[tor-bugs] #1751 [Tor Relay]: Project: Make it harder to use exits as one-hop proxies

[Tor Bug Tracker & Wiki]

#1751: Project: Make it harder to use exits as one-hop proxies
-----------------------+----------------------------------------------------
 Reporter:  nickm      |       Owner:                     
     Type:  task       |      Status:  needs_review       
 Priority:  normal     |   Milestone:  Deliverable-Sep2010
Component:  Tor Relay  |     Version:                     
 Keywords:             |      Parent:                     
-----------------------+----------------------------------------------------

Comment(by arma):

 Replying to [comment:17 nickm]:
 > Adding a reason for this seems like a fine thing to me.

 Yet another case where we wish we'd sorted out and implemented that
 proposal to reserve half the reason space as retriable reason codes.

 Oh hey, I just checked the code again, and apparently I already chose
 END_STREAM_REASON_MISC, not EXITPOLICY, as the reason code when we send
 back the end cell. I'm so smart. Nevermind my above comment about
 EXITPOLICY then.

 Does that mean the current plan is to use MISC for now (and for the next
 couple years), but teach clients to recognize some other reason, and
 eventually plan to move to sending that reason, at which point clients
 will be able to discover that they're being refused due to thinking
 they're trying to cheat?

 > For configuration, a tristate torrc option whose default value is "do
 whatever the consensus says" seems right.

 That does seem slightly smarter than what I was going to do, which was a
 boolean config option which turns it on no matter what when 1, and uses
 whatever's in the consensus when 0. If you have some implementation ideas
 in mind, please do run with them.

-- 
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/1751#comment:18>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online