Nagios Support Forum

Has anyone here attempted (or succeeded) in integrating Nagios and Netcool?

Specifically, I want Nagios to send certain events to Netcool (via SNMP)
and have Netcool ACK the event and send back to Nagios.

The real world reason I want this:

Netcool is a part of our infrastructure and there are people watching
the Netcool board 24/7 (NOC). Nagios is looking at other issues
and we wanted Nagios to send a subset of these issues
to Netcool for escalation and notification purposes.

So, I checked out this site: http://www.robinharwani.net/tag/nagios- ... tegration/

But it only documents half of the conversation: Nagios->Netcool
but not the reverse.

Any help is appreciated.

The link does look thin on explanation/examples. Have tried contacting the author of the post?

That post is very old. I did email the author but I am not expecting a reply. The blog itself
hasn't been updated in a few years.

-G

I found the following post in the old archive. He uses nsca on the netcool device to send back commands to the nagios server by way of a dummy host/service. I don't know if this will help, but it is something.

I wish the blog author from your first post got around to making part 2, but alas, even the Mona Lisa is falling apart.

EDIT forgot the link:
https://www.mail-archive.com/nagios-use ... 28699.html

If all else fails, I'll solve this the old fashioned way:
coffee and some long shifts in front of the computer.

When I start talking code in my sleep, I know I'm on the
right track.

BTW, that really does happen with me

So, let's leave this open and I'll update as I make
progress.

Cheers!

Great, I am replying to get it off our dashboard. Be kind and don't reply until you have more actionable information

Well, I have made great progress on this.

There are two more issues left.
Issue 1:
I noticed that the "--serviceeventid=" parameter and the
"--servicestateid=" parameter that is passed to my script,
from nagios, look like unique numbers.

I could combine these two numbers say: to send netcool a unique number along with the alert.

This way when they ACK back to me
(via a DB row update which netcool would do)
I can trace the ACK back to the issue.

Does this sound like a good approach?
Does the contatenation of serviceeventid and servicestateid
guarantee uniqueness?

Issue 2:
From the command line, how would I ACK a nagios event
and possible add a comment such as:

"This event has been ACked from Netcool"

We will have to look into the exact command, but to ack via cli you would want to submit to the nagios.cmd pipe most likely or script a web connection to the local host. As for your values not being passed, you need to add a $ at the end as well, otherwise core will not expand them. Finally, I cannot guarantee that they will make a unique ID, however it is very very likely.

I was only using that statement as a logic example.
I get the variables just fine.

Well, I was using this code (from bash) and I definitely got
a unique ID but I did not want to go overboard. Looks like
I was doing the right thing after all. And I can pass that number to netcool. This part works like a charm.

Ok, issue 1 is solved then.

Next, is the issue where I want to Acknowledge the event from the command line.
I assume that I have to give Nagios the serviceeventid and/or the serviceproblemid
that corresponds to this alert somehow.

This is that part that I'm missing.

One more thing: How do I add a comment (from the command line) to a specific alert?

Thanks in advance on this and previous post as it is rather critical.

GKH