NCPA CPU Check

[AMacintosh]

I am curious as to how other Nagios NCPA users are using the CPU check?

We are currently getting alerted for a CPU issue on a Windows host that does not appear to have an overall CPU Usage issue. It looks like one of the CPU's or cores may be utilized..

CRITICAL: percent was 0%,0%,0%,100%,0%,0%,0%,0%

[lmiltchev]

What is the NCPA version that you are currently using? Can you show us the actual command that you are running from the command line, along with the output of it?

Example:

Code: Select all

[root@testbox libexec]# /usr/local/nagios/libexec/check_ncpa.py -H 192.168.x.x -t 'token' -P 5693 -M cpu/percent -w 20 -c 40
WARNING: percent was 27%,0%,28%,0%,18%,0%,0%,3% | 'percent_0'=27%;20;40; 'percent_1'=0%;20;40; 'percent_2'=28%;20;40; 'percent_3'=0%;20;40; 'percent_4'=18%;20;40; 'percent_5'=0%;20;40; 'percent_6'=0%;20;40; 'percent_7'=3%;20;40;

[AMacintosh]

We are using the following command which seems to alert if a single CPU or Core is over the threshold rather than reporting on the overall usage.

/usr/local/nagios/libexec/check_ncpa.py -H 0.0.0.0 -t '<token>' -P 5693 -M cpu/percent -w 85 -c 95

[lmiltchev]

You are correct. This is how CPU check is working at the moment. It's a known issue. You can check out this post:

https://github.com/NagiosEnterprises/ncpa/issues/56

[AMacintosh]

Has anyone implemented a fix so that it gets an aggregate instead of alerting if a single core meets the threshold? This is causing a lot of false alerts.

[lmiltchev]

anyone implemented a fix so that it gets an aggregate instead of alerting if a single core meets the threshold? This is causing a lot of false alerts.

I don't think so. I expect this to be implemented in the next version of NCPA, but I cannot be 100% sure - it is up to the developer. You can post a question for the developer here: https://github.com/NagiosEnterprises/ncpa/issues/56