Dataphin:Manage submitted instances

Block

Description

① Filter and search area

You can filter submitted instances by entering the instance object name or ID. This is useful when you know the instance object ID or name and want to filter out the required submitted instances.

• Submitted By Me: Select this option to filter out submitted instances that were submitted by the current operating account.

• Data Backfill Failed: Select this option to filter out submitted instances that were submitted by the current operating account and have a data backfill status of failed.

• Data Backfill Incomplete: Select this option to filter out submitted instances that were submitted by the current operating account and have a data backfill status of running or incomplete.

• Select Project: Filter by the project to which the node belongs. You can select multiple projects for which the current user has O&M - Access Folder permissions.

  Note

  After you filter by project, the list displays only the instances of the selected projects in the data backfill workflow.

• Filter: Click Filter to expand the filter area. To filter again, you can click Reset to quickly clear all filter conditions and restore the system defaults. The supported filter options include the following:

  • Instance Type: Filter by instance type. You can select one or more instance types that the current tenant supports.

  • Instance Running Status: You can filter submitted instances by their running status. The supported statuses include Success, Failed, Running, Not Run, Waiting for Schedule Resource, Waiting for Scheduled Time, and Throttling.

  • Data Backfill Status: You can filter submitted instances based on their data backfill status. Data backfill statuses include Succeeded, Failed, Not Running, and Running. For more information, see Data backfill instance status description.

  • Instance Owner: You can filter submitted instances based on their owner.

  • Data Backfill Submitter: You can filter submitted instances based on their submitter.

  • Data Timestamp: By default, all submitted instances are displayed. You can choose to view submitted instances for Today, Yesterday, or A Specific Date Range.

  • Start Running Time: By default, submitted instances that started running today are displayed. You can choose to view submitted instances that started running Today, Yesterday, or On A Specific Day.

  • Submitted On: You can filter by submission date, including Today, Yesterday, All, or submitted instances on a specific day.

  • Resource Group: Filter by the schedule resource group specified for the instance at runtime.

  • Tag: Select one or more node tag values to filter the instances.

② List operations

• Switch View - Tree View: Click the icon. The system automatically merges instances that have the same data backfill node name and groups them by data timestamp. For more information, see View the submitted instances list in tree view.

• Switch View - Tile View: Click the icon. All data backfill instances are displayed in a tile view.

• Customize List Header: Click the icon to select the columns to display in the list and adjust their order.

  Note

  • The first column and the Actions column cannot be hidden or reordered.

  • If you clear the browser cache or switch to a different browser, the header settings are reset to default.

• Refresh List: Click the icon to refresh the current list.

③ Submitted instances list

The submitted instances page lists the submitted instances generated from auto triggered tasks. The list includes Data Backfill Name-Data Timestamp, Data Backfill Status, Data Backfill Flow, Submitter/Time, Data Timestamp, Running Duration, Node Count, Tag, and supported Actions.

④ Batch operations area

You can use the batch operation functions provided in the batch operations area to process data backfill instances in batches, improving operational efficiency. For specific information, see Batch operations supported for submitted instances.

Judgment Order

Normalized Status

Judgment Logic

1

Not Running

If the instance is in the initialization state or all instances have a status of Not Running, the data backfill instance status is Not Running.

2

Running

If there are instances in the waiting or running state and no instances have failed, the data backfill instance status is Running.

3

Failed

If there are failed instances, the data backfill instance status is Failed.

4

Succeeded

If all instances have a status of Succeeded, the data backfill instance status is Succeeded.

Section

Description

①First layer

Data backfill task layer: This is the data backfill task, named as the instance name configured when executing the data backfill operation. Each data backfill operation submission generates a corresponding data backfill task layer.

②Second layer

Data timestamp layer: This displays all data timestamps that need data backfill under the data backfill instance, with each data timestamp corresponding to a folder.

③Third layer

Data backfill instance layer: This displays the instances generated after executing the data backfill operation on the task. You can also view the DAG graph that the instance depends on. The DAG for data backfill instances under each data timestamp is the same.

Operation

Description

Rerun

After selecting multiple data backfill instances, choose Rerun All Failed Instances, Rerun All Instances In Desired State, or Force Rerun All Instances at the bottom of the list. After rerunning, the data of the current instance node is directly updated. Downstream nodes that have not started running due to the current instance's failure will automatically start running after the current instance is successfully rerun.

• Rerun All Failed Instances: Reruns nodes with a running status of Failed.

• Rerun All Instances In Desired State: Reruns nodes with a running status of Failed or Succeeded.

• Force Rerun All Instances: Except for logical table instances where the logical table model has changed, all nodes or fields in any status can be force rerun.

Stop

Only instances with a status of Succeeded or Not Running do not support the stop operation. Instances with other statuses support the stop operation. After stopping, the running status of the instance changes to Failed. After the current instance node stops running, its downstream nodes will not run either.

The stop operation is often used in the following scenarios:

• You should stop an instance in the event of a Dataphin abnormality or an operational error.

• When an instance shows as running but data is not being produced. You can manually stop the run and check for exceptions through the run log.

Continue Scheduling on Success

Manually sets multiple instance nodes with a running status of Failed or Not Running to Succeeded and continues to participate in scheduling. Downstream instances that were only blocked from running due to the current instance's failure will start running.

Pause

Only instances with a status of Not Running, Waiting for Scheduling Time, or Throttled support pausing.

Resume

Only paused instances support resuming.

Modify Priority

Click the icon and select Modify Priority. In the Batch Modify Priority dialog box, modify the priority of the selected nodes. You can select Highest, High, Medium, Low, or Lowest.

Download All

Select multiple instances, click the icon at the bottom of the list, and select Download All. The system downloads the information of all instances across all pages in the current list. The current list is the result of your filtering and search.