Christian,

> Christian Huitema wrote:
> we want to remove ambiguity, which is the root cause
> of many problems occuring when scoped addresses leak.

If we want these addresses to be used, there are two things we need to
do:

1. Make these addresses globally unique, which is effectively removing
ambiguity. As of today, I don't see how we could achieve this without a
uniqueness database. It probably means some kind of a registration and
possibly a fee. We absolutely need to make the registration easy and the
fee low if there is one.

I invite people to have a look at:
http://www.ietf.org/internet-drafts/draft-py-multi6-gapi-00.txt
and its (temporary) results:
http://arneill-py.sacramento.ca.us/ipv6mh/geov6.txt
This scheme would use 1/64th of the FEC0::/10 space (replace 2346::/16
with FEFE::/16)

2. Make these addresses not globally routable, not only by decree but by
requiring them being blocked by default and also BGP routes for this
range being rejected by default. Ambiguity is somehow a guarantee that
these addresses are not publicly routable. If we remove ambiguity, we
need to provide something instead to address this.

Bob Hinden and I have contributed some interesting suggestions about
this earlier, but they were lost in the email volume. If my memory is
correct, Bill Manning was the only one to pick it; Bill, I would like
more of your comments.
 

> we may or may not want to prevent routing of these
> addresses between sites. I guess we should certainly
> prevent routing between non-consenting sites.

See above.
 

> we definitely want the addresses to be provider
> independent, so they can survive renumbering or
> intermittent connectivity.
 
Definitely.


> indeed, it would be desirable that the addresses
> be usable in sites that are not connected.

Yes.


>and we would definitely want the addresses to be free.

I think we have to accept the compromise of a low fee, as the uniqueness
database would likely be handled by the RIRs and we don't want to
bankrupt them.
 
> I guess that the consensus is, "just like site local
> addresses." We don't want to prevent usage in connected
> sites, but we expect that in these sites the hosts will
> also have provider based addresses, and that traffic
> routed out of the site will use the provider addresses.

Agreed.

Michel.


--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to