My webserver (Debian BullsEye) has an internal 2TB harddrive which sole purpose is to hold a backup of the server's main HD. Every day a cronjob is started that creates a backup using rsync. But every few days, the backup drive gets corrupted and backups fail. Doing an fsck gives me tons of errors like
Entry 'de_language_import.lng' in /backup.2/backup.1/usr/local/ispconfig/interface/web/admin/lib/lang (24910326) has deleted/unused inode 18221087. Clear?
Free blocks count wrong for group #2226 (2, counted=18).
Free inodes count wrong for group #2224 (541, counted=557).
And lots more.
I then go to reformat the drive, after which backups succeed for a few days, after which the drive gets corrupted again.
I've checked the drives SMART data and it shows the HD in perfect health. I also did a
badblocks -v /dev/sdb1 > badsectors.txt
which results in an empty badsectors.txt file (which I assume means no bad sectors were found).
And that concludes my knowledge of trying to figure out what could be wrong with the HD. So my question is: what else can I do to figure out what is going wrong?