Re: NAT Not Needed To Make Renumbering Easy

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

 



Andrew G. Malis wrote:

> One thing that IPv6 NAT has in advantage to IPv4 NAT is that it can be
> stateless, isomorphic, and port transparent by just translating the
> upper part of the address,

Not at all. Unless the NAT have end to end transparency,
statefull trasnration of raw IP addresses in payload is
still necessary.

Worse, port transparency is not very useful for IPv6, because,
transport check sum to be modified may be located in the second
or latter fragment, which means stateful reassembly is necessary
for IPv6 NAT.

And, remember to say IPSEC, which is *REQUIRED* by IPv6.

> This allows easy multihoming without needing to punch holes
> in ISP address blocks.

Wrong. It has nothing to do with NAT nor IPv6.

Punch holes are not necessary, if hosts have multiple addresses
assigned from ISPs *AND* transport and/or application layer of
their peers try all the IP addresses of the hosts, which is the
case with DNS/SMTP but not UDP/TCP and almost all the applications.

If we can fix transport/application over IP, IPv4 does not need
punch holes.

Of course, multiple addresses may be held not by individual end
hosts but by a middlebox (either IPv4 or Ipv6) and only UDP/TCP
of the middlebox may be modified, *IF* you sacrifice end to end
transparency.

						Masataka Ohta


_______________________________________________

Ietf@xxxxxxxx
https://www.ietf.org/mailman/listinfo/ietf

[Index of Archives]     [IETF Annoucements]     [IETF]     [IP Storage]     [Yosemite News]     [Linux SCTP]     [Linux Newbies]     [Fedora Users]