Auto Scaling：CreateScalingGroup

The name of the scaling group. The name of each scaling group must be unique in a region.

The name must be 2 to 64 characters in length, and can contain letters, digits, underscores (_), hyphens (-), and periods (.). The name must start with a letter or a digit.

If you do not specify this parameter, the value of the ScalingGroupId parameter is used.

The ID of the launch template that provides instance configurations for Auto Scaling to create instances.

The version number of the launch template. Valid values:

• A fixed template version number.
• Default: the default template version.
• Latest: the latest template version.

The ID of the ECS instance. When you create a scaling group, you can specify an existing ECS instance. Auto Scaling obtains the configurations of the ECS instance and automatically creates a scaling configuration from the obtained configurations.

The region ID of the scaling group.

The minimum number of instances that must be contained in the scaling group. When the total number of ECS instances in the scaling group is less than the value of MinSize, Auto Scaling automatically creates ECS instances in the scaling group until the total number reaches the minimum number.

The maximum number of instances that can be contained in the scaling group. When the total number of ECS instances in the scaling group exceeds the value of MaxSize, Auto Scaling automatically removes ECS instances from the scaling group until the total number equals the maximum number.

The value range of MaxSize is directly correlated with the degree of dependency your business has on Auto Scaling. You can go to Quota Center to check the maximum number of instances that a single scaling group can contain.

If a single scaling group can contain up to 2,000 ECS instances, the value range of MaxSize is 0 to 2,000.

The cooldown period of the scaling group after a scaling activity is complete in the scaling group. Valid values: 0 to 86400. Unit: seconds.

During the cooldown period, Auto Scaling does not execute scaling activities that are triggered by CloudMonitor event-triggered tasks.

Default value: 300.

The IDs of the CLB instances that you want to associate with the scaling group. The value can be a JSON array that contains multiple CLB instance IDs. Separate multiple IDs with commas (,).

You can associate only a limited number of CLB instances with a scaling group. Go to Quota Center to check the maximum number of CLB instances that you can associate with a scaling group.

The IDs of the ApsaraDB RDS instances that you want to associate with the scaling group. The value can be a JSON array that contains multiple ApsaraDB RDS instance IDs. Separate multiple IDs with commas (,).

You can associate only a limited number of ApsaraDB RDS instances with a scaling group. Go to Quota Center to check the maximum number of ApsaraDB RDS instances that you can associate with a scaling group.

The instance removal policies. Valid values:

• OldestInstance: removes ECS instances that are added at the earliest point in time to the scaling group.
• NewestInstance: removes ECS instances that are most recently added to the scaling group.
• OldestScalingConfiguration: removes ECS instances that are created based on the earliest scaling configuration.
• CustomPolicy: removes ECS instances based on the custom scale-in policy (Function).

The scaling configuration source specified by the OldestScalingConfiguration setting can be a scaling configuration or a launch template. The CustomPolicy setting takes effect only if you specify it as the first step to remove instances. If you specify CustomPolicy, you must also specify the CustomPolicyARN parameter.

The ID of the vSwitch. If you specify the VSwitchId parameter, the network type of the scaling group is VPC.

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

• PRIORITY: scale ECS instances based on the priority of the vSwitches specified by VSwitchIds. Auto Scaling preferentially scales instances in the zone where the vSwitch of the highest priority resides. If the scaling fails, Auto Scaling scales instances in the zone where the vSwitch of the next highest priority resides.

• COST_OPTIMIZED: create ECS instances that have the lowest unit price of vCPUs during scale-out events and removes ECS instances that have the highest unit price of vCPUs during scale-in events. If you specify preemptible instance types in your scaling configuration, Auto Scaling will preferentially create preemptible instances. You can also specify CompensateWithOnDemand to allow Auto Scaling to create pay-as-you-go instances in the case that preemptible instances cannot be created due to limited stock.

  **

  Note The COST_OPTIMIZED setting takes effect only when your scaling configuration contains multiple instance types or specifically contains preemptible instance types.

• BALANCE: evenly distribute ECS instances across the zones that are specified for the scaling group. If ECS instances are unevenly distributed across the specified zones due to insufficient inventory, you can call the RebalanceInstance operation to evenly distribute the instances across the zones.

  **

  Note When you set MultiAZPolicy to BALANCE, this setting has an equivalent effect to setting MultiAZPolicy to COMPOSABLE and enabling AzBalance to true.

• COMPOSABLE: combine the preceding policies into a custom scaling policy based on your business requirements. Alternatively, you can specify custom parameters to finely control the capacity of the scaling group.

Default value: PRIORITY.

The health check mode of the scaling group. Valid values:

• NONE: Auto Scaling does not check the health status of instances.
• ECS: Auto Scaling checks the health status of instances in the scaling group. If you want to enable instance health check, you can set the value to ECS, regardless of whether the scaling group is of ECS type or Elastic Container Instance type.
• LOAD_BALANCER: Auto Scaling checks the health status of instances in the scaling group based on the health check results of load balancers. The health check results of Classic Load Balancer (CLB) instances are not supported as the health check basis for instances in the scaling group.

Default value: ECS.

The reclaim mode of the scaling group. Valid values:

• recycle: the economical mode

• release: the release mode

• forcerelease: the forced release mode

  **

  Note If you set the value to forcerelease, Auto Scaling will forcibly release the ECS instances that are in the Running state during the scale-out events. Forced release equates to an immediate power-off, resulting in the irreversible deletion of all ephemeral data stored on the instance. Once executed, this action cannot be undone and the lost data cannot be recovered. Exercise caution when you select this option.

• forcerecycle: the forced recycle mode

  **

  Note If you set the value to forcerecycle, Auto Scaling will forcibly shut down the ECS instances that are in the Running state during the scale-out events. Forced recycle equates to an immediate power-off, resulting in the irreversible deletion of all ephemeral data stored on the instance. Once executed, this action cannot be undone and the lost data cannot be recovered. Exercise caution when you select this option.

ScalingPolicy specifies the reclaim mode of the scaling group. RemovePolicy of the RemoveInstances operation specifies the specific instance removal action. For more information, see RemoveInstances .

The client token that is used to ensure the idempotence of the request.

You can use the client to generate the token, but you must make sure that the token is unique among different requests. The token can contain only ASCII characters and cannot exceed 64 characters in length. For more information, see Ensure idempotence.

The minimum number of pay-as-you-go instances that must be contained in the scaling group. Valid values: 0 to 1000. If the number of pay-as-you-go instances is less than the value of this parameter, Auto Scaling preferentially creates pay-as-you-go instances.

The percentage of pay-as-you-go instances in the excess instances when the minimum number of pay-as-you-go instances reaches the requirement. Valid values: 0 to 100.

Specifies whether to supplement preemptible instances. If you set this parameter to true, Auto Scaling creates an instance to replace a preemptible instance when Auto Scaling receives a system message which indicates that the preemptible instance is to be reclaimed.

Specifies whether to automatically create pay-as-you-go instances to meet the requirement on the number of ECS instances when the expected capacity of preemptible instances cannot be provided due to reasons such as cost-related issues and insufficient resources. This parameter is available only if you set the MultiAZPolicy parameter to COST_OPTIMIZED. Valid values:

• true
• false

Default value: true.

The number of available instance types. Auto Scaling evenly creates preemptible instances of multiple instance types that are provided at the lowest cost in the scaling group. Valid values: 1 to 10.

The expected number of ECS instances in the scaling group. Auto Scaling automatically maintains the specified expected number of ECS instances. The DesiredCapacity value cannot be greater than the MaxSize value or less than the MinSize value.

Specifies whether to enable deletion protection for the scaling group. Valid values:

• true: enables deletion protection for the scaling group. This way, the scaling group cannot be deleted.
• false: disables deletion protection for the scaling group.

Default value: false.

The type of the instances that are managed by the scaling group. Valid values:

• ECS: ECS instances.
• ECI: elastic container instances.

Default value: ECS.

The ID of the elastic container instance.

The IDs of the vSwitches. If you specify VSwitchIds, VSwitchId is ignored. If you specify VSwitchIds, the network type of the scaling group is VPC.

If you specify multiple vSwitches, take note of the following items:

• The vSwitches must belong to the same VPC.
• The vSwitches can belong to different zones.
• vSwitches are sorted in ascending order based on their priorities. The first vSwitch has the highest priority. If Auto Scaling fails to create ECS instances in the zone where the vSwitch of the highest priority resides, Auto Scaling attempts to create ECS instances in the zone where the vSwitch of the next highest priority resides.

The lifecycle hooks.

The backend vServer group that you want to associate with the scaling group.

The tags that you want to add to the scaling group.

Details of the instance types that you specify by using the Extended Configurations feature of the launch template.

The Application Load Balancer (ALB) server groups.

The server groups.

Specifies whether to evenly distribute instances in the scaling group across multiple zones. This parameter takes effect only if you set MultiAZPolicy to COMPOSABLE. Valid values:

• true
• false

Default value: false.

The allocation policy of instances. Auto Scaling selects instance types based on the allocation policy to create the required number of instances. The policy can be applied to pay-as-you-go instances and preemptible instances. This parameter takes effect only when you set the MultiAZPolicy parameter to COMPOSABLE. Valid values:

• priority: Auto Scaling selects instance types based on the specified order of the instance types to create the required number of instances.
• lowestPrice: Auto Scaling selects instance types that have the lowest unit price of vCPUs to create the required number of instances.

Default value: priority.

The allocation policy of preemptible instances. You can use this parameter to individually specify the allocation policy of preemptible instances. This parameter takes effect only if you set the MultiAZPolicy parameter to COMPOSABLE. Valid values:

• priority: Auto Scaling selects instance types based on the specified order of the instance types to create the required number of preemptible instances.
• lowestPrice: Auto Scaling selects instance types that have the lowest unit price of vCPUs to create the required number of preemptible instances.

Default value: priority.

The maximum life span of an instance in the scaling group. Unit: seconds.

Valid values: 86400 to the value of the Integer.maxValue parameter.

Default value: null.

The Alibaba Cloud Resource Name (ARN) of the custom scale-in policy (Function). This parameter is available only if you specify CustomPolicy as the first step to remove instances.

The ID of the resource group to which you want to add the scaling group.

The load balancer configurations.

The health check mode of the scaling group.

The databases that you want to attach to the scaling group.

The period of time required by the ECS instance to enter the Stopped state. Unit: seconds. Valid values: 30 to 240.

• This parameter takes effect only if you set ScalingPolicy to release.

• If you specify this parameter, the system will wait for the ECS instance to enter the Stopped state for the specified period of time before continuing with the scale-in operation, regardless of the status of the ECS instance.

• If you do not specify this parameter, the system will wait for the ECS instance to stop before continuing with the scale-in operation. If the ECS instance is not successfully stopped, the scale-in process will be rolled back and considered failed.

The capacity options.