct-2011 10:31:48] <JohnnyNOC> and that OID says: "I am a 1600"
[27-Oct-2011 10:32:17] <JohnnyNOC> SNMPv2-SMI::enterprises.3375.2.1.3.4.21 = No Such Object available on this agent at this OID
[27-Oct-2011 10:32:18] <dpetzel> JohnnyNOC: ahh. I think I follow you now
[27-Oct-2011 10:32:40] <JohnnyNOC> dpetzel: cool
[27-Oct-2011 10:34:07] <dpetzel> JohnnyNOC: Yeah I'm totally with you know, I miss read your first response. I should be able to get this updated to work properly
[27-Oct-2011 10:34:40] <bseklecki> IF-MIB::ifTable: No entries
[27-Oct-2011 10:34:47] <bseklecki> when I'm god, IBM dies.
[27-Oct-2011 10:36:05] <dgarcia1> zaoul: deleting rrd/events (related to a device) is optional when removing a device
[27-Oct-2011 10:36:57] <zaoul> I'm scanning a kvm hypervisor and it aggregates the network interfaces, which is nice.. but I assume after some time they will be gone and it will not matter anymore.
[27-Oct-2011 10:37:26] <zaoul> I read on the forms to use find and mtime, but im trying to get away from cruft scripts.
[27-Oct-2011 10:58:11] <nyeates> Hi
[27-Oct-2011 10:58:35] <nyeates> dev chat in here in a few
[27-Oct-2011 10:59:22] <jmp242> Hi all
[27-Oct-2011 10:59:48] <nyeates> Hope you like the jive access James :-)
[27-Oct-2011 11:00:17] <jmp242> I haven't had much time, but yes thanks
[27-Oct-2011 11:00:37] <jmp242> Anyone have luck with 3.2.1 core auth to AD?
[27-Oct-2011 11:00:46] <jmp242> I might be trying that in the future
[27-Oct-2011 11:00:53] <simonjj> Hello everyone, hope all is well
[27-Oct-2011 11:01:45] <simonjj> anyone out there test-driven 3.2.1 at this time ?
[27-Oct-2011 11:02:03] <jmp242> Yea
[27-Oct-2011 11:02:10] <jmp242> well, just did the update on my production server
[27-Oct-2011 11:02:15] <jmp242> all is well after clearing browser cache
[27-Oct-2011 11:02:49] <simonjj> feedback ?
[27-Oct-2011 11:03:01] <nyeates> jmp242: noticed if any graphs have issues changing their range?
[27-Oct-2011 11:03:13] <jmp242> mmm
[27-Oct-2011 11:03:20] <jmp242> haven't done much with it since yesterday
[27-Oct-2011 11:03:27] <jmp242> do you mean zooming in and out on graphs?
[27-Oct-2011 11:04:33] <nyeates> I think its more the hourly, daily, monthly button
[27-Oct-2011 11:04:44] <simonjj> overall it evolutionary, but I'd like to think that most things we did are for the better
[27-Oct-2011 11:05:49] <nyeates> revolutionary will hopefully come next major release ;-)
[27-Oct-2011 11:06:00] <jmp242> mmm
[27-Oct-2011 11:06:10] <jmp242> I didn'g get it to do anything
[27-Oct-2011 11:06:27] <jmp242> I tried on a test device, changed to weekly, hit reset, hit refresh, nada
[27-Oct-2011 11:06:32] <jmp242> in Firefox 3.6.23
[27-Oct-2011 11:06:36] <nyeates> nada as in....
[27-Oct-2011 11:06:47] <jmp242> graphs don't change
[27-Oct-2011 11:06:49] <jmp242> at all
[27-Oct-2011 11:07:43] <jmp242> same in Opera 11.52
[27-Oct-2011 11:08:27] <dgarcia1> jmp242: is this a firefox only problem? or can you reproduce on another browser?
[27-Oct-2011 11:08:55] <jmp242> same in Opera 11/52
[27-Oct-2011 11:09:09] <dgarcia1> did you get any javascript errors?
[27-Oct-2011 11:09:16] <jmp242> mmm
[27-Oct-2011 11:09:20] <jmp242> let me check error console
[27-Oct-2011 11:10:02] <jmp242> not in Opera
[27-Oct-2011 11:10:27] <jmp242> well, at least not when using the resize
[27-Oct-2011 11:10:33] <jmp242> let me try a reload of the whole page
[27-Oct-2011 11:10:42] <dgarcia1> try firebug/firefox
[27-Oct-2011 11:11:07] <jmp242> A bunch of CSS errors
[27-Oct-2011 11:11:11] <jmp242> but no javascript ones
[27-Oct-2011 11:11:22] <jmp242> I don't have firebug in Firefox, but maybe someone else can check
[27-Oct-2011 11:11:47] <nyeates> yeah - ok - its been verified in a couple places....someone said maybe its an AJAX thing....its been marked as a priority 1 bug
[27-Oct-2011 11:12:35] <jmp242> Firefox shows the same sorts of CSS errors
[27-Oct-2011 11:12:43] <jmp242> but no javascript errors in it's error console
[27-Oct-2011 11:14:09] <jmp242> oh hey
[27-Oct-2011 11:14:13] <jmp242> if I go off of graphs
[27-Oct-2011 11:14:15] <jmp242> say to events
[27-Oct-2011 11:14:17] <jmp242> and then back
[27-Oct-2011 11:14:21] <jmp242> the ranges change then in FF
[27-Oct-2011 11:14:37] <emillerzenoss> interesting
[27-Oct-2011 11:14:41] <nyeates> weird
[27-Oct-2011 11:14:58] <jmp242> same in Opera
[27-Oct-2011 11:15:08] <jmp242> this is in the left hand nav
[27-Oct-2011 11:15:11] <jmp242> for the device
[27-Oct-2011 11:15:16] <jmp242> not to the main event console
[27-Oct-2011 11:17:39] <nyeates> Hey, while we do have some devs in here, I want to point you guys at Jane Curry's latest predicament. She has posted non-working code here: https://github.com/jcurry/ZenPacks.skills1st.MenuExamples
[27-Oct-2011 11:17:54] <nyeates> In browser.configure.zcml, the js-myFooterMenu stanza completely naffs everything. No panel that has a footer renders at all.
[27-Oct-2011 11:18:40] <nyeates> 2) The run_my_predefined_command menu does appear but doesn't produce output from the command. I have debugging in libexec/mywrapper_script1 that shows that control is indeed passed there, but my output window is blank. I originally got this code snippet from Kells.
[27-Oct-2011 11:19:20] <JohnnyNOC> wondering if someone could give me some advice on how to properly add ssCpuRawNice
[27-Oct-2011 11:19:58] <JohnnyNOC> i can't just add the single OID as it's a sum of all the CPUs.. but i'm not sure how to capture the number of cpus and do the math
[27-Oct-2011 11:22:08] <dgarcia1> JohnnyNOC: do you think docs/DOC-10224 may solve your problem?
[27-Oct-2011 11:23:19] <JohnnyNOC> dgarcia1: i believe so
[27-Oct-2011 11:23:23] <nyeates> dgarcia1: I was going to offer the same advise. Rmatte zenpack
[27-Oct-2011 11:23:52] <nyeates> it works well - just note that it doesnt scale to many many devices well yet
[27-Oct-2011 11:24:18] <emillerzenoss> i could be completely off here, but wouldn't the number of cpus be in the model and not in an rrd? and would that require a custom modeler plugin? if so docs/DOC-5859
[27-Oct-2011 11:26:02] <JohnnyNOC> emillerzenoss: yes, but i presume there's an OID available i can query to get the num cpus
[27-Oct-2011 11:26:15] <nyeates> "You can also call on values such as here.hw.totalMemory."
[27-Oct-2011 11:26:23] <emillerzenoss> ahh
[27-Oct-2011 11:32:04] <JohnnyNOC> hmm
[27-Oct-2011 11:32:07] <JohnnyNOC> of course it can't be "easy"
[27-Oct-2011 11:32:09] <JohnnyNOC>
[27-Oct-2011 11:34:37] <JohnnyNOC> i>>> len(d.hw.cpus()) 2
[27-Oct-2011 11:35:48] <aclark> Anyone used docs/DOC-3501 (mysql monitor) on 3.2.1? Just installed it but Zenoss says it's broken even after restart
[27-Oct-2011 11:38:50] <nyeates> aclark: what happens or what does zenoss say?
[27-Oct-2011 11:39:37] <Jane_Curry> Thanks Nick - anyone any ideas to contribute to my MenuExamples - https://github.com/jcurry/ZenPacks.skills1st.MenuExamples ?????????????
[27-Oct-2011 11:39:39] <nyeates> also, did you install it as part of the zenoss zenpack bundle, or seperately? I think we QA the install of the bundle
[27-Oct-2011 11:41:18] <jmp242> So I'm updating the Forum FAQs and integrating them to one Wiki page on zcaportal.org so I'm no longer across so many pages for users
[27-Oct-2011 11:41:26] <jmp242> What browsers is Zenoss now supporting?
[27-Oct-2011 11:44:03] <jmp242> nyeates: ? I'm assuming IE8-9? Firefox (some version)...
[27-Oct-2011 11:45:11] <nyeates> ff, ie, maybe chrome....im looking
[27-Oct-2011 11:46:24] <nyeates> Firefox (verified with 3.6.x, 4, 5) or Internet Explorer (verified with 7, 8, 9)
[27-Oct-2011 11:46:42] <nyeates> we also run it internally on Chrome, but no official support
[27-Oct-2011 11:50:11] <dgarcia1> @jane i can't get it to install cleanly, working on it
[27-Oct-2011 11:51:16] <spawnwho> Hi All I would like to ask regarding enterprise version of zenoss. What is the main benefits of going for Enterprise version compared to Zenoss core?
[27-Oct-2011 11:51:32] <aclark> nyeates: in zenpacks i see, This pack was just installed, or is missing or broken. Please restart Zope. (even though i restarted)
[27-Oct-2011 11:52:42] <nyeates> spawnwho: I will take that question with you in private. We are in a dev chat session now.
[27-Oct-2011 11:52:51] <aclark> nyeates: might be a perms issue actually, installed as root. i'm chowning to zenoss…
[27-Oct-2011 11:53:23] <nyeates> su - zenoss
[27-Oct-2011 11:53:34] <nyeates> then install
[27-Oct-2011 11:53:49] <nyeates> - brings environment vars needed
[27-Oct-2011 11:53:54] <spawnwho> @nyeates sure. Thanks
[27-Oct-2011 11:54:02] <JohnnyNOC> could anyone give suggestions as to what I should look into when I have a datasource with a threshold (that has been disabled) which keeps sending alerts to the dashboard?
[27-Oct-2011 11:54:38] <JohnnyNOC> i have a monitoring template for some mongo servers of ours which had thresholds enabled that i've disabled (in the lil' checkbox thing) but the alerts keep hitting the dashboard and i'd like to try and understand why
[27-Oct-2011 11:54:57] <aclark> nyeates: yep, perms! sorry
[27-Oct-2011 11:55:20] <nyeates> aclark: glad we stomped that to be a no-issue :-)
[27-Oct-2011 11:55:54] <aclark> nyeates: yeah! :-) thanks for the reply
[27-Oct-2011 11:58:00] <kenhen93> is there any kind of bug that would if a data source or monitoring template contains a space, would prevent it from being added to a zenpack properly?
[27-Oct-2011 12:00:01] [disconnected at Thu Oct 27 12:00:01 2011]
[27-Oct-2011 12:00:02] [connected at Thu Oct 27 12:00:02 2011]
[27-Oct-2011 12:00:13] [zenoss-logger (logger bot) has joined #zenoss]
[27-Oct-2011 12:01:48] <JohnnyNOC> OK how about another question
[27-Oct-2011 12:02:11] <emillerzenoss> Jane, I looked at another implementation of stream() in a StreamingView subclass, and rather than returning output, it called self.write() in a loop for each bit of output expected
[27-Oct-2011 12:02:29] <JohnnyNOC> I have filesystem thresholds set at 95% for disk usage, however if i restart zenoss I get alerts as low as 1.9% utilization for all my devices.... they eventually catch up and show the right usage but why am i getting alerts about utilization way under my set threshold?
[27-Oct-2011 12:04:16] <Jane_Curry> emillerzenoss: Can you point me at that sample - is it in Zenoss Core?
[27-Oct-2011 12:05:00] <emillerzenoss> http://dev.zenoss.com/trac/browser/trunk/Products/ZenUI3/browser/streaming.py
[27-Oct-2011 12:05:12] <emillerzenoss> class TestStream at the end
[27-Oct-2011 12:08:58] <nyeates> emillerzenoss: what does that stream code do or relate to?
[27-Oct-2011 12:09:29] <emillerzenoss> the run_my_predefined_command not showing output
[27-Oct-2011 12:15:02] <nyeates> ok gotcha....I am curious as to what that streaming.py, in particular, is doing....it spits back html upon request of something else? something to do with ajax or other data exchange?
[27-Oct-2011 12:16:27] <emillerzenoss> it writes to the response. remodel a device and see the modeler output stream, for example.
[27-Oct-2011 12:17:53] <nyeates> ahhhh gotcha, that translucent popup that streams in new data (maybe from a log or command) as stuff happens
[27-Oct-2011 12:17:57] <JohnnyNOC> nyeates: i'm using 3.1.0 and have the DeviceSearch zenpack installed. will i have to remove it before i upgrade to 3.2.1 ?
[27-Oct-2011 12:18:18] <JohnnyNOC> also, i just noticed that i don't see that zenpack listed in my list of zenpacks.. but i do have the global device search bar in the upper right hand corner. do you know if that's to be expected?
[27-Oct-2011 12:20:09] <JohnnyNOC> hmm.. it's listed in my list of zenpacks on my dev instance of zenoss
[27-Oct-2011 12:20:16] <JohnnyNOC> so i would say probably not expected
[27-Oct-2011 12:22:30] <JohnnyNOC> any ideas if/how that may affect a Zenoss upgrade? should i remove the zenpack manually first?
[27-Oct-2011 12:23:22] <nyeates> looking
[27-Oct-2011 12:23:28] <JohnnyNOC> i haven't had good experiences trying to upgrade zenoss and i'm afraid something small like this is going to totally bomb
[27-Oct-2011 12:23:33] <JohnnyNOC> the upgrade
[27-Oct-2011 12:23:33] <JohnnyNOC>
[27-Oct-2011 12:24:17] <nyeates> talking about this ZP docs/DOC-7453
[27-Oct-2011 12:24:18] <nyeates> ?
[27-Oct-2011 12:24:29] <JohnnyNOC> yes
[27-Oct-2011 12:24:42] <JohnnyNOC> i have it installed on two separate 3.1.0 installations and my prod install doesn't list it in the list of zenpacks
[27-Oct-2011 12:24:46] <JohnnyNOC> but the device search bar is still there
[27-Oct-2011 12:25:03] <JohnnyNOC> and i'm just afraid that's one thing that could possibly break a future upgrade so i thought it wouldn't hurt to ask
[27-Oct-2011 12:25:29] <nyeates> ahhh gotcha
[27-Oct-2011 12:26:08] <nyeates> yeah that is a discrepancy, isnt it - I too would want to make sure that it is actually gone
[27-Oct-2011 12:26:25] <JohnnyNOC> and then i have problems with some thresholds that don't seem to be working correctly such as the disk usage i mentioned earlier, or this custom performance template who's thresholds have been dsiabled but alerts are still hitting my dashboard
[27-Oct-2011 12:26:40] <nyeates> ummmm, seems like you can use some scrpt or something to rid of sticky zenpacks
[27-Oct-2011 12:26:57] <JohnnyNOC> both of which i'd like to try and understand so i could fix and successfully upgrade
[27-Oct-2011 12:27:05] <nyeates> yeah i saw, i didnt know where to go with those
[27-Oct-2011 12:27:25] <nyeates> Thanks all for Dev chat - have a good one!
[27-Oct-2011 12:27:48] <nyeates> Oh BTW, one announce...
[27-Oct-2011 12:28:13] <nyeates> We are changing the way that ZenPacks are submitted and offered
[27-Oct-2011 12:28:34] <nyeates> Look for an upcoming announcement in the blog and forums
[27-Oct-2011 12:29:17] <JohnnyNOC> nyeates: is there an existing script? or should i just start poking around in zendmd
[27-Oct-2011 12:29:36] <nyeates> JohnnyNOC: yeah let me see if i can determine what script might do that
[27-Oct-2011 12:33:20] <nyeates> zenfixit.py does it
[27-Oct-2011 12:33:33] <nyeates> it has a routine called def removeBrokenZenPacks(self):
[27-Oct-2011 12:34:01] <nyeates> also def checkZenPacksDirectory(self):
[27-Oct-2011 12:34:08] <nyeates> """
[27-Oct-2011 12:34:08] <nyeates> 107 Sometimes the zenpack install or remove doesn't get a chance to complete, so the
[27-Oct-2011 12:34:09] <nyeates> 108 Python include list can get messed up.
[27-Oct-2011 12:34:09] <nyeates> 109 """
[27-Oct-2011 12:34:18] <JohnnyNOC> hm
[27-Oct-2011 12:34:26] <JohnnyNOC> i'm running it now
[27-Oct-2011 12:34:33] <JohnnyNOC> and have run it in the recent past
[27-Oct-2011 12:34:43] <JohnnyNOC> and the zenpack is not in the list, but like i said, the search bar is still there
[27-Oct-2011 12:35:03] <JohnnyNOC> i guess i'll see what happens this time
[27-Oct-2011 12:35:34] <nyeates> is the base of the code same as mine?...
[27-Oct-2011 12:35:59] <JohnnyNOC> apparently not
[27-Oct-2011 12:36:01] <nyeates> 62 for pack in packs():
[27-Oct-2011 12:36:01] <nyeates> 63 try:
[27-Oct-2011 12:36:01] <nyeates> 64 unused = pack.primaryAq()
[27-Oct-2011 12:36:01] <nyeates> 65 print "%s is fine." % (pack.id,)
[27-Oct-2011 12:36:01] <nyeates> 66 except AttributeError:
[27-Oct-2011 12:36:01] <nyeates> 67 print "Problem with %s ZenPack. Forcing removal." % (pack.id,)
[27-Oct-2011 12:36:03] <nyeates> 68 packs._remove(pack)
[27-Oct-2011 12:36:05] <JohnnyNOC> i don't seem to have a removeBrokenenPAcks thingy
[27-Oct-2011 12:36:22] <nyeates> that is in remove broken packs
[27-Oct-2011 12:36:32] <JohnnyNOC> is that not part of zenfixit.py ?
[27-Oct-2011 12:36:45] <nyeates> let me just paste pin you that routine and you can add a line for it in the def run
[27-Oct-2011 12:36:48] <nyeates> make sense?
[27-Oct-2011 12:37:03] <JohnnyNOC> yes
[27-Oct-2011 12:37:13] <JohnnyNOC> but should i be looking for a newer/updated version of zenfixit?
[27-Oct-2011 12:37:23] <JohnnyNOC> or do you have some super secret special squirrel internal version
[27-Oct-2011 12:39:16] <nyeates> yeah im not sure either....could be the later
[27-Oct-2011 12:43:41] <JohnnyNOC> i'm afraid to try it now
[27-Oct-2011 12:43:48] <JohnnyNOC> i'll give it a go after the market closes and i've down a complete backup
[27-Oct-2011 12:46:29] <nyeates> yeah im still looing for newest zenfixit.py
[27-Oct-2011 12:46:34] <nyeates> that is public
[27-Oct-2011 12:46:40] <nyeates> if not that, ill giv eu the pastbin
[27-Oct-2011 12:47:28] <JohnnyNOC> cool, i was just going to add the snippet above
[27-Oct-2011 12:49:40] <JohnnyNOC> i think i've mentioned this before, but i (and i'm sure others would as well!) would LOOOOOVE something to somehow 'audit' my setup to tell me what's right/wrong
[27-Oct-2011 12:49:49] <nyeates> pastebin it is
[27-Oct-2011 12:49:52] <nyeates> http://pastebin.com/s85V1wsk
[27-Oct-2011 12:49:53] <JohnnyNOC> ZCA are you listening?!
[27-Oct-2011 12:49:54] <JohnnyNOC>
[27-Oct-2011 12:51:22] <nyeates> heh good luck dev'ing that....even the best 20 python zenoss devs in the world have issues with controlling stray object model relations
[27-Oct-2011 12:51:28] <nyeates> each case is a new thing and one off
[27-Oct-2011 12:52:59] <jmp242> JohnnyNOC: let me know if that all works
[27-Oct-2011 12:53:12] <nyeates> from what i understand, which is not direct knowledge, is that there are tons of relationships amongst all the objects in the object oriented database. When you got to move or remove or alter these relations - maybe removing a zenpack - unexplained stuff can go wrong and cause some of the relations to stay and even for the object(s) to stay
[27-Oct-2011 12:53:13] <JohnnyNOC> jmp242: the zenfixit thing?
[27-Oct-2011 12:53:29] <JohnnyNOC> nyeates: ok. well, that sucks.
[27-Oct-2011 12:54:24] <jmp242> yea
[27-Oct-2011 12:54:31] <jmp242> If it's useful I do want to document it
[27-Oct-2011 12:54:43] <JohnnyNOC> jmp242: will do
[27-Oct-2011 12:55:04] <nyeates> later guys
[27-Oct-2011 12:55:51] <nyeates> James, talk to you tomorrow.....ill be reviewing the changes to Community ZenPacks and demo how to post one to the new many-repo's github schema
[27-Oct-2011 12:55:54] <JohnnyNOC> thanks for the help nyeates
[27-Oct-2011 13:01:05] <jmp242> ok
[27-Oct-2011 13:02:23] * aclark <3s zenbatchload (and zenpatch for that matter, too)
[27-Oct-2011 13:33:09] <jb> hrm, I have apachemonitor bound to a device, and verified that it is collecting data (via zencommand), but none of the graphs are updating and continue to say nan.
[27-Oct-2011 13:35:36] <jb> anyone see a problem with this? http://pastebin.com/aFyxuyEQ
[27-Oct-2011 13:40:17] <kenhen93> have you checked your command interval on the apache command?
[27-Oct-2011 13:41:08] <jb> command interval?
[27-Oct-2011 13:41:15] <jb> you mean zCommandCycleTime?
[27-Oct-2011 13:43:04] <kenhen93> no, on the actual apache status command
[27-Oct-2011 13:43:27] <kenhen93> rrd won't graph anything if it does not get data over a consistent time
[27-Oct-2011 13:44:36] <kenhen93> under your data source apache
[27-Oct-2011 13:44:56] <kenhen93> there is a cycle time
[27-Oct-2011 13:45:35] <jb> hrm, lemme check
[27-Oct-2011 13:45:40] <kenhen93> i just checked an apache template I have and the cycle time is set to 30 seconds
[27-Oct-2011 13:46:11] <jb> hrm its set to 300
[27-Oct-2011 13:46:47] <jb> its set to 300 on my old 3.1 instance which is working fine, though.
[27-Oct-2011 13:47:13] <jb> it's not writing the RRD's for some reason.
[27-Oct-2011 13:47:42] <kenhen93> the .rrd files are not showing up at all?
[27-Oct-2011 13:47:55] <jb> it creates one, the apacheSlotDNSLookup.
[27-Oct-2011 13:47:57] <jb> and then quits
[27-Oct-2011 13:48:07] <jb> 2011-10-27 12:47:31,628 DEBUG zen.RRDUtil: Creating new RRD file /opt/zenoss/perf/Devices/fheg-eportal01.fheg.follett.com/apache_slotDNSLookup.rrd
[27-Oct-2011 13:48:50] <jb> 2011-10-27 12:47:31,624 DEBUG zen.zencommand: Datasource: apache Received exit code: 0 Output:
[27-Oct-2011 13:48:52] <jb> 'STATUS OK|slotDNSLookup=0 totalAccesses=163947573 slotReadingRequest=0 totalKBytes=345053153
[27-Oct-2011 13:48:57] <jb> but it's obviously getting the data
[27-Oct-2011 13:50:58] <jb> yeah, its not working on any devices. hmm.
[27-Oct-2011 13:53:40] <Ve2dmn1> jb: how did you get that RRDUtils Debug output? I have a similar problem with a device having 4 days of NaN graphs....
[27-Oct-2011 13:54:07] <kenhen93> zencommand debug
[27-Oct-2011 13:54:14] <jb> message/61046
[27-Oct-2011 13:54:15] <kenhen93> run that as the zenoss user
[27-Oct-2011 13:54:18] <jb> sec
[27-Oct-2011 13:54:40] <jb> ve2dmn: run
[27-Oct-2011 13:54:47] <jb> "zencommand run -v10 -d devicename"
[27-Oct-2011 13:55:13] <jb> i think this may be a bug.
[27-Oct-2011 13:58:41] <Ve2dmn1> I get the same output for any "devicename"... which is almost nothing
[27-Oct-2011 13:58:42] <jb> I opened a support ticket.
[27-Oct-2011 13:58:46] <kenhen93> try changing the interval
[27-Oct-2011 13:58:50] <kenhen93> this helped me
[27-Oct-2011 13:58:57] <kenhen93> i had to delete the graphs
[27-Oct-2011 13:59:06] <Ve2dmn1> I only have 1 switch out of 4 with empty graphs
[27-Oct-2011 14:00:31] <Ve2dmn1> that device has missing files in /perf/Device/devicename
[27-Oct-2011 14:02:08] <Ve2dmn1> funny thing is: it used to work before...
[27-Oct-2011 14:02:55] <Ve2dmn1> I moved the device from one loaction to another (without changing the IP) and decided to simple delete the device and re-create it under a new name...
[27-Oct-2011 14:03:12] <Ve2dmn1> Empty graphs since...
[27-Oct-2011 14:03:43] <Ve2dmn1> spent all my free "work" hours in the last 4 days on this problem...
[27-Oct-2011 14:04:03] <jb> hm, I dind't realize enterprise 4.1.0 was out.
[27-Oct-2011 14:04:06] <jb> maybe I should upgrade.
[27-Oct-2011 14:04:08] <jb> im @ 4.0.2.
[27-Oct-2011 14:11:51] <jb> hrmf.. lack of communication is really starting to get frustrating.
[27-Oct-2011 14:12:20] <jb> any zen guys here?
[27-Oct-2011 14:12:48] <rocket> jb whats up?
[27-Oct-2011 14:14:25] <jb> well, I'm trying to determine if I need to go from 4.0.2 -> 4.1.0, but the release notes are basically zilch from 4.1.0.
[27-Oct-2011 14:14:29] <jb> so, I don't know what changed.
[27-Oct-2011 14:14:36] <jb> (I didn't even know 4.1.0 was out)
[27-Oct-2011 14:15:06] <jb> i'm hitting what I believe to be a bug in apachemonitor (maybe not directly caused by apachemonitor), and I'm wondering if this is perhaps fixed in 4.1.0
[27-Oct-2011 14:15:24] <rocket> I am looking at the release notes now
[27-Oct-2011 14:15:57] <rocket> jb did you get the latest release note file?
[27-Oct-2011 14:16:04] <jb> yeah, i just looked at it
[27-Oct-2011 14:16:11] <jb> i don't see any info in it though
[27-Oct-2011 14:16:46] <jb> actually, this is all that I see on the site: Global_Operations_Manager_Release_Notes_34-092011-4.1-v01.pdf
[27-Oct-2011 14:17:16] <jb> ah, now I see the SD release notes.
[27-Oct-2011 14:17:25] <jb> navigation is confusing on the new support site
[27-Oct-2011 14:18:06] <rocket> Zenoss_Resource_Manager_Release_Notes_23-092011-4.1-v01.pdf is the one you want
[27-Oct-2011 14:18:13] <jb> got it.
[27-Oct-2011 14:18:40] <rocket> I am not sure your particular issue is resolved but there are a fair number of bug fixes applied
[27-Oct-2011 14:18:59] <rocket> as well as some new auditing code to track ui changes etc
[27-Oct-2011 14:19:40] <jb> sweet, I'll go ahead and upgrade and see what happens.
[27-Oct-2011 14:21:07] <jb> lots of bugfixes
[27-Oct-2011 14:22:48] <kenhen93> @rocket: how did you fix this issue? message/62207#62207
[27-Oct-2011 14:23:00] <rocket> jb its always hard to say if they cover your particular issue though without more information
[27-Oct-2011 14:23:14] <jb> rocket: I have a support ticket open, so we'll see what they say,.
[27-Oct-2011 14:23:18] <jb> i'm going to upgrade now.
[27-Oct-2011 14:26:06] <rocket> kenhen93: try just componentScanValue = 'id'
[27-Oct-2011 14:26:21] <kenhen93> ok, let me try that
[27-Oct-2011 14:27:24] <rocket> kenhen93: sorry that was an ancient post from before becoming an employee .. I didnt even remember asking it
[27-Oct-2011 14:27:48] <kenhen93> no worries, I really hope this works! Appreciate your help
[27-Oct-2011 14:28:56] <rocket> kenhen93: to be honest I did resort to import pdb;pdb.set_trace() debugging the daemons to figure out my issues back then ..
[27-Oct-2011 14:29:27] <jb> rocket: heard of any more dev on the AIX stuff?
[27-Oct-2011 14:31:35] <rocket> jb: no I havent, I have been swamped in the support side lately.
[27-Oct-2011 14:31:51] <jb> ahh
[27-Oct-2011 14:32:20] <rocket> jb: I doubt there has been, but I could be wrong
[27-Oct-2011 14:32:40] <jb> im guessing no n:)
[27-Oct-2011 14:32:42] <jb> no.
[27-Oct-2011 14:33:03] <rocket> I have just started working on a new zenpack, but this is the first one I have worked on in almost a year
[27-Oct-2011 14:33:09] <jb> some of the enterprise packs just don't cut it anymore..
[27-Oct-2011 14:33:15] <jb> and nobody has looked at them in years
[27-Oct-2011 14:33:18] <jb> (AIX, bigip, etc)
[27-Oct-2011 14:33:27] <jb> what ya working on?
[27-Oct-2011 14:33:37] <rocket> a puppet zenpack
[27-Oct-2011 14:33:42] <jb> hrm, whats it do?
[27-Oct-2011 14:33:57] <dpetzel> Anyone aware of issues with infrastructure page issues on a 3.1 --> 3.2.1 upgrade? My problem is it sitts spinning for 5-10 minutes at least, it eventually comes up with no devices listed, however if I say type something in the device name filter box they all show up
[27-Oct-2011 14:34:16] <rocket> I have pushed out my puppet management pieces already .. eg it has the puppet modules to install zenoss etc
[27-Oct-2011 14:34:49] <jb> ah so its more for management and not integration?
[27-Oct-2011 14:34:54] <rocket> jb: this was just stuff I was using for my lab etc .. it needs to be enhanced for 4.X but its something
[27-Oct-2011 14:35:09] <jb> i'd like to see those modules.
[27-Oct-2011 14:35:10] <rocket> jb: I just started adding the zenoss integration pieces now
[27-Oct-2011 14:36:08] <rocket> https://github.com/zenoss/ZenPacks.zenoss.Puppet
[27-Oct-2011 14:36:49] <jb> my puppet is so rusty.. haven't had time for it lately :/
[27-Oct-2011 14:37:01] <kenhen93> @rocket: now my parser actually does not error out but getting no results :-/
[27-Oct-2011 14:37:03] <kenhen93> 2011-10-27 14:36:00,830 DEBUG zen.ComponentCommandParser: ParsedResults events: [] values: []}
[27-Oct-2011 14:37:08] <rocket> jb: eg I basically am adding new components to a device that lists the puppet clients
[27-Oct-2011 14:37:22] <jb> ah
[27-Oct-2011 14:37:36] <mattray> rocket: that README.txt isn't doing much for me
[27-Oct-2011 14:38:04] <rocket> mattray: the whole zenpack is incomplete ..
[27-Oct-2011 14:38:15] <rocket> all thats in there is the puppet_modules folder
[27-Oct-2011 14:38:42] <rocket> mattray: you know how things go .. I havent got any time for anything .. I am too busy saving the day with the support cases
[27-Oct-2011 14:39:12] <mattray> good times
[27-Oct-2011 14:39:16] <rocket> once I get the components for puppet clients, I would like to see what data is available to graph per client etc
[27-Oct-2011 14:39:42] <rocket> I then plan on adding a modeller that will take the puppetca information and store that as well
[27-Oct-2011 14:39:48] <mattray> yeah, I was just curious what kind of stuff you'd be monitoring
[27-Oct-2011 14:39:50] <jb> rocket: runtime, resources, etc.
[27-Oct-2011 14:39:54] <rocket> eg how long the certificates are valid for etc
[27-Oct-2011 14:39:54] <jb> all of that can be graphed.
[27-Oct-2011 14:40:23] <Hackman238> CbQoS modeller in 5 days
[27-Oct-2011 14:40:32] <jb> ?
[27-Oct-2011 14:40:39] <rocket> I think I will take some of the community puppet report/zenoss integration pieces and update them to use the new zenoss api
[27-Oct-2011 14:40:45] <mattray> rocket: you should look at the Chef zenpack, maybe you can use Facter as your modeler
[27-Oct-2011 14:40:57] <rocket> eg have events fire when an error has occurred etc
[27-Oct-2011 14:41:00] <Hackman238> Pack of mine will be releasing.
[27-Oct-2011 14:41:10] <jb> ah, what is it?
[27-Oct-2011 14:41:34] <mattray> rocket: I never finished it, but the Chef ZenPack just provides modeling
[27-Oct-2011 14:41:37] <rocket> Hackman238: I have children and a very busy day job .. its a miracle I have anything done
[27-Oct-2011 14:41:47] <Hackman238> jb: Models Service Policies, Class Maps and Police QoS policies
[27-Oct-2011 14:42:01] <Hackman238> rocket: I hear that.
[27-Oct-2011 14:42:06] <jb> ah ok
[27-Oct-2011 14:42:08] <Hackman238> rocket: Also, you were right. It was in the gc
[27-Oct-2011 14:42:19] <jb> Hackman238: still looking forward to the bigip zenpack
[27-Oct-2011 14:42:34] <Hackman238> jb: It's coming
[27-Oct-2011 14:43:08] <rocket> Hackman238: you got it cleaned up?
[27-Oct-2011 14:43:09] <Hackman238> rocket: You deserve a raise!
[27-Oct-2011 14:43:15] <Hackman238> rocket: Yep.
[27-Oct-2011 14:43:34] <rocket> Hackman238: lol its what I do .. ... every day ...
[27-Oct-2011 14:43:45] <Hackman238> rocket: Looks like those pages were tracing since it was using tales that called a method that was fetching from catalog
[27-Oct-2011 14:44:02] <rocket> which pages again?
[27-Oct-2011 14:44:25] <Hackman238> rocket: interfaceGraphs, shows all the interface graphs on one page
[27-Oct-2011 14:44:40] <Hackman238> rocket: And implements date/time menus
[27-Oct-2011 14:45:12] <rocket> ah yes that one ..
[27-Oct-2011 14:45:17] <rocket> so that is resolved as well?
[27-Oct-2011 14:45:27] <Hackman238> rocket: Looks as so.
[27-Oct-2011 14:45:43] <Hackman238> rocket: Still getting conflicts though
[27-Oct-2011 14:46:04] <Hackman238> rocket: Might need help to tweak the mysql more. Already caches 100% of the db.
[27-Oct-2011 14:46:25] <Hackman238> ...less it grew substantially in a few weeks.
[27-Oct-2011 14:46:29] <rocket> Hackman238: that is extremely strange .. something is writing alot ...
[27-Oct-2011 14:46:57] <Hackman238> rocket: I'll try backing off the zopes. Might be too many players to take turns
[27-Oct-2011 14:47:33] <rocket> yeah the most I have seen is 20 myself ..
[27-Oct-2011 14:48:39] <Hackman238> I'll give that a shot
[27-Oct-2011 14:49:27] <Hackman238> rocket: poop size set to 22GB
[27-Oct-2011 14:49:30] <Hackman238> *pool
[27-Oct-2011 14:50:38] <Hackman238> rocket: heap 32GB
[27-Oct-2011 14:58:53] <kenhen93> is there any documentation on zWinEventlogClause?
[27-Oct-2011 15:07:06] <jb> ok, upgrade all done.
[27-Oct-2011 15:08:09] <jb> yep, my apachemonitor bug was fixed.
[27-Oct-2011 15:08:10] <jb> yay!
[27-Oct-2011 15:10:13] <rocket> kenhen93: not really unfortunately .. see http://dev.zenoss.org/trac/ticket/5785
[27-Oct-2011 15:10:57] <kenhen93> thanks, I saw that, I was hoping to widdle it down to system log and one source
[27-Oct-2011 15:11:16] <rocket> kenhen93: eg it is just a part of the sql syntax for the windows event system
[27-Oct-2011 15:11:21] <rocket> eg its the where part
[27-Oct-2011 15:11:38] <rocket> where .... TargetInstance.Logfile = "Application"
[27-Oct-2011 15:11:42] <rocket> for example
[27-Oct-2011 15:11:51] <kenhen93> so for source
[27-Oct-2011 15:12:34] <kenhen93> TargetInstance.Logfile = "System", TargetInstance.SourceName = "source" ?
[27-Oct-2011 15:14:51] <rocket> instead of the , you may need an and
[27-Oct-2011 15:15:21] <rocket> kenhen93: to be honest I am not very familiar with the windows syntax on those sql queries
[27-Oct-2011 15:15:49] <kenhen93> ok thanks
[27-Oct-2011 15:16:17] <rocket> kenhen93: at most customers we just grab all of the logs and do a transform to drop unneeded ones
[27-Oct-2011 15:16:34] <jb> so is the consensus that 4.x is faster than 3.x?
[27-Oct-2011 15:16:54] <rocket> jb it should be. The event system is tons faster
[27-Oct-2011 15:17:03] <jb> my 3.x instance is faster
[27-Oct-2011 15:17:05] <kenhen93> @rocket: I have some very chatty windows servers
[27-Oct-2011 15:17:12] <jb> just browsing the interface, etc.
[27-Oct-2011 15:17:18] <jb> and it has about 3 times the devices :/
[27-Oct-2011 15:17:58] <rocket> jb increase the /etc/sysconfig.d/memcache settings from teh default of 512 to 2048 or higher
[27-Oct-2011 15:18:06] <rocket> jb: restart memcached
[27-Oct-2011 15:18:28] <rocket> jb have you tuned the zope.conf settings as well?
[27-Oct-2011 15:18:56] <jb> nah, havent done anything yet.
[27-Oct-2011 15:18:58] <jb> lets see..
[27-Oct-2011 15:19:18] <rocket> jb look at the install guide I believe there is a tuning section .. please follow those steps
[27-Oct-2011 15:19:34] <jb> k
[27-Oct-2011 15:19:35] <rocket> jb you may also need to tune zends
[27-Oct-2011 15:19:59] <rocket> jb because everything is stored in zends(mysql) now .. if that is not tuned it will be slower etc
[27-Oct-2011 15:20:04] <jb> you referring to CACHESIZE in memcached?
[27-Oct-2011 15:20:08] <jmp242> Ok
[27-Oct-2011 15:20:09] <jb> right
[27-Oct-2011 15:20:09] <rocket> yes
[27-Oct-2011 15:20:13] <jb> its 64 on my box.
[27-Oct-2011 15:20:19] <rocket> err wait
[27-Oct-2011 15:20:31] <jmp242> I've merged all the various Zenoss Forum FAQs here: https://zcaportal.org/wiki/bin/view/ZCA/ZCAFAQ
[27-Oct-2011 15:20:44] <jmp242> and done my best to update the old stuff or leave out obsolete things re 2.4.x etc
[27-Oct-2011 15:20:53] <rocket> jb yes CACHESIZE
[27-Oct-2011 15:21:08] <rocket> change it from 64mb to 2Gb or 2048
[27-Oct-2011 15:21:16] <jb> k, thats not in the perf tuningn part
[27-Oct-2011 15:21:23] <jb> k
[27-Oct-2011 15:21:43] <rocket> jb please open a ticket so we can get the docs updated
[27-Oct-2011 15:21:50] <jb> sure
[27-Oct-2011 15:22:57] <jb> done
[27-Oct-2011 15:25:18] <jb> import math; from test import pystone; int(math.ceil(sum(s[1] for s \
[27-Oct-2011 15:25:19] <jb> in (pystone.pystones() for i in range(3)))/150.0))
[27-Oct-2011 15:25:23] <jb> that should just be ran from a shell?
[27-Oct-2011 15:25:33] <jb> or from a python/zendmd shell?
[27-Oct-2011 15:25:41] <rocket> zendmd shell
[27-Oct-2011 15:27:00] <jb> ok cool.. thanks. we'll see if that makes a difference.
[27-Oct-2011 15:28:09] <rocket> most likely zends really needs the tunings however ..
[27-Oct-2011 15:28:56] <kenhen93> @rocket using and between the statements was correct, thanks!
[27-Oct-2011 15:29:38] <jb> rocket: any docs on that?
[27-Oct-2011 15:30:06] <rocket> jb follow the same recommendations for tuning mysql
[27-Oct-2011 15:30:51] <rocket> eg you probably want to take the stuff from your mysql.conf on your fast system and apply it to the zends.conf file
[27-Oct-2011 15:31:01] <jb> gotcha
[27-Oct-2011 15:31:02] <jb> i'll check.
[27-Oct-2011 15:31:33] <rocket> you can also run mysqltuner.pl against it the port is different on zends than a default mysql install
[27-Oct-2011 15:31:40] <rocket> 13306 I believe
[27-Oct-2011 15:37:42] <Hackman238> rocket: whats mysqltuner do?
[27-Oct-2011 15:38:03] <rocket> wget mysqltuner.pl && ./mysqltuner.pl
[27-Oct-2011 15:38:19] <jb> yeah, just a script.
[27-Oct-2011 15:38:20] <rocket> it just prints a report of how your server is working and makes recommendations
[27-Oct-2011 15:38:29] <jb> that doesn't like zends' mysqladmin
[27-Oct-2011 15:38:39] <Hackman238> rocket: Ah, safe to run against production box?
[27-Oct-2011 15:38:43] <jb> yeah
[27-Oct-2011 15:38:46] <jb> it doesn't make changes.
[27-Oct-2011 15:38:47] <rocket> yea its readonly
[27-Oct-2011 15:38:50] <jb> just reccomends.
[27-Oct-2011 15:38:55] <Hackman238> rocket: What do you thin about the error in this thread? message/62237#62237
[27-Oct-2011 15:39:02] <Hackman238> Nice, I'll give it a shot. TY
[27-Oct-2011 15:42:03] <rocket> Hackman238: uhmmm I dont .. other than there is a keyerror on the DeviceClass... he might try rebuilding the globalcatalog
[27-Oct-2011 15:42:10] <rocket> see zencatalog
[27-Oct-2011 15:42:31] <Hackman238> rocket: I had him try that, no luck.
[27-Oct-2011 15:42:44] <Hackman238> rocket: even --forceindex --createcatalog
[27-Oct-2011 15:42:55] <rocket> I guess I would see what zenpacks are installed
[27-Oct-2011 15:43:04] <rocket> there might be a bad one in there ..
[27-Oct-2011 15:43:17] <rocket> this is a core install correct?
[27-Oct-2011 15:43:23] <Hackman238> Yep
[27-Oct-2011 15:45:42] <rocket> yeah I would see about the zenpacks .. eg any that would touch the catalogs
[27-Oct-2011 15:47:59] <Hackman238> rocket: Gotcha. Thanks
[27-Oct-2011 15:49:15] <rocket> Hackman238: or it is just a bad catalog somewhere
[27-Oct-2011 15:49:25] <rocket> so clear the contents and start reindexing etc
[27-Oct-2011 15:49:31] <jb> the update.zenoss.org query in 4.0.2 makes things slower too.
[27-Oct-2011 15:49:35] <jb> wonder if that can be disabled?
[27-Oct-2011 15:49:36] <rocket> the traceback is in the catalogs
[27-Oct-2011 15:49:55] <Hackman238> rocket: Can't hurt
[27-Oct-2011 15:50:20] <Hackman238> rocket: any trick to mysqltuner.pl? seems to not like a blank root
[27-Oct-2011 15:50:41] <rocket> --help should show you how to use a username and password etc
[27-Oct-2011 15:51:04] <Hackman238> rocket: Yeah, looks like its formulated right.
[27-Oct-2011 15:51:13] <rocket> jb: can you have the zenoss server instance contact update.zenoss.org?
[27-Oct-2011 15:52:06] <rocket> jb: it falls back to a browser based query if it cannot .. but I havent looked into how that stuff works
[27-Oct-2011 15:53:27] <jb> well, it is able to reach updates.zenoss.org.
[27-Oct-2011 15:53:38] <jb> but it's often delayed for whatever reason, which causes delays during page loads.
[27-Oct-2011 15:54:48] <jb> hrm, browsing the INFRASTRUCTURE tab is super slow for some reason.. mysql looks fine.
[27-Oct-2011 15:55:19] <jb> after I hit the INFRASTRUCTURE button, its at least 20 seconds usually before anything appear
[27-Oct-2011 15:55:21] <jb> 15-20secs
[27-Oct-2011 15:55:32] <Hackman238> '*** MySQL's maximum memory usage is dangerously high ***'
[27-Oct-2011 15:55:34] <Hackman238>
[27-Oct-2011 15:55:56] <dpetzel> jb: I'm fighting a very similar issue after upgrading to 3.2.1 today
[27-Oct-2011 15:56:07] <dpetzel> except I'm in the minute range
[27-Oct-2011 15:56:12] <jb> no organizers or anything appear
[27-Oct-2011 15:56:23] <dpetzel> jb same here, totally blank
[27-Oct-2011 15:56:29] <dpetzel> however if I type a letter into the device filter
[27-Oct-2011 15:56:32] <dpetzel> I see devices
[27-Oct-2011 15:56:42] <dpetzel> but no ogranizers
[27-Oct-2011 15:57:00] <jb> weird
[27-Oct-2011 15:57:02] <rocket> dpetzel rebuild your global catalog
[27-Oct-2011 15:57:40] <dpetzel> rocket: You have a link handy on how that is done?
[27-Oct-2011 15:57:43] <dpetzel> Ive never done it
[27-Oct-2011 15:57:45] <rocket> take a look at the zencatalog
[27-Oct-2011 15:57:59] <rocket> zencatalog --forceindex --createcatalog
[27-Oct-2011 15:58:24] <dpetzel> safe? shoudl I back up first or anything?
[27-Oct-2011 15:58:25] <rocket> dpetzel: you may also try running zenmigrate with "zenmigrate"
[27-Oct-2011 15:58:41] <rocket> dpetzel: I would backup just in case .. but its reasonably safe
[27-Oct-2011 15:58:50] <dpetzel> rocket: Thanks
[27-Oct-2011 15:58:53] <jb> maybe I should try zencatalog as well, rocket?
[27-Oct-2011 15:58:53] <rocket> dpetzel: its possible part of the zenmigrate failed
[27-Oct-2011 15:59:06] <rocket> jb: i dont see why you should do that ...
[27-Oct-2011 15:59:13] <dpetzel> rocket: should I try zenmigrate before zencatalog?
[27-Oct-2011 15:59:16] <jb> hmm
[27-Oct-2011 15:59:20] <rocket> dpetzel: probably
[27-Oct-2011 15:59:48] <dpetzel> rocket: that ran fast, like instant. any options I should pass?
[27-Oct-2011 15:59:57] <rocket> dpetzel: no
[27-Oct-2011 16:00:08] <rocket> well what version did you come from?
[27-Oct-2011 16:00:16] <dpetzel> 3.1 --> 3.2.1
[27-Oct-2011 16:00:51] <rocket> dpetzel: yeah redo the zencatalog next
[27-Oct-2011 16:01:05] <dpetzel> rocket: OK will do in a little bit
[27-Oct-2011 16:02:32] <rocket> jb: I would open a support ticket if it doesnt get any better. There might be something else going on there
[27-Oct-2011 16:07:25] <jb> will do
[27-Oct-2011 16:23:14] <jmp242> Hackman238: you got a minute? Can you look at the thread: message/62268#62268
[27-Oct-2011 16:23:35] <jmp242> I've not seen an event mapped to an existing class with an error event that is a clear event *not* clear the error event
[27-Oct-2011 16:33:54] <deadbum> can anyone point me to an article on adding nagios plugins to zenoss 3.0?
[27-Oct-2011 16:34:07] <deadbum> I've got one added but it doesn't seem to actually alert when the state changes
[27-Oct-2011 16:34:19] <deadbum> the "test" feature works as expected.
[27-Oct-2011 16:43:57] <humanix> hi there guys
[27-Oct-2011 16:44:07] <humanix> hi rmatte, Hackman238
[28-Oct-2011 00:00:01] [disconnected at Fri Oct 28 00:00:01 2011]
[28-Oct-2011 00:00:02] [connected at Fri Oct 28 00:00:02 2011]
[28-Oct-2011 00:00:12] [zenoss-logger (logger bot) has joined #zenoss]
[28-Oct-2011 00:17:14] <Tabmow> nyeates: we do have the Enterprise version of Zenoss fwiw
[28-Oct-2011 01:13:43] <QubeZ> hello all
[28-Oct-2011 01:14:22] <QubeZ> i keep getting access denied to read windows event logs. I opened ports 161 on the server udp and tcp then validated the connectin from another windows machine using wbemtest. Still Zenoss generates these messages.
[28-Oct-2011 02:55:34] <fragfutter> Quadro: 161 is snmp. you won't get event logs there
[28-Oct-2011 02:56:29] <fragfutter> oh, already gone. well...
[28-Oct-2011 04:38:13] <enkrypt> hello! is there any way to reset or clean or delete the network topology that zenoss has built?
[28-Oct-2011 04:38:18] <enkrypt> or to simply disable that feature?
[28-Oct-2011 05:02:18] <enkrypt> I'm looking for a solution to http://dev.zenoss.org/trac/ticket/4104
[28-Oct-2011 07:32:33] <enkrypt> Hackman238, nyeates, rmatte, rocket, psyklone, anyone alive?
[28-Oct-2011 08:57:27] <jb> mornin
[28-Oct-2011 08:58:05] <psyklone> uuuuuuuuugh
[28-Oct-2011 08:58:37] <jmp242> Good morning everyone
[28-Oct-2011 08:58:54] <psyklone> coffee
[28-Oct-2011 09:16:27] <kiven> hello
[28-Oct-2011 09:36:00] <dpetzel> rocket: u around?
[28-Oct-2011 09:50:53] <jb> Hackman238: the newest BIG-IP pack actually has some additional datasources.
[28-Oct-2011 10:01:24] <rocket> dpetzel: yeah
[28-Oct-2011 10:02:20] <dpetzel> rocket: Hey, just wanted to circle back from yesterday running catalag didnt seem to help my cause. I tried to write down all the revelant details: message/62253#62253 (If you have a minute)
[28-Oct-2011 10:02:34] <rocket> for all in the halloween spirit http://www.youtube.com/watch?v=UfcNoMnKjrY&feature=player_embedded
[28-Oct-2011 10:06:42] <rocket> dpetzel: is there a traceback in the event.log?
[28-Oct-2011 10:07:15] <dpetzel> rocket: I didnt see one
[28-Oct-2011 10:09:08] <dpetzel> eventlog doesnt reall seem to log much after statup: http://pastebin.com/kf6dMWMp
[28-Oct-2011 10:31:46] <jb> rocket: just opened a ticket about the slowness.
[28-Oct-2011 10:36:50] <enkrypt> check out what my zenhub.py at 100% cpu load is causing, several times per day:
[28-Oct-2011 10:36:52] <enkrypt> http://dl.dropbox.com/u/13259098/z_configtimes.png
[28-Oct-2011 10:36:54] <enkrypt> http://dl.dropbox.com/u/13259098/z_datapoints.png
[28-Oct-2011 10:36:56] <enkrypt> http://dl.dropbox.com/u/13259098/z_worklistlength.png
[28-Oct-2011 10:37:32] <enkrypt> zenhub worklistlength goes very high, config times go through the roof, data points drop to zero, and events get queued up by the thousands
[28-Oct-2011 10:37:50] <enkrypt> and the end result: my graphs are showing gaps
[28-Oct-2011 10:40:59] <rocket> enkrypt: can you split the hub into two? and then balance the collectors against that?
[28-Oct-2011 10:41:34] <enkrypt> rocket: unfortunately not possible at the moment
[28-Oct-2011 10:41:46] <enkrypt> the thing is it didn't used to be like this
[28-Oct-2011 10:41:54] <enkrypt> the problem started showing up like 2 weeks ago
[28-Oct-2011 10:41:59] <enkrypt> I'm debugging it with psyklone
[28-Oct-2011 10:42:13] <rocket> ok .. if your debugging it with him I will let him continue
[28-Oct-2011 10:43:08] <enkrypt> yesterday we thought we had it, but today the problem reappeared
[28-Oct-2011 10:49:08] <jb> hrm, since upgrading, a few of my graphs are showing 46.2T instead of 46.2M
[28-Oct-2011 10:49:36] <jb> what am I missing here?
[28-Oct-2011 10:49:43] <jb> the datasource has an RPN of 8,*
[28-Oct-2011 10:54:04] <enkrypt> about this: http://dev.zenoss.org/trac/ticket/4104 is there a way to simply disable the topology stuff? because i'm using a lot of VRF and crazy network stuff, and the topology detection doesn't work well, I would prefer to disable it outright
[28-Oct-2011 11:00:13] <rocket> enkrypt: stop modelling the routes and then remove the routes on the devices
[28-Oct-2011 11:00:31] <enkrypt> rocket: aha! ok that sounds interesting
[28-Oct-2011 11:00:37] <rocket> enkrypt: that stuff is totally different in v4 btw... we no longer rely on route detection from modelling
[28-Oct-2011 11:00:59] <enkrypt> rocket: does that have any negative consequences? anyway my topology doesn't get built succesfully, so I would prefer to turn the feature off entirely
[28-Oct-2011 11:01:05] <rocket> enkrypt: in your case you will see more not in topology messages but they can be ignored
[28-Oct-2011 11:01:38] <rocket> enkrypt: if your not using it, leaving the routes out of the database will improve performance
[28-Oct-2011 11:01:57] <enkrypt> yeah I was thinking in that direction
[28-Oct-2011 11:05:51] <nyeates> any ZCA ppl, call ongoing
[28-Oct-2011 11:05:55] <nyeates> om me if you need URL
[28-Oct-2011 11:06:00] <nyeates> om = pm
[28-Oct-2011 11:16:16] <aclark> I'm using mysqlmonitor zenpack and i want to change the poll interval… had 4 minutes of downtime with no notice, does zProdstateThreshold do that? Or something else…
[28-Oct-2011 11:21:06] <aclark> Oh here we go, docs/DOC-2445#Howdoesonechangethepingpollingintervalforaspecificdeviceclass
[28-Oct-2011 11:21:14] <aclark> change zCommandCycleTime
[28-Oct-2011 11:21:16] <dhopp> aclark: in the data source there is a "Cycle Time" that you can change down to 60 seconds if you want
[28-Oct-2011 11:21:23] <dhopp> but you will need to delete your RRD files
[28-Oct-2011 11:21:50] <aclark> dhopp: ah otherwise they'll be wrong-ish?
[28-Oct-2011 11:22:47] <dhopp> aclark: yes..because right now the rrd would have a step of 300 but data would be entered every 60 seconds...
[28-Oct-2011 11:22:53] <dhopp> I'm not sure what it would look like
[28-Oct-2011 11:23:11] <aclark> heh
[28-Oct-2011 11:23:12] <aclark> k
[28-Oct-2011 11:23:18] <dhopp> aclark: you just need to remove the RRD files for those datapoints...
[28-Oct-2011 11:23:35] <aclark> dhopp: cool, i'll look for them
[28-Oct-2011 11:32:28] <jb> aclark: they are in $ZENHOME/perf/Devices/<devicename>
[28-Oct-2011 11:32:34] <rocket> dhopp: it would look the same as before .. it would only save the 1st of the 5 datapoints in the rrd file and would exclude the rest as garbage
[28-Oct-2011 11:32:57] <dhopp> rocket: ahh…I wasn't sure how it would handle that
[28-Oct-2011 11:34:04] <aclark> jb, dhopp all the zencommand_ rrds? I don't see anything mysql specific
[28-Oct-2011 11:36:24] <dhopp> aclark: the filenames should have mysql in the name
[28-Oct-2011 11:37:10] <aclark> dhopp: found'em. thanks
[28-Oct-2011 11:38:50] <aclark> dhopp: do i need to restart anything?
[28-Oct-2011 11:43:23] <dhopp> you might need to restart zenhub and zencommand
[28-Oct-2011 11:43:32] <aclark> dhopp: k thanks
[28-Oct-2011 11:50:13] <aclark> dhopp: and… cycle time already defaults to 60 :-)
[28-Oct-2011 11:50:49] <aclark> guessing my zenoss did not loose contact w/mysql but that maybe some other unmonitored resource did
[28-Oct-2011 12:00:01] [disconnected at Fri Oct 28 12:00:01 2011]
[28-Oct-2011 12:00:01] [connected at Fri Oct 28 12:00:01 2011]
[28-Oct-2011 12:00:11] [zenoss-logger (logger bot) has joined #zenoss]
[28-Oct-2011 13:18:07] <Sashness> does anyone know how i could export the list of devices Zenoss found a process running on
[28-Oct-2011 13:18:19] <Sashness> i see the list in Zenoss but there doesn't seem to be any quick way to export hte list
[28-Oct-2011 13:18:30] <Sashness> and i can't find anything via zendmd
[28-Oct-2011 13:25:07] <Sashness> well i just found d.os.processes()
[28-Oct-2011 14:07:13] <humanix> zup rmatte
[28-Oct-2011 14:07:57] kocolosk|away is now known as kocolosk
[28-Oct-2011 14:12:57] <jb> crap, i lost my f5 transforms some how.
[28-Oct-2011 15:03:48] <Hackman238> Hey all!
[28-Oct-2011 15:06:16] <nyeates> hilo
[28-Oct-2011 15:07:05] <nyeates> jb: hope you had some backups
[28-Oct-2011 15:12:21] <jmp242> Hi Hackman238
[28-Oct-2011 15:12:34] <jmp242> hey do you remember the graph range ticket #?
[28-Oct-2011 15:12:53] <Hackman238> jmp242: No, sorry
[28-Oct-2011 15:13:02] <Hackman238> nyeates: Can you op me?
[28-Oct-2011 15:13:17] <dhopp> jmp242: which one? there are a couple
[28-Oct-2011 15:13:19] <dhopp> http://dev.zenoss.org/trac/ticket/7879
[28-Oct-2011 15:13:23] <Hackman238> nyeates: and jmp242s also
[28-Oct-2011 15:13:40] <nyeates> we fixed the graph range thing internally
[28-Oct-2011 15:13:52] <nyeates> but i think it involved 4.* things :-/
[28-Oct-2011 15:13:53] <dhopp> http://dev.zenoss.org/trac/ticket/7873
[28-Oct-2011 15:14:13] <jmp242> ahh, the first
[28-Oct-2011 15:14:14] <jmp242> thanks
[28-Oct-2011 15:14:39] <nyeates> as to the ops: yes sure! do either of you have your nicks registered on freenode?
[28-Oct-2011 15:14:43] <nyeates> if so, ill set auto ops
[28-Oct-2011 15:14:59] <jmp242> mine is
[28-Oct-2011 15:15:03] <jmp242> registered on freenode
[28-Oct-2011 15:15:06] <jmp242> makes me log in all the time
[28-Oct-2011 15:15:10] <nyeates> coolio, will do now
[28-Oct-2011 15:15:41] <nyeates> gotta find that command
[28-Oct-2011 15:15:46] <nyeates> i need to doc those
[28-Oct-2011 15:16:25] <jmp242> lol
[28-Oct-2011 15:16:40] <Hackman238> nyeates: Yep
[28-Oct-2011 15:17:44] <jmp242> Hackman238: can you look at this? message/62268#62268
[28-Oct-2011 15:17:53] <nyeates> cmon ChanServ....dont let me down
[28-Oct-2011 15:18:00] <jmp242> I've described exactly what I've done in some situations
[28-Oct-2011 15:18:10] <jmp242> yet his doesn't work for *one* situation, but does work for others
[28-Oct-2011 15:22:09] <nyeates> [15:21] -ChanServ- 12 jmp242 +O [modified 4 minutes, 31 seconds ago]
[28-Oct-2011 15:22:09] <nyeates> [15:21] -ChanServ- 13 Hackman238 +O [modified 32 seconds ago]
[28-Oct-2011 15:22:12] <nyeates> but why no ops?
[28-Oct-2011 15:22:30] <nyeates> leave and come back
[28-Oct-2011 15:25:44] <Hackman238> jmp242: Sure.
[28-Oct-2011 15:25:44] <jmp242> leave and come back? lol
[28-Oct-2011 15:25:46] <jmp242> ok
[28-Oct-2011 15:26:02] <jmp242> Yay, I'm back
[28-Oct-2011 15:26:05] <nyeates> booya
[28-Oct-2011 15:26:12] <Hackman238> I'm op now
[28-Oct-2011 15:26:13] <nyeates> double booya
[28-Oct-2011 15:26:17] <jmp242> lol
[28-Oct-2011 15:26:20] <jmp242> happy friday!
[28-Oct-2011 15:26:26] <Hackman238> nyeates: Did you head about the meeting results?
[28-Oct-2011 15:26:29] * nyeates clanks beer
[28-Oct-2011 15:26:41] <nyeates> dont think so
[28-Oct-2011 15:29:10] <Hackman238> nyeates: PM'd you
[28-Oct-2011 16:21:41] <tightwork> Hello. I have two machines that have the same private network 172.16.254.0 but are actually physically different, yet zenoss sees them as together. Is there anyway to separate them?
[28-Oct-2011 16:29:14] <tightwork> mork calling orson?
[28-Oct-2011 16:29:44] <nexex> did you add them by dns names?
[28-Oct-2011 16:33:20] <tightwork> I added my cisco switch, from there I discovered devices by snmp and both devices have the same network. Looking at device map shows them both connected but in reality they are not
[28-Oct-2011 16:36:51] <tightwork> pretty big issue
[28-Oct-2011 16:53:41] <Hackman238> ....Oh man...http://seattle.cbslocal.com/2011/10/28/not-a-halloween-costume-washington-man-cuts-off-arm-with-guillotine/
[28-Oct-2011 17:00:23] <Hackman238> Later all!
[28-Oct-2011 17:07:32] <Ve2dmn> I have SNMP fails
[28-Oct-2011 17:08:00] <Ve2dmn> yet, If I do it manually, it works...
[28-Oct-2011 17:09:10] <Ve2dmn> What's worse is googling for the exact error message only get me my pastebin from 2 days ago...
[28-Oct-2011 17:12:18] <tightwork> firewall?
[28-Oct-2011 17:13:21] kocolosk is now known as kocolosk|away
[28-Oct-2011 17:13:28] <Ve2dmn> It's a switch...
[28-Oct-2011 17:15:53] <dpetzel> anyone know why this fix isnt making it into subsequent released. Just noticed that during my 3.2.1 upgrade the patch is gone again
[28-Oct-2011 17:15:54] <dpetzel> http://dev.zenoss.com/trac/ticket/7442
[28-Oct-2011 17:19:38] <tightwork> Ve2dmn: you can snmpget the node from the zenoss box?
[28-Oct-2011 17:28:05] <Ve2dmn> What I tried is to go to Monitoring templates and test the datasource...
[28-Oct-2011 17:29:07] <Ve2dmn> (ie snmpwalk -cpublic -v1 192.168.2.230 1.3.6.1.2.1.2.2.1.16 against 192.168.2.230 )
[28-Oct-2011 17:29:19] <Ve2dmn> it works...
[28-Oct-2011 17:29:48] <Ve2dmn> but with zenperfsnmp: WARNING zen.zenperfsnmp: Continuing to wait for response from 192.168.2.230 after 149.99902606 seconds
[28-Oct-2011 17:46:52] <humanix> anyone knows if there is a zenpack to monitor a machine disk via ssh's df -h?
[28-Oct-2011 17:47:47] <humanix> cause the appliance doensn't seem to have disk mibs
[28-Oct-2011 18:17:31] <Praxi> so for a custom monitoring template, I am trying to set the min and max y in the graph definition. It lets me type in -25, but it does not appear to show the graph below 0. My data is coming in around 1 to 3 , so its hard to tell if it goes lower, whether or not it will push it down. anyone have any ideas on this?
[29-Oct-2011 00:00:01] [disconnected at Sat Oct 29 00:00:01 2011]
[29-Oct-2011 00:00:01] [connected at Sat Oct 29 00:00:01 2011]
[29-Oct-2011 00:00:13] [zenoss-logger (logger bot) has joined #zenoss]
[29-Oct-2011 09:57:36] <tja> hi, anyone here who is familiar with monitoring via ssh and could help the noob ?
[29-Oct-2011 11:13:45] <enkrypt> hi guys. still having the issue where my zenhub.py is periodically taking 100% cpu and destroying all performance
[29-Oct-2011 12:00:01] [disconnected at Sat Oct 29 12:00:01 2011]
[29-Oct-2011 12:00:02] [connected at Sat Oct 29 12:00:02 2011]
[29-Oct-2011 12:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[30-Oct-2011 00:00:01] [disconnected at Sun Oct 30 00:00:01 2011]
[30-Oct-2011 00:00:02] [connected at Sun Oct 30 00:00:02 2011]
[30-Oct-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]
[30-Oct-2011 12:00:01] [disconnected at Sun Oct 30 12:00:01 2011]
[30-Oct-2011 12:00:01] [connected at Sun Oct 30 12:00:01 2011]
[30-Oct-2011 12:00:16] [zenoss-logger (logger bot) has joined #zenoss]
[30-Oct-2011 12:11:59] <hmp> just stumbeled upon this, probably everybody knows about it, but wth: http://is.gd/wHVgMj
[30-Oct-2011 13:40:23] kocolosk|away is now known as kocolosk
[30-Oct-2011 13:46:23] * aclark goes for the redis zenpack gold today
[31-Oct-2011 00:00:01] [disconnected at Mon Oct 31 00:00:01 2011]
[31-Oct-2011 00:00:01] [connected at Mon Oct 31 00:00:01 2011]
[31-Oct-2011 00:00:13] [zenoss-logger (logger bot) has joined #zenoss]
[31-Oct-2011 08:28:13] <aryan> hihi! anyone using OpenStack zenpack?
[31-Oct-2011 08:46:09] <jb> morning
[31-Oct-2011 08:46:26] <jmp242> morning
[31-Oct-2011 08:47:50] <aryan> jmp242: I'm experiencing an issue on OpenStack zenpack, wondering if you can help me with it
[31-Oct-2011 08:48:28] <aryan> http://pastebin.com/i7KFqEwH
[31-Oct-2011 08:49:47] <jmp242> I don't use the OpenStack zenpack so can't really help
[31-Oct-2011 08:50:34] <Sashness> aryan: looksl ike its calling an API thats not there. I see a 404 error at the very end of the your pastebin
[31-Oct-2011 08:52:37] <aryan> Sashness: actually, that is the point. I know that the API is there
[31-Oct-2011 08:52:57] <aryan> Sashness: because other apps are working fine
[31-Oct-2011 08:53:41] <aryan> Sashness: I afraid that the passed parameter is somehow modified in the zenpack
[31-Oct-2011 08:53:51] <jb> hrm, wondering if someone could take a look at this transform, it seems to be bailing at the gethostbyaddr(ip) function, even though I thought I was importing it correctly (worked in previous versions). http://pastebin.com/GMZpakpM
[31-Oct-2011 08:53:55] <Sashness> hmm. guessing that maybe its calling it in a weird way then. Sorry, I don't use OpenStack so i can't test much
[31-Oct-2011 08:54:36] <aryan> thanks :), Sashness: do you know where can I find Nick?
[31-Oct-2011 08:54:53] <aryan> Nick Yeates
[31-Oct-2011 08:55:04] <aryan> He is the author
[31-Oct-2011 08:55:10] <jb> hrm, i think its actually bailing because the host is not resolvablne.
[31-Oct-2011 08:55:17] <jb> but, I don't know which host it's trying to resolve, so I can't fix it.
[31-Oct-2011 08:55:56] <rocket> aryan: Nick is NOT the author for that .. he is just the publisher
[31-Oct-2011 08:56:13] <jb> herror: [Errno 1] Unknown host
[31-Oct-2011 08:56:25] <aryan> rocket: neat, then who is the author
[31-Oct-2011 08:56:29] <jb> so, maybe I need to rewrite it to display the IP only if gethostbyipaddr can't resolve it?
[31-Oct-2011 08:56:41] <rocket> aryan Chet Luther
[31-Oct-2011 08:57:55] <jb> no idea how to do that tho :/
[31-Oct-2011 08:58:34] <aryan> rocket: what about the Chet's nick?
[31-Oct-2011 08:58:57] <jb> aryan: cluther.. he's not online.
[31-Oct-2011 09:03:37] kocolosk is now known as kocolosk|away
[31-Oct-2011 09:14:52] kocolosk|away is now known as kocolosk
[31-Oct-2011 10:21:15] <rmatte> jb: http://pastebin.com/6GB8NT72
[31-Oct-2011 10:21:22] <rmatte> that's the solution to your transform problem
[31-Oct-2011 10:23:04] <SDuensin> Greetings.
[31-Oct-2011 10:24:44] <Sashness> salutations
[31-Oct-2011 10:37:38] <SDuensin> Hey Sashness
[31-Oct-2011 10:37:47] * SDuensin thinks today is going to be a Zenoss day.
[31-Oct-2011 10:39:05] <Ve2dmn> I need some much needed help with some (now) week old NaN graphs...
[31-Oct-2011 10:39:20] <Ve2dmn> Should I be asking the question here or in the forum?
[31-Oct-2011 10:39:27] <SDuensin> Heck. I need help with *all* graphs. :-)
[31-Oct-2011 10:39:48] <Ve2dmn> It's not "critical" but it's getting anoying....
[31-Oct-2011 10:39:49] <rmatte> Ve2dmn: we need more info about the graphs
[31-Oct-2011 10:39:59] <rmatte> i.e, what kind of device, is it all graphs for that device or a specific one?
[31-Oct-2011 10:40:06] <Ve2dmn> I'm down to doing tcpdumps now and comparing the results but I don't know SNMP enough to know what I'm reading
[31-Oct-2011 10:40:21] <Ve2dmn> Ok. The device itself is a D-link switch.
[31-Oct-2011 10:40:38] <SDuensin> I've been reading code and can't find where/how to get my custom SNMP OIDs mapped to things like CPU usage, disk space usage, interface usage, etc.
[31-Oct-2011 10:40:42] <Ve2dmn> I'm monitoring the differents eth interfaces.
[31-Oct-2011 10:41:23] <rmatte> and which graphs are failing?
[31-Oct-2011 10:41:36] <rmatte> SDuensin: not really sure what you mean by that
[31-Oct-2011 10:41:54] <rmatte> SDuensin: in the context of how you explained it, that's done by collector plugins
[31-Oct-2011 10:41:55] <Ve2dmn> This device used to be monitored correctly. I've unpplugged it and moved it elsewhere.
[31-Oct-2011 10:42:35] <rmatte> Ve2dmn: ok, and which graphs aren't working?
[31-Oct-2011 10:42:40] <Ve2dmn> and now ALL the graphs are basicly NaN
[31-Oct-2011 10:42:52] <SDuensin> rmatte, I'll look again, but for example, I can't find where to write the NIC data usage stats into Zenoss in my plugin.
[31-Oct-2011 10:42:58] <rmatte> even under the Graphs section?
[31-Oct-2011 10:43:34] <rmatte> SDuensin: the actual data collection itself is done by templates
[31-Oct-2011 10:43:43] <rmatte> SDuensin: the plugins just create the objects that the templates use
[31-Oct-2011 10:43:59] <rmatte> the templates contain the OIDs for each value minus the index value (which is what is picked up by the modeler)
[31-Oct-2011 10:44:23] <Ve2dmn> rmatte: I erased the decive and re-added it under a different name, if that changes anything...
[31-Oct-2011 10:44:24] <rmatte> SDuensin: what exactly are you trying to do?
[31-Oct-2011 10:44:40] <rmatte> Ve2dmn: have you verified that you can at least snmpwalk the device?
[31-Oct-2011 10:45:02] <rmatte> i.e: snmpwalk -v2c -c your-snmp-string your-device ifName
[31-Oct-2011 10:45:11] <SDuensin> rmatte, I've got Windows clients with net-snmp on them. I'm using the "extend" feature to do data collection on them. So my OIDs aren't the "usual" ones.
[31-Oct-2011 10:45:32] <rmatte> SDuensin: but they are related to existing components?
[31-Oct-2011 10:45:57] <Ve2dmn> rmatte: I can no problem.
[31-Oct-2011 10:46:18] <SDuensin> rmatte, yes
[31-Oct-2011 10:46:30] <rmatte> Ve2dmn: ssh to your zenoss server, as the zenoss user issue the following: zenperfsnmp run -v10 -d yourdevicename
[31-Oct-2011 10:46:37] * SDuensin is still confused by the whole template thing
[31-Oct-2011 10:46:39] <rmatte> Ve2dmn: fpaste.org the results
[31-Oct-2011 10:46:44] <Ve2dmn> And paste to pastebin?
[31-Oct-2011 10:46:52] <rmatte> use fpaste.org, it's simpler
[31-Oct-2011 10:47:47] <rmatte> SDuensin: it's really very simple... each component has a python class. Templates have a target class value which coincides with this, i.e Products.ZenModel.IpInterface
[31-Oct-2011 10:48:07] <rmatte> So when the components are mapped to that class, they know to use templates with that target class assigned to them
[31-Oct-2011 10:48:37] <rmatte> when the template is used the proper snmp suffix is added when performing the polling for each component
[31-Oct-2011 10:49:02] <SDuensin> But how can it know the suffix if I'm not using "standard" OIDs?
[31-Oct-2011 10:49:03] <rmatte> so if an interface has an index of 28, and ifInErrors is 1.3.6.1.2.1.2.2.1.14, the end OID is 1.3.6.1.2.1.2.2.1.14.28 for that component
[31-Oct-2011 10:49:55] <rmatte> That's the thing, if you're not generating OIDs with matching index values, then you would need to create a whole new component table and a whole new python class specifically for what you're trying to collect
[31-Oct-2011 10:49:55] <Ve2dmn> all the same to me
[31-Oct-2011 10:50:15] <rmatte> you can only put the graphs under an existing section if the index values match up with the standard ones
[31-Oct-2011 10:51:00] <SDuensin> rmatte, might I just mention that I HATE WINDOWS.
[31-Oct-2011 10:51:18] <Ve2dmn> http://fpaste.org/NFFn/
[31-Oct-2011 10:51:20] <rmatte> So you're installing net-snmp in windows?
[31-Oct-2011 10:51:25] <SDuensin> yup
[31-Oct-2011 10:51:30] <rmatte> Why?
[31-Oct-2011 10:52:05] <rmatte> Ve2dmn: 2011-10-31 10:19:20,285 DEBUG zen.zenperfsnmp: Device 192.168.2.230 timed out
[31-Oct-2011 10:52:13] <Ve2dmn> I know
[31-Oct-2011 10:52:14] <SDuensin> Ideally, so I can expand on what I can collect. And all my reading says the MS SNMP stuff is horribly broken.
[31-Oct-2011 10:52:15] <rmatte> Ve2dmn: when you snmpwalk the device does it walk it really slowly?
[31-Oct-2011 10:52:21] <Ve2dmn> nope
[31-Oct-2011 10:52:51] <Ve2dmn> it replies within 0.3sec (according to my tcpsump)
[31-Oct-2011 10:53:17] <rmatte> It sounds like your timeout values are set too low in zProperties
[31-Oct-2011 10:53:22] <rmatte> you should try bumping them up
[31-Oct-2011 10:53:51] <rmatte> zCollectorClientTimeout and zSnmpTimeout
[31-Oct-2011 10:55:42] <Ve2dmn> I've tried that put zSnmpTimeout at 50sec and it doesn't do anything... I didn't try messing with zCollectorClientTimeout et...
[31-Oct-2011 10:55:49] <Ve2dmn> *yet
[31-Oct-2011 10:56:34] <rmatte> zCollectorClientTimeout should be 300
[31-Oct-2011 10:57:03] <rmatte> this output suggests that it's only having problems walking things like ifInErrors
[31-Oct-2011 10:57:12] <rmatte> it's able to walk ifOperStatus fine
[31-Oct-2011 10:57:27] <Ve2dmn> I've bump it up to 600 to see if it does anything...
[31-Oct-2011 10:58:07] <rmatte> try: snmpwalk -v2c -c your-string device-name ifInErrors
[31-Oct-2011 10:58:16] <jb> rmatte: hey.. thanks!
[31-Oct-2011 10:58:31] <rmatte> jb: np
[31-Oct-2011 10:58:44] <jb> i have another on /Change/Set/Status
[31-Oct-2011 10:58:46] <jb> not sure whats causing that one.
[31-Oct-2011 10:59:09] <Ve2dmn> Any reason why everything else works with zCollectorClientTimeout at 180...
[31-Oct-2011 10:59:49] <rmatte> Ve2dmn: I'm only suggesting that as a possible solution, that might not even be the issue
[31-Oct-2011 10:59:57] <rmatte> and 180 is the default, but I personally bump it to 300
[31-Oct-2011 11:00:05] <rmatte> since I monitor some devices that are slow to respond
[31-Oct-2011 11:00:28] <rmatte> like I said, please try to specifically snmpwalk ifInErrors
[31-Oct-2011 11:00:38] <rmatte> and ifInOctets
[31-Oct-2011 11:00:47] <rmatte> see if the device actually returns that data to you
[31-Oct-2011 11:00:57] <rmatte> from what I can see, it's choking when trying to poll those values
[31-Oct-2011 11:01:02] <rmatte> but it's getting ifOperStatus fine
[31-Oct-2011 11:01:30] <Ve2dmn> If I do "Test data source" it works just fine....
[31-Oct-2011 11:01:53] <Ve2dmn> but then, it only request 1 value at a time....
[31-Oct-2011 11:02:16] <Ve2dmn> the zenperfsnmp does several in a single resquest...
[31-Oct-2011 11:02:44] <rmatte> here's something to try... lower zMaxOIDPerRequest from 40 to 10
[31-Oct-2011 11:02:53] <rmatte> then try zenperfsnmp run -v10 -d again
[31-Oct-2011 11:03:18] <Ve2dmn> btw, some of the graphs in /opt/zenoss/perf/Devices/192.168.2.230/ are missing...
[31-Oct-2011 11:03:37] <rmatte> yeh, they will be if it wasn't able to gather the initial data to create RRDs
[31-Oct-2011 11:03:50] <rmatte> they'll be automatically generated when it can
[31-Oct-2011 11:04:11] <jb> rmatte: the error on /Change/Set/Status is "AttributeError: firstTime"
[31-Oct-2011 11:04:15] <jb> let me get the line.
[31-Oct-2011 11:04:43] <rmatte> I don't have much time left here before I need to get back to work
[31-Oct-2011 11:04:47] <jb> cool
[31-Oct-2011 11:04:58] <jb> it says line 474, but there are only like 6 lines in this transform
[31-Oct-2011 11:05:12] <rmatte> it's talking about some line in a library file
[31-Oct-2011 11:05:18] <rmatte> not in your transform
[31-Oct-2011 11:05:22] <jb> ah
[31-Oct-2011 11:05:33] <rmatte> don't focus on that, focus on the error itself
[31-Oct-2011 11:05:34] <Sam-I-Am> morning
[31-Oct-2011 11:05:42] <rmatte> do you have the fill traceback?
[31-Oct-2011 11:05:44] <rmatte> full*
[31-Oct-2011 11:06:10] <jb> i'm not seeing where I can get that.. I see the
[31-Oct-2011 11:06:13] <jb> "transform" details
[31-Oct-2011 11:06:30] <rmatte> paste the transform so I can review it
[31-Oct-2011 11:06:58] <jb> if evt.firstTime == evt.lastTime and hasattr(evt, 'compClass'):
[31-Oct-2011 11:06:59] <jb> for comp in device.getMonitoredComponents(type=evt.compClass):
[31-Oct-2011 11:06:59] <jb> if comp.id != evt.component: continue
[31-Oct-2011 11:06:59] <jb> comp.status = int(evt.compStatus)
[31-Oct-2011 11:06:59] <jb> from transaction import commit
[31-Oct-2011 11:06:59] <jb> commit()
[31-Oct-2011 11:07:00] <jb> break
[31-Oct-2011 11:07:01] <jb> crap
[31-Oct-2011 11:07:04] <rmatte>
[31-Oct-2011 11:07:05] <jb> i meant to paste the URL.
[31-Oct-2011 11:07:06] <Ve2dmn> brb
[31-Oct-2011 11:07:12] <rmatte> k well, good enough
[31-Oct-2011 11:10:44] <rmatte> where are these compClass and compStatus values coming from?
[31-Oct-2011 11:10:59] <Ve2dmn> well, well.... it worked....
[31-Oct-2011 11:11:17] <jb> rmatte: i have no idea.
[31-Oct-2011 11:11:21] <Ve2dmn> not 100% sure why it didn't before...
[31-Oct-2011 11:11:22] <jb> rmatte: maybe from egor's dellmonitor?
[31-Oct-2011 11:11:45] <rmatte> Ve2dmn: did you lower zMaxOIDPerRequest or did it just magically decide to start working?
[31-Oct-2011 11:11:54] <Ve2dmn> lower zMaxOIDPerRequest
[31-Oct-2011 11:12:08] <rmatte> jb: I've never seen a ZenPack actually add properties to events before
[31-Oct-2011 11:12:17] <rmatte> jb: I use the dell mon pack and it didn't do that to my install
[31-Oct-2011 11:12:28] <jb> k, well thats about the only 3rd party pack I have installed.
[31-Oct-2011 11:12:44] <jb> maybe an enteprise pack did it.
[31-Oct-2011 11:12:46] <rmatte> Ve2dmn: some devices are sensitive regarding the rate at which SNMP is polled at
[31-Oct-2011 11:12:58] <rmatte> Zenoss polls at an extremely high rate, some devices can't handle it
[31-Oct-2011 11:13:06] <Ve2dmn> It worked before, that's what even more weird....
[31-Oct-2011 11:13:09] <rmatte> Fortigate firewalls and Cisco ASAs almost always have that problem
[31-Oct-2011 11:13:21] <rmatte> well, maybe it's sitting behind one of those types of firewalls now?
[31-Oct-2011 11:13:49] <Ve2dmn> but the monitor was plugged in before... now it's 1 hop away...
[31-Oct-2011 11:13:54] <rmatte> It's not always the device itself causing the issue, it can be a device that the traffic is passing through along the way
[31-Oct-2011 11:14:19] <rmatte> either way, there you go
[31-Oct-2011 11:14:33] <rmatte> jb: probably an enterprise pack, yeh
[31-Oct-2011 11:14:56] <Ve2dmn> great. I can now find who's been using all the bandwith latetly
[31-Oct-2011 11:15:32] <rmatte> jb: The only thing I can think of is that evt.firstTime and evt.lastTime are written after the transform is executed so they don't exist during execution (like the count attribute)
[31-Oct-2011 11:15:40] <rmatte> jb: which is why you can't use them in your transform
[31-Oct-2011 11:16:17] <rocket> most likely you have to look up the event to pull in those variables
[31-Oct-2011 11:16:51] <rmatte> rocket: which can only be done if the event already exists in the database, no?
[31-Oct-2011 11:17:00] <rocket> correct
[31-Oct-2011 11:17:22] <rmatte> so what he's trying to do is essentially impossible... he's basically trying to do the equivalent of: if evt.count == 1:
[31-Oct-2011 11:17:25] <rmatte> which also doesn't work
[31-Oct-2011 11:17:47] <rocket> but if you cant find an existing event you know your the first event etc
[31-Oct-2011 11:17:55] <rmatte> true
[31-Oct-2011 11:18:18] <rmatte> too bad there isn't some built in system for doing that, gets a bit intensive to do that each time the transform is run
[31-Oct-2011 11:18:22] <rocket> commits in a transform are bad too
[31-Oct-2011 11:18:38] <rmatte> I thought txnCommit() was the preferred method in a transform?
[31-Oct-2011 11:19:11] <rocket> really you should import transact and wrap a method using the transact decorator
[31-Oct-2011 11:19:31] <rocket> then it will retry etc and not lock things up if there are database conflict errors etc
[31-Oct-2011 11:19:39] <jb> well, first of all, this is not my transform
[31-Oct-2011 11:19:50] <jb> it's either from egor's dell pack or from an enterprise zenpack :/
[31-Oct-2011 11:20:13] <rmatte> jb: so we've determined that if evt.firstTime == evt.lastTime will not work, and that commit() in a transform is a bad idea
[31-Oct-2011 11:20:28] <jb> rocket: can you check and see if that's a stock transform in 4.1.0?
[31-Oct-2011 11:20:30] <rmatte> If it's from either of those I'd be flabergasted
[31-Oct-2011 11:21:14] <rmatte> anyways, time for me to get back to my work, cheers
[31-Oct-2011 11:21:32] <jb> i don't have anything else installed on this instance, other than some performance template-only zenpacks that are my own.
[31-Oct-2011 11:26:42] <rocket> jb if I get a chance I can .. but I am fairly busy this morning
[31-Oct-2011 11:27:24] <jb> k
[31-Oct-2011 11:29:43] <jb> anyone else here running 4?
[31-Oct-2011 11:35:22] <gwb2351> a little?
[31-Oct-2011 11:35:37] <gwb2351> only as a "tech preview"
[31-Oct-2011 11:43:03] <Hackman238> jb: Yep
[31-Oct-2011 11:43:07] <Hackman238> jb: Production.
[31-Oct-2011 12:00:01] [disconnected at Mon Oct 31 12:00:01 2011]
[31-Oct-2011 12:00:01] [connected at Mon Oct 31 12:00:01 2011]
[31-Oct-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]
[31-Oct-2011 12:03:40] <jb> Hackman238: can you check for a transform at /Change/Set/Status ?
[31-Oct-2011 12:04:36] <Hackman238> jb: I have one, but I put it there
[31-Oct-2011 12:04:55] <jb> hrm, i have no idea where this one came from, then.
[31-Oct-2011 12:05:27] <Hackman238> jb: whats your transform say?
[31-Oct-2011 12:05:38] <jb> http://pastebin.com/CSXCeR5V
[31-Oct-2011 12:05:44] <jb> its causing a transform error
[31-Oct-2011 12:07:15] <Hackman238> jb: Oh wow. Yeah thats not an out of box transform
[31-Oct-2011 12:07:31] <jb> i thought it may have been from egor's DellMon pack.
[31-Oct-2011 12:07:40] <jb> but, rmatte says he doesn't have it
[31-Oct-2011 12:10:08] <Hackman238> jb: grep "compClass" $ZENHOME/ZenPacks/ -Rf
[31-Oct-2011 12:10:27] <Hackman238> jb: That'll look through all the ZenPack source for anything that would insert the transform
[31-Oct-2011 12:11:02] <jb> hrm, egor's deviceAdvDetail pack.
[31-Oct-2011 12:11:40] <Hackman238> jb Ah
[31-Oct-2011 12:11:48] <jb> i'll ask him about it
[31-Oct-2011 12:14:56] <rocket> a commit inside a transform typically is a bad idea btw
[31-Oct-2011 12:16:31] <jb> i just deleted it for now.
[31-Oct-2011 12:56:27] <kenhen93> has anyone made a succesful parser using componentcommandparser?
[31-Oct-2011 12:57:41] <rocket> yes .. I *think* the aix zenpack is using that if I recall correctly
[31-Oct-2011 12:58:52] <kenhen93> i feel like I am very close to getting mine working
[31-Oct-2011 12:59:05] <kenhen93> but i don't quit understand the componentsplit
[31-Oct-2011 13:00:00] <kenhen93> why did you use ^$ in your parser message/36229#36229
[31-Oct-2011 14:31:43] <jb> rmatte: here?
[31-Oct-2011 15:14:37] <kenhen93> I think I found a potential bug, how would I open a ticket to get someone to look at it
[31-Oct-2011 15:15:46] <kenhen93> I finally got my parser to work by commenting out some zenoss code in ComponentCommandParser.py
[31-Oct-2011 15:16:37] <jb> kenhen93:
[31-Oct-2011 15:16:41] <jb> http://dev.zenoss.org/trac/
[31-Oct-2011 15:17:07] <kenhen93> thanks
[31-Oct-2011 15:18:43] <jb> np
[31-Oct-2011 15:21:24] <kenhen93> do i need my own account or can i log it as zenoss?
[31-Oct-2011 15:22:24] <jmp242> yse the zenoss one
[31-Oct-2011 15:22:35] <jmp242> unless you want to take time to set up your own account
[31-Oct-2011 15:23:25] <kenhen93> maybe i will if this actually turns out to be a bug :-)
[31-Oct-2011 15:32:58] <jmp242> rocket: you there? Does Zenoss 3.x or 4.x support MySQL 5.5 yet?
[31-Oct-2011 15:33:44] <rocket> 4.x uses a rebranded 5.5 I believe
[31-Oct-2011 15:36:08] <jmp242> Ahh ok
[31-Oct-2011 15:36:11] <jmp242> thanks
[31-Oct-2011 16:41:25] <aclark> so, i've got some custom graphs to create consisting of anywhere from 2 to 5 or 6 data points over time… should I do this TTW or can a create a zenpack for this?
[31-Oct-2011 16:49:55] <aclark> looks like i need a new monitoring template
[31-Oct-2011 16:50:06] <aclark> and the data comes from a shell command, hope that's OK
[31-Oct-2011 16:54:44] <aclark> Jane_Curry!
[31-Oct-2011 17:00:47] <diatonic> Really strange, trying to update IPs on some DHCP devices but Zenoss is still showing old IPs. Names resolve to correct IP from shell :-/
[31-Oct-2011 17:00:53] <diatonic> 3.2.1
[31-Oct-2011 17:03:09] kocolosk is now known as kocolosk|away
[31-Oct-2011 17:27:41] <dhopp> diatonic: did you remodel?
[31-Oct-2011 17:28:13] <dhopp> diatonic: and you are talking about the manage IP or the ips underneath interfaces?
[31-Oct-2011 17:36:36] <diatonic> dhopp, Just went to device page, clicked the little "gear" icon at the bottom and chose "Reset/Change IP" - Says to leave blank to update using DNS
[31-Oct-2011 17:36:42] <diatonic> but keeps old IP
[31-Oct-2011 17:37:04] <dhopp> but after you did the reset of the IP did you do a remodel?
[31-Oct-2011 17:37:11] <diatonic> nope
[31-Oct-2011 17:37:22] <dhopp> push the gear icon and to "Push Changes"
[31-Oct-2011 17:37:31] <dhopp> and then the gear icon again and do "Remodel"
[31-Oct-2011 17:42:28] <diatonic> dhopp, thanks.
[31-Oct-2011 17:46:21] <dhopp> rmatte you there?
[31-Oct-2011 19:47:51] kocolosk|away is now known as kocolosk
[01-Nov-2011 00:00:01] [disconnected at Tue Nov 1 00:00:01 2011]
</pre></html>
