[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Misusing registries for uniqueness

On 2003-05-30 at 11:09, Robert Elz wrote:

> Even adding in registries, as Benny Amorsen proposed, doesn't really
> solve the problems.  To be unique, there'd need to be one overall
> registry (perhaps sub-dividing the space to others) - which again means
> one monopoly, with no real justification for anyone to be the owner of
> that.

The advantage of using multiple registries is that the problems are
obvious to everyone. No one ends up thinking that their number is
guaranteed to be unique. 
As you point out later, the addresses /cannot/ be enforced to be unique.
There were organisations which used 9/8 or 192/8 (not 192.168/16)
addresses for site local addressing in IPv4. An Internet Standard cannot
force people to comply.

I do not think there should be one top-level registry that subdivides
space to other registries. Rather I think each registry should pick
whatever bit of address space they feel like, within a single global
allocation for site-local addresses. If two registries pick the same
address space, too bad. Either they figure out a way to reconcile, or
customers will avoid them. If a registry takes too much space, the other
registries will ignore them.

Either way, anyone can set up their own registry of site-local IPv6
numbers. I think the working group should try to stay out of the fray,
except for setting aside a reasonably large space for it. A /16 would
seem appropriate.

Should site-local addresses end up not being standardised, there is
always 10/8 mapped with 6to4...


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 majordomo@sunroof.eng.sun.com