How to recreate X-RAID volume after accidental removal of two drives

[Chrobrego]

Hi Community,

I need hep here to recover data out of 6 drives. I have lost volume information in readynas FrontView. I first thought about consequence of upgrading to latest firmware 6.9.4 since I saw problem on first reboot after that event, but after investigating more it looks it may have been consequence of the accidental and temporary removal of two drives out of the X-Raid Array

Now Frontview shows 3 inactive volumes:

* One named data-0 (13,62 TB), type RAID5 indicating that I should remove inactive volume to use the disk. Disk #1 #2 #3,#4 #5 #6

* One named data-1 (8,19 TB), type RAID5 indicating that I should remove inactive volume to use the disk. Disk #1 #2 #3,#4

* One named data (1,82 TB), type RAID1 indicating that I should remove inactive volume to use the disk. Disk #3,#4

All disks show Red but show as part of data-0 volume, their disk state is online and shows a green light (so they are alive)

Obviously with X-RAID I should have only one volume - but I do not know how to recreate that volume while keeping the data

PS: Support did not want to help me as I have a Pro 6 hardware and that is not officially supported with OS6, however I am pretty sure that is not the reason for the problem

Thanks in advance for any advise!

[StephenB · Accepted solution]

---

@Chrobrego wrote:

PS: Support did not want to help me as I have a Pro 6 hardware and that is not officially supported with OS6,

---

Correct.  That's what "unsupported" means.

---

@Chrobrego wrote:

Obviously with X-RAID I should have only one volume - but I do not know how to recreate that volume while keeping the data

---

X-RAID only has one volume, but there will be multiple RAID groups underneath when you have mixed disk sizes (or vertically expanded in the past).  When the volume is inactive, the system shows you these RAID groups.

In your case it appears that you have

• 6 TB drives in bays 1,2
• 8 TB drives in bays 3,4
• 3 TB drives in bays 5,6

Is that correct?  

Recovering data on your own is probably best done in with a recovery tool like ReclaiMe ( https://www.reclaime.com/).  This would require that you get all the disks connected to a PC (likely with one or more USB docks).

Alternatively you could try going in with ssh, and forcing the out of sync RAID volume to assemble with mdadm, which might need the --force option.  You'd then have to manually mount the BTRFS data volume.  There likely is data corruption, so you should mount the volume as read-only.  If the data is still there, immediately back it up.  Personally I'd then create a new volume and restore the data.   

This is a risky thing to try, spending the money on the USB dock(s) and ReclaiMe is a better option.  The basic steps are here: https://community.netgear.com/t5/Using-your-ReadyNAS/Disk-volume-no-longer-active-force-activation/m... (though you don't need to install anything if you are doing this on the NAS itself).

[StephenB · Accepted solution]

---

@Chrobrego wrote:

PS: Support did not want to help me as I have a Pro 6 hardware and that is not officially supported with OS6,

---

Correct.  That's what "unsupported" means.

---

@Chrobrego wrote:

Obviously with X-RAID I should have only one volume - but I do not know how to recreate that volume while keeping the data

---

X-RAID only has one volume, but there will be multiple RAID groups underneath when you have mixed disk sizes (or vertically expanded in the past).  When the volume is inactive, the system shows you these RAID groups.

In your case it appears that you have

• 6 TB drives in bays 1,2
• 8 TB drives in bays 3,4
• 3 TB drives in bays 5,6

Is that correct?  

Recovering data on your own is probably best done in with a recovery tool like ReclaiMe ( https://www.reclaime.com/).  This would require that you get all the disks connected to a PC (likely with one or more USB docks).

Alternatively you could try going in with ssh, and forcing the out of sync RAID volume to assemble with mdadm, which might need the --force option.  You'd then have to manually mount the BTRFS data volume.  There likely is data corruption, so you should mount the volume as read-only.  If the data is still there, immediately back it up.  Personally I'd then create a new volume and restore the data.   

This is a risky thing to try, spending the money on the USB dock(s) and ReclaiMe is a better option.  The basic steps are here: https://community.netgear.com/t5/Using-your-ReadyNAS/Disk-volume-no-longer-active-force-activation/m... (though you don't need to install anything if you are doing this on the NAS itself).

[Chrobrego]

---

Hi StephenB,

@StephenB wrote:

In your case it appears that you have

• 6 TB drives in bays 1,2
• 8 TB drives in bays 3,4
• 3 TB drives in bays 5,6

Is that correct?  

Yes, this is correct! 

I will evaluate the suggestions you provide and let you know.

Thanks for the insight!

[Chrobrego]

This is the output I get using ssh:

root@readynas:/dev# mdadm --assemble --scan
mdadm: /dev/md/data-0 assembled from 4 drives - not enough to start the array.
mdadm: /dev/md/data-1 assembled from 2 drives - not enough to start the array.
mdadm: No arrays found in config file or automatically

root@readynas:/dev# btrfs device scan
Scanning for Btrfs filesystems

root@readynas:/dev# btrfs fi show
Label: '540ed253:root' uuid: 96ca5c06-fcd1-45b5-8d56-b1aaba579bc3
Total devices 1 FS bytes used 688.46MiB
devid 1 size 4.00GiB used 3.98GiB path /dev/md0

warning, device 1 is missing
warning, device 1 is missing
bytenr mismatch, want=21135360, have=0
ERROR: cannot read chunk root
Label: '540ed253:data' uuid: e3fc18a6-b9f0-45f0-8360-c4656bea3f20
Total devices 3 FS bytes used 21.23TiB
devid 3 size 1.82TiB used 729.00GiB path /dev/md127
*** Some devices missing

root@readynas:/dev# mount -o ro -t btrfs /dev/md127 /mnt
mount: wrong fs type, bad option, bad superblock on /dev/md127,
missing codepage or helper program, or other error

In some cases useful info is found in syslog - try
dmesg | tail or so.

root@readynas:/dev# dmesg | tail
[24649.095056] md: export_rdev(sda4)
[24649.095069] md: unbind<sdd4>
[24649.099080] md: export_rdev(sdd4)
[24649.099093] md: unbind<sdc4>
[24649.101660] md: export_rdev(sdc4)
[24649.101671] md: unbind<sdb4>
[24649.105062] md: export_rdev(sdb4)
[24799.922358] BTRFS info (device md127): has skinny extents
[24799.922708] BTRFS error (device md127): failed to read the system array: -5
[24799.928021] BTRFS error (device md127): open_ctree failed

Outstanding error seems to be that Device 1 is missing?

[Chrobrego]

Update:

I now applied the --force option (or indeed --really-force) on the command "mdadm --assemble --scan"

... and it magically solved the whole problem!!

The data is back!

BIG THANKS!