- High: It blocks me to complete my task.
I need to periodically propagate log files from worker nodes to S3, so they will be available permanently. I have to do this incrementally while the program is running, and not wait until the end, in case something crashes. I think my driver program should be able to to rclone from the session logs directory to my intended destination. But how can my driver program know the path of the logs directory?
I see in the documentation that it’s at /tmp/ray/session_*/logs, but what is the api, if any, for getting the session id?
