Kilrah

You're on 6.12, click the green padlock on the top right row of icons to unlock reordering.

Go into the plugin settings and you can enable loading the modules there.
 
Unraid doesn't care, but depending on what you installed the USB Manager plugin for you may need them loaded for the desired plugin functionality to work.

Yeah, just not possible on Intel at this point, seems AMD is fine.
 

Nope, either unbalance or manual move.

Yeah, the CPU has no actual issues with nested virt but something's broken in the kernel specifically for nested Hyper-V on intel... hopefully gets fixed at some point

You can install the nerd tools plugin, install ncdu there, then run 
ncdu --exclude /mnt --exclude /var/lib/docker --exclude-kernfs /  
Should show you what's big in the rootfs.

You can't do RAID with XFS, only btrfs, so if you add more than one slot it'll switch the pool to expecting btrfs, now since the drive is xfs it can't mount it.
 
You should be able to remove the 2nd drive, switch the pool back to one slot, then click on the pool name and select xfs again as filesystem. Should mount again.

As you've seen a single file cannot span multiple drives.
 
The common way of setting things up is to have a small vdisk, on an SSD for the OS and app and have it store the actual data/content to a network share, which is an unraid share and that will be able to scatter the files across drives.

No point, they're just doing what they're designed for. 
Every drive will fail one day or another. That's why you have redundancy, so it can be replaced when it does. No big deal.

If you have enabled reconstruct writes in disk settings than any write to any disk will spin them all up. 

https://wiki.unraid.net/Manual/Storage_Management#Removing_data_disk.28s.29

See the help text. In your current situation it'll only move to the next once disk3 has less than 1TB free.
 

Try chown 1001:1001 /mnt/user/appdata/TGTG_whatevergotblurred
That's the user/group this image expects to have
 
https://github.com/Der-Henning/tgtg/blob/main/Dockerfile

Back up your original usb, the keyfile is a pretty obvious "Plus.key" in the config folder.
Only restore it to the original drive it's for since it's tied to it. 

Would create a VM, pass it the disk and a gparted live iso and do the resizing there.

It should be in the doc, or at least the docker image you're using should have an example of how to run it. If not then either it doesn't actually need local storage of its own, or is poorly maintained and it's going to be a bit of a pain indeed.
 
Edit: Looked at those 2 projects and they seem to be "give them input, they spit output", no storage of their own.

If you screenshot your Main page so you're sure of what drive belongs where and then delete the whole config folder on the flash drive you'll be starting clean.
You'll have to create a password, reenter all your network and other settings, reassign the drives the same way, starting the array will create shares for the existing folders but you'll then need to reconfigure each of them how you want them as they'll be created with defaults... You'll have lost your app templates so any Dockers will be orphan and need to be removed/recreated with all settings, unless you kept the Dockerman plugin folders...
 
Starting fresh while keeping all data is possible when you know what you're doing but not exactly straightforward.
 
When using the delete option in the share, which requires first deleting all contents separately it does remove the config file for me.
If you just delete the folders on the disks then it of course won't.
 
  
I agree that uninstalling a plugin should have a checkbox to remove any configuration files too. 

A rebuild is not a check. If parity was inconsistent when a rebuild is done then that can't be detected (at least not with one parity drive).

Just that, lack of protection.
That's why the standard way of using Unraid is having an SSD cache pool on which the VM images are and which you can set up as RAID1 for protection.

Is your mariadb container actually called "mariadb"?
 
Theres a "db-backup" app in CA that can do the job more cleanly.

You're probably on unraid 6.11.2 which had issues formatting drives. Update.
 
  
When a drive is replaced it gets rebuilt, that's the whole point of parity. The system doesn't even know whether there is something useful on the filesystem or not.

Clearing a 3TB drive like that should take less than a day, but if there are write errors it can be anything, completely unpredictable...

See 
 

Sure as long as you only go between /mnt/diskX and /mnt/diskY, not mixing /mnt/user with /mnt/disk