
jeroen at unfix
May 4, 2012, 7:11 AM
Post #6 of 7
(571 views)
Permalink
|
On 2012-05-04 15:19 , Mark Pace Balzan wrote: > > Marco- thanks, its possibly bit torrent, but as Jeroen points out > both src and dst are 2001:0 (teredo) > > Jeroen - yes RPF is of course critical, but im also interested in why > this is happening at all... Obviously you are not applying proper RPF for your traffic otherwise... > >> But, as we do not know if he has a default route or anything else >> on it it is hard to tell why his box is even forwarding these >> packets. > > v6 default on the relay points out to the v6 internet and purpose in > life of this box is just 6to4 :) > > Performing a packet trace on packets (v4 and v6) incoming into the > relay box, shows that said teredo packets (ie 2001:0 in both src and > dst) have v4 source address belonging to v4 unicast users on my > network and v4 destination being 192.88.99.1 which is the 6to4 > anycast ip. 6to4 tunnels are just proto-41 tunnels, with one difference, they should only handle packets where a source address on the tunnel side is 6to4 (thus 2002::/16). Clearly you are accepting proto-41 traffic with any kind of source address, otherwise this could not be happening. Greets, Jeroen
|