As you can see, my node is saying it’s overused. However that’s is not true. This is a 4TB drive and there is still a TB free. I recently set in motion the hashstore upgrade and its going well tho still in progress. My two other nodes have gone thru the process and no problems. This one is weird. It appears to be working fine but due to the overuse there is no ingress.
The unofficial fix:
Stop your node. Delete/rename “used_space_per_prefix.db” and startup your node.
On startup it will recreate database, and trigger file Walker to do a scan. This has proven to resolve this issue.
3 Likes
Not sure why it got out of sync but that fix work and its now reading the space correctly.
That is because it counts the migrated pieced when it is written to hashstore, but does not remove the space used for the deleted piecestore files.
1 Like