Dataphin:Recurring task O&M

Area

Description

① Filter and search area

• Filter: Quick filters include My Tasks, Suspended Tasks, and Dry Run Tasks, plus a project filter. Additional options:

  • My Tasks: Lists tasks owned by the current user.

  • Suspended Tasks: Lists tasks in the current project that are configured for suspended scheduling.

  • Dry Run Tasks: Lists tasks in the current project that are configured for dry run scheduling.

  • Project: Filters tasks by the project they belong to. You can select multiple projects for which you have O&M - Access Directory permissions.

  • Task Type: For Data_distill projects, you can select script tasks. For other projects, you can select integration and computing tasks and modeling tasks.

    • Integration and computing tasks: The supported script tasks vary by compute engine. See the page for the available options.

    • Modeling tasks: Includes dimension logic tables, fact logic tables, and aggregate logic tables.

  • Scheduling Mode: Filters tasks by normal, dry run, or suspended scheduling. Scheduling modes.

  • Priority: Filters tasks by the runtime priority of their instances: Highest, High, Medium, Low, or Lowest.

    Note

    If the baseline feature is enabled, only Medium, Low, and Lowest priorities are supported for tasks.

  • O&M Engineer: Filters tasks by their assigned O&M engineer.

  • Scheduling Period: Filters tasks by their scheduling period: minute, hour, day, week, month, or year.

  • Last Updated: Filters tasks by a specific update date or a date range.

  • Scheduling Resource Group: Filters tasks by their configured scheduling resource group.

    Note

    This filter is available only for integration and computing tasks.

  • Tag: Filters tasks by one or more tag values.

  To clear all filters, click Reset.

• Search: Enter a node ID or task name for fuzzy search. Click the icon to open the Batch Search dialog. In Batch Search, enter multiple task IDs or upload a file. Combines with other filters.

  • Batch Search - Manual Input: In Batch Search, select Manual Input, enter up to 5000 task IDs (one per line), and click OK.

  • Batch Search - Upload File: In Batch Search, select Upload File and click Upload to upload an Excel file. The system reads task IDs from the first column (rows 2–5001) of the first sheet. Click Download Task ID List Template for a template.

② List operations

• Customize Columns: Click the icon to select which columns to show and reorder them.

  Note

  • The first column and the Actions column cannot be hidden or reordered.

  • Clearing your browser cache or switching browsers resets the table header settings to their defaults.

• Refresh: Click the icon to refresh the current list.

③ Recurring task list area

Lists all recurring tasks from projects where you have O&M - Access Directory permissions, organized into Integration and Computing Tasks and Modeling Tasks tabs. The Actions column shows available operations. Column widths are adjustable and saved in your browser (cleared with history/cookies).

• Integration and computing tasks: Supported operations for integration and computing tasks.

• Modeling tasks: Supported operations for modeling tasks.

④ Batch operations area

Manage multiple recurring tasks at once with batch operations.

• Integration and computing tasks: Supported batch operations for integration and computing tasks.

• Modeling tasks: Supported batch operations for modeling tasks.

Icon

Scheduling mode

Description

None

Normal scheduling

The task runs normally according to the configured scheduling period and start time.

Dry run scheduling

The task triggers on schedule but only performs a dry run, which immediately returns success without executing the task logic.

Suspended scheduling

The system generates instances in a suspended state. Click Resume to run an instance. Use this mode when a task and its downstream dependencies do not need to run temporarily. For example, suspend a task to adjust its calculation logic without affecting downstream data.

Details:

• After suspension, new instances are still generated on schedule but remain in a suspended state. This is ideal when a node and its downstream nodes do not need scheduling, but the task should not be taken offline.

• Resuming a task changes its scheduling from suspended to normal. To run only specific instances, manually resume them individually.

Area

Description

① Node information

Shows a summary of the selected node. Click View Node Details for full details.

• Task Overview: Displays basic information about the current node, such as its name, ID, type, tenant, project, and data domain.

• Operation Logs: Shows the operation time, operator, and the action performed.

• Node Code/Materialized Code: Shows the code for the current node. You can view all its versions in the production environment. Click the icon to compare versions. Click the icon to view version information, including creation time and modifier.

• Task Parameters: Shows the task parameters for the current node.

Integration and computing task node: Includes Task Overview, Operation Logs, and Node Code.

Modeling task node: Includes Task Overview, Operation Logs, and Materialized Code.

Real-time task: Includes Node Details, Operation Logs, Node Code, and Task Parameters.

Cross-tenant node (a node that belongs to another tenant): Shows only the Task Overview.

② Node filter area

Set the number of upstream and downstream levels to expand. Search by node ID or name to locate a specific node in the DAG.

③ Scheduling dependency graph

Displays the scheduling dependency graph. Expand upstream and downstream nodes to perform O&M operations. Hover over a node to view its name, type, scheduling mode, O&M engineer, and description. Cross-tenant nodes are tagged and limited to expanding immediate parent and child nodes.

• Integration and computing task DAG nodes: Supported operations for integration and computing task DAG nodes.

• Modeling task DAG nodes: Supported operations for modeling task DAG nodes.

④ Canvas adjustment area

Controls for the DAG display: set zoom level (default 100%, range 20%–200%), fit to canvas, or switch to full-screen mode. Also shows the main node ID and name.