[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Teredo routing status?



On Wed, 2006-05-17 at 16:09 +0300, Anssi Porttikivi wrote:
> I found in the Wikipedia Teredo article that
> http://www.sixxs.net/tools/grh/lg/?find=2001::/32
> can be used to verify that the new Teredo prefix is indeed widely
> available as an anycast route. There was also a reference to this list
> and a hint by David Thaler from Micorosoft that the prefix can be
> configured to Windows:
> 
> 	http://psg.com/lists/v6ops/v6ops.2006/msg00064.html 
> 
> Done that, and my Teredo does not work. Windows Teredo says "secondary
> server address unreachable".

Because you are most likely still using the old servers which only
support (afaik) the old 6bone prefix and not yet the 2001::/32 one.
Also, did you reboot? :)

> What is this configuration changed supposed to affect, exactly? Does the
> MS Teredo server accept a client request for address with this prefix?
> Is the global routing for it really working?

The v6 path is there, that is what GRH shows, though that still depends
on the viewpoint and if the route is actually made visible globally or
only for a limitted set of networks (eg using no-export or filters).

The v4 path is the other question and you will have to configure it to
use a server which supports the RIR prefix.

That said, you will want to contact the operator of your Teredo server
you are using, though you might some of them here, if you really want
operational people and not standards folks, try:
http://lists.cluenet.de/mailman/listinfo/ipv6-ops
Though I have to note that that is not a help-me list either.

> Is Teredo reliably and universally working with ANY configuration?

According to the people who tested it it seems to work except for the
cone-nat situation (or was it the other one, see wikipedia ;)

> I have fond memories of those few times last year when a South Korean
> public relay actually advertised the old prefix in global BGP, and
> Teredo actually worked globally.

Which was most likely quite useless latency wise when your packets had
to travel twice around the world; it also makes debugging a lot of
'fun'.

Greets,
 Jeroen

Attachment: signature.asc
Description: This is a digitally signed message part