7.1.3 Investigating the cause of a failover (cluster system)
You must investigate the cause of a failover that occurs when ITSLM is running in a cluster system. Failover occurs when an event that leads to failover occurs on the active server. For the types of failures that result in failover, see 6.1.3 Failover timing.
- Organization of this subsection
(1) Before you start
-
Check if a message has been output.
For details about troubleshooting when no message is output, see 7.1.6 Collecting the data needed for determining the cause of a problem.
(2) Procedure
To investigate the cause of a failover:
-
Check the message output at the active server.
Messages are output to the cluster software's logs in addition to the ITSLM - Manager window, event log, integrated trace log, or message log.
-
For details about the cluster software's logs, see the cluster software documentation.
-
For details about the event log, integrated trace log, and message log output destinations, see 7.1.1 Checking and handling the output messages.
-
-
If possible, take the appropriate corrective action based on the information provided in the message.
For details about the messages see 11.3 Messages.
Message checking and taking of corrective action are now complete. However, if the message cannot be handled or the corrective action taken based on the message does not resolve the problem, you must collect data needed for determining the cause of the problem and contact the system administrator.
For details about how to collect the data needed for determining the cause, see 7.1.6 Collecting the data needed for determining the cause of a problem.