On Fri, 18 Nov 2016 13:40:36 -0500, Jim Mulder  wrote:

>> So ... if our messages could readily say "not authorized -- check
>> STEPLIB(+2)"
>
>  So you would like OPEN, when it builds a DEB in which the 
>DEBAPFIN is off, to provide in some DEB extension the 
>concatenation number of some data set which it found to be not 
>APF Authorized.
>
>  That may be a reasonable thing to submit to DFSMS as a requirement 
>or request or whatever we call those things now. 
> 
Absolutely.  And more usefully, if such an indication existed, when a
module was loaded from a catenand not marked unauthorized it could
proceed with the address space authorized.

(I think it's called an RFE.)

"extension"?  Is there today no uncommitted bit in the DEB that could
be exploited for the purpose?

What does the DEB entry for a Program Object library catenand look
like?

Jesse Robinson cited insufficient information in the DEB as reason for
the limitation.  Such a clear understanding of tie cause is a large
initial step toward designing a solution.

Another ply (which I can't find) questioned the usefulness of mixing
APF/non-APF in STEPLIB.  By analogy, I understand that in bygone
days LINKLIST was all-or-nothing; no mixture.  Some customers
must have reported sufficient need for a mixed LINKLIST that IBM
relieved the restriction many releases ago.  Similar arguments apply
to STEPLIB.  More so with JCLLIB.  A user may choose not to undertake
the effort of maintaining authorized and unauthorized but otherwise
similar STEPLIB members for the respective purposes.

-- gil

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

Reply via email to