Nagios Support Forum

Hi,

I am monitoring my sql parameters using check_nt plugin in Nagios.

I am using below command to monitor Sql server CPU utilization for one of my SQL server:
./check_nt -H <IP> -p 12489 -v COUNTER -l "\\Process(sqlservr)\\% Processor Time","CPU Time for sqlservr is %.f.0 %%","Processor Time" -w 80 -c 100

Output: CPU Time for sqlservr is 100.0 % | 'CPU Time for sqlservr is %.f.0 %%'=100.000000%;80.000000;100.000000;

I wonder the CPU utilization is giving 100% . But actually it is not taking that much utilization.

Any help/suggestions please.

Thanks & Regards
Kalyan

Well, that is what the perf counter is reporting. Have you checked the perf counter on the windows system for this process? What does it report for cpu time utilization?

Now I am just guessing here but...

Could it be that the counter in question is affected by cores?
In which case the caps is not 100%?
For such counters you need to switch to check_nrpe and use the nocap100 flag to get values above 100%.

This is a "feature" of the performance data counterd that many % counters are not in fact 0-100 instead 0 -100x<multiplier> which kind of sucks since the returned value is by default capped to 100 (unless you set the nocap100 flag).

// Michael Medin

mickem wrote:For such counters you need to switch to check_nrpe and use the nocap100 flag to get values above 100%.

I was not aware of this flag. The more you know . . .