[6bone] RE: [ipv6-wg@ripe.net] Update on IPv6 filter recommendation

[Michel Py]

Christian,


> Christian Kuhtz wrote:
> Why not ask Rob Thomas do the same for IPv6?  

I did and I know he's working on it.


> We don't need to reinvent the wheel and have multiple
> sources for this stuff.

I'm not sure about that.

- I like redundancy.

- The IPv6 situation is quite different than the IPv4. In IPv4, the
bogon list is constructed by taking out what is not allocated. In IPv6,
this would not be enough as there are so many huge holes in allocated
space that hijacking applies both to allocated and unallocated.

- Therefore, things such as what Jeroen does which is a live analysis of
the actual routing table have value. One can argue that the name for
this would be a real-time blackholing table and not a bogon list, but
nevertheless a route-server for these would be welcomed.

Jeroen please correct me if I'm wrong but what you call a bogon is a
route that you have actually seen in the GRT and that has no business
there?

So, I don't see any re-inventing of the wheel here. It is clear that
Rob's IPv6 bogon list would include 4000::/4 and FE80::/10 among other
things, but a more dynamic blocking such as what Jeroen could do would
be welcomed too.

Michel.