check-lxd

nagios plugin for check lxd containers

Description

Check the status of the containers from lxd server, without configure containers itself. Generate standard status codes for use it with a nagios-like monitoring server.

Requirements

Python3 with standard libraries. Must be executed by user of lxd group. for example, in nagios, you can add the default nrpe user to lxd group:

sudo usermod -a -G nagios lxd

Install

Put the script in path of your convenience.

nrpe

For example, in your nagios server, you can create a service with nrpe for use this check:

define service{
        use                             generic-service,graphed-service
        host_name                       lxd-server
        service_description             mycontainer_status
        check_command                   check_nrpe!check_mycontainer
        }

and in your nrpe.cfg in lxd server:

command[check_mycontainer]=/usr/lib/nagios/plugins/custom/check_lxd state mycontainer

Usage

usage: check_lxd.py [-h] option [thresholds] container_name

options

state - Check the state of the container and generate CRITICAL state if not running

example:

check_lxd state mycontainer

run - Run a custom command into container.

example:

check_lxd run -c "/bin/bash /home/foo/foo.sh"

procs - Check processes of the container and generates WARNINGor CRITICAL states if the number is outside of the required threshold ranges.

example:

check_lxd procs -w 12-14 -c 12-20

mem - Check the amount of memory used by the container and generates WARNING or CRITICAL states if the number is higher of the threshold defined in MB.

example:

check_lxd mem -w 1024 -c 2048

optional arguments: -h, --help show this help message and exit

TO DO

• connect to externals lxd servers through https
• other checks: network, disk, custom devices...