No duplicate detection in the database.
It's possible to import the same track twice (different file paths, slightly different artist capitalisation) and the scheduler will treat them as different tracks. A duplicate detector based on title + artist + duration matching would prevent most of this.