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

Can't reach anything but endpoints

Started by mattventura, November 28, 2009, 08:23:43 PM

Previous topic - Next topic

mattventura

I configured my openWRT using the tutorial on the openWRT wiki, with this script:
#!/bin/sh /etc/rc.common
LOCALIP=71.231.208.248
POPIP=216.218.226.238
LOCTUN=2001:470:a:122::2
REMTUN=2001:470:a:122::1

START=50

start()
{
       echo -n "Starting  he.net IPv6 tunnel: "
       ip tunnel add he-ipv6 mode sit  remote $POPIP local $LOCALIP ttl 255
       ip link set he-ipv6 up
       ip link set mtu 1280 dev he-ipv6
       ip tunnel change he-ipv6 ttl 64

       ip -6 addr add $LOCTUN/64 dev he-ipv6
       ip -6 route add ::/0 dev he-ipv6
       ip -6 ro add default via $REMTUN dev he-ipv6
       ip -f inet6 addr
       echo "Done."
}
stop()
{
       echo -n "Stopping he.net IPv6 tunnel: "
       ip link set he-ipv6 down
       ip tunnel del he-ipv6
       echo "Done."
}
restart()
{
       stop
       start
}


Now, the tunnel worked fine until now. But starting a few hours ago, I can only ping the tunnel endpoints (2001:470:a::1 and ::2). If I traceroute6 from my server to something else it gets to my router, then to the endpoint, but then dies.


traceroute to www.kame.net (2001:200:0:8002:203:47ff:fea5:3085) from 2001:470:b:122:20a:5eff:fe77:bd47, 30 hops max, 16 byte packets
1  2001:470:b:122::1 (2001:470:b:122::1)  0.501 ms  0.678 ms  0.589 ms
2  mattventura-1.tunnel.tserv14.sea1.ipv6.he.net (2001:470:a:122::1)  60.005 ms  60.25 ms  59.503 ms
3  * * *
4  * * *
...
29  * * *
30  * * *


My IPv4 address is still the same, and I wasn't doing anything with the router at that time. Is this just an outage at the PoP or something?

Update: when I do a reverse DNS lookup on my tunnel's endpoints, I see two names: mattventura-1.tunnel.tserv14.sea1.ipv6.he.net and incdona-1.tunnel.tserv14.sea1.ipv6.he.net. Is this normal?

broquea

Needed to flush configuration issues out of the system, which restored
service.

Looks like the point-to-point /64 didn't get a proper rDNS scrubbing
from when it was de-allocated previously. I can remove the extra entries.