Backfill metrics

Create a backfill task

1. On the Instance Overview page of the IoT Platform console, click the ID or alias of the target Enterprise instance.

2. In the left-side navigation pane, select Data Service > Data Metric, and single click Customize Metric.

3. On the Customized Metric tab, click Metric Task in the upper-right corner.

4. Click the Backfill Instance tab, and then click Create Backfill Task.

5. In the Create Backfill Task panel, configure the following parameters and select the metrics to backfill.

   Parameter	Description
   Select backfill date	The end date defaults to the previous day. You can select a date range of up to 30 days.
   Backfill downstream metrics	Specifies whether to backfill the downstream metrics of the selected metric. Note All metrics derived from a specific metric are considered its downstream metrics. If you select this checkbox, all downstream metrics are also backfilled. If you clear this checkbox, only the selected metric is backfilled.

6. Click OK.

   After you create the task, a list of tasks with their statuses appears on the Backfill Instance tab.

   To run the task again, click rerun.

   Important

   After creating a backfill task, wait for its status to change to succeeded before using the corresponding metrics.

   The status of a backfill task changes from initializing to running and then to succeeded.

7. (Optional) In the Actions column of a backfill task, click view. On the Backfill Details page, select a run record to view its run log.

   The Instance Information section at the top of the Backfill Details page displays fields such as Metric Task Name, Metric Code, Service Date, Last Submission Time, Last End Time, Last Execution Status, and Entity. The run log area below displays INFO-level task execution logs organized by timestamp.

View periodic instances

The Periodic Instance tab displays a list of periodically generated statistics for your metrics.

In the Actions column of a backfill task, click view to see statistics for each data cycle, including Service Date, Last Submission Time, Last End Time, Last Execution Status, and run log.

The Instance Information section on the periodic instance details page also includes the Metric Task Name, Metric Code, and Entity fields. In the run record area, you can view the run time and detailed DAG scheduling log for each run. If a periodic instance has a failed status, you can click rerun. If it fails repeatedly, click view to check the failure details, identify the cause of the error, and update the metric configuration.