|
Ticket ID: SIXXS #14350609 Ticket Status: Resolved PoP: uschi03 - Team Cymru (Chicago, Illinois)
Routing issues
![[us]](/s/countries/us.gif) Shadow Hawkins on Saturday, 03 October 2015 09:06:33
Uschi03 seems to have a routing problem after it restarted where things used to work fine.
On a machine that uses sixxs for IPv6 connectivity a ping to google works fine:
[leendert@amsterdam ~]$ ping6 www.google.com
PING www.google.com(dfw06s32-in-x10.1e100.net) 56 data bytes
64 bytes from dfw06s32-in-x10.1e100.net: icmp_seq=1 ttl=55 time=77.1 ms
64 bytes from dfw06s32-in-x10.1e100.net: icmp_seq=2 ttl=55 time=70.0 ms
^C
However, if I try ping a different IPV6 connected machine things don't get through:
[leendert@amsterdam ~]$ ping6 madrid.paramecium.org
PING madrid.paramecium.org(madrid.paramecium.org) 56 data bytes
^C
--- madrid.paramecium.org ping statistics ---
39 packets transmitted, 0 received, 100% packet loss, time 37999ms
On the remote machine that's not connected to IPv6 using sixxs (aka madrid) connectivity to google works fine:
leendert@madrid ~]$ ping6 dfw06s32-in-x10.1e100.net
PING dfw06s32-in-x10.1e100.net(dfw06s32-in-x10.1e100.net) 56 data bytes
64 bytes from dfw06s32-in-x10.1e100.net: icmp_seq=1 ttl=59 time=0.534 ms
64 bytes from dfw06s32-in-x10.1e100.net: icmp_seq=2 ttl=59 time=0.504 ms
^C
but trying to reach the sixxs connected machine (amsterdam) from madrid fails:
[leendert@madrid ~]$ ping6 berlin.paramecium.org
PING berlin.paramecium.org(berlin.paramecium.org) 56 data bytes
From ae4-51.dal33.ip6.gtt.net icmp_seq=1 Destination unreachable: No route
From ae4-51.dal33.ip6.gtt.net icmp_seq=2 Destination unreachable: No route
From ae4-51.dal33.ip6.gtt.net icmp_seq=3 Destination unreachable: No route
^C
--- berlin.paramecium.org ping statistics ---
3 packets transmitted, 0 received, +3 errors, 100% packet loss, time 2596ms
A traceroute from amsterdam to madrid (sixxs to non-sixxs connected machine) seems to indicate that things get stuck at uschi03.
It looks like some configuration/route didn't get restored after the recent outage of uschi03 because I haven't had these problems before.
State change: remoteproblem
The state of this ticket has been changed to remoteproblem
Routing issues
Pings are not very informative, they tell if something works or not.
Please provide traceroutes.
Routing issues
Shadow Hawkins on Saturday, 03 October 2015 16:17:05
Added the requested traceroutes.
From the sixxs connected machine (amsterdam):
[leendert@amsterdam ~]$ traceroute -6 madrid.paramecium.org
traceroute to madrid.paramecium.org (2607:f740:100::f18), 30 hops max, 80 byte packets
1 gw-442.chi-03.us.sixxs.net (2604:8800:100:1b9::1) 82.368 ms 82.338 ms 82.141 ms
2 uschi03.sixxs.net (2620:0:6b0:a:250:56ff:fe99:78f7) 81.960 ms 81.938 ms *
3 * * *
4 * * *
5 * * *
6 * * *
7 * * *
8 * * *
9 * * *
10 * * *
11 * * *
12 * * *
13 * * *
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *
19 * * *
20 * * *
21 * * *
22 * * *
23 * * *
24 * * *
25 * * *
26 * * *
27 * * *
28 * * *
29 * * *
30 * * *
And from the non-sixxs connected machine:
[leendert@madrid ~]$ traceroute -6 amsterdam
traceroute to amsterdam (2604:8800:164::1), 30 hops max, 80 byte packets
1 2607:f740:100::2 (2607:f740:100::2) 2.866 ms 2.773 ms 2.724 ms
2 2607:f740:100::3 (2607:f740:100::3) 0.358 ms 0.312 ms 0.254 ms
3 ae4-51.dal33.ip6.gtt.net (2001:668:0:3::8000:4a1) 0.533 ms !N 0.487 ms !N 0.425 ms !N
Interestingly, routing is even broken within the sixxs enclave. I have a second tunnel that uses a Boston POP and it gets stuck too:
[leendert@peanut ~]$ traceroute -6 amsterdam.paramecium.org
traceroute to amsterdam.paramecium.org (2604:8800:164::1), 30 hops max, 80 byte packets
1 gw-460.bos-01.us.sixxs.net (2001:4830:1100:1cb::1) 92.380 ms 92.201 ms 92.243 ms
2 sixxs-ic-1139-bos.customer.occaid.net (2001:4830:e1:b::2) 92.528 ms 92.760 ms 93.237 ms
3 dcr01-ve402.bstn01.occaid.net (2001:4830:e1:b::1) 93.176 ms 93.063 ms 93.311 ms
4 dcr01-ve402.bstn01.occaid.net (2001:4830:e1:b::1) 93.232 ms !N 93.685 ms !N 93.612 ms !N
Hopefully this helps diagnosing the problem.
Thanks, Leendert
Routing issues
That is obviously a transit issue. I'll forward to Team Cymru to see if they can peek at it as it looks like one of their transit providers is doing something odd.
State change: resolved
The state of this ticket has been changed to resolved
|
![[ch]](/s/countries/ch.gif)
on Saturday, 03 October 2015 09:16:21
Posting is only allowed when you are