[tor-bugs] #15482 [Tor]: Don't surprise users with new circuits in the middle of browsing

[Tor Bug Tracker & Wiki]

#15482: Don't surprise users with new circuits in the middle of browsing
-------------------------+-------------------------------------------------
     Reporter:           |      Owner:  yawning
  mikeperry              |     Status:  assigned
         Type:           |  Milestone:  Tor: 0.2.7.x-final
  enhancement            |    Version:  Tor: unspecified
     Priority:  normal   |   Keywords:  tbb-usability, tbb-wants, tor-core,
    Component:  Tor      |  TorCoreTeam201508
   Resolution:           |  Parent ID:
Actual Points:           |
       Points:           |
-------------------------+-------------------------------------------------

Comment (by mikeperry):

 FWIW, I like the idea behind rustybird's second patch
 (https://trac.torproject.org/projects/tor/attachment/ticket/15482/IsolateKeepAliveSOCKSAuth.patch)
 minus the needless whitespace changes.

 I think any form of max lifespan opens up the user to both guard discovery
 attacks as well as increased exit node and correlation exposure (because a
 max lifespan allows an application to be induced to continually reconnect
 until a compromised middle or exit node is chosen on a new circuit).

 Beyond the security concerns (which should be sufficient by themselves),
 it also terrible for usability. The lifespan of HTTP connections is a
 relic of the shittiness of HTTP/1.x. Both HTTP/2 and QUIC fix this, and
 keep connections opened forever, because that is how sessions actually
 work on the web. To drive home the usability impact of enforcing this max
 lifespan: would we ever force people to reconnect to their SSH servers
 every X minutes/hours/days through Tor? If we're not willing to do that,
 we shouldn't to the equivalent to the web.

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/15482#comment:31>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online