Check load alarming incorrectly?

[LifeTag]

Hello everyone,

I am running Nagios Core 4.3.2
The box i want to monitor is CentOS 7.4

In the config file I am using the check load command:

Code: Select all

check_command                       check_nrpe_args!check_load! -r -w 1.2,1.10,1.05 -c 1.75,1.6,1.5

If I run this command on the box itself, it reports back ok:

Code: Select all

./check_load -r -w  1.2,1.10,1.05 -c 1.75,1.6,1.5
OK - load average per CPU: 0.46, 0.31, 0.29|load1=0.460;1.200;1.750;0; load5=0.312;1.100;1.500;0; load15=0.290;1.300;1.600;0;

But on the Nagios Core server, it shows this box as Alarming Critical:

Code: Select all

CRITICAL - load average per CPU: 0.28, 0.28, 0.28 

Any idea as to why this is happening?

[tgriep]

I feel that the check_load command that is defined in the NRPE config file has the arguments hard coded so when the check_nrpe is ran from the Nagios server, it is ignoring the arguments that it is passing to the agent.

Login to the Centos7.4 server and make sure the command is defined line the example below. (You may have to edit the path to match your server)

Code: Select all

command[check_load]=/usr/local/nagios/libexec/check_load $ARG1$

Save the change and restart the NRPE agent.
Then test the command again from the Nagios server and post if it worked or not.