6to4 status (again)

[Phil Mayers]

On 01/03/13 15:40, Brian E Carpenter wrote:
> On 01/03/2013 14:58, Ignatios Souvatzis wrote:
>> On Fri, Mar 01, 2013 at 02:04:50PM +0000, Brian E Carpenter wrote:
>>
>>> You can't fix legacy clients and CE devices, that's the problem. If they are
>>> young enough to have v6 enabled by default, but old enough to have anycast 6to4
>>> enabled by default and v6 always preferred, they will try 6to4 and will not fall
>>> back to v4.
>>>
>>> (The story is similar for Teredo.)
>>
>> Shouldn't - Teredo is explicitly specified to be last-resort, IIRC.
>> Do the prevalent implementations disagree?
>
> As I understand the RFC3484 default, which would apply to legacy
> systems, if there's a Teredo address it will beat IPv4. I don't
> know if the older Windows stacks actually do that, personally.

Any version of windows with Teredo has it de-prefed IIRC; my Win7 box has:

netsh interface ipv6>show pref
Querying active state...

Precedence  Label  Prefix
----------  -----  -------------
         50      0  ::1/128
         40      1  ::/0
         30      2  2002::/16
         20      3  ::/96
         10      4  ::ffff:0:0/96
          5      5  2001::/32


XP didn't have IPv6 enabled by default, and IIRC didn't have Teredo even 
if you enabled IPv6?