Chris wrote:
Following through on that, and sorry it took nine days. Tom shipped the fix the same day. The reason your clean reinstall didn't help: Immich's Postgres database lives in its own volume and survives an uninstall, so the scrambled settings come straight back — reinstalling can never clear it. sudo sparkbox doctor now says that plainly instead of sending you round again. He also built a rescue command for exactly this. Start with the read-only step: sudo sparkbox upgrade sudo sparkbox restore-immich --list That only lists which backups carry a usable database dump — it changes nothing. Tell me what it prints and we'll go from there. Your photo files stay untouched throughout.
TomQ wrote:
Update I ran sudo sparkbox doctor and it kicked out this [Container Health] [ERROR] sb-immich-server is stuck in a crash-restart loop (35 restarts so far) │ at async /usr/src/app/server/dist/repositories/database.repository.j s:389:13 │ microservices worker exited with code 1 │ Killing api process Try: sudo sparkbox up (and 'sparkbox logs <name' to see why a container st
Chris wrote:
Good news first: you didn't lose your photos. Immich keeps your actual photo files on disk, separate from the little settings database that got scrambled here — so nothing you did deleted your pictures. That crash-restart loop is Immich's database not coming back up cleanly, and a plain reinstall won't clear that on its own. One safe thing to try first, since Tom's shipped several Immich stability fixes lately: make sure your box is fully up to date. Run: sudo sparkbox update Let it finish, then give Immich a few minutes — if it recovers, its tile turns green on the Apps page. If it's still stuck in that restart loop after updating, don't delete or reinstall anything else; that database repair isn't something to guess at. I've flagged this to Tom so he can get you the exact safe steps.