On Sun, Jan 28, 2001 at 02:22:46AM +0100, Andrew Miehs wrote: Hi Andrew, This is an interesting proposal. However I think the big problem is that if you do not have EUI-64 then you throw out address auto-configuration, which is a big win in too many situations to really just throw away. (Arp still exists in a way; it's sort of in neighbor discovery.) Also, a fixed boundary really helps when you are building routing software - it makes it simpler, cheaper and faster. Well, it *will* make it so :-)
- Running out of AS Numbers (prediction mid 2005) (It will become obvious later why I mention this here)
With the current proposal, I do not see that any of these problems will be adequately (or if at all) addressed.
This is certainly a more serious problem than it might at first appear, as several members of the community have said to me. The problem is not really the difficulty of preparing a new standard, or extending the existing in some way (although it's clear that there should *not* be a BGP 5 which extends AS space one month, and a BGP 6 which does v6 the next, for all the usual management reasons) -- the problem is that BGP-speaking core routers only get upgraded very rarely, and if we want to have a standard ready in time *we have to start now*. In fact it would be really great if we finished tomorrow :-)
The most obvious problem is that the 13 bits of sTLA space that is being given to providers today is TOO small.
There is a document being prepared on this (and other sTLA issues) as we speak.
There is no medium that I know of that can support 2^64 addresses in the one segment, and I see no advantage of shipping this information across the Internet.
Well, 64 bits is very handy because it nicely encapsulates ethernet and leaves room for future growth. Remember - the 3G people are coming, and they might use the phone number in the interface token, or might not, but it's useful to have more room anyway.
Mutlihoming and Renumbering are still very much up in the air.
You said it. That's definitely one of the things everyone would be much happier if it were sorted out as soon as possible. A good sign is that (apparently) an explicit IETF working group has been created for it. Your guess is as good as mine as to whether the standards process will produce a result quickly enough before 2005. (Of course you could always run this proposal past them as well.)
I would be interested in hearing any opinions,
I will have to look at this more closely (after IPv6 forum has closed...) Niall -- Enigma Consulting Limited 45 Dawson Street, Dublin 2, Ireland.