Page 1 of 1

Need to monitor each CPU core's

Posted: Fri Feb 19, 2021 2:31 pm
by manimurugesan
Hello Team,

if a server has multiple CPUs these need to monitoring individually. In the example it is CPU0, 1, 2,3 .

Could you please provide plugin ,i have checked below but i don't want to monitor using ncpa i have already installed Nsclient and nrpe .

https://support.nagios.com/kb/article/c ... s-773.html

Please suggest how to monitor if server has multiple cpu i want to monior individually using nrpe or nsclient .

Please provide step by step configuration

Re: Need to monitor each CPU core's

Posted: Mon Feb 22, 2021 11:29 am
by benjaminsmith
Hi,

There are fewer CPU plugins available for Windows than Linux on the Nagios Exchange. And while NSClient does not have checks for CPU usage, it does have CPU load checks. The documentation is available on the following page:

https://docs.nsclient.org/checks/

There is an option to filter on specific CPU cores. An example check:

Code: Select all

./check_nrpe -H 192.168.254.27 -c check_cpu -a 'warn=load>80' 'crit=load>90' "filter=core = 'core 5'"
Let me know if that works on your system. If so, the following guide will help you configure the service check-in the Core Config Manager.

Managing Plugins In Nagios XI

Re: Need to monitor each CPU core's

Posted: Mon Feb 22, 2021 3:23 pm
by manimurugesan
Hi

It's working but it's Simply showing "cpu load" is ok i am not getting any utilization value .

Please find the attached screen shot and let me know is there any way to get utilization value in status information .

Re: Need to monitor each CPU core's

Posted: Tue Feb 23, 2021 2:32 pm
by benjaminsmith
Hi @manimurugesan,

For this particular check, the performance data output is not being picked up by the performance data processor (NPCD).

Another option would be to use NCPA, as it will graph each logical core, however, the thresholds would apply globally to any core on the CPU.
cpu-performance-ncpa.png
Let me know if that would work for you. In most cases, averages or max thresholds are the most relevant metric to monitor since these values can fluctuate, this is normal for CPU activity.

Benjamin