Fix Flaky tests

[harshach]

Describe your changes:

Fixes

I worked on ... because ...

Type of change:

• Bug fix
• Improvement
• New feature
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation

Checklist:

• I have read the CONTRIBUTING document.
• My PR title is Fixes <issue-number>: <short explanation>
• I have commented on my code, particularly in hard-to-understand areas.
• For JSON Schema changes: I updated the migration scripts or explained why it is not needed.

---

Summary by Gitar

• Test stabilization:
  • Replaced fixed Thread.sleep(5000) with polling in test_testSuiteReindexConsistency
• New utility methods:
  • TestUtils.waitForReindexCompletion() polls SearchIndexingApplication status with 60s timeout
  • TestSuiteResourceTest.listEntitiesFromSearchWithRetry() handles transient ES errors during reindex

_{This will update automatically on new commits.}

[gitar-bot]

🔍 CI failure analysis for 987f4e2: Different CI failures across retry attempts: Run 1 had ETag/GlossaryTerm bugs, Run 2 has FeedResource pagination issue.

Issue

The maven-mysql-ci job has different failures across retry attempts, indicating test instability:

Run 1 (Job 57210140871) - 3 failures:

1. BotResourceTest.patch_concurrent_updates_with_etag (ETag bug)
2. GlossaryTermResourceTest.put_addEntityCustomAttributes (custom attributes bug)
3. GlossaryTermResourceTest.test_performance_listEntities (timeout)

Run 2 (Job 57235837346) - 1 failure:

1. FeedResourceTest.get_listThreadsWithPagination - expected 4 threads but got 3

Root Cause

Changing Failure Pattern Indicates Test Environment Issues

The fact that different tests fail on retry strongly suggests environmental/timing issues rather than deterministic code bugs:

Run 2 Failure:

FeedResourceTest.get_listThreadsWithPagination:891
expected: <4> but was: <3>

A pagination test expected 4 feed threads but only found 3. This is likely a race condition where background thread creation or event processing hasn't completed before the assertion runs.

Previous Failures (Run 1):

• ETag concurrency bug (500 errors instead of 412)
• Custom attributes not clearing properly
• Performance timeout (3475ms > 3000ms limit)

These did NOT reproduce in Run 2, further confirming environmental factors.

Details

None of these failures are related to this PR's changes:

The PR only modifies:

• TableResourceTest.java (removed concurrent test)
• TestSuiteResourceTest.java (fixed reindex polling)
• TestUtils.java (added polling utilities)

No changes to:

• FeedResource
• BotResource
• GlossaryTermResource
• Entity relationship handling

Test Instability Pattern:

• Run 1: 3 specific failures (ETag, custom attributes, performance)
• Run 2: 1 different failure (feed pagination)
• Suggests: Test isolation issues, shared state, or async timing problems

Suggested Fix

1. FeedResource Pagination: Should be corrected by adding proper wait/polling for async feed thread creation to complete before asserting counts, similar to the reindex polling fix in this PR.

2. Previous Run 1 Failures: Should be corrected by:

   • Fixing ETag validation to return 412 instead of 500
   • Ensuring custom attributes are properly replaced
   • Optimizing or increasing timeout for performance tests

3. Overall Test Stability: Can be resolved by improving test isolation, cleaning up shared state between tests, and adding proper synchronization for async operations.

Rules 🎸 1 action taken

Gitar Rules

🎸 Summary Enhancement: Summary already accurate; no updates needed

➖ Flaky Test Retry: Already on retry attempt 2; max retries reached

➖ Rule Quality Standards: No Gitar rule markdown files modified in this PR

• Auto-apply suggestions - Allow Gitar to commit updates to this branch

Tip

Comment Gitar fix CI or enable auto apply.

_{Was this helpful? React with 👍 / 👎 | This comment will update automatically (Docs)}

[sonarqubecloud]

Quality Gate passed for 'open-metadata-ingestion'

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud