We have a client running Backup Exec 10.0 with a DDS4 tape drive. The jobs almost always worked fine, yet they have failed for the last 3 days in a row. Each night the job runs a little over 2 gb of the job (26 gb total), and then stops and fails. The log comes back saying: The job was canceled by user Recovery. Then if I go into the event viewer there is an error that the drive hardware is offline from the night before. I can see the tape drive, inventory the tape that is in it, and it shows up. I have stopped and restarted all the backup exec services to no avail, and tested inventorying the tape before the nightly job started. Any ideas? To my knowledge nothing new has been added to the server it is on, and the jobs worked fine for a long time before this started.