I'm trying to get some third-party and my own custom checks running.
The current test setup consists of one icinga(web)2 CentOS 7 server and one CentOS 7 client with NRPE installed within the same network and firewalld has been disabled.
The check_service plugin I'm trying to run uses systemctl is-active.
Running the check on the client gives back the right result:
[root@localhost ~]# sudo -u nrpe /usr/lib64/nagios/plugins/check_service -s docker
OK: Service docker is running!
When ran from the Icinga2 server it results in a 'Service not running' message even though the service is running on the client:
[root@icinga2 ~]# /usr/lib64/nagios/plugins/check_nrpe -H 192.168.100.153 -c check_docker_service
CRITICAL: Service docker is not running!
All of the standard tests are running correctly when called from Icinga2.
Client:
/etc/sudoers
[...]
nrpe ALL=NOPASSWD:/bin/systemctl is-active docker,/usr/lib64/nagios/plugins/check_service
nagios ALL=NOPASSWD:/bin/systemctl is-active docker,/usr/lib64/nagios/plugins/check_service
[...]
/etc/nagios/nrpe.cfg
log_facility=daemon
debug=0
pid_file=/var/run/nrpe/nrpe.pid
server_port=5666
nrpe_user=nrpe
nrpe_group=nrpe
allowed_hosts=127.0.0.1,::1,192.168.100.151
dont_blame_nrpe=0
allow_bash_command_substitution=0
command_timeout=60
connection_timeout=300
[...some working default checks...]
command[check_docker_service]=/usr/lib64/nagios/plugins/check_service -s docker
include_dir=/etc/nrpe.d/
Server:
/etc/icinga2/conf.d/hosts/test.conf
object Host "Host 1" {
import "generic-host"
address = "192.168.100.153"
vars.os = "Linux"
vars.sla = "24x7"
}
object Service "Docker" {
import "generic-service"
host_name = "Host 1"
check_command = "nrpe"
vars.nrpe_command = "check_docker_service"
}