RE: CICS error

New Message Reply Date view Thread view Subject view Author view Other groups

Subject: RE: CICS error
From: Wakser, David (David.Wakser@infocrossing.com)
Date: Mon Apr 07 2008 - 14:33:43 EDT


Issue CEMT INQ SYSTEM and you will see the DSA limit - you can change it
on that screen.

David Wakser 

-----Original Message-----
From: owner-vse-l@Lehigh.EDU [mailto:owner-vse-l@Lehigh.EDU] On Behalf
Of Ron Ashley
Sent: Monday, April 07, 2008 2:17 PM
To: VSE Discussion List
Subject: RE: CICS error

Can you safely lower the DSALIMIT - for instance move it from 6912k to
6784k. And how do you do it dynamically?

Thanks

-----Original Message-----
From: owner-vse-l@Lehigh.EDU [mailto:owner-vse-l@Lehigh.EDU] On Behalf
Of Tom Duerbusch
Sent: Monday, April 07, 2008 12:50 PM
To: VSE Discussion List
Subject: RE: CICS error

Obviously, watch for just "upping the DSA Limit below the line".....

Do a 3,6,3 and look at F2, for the ICCF system.

Look at the high water mark for below the line.  You don't want to
infringe on that.
Take the HWM and subtract the currently used, to get a true "available".
Round that down to the next 256K boundry, and that is what you can
safely raise the DSALIMIT by.  Note, you can dynamically raise it at
anytime (it is reset back to the SIT when CICS is restarted).  You may
find that 24 bit getvis is mostly needed at startup.

If this is the ICCF partition, you can also monitor and shutdown a
couple ICCF pseudo partitions.  Those are carved out of the 24 bit area.
Issue a "/map" from the console to see your current 5 (default) pseudo
partitions.
Don't terminate the Class=I partition, but you may be able to get by
with fewer Class=A and Class=B pseudo partitions.

For all CICS partitions, look at your partitions start address.  For
years, we have been able to start the partitions at 5 MB.  Then, it
creeped up to 6 MB.  Do a GETVIS SVA to see if you might be able to
rearrange things to reduce the size to get the partition start address
back to 5 MB.  If so, that is 1 MB more that is available to DSALIMIT.
The next time you upgrade vendor products, check the requirements on how
they are loaded in the SVA.
When your VSE release was initially installed (VSE/ESA 2.4 or later) and
FSU'ed to your current release, your vendors may required their SVA
programs in the 24 bit area.  When they upgrade their programs to load
in the 31 bit SVA, their documentation really doesn't "hit you over the
head" that this has changed.  (This includes IBM when you FSU a system.)
Sometimes, you are loading in the 24 bit area, when you don't need to.
With enough of these changes, you might be able to get back to a 5 MB
partition start address.

And obviously, compile your programs with Cobol/VSE and run them above
the line.  You may need to change the PPT to say "datalocation below" if
the program will communicate with a 24 bit program, but getting the code
above the line, is a good deal.

Tom Duerbusch
THD Consulting

Law of Cat Acceleration

  A cat will accelerate at a constant rate, until he gets good and
  ready to stop.


>>> "Dan Gofton" <dgofton@saginawcounty.com> 4/7/2008 6:42 AM >>>
Yes, CICS/TS for a lot of years now and programs above the line.  I'm
also increasing DSA which will take effect next CICS cycle (tonight).

Thanks for the suggestion.

Dan Gofton
Systems Programmer
County of Saginaw, MI
989 790 5525 

> -----Original Message-----
> From: owner-vse-l@Lehigh.EDU [mailto:owner-vse-l@Lehigh.EDU] On Behalf

> Of Dennis McLoud
> Sent: Friday, April 04, 2008 3:49 PM
> To: VSE Discussion List
> Subject: RE: CICS error
> 
> Are you running CICS/TS? Are the programs all running below the line? 
> We have not seen this problem since going to CICS/TS and moving most 
> of our programs above the line. You should only see this error after 
> program space compression.
> CICS doesn't need the pointer to the library unless the program is 
> flushed from memory. You could at least make the problem less frequent

> by giving CICS more DSA.
> 
> Dennis McLoud
> Systronics, Inc.
> 913-829-9229
> 
> 
> 


New Message Reply Date view Thread view Subject view Author view Other groups

This archive was generated by hypermail 2b25 : Fri May 16 2008 - 12:20:06 EDT