Re: Strange problem with Device Mapper

From: Andrew Morton
Date: Fri Apr 27 2007 - 00:05:59 EST


On Thu, 26 Apr 2007 19:12:27 +0200 "Paul Rolland" <rol@xxxxxxxxxx> wrote:

> Hello,
>
> I've a machine that has been installed with Intel Soft Raid on top
> of 2 SATA disks.
> I'm trying to have this work as a RAID-1 array.
> Bios configuration has been done, using 128K chunk, and the kernel
> (2.6.20.7) sees perfectly /dev/mapper/isw_xxxxxxxx_RAID1
>
> But, I'm facing two problems :
> 1 - If i try to create partitions on this device, it does fail (the
> values are not interpreted correctly)
>
> 2 - To avoid 1), I stop the RAID array (dmraid -an), then I do create
> exactly the same partition set on /dev/sda, and /dev/sdb, and
> then I reactivate RAID (dmraid -ay).
> This allows me to see all the /dev/mapper/isw_xxxxxxx_RAID1p1, p2, ...
> But, running fsck -t ext2 /dev/mapper/isw_xxxxxxx_RAID1p1
> ends in a lock when the partition is larger than 10Go (well, it is
> Ok on the 10Go one, and it locks on the 100Go).
> Not a real hardlock, I still can switch to a new VC, but it's not
> more possible to start a new command or to stop fsck.
>
> Should I do everything on the physical disks before activating RAID ?
> Is it a normal behavior ?
>
> BTW, I also noticed that such a device doesn't support BLKRRPART ioctl...
> So it is possible that what I'm doing is wrong...
>
> Any idea ?
>

cc added.
-
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/