NSClient++ Help (#1) - CheckCPU not returning performance data (#762) - Message List
I've got NSClient++ 0.3.8.76 installed on a Windows Server 2008 R2 box.
I need to get "% cpu used" and "% memory used" back to Nagios and RRD/PNP so I can get trending graphs.
Using Nagios Core 3.0.3, I do a couple of queries interactively:
#> ./check_nrpe -H 172.20.8.26 -c CheckMem -a MaxWarn=80% MaxCrit=90% type=physical ShowAll=long OK: physical memory: Total: 4G - Used: 1.37G (34%) - Free: 2.63G (66%)|'physical memory %'=34%;80;90; 'physical memory'=1.36G;3.;3.;0;4; #> #> ./check_nrpe -H 172.20.8.26 -c CheckCPU -a warn=80% crit=90% ShowAll=long OK CPU Load ok.
The memory check seems to be working fine, but the CPU check doesn't seem to return performance data. It's like it's not following the plug-in development guidelines, ie nothing after the "|".
Am I using the wrong parameters? Do I have to go to an external plugin to check this?
Thanks for any advice...Lyle
I actually noticed a but related to this when fixing some stuff the other day so it is on my todo list... I have never noticed it myself but this could be one of them "random" things...
SO let me get back to you, and feel free to file a ticket in the mean time...
Michael Medinmickem03/30/11 08:07:39 (2 years ago)
Thanks for the reply, Michael, and thanks for the development work you do.
I've filed a bug report. It's probably not the highest priority for most, but it's critical for me at this point.
As a temporary workaround, is there an external script you could recommend to return "% cpu time" idle or used?
I looked at using the check_nt transport with CPULOAD, but it looks like it only returns load, not % used.
...LyleLyle03/30/11 18:44:27 (2 years ago)
I'm still in need of collecting cpu performance data from a Windows client. Load or %Idle/%Used, either would work.
If I found a plugin, I could call it from nscp, and that would get me by until the native call gets fixed. But the only plugins I've found on Nagios Exchange are made for snmp; maybe they would work with nscp too, I don't know.
Maybe I could try installing Cygwin and using a shell script plugin, but that's a chunk of work too.
Thanks for any advice....LyleLyle04/05/11 23:57:28 (2 years ago)
A bit of information that may help with my original request:
I found the NRPE_NT_Perf package on Monitoring Exchange. Seemed like it had been around for a while, and was the top download for cpu performance on Windows (sans SNMP). From that, I pulled "cpuload_nrpe_nt.exe", the plugin for monitoring cpu load.
cpuload_nrpe_nt.exe reported correctly when I ran it interactively on my XP laptop. But I moved it to one of the Windows 2008 clients I'm trying to monitor, and it only reports 0% cpu usage, though I've loaded the system with tasks.
I checked this plugin on a totally unrelated Windows 2008 box (hardware vs the previous VM one) and it also reports 0% cpu usage.
I'm wondering if there's anything odd about Windows 2008 R2 that keeps this plugin and NSCP from reporting correctly.
Thanks....LyleLyle04/06/11 02:01:19 (2 years ago)
Sorry... I just saw your check. There is no "time" specified so thus it does nothing. I should improve the error message I guess...
Michael Medinmickem04/06/11 07:03:34 (2 years ago)