In the Data Development module, the tasks for which the scheduling type is set as cyclic tasks are referred as the Cyclic tasks. If you enable cyclic tasks in the scheduling system, the tasks run automatically once submitted.
Default display rules are shown as follows.
- Displays workflow tasks
- Displays the tasks under the account currently logged on
Note: For a task submitted before 23:00, an instance is generated automatically at 23:30 the next day to run the task. For a task submitted after 23:30, an instance is generated on the third day to run the task.
Do not operate on the project_etl_start node because it is the root node of a project and the instances of cyclic tasks depend on it. If this node is frozen, instances of cyclic tasks do not run.
This list displays submitted cyclic tasks. You can perform such operations as testing tasks, completing data, adding alarms, changing owners, changing resource groups, and freezing or unfreezing tasks.
Filter: As shown in area ① in the preceding figure, the filtering conditions can be used to filter out the tasks you want. You can make accurate filtrations by task type, task name, owner, tasks modified today, and frozen status.
- The task types for filtering are workflow, node task, and internal node (which refers to a node task in a workflow).
Note: The result of search by task name is subject to other filtering conditions. Only items that match to all those filtering conditions are displayed.
Operation: As shown in area ② in the preceding figure, you can test a task, complete data, freeze, or unfreeze a task, view its instances, add alarms, and change the owner.
Freeze: For frozen tasks, the generated instances are also frozen and do not run directly. To run them, you must unfreeze them and click Re-run.
PatchData: The patch data feature supports parallel submission. After the submission, the same patch data task are divided into several groups. The generated patch data instances run in parallel, which can greatly improve the operation efficiency of the patch data instances.
Batch operation: As shown in area ③ in the preceding figure, you can select tasks in batch to add alarms, change owners, change resource groups, and freeze, or unfreeze tasks.
Note: Changing the resource group is not supported for workflow tasks. The generated testing or data completing instances of frozen tasks are also frozen, and the tasks cannot run until they are unfrozen.
To view the details of a task, you can directly click the task name or workflow name in the task list, and then a DAG of the task information appears.
In the DAG, you can select a task or workflow to view its properties and operation logs. You can also modify properties of the task.
Note: The properties of workflows are different from those of internal nodes in that the code content on the latter can be viewed.
For a workflow task, you can only change its owner and configure the monitoring but cannot change the resource group. In addition, you can view its operation log, which records the operations performed on this task, such as updating workflow content, freezing instance scheduling, completing data, and performing smoke tests.
- Internal node of a workflow
For an internal node of a workflow, you can change its owner and resource group, and configure the monitoring. You can also view its operation log, which records the operations performed on this task, such as updating workflow content, freezing instance scheduling, completing data, and performing smoke tests. You can view but cannot modify the code of the node.
- Right-click a workflow or an internal node
After you right-click a workflow or an internal node, an Action window is displayed, and you can choose to test the task, complete data, freeze, or unfreeze the task, and view its instances.
To view the upstream and downstream dependencies of a workflow or node and to expand or collapse a workflow
By clicking the symbol in area ①, you can expand the upstream and downstream dependencies. By clicking the symbol in area ②, you can expand or collapse the workflow.