The following layout test is flaky on macOS http/wpt/mediarecorder/MediaRecorder-AV-audio-video-dataavailable.html Probable cause: Unknown, this test has been flaky for a while and the regression point can not be determined by looking at the dashboard. Flakiness Dashboard: https://webkit-test-results.webkit.org/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=http%2Fwpt%2Fmediarecorder%2FMediaRecorder-AV-audio-video-dataavailable.html --- /Volumes/Data/slave/mojave-leaks/build/layout-test-results/http/wpt/mediarecorder/MediaRecorder-AV-audio-video-dataavailable-expected.txt +++ /Volumes/Data/slave/mojave-leaks/build/layout-test-results/http/wpt/mediarecorder/MediaRecorder-AV-audio-video-dataavailable-actual.txt @@ -1,4 +1,7 @@ +CONSOLE MESSAGE: line 2659: Error: assert_approx_equals: Red channel of the pixel at (25, 25) expected 255 +/- 20 but got 0 -PASS MediaRecorder can successfully record the video for a audio-video stream +Harness Error (FAIL), message = Error: assert_approx_equals: Red channel of the pixel at (25, 25) expected 255 +/- 20 but got 0 +TIMEOUT MediaRecorder can successfully record the video for a audio-video stream Test timed out +
<rdar://problem/50554421>
This is still happening, we should mark the test as flaky.
This test is flaky on iOS too, re-titling. Also seeing less frequent timeouts, especially on internal testers. Added expectations in http://trac.webkit.org/r253891