Hey all, I’ve got ZFS pool created and just create a VM drive in that pool like normal, then Jellyfin just has that drive mounted. I think I’m losing the best parts of ZFS through this manner.
How should I set this up properly? Create a media pool or something and have VMs accessing the pool directly?
You’re doing it fine. You can overprovision your zpool that way for other VMs with datasets on the pool, and it only uses what it needs for each set. And you can exclude the media pool from backups and repllications if you like. There’s no performance hit the way you’r doing it.
I sure wouldn’t bother with any sort of NAS intermediary, you can share out the media pool via the Jellyfin VM if you really need to, but I can’t see why you would if you’re using Arr stack components already.
OK, sweet thanks. I just thought having the media files directly in the ZFS pool (zfs pool > files) instead of ZFS Pool > Qemu img > Files would be smarter
Then you’d have to share them back from the host to the guest using NFS or samba, and there’s no reason to have them on the host. No, you’re fine. That’s what I would consider the right way to do it for what you describe as your purpose. Putting more layers in there isn’t serving a useful purpose, which is why I say to not bother with a NAS to store them, if your primary use of them is via the Arr stack VM.