[tor-bugs] #20348 [Metrics/Censorship analysis]: Kazakhstan blocking of vanilla Tor and obfs4, 2016-06

[Tor Bug Tracker & Wiki]

#20348: Kazakhstan blocking of vanilla Tor and obfs4, 2016-06
-----------------------------------------+--------------------------
 Reporter:  dcf                          |          Owner:
     Type:  project                      |         Status:  reopened
 Priority:  Medium                       |      Milestone:
Component:  Metrics/Censorship analysis  |        Version:
 Severity:  Normal                       |     Resolution:
 Keywords:  censorship block kz          |  Actual Points:
Parent ID:                               |         Points:
 Reviewer:                               |        Sponsor:
-----------------------------------------+--------------------------

Comment (by dcf):

 Replying to [comment:35 cypherpunks]:
 > > It sounds like kzblocked has actually gotten it to happen, maybe on a
 slow link.
 >
 > This is something about conn.Conn.Write(frameBuf.Bytes()), if frameBuf
 is too long (say 4-7 frames) then everything broken. Can't reproduce short
 write exactly, as log and tcpdump show everything goes to wire, but server
 dislikes result anyway. If to conn.Conn.Write every frame separately then
 server dislike data anyway, but if to delay every conn.Conn.Write then
 everything works ok. I'm lost

 It might be a race condition... Can you try compiling the obfs4proxy code
 with the race detector?
   https://blog.golang.org/race-detector
   https://golang.org/doc/articles/race_detector.html
 I think you can just do `go build -race` in the obfs4proxy directory. Then
 export the environment variable
 `GORACE='log_path=/tmp/obfs4proxyrace.txt'` before you run anything.

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/20348#comment:37>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online