[tor-bugs] #13718 [Tor]: Reachability Tests aren't conducted if there are no exit nodes

Tor Bug Tracker & Wiki blackhole at torproject.org
Fri Nov 21 14:55:16 UTC 2014


#13718: Reachability Tests aren't conducted if there are no exit nodes
------------------------+--------------------------------------------------
     Reporter:  tom     |      Owner:  teor
         Type:  defect  |     Status:  assigned
     Priority:  normal  |  Milestone:  Tor: 0.2.???
    Component:  Tor     |    Version:  Tor: 0.2.6.1-alpha
   Resolution:          |   Keywords:  tor-relay test-network lorax chutney
Actual Points:          |  Parent ID:
       Points:          |
------------------------+--------------------------------------------------

Comment (by nickm):

 Replying to [comment:10 teor]:
 > Happy to make these changes, Nick.
 >
 > I've now seen the statuses pop up when launching TorBrowser using this
 build, so I understand the need to comprehensibility.
 >
 > > I kinda want an enum for the argument to
 router_have_minimum_dir_info(), rather than a boolean. It seems like it
 would be clearer that way. Or possibly, there should be two wrappers
 around it: have_minimum_dir_info_for_exit_circ(),
 > have_minimum_dir_info_for_internal_circ().
 >
 > Is there the possibility of needing to calculate weights for guard,
 middle, and exit nodes in arbitrary combinations? (i.e. before choosing a
 guard node, ensure minimum guard bandwidth) If so, we could use a set of
 bit-shift flags.

 I don't think so.  It would be likelier to have to calculate weights for
 different kinds of circuits, I imagine.

 > If not, I'm happy to set up an enum with the two current values of Exit
 and Internal, and possibly an aliased value for those circumstances where
 we want a default option.

 Sounds good.

 > We may also need to update the status/enough-dir-info GETINFO control
 event  - should we add status/enough-dir-info/exit and status/enough-dir-
 info/internal (we default status/enough-dir-info to exit for backwards
 compatibility).

 Sounds fine, though it could be a separate ticket.

 >I also wonder about the impact of changing the invocation of
 circuit_build_needed_circs() so that it runs when we know we have internal
 circuits, rather than waiting for exit circuits.
 >Should we split it into internal and exit versions? If so, which types of
 circuits go in each category?

 That's an interesting question, but it sounds like a separate ticket.
 Generally, anything that is a predicted circuit, or anything that might
 carry user traffic, is an exit circuit.  Anything else is an internal
 circuit.

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/13718#comment:12>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online


More information about the tor-bugs mailing list