Data Governance Center provides a complete node undeployment solution for invalid nodes that are identified during data governance. The solution supports various features such as automatically identifying and recommending nodes that need to be undeployed or allowing you to specify the nodes that you want to undeploy, evaluating the risks of node undeployment, sending notifications about node undeployment, and managing the node undeployment process. The solution helps you resolve various issues. For example, you must manually identify node undeployment risks, you cannot undeploy nodes whose descendant nodes are normally scheduled, and you dare not undeploy nodes easily when unpredictable risks exist. The solution helps improve the node undeployment efficiency. This topic describes how to create and manage node undeployment plans.
Background information
Graceful undeployment supports analysis of undeployment impacts and undeployment process management that are not supported in the traditional undeployment. To-be-undeployed notifications of nodes can be sent to related personnel at each phase of the node undeployment process by using various methods.
No. | Operation | Description |
1 | You can manually specify the nodes that you want to undeploy or directly select the nodes that are automatically recommended by the system. You can evaluate the impact scope and risks of node undeployment. | |
2 | You can manage the entire node undeployment process:
Note Silence period: the period of time you must wait before you can perform the next step after the step in the current phase starts to be performed. For example, if you specify a silence period for a node whose scheduling you want to suspend, after the silence period expires, you can undeploy the node. | |
3 | The step in each phase of the node undeployment process can be manually or automatically triggered. | |
4 | You can view the details of a node undeployment plan and perform various operations on the nodes that are specified in the plan, such as viewing the details of the nodes, canceling node undeployment, or rolling back the undeployment operation. |
Limits
Only users that are assigned the Workspace Manager or O&M role can create node undeployment plans. For information about how to assign roles to workspace members, see the "Add a RAM user to a workspace as a member and assign roles to the member" section in Manage permissions on workspace-level services.
You can add nodes only in the same workspace to a node undeployment plan.
Only users that are assigned the Workspace Manager role have all permissions on a node undeployment plan. Users that are assigned the O&M role have permissions only on the node undeployment plans that are created by themselves. They have only the Read permissions on the node undeployment plans that are created by other users.
A maximum of 200 nodes can be added to a node undeployment plan.
Precautions
You cannot select a node that has been added to a node undeployment plan. This indicates that a node can be added to only one node undeployment plan.
Special types of nodes cannot be added to a node undeployment plan. For example, you cannot add the root node of a workspace to a node undeployment plan or add a zero load node that you create for a node undeployment plan to the node undeployment plan.
If a node that you want to undeploy has descendant nodes and the scheduling dependencies between the descendant nodes and the node are not removed within the specified period of time, the system generates a zero load node and configures the zero load node as the ancestor node of the descendant nodes.
The operations that you need to perform to create a node undeployment plan vary based on the type of node. You can follow on-screen instructions to create a node undeployment plan for a type of node.
Create a node undeployment plan
Go to the Data Governance Center page.
Log on to the DataWorks console. In the left-side navigation pane, choose . On the page that appears, select the desired workspace from the drop-down list and click Go to DataStudio.
]On the DataStudio page, click the
icon in the upper-left corner and choose .
In the top navigation bar of the Data Governance Center page, click Workbench. The Governance issues page appears.
In the left-side navigation pane, click Elegant offline in the Special governance section.
In the upper part of the Elegant offline page, select the workspace to which the nodes you want to undeploy belong from the Workspace drop-down list. Only the workspaces to which you are added as a member are displayed in the drop-down list.
To join a workspace, you must contact the administrator of the workspace. For more information, see the "Add a RAM user to a workspace as a member and assign roles to the member" section in Manage permissions on workspace-level services.
On the Elegant offline page, click New plan.
You can also find the nodes that you want to undeploy from the workspace perspective on the Governance issues page, and click New plan in the Actions column of the nodes to go to the New plan page.
Select the nodes that you want to undeploy
You can click Add task in the Offline object section on the New plan page to add the nodes that you want to undeploy. In the Add task dialog box, you can view the related information about the nodes. For example, you can view the level-1 descendant nodes of the nodes, the number of tables generated by the nodes, and whether data in the generated tables is consumed. The information helps you evaluate the impacts and risks of the node undeployment.
You cannot select a node that has been added to a node undeployment plan. This indicates that a node can be added to only one node undeployment plan.
Special types of nodes cannot be added to a node undeployment plan. For example, you cannot add the root node of a workspace to a node undeployment plan or add a zero load node that you create for a node undeployment plan to the node undeployment plan.
Configure settings for node undeployment
You can configure fine-grained settings for the entire node undeployment process in the Processing flow section of the New plan page. The following phases are included: start undeployment, delay node scheduling, suspend node scheduling, undeploy nodes, rename output tables of the nodes, remove the output tables, and complete node undeployment. You can decide whether to perform the step in a specific phase of the entire node undeployment process. If you do not select the check box of a specific phase, you do not need to configure the parameters in this phase and you are navigated to the next phase.
Parameter | Description |
Send Notification | The method that is used to send node undeployment notifications. Valid values: In-system prompts, Mail, and DingTalk robot. For information about how to configure a DingTalk chatbot, see the "Scenario practices: Send alert notifications to a DingTalk group" section in Create a custom alert rule. Note For information about system prompt messages, go to the Data Governance Center page and click Workbench. In the left-side navigation pane, click Governance reminder in the My agent section. |
Notify Personnel | The personnel to whom a node undeployment notification is sent. The owner of a to-be-undeployed node, the owner of the output table of a to-be-undeployed node, and the owner of the baseline to which a to-be-undeployed node belongs are supported. The valid values of the parameter vary based on the undeployment phase. |
Silence Duration | The period of time you must wait before you can perform the next step after the step in the current phase starts to be performed. For example, if you specify a silence period for a node whose scheduling you want to suspend, after the silence period expires, you can undeploy the node. |
Specify the method to execute the node undeployment plan
You can specify the method to perform the step in each phase of the node undeployment plan. The step in each phase can be manually or automatically triggered.
Manual trigger: The step in each phase of the node undeployment process must be confirmed by the creator of the node undeployment plan before the next step can be performed.
Automatic trigger: The step in each phase of the node undeployment process starts to be performed after the configured silence period expires. Then, you are navigated to the step in the next phase.
View and manage the node undeployment plan
On the Elegant offline page, you can view the list of node undeployment plans. You can find the desired node undeployment plan and the desired nodes to execute the plan for, cancel the plan for, or remove the nodes in the node undeployment plan.
The nodes that have been undeployed are not affected if you cancel node undeployment for nodes that are in the Undeploying state in the same node undeployment plan.
If you delete a node undeployment plan, the system first cancels node undeployment for nodes that are in the Undeploying state in the plan. Nodes that have been undeployed are not affected. After the node undeployment for the nodes is canceled, the plan is deleted. The deletion operation is irreversible.
You can click Execution Details in the Actions column of a node undeployment plan to view the configuration details of the plan. You can also perform the operations that are described in the following table on a single node that you want to undeploy in the plan.
Operation | Description |
View details | You can view the undeployment details of a single node, such as the undeployment progress, node status, and the execution details of each phase. You can also perform various operations on the node, such as canceling undeployment for the node and rolling back the undeployment operation for the node.
|
Go to Operation Center | You can go to the Operation Center in the production environment to view the O&M details of the node. |
Go to the configuration tab of the node | You can go to the configuration tab of the node in DataStudio to edit the node. |