Subject: Re: chpass assumptions
To: John Nemeth <jnemeth@victoria.tc.ca>
From: Luke Mewburn <lukem@NetBSD.org>
List: current-users
Date: 04/11/2005 09:12:31
--4N29bJelYIMOTi0e
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline

On Sun, Apr 10, 2005 at 01:53:46AM -0700, John Nemeth wrote:
  |      chpass appears to be aimed at updating "user database information"
  | stored in either YP/NIS (it's manpage should be updated to say NIS) or
  | /etc/passwd.  At one point, its code checks NIS, then has a comment
  | saying that since the user wasn't found in NIS, check the local
  | database, and then it promptly calls getpwnam().  For quite sometime,
  | NetBSD has supported the remote storage of "user database information"
  | in the form of Hesiod.  With the advent of dynamic nsswitch support,
  | "user database information" could quite literally be stored anywhere.
  | What do we want to do about the grossly out of date assumption that
  | getpwnam() only looks in /etc/passwd and NIS?

Document it that way ...

Seriously; there's not much we can do in an nsswitch world.
Except possibly simplify chpass and say that it's a "local files only"
tool (i.e, remove NIS support).

--4N29bJelYIMOTi0e
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.6 (NetBSD)

iD8DBQFCWbLfpBhtmn8zJHIRAkhhAJ0dPQ49JPrCrF0P0lSe7fcZRQ3TLACferI5
6HgRdrhEEYdQMiBxF32hH9A=
=ts+z
-----END PGP SIGNATURE-----

--4N29bJelYIMOTi0e--