RE: QSERVER subsystem problems -- MIDRANGE-L

Hi, the same thing happened to me some time ago.  After a lot of searching
and headaches we found that the cause of the problem was an LU session of
client Access (anynet over tcpip), so we finish and deactivate all the
devices connected this way to our as/400 and let the subsystem finish
normally.  I think we were at v3r2m0 at the time.

Emilio Padilla

-----Original Message-----
From: owner-midrange-l@midrange.com [mailto:owner-midrange-l@midrange.com]On
Behalf Of Eric Sundell PMI
Sent: Wednesday, March 31, 1999 3:20 PM
To: "pytel@us.ibm.com" ; "MIDRANGE-L@midrange.com" ; MIDRANGE-L@midrange.com
Subject: RE: QSERVER subsystem problems

Can't you end a job like that 'Abnormaly' (ENDJOBABN)

____________________Reply Separator____________________
Subject:    Re: QSERVER subsystem problems
Author: "pytel@us.ibm.com" [SMTP:pytel@us.ibm.com]
Date:       3/30/99 8:52 AM

I am afraid the only thing you can do is to watch APARs for a PTF to fix
the problem.
Job will not end if it cannot free some resource, which it should free.
It is a code problem. Most probably it is some difficult to reproduce
timing problem, so you are unlikely to see this problem reappear.

Best regards
    Alexey Pytel



Matthias Oertli <oertlim@s054.aone.net.au> on 03/30/99 10:43:20 AM

Please respond to MIDRANGE-L@midrange.com

To:   midrange-l@midrange.com
cc:    (bcc: Alexei Pytel/Rochester/IBM)
Subject:  QSERVER subsystem problems





I've had a curious problem: The QSERVER subsystem was ended
(ENDSBS QSERVER OPTION(*IMMED)) and most jobs within it
ended. But one job (QPWFSERVSD) didn't end, instead it
grabbed 80% of the CPU and just sat there at status END.
Hence the QSERVER subsystem also didn't end it just sat
there at status END.
A short while after this a backup kicked in. Because it
didn't get much CPU it didn't get anything done. Therefore
it was ended (*CTRLD). But now, it too didn't end, it just
sat there at status END consuming a few tenth of CPU
percents.
IBM support recommended shutting down tcpip and see if that
allows jobs to end. It didn't. QSERVER didn't accept any end
requests. Then they recommended PWRDWNSYS which was done.
This timed out. (though SRC B900BF10 was avoided by
specifying TIMOUTOPT(*CONTINUE)). The machine then IPL'd
normally and all seems well. IBM recommended a RCLSTG asap
which I will do.

But... I would like to know what really went on. What went
wrong and what can I do to avoid this problem in the future.
Joblogs et al aren't helping. No indication of a problem
anywhere that I can see. Btw: I'm on v4r3 cum 9054.

Any insight would be appreciated.
Regards,
Matthias

--------------------------------------------------
Matthias Oertli   oertlim@s054.aone.net.au
+---
| 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
+---



+---
| 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
+---
+---
| 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
+---

+---
| 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
+---