After you configure nodes, you need to commit the nodes to Operation Center for testing.


  1. Optional. Commit nodes. If you have committed nodes after configuring them and have not modified the nodes, skip this step.
    1. On the Data Analytics tab, double-click the Workshop workflow. On the workflow editing canvas that appears, click the icon in the upper-left corner.
    2. Select all nodes that can be committed, set Description, select Ignore I/O Inconsistency Alerts, and then click Commit.
    Note In a workspace in the standard mode, after you commit the nodes, you must click the Deploy icon to deploy the nodes to the production environment.
  2. Click Operation Center in the upper-right corner.
  3. On the page that appears, choose Cycle Task Maintenance > Cycle Task in the left-side navigation pane. On the page that appears, click start in the Name column.
  4. In the directed acyclic graph (DAG) that appears on the right, right-click the start node and choose Run > Current and Descendant Nodes Retroactively.
  5. In the Patch Data dialog box that appears, select all nodes, set Data Timestamp to the last week, and then click OK.
    Note For more information about retroactive instances, see Retroactive instances.
  6. On the Patch Data page that appears, view the running status of retroactive instances. You can click Refresh in the upper-right corner to refresh the page and view the running status of the retroactive instances in real time.
    If a retroactive instance fails to be run, right-click the corresponding node in the DAG, select View Runtime Log, and then fix the issue based on logs.
  7. Verify retroactive data generation.
    1. On the Data Analytics tab in DataStudio, choose Business Flow > Workshop > MaxCompute. Right-click Data Analytics and choose Create > ODPS SQL. In the Create Node dialog box that appears, set Node Name to test and click Commit.
    2. Enter the following SQL statement in the SQL editor of the node to query data in the time range from June 11, 2019 to June 17, 2019 in the rpt_user_trace_log table:
      select * from rpt_user_trace_log where dt BETWEEN '20190611' and '20190617' limit 1000;
    3. Click the icon in the upper-left corner. The following figure shows the query result.