RE: **SPAM** Re: activation groups and rclrsc

["Holden Tommy" <Tommy.Holden@xxxxxxxxxxxxxxxxx>]

 <snip>Use one name, say your company initials or even QILE.  Leave 
the service programs *CALLER.  I get... debate from this recommendation,
</snip>

*CALLER is fine for such items as service programs as far as I'm
concerned.  The issue is that "most" shops end up compiling every
freakin' thing *CALLER (which results in everything runs in the DAG).

I recommend using a named AG (perhaps for the application i.e. AP_AG)
with service programs compiled to run in the named AG or *CALLER.  Named
AGs are more secure since you KNOW it will never get "locked" into the
DAG.

Just my 2 coppers...


Thanks,
Tommy Holden


-----Original Message-----
From: rpg400-l-bounces@xxxxxxxxxxxx
[mailto:rpg400-l-bounces@xxxxxxxxxxxx] On Behalf Of Buck
Sent: Tuesday, January 10, 2006 2:55 PM
To: rpg400-l@xxxxxxxxxxxx
Subject: **SPAM** Re: activation groups and rclrsc

> We are a small company with 2 developers and I am trying to
> keep the activation group thing as simple as possible.

The simplest solution is to compile all your RPG IV programs with a 
named AG.  Use one name, say your company initials or even QILE.  Leave 
the service programs *CALLER.  I get... debate from this recommendation,

so you should think on it a bit.  Compile all your stuff in the test 
library and try it before going to production.

I've been using a single named AG here for several years because of the 
need to keep it simple.  Considerations are to RCLACTGRP() when changing

*LIBL and to change OVRDBF's for OPNQRYF.  That's what comes to the top 
of my head.
   --buck