coveralls seems broken <title>Coveralls :: Oops, something went wrong (500)</title>

Ignore it, I'm seeing something similar it might be cloudfront related, that is the only host holding things up.

d3dy5gmtp8yhk7.cloudfront.net

== TIMING INFO ============
ClientConnected: 02:54:49.291
ClientBeginRequest: 02:54:49.320
GotRequestHeaders: 02:54:49.320
ClientDoneRequest: 02:54:49.320
Determine Gateway: 0ms
DNS Lookup: 70ms
TCP/IP Connect: 20999ms
HTTPS Handshake: 0ms
ServerConnected: 02:55:10.392
FiddlerBeginRequest: 00:00:00.000
ServerGotRequest: 00:00:00.000
ServerBeginResponse: 00:00:00.000
GotResponseHeaders: 00:00:00.000
ServerDoneResponse: 00:00:00.000
ClientBeginResponse: 02:55:10.392
ClientDoneResponse: 02:55:10.392

Overall Elapsed: 0:00:21.072

Thanks!

The removed torture tests took the coverage down from ~75% to ~70%. Clearly lots of exit paths that the normal tests never hit. But now the coverage build completed in 32 minutes instead of timing out after 50 so I see no real choice for us here. An alternative would be to run a few torture tests just to exercise the most common exit paths... I'll add a few and see how this affects the build.

With just a small set of torture tests, we're back on 74.589%