Date: Mon, 17 Jan 2005 10:42:12 -0800 From: Brooks Davis <brooks@one-eyed-alien.net> To: Shaun Jurrens <shaun@shamz.net> Cc: freebsd-geom@freebsd.org Subject: Re: partitioning and labeling geom devices Message-ID: <20050117184212.GB7538@odin.ac.hmc.edu> In-Reply-To: <20050117161647.GD75478@atreides.freenix.no> References: <20050111110012.GB75478@atreides.freenix.no> <20050111161324.GA8528@odin.ac.hmc.edu> <20050117161647.GD75478@atreides.freenix.no>
next in thread | previous in thread | raw e-mail | index | archive | help
--Fba/0zbH8Xs+Fj9o
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable
On Mon, Jan 17, 2005 at 05:16:47PM +0100, Shaun Jurrens wrote:
> On Tue, Jan 11, 2005 at 08:13:24AM -0800, Brooks Davis wrote:
> #> On Tue, Jan 11, 2005 at 12:00:12PM +0100, Shaun Jurrens wrote:
> #> > Now I have /dev/stripe/stripe0 and i want to subdivide this into two=
slices
> #> > (aka. DOS partitions) and then partition the slices into various Fre=
eBSD
> #> > partitions, (so the a-h thingies, for those catching up). This is s=
ort of
> #> > where the concept goes to hell.
> #> >=20
> [....]
> #>=20
> #> > Do you want to change our idea of what BIOS thinks ?
> #> >=20
> #> > (I have no clue what to answer here, so 'no' is the answer because I
> #> > couldn't tell you the acceptable values for this if you beat me sens=
eless)
> #>=20
> #> You probalby need to answer yes yere. You don't care what the BIOS
> #> might think, you care what freebsd thinks.
>=20
> If I answer yes, it requires that I supply the parameters myself. I
> already wrote that I do not know what acceptable parameters would be.
> #>=20
> #> > Strangely enough the values from previous fdisk runs are still there=
when I
> #> > continue... I've 'cleared' the plexes and stripes twice now... And i=
f the
> #> > partition/slice confusion isn't complete enough, fdisk tells you abo=
ut
> #> > partitions when the rest of your life in FreeBSD, you'll call them
> #> > 'slices'...
> #> >=20
> [...]
> #> >=20
> #> > with a date from a week ago! Really, the box is sync'd with ntpd... =
There's
> #> > probably an explanation somewhere.
> #>=20
> #> Times in /dev are irrelevent, ignore them. I suspect the problem is
> #> that you have an entierly bogus MBR partion table on the disk and
> #> now you've modified it, possiably exposing a bogus bsdlabel. Before
> #> attempting to label a disk, it's generally best to do something like:
> #>=20
> #> dd if=3D/dev/zero of=3D/dev/<disk> count=3D64
> #>=20
> #> That way you've nuked the MBR partition table and any bsdlabel labels =
in
> #> the first slice.
> #>=20
> #> You might take a look at the sysutils/diskprep port. It's a config fi=
le
> #> driven disk labeler.
>=20
> I gave diskprep a go... I still have not gotten the desired result...
>=20
> paracles:/root#> diskprep -config diskprep.array.cf stripe/stripe0
> =3D=3D=3D> Initalizing slices
> ------> Hog slice is 2.
> ******* Working on device /dev/stripe/stripe0 *******
> fdisk: WARNING: adjusting size of partition 1 from 10485760 to 10474317
> to end on a cylinder boundary
> fdisk: WARNING: adjusting start offset of partition 2
> from 10485823 to 10485846, to fall on a head boundary
> fdisk: WARNING: adjusting size of partition 2 from 42576872 to 42576849
> to end on a cylinder boundary
> =3D=3D=3D> Getting adjusting sizes
> ------> slice 1 is 10474317 sectors
> ------> slice 2 is 42576849 sectors
> =3D=3D=3D> initializing slice 1
> ------> Hog partition is 'a'
> Argument "*" isn't numeric in subtraction (-) at /usr/local/sbin/diskprep
> line 232.
> disklabel: /dev/stripe/stripe0s1: No such file or directory
> =3D=3D=3D> initializing slice 2
> ------> Hog partition is 'c'
> disklabel: /dev/stripe/stripe0s2: No such file or directory
> newfs: /dev/stripe/stripe0s1a: could not find special device
> newfs: /dev/stripe/stripe0s1d: could not find special device
> newfs: /dev/stripe/stripe0s1e: could not find special device
>=20
>=20
> The config file is simple enough:
>=20
> $slices{1}{type} =3D $fbsd_type;
> $slices{1}{size} =3D 5120*1024*2;
> $slices{1}{'d'}{size} =3D 768*1024*2;
> $slices{1}{'d'}{type} =3D "4.2BSD";
> $slices{1}{'e'}{size} =3D "*";
> $slices{1}{'e'}{type} =3D "4.2BSD";
>=20
> #$hog_slice =3D 2;
> $slices{2}{size} =3D "*";
> $slices{2}{type} =3D $fbsd_type;
> $slices{2}{'c'}{type} =3D "4.2BSD";
> $slices{2}{'c'}{size} =3D "*";
>=20
> =09
> So I could still use some suggestions... I'm not getting any wiser
> here. I have no "slice" devices or partitions. Is this devd's fault
> or geom's ?
I'd guess from the errors that diskprep has a bug in the case that your
device is in a directory in /dev, but there could be a config bug. I
don't have time or resouces to figure out which. Rather then trying
your rather complicated config first, try a minimal config and see if
you can get that working. You can also read the source of diskprep and
try to do what it is attempting to do by hand to see where things work
and where things don't.
-- Brooks
--=20
Any statement of the form "X is the one, true Y" is FALSE.
PGP fingerprint 655D 519C 26A7 82E7 2529 9BF0 5D8E 8BE9 F238 1AD4
--Fba/0zbH8Xs+Fj9o
Content-Type: application/pgp-signature
Content-Disposition: inline
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.1 (GNU/Linux)
iD8DBQFB7AcEXY6L6fI4GtQRAtdqAKCetsqlKNx+F9buWfUvdSZlSLqmAACgzJOz
jRJeYyrP0h1xudV6U6khiL0=
=LmcH
-----END PGP SIGNATURE-----
--Fba/0zbH8Xs+Fj9o--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050117184212.GB7538>