Restart OpenLDAP event handler

Presentation

The script restart_slapd.sh is an event handler designed for Nagios.

It restart OpenLDAP if the service is CRITICAL in HARD state. OpenLDAP is killed with force if a normal kill takes too much time.

Warning

This script requires root privileges, because it uses the kill command. You have to run Nagios as root or set a SUID bit on the script.

Configuration

Edit the script to set these parameters:

  • SLAPD_PID_FILE: File where OpenLDAP stores its PID

  • SLAPD_START_CMD: How OpenLDAP is started

  • TIMEOUT: time to wait before doing a kill -KILL

Download

This plugin, along with all other Nagios plugins from this site, can be downloaded in a single archive.

Installation

Copy the script with other Nagios eventhandlers, for example in /usr/local/nagios/libexec/eventhandlers/:

# cp restart_slapd.sh /usr/local/nagios/libexec/eventhandlers/
# chmod +x /usr/local/nagios/libexec/eventhandlers/restart_slapd.sh

Edit Nagios commands file and add:

define command{
    command_name    restart_slapd
    command_line    /usr/local/nagios/libexec/eventhandlers/restart_slapd.sh  $SERVICESTATE$ $SERVICESTATETYPE$ $SERVICEATTEMPT$
}

Add the event handler to an LDAP check service:

define service{
    host_name                somehost
    service_description      LDAP
    max_check_attempts       4
    event_handler            restart_slapd
    ...other service variables...
}