On 2009-02-18 17:20, David W. Hankins wrote:
> On Wed, Feb 18, 2009 at 03:45:45PM +1300, Brian E Carpenter wrote:
>> Wouldn't you want a short overlap to allow seamless migration
>> of application sessions? Break-before-make is not good for
>> real time streams or for any kind of QOS assurance.
> 
> I think two Softwires endpoints will cover different IPv4 address
> ranges on the 'external' side of the CGN.  So even make-before-break
> would not provide what you are looking for?

Actually, wouldn't it allow a clever NAT to update session state
automatically? In any case, we can't be sure that this is the only
usage scenario for ever.

   Brian
_______________________________________________
Softwires mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/softwires

Reply via email to