[Info-vax] implementing IPv6 on the internet

[Dirk Munk]

Chris wrote:
> On 09/23/16 21:22, Dirk Munk wrote:
>
>>
>> And now we get a problem. How is the router suppose to know it has to
>> build a new IPv6 packet with the same payload, but destination address
>> X6, and sending address WAN6?
>
> The router will already have it's own V6 address, supplied by the isp.
> It will also know about the V4 addresses on it's own private subnet
> and how to work NAT for those addresses.

Great, and how does the router know the IPv6 address of Microsoft, or 
even better, how does the router know it has to send the packet to the 
IPv6 address of Microsoft?

The PC did a nslookup remember, the router didn't. The router doesn't 
know it has to exchange the IPv4 address of microsoft by the IPv6 
address, how could it? If the PC sends something to an IPv4 address, how 
can the router know if this destination also has an IPv6 address?

> Basically, all that needs
> to be done is to translate between the packet format of one
> to the other. Not a difficult task at all and one that a V4
> router already does via it's NAT already. NAT translates
> internal addresses in both directions, so that the internal addresses
> never appear on the wan side of the router.
>
> As for cpu throughput, we work embedded systems here and can tell you
> that even an ancient pentium P500 has enough throughput to handle NAT
> and do dpi (deep packet inspection) and maintain state tables for
> dozens or more connections at once and in real time. Modern Arm
> processors are orders of magnitude faster than that and will handle
> it easily.
>
> As I said earlier, it's just a software engineering problem and if
> there's a demand for V6 to V4 conversion it will happen. Your isp
> may already use it for backbone to local subscriber conversion,
> so why don't you ask them ?.
>
> As for looking at "Belgium", I'd rather not, if you don't mind :-)...
>
> Regards,
>
> Chris