Anyone else notice slow iCloud photo syncing on 2degrees when IPv6 is enabled?
I did a fresh install of Mojave on my MBP so I'm currently downloading my photos and it's been going for several days and barely made a dent, only 456 photos/videos, out of 2495 so its not a huge library.
Seems to be going at ~3-5Mbit.
Wondering what was going on I had a look at nettop and see the following:
nsurlsessiond.250 143 MiB 51 KiB 0 B 5672 B 1580 B
tcp6 x:cd06:164a:ee5c:5d0e.51830<->th-in-x80.1e100.net.443 en0 Established 6747 KiB 1270 B 0 B 0 B 0 B 148.06 ms 83 KiB 60 KiB BK_SYS -
tcp6 x:cd06:164a:ee5c:5d0e.51831<->2403:300:a41:400::7.443 en0 Established 1557 KiB 2321 B 0 B 0 B 0 B 325.09 ms 64 KiB 34 KiB BK_SYS -
tcp6 x:cd06:164a:ee5c:5d0e.51832<->th-in-x80.1e100.net.443 en0 Established 5628 KiB 1230 B 0 B 0 B 0 B 145.34 ms 162 KiB 60 KiB BK_SYS -
tcp6 x:cd06:164a:ee5c:5d0e.51833<->th-in-x80.1e100.net.443 en0 Established 4826 KiB 1230 B 0 B 0 B 0 B 129.31 ms 51 KiB 60 KiB BK_SYS -
tcp6 x:cd06:164a:ee5c:5d0e.51834<->th-in-x80.1e100.net.443 en0 Established 6995 KiB 1230 B 0 B 0 B 0 B 130.44 ms 194 KiB 60 KiB BK_SYS -
tcp6 x:cd06:164a:ee5c:5d0e.51844<->2403:300:a41:400::7.443 en0 Established 141 B 1969 B 0 B 0 B 0 B 280.75 ms 128 KiB 28 KiB BK_SYS -
Doing a traceroute to2403:300:a41:400::7, it seems to take a scenic route to San Jose (i think) then Japan and I'm not sure what the final destination is but the latency is high.
traceroute6 to 2403:300:a50:105::8 (2403:300:a50:105::8) from x:cd06:164a:ee5c:5d0e, 64 hops max, 12 byte packets
1 ::1 1.012 ms 2.253 ms 1.447 ms
2 2406:e000:2801:c::1 3.210 ms 4.061 ms 2.235 ms
3 as23655.auckland.megaport.com 15.115 ms 16.333 ms 15.505 ms
4 2001:7fa:11:6::1b1b:0:1 15.881 ms 14.635 ms 14.317 ms
5 10ge2-3.core1.syd1.he.net 39.391 ms 39.118 ms 48.607 ms
6 10ge3-5.core1.sjc1.he.net 181.170 ms 189.033 ms 180.653 ms
7 10ge7-2.core1.sjc2.he.net 180.222 ms
100ge1-1.core1.sjc2.he.net 180.409 ms
10ge7-2.core1.sjc2.he.net 181.625 ms
8 kddi-as2516.10gigabitethernet1-1-1.switch2.sjc2.he.net 181.098 ms 181.841 ms 182.752 ms
9 6sjegcs002.int-gw.kddi.ne.jp 162.361 ms
6sjegcs001.int-gw.kddi.ne.jp 162.325 ms
6sjegcs002.int-gw.kddi.ne.jp 163.800 ms
10 oteacs001.int-gw.kddi.ne.jp 264.717 ms
2001:268:fb81:72::1 268.788 ms
2001:268:fb81:6a::1 266.005 ms
11 2001:268:fa00:2fc::2 265.173 ms
2001:268:fa00:2f9::2 269.729 ms 262.656 ms
12 6kotjin204.int-gw.kddi.ne.jp 273.841 ms 268.750 ms
6kotjin204.int-gw.kddi.ne.jp 265.031 ms
13 2001:268:f702:100::2 282.773 ms 288.522 ms 288.598 ms
14 2403:300:0:80f::21 221.936 ms 223.108 ms 217.024 ms
15 2403:300:0:80f::1 220.787 ms 220.023 ms 217.712 ms
16 2403:300:0:810::52 221.658 ms 216.997 ms 222.670 ms
17 2403:300:a09:2::302 217.379 ms 223.525 ms 218.987 ms
18 2403:300:a50:105::8 220.699 ms 220.420 ms 217.889 ms
And the other host, not sure where that ends up but judging by latency US West somewhere?
traceroute to th-in-x80.1e100.net (2404:6800:4008:c03::80), 30 hops max, 80 byte packets
1 x (x) 0.292 ms 0.256 ms 0.278 ms
2 2406:e000:2801:c::1 (2406:e000:2801:c::1) 1.380 ms 1.364 ms 1.344 ms
3 * * *
4 2001:4860:1:1:0:5c67:0:1 (2001:4860:1:1:0:5c67:0:1) 34.778 ms 34.755 ms 34.746 ms
5 2001:4860:1:1:0:5c67:: (2001:4860:1:1:0:5c67::) 34.697 ms 34.715 ms 34.669 ms
6 2001:4860:0:1107::b (2001:4860:0:1107::b) 35.080 ms 39.788 ms 39.736 ms
7 2001:4860::c:4001:11b6 (2001:4860::c:4001:11b6) 142.799 ms 2001:4860::c:4000:f918 (2001:4860::c:4000:f918) 139.317 ms 139.288 ms
8 2001:4860::c:4000:e43d (2001:4860::c:4000:e43d) 162.983 ms 2001:4860::c:4000:e43e (2001:4860::c:4000:e43e) 162.901 ms 2001:4860::8:0:c70b (2001:4860::8:0:c70b) 163.194 ms
9 2001:4860::cc:4001:1e43 (2001:4860::cc:4001:1e43) 162.057 ms 162.065 ms 2001:4860::cc:4001:1c1a (2001:4860::cc:4001:1c1a) 162.785 ms
10 * * *
11 * * *
12 * * *
13 * * *
14 * * *
15 * * *
16 * * *
17 * * *
18 * * *
19 * th-in-x80.1e100.net (2404:6800:4008:c03::80) 162.168 ms 162.891 ms
Disabling IPv6 on the Mac, it's sped up but not by a huge amount ~10-15Mbit but that could be apple throttling.
Nettop:
nsurlsessiond.250 300 MiB 117 KiB 0 B 33 KiB 2106 B
tcp4 192.168.1.120:51892<->tg-in-f128.1e100.net:443 en0 Established 5681 KiB 1228 B 0 B 2836 B 0 B 169.09 ms 124 KiB 61 KiB BK_SYS -
tcp4 192.168.1.120:51893<->tg-in-f128.1e100.net:443 en0 Established 4122 KiB 1226 B 0 B 2836 B 0 B 159.03 ms 126 KiB 61 KiB BK_SYS -
tcp4 192.168.1.120:51894<->tg-in-f128.1e100.net:443 en0 Established 4997 KiB 1242 B 0 B 2836 B 0 B 170.31 ms 105 KiB 61 KiB BK_SYS -
tcp4 192.168.1.120:51895<->tg-in-f128.1e100.net:443 en0 Established 5055 KiB 1244 B 0 B 2836 B 0 B 168.66 ms 110 KiB 61 KiB BK_SYS -
tcp4 192.168.1.120:51896<->tg-in-f128.1e100.net:443 en0 Established 3687 KiB 1242 B 0 B 0 B 0 B 158.97 ms 77 KiB 61 KiB BK_SYS -
tcp4 192.168.1.120:51901<->tg-in-f128.1e100.net:443 en0 Established 1369 KiB 1240 B 0 B 0 B 0 B 161.22 ms 83 KiB 61 KiB BK_SYS -
Traceroute lands in Australia judging by the latency?:
traceroute tg-in-f128.1e100.net
traceroute: Warning: tg-in-f128.1e100.net has multiple addresses; using 216.239.38.128
traceroute to tg-in-f128.1e100.net (216.239.38.128), 64 hops max, 52 byte packets
1 router (192.168.1.1) 3.169 ms 1.248 ms 2.617 ms
2 69.7.69.111.static.snap.net.nz (111.69.7.69) 2.797 ms 2.163 ms 2.221 ms
3 * * *
4 72.14.215.151 (72.14.215.151) 37.438 ms 36.263 ms 35.343 ms
5 72.14.215.150 (72.14.215.150) 36.476 ms 36.423 ms 35.640 ms
6 108.170.247.49 (108.170.247.49) 37.022 ms 37.648 ms 38.965 ms
7 209.85.248.253 (209.85.248.253) 36.216 ms
209.85.254.119 (209.85.254.119) 36.254 ms
209.85.255.165 (209.85.255.165) 36.416 ms
8 any-in-2680.1e100.net (216.239.38.128) 36.358 ms 36.062 ms 36.303 ms
I believe Apple use Google Cloud Compute hence the hosts/IP's belonging to Google.
Any idea why when using IPv6 Apple/Google/someone isn't preferring Australian hosts?
edit: Although having said that I just noticed the latency figures for ipv4 are all over 100 which suggests it's not hitting Australia either :/. Guessing this is just all down to Apple and not much can be done?