Re: [DNA-BOF] Initial Reading List ideas for DNA

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

Hi Gabriel,

I have to say that's the last time I post URLs
without checking them.

Thanks for your updates.

gabriel montenegro wrote:
> Greg Daley wrote:
> 
>> Mobile IPv4 Movement Detection:
>>
>> Mobile Ipv4 (RFC 3220, section 2.4.2)
>> http://www.ietf.org/rfc3220.txt
> 
> 
> That link doesn't work. But never mind 3220 anyways, the real
> version and link is:
> 
> http://www.ietf.org/rfc/rfc3344.txt
> 
> How about:
>
 >
> Optimistic DAD (odad)
> http://www.ietf.org/internet-drafts/draft-moore-ipv6-optimistic-dad-00.txt

this is now at -02.
( I think there may be a new revision out soon )

> MLD to optimize DAD
> http://www.ietf.org/internet-drafts/draft-daley-ipv6-mcast-dad-02.txt
>
> Alper Yegin had a draft on using the router as an oracle (can't find it 
> though).

Alper, can you indicate where that work is at?

> Advanced DAD (adad)
> http://www.ietf.org/internet-drafts/draft-han-mobileip-adad-00.txt
> Which was preceded by several years by similar functionality in:
> http://www.cs.ucl.ac.uk/staff/t.pagtzis/papers/draft-pagtzis-mobileip-proactivev6-00.txt 

Do you have a section number(s) which describes this function?
I was unable to find the section related to aDAD.


I'm interested in DAD optimization myself, but was wondering
if people thought this was part of network attachment detection?

The issue is largely whether the detection procedures are
separate from the configuration procedures.  For example
in IPv4, detecting the identity of the link
could use a combination of ARP to the current AR and DHCP,
to confirm the prefix.

In IPv6, Router Discovery may be able to provide prefix information
without resorting to explicit configuration procedures,
although stateless address autoconfiguration follows in the
case of arrival on a new network.

What is the feeling at the moment?

I think that we'll have to also ask the ADs whether
they had something else in mind for DAD optimization.
In the mean time, we can add them to the list for discussion.

Greg