cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Performance Warehouse Offline Alerting Workaround

jason_ostroski
Dynatrace Pro
Dynatrace Pro

Reaching out to see if others have run into this scenario and can offer any suggestions.

We've been utilizing the OOTB Performance Warehouse Offline alert, however, we have been frequently alerted (once or twice a night) on just 3-5 second incidents. This is waking up our on-call via paging and the performance warehouse is back online before he can get to his laptop.

Since this is an OOTB incident, there is no way to configure the timeframe and we would like to only get alerted if it is down for more than 1 minute. While there may be options for fixing this in BPPM (our alerting system), we would prefer to have a comparable alert in App Mon.

Has anyone else run into this scenario and have you been able to utilize other App Mon performance warehouse metrics to reliably alert on the performance warehouse offline?

Thanks,

Jason

2 REPLIES 2

peter_karlhuber
Dynatrace Pro
Dynatrace Pro

Hi Jason,

this incident is assembled in the server internally and not governed by normal incident logic. Whenever we discover that the pwh is disconnected we fire it without any delay.

Therefore you also can't create a custom incident based on the same logic.

However, you could change what the incident does when it fires instead of sending an email (in debug mode). Probably the most flexible solution would be to call an ActionPlugin of your own design that collects the incidents and only sends an email (or whatever) once it determines that the desired threshold has been reached.

Hope that helps, cheers,

Peter

Thanks for the idea Peter.

We decided it was best to send both the start and end of the incident to BPPM. If the length of incident is significant, we will get alerted from BPPM.