Test Failures on Linux

I’m getting these test failures on linux righ tnow.

I’m still getting these test failures on Linux:

FAIL: LLDB (suite) :: TestDataFormatterAdv.py (Linux zturner-goobuntu.mtv.corp.google.com 3.13.0-45-generic #74-Ubuntu SMP Tue Jan 13 19:36:28 UTC 2015 x86_64 x86_64)
FAIL: LLDB (suite) :: TestDataFormatterCpp.py (Linux zturner-goobuntu.mtv.corp.google.com 3.13.0-45-generic #74-Ubuntu SMP Tue Jan 13 19:36:28 UTC 2015 x86_64 x86_64)
FAIL: LLDB (suite) :: TestDataFormatterSmartArray.py (Linux zturner-goobuntu.mtv.corp.google.com 3.13.0-45-generic #74-Ubuntu SMP Tue Jan 13 19:36:28 UTC 2015 x86_64 x86_64)

Are these new test failures? Are they known about? If they’re known, can we XFAIL them to get a clean run? If they’re new, does anyone know which CL caused it? A discussion on IRC this morning involved someone who committed something last week which caused 20 new failures. But he never got an email about it so he didn’t know to fix it, because the bots are always red. This is why we need to immediately revert things that break the build, so that anyone who breaks the build knows about it and can fix it, and not let it linger to the point that it’s impossible to revert.

These are fairly old test failures, they should not be related to the recent breakage. I believe we have someone looking into them, but I don’t know who it is. o:)

I'm getting these test failures on linux righ tnow.

I'm still getting these test failures on Linux:

FAIL: LLDB (suite) :: TestDataFormatterAdv.py (Linux
zturner-goobuntu.mtv.corp.google.com 3.13.0-45-generic #74-Ubuntu SMP Tue
Jan 13 19:36:28 UTC 2015 x86_64 x86_64)
FAIL: LLDB (suite) :: TestDataFormatterCpp.py (Linux
zturner-goobuntu.mtv.corp.google.com 3.13.0-45-generic #74-Ubuntu SMP Tue
Jan 13 19:36:28 UTC 2015 x86_64 x86_64)
FAIL: LLDB (suite) :: TestDataFormatterSmartArray.py (Linux
zturner-goobuntu.mtv.corp.google.com 3.13.0-45-generic #74-Ubuntu SMP Tue
Jan 13 19:36:28 UTC 2015 x86_64 x86_64)

As a point of comparison these three are not failing on FreeBSD.

... he never got an email
about it so he didn't know to fix it, because the bots are always red. This
is why we need to immediately revert things that break the build, so that
anyone who breaks the build knows about it and can fix it, and not let it
linger to the point that it's impossible to revert.

Completely agreed. Right now on FreeBSD only the lldb-mi tests are
failing, presumably due to the thread race conditions that have been
discussed on this list. I'm considering having FreeBSD just skip the
lldb-mi tests for now, until there are changes there to test.

I uploaded a CL to fix these yesterday.