[tor-qa] Experimental 3.5.2 bundles with tor-fw-helper (trial 2)

[David Fifield]

(Devs are in Iceland, post bundles.)

Here are revised bundles in response to some of the problems we've seen
so far (thanks especially to Lunar).

https://people.torproject.org/~dcf/pt-bundle/3.5.2-fwhelper-2/

The changes relative to 3.5.2-fwhelper-1:
 * NAT-PMP is completely disabled and removed. it didn't seem to be
   working (maybe it is only for Apple routers), and we saw one of
   libnatpmp's programs segfault (though not tor-fw-helper itself).
   https://trac.torproject.org/projects/tor/ticket/5213#comment:14
 * The client is forced to use IPv4 with the -4 option. Though
   flashproxy-client supports IPv6, and IPv6 can sometimes work to get
   around NAT, tor-fw-helper only does IPv4 port forwardings, and that's
   what we're trying to test. We encountered a problem where the
   external IPv6 address was being registered, rather than the IPv4
   address that was forwarded.
 * Uses a static port 9000 for the external listener, instead of an
   ephemeral port. This is to make it easier for testers to clean up
   their port forwarding table, because tor-fw-helper appears to leave
   the forwardings active.
   https://lists.torproject.org/pipermail/tor-qa/2014-February/000335.html

Like before, you just have to click "Connect" and wait up to 60 seconds.
You don't have to do any special tests; just see whether it bootstraps
or not.

These bundles are still nothing close to a release candidate. We'll need
to find a way to limit how long port forwarding lasts, at least, and
make sure that the UPnP library is safe to deploy.

Changes relative to master:
https://gitweb.torproject.org/user/dcf/tor-browser-bundle.git/commitdiff/3.5.2-fwhelper-2?hp=tbb-3.5.2-build5
Changes relative to 3.5.2-fwhelper-1:
https://gitweb.torproject.org/user/dcf/tor-browser-bundle.git/commitdiff/3.5.2-fwhelper-2?hp=3.5.2-fwhelper-1

David Fifield