Re: devfs persistence

From: Richard Gooch (rgooch@ras.ucalgary.ca)
Date: Wed May 03 2000 - 16:42:35 EST


Stephen C. Tweedie writes:
> Hi,
>
> On Wed, May 03, 2000 at 12:31:20PM +0200, Andries Brouwer wrote:
> > On Wed, May 03, 2000 at 11:15:49AM +0100, Stephen C. Tweedie wrote:
> >
> > > The only advantage you'd get by adding extra linux-only partitioning
> > > would be to let you subdivide the DOS-type partitions even further
> >
> > No. A DOS type partition table tells you where the partitions are
> > and has a partition type but no room for UUIDs, volume labels
> > and the like.
>
> You quoted me completely out of context! The only advantage of
> linux-only partitioning *over a raid-style superblock* is that
> a linux partition table would be able to describe multiple devices
> at once, whereas a raid md0 superblock only describes a single
> logical device. LVM lets you provide UUIDs for multiple logical
> devices in a single partition.
>
> The LVM superblock is exactly what you are after, isn't it? You
> place one of those on a DOS partition, and then you can carve off
> fixed-name portions by allocating physical extents from the LVM pool
> to specific logical volumes. We don't need to discard the existing,
> portable partitioning mechanisms in order to achieve this.

But you're really just moving the problem, since now you have to
configure LVM/MD to say where the real devices are. And they can shift
around (which is where this dicussion started).

If you knew that "the second primary DOS partition always contains
LVM", it would work. But you can't really impose that.

                                Regards,

                                        Richard....
Permanent: rgooch@atnf.csiro.au
Current: rgooch@ras.ucalgary.ca

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



This archive was generated by hypermail 2b29 : Sun May 07 2000 - 21:00:13 EST