itimpi

I've a feeling that something like ZeroTier might work without the need for port forwarding.  Not used it myself so not sure.

No.
 
The drives in the array can use any mix of the file systems supported by Unraid - no requirement for them to all be the same.

You were getting continual errors on that drive and no valid SMART data could be read.    I would carefully check all cabling to the drive.  If it is not that then the drive itself could be faulty.

According to the diagnostics you were getting I/O errors on both parity drives and so they have been disabled.   As a result parity will need rebuilding on them.
 
The SMART reports for those drives look OK so maybe you disturbed the cabling when adding the new drive so carefully check both SATA and power cabling to the parity drives.   If you want to test the drives then you can run an extended SMART test on the drives but I think the cabling is a more likely suspect.  You can use the process described here in the online documentation accessible via the ‘Manual’ link at the bottom of the GUI or the DOCS link at the top of each forum page to rebuild parity on these two drives (they can be done at the same time).

As long as where you connect the SATA drive passes the serial number through correctly to Unraid, then it does not matter where they are connected.

This will be be because:
Dec 9 09:07:27 ACME-TOWER kernel: sd 3:0:0:0: [sdb] Enabling DIF Type 2 protection I do not have to hand the instructions to remove that protection but it should be easy to find them in the forum.

No data drive can be larger than the smallest parity drive.
 
you can use the Parity Swap procedure to simultaneously upgrade a parity drive to a larger one, and use the old parity drive to replace an array data drive.   With dual parity you can do this to two drives at the same time.  
 
After a data drive has been rebuilt then if it is a larger drive than the one it replaces the file system then gets expanded to fill the whole disk.

Have you made sure you have Disk Shares enabled under Settings->Global Share Settings?
 
if so check for each drive that you have the drive exported under its SMB settings as due to tightening security it now defaults to No.

@JorgeB as the expert on these things will probably confirm but I think the steps are:
stop array unassign disk start the array.   At this point Unraid should be emulating the drive and it’s contents should be visible stop array assign disk Start array to rebuild.

It is worth noting that Split Level takes precedence over Allocation method as described here in the online documentations accessible via the ‘Manual’ link at the bottom of the GUI.   This probably explains what you are seeing?    If not then tell us the name of a share that you think is not behaving as expected so we know what to look for in the diagnostics. 

If a drive cannot pass a SMART test without errors then it needs to be replaced.

You probably need to ensure that at the Windows level user ‘everyone’ has full control over the folders/files you want to manipulate remotely.  Note that this permission needs to be correct both directly a the File Manager level, and also in Windows at the Share level.

If the permissions are wrong you need to look at how the files were created that way.    It is not unusual for docker containers that create files to not be configured correctly to set the right permissions.
 
You can use Tools->New permissions to reset permissions to Unraid defaults.   This should not be run against the appdata share as that can cause problems as many containers have specific requirements around permissions.
 
you can also use the Dynamix File Manager (available for recent Unraid releases) to manipulate files from the GUI regardless of permissions.

You should NEVER use a format during a rebuild as the format operation updates parity to say the disk has an empty file system on it so you just end up rebuilding an empty file system.  If you still have the original disks intact you should be able to mount them outside the main array and copy back to the array the data they contain.

Sounds as if the flash drive dropped offline?
 
The flash drive must stay online all the time Unraid is running, and Unraid cannot handle it not being present.  If it drops offline then the only possible recovery action is to reboot the server.

Plugins can do just about anything so in that sense they are a security risk.   Users should definitely not install any random plugin.   If a malicious user gains access to your server then they could install a rogue plugin.  However a plugin will not survive a server reboot unless there is a corresponding .plg file in the ‘config/plugins’ folder on the flash drive so at least it will be visible.

it is expected  that plugins make their source code publicly available (not sure if there are any that do not conform to this rule).   I know that they are not added to CA unless their is a chance to review their code.

Yes if you use the New Config tool and change the disk assignments to the slots you want.  The data on the drives will be left intact.   However changing the drive slots invalidates parity2 so that needs rebuilding after making the change.

If you click on any of the shares on the Shares tab then there should be an Export option.   It needs to be set to Yes to make it visible on the network (the default is No).   This applies to both User Shares and Disk Shares.

There is a problem with formatting disks bigger than 2TB on the 6.11.2 release.    This is fixed in the 6.11.3 release.

Mover ignores shares which are set to Use Cache=No.   if you want mover to transfer the files to the array then you need Use Cache=Yes setting.

I believe Unraid displays what the BIOS reports

UFS Explorer is not free, but you can run it in a trial mode that will give you a list of the files it found before you make any decision about whether it is worth purchasing it.    I bought myself a copy of the Standard edition some time ago and it has proved invaluable the few times I have had to use it.

The CA Appdata backup plugin.

You should contact support to get this resolved - the forum is not the right way.

Looks like it IS activated.   Thread pairings relate to a core that has an associated hyperthread so it appears as 2 logical cores.   If you are assigning cores to a specific task you normally want both the master core and its associated hyperthread for optimum performance.