Call for testers: RFC 5569 (6rd) support in stf(4)

Ondoy loki.fab at gmail.com
Fri Sep 24 03:32:12 UTC 2010


Hello Hiroki,

This is great.
I've been looking at 6rd including the srd stuff from
http://bougaidenpa.org/masakazu/archives/54
(which I guess you also got wind of).

I haven't tried your patch yet but I need some clarifications.
RFC 5969 has the following elements for 6rd configuration:
IPv4MaskLen, 6rdPrefix, 6rdPrefixLen, 6rdBRIPv4Address.

>From your example, I think the following takes care of
6rdPrefix and 6rdPrefixLen:
  # ifconfig stf0 inet6 2001:db8:c000:0202::/32

How can we configure the IPv4MaskLen and 6rdBRIPv4Address?
I suspect IPv4MaskLen is fixed at zero (use all 32 bits) in this patch.(?)
As I understand from the specs, this can be variable.
It seems that at the CE, the IPv4 address of the BR is automatically
calculated from the dst/src IPv6 address. But what if the embedded
IPv4 address is not the whole 32 bits (IPv4MaskLen > 0)?

Also, you mentioned that prefixlen of stf is from 0 to 32. Is this
the same as 6rdPrefixLen?
I think 6rdPrefixLen can be more than 32.

Lastly, I'm a bit confused with the IPv6 default route at the CE
(set to 2001:db8:c000:0201::1) since there's supposed to be
no IPv6 connection between the CE and the BR, only IPv4.

Regards,
Ondoy


On Fri, Sep 24, 2010 at 12:15 AM, Hiroki Sato <hrs at freebsd.org> wrote:
> Hiroki Sato <hrs at freebsd.org> wrote
>  in <20100923.053236.231630719.hrs at allbsd.org>:
>
> hr> Hello,
> hr>
> hr>  Can anyone try a patch for adding 6rd (RFC 5569) support to stf(4)?
> hr>  The patch for HEAD can be found at:
> hr>
> hr>   http://people.allbsd.org/~hrs/FreeBSD/stf_6rd_20100921-1.diff
>
>  Grr, I noticed it broke 6to4 functionality.  Please use the fixed
>  version here:
>
>  http://people.allbsd.org/~hrs/FreeBSD/stf_6rd_20100923-1.diff
>
> -- Hiroki
>


More information about the freebsd-net mailing list