CheckMK 2.0.0p9 - set all sub checks to CRIT

Hi all!

Is it somehow possible to set all sub-checks as CRIT when in a stale/not polled state when the Check_MK check is critical for more than 20 minutes? See screenshot for more details.

Why would you do that? You get a notification of the Check_MK service telling you there is something wrong. You then fix that and everything is fine. Why would you trigger X alerts when only one thing is wrong in the first place?

2 Likes

Beside the arguments from @thorian93 i would also say that a error message for all the services would be a technically false alert. It could be that only the check_mk service is failed and anything else is working as expected.

Actually, I just was asked by my colleagues to know if this is possible. Maybe, there is some data flowing into OpsGenie - I don’t know that.

I’m asked if this solution is possible and easy to make.

Regarding other checks - those, which are stale/not refreshed should be CRIT in any case.

Well - excuse my bluntness - no. If a service is stale that means by definition Checkmk cannot know whether the state has changed since it has received the last status update. So marking those services as CRIT would be - as @andreas-doehler already explained - a possible false alert.

May I suggest to inquire what outcome your colleagues expect and maybe explain to them, that the CRIT alert for the Check_MK service would suffice to indicate a problem on that host? Why would would they want 10-1000 alerts when they can be done with just one?

Please do not misread our tone, we do not intent to patronize you, let alone rant, we merely want to understand where your question comes from and try and direct you to a proper solution. :slight_smile:

1 Like

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed. Contact an admin if you think this should be re-opened.