RE: issue 11 -- window size

[email protected] writes:
> Mohan Parthasarathy wrote:
> 
> > > - In MOPO-IKE, path test is a separate message primarily
> > >   because then it cannot interfere with anything else. Here, 
> > >   host A would not receive B's reply to the PATH_TEST 
> > >   message, and traffic would stay on path (A2,B2) which 
> > >   is working fine.
> >
> > If the window size is 1, you have to retransmit the last message 
> > with a new address (discovered through PATH_TEST message in your 
> > case). If  "A" retransmits the message with new address (A1, B1), 
> > why does'nt B move all traffic to path (A1, B1) ? How is this 
> > prevented in MOPO-IKE ?
> 
> In the example I gave the current path (A2,B2) was working all the
> time, so there's no need to retransmit the last message over (A1,B1).
> 
> But if (A2,B2) stops working, and we discover that (A3,B3) works,
> then MOPO-IKE retransmits any outstanding requests using (A3,B3).
> If NAT-T was enabled, this can cause the responder to move all traffic 
> to (A3,B3).... but in this case it's not a problem, because that's
> what we wanted to do anyway! 

Same happens in my proposal too. When the A does not get answer back
from the (A2,B2) of its update message, it starts path testing, and
finally finds the (A3, B3) that works, and uses that. Now depending on
the protocol the B has already moved traffic to (A3, B3), or A
manually does that after noticing the situation. 
-- 
[email protected]