Re: [DNA-BOF] Potential Workplan for comment

[Greg Daley <greg.daley@eng.monash.edu.au>]

Hi Bernard,

Bernard Aboba wrote:
>>Is there interest at this stage to assume this
>>work in either DHC or zeroconf?  This would
>>make it easier for us to see what role this BoF has for v4.
> 
> 
> There was a section in the IPv4LL spec (which I'm editing), but the AD
> (Thomas Narten) felt it would be better off as a separate document --
> and thus the IPv4 NAD I-D was born.
> 
> A bunch of folks from DHC WG were part of the discussion of the issues,
> and much of the document does touch on issues relating to RFC 2131. But I
> won't presume to speak for DHC WG or the Internet ADs in terms of whether
> the IPv4 NAD document will be accepted within DHC WG, once it's presented.

Thanks for the information.  At this stage, I think
we will have to address the possibility of the DHC related
work coming to DNA as well, until we can confirm what
Thomas had in mind (but we won't burn too many cycles
on it worrying).

It will definitely be part of the discussion in the BoF,
even if DNA isn't going to handle this (especially so
because we want people to decide/agree about this).


>>I think that you mentioned earlier that there are
>>common reliances between v4 and v6 systems, in
>>terms of underlying Link-triggers.
>>Do you think that this work should be done in common
>>(in DNA) even if another group handles updates to
>>existing documents?
> 
> 
> Well, a section could be written for one doc that could be referenced from
> the other. So the important thing is not so much common work as common
> review.  I suspect that where DHCPv6 is deployed, IPv6 DNA will
> end up having some similarities to IPv4 DNA. Is it true to say that the
> major differences between these two scenarios result from differences
> been ARP and ND/NS?

Actually the architecture of IPV6 ND provides a
lot of information which can be used for detection
before configuration procedures are undertaken.
So it may be that pure detection doesn't (necessarily)
include DHCPv6.

If DNA does have a role in speeding address configuration
(and especially DAD), then certainly, DHCPv6 is likely
to be involved.  Whether the work is done in one group
or the other there will still certainly be common review.


>>Is it worth asking some of these questions on the DHC
>>list?
> 
> 
> I think it is worth talking about DHCP-based DNA scenarios on the DHC
> list, yes.


Hmm. I think it may take some delicate phrasing.

I will ponder.

Greg