If the schedule server and the execution server are both running on the same cluster system, any jobs or job nets that are executing when failover occurs will be displayed as having terminated abnormally (execution suspended status: completion code =239) when the failover is completed.
Recovery processing (such as restarting) for these jobs and job nets whose execution was suspended when the failover occurred is not performed automatically. Check the abnormal termination, and take the appropriate action (restarting, etc.).
In order to continue operation after the failover occurred, create a job to perform the following processing, and then register this job as the job net to be executed when the server starts up. Be sure to set startup days for the job net to be registered. If not set the startup days, the job net will not start after the failover.
Output a job net status list with the jobschprint -n command. Redirect the output results to a file and save the file.
Read the file saved in step 1, and check whether the job net is in suspended status by checking whether the completion code is "239".
If the job net was suspended, restart it with the jobschcontrol rerun command.
Repeat steps 2 and 3 for each job net in the list output in step 1.
Sample source code that performs this processing can be found at the following location:
For PRIMECLUSTER (for Solaris or Linux version), Oracle Solaris Cluster
/opt/FJSVJOBSC/sample/abendrerun.c
For MC/ServiceGuard
/opt/FHPJOBSCH/sample/abendrerun.c
For HACMP
/opt/FAIXJOBSC/sample/abendrerun.c
Refer to the Systemwalker Operation Manager Reference Guide for details on the jobschprint command and the jobschcontrol command.
