You can configure task-level runtime timeout and rerun policies for offline computing tasks based on your business scenarios. This helps prevent resource waste caused by long-running computing tasks while improving the reliability of task execution. This topic describes how to configure runtime settings for offline computing tasks.
Procedure
In the top navigation bar of the Dataphin homepage, choose Development > Data Development.
On the Development page, select a Project from the top navigation bar (In Dev-Prod mode, you need to select an Environment).
In the left-side navigation pane, choose Data Processing > Script Task.
In the computing task list, click the target computing task to open its tab.
Click Property in the right sidebar to open the Property panel, and configure the following parameters in the Runtime Configuration section.
Parameter
Description
Runtime Timeout
If a single run exceeds the set threshold without completion, the task will be automatically terminated and marked as failed.
System Configuration: The default option uses the system configuration, which is based on the Recurring Task parameter set in the tenant-level runtime configuration in the O&M module. For more information, see Runtime Configuration.
Custom: You can enter a number between 0 and 168 with up to 2 decimal places.
Auto Rerun After Timeout
Determines whether to automatically rerun recurring instances or data backfill instances that are terminated and marked as failed after exceeding the configured runtime.
System Configuration: The default option uses the system configuration, which can be set to either not auto rerun or auto rerun based on the tenant configuration.
No Auto Rerun: When the current tenant has not enabled auto rerun after timeout in O&M > Global Configuration > Rerun Configuration, this displays as System Configuration: No Auto Rerun.
Auto Rerun: When the current tenant has enabled auto rerun after timeout in O&M > Global Configuration > Rerun Configuration, this displays as System Configuration: Auto Configuration.
Enable: If enabled, the tenant configuration in O&M > Global Configuration > Rerun Configuration will be ignored. When an instance times out, is terminated by the system, and is marked as failed, it will be automatically rerun.
Disable: If disabled, the tenant configuration in O&M > Global Configuration > Rerun Configuration will be ignored. When an instance times out, is terminated by the system, and is marked as failed, it will not be automatically rerun.
Auto Rerun After Failure
Determines whether to automatically rerun task instances or data backfill instances that have failed based on the configuration.
System Configuration: The default option uses the system configuration, which is based on the Auto Rerun After Failure parameter set in the tenant-level runtime configuration in the O&M module. For more information, see: Runtime Configuration.
Custom: The rerun count supports integers between 0 and 10, and the rerun interval supports integers between 1 and 60.
Click OK to complete the computing task runtime configuration.