The IP over ATM Mailing List Archive by date

[Andy Malis <malis@enigma.acton.timeplex.com>]

I'm going to answer both Mark and Keith in the same message ...

Mark Laubach writes:
> Andy, please clarify, do you mean a morf'd option 3 & 4?; i.e.:
> 
>   [0xAA-AA-03] [0x00-00-5E] [New PID] [2octet CMI] [Old PID] [MP Packet]
> 
> Where: [New PID] is the IANA single allocation, and 
>        [Old PID] is an Ethertype

Yes, I do - thanks much for pointing this out.  I should have been
more explicit in my message, but I've been thinking about the new OUI
to the extent that I forgot that option 4 originally suggested using a
PID in 00-00-00.

The reasons I don't like using a PID in the 00-00-00 space are that
it's really NOT an Ethertype, there's little or no precedent for
further network layer protocol demux once you've demuxed on an
Ethertype in the 00-00-00 space, and I thought it would be pretty neat
for us to be the first users of the IANA's OUI-PID space.

Keith McCloghrie writes:
> I agree that option 4 is the best of the bunch.  To extend your
> reasoning, it seems to me that MARS is a new protocol - so, using a new
> PID value, rather than a new OUI value, is more appropriate, as well as
> "cheaper", to indicate the use of MARS.

Well, since MARS is a REALLY new protocol, there's no reason to not
use a new OUI as well! :-)

If the WG would like to stay in OUI 00-00-00 (pure option 4), that's
really OK with me too.  I much prefer either pure option 4 or my
"morf'd option 3 & 4", as Mark so accurately put it, to any of the
other alternatives.

Cheers,
Andy
__________________________________________________________________________
Andrew G. Malis   malis@maelstrom.timeplex.com             +1 508 266-4522
Ascom Nexion      289 Great Rd., Acton MA 01720 USA   FAX: +1 508 266-2300