Re: Fwd: Teredo and Tunneling WGLC

[Rémi Denis-Courmont <rdenis@simphalempin.com>]

> > From: Fred Baker <FredBakerSBA@gmail.com>
> > Date: August 4, 2008 4:22:05 PM JST
> > To: v6ops@ops.ietf.org
> > Cc: Kurt Erik Lindqvist <kurtis@kurtis.pp.se>, Ron Bonica
> > <rbonica@juniper.net
> >
> > >, Mark Townsley <townsley@cisco.com>
> >
> > Subject: Teredo and Tunneling WGLC
> >
> > This is to initiate a two week working group last call of two
> > documents.
> >
> > http://tools.ietf.org/html/draft-krishnan-v6ops-teredo-update
> >   "Teredo Security Updates", Dave Thaler, Suresh Krishnan, James
> > Hoagland,
> >   14-Jul-08, <draft-krishnan-v6ops-teredo-update-03.txt>

I have already expressed myself on this one here (no reply):
http://www.ops.ietf.org/lists/v6ops/v6ops.2008/msg01240.html

> > http://tools.ietf.org/html/draft-thaler-v6ops-teredo-extensions
> >   "Teredo Extensions", Dave Thaler, 14-Jul-08,
> >   <draft-thaler-v6ops-teredo-extensions-01.txt>

And here http://www.ops.ietf.org/lists/v6ops/v6ops.2008/msg01183.html

I am also concerned that the new message formatting that this proposes is not 
backward compatible with RFC4380. Depends how one interprets RFC4380, page 
22:

   An IPv6 packet is deemed valid if it conforms to [RFC2460]: the
   protocol identifier should indicate an IPv6 packet and the payload
   length should be consistent with the length of the UDP datagram in
   which the packet is encapsulated.

draft-thaler-v6ops-teredo-extensions implicitly assumes that a "consistent" 
length means a smaller or equal length. I know deployed Teredo relays which 
considers a "consistent" length to be stricly equal. As far as I can tell, 
such a Teredo node will not interoperate with a 
draft-thaler-v6ops-teredo-extensions-01 Teredo node. It is debatable whether 
such a legacy node is broken. In any case, this issue should at the very 
least be mentioned in the document, in my humble opinion.

-- 
Rémi Denis-Courmont
http://www.remlab.net/