Most of the Storage Automated Diagnostic Environment events are based on health transitions. When, for example, the state of a device goes from 'online' to 'offline' , a health transition occurs. It is the transition from 'offline' to 'online' that generates an event, not the actual value 'offline'. If the state alone was used to generate events, the same events would be generated all the time. Transitions cannot be used when monitoring logfiles, so log events can be very repetitive. This problem is minimized by attaching thresholds to entries in the logfiles.
Thresholds ensure that a minimum number (up to eight) of logfile entries within a certain time period occur before an event is generated. The Storage Automated Diagnostic Environment also includes an event maximums database that keeps track of the number of events generated about the same subject in the same 8-hour timeframe. This database is used to stop the generation of repetitive events when there is no other way to do it. For example, if the port of a switch was toggling between offline and online every few minutes, the event maximums database would ensure that this toggling is reported only once every 8 hours instead of every five minutes.
Events are usually generated using the following rules: