nagios check for AIX CPU
perl, ssh on nagios server
This script is executed remotely on a monitored system by the NRPE or check_by_ssh methods available in nagios.
If you hare using the check_by_ssh method, you will need a section in the services.cfg file on the nagios server that looks similar to the following. This assumes that you already have ssh key pairs configured.
define service {
use generic-service
hostgroup_name all_aix
service_description AIX CPU
check_command check_by_ssh!/usr/local/nagios/libexec/check_aix_cpu
}
Alternatively, if you are using the NRPE method, you should have a section similar to the following in the services.cfg file:
define service{
use generic-service
hostgroup_name all_aix
service_description AIX CPU
check_command check_nrpe!check_aix_cpu -t 30
notification_options c,r ; Send notifications about critical and recovery events
}
If you are using the NRPE method, you will also need a command definition similar to the following on each monitored host in the /usr/local/nagios/nrpe/nrpe.cfg file:
command[check_aix_cpu]=/usr/local/nagios/libexec/check_aix_cpu
Please note that the default thresholds for alerting are as follows:
WARN on iowait cpu utilization 50% or more
CRIT on iowait cpu utilization 80% or more
WARN on user cpu utilization 75% or more
CRIT on user cpu utilization 90% or more
WARN on sys cpu utilization 75% or more
CRIT on sys cpu utilization 90% or more
WARN on idle cpu utilization 10% or less
CRIT on idle cpu utilization 5% or less
WARN on entitled cpu utilization 500% or more
CRIT on entitled cpu utilization 800% or more