Re: [dna] Comments on Simple DNA

[JinHyeock Choi <jinchoe@gmail.com>]

Dear Bernard

in-line comments.

> I believe that the documents needs a detailed description of the contents of
> ND and RS packets (such as included options,
> and source/dest addresses).   For example, I believe ND packets are sent to
> the unicast linkscope address of the router,
> not a multicast address, whereas RS packets are sent to a multicast address.

agree. that would help to clarify the procedures.

> 2. Separation of DAD advice for DHCPv6 and stateless auto-config.  During
> the meeting, it was said that the document
> mentioned that DAD was not needed for still-valid DHCPv6 obtained addresses,
> but on reading the document, I could not
> find this statement (or a section specifically about behavior with
> DHCPv6-obtained addresses).

In theory, DAD is needed even for a valid DHCPv6 obtained address.

As of RFC 4862,

               Both
    stateless address autoconfiguration and DHCPv6 may be used
    simultaneously.

So when a host with a DHCP address detaches for a while,
during the host's absence, it's possible (while not likely) that
another host may configure the same address in stateless
address autoconfiguration.

I guess that's why RFC 4862 mandates DAD
even for DHCP addresses as below

                                                                The Duplicate
   Address Detection algorithm is performed on all addresses,
   independently of whether they are obtained via stateless
   autoconfiguration or DHCPv6.

IMO simple DNA would better not delve into DAD problem too much
lest it should become simple no more.

Thanks for your kind consideration.

best regards

JinHyeock
_______________________________________________
dna mailing list
dna@ietf.org
https://www.ietf.org/mailman/listinfo/dna