Disk not recognised after firmware installation

[scrambldchannel]

I have a cloudbox and the 3tb Seagate Barracude disk it came with. It had been running fine for a couple of years with the lacie firmware but recently stopped working. I tried a factory reset with no luck. I was able to recover the data by connecting the disk to my linux machine using aa sata to usb bridge and re-assembling and mounting using mdadm and lvm. Given the disk seemed to work, I thought I would try to get it working again using the fvdw-sl firmware and using fvdw-sl-console from windows I went through the installation steps seemingly successfully.

Upon a reboot though, the NAS reverted to the fast flashing blue led. I left it for an hour or so, just in case it needed a bit of time on a first boot, but (perhaps unsurprisingly) it didn't fix itself. No worries I thought, I'll just try again and make sure I didn't give an incorrect answer. When I ran fvdw-sl-console again, I was able to get the standalone kernel running and connect via telnet but the installer failed saying that there was no disk found and fdisk suggested it didn't have a valid partition table. I tried connecting the drive to my laptop using the sata bridge again but now nothing shows up. Has the disk just coincidentally now failed or have I done something stupid? I'm happy enough buying a replacement disk but don't want to if something else is causing the problem.

Thanks for any suggestions in advance!

[fvdw]

Hi, mmm...strange behavior and not a good sign.
Your disk may be suffering from bad blocks.

What you could do is load the the standalone kernel with the disk in the cloudbox. After you have telnet connection give command

Code: Select all

dmesg

and post output.
Also of the commands

Code: Select all

fdisk -l /dev/sda

And

Code: Select all

gdisk -l /dev/sda

If gdisk shows a partion table delete it using the comman line commands of gdisk

[scrambldchannel]

Thanks for the quick reply!

Interestingly, after putting the disk back in the casing and rebooting with the standalone kernel, the filesystem was recognised again. I've attached the output of the three commands and I don't see anything obviously wrong but when I tried to reboot it still didn't work. So I deleted the gpt partition by using the "o" and then "w" options in gdisk and tried the firmware installation again. Again it seemed to complete successfully but the NAS wouldn't boot properly. I've also attached a copy of the installer.log in case that helps.

If I attach the disk via the sata bridge it does now register as a block device again and shows the 8 partitions so I'm not sure what to make of it. I had kind of assumed that a disk would either work or it wouldn't rather than working sometimes. Annoyingly, I don't have a spare disk to test it with but can I safely assume the enclosure itself is OK given that it boots into the standalone kernel or is that not necessarily true? A new disk would be cheaper than a whole new NAS, I just want to avoid buying a disk and then discovering there is a problem with the box itself.

[scrambldchannel]

OK, so in the end, I just bought a new disk figuring I'd probably just roll my own NAS if I couldn't get this working and could use it there. Anyway, it worked straight away, I guess the old disk was just borked.

[fvdw]

hi

sorry to keep you waiting, i checked the output you posted but that all seems normal, the next check would have been a disk check for bad sectors.
Anyhow if the new disk works without problems your conclusion is correct that the old disk has problems.

[scrambldchannel]

Oh that's totally fine, I wasn't expecting an instant answer and I was pretty sure the disk was dead anyway.