Can't reach Google IPv6, TWC -> XO -> ???
I suspect this is just me, but putting it out there. Coming from Time Warner Cable trying to reach Google, it dies inside XO. Anyone else seeing this? netsville@silliac2:~$ mtr -rw -c 3 google.com Start: Sat Sep 10 13:26:29 2016 HOST: silliac2 Loss% Snt Last Avg Best Wrst StDev 1.|-- roc-core2-gig21.netsville.com 0.0% 3 0.3 0.3 0.3 0.4 0.0 2.|-- 2604:6000:0:8::f:214 0.0% 3 0.4 0.4 0.4 0.4 0.0 3.|-- 2604:6000:0:4::f:155 0.0% 3 9.2 7.6 6.6 9.2 1.4 4.|-- 2604:6000:0:4:0:2006:0:21f2 0.0% 3 1.2 1.3 1.2 1.3 0.0 5.|-- 2604:6000:0:4::d0 0.0% 3 8.2 9.6 8.2 10.8 1.0 6.|-- 2001:1998:0:8::2a0 0.0% 3 34.0 34.9 34.0 35.4 0.7 7.|-- 2001:1998:0:4::102 0.0% 3 28.6 28.6 28.5 28.6 0.0 8.|-- 2610:18:10c:9000::15 0.0% 3 28.9 28.9 28.8 28.9 0.0 9.|-- 2610:18::5101 0.0% 3 29.2 29.0 28.9 29.2 0.0 10.|-- 2610:18::2071 0.0% 3 29.1 29.1 29.1 29.1 0.0 11.|-- ??? 100.0 3 0.0 0.0 0.0 0.0 0.0
If anyone has IPv6 eyes inside XO (their Looking Glass page doesn't support it), can you trace to 2606:c380::1994? I would appreciate it. On 09/10/2016 01:34 PM, Micah Brandon via Outages wrote:
I suspect this is just me, but putting it out there. Coming from Time Warner Cable trying to reach Google, it dies inside XO. Anyone else seeing this?
I couldn't hit. Netflix earlier via IPv6 HE tunnel. Had to disable it and use IPv4 On Sat, Sep 10, 2016 at 8:56 PM Micah Brandon via Outages < outages@outages.org> wrote:
If anyone has IPv6 eyes inside XO (their Looking Glass page doesn't support it), can you trace to 2606:c380::1994? I would appreciate it.
On 09/10/2016 01:34 PM, Micah Brandon via Outages wrote:
I suspect this is just me, but putting it out there. Coming from Time Warner Cable trying to reach Google, it dies inside XO. Anyone else seeing this?
_______________________________________________ Outages mailing list Outages@outages.org https://puck.nether.net/mailman/listinfo/outages
The Looking Glass page is not maintained by Engineering. I did ask our Ops team to take a look into it, as we don't appear to be learning 2606:c380::/32 directly from 7843 and to ensure there are no other complaints about reachability to Google. There are a couple of issues relating to the return path and it definitely needs TWC's involvement to address so if your concern is about the return path, any response is likely to have TWC open a ticket. The last hop in your traceroute is where we hand it off to Google's backbone AS and the traffic enters and exits in NY as expected based on your mtr. To get back to you we use HE and you are reachable via ICMP. I can't speak to how the first starred hop in your mtr reaches you. I didn't bother sending a bunch of other trace routes which show things working, but from that last hop here is your return traceroute. traceroute inet6 2606:c380::1994 traceroute6 to 2606:c380::1994 (2606:c380::1994) from 2610:18::2071, 64 hops max, 12 byte packets 1 2610:18:10:1::2 (2610:18:10:1::2) 11.641 ms 1.184 ms 0.611 ms 2 twcable-backbone-as7843.10gigabitethernet2.switch1.nyc7.he.net (2001:470:0:2bd::2) 1.386 ms 0.628 ms 0.634 ms 3 2001:1998:0:4::105 (2001:1998:0:4::105) 6.064 ms 5.019 ms 2001:1998:0:4::394 (2001:1998:0:4::394) 4.840 ms 4 2001:1998:0:4::a2 (2001:1998:0:4::a2) 31.979 ms 29.301 ms 31.959 ms 5 2001:1998:0:8::2a5 (2001:1998:0:8::2a5) 43.977 ms 2001:1998:0:8::2a7 (2001:1998:0:8::2a7) 42.679 ms 2001:1998:0:8::2a5 (2001:1998:0:8::2a5) 42.380 ms 6 2604:6000:0:4::cd (2604:6000:0:4::cd) 40.389 ms 40.143 ms 40.418 ms 7 2604:6000:0:4:0:2006:0:21f1 (2604:6000:0:4:0:2006:0:21f1) 43.659 ms 42.356 ms 43.579 ms 8 2604:6000:0:4::f:156 (2604:6000:0:4::f:156) 28.867 ms 28.846 ms 28.924 ms 9 2604:6000:0:8::f:215 (2604:6000:0:8::f:215) 29.048 ms 28.992 ms 29.026 ms 10 * * * 11 * * * ^C {master}
From Hop #10 in mtr 2610:18::2071
ping 2606:c380::1994 PING6(56=40+8+8 bytes) 2610:18::2071 --> 2606:c380::1994 16 bytes from 2606:c380::1994, icmp_seq=0 hlim=56 time=29.121 ms 16 bytes from 2606:c380::1994, icmp_seq=1 hlim=56 time=29.029 ms 16 bytes from 2606:c380::1994, icmp_seq=2 hlim=56 time=29.059 ms 16 bytes from 2606:c380::1994, icmp_seq=3 hlim=56 time=29.638 ms 16 bytes from 2606:c380::1994, icmp_seq=4 hlim=56 time=29.065 ms 16 bytes from 2606:c380::1994, icmp_seq=5 hlim=56 time=29.099 ms 16 bytes from 2606:c380::1994, icmp_seq=6 hlim=56 time=29.022 ms 16 bytes from 2606:c380::1994, icmp_seq=7 hlim=56 time=29.122 ms 16 bytes from 2606:c380::1994, icmp_seq=8 hlim=56 time=29.164 ms 16 bytes from 2606:c380::1994, icmp_seq=9 hlim=56 time=29.089 ms ^C --- 2606:c380::1994 ping6 statistics --- 10 packets transmitted, 10 packets received, 0% packet loss round-trip min/avg/max/std-dev = 29.022/29.141/29.638/0.171 ms Sean 2016-09-10 20:55 GMT-04:00 Micah Brandon via Outages <outages@outages.org>:
If anyone has IPv6 eyes inside XO (their Looking Glass page doesn't support it), can you trace to 2606:c380::1994? I would appreciate it.
On 09/10/2016 01:34 PM, Micah Brandon via Outages wrote:
I suspect this is just me, but putting it out there. Coming from Time Warner Cable trying to reach Google, it dies inside XO. Anyone else seeing this?
_______________________________________________ Outages mailing list Outages@outages.org https://puck.nether.net/mailman/listinfo/outages
Regarding this issue, if anyone from Google can respond to me offlist or point me in the right direction to report IPv6 issue -- or perhaps do a trace for me to 2606:c380::1994 to see where it stops. It appears that Google has filtered my 2606:c380::/32 block for some reason. It started ~9am EDT Friday 9/9. Last successful gmail msg left here 8:54a and then all IPv6 attempts timeout after 9:03a. Another user also reported their Gmail stopped picking up mail from here almost the same time. Had to use an IPv4 host to get that to work again. Most places at least see the TWC -> HE path and make it back to me just fine (including XO). So I'm unable to find another site that can't reach me. On 09/10/2016 01:34 PM, Micah Brandon via Outages wrote:
I suspect this is just me, but putting it out there. Coming from Time Warner Cable trying to reach Google, it dies inside XO. Anyone else seeing this?
netsville@silliac2:~$ mtr -rw -c 3 google.com Start: Sat Sep 10 13:26:29 2016 HOST: silliac2 Loss% Snt Last Avg Best Wrst StDev 1.|-- roc-core2-gig21.netsville.com 0.0% 3 0.3 0.3 0.3 0.4 0.0 2.|-- 2604:6000:0:8::f:214 0.0% 3 0.4 0.4 0.4 0.4 0.0 3.|-- 2604:6000:0:4::f:155 0.0% 3 9.2 7.6 6.6 9.2 1.4 4.|-- 2604:6000:0:4:0:2006:0:21f2 0.0% 3 1.2 1.3 1.2 1.3 0.0 5.|-- 2604:6000:0:4::d0 0.0% 3 8.2 9.6 8.2 10.8 1.0 6.|-- 2001:1998:0:8::2a0 0.0% 3 34.0 34.9 34.0 35.4 0.7 7.|-- 2001:1998:0:4::102 0.0% 3 28.6 28.6 28.5 28.6 0.0 8.|-- 2610:18:10c:9000::15 0.0% 3 28.9 28.9 28.8 28.9 0.0 9.|-- 2610:18::5101 0.0% 3 29.2 29.0 28.9 29.2 0.0 10.|-- 2610:18::2071 0.0% 3 29.1 29.1 29.1 29.1 0.0 11.|-- ??? 100.0 3 0.0 0.0 0.0 0.0 0.0
Just wanted to close this out. The issue has been resolved. If anyone wants details, they can email me. Thanks to everyone that provided information and traces, especially to XO that provided multiple peeks into their network. On 09/10/2016 01:34 PM, Micah Brandon via Outages wrote:
I suspect this is just me, but putting it out there. Coming from Time Warner Cable trying to reach Google, it dies inside XO. Anyone else seeing this?
netsville@silliac2:~$ mtr -rw -c 3 google.com Start: Sat Sep 10 13:26:29 2016 HOST: silliac2 Loss% Snt Last Avg Best Wrst StDev 1.|-- roc-core2-gig21.netsville.com 0.0% 3 0.3 0.3 0.3 0.4 0.0 2.|-- 2604:6000:0:8::f:214 0.0% 3 0.4 0.4 0.4 0.4 0.0 3.|-- 2604:6000:0:4::f:155 0.0% 3 9.2 7.6 6.6 9.2 1.4 4.|-- 2604:6000:0:4:0:2006:0:21f2 0.0% 3 1.2 1.3 1.2 1.3 0.0 5.|-- 2604:6000:0:4::d0 0.0% 3 8.2 9.6 8.2 10.8 1.0 6.|-- 2001:1998:0:8::2a0 0.0% 3 34.0 34.9 34.0 35.4 0.7 7.|-- 2001:1998:0:4::102 0.0% 3 28.6 28.6 28.5 28.6 0.0 8.|-- 2610:18:10c:9000::15 0.0% 3 28.9 28.9 28.8 28.9 0.0 9.|-- 2610:18::5101 0.0% 3 29.2 29.0 28.9 29.2 0.0 10.|-- 2610:18::2071 0.0% 3 29.1 29.1 29.1 29.1 0.0 11.|-- ??? 100.0 3 0.0 0.0 0.0 0.0 0.0
participants (3)
-
Jamie Stephens -
L Sean Kennedy -
Micah Brandon