Hey all,

We've just had two reports on [email protected] about ipv6 connectivity issues. Normally they are caused by client / route issues, however if I try via claranet I get failures too:

http://noc.eu.clara.net/lg.php
Router: US
Command: ping ipv6 python.org

Sending 5, 100-byte ICMP Echos to 2001:888:2000:D::A2, timeout is 2 seconds:
.....
Success rate is 0 percent (0/5)

Reports below:

It seems that your IPv6 network is down:

[luk@bender:~]$ traceroute6 python.org # from my workplace
traceroute to python.org (2001:888:2000:d::a2), 30 hops max, 80 byte packets
 1  spider1.mimuw.edu.pl (2001:6a0:5001:1::1)  0.273 ms  0.295 ms  0.293 ms
2 kp-mm-r3.rtr.net.icm.edu.pl (2001:6a0:4001:1::1) 4.794 ms 4.826 ms 4.822 ms 3 kp-mm-r3.rtr.net.icm.edu.pl (2001:6a0:1:4::1) 5.045 ms 5.069 ms 5.254 ms
 4  2001:6a0:0:19::1 (2001:6a0:0:19::1)  1.396 ms  1.394 ms  1.401 ms
5 2001:b10:c000:204::1 (2001:b10:c000:204::1) 6.499 ms 6.501 ms 6.513 ms 6 pionier-bckp.rt1.poz.pl.geant2.net (2001:798:23:10aa::5) 6.476 ms 5.960 ms 5.975 ms 7 so-7-1-0.rt1.fra.de.geant2.net (2001:798:cc:1401:2301::1) 18.706 ms 18.842 ms 18.876 ms 8 2001:450:2002:70::1 (2001:450:2002:70::1) 148.335 ms 144.553 ms 143.525 ms 9 2001:450:2002:1e2::2 (2001:450:2002:1e2::2) 27.310 ms 27.305 ms 27.300 ms
10  * * *
11  * * *
12  * * *
13 kpn-ic-135436-adm-b5.c.telia.net (2001:2000:3080:6f::2) 48.579 ms 48.601 ms 48.554 ms
14  * * *
15  * * *
16  * * *
17  * * *
18  * * *
19  * * *
20  * * *
21  * * *
22  * * *
23  * * *
24  * * *
25  * * *
26  * * *
27  * * *
28  * * *
29  * * *
30  * * *


I'd like to just report that python.org (and subdomains) is not avaiable (lost 100% packets, no ICMP responses) from some locations over IPv6. The problem started yesterday in the evening CEST. I'm observing the problem from the University of Warsaw network which is one of the largest .edu networks in Poland. Traceroute6 reports:

root@schowek:/home/accek/foo# traceroute6 pypi.python.org
traceroute to pypi.python.org (2001:888:2000:d::a8) from 2001:6a0:5001:2:212:3fff:fe78:660, 30 hops max, 16 byte packets
 1  spider2.mimuw.edu.pl (2001:6a0:5001:2::1)  0.237 ms  0.202 ms  0.207 ms
2 kp-mm-r3.rtr.net.icm.edu.pl (2001:6a0:4001:1::1) 4.229 ms 4.175 ms 4.115 ms 3 kp-mm-r3.rtr.net.icm.edu.pl (2001:6a0:1:4::1) 4.085 ms 3.824 ms 4.079 ms
 4  2001:6a0:0:19::1 (2001:6a0:0:19::1)  6.328 ms  0.821 ms  0.736 ms
5 2001:b10:c000:204::1 (2001:b10:c000:204::1) 6.136 ms 5.953 ms 5.894 ms 6 * pionier-bckp.rt1.poz.pl.geant2.net (2001:798:23:10aa::5) 6.249 ms 5.952 ms 7 so-7-1-0.rt1.fra.de.geant2.net (2001:798:cc:1401:2301::1) 18.643 ms 18.661 ms 18.659 ms 8 2001:450:2002:70::1 (2001:450:2002:70::1) 112.468 ms 76.75 ms 18.881 ms 9 2001:450:2002:1e2::2 (2001:450:2002:1e2::2) 44.23 ms 27.581 ms 27.315 ms
10  * * *
11  * * *
12  * * *
13 kpn-ic-135436-adm-b5.c.telia.net (2001:2000:3080:6f::2) 49.17 ms 49.016 ms 77.695 ms
14  * * *
15  * * *
16  * * *
17  * * *
18  * * *
19  * * *
20  * * *

It looks like a routing problem in the core IPv6. I don't know the exact area affected, but entries like 'telia.net' or 'geant2.net' suggest it's not a local issue. Access via some Teredo server in the Netherlands seems ok.

Maybe it's a good idea to drop AAAA DNS records for python.org temporarily, especially that easy_install and such are broken on affected networks with IPv6-enabled hosts (and automatic fallback to IPv4 does not always work, at least in my case).

All the best,

Michael Foord

--
http://www.voidspace.org.uk/

May you do good and not evil
May you find forgiveness for yourself and forgive others
May you share freely, never taking more than you give.
-- the sqlite blessing http://www.sqlite.org/different.html

_______________________________________________
pydotorg-www mailing list
[email protected]
http://mail.python.org/mailman/listinfo/pydotorg-www

Reply via email to