Merge pull request #12918 from sbueringer/pr-fix-sync-machines-flake

🌱 Fix flakes in syncMachines unit test

Author: k8s-ci-robot

internal/controllers/machineset/machineset_controller_test.go

@@ -1395,6 +1395,11 @@ func TestMachineSetReconciler_syncMachines(t *testing.T) {
 	g.Expect(ssa.Patch(ctx, env.Client, machineSetManagerName, bootstrapConfig)).To(Succeed())
 	g.Expect(ssa.RemoveManagedFieldsForLabelsAndAnnotations(ctx, env.Client, env.GetAPIReader(), bootstrapConfig, machineSetManagerName)).To(Succeed())
 
+	// Set ownerReferences of the Machines to the MachineSet to ensure deterministic field ownership by capi-machineset below.
+	// Note: There is code in the Machine controller that sets ownerRefs to the Cluster on standalone Machines.
+	inPlaceMutatingMachine.SetOwnerReferences([]metav1.OwnerReference{*metav1.NewControllerRef(ms, machineSetKind)})
+	deletingMachine.SetOwnerReferences([]metav1.OwnerReference{*metav1.NewControllerRef(ms, machineSetKind)})
+
 	// Create Machines (same as in syncReplicas)
 	g.Expect(ssa.Patch(ctx, env.Client, machineSetManagerName, inPlaceMutatingMachine)).To(Succeed())
 	g.Expect(ssa.Patch(ctx, env.Client, machineSetManagerName, deletingMachine)).To(Succeed())
@@ -1573,20 +1578,20 @@ func TestMachineSetReconciler_syncMachines(t *testing.T) {
 	g.Eventually(func(g Gomega) {
 		updatedDeletingMachine := deletingMachine.DeepCopy()
 		g.Expect(env.GetAPIReader().Get(ctx, client.ObjectKeyFromObject(updatedDeletingMachine), updatedDeletingMachine)).To(Succeed())
-		statusManagedFields := managedFieldsMatching(updatedInPlaceMutatingMachine.ManagedFields, "manager", metav1.ManagedFieldsOperationUpdate, "status")
+		statusManagedFields := managedFieldsMatching(updatedDeletingMachine.ManagedFields, "manager", metav1.ManagedFieldsOperationUpdate, "status")
 		g.Expect(statusManagedFields).To(HaveLen(1))
 		g.Expect(cleanupTime(updatedDeletingMachine.ManagedFields)).To(ConsistOf(toManagedFields([]managedFieldEntry{{
 			// capi-machineset owns almost everything.
 			Manager:    machineSetManagerName,
 			Operation:  metav1.ManagedFieldsOperationApply,
 			APIVersion: clusterv1.GroupVersion.String(),
-			FieldsV1:   "{\"f:metadata\":{\"f:finalizers\":{\"v:\\\"testing-finalizer\\\"\":{}}},\"f:spec\":{\"f:bootstrap\":{\"f:dataSecretName\":{}},\"f:clusterName\":{},\"f:infrastructureRef\":{\"f:apiGroup\":{},\"f:kind\":{},\"f:name\":{}}}}",
+			FieldsV1:   fmt.Sprintf("{\"f:metadata\":{\"f:finalizers\":{\"v:\\\"testing-finalizer\\\"\":{}},\"f:ownerReferences\":{\"k:{\\\"uid\\\":\\\"%s\\\"}\":{}}},\"f:spec\":{\"f:bootstrap\":{\"f:dataSecretName\":{}},\"f:clusterName\":{},\"f:infrastructureRef\":{\"f:apiGroup\":{},\"f:kind\":{},\"f:name\":{}}}}", ms.UID),
 		}, {
 			// manager owns the fields that are propagated in-place for deleting Machines in syncMachines via patchHelper.
 			Manager:    "manager",
 			Operation:  metav1.ManagedFieldsOperationUpdate,
 			APIVersion: clusterv1.GroupVersion.String(),
-			FieldsV1:   fmt.Sprintf("{\"f:metadata\":{\"f:ownerReferences\":{\".\":{},\"k:{\\\"uid\\\":\\\"%s\\\"}\":{}}},\"f:spec\":{\"f:deletion\":{\"f:nodeDrainTimeoutSeconds\":{},\"f:nodeVolumeDetachTimeoutSeconds\":{}},\"f:minReadySeconds\":{},\"f:readinessGates\":{\".\":{},\"k:{\\\"conditionType\\\":\\\"foo\\\"}\":{\".\":{},\"f:conditionType\":{}}}}}", testCluster.UID),
+			FieldsV1:   "{\"f:spec\":{\"f:deletion\":{\"f:nodeDrainTimeoutSeconds\":{},\"f:nodeVolumeDetachTimeoutSeconds\":{}},\"f:minReadySeconds\":{},\"f:readinessGates\":{\".\":{},\"k:{\\\"conditionType\\\":\\\"foo\\\"}\":{\".\":{},\"f:conditionType\":{}}}}}",
 		}, {
 			// manager owns status.
 			Manager:    "manager",