v1alpha3 API を使用すると、スーパーバイザー の TKG を使用して TanzuKubernetesCluster をプロビジョニングできます。v1alpha3 API のドキュメントについては、このトピックを参照してください。
TanzuKubernetesCluster v1alpha3 API
この仕様には、v1alpha3 API を使用して
TanzuKubernetesCluster をプロビジョニングする際に使用できるすべてのパラメータがリストされています。
重要： 有効なキー名は、英数字、ダッシュ（
key-name など）、アンダースコア（
KEY_NAME など）、ドット（
key.name など）のみで構成する必要があります。キー名にスペース文字を使用することはできません。
apiVersion: run.tanzu.vmware.com/v1alpha3
kind: TanzuKubernetesCluster
metadata:
name: string
namespace: string
annotations:
run.tanzu.vmware.com/resolve-os-image: os-name=string
spec:
topology:
controlPlane:
replicas: int32
vmClass: string
storageClass: string
volumes:
- name: string
mountPath: string
capacity:
storage: size in GiB
tkr:
reference:
name: string
nodeDrainTimeout: string
nodePools:
- name: string
failureDomain: string
labels: map[string]string
taints:
- key: string
value: string
effect: string
timeAdded: time
replicas: int32
vmClass: string
storageClass: string
volumes:
- name: string
mountPath: string
capacity:
storage: size in GiB
tkr:
reference:
name: string
nodeDrainTimeout: string
settings:
storage:
classes: [string]
defaultClass: string
network:
cni:
name: string
pods:
cidrBlocks: [string]
services:
cidrBlocks: [string]
serviceDomain: string
proxy:
httpProxy: string
httpsProxy: string
noProxy: [string]
trust:
additionalTrustedCAs:
- name: string
data: string
TanzuKubernetesCluster v1alpha3 API – 注釈付き
この注釈付きの仕様には、v1alpha3 API を使用して TanzuKubernetesCluster をプロビジョニングする際に使用できるすべてのパラメータと、各フィールドのドキュメントがリストされています。
apiVersion: run.tanzu.vmware.com/v1alpha3
kind: TanzuKubernetesCluster
#valid config key must consist of alphanumeric characters, '-', '_' or '.'
#metadata defines cluster information
metadata:
#name for this Tanzu Kubernetes cluster
name: string
#namespace vSphere Namespace where to provision this cluster
namespace: string
#Use annotation to provision non-default OS for the VM nodes
#PhotonOS is the default OS; use "ubuntu" to specify Ubuntu TKR
annotations:
run.tanzu.vmware.com/resolve-os-image: os-name=string
#spec defines cluster configuration
spec:
#topology describes the number, purpose, organization of nodes
#and the resources allocated for each
#nodes are grouped into pools based on their purpose
#controlPlane is special kind of a node pool
#nodePools is for groups of worker nodes
#each node pool is homogeneous: its nodes have the same
#resource allocation and use the same storage
topology:
#controlPlane defines the topology of the cluster
#controller, including the number of nodes and
#the resources allocated for each
#control plane must have an odd number of nodes
controlPlane:
#replicas is the number of nodes in the pool
#the control plane can have 1 or 3 nodes
#NOTE: production deployments require 3 control plane nodes
#defaults to 1 if nil (empty)
replicas: int32
#vmClass is the name of the VirtualMachineClass
#which describes the virtual hardware settings
#to be used for each node in the node pool
#vmClass controls the CPU and memory available
#to the node and the requests and limits on
#those resources; to list available vm classes run
#kubectl get virtualmachineclass
vmClass: string
#storageClass to be used for storage of the disks
#which store the root filesystems of the nodes
#to list available storage classes run
#kubectl describe storageclasses
storageClass: string
#volumes is the optional set of PVCs
#to create and attach to each control plane node
volumes:
#name of the PVC to be used as the suffix (node.name)
- name: string
#mountPath is the directory where the volume
#device is mounted; takes the form /dir/path
mountPath: string
#capacity is the PVC capacity
capacity:
#storage to be used for the disk
#volume; if not specified defaults to
#spec.controlPlane.storageClass
storage: size in GiB
#tkr.reference.name is the TKR NAME
#to be used by control plane nodes
#format is v1.27.11---vmware.1-fips.1-tkg.2
#currently all tkr.reference.name fields must match
tkr:
reference:
name: string
#nodeDrainTimeout is the total amount of time
#the controller will spend draining a node
#the default value is 0 which means the node is
#drained without any time limit
nodeDrainTimeout: string
#nodePools is an array that describes a group of
#worker nodes in the cluster with the same configuration
nodePools:
#name of the worker node pool
#must be unique in the cluster
- name: string
#failureDomain is the name of a vSphere Zone
#failureDomain is required for multi-zoned Supervisor
#in a multi-zoned Supervisor, you will have 3 node pools
#each refernecing a different failureDomain zone name
#refer to the examples
failureDomain: string
#labels are an optional map of string keys and values
#to organize and categorize objects
#propagated to the created nodes
labels: map[string]string
#taints specifies optional taints to register the
#Node API object with; user-defined taints are
#propagated to the created nodes
taints:
#key is the taint key to be applied to a node
- key: string
#value is the taint value corresponding to the key
value: string
#effect is the effect of the taint on pods
#that do not tolerate the taint; valid effects are
#NoSchedule, PreferNoSchedule, NoExecute
effect: string
#timeAdded is the time when the taint was added
#only written by the system for NoExecute taints
timeAdded: time
#replicas is the number of nodes in the pool
#worker nodePool can have from 0 to 150 nodes
#value of nil means the field is not reconciled,
#allowing external services like autoscalers
#to choose the number of nodes for the nodePool
#by default CAPI's MachineDeployment will pick 1
#NOTE: a cluster provisioned with 0 worker nodes/nodepools
#is not assigned any load balancer services
replicas: int32
#vmClass is the name of the VirtualMachineClass
#which describes the virtual hardware settings
#to be used for each node in the pool
#vmClass controls the CPU and memory available
#to the node and the requests and limits on
#those resources; to list available vm classes run
#kubectl get virtualmachineclass
vmClass: string
#storageClass to be used for storage of the disks
#which store the root filesystems of the nodes
#to list available storage classes run
#kubectl describe ns
storageClass: string
#volumes is the optional set of PVCs to create
#and attach to each node for high-churn worker node
#components such as the container runtime
volumes:
#name of this PVC to be used as the suffix (node.name)
- name: string
#mountPath is the directory where the volume
#device is mounted; takes the form /dir/path
mountPath: string
#capacity is the PVC capacity
capacity:
#storage to be used for the disk
#volume; if not specified defaults to
#topology.nodePools[*].storageClass
storage: size in GiB
#tkr.reference.name points to the TKR NAME
#to be used by spec.topology.nodePools[*] nodes
#format is v1.27.11---vmware.1-fips.1-tkg.2
#currently all tkr.reference.name fields must match
tkr:
reference:
name: string
#nodeDrainTimeout is the total amount of time
#the controller will spend draining a node
#the default value is 0 which means the node is
#drained without any time limit
nodeDrainTimeout: string
#settings are optional runtime configurations
#for the cluster, including persistent storage
#for pods and node network customizations
settings:
#storage defines persistent volume (PV) storage entries
#for container workloads; note that the storage used for
#node disks is defined by topology.controlPlane.storageClass
#and by spec.topology.nodePools[*].storageClass
storage:
#classes is a list of persistent volume (PV) storage
#classes to expose for container workloads on the cluster
#any class specified must be associated with the
#vSphere Namespace where the cluster is provisioned
#if omitted, all storage classes associated with the
#namespace will be exposed in the cluster
classes: [string]
#defaultClass treats the named storage class as the default
#for the cluster; because all namespaced storage classes
#are exposed if specific classes are not named,
#classes is not required to specify a defaultClass
#many workloads, including TKG Extensions and Helm,
#require a default storage class
#if omitted, no default storage class is set
defaultClass: string
#netowrk defines custom networking for cluster workloads
network:
#cni identifies the CNI plugin for the cluster
#use to override the default CNI set in the
#tkgservicesonfiguration spec, or when customizing
#network settings for the default CNI
cni:
#name is the name of the CNI plugin to use
#supported values are antrea, calico, antrea-nsx-routed
name: string
#pods configures custom networks for pods
#defaults to 192.168.0.0/16 if CNI is antrea or calico
#defaults to empty if CNI is antrea-nsx-routed
#custom subnet size must equal or exceed /24
#use caution before seeting CIDR range other than /16
#cannot overlap with Supervisor workload network
pods:
#cidrBlocks is an array of network ranges
#multiple ranges may not be supported by all CNI plugins
cidrBlocks: [string]
#services configures custom network for services
#defaults to 10.96.0.0/12
#cannot overlap with Supervisor workload network
services:
#cidrBlocks is an array of network ranges
#multiple ranges many not be supported by all CNI plugins
cidrBlocks: [string]
#serviceDomain specifies the service domain for the cluster
#defaults to cluster.local
serviceDomain: string
#proxy configures proxy server to be used inside the cluster
#if omitted no proxy is configured
proxy:
#httpProxy is the proxy URI for HTTP connections
#to endpoints outside the cluster
#takes form http://<user>:<pwd>@<ip>:<port>
httpProxy: string
#httpsProxy is the proxy URL for HTTPS connections
#to endpoints outside the cluster
#takes the frorm http://<user>:<pwd>@<ip>:<port>
httpsProxy: string
#noProxy is the list of destination domain names, domains,
#IP addresses, and other network CIDRs to exclude from proxying
#must include Supervisor Cluster Pod, Egress, Ingress CIDRs
noProxy: [string]
#trust configures additional certificates for the cluster
#if omitted no additional certificate is configured
trust:
#additionalTrustedCAs are additional trusted certificates
#can be additional CAs or end certificates
additionalTrustedCAs:
#name is the name of the additional trusted certificate
#must match the name used in the filename
- name: string
#data holds the contents of the additional trusted cert
#PEM Public Certificate data as a base64-encoded string
#such as LS0tLS1C...LS0tCg== where "..." is the
#middle section of the long base64-encoded string
data: string
