15:58:54 #startmeeting tor anti-censorship meeting 15:58:54 Meeting started Thu Jan 25 15:58:54 2024 UTC. The chair is onyinyang. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:58:54 Useful Commands: #action #agreed #help #info #idea #link #topic. 15:58:54 hello everyone! 16:00:25 * onyinyang 's meeting pad link is out of date, one second 16:01:08 here is our meeting pad: https://pad.riseup.net/p/r.9574e996bb9c0266213d38b91b56c469 16:01:13 hi~ 16:01:14 hi~ 16:01:16 hi 16:06:27 Ok we don't have any announcements this week but we have a couple of discussion points and an interesting link 16:07:33 Actually, it seems that most of the discussion points are from last week 16:07:48 I didn't see any new discussion points 16:07:54 yes.... 16:08:02 if there are no objections, i'll try to deploy the SQS rendezvous later today 16:08:09 no objection from me 16:08:19 the interesting link is new 16:09:07 ok cool, if there's nothing else to discuss, then let's move on to the interesting link 16:09:32 https://github.com/net4people/bbs/issues/325 "Default Snowflake bridges in Tor browser 13.0.8 stopped working properly in China since around 2024-01-12" 16:10:11 the bridge status also shows issue with snowflake 16:11:02 it hasn't shown up in the aggregate graph https://people.torproject.org/~dcf/metrics-country.html?start=2023-11-01&end=2024-01-25&country=cn 16:11:13 shelikhoo: what do you mean the bridge status? 16:13:41 https://gitlab.torproject.org/tpo/anti-censorship/connectivity-measurement/bridgestatus/-/blob/main/recentResult_cnnext?ref_type=heads 16:13:44 this 16:14:04 it also includes packet capture 16:14:59 permalink for the meeting log https://gitlab.torproject.org/tpo/anti-censorship/connectivity-measurement/bridgestatus/-/blob/3da850b7399a735f52daa8059694bbd9947594b1/recentResult_cnnext 16:15:22 You are referring to all the snowflake "10" results at the bottom? 16:15:30 https://gitlab.torproject.org/tpo/anti-censorship/connectivity-measurement/bridgestatus/-/blob/3da850b7399a735f52daa8059694bbd9947594b1/recentResult_cnnext#L86-125 16:15:53 snowflake-probe-snowflake1-0-tor 16:16:12 line 86 to line 105 16:16:25 snowflake over amp is known to be not working in China 16:17:39 do you know, was there a change observable on 2024-01-12 to match the timing of the report? 16:20:45 sorry, gitlab was too slow for me to check it in real time... 16:21:41 https://gitlab.torproject.org/tpo/anti-censorship/connectivity-measurement/bridgestatus/-/blob/1b857f0389539e38974c76a68e633d127c7fd188/recentResult_cnnext 16:22:12 In Dec last year, there was still instance where things can bootstrap successfully 16:23:30 This is 2024-01-10 https://gitlab.torproject.org/tpo/anti-censorship/connectivity-measurement/bridgestatus/-/blob/921f6cd777f68d551d06d7f78d075b34392cb154/recentResult_cnnext#L85-104 16:23:41 Still mostly 10, but a few 100 16:24:02 ok thanks for that information 16:24:20 do you want me to share the user logs and pcaps with you? I haven't been able to look at them yet. 16:24:26 shelikhoo is having network problems apparently 16:24:33 yes... sorry... 16:24:41 oh yay you're back 16:25:05 yes.., unsure why there was a few minute things don't work 16:25:27 I have checked the packet capture from vantage point 16:26:07 if these are tor logs, the SnowflakeEvents might help us quickly see which part is failing 16:26:29 there's a tor log and a snowflake-client log 16:26:29 it seems there was a combination of full block of proxy(one way stun) 16:26:36 https://mastodon.social/@tor4zh/111811643889038641 16:26:56 and block of server hello packet(keep receiving client hello) 16:27:13 we didn't have any uncommon amount of tickets from CN 16:27:27 dcf1: yes, I am interested to see the logs and packet capture 16:27:43 in the snowflake-client log I see rendezvous happeneing, stun nat check happening, DataChannel.OpOpen, a small number of DTLS bytes being trasferred, then disconnection and a new rendezvous 16:28:25 shelikhoo: for logcollector output: it will show the bootstrap status after how many minutes? 16:28:40 hmm weird 16:28:52 or asking in another way: how long it takes the test? 16:29:10 https://share.riseup.net/#sUbkhMtY_RayUx70FnImaQ 16:29:59 3 min for snowflake before timeout 16:30:07 1 min for obfs4 16:32:06 I wasn't able to confirm whether it is just packet loss or censorship from the packet capture 16:32:32 but from the screenshot, something wired is going on 16:32:42 i'll send you the logs 16:33:15 yes, thanks! I will also send the log from logcollector to you and cohosh 16:34:23 (maybe I can send to onyinyang as well if you are interested...) 16:35:02 sure 16:36:48 anything else on this or any other topic? 16:37:15 nothing from me, I will have a closer look on this next week 16:40:13 ok, I will end the meeting now then :) Thanks everyone! 16:40:24 #endmeeting