All Products
Search

This Product

    DataWorks:Dry-runs of nodes

    Document Center

    DataWorks:Dry-runs of nodes

    Last Updated:Mar 26, 2026

    Node dry-run

    When a node is scheduled by week or month, the scheduling system runs it only on the configured day. On all other days in the scheduling cycle, the system generates a dry-run instance instead of running the node. A dry-run instance completes immediately without processing data, so the scheduling chain stays unbroken even on days when no actual computation is needed.

    A dry-run instance:

    • Returns a success response immediately — running duration is 0 seconds

    • Generates no run logs

    • Produces no data

    • Does not block descendant nodes

    • Occupies no resources

    Dry-run scenarios

    The following scenarios trigger a dry-run. Use the table to identify which scenario applies to your situation, then follow the corresponding steps.

    Scenario Status shown in Operation Center
    Node scheduled for a specific day of the week or month Dry-run
    Real-time instance generated within 10 minutes of scheduling time Dry-run (expired instance)
    Instance manually set to successful Instance Set Successfully
    Scheduling Type set to Dry-run Dry-run
    Node excluded from a temporary workflow during data backfill Unselected Instance in Temporary Workflow

    Scenario 1: Node scheduled for a specific day of each week or month

    Nodes with a weekly or monthly Scheduling Cycle run only on the configured day. The scheduling system generates dry-run instances for all other days in the cycle.

    To configure the scheduling settings:

    1. Double-click the node to open its configuration tab. In the right-side navigation pane, click Properties, then go to the Scheduling Time section. Set the following parameters:

      Parameter Value
      Scheduling Cycle Week or Month
      Custom Time The specific day of each week or month
      Scheduling Time For example, 10:10

      image

    2. Save and deploy the node.

    After deployment, the node generates data only at the scheduling time on the configured day. Dry-run instances are generated for all other days.

    Avoid dry-run instances during data backfill

    To prevent dry-run instances when backfilling data for a weekly- or monthly-scheduled node, set the data timestamp to a day when the node would actually run.

    Example: If the node runs on the first day of every month, set the data timestamp to the last day of the previous month.

    Scenario 2: Real-time instance generated within 10 minutes of scheduling time

    If Instance Generation Mode is set to Immediately After Deployment, the system generates a real-time instance when you deploy the node. Whether that instance generates data depends on the gap between deployment time and scheduling time:

    Deployment time Result
    More than 10 minutes before scheduling time Instance runs normally and generates data
    10 minutes or less before scheduling time Dry-run instance (expired instance generated in real time)

    Example: If the Scheduling Time is 10:09 and the node is deployed at 10:00, the gap is 9 minutes — the system generates a dry-run instance.

    Scenario 3: Instance status manually set to successful

    If an instance is in the Failed status, set it to successful to unblock descendant nodes without generating data:

    1. Go to Operation Center.

    2. In the left-side navigation pane, choose Auto Triggered Node O&M > Auto Triggered Instances.

    3. Find the instance, then choose More > Set Status to Successful in the Actions column.

    The instance status changes to Instance Set Successfully. The instance does not generate data and does not block its descendant nodes.

    image

    Scenario 4: Scheduling Type set to Dry-run

    If the Scheduling Type property is set to Dry-run

    To check or change this setting:

    1. Double-click the node to open its configuration tab. In the right-side navigation pane, click Properties, then go to the Scheduling Policies section.

    2. Check whether Scheduling Type is set to Dry-run.

    image

    Scenario 5: Node excluded from a temporary workflow

    When you backfill data for specific nodes, any intermediate nodes that are not selected become Unselected Instance in Temporary Workflow

    Example: Node C depends on Node B, which depends on Node A. If you backfill data for Node A and Node C only, Node B generates a dry-run instance with the status Unselected Instance in Temporary Workflow.

    To view the affected nodes:

    1. In the upper-left corner, click the image icon, then choose All Products > Data Development and Task Operation > Operation Center.

    2. In the left-side navigation pane, choose Auto Triggered Node O&M > Auto Triggered Nodes.

      image

    3. Select the auto triggered node, then backfill data for the node and its descendant nodes to generate a temporary workflow.

    Troubleshooting

    Daily-scheduled node performs a dry-run unexpectedly

    If a node with a daily Scheduling Cycle is generating dry-run instances, check whether Scheduling Type in the Scheduling Policies section on the Properties tab is set to Dry-run. If so, change it to the appropriate scheduling type.

    T+1 means the scheduling system runs nodes on the next day using data generated on the current day.