W dniu 03.02.2022 o 21:25, Tony Harminc pisze:
On Thu, 3 Feb 2022 at 12:59, Radoslaw Skorupka <[email protected]> wrote:
BTW: current z/OS 2.4 still supports UNIT=SEP construct in JCL. Yes, it
is ignored, but still syntactically correct.
AFAIK this parameter predates MVS. :-)
Not only does it predate MVS, but the very first release of MVS
("OS/VS2 Release 2") ignores SEP so that SRM (remember that...?) can
swap address spaces with the expectation that their effect +- on I/O
will be less spread around.
So SEP= has been ignored for about 50 years, but as you say, it's
still accepted. And it's not even so unusual to see JCL with it.
I found UNIT=SEP in a job created in 1999. I don't know why author used
it, but he... he wasn't well educated.
It was the only case.
Few years later I started collecting obsoleted words in JCL, just for
curiosity. I vaguely remember only few of them are still syntactically
correct. Others cause regular JCL ERROR.
BTW: few weeks ago I found some JCL procedure dated on 1981.
--
Radoslaw Skorupka
Lodz, Poland
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
