On 2009-11-17 21:20, Scott Brim wrote: > Brian E Carpenter allegedly wrote on 11/17/2009 7:52 AM: >>> We need the a box that translates local IPv6 addresses into global IPv6 >>> addresses using a stateless NAT using a one-to-one reversible mapping >>> algorithm for single-homed edge networks. That is precisely what NAT66 >>> as described in the draft does according to our understanding. >> Clarification: when I say 'reversible', I mean that the translation >> is reversible before the packet is delivered to the transport layer at >> the remote end (as in 8+8/ILNP or SHIM6). >> >> Obviously, any NAT has to perform its local mapping in both directions, >> but tha's not what I meant. > > I am concerned about requiring the remote site to be able to reverse the > sender's mappings. For that to be the case, everyone must be able to > run the same algorithm for translation, and changing it will be very > very difficult, at least as difficult as changing IP.
To be pedantic, the condition is a little weaker: any given pair of correspondents must run the same algorithm for translation. But I agree that it is very tricky, and I think any viable solution will require e2e signals of some kind. encap/decap is certainly more robust. Brian _______________________________________________ rrg mailing list [email protected] http://www.irtf.org/mailman/listinfo/rrg
