Hi Tore,

> Over the years operational experience has revealed that the 6to4
> protocol suffers from so severe operational problems that it simply
> cannot be fixed. Therefore, the IETF recently decided to deprecate 6to4:
> 
> http://tools.ietf.org/html/draft-ietf-v6ops-6to4-to-historic
> 
> The document is approved for publication by the IESG and is just
> awaiting the finishing editorial touches by the RFC Editor.
> 
> The important stuff is in section 5:
> 
>   It is NOT RECOMMENDED to include the anycast 6to4 transition
>   mechanism in new implementations.  If included in any
>   implementations, the anycast 6to4 mechanism MUST be disabled by
>   default.
> 
>   In host implementations, unicast 6to4 MUST also be disabled by
>   default.  All hosts using 6to4 MUST support the IPv6 address
>   selection policy described in [RFC6724].
> 
>   In router implementations, 6to4 MUST be disabled by default.  In
>   particular, enabling IPv6 forwarding on a device MUST NOT
>   automatically enable 6to4.
> 
> ConnMan does currently automatically enable 6to4, so will violate this
> new requirement. I'm thinking that the easiest way to deal with that is
> to simply rip out the 6to4 code completely, since the operational
> problems with 6to4 (see RFC 6343) makes it such that the end users are
> far better off by staying IPv4-only than "dual-stacked" with 6to4. But I
> thought I'd ask before sending such a patch. The alternative would be
> to keep the code, but disable it by default. Any thoughts?

I would initially hide it behind a main.conf option, but have it defaulting to 
disable to comply with this updated RFC once it has been accepted.

However with 6to4, I actually do not know how many users are actually depending 
on our support of 6to4. Others that have used this feature, might want to 
comment here.

Regards

Marcel

_______________________________________________
connman mailing list
[email protected]
https://lists.connman.net/mailman/listinfo/connman

Reply via email to