REL (Remote Execution Layer) Addon

The Nagios REL addon provides an alternative way to send service check results from remote hosts back to nagios. It was developed for scenarios where the customer's security policy does not allow extra ports through the firewall for NRPE orNSCA, which both depend on TCP/IP connections. Currently, REL uses SMTP to send the service check results to the Nagios server, but it can easily be extended to use other protocols like nntp or ssh. The image below shows how REL works.

Let's have a look on an example:

For this example we will setup a process check with our check_process plugin (but you can use any other Nagios plugin as well.) First of all, we need a service definition on our Nagios server. We have defined a passive-check-template which can be used for all passive service checks.

define service{
        register                        0  name                            passive-check-template
        use                             generic-service
        check_freshness                 1
        check_period                    none
        passive_checks_enabled          1
        max_check_attempts              1
        check_command                   passive_check_missing
        freshness_threshold             600
}
define service{
        use                             passive-check-template
        host_name                       webserver
        service_description             Syslog Process
        contact_groups                  nagios,unix-admins
}

We do not disable active service checks. Instead we set the check_period to 'none' (this avoids the cross in the CGIs, indicating that there is something switched of). The check_command will be called if there was no passive service check result in the last 10 minutes (600 sec). The script looks like this:

#!/bin/sh

/bin/echo "CRITICAL: Passive Service check is missing!"

exit 2

Now it's time to configure sendmail (qmail works as well). Just add the following alias to your /etc/aliases file and recreate the database.

nagiossmtp: "|/usr/bin/perl -w /usr/local/nagios/bin/smtpreceiver.pl"

Make sure, the sendmail user has write access to the nagios external command file. That's it.

On the remote host, you will need to edit the relconfig file. The HOST should match the host_name used in nagios. Protocol defines which protocol to use (currently only SMTP). The DEST_MAIL should match the alias we have just defined.

TARGET_HOST=127.0.0.1
HOST=webserver
protocol=SMTP
SENDMAIL_COMMAND=/usr/sbin/sendmail -oi -t -f SOURCE_MAIL -od
DEST_MAIL=nagiossmtp@my-nagios-host.com
SOURCE_MAIL=nagios@somewhat.com

And finally you have to create a cron entry that starts the plugin every few minutes using the following command.

# -c specifies the configuration file to use. if ommitted, then /etc/relconfig will be tried.
# -s specifies the Service to which this check will be associated
# next parameters are the plugin to execute and the plugin parameter.
./plugin_wrapper.pl -c $NAGIOSPATH/etc/relconfig -s "Syslog Process" $PLUGINPATH/check_process.pl -c /sbin/syslogd