On Thu, 5 Mar 2020 21:07:32 -0500, Tom Conley wrote:
>
> ..., I've used the following JCL
>for years to check out procs that fail (typically my STC sysout goes to
>the bit bucket):
>
>//IBMUSERC JOB 'IBMUSER',
>// CLASS=A,
>// NOTIFY=&SYSUID,
>// MSGCLASS=X,REGION=0M,MSGLEVEL=(1,1)
>//TESTPROC EXEC procname
>
>The error becomes immediately apparent.
>
In order to extend my understanding, I looked at:
https://www.ibm.com/support/knowledgecenter/en/SSLTBW_2.1.0/com.ibm.zos.v2r1.ieab600/iea3b6_Examples_of_the_TERM_parameter.htm
Examples of the TERM parameter
z/OS MVS JCL Reference SA23-1385-00
...
//DD1 DD TERM=TS,SYSOUT=*
In a background or batch job, the system ignores TERM=TS and
recognizes a sysout data set. (An allocation error occurs if SYSOUT=*
is not coded with TERM=TS.)
...
I immediately had a problem -- I am very accustomed to coding
//SYSPRINT DD SYSOUT=*
by itself, not with TERM=TS. Ah, now I get it. Scope of quantifiers
in English can be ambiguous. It would be clearer if it said:
(An allocation error occurs if TERM=TS is coded without SYSOUT=*.)
Must it be SYSOUT=*? Will neither SYSOUT=specific-class nor
SYSOUT=(,) be accepted? What if I use OUTPUT statements to
direct my JES data sets to three different classes and don't specify
MSGCLASS?
And, "...the system ignores TERM=TS ...", but "An allocation error occurs ..."
If it's truly ignored, it can't cause an error. How about:
... if TERM=TS is coded without SYSOUT=* an allocation error occurs.
Otherwise, TERM=TS is ignored.
Feels like an RCF to me.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
