Nagios Support Forum

Hello,

I am having an issue where I am trying to have CPU Load alerts only alert when the last 60minute average is over 80 percent (warning) and 90 percent (critical). For some reason, the command setting doesn't appear to be respecting using the last 60 minute average. (command, screenshots below).

Command - $USER1$/check_nt -H $HOSTADDRESS$ -p 12489 -v $ARG1$ $ARG2$
ARG1 - CPULOAD
ARG2 -l 60,80,90

Screenshot 1 - Real-time Graph from VMware (showing high usage, but not for 60minutes)
Screenshot 2 - Graph from Nagios XI (I realize that when it gets to Nagios XI, the average is already calculated, so this spike will naturally be smaller)
Any insight is appreciated!

Regards,
Mike

Depending on your version of NSClient this was a long standing bug that it was using seconds, instead of minutes

https://forums.nsclient.org/t/check-nt- ... nutes/4523

thanks for that suggestion! I will try and put the items in their "seconds" equivalents and see if that resolves it and confirms the bug we may be seeing.

MikeMAN987 wrote:thanks for that suggestion! I will try and put the items in their "seconds" equivalents and see if that resolves it and confirms the bug we may be seeing.

Sounds good

Scott, thank you for helping validate this. We are making arrangements to temporarily change all of our checks to their second(s) equivalent and re-write using NRPE. Feel free to close!