Add clusterClass generation check to Cluster reconciler

Author: killianmuldoon

api/v1beta1/condition_consts.go

@@ -276,6 +276,11 @@ const (
 	// TopologyReconciledHookBlockingReason (Severity=Info) documents reconciliation of a Cluster topology
 	// not yet completed because at least one of the lifecycle hooks is blocking.
 	TopologyReconciledHookBlockingReason = "LifecycleHookBlocking"
+
+	// TopologyReconciledClusterClassNotReconciledReason (Severity=Info) documents reconciliation of a Cluster topology not
+	// yet completed because the ClusterClass has not reconciled yet. If this condition persists there may be an issue
+	// with the ClusterClass surfaced in the ClusterClass status or controller logs.
+	TopologyReconciledClusterClassNotReconciledReason = "ClusterClassNotReconciled"
 )
 
 // Conditions and condition reasons for ClusterClass.

internal/controllers/topology/cluster/cluster_controller.go

@@ -221,6 +221,31 @@ func (r *Reconciler) Reconcile(ctx context.Context, req ctrl.Request) (_ ctrl.Re
 func (r *Reconciler) reconcile(ctx context.Context, s *scope.Scope) (ctrl.Result, error) {
 	var err error
 
+	// Get ClusterClass.
+	clusterClass := &clusterv1.ClusterClass{}
+	key := client.ObjectKey{Name: s.Current.Cluster.Spec.Topology.Class, Namespace: s.Current.Cluster.Namespace}
+	if err := r.Client.Get(ctx, key, clusterClass); err != nil {
+		return ctrl.Result{}, errors.Wrapf(err, "failed to retrieve ClusterClass %s", s.Current.Cluster.Spec.Topology.Class)
+	}
+
+	s.Blueprint.ClusterClass = clusterClass
+	// If the ClusterClass `metadata.Generation` doesn't match the `status.ObservedGeneration` return as the ClusterClass
+	// is not up to date.
+	// Note: This doesn't require requeue as a change to ClusterClass observedGeneration will cause an additional reconcile
+	// in the Cluster.
+	if clusterClass.GetGeneration() != clusterClass.Status.ObservedGeneration {
+		return ctrl.Result{}, nil
+	}
+
+	// Default and Validate the Cluster based on information from the ClusterClass.
+	// This step is needed as if the ClusterClass does not exist at Cluster creation some fields may not be defaulted or
+	// validated in the webhook.
+	if errs := webhooks.DefaultVariables(s.Current.Cluster, clusterClass); len(errs) > 0 {
+		return ctrl.Result{}, apierrors.NewInvalid(clusterv1.GroupVersion.WithKind("Cluster").GroupKind(), s.Current.Cluster.Name, errs)
+	}
+	if errs := webhooks.ValidateClusterForClusterClass(s.Current.Cluster, clusterClass, field.NewPath("spec", "topology")); len(errs) > 0 {
+		return ctrl.Result{}, apierrors.NewInvalid(clusterv1.GroupVersion.WithKind("Cluster").GroupKind(), s.Current.Cluster.Name, errs)
+	}
 	// Gets the blueprint with the ClusterClass and the referenced templates
 	// and store it in the request scope.
 	s.Blueprint, err = r.getBlueprint(ctx, s.Current.Cluster, s.Blueprint.ClusterClass)

internal/controllers/topology/cluster/conditions.go

@@ -37,6 +37,7 @@ func (r *Reconciler) reconcileConditions(s *scope.Scope, cluster *clusterv1.Clus
 // cluster are in sync with the topology defined in the cluster.
 // The condition is false under the following conditions:
 // - An error occurred during the reconcile process of the cluster topology.
+// - The ClusterClass has not been successfully reconciled with its current spec.
 // - The cluster upgrade has not yet propagated to all the components of the cluster.
 //   - For a managed topology cluster the version upgrade is propagated one component at a time.
 //     In such a case, since some of the component's spec would be adrift from the topology the
@@ -58,6 +59,22 @@ func (r *Reconciler) reconcileTopologyReconciledCondition(s *scope.Scope, cluste
 		return nil
 	}
 
+	// If the ClusterClass `metadata.Generation` doesn't match the `status.ObservedGeneration` requeue as the ClusterClass
+	// is not up to date.
+	if s.Blueprint != nil && s.Blueprint.ClusterClass != nil &&
+		s.Blueprint.ClusterClass.GetGeneration() != s.Blueprint.ClusterClass.Status.ObservedGeneration {
+		conditions.Set(
+			cluster,
+			conditions.FalseCondition(
+				clusterv1.TopologyReconciledCondition,
+				clusterv1.TopologyReconciledClusterClassNotReconciledReason,
+				clusterv1.ConditionSeverityInfo,
+				"ClusterClass is outdated. If this condition persists please check ClusterClass status.",
+			),
+		)
+		return nil
+	}
+
 	// If any of the lifecycle hooks are blocking any part of the reconciliation then topology
 	// is not considered as fully reconciled.
 	if s.HookResponseTracker.AggregateRetryAfter() != 0 {

internal/controllers/topology/cluster/conditions_test.go

@@ -22,6 +22,7 @@ import (
 	. "github.com/onsi/gomega"
 	"github.com/pkg/errors"
 	corev1 "k8s.io/api/core/v1"
+	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 
 	clusterv1 "sigs.k8s.io/cluster-api/api/v1beta1"
 	runtimehooksv1 "sigs.k8s.io/cluster-api/exp/runtime/hooks/api/v1alpha1"
@@ -48,6 +49,26 @@ func TestReconcileTopologyReconciledCondition(t *testing.T) {
 			wantConditionReason: clusterv1.TopologyReconcileFailedReason,
 			wantErr:             false,
 		},
+		{
+			name:    "should set the condition to false if the ClusterClass is out of date",
+			cluster: &clusterv1.Cluster{},
+			s: &scope.Scope{
+				Blueprint: &scope.ClusterBlueprint{
+					ClusterClass: &clusterv1.ClusterClass{
+						ObjectMeta: metav1.ObjectMeta{
+							Name:       "class1",
+							Generation: 10,
+						},
+						Status: clusterv1.ClusterClassStatus{
+							ObservedGeneration: 999,
+						},
+					},
+				},
+			},
+			wantConditionStatus: corev1.ConditionFalse,
+			wantConditionReason: clusterv1.TopologyReconciledClusterClassNotReconciledReason,
+			wantErr:             false,
+		},
 		{
 			name:         "should set the condition to false if the there is a blocking hook",
 			reconcileErr: nil,