Detected 5 new flakes for test/step "LabelTest.FontPropertySymbol". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyJwsSBUZsYWtlIhxMYWJlbFRlc3QuRm9udFByb3BlcnR5U3ltYm9sDA. This message was posted automatically by the chromium-try-flakes app. Since flakiness is ongoing, the issue was moved back into Sheriff Bug Queue (unless already there).

The LabelTest.FontPropertySymbol failed during a try run for the CL when the Skia DEPS were rolled and has since been flaking out, as per above. Can you see if it is related and if not, re-assign to who you think is most appropriate to handle it?

Detected 6 new flakes for test/step "LabelTest.FontPropertySymbol". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyJwsSBUZsYWtlIhxMYWJlbFRlc3QuRm9udFByb3BlcnR5U3ltYm9sDA. This message was posted automatically by the chromium-try-flakes app. Since flakiness is ongoing, the issue was moved back into Sheriff Bug Queue (unless already there).

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/5db81e12c9e4b5c39123488df7e367afcb86706e

commit 5db81e12c9e4b5c39123488df7e367afcb86706e
Author: gogerald <gogerald@chromium.org>
Date: Thu May 19 17:03:49 2016

Disable flaky test LabelTest.FontPropertySymbol on Linux

This test gets flaky and failed frequently on Linux recently.

BUG=612406
TBR=sky@chromium.org

Review-Url: https://codereview.chromium.org/1995003003
Cr-Commit-Position: refs/heads/master@{#394800}

[modify] https://crrev.com/5db81e12c9e4b5c39123488df7e367afcb86706e/ui/views/controls/label_unittest.cc

Disabled the test on Linux platform.

The one change in that Skia roll could not have possibly affected the result being asserted. Interestingly, this is only 'flaky' in the sense that some runs consistently fail (if a run fails this test the first time, the reties never appear to succeed). This seems to happen once every ~100 runs or so of the linux_chromium_rel_ng trybot, which appears to have ~300 slaves.

Interestingly LabelTest.FontPropertyArial which is an exact copy paste of LabelTest.FontPropertySymbol with the exception of using "arial" instead of "symbol" never seems to fail. Since all of the 'test' setups include an "arial" but none of them seem to contain "symbol" it seems likely that this is some sort of race with some other test mucking about with the (unfortunately) global FontConfig instance. However, it isn't clear which test might be triggering this.

After taking a look at gfx::SetUpFontconfig() and generally where FcInit is used (and it never should be used) it seems strange that this sort of thing wouldn't be happening all the time.

Detected 4 new flakes for test/step "LabelTest.FontPropertySymbol". To see the actual flakes, please visit https://chromium-try-flakes.appspot.com/all_flake_occurrences?key=ahVzfmNocm9taXVtLXRyeS1mbGFrZXNyJwsSBUZsYWtlIhxMYWJlbFRlc3QuRm9udFByb3BlcnR5U3ltYm9sDA. This message was posted automatically by the chromium-try-flakes app. Since flakiness is ongoing, the issue was moved back into Sheriff Bug Queue (unless already there).

Adding Test-Disabled label, the Tests>Disabled component is going away.