Merge remote-tracking branch 'origin/master' into mvp_demo

.github/workflows/test-on-comment.yaml

@@ -3,28 +3,63 @@ name: Kruize Remote monitoring functional tests
 on:
   issue_comment:
     types: [created]
+    issue_type: [pull_request]
   pull_request_review:
     types: [submitted]
 
 jobs:
   functest:
 
-    # Only run this if the user asked us to
-    if: github.event.comment.body == 'run functional tests'
+    if: contains(github.event.comment.body, 'run functional tests')
 
     runs-on: ubuntu-20.04
     steps:
-      - uses: actions/checkout@v3
+      - name: Get workflow run info
+        run: |
+          echo "url=$GITHUB_SERVER_URL/$GITHUB_REPOSITORY/actions/runs/$GITHUB_RUN_ID" >> $GITHUB_OUTPUT
+          echo "id=$GITHUB_RUN_ID" >> $GITHUB_OUTPUT
+          echo "url=$GITHUB_SERVER_URL/$GITHUB_REPOSITORY/actions/runs/$GITHUB_RUN_ID"
+          echo "id=$GITHUB_RUN_ID"
+        id: workflow_run_info
+
+      - name: Debug Comment Content
+        run: echo "Comment :${{ github.event.comment.body }}"
+
+
+      - uses: actions/github-script@v6
+        id: get-pr
+        with:
+          script: |
+            const request = {
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              pull_number: context.issue.number
+            }
+            core.info(`Getting PR #${request.pull_number} from ${request.owner}/${request.repo}`)
+            try {
+              const result = await github.rest.pulls.get(request)
+              return result.data
+            } catch (err) {
+              core.setFailed(`Request failed with error ${err}`)
+            }
+      
+      - name: Check out code
+        uses: actions/checkout@v4
+        with:
+          ref: refs/pull/${{ github.event.issue.number }}/head
+
       - name: Setup Minikube
         uses: manusa/[email protected]
         with:
           minikube version: 'v1.16.0'
           kubernetes version: 'v1.19.2'
+
       - name: Build crc
         run: |
           echo Build crc
           ./build.sh -i autotune_operator:test_build
           docker images | grep autotune
+
       - name: Install Prometheus on minikube
         run: |
           echo Install Prometheus on minikube
@@ -60,3 +95,67 @@ jobs:
            name: kruize_test_results
            path: ./kruize_test_results.tar
            retention-days: 2
+
+  reportFailure:
+    runs-on: ubuntu-latest
+    needs: [functest]
+    if: failure()
+    steps:
+    - name: Create comment
+      uses: actions/github-script@v6
+      with:
+        github-token: ${{secrets.GITHUB_TOKEN}}
+        script: |
+          comment_body = `
+          @${{ github.actor }} Build(s) failed.
+          Workflow Run ID: [${{ needs.functest.outputs.workflow_id }}](${{ needs.functest.outputs.workflow_url }})
+          `;
+          github.rest.issues.createComment({
+            issue_number: context.issue.number,
+            owner: context.repo.owner,
+            repo: context.repo.repo,
+            body: comment_body
+          })
+
+  reportCancelled:
+    runs-on: ubuntu-latest
+    needs: [functest]
+    if: cancelled()
+    steps:
+    - name: Create comment
+      uses: actions/github-script@v6
+      with:
+        github-token: ${{secrets.GITHUB_TOKEN}}
+        script: |
+          comment_body = `
+          @${{ github.actor }} Build(s) cancelled.
+          Workflow Run ID: [${{ needs.functest.outputs.workflow_id }}](${{ needs.functest.outputs.workflow_url }})
+          `;
+          github.rest.issues.createComment({
+            issue_number: context.issue.number,
+            owner: context.repo.owner,
+            repo: context.repo.repo,
+            body: comment_body
+          })
+          
+  reportSuccess:
+    runs-on: ubuntu-latest
+    needs: [functest]
+    if: success()
+    steps:
+    - name: Create comment
+      uses: actions/github-script@v6
+      with:
+        github-token: ${{secrets.GITHUB_TOKEN}}
+        script: |
+          
+          comment_body = `
+          @${{ github.actor }} Build(s) successful.
+          Workflow Run ID: [${{ needs.functest.outputs.workflow_id }}](${{ needs.functest.outputs.workflow_url }})
+          `;
+          github.rest.issues.createComment({
+            issue_number: context.issue.number,
+            owner: context.repo.owner,
+            repo: context.repo.repo,
+            body: comment_body
+          })

src/main/java/com/autotune/database/dao/ExperimentDAOImpl.java

@@ -163,8 +163,8 @@ public List<KruizeResultsEntry> addToDBAndFetchFailedResults(List<KruizeResultsE
         String statusValue = "failure";
         Timer.Sample timerAddBulkResultsDB = Timer.start(MetricsConfig.meterRegistry());
         try (Session session = KruizeHibernateUtil.getSessionFactory().openSession()) {
-            tx = session.beginTransaction();
             for (KruizeResultsEntry entry : kruizeResultsEntries) {
+                tx = session.beginTransaction();
                 try {
                     session.persist(entry);
                     session.flush();
@@ -199,16 +199,13 @@ public List<KruizeResultsEntry> addToDBAndFetchFailedResults(List<KruizeResultsE
                         entry.setErrorReasons(List.of(e.getMessage()));
                         failedResultsEntries.add(entry);
                     }
-                    tx.commit();
-                    tx = session.beginTransaction();
                 } catch (Exception e) {
                     entry.setErrorReasons(List.of(e.getMessage()));
                     failedResultsEntries.add(entry);
+                } finally {
                     tx.commit();
-                    tx = session.beginTransaction();
                 }
             }
-            tx.commit();
             statusValue = "success";
         } catch (Exception e) {
             LOGGER.error("Not able to save experiment due to {}", e.getMessage());

tests/scripts/remote_monitoring_tests/fault_tolerant_tests/kruize_pod_restart_test.py

@@ -133,7 +133,12 @@ def main(argv):
 
         # Fetch listExperiments
         list_exp_json_file_before = list_exp_json_dir + "/list_exp_json_before_" + str(i) + ".json"
-        response = list_experiments()
+        # assign params to be passed in listExp
+        results = "true"
+        recommendations = "true"
+        latest = "false"
+        experiment_name = None
+        response = list_experiments(results, recommendations, latest, experiment_name)
         if response.status_code == SUCCESS_200_STATUS_CODE:
            list_exp_json = response.json()
         else:
@@ -173,7 +178,11 @@ def main(argv):
 
         # Fetch listExperiments
         list_exp_json_file_after = list_exp_json_dir + "/list_exp_json_after_" + str(i) + ".json"
-        response = list_experiments()
+        results = "true"
+        recommendations = "true"
+        latest = "false"
+        experiment_name = None
+        response = list_experiments(results, recommendations, latest, experiment_name)
         if response.status_code == SUCCESS_200_STATUS_CODE:
             list_exp_json = response.json()
         else:

tests/scripts/remote_monitoring_tests/helpers/kruize.py

@@ -216,13 +216,25 @@ def create_performance_profile(perf_profile_json_file):
 
 # Description: This function obtains the experiments from Kruize Autotune using listExperiments API
 # Input Parameters: None
-def list_experiments():
-    PARAMS = {'results': "true", 'recommendations': "true", "latest": "false"}
+def list_experiments(results=None, recommendations=None, latest=None, experiment_name=None):
     print("\nListing the experiments...")
-    url = URL + "/listExperiments"
-    print("URL = ", url)
+    query_params = {}
 
-    response = requests.get(url=url, params=PARAMS)
+    if experiment_name is not None:
+        query_params['experiment_name'] = experiment_name
+    if latest is not None:
+        query_params['latest'] = latest
+    if results is not None:
+        query_params['results'] = results
+    if recommendations is not None:
+        query_params['recommendations'] = recommendations
 
+    query_string = "&".join(f"{key}={value}" for key, value in query_params.items())
+
+    url = URL + "/listExperiments"
+    if query_string:
+        url += "?" + query_string
+    print("URL = ", url)
+    response = requests.get(url)
     print("Response status code = ", response.status_code)
     return response

tests/scripts/remote_monitoring_tests/helpers/utils.py

@@ -25,13 +25,16 @@
 SUCCESS_200_STATUS_CODE = 200
 ERROR_STATUS_CODE = 400
 ERROR_409_STATUS_CODE = 409
+DUPLICATE_RECORDS_COUNT = 5
 
 SUCCESS_STATUS = "SUCCESS"
 ERROR_STATUS = "ERROR"
 UPDATE_RESULTS_SUCCESS_MSG = "Results added successfully! View saved results at /listExperiments."
 UPDATE_RESULTS_DATE_PRECEDE_ERROR_MSG = "The Start time should precede the End time!"
 UPDATE_RESULTS_INVALID_METRIC_VALUE_ERROR_MSG = "Performance profile: avg cannot be negative or blank for the metric variable: "
 UPDATE_RESULTS_INVALID_METRIC_FORMAT_ERROR_MSG = "Performance profile:  Format value should be among these values: [cores, MiB]"
+UPDATE_RESULTS_FAILED_RECORDS_MSG = f"Out of a total of 100 records, {DUPLICATE_RECORDS_COUNT} failed to save"
+DUPLICATE_RECORDS_MSG = "An entry for this record already exists!"
 CREATE_EXP_SUCCESS_MSG = "Experiment registered successfully with Kruize. View registered experiments at /listExperiments"
 CREATE_EXP_BULK_ERROR_MSG = "At present, the system does not support bulk entries!"
 UPDATE_RECOMMENDATIONS_MANDATORY_DEFAULT_MESSAGE = 'experiment_name is mandatory'
@@ -335,6 +338,29 @@ def validate_reco_json(create_exp_json, update_results_json, list_reco_json, exp
                                 list_reco_kubernetes_obj, expected_duration_in_hours, test_name)
 
 
+def validate_list_exp_results_count(expected_results_count, list_exp_json):
+
+    # Get the count of objects in all results arrays
+    list_exp_results_count = count_results_objects(list_exp_json)
+    print("results_count = ", expected_results_count)
+    print("list_exp_results_count = ", list_exp_results_count)
+
+    assert expected_results_count == list_exp_results_count
+
+
+# Function to count objects in results arrays
+def count_results_objects(list_exp_json):
+    count = 0
+    container_count = 1
+    for k8s_object in list_exp_json.get("kubernetes_objects", []):
+        container_count = len(k8s_object.get("containers"))
+        for container in k8s_object.get("containers", {}).values():
+            results = container.get("results", {})
+            count += len(results)
+
+    return count/container_count
+
+
 def validate_kubernetes_obj(create_exp_kubernetes_obj, update_results_kubernetes_obj, update_results_json,
                             list_reco_kubernetes_obj, expected_duration_in_hours, test_name):
     # Validate type, name, namespace