--ibTvN161/egqYuK8
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Nov 14, 2006 at 05:16:52PM -0500, Blair Sadewitz wrote:
> This will not help in my case--at least I don't think--because sd4
> only appears if the flash drive is plugged into the USB port.  If it's
> not, there is no device to pin down.  If the device is plugged in, a

Oh, but it is.  That pinned down device is reserved just for what
it matches.

> wedge (dk device) is configured for it.  My flash card reader gets
> probed as sd0-sd3, so if I insert cards in there, those get wedges,
> too.  All of this autoconfiguration messes up the wedge assigments to
> my fixed disks, which borks their assignment in fstab; and then I
> can't boot to multiuser mode!

You need to pin down those essential drives.  Unless the wedge code
runs in undefined order, this should allow for predictable results.

>=20
> IMHO, this is significant concern re: the wedge implementation, unless
> in the future things will be done differently. If there's something
> I'm missing here, I request enlightenment/admonition.

On one of my machines where I need this, I have virtually
all possible drives pinned into place.  Again, assuming
the wedge discovery code scans disks in order, this will
allow for repeatable scan results.

Of course, wedges will be created in the order the drives
are scanned, this would be problematic if sd(4)s are scanned
before wd(4)s.  Perhaps some config(8) machinery is needed
to allow for groups of dk(4)s to be allocated to specific disks.

	Jonathan Kollasch

--ibTvN161/egqYuK8
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (NetBSD)

iD8DBQFFWlUAOjx1ye3hmokRAnv1AJkBR2aylFTE1z+J1y9/AEajGA8xfwCghCm3
Nqb1ppvA7ca7yS0dMAY8BOY=
=Gw9Y
-----END PGP SIGNATURE-----

--ibTvN161/egqYuK8--