[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Linux-NTFS-Dev] Windows Dynamic Disks, Parted
From: |
Anton Altaparmakov |
Subject: |
Re: [Linux-NTFS-Dev] Windows Dynamic Disks, Parted |
Date: |
Sun, 5 Sep 2004 06:52:58 +0100 (BST) |
On Sun, 5 Sep 2004, Andrew Clausen wrote:
> Hi Szaka,
>
> On Sat, Sep 04, 2004 at 12:50:22PM +0200, Szakacsits Szabolcs wrote:
> > > If Parted sees that a partition table is a LDM disk (i.e. has a
> > > partition of type 0x42), should it simply refuse to recognize it? i.e.
> > > say that it isn't a partition table.
> >
> > Well, in the above email I wrote,
> >
> > "Partitioners must detect if one has a dynamic disk (it was designed to
> > be very easy to detect) and refuse to progress or implement dynamic disk
> > resizing."
> >
> > But I've meant it in its context, namely non-destructive NTFS resizing.
> >
> > General support is much more complex. You also have to think about cases
> > like:
> >
> > - user wants to get rid of windows dynamic partitions
> >
> > - user wants to fix spoiled windows dynamic partition entries
>
> Can you give a practical example of this?
Made up example: "I just ran fdisk in Linux and now Windows doesn't
boot. Help."
> > - user wants to edit non-dynamic disk partition entries (I've thought
> > it was impossible mixing windows dynamic partitons and basic
> > partitons but I've seen them working together)
Yes, this is possible. I have seen it in action.
> > - perhaps other scenarios
>
> Shouldn't all of this stuff be provided by a Windows Dynamic Disk program
> rather than Parted? Therefore, shouldn't Parted always refuse to touch
> it?
Perhaps. OTOH parted could become such a program. There is no LDM editor
under Linux yet (mostly due to lack of requests for it).
> The counter-argument is that the Windows Dynamic Disk program might
> want to use libparted to handle the partition table stuff. Is this
> likely? This is rather hypothetical though, since the linux-ntfs
> projects' LDM program is still rather immature (is that right?).
Well it only does "fdisk -l" equivalent for LDM and does no editing. But
it is fully stable doing that. (-:
> In any case, I don't have access to Windows XP Professional, and I
> can't test this stuff out myself. I am comfortable with doing the
> following things immediately:
>
> * option 1: simply refuse to touch them.
>
> * option 2: make a new partition table type called dynamic-disk,
> and make it read-only. (It would share the disk_dos.c code).
>
> * option 3: make a new partition table type called dynamic-disk,
> and make it only permit modification of non-0x42 partitions.
This is probably the best one. Perhaps add a further restriction: Do not
allow to add partitions to a dynamic disk. They would conflict with the
LDM anyways...
Best regards,
Anton
--
Anton Altaparmakov <aia21 at cam.ac.uk> (replace at with @)
Unix Support, Computing Service, University of Cambridge, CB2 3QH, UK
Linux NTFS maintainer / IRC: #ntfs on irc.freenode.net
WWW: http://linux-ntfs.sf.net/ & http://www-stu.christs.cam.ac.uk/~aia21/