Re: new draft on IPv6 CPE router available for review

[Shin Miyakawa <miyakawa@nttv6.jp>]

Hemant and Mikael,

Let me clarify that your idea is that 

   One CPE MUST support BOTH models ?

This is quite important, because it defines operation scheme.

Shin Miyakawa



From: "Hemant Singh (shemant)" <shemant@cisco.com>
Subject: RE: new draft on IPv6 CPE router available for review
Date: Tue, 1 Jul 2008 16:05:07 -0400

> Honestly, I don't see any downside to making support for both models as
> a MUST. If anyone any objections to making both models a MUST, it's time
> to speak up.
> 
> Hemant 
> 
> -----Original Message-----
> From: Mikael Abrahamsson [mailto:swmike@swm.pp.se] 
> Sent: Tuesday, July 01, 2008 3:50 PM
> To: Hemant Singh (shemant)
> Cc: Shin Miyakawa; Wes Beebee (wbeebee); v6ops@ops.ietf.org
> Subject: RE: new draft on IPv6 CPE router available for review
> 
> On Tue, 1 Jul 2008, Hemant Singh (shemant) wrote:
> 
> > router supporting the link-local only WAN model. Of course, we totally
> 
> > want the first model to be a MUST for the CPE router to support - in 
> > this model the WAN interface acquires a global IPv6 address.
> 
> What are the downsides of making both models a MUST? I think any router
> should (MUST) be able to source packets from any interface chosen,
> including one that is not the outgoing interface.
> 
> This would of course have the implications that some primary host
> implementations with optional routing capability might not comply with
> this right now, but would put pressure on vendors to implement this
> properly.
> 
> I know there's been discussions regarding traceroute behaviour as well,
> with the industry standard being that the ICMP TTL-EXCEEDED message
> being sourced from the router IP where it was received, but goes out any
> interface (as per the routing table to the destination).
> 
> I just would like to get future IPv6 implementations for routers working
> properly from the start, as most residential CPE vendors seem to not
> have implemented IPv6 support yet, I would like them to do both models
> as the model I'm advocating would have long term benefits for network
> stability.
> 
> -- 
> Mikael Abrahamsson    email: swmike@swm.pp.se
>