[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#106201: marked as done (Clearer instructions for driver-# floppies)



Your message dated 25 Sep 2001 15:54:35 -0400
with message-id <oavgi7avwk.fsf@arroz.onshored.com>
and subject line bugs closed in boot-floppies 3.0.14
has caused the attached Bug report to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what I am
talking about this indicates a serious mail system misconfiguration
somewhere.  Please contact me immediately.)

Darren Benham
(administrator, Debian Bugs database)

--------------------------------------
Received: (at submit) by bugs.debian.org; 22 Jul 2001 14:28:57 +0000
>From bounce@foobar.clara.co.uk Sun Jul 22 09:28:57 2001
Return-path: <bounce@foobar.clara.co.uk>
Received: from oracle.clara.net [::ffff:195.8.69.94] 
	by master.debian.org with esmtp (Exim 3.12 1 (Debian))
	id 15OKE5-0003to-00; Sun, 22 Jul 2001 09:28:57 -0500
Received: from [212.126.132.26] (helo=-)
	by oracle.clara.net with smtp (Exim 3.11 #5)
	id 15OKDw-0001Uc-00; Sun, 22 Jul 2001 15:28:49 +0100
Date: Sun, 22 Jul 2001 15:28:30 +0100
From: B Boudet <bboudet@foobar.clara.co.uk>
To: submit@bugs.debian.org
Subject: Clearer instructions for driver-# floppies
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
X-Mailer: Becky! ver. 2.00.01
Message-Id: <E15OKDw-0001Uc-00@oracle.clara.net>
Delivered-To: submit@bugs.debian.org

Package: boot-floppies
Version: 2.2.23

This is a bug report against the Debian Install Manual.  At the end of
this message is a suggested fix.

The manual is lacking in a description of how to deal with installs
where the hd controller is not recognised on the initial boot and
therefore the hd cannot be partitioned.  

This can be either because the rescue floppy does not have the necessary
driver compiled in, or the user failed to enter correct boot parameters.
However, a new user can be unaware this is the cause and may instead
believe the driver-# floppies are required in order to load the driver
as a module.

It's not clear to a new user whether the setup process should prompt for
the driver floppies automatically, or whether the user needs to do
something to make the setup process prompt for the floppies.  So the
user may wonder if the problem is due to user error or a bug.

After the hd is partitioned, it becomes clear, but this is no help if
the hd is not accessible due to driver problems.  

I'm reporting this because I had trouble with my first floppy install of
Debian, for which I could not find the answer in the install manual.  My
search of the debian-user list for recent months only found similar
confusion about how to make use of driver-# floppies:

http://lists.debian.org/debian-user-0106/msg02662.html
http://lists.debian.org/debian-user-0106/msg02685.html

I think bug #96237 is related, but I'm not sure, which is why I opened a
new bug.  The response to that report was to RTM.  Well I *have* RTFM
and the problem stands.

Below is a modification to the Install Manual text that I know would
have helped me, and I think would help other new users too:

5.4.3. Driver Files
-------------------

     These files contain kernel modules, or drivers, for all kinds of
     hardware that are not necessary for initial booting.  Getting the
     drivers you want is a two step process: first you identify an archive
     of drivers you want to use, and then you select which particular
     drivers you want.
     
     The driver archive floppies are not used until after the hard drive
     has been partitioned and the kernel has been installed.  If you need a
     particular driver for initial booting, or to access the hard drive,
     choose a kernel with the necessary driver compiled in and supply the
     correct boot parameter arguments.  Please see Section 5.2. `Choosing
     the Right Installation Set' and Section 6.1, `Boot Parameter
     Arguments'.

     Remember that your driver archive must be consistent with your initial
     kernel choice.

Thanks,

-Bernie Boudet.


---------------------------------------
Received: (at 106201-done) by bugs.debian.org; 25 Sep 2001 19:40:32 +0000
>From adam@onshore-devel.com Tue Sep 25 14:40:32 2001
Return-path: <adam@onshore-devel.com>
Received: from arroz.onshored.com [216.220.101.2] (postfix)
	by master.debian.org with esmtp (Exim 3.12 1 (Debian))
	id 15ly4F-0007n8-00; Tue, 25 Sep 2001 14:40:31 -0500
Received: from arroz.onshored.com (localhost [127.0.0.1])
	by arroz.onshored.com (Postfix) with ESMTP
	id 8FE3993808; Tue, 25 Sep 2001 15:54:35 -0400 (EDT)
Sender: adam@arroz.onshored.com
To: 106201-done@bugs.debian.org, 107598-done@bugs.debian.org,
	108103-done@bugs.debian.org, 109925-done@bugs.debian.org,
	110255-done@bugs.debian.org, 110710-done@bugs.debian.org,
	112569-done@bugs.debian.org, 84388-done@bugs.debian.org, 
Subject: bugs closed in boot-floppies 3.0.14
Organization: onShore Development, Inc
From: Adam Di Carlo <adam@onshore-devel.com>
Date: 25 Sep 2001 15:54:35 -0400
Message-ID: <oavgi7avwk.fsf@arroz.onshored.com>
Lines: 107
User-Agent: Gnus/5.0808 (Gnus v5.8.8) Emacs/20.7
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
Delivered-To: 106201-done@bugs.debian.org


We believe the problem you reported has been closed in boot-flopppies
3.0.14.  The changelog is included below.  We invite you to test that
new version.  If your problem is still occuring, we can reopen the
bug, or you can file a new one.

--=20
...Adam Di Carlo..<adam@onshore-devel.com>...<URL:http://www.onshored.com/>

 boot-floppies (3.0.14) unstable; urgency=3Dlow
 .
   * Stefan Gybas
     - use kernel 2.4.7 on s390
     - tweaked contents of s390 initrd
     - ignore devfs partitons in libfdisk
   * Gerhard Tonn
     - add support for IBM partition table on s390
     - add support for rescue disk
     - .tar-place-holder file added
     - use kernel/modules 2.4.7 on s390
     - special inittab.install for s390 added
   * David Kimdon
     - add clearer instructions for driver-# floppies from
       Bernie Boudet <bboudet@foobar.clara.co.uk>
       closes: #106201
     - 2 Danish translation updates thanks to
       Claus Hindsgaul <claus_h@image.dk>
     - update url of PC Hardware FAQ; closes: #110255
     - make tftpboot images for i386; closes: #108103
     - fix missing fdisk.txt; closes: #107598
     - document 'Report a Problem' menu choice
     - Policy: do debug mode if DEB_BUILD_OPTIONS says so
   * Phil Blundell
     - use glibc 2.2.4 for arm
     - bump RiscPC root disks to 4200 blocks, and netwinder to 4000,
       so modules and i18n support will fit
     - include adfs.o and floppy.o on arm root disks; actually try to
       load the SCSI modules added back in 3.0.8
     - detect revision J/K SA-110 chips; if found, advise the user
       that trouble may lie ahead
     - try to stop `images-2.88' popping up in the arm files list
     - arm 2.4.5 kernel doesn't seem too stable; drop back to 2.2.19
       for netwinder and riscpc (shark is only supported in 2.4)
     - try to make arm builds work with USE_LANGUAGE_CHOOSER
   * Eduard Bloch:
     - reformated German boot message to fit on 80x25
     - German translation updates
     - started building udma100-ext3 flavor on i386, ext3 autodetection
     - user query about ReiserFS v2, fs selection fixes
     - updated copyright year in the f10.txt files; closes: #110710
     - integrated testing parted option on i386_udma-100 flavor.
   * Christian T. Steigies
     - m68k: add arch specific posttask to make amiboot and dmesg executable
   * Ethan Benson
     - Make build fail if rootimage is larger then the floppy image.
     - Redesign filesystem creation/mounting to cleanly, and dynamically
       support multiple filesystems, redo the ext3/reiserfs code.
       Now dbootstrap will offer a menubox list of supported filesystems.
       A filesystem is offered as a choice when both a) the running kernel
       supports it, and b) there is a mkfs for that filesystem.  If only Ex=
t2
       is supported the behavior is exactly as its always been.
     - Add is_filesystem_supported() function to util.c.
     - Add XFS filesystem support.
     - Fix partition mounting so it works properly again.
     - Remove bogus and non-existent reiserfs partition type, reiserfs is
       not, and should not get its own partition type, it should use the
       "Linux Native" partition types that already exist on most/all
       architectures.  The redesigned mounting code does this.
     - Redo fstab generation to work properly with multiple filesystems.
     - Moved is_fstype() function out of baseconfig.c and into util.c.  Its
       useful elsewhere (now also used in bootconfig.c).
     - Quik setup aborts if root partition is not ext2.
     - Eliminate ioctl messages that were spewed all over the screen
       during kernel installation.  (perror() -> ERRMSG() don't use
       perror() in boot-floppies!)
     - Check for parted correctly.
     - Remove ddisk and cfdisk from pmac root floppy, we don't need them and
       there is not sufficient space.
   * Richard Hirst
     - redirect info output from downloadpkg() to stderr, so offline_mode=
=3Dtrue
       works again
     - ia64 now uses kernel 2.4.9 and libc 2.2.4
     - hppa now uses kernel 2.4.9
     - ia64 now uses compressed kernels
     - hppa: move vmlinux symlink in to /boot for palo
     - ia64 and hppa: add a kernel-choice section to docs
     - ia64 now uses mckinley-smp kernel
     - ia64: add efi/boot/bootia64.efi to rescue.bin
     - hppa: add warning about 2GB limit for /boot to PALO screen.
   * Adam Di Carlo:
     - only include reiserprogs stuff on the reiserfs flavor; this may help
       with the 1.2MB images
     - fix minor SGML build issues
     - integrate and rework patch from Samy PERSEHAIE, now you can set
       KBD in config when you build, and that still will be suppressed
       during installation; obviously, that is turned off by default
     - also from Samy PERSEHAIE, French listed as a Latin-1 language
     - similarly, R=FCdiger Kuhlmann pointed out German should be there too;
       I'm sure there are many more, I wish someone would just do a full
       pass over that...; closes: #112569
     - synchronize, for all architectures, the keymaps we ship and the
       keymaps used by dbootstrap; closes: #84388, #109925
   * Chris Tillman
     - numerous documentation fixes and updates
   * Marcin Owsiany
     - various language chooser fixes; this is reputed to work now,
       although the on-screen fonts are scrambled



Reply to: