Check_nt CPULOAD returns data for seconds rather than minutes



I’m running NSClient++ and have run into an odd issue with check_nt CPULOAD. The first parameter after -l is supposed to be the timeframe in minutes. However, NSClient++ is interpreting that timeframe as being in seconds.

Here’s a shortened version of the command I’m using:

check_nt -v CPULOAD -l 5,80,90

That command returns the 5-second average, not 5-minute average, with this version of NSClient++. I’ve verified this using a load generator. Any thoughts? I’ve asked the Nagios XI team and they say this is an NSClient++ bug I guess I’d have to agree if that value is supposed to be in minutes, not seconds. I can’t specify “5m” or anything like that since that’s invalid syntax for -l for check_nt CPULOAD. I may have to convert all my hundreds of CPU monitors to NRPE if this isn’t considered a bug in NSClient++ and I’m hoping that’s not the answer. Thanks for any assistance you can provide.



Looks like this is a longstanding bug in NSClient++ that will not be fixed. I’m going to have to transition to NRPE.


This has been fixed in 0.5.1