Thanks for the PR! Do we know why this test stopped working? Although if this unblocks CI I am ok to merge nonetheless.

Hi @kongtiaowang, I've gone through your changes and even I had noticed this test and submitted a fix in #10355. Based on your PR, I have a few questions for my own understanding.

While changing the search term to 430 (expecting 0 results) does fix the CI failure, I'm concerned it might accidentally dilute the test's effectiveness. By changing it to a negative test (checking for no results), we lose the ability to verify that the filter can actually find matching records. A broken filter that always returns 0 rows would still pass this new version.

In my PR, I kept the original test intent (filtering for 4300 and expecting 1 row) but fixed the underlying race condition by adding WebDriverWait. This ensures the test waits for React to re-render before checking the results, which makes it robust without sacrificing the positive test coverage.

Your fix is definitely great for getting CI green quickly! But long-term, do you think we should aim to keep the positive test case to ensure the feature is fully verified?

Would love to hear your thoughts in case I'm missing something.

Hi @arnav-makkar Thanks for your suggestion! I’ve already tried the approach you shared (#10362), but the test still occasionally fails in CI. Since PHP 8.3 will be deprecated in a few months, I think the best course for now is to ensure CI passes. I plan to revisit and optimize this test case when we upgrade to PHP 8.5. Really appreciate your input!

@arnav-makkar I saw your PR, it has been fixed the issue. I will close this one.