[tor-bugs] #6341 [Tor Relay]: connection_or_flush_from_first_active_circuit() does wrong thing when ewma_enabled

[Tor Bug Tracker & Wiki]

#6341: connection_or_flush_from_first_active_circuit() does wrong thing when
ewma_enabled
-----------------------+----------------------------------------------------
 Reporter:  arma       |          Owner:                    
     Type:  defect     |         Status:  needs_review      
 Priority:  normal     |      Milestone:  Tor: 0.2.3.x-final
Component:  Tor Relay  |        Version:                    
 Keywords:             |         Parent:                    
   Points:             |   Actualpoints:                    
-----------------------+----------------------------------------------------

Comment(by robgjansen):

 Replying to [comment:48 arma]:
 > If they work, by all means keep them for now.

 It turns out it made things worse.

 I ran 24 experiments with "FetchDirInfoEarly 1", and 24 without. Then I
 graphed the time that each node in each experiment reached 100% bootstrap
 status as a CDF. The
 [https://trac.torproject.org/projects/tor/attachment/ticket/6341/time-to-
 bootstrap.pdf results are here]. It seems to have not affected relay
 bootstrapping times, and made it worse for clients. (Note that clients in
 my experiments don't begin until around 15 minutes = 900 seconds, hence
 the large jump in the client plots.)

 > As another fun example of bizarre behavior, [...]

 I guess we won't have to worry about that. So it seems that, after all of
 this, the fix isn't in the config and we don't have any answers except to
 lower the retry schedule (#6752)? Do you expect the fixes in 0.2.3.21-rc
 to improve bootstrapping?

-- 
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/6341#comment:49>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online