Re: GRTOBJAUT...does anyone know

[Vern Hamberg <vhamberg@xxxxxxxxxxxxxxx>]

I think you are better off getting a list of those that need the change. 
I suggest using SQL - first use one of the security services that can 
filter by authorities in a CTE - common table expression. Use columns 
from the CTE in a SELECT that uses the QCMDEXC function to execute the 
GRTOBJAUT command. Here is a simple example that you might be able to 
modify - you would probably use :

WITH FF AS (SELECT TRIM(TABLE_SCHEMA) || '/' || TRIM(TABLE_PARTITION) 
PGMNAME,
               TRIM(TABLE_SCHEMA) || '/' || TRIM(TABLE_NAME) SRCNAME
            FROM SYSPARTITIONSTAT
            WHERE TABLE_SCHEMA = 'VHAMBERG' AND TABLE_NAME = 'GOODBADSRC')
    SELECT SRCNAME,
           PGMNAME,
           CASE
             WHEN QSYS2.QCMDEXC('CRTBNDCL PGM(' || PGMNAME || ') 
SRCFILE(' || SRCNAME || ')') = 1
               THEN 'Compile succeeded'
             ELSE 'Compile failed' END AS COMPILE_RESULT
    FROM FF;

*Regards*

*Vern Hamberg*

IBM Champion 2025 <cid:part1.H291BU20.6TspYA3O@centurylink.net> CAAC 
(COMMON Americas Advisory Council) IBM Influencer 2023


On 5/11/2026 8:12 AM, smith5646midrange@xxxxxxxxx wrote:
> We are in the process of cleaning up / tightening up security on our
> production machines.
>
>   
>
> One of the many libraries has about 8,000 objects in it and all but about
> 100 have the correct authorities.  If I do a GRTOBJAUT for *ALL objects in
> the library, does anyone know if the command is smart enough to only look at
> and only adjust the objects that do not have the particular supplemental
> group or does it try to adjust all 8,000?  I'm trying to figure out if it
> would be more efficient to build a CL with just the objects that are
> problems.
>
>   
>
> And please do not suggest using authorization lists.  I'm working toward
> that but trying to get a handle on why the authorities are what they are
> (and that would politely be called "a mess") has been like trying to turn a
> barge with an oar.  I'm slowing getting there but is it taking forever so we
> are taking an intermediate step first.
>
>   
>
>