Pedro,

That's fine, about what I expected. But it will be a lot clearer
if you add two sentences
a) clarifying that IPv6 is future work
  [Personlly I'd be tempted to reserve a code for the flow label.
   After all, draft-dreibholz-ipv4-flowlabel-09.txt is around.]
b) clarifying the as# point, maybe even say MUST be set to zero
   when no 2-byte as# is available.

Regards
   Brian Carpenter
   University of Auckland

On 2009-01-08 13:26, Pedro Roque Marques wrote:
> Brian,
> 
> Regarding the traffic-rate community referred to in section 6, the
> 2-byte as# is purely informational. It has no semantics and can be set
> to 0 when a 2-byte as# is not available. The field is there in order to
> be consistent with the definition of the traffic-rate community which
> existed in a early version of the extended community document and for
> which there are implementations.
> 
> Regarding what protocols are supported. Only IPv4 is supported by this
> version of the document. As you mention the same approach can be used to
> support IPv6 (using the IPv6 AFI).
> 
> The current document interoperable implementations as per IDR policies.
> These implementations do not support IPv6 at the moment as the types of
> problems they are ment to address have not yet been seen with IPv6
> deployments.
> 
>   Pedro.
> 
> On Jan 7, 2009, at 3:03 PM, Brian E Carpenter wrote:
> 
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> <draft-ietf-idr-flow-spec-03-carpenter.txt>
> 
> 
_______________________________________________
Gen-art mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/gen-art

Reply via email to