Re: [dhcwg] Feedback on Layer 2 Relay Agent functionality

[David Miles <davidm at thetiger.com>]

Thanks David,

This aligns with what I had assumed for option 1. I'd assume its also  
legal to set the IP destination of this DHCPv6 Relay-Forward to the  
All-DHCPv6-Servers MC address (fe02::1:2)? One consequence of  
implementing a full DHCPv6 relay on the DSLAM is that the DSLAM is now  
an IPv6 host within each VLAN and must therefore respond to DAD and ND  
from both clients and the BNG (as an example, of course there is far  
more than this). This could introduce significant complexity into the  
implementation with little apparent benefit, after all the relay agent  
in the DSLAM is not trying to unicast the traffic directly to a server  
- it is simply inserting Remote-ID.

When you say it has been discussed, does that imply a preliminary  
conclusion has been reached? I don't suppose you know roughly when it  
was discussed so I could trawl through the archives? While DHCPv6  
relay is an integral part of RFC 3315 I'm not aware of a discussion  
around TR-101 L2 DHCPv6 Relay, though I see there was one recently for  
IPv4.

Regards,

-David



On 07/11/2007, at 5:15 AM, David W. Hankins wrote:

> On Tue, Nov 06, 2007 at 08:21:29PM +1100, David Miles wrote:
> > With the recent discussions around Layer 2 Relay Agents (such as that
> > described in DSL Forum TR-101 Appendix B) has there been  
> > consideration
> > of the equivalent for IPv6?
>
> I believe it has been discussed...it's an integral part of DHCPv6,
> this seems to have been a design consideration.  The things we've been
> talking about are places where perhaps some clarification is in order;
> in particular that they will simply behave like a regular DHCPv6 relay
> agent - the only exception that they will send a zero-filled
> link-address value (3315 makes no explicit statement about what to do
> here, and some implementers have asked if they should fill it with a
> link-local address...).
>
> For this reason, servers use the outermost link-address value that
> is nonzero for client link detection ("unless otherwise configured"),
> something that is also not concretely defined ("exampled?") in 3315.
>
> --
> Ash bugud-gul durbatuluk agh burzum-ishi krimpatul.
> Why settle for the lesser evil?	 https://secure.isc.org/store/t-shirt/
> --
> David W. Hankins	"If you don't do it right the first time,
> Software Engineer		     you'll just have to do it again."
> Internet Systems Consortium, Inc.		-- Jack T. Hankins
> _______________________________________________
> dhcwg mailing list
> dhcwg at ietf.org
> https://www1.ietf.org/mailman/listinfo/dhcwg


_______________________________________________
dhcwg mailing list
dhcwg at ietf.org
https://www1.ietf.org/mailman/listinfo/dhcwg