Bug 262134 - REGRESSION(268154@main): [ macOS ] TestWebKitAPI.WKNavigation.WebProcessLimit is a constant failure
Summary: REGRESSION(268154@main): [ macOS ] TestWebKitAPI.WKNavigation.WebProcessLimit...
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: WebKit Process Model (show other bugs)
Version: WebKit Nightly Build
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Chris Dumez
URL:
Keywords: InRadar
: 261946 (view as bug list)
Depends on:
Blocks:
 
Reported: 2023-09-26 11:53 PDT by Ben Schwartz
Modified: 2023-09-27 18:12 PDT (History)
4 users (show)

See Also:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Ben Schwartz 2023-09-26 11:53:40 PDT 
TestWebKitAPI.WKNavigation.WebProcessLimit

This API test is constantly failing on macOS.

HISTORY:

https://results.webkit.org/?suite=api-tests&test=TestWebKitAPI.WKNavigation.WebProcessLimit&platform=mac

STDIO LOG:

    TestWebKitAPI.WKNavigation.WebProcessLimit
        
        /Volumes/Data/worker/Apple-Ventura-Release-Build/build/Tools/TestWebKitAPI/Tests/WebKitCocoa/WebContentProcessDidTerminate.mm:445
        Value of: didCrash
          Actual: true
        Expected: false

REPRODUCIBILITY:

I was able to reproduce this issue on macOS Ventura (13.5, 22G74/arm64) ToT using the following command:

run-api-tests --debug --verbose TestWebKitAPI.WKNavigation.WebProcessLimit

REGRESSION:

I was able to identify a potential regression point at 268154@main. As I don't have access to a reliable testing environment, I can't bisect to be sure; however, changes at this point seem directly related to the issue and the first failure instance is only a few commits later at 268157@main.
Comment 1 Radar WebKit Bug Importer 2023-09-26 11:53:53 PDT 
<rdar://problem/116069794>
Comment 2 EWS 2023-09-26 14:02:11 PDT 
Test gardening commit 268478@main (39614e3c76d4): <https://commits.webkit.org/268478@main>

Reviewed commits have been landed. Closing PR #18243 and removing active labels.
Comment 3 Ben Schwartz 2023-09-26 14:03:09 PDT 
FULL STDIO:

TestWebKitAPI.WKNavigation.WebProcessLimit
        
        /Volumes/Data/worker/Apple-Ventura-Release-Build/build/Tools/TestWebKitAPI/Tests/WebKitCocoa/WebContentProcessDidTerminate.mm:445
        Value of: didCrash
          Actual: true
        Expected: false
        
        
        /Volumes/Data/worker/Apple-Ventura-Release-Build/build/Tools/TestWebKitAPI/Tests/WebKitCocoa/WebContentProcessDidTerminate.mm:447
        Expected: ([view _webProcessIdentifier]) != (0), actual: 0 vs 0
        
        
        /Volumes/Data/worker/Apple-Ventura-Release-Build/build/Tools/TestWebKitAPI/Tests/WebKitCocoa/WebContentProcessDidTerminate.mm:452
        Expected equality of these values:
          views[0]
            Which is: 1
          view
            Which is: 0x13e609f50
        
        
        /Volumes/Data/worker/Apple-Ventura-Release-Build/build/Tools/TestWebKitAPI/Tests/WebKitCocoa/WebContentProcessDidTerminate.mm:462
        Expected: ([views[i] _webProcessIdentifier]) != (0), actual: 0 vs 0
        
        
        /Volumes/Data/worker/Apple-Ventura-Release-Build/build/Tools/TestWebKitAPI/Tests/WebKitCocoa/WebContentProcessDidTerminate.mm:467
        Expected equality of these values:
          views[1]
            Which is: 1
          view
            Which is: 0x11ea0b3d0
        
        
        /Volumes/Data/worker/Apple-Ventura-Release-Build/build/Tools/TestWebKitAPI/Tests/WebKitCocoa/WebContentProcessDidTerminate.mm:477
        Expected: ([views[i] _webProcessIdentifier]) != (0), actual: 0 vs 0
Comment 4 Chris Dumez 2023-09-26 14:58:10 PDT 
Pull request: https://github.com/WebKit/WebKit/pull/18246
Comment 5 EWS 2023-09-26 18:58:19 PDT 
Committed 268492@main (50b0cd282f38): <https://commits.webkit.org/268492@main>

Reviewed commits have been landed. Closing PR #18246 and removing active labels.
Comment 6 Alexey Proskuryakov 2023-09-27 18:12:29 PDT 
*** Bug 261946 has been marked as a duplicate of this bug. ***