Buffalo Forums

Products => Storage => Topic started by: yorkshirebeef on August 12, 2014, 04:13:14 AM

Title: Dead LS-WXL Disks ok?
Post by: yorkshirebeef on August 12, 2014, 04:13:14 AM
Hi

This is my first post :-)

I've had a Buffalo LinkStation Duo 2TB (2 x 1tb) since 2011 and its been working fine with the drives lumped together into one partition, ie not mirrored. The week before last, lightening struck our house during a storm and killed a number of electrical items including the linkstation (surge protection made zero difference). My assumption currently is that the power surge fried the motherboard so the disks should be fine (i've not tested the assumption yet).

My question is around how I can get the data back. The house insurance is replacing the NAS so if its another linkstation duo can I put these existing disks in to access the data? Can I plug them into a pc/linux box to access the data?

Any help will be gratefully received. I have backups of the data but not sure how recent as the linux server I backed them up to also fell fowl of the lightening but I know the motherboard is definitely fried on that one so I can replace the server and the disks will be fine as they are just configured as single disks.

Thanks in advance.

Mark
Title: Re: Dead LS-WXL Disks ok?
Post by: davo on August 14, 2014, 12:24:24 PM
Technically if there is nothing wrong with the disks then you should be able to insert them into a new unit and regain access. Although it is possible that the partitions on one or both of the disks may of become corrupt as a result of the power surge.

If you plug them into a Windows PC then you would need to use UFS explorer to recover the data.
Title: Re: Dead LS-WXL Disks ok?
Post by: yorkshirebeef on August 30, 2014, 04:25:37 PM
Success! It was surprisingly easy in the end. This is what I did (for anyone else in a similar predicament):

I used my HP microserver which is running Ubuntu 12.04.

1. Check raid support is loaded into the kernel

cat /proc/mdstat

Mine was blank so I ran:

modprobe raid0
modprobe raid1

2. Install mdadm

sudo apt-get mdadm


3. Shutdown the server/pc and install the disks

4. As part of the boot process (I'm assuming) the disks are scanned and the raid information is read, which populated my /proc/mdconf and created /dev/md devices, ie

root@morpheus:/mnt# cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md124 : active raid1 sdd2[1] sdc2[0]
      4999936 blocks [2/2] [UU]

md125 : active raid0 sdc6[0] sdd6[1]
      1923499776 blocks 64k chunks

md126 : active (auto-read-only) raid1 sdd5[1] sdc5[0]
      999872 blocks [2/2] [UU]

md127 : active raid1 sdc1[0] sdd1[1]
      999872 blocks [2/2] [UU]

unused devices: <none>

5. So basically all you then need to do is mount the devices

mount -t auto /dev/md124 /mnt

Bobs your uncle!

The most useful doc I used was: https://raid.wiki.kernel.org/index.php/RAID_setup

thanks,
Mark
Browser ID: smf (is_webkit)
Templates: 1: Printpage (default).
Sub templates: 4: init, print_above, main, print_below.
Language files: 1: index+Modifications.english (default).
Style sheets: 0: .
Hooks called: 59 (show)
Files included: 27 - 1055KB. (show)
Memory used: 720KB.
Tokens: post-login.
Queries used: 14.

[Show Queries]