Feature/openj9 integration

.github/actions/run-gradle/action.yml

@@ -16,8 +16,8 @@
 # under the License.
 #
 ---
-name: "Gradle Setup"
-description: "Setup Java and Gradle"
+name: "Run Tests with Gradle"
+description: "Run specified Gradle test tasks with configuration for timeout and test catalog."
 inputs:
   # Composite actions do not support typed parameters. Everything is treated as a string
   # See: https://github.com/actions/runner/issues/2238

.github/scripts/develocity_reports.py

@@ -314,7 +314,7 @@ def get_build_info(self, build_ids: List[str], project: str, test_type: str, que
 
             # Remove already found builds from the search list
             build_ids = [bid for bid in build_ids if bid not in builds]
-        
+
         if not build_ids:
             logger.info("All builds found in cache")
             return builds
@@ -733,7 +733,7 @@ def print_summary(problematic_tests: Dict[str, Dict], flaky_regressions: Dict[st
 
     # Combine and sort all test cases by failure rate
     all_problem_cases = []
-    
+
     # Process problematic quarantined tests
     if len(problematic_tests) > 0:
         print(f"Found {len(problematic_tests)} tests that have been quarantined for a while and are still flaky.")
@@ -742,15 +742,15 @@ def print_summary(problematic_tests: Dict[str, Dict], flaky_regressions: Dict[st
             total_runs = test_case.outcome_distribution.total
             method_name = test_case.name.split('.')[-1]
             if total_runs > 0:
-                failure_rate = (test_case.outcome_distribution.failed + 
+                failure_rate = (test_case.outcome_distribution.failed +
                               test_case.outcome_distribution.flaky) / total_runs
                 all_problem_cases.append({
                     'class': full_class_name,
                     'method': method_name,
                     'failure_rate': failure_rate,
                     'total_runs': total_runs
                 })
-    
+
     # Process flaky regressions
     if len(flaky_regressions) > 0:
         print(f"Found {len(flaky_regressions)} tests that have started recently failing.")
@@ -763,8 +763,8 @@ def print_summary(problematic_tests: Dict[str, Dict], flaky_regressions: Dict[st
         })
 
     # Sort by failure rate descending
-    sorted_cases = sorted(all_problem_cases, 
-                         key=lambda x: x['failure_rate'], 
+    sorted_cases = sorted(all_problem_cases,
+                         key=lambda x: x['failure_rate'],
                          reverse=True)
 
     # Group by class
@@ -790,15 +790,12 @@ def main():
     token = None
     if os.environ.get("DEVELOCITY_ACCESS_TOKEN"):
         token = os.environ.get("DEVELOCITY_ACCESS_TOKEN")
-    elif os.environ.get("GE_ACCESS_TOKEN"):
-        # Special case for when we run in GHA
-        token = os.environ.get("GE_ACCESS_TOKEN").removeprefix("ge.apache.org=")
     else:
         print("No auth token was specified. You must set DEVELOCITY_ACCESS_TOKEN to your personal access token.")
         exit(1)
 
     # Configuration
-    BASE_URL = "https://ge.apache.org"
+    BASE_URL = "https://develocity.apache.org"
     PROJECT = "kafka"
     QUARANTINE_THRESHOLD_DAYS = 7
     MIN_FAILURE_RATE = 0.1
@@ -862,7 +859,7 @@ def main():
                 print(f"\nRecent Executions (last {len(details['recent_executions'])} runs):")
                 for entry in sorted(details['recent_executions'], key=lambda x: x.timestamp)[-5:]:
                     print(f"  {entry.timestamp.strftime('%Y-%m-%d %H:%M')} - {entry.outcome}")
-        
+
         # Print Cleared Tests
         print("\n## Cleared Tests (Ready for Unquarantine)")
         if not cleared_tests:
@@ -881,19 +878,19 @@ def main():
                 print(f"\nRecent Executions (last {len(details['recent_executions'])} runs):")
                 for entry in sorted(details['recent_executions'], key=lambda x: x.timestamp):
                     print(f"  {entry.timestamp.strftime('%Y-%m-%d %H:%M')} - {entry.outcome}")
-        
+
         # Print Defective Tests
         print("\n## High-Priority Quarantined Tests")
         if not problematic_tests:
             print("No high-priority quarantined tests found.")
         else:
             print("These are tests which have been quarantined for several days and need attention.")
             sorted_tests = sorted(
-                problematic_tests.items(), 
+                problematic_tests.items(),
                 key=lambda x: (x[1]['failure_rate'], x[1]['days_quarantined']),
                 reverse=True
             )
-            
+
             print(f"\nFound {len(sorted_tests)} high-priority quarantined test classes:")
             for full_class_name, details in sorted_tests:
                 class_result = details['container_result']
@@ -916,21 +913,21 @@ def main():
                     key=lambda x: (x.outcome_distribution.failed + x.outcome_distribution.flaky) / x.outcome_distribution.total if x.outcome_distribution.total > 0 else 0,
                     reverse=True
                 )
-                
+
                 for test_method in sorted_methods:
                     total_runs = test_method.outcome_distribution.total
                     if total_runs > 0:
                         failure_rate = (test_method.outcome_distribution.failed + test_method.outcome_distribution.flaky) / total_runs
-                        
+
                         # Extract the method name from the full test name
                         method_name = test_method.name.split('.')[-1]
-                        
+
                         print(f"\n  → {method_name}")
                         print(f"    Failure Rate: {failure_rate:.2%}")
                         print(f"    Runs: {total_runs:3d} | Failed: {test_method.outcome_distribution.failed:3d} | "
                               f"Flaky: {test_method.outcome_distribution.flaky:3d} | "
                               f"Passed: {test_method.outcome_distribution.passed:3d}")
-                        
+
                         # Show test method timeline
                         if test_method.timeline:
                             print(f"\n    Recent Executions (last {min(3, len(test_method.timeline))} of {len(test_method.timeline)} runs):")
@@ -939,7 +936,7 @@ def main():
                             for entry in sorted(test_method.timeline, key=lambda x: x.timestamp)[-3:]:
                                 date_str = entry.timestamp.strftime('%Y-%m-%d %H:%M')
                                 print(f"    {date_str:<17} {entry.outcome:<10} {entry.build_id}")
-                
+
     except Exception as e:
         logger.exception("Error occurred during report generation")
         print(f"Error occurred: {str(e)}")

.github/scripts/junit.py

@@ -91,7 +91,10 @@ class TestSuite:
 def clean_test_name(test_name: str) -> str:
     cleaned = test_name.strip("\"").rstrip("()")
     m = method_matcher.match(cleaned)
-    return m.group(1)
+    if m is None:
+        raise ValueError(f"Could not parse test name '{test_name}'. Expected a valid Java method name.")
+    else:
+        return m.group(1)
 
 
 class TestCatalogExporter:
@@ -402,10 +405,19 @@ def split_report_path(base_path: str, report_path: str) -> Tuple[str, str]:
     print("<hr/>")
 
     # Print special message if there was a timeout
-    exit_code = get_env("GRADLE_EXIT_CODE", int)
-    if exit_code == 124:
+    test_exit_code = get_env("GRADLE_TEST_EXIT_CODE", int)
+    quarantined_test_exit_code = get_env("GRADLE_QUARANTINED_TEST_EXIT_CODE", int)
+
+    if test_exit_code == 124 or quarantined_test_exit_code == 124:
+        # Special handling for timeouts. The exit code 124 is emitted by 'timeout' command used in build.yml.
+        # A watchdog script "thread-dump.sh" will use jstack to force a thread dump for any Gradle process
+        # still running after the timeout. We capture the exit codes of the two test tasks and pass them to
+        # this script. If either "test" or "quarantinedTest" fails due to timeout, we want to fail the overall build.
         thread_dump_url = get_env("THREAD_DUMP_URL")
-        logger.debug(f"Gradle command timed out. These are partial results!")
+        if test_exit_code == 124:
+            logger.debug(f"Gradle task for 'test' timed out. These are partial results!")
+        else:
+            logger.debug(f"Gradle task for 'quarantinedTest' timed out. These are partial results!")
         logger.debug(summary)
         if thread_dump_url:
             print(f"\nThe JUnit tests were cancelled due to a timeout. Thread dumps were generated before the job was cancelled. "
@@ -414,7 +426,7 @@ def split_report_path(base_path: str, report_path: str) -> Tuple[str, str]:
         else:
             logger.debug(f"Failing this step because the tests timed out. Thread dumps were not archived, check logs in JUnit step.")
         exit(1)
-    elif exit_code in (0, 1):
+    elif test_exit_code in (0, 1):
         logger.debug(summary)
         if total_failures > 0:
             logger.debug(f"Failing this step due to {total_failures} test failures")
@@ -425,5 +437,5 @@ def split_report_path(base_path: str, report_path: str) -> Tuple[str, str]:
         else:
             exit(0)
     else:
-        logger.debug(f"Gradle had unexpected exit code {exit_code}. Failing this step")
+        logger.debug(f"Gradle had unexpected exit code {test_exit_code}. Failing this step")
         exit(1)

.github/workflows/build.yml

@@ -21,14 +21,10 @@ name: Check and Test
 on:
   workflow_call:
     inputs:
-      gradle-cache-read-only:
-        description: "Should the Gradle cache be read-only?"
+      is-trunk:
+        description: "Is this a trunk build?"
         default: true
         type: boolean
-      gradle-cache-write-only:
-        description: "Should the Gradle cache be write-only?"
-        default: false
-        type: boolean
       is-public-fork:
         description: "Is this CI run from a public fork?"
         default: true
@@ -105,9 +101,9 @@ jobs:
         uses: ./.github/actions/setup-gradle
         with:
           java-version: 23
-          gradle-cache-read-only: ${{ inputs.gradle-cache-read-only }}
-          gradle-cache-write-only: ${{ inputs.gradle-cache-write-only }}
-          develocity-access-key: ${{ secrets.GE_ACCESS_TOKEN }}
+          gradle-cache-read-only: ${{ !inputs.is-trunk }}
+          gradle-cache-write-only: ${{ inputs.is-trunk }}
+          develocity-access-key: ${{ secrets.DEVELOCITY_ACCESS_KEY }}
       - name: Compile and validate
         env:
           SCAN_ARG: ${{ inputs.is-public-fork && '--no-scan' || '--scan' }}
@@ -117,7 +113,7 @@ jobs:
         # --scan:         Publish the build scan. This will only work on PRs from apache/kafka and trunk
         # --no-scan:      For public fork PRs, we won't attempt to publish the scan
         run: |
-          ./gradlew --build-cache --info $SCAN_ARG check -x test
+          ./gradlew --build-cache --info $SCAN_ARG check siteDocTar -x test
       - name: Archive check reports
         if: always()
         uses: actions/upload-artifact@v4
@@ -137,6 +133,16 @@ jobs:
         run: python .github/scripts/rat.py
         env:
           GITHUB_WORKSPACE: ${{ github.workspace }}
+      - name: Check generated documentation
+        # Check if there are any empty files under ./site-docs/generated, If any empty files are found, print an error 
+        # message and list the empty files
+        run: |
+          tar zxvf core/build/distributions/kafka_2.13-$(./gradlew properties | grep version: | awk '{print $NF}' | head -n 1)-site-docs.tgz
+          if find ./site-docs/generated -type f -exec grep -L "." {} \; | grep -q "."; then
+            echo "One or more documentation files are empty!" >&2
+            find ./site-docs/generated -type f -exec grep -L "." {} \; >&2
+            exit 1
+          fi
 
   test:
     needs: [validate, load-catalog]
@@ -146,6 +152,8 @@ jobs:
       fail-fast: false
       matrix:
         java: [ 23, 17 ]  # If we change these, make sure to adjust ci-complete.yml
+    outputs:
+      timed-out: ${{ (steps.junit-test.outputs.gradle-exitcode == '124' || steps.junit-quarantined-test.outputs.gradle-exitcode == '124') }}
     name: JUnit tests Java ${{ matrix.java }}
     steps:
       - name: Checkout code
@@ -158,9 +166,9 @@ jobs:
         uses: ./.github/actions/setup-gradle
         with:
           java-version: ${{ matrix.java }}
-          gradle-cache-read-only: ${{ inputs.gradle-cache-read-only }}
-          gradle-cache-write-only: ${{ inputs.gradle-cache-write-only }}
-          develocity-access-key: ${{ secrets.GE_ACCESS_TOKEN }}
+          gradle-cache-read-only: ${{ !inputs.is-trunk }}
+          gradle-cache-write-only: ${{ inputs.is-trunk }}
+          develocity-access-key: ${{ secrets.DEVELOCITY_ACCESS_KEY }}
 
       # If the load-catalog job failed, we won't be able to download the artifact. Since we don't want this to fail
       # the overall workflow, so we'll continue here without a test catalog.
@@ -176,7 +184,7 @@ jobs:
         uses: ./.github/actions/run-gradle
         with:
           test-task: quarantinedTest
-          timeout-minutes: 30
+          timeout-minutes: 180
           test-catalog-path: ${{ steps.load-test-catalog.outputs.download-path }}/combined-test-catalog.txt
           build-scan-artifact-name: build-scan-quarantined-test-${{ matrix.java }}
 
@@ -225,7 +233,8 @@ jobs:
           GITHUB_WORKSPACE: ${{ github.workspace }}
           JUNIT_REPORT_URL: ${{ steps.junit-upload-artifact.outputs.artifact-url }}
           THREAD_DUMP_URL: ${{ steps.thread-dump-upload-artifact.outputs.artifact-url }}
-          GRADLE_EXIT_CODE: ${{ steps.junit-test.outputs.gradle-exitcode }}
+          GRADLE_TEST_EXIT_CODE: ${{ steps.junit-test.outputs.gradle-exitcode }}
+          GRADLE_QUARANTINED_TEST_EXIT_CODE: ${{ steps.junit-quarantined-test.outputs.gradle-exitcode }}
 
       - name: Archive Test Catalog
         if: ${{ always() && matrix.java == '23' }}
@@ -239,7 +248,7 @@ jobs:
   update-test-catalog:
     name: Update Test Catalog
     needs: test
-    if: ${{ always() && !inputs.is-public-fork }}
+    if: ${{ always() && inputs.is-trunk && needs.test.outputs.timed-out == 'false' }}
     runs-on: ubuntu-latest
     permissions:
       contents: write

.github/workflows/ci-complete.yml

@@ -30,7 +30,7 @@ run-name: Build Scans for ${{ github.event.workflow_run.display_title}}
 #
 # This "workflow_run" triggered workflow is run in a privileged context and so does have access to
 # the repository secrets. Here we can download the build scan files produced by a PR and publish
-# them to ge.apache.org.
+# them to develocity.apache.org.
 #
 # If we need to do things like comment on, label, or otherwise modify PRs from public forks. This
 # workflow is the place to do it. PR number is ${{ github.event.workflow_run.pull_requests[0].number }}
@@ -59,7 +59,7 @@ jobs:
         uses: ./.github/actions/setup-gradle
         with:
           java-version: ${{ matrix.java }}
-          develocity-access-key: ${{ secrets.GE_ACCESS_TOKEN }}
+          develocity-access-key: ${{ secrets.DEVELOCITY_ACCESS_KEY }}
       - name: Download build scan archive
         id: download-build-scan
         uses: actions/download-artifact@v4

.github/workflows/ci.yml

@@ -19,6 +19,7 @@ on:
   push:
     branches:
       - 'trunk'
+      - '4.0'
 
   schedule:
     - cron: '0 0 * * 6,0'    # Run on Saturday and Sunday at midnight UTC
@@ -27,17 +28,17 @@ on:
     types: [ opened, synchronize, ready_for_review, reopened ]
     branches:
       - 'trunk'
+      - '4.0'
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
-  cancel-in-progress: ${{ github.ref != 'refs/heads/trunk' }}
+  cancel-in-progress: ${{ startsWith(github.ref, 'refs/pull/') }}
 
 jobs:
   build:
     uses: ./.github/workflows/build.yml
     with:
-      gradle-cache-read-only: ${{ github.ref != 'refs/heads/trunk' }}
-      gradle-cache-write-only: ${{ github.ref == 'refs/heads/trunk' }}
+      is-trunk: ${{ github.ref == 'refs/heads/trunk' }}
       is-public-fork: ${{ github.event.pull_request.head.repo.fork || false }}
     secrets:
       inherit

.github/workflows/deflake.yml

@@ -50,7 +50,7 @@ jobs:
         with:
           java-version: ${{ inputs.java-version }}
           gradle-cache-read-only: true
-          develocity-access-key: ${{ secrets.GE_ACCESS_TOKEN }}
+          develocity-access-key: ${{ secrets.DEVELOCITY_ACCESS_KEY }}
       - name: Test
         timeout-minutes: 60
         id: junit-test