check_usolved_barracuda_temperature

Overview

This Python Nagios/Icinga plugin checks the temperature and fan state of Barracuda firewalls. If the temperature is higher than the given treshold you'll be informed. You'll also get noticed if a fan isn't working anymore with the integrated hardware check.

The plugin also returns performance data for the temperature sensors.

Authors

Ricardo Klement (www.usolved.net)

Installation

Just copy the file check_usolved_barracuda_temperature.py into your Nagios plugin directory. For example this path: /usr/local/nagios/libexec/

Give check_usolved_barracuda_temperature.py execution rights for the nagios user. This plugin needs Python 2 to be installed and uses the libraries sys, os and optparse.

Why not Python 3 you may ask? Most Nagios / Icinga installations are already using other plugins which are written in Python 2. So for compatibility reasons I've decided to use Python 2 as well.

Make sure you've enabled the SNMP service on your Barracuda firewall. If you have a cluster it's good to configure the SNMP service on the virtual server layer on your Barracuda. Details to find here.

I've tested the plugin on Barracuda appliances F100b, F200b, F200c, F600c and F800b.

Usage

Test on command line

If you are in the Nagios plugin directory execute this command:

./check_usolved_barracuda_temperature.py -H ip_address_of_barracuda -C snmp_community

The output could be something like this:

OK - 8 fans / CPU Temperature: 49.5 *C, Mainboard Temperature: 23.0 *C
Chassis Fan 1A: 1917 RPM
Chassis Fan 1B: 2109 RPM
Chassis Fan 2A: 2343 RPM
Chassis Fan 2B: 2812 RPM
Chassis Fan 3A: 2481 RPM
Chassis Fan 3B: 2721 RPM
Chassis Fan 4A: 2481 RPM
Chassis Fan 4B: 2909 RPM
CPU Temperature: 49.5 *C
Mainboard Temperature: 23.0 *C

Here are all arguments that can be used within this plugin:

-H 
Required: IP or hostname of the Barracuda firewall node with a running snmp service

[-C ]
Optional: SNMP Community String

[-V ]
Optional: SNMP version 1 or 2c are supported, if argument not given version 2 is used by default

[-O ]
Optional: temp or fan. If no object is given fans and temperature will be checked

[-w ]
Optional: Warning threshold for the temperature in degree celsius

[-c ]
Optional: Critical threshold for the temperature in degree celsius

[-P ]
Optional: Write -P yes if you like to have performance data

[-T ]
Optional: SNMP timeout in seconds. Default is 30 seconds.

Install in Nagios

Edit your commands.cfg and add the following.

Example for checking the temperature with performande data enabled (detecting not working fans works by default):

define command {
    command_name    check_usolved_barracuda_temperature
    command_line    $USER1$/check_usolved_barracuda_temperature.py -H $HOSTADDRESS$ -P yes -C public -w $ARG1$ -c $ARG2$
}

Example for just checking the fans (if they are not spinning anymore a critical will be thrown):

define command {
    command_name    check_usolved_barracuda_temperature
    command_line    $USER1$/check_usolved_barracuda_temperature.py -H $HOSTADDRESS$ -P yes -C public -O fan
}

Edit your services.cfg and add the following.

Example for checking the temperature:

define service{
    host_name               Test-Server
    service_description     Barracuda-Temperature
    use                     generic-service
    check_command           check_usolved_barracuda_temperature!60!70
}

You could also use host macros for the snmp community.

What's new

v1.2 2015-11-10 Fixed perfdata and replaced spaces with a "-"

v1.1 2015-07-22 Fixed output when snmpwalk returned quotes in string

v1.0 2015-07-17 Initial release