Sqlite database corruption has been an issue for SOME people using /mnt/user/appdata/ for a long time, however nobody has been able to isolate why it affects some users and not others. If you read what I wrote I didn't say using /mnt/disk1/appdata would fix an array being spun up all the time, but it stands to reason if you have a load of docker containers metadata then whilst those containers are running, applications will be accessing it, now if that data is spread across various array drives and not confined to a single disk, it MAY, keep drives spun up unnecessarily. Regardless using your protected array for your appdata isn't a great idea due to the performance hit on writing to the protected array, which is why I would always recommend using at least a cache disk or, even better, an SSD for this. Personally I have spinning rust as my cache disk as it's used as a recording/timeshift store for my PVR and I use a SSD mounted with unassigned devices to store my docker.img and appdata and noticed a massive difference with Emby and Plex in terms of webui responsiveness, compared with when it was on the cache disk. The downside to running everything on cache, or an unassigned device is of course no parity protection, however as parity is not a backup strategy, everyone should have mechanisms in place to ensure their appdata is regularly backed up so in the event of issues they have a recent source to restore from. I personally use CA backup/restore on a weekly basis for this, copying my backups to my protected array and then I rclone it to a cloud provider. Sent from my Mi A1 using Tapatalk