Service out of operation system alarms are remains in ACT for many days even system running normal
In one of the running 800xA system 5.0, there is a "service out of operation" system alarm all the services related to PriCS and PriAS for many days even though system running normal with sytem status viewer showing healthy/green.
The service alarms are sometimes not correctly reflected by the SoftAlarm Service. During server or service failover, state changes may go lost or become corrupted. Later versions of System 800xA handle this a bit better (but not perfect...).
If you find obsolete service alarms, just have them deleted.
Deletion of object owning a not idle soft alarm (eg a service provider or group) will lead to an orphaned and lingering alarm that may return with the restart of an aspect server. To permanently remove such alarm, a cold start of the SoftAlarm service is required (search AKS for procedure and side effects).
My main source of system health state over time I derive from the System Event List.