[tor-bugs] #19989 [Core Tor/Tor]: Tor fails to bootstrap with an Exit as EntryNode

Tor Bug Tracker & Wiki blackhole at torproject.org
Mon Sep 5 07:39:31 UTC 2016


#19989: Tor fails to bootstrap with an Exit as EntryNode
--------------------------+------------------------------
 Reporter:  teor          |          Owner:
     Type:  defect        |         Status:  new
 Priority:  Medium        |      Milestone:  Tor: 0.2.???
Component:  Core Tor/Tor  |        Version:
 Severity:  Normal        |     Resolution:
 Keywords:  fallback      |  Actual Points:
Parent ID:                |         Points:  1
 Reviewer:                |        Sponsor:
--------------------------+------------------------------

Comment (by teor):

 Here is the debug output from a similar issue, where Tor refuses to choose
 a single middle node, even though it fits all the criteria:

 {{{
 Sep 05 17:30:04.000 [debug] router_choose_random_node(): We found 0
 running nodes.
 Sep 05 17:30:04.000 [debug] router_choose_random_node(): We removed 0
 excludednodes, leaving 0 nodes.
 Sep 05 17:30:04.000 [debug] router_choose_random_node(): We removed 3
 excludedsmartlist, leaving 0 nodes.
 Sep 05 17:30:04.000 [info] compute_weighted_bandwidths(): Empty routerlist
 passed in to consensus weight node selection for rule weight as middle
 node
 Sep 05 17:30:04.000 [info] router_choose_random_node(): We couldn't find
 any live, stable routers; falling back to list of all routers.
 Sep 05 17:30:04.000 [debug] router_choose_random_node(): We found 1
 running nodes.
 Sep 05 17:30:04.000 [debug] router_choose_random_node(): We removed 0
 excludednodes, leaving 1 nodes.
 Sep 05 17:30:04.000 [debug] router_choose_random_node(): We removed 3
 excludedsmartlist, leaving 1 nodes.
 Sep 05 17:30:04.000 [debug] compute_weighted_bandwidths(): Got negative
 bandwidth weights. Defaulting to naive selection algorithm.
 Sep 05 17:30:04.000 [debug] compute_weighted_bandwidths(): Generated
 weighted bandwidths for rule weight as middle node based on weights
 Wg=1.000000 Wm=1.000000 We=1.000000 Wd=1.000000 with total bw 0
 Sep 05 17:30:04.000 [info] rend_consider_services_intro_points(): Unable
 to find an intro point that we can connect to directly for
 4x463d2itvwimvzw, falling back to a 3-hop path.
 }}}

 But when there are two nodes, Tor happily chooses exactly the same node as
 the single node above (I know it's the single node, because it's the only
 one set as reachable directly in this network):
 {{{
 Sep 05 17:30:04.000 [debug] router_choose_random_node(): We found 3
 running nodes.
 Sep 05 17:30:04.000 [debug] router_choose_random_node(): We removed 0
 excludednodes, leaving 3 nodes.
 Sep 05 17:30:04.000 [debug] router_choose_random_node(): We removed 3
 excludedsmartlist, leaving 0 nodes.
 Sep 05 17:30:04.000 [info] compute_weighted_bandwidths(): Empty routerlist
 passed in to consensus weight node selection for rule weight as middle
 node
 Sep 05 17:30:04.000 [info] router_choose_random_node(): We couldn't find
 any live, stable routers; falling back to list of all routers.
 Sep 05 17:30:04.000 [debug] router_choose_random_node(): We found 5
 running nodes.
 Sep 05 17:30:04.000 [debug] router_choose_random_node(): We removed 0
 excludednodes, leaving 5 nodes.
 Sep 05 17:30:04.000 [debug] router_choose_random_node(): We removed 3
 excludedsmartlist, leaving 2 nodes.
 Sep 05 17:30:04.000 [debug] compute_weighted_bandwidths(): Got negative
 bandwidth weights. Defaulting to naive selection algorithm.
 Sep 05 17:30:04.000 [debug] compute_weighted_bandwidths(): Generated
 weighted bandwidths for rule weight as middle node based on weights
 Wg=1.000000 Wm=1.000000 We=1.000000 Wd=1.000000 with total bw 0
 Sep 05 17:30:04.000 [debug] extend_info_from_node(): using 127.0.0.1:5000
 for test000a
 }}}

 This debug log should allow us to identify where single nodes are failing
 to be selected, at least in the intro case.

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/19989#comment:5>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online


More information about the tor-bugs mailing list