Re: [sunset4] An IPv4 sunset problem (?)

Hi Marc,

Please see inline.

Cheers,
Med 

>-----Message d'origine-----
>De : Marc Blanchet [mailto:marc.blanchet@viagenie.ca] 
>Envoyé : mercredi 22 août 2012 18:59
>À : BOUCADAIR Mohamed OLNC/NAD/TIP
>Cc : Simon Perreault; sunset4@ietf.org
>Objet : Re: [sunset4] An IPv4 sunset problem (?)
>
>
>Le 2012-08-22 à 12:54, <mohamed.boucadair@orange.com> a écrit :
>
>> Hi Simon,
>> 
>> This issue has been already documented in depth in 
>http://tools.ietf.org/html/draft-ietf-behave-nat64-learn-analysis-03.
>> 
>
>maybe, but for a document (gap analysis) that should state 
>potential issues when turning IPv4 off in a network,

Med: If IPv4 is turned off, then no host will be dual-stack serviced...so there is no issue at all with NAT64/DNS64. The issue is when there are a mixture of IPv6-only connectivity and DS connectivity; but that one can be solved in various was as having distinct dns servers or using one of the solutions listed in the behave draft. 

 I think 
>that issue (two sets of stacks in an IPv6-only network with 
>NAT64)  is relevant to be listed.  
>
>Maybe it is a short paragraph that points to the others (such 
>as the one you are referencing to), so we avoid multiple 
>discussions. I leave it to the editors to review the 
>referenced texts and write whatever is appropriate.
>
>Marc.
>
>> I'm not sure if there is a value to re-discuss this point in 
>the gap analysis document.
>> 
>> Cheers,
>> Med
>> 
>> 
>>> -----Message d'origine-----
>>> De : sunset4-bounces@ietf.org 
>>> [mailto:sunset4-bounces@ietf.org] De la part de Simon Perreault
>>> Envoyé : mercredi 22 août 2012 17:54
>>> À : sunset4@ietf.org
>>> Objet : [sunset4] An IPv4 sunset problem (?)
>>> 
>>> Sunsetters,
>>> 
>>> Recently there has been discussion in v6ops about a problem 
>>> that arises 
>>> when a network infrastructure featuring NAT64+DNS64 is used by 
>>> IPv6-only 
>>> as well as dual-stack hosts. The problem is that you want the 
>>> IPv6-only 
>>> hosts to use the DNS64 but not the dual-stack hosts.
>>> 
>>> This is a well-known problem that has been discussed in 
>behave and for 
>>> which there is a solution that involves the hosts 
>>> (draft-ietf-behave-nat64-discovery-heuristic), and another 
>>> that involves 
>>> only the network (provision different DNS resolver addresses to 
>>> IPv6-only and dual-stack hosts). Those solutions may not 
>apply in many 
>>> cases: the first solution requires that all hosts be modified, 
>>> while the 
>>> second solution requires that one be able to differentiate 
>IPv6-only 
>>> from dual-stack hosts.
>>> 
>>> Question for the WG: should this be discussed in 
>>> draft-ietf-sunset4-gapanalysis? If so, we would appreciate 
>additional 
>>> input. Additional solution ideas would also be appreciated...
>>> 
>>> Thanks,
>>> Simon
>>> -- 
>>> DTN made easy, lean, and smart --> http://postellation.viagenie.ca
>>> NAT64/DNS64 open-source        --> http://ecdysis.viagenie.ca
>>> STUN/TURN server               --> http://numb.viagenie.ca
>>> _______________________________________________
>>> sunset4 mailing list
>>> sunset4@ietf.org
>>> https://www.ietf.org/mailman/listinfo/sunset4
>>> 
>> _______________________________________________
>> sunset4 mailing list
>> sunset4@ietf.org
>> https://www.ietf.org/mailman/listinfo/sunset4
>
>