Bump CAPI to v1.7.5

.go-version

@@ -1 +1 @@
-1.21.11
+1.22.5

.golangci.yml

@@ -104,9 +104,9 @@ linters-settings:
       - pkg: github.com/smartxworks/cluster-api-provider-elf/api/v1beta1
         alias: infrav1
   staticcheck:
-    go: "1.21"
+    go: "1.22"
   stylecheck:
-    go: "1.21"
+    go: "1.22"
 issues:
   max-same-issues: 0
   max-issues-per-linter: 0
@@ -169,7 +169,7 @@ issues:
 
 run:
   timeout: 10m
-  go: "1.21"
+  go: "1.22"
   skip-files:
   - "zz_generated.*\\.go$"
   allow-parallel-runners: true

Dockerfile

@@ -18,7 +18,7 @@
 ARG ARCH
 
 # Build the manager binary
-FROM golang:1.21.11 as builder
+FROM golang:1.22.5 as builder
 WORKDIR /workspace
 
 # Run this with docker build --build_arg $(go env GOPROXY) to override the goproxy

Makefile

@@ -25,7 +25,7 @@ VERSION ?= $(shell cat clusterctl-settings.json | jq .config.nextVersion -r)
 #
 # Go.
 #
-GO_VERSION ?= 1.21.11
+GO_VERSION ?= 1.22.5
 
 # Get the currently used golang install path (in GOPATH/bin, unless GOBIN is set)
 ifeq (,$(shell go env GOBIN))
@@ -151,23 +151,23 @@ test-e2e: ginkgo kustomize kind ## Run e2e tests
 
 KUSTOMIZE = $(shell pwd)/bin/kustomize
 kustomize: ## Download kustomize locally if necessary.
-	$(call go-get-tool,$(KUSTOMIZE),sigs.k8s.io/kustomize/kustomize/[email protected])
+	$(call go-get-tool,$(KUSTOMIZE),sigs.k8s.io/kustomize/kustomize/[email protected])
 
 CONTROLLER_GEN = $(shell pwd)/bin/controller-gen
 controller-gen: ## Download controller-gen locally if necessary.
 	$(call go-get-tool,$(CONTROLLER_GEN),sigs.k8s.io/controller-tools/cmd/[email protected])
 
 GINKGO := $(shell pwd)/bin/ginkgo
 ginkgo: ## Download ginkgo locally if necessary.
-	$(call go-get-tool,$(GINKGO),github.com/onsi/ginkgo/v2/ginkgo@v2.13.1)
+	$(call go-get-tool,$(GINKGO),github.com/onsi/ginkgo/v2/ginkgo@v2.17.1)
 
 KIND := $(shell pwd)/bin/kind
 kind: ## Download kind locally if necessary.
-	$(call go-get-tool,$(KIND),sigs.k8s.io/kind@v0.20.0)
+	$(call go-get-tool,$(KIND),sigs.k8s.io/kind@v0.22.0)
 
 GOLANGCI_LINT := $(shell pwd)/bin/golangci-lint
 golangci-lint: ## Download golangci-lint locally if necessary.
-	$(call go-get-tool,$(GOLANGCI_LINT),github.com/golangci/golangci-lint/cmd/golangci-lint@v1.55.2)
+	$(call go-get-tool,$(GOLANGCI_LINT),github.com/golangci/golangci-lint/cmd/golangci-lint@v1.57.2)
 
 ## --------------------------------------
 ## Linting and fixing linter errors

api/v1beta1/types.go

@@ -104,7 +104,7 @@ type NetworkSpec struct {
 }
 
 func (n *NetworkSpec) RequiresStaticIPs() bool {
-	for i := 0; i < len(n.Devices); i++ {
+	for i := range len(n.Devices) {
 		if n.Devices[i].NetworkType == NetworkTypeIPV4 && len(n.Devices[i].IPAddrs) == 0 {
 			return true
 		}

controllers/elfcluster_controller.go

@@ -285,7 +285,7 @@ func (r *ElfClusterReconciler) cleanOrphanLabels(ctx goctx.Context, clusterCtx *
 	keys := []string{towerresources.GetVMLabelClusterName(), towerresources.GetVMLabelVIP(), towerresources.GetVMLabelNamespace()}
 	labelIDs, err := clusterCtx.VMService.CleanUnusedLabels(keys)
 	if err != nil {
-		log.Error(err, fmt.Sprintf("Warning: failed to clean orphan labels in Tower %s", clusterCtx.ElfCluster.Spec.Tower.Server))
+		log.Error(err, "Warning: failed to clean orphan labels in Tower "+clusterCtx.ElfCluster.Spec.Tower.Server)
 
 		return
 	}

controllers/elfcluster_controller_test.go

@@ -270,7 +270,7 @@ var _ = Describe("ElfClusterReconciler", func() {
 			mockVMService.EXPECT().CleanUnusedLabels(keys).Return(nil, unexpectedError)
 			reconciler := &ElfClusterReconciler{ControllerManagerContext: ctrlMgrCtx, NewVMService: mockNewVMService}
 			reconciler.cleanOrphanLabels(ctx, clusterCtx)
-			Expect(logBuffer.String()).To(ContainSubstring(fmt.Sprintf("Warning: failed to clean orphan labels in Tower %s", elfCluster.Spec.Tower.Server)))
+			Expect(logBuffer.String()).To(ContainSubstring("Warning: failed to clean orphan labels in Tower " + elfCluster.Spec.Tower.Server))
 
 			logBuffer.Reset()
 			mockVMService.EXPECT().CleanUnusedLabels(keys).Return(nil, nil)

controllers/elfmachine_controller.go

@@ -31,9 +31,8 @@ import (
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
 	apitypes "k8s.io/apimachinery/pkg/types"
 	"k8s.io/klog/v2"
-	"k8s.io/utils/pointer"
+	"k8s.io/utils/ptr"
 	clusterv1 "sigs.k8s.io/cluster-api/api/v1beta1"
-	capierrors "sigs.k8s.io/cluster-api/errors"
 	capiutil "sigs.k8s.io/cluster-api/util"
 	"sigs.k8s.io/cluster-api/util/annotations"
 	"sigs.k8s.io/cluster-api/util/conditions"
@@ -540,15 +539,15 @@ func (r *ElfMachineReconciler) reconcileVM(ctx goctx.Context, machineCtx *contex
 			if canRetry, err := canRetryVMOperation(ctx, machineCtx, r.Client); err != nil {
 				return nil, false, err
 			} else if !canRetry {
-				log.V(1).Info(fmt.Sprintf("%s, skip creating VM", message))
+				log.V(1).Info(message + ", skip creating VM")
 				return nil, false, nil
 			}
 
-			log.V(1).Info(fmt.Sprintf("%s and the retry silence period passes, will try to create the VM again", message))
+			log.V(1).Info(message + " and the retry silence period passes, will try to create the VM again")
 		}
 
 		if ok, msg := acquireTicketForCreateVM(machineCtx.ElfMachine.Name, machineutil.IsControlPlaneMachine(machineCtx.ElfMachine)); !ok {
-			log.V(1).Info(fmt.Sprintf("%s, skip creating VM", msg))
+			log.V(1).Info(msg + ", skip creating VM")
 			return nil, false, nil
 		}
 
@@ -644,8 +643,9 @@ func (r *ElfMachineReconciler) reconcileVM(ctx goctx.Context, machineCtx *contex
 	// The VM was moved to the recycle bin. Treat the VM as deleted, and will not reconganize it even if it's moved back from the recycle bin.
 	if service.IsVMInRecycleBin(vm) {
 		message := fmt.Sprintf("The VM %s was moved to the Tower recycle bin by users, so treat it as deleted.", machineCtx.ElfMachine.Status.VMRef)
-		machineCtx.ElfMachine.Status.FailureReason = capierrors.MachineStatusErrorPtr(capeerrors.MovedToRecycleBinError)
-		machineCtx.ElfMachine.Status.FailureMessage = pointer.String(message)
+		machineStatusError := capeerrors.MovedToRecycleBinError
+		machineCtx.ElfMachine.Status.FailureReason = &machineStatusError
+		machineCtx.ElfMachine.Status.FailureMessage = ptr.To(message)
 		machineCtx.ElfMachine.SetVM("")
 		log.Error(stderrors.New(message), "")
 
@@ -700,8 +700,9 @@ func (r *ElfMachineReconciler) getVM(ctx goctx.Context, machineCtx *context.Mach
 		}
 
 		// If the machine was not found by UUID and timed out it means that it got deleted directly
-		machineCtx.ElfMachine.Status.FailureReason = capierrors.MachineStatusErrorPtr(capeerrors.RemovedFromInfrastructureError)
-		machineCtx.ElfMachine.Status.FailureMessage = pointer.String(fmt.Sprintf("Unable to find VM by UUID %s. The VM was removed from infrastructure.", machineCtx.ElfMachine.Status.VMRef))
+		machineStatusError := capeerrors.RemovedFromInfrastructureError
+		machineCtx.ElfMachine.Status.FailureReason = &machineStatusError
+		machineCtx.ElfMachine.Status.FailureMessage = ptr.To(fmt.Sprintf("Unable to find VM by UUID %s. The VM was removed from infrastructure.", machineCtx.ElfMachine.Status.VMRef))
 		log.Error(err, fmt.Sprintf("failed to get VM by UUID %s in %s", machineCtx.ElfMachine.Status.VMRef, infrav1.VMDisconnectionTimeout.String()), "message", machineCtx.ElfMachine.Status.FailureMessage)
 
 		return nil, err
@@ -779,7 +780,7 @@ func (r *ElfMachineReconciler) reconcileVMStatus(ctx goctx.Context, machineCtx *
 		// See issue http://jira.smartx.com/browse/SKS-1351 for details.
 		return false, r.powerOffVM(ctx, machineCtx)
 	default:
-		log.Info(fmt.Sprintf("The VM is in an unexpected status %s", string(*vm.Status)), "vmRef", machineCtx.ElfMachine.Status.VMRef)
+		log.Info("The VM is in an unexpected status "+string(*vm.Status), "vmRef", machineCtx.ElfMachine.Status.VMRef)
 
 		return false, nil
 	}
@@ -814,7 +815,7 @@ func (r *ElfMachineReconciler) powerOffVM(ctx goctx.Context, machineCtx *context
 	log := ctrl.LoggerFrom(ctx)
 
 	if ok := acquireTicketForUpdatingVM(machineCtx.ElfMachine.Name); !ok {
-		log.V(1).Info(fmt.Sprintf("The VM operation reaches rate limit, skip powering off VM %s", machineCtx.ElfMachine.Status.VMRef))
+		log.V(1).Info("The VM operation reaches rate limit, skip powering off VM " + machineCtx.ElfMachine.Status.VMRef)
 
 		return nil
 	}
@@ -849,11 +850,11 @@ func (r *ElfMachineReconciler) powerOnVM(ctx goctx.Context, machineCtx *context.
 			return nil
 		}
 
-		log.V(1).Info(fmt.Sprintf("%s and the retry silence period passes, will try to power on the VM again", message))
+		log.V(1).Info(message + " and the retry silence period passes, will try to power on the VM again")
 	}
 
 	if ok := acquireTicketForUpdatingVM(machineCtx.ElfMachine.Name); !ok {
-		log.V(1).Info(fmt.Sprintf("The VM operation reaches rate limit, skip power on VM %s", machineCtx.ElfMachine.Status.VMRef))
+		log.V(1).Info("The VM operation reaches rate limit, skip power on VM " + machineCtx.ElfMachine.Status.VMRef)
 
 		return nil
 	}
@@ -885,7 +886,7 @@ func (r *ElfMachineReconciler) updateVM(ctx goctx.Context, machineCtx *context.M
 	log := ctrl.LoggerFrom(ctx)
 
 	if ok := acquireTicketForUpdatingVM(machineCtx.ElfMachine.Name); !ok {
-		log.V(1).Info(fmt.Sprintf("The VM operation reaches rate limit, skip updating VM %s", machineCtx.ElfMachine.Status.VMRef))
+		log.V(1).Info("The VM operation reaches rate limit, skip updating VM " + machineCtx.ElfMachine.Status.VMRef)
 
 		return nil
 	}
@@ -1005,8 +1006,9 @@ func (r *ElfMachineReconciler) reconcileVMFailedTask(ctx goctx.Context, machineC
 	conditions.MarkFalse(machineCtx.ElfMachine, infrav1.VMProvisionedCondition, infrav1.TaskFailureReason, clusterv1.ConditionSeverityInfo, errorMessage)
 
 	if service.IsCloudInitConfigError(errorMessage) {
-		machineCtx.ElfMachine.Status.FailureReason = capierrors.MachineStatusErrorPtr(capeerrors.CloudInitConfigError)
-		machineCtx.ElfMachine.Status.FailureMessage = pointer.String(fmt.Sprintf("VM cloud-init config error: %s", service.FormatCloudInitError(errorMessage)))
+		machineStatusError := capeerrors.CloudInitConfigError
+		machineCtx.ElfMachine.Status.FailureReason = &machineStatusError
+		machineCtx.ElfMachine.Status.FailureMessage = ptr.To("VM cloud-init config error: " + service.FormatCloudInitError(errorMessage))
 	}
 
 	log.Error(errors.New("VM task failed"), "", "vmRef", vmRef, "taskRef", taskRef, "taskErrorMessage", errorMessage, "taskErrorCode", service.GetTowerString(task.ErrorCode), "taskDescription", service.GetTowerString(task.Description))
@@ -1038,13 +1040,13 @@ func (r *ElfMachineReconciler) reconcileVMFailedTask(ctx goctx.Context, machineC
 		setVMDuplicate(machineCtx.ElfMachine.Name)
 	case service.IsStorageInsufficientError(errorMessage):
 		recordElfClusterStorageInsufficient(machineCtx, true)
-		message := fmt.Sprintf("Insufficient storage detected for the ELF cluster %s", machineCtx.ElfCluster.Spec.Cluster)
+		message := "Insufficient storage detected for the ELF cluster " + machineCtx.ElfCluster.Spec.Cluster
 		log.Info(message)
 
 		return errors.New(message)
 	case service.IsMemoryInsufficientError(errorMessage):
 		recordElfClusterMemoryInsufficient(machineCtx, true)
-		message := fmt.Sprintf("Insufficient memory detected for the ELF cluster %s", machineCtx.ElfCluster.Spec.Cluster)
+		message := "Insufficient memory detected for the ELF cluster " + machineCtx.ElfCluster.Spec.Cluster
 		log.Info(message)
 
 		return errors.New(message)
@@ -1075,7 +1077,7 @@ func (r *ElfMachineReconciler) reconcileProviderID(ctx goctx.Context, machineCtx
 	}
 
 	if machineCtx.ElfMachine.Spec.ProviderID == nil || *machineCtx.ElfMachine.Spec.ProviderID != providerID {
-		machineCtx.ElfMachine.Spec.ProviderID = pointer.String(providerID)
+		machineCtx.ElfMachine.Spec.ProviderID = ptr.To(providerID)
 
 		log.Info("updated providerID", "providerID", providerID)
 	}
@@ -1185,7 +1187,7 @@ func (r *ElfMachineReconciler) reconcileNetwork(ctx goctx.Context, machineCtx *c
 		return false, err
 	}
 
-	for i := 0; i < len(nics); i++ {
+	for i := range len(nics) {
 		nic := nics[i]
 		ip := service.GetTowerString(nic.IPAddress)
 
@@ -1281,7 +1283,7 @@ func (r *ElfMachineReconciler) reconcileLabels(ctx goctx.Context, machineCtx *co
 
 	// If the virtual machine has been labeled with managed label,
 	// it is considered that all labels have been labeled.
-	for i := 0; i < len(vm.Labels); i++ {
+	for i := range len(vm.Labels) {
 		if *vm.Labels[i].ID == *capeManagedLabel.ID {
 			return true, nil
 		}
@@ -1429,14 +1431,14 @@ func (r *ElfMachineReconciler) deleteDuplicateVMs(ctx goctx.Context, machineCtx
 
 	if machineCtx.ElfMachine.Status.VMRef == "" {
 		vmIDs := make([]string, 0, len(vms))
-		for i := 0; i < len(vms); i++ {
+		for i := range len(vms) {
 			vmIDs = append(vmIDs, *vms[i].ID)
 		}
 		log.Info("Waiting for ElfMachine to select one of the duplicate VMs before deleting the other", "vms", vmIDs)
 		return reconcile.Result{RequeueAfter: config.DefaultRequeueTimeout}, nil
 	}
 
-	for i := 0; i < len(vms); i++ {
+	for i := range len(vms) {
 		// Do not delete already running virtual machines to avoid deleting already used virtual machines.
 		if *vms[i].ID == machineCtx.ElfMachine.Status.VMRef ||
 			*vms[i].LocalID == machineCtx.ElfMachine.Status.VMRef ||

controllers/elfmachine_controller_gpu.go

@@ -22,7 +22,7 @@ import (
 	"github.com/pkg/errors"
 	"github.com/smartxworks/cloudtower-go-sdk/v2/models"
 	"k8s.io/apimachinery/pkg/util/sets"
-	"k8s.io/utils/pointer"
+	"k8s.io/utils/ptr"
 	clusterv1 "sigs.k8s.io/cluster-api/api/v1beta1"
 	"sigs.k8s.io/cluster-api/util/conditions"
 	ctrl "sigs.k8s.io/controller-runtime"
@@ -50,7 +50,7 @@ func (r *ElfMachineReconciler) selectHostAndGPUsForVM(ctx goctx.Context, machine
 	log := ctrl.LoggerFrom(ctx)
 
 	if !machineCtx.ElfMachine.RequiresGPUDevices() {
-		return pointer.String(""), nil, nil
+		return ptr.To(""), nil, nil
 	}
 
 	defer func() {
@@ -129,7 +129,7 @@ func (r *ElfMachineReconciler) selectHostAndGPUsForVM(ctx goctx.Context, machine
 		unsortedHostIDs = hostIDSet.UnsortedList()
 	}
 
-	for i := 0; i < len(unsortedHostIDs); i++ {
+	for i := range len(unsortedHostIDs) {
 		hostGPUVMInfos, ok := hostGPUVMInfoMap[unsortedHostIDs[i]]
 		if !ok {
 			continue
@@ -173,14 +173,14 @@ func selectGPUDevicesForVM(hostGPUVMInfos service.GPUVMInfos, requiredGPUDevices
 	})
 
 	var selectedGPUDeviceInfos []*service.GPUDeviceInfo
-	for i := 0; i < len(requiredGPUDevices); i++ {
+	for i := range len(requiredGPUDevices) {
 		gpuVMInfos, ok := modelGPUVMInfoMap[requiredGPUDevices[i].Model]
 		if !ok || len(gpuVMInfos) < int(requiredGPUDevices[i].Count) {
 			return nil
 		}
 
 		gpuInfos := gpuVMInfos[:int(requiredGPUDevices[i].Count)]
-		for j := 0; j < len(gpuInfos); j++ {
+		for j := range len(gpuInfos) {
 			selectedGPUDeviceInfos = append(selectedGPUDeviceInfos, &service.GPUDeviceInfo{ID: *gpuInfos[j].ID, AllocatedCount: 1, AvailableCount: 1})
 		}
 	}
@@ -202,15 +202,15 @@ func selectVGPUDevicesForVM(hostGPUVMInfos service.GPUVMInfos, requiredVGPUDevic
 	})
 
 	var selectedGPUDeviceInfos []*service.GPUDeviceInfo
-	for i := 0; i < len(requiredVGPUDevices); i++ {
+	for i := range len(requiredVGPUDevices) {
 		gpuVMInfos, ok := typeVGPUVMInfoMap[requiredVGPUDevices[i].Type]
 		if !ok {
 			return nil
 		}
 
 		var gpuInfos []*service.GPUDeviceInfo
 		requiredCount := requiredVGPUDevices[i].Count
-		for j := 0; j < len(gpuVMInfos); j++ {
+		for j := range len(gpuVMInfos) {
 			availableCount := service.GetAvailableCountFromGPUVMInfo(gpuVMInfos[j])
 			if availableCount <= 0 {
 				continue
@@ -249,7 +249,7 @@ func (r *ElfMachineReconciler) reconcileGPUDevices(ctx goctx.Context, machineCtx
 
 	// Ensure GPUStatus is set or up to date.
 	gpuDevices := make([]infrav1.GPUStatus, len(vm.GpuDevices))
-	for i := 0; i < len(vm.GpuDevices); i++ {
+	for i := range len(vm.GpuDevices) {
 		gpuDevices[i] = infrav1.GPUStatus{GPUID: *vm.GpuDevices[i].ID, Name: *vm.GpuDevices[i].Name}
 	}
 	machineCtx.ElfMachine.Status.GPUDevices = gpuDevices
@@ -272,7 +272,7 @@ func (r *ElfMachineReconciler) reconcileGPUDevices(ctx goctx.Context, machineCtx
 	}
 
 	gpuIDs := make([]string, len(vm.GpuDevices))
-	for i := 0; i < len(vm.GpuDevices); i++ {
+	for i := range len(vm.GpuDevices) {
 		gpuIDs[i] = *vm.GpuDevices[i].ID
 	}
 
@@ -338,14 +338,14 @@ func (r *ElfMachineReconciler) removeVMGPUDevices(ctx goctx.Context, machineCtx
 			return err
 		}
 
-		for i := 0; i < len(vmGPUInfo.GpuDevices); i++ {
+		for i := range len(vmGPUInfo.GpuDevices) {
 			staleGPUs = append(staleGPUs, &models.VMGpuOperationParams{
 				GpuID:  vmGPUInfo.GpuDevices[i].ID,
 				Amount: vmGPUInfo.GpuDevices[i].VgpuInstanceOnVMNum,
 			})
 		}
 	} else {
-		for i := 0; i < len(vm.GpuDevices); i++ {
+		for i := range len(vm.GpuDevices) {
 			staleGPUs = append(staleGPUs, &models.VMGpuOperationParams{
 				GpuID:  vm.GpuDevices[i].ID,
 				Amount: service.TowerInt32(1),