[tor-bugs] #7959 [Tor]: onionskin_answer(): Bug: couldn't format created cell

Tor Bug Tracker & Wiki blackhole at torproject.org
Tue Jan 15 17:18:57 UTC 2013


#7959: onionskin_answer(): Bug: couldn't format created cell
-----------------------+----------------------------------------------------
    Reporter:  sysrqb  |       Owner:                    
        Type:  defect  |      Status:  closed            
    Priority:  major   |   Milestone:  Tor: 0.2.4.x-final
   Component:  Tor     |     Version:  Tor: 0.2.4.8-alpha
  Resolution:  fixed   |    Keywords:  tor-relay         
      Parent:          |      Points:                    
Actualpoints:          |  
-----------------------+----------------------------------------------------
Changes (by sysrqb):

  * status:  needs_review => closed
  * resolution:  => fixed


Comment:

 Replying to [comment:9 nickm]:
 >
 > Make sure it's running as a relay, and set UseNTorHandshake 1 ; that'll
 be good for something. In fact, I believe that when it tries to make a
 testing circuit to probe itself, it'll use the NTor handshake to do so --
 probably through a node that doesn't support ntor.  So if it decides that
 its ORPort is reachable, then this bug is fixed.

 Yeah, looking through the logs a little more thoroughly the socket that
 resulted in this was a cpu worker connected to localhost (I'm not too
 familiar with the cpu worker code right now). The relay eventually figured
 out it was reachable after successfully completing appropriate external
 connections.


 > (I've merged the code, since it's *a* correct thing to do, but I'm still
 hoping for more confirmation that it *does* fix this issue.)

 {{{
 Jan 15 09:44:32.000 [debug] conn_write_callback(): socket 16 wants to
 write.
 Jan 15 09:44:32.000 [debug] cpuworker_main(): onion_skin_server_handshake
 succeeded.
 Jan 15 09:44:32.000 [debug] cpuworker_main(): finished writing response.
 Jan 15 09:44:32.000 [debug] conn_read_callback(): socket 16 wants to read.
 Jan 15 09:44:32.000 [debug] read_to_chunk(): Read 620 bytes. 620 on inbuf.
 Jan 15 09:44:32.000 [debug] connection_cpu_process_inbuf(): Unpacking
 cpuworker reply, chan_id is 2, circ_id is 56640
 Jan 15 09:44:32.000 [debug] circuit_get_by_circid_channel_impl():
 circuit_get_by_circid_channel_impl() returning circuit 0x7f7f6c03a8b0 for
 circ_id 56640, channel ID 2 (0x7f7f6c014b90)
 Jan 15 09:44:32.000 [debug] onionskin_answer(): init digest forward
 0x58de764c, backward 0x02b123d9.
 Jan 15 09:44:32.000 [debug] append_cell_to_circuit_queue(): Made a circuit
 active.
 Jan 15 09:44:32.000 [debug] append_cell_to_circuit_queue(): Primed a
 buffer.
 Jan 15 09:44:32.000 [debug] channel_write_packed_cell(): Writing
 packed_cell_t 0x7f7f6c03b8a8 to channel 0x7f7f6c014b90 with global ID 2
 Jan 15 09:44:32.000 [debug] channel_flush_from_first_active_circuit():
 Made a circuit inactive.
 Jan 15 09:44:32.000 [debug] onionskin_answer(): Finished sending 'created'
 cell.
 Jan 15 09:44:32.000 [notice] Self-testing indicates your ORPort is
 reachable from the outside. Excellent. Publishing server descriptor.
 }}}

 and socket 16 is a connection to localhost. :)

 Looks like this one is good. Closing

-- 
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/7959#comment:12>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online


More information about the tor-bugs mailing list