[random test failure] Try to un-skip and fix InlineBlockTest::testDeletion in context of [#3353085]

Test seems too slow for 1500x, let's try 1000x.

Right...

Let's do what I've should have done from the start, test _only_

Very minimal difference, only one fail in the should_fail, but it is with

1) Drupal\Tests\layout_builder\FunctionalJavascript\InlineBlockTest::testDeletion
Error: cannot call methods on dialog prior to initialization; attempted to call method 'widget'
    at Function.error (http://php-apache-jenkins-drupal-patches-177329/subdirectory/core/assets/vendor/jquery/jquery.min.js?v=3.6.3:2:2616)
    at HTMLDivElement. (http://php-apache-jenkins-drupal-patches-177329/subdirectory/core/assets/vendor/jquery.ui/ui/widget-min.js?v=10.1.0-dev:9:2224)
    at Function.each (http://php-apache-jenkins-drupal-patches-177329/subdirectory/core/assets/vendor/jquery/jquery.min.js?v=3.6.3:2:3003)
    at E.fn.init.each (http://php-apache-jenkins-drupal-patches-177329/subdirectory/core/assets/vendor/jquery/jquery.min.js?v=3.6.3:2:1481)
    at t.fn. [as dialog] (http://php-apache-jenkins-drupal-patches-177329/subdirectory/core/assets/vendor/jquery.ui/ui/widget-min.js?v=10.1.0-dev:9:1962)
    at Object.getContainer (http://php-apache-jenkins-drupal-patches-177329/subdirectory/core/misc/dialog/off-canvas/js/off-canvas.js?v=10.1.0-dev:302:23)
    at resetSize (http://php-apache-jenkins-drupal-patches-177329/subdirectory/core/misc/dialog/off-canvas/js/off-canvas.js?v=10.1.0-dev:207:42)
    at later (http://php-apache-jenkins-drupal-patches-177329/subdirectory/core/misc/debounce.js?v=10.1.0-dev:37:23)

which is the exact failure that was fixed in 🐛 [random test failure] Drupal\Tests\layout_builder\FunctionalJavascript\LayoutBuilderUiTest::testReloadWithNoSections() Fixed , so I believe this one to be fixed as well.

Amazing.

Both should_fail and non_fail patch ran 5 times. That's 7500 individual tests per patch.

Back to RTBC.

I've just committed 📌 [random test failure] Try to un-skip and fix InlineBlockPrivateFilesTest::testPrivateFiles() in context of [#3353085] Fixed , will aim to commit this one (or another similar RTBC one) in roughly 24 hours.

Our fail patch is not failing here. Queueing more jobs to see if it's just unlucky or a lower fail rate. If it continues to pass we should double-check that we're fixing the right thing, or look to external factors like the AWS instances.

Blergh, fixing attribution.

One of the should-fail-jobs in #5 shows the expected fail, so we are good there. See what I mean about needing more test runs. :)

To be fair, it did already fail: https://www.drupal.org/pift-ci-job/2639266 → , but more tests are better tests.

Based on the current fail rate, even 24,000 jobs (16 test runs queued) would still have a 1% chance of a false pass, so queueing even more. I think this is definitely one where the race condition is worse depending on external factors like the AWS instance or such (which is also why it's important to queue the pass- and fail-patch test jobs at the same time on the same environments).

Patch no longer applied for some reason, but I just made the change manually. Committed/pushed to 10.1.x, thanks!

Moving to 10.0.x for backport.

Patch applies to 10.0.x, 9.5.x and 9.4.x

Seems several testing rounds happened in #6

Using 'to be ported' as 'need to cherry-pick this once all the related patches are in 10.1' so it's easier to tell which have already been committed vs. not.

Cherry-picked to 10.0.x and 9.5.x, thanks!

Automatically closed - issue fixed for 2 weeks with no activity.