Skip navigation
Currently Being Moderated

IRC November 2011

Created on: Dec 12, 2011 5:08 PM by Nick Yeates - Last Modified:  Dec 12, 2011 5:14 PM by Nick Yeates

[01-Nov-2011 00:00:02] [connected at Tue Nov  1 00:00:02 2011]

[01-Nov-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]

[01-Nov-2011 02:06:03] <zaoul1> hi

[01-Nov-2011 02:09:38] <zaoul1> Anyone thought about interface thresholds based on cisco service policy thresholds? Maybe possible using snmp informs?

[01-Nov-2011 06:57:04] <zaoul1> ?

[01-Nov-2011 08:35:09] <jmp242> morning all

[01-Nov-2011 08:37:20] kocolosk is now known as kocolosk|away

[01-Nov-2011 08:40:45] <zaoul1> hey

[01-Nov-2011 08:41:24] <zaoul1> Anyone know what I can do about multiple devices that share common private networks but are actually physically independent?

[01-Nov-2011 08:43:03] <jmp242> umm, I'm not sure I understand

[01-Nov-2011 08:43:11] <jmp242> what's the problem? that there are common networks?

[01-Nov-2011 09:53:33] <dhopp> if I'm monitoring a process with a regex that includes the process name and part of the options, what happens in Zenoss if the process is restarted with a slightly different options but my regex will still find it?

[01-Nov-2011 09:53:43] <dhopp> does it create a new process to monitor?

[01-Nov-2011 10:13:18] <Gat0rvean> could anyone spare a minute to help me setup an event, to clear another event

[01-Nov-2011 10:15:46] <Gat0rvean> I'm using traps from this particular device, and I have them mapped to the same class (/Net/Link), I've set one as the "Set" with the zEventSeverity as Error, and the "Clear" trap with the zEventSeverity as "Clear"; I've also set the zEventClearClasses on the "Set" event, to the /Net/Link/ClearEvent, but the only thing that happens is the Set event gets triggered, and the "Clear" event gets sent, but does nothing

[01-Nov-2011 10:28:52] <jmp242> dhopp: I think it might

[01-Nov-2011 10:29:15] <jmp242> process monitoring has always been less deterministic than  I would like on Zenoss through 3.1 anyway

[01-Nov-2011 10:32:34] <dhopp> jmp242:  in my quick little test it doesn't seem to…however, the options that I changed don't show up in the "name" field when looking at it in zenoss

[01-Nov-2011 10:32:48] <dhopp> but they do show up in a ps aux

[01-Nov-2011 10:35:44] <jmp242> well zenoss isn't reading the processes

[01-Nov-2011 10:35:47] <jmp242> via ps aux

[01-Nov-2011 10:35:52] <dhopp> jmp242:  I know that

[01-Nov-2011 10:35:56] <jmp242> but via whatever snmp shows in that OID

[01-Nov-2011 10:36:31] <jmp242> are you using the ignore arguments in Zenoss

[01-Nov-2011 10:36:36] <jmp242> but using your regex to match some args?

[01-Nov-2011 10:36:46] <dhopp> no because I need to look at arguments to find this process

[01-Nov-2011 10:37:15] <jmp242> mmm

[01-Nov-2011 10:37:45] <jmp242> I've noticed that Zenoss seems to (for me) change processes with ignore args *OFF* with *any* argument change

[01-Nov-2011 10:38:04] <jmp242> though I haven't been doing advanced regexes either

[01-Nov-2011 10:39:35] <dhopp> ok..I just verified it will make a new process if the argument that changed is in the "name" field in zenoss…but if the argument that changes does not fit in whatever limit that is then it will not

[01-Nov-2011 10:40:25] <dhopp> on our production systems we might have 4 tomcat instances running…all started with the process name of jsvc.exec and then started with arguments that signify which tomcat it is

[01-Nov-2011 10:40:47] <dhopp> so I have to use arguments so I can catch the tomcat instance and have it show up as 4 different instances when monitoring

[01-Nov-2011 10:55:05] <rocket> dhopp: in snmp the arguement list is truncated to 255 chars

[01-Nov-2011 10:55:25] <rocket> dhopp: so typically you need the unique items first if possible

[01-Nov-2011 10:56:28] <rocket> I believe there is a bug in 3.1 where the args are not ignored properly.

[01-Nov-2011 10:56:55] <dhopp> rocket:  well I have one argument that may change (it's rare and it's because somebody manually changed it)… the argument is -wait followed by a number of seconds to wait (so -wait 10)…

[01-Nov-2011 10:56:57] <dhopp> I'm using 3.2

[01-Nov-2011 10:57:49] kocolosk|away is now known as kocolosk

[01-Nov-2011 10:58:01] <rocket> dhopp: I would have to look it up but I have a meeting in 3mins

[01-Nov-2011 10:58:47] <dhopp> rocket:  well I don't want to ignore arguments in this case

[01-Nov-2011 10:58:58] <dhopp> so I don't think I would be hurt by that bug anyway

[01-Nov-2011 10:59:18] <dhopp> I guess I'll just have to live with the possibility of the wait changing...

[01-Nov-2011 11:00:03] <dhopp> I was hoping I could do a regex of something that included -wait (\d+) to tell it I expect any number here..but that seems to break catching the process altogether

[01-Nov-2011 11:09:18] <rocket> your regex is wrong

[01-Nov-2011 11:09:35] <rocket> the -wait would have to be optional too unless that is always there

[01-Nov-2011 11:09:40] <rocket> so maybe it wasnt wrong

[01-Nov-2011 11:09:57] <dhopp> it is always there

[01-Nov-2011 11:10:40] <dhopp> rocket:  this is what I have that works: ^[^ ]*jsvc\.exec*(tomcat)*

[01-Nov-2011 11:10:52] <dhopp> (I'm not sure why I have to have tomcat in () but if I don't it doesn't work either)

[01-Nov-2011 11:11:40] <dhopp> and this is the process that it is matching (snipped the arguments for clarity)

[01-Nov-2011 11:11:45] <dhopp> jsvc.exec -home /usr/local/java -user CLI-AppUser -wait 10 -outfile /usr/local/tomcat-01/logs/catalina.out -errfile /usr/local/tomcat-01/logs/catalina.err -pidfile /var/run/

[01-Nov-2011 11:40:41] <dhopp> rocket:  actually…does the regex even try to match the arguments or does it strictly look at the process name but then use the argument list to tell if it's a different process or not?

[01-Nov-2011 12:00:01] [disconnected at Tue Nov  1 12:00:01 2011]

[01-Nov-2011 12:00:02] [connected at Tue Nov  1 12:00:02 2011]

[01-Nov-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]

[01-Nov-2011 12:00:59] <dhopp> rocket: nevermind…I really shouldn't do semi complex regex's without caffiene

[01-Nov-2011 12:01:25] <dhopp> I just looked at my regex and realized it didn't do anything that I thought it did

[01-Nov-2011 12:01:26] <dhopp> lol

[01-Nov-2011 12:05:03] <rocket> dhopp: to be honest I would have to look at the code again.  Its been a couple of months.

[01-Nov-2011 12:05:41] <dhopp> rocket: I did look at the code and it looks like the process name gets set as a catenation of 2 strings (the second being the arguments) if ignore arguments is false

[01-Nov-2011 12:05:46] <dhopp> or no

[01-Nov-2011 12:05:50] <dhopp> whatever the setting is

[01-Nov-2011 12:06:47] <fragfutter> dhopp: sounds about right from what i remember

[01-Nov-2011 13:16:47] <subbu> Hi all We are unable to add TCP IP ports to monitoring in most of the Linux Devices

[01-Nov-2011 13:17:17] <subbu> We  already tried by  adding the port manually from  the global IP services  list , after adding we model the device and  the port went missing .

[01-Nov-2011 13:17:26] <subbu> Can any one please help me out in this

[01-Nov-2011 14:21:08] <SDuensin> So has anyone _ever_ made this work with WMI and machines in a Windows Domain?

[01-Nov-2011 14:34:50] <mducharme-work> sduensin - we are using local accounts on the systems for zenoss currently, but that's because we didn't have active directory before when we set them up

[01-Nov-2011 14:35:04] <mducharme-work> the wmi is still working now that we have our domain, zenoss is still using the locally created accounts for WMI

[01-Nov-2011 14:35:49] <SDuensin> mducharme-work, man, I hate to create accounts on 300+ machines.  Guess it's a solution though.

[01-Nov-2011 14:36:47] <jmp242> SDuensin: I use a single "service account" that is a domain admin

[01-Nov-2011 14:36:53] <jmp242> works fine for me

[01-Nov-2011 14:37:00] <jmp242> Core 3.2.1

[01-Nov-2011 14:37:07] <SDuensin> jmp242, that's kind of what I have here.  Except for the working part.  :-)

[01-Nov-2011 14:37:08] <jmp242> Server 2008 R2

[01-Nov-2011 14:37:43] <ablew> anybody experienced a memory leak with zentrap and zenoss-3.2.1 core?

[01-Nov-2011 14:38:31] <jmp242> you using the CIM / egors pack?

[01-Nov-2011 14:39:05] <SDuensin> jmp242, me?

[01-Nov-2011 14:41:47] <ablew> jmp242: pretty sure i'm not

[01-Nov-2011 14:49:17] <jmp242> that was to SDuensin

[01-Nov-2011 14:51:58] <SDuensin> I can't get WMI to work at all now.  This has gone from bad to worse.

[01-Nov-2011 14:52:14] <SDuensin> It says "User credentials cannot be used for local connections"

[01-Nov-2011 14:54:34] <jmp242> never seen that

[01-Nov-2011 14:54:59] <SDuensin> I'm good at finding new problems.  :-)

[01-Nov-2011 15:00:25] <SDuensin> How about "failed NT status (c0000022) in dcerpc_pipe_connect_b_recv"?

[01-Nov-2011 15:11:08] <jmp242> sounds like it doesn't like something

[01-Nov-2011 15:11:16] <jmp242> but these seem new to me

[01-Nov-2011 15:11:30] <jmp242> And these are domain admin accounts, that have login rights to the servers

[01-Nov-2011 15:11:31] <dhopp> SDuensin: are you trying to monitor WMI through a NAT?

[01-Nov-2011 15:12:19] <SDuensin> dhopp - oh man.  That may actually be the case.  I'll have to check with my network dorks.

[01-Nov-2011 15:12:25] * SDuensin smacks himself in the head.

[01-Nov-2011 15:13:52] <dhopp> WMI uses DCOM and DCOM needs access to the real IP of the server…NATs don't work

[01-Nov-2011 15:14:40] <SDuensin> We've got a hodge podge here.  Some of our clients are NATted and some are routed.  Guess which one my test network is?

[01-Nov-2011 15:14:45] * SDuensin feels like an idiot.

[01-Nov-2011 15:15:08] <dhopp> SDuensin:  no took me a few hours to figure that problem out myself

[01-Nov-2011 15:19:01] <johnnynoc> i'm having problem with a zenpack i'm hoping someone can help with

[01-Nov-2011 15:19:19] <johnnynoc> i've removed the zenpack, but it left some artifacts that are causing some serious problems

[01-Nov-2011 15:19:43] <johnnynoc> i've tried to remove every reference to the zenpack but still can't seem to figure out how to remove what appears to eb a problematic device

[01-Nov-2011 15:20:12] <johnnynoc> when i run zenchkrels -r -x1 it dies with this error:  AttributeError: 'IMMDevice' object has no attribute 'getPrimaryPath'

[01-Nov-2011 15:20:26] <johnnynoc> and that's when it's checking relationships on the monitor

[01-Nov-2011 15:20:27] <johnnynoc> INFO:zen.Relations:checking relations on object /zport/dmd/Monitors/Performance INFO:zen.Relations:checking relations on object /zport/dmd/Monitors/Performance/localhost

[01-Nov-2011 15:20:53] <johnnynoc> when i look at that monitor i don't see any of these IMMDevice's - is there something else i could be looking at?

[01-Nov-2011 15:21:18] <johnnynoc> and when i poke around in the zope interface i still can't seem to find any mention of this

[01-Nov-2011 15:21:31] <johnnynoc> i tried using and it gives me this error: AttributeError: 'IMMDevice' object has no attribute '__of__'

[01-Nov-2011 15:21:45] <johnnynoc>   File "./", line 118, in fixMonitorRelations

[01-Nov-2011 15:21:57] <johnnynoc> i can't seem to delete the localhost monitor

[01-Nov-2011 15:24:49] <johnnynoc> :'(

[01-Nov-2011 15:33:55] <dhopp> johnnynoc:  what zenpack is this from?

[01-Nov-2011 15:44:29] <johnnynoc> dhopp it's from an IBM IMM Device zenpack i'm helping them with

[01-Nov-2011 15:44:36] <johnnynoc> fortuntely this is all on my dev instance of zenoss

[01-Nov-2011 15:44:42] <johnnynoc> but i'd still like to figure out how to get myself out of this problem

[01-Nov-2011 15:44:54] <johnnynoc> at this point, it seems zenoss is looking for a device that doesn't exist

[01-Nov-2011 15:45:00] <johnnynoc> but i can't seem to get everything back in working order

[01-Nov-2011 15:45:23] <johnnynoc> something abou tit's relationship with the localhost monitor

[01-Nov-2011 15:45:52] <johnnynoc> is there any way i can remove the localhost monitor and re-add it?

[01-Nov-2011 15:46:10] <johnnynoc> or something specific i can do in zend or zope to try fix this relationship?

[01-Nov-2011 15:49:26] <johnnynoc> and zenchkrels -x -r1 have been of no help thus far

[01-Nov-2011 15:49:26] <johnnynoc> :/

[01-Nov-2011 15:56:34] <SDuensin> dhopp, ok, we've made sure there's no firewalls, NAT, etc.  I'm still getting:

[01-Nov-2011 15:56:52] <dhopp> SDuensin:  I'm not sure then...

[01-Nov-2011 15:56:59] <SDuensin> I've tried my domain admin account and a local account with admin rights.

[01-Nov-2011 17:05:35] <cluther> Does anyone out there have a Solaris host running LDOMs?

[01-Nov-2011 17:07:12] <rmatte> cluther: unfortunately not.

[01-Nov-2011 18:22:26] <nexex> anyone tried ext4 with zenoss?

[01-Nov-2011 18:48:02] <isotope> is there an updated doc for how to use a nagios plugin with in zenoss 3.x?

[01-Nov-2011 18:48:18] <isotope> This is the only guide I can find and its pretty old

[01-Nov-2011 18:48:19] <isotope>

[01-Nov-2011 20:16:24] kocolosk is now known as kocolosk|away

[01-Nov-2011 20:25:03] kocolosk|away is now known as kocolosk

[01-Nov-2011 21:32:03] sfsdsddss is now known as isotope

[02-Nov-2011 00:00:01] [disconnected at Wed Nov  2 00:00:01 2011]

[02-Nov-2011 00:00:03] [connected at Wed Nov  2 00:00:03 2011]

[02-Nov-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]

[02-Nov-2011 00:55:46] <zaoul1> what I can do about multiple devices that share common private networks but are actually physically independent? That is; I have two PUBLIC nodes, that BOTH have PRIVATE - PHYSICALLY seperate networks... yet zenoss will link my nodes together through this private network.

[02-Nov-2011 06:49:27] <zaoul> this is bad

[02-Nov-2011 08:38:40] kocolosk is now known as kocolosk|away

[02-Nov-2011 09:17:11] kocolosk|away is now known as kocolosk

[02-Nov-2011 09:52:52] kocolosk is now known as kocolosk|away

[02-Nov-2011 10:42:46] kocolosk|away is now known as kocolosk

[02-Nov-2011 10:43:42] kocolosk is now known as kocolosk|away

[02-Nov-2011 11:33:02] kocolosk|away is now known as kocolosk

[02-Nov-2011 11:48:28] <subbu259> Hi All .. I would like to trigger an alert from zenoss if memory utilization reaches to 95%. I am monitoring this server by WMI

[02-Nov-2011 11:48:44] <subbu259> Can any one please provide me the Threshold expression for the same

[02-Nov-2011 11:57:07] <jb> subbu259: you just need to add a "MinMax" threshold to the data-source.

[02-Nov-2011 11:57:35] <jb> leave the "min" value empty, and put just use the "max" value to specify 95%.

[02-Nov-2011 11:58:33] <subbu259> @jb:I used the datapoint as MemoryAvialbleKBytes_memory AvailBytes

[02-Nov-2011 11:58:45] <subbu259> will it works if i give 95%

[02-Nov-2011 11:59:06] <subbu259> nd gave the transform as /Perf/Memory

[02-Nov-2011 12:00:01] [disconnected at Wed Nov  2 12:00:01 2011]

[02-Nov-2011 12:00:01] [connected at Wed Nov  2 12:00:01 2011]

[02-Nov-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]

[02-Nov-2011 12:01:24] <subbu259> @jb:Can you please help me out in this

[02-Nov-2011 12:48:38] <subbu046> hi all can any one please help me how to create a thresold for the memory utilization .as i am monitoring this server through WMI

[02-Nov-2011 12:52:45] <subbu046> @Jmp242:Can you please help me out in this requeswt

[02-Nov-2011 13:03:34] <subbu046> u r help is much appreciated

[02-Nov-2011 13:20:52] <subbu441> hi all can any one please provide me the threshold for Memory usage if exceeds 95%

[02-Nov-2011 13:21:05] <subbu441> threshold Expression

[02-Nov-2011 13:53:47] <johnnynoc> who wants to help me try to delete a phantom device?

[02-Nov-2011 13:53:48] <johnnynoc>

[02-Nov-2011 13:54:50] <johnnynoc> zenchkrels -x -r1, both die when i run them

[02-Nov-2011 13:55:02] <johnnynoc> and the device isn't in zope, but is still somehow attached to my localhost performance monitor

[02-Nov-2011 13:55:05] <johnnynoc> (at least that's what it seems like)

[02-Nov-2011 13:55:41] <subbu441> hey can any one help me with the memory threshold expression

[02-Nov-2011 13:56:47] <subbu441> @johnnynoc:Can you please help me out in this

[02-Nov-2011 13:57:57] <johnnynoc> subbu441: i can try but i don't know your question/problem

[02-Nov-2011 13:58:02] <johnnynoc> i need help too man!

[02-Nov-2011 13:58:03] <johnnynoc>

[02-Nov-2011 13:58:53] <subbu441> i want to create a trigger if the memory utilization is more than 95% .Can you please give me the threshold expression for the same

[02-Nov-2011 13:59:01] <subbu441> i am monitoring it with WMI

[02-Nov-2011 13:59:57] <johnnynoc> are you using bigegor's wmi zenpacks?

[02-Nov-2011 14:00:20] <subbu441> no

[02-Nov-2011 14:00:34] <subbu441> i am using the default wmi zenpacks which came up with Zenoss Ent

[02-Nov-2011 14:01:18] <johnnynoc> ok

[02-Nov-2011 14:01:25] <johnnynoc> i use zenoss core, so i'm not sure what you're dealing with exactly

[02-Nov-2011 14:01:37] <johnnynoc> but from bigegor's zenpack i can shwo you the datasource he's using to monitor the memory

[02-Nov-2011 14:01:47] <johnnynoc> SELECT AvailableKBytes, CacheBytes, PagesPerSec FROM Win32_PerfRawData_PerfOS_Memory

[02-Nov-2011 14:02:08] <johnnynoc> and he puts a threshold on the pagespersec

[02-Nov-2011 14:02:24] <subbu441> i want to know the datapoint used in the threshold section

[02-Nov-2011 14:03:24] <talntid>  Any contractors here? I'm looking to pay someone to setup ZenOSS in my environment.

[02-Nov-2011 14:03:52] <johnnynoc> subbu441: it's pagespersec

[02-Nov-2011 14:04:04] <johnnynoc> availablekbytes, cachebytes, and pagespersec are the datapoints being used

[02-Nov-2011 14:04:10] <johnnynoc> and in the threshold, he's using pagespersec

[02-Nov-2011 14:04:24] <subbu441> @johnnynoc:I am using the memoryavailKbytes_Memoryavilable bytes

[02-Nov-2011 14:04:41] <johnnynoc> ok, if that's how you want to do it that's fine

[02-Nov-2011 14:04:46] <johnnynoc> you just need to use the correct datapoint in your threshold

[02-Nov-2011 14:04:50] <johnnynoc> not sure what the confusion there is about

[02-Nov-2011 14:05:06] <subbu441> i gave the threshold value as 131072

[02-Nov-2011 14:05:16] <subbu441> i useed in the minimum

[02-Nov-2011 14:05:51] <subbu441> if i put 95 in the Max will it trigger

[02-Nov-2011 14:05:59] <subbu441> if it reaches to 95 %

[02-Nov-2011 14:07:46] <subbu441> @Johnnynoc:Can you please help me with this

[02-Nov-2011 14:09:07] <johnnynoc> ok, i understand a little better now

[02-Nov-2011 14:11:06] <subbu441> oh can u please help me now

[02-Nov-2011 14:11:10] <subbu441> how to figure out this

[02-Nov-2011 14:11:40] <subbu441> how to fix this

[02-Nov-2011 14:13:38] <johnnynoc> hmm

[02-Nov-2011 14:13:42] <johnnynoc> let me look at a couple things

[02-Nov-2011 14:13:52] <johnnynoc> as i said, my memory threshold is based on pagespersec which doesn't require anything fancy

[02-Nov-2011 14:14:44] <subbu441> k

[02-Nov-2011 14:16:22] <subbu441> is there any zenoss Support guys available on the chat

[02-Nov-2011 14:16:42] <subbu441> I am using zenoss ENT

[02-Nov-2011 14:16:49] <subbu441> so need the help immediately

[02-Nov-2011 14:17:01] <johnnynoc> well, can you open a support ticket with them?

[02-Nov-2011 14:17:16] <johnnynoc> i'm sorry but i don't have the time to figure out what here.<something> expression you need for the available memory

[02-Nov-2011 14:17:17] <subbu441> i am already working on the other issue

[02-Nov-2011 14:17:24] <subbu441> yeah

[02-Nov-2011 14:17:30] <johnnynoc> our thresholds are easy with regards to memory, we don't care about the available memory only if we're paging/swapping

[02-Nov-2011 14:17:39] <johnnynoc> :/

[02-Nov-2011 14:17:41] <johnnynoc> sorry

[02-Nov-2011 14:17:55] <subbu441> hh

[02-Nov-2011 14:39:53] <SDuensin> Ok, I have a _big_ problem here that I hope is just a misunderstanding.

[02-Nov-2011 14:40:29] <SDuensin> My /home/zenoss folder now shows almost every file being owned by user "1337" and there's a ".smirc" file.

[02-Nov-2011 14:41:32] <SDuensin> This box is fairly isolated from the outside world.  Any chance the CentOS 32 bit stack installer created these files?

[02-Nov-2011 14:43:04] <pipeline> no.

[02-Nov-2011 14:43:10] <johnnynoc> i seriously doubt it

[02-Nov-2011 14:43:13] <pipeline> disconnect network ASAP

[02-Nov-2011 14:43:16] <pipeline> analyze offline

[02-Nov-2011 14:43:24] <pipeline> you been pwned son

[02-Nov-2011 14:43:26] <johnnynoc> SDuensin: sounds like you've been 'owned'

[02-Nov-2011 14:43:29] <SDuensin> pipeline, we're checking our firewall logs.  Nothing going on outbound from that box.

[02-Nov-2011 14:43:36] <SDuensin> johnnynoc, that was my thought.

[02-Nov-2011 14:46:09] <SDuensin> We originally installed it via RPM - something there created it.  No idea where my cohort got the RPMs though.

[02-Nov-2011 14:46:20] * SDuensin used the stack installer, which looks fine.

[02-Nov-2011 14:50:04] <johnnynoc> that just seems so suspect

[02-Nov-2011 14:50:11] <johnnynoc> but at the same time, so obvious makes me curious

[02-Nov-2011 14:50:12] <johnnynoc>

[02-Nov-2011 14:51:45] <SDuensin> Exactly.

[02-Nov-2011 14:52:02] <SDuensin> I downed the box.  We're starting with a fresh machine.

[02-Nov-2011 14:52:16] <johnnynoc>

[02-Nov-2011 15:09:05] <jmp242> Anyone here have some time to look at some tricky forum posts?

[02-Nov-2011 15:28:55] <rmatte> jmp242: I could take a look

[02-Nov-2011 15:32:14] <jmp242>

[02-Nov-2011 15:32:36] <jmp242> this one looks like a RPN issue to me, but I'm not really sure why only one device and filesystem would be affected

[02-Nov-2011 15:33:11] <jmp242> I mean, actually I'm pretty sure Zenoss doesn't have problems such as that for me

[02-Nov-2011 15:36:53] <rmatte> Do you have any terabyte filesystems under monitoring?

[02-Nov-2011 15:39:00] <rmatte> I just posted to that thread, need to find out if it's an NFS filesystem

[02-Nov-2011 15:39:10] <rmatte> since net-snmp doesn't handle NFS filesystems above 1TB well

[02-Nov-2011 15:39:47] <jmp242> Yea, I think we do

[02-Nov-2011 15:39:56] <rmatte> "net-snmp does not deal with terabyte file systems well if they are NFS mounted - they will have small hrStorageAllocationUnits, and hrStorageSize and hrStorageUsed will wrap, so you may see that the Used space is greater than the total Size.

[02-Nov-2011 15:39:57] <rmatte> There is currently no fix for this in the Net-snmp snmp agent, so the workaround is to monitor the NFS server, not the system that mounts them."

[02-Nov-2011 15:40:15] <jmp242> but now IDK for sure - we have 6 terrabyte systems, but I don't know if the filesystems are that big

[02-Nov-2011 15:40:29] <Hackman238_home> Hey all

[02-Nov-2011 15:40:45] <jmp242> Here's another - I haven't done much with CPU changes in systems but is this a bug?

[02-Nov-2011 15:40:46] <jmp242>

[02-Nov-2011 15:40:46] <Hackman238_home> Sorry I've been aloof. Been battling a migration.

[02-Nov-2011 15:40:54] <jmp242> Hey Hackman238_home

[02-Nov-2011 15:43:31] <rmatte> jmp242: well, for one thing, the fact that the load average went from 8 to 16 has nothing to do with Zenoss, that's a system level thing

[02-Nov-2011 15:43:37] <rmatte> number of processes running or waiting for CPU

[02-Nov-2011 15:44:05] <jmp242> Yea, I'm more interested in the CPU number

[02-Nov-2011 15:44:15] <jmp242> I understand Load has 0 to do with CPUs in the system

[02-Nov-2011 15:44:18] <jmp242> "ever, in the "Components -> Processors" info the old number of CPUs (4) is displayed, instead of the new # (8)."

[02-Nov-2011 15:44:28] <jmp242> I'm wondering why a remodel wouldn't pick this up

[02-Nov-2011 15:46:53] <rmatte> I added a couple of pieces of info to this conversation:

[02-Nov-2011 15:47:02] <rmatte> They are confusing processors with cores

[02-Nov-2011 15:47:05] <rmatte> not the same thing at all

[02-Nov-2011 15:49:31] <jmp242> Solves my confusion lol

[02-Nov-2011 15:49:42] <jmp242> I'm never sure if a tool is going to print cores or processors anymore

[02-Nov-2011 15:49:56] <rmatte> lol

[02-Nov-2011 15:50:07] <rmatte> I just added a couple more lines to it for additional clarification

[02-Nov-2011 15:50:10] <jmp242> To my mind - as far as software cares cores *are* processors

[02-Nov-2011 15:50:19] <jmp242> it's just down to the physical level

[02-Nov-2011 15:50:21] <jmp242> when you have to care

[02-Nov-2011 15:50:32] <jmp242> and really only for replacing or upgrading in my mind

[02-Nov-2011 15:50:51] <rmatte> there, my full reply to that thread seals it off quite nicely I believe

[02-Nov-2011 15:51:11] <rmatte> yeh, the processors section in Zenoss though only shows the actual physical processors

[02-Nov-2011 15:51:26] <jmp242> Here's the trickiest one I've been holding:

[02-Nov-2011 15:51:38] <jmp242> well, one of the more zenoss internals tricky

[02-Nov-2011 15:52:22] <rmatte> That drive space one is almost certainly a net-snmp limitation, I'm going to suggest that they monitor via SSH to overcome it

[02-Nov-2011 15:52:50] <jmp242> Yea - that does make sense to em

[02-Nov-2011 15:52:51] <jmp242> me

[02-Nov-2011 15:53:08] <jmp242> so it sounds I should put that in a best practice or FAQ

[02-Nov-2011 15:54:03] <rmatte> there, finished this one off pretty well too:

[02-Nov-2011 15:54:11] <rmatte> I'll check out that other one you just pasted

[02-Nov-2011 15:57:41] <rmatte> There, just answered

[02-Nov-2011 15:57:49] <rmatte> anything else for me to knock down?

[02-Nov-2011 15:57:50] <rmatte> lol

[02-Nov-2011 15:58:26] <bseklecki> /me passes rmatte the |

[02-Nov-2011 15:58:37] <rmatte> ?

[02-Nov-2011 15:58:41] <bseklecki> take a tote on that

[02-Nov-2011 15:58:47] <rmatte> haha

[02-Nov-2011 15:58:56] <bseklecki> knock it back(R)

[02-Nov-2011 15:59:08] <rmatte> I'd rather have a beer

[02-Nov-2011 15:59:17] <rmatte> |^|:

[02-Nov-2011 16:00:07] <rmatte> damnit, I just found out that 8 reports I ran didn't run properly because I had a = instead of a .= in the code after modifying it to fix something

[02-Nov-2011 16:00:09] <bseklecki> I'm loving my ZenOss over here

[02-Nov-2011 16:00:09] <rmatte> bleh

[02-Nov-2011 16:00:25] <rmatte> time to run them again

[02-Nov-2011 16:00:49] <rmatte> they take forever since they run like 30 different custom coded Zenoss reports and a bunch of reports from our ticketing system

[02-Nov-2011 16:00:55] <jmp242> Two more I guess

[02-Nov-2011 16:00:56] <jmp242>

[02-Nov-2011 16:01:02] <jmp242>

[02-Nov-2011 16:01:15] <jmp242> Thanks for the assist

[02-Nov-2011 16:01:35] <rmatte> np

[02-Nov-2011 16:01:53] <rmatte> hmmm, that catalyst one is interesting, almost sounds like something a reindex() might fix honestly

[02-Nov-2011 16:04:22] <jmp242> want me to suggest that then?

[02-Nov-2011 16:05:55] <rmatte> See my reply to

[02-Nov-2011 16:06:05] <rmatte> I'm not sure what to really suggest to them other than that

[02-Nov-2011 16:06:43] <rmatte> suggest a reindex for the catalyst one as a starting point

[02-Nov-2011 16:07:02] <rmatte> actually wait

[02-Nov-2011 16:07:14] <rmatte> no no, I think I see where they are going wrong, give me a second...

[02-Nov-2011 16:07:26] <rmatte> need to double check one of my instances to make sure

[02-Nov-2011 16:08:51] <jmp242> Well, sounds ok to me. I might pass it along to the mailing list IDK

[02-Nov-2011 16:09:02] <jmp242> re the first one

[02-Nov-2011 16:10:41] <rmatte> ah nope, still think a reindex() is best bet for a starting point on that catalyst interface question

[02-Nov-2011 16:10:57] <rmatte> I'll add a note asking them to try that

[02-Nov-2011 16:13:20] <rmatte> alright, any others?

[02-Nov-2011 16:13:41] <jmp242> nope

[02-Nov-2011 16:13:44] <jmp242> I'm caught up

[02-Nov-2011 16:13:45] <jmp242> thanks

[02-Nov-2011 16:13:46] <rmatte> cool

[02-Nov-2011 16:13:47] <rmatte> np

[02-Nov-2011 16:14:01] <jmp242> Good to have more than just Shane to bug lol

[02-Nov-2011 16:14:04] <rmatte> I like challenging forum posts, not the same old stuff lol

[02-Nov-2011 16:14:28] <jmp242> I like ones I know the answers to lol, though that's maybe 40% - mostly the same old suff.

[02-Nov-2011 16:15:00] <tightwork> Anyone help me with my private network problem?

[02-Nov-2011 16:15:09] <rmatte> private network problem?

[02-Nov-2011 16:16:14] <tightwork> I have public nodes that also have the same private network but are actually physically independent, zenoss shows them connected, either by the subnet ip list and especially the map tool shows them directly connected by this private network.

[02-Nov-2011 16:17:35] <rmatte> All that it does to calculate the links on the map tool is "Are these devices at different sites in the same subnet?  Yes?  Ok, then they must be linked."

[02-Nov-2011 16:17:47] <rmatte> There isn't any sort of complicated calculations done.

[02-Nov-2011 16:18:04] <rmatte> I assume you mean the google maps tool?

[02-Nov-2011 16:18:12] <tightwork> sure in 'theory' they are the same subnet, but its physically independent

[02-Nov-2011 16:18:14] <rmatte> or are you referring to the flash network map?

[02-Nov-2011 16:18:17] <tightwork> flash network map

[02-Nov-2011 16:18:31] <rmatte> ah, that network map is fairly useless, I haven't even looked at it in over a year

[02-Nov-2011 16:18:48] <rmatte> They need to get a replacement for that

[02-Nov-2011 16:19:21] <rmatte> Zenoss is not Layer 2 aware, which is why physical dependence isn't detected

[02-Nov-2011 16:19:52] <tightwork> need someway to say 'physically independent', yes for layer2

[02-Nov-2011 16:20:18] <rmatte> layer 2 dependencies have been a requested feature for years now

[02-Nov-2011 16:20:31] <tightwork> i really dont care too much about scanning through nat, but I do like my network map to work out

[02-Nov-2011 16:20:50] <tightwork> ive already found some stuff that is wtf, orly factor

[02-Nov-2011 16:21:11] <tightwork> cool guys creating vlans off our management interface

[02-Nov-2011 16:23:39] <tightwork> arg

[02-Nov-2011 16:23:53] <rmatte> lol

[02-Nov-2011 16:24:20] <rmatte> If you need a reliable network map, you should find some other tool to use for the mapping

[02-Nov-2011 16:25:01] <rmatte> My personal favourite:

[02-Nov-2011 16:25:06] <rmatte> in terms of free software

[02-Nov-2011 16:26:13] <rmatte> I just discovered rundeck (it was mentioned in Zenoss blog post)

[02-Nov-2011 16:26:25] <rmatte> it's exactly the sort of thing I need, I'll be setting it up on a server soon enough

[02-Nov-2011 16:26:34] <tightwork> could netdisco be incorporated into zenoss if the map is to ever be updated

[02-Nov-2011 16:26:48] <tightwork> rundeck eh..

[02-Nov-2011 16:26:58] <rmatte> rundeck is off topic

[02-Nov-2011 16:27:09] <rmatte> I'm not sure about netdisco integration with Zenoss, you'd have to ask an actual developer

[02-Nov-2011 16:27:33] <rmatte> But you can easily run netdisco in addition to Zenoss, it really just does mapping, not monitoring

[02-Nov-2011 16:28:17] <tightwork> yeah, problem is 'splaining to everyone this :-\

[02-Nov-2011 16:28:36] <rmatte> netdisco does some extra stuff like letting you actually enable/disable switch ports from the web interface if you want

[02-Nov-2011 16:28:50] <rmatte> identifying duplex mismatches

[02-Nov-2011 16:28:51] <tightwork> thats pretty nice

[02-Nov-2011 16:28:52] <rmatte> etc...

[02-Nov-2011 16:29:13] <tightwork> The infrastructure im working with now im probably 10 years old,  home brew php crud

[02-Nov-2011 16:29:23] <tightwork> its terrible

[02-Nov-2011 16:29:25] <rmatte>

[02-Nov-2011 16:29:32] <rmatte> that's the style of network map it uses

[02-Nov-2011 16:29:35] <rmatte> more structured

[02-Nov-2011 16:29:56] <rmatte> eugh, vim

[02-Nov-2011 16:29:58] * rmatte shivers

[02-Nov-2011 16:31:13] kocolosk is now known as kocolosk|away

[02-Nov-2011 16:31:24] <tightwork> I do like how in zenoss.. I can add a switch and discover by subnet.. last time I tried to get a new network monitoring going with opennms.. I tried to go the real lazy way and add our /17 and got some angry calls :-p

[02-Nov-2011 16:31:39] <rmatte> lol

[02-Nov-2011 16:32:13] <tightwork> currently using ipswitch whats up gold... thing is horrible.

[02-Nov-2011 16:32:27] <rmatte> whats up gold blows hard

[02-Nov-2011 16:32:40] <tightwork> it will crash and the interface will be up and alerts wont get sent out.. only way you know it crashed is ... 'oh snap is been quiet for way too long'

[02-Nov-2011 16:33:42] <rmatte> lol

[02-Nov-2011 16:34:05] <rmatte> well, that can technically happen with Zenoss daemons too, but I monitor mine from a central Zenoss instance

[02-Nov-2011 16:34:13] <rmatte> and then have something monitoring that

[02-Nov-2011 16:34:25] <rmatte> they rarely ever go down

[02-Nov-2011 16:35:33] <tightwork> yeah redundancy is key

[02-Nov-2011 16:37:05] <rmatte> I have 16 Production Zenoss servers (all full instances of Core)

[02-Nov-2011 16:37:08] <rmatte> lots of fun to manage

[02-Nov-2011 16:40:00] <Ve2dmn1> I have a small question: How do I rename an IPInterface? I monitor a switch which has 48 ports and they are all named "Ehternet Interface"... the reports are fun to read...

[02-Nov-2011 16:40:19] <rmatte> Ve2dmn1: what kind of switch is it?

[02-Nov-2011 16:40:22] <Ve2dmn1> my Google skills are failing me...

[02-Nov-2011 16:40:26] <Ve2dmn1> D-link

[02-Nov-2011 16:40:37] <rmatte> try this... go in to collector plugins for the device

[02-Nov-2011 16:41:14] <rmatte> you'll see a plugin called zenoss.snmp.InterfaceAliasMap

[02-Nov-2011 16:41:25] <rmatte> move that to the active plugins list

[02-Nov-2011 16:41:41] <rmatte> leave the existing InterfaceMap plugin since the InterfaceAliasMap plugin needs it to work

[02-Nov-2011 16:41:44] <rmatte> then remodel the device

[02-Nov-2011 16:42:01] <rmatte> what that will do is it'll pick up the interface names from ifName instead of ifDescr

[02-Nov-2011 16:43:09] <Ve2dmn1> I don't have zenoss.snmp.InterfaceAliasMap, only zenoss.snmp.InterfaceMap

[02-Nov-2011 16:43:29] <rmatte> It comes with Zenoss, you have to have it lol

[02-Nov-2011 16:43:34] <Ve2dmn1> Forgot to mention that I'm running Zenoss 3.0.0

[02-Nov-2011 16:43:47] <rmatte> it'll be in the right hand pane

[02-Nov-2011 16:43:48] <rmatte> there's a link to reveal it that you need to click

[02-Nov-2011 16:43:49] <rmatte> it's hidden by default

[02-Nov-2011 16:44:03] <rmatte> not pane, but right hand list rather

[02-Nov-2011 16:44:13] <Ve2dmn1> all right. Found

[02-Nov-2011 16:45:22] <rmatte> so drag it from the right list to the left list

[02-Nov-2011 16:45:23] <rmatte> then save

[02-Nov-2011 16:45:26] <rmatte> then remodel the device

[02-Nov-2011 16:46:05] <Ve2dmn1> Hum... Got a Traceback ...

[02-Nov-2011 16:46:13] <Ve2dmn1> ERROR zen.ZenModeler: Problem while executing plugin zenoss.snmp.InterfaceAliasMap

[02-Nov-2011 16:46:17] <rmatte> oh, that was actually a bug in 3.0.0

[02-Nov-2011 16:46:24] <Ve2dmn1> I see...

[02-Nov-2011 16:46:24] <rmatte> 3.0.0 was the buggiest version of Zenoss ever released

[02-Nov-2011 16:46:31] <rmatte> not sure why you're still running that version?

[02-Nov-2011 16:46:43] <rmatte> anyways, let me find the patch that fixed that

[02-Nov-2011 16:46:52] <Ve2dmn1> Because my boss is an EE and changing version is a big NO

[02-Nov-2011 16:46:59] <rmatte> lame

[02-Nov-2011 16:47:09] <rmatte> there were so many bad bugs in that version

[02-Nov-2011 16:47:18] <Ve2dmn1> In electronic design an exact version is VERY important

[02-Nov-2011 16:47:49] <rmatte> Fix InterfaceAliasMap issue in 3.0: zenpatch 22540

[02-Nov-2011 16:47:49] <rmatte> Fix IP Service Monitor issue in 3.0: zenpatch 25791

[02-Nov-2011 16:47:54] <rmatte> a couple of patches you can apply

[02-Nov-2011 16:47:59] <rmatte> the first one fixes that issue you're seeing

[02-Nov-2011 16:48:06] <rmatte> It'll require a restart

[02-Nov-2011 16:48:06] <Ve2dmn1> between 2 version of Xilinx tools, your design might break for the exact same hardware...

[02-Nov-2011 16:48:37] <rmatte> Yeh, well, 3.0 was the wrong version to pick

[02-Nov-2011 16:48:38] <Ve2dmn1> Ok. I'll talk to him about an upgrade to 3.2.1 instead.

[02-Nov-2011 16:48:46] <rmatte> Yeh

[02-Nov-2011 16:48:55] <rmatte> good idea

[02-Nov-2011 16:50:49] <Ve2dmn1> I should probably start with the server itself... it's running Centos 4.5

[02-Nov-2011 16:50:57] <rmatte> anyways, if you apply the AliasMap patch in the mean time and do a quick restart of Zenoss, you'll get better interface names

[02-Nov-2011 16:51:01] <Ve2dmn1> Sorry, that's another one

[02-Nov-2011 16:51:05] <rmatte> provided ther ifName values don't suck too

[02-Nov-2011 16:51:10] <Ve2dmn1> it's actually 5.x

[02-Nov-2011 16:51:33] <Ve2dmn1> Will it automagically ajust the reports also?

[02-Nov-2011 16:51:37] <rmatte> I still have Ubuntu Server LTE 8.04 running on a bunch of mine, I'm slowly working through to update them all to 10.04

[02-Nov-2011 16:52:01] <rmatte> well it depends on what report you're talking about

[02-Nov-2011 16:52:20] <Ve2dmn1> I'm planning to migrate everything to Ubuntu 12.04 LTS in June

[02-Nov-2011 16:52:26] <rmatte> if it's custom graph report where you specifically selected certain interfaces, no.  If it's one of the stock interface reports, yes

[02-Nov-2011 16:52:40] <rmatte> since the interface names are changing you'll lose your graph data though

[02-Nov-2011 16:52:49] <Ve2dmn1> It's just basic reports. We use it to punish people who abuse bandwith

[02-Nov-2011 16:52:56] <rmatte> unless you somehow figure out which is which and copy the RRD files over to the new folders from the old ones

[02-Nov-2011 16:53:17] <rmatte> If you can't see which was which the data is basically useless anyways

[02-Nov-2011 16:53:40] <rmatte> You're talking about Performance Reports -> Interfaces?

[02-Nov-2011 16:53:51] <rmatte> If so, yes that'll update automatically

[02-Nov-2011 16:54:06] <rmatte> It always pulls the most recent componentns

[02-Nov-2011 16:54:09] <rmatte> components*

[02-Nov-2011 16:56:11] <rmatte> 2 more days and I'm on vacation.  Can't wait.

[02-Nov-2011 16:56:39] <Ve2dmn1> Priority one is making sure the employees can work without any network disturbance. I think I can live with losing the past graphs

[02-Nov-2011 16:57:46] <rmatte> yup

[02-Nov-2011 16:58:17] <johnnynoc> hey rmatte

[02-Nov-2011 16:58:21] <rmatte> hey

[02-Nov-2011 16:58:28] <Ve2dmn1> Anyway. Thanks for the help

[02-Nov-2011 16:58:29] <johnnynoc> i just saw a post you made about the guy having a problem with zenchkrels

[02-Nov-2011 16:58:35] <rmatte> yup

[02-Nov-2011 16:58:47] <johnnynoc> i have a similar issue stemming from a broken install/uninstall of a zenpack i was wondering if you could help me track down

[02-Nov-2011 16:58:57] <johnnynoc> when i ran that snippet of code i received this error

[02-Nov-2011 16:59:02] <johnnynoc> 2011-11-02 15:56:04 WARNING OFS.Uninstalled Could not import class 'IMMDevice' from module ''

[02-Nov-2011 16:59:20] <rmatte> is that ZenPack still actually installed on your system?

[02-Nov-2011 16:59:22] <johnnynoc> that's from a zenpack that i've been working with IBM on developing..  i tried to upgrade the zenpack when they had a new version but that made things very unhappy

[02-Nov-2011 16:59:32] <johnnynoc> not anymore

[02-Nov-2011 16:59:48] <rmatte> ok, so it's been removed but it's still complaining?

[02-Nov-2011 17:00:06] <rmatte> It doesn't show up in the ZenPack list even though it's removed does it?

[02-Nov-2011 17:01:08] <rmatte> cluther: good day

[02-Nov-2011 17:01:59] <johnnynoc> no it does not show up in the list

[02-Nov-2011 17:02:30] <johnnynoc> however, it looks like it still exists in $ZENHOME/zenpacks

[02-Nov-2011 17:02:30] <rmatte> ok... if you navigate to $ZENHOME/ZenPacks, is there still a directory for it in there?

[02-Nov-2011 17:02:43] <johnnynoc> (mind you, i've removed/readded this zenpack multiple times trying to figure this out)

[02-Nov-2011 17:02:43] <rmatte> ok, start off my moving (not deleting) that directory to somewhere else

[02-Nov-2011 17:02:54] <rmatte> see if it continues to do that afterwards

[02-Nov-2011 17:02:59] <rmatte> by moving*

[02-Nov-2011 17:03:13] <rmatte> by somewhere else I mean your home directory or whatever

[02-Nov-2011 17:03:14] <rmatte>

[02-Nov-2011 17:03:31] <johnnynoc> ok, and then do i need to restart it?

[02-Nov-2011 17:03:39] <rmatte> might as well

[02-Nov-2011 17:03:48] <rmatte> just to be sure

[02-Nov-2011 17:04:16] <johnnynoc> ok

[02-Nov-2011 17:04:18] cluther_ is now known as cluther

[02-Nov-2011 17:06:51] <johnnynoc> it didn't complain this time when i ran it but i'm still having the same problem i had before where the infrastrcuture page is stuck

[02-Nov-2011 17:07:10] <johnnynoc> and i see this KeyError in event.log

[02-Nov-2011 17:07:11] <johnnynoc> KeyError: 'IMM'

[02-Nov-2011 17:08:05] <johnnynoc> starts like this: 2011-11-02T16:06:31 ERROR extdirect 'IMM'

[02-Nov-2011 17:08:09] <johnnynoc> then the traceback

[02-Nov-2011 17:08:16] <rmatte> Do you have the pack on hand to reinstall it?

[02-Nov-2011 17:08:27] <rmatte> Reinstall/uninstall might work

[02-Nov-2011 17:09:35] <johnnynoc> yes i do

[02-Nov-2011 17:09:45] <johnnynoc> unfortunately i've tried that with no success but i'll giv eit another go now

[02-Nov-2011 17:09:51] <rmatte> k

[02-Nov-2011 17:10:44] <rmatte> make sure you restart zope after installing it before you attempt the removal

[02-Nov-2011 17:11:22] <rmatte> and do both the install and the removal from the commandline, not the UI

[02-Nov-2011 17:11:49] <johnnynoc> *nod*

[02-Nov-2011 17:11:50] <johnnynoc> reinstalled

[02-Nov-2011 17:11:52] <johnnynoc> restarted zope

[02-Nov-2011 17:12:11] <rmatte> k

[02-Nov-2011 17:12:14] <johnnynoc> now event.log is complaining about a device that used to be around

[02-Nov-2011 17:12:29] <rmatte> did you try that code I put in that post?

[02-Nov-2011 17:12:53] <johnnynoc> yes

[02-Nov-2011 17:12:57] <johnnynoc> just tried it, and am now restarting zenoss

[02-Nov-2011 17:13:16] <rmatte> I'm thinking there was some ghost device that was still hanging on to attributes provided by the pack

[02-Nov-2011 17:13:25] <rmatte> which was in turn preventing a proper uninstall of it

[02-Nov-2011 17:13:36] <rmatte> that code will actually print something if it removed the device

[02-Nov-2011 17:13:41] <rmatte> did it print anything?

[02-Nov-2011 17:14:00] <johnnynoc> no

[02-Nov-2011 17:14:03] <rmatte> hmmm

[02-Nov-2011 17:14:08] <johnnynoc> it did the first time i ran it

[02-Nov-2011 17:14:11] <johnnynoc> but it said it was removing 'broken'

[02-Nov-2011 17:14:11] <johnnynoc> heh

[02-Nov-2011 17:14:12] <rmatte> ah

[02-Nov-2011 17:14:28] <rmatte> still if it printed anything that's good

[02-Nov-2011 17:14:33] <rmatte> lol

[02-Nov-2011 17:14:41] <johnnynoc> heh, i have the same isssue

[02-Nov-2011 17:14:45] <rmatte> have you tried a reindex?

[02-Nov-2011 17:14:47] <johnnynoc> saying NotFound: somefuckingassholedevice

[02-Nov-2011 17:14:55] <johnnynoc> i tried that the other day, but if i recall correctly it failed

[02-Nov-2011 17:14:57] <johnnynoc> let me try again

[02-Nov-2011 17:15:00] <rmatte> k

[02-Nov-2011 17:15:14] <johnnynoc> i even tried re-cataloging or whatever

[02-Nov-2011 17:15:16] <johnnynoc> but it died

[02-Nov-2011 17:15:17] <johnnynoc>

[02-Nov-2011 17:16:56] <johnnynoc> i get that damn attributeerror

[02-Nov-2011 17:17:00] <johnnynoc> AttributeError: primaryAq

[02-Nov-2011 17:17:34] <rmatte> hmm

[02-Nov-2011 17:18:33] <rmatte> remove the ZenPack

[02-Nov-2011 17:18:37] <johnnynoc> when i run some of the suggested fixes such as zenchkrels or zenfixit i can see the problem is related to my performance monitor

[02-Nov-2011 17:18:43] <johnnynoc> like, its still attached somehow

[02-Nov-2011 17:18:48] <johnnynoc> this fucking bastard device!

[02-Nov-2011 17:18:54] <rmatte> which is what the code I gave is supposed to fix

[02-Nov-2011 17:19:00] <rmatte> lol

[02-Nov-2011 17:19:48] <johnnynoc> i know

[02-Nov-2011 17:19:54] <johnnynoc> i saw it and got all excited!

[02-Nov-2011 17:20:03] <rmatte> lol

[02-Nov-2011 17:20:08] <johnnynoc> i'm just trying to avoid having to build this server all over again..  it's just a dev server so not a big deal but still a p.i.t.a.

[02-Nov-2011 17:20:09] <johnnynoc> :/

[02-Nov-2011 17:20:18] <johnnynoc> and figure this is an opportunity to get more familiar with some internals

[02-Nov-2011 17:20:56] <rmatte> yeh, better to fix it in case it ever happens again

[02-Nov-2011 17:21:44] <johnnynoc> yea i mean, this seems like somewhat of a common problem..  but i know my situation may be a little different as there was a pretty significant change to the zenpack which should have been deleted instead of installed on top of

[02-Nov-2011 17:21:47] <johnnynoc> oh well

[02-Nov-2011 17:25:14] <johnnynoc> so when i run zenchkrels i get this

[02-Nov-2011 17:25:16] <johnnynoc> AttributeError: 'IMMDevice' object has no attribute 'id'

[02-Nov-2011 17:25:52] <johnnynoc> in

[02-Nov-2011 17:25:53] <rmatte> yeh, there's clearly some IMMDevice object sitting around in there

[02-Nov-2011 17:26:13] <johnnynoc> IMMDevice?  Y U NO DELETE

[02-Nov-2011 17:26:23] <hmp> johnnynoc:

[02-Nov-2011 17:26:34] <johnnynoc> hehehe

[02-Nov-2011 17:26:59] <johnnynoc> rmatte: anything else you could possibly suggest before i become a quitter

[02-Nov-2011 17:27:15] <rmatte> yeh hold on

[02-Nov-2011 17:27:24] <johnnynoc> oooh

[02-Nov-2011 17:27:27] <rmatte> I have an alternative to zencheckrels, let me find it

[02-Nov-2011 17:27:29] <johnnynoc> i ran and it didn't complain

[02-Nov-2011 17:27:32] <rmatte> it has a better success rate

[02-Nov-2011 17:27:33] <johnnynoc> Removing IP addresses linked to invalid devices.   - removing

[02-Nov-2011 17:27:35] <johnnynoc> and i see this

[02-Nov-2011 17:27:37] <johnnynoc> ooh ooh ooh

[02-Nov-2011 17:27:39] * johnnynoc does a lil dance

[02-Nov-2011 17:28:39] <rocket> why are you dancing?

[02-Nov-2011 17:29:22] <johnnynoc> well, becuase i thought i was on my way to seeing this fixed

[02-Nov-2011 17:29:23] <rmatte> johnnynoc:

[02-Nov-2011 17:29:26] <johnnynoc> however, now i'm crying in my beer

[02-Nov-2011 17:29:32] <rmatte> try that

[02-Nov-2011 17:29:50] <johnnynoc> trying

[02-Nov-2011 17:31:47] <rmatte> If you see any Failed on lines paste them here

[02-Nov-2011 17:31:55] <rmatte> since we'll have to tackle them manually at that point

[02-Nov-2011 17:32:43] <johnnynoc> ok

[02-Nov-2011 17:36:35] <rmatte> It could take a while to run since it'll tear through every relation in the database

[02-Nov-2011 17:36:52] <rmatte> zencheckrels stops as soon as it hits an error

[02-Nov-2011 17:37:14] <johnnynoc> ah i gotcha

[02-Nov-2011 17:37:22] <johnnynoc> fortunately there isn't too much going on on this server but it is still running

[02-Nov-2011 17:37:42] <rmatte> no failed line yet?

[02-Nov-2011 17:38:07] <johnnynoc> nope, not yet

[02-Nov-2011 17:38:12] <johnnynoc> might be running slow - this is on a vm

[02-Nov-2011 17:39:12] <johnnynoc> hmm

[02-Nov-2011 17:39:20] <johnnynoc> ok so it's complaining about some ip services

[02-Nov-2011 17:39:26] <johnnynoc> some routes

[02-Nov-2011 17:39:34] <johnnynoc> that's all i see

[02-Nov-2011 17:39:55] <rmatte> it finished running>?

[02-Nov-2011 17:40:03] <johnnynoc> yes

[02-Nov-2011 17:40:16] <rmatte> ok, commit and try zencheckrels again

[02-Nov-2011 17:40:17] <johnnynoc> same error in event.log

[02-Nov-2011 17:40:17] <johnnynoc>

[02-Nov-2011 17:40:20] <rmatte> see if you get the same error

[02-Nov-2011 17:40:45] <rmatte> so you did actually commit yet?

[02-Nov-2011 17:40:46] <rmatte> yeh*

[02-Nov-2011 17:41:02] <johnnynoc> yes

[02-Nov-2011 17:41:08] <johnnynoc> i commited, zopectl restart

[02-Nov-2011 17:41:41] <rmatte> wish I could get my hands on the server myself to screw around with it lol

[02-Nov-2011 17:41:53] <rmatte> at this point it'd be trial and error

[02-Nov-2011 17:42:46] <johnnynoc> hrm

[02-Nov-2011 17:43:24] <johnnynoc> can i just remove the localhost performance collector?

[02-Nov-2011 17:43:24] <johnnynoc>

[02-Nov-2011 17:43:29] <rmatte> have yout tried going in to the management interface to track this object down?

[02-Nov-2011 17:43:35] <rmatte> I mean, it has to be somewhere

[02-Nov-2011 17:44:58] <johnnynoc> yes i have

[02-Nov-2011 17:45:00] <johnnynoc> yes, i agree!

[02-Nov-2011 17:45:13] <johnnynoc> it's funny, so i try to load the infrastructure page and i see the loading please wait message

[02-Nov-2011 17:45:18] <johnnynoc> but i see the device class and the (1) device in it

[02-Nov-2011 17:45:22] <johnnynoc> GRR

[02-Nov-2011 17:45:23] <rmatte> it's too bad that you don't have a better idea of where the object actually is in the database pathwise

[02-Nov-2011 17:45:30] <rmatte> just IMMWhatever isn't very helpful

[02-Nov-2011 17:45:31] <johnnynoc> well

[02-Nov-2011 17:45:33] <johnnynoc> i think i do

[02-Nov-2011 17:45:38] <johnnynoc> i know the device class

[02-Nov-2011 17:45:42] <johnnynoc> i know the name

[02-Nov-2011 17:45:43] <rmatte> which class?

[02-Nov-2011 17:45:49] <johnnynoc> when i tried to go to that directly in the url it dies

[02-Nov-2011 17:45:58] <johnnynoc>  /Devices/Server/IMM

[02-Nov-2011 17:46:01] <rmatte> right, but we might be able to do something via zendmd

[02-Nov-2011 17:46:38] <rmatte> have you uninstalled the ZenPack at this point?

[02-Nov-2011 17:47:02] <johnnynoc> no, not yet

[02-Nov-2011 17:47:05] <rmatte> if not, might have to remove it then try checking the relations again

[02-Nov-2011 17:47:14] <rmatte> since it'll only technically be broken when the class no longer exists

[02-Nov-2011 17:47:33] <rmatte> if it actually sticks around after removal

[02-Nov-2011 17:47:42] <johnnynoc> ok

[02-Nov-2011 17:47:44] <johnnynoc> removed the zenpack

[02-Nov-2011 17:47:46] <johnnynoc> restarted zope

[02-Nov-2011 17:48:09] <johnnynoc> going to try everything over again i guess

[02-Nov-2011 17:48:20] <rmatte> check if the ZenPack directory is still in $ZENHOME/ZenPacks

[02-Nov-2011 17:48:36] <johnnynoc> it is

[02-Nov-2011 17:48:41] <johnnynoc> you son of a bitch

[02-Nov-2011 17:48:44] <rmatte> that's bizarre, it should get removed

[02-Nov-2011 17:48:48] <rmatte> no errors when removing it?

[02-Nov-2011 17:48:57] <johnnynoc> no sir

[02-Nov-2011 17:49:12] <rmatte> hmmm

[02-Nov-2011 17:49:27] <rmatte> anyways, try it all again now

[02-Nov-2011 17:49:31] <rmatte> we'll see if it does anything

[02-Nov-2011 17:49:42] <johnnynoc> i moved it

[02-Nov-2011 17:49:48] <rmatte> k

[02-Nov-2011 17:50:05] <rmatte> try the zendmd code to check relations again

[02-Nov-2011 17:50:07] <rmatte> for starters

[02-Nov-2011 17:50:52] <johnnynoc> no complaints

[02-Nov-2011 17:51:05] <rmatte> not failed lines at all this time?

[02-Nov-2011 17:51:07] <rmatte> no*

[02-Nov-2011 17:51:09] <johnnynoc> zenfixit didn't complain

[02-Nov-2011 17:51:10] <johnnynoc> no, none

[02-Nov-2011 17:51:20] <rmatte> oh zenfixit, but you tried the other code I gave you/

[02-Nov-2011 17:51:21] <rmatte> ?

[02-Nov-2011 17:51:43] <johnnynoc> yes

[02-Nov-2011 17:51:47] <johnnynoc> i ran your code in zendmd, and nothing

[02-Nov-2011 17:51:51] <rmatte> k

[02-Nov-2011 17:51:52] <johnnynoc> no errors or anything

[02-Nov-2011 17:52:00] <rmatte> does /Devices/Server/IMM no longer exist?

[02-Nov-2011 17:52:20] <johnnynoc> nope, i don't see it anymore

[02-Nov-2011 17:52:27] <johnnynoc> mind you, this is all greyed out under the 'loading please wait' message

[02-Nov-2011 17:52:30] <rmatte> can you reproduce the error?

[02-Nov-2011 17:52:36] <johnnynoc> yes

[02-Nov-2011 17:52:41] <johnnynoc> all i have to do is navigate to the infrastructure page

[02-Nov-2011 17:52:49] <johnnynoc> and it's forever loading

[02-Nov-2011 17:53:16] <rmatte> AttributeError: 'IMMDevice' object has no attribute 'id'...

[02-Nov-2011 17:53:22] <rmatte> was IMMDevice the name of the template?

[02-Nov-2011 17:53:43] <johnnynoc> I don't recall specifically, but i believe so yes

[02-Nov-2011 17:53:49] <johnnynoc> hang on, let me look on my prod server

[02-Nov-2011 17:54:17] <johnnynoc> hmm

[02-Nov-2011 17:54:17] <rmatte> k

[02-Nov-2011 17:54:21] <johnnynoc> i'm not sure to be honest, doesn't look like it

[02-Nov-2011 17:54:40] <rmatte> it'd help a lot to know what it actually was

[02-Nov-2011 17:54:40] <johnnynoc> i don't have an IMMDevice template on my prod server

[02-Nov-2011 17:54:46] <rmatte> a template, a component, a device

[02-Nov-2011 17:55:42] <johnnynoc> i think it's the name of a python class

[02-Nov-2011 17:55:45] <johnnynoc> from zPythonClass

[02-Nov-2011 17:55:46] <johnnynoc>

[02-Nov-2011 17:56:06] <rmatte> hmmm

[02-Nov-2011 17:56:19] <rmatte> can I see the full traceback that's in the log?

[02-Nov-2011 17:56:30] <johnnynoc> yes, once this zenchkrels is finished

[02-Nov-2011 17:56:34] <rmatte> k

[02-Nov-2011 17:56:37] <johnnynoc> well

[02-Nov-2011 17:56:37] <johnnynoc> nm

[02-Nov-2011 17:56:38] <johnnynoc> hang on

[02-Nov-2011 17:56:40] <johnnynoc> i'll just pause it

[02-Nov-2011 17:56:45] <rmatte> lol

[02-Nov-2011 17:56:59] <johnnynoc> er

[02-Nov-2011 17:57:06] <johnnynoc> looks like the /query might be hard to read

[02-Nov-2011 17:57:08] <johnnynoc> should i just pastebin it ?

[02-Nov-2011 17:57:30] <rmatte> please

[02-Nov-2011 17:57:37] <rmatte> or, whichever

[02-Nov-2011 17:57:47] <johnnynoc>

[02-Nov-2011 17:57:50] <johnnynoc> thats from event.log

[02-Nov-2011 17:58:21] <johnnynoc> thats what i get when i click infrastructure

[02-Nov-2011 17:58:48] <johnnynoc> i can paste output of zenchkrels, zenfixit, when i'm back from using the little boys room

[02-Nov-2011 17:58:49] <johnnynoc> brb

[02-Nov-2011 18:07:47] <rmatte> so yeh, after checking out line 128 in /usr/local/zenoss/zenoss/Products/ZenUtils/extdirect/, it's basically cycling through a bunch of objects and not finding an ID for that object when it's expecting one...

[02-Nov-2011 18:07:56] <rmatte> which is what I figured at the start

[02-Nov-2011 18:08:32] <rmatte> I just don't know what value it's getting the "IMM" from

[02-Nov-2011 18:09:13] <rmatte> If I knew I could write a script that scans through and deletes an object when it matches that

[02-Nov-2011 18:09:30] <rmatte> but the IMM obviously isn't from the object id property since there isn't one

[02-Nov-2011 18:10:53] <johnnynoc> hm

[02-Nov-2011 18:11:02] <rmatte> well, it seems to be coming from...

[02-Nov-2011 18:11:09] <rmatte> e.__name__

[02-Nov-2011 18:11:22] <rmatte> where e is the exception

[02-Nov-2011 18:12:39] <johnnynoc> well, fuck it i guess

[02-Nov-2011 18:12:52] <johnnynoc> i've struggled with this for long enough where i should just start over if i can't figure it out by the end of the day

[02-Nov-2011 18:12:58] <johnnynoc> seems like after everything i've done, i'm still back at square 1

[02-Nov-2011 18:13:03] <johnnynoc> KeyError: 'IMM'

[02-Nov-2011 18:13:05] <johnnynoc>

[02-Nov-2011 18:15:58] <rmatte> I'm thinking you'd have to add something to $ZENHOME/Products/ZenUtils/extdirect/ on line 130 like...

[02-Nov-2011 18:15:59] <rmatte> if e.__name__ == "IMM":

[02-Nov-2011 18:15:59] <rmatte>     self.getPrimaryParent()._remove(self)

[02-Nov-2011 18:16:11] <rmatte> though no idea if that particular code would work

[02-Nov-2011 18:16:25] <rocket> rmatte if e.__name__ == "IMM":

[02-Nov-2011 18:16:32] <johnnynoc> i mean, at this point it can't hurt

[02-Nov-2011 18:16:32] <rocket>        import pdb;pdb.set_trace()

[02-Nov-2011 18:16:32] <johnnynoc>

[02-Nov-2011 18:16:40] <rocket> then you can figure out more about the object

[02-Nov-2011 18:16:46] <rmatte> ah

[02-Nov-2011 18:17:02] <rocket> learn more about it

[02-Nov-2011 18:17:16] <rocket> technically do you guys know about pdb.post_mortem() ?

[02-Nov-2011 18:17:16] <rmatte> I don't even know if this line is right though, I'm just guessing based on what I see in the code...

[02-Nov-2011 18:17:22] <rmatte> if e.__name__ == "IMM":

[02-Nov-2011 18:17:40] <rmatte> nope, don't know about post_mortem, but I could guess what it does

[02-Nov-2011 18:17:51] <rmatte> name is a bit of a giveaway

[02-Nov-2011 18:18:18] <rocket>

[02-Nov-2011 18:18:41] <rmatte> I've read the pdb docs, but a lot of the time I prefer just debugging it myself via debug lines

[02-Nov-2011 18:18:48] <rmatte> less information to wade through

[02-Nov-2011 18:19:05] <rocket> you often have less info than you need

[02-Nov-2011 18:19:07] <rmatte> I mean right here we just need to remove that object wherever it may be

[02-Nov-2011 18:19:19] <rmatte> leftovers from the ZenPack

[02-Nov-2011 18:19:48] <johnnynoc> i'm afraid of what pdb.set_trace() is going to show me

[02-Nov-2011 18:20:44] <rmatte> I figure if extdirect/router can be modified and have a couple lines added to the Exception part of it to delete the object on exception, it could help

[02-Nov-2011 18:20:51] <rmatte> it's just a dev box anyways

[02-Nov-2011 18:21:17] <rmatte> either way, I need to get out of here and head home

[02-Nov-2011 18:21:33] <johnnynoc> yes

[02-Nov-2011 18:21:36] <johnnynoc> i appreciate the help

[02-Nov-2011 18:21:38] <johnnynoc> as always

[02-Nov-2011 18:22:33] <rmatte> no problem

[02-Nov-2011 18:36:10] kocolosk|away is now known as kocolosk

[02-Nov-2011 20:30:22] <[hC]> I cannot for the life of me figure out how to change a device in zenoss from snmp v2c to snmp v1

[02-Nov-2011 20:30:25] <[hC]> Anyone have any insight there?

[02-Nov-2011 20:30:33] <[hC]> Just an individual device.

[02-Nov-2011 20:30:41] <hmp> Go to that device

[02-Nov-2011 20:30:46] <hmp> then configuration properties

[02-Nov-2011 20:30:49] <hmp> and change it there

[02-Nov-2011 20:30:53] <hmp> it becomes localy defined

[02-Nov-2011 20:31:03] <hmp> its something like

[02-Nov-2011 20:31:08] <hmp> zSnmpVer

[02-Nov-2011 20:31:44] <[hC]> Aaahhhhhh you have to start typing in snmp there to see what the available options are.

[02-Nov-2011 20:31:58] <[hC]> oh. you could scroll and i missed it.

[02-Nov-2011 20:32:10] <[hC]> Wow. Thanks.

[02-Nov-2011 20:32:24] <hmp> np

[02-Nov-2011 20:33:13] <[hC]> While I've got you, is there a similarly easy way, when a device has http monitoring turned on, to specify a particular url to test?

[02-Nov-2011 20:33:36] <hmp> sorry, i mostly monitor network devices

[02-Nov-2011 20:33:42] <[hC]> no problem. Thx

[02-Nov-2011 23:00:48] <bseklecki> time to build some RHEL6 RPMs

[02-Nov-2011 23:01:08] <bseklecki> and its time to build them for POWER and Itanium2 on RHEL5

[02-Nov-2011 23:41:00] kocolosk is now known as kocolosk|away

[03-Nov-2011 00:00:01] [disconnected at Thu Nov  3 00:00:01 2011]

[03-Nov-2011 00:00:01] [connected at Thu Nov  3 00:00:01 2011]

[03-Nov-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]

[03-Nov-2011 07:32:58] DW-10297 is now known as DogWater

[03-Nov-2011 07:34:09] <enkrypt> I'm often getting "Unable to read processes on device;error: first argument must be string or compiled pattern"

[03-Nov-2011 07:34:33] <enkrypt> and events caused by "unable to read process" in general. they appear, and a few minutes later they disappear. and so on.

[03-Nov-2011 07:34:41] <enkrypt> is this a known issue? zenoss enterprise 3.2.1

[03-Nov-2011 09:10:04] <hmp> hello

[03-Nov-2011 09:10:18] <hmp> anyone using zInterfaceMapIgnoreTypes to ignore Juniper virtual interfaces?

[03-Nov-2011 09:11:48] <hmp> if I go to interface details in a device, it says: "Type: Encapsulation Interface_64", so my zInterfaceMapIgnoreTypes looks like: ^(...|...|...|Encapsulation Interface_64)

[03-Nov-2011 09:12:07] <hmp> i tried remodeling, delete and recreate but they always show up in interfaces tab of that juniper device

[03-Nov-2011 09:29:09] <jiveturkey> Anybody know how to specify a date range for interface graphs through the GUI in Zenoss 4?

[03-Nov-2011 09:29:51] <dhopp> jiveturkey:  I'm not an enterprise customer so I haven't used version 4…but in version 3 that's not possible

[03-Nov-2011 09:30:23] <jiveturkey> I'm going straight from 2.5.2 so I never actually used 3.  However, as I understand it, the GUIs are pretty much the same.

[03-Nov-2011 09:30:30] <jiveturkey> In sum, curses.

[03-Nov-2011 09:31:02] <jiveturkey> I guess we're gonna have to write a ZenPack for this.

[03-Nov-2011 09:31:36] <jiveturkey> For now, though, it's back to deconstructing base64, editing RRD graph commands, reconstructing base64, and feeding it to zenrender.

[03-Nov-2011 09:31:39] <jiveturkey> My favorite.

[03-Nov-2011 12:00:01] [disconnected at Thu Nov  3 12:00:01 2011]

[03-Nov-2011 12:00:02] [connected at Thu Nov  3 12:00:02 2011]

[03-Nov-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]

[03-Nov-2011 13:31:03] <twm1010> Anyone know if the virtual appliance has any tweaking/optimization done already?

[03-Nov-2011 13:34:18] <jb> i'm guessing not..

[03-Nov-2011 13:38:26] <twm1010> I'm FINALLY building myself a template VM, just wanted to know where to start :|

[03-Nov-2011 13:39:27] <twm1010> the appliance is centos based, i started building an ubuntu base yesterday, but I think this will save me a little effort

[03-Nov-2011 13:48:20] <twm1010> interesting, its centos 5.4, a bit old now

[03-Nov-2011 13:59:11] <jmp242> So anyone else seen the update to the format for the Zenpacks page?

[03-Nov-2011 14:11:01] <twm1010> @jmp242 - I like the separation between legacy and current

[03-Nov-2011 14:12:04] <johnnynoc> rmatte:

[03-Nov-2011 14:12:04] <johnnynoc> where you be

[03-Nov-2011 14:13:01] <jmp242> not here lol

[03-Nov-2011 14:13:15] <jmp242> oh no, it does look like he's here

[03-Nov-2011 14:15:12] <johnnynoc> hah

[03-Nov-2011 14:15:17] <johnnynoc> have a questino abotu his formula datasource zenpack

[03-Nov-2011 14:15:25] <johnnynoc> not sure why this datapoitn isn't being calculated properly

[03-Nov-2011 14:18:50] <johnnynoc> he's going to have to give me his cell #

[03-Nov-2011 14:18:53] <johnnynoc> on-call, 24/7 support

[03-Nov-2011 14:18:55] <johnnynoc>

[03-Nov-2011 14:20:45] <johnnynoc> anybody use his formula datasource zenpack?

[03-Nov-2011 14:29:46] <jmp242> johnnynoc: pay him enough and he might lol

[03-Nov-2011 14:29:52] <jmp242> I don't use it

[03-Nov-2011 14:30:04] <jmp242> maybe put in a forum thread?

[03-Nov-2011 14:30:14] <johnnynoc> right on, no big deal

[03-Nov-2011 14:30:16] <johnnynoc> i'll bug him when he comes back around

[03-Nov-2011 14:30:26] <johnnynoc> not a bad idea

[03-Nov-2011 14:33:10] <rocket> Is this the zenpack that takes and combines two datasources?

[03-Nov-2011 14:33:17] <johnnynoc> yes

[03-Nov-2011 14:33:31] <rocket> you are aware of its performance limitations?

[03-Nov-2011 14:33:41] <johnnynoc> yes

[03-Nov-2011 14:33:58] <johnnynoc> figured i'd see just how bad it was if i tried to calculate a single datasource for our linux machines

[03-Nov-2011 14:34:14] <johnnynoc> i'm trying to caluclate %nice

[03-Nov-2011 14:34:25] <johnnynoc> and i have to determine the number of cpus/cores to get this number right

[03-Nov-2011 14:34:31] <johnnynoc> have another suggestion on how i could accomplish that? 

[03-Nov-2011 14:35:47] <johnnynoc> ssems like ssCpuRawWait suffers from a similar issue

[03-Nov-2011 14:36:03] <johnnynoc> ie, i shouldn't see percentages above 100%, which i assume i see because of multi-processors

[03-Nov-2011 14:41:18] kocolosk|away is now known as kocolosk

[03-Nov-2011 14:54:19] <johnnynoc> figured out my problem

[03-Nov-2011 14:54:29] <johnnynoc> the datapoint i want to mess with needs to actually be in an rrd

[03-Nov-2011 14:54:37] <johnnynoc> i was hoping that wasn't necessary :/

[03-Nov-2011 14:56:25] <jmp242> I think all the datapoints have to be in an rrd

[03-Nov-2011 15:02:28] <johnnynoc> *nod*

[03-Nov-2011 15:02:40] <johnnynoc> not until i added a graph for this datapoint did it start to work

[03-Nov-2011 15:03:01] <jmp242> Mmm

[03-Nov-2011 15:03:28] <jmp242> I didn't think you had to have a graph to get an RRD, but I might be wrong, cause there was an invisible graph zenpack, and why would you need it if you could just not have a graph

[03-Nov-2011 15:03:53] <johnnynoc> heh

[03-Nov-2011 15:03:57] <johnnynoc> good point

[03-Nov-2011 15:04:14] <johnnynoc> i actually added the rrd to an existing graph

[03-Nov-2011 15:04:19] <johnnynoc> so now my cpu utilization graph shows how many cores

[03-Nov-2011 15:04:23] <johnnynoc> not ideal, but whatever

[03-Nov-2011 15:04:24] <johnnynoc> :/

[03-Nov-2011 15:04:32] <jmp242> Hmm, well you could make it a different graph

[03-Nov-2011 15:05:00] <jmp242> and

[03-Nov-2011 15:05:04] <jmp242> hide it

[03-Nov-2011 15:05:27] <johnnynoc> hmm

[03-Nov-2011 15:05:32] <johnnynoc> something i'll consider

[03-Nov-2011 15:05:40] <johnnynoc> i'm concerned i'll have to bail on this whole idea if it turns out to be really slow for all of our linux servers

[03-Nov-2011 15:05:54] <johnnynoc> i'm trying to keep the cycle times below 60s so we have 60s precision in our graphs

[03-Nov-2011 15:05:54] <jmp242> mmm ok

[03-Nov-2011 15:06:02] <johnnynoc> and if this means zencommand takes 5 minutes to run, that will be a problem

[03-Nov-2011 15:06:08] <jmp242> Yea

[03-Nov-2011 15:06:13] <johnnynoc> i say that because the zenpack is kinda slow when calculating this stuff

[03-Nov-2011 15:06:15] <jmp242> isn't the default 5 min though

[03-Nov-2011 15:06:15] <johnnynoc> (known issue)

[03-Nov-2011 15:06:25] <johnnynoc> yes but i turned it up to 1min

[03-Nov-2011 15:06:30] <jmp242> mmm, never a great idea

[03-Nov-2011 15:06:33] <johnnynoc> 5min precision wasn't good enough

[03-Nov-2011 15:06:35] <jmp242> you might have to throw bigger hardware at it

[03-Nov-2011 15:06:42] <johnnynoc> *nod*

[03-Nov-2011 15:06:45] <johnnynoc> that's been done

[03-Nov-2011 15:06:45] <jmp242> or try multiple collectors

[03-Nov-2011 15:06:51] <johnnynoc> *nod*

[03-Nov-2011 15:06:52] <johnnynoc> and we have distributed collectors

[03-Nov-2011 15:06:57] <johnnynoc> my prob now is i'm running out of space on my SSD used for RRDs

[03-Nov-2011 15:07:01] <jmp242> well, maybe if it's taking too long

[03-Nov-2011 15:07:10] <jmp242> add a few more collectors and balance it out?

[03-Nov-2011 15:07:20] <jmp242> Is there a feeling that the zenpack can be improved?

[03-Nov-2011 15:07:22] <johnnynoc> easier said than done

[03-Nov-2011 15:07:22] <johnnynoc>

[03-Nov-2011 15:07:23] <jmp242> to speed it up?

[03-Nov-2011 15:07:39] <johnnynoc> yes, rmatte says if he daemonized it into a single thread it would be much faster

[03-Nov-2011 15:07:43] <johnnynoc> but he's been too busy to add that functionality

[03-Nov-2011 15:07:47] <johnnynoc> ZCA?

[03-Nov-2011 15:07:48] <johnnynoc>

[03-Nov-2011 15:07:49] <jmp242> Yea, you'd probably need to fund it

[03-Nov-2011 15:08:01] <jmp242> but yes, it might be possible

[03-Nov-2011 15:08:02] <johnnynoc> hold on

[03-Nov-2011 15:08:06] <johnnynoc> let me see what i have in my pocket

[03-Nov-2011 15:08:19] <jmp242> lol

[03-Nov-2011 15:08:24] <johnnynoc> $85!

[03-Nov-2011 15:17:29] <dhopp> johnnynoc:  what are you monitoring that 5 minutes isn't good enough?

[03-Nov-2011 15:31:10] <jb> Hackman238: around?

[03-Nov-2011 15:31:22] <jmp242> I don't think he is

[03-Nov-2011 15:31:39] <jb> k

[03-Nov-2011 15:32:25] <johnnynoc> dhopp: just servers

[03-Nov-2011 15:32:28] <johnnynoc> they wanted better than 5min precision

[03-Nov-2011 17:37:47] <johnnynoc> anybody know how i could setup a script to query zendmd on a remote server?

[03-Nov-2011 17:37:59] <johnnynoc> i feel like i've seen it done somewhere but can't seem to figure out how

[03-Nov-2011 18:01:43] <syoma> hi guys

[03-Nov-2011 18:02:59] <syoma> how can I set an threshold on interface component ?

[03-Nov-2011 18:03:21] <syoma> the device is integrated with snmp and default template

[03-Nov-2011 18:03:35] <syoma> can somebody help me ?

[03-Nov-2011 18:04:03] <syoma> sorry, is the linux default template

[03-Nov-2011 18:08:41] <johnnynoc_> still there?

[03-Nov-2011 18:08:52] <johnnynoc_> what version of zenoss

[03-Nov-2011 18:12:10] <johnnynoc_> syoma: ?

[03-Nov-2011 18:17:59] <johnnynoc_> ok gotta go

[03-Nov-2011 19:10:42] kocolosk is now known as kocolosk|away

[03-Nov-2011 19:59:30] <nachox> guys, is it possible get monitoring the way ntop does it with zenoss? ntop monitors network flows using pcap

[03-Nov-2011 19:59:47] <nachox> ntop is a bit broken at the moment so i cant really use that

[03-Nov-2011 20:07:28] <nachox> i'd use netflow but i cannot...

[03-Nov-2011 20:31:07] <bseklecki> <-- still valid?

[03-Nov-2011 20:35:57] <bseklecki> doesn't seem to work for me =/

[03-Nov-2011 20:44:13] kocolosk|away is now known as kocolosk

[03-Nov-2011 20:49:19] <bseklecki> it geocodes, but then it remains zoomed out

[03-Nov-2011 20:49:23] <bseklecki> any way to turn on debugging?

[03-Nov-2011 20:52:26] <bseklecki> meh, new key seems to fix it

[03-Nov-2011 20:52:31] <bseklecki> love windows-style debugging

[04-Nov-2011 00:00:01] [disconnected at Fri Nov  4 00:00:01 2011]

[04-Nov-2011 00:00:02] [connected at Fri Nov  4 00:00:02 2011]

[04-Nov-2011 00:00:16] [zenoss-logger (logger bot) has joined #zenoss]

[04-Nov-2011 08:19:52] <jmp242> TGIF everyone!!

[04-Nov-2011 08:37:28] <ion-fan> whats TGIF???

[04-Nov-2011 08:37:28] <Jane_Curry> TGIF ?????

[04-Nov-2011 08:46:00] <dpetzel> Thank God Its Friday

[04-Nov-2011 08:46:04] <enkrypt> ^^ yeah!

[04-Nov-2011 08:46:42] <jb> i think it's a .US thing

[04-Nov-2011 08:58:02] <jmp242> Might be

[04-Nov-2011 08:58:21] <jmp242> it was a 90s tv block on one of the broadcast channels too

[04-Nov-2011 09:00:35] <Jane_Curry> Ahh - you mean POETS day

[04-Nov-2011 09:01:24] <jmp242> sure Jane_Curry - is that a .UK thing?

[04-Nov-2011 09:03:19] <jmp242> BTW - Zenoss programming wizzes: two threads that I can't help with:

[04-Nov-2011 09:03:21] <jmp242>

[04-Nov-2011 09:03:32] <jmp242>

[04-Nov-2011 09:03:52] <jmp242> Though the second one - maybe he could go into Zope and delete it, but I'm afraid of it breaking relationships and making things worse

[04-Nov-2011 09:42:15] kocolosk is now known as kocolosk|away

[04-Nov-2011 09:54:08] kocolosk|away is now known as kocolosk

[04-Nov-2011 11:15:56] <nachox> why wont zenoss graph cpu usage? i am using snmp to monitor the equipement

[04-Nov-2011 11:16:11] <nachox> zenoss shows the processors to run at 1.9ghz

[04-Nov-2011 11:16:21] <nachox> err 0 mhz

[04-Nov-2011 11:24:40] <jmp242> nachox: we'll need a lot more info

[04-Nov-2011 11:24:45] <jmp242> what version of zenoss?

[04-Nov-2011 11:24:50] <jmp242> what os are you monitoring?

[04-Nov-2011 11:24:59] <jmp242> what zenpacks, if any, are you using for the monitoring

[04-Nov-2011 11:25:14] <nachox> the latest core from the page

[04-Nov-2011 11:25:41] <nachox> snmpwalk returns my performance counter in hrSWRunPerfCPU.X

[04-Nov-2011 11:25:53] <nachox> i am monitoring ubuntus

[04-Nov-2011 11:26:18] <nachox> and i dont think i installed any zenpacks

[04-Nov-2011 11:26:35] <nachox> actually, i know i didnt install any

[04-Nov-2011 11:29:24] <syoma> hi guys

[04-Nov-2011 11:30:20] <syoma> how I can create a threshold that indicates when to stop the rrd graph

[04-Nov-2011 11:30:34] <syoma> create an alert

[04-Nov-2011 11:30:36] <syoma> ?

[04-Nov-2011 11:36:43] <rmatte> syoma: not sure what you mean

[04-Nov-2011 11:37:50] <rmatte> "stop the rrd graph"?

[04-Nov-2011 11:38:01] <rmatte> you mean setting a max value?

[04-Nov-2011 11:39:07] <syoma> hi rmatte

[04-Nov-2011 11:40:09] <syoma> rmatte: I make a piece of software (perl script)

[04-Nov-2011 11:40:40] <syoma> and that piece it's integrated on zenoss with a template

[04-Nov-2011 11:41:27] <rmatte> right, a command datasource, ok

[04-Nov-2011 11:41:30] <rmatte> ...and?

[04-Nov-2011 11:41:41] <syoma> but in somes data source, the value it returnes is 0

[04-Nov-2011 11:42:23] <syoma> so, the rrd graph is not created on time interval

[04-Nov-2011 11:42:25] <jmp242> nachox: sorry - I was afk

[04-Nov-2011 11:42:43] <jmp242> Is the device that you're monitoring

[04-Nov-2011 11:42:50] <jmp242> in the /Server/Linux class?

[04-Nov-2011 11:42:54] <nachox> yes

[04-Nov-2011 11:42:55] <syoma> I need create an treshold when the rrd graph is not created...

[04-Nov-2011 11:43:06] <jmp242> hmmm, I don't use ubuntu

[04-Nov-2011 11:43:25] <jmp242> but I think you need to look at if the template is looking for a different OID than ubuntu net-snmp is returning

[04-Nov-2011 11:43:30] <syoma> rmatte: do you have some idea?

[04-Nov-2011 11:44:21] <rmatte> nachox: you probably don't have net-snmp configured properly on the server.  net-snmp defaults to a paranoid security level which only allows you to poll a limited number of OIDs.  You need to reconfigure it.

[04-Nov-2011 11:44:56] <syoma> :/

[04-Nov-2011 11:44:59] <rmatte> syoma: I still don't understand honestly...

[04-Nov-2011 11:45:08] <nachox> rmatte, i used a very simplistic configuration. i dont know what oid zenoss is looking for though

[04-Nov-2011 11:45:08] <rmatte> you want to set a minimum required value before an RRD gets created?

[04-Nov-2011 11:45:20] <rmatte> nachox: have you looked in the template?

[04-Nov-2011 11:45:36] <nachox> i just noticed the set of numbers in the template

[04-Nov-2011 11:45:42] <rmatte> or are you talking about in the actual processors component section?

[04-Nov-2011 11:46:01] <syoma> rmatte: I upload an image, please wait... and thank for your collaboration

[04-Nov-2011 11:46:45] <rmatte> nachox: in any case, I just know that a fresh install of net-snmp with default template will not present any of the performance values like cpu, memory, load average

[04-Nov-2011 11:47:15] <nachox> snmpwalk -c public -v1 localhost ssCpuIdle

[04-Nov-2011 11:47:15] <nachox> UCD-SNMP-MIB::ssCpuIdle.0 = INTEGER: 97

[04-Nov-2011 11:47:31] <rmatte> ok, so it's obviously getting it

[04-Nov-2011 11:47:38] <rmatte> you are looking at the actual graphs section right?

[04-Nov-2011 11:47:43] <rmatte> not under processors

[04-Nov-2011 11:48:14] <nachox> i was looking under processors

[04-Nov-2011 11:48:23] <rmatte> yeh, there are no graphs there, that's normal

[04-Nov-2011 11:48:25] <syoma> rmatte:

[04-Nov-2011 11:48:42] <rmatte> click on the actual graphs component on the left

[04-Nov-2011 11:48:49] <rmatte> and you'll see memory/cpu/load average

[04-Nov-2011 11:48:56] <nachox> rmatte, isnt that odd? i mean, i would expect to see those graphs there as well

[04-Nov-2011 11:49:08] <rmatte> no it's not odd because Zenoss doesn't come with a template for those

[04-Nov-2011 11:49:12] <rmatte> you could make one if you wanted to

[04-Nov-2011 11:49:34] <syoma> rmatte: you can view the image?

[04-Nov-2011 11:49:39] <rmatte> I just don't know what performance values you could pull from a per processor

[04-Nov-2011 11:49:45] <rmatte> since what you want to really pull is per core

[04-Nov-2011 11:49:52] <rmatte> and processors shows physical processors, not cores

[04-Nov-2011 11:50:05] <rmatte> syoma: yes

[04-Nov-2011 11:50:20] <rmatte> so I see the graph, but I still don't understand your question

[04-Nov-2011 11:50:21] <nachox> oh, i see, well, thanks for clearing up my confusion

[04-Nov-2011 11:50:38] <rmatte> no prob

[04-Nov-2011 11:50:41] <syoma> how can alert when the data source Nav_Indirecta not graph

[04-Nov-2011 11:50:45] <syoma> ?

[04-Nov-2011 11:51:36] <rmatte> well, there are two scenarios...

[04-Nov-2011 11:51:51] <rmatte> either no value is returned at all in which no RRD is generated and a debug event is created

[04-Nov-2011 11:51:59] <rmatte> or your script returns 0 and it creates the RRD

[04-Nov-2011 11:52:23] <syoma> the script returns 0

[04-Nov-2011 11:52:32] <syoma> but the thrshold not work for me

[04-Nov-2011 11:52:38] <rmatte> ok, and what's the minimum value that it can be?

[04-Nov-2011 11:52:43] <syoma> 0

[04-Nov-2011 11:52:46] <syoma> :X

[04-Nov-2011 11:52:48] <rmatte> no nocan it be: 0.01

[04-Nov-2011 11:52:59] <rmatte> I mean the minimum value that it SHOULD be besides 0

[04-Nov-2011 11:53:04] <rmatte> how low can it go basically

[04-Nov-2011 11:53:24] <rmatte> Is it possible for the value to be 0.01, or does it start at 1?

[04-Nov-2011 11:54:05] <rmatte> ah it's bytes per second

[04-Nov-2011 11:54:11] <rmatte> so 1 would be the minimum

[04-Nov-2011 11:54:25] <syoma> ah Ok

[04-Nov-2011 11:54:34] <rmatte> ok, so just create a MinMax threshold, and instead of setting a max value, set a min value of 1

[04-Nov-2011 11:54:41] <rmatte> which means it has to be at least 1 or it'll alert

[04-Nov-2011 11:54:46] <rmatte> meaning it'll alert if it's 0 or less

[04-Nov-2011 11:54:55] <syoma> thank you man

[04-Nov-2011 11:54:58] <rmatte> np

[04-Nov-2011 11:55:05] <enkrypt> I have disabled all of the Software modelers on all templates, and I wanted to delete all existing Software components from zenoss. I do this by looping over and manage_deleteComponent() and commit(). but apparently this doesn't delete the things?

[04-Nov-2011 11:56:05] <rmatte> enkrypt:

[04-Nov-2011 11:56:13] <rmatte> old thread with a solution to that

[04-Nov-2011 11:58:09] <rmatte> >>> for device in dmd.Devices.getSubDevices():

[04-Nov-2011 11:58:09] <rmatte> ...

[04-Nov-2011 11:58:09] <rmatte> ...

[04-Nov-2011 11:58:09] <rmatte> >>> commit()

[04-Nov-2011 11:58:48] <enkrypt> doing that now I had a similar approach for deleting all existing routes, and that DID work with manage_deleteComponent(). but apparently software needs a different approach

[04-Nov-2011 11:59:37] <enkrypt> I disabled route + software modeling (because the topology feature was causing problems due to my VRF routing, and because I've got performance issues in general that I'm trying to cure with Zack Salinas)

[04-Nov-2011 12:00:01] [disconnected at Fri Nov  4 12:00:01 2011]

[04-Nov-2011 12:00:02] [connected at Fri Nov  4 12:00:02 2011]

[04-Nov-2011 12:00:14] [zenoss-logger (logger bot) has joined #zenoss]

[04-Nov-2011 12:04:25] <rmatte> I see

[04-Nov-2011 12:05:28] <enkrypt> rmatte: what is the equivalent for Networks? because I've already removed the network relationship for each asset, but the networks themselves still "exist" in dmd.Networks

[04-Nov-2011 12:28:55] <rmatte> I'm not sure off the top of my head

[04-Nov-2011 12:30:03] <rmatte> looks like it should be dmd.Networks

[04-Nov-2011 12:30:23] <rmatte> though you'd probably have to use getSubObjects

[04-Nov-2011 12:31:51] <rmatte> something like...

[04-Nov-2011 12:32:27] <rmatte> for net in dmd.Networks.getSubObjects(): dmd.Networks.delObject(net)

[04-Nov-2011 12:32:34] <rmatte> though I don't know if delObject is the proper function

[04-Nov-2011 12:47:31] <Peanut> Hi folks - I just did an 'apt-get install zenoss' after adding the repository, but the package install seems stuck at 'Setting up zenoss-stack (3.2.1-0) and uses 100% cpu in .python.bin :-(

[04-Nov-2011 12:48:51] <Peanut> And I'm apparently not the first one to run into this problem, as someone reported the same problem on Ubuntu a month ago.

[04-Nov-2011 13:05:50] <rmatte> Peanut: it's not stuck

[04-Nov-2011 13:06:01] <rmatte> it's setting up the zope database

[04-Nov-2011 13:06:12] <rmatte> You need to be patient

[04-Nov-2011 13:06:15] <rmatte>

[04-Nov-2011 13:06:57] <rmatte> (I assume you mean stuck as in it's sitting at what appears to be 100% on the progress bar?)

[04-Nov-2011 13:17:19] <jb> hey rmatte

[04-Nov-2011 13:17:28] <jb> you know that transform you helped me with the other day?

[04-Nov-2011 13:17:50] <rmatte> yes?

[04-Nov-2011 13:17:53] <jb> it seems to be throwing an error. .Problem with line 7: except socket.herror:

[04-Nov-2011 13:18:05] <jb> sec

[04-Nov-2011 13:18:09] <rmatte> do you know what error it's throwing exactly?

[04-Nov-2011 13:18:10] <rmatte>

[04-Nov-2011 13:18:49] <jb> btw the small transform edit box is annoyingly small

[04-Nov-2011 13:19:08] <rmatte> that's why I copy transforms in to a notepad type app to edit them

[04-Nov-2011 13:19:11] <rmatte> then paste them back

[04-Nov-2011 13:19:14] <rmatte> best way to do it

[04-Nov-2011 13:19:30] <jb>

[04-Nov-2011 13:19:47] <rmatte> I already see the problem

[04-Nov-2011 13:19:51] <jb> hm

[04-Nov-2011 13:20:01] <rmatte> the transform I gave you had "import socket" for a good reason

[04-Nov-2011 13:20:09] <rmatte> because it's required for socket.herror to be recognized

[04-Nov-2011 13:20:17] <jb> oh, I must have missed that?

[04-Nov-2011 13:20:48] <jb> should I just use "import socket" instead of the "from socket import gethostbyaddr" ?

[04-Nov-2011 13:20:59] <rmatte>

[04-Nov-2011 13:21:06] <jb> k

[04-Nov-2011 13:21:19] <rmatte> you need to just use import socket and change gethostbyaddr to socket.gethostbyaddr

[04-Nov-2011 13:21:48] <jb> thanks

[04-Nov-2011 13:21:53] <rmatte> no prob

[04-Nov-2011 13:25:09] <jmp242> rmatte: did you see / have time to look at:

[04-Nov-2011 13:25:09] <jmp242> BTW - Zenoss programming wizzes: two threads that I can't help with:

[04-Nov-2011 13:25:10] <jmp242>           jmp242

[04-Nov-2011 13:25:10] <jmp242>           jmp242

[04-Nov-2011 13:25:10] <jmp242>           jmp242          Though the second one - maybe he could go into Zope and delete it, but I'm afraid of it breaking relationships and making things worse

[04-Nov-2011 13:25:55] <rmatte> jmp242: the first one I'm really not sure of, it's at the point where I'd need to get in there myself and trial and error it

[04-Nov-2011 13:26:05] <rmatte> database issues like that are always a total pain to troubleshoot

[04-Nov-2011 13:26:17] <Peanut> rmatte: thanks for your reply, just -how- patient would I have to be?

[04-Nov-2011 13:26:21] <billc> having trouble resetting the admin password as zenoss use w/ zenpass tool.  getting from the script.

[04-Nov-2011 13:26:36] <billc> : No such file or directory

[04-Nov-2011 13:26:36] <billc> /usr/local/zenoss/zenoss/bin/ line 42: import: command not found

[04-Nov-2011 13:26:36] <billc> /usr/local/zenoss/zenoss/bin/ line 44: try:: command not found

[04-Nov-2011 13:26:36] <billc> /usr/local/zenoss/zenoss/bin/ line 45: from: command not found

[04-Nov-2011 13:26:48] <rmatte> Peanut: depends on the hardware, usually takes between 15 and 30 minutes

[04-Nov-2011 13:26:49] <billc> Ideas oh gods of zenosss....

[04-Nov-2011 13:26:49] <jb> whats the script's interpreter?

[04-Nov-2011 13:27:12] <rmatte> Peanut: is it sitting at a progress bar showing 100%?

[04-Nov-2011 13:27:19] <jb> head -n1

[04-Nov-2011 13:27:27] <billc> zenpass is #! /usr/bin/env bash

[04-Nov-2011 13:28:04] <rmatte> billc: go in zendmd and do this:

[04-Nov-2011 13:28:05] <rmatte> app.acl_users.userManager.updateUserPassword('admin', 'newPassword')

[04-Nov-2011 13:28:06] <rmatte> commit()

[04-Nov-2011 13:28:24] <Peanut> rmatte: I'm doing an 'apt-get install' on Debian, there's no progress bar.

[04-Nov-2011 13:28:49] <rmatte> Peanut: you might be better off just using the stack installer and forgetting about the .deb.  The .deb is just the stack installer wrapped up anyways.

[04-Nov-2011 13:29:06] <jb> or just install EL

[04-Nov-2011 13:29:10] <rmatte> I've never personally tried installing from the deb since the stack is so nice to work with

[04-Nov-2011 13:30:05] <Peanut> rmatte: I'm completely new to zenoss, we're evaluating it against zabbix and a few others, and I have to agree with the 'nice to work with' so far :-)

[04-Nov-2011 13:30:10] <rmatte> though it could be setting up the database and the deb is just suppressing the output

[04-Nov-2011 13:30:13] <rmatte> I really don't know

[04-Nov-2011 13:30:49] <Peanut> I'm not the first one to run into that problem though, so that might actually hurt the adoption of zenoss.

[04-Nov-2011 13:30:51] <rmatte> I've tried zabbix and Zenoss blows it away

[04-Nov-2011 13:31:25] <rmatte> What I would suggest is that you download the latest .bin (the stack installer), chmod +x it and run it

[04-Nov-2011 13:31:31] <rmatte> it's a super painless install

[04-Nov-2011 13:32:04] <rmatte> trust me, you don't want automated updates with Zenoss.  Enough things change between versions lately that you'll want to do your upgrades by hand...

[04-Nov-2011 13:32:29] <rmatte> also never upgrade production systems immediately following a new release since bugs are generally found/fixed in a second release weeks later

[04-Nov-2011 13:32:43] <jb> im sure you are really encouraging him to stick with zenoss

[04-Nov-2011 13:32:47] <rmatte> though you can use the "zenpatch" command to easily apply patches to bugs as they come out

[04-Nov-2011 13:33:01] <Peanut> Well, if the .deb were part of a released Debian version, then the only changes would be backported security fixes, which sounds just great. Thanks for your warnings regarding the stability of zenoss though.

[04-Nov-2011 13:33:35] <rmatte> It's not a warning regarding the stability of the application itself, it's regarding actual releases

[04-Nov-2011 13:33:40] <rmatte> when everything is working well it's fantastic

[04-Nov-2011 13:34:01] <rmatte> Development is just progressing at such a rapid pace that there's bound to be bugs

[04-Nov-2011 13:34:16] <rmatte> but like I said, they are generally fixed relatively quickly

[04-Nov-2011 13:34:24] <rmatte> so if you time your upgrades right there's no issue

[04-Nov-2011 13:38:21] <rmatte> jmp242: that second post is insane, I have no idea how they even got to the point of having commas in their device names.

[04-Nov-2011 13:39:55] <rmatte> jmp242: If a rename isn't working, there's really not much I can think of to fix that without breaking a whole bunch of stuff

[04-Nov-2011 13:40:21] <rmatte> jmp242: That shouldn't even be possible though, so would be worthwhile having them open a bug ticket for it

[04-Nov-2011 13:40:41] <jmp242> Ok

[04-Nov-2011 13:40:43] <jmp242> lol

[04-Nov-2011 13:40:52] <jmp242> zenmagets on youtube - exploding tops

[04-Nov-2011 13:40:56] <jmp242> crazy

[04-Nov-2011 13:41:01] <rmatte> ?

[04-Nov-2011 13:41:09] <jmp242>

[04-Nov-2011 13:41:21] <rmatte> oh, those things

[04-Nov-2011 13:41:29] <jmp242> I had no idea such a thing was possible

[04-Nov-2011 13:42:40] <rmatte> lol nice

[04-Nov-2011 13:43:36] <jmp242> the first time in a long time my mind's been blown by magnets lol

[04-Nov-2011 13:43:59] <rmatte> If you think that's cool, check this out

[04-Nov-2011 13:44:01] <rmatte>

[04-Nov-2011 13:44:14] <jmp242> so what did you think of the zenpack wiki edit with top and bottom lists

[04-Nov-2011 13:44:36] <rmatte> I honestly haven't even had time to check it out hehe

[04-Nov-2011 13:46:18] <jmp242> That video is insane

[04-Nov-2011 13:46:22] <rmatte> yeh

[04-Nov-2011 13:46:25] <rmatte> really cool

[04-Nov-2011 13:59:24] <rmatte> I'll check the ZenPack page in a minute, just finishing something up

[04-Nov-2011 14:14:03] <rmatte> man, rundeck is a godsent

[04-Nov-2011 14:14:13] <rmatte> makes managing my 15 Zenoss boxes a lot easier

[04-Nov-2011 14:17:04] <rmatte> actually, up to 16 now, but back down to 15 in a week

[04-Nov-2011 14:20:33] <nachox> its like puppet but more web'ish?

[04-Nov-2011 14:21:39] <nachox> i'd love some zenpack that does what whatsup flow monitor does

[04-Nov-2011 14:22:22] <d4gg3r> hey guys, I have installed zenoss via apt repo and installed and configured snmpd to give read-only access to zenoss via public community string

[04-Nov-2011 14:22:30] <d4gg3r> but graphs are always empty

[04-Nov-2011 14:22:30] <nachox> i'm using a portlet with ntop right now, but it' kind of broken

[04-Nov-2011 14:22:57] <d4gg3r> any idea how to debug this?

[04-Nov-2011 14:23:30] <nachox> have you checked zenoss is actually getting the data?

[04-Nov-2011 14:23:36] <nachox> what graphs are you seeing?

[04-Nov-2011 14:24:19] <d4gg3r> i go to devices and select my server (

[04-Nov-2011 14:24:22] <d4gg3r> and graphs are

[04-Nov-2011 14:24:37] <d4gg3r> load, cpu,...

[04-Nov-2011 14:24:48] <d4gg3r> memory

[04-Nov-2011 14:24:50] <d4gg3r> and IO

[04-Nov-2011 14:25:12] <d4gg3r> how can I check if zenoss is getting the data?

[04-Nov-2011 14:25:54] <nachox> in your device, look in the lower left corner, and you'll see a commands option

[04-Nov-2011 14:25:58] <nachox> pick snmpwalk

[04-Nov-2011 14:26:47] <d4gg3r> okay

[04-Nov-2011 14:26:55] <d4gg3r> pastebin the output?

[04-Nov-2011 14:27:20] <d4gg3r>

[04-Nov-2011 14:28:09] <rmatte> nachox: all it does is allow bulk command execution across devices, and yeh, web based

[04-Nov-2011 14:28:30] <rmatte> you can setup scheduled jobs and stuff with it too

[04-Nov-2011 14:28:38] <nachox> rmatte, there is no zenpack like the one i need, right?

[04-Nov-2011 14:29:43] <rmatte> d4gg3r: net-snmp has paranoid security settings by default, you need to modify the configuration to disable them

[04-Nov-2011 14:29:50] <rmatte> that's why you're missing that data

[04-Nov-2011 14:30:06] <rmatte> nachox: you'd have to describe what you're wanting more

[04-Nov-2011 14:30:15] <rmatte> I'm not familiar with "flow monitor"

[04-Nov-2011 14:30:20] <d4gg3r> rmatte: well, i try to play with the config file

[04-Nov-2011 14:31:03] <rmatte> d4gg3r:

[04-Nov-2011 14:31:10] <rmatte> use that config and modify what you need to

[04-Nov-2011 14:31:12] <rmatte> that'll work

[04-Nov-2011 14:31:23] <nachox> rmatte, and with ntop? it basically tells you who is using your network, connectiong to what and at what speed

[04-Nov-2011 14:32:06] <rmatte> nachox: that's netflow, no there are no ZenPacks for that

[04-Nov-2011 14:32:31] <nachox> if you have a proper cisco, then yes, its netflow

[04-Nov-2011 14:32:50] <rmatte> Well, Netflow, RTR, SAA, whatever you want to call it lol

[04-Nov-2011 14:33:07] <rmatte> actually sorry, those other two are sla prots, not flow prots

[04-Nov-2011 14:33:18] <rmatte> but yeh

[04-Nov-2011 14:33:23] <d4gg3r> rmatte: so, can I replace this config or have modify it btw/

[04-Nov-2011 14:33:23] <d4gg3r> ?

[04-Nov-2011 14:33:37] <rmatte> We use ManageEngine Netflow Analyzer to monitor Netflow enabled devices

[04-Nov-2011 14:33:54] <rmatte> d4gg3r: you need to modify certain lines in it, but you can replace your whole config with that

[04-Nov-2011 14:34:17] <rmatte> modify the trap lines, the community string lines, and the location/contact lines.

[04-Nov-2011 14:34:36] <rmatte> You can just comment out the trap lines if you don't want to use traps

[04-Nov-2011 14:35:15] <d4gg3r> just done, let me take a look

[04-Nov-2011 14:36:43] <d4gg3r> snmpwalk :

[04-Nov-2011 14:36:46] <d4gg3r> after changes

[04-Nov-2011 14:36:49] <d4gg3r> rmatte:

[04-Nov-2011 14:37:05] <rmatte> that walk is useless though since it's a system walk, that's always going to return data

[04-Nov-2011 14:37:36] <rmatte> hold on

[04-Nov-2011 14:38:22] <d4gg3r> ok

[04-Nov-2011 14:38:37] <rmatte> Try: snmpwalk -v1 -cpublic laLoadInt15

[04-Nov-2011 14:39:04] <rmatte> actually wait

[04-Nov-2011 14:39:23] <rmatte> snmpwalk -v1 -cpublic laLoadInt1

[04-Nov-2011 14:39:25] <rmatte> like that

[04-Nov-2011 14:39:32] <rmatte> you should get 3 lines back

[04-Nov-2011 14:40:09] <rmatte> do that from the commandline of the server

[04-Nov-2011 14:42:36] <d4gg3r> ok

[04-Nov-2011 14:43:09] <d4gg3r> laLoadInt1: Unknown Object Identifier (Sub-id not found: (top) -> laLoadInt1)

[04-Nov-2011 14:43:37] <rmatte> without the 1

[04-Nov-2011 14:43:42] <rmatte> snmpwalk -v1 -cpublic laLoadInt

[04-Nov-2011 14:43:45] <d4gg3r> same

[04-Nov-2011 14:43:46] <rmatte> sorry

[04-Nov-2011 14:43:50] <rmatte> ok, then it didn't work

[04-Nov-2011 14:43:56] <d4gg3r> yeah

[04-Nov-2011 14:44:07] <rmatte> actually...

[04-Nov-2011 14:44:19] <d4gg3r> i need to define those objects somewhere?

[04-Nov-2011 14:44:25] <rmatte> no no, wait one sec

[04-Nov-2011 14:44:29] <d4gg3r> or they are shipped with the default...

[04-Nov-2011 14:44:30] <d4gg3r> ok

[04-Nov-2011 14:44:31] <rmatte> you might not have proper mibs installed

[04-Nov-2011 14:44:57] <rmatte> snmpwalk -v1 -cpublic

[04-Nov-2011 14:44:59] <rmatte> try that

[04-Nov-2011 14:46:02] <d4gg3r> got the 3 lines

[04-Nov-2011 14:46:03] <d4gg3r> iso. = INTEGER: 1

[04-Nov-2011 14:46:03] <d4gg3r> iso. = INTEGER: 4

[04-Nov-2011 14:46:03] <d4gg3r> iso. = INTEGER: 5

[04-Nov-2011 14:46:09] <rmatte> then it's working

[04-Nov-2011 14:46:19] <rmatte> you should see data in Zenoss

[04-Nov-2011 14:46:21] <d4gg3r> so... whats wrong with zenoss

[04-Nov-2011 14:46:22] <rmatte> if you don't wait a bit

[04-Nov-2011 14:46:31] <rmatte> it takes 3 polling cycles to graph the data

[04-Nov-2011 14:46:36] <rmatte> 5 minutes per polling cycle, 15 minutes

[04-Nov-2011 14:46:58] <rmatte> you did apply the new config I gave you right?

[04-Nov-2011 14:47:04] <d4gg3r> yeah

[04-Nov-2011 14:47:10] <d4gg3r> and commented trap lines

[04-Nov-2011 14:47:13] <rmatte> ok, then that's what fixed it, just give Zenoss some time

[04-Nov-2011 14:47:25] <d4gg3r> and I am now searching for some way to reduce the polling cycle

[04-Nov-2011 14:47:42] <rmatte> make sure you update the contact/location lines in the config, otherwise your location will be "somewhere over the rainbow" I believe is what I set it to

[04-Nov-2011 14:47:42] <rmatte> lol

[04-Nov-2011 14:48:06] <d4gg3r> this is just for testing

[04-Nov-2011 14:48:07] <rmatte> ah, nah, I have it at "Set this to the location"

[04-Nov-2011 14:48:11] <rmatte> I should make it that though

[04-Nov-2011 14:48:12] <rmatte> k

[04-Nov-2011 14:48:23] <d4gg3r> when I deploy I will change those all

[04-Nov-2011 14:48:24] <d4gg3r>

[04-Nov-2011 14:48:36] <rmatte> Cool

[04-Nov-2011 14:48:46] <d4gg3r> so I'll be waiting

[04-Nov-2011 14:48:49] <d4gg3r> ...

[04-Nov-2011 15:12:21] <rmatte> This is so true:

[04-Nov-2011 15:12:29] <rmatte> actually blah, that won't work

[04-Nov-2011 15:12:30] <rmatte> hold on

[04-Nov-2011 15:12:37] <rmatte>

[04-Nov-2011 15:20:08] <rmatte> whoever made ctrl-w a key combination to close firefox should be shit

[04-Nov-2011 15:20:09] <rmatte> shot*

[04-Nov-2011 15:20:18] <rmatte> bad typo lol

[04-Nov-2011 15:20:30] <rmatte> but yeh, ctrl-w is to search in nano

[04-Nov-2011 15:20:40] <rmatte> I find myself accidentally doing it in Firefox, and it goes byebye

[04-Nov-2011 15:21:51] <hmp> i dont mind ctrl-w for tab closing, but as you say closing a browser when accidentally closing the last tab is just supid

[04-Nov-2011 15:24:09] <d4gg3r> rmatte: nothing plotted yet

[04-Nov-2011 15:24:48] <rmatte> d4gg3r: all your zenoss daemons are running yes?

[04-Nov-2011 15:25:08] <rmatte> d4gg3r: you have the correct SNMP string defined in zProperties?

[04-Nov-2011 15:25:25] <d4gg3r> root@hossein-ThinkPad-T420s:/etc/snmp# service zenoss-stack status

[04-Nov-2011 15:25:25] <d4gg3r> Daemon: zeoctl program running; pid=1796

[04-Nov-2011 15:25:26] <d4gg3r> Daemon: zopectl program running; pid=1829

[04-Nov-2011 15:25:26] <d4gg3r> Daemon: zenhub program running; pid=1872

[04-Nov-2011 15:25:26] <d4gg3r> Daemon: zenjobs program running; pid=1917

[04-Nov-2011 15:25:28] <d4gg3r> Daemon: zenping program running; pid=1997

[04-Nov-2011 15:25:30] <d4gg3r> Daemon: zensyslog program running; pid=2065

[04-Nov-2011 15:25:32] <d4gg3r> Daemon: zenstatus program running; pid=2042

[04-Nov-2011 15:25:34] <d4gg3r> Daemon: zenactions program running; pid=2154

[04-Nov-2011 15:25:36] <d4gg3r> Daemon: zentrap program running; pid=2307

[04-Nov-2011 15:25:38] <d4gg3r> Daemon: zenmodeler program running; pid=2312

[04-Nov-2011 15:25:40] <d4gg3r> Daemon: zenperfsnmp program running; pid=2392

[04-Nov-2011 15:25:42] <d4gg3r> Daemon: zencommand program running; pid=2446

[04-Nov-2011 15:25:44] <d4gg3r> Daemon: zenprocess program running; pid=2476

[04-Nov-2011 15:25:46] <d4gg3r> Daemon: zenwin program running; pid=2522

[04-Nov-2011 15:25:48] <d4gg3r> Daemon: zeneventlog program running; pid=2604

[04-Nov-2011 15:25:50] <d4gg3r> mysql already running

[04-Nov-2011 15:25:55] <rmatte> d4gg3r: you could have just said yes lol

[04-Nov-2011 15:25:58] <rmatte>

[04-Nov-2011 15:26:03] <d4gg3r> yes lol

[04-Nov-2011 15:26:04] <d4gg3r>

[04-Nov-2011 15:26:06] <rmatte> as the zenoss user do the following:

[04-Nov-2011 15:26:15] <rmatte> zenperfsnmp run -v10 -d

[04-Nov-2011 15:26:27] <rmatte> see if you can spot any errors

[04-Nov-2011 15:26:35] <rmatte> also, how do you know it's not plotting yet?

[04-Nov-2011 15:26:40] <rmatte> has no RRD been created?

[04-Nov-2011 15:26:46] <rmatte> is it staying at 'nan' values?

[04-Nov-2011 15:27:05] <d4gg3r> everything is nan

[04-Nov-2011 15:27:36] <rmatte> 'nan' is not a bad thing, it means it's actually getting some values, otherwise the RRD files would have never gotten created

[04-Nov-2011 15:27:48] <rmatte> but 'nan' should only last 3 polling cycles, 15 minutes

[04-Nov-2011 15:29:31] <d4gg3r> where should the rrd files be?

[04-Nov-2011 15:30:03] <d4gg3r> zenperfsnmp output

[04-Nov-2011 15:30:05] <d4gg3r>

[04-Nov-2011 15:30:41] <rmatte> RRD files get stored in $ZENHOME/perf/Devices/

[04-Nov-2011 15:30:47] <rmatte> for that device

[04-Nov-2011 15:31:05] <rmatte> or possibly $ZENHOME/perf/Devices/localhost

[04-Nov-2011 15:31:09] <rmatte> depending on the name of the device

[04-Nov-2011 15:32:05] <d4gg3r> there is a single rrd about sysUpTime

[04-Nov-2011 15:32:12] <d4gg3r> that I think is not relevant to snmp

[04-Nov-2011 15:34:04] <rmatte> yeh, it's not relevant, if there are no other RRDs then I have no idea how you're seeing "nan"

[04-Nov-2011 15:34:38] <rmatte> checking your zenperfsnmp output now...

[04-Nov-2011 15:35:54] <rmatte> that's so weird, it's like it's not even trying to collect the data

[04-Nov-2011 15:36:14] <rmatte> 2011-11-04 16:27:32,584 DEBUG zen.zenperfsnmp: Gathering performance data for

[04-Nov-2011 15:36:14] <rmatte> 2011-11-04 16:27:32,584 INFO zen.zenperfsnmp: Removing device localhost

[04-Nov-2011 15:36:25] <rmatte> there should have been a bunch of lines between those showing the values being gathered

[04-Nov-2011 15:37:15] <rmatte> It's definitely seeing the template because of: 2011-11-04 16:27:32,583 DEBUG zen.thresholds: Updating threshold ('high load', ('localhost', ''))

[04-Nov-2011 15:37:25] <rmatte> which is one of the thresholds in the template

[04-Nov-2011 15:37:36] <rmatte> you haven't changed anything in that template have you?

[04-Nov-2011 15:39:07] <rmatte> Try: snmpwalk -v1 -cpublic

[04-Nov-2011 15:39:18] <d4gg3r> no

[04-Nov-2011 15:39:24] <d4gg3r> i haven't changed

[04-Nov-2011 15:39:44] <d4gg3r> 4 values given

[04-Nov-2011 15:39:47] <d4gg3r> 4 lines

[04-Nov-2011 15:39:52] <rmatte> 2011-11-04 16:27:32,590 INFO zen.zenperfsnmp: Sent 0 OID requests

[04-Nov-2011 15:39:52] <rmatte> 2011-11-04 16:27:32,590 INFO zen.zenperfsnmp: Queried 0 devices

[04-Nov-2011 15:39:53] <rmatte> 2011-11-04 16:27:32,590 INFO zen.zenperfsnmp:   0 in queue still unqueried

[04-Nov-2011 15:39:58] <d4gg3r> snmpwalk -v1 -cpublic

[04-Nov-2011 15:40:00] <rmatte> It's not even trying to collect data

[04-Nov-2011 15:40:11] <d4gg3r> HOST-RESOURCES-MIB::hrProcessorLoad.768 = INTEGER: 10

[04-Nov-2011 15:40:11] <d4gg3r> HOST-RESOURCES-MIB::hrProcessorLoad.769 = INTEGER: 8

[04-Nov-2011 15:40:11] <d4gg3r> HOST-RESOURCES-MIB::hrProcessorLoad.770 = INTEGER: 5

[04-Nov-2011 15:40:11] <d4gg3r> HOST-RESOURCES-MIB::hrProcessorLoad.771 = INTEGER: 5

[04-Nov-2011 15:40:13] <rmatte> What class do you have the device in right now?

[04-Nov-2011 15:40:35] <d4gg3r> server/linux

[04-Nov-2011 15:41:20] <rmatte> It's almost like you have zSnmpMonitorIgnore enabled for that class, or for the device or something

[04-Nov-2011 15:42:07] <d4gg3r> for the device its false

[04-Nov-2011 15:42:18] <rmatte> It may be ignoring it because it's the localhost address

[04-Nov-2011 15:42:32] <rmatte> try changing the IP of the device to the IP of the actual network interface, not the loopback

[04-Nov-2011 15:43:25] <rmatte> then run zenperfsnmp again and post the output

[04-Nov-2011 15:46:45] <d4gg3r> ok

[04-Nov-2011 15:51:16] <d4gg3r> why dies the ip reset insists to use a dns lookup? :|

[04-Nov-2011 15:51:23] <d4gg3r> no manual ip change option?

[04-Nov-2011 15:53:40] <d4gg3r> rmatte: worked for the IP address

[04-Nov-2011 15:53:44] <d4gg3r> thank you!!!

[04-Nov-2011 15:53:48] <d4gg3r> im so happy!

[04-Nov-2011 15:53:49] <d4gg3r>

[04-Nov-2011 16:03:04] <InvidiKidwell> can someone help me with a question? How do I get zenoss to stop monitoring disk space threshold on my cdrom? I removed it from the device list yet I keep getting errors. I would prefer to stop monitoring it

[04-Nov-2011 16:08:01] <rmatte> no prob

[04-Nov-2011 16:08:29] <rmatte> and you can manually set an IP, you don't have to use a lookup

[04-Nov-2011 16:09:35] <rmatte> There's a Reset IP option in the bottom left hand menu when you're at a device's page

[04-Nov-2011 16:11:00] <InvidiKidwell> I added it manually

[04-Nov-2011 16:15:30] <InvidiKidwell> how will resetting the IP make it stop scanning the cdrom?

[04-Nov-2011 16:30:45] <rmatte> had he stuck around for 2 more minutes I would have answered his question

[04-Nov-2011 16:30:51] <rmatte> impatient people = fail

[04-Nov-2011 16:31:30] <rmatte> plus I wasn't even talking to him lol

[04-Nov-2011 16:32:11] <cymruman> i'm still here.. sorry.. I had to log on with my other laptop

[04-Nov-2011 16:35:26] <cymruman> rmatte.. can you please answer my question

[04-Nov-2011 16:36:22] <rmatte> yes

[04-Nov-2011 16:36:36] <rmatte> ok, so let me get it straight...

[04-Nov-2011 16:36:49] <rmatte> the filesystem no longer actually shows in the list but it's still monitoring it?

[04-Nov-2011 16:36:57] <rmatte> or does it keep coming back?

[04-Nov-2011 16:37:58] <cymruman> correct.. it is no longer in the filesystem list

[04-Nov-2011 16:38:09] <cymruman> but it still errors out

[04-Nov-2011 16:38:39] <rmatte> ok, let's try something...

[04-Nov-2011 16:38:48] <rmatte> remodel the device so that the filesystem comes back

[04-Nov-2011 16:38:54] <rmatte> let me know when it's back

[04-Nov-2011 16:40:17] <cymruman> ok.. its back i.e.  "/mnt/cdrom"

[04-Nov-2011 16:40:27] <rmatte> click on it and view the properties for it

[04-Nov-2011 16:40:35] <rmatte> tell me what "Type:" it is

[04-Nov-2011 16:41:07] <cymruman> type? where does it say that

[04-Nov-2011 16:41:21] <rmatte> when you click on the filesystem, you should see a dropdown that says "Graphs"

[04-Nov-2011 16:41:28] <rmatte> open that dropdown

[04-Nov-2011 16:41:36] <rmatte> there should be Properties (I'm doing this off memory)

[04-Nov-2011 16:41:51] <rmatte> I don't have a functional 3.x instance to use as reference at the moment.

[04-Nov-2011 16:42:09] <rmatte> It's either "Properties" or "Details"

[04-Nov-2011 16:42:25] <rmatte> and you should see Type in there

[04-Nov-2011 16:42:42] <cymruman> graphs isn't a drop down. it is part of the list on the left below components and software

[04-Nov-2011 16:42:56] <cgibbons> hrm

[04-Nov-2011 16:43:06] <rmatte> no no, I mean just below the top pane

[04-Nov-2011 16:43:15] <cgibbons> anyone know difference between the 3.0 and 4.0 API docs?

[04-Nov-2011 16:43:23] <rmatte> cgibbons: no idea

[04-Nov-2011 16:44:06] <rmatte> cymruman:

[04-Nov-2011 16:44:20] <rmatte> In this screenshot, the dropdown menu that says Display: with "Events" in it

[04-Nov-2011 16:44:24] <rmatte> that's the one I'm talking about

[04-Nov-2011 16:44:31] <rmatte> You select Details in there

[04-Nov-2011 16:45:45] <cymruman> ok.. type = fixed disk

[04-Nov-2011 16:46:11] <rmatte> weird that it shows up as that but ok, what's the name of the filesystem?

[04-Nov-2011 16:47:35] <cymruman> It doesn't show "name" only Mount Point

[04-Nov-2011 16:47:49] <rmatte> the mount point then

[04-Nov-2011 16:48:10] <cymruman> "/mnt/cdrom"

[04-Nov-2011 16:48:43] <rmatte> ok, so now... go to the Infrastructure page

[04-Nov-2011 16:48:56] <cymruman> ok

[04-Nov-2011 16:49:01] <rmatte> You'll see a button shaped like an arrow with "Details" written on it

[04-Nov-2011 16:49:03] <rmatte> click on it

[04-Nov-2011 16:49:09] <rmatte> then you'll see a new menu on the left

[04-Nov-2011 16:49:13] <rmatte> click on Properties

[04-Nov-2011 16:49:30] <rmatte> then on the Properties page, find zFileSystemMapIgnoreNames

[04-Nov-2011 16:49:34] <cymruman> Configuration Properties

[04-Nov-2011 16:49:50] <rmatte> yeh

[04-Nov-2011 16:49:56] <cymruman> ok

[04-Nov-2011 16:49:57] <rmatte> like I said, doing this off the top of my head

[04-Nov-2011 16:50:01] <rmatte> lol

[04-Nov-2011 16:50:20] <rmatte> Once you find zFileSystemMapIgnoreNames put the following in it: ^/mnt/cdrom

[04-Nov-2011 16:50:32] <rmatte> Then scroll down to the bottom of the page and click Save

[04-Nov-2011 16:50:56] <rmatte> Then remodel your device and the filesystem should disappear from the list

[04-Nov-2011 16:51:30] <cymruman> ok

[04-Nov-2011 16:51:52] <rmatte> Then to make extra sure that it's not still being monitored, restart the zenperfsnmp daemon...

[04-Nov-2011 16:51:59] <rmatte> and at the commandline as the zenoss user type zendmd

[04-Nov-2011 16:52:04] <rmatte> then when zendmd opens do...

[04-Nov-2011 16:52:06] <rmatte> reindex()

[04-Nov-2011 16:52:09] <rmatte> commit()

[04-Nov-2011 16:52:12] <rmatte> ctrl-d to quit

[04-Nov-2011 16:52:14] <rmatte> that should do it

[04-Nov-2011 16:52:24] <rmatte> (and prevent it from happening with other devices in the future)

[04-Nov-2011 16:52:45] <rmatte> If you need to add more filesystem names to ignore in the future you'd do something like:

[04-Nov-2011 16:53:06] <rmatte> ^(/mnt/cdrom|/some/other/mount)

[04-Nov-2011 16:53:18] <rmatte> It uses regex to do the matching

[04-Nov-2011 16:54:06] <rmatte> afk for a couple

[04-Nov-2011 16:54:09] <cymruman> ok.. but the hat belongs in the front

[04-Nov-2011 16:54:30] <rmatte> The hat means it matches anything that starts with

[04-Nov-2011 16:54:47] <rmatte> by moving it outside the ( it applies it to anything within the ()

[04-Nov-2011 16:54:57] <cymruman> ok, and in parenthesis for multiples?

[04-Nov-2011 16:55:06] <cymruman> ok

[04-Nov-2011 16:55:13] <cymruman> got it

[04-Nov-2011 16:55:14] <cymruman> thx

[04-Nov-2011 16:55:21] <rmatte> yeh, multiple you do ^(/one|/two|/three)

[04-Nov-2011 16:55:25] <rmatte> np

[04-Nov-2011 16:55:45] <cymruman> I will work on it

[04-Nov-2011 16:59:23] <rmatte> and if you want to be even more accurate...

[04-Nov-2011 16:59:34] <rmatte> ^/mnt/cdrom$

[04-Nov-2011 17:00:01] <cymruman> what does the $ do?

[04-Nov-2011 17:00:32] <rmatte> well, if you do ^/mnt/cdrom then anything that starts with /mnt/cdrom gets blocked... meaning /mnt/cdromblah or whatever

[04-Nov-2011 17:00:51] <rmatte> If you specify the $ at the end, then only an exact match of /mnt/cdrom will get blocked

[04-Nov-2011 17:01:07] <cymruman> oh ok

[04-Nov-2011 17:01:46] <rmatte> not really necessary in this case since ideally you want anything starting with /mnt/cdrom to be blocked regardless of what the name finishes with

[04-Nov-2011 17:02:07] <cymruman> true, but good for future reference

[04-Nov-2011 17:02:17] <rmatte> yup

[04-Nov-2011 17:02:30] <cgibbons> hurm

[04-Nov-2011 17:02:46] <cgibbons> so how about this one… can a zencommand parse multiple nagios check values at once?

[04-Nov-2011 17:03:01] <rmatte> cgibbons: not sure I get what you mean?

[04-Nov-2011 17:03:29] <cgibbons> like say i made a zencommand that was check_http; check_process; check_https (poor example, bear with me)

[04-Nov-2011 17:03:44] <cgibbons> or is the parser only smart enough to do one line of output from one chefk?

[04-Nov-2011 17:03:46] <cgibbons> er check

[04-Nov-2011 17:03:54] <rmatte> as long as you arrange the output properly it doesn't matter how many scripts you execute

[04-Nov-2011 17:04:22] <cgibbons> so if I've got say 8 check calls and each output is online, it could parse that and I could arrive with 8 data points?

[04-Nov-2011 17:04:22] <rmatte> you'd be best to make some sort of wrapper script that executes it all, then stored it in variables and outputs the results in nagios format at the end

[04-Nov-2011 17:04:54] <rmatte> All it cares about is the output, not where the data comes from

[04-Nov-2011 17:06:13] <rmatte> alright, I have some device commissions to do so I'm afk for a few

[04-Nov-2011 17:06:39] <rmatte> I'm on vacation for 2 weeks after today

[04-Nov-2011 17:31:23] <cgibbons> hurm

[04-Nov-2011 17:46:12] kocolosk is now known as kocolosk|away

[04-Nov-2011 18:34:39] <rmatte> Out of Office Notification Set: Check, 2 weeks off: Check, Time to party: hellz yeh!

[04-Nov-2011 18:38:15] kocolosk|away is now known as kocolosk

[04-Nov-2011 20:20:29] kocolosk is now known as kocolosk|away

[05-Nov-2011 00:00:01] [disconnected at Sat Nov  5 00:00:01 2011]

[05-Nov-2011 00:00:02] [connected at Sat Nov  5 00:00:02 2011]

[05-Nov-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]

[05-Nov-2011 11:21:36] <tightwork> What can I do about a node that constantly sends out "unable to read processes" .. "Process table up" .. Its one thing to auto resolve the event but the fact I thousands of emails for a single node.

[05-Nov-2011 12:00:01] [disconnected at Sat Nov  5 12:00:01 2011]

[05-Nov-2011 12:00:02] [connected at Sat Nov  5 12:00:02 2011]

[05-Nov-2011 12:00:16] [zenoss-logger (logger bot) has joined #zenoss]

[05-Nov-2011 12:48:20] <tightwork> I made a whole seperate class /Server/Extend for nsExtend, added my oid.. I can snmpwalk and snmpget from the zenoss and get my value from my extend script. Yet periodically will tell me oid is bad

[05-Nov-2011 12:53:00] <tightwork> Oh yeah, Saturday, screwed

[05-Nov-2011 13:02:32] <tightwork> thats really neat, basic functionality of a network monitoring system, adding snmp extends is like brain surgery

[05-Nov-2011 13:08:01] <tightwork> Im disappointed that I added a switch, scanned subnet and looking at the node it does not tell me what switch port it is on.

[05-Nov-2011 13:08:04] <tightwork> like DUH

[05-Nov-2011 13:08:22] <tightwork> derp

[05-Nov-2011 13:09:22] <tightwork> is there anything that works

[05-Nov-2011 20:18:11] kocolosk|away is now known as kocolosk

[05-Nov-2011 22:35:47] <SilverShot> kl'

[06-Nov-2011 00:00:01] [disconnected at Sun Nov  6 00:00:01 2011]

[06-Nov-2011 00:00:02] [connected at Sun Nov  6 00:00:02 2011]

[06-Nov-2011 00:00:16] [zenoss-logger (logger bot) has joined #zenoss]

[06-Nov-2011 12:00:01] [disconnected at Sun Nov  6 12:00:01 2011]

[06-Nov-2011 12:00:01] [connected at Sun Nov  6 12:00:01 2011]

[06-Nov-2011 12:00:11] [zenoss-logger (logger bot) has joined #zenoss]

[07-Nov-2011 00:00:01] [disconnected at Mon Nov  7 00:00:01 2011]

[07-Nov-2011 00:00:01] [connected at Mon Nov  7 00:00:01 2011]

[07-Nov-2011 00:00:13] [zenoss-logger (logger bot) has joined #zenoss]

[07-Nov-2011 06:37:13] <rcngeoff> Hi.  I've got some scripts which connect to a database to do some basic app monitoring.  Zenoss currently executes them via snmp extenstions, I'd like to integrate these directly into zenoss without having to use snmp exentsions.  Can anyone point me in the right direction to get started?

[07-Nov-2011 08:11:16] <jmp242> Uggg Monday - and DST change\

[07-Nov-2011 08:13:29] <enkrypt> rmatte: hi Ryan, are you there?

[07-Nov-2011 09:13:18] <rcngeoff_> Hi.  I've got some scripts which connect to a database to do some basic app monitoring.  Zenoss currently executes them via snmp extenstions, I'd like to integrate these directly into zenoss without having to use snmp exentsions.  Can anyone point me in the right direction to get started?

[07-Nov-2011 09:14:08] <jmp242> rcngeoff_: doing it via SNMP extensions actually isn'

[07-Nov-2011 09:14:11] <jmp242> t

[07-Nov-2011 09:14:14] <jmp242> a bad method

[07-Nov-2011 09:14:25] <jmp242> it's recommended by some forum members

[07-Nov-2011 09:14:43] <jmp242> you could use command based datasources, but you'll increase the load on the zenoss server

[07-Nov-2011 09:15:09] <jmp242> and probably keep the load about the same on the monitored devices, plus require logins etc or some method to get to the data

[07-Nov-2011 09:15:32] <rcngeoff_> @jmp242:  indeed, it has been working well for us, but we are scaling out to new datacenters and it would be more simple to keep the checks in zenoss, if possible.

[07-Nov-2011 09:17:24] jb__ is now known as 36DAAN1PL

[07-Nov-2011 09:18:47] <rcngeoff_> @jmp242:  thanks the info, that looks it will get me going

[07-Nov-2011 09:22:48] <jmp242> If you have your script

[07-Nov-2011 09:22:54] <jmp242> that's just connecting to a database and doing quereys

[07-Nov-2011 09:23:14] <jmp242> you could also look into the ODBC datasource zenpack. . .

[07-Nov-2011 09:23:15] <jmp242> also

[07-Nov-2011 09:23:36] <rcngeoff_> that is just what the script does, I'll look into that

[07-Nov-2011 09:24:10] <jmp242>

[07-Nov-2011 09:42:35] kocolosk is now known as kocolosk|away

[07-Nov-2011 09:42:54] <fragfutter> egor wrote a newer version of sql datasource

[07-Nov-2011 09:57:46] <jmp242> So should that supersede the ODBC one?

[07-Nov-2011 10:14:23] <fragfutter> jmp242: i would try to stay away from odbc. but i never looked into the code to see if one of those packs is better or nicer than the other.

[07-Nov-2011 11:24:36] 36DAAN1PL is now known as jb

[07-Nov-2011 12:00:01] [disconnected at Mon Nov  7 12:00:01 2011]

[07-Nov-2011 12:00:02] [connected at Mon Nov  7 12:00:02 2011]

[07-Nov-2011 12:00:16] [zenoss-logger (logger bot) has joined #zenoss]

[07-Nov-2011 12:20:22] <cgibbons> so… remind me. if you are adding a new component to a device, do you need to override the Device object as well?

[07-Nov-2011 12:28:56] froztbyt1 is now known as froztbyte

[07-Nov-2011 13:38:00] kocolosk|away is now known as kocolosk

[07-Nov-2011 14:06:11] <Jane_Curry> Hiya

[07-Nov-2011 14:15:22] <Jane_Curry> Anyone there exttended the Action "gear" menu from the device details panel??

[07-Nov-2011 14:16:58] <Jane_Curry> I can add to the gear action on the device listing panel...

[07-Nov-2011 14:34:22] <jb> anyone decent at regex?

[07-Nov-2011 14:34:31] <hmp> shoot

[07-Nov-2011 14:34:41] <jb> rather simple in theory

[07-Nov-2011 14:35:09] <jb> Certificate '' in file will expire on Fri Nov 11 23:59:59 2011 GMT

[07-Nov-2011 14:35:28] <jb> i basically need to match "Certificate %var in file %var will expire"

[07-Nov-2011 14:35:56] <hmp> so you just want to cut off the last part?

[07-Nov-2011 14:36:06] <jb> i just need to match that line

[07-Nov-2011 14:36:17] <jb> but yeah

[07-Nov-2011 14:38:46] <hmp> not sure what you mean, but something like "Certificate '\w+' in file \w+ will expire on" should do the trick

[07-Nov-2011 14:39:00] <hmp> now you just backreference the parts you need

[07-Nov-2011 14:39:25] <jb> i'm basically trying to parse this message out of syslog

[07-Nov-2011 14:39:33] <jb> so everytime there is a message similar to this, I need to know about it.

[07-Nov-2011 14:39:49] <jb> and I can do so by matching it using a regexp

[07-Nov-2011 14:39:51] <hmp> if these are syslog messages, the format should stay the same

[07-Nov-2011 14:40:00] <hmp> so you can use that up there

[07-Nov-2011 14:40:17] <jb> so, in essence, I just need to replace '' with a wildcard.

[07-Nov-2011 14:42:15] <hmp> sorry, not \w then, use something like [\w\.]

[07-Nov-2011 14:42:22] <hmp> couse you need the dots to mach

[07-Nov-2011 14:42:24] <hmp> match*

[07-Nov-2011 14:42:27] <jb> ah ok

[07-Nov-2011 14:42:33] <jb> in single quotes?

[07-Nov-2011 14:42:44] <hmp> you might want to escape them

[07-Nov-2011 14:42:50] <hmp> like \'[\w\.]+\'

[07-Nov-2011 14:42:57] <hmp> sorry, i'm more of a perl guy

[07-Nov-2011 14:43:05] <jb> thanks!  i'll give that a try.

[07-Nov-2011 14:43:18] <hmp> no problem

[07-Nov-2011 14:52:24] <cgibbons> grrr

[07-Nov-2011 16:31:57] <draagon> can I ask a question about 3.2.1...fresh install on centos 5.7, updated packages through yum, downloaded 64bit of zenoss and trying to start zenoss fails

[07-Nov-2011 16:33:50] <draagon> let it run for hours with this message:  Loading initial Zenoss objects into the Zeo database (this can take a few minutes)

[07-Nov-2011 16:34:30] <draagon> zeo log shows CRITICAL root restarting too frequently; quit

[07-Nov-2011 17:30:53] kocolosk is now known as kocolosk|away

[07-Nov-2011 18:20:15] kocolosk|away is now known as kocolosk

[07-Nov-2011 19:35:28] <wobblyonions> hi all

[07-Nov-2011 19:39:43] <qkit> morning

[07-Nov-2011 19:39:59] <qkit> guys when i installed the zenoss it get hang at Loading initial Zenoss objects into the Zeo database hang

[07-Nov-2011 19:40:14] <qkit> i have leave it run 24 hours and it still hang there.

[07-Nov-2011 20:05:37] <wobblyonions> guys is it possible to have notifiactions sent to specfic people in the event a devices goes down, but per device so say device A goes down people xyz are emailed but if its device b then people abc get emailed?

[07-Nov-2011 20:56:11] <tssupremacy> sure is, check Advanced > Users > admin > Alerting Rules

[07-Nov-2011 21:36:55] kocolosk is now known as kocolosk|away

[08-Nov-2011 00:00:01] [disconnected at Tue Nov  8 00:00:01 2011]

[08-Nov-2011 00:00:02] [connected at Tue Nov  8 00:00:02 2011]

[08-Nov-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]

[08-Nov-2011 08:37:06] <Rewt`_> will zenoss perform software license auditing?

[08-Nov-2011 08:37:51] <jmp242> Rewt`_: not that I know of

[08-Nov-2011 09:37:42] marc is now known as Guest46534

[08-Nov-2011 10:04:20] <humanix> hi there guys

[08-Nov-2011 10:04:24] <humanix> hi rmatte Hackman238

[08-Nov-2011 10:04:32] <humanix> how are you?

[08-Nov-2011 10:05:12] kocolosk|away is now known as kocolosk

[08-Nov-2011 10:05:28] <humanix> i have a doubt

[08-Nov-2011 10:05:53] <humanix> is there any zenpack that allows to filter graphics in order to see an exact date range?

[08-Nov-2011 10:06:14] <humanix> cause it has daily, hourly, weekly, yearly

[08-Nov-2011 10:06:22] <hmp> afaik, no

[08-Nov-2011 11:01:43] <Jane_Curry> Hiya - anyone there with experience writing Zenoss 3 GUI code????

[08-Nov-2011 11:02:37] <rocket_> Rewt`_: auditing in which manner?  It might be possible to extend zenoss to do something along those lines but it is definately not its core functionality

[08-Nov-2011 11:51:15] <jb> rocket_: hey

[08-Nov-2011 11:56:50] <jb> any python folks here?

[08-Nov-2011 11:56:55] <jb> trying to tokenize/split a string for a transform

[08-Nov-2011 12:00:01] [disconnected at Tue Nov  8 12:00:01 2011]

[08-Nov-2011 12:00:02] [connected at Tue Nov  8 12:00:02 2011]

[08-Nov-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]

[08-Nov-2011 12:09:17] <jb> trying to do something like this

[08-Nov-2011 12:09:17] <jb> evt.component =  getattr(evt, s.split('bigipNotifyObjMsg')[1], 'Unknown')

[08-Nov-2011 12:16:21] <Hackman238> >>> a="a.b.c.d"

[08-Nov-2011 12:16:21] <Hackman238> >>> import re

[08-Nov-2011 12:16:21] <Hackman238> >>> a.split('.')

[08-Nov-2011 12:16:21] <Hackman238> ['a', 'b', 'c', 'd']

[08-Nov-2011 12:16:21] <Hackman238> >>> s=a.split('.')

[08-Nov-2011 12:16:23] <Hackman238> >>> s[0]

[08-Nov-2011 12:16:26] <Hackman238> 'a'

[08-Nov-2011 12:16:28] <Hackman238> >>> s[1]

[08-Nov-2011 12:16:31] <Hackman238> 'b'

[08-Nov-2011 12:22:22] <whyzgeek> hi guys I am checking zenoss master server secure log and it seems all of the collectors ssh to that box for some reason, has anybody got any idea as why?

[08-Nov-2011 12:22:35] <whyzgeek> it is through zenoss user

[08-Nov-2011 12:29:40] <whyzgeek> ping

[08-Nov-2011 12:40:45] <Hackman238> whyzgeek: Thats odd. The collectors ssh to the master?

[08-Nov-2011 12:43:40] <whyzgeek> yep

[08-Nov-2011 12:43:46] <whyzgeek> very strange

[08-Nov-2011 12:45:18] <jmp242> hey Hackman238 how's it going?

[08-Nov-2011 12:46:45] <Hackman238> jmp242: Not good. Lots of migrations and moving around eating all resources.

[08-Nov-2011 12:47:10] <Hackman238> jmp242: Burning the candle at both ends and still only just enough time to get sev 1 items handled.

[08-Nov-2011 12:47:30] <whyzgeek> Hackman238: the more interesting bit is that all of the collectors does that not just the collector for master server

[08-Nov-2011 12:47:45] <whyzgeek> and I have several instances of zenoss which do exactly this

[08-Nov-2011 12:48:20] <Hackman238> whyzgeek: I'm not sure why the collectors would do that less they are fetching updates by rsync or something

[08-Nov-2011 12:48:50] <Hackman238> jmp242: Plus working with staff who are very upset over the new UI speed issues.

[08-Nov-2011 12:48:59] <whyzgeek> can you check if your instance does that?

[08-Nov-2011 12:49:22] <Hackman238> whyzgeek: Ours doesn't so far as I see. Core or Ent?

[08-Nov-2011 12:49:28] <jmp242> Oh, that sucks

[08-Nov-2011 12:49:29] <whyzgeek> Ent

[08-Nov-2011 12:49:51] <whyzgeek> very strange it is 2.5.2

[08-Nov-2011 12:50:03] <whyzgeek> no idea what triggers it

[08-Nov-2011 12:50:12] <Hackman238> jmp242: It does indeed. I project I'll be caught back up by Friday.

[08-Nov-2011 12:50:36] <Hackman238> jmp242: I do have your account, I jsut hadn't had a chance to email you.

[08-Nov-2011 12:50:48] <Hackman238> jmp242: A ZCA post, if still applicable, would be good.

[08-Nov-2011 12:51:16] <Hackman238> jmp242: Also the layer 2 dep pack is nearing completeion.

[08-Nov-2011 12:53:05] <Hackman238> Gotta run. Have to defend against staff with pitch forks.

[08-Nov-2011 12:53:50] <jmp242> Ok

[08-Nov-2011 13:27:36] <jb> Hackman238: do I need to import re?

[08-Nov-2011 13:28:47] <rocket_> hey jb

[08-Nov-2011 13:29:28] <jb> sup

[08-Nov-2011 13:32:17] <jb> rocket_: I updated that ticket.

[08-Nov-2011 13:33:18] <rocket_> jb: I am not clear .. by tomorrow do you mean wednesday? or today now?

[08-Nov-2011 13:33:52] <jb> i have some time now, if you do?

[08-Nov-2011 13:34:10] <rocket_> I am on a call at the moment .. multitasking

[08-Nov-2011 13:34:21] <rocket_> so it would have to wait for about an hour and a half

[08-Nov-2011 13:35:04] <jb> have another meeting at 2:45

[08-Nov-2011 13:35:07] <jb> so just ping me when you are done

[08-Nov-2011 13:37:45] <rocket_> sure .. in the mean time can you update the case with the number of events in the system and in the archive?

[08-Nov-2011 13:38:44] <jb> sweet, got the transform working.

[08-Nov-2011 13:38:51] <jb> yeah

[08-Nov-2011 13:38:58] <jb> there aren't many.

[08-Nov-2011 14:29:29] <Glacee> goodjob on the swift zenpack!

[08-Nov-2011 14:43:30] <dhopp1> Glacee:  swift zenpack?

[08-Nov-2011 14:49:48] <jmp242> random OT question - anyone found a screen saver or gnome configuration program that would enable features like : on RHEL?

[08-Nov-2011 14:53:18] kocolosk is now known as kocolosk|away

[08-Nov-2011 15:02:04] <Glacee> yes

[08-Nov-2011 15:02:11] <Glacee> openstack swift zenpack

[08-Nov-2011 15:08:08] <rocket_> Glacee: ?

[08-Nov-2011 15:32:28] <rcngeoff> hi folks, having a little trouble with the sql data source.. I have it installed and it seems to be running ok, it is just that when I test it, I see query run but get no output back

[08-Nov-2011 15:32:43] <rcngeoff> any pointers?

[08-Nov-2011 15:57:33] <rcngeoff> ah ha.. I figured it out.  I just hadn't added the data point

[08-Nov-2011 16:29:06] <blah37> Hi I am new to zenoss and having some issue with mysql zenpack can anyone help me please

[08-Nov-2011 16:40:27] <blah37> anyone?

[08-Nov-2011 16:47:32] <blah37> or can someone direct me to good tutorial or documentation others than the zenoss zite one?

[08-Nov-2011 17:04:26] <blah37> zen.Classifier: No classifier defined getting this when modeling my device using the mysql zenpack what does that mean?

[08-Nov-2011 17:10:35] kocolosk|away is now known as kocolosk

[08-Nov-2011 17:24:03] <blah37> zen.Classifier: No classifier defined can someone tell me what does that mean plz

[08-Nov-2011 17:41:17] kocolosk is now known as kocolosk|away

[08-Nov-2011 18:20:51] kocolosk|away is now known as kocolosk

[08-Nov-2011 18:59:48] kocolosk is now known as kocolosk|away

[08-Nov-2011 20:16:37] kocolosk|away is now known as kocolosk

[08-Nov-2011 22:46:04] kocolosk is now known as kocolosk|away

[08-Nov-2011 23:54:48] rcngeoff__ is now known as rcngeoff

[09-Nov-2011 00:00:01] [disconnected at Wed Nov  9 00:00:01 2011]

[09-Nov-2011 00:00:02] [connected at Wed Nov  9 00:00:02 2011]

[09-Nov-2011 00:00:12] [zenoss-logger (logger bot) has joined #zenoss]

[09-Nov-2011 01:33:54] <froztbyte> jmp242: I'm not sure if they still exist, but the old "distort" screensavers and such come to mind as an option

[09-Nov-2011 01:34:07] <froztbyte> since you could load the apps you want, and then just distort them all

[09-Nov-2011 01:34:21] <froztbyte> jmp242: alternatively, second X instance with just the limited things open?

[09-Nov-2011 03:52:19] <Tabmow> Is there anyway of having the Zenoss http checks adhering to the HTTP 1.1 spec?

[09-Nov-2011 05:10:26] bzed_ is now known as bzed

[09-Nov-2011 06:11:37] <daveloper> Tabmow: what do you mean.

[09-Nov-2011 06:11:52] <daveloper> ?

[09-Nov-2011 08:00:22] <enkrypt> welke/names

[09-Nov-2011 08:09:46] <syoma> hi !

[09-Nov-2011 08:09:53] <enkrypt> hiya

[09-Nov-2011 08:36:04] <jb> rocket?

[09-Nov-2011 08:36:53] <enkrypt> I am getting so tired of battling performance issues am doing this together with Zack S. I thought I had it for a while, had a period of 48hours of good performance

[09-Nov-2011 08:37:12] <enkrypt> I tried re-activing vmware esx infrastructure monitoring, performance stayed good for 12 hours

[09-Nov-2011 08:37:19] <enkrypt> but this morning: boom. workListLength at 300

[09-Nov-2011 08:37:53] <enkrypt> restarted everything since then, removed all .pickle files, removed vmware monitoring completely, but no go, zenhub keeps struggling at 100% cpu, and my workListLength is too high, causing gaps in the graphs

[09-Nov-2011 08:38:07] <enkrypt> really wished I knew how to fix this once and for all

[09-Nov-2011 08:38:16] <jmp242> More hardware?

[09-Nov-2011 08:38:24] <jmp242> more zenhub workers?

[09-Nov-2011 08:38:56] <enkrypt> the hardware is doing nothing as it is. zero load, almost no CPU usage (except for keeping 1 of the 8 cores at 100% CPU)

[09-Nov-2011 08:39:17] <enkrypt> zenhub workers are at 8 currently (with 8 physical cpu cores, 16 "hyperthreaded" cores)

[09-Nov-2011 08:39:34] <jmp242> shouldn't zenhub workers be n-1 of cores/

[09-Nov-2011 08:39:49] <enkrypt> I sent Zack an email asking him if he changed anything maybe, because for 48 hours I finally had splendid performance. but it's super fragile apparently

[09-Nov-2011 08:39:56] <jmp242> not that I really know how to help here

[09-Nov-2011 08:40:05] <enkrypt> jmp242: it used to be at 7 yeah. but I think Zack changed it

[09-Nov-2011 08:40:20] <enkrypt> I'm battling this issue for a month now, would like to get it sorted out

[09-Nov-2011 08:41:11] <jb> im having performance issues too :/

[09-Nov-2011 08:42:06] <enkrypt> I'm looking forward to a stable version of 4.1 or 4.2 so that we can migrate to that one, because the 3.x branch is damn tricky to get working correctly when you're talking about multiple thousands of assets

[09-Nov-2011 09:15:08] <tsener> hello

[09-Nov-2011 09:18:20] <enkrypt> iya

[09-Nov-2011 09:18:22] <enkrypt> +h

[09-Nov-2011 09:42:17] <enkrypt> in zendmd how do I give a device a different id?

[09-Nov-2011 09:49:58] <rocket_> jb pong

[09-Nov-2011 09:52:58] <rocket_> dev=find('device_name')

[09-Nov-2011 09:53:09] <rocket_> dev.renameDevice('new_device_name')

[09-Nov-2011 09:53:18] <rocket_> you may just want to update the title though

[09-Nov-2011 10:26:05] <jb> rocket_: sup

[09-Nov-2011 10:26:14] <jb> rocket_: got time now, or you want to do this later?

[09-Nov-2011 10:39:56] kocolosk|away is now known as kocolosk

[09-Nov-2011 10:40:28] <bseklecki> Anyone managing HP C3000 series blade enclosure w/ zenoss?

[09-Nov-2011 10:40:32] <bseklecki> s/managing/monitoring/g

[09-Nov-2011 11:35:33] <enkrypt> is there a way to clean the entire ZODB. my Data.fs has become big, and zeopack doesn't do much good. I would like to start with a clean slate. get rid of all discovered data, all monitored data, everything fresh. throw away all routes, Software things etc. can that be done, without losing the device configs and templates and so on ?

[09-Nov-2011 12:00:01] [disconnected at Wed Nov  9 12:00:01 2011]

[09-Nov-2011 12:00:02] [connected at Wed Nov  9 12:00:02 2011]

[09-Nov-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]

[09-Nov-2011 12:14:56] <nexex> bseklecki: a C7000

[09-Nov-2011 16:41:04] <fritz101> hello all.  i've a problem:  created a script and put it into a template to run against a machine.  it 'tests' just fine and returns a value: count=3 for example.  I added a datapoint named 'count'.  but no rrd's are being created and I'm not sure why.

[09-Nov-2011 17:18:30] kocolosk is now known as kocolosk|away

[09-Nov-2011 18:05:23] <designated> anyone here use shane scott's ip slao monitor zenpack?

[09-Nov-2011 18:05:30] <designated> ip sla*

[09-Nov-2011 18:37:28] <designated> nice to see this place is so active

[09-Nov-2011 19:03:52] <wobblyonions> guys , can you use Zenoss for sflow at all etc?

[10-Nov-2011 00:00:01] [disconnected at Thu Nov 10 00:00:01 2011]

[10-Nov-2011 00:00:01] [connected at Thu Nov 10 00:00:01 2011]

[10-Nov-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]

[10-Nov-2011 01:00:23] [disconnected at Thu Nov 10 01:00:23 2011]

[10-Nov-2011 12:00:02] [connected at Thu Nov 10 12:00:02 2011]

[10-Nov-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]

[10-Nov-2011 12:04:36] <superdug> the email alert template, where is that stored or if it's in a different spot everywhere, what's the filename called?

[10-Nov-2011 12:05:54] <Jane_Curry> jrh0090: I suspect its me that is confused - I don't really have all the correct vocabulary to have this conversation

[10-Nov-2011 12:07:23] <superdug> Jane_Curry:

[10-Nov-2011 12:07:34] <superdug> but it does not say, where ... to paste it

[10-Nov-2011 12:10:55] <ablew> anybody seen memory leaks in zentrap?

[10-Nov-2011 12:11:20] <superdug> no one knows how to change the format of the alert emails in zenoss?

[10-Nov-2011 12:12:12] <ablew> superdug: running 3.x?

[10-Nov-2011 12:13:01] <ablew> in 3.x you just go into the alerting rule and select "message" on the left side once you're editing the alert

[10-Nov-2011 12:18:54] <superdug> ablew: I assume you mean in the GUI, and I know I've seen the page you were looking at ... and it gave a pulldown of "email or page"

[10-Nov-2011 12:19:08] <superdug> I have no idea where that was though

[10-Nov-2011 12:19:50] <ablew> superdug: on the left of that there's a menu (the blue column) that says "edit, message, schedule"

[10-Nov-2011 12:19:54] <ablew> click message

[10-Nov-2011 12:20:13] <designated> anyone here use shane scott's ip sla monitor zenpack?

[10-Nov-2011 12:20:27] <superdug> how do I get to there though?, I'm sorry I've been mainly perusing the filesystem via ssh

[10-Nov-2011 12:20:34] <superdug> I have little experience with the gui

[10-Nov-2011 12:21:20] <ablew> superdug: it's under advanced->users->group->alerting rules

[10-Nov-2011 12:25:07] <designated> let me try a different question.  Does anyone here know of an ipsla zenpack that actually works with 3.2.0?

[10-Nov-2011 12:26:40] <simonjj> what's the error you are seeing ?

[10-Nov-2011 12:26:50] <superdug> ablew: thank you very much

[10-Nov-2011 12:27:20] <designated> simonjj, are you asking me?

[10-Nov-2011 12:27:30] <ablew> superdug: sure thing

[10-Nov-2011 12:42:10] <enkrypt> I seem to have deleted too much information... namely all dmd.Manufacturers any way to restore that? I've get zenbackups, but those just give me the big .fs file at once

[10-Nov-2011 12:42:59] <enkrypt> I've got zenbackups*

[10-Nov-2011 13:09:40] <enkrypt> Jane_Curry, Hackman238, jmp242 or rocket_: any idea how I can restore dmd.Manufacturers ? it seems I deleted too much while cleaning...

[10-Nov-2011 13:29:56] kocolosk|away is now known as kocolosk

[10-Nov-2011 19:06:39] kocolosk is now known as kocolosk|away

[10-Nov-2011 20:15:35] kocolosk|away is now known as kocolosk

[10-Nov-2011 22:21:19] <brockp> Where does zenoss store the pid it expects to see when monitoring a process?

[11-Nov-2011 00:00:01] [disconnected at Fri Nov 11 00:00:01 2011]

[11-Nov-2011 00:00:02] [connected at Fri Nov 11 00:00:02 2011]

[11-Nov-2011 00:00:17] [zenoss-logger (logger bot) has joined #zenoss]

[11-Nov-2011 05:10:19] <froztbyte> so I'm developing a zenpack for Meinberg timeservers

[11-Nov-2011 05:10:31] <froztbyte> anyone got an opinion on what classpath I should use?

[11-Nov-2011 05:11:21] <froztbyte> I'm currently thinking /Devices/Server/Time, although /Devices/Timeserver comes to mind as well since it's a racked appliance..

[11-Nov-2011 06:03:11] <ali3n0> I'd go for /Device/Server/Time if it's a "common" server, otherwise I'd go for the second

[11-Nov-2011 06:24:23] <ali3n0> is there a way to have a graph reporting the sum of two datapoints values without using Formula zenpack?

[11-Nov-2011 08:14:57] <froztbyte> ali3n0: I settled on /Device/Server/Time/Meinberg

[11-Nov-2011 08:15:12] <ali3n0> froztbyte: 

[11-Nov-2011 08:15:14] <froztbyte> but I could shift it up one level

[11-Nov-2011 08:15:18] <ali3n0> so it is a server

[11-Nov-2011 08:15:25] <froztbyte> dunno how much abstraction there is in other timeservers

[11-Nov-2011 08:15:36] <froztbyte> the plan is to post the zenpack publically later

[11-Nov-2011 08:15:42] <ali3n0> I see

[11-Nov-2011 08:15:51] <ali3n0> is it ntp based?

[11-Nov-2011 08:22:49] <froztbyte> it's a stratum1 NTP server

[11-Nov-2011 08:23:49] <froztbyte> ali3n0:

[11-Nov-2011 08:24:09] <froztbyte> clickthrough on the links

[11-Nov-2011 08:25:49] <ali3n0> froztbyte: nice

[11-Nov-2011 09:08:13] kocolosk is now known as kocolosk|away

[11-Nov-2011 09:30:24] kocolosk|away is now known as kocolosk

[11-Nov-2011 09:39:56] <tightwork> Anyone able to get the ApacheMonitor to graph worker processes correctly? The graphs just show nan

[11-Nov-2011 09:44:46] <froztbyte> I'm guessing that means it didn't manage to collect the data

[11-Nov-2011 09:45:04] <froztbyte> turn the appropriate collector's log level up to 11 (debug) and watch the log

[11-Nov-2011 09:45:09] <froztbyte> might need a collector restart as well

[11-Nov-2011 09:45:38] <froztbyte> ali3n0: uhm, "yes", but it's a bit hacky

[11-Nov-2011 09:45:50] <froztbyte> ali3n0: on the graph, go to the custom definitions tab

[11-Nov-2011 09:46:11] <froztbyte> ali3n0: you can add a rrdtool expression there, and in it do whatever you want as long as you understand rrdtool syntax

[11-Nov-2011 09:46:19] <froztbyte> the main graph's variables are still available

[11-Nov-2011 09:46:23] <ali3n0> I have to decrypt that stuff I guess

[11-Nov-2011 09:46:29] <ali3n0> I see, thanks for the hing

[11-Nov-2011 09:46:30] <ali3n0> hint

[11-Nov-2011 10:05:07] <tightwork> How can I increase the maximum number of routes? Exceeding at 500... I am polling a main edge router (yea probably half are junk)

[11-Nov-2011 10:05:52] <tightwork> Added a /30 to a interface the other day and brought up a old static route that was never removed and smashed another customer was super awesome

[11-Nov-2011 10:19:29] <Jane_Curry> hiya

[11-Nov-2011 10:26:50] <ali3n0> froztbyte: I've added this to my graph, doesn't seem to work. can you spot any error?

[11-Nov-2011 10:28:53] <jmp242> Morning all

[11-Nov-2011 10:28:58] <jmp242> and a friday morning at that

[11-Nov-2011 10:31:30] <Jane_Curry> its POETS day again

[11-Nov-2011 10:38:16] <jmp242>

[11-Nov-2011 10:39:22] <jmp242> Veterens day in the US, but we still work ...

[11-Nov-2011 10:41:31] <tightwork> Im looking at logs and tcpdump and zenoss isnt polling at all for the apache server-status with the Apache template bound to my device, of course graphs show nan.

[11-Nov-2011 10:42:10] <tightwork> I can use the -H host -p 80 -u '/server-status?auto' no problem though

[11-Nov-2011 11:04:22] <jmp242> back from the local network outage

[11-Nov-2011 11:06:00] <jmp242> Hey, any IRC / ZCA / whoever want to look at some forum posts that are trickier than I can help with?

[11-Nov-2011 11:07:30] <ali3n0> froztbyte: nevermind I sorted it out somehow

[11-Nov-2011 11:09:00] <Hackman238> jmp242: Morning

[11-Nov-2011 11:09:08] <Hackman238> jmp242: I'll take a glance.

[11-Nov-2011 11:09:24] <jmp242> hey - the long lost Hackman238

[11-Nov-2011 11:09:28] <jmp242> how's it going

[11-Nov-2011 11:09:30] <Hackman238> jmp242: Might be able to answer them between fires

[11-Nov-2011 11:09:39] <Hackman238> jmp242: Its been a long month.

[11-Nov-2011 11:10:01] <jmp242>

[11-Nov-2011 11:10:07] <jmp242> err

[11-Nov-2011 11:10:16] <jmp242> I'll just list some for anyone to jump in on

[11-Nov-2011 11:10:20] <jmp242>

[11-Nov-2011 11:11:01] <jmp242>

[11-Nov-2011 11:11:20] <jmp242>

[11-Nov-2011 11:11:31] <jmp242>

[11-Nov-2011 11:11:35] <jmp242> Not too bad, only 4

[11-Nov-2011 11:11:50] <jmp242> One solved it himself which was nice lol

[11-Nov-2011 11:14:38] <Hackman238> jmp242:

[11-Nov-2011 11:15:05] <jmp242> Aaaannnd we just are getting over a network outage from an overloaded /burned power controller

[11-Nov-2011 11:15:16] <jmp242> so get to watch our Zenoss as servers come back up

[11-Nov-2011 11:17:12] <Hackman238> jmp242: Fun fun

[11-Nov-2011 11:17:34] <johnnynoc> hello Hackman238

[11-Nov-2011 11:17:36] <johnnynoc> hello jmp242

[11-Nov-2011 11:18:04] <Hackman238> johnnynoc: Hey. How goes it?

[11-Nov-2011 11:18:37] <johnnynoc> going pretty well

[11-Nov-2011 11:18:46] <johnnynoc> setting up a little zenoss lab for myself

[11-Nov-2011 11:19:03] <johnnynoc> to better replicate what we have in prod..  and because i hosed my other dev instance

[11-Nov-2011 11:19:03] <johnnynoc> heheh

[11-Nov-2011 11:19:11] <johnnynoc> how have you been?

[11-Nov-2011 11:19:17] <johnnynoc> It's Friday!!!!   >:D

[11-Nov-2011 11:29:29] <Hackman238> johnnynoc: Glad to hear its been going well!

[11-Nov-2011 11:29:47] <Hackman238> johnnynoc: Running a bit ragged for about a month. It's been panifully busy. Fire after fire.

[11-Nov-2011 11:30:00] <jmp242> It's got to get better, right?

[11-Nov-2011 11:30:20] <Hackman238> jmp242: ...3 of 12 instances migrated

[11-Nov-2011 11:30:49] <jmp242> wow

[11-Nov-2011 11:30:53] <jmp242> lots of work then

[11-Nov-2011 11:31:44] <Hackman238> jmp242: Not just Zenoss stuff though, we're prepping to move to our new location, puppet undid reality for us a few times, lots of new contracts on my end

[11-Nov-2011 11:31:47] <johnnynoc> well, glad to know we're all busy

[11-Nov-2011 11:31:58] <johnnynoc> we could be occupying wall street

[11-Nov-2011 11:33:30] <tightwork> Uninstalled ApacheMonitor, Shutdown zenoss-stack, Upgraded to 3.2.1.. restarted zenoss-stack, installed ApacheMonitor zenpack, restarted again... graph still nan

[11-Nov-2011 11:34:23] <Hackman238> johnnynoc: LOL I'll skip that.

[11-Nov-2011 11:34:42] <johnnynoc> Hackman238: :-P

[11-Nov-2011 11:35:00] <Hackman238> tightwork: at the cli, zencommand run now -v10 -d 'yourdevice' and look to se if the apache command fails

[11-Nov-2011 11:35:17] <ali3n0> anyone here using postgres zenoss zenpack? I get frequent disconnection events but I'm quite sure it's not a network problem (any other ssh command is ok)

[11-Nov-2011 11:35:51] <jmp242> puppet uggg. Can be good can be a pain

[11-Nov-2011 11:38:19] <tightwork> Hackman238: says nothing about apache, I did bind the template to the device and from the UI I can see the template and the nan graph.

[11-Nov-2011 11:40:28] <johnnynoc> Hackman238: did I already ask you if you dressed up for halloween?

[11-Nov-2011 11:40:29] <Hackman238> tightwork: Is it using the http datasource?

[11-Nov-2011 11:40:29] <johnnynoc>

[11-Nov-2011 11:41:50] <Hackman238> johnnynoc: You did To work I wore a black suit with a bright red silk tie. Very frightening to people at rackspace. Nobody here wears suites or ties let alone high contrast.

[11-Nov-2011 11:42:05] <johnnynoc> hah

[11-Nov-2011 11:42:09] <johnnynoc> so many suits around here

[11-Nov-2011 11:42:10] <Hackman238> johnnynoc: I didnt get a chance to dress up and go out though.

[11-Nov-2011 11:42:16] <johnnynoc> ah, boo

[11-Nov-2011 11:42:18] <Hackman238> johnnynoc: Yep. NE, right?

[11-Nov-2011 11:43:02] <Hackman238> johnnynoc: Seems in Texas the only type of fabric known to people is denim

[11-Nov-2011 11:43:13] <johnnynoc> well, i'm in Chicago

[11-Nov-2011 11:43:17] <johnnynoc> so many suits becuase we're in the financial industry

[11-Nov-2011 11:43:29] <Hackman238> johnnynoc: Oh right right, I recall now.

[11-Nov-2011 11:43:29] <tightwork> Hackman238: source is for ${dev/manageIp} which I assume is a variable for the device and IP, The other field variable for /server-status?auto is also set.  Combining these for my host/server-status?auto I can get server-status from my browser.

[11-Nov-2011 11:43:34] <Hackman238> johnnynoc: Investment, right?

[11-Nov-2011 11:43:35] <johnnynoc> you know, i have to complain, i'm tired of mario and luigi costuimes

[11-Nov-2011 11:43:55] <Hackman238> johnnynoc: LOL!

[11-Nov-2011 11:44:10] <jmp242> Mmm, I grew out of holloween long ago

[11-Nov-2011 11:44:31] <Hackman238> johnnynoc: Some one should dress up as one of the less knwon characters. Like the guy in the cloud that tosses down mines.

[11-Nov-2011 11:44:35] <johnnynoc> jmp242: you should be in Grumpy Old Men 3

[11-Nov-2011 11:44:38] <johnnynoc>

[11-Nov-2011 11:44:43] <jmp242> Maybe

[11-Nov-2011 11:44:44] <johnnynoc> Hackman238: hell yea!

[11-Nov-2011 11:45:22] <Hackman238> tightwork: Sounds like this is indeed a zenoss problem with running the test

[11-Nov-2011 11:45:33] <tightwork> that ${dev/manageIp} is default for the zenpack datasource. I just put in the hostname of my web server and it  seems to work....

[11-Nov-2011 11:45:47] <Hackman238> tightwork: Yep, thats right

[11-Nov-2011 11:45:53] <tightwork> What should ${dev/manageIp} expand to exactly ?

[11-Nov-2011 11:45:58] <Hackman238> tightwork: ip address

[11-Nov-2011 11:46:03] <tightwork> hmm

[11-Nov-2011 11:46:12] <Hackman238> tightwork: less your using vhosts should be ok

[11-Nov-2011 11:46:36] <tightwork> its for the main host, monitoring the overal health of the shared server

[11-Nov-2011 11:46:50] <Hackman238> tightwork: should be ok then

[11-Nov-2011 11:47:05] <Hackman238> fromt he device screen press the gear icon -> push changes

[11-Nov-2011 11:48:23] <tightwork> hmm this is weird.. I can get to server-status from the IP also. Pretty sure I have a main IP set for the server and an additional interface for shared plus interfaces for unique ip if they are ssl

[11-Nov-2011 11:48:56] <tightwork> Id like to use the variable since I made a WebServer class with this template.

[11-Nov-2011 11:49:06] <tightwork> drag and drop goodness

[11-Nov-2011 11:50:12] <Hackman238> tightwork: Gotcha

[11-Nov-2011 11:50:32] <tightwork> I put the IP as well and it still works... overall seems hard coding the host and IP works but not by variable.

[11-Nov-2011 11:51:17] <Hackman238> tightwork: Lets try something

[11-Nov-2011 11:51:21] <Hackman238> tightwork: zendmd

[11-Nov-2011 11:51:35] <Hackman238> then d=dmd.Devices.findDevice('yourdevice')

[11-Nov-2011 11:51:42] <Hackman238> then d.manageIp

[11-Nov-2011 11:51:47] <Hackman238> Does it return the ip?

[11-Nov-2011 11:52:16] <dhopp> anybody know of a zenpack that will check a page and if it fails store the resulting page in a DB to keep a history of that page?

[11-Nov-2011 11:52:49] <tightwork> Hackman238: ooh, python shell nice... it does return the IP :-|

[11-Nov-2011 11:53:32] <Hackman238> tightwork: does d.title return anything?

[11-Nov-2011 11:53:41] <Hackman238> If so, is it the fqdn?

[11-Nov-2011 11:53:47] <tightwork> yes, and yes

[11-Nov-2011 11:54:02] <Hackman238> tightwork: Use ${dev/title}

[11-Nov-2011 11:55:14] <tightwork> Cool thank you so much

[11-Nov-2011 11:55:51] <Hackman238> tightwork: NP

[11-Nov-2011 12:00:01] [disconnected at Fri Nov 11 12:00:01 2011]

[11-Nov-2011 12:00:02] [connected at Fri Nov 11 12:00:02 2011]

[11-Nov-2011 12:00:16] [zenoss-logger (logger bot) has joined #zenoss]

[11-Nov-2011 12:09:30] <Jane_Curry> Anyone comment on significant changes to zentrap with 3.2 ????

[11-Nov-2011 12:10:02] <Jane_Curry> zentrap.log now seems to have lots of zen.collector.scheduler lines every 5 mins that never used to appear.....

[11-Nov-2011 12:31:42] <jmp242> Jane_Curry: I can't really, don't use it too much, but there was a forum post

[11-Nov-2011 12:31:54] <jmp242>

[11-Nov-2011 12:32:02] <jmp242> talking about issues with it maybe having a memory leak

[11-Nov-2011 12:34:36] <Hackman238> Jane_Curry: Replied to post

[11-Nov-2011 12:34:54] <jmp242> yea

[11-Nov-2011 12:34:57] <jmp242> I just saw that

[11-Nov-2011 12:35:01] <jmp242> I was eating lunch

[11-Nov-2011 12:35:09] <jmp242> oh well

[11-Nov-2011 12:35:55] <SEJeff> Has Service Dynamics 4.1 been released yet?

[11-Nov-2011 12:37:57] <Hackman238> SEJeff: Ent, yes

[11-Nov-2011 12:38:03] <Hackman238> SEJeff: 4.1.1 soon.

[11-Nov-2011 12:38:14] <SEJeff> Hackman238, Are you an employee?

[11-Nov-2011 12:38:43] <SEJeff> I'm looking at upgrading the latest 3.x directly to 4.1.x

[11-Nov-2011 12:38:45] <SEJeff> possible?

[11-Nov-2011 12:38:53] <Hackman238> SEJeff: Negative. I work for Rackspace hosting to handel Zenoss and I do Zenoss contracting.

[11-Nov-2011 12:39:07] <Hackman238> SEJeff: That will work.

[11-Nov-2011 12:39:58] <SEJeff> Perhaps I should wait until 4.1.1 is out

[11-Nov-2011 12:40:00] <SEJeff> Any idea how soon?

[11-Nov-2011 12:40:17] <SEJeff> And is there an announce list where we can subscribe for only release announcements?

[11-Nov-2011 12:41:24] <Hackman238> SEJeff: Unofficially, 4.1.1 sometime in Dec.

[11-Nov-2011 12:41:37] <Hackman238> SEJeff: Subject to change and my highly volitite memory

[11-Nov-2011 12:41:47] <Hackman238> *volatile

[11-Nov-2011 12:42:55] <Hackman238> SEJeff: email, Rusty. He'll be able to get you on the list.

[11-Nov-2011 12:44:02] <SEJeff> Will do

[11-Nov-2011 12:44:23] <Jane_Curry> Hackman238: Thanks for confirmation.  Don't have zenhub debug on but just turned it on - will report back on forum....

[11-Nov-2011 12:44:33] <Hackman238> Jane_Curry: Thansk Jane

[11-Nov-2011 12:44:34] <Jane_Curry> Do you have a ticket in for this???

[11-Nov-2011 12:44:48] <Hackman238> Jane_Curry: I do, its being worked now.

[11-Nov-2011 12:45:22] <Hackman238> Jane_Curry: If we figure that its a zenhub problem we'll take a look at the 3.x source and make sure to fix it where possible.

[11-Nov-2011 12:55:50] <Jane_Curry> Is there a 3.x ticket or should I open one??

[11-Nov-2011 13:01:11] <Hackman238> Jane_Curry: Its a 4.1 ticket I have open

[11-Nov-2011 13:10:32] <Hackman238> "No one really understands it[zope]. Like a hypercube or M theory. :)"

[11-Nov-2011 13:31:07] <cek> i'm using /Power/ApcPdu template to monitor our apc pdus and somehow, "PDU banks" graph is empty now

[11-Nov-2011 13:39:09] <Hackman238> cek: Never managed to get that pack to work

[11-Nov-2011 13:39:46] <cek> you know what's most pathetic . is that the screenshot on is showing "unknown" on graphs

[11-Nov-2011 13:40:14] <cek> but it should draw the load at least. i mean, i see it getting off pdus

[11-Nov-2011 13:44:50] <jmp242> Jane might comment . . .

[11-Nov-2011 13:45:14] <jmp242> Did you meet the dependencies?

[11-Nov-2011 13:46:17] <cek> yes, i met them in person. they were tidy and shy

[11-Nov-2011 13:47:48] <jmp242> ?? I mean : The APC PDU MIB needs to be available on target devices

[11-Nov-2011 13:48:14] <jmp242> I wonder if Jane wrote that pack or if the Author stuff is messed up

[11-Nov-2011 13:48:21] <jmp242> Her stuff generally work

[11-Nov-2011 13:48:22] <jmp242> s

[11-Nov-2011 13:51:00] <cek> yes dear, it's all there. it used to draw nice graph of overall load until i added overload notification rule

[11-Nov-2011 13:54:22] <jmp242> Notification rule?

[11-Nov-2011 13:54:32] <jmp242> do you mean you added a threshold to the datapoint / graph?

[11-Nov-2011 13:54:39] <cek> i think so

[11-Nov-2011 13:54:39] <jmp242> or are you talking about an alerting rule?

[11-Nov-2011 13:54:46] <cek> hodl on:)

[11-Nov-2011 14:00:36] <cek> well, i can't find that alerting rule

[11-Nov-2011 14:00:46] <cek> it would alert if load was near 14 amps

[11-Nov-2011 14:04:02] <cek> oh yeah, found it. in "monitoring templates" -> /power/apcpdu/apcpdubank

[11-Nov-2011 14:04:17] <cek> i added load threshold and it looks it stopped working

[11-Nov-2011 14:04:44] <jmp242> Ok, so it was adding a threshold

[11-Nov-2011 14:05:00] <jmp242> that does change the graph a little, though I don't know why it would stop working

[11-Nov-2011 14:08:14] <cek> i can even view the graph of inteface utilization, on a pdu! but the most important thing - power load is not working!

[11-Nov-2011 14:09:14] <jmp242> any log entries

[11-Nov-2011 14:09:20] <jmp242> that might be relevant...

[11-Nov-2011 14:10:52] <jmp242> how's the threshold defined?

[11-Nov-2011 14:11:05] <jmp242> did you add the line to the graph or only put the threshold in the datasource

[11-Nov-2011 14:19:08] <cek> alrighty, where should I look at ? what log file and for what line? yes, i just added the threshold, that's all

[11-Nov-2011 14:20:25] <cek> ./zenperfsnmp.log.1:98446:2011-11-04 16:21:18,347 WARNING zen.zenperfsnmp: Error reading value for "Object_Id_0" on xxx (oid . is bad)

[11-Nov-2011 14:20:37] <jmp242> If you remove the threshold, does the graph start coming back? (after 20 minutes?)

[11-Nov-2011 14:20:51] <jmp242> yea, that does look like such an issue

[11-Nov-2011 14:20:56] <jmp242> what device is that?

[11-Nov-2011 14:21:20] <cek> SNMPv2-SMI::enterprises.318. = No Such Instance currently exists at this OID

[11-Nov-2011 14:21:22] <jmp242> try an snmpget on that OID - it sounds like it may not be returning anything - but that would be separate from the threshold

[11-Nov-2011 14:21:23] <cek> SNMPv2-SMI::enterprises.318. = Gauge32: 106

[11-Nov-2011 14:21:39] <jmp242> did you edit the SNMP datapoint/

[11-Nov-2011 14:21:40] <jmp242> ?

[11-Nov-2011 14:21:42] <cek> now I wonder why did it add an ".1" post that value?:/

[11-Nov-2011 14:21:53] <cek> Im sure I didn't edit it. Is it safe to do it now?

[11-Nov-2011 14:22:00] <jmp242> I doubt it could hurt

[11-Nov-2011 14:22:02] <jmp242> try that

[11-Nov-2011 14:22:51] <cek> well, you know what:) here's what I have in OID field:  -- that is, without ".1". Why is snmpref.log showing it with ".1"?

[11-Nov-2011 14:24:58] <cek> oh, and it had a snowflake in component's icon - didn't pay attention to that

[11-Nov-2011 14:25:25] <jmp242> I wonder if it's reading some sort of SNMP table

[11-Nov-2011 14:25:39] <jmp242> Have you restarted Zenoss since this started? Or re-modeled the device?

[11-Nov-2011 14:25:53] <cek> yep, restarted a lot, remodelling happens automatically

[11-Nov-2011 14:26:10] <jmp242> well, it's supposed to, but sometimes it hangs - though a restart will fix that

[11-Nov-2011 14:26:12] <jmp242> mmm

[11-Nov-2011 14:26:23] <jmp242> I might suggest posting a thread in the Zenpacks forum

[11-Nov-2011 14:38:37] <cek> alright, it started collecting

[11-Nov-2011 14:41:00] <cek> now I want to have all those 4 pdu bank load graphs on one page. How do I do that?

[11-Nov-2011 14:49:11] <mloven> create a report?

[11-Nov-2011 15:26:44] <johnnynoc> cek what mloven said

[11-Nov-2011 15:26:45] <johnnynoc>

[11-Nov-2011 15:27:09] <cek> yeah mclovin being genious today

[11-Nov-2011 15:27:12] <johnnynoc> i have multi-graph reports if we want to see the graphs and then regular device reports to tell us cur, avg, max utilization of a pdu bank and the aggregate power

[11-Nov-2011 15:27:26] <cek> how do i set them up?

[11-Nov-2011 15:27:39] <cek> i didn't read the manual, but i guess even after reading it, i'll be stuck

[11-Nov-2011 15:27:56] <johnnynoc> you need to learn how to create reports

[11-Nov-2011 15:28:01] <johnnynoc> it's fairly easy, toy around, i'm sure you'll figure i tout

[11-Nov-2011 15:28:19] <cek> i have already toiled around

[11-Nov-2011 15:28:27] <cek> and i don't like what i see

[11-Nov-2011 15:28:32] <johnnynoc> the graph reports were easy, the custom device report took a little bit of work

[11-Nov-2011 15:29:09] <cek> the problem is that we've got 48 port switches and i gave up on 3rd "custom report whatever" component whatever addition of inteface bw graph

[11-Nov-2011 15:29:27] <cek> i just want them all on 1 page for a switch for example

[11-Nov-2011 15:31:45] <johnnynoc> let me test something on one of my switches

[11-Nov-2011 15:33:01] <johnnynoc> so, i chose my device

[11-Nov-2011 15:33:07] <johnnynoc> chose the 48 interfaces

[11-Nov-2011 15:33:09] <johnnynoc> chose the throughput graph

[11-Nov-2011 15:33:20] <johnnynoc> now i have a graph report for a single switch with graphs for all 48 interfaces on one page

[11-Nov-2011 15:33:34] <johnnynoc> sometimes when you're selecting the components the graph options don't always pop up

[11-Nov-2011 15:33:40] <johnnynoc> i had to choose each interface individually

[11-Nov-2011 15:33:48] <enkrypt> I've got a problem, I deleted something from dmd.Manufacturers which I shouldn't have deleted. namely productSearch. I already restored the manufacturers themselves, by doing "zenload -i /opt/zenoss/Products/ZenModel/data/manufacturers.xml' but my system is still messed up because I don't have productSearch. where/how can I restore that?

[11-Nov-2011 15:34:13] <johnnynoc> cek perhaps i'm misunderstanding something but i don't believe it should be too difficult

[11-Nov-2011 15:34:15] <cek> that's shitload of work.

[11-Nov-2011 15:34:19] <cek> okay, thanks

[11-Nov-2011 15:37:28] <enkrypt> is there some way I can do the initial data import again, that happens when you install zenoss? to restore the stuff from dmd.Manufacturers that I deleted accidentaly... need help :-s

[11-Nov-2011 15:41:47] <enkrypt> rmatte / roket_ / Hackman238 any ideas?

[11-Nov-2011 15:41:50] <enkrypt> rocket_ *

[11-Nov-2011 15:42:37] <Hackman238> cek: You'll want to code a template which iteratively shows the graphs for each device

[11-Nov-2011 15:42:56] <Hackman238> enkrypt: Howe long ago?

[11-Nov-2011 15:43:05] <enkrypt> could someone provide me with an export of their dmd.Manufacturers for example? preferably from an Enterprise install of 3.2.1

[11-Nov-2011 15:43:11] <enkrypt> Hackman238 yesterday

[11-Nov-2011 15:43:24] <Hackman238> enkrypt: check -> undo tab

[11-Nov-2011 15:43:51] <enkrypt> Hackman238 : unfortunately a zeopack happened in the meanwhile

[11-Nov-2011 15:43:55] <Hackman238> enkrypt: Undo is transactional though, so it might undo what has been done after the deletion which you do not wish to undo

[11-Nov-2011 15:44:02] <Hackman238> enkrypt: Ah

[11-Nov-2011 15:44:35] <enkrypt> Hackman238: I already recovered the manufacturers themselves from the .xml file with zenload, but I'm missing some important stuff like productSearch, which causes my modeling to fail

[11-Nov-2011 15:44:50] <Hackman238> enkrypt: Restore from a pack: zenoss stop; mv $ZENHOME/var/Data.fs $ZENHOME/var/Data.fs.whoops; mv $ZENHOME/var/Data.fs.old $ZENHOME/var/Data.fs; zenosss start

[11-Nov-2011 15:45:20] <Hackman238> enkrypt: productSearch can be rebuilt. Just a catalog, right?

[11-Nov-2011 15:45:38] <cek> Hackman238, indeed i might want to code a template - the multiselection of interfaces messed up the sorting

[11-Nov-2011 15:45:42] <enkrypt> Hackman238 I'm not sure what it is exactly, I think it's some kind of catalog or index yeah

[11-Nov-2011 15:45:58] <cek> now it goes fa0/1, fa0/10, fa0/11

[11-Nov-2011 15:46:07] <Hackman238> enkrypt: 1 sec

[11-Nov-2011 15:47:41] <Hackman238> enkrypt: Can you paste bin some of the trace around productSearch?

[11-Nov-2011 15:47:47] <Hackman238> enkrypt: IDK where that catalog is

[11-Nov-2011 15:47:59] <enkrypt> dmd.Manufacturers.productSearch

[11-Nov-2011 15:48:24] <Hackman238> enkrypt: ok. 1 sec

[11-Nov-2011 15:49:34] <enkrypt> I'm pretty sure that is all I'm missing

[11-Nov-2011 15:49:49] <enkrypt> but what would be even better is an export of the entire dmd.Manufacturers maybe, if that isn't too big

[11-Nov-2011 15:49:50] <Hackman238> enkrypt: zendmd

[11-Nov-2011 15:49:54] <Hackman238> from Products.ZCatalog.ProgressHandler import StdoutHandler

[11-Nov-2011 15:50:19] <Hackman238> dmd.Manufacturers.refreshCatalog(clear=1,pghandler=StdoutHandler())

[11-Nov-2011 15:50:22] <Hackman238> commit()

[11-Nov-2011 15:50:36] <Hackman238> enkrypt: That'll rebuild it. Better, stronger, faster

[11-Nov-2011 15:51:08] <enkrypt> oh damn

[11-Nov-2011 15:51:13] <enkrypt> guess what

[11-Nov-2011 15:51:18] <enkrypt> refreshCatalog itself isn't there

[11-Nov-2011 15:51:18] <johnnynoc> chicken butt?

[11-Nov-2011 15:51:22] <johnnynoc> oh, nvm

[11-Nov-2011 15:51:27] <Hackman238> LOL

[11-Nov-2011 15:51:29] <Hackman238> What?

[11-Nov-2011 15:51:31] <enkrypt> damn oh damn

[11-Nov-2011 15:51:47] <enkrypt> the thing is I really pushed a wrong button once, inside the zope admin thing

[11-Nov-2011 15:51:55] <rocket_> enkrypt: 1 sec I think I can get you the code to recreate productSearch

[11-Nov-2011 15:51:59] <enkrypt> and away went everything inside Manufacturers

[11-Nov-2011 15:52:17] <enkrypt> anyway I think I need refreshCatalog as well no ?

[11-Nov-2011 15:52:53] <cek> damn, it's possible to have only 40 graphs on 1 page of user graph report

[11-Nov-2011 15:53:41] <enkrypt> rocket_ I would need to restore all code related things in Manufacturers in fact. currently I've got 40 items in the dmd.Manufacturers tree, but I think the right number was 42 or 43 before I accidentilly threw it away

[11-Nov-2011 15:53:47] <Hackman238> enkrypt: afterwards, maybe

[11-Nov-2011 15:53:57] <enkrypt> one of them is refreshCatalog, other is productSearch

[11-Nov-2011 15:54:23] <Hackman238> enkrypt: refreshCatalog is a method of productSearch

[11-Nov-2011 15:54:29] <enkrypt> this is the first & the last time I went clicking around in the zope admin, promised

[11-Nov-2011 15:55:05] <Hackman238> enkrypt: 

[11-Nov-2011 15:55:18] <enkrypt> Hackman238 are you sure? cause you asked me to perform dmd.Manufacturers.refreshCatalog() in zendmd just now

[11-Nov-2011 15:55:56] <Hackman238> enkrypt: My bad, I meant dmd.Manufacturers.productSearch.refreshCatalog()

[11-Nov-2011 15:56:10] <enkrypt> ah ok that makes sense

[11-Nov-2011 15:56:14] <Hackman238> enkrypt: Sorry about that 

[11-Nov-2011 15:56:52] <designated> anyone here use shane scott's ip sla monitor zenpack?

[11-Nov-2011 15:57:26] <Hackman238> designated: I do

[11-Nov-2011 15:57:35] <rocket_> enkrypt: there is a migrate script in ZenModel/migrate/

[11-Nov-2011 15:57:37] <Hackman238> designated: Plus I'm Shane. What can I do for you?

[11-Nov-2011 15:57:44] <rocket_> I believe the code for building that catalog is in there

[11-Nov-2011 15:58:00] <enkrypt> rocket_ already tried running that one in fact

[11-Nov-2011 15:58:09] <rocket_> designated: dont listen to him I am Shane

[11-Nov-2011 15:58:20] <Hackman238>

[11-Nov-2011 15:58:25] <designated> Hackman238, i can't get it to graph the sla data unless the device is in /Network/IPSLA

[11-Nov-2011 15:58:34] <Hackman238> designated: Thats correct.

[11-Nov-2011 15:58:51] <Hackman238> designated: Devices much be in a organizer where the zPythonClass is set to the IPSLADevice class

[11-Nov-2011 15:58:57] <designated> despite having set the zpython class of the other organizer correctly and adding the appropriate modeler

[11-Nov-2011 15:59:06] <enkrypt> rocket_: in the code of that migrate script they require productSearch to exist already

[11-Nov-2011 15:59:12] <Hackman238> designated: You can set the zPythonClass on any organizer to the samething the IPSLA one is and restart zenoss

[11-Nov-2011 15:59:25] <designated> it doesn't work on zenoss 3.2.0

[11-Nov-2011 15:59:36] <designated> I've tried and the graphs never get generated

[11-Nov-2011 15:59:42] <Hackman238> designated: Interesting. Do no graphs show or are they nan?

[11-Nov-2011 15:59:48] <designated> no graphs

[11-Nov-2011 16:00:06] <Hackman238> designated: Copy the SLA templates to the target device class

[11-Nov-2011 16:00:14] <Hackman238> designated: pretty sure thats the problem

[11-Nov-2011 16:00:42] <designated> i also cannot get the sla generators to work

[11-Nov-2011 16:00:54] <designated> it keeps spitting out an error so i got to digging around in the logs

[11-Nov-2011 16:00:59] <designated> let me grab that error

[11-Nov-2011 16:01:09] <Hackman238> designated: Yep, thats the problem. You'll need to copy the templates to the new device class. The new version of the pack do out soon fixes all this complication

[11-Nov-2011 16:01:40] <rocket_> dmd.Manufacturers.createCatalog() ?

[11-Nov-2011 16:01:43] <designated>   Module Products.ZenModel.Commandable, line 151, in manage_doUserCommand

[11-Nov-2011 16:01:43] <designated> AttributeError: 'NoneType' object has no attribute 'RESPONSE'

[11-Nov-2011 16:02:07] <Hackman238> designated: Interesting.

[11-Nov-2011 16:02:12] <Hackman238> Zenoss 3.2.0?

[11-Nov-2011 16:02:15] <enkrypt> rocket_: method present, but doesnt seem to do anything

[11-Nov-2011 16:02:24] <designated> Hackman238, yes

[11-Nov-2011 16:02:44] <rocket_> it wont show anything commit when done

[11-Nov-2011 16:03:08] <rocket_> should create a new dmd.Manufacturers.productSearch object catalog

[11-Nov-2011 16:03:08] <Hackman238> designated: That shouldnt be. I'll spin up a v 3.2.0 and figure whats wrong so as to make sure its fixed for the next release and figure a short term solution for you.

[11-Nov-2011 16:03:12] <enkrypt> rocket_ i think it worked

[11-Nov-2011 16:03:15] <Hackman238> designated: Can you PM me your email?

[11-Nov-2011 16:03:27] <Hackman238> designated: Or better yet, email me a reminder:

[11-Nov-2011 16:03:41] <cek> help! report edit page is limited to 40 elemnts. now I can't delete report for those elements over the limit :/

[11-Nov-2011 16:03:49] <cek> though i can see them in zope manage

[11-Nov-2011 16:04:00] <rocket_> enkrypt: now you may want to rerun the migrate steps to populate the catalog

[11-Nov-2011 16:04:06] <Hackman238> cek: In your user properties change the limit from 40 to higher

[11-Nov-2011 16:04:08] <cek> Type:  Value: xxx Vlan99 Throughput does not exist

[11-Nov-2011 16:04:13] <cek> oh

[11-Nov-2011 16:04:53] <cek> but it should have paginator there

[11-Nov-2011 16:05:23] <enkrypt> rocket_: so I run the migrate script you talked about earlier? just "python" or with zenmigrate?

[11-Nov-2011 16:05:44] <cek> still same result - 40 elements

[11-Nov-2011 16:05:56] <Hackman238> cek: Try logging out and in.

[11-Nov-2011 16:05:56] <rocket_> manually run it from zendmd

[11-Nov-2011 16:05:57] <cek>  /zport/dmd/Reports/Graph%20Reports/Interface%20utilization/editGraphReport

[11-Nov-2011 16:06:21] <rocket_> eg copy and paste the contents into zendmd if that makes sense

[11-Nov-2011 16:06:31] <enkrypt> ok

[11-Nov-2011 16:06:38] <cek> logged in, logged out - 40!

[11-Nov-2011 16:06:39] <rocket_> you might be able to tell zenmigrate to just run that step

[11-Nov-2011 16:07:15] <rocket_> I would have to lookup the syntax

[11-Nov-2011 16:07:18] <Hackman238> cek: Thats odd

[11-Nov-2011 16:07:43] <cek> <!-- The main body into which we dump rows --> -- right this table has 40 elements and no more

[11-Nov-2011 16:07:50] <cek> and i can't delete them in zenmanage

[11-Nov-2011 16:09:09] <enkrypt> rocket_: zenmigrate --step=NewProductIndexes

[11-Nov-2011 16:09:22] <rocket_> ah yes .. that should do it

[11-Nov-2011 16:10:05] <cek>

[11-Nov-2011 16:10:10] <enkrypt> ok everything seems back to normal again... phew!

[11-Nov-2011 16:10:21] <enkrypt> thanks Hackman238 & rocket_ , you're lifesavers, as always

[11-Nov-2011 16:10:37] <enkrypt> rocket_ you are not eric edgar by any chance?

[11-Nov-2011 16:10:39] <cek> resolution set to Not Reproducible

[11-Nov-2011 16:10:42] <cek> lol

[11-Nov-2011 16:14:51] <enkrypt> ok anyway thanks guys, back to the weekend!

[11-Nov-2011 16:16:01] <cek> holly f

[11-Nov-2011 16:16:10] <cek> restarted the browser, it all went back to normal

[11-Nov-2011 16:16:20] <cek> to that configured value, which i set to 80

[11-Nov-2011 16:16:24] <cek> but still, it's a bug

[11-Nov-2011 16:17:25] <SEJeff> So I've got a really stupid zenoss question.

[11-Nov-2011 16:17:39] <SEJeff> ZenOSS 3.2, from a device view

[11-Nov-2011 16:17:47] <SEJeff> I can click Configuration Properties or Custom Properties

[11-Nov-2011 16:17:55] <SEJeff> and I don't see all of the zproperties for that device

[11-Nov-2011 16:18:01] <SEJeff> Only a couple that are overrideen

[11-Nov-2011 16:18:14] <SEJeff> I'd like to see all of them so I can customize a few. Am I clicking in the wrong place?

[11-Nov-2011 16:18:18] <rocket_> cek sure its a bug but it most likely was closed because it was against such an old version of zenoss

[11-Nov-2011 16:19:02] <rocket_> if it happens in a recent version of zenoss reopen the bug against the appropriate version and the defect review board will go over it

[11-Nov-2011 16:19:54] <rocket_> its nearly impossible to go through every old bug and test if its still valid .. some pages/pieces of the product change too fast etc

[11-Nov-2011 16:20:40] <rocket_> its unfortunate some get closed wrongly but there is limited bandwidth to fix stuff in recent versions much less go back and validate a really old bug is still applicable

[11-Nov-2011 16:21:13] <cek> not a problem . thanks for your time

[11-Nov-2011 16:22:55] <SEJeff> rocket_, Should I wait for ZSD 4.1.1 enterprise or upgrade from 3.2.0 to 4.1?

[11-Nov-2011 16:23:26] <jmp242> SEJeff: have you edited the filters or something?

[11-Nov-2011 16:23:41] <jmp242> Cause under Configuration properties, I certainly see (in 3.2.1)

[11-Nov-2011 16:23:44] <jmp242> all the zprops

[11-Nov-2011 16:23:58] <SEJeff> jmp242, Hmmm must be saved in a cookie. Yup that was it. It has been a very rough day today :/

[11-Nov-2011 16:24:00] <Hackman238> SEJeff: What scale?

[11-Nov-2011 16:24:13] <SEJeff> Hackman238, around 400 devices with 4 distributed collectors

[11-Nov-2011 16:24:23] <SEJeff> 4 continents

[11-Nov-2011 16:24:56] <rocket_> SEJeff: I would wait for 4.1.1 it will be out relatively soon and there are some good changes for performance of the ui and snmp collection coming

[11-Nov-2011 16:25:15] <rocket_> if you arent in a rush

[11-Nov-2011 16:25:22] <rocket_> otherwise 4.1 is a fine product

[11-Nov-2011 16:25:49] <Hackman238> SEJeff: I'd wait.

[11-Nov-2011 16:25:56] <SEJeff> rocket_, Hackman238 Thanks

[11-Nov-2011 16:25:58] <Hackman238> SEJeff: 4.1.1 has some important updates

[11-Nov-2011 16:26:01] <SEJeff> Around a christmas timeframe?

[11-Nov-2011 16:26:08] <Hackman238> SEJeff: Thats what I heard.

[11-Nov-2011 16:26:23] <jmp242> Is the Core 4.2 beta still December timeframe?

[11-Nov-2011 16:26:27] <SEJeff> Correct me if I'm wrong...

[11-Nov-2011 16:26:28] <jmp242> or is that pushed to Jan?

[11-Nov-2011 16:26:43] <SEJeff> ZSD includes the "root cause analysis" thing that will let you turn a load of events into a few, right?

[11-Nov-2011 16:27:00] <SEJeff> We get so many events in zenoss when a circuit flaps it is unusable and are really anxious to test the new stuff out.

[11-Nov-2011 16:27:25] <Hackman238> SEJeff: If you're using Impact, I think.

[11-Nov-2011 16:27:34] <Hackman238> SEJeff: rocket_ can better answer that question.

[11-Nov-2011 16:27:50] <SEJeff> Oh, do you need to pay additionally for that?

[11-Nov-2011 16:28:16] <rocket_> SEJeff: Impact more than likely will address that issue for you, but it is an additional entitlement.

[11-Nov-2011 16:28:30] <rocket_> SEJeff: thats a matter for your sales rep to go over with you.

[11-Nov-2011 16:28:41] <SEJeff> I see

[11-Nov-2011 16:28:44] <SEJeff> What is list for that?

[11-Nov-2011 16:28:57] <Hackman238> SEJeff: No idea. Rackspace doesnt use it

[11-Nov-2011 16:29:04] <rocket_> Honestly I dont have a clue .. I dont see that side of the equation

[11-Nov-2011 16:29:11] <nexex> I always that service dynamics was just a rename of enterprise

[11-Nov-2011 16:29:12] <SEJeff> Fair enough

[11-Nov-2011 16:29:17] <rocket_> I just fix the broken fun ..

[11-Nov-2011 16:29:32] <SEJeff> nexex, Yeah with some backend parts rewritten that suck in 3.x

[11-Nov-2011 16:29:39] <SEJeff> The events stuff has a much better design

[11-Nov-2011 16:30:14] <rocket_> well the service dynamics umbrella is really 3 parts together

[11-Nov-2011 16:30:46] <rocket_> impact,insight,resource manager (old enterprise product)

[11-Nov-2011 16:31:36] <SEJeff> When you pay for enterprise

[11-Nov-2011 16:31:43] <SEJeff> and then you want to upgrade to SD

[11-Nov-2011 16:31:46] <SEJeff> what do you get for free?

[11-Nov-2011 16:31:55] <SEJeff> "free" being relative

[11-Nov-2011 16:31:59] <rocket_> resource manager

[11-Nov-2011 16:32:20] <SEJeff> Rachel came in awhile ago and tried to sell us on insight

[11-Nov-2011 16:32:24] <SEJeff> Seemed nice

[11-Nov-2011 16:33:16] <rocket_> honestly I havent had to fix much with that one so I would agree

[11-Nov-2011 16:34:10] <cek> gais, help me delete report element in zenmanage!

[11-Nov-2011 16:34:38] <cek> i messed it up, tried to delete its child "elements" in zenmanage and now its throwing errors on trying to delete it from zenoss

[11-Nov-2011 16:35:05] <rocket_> restore from backup .. sorry I dont really support anyone mucking about in zenmanage

[11-Nov-2011 16:35:14] <cek> hehe

[11-Nov-2011 16:35:25] <rocket_> if I could I would turn the whole access to that off

[11-Nov-2011 16:36:09] <rocket_> think if zenmanage == backup first and plan on using it .. :/

[11-Nov-2011 16:37:29] <rocket_> SEJeff: can you put a transform in or a delay on the initial event notification?

[11-Nov-2011 16:37:34] <rocket_> for the flapping issue?

[11-Nov-2011 16:37:48] <SEJeff> rocket_, Perhaps we could

[11-Nov-2011 16:37:52] <SEJeff> but for 30+ clusters?

[11-Nov-2011 16:38:01] <SEJeff> Not so pleasant to manage

[11-Nov-2011 16:38:06] <SEJeff> and 4 collectors

[11-Nov-2011 16:39:03] <rocket_> SEJeff: I would ask about GOM when you move to 4.X

[11-Nov-2011 16:39:11] <SEJeff> GOM?

[11-Nov-2011 16:39:22] <SEJeff> We just need to upgrade to 4.x and then get PS in

[11-Nov-2011 16:39:28] <SEJeff> Thats the plan, but we're waiting for 4.1.1

[11-Nov-2011 16:39:31] <Hackman238> SEJeff: We use it here. Its a single instance that connects to other instances to aggergate events

[11-Nov-2011 16:39:40] <rocket_> SEJeff: Global Operations Manager .. it only manages events for now

[11-Nov-2011 16:39:44] <SEJeff> Oh the global dashboard?

[11-Nov-2011 16:39:52] <Hackman238> SEJeff: Yep. Mostly works.

[11-Nov-2011 16:39:58] <SEJeff> We have a copy of the very very old zenoss dashboard

[11-Nov-2011 16:40:01] <SEJeff> It didn't work so well

[11-Nov-2011 16:40:08] <rocket_> no this is different

[11-Nov-2011 16:40:15] <SEJeff> rocket_, Can you get it from the enterprise login place?

[11-Nov-2011 16:40:17] <Hackman238> SEJeff: New one rewritten for rabbit. Its fast and reliable

[11-Nov-2011 16:40:20] <SEJeff> The new support portal?

[11-Nov-2011 16:40:24] <SEJeff> rabbit being rabbitmq?

[11-Nov-2011 16:40:33] <Hackman238> SEJeff: rabbitmq, yes

[11-Nov-2011 16:40:37] <SEJeff> Hackman238, Oh SWEET

[11-Nov-2011 16:40:39] <rocket_> SEJeff: this is also an additional purchase

[11-Nov-2011 16:40:54] <SEJeff> Alrighty

[11-Nov-2011 16:41:01] <SEJeff> We'll wait for 4.1.1

[11-Nov-2011 16:41:01] <Hackman238> rocket_: It is?

[11-Nov-2011 16:41:04] <rocket_> SEJeff: but the GOM forwards events from slave event consoles to a central console

[11-Nov-2011 16:41:19] <rocket_> Hackman238: you had ZEC so you didnt have to pay for GOM if I recall

[11-Nov-2011 16:41:31] <rocket_> I could be incorrect, I dont deal with the pay parts

[11-Nov-2011 16:41:33] <Hackman238> rocket_: Ah gotcha.

[11-Nov-2011 16:41:40] <Hackman238> rocket_: Can't blame you.

[11-Nov-2011 16:41:42] <Hackman238>

[11-Nov-2011 16:41:50] <Hackman238> I have to run.

[11-Nov-2011 16:41:54] <Hackman238> Have a good weekend all!

[11-Nov-2011 16:41:57] <Hackman238> Later!

[11-Nov-2011 16:42:19] <rocket_> SEJeff: the point is you can forward all of your events to the main console and from there apply the transforms and send alerts etc

[11-Nov-2011 16:42:36] <SEJeff> Of of course

[11-Nov-2011 16:42:40] <SEJeff> We have 1 "console" now

[11-Nov-2011 16:42:45] <rocket_> acknowledge events on the main console and the acknowledgement flows backwards

[11-Nov-2011 16:42:52] <rocket_> ok I misunderstood the setup then

[11-Nov-2011 16:42:52] <SEJeff> I guess I'm not honestly sure where to get started

[11-Nov-2011 16:43:03] <SEJeff> We have 1 master zenoss server running zope and all of that fun stuff

[11-Nov-2011 16:43:13] <SEJeff> Then 4 distributed collectors

[11-Nov-2011 16:43:20] <SEJeff> So 5 servers total

[11-Nov-2011 16:43:30] <rocket_> ok then fixing the transforms for your case shouldnt be a big issue

[11-Nov-2011 16:43:49] <rocket_> it would just be working inside the one instance ..

[11-Nov-2011 16:43:55] <SEJeff> Right

[11-Nov-2011 16:44:52] <rocket_> gotta run .. have fun folks

[11-Nov-2011 16:44:59] <SEJeff> thanks!

[11-Nov-2011 16:45:00] rocket_ is now known as rocket

[11-Nov-2011 16:45:20] rocket is now known as rocket_away

[11-Nov-2011 16:48:26] <cek> well i just deleted the report, thats all

[11-Nov-2011 16:48:38] <cek> btw, it's too easy to delete a report - you click the button and 1 hr work is gone

[11-Nov-2011 16:59:57] <cek> now the problem with user graph report is that graphs have different scales

[11-Nov-2011 17:00:11] <cek> it's hard to compare PDU bank load when you have different scales there

[11-Nov-2011 17:14:10] <gloin> Hey, any way to make the Z2.log less humungous?

[11-Nov-2011 18:05:35] kocolosk is now known as kocolosk|away

[12-Nov-2011 00:00:01] [disconnected at Sat Nov 12 00:00:01 2011]

[12-Nov-2011 00:00:02] [connected at Sat Nov 12 00:00:02 2011]

[12-Nov-2011 00:00:16] [zenoss-logger (logger bot) has joined #zenoss]

[12-Nov-2011 12:00:01] [disconnected at Sat Nov 12 12:00:01 2011]

[12-Nov-2011 12:00:02] [connected at Sat Nov 12 12:00:02 2011]

[12-Nov-2011 12:00:14] [zenoss-logger (logger bot) has joined #zenoss]

[12-Nov-2011 12:00:14] <> [freenode-info] please register your nickname...don't forget to auto-identify!

[13-Nov-2011 00:00:01] [disconnected at Sun Nov 13 00:00:01 2011]

[13-Nov-2011 00:00:02] [connected at Sun Nov 13 00:00:02 2011]

[13-Nov-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]

[13-Nov-2011 07:22:56] <fateme> hey all, I want to create a zenpack but I have a little problem, is someone here eager to help me?

[13-Nov-2011 07:25:01] <fateme> hellooooooo

[13-Nov-2011 07:29:12] <tightwork> fateme: this place shutsdown on the weekends

[13-Nov-2011 07:29:32] <tightwork> its a m-f 9-5 est deal as I see it

[13-Nov-2011 07:30:25] <fateme> tightwork: thanks dude

[13-Nov-2011 07:30:53] <tightwork> and yea, all I know is create an egg

[13-Nov-2011 07:31:21] <tightwork> well I guess it can be done just drop it in .. ye ai dunno

[13-Nov-2011 07:35:08] <fateme> good thanks alot

[13-Nov-2011 12:00:01] [disconnected at Sun Nov 13 12:00:01 2011]

[13-Nov-2011 12:00:02] [connected at Sun Nov 13 12:00:02 2011]

[13-Nov-2011 12:00:14] [zenoss-logger (logger bot) has joined #zenoss]

[13-Nov-2011 16:58:49] <LostyJai> hi guys, is this the channel for the community version of zenoss?

[13-Nov-2011 20:10:58] kocolosk|away is now known as kocolosk

[13-Nov-2011 22:19:29] <LostyJai> hello

[14-Nov-2011 00:00:01] [disconnected at Mon Nov 14 00:00:01 2011]

[14-Nov-2011 00:00:02] [connected at Mon Nov 14 00:00:02 2011]

[14-Nov-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]

[14-Nov-2011 04:27:16] <zykes-> Error: no program specified; use -p or -C < how can i fix that ?

[14-Nov-2011 04:41:18] <fcolista> hi2all

[14-Nov-2011 04:43:53] <fcolista> I've some problem to find the zenoss class that return firmware version of a procurve switch. Seems that only snmpDescr contain this info. I tryed getOS* without luck. My goal is to have a reports with all switch firmware versions.

[14-Nov-2011 04:44:27] <fcolista> I can get this info via snmp

[14-Nov-2011 04:44:47] <fcolista> but in the reports i've not found a way to make a snmp query instead of ZMI

[14-Nov-2011 04:45:15] <fcolista> Any link is appreciated.

[14-Nov-2011 04:45:20] <fcolista> Or hints

[14-Nov-2011 04:45:23] <fcolista> Thanks in advance.

[14-Nov-2011 05:23:50] <fragfutter> fcolista: is snmpDescr enough?

[14-Nov-2011 05:24:58] <fcolista> fragfutter: thanks for your answer. Could be enough, but i should awk or sed in someway to delete what i'm not interested in.

[14-Nov-2011 05:25:15] <fcolista> I'm figuring out how to obtain only firmware infos

[14-Nov-2011 05:27:51] <fragfutter> can you give an example of snmpDescr and what you want to extract?

[14-Nov-2011 05:32:56] <fcolista> sure

[14-Nov-2011 05:33:00] <fcolista> jsut one moment

[14-Nov-2011 05:38:47] <fcolista> ProCurve J8697A Switch 5406zl, revision K.15.04.0003, ROM K.15.12 (/sw/code/build/btm(K_15_04))

[14-Nov-2011 05:39:06] <fcolista> i would only "revision K.15.04.0003, ROM K.15.12"

[14-Nov-2011 05:42:53] <fragfutter> python:",".join(dev.snmpDescr.split(",")[1:]).split('(')[0].strip()

[14-Nov-2011 05:42:57] <fragfutter> and hope for the best

[14-Nov-2011 05:43:55] <fragfutter> and no, you won't be able to use a regexp at this point

[14-Nov-2011 06:30:11] <froztbyte> fragfutter: where would you use that?

[14-Nov-2011 06:30:16] <froztbyte> in the collector somewhere?

[14-Nov-2011 06:30:19] <froztbyte> zykes-: contexst?

[14-Nov-2011 06:30:21] <froztbyte> s/st/t/

[14-Nov-2011 06:30:25] <froztbyte> mein gott I can't type today -_-

[14-Nov-2011 06:30:38] <zykes-> froztbyte: post upgrade

[14-Nov-2011 06:31:50] <fragfutter> fcolista: as a column in your report

[14-Nov-2011 06:39:41] <zykes-> should i be able to upgrade from 3.1 > 3.2.1 directory ?

[14-Nov-2011 06:39:45] <zykes-> directly

[14-Nov-2011 06:41:58] <zykes-> oh, apparantly the upgrade hadn't finished the last time

[14-Nov-2011 06:59:24] <zykes-> what will version v4.0 improve or change ?

[14-Nov-2011 07:00:12] <zykes-> jmp242: btw, regarding your post about JSON api, which bindings are you using and is it python ?

[14-Nov-2011 07:03:01] <froztbyte> fragfutter: that was actually me asking, but I see

[14-Nov-2011 07:21:22] <fcolista> back from lunch

[14-Nov-2011 07:21:27] <fcolista> thanks fragfutter

[14-Nov-2011 07:22:27] <fcolista> actually seems works

[14-Nov-2011 07:23:17] <fcolista> now is matter of restrict this snmpDescr field only to a particular subnet of devices

[14-Nov-2011 07:34:53] <fcolista> umh

[14-Nov-2011 07:34:54] <fcolista> Type:

[14-Nov-2011 07:34:54] <fcolista> Value: cannot concatenate 'str' and 'int' objects

[14-Nov-2011 07:36:54] kocolosk is now known as kocolosk|away

[14-Nov-2011 07:42:39] kocolosk|away is now known as kocolosk

[14-Nov-2011 08:30:25] kocolosk is now known as kocolosk|away

[14-Nov-2011 09:14:47] kocolosk|away is now known as kocolosk

[14-Nov-2011 09:23:10] <enkrypt> when I loaded some MIB's into zenoss, do I need to restart anything to make it work?

[14-Nov-2011 09:28:25] <enkrypt> I imported some mibs, but still my traps remain in the cryptic OID form, not showing any explanation

[14-Nov-2011 10:43:06] rocket_away is now known as rocket

[14-Nov-2011 10:46:00] <cek> Hi. I'm trying to build a multigraph report. I've made everything according to admin guide, but graphs are empty. I'm using  ifInOctets_ifInOctets datapoint to graph iface utilization

[14-Nov-2011 10:46:26] <cek> Do I understand correctly that the graphs should appear immediately (vs. how regular graphs are done?)

[14-Nov-2011 11:03:43] <cek> alright, it's because it's  ifHCInOctets_ifHCInOctets 64bit version instead of 32bit counters one

[14-Nov-2011 11:03:50] <cek> graphs appear immediately.

[14-Nov-2011 11:04:25] <cek> Now, how do I do a graph with aggregate of interface utilization on a router? That is, 1 graph with stacked areas of all ports.

[14-Nov-2011 12:00:01] [disconnected at Mon Nov 14 12:00:01 2011]

[14-Nov-2011 12:00:03] [connected at Mon Nov 14 12:00:03 2011]

[14-Nov-2011 12:00:16] [zenoss-logger (logger bot) has joined #zenoss]

[14-Nov-2011 12:00:21] <zykes-> do i need to place my mib's any place special ?

[14-Nov-2011 12:09:02] <Hackman238> zykes-: $ZENHOME/share/mibs

[14-Nov-2011 12:09:14] <Hackman238> zykes-: Then run zenmib

[14-Nov-2011 12:09:21] <Hackman238> zykes-: Mibs do not create templates.

[14-Nov-2011 12:29:37] <cek> what's the name of a graphpoint for eth0 inbound rate when I'm applying that graphdef to a group of hosts?

[14-Nov-2011 13:03:23] <fus10nx> Hi All -- so I have an event that has a count of there a way to see each individual time that was logged?

[14-Nov-2011 13:03:29] <fus10nx> instead of just the first seen and last seen

[14-Nov-2011 13:05:46] <rocket> fus10nx: no

[14-Nov-2011 13:05:51] <fus10nx> that sucks

[14-Nov-2011 13:06:09] <rocket> specifically in 3.X the answer is no

[14-Nov-2011 13:06:10] <fus10nx> is there any way to turn that type of logging on ?

[14-Nov-2011 13:06:28] <rocket> technically you could use a transform that updated a file

[14-Nov-2011 13:06:33] <rocket> but its not there out of the box

[14-Nov-2011 13:06:39] <rocket> and I wouldnt do that for every event

[14-Nov-2011 13:06:43] <fus10nx> is there a common zenpack for this or something ?

[14-Nov-2011 13:06:47] <rocket> no

[14-Nov-2011 13:06:53] <fus10nx> it's basically logging an event each time our software crashes

[14-Nov-2011 13:07:00] <fus10nx> so its very helpful to log thta each time to help QA

[14-Nov-2011 13:07:29] <rocket> you could transform the subject every time to add a timestamp and update the dedupid etc

[14-Nov-2011 13:07:48] <fus10nx> is that difficult ?

[14-Nov-2011 13:07:53] <rocket> that way each event is technically new

[14-Nov-2011 13:07:57] <rocket> no

[14-Nov-2011 13:07:59] <fus10nx> damn

[14-Nov-2011 13:08:00] <fus10nx> haha

[14-Nov-2011 13:08:02] <fus10nx> of course not

[14-Nov-2011 13:08:14] <rocket> ?

[14-Nov-2011 13:08:24] <fus10nx> of cours eit's not easy is wwhat i meant

[14-Nov-2011 13:08:35] <fus10nx> oh you said it's NOT difficult, sorry, just waking up

[14-Nov-2011 13:08:42] <fus10nx> is there a good tutorial on how to do that?

[14-Nov-2011 13:09:01] <rocket> evt.summary = event.summary +

[14-Nov-2011 13:09:15] <rocket> would be an example of updating the summary message

[14-Nov-2011 13:09:27] <fus10nx> thats all ?

[14-Nov-2011 13:09:43] <rocket> you may need and "import time"

[14-Nov-2011 13:10:23] <fus10nx> ill have to investigate

[14-Nov-2011 13:10:31] <fus10nx> one last thing...we have these temp sensors

[14-Nov-2011 13:10:40] <fus10nx> and they provide a link to their MIB files --

[14-Nov-2011 13:10:47] <fus10nx> sorry -- here

[14-Nov-2011 13:10:48] <fus10nx>

[14-Nov-2011 13:11:04] <fus10nx> but whenever i try to import, zenoss doesnt like it but they show that it works fine w/ MRTG, Nagios, Cacti

[14-Nov-2011 13:12:01] <rocket> technically you dont need the mibs .. all you need is the oid and will need to make new snmp based templates to handle the data you want.

[14-Nov-2011 13:12:17] <fus10nx> man, I should really hire a zenoss consultant to deal w/ all this

[14-Nov-2011 13:27:55] <jmp242> fus10nx: there certainly are community consultants, and of course you could always purchase Zenoss Service Dynamics (enterprise or whatever name you like)

[14-Nov-2011 13:42:47] <fus10nx> Is there a list of them in the NYC area?

[14-Nov-2011 13:44:07] <jmp242> Mmmm, not that I know of

[14-Nov-2011 13:44:13] <jmp242> most of the stuff can be done remotely

[14-Nov-2011 13:44:26] <jmp242> The ZCA has several consultant members

[14-Nov-2011 13:49:48] <fus10nx> do you have a link to the ZCA?

[14-Nov-2011 13:49:56] <jmp242> Yea

[14-Nov-2011 13:50:14] <jmp242> Our most recent update is linked in the topic:

[14-Nov-2011 13:50:22] <jmp242> that also gets you to the space

[14-Nov-2011 13:50:37] <jmp242>

[14-Nov-2011 13:50:47] <jmp242> sorry the first link was to the forum post on it

[14-Nov-2011 13:51:31] <fus10nx> thanks, ill certainly try to find someone

[14-Nov-2011 13:51:52] <jmp242> well, one of the things the ZCA tries to do is get some matchmaking done

[14-Nov-2011 13:52:11] <fus10nx> sure

[14-Nov-2011 13:52:11] <jmp242> I'm a ZCA member and can give you some ideas on who to contact if you give me an idea what you're looking to have done

[14-Nov-2011 13:52:59] <jmp242> One place you might also consider posting is the Zenoss-Jobs forum

[14-Nov-2011 13:53:06] <fus10nx> Sure...I'm talking w/ one of the guys now as well....basically just looking to have a good shell setup and make sure we are focusing on reporting when our applications lock up/crash/etc. and that hte network is always up

[14-Nov-2011 13:53:20] <fus10nx> i certianly will, great to know this is out there !

[14-Nov-2011 13:53:36] <jmp242> Glad to help

[14-Nov-2011 13:54:47] <fus10nx> thanks !

[14-Nov-2011 14:14:48] <humanix> hi there guys

[14-Nov-2011 14:15:05] <humanix> i have a problem with linux monitor zenpack via ssh with zenoss 2.5.2

[14-Nov-2011 14:21:36] <humanix> when i run a test, it works

[14-Nov-2011 14:21:39] <humanix> but it cannot model

[14-Nov-2011 14:30:29] <sud03r_0> hello, has it been implemented in Zenoss the use of zentrap to get IPMI PET (Platform Event Traps) and have zentrap convert its hex value to string?

[14-Nov-2011 14:32:49] kocolosk is now known as kocolosk|away

[14-Nov-2011 14:32:51] <rocket> sud03r_0: you most likely need a transform to convert the hex to string once the trap has been recieved ..

[14-Nov-2011 14:34:31] <sud03r_0> indeed. could a transform call a command i.e. "ipmiutil" ?

[14-Nov-2011 14:35:21] <sud03r_0> thanks rocket. the only thing done so far on transforms is altering any values from the event console. but this one would involve getting that hex value converted to another thing/animal

[14-Nov-2011 14:35:49] <rocket> sud03r_0: it can but be careful it could really harm performance ...

[14-Nov-2011 14:36:30] <rocket> sud03r_0: you can do anything you would do from a python shell ...

[14-Nov-2011 14:36:33] <sud03r_0> yeah. that's the big concern esp with a hub running with 10k to 50k events per hour

[14-Nov-2011 14:36:55] <sud03r_0> great. i'll try that out. would something like an OS call from the python shell do that right off

[14-Nov-2011 14:36:59] <sud03r_0> thanks rocket

[14-Nov-2011 14:37:01] <rocket> if you are only doing this transform on specific events you are fine

[14-Nov-2011 14:37:26] <rocket> eg if subject = "certain trap:  then do the ipmiutil call

[14-Nov-2011 14:39:43] <sud03r_0> great

[14-Nov-2011 15:54:58] <humanix> hi there

[14-Nov-2011 16:36:17] <fus10nx> Does ZenOSS act as a syslog server as well?

[14-Nov-2011 16:54:42] <dhopp> fus10nx:  it can

[14-Nov-2011 17:32:52] <rhce7320> anybody know of a smokeping irc channel?

[14-Nov-2011 17:38:11] <designated> there is no official smoke ping irc channel

[14-Nov-2011 17:38:37] <designated> try the mailing list

[14-Nov-2011 17:51:16] <rhce7320> tks

[15-Nov-2011 00:00:01] [disconnected at Tue Nov 15 00:00:01 2011]

[15-Nov-2011 00:00:01] [connected at Tue Nov 15 00:00:01 2011]

[15-Nov-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]

[15-Nov-2011 04:48:48] <rcngeoff> Hi.. I have an snmp script extension check that returns 0 or 1.  I'd like zenoss to alert only if it gets 3 three consecutive failures (value = 1).  I can't seem to find a way to do that.

[15-Nov-2011 04:56:40] <zykes-> anyone seen this before? "Unable to read processes on device x; error: bin_sh"

[15-Nov-2011 05:14:20] <danuvius> good morning....

[15-Nov-2011 05:30:36] <zykes-> i get the error on one of my nodes

[15-Nov-2011 05:46:36] <konvert> hello

[15-Nov-2011 05:46:51] <konvert> want to do an upgrade to 3.2.1 from 3.2.0

[15-Nov-2011 05:46:59] <konvert> did a backup of all files

[15-Nov-2011 05:47:11] <konvert> can i do upgrade now, or did i forgot something

[15-Nov-2011 08:06:48] <nachox> guys, is it normal for zenoss to consume about 1.5 gb of ram after 3 days of usage?

[15-Nov-2011 08:07:22] <nachox> the machine in which my instance is installed has about 2gb so it is trashing the server

[15-Nov-2011 08:08:55] <nachox> i am monitoring about 5 devices with it

[15-Nov-2011 08:10:12] <fragfutter> nachox: welcome to zope

[15-Nov-2011 08:10:52] <nachox> meaning there is nothing i can do?

[15-Nov-2011 08:12:22] <fragfutter> nachox: it's an application server. you can tune the amount of memory it will consume in zope.conf. There (should) is no memory leak, it will simply start to consume memory while it builds up it's caches and then settles somewhere.

[15-Nov-2011 08:13:15] <fragfutter> i would not try to run it with 2GB ram.

[15-Nov-2011 08:13:45] <nachox> meaning its too little even to monitor just 5 devices? impressive

[15-Nov-2011 08:14:14] <fragfutter> nachox: base memory consumption of zope is around 1GB never seen any smaller (long running) zope instance.

[15-Nov-2011 08:18:11] <humanix> hi there guys

[15-Nov-2011 08:18:16] <humanix> hi rmatte Hackman238

[15-Nov-2011 08:19:24] <SEJeff> So the past 3 times I've went to look at the zenoss webui, zope is hung on something and it makes the webui unresponsive

[15-Nov-2011 08:19:35] <SEJeff> As in it will spin for 10 minutes and not even display the login page

[15-Nov-2011 08:19:47] <SEJeff> Has anyone else seen this with 3.2.0?

[15-Nov-2011 08:21:38] <nachox> fragfutter, zope.conf is not particularly descriptive as to what can control memory usage

[15-Nov-2011 08:25:06] <danuvius> I'm having some issues with the zenperfsnmp (at least I think). When I run zenperfsnmp by hand against a device I instantly gat results no issues however when running it against all devices some devices time out (the lines are busy).

[15-Nov-2011 08:25:23] <danuvius> I'm using snmp v3 formonitoring...

[15-Nov-2011 09:06:46] <zykes-> anyone here built packages for RHEL6 ?

[15-Nov-2011 09:08:20] <zykes-> jmp242: did you get my message earlier regarding json stuff ?

[15-Nov-2011 09:10:41] <jmp242> no

[15-Nov-2011 09:10:44] <jmp242> my PC rebooted

[15-Nov-2011 09:10:47] <zykes-> ah

[15-Nov-2011 09:10:51] <jmp242> and then hung up trying to log in for like 1 hr

[15-Nov-2011 09:10:51] <zykes-> jmp242: btw, regarding your post about JSON api, which bindings are you using and is it python ?

[15-Nov-2011 09:10:59] <jmp242> but I don't use JSON

[15-Nov-2011 09:11:05] <jmp242> so what post are you referring to?

[15-Nov-2011 09:11:15] <bseklecki> "We are very excited to announce the Public Beta of a new Amazon EC2 Cluster Compute instance"

[15-Nov-2011 09:11:28] <zykes-> i was reading a blog or something regarding some bindings

[15-Nov-2011 09:11:34] <bseklecki> oh fuck yea; we causing cancer or curing it this week?

[15-Nov-2011 09:11:54] <jmp242> can you give me a link?

[15-Nov-2011 09:12:01] <jmp242> I might have been bringing information together

[15-Nov-2011 09:12:02] <zykes-> jmp242: sec

[15-Nov-2011 09:22:51] <humanix> im having a problem with a zenpack

[15-Nov-2011 10:35:04] <kenhen93> Hi

[15-Nov-2011 10:35:21] <kenhen93> what would be the best way to get an export list of device classes and event classes out of zenoss?

[15-Nov-2011 11:03:10] <designated> Hackman238, you around?

[15-Nov-2011 11:13:36] <Hackman238> designated: I am, but my time is sparse today. Rain + Texas = Disaster

[15-Nov-2011 11:13:40] <Hackman238> designated: Whats up?

[15-Nov-2011 11:37:36] <designated> Hackman238, I was just wondering if you were planning on adding support for the additional SLA operations, like "UDP Jitter for VoIP"

[15-Nov-2011 11:37:49] <designated> Hackman238, within the zenoss ipsla zenpack

[15-Nov-2011 11:38:52] <Hackman238> designated: Yes.

[15-Nov-2011 11:39:44] <designated> Hackman238, any idea on a time line?

[15-Nov-2011 11:40:06] <Hackman238> designated: This 19th

[15-Nov-2011 11:40:16] <designated> perfect thank you

[15-Nov-2011 11:40:21] <Hackman238> designated: NP

[15-Nov-2011 12:00:01] [disconnected at Tue Nov 15 12:00:01 2011]

[15-Nov-2011 12:00:02] [connected at Tue Nov 15 12:00:02 2011]

[15-Nov-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]

[15-Nov-2011 12:01:40] <Balistic> I am getting "Problem while executing plugin zenoss.snmp.NewDeviceMap" on linux boxes alot lately, the emails contain a stack trace complaining about not finding KeyError: (1, 3, 6, 1, 2, 1, 1, 2, 0)

[15-Nov-2011 12:01:55] <Balistic> in oidmap. Any ideas?

[15-Nov-2011 12:09:00] * davetoo always gets a kick out of "ZCA"... Zenoss Component Alliance

[15-Nov-2011 12:29:37] * davetoo hums an old Elton John tune

[15-Nov-2011 12:35:13] <johnnynoc> Hackman238 4ever

[15-Nov-2011 13:23:06] <Hackman238> johnnynoc: LOL thanks

[15-Nov-2011 13:23:36] <jmp242> Afternoon Hackman238

[15-Nov-2011 13:24:01] <Hackman238> jmp242: Hey! How goes the war?

[15-Nov-2011 13:25:00] <jmp242> ok

[15-Nov-2011 13:25:35] <jmp242> how goes you're battles?

[15-Nov-2011 13:25:37] <jmp242> err

[15-Nov-2011 13:25:38] <jmp242> your

[15-Nov-2011 13:25:39] <Hackman238> jmp242: Not good.

[15-Nov-2011 13:25:40] <jmp242> lol

[15-Nov-2011 13:25:44] <Hackman238> LOL

[15-Nov-2011 13:25:57] <jmp242> I'm sorry to hear of the problems

[15-Nov-2011 13:26:23] <Hackman238> jmp242: It happens. We'll all live...I hope

[15-Nov-2011 13:26:41] <jmp242> I'm trying to get remotely anywhere quotes ... web site goes into black hole apparently

[15-Nov-2011 13:26:44] <jmp242> now on hold

[15-Nov-2011 13:26:53] <Hackman238> jmp242: Next month things should start smoothing out.

[15-Nov-2011 13:27:02] <Hackman238> jmp242: Oh man. That's a pain.

[15-Nov-2011 13:27:17] <jmp242> yea

[15-Nov-2011 13:27:26] <Hackman238> jmp242: Never heard of the software. Is it like logmein?

[15-Nov-2011 13:27:33] <jmp242> it is logmein

[15-Nov-2011 13:27:38] <jmp242> just w/out the website

[15-Nov-2011 13:27:45] <Hackman238> jmp242: Gotcha

[15-Nov-2011 13:27:46] <jmp242> it's the corporate one

[15-Nov-2011 13:27:52] <jmp242> so you don't need to use their servers

[15-Nov-2011 13:27:52] <Hackman238> jmp242: Ah

[15-Nov-2011 13:28:02] <Hackman238> jmp242: I didnt know they had a service like that

[15-Nov-2011 13:28:07] <jmp242> yea

[15-Nov-2011 13:28:09] <jmp242> they hide some things

[15-Nov-2011 13:28:14] <Hackman238> jmp242: I'll have to check it out when they decide to fix their site LOL

[15-Nov-2011 13:28:15] <jmp242> but it's like LogMeIn pro

[15-Nov-2011 13:28:49] <Hackman238> jmp242: Gotcha.

[15-Nov-2011 13:28:56] <Hackman238> jmp242: Site is working here.

[15-Nov-2011 13:28:57] <jmp242> w/out subscriptions, more perpetual license + maintenance if desired

[15-Nov-2011 13:29:02] <jmp242> yea

[15-Nov-2011 13:29:04] <jmp242> the site works

[15-Nov-2011 13:29:15] <jmp242> the quote request just doesn't get answered

[15-Nov-2011 13:29:53] <Hackman238> jmp242: Oh gotcha.

[15-Nov-2011 13:30:04] <Hackman238> jmp242: It goes to 'that address nobody checks'

[15-Nov-2011 13:30:08] <Hackman238>

[15-Nov-2011 13:30:08] <jmp242> yea

[15-Nov-2011 13:30:20] <jmp242> the we don't want to answer cause you might give us money one lol

[15-Nov-2011 13:30:35] <Hackman238> LOL

[15-Nov-2011 13:31:00] <jmp242> I perservere

[15-Nov-2011 13:31:08] <jmp242> left a voice mail

[15-Nov-2011 13:31:09] <jmp242> lol

[15-Nov-2011 13:31:19] <jmp242> I guess sales reps don't do much selling there IDK

[15-Nov-2011 13:31:32] <Hackman238> jmp242: LOL Too busy surfing facebook and youtube I bet

[15-Nov-2011 13:33:28] <jmp242> I also love windows and samba shares here

[15-Nov-2011 13:33:39] <jmp242> no one reads the web docs

[15-Nov-2011 13:33:56] <jmp242> they are told they have to put domain\username in the username field for it to work

[15-Nov-2011 13:33:57] <jmp242> lol

[15-Nov-2011 13:37:24] <Hackman238> LOL

[15-Nov-2011 13:38:00] <Hackman238> You'd think people would just take two seconds, remove their head from their ass, read the one paragraph in the 'how to setup your shares' email and just make it work.

[15-Nov-2011 13:38:13] <Hackman238> You know, acctually learn something new rather than just ask for help.

[15-Nov-2011 13:52:28] <jmp242> yea, but they think it's faster to walk out to a different building and ask me

[15-Nov-2011 13:53:01] <jmp242> I also love that most users seem to think that the e-mails with links to documentation are sent because we like to fill up their inbox

[15-Nov-2011 13:53:08] <jmp242> rather than it's info they'll need

[15-Nov-2011 13:53:24] <Hackman238> jmp242: " mean I have to read? What is this 'read' you speak of? Sound hard!"

[15-Nov-2011 13:56:02] <jmp242> lol

[15-Nov-2011 14:16:20] <rocket> I dotn raed ... I treid ist oto ahrd ...

[15-Nov-2011 14:22:26] <SCR512> ehlo, is there a way to override a local value, for instance SNMP community globally or am I stuck going to each device I monitor and removing the local value?

[15-Nov-2011 14:25:46] <gloin> Zenoss 3.2.1, trying to enable smtp monitoring for a specific host by browsing to it in IP Services and clicking the gear drop-down but "Monitoring" is greyed out.  I click on SMTP and changed "Set local value" to Yes in enable monitoring and save, but it's still greyed out in the IP services list for that host. Help?

[15-Nov-2011 14:31:44] <gloin> Darmok and Jalad at Tanagra

[15-Nov-2011 14:40:08] <nexex> Temba, his arms wide!

[15-Nov-2011 14:49:45] <jmp242> SCR512: I believe if you say reset bindings to a device class, it's supposed to reset all the local values, but I haven't really seen that actually work

[15-Nov-2011 14:49:56] <jmp242> you could code something up in zendmd

[15-Nov-2011 14:50:27] <jmp242> I think it was referenced in some forum post - try searching?

[15-Nov-2011 14:52:58] <jmp242> gloin: I see something similar

[15-Nov-2011 14:53:14] <jmp242> it's odd to me, if there's a monitored entry in that column, you can change it

[15-Nov-2011 14:53:16] <jmp242> but if not, no

[15-Nov-2011 14:53:19] <jmp242> and I'm not sure

[15-Nov-2011 14:53:24] <jmp242> Hackman238: any ideas?

[15-Nov-2011 14:59:19] <Hackman238> Sorry, I was in limbo. What'd I miss?

[15-Nov-2011 15:01:34] <gloin> Shaka, when the walls fell

[15-Nov-2011 15:01:41] <SCR512> jmp242: Yeah, I found the code

[15-Nov-2011 15:02:03] <gloin> aka I lost my webchat connection right when people started talking (hopefully about the perma-gray smtp monitoring issue)

[15-Nov-2011 15:02:05] <SCR512> I ended up doing it un-boss like and sifting through my 100+ window devices and fixing it

[15-Nov-2011 15:07:49] <Hackman238> SCR512: It's doable in DMD. Sorry I missed your question.

[15-Nov-2011 15:11:19] <SCR512> so in DMD

[15-Nov-2011 15:12:03] <SCR512> is it similar to this post?

[15-Nov-2011 15:14:00] <Hackman238> SCR512: Thats exactly how

[15-Nov-2011 15:14:18] <SCR512> ok

[15-Nov-2011 15:14:26] <SCR512> time for me to RTFM on how to interact with DMD

[15-Nov-2011 15:14:50] <Hackman238> SCR512: DMD is the savior of Zope, Zenoss and Mankind alike.

[15-Nov-2011 15:14:59] <Hackman238> SCR512: Without it there is only madness

[15-Nov-2011 15:15:01] <Hackman238>

[15-Nov-2011 15:15:08] <SCR512> haha, true

[15-Nov-2011 15:15:41] <SCR512> ah, this sounds like my solution

[15-Nov-2011 15:15:42] <SCR512>

[15-Nov-2011 15:17:23] <Hackman238> SCR512: Thats how you can write entire scripts that interact via dmd. highly useful.

[15-Nov-2011 15:18:08] <Hackman238> SCR512: Your previous link, DOC-2533, will work with any zProperty and cProperty on any iterable object

[15-Nov-2011 15:25:38] <Hackman238> From one of my users today: "In other news, Zenoss is no longer loading for me at all (cookies/cache cleared, goats sacrificed, etc).  Others are still able to access it.  Perhaps I have angered the Zenoss gods? "

[15-Nov-2011 15:35:58] <SCR512> ha, Zenoss does have a multi-god dependency so it's likely

[15-Nov-2011 15:39:47] <Hackman238> SCR512: /opt/zenoss/lib/python/AccessControl/    "AccessControl.PermissionRole": ("_what_not_even_god_should_do",

[15-Nov-2011 15:40:01] <Hackman238> LOL

[15-Nov-2011 15:44:17] <jmp242> Hackman238: the idea was regarding the monitoring options being greayed out

[15-Nov-2011 15:44:21] <jmp242> grayed

[15-Nov-2011 15:44:27] <jmp242> but not consistently for a process

[15-Nov-2011 15:44:39] <jmp242> it's odd in my 3.2.1 install, and it looks like gloin has a similar issue

[15-Nov-2011 15:47:08] <jmp242> This may be a thread by the same person:

[15-Nov-2011 16:14:41] <bseklecki>!?!

[15-Nov-2011 16:14:55] <bseklecki>

[15-Nov-2011 16:15:45] <Hackman238> bseklecki: Yikes.

[15-Nov-2011 16:16:22] <bseklecki> ("The Usual Suspects" reference)

[15-Nov-2011 17:03:37] <Hackman238> Later all!

[15-Nov-2011 17:25:05] <jb> whats the easiest way to make a custom alert template?

[15-Nov-2011 17:25:23] <jb> i need to change the entire format of the generic/out of the box alert message.

[15-Nov-2011 17:25:27] <jb> it just doesn't work very well for SMS.

[15-Nov-2011 17:25:40] <jb> ahh nevermind

[15-Nov-2011 17:25:43] <jb> i see now

[15-Nov-2011 17:45:21] <zykes-> jmp242:

[15-Nov-2011 17:45:22] <zykes-> btw

[15-Nov-2011 17:45:42] <zykes-> "d) interoperation programming (using JSON and other APIs to integrate Zenoss with other tools)"

[15-Nov-2011 17:45:52] <zykes-> i've made a python binding already for the 3.x api

[15-Nov-2011 17:46:15] <zykes-> don't know if it's missing features though for v3.2.x

[15-Nov-2011 17:46:18] <zykes-> or 3.1

[15-Nov-2011 17:46:44] <zykes-> has anyone ever install zenoss on a RHEL6.1 derivative ?

[15-Nov-2011 17:49:16] <nexex> zykes-: I tried the RPMs, they didn't work

[15-Nov-2011 17:49:59] <zykes-> nexex: yeah,  wonder seriously why they don't get a el6 build

[15-Nov-2011 17:50:14] <zykes-> installing el5 is pain... it's like having a t-rex

[15-Nov-2011 17:50:21] <zykes-> old and slower

[15-Nov-2011 17:51:12] <nexex> zykes-: yea I hope 4.0 will have one, especially since they are using github for the zenpack stuff, and rhel5 doesn't have git in its repositories

[15-Nov-2011 17:52:17] <zykes-> i don't see the problem for them in building a v3.2 build for el6...

[15-Nov-2011 17:52:21] <zykes-> just to change the spec a bit

[15-Nov-2011 17:53:22] <nexex> Anyone know about reportmail sending blank reports (

[15-Nov-2011 17:53:43] <zykes-> stupid to have to install centos 5 and then when v4 comes you have to upgrade both os and zenoss now

[15-Nov-2011 17:54:25] <zykes-> maybe i can rebuild a package from the el5 srpm

[15-Nov-2011 17:54:33] <zykes-> worth the shot

[15-Nov-2011 17:55:00] <zykes-> Hackman238: around ?

[15-Nov-2011 18:06:18] <rocket> zykes-: its more a matter of testing for el6

[15-Nov-2011 18:06:40] <rocket> zykes-: there are rounds of testing now for a 4.X el6 but I do not think there will be a 3.X el6

[15-Nov-2011 18:07:49] <zykes-> rocket: what's new in 4.x ?

[15-Nov-2011 18:07:56] <rocket> SCR512: are you still here?  are you trying to change that setting globally?

[15-Nov-2011 18:08:01] <zykes-> for core, anything revolutionary ?

[15-Nov-2011 18:08:18] <rocket> zykes-: I dont know all of the details for the core version

[15-Nov-2011 18:08:27] <rocket> but I know that the new event console will be there

[15-Nov-2011 18:08:37] <zykes-> will the amqp goodies be in ?

[15-Nov-2011 18:08:47] <zykes-> event engine

[15-Nov-2011 18:09:14] <rocket> zykes-: yes as far as I have heard, however there are a few performance bolt-ons for the super massive events that will not be there

[15-Nov-2011 18:09:26] <rocket> it will still be a major improvement over 3.x

[15-Nov-2011 18:09:37] <zykes-> super massive events ?

[15-Nov-2011 18:10:01] <rocket> think handling 10k events a second rather than the out of the box 1000 events

[15-Nov-2011 18:10:14] <rocket> zenoss 3 handles 100events a second for instance

[15-Nov-2011 18:10:30] <rocket> so zenoss4 core should be an order of magnitude faster from that standpoint

[15-Nov-2011 18:10:56] <zykes-> ah

[15-Nov-2011 18:11:08] <zykes-> is there any roadmap for it that i can see or ?

[15-Nov-2011 18:11:09] <rocket> between 3 and 4 there are massive core rewrites of the collector daemons

[15-Nov-2011 18:11:26] <rocket> tentatively 1st quarter 2012

[15-Nov-2011 18:11:54] <rocket> definately do not consider that official

[15-Nov-2011 18:12:23] <zykes-> what's v4 in enterprise then ?

[15-Nov-2011 18:12:34] <rocket> ?

[15-Nov-2011 18:13:14] <rocket> I dont understand that last question can you say it differently?

[15-Nov-2011 18:13:56] <zykes-> i remember there being some funkyness around v3 core now and enterprise now or something

[15-Nov-2011 18:14:06] <zykes-> that "core" skipped a version or something

[15-Nov-2011 18:14:14] <rocket> well there is not currently a 4 core at this time

[15-Nov-2011 18:14:40] <rocket> at 4.2 there will be a 4 core again if my information is accurate

[15-Nov-2011 18:14:57] <zykes-> eh, so the 4.x which is now is ent only +

[15-Nov-2011 18:15:06] <rocket> yes

[15-Nov-2011 18:15:22] <zykes-> is 4.x out now for ent customers ?

[15-Nov-2011 18:15:40] <rocket> it has been for some time now .. since around july

[15-Nov-2011 18:15:51] <rocket> we are currently on 4.1 for enterprise customers

[15-Nov-2011 18:16:14] <zykes-> hmmms, funky to skip the community for allmost a year :/

[15-Nov-2011 18:16:27] <rocket> community has NOT been skipped

[15-Nov-2011 18:16:54] <rocket> 3.2.1 was and is also supported as an enterprise product and most customers for enterprise are on that version

[15-Nov-2011 18:17:10] <rocket> 3.2.1 was recently released with a significant number of patches

[15-Nov-2011 18:18:02] <zykes-> well still the major focus has been on v4 which is the new hot fuzz

[15-Nov-2011 18:18:47] <rocket> it is the new hot fuzz but that hasnt made it all better.  We are reworking some areas which is causing the delays in 4.2 getting out the door

[15-Nov-2011 18:19:13] <zykes-> event stuff or engine stuff ?

[15-Nov-2011 18:19:45] <rocket> both ... tweaks to the event engine for some unforseen defects and the core daemons have needed some work.

[15-Nov-2011 18:20:23] <rocket> in fact we have told many customers to stay on 3.X as we have found some issues

[15-Nov-2011 18:20:34] <zykes-> ok

[15-Nov-2011 18:20:46] <zykes-> but for 4.2 then it will be core upped to that as well then ?

[15-Nov-2011 18:20:51] <rocket> yes

[15-Nov-2011 18:20:59] <zykes-> goodie goodie

[15-Nov-2011 18:21:26] <zykes-> looking forwards to be able to listen in to a amqp queue instead of having to do hackish ways like now

[15-Nov-2011 18:23:20] <rocket> yeah .. do keep in mind the system requirements for ram etc does go up as there are more daemons etc in 4.2

[15-Nov-2011 18:23:26] <rocket> if your looking to budget hardware

[15-Nov-2011 18:23:39] <zykes-> we've got it in a virtual environment so

[15-Nov-2011 18:25:09] <rocket> gotta run ... bbl

[15-Nov-2011 18:29:35] <GameMagister> Hi all

[15-Nov-2011 18:30:04] <GameMagister> i am new to zenoss and i am working on getting the remote collectors working...

[15-Nov-2011 18:31:04] <GameMagister> i am stuck at trying to get the device on the remote collector to model or snmpwalk.. i can snmpwalk from the commandline

[15-Nov-2011 18:36:08] <nexex> GameMagister: I would try asking again tomorrow earlier in the day, this channel dies down after 05:00 PM EST

[15-Nov-2011 18:36:48] <GameMagister> thanks... i have been working on this most of the day and found a reference to this channel and thought i would give it a try

[15-Nov-2011 18:36:55] <GameMagister> i will try tomorrow...

[15-Nov-2011 19:16:08] <zykes-> rocket: around ?

[15-Nov-2011 19:47:07] <zykes-> nexex: i built a rpm for EL6.1

[15-Nov-2011 19:47:11] <zykes-> works gloriously

[15-Nov-2011 20:09:37] <lawnchair> im installing zenoss into my environment... is snmp v3 recommended?

[15-Nov-2011 23:15:31] <bseklecki> snmpv3 is never recommend

[15-Nov-2011 23:15:36] <bseklecki> speaking for myself

[16-Nov-2011 00:00:01] [disconnected at Wed Nov 16 00:00:01 2011]

[16-Nov-2011 00:00:01] [connected at Wed Nov 16 00:00:01 2011]

[16-Nov-2011 00:00:11] [zenoss-logger (logger bot) has joined #zenoss]

[16-Nov-2011 00:00:50] <Balistic> If you are not going to use snmpv3 then dont use snmp(From a security perspective)

[16-Nov-2011 00:44:56] kocolosk|away is now known as kocolosk

[16-Nov-2011 03:43:43] <danuvius> good morning

[16-Nov-2011 05:06:20] fade1 is now known as fade_uk

[16-Nov-2011 05:10:14] <fade_uk> Hi, I'm trying to graph a grouped data set, and struggling to find what I need in the docs

[16-Nov-2011 05:10:34] <fade_uk> specifically I want an area graph of: SELECT user, COUNT(*) AS connections FROM processlist GROUP BY user;

[16-Nov-2011 05:36:12] <fragfutter> fade_uk: you want to install sql data source zenpack

[16-Nov-2011 05:38:53] <fade_uk> fragfutter: yep, got that and it's worked for a simple select

[16-Nov-2011 05:39:32] <fade_uk> where I'm struggling is mapping the more complicated query to a graph

[16-Nov-2011 06:24:08] <zykes-> anyone of you done multiple collectors ?

[16-Nov-2011 06:49:16] <zykes-> how do i install a zenpack from a directory from github ?

[16-Nov-2011 08:12:07] <jmp242> zykes-: you'd need to use git to get the zenpack dir

[16-Nov-2011 08:12:43] <jmp242> and do a  zenpack --link --install <directoryname>

[16-Nov-2011 08:12:52] <jmp242> see:

[16-Nov-2011 08:12:54] <jmp242> for the docs

[16-Nov-2011 08:40:20] <jb> ugh, wonder why this isn't working on zFileSystemIgnoreNames: (^/oracle-data/|^/fc-tsm03/|^/tsm|/cdrom)

[16-Nov-2011 08:43:18] <jmp242> mmm

[16-Nov-2011 08:43:27] <jmp242> why the (^/

[16-Nov-2011 08:43:30] <jmp242> we use

[16-Nov-2011 08:43:47] <jmp242> like this /afs|/home/|/media/

[16-Nov-2011 08:43:52] <jmp242> which seems to work

[16-Nov-2011 08:55:26] <johnnynoc> anyone have a problem with custom properties/schema in 3.2.1?

[16-Nov-2011 08:55:58] <johnnynoc> i have a zenpack that relies on them (aggregate apc pdu zenpack) and i can't seem to create them manually through the zope management interface

[16-Nov-2011 09:44:54] kocolosk is now known as kocolosk|away

[16-Nov-2011 10:10:55] <jb> any of you sending real SMS from zen?

[16-Nov-2011 10:14:47] <straterra> Define real SMS

[16-Nov-2011 10:19:09] <jb> not using a SMS-to-email gateway

[16-Nov-2011 10:19:13] <jb> but sending real SMS messages

[16-Nov-2011 10:22:59] <f00fSteR> Hackman238, heyy

[16-Nov-2011 10:45:12] <GameMagister> morning all...

[16-Nov-2011 10:45:20] <GameMagister> i am new to zenoss and i am working on getting the remote collectors working...

[16-Nov-2011 10:45:26] <GameMagister> i am stuck at trying to get the device on the remote collector to model or snmpwalk.. i can snmpwalk from the commandline

[16-Nov-2011 11:03:44] kocolosk|away is now known as kocolosk

[16-Nov-2011 11:07:28] <GameMagister> found my issue...

[16-Nov-2011 11:20:31] <blankaman> >>> z._router_request('DeviceRouter', 'getDevices', cache=0, data=[{'uid': '/zport/dmd/Systems/WWW'}]) {u'message': u"AttributeError 'NoneType' object has no attribute 'iteritems'", u'type': u'exception'}

[16-Nov-2011 11:20:56] <blankaman> anyone know why the "getDevices" api method would suddenly stop working?

[16-Nov-2011 11:21:00] <blankaman> we're running 3.2.0

[16-Nov-2011 11:56:43] <johnnynoc> anyone using 3.2.1 successfully with custom properties?

[16-Nov-2011 11:56:53] <johnnynoc> i can't seem to set them through the zope management interface

[16-Nov-2011 11:57:03] <johnnynoc> i see the properties in the zmi, but not on the device

[16-Nov-2011 11:57:03] <johnnynoc>

[16-Nov-2011 12:00:01] [disconnected at Wed Nov 16 12:00:01 2011]

[16-Nov-2011 12:00:02] [connected at Wed Nov 16 12:00:02 2011]

[16-Nov-2011 12:00:14] [zenoss-logger (logger bot) has joined #zenoss]

[16-Nov-2011 12:11:15] kocolosk is now known as kocolosk|away

[16-Nov-2011 12:12:34] kocolosk|away is now known as kocolosk

[16-Nov-2011 12:16:02] <GameMagister> does any one know what the url form is to test the zrender call on a remote collector?

[16-Nov-2011 12:16:22] <blankaman> lots of questions and no answers ...

[16-Nov-2011 12:21:23] kocolosk is now known as kocolosk|away

[16-Nov-2011 12:22:54] tvincent_ is now known as tvincent

[16-Nov-2011 12:23:47] kocolosk|away is now known as kocolosk

[16-Nov-2011 12:36:11] kocolosk is now known as kocolosk|away

[16-Nov-2011 12:50:58] <zykes-> is there life?

[16-Nov-2011 12:51:07] <GameMagister> i am here

[16-Nov-2011 12:51:21] <GameMagister> but i am not life

[16-Nov-2011 12:53:13] <blankaman> anyone know why the "getDevices" api method would suddenly stop working in 3.2.0?

[16-Nov-2011 12:56:47] <zykes-> blankaman: json api or  ?

[16-Nov-2011 13:07:27] <zykes-> anyone here monitoring switches ?

[16-Nov-2011 13:10:09] <zykes-> fragfutter: around ?

[16-Nov-2011 13:10:30] <GameMagister> i am not monitoring any switches just routers and servers sorry

[16-Nov-2011 13:10:52] <zykes-> GameMagister: any specific router or ?

[16-Nov-2011 13:11:19] <GameMagister> ciscos

[16-Nov-2011 13:22:26] <zykes-> when using remote collectors is there a way to debug why i'm not getting graphs in the ui ?

[16-Nov-2011 13:24:57] <GameMagister> what i found is that your workstation that you are pulling up the graph from has to have access to the remote collector

[16-Nov-2011 13:25:31] <GameMagister> i had to redo my setup so that i could access the remote collector from my desk instead of from the datacenter

[16-Nov-2011 13:26:25] <mloven> GameMagister: if you're using Firefox, you can right click on the window and select "view page info", then select the "media" tab.   You should see the graphs that didn't load in there.

[16-Nov-2011 13:26:48] <zykes-> ok

[16-Nov-2011 13:26:55] <blankaman> zykes - yes json api

[16-Nov-2011 13:27:06] <zykes-> blankaman: python or perl ?

[16-Nov-2011 13:27:08] <blankaman> >>> z._router_request('DeviceRouter', 'getDevices', cache=0, data=[{'uid': '/zport/dmd/Systems/WWW'}]) {u'message': u"AttributeError 'NoneType' object has no attribute 'iteritems'", u'type': u'exception'}

[16-Nov-2011 13:27:10] <blankaman> python

[16-Nov-2011 13:27:28] <zykes-> which bindings ?

[16-Nov-2011 13:27:30] <blankaman> i'm actually using the api example lib and making a manual call through the devicerouter

[16-Nov-2011 13:27:42] <blankaman> i've pasted what i did

[16-Nov-2011 13:27:46] <blankaman> and the response

[16-Nov-2011 13:27:57] <blankaman> says the object has no attribute 'iteritems'

[16-Nov-2011 13:29:10] <zykes-> blankaman: i made a more extensive api binding for v 3.x.x

[16-Nov-2011 13:29:39] <blankaman> oh hmm

[16-Nov-2011 13:30:39] <blankaman> will take a look. is this from scratch?

[16-Nov-2011 13:30:49] <blankaman> or does it borrow from the original example

[16-Nov-2011 13:31:36] <blankaman> zykes-: sorry I should be prepending your nick

[16-Nov-2011 13:32:14] <blankaman> zykes-: also curious do you work for zenoss?

[16-Nov-2011 14:18:57] kocolosk|away is now known as kocolosk

[16-Nov-2011 14:27:23] <blankaman> zykes-: do i need to populate the .zenossrc with user=, password= and url= ?

[16-Nov-2011 14:46:51] kocolosk is now known as kocolosk|away

[16-Nov-2011 14:47:28] kocolosk|away is now known as kocolosk

[16-Nov-2011 14:48:00] <zykes-> if you don't provide it when instantiating the api binding i think yeah

[16-Nov-2011 14:48:10] <zykes-> blankaman: no i don't work for them

[16-Nov-2011 15:31:48] <zykes-> fragfutter: ?

[16-Nov-2011 16:00:07] kocolosk is now known as kocolosk|away

[16-Nov-2011 16:00:16] <johnnynoc> my question about the custom properties was an id10t error

[16-Nov-2011 16:12:54] <blankaman> zykes-: do you have any documentation for your api?

[16-Nov-2011 16:13:01] <blankaman> zykes-: slightly lost

[16-Nov-2011 16:20:03] <wvcain> Having trouble packing my zeo database.  trying "zeopack -p 8100" as user zenoss  from /usr/local/zenoss/python/bin/  It times out.  Ideas?

[16-Nov-2011 16:21:12] <dhopp> wvcain:  I think you need zeopack -h localhost -p 8100

[16-Nov-2011 16:22:08] <wvcain> The script is kind enough to tell me that I am using the localhost; rather `localhost` which resolves and pings

[16-Nov-2011 16:22:28] <wvcain> That said, I just had success...

[16-Nov-2011 16:22:49] <wvcain> rather I got my prompt back with out an error.

[16-Nov-2011 16:24:15] <wvcain> And I have a started and complete in the log...  thanks for instant gratification.

[16-Nov-2011 17:00:59] kocolosk|away is now known as kocolosk

[16-Nov-2011 17:37:37] kocolosk is now known as kocolosk|away

[17-Nov-2011 00:00:01] [disconnected at Thu Nov 17 00:00:01 2011]

[17-Nov-2011 00:00:02] [connected at Thu Nov 17 00:00:02 2011]

[17-Nov-2011 00:00:13] [zenoss-logger (logger bot) has joined #zenoss]

[17-Nov-2011 02:39:58] <fragfutter> zykes-: now

[17-Nov-2011 02:50:25] <zykes-> fragfutter: got a suggestion on what i need to do to get your lldp thing to work with procurve switches?

[17-Nov-2011 02:50:48] <fragfutter> zykes-: should work right out of the box.

[17-Nov-2011 02:53:46] <fragfutter> zykes-: you need to add the modeller to your deviceclass

[17-Nov-2011 03:17:33] <thevoke_>   /win 111

[17-Nov-2011 03:18:09] <fragfutter> thevoke_: no, the winning number is 8812482

[17-Nov-2011 03:33:53] <zykes-> fragfutter: i've done that, still doesn't work

[17-Nov-2011 03:34:03] <fragfutter> so what is your problem?

[17-Nov-2011 05:08:06] <zykes-> fragfutter: it doesn't seem to be working :/

[17-Nov-2011 05:10:23] <fragfutter> zykes-: is the modeller running? does the implement .1.0.8802.

[17-Nov-2011 05:11:05] <zykes-> how to check fragfutter ?

[17-Nov-2011 05:11:31] <fragfutter> run modeller from CLI, do an snmpwalk

[17-Nov-2011 05:54:09] <rcngeoff> Hi... I'm having some graph report trouble.  I have a multi-graph configured for load (laLoadInt1) with an RPN value of "100,/" but it seems to only be converting the "max" values.  The avg and cur values do not appear in a accurate manner.

[17-Nov-2011 05:54:16] <rcngeoff> any ideas?

[17-Nov-2011 07:04:30] <rcngeoff> fwiw, I just discovered the answer to my own question... the real problem was the format option had %s, which in case I did not want.  so the values were being scaled incorrectly.

[17-Nov-2011 07:04:44] <rcngeoff> in case/in this case

[17-Nov-2011 08:07:19] <enkrypt> I've loaded some MIB's into zenoss, but my snmp traps keep containing the numerical OID, and not the corresponding explanation. do I need to restart anything for this to work? I am using distributed collectors, do I need to update them in order for the MIBs to take effect?

[17-Nov-2011 08:12:23] <enkrypt> should I add mibs through the GUI, or with zenmib command?

[17-Nov-2011 08:13:01] <enkrypt> I imported a few dozen MIB's succesfully, but they don't seem to work, my traps remain numeric. when I run "zenmib run -v10" it tells me my /opt/zenoss/share/mibs/site is empty

[17-Nov-2011 08:49:34] <jmp242> morning all

[17-Nov-2011 08:52:33] <rocket> morning jmp242

[17-Nov-2011 08:53:58] <Hackman238> Morning!

[17-Nov-2011 08:56:42] <ion-fan> good early evening^^

[17-Nov-2011 08:57:35] <jmp242> I wish it was good lol

[17-Nov-2011 08:59:05] <Hackman238> How goes the war? Glorious I hope.

[17-Nov-2011 08:59:19] <Hackman238> ion-fan: Where are you at?

[17-Nov-2011 08:59:19] <jmp242> Not so much yesterday and today

[17-Nov-2011 08:59:33] <jmp242> my desktop decided to hate me and only will log in if I disconnect from the network

[17-Nov-2011 08:59:38] <Hackman238> jmp242: Sorry to hear. Been in the trenchs myself.

[17-Nov-2011 09:00:09] <Hackman238> jmp242: Huh. Thats odd. Have you tried 'net Workstation restart'?

[17-Nov-2011 09:00:10] <jmp242> and some Win7 computers are deciding they intermittantly, (once a day or a couple times a week), don't want to continue to talk to our samba server via the samba cname

[17-Nov-2011 09:00:21] <jmp242> which is more important lol

[17-Nov-2011 09:00:48] <Hackman238> jmp242: Ah yeah. Theres a reason for that

[17-Nov-2011 09:00:50] <jmp242> it's so weird, I wanted help from MS technet:

[17-Nov-2011 09:00:57] <ion-fan> hackman238: thank you now the zenpack runs good!!! but there are some design bugs... and after installation i had to copy my old version of transforms into zenoss because the new one doesnt work either...

[17-Nov-2011 09:00:57] <jmp242> but not so much help so far ... lol

[17-Nov-2011 09:01:55] <ion-fan> hackman238: the design bug is that there is not enough space to view all information in the search field because it is right aligned

[17-Nov-2011 09:01:59] <Hackman238> jmp242: I hate to desicrate our lovely channel with this, you have a WINS server?

[17-Nov-2011 09:02:05] <jmp242> yea

[17-Nov-2011 09:02:20] <jmp242> still some NT4 so . . .

[17-Nov-2011 09:02:31] <Hackman238> jmp242: Your SMB, no matter what version of Windows serving, will be a nightmare without WINS properly working.

[17-Nov-2011 09:02:43] <Hackman238> jmp242: Good ole NT4

[17-Nov-2011 09:02:53] <jmp242> yea lol

[17-Nov-2011 09:03:09] <jmp242> were in a drawn out migration from our NT4 domain to our AD domain lol

[17-Nov-2011 09:03:17] <Hackman238> jmp242: I guess so

[17-Nov-2011 09:03:31] <Hackman238> jmp242: Heres the problem. The windows DNS resolver is very stupid

[17-Nov-2011 09:03:52] <Hackman238> jmp242: You'll need to make sure that the cname and netbios name of the share host match

[17-Nov-2011 09:03:57] <ion-fan> hackman238: can i do a quick hack in the python file to change the appeareance of the search field?

[17-Nov-2011 09:04:03] <Hackman238> jmp242: Otherwise it'll be a battle of negative caches

[17-Nov-2011 09:04:26] <Hackman238> ion-fan: Which pack?

[17-Nov-2011 09:04:43] <Hackman238> ion-fan: DeviceSearch?

[17-Nov-2011 09:05:06] <ion-fan> yes

[17-Nov-2011 09:05:56] <ion-fan> hackman238: the search field in the top bar is too near to the right end of the window...

[17-Nov-2011 09:06:34] <Hackman238> ion-fan: Oh gotcha. Let me take a glance

[17-Nov-2011 09:06:41] <Hackman238> jmp242: Do they match?

[17-Nov-2011 09:07:38] <ion-fan> hackman238: thank you .... ill post the problem with the transform in the forums.. so that you can take a look at it when you've got time to do it...

[17-Nov-2011 09:07:48] <Hackman238> ion-fan: Thanks man

[17-Nov-2011 09:14:06] <jmp242> let me look, thanks Hackman238

[17-Nov-2011 10:36:18] mau is now known as Guest5527

[17-Nov-2011 10:36:36] <Guest5527> Can any one help me how to add external URLS to zenoss moniotoring

[17-Nov-2011 10:41:16] <Guest5527> any help is much appreciated

[17-Nov-2011 10:50:37] <Guest5527> is there any one from Zenoss Support team

[17-Nov-2011 11:07:13] <Hackman238> Guest5527:

[17-Nov-2011 11:07:17] <Hackman238> Guest5527: Hey there

[17-Nov-2011 11:08:03] <Hackman238> Guest5527: You'll want to use this zenpack

[17-Nov-2011 11:08:20] <Guest5527> yeah

[17-Nov-2011 11:08:23] <Guest5527> Hackman

[17-Nov-2011 11:09:42] <Hackman238> Guest5527: Let me know if you need help

[17-Nov-2011 11:10:27] <Guest5527> sure:

[17-Nov-2011 11:15:10] <Guest5527> @Hackman23:I have insatlled all the zenpacks in th esystem

[17-Nov-2011 11:17:18] <Guest5527> can you please help me out in this

[17-Nov-2011 11:23:01] <Hackman238> Guest5527: Okay. What you'll want to do is go to the device that you wish to monitor the website from and add a new local template. In this template add a new HTTP datasource and populate the required fields.

[17-Nov-2011 11:23:50] <Guest5527> this is the external URL hosted on External server

[17-Nov-2011 11:24:20] <Hackman238> Guest5527: Thats ok

[17-Nov-2011 11:24:38] <Hackman238> Guest5527: In that case just pick a device you'd preffer to be associated with the test

[17-Nov-2011 11:24:43] <Guest5527> that too its an ssl

[17-Nov-2011 11:24:56] <Hackman238> Guest5527: The test supports ssl

[17-Nov-2011 11:24:59] <Guest5527> i dont want to assosiate any device on that

[17-Nov-2011 11:26:04] <Guest5527> can i add a device with the name test in www class and test it'

[17-Nov-2011 11:26:50] <Hackman238> Guest5527: Yep, if you like. On this dummy device be sure to set its zPingIgnore and zSnmpIgnore to true so it wont bug you about being down

[17-Nov-2011 11:28:11] <Guest5527> but ti want the trigger if the URL goes down

[17-Nov-2011 11:36:42] <Lorin> @Hackman:I am guest

[17-Nov-2011 11:36:45] <Lorin> its dc

[17-Nov-2011 11:36:54] <Lorin> can you please help me out in this

[17-Nov-2011 11:45:53] <nyeates> Hiya anyone

[17-Nov-2011 11:49:11] <Hackman238> Lorin: Hey

[17-Nov-2011 11:49:21] <Hackman238> nyeates: Hey Nick

[17-Nov-2011 11:49:28] <Hackman238> nyeates: How goes the war?

[17-Nov-2011 11:49:47] <jmp242> hi nyeates

[17-Nov-2011 11:51:44] <fragfutter> i somehow wonder why hackman is always asking about the war. maybe he is some kind of arms-dealer.

[17-Nov-2011 11:52:13] <jmp242> it's the constant IT war

[17-Nov-2011 11:52:21] <jmp242> against users, vendors, and buggy products

[17-Nov-2011 11:52:36] <Hackman238> fragfutter: Because I'm the worst kind of person...small business entrepreneur

[17-Nov-2011 11:53:59] <Hackman238> E.g. beating quarterly expectations or winning a large customer is equal to winning a glorious battle.

[17-Nov-2011 11:54:22] <Lorin> @nick:can u please hlep me how to monitor external URL

[17-Nov-2011 11:54:23] <Lorin> please

[17-Nov-2011 11:55:36] <Hackman238> Lorin:

[17-Nov-2011 11:55:54] <Hackman238> Lorin: Half way down at 'What is HTTP Monitior'

[17-Nov-2011 11:55:59] <Hackman238> Lorin: Step by step

[17-Nov-2011 11:56:36] <Lorin> @hackman:I gone though dan

[17-Nov-2011 11:57:57] <Hackman238> Lorin: What part do you need clarification on?

[17-Nov-2011 11:58:18] <Lorin> I need clarification on SSL monoitoring the external servers

[17-Nov-2011 11:59:01] <Lorin> external URLS

[17-Nov-2011 11:59:32] <Hackman238> Lorin: When setting up the datasource put in the external URL to monitor in the hostname field of the datasource and check off 'SSL' box.

[17-Nov-2011 12:00:01] [disconnected at Thu Nov 17 12:00:01 2011]

[17-Nov-2011 12:00:02] [connected at Thu Nov 17 12:00:02 2011]

[17-Nov-2011 12:00:16] [zenoss-logger (logger bot) has joined #zenoss]

[17-Nov-2011 12:01:41] <Lorin> k

[17-Nov-2011 12:03:33] <Hackman238> Lorin: I'd also set the ip field to the IP the domain resolves to. I do not know if that will matter.

[17-Nov-2011 12:03:41] <Hackman238> Anyone know? ^

[17-Nov-2011 12:06:10] <Lorin> hmm will chk this

[17-Nov-2011 13:05:31] <Hackman238> All - Help stop the internet censorship bill! -

[17-Nov-2011 13:05:46] <Hackman238> No, not spam

[17-Nov-2011 14:30:46] kocolosk|away is now known as kocolosk

[17-Nov-2011 14:42:37] <opapo> Hello all

[17-Nov-2011 14:43:27] <opapo> How do you snmpwalk an address that is not the localhost?

[17-Nov-2011 14:43:57] <opapo> changing agentaddress in the snmpd.conf doesn't work

[17-Nov-2011 14:44:24] <opapo> changing snmpdopts in /etc/defaults/snmpd doesn't work either

[17-Nov-2011 14:45:09] <Hackman238> opapo: snmpwalk -v2c -c community deviceip

[17-Nov-2011 14:45:51] <opapo> Hackman238: I get a timeout error with the deviceip

[17-Nov-2011 14:45:59] <nyeates> you can run snmpwalk commands in the zenoss UI too...

[17-Nov-2011 14:46:00] <opapo> I can only get localhost to work

[17-Nov-2011 14:46:23] <nyeates> if you cant get it to work at command line have network or network settings issues likely

[17-Nov-2011 14:46:48] <opapo> I am checking the end node on the end node

[17-Nov-2011 14:46:57] <nyeates> make sure host and DNS are all set correctly....try pinging machines you are trying to snmpwalk - can you even reach them?

[17-Nov-2011 14:47:07] <nyeates> if you can ping them....what is snmpwalk saying?

[17-Nov-2011 14:48:14] <opapo> If I am on hostA and I am using "snmpwalk -v2c -c public hostA" I get a timeout

[17-Nov-2011 14:48:30] <opapo> If I am on hostA and I am using "snmpwalk -v2c -c public localhost" I get data

[17-Nov-2011 14:49:56] <Hackman238> opapo: Does hostA have snmpd running? Is hostA snmp community = public? Does hostA ahve iptables running? If so, you'll need to iptables --list and ensure a rule for port 161 exists

[17-Nov-2011 14:50:48] <opapo> snmpd is running. the community is correct.  iptables has been flushed

[17-Nov-2011 14:51:13] <opapo> There was a setting in /etc/default/snmpd that had to be changed I thought

[17-Nov-2011 14:57:19] <Hackman238> opapo: Shouldnt be

[17-Nov-2011 14:57:23] <Hackman238> Not to jsut hit it

[17-Nov-2011 14:57:40] <Hackman238> opapo: you'll need to tweak snmpd.conf for some platforms to get the whole tree

[17-Nov-2011 14:57:56] <Hackman238> opapo: Can you telnet hostA 161?

[17-Nov-2011 14:58:10] <Hackman238> opapo: If it connects ok you'll get an escape character notification

[17-Nov-2011 15:01:19] <opapo> "telnet ipaddr 161" returns "Connetion refused"

[17-Nov-2011 15:01:57] <opapo> "iptables -L" shows no rules

[17-Nov-2011 15:02:40] <opapo> "telnet localhost 161" returns "connection refused"

[17-Nov-2011 15:03:29] <opapo> is this because telnet works in udp not tcp?

[17-Nov-2011 15:03:38] <Hackman238> opapo: are there routers or firewalls between these two devices?

[17-Nov-2011 15:03:47] <opapo> this is localhost

[17-Nov-2011 15:04:28] <Hackman238> opapo: snmp is udp

[17-Nov-2011 15:05:24] <Hackman238> opapo: Whoops. telnet is tcp

[17-Nov-2011 15:05:26] <opapo> "netstat -uln" shows

[17-Nov-2011 15:05:29] <Hackman238> *face palm*

[17-Nov-2011 15:05:30] <Hackman238>

[17-Nov-2011 15:06:13] <Hackman238> snmpwalk working for localhost?

[17-Nov-2011 15:06:21] <opapo> yes

[17-Nov-2011 15:06:42] <mloven> "service iptables stop" and try again?

[17-Nov-2011 15:06:44] <Hackman238> Okay. I think theres a router or bridge or another blocking device between your box and the other device you want to test

[17-Nov-2011 15:07:09] <opapo> they are the same box

[17-Nov-2011 15:07:29] <opapo> works, but not the actual ip

[17-Nov-2011 15:07:36] <Hackman238> opapo: Oh

[17-Nov-2011 15:07:40] <Hackman238> opapo: That wont work

[17-Nov-2011 15:07:43] <Hackman238> LOL

[17-Nov-2011 15:08:25] <opapo> what are the settings in snmpd.conf that I need to change?

[17-Nov-2011 15:10:21] <Hackman238> opapo: I'm not sure what you're needing to change

[17-Nov-2011 15:10:48] <Hackman238> opapo: I know on my boxes I can't snmpwalk my local ip if I bind snmpd to

[17-Nov-2011 15:10:59] <Hackman238> opapo: Not without messing with my routing table

[17-Nov-2011 15:11:34] <Hackman238> opapo: Only way works for me is or via the v6 loopback

[17-Nov-2011 15:11:48] <Hackman238> Anyone know how to get this to work?

[17-Nov-2011 17:04:40] <opapo> If you have a project that requires you to use the ip address, make sure you know the correct ip.

[17-Nov-2011 17:37:45] <opapo> When I click on a device I can click on Graphs and see the performance graphs.

[17-Nov-2011 17:38:05] <opapo> If I click on interfaces I can see the throughput for an interface.

[17-Nov-2011 17:38:20] <opapo> How do I see all of those in one list?

[17-Nov-2011 17:41:31] <opapo> Sorry I got that.

[17-Nov-2011 18:27:55] kocolosk is now known as kocolosk|away

[17-Nov-2011 19:35:36] kocolosk|away is now known as kocolosk

[17-Nov-2011 21:38:20] kocolosk is now known as kocolosk|away

[17-Nov-2011 21:39:58] kocolosk|away is now known as kocolosk

[17-Nov-2011 22:04:25] <milad> Hi, anyone knows how can I just see production servers in event console ?

[17-Nov-2011 22:05:25] <milad> in zenoss API in querying event there is a criteria option but I dunno who can I send it through API

[18-Nov-2011 00:00:01] [disconnected at Fri Nov 18 00:00:01 2011]

[18-Nov-2011 00:00:01] [connected at Fri Nov 18 00:00:01 2011]

[18-Nov-2011 00:00:13] [zenoss-logger (logger bot) has joined #zenoss]

[18-Nov-2011 03:21:06] <zykes-> jmp242: around ?

[18-Nov-2011 05:07:01] <danuvius> Good morning

[18-Nov-2011 05:07:53] <Bart> is there a script to properly restart zenoss? regular restart doesn't wait for zenhub to terminate before starting the processes again

[18-Nov-2011 05:08:38] <danuvius> In the infrastructure overview with some devices the oid is displayed instead of the name inf the hardware model column. How to fix?

[18-Nov-2011 05:11:15] <danuvius> Bart: I'm using the ubuntu stack. and when using the init script it restarts correctly..

[18-Nov-2011 08:11:22] <jmp242> Bart: zenoss restart always works for me . . .

[18-Nov-2011 09:03:09] kocolosk is now known as kocolosk|away

[18-Nov-2011 09:09:22] <Simon4> hey all… anyone had zenperfsnmp silently fail to collect for a device, yet if you run zenperfsnmp from the command-line manually it works just fine?

[18-Nov-2011 09:10:27] * Simon4 changed the snmp v3 priv password yesterday for a bunch of devices and now they're all in that state :/

[18-Nov-2011 09:10:44] kocolosk|away is now known as kocolosk

[18-Nov-2011 09:29:49] <Jane_Curry> Hiya

[18-Nov-2011 10:16:57] <danuvius> simon4: At the moment I also have some issues with snmp3 and collection. When running from cli on all devices it fails. However one device at a time doesn't give any problems at all..

[18-Nov-2011 10:17:21] <danuvius> simon4: I haven't been able to fix it though

[18-Nov-2011 10:17:32] <Simon4> mine was fine until I added priv to our switches, so they're now authpriv

[18-Nov-2011 10:17:37] <Simon4> then it all fell apart :/

[18-Nov-2011 10:33:18] * Simon4 finds it

[18-Nov-2011 10:33:37] <jmp242> what did you find?

[18-Nov-2011 10:33:49] <Simon4> so if you have incorrect v3 auth parameters, zenperfsnmp doesn't report such things, it just tries said devices indefinitely which stops collection for pretty much everything for that cycle as things all time out

[18-Nov-2011 10:34:05] <jmp242> oh

[18-Nov-2011 10:34:09] * Simon4 fixed the class of devs with the typo'ed password and magically it all started again

[18-Nov-2011 10:34:10] <jmp242> report the bug?

[18-Nov-2011 10:34:13] <Simon4> that's a crappy bug

[18-Nov-2011 10:34:45] <Simon4> I have a horrible feeling it's been reported, and I've found it before.. it's just been 6 months since I did anything this serious with zenoss

[18-Nov-2011 10:35:14] <jmp242> oh

[18-Nov-2011 10:35:15] <jmp242>

[18-Nov-2011 10:35:43] * Simon4 will check though

[18-Nov-2011 10:35:50] <Simon4> we're also back on 2.5.2, so it may have been sorted in future versions

[18-Nov-2011 10:35:56] <jmp242> oh

[18-Nov-2011 10:35:59] <Simon4> about to fire up some avalon on a box anyway, so will test it

[18-Nov-2011 10:35:59] <jmp242> well, that is old

[18-Nov-2011 10:36:10] <Simon4>

[18-Nov-2011 10:42:57] <jmp242> I can't wait for betas of 4.2

[18-Nov-2011 10:43:09] <jmp242> I'll finally spin up a rackspace slice or EC2 instance to test on

[18-Nov-2011 11:07:34] <designated> Hackman238, you around?

[18-Nov-2011 11:14:10] <Jane_Curry> Anyone else here seen problems with zentrap on 3.2.1????

[18-Nov-2011 12:00:01] [disconnected at Fri Nov 18 12:00:01 2011]

[18-Nov-2011 12:00:02] [connected at Fri Nov 18 12:00:02 2011]

[18-Nov-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]

[18-Nov-2011 12:00:15] <> [freenode-info] why register and identify? your IRC nick is how people know you.

[18-Nov-2011 13:04:43] <nyeates> hi

[18-Nov-2011 13:06:52] <Jane_Curry> hi nick

[18-Nov-2011 14:16:28] <nexex> ohh, now I really wish I was going

[18-Nov-2011 14:43:57] <jmp242> Hey all

[18-Nov-2011 14:44:26] <jmp242> rocket, rmatte, Hackman238 ... Anyone see how some IP services (like snmp) don't let you select the monitoring status in 3.2.1?

[18-Nov-2011 14:52:36] <mattt> derp, install of zenoss on a server w/ 256 MB fails

[18-Nov-2011 14:52:42] <mattt> *installation

[18-Nov-2011 15:00:27] <jmp242> lol

[18-Nov-2011 15:00:30] <jmp242> yea, it will

[18-Nov-2011 15:06:14] <mattt> probably ought to read before i try to install stuff

[18-Nov-2011 15:08:23] <jmp242> it is important to check out the docs for Zenoss

[18-Nov-2011 15:08:39] <mattt> yep, reading them now

[18-Nov-2011 15:08:47] <mattt> is there a preferred platform to run zenoss on?

[18-Nov-2011 15:09:09] <jmp242> Well

[18-Nov-2011 15:09:20] <jmp242> I prefer a Redhat derivative

[18-Nov-2011 15:09:26] <jmp242> Scientific Linux 5 is what I use

[18-Nov-2011 15:09:32] <jmp242> CENTOS will be about the same

[18-Nov-2011 15:17:00] <mattt> jmp242: fair enough, thanks

[18-Nov-2011 16:17:34] <Glacee> gI cant wait to try the swift zenpack

[18-Nov-2011 17:20:42] <cymruman> can someone help me with an "SNMP Agent Down" error please

[18-Nov-2011 17:25:29] <mattt> cymruman: new to zenoss, but what you got?

[18-Nov-2011 17:25:29] kocolosk is now known as kocolosk|away

[18-Nov-2011 17:28:15] <cymruman> I have 1 device out of 4 setup that show snmp agent down

[18-Nov-2011 17:31:16] <mattt> cymruman: and snmpd is def. not down on that 1 host? 

[18-Nov-2011 17:32:48] <cymruman> no, I checked and it shows that its running

[18-Nov-2011 17:33:19] <mattt> cymruman: it's not bound to localhost only or anything is it?

[18-Nov-2011 17:33:30] <mattt> i just installed on debian, and noticed snmpd binds to localhost only by default

[18-Nov-2011 17:34:03] <cymruman> where do I check that

[18-Nov-2011 17:34:35] <mattt> cymruman: what are you running?

[18-Nov-2011 17:35:00] <cymruman> zenoss on redhat 5.3

[18-Nov-2011 17:35:19] <mattt> cymruman: the host that's down i mean

[18-Nov-2011 17:35:36] <cymruman> redhat 5.3

[18-Nov-2011 17:37:39] <mattt> try this

[18-Nov-2011 17:37:42] <mattt> from the zenoss host

[18-Nov-2011 17:38:05] <mattt> snmpwalk -c public -v1 <ip of box w/ snmp down>

[18-Nov-2011 17:39:09] <cymruman> ok

[18-Nov-2011 17:39:11] <cymruman> done

[18-Nov-2011 17:39:27] <mattt> and you got a response?

[18-Nov-2011 17:39:41] <cymruman> yeah, a ton of stuff

[18-Nov-2011 17:39:52] <mattt> so it's up, not sure why zenoss is reporting it down

[18-Nov-2011 17:39:59] <mattt> i'm new to zenoss, just installed it this evening

[18-Nov-2011 17:41:03] <cymruman> where do I check the localhost thing you were talking about

[18-Nov-2011 17:41:58] <mattt> unnecessary, if you can connect to snmpd from the zenoss host using snmpwalk

[18-Nov-2011 17:42:50] <cymruman> oh

[18-Nov-2011 17:43:10] <mattt> cymruman: what if you accept and close the alert?

[18-Nov-2011 17:43:11] <mattt> does it come back?

[18-Nov-2011 17:43:27] <cymruman> yes

[18-Nov-2011 17:43:47] <mattt> not sure, hopefully someone more experienced can help :/

[18-Nov-2011 17:44:35] <cymruman> ok, thanks

[19-Nov-2011 00:00:01] [disconnected at Sat Nov 19 00:00:01 2011]

[19-Nov-2011 00:00:02] [connected at Sat Nov 19 00:00:02 2011]

[19-Nov-2011 00:00:16] [zenoss-logger (logger bot) has joined #zenoss]

[19-Nov-2011 01:24:58] <Tabmow> Anyone good with the regexp patterns?

[19-Nov-2011 01:25:59] <Tabmow> <-- not sure why this isn't working

[19-Nov-2011 12:00:01] [disconnected at Sat Nov 19 12:00:01 2011]

[19-Nov-2011 12:00:02] [connected at Sat Nov 19 12:00:02 2011]

[19-Nov-2011 12:00:13] [zenoss-logger (logger bot) has joined #zenoss]

[19-Nov-2011 12:51:07] <faiz> Good evening all

[19-Nov-2011 13:09:00] <faiz> hi

[19-Nov-2011 14:14:33] tvincent_ is now known as tvincent

[20-Nov-2011 00:00:01] [disconnected at Sun Nov 20 00:00:01 2011]

[20-Nov-2011 00:00:02] [connected at Sun Nov 20 00:00:02 2011]

[20-Nov-2011 00:00:12] [zenoss-logger (logger bot) has joined #zenoss]

[20-Nov-2011 01:07:45] <zooko> Is it true that you actually *require* flash to use Zenoss?

[20-Nov-2011 01:07:53] <zooko> I'm reading the install docs for 3.2.x and it seems to say that.

[20-Nov-2011 05:32:33] <faiz> Hi All

[20-Nov-2011 12:00:01] [disconnected at Sun Nov 20 12:00:01 2011]

[20-Nov-2011 12:00:01] [connected at Sun Nov 20 12:00:01 2011]

[20-Nov-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]

[21-Nov-2011 00:00:01] [disconnected at Mon Nov 21 00:00:01 2011]

[21-Nov-2011 00:00:01] [connected at Mon Nov 21 00:00:01 2011]

[21-Nov-2011 00:00:02] [disconnected at Mon Nov 21 00:00:02 2011]

[21-Nov-2011 00:00:03] [connected at Mon Nov 21 00:00:03 2011]

[21-Nov-2011 00:00:16] [zenoss-logger (logger bot) has joined #zenoss]

[21-Nov-2011 03:36:15] Guest55170 is now known as bzed

[21-Nov-2011 04:12:24] Tabstar is now known as Guest75188

[21-Nov-2011 06:02:06] <mf2hd> hi, has any1 succeeded with windows 2008 & winexe? I'm trying to collect iperf data with winexe and all i get is "Error: error Creating process(iperf <params>)" 2

[21-Nov-2011 06:38:07] <zykes-> winexe?

[21-Nov-2011 08:15:40] <Hackman238> Morning all!

[21-Nov-2011 08:34:48] <jmp242> morning!

[21-Nov-2011 08:35:02] <jmp242> final rundown to thanksgiving here in the US

[21-Nov-2011 08:39:15] kocolosk|away is now known as kocolosk

[21-Nov-2011 09:02:31] <Hackman238> jmp242: That it is

[21-Nov-2011 09:04:27] <jmp242> Hey, do you have a minute to look at a couple forum threads?

[21-Nov-2011 09:04:58] <Hackman238> jmp242: Sure

[21-Nov-2011 09:05:02] <Hackman238> jmp242: Hows by you?

[21-Nov-2011 09:05:33] <jmp242> mmm, waking up lol

[21-Nov-2011 09:05:39] <jmp242>

[21-Nov-2011 09:05:53] <jmp242> this one had a final query at the end, maybe should have been a new thread

[21-Nov-2011 09:06:28] <jmp242> This one I can confirm on 3.2.1 - it's quite odd as it doesn't apply consistently on my install, but to some processes...

[21-Nov-2011 09:10:32] <Hackman238> jmp242: Off the top of your head do you know how to force zenoss to use UTC time?

[21-Nov-2011 09:11:16] <Hackman238> jmp242: I have a mom with torches and pitch forks since Zenoss is paging first shift folks 6 hours earlier than scheduled since we're in GMT-6.

[21-Nov-2011 09:11:21] <Hackman238> *mob

[21-Nov-2011 09:15:18] <jmp242> I was wondering about a MOM lol

[21-Nov-2011 09:15:26] <jmp242> no, can't say that I know how to make zenoss use UTC

[21-Nov-2011 09:16:20] <Hackman238> LOL

[21-Nov-2011 09:16:27] <Hackman238> jmp242: Alrighty, thanks anyhow man

[21-Nov-2011 09:19:16] <Hackman238> jmp242: Tricky threads

[21-Nov-2011 09:19:48] <jmp242> yea

[21-Nov-2011 09:19:59] <jmp242> The second one interests me as well, as I see it on my install

[21-Nov-2011 09:20:20] <jmp242> where some processes I can select / set monitoring for, others, like smtp, I can't - it's greyed out with no monitored status

[21-Nov-2011 09:20:21] <Hackman238> jmp242: I don't think theres a solution to that

[21-Nov-2011 09:20:32] <jmp242> too bad - should I report as a bug?

[21-Nov-2011 09:20:38] <jmp242> or is it intended?

[21-Nov-2011 09:20:59] <Hackman238> jmp242: I see the same thing in Ent 4.1.1

[21-Nov-2011 09:21:12] <Hackman238> jmp242: It might be intended, though I don't understand the logic

[21-Nov-2011 09:21:36] <jmp242> me either

[21-Nov-2011 09:21:38] <Hackman238> jmp242: I can't claim to make use of that functionality enough to be sure though

[21-Nov-2011 09:22:02] <jmp242> yea, to be honest, I don't play with it a lot

[21-Nov-2011 09:22:17] <jmp242> but I can see it potentially being important for some users or in some instance down the road

[21-Nov-2011 09:22:33] <jmp242> It just would seem odd to use an event transform to supress that when the GUI implies you should be able to manage it

[21-Nov-2011 09:25:27] <Hackman238> jmp242: true

[21-Nov-2011 09:33:42] <mattt> seeing this in zenperfsnmp.log: ERROR zen.zenperfsnmp: string is too long to send

[21-Nov-2011 09:33:45] <mattt> any idea?

[21-Nov-2011 09:43:53] * froztbyte frowns at his zenoss' zentrap

[21-Nov-2011 09:49:40] <Hackman238> mattt: A string being returned is too large to return. Are you using 64-bit counters?

[21-Nov-2011 09:49:46] <Hackman238> froztbyte: Whats up?

[21-Nov-2011 09:50:33] <froztbyte> it doesn't appear to be processing traps

[21-Nov-2011 09:50:42] <froztbyte> even with captureAll

[21-Nov-2011 09:50:55] <froztbyte> unless captureAll does not-what-I-think-it-does

[21-Nov-2011 09:51:12] <mattt> Hackman238: no idea, how do i check?  :/

[21-Nov-2011 09:51:18] <froztbyte> gonna write some stuff out now with captureFilePrefix though

[21-Nov-2011 09:51:21] <froztbyte> see what's potting

[21-Nov-2011 09:51:32] <froztbyte> (logging already turned up to 11 (Debug), no useful tips there)

[21-Nov-2011 09:51:37] <Hackman238> mattt: Are you using snmp v2c?

[21-Nov-2011 09:51:45] <Hackman238> mattt: And whats the source device?

[21-Nov-2011 09:52:09] <Hackman238> froztbyte: Not good. Did you get an unjellyable error in zenhuB?

[21-Nov-2011 09:52:13] <mattt> Hackman238: it doesn't actually say what device is tripping it up

[21-Nov-2011 09:52:48] <mattt> Hackman238: but yes, v2

[21-Nov-2011 09:52:53] <mattt> (v2c)

[21-Nov-2011 09:53:00] <froztbyte> Hackman238: no, I just noticed that it doesn't appear to be doing anything

[21-Nov-2011 09:53:22] <froztbyte> *frown*

[21-Nov-2011 09:54:24] <Hackman238> mattt: as the zenoss user, zenperfsnmp debug

[21-Nov-2011 09:54:53] <Hackman238> mattt: tail -f the zenperfsnmp log and grep for the error. When it pops back up we should have more debugging info

[21-Nov-2011 09:55:13] <froztbyte> what in the actual fuck is this thing doing

[21-Nov-2011 09:55:27] <Hackman238> froztbyte: If you zenhub debug then stop zentrap and zentrap run now -v10, anything interesting show up in zenhub log?

[21-Nov-2011 09:55:34] <froztbyte> the daemon's running, the process ID's that are listening on :162 match up with what the rest of the things are

[21-Nov-2011 09:55:41] <froztbyte> yet there's nothing pitching up in the logs

[21-Nov-2011 09:55:56] <froztbyte> Hackman238: yeah, I was gonna go with that now

[21-Nov-2011 09:56:02] <froztbyte> bloody silly of it, really

[21-Nov-2011 09:56:16] <Hackman238> froztbyte: Yep

[21-Nov-2011 09:57:06] <mattt> Hackman238: have limited access to this box, will let you know once i've run the zenperfsnmp (thanks btw!)

[21-Nov-2011 09:57:48] <Hackman238> mattt: ty

[21-Nov-2011 09:58:08] <froztbyte> 2011-11-21 16:57:30,881 INFO zen.ZenTrap: getting OID -> name mappings

[21-Nov-2011 09:58:11] <froztbyte> 2011-11-21 16:57:31,205 DEBUG zen.ZenTrap: Sent a 'stop' event

[21-Nov-2011 09:58:14] <froztbyte> 2011-11-21 16:57:31,206 DEBUG zen.ZenTrap: stop() called when not running

[21-Nov-2011 09:58:17] <froztbyte> 2011-11-21 16:57:31,206 INFO zen.ZenTrap: Daemon ZenTrap shutting down

[21-Nov-2011 09:58:20] <froztbyte> when I run it manually

[21-Nov-2011 09:58:37] <Hackman238> froztbyte: Okay thats good. Nothing funky in zenhub?

[21-Nov-2011 09:58:49] <froztbyte> nothing

[21-Nov-2011 09:59:03] <froztbyte> no bitching about incorrect passwords, etc

[21-Nov-2011 09:59:20] <froztbyte> this thing appears to be totally healthy except for the bit where it's supposed to do stuff

[21-Nov-2011 09:59:28] <froztbyte> I'll probably collect some information and post to list

[21-Nov-2011 09:59:28] <Hackman238> froztbyte: XD LOL

[21-Nov-2011 09:59:39] <froztbyte> easier to keep the info persistent that way

[21-Nov-2011 10:00:03] <Hackman238> froztbyte: yeah, good idea

[21-Nov-2011 10:00:26] <Hackman238> froztbyte: snmp-trapd isnt running, is it?

[21-Nov-2011 10:00:44] <froztbyte> nope

[21-Nov-2011 10:00:46] <froztbyte> (did check)

[21-Nov-2011 10:00:58] <Hackman238> froztbyte: and a ps aux | grep zentrap shows nothing?

[21-Nov-2011 10:01:02] <Hackman238> (if you stop it)

[21-Nov-2011 10:01:25] <froztbyte> yup

[21-Nov-2011 10:01:36] <Hackman238> froztbyte: Definately not a port problem?

[21-Nov-2011 10:01:44] <froztbyte> no dead daemon hanging around, no hidden process, no funky lockfile issues that I've seen, nothing

[21-Nov-2011 10:01:50] <froztbyte> Hackman238: not that I can see

[21-Nov-2011 10:02:01] <froztbyte> zentrap process runs on tcp:162 and udp:162

[21-Nov-2011 10:02:10] <froztbyte> then when i stop it, it stops listening there

[21-Nov-2011 10:02:22] <froztbyte> if I `tcpdump -nvi any port 162`, I see those packets

[21-Nov-2011 10:02:33] <Hackman238> Gotcha. For the heck of it...nvm was just going to ask that

[21-Nov-2011 10:02:38] <froztbyte>

[21-Nov-2011 10:02:44] <froztbyte> <froztbyte> what in the actual fuck is this thing doing

[21-Nov-2011 10:02:55] <Hackman238> froztbyte: Nothing from the sounds of it

[21-Nov-2011 10:03:07] <froztbyte> box is good on memory, so it's not swapping issues or whatnot

[21-Nov-2011 10:03:15] <froztbyte> it's just pissing me off, really

[21-Nov-2011 10:03:21] <froztbyte> anyway, zenoss-users soon

[21-Nov-2011 10:05:53] <froztbyte> just need to get my washing first

[21-Nov-2011 10:06:35] <Hackman238> froztbyte: Alrighty

[21-Nov-2011 10:06:42] <Hackman238> froztbyte: No stress

[21-Nov-2011 10:07:58] <mattt> Hackman238: do you just restart zenperfsnmp to get it out of debug?

[21-Nov-2011 10:08:39] <Hackman238> mattt: nein, just zenperfsnmp debug again

[21-Nov-2011 10:08:43] <Hackman238> mattt: toggles debug mode

[21-Nov-2011 10:08:53] <mattt> Hackman238: ah, ok

[21-Nov-2011 10:16:15] <syoma> hi

[21-Nov-2011 10:18:35] <Hackman238> syoma: Hey

[21-Nov-2011 10:18:54] <syoma> as I can indicate that the third fall event generates an alert?

[21-Nov-2011 10:20:46] <Hackman238> syoma: ??

[21-Nov-2011 10:21:29] <syoma> for example when a device goes down

[21-Nov-2011 10:22:02] <syoma> in a time interval of 5 minutes

[21-Nov-2011 10:22:16] <syoma> or more

[21-Nov-2011 10:22:24] <syoma> zenoss generate an alert

[21-Nov-2011 10:22:46] <syoma> is possible to define the threshold

[21-Nov-2011 10:22:48] <syoma> ?

[21-Nov-2011 10:23:06] <mattt> Hackman238: seems to have spat out this wicked huge string ... trying to determine what exactly it is

[21-Nov-2011 10:24:02] <Hackman238> mattt: paste bin it

[21-Nov-2011 10:24:05] <Hackman238> plz

[21-Nov-2011 10:25:26] <mattt> Hackman238: can't, it's that large   but i think the gizt of it is that it's complaining about rrd files not updating

[21-Nov-2011 10:25:32] <mattt> wonder if it's this huge string that's causing it to crap out

[21-Nov-2011 10:25:55] <syoma> Hackman238: do yo have some idea?

[21-Nov-2011 10:25:58] <syoma> XD

[21-Nov-2011 10:27:57] <Hackman238> syoma: Yes, the threshold can be defined at the device template. the alerting conditions can be set seperately on each user or group

[21-Nov-2011 10:30:54] <mattt> Hackman238: btw, i tried to take zenperfsnmp out of debug, but the log file is no longer appending (the process is still running tho)

[21-Nov-2011 10:30:58] <mattt> safe to restart that?

[21-Nov-2011 10:31:15] <Hackman238> mattt: do zenperfsnmp status

[21-Nov-2011 10:31:21] <Hackman238> mattt: To see if it blew up

[21-Nov-2011 10:32:37] <mattt> Hackman238: it's still running, but nothing has been logged for > 10 mins

[21-Nov-2011 10:33:55] <Hackman238> mattt: Might better restart it

[21-Nov-2011 10:34:06] <Hackman238> mattt: Sounds like something went sideways

[21-Nov-2011 10:34:43] <syoma> Hackman238: thanks

[21-Nov-2011 10:35:11] <syoma> in the device classes

[21-Nov-2011 10:35:34] <syoma> I can set the value of the alert condition?

[21-Nov-2011 10:36:28] <Hackman238> syoma: Yeah, you'll need to edit the device template from the 'Advanced' -> 'Monitoring Templates' page

[21-Nov-2011 10:37:09] <Hackman238> syoma: The alerting rule itself can be edited for conditions as well on 'Advanced' -> 'Users' -> Pick a user or group -> 'Alerting rules'

[21-Nov-2011 10:38:10] <syoma> Hackman238: in adavanced tab -> monitoring template -> template of device I have onley set cpu and mem thresholds

[21-Nov-2011 10:38:28] <syoma> not threhold of down device

[21-Nov-2011 10:39:03] <Hackman238> syoma: oh up down device is jsut based on if it can be hit by the daemon.

[21-Nov-2011 10:39:05] <syoma> by the intermittent link that I have a need to alert after 5 minutes the zenoss see that the device is down

[21-Nov-2011 10:39:35] <Hackman238> syoma: Gotcha. In that case go to the alerting rule on the user or group and add the condition 'Count > 3'

[21-Nov-2011 10:39:54] <syoma> thank !

[21-Nov-2011 10:39:58] <Hackman238> syoma: That'll make zenactions wait for the threshold to be true 3 times before emailing/paging

[21-Nov-2011 10:40:14] <Hackman238> NP

[21-Nov-2011 10:40:32] <syoma> Hackman238: thank you very muh

[21-Nov-2011 10:40:35] <syoma> *much

[21-Nov-2011 10:40:35] <syoma> XD

[21-Nov-2011 10:53:02] <dhopp> ugh...

[21-Nov-2011 10:56:11] <dhopp> ok…I need some brain power…anybody got some to spare?

[21-Nov-2011 11:03:03] <whyzgeek> hi guys any body knows how to give privilege to zenusers to create event views. on 2.5.2 they can create alert rules but not event views.

[21-Nov-2011 11:03:57] <Hackman238> dhopp: Whats up?

[21-Nov-2011 11:04:18] <Hackman238> whyzgeek: Hum. Jane would be best cannidate to figure that for youi

[21-Nov-2011 11:04:35] <dhopp> Hackman238:  well I've run into a little bit of a problem with my linux monitoring.

[21-Nov-2011 11:05:17] <whyzgeek> Thanks Hackman238          ,  Jane_Curry any comments?

[21-Nov-2011 11:05:40] <dhopp> We still have some RHEL 4 boxes and it seems that after the server has been up for a while the values for ssCpuIdle,ssCpuSystem and ssCpuUser will start to return 0 always until snmpd is restarted.  Looking into it the net-snmp package basically says that those values shouldn't be used and the Raw values should be...

[21-Nov-2011 11:05:58] <dhopp> however, using Raw the graphs don't make a lot of sense without knowing how many cores are in a server

[21-Nov-2011 11:06:10] <dhopp> so now I'm trying to figure out how to make the graphs make some sense

[21-Nov-2011 11:06:47] <Hackman238> dhopp: Ah gotcha. To be honest...cron a restart of snmpd

[21-Nov-2011 11:07:25] <Hackman238> dhopp: You can also poll to get the number of cores if you like

[21-Nov-2011 11:07:50] <Hackman238> dhopp: I released a modeler and template that uses the colelcted value for this same thing

[21-Nov-2011 11:08:08] <dhopp> Hackman238:  yeah, but then I have to do math on the value and that could have a crap load of overhead

[21-Nov-2011 11:08:32] <Hackman238> dhopp: Only on the graph draw

[21-Nov-2011 11:08:48] <dhopp> although, I guess I could collect the number of cores, store them in an RRD that isn't easily accessible and then have the graph command do the math

[21-Nov-2011 11:09:17] <dhopp> where is your modelere/template?

[21-Nov-2011 11:09:25] <dhopp> modeler*

[21-Nov-2011 11:10:11] <Hackman238> dhopp: What the modeler does it get the core count then store it in a cProp

[21-Nov-2011 11:10:33] <dhopp> ah

[21-Nov-2011 11:10:34] <Hackman238> dhopp: graphs can then use ${here/cPropName} to get the value

[21-Nov-2011 11:11:02] <Hackman238> dhopp: Its on my site

[21-Nov-2011 11:11:16] <Hackman238> dhopp: Make sure you do a back up of zope or test in dev

[21-Nov-2011 11:11:27] <Hackman238> dhopp: It does edit templates and make global changes.

[21-Nov-2011 11:12:11] <dhopp> what templates does it edit?

[21-Nov-2011 11:13:53] <Hackman238> dhopp: Edits cpu templates and adds properties to all devices

[21-Nov-2011 11:14:17] <dhopp> which cpu templates?  Linux and Windows or just Linux?

[21-Nov-2011 11:14:33] <Hackman238> dhopp: I might suggest to just unzip it and look at the snmp value I'm collecting and how I'm using it in the graphs unless you want to go that way 100%

[21-Nov-2011 11:14:47] <Hackman238> dhopp: I forget. Definately linux. Cant recall if windows

[21-Nov-2011 11:15:02] <dhopp> I'll take a look..thanks

[21-Nov-2011 11:15:10] <dhopp> I have a dev instance I can blow up if necessary :-)

[21-Nov-2011 11:23:43] <Hackman238> dhopp:

[21-Nov-2011 11:26:19] <dhopp> looking at your objects.xml it looks like you only mess with the Linux templates

[21-Nov-2011 11:26:33] <dhopp> and Solaris

[21-Nov-2011 11:26:35] <dhopp> so *nix

[21-Nov-2011 11:26:49] <dhopp> I don't have any solaris so I don't care :-)

[21-Nov-2011 11:29:08] <dhopp> actually this ZenPack is relatively simple…basically contains objects.xml, a simple modeler and some custom code in for setting up the environment

[21-Nov-2011 11:29:34] <dhopp> I'm sure it was more work then I'm given credit for to actually do…but it looks simple!

[21-Nov-2011 11:30:45] <Hackman238> dhopp: its a way simple pack. problem is I dont use it myself so I can't recall how it works well.

[21-Nov-2011 12:00:01] [disconnected at Mon Nov 21 12:00:01 2011]

[21-Nov-2011 12:00:02] [connected at Mon Nov 21 12:00:02 2011]

[21-Nov-2011 12:00:13] [zenoss-logger (logger bot) has joined #zenoss]

[21-Nov-2011 12:42:43] <rmatte> I'm back from vacation

[21-Nov-2011 13:11:46] <jmp242> welcome back rmatte

[21-Nov-2011 13:13:36] <rmatte> tnx

[21-Nov-2011 14:04:58] <davetoo> Nice.  is Skype down again?

[21-Nov-2011 14:13:37] <rmatte> no idea, I don't use skype much

[21-Nov-2011 14:14:50] <Hackman238> rmatte: Have a good holiday rmatte?

[21-Nov-2011 14:15:12] <Hackman238> davetoo: my skype is up

[21-Nov-2011 14:16:03] <rmatte> Hackman238: yup, quite relaxing

[21-Nov-2011 14:16:17] <Hackman238> rmatte: Nice.

[21-Nov-2011 14:18:23] <rmatte> So I have yet another monitoring tool to review now...

[21-Nov-2011 14:18:32] <rmatte> statseeker

[21-Nov-2011 14:18:44] <rmatte> It looks alright, but it looks very network monitoring oriented

[21-Nov-2011 14:18:55] <rmatte> almost reminds me of OpenNMS with a simpler interface

[21-Nov-2011 14:23:25] <rmatte> well, time to review every commission request that was completed while I was gone to make sure they were done properly

[21-Nov-2011 14:23:26] <rmatte> eugh

[21-Nov-2011 14:25:53] <Hackman238> rmatte: NOO!!! Moving away from Zenoss?

[21-Nov-2011 14:31:54] <froztbyte> --- wonder if I should just work up a patch sometime...

[21-Nov-2011 14:32:13] <froztbyte> how is upstream's attitude towards accepting community patches?

[21-Nov-2011 14:32:19] <froztbyte> I'm not at all familiar with the dev process

[21-Nov-2011 14:32:36] <jmp242> Umm

[21-Nov-2011 14:32:46] <jmp242> I think you have to submit the contribution agreement

[21-Nov-2011 14:32:57] <jmp242> giving them copyright for core code changes

[21-Nov-2011 14:33:03] <jmp242> vs zenpacks for instance

[21-Nov-2011 14:34:01] <froztbyte> hmmm

[21-Nov-2011 14:34:22] <froztbyte> I might be okay enough with that

[21-Nov-2011 14:34:27] <froztbyte> guess I'll have to do some reading

[21-Nov-2011 14:34:51] <jmp242> Yea, and you may want to e-mail Nick directly

[21-Nov-2011 14:35:08] <froztbyte> but yeah, that functionality is a serious bugbear, and emulating it in transforms and stuff is really sub-optimal

[21-Nov-2011 14:36:10] <froztbyte> based on the first time I did, it didn't even seem like anyone else (who was around) knew how to go about for that

[21-Nov-2011 14:36:25] <jmp242> hmm may be

[21-Nov-2011 14:36:53] <jmp242> would you be OK with putting that info on the ZCA wiki (or having me do it?)

[21-Nov-2011 14:37:05] <jmp242> I like to collect useful zenoss snippets

[21-Nov-2011 14:37:18] <froztbyte> sure, just give me a citation link or whatever

[21-Nov-2011 14:37:29] <jmp242> though, I guess I need to test against 3.x

[21-Nov-2011 14:37:30] <froztbyte> I am on the zenoss forums or whatever too, if that is easier

[21-Nov-2011 14:37:34] <froztbyte> jmp242: yeah, please do

[21-Nov-2011 14:37:48] <froztbyte> daffy didn't get around to doing so for me, and I've still not upgraded to 3.x :/

[21-Nov-2011 14:37:59] <froztbyte> stuff keeps interrupting my time on that

[21-Nov-2011 14:39:19] <jmp242> looks good to me on 3.2.1

[21-Nov-2011 14:39:21] <jmp242> on SL5.7

[21-Nov-2011 14:40:41] <froztbyte> \o/

[21-Nov-2011 14:40:44] * froztbyte updates the post

[21-Nov-2011 14:40:54] <froztbyte> any preferred name?

[21-Nov-2011 14:41:29] <froztbyte> or is "jmp242 (from #zenoss)" fine?

[21-Nov-2011 14:41:53] <Hackman238> v4.2 Soon!

[21-Nov-2011 14:42:00] <froztbyte> stfu you

[21-Nov-2011 14:42:18] <froztbyte> but I'm painfully aware of this fact :/

[21-Nov-2011 14:42:45] <Hackman238> froztbyte: Oh man. Sorry LOL

[21-Nov-2011 14:43:54] <jmp242> that's fine

[21-Nov-2011 14:43:55] <jmp242> lol

[21-Nov-2011 14:47:38] <jmp242>

[21-Nov-2011 14:48:08] <jmp242> Thanks froztbyte

[21-Nov-2011 14:50:10] <rmatte> that's a good tip

[21-Nov-2011 14:51:36] <froztbyte>

[21-Nov-2011 14:54:03] <froztbyte> post updated

[21-Nov-2011 14:55:14] <dhopp> Can somebody give me a good reason why I would want to graph CpuIdle?  I mean, if I graph the other aspects (CpuSystem,CpuUser, etc.) what value is showing the remaining cpu?

[21-Nov-2011 14:56:06] <froztbyte> making sure all the numbers match up?

[21-Nov-2011 14:56:33] <froztbyte> I remember there were some cases that I'd picked up before where it was useful to do "negative space" monitoring, but can't recall what they were now

[21-Nov-2011 14:57:29] <dhopp> froztbyte:  I'm going from a fairly basic performance metrics to more detailed and I can see somebody not knowing what they are looking at going "OMG we are at 90% cpu"…uhm, no you are not..that's 90% of your cpu is idle

[21-Nov-2011 14:57:29] <dhopp> heh

[21-Nov-2011 14:58:24] <froztbyte> dhopp: I think the right answer is that you shouldn't just be using your CPU graphs

[21-Nov-2011 14:58:25] <jb> anybody monitoring windows clusters?

[21-Nov-2011 14:58:42] <froztbyte> performance metrics are very much tied to the performance of whatever it is that you're monitoring

[21-Nov-2011 14:59:19] <dhopp> froztbyte:  well true…but CPU can be a signal of other problems.

[21-Nov-2011 14:59:23] <froztbyte> so very high amounts of CPU Idle but very low amounts of free memory and lots of used swap might for instance indicate an application that's slowly leaking memory while not doing anything

[21-Nov-2011 15:00:03] <froztbyte> high CPU User, Memory and Swap might indicate processes getting out of hand

[21-Nov-2011 15:00:04] <froztbyte> etc

[21-Nov-2011 15:00:15] <dhopp> froztbyte:  true..but a network device that is operating at 99% CPU and all of a sudden is expected to handle 25% more traffic isn't going to work either

[21-Nov-2011 15:00:19] <froztbyte> tie everything together as best you can

[21-Nov-2011 15:00:38] <froztbyte> dhopp: if your networking device is at 99% CPU, you're possibly fucked 3 weeks ago already

[21-Nov-2011 15:00:47] <froztbyte> (or 10 minutes ago, in the case of sudden spikes)

[21-Nov-2011 15:01:02] <froztbyte> (also depends if you're using x86-based routers or stuff with ASICs, etc)

[21-Nov-2011 15:01:44] <dhopp> froztbyte:  we did a load test for a client that exploded when we entered load level 4….load level 3 had the load balancer running at 99% and I was like "well what did you expect the load balancer to do?"

[21-Nov-2011 15:01:45] <dhopp> heh

[21-Nov-2011 15:01:49] <dhopp> so we had to buy a bigger load balancer

[21-Nov-2011 15:01:57] <froztbyte> I tend to change my alert conditions so that "critical" means "we're going to run into a problem very soon", as opposed to "oh shit, we're fucked"

[22-Nov-2011 00:00:01] [disconnected at Tue Nov 22 00:00:01 2011]

[22-Nov-2011 00:00:01] [connected at Tue Nov 22 00:00:01 2011]

[22-Nov-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]

[22-Nov-2011 06:45:01] <Jane_Curry> Mornin

[22-Nov-2011 07:31:24] <mattt> Jane_Curry: hey

[22-Nov-2011 07:37:18] kocolosk|away is now known as kocolosk

[22-Nov-2011 07:43:41] <jmp242> morning

[22-Nov-2011 07:59:01] <mattt> jmp242: howdy!

[22-Nov-2011 08:07:54] <mattt> can someone point me to the best documentation for interfacing w/ zenoss via python?

[22-Nov-2011 08:08:32] <Jane_Curry> Depends a bit where you are starting from - what do you want to do?

[22-Nov-2011 08:09:59] <mattt> basic things, like being able to put a box in maintenance via an IRC bot, etc.

[22-Nov-2011 08:10:16] <mattt> or have something query current event and post them to IRC

[22-Nov-2011 08:10:18] <mattt> that sort of thing

[22-Nov-2011 08:10:24] <mattt> *current events

[22-Nov-2011 08:22:11] kocolosk is now known as kocolosk|away

[22-Nov-2011 08:25:20] <SEJeff> mattt, There is an xmpp bot that works pretty well

[22-Nov-2011 08:25:31] <SEJeff> You might take a look at that, but I think it is written in ruby or some crack like that

[22-Nov-2011 08:44:26] <whyzgeek> hi everyone, I am experiencing some strage thing. After a day or so, modeling changes don't get applied with no evident errors. So devices gets modeled but Last Collection stamp doesn't change. Then I have to restart zenhub and it will fix the issue for a while and again it happens.

[22-Nov-2011 08:44:48] <whyzgeek> any ideas?

[22-Nov-2011 09:34:49] kocolosk|away is now known as kocolosk

[22-Nov-2011 09:37:03] <mattt> SEJeff: seems to work nicely w/ supybot, got it sucking in alerts

[22-Nov-2011 09:38:45] <SEJeff> win

[22-Nov-2011 10:10:47] <tightwork> Hi. I have two events that show Critical and Error in Events list but not in the dashboard portlet. What can I do?

[22-Nov-2011 11:08:53] <nyeates> Hi

[22-Nov-2011 11:13:06] <Hackman238> nyeates: Hey, how goes it?

[22-Nov-2011 11:13:21] <Hackman238> tightwork: Which portlet?

[22-Nov-2011 11:13:49] <tightwork> Hackman238: the device issues

[22-Nov-2011 11:14:21] <tightwork> I have a service down in the main Events, but it does not show in the portlet.

[22-Nov-2011 11:16:23] <jb> ok, so i'm trying to stop monitoring a particular network interface, so I highlight it, click the gear, and uncheck Monitoring.

[22-Nov-2011 11:16:32] <jb> but, it always re-enables it's self.

[22-Nov-2011 11:16:36] <jb> do I need to reindex or something?

[22-Nov-2011 11:17:15] <Hackman238> tightwork: Are the devices in production?

[22-Nov-2011 11:21:12] <tightwork> Hackman238: oh sob, derp   no their in maintenance. Ah the simplest things. Thank you

[22-Nov-2011 11:28:13] <Hackman238> tightwork: No problem

[22-Nov-2011 11:39:13] <Hackman238> jb: Lock it

[22-Nov-2011 11:39:24] <Hackman238> jb: the modeler is reenabling it I think

[22-Nov-2011 11:41:14] <Hackman238> Anyone use the Applciation Profile zenpack?

[22-Nov-2011 11:45:57] <nyeates> heya hackman

[22-Nov-2011 11:46:45] <nyeates> that app profile code, i beleive, has been updated recently

[22-Nov-2011 11:46:50] <nyeates> on github

[22-Nov-2011 11:46:57] <nyeates> but maybe not as the eg on that link above

[22-Nov-2011 11:48:04] <nyeates>

[22-Nov-2011 11:51:53] <nyeates> was just version changes

[22-Nov-2011 11:53:25] <nyeates> oh wait...yes...there are changes here:

[22-Nov-2011 11:53:31] <nyeates> that are not in his 1.7 version

[22-Nov-2011 12:00:01] [disconnected at Tue Nov 22 12:00:01 2011]

[22-Nov-2011 12:00:02] [connected at Tue Nov 22 12:00:02 2011]

[22-Nov-2011 12:00:16] [zenoss-logger (logger bot) has joined #zenoss]

[22-Nov-2011 12:03:15] <dhopp> is there a known bug when looking at a list of devices if there is exactly 25 devices that it doesn't actually show #25?

[22-Nov-2011 12:03:32] <Hackman238> nyeates: Thanks Nick

[22-Nov-2011 12:03:51] <Hackman238> dhopp: Where do you see the problem?

[22-Nov-2011 12:21:30] <dhopp> in groups

[22-Nov-2011 12:21:58] <dhopp> I have 25 devices that is a member of a group

[22-Nov-2011 12:24:00] <dhopp> it says showing 1-25 of 25..but there are only 24 devices in the list

[22-Nov-2011 12:25:08] <dhopp> if I resort the list (i.e z-a rather then a-z) I see the missing device but don't see the original first device in the list

[22-Nov-2011 12:30:09] <dhopp> Hackman238: my wireless crapped out..not sure if you got my message

[22-Nov-2011 12:32:19] <Hackman238> dhopp: Huh. I tried it here but no problem

[22-Nov-2011 12:32:36] <dhopp> Hackman238:  what version of zenoss?

[22-Nov-2011 12:32:52] <Hackman238> dhopp: 3.2.0

[22-Nov-2011 12:33:05] <Hackman238> dhopp: The same one I'm fixing the CPU modeler on

[22-Nov-2011 12:33:19] <dhopp> lol

[22-Nov-2011 12:34:47] <dhopp> hmm..I'm still seeing it..I cleared browser cache and cookies…and tried a different browser

[22-Nov-2011 12:35:03] <dhopp> is there a setting to tell zope to pull a certain number of devices by default?

[22-Nov-2011 12:35:35] <Hackman238> dhopp: There are list limits, the default is 40 in zenoss

[22-Nov-2011 12:35:45] <dhopp> Hackman238:  yeah I found that...

[22-Nov-2011 12:35:52] <dhopp> going to play with that number and see what it does

[22-Nov-2011 12:35:56] <Hackman238> dhopp: IDK what else could cause it

[22-Nov-2011 12:36:14] <Hackman238> dhopp: Unless one of the objects has an invalid title but an exception isnt being thrown

[22-Nov-2011 12:38:43] <dhopp> Hackman238:  it seems to only happen when I'm sorted by Device...

[22-Nov-2011 12:39:17] <Hackman238> dhopp: Very strange. Does a zenchkrels cmoe out clean?

[22-Nov-2011 13:13:46] <jb> Hackman238: ahh locking.  thanks!

[22-Nov-2011 13:13:59] <Hackman238> jb: NP

[22-Nov-2011 14:04:18] <Hackman238> OT: "There isn't a headphone jack, which is only a problem if the folks around me don't  appreciate yodeling, (which almost NEVER happens). Also, I was accustomed to carrying the  iPod strapped to my arm with the elastic armband accessory. Nothing like this is available  for the pickle. On my beach walks, I've found that the pickle can be carried around by

[22-Nov-2011 14:04:23] <Hackman238>  convenientl tucking it down into the front of dmy Speedo."

[22-Nov-2011 14:04:42] <Hackman238> Commetn from yodling pickle: "There isn't a headphone jack, which is only a problem if the folks around me don't  appreciate yodeling, (which almost NEVER happens). Also, I was accustomed to carrying the  iPod strapped to my arm with the elastic armband accessory. Nothing like this is available  for the pickle. On my beach walks, I've found that the pickle can be carried around by

[22-Nov-2011 14:04:55] <jmp242> ??

[22-Nov-2011 14:05:01] <Hackman238> My bad all. Copy paste error.

[22-Nov-2011 14:05:07] <jmp242> ok

[22-Nov-2011 14:05:12] <jmp242> I was quite confused

[22-Nov-2011 14:05:14] <Hackman238>

[22-Nov-2011 14:05:24] <Hackman238> Thats what I meant to paste.

[22-Nov-2011 14:05:29] <jmp242> I see

[22-Nov-2011 14:05:35] <jmp242> quite useless as far as I can see

[22-Nov-2011 14:05:52] <Hackman238> jmp242: The comments are whats great

[22-Nov-2011 14:06:01] <jmp242> Ahh

[22-Nov-2011 14:06:04] <jmp242> I prefer

[22-Nov-2011 14:06:04] <jmp242>

[22-Nov-2011 14:06:20] <jmp242> I hope I make it into the January manufacturing run

[22-Nov-2011 14:07:02] <Hackman238> jmp242: LOL

[22-Nov-2011 14:09:03] <jmp242> you ever watch the ashens reviews on youtube?

[22-Nov-2011 14:09:15] <jmp242> he had a 47 minute review of it which made me want it lol

[22-Nov-2011 14:13:36] <Hackman238> jmp242: I havent. Sounds good though. LOL

[22-Nov-2011 14:14:13] <syoma> hi dears

[22-Nov-2011 14:16:44] <Hackman238> syoma: Hey, how goes the war?

[22-Nov-2011 14:39:59] kocolosk is now known as kocolosk|away

[22-Nov-2011 14:49:41] <syoma> juajuajua Hackman238 the war of chile ?

[22-Nov-2011 14:50:11] <syoma> all students go to war

[22-Nov-2011 14:50:12] <syoma> XD

[22-Nov-2011 14:53:07] <syoma> I've a question with the alerting rule...

[22-Nov-2011 14:55:00] <syoma> if possible generate alerts based on device dependencies, for example, if fail a group device (sensor ips) and behind this device thear more devices

[22-Nov-2011 14:55:14] <syoma> that are also monitoring by zenoss

[22-Nov-2011 14:56:06] <syoma> how is it possible that only alerts this sensor and not the rest

[22-Nov-2011 14:57:59] <syoma> the sensor is implemented in bridge mode, therefore, when the connectivity is lost, the other devices that are behind the sensor are alerted.

[22-Nov-2011 15:49:27] <Hackman238> syoma: You'll need to setup transforms to handle this.

[22-Nov-2011 15:53:25] <syoma> Hackman238:

[22-Nov-2011 15:53:33] <syoma> the event transforms ?

[22-Nov-2011 15:53:34] <syoma> or

[22-Nov-2011 15:53:42] <syoma> other event type

[22-Nov-2011 15:54:30] <syoma> Hackman238: tomorrow can you help me ?

[22-Nov-2011 15:54:38] <syoma> in this moment I go to ~

[22-Nov-2011 15:54:38] <syoma> XD

[22-Nov-2011 15:54:48] <syoma> thank you for your collaboration

[22-Nov-2011 15:54:51] <syoma> thank you very much

[22-Nov-2011 15:55:00] <syoma> good luck guys

[22-Nov-2011 16:25:10] kocolosk|away is now known as kocolosk

[22-Nov-2011 16:57:13] <Hackman238> Later all!

[22-Nov-2011 17:24:03] kocolosk is now known as kocolosk|away

[22-Nov-2011 20:13:07] kocolosk|away is now known as kocolosk

[23-Nov-2011 00:00:01] [disconnected at Wed Nov 23 00:00:01 2011]

[23-Nov-2011 00:00:01] [connected at Wed Nov 23 00:00:01 2011]

[23-Nov-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]

[23-Nov-2011 04:52:20] <[tla]> hi.  first time user - so be gentle with me! zenoss core v3.2.1 on ESXi 4.1. getting event "sendto error localhost and localhost.localdomain are both using ip". i can ping locallhost{.localdomain} and SNMPwalk the local snmpd from the command line.  changes i made: static IP, timezone, ntpd.conf. initiated discovery with 2 new community strings (removed public/private).  any ideas?...

[23-Nov-2011 04:52:22] <[tla]> ...thanks.

[23-Nov-2011 07:36:03] <tightwork> I had a linux vlan interface for eth0.27 but I turned off vlan and just made it a sub eth0:27 .. I remodel the server but it does not show eth0:27?

[23-Nov-2011 07:36:38] <tightwork> I have aliasmap modeler, ipmap

[23-Nov-2011 07:59:34] <fragfutter> tightwork: afaik subinterfaces are not exposed via snmp

[23-Nov-2011 09:24:19] kocolosk is now known as kocolosk|away

[23-Nov-2011 09:37:07] kocolosk|away is now known as kocolosk

[23-Nov-2011 11:17:02] <jb> here we go again.. trying to ignore all filesystems that contain "Volume"

[23-Nov-2011 11:17:04] <jb> any ideas?

[23-Nov-2011 11:22:44] <dpetzel> jb: You want to exclude filesystems that have the word "Volume" in them? does zFileSystemMapIgnoreNames not do the trick?

[23-Nov-2011 11:22:49] <fragfutter> zFileSystemMapIgnoreNames

[23-Nov-2011 11:23:19] <dhopp> is there a way to monitor a process (count, cpu, memory utilization) but NOT alert if it's not found?

[23-Nov-2011 11:24:00] <fragfutter> dhopp: set event severity to info?

[23-Nov-2011 11:24:31] <dhopp> fragfutter: hmmm..I guess I could try that and then do an event transform to just drop the event because I don't care about it

[23-Nov-2011 11:25:17] <dhopp> we have an application that calls an external process and just had a problem were there were over 200 of those external processes running and consumed all memory and swap on the server

[23-Nov-2011 11:25:30] <dhopp> I just want to monitor for a trend

[23-Nov-2011 11:26:18] <fragfutter> dhopp: you could write a command check to count the number of these processes and monitor that?

[23-Nov-2011 11:26:32] <dhopp> yeah I could do that but I want to graph memory too

[23-Nov-2011 11:26:46] <dhopp> figured it would be easier to just use what Zenoss already does

[23-Nov-2011 11:48:32] <jb> dpetzel: it does, im just having a hard time wit the regexp.

[23-Nov-2011 11:49:52] <dpetzel> jb: Assuming its currently an empty field you should just be able to put the word "Volume" without the quotes, thats not working?

[23-Nov-2011 11:50:26] <nyeates> I need a good doc on learning regex.... i never quite "got it"

[23-Nov-2011 11:50:33] <jb> me too.

[23-Nov-2011 11:50:38] <jb> frankly, I -suck- at it.

[23-Nov-2011 11:50:52] <nyeates> its tough to decypher it

[23-Nov-2011 11:50:57] <davetoo> it's worse than you think

[23-Nov-2011 11:51:00] <rocket> nyeates: the oreilly regex book is really good

[23-Nov-2011 11:51:11] <davetoo> zero-width positive lookbehind

[23-Nov-2011 11:51:21] <jb> i just have a hard time reading a whole book about something as boring as regexp

[23-Nov-2011 11:51:32] <nyeates> ill check it out rocket..... I was hoping that I wouldnt have to read an entire book to get it heh

[23-Nov-2011 11:51:47] <rocket> there are certain chapters etc that cover the basics

[23-Nov-2011 11:51:53] <nyeates> k

[23-Nov-2011 11:51:55] <rocket> then others just go into more depth if needed

[23-Nov-2011 11:52:28] <rocket> so really you dont read the whole book, but its a reference

[23-Nov-2011 11:52:30] <davetoo> what I like to do to learn these things (sometimes) is write a standalone unit-test

[23-Nov-2011 11:52:41] <davetoo> with progressively more complicated usages

[23-Nov-2011 11:52:51] <jb> dpetzel: that did it.. I was trying to over-complicate things, I guess.

[23-Nov-2011 11:53:06] <dpetzel> jb: Cool

[23-Nov-2011 11:53:08] <davetoo> thing to remember about regex's: they're slow

[23-Nov-2011 11:54:02] <rocket> well they dont have to be, thats one of the advanced topics of the book....

[23-Nov-2011 11:54:10] <rocket> precompiling the regex if its used alot etc

[23-Nov-2011 11:54:34] <davetoo> yes

[23-Nov-2011 11:54:46] <rocket> ordering to short circut the rest of the tests etc ..

[23-Nov-2011 11:54:51] <davetoo> and the book will tell you the relative costs...

[23-Nov-2011 11:54:53] <davetoo> yeah, that

[23-Nov-2011 11:55:04] <nyeates> rocket: which oreilly?

[23-Nov-2011 11:55:28] <jb> so I have a zenpack that has nothing but transforms in it

[23-Nov-2011 11:55:50] <jb> should I be able to take the objects.xml from it, and copy it to a new zenpack.. and these transforms would migrate over?

[23-Nov-2011 11:55:51] * davetoo <hearts> his Safari subscription

[23-Nov-2011 11:56:09] <rocket> I have the third one .. the upside down bat looking picture

[23-Nov-2011 11:56:21] <davetoo> jb: if the objects that the transforms belong to are there.

[23-Nov-2011 11:56:35] <davetoo> I.e. if they're all in stock locations, probably...

[23-Nov-2011 11:56:46] <jb> oh yeah

[23-Nov-2011 11:56:52] <jb> i probably need to just create the classes

[23-Nov-2011 11:56:53] <davetoo> get an xml editor and look at it.

[23-Nov-2011 11:57:19] <rocket> nyeates: actually its the owl book I just looked again and doublechecked

[23-Nov-2011 11:57:20] <davetoo> or something that at least folds xml to make it readable.

[23-Nov-2011 11:57:51] <nyeates> rocket: thx

[23-Nov-2011 11:58:50] <whyzgeek>           hi everyone, I am experiencing some strage thing. After a day or so, modeling changes don't get applied with no evident errors. So devices gets modeled but Last Collection stamp doesn't change. Then I have to restart zenhub and it will fix the issue for a while and again it happens.

[23-Nov-2011 11:58:57] <whyzgeek> any ideas?

[23-Nov-2011 12:00:02] [disconnected at Wed Nov 23 12:00:02 2011]

[23-Nov-2011 12:00:02] [connected at Wed Nov 23 12:00:02 2011]

[23-Nov-2011 12:00:15] [zenoss-logger (logger bot) has joined #zenoss]

[23-Nov-2011 12:08:31] <nyeates> whyzgeek....notice what davetoo put into the topic....

[23-Nov-2011 12:08:49] <nyeates> I think it would help to get some more details

[23-Nov-2011 12:09:25] <whyzgeek> opps I got it! 2.5.2 Enterprise

[23-Nov-2011 12:38:29] <nyeates> whyzgeek: if no one comes back with anything....i would try a test install of zenoss 3.2.1 and see if the same machines model right on that machine

[23-Nov-2011 12:38:46] <nyeates> there are literally multiple hundreds of fixes from 2.5.2 to 3.2.1

[23-Nov-2011 13:54:24] <jmp242> hey, anyone want to take a look at:

[23-Nov-2011 13:54:24] <jmp242>

[23-Nov-2011 14:23:32] <davetoo> I am not aware of a "component" template for CPUs for general pupose servers.

[23-Nov-2011 14:23:56] <davetoo> Some of the latest enterprise Cisco zenpacks break them down now, but ... that's not a very close match either.

[23-Nov-2011 14:24:17] <davetoo> oh, sorry, I'm misreading.

[23-Nov-2011 14:24:22] <davetoo> (and I should answer there.. nevermind)

[23-Nov-2011 14:49:57] <tightwork> Quite often all zen services will loose heartbeat, any thought?

[23-Nov-2011 14:50:08] <tightwork> It does clear but..

[23-Nov-2011 14:52:42] <jmp242> sounds like you may have a load issue on the zenoss serve

[23-Nov-2011 14:52:43] <jmp242> r

[23-Nov-2011 14:54:22] <rocket> increase the number of workers for zenhub

[23-Nov-2011 14:57:00] <dhopp> rocket: I have a weird problem…I have a few linux servers that have 4 or 5 filesystems.  monitoring works and then all of a sudden 3 of them will give me a component error: . is bad (the oid changes slightly for each filesystem)

[23-Nov-2011 14:57:54] <dhopp> rocket: then of the 2 that don't give that error disk utilization is all out of whack because it thinks it's space used is really high and the number that it shows is actually the space used by one of the other volumes

[23-Nov-2011 15:10:40] <tightwork> Ok great. I'll double check load and zenhub workers.  I know my IO isnt great.(quad xeon, 6GB ram, raid 10) I do have 55 nodes which can create exponential metrics  I'm looking at the perf graph for the collector says zenmodeler runs for 12 hours at around ~250?

[23-Nov-2011 15:24:51] <tightwork> web servers make the most interesting graphs

[23-Nov-2011 15:25:56] <rocket> dhopp: is your snmp daemon crashing?

[23-Nov-2011 15:26:09] <rocket> dhopp: it sounds like its not maintaining the oids etc

[23-Nov-2011 15:26:26] <rocket> or if you changed mounts on the filesystems etc perhaps you need to remodel

[23-Nov-2011 15:26:34] <rocket> I havent seen that behavior before though

[23-Nov-2011 15:34:04] <dhopp> rocket: a remodel does seem to fix it. snmpd hasn't crahsed

[23-Nov-2011 15:35:41] <dhopp> actually I take that back…snmpd restarted about an hour ago..hrm...

[23-Nov-2011 15:36:05] <dhopp> so will that happen after every time snmpd restarts?

[23-Nov-2011 15:36:15] <dhopp> that could be very problematic for our patching process

[23-Nov-2011 15:37:37] <jmp242> dhopp: yes, I believe it's been reported

[23-Nov-2011 15:37:44] <dhopp> oh god

[23-Nov-2011 15:37:46] <jmp242> that SNMP restarting changes / can change

[23-Nov-2011 15:37:52] <jmp242> the filesystem OID mapping

[23-Nov-2011 15:37:54] <jmp242> for some reason

[23-Nov-2011 15:38:03] <jmp242> you may want more often re-models

[23-Nov-2011 15:38:11] <dhopp> I'm going to have to find a different way to monitor filesystems...

[23-Nov-2011 15:38:19] <jmp242> it's how net-snmp works

[23-Nov-2011 15:38:30] <jmp242> which I would consider a bug of sorts or bad design

[23-Nov-2011 15:38:31] <jmp242> but hey

[23-Nov-2011 15:38:48] <jmp242> How often are you restarting SNMP

[23-Nov-2011 15:39:09] <dhopp> we patch servers monthly

[23-Nov-2011 15:39:15] <dhopp> so it would be at least every month

[23-Nov-2011 15:39:17] <jmp242> could a cron job basically fix it? or could the patching program or the service definition for SNMP on a start command send something off to Zenoss to then re-model

[23-Nov-2011 15:39:33] <dhopp> we reboot monthly as part of the patching process

[23-Nov-2011 15:39:58] <dhopp> I guess I could schedule remodel as part of the patching process

[23-Nov-2011 15:40:09] <jmp242> Oh, well what I'd do is have maintenance windows on the reboot (if it's a set time), and then have a cron task that restarts zenmodeler

[23-Nov-2011 15:40:18] <jmp242> so it can re-model everything before coming out of maintenance

[23-Nov-2011 15:42:29] <dhopp> our server configs (production anyway) don't change all that often…I'm sort of thinking maybe we don't remodel very often and just do it as part of our patching process….but then I lose some of the cool functionality of zenoss

[23-Nov-2011 15:43:44] <tightwork> Is there a TALES expression to get the services of a device? I'd like to create a custom report for all devices that have X ipProcess, or more useful: all devices that have X ipProcess and are not monitored.

[23-Nov-2011 15:44:57] <jmp242> dhopp: want to explain - what cool functionality do you lose?

[23-Nov-2011 15:45:17] <jmp242> the remodel from cron can be totally separate from the zenoss scheduled ones - it's no problem

[23-Nov-2011 15:52:27] <dhopp> jmp242:  the ability to pickup changes automatically…(processes running, etc.)

[23-Nov-2011 15:52:57] <rocket> dhopp: just schedule a separate zenmodeller instance per device or what ever from the command line

[23-Nov-2011 15:52:58] <dhopp> I understand I can cron it…but that doesn't really solve the problem…what if I have a server crash on a tuesday and my model doesn't run until thursday

[23-Nov-2011 15:53:10] <rocket> zenmodeller run -d <device_in_maintenance>

[23-Nov-2011 15:53:27] <dhopp> rocket:  again…that doesn't solve the real problem

[23-Nov-2011 15:53:30] <rocket> the zenping events can trigger a remodel

[23-Nov-2011 15:53:41] <dhopp> rocket:  oh I didn't know that

[23-Nov-2011 15:53:45] <rocket> as an event command .. on ping clear issue a remodel command

[23-Nov-2011 15:53:57] <rocket> as an example

[23-Nov-2011 15:54:23] <rocket> or switch to ssh monitoring of linux servers

[23-Nov-2011 15:54:31] <dhopp> rocket: I might look into that..we don't have zenoss doing snmp traps..but I guess if I had it doing that I could do a remodel on a snmp start event

[23-Nov-2011 15:54:44] <rocket> yes that would work to

[23-Nov-2011 15:55:02] <dhopp> rocket: I might switch to ssh monitoring just for disk space

[23-Nov-2011 16:00:13] <rocket> or upgrade netsnmp  maybe new versions dont have that issue

[23-Nov-2011 16:00:42] <dhopp> that's not an option unfortunately…we have to remain with the vendor supplied patches (Red Hat)

[23-Nov-2011 16:01:10] <rocket> sure I understand

[23-Nov-2011 16:01:12] <dhopp> if we go off that then we have to maintain the patches ourselves for PCI compliance and that is a headache we don't want to deal with

[23-Nov-2011 16:01:44] <rocket> bug redhat about it?

[23-Nov-2011 17:01:49] <nyeates> anyone around?

[23-Nov-2011 17:02:17] <nyeates> need a second pair of eyes

[23-Nov-2011 17:03:52] <dhopp> nyeates:  I'm sort of around...

[23-Nov-2011 17:03:52] <dhopp> heh

[23-Nov-2011 17:03:55] <dhopp> what do you need?

[23-Nov-2011 17:04:11] <nyeates> eye ball a blog post that is going to devs....lemme pm u

[23-Nov-2011 17:06:37] kocolosk is now known as kocolosk|away

[23-Nov-2011 17:31:56] <nyeates>

[23-Nov-2011 17:32:08] <nyeates> if you care to view dhopp

[23-Nov-2011 17:32:19] <nyeates> happy turkey weekend everyone! bye

[23-Nov-2011 18:23:28] tvincent_ is now known as tvincent

[24-Nov-2011 00:00:01] [disconnected at Thu Nov 24 00:00:01 2011]

[24-Nov-2011 00:00:01] [connected at Thu Nov 24 00:00:01 2011]

[24-Nov-2011 00:00:13] [zenoss-logger (logger bot) has joined #zenoss]

[24-Nov-2011 10:29:26] <Appiah> I'm having a issue adding a Vmware infrastructure, I just get connection timed out in the zenvmwareeventlog. However the collector can reach the vcenter on 80/443/8080/8081

[24-Nov-2011 10:31:04] <Appiah> any tips on troubleshooting this error?

[24-Nov-2011 10:48:57] <rmatte> Appiah: you're sure the credentials are correct?

[24-Nov-2011 10:49:09] <rmatte> also, is that with the enterprise pack?

[24-Nov-2011 11:21:25] <rmatte> does anyone know of a ZenPack that just adds a new section to the OS section in Zenoss 2.5?

[24-Nov-2011 11:26:14] <rmatte> I'm trying to create a method to monitor windows servers via SNMP, but I've never actually coded a new section for the OS section, so I need some reference code

[24-Nov-2011 11:27:51] <rmatte> I may just end up creating a new tab

[24-Nov-2011 11:28:09] <Appiah> it's enterprise

[24-Nov-2011 11:28:20] <Appiah> if the credentials are wrong should it not say something different?

[24-Nov-2011 11:28:41] <rmatte> I'm not sure about that since I don't use enterprise, but credentials seem like the most likely cause if you've verified connectivity

[24-Nov-2011 11:28:49] <Appiah> hmm

[24-Nov-2011 11:29:12] <Appiah> maybe it cant use ldap/ad accounts

[24-Nov-2011 11:29:17] <Appiah> i'll try a local account...

[24-Nov-2011 11:29:52] <rmatte> Permissions -> Right Click -> Add Permission -> Click Add... -> Select user and click OK -> Make sure that Read-only is selected on the right dropdown -> Click OK.

[24-Nov-2011 11:30:04] <rmatte> That's what you need to do in ESX

[24-Nov-2011 11:30:28] <rmatte> Also, you probably want to be pointing it directly at the ESX servers, not at the vSphere server (if you have one)

[24-Nov-2011 11:32:16] <Appiah> says here it spports vcenters

[24-Nov-2011 11:32:22] <Appiah> supports*

[24-Nov-2011 11:33:45] <rmatte> I'm sure it does, I just don't personally see the point in monitoring the environment at that level

[24-Nov-2011 11:34:49] <Appiah> I thought if you add the vcenter it will add the whole infrastruce including the esx's running in that environment

[24-Nov-2011 11:35:51] <rmatte> it's possible that it will, as I said, I've never used enterprise, I've only used the community packs

[24-Nov-2011 11:37:31] <Appiah> I'll update you when I've tried

[24-Nov-2011 11:37:37] <rmatte> k

[24-Nov-2011 12:00:01] [disconnected at Thu Nov 24 12:00:01 2011]

[24-Nov-2011 12:00:02] [connected at Thu Nov 24 12:00:02 2011]

[24-Nov-2011 12:00:13] [zenoss-logger (logger bot) has joined #zenoss]

[24-Nov-2011 12:33:52] <rmatte> hmmm, I seem to have found a bug in Egor's HP Proliant Monitor pack

[24-Nov-2011 12:34:11] <rmatte> the temperature template doesn't get bound to the sensor objects after modelling

[24-Nov-2011 12:48:16] <rmatte> ah, the collector plugins weren't in the right order

[24-Nov-2011 12:48:19] <rmatte> it's working now

[24-Nov-2011 13:02:39] <rmatte> hmmm, a purple status dot, that's new

[24-Nov-2011 15:34:54] <rmatte> Hackman238: you around?

[24-Nov-2011 15:51:04] <dhopp> rmatte: it's thanksgiving!

[24-Nov-2011 15:51:16] <dhopp> I know you Canadian's don't celebrate it :-P

[24-Nov-2011 15:51:32] <rmatte> we do celebrate it actually, just not on the same date as you guys do

[24-Nov-2011 15:51:37] <dhopp> well right

[24-Nov-2011 15:51:46] <dhopp> I'm only online because I'm the poor sap that is on call this week

[24-Nov-2011 15:51:56] <rmatte> haha

[24-Nov-2011 15:52:15] <rmatte> I'm trying to figure out how to move a tab to a specific position in the Zenoss UI

[24-Nov-2011 15:52:25] <rmatte> it's at the end of the tabs at the moment, I want to move it beside the OS tab

[24-Nov-2011 15:52:28] <rmatte> (this is in 2.5.2)

[24-Nov-2011 15:53:12] <rmatte> guess I'll just keep working on it and figure that out later

[24-Nov-2011 15:53:37] <rmatte> I'm making a ZenPack for monitoring windows servers via SNMP

[24-Nov-2011 15:53:46] <rmatte> It'll work in both 2.5 and 3.x

[24-Nov-2011 16:07:35] <dhopp> I found a very annoying bug

[24-Nov-2011 16:07:40] <dhopp> I don't know if it's 3.x specific

[24-Nov-2011 16:07:57] <dhopp> but if you create a graph report that has more then 10 graphs you can't edit the sequence of any of the graphs past #10

[24-Nov-2011 16:08:01] <dhopp> they never show up in the list

[24-Nov-2011 16:08:08] <dhopp> but the report shows them

[24-Nov-2011 16:08:36] <rmatte> dhopp: the old interface had the same bug but it started at 40 graphs

[24-Nov-2011 16:09:05] <rmatte> and yes, extremely annoying

[24-Nov-2011 16:09:15] <dhopp> 40 graphs is a lot so that would be fairly rare…10 graphs is easy to get to...

[24-Nov-2011 16:09:25] <dhopp> I mean you have 5 servers that you want to see cpu and memory for a bam you are at 10

[24-Nov-2011 16:09:31] <rmatte> actually it may have been 20 or 30, I don't remember exactly

[24-Nov-2011 16:09:36] <rmatte> I know I've personally run in to the issue

[24-Nov-2011 16:09:47] <rmatte> but yeh, that is a lame bug

[24-Nov-2011 16:10:37] <dhopp> I haven't looked at the code, but it seems like it could be something easy like the display page only loops to 9 (since the count is 0-9)

[24-Nov-2011 16:11:37] <dhopp> and I'm going to have to write a new check_http plugin…I need to be able to save the output of the page if it failed into a DB for later analysis

[24-Nov-2011 16:13:19] <rmatte> you could just write a wrapper script to do that using curl

[24-Nov-2011 16:13:54] <rmatte> have it execute check_http, return the output, if it returns with an error code launch curl and save the page wherever

[24-Nov-2011 16:15:22] <dhopp> true

[24-Nov-2011 16:15:30] <dhopp> not a bad idea

[24-Nov-2011 16:16:24] <dhopp> I could then rename check_http to something and then name the wrapper script check_http…of course that could be problematic for upgrades

[24-Nov-2011 16:17:50] <dhopp> I guess I could just create a new zenpack and modify it's call to check_http to be to my wrapper….although that could get messy too

[24-Nov-2011 16:17:51] <dhopp> hrmph

[24-Nov-2011 16:53:36] <dhopp> man I need to get up to speed on tales

[24-Nov-2011 17:03:04] <rmatte> hehe

[24-Nov-2011 17:08:47] <dhopp> rmatte:  I'm trying to see if I can find the bug in the graph report code but trying to figure out all of the nomenclature isn't as easy as I thought

[24-Nov-2011 17:09:09] <rmatte> yeh, it can be a bit convoluted

[24-Nov-2011 17:10:06] <dhopp> I think I'm going to have to figure out where to put debug statements in so I can see how it is looping through the code

[24-Nov-2011 17:17:16] <dhopp> rmatte I found part of the problem

[24-Nov-2011 17:17:38] <dhopp> it only shows the number of graphs up to the value of the users "default page size" value

[24-Nov-2011 17:17:41] <dhopp> I had mine set to 10

[24-Nov-2011 17:18:30] <dhopp> That defaults to 40 which would explain what you were seeing

[25-Nov-2011 00:00:01] [disconnected at Fri Nov 25 00:00:01 2011]

[25-Nov-2011 00:00:02] [connected at Fri Nov 25 00:00:02 2011]

[25-Nov-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]

[25-Nov-2011 07:21:37] <sena1> hello

[25-Nov-2011 07:24:49] <sena1> Don't know if this is possible, but here it goes: I'm finishing a ZenOSS installation, everything is going fine, but a couple of devices (Wireless APs and switches) are generating lots of DEBUG events. Is there a way to ignore (for example, automatically dropping those events), so that they are never seen?

[25-Nov-2011 07:25:09] <sena1> Or automatically moving them to the event history?

[25-Nov-2011 08:26:50] <zykes-> has anyone here done reboot detection with zenoss ?

[25-Nov-2011 08:32:29] <mattt> zykes-: no, but that's something i'd like

[25-Nov-2011 08:32:41] <mattt> zykes-: ie. send alert if uptime resets

[25-Nov-2011 08:33:20] <zykes-> hmm, how can one check that ?

[25-Nov-2011 08:33:47] <zykes-> sooo looking forward to getting v4 with amqp if that's a fact

[25-Nov-2011 09:50:42] <rmatte> zykes-: reboot detection for servers?

[25-Nov-2011 09:51:22] <rmatte> Servers send a coldStart trap when the snmp agent restarts

[25-Nov-2011 09:51:36] <rmatte> you can just do what I did and transform that trap so that it says "Device Reboot"

[25-Nov-2011 09:51:56] <rmatte> as for network devices, you'd have to have something checking the uptime each cycle and generating an event if it changes

[25-Nov-2011 09:52:18] <rmatte> you could do that with a simple snmp template

[25-Nov-2011 09:58:07] <rmatte> YES! I just got a successful model of windows servers via SNMP, that's the hard part over with, now to design the back end templates and code the rest of what I need.

[25-Nov-2011 09:58:21] <rmatte> windows services* rather

[25-Nov-2011 10:04:57] <fragfutter> rmatte: what is this windows you are talking about?

[25-Nov-2011 10:17:46] <rmatte>

[25-Nov-2011 10:17:59] <rmatte> It's the bane of my existance when it comes to monitoring (WMI)

[25-Nov-2011 10:18:23] <rmatte> This snmp windows service monitoring pack is going to be lovely

[25-Nov-2011 10:18:50] <dhopp> going to release it to the community?

[25-Nov-2011 10:19:07] <mattt> can anyone recommend a good snmp book?

[25-Nov-2011 10:19:17] <mattt> i'm more used to nagios and doing everything w/ nrpe

[25-Nov-2011 10:19:20] <rmatte> dhopp: I release basically everything I make to the community, so yes

[25-Nov-2011 10:19:45] <rmatte> mattt:

[25-Nov-2011 10:20:00] <rmatte> though SNMP is pretty simple to pick up on

[25-Nov-2011 10:20:10] <rmatte> that book goes really in depth

[25-Nov-2011 10:20:52] <dhopp> rmatte:  you do man

[25-Nov-2011 10:22:03] <rmatte>

[25-Nov-2011 10:22:32] <dhopp> well that came out wrong

[25-Nov-2011 10:22:43] <rmatte> I got the just of it

[25-Nov-2011 10:22:48] <dhopp> I didn't mean to imply you DO men...

[25-Nov-2011 10:22:49] <rmatte> anywho, back to work...

[25-Nov-2011 10:22:52] <rmatte> lol

[25-Nov-2011 10:22:53] <dhopp> you da man

[25-Nov-2011 10:22:54] <dhopp> hehe

[25-Nov-2011 10:23:11] <dhopp> I'm a little called at 3 AM

[25-Nov-2011 10:23:19] <dhopp> sleepy*

[25-Nov-2011 10:23:20] <rmatte> I'm actually recycling some code from a Jane Curry ZenPack for this, making development faster

[25-Nov-2011 10:23:20] <dhopp> ugh

[25-Nov-2011 10:23:27] <mattt> rmatte: yeah, i pick that up ... thanks!

[25-Nov-2011 10:23:36] <rmatte> mattt: np

[25-Nov-2011 10:28:07] <tightwork> I am scanning a 7506 cisco router. I modeled it with the cisco mibs for interface and route map.  The vlans show up as Vl2 VL2.. graphs are also being shown but for some I get bad oid for Vlx makes no sense because zenoss is getting data, its a valid oid

[25-Nov-2011 10:33:08] <rmatte> tightwork: virtual interfaces don't support error OIDs

[25-Nov-2011 10:33:41] <rmatte> you should create a custom interface template for it (copy the ethernetCsmacd and change the name of the template to the type of the interface, which you'll see in the details)

[25-Nov-2011 10:33:52] <rmatte> then remove the error OIDs, threshold, and graph

[25-Nov-2011 10:34:18] <rmatte> or copy ethernetCsmacd_64 if the type shows up with _64 at the end of it

[25-Nov-2011 10:34:26] <tightwork> oh is that what it is... I didnt resolve the oid but indeed there are no errors on that graph

[25-Nov-2011 10:35:05] <tightwork> ok thank you very much

[25-Nov-2011 10:35:44] <rmatte> np

[25-Nov-2011 10:36:15] <rmatte> by the way, when I say the type of interface, I don't mean "Vl" or "vlan"

[25-Nov-2011 10:36:29] <rmatte> if you select the interface, you'll see the graphs with a dropdown menu above them

[25-Nov-2011 10:36:43] <rmatte> in that menu there should be something like "Details" or "Properties"

[25-Nov-2011 10:36:53] <rmatte> select it and there will be "Type" as one of the values

[25-Nov-2011 10:37:32] <rmatte> I don't have a Zenoss 3.x instance loaded up at the moment so I'm just doing this from memory

[25-Nov-2011 10:38:08] <rmatte> eugh, this is driving me crazy, this ZenPack shows a "Status Value" of "Not active (1)"

[25-Nov-2011 10:38:19] <rmatte> so I grep through the whole pack to see if there's a spot to change those

[25-Nov-2011 10:38:23] <rmatte> but nope, nothing

[25-Nov-2011 10:38:31] <rmatte> then I grep through the whole Zenoss directory, still nothing

[25-Nov-2011 10:38:39] <rmatte> wondering where the heck it's getting that from

[25-Nov-2011 10:38:46] <rmatte> maybe it's pulling it from a Mib or something, hmmm

[25-Nov-2011 10:40:53] <rmatte> Jane_Curry: you around?

[25-Nov-2011 11:24:08] <rmatte> eugh, I hate crap like this where what I'm seeing makes zero sense

[25-Nov-2011 11:35:58] <rmatte> ah, think I might have figured this out.

[25-Nov-2011 11:37:44] <rmatte> yup, got it, awesome

[25-Nov-2011 11:54:11] <rmatte> hmmm, now I need to figure out how to pull the oids out of the data table since these snmp indexes are handled really strangely.

[25-Nov-2011 12:00:01] [disconnected at Fri Nov 25 12:00:01 2011]

[25-Nov-2011 12:00:01] [connected at Fri Nov 25 12:00:01 2011]

[25-Nov-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]

[25-Nov-2011 12:15:13] <rmatte> does anyone in here know how I'd go about providing the service name and retrieving the OID in this dictionary?

[25-Nov-2011 12:15:22] <rmatte> I've tried all sorts of stuff, can't seem to figure it out

[25-Nov-2011 12:15:39] <rmatte> might have to use a for loop or something

[25-Nov-2011 12:15:45] <rmatte> hmmm, I'll give that a go

[25-Nov-2011 12:21:59] <rmatte> ah, I think I got it

[25-Nov-2011 12:27:35] <rmatte> bingo, got it, woot

[25-Nov-2011 12:41:57] <rmatte> hmmm

[25-Nov-2011 12:42:10] <rmatte> Zenoss expect the snmpindex value to be an integer, that's dumb

[25-Nov-2011 12:42:45] <rmatte> ls

[25-Nov-2011 12:42:47] <rmatte> oops

[25-Nov-2011 12:52:11] <rmatte> is anyone around who has experience writing collector plugins?

[25-Nov-2011 13:08:39] <rmatte> oh no, apparently it can be a string, good

[25-Nov-2011 13:31:37] <froztbyte> rmatte: why is it dumb?

[25-Nov-2011 13:41:33] <rmatte> because if it didn't allow strings you couldn't make the index something like: 10.23.576

[25-Nov-2011 13:41:45] <rmatte> it'd be restricted to one number

[25-Nov-2011 13:41:57] <rmatte> but anyways, I've sorted it out

[25-Nov-2011 13:42:34] <rmatte> with any luck I'll be done this pack by end of day, though probably not since I still have to modify/test it to make sure it works with Zenoss 3.x too

[25-Nov-2011 13:42:50] <davetoo> what are you working on?

[25-Nov-2011 13:46:04] <rmatte> A ZenPack to monitor windows services via SNMP

[25-Nov-2011 13:46:21] <davetoo> With which snmp agent?

[25-Nov-2011 13:46:28] <rmatte> the default windows one

[25-Nov-2011 13:46:34] <davetoo> *nod*

[25-Nov-2011 13:46:54] <davetoo> Haven't had the cause yet to dig very far into that one

[25-Nov-2011 13:47:15] <rmatte> We have so many servers reporting WMI errors, we constantly fix them, and they come back a few days later

[25-Nov-2011 13:47:19] <rmatte> it's becoming unmanageable

[25-Nov-2011 13:47:32] <rmatte> and we're really only using WMI for service monitoring on 95% of the devices

[25-Nov-2011 13:47:42] <rmatte> so when I finish this we can use WMI in a more targetted fashion

[25-Nov-2011 13:47:43] <davetoo> yeah, and MS and Cisco seemed to try really hard to frustrate deduplication algorithms

[25-Nov-2011 13:48:30] <davetoo> "11:01:01: this has happened 325 times"

[25-Nov-2011 13:48:40] <davetoo> "11:02:01: this has happened 326 times"

[25-Nov-2011 13:51:04] <rmatte> hmmm, this is bizarre, I installed the advanced device details pack so that we could use the HP Proliant monitoring pack

[25-Nov-2011 13:53:53] <rmatte> but now when I model a device it's only picking up on some of the interfaces

[25-Nov-2011 13:53:56] <Bart_> i was considering switching back to good old snmp because i'm tired of fixing wmi issues

[25-Nov-2011 13:54:57] <rmatte> damnit, I wish egor was around

[25-Nov-2011 13:55:11] <rmatte> Bart_: yeh, that's what we're in the process of doing

[25-Nov-2011 13:55:18] <davetoo> I need to completely rip apart and redesign my VM lab environment today

[25-Nov-2011 13:55:45] <davetoo> sort of "not working, but working" today :/

[25-Nov-2011 13:56:12] <Bart_> rmatte: i like seeing the windows event logs though, we see a lot we wouldn't even have noticed otherwise

[25-Nov-2011 13:56:30] <Bart_> but the windows machines are less than 5% of total but i spend most of my time on them

[25-Nov-2011 13:56:36] <davetoo> hah

[25-Nov-2011 13:56:38] <davetoo> wonder why

[25-Nov-2011 13:56:58] <rmatte> Bart_: I don't even think Zenoss uses WMI for the event logs gathering, it uses some other protocol

[25-Nov-2011 13:57:21] <davetoo> at my last job, the Windows dev/qa team's labs took up so much more lab space, power, and cooling than the equivalent linux/core teams, per unit of work

[25-Nov-2011 13:57:37] <rmatte> hmmmm, so in ZMI I see the interfaces listed, something is screwed with this Advanced Device Details pack when it comes to the 2.5 interface

[25-Nov-2011 13:57:40] <rmatte> time to figure out what

[25-Nov-2011 13:58:58] <Bart_> davetoo: heh, monitoring takes up most of our lab resources, but thats just because puppet makes it way to easy for me to start new test machines and forget about the old ones

[25-Nov-2011 13:59:05] <davetoo> I got brave last night and installed the latest firefox nightly build (win7x64)

[25-Nov-2011 13:59:11] <davetoo> hahah

[25-Nov-2011 13:59:39] <davetoo> I'm trying to decide whether to go harass Mark Burgess next week at the local LISA chapter meeting

[25-Nov-2011 14:00:17] <davetoo> ask him WTF he's talking about, because fundamentally I like the staggered "triggered pull" method that cfengine uses

[25-Nov-2011 14:00:51] <davetoo> but he's making the terminology too clever/complicated

[25-Nov-2011 14:01:08] <Bart_> I never really understood the cfengine syntax. thought puppet was easier

[25-Nov-2011 14:01:23] <Bart_> not that i'm an expert either, but it works

[25-Nov-2011 14:01:51] <davetoo> ask him why is "promise theory" different than, say, "design by contract" or even "postcondition"

[25-Nov-2011 14:02:19] <davetoo> cfengine is.. more like a bag of different billets of metal

[25-Nov-2011 14:02:32] <davetoo> where puppet is more like an erector set

[25-Nov-2011 14:02:52] <davetoo> and chef is more of a bag of modules with connectors

[25-Nov-2011 14:02:59] <Bart_> my problem is still that people mess with configs out of puppets control, so i never see those untill i re-image the system

[25-Nov-2011 14:03:11] <davetoo> in my very cursory look at the latter two

[25-Nov-2011 14:03:24] <davetoo> well,

[25-Nov-2011 14:04:06] <davetoo> My few cfengine deployments, cfengine itself would "correct" those over and over until the perpetrators came to me and complained,

[25-Nov-2011 14:04:09] <davetoo> then I could educate them

[25-Nov-2011 14:05:26] <rmatte> lol

[25-Nov-2011 14:05:28] <Bart_> yeah, every file under puppet control gets overwritten every 2hrs. But if someone installs an additional package, that package may include more scripts that affect the system behavior

[25-Nov-2011 14:05:36] <rmatte> I have a good question if anyone knows the answer...

[25-Nov-2011 14:05:38] <davetoo> sure

[25-Nov-2011 14:05:48] <davetoo> (sure, Bart)

[25-Nov-2011 14:05:50] <rmatte> so for certain ZenPacks to work, you have to define a zPythonClass property on the class

[25-Nov-2011 14:05:59] <davetoo> cats are hard to herd

[25-Nov-2011 14:06:00] <rmatte> what if you have 2 ZenPacks that require this that you both want to use

[25-Nov-2011 14:06:13] <rmatte> out of luck?

[25-Nov-2011 14:08:43] <Bart_> davetoo: main problem at the moment is that the order in which packages are installed may affect the system state. (install ssmtp, postfix, then remove ssmtp and the sendmail symlinks are screwed). Very difficult to predict as it only happens every now and then

[25-Nov-2011 14:08:45] <davetoo> argh, Jive link spam out of control again

[25-Nov-2011 14:09:44] <davetoo> *nod*

[25-Nov-2011 14:10:16] <davetoo> Anybody here done much with SMPP?

[25-Nov-2011 14:12:04] <Bart_> rmatte: i had the same question, but seems to me you need to pick one, or create your own class that invokes methods from the other two (if at all possible - i haven't tried that bit yet)

[25-Nov-2011 14:15:51] <rmatte> It just seems very limiting

[25-Nov-2011 14:18:43] <davetoo> Safari Books Online FTW! (again)

[25-Nov-2011 14:54:11] <rmatte> hmmm, renaming all the variables and files in this ZenPack once I have it working is going to be a pain

[25-Nov-2011 14:54:27] <rmatte> I did a sed to replace "Port" with "Service" and even though it should have worked fine, it didn't for some reason

[25-Nov-2011 14:54:40] <rmatte> oh well

[25-Nov-2011 15:26:41] <rmatte> well, I've got it collecting data now... still have some slight kinks to work out

[25-Nov-2011 15:46:05] <rmatte> hmmm, it only displays services that are actually running, when they are stopped they dissapear from SNMP

[25-Nov-2011 15:46:10] <rmatte> that's going to be a bit of a challenge

[25-Nov-2011 16:15:06] <rmatte> looks like I'll have to use a script to do this properly.

[25-Nov-2011 16:36:19] <rmatte> eugh, this is really starting to drive me nuts... I add a command datasource to the same template that was working fine with a snmp type datasource and it returns no data

[25-Nov-2011 16:37:53] <rmatte> as in it doesn't even attempt to execute the script

[26-Nov-2011 00:00:01] [disconnected at Sat Nov 26 00:00:01 2011]

[26-Nov-2011 00:00:01] [connected at Sat Nov 26 00:00:01 2011]

[26-Nov-2011 00:00:01] [disconnected at Sat Nov 26 00:00:01 2011]

[26-Nov-2011 00:00:02] [connected at Sat Nov 26 00:00:02 2011]

[26-Nov-2011 00:00:15] [zenoss-logger (logger bot) has joined #zenoss]

[26-Nov-2011 12:00:01] [disconnected at Sat Nov 26 12:00:01 2011]

[26-Nov-2011 12:00:01] [connected at Sat Nov 26 12:00:01 2011]

[26-Nov-2011 12:00:11] [zenoss-logger (logger bot) has joined #zenoss]

[26-Nov-2011 15:42:09] <hetz> Hi people. Just installed ZenOSS latest and configured some stuff, but I don't see the alert menu in advanced. Have I missed anything?

[26-Nov-2011 15:45:12] <hetz> oops, found out why, sorry

[26-Nov-2011 17:31:10] <hetz> Is there a way to simply add ip service to a machine I already added? (it seems when I added it found everything, but not http)..

[26-Nov-2011 18:16:07] <mattt> hetz: believe so

[26-Nov-2011 18:16:46] <mattt> if you're looking at the device under infrastructure

[26-Nov-2011 18:16:53] <mattt> just click the + in the bottom left corner

[26-Nov-2011 18:17:08] <mattt> 'Add IpService'

[26-Nov-2011 18:28:01] <hetz> wow, why is it there? this button should be in the upper right part.

[26-Nov-2011 18:29:28] <hetz> but thanks mattt

[26-Nov-2011 18:39:48] <mattt> hetz: no idea, still learning zenoss myself :/

[26-Nov-2011 22:34:41] <davetoo> ergh

[26-Nov-2011 22:35:23] <davetoo> I sure hope that Eclipse Indigo and Helios .project files are compatible.  Simultaneous install migrating to the former.

[27-Nov-2011 00:00:01] [disconnected at Sun Nov 27 00:00:01 2011]

[27-Nov-2011 00:00:02] [connected at Sun Nov 27 00:00:02 2011]

[27-Nov-2011 00:00:12] [zenoss-logger (logger bot) has joined #zenoss]

[27-Nov-2011 12:00:01] [disconnected at Sun Nov 27 12:00:01 2011]

[27-Nov-2011 12:00:02] [connected at Sun Nov 27 12:00:02 2011]

[27-Nov-2011 12:00:14] [zenoss-logger (logger bot) has joined #zenoss]

[27-Nov-2011 13:24:45] kocolosk|away is now known as kocolosk

[27-Nov-2011 17:49:18] <hetz> I get a keyerror here which seems related to extdirect. How can It be dealt?

[28-Nov-2011 00:00:01] [disconnected at Mon Nov 28 00:00:01 2011]

[28-Nov-2011 00:00:03] [connected at Mon Nov 28 00:00:03 2011]

[28-Nov-2011 00:00:14] [zenoss-logger (logger bot) has joined #zenoss]

[28-Nov-2011 06:53:41] <froztbyte> hmm

[28-Nov-2011 06:54:21] <froztbyte> how well does the multi-collector stuff work over non-low-latency (sometimes lossy) links?

[28-Nov-2011 06:54:28] <froztbyte> if anyone's tried, I guess

[28-Nov-2011 08:49:22] kocolosk is now known as kocolosk|away

[28-Nov-2011 09:04:34] kocolosk|away is now known as kocolosk

[28-Nov-2011 09:24:58] <Gat0rvean> Morning all

[28-Nov-2011 09:25:26] <Gat0rvean> Is anyone very familiar with the "ZenSystems.Juniper" zenpack by Jane Curry?

[28-Nov-2011 09:28:04] <fragfutter> Gat0rvean: i would guess jane if you stick around she might drop by (jcurry)

[28-Nov-2011 09:32:01] <Gat0rvean> fragfutter: thanks

[28-Nov-2011 10:49:31] <rmatte> I assume everyone is back from thanksgiving today?

[28-Nov-2011 11:03:09] <jb___> so, whats an easy way to clear drop all history/current events?

[28-Nov-2011 11:03:22] jb___ is now known as jb

[28-Nov-2011 11:05:21] <rmatte> jb__: you can just connect to the mysql server (zenoss database) and do: truncate table history; truncate table status;

[28-Nov-2011 11:05:40] <rmatte> that'll be quicker than any other method

[28-Nov-2011 11:05:58] <rmatte> should only take a few seconds since it's actually dropping the tables then re-creating them

[28-Nov-2011 11:07:22] <jb> thanks

[28-Nov-2011 11:08:04] <rmatte> np

[28-Nov-2011 11:23:53] <fixxxermet> Hey everyone, I'm having trouble finding the right documentation for writing my own monitoring templates, along with data sources and all that jazz.

[28-Nov-2011 11:23:59] <fixxxermet> Could someone give me a pointer?

[28-Nov-2011 11:30:18] <rmatte> fixxxermet: it's detailed in the Zenoss Admin guide

[28-Nov-2011 11:30:38] <rmatte>

[28-Nov-2011 11:31:26] <rmatte> ...and even more technical:

[28-Nov-2011 11:44:58] <fixxxermet> thank ya rmatte

[28-Nov-2011 11:55:59] <Glacee> is there any kind of SSO integration beside LDAP?

[28-Nov-2011 11:56:26] <Glacee> could not find documentation on it

[28-Nov-2011 12:00:01] [disconnected at Mon Nov 28 12:00:01 2011]

[28-Nov-2011 12:00:02] [connected at Mon Nov 28 12:00:02 2011]

[28-Nov-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]

[28-Nov-2011 12:05:16] <rmatte> Glacee: you can also do Active Directory

[28-Nov-2011 12:05:22] <rmatte> but those are the only two

[28-Nov-2011 12:05:47] <rmatte> it doesn't support RADIUS, or TACACS or whatever else

[28-Nov-2011 12:08:49] <rmatte> Hackman238: you around?

[28-Nov-2011 12:21:15] <humanix> hi there

[28-Nov-2011 12:21:21] <humanix> hi rmatte Hackman238

[28-Nov-2011 12:21:57] <rmatte> hi

[28-Nov-2011 12:43:52] <rmatte> grrrrr, I have a templat the zenperfsnmp is using perfectly fine, but when I add a zencommand datasource to it, zencommand completely ignores it

[28-Nov-2011 12:43:56] <rmatte> and I can't figure out why

[28-Nov-2011 12:44:02] <rmatte> template*

[28-Nov-2011 12:44:26] <rmatte> I have a templat that* zenperfsnmp is using perfectly fine

[28-Nov-2011 12:48:34] <Glacee> rmatte: thanks for your answer

[28-Nov-2011 12:49:00] <Glacee> any plans on for SAML?

[28-Nov-2011 12:50:40] <Glacee> or OpenID?

[28-Nov-2011 12:51:37] <rmatte> The LDAP and AD authentication are just Zope plugins, Zenoss Inc. did not develop them.

[28-Nov-2011 12:51:45] <Glacee> oh ok thanks

[28-Nov-2011 12:53:46] <rmatte> If you can find plugins for Zope that handle those then you can probably use them.

[28-Nov-2011 12:59:39] <Glacee>

[28-Nov-2011 13:00:15] <rmatte> you'd have to try it to see if it'll work properly

[28-Nov-2011 13:00:23] <Glacee> yeah

[28-Nov-2011 13:00:44] <Glacee> lot of integration work anyway for what we want to do

[28-Nov-2011 13:00:50] <tightwork> rmatte: Hey remember I had that problem for bad oids on the virtual interfaces? Rather than remove the oid, is there anyway to just filter that oid out for the virtual interfaces? I'd like to have errors for regular interfaces.

[28-Nov-2011 13:09:14] <rmatte> tightwork: you can write a transform to filter out the events, or you can create local copies of the template on those interfaces and remove those OIDs... though I told you to copy the ethernetCsmacd template and rename it to match the type of those specific interfaces, then remove the errors datasources/graphs from that template

[28-Nov-2011 13:09:30] <rmatte> Interfaces automatically use whatever template that has a name matching the type of the interface

[28-Nov-2011 13:10:10] <rmatte> also, there's no such thing as a "regular interface"

[28-Nov-2011 13:11:32] <tightwork> true, bad wordage. Ok thanks again. just clarifying before I go and do this

[28-Nov-2011 13:12:17] <rmatte> I'll give you an example... let's say I have a Cisco switch... It has several GigabitEthernet interfaces with a type of ethernetCsmacd_64 and it also has several Vlan interfaces with type of propVirtual_64

[28-Nov-2011 13:12:40] <rmatte> I would copy the ethernetCsmacd_64 template, and then rename the copy to propVirtual_64

[28-Nov-2011 13:12:54] <rmatte> I would then modify that template and remove the error datasources/graph

[28-Nov-2011 13:12:59] <tightwork> then delete the bad errors

[28-Nov-2011 13:13:03] <tightwork> ah ok I get it

[28-Nov-2011 13:13:27] <tightwork> pretty intuitive the types relate to the templates like that

[28-Nov-2011 13:13:58] <rmatte> yeh, and remember that the type is not the actual name of the interface

[28-Nov-2011 13:14:04] <rmatte> it's an actual property given by SNMP

[28-Nov-2011 13:14:08] <rmatte> it's available in the interface details

[28-Nov-2011 13:14:22] <rmatte> which is accessible via the dropdown, which by default is set to show graphs

[28-Nov-2011 13:15:56] <rmatte> It might be called Interface Properties, I can't remember

[28-Nov-2011 13:16:14] <rmatte> My Zenoss 3.x instance is fubar at the moment so I can't really check

[28-Nov-2011 13:25:32] <tightwork> Shows as Details in the drop down. Yes I see the type.. good to know

[28-Nov-2011 13:25:57] <rmatte> cool

[28-Nov-2011 13:26:30] <tightwork> yea even cooler this is where shows me duplex mode.. was thinking earlier where that is

[28-Nov-2011 13:28:59] <rmatte> yup

[28-Nov-2011 13:33:41] <rmatte> man, did everyone take today off or something?

[28-Nov-2011 13:34:07] <rmatte> I need to speak with Hackman, or Egor, or rocket about some issues I'm having with this ZenPack that I'm making

[28-Nov-2011 13:38:37] tvincent_ is now known as tvincent

[28-Nov-2011 13:39:07] tvincent_ is now known as tvincent

[28-Nov-2011 13:39:55] tvincent_ is now known as tvincent

[28-Nov-2011 14:22:44] <rmatte> I'm going to tear my hair out soon if I don't figure out why the hell zencommand won't recognize command datasources in this template

[28-Nov-2011 14:22:56] <rmatte> It was working yesterday

[28-Nov-2011 14:34:22] <rmatte> yes, finally got the commands to execute

[28-Nov-2011 14:35:25] <rmatte> Had to do...

[28-Nov-2011 14:35:26] <rmatte> from import getDeviceCommands

[28-Nov-2011 14:35:33] <rmatte> cmds = getDeviceCommands(find("nameofyourdevice"))

[28-Nov-2011 14:35:36] <rmatte> in zendmd

[28-Nov-2011 14:35:59] <rmatte> now to see if it'll still run them after a restart of Zenoss

[28-Nov-2011 14:37:04] <rmatte> it works, a bit intensive though

[28-Nov-2011 14:50:26] <LIBERIA> how do i set up alert

[28-Nov-2011 14:51:47] <LIBERIA> i am new to zenoss

[28-Nov-2011 14:52:49] <rmatte> LIBERIA: Read -

[28-Nov-2011 15:09:32] <rmatte> If you read that top to bottom it'll answer about 90% of the questions that you're likely to ask

[28-Nov-2011 16:40:55] <LIBERIA> thank you

[28-Nov-2011 16:42:23] <rmatte> no problem

[28-Nov-2011 17:38:27] kocolosk is now known as kocolosk|away

[28-Nov-2011 18:10:49] <w0lv3> hi

[28-Nov-2011 18:10:54] <w0lv3> I can generate a report the Tresholds in zdmd?

[28-Nov-2011 18:11:34] <w0lv3> "Thresholds"

[28-Nov-2011 19:48:35] kocolosk|away is now known as kocolosk

[29-Nov-2011 00:00:01] [disconnected at Tue Nov 29 00:00:01 2011]

[29-Nov-2011 00:00:02] [connected at Tue Nov 29 00:00:02 2011]

[29-Nov-2011 00:00:13] [zenoss-logger (logger bot) has joined #zenoss]

[29-Nov-2011 00:04:08] kocolosk is now known as kocolosk|away

[29-Nov-2011 08:22:58] <humanix> hi there

[29-Nov-2011 08:52:01] <fixxxermet> hai

[29-Nov-2011 09:29:14] <enkrypt> hi, is there any Zenoss support staf present that is up to date on my enterprise issue 16052-2480 ?

[29-Nov-2011 09:44:49] <enkrypt> is anyone here familiar with the zenoss enterprise permission system?

[29-Nov-2011 09:45:03] <enkrypt> I want to give a user permission to administer 1 specific asset, change the filesystem thresholds etc

[29-Nov-2011 09:45:38] <enkrypt> when I give him ZenManager on that object, he can change the thresholds, but he can even do that without doing a local copy first, and thereby he changes the GLOBAL template, which is quite dangerous

[29-Nov-2011 09:45:44] <enkrypt> how do I solve this?

[29-Nov-2011 10:16:29] <jmp242> Morning all

[29-Nov-2011 10:18:30] <rocket> Morning jmp242

[29-Nov-2011 10:36:26] <jmp242> Quiet today. Is everyone still recovering from the holidays?

[29-Nov-2011 10:47:29] <jb> enkrypt: let me know if you figure that out

[29-Nov-2011 10:50:09] <jb> still can't get this objects.xml to load correctly.

[29-Nov-2011 10:50:36] <jb> i'm trying to take a objects.xml from another zenpack that contains SNMP transforms only.. and put it in a new zenpack.  i created the required event classes and all, and the transforms are still missing.

[29-Nov-2011 10:50:38] <jb> any ideas?

[29-Nov-2011 10:50:38] <dhopp> jmp242:  I'm still recovering from last night

[29-Nov-2011 10:50:46] <dhopp> CRAZY cyber monday where I work

[29-Nov-2011 10:51:40] <jmp242> wow

[29-Nov-2011 10:55:17] kocolosk|away is now known as kocolosk

[29-Nov-2011 10:56:37] <rmatte> wow, there's actually life in here today

[29-Nov-2011 10:57:30] <rmatte>

[29-Nov-2011 10:57:53] <rmatte> If anyone can provide some suggestions of how I can do what I'm trying there it would be appreciated

[29-Nov-2011 11:06:18] <fragfutter> rmatte: is the snmpd native/part of windows?

[29-Nov-2011 11:06:43] <rmatte> fragfutter: yup

[29-Nov-2011 11:07:14] <rmatte> I stay away from snmp-informant at all costs lol

[29-Nov-2011 11:07:28] <fragfutter> rmatte: i stay away from windows...

[29-Nov-2011 11:07:42] <rmatte> unfortunately our clients don't

[29-Nov-2011 11:07:46] <rmatte>

[29-Nov-2011 11:08:27] <rmatte> I'll be so happy when this pack is done... WMI really really sucks

[29-Nov-2011 11:09:24] <fragfutter> so the oid is there, then the service is running, oid missing, service is stopped?

[29-Nov-2011 11:09:26] <jmp242> why does WMI suck so bad?

[29-Nov-2011 11:09:51] <rmatte> jmp242: because I currently have 60 unresolved tickets for WMI failing on servers

[29-Nov-2011 11:10:00] <rmatte> we fix them, and then a few weeks later, WMI craps out again

[29-Nov-2011 11:10:03] <rmatte> it's a never ending circle

[29-Nov-2011 11:10:23] <rmatte> and it's not isolated to any particular environment, happens for all of our customers

[29-Nov-2011 11:10:52] <fragfutter> rmatte: do you need anything below the oid?

[29-Nov-2011 11:10:57] <rmatte> once I get SNMP service monitoring we can stop using WMI on the majority of them and only use it in an extremely targetted fashion

[29-Nov-2011 11:11:12] <rmatte> fragfutter: not sure I follow?

[29-Nov-2011 11:11:15] <rmatte> below?

[29-Nov-2011 11:11:59] <fragfutter> rmatte: you have a long oid (containing the textual representation of the service name). Do you simply check if this oid exists or do you pull some values from there.

[29-Nov-2011 11:12:44] <rmatte> is service names, is service status, but when a service is stopped, instead of the status value actually changing, the OID gets removed

[29-Nov-2011 11:13:05] <rmatte> I think the status value is only used in specific cases

[29-Nov-2011 11:13:40] <rmatte> but therein lies the issue, I can't just use a threshold like I want to unless I use a command datasource instead of SNMP that returns a 0 if the OID doesn't exist

[29-Nov-2011 11:14:02] <rmatte> but the problem with that is that it's not going to scale well, can't have 50+ scripts kicking off each polling cycle for every server

[29-Nov-2011 11:14:08] <rmatte> will bring the monitoring tools to their knees

[29-Nov-2011 11:15:28] <rmatte> I was thinking of using a transform on the debug event, but the problem with that is, how to handle an auto-clear

[29-Nov-2011 11:16:46] <rmatte> I figured that I'd see if anyone out there has a good idea that I haven't thought of.

[29-Nov-2011 11:16:48] <Crashilda-Bot>  lol

[29-Nov-2011 11:17:04] <rmatte> lol indeed

[29-Nov-2011 11:17:21] <fragfutter> rmatte: might be necessary to write your own monitoring daemon

[29-Nov-2011 11:17:59] <rmatte> fragfutter: yeh, it may be, but I want to avoid that if possible.  Seems like you need to write your own daemon to do anything with Zenoss these days, lol

[29-Nov-2011 11:57:23] kocolosk is now known as kocolosk|away

[29-Nov-2011 11:57:29] kocolosk|away is now known as kocolosk

[29-Nov-2011 12:00:01] [disconnected at Tue Nov 29 12:00:01 2011]

[29-Nov-2011 12:00:02] [connected at Tue Nov 29 12:00:02 2011]

[29-Nov-2011 12:00:14] [zenoss-logger (logger bot) has joined #zenoss]

[29-Nov-2011 13:31:27] <jmp242> Hey, anyone have a comment for:  ??

[29-Nov-2011 13:31:54] <jmp242> do you ever stop polling devices for a set time and then start? Kind of like maintenance windows, but not just for alerting rules - actually stop sending SNMP polls etc...

[29-Nov-2011 13:40:22] <rmatte> Zenoss actually polls every 300 seconds by default, not 60 seconds

[29-Nov-2011 13:42:03] <otakup0pe> easy to change to 60s though ! but it gets hairy if you go below that.

[29-Nov-2011 13:42:28] <otakup0pe> hey is there any way to specify the 'production state' for a infrastructure/devices view via the url ? query string or something ?

[29-Nov-2011 13:43:47] <jmp242> mmm

[29-Nov-2011 13:43:51] <jmp242> was it pings that are 60 seconds

[29-Nov-2011 13:44:10] <jmp242> there are some things that are every 60, 180 and 300 that I recall by default, I just don't recall the specifics

[29-Nov-2011 13:44:26] <jmp242> not to mention the modeler which is every 12 hrs IIRC

[29-Nov-2011 13:45:13] <rmatte> pings are 60, yes

[29-Nov-2011 13:48:14] <dhopp> defaults are: pings 60, process 180, snmp 300

[29-Nov-2011 13:48:31] <dhopp> status I think is 60

[29-Nov-2011 13:48:31] <otakup0pe> all commands (not just snmp) default to 300

[29-Nov-2011 13:48:34] <dhopp> as well

[29-Nov-2011 13:51:38] <jmp242> yea, process is SNMP though

[29-Nov-2011 13:52:07] <jmp242> so the guy wanted to prevent all SNMP, so I was a little off, but it is more often than 300 seconds

[29-Nov-2011 13:52:53] <rmatte> nah he didn't want all SNMP, only on specific devices

[29-Nov-2011 13:53:05] <rmatte> What I just replied with should be sufficient for him

[29-Nov-2011 13:53:12] <jmp242> sorry, yes you're right

[29-Nov-2011 13:53:19] <jmp242> I meant all snmp for the specific devices

[29-Nov-2011 13:53:29] <jmp242> which process would count for right?

[29-Nov-2011 13:53:34] <jmp242> I do like your reply

[29-Nov-2011 13:53:54] <jmp242> it's good to know that we could use something that acts like "decomissioned" without calling it that (to prevent user confusion)

[29-Nov-2011 13:54:03] <jmp242> I assume I can just create that in the GUI

[29-Nov-2011 13:54:22] <rmatte> yup, we do it for certain devices that we want in the database simply for tracking purposes but not monitoring purposes

[29-Nov-2011 13:54:35] <rmatte> and yes, you can just create that in the GUI

[29-Nov-2011 13:57:00] <jmp242> rmatte:  while we're talking about small config stuff, what do you think re:

[29-Nov-2011 13:57:58] <jmp242> process matching is always a bit fidly

[29-Nov-2011 13:58:10] <jmp242> though we here seem to have less issues than many forum posters

[29-Nov-2011 14:01:13] <otakup0pe> as fiddly as regexp's tend to be

[29-Nov-2011 14:37:37] <nyeates> yo yo yo, merry xmas

[29-Nov-2011 14:50:40] <froztbyte> <3 catb-esr

[29-Nov-2011 15:07:59] <rmatte> jmp242: sorry, I went awol there for a bit hehe

[29-Nov-2011 15:08:25] <jmp242> no problem

[29-Nov-2011 15:08:26] <jmp242> I did too lol

[29-Nov-2011 15:09:14] <rmatte> Regarding that post you pointed out:

[29-Nov-2011 15:09:22] <rmatte> known bug

[29-Nov-2011 15:10:16] <jmp242> yea

[29-Nov-2011 15:10:17] <jmp242> ok

[29-Nov-2011 15:10:34] <rmatte> I just replied with that link

[29-Nov-2011 15:10:39] <jmp242> oh ok

[29-Nov-2011 15:10:41] <jmp242> saved me doing it

[29-Nov-2011 15:10:45] * rmatte nods

[29-Nov-2011 15:11:53] <rmatte> I wish there was an easy way to tell Zenoss to return a 0 value instead of a debug event if it can't poll a certain OID

[29-Nov-2011 15:13:44] <jmp242> lol

[29-Nov-2011 15:13:47] <jmp242> I heard

[29-Nov-2011 15:13:51] <jmp242> submit a feature request?

[29-Nov-2011 15:14:01] <rmatte> meh, they never make it in lol

[29-Nov-2011 15:14:03] <jmp242> it's probably something you could do in editing the code in zenperfsnmp

[29-Nov-2011 15:14:16] <rmatte> there are tons of feature requests that I'm dying to see go in that haven't been touched

[29-Nov-2011 15:14:26] <rmatte> they develop based on what their sales team tells them to develop

[29-Nov-2011 15:14:48] <jmp242> yes, that is probably true

[29-Nov-2011 15:14:59] <rmatte> oh, I could easily add it in myself, but I'd rather not because I am planning to upgrade

[29-Nov-2011 15:15:04] <jmp242> true

[29-Nov-2011 15:15:12] <jmp242> That's the issue, unless you do it as a Zenpack

[29-Nov-2011 15:15:18] <jmp242> you can't really get it into Core

[29-Nov-2011 15:15:38] <rmatte> it'd have to use monkeypatching or somesuch which is messy

[29-Nov-2011 15:16:30] <jmp242> oh

[29-Nov-2011 15:16:31] <jmp242> crap

[29-Nov-2011 15:16:54] <jmp242> if you're willing to donate the code copyright to zenoss lol, they might take a patch??

[29-Nov-2011 15:17:34] <rmatte> meh, I'll just code my own collector I guess lol

[29-Nov-2011 15:19:24] <jmp242> yea

[29-Nov-2011 15:19:55] <rmatte> It just seems like a really extreme thing to code a whole collector for

[29-Nov-2011 15:20:00] <rmatte>

[29-Nov-2011 15:20:21] <rmatte> and what do I even call it?  zenwinservicesnmp?

[29-Nov-2011 15:20:22] <rmatte> lol

[29-Nov-2011 15:20:58] <jmp242> mmm

[29-Nov-2011 15:21:00] <jmp242> maybe?

[29-Nov-2011 15:21:14] <jmp242> zenwinperfsnmp

[29-Nov-2011 15:21:30] <jmp242> maybe?

[29-Nov-2011 15:21:31] <rmatte> it's not really collecting performance though, just up/down stats

[29-Nov-2011 15:21:32] <jmp242> IDK

[29-Nov-2011 15:21:35] <jmp242> mmm

[29-Nov-2011 15:21:36] <rmatte>

[29-Nov-2011 15:21:37] <jmp242> that's painful

[29-Nov-2011 15:21:44] <rmatte> anyways, I'll sort it out

[29-Nov-2011 15:21:55] <jmp242> good luck

[29-Nov-2011 15:21:58] <rmatte> I need this monitoring badly hehe

[29-Nov-2011 15:28:00] <jmp242> god I love tech support chat

[29-Nov-2011 15:28:09] <jmp242> always a fight because we have a campus address

[29-Nov-2011 15:28:14] <jmp242> and it's not a street address

[29-Nov-2011 15:36:31] <nyeates> just cought up on some of above...

[29-Nov-2011 15:37:08] <nyeates> I dont quite understand the issue

[29-Nov-2011 15:38:10] <nyeates> Sorry if over the last few months feature req's have been lower than normal even....I stopped going to our defect review meetings b/c im focusing on other stuff

[29-Nov-2011 15:38:26] <nyeates> This means 1 less person pushing for tickets to actually be handled there

[29-Nov-2011 15:38:37] <nyeates> community tickets rather

[29-Nov-2011 15:39:30] <rmatte> In all the time I've been using Zenoss I don't think I've actually seen a community feature request implemented.  Any new features are always sales driven.

[29-Nov-2011 15:39:44] <rmatte> I've seen lots of bug fixes though.

[29-Nov-2011 15:39:49] <nyeates> ahhh feature rquest

[29-Nov-2011 15:40:57] <nyeates> Yeah those are difficult to come by.... its a good thing that 4.2 is going to introduce a metric ton of features to core

[29-Nov-2011 15:40:58] <rmatte> It gets a bit depressing for the community after a while

[29-Nov-2011 15:41:13] <rmatte> that does sound lovely

[29-Nov-2011 15:41:18] <rmatte> can't wait to see the list of new stuff

[29-Nov-2011 15:41:57] <nyeates> I have not event seen an official internal announcement - but from what i heard - all new event architecture and DB schema - this includes the new actions and triggers

[29-Nov-2011 15:42:01] <nyeates> ipv6 stuffs

[29-Nov-2011 15:42:20] <nyeates> some scaling stuff

[29-Nov-2011 15:42:42] <rmatte> one big thing that I really wanted to see was the ability to "tag" devices/components

[29-Nov-2011 15:42:56] <rmatte> so that I can actually define what interfaces are WAN interfaces for instance

[29-Nov-2011 15:43:09] <nyeates> like instead of the hierarchical relations?

[29-Nov-2011 15:43:22] <rmatte> so that I could do something like...

[29-Nov-2011 15:43:31] <jmp242> god, acer is looking that they may not honor their warranties if their system doesn't like your address

[29-Nov-2011 15:43:32] <rmatte> for i in os.interfaces():

[29-Nov-2011 15:43:43] <rmatte>  if "WAN" in i.tag:

[29-Nov-2011 15:43:48] <rmatte>   do whatever

[29-Nov-2011 15:44:13] <rmatte> jmp242: acer are the worst maker of computers on the planet, followed closely by gateway

[29-Nov-2011 15:44:25] <rmatte> you couldn't pay me to own one

[29-Nov-2011 15:44:26] <rmatte> lol

[29-Nov-2011 15:44:28] <jmp242> really? I've liked their stuff up till I'm trying to get a monitor repaired

[29-Nov-2011 15:44:52] <rmatte> In my experience (and friend's experiences) their stuff is very poor quality

[29-Nov-2011 15:44:54] <jmp242> Though I only buy their monitors

[29-Nov-2011 15:45:14] <rmatte> I did own two acer monitors and they lasted the test of time, but I've seen lots of friend's acer computers die

[29-Nov-2011 15:45:19] <jmp242> yea

[29-Nov-2011 15:45:24] <jmp242> same happened to my sisters laptop

[29-Nov-2011 15:45:33] <rmatte> yeh, their laptops are absolutely horrible

[29-Nov-2011 15:45:34] <rmatte> lol

[29-Nov-2011 15:45:35] <jmp242> though it did manage, after some repairs, to last 3-4 years

[29-Nov-2011 15:45:46] <jmp242> she liked it a lot though it didn't like sawdust

[29-Nov-2011 15:46:00] <rmatte> I have a friends that went through three of them before I told them to stop buying acer laptops

[29-Nov-2011 15:46:08] <rmatte> they switched to an HP laptop and it's still working

[29-Nov-2011 15:46:09] <rmatte> lol

[29-Nov-2011 15:46:12] <jmp242> I only recommend Macs and Lenovo for laptops myself

[29-Nov-2011 15:46:14] <rmatte> a friend*

[29-Nov-2011 15:46:51] <rmatte> ah, the brands I recommend are Panasonic, HP, Lenovo, Mac, and Asus

[29-Nov-2011 15:46:53] <jmp242> though I don't use or know macs at all, I hear they're well built from my colleague who supports all the macs here

[29-Nov-2011 15:46:59] <nyeates> rmatte: yeah its a good idea....ive seen similar things as internal feature requests for 4+ yrs.

[29-Nov-2011 15:47:08] <rmatte> Mac are really well built but incredible overpriced for what they are

[29-Nov-2011 15:47:43] <jmp242> HP hasn't wowed me recently with the PCs going, then not, plus the crap for business desktops we got recently (though it was a great "Cornell" deal and really cheap)

[29-Nov-2011 15:47:55] <rmatte> nyeates: the main feature request, to allow device dependencies/relations still hasn't been addressed after years and years of the community begging for it

[29-Nov-2011 15:48:05] <rmatte> and then they go and stick in an enterprise only package that does it

[29-Nov-2011 15:48:09] * rmatte cries

[29-Nov-2011 15:48:35] <jmp242> I think Hackman238 is writing a zenpack for that

[29-Nov-2011 15:48:44] <rmatte> HP is actually going to stop building PCs soon, they are selling off their production lines

[29-Nov-2011 15:48:47] <nyeates> Yeah i mean.... that is a *huge* money maker. We cant give innovation away.

[29-Nov-2011 15:48:52] <rmatte> they are going to be focusing on server only like IBM

[29-Nov-2011 15:49:05] <rmatte> Actually, I forgot to mention Dell, their stuff is quite good

[29-Nov-2011 15:49:09] <jmp242> LOL

[29-Nov-2011 15:49:13] <jmp242> I avoid Dell at all costs

[29-Nov-2011 15:49:19] <jmp242> nothing but crap in all my experiances

[29-Nov-2011 15:49:28] <nyeates> Rahter...We can give *some* innocation away....and that will come in 4.2  :-)

[29-Nov-2011 15:49:32] <rmatte> Some of it is crap, you just have to buy the higher end stuff

[29-Nov-2011 15:49:35] <froztbyte> "oh hell, it's a dell"

[29-Nov-2011 15:49:46] <rmatte> their cheap lower end laptops are fisher price

[29-Nov-2011 15:49:50] <jmp242> Oh, we've bought precision workstations which are ... ok

[29-Nov-2011 15:50:06] <rmatte> Dell's monitors are the best out there in my opinion

[29-Nov-2011 15:50:09] <jmp242> but cost more than the Lenovo S20s which are comparable - and the support is just nowhere

[29-Nov-2011 15:50:10] <rmatte> in terms of quality vs price

[29-Nov-2011 15:50:28] <froztbyte> I'm strongly considering apple kit for my next laptop

[29-Nov-2011 15:50:32] <jmp242> Our worst experiances with Dell were actually in the servers

[29-Nov-2011 15:50:37] <froztbyte> the build quality on them is just totally awesome

[29-Nov-2011 15:50:46] <rmatte> I wouldn't touch dell for servers

[29-Nov-2011 15:50:56] <rmatte> I'm personally a fan of HP for servers

[29-Nov-2011 15:51:07] <froztbyte> amen

[29-Nov-2011 15:51:15] <jmp242> I've been a big IBM fan, though I think the 40% discount is part of it

[29-Nov-2011 15:51:16] <froztbyte> iLO >> DRAC

[29-Nov-2011 15:51:24] <rmatte> IBM is a close second to HP for me

[29-Nov-2011 15:51:30] <nyeates> Dell PC treated me well in past

[29-Nov-2011 15:51:32] <jmp242> but we just get the cheap X series ones

[29-Nov-2011 15:51:32] <froztbyte> jmp242: yeah, it seems like IBM's a big thing in the university space

[29-Nov-2011 15:51:42] <nyeates> id never get their laptops tho

[29-Nov-2011 15:51:44] <jmp242> Well, they never seem to break

[29-Nov-2011 15:51:45] <froztbyte> I haven't ever had much chance to play with their gear

[29-Nov-2011 15:51:47] <jmp242> or not work

[29-Nov-2011 15:51:58] <rmatte> I have an xSeries machine sitting in my basement rotting away

[29-Nov-2011 15:51:59] <froztbyte> actually no, I lie

[29-Nov-2011 15:52:00] <jmp242> they just go so long we don't know what warranty support for the servers is

[29-Nov-2011 15:52:00] <rmatte> lol

[29-Nov-2011 15:52:11] <froztbyte> I have one hand-me-down IBM server in service now

[29-Nov-2011 15:52:14] <jmp242> once when we had a desktop they sent out a tech

[29-Nov-2011 15:52:19] <jmp242> same day to replace the HD

[29-Nov-2011 15:52:21] <froztbyte> got it from a previous client of ours who shut down

[29-Nov-2011 15:52:38] <rmatte> jmp242: yeh, when we call them for hardware replacement at customer sites the service is quite good

[29-Nov-2011 15:52:48] <froztbyte> and I've been reasonably happy with it so far

[29-Nov-2011 15:52:56] <jmp242> Lenovo seemed to keep the same level support for the Think products

[29-Nov-2011 15:53:00] <jmp242> so I recommend them to many people

[29-Nov-2011 15:53:05] <nyeates> For laptops, for me, its apple or ibm or sony (maybe)

[29-Nov-2011 15:53:15] <jmp242> I just tend to get right through

[29-Nov-2011 15:53:17] <jmp242> w/out the bullshit

[29-Nov-2011 15:53:35] <froztbyte> we have quite a few lenovo/thinkpad units at the office

[29-Nov-2011 15:53:47] <froztbyte> and there's a strong preference for their X series

[29-Nov-2011 15:53:57] <jmp242> and they send a box for repair

[29-Nov-2011 15:54:00] <froztbyte> but yeah, MBP or MBA it probably is in my near future

[29-Nov-2011 15:54:09] <froztbyte> I can live with OSX, and I'll dualboot a linux

[29-Nov-2011 15:54:23] <jmp242> yea - I don't use laptops but for work

[29-Nov-2011 15:54:24] <nyeates> I still dont get why, to this day, all the pc vendors think its grand to put all  kinds of their own special software on their laptops and pc's when shipped.

[29-Nov-2011 15:54:26] <jmp242> so I get T series

[29-Nov-2011 15:54:32] <jmp242> yea

[29-Nov-2011 15:54:39] <jmp242> Lenovos are better w/out the special software IMO

[29-Nov-2011 15:54:54] <jmp242> and their stuff isn't as bad as the crap at Best Buy lol

[29-Nov-2011 15:55:15] <froztbyte> the only special software I really need on my computers is a linux distribution I can tweak to my heart's desire

[29-Nov-2011 15:55:40] <froztbyte> my work laptop accidentally turned into my main workspace

[29-Nov-2011 15:55:54] <froztbyte> which is surprising, since it's an SL400 with spinning rust for storage, etc

[29-Nov-2011 15:56:16] <nyeates> What kind of server hardware do you guys get for general purpose computing or to become *nix servers or to become virtualized, etc?

[29-Nov-2011 15:56:22] <nyeates> does it run the gamut?

[29-Nov-2011 15:57:11] <froztbyte> our hardware stack on the server side is a mix of DL180s, DL360s and BL3000c chassis installs

[29-Nov-2011 15:57:30] <froztbyte> depends what time stuff was installed and whether we had a spare box from elsewhere lying around

[29-Nov-2011 15:57:48] <froztbyte> usually G5 through G7 hardware, although some of our blade stuff is still G2 and G3, I think

[29-Nov-2011 15:57:59] <jmp242> We have some Dell PowerEdge 2900s that suck horribly

[29-Nov-2011 15:58:02] <jmp242> but we have to use to kill them

[29-Nov-2011 15:58:20] <jmp242> all new servers are IBM X 3500s and up, many 1U 3650s IIRC

[29-Nov-2011 15:58:31] <rmatte> I forget what the exact hardware that we use is, but it's an HP blade chassis with 5 HP blade servers in it

[29-Nov-2011 15:58:35] <jmp242> And we're maybe considering some X 3900

[29-Nov-2011 15:58:40] <froztbyte> we're not a cloud house or somesuch, though. carrier network / network service provider

[29-Nov-2011 15:58:40] <jmp242> we don't use blade chassis

[29-Nov-2011 15:58:41] <rmatte> we have close to 100 VMs running on them

[29-Nov-2011 15:58:43] <nyeates> the DL and BL stuff is all HP?

[29-Nov-2011 15:58:52] <rmatte> with capacity for around 3 times that

[29-Nov-2011 15:59:06] <froztbyte> rmatte: 5? some storage blades in there as well?

[29-Nov-2011 15:59:08] <froztbyte> nyeates: yes

[29-Nov-2011 15:59:16] <rmatte> froztbyte: 3 SANs

[29-Nov-2011 15:59:24] <froztbyte> nyeates: BL is the blade series product code, DL is the rest

[29-Nov-2011 15:59:26] <rmatte> linked via fiber

[29-Nov-2011 15:59:31] <froztbyte> ML is the standing chassis line

[29-Nov-2011 15:59:39] <froztbyte> rmatte: ah right

[29-Nov-2011 15:59:58] <froztbyte> FC to EMC clusters, or what?

[29-Nov-2011 16:00:32] <rmatte> lol, we got our first SAN and it was slower than all hell, so we switched to a newer faster SAN, but that still wasn't cutting it so we just got an HP Eva the other day

[29-Nov-2011 16:00:43] <froztbyte> well

[29-Nov-2011 16:00:50] <rmatte> all 3 of them are running, we just balance stuff out between them

[29-Nov-2011 16:00:56] <jmp242> Lucky

[29-Nov-2011 16:01:04] * froztbyte finds a URL

[29-Nov-2011 16:01:07] <nyeates> rmatte: hopefully you all didnt have to keep each of those things you purchased and that didnt work out

[29-Nov-2011 16:01:18] <nyeates> ouch

[29-Nov-2011 16:01:23] <jmp242> we only can afford Infortrend and iSCSI on 10Gb ethernet

[29-Nov-2011 16:01:27] <rmatte> it's not bad to have the 3 actually

[29-Nov-2011 16:01:40] <rmatte> we can put the servers that user IO less intesively on the slower SANs

[29-Nov-2011 16:01:53] <rmatte> and keep the fast ones for the monitoring servers and such

[29-Nov-2011 16:02:32] <nyeates> Have any of you used public or internal private cloud / virt resources to replace or append or burst any existing or new infrastructure?

[29-Nov-2011 16:03:05] <jmp242> I haven't

[29-Nov-2011 16:03:08] <froztbyte> holy dear god it is difficult getting a URL for an old tweet out of twitter

[29-Nov-2011 16:03:10] <jmp242> not for lack of wanting perhaps

[29-Nov-2011 16:03:25] <froztbyte> anyway

[29-Nov-2011 16:03:36] <froztbyte> Regarding SANs, shared storage and FCoE, this is basically the general problem with it:

[29-Nov-2011 16:03:38] <jmp242> I'd like to potentially use Rackspace or Amazon for some testing Zenoss installs

[29-Nov-2011 16:03:38] <rmatte> nyeates: I wouldn't trust cloud for anything other than hosting web apps personally

[29-Nov-2011 16:03:41] <froztbyte> rmatte: ^

[29-Nov-2011 16:03:55] <jmp242> but time time time lol

[29-Nov-2011 16:04:13] <nyeates> rmatte: huh. I would. Depends what you are doign for sure though.

[29-Nov-2011 16:04:25] <rmatte> froztbyte: hahaha, yeh, tell me about it

[29-Nov-2011 16:04:57] <froztbyte> nyeates: I think very few places in the world do "private cloud" the right way

[29-Nov-2011 16:04:58] <nyeates> Also I wasnt saying only just public cloud....therea re also internal cloud resources if you work at govt or orgs that decide to do this

[29-Nov-2011 16:05:03] <froztbyte> usually it just ends up being a VM farm

[29-Nov-2011 16:05:11] <rmatte> nyeates: an internal cloud is different

[29-Nov-2011 16:05:18] <nyeates> froztbyte: true

[29-Nov-2011 16:06:07] <froztbyte> hell

[29-Nov-2011 16:06:35] <froztbyte> most of these tards who program "cloud systems" and tout scalability hardly even get that right the first few iterations

[29-Nov-2011 16:06:35] <nyeates> It would seem like you would need a pretty big org or one that is heavy on IT use, to use an internal / priv cloud

[29-Nov-2011 16:06:49] <froztbyte> nyeates: depends

[29-Nov-2011 16:07:13] <froztbyte> we're a 12-person company, but we've considered building a n-host vm platform on openstack or whatever

[29-Nov-2011 16:07:22] <froztbyte> just to make provisioning on the dev side easier

[29-Nov-2011 16:07:39] <rmatte> well... honestly I haven't really seen a good definition of exactly what a "cloud" entices... I mean, my 5 blade servers running vmware across them could technically be deemed a "cloud"

[29-Nov-2011 16:07:42] <froztbyte> we specifically need to go with that option because our software doesn't play well with EC2 and friends

[29-Nov-2011 16:07:51] <nyeates> so basically a nicer acting VM provisioning yes?

[29-Nov-2011 16:07:58] <froztbyte> nyeates: yeah

[29-Nov-2011 16:08:21] <froztbyte> rmatte: "cloud" is supposed to mean "as much resource available as you want without you having to worry about the hardware"

[29-Nov-2011 16:08:26] <froztbyte> nothing more than that

[29-Nov-2011 16:08:35] <froztbyte> then the marketroids happened.

[29-Nov-2011 16:09:10] <rmatte> right, so buy more hardware than you could possibly need and spend ridiculous amounts of money or scale your infrastructure up intelligently

[29-Nov-2011 16:09:22] <nyeates> best definition ive seen is by NIST:

[29-Nov-2011 16:09:32] <nyeates> Cloud computing is a model for enabling convenient, on-demand network access to a shared

[29-Nov-2011 16:09:33] <nyeates> pool of configurable computing resources (e.g., networks, servers, storage, applications, and

[29-Nov-2011 16:09:33] <nyeates> services) that can be rapidly provisioned and released with minimal management effort or

[29-Nov-2011 16:09:33] <nyeates> service provider interaction

[29-Nov-2011 16:09:47] <froztbyte> rmatte: well, not necessarily buy your own infrastructure. that's the bit where EC2 makes their money

[29-Nov-2011 16:10:05] <rmatte> right, I was more referring to internal cloud

[29-Nov-2011 16:10:07] <froztbyte> "it's up there in the cloud and it just works" is really as fluffy and good an explanation as I can give to the idea

[29-Nov-2011 16:10:21] <froztbyte> it's literally supposed to be that sort of airy-fairy way to deal with how you get a box

[29-Nov-2011 16:10:25] <froztbyte> rmatte: ah

[29-Nov-2011 16:10:43] <froztbyte> rmatte: well, like I say, I don't think very many of these places ever do "private cloud" the right way

[29-Nov-2011 16:11:22] <froztbyte> the one client I've spoken to has a 3-pager requisition form for spinning up a box

[29-Nov-2011 16:11:37] <froztbyte> even if you might only be using that box for a few hours and then nuking it

[29-Nov-2011 16:11:47] <rmatte> I have a 1 word requesition from: "clone"

[29-Nov-2011 16:11:53] <rmatte> form*

[29-Nov-2011 16:11:58] <nyeates> We are goign to be putting together a test matrix for ZenPacks. Everytime someone changes ZenPack code on github, we will detect it, fire up an EC2 instance of a few versions of zenoss

[29-Nov-2011 16:12:01] <nyeates> and test away

[29-Nov-2011 16:12:13] <rmatte> cool

[29-Nov-2011 16:12:17] <nyeates> pretty neat use case of public cloud

[29-Nov-2011 16:12:28] <froztbyte> that's the way it's supposed to be done

[29-Nov-2011 16:12:48] <froztbyte> I read an article some time ago of one of the big distros' build process, but couldn't find it again

[29-Nov-2011 16:13:00] <froztbyte> what they have is a base farm with all the resources close at hand

[29-Nov-2011 16:13:38] <froztbyte> then when they build and release things, a vm gets spun up completely virgin, install process takes place, everything logs to useful places, and then gets shitcanned if it all works

[29-Nov-2011 16:13:48] <froztbyte> if it didn't, they had a few options of what they could do with it

[29-Nov-2011 16:14:04] <froztbyte> either just log the results for later testing, or image the box and stash it somewhere

[29-Nov-2011 16:14:07] <froztbyte> etc

[29-Nov-2011 16:14:19] <froztbyte> <3 automated testing

[29-Nov-2011 16:14:43] <nyeates> huh cool

[29-Nov-2011 16:14:49] <nyeates> what distro u know?

[29-Nov-2011 16:15:23] <froztbyte> if I find the article again, I'll send you the link

[29-Nov-2011 16:18:11] <nyeates> Ive gone to a few local devops meetings. Some of the stuff they are doing is way way over my head. Using like configuration Mgmt systems to deploy their entire app infrastructures, assure that if boxes go down, new ones automatically come up (and they do because they run internal cloud environs that enable instant machine spin ups). Even how the machines talk to one aonther, say a web app and its DB, the config mgmt can rees

[29-Nov-2011 16:18:11] <nyeates> taablish a totally new virtual box and understand how to change settings on that box to get it quickly talking to the other one the way it expects.

[29-Nov-2011 16:19:12] <froztbyte> you're still working at zenoss, right?

[29-Nov-2011 16:19:23] <nyeates> yes

[29-Nov-2011 16:19:40] <froztbyte> awesome

[29-Nov-2011 16:19:45] <nyeates> ??

[29-Nov-2011 16:19:50] <froztbyte> the fact that some people there are aware of these things is fantastic

[29-Nov-2011 16:20:33] <nyeates> thanks :-)

[29-Nov-2011 16:20:49] <froztbyte> looks like a thing

[29-Nov-2011 16:20:58] <froztbyte> but that appears to only have been the draft proposal

[29-Nov-2011 16:21:01] <nyeates> I dont know the nitty gritty of it....but im following it at a higher level at least i guess

[29-Nov-2011 16:21:27] <nyeates> Matt Ray, who i replaced here in the community, works on this stuff for Chef

[29-Nov-2011 16:21:40] <froztbyte> yeah, I recall

[29-Nov-2011 16:22:12] <nyeates> he made a cookbook that ties Chefs config mgmt stuff into zenoss

[29-Nov-2011 16:23:15] <froztbyte> I *really* need to get my installations over to 3.x from 2.5 :/

[29-Nov-2011 16:23:30] <froztbyte> that way I can actually start using the JSON API instead of hacking at things through the dmd

[29-Nov-2011 16:23:33] <nyeates> or just wait a few months and go to 4.2

[29-Nov-2011 16:23:59] <froztbyte> would I even have the option of jumping straight to 4.2 from 2.5.2?

[29-Nov-2011 16:24:05] <froztbyte> (.deb installation)

[29-Nov-2011 16:24:09] <nyeates> no

[29-Nov-2011 16:24:22] <froztbyte> see, this is why I worry about these things

[29-Nov-2011 16:24:43] <nyeates> How much customization is in place?

[29-Nov-2011 16:24:55] <froztbyte> varies from server to server

[29-Nov-2011 16:25:02] <froztbyte> but none of it that I can't deal with

[29-Nov-2011 16:25:13] <froztbyte> the main issue is just that I always get interrupted

[29-Nov-2011 16:25:33] <froztbyte> and then only get back to it like 3 weeks later

[29-Nov-2011 16:25:54] <nyeates> Might be able to export much of what you need to XML and then suck as much of it back in, and then spend some time with the leftovers.

[29-Nov-2011 16:26:04] <froztbyte> usually it's actually just core + zenpacks + transforms

[29-Nov-2011 16:26:25] <froztbyte> and the only slightly tricky bit is the transforms, and I've already solved that one ages ago

[29-Nov-2011 16:26:46] <froztbyte> nyeates: hmm, yeah, I've been wondering about that before

[29-Nov-2011 16:26:59] <froztbyte> that doesn't preserve the events DBs and such though, does it?

[29-Nov-2011 16:27:08] <nyeates> mmm no

[29-Nov-2011 16:27:14] <nyeates> events db is totally changing schemas

[29-Nov-2011 16:27:25] <nyeates> so you would NEED to be at 3.2.1 and upgrade

[29-Nov-2011 16:27:29] <nyeates> which has a conversion

[29-Nov-2011 16:27:44] <froztbyte> yeah, the goal is to go 2.5.2 -> 3.0 -> 3.2 -> 3.2.1

[29-Nov-2011 16:27:51] <froztbyte> I think that's the flow I had planned, anyway

[29-Nov-2011 16:28:03] <froztbyte> 2 -> 3 is the most major pain

[29-Nov-2011 16:28:31] <froztbyte> I should get back to my attempt of moving the events DB code over to MongoDB

[29-Nov-2011 16:28:46] <nyeates> one thing i find that helps in a large back to back upgrades

[29-Nov-2011 16:28:54] <nyeates> get your system spick and span clean before any upgrades

[29-Nov-2011 16:28:58] <froztbyte> and then a little bit later a cassandra backend too

[29-Nov-2011 16:29:10] <froztbyte> nyeates: how do you mean?

[29-Nov-2011 16:29:13] <nyeates> no or few unknown errors in log files

[29-Nov-2011 16:29:28] <nyeates> zope or zodb will sometimes spew stuff if relations are bad, etc

[29-Nov-2011 16:29:34] <froztbyte> oh

[29-Nov-2011 16:29:37] <froztbyte> good question, that

[29-Nov-2011 16:29:38] <nyeates> get rid of those type things and upgrades go cleaner

[29-Nov-2011 16:29:40] * froztbyte tails some logs

[29-Nov-2011 16:30:41] <nyeates> 2.52

[29-Nov-2011 16:30:43] <nyeates> oops

[29-Nov-2011 16:31:03] <froztbyte> nyeates: oh, a question

[29-Nov-2011 16:31:27] <froztbyte> have there been any major changes to the topology code from 2.5.x to the newer versions?

[29-Nov-2011 16:32:27] <rmatte> froztbyte: can you document the steps that you take during your upgrade testing?

[29-Nov-2011 16:32:28] <nyeates> froztbyte: you will want to do 2.5.2 -> 3.0.3  -> 3.1.0 -> 3.2.1

[29-Nov-2011 16:32:37] <rmatte> froztbyte: and I'll do the same

[29-Nov-2011 16:32:46] <rmatte> yeh, I tried 2.5.2 to 3.2, doesn't work at all lol

[29-Nov-2011 16:32:47] <Hackman238> froztbyte: You'll have to go from 2.5.2 -> 3.2.1 -> 4.1.1

[29-Nov-2011 16:32:48] <froztbyte> rmatte: yeah, I'm planning to do so

[29-Nov-2011 16:32:49] <rmatte> complete disaster

[29-Nov-2011 16:33:04] <froztbyte> especially for any random caveats I run into

[29-Nov-2011 16:33:16] <froztbyte> nyeates: thanks for the heads up

[29-Nov-2011 16:33:18] <Hackman238> froztbyte: Skip all the 3.x upgrades, just go to 3.2.1 since its just a stepping stone.

[29-Nov-2011 16:33:19] <rmatte> I'll try the upgrade path that nyeates suggested

[29-Nov-2011 16:33:35] <froztbyte> Hackman238: nah, I tried that once before, it blew up

[29-Nov-2011 16:33:50] <nyeates> you can see it in latest release notes here:

[29-Nov-2011 16:33:50] <Hackman238> froztbyte: Yeah it'll probably do the same the otherway

[29-Nov-2011 16:34:20] <Hackman238> froztbyte: The real tricky problem is that those of us running long standing v2.5.2 instances have orphan objects, rel problems and other abnormalities

[29-Nov-2011 16:34:55] <Hackman238> froztbyte: Some are hard to work around. You'll need to comb the bugs out of the v2.5.2 insance (even if they dont cause v2.5.2 problems) little by little.

[29-Nov-2011 16:35:36] <nyeates> BEFORE upgrading from 2.5.2, install the PreUpgrade ZenPack

[29-Nov-2011 16:35:39] <Hackman238> froztbyte: Took us months here at rackspace.

[29-Nov-2011 16:35:56] <rmatte> I'm just wondering what the impact of having packs marked as py2.4 on install is going to have on Zenoss 3.x

[29-Nov-2011 16:36:23] <nyeates> rmatte: uninstall them first

[29-Nov-2011 16:36:25] <nyeates> then upgrade

[29-Nov-2011 16:36:31] <Hackman238> rmatte: You'll want to create copies wih the new name and install them.

[29-Nov-2011 16:36:50] <rmatte> nyeates: the problem with uninstalling is that several packs have device classes, so bye bye goes the pack and then bye bye go the devices

[29-Nov-2011 16:37:00] <Hackman238> rmatte: Good case.

[29-Nov-2011 16:37:32] <nyeates> rmate: yes....this is true :-( hopefully there is a py 2.6 version?

[29-Nov-2011 16:37:52] <Hackman238> rmatte: In that case what you'll want to do is make a duplicate pack upgraded for v3.x or v4.x and install them after the rpm -U and zenmigrate

[29-Nov-2011 16:37:58] <rmatte> I'm pretty sure you could get away with just installing the py2.6 labeled pack over top of the old one

[29-Nov-2011 16:38:17] <nyeates> and if not, and its crucially important....gotta dev the zenpack to update it to work on new zenoss, or stay on old zenoss

[29-Nov-2011 16:38:24] <rmatte> Hackman238: have you done your upgrades from 2.5 to 3.x yet?

[29-Nov-2011 16:38:35] <froztbyte> Hackman238: I've actually heard lots of people mention rel issues, but I don't think I've ever run into one

[29-Nov-2011 16:38:41] <froztbyte> which is probably just my luck

[29-Nov-2011 16:38:47] <nyeates> rmatte: im sure u are all too aware of this

[29-Nov-2011 16:38:49] <Hackman238> We have 3 DC's upgraded from v2.5.2 to v4.1.1

[29-Nov-2011 16:38:55] <rmatte> nyeates: I don't think I have any ZenPacks that wouldn't work on the new version at this point

[29-Nov-2011 16:39:00] <froztbyte> nyeates: yus, I'm well aware of the pre-upgrade zenpack

[29-Nov-2011 16:39:19] <froztbyte> there's another question I have, although I don't know how many people sit in the same situation

[29-Nov-2011 16:39:27] <rmatte> Hackman238: did you happen to make notes as to what steps you followed?

[29-Nov-2011 16:39:51] <froztbyte> I have a lot of remote sites that are >=250ms away, sometimes the lowest they get is 900ms close

[29-Nov-2011 16:39:53] <rmatte> Hackman238: I'm trying to gather as much upgrade intel as possible before I attempt it myself

[29-Nov-2011 16:40:08] <froztbyte> and I'm wondering about setting up a remote collector on the edge of the site and collecting the hosts there from that single collector

[29-Nov-2011 16:40:20] <froztbyte> as opposed to my core collector fetching them all by itself

[29-Nov-2011 16:40:25] <rmatte> froztbyte: I doubt you'd have a problem doing that

[29-Nov-2011 16:40:27] <froztbyte> anyone been in that boat?

[29-Nov-2011 16:40:38] <Hackman238> rmatte: Yes, I have a  script that unfucks v2.5.2 *pretty* well.

[29-Nov-2011 16:40:54] <rmatte> Hackman238: would you mind firing it over at some point?

[29-Nov-2011 16:40:55] <Hackman238> rmatte: But it'll still likely take a couple of tries

[29-Nov-2011 16:41:11] <Hackman238> rmatte: Are you ent now?

[29-Nov-2011 16:41:21] <rmatte> oh well, I need to get my boxes upgraded one way or another so I'll do what it takes

[29-Nov-2011 16:41:28] <rmatte> Hackman238: no, just core

[29-Nov-2011 16:41:28] <froztbyte> oh also, that's the other thing

[29-Nov-2011 16:41:34] <Hackman238> rmatte: Oh gotcha. Sure man

[29-Nov-2011 16:41:37] <froztbyte> I'm not running enterprise anywhere

[29-Nov-2011 16:41:42] <rmatte> I can strip any ent specific stuff out of the script if it has any

[29-Nov-2011 16:41:50] * froztbyte is Zenoss Support for the company

[29-Nov-2011 16:42:04] <rmatte> I'm Zenoss Support/everything else man

[29-Nov-2011 16:42:10] <rmatte> lol

[29-Nov-2011 16:42:22] <Hackman238> LOL

[29-Nov-2011 16:42:28] <froztbyte> rmatte: but concerns were mostly around how latency-and-shitty-link-sensitive the collector stuff is

[29-Nov-2011 16:42:33] <froztbyte> but I guess I'll try it in a dev lab

[29-Nov-2011 16:42:43] <rmatte> Hackman238: also, any ideas in regards to the snag I've hit with my latest project?:

[29-Nov-2011 16:43:03] <froztbyte> although actually I lie in that previous remark...I essentially run all the monitoring that matters in our network, and any issues always get escalated to me

[29-Nov-2011 16:43:13] <rmatte> froztbyte: I'm pretty sure it caches the data before sending it, so it should work fine in theory

[29-Nov-2011 16:43:24] <froztbyte> (which does get annoying when you have cacti and similar shit in your life as well)

[29-Nov-2011 16:43:34] <rmatte> ewwww, cacti

[29-Nov-2011 16:43:51] <froztbyte> the really crappy part is that it's still the best for what it does

[29-Nov-2011 16:43:57] <rmatte> lol

[29-Nov-2011 16:44:01] <froztbyte> unless ganglia has SNMP support I don't know about

[29-Nov-2011 16:44:04] <froztbyte> mrtg is just fail

[29-Nov-2011 16:44:04] <Hackman238> To be honest you'll definately want a second box you can dedicate to the migration. On this box install zenoss v2.5.2, start, stop, extract a tar of the zenoss to migrate (excluding perf), start, install preupgrade, run, run zencatalog, stop, rpm -U v3, start, zenmigrate, reindex

[29-Nov-2011 16:44:35] <rmatte> that's fine, I can just clone the active boxes

[29-Nov-2011 16:44:40] <rmatte> then copy over RRD data when done

[29-Nov-2011 16:45:24] <rmatte> (I love VMs)

[29-Nov-2011 16:45:25] <rmatte> lol

[29-Nov-2011 16:45:30] <froztbyte> hehe

[29-Nov-2011 16:45:37] <froztbyte> sigh, I really need to write a cacti replacement

[29-Nov-2011 16:45:38] <Hackman238> rmatte: Huh. Thats a tricky problem.

[29-Nov-2011 16:45:45] <rmatte> yeh

[29-Nov-2011 16:45:50] <Hackman238> rmatte: Thats a weird way to handle oids too. LOL

[29-Nov-2011 16:45:59] <rmatte> that's micro$oft for you

[29-Nov-2011 16:46:19] <froztbyte> that's not particularly strange

[29-Nov-2011 16:46:25] <froztbyte> well, let me quantify that statement

[29-Nov-2011 16:46:28] <rmatte> but yeh, can you think of anything that could be done with a transform or something to get around this problem?

[29-Nov-2011 16:46:29] <Hackman238> Let me think it over.

[29-Nov-2011 16:46:30] <froztbyte> I've seen other vendors do worse

[29-Nov-2011 16:46:39] <rmatte> or is a custom datasource/collector inevitable?

[29-Nov-2011 16:46:48] <Hackman238> I have to bail.

[29-Nov-2011 16:46:51] <rmatte> k

[29-Nov-2011 16:46:54] <rmatte> ttyl

[29-Nov-2011 16:47:02] <Hackman238> rmatte: you might get away with a custom parser and datasource

[29-Nov-2011 16:47:09] <rmatte> and yeh, I've seen some vendors with horrible horrible snmp implementation

[29-Nov-2011 16:47:22] <Hackman238> I'll mull it over on my grueling drive home

[29-Nov-2011 16:47:25] <rmatte> such as displaying a temperature with a character after it so that it's a string instead of an integer

[29-Nov-2011 16:47:44] <rmatte> or naming every interface the same in ifName and not setting anything in ifAlias

[29-Nov-2011 16:47:56] <rmatte> Hackman238: thanks

[29-Nov-2011 16:48:27] <rmatte> Hackman238: wouldn't I need to use a command datasource for a parser?

[29-Nov-2011 16:48:36] <rmatte> or can you actually use a parser with zenperfsnmp?

[29-Nov-2011 16:50:26] <rmatte> guess he's gone lol

[29-Nov-2011 16:50:45] <rmatte> alrighty, back to setting thresholds for IPSLA...

[29-Nov-2011 16:51:26] <rmatte> jmp242: thanks for the RTFM comment on that thread

[29-Nov-2011 16:52:49] <froztbyte> nyeates: so I've been keeping an eye on my logs for the last few minutes now, looks like the only "error" thing I have is the xmpp zenpack which performs a bit of a hack on the method it uses to do notifications

[29-Nov-2011 16:53:08] <froztbyte> so I guess that's also not much of a worry for me

[29-Nov-2011 16:54:06] <rmatte> yup

[29-Nov-2011 16:54:28] <rmatte> I have 15 Zenoss servers to upgrade from 2.5.2 to 3.x

[29-Nov-2011 16:54:32] <rmatte> it shall be fun

[29-Nov-2011 16:54:45] <froztbyte> yours are uniform-ish, though?

[29-Nov-2011 16:54:54] <rmatte> fairly, yes

[29-Nov-2011 16:54:59] <froztbyte> lucky

[29-Nov-2011 16:55:04] <rmatte> hehe

[29-Nov-2011 16:55:16] <froztbyte> I have 4 (down from 6) reasonably distinctly differently tweaked installations

[29-Nov-2011 16:55:36] <froztbyte> and probably one more coming up soon, I guess

[29-Nov-2011 16:55:43] * froztbyte should start training some people on this stuff

[29-Nov-2011 16:55:46] <rmatte> well, the configuration is tweaked on each one to suit the client

[29-Nov-2011 16:55:53] <rmatte> but other than that, fairly similar

[29-Nov-2011 16:55:58] <rmatte> some have ZenPacks that others don't have

[29-Nov-2011 16:56:12] <froztbyte> yar...on mine they all typically get the same zenpacks

[29-Nov-2011 16:56:20] <froztbyte> but it's different hardware and hacks in each network

[29-Nov-2011 16:56:34] <froztbyte> so each has a very different signature

[29-Nov-2011 16:56:37] <rmatte> ah, mine are all VMs which makes it easy to deal with

[29-Nov-2011 16:56:53] <froztbyte> that's also the other thing...we almost exclusively deal with network devices, not servers

[29-Nov-2011 16:57:02] <rmatte> lucky you

[29-Nov-2011 16:57:04] <froztbyte> thus all those nice vendor hacks I deal with

[29-Nov-2011 16:57:05] <rmatte> (seriously)

[29-Nov-2011 16:57:12] <froztbyte> rmatte: bwahahahahahano.

[29-Nov-2011 16:57:19] <nyeates> sorry back

[29-Nov-2011 16:57:23] <rmatte> lol

[29-Nov-2011 16:57:34] <rmatte> Server monitoring has it's own little quirks

[29-Nov-2011 16:57:40] <froztbyte> rmatte: servers are easier to deal with, you get to fix the software yourself if necessary

[29-Nov-2011 16:57:56] <rmatte> you can't fix a shitty microsoft SNMP agent

[29-Nov-2011 16:58:02] <rmatte> you just have to overcome

[29-Nov-2011 16:58:05] <rmatte> lol

[29-Nov-2011 16:58:09] <froztbyte> vendor-provided things are basically just something that you get to go "so I need to get information out of this thing.....somehow"

[29-Nov-2011 16:58:19] <rmatte> true

[29-Nov-2011 16:58:27] <rmatte> I deal with that kind of stuff too

[29-Nov-2011 16:58:33] <froztbyte> there was this one transform I put together some time ago..

[29-Nov-2011 16:58:35] * froztbyte finds it

[29-Nov-2011 16:58:39] <rmatte> our sales guy always manages to find customers with equipment that I've never even heard of

[29-Nov-2011 16:59:01] <rmatte> the worst snmp implementation I've ever seen in my life was on a device called a firebox

[29-Nov-2011 16:59:09] <nyeates> froztbyte: which logs did you inspect in particular?

[29-Nov-2011 16:59:18] <rmatte> It wouldn't allow polling from an IP address outside of it's subnet

[29-Nov-2011 16:59:24] <froztbyte> tail -f /usr/local/zenoss/zenoss/log/*.log

[29-Nov-2011 16:59:34] <froztbyte> and then the two others that don't end in the right extension

[29-Nov-2011 17:00:03] <froztbyte> rmatte: 1x edge-of-subnet NAT coming right up!

[29-Nov-2011 17:00:22] <rmatte> bleh, so lame

[29-Nov-2011 17:00:33] <rmatte> lol

[29-Nov-2011 17:00:39] <nyeates> some active logs are in the collector in $ZENHOME/log/<collector name>  (often $ZENHOME/log/localhost )

[29-Nov-2011 17:01:34] <nyeates> zenhub is in there i think

[29-Nov-2011 17:01:56] <froztbyte> rmatte:

[29-Nov-2011 17:02:09] <nyeates> zeneventlog.log

[29-Nov-2011 17:02:36] <froztbyte> rmatte: that's my "deal-with-a-goddamned-cisco-ASA" transform

[29-Nov-2011 17:02:44] <nyeates> zenstatus.log, zenmodeler.log

[29-Nov-2011 17:02:44] <froztbyte> those things are *spammy*

[29-Nov-2011 17:03:18] <froztbyte> also the zenoss syslog collector does remarkably well with ~3000 lines a minute (until mysql starts being a little bitch)

[29-Nov-2011 17:03:35] <froztbyte> box with 2GB RAM actually kept up with the things reasonably well until mysql exploded

[29-Nov-2011 17:03:58] <rmatte> interesting transform

[29-Nov-2011 17:04:26] <froztbyte> I had a further refined version of it that effectively turned into a trie-based filter

[29-Nov-2011 17:04:48] <froztbyte> but I'd never speed-optimized it, and I was throwing too much data at it

[29-Nov-2011 17:05:11] <froztbyte> this one runs pretty damn quickly and is very versatile as a "if this then that" sort of dropfilter

[29-Nov-2011 17:05:20] <rmatte> I use syslog-ng as a layer between devices and Zenoss

[29-Nov-2011 17:05:24] * froztbyte too

[29-Nov-2011 17:05:30] <nyeates> add that transform to our event transforms and examples wiki page:

[29-Nov-2011 17:05:40] <rmatte> yeh, please do

[29-Nov-2011 17:05:42] <froztbyte> nyeates: I'll do it tomorrow

[29-Nov-2011 17:05:58] <froztbyte> need to scrub it a bit more

[29-Nov-2011 17:06:10] <nyeates> i like the soulmate wording the best :-D

[29-Nov-2011 17:06:15] <froztbyte> and right now my mind is far too tired to do so effectively

[29-Nov-2011 17:06:20] <froztbyte> nyeates:

[29-Nov-2011 17:06:31] <froztbyte> my code is full of little jokes that will likely only ever amuse me

[29-Nov-2011 17:06:36] <froztbyte> or whoever one day replaces me

[29-Nov-2011 17:06:51] <froztbyte> (since I'm the only one who ever deals with it)

[29-Nov-2011 17:07:07] <rmatte> my code is full of "Feature will be implemented some day over the rainbow" type comments

[29-Nov-2011 17:07:28] <froztbyte> my ssh keys package has a "extra_extra_key_all_about_it" resource function which gets called in to prep a whole bunch of stuff that other things will use

[29-Nov-2011 17:07:43] <rmatte> haha

[29-Nov-2011 17:10:07] <froztbyte> my notifications framework has so many puns that form when different codepaths are followed, it's actually pretty hilarious to watch someone read the logs later and figure out the hidden sentences

[29-Nov-2011 17:10:24] <froztbyte> usually takes a few tries, because it needs enough activity to trigger

[29-Nov-2011 17:10:28] <froztbyte> easter eggs++

[29-Nov-2011 17:10:39] <rmatte>

[29-Nov-2011 18:02:49] kocolosk is now known as kocolosk|away

[29-Nov-2011 18:10:58] kocolosk|away is now known as kocolosk

[29-Nov-2011 21:33:17] <Corey> Is the 4.2 new architecture built on Java?

[29-Nov-2011 21:56:52] <qkit> Morning,

[29-Nov-2011 21:57:09] <qkit> wonder any one success running zenoss to monitor hpux 11.11 ?

[29-Nov-2011 21:57:28] <qkit> if say i using ssh to monitor my hpux , do i needed to install an agent in my hpux?

[29-Nov-2011 23:44:26] kocolosk is now known as kocolosk|away

[30-Nov-2011 00:00:01] [disconnected at Wed Nov 30 00:00:01 2011]

[30-Nov-2011 00:00:02] [connected at Wed Nov 30 00:00:02 2011]

[30-Nov-2011 00:00:12] [zenoss-logger (logger bot) has joined #zenoss]

[30-Nov-2011 09:17:29] kocolosk|away is now known as kocolosk

[30-Nov-2011 10:32:57] <mattt> hey all, i've got some devices that went down a few days ago, and were brought up this morning

[30-Nov-2011 10:33:13] <mattt> but the status within zenoss still appears to be down

[30-Nov-2011 10:34:11] <stevex> Are the devices monitored over snmp?

[30-Nov-2011 10:35:06] <mattt> yeah, and snmp is up

[30-Nov-2011 10:35:09] <mattt> i can snmpwalk from zenoss

[30-Nov-2011 10:35:17] <mattt> i've given it about 30-45 mins to allow it to clear

[30-Nov-2011 10:38:19] <mattt> stevex: i know it's auto-clearing some events, if i kill snmpd on a host i see the alert, then if i bring it back it auto-clears

[30-Nov-2011 10:40:11] <stevex> Yeah I am not too sure -- I am still learning the system myself.  I was going to have you try the snmpwalk which you already did.

[30-Nov-2011 10:40:39] <rmatte> mattt: I know what the issue is, it's a stupid "feature" that I've petitioned in the past to have removed...

[30-Nov-2011 10:41:01] <mattt> rmatte: is it due to the length it was down for?

[30-Nov-2011 10:41:16] <mattt> as it's down for longer, check intervals increase or something?

[30-Nov-2011 10:41:23] <rmatte> yeh... are you using a single Zenoss instance or multiple collectors?

[30-Nov-2011 10:41:30] <rmatte> no, that's not it, I'll explain...

[30-Nov-2011 10:41:35] <cluther> ugh, yes.. maximum ping failures.

[30-Nov-2011 10:41:39] <rmatte> yup

[30-Nov-2011 10:41:48] <mattt> single zenoss!

[30-Nov-2011 10:41:59] <rmatte> under the collector settings for each collector, there is a setting called "Maximum Ping Failures"

[30-Nov-2011 10:42:20] <rmatte> by default it's set to the equivalent of 1 day

[30-Nov-2011 10:42:42] <mattt> looking now

[30-Nov-2011 10:42:43] <rmatte> and what it does is, if Zenoss sees the device down for more than that set amount of time, it completely ignores the device until Zenoss is restarted or the device is remodeled

[30-Nov-2011 10:43:02] <mattt> oh jeez

[30-Nov-2011 10:43:10] <rmatte> it's the stupidest "feature" in Zenoss, when you find the setting, change it from 1440 (which I think is what it is by default) to 99999999999

[30-Nov-2011 10:43:11] <mattt> Maximum Ping Failures1440

[30-Nov-2011 10:43:28] <rmatte> then either restart Zenoss or remodel the devices

[30-Nov-2011 10:44:33] <mattt> rmatte: brilliant, doing now ... thanks for that ... scanned the forums, etc. but didn't see anything obvious

[30-Nov-2011 10:44:39] <rmatte> no problem

[30-Nov-2011 10:50:32] <whyzgeek> hi guys, does any body know how to get SNMP ENUM status change tracking in Zenoss. Like lets say two clusters change Status from Sync to out of Sync or degraded

[30-Nov-2011 10:51:18] <whyzgeek> I know I can just create a template and set some threshold on it which always triggers it and then use a tranform to catch status changes

[30-Nov-2011 10:51:33] <whyzgeek> I just want to see if there is any easier way

[30-Nov-2011 10:52:03] <rocket> whyzgeek: you could create a nagios like script that monitors that .. and throws an event on the error condition

[30-Nov-2011 10:53:13] <whyzgeek> rocket thanks

[30-Nov-2011 10:53:26] <whyzgeek> but I was hoping it would be neater

[30-Nov-2011 10:53:42] <whyzgeek> solution available

[30-Nov-2011 10:53:59] <rocket> well if its not numerical you need some sort of code to convert it to a number

[30-Nov-2011 10:54:16] <rocket> or you could use cron and the zensendevent utility to check that etc

[30-Nov-2011 10:54:47] <whyzgeek> yep basically it depends on some external script then

[30-Nov-2011 10:54:58] <whyzgeek> I could do a trap as well

[30-Nov-2011 10:54:59] <rocket> or it depends on a custom parser

[30-Nov-2011 10:55:20] <rocket> eg the parser converts the snmp data to 1 or 0 etc

[30-Nov-2011 10:56:26] <whyzgeek> yep probably parser is good idea

[30-Nov-2011 10:56:53] <whyzgeek> if the numbers are mapped to continous integers

[30-Nov-2011 10:57:03] <whyzgeek> probably the graphs of it will help as well

[30-Nov-2011 10:57:09] <rocket> yeah

[30-Nov-2011 10:57:10] <whyzgeek> to track all status changes

[30-Nov-2011 10:57:30] <whyzgeek> thanks rocket this makes sense

[30-Nov-2011 10:57:57] <rocket> put it in a zenpack and you can keep track of the changes

[30-Nov-2011 10:58:10] <whyzgeek> yep agree

[30-Nov-2011 11:00:14] <rmatte> rocket: are parsers specific to zencommand or can they be made to work with any daemon?

[30-Nov-2011 11:04:53] kocolosk is now known as kocolosk|away

[30-Nov-2011 11:05:07] kocolosk|away is now known as kocolosk

[30-Nov-2011 11:49:01] <Gat0rvean> Is anyone very familiar with the "ZenSystems.Juniper" zenpack by Jane Curry?

[30-Nov-2011 12:00:01] [disconnected at Wed Nov 30 12:00:01 2011]

[30-Nov-2011 12:00:02] [connected at Wed Nov 30 12:00:02 2011]

[30-Nov-2011 12:00:12] [zenoss-logger (logger bot) has joined #zenoss]

[30-Nov-2011 12:00:13] <> [freenode-info] channel flooding and no channel staff around to help? Please check with freenode support:

[30-Nov-2011 12:33:28] <whyzgeek> anybody knows why zenoss daemons are all twisted select reactors rather than epoll?

[30-Nov-2011 12:40:50] <cluther> whyzgeek: Good question. I don't know why the select reactor was chosen for all Zenoss daemons, but I know that we did flip-flopped from epoll to select then back to epoll for the Windows daemons in the past.

[30-Nov-2011 12:40:51] <rmatte> whyzgeek: Because twisted rocks?

[30-Nov-2011 12:40:52] <cluther> whyzgeek:

[30-Nov-2011 12:41:03] <rmatte> lol

[30-Nov-2011 12:42:11] <whyzgeek> ic good to know!

[30-Nov-2011 12:42:20] <cluther> whyzgeek: My best guess would be that it's a relic of the past when we tried a lot harder to be portable.

[30-Nov-2011 12:42:25] <whyzgeek> I was about to do a patch to change it!

[30-Nov-2011 12:42:37] <whyzgeek> But currently I don't use zenwin anyway

[30-Nov-2011 12:43:38] <whyzgeek> I am not sure if this is related or not

[30-Nov-2011 12:43:48] <whyzgeek> but there is MAXSNMPREQUEST in zenperfsnmp

[30-Nov-2011 12:44:00] <whyzgeek> when I increase that beyond default 20

[30-Nov-2011 12:44:28] <whyzgeek> it hangs the daemon and I see it gets to the 1000 limit of pselect

[30-Nov-2011 12:45:09] <whyzgeek> so it end up having some sort of FD leak

[30-Nov-2011 12:45:41] <whyzgeek> however these child snmp-gets are running under another process

[30-Nov-2011 12:45:51] <whyzgeek> so I am not quite sure if this applies or not

[30-Nov-2011 12:45:55] <rmatte> why are you increasing that value in the first place?

[30-Nov-2011 12:46:08] <rmatte> are you polling a gazillion datapoints?

[30-Nov-2011 12:46:15] <whyzgeek> because I want to pack more devices per collector

[30-Nov-2011 12:46:20] <whyzgeek> yep

[30-Nov-2011 12:46:32] <rmatte> how many do you have per collector currently?

[30-Nov-2011 12:46:49] <whyzgeek> 1200

[30-Nov-2011 12:47:03] <rmatte> I'm guessing around 45k datapoints?

[30-Nov-2011 12:47:13] <whyzgeek> 122K

[30-Nov-2011 12:47:18] <rmatte> ah, quite a few

[30-Nov-2011 12:48:03] <rmatte> Have you tried increasing the zMaxOIDPerRequest value?

[30-Nov-2011 12:48:14] <whyzgeek> yep that's already 40

[30-Nov-2011 12:48:22] <whyzgeek> I think I can't go beyond that

[30-Nov-2011 12:48:24] <rmatte> 40 is the default, but can it go higher I wonder

[30-Nov-2011 12:48:36] <whyzgeek> well that's make a 64K ud[

[30-Nov-2011 12:48:39] <whyzgeek> udp packet

[30-Nov-2011 12:48:42] <whyzgeek> some times

[30-Nov-2011 12:48:44] <cluther> whyzgeek: That's strange. I've bumped MAX_SNMP_REQUESTS up to 100 without hitting that problem.

[30-Nov-2011 12:49:16] <whyzgeek> may be it is per device, so on average how many OIDs have you got per device

[30-Nov-2011 12:49:19] <whyzgeek> ?

[30-Nov-2011 12:49:41] <cluther> No, it's total for the daemon. Number of in-flight requests.

[30-Nov-2011 12:49:59] <rmatte> yeh, the daemons just fires them out in whatever order and waits for them to return

[30-Nov-2011 12:50:06] <rmatte> doesn't care what devices they belong to

[30-Nov-2011 12:50:25] <whyzgeek> yep all requests gets chained together

[30-Nov-2011 12:50:32] <whyzgeek> per device

[30-Nov-2011 12:50:49] <whyzgeek> so I am thinking if you have 100 devices like this

[30-Nov-2011 12:50:58] <whyzgeek> each launch their own chain

[30-Nov-2011 12:51:15] <whyzgeek> then that certainly goes beyond the 1000 limit

[30-Nov-2011 12:53:38] <cluther> hmm.. those chains are sequential though as far as I can tell.

[30-Nov-2011 12:54:49] <whyzgeek> hmm, you are right. because it is deferred object I was thinking that they might right in parallel

[30-Nov-2011 12:55:04] <whyzgeek> but it seems that they are wired to run serial

[30-Nov-2011 12:58:15] <whyzgeek> I think most of the time is wasted just because of some slow devices waiting to timeout

[30-Nov-2011 12:58:54] <whyzgeek> and they are the ones clogging the request pool

[30-Nov-2011 13:03:56] <whyzgeek> one more questions why zenperfsnmp takes time to collect full devices? I see the first Cycle is 0sec waiting for 5min then it only takes around half of the devices in the second cycle and in the third cycle it goes for whole bunch.

[30-Nov-2011 13:04:19] <whyzgeek> I mean when you start the daemon

[30-Nov-2011 13:14:45] <rmatte> whyzgeek: it collects the configuration of the devices, as in what it needs to poll

[30-Nov-2011 13:14:48] <rmatte> if that's what you mean

[30-Nov-2011 13:15:28] <whyzgeek> yep but why not all in one go?

[30-Nov-2011 13:16:16] <rmatte> no idea

[30-Nov-2011 13:17:03] <jmp242> cluther might know, but he's lurking

[30-Nov-2011 13:17:54] <cluther> whyzgeek: It's related to the DEVICE_LOAD_CHUNK_SIZE global in

[30-Nov-2011 13:18:40] <cluther> We ran into problems on very large collectors where it would take zenperfsnmp way too long to startup and start doing anything. So we implemented incremental device configuration loading.

[30-Nov-2011 13:18:43] <whyzgeek> yes exactly

[30-Nov-2011 13:19:10] <whyzgeek> ic but how that logic works and where is it?

[30-Nov-2011 13:19:26] <whyzgeek> because this way it creating gaps in graphs for me which is not good

[30-Nov-2011 13:19:27] <cluther> You could increase that DEVICE_LOAD_CHUNK_SIZE to some value greater than your number of devices and just wait for the full config before the first cycle.

[30-Nov-2011 13:20:07] <cluther> It's in the fetchDevices inner method of the updateDeviceList method in

[30-Nov-2011 13:21:15] <whyzgeek> I see, thanks cluther that makes sense now

[30-Nov-2011 13:21:35] <whyzgeek> I wonder why I didn't notice it before!

[30-Nov-2011 13:28:27] <rmatte> cluther: we need to find some way to hook up your brain to a computer and download all your Zenoss knowledge to an appendix.

[30-Nov-2011 13:28:58] <cluther> I'd pay for that. My arms/hands/fingers connection is suboptimal.

[30-Nov-2011 13:29:07] <whyzgeek> lol

[30-Nov-2011 13:29:08] <cluther> Too much impedance I think.

[30-Nov-2011 13:29:15] <rmatte> indeed

[30-Nov-2011 13:31:04] <cluther> I'm doing a webinar with David Nalley from Citrix on monitoring CloudStack with Zenoss in about 30 minutes if anyone is interested.

[30-Nov-2011 13:31:08] <cluther>

[30-Nov-2011 13:31:32] <rmatte> sure, I'll take a look

[30-Nov-2011 13:33:02] <whyzgeek> I had to override this failed method in Chain to just ignore the device that times out at any time

[30-Nov-2011 13:33:12] <whyzgeek> now I can collect 132K in 100sec

[30-Nov-2011 13:34:11] <whyzgeek> failed/failure

[30-Nov-2011 13:34:45] <rmatte> nice

[30-Nov-2011 15:00:39] <otakup0pe> hey so i have a device complaining about an OSProcess being missing but this OSPRocess isn't listed in the OSPRocess tab

[30-Nov-2011 15:00:46] <otakup0pe> please tell me the solution isn't to delete/re-add the device

[30-Nov-2011 15:01:01] <otakup0pe> i clear the event and it comes right back.

[30-Nov-2011 15:02:05] <rmatte> otakup0pe: restart zenprocess and if that doesn't work, run zendmd as the zenoss user and do reindex() then commit()

[30-Nov-2011 15:02:38] <otakup0pe> hmm reindex is different thatn remodel ?

[30-Nov-2011 15:02:45] <rmatte> completely different

[30-Nov-2011 15:02:52] <otakup0pe> ok i'll try thx

[30-Nov-2011 15:03:16] <rmatte> zendmd is a command line command to be clear

[30-Nov-2011 15:03:24] <otakup0pe> yeah i've used that before

[30-Nov-2011 15:03:29] <rmatte> k

[30-Nov-2011 15:03:36] <rocket> whyzgeek btw the zenperfsnmp daemon is being totally rewritten for 4.1.1 to address some of the issues with slow devices

[30-Nov-2011 15:03:36] <otakup0pe> hey while i have your attention do you know of a way to progrmatically get a couynt of how many graphs are in zenoss ?

[30-Nov-2011 15:03:44] <otakup0pe> i tried using zendmd before but got confused

[30-Nov-2011 15:04:09] <rmatte> not off the top of my head, I'd have to mess around with zendmd to get that and I don't have the time at the moment

[30-Nov-2011 15:04:16] <otakup0pe> no worries

[30-Nov-2011 15:04:25] <otakup0pe> i just kinda want a graph of how many graphs we have.

[30-Nov-2011 15:06:04] <cluther> otakup0pe: Total graphs, or total types of graphs?

[30-Nov-2011 15:06:10] <otakup0pe> total

[30-Nov-2011 15:07:11] <rocket> Yea we need a robochet interface ...

[30-Nov-2011 15:07:23] <rmatte> hehe

[30-Nov-2011 15:08:29] <cluther> This zendmd snippet should work:

[30-Nov-2011 15:08:54] <otakup0pe> ooh

[30-Nov-2011 15:14:50] <otakup0pe> works ! now to get at it progmatically

[30-Nov-2011 15:16:54] <rmatte> otakup0pe:

[30-Nov-2011 15:17:38] <otakup0pe> already on that page but thx yo <3

[30-Nov-2011 15:17:46] <otakup0pe> i've wanted this graph for months ^_^

[30-Nov-2011 15:17:53] <rmatte> lol

[30-Nov-2011 15:59:52] <tightwork> I recall a setting to expire data, anyone know where this is? Ideally I would like to automatically delete interfaces if they are down/down after sometime.

[30-Nov-2011 16:00:14] <tightwork> This is a hypervisor so interfaces come and go.

[30-Nov-2011 16:00:30] <nyeates> interfaces would be more than data

[30-Nov-2011 16:01:10] <mloven> It's all 1's and 0's Nick. 

[30-Nov-2011 16:01:13] <nyeates> there is built-in ability to remove old event data

[30-Nov-2011 16:01:19] <nyeates> heh true

[30-Nov-2011 16:03:30] <nyeates> I am not familiar with built-in settings or functionality that relate to expiring interfaces

[30-Nov-2011 16:03:51] <nyeates> Maybe it was a customization you saw? a dmd script?

[30-Nov-2011 16:04:41] <tightwork> no, it was a field in days

[30-Nov-2011 16:04:49] <tightwork> cant remember where the heck it was

[30-Nov-2011 16:04:49] <nyeates> hmm

[30-Nov-2011 16:05:10] <nyeates> dont forget to look under the collector settings

[30-Nov-2011 16:13:49] <coldboot> I have a custom HTTP service, and I'd like to have Zenoss monitor it's "average query time" every minute. Can someone point me to the documentation for this? I'm not sure what to call it, so it's hard to search for it.

[30-Nov-2011 16:14:20] <coldboot> By "average query time" I just mean the HTTP service I wrote will simply provide that number, it's arbitrary.

[30-Nov-2011 16:15:22] <nyeates> explain what you mean by HTTP service

[30-Nov-2011 16:16:03] <nyeates> like, some outside thing hits a URL, gets blah data back from it in return, and you want to know how long that transaction took?

[30-Nov-2011 16:16:08] <coldboot> It's some webserver running a web application.

[30-Nov-2011 16:16:18] <otakup0pe> you want to graph the value returned by the web service ?

[30-Nov-2011 16:16:22] <coldboot> nyeates: I want it to graph the value, yeah.

[30-Nov-2011 16:16:26] <coldboot> otakup0pe: Yep

[30-Nov-2011 16:16:33] <nyeates> Use httpmonitor

[30-Nov-2011 16:16:38] <nyeates> comes built in with zenoss

[30-Nov-2011 16:16:51] <otakup0pe> nyeates: does that let you use the return value ?

[30-Nov-2011 16:16:58] <coldboot> Is it in this documentation?

[30-Nov-2011 16:17:30] <nyeates> oh wait im rereaeding above what u want

[30-Nov-2011 16:17:33] <otakup0pe> yeah i don't think httpmonitor will work if you want to *graph* it

[30-Nov-2011 16:17:39] <otakup0pe> if you just want to throw an alert it will

[30-Nov-2011 16:17:44] <nyeates> so the web service gives some value back and u want to read that and then graph it?

[30-Nov-2011 16:17:49] <coldboot> Yeah you can't graph it.

[30-Nov-2011 16:18:01] <nyeates> or u wnat to know how long the transaction took to get the data back?

[30-Nov-2011 16:18:03] <otakup0pe> if you want to graph, custom zenoss/nagios plugin

[30-Nov-2011 16:18:06] <otakup0pe> not tooooo hard

[30-Nov-2011 16:18:10] <coldboot> you can get 2 datapoints: - seconds taken to perform the test, - size of page loaded

[30-Nov-2011 16:18:37] <coldboot> nyeates: Yeah, I want to read the value and graph it. Not response time.

[30-Nov-2011 16:19:03] <otakup0pe> does the 'make a custom nagios plugin' process exist documented somewhere ?

[30-Nov-2011 16:19:20] <coldboot> Deprecated:

[30-Nov-2011 16:19:30] <coldboot> At the top of Google rankings for "zenoss plugin"

[30-Nov-2011 16:19:55] <coldboot> Is this sensible?

[30-Nov-2011 16:21:15] <otakup0pe> oh cool

[30-Nov-2011 16:21:25] <otakup0pe> nyeates: yeah look at that second url

[30-Nov-2011 16:21:31] <otakup0pe> you'll want to make your own plugin probably tho

[30-Nov-2011 16:21:39] <otakup0pe> which can maybe even be bash + curl

[30-Nov-2011 16:23:22] <nyeates> yeah i think you will have to make something that parses the output from the url and puts that parsed data into nagios format so that zenoss can suck it in smoothly

[30-Nov-2011 16:23:41] <nyeates> im still looking around for a better solution

[30-Nov-2011 16:24:07] <coldboot> It would be cool if you could just click on something in Zenoss that says "Graph the number from this URL every $interval".

[30-Nov-2011 16:24:59] <nyeates> Ahh.... there is a feature for this, sorta.....and its for Enterprise, not Core

[30-Nov-2011 16:25:01] <nyeates>

[30-Nov-2011 16:25:11] <nyeates> data in XML can be prsed and brought in

[30-Nov-2011 16:28:01] <coldboot> But we have to pay for Enterprise to get that?

[30-Nov-2011 16:28:11] <otakup0pe> it's not hard to write a nagios plugin

[30-Nov-2011 16:28:25] <nyeates> yea....seeing now an old zenpack that does synthetic web transaction

[30-Nov-2011 16:28:50] <nyeates> its not been fixed to work on 3.* though and im not sure if it supports the same XML parsing stufff....but it may

[30-Nov-2011 16:29:56] <nyeates> Ive never tackled the write-your-own-plugin approach.... otakup0pe: what are the highe level steps?

[30-Nov-2011 16:38:50] <nyeates> coldboot: see this forum post:

[30-Nov-2011 16:58:14] <trnzmeta> hi guys: why is there a 300Gb min req for install?

[30-Nov-2011 16:58:23] <trnzmeta> does zenoss actually use all this?

[30-Nov-2011 17:00:15] <coldboot> nyeates: Thanks

[30-Nov-2011 17:20:54] <cluther> trnzmeta: Not even close, I just checked my usage and its under 1GB. That space is primarily for event storage.

[30-Nov-2011 17:28:02] <rmatte> most of my Zenoss VMs have between 10GB and 20GB filesystems

[30-Nov-2011 17:28:07] <rmatte> it really doesn't use much space

[30-Nov-2011 17:45:06] kocolosk is now known as kocolosk|away

[30-Nov-2011 17:57:07] <trnzmeta> thanks guys

[30-Nov-2011 18:08:15] kocolosk|away is now known as kocolosk

[30-Nov-2011 19:33:26] <sam___> hellp

[30-Nov-2011 19:33:29] <sam___> hello

[30-Nov-2011 19:33:33] <sam___> anyone can help?

[30-Nov-2011 19:33:47] <rmatte> don't ask to ask

[30-Nov-2011 19:34:06] <sam___> how do you install zenpacks on Centos?

[30-Nov-2011 19:35:10] <rmatte> Installing ZenPacks is explained in detail in the Zenoss Admin Guide:

[30-Nov-2011 19:35:59] <sam___> zenpack --install <filename> does not work

[30-Nov-2011 19:36:16] <rmatte> you have to do it as the zenoss user

[30-Nov-2011 19:36:18] <rmatte> sudo su - zenoss

[30-Nov-2011 19:38:09] <sam___> ok thanks

[01-Dec-2011 00:00:01] [disconnected at Thu Dec  1 00:00:01 2011]

Comments (0)