Escalation for flapping alerts

Support forum for Nagios Core, Nagios Plugins, NCPA, NRPE, NSCA, NDOUtils and more. Engage with the community of users including those using the open source solutions.
Locked
unixninja
Posts: 1
Joined: Thu May 29, 2014 3:08 pm

Escalation for flapping alerts

Post by unixninja »

Nagios Core 3.4.1

We are doing all of our notifications through escalations, but it looks like there is no way to configure an escalation to send flapping start/stop alerts... It would be nice if someone who receives notifications could know that a service/system is flapping so they understand why they are no longer receiving alerts. Is this not implemented by design, is it an oversight, is it a bug, or ??

Example errors when trying to add a flapping option to escalations (No flapping start/stop alerts are sent without adding the options either):

Error: Invalid escalation option 'f' in hostescalation definition.
Error: Could not add object property in file '/opt/nagios/etc/escalations.cfg' on line 22.

Error: Invalid escalation option 'f' in serviceescalation definition.
Error: Could not add object property in file '/opt/nagios/etc/escalations.cfg' on line 240.

Thank you,
sreinhardt
-fno-stack-protector
Posts: 4366
Joined: Mon Nov 19, 2012 12:10 pm

Re: Escalation for flapping alerts

Post by sreinhardt »

I would consider it more a design choice than anything. Let's look at a few possibilities here. Firstly, flapping notifications are not counted in the normal notification number, so it would be very difficult to include them in the escalation logic, additionally they would artificially bump your notification count if the were. Secondly, escalations are only counted when a service is repeatedly sending the same notification over a period of time, again notifications really don't fit this description as they by definition are a continual OK\Not OK state changes. Thirdly, flapping start and stop can happen both on OK states and Warn\Crit states, so we are kind of back to the artificial incrementing of notification number, and issues of where to place them in the mix.

With all that, if you wanted, you are welcome to post a feature request to either sourceforge or github. I can honestly say that I do not believe it will get included, as the logic changes that would be needed pose great risk to causing other issues with minimal reward. I'm sorry I don't have better news here, as I agree it would probably be very confusing to not have a recovery notification but also not have further warn\critical notifications.
Nagios-Plugins maintainer exclusively, unless you have other C language bugs with open-source nagios projects, then I am happy to help! Please pm or use other communication to alert me to issues as I no longer track the forum.
Locked