The ICIM defines two types of notifications: momentary and durable. A momentary notification has no duration, it describes an event that happened at a specific time. An authentication failure is an example of a momentary event. A durable notification describes an event that is active over a period of time. While the event is active, the problem it causes is still in effect. An example of a durable event is a link failure. You can determine a notification’s type by checking the value of the EventType notification attribute.
A notification’s type determines how the value of the OccurrenceCount attribute is increased by a Global Manager. In addition, the source of the notification and the conditions under which a notification is sent, also determine how a notification’s count is increased.
For example, consider the scenario where notification N_1 is active in a Global Manager and there are two different sources (S1 and S2) for the notification:
For durable notifications, the OccurrenceCount represents how many times the notification has become active (notified) during the notification’s life cycle. The count is incremented when the notification reoccurs after it was cleared and renotified.
Using the example above, when S1 notifies, the Global Manager increments the OccurrenceCount to 1. If S2 then notifies, the OccurrenceCount remains 1.
For momentary notifications, the OccurrenceCount represents how many times the Global Manager has received a notify message regarding the notification. The current state of the notification has no effect.
Using the example above, when S1 notifies, the Global Manager increments the OccurrenceCount to 1. If S2 then notifies, the OccurrenceCount increments to 2.
The Global Manager would increment the OccurrenceCount on a re-notify by 1, regardless of whether the notification was active. In fact, if notification N_1 occurred 50 times in the underlying domain, the Global Manager would increase the value of the OccurrenceCount by 50.
If the Event Type is durable or the value for the SM_OCCURANCECOUNT flag is set to FALSE in the runcmd_env.sh file while the event type is momentary, the server always ignores any re-occurrences of an active notification that has occurred during the disconnected period. If the Event Type is momentary and the value for the SM_OCCURANCECOUNT flag is set to TRUE in the runcmd_env.sh file, the server will update the re-occurrences of an active notification that has occurred during the disconnected period.