[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Posted a new copy of CPE Rtr draft
On 20 jul 2009, at 19:34, james woodyatt wrote:
Careful with that axe... if we're going to require prefix sub-
delegation in best-practice CPE routers, then I contend we
absolutely need to specify how that's done with zero configuration.
That would certainly be best, but it would be acceptable if you'd have
to go into the "bad" CPE to make it work correctly in the cascading
case, IMO.
Unfortunately, I don't think it's possible to manually configure an
alternative for subdelegation, assuming the (sub)delegation isn't a
fixed prefix.
If that's too hard, then we must mandate the ability for ISP-
provided CPEs to run in bridge mode. (But then that CPE would
probably still need a global address for management.) Preferably
the ISP-provided CPE would go in bridge mode automatically when a
more capable CPE is present. I guess this could happen if the ISP-
CPE receives a PD request.
My understanding is that some providers expect to provision their
networks in an unnumbered fashion, i.e. they MUST be a router, not a
bridge, and their integrated hosts obtain global addresses by self-
assigning from the router's delegated prefix. Having another router
come along and dynamically replace the integrated router without
disturbing the host strikes me as a tricky problem.
I don't think that's necessary. We could make it such that a CPE
detects whether other CPEs or non-CPE devices are connected to its LAN
ports. Then, if only one LAN port is active and that LAN port connects
to a device identifying as a CPE, the first CPE disables its WLAN
interfaces, if any, and transparently bridges between its WAN and LAN
ports. The only question in this scenario is how management of the
original CPE would happen in the unnumbered scenario. But if that's a
problem people can always use the numbered scenario...
Iljitsch