On Tue, 18 Jun 2013 08:30:46 -0500, Joel C. Ewing wrote:
>
>... This makes me suspect that the current IBM
>implementation of associating the home path with a RACF userid profile
>rather than somehow with a UID-related profile violates the spirit if
>not the letter of the UNIX standards.
>
Yup. But it has been written in these lists that there are other UNIX
implementations that likewise "violate" the spirit. IBM could still
backfill with a PARMLIB option, "BPX_UIDS_UNIQUE={YES|NO}",
and enforce uniqueness at OMVS segment definition if selected.
IBM has some uncommon facilities such as the _BPX_SUPERUSER
attribute to remove the need to give multiple users UID 0.
>> On Tue, Jun 18, 2013 at 5:53 AM, Ted MacNEIL wrote:
>>
>>> I believe it's based on the first id retrieved from the RACF database that
>>> matches.
>>>
I understand it's cached. If only one OMVS segment has been touched
lately and all the others have aged out, you'll get that one. Otherwise
it's more unpredictable.
>>> -----Original Message-----
>>> From: Adam
>>> Date: Tue, 18 Jun 2013 05:38:56
>>>
>>> We have a system where two RACF userids are defined with the same uid.
>>> (This is deliberate and is intended to simplify access using NFS and other
>>> OS.)
>>>
Don't do that. It hurts. If you want data to be sharable, that's what GIDs
are for.
>>> When there are two (or more) RACF userids with the same uid in the OMVS
>>> segment, how is the value in username determined?
>>>
Years ago, there was some discussion here (perhaps Walt Farrell contributed)
that IBM seemed to be committed to making it predictable: the oldest OMVS
segment having a given UID would always be returned. I suggested that it
would be more useful to return the OMVS segment most recently updated
because then an administrator could select from several one to be preferred
simply by making a trivial change to that OMVS segment. This was a distinct
minority view.
In any case, neither ever happened.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
