JorgeB

A reboot should fix it.

USB is not recommended for array or pool devices.

Mar 3 05:45:59 Bifrost emhttpd: unclean shutdown detected A few sync errors are normal, even expected, after this, just run a correcting check.

If the log is the only issue this might help:
 
btrfs rescue zero-log /dev/nvme0n1p1  
If it works make sure backups are up date since there a chance it can re-occur.

Replace cables for
 
cache device: (sds) ADATA_SU655_2L1929199ERW  
Or connect it to an onboard SATA port if available.
 
Emulated disk is mounting assuming contents look correct you can rebuild on top (and re-sync parity2 at the same time).

They should be all valid, assuming parity was in sync.

Logical block size: 520 bytes This is not supported, you'll need to re-format them with 512B sector size.

Set both appropriately for SSDs, defaults are good gor HDDs only, I believe the array status only fails if critical temp is hit.

Look for the tips and tweaks plugin, there are some NIC settings there that can make a difference.

Start here:
https://forums.unraid.net/topic/46802-faq-for-unraid-v6/?do=findComment&comment=819173
 

ZFS doesn't distribute files, it stripes based on blocks for the drives in the pool, if you want that best to wait for v6.12 and use a zfs pool instead of the array.

Everything on cache is using raid1, including all folders you've moved back.
 

Samsung is usually good, there have been some issues but from what I remember they are solved with a firmware update.

Delete/rename /boot/config/ssh and reboot, note that any saved keys will be lost.

No ATA errors on disk6 for now, check filesystem. 

v6.12 is expected soon, it will use kernel 6.1 or higher.
 
Yes.

Mar 9 22:42:41 Tower12 kernel: md: sync done. time=36575sec Mar 9 22:42:41 Tower12 kernel: md: recovery thread: exit status: 0 Rebuild finished.
 
btrfs errors are about the docker image, delete and re-create, and try switching to ipvlan (Settings -> Docker Settings -> Docker custom network type -> ipvlan (advanced view must be enabled, top right)), this might have been the source of the main issues.

I would recommend at least checking the cables, both power and SATA, or ideally replacing them to rule them out.

You are correct, behavior changed in the current release, going to find out when and report it since like you've experience this is a problem.
 
Not really my area but I'll also inquire about that, it would make things easier if that's possible to do.

See if this helps:
 
 

It's not logged as a disk problem and SMART looks OK, check/replace cables to rule that out and re-sync parity.

Controller problem:
 
Feb 22 18:52:07 Mordor kernel: megaraid_sas 0000:02:00.0: 236535 (730425127s/0x0020/DEAD) - Fatal firmware error: Line 255 in ../../hw/tempsensor.c Feb 22 18:52:07 Mordor kernel: Feb 22 18:52:07 Mordor kernel: megaraid_sas 0000:02:00.0: megasas_disable_intr_fusion is called outbound_intr_mask:0x40000009 Feb 22 18:52:07 Mordor kernel: megaraid_sas 0000:02:00.0: FW in FAULT state Fault code:0x10000 subcode:0x0 func:megasas_wait_for_outstanding_fusion Feb 22 18:52:07 Mordor kernel: megaraid_sas 0000:02:00.0: resetting fusion adapter scsi1. Feb 22 18:52:07 Mordor kernel: megaraid_sas 0000:02:00.0: Outstanding fastpath IOs: 160 Feb 22 18:52:07 Mordor kernel: megaraid_sas 0000:02:00.0: Reset not supported, killing adapter scsi1.  
Also note that RAID controllers are not recommended.

Unraid driver is crashing, reboot and post new diags after trying to format again.

Check filesystem on disk2.

Now it's showing a different issue:
 
Feb 28 18:56:07 Tower kernel: BTRFS error (device loop2): parent transid verify failed on 30425088 wanted 58737 found 58225  
Docker image is corrupt, you need to re-create.