Unraid Disk become Unmountable with "No file system"

[artstack]

I'm still using Unraid 6.9.0 Beta 25. I just successfully added a new disk with following steps but after file copy to new disk, it become un-mountable.

 

1, Preclear disk

2.Set disk to btrfs filesystem 

3.Stop Array and add new disk (DISK9)

4.Start array and Format the disk

5.After completed  the array are good.

6.Use Krusader to copy file from /mnt/disk8/* to /mnt/disk9/

7.Successfully copy many files but suddenly Krusader start to report that it cannot write file

8.Test to creat small text file directly into /mnt/disk9/test.txt and got error the disk is Read-Only

9.Reboot Unraid and now disk become "Unmountable:No FIle System"

 

Attached the Diagnostic file below

 

Did I just lost the file on disk ? It strange that the circle status indicator in front of the disk still show green.

 

 

SMART did not show any sign of issue.

 

 

Now Unraid asked me if I want to format this disk.

 

 

 

chbvault-diagnostics-20201227-2157.zip

[itimpi]

Dealing with unmountable disks is covered here in the online documentation accessible via the ‘Manual’ link at the bottom of the Unraid GUI.

[JorgeB]

18 hours ago, artstack said:

It strange that the circle status indicator in front of the disk still show green.

That's the disks status, there is a filesystem problem:

parent transid verify failed on 22183936 wanted 519 found 510

 

This error means some writes were missed due to unclean shutdown or device issues, and it's fatal, there are some recovery options here, btrfs restore (option 2) is the most likely to work for this.

[artstack]

Thanks Itimpi and JorgeB for the recommendation and guide me to the article that help to recover the filesystem. I still have a question since I forgot to share that I have a perfect full backup on another Unraid server in a different location. In this case, is it easier for me to just pull off the physical drive and make the disk failed and Unraid mark status as "Disabled". Then I can copy file over again ? Also do you think that the Parity I have on this server with failed filesystem can save me by rebuilding after I force the Drive into "Disabled" status ? 

 

To be specific, will any of the below method works ?

 

Method 1 (Force the drive to fail, re-format as BTRFS and then rebuild onto the drive using Parity)

1. Make the physical drive "Disabled". (By shutdown server and pull SATA power cable off)

2. Try to format in Unassigned device with BTRFS filesystem

3. Add the Disk back to original DISK9 location

4. Parity rebuild.

 

Method 2 (Force the drive to fail --> New configuration --> Build parity --> Copy missing file from backup server)

1.  Make the physical drive "Disabled". (By shutdown server and pull SATA power cable off)

2. Go to New Configuration and assign the drive location.

3. Build  Partity, knowing that there will be some missing file.

4. Do file Compare and Sync with backup server and see which file is missing and copy file from another server. 

 

 

 

[itimpi]

36 minutes ago, artstack said:

Method 1 (Force the drive to fail, re-format as BTRFS and then rebuild onto the drive using Parity)

1. Make the physical drive "Disabled". (By shutdown server and pull SATA power cable off)

2. Try to format in Unassigned device with BTRFS filesystem

3. Add the Disk back to original DISK9 location

4. Parity rebuild.

All this will give you is exactly what you have now so no point that I can see.

 

38 minutes ago, artstack said:

Method 2 (Force the drive to fail --> New configuration --> Build parity --> Copy missing file from backup server)

1.  Make the physical drive "Disabled". (By shutdown server and pull SATA power cable off)

2. Go to New Configuration and assign the drive location.

3. Build  Partity, knowing that there will be some missing file.

4. Do file Compare and Sync with backup server and see which file is missing and copy file from another server. 

This is a possibility.

 

However if you have already managed a recovery then the chances are high that all your data is intact.   You could just run a compare with your backup system.

[JorgeB]

You don't need to disable the disk, just format it and restore the data from the backup server.

[artstack]

Happy New Year !!! I just back to my server after a long vacation. I did try to follow the recommendation from JorgeB by follow the Step#2. Here is the result I got. Please advise.

 

root@chbvault:/# btrfs restore -v /dev/sdg1 /mnt/disk8/restore
parent transid verify failed on 22183936 wanted 519 found 510
parent transid verify failed on 22183936 wanted 519 found 510
parent transid verify failed on 22183936 wanted 519 found 510
Ignoring transid failure
ERROR: child eb corrupted: parent bytenr=22052864 item=4 parent level=1 child level=1
Couldn't read chunk tree
Could not open root, trying backup super
parent transid verify failed on 22183936 wanted 519 found 510
parent transid verify failed on 22183936 wanted 519 found 510
parent transid verify failed on 22183936 wanted 519 found 510
Ignoring transid failure
ERROR: child eb corrupted: parent bytenr=22052864 item=4 parent level=1 child level=1
Couldn't read chunk tree
Could not open root, trying backup super
parent transid verify failed on 22183936 wanted 519 found 510
parent transid verify failed on 22183936 wanted 519 found 510
parent transid verify failed on 22183936 wanted 519 found 510
Ignoring transid failure
ERROR: child eb corrupted: parent bytenr=22052864 item=4 parent level=1 child level=1
Couldn't read chunk tree
Could not open root, trying backup super

[artstack]

Happy New Year !!! I just back to my server after a long vacation. I did try to follow the recommendation from JorgeB by follow the Step#2. Here is the result I got. Please advise.

 

root@chbvault:/# btrfs restore -v /dev/sdg1 /mnt/disk8/restore
parent transid verify failed on 22183936 wanted 519 found 510
parent transid verify failed on 22183936 wanted 519 found 510
parent transid verify failed on 22183936 wanted 519 found 510
Ignoring transid failure
ERROR: child eb corrupted: parent bytenr=22052864 item=4 parent level=1 child level=1
Couldn't read chunk tree
Could not open root, trying backup super
parent transid verify failed on 22183936 wanted 519 found 510
parent transid verify failed on 22183936 wanted 519 found 510
parent transid verify failed on 22183936 wanted 519 found 510
Ignoring transid failure
ERROR: child eb corrupted: parent bytenr=22052864 item=4 parent level=1 child level=1
Couldn't read chunk tree
Could not open root, trying backup super
parent transid verify failed on 22183936 wanted 519 found 510
parent transid verify failed on 22183936 wanted 519 found 510
parent transid verify failed on 22183936 wanted 519 found 510
Ignoring transid failure
ERROR: child eb corrupted: parent bytenr=22052864 item=4 parent level=1 child level=1
Couldn't read chunk tree
Could not open root, trying backup super

 

====================================

 

By the way, if the file on the unmountable is acceptable to be lost, what happen if I format the unmountable disk , will I just lost the file on the unmountable disk only without impacting the Parity validity or I will need to rebuild the parity all over again ?

[JorgeB]

You mentioned you had backups, just format that disk and restore from them, parity remains in sync after formatting a disk.

[artstack]

Okay, I've format the disk and now copying from backup. Looks OK so far. Thank you so much for all the helpful recommendation in this forum.