this is obsolete doc -- see http://doc.nethence.com/ instead

Setting up NRPE on Nagios server and clients 

Requirements on the server 

Install NRPE, 

apt-get install nagios-nrpe-server nagios-nrpe-plugin 

Register the check_nrpe command into Nagios, 

cd /etc/nagios/ 

vi commands.cfg 

# 

# NRPE 

# 

# 'check_nrpe' command definition 

define command{ 

command_name check_nrpe 

command_line $USER1$/check_nrpe -H $HOSTADDRESS$ -c $ARG1$ 

} 

Note. No need to add a plugin to /usr/lib/nagios/plugins/. The plugin is located on the host running the nrpe agent. No need to configure commands in /etc/nagios-plugins/config/ either. 

Setting up the alarms on the server 

Make sure the NRPE calls work from the server to the client. Execute the script remotely to check the communication and that the script are working fine, 

host=host_ 

/usr/lib/nagios/plugins/check_nrpe -H $host_ 

/usr/lib/nagios/plugins/check_nrpe -H $host_ -c check_load 

/usr/lib/nagios/plugins/check_nrpe -H $host_ -c check_space 

/usr/lib/nagios/plugins/check_nrpe -H $host_ -c check_swap 

/usr/lib/nagios/plugins/check_nrpe -H $host_ -c check_memsh 

#/usr/lib/nagios/plugins/check_nrpe -H $host_ -c check_omreport 

/usr/lib/nagios/plugins/check_nrpe -H $host -c diskusage4nagios 

unset $host_ 

Add the nrpe call for the host (assuming the host is alreay defined), 

cd /etc/nagios/conf.d/ 

vi hosts.cfg 

define service{ 

use generic-service 

host_name host 

service_description Open Manage disk status 

check_command check_nrpe!check_omreport 

} 

define service{ 

use generic-service 

host_name host 

service_description memory 

check_command check_nrpe!check_mem 

} 

define service{ 

use generic-service 

host_name host 

service_description cpu load 

check_command check_nrpe!check_load 

} 

define service{ 

use generic-service 

host_name host 

service_description disk usage 

check_command check_nrpe!check_space 

} 

define service{ 

use generic-service 

host_name host 

service_description should not swap 

check_command check_nrpe!check_swap 

} 

define service{ 

use generic-service 

host_name host 

service_description memory 

check_command check_nrpe!check_memsh 

} 

Reload the nagios configuration on the server, 

/etc/init.d/nagios3 reload

Troubleshooting 

If you get this error, 

CHECK_NRPE: Socket timeout after 10 seconds.

==> make sure nrpe server is running on the client and its port e.g. 5999 is open 

If you get this error, 

CHECK_NRPE: Error - Could not complete SSL handshake.

==> allow the nagios server to connect on client's nrpe.cfg 

References 

Surveiller vos serveurs Linux avec Nagios et NRPE: http://blog.nicolargo.com/2007/10/surveiller-vos-serveurs-linux-avec-nagios-et-nrpe.html 

Configuring Nagios to monitor remote load, disk, swap etc using NRPE: https://kura.io/2010/03/21/configuring-nagios-to-monitor-remote-load-disk-using-nrpe/