RE: Comments on draft-wbeebee-ipv6-cpe-router-01.txt

["Hemant Singh (shemant)" <shemant@cisco.com>]

BTW, Ole, the high level requirements for DSL are already in the draft,
like totally high level features such as PPP and then new Softwires.
It's more detailed requirements such as WAN with link-local only vs.
only allowing WAN with GUA that are pending.  Also as I said in the past
there is one stock change in the -02 draft replacing stateless DHCPv6
with stateful DHCPv6 for IA_PD acquisition and then draft it fixed -
these are 3-4 places. 

IRB is something we have used time and time on routers and as I said to
this mailer, we have a working CPE Router using IRB in a cable modem at
Cisco. Any interface on our cable modem router can receive the RA sent
to the modem WAN interface if the interface has joined the bridged
group.  Anyhow, we are open to replacing IRB for a different solution.
For example, let users just configure the Loopback interface manually or
automatically (where the interface gets IPv6 address from IA_PD) if NTT
would like that for the CPE Router. 

Some folks have asked for pictures.  We plan to include pictures showing
external and internals of the CPE Router in our IETF 72 presentation. 

Hemant 

-----Original Message-----
From: owner-v6ops@ops.ietf.org [mailto:owner-v6ops@ops.ietf.org] On
Behalf Of Hemant Singh (shemant)
Sent: Tuesday, July 22, 2008 11:45 AM
To: Ole Troan
Cc: Iljitsch van Beijnum; IPv6 Operations
Subject: RE: Comments on draft-wbeebee-ipv6-cpe-router-01.txt

-----Original Message-----
From: ichiroumakino@gmail.com [mailto:ichiroumakino@gmail.com] On Behalf
Of Ole Troan
Sent: Monday, July 21, 2008 5:25 PM
To: Hemant Singh (shemant)
Cc: Iljitsch van Beijnum; IPv6 Operations
Subject: Re: Comments on draft-wbeebee-ipv6-cpe-router-01.txt

> We are not even concerned about the draft being a WG work item.  We 
> have not even completed requirements for the device.


>> perhaps you would have more success if you wrote down the
requirements instead of arguing for solutions.


I think so too.  We thought we could provide solutions to some early
requirements from DSL folks, but as you can see, those solutions need
some ironing out.  We have to get consensus between AT&T, NTT, Mikael
Abrahamsson, and David Miles who works with DSL IPv6 Forum for
requirements.  In that process we also have to filter some requirements
because a requirement might not be possible.  We heard some consensus
between AT&T and NTT when AT&T agreed that they also support a MUST for
global IPv6 address on the WAN interface.  There are other requirements
to be ironed out.  I had told the mailer, unless I meet NTT folks at
IETF 72, I am not clear on their requirements and our draft is subject
to change.  We understand David Miles' requirements fine.  We'll see
what we can do to add a requirements section to the document and
eventually that section also the lists what other section to go to for
what requirement so that any given solution is identifiable for a given
requirement.  If you see the first email from Mikael on our draft,
that's his requirement.  Then see the first email from Shin - that's
NTT's requirement.  Then see the first email from David Miles to see his
requirement.  AT&T's requirements are best seen from any of Barbara's
recent emails marching backwards.

Hemant