[tor-bugs] #28780 [Core Tor/Tor]: circpadding: Add machine flag for not closing circuit if machine is active

[Tor Bug Tracker & Wiki]

#28780: circpadding: Add machine flag for not closing circuit if machine is active
-------------------------------------------------+-------------------------
 Reporter:  asn                                  |          Owner:  (none)
     Type:  defect                               |         Status:
                                                 |  needs_revision
 Priority:  Very High                            |      Milestone:  Tor:
                                                 |  unspecified
Component:  Core Tor/Tor                         |        Version:
 Severity:  Normal                               |     Resolution:
 Keywords:  wtf-pad, tor-relay, tor-cell,        |  Actual Points:
  padding, 041-proposed, network-team-           |
  roadmap-2019-Q1Q2                              |
Parent ID:  #28634                               |         Points:  5
 Reviewer:                                       |        Sponsor:
                                                 |  Sponsor2
-------------------------------------------------+-------------------------

Comment (by asn):

 Hmm, I can't get this to work with #28634. I did a quick rebase here:
 https://github.com/torproject/tor/pull/927

 I also added this log to signify when we reject to close a circuit:
 {{{
    if (!circpad_circuit_should_be_marked_for_close(circ, reason)) {
 +    log_warn(LD_GENERAL, "Don't close this yet!!!");
     return;
    }
 }}}

 What I see happening is that the client will send an `INTRODUCE_ACK`, then
 the above log will trigger and the circuit will stay open, but no more
 padding will be sent whatsoever. BTW, due to the purposes approach I also
 had to add the CIRCUIT_PURPOSE_C_CIRCUIT_PADDING purpose to the intro
 machine otherwise the machine would be shutdown when the purpose change
 happened.

 You can test this in chutney, and if you want to test it on the real net
 you can use this `MiddleNodes 56927E61B51E6F363FB55498150A6DDFCF7077F2`
 which is running an old version of the #28634 branch (commit `15e0f36`).
 Testing this on chutney will illustrate the problem, but note that the
 histogram latencies are too slow for localhost so you won't see much
 padding.

 Finally, I also get `pathbias_should_count(): Bug: Circuit 3 is now being
 counted despite being ignored in the past. Purpose is Circuit kept open
 for padding, path state is already counted (on Tor 0.4.1.0-alpha-dev
 2a0158f4a3873845)`.

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/28780#comment:21>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online