Nagios Module

NOTE: This module was initially hard-coded as a quick stop-gap and I am only now in process of parameterising the module. This currently assumes a RHEL family O/S.

This module will manage server and client nodes. It relies on exported resources to auto-add clients, thus requires puppetDB to be used.

The module also optionally installs and configures Graphios in order to send results to a Graphite server.

Standard checks deployed on all hosts:

• Load Per Core
• Swap Usage
• Logged On Users
• Disk Usage
• Zombie Processes
• Memory Usage
• Puppet Agent Status
• SSH
• PING

Usage

Server:

node nagios {
  class { 'nagios':
    role => 'server',
    graphios_install => true,
    graphite_host => $graphite_host
  }
}

Client

node client {
  include nagios
}

Hieradata

Extra plugins can be added in hiera as follows:

nagios_plugins:
  - nagios::plugin::notify

This will include the notify plugin on this host.

Hostgroups

Hostgroups have two steps, first create all the hostgroups in quetion:
nagios::hostgroups:
  Linux:
    alias: 'Linux Servers'
  admin-prod:
    alias: 'Admin Servers - Production'

Then add the servers into the groups by adding a lookup in the node:
nagios_hostgroup: 'admin-prod'

This will add the host to both the Linux and Production Admin hostgroups.

NRPE

nrpe::dont_blame_nrpe: 0
nrpe::allowed_hosts:
  - '127.0.0.1'
  - '192.168.0.1'

Contacts

nagios::contacts:
  nagiosadmin:
    contact_name: nagiosadmin
    alias: 'Nagios Admin'
    ensure: 'present'
    email: nagiosadmin@example.com

Override defaults

nagios::plugin::nrpe_core_load::warn: 4
nagios::plugin::nrpe_core_load::crit: 8

Credits

This is a collection of modules written by their respective owners.
All credit goes to the respective authors