Applicable versions and levels
Solaris versions: 5.0 or later
Linux versions: 5.2, V10.0L10 or later
Linux for Itanium versions: V12.0L10 or later
HP-UX versions: 10.0 or later
AIX versions: 13.2.0
A
Testing failover
When testing failover, send the SIGTERM signal to the following process to stop it:
Jobscheduler: tskwnsrv
Failover does not occur even if you use the poperationmgr command to stop a daemon.
This is because in a cluster configuration, daemons (Jobscheduler and Job Execution Control) to be registered in a cluster are excluded from the automatic startup and stop targets of the services that are stopped by the poperationmgr command.
Triggers for failover
There are numerous failover triggers. Do not, however, initiate failover by forcibly unmounting a shared disk.
On Systemwalker Operation Manager, shared disk information is referred to during the Jobscheduler daemon stop process. Therefore, the stop process of the Jobscheduler daemon no longer runs properly after the shared disk has been forcibly unmounted. (Failover is completed.)
If unmounting the shared disk causes failover, send the SIGTERM signal to the tskwnsrv process on the node where the shared disk was unmounted. Do not send the SIGKILL signal.
