- observium - lists.observium.org

ping alerts
by Joshua Elson 13 May '14

13 May '14

Is there a way to disable alerts for ping when snmpd isn't accessible? We get lots of false alerts when servers are under high load. We basically just want to be notified if a host is actually down (not pinging) and not when snmpd chokes out. Thanks! Josh Elson | Linux Systems Administrator MaxLinear | www.maxlinear.com<http://www.maxlinear.com/> 2051 Palomar Airport Road, Suite 100 | Carlsbad, CA 92011 [Description: Description: http://wiki/download/attachments/1704274/Maxlinear+Logo.png?version=1&modif…] ________________________________ This electronic mail (including any attachments) may contain information that is privileged, confidential, and/or otherwise protected from disclosure to anyone other than its intended recipient(s). Any dissemination or use of this electronic mail or its contents (including any attachments) by persons other than the intended recipient(s) is strictly prohibited. If you have received this message in error, please notify us immediately by reply e-mail so that we may correct our internal records. Please then delete the original message (including any attachments) in its entirety.

3 3

Disk i/o statistics
by Fabrice Charlier 12 May '14

12 May '14

Hi all, We are currently moving our observium server from an old community edition (0.13.10.4585) to a new powerful server running Professional edition (0.14.4.5391). On the new one, all disk i/o statistics for server devices don't appear in the web interface. After some research, I found that rrd files for disks i/o are not generated. For example, same server monitored by old instance: http://pastebin.com/gF7RA9bU and on new one: http://pastebin.com/Y9ePqyst Maybe I'm missing some new configuration option? Thanks, Fabrice

4 7

Add device failing
by Robert Williams 12 May '14

12 May '14

Hi, Just tried to add a new device (r5478) but it’s failing with no feedback, even at the command line. root@OBSERVIUM:/opt/observium # php add_device.php server.name.domain Observium v0.14.5.5478 Add Device(s) Try to add server.name.domain: Trying v2c community xyz ... root@OBSERVIUM:/opt/observium# It was ok a couple of days ago and I’m not sure how to try to diagnose this further as there is no feedback either at the command line or the web interface (which just says “Trying v2c community xyx …” the same as the command line does). To rule out a duff device I removed an existing working device and tried to re-add it. Fails with the same error also and I’m unable to re-add that device. Any pointers please? Cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

3 4

Calix E7-2 port disabling
by Crozier, Doug 11 May '14

11 May '14

I have 14 Calix E7-2 OLTs in this environment that I want to monitor. The first two I added to Observium(PE 5447) showed up with over 1000 ports each. I'd like to disable polling/alerting on all but a handful of ports on each of these devices. I used the "All" button on the device's Ports tab to get this done, but it only disables the first 333 ports. Same behavior on the second OLT. The remaining 900+ ports are all still enabled for polling. Manually selecting the slider for 50 or so of the remaining ports to disable them fails to take effect after clicking "Save". Is there a way to disable the remaining ports, then enable the few that I am interested in? Thanks Doug

3 3

dpkg agent oddness
by Robert Williams 10 May '14

10 May '14

Hi All, I’ve just added the dpkg agent script for the agent to a server, it already had other agents on it which are all working OK (apache, ntp etc.). Now, when I ./dpkg locally it gives the package list just fine. However, Observium shows the package list with some extra items at the top which read <blank> and “Fatal error:” [cid:image001.png@01CF6C41.93B4FDE0] Now on every poll (manual or scheduled) I also get the event log entry “System Arch -> Directive” [cid:image002.png@01CF6C41.F2135120] This dpkg script is working fine on all the other hosts I run it on, and everything else is working fine on this particular host apart from this element of the dpkg script. If I run the agent module in debug then I notice that at the end of the script I get: Fatal error: Directive 'allow_call_time_pass_reference' is no longer available in PHP in Unknown on line 0 Fatal error: Directive 'allow_call_time_pass_reference' is no longer available in PHP in Unknown on line 0 ) So I guess this is what is being interpreted by the poller as a ‘package’ and the ‘architecture’ of the system. However, I do not get this error when I run it locally, only when it’s run from Observium remotely. So - Any ideas or suggestions what to try next? Target host is Ubuntu 14.04 LTS (3.13.0-24-generic x86_64) and I have several others running 14.04 which are polling fine as far as I’m aware. Cheers! Robert Williams Custodian Data Centre Email: Robert(a)CustodianDC.com http://www.CustodianDC.com

1 1

Relative times not showing properly...
by Michael Loftis 09 May '14

09 May '14

I'm actually not certain when this started to happen as I've noticed it before, but it seemed like it was just in old events, but now, I've noticed that all relative times are getting computed incorrectly, they'll all show like "364 days, 17h 13m 3s" when you hover for an actual time (system event log particularly) the timestamps are right...it's something going wonk doing date math I'm guessing... 0.14.5.5458 - php 5.4.4 (just svn updated to make sure it wasn't fixed) not sure what else is relevant... -- "Genius might be described as a supreme capacity for getting its possessors into trouble of all kinds." -- Samuel Butler

3 4

weekly/monthly reporting tool?
by Joshua Elson 09 May '14

09 May '14

Hi all, I was wondering if there is any sort of reporting tool for Observium? We are looking to get a weekly and/or monthly break down of CPU and memory for all 100 or so of our servers. It has become tedious to go through each server one by one each time we need to gather all these details. As always, thanks in advance! Josh ________________________________ This electronic mail (including any attachments) may contain information that is privileged, confidential, and/or otherwise protected from disclosure to anyone other than its intended recipient(s). Any dissemination or use of this electronic mail or its contents (including any attachments) by persons other than the intended recipient(s) is strictly prohibited. If you have received this message in error, please notify us immediately by reply e-mail so that we may correct our internal records. Please then delete the original message (including any attachments) in its entirety.

3 4

Ruckus ZoneDirector
by Eduardo Schoedler 09 May '14

09 May '14

Hi, Do you plan to provide the discovery of the Ruckus ZoneDirector? Thanks. Regards, -- Eduardo Schoedler

3 3

Alerts only with alert checker
by Eduardo Schoedler 09 May '14

09 May '14

Hi, We are using Observium Pro, congratulations, it's a great tool! But I'm having trouble with alerts and need your help. I have created an alert check for bgp peers, and is ok. But Observium still want send alerts from himself :). How to disable it? Thanks in advance. Regards, -- Eduardo Schoedler

3 4

Add disk status to Buffalo Terastation
by Stefano Cordio 09 May '14

09 May '14

Hey guys, I'd like to implement an addition for my NAS box, a Buffalo Terastation. This NAS sends a daily report about the available space and the state of the disks. Observium already grabs details about the space, so I'd like to also add support to disk status/diagnostics (I already have the needed MIB files). The following is a MIB snippet: ------------------------------------------------------ nasDiskStatus OBJECT-TYPE SYNTAX INTEGER { notSupport(-1), normal(1), array1(2), array2(3), standby(4), degrade(5), remove(6), standbyRemoved(7), degradeRemoved(8), removeRemoved(9) } MAX-ACCESS read-only STATUS current DESCRIPTION "The state of local disks. This value is read from 'diskx' line in file /etc/melco/diskinfo. notSupport(-1) indicates the disk slot is not exist. normal(1) indicates the disk is not part of RAID array. array1(2) indicates the disk is part of array1. array2(3) indicates the disk is part of array2. standby(4) indicates the disk is standby and power-off. degrade(5) indicates the disk has degraded and removed from array, and power-off. remove(6) indicates the disk has removed from array by user and power-off. standbyRemoved(7) indicates the disk is standby and removed from NAS. degradeRemoved(8) indicates the disk has degraded and removed from array and NAS. removeRemoved(9) indicates the disk has removed from array and NAS." ::= { nasDiskEntry 2 } ------------------------------------------------------ I'm not sure which is the best location in the device page for showing these informations. Could we consider these as a kind of sensor or is it something inventory-related? Actually I need a design advice. Would be really nice if you could also give me a pointer to a similar existing code. I strongly believe in learning by example :-) Thanks, Stefano

2 2