NAS325-v2 volume missing after shut down.

Mijzelf

No messages is good. That means there are no errors. You should be able to find your files in /mnt/attempt2/. Right?

cd /mnt/attempt2

ls

cd <sharename>

ls

cd ..

cd <othersharename>

ls

tycoonlee

Hi, Heroic Mijzelf, Yes, Yes, Yes.
I could access my file and see the list.
How could copy it to another USB drive?

You are really a hero of my NAS.

Mijzelf

Plug the USB disk in the NAS. Preferably the front port, that is USB3. The disk will automatically be mounted, somewhere in /e-data/. You can see the mountpoint with

cat /proc/mounts | grep 'e-data'

Now enter that directory

cd /e-data/<some-long-hex-code>/

and copy everything over.

cp -a /mnt/attempt2/* .

There is a dot at the end, which defines the target directory. (Current dir). The copying will take several hours, in which you have to keep your terminal open. If you have an USB3 disk, you can expect about 35MB/sec on big files. Smaller ones are slower due to the random access time. If you don't know how much date there's on the disk,  'df -h' can tell you.

Here I assume the USB disk is big enough to contain all the data, as this way there is no way to control which data is copied first, and so you won't easily know which data is not yet copied when the disk is full.

When you have created your backup, we can try to repair the internal volume. If that fails you can create a new volume, if it succeeds you'll have your setup back in 5 minutes.

tycoonlee

Hi, Mijzelf and Happy New Year 2020!!!
I follow your instruction and it seems like the copying is in progress.

/mnt/attempt2/video # cat /proc/mounts | grep 'e-data'

/dev/sdc2 /e-data/17964beb84e029e39aea525626386d22 tntfs rw,relatime,uid=99,gid=0,umask=00,nls=utf8,case_sensitive,nocache,errors=continue,mft_zone_multiplier=1 0 0

/mnt/attempt2/video # cd /e-data/17964beb84e029e39aea525626386d22/

/e-data/17964beb84e029e39aea525626386d22 # cp -a /mnt/attempt2/* .

It would be the best new year gift if I get all the data back.
Thank you. Mijzelf!

PS: how do I repair my volume after data backup is finished?

Mijzelf

how do I repair my volume after data backup is finished?

First you have to find out if the array should be rebuild degraded, or full. To do so you'll have to try to mount sda2 the same way like you mounted sdb2, only use loop2, attemp1 and sda2, and check if your files are there.

If this fails than either that partition is more damaged than sdb2 is, or the transplanted partition table is not compatible after all. (Or you or I made a typo somewhere, but that's hardly possible, is it?)

First reboot the box to get rid of the mounted loopdevices, and remove that USB disk. Better put it safe in a drawer.

When sda2 did mount, the command to create the array is

mdadm --create --assume-clean --level=2  --raid-devices=2 --metadata=1.2 /dev/md0 /dev/sda2 /dev/sdb2

(that's a single line)

When sda2 did not mount the command to create the array degraded is

mdadm --create --assume-clean --level=2  --raid-devices=2 --metadata=1.2 /dev/md0 missing /dev/sdb2

It is possible that directly after the create the array is automatically mounted and shown in the webinterface. If not, reboot the box.

tycoonlee

Hi, I have backed up all the data on my USB drive and keep it in my safe.
I did mount the other drive successfully.
Then I input the array command, it shows "mdadm: invalid raid level: 2".
And I change level 2 to level 1. It shows array started.
The volume does show in my GUI but with Volume down.(The attachment) 
Here is output.

 ~ # losetup /dev/loop1 -o 1048576 /dev/sdb2

~ # mkdir -p /mnt/attempt2

~ # mount -o ro /dev/loop1 /mnt/attempt2

~ # losetup /dev/loop2 -o 1048576 /dev/sda2

~ # mkdir -p /mnt/attempt1

~ # mount -o ro /dev/loop2 /mnt/attempt1

~ # mdadm --create --assume-clean --level=2  --raid-devices=2 --metadata=1.2 /de

v/md0 /dev/sda2 /dev/sdb2

mdadm: invalid raid level: 2

~ # mdadm --create --assume-clean --level=1  --raid-devices=2 --metadata=1.2 /de

v/md0 /dev/sda2 /dev/sdb2

mdadm: array /dev/md0 started.

Should I re-mount the drive again? Or I miss some steps?
Does my volume be repaired?

Mijzelf

Then I input the array command, it shows "mdadm: invalid raid level: 2".
And I change level 2 to level 1. It shows array started.

Good catch. So I did make a typo after all.

Should I re-mount the drive again? Or I miss some steps?
Does my volume be repaired?

The volume is not being repaired. Did you reboot the box already? If yes, can you mount /dev/md0? If no, can you post the output of

mdadm --examine /dev/sd[ab]2

tycoonlee

Hi, Mijzelf, I do reboot the device several times.
Here is the output.

           Name : NSA325-v2:0  (local to host NSA325-v2)

  Creation Time : Wed Jan  1 20:36:18 2020

     Raid Level : raid1

   Raid Devices : 2

 Avail Dev Size : 976244928 (931.02 GiB 999.67 GB)

     Array Size : 976244792 (931.02 GiB 999.67 GB)

  Used Dev Size : 976244792 (931.02 GiB 999.67 GB)

    Data Offset : 2048 sectors

   Super Offset : 8 sectors

          State : clean

    Device UUID : 39ef2f0e:1d21b2f5:7071989d:10a45d92

    Update Time : Wed Jan  1 20:37:40 2020

       Checksum : 9aa84318 - correct

         Events : 2

   Device Role : Active device 0

   Array State : AA ('A' == active, '.' == missing)

/dev/sdb2:

          Magic : a92b4efc

        Version : 1.2

    Feature Map : 0x0

     Array UUID : a75ab14c:888457ee:92a13b50:771ae6e3

           Name : NSA325-v2:0  (local to host NSA325-v2)

  Creation Time : Wed Jan  1 20:36:18 2020

     Raid Level : raid1

   Raid Devices : 2

 Avail Dev Size : 976244928 (931.02 GiB 999.67 GB)

     Array Size : 976244792 (931.02 GiB 999.67 GB)

  Used Dev Size : 976244792 (931.02 GiB 999.67 GB)

    Data Offset : 2048 sectors

   Super Offset : 8 sectors

          State : clean

    Device UUID : 0db7287d:0c743cf5:a3e52356:7e65a061

    Update Time : Wed Jan  1 20:37:40 2020

       Checksum : 90f9937d - correct

         Events : 2

   Device Role : Active device 1

   Array State : AA ('A' == active, '.' == missing)

Mijzelf

  Data Offset : 2048 sectors

OK, that is the right offset. Can you check if the array is mounted?

cat /proc/mounts | grep md0

If not, can you mount it manually?

mkdir /mnt/somename

mount /dev/md0 /mnt/somename

tycoonlee

Hi,Mijzelf:
I did all the command and found there is a scan action in GUI.
After scanning, my volume is in a healthy status but I still could not access my shared folders.
It seems like there are two duplicate share folders in my list and could not be deleted at old one.
Does it mean I need to recreate all the share folder metadata with the list database?