Hello all, I think I've made a major stuff up while trying to replace a 6tb drive with a new 20tb drive. Pre clear and formatting of the new 20tb drive went to plan, however, I wasn't able to remove the 6tb drive (as disk 5) in my array.

So I foolishly went ahead and proceeded into Tools > New Config and under Preserve current assignments: checked 'Array Slots'.

This has created a new config which I'm pretty sure has wiped my Array/Pool config. I've yet to start the array as then message 'All existing data on this device will be OVERWRITTEN when array is Started' is appearing next to the parity drive.

Is there a way I can recover the pool settings from the USB drive?

53 minutes ago, DrewFechn said:

wasn't able to remove the 6tb drive (as disk 5)

Could you explain this in more detail?

Is there anything wrong with that drive?

Attach Diagnostics ZIP to your NEXT post in this thread.

My 6tb drive (disk 5) has had increasing 'UDMA CRC error count' errors/warnings. I suspect this is due to the pins being damaged/stripped as I have an adapter on the drive, image attached. But it has been working ok. This changed when I changed the "Default spin down delay:" from none to 1 hour, as the next morning the drive was not detected. This also occurs during some reboots/power outages.

In the meantime I had started to clear and format the 20tb drive (disk 6), I stopped the array placed it in maintaainence mode to re discover the 6tb drive. It did start a data rebuild, but failed pretty quick due to read errros. Instead of rebooting like I normally do to see if the disk would be again picked up, I went into mainatinence mode, removed the 6 tb drive, then tried to replace it with the 22tb drive in disk 6. However, I couldn't start the array to do incorrect config, it did try a rebuild and this is when I decided to create a new config.

While waiting for my post to be approved, I got the usb drive and renamed 'super.dat' to super1.dat and then changed super.old into super.dat.

As per the screenshot, it appears to have assigned the drives as they were, with "All existing data on this device will be OVERWRITTEN when array is Started" next to disk 5 (6tb) and the message "Start will start Parity-Sync and/or Data-Rebuild." near the start array prompt. I'm concerned the FS is picking up as 'auto'.

NOTE: Diagnostics dump has been taken after I renamed super.old into super.dat and rebooted. I then powered the system down last night.

Screens after booting this morning :

ur-diagnostics-20250703-1939.zip

Edited July 4, 20251 yr by Drew_F
Attached image of sata adapter in use / Order of images & additional info

Disk5 looks healthy, best to probably just do a new config and resync parity with that disk in place, do you also want disk6 to be added to the array as a new disk?

Yes please.

What steps do I need to follow?

Tools - New config - Keep all

Then check assignments are correct and start the array to begin a parity sync, disk6 will need to be formatted.

I clicked apply

Then drives appear in order, however, it looks like it's suggesting the parity drive will be erased?

Edited July 4, 20251 yr by Drew_F
Additional Context

That is normal, parity will be resynced, but before starting can you confirm if this refers to disk5?

10 hours ago, Drew_F said:

It did start a data rebuild, but failed pretty quick due to read errros.

Yes, disk 5 (aka 6tb) was the drive that had errors.

Hopefully, that didn't do damage, but in that case, before starting with the new config, I would recommend mounting disk5 read-only with the UD plugin to confirm it mounts and contents look correct

I 'unassigned' disk 5, then went to Unassigned Devices, settings, set as 'Read Only'. Then when I hit mount it errors out.

Below is from the syslog.

Jul  4 19:32:04 omnicorp kernel: XFS (sdc1): Corruption warning: Metadata has LSN (7:6) ahead of current LSN (1:0). Please unmount and run xfs_repair (>= v4.3) to resolve.
Jul  4 19:32:04 omnicorp kernel: XFS (sdc1): log mount/recovery failed: error -22
Jul  4 19:32:04 omnicorp kernel: XFS (sdc1): log mount failed
Jul  4 19:32:05 omnicorp unassigned.devices: Mount of 'sdc1' failed: 'mount: /mnt/disks/WD-WX21DA8DEZ28: wrong fs type, bad option, bad superblock on /dev/sdc1, missing codepage or helper program, or other error.        dmesg(1) may have more information after failed mount system call.'
Jul  4 19:32:05 omnicorp unassigned.devices: Partition 'WD-WX21DA8DEZ28' cannot be mounted.

I suspect I will need to follow the Space Invaders tutorial below to fix the drive.

Bit of a dense question, if I start the array in it's current state in 'Maintenance mode', nothing will be changed?

https://www.youtube.com/watch?v=JVjB0ROdLBA

Check filesystem on that disk, you can do it in UD, or just type

xfs_repair -v /dev/sdc1

Then try to mount again

Excellent, that appears to have worked and I can mount and view the directories ok in UD as Read Only.

So I can unmount, reassign and kick off the parity sync now?

Edited July 4, 20251 yr by Drew_F

41 minutes ago, Drew_F said:

So I can unmount, reassign and kick off the parity sync now?

Yep

Thank you! Will report back once its done fully.

Only 0.1% in and assume these errors are nothing major?

Edited July 4, 20251 yr by Drew_F

Any read errors are not great, post new diags.

When I woke up this morning, so I've paused the sync for now.

Diags attached.

ur-diagnostics-20250705-0910.zip

So I've stopped the array, shutdown completely. I suspected the cable from the HBA card do not like this adapter on drive, so I have replaced it with a right angled SATA cable into my motherboard directly. Then rebooted. I was able to mount the drive in UD and view directories and files. So I've retsarted the array and parity sync, but am still getting errors. I suspect this drive might be on its way out.

For now have left parity sync running.

ur-diagnostics-new-sata-cable-20250705-0952.zip

Have paused the sync. Errors still appearing.

SMART still looks good, but the errors are logged as an actual disk problem, I would recommend trying to clone that disk with ddrescue, or manually copying what you can to another disk.

Re: "manually copying what you can to another disk."

I have paused the Parity Sync and have started to copy the existing directories from disk 5 (6tb) onto disk 6 (20tb) with krusader.

Once this is completed, my understanding is I will need to do the following:

Stop the array, unassign both disk 5 and disk 6.

Shutdown.

Disconnect disk 5 (6tb) completely.

Reboot

Create a new config with the 20tb drive as disk 5, and start a parity sync.

Edited July 5, 2025Jul 5 by Drew_F

You just need to do a new config and not reassign disk5, no need to actually disconnect it, but you can also do that of course.

Yep, I'll remove it and use it as some offline storage for non critical stuff. I don't trust it as part of the array!

With the new config, will the order of disks matter? Obviously I will keep the existing parity drive as parity, my understanding of how the disks are labelled and handled (eg mnt/disk1 etc) is limited.

38 minutes ago, Drew_F said:

With the new config, will the order of disks matter?

Nope, you can reorder them if you like.