User-agent: Mozilla/5.0 (Macintosh; U; PPC Mac OS X Mach-O; en-US; rv:1.5) Gecko/20031007
So in the Internet architecture, IP addresses serve at least these three
functions:
1 - Identify an end-end entity
2 - Describe where its interface(s) is in the network (location)
3 - Serve as a forwarding tag for packets.
I think this is a very important point, and worth pursuing much more.
(I am also looking forward to more brain torque with more functions
of IP addresses...) To start with, a small observation:
- In a vanilla IP-layer router, the forwarding tag is the
destination address.
- In a QoS-enabled router, the forwarding tag is something more,
e.g. <dst addr, flow label>
- In a NAT box, the forwarding tag depends on the direction of the
traffic, and for inbound traffic it is typically <dst, proto, dport>
but may be even smaller (e.g <proto, dport>) or larger
(<src, dst, proto, sport, dport>), depending on implementation.
[I hope I got it right, I am not a NATologist.]
And the maybe more important one:
- If IPsec is used, or if a new "session ID" is introduced (as in SIM),
the <dst addr, SPI> or <dst addr, session ID> could be used as a
forwarding tag, thereby enabling cross-realm communication.
Hence, the important question is whether we want to limit our
considerations to solutions where the forwarding tag is solely
the IP address or whether we want to consider the cases where
it actually is or can be something more. A related question is
whether it is acceptable to rewrite forwarding tags on the fly.
--Pekka Nikander
_______________________________________________
Saad mailing list
Saad@ietf.org
https://www1.ietf.org/mailman/listinfo/saad