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

Incident evaluation of 72 hours?

tarjei
Organizer

Hi,

We have a number of Matlab servers with host agents, that has the following SLA criteria:

If the server has a CPU of 100% for 72hours, this means that the matlab has hung, and needs to be fixed.

Anything inside of this is OK.

Any suggestions of how to do this?

Btw: I can use Windows Perfmon Monitor plugin if that makes it easier

1 REPLY 1

JamesKitson
Dynatrace Leader
Dynatrace Leader

Unfortunately there is not a way to get this behavior that I am aware of. The reason that the incident evaluation window is limited to 1 hour is this is the amount of time that time-series data (measures) are cached on the server itself and all custom incidents are in some way based off of measures.

There might be some technical reasons why it is implemented this way but feel free to post an RFE/idea in the AppMon and UEM Product Ideas forums to see if there is a demand for this ability. I know I have had some cases where a longer evaluation window is desirable.

James