check_nrpe chop perf data

[alokispandey]

Hi

I am facing an issue where check_nrpe o/p and script o/p differs. We are not getting full o/p of script while executing it under check_nrpe. And this is happening only with some client.
Check_mem performance data differs for some client even i am using same script.

eg:[ mind the return value of ServerType=]

Code: Select all

non-workinghost# check_nrpe -H localhost  -c Dcheck_mem -t 60 -a 90 95
MEMORY OK - total: 98944576 kb used: 13850692 kB cached: 1412608 kB buffers: 419148 kB free: 83262128 kB | total=101319245824;; used=14183108608;; cached=1446510592;; buffers=429207552;; free=85260419072;; [color=#FF4000]ServerType=;;[/color]

NRPE.cfg :

Code: Select all

non-workinghost#/etc/nagios/nrpe.d/DefaultNrpe_checks.cfg:command[Dcheck_mem]=/usr/lib64/nagios/plugins/check_mem -w $ARG1$ -c $ARG2$

Executing that script itself:

Code: Select all

non-workinghost#/usr/lib64/nagios/plugins/check_mem -w 90 -c 99
MEMORY OK - total: 98944576 kb used: 13849208 kB cached: 1407108 kB buffers: 419148 kB free: 83269112 kB | total=101319245824;; used=14181588992;; cached=1440878592;; buffers=429207552;; free=85267570688;;[color=#40BF40] ServerType=physical[/color]
;;

O/P of working host

Code: Select all

onworkkinghost]# /usr/local/nagios/libexec/check_nrpe -H onworkkinghost -c Dcheck_mem -t 60 -a 90 95
MEMORY OK - total: 99074388 kb used: 84080736 kB cached: 623412 kB buffers: 371832 kB free: 13998408 kB | total=101452173312;; used=86098673664;; cached=638373888;; buffers=380755968;; free=14334369792;; [color=#40BF80]ServerType=physical[/color]
;;

By above it seems there is some issue with check_nrpe, check_mem or nrpe itself. And in order to investigate it, i've executed below steps..

Nrpe version on both host

Code: Select all

[root@compute01 ~]# /usr/local/nagios/libexec/check_nrpe -H 169.144.104.6
NRPE v2.15
[root@compute01 ~]# /usr/local/nagios/libexec/check_nrpe -H localhost
NRPE v2.15

Check_nrpe version on both host

Code: Select all

Version: 2.12

Any difference in check_mem script on both host

Code: Select all

none

Any difference in kernel parameter, ulimit value for user

Code: Select all

none

Tested by reinstalled nrep on client

Code: Select all

 yes

Now i am running short on clue how to proceed next, hence this request...!

Suggest,..!

[rkennedy]

I'm having a hard time understanding what's missing, are you referring to this part?

Code: Select all

ServerType=physical;;

This doesn't look to be the proper performance data output. I don't believe it's going to be a value that you can graph, though. Can you show us a screenshot of the 'Service State Information' page for the service? I'd like to see the performance data that Nagios is seeing.

[alokispandey]

Thanks for your time, we managed to get it fixed. One of the dependent package "facter (puppet labs) was outdated".

[ischwartz]

I will go a head and close this