This section describes the following:
Each server runs syscheck periodically and reports any problems found through platform alarms. The severity of platform alarms is one of the following:
When one or more problems in a given category has been found, the server reports one corresponding event notification to its Surveillance log and its serial port 3. If the server is not the active server, it also sends the event notification to the active server. The active server reports its own platform events to its own Surveillance log and to its Serial Port 3, and also sends an SNMP trap and displays a GUI notification for either its own platform events or for the non-active server’s platform events.
Each of the events 4100, 4200, and 4300 contain a 16-character hexadecimal bitmasked string that indicates all of the platform events in that category that currently exist. To decode which platform events exist, use the procedure described in “How to Decode Platform Alarms”.
Each time the combination of platform events in a given category changes, a new event is reported. Following is an example of how platform events are reported:
At first, only one major platform event is reported on the standby server. A 4200 event with the alarm number of the event is reported.
One minute later, another platform event exists on the standby server (and the first one still exists). Another 4200 event is reported, with a bitmasked string that indicates both of the platform events that exist.
One minute later, another platform event exists on the standby server (and the previous ones still exist). Another 4200 event is reported, with a bitmasked string that indicates all of the platform events that exist.
One minute later, the first platform event is cleared. Another 4200 event is reported, with a bitmasked string that indicates the two platform events that still exist.