RE: Activation group *ENTMOD vs. *NEW in CRTPGM

["Bob Cozzi" <cozzi@xxxxxxxxx>]

I think what's happening is that you can insert ACTGRP('QILE') into one
source member, and ACTGRP('XYZ') into another.  Then at bind-time which one
has priority? By changing the "default" from *NEW to *ENTMOD, it is say to
use the ACTGRP() keyword used in the entry module's source member (on its
header specification). 

Embedding the compiler parameter into the source code is the _best_ practice
in my opinion, and thus gave IBM the justification to make this subtle
change that should not really impact most (any?) code unless you've done the
following:
Secondary *Module(s):  CRTRPGMOD
Main *Module:  CRTBNDRPG

-Bob Cozzi
www.RPGxTools.com
RPG xTools - Enjoy programming again.


-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx [mailto:rpg400-l-bounces@xxxxxxxxxxxx]
On Behalf Of Scott Klement
Sent: Thursday, November 17, 2005 11:30 AM
To: RPG programming on the AS400 / iSeries
Subject: Re: Activation group *ENTMOD vs. *NEW in CRTPGM


> Why does IBM has changed the default value for the activation group of
> created programs from *NEW to *ENTMOD?

Good question.  I know that a lot of people didn't like having *NEW as the 
default, but I'm not exactly sure why IBM changed it now after all of 
these releases.


> I've created my tools with default value for ACTGROUP. Until V5R3 this was
> *NEW. Resources for those tools had been relaesed after the tool-programm
> ended.  Now with *ENTMOD, resources are released when QILE ends. This
> behaviour causes some problems when more than one tool-program is on the
> call-stack or when a program ends abnormaly.

Here's what I did when I upgraded to V5R3:

     CHGCMDDFT CMD(CRTPGM) NEWDFT('ACTGRP(*NEW)')

Now, *NEW is the default again.