Re: [2.6.20.11] File system corruption with degraded md/RAID-5 device

[Neil Brown]

On Wednesday May 16, koan00@xxxxxxxxx wrote:
> 
> Here is what I am doing to test:
> 
> fdisk /dev/sda1 and /dev/sb1 to type fd/Linux raid auto
> mdadm --create /dev/md1 -c 128 -l 5 -n 3 /dev/sda1 /dev/sdb1 missing
> mke2fs -j -b 4096 -R stride=32 /dev/md1
> e2fsck -f /dev/md1
> ---------------------
> Result: FAILS - fsck errors (Example: "Inode 3930855 is in use, but
> has dtime set.")

Very odd.  I cannot reproduce this, but then my drives are somewhat
smaller than yours (though I'm not sure how that could be
significant).

Can you try a raid0 across 2 drives?  That would be more like the
raid5 layout than raid1.

My guess is some subtle hardware problem,  as I would be very
surprised in the raid5 code is causing this.  Maybe run memtest86? 

NeilBrown
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/