[tor-bugs] #30733 [Core Tor/sbws]: sbws does not detect changes in descriptor bandwidth values

Tor Bug Tracker & Wiki blackhole at torproject.org
Wed Feb 5 10:25:17 UTC 2020


#30733: sbws does not detect changes in descriptor bandwidth values
-------------------------------------------------+-------------------------
 Reporter:  starlight                            |          Owner:  juga
     Type:  defect                               |         Status:
                                                 |  merge_ready
 Priority:  Very High                            |      Milestone:  sbws:
                                                 |  1.1.x-final
Component:  Core Tor/sbws                        |        Version:  sbws:
                                                 |  1.1.0
 Severity:  Critical                             |     Resolution:
 Keywords:  sbws-majority-blocker, sbws-roadmap  |  Actual Points:
Parent ID:  #33121                               |         Points:  1
 Reviewer:  asn                                  |        Sponsor:
-------------------------------------------------+-------------------------

Comment (by juga):

 Replying to [comment:41 asn]:
 > It would be nice if the new code was also tested.

 agree. It's hard to test right now because:
 - the test network is tear down after 1st consensus and download. If we'd
 use chutney, then i think we could check for changes in descriptors
 (#33150)
 - there could be unit tests where we initialize RelayList with different
 consensus/descriptors, but that won't be possible until #29717
 - we could check that the production bandwidth files detect changes in the
 descriptors via an external tool (#33152)

 Maybe i should create a child ticket for #33149 to test this so we don't
 forget?

 > Also, you might want to check dormant mode and see if you need to
 disable that too, since if your tor thinks you are not using it, it will
 go dormant in 24h and stop downloading consensuses etc. The right way to
 disable dormant mode is to send the `ACTIVE` signal with the controller
 every once in a while. I'm not sure if this applies in your case, but I
 noticed the new torrc options you added are relevant.

 Questions about the dormant mode:
 - could tor become dormant when it is constantly creating new circuits?.
 - would be enough to set the option `DormantCanceledByStartup 1` to ensure
 that tor won't ever become dormant?
 - in which tor version was introduced the dormant mode?, 0.4.0?
 Maybe i could create other ticket to test this once we also run tests with
 different tor versions (#30067)?

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/30733#comment:42>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online


More information about the tor-bugs mailing list