Buffalo NAS-Central Forums

I was out of town when we had a power outage and when I got back and booted up the terastation it reported a raid error on all 4 drives. It's clear that at least one drive is okay since the unit rebooted successfully (since the OS partition is mirrored over all 4 drives) so my guess is that the raid filesystem information on partition 3 was corrupted. I have 2nd brand new terastation that was not online during the power failure and have tried swaping the disk cages between this new unit and the old one. Unfortunately, this didn't work (i.e. the unit with the old disks show the same error while the new disks are fine) which is why I think it's a problem with the data on the disk and not a hardware issue.

While much of the data was backup and the original files are still around, there is a small amount of personal information that was not stored elsewhere and I'd really like to get it back. The old unit is still under warranty and Buffalo will replace it, but I suspect that a reformat would do as well. I just want to make all reasonable attempts to recover the data before I do that. Any suggestions would be very much appreciated.

For what it's worth, the old system was running firmware rev 1.08, the new one is 1.12.

Thanks in advance

have you tried accessing via EM-mode?? by holding the init button and then simultaneously pressing the power button, that might get you access to the device, however, I wouldn't know if you would get access to the data.

Also maybe consider flashing your firmware.

_________________
1 LS-GL: (FreeLink - 2.6.22 kernel)- backup and media server...
As of June 11, 2010:
admin@LS320:~$ uptime
20:23:40 up 417 days, 20:09, 1 user, load average: 0.20, 0.21, 0.30

That's only because I had to move that time ago. Love this box and this community, wish I had more time to dedicate to it.

Don't know much about Raid 5 or TS but had the odd disk or two fail so sorry to hear that.
Usually one error fails at a time -although this isn't always true for a power cut or say a cup of coffee poured in.

Anyway, before wiping data is there a way of finding out how many disks/stripes went down - if only one it should still be able to recover something -no?

http://en.wikipedia.org/wiki/Redundant_array_of_independent_disks#RAID_5 mentions a 'write hole' vunerability in RAID 5

I can see its tricky/impossible? to recover data by one pulling out the disks individually and connecting each HD into a PC since its striped.

Hi, i've had exactly the same problem after our UPS got overloaded and shut down power, i eventually got the data back despite buffalo saying there was no solution - heres the sequence of events :

Using buffalo helpline :

1 - UPS fails after which disks failed
2 - contact buffalo - advised to rebuild array - failed
3 - contact buffalo - given newer firmware - firmware update failed and unit only loaded enginering mode

Me playing around

1- take out cage, wipe partion on a fault disk - stick in other terrastation
2 - other terrastation repairs/rebuilds array ? then fails on repair/rebuild ( think the repair worked, the rebuild failed )
3 - take cage out and move to original terrastation - flash with old firmware and the terrastation repairs/rebuilds array
4 - the original terrastations still ----ed since on power of loses firmware and goes to engineering mode
5 - move disk cage to other terrastation and works fine