Re: changes to draft-ietf-v6ops-nat64-pb-statement-req-00.txt

[Alain Durand <alain_durand@cable.comcast.com>]

On 7/17/08 11:40 AM, "marcelo bagnulo braun" <marcelo@it.uc3m.es> wrote:
    
>>> well, it seems that you will be having legacy IPv6 hosts (i.e. no new
>>> mechanisms) that are running IPv6 apps that want to communicate with v4
>>> hosts that are running ipv4 apps and in that case, this doesn't seem to work
>>>     
>> 
>> The only thing that is required for dual-stack lite is a v4/v6 tunnel that
>> is present on most implementation I know, so no new code is needed.
>> And if this is still a problem, you can always use a router upstream to do
>> the encap for you.
>>   
> but you would need to have v4 on your hosts, right?

No, you need a v4 stack, not a v4 address. That is the whole point of the
dual-stack lite model.
 
> the assumption for translation is that you have v6 only nodes on one side

This is the part I'm challenging by making a distinction between v6-only,
dual-stack provisioned (v4 & v6) and dual-stack lite, where the host is
dual-stack capable but only provisioned with v6.

  - Alain.