On Dec 19, 2007, at 8:52 AM, Thompson, Steve wrote:
-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On
Behalf Of Ed Gould
Sent: Tuesday, December 18, 2007 7:09 PM
To: [email protected]
Subject: Re: Bad JOB card through NJE
On Dec 18, 2007, at 5:30 PM, Thompson, Steve wrote:
------------SNIP------------
Has something changed?
<SNIP>
We seem to have a problem with RACF in this case. The JOB gets to the
receiving node and gets a syntax error. However, my point is, the JOB
card was invalid to start, it and the rest of the JOB should never
have been shipped. If you want blind ship then use /*XMIT not /*XEQ
<SNIP>
Let me try to bring this back out of the weeds (from my perspective).
Is it a bug for JES to do minimal JOB statement checking before
shipping
the JOB to another system, when the JOB statement contains a fatal
syntax error?
What I think started this was the differences in JCL between MVS/
SP1 and
MVS/XA.
I have been given some answers outside of the group on this, and while
USERID is part of JECL (VSE) it is not part of the KEYWORD set for the
JOB card of either VSE or z/OS.
So what we have happen here is the JOB goes to the other LPAR. That
LPAR
fails the JOB because it can't scan the JOB card and find who the JOB
belongs to (syntax error in the JOB). So now what does it do with
this,
it says that the profile for the JOB is invalid.
This is a Microsoft answer. Absolutely correct, but is not the
cause of
the problem.
Regards,
Steve Thompson
Steve,
I think I understand your issue, but to muddy the water (a little) we
also transmitted (and received) jobs to dos again this was ages ago
so something might have changed, I have not kept up with the issues
since then. We transmitted to just about every IBM platform out there
(except SERIES 1) and another that is blanked out of my memory at the
moment) but any IBM system that talked NJE we probably transmitted
(and received) data to and we did it on a daily basis. Sorry about
the omission earlier but IIRC we had 1 dos system. The more primitive
the system it was harder for set up, IMO. Once it was set up it
worked PERIOD.
We did have issues with someone that had a bisync RJE emulator But
our network people (they were really good) worked to get them up. I
do remember our guy swearing at bisync and emulators a few times. I
would have liked the policy of SNA only but I could not enforce it as
it was politics and those trumped everything.
I was able to enforce the rule if were JES2 then you had to be SNA
but JES3 didn't support it unless you had some package (I don't
remember the name, "BDT" but I am not sure) so we had to stay Bisync
with them. Luckily there was only one JES3 user.
I used to talk 1 on 1 with the technical people and that what was
needed to get the information to the right people. More than a few
times I had to "baby sit" with the people as they had no NJE back
ground and yes sometimes I told them what to code.
Ed
----------------------------------------------------------------------
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
