It works for me.

There are two logs because one corresponds to stdout and the other to stderr. I'm not sure if that distinction is present on Windows. It so happens that the test runner uses stdout, so that's the interesting one.

Searching for the test name reveals:

Timeout happened...killed process "cd "/tmp/ghctest-wSvwjx/test spaces/./rename/should_compile/rn003.run" && "/var/drydock/workingcopy-13/repo/glasgow-haskell-compiler/bindisttest/install dir/bin/ghc" -c rn003.hs -dcore-lint -dcmm-lint -dno-debug-output -no-user-package-db -rtsopts -fno-warn-missed-specialisations -fshow-warning-groups "...

Which can mean one of the these things:
1) The test legitimately got slower or loops, resulting in a timeout

2) Harbormaster machine was slightly overloaded when running this test

3) This test is flaky

The fact that it works on my machine (with DEBUG even!) suggests that 1) is unlikely.

There's another commit being tested right now (see the top of https://phabricator.haskell.org/diffusion/GHC/history/master/), unless the test fails on that as well, I would classify this as infra failure.

2016-10-01 1:43 GMT+09:00 Simon Peyton Jones via ghc-devs <ghc-devs@haskell.org>:

Thanks Chris, that’s helpful. I got to your final link, but it didn’t show me what rn003 did. There seem to be two logs for some reason. I tried clicking “Unlimited” lines, but I still didn’t see a clear failure from rn003.

Does it fail for anyone else?