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

RE: WPD-6, WAA-8, and WAA-9 of draft-ietf-v6ops-ipv6-cpe-router-04



> it doesn't. the need for the 'unnumbered' WAN interface does.
> I don't know of any router implementation which doesn't implement the
weak host model, but I'm not 
> aware of that being specified anywhere either.

Shin Miyakawa from NTT explained to me one time that they had people
implementing CPE routers based on a Windows Vista stack that used the
strong host model.  This presented quite some difficulty for NTT as an
operator - as they had to restrict themselves in what they could do in
terms of address space conservation in order to support a strong host
model CPE router - and this warning was in a previous version of the CPE
router specification.  We should actively discourage people from
building and using strong host model CPE routers.

This was the previous text:

"The Unnumbered model is incompatible with the strong host model
[RFC1122] on the CPE router.  The unnumbered model may be inappropriate
for use with certain deployments where a device that uses the strong
host model can operate as a CPE router."

- Wes