Re: NATLL6 [was Re: Forwarding Packets With Link Local Destination Addresses]

[Alejandro Acosta <[e-mail]>]

Hello Fernando, All,

   Here are few links if you want to check out:

https://github.com/LACNIC/natmeter

https://www.youtube.com/watch?v=XbMW0shzrPQ

https://www.youtube.com/watch?v=UTP_sOehrxU


Thanks,


Alejandro,



On 8/1/21 3:51 PM, Fernando Gont wrote:
> On 8/1/21 16:37, Brian E Carpenter wrote:
> [...]
>>>>
>>>> I interpret this statement to include packets that contain routing 
>>>> headers. For example, it forbids an SRv6 packet whose final segment 
>>>> has a locator that begins with FE80.
>>>>
>>>    In LACNIC we ran a project called Natmeter [1] for about 2 years, 
>>> we obtained a lot of interesting data during this period [2].
>>>
>>>    The case is the following, we detected some end-user devices with 
>>> only Link Local addresses (exactly, no GUA nor ULA) that were 
>>> successfully natted and using the web.
>>>
>>>    Is it ok?, do you consider it as a forwarding of a packet? was it 
>>> a crazy result? (we saw few samples of this)
>>
>> NAT66 is not defined and not recommended, so there really isn't an 
>> answer to your questions. I can't think of a reason why such a hack 
>> wouldn't work, though. As long as the LL address is translated, it 
>> will not escape from the LAN.
>
> I would expect that something else (other than IPv6 NAT) is going on. 
> IN that sense, it would be interesting to know more details about the 
> methodology they employ to measure the use of NATs.
>
> Thanks!
>
> Regards,