Le 2012-02-08 à 19:39, Lee, Yiu a écrit :

> Hi Remi,
> 
> I know this is possible to do, in theory. However my question is more
> toward manageability of the network. IMHO, layering one tunnel (or
> translation) protocol on another tunnel protocol is asking for trouble.

No obligation to do it, of course.
Noting that it is possible doesn't hurt though.
Personally, I find it legitimate to be careful before pretending to offer more 
service than before, but I don't see which trouble would result from IPv6 (with 
IPv4 embedded addresses) encapsulated in IPv4 (with RFC1918 addresses). 

Cheers,
RD

> 
> Cheers,
> /Yiu
> 
> On 2/8/12 2:13 AM, "Rémi Després" <[email protected]> wrote:
> 
>> 6rd can be deployed over Net-10 networks (RFC1918) to deployIPv6.
>> Shared public IPv4 addresses can then be offered to customers via this
>> IPv6.
>> Such a use case, based on the header-mappiong variant of 4rd-U rather
>> than on a double MAP encapsulation, is described in
>> tools.ietf.org/html/draft-despres-softwire-4rd-u-03#section-5.4.
>> 
>> RD
>> 
> 
_______________________________________________
Softwires mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/softwires

Reply via email to