[tor-bugs] #5156 [Obfsproxy]: assertion failure at src/network.c:931: circ

[Tor Bug Tracker & Wiki]

#5156: assertion failure at src/network.c:931: circ
-----------------------+----------------------------------------------------
 Reporter:  arma       |          Owner:  asn
     Type:  defect     |         Status:  new
 Priority:  normal     |      Milestone:     
Component:  Obfsproxy  |        Version:     
 Keywords:             |         Parent:     
   Points:             |   Actualpoints:     
-----------------------+----------------------------------------------------

Comment(by nickm):

 Replying to [comment:4 asn]:
 > Replying to [comment:3 nickm]:
 > > Ah, it looks like there's an ordering bug here.
 bufferevent_connect_hostname is allowed to complete immediately while it
 is called, but the function calling it assumes that it's okay to set up
 the ->circuit pointer after the open_outbound_hostname call returns.
 > >
 > > So the right fix here seems to be setting the circuit earlier, I
 guess?
 >
 > So I see, I didn't know that bufferevent callbacks are triggered
 immediately. I thought they were triggered on the next libevent loop;
 seems like this is the behavior of deferred callbacks, though.

 Most callbacks are; some aren't.  Connect/resolve callbacks, especially
 failing ones, can happen immediately.  (Might be nice to get this fixed in
 libevent 2.1)

 > So, does `simple_server_listener_cb()` and `simple_client_listener_cb()`
 have the same problem? `open_outbound()` seems to do `bufferevent_setcb`
 and `bufferevent_socket_connect` before setting up the circuit.

 It might on some BSDs.  The resolve in bufferevent_connect_hostname seems
 to be the main issue with this case, but ISTR that on some BSDs, connect()
 can succeed/fail immediately.

-- 
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/5156#comment:5>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online