Re: [v4v6interim] "IPv4->IPv6 is hard"

marcelo bagnulo braun 写道:
>>
>> Some additional notes based on CERNET two years experience of running a
>> pure IPv6 backbone.
>>
>> We believe the selection of the scenarios could be based on the
>> incentive of the owner of the IPv6 hosts.
>> (1) If the IPv6 host needs to be an IPv4-accessible server, they will
>> use the IVI address. Otherwise, it will use non-IVI address. This is to
>> say that there is no need to provide a general scheme for any IPv4 host
>> initiates communication with ANY IPv6 host, since the owner of the IPv6
>> host can select the suitable category of the IPv6 addresses.
> here is where i disagree
> I think we should support communications to some v6 addresses creating 
> some manual or automatic 1:1 mapping because it is the best that we 
> can do without having to use very nasty hacks like the ones defined in 
> the original natpt, not because there is no need for such scheme
Why not? This is What IVI is doing in CERNET2. The 1:1 stateless is for 
the IPv4 initiated communication.
> Many many hosts today run p2p applications. they certainly can use a 
> reachable address wihtout neededing to rely on nat traversal 
> techniques. and this is just one example. you can think about other 
> cases, like your home camrea that you want to access from the internet 
> and so on.
>
> So, i think it makes a lot of sense if any v6 address could be reached 
> from the v4 internet and i think this would result in faster adoption 
> of v6, but i don't think we can achieve this in a sufficiently clean 
> way (not yet at least, maybe in some near future, the natpt would be 
> sufficiently clean :-)
IVI 1:N mapping can solve this problem. The 1:N Stateful is for the IPv6 
initiated communication.

xing
>
>
>
>> (2) As the ISP, we price IVI addresses (1:1 mapping of IPv4 address,
>> scarce resource) and non-IVI addresses (non-scarce resource)
>> differently. So it is also economically reasonable.
> Right, this is exactly what i would like to prevent, that ISPs have 
> control on whether their clients are able or not to publish content 
> from their sites
>
> Regards, marcelo
>
>
>
>>
>> Just our two cents.
>>
>> xing
>>>
>>> _______________________________________________
>>> v4v6interim mailing list
>>> v4v6interim@ietf.org
>>> https://www.ietf.org/mailman/listinfo/v4v6interim
>>>
>>>
>>
>>
>>
>>
>> _______________________________________________
>> v4v6interim mailing list
>> v4v6interim@ietf.org
>> https://www.ietf.org/mailman/listinfo/v4v6interim
>
>

_______________________________________________
v4v6interim mailing list
v4v6interim@ietf.org
https://www.ietf.org/mailman/listinfo/v4v6interim