[tor-relays] non-exit risks?

David Carlson david.carlson.417 at gmail.com
Fri Sep 20 17:13:08 UTC 2013


On 9/20/2013 10:23 AM, Moritz Bartl wrote:
> On 09/20/2013 04:17 PM, That Guy wrote:
>> While I know you have vastly more knowledge and experience than me and
>> often my observations or worries are doe to me mis-understanding
>> something but about the comment above,  I have experienced issues
>> running a non-exit relay pretty soon after it going up and though I have
>> no idea if there is a connection, I started to get more trouble after
>> the doubling of connections at the end of July.
> Yes, unfortunately more and more sites block Tor relay IPs, regardless
> of whether they allow exiting (to that site) or not. All that helps here
> is friendly education. Whenever you notice something like that, contact
> the site owner or blocklist maintainer and teach them about the problem
> and how to properly detect Tor exits, namely
>
> https://check.torproject.org/cgi-bin/TorBulkExitList.py
> https://www.torproject.org/projects/tordnsel.html.en
>
> This is not the kind of "trouble" Gordon M. is referring to. Especially
> exit relays should be run on dedicated IPs, and, if possible, dedicated
> machines. Non-exit relays face no such "trouble", there is no known case
> of law enforcement asking for customer information or abuse complaints
> directed at a non-exit relay.
>
That may be true, but as you point out, there are more and more cases of
non-exit relay operators being denied the opportunity to spend money at
certain websites or to use the services provided by certain websites. 
The corollary is that it is also harder for legitimate Tor browser users
to access those same websites. 

In one case I tried to 'educate' the person in charge of the blacklist
at one large group of websites.  Since he knew that it was a huge risk
to allow any IP address identified as being used by any Tor relay, he
would not consider any change to his policy.  Whatever method he used to
generate his list was certainly the only correct method.  Considering
more sophisticated methods to detect and differentiate legitimate
activity from nefarious activity would be too difficult, i suppose.

Suggesting that he confer with other experts, for example at EFF, was
pointless, because he knew that he was right.  I think that he also knew
that the Earth is flat. 

That's my 2¢ US (btc are too valuable).

David C


More information about the tor-relays mailing list