JorgeB

Most likely, the backplane is transparent to the OS.

Please post the diagnostics: Tools -> Diagnostics

If you can try a couple of different brand/model SSDs.

If you knew the problem was only on the actual disk and parity was still correct you could rebuild the disk to correct the error.

See here: https://wiki.unraid.net/Upgrading_to_UnRAID_v6

If it's always the same port (and just that one) I would just avoid using it.

https://unraid.net/blog/top-support-and-forum-questions-answered

Thanks for reporting back, AV software can have a noticeable performance impact in various areas.

This is the correct forum, if typing "reboot" didn't work you'd need to manually stop whatever is using the disks and then unmount them or force a reboot.

Parity dropped offline: Mar 27 17:28:14 Tower kernel: ata11: softreset failed (1st FIS failed) Mar 27 17:28:14 Tower kernel: ata11: reset failed, giving up Mar 27 17:28:14 Tower kernel: ata11.00: disabled Mar 27 17:28:14 Tower kernel: ata11: EH complete Mar 27 17:28:14 Tower kernel: sd 11:0:0:0: [sdi] tag#8 UNKNOWN(0x2003) Result: hostbyte=0x04 driverbyte=0x00 Mar 27 17:28:14 Tower kernel: sd 11:0:0:0: [sdi] tag#8 CDB: opcode=0x8a 8a 00 00 00 00 01 50 db ed 40 00 00 04 00 00 00 Mar 27 17:28:14 Tower kernel: print_req_error: I/O error, dev sdi, sector 5651557696 Mar 27 17:28:14 Tower kernel: md: disk0 write error, sector=5651557632 Since it dropped there's no SMART, so check/replace cables and post new diags. Also performance wise (and for other reasons) would be a good idea to convert all your reiserfs disks, reiserfs it's not recommended for years now.

Unraid UEFI boot is disable by default, you need to rename the EFI- folder to EFI, also make sure secure boot is disable in the NUC BIOS.

Move back the appdata folder then recreate the docker image.

They are saved to the logs folder on your flash drive.

With the array stopped check cache is still sdj and wipe the drive with: blkdiscard /dev/sdj Then start array and format.

There's no problem running 4kn drives on a 9211 in IT mode.

On the last diags the cache pool balance operation failed, possibly because of fstrim running and crashing, you can reboot and try again or backup cache pool, re-format and restore.

Cache filesystem is corrupt, best to backup, re-format and restore data.

Yep, looks good so far, speeds are around 260MB/s which about the expected for those disks on the outer cylinders, it will slow down a little as it moves to the inner ones.

Best to wait for the other tests.

Very strange, it would have been one of the last things I suspected, was the BIOS up to date?

Longer than normal time means disk might be marginal, but if there's no error it's OK for now, all other disks I mentioned above also finished without error?

Yes, IIRC, not 100% certain.

Yep

You'd need previously created checksums (or be using btrfs), without that you'd need to compare all files to known good ones or try them one by one.

The metadata part of the filesystem should be OK, so data will most likely look fine and complete, but any data that was rebuilt past disk12's failure will be corrupt/unavailable when you try to access it.