Cisco CUCM/CUBE monitoring via SNMP

[Unigarant]

Hi folks,

We want to monitor our telephony enviroment with Nagios. This is however proving to be quite hard since there is no "real" plugin/wizard for it and you have over a 1000 objects you can monitor. We received oid's from our supplier, I have imported these in NagiosXI thinking if I do an snmp walk I would get some meaningfull results. Sadly not

I have been in touch with our NagiosXI support. They asked me to put in /etc/snmp/snmp.conf the following:"mibs +ALL". But this file does not exist, we do have an snmpd.conf, I don't know if that is the correct file? And if so: where do I put it and is this the exact syntax? Please note: our Cisco support says we should not be using snmp traps.

In other monitoring software we can just add a callmanager, it does an snmp walk and it recognizes the version and even pre fills the "default" values for errors.

We would love to monitor our callmanager with Nagios, but manually adding every line we get balk from the snmp walk is just insane. There are more users requesting Cisco Unified Communications Manager monitoring, so I think this would definetly be beneficial for NagiosXI as well. Any help in getting this to work would be appreciated! We are willing to help build a wizard for this, just contact me via mail.

[jdalrymple]

We received oid's from our supplier

Have you compared those OIDs to the OIDs monitored by the existing CCM plugin on the Exchange? I'm personally not sure how much changed when they turned CCM into CUCM, but it is definitely worth looking at. If you want to post those OIDs here I'll be happy to compare them to that plugin for you.

[WillemDH]

Unigarant,

This pdf => https://assets.nagios.com/downloads/nag ... ios-XI.pdf has some guidelines for writing custom wizards.
We are also using Cisco Callmanager. I'll have a look for you what I'm monitoring and will update this thread.

EDIT:

DRV_Root => $USER1$/check_snmp_storage_wizard.pl -H $HOSTADDRESS$ $ARG1$
-C community --v2c -m "^/$" -w 80 -c 95 -f

SRV_CPU => $USER1$/check_snmp_load_wizard.pl -H $HOSTADDRESS$ $ARG1$
-C community --v2c -w 80 -c 90 -f

SRV_Memory => $USER1$/check_snmp_storage_wizard.pl -H $HOSTADDRESS$ $ARG1$
-C community --v2c -m 'Memory' -w 80 -c 90 -f

SNMP_Registered_Phones => $USER1$/check_snmp -H $HOSTADDRESS$ $ARG1$
-o .1.3.6.1.4.1.9.9.156.1.5.5.0 -C community -P 2c -l "Registered_Phones" -w 250 -c 300

I hope it gets you started. Let me know if you find any other useful info.

Grtz

Willem

[jolson]

Unigarant,

Let us know if you have any further questions after reading through the above posts. Thanks!

[Unigarant]

Thank you for your replies.

I am not a fulltime Nagios administrator, we are a small club that manages the entire IT infrastructure for Unigarant.
We are willing to put time and effort in getting this to work, but we do not have the time/knowledge to build an entire wizard on our own unfortunately. A template, or something simular would be great.

Also, the things we want to monitor are much more specific than memory/cpu/registered phones. Things like jitter, line drops, total lines in use on the sip trunk, etc. etc.
The OID's on Nagios Exchange do not contain these as far as I can tell. But even then: I am still not sure how to get them in NagiosXI.

The last example by WillemDH could be usefull, will try and see if we can do something with it .

I think getting a wizard for CUCM in Nagios with loads of troubleshooting/monitoring presets would benefit NagiosXI.

[lmiltchev]

The last example by WillemDH could be usefull, will try and see if we can do something with it .

Let us know how it went.

I think getting a wizard for CUCM in Nagios with loads of troubleshooting/monitoring presets would benefit NagiosXI.

I would recommend posting a feature request on our public bug tracker here:

http://tracker.nagios.com

We can also post an internal feature request if you want us to.

[WillemDH]

Just a small addition. Answering this thread reminded me of a small issue I had with the registered phones check. As the snmp check seems to add a 'c' at the end of the performance data, a wrapper script is needed in order to generate the correct graphs. I dug up an old thread about this and created the wrapper script today.
Here it is in case you need it:

Code: Select all

#!/bin/bash

Hostaddress=$1
Oid=$2
Community=$3
Version=$4
Item=$5
Warning=$6
Critical=$7
Output="$(/usr/local/nagios/libexec/check_snmp -H $Hostaddress -o $Oid -C $Community -P $Version -l $Item -w $Warning -c $Critical)"
CleanOutput="$(sed 's/c//' <<<$Output)"
echo $CleanOutput

You will need to create a new command with this:

$USER1$/snmp_wrapper.sh $HOSTADDRESS$ $ARG1$

And the service arg1:
.1.3.6.1.4.1.9.9.156.1.5.5.0 community 2c "Registered_Phones" 250 300

Grtz

Willem

[lmiltchev]

Thanks for the script, WillemDH!

[Unigarant]

I have put up a feature request for this: click.

[ssax]

Great, thank you for posting the feature request. Can we mark this as resolved and lock the topic or do you still have questions?