Container Service for Kubernetes:nodepool

Specifies whether to enable auto scaling.

• true: Enables the auto scaling feature for the node pool.

• false: Disables auto scaling. If this parameter is set to false, other configuration parameters in auto_scaling do not take effect.

Default value: false.

The maximum number of instances in the auto scaling group.

The minimum number of instances in the auto scaling group.

The auto scaling type, classified by instance type. Valid values:

• cpu: regular instance type.

• gpu: GPU instance type.

• gpushare: GPU sharing type.

• spot: spot instance type.

Default value: cpu.

[This field is deprecated].

Specifies whether to associate an EIP. Valid values:

• true: Associates an EIP.

• false: Does not associate an EIP.

Default value: false.

[This field is deprecated].

The billing method of the EIP. Valid values:

• PayByBandwidth: pay-by-bandwidth.

• PayByTraffic: pay-by-traffic.

Default value: PayByBandwidth.

[This field is deprecated]

The peak bandwidth of the EIP. Unit: Mbit/s.

Specifies whether to install the CloudMonitor agent on ECS nodes. After installation, you can view monitoring information of the created ECS instances in the CloudMonitor console. We recommend that you enable this feature. Valid values:

• true: Installs the CloudMonitor agent on ECS nodes.

• false: Does not install the CloudMonitor agent on ECS nodes.

Default value: false.

The CPU management policy for nodes. The following two policies are supported for clusters of version 1.12.6 or later:

• static: allows pods with certain resource characteristics on the node to be granted enhanced CPU affinity and exclusivity.

• none: uses the existing default CPU affinity scheme.

Default value: none.

The container runtime. Valid values:

• containerd: recommended. Supports all cluster versions.

• Sandboxed-Container.runv: sandboxed container that provides higher isolation. Supports clusters of version 1.24 or earlier.

• docker: supports clusters of version 1.22 or earlier.

Default value: containerd.

The container runtime version.

The custom data for nodes.

A node name consists of three parts: prefix + node IP + suffix:

• The prefix and suffix can each contain one or more parts separated by periods (.). Each part can contain lowercase letters, digits, and hyphens (-). The node name must start and end with a lowercase letter or digit.

• The node IP is the complete private IP address of the node.

The parameter contains four parts separated by commas. For example, if you pass the string "customized,aliyun,ip,com" (where "customized" and "ip" are fixed strings, "aliyun" is the prefix, and "com" is the suffix), the node name is aliyun.192.168.xxx.xxx.com.

The node pool name.

The ID of the resource group to which the node pool belongs.

The node pool type. Valid values:

• ess: regular node pool.

• edge: edge node pool.

Specifies whether to enable auto-renewal for the node pool. This parameter takes effect only when instance_charge_type is set to PrePaid. Valid values:

• true: Auto-renewal is enabled.

• false: Auto-renewal is disabled.

Default value: true.

The auto-renewal epoch of the node pool. This parameter takes effect and is required only when instance_charge_type is set to PrePaid.

When PeriodUnit=Month, valid values: {1, 2, 3, 6, 12}.

Default value: 1.

The custom image ID. By default, the system-provided image is used.

The billing method for nodes in the node pool. Valid values:

• PrePaid: subscription.

• PostPaid: pay-as-you-go.

Default value: PostPaid.

The key pair name. Specify either this parameter or login_password.

The SSH logon password. Specify either this parameter or key_pair. The password must be 8 to 30 characters in length and must contain at least three of the following character types: uppercase letters, lowercase letters, digits, and special characters.

Specifies whether to log on to the scaled-out ECS instance as a non-root user.

The subscription duration of the node pool. This parameter takes effect and is required only when instance_charge_type is set to PrePaid. When period_unit is set to Month, valid values of period: {1, 2, 3, 6, 12}.

Default value: 1.

The unit of the subscription period for nodes in the node pool. This parameter must be specified when instance_charge_type is set to PrePaid.

Month: Currently, only monthly billing is supported.

The operating system distribution. Valid values:

• CentOS

• AliyunLinux

• Windows

• WindowsCore

Default value: AliyunLinux.

The type of spot instance. Valid values:

• NoSpot: non-spot instance.

• SpotWithPriceLimit: spot instance with a price limit.

• SpotAsPriceGo: spot instance priced at the market price at the time of purchase.

For more information, see Spot instances.

The scaling group mode. Valid values:

• release: Standard mode. ECS instances are created and released based on resource usage.

• recycle: Swift mode. ECS instances are created, stopped, and restarted for scaling, which accelerates subsequent scaling operations. Stopped instances are not charged for compute resources. Only storage fees are charged, except for instances with local disks.

Default value: release.

The security group ID of the node pool. Specify either this parameter or security_group_ids. We recommend that you use security_group_ids.

The type of the node system cloud disk. Valid values:

• cloud_efficiency: ultra cloud disk.

• cloud_ssd: standard SSD.

• cloud_essd: ESSD cloud disk.

• cloud_auto: ESSD AutoPL cloud disk.

• cloud_essd_entry: ESSD Entry disk.

Default value: cloud_efficiency.

The size of the node system cloud disk. Unit: GiB.

Valid values: [40,500].

The performance level of the node system cloud disk. This parameter takes effect only for ESSD cloud disks.

• PL0: moderate maximum concurrent I/O performance with relatively stable read/write latency.

• PL1: moderate maximum concurrent I/O performance with relatively stable read/write latency.

• PL2: high maximum concurrent I/O performance with stable read/write latency.

• PL3: ultra-high maximum concurrent I/O performance with extremely stable read/write latency.

Specifies whether to encrypt the system cloud disk. Valid values: true: Encrypted. false: Not encrypted.

The KMS key ID used by the node system cloud disk.

The encryption algorithm used for the system cloud disk of a node. Valid values: aes-256.

The provisioned read/write IOPS of the system cloud disk of a node. Valid values: 0 to min{50,000, 1000*Capacity-Baseline performance}. Baseline performance = min{1,800+50*Capacity, 50000}.

This parameter is supported only when SystemDiskCategory is set to cloud_auto. For more information, see ESSD AutoPL cloud disks.

Specifies whether to enable burst (performance bursting) for the system cloud disk of a node. Valid values:

• true: Enable burst.

• false: Disable burst.

This parameter is supported only when SystemDiskCategory is set to cloud_auto. For more information, see ESSD AutoPL cloud disks.

The multi-zone scaling policy for ECS instances in the scaling group. Valid values:

• PRIORITY: Scales instances based on the vSwitches (VSwitchIds.N) that you define. When ECS instances cannot be created in the zone of the vSwitch with the highest priority, the system automatically uses the vSwitch with the next highest priority to create ECS instances.

• COST_OPTIMIZED: Attempts to create instances in ascending order of vCPU unit price. When the scaling configuration sets multiple instance types with the spot billable methods, spot instances are created first. You can use the CompensateWithOnDemand parameter to specify whether the system is allowed to automatically attempt to create pay-as-you-go instances when spot instances cannot be created due to insufficient inventory or other reasons.

  説明

  COST_OPTIMIZED takes effect only when the scaling configuration specifies multiple instance types or uses spot instances.

• BALANCE: Evenly distributes ECS instances across the multiple zones specified in the scaling group. If the zones become unbalanced due to insufficient inventory or other reasons, you can call the RebalanceInstances operation to rebalance resources. For more information, see RebalanceInstances.

Default value: PRIORITY.

The minimum number of pay-as-you-go instances required in the scaling group. Valid values: 0 to 1000. When the number of pay-as-you-go instances is less than this value, pay-as-you-go instances are created first.

The percentage of pay-as-you-go instances among the excess instances after the minimum number of pay-as-you-go instances (on_demand_base_capacity) is met. Valid values: 0 to 100.

The number of available instance types. The scaling group creates spot instances of multiple types at the lowest cost. Valid values: 1 to 10.

Specifies whether to enable supplementation of spot instances. If enabled, the scaling group attempts to create new instances to replace spot instances that are about to be reclaimed when the system sends a reclamation notification. Valid values:

• true: Enable supplementation of spot instances.

• false: Disable supplementation of spot instances.

When multi_az_policy is set to COST_OPTIMIZED, specifies whether to allow automatic creation of pay-as-you-go instances to meet the required number of ECS instances when spot instances cannot be created due to cost, inventory, or other reasons. Valid values:

• true: Allowed.

• false: Not allowed.

The billing method for the public IP address. Valid values:

• PayByBandwidth: billed on a fixed-bandwidth basis.

• PayByTraffic: billed on a pay-by-data-transfer basis.

The maximum outbound bandwidth of the public IP address of a node. Unit: Mbit/s. Valid values: 1 to 100.

The operating system image type. Specify either this parameter or the platform parameter. Valid values:

• AliyunLinux: Alinux 2 image.

• AliyunLinux3: Alinux 3 image.

• AliyunLinux3Arm64: Alinux 3 ARM image.

• AliyunLinuxUEFI: Alinux 2 UEFI image.

• CentOS: CentOS image.

• Windows: Windows image.

• WindowsCore: Windows Core image.

• ContainerOS: container-optimized image.

The ID of the deployment set.

The desired number of nodes in the node pool.

The name of the worker RAM role.

The metadata access configuration for ECS instances. This parameter is currently available only to users in the whitelist. Submit a ticket to request access.

Specifies whether the node pool is a confidential computing node pool.

Specifies whether to enable the managed node pool. Valid values:

• true: Enables the managed node pool.

• false: Does not enable the managed node pool. Other related configurations take effect only when enable=true.

Specifies whether to enable automatic repair. This parameter takes effect only when enable=true.

• true: Enables automatic repair.

• false: Does not enable automatic repair.

Specifies whether to automatically fix CVEs.

Specifies whether to enable auto upgrade.

This field is deprecated.

The instance ID of the Cloud Enterprise Network (CEN) attached to the enhanced edge node pool.

This field is deprecated.

The instance ID of the Cloud Connect Network (CCN) attached to the enhanced edge node pool.

This field is deprecated.

The region of the CCN instance associated with the enhanced edge node pool.

This field is deprecated.

The network bandwidth of the enhanced edge node pool. Unit: Mbps.

This field is deprecated.

The subscription duration of the enhanced edge node pool. Unit: months.

The kubelet parameter settings.