You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When working with SQLite one of the worst errors a user can encounter is a "malformed disk image." There are many causes, including known bugs in SQLite. Often, the data is actually fine and usable.
This ticket is to track work for finding ways to gracefully work around this situation.
export the db to a file, delete the DB, then recreate it from the export
The ability to pause the synchronizer so that a corrective action can be taken without the DBs being accessed
perhaps we just prevent access to the datadb yet allow caching to still occur
copying the DB file might work (but probably not but it's worth a try)
add some sort of fast re-scan, where we know which blocks we cared about, before, so we can take some shortcuts
this might be more difficult to do once we're deleting the cached blocks but we could maintain a secondary database of just metadata for primitives
consider adding an "export DB" option that allows the user to grab their DB file out of the app. This would work similar to logs and potentially allow the DB to submitted for troubleshooting or recovery.
As the data access API evolves, we should explore more options in this area.
The text was updated successfully, but these errors were encountered:
When working with SQLite one of the worst errors a user can encounter is a "malformed disk image." There are many causes, including known bugs in SQLite. Often, the data is actually fine and usable.
This ticket is to track work for finding ways to gracefully work around this situation.
As the data access API evolves, we should explore more options in this area.
The text was updated successfully, but these errors were encountered: