Hello out there, anyone know of a way to 'capture' a jcl error and force it to display a message to the console. This can be done easily when a job abends using cond code checking or if then else statements, but to the best of my knowledge, if the error is a JCL error the job just STOPS and no further steps will execute. I am particularly concerned in a case whereby we get a JCL error 'space requested not available' - we have procs that run that have the potential to get this error and we need to know about it right away if this condition is encountered...any ideas.
We have job scheduling tools (ca7 to be exact)...in this case, however, the 'job' in question is actually forced by a user exit in our ADABAS database...so there is no control by CA7 unfortunately. Anybody else have any ideas?
Joined: 01 Mar 2006 Posts: 290 Location: Basel, Switzerland
if you have any scheduler, then you should not submit
any other job directly without control of your scheduler.
check how your scheduler schedules defined applications,
then create one job that will schedule a scheduler application,
which includes your job as you wanted, but then under control of
I am sort of in a catch-22 with the scheduler tho because if I allow the scheduler to submit this job, then my database will 'think' that it has copied off its protection logs when in fact it may not have. The user exit handles all the business between the protection logs and the database. This would not be a good situation to be in. I guess bottom line, there is no way to capture jcl errors and cause another even to take place...unless I go the job scheduler route...which in this case I can't do.
I am trying to capture JCL errors after the job has started (not syntax errors). My biggest concern is a 'space requested not available' jcl error. This happens once in a blue moon. I am allocating a rather large first extent in a couple of my procsteps and if the SMS pool of volumes is rather fragmented or full due to a busy batch window this can spell trouble.