RE: Commands to joblog

On 17 Jan 2002 at 16:28, Gene_Gaunt@ReviewWorks.com wrote:

>
> >> "If anything were to go wrong with the program after sending the
> >> *RQS message but before receiving it to make it old, the request
> >> will be processed after the program ends... and I don't like that."
>
> Then run your programs under your own custom routing entry program
> instead of QCMD.  Then you will be able to stop those dangling request
> messages from running as CL commands if programs fail.
Thanks, Gene.  I considered that.  But it's not the biggest problem.

The biggest problem I see is that when a person presses F9, of course, *RQS 
messages are
recalled.  This includes any *RQS messages that were sent via my "tool."  This 
effect, I don't
think there's an acceptable circumvention for, and for the unwary user, it 
could cause
problems.  I might as well switch to *INFO messages and be done with it... but 
I figured this
would be the one place where I'd find out enough internal information to be 
able to simulate
the LOGCLPGM(*YES) effect.  How disappointing that IBM doesn't seem to provide 
any
means for that!

Anyway, I do appreciate the reply.  Thanks!
--
Qiu Min & Dennis Lovelady
mail: dennis@lovelady.com
URL:  http://www.lovelady.com
ICQ:  5734860
--
"I studied the lives of great men and famous women, and I found that the men 
and women
who got to the top were those who did the jobs they had in hand, with 
everything they had of
energy and enthusiasm and hard work."
        - Harry S. Truman