Sorry guys, so what is the solution to get Syncthing back up again?
2026-01-31 18:02:10 INF syncthing v2.0.13 “Hafnium Hornet” (go1.24.12 linux-amd64) root@buildkitsandbox 2026-01-17 02:31:28 UTC [modernc-sqlite, noupgrade] (log.pkg=main)
2026-01-31 18:02:10 ERR Error opening database (error=“openbase (PRAGMA optimize = 0x10002): database disk image is malformed (11)” log.pkg=main)
At that point, you’ll have to nuke the database. The presence of noupgrade looks like some Docker deployment? Open the volume or bind mounted folder, delete the index-v2 folder and restart.
OT: Judging from the presence of the modernc driver this is not one of our releases?
Thank you, just deleted the index-v2 folder and started fine.
I’m on unraid 7.2.3 and using the Docker app based on the lscr.io/linuxserver/syncthing repo.
I strongly recommend switching over to our image: syncthing/syncthing - Docker Image
We’re listed on Dockerhub. No idea why it’s not part of the search results.
I suspect these images might be specifically done for Unraid.
I also had this issue on Unraid after running a manual appdata backup via the plugin.
I’m going to try changing all my docker appdata paths to /mnt/cache/appdata/ and see if it happens again. I’ll also change the default (settings → docker) for future containers.
I also noticed the default Docker shutdown timeout in Unraid is only 10 seconds. I’m going to change it to 60. Even so, syncthing should ideally recover from an unclean shutdown; not sure what’s up with that. Also, it shouldn’t retry forever, as that fails to notify that the container is not working.
Re. the image source, Unraid Community Applications does not automatically index everything on Dockerhub as they need additional configuration for Unraid. Somebody needs to make a Community Application template that points to the official docker.
