Cisco CallManager MGCP PRI usage

[nduda78]

This is my first cacti template/script to share. We run a VOIP callcenter using multiple MGCP configed routers with PRI circuits. We needed to have a way to monitor PRI usage/capacity to plan for corrective action and new PRI circuits. The following scripts/template monitor each routers PRI capacity seperatly and in total.

This requires NSCLIENT as it runs on a Linux cacti install obtaining its info from a Windows CCM.

Obviously youll need to customize the script to get the counter info you need. You can add on to the script and datasource/templates for more than 2 routers.

[Brettw]

Any Idea what the oid would be for the number of calls on a dual E1 card?

Ive tried this the details here --> http://www.eatworms.org.uk/cacti/isdncalls.php

However no joy

Thanks

Brett

[techguy]

Any thoughts on how to get this to work from a windows machine running Cacti? This would be a great asset to me.

[crackrocksteady]

I was thinking about doing this with Win32::PerfLib as a perl script to pull this info out perfmon. Has anybody attempted this? any tips?


Fletch

[crackrocksteady]

Ok,

I had no luck with win32:perflib, but I did have success using Win32::Perfmon perl module.

The module is not available through the activestate repo's, so to compile the module you need to use virtual C++ 6 , which I couldn't find a copy of unfortunately. Luckily I was able to find a compiled copy on a third party repo, you can install Win32:Perfmon through ppm with the following:

ppm install http://www.bribes.org/perl/ppm/Win32-PerfMon.ppd

I have taken the original template and modified it slightly, allowing the script to be a little more dynamic and the labels a bit more generic. The call manager to get the data from is parsed through the hostname variable and the gateways are per-data source values that you define. It is currently setup for 2 gateways, but you could modify it too support more.


To Install
1) install Win32-Perfmon
2) place mgcp.pl in the scripts directory
3) Import the data template
4) Import the Graph template.

Note: I am not a programmer and this is my first contribution to the cacti community. If you spot anything stupid that I've done please let me know....

Thanks

Fletch

[crackrocksteady]

Here are the files.

[evanover]

Could you give a detailed explanation of the total PRI script elements? I am trying to modify to suit our environment, and I am having trouble try to figure out the the counter info.

[crackrocksteady]

Can you elaborate on what you mean? I don't quite understand.

[evanover]

In the script there is the path to check_nt (I.E. pria=`/usr/local/nagios/libexec/check_nt -H ).
Then the ip address of the Call Manager (10.X.X.X -p 1248 -v)
Then the counter this is where it is not clear what the counter should be ( COUNTER -l "\\Cisco MGCP Gateways(BCXXXX)\\PRIChannelsActive"`.

[crackrocksteady]

ok,

I don't use that check_nt script, have a look at my perl script for a comparison of another way to do it.

I am guessing that the name in the brackets is the name of your mgcp gateway. This is the name that your CCM gives it in perfmon. To find out what yours are called connect perfmon to your CCM and add the counter and see what they are called.

If your Call Manager is running on linux your not going to be able to pull this information out easily.

[evanover]

I

[evanover]

I

[evanover]

Thanks I will play with it and see what I can come up with. Unfortunately the only Win32-PerrfMon and Win32-Perflib I can find are for the Windoz platform. I am running cacti on a Linux box while CCM is on a Windoz box.

[crackrocksteady]

ahhh k, I didn't think about the fact that win32-perlib needs to be compiled using Visual C.

It means you will need to use check_nt. I think this is an offshoot nagios thing and you might need to run an agent on your ccm.

[poezie]

Hi crackrocksteady

Does your mgcp.pl work on linux as I get some errors ?