Walt,

> I believe that is an incorrect inference, Barbara. As I remember, that 
> documentation is specific to the case it describes, having an exactly 
> matching name (ABC) and that same exactly matching name but extended by the * 
> (ABC*). 
> 
> For the case of ABC and * RACF should still use ABC if the library 
> specification is appropriate. Of course, you can easily confirm that by using 
> some unimportant program and running the experiment to see.

thanks for the confirmation. But: I have no clue how to run an experiment on 
this. I guess I'll be keeping what's left in the program class. (The * profile 
with certain data sets that I have shown).

It's actually a pity that IBM is incapable of implementing the 'good stuff' (in 
this case enhanced program mode) in the things IBM delivers themselves. I am 
just trying to get a workable system out of the IBM delivered ADCD system. Or 
rather one that isn't wide open anymore with everybody having SPECIAL and 
OPERATIONS. The ADCD system doesn't even implement PROTECTALL, much less has 
dataset profiles for the system datasets. Not even for the APF auth'd datasets. 
And that is straight out of the box as delivered by IBM. I am not courageous 
enough to try to establish protectall security by defining the missing profiles 
- my chances of being left with an unusable system are too high, and in a 
monoplex, there isn't any testing ground other than the production system. Same 
with WHEN(PROGRAM) - once OMVS is screwed up, I cannot get at the system 
anymore.

Barbara

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to