Nagios Support Forum

Hello,

I am having an issue where I run check_cpu on my server and it takes longer than expected. I have thousands of other servers running this check and they return in the 60 second allotted but on my one server it takes upwards of 1m 20s and sometimes it runs just fine on that server. Any ideas would be greatly appreciated. I have timed it running the check locally on the machine so I know it has nothing to do with networks.

Hello, @zjposz. I assume you're using the NRPE agent on the remote servers? What does the load look like on the server when the CPU check is taking long time? You can check with the TOP command. If the server is overloaded it can become slow to respond.
What plugin are you using for the CPU check? What is the OS on that remote server?
Also, have you looked into /var/log/syslog or /var/log/messages on the remote server for anything out of the ordinary?

Yeah I am using NRPE agent. The load is usually quite low which is what is throwing me off. When I say I run the script locally I mean I run it one after the other and the load does not change in that time and they are coming back with huge time differences.

@zjposz, The system load on the server could be normal when you ran the plugin from the command line but not when NRPE was timing out. Can you clarify what plugin you're using?