[tor-bugs] #6475 [Tor Client]: circuit_send_next_onion_skin(): Bug: Unexpectedly high circuit_successes

Tor Bug Tracker & Wiki torproject-admin at torproject.org
Tue Aug 21 01:31:48 UTC 2012


#6475: circuit_send_next_onion_skin(): Bug: Unexpectedly high circuit_successes
---------------------------+------------------------------------------------
    Reporter:  grarpamp    |       Owner:  mikeperry         
        Type:  defect      |      Status:  reopened          
    Priority:  major       |   Milestone:  Tor: 0.2.3.x-final
   Component:  Tor Client  |     Version:  Tor: 0.2.3.19-rc  
  Resolution:              |    Keywords:  MikePerry201208   
      Parent:              |      Points:  8                 
Actualpoints:  8           |  
---------------------------+------------------------------------------------

Comment(by mikeperry):

 Replying to [comment:24 mikeperry]:
 > Ok, we're hitting the debugging notice lines in the field. We've got two
 sets: One is related to config->UseEntryGuards == 0 (we should just
 disable pathbias accounting for this case), and the other is related to
 hidserv code:
 >
 > Aug 20 19:10:09.000 [notice] pathbias_count_success(): Bug: Opened
 circuit is in strange path state new. Circuit is a Hidden service:
 Connecting to rendezvous point currently open.  [9 similar message(s)
 suppressed in last 600 seconds]

 I think the culprit for this one is rend_service_introduce(). It looks
 like it can launch CIRCUIT_PURPOSE_S_CONNECT_REND circuits and call
 circuit_init_cpath_crypto() without ever going through
 circuit_finish_handshake() to do so. Thus its first hops don't get
 counted. This could very well be the cause of our original warn bug.

-- 
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/6475#comment:28>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online


More information about the tor-bugs mailing list