Understanding a task's status and details - FSx for Lustre
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

Understanding a task's status and details

A data repository task can have one of the following statuses:

  • PENDING indicates that Amazon FSx has not started the task.

  • EXECUTING indicates that Amazon FSx is processing the task.

  • FAILED indicates that Amazon FSx didn't successfully process the task. For example, there might be files that the task failed to process. The task details provide more information about the failure. For more information about failed tasks, see Troubleshooting data repository task failures.

  • SUCCEEDED indicates that Amazon FSx completed the task successfully.

  • CANCELED indicates that the task was canceled and not completed.

  • CANCELING indicates that Amazon FSx is in the process of canceling the task.

After a task is created, you can view the following detailed information for a data repository task using the Amazon FSx console, CLI, or API:

  • The task type:

    • EXPORT_TO_REPOSITORY indicates an export task.

    • IMPORT_METADATA_FROM_REPOSITORY indicates an import task.

    • RELEASE_DATA_FROM_FILESYSTEM indicates a release task.

  • The file system that the task ran on.

  • The task creation time.

  • The task status.

  • The total number of files that the task processed.

  • The total number of files that the task successfully processed.

  • The total number of files that the task failed to process. This value is greater than zero when the task status is FAILED. Detailed information about files that failed is available in a task completion report. For more information, see Working with task completion reports.

  • The time that the task started.

  • The time that the task status was last updated. Task status is updated every 30 seconds.

For more information about accessing existing data repository tasks, see Accessing data repository tasks.