[tor-bugs] #1826 [Tor Relay]: Sending packets after RELAY_END

[Tor Bug Tracker & Wiki]

#1826: Sending packets after RELAY_END
-----------------------+----------------------------------------------------
 Reporter:  mwenge     |       Owner:                    
     Type:  defect     |      Status:  new               
 Priority:  normal     |   Milestone:  Tor: 0.2.2.x-final
Component:  Tor Relay  |     Version:  Tor: unspecified  
 Keywords:             |      Parent:                    
-----------------------+----------------------------------------------------

Comment(by mwenge):

 I think this might be a useful clue:

 The total number of RELAY_DATA cells received for the three streams
 combinaed, after a RELAY_END for each had been sent, is exactly 1000.

 I believe that's the size of the cell_queue for a circuit - the spec calls
 it a 'window', I assume that's what it's referring to.

 I think this tells us that the exit used up the entire cell_queue and only
 stopped sending when it received no further SEND_MEs for the individual
 streams.

 However, even though the client had sent RELAY_ENDs for each of the
 streams, as it continued to receive cells on the circuit it continued to
 send SENDMEs with a Circuit ID of zero every hundred of the unwanted
 RELAY_DATA cells it received on the circuit. Even though it had no active
 streams on the circuit.

 If my understanding of the spec is correct, had it not send those SENDMEs
 the exit might have stopped pumping the unwanted RELAY_DATAs sooner. The
 client could have observed it had no active streams on the circuit and
 sent no SENDMEs at all.

 I should be able to test this theory here.

-- 
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/1826#comment:5>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online