Flapping question

Icinga Web Modules
,
1

Hello Icinga community,

It’s been a while since I’ve posted here.

I have a general question about flapping, which I haven’t found any reference to in previous threads.
Few services are constantly appearing with the flapping sign on the dashboard.
I’ve tried to find out the cause many times in the past with no avail.
Today a colleague has asked to change the check interval of a specific check to 10 minutes (used to be 1 minute). coincidently, this service is one of these services that have the flapping issue.
After the change, the flapping issue seem to have stopped.
Could it be that when the checks are so close together, whatever is checked is not yet ready for new connections, thus, causing the results to “flap”?

Best regards,
An IT guy

2

I don’t understand your question, hence, I can only direct you to the flapping documentation.

3

Hello @rsx ,
My question is - can flapping be caused by repeated checks in a short time interval?

When the check interval was 1 minute, flapping occurred non-stop.
Now that the interval is 10 mintutes, flapping had stopped.

Best regards,
IT guy.

4

Possible but unlikely.

From memory, flappig uses a formula involving a sliding windows of the last x checks and the number of changes in there. The goal is to not flood the admins with notifications while the service still has problems - for example while it gets fixed and nobody set a downtime.

Only an analysis of the service history in Icinga2 and the services logs and/or manual tests could tell you why this happens.