Re: [DNA] Proposed resolution to issue 9 fordraft-pentland-dna-protocol-01.txt

[Syam Madanapalli <smadanapalli@gmail.com>]

Hi Brett,

The CPL requires to build complete prefix list on both the previous 
link as well as on the new link. Is this correct? 

Also Greg said the MN checks all RAs as they arrive. This means
MN cannot make decision based on single RA with NO answer, unless
the RA is a 'Complete RA'.

Also I am wondering how would CPL works without building complete
prefix list on previous link? Are you assuming 'Complete RA'? 

I am sorry, I am getting confused.

-Syam


On 8/4/05, Brett Pentland <brett.pentland@eng.monash.edu.au> wrote:
> Hi Syam,
> 
> Syam Madanapalli wrote:
> > Hi Greg,
> >
> >
> >>Syam Madanapalli wrote:
> >>
> >>>Hi Brett,
> >>>
> >>>I think we may need to check for the prefixes that the other routers
> >>>on the link advertising. That is we need to build CPL on the new
> >>>link.
> >>
> >>Hi Syam,
> >>
> >>
> >>I think that as each RA arrives, it can be compared to the
> >>CPL, following this rule.
> >
> >
> > Yes I got it.  Now the question is how long the MN should wait
> > for receiving all RAs from the routers? In between if an RA comes
> > with YES answer then we can assume link change otherwise, the
> > MN needs to wait for all RAs.
> 
> Why would it wait.  If the host has the CPL on the old link, then it
> can make the decision immediately on recption of the first RA.  If we
> specify that the NO answer include a DNAO, then the host doesn't even
> need to have the CPL on the previous link to make an immediate
> decision.
> 
> Brett.
>