On Sat, 24 May 2008 21:12:25 -0400, Robert A. Rosenberg wrote:
>At 14:46 -0500 on 05/22/2008, William H. Blair wrote about Re:
>80-Column Minds (Was: SMP/E question):
>
>>But I was surprised and pleased to find that it just
>>worked in JES2 (at least by the time we got around to
>>using MVS 3.8). So as far as I know, this was never a
>>problem or restriction in JES2 (like it was in HASP).
>>I just wrote the long records to a SYSOUT=(x,INTRDR)
>>data set. The only thing you had to watch was that
>>the actual JCL and JECL could not be anything other
>>than LRECL=80. So, the program actually used two DCBs
>>to write the records to the INTRDR SYSOUT data set,
>>depending on and using the implied SYSOUT DISP=MOD.
>>The '//XXXXXX DD *' record was written with the DCB
>>that had LRECL=80, but the in-stream input records
>>were written with the LRECL=20x DCB. It just worked.
>
>I may be misremembering but could you not just use RECFM=V and send
>84 byte JCL and 208/212 LRECL SYSIN Data?
>
By experiment (nowadays, JES2) JCL statements are quietly
truncated at 80 columns; data beyond column 80 does not
even appear in the listing. But earlier (perhaps pre 1.7,
JES2) RECFM=V caused serious I/O errors on SYSIN. JES3
has long worked better; perhaps that's part of what you
pay for.
From:
Title: z/OS V1R1.0 CS: IP Configuration Reference
Document Number: SC31-8776-00
2.8.44 "z/OS V1R1.0 CS: IP Configuration Reference"
_____________________
2.8.44 JESRECFM Statement
Usage Notes
* Only use the value V when running on JES3 systems.
It's dismaying that restrictions such as this must be inferred
from a TCP/IP manual, rather than appearing in the JCL RM.
BTW, the restriction is still stated in V1R9.0. I suspect it's
outdated. Perhaps time for a RCF.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
