New Failure LldbGdbServerTestCase.test_Hc_then_Csignal_signals_correct_thread_launch_llgs_dwarf just started failing in top-of-tree

http://lab.llvm.org:8011/builders/lldb-x86_64-ubuntu-14.04-cmake/builds/726

======================================================================
ERROR: test_Hc_then_Csignal_signals_correct_thread_launch_llgs_dwarf (TestLldbGdbServer.LldbGdbServerTestCase)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/lldb-buildbot/lldbSlave/buildWorkingDir/llvm/tools/lldb/test/lldbtest.py", line 513, in wrapper
    return func(self, *args, **kwargs)
  File "/lldb-buildbot/lldbSlave/buildWorkingDir/llvm/tools/lldb/test/lldbtest.py", line 479, in wrapper
    return func(self, *args, **kwargs)
  File "/lldb-buildbot/lldbSlave/buildWorkingDir/llvm/tools/lldb/test/tools/lldb-server/TestLldbGdbServer.py", line 869, in test_Hc_then_Csignal_signals_correct_thread_launch_llgs_dwarf
    self.Hc_then_Csignal_signals_correct_thread(signal.SIGSEGV)
  File "/lldb-buildbot/lldbSlave/buildWorkingDir/llvm/tools/lldb/test/tools/lldb-server/TestLldbGdbServer.py", line 790, in Hc_then_Csignal_signals_correct_thread
    context = self.expect_gdbremote_sequence(timeout_seconds=10)
  File "/lldb-buildbot/lldbSlave/buildWorkingDir/llvm/tools/lldb/test/tools/lldb-server/gdbremote_testcase.py", line 501, in expect_gdbremote_sequence
    return expect_lldb_gdbserver_replay(self, self.sock, self.test_sequence, timeout_seconds, self.logger)
  File "/lldb-buildbot/lldbSlave/buildWorkingDir/llvm/tools/lldb/test/tools/lldb-server/lldbgdbserverutils.py", line 234, in expect_lldb_gdbserver_replay
    raise Exception("timed out while waiting for packet match (receive buffer: {})".format(pump.get_receive_buffer()))
Exception: timed out while waiting for packet match (receive buffer: )
Config=x86_64-clang

Thanks,
Ilia

I had a major re-factor around the TestLldbGdbServer test cases (http://reviews.llvm.org/rL231922) but based on the logs the failure is unrelated to it.

I run the test locally several time (cmake + ninja, linux - x86-64) and seen the first failure after more than 100 runs, so I can imagine that this test case is a little bit flaky and we haven’t seen the failure before because we haven’t hit the specific failure case. The linked build bot haven’t caught this failure again since then.

I attached the packet log and the process log for the failure if somebody have any idea about what caused it.

Thanks,
Tamas

packets-TestLldbGdbServer.LldbGdbServerTestCase.test_Hc_then_Csignal_signals_correct_thread_launch_llgs_dwarf.log (3.19 KB)

process-TestLldbGdbServer.LldbGdbServerTestCase.test_Hc_then_Csignal_signals_correct_thread_launch_llgs_dwarf.log (35.3 KB)

If you feel like looking into this now, and if you still have the same build of lldb, I would run the test again and compare the traces (if it passes). BeyondCompare does a great job of comparing files with minor differences (timestamps, process ids, etc)