Kirk and Rob
I was using both screens the WRKSYSSTS to get Total Jobs in the system
to go with QTOTJOB and QADLTOTJ system values and WRKACTJOB for the
others, using the WRKSYSSTS ASTLVL(*intermed). If I remember correctly
you are right that any job (even ended jobs w/spoolfiles) count in this
total, and that the system has to keep track of these jobs and that is
what the QTOTJOB and QADLTOTJ values correspond to. When this table in
the OS or the Active Job table exceed the initial limit then it has to
be expanded. Used to be when that happened the system kind of suspended
jobs while it did housekeeping and expanded the table(s).
I was just wondering in v6.1 if that still held true. Since we seem to
be bouncing back and forth over limits as our number grow and shrink.
Maybe it doesn't matter as much coming back under a thresh hold, but I
thought the system freed up that space meaning it would have to
reallocate it again when crossing the other direction.
Like I said this was some 15 years ago that I was involved in doing some
heavy tuning and I know a lot has changed. We are having some timeout
issues in some jobs that can't be explained and they could be elsewhere
network, MS servers, etc. but haven't been found yet so fingers are
pointed our way. That was when I noticed the values trying to help
someone figure out the problem and thought from past experience that
they didn't seem to be set to values I would have thought they should.
date: Wed, 5 Sep 2012 21:17:29 -0700
from: Kirk Goins <kirkgoins@xxxxxxxxx>
subject: Re: Performance tuning
First what screen are you looking at WRKACTJOB or WRKSYSSTS? If I
correctly WRKSYSSTS's Jobs in system value includes jobs ( even ended
) that still have spoolfiles on the system. WRKACTJOB show real running