Manage node pools

This topic describes how to manage node pools in the Kingsoft Cloud Container Engine (KCE) console. For example, you can modify node pool information, node template configuration, overall configuration of a node pool, and auto scaling configuration, and manage specific nodes in node pools.

Modify node pool information

You can modify node pool information, such as the name, auto scaling, node quantity, node quantity range, labels, and taints of a node pool.

1. Log in to the KCE console.
2. In the left navigation pane, click Cluster.
3. Click the ID of the cluster in which you want to modify node pool information. The cluster details page appears.
4. In the left navigation pane, choose Manage Nodes > Node Pool. The Node Pool page appears.

5. Find the target node pool and click Modify Node Pool Information in the Operation column.

6. In the dialog box that appears, modify the following node pool information as required:

   • Name: the custom name of the node pool. The name must be 2 to 64 characters in length and can contain only letters, digits, and the following special characters: - , . ! $ ^ * ( ) % # & + / : ; < = > [ ] _ ` { | } ~.

   • Auto Scaling: specifies whether to disable auto scaling.

     Note

     • You can disable auto scaling as required. After you disable auto scaling, you cannot enable it again.

   • Nodes: the number of nodes in the node pool. You can modify this parameter within the node quantity range.
   • Node Quantity Range: the quantity range of nodes in the node pool. The number of nodes is adjusted within this range.
   • Label: the label to be attached to the nodes in the node pool. Labels allow you to implement flexible scheduling. You can customize labels based on your business requirements.
   • Taints: the taints to be attached to nodes in the node pool.
   • Updates on labels and taints apply to all existing nodes in the node pool: specifies whether to apply the updates on labels and taints to all existing nodes in the node pool. You can select or clear this option as required.

     Note

     • 1. If you select this option, the update on labels and taints takes effect on the existing and newly added nodes in the node pool, which may cause the rescheduling of pods on the existing nodes. Exercise caution when you perform this operation.
     • 2. If you do not select this option, the update on labels and taints takes effect only on newly added nodes.

7. Click OK.

Modify node template configuration

You can modify the following configuration of a node template: node configuration, node type, operating system, system disk, data disk, data disk mount, container storage directory, subnet, network expansion policy, security group, scripts to run before and after deployment, and cordon settings.

1. On the Node Pool page, click the name of the node pool for which you want to modify the node template configuration. The node pool details page appears.
2. Click Edit next to Node Template Configuration, as shown in the following figure.
3. Modify the configuration as required and click Update.

   Note

   • The update on node template configuration takes effect only on newly added nodes, without affecting the existing nodes in the node pool.

Modify overall configuration of a node pool

You can modify node pool information and node template configuration separately or modify the overall configuration of a node pool at a time.

1. On the Node Pool page, find the target node pool and click Manage in the Operation column.

2. On the Manage Node Pool page, modify the configuration as required and click Update.

Modify auto scaling configuration

If auto scaling is enabled for a node pool, you can modify the scale-out policy and automatic scale-in configuration for the corresponding cluster as required.

1. On the Node Pool page, click the edit icon for Cluster Scaling-Up Policy or Cluster Automatic Scaling-Down.

2. In the dialog box that appears, modify the following information as required:

   • Cluster Scaling-Up Policy: If a cluster has more than one node pool with auto scaling enabled, you can specify the policy for selecting the node pool to scale out. Valid values:
     • random: randomly selects a node pool.
     • most-pods: selects the node pool with the maximum capacity. This node pool has more resources to schedule pods.
     • least-waste: selects the node pool with the least available resources after pod scheduling.
   • Cluster Automatic Scaling-Down: This option is cleared by default. If you select this option, a scale-in will be triggered when a specific proportion of node resources in the cluster are idle. For more information, see Configure auto scaling for a cluster.
   • Scaling-Down Configuration: This configuration is available only when automatic scale-in is enabled.
     • Max. Scaling-Down Concurrency: the maximum number of idle nodes that can be removed at a time in a scale-in. If pods exist, only one node can be removed at a time. The default value is 10. Set this parameter as required.
     • The Allocated Or Available Node Resources Is Less Than: the maximum proportion of allocated node resources in all available node resources for triggering analytics on the scale-in condition. The value ranges from 0 to 80. The default value is 50. Set this parameter as required.
     • Node Satisfies The Scaling-Down Condition: the duration in which the node meets the scale-in condition, measured in minutes. After the duration elapses, the node will be removed for a scale-in. The default value is 10. Set this parameter as required.
     • Cluster Scaling-Up: the duration in which no scale-out is performed, measured in minutes. After the duration elapses, the system analyzes the scale-in condition. The default value is 10. Set this parameter as required.
3. Click OK.

Note

• 1. The update on auto scaling configuration takes effect only on node pools with auto scaling enabled.
• 2. By default, the Cluster Scaling-Up Policy parameter is set to random and the Cluster Automatic Scaling-Down option is cleared.

Manage nodes in a node pool

Add existing nodes to a node pool

To add existing nodes in a cluster to a node pool, perform the following steps. Make sure that the nodes have not been added to other node pools.

1. On the Node Pool page, click the name of the target node pool. The node pool details page appears.
2. Click Manage Nodes.
3. On the Manage Nodes tab, click Adding Existing Nodes.
4. On the Add Existing Nodes page, select available nodes from the VPC where the cluster resides and the subnet where the node pool resides to add the nodes to the node pool.
5. Click OK.

   Note

   • 1. Only nodes in Pay-By-Daily-Config or Pay-As-You-Go billing mode in the cluster can be added to the node pool.
   • 2. You can add only started instances in the cluster to the node pool.

Enable/Disable scale-in protection

You can enable or disable scale-in protection as required. After scale-in protection is enabled, the node is not removed during scale-in or node quantity adjustment. To enable scale-in protection, perform the following steps:

1. On the Node Pool page, click the name of the target node pool. The node pool details page appears.
2. Click Manage Nodes.
3. On the Manage Nodes tab, find the target node and click Scale-in Protection in the Operation column, as shown in the following figure.
4. In the message that appears, click OK.

   You can also disable scale-in protection by following the preceding procedure. After scale-in protection is disabled, the node may be removed for scale-in and node quantity adjustment.

Remove nodes from a node pool

1. On the Node Pool page, click the name of the target node pool. The node pool details page appears.
2. Click Manage Nodes.
3. On the Manage Nodes tab, find the node that you want to remove and click Remove Node in the Operation column, as shown in the following figure.
4. In the dialog box that appears, select If you delete a node, the data on its system disk and data disks is deleted and cannot be recovered. Exercise caution when you perform this operation. as required, and click OK.

Other operations

You can cordon or drain nodes in a node pool as required. For more information, see Cordon or drain a node.