[tor-bugs] #21621 [Core Tor/Tor]: Intro points can get stuck in CIRCUIT_PURPOSE_S_ESTABLISH_INTRO

[Tor Bug Tracker & Wiki]

#21621: Intro points can get stuck in CIRCUIT_PURPOSE_S_ESTABLISH_INTRO
--------------------------+------------------------------------
 Reporter:  teor          |          Owner:  teor
     Type:  defect        |         Status:  new
 Priority:  Medium        |      Milestone:  Tor: 0.3.1.x-final
Component:  Core Tor/Tor  |        Version:  Tor: 0.2.7.2-alpha
 Severity:  Normal        |     Resolution:
 Keywords:  tor-hs        |  Actual Points:
Parent ID:  #21446        |         Points:  1
 Reviewer:                |        Sponsor:
--------------------------+------------------------------------

Comment (by teor):

 Replying to [comment:3 teor]:
 > Replying to [comment:2 dgoulet]:
 > > I think we do.
 > >
 > > `intro->circuit_established = 1;` is what signals that the intro
 circuit is established and ready to be used?
 >
 > We only do that when the INTRO_ESTABLISHED cell is received.
 > Until then, `intro->circuit_established` is 0, and it can remain in that
 state forever:
 > * What if the introduction point has a bug and never sends it?

 This can't happen, at least in the latest version: the intro point either
 sends INTRO_ESTABLISHED or closes the circuit.

 It's also not possible for an OOM to cause this: on OOM, the circuit is
 closed when the cells are discarded.

 >   * What do introduction points do when they detect a replay?

 There is no replay cache: that's rend points.

 (If the key is the same, the old circuit is replaced with the new circuit,
 and the old circuit is closed.)

 > * What if there's a deliberate attack where an intro point maintains the
 circuit but never sends the INTRO_ESTABLISHED cell?

 Or another bug?

 Then we'd need to fix this issue.

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/21621#comment:4>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online