[916] in arla-drinkers
Re: User level permissions
daemon@ATHENA.MIT.EDU (Chris Wing)
Mon Jun 28 13:20:18 1999
From owner-arla-drinkers@stacken.kth.se Mon Jun 28 17:20:17 1999
Return-Path: <owner-arla-drinkers@stacken.kth.se>
Delivered-To: arla-drinkers-mtg@bloom-picayune.mit.edu
Received: (qmail 14550 invoked from network); 28 Jun 1999 17:20:14 -0000
Received: from unknown (HELO sundance.stacken.kth.se) (130.237.234.41)
by bloom-picayune.mit.edu with SMTP; 28 Jun 1999 17:20:14 -0000
Received: (from majordom@localhost)
by sundance.stacken.kth.se (8.8.8/8.8.8) id TAA00798
for arla-drinkers-list; Mon, 28 Jun 1999 19:14:27 +0200 (MET DST)
Received: from shaft.engin.umich.edu (wingc@shaft.engin.umich.edu [141.213.33.85])
by sundance.stacken.kth.se (8.8.8/8.8.8) with ESMTP id TAA00789
for <arla-drinkers@stacken.kth.se>; Mon, 28 Jun 1999 19:14:18 +0200 (MET DST)
Received: from localhost (wingc@localhost)
by shaft.engin.umich.edu (8.9.3/8.9.3) with ESMTP id NAA30352;
Mon, 28 Jun 1999 13:14:14 -0400
X-Authentication-Warning: shaft.engin.umich.edu: wingc owned process doing -bs
Date: Mon, 28 Jun 1999 13:14:14 -0400 (EDT)
From: Chris Wing <wingc@engin.umich.edu>
To: Johan Danielsson <joda@pdc.kth.se>
cc: arla-drinkers@stacken.kth.se
Subject: Re: User level permissions
In-Reply-To: <xofzp1k5ovi.fsf@blubb.pdc.kth.se>
Message-ID: <Pine.LNX.4.10.9906281258390.30322-100000@shaft.engin.umich.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-arla-drinkers@stacken.kth.se
Precedence: bulk
Johan:
> > It's worth solving if you want to emulate the behavior of the
> > original AFS.
>
> Ok. I've never felt I had to do that (but we're running as little as
> possible of the transarc stuff).
I agree that emulating every minor detail is a waste of time. However,
regarding the output of 'tokens', printing out the correct AFS ID is a
useful feature in my opinion, and I think it should be implemented in
Arla/KTH-KRB.
I don't even particularly like the behavior and syntax of the programs
from official AFS. 'pts' only performs one operation at a time, while 'fs'
gives you a command prompt. Some of the options are poorly worded ("klog
-tmp", for instance). Howover, having a consistent interface across
different platforms is _very_ important to users. At U-M we have thousands
of Unix workstations running Transarc AFS, and if "klog" doesn't work as
expected on the Linux/BSD machines running Arla, a lot of people will be
confused and annoyed.
> > Like I said, is anyone else interested in a clone of klog that would
> > do this and follow the syntax of official AFS?
>
> Sure. If you can make it reasonably clean, we could add this to kauth
> too (perhaps as a compile-time option or something).
There is a slight chicken-and-egg problem there; you can't build Arla
without installing a Kerberos library, but you'd need arlalib available in
order to build kauth.
I think the right thing to do is to make sure that a working klog is
bundled with Arla.
-Chris
wingc@engin.umich.edu
