IPv6 Over Level3 Down?

Portal Home > Knowledgebase > Industry Announcements > Web Hosting Main Forums > Providers and Network Outages and Updates > IPv6 Over Level3 Down?

Posted by kbeezie, 06-05-2012, 06:45 AM
I have a personal VPS down in the Texas Location that appears to be having problems with IPv6 it doesn't seem to be host specific so wondering if something is up with level3.

For example using my tunneled IPv6 connection from He.net from home, as well as my dedicated server in Kansas City, Mo (which also happens to be over he.net IPv6 native), I can ping6/traceroute6 ipv6.google.com like so:

From Home:

Quote:

traceroute6 ipv6.google.com
traceroute6 to ipv6.l.google.com (2607:f8b0:400f:801::1013) from 2001:470:1f11:a45::, 64 hops max, 12 byte packets
1 kbeeziecom-1.tunnel.tserv9.chi1.ipv6.he.net 47.964 ms 48.006 ms 49.126 ms
2 gige-g3-4.core1.chi1.he.net 48.442 ms 49.411 ms 49.573 ms
3 2001:470:0:bf::2 44.790 ms 44.076 ms 44.023 ms
4 2001:4860::1:0:92e 44.508 ms
2001:4860::1:0:3f7 51.808 ms
2001:4860::1:0:92e 44.636 ms
5 2001:4860::8:0:2fea 46.358 ms
2001:4860::8:0:2fe9 45.619 ms 63.565 ms
6 2001:4860::8:0:281e 59.363 ms
2001:4860::8:0:281d 55.687 ms 56.293 ms
7 2001:4860::8:0:3426 65.842 ms 65.238 ms 87.545 ms
8 2001:4860::1:0:7a4 65.260 ms 65.966 ms 66.561 ms
9 2001:4860:0:1::593 89.957 ms 67.285 ms 65.412 ms
10 2607:f8b0:400f:801::7 64.867 ms 66.306 ms 64.620 ms

and from the dedicated server in Missouri

Quote:

traceroute6 ipv6.google.com
traceroute to ipv6.google.com (2607:f8b0:4009:801::1012), 30 hops max, 40 byte packets
1 (2001:4858:aaaa:6::1) 0.556 ms 0.758 ms 0.974 ms
2 (2001:4858::125) 1.157 ms 1.250 ms 1.353 ms
3 10gigabitethernet1-3.core1.mci2.he.net (2001:470:1:252::1) 1.123 ms 1.121 ms 1.216 ms
4 10gigabitethernet1-4.core1.mci1.he.net (2001:470:0:179::1) 1.103 ms 1.199 ms 1.192 ms
5 10gigabitethernet6-1.core1.chi1.he.net (2001:470:0:15b::1) 13.019 ms 13.012 ms 13.123 ms
6 2001:470:0:bf::2 (2001:470:0:bf::2) 11.477 ms 10.821 ms 10.874 ms
7 2001:4860::1:0:92e (2001:4860::1:0:92e) 11.552 ms 11.544 ms 11.527 ms
8 2001:4860:0:1::340 (2001:4860:0:1::340) 11.785 ms 12.021 ms 11.869 ms
9 2607:f8b0:8000:a::8 (2607:f8b0:8000:a::8) 11.058 ms 11.056 ms 11.047 ms

But from the VPS in Texas which appears to be on Level3

Quote:

raceroute6 ipv6.google.com
traceroute6 to ipv6.l.google.com (2607:f8b0:4002:801::1013) from 2607:fc50:1000:e100::2, 64 hops max, 12 byte packets
1 2607:fc50:1000:e100::1 0.584 ms 0.362 ms 0.340 ms
2 2607:fc50:1000:2::1 0.397 ms 0.259 ms 0.277 ms
3 2607:fdb8:8000:700::1 0.597 ms 0.485 ms 0.454 ms
4 2607:fdb8:0:118::1 33.258 ms 33.067 ms 33.105 ms
5 2607:fdb8::1:2:0:1 33.108 ms 33.032 ms 33.118 ms
6 2001:438:fffe::219 1.060 ms
te-5-3-0.edge3.Dallas1.Level3.net 33.176 ms
2001:438:fffe::219 1.054 ms
7 vl-80.edge2.Dallas3.Level3.net 33.862 ms
2001:438:ffff::407d:1acd 1.287 ms
vl-90.edge2.Dallas3.Level3.net 33.469 ms
8 * * *
9 * * *
10 * * *
11 * * *
12 * * *
13 * * *
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *
19 * * *
20 * * *
21 * * *
22 * * *
23 * * *
24 * * *
25 * * *
26 * * *
27 * * *
28 * * *
29 * * *
30 * * *
31 *^C

Traceroute6 from Dedicated Server to VPS in Texas seems fine

Quote:

traceroute6 2607:fc50:1000:e100:216:3eff:fe49:6e69
traceroute to 2607:fc50:1000:e100:216:3eff:fe49:6e69 (2607:fc50:1000:e100:216:3eff:fe49:6e69), 30 hops max, 40 byte packets
1 (2001:4858:aaaa:6::1) 0.482 ms 0.677 ms 0.818 ms
2 (2001:4858::125) 1.127 ms 1.121 ms 1.250 ms
3 (2001:550:2:44::4:1) 1.425 ms 1.661 ms 1.774 ms
4 te0-0-0-5.ccr22.dfw01.atlas.cogentco.com (::ffff:154.54.5.157) 11.496 ms te0-4-0-5.ccr22.dfw01.atlas.cogentco.com (::ffff:154.54.46.214) 11.483 ms 11.657 ms
5 (2001:550:2:33::1) 11.767 ms 11.921 ms 12.057 ms
6 (2001:550:2:33::2) 11.327 ms 10.570 ms 10.655 ms
7 2607:fc50:1000:2::2 (2607:fc50:1000:2::2) 10.676 ms 10.728 ms 10.809 ms
8 bsd.ionvz.com (2607:fc50:1000:e100:216:3eff:fe49:6e69) 10.795 ms 10.691 ms 10.784 ms

tracepath6 2607:fc50:1000:e100:216:3eff:fe49:6e69
1?: [LOCALHOST] pmtu 1500
1: 2001:4858:aaaa:6::1 0.811ms
2: 2001:4858::125 0.835ms
3: 2001:550:2:44::4:1 1. 29ms
4: te0-0-0-5.ccr22.dfw01.atlas.cogentco.com asymm 8 11.101ms
5: 2001:550:2:33::1 11.275ms
6: 2001:550:2:33::2 11. 63ms
7: 2607:fc50:1000:2::2 11.548ms
8: bsd.ionvz.com 11.336ms reached
Resume: pmtu 1500 hops 8 back 8

But if I try to traceroute6 to my VPS from Home it's

Quote:

traceroute6 2607:fc50:1000:e100:216:3eff:fe49:6e69
traceroute6 to 2607:fc50:1000:e100:216:3eff:fe49:6e69 (2607:fc50:1000:e100:216:3eff:fe49:6e69) from 2001:470:1f11:a45::, 64 hops max, 12 byte packets
1 kbeeziecom-1.tunnel.tserv9.chi1.ipv6.he.net 49.134 ms 49.381 ms 51.523 ms
2 gige-g3-4.core1.chi1.he.net 47.367 ms 54.181 ms 46.929 ms
3 10gigabitethernet3-2.core1.den1.he.net 68.106 ms 68.059 ms 69.317 ms
4 10gigabitethernet11-4.core1.sjc2.he.net 104.595 ms 97.682 ms 97.861 ms
5 10gigabitethernet3-2.core1.pao1.he.net 95.027 ms 102.883 ms 99.777 ms
6 * * *
7 * * *
8 * * *

From VPS to Home

Quote:

traceroute6 2001:470:1f11:a45::
traceroute6 to 2001:470:1f11:a45:: (2001:470:1f11:a45: from 2607:fc50:1000:e100::2, 64 hops max, 12 byte packets
1 2607:fc50:1000:e100::1 0.593 ms 0.350 ms 0.354 ms
2 2607:fc50:1000:2::1 0.935 ms 0.333 ms 0.326 ms
3 2607:fdb8:8000:700::1 0.638 ms 0.574 ms 0.461 ms
4 2607:fdb8:0:118::1 33.671 ms 33.127 ms 33.183 ms
5 2607:fdb8::1:2:0:1 33.194 ms 33.090 ms 33.124 ms
6 2001:438:fffe::219 1.043 ms
te-5-3-0.edge3.Dallas1.Level3.net 33.376 ms
2001:438:fffe::219 1.294 ms
7 vl-80.car1.Dallas1.Level3.net 33.569 ms
8 * * *
9 * * *
10 * * *
11 * * *
12 * * *
13 * * *
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *
19 * * *
20 * * *
21 * * ^C

So wondering if maybe there's some IPv6 routing issues over Level3 as I cannot seem to ping, connect, etc to my VPS via IPv6
Posted by FastServ, 06-05-2012, 08:09 AM
You could post this type of thing on the outages mailing list where there are ipv6 savvy network operators rather than web hosts.

http://puck.nether.net/mailman/listinfo/outages
Posted by kbeezie, 06-05-2012, 08:12 AM

Quote:

Originally Posted by FastServ

You could post this type of thing on the outages mailing list where there are ipv6 savvy network operators rather than web hosts.

http://puck.nether.net/mailman/listinfo/outages

The VPS provider was able to reproduce the problem within their network as such : " have put in a ticket with one of our upstreams where I see ipv6 trace route dying"

I don't do too well with mailing lists seems like an odd means given the means of communications we have available to us today. (i.e.: forums are so much easier to follow and search).