[EMAIL PROTECTED] writes:

> I was attempting to get to the bottom of what is, to me, a mystery:

> Why does IPv6 use EUI-64 for Interface Identifiers, instead of MAC-48?

> The previous version of the RFCs used MAC-48. There seems to have been,
> at some point, some discussion regarding the difference between EUI-48 and
> MAC-48, as relate to their use in EUI-64.

> And the mailing list archives only go back to some time in 2003, while the
> RFCs are 1998 (current) and 1996 (replaced by 1998 versions).

The IPv6 WG was (previousl) the ipng WG. See the archives at
ftp://ftp.ietf.org/ietf-mail-archive/ipngwg/

> Did we collectively just throw away 16 bits of usable space from our 128,
> for no good reason?

It kind of looks that way today. :-(

> If someone can explain the rationale, that'd be great.

Well, there were multiple ones. But the main one I recall was that it
made the address format more "friendly" to the 8+8/GSE proposal, with
the hope being that maybe we could still develop 8+8/GSE and have it
work with easily with IPv6. As we know now, that never happened.

The other factor was that there was the thought that future LAN
devices would use 64-bit MAC addresses, i.e., EUI-64
identifiers. Thus, we simply made stateless address autoconfiguration
forwards compatible.

The reality, of course, is that the Ethernet frame format continues
to dominate networking at the edges, and probably will more-or-less
forever.

> (I suppose it's a bit late in the game to go back to MAC-48 as II?)

Yes. The difficulty is figuring out how to move to something different
in a backwards-compatable way. I.e., we'd need a transition strategy.

Thomas

--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------

Reply via email to