Really weird issue with nrpe

Support forum for Nagios Core, Nagios Plugins, NCPA, NRPE, NSCA, NDOUtils and more. Engage with the community of users including those using the open source solutions.
Locked
jose_a_guardia
Posts: 13
Joined: Fri Dec 04, 2015 2:40 am

Really weird issue with nrpe

Post by jose_a_guardia »

Hi!

I've a lot of NRPE servers, and works fine. However, in one server I get this issue:

SERVER NRPE
NRPE v.2.13
I run this command to check glassfish process status:

Code: Select all

root@srvglassfish:~# /usr/lib/nagios/plugins/check_procs -a glassfish -c 1:3
PROCS OK: 1 process with args 'glassfish'
This is OK, I have 1 proccess 'glassfish'

In nrpe.cfg I set up this:

Code: Select all

command[check_procs_glassfish]=/usr/lib/nagios/plugins/check_procs -a glassfish -c 1:3


SERVER NAGIOS

Nagios 4.1.1

In the Nagios server, I run this check manually:

Code: Select all

/usr/lib/nagios/plugins/check_nrpe -H NrpeServerIP -c check_procs_glassfish
PROCS OK: 2 processes with args 'glassfish'
Two processes???!!


If I stop the only glassfish process, the machine itself detect 0 process, and Nagios Server 1 process.

Thanks in advance.

Regards.
User avatar
lmiltchev
Former Nagios Staff
Posts: 13587
Joined: Mon May 23, 2011 12:15 pm

Re: Really weird issue with nrpe

Post by lmiltchev »

Can you run the following commands on the remote box and show the output?

Code: Select all

ps -ef | grep glassfish
/usr/lib/nagios/plugins/check_procs -a glassfish -c 1:3
Run these commands twice - once when the glassfish service running, and again when it is stopped.
Be sure to check out our Knowledgebase for helpful articles and solutions!
jose_a_guardia
Posts: 13
Joined: Fri Dec 04, 2015 2:40 am

Re: Really weird issue with nrpe

Post by jose_a_guardia »

With Glassfish RUNNING:

Code: Select all

root@glassfish:~# ps -ef | grep glassfish
root     22554 22535  0 10:48 pts/0    00:00:00 grep glassfish
root     31983     1  2 Aug23 ?        01:56:21 /opt/jdk1.6.0_45/bin/java -cp /opt/glassfish3/glassfish/modules/glassfish.jar  -...

Code: Select all

root@glassfish:~# /usr/lib/nagios/plugins/check_procs -a glassfish -c 1:3
PROCS OK: 1 process with args 'glassfish'

And now, with Glassfish process STOPPED:

Code: Select all

root@glassfish:~# ps -ef | grep glassfish
root     22856 22535  0 10:51 pts/0    00:00:00 grep glassfish

Code: Select all

root@glassfish:~# /usr/lib/nagios/plugins/check_procs -a glassfish -c 1:3
PROCS CRITICAL: 0 processes with args 'glassfish'
In localhost all works fine.

Thanks!
User avatar
lmiltchev
Former Nagios Staff
Posts: 13587
Joined: Mon May 23, 2011 12:15 pm

Re: Really weird issue with nrpe

Post by lmiltchev »

I was not able to recreate the issue. :(

I will have to discuss the problem with our developers.

I defined my command on the client as:

Code: Select all

command[check_procs_noargs]=/usr/local/nagios/libexec/check_procs -a httpd -c 1:3
I am using check_procs v.2.0.3.

Code: Select all

[root@192 ~]# /usr/local/nagios/libexec/check_procs -V
check_procs v2.0.3 (nagios-plugins 2.0.3)
I have 21 httpd processes running at the moment:

Code: Select all

[root@192 ~]# ps -ef | grep httpd | grep -v grep | wc -l
21
[root@192 ~]#
Testing the check locally on the client:

Code: Select all

[root@192 ~]# /usr/local/nagios/libexec/check_procs -a httpd -c 1:3
PROCS CRITICAL: 21 processes with args 'httpd' | procs=21;;1:3;0;
Testing it on the nagios server:

Code: Select all

[root@localhost libexec]# ./check_nrpe -H <client ip> -c check_procs_noargs
PROCS CRITICAL: 21 processes with args 'httpd' | procs=21;;1:3;0;
Be sure to check out our Knowledgebase for helpful articles and solutions!
Locked