Re: How to use an IP-header bit for Path MTU discovery.

mogul (Jeffrey Mogul) Mon, 26 February 1990 20:53 UTC

Received: by acetes.pa.dec.com (5.54.5/4.7.34) id AA02280; Mon, 26 Feb 90 12:53:34 PST
From: mogul (Jeffrey Mogul)
Message-Id: <9002262053.AA02280@acetes.pa.dec.com>
Date: 26 Feb 1990 1253-PST (Monday)
To: Philippe Prindeville <philipp@Gipsi.Gipsi.Fr>
Cc: MTU Discovery <mtudwg>
Subject: Re: How to use an IP-header bit for Path MTU discovery.
In-Reply-To: Philippe Prindeville <philipp@Gipsi.Gipsi.Fr> / Sun, 25 Feb 90 14:03:50 -0100. <9002251303.AA15531@gipsi.Gipsi.Fr>

	Would be it be so bad to require intra-subnet fragmentation/reassembly
	on tiny-gram subnets?  It could be done either at level 2 or 3...

This would have been the right thing to do a decade ago.  It is still
the right thing to recommend ... but
	(1) It's too late to change most of the tinygram data links,
	they're already "mature" techology and there probably will
	be few new ones devised

	(2) Changing the rules for these data links (essentially, changing
	the specs for how IPs are encapsulated) would be a major
	political undertaking.  I encourage you to form your own working
	group; it's outside the charter of this one.

-Jeff