Re: [DNA] Re: Scope of DNA BCP (Previously Re: Submission_for_...)

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

Hi JinHyeock,

JinHyeock Choi wrote:
> Dear Sathya
> 
> [omitted]
> 
>>>I hope we take pragmatic approach. If it'll help DNA to mention DAD in BCP, it's fine with 
>>>me. If not, it's no point to include DAD in BCP. 
>>
>>I agree - lets include text only if we have consensus that it is useful in providing required > background for
>>our work.
> 
> 
> ok. Reasonable enough. 

Perhaps the questions regarding DAD and configuration
come about because of the background in section
4 "Current Practice for Hosts Supporting DNA "


There is a lot of background information regarding the
use of ND for DNA, and its relationship to configuration
procedures such as DAD.

Is the role of this section appropriate?

Does it need to be reviewed?


>>>I don't think it makes sense for BCP to include 'everything that needs to be done' for the 
>>>IP layer to become operational'. We'll not contemplate Binding Update delay, will we? It 
>>>may detract us from the real thing to be done, 'checking for link change'.
>>
>>Binding Update itself does not fall under 'everything that needs to be done' becos BU is to maitaing continuity
>>of higher layer session, 'IP layer is operational' before we send the BU.   
> 
> 
> 'Everything that needs to be done' is still fuzzy to me. IP layer is operational in various degrees.   
> 
> 1) When a host has a link-local address, it can communicate on link. 
> 2) When a host has a global IP address (for example, CoA), it can start 
>     a new session but the existing one will be severed. 
> 3) Only after a host sends BU and receives BA, the existing session will be 
>      resumed. 

Indeed, if we do MIPv6 :)

I think we need to determine that we've got a usable
connection to the Internet, which allows us
to configure addresses etc (without worrying about
how to do it).


Greg