[Wylug-help] Inactive RAID 10 Array

[Dave Fisher]

On Tue, Apr 14, 2009 at 03:24:49PM +0100, "richard bruce" wrote:
> on 14/4/09 3:00 PM, Dave Fisher <wylug-help at davefisher.co.uk> wrote:
> > I need some help recovering from a RAID 10 fault.  Unfortunately, I really
> 
> which raid controller are you using?   I can try a hard drive recovery it
> the worst happens given its such a large drive it will take several days
> just to scan it before I can even do anything interesting.  No charge.

Hi Rich,

Thanks very much for the offer, but I'm not sure that I've reached the
point where I have to resort to extracting individual files with a hex
editor.

I'm using Linux software RAID so I don't think the controller should
matter, but I'll fish around a bit to find out which SATA controllers
I'm using.

Although I freely admit my ignorance of RAID, my understanding is that,
providing that only one partition has failed and providing that I can
get the spare in play, I should be able to rebuild the RAID 10 array
with that spare ... using mdadm.  N.B. I still have a root shell on the
machine.

I'm fairly sure that I tested such a procedure when I built the arrays
in the first place, but I've forgotten the details and can't refind the
relevant bits of documentation.

In any case, I'd still like to take my own copies of the relevant
partitions and superblocks, hence my original questions about where the
RAID metadata is stored.

BTW, I now have 5x spare 1TB HDDs, so backing up everything should be
possible, providing I can find out exactly what I need to back up.

Dave