The IP over ATM Mailing List Archive by date

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

Bryan,

(omissions are for brevity)
> A new version of the encapsulation Internet-Draft has been
> submitted. It is <draft-armitage-ipatm-encaps-01.txt> ...
> I no longer believe that the new SSCS option is the best 
> solution to the problem, although it is included as option 5. ...
> So I think the best options are 3 and 4 (you'll need to read the
> draft). One differentiates based on OUI, the other on PID.

I read through the draft and thought a bit about the implications of
each.  I completely agree that while you included options 1, 2, and 5
for completeness' sake, they are not suitable solutions.  Between
options 3 and 4, as you mentioned, it's a harder choice.  One of the
implications of option 3 is that the entire PID space for the new OUI
is basically used up by completely replicating the PID space for OUI
00-00-00, while option 4 only requires one new PID to be allocated.

FYI, I checked with the IANA to see if there were any current uses of
the IANA's OUI in SNAP headers, and they said:

   The answer is "no", we are not aware of any uses of the IANA's OUI
   (00-00-5E) as the OUI in SNAP headers, and the assigment of PIDs to
   go along with the OUI.

So while the space is available, I would hate to ask them to allocate
it all in one fell swoop!  So, I've come to the conclusion that option
4 is best, and we should ask the IANA to allocate us one PID in its
OUI space for this use.

Regards,
Andy Malis
__________________________________________________________________________
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