Question What are my options to try and resurrect a dead Raid1 HDD?

[ThePiston]

I've got a failed Raid1 array with:

1. one drive dead (offline member)
2. and the other shows "Error Occurred(0)".

What are my options for resurrecting this array using the one drive that spins? This PC is old and using Intel RST 3.6.0. Also, this is the OS array.

The drive that does spin passes SMART tests so I'm not sure what's wrong - maybe it's just the raid software (?)

Can I update the Raid drivers without killing the array perhaps? Or, can I use Acronis to clone the one drive that spins to try and clone that onto a working drive perhaps? Maybe Spinrite? I need to get this array up ASAP. Any other options?

It's the 465GB array in this image:

 

[Tech Junky]

Remove it from the software and then readd it.

If that doesn't work format it and then try to introduce it to the Raid again.

If that doesn't work there's something else going on either with the controller or cable causing it to be offline.

If all else fails then replace the drive. If that doesn't work then you have bigger issues that might require a fresh OS install. If the "non-working" drive is working though you can copy the data to it and swap it with the working one and potentially rebuild.

 

[ThePiston]

I can't format it - it's the only source of the data. I'm looking for ideas of trying to fix it either in another process (Spinrite, clone onto another drive, something else(?)) or maybe update the raid driver to get a new RST version on the mobo.

 

[Tech Junky]

I gave you 5 options.

 

[ThePiston]

thanks, but I have to tune out when the 2nd option is to nuke the only working drive in the array

 

[OlyAR15]

Maybe remove the drive, put it in an enclosure and see if you can read it with another computer?

Also, RAID 1 systems are not designed as a backup. You still need to back up your data.

 

[Tech Junky]

Windows won't let you format your system disk while you're booted into windows anyway, Remove the drive from the array and then try to initialize it again into the array. Otherwise do the enclosure / cable thing on the same PC or another to see if it's readable. Use another PC to format it and introduce it to the array again and let it sync.

 

[mv2devnull]

I've got a failed Raid1 array with:

1. one drive dead (offline member)
2. and the other shows "Error Occurred(0)".

Did these both occur simultaneously? Why is the other array in degraded state? You have had no time to replace the dead disk from it yet?


RAID 1, aka "mirror", should have rather trivial array metadata. Like other has said, take the "maybe not dead" drive out and connect it to some other computer (that knows nothing about RAID). Connect a healthy empty disk too. Then a low-level copy (aka clone) of entire suspect drive into healthy disk. My pick would be copy with 'dd' on Linux, but there are many cloning apps/systems now. There are even external "plug two drives and press button" cloning devices that need no attached computer.

You will either clone the image successfully -- including any corruption -- or the suspect disk dies in the operation. This is a "read-only" approach; nothing on the suspect disk will be modified, so less dangerous operations. Whenever one does "use" a filesystem, one probably does writes too and that is not healthy on bad disk.

When you have copy on healthy disk, then you can focus on accessing the files from it.