I am planning to import a large schema from one 11.2.0.3 database into another of the same version on a different host, both running RHEL. I am expecting this will take a few days to complete and will be over a db link. I plan to start this as a background process (nohup &) so it will continue if I get disconnected. I know there is a timeout window if something like tablespace limits occurs, but is there any way to restart the job if it happens to fail and stop? The documentation I read said that the START_JOB command could be used to restart a job to which you are attached, but that won't be the case here since I will run it nohup.