The first failure reported by the bots it's in r221942. Since then it has flaky behavior. The same has happened almost simultaneously with this WebRTC test, so it may be related (e.g. a regression in MediaPlayer maybe?), but it could have also been by chance: webrtc/clone-audio-track.html https://bugs.webkit.org/show_bug.cgi?id=176850
Marked as expected crash in r221992. Please update expectations when fixed.
That test has been fixed in the new implementation of MediaStream (r232589) and is now enabled in the exectations. Closing