Hey everyone,

Woke up the other monring and all my dockers were gone. I set them all up again to notice that one of my discs now says: unmountable: wrong or no file system.

If I start the array in maintaince mode, I get the following message under xfs_repair status:

Phase 1 - find and verify superblock...
superblock read failed, offset 0, size 524288, ag 0, rval -1

fatal error -- Input/output error

I'm going to take this as the drive is dead ? If I was to formatt it and it works, am I going to have the same issue again ?

Please post the diagnostics.

Sorry. Attached. I get the feeling the drive is gone. I can here it make that click sound every now and then. tower-diagnostics-20250521-2016.zip

Which disk was it?

Just now, itimpi said:

Which disk was it?

Its Disk 3 • WDC_WD20EZAZ-00GGJB0_WD-WX31A39445RD (sdb)

Thought it might be as that drive appeared to not be online and able to give a SMART report.

Before deciding the drive really has died I suggest carefully checking the cabling (in particular power) in case on of them has worked loose due to vibration.

Since no parity drive shows in the diagnostics I hope you have backups of any critical data on the drive?

25 minutes ago, itimpi said:

Thought it might be as that drive appeared to not be online and able to give a SMART report.

Before deciding the drive really has died I suggest carefully checking the cabling (in particular power) in case on of them has worked loose due to vibration.

Since no parity drive shows in the diagnostics I hope you have backups of any critical data on the drive?

I will give it a wiggle. It was just media which I can download again. I'm not thrilled but its not dire.

If i do check the cables and it does not fix, if I formatt it and itw orks again, should i not trust it ?

It would be good to see a SMART report, or run an extended SMART test.

Yep, pulling it out did not fix it. Running an extended one now but attached is the short one. I think it is cooked. tower-smart-20250521-0435.zip

Test aborted, make sure nothing is using the device and try again.

22 hours ago, JorgeB said:

Test aborted, make sure nothing is using the device and try again.

Sorry, the full test too a while.

tower-smart-20250522-0307.zip

Test passed, replace both the power and SATA cables for that disk, since there were constant ATA errors before, and post new diags after array start.

Thanks. I'll give it a shot. I have all the hard drives running of one sata power rail, so that might be the issue.

Edited May 22, 20251 yr by Wintermute1987

1 hour ago, Wintermute1987 said:

I have all the hard drives running of one sata power rail

If you mean a SATA splitter from one connector to more than two SATA connectors, that is not a good idea.

On 5/22/2025 at 11:33 PM, JorgeB said:

If you mean a SATA splitter from one connector to more than two SATA connectors, that is not a good idea.

That is what I mean. 5 drives off sata rail.

So, I swapped out all the data cables, added another power rail for the affected drive. Still the same issue.

Attached is the new smart report.

I think the drive is cooked.

tower-smart-20250524-0109.zip

1 hour ago, Wintermute1987 said:

5 drives off sata rail.

That's a very bad idea.

1 hour ago, Wintermute1987 said:

Attached is the new smart report.

SMART looks good, and there's a recent successful extended test, but run another one.

16 hours ago, JorgeB said:

That's a very bad idea.

SMART looks good, and there's a recent successful extended test, but run another one.

Most recent. I think the drive is done.

tower-smart-20250524-1908.zip

That's the same as before, there's no new SMART test.

3 hours ago, JorgeB said:

That's the same as before, there's no new SMART test.

Strange. I ran it again. Pressed download. Is that not the correct process ?

You first need to click run the extended test, wait for it to finish, it will take a few hours, then download a new report.

18 hours ago, Wintermute1987 said:

I ran it again.

When you say you ran it - did you wait for it to complete? The Extended test takes something like 1.5 to 2 hours per TB of the disk being tested so it is not quick. Also the progress typically only updates at 10% intervals so do not assume it is stuck if the progress does not frequently update.

On 5/26/2025 at 5:40 PM, itimpi said:

When you say you ran it - did you wait for it to complete? The Extended test takes something like 1.5 to 2 hours per TB of the disk being tested so it is not quick. Also the progress typically only updates at 10% intervals so do not assume it is stuck if the progress does not frequently update.

I did. Clicked extended test. Came back next morning and clicked download smart test. I'll do it again and see.

At this point, I think I will just formatt the drive. Could it be an Unraid issue ?