[tor-bugs] #25687 [Core Tor/Tor]: over-report of observed / self-measure bandwidth on fast hardware -- important to torflow / peerflow

Tor Bug Tracker & Wiki blackhole at torproject.org
Tue Sep 18 16:30:46 UTC 2018


#25687: over-report of observed / self-measure bandwidth  on fast hardware --
important to torflow / peerflow
--------------------------+----------------------------------
 Reporter:  starlight     |          Owner:  (none)
     Type:  defect        |         Status:  new
 Priority:  Medium        |      Milestone:  Tor: unspecified
Component:  Core Tor/Tor  |        Version:  Tor: 0.2.6.10
 Severity:  Normal        |     Resolution:
 Keywords:  tor-bwauth    |  Actual Points:
Parent ID:                |         Points:
 Reviewer:                |        Sponsor:
--------------------------+----------------------------------

Comment (by juga):

 Replying to [comment:10 starlight]:
 > The concern of this ticket is with regard to the observed value
 published by the daemon rather than the Torflow calculated vote resulting
 from it.

 if by daemon you mean little-t-tor, i know

 >I suspect that changes to measurement logic in the daemon have likely
 reduced or eliminated the error last definitively observed in 2.6.10.  I
 may yet write a monitor to evaluate the quality of daemon measurements.

 if you do, would be awesome you point us to the code and results

 > Torflow adjusts self-report with the ratio

 exactly, and that ratio is a small number (0.8-8 in the samples i took),
 therefore observed / self-measured bandwidth affects a lot on what Torflow
 finally reports.

 > of the measured bandwidth for a relay to the average of all measured
 bandwidths in a manner short on nuance.  Wrote about possible ways to
 address that here:
 >
 > https://github.com/torproject/sbws/issues/182#issuecomment-410745893

 sorry, i found that comment hard to read

 FWIW, we're continuing that ticket in multiple others. Follow the links of
 https://trac.torproject.org/projects/tor/ticket/27107

 > Have considered opening a Trac ticket encompassing the ideas, but am
 waiting for an opportune moment at a point where serious comparison work
 is evident.

 on what you would like to have a serious comparison?

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/25687#comment:11>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online


More information about the tor-bugs mailing list