[6bone] separating IPv6 experimental from production traffic
Gert Doering
gert@space.net
Thu, 22 Aug 2002 15:12:42 +0200
Hi,
On Thu, Aug 22, 2002 at 10:55:01AM +0300, Pekka Savola wrote:
> On 22 Aug 2002, Nicolas DEFFAYET wrote:
> > > 5) production networks announce all prefixes to everybody
> >
> > I add:
> >
> > 6) Don't provide by default full transit to peers, provide full
> > transit only if peer request it.
>
> Automatic full transit per request is a road to disaster, *especially* if
> said peer will also give transit to others.
"Full transit to everybody" was a good idea to get a pretty tightly
meshed IPv6 network into place, which is *good* because it means that
you have (typically) only few AS hops to traverse - and thus fewer
networks in between that can mess up your routing.
Won't help against Ghosts, of course, but there is nothing besides
establishment of a clear upstream/downstream structure and strong
filtering on the downstream BGP sessions to fix *that*.
Gert Doering
-- NetMaster
--
Total number of prefixes smaller than registry allocations: 45583 (46871)
SpaceNet AG Mail: netmaster@Space.Net
Joseph-Dollinger-Bogen 14 Tel : +49-89-32356-0
80807 Muenchen Fax : +49-89-32356-299