[39194] in North American Network Operators' Group
Re: Cable Modem [really responsible engineering]
daemon@ATHENA.MIT.EDU (Fletcher E Kittredge)
Thu Jun 28 15:12:25 2001
Message-Id: <200106281911.f5SJBot05915@smtp.gwi.net>
To: dhcp-server@isc.org
Cc: nanog@merit.edu (North America Network Operators Group Mailing List)
Reply-To: dhcp-server@isc.org
In-reply-to: Your message of "Thu, 28 Jun 2001 10:54:14 PDT."
<200106281754.f5SHsEL00489@grosse.bisbee.fugue.com>
Date: Thu, 28 Jun 2001 15:11:50 -0400
From: Fletcher E Kittredge <fkittred@gwi.net>
Errors-To: owner-nanog-outgoing@merit.edu
On Thu, 28 Jun 2001 10:54:14 -0700 Ted Lemon wrote:
> You're missing a piece, actually - sorry about that. You need to
> read RFC3046 as well. I thought DOCSIS referred to it, but maybe
> not.
Hey! That is a *very* cool spec. Do you know if any cable equipment
implements it yet (Motorola I assume?) From my packet sniffing, I
don't think the DOCSIS 1.0 and 1.1 gear we use implements RFC3046 yet.
From a casual skim RFC3046, "DHCP Relay Agent Information Option",
looks like a great idea that would solve many of the problems
mentioned in this thread. Rather than ask the customer for their
MACs, it would be possible to monitor DHCP packets and populate a
database from the Relay information.
I am not involved in the DOCSIS standards process, but future
DOCSIS specs should require RFC3046. Since Motorola is a major player
in cable modems, I bet RFC3046 will make it into future standards.
Thanks Ted, I appreciate the pointers. I will lobby to try to make
the above happen. This is the Internet at its best.
regards,
fletcher
