On Thu, Nov 13, 2008 at 1:16 PM, Scott Brim <[EMAIL PROTECTED]> wrote:
> I was thinking about unilateral mode.  In Unilateral Mode, Six/One
> Router does a 1:1 translation at a site boundary, without involvement of
> mapping services or a requirement for reverse translation.  I get the
> feeling unilateral mode may be OBEd by NAT66, but I haven't asked Christian.

Outbound with clients behind the Six/One Router, that looks like plain
old client NAT. Client NAT is a fine thing, but it's already widely
deployed so it isn't likely to improve the routing table versus what
we have today.

Inbound with servers behind the Six/One Router, I'm pretty sure it
doesn't work. The raw networking part works. The fault isn't obvious
until you step back and take a holistic view, but it would require an
impractically complex split-horizon DNS system. Hosts on legacy
networks must find the provider-assigned address and hosts on Six/One
Router-served networks must find the PI address.

I say impractically complex because all DNS requests are relayed by a
caching resolver, usually on a different machine than the client host.
The AAAA DNS query may even come from an IPv4 address! As a result,
you can't just build a smart authoritative DNS server that hands out
different IPs depending on the request source address. And you can't
just err on the side of handing out the provider-assigned address:
that address is unreachable to any user behind a Six/One Router
because the user's Six/One Router will alter the address in the
response!

IMO, this isn't deadly to Six/One Router. There is such a dearth of
IPv6 deployment that one could reasonably just say that "legacy" IPv6
users are required to add a Six/One Router at their border.


>> Maybe that's an engineering compatibility difference like the
>> gyrations for PMTU in map-encap rather than an architectural
>> difference? What do you think?

I'm gonna back off on this assertion and the earlier one about
backwards compatibility not being an architectural issue. If I add the
following to strategy A, does it cover the strategies you've talked
about?

Strategy A.

Compatibility approaches include:

A4a. New IP protocol. Not compatible with IPv4 and IPv6.

A4b. Modified IP protocol. Not compatible with deployed IPv4 and IPv6.

A4c. Standard IPv4 and IPv6 packets are encapsulated in a tunnel
packet while they transit the core. Path-MTU issues are addressed by
setting an Internet-wide maximum packet size enforced by the encoders
and assuring that all core links support that size.

A4d. Standard IPv4 and IPv6 packets are encapsulated in a tunnel
packet while they transit the core. Path-MTU issues are addressed by
returning packets which breach the MTU while in the core to the
encoder who must act as a proxy by returning a sensible packet-too-big
message to the originating host.

A4e. The IPv6 address space is partitioned into end-user address space
and Internet core address space. The address to RLOC map is symmetric.
Part of the IPv6 end-user address is swapped for the RLOC when the
packet enters the Internet core and then restored when it leaves the
core. IPv4 is abandoned.


Regards,
Bill Herrin


-- 
William D. Herrin ................ [EMAIL PROTECTED]  [EMAIL PROTECTED]
3005 Crane Dr. ...................... Web: <http://bill.herrin.us/>
Falls Church, VA 22042-3004
_______________________________________________
rrg mailing list
[email protected]
https://www.irtf.org/mailman/listinfo/rrg

Reply via email to