Nagios Support Forum

Hi, all.

I have a requirement from management for our Nagios (Core VM 3.5.0) installation to with notifications and escalations etc.

1. If there is a service problem, I need to notify a group of people every 1 minute, the "phase1" contact list.
2. If the problem has not been acknowledged within 15 minutes, I need to also send an alert to management, the "phase2" list.
3. If the problem has not been acknowledged within 30 minutes, I need to also send an alert to senior management, the "phase3" list.
4. If the problem has been acknowledged, only send an alert to phase1 every 30 minutes, and do not escalate to 2 and 3.

I think I can manage 1-3 using the normal service escalation configuration, but I can't see how to keep alerting once a problem has been acknowledged.

thanks.

Here's what I would do:

Create a second service that does the exact same check as the first service, but make it DEPENDENT upon the first service being in a CRITICAL state. So it only runs if the first one is CRITICAL. Then set up notifications on this second service to notify every 30 minutes to the phase 1 people if the second service check (which is the exact same check as the first one) is CRITICAL. This way, you have a second check that is separate from the acknowledged check. Of course, you have to teach people NOT to acknowledge the second check.

Dependency documentation can be found here: http://nagios.sourceforge.net/docs/3_0/ ... ncies.html

Hmm, that might just work

I'll just have to put something like "NO_ACK" in the name.

Or you could put something like "FIX THE @#$% PROBLEM!" in the title.

@cylindric, did that work for you? If you have no further questions I'd like to close this thread.

Just back into work - I'll test it this morning and report back.

Haven't heard from you in a bit. Did this work out?