• Welcome to Hurricane Electric's IPv6 Tunnel Broker Forums.

Router can ping6, but connected computers can't.

Started by colincogle, October 10, 2008, 12:26:13 AM

Previous topic - Next topic

colincogle

This is an odd issue.

I've got a router running OpenWRT 7.09 "Kamikaze", and several computers (Mac, Windows, and Linux) on my home network.  I've set up the tunnel on my router, and I can ping6, tcptraceroute6, etc. without a problem.

Other computers on the network, however, can't do anything.  They get a stateless IPv6 address from my /64 subnet almost instantly after joining the network, but all IPv6 traffic after that seems to blackhole.  I can type ipv6.google.com into my address bar, or try to ping6 tunnelbroker.net, but everything just times out;  nslookup can get AAAA records from my router, though I bet the actual lookup is over IPv4.

I've followed many guides, but it seems like I've forgotten a step.

Router's /etc/radvd.conf
interface br-lan
{
   AdvSendAdvert on;
   AdvHomeAgentFlag on;
   MinRtrAdvInterval 3;
   MaxRtrAdvInterval 10;

   prefix 2001:470:1f07:951::/64
   {
      AdvOnLink on;
      AdvAutonomous on;
      AdvRouterAddr on;
   };

};


Router's IP addresses (sans MAC addresses, link-local addresses, etc):
1: lo: <LOOPBACK,UP> mtu 16436 qdisc noqueue 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 brd 127.255.255.255 scope host lo
    inet6 ::1/128 scope host
2: eth0: <BROADCAST,MULTICAST,PROMISC,UP> mtu 1500 qdisc pfifo_fast qlen 1000
    link/ether 00:16:b6:xx:xx:xx brd ff:ff:ff:ff:ff:ff
    inet6 fe80::216:dead:beef:cafe/64 scope link
3: eth0.0: <BROADCAST,MULTICAST,PROMISC,UP> mtu 1500 qdisc noqueue
    link/ether 00:16:b6:xx:xx:xx brd ff:ff:ff:ff:ff:ff
    inet6 fe80::216:dead:beef:cafe/64 scope link
4: eth0.1: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc hfsc qlen 5
    link/ether 00:16:b6:xx:xx:xx brd ff:ff:ff:ff:ff:ff
    inet MY_IPV4_ADDRESS/23 brd 255.255.255.255 scope global eth0.1
    inet6 fe80::216:dead:beef:cafe/64 scope link
5: sit0@NONE: <NOARP> mtu 1480 qdisc noop
    link/sit 0.0.0.0 brd 0.0.0.0
6: br-lan: <BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue
    link/ether 00:16:b6:xx:xx:xx brd ff:ff:ff:ff:ff:ff
    inet 192.168.1.1/24 brd 192.168.1.255 scope global br-lan
    inet6 fe80::200:ff:fe00:0/64 scope link
    inet6 2001:470:myv6:subnet::1/64 scope global
7: wl0: <BROADCAST,MULTICAST,PROMISC,UP> mtu 1500 qdisc pfifo_fast qlen 1000
    link/ether 00:16:b6:xx:xx:zz brd ff:ff:ff:ff:ff:ff
    inet6 fe80::216:dead:beef:cafe/64 scope link
22: he-ipv6@NONE: <POINTOPOINT,NOARP,UP> mtu 1480 qdisc noqueue
    link/sit MY_IPV4_ADDRESS peer 209.51.161.14
    inet6 2001:470:myv6:subnet::2/64 scope global
    inet6 fe80::43ac:272f/128 scope link

broquea

When you run a traceroute6 from a machine that got autoconfigured, does it at least make it 1 hop to the OpenWRT router and then fail, or simply no where at all?
I've verified on the tunnel-server that your routed /64 is indeed statically routed to your side of the tunnel.
Your radvd.conf looks fine, intervals are pretty low, but that just ensures things happen faster :)

kriteknetworks