Sometimes a job might always fail on a particular cloud, but work on others. Should there be a way to allow failed jobs to be only retried on different clouds? This should be optional, as there as some situations in which you might not want this to happen.
Sometimes a job might always fail on a particular cloud, but work on others. Should there be a way to allow failed jobs to be only retried on different clouds? This should be optional, as there as some situations in which you might not want this to happen.