0

Routing to Sydney instead of London

Morning,

The routing seems to have gone badly wrong for the secondary anycast, I'm being routed to Sydney instead of London for the secondary IPv6 address.

Diag here: https://nextdns.io/diag/f6e2a510-f981-11ee-abc8-e3316a216bff

For some very strange reason the diag is stating I do not have IPv6, which I do, and it's working fine as can be clearly seen by the below screenshots, so I don't know what's wrong with the diag.

 

 

curl -k "https://[2a07:a8c0::]/info"
{"locationName": "馃嚞馃嚙 London, United Kingdom", "pop": "zepto-lon", "rtt": 11828}%

curl -k "https://[2a07:a8c1::]/info"
{"locationName": "馃嚘馃嚭 Sydney, Australia", "pop": "zetta-syd", "rtt": 258173}%

 I just find it really bizarre that it would route to Sydney of all places, what is going on?

Thanks,

-Will

16 replies

null
    • NextDNs
    • 7 mths ago
    • Reported - view

    Your diag does not seem detect IPv6 enabled on your network, so it does not include the necessary information for us to investigate. Did you run the diag on the system experiencing the issue?

      • Will_Tisdale
      • 7 mths ago
      • Reported - view

       the diag tool is not working correctly, that鈥檚 why. There鈥檚 nothing wrong with my IPv6 connectivity.

      I pointed that fact out in my original post and proved that my IPv6 is working correctly.

      Yes, the diag was ran from the system experiencing the issue. 

      • Will_Tisdale
      • 7 mths ago
      • Reported - view

      another diag, same issue where it claims no ipv6. https://nextdns.io/diag/713068f0-fa5b-11ee-80b9-151cb933a77a
       

      followed by a traceroute to the ipv6 anycast 2 (same box obviously). Sorry for the screenshot, best I can do on my phone. 
       

      • Will_Tisdale
      • 7 mths ago
      • Reported - view

       Back home now and I'm at a complete loss as to why the diag will not detect that I have IPv6 - pings work, traceroute works, ipv6-test and test-ipv6 show no issues - I suspect a bug in the diag tool, I'm happy to assist to debug that.

      Same box as previous message:

      nas.z3n.uk (2001:8b0:6401:82a0:0:ffff:a0a:a05) -> 2a07:a8c1:: (2a07:a8c1::)                                                2024-04-14T18:21:17+0100
      Keys:  Help   Display mode   Restart statistics   Order of fields   quit
                                                                                                                 Packets               Pings
       Host                                                                                                    Loss%   Snt   Last   Avg  Best  Wrst StDev
       1. AS20712  2001:8b0:1111:1111:0:ffff:51bb:b98e                                                          0.0%    13    3.5   3.2   2.2   6.3   1.2
       2. AS20712  2001:8b0:0:53::133                                                                           0.0%    13   30.7  28.9  11.1  56.6  16.3
       3. AS20712  2001:8b0:0:53::104                                                                           0.0%    12   43.6  29.0  11.9  61.2  16.8
       4. (waiting for reply)
       5. (waiting for reply)
       6. (waiting for reply)
       7. (waiting for reply)
       8. (waiting for reply)
       9. AS6939   2001:470:0:2d9::2                                                                            0.0%    12  244.3 246.5 223.4 285.6  15.9
      10. (waiting for reply)
      11. AS???    2001:7fa:11:4:0:1db4:0:1                                                                     8.3%    12  293.9 290.5 261.4 314.6  16.7
      12. AS7604   2403:3400:7604:603::13                                                                       0.0%    12  280.8 290.4 261.6 322.2  16.9
      13. AS34939  2a07:a8c1::                                                                                  0.0%    12  297.8 300.5 268.6 323.2  18.9

      A box I have on Oracle cloud (completely different network) - also going to Sydney for the secondary anycast IPv6:

      cloud (2603:c020:c003:e9ff:8eb8:9447:4bd:9efb) -> 2a07:a8c1:: (2a07:a8c1::)                                2024-04-14T18:23:25+0100
      Keys:  Help   Display mode   Restart statistics   Order of fields   quit
                                                                                                 Packets               Pings
       Host                                                                                    Loss%   Snt   Last   Avg  Best  Wrst StDev
       1. AS31898  2603:c000:300::8c5b:c864                                                     0.0%    25    0.2   0.3   0.2   0.6   0.1
       2. AS???    2001:7f8:be::3:1898:1                                                        0.0%    25    0.9   1.1   0.5  10.0   1.9
       3. AS???    2001:7f8:be::6939:1                                                          0.0%    24    2.0   2.2   1.9   5.0   0.6
       4. (waiting for reply)
       5. (waiting for reply)
       6. (waiting for reply)
       7. (waiting for reply)
       8. AS6939   2001:470:0:2d9::2                                                            0.0%    24  205.3 217.2 204.1 229.8   9.5
       9. (waiting for reply)
      10. AS???    2001:7fa:11:4:0:1db4:0:1                                                     0.0%    24  256.1 263.0 249.0 275.4  10.0
      11. AS7604   2403:3400:7604:603::13                                                       0.0%    24  259.9 264.9 248.9 275.8   9.6
      12. AS34939  2a07:a8c1::                                                                  0.0%    24  273.2 266.7 248.1 273.9   8.9
      

      Something is definitely screwed up in the UK.

      Cheers,

      -Will

      • Will_Tisdale
      • 7 mths ago
      • Reported - view

       Finally got a diag where IPv6 worked, by switching the resolvers from NextDNS to Adguard (go figure). Switching resolvers to my ISP also worked. IPv6 is only not detected when NextDNS resolvers are used, so I've no idea what's going on there.

      https://nextdns.io/diag/fca22330-fa86-11ee-81d0-1b45e70a7a31

      • Will_Tisdale
      • 7 mths ago
      • Reported - view

       This is what is breaking the IPv6 diag:

      • Will_Tisdale
      • 7 mths ago
      • Reported - view

       and another diag, with the aforementioned "AI" related issue resolved:

      https://nextdns.io/diag/6116f700-fb04-11ee-a8ed-31ce4a2f7208

      Hopefully you'll have all you need to resolve this now, but let me know if you need something else. 馃檪

      • NextDNs
      • 7 mths ago
      • Reported - view

       thanks we are looking into it

      • Will_Tisdale
      • 7 mths ago
      • Reported - view

       any progress on this?
       

      I鈥檓 still being routed to Sydney for the secondary anycast IPv6 address, which is just bizarre.

      • NextDNs
      • 7 mths ago
      • Reported - view

       Can you please try another diag?

      • Will_Tisdale
      • 7 mths ago
      • Reported - view

       It's got worse - it's now going to zetta-per! What's going on?

      https://nextdns.io/diag/14044dd0-00e2-11ef-bf0b-916dc33e67bd

      • Will_Tisdale
      • 7 mths ago
      • Reported - view

       My Oracle cloud instance is showing similar too (zetta-per for 2a07:a8c1::), but the diag won't complete - the diag tool doesn't seem to be able to deal with IPv6 traceroute on linux timing out for some reason, but here's an mtr from that box:

      cloud (2603:c020:c003:e9ff:8eb8:9447:4bd:9efb) -> 2a07:a8c1:: (2a07:a8c1::)                                                  2024-04-22T20:58:49+0100
      Keys:  Help   Display mode   Restart statistics   Order of fields   quit
                                                                                                                   Packets               Pings
       Host                                                                                                      Loss%   Snt   Last   Avg  Best  Wrst StDev
       1. 2603:c000:300::8c5b:c864                                                                                0.0%    11    0.2   0.2   0.2   0.3   0.0
       2. 2001:7f8:be::3:1898:1                                                                                   0.0%    11    0.6   0.7   0.5   1.2   0.2
       3. 2001:7f8:be::6939:1                                                                                     0.0%    11    1.8   2.2   1.8   3.2   0.5
       4. (waiting for reply)
       5. (waiting for reply)
       6. (waiting for reply)
       7. (waiting for reply)
       8. (waiting for reply)
       9. 2001:470:0:533::2                                                                                      20.0%    10  126.9 126.8 126.5 126.9   0.1
      10. 2001:470:0:553::2                                                                                       0.0%    10  253.1 253.1 252.8 253.3   0.2
      11. 2001:470:0:59a::2                                                                                       0.0%    10  254.9 255.2 254.9 256.5   0.5
      12. (waiting for reply)
      13. 2001:470:0:62b::2                                                                                       0.0%    10  286.7 288.2 286.5 301.6   4.7
      14. 2001:dea:0:30::a                                                                                        0.0%    10  295.6 294.1 286.1 354.8  21.5
      15. 2403:3400:7604:861::34                                                                                  0.0%    10  285.5 285.7 285.5 285.8   0.1
      16. 2403:3400:7604:950:952:11:0:126                                                                         0.0%    10  337.8 328.4 321.5 337.8   6.7
      17. 2a07:a8c1::  reply)                                                                                     0.0%    10  319.0 318.9 318.9 319.0   0.0
      
      • NextDNs
      • 7 mths ago
      • Reported - view

       anycast is hard :) What about now?

      • Will_Tisdale
      • 7 mths ago
      • Reported - view

       Fixed - thank you!

      It's back routed to anexia-lon where it was before it went wrong: https://nextdns.io/diag/be1502d0-00ea-11ef-b373-451598c64e09

      2a07:a8c1:: is also being routed to anexia-lon from my Oracle cloud instance too, so the secondary is fixed.

      However, there is a new issue that I've noticed with 2a07:a8c0:: on Oracle cloud - it's now being routed to zepto-sin - will try to get a diag but don't think it will complete as it's already hanging after hop 3 of the traceroute6 but here's an mtr:

      cloud (2603:c020:c003:e9ff:8eb8:9447:4bd:9efb) -> 2a07:a8c0:: (2a07:a8c0::)                                                  2024-04-22T22:20:16+0100
      Keys:  Help   Display mode   Restart statistics   Order of fields   quit
                                                                                                                   Packets               Pings
       Host                                                                                                      Loss%   Snt   Last   Avg  Best  Wrst StDev
       1. 2603:c000:300::8c5b:c86c                                                                                0.0%    12    0.3   0.3   0.2   0.5   0.1
       2. 2001:7f8:be::3:1898:1                                                                                   0.0%    12    0.6   0.7   0.6   1.0   0.1
       3. 2001:7f8:be::6939:1                                                                                     0.0%    12    2.0   2.9   1.9  11.4   2.7
       4. (waiting for reply)
       5. (waiting for reply)
       6. (waiting for reply)
       7. (waiting for reply)
       8. (waiting for reply)
       9. (waiting for reply)
      10. (waiting for reply)
      11. (waiting for reply)
      12. (waiting for reply)
      13. 2403:e800:508:300::5a                                                                                   0.0%    11  312.0 330.2 306.7 346.2  15.6
      14. 2401:3cc0:600::f                                                                                       27.3%    11  302.7 321.7 302.7 342.2  16.0
      15. 2a0b:4342::19                                                                                           0.0%    11  300.1 325.2 300.1 340.5  17.1
      16. 2407:b9c0:e000::5                                                                                      10.0%    11  346.5 328.2 304.1 346.5  17.0
      17. 2a07:a8c0::  reply)                                                                                    10.0%    11  345.5 328.2 309.9 345.5  16.5
      
    • entrepreneur, academic educator
    • Erwin
    • 7 mths ago
    • Reported - view

    Yeah @nextdns I have been routed to Sydney from Melbourne for a while as well...
    https://nextdns.io/diag/b66ed2a0-004e-11ef-8cec-294e071f5b41

      • Will_Tisdale
      • 7 mths ago
      • Reported - view

       that is clearly a different issue which is completely unrelated to my issue. You should have started your own thread.

Content aside

  • 7 mths agoLast active
  • 16Replies
  • 139Views
  • 4 Following