JorgeB

Try a different PCIe slot if possible, you can also try the bellow, though not sure it helps with this error:
 
https://forums.unraid.net/topic/132930-drives-and-usb-devices-visible-in-bios-not-available-once-booted-asus-wrx80-sage-5965wx/?do=findComment&comment=1208035
 

https://docs.unraid.net/unraid-os/release-notes/6.12.4/#fix-for-macvlan-call-traces

The main issue I see was an OOM error:
 
Out of memory: Killed process 32593 (jellyfin)  
That container was using a lot of RAM, so check the config or limit its RAM usage.

I would recommend using the form again, try a different brwoser, until you get an automatic reply, AFAIK there's no direct email you can use.

Diags are after rebooting so we can't see what happened, but the disk looks healthy, since the emulated disk is amounting, and assuming contents look correct, you can rebuild on top, recommend replacing the cables first to rule that out, if it happens again to the same disk.

gdisk output confirms the partitions got clobbered, likely by the RAID controller, you can try to fix one with gdisk to see if it works after, to play it safer I would recommend cloning the disk first with dd and do it on the clone.

If the log tree is the only issue this may help:
 
btrfs rescue zero-log /dev/nvme0n1p1  
Then restart the array.

ZFS raidz1 is far from experimental, it has been considerable stable for a long time.
 
That suggests the pool is crashing the server on mount, before starting the server type:
 
zpool import -o readonly=on cache  
If successful then start the array, the GUI will show the pool unmountable but the data should be under /mnt/cache, then backup and re-create the pool
 

This usually means bad RAM or other kernel memory corruption, start by running memtest. 

Try this first:
 
https://forums.unraid.net/topic/118286-nvme-drives-throwing-errors-filling-logs-instantly-how-to-resolve/?do=findComment&comment=1165009
 

That will only happen if you set the share to use a pool as the primary storage, currently is set to array.
 
Note that you can also get better performance writing directly to the array with turbo write.
 
 

See if Windows is set to sleep/hibernate after a while.

Mar 13 16:16:45 Tower kernel: ata1: link is slow to respond, please be patient (ready=0) Mar 13 16:16:45 Tower kernel: ata2: link is slow to respond, please be patient (ready=0) Mar 13 16:16:49 Tower kernel: ata1: COMRESET failed (errno=-16) Mar 13 16:16:49 Tower kernel: ata1: hard resetting link Mar 13 16:16:49 Tower kernel: ata2: COMRESET failed (errno=-16) Mar 13 16:16:49 Tower kernel: ata2: hard resetting link Mar 13 16:16:55 Tower kernel: ata2: link is slow to respond, please be patient (ready=0) Mar 13 16:16:55 Tower kernel: ata1: link is slow to respond, please be patient (ready=0) Mar 13 16:16:59 Tower kernel: ata1: COMRESET failed (errno=-16) Mar 13 16:16:59 Tower kernel: ata1: hard resetting link Mar 13 16:16:59 Tower kernel: ata2: COMRESET failed (errno=-16) Mar 13 16:16:59 Tower kernel: ata2: hard resetting link Mar 13 16:17:03 Tower kernel: ata2: SATA link up 1.5 Gbps (SStatus 113 SControl 310) Mar 13 16:17:03 Tower kernel: ata2.00: configured for UDMA/100  
Constant ATA errors from parity and disk1, check/replace cables (power and SATA) for both and post new diags after array start.

Looks to me like Frigate is the problem, check the configuration or try limiting its RAM usage.

This is why it's failing, but not sure what's causing it:
failed to start containerd: timeout waiting for containerd to start  
You can try recreating the docker image:
https://docs.unraid.net/unraid-os/manual/docker-management/#re-create-the-docker-image-file
Also see below if you have any custom docker networks:
https://docs.unraid.net/unraid-os/manual/docker-management/#docker-custom-networks

Should be perfectly safe, if you follow the steps correctly, I did it with my TrueNAS CORE pool, just for testing, since I want to keep TrueNAS on this server, booted with an Unraid flash drive, pool before the changes:
   pool: tank      id: 11986576849467638030   state: ONLINE status: The pool was last accessed by another system.  action: The pool can be imported using its name or numeric identifier and         the '-f' flag.    see: https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-EY  config:         tank        ONLINE           raidz3-0  ONLINE             sdk2    ONLINE             sdg2    ONLINE             sdc2    ONLINE             sdd2    ONLINE             sdf2    ONLINE             sdi2    ONLINE             sde2    ONLINE             sdh2    ONLINE             sdm2    ONLINE             sdj2    ONLINE             sdl2    ONLINE  
After running fdisk on each device to delete parttion1 and make partition2 > partition1:
   pool: tank      id: 11986576849467638030   state: ONLINE status: The pool was last accessed by another system.  action: The pool can be imported using its name or numeric identifier and         the '-f' flag.    see: https://openzfs.github.io/openzfs-docs/msg/ZFS-8000-EY  config:         tank        ONLINE           raidz3-0  ONLINE             sdk1    ONLINE             sdg1    ONLINE             sdc1    ONLINE             sdd1    ONLINE             sdf1    ONLINE             sdi1    ONLINE             sde1    ONLINE             sdh1    ONLINE             sdm1    ONLINE             sdj1    ONLINE             sdl1    ONLINE
After doing this, the pool imported normally with Unraid 6.12.8:

 
Rebooted the server and booted TrueNAS, pool imported as if nothing changed:

 
 
 
 

Yes, it does support zfs striped mirrors.

Should be OK but recommend reading the release notes for any major release, 6.8.0, 6.9.0, etc

Yes, it can be more complete, once a week should be fine.

If the log tree is the main issue this may help, type:
 
btrfs rescue zero-log /dev/sdd1  
Then restart the array

Try booting in safe mode and/or using a different browser.

Yep, it will be supported.

Stop the array, unassign the pool device, start array, stop array, set the pool to one slot, assign the pool device, start array, that should do it.

Not sure what you mean here, as long as the new devices are same size or large there won't be a problem.

type
zpool import -F cache if successful then type
zpool export cache After that it should be mountable with UD or Unraid as a pool, and run a scrub.