Itojun,
I subscribe to the model that all nodes may indeed be routers.
This comes mainly from my background in MANET, but I see
the paradigm as possibly appropriate here as well. If each node
in the site were a router, we would be sending an RS in the
expectation of getting back an RA with more-specific routes
and (in most cases) zero in the router_lifetime field.
We will certainly still have our default router(s) as told by the
Potential Router List. But, if we view the global DNS as an
extension to the PRL, we can select ISATAP routers for more
specific routes on an on-demand basis.
Sorry for not saying all of this at the microphone yesterday, but
I was preoccupied with other matters at the time.
Thanks - Fred
[EMAIL PROTECTED]
[EMAIL PROTECTED] wrote:
>I have not studied this space, but it occurs to me that ISATAP
>could be tried as a first alternative to check whether the two
>hosts are separated by a NAT. If there is no intervening NAT,
>it seems to me that ISATAP would provide the benefit of not
>needing the UDP header and "bubble" packets, yielding
>greater efficiency. Otherwise, if blocked by a NAT the
>initiating host coud after a short timeout try again with
>Teredo.
ISATAP requires an ISATAP router in a site to advertise RA in a
pretty weird fashion, which is clearly a drawback.
itojun
