Re: decreases in MTU

Philippe Prindeville <philipp@Gipsi.Gipsi.Fr> Mon, 05 March 1990 17:27 UTC

Received: from decwrl.dec.com by acetes.pa.dec.com (5.54.5/4.7.34) id AA16111; Mon, 5 Mar 90 09:27:48 PST
Received: by decwrl.dec.com; id AA05637; Mon, 5 Mar 90 09:27:32 -0800
Received: from [192.33.166.11] by inria.inria.fr (5.61+/89.0.8) via Fnet-EUnet id AA18531; Mon, 5 Mar 90 18:23:20 +0100 (MET)
Received: by Gipsi.Gipsi.Fr (4.12/4.8) id AA15979; Mon, 5 Mar 90 18:23:31 -0100 (MET)
Date: Mon, 5 Mar 90 18:23:31 -0100
From: Philippe Prindeville <philipp@Gipsi.Gipsi.Fr>
Message-Id: <9003051723.AA15979@Gipsi.Gipsi.Fr>
X-Phone: +33 1 30 60 75 25 / +33 1 47 34 42 74
To: deering@pescadero.stanford.edu
Subject: Re: decreases in MTU
Cc: MTU Discovery <mtudwg>

It would (should?) not be hard to mandate that all sub-576 MTU paths
be the first to employ the new type of Can't Fragment message.  Anyway,
there can't be all that many such paths (I hope).  The only one I can
think of would be ARCnet (with a 511-byte MTU) -- and now I feel guilty
for bring IP to it (or rather, bringing ARCnet to the IP community).
Fortunately, it is seldom (if ever) used as a transit network...

-Philip