RE: survivability, rewriting

["marcelo bagnulo" <mbagnulo@ing.uc3m.es>]

> One could allocate a prefix, such a 0::/48 if you ignore that
> there is some other use for that, for use as the source locator.
> Thus the host could send packets with that source *IF* it knows that there
> is a rewriting router in the path which will insert the correct
> high-order bits in the source locator.
> But this is tricky to migrate to since the host would need to be able
> to tell whether its site has such rewriting routers at the boundary;
> if the 0::/48 makes it to a remote peer it will not be able to respond.
> How can it do this?

I guess that if a host has the M6 mechanism working, it would be reasonable
to assume that the site supports M6 and so that exit routers would rewrite
the source address.
But anyway you could also use a bit in the router advertisement to
communicate it to the hosts, right?

> Having the source send packets with the actual source locator prefix
> plus indicating that it is ok to rewrite it doesn't have this issue;
> if there is no rewriting router in the path the source locator
> stays unmodified but it still usable for the peer to respond.
>
> Second part is just the observation that having the node send packets with
> a fixed source prefix (such as 0::/48) doesn't remove the need for
> the ULP/apps to know the actual source locator, since it might need
> this as part of referals.
>

Agree but why this a problem? If i undestand correclty, Iljistch just
wanted:

"The obvious place to put an indication that the address may
 be rewritten is... in the address. "

I don't see why the second part is conflicting with Iljistch goal...

Regards, marcelo

>    Erik
>