feat: adding startup and readiness probe

caraml-dev · Aug 20, 2024 · 24f3114 · 24f3114
1 parent 7c30d7c
commit 24f3114
Show file tree

Hide file tree

Showing 5 changed files with 70 additions and 18 deletions.
diff --git a/.gitignore b/.gitignore
@@ -31,3 +31,5 @@ override.tf.json
 
 # Ignore inner charts dir
 charts/**/charts/*.tgz
+
+.idea/
diff --git a/charts/caraml-store/Chart.yaml b/charts/caraml-store/Chart.yaml
@@ -20,4 +20,4 @@ maintainers:
 - email: caraml-dev@caraml.dev
   name: caraml-dev
 name: caraml-store
-version: 0.1.17
+version: 0.1.18
diff --git a/charts/caraml-store/README.md b/charts/caraml-store/README.md
@@ -1,6 +1,6 @@
 # caraml-store
 
-![Version: 0.1.17](https://img.shields.io/badge/Version-0.1.17-informational?style=flat-square) ![AppVersion: 0.1.3](https://img.shields.io/badge/AppVersion-0.1.3-informational?style=flat-square)
+![Version: 0.1.18](https://img.shields.io/badge/Version-0.1.18-informational?style=flat-square) ![AppVersion: 0.1.3](https://img.shields.io/badge/AppVersion-0.1.3-informational?style=flat-square)
 
 CaraML store registry: Feature registry for CaraML store.
 
@@ -85,6 +85,11 @@ CaraML store registry: Feature registry for CaraML store.
 | serving.image.tag | string | `""` |  |
 | serving.imagePullSecrets | list | `[]` |  |
 | serving.javaOpts | string | `nil` |  |
+| serving.livenessProbe.enabled | bool | `true` | Flag to enable the liveness probe |
+| serving.livenessProbe.failureThreshold | int | `2` | Min consecutive failures for the liveness probe to be considered failed |
+| serving.livenessProbe.initialDelaySeconds | int | `10` | Delay before the liveness probe is initiated |
+| serving.livenessProbe.periodSeconds | int | `10` | How often to perform the liveness probe |
+| serving.livenessProbe.timeoutSeconds | int | `2` | When the liveness probe times out |
 | serving.minReadySeconds | int | `0` | The minimum number of seconds for which a newly created Pod should be ready without any of its containers crashing, for it to be considered available. |
 | serving.name | string | `"serving"` |  |
 | serving.nameOverride | string | `""` |  |
@@ -93,19 +98,23 @@ CaraML store registry: Feature registry for CaraML store.
 | serving.podDisruptionBudget | object | `{}` | This value is used to configure a Kubernetes PodDisruptionBudget for Serving deployment |
 | serving.podLabels | object | `{}` |  |
 | serving.prometheus.monitor.enabled | bool | `false` | Create a ServiceMonitor resource to expose Prometheus metrics |
-| serving.readinessProbe.enabled | bool | `true` | Flag to enable the probe |
-| serving.readinessProbe.failureThreshold | int | `5` | Min consecutive failures for the probe to be considered failed |
-| serving.readinessProbe.initialDelaySeconds | int | `20` | Delay before the probe is initiated |
-| serving.readinessProbe.periodSeconds | int | `10` | How often to perform the probe |
-| serving.readinessProbe.successThreshold | int | `1` | Min consecutive success for the probe to be considered successful |
-| serving.readinessProbe.timeoutSeconds | int | `10` | When the probe times out |
+| serving.readinessProbe.enabled | bool | `true` | Flag to enable the readiness probe |
+| serving.readinessProbe.failureThreshold | int | `2` | Min consecutive failures for the readiness probe to be considered failed |
+| serving.readinessProbe.initialDelaySeconds | int | `2` | Delay before the readiness probe is initiated |
+| serving.readinessProbe.periodSeconds | int | `2` | How often to perform the readiness probe |
+| serving.readinessProbe.successThreshold | int | `2` | Min consecutive success for the readiness probe to be considered successful |
+| serving.readinessProbe.timeoutSeconds | int | `2` | When the readiness probe times out |
 | serving.replicaCount | int | `1` |  |
 | serving.resources | object | `{}` |  |
 | serving.secrets | list | `[]` |  |
 | serving.service.grpc.nodePort | string | `nil` | Port number that each cluster node will listen to |
 | serving.service.grpc.port | int | `6566` | Service port for GRPC requests |
 | serving.service.grpc.targetPort | int | `6566` | Container port serving GRPC requests |
 | serving.service.type | string | `"ClusterIP"` | Kubernetes service type |
+| serving.startupProbe.enabled | bool | `true` | Flag to enable the startup probe |
+| serving.startupProbe.failureThreshold | int | `6` | Min consecutive failures for the startup probe to be considered failed |
+| serving.startupProbe.periodSeconds | int | `5` | How often to perform the startup probe |
+| serving.startupProbe.timeoutSeconds | int | `1` | When the startup probe times out |
 | serving.strategy | object | `{}` | Strategy used to replace old Pods by new ones. .spec.strategy.type can be "Recreate" or "RollingUpdate". "RollingUpdate" is the default value. |
 | serving.tolerations | list | `[]` |  |
 

diff --git a/charts/caraml-store/templates/serving/deployment.yaml b/charts/caraml-store/templates/serving/deployment.yaml
@@ -106,6 +106,15 @@ spec:
             - name: http
               containerPort: {{ .Values.serving.actuator.port }}
               protocol: TCP
+          {{- if .Values.serving.startupProbe.enabled }}
+          startupProbe:
+            httpGet:
+              path: /actuator/health
+              port: {{ .Values.serving.actuator.port }}
+            periodSeconds: {{ .Values.serving.startupProbe.periodSeconds }}
+            timeoutSeconds: {{ .Values.serving.startupProbe.timeoutSeconds }}
+            failureThreshold: {{ .Values.serving.startupProbe.failureThreshold }}
+          {{- end }}
           {{- if .Values.serving.readinessProbe.enabled }}
           readinessProbe:
             httpGet:
@@ -117,6 +126,16 @@ spec:
             timeoutSeconds: {{ .Values.serving.readinessProbe.timeoutSeconds }}
             failureThreshold: {{ .Values.serving.readinessProbe.failureThreshold }}
           {{- end }}
+          {{- if .Values.serving.livenessProbe.enabled }}
+          livenessProbe:
+            httpGet:
+              path: /actuator/health
+              port: {{ .Values.serving.actuator.port }}
+            initialDelaySeconds: {{ .Values.serving.livenessProbe.initialDelaySeconds }}
+            periodSeconds: {{ .Values.serving.livenessProbe.periodSeconds }}
+            timeoutSeconds: {{ .Values.serving.livenessProbe.timeoutSeconds }}
+            failureThreshold: {{ .Values.serving.livenessProbe.failureThreshold }}
+          {{- end }}
           resources:
             {{- toYaml .Values.serving.resources | nindent 12 }}
       {{- with .Values.serving.nodeSelector }}

diff --git a/charts/caraml-store/values.yaml b/charts/caraml-store/values.yaml
@@ -160,19 +160,41 @@ serving:
 
   javaOpts:
 
+  startupProbe:
+    # serving.startupProbe.enabled -- Flag to enable the startup probe
+    enabled: true
+    # serving.startupProbe.periodSeconds -- How often to perform the startup probe
+    periodSeconds: 5
+    # serving.startupProbe.timeoutSeconds -- When the startup probe times out
+    timeoutSeconds: 1
+    # serving.startupProbe.failureThreshold -- Min consecutive failures for the startup probe to be considered failed
+    failureThreshold: 6
+
   readinessProbe:
-    # serving.readinessProbe.enabled -- Flag to enable the probe
+    # serving.readinessProbe.enabled -- Flag to enable the readiness probe
     enabled: true
-    # serving.readinessProbe.initialDelaySeconds -- Delay before the probe is initiated
-    initialDelaySeconds: 20
-    # serving.readinessProbe.periodSeconds -- How often to perform the probe
+    # serving.readinessProbe.initialDelaySeconds -- Delay before the readiness probe is initiated
+    initialDelaySeconds: 2
+    # serving.readinessProbe.periodSeconds -- How often to perform the readiness probe
+    periodSeconds: 2
+    # serving.readinessProbe.timeoutSeconds -- When the readiness probe times out
+    timeoutSeconds: 2
+    # serving.readinessProbe.successThreshold -- Min consecutive success for the readiness probe to be considered successful
+    successThreshold: 2
+    # serving.readinessProbe.failureThreshold -- Min consecutive failures for the readiness probe to be considered failed
+    failureThreshold: 2
+
+  livenessProbe:
+    # serving.livenessProbe.enabled -- Flag to enable the liveness probe
+    enabled: true
+    # serving.livenessProbe.initialDelaySeconds -- Delay before the liveness probe is initiated
+    initialDelaySeconds: 10
+    # serving.livenessProbe.periodSeconds -- How often to perform the liveness probe
     periodSeconds: 10
-    # serving.readinessProbe.timeoutSeconds -- When the probe times out
-    timeoutSeconds: 10
-    # serving.readinessProbe.successThreshold -- Min consecutive success for the probe to be considered successful
-    successThreshold: 1
-    # serving.readinessProbe.failureThreshold -- Min consecutive failures for the probe to be considered failed
-    failureThreshold: 5
+    # serving.livenessProbe.timeoutSeconds -- When the liveness probe times out
+    timeoutSeconds: 2
+    # serving.livenessProbe.failureThreshold -- Min consecutive failures for the liveness probe to be considered failed
+    failureThreshold: 2
 
   resources: {}