Answer-first summary for fast verification
Answer: Utilize the repair feature to re-execute only the failed tasks and their dependencies.
Repairing a failed multi-task job allows for the re-execution of only the unsuccessful tasks and any tasks that depend on them. This approach significantly reduces the time and resources needed for recovery by not rerunning tasks that have already completed successfully. For more details, refer to the [Databricks documentation on repairing job failures](https://docs.databricks.com/workflows/jobs/repair-job-failures.html).
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
A data engineer is managing a Job that consists of multiple tasks, taking over 2 hours to complete. During the last execution, the final task failed unexpectedly. What action should the data engineer take to complete this run with minimal additional execution time?
A
Execute the entire Job Run again to run all tasks.
B
Utilize the repair feature to re-execute only the failed tasks and their dependencies.
C
Delete the failed Run and initiate a new Run for the Job.
D
Leave the failed Run as is and start a new Run for the Job.
E
Run the Job in Production mode for automatic retries on errors.
No comments yet.