[tor-bugs] #7656 [TorBrowserButton]: Save network, delete polluting, disable network.http.connection-retry-timeout

[Tor Bug Tracker & Wiki]

#7656: Save network, delete polluting, disable network.http.connection-retry-
timeout
---------------------------------+------------------------------------------
    Reporter:  cypherpunks       |       Owner:  mikeperry
        Type:  defect            |      Status:  reopened 
    Priority:  blocker           |   Milestone:           
   Component:  TorBrowserButton  |     Version:           
  Resolution:                    |    Keywords:           
      Parent:                    |      Points:           
Actualpoints:                    |  
---------------------------------+------------------------------------------

Comment(by arma):

 {{{
 <oftc_must_be_destroyed> because it overloaded by firefox attempt second
 conn
 > hey, speaking of which. has anybody tracked down when and why that
 firefox-second-attempt happens?
 > (and can we turn it off?)
 <oftc_must_be_destroyed> yes we can!
 <oftc_must_be_destroyed> everything tracked
 <oftc_must_be_destroyed> but nobody cares
 <oftc_must_be_destroyed> #7656
 > fun. does that behavior happen when there's a proxy set too?
 > "At a glance, it looks like this only applies to native socket
 transports, not proxies"
 <oftc_must_be_destroyed> yes
 <oftc_must_be_destroyed> happen
 > i guess it is especially bad for tor, since it very likely just reuses
 the same circuit as the first request
 <oftc_must_be_destroyed> yes
 }}}

 Looks like http://hg.mozilla.org/mozilla-central/rev/be4b69a4babf is where
 the dreaded "feature" went into Firefox. Turning it off does indeed sound
 like a wise plan.

-- 
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/7656#comment:6>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online