Hi @Michael_Long,
Thanks for your patience, the diagnostics came through. Two things are going on, and neither is a permissions or format issue.
Every failed run ends the same way:
GetFSDir: volume was not available Roon PNY USB 3.0 FD : !volume.IsAvailable
failed sync prepare1 ... Result[Status=NotAvailable]
Roon Server never gets as far as writing, it can’t resolve the volume.
1. The drive is mounted by your desktop session, not by the system. It sits at /run/media/michael/Roon, a udisks2 mount that only exists once someone logs into GNOME. After the 08/21 restart there was no mount event for ten days, and all seven scheduled runs in that window failed. Not one scheduled backup failed while the drive was actually mounted, manual backups worked because you were at the machine with it freshly mounted.
2. The ext4 reformat orphaned the saved destination. On 09/01 the drive came back as /run/media/michael/ROON, but the schedule is still bound to the old Roon volume. Roon treats those as two different volumes, so the picker shows the drive and Backup Now succeeds while the schedule resolves something that no longer exists. Editing the entry only changes the time and keeps the old location.
Next steps to try:
Delete the scheduled backup entry (don’t edit it) and re-add it, browsing to the drive as it appears now.
That will break again on the next reboot, so we’d also give the drive a permanent mount:
sudo blkid /dev/sda1 # get the UUID
# /etc/fstab
UUID=xxxx-xxxx /mnt/roonbackup ext4 defaults,nofail,x-systemd.device-timeout=10 0 0
sudo mkdir -p /mnt/roonbackup && sudo mount -a
Then add the backup location by browsing through the root entry (/) to /mnt/roonbackup/RoonBackups instead of picking the removable drive. Your 09/03 10:00 scheduled backup to /home/michael/Downloads/RoonBackups ran cleanly, path-based destinations survive reboots and relabels, volume-based ones on a session mount don’t.
Let us know how the next scheduled window goes. 