How about adding an MPLUS job monitor looking for jobs on hold?
Or one that looks for certain jobs to be completed at a certain time every
On Wed, May 21, 2014 at 3:16 PM, Steinmetz, Paul <PSteinmetz@xxxxxxxxxx>wrote:
On our R&D system, we had a tape issue, which caused the jobs to stack up
in BRMS jobq.
Tape issue was resolved.
BRMS jobs all ran, except for the save, (which was previously put on)
Nightly save on the LPAR didn't run for five days, because of the held
job. (operator had a senior moment)
AJS calendars and scheduling are actually very flexible. Also, it's quite
simple to run a job every 5 minutes, we have several.
Now the base job scheduler is a different story. There you had to create
duplicate jobs with the same call with a different schedule.
From: MIDRANGE-L [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of
Sent: Wednesday, May 21, 2014 3:02 PM
To: Midrange Systems Technical Discussion
Subject: Re: AJS (Advance Job Scheduler) issue
From: "Steinmetz, Paul" <PSteinmetz@xxxxxxxxxx>
Did you know, that if a job submitted from AJS, is put on hold in the
jobq before it becomes active, AJS will not submit any future jobs for
That's interesting. I don't think I've run into this before. Thank you.
I must admit that I too think this is probably the Right Thing. A
user-settable option might be nice -- a per-job user-settable option even
better -- but in general I think the assumption is that a given job will
perform the same processing no matter when it is submitted; and therefore
if you've held one instance of the job, there would be no reason you would
want another instance jumping in.
So, actually I would be curious about what the specifics of your case are.
Is there logic in the program called that does different things depending
on when it is run? Or does the job depend on its own output from a previous
run? -- something like that?
Unfortunately, we have some things set up like that where I work (set up
that way before my time, mind) but fortunately we don't have many jobs
scheduled to run more than once a day, so this hasn't come up. In a few
cases we have created duplicate jobs with the same call but a different
(FOOJOB1 and FOOJOB2, for example). We've done that not because of the
issue you reported, but just because AJS's calendar is so inflexible. (And
it would get out of hand quickly if you wanted to run the job every five
<i><b>NOTICE: This email and any attachments may contain proprietary or
privileged information for the use of the intended recipient. Unauthorized
use or distribution is prohibited. If you received this email in error
please notify the sender and delete the email. Thank you.</b></i>
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing
list To post a message email: MIDRANGE-L@xxxxxxxxxxxx To subscribe,
unsubscribe, or change list options,
or email: MIDRANGE-L-request@xxxxxxxxxxxx Before posting, please take a
moment to review the archives at http://archive.midrange.com/midrange-l.
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives