Re: Source address selection insufficient?

[Masataka Ohta <mohta@necom830.hpcl.titech.ac.jp>]

Joel M. Halpern;

> I think that this is getting more complex than necessary.

Because of the introduction of connection at the IP layer, yes.

The issue is simple if you accept the IP layer is connectionless
and unidirectional.

> It is true that routing is frequently asymmetric.
> However, that is a very different statement from saying that 
> connectivity is asymettric.  While the routing path among ISPs for a 
> given address pair (src, dst) may be different in the forward and 
> reverse direction, it would take a very strange situation for a pair to 
> work one way, and not when reversed.  It would take an even stranger 
> situation for there to be no pair that worked in both directions.
>
> As such, I think we ought to be able to assume that there exists (at any 
> given time) an address pair that is useable in both directions.

That is a useful property because it makes ICMP returned most of
the time, though we shouldn't rely on them too much (that is,
PMTUD is not a good idea).

However, as the destination locator is chosen by the source
local policy, bidirectional transport layer can not assume
the locator pairs are same for the both directions.

							Masataka Ohta