MrDatum

Ah, that clears up the mystery. I ended up deleting the pool and recreating it and now all is well. Other than mover leaving duplicates on the array after setting up the cache pool and running mover with array-->cache, but that is a separate issue and one I can just clean up manually.

I'm running Unraid 7.0.1. One of the SSD drives in my cache pool is failing and I want to move to two nvme's instead of the two SSD's. I disabled the Docker and VM services, moved all of my appdata/domain/system shares the array using mover and then stopped the array. I tried: 1. In the 'main' tab, for the two cache drives after stopping the array, I changed the drives to the two nvme drives, then started the array. I get the Wrong Pool State error message 2. With the two old SSD's showing in the cache pool, I stopped the array and turned the server off, then unplugged the two SSD drives, rebooted and for both drives in the cache pool it showed, 'no device'. I changed those to the two nvme drives and started the array and get the Wrong Pool State message 3. I set both cache drives to 'no device', then started the array and still get the Wrong Pool State error Am I doing something wrong? Attached is my diagnostics. tower-diagnostics-20250520-1134.zip

Awesome, thanks!

Would using BTRFS in single profile allow me to expand the cache easily by adding another drive?

I'm using BTRFS in Single profile.

Thanks for the help, it is working and I have that warm fuzzy feeling from being fully backed-up In the future if I need more space in the pool can I just expand it by shutting down the array, adding another dive and then assigning that drive to the existing pool?

I'm about to setup my new pool but wanted to confirm that a single large multi-TB file can span across more than one of the drives in the pool so that I don't run into the issue of my backups failing because there isn't enough space. For example, if I have a 3TB and another 3TB drive, and my backup file is 4TB, setting up a pool and selecting the Single profile will allow the 4TB file to span across both drives?

Thanks, I'll give this a shot.

I want to backup a number of my home computers to my Unraid server. Some of the backup files are multi-terrabytes. I have been using a share with a couple of drives on the array, but I'm running into an issue where the backup fails as their isn't enough room, although I have TB's of space available. However, that space is spread over a couple of drives and I don't have the size needed on any one of those drives for the large multi-TB backup file. 1. Is it possible to create a JBOD pool where all of the drives are combined into one large drive such that a single large multi-TB file can span across more than one of the drives? 2. If so, can the drives be of different sizes? Would this be the best way to accomplish what I am trying to do with my backups?

That makes sense, I'll give that a shot.

I think I have the docker issue sorted. I ended up reverting to V2.0 versions of both and that seemed to fix the issue as they both now load. Still not sure what is going on with the appdata folder being on both the cache and the array. I did stop the docker service before running mover but that did not seem to fix the issue.

Any thoughts on this?

I had an issue with my Unraid server a few weeks ago where my boot drive failed and my cache pool drives were showing up in Unassigned devices. I made a new boot drive and was able to recreate the cache pool. However, two of my dockers (Radarr and Sonarr) would not load while my Plex and NZBget dockers loaded fine. The logs for both says the database is corrupt: "EPIC FAIL! NzbDrone.Core.Datastore.CorruptDatabaseException: Database file: /config/sonarr.db is corrupt, restore from backup if available." Something else I’ve noticed is that my appdata folder is, for some reason, present on both my cache pool as well as on one of my drives in the array. The appdata share setting had Use Cache Pool set to “Only” so I don’t know how it wound up being also on one of my array drives. I stopped Docker, set the share settings for appdata to “prefer” for the cache pool and ran the mover. However, I am still seeing appdata on one of my array drives. What I’ve tried to get Sonarr and Radarr running: 1. In Settings/Docker I ran a scrub with “correct file system errors” 2. I forced an update of both Dockers so the Docker would be recreated 3. I removed the Dockers and added them back with “Add Container” and selected the appropriate template 4. I stopped the Docker service and deleted the vdisk file, restarted Docker and reinstalled all of my Dockers using the templates. Again, Sonarr and Radarr will not load (same corrupt database error in the log) while NZBget and Plex do load fine. So I’m not sure what the next step is. If I can’t get Sonarr/Radarr to run by deleting the Docker image and reinstalling, what else can I try? tower-diagnostics-20230618-1509.zip

Would I just copy the old appdata folder over top of the existing and then delete the docker and then recreate the docker using the template?

No, I haven't done that yet. I do have a few points in time of the CA backups, is there now way to use those to rebuild the Docker without losing data/settings?