RE: Optimizer/query plan issue

Nope - all are 3 - 10 meg.

These are stored procedures called via ODBC from a VB app.

Further update:  Received several ptfs Friday.  Some to prevent the
rebuilds, some to gather more data when it does occur.  

Loaded them Saturday, and on a dedicated system 'ran' all possible
combinations of each stored procedure to rebuild the plans.  This morning
almost EVERY stored procedure is rebuilding - and not saving, of course.  We
are pegged at +++ constantly.  IBM is dialed in looking, but it appears
their 'fix' torched us...

Happy Monday.

JN 

-----Original Message-----
From: Neil Palmer [mailto:neilp@dpslink.com]
Sent: Saturday, August 04, 2001 5:23 PM
To: MIDRANGE-L@midrange.com
Subject: Re: Optimizer/query plan issue


Info APAR II11801 is no longer valid.
See II12963.

...Neil





"Clare Holtham" <Clare.Holtham@btinternet.com>
Sent by: owner-midrange-l@midrange.com
2001/08/04 11:03
Please respond to MIDRANGE-L

 
        To:     <MIDRANGE-L@midrange.com>
        cc: 
        Subject:        Re: Optimizer/query plan issue


Hi Jim,
Was the program whose access plan can't be updated > 16Mb when you 
upgraded?
If so, see that Apar I mentioned - II11801.
Hope this helps,
Clare
----- Original Message -----
From: <Jim.Nelson@RCIS-NET.COM>
To: <MIDRANGE-L@midrange.com>
Sent: Friday, August 03, 2001 8:30 PM
Subject: Optimizer/query plan issue


>
> We have been struggling for two months with the following:
>
> Upgraded to V4R5, and immediately started having performance 
difficulties.
> After a great runaround with IBM support, two problems were identified.
> * Optimizer is deciding to rebuild the access plan for seemingly no 
reason
> * After the access plan is rebuilt, the operating system is unable to 
save
> the new plan.  The system cannot gain a 'lock' on the object, even 
though
> nobody else is running the stored procedure at that moment.
>
> We've loaded several ptfs, some to allow IBM to better monitor our 
system,
> and some to try and reduce the instances that the optimizer decided to
> rebuild the plan.  The number of rebuilds has been reduced some, but it
> still too high (hundreds per day), and we are not yet able to save the
plan.
> This causes every instance of that stored procedure to perform very 
poorly
> (what would take .03 seconds now takes 17.0 or more) the rest of the 
day.
>
> Has anybody experienced similar problems?  If so, are you getting 
anywhere
> on it?  We've been reduced to shutting down/restarting QSRV nightly and,
on
> a dedicated system, running all of our stored procedures.  This will put
> them back 'on track', and prepare us for the day - until they start
falling
> out again.
>
> I'm running out of ways to explain this to users and Sr. Management. Any
> suggestions would be welcome.
>
> Thanks, JN




+---
| 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
+---