Quality of Service measurement

Issue

The measurement of the Quality of Service is not a functionnality originally from Nagios; aggregating states of the monitored elements is not possible.

Centreon allows access to performance data thanks to the Meta Services plugin. It also offers views on metrics' evolution, but there is no information about aggregated states.

Solutions

Merethis created a plugin that can consolidate a quality of service from "real time" working states of a component list. Each of these components have weighting properties and blocking states, thus affecting more or less the level of Quality of Service it is attached to.
Since it is important to have an eye on this QoS in real time, and over the elapsed time, Merethis reassembles the metrics from lower levels so that they can be graphed and followed up over the time, just like the regular Centreon services.

Proper to Nagios' system, it is possible to be notified when the QoS reaches below a defined limit. Features of Nagios such as the dependencies and escalation are also kept. It is then possible to be notified only by the QoS indicators instead of unitary hosts/services.
The Quality of Service indicators can themselves be part of other QoS measures, the inheritance is done on n levels,

Description

Picture 1: Synthesis of states of Quality of Service in "real time"

Picture 2: Details of a Quality of Service in "real time""

Picture 3: Evolution of performance data

Picture 4: Control Board