XI integration to RemedyForce Incident Management

[djohansen]

Hi, very new to Nagios...

Been given direction to take ownership and move existing functionality to our upcoming Nagios XI/Core 4 implementation from our existing HPOM implementation. I am currently the HPOM SME.

A little background. We current have HP Operations Manager (HPOM) V9.11 with integration to RemedyForce (RF) ITSM via a very nice Perl script that opens a call via Web Services to RF and returns the incident number back to the incident event in HPOM. We are in the process of installing Nagios Core 4/Nagios XI in an eval POC.

My employer wants me to recreate this ticketing functionality in Nagios, but I am at a loss as WHERE and HOW to do this within Nagios/Nagios XI.. I see a lot of "you should be able to do that", but unfortunately not many concrete examples.
I figure this is either so blatantly obvious that I should see it, or maybe it's more involved than that.. I am going to repurpose my existing connector to act as the new method, but where do I begin in Nagios/XI??

Can/Should I add this to Core? I'd also like to be able to add this to the menu/tabs in XI to make it super easy for the next guy..

Thanks for any help...

-Dave

[mcapra]

An event handler is likely the best option:
https://assets.nagios.com/downloads/nag ... ios-XI.pdf

Basically, this event handler would be a modified version of your Perl script (or maybe not even modified) which accepts arguments provided by Nagios XI (what host, what service, status output, etc) when a problem occurs. Let us know if you have additional questions regarding event handlers!

[djohansen]

Many Thanks for the reply!

I just printed this very document out.. And I *sure* I'm going to have a bunch more questions!!

Thanks

-Dave

[tmcdonald]

If you have any further questions about this topic please let us know here, otherwise for unrelated questions please open a new forum thread.

[djohansen]

Thanks... I read the document and went over this with my coworkers and there was a concern about using the event handlers in the "event" that we need notification and event handling. Can you add additional event handlers to a service/host? or is it a single one to one mapping.. I proposed a dual function event handler 1) fixit (or not) routine, and then (2) execute a notification (or not) routine..

However they are pushing that we add an additional notification method to avoid using the event handler for other than it's "intended" purpose. Like I said I'm new to all this so I'm looking at all the possibilities.

Thanks
-Dave

[dwhitfield]

I don't understand the concern:

Event handlers are executed for HARD problem and recovery states immediately after notifications are sent out.

- https://assets.nagios.com/downloads/nag ... dlers.html

What exactly seems to be causing the confusion?

[djohansen]

Not sure about your question on my confusion.

It seems that putting the notification into the event handler would negate the ability to execute a reaction to an event if so desired (unless I write some sort of dual purpose handler).

From what is said about the event handler executing AFTER the notification is sent is sort of out of scope of what I'm trying to figure out. I get that it is a reaction to a state change.

What I want to do is link my alerts to tickets in our ITSM system instead of just straight email or SMS alerting. Doesn't seem to be a lot of documentation on this (Update: I found documentation on how to integrate a pager notification system to XI)

It has me create a command, and then tie that command to either a service/host in the notification area, so this is promising

-D

[mcapra]

Here's the full rundown on how Event handlers work:
https://assets.nagios.com/downloads/nag ... dlers.html

A brief summary that addresses some of your concerns:

... global host and service event handlers are executed immediately before host- or service-specific event handlers.

So you could have both a global event handler as well as individual service/host event handlers. However, you are correct that services/hosts can only have one local event handler defined. So if your event handler needs multiple steps or richer logic, that has to come from the event handler itself.

If you wanted just alerts to go to the ITSM system, you would drop in your custom notification handler in place of the stock one included with Nagios XI. If you wanted to leverage your custom handler in addition to the stock one included with Nagios XI, I think the best way to handle that would be to call the XI handler as part of your custom ITSM handler.

But there's nothing native to Nagios XI that could say if [condition1] run(event_handler1) else if [condition2] run(event_handler2) or anything like that. You would need to bake that into your event handler.

[djohansen]

Hey Thanks again for the VERY informative response.. This clears up a lot.

With Nagios it eems like there is always more than one way to bake that cake!

[Box293]

Excellent, glad you were able to understand the different cake recipes there are for Nagios

Is there anything else we can help with in relation to this or is it OK to close this thread?