I got an alert about a disk read error on my server. From searching around I decided to stop and rerun the array (not reboot), and a message saying that the "array turned good" popped up.

I am not sure how to proceed from here. The server itself is barely 3 months old and has worked perfectly fine up till now, so I'm not sure if or why one of them crapped out. Anyways I have attached the diagnostics.

babel-diagnostics-20260127-2243.zip

Disk dropped offline, check/replace its cables and post new diags fater array start.

I reseated all SATA cables and was able to at least get some SMART information in the diagnostics. Changing the SATA cable and the port didn't result in any further improvements.

However trying to run a short SMART test on the affected drive doesn't work.

New diagnostics attached

babel-diagnostics-20260128-1301.zip

Edited January 28Jan 28 by Wazzy

Disk is not providing a valid SMART report, if the cables have been replaced, it could be a failing disk.

Diags are with the array stopped.

10 hours ago, JorgeB said:

new diags fater array start.

2 hours ago, JorgeB said:

Disk is not providing a valid SMART report, if the cables have been replaced, it could be a failing disk.

Diags are with the array stopped.

That second set of diagnostics was after starting the array in maintenance mode.

This third set of diagnostics is with the array started and is not in maintenance mode.

I am still unable to run a short SMART self-test on the affected drive.

Thank you for your help!

babel-diagnostics-20260128-1603.zip

48 minutes ago, Wazzy said:

array started and is not in maintenance mode

Emulated disk1 is mountable so it should be OK to rebuild. The only question is whether to replace.

49 minutes ago, Wazzy said:

still unable to run a short SMART self-test

How are you trying to run it? The drive says no self-test has been run.

I realized the error I was making - I thought the errored disk was spun up by default (because hovering over the red x said that clicking it would spin the drive down). I manually spun the drive up and I'm currently running an extended SMART test on it. A short SMART test ran without errors. In the meantime here's the actual diagnostics.

I noticed that when booting up the server, the hard drives made a different sound than normal (there were some clicking sounds), which disappeared after a few seconds.

babel-diagnostics-20260128-2129.zip

Edited January 29Jan 29 by Wazzy

SMART attributes for disk1 look OK but might as well let it finish the extended test. It is going to take many hours.

10 hours ago, Wazzy said:

in maintenance mode.

Must be in normal mode to see if the emulated disk is mounting; it is, so assuming the contents look correct, you can replace it; disk 1 still doesn't have a valid SMART.

17 hours ago, trurl said:

SMART attributes for disk1 look OK but might as well let it finish the extended test. It is going to take many hours.

Extended SMART test returned no errors.

13 hours ago, JorgeB said:

Must be in normal mode to see if the emulated disk is mounting; it is, so assuming the contents look correct, you can replace it; disk 1 still doesn't have a valid SMART.

This time I have started the array not in maintenance mode (which I presume to be "normal mode"), and I can see SMART info through the GUI on the affected disk. Diagnostics attached. Thanks in advance

babel-diagnostics-20260129-1611.zip

Emulated disk1 mounted, and disk OK. You can rebuild on top.

https://docs.unraid.net/unraid-os/using-unraid-to/manage-storage/array/replacing-disks-in-array/#re-enabling-a-disabled-disk-rebuilding-onto-itself

In case the SMART data doesn't show up again, here's a direct screenshot from the GUI

You can examine your diagnostics yourself. SMART reports for all attached disks is in the smart folder.

Is it time to RMA the drive? More read errors were encountered while rebuilding on the same disk. From the logs, example: Jan 29 22:15:25 Babel kernel: md: disk1 write error, sector=10097871664

Everything critical is backed up so I'm not worried if the parity (from the same batch of drives) also happens to fail. Diagnostics (maintenance mode) attached

babel-diagnostics-20260129-2224.zip

Edited January 30Jan 30 by Wazzy
text didnt show up for some reason?

Better if we get diagnostics in normal (not maintenance) mode. Can't tell anything about your filesystems in maintenance mode.

Disk1 has disconnected so there is no SMART report.

Check connections, both ends, SATA and power, including splitters.

Then post new diagnostics

Doesn't look like you are using this controller (and probably shouldn't) but why do you even have it?

00:0e.0 RAID bus controller [0104]: Intel Corporation RST Volume Management Device Controller [8086:a77f]
	DeviceName: RAID Controller
	Subsystem: ASUSTeK Computer Inc. Device [1043:8882]
	Kernel driver in use: vmd
	Kernel modules: vmd

babel-diagnostics-20260213-2128.zip

On 1/29/2026 at 10:54 PM, trurl said:

Doesn't look like you are using this controller (and probably shouldn't) but why do you even have it?

00:0e.0 RAID bus controller [0104]: Intel Corporation RST Volume Management Device Controller [8086:a77f]
	DeviceName: RAID Controller
	Subsystem: ASUSTeK Computer Inc. Device [1043:8882]
	Kernel driver in use: vmd
	Kernel modules: vmd

I disabled the VMD controller in the BIOS. I never touched this setting beforehand so it seems to have been turned on by default during the initial setup.

All cables were reseated, including power, fans, SATA cables etc.

New diagnostics attached, this time with the array started.

Edited February 14Feb 14 by Wazzy

Looks OK (like before)

On 1/29/2026 at 4:18 PM, trurl said:

Emulated disk1 mounted, and disk OK. You can rebuild on top.

https://docs.unraid.net/unraid-os/using-unraid-to/manage-storage/array/replacing-disks-in-array/#re-enabling-a-disabled-disk-rebuilding-onto-itself

Rebuilding on top constantly fails. I have changed the SATA cables again, using Startech instead of Aliexpress stuff, and the rebuild failed via the linked directions. Also, the disk dropped off the array completely as if it were removed from its bay. So I tried again, changing the bay the affected hard drive was in, in case there was an issue with the SATA backplane. Same issues as before, with the rebuild failing. Diagnostics attached along with a screenshot of the log after the second rebuild attempt failed.

I am not sure where to proceed as I have replaced all SATA cables with brand-name stuff, changed ports on the motherboard (I have 6 SATA ports), and reseated all other cables. Is it time to RMA the drive? It's making a different noise than usual upon startup, with more clicking than usual.

babel-diagnostics-20260218-1712.zip

Edited February 18Feb 18 by Wazzy

Yeah, seems it's time to consider this drive dead

9 hours ago, Kilrah said:

Yeah, seems it's time to consider this drive dead

If you suspect a drive is faulty you should try to run the extended SMART test on the drive. Failing that will confirm that it needs replacing, whereas passing normally means the drive is OK. Earlier on it was mentioned that the drive could not even complete the Short SMART test which sounded ominous.

7 hours ago, itimpi said:

If you suspect a drive is faulty you should try to run the extended SMART test on the drive. Failing that will confirm that it needs replacing, whereas passing normally means the drive is OK. Earlier on it was mentioned that the drive could not even complete the Short SMART test which sounded ominous.

I can't run the SMART test at all; when I click to run it, the button greys out for a split second then goes back to orange. Running it in the command line (smartctl -t short /dev/sdc) says "Read SMART Data failed: scsi error aborted command". The disks are spun up this time so I'm not making the same mistake as before.

And now the drive sometimes doesn't show up at all in the Dashboard and only reappears after a reboot. Rebuilding on it fails after minutes instead of after a few hours.

I think the writing is on the wall and I'll be starting an RMA on the drive. It's odd because I got the drive about 3 months ago and would've expected it to last longer

2 minutes ago, Wazzy said:

scsi error aborted command

This seems more like a problem communicating with the disk instead of an actual disk problem. But we can move on. It remains to be seen if you can rebuild to a different disk.

Do you have another disk you can use for rebuild?

5 minutes ago, trurl said:

This seems more like a problem communicating with the disk instead of an actual disk problem. But we can move on. It remains to be seen if you can rebuild to a different disk.

Do you have another disk you can use for rebuild?

I don't have another drive for rebuild (I have a 26TB external drive used for backups, and I was able to successfully back my critical files overnight without issue via Krusader).

I have formatted the drive and am currently preclearing it again from the beginning to see if that resolves the issue.

32 minutes ago, Wazzy said:

formatted the drive and am currently preclearing it again

format before preclear is pointless since preclear writes zeros to the entire disk, wiping out whatever filesystem you formatted. Preclear might be a good test of the disk. No point in formatting it after preclear. And you absolutely MUST NOT format the disk after you put it back in the array for rebuild.