[tor-bugs] #12752 [Tor]: stem: Microdescript bandwidth considered unmeasured by default

Tor Bug Tracker & Wiki blackhole at torproject.org
Sun Aug 24 17:40:39 UTC 2014


#12752: stem: Microdescript bandwidth considered unmeasured by default
------------------------+--------------------
     Reporter:  asn     |      Owner:  atagar
         Type:  defect  |     Status:  new
     Priority:  normal  |  Milestone:
    Component:  Tor     |    Version:
   Resolution:          |   Keywords:
Actual Points:          |  Parent ID:
       Points:          |
------------------------+--------------------

Comment (by nickm):

 Replying to [comment:1 atagar]:
 > Hi George, really sorry about the long delay. I'm still catching up on
 tickets. According to [https://gitweb.torproject.org/torspec.git/blob/HEAD
 :/dir-spec.txt#l1651 the dir-spec on this]...
 >
 >  * 'Bandwidth=INT' is 'an estimate of the bandwidth of this relay'
 though it's really nebulous about what that means.

 That's correct.  From a client's POV, the units and accuracy of this value
 are unspecified intentionally, so that we could make some uniform change
 to the units and/or measurement scheme as needed.

 >  * 'Measured=INT' is the vote of a bandwidth authority.
 >  * 'Unmeasured=1' has a specific meaning, which is that the
 'Bandwidth=INT' isn't based on three or more measurements.
 >
 > Stem's docs on this are...
 >
 > {{{
 > :var int bandwidth: bandwidth claimed by the relay (in kb/s)
 > :var int measured: bandwidth measured to be available by the relay
 > :var bool is_unmeasured: bandwidth measurement isn't based on three or
 more
 >   measurements
 > }}}
 >
 > On reading this again I suspect the behavior is as follows...
 >
 >  * Votes from bandwidth authorities have 'Measured=' values.

 Right.

 >  * These votes are transformed into 'Bandwidth=' for the consensus, with
 the 'Unmeasured=1' if it's not coming from 3 or more authorities (though
 the wording on this is vague enough that it could mean 3 or more samplings
 by a single authority).

 Yes.

 > Either way, I'm happy to revise Stem if the dir-spec is clarified.
 Sending this over to the Tor component.

 What revisions to dir-spec do you suggest?

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/12752#comment:2>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online


More information about the tor-bugs mailing list