TestWebKitAPI.EventAttribution.Basic has become extremely flaky on api-ios EWS since last few days. It is frequently causing false positives on EWS, blaming innocent patches for causing this test failure. e.g.: https://ews-build.webkit.org/#/builders/9/builds/53916 https://ews-build.webkit.org/#/builders/9/builds/53850 https://ews-build.webkit.org/#/builders/9/builds/53855 https://ews-build.webkit.org/#/builders/9/builds/53858 https://ews-build.webkit.org/#/builders/9/builds/53871 https://ews-build.webkit.org/#/builders/9/builds/53878 https://ews-build.webkit.org/#/builders/9/builds/53884
Unfortunately https://results.webkit.org/?suite=api-tests&test=TestWebKitAPI.EventAttribution.Basic doesn't have lot of data for iOS release because of Bug 229994.
Recent changes in were in r281723 and r281779.
Recent changes in EventAttribution.mm were in r281723 and r281779
Committed r282079 (241379@main): <https://commits.webkit.org/241379@main>
Tried on both iOS 14 simulator and BigSur, but I was not able to reproduce the timeout locally using run-api-tests --iterations 50 --ios-simulator TestWebKitAPI.EventAttribution.Basic --release.
<rdar://problem/82826500>
(In reply to ayumi_kojima from comment #5) > Tried on both iOS 14 simulator and BigSur, but I was not able to reproduce > the timeout locally using run-api-tests --iterations 50 --ios-simulator > TestWebKitAPI.EventAttribution.Basic --release. Try just `TestWebKitAPI.EventAttribution`, bet we have some test results poisoning one another.
I'm looking into this. I bet adding more tests caused state to be left on disk from one test affecting a future test.
Created attachment 437531 [details] Patch
Created attachment 437533 [details] Patch
Created attachment 437534 [details] Patch
Committed r282102 (241400@main): <https://commits.webkit.org/241400@main> All reviewed patches have been landed. Closing bug and clearing flags on attachment 437534 [details].
*** Bug 229852 has been marked as a duplicate of this bug. ***