Everything is automatic.
On piecestore every few days a new batch of data gets flagged as trash, and seven days later automatically erased from disk. On hashstore it gets flagged as trash the same way… still has the week delay.. then a job automatically runs that decides when compaction should happen to which logs (which is when their trash is erased from disk).
Running a Storj node should consume basically none of your time once you set it up. It upgrades itself, stores new data, deletes old data - and you don’t have to touch it. Ignore it unless you receive an email saying your node is offline 