add liveness, readiness, and startup probes to noobaa-operator

- integrate controller-runtime healthz and readyz endpoints into the namespace-scoped manager
- expose probe server on :8081 and register default checks
- update operator Deployment with liveness, readiness, and startup probes

Signed-off-by: Oded Viner <[email protected]>

Author: OdedViner

deploy/operator.yaml

@@ -41,6 +41,13 @@ spec:
       containers:
         - name: noobaa-operator
           image: NOOBAA_OPERATOR_IMAGE
+          args:
+            - operator  
+            - run       
+            - --health-probe-bind-address=:8081
+          ports:
+            - name: healthz
+              containerPort: 8081
           volumeMounts:
           - name: bound-sa-token
             mountPath: /var/run/secrets/openshift/serviceaccount
@@ -50,6 +57,29 @@ spec:
           # SHOULD BE RETURNED ONCE COSI IS BACK
           # - name: socket
           #   mountPath: /var/lib/cosi
+          readinessProbe:
+            httpGet:
+              path: /readyz
+              port: healthz
+            initialDelaySeconds: 5
+            periodSeconds: 10
+            timeoutSeconds: 5
+            failureThreshold: 3
+          livenessProbe:
+            httpGet:
+              path: /healthz
+              port: healthz
+            initialDelaySeconds: 15
+            periodSeconds: 10
+            timeoutSeconds: 5
+            failureThreshold: 3
+          startupProbe:
+            httpGet:
+              path: /healthz
+              port: healthz
+            periodSeconds: 10
+            timeoutSeconds: 5
+            failureThreshold: 30
           resources:
             limits:
               cpu: "250m"

pkg/bundle/deploy.go

@@ -6189,7 +6189,7 @@ spec:
   sourceNamespace: default
 `
 
-const Sha256_deploy_operator_yaml = "aa8da1c289a05b3c94b9393b04307d38814a67625ac6a8006dace4d09366f35b"
+const Sha256_deploy_operator_yaml = "1e940d1e2206f9b692fdcfa0f9bd80fe6a8247f70429216c359ec4851219c2e5"
 
 const File_deploy_operator_yaml = `apiVersion: apps/v1
 kind: Deployment
@@ -6234,6 +6234,13 @@ spec:
       containers:
         - name: noobaa-operator
           image: NOOBAA_OPERATOR_IMAGE
+          args:
+            - operator  
+            - run       
+            - --health-probe-bind-address=:8081
+          ports:
+            - name: healthz
+              containerPort: 8081
           volumeMounts:
           - name: bound-sa-token
             mountPath: /var/run/secrets/openshift/serviceaccount
@@ -6243,6 +6250,29 @@ spec:
           # SHOULD BE RETURNED ONCE COSI IS BACK
           # - name: socket
           #   mountPath: /var/lib/cosi
+          readinessProbe:
+            httpGet:
+              path: /readyz
+              port: healthz
+            initialDelaySeconds: 5
+            periodSeconds: 10
+            timeoutSeconds: 5
+            failureThreshold: 3
+          livenessProbe:
+            httpGet:
+              path: /healthz
+              port: healthz
+            initialDelaySeconds: 15
+            periodSeconds: 10
+            timeoutSeconds: 5
+            failureThreshold: 3
+          startupProbe:
+            httpGet:
+              path: /healthz
+              port: healthz
+            periodSeconds: 10
+            timeoutSeconds: 5
+            failureThreshold: 30
           resources:
             limits:
               cpu: "250m"
@@ -6926,4 +6956,3 @@ metadata:
     app: prometheus-adapter
   name: custom-metrics-prometheus-adapter
 `
-

pkg/operator/manager.go

@@ -24,6 +24,7 @@ import (
 
 	"github.com/operator-framework/operator-lib/leader"
 	"sigs.k8s.io/controller-runtime/pkg/cache"
+	"sigs.k8s.io/controller-runtime/pkg/healthz"
 	"sigs.k8s.io/controller-runtime/pkg/manager"
 	"sigs.k8s.io/controller-runtime/pkg/manager/signals"
 	metricsServer "sigs.k8s.io/controller-runtime/pkg/metrics/server"
@@ -44,6 +45,8 @@ func RunOperator(cmd *cobra.Command, args []string) {
 		util.InitLogger(logrus.DebugLevel)
 	}
 	version.RunVersion(cmd, args)
+	// Probe address from CLI flag (defaults to :8081)
+	probeAddr, _ := cmd.Flags().GetString("health-probe-bind-address")
 
 	config := util.KubeConfig()
 
@@ -66,6 +69,7 @@ func RunOperator(cmd *cobra.Command, args []string) {
 		Metrics: metricsServer.Options{
 			BindAddress: fmt.Sprintf("%s:%d", metricsHost, metricsPort),
 		},
+		HealthProbeBindAddress: probeAddr, // Serve /healthz and /readyz here
 	})
 	if err != nil {
 		log.Fatalf("Failed to create manager: %s", err)
@@ -97,6 +101,16 @@ func RunOperator(cmd *cobra.Command, args []string) {
 		log.Fatalf("Failed AddToClusterScopedManager: %s", err)
 	}
 
+	// Register health and readiness endpoints on mgr
+	if err := mgr.AddHealthzCheck("healthz", healthz.Ping); err != nil {
+		log.Fatalf("Failed to add health check: %s", err)
+	}
+
+	if err := mgr.AddReadyzCheck("readyz", healthz.Ping); err != nil {
+		log.Fatalf("Failed to add readiness check: %s", err)
+	}
+
+
 	util.Panic(mgr.Add(manager.RunnableFunc(func(ctx context.Context) error {
 		system.RunOperatorCreate(cmd, args)
 		return nil

pkg/operator/operator.go

@@ -33,6 +33,8 @@ func Cmd() *cobra.Command {
 		Use:   "operator",
 		Short: "Deployment using operator",
 	}
+	// health-probe flag available on operator and inherited by subcommands
+	cmd.PersistentFlags().String("health-probe-bind-address", ":8081", "HTTP address for health/readiness probes (e.g., :8081)")
 	cmd.AddCommand(
 		CmdInstall(),
 		CmdUninstall(),