[tor-talk] SKKU43 Tor relay (was Re: Metrics fast exit page)

[Roger Dingledine]

On Mon, Feb 10, 2014 at 06:36:36PM +0900, Hyoung-Kee Choi wrote:
> We are the academic research group located in Seoul, Korea. One of my
> students happened to misconfigure the bandwidth parameter in his Tor during
> his research on the Tor bandwidth scanner. Then, the advertised bandwidth
> jumped up to the high number and drew a number of circuits in the last days
> or so. We switched the number back to the automatic adjustment right after
> checking your email in this list. We are still running SKKU43  in order for
> the advertised bwauth value to return to normal.
> 
> I apologize for this happening to you and the Tor community.

Hi Hyoung-Kee,

Thanks for letting us know. And now we're all curious: what was this
research on the Tor bandwidth scanner?

TorFlow really needs a maintainer currently, and indeed there are
many known attacks to manipulate the results, for example because
it calculates your consensus weight as a function of your current
self-measured bandwidth.

See also
https://lists.torproject.org/pipermail/tor-relays/2014-January/thread.html#3717

Thanks!
--Roger