2.2.2 Corrective action when job execution fails
When you have decided the order of job execution, you must then think about the corrective action to take when a job fails to execute properly.
Consider the following:
-
How to execute processing automatically after a job fails
-
How to run the job again
JP1/AJS3 judges whether processing has ended normally or abnormally based on the return code of the executable file. This allows processing, such as sending an email informing the system administrator of the problem, to be executed automatically when an abnormal return code is output.
If you choose to rerun a failed job, you must consider whether to re-execute all the jobs from the beginning of the job flow, or just some of them.
The following figure lists the considerations for re-executing a job.
|
|
![[Figure]](GRAPHICS/ZG020065.GIF)
Whichever method you decide on, you can put a job on hold temporarily before rerunning it. Because execution will not proceed until the "on hold" status is released, you can temporarily alter the execution schedule or suspend execution.
If you choose to re-execute only some jobs, consider whether you want to re-execute the job that ended abnormally, or re-start execution from the next job.