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

Re: GSE IDs [Re: IETF multihoming powder: just add IPv6 and stir]

On dinsdag, mei 13, 2003, at 17:26 Europe/Amsterdam, Christian Huitema wrote:

Mobile IP in IPv6 uses a 24 byte header to carry the original source
address, right?



It does, but it only needs to do so when negotiating a binding update.
The trick is to design a usage model of MIPv6 in which binding updates
are used to redirect a TCP connection (or a UDP flow) to a new address.




Christian, why did you have to do this???



All this time I managed to avoid reading the MIPv6 draft (no RFC 
yet...) but now I had to in order to be able to disagree with you.



On page 78 it says that there must be state in the binding cache AND 
the correspondent has to use a routing header. So there's something to 
hate there for everyone.



(I know I was talking about the source and not the dest address but 
that one is significantly more convoluted in the spec. Looks like the 
same thing applies, though.)



Unfortunately it isn't entirely trivial to remove the routing header 
and have the identifier^H^H^H^H^Hhome address be implied as this 
requires a mechanism to detect whether something is directed at the 
home address or the care-of address, which we need to know to do the 
checksums. Who's bright idea was it to include the addresses in the 
transport checksums anyway???