Migration controller pod launch

Starts the migration process, through the launching of the target pod.
Increased unit test coverage for both migration and VMs

cmd/virt-controller/virt-controller.go

@@ -9,6 +9,7 @@ import (
 
 	"github.com/emicklei/go-restful"
 	"github.com/facebookgo/inject"
+	clientrest "k8s.io/client-go/rest"
 	"kubevirt.io/kubevirt/pkg/kubecli"
 	"kubevirt.io/kubevirt/pkg/logging"
 	"kubevirt.io/kubevirt/pkg/virt-controller/rest"
@@ -40,7 +41,14 @@ func main() {
 		golog.Fatal(err)
 	}
 
+	var restClient *clientrest.RESTClient
+	restClient, err = kubecli.GetRESTClient()
+	if err != nil {
+		golog.Fatal(err)
+	}
+
 	g.Provide(
+		&inject.Object{Value: restClient},
 		&inject.Object{Value: clientSet},
 		&inject.Object{Value: templateService},
 		&inject.Object{Value: vmService},
@@ -57,15 +65,14 @@ func main() {
 	defer close(stop)
 
 	// Start wachting vms
-	restClient, err := kubecli.GetRESTClient()
+	restClient, err = kubecli.GetRESTClient()
 	if err != nil {
 		golog.Fatal(err)
 	}
 	vmCache, vmController := watch.NewVMController(vmService, nil, restClient)
 
 	vmController.StartInformer(stop)
 	go vmController.Run(1, stop)
-
 	// Wait until VM cache has warmed up before we start watching pods
 	vmController.WaitForSync(stop)
 
@@ -74,6 +81,11 @@ func main() {
 	podController.StartInformer(stop)
 	go podController.Run(1, stop)
 
+	_, migrationController := watch.NewMigrationController(vmService, nil, restClient)
+	migrationController.StartInformer(stop)
+	go migrationController.Run(1, stop)
+	migrationController.WaitForSync(stop)
+
 	httpLogger := logger.With("service", "http")
 
 	httpLogger.Info().Log("action", "listening", "interface", *host, "port", *port)

pkg/api/v1/types.go

@@ -240,17 +240,26 @@ func (s SyncEvent) String() string {
 
 func NewMinimalVM(vmName string) *VM {
 	precond.CheckNotEmpty(vmName)
-	return &VM{
-		Spec: VMSpec{Domain: NewMinimalDomainSpec(vmName)},
+	vm := NewVMReferenceFromName(vmName)
+	vm.Spec = VMSpec{Domain: NewMinimalDomainSpec(vmName)}
+	vm.TypeMeta = metav1.TypeMeta{
+		APIVersion: GroupVersion.String(),
+		Kind:       "VM",
+	}
+	return vm
+}
+
+// TODO Namespace could be different, also store it somewhere in the domain, so that we can report deletes on handler startup properly
+func NewVMReferenceFromName(name string) *VM {
+	vm := &VM{
 		ObjectMeta: v1.ObjectMeta{
-			Name:      vmName,
+			Name:      name,
 			Namespace: kubeapi.NamespaceDefault,
-		},
-		TypeMeta: metav1.TypeMeta{
-			APIVersion: GroupVersion.String(),
-			Kind:       "VM",
+			SelfLink:  fmt.Sprintf("/apis/%s/namespaces/%s/vms/%s", GroupVersion.String(), kubeapi.NamespaceDefault, name),
 		},
 	}
+	vm.SetGroupVersionKind(schema.GroupVersionKind{Group: GroupVersion.Group, Kind: "VM", Version: GroupVersion.Version})
+	return vm
 }
 
 type Spice struct {
@@ -280,33 +289,28 @@ func NewSpice(vmName string) *Spice {
 	}
 }
 
-// TODO Namespace could be different, also store it somewhere in the domain, so that we can report deletes on handler startup properly
-func NewVMReferenceFromName(name string) *VM {
-	vm := &VM{
-		ObjectMeta: v1.ObjectMeta{
-			Name:      name,
-			Namespace: kubeapi.NamespaceDefault,
-			SelfLink:  fmt.Sprintf("/apis/%s/namespaces/%s/vms/%s", GroupVersion.String(), kubeapi.NamespaceDefault, name),
-		},
+//TODO validate that this is correct
+func NewMinimalMigration(name string, vmName string) *Migration {
+	migration := NewMigrationReferenceFromName(name)
+	migration.Spec = MigrationSpec{
+		MigratingVMName: vmName,
 	}
-	vm.SetGroupVersionKind(schema.GroupVersionKind{Group: GroupVersion.Group, Kind: "VM", Version: GroupVersion.Version})
-	return vm
+	return migration
 }
 
-//TODO validate that this is correct
-func NewMinimalMigration(name string, vmName string) *Migration {
+func NewMigrationReferenceFromName(name string) *Migration {
 	migration := &Migration{
 		ObjectMeta: v1.ObjectMeta{
 			Name:      name,
 			Namespace: kubeapi.NamespaceDefault,
-			SelfLink:  fmt.Sprintf("/apis/%s/namespaces/%s/migrations/%s", GroupVersion.String(), kubeapi.NamespaceDefault, name),
+			SelfLink:  fmt.Sprintf("/apis/%s/namespaces/%s/%s", GroupVersion.String(), kubeapi.NamespaceDefault, name),
 		},
 		TypeMeta: metav1.TypeMeta{
 			APIVersion: GroupVersion.String(),
 			Kind:       "Migration",
 		},
-		Spec: MigrationSpec{
-			MigratingVMName: vmName,
+		Status: MigrationStatus{
+			Phase: MigrationUnknown,
 		},
 	}
 	return migration
@@ -341,6 +345,9 @@ type MigrationSpec struct {
 type MigrationPhase string
 
 const (
+	// Create Migration has been called but nothing has been done with it
+	MigrationUnknown MigrationPhase = ""
+
 	// Migration has been scheduled but no update on the status has been recorded
 	MigrationPending MigrationPhase = "Pending"
 
@@ -358,12 +365,24 @@ const (
 // MigrationStatus is the last reported status of a VM Migratrion. Status may trail the actual
 // state of a migration.
 type MigrationStatus struct {
-	Phase MigrationPhase `json:"message,omitempty"`
+	Phase MigrationPhase `json:"phase,omitempty"`
 }
 
 // Required to satisfy ObjectMetaAccessor interface
-func (v *Migration) GetObjectMeta() meta.Object {
-	return &v.ObjectMeta
+func (m *Migration) GetObjectMeta() meta.Object {
+	return &m.ObjectMeta
+}
+
+func (m *Migration) UnmarshalJSON(data []byte) error {
+	type MigrationCopy Migration
+	tmp := MigrationCopy{}
+	err := json.Unmarshal(data, &tmp)
+	if err != nil {
+		return err
+	}
+	tmp2 := Migration(tmp)
+	*m = tmp2
+	return nil
 }
 
 //A list of Migrations

pkg/virt-api/rest/kubeproxy_test.go

@@ -26,7 +26,6 @@ import (
 	rest2 "kubevirt.io/kubevirt/pkg/rest"
 	. "kubevirt.io/kubevirt/pkg/virt-api/rest"
 	. "kubevirt.io/kubevirt/test"
-
 )
 
 const vmResource = "vms"

pkg/virt-controller/services/vm.go

@@ -6,29 +6,36 @@ import (
 	"k8s.io/client-go/pkg/api/v1"
 	"k8s.io/client-go/pkg/fields"
 	"k8s.io/client-go/pkg/labels"
+	"k8s.io/client-go/rest"
 	corev1 "kubevirt.io/kubevirt/pkg/api/v1"
 	"kubevirt.io/kubevirt/pkg/middleware"
 	"kubevirt.io/kubevirt/pkg/precond"
 )
 
 type VMService interface {
-	StartVM(*corev1.VM) error
-	DeleteVM(*corev1.VM) error
-	PrepareMigration(*corev1.VM) error
-	GetRunningPods(*corev1.VM) (*v1.PodList, error)
+	StartVMPod(*corev1.VM) error
+	DeleteVMPod(*corev1.VM) error
+	GetRunningVMPods(*corev1.VM) (*v1.PodList, error)
+	DeleteMigration(*corev1.Migration) error
+	GetRunningMigrationPods(*corev1.Migration) (*v1.PodList, error)
+	SetupMigration(migration *corev1.Migration, vm *corev1.VM) error
+	UpdateMigration(migration *corev1.Migration) error
+	FetchVM(vmName string) (*corev1.VM, error)
 }
 
 type vmService struct {
 	KubeCli         *kubernetes.Clientset `inject:""`
+	RestClient      *rest.RESTClient      `inject:""`
 	TemplateService TemplateService       `inject:""`
 }
 
-func (v *vmService) StartVM(vm *corev1.VM) error {
+func (v *vmService) StartVMPod(vm *corev1.VM) error {
+
 	precond.MustNotBeNil(vm)
 	precond.MustNotBeEmpty(vm.GetObjectMeta().GetName())
 	precond.MustNotBeEmpty(string(vm.GetObjectMeta().GetUID()))
 
-	podList, err := v.GetRunningPods(vm)
+	podList, err := v.GetRunningVMPods(vm)
 	if err != nil {
 		return err
 	}
@@ -49,59 +56,97 @@ func (v *vmService) StartVM(vm *corev1.VM) error {
 	return nil
 }
 
-func (v *vmService) DeleteVM(vm *corev1.VM) error {
+func (v *vmService) DeleteVMPod(vm *corev1.VM) error {
 	precond.MustNotBeNil(vm)
 	precond.MustNotBeEmpty(vm.GetObjectMeta().GetName())
 
-	if err := v.KubeCli.Core().Pods(v1.NamespaceDefault).DeleteCollection(nil, UnfinishedVMPodSelector(vm)); err != nil {
+	if err := v.KubeCli.CoreV1().Pods(v1.NamespaceDefault).DeleteCollection(nil, UnfinishedVMPodSelector(vm)); err != nil {
 		return err
 	}
 	return nil
 }
 
-func (v *vmService) GetRunningPods(vm *corev1.VM) (*v1.PodList, error) {
-	podList, err := v.KubeCli.Core().Pods(v1.NamespaceDefault).List(UnfinishedVMPodSelector(vm))
+func (v *vmService) GetRunningVMPods(vm *corev1.VM) (*v1.PodList, error) {
+	podList, err := v.KubeCli.CoreV1().Pods(v1.NamespaceDefault).List(UnfinishedVMPodSelector(vm))
+
 	if err != nil {
 		return nil, err
 	}
 	return podList, nil
 }
 
+func (v *vmService) UpdateMigration(migration *corev1.Migration) error {
+	migrationName := migration.ObjectMeta.Name
+	_, err := v.RestClient.Put().Namespace(v1.NamespaceDefault).Resource("migrations").Body(migration).Name(migrationName).Do().Get()
+	return err
+}
+
+func (v *vmService) FetchVM(vmName string) (*corev1.VM, error) {
+	resp, err := v.RestClient.Get().Namespace(v1.NamespaceDefault).Resource("vms").Name(vmName).Do().Get()
+	if err != nil {
+		return nil, err
+	}
+	vm := resp.(*corev1.VM)
+	return vm, nil
+}
+
 func NewVMService() VMService {
 	svc := vmService{}
 	return &svc
 }
 
-func (v *vmService) PrepareMigration(vm *corev1.VM) error {
-	precond.MustNotBeNil(vm)
-	precond.MustNotBeEmpty(vm.GetObjectMeta().GetName())
-	precond.MustBeTrue(len(vm.Spec.NodeSelector) > 0)
-	podList, err := v.GetRunningPods(vm)
+func UnfinishedVMPodSelector(vm *corev1.VM) v1.ListOptions {
+	fieldSelector := fields.ParseSelectorOrDie(
+		"status.phase!=" + string(v1.PodFailed) +
+			",status.phase!=" + string(v1.PodSucceeded))
+	labelSelector, err := labels.Parse(fmt.Sprintf(corev1.DomainLabel+" in (%s)", vm.GetObjectMeta().GetName()))
 	if err != nil {
-		return err
+		panic(err)
 	}
+	return v1.ListOptions{FieldSelector: fieldSelector.String(), LabelSelector: labelSelector.String()}
+}
 
-	// If there are more than one pod in other states than Succeeded or Failed we can't go on
-	if len(podList.Items) > 1 {
-		return middleware.NewResourceConflictError(fmt.Sprintf("VM %s is already migrating", vm.GetObjectMeta().GetName()))
+func (v *vmService) SetupMigration(migration *corev1.Migration, vm *corev1.VM) error {
+	pod, err := v.TemplateService.RenderLaunchManifest(vm)
+	if err == nil {
+		_, err = v.KubeCli.CoreV1().Pods(v1.NamespaceDefault).Create(pod)
+	}
+	if err == nil {
+		migration.Status.Phase = corev1.MigrationInProgress
+	} else {
+		migration.Status.Phase = corev1.MigrationFailed
 	}
 
-	pod, err := v.TemplateService.RenderLaunchManifest(vm)
-	if err != nil {
-		return err
+	err2 := v.UpdateMigration(migration)
+	if err2 != nil {
+		err = err2
 	}
-	if _, err := v.KubeCli.Core().Pods(v1.NamespaceDefault).Create(pod); err != nil {
+	return err
+}
+
+func (v *vmService) DeleteMigration(migration *corev1.Migration) error {
+	precond.MustNotBeNil(migration)
+	precond.MustNotBeEmpty(migration.GetObjectMeta().GetName())
+
+	if err := v.KubeCli.CoreV1().Pods(v1.NamespaceDefault).DeleteCollection(nil, unfinishedMigrationPodSelector(migration)); err != nil {
 		return err
 	}
 	return nil
 }
 
-// Visible for tests
-func UnfinishedVMPodSelector(vm *corev1.VM) v1.ListOptions {
+func (v *vmService) GetRunningMigrationPods(migration *corev1.Migration) (*v1.PodList, error) {
+	podList, err := v.KubeCli.CoreV1().Pods(v1.NamespaceDefault).List(unfinishedMigrationPodSelector(migration))
+	if err != nil {
+		return nil, err
+	}
+	return podList, nil
+}
+
+func unfinishedMigrationPodSelector(migration *corev1.Migration) v1.ListOptions {
 	fieldSelector := fields.ParseSelectorOrDie(
 		"status.phase!=" + string(v1.PodFailed) +
 			",status.phase!=" + string(v1.PodSucceeded))
-	labelSelector, err := labels.Parse(fmt.Sprintf(corev1.DomainLabel+" in (%s)", vm.GetObjectMeta().GetName()))
+	labelSelector, err := labels.Parse(fmt.Sprintf(corev1.DomainLabel+" in (%s)", migration.GetObjectMeta().GetName()))
 	if err != nil {
 		panic(err)
 	}