Recursive Triggers and ILE/RPG

>>  and I am looking
at 30 to 40,000 base updates each night, resulting in the creation and
destruction of 150 to 200,000 activation groups in this job. Now, on a
530 this is not the end of the world, but I would like a better
solution. Any ideas? <<

Walden,

Maybe you could leave the triggers activated - use RETURN with *INLR
*OFF, and just do a RCLACTGRP *ELIGIBLE at the end of the job. I'd state
that this would work with confidence if it was a normal CALL, but in
this case, I'd try it in a controlled manner, as I'm not 100% certain
that each subsequent level will always end up using the same invocation,
so this could conceivably end up creating a new invocation anyway, and
then leaving it there as a permanent orphan. How could you test this
(without crashing the system of course)? Monitoring non-database faults?
Maybe someone else has an idea, or better yet, maybe someone can confirm
that this is not a problem, or has another solution entirely, or ...

Pete
 


--
 - Pete Hall peteh@earth.inwave.com
http://www.inwave.com/~peteh/

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* This is the Midrange System Mailing List!  To submit a new message,   *
* send your mail to "MIDRANGE-L@midrange.com".  To unsubscribe from     *
* this list send email to MAJORDOMO@midrange.com and specify            *
* 'unsubscribe MIDRANGE-L' in the body of your message.  Questions      *
* should be directed to the list owner / operator: david@midrange.com   *
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *