What's your setting on the QZBSJOBD jobd for Job log output? Mine is set to *PND
-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx [mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Pete Massiello
Sent: Monday, April 21, 2008 4:34 PM
To: 'Midrange Systems Technical Discussion'
Subject: RE: IPL Job Table Cleanup
What I do is do a WRKACTJOB and then make my QACTJOB greater than that value
for number of jobs on WRKACTJOB, and then do a WRKSYSSTS and make QTOTJOB
greater than that value for number of jobs on WRKSYSSTS.
Of Course, you should check with your Application Vendor to see if they
support making intelligent decisions ;) (See Supported configurations for
further humor).
Pete
-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Mike Cunningham
Sent: Monday, April 21, 2008 4:21 PM
To: 'Midrange Systems Technical Discussion'
Subject: RE: IPL Job Table Cleanup
Thanks Pete. That sounds like a very interesting solution. I was doing some
additional reading and was going to use the *ABONRMAL option but I like this
one much better. I was also looking at the data from the DSPJOBTBL command
and wonder how our settings compare to others. Ours are below. In-use jobs
look way to high to me. I can't find any jobs with a job log pending.
CLEANUP runs daily with a setting for Joblogs of 7 days
Permanent job structures: Temporary job structures:
Initial . . . . : 10000 Initial . . . . : 1500
Additional . . . : 50 Additional . . . : 50
Available . . . : 20 Available . . . : 50
Total . . . . . : 50352
Maximum . . . . : 163520
---------------------Entries----------------------
Table Size Total Available In-use Other
1 16752384 16352 0 16352 0
2 16752384 16352 0 16352 0
3 16749312 16352 0 16352 0
4 1348352 1296 20 1276 0
------------------In-use Entries------------------
Job Output Job Log
Table Active Queue Queue Pending
1 34 0 21 16297
2 1 0 47 16304
3 522 0 1207 14623
4 617 1 464 185
-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Pete Massiello
Sent: Monday, April 21, 2008 3:06 PM
To: 'Midrange Systems Technical Discussion'
Subject: RE: IPL Job Table Cleanup
You don't require the job table to be cleaned up at each IPL. You can set
CHGIPLA CPRJOBTBL(*NONE) or even set it to *ABNORMAL so it doesn't do it
everytime you IPL, which seems excessive. At a few of our customers, we
have a job that runs every quarter that does a CHGIPLA CPRJOBTBL(*NEXT).
Then on the next IPL it compresses the job table, and then sets the value to
*NONE.
Pete Massiello
-----Original Message-----
From: midrange-l-bounces@xxxxxxxxxxxx
[mailto:midrange-l-bounces@xxxxxxxxxxxx] On Behalf Of Mike Cunningham
Sent: Monday, April 21, 2008 11:39 AM
To: 'midrange-l@xxxxxxxxxxxx'
Subject: IPL Job Table Cleanup
Is there a way to cleanup job tables prior to an IPL while the system is
running? I happen to be here to watch our IPL this past weekend when we
switched to long passwords and noticed that the Job Table Cleanup step was
the part of the IPL that took the longest (14 minutes). Is was wondering if
there might be way to do this cleanup task prior to an IPL so the IPL does
not take as long. I found a few documents (this one being the best
http://publib.boulder.ibm.com/iseries/v5r2/ic2924/info/experience/ipla2.pdf
) about reducing the IPL time and some talked about how to reduce the number
of job table entries but the latest info was from 2004. Maybe something
newer in V5R4 or V6R1?
--
--
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,
visit:
http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at
http://archive.midrange.com/midrange-l.
midrange.com
