WeichenXu123 edited a comment on pull request #32399: URL: https://github.com/apache/spark/pull/32399#issuecomment-842252413
> re: [#32399 (comment)](https://github.com/apache/spark/pull/32399#discussion_r629087709) > > How does it guarantee that it kill the Spark jobs? After one second, we might be on the step 3, and it ends up with not killing any job. no guarantee…:joy: But the assumption is: ``` def trial_thread_target(): if subTaskFailed: raise Error() else: # 1. run some code here # 2. launch a spark job... # 3. run some code here # 4. launch a second spark job... # .... ``` the 1/3/.. step took little time because most of training workloads will be run via spark job. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
