nfd-gc: add healthz endpoint

Author: Kaston-C

cmd/nfd-gc/main.go

@@ -84,7 +84,7 @@ func initFlags(flagset *flag.FlagSet) *nfdgarbagecollector.Args {
 	flagset.StringVar(&args.Kubeconfig, "kubeconfig", "",
 		"Kubeconfig to use")
 	flagset.IntVar(&args.Port, "port", 8080,
-		"Port on which to expose metrics.")
+		"Port which metrics and healthz endpoints are served on")
 
 	klog.InitFlags(flagset)
 

deployment/base/gc/gc.yaml

@@ -19,6 +19,19 @@ spec:
         - name: nfd-gc
           image: gcr.io/k8s-staging-nfd/node-feature-discovery:master
           imagePullPolicy: Always
+          livenessProbe:
+            httpGet:
+              path: /healthz
+              port: http
+            initialDelaySeconds: 10
+            periodSeconds: 10
+          readinessProbe:
+            httpGet:
+              path: /healthz
+              port: http
+            initialDelaySeconds: 5
+            periodSeconds: 10
+            failureThreshold: 10
           resources:
             limits:
               cpu: 20m

deployment/helm/node-feature-discovery/templates/nfd-gc.yaml

@@ -44,6 +44,41 @@ spec:
       - name: gc
         image: "{{ .Values.image.repository }}:{{ .Values.image.tag | default .Chart.AppVersion }}"
         imagePullPolicy: "{{ .Values.image.pullPolicy }}"
+        livenessProbe:
+          httpGet:
+            path: /healthz
+            port: http
+        {{- with .Values.gc.livenessProbe.initialDelaySeconds }}
+          initialDelaySeconds: {{ . }}
+        {{- end }}
+        {{- with .Values.gc.livenessProbe.failureThreshold }}
+          failureThreshold: {{ . }}
+        {{- end }}
+        {{- with .Values.gc.livenessProbe.periodSeconds }}
+          periodSeconds: {{ . }}
+        {{- end }}
+        {{- with .Values.gc.livenessProbe.timeoutSeconds }}
+          timeoutSeconds: {{ . }}
+        {{- end }}
+        readinessProbe:
+          httpGet:
+            path: /healthz
+            port: http
+        {{- with .Values.gc.readinessProbe.initialDelaySeconds }}
+          initialDelaySeconds: {{ . }}
+        {{- end }}
+        {{- with .Values.gc.readinessProbe.failureThreshold }}
+          failureThreshold: {{ . }}
+        {{- end }}
+        {{- with .Values.gc.readinessProbe.periodSeconds }}
+          periodSeconds: {{ . }}
+        {{- end }}
+        {{- with .Values.gc.readinessProbe.timeoutSeconds }}
+          timeoutSeconds: {{ . }}
+        {{- end }}
+        {{- with .Values.gc.readinessProbe.successThreshold }}
+          successThreshold: {{ . }}
+        {{- end }}
         env:
         - name: NODE_NAME
           valueFrom:

deployment/helm/node-feature-discovery/values.yaml

@@ -565,6 +565,18 @@ gc:
 
   podSecurityContext: {}
 
+  livenessProbe:
+    initialDelaySeconds: 10
+    # failureThreshold: 3
+    # periodSeconds: 10
+    # timeoutSeconds: 1
+  readinessProbe:
+    initialDelaySeconds: 5
+    failureThreshold: 10
+    # periodSeconds: 10
+    # timeoutSeconds: 1
+    # successThreshold: 1
+
   resources:
     limits:
       memory: 1Gi

docs/deployment/helm.md

@@ -330,7 +330,7 @@ API's you need to install the prometheus operator in your cluster.
 | `gc.podDisruptionBudget.minAvailable`               | integer | 1                         | Specifies minAvailable for the podDisruptionBudget configuration for nfd-gc                                                                                    |
 | `gc.podDisruptionBudget.maxUnavailable`             | integer | NULL                      | Specifies maxUnavailable for the podDisruptionBudget configuration for nfd-gc                                                                                  |
 | `gc.podDisruptionBudget.unhealthyPodEvictionPolicy` | string  | AlwaysAllow               | Specifies unhealthyPodEvictionPolicy for the podDisruptionBudget configuration for nfd-gc                                                                      |
-| `gc.port`                                           | integer | 8080                      | Port on which to serve Prometheus metrics.                                                                                                                     |
+| `gc.port`                                           | integer | 8080                      | Port on which to serve http for metrics and health endpoints.                                                                                                                     |
 | `gc.nodeSelector`                                   | dict    | {}                        | Garbage collector pod [node selector][nodeselector]                                                                                                            |
 | `gc.tolerations`                                    | dict    | {}                        | Garbage collector pod [node tolerations][toleration]                                                                                                           |
 | `gc.annotations`                                    | dict    | {}                        | Garbage collector pod [annotations][annotations]                                                                                                               |

pkg/nfd-gc/nfd-gc.go

@@ -82,6 +82,10 @@ func New(args *Args) (NfdGarbageCollector, error) {
 	}, nil
 }
 
+func (n *nfdGarbageCollector) Healthz(writer http.ResponseWriter, _ *http.Request) {
+	writer.WriteHeader(http.StatusOK)
+}
+
 func (n *nfdGarbageCollector) deleteNodeFeature(namespace, name string) {
 	kind := "NodeFeature"
 	if err := n.client.Resource(gvrNF).Namespace(namespace).Delete(context.TODO(), name, metav1.DeleteOptions{}); err != nil {
@@ -252,7 +256,8 @@ func (n *nfdGarbageCollector) Run() error {
 	httpMux.Handle("/metrics", promhttp.HandlerFor(promRegistry, promhttp.HandlerOpts{}))
 	registerVersion(version.Get())
 
-	// TODO: health probe endpoint could be added here
+	// Register health endpoint (at this point we're "ready and live")
+	httpMux.HandleFunc("/healthz", n.Healthz)
 
 	// Start HTTP server
 	httpServer := http.Server{Addr: fmt.Sprintf(":%d", n.args.Port), Handler: httpMux}