On Mon, Nov 06, 2006 at 02:21:37PM +0000, zarrar yousaf wrote:
> 
> Salah, thinking from another angle, i think the use of a HAO is
> universally more appealing especially during circumstances when the
> CN also happens to be a mobie entity present in some foreign
> network, communicating with our MN.In such a case both the MN and
> the CN will contain BCE (Binding Cache Entry table) and Binding
> Update List (BUL)
>
> Any comments ?

As already pointed out, Home Address Option and Routing Header Type 2
are two different things.  

HAO tells, "this is my real address, although that other address
appears on the source field."  Routing header tells, "This packet must
go via this address before ending up in the final destination."  If
there wasn't a HAO, destination would not know who the real sender
was.  If there wasn't a RH, packet could not be routed directly from
CN to MN.  If both ends are in fact MNs, you just use both.  Or you
don't use either, and you'd still have mobility but without route
optimization.  I really don't see any problem here, or why you should
talk about "RH vs HAO".  Neither one can replace the other.

A little loosely related history: A few years back, MIPv6 still used
Routing Header Type 0.  Then it was decided that type 0 was not
restricted enough.  Type 2 was added so there could only be one
intermediate destination before the final destination and the
restriction that intermediate destination and final destination must
be on the same host.  And if someone wanted to use source routing, it
still could be done with type 0, without interfering with MIPv6 RH
Type 2 use.  Also, there were IPsec considerations that made type 0
less desirable.

Regards,

Antti

_______________________________________________
mipl mailing list
[email protected]
http://www.mobile-ipv6.org/cgi-bin/mailman/listinfo/mipl

Reply via email to