:Manage nodes

Procedure

1. Open the Nodes page.

   1. Log on to the Elastic High Performance Computing (E-HPC) console.

   2. In the top navigation bar, select a region.

   3. In the left-side navigation pane, choose Resource Management > Nodes.

2. Select the cluster that you want to manage from the Cluster drop-down list, select the node type from the Node Type drop-down list, and then perform operations based on your requirements.

   Note

   Different types of nodes support different operations. For example, management nodes cannot be started, stopped, reset, or deleted. The operations that are displayed in the E-HPC console take precedence.

   Operation	Description	Procedure
   Stop	If the economical mode is enabled for your idle pay-as-you-go compute nodes, you are no longer charged for the computing resources on the nodes after the nodes are stopped. For more information, see Economical mode. Important If a compute node is stopped, the jobs that are running on the node are stopped. Make sure that no jobs are running on a compute node before you stop the compute node.	Select the node that you want to manage from the node list. In the lower part of the node list, choose More Actions > Stop. Check the tips and then click Confirm.
   Start	You can start a node that is in the Stopped state to continue to use the node.	Select the node that you want to manage from the node list. In the lower part of the node list, choose More Actions > Start. Check the tips and then click Confirm.
   Restart	You can restart a node on which exceptions occur. You can restart a node in one of the following modes: Normal restart: If you set Reboot Mode to Normal Reboot, restart commands are sent to the node. Then, the operating system automatically terminates all processes and is restarted. Force restart: If you set Reboot Mode to Force Reboot, data may be lost. We recommend that you perform a force restart only when a normal restart fails. Important After a node is restarted, the jobs that are running on the node are stopped. Make sure that no jobs are running on the node that you want to restart before you restart the node.	Find the node that you want to manage in the node list. Click Reboot in the Operation column that corresponds to the node. Select a restart mode and click Confirm.
   Reset	If you cannot fix a node fault, you can try to reset the node to restore the operating system and software to their initial state. Warning Resetting a node may affect the cluster. Reset a node that is in the Running state only if necessary, for example, to fix faults.	Find the node that you want to manage in the node list. Click Reset in the Operation column that corresponds to the node. Check the tips and then click Confirm.
   Delete	You can delete the compute nodes that you no longer need in a cluster to scale down the cluster. A deleted compute node can be released or retained based on its billing method. If the billing method of the node is pay-as-you-go, the node is released when it is deleted. If the billing method of the node is subscription, the node is retained after it is deleted. If you want to release the node before it expires, you must request for a refund or change the billing method of the node to pay-as-you-go. For more information, see Release an instance. Important Data stored on the node cannot be restored after a node is released. If you want to retain the data on the node, we recommend that you create a snapshot to back up the data before you delete the node. For more information, see Create a snapshot for a disk.	Select the node that you want to manage from the node list. In the lower part of the node list, click Batch Delete. If you want to delete a single node, you can also click Delete in the Operation column that corresponds to the node. Check the tips and then click Confirm.
   Run commands	If you want to quickly maintain a node, for example, to install software or execute O&M scripts, you can run commands.	Select the node that you want to manage from the node list. In the lower part of the node list, click Bulk Run Commands. If you want to run commands on a single node, you can also click Run Command in the Operation column that corresponds to the node. Configure the parameters for the command, enter the command content, and then click Run. For more information about the command parameters, see Send remote commands.
   Change billing methods	ECS instances support the subscription and pay-as-you-go billing methods. You can exchange the billing methods to meet your cost and usage requirements.	Select the node that you want to manage from the node list. In the lower part of the node list, choose More Actions > Change to Subscription or More Actions > Change to Pay-as-you-go. Follow the on-screen instructions to perform subsequent operations.
   Renew	You can renew a subscription node to extend its validity period.	Select the node that you want to manage from the node list. In the lower part of the node list, choose More Actions > Renew. Follow the on-screen instructions to select a renewal duration and complete the payment.
   Change instance types	If the configurations of a node cannot meet your business requirements, you can upgrade or downgrade the instance type of the ECS instance (node). Before you change the instance type of the ECS instance, make sure that the current instance type can be changed. For more information, see Instance families that support instance type changes. Important If you change the instance type of an ECS instance, the ECS instance is stopped, which affects your business. We recommend that you change the instance type when no job computing is performed on the ECS instance. The impacts of instance type changes vary based on the type of the node. The following items list the impacts: Login nodes: You cannot log on to the cluster during instance type changes. Management nodes: You cannot create or view jobs, and scheduling exceptions may occur during instance type changes. However, jobs that are in the Running state are not affected. Compute nodes: Jobs that are running on the node are stopped.	Find the node that you want to manage in the node list. Click the node ID to go to the Instance Details page in the ECS console. Click Change Instance Type next to Type in the Basic Information section. Select the instance type to which you want to change and complete the subsequent operations. For more information, see Change instance types.
   Add to a queue	You can add compute nodes that execute different jobs or tasks to different queues and then manage the nodes by group. For more information, see Manage a queue.	Select the node that you want to manage from the node list. In the lower part of the node list, click Move to. Follow the on-screen instructions to perform subsequent operations.