NRPE check_cpu_stats unable to read, but other NRPE are fine

[encee]

Hello,

I recently installed nrpe on a few more servers and as far as I can tell theres no difference in the install process.

the nrpe check cpu fails with NRPE: Unable to read output but I can execute the command on the server as the nagios user with no issues.

Check memory/load other nrpe checks all pass successfully as well.

Does anyone have an idea as to why it wont run?

[encee]

To elaborate, I can run the command on the destination server for the cpu check fine. The check fails from the xi server.

[Box293]

Please read through this guide as it should address your issue:

https://support.nagios.com/kb/article.php?id=140

If this doesn't fix your problem, please let us know what you tried (including commands typed) and what the output/result is.

[encee]

Followed step - VII. NRPE: Unable To Read Output

is nrpe installed? yes

check path -

command[check_cpu_stats]=/usr/local/nagios/libexec/check_cpu_stats.sh $ARG1$

/usr/local/nagios/libexec $ ls -d $PWD/*
/usr/local/nagios/libexec/check_cpu_stats.sh

paths match. and sh file executes fine with proper return.

/bin/bash /usr/local/nagios/libexec/check_cpu_stats.sh -w 85 -c 95
CPU STATISTICS OK: user=0.00% system=0.50% iowait=0.00% idle=99.50% | user=0.0% system=0.0% iowait=0.0%;85;95 idle=99.50%

nrpe is running - tcp 0 0 0.0.0.0:5666 0.0.0.0:* LISTEN 12014/xinetd


Plugin does not require sudo privileges.

[rkennedy]

What are the permissions of /usr/local/nagios/libexec/check_cpu_stats.sh? ls -l /usr/local/nagios/libexec/check_cpu_stats.sh

[encee]

-rwxr-xr-x 1 root root 5580 May 6 19:50 /usr/local/nagios/libexec/check_cpu_stats.sh

This is the same as my other servers.

[tgriep]

When you run the check from the command line on the remote server, are you using /bin/bash to run the command?
If so, add that to the script or edit the command in the nrpe.cfg file and add that like below.

Code: Select all

command[check_cpu_stats]=/bin/bash /usr/local/nagios/libexec/check_cpu_stats.sh $ARG1$

What is the OS and version of the remote system you are trying to monitor?

[encee]

That fixed the issue.

After investigating further it looks like we are creating the nagios user ahead of time to ensure its uid/gid don't overlap on other servers, what should the shell be set to for this? Its currently using /bin/bash.

from /etc/passwd

old servers without precreating user - nagios1012:/home/nagios:
new servers with precreating user - nagios1013:/home/nagios:/bin/bash

Is there any documentation on the proper way to specify a uid/gid for nagios upon installing nrpe?

[tgriep]

The default shell is /bin/bash for the nagios user account.
Is there a reason you want to set the same uid/gid for the nagios account?

Here are the instructions for Installing NRPE. Take a look at them for any details.

Code: Select all

https://assets.nagios.com/downloads/nagiosxi/docs/How-To-Configure-NRPE-and-Install-From-Source-with-Nagios-XI.pdf

[encee]

As of this moment it is for consistency with our automation. At this moment I am not seeing anything in the manual source compile that would affect this, also this issue arose even after removing the user and the local files and reinstalling. Same problem with the output.