Re: [BEHAVE] I-D Action:draft-ietf-behave-v6v4-xlate-03.txt

[Simon Perreault <simon.perreault at viagenie.ca>]

Xing Li wrote:
> HUANG, JERRY (ATTLABS) 写道:
>> Does it make sense to specify only one option (to decrement the
>> IPv4 TTL and check for zero before translation and send ICMPv4 "TTL
>> Exceeded")? It doesn't seem to make sense to translate and find out TTL
>> would be zero and send ICMPv6 "Time Exceeded" which would then need to
>> translated again into ICMPv4 to send to the IPv4 host.   
> 
> Well, we think this should be vendor's decision.

Indeed. Our implementation does it before translation. The spec doesn't
need to be changed for what amounts to an implementation detail.

Simon
-- 
DNS64 open-source   --> http://ecdysis.viagenie.ca
STUN/TURN server    --> http://numb.viagenie.ca
vCard 4.0           --> http://www.vcarddav.org