CPU Usage alarm exception

rgivens

I have an alert configured to notify me if the CPU load on production machines is over 80% for a set period of time.  Works as expected and is a fairly standard alert.  I have several machines that are chronically over 100% so I setup a custom property so that this specific alert ignores those machines.  I have one machine though that runs a nightly job that causes the cpu to spike which triggers the alarm.  I would like to ignore this particular alarm also.  What would be the best way to add an exception for this one specific server just for the timeframe the nightly job is running?  I don't want to totally ignore the CPU load on this machine.  Just ignore it for a few hours.  I guess I could add the custom property that I've assigned to some the chronically high machines and then create a separate alarm just for this one machine, but I was wondering if there is a better way that I'm not thinking of.

m-milligan

Since it's just one machine, I suggest using the Unmanage Utility to unmanage the node during the nightly job. This will stop collection of all performance stats, though. If you simply want to mute alerts on that node during the nightly job, you can use cron or the Windows Task Scheduler to run one of the scripts linked below. You can either run the script to mute the alert for a specified amount of time, or you can run the script before and after the nightly job to mute and un-mute the node.

Unmanage/remanage or mute/resume an item or all items in a group

Unmanage or mute nodes Perl script for Linux

Unmanage/remanage/mute/unmute a node or group - Windows program in C#