Nagios/checks/solaris zpools

Nav: Nagios/checks/solaris zpools

Add Service

this will add a new service check to all hosts in the "zpool" 'hostgroup'

add this to the local services folder on the Nagios server '/opt/nagios/etc/local/services'

#cat zfs_zpool.cfg 
define service {
        service_description     zfs_zpool
        display_name            ZFS pool Health
        check_command           check_nrpe!check_zpools
        use                     generic-service
        hostgroup_name          zfs_disk
}

Define the zfs hostgroup

this goes on the server in the local host groups folder '/opt/nagios/etc/global/hostgroups'

#cat zfs_disk.cfg
define hostgroup {
       hostgroup_name  zfs_disk
       alias           ZFS Disks
}

Define a host-template

cat zfs.cfg 
define host{
        name            zfs
        hostgroups      +zfs_disk
        register        0
}

Add check to host

define host {
        use             zfs,solaris,telehouse
        host_name       server01
        address         192.168.100.100
        parents
}

Check config

/opt/nagios/bin/nagios -v /opt/nagios/etc/nagios.cfg '

NRPE Command defanition

Add this to the client in the commands folder '/opt/nagios/nrpe/nrpe-commands"

#cat zfs.cfg
command[check_zpools]=/opt/nagios/plugins/libexec/check_zpools

add the check script to plugins folder

'/opt/nagios/plugins/libexec/'

then make it exacutable

chmod u+x /opt/nagios/plugins/libexec/check_zpools

and a quick `chown -R nagios:nagios /opt/nagios` never did any harm

then restart nrpe '/etc/init.d/nrpe restart'

Check script [check_zpools]

#!/bin/sh

#
# By Aldo Fabi 01-09-2006
# LTP: 04-10-2006
#
# Nagios solaris 10 zpools Checker
#

### Begin vars
exit_code=0
exit_text=""
### End vars

POOLS=`/usr/sbin/zpool list -Ho name`
if [ $1 ]
then
	THRESHOLD=$1
else
	THRESHOLD=90
fi

errors_checker(){

        if [ "${exit_text}" ]
        then
		if [ $2 -gt ${THRESHOLD} -o $3 != "ONLINE" ]
		then
			exit_code=2
                	exit_text=${exit_text}" NOK:$1:$2%:$3"
		else
			exit_text=${exit_text}" OK:$1:$2%:$3"
		fi
        else
		if [ $2 -gt ${THRESHOLD} -o $3 != "ONLINE" ]
		then
			exit_code=2
			exit_text="NOK:$1:$2%:$3"
		else
                	exit_text="OK:$1:$2%:$3"
		fi
        fi
}

for POOL in `echo ${POOLS}`
do
	CAPACITY=`/usr/sbin/zpool list -Ho capacity ${POOL}|cut -d"%" -f1`
	HEALTH=`/usr/sbin/zpool list -Ho health ${POOL}`
	errors_checker ${POOL} ${CAPACITY} ${HEALTH}
done

echo ${exit_text}
exit ${exit_code}