Wanted to enlarge the array, bought two 16tb drives. Swapped the 8tb parity drive for a 16tb and rebuilt. I then removed a 4tb drive (disk3), added the other 16tb drive, and rebuilt the data.

 

There some disk read errors on disk1, but notifications said the data-rebuild finished without errors, but the logs shows errors. Which one is it? If there are errors does that mean the rebuilt disk3 data is corrupted?

 

Dec  7 10:55:03 Tower kernel: md: disk1 read error, sector=3283476200
Dec  7 10:55:03 Tower kernel: md: recovery thread: multiple disk errors, sector=3283476200

 

| 08-12-2024 12:14 | Unraid Data-Rebuild              | Notice [TOWER] - Data-Rebuild finished (0 errors) | Duration: 1 day, 5 hours, 33 minutes, 19 seconds. Average speed: 150.4 MB/s | warning |
| 07-12-2024 11:21 | Unraid Disk 1 SMART health [197] | Warning [TOWER] - current pending sector is 16    | ST8000DM004-2CX188_WCT0GB40 (sdf)                                           | warning |
| 07-12-2024 11:21 | Unraid Disk 1 SMART health [5]   | Warning [TOWER] - reallocated sector ct is 8      | ST8000DM004-2CX188_WCT0GB40 (sdf)                                           | warning |
| 07-12-2024 10:56 | Unraid array errors              | Warning [TOWER] - array has errors                | Array has 1 disk with read errors                                           | warning |

 

 

I tried to run an extended smart check on disk1 and that seems to have let to it dying, it's unresponsive and is making bad sounds.

 

If there are data-rebuild errors I'd like to put the old disk3 back and rebuild disk1, and then redo the expansion of disk3. I don't think I touched the array since rebuilding the parity, is there a way to see in the logs? I suppose I'd have to force the a new config to go back to as it was.

 

If there aren't any data-rebuild errors, I guess I can just rebuild disk1.

 

In summary, is the rebuilt disk3 corrupt, and if so is there any risk trying to restore the old disk3 and rebuilding disk1?

 

Thanks in advance.

tower-diagnostics-20241209-1740.zip

1 hour ago, cloudsox said:

There some disk read errors on disk1, but notifications said the data-rebuild finished without errors, but the logs shows errors. Which one is it?

Notifications don't always show the last result, see on main page, but if there are errors in the log, there were errors.

I think I might have messed up. I tried to reinstate the old disk3 via new config in order to rebuild disk1 but disk1 is not being emulated. Any advice?

You can try this, but it will only work if parity is still valid with the old disk:

 

-Tools -> New Config -> Retain current configuration: All -> Apply
-Check all assignments and assign any missing disk(s) if needed, including the new disk you want to rebuild, replacement disk should be same size or larger than the old one
-IMPORTANT - Check both "parity is already valid" and "maintenance mode" and start the array (note that the GUI will still show that data on parity disk(s) will be overwritten, this is normal as it doesn't account for the checkbox, but it won't be as long as it's checked)
-Stop array
-Unassign disk1
-Start array (in normal mode now) and post new diags

I'm unable to assign disk1 because that disk is now unresponsive. Should that make a difference when starting the array in maintenance? I don't see it being emulated.

 

I've attached two diagnostics, the earlier with the original parity and disk3, and the later with the rebuilt parity and disk3.

tower-diagnostics-20241218-1500.zip tower-diagnostics-20241211-1238.zip

2 hours ago, cloudsox said:

I'm unable to assign disk1 because that disk is now unresponsive.

You will need to use a new disk1, same size or larger than the old one.

So new config retaining current, keep parity, d2, d3 from original, new d1. Then start in maintenance, stop, remove d1, restart?

 

In normal mode it says d1 "Unmountable: Unsupported or no file system".

tower-diagnostics-20241218-2014.zip

Check filesystem on the emulated disk1, run it without -n, and if it asks for -L use it.

No check filesystem options appear for the emulated disk1.

edit: forced filesystem from auto to xfs_encrypted and options appeared.  After repairs disk1 is now showing, thanks a million! Showing some files under "lost+found". Could you explain to me what has happened and how I might further fix my files? Thanks

 

Edited December 18, 20241 yr by cloudsox
update

Files that xfs_repair can't recover totally, or are missing some details like the file name, are moved to lost+found, you will need to sort them manually.