Task Management

Last updated: 2019-01-31 16:36:38

PDF

Operation Scenario

Sparkling allows you to manage and schedule DAG tasks, including notebook development tasks and data ingestion tasks. You can choose whether to turn on a task or execute it immediately and view its execution time and execution details. A notebook development task begins with "notebook_" in its name, while a data ingestion task "datasync_". This documents demonstrates how to manage and execute tasks in the Sparkling console.

Steps

Go to Cluster Management and click Tasks in the left pane to enter the task management page.

  • Turn on\off a task
    Click the button in the first column on the page to switch the task status. indicates that the task is turned on, while means that the task is turned off. After the page is refreshed, tasks turned off will not be displayed, but they can be retrieved by clicking Show Paused DAGs at the bottom left corner.
  • View task details
    Click the second column (DAG) to view the task's DAG details. Click the third column (Component) to view the task details. You will be able to view the information about data source, target table, extraction rule and timed tasks for a Datasync task, and you will be redirected to the Notebook page for a Notebook task.
  • View task scheduling time
    You can view the task's cron code in the fourth column (Schedule) and its last execution time in the seventh column (Last Run). Please note that the time zone in this column is UTC, which is 8 hours behind Beijing time.
  • View task status
    The last task execution status can be viewed in the sixth column (Recent Tasks).
  • View task execution history
    The eighth column (DAG Runs) shows the number of times the task succeeded and failed and the running status of the task.
  • Execute immediately\refresh a task
    Click the Trigger Dag button in the last column (Links) to manually trigger the task for immediate execution and click the refresh button to refresh the task status.