Hi 🙂

Swapped a 4tb drive for 8tb drive (disk-1). 

 

Ran pre-clear, default file system = XFS in disk settings, followed unRAID documentation and another post to change a data hard drive, ticked the box next to blue-box to rebuild from parity.  

14hrs+ later I get "Unmountable: Unsupported or no file system" and drive is formatted to XFS - See attached photos.  

Ran XFS_Repair on disk1 and "Sorry, could not find valid secondary superblock".

 

I'm unsure what to do after many hours reading posts and around the inter-webs.  I do not want to destroy any data on data disk1 and I do not have another HDD to throw in + I live around 1hr & 10mins from nearest computer store.  I did find this unRAID post/procedure, unsure if it'll work for me.

 

Looking at using either UFS Explorer or Recovery-Explorer to examine the unmountable array disk on another machine.  Annnd unfortunately I have formatted the 4tb.

 

Array remains stopped until I know how to proceed. 

Is this fixable please?

 

Thankyou in advance 🙂

 

Vers: 6.12.10

Diagnostics attached

 

Edited May 22, 20242 yr by RadRom

49 minutes ago, RadRom said:

Ran XFS_Repair on disk1 and

How did you run that? If it was on the CLI do it using the GUI, and run it without -n, check filesystem

As per documentation, stopped array and ran it in GUI terminal.

Quote

If you ever need to run a check on a drive that is not part of the array or if the array is not started then you need to run the appropriate command from a console/terminal session. As an example for an XFS disk you would use a command of the form:

xfs_repair -v /dev/sdb

 

2 minutes ago, RadRom said:

xfs_repair -v /dev/sdb

That won't work.

1 minute ago, RadRom said:

As per documentation, stopped array and ran it in GUI terminal.

 

Where in the current documentation did you see this?   When I looked it said:

If you ever need to run a check on a drive that is not part of the array or if the array is not started then you need to run the appropriate command from a console/terminal session. As an example for an XFS disk you would use a command of the form:

xfs_repair -v /dev/sdX1

 

16 minutes ago, JorgeB said:

How did you run that? If it was on the CLI do it using the GUI, and run it without -n, check filesystem

Running XFS_Repair without -n through GUI as exactly described in documentation.  Will report back when it's finished.

Edited May 22, 20242 yr by RadRom

6 minutes ago, itimpi said:

Where in the current documentation did you see this?   When I looked it said:

If you ever need to run a check on a drive that is not part of the array or if the array is not started then you need to run the appropriate command from a console/terminal session. As an example for an XFS disk you would use a command of the form:

xfs_repair -v /dev/sdX1

 

I probably mis-interpreted that and did it while array was stopped and moreso out of paranoia of data protection... Annnnnd thinking it would work, apparently won't.

Last line reads "fatal error -- File system metadata writeout failed, err=117. Re-run xfs_repair."

 

Now I'm getting nervous!

Full log attached.

XFS_Repair_Log.txt

That looks like a xfs_repair problem, do you still have the old disk intact?

3 minutes ago, JorgeB said:

That looks like a xfs_repair problem, do you still have the old disk intact?

Just my Irish luck!!  🙂

 

Yes, but I formatted it to XFS, no data written to it, just formatted it far too quickly.  I am hesitant to put that in incase I do lose data now.

 

So I guess a new HDD and throw that in??

15 minutes ago, RadRom said:

Yes, but I formatted it to XFS, no data written to it, just formatted it far too quickly.

You can try using something like testdisk, it may recover the old partition, if not, UFS explorer should be able to recover the data, but it's not free.

14 minutes ago, JorgeB said:

You can try using something like testdisk, it may recover the old partition, if not, UFS explorer should be able to recover the data, but it's not free.

Not concerned about the $65USD cost.

 

UFS Explorer Standard or Recovery-Explorer Standard?  Or they the same 'thing' essentially??

28 minutes ago, RadRom said:

UFS Explorer Standard

This is what other users have been using.

Update; this is works in progress:

 

 I just hope this thread helps some future person (and note to future-self) who sadly has a similar issue and can hopefully find a good solution 🙂

 

UFS Explorer under linux could not find any files on the 4tb I removed from the array and then prematurely formatted. <--- My bad there!!

 

I hijacked a 4tb from another unRAID (backup) server I have, ran pre-clear (non docker version) on that, then plugged that HDD in to my main unRAID array and received this message: "The replacement disk must be as big or bigger than the original."  Whilst Recovery Explorer on linux runs over my 8tb disk, it's now off to the computer shop for a new 8tb WD Red Pro or + to see if I can have that HDD rebuilt within my main array.

 

The replacement/upgraded 8tb HDD I inserted that caused me this issue was a 8tb WD Red+ brand new from shop, unformatted and passed pre-clear and swapped as per docs ..... Unsure why it or XFS_Reapir has errors - Guess it happens.

 

unRAID Version: 6.12.10

Format: XFS

XFS set as default in disk settings.

 

I'll check back in around 16+/- hours.

 

** I found no difference whatsoever between UFS Explorer and Recovery Explorer ---- Me thinks they're the exact same company  You decide there!

 

 

Edited May 23, 20242 yr by RadRom

Ok.  Same error with a brand new disk.

Ran pre-clear - logs attached.

I triple checked procedures and settings, still getting this unformatted error in my array.  Seems this will continue until I can resolve.

Any ideas please? 🙂  Cause I'm lost and scratching my head here!!

 

Recovery Explorer found files on both 4tb (prematurely formatted) and previous 8tb drive that had same unsupported format error.
** I found Recovery Explorer worked best and found files, whereas UFS Explorer could not find files on same formatted disk.  Only difference I can tell between the two apps is one is priced in $USD the other is priced in €Euro. **

 

preclear_report_WD-CA3S6T2K_2024.05.26_11.07.34.txt

Edited May 26, 20242 yr by RadRom

If you have already recovered the data form the disk you will need to re-format it before use.

15 hours ago, JorgeB said:

If you have already recovered the data form the disk you will need to re-format it before use.

Thankyou! 🙂

Q/. If I were to format to XFS and then place into main array, would that rebuild or would I have to copy the data over?

 

I'm just trying to understand what went wrong, how to prevent and how to fix.  I don't trust parity/rebuild anymore... I have used it in the past and it worked great 🙂

6 hours ago, RadRom said:

Q/. If I were to format to XFS and then place into main array, would that rebuild or would I have to copy the data over?

You format the disk already in the array, not outside, and then you would have to copy the data over.

I received a message if this question has been answered.... Not yet.  Will have an update soon.

 

Array remains offline at the moment.