Bug 206949

Summary: [ Mac ] fast/dom/Window/post-message-crash.html is flaky failing
Product: WebKit Reporter: Jason Lawrence <Lawrence.j>
Component: New BugsAssignee: Nobody <webkit-unassigned>
Status: RESOLVED FIXED    
Severity: Normal CC: cdumez, tsavell, webkit-bot-watchers-bugzilla, webkit-bug-importer
Priority: P2 Keywords: InRadar
Version: WebKit Nightly Build   
Hardware: Unspecified   
OS: Unspecified   
Attachments:
Description Flags
Update Test Expectations none

Description Jason Lawrence 2020-01-29 10:48:58 PST 
fast/dom/Window/post-message-crash.html

Description: 
This test is flaky failing on Mac. I see a potential regression point of 252824.

History:
https://results.webkit.org/?suite=layout-tests&test=fast%2Fdom%2FWindow%2Fpost-message-crash.html

Diff:
--- /Volumes/Data/slave/catalina-leaks/build/layout-test-results/fast/dom/Window/post-message-crash-expected.txt
+++ /Volumes/Data/slave/catalina-leaks/build/layout-test-results/fast/dom/Window/post-message-crash-actual.txt
@@ -1,3 +1,3 @@
 This ensures that postMessage called in a callback does not crash.
 
-PASS
+FAIL
Comment 1 Radar WebKit Bug Importer 2020-01-29 10:49:24 PST 
<rdar://problem/58997453>
Comment 2 Jason Lawrence 2020-01-29 10:54:48 PST 
Created attachment 389157 [details]
Update Test Expectations
Comment 3 Truitt Savell 2020-01-29 11:14:34 PST 
Comment on attachment 389157 [details]
Update Test Expectations

Clearing flags on attachment: 389157

Committed r255370: <https://trac.webkit.org/changeset/255370>
Comment 4 Alexey Proskuryakov 2020-01-29 23:09:09 PST 
This test was already marked as flaky on iOS. Clearly a regression on Mac, at least in frequency.
Comment 5 Chris Dumez 2020-01-30 08:50:58 PST 
The test relies on a 50ms timer :(

setTimeout(function() {
    if (window.testRunner)
        testRunner.notifyDone();
}, 50);
Comment 6 Chris Dumez 2020-01-30 08:56:17 PST 
Committed r255427: <https://trac.webkit.org/changeset/255427>
Comment 7 Alexey Proskuryakov 2020-01-30 09:00:24 PST 
This would have helped help if the test lacked output, but that's not the failure seen here.
Comment 8 Alexey Proskuryakov 2020-01-30 09:04:00 PST 
I was wrong, the test replaces FAIL with PASS.