RE: Comments on draft-yourtchenko-colitti-nd-reduce-multicast

I agree with Eric.   Without changing some IETF RFC's such as making sending DAD a MUST for every DHCPv6 acquired address or not changing RFCs we still will have to change ipv6 implementations to make DHCPv6 work.  
For this wifi services, the client and the access-point (AP) serving the client are in the same ipv6 ND link-local domain.  There are several solutions but implementations will change. 

A hack, but if the AP also includes DHCPv6 server, the AP knows from the upstream L2 frame of the DHCPv6 SOLICIT as to what the client mac-address and ipv6 link-local address is.   If the DHCPv6 server is north-bound of the AP, then the AP should support DHCPv6 relay services and the relay tracks client mac-address vs. ipv6 address(es) in the AP.  

Hemant

-----Original Message-----
From: ipv6 [mailto:ipv6-bounces@ietf.org] On Behalf Of Eric Levy- Abegnoli (elevyabe)
Sent: Friday, February 21, 2014 9:19 AM
To: Ole Troan; Erik Nordmark
Cc: Andrew Yourtchenko (ayourtch); 6man WG
Subject: Re: Comments on draft-yourtchenko-colitti-nd-reduce-multicast

DHCP won't work without changes which would be a lot bigger that anything envisioned for ND address registration.

DHCP does not maintain link-locals. That would be a major change to fix that, including on hosts, since address requests are responded in unicast.
DHCP does not maintain <IP, MAC> binding. It does not have the MAC (not to confuse with DUID).
DHCP is not on the forwarding path where we need the information quickly Eric

On 20/02/14 15:43, "Ole Troan" <otroan@employees.org> wrote:

><nochair>
>
>> 4.8 seems to conflate the address assignment with DAD. Just because 
>>we might want to centralize the DAD checks doesn't imply that we want 
>>to remove the ability for the host to pick its own privacy enhanced 
>>interface-IDs to form its addresses.
>> From a deployment perspective DHCPv6 is available for address 
>>assignment, but don't think we want to require that for WiFi or other 
>>links which have packet loss. (Packet loss occurs on wired networks as 
>>well, but the drop distribution is different - might happen during 
>>spanning tree reconvergence etc.)  Note that DHCPv6 (RFC 3315) has a 
>>SHOULD for doing DAD on the addresses received from the DHCP server - 
>>needed since the server could be confused.
>
>I'd like to explore the difference between DHCP address assignment and 
>ND address registration a little more.
>the state that is required in the network for "efficient ND", why 
>cannot that be created and maintained by DHCP?
>
>there are multiple ways of dealing with DAD in this scenario, including 
>solutions that don't require host changes.
>DHCP also supports temporary addresses btw.
>
>cheers,
>Ole
>
></nochair>

--------------------------------------------------------------------
IETF IPv6 working group mailing list
ipv6@ietf.org
Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------