IPv6 over gif(4) broken in 6.2-RELEASE?

[Bruce A. Mah]

If memory serves me right, John Hay wrote:
> On Sat, Jan 20, 2007 at 08:29:36AM -0800, Bruce A. Mah wrote:
>> I'm observing a problem with IPv6 over gif(4) tunnels on 6.2-RELEASE
>> and recent 6-STABLE, namely that I can't seem to be able to pass
>> traffic over them.
>>
>> Essentially, when I configure a gif interface like this:
>>
>> # ifconfig gif0 inet6 aaaa:bbbb:cccc:dddd::1 aaaa:bbbb:cccc:dddd::2 prefixlen 128
>>
>> the interface should add a host route to aaaa:bbbb:cccc:dddd::2
>> through gif0.  This is necessary to be able to pass traffic over the
>> tunnel, particularly since the source and destination addresses of the
>> link don't need to have any relationship to each other.
> 
> I only have one IPv6 over IPv4/gif tunnel and ther I use only my side
> of the address, something like this:
> 
> ifconfig gif0 inet6 2001:4200:ffff:5::2 prefixlen 64
> 
> And then bgp on top of this. It seems to work fine on -current built
> after my change. 

I believe the difference between your situation and mine is that your
gif0 interface is setup with a prefixlen < 128, which doesn't
specifically require a host route to the interface of the destination.
This is actually handled specially in several parts of the IPv6 stack.

> Well it seems that even my stuff does not always work perfectly with that
> change (1.48.2.15), so maybe we should revert it and I will search for
> yet other ways to make FreeBSD's IPv6 code to actually work for our stuff.
>
> My "stuff" is a wireless IPv6 only network running in adhoc mode with
> olsrd as the routing protocol. The problem is that all nodes on a subnet
> cannot "see" each other, so olsrd needs to add routes to a node through
> another node. Sometimes, just to complicate matters a little more, you
> would want to have more than one network card in a host, all with the same
> subnet address. (For instance on a high site, with sector antennas.)
> 
> The case that I found that still does not work reliably, is if olsrd add
> the route and route is not immediately used, then the nd code will time
> it out and remove it.
> 
> So, I guess if you guys think I should revert my stuff, just say so.
> 
> And if you have a solution for my problem, just say so too. :-)

This sounds kind of interesting!

I'm concerned that this bug seems (at least in my testing) to be present
in 6.2-RELEASE.  I'm not 100% sure what's the right thing to do at this
point.

Thanks,

Bruce.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 249 bytes
Desc: OpenPGP digital signature
Url : http://lists.freebsd.org/pipermail/freebsd-net/attachments/20070122/a9989735/signature.pgp