RE: ACCPTH(*YES) becomes ACCPTH(*MAYBE) in V5R2

I know that this question misses the point, but is there any obvious
criteria for skipping an access path on your save?

-Jim

-----Original Message-----
From: Al Barsa [mailto:barsa@xxxxxxxxxxxxxxxxxxx]
Sent: Friday, December 05, 2003 3:17 PM
To: Midrange Systems Technical Discussion
Subject: ACCPTH(*YES) becomes ACCPTH(*MAYBE) in V5R2



Hi,

We were migrating a partition from an older box to a newer box using
SAVSTG, which failed.  I have never got SAVSTG to work successfully on
V5R2.  I heard later today that a PTF will be available by January 10.

So we went back and did GO SAVE Option 21, and restored that media.  We
were checking our job logs on the restore, and we discovered that many
access paths were rebuild, which is not what we had predicted.  So we
called this into service.

We went back and displayed the tape, and saw that many of the access paths
were never saved, which mystifies us.

I retrieved the CL code for GO SAVE option 21 (QMNSAVE, but I'm sure you
knew that off the top of your head ;-))))) and saw that ACCPTH(*YES) is
definitely specified.

This afternoon, I got a call from service saying that ACCPTH(*YES) was
really changed to ACCPTH(*MAYBE) in V5R2, and this feature was never
documented.    IMHO, this should have both been documented in the Memo to
Users and the help text.

The bottom line is that it's much faster to rebuild an access path with
zero records than to save and restore it,   So ACCPTH(*MAYBE) is the
performance you would like, but SUPRISES(*YES) is what you got, and it
should always be SUPRISES(*NO).

Al


Al Barsa, Jr.
Barsa Consulting Group, LLC

400>390

914-251-1234
914-251-9406 fax

http://www.barsaconsulting.com
http://www.taatool.com