This might be related to a new bug in the filewalker: High Trash usage - 27 TB - no more uploads - #29 by Alexey
I’m sure this is not recommended, but, on my raspbery pi, if you want the node to think you have zero trash, the following seems to work:
first stop the node
In your config file disable the file walker with setting storage2.piece-scan-on-startup: false
Go into the directory that contains your databases, make a backup of your piece_spaced_used.db for safety. Run the following command:
sudo sqlite3 piece_spaced_used.db "UPDATE piece_space_used SET total = 0 WHERE satellite_id = 'trashtotal';"
cross your fingers and start the node.
On windows you can edit the database with a graphical program like DBeaver.
I’m not sure this is really any easier than the recreation procedure you are trying to avoid but at least you can skip running the file walker.
Again, not recommended, only attempt if you enjoy tinkering and possibly breaking things…
Or, you can do the work around that Alexey describes in the link I posted. This should allow the file walker to run successfully which should update your trash values to what is actually on your disk, but that’s basically another database recreation procedure that you said you want to avoid.