check_nt returns super big CPU usage value

[caqin]

Hi NagiosXI team,
We're using check_nt to monitor CPU usage, but in one server, sometimes, it returns super big value. Like below.
[1570626026] SERVICE ALERT: SERVER_NAME_XXXX;System: CPU Usage;CRITICAL;SOFT;1;CPU Load 18446744044341600887% (5 min average)

CPU load is 18446744044341600887%, just wondering what caused this and how to avoid it? Is it RRD overflow?
Nagios XI version : 5.5.8
NSClient++ version : 0.5.0.62

Thanks

[scottwilkerson]

Can you post the actual command that Nagios is using when making this call to nsclient?

[caqin]

check_nt -H hostname -s "password" -p 12489 -v CPULOAD -l 5,80,90

This huge CPU usage returning value is not always happening. It happens for example, once or twice a week. Very concerning how it happens.

[scottwilkerson]

I've not run into this ever with CPULOAD, I'm not sure if it's a bug with the version or your specific server.

We are not the authors of NSClient however and it would have to be a bug in that, I would suggest reaching ot to the author, they may have some suggestions or things you can try to resolve the issue.

Their forum is here
https://forums.nsclient.org/

[tgriep]

Please upgrade to the latest 0.5.2.xx version of NSClient as there is a but that was fixed in newer versions that could effect that test you are running.
From the NSClient++ changelog

Code: Select all

2017-04-06 Michael Medin
 * check_nt CPU load using invalid time (second instead of minute) Fixes #251 and #403

[caqin]

Thanks guys!

[scottwilkerson]

Thanks guys!

Let us know if the upgrade resolves your issue.

Thanks!