On Sun, 1 Sep 2013 10:53:19 -0400, DanD wrote:
>
>Device allocation for JES2 occurs at step initiation.
>Each step would allocate a single device.  If RETAIN was coded AVR would
>recognize the mounted volume and reuse the same device for the next step.
> 
Let's talk about DYNALLOC.  AFAICT, DYNALLOC has no analogue of RETAIN.
It unloads the volume after each FREE, and it must be remounted.  In
desperation, I supplied a DD LABEL=(nn,BLP) for each file (including the
labels) so I could process the entire volume with a Rexx EXEC.

Bruce Black once generously performed an experiment and concluded that
there's no way to DYNALLOC two or more data sets on the same volume
concurrently although JCL DD allocation has no problem with it.  (Of course
at most one of those data sets can be open at a time.)

Why the DYNALLOC restriction?  The message adds insult to injury by
saying "... ANOTHER JOB OR USER."  False.  It's the same job and user.
I wonder whether the message alludes to the intent of the restriction,
and there's an aboriginal logic error in that DYNALLOC fails to recognize
that it's the same user, same job, and same job step?

-- gil

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to