feat: Support AnalysisRunMetadata and Dryrun for experiments via Rollout

Signed-off-by: Guillaume Doussin <[email protected]>

experiments/analysisrun_test.go

@@ -734,3 +734,118 @@ func TestTerminateAnalysisRuns(t *testing.T) {
 	patchedAr := f.getPatchedAnalysisRunAsObj(arPatchIdx)
 	assert.True(t, patchedAr.Spec.Terminate)
 }
+
+// TestCreateAnalysisRunWithMetadataAndDryRun ensures we create the AnalysisRun with the appropriate labels, annotations, and dry-run options when provided in the experiment
+func TestCreateAnalysisRunWithMetadataAndDryRun(t *testing.T) {
+	templates := generateTemplates("bar")
+	aTemplates := generateAnalysisTemplates("success-rate")
+	e := newExperiment("foo", templates, "")
+	e.Spec.Analyses = []v1alpha1.ExperimentAnalysisTemplateRef{
+		{
+			Name:         "success-rate",
+			TemplateName: aTemplates[0].Name,
+		},
+	}
+	e.Status.Phase = v1alpha1.AnalysisPhaseRunning
+	e.Status.AvailableAt = now()
+	e.Spec.AnalysisRunMetadata = v1alpha1.AnalysisRunMetadata{
+		Labels: map[string]string{
+			"foo":  "bar",
+			"foo2": "bar2",
+		},
+		Annotations: map[string]string{
+			"bar":  "foo",
+			"bar2": "foo2",
+		},
+	}
+	e.Spec.DryRun = []v1alpha1.DryRun{
+		{
+			MetricName: "someMetric",
+		},
+		{
+			MetricName: "someOtherMetric",
+		},
+	}
+	rs := templateToRS(e, templates[0], 1)
+	ar := analysisTemplateToRun("success-rate", e, &aTemplates[0].Spec)
+
+	f := newFixture(t, e, rs, &aTemplates[0])
+	defer f.Close()
+
+	analysisRunIdx := f.expectCreateAnalysisRunAction(ar)
+	patchIdx := f.expectPatchExperimentAction(e)
+	f.run(getKey(e, t))
+
+	patchedEx := f.getPatchedExperimentAsObj(patchIdx)
+	assert.Equal(t, v1alpha1.AnalysisPhasePending, patchedEx.Status.AnalysisRuns[0].Phase)
+
+	analysisRun := f.getCreatedAnalysisRun(analysisRunIdx)
+	assert.Len(t, analysisRun.ObjectMeta.Labels, 2)
+	assert.Equal(t, analysisRun.ObjectMeta.Labels["foo"], "bar")
+	assert.Equal(t, analysisRun.ObjectMeta.Labels["foo2"], "bar2")
+	assert.Len(t, analysisRun.ObjectMeta.Annotations, 2)
+	assert.Equal(t, analysisRun.ObjectMeta.Annotations["bar"], "foo")
+	assert.Equal(t, analysisRun.ObjectMeta.Annotations["bar2"], "foo2")
+
+	assert.Len(t, analysisRun.Spec.DryRun, 2)
+	assert.Equal(t, analysisRun.Spec.DryRun[0].MetricName, "someMetric")
+	assert.Equal(t, analysisRun.Spec.DryRun[1].MetricName, "someOtherMetric")
+}
+
+// TestCreateAnalysisRunWithMetadataAndDryRunWithClusterScope tests the same thing as TestCreateAnalysisRunWithMetadataAndDryRun, with a cluster scope analysis template
+func TestCreateAnalysisRunWithMetadataAndDryRunWithClusterScope(t *testing.T) {
+	templates := generateTemplates("bar")
+	aTemplates := generateClusterAnalysisTemplates("success-rate")
+	e := newExperiment("foo", templates, "")
+	e.Spec.Analyses = []v1alpha1.ExperimentAnalysisTemplateRef{
+		{
+			Name:         "success-rate",
+			TemplateName: aTemplates[0].Name,
+			ClusterScope: true,
+		},
+	}
+	e.Status.Phase = v1alpha1.AnalysisPhaseRunning
+	e.Status.AvailableAt = now()
+	e.Spec.AnalysisRunMetadata = v1alpha1.AnalysisRunMetadata{
+		Labels: map[string]string{
+			"foo":  "bar",
+			"foo2": "bar2",
+		},
+		Annotations: map[string]string{
+			"bar":  "foo",
+			"bar2": "foo2",
+		},
+	}
+	e.Spec.DryRun = []v1alpha1.DryRun{
+		{
+			MetricName: "someMetric",
+		},
+		{
+			MetricName: "someOtherMetric",
+		},
+	}
+	rs := templateToRS(e, templates[0], 1)
+	ar := analysisTemplateToRun("success-rate", e, &aTemplates[0].Spec)
+
+	f := newFixture(t, e, rs, &aTemplates[0])
+	defer f.Close()
+
+	analysisRunIdx := f.expectCreateAnalysisRunAction(ar)
+	patchIdx := f.expectPatchExperimentAction(e)
+	f.run(getKey(e, t))
+
+	patchedEx := f.getPatchedExperimentAsObj(patchIdx)
+	assert.Equal(t, v1alpha1.AnalysisPhasePending, patchedEx.Status.AnalysisRuns[0].Phase)
+
+	analysisRun := f.getCreatedAnalysisRun(analysisRunIdx)
+	assert.Len(t, analysisRun.ObjectMeta.Labels, 2)
+	assert.Equal(t, analysisRun.ObjectMeta.Labels["foo"], "bar")
+	assert.Equal(t, analysisRun.ObjectMeta.Labels["foo2"], "bar2")
+	assert.Len(t, analysisRun.ObjectMeta.Annotations, 2)
+	assert.Equal(t, analysisRun.ObjectMeta.Annotations["bar"], "foo")
+	assert.Equal(t, analysisRun.ObjectMeta.Annotations["bar2"], "foo2")
+
+	assert.Len(t, analysisRun.Spec.DryRun, 2)
+	assert.Equal(t, analysisRun.Spec.DryRun[0].MetricName, "someMetric")
+	assert.Equal(t, analysisRun.Spec.DryRun[1].MetricName, "someOtherMetric")
+}

experiments/experiment.go

@@ -101,7 +101,7 @@ func (ec *experimentContext) reconcile() *v1alpha1.ExperimentStatus {
 	}
 
 	for _, analysis := range ec.ex.Spec.Analyses {
-		ec.reconcileAnalysisRun(analysis, ec.ex.Spec.DryRun, ec.ex.Spec.MeasurementRetention)
+		ec.reconcileAnalysisRun(analysis, ec.ex.Spec.DryRun, ec.ex.Spec.MeasurementRetention, &ec.ex.Spec.AnalysisRunMetadata)
 	}
 
 	newStatus := ec.calculateStatus()
@@ -390,7 +390,7 @@ func calculateEnqueueDuration(ex *v1alpha1.Experiment, newStatus *v1alpha1.Exper
 
 // reconcileAnalysisRun reconciles a single analysis run, creating or terminating it as necessary.
 // Updates the analysis run statuses, which may subsequently fail the experiment.
-func (ec *experimentContext) reconcileAnalysisRun(analysis v1alpha1.ExperimentAnalysisTemplateRef, dryRunMetrics []v1alpha1.DryRun, measurementRetentionMetrics []v1alpha1.MeasurementRetention) {
+func (ec *experimentContext) reconcileAnalysisRun(analysis v1alpha1.ExperimentAnalysisTemplateRef, dryRunMetrics []v1alpha1.DryRun, measurementRetentionMetrics []v1alpha1.MeasurementRetention, analysisRunMetadata *v1alpha1.AnalysisRunMetadata) {
 	logCtx := ec.log.WithField("analysis", analysis.Name)
 	logCtx.Infof("Reconciling analysis")
 	prevStatus := experimentutil.GetAnalysisRunStatus(ec.ex.Status, analysis.Name)
@@ -446,7 +446,7 @@ func (ec *experimentContext) reconcileAnalysisRun(analysis v1alpha1.ExperimentAn
 			logCtx.Warnf("Skipping AnalysisRun creation for analysis %s: experiment is terminating", analysis.Name)
 			return
 		}
-		run, err := ec.createAnalysisRun(analysis, dryRunMetrics, measurementRetentionMetrics)
+		run, err := ec.createAnalysisRun(analysis, dryRunMetrics, measurementRetentionMetrics, analysisRunMetadata)
 		if err != nil {
 			msg := fmt.Sprintf("Failed to create AnalysisRun for analysis '%s': %v", analysis.Name, err.Error())
 			newStatus.Phase = v1alpha1.AnalysisPhaseError
@@ -493,13 +493,13 @@ func (ec *experimentContext) reconcileAnalysisRun(analysis v1alpha1.ExperimentAn
 // createAnalysisRun creates the analysis run. If an existing runs exists with same name, is
 // semantically equal, and is not complete, returns the existing one, otherwise creates a new
 // run with a collision counter increase.
-func (ec *experimentContext) createAnalysisRun(analysis v1alpha1.ExperimentAnalysisTemplateRef, dryRunMetrics []v1alpha1.DryRun, measurementRetentionMetrics []v1alpha1.MeasurementRetention) (*v1alpha1.AnalysisRun, error) {
+func (ec *experimentContext) createAnalysisRun(analysis v1alpha1.ExperimentAnalysisTemplateRef, dryRunMetrics []v1alpha1.DryRun, measurementRetentionMetrics []v1alpha1.MeasurementRetention, analysisRunMetadata *v1alpha1.AnalysisRunMetadata) (*v1alpha1.AnalysisRun, error) {
 	analysisRunIf := ec.argoProjClientset.ArgoprojV1alpha1().AnalysisRuns(ec.ex.Namespace)
 	args, err := ec.ResolveAnalysisRunArgs(analysis.Args)
 	if err != nil {
 		return nil, err
 	}
-	run, err := ec.newAnalysisRun(analysis, args, dryRunMetrics, measurementRetentionMetrics)
+	run, err := ec.newAnalysisRun(analysis, args, dryRunMetrics, measurementRetentionMetrics, analysisRunMetadata)
 	if err != nil {
 		return nil, err
 	}
@@ -635,7 +635,7 @@ func (ec *experimentContext) assessAnalysisRuns() (v1alpha1.AnalysisPhase, strin
 }
 
 // newAnalysisRun generates an AnalysisRun from the experiment and template
-func (ec *experimentContext) newAnalysisRun(analysis v1alpha1.ExperimentAnalysisTemplateRef, args []v1alpha1.Argument, dryRunMetrics []v1alpha1.DryRun, measurementRetentionMetrics []v1alpha1.MeasurementRetention) (*v1alpha1.AnalysisRun, error) {
+func (ec *experimentContext) newAnalysisRun(analysis v1alpha1.ExperimentAnalysisTemplateRef, args []v1alpha1.Argument, dryRunMetrics []v1alpha1.DryRun, measurementRetentionMetrics []v1alpha1.MeasurementRetention, analysisRunMetadata *v1alpha1.AnalysisRunMetadata) (*v1alpha1.AnalysisRun, error) {
 
 	if analysis.ClusterScope {
 		clusterTemplate, err := ec.clusterAnalysisTemplateLister.Get(analysis.TemplateName)
@@ -645,14 +645,26 @@ func (ec *experimentContext) newAnalysisRun(analysis v1alpha1.ExperimentAnalysis
 		name := fmt.Sprintf("%s-%s", ec.ex.Name, analysis.Name)
 
 		clusterAnalysisTemplates := []*v1alpha1.ClusterAnalysisTemplate{clusterTemplate}
-		run, err := analysisutil.NewAnalysisRunFromTemplates(nil, clusterAnalysisTemplates, args, dryRunMetrics, measurementRetentionMetrics, name, "", ec.ex.Namespace)
-		if err != nil {
-			return nil, err
-		}
+		runLabels := map[string]string{}
+		runAnnotations := map[string]string{}
+
 		instanceID := analysisutil.GetInstanceID(ec.ex)
 		if instanceID != "" {
-			run.Labels = map[string]string{v1alpha1.LabelKeyControllerInstanceID: ec.ex.Labels[v1alpha1.LabelKeyControllerInstanceID]}
+			runLabels[v1alpha1.LabelKeyControllerInstanceID] = ec.ex.Labels[v1alpha1.LabelKeyControllerInstanceID]
 		}
+		if analysisRunMetadata != nil {
+			for k, v := range analysisRunMetadata.Labels {
+				runLabels[k] = v
+			}
+			for k, v := range analysisRunMetadata.Annotations {
+				runAnnotations[k] = v
+			}
+		}
+		run, err := analysisutil.NewAnalysisRunFromTemplates(nil, clusterAnalysisTemplates, args, dryRunMetrics, measurementRetentionMetrics, runLabels, runAnnotations, name, "", ec.ex.Namespace)
+		if err != nil {
+			return nil, err
+		}
+
 		run.OwnerReferences = []metav1.OwnerReference{*metav1.NewControllerRef(ec.ex, controllerKind)}
 		return run, nil
 	} else {
@@ -663,14 +675,26 @@ func (ec *experimentContext) newAnalysisRun(analysis v1alpha1.ExperimentAnalysis
 		name := fmt.Sprintf("%s-%s", ec.ex.Name, analysis.Name)
 
 		analysisTemplates := []*v1alpha1.AnalysisTemplate{template}
-		run, err := analysisutil.NewAnalysisRunFromTemplates(analysisTemplates, nil, args, dryRunMetrics, measurementRetentionMetrics, name, "", ec.ex.Namespace)
-		if err != nil {
-			return nil, err
-		}
+		runLabels := map[string]string{}
+		runAnnotations := map[string]string{}
 		instanceID := analysisutil.GetInstanceID(ec.ex)
 		if instanceID != "" {
-			run.Labels = map[string]string{v1alpha1.LabelKeyControllerInstanceID: ec.ex.Labels[v1alpha1.LabelKeyControllerInstanceID]}
+			runLabels[v1alpha1.LabelKeyControllerInstanceID] = ec.ex.Labels[v1alpha1.LabelKeyControllerInstanceID]
+		}
+		if analysisRunMetadata != nil {
+			for k, v := range analysisRunMetadata.Labels {
+				runLabels[k] = v
+			}
+			for k, v := range analysisRunMetadata.Annotations {
+				runAnnotations[k] = v
+			}
 		}
+
+		run, err := analysisutil.NewAnalysisRunFromTemplates(analysisTemplates, nil, args, dryRunMetrics, measurementRetentionMetrics, runLabels, runAnnotations, name, "", ec.ex.Namespace)
+		if err != nil {
+			return nil, err
+		}
+
 		run.OwnerReferences = []metav1.OwnerReference{*metav1.NewControllerRef(ec.ex, controllerKind)}
 		return run, nil
 	}

manifests/crds/experiment-crd.yaml

@@ -84,6 +84,17 @@ spec:
                   - templateName
                   type: object
                 type: array
+              analysisRunMetadata:
+                properties:
+                  annotations:
+                    additionalProperties:
+                      type: string
+                    type: object
+                  labels:
+                    additionalProperties:
+                      type: string
+                    type: object
+                type: object
               dryRun:
                 items:
                   properties:

manifests/crds/rollout-crd.yaml

@@ -571,6 +571,26 @@ spec:
                                     - templateName
                                     type: object
                                   type: array
+                                analysisRunMetadata:
+                                  properties:
+                                    annotations:
+                                      additionalProperties:
+                                        type: string
+                                      type: object
+                                    labels:
+                                      additionalProperties:
+                                        type: string
+                                      type: object
+                                  type: object
+                                dryRun:
+                                  items:
+                                    properties:
+                                      metricName:
+                                        type: string
+                                    required:
+                                    - metricName
+                                    type: object
+                                  type: array
                                 duration:
                                   type: string
                                 templates:

manifests/install.yaml

@@ -9263,6 +9263,17 @@ spec:
                   - templateName
                   type: object
                 type: array
+              analysisRunMetadata:
+                properties:
+                  annotations:
+                    additionalProperties:
+                      type: string
+                    type: object
+                  labels:
+                    additionalProperties:
+                      type: string
+                    type: object
+                type: object
               dryRun:
                 items:
                   properties:
@@ -12465,6 +12476,26 @@ spec:
                                     - templateName
                                     type: object
                                   type: array
+                                analysisRunMetadata:
+                                  properties:
+                                    annotations:
+                                      additionalProperties:
+                                        type: string
+                                      type: object
+                                    labels:
+                                      additionalProperties:
+                                        type: string
+                                      type: object
+                                  type: object
+                                dryRun:
+                                  items:
+                                    properties:
+                                      metricName:
+                                        type: string
+                                    required:
+                                    - metricName
+                                    type: object
+                                  type: array
                                 duration:
                                   type: string
                                 templates:

pkg/apiclient/rollout/rollout.swagger.json

@@ -1955,6 +1955,17 @@
             "$ref": "#/definitions/github.com.argoproj.argo_rollouts.pkg.apis.rollouts.v1alpha1.RolloutExperimentStepAnalysisTemplateRef"
           },
           "title": "Analyses reference which analysis templates to run with the experiment\n+patchMergeKey=name\n+patchStrategy=merge"
+        },
+        "dryRun": {
+          "type": "array",
+          "items": {
+            "$ref": "#/definitions/github.com.argoproj.argo_rollouts.pkg.apis.rollouts.v1alpha1.DryRun"
+          },
+          "title": "DryRun object contains the settings for running the analysis in Dry-Run mode\n+patchMergeKey=metricName\n+patchStrategy=merge\n+optional"
+        },
+        "analysisRunMetadata": {
+          "$ref": "#/definitions/github.com.argoproj.argo_rollouts.pkg.apis.rollouts.v1alpha1.AnalysisRunMetadata",
+          "title": "AnalysisRunMetadata labels and annotations that will be added to the AnalysisRuns\n+optional"
         }
       },
       "title": "RolloutExperimentStep defines a template that is used to create a experiment for a step"

pkg/apis/api-rules/violation_exceptions.list

@@ -36,6 +36,7 @@ API rule violation: list_type_missing,github.com/argoproj/argo-rollouts/pkg/apis
 API rule violation: list_type_missing,github.com/argoproj/argo-rollouts/pkg/apis/rollouts/v1alpha1,RolloutAnalysis,MeasurementRetention
 API rule violation: list_type_missing,github.com/argoproj/argo-rollouts/pkg/apis/rollouts/v1alpha1,RolloutAnalysis,Templates
 API rule violation: list_type_missing,github.com/argoproj/argo-rollouts/pkg/apis/rollouts/v1alpha1,RolloutExperimentStep,Analyses
+API rule violation: list_type_missing,github.com/argoproj/argo-rollouts/pkg/apis/rollouts/v1alpha1,RolloutExperimentStep,DryRun
 API rule violation: list_type_missing,github.com/argoproj/argo-rollouts/pkg/apis/rollouts/v1alpha1,RolloutExperimentStep,Templates
 API rule violation: list_type_missing,github.com/argoproj/argo-rollouts/pkg/apis/rollouts/v1alpha1,RolloutExperimentStepAnalysisTemplateRef,Args
 API rule violation: list_type_missing,github.com/argoproj/argo-rollouts/pkg/apis/rollouts/v1alpha1,RolloutStatus,ALBs

pkg/apis/rollouts/v1alpha1/experiment_types.go

@@ -65,6 +65,9 @@ type ExperimentSpec struct {
 	// +patchStrategy=merge
 	// +optional
 	MeasurementRetention []MeasurementRetention `json:"measurementRetention,omitempty" patchStrategy:"merge" patchMergeKey:"metricName" protobuf:"bytes,8,rep,name=measurementRetention"`
+	// AnalysisRunMetadata labels and annotations that will be added to the AnalysisRuns
+	// +optional
+	AnalysisRunMetadata AnalysisRunMetadata `json:"analysisRunMetadata,omitempty" protobuf:"bytes,9,opt,name=analysisRunMetadata"`
 }
 
 type TemplateSpec struct {