This topic describes how to configure recurrence and dependencies for a node in the DataWorks console.

Note In this topic, batch synchronization node write_result created in Create a batch synchronization node is used as an example and the recurrence is set to weekly.

DataWorks has a powerful scheduling engine to trigger nodes based on the recurrence and dependencies of nodes. DataWorks guarantees that tens of millions of nodes run accurately and punctually per day based on directed acyclic graphs (DAGs). In the DataWorks console, you can set the recurrence to minutely, hourly, daily, weekly, or monthly. For more information, see Scheduling properties.

Configure recurrence for the batch synchronization node

  1. After creating batch synchronization node write_result, double-click the batch synchronization node to configure it.
  2. Click the Properties tab in the right-side navigation pane to configure recurrence for the batch synchronization node.
    Parameter Description
    Start Instantiation The time when an instance is generated for the node. Valid values: Next Day and Immediately After Deployment. Select an option as required.
    Execution Mode The mode in which the node is run. Valid values: Normal and Dry-Run. Select an option as required.
    Rerun Specifies whether to allow the node to be rerun. Valid values: Allow Regardless of Running Status, Allow upon Failure Only, and Disallow Regardless of Running Status. Select an option as required.
    Auto Rerun upon Error Specifies whether to automatically rerun the node when an error occurs. This parameter only appears when the Rerun parameter is set to Allow Regardless of Running Status or Allow upon Failure Only. After you select this check box, the node is automatically rerun when an error occurs. This parameter does not appear if you set the Rerun parameter to Disallow Regardless of Running Status. In this case, the node is not rerun when an error occurs.
    Start and End Dates The validity period of the node. Specify the start and end dates of the validity period as required.
    Skip Execution Specifies whether to skip execution of the node.
    Instance Recurrence The recurrence of the node. Valid values: Minute, Hour, Day, Week, and Month. In this example, set the value to Week.
    Customize Runtime Specifies whether to run the node at the specified time. This parameter is selected by default.
    Run At The specific day and time when the node is run. For example, you can configure the node to run at 02:00 every Tuesday.
    CRON Expression The CRON expression of the time you specified, which cannot be modified.
    Cross-Cycle Dependencies Specifies whether the node depends on the result of the last cycle.

Configure dependencies for the batch synchronization node

After configuring recurrence for batch synchronization node write_result, you can continue to configure dependencies for the batch synchronization node.

You can configure the ancestor node on which the batch synchronization node depends. After that, the scheduling system triggers the batch synchronization node only after the instance of the ancestor node is run.

For example, the instance of the batch synchronization node is not triggered until the instance of its ancestor node insert_data is run.

By default, the scheduling system creates a node named in the format of Workspace name_root for each workspace as the root node. If no ancestor node is configured for the batch synchronization node, the batch synchronization node depends on the root node.

Commit the batch synchronization node

Save the configuration of batch synchronization node write_result and click Commit icon to commit the node to the scheduling system.

A node must be committed to the scheduling system so that the scheduling system can automatically generate and run instances for the node. The scheduling system runs these instances at the specified time from the next day according to the recurrence property.
Note If you commit a node after 23:30, the scheduling system automatically generates and runs instances for the node from the third day.

What to do next

Now you have learned how to configure recurrence and dependencies for a batch synchronization node. You can proceed with the next tutorial. In the next tutorial, you will learn how to perform O&M on the committed node and troubleshoot errors based on the operational logs. For more information, see Run a node and troubleshoot errors.