Merge pull request kubernetes-sigs#9077 from killianmuldoon/pr-scale-test-upgrade

🌱 Add scale testing for upgrades

Author: k8s-ci-robot

test/e2e/scale.go

@@ -41,6 +41,7 @@ import (
 	"sigs.k8s.io/controller-runtime/pkg/client"
 
 	clusterv1 "sigs.k8s.io/cluster-api/api/v1beta1"
+	controlplanev1 "sigs.k8s.io/cluster-api/controlplane/kubeadm/api/v1beta1"
 	"sigs.k8s.io/cluster-api/test/e2e/internal/log"
 	"sigs.k8s.io/cluster-api/test/framework"
 	"sigs.k8s.io/cluster-api/test/framework/clusterctl"
@@ -123,6 +124,9 @@ type scaleSpecInput struct {
 	// if the test suit should fail as soon as c6 fails or if it should fail after all cluster creations are done.
 	FailFast bool
 
+	// SkipUpgrade if set to true will skip upgrading the workload clusters.
+	SkipUpgrade bool
+
 	// SkipCleanup if set to true will skip deleting the workload clusters.
 	SkipCleanup bool
 
@@ -133,7 +137,7 @@ type scaleSpecInput struct {
 	SkipWaitForCreation bool
 }
 
-// scaleSpec implements a scale test.
+// scaleSpec implements a scale test for clusters with MachineDeployments.
 func scaleSpec(ctx context.Context, inputGetter func() scaleSpecInput) {
 	var (
 		specName      = "scale"
@@ -253,7 +257,7 @@ func scaleSpec(ctx context.Context, inputGetter func() scaleSpecInput) {
 			Flavor:                   flavor,
 			Namespace:                scaleClusterNamespacePlaceholder,
 			ClusterName:              scaleClusterNamePlaceholder,
-			KubernetesVersion:        input.E2EConfig.GetVariable(KubernetesVersion),
+			KubernetesVersion:        input.E2EConfig.GetVariable(KubernetesVersionUpgradeFrom),
 			ControlPlaneMachineCount: controlPlaneMachineCount,
 			WorkerMachineCount:       workerMachineCount,
 		})
@@ -333,6 +337,47 @@ func scaleSpec(ctx context.Context, inputGetter func() scaleSpecInput) {
 			Fail("")
 		}
 
+		if !input.SkipUpgrade {
+			By("Upgrade the workload clusters concurrently")
+			// Get the upgrade function for upgrading the workload clusters.
+			upgrader := getClusterUpgradeAndWaitFn(framework.UpgradeClusterTopologyAndWaitForUpgradeInput{
+				ClusterProxy:                input.BootstrapClusterProxy,
+				KubernetesUpgradeVersion:    input.E2EConfig.GetVariable(KubernetesVersionUpgradeTo),
+				EtcdImageTag:                input.E2EConfig.GetVariable(EtcdVersionUpgradeTo),
+				DNSImageTag:                 input.E2EConfig.GetVariable(CoreDNSVersionUpgradeTo),
+				WaitForMachinesToBeUpgraded: input.E2EConfig.GetIntervals(specName, "wait-machine-upgrade"),
+				WaitForKubeProxyUpgrade:     input.E2EConfig.GetIntervals(specName, "wait-machine-upgrade"),
+				WaitForDNSUpgrade:           input.E2EConfig.GetIntervals(specName, "wait-machine-upgrade"),
+				WaitForEtcdUpgrade:          input.E2EConfig.GetIntervals(specName, "wait-machine-upgrade"),
+			})
+
+			clusterNamesToUpgrade := []string{}
+			for _, result := range clusterCreateResults {
+				clusterNamesToUpgrade = append(clusterNamesToUpgrade, result.clusterName)
+			}
+
+			// Upgrade all the workload clusters.
+			_, err = workConcurrentlyAndWait(ctx, workConcurrentlyAndWaitInput{
+				ClusterNames: clusterNamesToUpgrade,
+				Concurrency:  concurrency,
+				FailFast:     input.FailFast,
+				WorkerFunc: func(ctx context.Context, inputChan chan string, resultChan chan workResult, wg *sync.WaitGroup) {
+					upgradeClusterAndWaitWorker(ctx, inputChan, resultChan, wg, namespace.Name, input.DeployClusterInSeparateNamespaces, baseClusterTemplateYAML, upgrader)
+				},
+			})
+			if err != nil {
+				// Call Fail to notify ginkgo that the suit has failed.
+				// Ginkgo will print the first observed error failure in this case.
+				// Example: If cluster c1, c2 and c3 failed then ginkgo will only print the first
+				// observed failure among the these 3 clusters.
+				// Since ginkgo only captures one failure, to help with this we are logging the error
+				// that will contain the full stack trace of failure for each cluster to help with debugging.
+				// TODO(ykakarap): Follow-up: Explore options for improved error reporting.
+				log.Logf("Failed to upgrade clusters. Error: %s", err.Error())
+				Fail("")
+			}
+		}
+
 		// TODO(ykakarap): Follow-up: Dump resources for the failed clusters (creation).
 
 		clusterNamesToDelete := []string{}
@@ -494,11 +539,9 @@ func getClusterCreateAndWaitFn(input clusterctl.ApplyCustomClusterTemplateAndWai
 			CustomTemplateYAML:           clusterTemplateYAML,
 			ClusterName:                  clusterName,
 			Namespace:                    namespace,
-			CNIManifestPath:              input.CNIManifestPath,
 			WaitForClusterIntervals:      input.WaitForClusterIntervals,
 			WaitForControlPlaneIntervals: input.WaitForControlPlaneIntervals,
 			WaitForMachineDeployments:    input.WaitForMachineDeployments,
-			WaitForMachinePools:          input.WaitForMachinePools,
 			Args:                         input.Args,
 			PreWaitForCluster:            input.PreWaitForCluster,
 			PostMachinesProvisioned:      input.PostMachinesProvisioned,
@@ -594,7 +637,7 @@ func deleteClusterAndWaitWorker(ctx context.Context, inputChan <-chan string, re
 				return true
 			case clusterName, open := <-inputChan:
 				// Read the cluster name from the channel.
-				// If the channel is closed it implies there is not more work to be done. Return.
+				// If the channel is closed it implies there is no more work to be done. Return.
 				if !open {
 					return true
 				}
@@ -648,6 +691,111 @@ func deleteClusterAndWaitWorker(ctx context.Context, inputChan <-chan string, re
 	}
 }
 
+type clusterUpgrader func(ctx context.Context, namespace, clusterName string, clusterTemplateYAML []byte)
+
+func getClusterUpgradeAndWaitFn(input framework.UpgradeClusterTopologyAndWaitForUpgradeInput) clusterUpgrader {
+	return func(ctx context.Context, namespace, clusterName string, clusterTemplateYAML []byte) {
+		resources := getClusterResourcesForUpgrade(ctx, input.ClusterProxy.GetClient(), namespace, clusterName)
+		// Nb. We cannot directly modify and use `input` in this closure function because this function
+		// will be called multiple times and this closure will keep modifying the same `input` multiple
+		// times. It is safer to pass the values explicitly into `UpgradeClusterTopologyAndWaitForUpgradeInput`.
+		framework.UpgradeClusterTopologyAndWaitForUpgrade(ctx, framework.UpgradeClusterTopologyAndWaitForUpgradeInput{
+			ClusterProxy:                input.ClusterProxy,
+			Cluster:                     resources.cluster,
+			ControlPlane:                resources.controlPlane,
+			MachineDeployments:          resources.machineDeployments,
+			KubernetesUpgradeVersion:    input.KubernetesUpgradeVersion,
+			WaitForMachinesToBeUpgraded: input.WaitForMachinesToBeUpgraded,
+			WaitForKubeProxyUpgrade:     input.WaitForKubeProxyUpgrade,
+			WaitForDNSUpgrade:           input.WaitForDNSUpgrade,
+			WaitForEtcdUpgrade:          input.WaitForEtcdUpgrade,
+			// TODO: (killianmuldoon) Checking the kube-proxy, etcd and DNS version doesn't work as we can't access the control plane endpoint for the workload cluster
+			// from the host. Need to figure out a way to route the calls to the workload Cluster correctly.
+			EtcdImageTag:       "",
+			DNSImageTag:        "",
+			SkipKubeProxyCheck: true,
+		})
+	}
+}
+
+func upgradeClusterAndWaitWorker(ctx context.Context, inputChan <-chan string, resultChan chan<- workResult, wg *sync.WaitGroup, defaultNamespace string, deployClusterInSeparateNamespaces bool, clusterTemplateYAML []byte, upgrade clusterUpgrader) {
+	defer wg.Done()
+
+	for {
+		done := func() bool {
+			select {
+			case <-ctx.Done():
+				// If the context is cancelled, return and shutdown the worker.
+				return true
+			case clusterName, open := <-inputChan:
+				// Read the cluster name from the channel.
+				// If the channel is closed it implies there is no more work to be done. Return.
+				if !open {
+					return true
+				}
+				log.Logf("Upgrading cluster %s", clusterName)
+
+				// This defer will catch ginkgo failures and record them.
+				// The recorded panics are then handled by the parent goroutine.
+				defer func() {
+					e := recover()
+					resultChan <- workResult{
+						clusterName: clusterName,
+						err:         e,
+					}
+				}()
+
+				// Calculate namespace.
+				namespaceName := defaultNamespace
+				if deployClusterInSeparateNamespaces {
+					namespaceName = clusterName
+				}
+				upgrade(ctx, namespaceName, clusterName, clusterTemplateYAML)
+				return false
+			}
+		}()
+		if done {
+			break
+		}
+	}
+}
+
+type clusterResources struct {
+	cluster            *clusterv1.Cluster
+	machineDeployments []*clusterv1.MachineDeployment
+	controlPlane       *controlplanev1.KubeadmControlPlane
+}
+
+func getClusterResourcesForUpgrade(ctx context.Context, c client.Client, namespace, clusterName string) clusterResources {
+	cluster := &clusterv1.Cluster{}
+	err := c.Get(ctx, client.ObjectKey{Namespace: namespace, Name: clusterName}, cluster)
+	Expect(err).ToNot(HaveOccurred(), fmt.Sprintf("error getting Cluster %s: %s", klog.KRef(namespace, clusterName), err))
+
+	controlPlane := &controlplanev1.KubeadmControlPlane{}
+	err = c.Get(ctx, client.ObjectKey{Namespace: namespace, Name: cluster.Spec.ControlPlaneRef.Name}, controlPlane)
+	Expect(err).ToNot(HaveOccurred(), fmt.Sprintf("error getting ControlPlane for Cluster %s: %s,", klog.KObj(cluster), err))
+
+	mds := []*clusterv1.MachineDeployment{}
+	machineDeployments := &clusterv1.MachineDeploymentList{}
+	err = c.List(ctx, machineDeployments,
+		client.MatchingLabels{
+			clusterv1.ClusterNameLabel:          cluster.Name,
+			clusterv1.ClusterTopologyOwnedLabel: "",
+		},
+		client.InNamespace(cluster.Namespace),
+	)
+	Expect(err).ToNot(HaveOccurred(), fmt.Sprintf("error getting MachineDeployments for Cluster %s: %s", klog.KObj(cluster), err))
+	for _, md := range machineDeployments.Items {
+		mds = append(mds, md.DeepCopy())
+	}
+
+	return clusterResources{
+		cluster:            cluster,
+		machineDeployments: mds,
+		controlPlane:       controlPlane,
+	}
+}
+
 type workResult struct {
 	clusterName string
 	err         any

test/e2e/scale_test.go

@@ -25,6 +25,7 @@ import (
 )
 
 var _ = Describe("When testing the machinery for scale testing using in-memory provider", func() {
+	// Note: This test does not support MachinePools.
 	scaleSpec(ctx, func() scaleSpecInput {
 		return scaleSpecInput{
 			E2EConfig:                e2eConfig,

test/framework/cluster_topology_helpers.go

@@ -75,6 +75,7 @@ type UpgradeClusterTopologyAndWaitForUpgradeInput struct {
 	WaitForEtcdUpgrade                 []interface{}
 	PreWaitForControlPlaneToBeUpgraded func()
 	PreWaitForWorkersToBeUpgraded      func()
+	SkipKubeProxyCheck                 bool
 }
 
 // UpgradeClusterTopologyAndWaitForUpgrade upgrades a Cluster topology and waits for it to be upgraded.
@@ -123,13 +124,16 @@ func UpgradeClusterTopologyAndWaitForUpgrade(ctx context.Context, input UpgradeC
 		KubernetesUpgradeVersion: input.KubernetesUpgradeVersion,
 	}, input.WaitForMachinesToBeUpgraded...)
 
-	log.Logf("Waiting for kube-proxy to have the upgraded Kubernetes version")
 	workloadCluster := input.ClusterProxy.GetWorkloadCluster(ctx, input.Cluster.Namespace, input.Cluster.Name)
 	workloadClient := workloadCluster.GetClient()
-	WaitForKubeProxyUpgrade(ctx, WaitForKubeProxyUpgradeInput{
-		Getter:            workloadClient,
-		KubernetesVersion: input.KubernetesUpgradeVersion,
-	}, input.WaitForKubeProxyUpgrade...)
+
+	if !input.SkipKubeProxyCheck {
+		log.Logf("Waiting for kube-proxy to have the upgraded Kubernetes version")
+		WaitForKubeProxyUpgrade(ctx, WaitForKubeProxyUpgradeInput{
+			Getter:            workloadClient,
+			KubernetesVersion: input.KubernetesUpgradeVersion,
+		}, input.WaitForKubeProxyUpgrade...)
+	}
 
 	// Wait for the CoreDNS upgrade if the DNSImageTag is set.
 	if input.DNSImageTag != "" {

test/framework/controlplane_helpers.go

@@ -393,13 +393,6 @@ func UpgradeControlPlaneAndWaitForUpgrade(ctx context.Context, input UpgradeCont
 	}
 }
 
-// controlPlaneMachineOptions returns a set of ListOptions that allows to get all machine objects belonging to control plane.
-func controlPlaneMachineOptions() []client.ListOption {
-	return []client.ListOption{
-		client.HasLabels{clusterv1.MachineControlPlaneLabel},
-	}
-}
-
 type ScaleAndWaitControlPlaneInput struct {
 	ClusterProxy        ClusterProxy
 	Cluster             *clusterv1.Cluster

test/framework/machine_helpers.go

@@ -127,7 +127,13 @@ func GetControlPlaneMachinesByCluster(ctx context.Context, input GetControlPlane
 	Expect(input.ClusterName).ToNot(BeEmpty(), "Invalid argument. input.ClusterName can't be empty when calling GetControlPlaneMachinesByCluster")
 	Expect(input.Namespace).ToNot(BeEmpty(), "Invalid argument. input.Namespace can't be empty when calling GetControlPlaneMachinesByCluster")
 
-	options := append(byClusterOptions(input.ClusterName, input.Namespace), controlPlaneMachineOptions()...)
+	options := []client.ListOption{
+		client.InNamespace(input.Namespace),
+		client.MatchingLabels{
+			clusterv1.ClusterNameLabel:         input.ClusterName,
+			clusterv1.MachineControlPlaneLabel: "",
+		},
+	}
 
 	machineList := &clusterv1.MachineList{}
 	Eventually(func() error {

test/infrastructure/inmemory/pkg/server/api/handler.go

@@ -35,6 +35,7 @@ import (
 	"k8s.io/apimachinery/pkg/apis/meta/v1/unstructured"
 	"k8s.io/apimachinery/pkg/apis/meta/v1/validation"
 	"k8s.io/apimachinery/pkg/fields"
+	"k8s.io/apimachinery/pkg/labels"
 	"k8s.io/apimachinery/pkg/runtime"
 	"k8s.io/apimachinery/pkg/runtime/schema"
 	"k8s.io/apimachinery/pkg/runtime/serializer"
@@ -317,16 +318,24 @@ func (h *apiServerHandler) apiV1List(req *restful.Request, resp *restful.Respons
 		listOpts = append(listOpts, client.InNamespace(req.PathParameter("namespace")))
 	}
 
-	// TODO: The only field selector which works is for `spec.nodeName` on pods.
-	selector, err := fields.ParseSelector(req.QueryParameter("fieldSelector"))
+	// TODO: The only field Selector which works is for `spec.nodeName` on pods.
+	fieldSelector, err := fields.ParseSelector(req.QueryParameter("fieldSelector"))
 	if err != nil {
 		_ = resp.WriteErrorString(http.StatusInternalServerError, err.Error())
 		return
 	}
-	if selector != nil {
-		listOpts = append(listOpts, client.MatchingFieldsSelector{Selector: selector})
+	if fieldSelector != nil {
+		listOpts = append(listOpts, client.MatchingFieldsSelector{Selector: fieldSelector})
 	}
 
+	labelSelector, err := labels.Parse(req.QueryParameter("labelSelector"))
+	if err != nil {
+		_ = resp.WriteErrorString(http.StatusInternalServerError, err.Error())
+		return
+	}
+	if labelSelector != nil {
+		listOpts = append(listOpts, client.MatchingLabelsSelector{Selector: labelSelector})
+	}
 	if err := inmemoryClient.List(ctx, list, listOpts...); err != nil {
 		if status, ok := err.(apierrors.APIStatus); ok || errors.As(err, &status) {
 			_ = resp.WriteHeaderAndEntity(int(status.Status().Code), status)

test/infrastructure/inmemory/pkg/server/mux_test.go

@@ -36,6 +36,7 @@ import (
 	rbacv1 "k8s.io/api/rbac/v1"
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 	"k8s.io/apimachinery/pkg/fields"
+	"k8s.io/apimachinery/pkg/labels"
 	"k8s.io/apimachinery/pkg/runtime"
 	"k8s.io/klog/v2"
 	ctrl "sigs.k8s.io/controller-runtime"
@@ -155,7 +156,7 @@ func TestAPI_corev1_CRUD(t *testing.T) {
 		Spec:       corev1.PodSpec{NodeName: n.Name},
 	})).To(Succeed())
 	g.Expect(c.Create(ctx, &corev1.Pod{
-		ObjectMeta: metav1.ObjectMeta{Name: "notSelectedPod", Namespace: metav1.NamespaceDefault},
+		ObjectMeta: metav1.ObjectMeta{Name: "labelSelectedPod", Namespace: metav1.NamespaceDefault, Labels: map[string]string{"name": "labelSelectedPod"}},
 	})).To(Succeed())
 
 	pl := &corev1.PodList{}
@@ -166,8 +167,15 @@ func TestAPI_corev1_CRUD(t *testing.T) {
 	g.Expect(pl.Items).To(HaveLen(1))
 	g.Expect(pl.Items[0].Name).To(Equal("bar"))
 
-	// get
+	// list with label selector on pod
+	labelSelector := &client.ListOptions{
+		LabelSelector: labels.SelectorFromSet(labels.Set{"name": "labelSelectedPod"}),
+	}
+	g.Expect(c.List(ctx, pl, labelSelector)).To(Succeed())
+	g.Expect(pl.Items).To(HaveLen(1))
+	g.Expect(pl.Items[0].Name).To(Equal("labelSelectedPod"))
 
+	// get
 	n = &corev1.Node{}
 	err = c.Get(ctx, client.ObjectKey{Name: "foo"}, n)
 	g.Expect(err).ToNot(HaveOccurred())