There is a patch in ubuntu addressing this. It tries to unmount
repeatedly. I don't know how well that can work. Can't a mounted
filesystem be prevented from unmounting at all due to the desktop's file
manager/etc?

http://patches.ubuntu.com/o/os-prober/os-prober_1.23ubuntu1.patch

--
see shy jo

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFIQdird8HHehbQuO8RAqFcAKC6n0RqrVOsExToxjkd98TqJVep7wCeLL3K
W/0oHJvxe7aWxn8Dabo8TPk=
=pbrC
-----END PGP SIGNATURE-----

The branch, beta2 has been created
at 0a591478ee3e21843b08b682d98d0e90c54836cd (commit)

- Shortlog ------------------------------------------------------------
commit 0a591478ee3e21843b08b682d98d0e90c54836cd
Author: Joey Hess kodama.kitenet.net>
Date: Sat May 31 18:32:02 2008 -0400

brought in the 2.74.2 changes from the uploaded tarball

-----------------------------------------------------------------------

The branch, master has been updated
via ded4da274ea1205a01351268325dc2ccef248bc2 (commit)
from e805ffe705d2c00cafb1169f1422c3f47351fc2d (commit)

- Shortlog ------------------------------------------------------------
ded4da2 Tasksel has moved to a git repository.

Summary of changes:
debian/changelog | 2 ++
debian/control | 2 +-
2 files changed, 3 insertions(+), 1 deletions(-)
-----------------------------------------------------------------------
Details of changes:

commit ded4da274ea1205a01351268325dc2ccef248bc2
Author: Joey Hess kodama.kitenet.net>
Date: Sat May 31 18:26:49 2008 -0400

Tasksel has moved to a git repository.

Add Vcs-Git field.

The branch, newtasksel has been created
at a60f5318b51b6cd085488a06ec5292bb9566d8f2 (commit)

- Shortlog ------------------------------------------------------------
commit a60f5318b51b6cd085488a06ec5292bb9566d8f2
Author: George Papamichelakis users.alioth.debian.org>
Date: Fri Jun 25 07:05:36 2004 +0000

new file

commit 65e877020a717b4fafc02b47bed3551d939187ac
Author: Joey Hess debian.org>
Date: Wed Jun 23 14:49:28 2004 +0000

XSI:ism removal

commit 7c100653dcfc65b7447c279141bd425c6feeace6
Author: Lior Kaplan users.alioth.debian.org>
Date: Tue Jun 22 20:16:15 2004 +0000

finished translation for hebrew

commit a1886bd4b9748b68e43cdf1cf2444e44a266aa91
Author: Peter Mann users.alioth.debian.org>
Date: Tue Jun 22 14:35:10 2004 +0000

New and updated Slovak translation (sk.po)

The branch, master has been updated
via e805ffe705d2c00cafb1169f1422c3f47351fc2d (commit)
from f35c74a1a4d19ddc8786a367fb7d33faa0be7204 (commit)

- Shortlog ------------------------------------------------------------
e805ffe Ignore .git directory in places that ignore .svn.

Summary of changes:
debian/changelog | 1 +
doincludes.pl | 2 +-
listpackages.pl | 2 +-
makedesc.pl | 2 +-
makeoverride.pl | 2 +-
5 files changed, 5 insertions(+), 4 deletions(-)
-----------------------------------------------------------------------
Details of changes:

commit e805ffe705d2c00cafb1169f1422c3f47351fc2d
Author: Joey Hess kodama.kitenet.net>
Date: Sat May 31 17:46:09 2008 -0400

Ignore .git directory in places that ignore .svn.

The branch, cdd has been created
at 3ccd3d23af3db51e3e10c2d3b16d82e4159abbbb (commit)

- Shortlog ------------------------------------------------------------
commit 3ccd3d23af3db51e3e10c2d3b16d82e4159abbbb
Author: Otavio Salvador debian.org>
Date: Sun Apr 24 23:26:09 2005 +0000

Update the changelog and version because of last merge.

commit 1c0ca6f5dd86fd2601ed6bab16865ec1144df79c
Author: Otavio Salvador debian.org>
Date: Sun Apr 24 23:25:01 2005 +0000

Merge all previous changes from trunk.

commit 936dfc60740f5609c561832b17d8b445fd184ee9
Author: Otavio Salvador debian.org>
Date: Tue Jan 25 21:37:37 2005 +0000

[ This version should not be merged back to trunk before sarge
release. Use this branch to CDD related development. ]

* NOT RELEASED YET.

The branch, lenny has been created
at 23d717d813d059c50c2c76a0e02e3e0ba83f4a72 (commit)

- Shortlog ------------------------------------------------------------
commit 23d717d813d059c50c2c76a0e02e3e0ba83f4a72
Author: Joey Hess debian.org>
Date: Mon Dec 11 21:09:22 2006 +0000

create a temporary branch for post-etch development

this will be merged back to trunk later, and updated translations and task
changes in trunk will prevail; new code changes can temporarily go here

-----------------------------------------------------------------------

Processing commands for control@bugs.debian.org:

Bug#136455: [vim-runtime] wrong Perl here-document highlighting
Changed Bug submitter from Marcin Owsiany debian.org> to "Dmitry E. Oboukhov" debian.org>.

Bug#364941: mutt: wish less strict RFC 2047 header decoding
Changed Bug submitter from "Dmitry E. Oboukhov" avanto.org> to "Dmitry E. Oboukhov" debian.org>.

Bug number 379931 not found. (Is it archived?)

Bug number 379939 not found. (Is it archived?)

Bug number 383639 not found. (Is it archived?)

Bug#398126: fontconfig: root umask depended package script
Changed Bug submitter from dimka@avanto.org to "Dmitry E. Oboukhov" debian.org>.
(By the way, that Bug is currently marked as done.)

Bug number 399058 not found. (Is it archived?)

Hi,

While looking at the d-i files on a mirror (sid, current on
ftp.de.d.o), I found this in the ia64 MANIFEST file:

cdrom/boot.img -- CDROM image for Itanium systems
cdrom/initrd.gz -- ./tmp/cdrom/udeb.list
cdrom/vmlinuz --
[...]

ia64/cdrom.cfg only has a MANIFEST-BOOT entry, looks like it needs
MANIFEST-INITRD and MANIFEST-KERNEL entries also.

JB.

--
Julien BLACHE debian.org> | Debian, because code matters more
Debian & GNU/Linux Developer | <http://www.debian.org>
Public key available on <http://www.jblache.org> - KeyID: F5D6 5169
GPG Fingerprint : 935A 79F1 C8B3 3521 FD62 7CC7 CD61 4FD7 F5D6 5169

Package: installation-reports
Severity: important
Boot method: CD
Image version: debian-testing-i386-netinst.iso, 5/30/2008

Subject: order of IDE channels interchanged

I tested installing the debian-testing-i386-netinst.iso (5/30/2008)
on a double processor (Celeron 433) machine that uses two different
IDE chipsets. Installation itself worked well but first rebooting
failed after decompression of the kernel. The system hangs a few minutes
after printing "Loading, please wait..." and aborts with a message
that the root partition /dev/hde3 can not be found. Then an
ash (initramfs) is invoked, so I could type a "dmesg" to see what
happened. Obviously, the order of the IDE channels is interchanged as
shown in the following table: