[tor-bugs] #1138 [Tor Client]: If bridge authority is unreachable, client doesn't fallback to bridge

Tor Bug Tracker & Wiki torproject-admin at torproject.org
Wed Sep 1 21:49:52 UTC 2010


#1138: If bridge authority is unreachable, client doesn't fallback to bridge
-------------------------+--------------------------------------------------
  Reporter:  arma        |       Owner:  mwenge            
      Type:  defect      |      Status:  needs_review      
  Priority:  major       |   Milestone:  Tor: 0.2.2.x-final
 Component:  Tor Client  |     Version:  0.2.1.19          
Resolution:  None        |    Keywords:  easy              
    Parent:              |  
-------------------------+--------------------------------------------------

Comment(by mwenge):

 Sunday 29 August 2010] [16:22:17] Quit j27Z2 has left this server (Ping
 timeout: 480 seconds).
 [Sunday 29 August 2010] [16:33:10] Join j27Z2 has joined this channel
 (~x at 7R1AACWB7.tor-irc.dnsbl.oftc.net).
 [Sunday 29 August 2010] [16:35:32] <Sebastian> mwenge: reviewing your
 #1138 patch
 [Sunday 29 August 2010] [16:35:48] <mwenge> thanks
 [Sunday 29 August 2010] [16:36:20] <Sebastian> You're lacking
 documentation for was_ei in connection_dir_retry_bridges()
 [Sunday 29 August 2010] [16:36:42] <Sebastian> Also it seems better to not
 have that parameter at all
 [Sunday 29 August 2010] [16:36:57] <Sebastian> and rather still check the
 assert at the caller
 [Sunday 29 August 2010] [16:37:32] <Sebastian> Then you also don't need to
 document it :)
 [Sunday 29 August 2010] [16:38:44] <Sebastian> Calling the first argument
 "failed" seems weird. Maybe just call it descs
 [Sunday 29 August 2010] [16:39:22] Join alkovic has joined this channel
 (~Undefined at 7GDAAEFFT.tor-irc.dnsbl.oftc.net).
 [Sunday 29 August 2010] [16:40:33] <mwenge> ok
 [Sunday 29 August 2010] [16:40:52] <Sebastian> (Please don't just update
 the branch, but rather add a new patch to change these things. Maybe nickm
 started reviewing it too)
 [Sunday 29 August 2010] [16:40:58] <Sebastian> (We can still squash before
 merge)
 [Sunday 29 August 2010] [16:41:07] <mwenge> ah ok
 [Sunday 29 August 2010] [16:41:56] <Sebastian> in
 connection_dir_bridge_routerdesc_failed(), do we want to
 tor_assert(conn->requested_resource) ?
 [Sunday 29 August 2010] [16:42:48] <Sebastian> If we can't do that then we
 have a bug
 [Sunday 29 August 2010] [16:44:13] <Sebastian> What do you think?
 [Sunday 29 August 2010] [16:46:52] <Sebastian> next is that you're saying
 "conn->requested_resource + 3", please add a comment to say "skip over
 fp/" or just do + strlen("fp/")
 [Sunday 29 August 2010] [16:49:32] <Sebastian> and then finally I think
 we're using names like connection_dir_bridge_routerdesc_failed() to ask
 questions like "did this fail". Maybe we find a better name, maybe
 something like connection_dir_bridge_routerdesc_handle_failure
 [Sunday 29 August 2010] [16:49:34] <Sebastian> or such
 [Sunday 29 August 2010] [16:50:00] <Sebastian> Ok please pick from this
 list what you agree with :)
 [Sunday 29 August 2010] [16:50:45] <mwenge> Sebastian: thanks for all
 that, will work through them and respond on trac. i think i only need to
 think about the was_ei one.
 [Sunday 29 August 2010] [16:51:13] <Sebastian> Should I add everything to
 trac?
 [Sunday 29 August 2010] [16:51:26] <Sebastian> Happy to do that, though I
 like discussing feedback before adding it.
 [Sunday 29 August 2010] [16:52:03] <mwenge> no, it's ok i'll add it there.
 i agree with everything.

 I ended up disagreeing with the rename-functions suggestion. But the rest
 of Sebastian's suggestions have been added to my bug1138 branch.

-- 
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/1138#comment:11>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online


More information about the tor-bugs mailing list