Old TS-RVH8.0TL/R6 has crashed but drives are OK

[jason--r]

So my old reliable TS-RVH8.0TL/R6 has crashed on me but there were no trouble with the drives and they were all in a RAID5 array. I was reading that I could rebuild the firmware with a EM USB disk https://buffalonas.miraheze.org/wiki/Create_Recovery_Disk_for_Intel-based_Terastation but I couldn't find if it would wipe all the stored data or not if I did this.

Typically, the NAS died and then the server that it backed up died; so, kinda need the data back.

If that fails, since this is essentially Linux, would I also be able to put the drives in a computer and get Linux to rebuild the array with MDRaid?

Thanks, for any help

-Jason

[1000001101000]

I wouldn't recommend doing anything with the firmware updater if you have important data on the device, but If you don't check any of the boxes for reformat/repartition/etc it should attempt to re-install the firmware without wiping the data.

Since this is an Intel-Based device you can attach a keyboard/mouse/monitor and boot into a live Linux environment and access the data as you described. I've used the Ubuntu Live image on a usb drive for this purpose before (make sure to set the rear switch to USB).

Once you have your data backed up you can restore the firmware as you mentioned or even install Debian Linux:
https://github.com/1000001101000/Debian_on_Intel_Terastations

[jason--r]

Thanks for the reply.

Well, I've been doing some shenanigans myself to figure it out. I've pulled out the disks and chained them in a testbench. Then I loaded the latest System Rescue CD and lo and behold, it sees the data and reconstructed the raid ... BUT now I've run into these delightful errors:

Code: [Select]

[root@sysresccd /mnt]# mount -o ro /dev/md124 /mnt/md124
mount: /mnt/md124: mount(2) system call failed: Structure needs cleaning.
[root@sysresccd /mnt]# xfs_repair -n /dev/md124
Phase 1 - find and verify superblock...
xfs_repair: V1 inodes unsupported. Please try an older xfsprogs.

Which seems to either suggest the filsystem is fubar or it's actually just a versioning issue of the XFS tools and filesystem.

SO, my next step was to try and go back to an older version of SRCD but those only go back to 2017. Next try is with Knoppix. I tried version 7.0.5 (circa 2012) but that doesn't recognize the RAID at all.

When I try to boot with a KB and Mouse on the Buffalo NAS, I get hung up on the boot, and I'll probably run into the same XFS versioning issue with Ubuntu that I did with SRCD.

I think my next steps would be to DD the disks to file images and then try them again in the NAS with a recovered Firmware. That way if the most horrible thing happens, then at least I'll be able to try again or something else.

Incidentally, I have been digging around in the data recovery tools out there and found an interesting one from DiskInternals that seems to be able to reconstruct and read the filestructure (named RAID Recovery) using the trial. It's a decent chunk of cash $500, but you know, after these long nights, might be something I'm gong to be buying HAHA.

Other than that, I'm out of ideas.

Thanks again!

[1000001101000]

Knoppix/etc don't always have mdadm installed. In the past I have used an Ubuntu LTS 12.xx version and it was as simple as:

Code: [Select]

apt-get install mdadm

[jason--r]

Well darn haha. Already spent the money but good news it that that software did work. Thanks for the help