mechanisms for client to determine which username a server uid represents

[Gerald (Jerry) Carter]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Steven French wrote:
|
|
| I was thinking along the lines of
| 1) extended session setup response when Unix like
| server (perhaps POSIX extensions) negotiated.
| and
| 2) an smb trans2 which given a tid and uid (server uid,
| not smb uid) returned the full qualified username.
| Unfortunately this is in an almost empty trans2
| request but we have only the Unix range of
| trans2 requests at our disposal to extend for purposes
| like this

But neither of these solve the general problem from
what I can tell.  You will still have uid's that
are unmatched.  I guess maybe we are talking about
different problems here.

I'm thinking of the case where winbind replaces NIS
and LDAP.  The name2uid and uid2name calls are
provided via NSS.  winbindd would query the Samba DC
for the SID<->{g|u}id lookups.  User space processes
like 'ls -l' would just report the correct thing.

When I describe this I realize I may have missed the
exact problem you are trying to solve since I can't
see how this relates to a kernel vfs at all.









cheers, jerry
=====================================================================
Alleviating the pain of Windows(tm)      ------- http://www.samba.org
GnuPG Key                ----- http://www.plainjoe.org/gpg_public.asc
"I never saved anything for the swim back."     Ethan Hawk in Gattaca
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFCf8BWIR7qMdg1EfYRAmOEAKCy85ZDsrnRQgWeF7qrOuJ4Ot2ZLACghh0c
WH+WDhe4Wo3CAk+RddZJqXk=
=xD27
-----END PGP SIGNATURE-----