Hello,
As we can see on
http://www.tunnelbroker.net/usage/tunnels_by_latency.php, only about 19 tunnels on the Tokyo tunnel server have ping less than 100 msec (the 20th tunnel in the top list has 108ms).
The traceroutes look like this:
Host Loss% Snt Last Avg Best Wrst StDev
1. cloud1.japan.so.cm 0.0% 11 0.0 0.1 0.0 0.1 0.0
2. ge3-42.ar1.tko1.jp.scnet.net 0.0% 10 11.1 6.8 0.4 53.9 16.9
3. xe-0-0-0.cr1.tko1.jp.scnet.net 0.0% 10 0.3 2.0 0.3 17.0 5.3
4. ???
5. tserv1.tyo1.he.net 0.0% 10 120.1 120.2 120.1 120.2 0.0
core1.tyo1.he.net> traceroute 50.31.254.105 source-ip 27.50.33.5 numeric
1 126 ms 126 ms 126 ms 10gigabitethernet3-3.core1.sjc2.he.net (72.52.92.237)
2 134 ms 134 ms 127 ms 10gigabitethernet3-2.core1.pao1.he.net (72.52.92.69)
3 127 ms 127 ms 127 ms softbanktelecom.10gigabitethernet2-2.core1.pao1.he.net (216.218.244.234)
4 233 ms 233 ms 233 ms STOrw-51-tengiga7-4.gw.odn.ad.jp (210.142.163.169)
5 233 ms 233 ms 233 ms STOrk-07G4-0-0.nw.odn.ad.jp (143.90.33.117)
6 277 ms 401 ms 345 ms STOrw-12T2-1.nw.odn.ad.jp (143.90.192.134)
7 246 ms 240 ms 237 ms 26.143090232.odn.ne.jp (143.90.232.26)
8 118 ms 124 ms 128 ms xe-0-0-1.gw402.ty2.ap.equinix.com (183.177.32.153)
9 122 ms 125 ms 125 ms xe-0-1-1.gw401.ty2.ap.equinix.com (183.177.32.140)
10 125 ms 125 ms 125 ms xe-0-0-0.gw401.ty1.ap.equinix.com (202.177.203.49)
11 127 ms 125 ms 125 ms e-1-3.br401.ty1.ap.equinix.com (183.177.32.158)
12 120 ms 125 ms 125 ms 202.177.202.226
13 125 ms 128 ms 122 ms tge2-3.ar1.tko1.jp.scnet.net (75.102.60.153)
14 125 ms 125 ms 117 ms as32780.ge3-42.ar1.tko1.jp.scnet.net (75.102.60.22)
15 117 ms 117 ms 117 ms cloud1.japan.so.cm (50.31.254.105)
So it first goes to San Jose in the US for the outbond transfer from core1.tyo1.he.net, and presumably from the tunnel server too. It's not just to my particular network, according to
http://tunnelbroker.net/usage/tunnels_by_tserv.php there are 1490 tunnels on the Tokyo tunnel server, and judging from the top latency list, 1471 of them are affected by this issue and have ping higher than 100 msec to the tunnel server. Can you please look into fixing this situation?
Thanks