CPU Critical 100 alert sent as Warning?

[MonitorGuy]

Had a strange one this morning:

Using ncpa to monitor Windows servers

Check interval: 6 min
Retry interval: 1 min
Max retries: 4

cpu/percent -w '97' -c '100' -q 'aggregate=avg'

Alert notification:

Service: CPU Load \State: WARNING
Info: WARNING: Percent was 100.00 %

My guess? The monitor rechecked four times in four mins. so if the average calculated was less than "100" for the four checks (say 99.9%) the severity would be a Warning, not a Critical.

Unless someone else has a better explanation, I suggested changing the Critical threshold to 99%

check_ncpa.py, Version 1.2.4
Nagios XI 5.8.5

Thanks!

Craig

[ssax]

Correct, it's comparing with greater-than compared to greater-than-or-equal-to.

[MonitorGuy]

Thanks, will ask to change the Critical threshold at 99 instead of 100 to resolve.

Craig

[ssax]

Great, let us know when we're okay to lock this up and mark it as resolved.

[MonitorGuy]

I think we are good to go, Thanks!

[ssax]

If you want it to be greater-than-or-equal-to you could pass the thresholds like this:

Code: Select all

cpu/percent -w '~:97' -c '~:100' -q 'aggregate=avg'

Taken from here:

https://nagios-plugins.org/doc/guidelin ... HOLDFORMAT

Locking the ticket and marking as resolved.

Thank you!