We have seen this for too long.  We should not wait on this SL figuring it out anymore rip them out of IPv6.
 
/jim
-----Original Message-----
[EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, June 11, 2002 9:47 AM
To: [EMAIL PROTECTED]; [EMAIL PROTECTED]
Subject: Re: Fwd: IPv6 Scoped Addresses and Routing Protocols

On Tuesday, 06/11/2002 at 01:47ZE10, [EMAIL PROTECTED] wrote:
> Putting limited-scope addrs in DNS is a bad idea *unless*
> you have a way to uniquely identify the scope.
>
> Neither AAAA or A6 provide support for this.  I made a proposal
> that would have added this to A6 *before* there was any A6
> deployment.  It got totally ignored by the IPv6 community (with
> the exception of Matt).
>
> Now might be a good time to raise the idea again except I won't
> tie it to A6.
>
> <ownername> SA <IPV6 address> <scopename>
>
> SA  scoped address
> scopename is a domainname choosen to be globally unique.
>  global addresses have a scopename of ".".
>
> the conversion from scopename to scopeid / zoneid is a
> *local* resolution problem for the machine.


So, to make SL work, we need changes to the DNS name server, changes
to the resolver, routers to advertise the scope zones, hosts to
learn the scope zones from routers and include them on DDNS
registrations and use them in making routing decisions, updates to
routing protocols, IANA needs to manage a new zone scope naming
space, and who knows what else.

Wouldn't it make sense to hold off on SL until we at least have
some proposals on the table which describe the changes necessary to
make SL work, at which time we could have a discussion on the
technical merits of the proposal(s)?  Until that time, though, the
SL unicast address space should either be reserved or moved to
experimental.

Roy

Reply via email to