[Bug]: loadUrl sometimes returns wrong html content

[bayun2]

Preflight Checklist

• I have read the Contributing Guidelines for this project.
• I agree to follow the Code of Conduct that this project adheres to.
• I have searched the issue tracker for a bug report that matches the one I want to file, without success.

Electron Version

21.0.1

What operating system are you using?

macOS

Operating System Version

macOS Big Sur 11.5.2

What arch are you using?

x64

Last Known Working Electron version

No response

Expected Behavior

loadURL can return correct html content

Actual Behavior

loadURL can return correct html content at first, when i use cmd + R to refresh html, sometimes return js content as html.

html file content inside asar is always correct

Testcase Gist URL

No response

Additional Information

when i use cmd + R to refresh html, sometimes return js content as html.

html file content inside asar is always correct

loadUrl in the following way

windowConfig

static defaultWindowOptions: () => Electron.BrowserWindowConstructorOptions = () => ({
  ...LBConfig.defaultWindowSizeOptions,
  minWidth: 1440,
  minHeight: 800,
  titleBarStyle: 'hidden',
  frame: false,
  titleBarOverlay: false,
  acceptFirstMouse: true,
  webPreferences: {
    minimumFontSize,
    sandbox: false,
    nodeIntegration: false,
    contextIsolation: true,
    preload: __dirname + '/preload.js',
    devTools: true,
  }
})

Probably an error caused by autoupdate，The zip package has been downloaded, but the installation has not been triggered

[deermichel]

Thanks for reporting this and helping to make Electron better!

Because of time constraints, triaging code with third-party dependencies is usually not feasible for a small team like Electron's.

Would it be possible for you to make a standalone testcase with only the code necessary to reproduce the issue? For example, Electron Fiddle is a great tool for making small test cases and makes it easy to publish your test case to a gist that Electron maintainers can use.

Stand-alone test cases make fixing issues go more smoothly: it ensure everyone's looking at the same issue, it removes all unnecessary variables from the equation, and it can also provide the basis for automated regression tests.

I'm adding the blocked/need-repro label for this reason. After you make a test case, please link to it in a followup comment. This issue will be closed in 10 days if the above is not addressed.

Regarding the update hypothesis, there might be a small time window during app restart when an update is applied that causes undefined behavior (ref #36117) - however this doesn't affect the app during regular runtime (including reloads).

[bayun2]

@deermichel thank you for your reply.

I have found a path to reproduce.

When the software exits with quitAndInstall or autoInstallOnAppQuit = true, if the software is quickly opened again during Shipit running, the software is still running the old code. At this time, the bug described can be reproduced by refreshing with cmd+R.

I'm sorry, I can't provide a direct testcase because this bug depends on an update of the package to reproduce

[VerteDinde]

@bayun2 We have a fix for this Shipit bug incoming here: #36130

[codebytere]

Closing as #36130 has been merged.