OCPVE-663: feat: introduce k8s events to lvmcluster and vgmanager

bundle/manifests/lvms-operator.clusterserviceversion.yaml

@@ -834,6 +834,14 @@ spec:
           - get
           - patch
           - update
+        - apiGroups:
+          - ""
+          resources:
+          - events
+          verbs:
+          - create
+          - patch
+          - update
         serviceAccountName: vg-manager
     strategy: deployment
   installModes:

cmd/vgmanager/main.go

@@ -26,9 +26,11 @@ import (
 	"k8s.io/apimachinery/pkg/runtime"
 	utilruntime "k8s.io/apimachinery/pkg/util/runtime"
 	clientgoscheme "k8s.io/client-go/kubernetes/scheme"
+	"k8s.io/klog/v2"
 	// Import all Kubernetes client auth plugins (e.g. Azure, GCP, OIDC, etc.)
 	// to ensure that exec-entrypoint and run can make use of them.
 	_ "k8s.io/client-go/plugin/pkg/client/auth"
+
 	ctrl "sigs.k8s.io/controller-runtime"
 	"sigs.k8s.io/controller-runtime/pkg/healthz"
 	"sigs.k8s.io/controller-runtime/pkg/log/zap"
@@ -55,7 +57,9 @@ func main() {
 	opts := zap.Options{}
 	opts.BindFlags(flag.CommandLine)
 	opts.Development = developmentMode
-	ctrl.SetLogger(zap.New(zap.UseFlagOptions(&opts)))
+	logr := zap.New(zap.UseFlagOptions(&opts))
+	ctrl.SetLogger(logr)
+	klog.SetLogger(logr)
 
 	setupLog := ctrl.Log.WithName("setup")
 
@@ -73,10 +77,11 @@ func main() {
 	}
 
 	if err = (&vgmanager.VGReconciler{
-		Client:    mgr.GetClient(),
-		Scheme:    mgr.GetScheme(),
-		NodeName:  os.Getenv("NODE_NAME"),
-		Namespace: os.Getenv("POD_NAMESPACE"),
+		Client:        mgr.GetClient(),
+		EventRecorder: mgr.GetEventRecorderFor(vgmanager.ControllerName),
+		Scheme:        mgr.GetScheme(),
+		NodeName:      os.Getenv("NODE_NAME"),
+		Namespace:     os.Getenv("POD_NAMESPACE"),
 	}).SetupWithManager(mgr); err != nil {
 		setupLog.Error(err, "unable to create controller", "controller", "VGManager")
 		os.Exit(1)

config/rbac/vg_manager_role.yaml

@@ -58,3 +58,11 @@ rules:
   - get
   - patch
   - update
+- apiGroups:
+    - ""
+  resources:
+    - events
+  verbs:
+    - create
+    - patch
+    - update

controllers/lvmcluster_controller.go

@@ -33,6 +33,7 @@ import (
 	k8serrors "k8s.io/apimachinery/pkg/api/errors"
 	"k8s.io/apimachinery/pkg/runtime"
 	"k8s.io/apimachinery/pkg/types"
+	"k8s.io/client-go/tools/record"
 	corev1helper "k8s.io/component-helpers/scheduling/corev1"
 
 	ctrl "sigs.k8s.io/controller-runtime"
@@ -42,6 +43,13 @@ import (
 	"sigs.k8s.io/controller-runtime/pkg/reconcile"
 )
 
+type EventReasonInfo string
+type EventReasonError string
+
+const EventReasonErrorDeletionPending EventReasonError = "DeletionPending"
+const EventReasonErrorResourceReconciliationFailed EventReasonError = "ResourceReconciliationFailed"
+const EventReasonResourceReconciliationSuccess EventReasonInfo = "ResourceReconciliationSuccess"
+
 var lvmClusterFinalizer = "lvmcluster.topolvm.io"
 
 const (
@@ -64,6 +72,7 @@ type resourceManager interface {
 // LVMClusterReconciler reconciles a LVMCluster object
 type LVMClusterReconciler struct {
 	client.Client
+	record.EventRecorder
 	Scheme      *runtime.Scheme
 	ClusterType cluster.Type
 	Namespace   string
@@ -146,15 +155,17 @@ func (r *LVMClusterReconciler) reconcile(ctx context.Context, instance *lvmv1alp
 			return ctrl.Result{}, fmt.Errorf("failed to check if LogicalVolumes exist: %w", err)
 		}
 		if lvsExist {
+			waitForLVRemoval := time.Second * 10
+			err := fmt.Errorf("found PVCs provisioned by topolvm, waiting %s for their deletion: %w", waitForLVRemoval, err)
+			r.WarningEvent(ctx, instance, EventReasonErrorDeletionPending, err)
 			// check every 10 seconds if there are still PVCs present
-			return ctrl.Result{RequeueAfter: time.Second * 10},
-				fmt.Errorf("found PVCs provisioned by topolvm, waiting for their deletion: %w", err)
+			return ctrl.Result{RequeueAfter: waitForLVRemoval}, err
 		}
 
 		logger.Info("processing LVMCluster deletion")
 		if err := r.processDelete(ctx, instance); err != nil {
 			// check every 10 seconds if there are still PVCs present or the LogicalVolumes are removed
-			return ctrl.Result{Requeue: true}, fmt.Errorf("failed to process LVMCluster deletion")
+			return ctrl.Result{Requeue: true}, fmt.Errorf("failed to process LVMCluster deletion: %w", err)
 		}
 		return reconcile.Result{}, nil
 	}
@@ -199,11 +210,14 @@ func (r *LVMClusterReconciler) reconcile(ctx context.Context, instance *lvmv1alp
 
 	resourceSyncElapsedTime := time.Since(resourceSyncStart)
 	if len(errs) > 0 {
-		return ctrl.Result{}, fmt.Errorf("failed to reconcile resources managed by LVMCluster within %v: %w",
-			resourceSyncElapsedTime, errors.Join(errs...))
+		err := fmt.Errorf("failed to reconcile resources managed by LVMCluster: %w", errors.Join(errs...))
+		r.WarningEvent(ctx, instance, EventReasonErrorResourceReconciliationFailed, err)
+		return ctrl.Result{}, err
 	}
 
-	logger.Info("successfully reconciled LVMCluster", "resourceSyncElapsedTime", resourceSyncElapsedTime)
+	msg := "successfully reconciled LVMCluster"
+	logger.Info(msg, "resourceSyncElapsedTime", resourceSyncElapsedTime)
+	r.NormalEvent(ctx, instance, EventReasonResourceReconciliationSuccess, msg)
 
 	return ctrl.Result{}, nil
 }
@@ -358,15 +372,13 @@ func (r *LVMClusterReconciler) logicalVolumesExist(ctx context.Context) (bool, e
 		return false, fmt.Errorf("failed to get TopoLVM LogicalVolume list: %w", err)
 	}
 	if len(logicalVolumeList.Items) > 0 {
-
 		return true, nil
 	}
 	return false, nil
 }
 
 func (r *LVMClusterReconciler) processDelete(ctx context.Context, instance *lvmv1alpha1.LVMCluster) error {
 	if controllerutil.ContainsFinalizer(instance, lvmClusterFinalizer) {
-
 		resourceDeletionList := []resourceManager{
 			&topolvmVolumeSnapshotClass{},
 			&topolvmStorageClass{},
@@ -383,16 +395,29 @@ func (r *LVMClusterReconciler) processDelete(ctx context.Context, instance *lvmv
 
 		for _, unit := range resourceDeletionList {
 			if err := unit.ensureDeleted(r, ctx, instance); err != nil {
-				return fmt.Errorf("failed cleaning up: %s %w", unit.getName(), err)
+				err := fmt.Errorf("failed cleaning up %s: %w", unit.getName(), err)
+				r.WarningEvent(ctx, instance, EventReasonErrorDeletionPending, err)
+				return err
 			}
 		}
+	}
 
-		if update := controllerutil.RemoveFinalizer(instance, lvmClusterFinalizer); update {
-			if err := r.Client.Update(ctx, instance); err != nil {
-				return fmt.Errorf("failed to remove finalizer from LVMCluster %s: %w", instance.GetName(), err)
-			}
+	if update := controllerutil.RemoveFinalizer(instance, lvmClusterFinalizer); update {
+		if err := r.Client.Update(ctx, instance); err != nil {
+			return fmt.Errorf("failed to remove finalizer from LVMCluster %s: %w", instance.GetName(), err)
 		}
 	}
 
 	return nil
 }
+
+func (r *LVMClusterReconciler) WarningEvent(_ context.Context, obj client.Object, reason EventReasonError, err error) {
+	r.Event(obj, corev1.EventTypeWarning, string(reason), err.Error())
+}
+
+func (r *LVMClusterReconciler) NormalEvent(ctx context.Context, obj client.Object, reason EventReasonInfo, message string) {
+	if !log.FromContext(ctx).V(1).Enabled() {
+		return
+	}
+	r.Event(obj, corev1.EventTypeNormal, string(reason), message)
+}

controllers/suite_test.go

@@ -122,11 +122,12 @@ var _ = BeforeSuite(func() {
 	Expect(err).ToNot(HaveOccurred())
 
 	err = (&LVMClusterReconciler{
-		Client:      k8sManager.GetClient(),
-		Scheme:      k8sManager.GetScheme(),
-		ClusterType: clusterType,
-		Namespace:   testLvmClusterNamespace,
-		ImageName:   testImageName,
+		Client:        k8sManager.GetClient(),
+		EventRecorder: k8sManager.GetEventRecorderFor(ControllerName),
+		Scheme:        k8sManager.GetScheme(),
+		ClusterType:   clusterType,
+		Namespace:     testLvmClusterNamespace,
+		ImageName:     testImageName,
 	}).SetupWithManager(k8sManager)
 	Expect(err).ToNot(HaveOccurred())
 

controllers/utils.go

@@ -95,17 +95,17 @@ func verifyDaemonSetReadiness(ds *appsv1.DaemonSet) error {
 
 func verifyDeploymentReadiness(dep *appsv1.Deployment) error {
 	if len(dep.Status.Conditions) == 0 {
-		return fmt.Errorf("the Deployment is not ready: %s/%s. deployment cannot be ready as no condition was found", dep.Namespace, dep.Name)
+		return fmt.Errorf("the Deployment %s/%s is not ready as no condition was found", dep.Namespace, dep.Name)
 	}
 	for _, condition := range dep.Status.Conditions {
 		if condition.Type == appsv1.DeploymentAvailable && condition.Status == corev1.ConditionFalse {
-			return fmt.Errorf("the Deployment is not ready: %s/%s. deployment has not reached minimum availability and thus not ready: %v",
+			return fmt.Errorf("the Deployment %s/%s has not reached minimum availability and is not ready: %v",
 				dep.Namespace, dep.Name, condition)
 		} else if condition.Type == appsv1.DeploymentProgressing && condition.Status == corev1.ConditionFalse {
-			return fmt.Errorf("the Deployment is not ready: %s/%s. deployment has not progressed and is thus not ready: %v",
+			return fmt.Errorf("the Deployment %s/%s has not progressed and is not ready: %v",
 				dep.Namespace, dep.Name, condition)
 		} else if condition.Type == appsv1.DeploymentReplicaFailure && condition.Status == corev1.ConditionTrue {
-			return fmt.Errorf("the Deployment is not ready: %s/%s. deployment has a replica failure and is thus not ready: %v",
+			return fmt.Errorf("the Deployment %s/%s has a replica failure and is not ready: %v",
 				dep.Namespace, dep.Name, condition)
 		}
 	}

main.go

@@ -141,6 +141,7 @@ func main() {
 	if err = (&controllers.LVMClusterReconciler{
 		Client:                           mgr.GetClient(),
 		Scheme:                           mgr.GetScheme(),
+		EventRecorder:                    mgr.GetEventRecorderFor(controllers.ControllerName),
 		ClusterType:                      clusterType,
 		Namespace:                        operatorNamespace,
 		TopoLVMLeaderElectionPassthrough: leaderElectionConfig,

pkg/vgmanager/status.go

@@ -78,12 +78,7 @@ func (r *VGReconciler) setVolumeGroupStatus(ctx context.Context, status *lvmv1al
 	logger := log.FromContext(ctx)
 
 	// Get LVMVolumeGroupNodeStatus and set the relevant VGStatus
-	nodeStatus := &lvmv1alpha1.LVMVolumeGroupNodeStatus{
-		ObjectMeta: metav1.ObjectMeta{
-			Name:      r.NodeName,
-			Namespace: r.Namespace,
-		},
-	}
+	nodeStatus := r.getLVMVolumeGroupNodeStatus()
 
 	result, err := ctrl.CreateOrUpdate(ctx, r.Client, nodeStatus, func() error {
 		exists := false
@@ -171,3 +166,12 @@ func (r *VGReconciler) setDevices(status *lvmv1alpha1.VGStatus) (bool, error) {
 
 	return devicesExist, nil
 }
+
+func (r *VGReconciler) getLVMVolumeGroupNodeStatus() *lvmv1alpha1.LVMVolumeGroupNodeStatus {
+	return &lvmv1alpha1.LVMVolumeGroupNodeStatus{
+		ObjectMeta: metav1.ObjectMeta{
+			Name:      r.NodeName,
+			Namespace: r.Namespace,
+		},
+	}
+}