Re[2]: WRKSBMJOB, WRKUSRJOB slug-li

[eric.delong@xxxxxxxxxxxxxxxxx]

     Dan, see my post from earlier. It includes the instructions for the 
     data area fix. This fix will compress the job tables (WCBT) and 
     speed up your "JOB" commands.


______________________________ Reply Separator _________________________________
Subject: RE: WRKSBMJOB, WRKUSRJOB slug-li 
Author:  <MIDRANGE-L@midrange.com> at INET_WACO
Date:    9/29/00 1:37 PM


DSPJOBTBL was not found on either our V3R2 or V3R7 boxes.
     
We are going to proceed with changing the system values QACTJOB & QTOTJOB, if 
for no other reason than to follow the advice given in the IBM help text for 
those values.
     
The other thing I learned is that they never IPL the V3R2 box until they have 
to (PTFs, what else is there?).  Nobody knows for sure, but the last time they 
IPL'd this box was probably two or three months ago.
     
Dan Bale
IT - AS/400
Handleman Company
248-362-4400  Ext. 4952
     
-----Original Message-----
Type in dspjobtbl and press enter.  If it works at your release, it will 
show you how many job tables exist on the machine, how big they are, and 
some other data.  Long "job display" times usually arise when the total size 
of job tables is a big number.  There is a cleanup option described in
V4R3 - CHGIPLA - but I don't know if it works at V3R2.
     
Your other findings aren't the reason that it takes 30 seconds to scan all 
the jobs.  The system values are not causing this problem - although they 
may be causing other issues.
     
The RISC box manages job tables better than CISC but I seem to remember a 
PTF to help with job table management.
     
Richard Jackson
mailto:richardjackson@richardjackson.net 
http://www.richardjacksonltd.com
Voice: 1 (303) 808-8058
Fax:   1 (303) 663-4325
     
> |-----Original Message-----
> |From: owner-midrange-l@midrange.com
> |[mailto:owner-midrange-l@midrange.com]On Behalf Of D.BALE@handleman.com 
> |Sent: Friday, September 29, 2000 8:51 AM
> |To: MIDRANGE-L@midrange.com
> |Subject: Re: WRKSBMJOB, WRKUSRJOB slug-li 
> |
> |
> |In addition to my last post, WRKACTJOB shows Active jobs = 127. 
> |
> |Got smart (?) and looked up the help for the QACTJOB & QTOTJOB 
> |values.  Seems
> |clear how to set these at the lowest level.  Should I set a 
> |value high enough
> |for these so that the "additional" jobs never (or hardly ever?) 
> |has to kick
> |in?  What's the penalty for setting these too high? 
> |
> |When do these changes take effect?  Next IPL?  (Help didn't say.) 
> |
> |James,
> |
> |Thanks for the reply!  I had never known to look for this before 
> |but, needless
> |to say, I was a little shocked when I found the following numbers: 
> |
> |WRKSYSSTS shows Jobs in system = 514  (for the past 10 minutes, it's been 
> |hovering in the 512 - 515 range)
> |
> |Our system values:
> |QACTJOB  - Initial number of active jobs           20 
> |QADLACTJ - Additional number of active jobs        10 
> |QADLTOTJ - Additional number of total jobs         10 
> |QTOTJOB  - Initial total number of jobs            30 
> |
> |Are these the shipping defaults from the factory?  I _think_ I 
> |want to change
> |QTOTJOB to 500, but will wait for advice from those who know 
> |better on the
> |list.  I have no reason to believe that 514 number goes much 
> |higher; we run
> |fairly level all day long.
> |
> |Just to compare, we have a V3R7 box that currently has 799 "jobs in the 
> |system" and has the same system values as shown above.  However, both
> |WRKSBMJOB & WRKUSRJOB respond within 1-2 seconds.  Does the RISC 
> |box manage
> |fragmented job data better than CISC? 
> |
> |Dan Bale
> |IT - AS/400
> |Handleman Company
> |248-362-4400  Ext. 4952
> |
> |
> |----------------- Original Message ----------------- 
> |Dan,
> |
> |You may want to look at the system values for initial number of jobs and 
> |additional number of jobs, then the total number of jobs on the system. 
> |
> |I've seen this happen when job data is fragmented. 
> |
> |D.BALE@handleman.com wrote:
> |>
> |> Good grief!  WRKSBMJOB & WRKUSRJOB are absolutely slug-like on 
> |this V3R2
> |> system.  It can take up to 30 seconds to get a response, 
> |everything else
> |> (i.e., WRKSPLF, WRKACTJOB) is at or near sub-second response. 
> |This on a
> |> system where the CPU usage is 8.2% for a two minute interval 
> |(right now).
> | Is
> |> this a V3R2 issue?
> |>
> |<<snip>>
     


+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---