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

traceroute6 not showing actual routing?

Started by vpsnoc, January 16, 2010, 09:45:51 AM

Previous topic - Next topic

vpsnoc

Hello, I've just setup my tunnel one of our production servers. Everything appears to be working, however the traceroute output puzzles me. I was wondering if someone can elaborate.

[root@node47 ~]# traceroute6  ipv6.google.com
traceroute to ipv6.google.com (2001:4860:b009::63), 30 hops max, 40 byte packets
1  vpsnoc-1.tunnel.tserv9.chi1.ipv6.he.net (2001:470:1f10:450::1)  58.625 ms  58.590 ms  58.920 ms
2  1g-3-4.core1.chi1.ipv6.he.net (2001:470:0:6e::1)  58.901 ms  58.882 ms  58.861 ms
3  * * *
4  * * *
5  * * *
6  * * *
7  * * *
8  2001:4860:0:1::33 (2001:4860:0:1::33)  95.853 ms  91.540 ms  91.288 ms
9  * * *
10  * * *
11  * * *
12  * * *
13  * * *
14  * * *
15  * * *
16  * * *
17  * * *
18  * * *
19  * * *
20  * * *
21  * * *
22  * * *
23  * * *
24  * * *
25  * * *
26  * * *
27  * * *
28  * * *
29  * * *
30  * * *
[root@node47 ~]# traceroute6 -T ipv6.google.com
traceroute to ipv6.google.com (2001:4860:b009::68), 30 hops max, 40 byte packets
1  vpsnoc-1.tunnel.tserv9.chi1.ipv6.he.net (2001:470:1f10:450::1)  58.852 ms  59.904 ms  61.030 ms
2  1g-3-4.core1.chi1.ipv6.he.net (2001:470:0:6e::1)  61.296 ms  61.279 ms  61.261 ms
3  qy-in-x68.1e100.net (2001:4860:b009::68)  91.534 ms * *
[root@node47 ~]# traceroute6 -I ipv6.google.com
traceroute to ipv6.google.com (2001:4860:800e::63), 30 hops max, 40 byte packets
1  vpsnoc-1.tunnel.tserv9.chi1.ipv6.he.net (2001:470:1f10:450::1)  58.714 ms  59.836 ms  61.180 ms
2  1g-3-4.core1.chi1.ipv6.he.net (2001:470:0:6e::1)  61.315 ms  61.310 ms  61.606 ms
3  2001:4860:800e::63 (2001:4860:800e::63)  112.291 ms  107.404 ms  107.384 ms

Thank you.


NewtonNet

#1
Hi,

Someone might be able to second guess, but it might help if you said what aspect(s) of it puzzles you?  :)

If it's that the first trace doesn't complete this is likely due to filtering.

Note that if you're looking at drawing comparisons between traces you ought to stick to using the IP address as, as you can see, the DNS resolution can differ hence your destination will change.

Mathew

broquea

Google filters UDP most likely, seeing as ICMP or TCP packets work.

vpsnoc

I was just trying to checkout through where the packets are routed. I highly doubt my tunnel is just 3 hops away from google, more like the output isn't shown because the hosts didn't respond on time/dropped the packets.

broquea

Don't see why you wouldn't be 3 hops, we peer with google. In reality its more hops if you count all the v4 hops your encapsulated packet takes. But from our equipment, a google server would only be 1-2 hops away from the peering session, unless of course they are masking it with ipv6 over mpls.

vpsnoc

Ah I see, that explains everything.
Thank you very much.