Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 16 Feb 1998 13:32:40 -0500 (EST)
From:      Robert Watson <robert@cyrus.watson.org>
To:        Charles Owens <owensc@enc.edu>
Cc:        hackers list FreeBSD <freebsd-hackers@FreeBSD.ORG>, Peter Braam <braam@cs.cmu.edu>
Subject:   Re: Coda FS: FBSD port done!, but development favors Linux
Message-ID:  <Pine.BSF.3.96.980216132221.891F-100000@trojanhorse.pr.watson.org>
In-Reply-To: <Pine.BSF.3.95q.980216093920.24408B-100000@itsdsv2.enc.edu>

next in thread | previous in thread | raw e-mail | index | archive | help
On Mon, 16 Feb 1998, Charles Owens wrote:

> It would seem that my language used in starting this thread carried some
> un-intended connotations.  I was _attempting_ to say:
> 
>    Coda is cool, but development _such_as_it_is_occurring_right_now_
>    is focusing on a particular OS (that just happens to be Linux)... 
>    Let's make sure that that FreeBSD remains a viable Coda platform
>    (since FreeBSD is what we use)

While a lot of public discussion and development is happening under Linux,
the impending availability of Coda under various BSDs should even things
out a bit.  I have every intention of continuing my development on Coda on
FreeBSD, and there is strong support for FreeBSD as a platform in the Coda
group.  I personally, as I have mentioned, do my authentication-related
development under FreeBSD.  One cutting-edge area of Coda right now is the
PAG implementation under Linux -- this will help determine the future
directions of local management of Coda tokens and authentication.  I think
it would be highly appropriate that we have matching developments in
FreeBSD, although possibly with more eye towards other possible uses (such
as IPsec, SSH, AFS, other token/key systems).

> I can see, however, that my comments could have been construed as if
> intended to urge reaction against some supposed "Linux-camp" advancement
> in a Linux/FreeBSD war.

I think that some of the response was concern over the possibility of
platform-related disagreements as the merging of the Coda sub-communities
occurs.  Peter's choice of words in inviting others to join the existing
Coda development lists was a warning of a possibility to avoid rather than
that this had already occurred.  Specific avenues we hope to avoid, I
think, are "We would never do something like that in FreeBSD".

> Nope... nothing so diabolical.  I simply would like to implement Coda
> using FreeBSD clients and servers (since that is what I personally
> prefer), and therefore wanted to start some discussion about what I saw as
> potential roadblocks to that happening.  I am, as well, very encouraged to
> see the level of interest in Coda shown hereabouts.

I have similar hopes!  Coda is not yet ready, we feel, for a production
environment, but we hope to move it in that direction as rapidly as
possible.  On the other hand, it is important that design decisions be
made correctly now while there is no solid userbase, and while we are not
yet committed to a particular protocol arrangement.  For example, I am
changing the behavior of the RPC2 Bind in one of the upcoming releases in
such a way that old clients and servers will no longer interoperate with
new ones -- however, it will no longer be subject to a particular class of
replay attacks, and will support Kerberos authentication (among other
things).  Now is the time to make important design decisions -- I think
that the FreeBSD community has a lot to add to the process.  The level of
enthusiasm thus-far seen is very encouraging.


  Robert N Watson 

Carnegie Mellon University http://www.cmu.edu/
SafePort Network Services  http://www.safeport.com/
robert@fledge.watson.org   http://www.watson.org/~robert/


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.980216132221.891F-100000>