SKS-1390: Optimize the creation of virtual machines when insufficient storage detected for ELF cluster

api/v1beta1/conditions_consts.go

@@ -110,6 +110,10 @@ const (
 	// waiting for ELF cluster with sufficient memory to create or power on VM.
 	WaitingForELFClusterWithSufficientMemoryReason = "WaitingForELFClusterWithSufficientMemory"
 
+	// WaitingForELFClusterWithSufficientStorageReason (Severity=Info) documents an ElfMachine
+	// waiting for ELF cluster with sufficient storage to create or power on VM.
+	WaitingForELFClusterWithSufficientStorageReason = "WaitingForELFClusterWithSufficientStorage"
+
 	// WaitingForAvailableHostRequiredByPlacementGroupReason (Severity=Info) documents an ElfMachine
 	// waiting for an available host required by placement group to create VM.
 	WaitingForAvailableHostRequiredByPlacementGroupReason = "WaitingForAvailableHostRequiredByPlacementGroup"

controllers/elfmachine_controller.go

@@ -922,6 +922,7 @@ func (r *ElfMachineReconciler) reconcileVMTask(ctx *context.MachineContext, vm *
 
 		if service.IsCloneVMTask(task) || service.IsPowerOnVMTask(task) {
 			releaseTicketForCreateVM(ctx.ElfMachine.Name)
+			recordElfClusterStorageInsufficient(ctx, false)
 			recordElfClusterMemoryInsufficient(ctx, false)
 
 			if err := recordPlacementGroupPolicyNotSatisfied(ctx, false); err != nil {
@@ -958,17 +959,24 @@ func (r *ElfMachineReconciler) reconcileVMFailedTask(ctx *context.MachineContext
 	case service.IsCloneVMTask(task):
 		releaseTicketForCreateVM(ctx.ElfMachine.Name)
 
-		if service.IsVMDuplicateError(errorMessage) {
-			setVMDuplicate(ctx.ElfMachine.Name)
-		}
-
 		if ctx.ElfMachine.RequiresGPUDevices() {
 			unlockGPUDevicesLockedByVM(ctx.ElfCluster.Spec.Cluster, ctx.ElfMachine.Name)
 		}
 	case service.IsPowerOnVMTask(task) || service.IsUpdateVMTask(task) || service.IsVMColdMigrationTask(task):
 		if ctx.ElfMachine.RequiresGPUDevices() {
 			unlockGPUDevicesLockedByVM(ctx.ElfCluster.Spec.Cluster, ctx.ElfMachine.Name)
 		}
+	}
+
+	switch {
+	case service.IsVMDuplicateError(errorMessage):
+		setVMDuplicate(ctx.ElfMachine.Name)
+	case service.IsStorageInsufficientError(errorMessage):
+		recordElfClusterStorageInsufficient(ctx, true)
+		message := fmt.Sprintf("Insufficient storage detected for the ELF cluster %s", ctx.ElfCluster.Spec.Cluster)
+		ctx.Logger.Info(message)
+
+		return errors.New(message)
 	case service.IsMemoryInsufficientError(errorMessage):
 		recordElfClusterMemoryInsufficient(ctx, true)
 		message := fmt.Sprintf("Insufficient memory detected for the ELF cluster %s", ctx.ElfCluster.Spec.Cluster)

controllers/elfmachine_controller_test.go

@@ -262,7 +262,7 @@ var _ = Describe("ElfMachineReconciler", func() {
 			resetMemoryCache()
 			vm := fake.NewTowerVM()
 			vm.Name = &elfMachine.Name
-			elfCluster.Spec.Cluster = clusterKey
+			elfCluster.Spec.Cluster = clusterInsufficientStorageKey
 			task := fake.NewTowerTask()
 			withTaskVM := fake.NewWithTaskVM(vm, task)
 			ctrlutil.AddFinalizer(elfMachine, infrav1.MachineFinalizer)
@@ -271,12 +271,26 @@ var _ = Describe("ElfMachineReconciler", func() {
 
 			machineContext := newMachineContext(ctrlContext, elfCluster, cluster, elfMachine, machine, mockVMService)
 			machineContext.VMService = mockVMService
-			recordIsUnmet(machineContext, clusterKey, true)
-			reconciler := &ElfMachineReconciler{ControllerContext: ctrlContext, NewVMService: mockNewVMService}
+			recordIsUnmet(machineContext, clusterInsufficientStorageKey, true)
+			mockVMService.EXPECT().GetVMPlacementGroup(gomock.Any()).Return(placementGroup, nil)
 			elfMachineKey := capiutil.ObjectKey(elfMachine)
+			reconciler := &ElfMachineReconciler{ControllerContext: ctrlContext, NewVMService: mockNewVMService}
 			result, err := reconciler.Reconcile(ctx, ctrl.Request{NamespacedName: elfMachineKey})
 			Expect(result.RequeueAfter).NotTo(BeZero())
 			Expect(err).NotTo(HaveOccurred())
+			Expect(logBuffer.String()).To(ContainSubstring("Insufficient storage detected for the ELF cluster"))
+			expireELFScheduleVMError(machineContext, clusterInsufficientStorageKey)
+
+			logBuffer.Reset()
+			elfCluster.Spec.Cluster = clusterInsufficientMemoryKey
+			ctrlContext = newCtrlContexts(elfCluster, cluster, elfMachine, machine, secret, md)
+			fake.InitOwnerReferences(ctrlContext, elfCluster, cluster, elfMachine, machine)
+			machineContext = newMachineContext(ctrlContext, elfCluster, cluster, elfMachine, machine, mockVMService)
+			recordIsUnmet(machineContext, clusterInsufficientMemoryKey, true)
+			reconciler = &ElfMachineReconciler{ControllerContext: ctrlContext, NewVMService: mockNewVMService}
+			result, err = reconciler.Reconcile(ctx, ctrl.Request{NamespacedName: elfMachineKey})
+			Expect(result.RequeueAfter).NotTo(BeZero())
+			Expect(err).NotTo(HaveOccurred())
 			Expect(logBuffer.String()).To(ContainSubstring("Insufficient memory detected for the ELF cluster"))
 
 			logBuffer = new(bytes.Buffer)
@@ -285,7 +299,7 @@ var _ = Describe("ElfMachineReconciler", func() {
 			mockVMService.EXPECT().Get(*vm.ID).Return(vm, nil)
 			mockVMService.EXPECT().GetTask(*task.ID).Return(task, nil)
 			mockVMService.EXPECT().GetVMPlacementGroup(gomock.Any()).Return(placementGroup, nil)
-			expireELFScheduleVMError(machineContext, clusterKey)
+			expireELFScheduleVMError(machineContext, clusterInsufficientMemoryKey)
 
 			reconciler = &ElfMachineReconciler{ControllerContext: ctrlContext, NewVMService: mockNewVMService}
 			result, err = reconciler.Reconcile(ctx, ctrl.Request{NamespacedName: elfMachineKey})
@@ -826,20 +840,20 @@ var _ = Describe("ElfMachineReconciler", func() {
 				vm := fake.NewTowerVM()
 				vm.Host = &models.NestedHost{ID: service.TowerString(fake.ID())}
 				elfMachine.Status.VMRef = *vm.LocalID
-				elfCluster.Spec.Cluster = clusterKey
+				elfCluster.Spec.Cluster = clusterInsufficientMemoryKey
 				ctrlContext := newCtrlContexts(elfCluster, cluster, elfMachine, machine, secret, md)
 				fake.InitOwnerReferences(ctrlContext, elfCluster, cluster, elfMachine, machine)
 				machineContext := newMachineContext(ctrlContext, elfCluster, cluster, elfMachine, machine, mockVMService)
 				machineContext.VMService = mockVMService
-				recordIsUnmet(machineContext, clusterKey, true)
+				recordIsUnmet(machineContext, clusterInsufficientMemoryKey, true)
 				reconciler := &ElfMachineReconciler{ControllerContext: ctrlContext, NewVMService: mockNewVMService}
 				err := reconciler.powerOnVM(machineContext, vm)
 				Expect(err).NotTo(HaveOccurred())
 				Expect(logBuffer.String()).To(ContainSubstring("Insufficient memory detected for the ELF cluster"))
 
 				task := fake.NewTowerTask()
 				mockVMService.EXPECT().PowerOn(elfMachine.Status.VMRef, "").Return(task, nil)
-				expireELFScheduleVMError(machineContext, clusterKey)
+				expireELFScheduleVMError(machineContext, clusterInsufficientMemoryKey)
 				err = reconciler.powerOnVM(machineContext, vm)
 				Expect(err).NotTo(HaveOccurred())
 				Expect(logBuffer.String()).To(ContainSubstring("and the retry silence period passes, will try to power on the VM again"))
@@ -3185,11 +3199,28 @@ var _ = Describe("ElfMachineReconciler", func() {
 			Expect(err.Error()).To(ContainSubstring("The placement group policy can not be satisfied"))
 			Expect(logBuffer.String()).To(ContainSubstring("VM task failed"))
 
+			logBuffer.Reset()
 			ok, msg, err := isELFScheduleVMErrorRecorded(machineContext)
 			Expect(ok).To(BeTrue())
 			Expect(msg).To(ContainSubstring("Insufficient memory detected for the ELF cluster"))
 			Expect(err).ShouldNot(HaveOccurred())
 
+			resetMemoryCache()
+			logBuffer.Reset()
+			elfMachine.Status.TaskRef = *task.ID
+			task.ErrorMessage = service.TowerString(service.StorageInsufficientError)
+			ok, err = reconciler.reconcileVMTask(machineContext, nil)
+			Expect(ok).Should(BeTrue())
+			Expect(err.Error()).To(ContainSubstring("Insufficient storage detected for the ELF cluster"))
+			Expect(elfMachine.Status.TaskRef).To(Equal(""))
+			Expect(logBuffer.String()).To(ContainSubstring("VM task failed"))
+
+			logBuffer.Reset()
+			ok, msg, err = isELFScheduleVMErrorRecorded(machineContext)
+			Expect(ok).To(BeTrue())
+			Expect(msg).To(ContainSubstring("Insufficient storage detected for the ELF cluster"))
+			Expect(err).ShouldNot(HaveOccurred())
+
 			// Start VM
 			task.Status = models.NewTaskStatus(models.TaskStatusSUCCESSED)
 			task.Description = service.TowerString("Start VM")

controllers/tower_cache.go

@@ -50,7 +50,8 @@ type clusterResource struct {
 //
 // Includes these scenarios:
 // 1. ELF cluster has insufficient memory.
-// 2. Placement group not satisfy policy.
+// 2. ELF cluster has insufficient storage.
+// 3. Placement group not satisfy policy.
 func isELFScheduleVMErrorRecorded(ctx *context.MachineContext) (bool, string, error) {
 	lock.Lock()
 	defer lock.Unlock()
@@ -59,6 +60,10 @@ func isELFScheduleVMErrorRecorded(ctx *context.MachineContext) (bool, string, er
 		conditions.MarkFalse(ctx.ElfMachine, infrav1.VMProvisionedCondition, infrav1.WaitingForELFClusterWithSufficientMemoryReason, clusterv1.ConditionSeverityInfo, "")
 
 		return true, fmt.Sprintf("Insufficient memory detected for the ELF cluster %s", ctx.ElfCluster.Spec.Cluster), nil
+	} else if resource := getClusterResource(getKeyForInsufficientStorageError(ctx.ElfCluster.Spec.Cluster)); resource != nil {
+		conditions.MarkFalse(ctx.ElfMachine, infrav1.VMProvisionedCondition, infrav1.WaitingForELFClusterWithSufficientStorageReason, clusterv1.ConditionSeverityInfo, "")
+
+		return true, fmt.Sprintf("Insufficient storage detected for the ELF cluster %s", ctx.ElfCluster.Spec.Cluster), nil
 	}
 
 	placementGroupName, err := towerresources.GetVMPlacementGroupName(ctx, ctx.Client, ctx.Machine, ctx.Cluster)
@@ -85,6 +90,16 @@ func recordElfClusterMemoryInsufficient(ctx *context.MachineContext, isInsuffici
 	}
 }
 
+// recordElfClusterStorageInsufficient records whether the storage is insufficient.
+func recordElfClusterStorageInsufficient(ctx *context.MachineContext, isError bool) {
+	key := getKeyForInsufficientStorageError(ctx.ElfCluster.Spec.Cluster)
+	if isError {
+		inMemoryCache.Set(key, newClusterResource(), resourceDuration)
+	} else {
+		inMemoryCache.Delete(key)
+	}
+}
+
 // recordPlacementGroupPolicyNotSatisfied records whether the placement group not satisfy policy.
 func recordPlacementGroupPolicyNotSatisfied(ctx *context.MachineContext, isPGPolicyNotSatisfied bool) error {
 	placementGroupName, err := towerresources.GetVMPlacementGroupName(ctx, ctx.Client, ctx.Machine, ctx.Cluster)
@@ -116,7 +131,9 @@ func canRetryVMOperation(ctx *context.MachineContext) (bool, error) {
 	lock.Lock()
 	defer lock.Unlock()
 
-	if ok := canRetry(getKeyForInsufficientMemoryError(ctx.ElfCluster.Spec.Cluster)); ok {
+	if ok := canRetry(getKeyForInsufficientStorageError(ctx.ElfCluster.Spec.Cluster)); ok {
+		return true, nil
+	} else if ok := canRetry(getKeyForInsufficientMemoryError(ctx.ElfCluster.Spec.Cluster)); ok {
 		return true, nil
 	}
 
@@ -158,6 +175,10 @@ func getClusterResource(key string) *clusterResource {
 	return nil
 }
 
+func getKeyForInsufficientStorageError(clusterID string) string {
+	return fmt.Sprintf("insufficient:storage:%s", clusterID)
+}
+
 func getKeyForInsufficientMemoryError(clusterID string) string {
 	return fmt.Sprintf("insufficient:memory:%s", clusterID)
 }

controllers/tower_cache_test.go

@@ -34,8 +34,10 @@ import (
 )
 
 const (
-	clusterKey        = "clusterID"
-	placementGroupKey = "getPlacementGroupName"
+	clusterKey                    = "clusterID"
+	clusterInsufficientMemoryKey  = "clusterInsufficientMemory"
+	clusterInsufficientStorageKey = "clusterInsufficientStorage"
+	placementGroupKey             = "getPlacementGroupName"
 )
 
 var _ = Describe("TowerCache", func() {
@@ -44,7 +46,7 @@ var _ = Describe("TowerCache", func() {
 	})
 
 	It("should set memoryInsufficient/policyNotSatisfied", func() {
-		for _, name := range []string{clusterKey, placementGroupKey} {
+		for _, name := range []string{clusterInsufficientMemoryKey, clusterInsufficientStorageKey, placementGroupKey} {
 			resetMemoryCache()
 			elfCluster, cluster, elfMachine, machine, secret := fake.NewClusterAndMachineObjects()
 			elfCluster.Spec.Cluster = name
@@ -98,7 +100,7 @@ var _ = Describe("TowerCache", func() {
 	})
 
 	It("should return whether need to detect", func() {
-		for _, name := range []string{clusterKey, placementGroupKey} {
+		for _, name := range []string{clusterInsufficientMemoryKey, clusterInsufficientStorageKey, placementGroupKey} {
 			resetMemoryCache()
 			elfCluster, cluster, elfMachine, machine, secret := fake.NewClusterAndMachineObjects()
 			elfCluster.Spec.Cluster = name
@@ -154,13 +156,23 @@ var _ = Describe("TowerCache", func() {
 		Expect(err).ShouldNot(HaveOccurred())
 		expectConditions(elfMachine, []conditionAssertion{})
 
-		recordIsUnmet(machineContext, clusterKey, true)
+		elfCluster.Spec.Cluster = clusterInsufficientMemoryKey
+		recordIsUnmet(machineContext, clusterInsufficientMemoryKey, true)
 		ok, msg, err = isELFScheduleVMErrorRecorded(machineContext)
 		Expect(ok).To(BeTrue())
 		Expect(msg).To(ContainSubstring("Insufficient memory detected for the ELF cluster"))
 		Expect(err).ShouldNot(HaveOccurred())
 		expectConditions(elfMachine, []conditionAssertion{{infrav1.VMProvisionedCondition, corev1.ConditionFalse, clusterv1.ConditionSeverityInfo, infrav1.WaitingForELFClusterWithSufficientMemoryReason}})
 
+		resetMemoryCache()
+		elfCluster.Spec.Cluster = clusterInsufficientStorageKey
+		recordIsUnmet(machineContext, clusterInsufficientStorageKey, true)
+		ok, msg, err = isELFScheduleVMErrorRecorded(machineContext)
+		Expect(ok).To(BeTrue())
+		Expect(msg).To(ContainSubstring("Insufficient storage detected for the ELF cluster clusterInsufficientStorage"))
+		Expect(err).ShouldNot(HaveOccurred())
+		expectConditions(elfMachine, []conditionAssertion{{infrav1.VMProvisionedCondition, corev1.ConditionFalse, clusterv1.ConditionSeverityInfo, infrav1.WaitingForELFClusterWithSufficientStorageReason}})
+
 		resetMemoryCache()
 		recordIsUnmet(machineContext, placementGroupKey, true)
 		ok, msg, err = isELFScheduleVMErrorRecorded(machineContext)
@@ -220,8 +232,10 @@ func removeGPUVMInfosCache(gpuIDs []string) {
 }
 
 func getKey(ctx *context.MachineContext, name string) string {
-	if name == clusterKey {
+	if name == clusterInsufficientMemoryKey {
 		return getKeyForInsufficientMemoryError(name)
+	} else if name == clusterInsufficientStorageKey {
+		return getKeyForInsufficientStorageError(name)
 	}
 
 	placementGroupName, err := towerresources.GetVMPlacementGroupName(ctx, ctx.Client, ctx.Machine, ctx.Cluster)
@@ -231,9 +245,12 @@ func getKey(ctx *context.MachineContext, name string) string {
 }
 
 func recordIsUnmet(ctx *context.MachineContext, key string, isUnmet bool) {
-	if strings.Contains(key, clusterKey) {
+	if strings.Contains(key, clusterInsufficientMemoryKey) {
 		recordElfClusterMemoryInsufficient(ctx, isUnmet)
 		return
+	} else if strings.Contains(key, clusterInsufficientStorageKey) {
+		recordElfClusterStorageInsufficient(ctx, isUnmet)
+		return
 	}
 
 	Expect(recordPlacementGroupPolicyNotSatisfied(ctx, isUnmet)).ShouldNot(HaveOccurred())

pkg/service/errors.go

@@ -37,6 +37,7 @@
 	LabelAddFailed            = "LABEL_ADD_FAILED"
 	CloudInitError            = "VM_CLOUD_INIT_CONFIG_ERROR"
 	MemoryInsufficientError   = "HostAvailableMemoryFilter"
+	StorageInsufficientError  = "EAllocSpace"
 	PlacementGroupError       = "PlacementGroupFilter" // SMTX OS <= 5.0.4
 	PlacementGroupMustError   = "PlacementGroupMustFilter"
 	PlacementGroupPriorError  = "PlacementGroupPriorFilter"
@@ -111,6 +112,10 @@
 	return message[index:]
 }
 
+func IsStorageInsufficientError(message string) bool {
+	return strings.Contains(message, StorageInsufficientError)
+}
+
 func IsMemoryInsufficientError(message string) bool {
 	return strings.Contains(message, MemoryInsufficientError)
 }