On Wed, 31 Jul 2024 21:32:48 -0400, Steve Thompson  wrote:

>Thanks to Kolusu, I re-examined my DD statement, and UNIT=3390
>was one char too much and so it became UNIT=339.
>
(That  seems to have been an off-list communication; Idon't
see it in the archives.)

Did Kolusu explain what syntactic production resulted in UNIT=339?
in <https://www.ibm.com/docs/en/zos/3.1.0?topic=up-syntax-1>
I see only:
{UNIT=([ddd          ] [,unit-count] [,DEFER][,SMSHONOR]) }
       [/ddd         ] [,P         ] [,     ]
       [/dddd        ] [,          ]
       [device-type
       [group-name

{UNIT=AFF=ddname                                }
 
>Fixing that DD statement solved this problem.
Please show the DD statement before and after fix.

>But I would have thought the DD statement should have been failed
>for an incorrect value. Perhaps SMS saved me by applying default
>info.
>
IBM has a harmful custom of not fixing deficient syntax error
reporting.  Long ago, Peter Relson offered the explanation
that such undocumented constructs might have private uses
within IBM.  More recently, Jonathan Scott reported that a
manifest misbehavior could not be repaired because of
compatibility: some users may be relying on the incorrect
operation.

Beyond that, it was indolent design for a single keyword, UNIT,
to have (at least) three different semantics distinguished only
by different subparameter values. A more robust design would
have used three different keywords.

>When I get time, I'm going to experiment with this.

-- 
gil

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

Reply via email to