The Application Server allows to monitor the observations as they are created or modified and can trigger actions whenever those modified observations fulfill some predefined rules. One prominent use for observation monitoring is sending out messages whenever an alarm condition is detected (also known as alarming), but observation monitoring goes beyond that.

Observation monitoring is based on observation monitoring definitions that can be created by the user. Observation monitoring definitions are specified with a specific feature context and therefore apply to all features in that context that are marked as being spied. See chapter enabling observation monitoring for more details.

An observation monitoring definition consists of tree main elements triggers, conditions and actions. Triggers are the events that cause rules to be checked triggers can be creation of new observations  or modification of observations. Conditions are the rules that are checked whenever a trigger fires, they specify which characteristics have to be fulfilled by an observations for actions to be performed. Actions are the procedures that are carried out if a rule was fulfilled.