feat(lifecycle-operator): introduce a possibility to configure number…

… of retries and interval for KeptnEvaluationDefinition (#3141)

Signed-off-by: odubajDT <ondrej.dubaj@dynatrace.com>

.github/scripts/.helm-tests/default/result.yaml

@@ -2890,6 +2890,19 @@ spec:
                   - keptnMetricRef
                   type: object
                 type: array
+              retries:
+                default: 10
+                description: |-
+                  Retries indicates how many times the KeptnEvaluation can be attempted in the case of an error or
+                  missed evaluation objective, before considering the KeptnEvaluation to be failed.
+                type: integer
+              retryInterval:
+                default: 5s
+                description: |-
+                  RetryInterval specifies the interval at which the KeptnEvaluation is retried in the case of an error
+                  or a missed objective.
+                pattern: ^0|([0-9]+(\.[0-9]+)?(ns|us|µs|ms|s|m|h))+$
+                type: string
             required:
             - objectives
             type: object

.github/scripts/.helm-tests/lifecycle-only/result.yaml

@@ -2836,6 +2836,19 @@ spec:
                   - keptnMetricRef
                   type: object
                 type: array
+              retries:
+                default: 10
+                description: |-
+                  Retries indicates how many times the KeptnEvaluation can be attempted in the case of an error or
+                  missed evaluation objective, before considering the KeptnEvaluation to be failed.
+                type: integer
+              retryInterval:
+                default: 5s
+                description: |-
+                  RetryInterval specifies the interval at which the KeptnEvaluation is retried in the case of an error
+                  or a missed objective.
+                pattern: ^0|([0-9]+(\.[0-9]+)?(ns|us|µs|ms|s|m|h))+$
+                type: string
             required:
             - objectives
             type: object

.github/scripts/.helm-tests/lifecycle-with-certs/result.yaml

@@ -2867,6 +2867,19 @@ spec:
                   - keptnMetricRef
                   type: object
                 type: array
+              retries:
+                default: 10
+                description: |-
+                  Retries indicates how many times the KeptnEvaluation can be attempted in the case of an error or
+                  missed evaluation objective, before considering the KeptnEvaluation to be failed.
+                type: integer
+              retryInterval:
+                default: 5s
+                description: |-
+                  RetryInterval specifies the interval at which the KeptnEvaluation is retried in the case of an error
+                  or a missed objective.
+                pattern: ^0|([0-9]+(\.[0-9]+)?(ns|us|µs|ms|s|m|h))+$
+                type: string
             required:
             - objectives
             type: object

docs/docs/reference/api-reference/lifecycle/v1beta1/index.md

@@ -110,6 +110,23 @@ _Appears in:_
 | `message` _string_ | Message contains additional information about the evaluation of an objective. This can include explanations about why an evaluation has failed (e.g. due to a missed objective), or if there was any error during the evaluation of the objective. || ✓ |
 
 
+#### FailureConditions
+
+
+
+FailureConditions represent the failure conditions (number of retries and retry interval)
+for the evaluation to be considered as failed
+
+_Appears in:_
+- [KeptnEvaluationDefinitionSpec](#keptnevaluationdefinitionspec)
+- [KeptnEvaluationSpec](#keptnevaluationspec)
+
+| Field | Description | Default | Optional |
+| --- | --- | --- | --- |
+| `retries` _integer_ | Retries indicates how many times the KeptnEvaluation can be attempted in the case of an error or missed evaluation objective, before considering the KeptnEvaluation to be failed. |10| ✓ |
+| `retryInterval` _[Duration](https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.28/#duration-v1-meta)_ | RetryInterval specifies the interval at which the KeptnEvaluation is retried in the case of an error or a missed objective. |5s| ✓ |
+
+
 #### FunctionReference
 
 
@@ -521,6 +538,8 @@ _Appears in:_
 | Field | Description | Default | Optional |
 | --- | --- | --- | --- |
 | `objectives` _[Objective](#objective) array_ | Objectives is a list of objectives that have to be met for a KeptnEvaluation referencing this KeptnEvaluationDefinition to be successful. || x |
+| `retries` _integer_ | Retries indicates how many times the KeptnEvaluation can be attempted in the case of an error or missed evaluation objective, before considering the KeptnEvaluation to be failed. |10| ✓ |
+| `retryInterval` _[Duration](https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.28/#duration-v1-meta)_ | RetryInterval specifies the interval at which the KeptnEvaluation is retried in the case of an error or a missed objective. |5s| ✓ |
 
 
 #### KeptnEvaluationList
@@ -555,9 +574,9 @@ _Appears in:_
 | `appName` _string_ | AppName defines the KeptnApp for which the KeptnEvaluation is done. || ✓ |
 | `appVersion` _string_ | AppVersion defines the version of the KeptnApp for which the KeptnEvaluation is done. || ✓ |
 | `evaluationDefinition` _string_ | EvaluationDefinition refers to the name of the KeptnEvaluationDefinition which includes the objectives for the KeptnEvaluation. The KeptnEvaluationDefinition can be located in the same namespace as the KeptnEvaluation, or in the Keptn namespace. || x |
+| `checkType` _string_ | Type indicates whether the KeptnEvaluation is part of the pre- or postDeployment phase. || ✓ |
 | `retries` _integer_ | Retries indicates how many times the KeptnEvaluation can be attempted in the case of an error or missed evaluation objective, before considering the KeptnEvaluation to be failed. |10| ✓ |
 | `retryInterval` _[Duration](https://kubernetes.io/docs/reference/generated/kubernetes-api/v1.28/#duration-v1-meta)_ | RetryInterval specifies the interval at which the KeptnEvaluation is retried in the case of an error or a missed objective. |5s| ✓ |
-| `checkType` _string_ | Type indicates whether the KeptnEvaluation is part of the pre- or postDeployment phase. || ✓ |
 
 
 #### KeptnEvaluationStatus

docs/docs/reference/crd-reference/evaluationdefinition.md

@@ -18,6 +18,8 @@ kind: KeptnEvaluationDefinition
 metadata:
   name: <evaluation-name>
 spec:
+  retries: <number-of-retries>
+  retryInterval: <duration>
   objectives:
     - evaluationTarget: "<value>"
       keptnMetricRef:
@@ -56,6 +58,16 @@ spec:
           This is used to define success or failure criteria for the referenced `KeptnMetric` in order to pass or fail
           the pre- and post-evaluation stages
 
+    * **retries** -- specifies the number of times
+      an `Keptnevaluation` defined by the `KeptnEvaluationDefinition`
+      should be restarted if an attempt is unsuccessful.
+      The default value is `10`.
+    * **retryInterval** -- specifies the time
+      to wait between the retries.
+      The value supplied should specify the unit of measurement;
+      for example, `5s` indicates 5 seconds and `5m` indicates 5 minutes.
+      The default value is `5s`.
+
 ## Usage
 
 A `KeptnEvaluationDefinition` references one or more [KeptnMetric](metric.md) resources.
@@ -78,6 +90,8 @@ metadata:
   name: my-prometheus-evaluation
   namespace: example
 spec:
+  retries: 5
+  retryInterval: 10s
   source: prometheus
   objectives:
     - keptnMetricRef:

lifecycle-operator/apis/lifecycle/v1beta1/keptnappversion_types.go

@@ -428,8 +428,9 @@ func (a KeptnAppVersion) GenerateEvaluation(evaluationDefinition KeptnEvaluation
 			AppName:              a.Spec.AppName,
 			EvaluationDefinition: evaluationDefinition.Name,
 			Type:                 checkType,
-			RetryInterval: metav1.Duration{
-				Duration: 5 * time.Second,
+			FailureConditions: FailureConditions{
+				RetryInterval: evaluationDefinition.Spec.FailureConditions.RetryInterval,
+				Retries:       evaluationDefinition.Spec.FailureConditions.Retries,
 			},
 		},
 	}

lifecycle-operator/apis/lifecycle/v1beta1/keptnappversion_types_test.go

@@ -272,14 +272,25 @@ func TestKeptnAppVersion(t *testing.T) {
 		ObjectMeta: v1.ObjectMeta{
 			Name: "eval-def",
 		},
+		Spec: KeptnEvaluationDefinitionSpec{
+			FailureConditions: FailureConditions{
+				RetryInterval: v1.Duration{
+					Duration: 5 * time.Second,
+				},
+				Retries: 5,
+			},
+		},
 	}, common.PostDeploymentCheckType)
 	require.Equal(t, KeptnEvaluationSpec{
 		AppVersion:           app.GetVersion(),
 		AppName:              app.GetParentName(),
 		EvaluationDefinition: "eval-def",
 		Type:                 common.PostDeploymentCheckType,
-		RetryInterval: v1.Duration{
-			Duration: 5 * time.Second,
+		FailureConditions: FailureConditions{
+			RetryInterval: v1.Duration{
+				Duration: 5 * time.Second,
+			},
+			Retries: 5,
 		},
 	}, evaluation.Spec)
 

lifecycle-operator/apis/lifecycle/v1beta1/keptnevaluation_types.go

@@ -45,22 +45,12 @@ type KeptnEvaluationSpec struct {
 	// The KeptnEvaluationDefinition can be
 	// located in the same namespace as the KeptnEvaluation, or in the Keptn namespace.
 	EvaluationDefinition string `json:"evaluationDefinition"`
-	// Retries indicates how many times the KeptnEvaluation can be attempted in the case of an error or
-	// missed evaluation objective, before considering the KeptnEvaluation to be failed.
-	// +kubebuilder:default:=10
-	// +optional
-	Retries int `json:"retries,omitempty"`
-	// RetryInterval specifies the interval at which the KeptnEvaluation is retried in the case of an error
-	// or a missed objective.
-	// +optional
-	// +kubebuilder:default:="5s"
-	// +kubebuilder:validation:Pattern="^0|([0-9]+(\\.[0-9]+)?(ns|us|µs|ms|s|m|h))+$"
-	// +kubebuilder:validation:Type:=string
-	// +optional
-	RetryInterval metav1.Duration `json:"retryInterval,omitempty"`
 	// Type indicates whether the KeptnEvaluation is part of the pre- or postDeployment phase.
 	// +optional
 	Type common.CheckType `json:"checkType,omitempty"`
+	// FailureConditions represent the failure conditions (number of retries and retry interval)
+	// for the evaluation to be considered as failed
+	FailureConditions `json:",inline"`
 }
 
 // KeptnEvaluationStatus defines the observed state of KeptnEvaluation

lifecycle-operator/apis/lifecycle/v1beta1/keptnevaluationdefinition_types.go

@@ -25,6 +25,26 @@ type KeptnEvaluationDefinitionSpec struct {
 	// Objectives is a list of objectives that have to be met for a KeptnEvaluation referencing this
 	// KeptnEvaluationDefinition to be successful.
 	Objectives []Objective `json:"objectives"`
+	// FailureConditions represent the failure conditions (number of retries and retry interval)
+	// for the evaluation to be considered as failed
+	FailureConditions `json:",inline"`
+}
+
+// FailureConditions represent the failure conditions (number of retries and retry interval)
+// for the evaluation to be considered as failed
+type FailureConditions struct {
+	// Retries indicates how many times the KeptnEvaluation can be attempted in the case of an error or
+	// missed evaluation objective, before considering the KeptnEvaluation to be failed.
+	// +kubebuilder:default:=10
+	// +optional
+	Retries int `json:"retries,omitempty"`
+	// RetryInterval specifies the interval at which the KeptnEvaluation is retried in the case of an error
+	// or a missed objective.
+	// +kubebuilder:default:="5s"
+	// +kubebuilder:validation:Pattern="^0|([0-9]+(\\.[0-9]+)?(ns|us|µs|ms|s|m|h))+$"
+	// +kubebuilder:validation:Type:=string
+	// +optional
+	RetryInterval metav1.Duration `json:"retryInterval,omitempty"`
 }
 
 type Objective struct {

lifecycle-operator/apis/lifecycle/v1beta1/keptntask_types.go

@@ -55,7 +55,6 @@ type KeptnTaskSpec struct {
 	// Timeout specifies the maximum time to wait for the task to be completed successfully.
 	// If the task does not complete successfully within this time frame, it will be
 	// considered to be failed.
-	// +optional
 	// +kubebuilder:default:="5m"
 	// +kubebuilder:validation:Pattern="^0|([0-9]+(\\.[0-9]+)?(ns|us|µs|ms|s|m|h))+$"
 	// +kubebuilder:validation:Type:=string

lifecycle-operator/apis/lifecycle/v1beta1/keptnworkloadversion_types.go

@@ -429,8 +429,9 @@ func (w KeptnWorkloadVersion) GenerateEvaluation(evaluationDefinition KeptnEvalu
 			Workload:             w.GetParentName(),
 			EvaluationDefinition: evaluationDefinition.Name,
 			Type:                 checkType,
-			RetryInterval: metav1.Duration{
-				Duration: 5 * time.Second,
+			FailureConditions: FailureConditions{
+				RetryInterval: evaluationDefinition.Spec.RetryInterval,
+				Retries:       evaluationDefinition.Spec.Retries,
 			},
 		},
 	}

lifecycle-operator/apis/lifecycle/v1beta1/keptnworkloadversion_types_test.go

@@ -251,15 +251,26 @@ func TestKeptnWorkloadVersion(t *testing.T) {
 		ObjectMeta: v1.ObjectMeta{
 			Name: "eval-def",
 		},
+		Spec: KeptnEvaluationDefinitionSpec{
+			FailureConditions: FailureConditions{
+				RetryInterval: v1.Duration{
+					Duration: 5 * time.Second,
+				},
+				Retries: 5,
+			},
+		},
 	}, common.PostDeploymentCheckType)
 	require.Equal(t, KeptnEvaluationSpec{
 		AppName:              workload.GetAppName(),
 		WorkloadVersion:      workload.GetVersion(),
 		Workload:             workload.GetParentName(),
 		EvaluationDefinition: "eval-def",
 		Type:                 common.PostDeploymentCheckType,
-		RetryInterval: v1.Duration{
-			Duration: 5 * time.Second,
+		FailureConditions: FailureConditions{
+			RetryInterval: v1.Duration{
+				Duration: 5 * time.Second,
+			},
+			Retries: 5,
 		},
 	}, evaluation.Spec)
 

lifecycle-operator/chart/templates/keptnevaluationdefinition-crd.yaml

@@ -259,6 +259,19 @@ spec:
                   - keptnMetricRef
                   type: object
                 type: array
+              retries:
+                default: 10
+                description: |-
+                  Retries indicates how many times the KeptnEvaluation can be attempted in the case of an error or
+                  missed evaluation objective, before considering the KeptnEvaluation to be failed.
+                type: integer
+              retryInterval:
+                default: 5s
+                description: |-
+                  RetryInterval specifies the interval at which the KeptnEvaluation is retried in the case of an error
+                  or a missed objective.
+                pattern: ^0|([0-9]+(\.[0-9]+)?(ns|us|µs|ms|s|m|h))+$
+                type: string
             required:
             - objectives
             type: object

lifecycle-operator/config/crd/bases/lifecycle.keptn.sh_keptnevaluationdefinitions.yaml

@@ -251,6 +251,19 @@ spec:
                   - keptnMetricRef
                   type: object
                 type: array
+              retries:
+                default: 10
+                description: |-
+                  Retries indicates how many times the KeptnEvaluation can be attempted in the case of an error or
+                  missed evaluation objective, before considering the KeptnEvaluation to be failed.
+                type: integer
+              retryInterval:
+                default: 5s
+                description: |-
+                  RetryInterval specifies the interval at which the KeptnEvaluation is retried in the case of an error
+                  or a missed objective.
+                pattern: ^0|([0-9]+(\.[0-9]+)?(ns|us|µs|ms|s|m|h))+$
+                type: string
             required:
             - objectives
             type: object

lifecycle-operator/config/samples/lifecycle_v1beta1_keptnevaluationdefinition.yaml

@@ -4,6 +4,8 @@ metadata:
   name: keptnevaluationdefinition-sample
 spec:
   source: prometheus
+  retries: 5
+  retryInterval: 10s
   objectives:
     - name: prometheus
       query: "sum(prometheus_engine_query_duration_seconds_count)"