V58 db migrate fail - piece_expiration.db - 1.108.3

If the garbage collector has been moved the piece to the trash, it will not be found by the collector.
It’s pretty easy to check, search for the piece ID in the trash (remove 2 first characters):

find /mnt/storj/storagenode/storage/trash -type f -iname JZRTIJAPEHJ67GI6RFHTWJTRT5FWTC7NTVDZGJXANP7ZN542AA.sj1

You may also limit to the subfolder of each satellite (there you can see the names of the folders: Satellite info (Address, ID, Blobs folder, Hex)).

But likely @nerdatwork is correct and the collector didn’t remove the info from the database if it was interrupted during the process. With the batching it should delete it now.

1 Like