trurl

I don't think the parity check is important right now. You can do it again after things are working.

Have you done memtest?

See if you can get diagnostics immediately after rebooting, and setup syslog server.

Emulated disk2 shows plenty of data. You can browse its files if you want (you can even write to it).

Should be OK to rebuild to the new 12TB. Assign new 12TB as disk2 and start the array to begin rebuild.

syslog seems to indicate connection issues instead of bad drive, and drive passed extended self-test.

Check connections and try again.

You can't do a parity check with the array stopped, so that is why there is no option to do so.

Check connections, power and SATA, both ends, including splitters.

Make sure there is slack in cables so nothing is pulling on them at the connection.

Don't bundle data (SATA) cables.

Ideally, no more than 4 drives per PSU cable. MOLEX-SATA splitters are better than SATA-SATA splitters.

Try replacing cable.

Have you let Unraid format the pools?

Just replace and assign new disk to same slot. Contents will be rebuilt from parity.

https://docs.unraid.net/unraid-os/using-unraid-to/manage-storage/array/replacing-disks-in-array/

The behavior you describe isn't possible without that plugin, so that is where you should seek a solution.

You should post to the existing support thread for that plugin.

Delete or rename config/network.cfg on flash drive and try again.

If you did all that then they would be unmountable because they haven't been formatted. Have you tried formatting?

Default is btrfs raid1 mirror. Probably you don't need more than 500G anyway, so leaving it as raid1 mirror is probably good idea.

Go to Settings - Docker and Disable Docker.

Use the builtin File Manager to delete all the folders in the appdata share.

That should allow you to start from scratch.

Possibly disk5 can't be read well enough to allow disk3 to be emulated accurately. Emulation requires all other disks to be accurately read.

Since both of those diagnostics were taken without the array started in Normal (not Maintenance) mode, no filesystems are mounted so can't tell anything about the contents of disk5 or emulated disk3.

Syslog from those older diagnostics do seem to indicate disk5 mounted though, and emulated disk3 didn't.

SMART for disk3 looks OK, possibly a connection issue disabled it and bad disk5 kept it from being emulated accurately.

There is a way to get disk3 back into the array and disable disk5 instead so it can be rebuilt.
Tools - New Config - Retain All - Apply
Check BOTH Parity Valid box and Maintenance Mode box, and start the array. This will accept all disks just as they are into the array and won't modify parity or any data disk.
Stop the array, unassign disk5, then start the array in Normal mode. This will disable disk5 and mount all disks including emulated disk5.
Then post new Diagnostics.

Option 2

USB not recommended for assigned disks for many reasons. You seem to have encountered some of those reasons. Parity disconnected so it was disabled. It has since reconnected as a different device, but it has to be assigned and rebuilt, preferably without attempting to use USB for assigned disks.

Why weren't you using SATA port?

I don't see SMART report for an unassigned 12TB disk, so I guess you removed it. Likely nothing wrong with that disk itself, connection problems are more common when replacing disks.

To make Unraid allow you to put the original 6TB back into the array you will have to New Config

https://docs.unraid.net/unraid-os/using-unraid-to/manage-storage/array/array-health-and-maintenance/#reset-the-array-configuration

Users cannot create new threads in Docker Containers. You post to the existing support thread, as I see you already did.

It might be a bit confusing to name your pool Main, since that is the name of the Main page in your Unraid webUI, the page that you use to access your Main pool. But I will attempt to describe things as you have it named currently. If you decide to rename your pool adjust accordingly.

I assume you haven't enabled VM Manager yet since you don't have its default shares. So...

Go to Settings - Docker and disable Docker.

Go to appdata and system shares and set them both to Primary-Main; Secondary-Array; Mover Action-Array->Main.

Go to Main page - Array Operation, click Move, wait for it to complete.

Then

Attach Diagnostics ZIP to your NEXT post in this thread.

Should be OK to rebuild then.

You rebooted before getting diagnostics so can't see what caused it to become disabled. SMART attributes look OK, but extended test not completed.

Since you have a spare, safest approach would be to rebuild to it and keep original intact in case of problems. Be sure to check connections.

During rebuild, there should be lots of writes to the rebuilding disk, lots of reads from all other disks, and zero in the Errors column for all disks.

Post diagnostics again if there are problems.

Can you see the disks in BIOS?

After memory is tested and working perfectly (even 1 error is too many), you could New Config the original disk back into the array and let parity rebuild, then try the replace/rebuild again.

Did you click Apply on the New Config page?

Parity is updated at the same time as any write operation. Format is a write operation.

Even if parity had been removed before you did that, single parity can't recover 2 drives.

You should always have another copy of anything important and irreplaceable. Parity is not a substitute for backups. Plenty of ways to lose data besides a disk failure, including user error.