DataWorks:Node and workflow deployment

How it works

The deployment panel controls production task deployment and undeployment. It supports pre-deployment checks with custom check items for scheduling governance. Deploy new scheduling nodes to the production system to enable automated, recurring scheduling. To maintain stability, update task code in Data Studio and deploy through this panel — this prevents unauthorized modifications to production tasks.

DataWorks supports two deployment modes:

• incremental deployment: Deploys only modified nodes. Faster and ideal for a single node or small batch.

• full deployment: Deploys all nodes in the workflow. Ensures workflow integrity and consistency.

Prerequisites

Before deploying, debug and run all nodes and workflows. Complete scheduling configuration for any recurring workflows and tasks.

• Manually triggered tasks and workflows: Create a node for a manually triggered workflow and Create a manually triggered workflow.

• Recurring tasks and workflows: Create a node for a recurring workflow and Create a recurring workflow.

• Recurring scheduling: Configure node scheduling.

Usage notes

• Changes made during the full instance generation period (23:30 to 24:00 daily) take effect on the third day.

• Only users with O&M permissions can deploy a deployment package.

• In addition to role-based permission limits, deployment is restricted by check items in the process. If a deployment fails, check its status on the Publish panel.

Deploy a task

1. Go to the Workspaces page in the DataWorks console. In the top navigation bar, select a desired region. Find the desired workspace and choose Shortcuts > Data Studio in the Actions column.

2. Open the task details page.

   In Data Studio, open the node or workflow you want to deploy.

3. Deploy the task.

   On the toolbar, click Publish to open the deployment panel. Choose to deploy to the development environment only, or to both development and production.

   • Deploying to development generates a node ID. You can then debug and run the node in the development Operation Center.

     Important

     The development Operation Center does not automatically run scheduled tasks. Tasks on the Auto Triggered Nodes page will not generate recurring instances.

   • Deploying to production adds the node to the scheduling system for recurring runs. View and manage the node in the Operation Center.

Undeploy a task

1. Go to the Workspaces page in the DataWorks console. In the top navigation bar, select a desired region. Find the desired workspace and choose Shortcuts > Data Studio in the Actions column.

2. Undeploy or delete the task.

   In Data Studio, right-click the node or workflow to undeploy and select Delete. Two scenarios apply:

   • If the task is not deployed to production: Click Confirm in the confirmation dialog. The task is removed from the directory tree and moved to the recycle bin.

   • If the task is deployed to production: The deployment panel opens. Apply the deletion through the deployment process:

     1. At the Dev Offline step, the node remains in the directory tree but its node ID is removed. A version record for Dev Offline is generated.

     2. At the Prod Offline step, the task is undeployed from production. You can then Delete the node from the directory tree.

Deployment history

Click the icon on the deployment panel to view deployment history.

Next steps

After deployment, click Supplementary data or Go to operation and maintenance in the deployment panel.

• Supplementary data: Backfills data for the current node only. For advanced backfill features, use Manage data backfill instances in Operation Center.

• Go to operation and maintenance: Opens Operation Center, where you can view running status or manually trigger the task.