What is Nagios?

Nagios is an industry standard server monitoring service, allowing an at-a-glance overview of machines in their server farm. It allows flexibility for users to specify exactly what aspects of each machine they want to monitor, along with how often they are checked.

What are the checks we use?

For most typical build and test machines, we have the following checks in place:

Check	How Often	Warning Boundary	Critical Boundary
Check SSH	Every 15 minutes	-	Can't connect to machine
Current Load	Every 30 minutes	15,10,5	30,24,20
Disk Space Root Partition	Every 60 minutes	20% free	10% free
Check Jenkins connection	Every 30 minutes	Temporarily disabled	Fully disconnected
Ping	Every 15 minutes	rta 200, 20% packet loss	rta 500, 60% packet loss
Check RAM	Every 10 minutes	15% free	5% free
Check Timesync	Every 15 minutes	Time not synchronized / service not running	Can't find required info
Check Package Manager	Once a day	Any updates required	Critical Updates required

The check_ssh check output defines if the host is considered connected to the Nagios Server. If this is critical, it can be assumed no other checks will work either.

Note: For the checks in bold, the checks are platform specific.

An up-to-date version of the checks can be found in ansible/playbooks/Supporting_Scripts/Nagios_Ansible_Config_Tool/templates