Re: [dhcwg] draft-forte-dhc-passive-dad-02

Yes. I was actually looking at it in the other direction. Assuming that 
you have pDAD, either one of the two would work. Perhaps the second one 
(DAD after binding) would take care of those less common cases where 
there could be some vulnerability window right after a server reboots 
and before the AUC re-sends the information to the server.

-Andrea

David W. Hankins wrote:
> On Wed, Jun 28, 2006 at 05:38:53PM -0400, Andrea G Forte wrote:
>   
>>> Does it have to avoid doing DAD, or can it merely elide the check to
>>> after binding to the address (if there is either an indication or a
>>> presumption that pDAD is in use)?
>>>       
>>  
>> This is a good point. I guess you could follow either approach. Is there 
>> a reason to prefer one rather than the other?
>>     
>
> Doing DAD before binding is not a bet.  You're dead sure, to the extent
> you reasonably can be, there isn't a collision.
>
> Eliding DAD after binding is hedging a bet: you're betting the server
> is likely to know about conflicts beforehand.
>
> So having an indication (or presumption) of wether the server is
> involved in pDAD might be a consideration in choosing one of those
> two.
>
>   
> ------------------------------------------------------------------------
>
> _______________________________________________
> dhcwg mailing list
> dhcwg@ietf.org
> https://www1.ietf.org/mailman/listinfo/dhcwg
>   

_______________________________________________
dhcwg mailing list
dhcwg@ietf.org
https://www1.ietf.org/mailman/listinfo/dhcwg