We need a way to control the return status of Commands run via Spark. Currently there is no way to indicate to a calling script if a command has encountered a failure without intercepting and parsing the output.
Example: Consider the deployment scenario where additional steps are taken after successful migration. Running spark migrate will fail or succeed with no indication in the return code. If MigrateCommand (and then spark) could return 1 then the next step in deployment could adjust accordingly.
We need a way to control the return status of Commands run via Spark. Currently there is no way to indicate to a calling script if a command has encountered a failure without intercepting and parsing the output.
Example: Consider the deployment scenario where additional steps are taken after successful migration. Running
spark migratewill fail or succeed with no indication in the return code. If MigrateCommand (and thenspark) could return1then the next step in deployment could adjust accordingly.