DataWorks:Appendix: Forcefully rerun the descendant instances of an auto triggered node instance

Limits

• Only DataWorks Professional Edition or a more advanced edition supports forceful rerun of the descendant instances of an auto triggered node instance. If your DataWorks service does not meet the requirements, you must upgrade it to Professional Edition or a more advanced edition first. For more information, see Billing of DataWorks advanced editions.
• You can use only a workspace administrator, tenant administrator, or Alibaba Cloud account to forcefully rerun the descendant instances of an auto triggered node instance.

• When you filter the descendant instances that you want to rerun by workspace, you can select only the workspaces in which you are the administrator.

• You can forcefully rerun only the descendant instances of an auto triggered node instance that meets the following requirements: The data timestamp of the auto triggered node instance is the previous day or the day before the previous day. The state of the auto triggered node instance is Succeeded or Failed.

  If the auto triggered node instance does not meet the preceding requirements, the Force Rerun Downstream option is dimmed.

• Each tenant in a region can forcefully rerun the descendant instances of an auto triggered node instance only once at the same point in time.

Precautions

• If you forcefully rerun the descendant instances of an auto triggered node instance, the states of the descendant instances are set to Not running. If the descendant instances that you want to forcefully rerun are running, the descendant instances are forcefully stopped when you rerun them.
• For an auto triggered node, if you set Rerun to Disallow Regardless of Running Status in the Schedule section of the Properties tab, you will fail to forcefully rerun the descendant instances of the instance generated by the auto triggered node. For more information about how to change the value of the Rerun parameter, see Rerun properties.

Procedure

1. Go to the Force Rerun Downstream dialog box of an auto triggered node instance in Operation Center.
   Go to the Cycle Instance page in Operation Center, find the auto triggered node instance whose descendant instances you want to forcefully rerun, and then open the directed acyclic graph (DAG) of the auto triggered node instance. Make sure that the data timestamp of the auto triggered node instance is the previous day or the day before the previous day and that the state of the auto triggered node instance is Succeeded or Failed. In the DAG, right-click the auto triggered node instance and choose Emergency Operations > Force Rerun Downstream. The Force Rerun Downstream dialog box is displayed.
2. Configure the parameters in the Force Rerun Downstream dialog box.
   1. Configure the parameters in the Configure Basic Settings step.
      

      Parameter	Description
      Data Timestamp	The data timestamp of the auto triggered node instance whose descendant instances you want to forcefully rerun. You can forcefully rerun only the descendant instances of an auto triggered node instance whose data timestamp is the previous day or the day before the previous day.
      Start Node	The instance from which the forceful rerun starts in the current workspace.
      Rerun Start Point	Specifies whether the forceful rerun starts from the current auto triggered node instance or from the level-1 descendant instances of the auto triggered node instance.
      Rerun Cross-Day Instance	Specifies whether to forcefully rerun the instance (Instance B) that depends on the current instance (Instance A) when you rerun the current instance. The data timestamp of Instance B is the current day, and the data timestamp of Instance A is the previous day. For example, in the preceding figure, Instance B is generated by a node scheduled by day and depends on Instance A. The data timestamp of Instance A is the previous day, and the data timestamp of Instance B is the current day. If you want to forcefully rerun Instance A, you can determine whether to rerun Instance B. If you set Rerun Cross-Day Instance to Yes, Instance B is rerun after Instance A is rerun.

   2. Click Next. In the Please note message, click Confirm.
      Notice After you click Confirm, you cannot modify the parameter settings in the Configure Basic Settings step. Make sure that the parameter settings are correct. This reduces computing overheads and shortens your waiting time.
   3. Select workspaces to filter the descendant instances that you want to rerun by workspace.
      In the Workspaces section on the left, select the workspaces to which the descendant instances you want to rerun belong and click the icon to add the workspaces to the Selected Workspaces section on the right. The descendant instances in the selected workspaces will be forcefully rerun. Dry runs are performed for descendant instances in workspaces that are not selected.

      Note

      When you filter the descendant instances that you want to rerun by workspace, you can select only the workspaces in which you are the administrator.

   4. Click Next. In the Please note message, click Confirm.
      Notice After you click Confirm, you cannot modify the parameter settings in the Configure Basic Settings step. Make sure that the parameter settings are correct. This reduces computing overheads and shortens your waiting time.
   5. Configure the parameters in the Confirm Settings step.

      Parameter	Description
      Total Forcefully Rerun Instances	The total number of instances to be forcefully rerun.
      Affected Baselines	The baselines whose output data will be affected by the forceful rerun. For more information about baseline monitoring, see Overview.
      Force Rerun Downstream Reason	The reason why you want to forcefully rerun the descendant instances of the current auto triggered node instance.

   6. Select the check box and click Confirm Rerun. Then, the system changes the states of the related descendant instances to Not running and forcefully rerun the descendant instances.

Example

• Scenario: You want to forcefully rerun the descendant instances of an auto triggered node instance that is generated by Node A. You select to rerun the descendant instances only in Workspace A.
• Rerun details: The instance that is generated by Node C in Workspace A is forcefully rerun. The instance that is generated by Node B in Workspace B is not forcefully rerun.