Re: [BEHAVE] How to set the DF and the ID values for IPv4 packets (was Re: Amount of fragmentation resulting from translation

[Reinaldo Penno <rpenno at juniper.net>]

Hi Marcelo,

Those scenarios are fine from the point of view of mapping to the framework
doc. But given the discussion we had/are having it seems to me that who runs
the GW and where it could be located in the network was not clear.

Small nit:

Section 8.2 is labeled as 'scenario 3'. There is no scenario 2.


On 10/30/09 11:45 AM, "marcelo bagnulo braun" <marcelo at it.uc3m.es> wrote:

> The draft has section    8.  Application scenarios which describe the
> scenarios that are defined in the framework document that the nat64
> apply. Not sure if more is needed.
> If you think it does, please propose text so that wg can comment.

Sure, something like (if there is consensus):

X. Applicability Statement

The GW could be located on the demarcation point between the ISP and its
customer or deeper within the ISP network. When the GW is located within a
CPE it could be managed by the customer or the ISP.

note: 'CPE' here should not be seen only as a small modem Home network, but
anything from a home network, SOHO till full fledge enterprise.

> 
> Regards, marcelo
> 
> 
> 
> Reinaldo Penno escribió:
>> Yes, it could. There was a discussion about NAT64 scenarios before and it
>> served to point out that people have different views on the deployment
>> scenarios. 
>> 
>> As far as the ISP running the NAT64, one could argue that even if the NAT64
>> is in the CPE, that is 'run by the ISP' as well. There was also a Microsoft
>> direct access scenario sent to the list.
>> 
>> 
>> Suggestion: 
>> 
>> My suggestion then was that the NAT64 document should have an
>> 'applicability' section (was many RFCs have) so that it is clear the
>> scenarios the solution covers.
>> 
>> Personally I feel the NAT64 could be at the CPE or deeper within the
>> network. It could be run by the ISP or an Enterprise.
>> 
>> Thanks,
>> 
>> Reinaldo
>> 
>> 
>> On 10/30/09 11:27 AM, "marcelo bagnulo braun" <marcelo at it.uc3m.es> wrote:
>> 
>>   
>>> Iljitsch van Beijnum escribió:
>>>     
>>>> On 30 okt 2009, at 17:30, Reinaldo Penno wrote:
>>>> 
>>>>       
>>>>> It is the norm in on ADSL/PPPoE deployments.
>>>>>         
>>>> But weren't we talking about the translator's MTU? I would assume
>>>> those aren't hooked up to an ADSL link.
>>>> 
>>>>       
>>> why not?
>>> you could put a nat64 in the cpe, right?
>>> 
>>>     
>> 
>> 
>>   
>