Datavolume doesn't exist ! message

[illogical]

Wayne, Txs for the heads up re the Datavolume doesn't exist. My drives all read good, but unassigned and went this way one day after my transferring a great deal of data across to them. I don't get why all four would fail at once.

I have spent 1/2 day trying different things after a few calls to the WD 'help' line. Here's what I've found to date, and I find it 'illogical' hence my blog name tag.

I tried installing my old HD's across to the working RMA NAS replacement unit, to try and recover my data without success. What I found really strange was that on any drive inserted it came up with firmware version 2.1.92 was now installed on the unit whereas it had been 2.1.96 before indicating that the firmware is loaded on the HD's and not the unit its self .

As I was unable to install and read my old drives(HD's unassigned) WD help then suggested that I update my 'old' HD's with the latest firmware, trouble was that  the units auto update told me I had the latest firmware installed albeit 2.1.92. There is a manual update option but you need  the file and it's nowhere to be found as i can see on the WD download site, and anyway it's not as upto date as my RMA replacement anyway. I spoke to WD again and they to date were unable to help me with a file copy of 2.1.96.

During the swapping of drives i managed to loose 1 drive of the RMA replacement and only a hard reboot (push button on rear of the unit) recovered it for me. I went systematically through my 'old' HD's into the RMA replacement until finally it gave up the ghost totally and refused to boot or turn off, which was exactly the fault i had on my first unit which i bought back in NOV 2009! (this is my third replacement).  Pushing the reset button had no effect.

Next  I reloaded my 'old' drives into the RMA replacement and it booted and switched off OK, but of course no data as all the drives are unassigned. Therefore it appears that the new RMA drives are knackered, having self destructed during a boot sequence, and it is they that prevented the unit booting correctly.

What I just don't get is why this unit is so flaky on booting up and why on earth would you have firmware on essentially 'dumb' HD's. If they don't have an answer for me tomorrow WD will be getting their unit back as is, and I'm stuck with trying to get my money back from my original supplier.

Given all the feedback of others with data loss I really think WD should offer a  mechanism where you can get your data retrieved FOC within a say 12 month time frame, and perhaps offer an out of warranty discount scheme to assist other older WD users who suffer at the hands of this NAS unit losing their data, after all that's why we bought it in the first place.

Better still have a service centre where you can send or drop the unit into worldwide, rather that this convoluted RMA system that takes forever and a day, in my case THREE WEEKS for the last one, and your loathe to use anyway as you need to keep your data.

I have absolutely no faith in this unit keeping my data long term given the troubles I've seen and I've only used it TWO DAYS in total. It's by far the worst purchase I've ever made and to think I trusted it with my data what an idiot..you live and learn.

[fibreiv]

I had the exact same problem and 16+ hours of work later I have recovered all the data and am copying it off.

Disclaimer:  This will not restore your device to normal working condition but I was able to connect to it over the network and copy my data after these steps.  Also the info here is for one RAID 5 2.7TB  volume on the 4TB Sharespace NAS.

Here is what I did.  First you will need to SSH into the device.  You might need to go ahead and update the firmware to the latest to get the option.  I did.  Under the network services enable SSH.

Connect to the device.  I used the free program Putty

username: admin

password: welc0me

This is Linux so everything is case sensitive.

at the prompt type: mdadm -D /dev/md2

You should get a message about it not existing.  If it does then this might not be the solution for you.

Then type: mdadm --assemble /dev/md2 /dev/sda4 /dev/sdb4 /dev/sdc4 /dev/sdd4

Then type: pvcreate /dev/md2

Then type: vgcreate lvmr /dev/md2

Then type: lvcreate -l 714329 lvmr -n lvm0

Now we need to rebuild the journal and check the FS.

Type: **bleep**.ext3 /dev/lvmr/lvmr0

This will take a really long time.  I then asked me a ton of Yes/No questions.  Prob 1000+ to answer.  I ended up just holding down the enter key to just shoot through them.

After it was finally done I just typed: mount -t ext3 /dev/lvmr/lvmr0 /DataVolume -o rw,noatime

I was then able to access it from my Windows 7 computer using the UNC (\\192.168.1.6) and copy the data off.

Be warned if you reboot the NAS you will have to do this all again.  I'm not spendning any time to learn how to make the config work ever time.  I am just copying off my data and then resetting this thing back to factory defaults wipping it out completly and then copying the files back.

Hope this helps!

[fibreiv]

The **bleep** is suppose to be

f

s

c

k

.

e

x

t

3

Stupid forum software thought I was trying to cuss.  hehe

[illogical]

Txs Fibreiv. Too late for me though as i had to reboot my unit to default (which restored it to working i might add) so that i could return it for credit, and i'm now attempting to recover data from my previous HD's but its a mess doing it this way.

I'm sure your info will help a lot of people in getting their data back. it's a shame though that WD could not post this or help it any shape or form via the 'help' lines. My third unit is going back to WD today as it happens, as this failed to boot up during drive 'swaps' between units which is another failure of this unit.

Anyway many thanks for posting this and well done, at least your 16+ hours was productive unlike mine:-)

[see_serge]

I cannot login to SSH (access denied)

How can I recover my data (about 3Tb lost)

I have activated SSH obtained the latest drivers but now....

With Raid5 it should be secure !?

Any idea ?

[NJliverpool]

I have had a similar problem with a 8TB WD Share space after 8 weeks.

WD have failed to provide any information on this issue and I may have lost 2 years worth of scientific data thinking it was safe on RAID5.

Has anyone come up with a solution for stripping data off the drives or is it gone forever?

I am giving WD 14 days to come up with a solution to this known problem otherwise I will be lobbying the Universities over here in the UK to blacklist them for procurement purposes

NJ

[Cohiba88]

Thanks Fibreiv! I know what happened to my worldspace which caused teh problem, since it has happened for the second time. The POWER was shut down, due the the UPS running dry WHILE it was consistency checking from a previous power outage. The consistency check usually takes 8++ hours. I managed to get my data back. The trick was to figure out which drives to assemble. Out of 4 drives a b c d, it seemed that 2 of them were not working. But when I checked drive sde4, the drive was fine! So I assume that what happened is that when the unit rebooted, it tried to assemble a RAID5 using drives sda4 sdb4 sdc4 sdd4, and 2 were not working. I manually created a raid5 using a b c e and it accepted it using the commands you provided. Here are some of the commands I used to see the status mdadm --detail /dev/md[0123] when it did not allow me to add e drive mdadm --assemble --force status messages mdadm --monitor more investigation messages mdadm --examine /dev/sd[abcde]4 stop the raid mdm -S /dev/md0 start raid mdm -R /dev/md0 see status cat /proc/mdstat more status, which is like what you will see int eh web interface tail -f /var/log/messages I wish WD would somehow update their software to be able to recover from these issues, since the data was acutally not damaged. And there are a lot of people getting this error and losing their data. -- although not the complete process, here is a capture of the last moments of recovery. Note that the lvmr figure had to be changed from what fibreiv wrote. When it worked, it was tears of joy!!! dev $ pvcreate /dev/md2 No physical volume label read from /dev/md2 Physical volume "/dev/md2" successfully created /dev $ vgcreate lvmr /dev/md2 Volume group "lvmr" successfully created /dev $ lvcreate -l 714329 lvmr -n lvm0 Insufficient free extents (714218) in volume group lvmr: 714329 required /dev $ lvcreate -l 714218 lvmr -n lvm0 Logical volume "lvm0" created /dev $ **bleep**.ext **bleep**.ext2 **bleep**.ext3 /dev $ **bleep**.ext3 /dev/lvmr/lvm0 e2fsck 1.38 (30-Jun-2005) Couldn't find ext2 superblock, trying backup blocks... ext3 recovery flag is clear, but journal has data. Recovery flag not set in backup superblock, so running journal anyway. NASRAID: recovering journal

[Cohiba88]

*sorry for the repost, the previous post did not format properly* Thanks Fibreiv! I know what happened to my worldspace which caused teh problem, since it has happened for the second time. The POWER was shut down, due the the UPS running dry WHILE it was consistency checking from a previous power outage. The consistency check usually takes 8++ hours. I managed to get my data back. The trick was to figure out which drives to assemble. Out of 4 drives a b c d, it seemed that 2 of them were not working. But when I checked drive sde4, the drive was fine! So I assume that what happened is that when the unit rebooted, it tried to assemble a RAID5 using drives sda4 sdb4 sdc4 sdd4, and 2 were not working. I manually created a raid5 using a b c e and it accepted it using the commands you provided. Here are some of the commands I used to see the status mdadm --detail /dev/md[0123] when it did not allow me to add e drive mdadm --assemble --force status messages mdadm --monitor more investigation messages mdadm --examine /dev/sd[abcde]4 stop the raid mdm -S /dev/md0 start raid mdm -R /dev/md0 see status cat /proc/mdstat more status, which is like what you will see int eh web interface tail -f /var/log/messages I wish WD would somehow update their software to be able to recover from these issues, since the data was acutally not damaged. And there are a lot of people getting this error and losing their data. -- although not the complete process, here is a capture of the last moments of recovery. Note that the lvmr figure had to be changed from what fibreiv wrote. When it worked, it was tears of joy!!! dev $ pvcreate /dev/md2 No physical volume label read from /dev/md2 Physical volume "/dev/md2" successfully created /dev $ vgcreate lvmr /dev/md2 Volume group "lvmr" successfully created /dev $ lvcreate -l 714329 lvmr -n lvm0 Insufficient free extents (714218) in volume group lvmr: 714329 required /dev $ lvcreate -l 714218 lvmr -n lvm0 Logical volume "lvm0" created /dev $ **bleep**.ext **bleep**.ext2 **bleep**.ext3 /dev $ **bleep**.ext3 /dev/lvmr/lvm0 e2fsck 1.38 (30-Jun-2005) Couldn't find ext2 superblock, trying backup blocks... ext3 recovery flag is clear, but journal has data. Recovery flag not set in backup superblock, so running journal anyway. NASRAID: recovering journal

[Brian731]

Ok, I too am having this problem (Datavolume doesn't exist), but I am not set up in RAID5. I made the mistake of going with stripped (RAID0) under the impression that only a proper hard-disk failure could bring the system down

I almost went with RAID 5, but it seems as though people haven't faired much better with that, but i digress.

I am using a 2TB (4x500gb) WD Sharespace, and there appears to have been a minor hicup on drive 3. At first I could still see the Files, however after a restart I am now getting the "datavolume doesn't exist" message. All drives appear to be 'good' but 'unassigned' under disk manager in the WD gui. I've enabled SSH, downloaded putty, and managed to log in as "root" with the default password "welc0me". I'm not familiar with linux commands, and am only slightly tech savy. I've managed to run the "df" and "mdadm --detail /dev/md[0123]" commands, but as I am not using RAID5 (as others) I'm not sure where to go from there to, do anything at all to, access the files.

md2 and md3 do not appear active on my NAS, but i'm not sure if that's because they're corrupt, or because of the way RAID0 works... I was hoping some of you lovely folks who are more familiar with these commands could help point me in the right direction... it seems a rather steap learning curve for someone who cringes at the very sight of a command line interface :'(

Any help would be greatly appreciated

-Brian

[Storm8]

Hello.

Guess what problem do I have on my WD SS 4TB in RAID5

Can pretty please with sugar on top someone help me with the data recovery trough SSH? I know nothing about linux...

Basic info:

accessing with Win 7 Ultimate x64

WD connected directly to PC

Firware: 2.1.92 (and says its the latest)

Trough Putty I managed to access the drive

login: root

pw: welc0me

- i get ASCI logo saying S6M-4MC

then I followed the guide:

mdadm -D /dev/md2

       --> mdadm: md device /dev/dev2 does not appear to be active.

mdadm --assemble /dev/md2 /dev/sda4 /dev/sdb4 /dev/sdc4 /dev/sdd4

       --> mdadm: dev/md2 assembled from 2 drives - not enough to start the array.

following commands just show me various errors.

Any clues?

Thank you very much for any kind of help!!