Question on window's checks

[jenstar13]

Sorry, but I'm Linux background but because of the improvements to Nagios, our company has started monitoring the windows servers.
Apparently the cpu management and the memory management leaves something to be desired

that means our CPU. memory, and swap checks max out at 100% and send alerts, then they go away
is there any docs on better checks for windows machines, or a way to have it do an average of say one hour at a time?
I thought of setting the checks to every 60 minutes, but if it catches the 100% at the check time, then it goes red then
Same for alert notifications.
Is there anything you can suggest?

Thank you in advance

[lmiltchev]

If you are using NSClient++ agent to monitor Windows machines, you could pass a minute range to the CPULOAD check.

From the check_nt plugin's usage:

CPULOAD =
Average CPU load on last x minutes.
Request a -l parameter with the following syntax:
-l <minutes range>,<warning threshold>,<critical threshold>.
<minute range> should be less than 24*60.

Example:

Code: Select all

/usr/local/nagios/libexec/check_nt -H x.x.x.x -p 12489 -s 'password' -v CPULOAD -l 1380,80,90
CPU Load 19% (1380 min average) |   '1380 min avg Load'=19%;80;90;0;100

I would recommend that you play with the check settings to see what is going to work the best in your environment. For example, you could increase the retry_interval and the max_check_attempts values to allow the check to take longer time before determining if this is an actual problem.

[jenstar13]

your sales team sold us on NCPA, we use it for everything we can

The load check is
$USER1$/check_ncpa.py -H $HOSTADDRESS$ $ARG1$
-t 'radial' -P 5693 -M cpu/percent -w 100 -c 100 -q 'aggregate=avg'

swap is
$USER1$/check_ncpa.py -H $HOSTADDRESS$ $ARG1$
-t 'radial' -P 5693 -M memory/swap -u Gi -w 85 -c 90

Can those options be used for NCPA?

[lmiltchev]

Currently, there is no way to monitor CPU usage over period of time via NCPA. There is a feature request for adding this functionality here:

https://github.com/NagiosEnterprises/ncpa/issues/180

Having said that, if you had a custom, e.g. PowerShell script that could do that, you could easily call it via check_ncpa.py. You may be able to obtain this info from a Windows Performance counter too.

Again, I would recommend that you increase the retry_interval and the max_check_attempts values to allow the check to take longer time before determining if this is an actual problem or just a temporary peak.

[jenstar13]

wow 3 years, for the feature to be sitting out there?
Can you add our account to the request, it seems that if your sales people are selling NCPA, and pushing that Nagios is good for Windows too, not to have this feature.

That being said, I will try your suggestion and you can close this ticket

thank you for your help!!