We like to generate a alert if a particular process on windows server is running longer than specific time period which indicate that process is hanging. Please advise if there is way to do this. Thank you.
Check out Mike Driskell's power shell template for longest process run time. You may be able to wrap a threshold around it to trigger when over a certain limit.
https://thwack.solarwinds.com/docs/DOC-166985
Thanks for the shout out. Yes that template was designed just for that. We had a process in which if it ran more than 30 minutes it had hung so we set monitoring around just that.
Alternatively you could create a Windows Performance Counter monitor for that processes elapsed time and configure a warning or critical threshold appropriately. This value is report the amount of time in seconds the process has been running.
Is there a template for Linux servers?
There are a couple of Nagios scripts that can be used with SAM's Nagios Script Monitor to monitor process run duration.
aLTeReGo wrote:Alternatively you could create a Windows Performance Counter monitor for that processes elapsed time and configure a warning or critical threshold appropriately. This value is report the amount of time in seconds the process has been running.
aLTeReGo wrote:
aLTeReGo: will this work for process instances? (E.g. transcodes or encodes - where the issue is only when the oldest instance of a process is running longer than a certain time.)
Also: if the process in question is not running, the monitor goes into "unknown" mode:
Node status is Up, One or more application monitors is in an Unknown state.
Is there a way to avoid it? E.g. "no problem is it's not running"?
