[tor-bugs] #3443 [Tor Client]: Client with low CBT can't switch to a bridge if it's slow

[Tor Bug Tracker & Wiki]

#3443: Client with low CBT can't switch to a bridge if it's slow
------------------------+---------------------------------------------------
 Reporter:  arma        |          Owner:                    
     Type:  defect      |         Status:  new               
 Priority:  major       |      Milestone:  Tor: 0.2.3.x-final
Component:  Tor Client  |        Version:                    
 Keywords:  regression  |         Parent:                    
   Points:              |   Actualpoints:                    
------------------------+---------------------------------------------------

Comment(by arma):

 In the situations we've seen so far, especially in the bridge case, we do
 not reach 20 circuits (or even 18) before going dead.

 In the extreme case of one bridge, we make one circuit, fail it, and fall
 dead.

 I'm increasingly thinking we should turn cbt learning off by default: it
 produces a slow trickle of users in #tor who have been bitten by the bug
 (their Tor was fast for a while, then suddenly it won't work and the only
 fix is flushing their state file). The other option is tuning it better,
 but to do that correctly we need to run it in all possible environments,
 and it's increasingly becoming clear that we have only looked at a tiny
 subset of the environments people have.

-- 
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/3443#comment:6>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online