Nagios Plugin for checking failed SimpliVity OmniCube backups and assigned policies.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
README.md
check_usolved_omnicube_backup.py
icingaexchange.yml

README.md

check_usolved_omnicube_backup

Overview

This Python Nagios plugin checks for any failed VM backups on SimpliVity Omnicubes. There's also a second mode in this plugin to display all VMs with a specific backup policiy.

Authors

Ricardo Klement (www.usolved.net)

Installation

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

Give check_usolved_omnicube_backup.py execution rights for the nagios user. This plugin needs Python 2 to be installed.

Because this plugin uses SSH to connect to the OmniCube you need the Python module "pexpect" installed which contains "pxssh".

sudo yum install pexpect.noarch

or

sudo apt-get install python-pexpect

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.

Usage

Test on command line

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

./check_usolved_omnicube_backup.py -H ip_address_of_omnicube -U "domain\\\myuser" -P yourpassword -M status

This should output something like this:

OK - All backups were successful

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

-H <host address>
Give the host address of an OmniCube with the IP address or FQDN

-U <ssh user>
OmniCube SSH username (remember to escape the backslash in the username. For example domain\myuser would be "domain\\\myuser" or domain\\\\myuser as argument)

-P <ssh password>
OmniCube SSH password

-M <mode>
Plugins mode (-M status (to check if all backups were successful), -M status:notstarted (to check if all backups were successful including not started backups) or -M policy (to check if all VMs have policies assigned))

[-N <policy name>]
When you use -M policy then you have to set the policy name here

[-E <exclude VMs>]
Optional: Exclude comma separated hosts from the check for policy check

[-D <backup date>]
Optional: Without an argument the backups status check gets the status from yesterday.
If you wish to check for other time ranges, give the argument -D YYYY-MM-DD

[-T <timeout>]
Optional: SSH and OmniCube command timeout in seconds. Default is 45 seconds.

Install in Nagios

Edit your commands.cfg and add the following.

Example for checking the backup status:

define command {
    command_name    check_usolved_omnicube_backup_status
    command_line    $USER1$/check_usolved_omnicube_backup.py -H $HOSTADDRESS$ -U $ARG1$ -P $ARG2$ -M status
}

Example for checking the assigned backup policies:

define command {
    command_name    check_usolved_omnicube_backup_policy
    command_line    $USER1$/check_usolved_omnicube_backup.py -H $HOSTADDRESS$ -U $ARG1$ -P $ARG2$ -M policy -N policyname -E $ARG3$
}

Edit your services.cfg and add the following.

Example for checking the backup status:

define service{
	host_name				Test-Server
	service_description		OmniCube-Backup-Status
	use						generic-service
	check_command			check_usolved_omnicube_backup_status!domain\\\\username!password
}

Example for checking the assigned backup policies:

define service{
	host_name				Test-Server
	service_description		OmniCube-Backup-Policies
	use						generic-service
	check_command			check_usolved_omnicube_backup_policy!domain\\\\username!password!hostnotwant1,hostnotwant2
}

You could also use host macros for the username and password. Again: Pay attention for escaping the domain and username with the backslashes. If you use a frontend like Centreon you need to put 4 backslashes between the domain and username.

Don't worry if the check takes a while for returning the info. The OmniCube SSH commands will take a while for executing.

Errors you may encounter

If you get some cryptic erorrs with pxssh on CentOS 6 you need to edit the fallowing file:

vi /usr/lib/python2.6/site-packages/pxssh.py

Add the following lines of code:

self.sendline() #Line 134
time.sleep(0.5) #Line 135

Should be inserted above the fallowing line:

self.read_nonblocking(size=10000,timeout=1) # GAS: Clear out the cache before getting the prompt

What's new

v1.3
Added enhanced argument status:notstarted to check for not started backups (OmniCube command svt-backup-show just includes started backups)
Bugfix for parameter -D. The --until argument is now being used for the OmniCube command
Timeout parameter is now also being used for the OmniCube commands and not just the SSH connect

v1.2
Changed policy check that you can list hosts with a specific policy name

v1.1
Bugfix for the backup status. When the retried backup succeeded you'll get an OK status and an extended info which hosts needed more than one try

v1.0
Initial release