#25882: clients not detecting stale onion service introduction points --------------------------+------------------------------------ Reporter: cypherpunks | Owner: (none) Type: defect | Status: assigned Priority: Medium | Milestone: Tor: 0.3.4.x-final Component: Core Tor/Tor | Version: Severity: Normal | Resolution: Keywords: tor-hs | Actual Points: Parent ID: #22455 | Points: Reviewer: | Sponsor: --------------------------+------------------------------------ Changes (by arma):
* cc: dgoulet (added) Comment: I think #21600 is not as related as you might think, since I think that ticket is for the onion service side. Theory 1: the intro point went down or was otherwise unreachable, and the Tor client made a circuit to try to reach it, but the circuit timed out, and so we launch a new parallel one every second, rather than once like we had intended. See {{{intro_going_on_but_too_old}}} in circuituse.c. Also, while you're looking at that clause: {{{ /* Log an info message if we're going to launch a new intro circ in * parallel */ if (purpose == CIRCUIT_PURPOSE_C_INTRODUCE_ACK_WAIT && !must_be_open && origin_circ->hs_circ_has_timed_out && !circ->marked_for_close) { intro_going_on_but_too_old = 1; continue; } }}} It isn't until later, in circuit_is_acceptable, which we don't call because of the "continue", that we check that the origin_circ->hs_circ_has_timed_out was for *our* intended intro point, right? What's up with that? I'm cc'ing dgoulet since he touched that code recently. I guess it isn't *so* bad, since that bug only controls whether we log a (potentially confusing) info-level log? -- Ticket URL: <https://trac.torproject.org/projects/tor/ticket/25882#comment:6> Tor Bug Tracker & Wiki <https://trac.torproject.org/> The Tor Project: anonymity online
_______________________________________________ tor-bugs mailing list tor-bugs@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs