✨ Update API with ClusterClass MachinePool support

api/v1alpha4/conversion.go

@@ -73,6 +73,8 @@ func (src *Cluster) ConvertTo(dstRaw conversion.Hub) error {
 				dst.Spec.Topology.Workers.MachineDeployments[i].Strategy = restored.Spec.Topology.Workers.MachineDeployments[i].Strategy
 				dst.Spec.Topology.Workers.MachineDeployments[i].MachineHealthCheck = restored.Spec.Topology.Workers.MachineDeployments[i].MachineHealthCheck
 			}
+
+			dst.Spec.Topology.Workers.MachinePools = restored.Spec.Topology.Workers.MachinePools
 		}
 	}
 
@@ -125,6 +127,7 @@ func (src *ClusterClass) ConvertTo(dstRaw conversion.Hub) error {
 	dst.Spec.ControlPlane.NodeDrainTimeout = restored.Spec.ControlPlane.NodeDrainTimeout
 	dst.Spec.ControlPlane.NodeVolumeDetachTimeout = restored.Spec.ControlPlane.NodeVolumeDetachTimeout
 	dst.Spec.ControlPlane.NodeDeletionTimeout = restored.Spec.ControlPlane.NodeDeletionTimeout
+	dst.Spec.Workers.MachinePools = restored.Spec.Workers.MachinePools
 
 	for i := range restored.Spec.Workers.MachineDeployments {
 		dst.Spec.Workers.MachineDeployments[i].MachineHealthCheck = restored.Spec.Workers.MachineDeployments[i].MachineHealthCheck
@@ -375,3 +378,13 @@ func Convert_v1beta1_ClusterClass_To_v1alpha4_ClusterClass(in *clusterv1.Cluster
 	// ClusterClass.Status has been added in v1beta1.
 	return autoConvert_v1beta1_ClusterClass_To_v1alpha4_ClusterClass(in, out, s)
 }
+
+func Convert_v1beta1_WorkersClass_To_v1alpha4_WorkersClass(in *clusterv1.WorkersClass, out *WorkersClass, s apiconversion.Scope) error {
+	// WorkersClass.MachinePools has been added in v1beta1.
+	return autoConvert_v1beta1_WorkersClass_To_v1alpha4_WorkersClass(in, out, s)
+}
+
+func Convert_v1beta1_WorkersTopology_To_v1alpha4_WorkersTopology(in *clusterv1.WorkersTopology, out *WorkersTopology, s apiconversion.Scope) error {
+	// WorkersTopology.MachinePools has been added in v1beta1.
+	return autoConvert_v1beta1_WorkersTopology_To_v1alpha4_WorkersTopology(in, out, s)
+}

api/v1beta1/cluster_types.go

@@ -147,6 +147,10 @@ type WorkersTopology struct {
 	// MachineDeployments is a list of machine deployments in the cluster.
 	// +optional
 	MachineDeployments []MachineDeploymentTopology `json:"machineDeployments,omitempty"`
+
+	// MachinePools is a list of machine pools in the cluster.
+	// +optional
+	MachinePools []MachinePoolTopology `json:"machinePools,omitempty"`
 }
 
 // MachineDeploymentTopology specifies the different parameters for a set of worker nodes in the topology.
@@ -238,6 +242,66 @@ type MachineHealthCheckTopology struct {
 	MachineHealthCheckClass `json:",inline"`
 }
 
+// MachinePoolTopology specifies the different parameters for a pool of worker nodes in the topology.
+// This pool of nodes is managed by a MachinePool object whose lifecycle is managed by the Cluster controller.
+type MachinePoolTopology struct {
+	// Metadata is the metadata applied to the MachinePool.
+	// At runtime this metadata is merged with the corresponding metadata from the ClusterClass.
+	// +optional
+	Metadata ObjectMeta `json:"metadata,omitempty"`
+
+	// Class is the name of the MachinePoolClass used to create the pool of worker nodes.
+	// This should match one of the deployment classes defined in the ClusterClass object
+	// mentioned in the `Cluster.Spec.Class` field.
+	Class string `json:"class"`
+
+	// Name is the unique identifier for this MachinePoolTopology.
+	// The value is used with other unique identifiers to create a MachinePool's Name
+	// (e.g. cluster's name, etc). In case the name is greater than the allowed maximum length,
+	// the values are hashed together.
+	Name string `json:"name"`
+
+	// FailureDomains is the list of failure domains the machine pool will be created in.
+	// Must match a key in the FailureDomains map stored on the cluster object.
+	// +optional
+	FailureDomains []string `json:"failureDomains,omitempty"`
+
+	// NodeDrainTimeout is the total amount of time that the controller will spend on draining a node.
+	// The default value is 0, meaning that the node can be drained without any time limitations.
+	// NOTE: NodeDrainTimeout is different from `kubectl drain --timeout`
+	// +optional
+	NodeDrainTimeout *metav1.Duration `json:"nodeDrainTimeout,omitempty"`
+
+	// NodeVolumeDetachTimeout is the total amount of time that the controller will spend on waiting for all volumes
+	// to be detached. The default value is 0, meaning that the volumes can be detached without any time limitations.
+	// +optional
+	NodeVolumeDetachTimeout *metav1.Duration `json:"nodeVolumeDetachTimeout,omitempty"`
+
+	// NodeDeletionTimeout defines how long the controller will attempt to delete the Node that the MachinePool
+	// hosts after the MachinePool is marked for deletion. A duration of 0 will retry deletion indefinitely.
+	// Defaults to 10 seconds.
+	// +optional
+	NodeDeletionTimeout *metav1.Duration `json:"nodeDeletionTimeout,omitempty"`
+
+	// Minimum number of seconds for which a newly created machine pool should
+	// be ready.
+	// Defaults to 0 (machine will be considered available as soon as it
+	// is ready)
+	// +optional
+	MinReadySeconds *int32 `json:"minReadySeconds,omitempty"`
+
+	// Replicas is the number of nodes belonging to this pool.
+	// If the value is nil, the MachinePool is created without the number of Replicas (defaulting to 1)
+	// and it's assumed that an external entity (like cluster autoscaler) is responsible for the management
+	// of this value.
+	// +optional
+	Replicas *int32 `json:"replicas,omitempty"`
+
+	// Variables can be used to customize the MachinePool through patches.
+	// +optional
+	Variables *MachinePoolVariables `json:"variables,omitempty"`
+}
+
 // ClusterVariable can be used to customize the Cluster through patches. Each ClusterVariable is associated with a
 // Variable definition in the ClusterClass `status` variables.
 type ClusterVariable struct {
@@ -268,6 +332,13 @@ type MachineDeploymentVariables struct {
 	Overrides []ClusterVariable `json:"overrides,omitempty"`
 }
 
+// MachinePoolVariables can be used to provide variables for a specific MachinePool.
+type MachinePoolVariables struct {
+	// Overrides can be used to override Cluster level variables.
+	// +optional
+	Overrides []ClusterVariable `json:"overrides,omitempty"`
+}
+
 // ANCHOR_END: ClusterSpec
 
 // ANCHOR: ClusterNetwork

api/v1beta1/clusterclass_types.go

@@ -133,6 +133,11 @@ type WorkersClass struct {
 	// a set of worker nodes.
 	// +optional
 	MachineDeployments []MachineDeploymentClass `json:"machineDeployments,omitempty"`
+
+	// MachinePools is a list of machine pool classes that can be used to create
+	// a set of worker nodes.
+	// +optional
+	MachinePools []MachinePoolClass `json:"machinePools,omitempty"`
 }
 
 // MachineDeploymentClass serves as a template to define a set of worker nodes of the cluster
@@ -244,6 +249,69 @@ type MachineHealthCheckClass struct {
 	RemediationTemplate *corev1.ObjectReference `json:"remediationTemplate,omitempty"`
 }
 
+// MachinePoolClass serves as a template to define a pool of worker nodes of the cluster
+// provisioned using `ClusterClass`.
+type MachinePoolClass struct {
+	// Class denotes a type of machine pool present in the cluster,
+	// this name MUST be unique within a ClusterClass and can be referenced
+	// in the Cluster to create a managed MachinePool.
+	Class string `json:"class"`
+
+	// Template is a local struct containing a collection of templates for creation of
+	// MachinePools objects representing a pool of worker nodes.
+	Template MachinePoolClassTemplate `json:"template"`
+
+	// FailureDomains is the list of failure domains the MachinePool should be attached to.
+	// Must match a key in the FailureDomains map stored on the cluster object.
+	// NOTE: This value can be overridden while defining a Cluster.Topology using this MachinePoolClass.
+	// +optional
+	FailureDomains []string `json:"failureDomains,omitempty"`
+
+	// NodeDrainTimeout is the total amount of time that the controller will spend on draining a node.
+	// The default value is 0, meaning that the node can be drained without any time limitations.
+	// NOTE: NodeDrainTimeout is different from `kubectl drain --timeout`
+	// NOTE: This value can be overridden while defining a Cluster.Topology using this MachinePoolClass.
+	// +optional
+	NodeDrainTimeout *metav1.Duration `json:"nodeDrainTimeout,omitempty"`
+
+	// NodeVolumeDetachTimeout is the total amount of time that the controller will spend on waiting for all volumes
+	// to be detached. The default value is 0, meaning that the volumes can be detached without any time limitations.
+	// NOTE: This value can be overridden while defining a Cluster.Topology using this MachinePoolClass.
+	// +optional
+	NodeVolumeDetachTimeout *metav1.Duration `json:"nodeVolumeDetachTimeout,omitempty"`
+
+	// NodeDeletionTimeout defines how long the controller will attempt to delete the Node that the Machine
+	// hosts after the Machine Pool is marked for deletion. A duration of 0 will retry deletion indefinitely.
+	// Defaults to 10 seconds.
+	// NOTE: This value can be overridden while defining a Cluster.Topology using this MachinePoolClass.
+	// +optional
+	NodeDeletionTimeout *metav1.Duration `json:"nodeDeletionTimeout,omitempty"`
+
+	// Minimum number of seconds for which a newly created machine pool should
+	// be ready.
+	// Defaults to 0 (machine will be considered available as soon as it
+	// is ready)
+	// NOTE: This value can be overridden while defining a Cluster.Topology using this MachinePoolClass.
+	MinReadySeconds *int32 `json:"minReadySeconds,omitempty"`
+}
+
+// MachinePoolClassTemplate defines how a MachinePool generated from a MachinePoolClass
+// should look like.
+type MachinePoolClassTemplate struct {
+	// Metadata is the metadata applied to the MachinePool.
+	// At runtime this metadata is merged with the corresponding metadata from the topology.
+	// +optional
+	Metadata ObjectMeta `json:"metadata,omitempty"`
+
+	// Bootstrap contains the bootstrap template reference to be used
+	// for the creation of the Machines in the MachinePool.
+	Bootstrap LocalObjectTemplate `json:"bootstrap"`
+
+	// Infrastructure contains the infrastructure template reference to be used
+	// for the creation of the MachinePool.
+	Infrastructure LocalObjectTemplate `json:"infrastructure"`
+}
+
 // IsZero returns true if none of the values of MachineHealthCheckClass are defined.
 func (m MachineHealthCheckClass) IsZero() bool {
 	return reflect.ValueOf(m).IsZero()
@@ -472,6 +540,11 @@ type PatchSelectorMatch struct {
 	// .spec.workers.machineDeployments.
 	// +optional
 	MachineDeploymentClass *PatchSelectorMatchMachineDeploymentClass `json:"machineDeploymentClass,omitempty"`
+
+	// MachinePoolClass selects templates referenced in specific MachinePoolClasses in
+	// .spec.workers.machinePools.
+	// +optional
+	MachinePoolClass *PatchSelectorMatchMachinePoolClass `json:"machinePoolClass,omitempty"`
 }
 
 // PatchSelectorMatchMachineDeploymentClass selects templates referenced
@@ -482,6 +555,14 @@ type PatchSelectorMatchMachineDeploymentClass struct {
 	Names []string `json:"names,omitempty"`
 }
 
+// PatchSelectorMatchMachinePoolClass selects templates referenced
+// in specific MachinePoolClasses in .spec.workers.machinePools.
+type PatchSelectorMatchMachinePoolClass struct {
+	// Names selects templates by class names.
+	// +optional
+	Names []string `json:"names,omitempty"`
+}
+
 // JSONPatch defines a JSON patch.
 type JSONPatch struct {
 	// Op defines the operation of the patch.

api/v1beta1/common_types.go

@@ -57,6 +57,10 @@ const (
 	// a classy Cluster to define the maximum concurrency while upgrading MachineDeployments.
 	ClusterTopologyUpgradeConcurrencyAnnotation = "topology.cluster.x-k8s.io/upgrade-concurrency"
 
+	// ClusterTopologyMachinePoolNameLabel is the label set on the generated  MachinePool objects
+	// to track the name of the MachinePool topology it represents.
+	ClusterTopologyMachinePoolNameLabel = "topology.cluster.x-k8s.io/pool-name"
+
 	// ClusterTopologyUnsafeUpdateClassNameAnnotation can be used to disable the webhook check on
 	// update that disallows a pre-existing Cluster to be populated with Topology information and Class.
 	ClusterTopologyUnsafeUpdateClassNameAnnotation = "unsafe.topology.cluster.x-k8s.io/disable-update-class-name-check"