In message
<20111204155527(_dot_)BE11218C131(_at_)mercury(_dot_)lcs(_dot_)mit(_dot_)edu>,
Noel Chiappa write
s:
> From: Mark Andrews <marka(_at_)isc(_dot_)org>
> The CGN boxes are new. The customer boxes which are being allocated the
> addresses are old. Lots of these boxes will not work with a 240/4
> address.
Here's a question, though: would a mix of a smaller block of 'classic' IPv4
space, _along with_ a big chunk of 240/4, be of any use?
I ask because I gather there are a lot of situations where e.g. a cable modem
has an ISP-local address on its ISP-facing side, and a global IP address
(which the customer gets) on the customer side. (I see this in checking out m
y
path to the Internet from my house - I see some 10.xx addresses on the other
side of the cable modem.) Could a block of 240/ space be used for things like
that?
Sure, the 240/ space wouldn't be useful as a source of addresses for
_customer-owned_ equipment - but could it be useful elsewhere inside an ISP's
network? (And perhaps also as a source of addresses for CPE equipment which
_will_ tolerate being given a 240/ space address, although I concede that may
be very few boxes, at least at first.)
I suspect most Cable ISP's will do what Comcast is doing and migrate
to IPv6 for management of the modems. DOCSIS 3 modems support IPv6
and I believe some DOCSIS 2 modems also support IPv6 with newer
firmware. This part will fix itself as modems fail and are replaced
as long as you require the new cable modem to support IPv6 on the
management side. You can deal with the few customers bringing their
old modem with them as they move between providers.
That said I'm not employed by a Cable ISP and the cable modem at
home is a DOCSIS 1 box.
Noel
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf
--
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742 INTERNET: marka(_at_)isc(_dot_)org
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf