Re: Service program authority adoption

["Simon Coulter" <shc@xxxxxxxxxxxxxxxxx>]

Hello David,

You wrote:
>In general yes, unless the program specifies useadpaut(*no).

USEADPAUT(*NO) controls what happens above the caller.  As in:

        PGMA  -- adopts QGOD
          calls PGMB which says USEADPT(*NO) therefore doesn't have QGOD
                (unless it adopts QGOD itself)

The system does support the ability to stop called programs from adopting
via a propogate authority attribute but Rochester haven't seen fit to
expose that.  I keep asking for it but I guess I'm alone.  You see my view
is that it is the caller's responsibility to determine whether inherited
authority is available to the callee, NOT the callee's job to say thanks
for the God-like rights but I really don't want them.  That way programs
that call exit programs can ensure that the exit program runs only with
the user's real authorities.  The current mechanism is simply
arse-backwards.

>There are some exceptions that can be significant, like triggers,
>which end adoption.

I didn't know that.  I can understand a trigger not inheriting adopted
authority from earlier in the stack but I doubt they stop the trigger
itself from adopting authority via USRPRF(*OWNER).  Hmm, I can feel an
investigation coming on -- it'll have to wait a bit though ... I'm busy!

Regards,
Simon Coulter.

--------------------------------------------------------------------
   FlyByNight Software         AS/400 Technical Specialists
   http://www.flybynight.com.au/

   Phone: +61 3 9419 0175   Mobile: +61 0411 091 400        /"\
   Fax:   +61 3 9419 0175   mailto: shc@flybynight.com.au   \ /
                                                             X
                 ASCII Ribbon campaign against HTML E-Mail  / \
--------------------------------------------------------------------