JorgeB

https://wiki.unraid.net/The_parity_swap_procedure

There was a QEMU related crash:
 
Jul 4 17:06:42 Tower kernel: qemu-event[9717]: segfault at 0 ip 0000148e82990396 sp 0000148e7e03d930 error 4 in libvirt.so.0.6005.0[148e82846000+1f6000]  
Rebooting should fix it, if it doesn't post new diags.

Start by running memtest, also a good idea to replace that old JMB controller, it's causing ATA errors on both connected devices.

Apple and Samba..., several Samba errors in the log, see if disabling the SMB macOS specific options helps or not.

Try with just one DIMM at a time and see if those hardware errors go away.

Try a different NIC if available, ideally an Intel one.

Problem with the controller:
 
Jul 2 04:16:27 CMNSVR01 kernel: smartpqi 0000:0b:00.0: controller is offline: status code 0x1111 Jul 2 04:16:27 CMNSVR01 kernel: smartpqi 0000:0b:00.0: controller offline  
Power cycle the server and if the controller comes online, and it should, post new diags after array start.

Not a network guy but pretty sure you should only have one gateway, remove the gateway from eth1 and try again.

Some strange crashes, including the Unraid driver, rebooting should fix the problem for now.

You'll need to replace the flash drive, to get the array up see here:
 
https://wiki.unraid.net/Manual/Changing_The_Flash_Device#What_to_do_if_you_have_no_backup_and_do_not_know_your_disk_assignments

As expected no fs detected, but since current cache is xfs lets assume old one was the same, so try this:
 
xfs_repair -v /dev/sdb1  
This will take some time while it looks for a backup superblock.

Jul 3 11:49:01 Darkness kernel: kernel BUG at fs/btrfs/extent-tree.c:1107! Btrfs crashed during the balance, this suggests there are fs issues, reboot, force it you have to, and try again, though most likely the same will happen, if yes best bet is to mount the pool read only, backup and re-format.

If you think the device was wiped the correct command would be:
 
btrfs-select-super -s 1 /dev/nvme1n1p1  

Command line: BOOT_IMAGE=/bzimage vfio-pci.ids=8086:10c9 initrd=/bzroot  
They are stubbed to vfio-pci, remove that from syslinux.cfg and reboot.

Docker image is corrupt, delete and re-create.

Reboot, it's no longer part of the array.

You cannot create new threads, you're supposed to use the existing docker support thread,  in this case:
 
 

Unraid doesn't support SAS multipath, connect just one cable from the HBA to each enclosure, i.e., only use one module in the enclosure.

That's because of this pesky "stale configuration" error, array is starting, just not showing in the GUI, make sure there are no other browser windows open in the GUI, including phones or tab, also clear browser cache.

@songI can replicate your issue with a GTX670, works fine with v6.9 and get code 43 after updating to v6.10, I found that this works for me, please try it, edit the XML and in the features section, below </hyperv> add this:
 
<kvm> <hidden state='on'/> </kvm>  
Note that if you have "fast startup" enabled in Windows, and it's enabled by default, first boot after the change will still have the problem, you need to re-start the VM (not shutdown and turn on again).
 
P.S. besides the code 43 issue I also had to dump the VBIOS to use the GTX670 with v6.10, but that's a different issue since there's no display output without it.

Based on the dates those diags look old, if Unraid is not finishing boot start by trying to boot in safe mode, if still the same try with a new trial flash drive, just to see if it boots.

Cache device dropped offline:
 
Jun 23 07:43:03 ForwardUntoDawn kernel: ata2.00: failed to IDENTIFY (I/O error, err_mask=0x100) Jun 23 07:43:03 ForwardUntoDawn kernel: ata2.00: revalidation failed (errno=-5) Jun 23 07:43:09 ForwardUntoDawn kernel: ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 310) Jun 23 07:43:09 ForwardUntoDawn kernel: ata2.00: failed to IDENTIFY (I/O error, err_mask=0x100) Jun 23 07:43:09 ForwardUntoDawn kernel: ata2.00: revalidation failed (errno=-5) Jun 23 07:43:09 ForwardUntoDawn kernel: ata2.00: disabled  
Then it came back with another identifier, suggest you shut down, check/replace cables, power back up and the device should be correctly assigned, if yes just start the array, if not start it without that device assigned, then stop, re-assign it and start again.
 

It's a problem with that device and newer kernels:
 
https://us.community.samsung.com/t5/Monitors-and-Memory/SSD-980-heat-spikes-to-84-C-183-F/td-p/2002779
 
 

You just need to unassign them all, them when the disks are re-formatted xfs (or btrsf encrypted) just re-assign all the pool devices and start the array, it will import the existing pool.

Any shares that have data on a single device pool will appear as unprotected, since there's no redundancy, that's normal.