[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [tor-bugs] #33745 [Circumvention/Snowflake]: Merge a turbotunnel branch
#33745: Merge a turbotunnel branch
-------------------------------------+--------------------------
Reporter: dcf | Owner: dcf
Type: task | Status: assigned
Priority: Medium | Milestone:
Component: Circumvention/Snowflake | Version:
Severity: Normal | Resolution:
Keywords: turbotunnel | Actual Points:
Parent ID: #19001 | Points:
Reviewer: | Sponsor:
-------------------------------------+--------------------------
Comment (by dcf):
Replying to [comment:1 arma]:
> > * Test without `LearnCircuitBuildTimeout 0` and find another
workaround, if necessary. See comment:15:ticket:33336.
>
> I did all my quic tests with those torrc changes commented out (i.e. I
used what normal Tor Browser would have given me).
Thanks, that's good to know. I added the `CircuitBuildTimeout` changes in
an uncontrolled way at the same time as the idle timeout fixes in #33336,
so it's possible they are not necessary. I'll try just removing them and
see how it goes. What I was afraid of is a case where the first time you
connect, you get lucky and get a working proxy on the first try, and tor
learns a `CircuitBuildTimeout` of 60 seconds. Then then next time you
connect, maybe it takes four tries to get a working proxy, but by then tor
has given up according to the shorter timeout it learned. But I don't know
if it really works like that or how eagerly tor adjusts the learned
timeout.
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/33745#comment:3>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
_______________________________________________
tor-bugs mailing list
tor-bugs@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs