Nagios Support Forum

Hey Guys,

I am trying to monitor load on servers, but been having issues setting the threshold value.

Server Specs:
1 server: 12 cores
1 server: 4 cores

Based on the rule(y = c * p / 100) I have tried the below values as well, still getting warning, even when there is no load on server(1.03, 1.04, 1.05). For 12 cores
-a '-w 10.8,9.6,8.4 -c 12.0,10.8,9.6'
Output of command:
/usr/local/nagios/libexec/check_nrpe -H HOSTADDRESS -t 30 -c check_load -a '-w 10.8,9.6,8.4 -c 12.0,10.8,9.6'
WARNING - load average per CPU: 0.10, 0.09, 0.09|load1=0.096;0.150;0.300;0; load5=0.089;0.100;0.250;0; load15=0.088;0.050;0.200;0;

My goal is to get warning on 90% and critical on 100%.
OR WARNING when 11 cores are 100% used and CRITICAL when 12 cores are 100% used

Have read posts/documentation regarding setting nagios thresholds but still not getting the right solution.

Thanks in advance.

Can you share the command definition you have in your nrpe.cfg for check_load

Thanks

Hey Scott, just reading your reply enlightened me where I am wrong. I just changed the default command definition to use my custom command. It worked!

Thanks a ton!!!

ansaaft wrote:Hey Scott, just reading your reply enlightened me where I am wrong. I just changed the default command definition to use my custom command. It worked!

Thanks a ton!!!

Glad it is resolved!

Locking thread