Investigate Tor Browser 4.5's new circuit lifetime behaviour

For various reasons, Tor Browser 4.5 raises MaxCircuitDirtiness to 2 hours:

• https://trac.torproject.org/projects/tor/ticket/13766
• https://gitweb.torproject.org/builders/tor-browser-bundle.git/commit/?id=824116042a03683ae20fa6e7d0768d8fd6fd6f95

It makes sense since that version has per-tab / per-top-level-domain / something circuit isolation. They can easily do it since their tor is only meant to be used by the browser.

Does our current circuit isolation setup allow us to safely do the same globally? If not, then this issue might raise a few non-trivial ones, so better answer this question ASAP => hence the >normal priority.

Feature Branch: feature/8925-tor-0.2.6

Parent Task: #9031 (closed)

Related issues

• Blocked by #7934 (closed)

Original created by @intrigeri on 9114 (Redmine)