Bug#60180: marked as done (Should detect previous installation attempt and explicitly ask for really installing over)
Your message dated 23 Apr 2000 15:48:14 -0400
with message-id <oapurgd2f5.fsf@arroz.fake>
and subject line bugs closed in boot-floppies 2.2.11
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; 11 Mar 2000 05:54:33 +0000
Received: (qmail 17197 invoked from network); 11 Mar 2000 05:54:33 -0000
Received: from gate.westel900.hu (194.176.224.33)
by master.debian.org with SMTP; 11 Mar 2000 05:54:33 -0000
Date: 11 Mar 2000 06:52:23 +0100
From: "János Holányi" <HOLANYIJ@westel900.hu>
To: submit@bugs.debian.org
Subject: Should detect previous installation attempt and explicitly ask for really installing over
X-Orcl-Application: InterOffice Version4.1.2.22.20
MIME-Version: 1.0
X-Mailer: Oracle InterOffice (version 4.1.1.3.40)
Content-Type: text/plain; charset="iso-8859-2"
Content-Transfer-Encoding: 8bit
Message-ID:<12TerH-0002ng-00@gate.westel900.hu>
Package: boot-floppies
Version: 2.2.8
Severity: critical
Currently, if I install the drivers again from the driver floppies without
reformatting the target partition, ("Installing Operating System and
Modules" in the menu) then the next choice after loading and installing from
drivers.tgz is "Make Linux Bootable", so I don't even get offered the chance
to reconfigure the Device Driver Modules.
The system should explicitly ask me if I want to reconfigure the modules
(something like "Previous driver configuration was found. Do you want to
reconfigure?), and then "Configure Network" and so on...
Moreover, if I manually choose Configure Device Driver Modules from the
menu, strange things happen:
1. right after pressing Enter on "Configure Device...", a message like
Mar 11 04:49:09 (none) user.info dbootstrap[30]: depmod: error reading ELF
header: No such file or directory
appears on tty3;
2. At the same time, on tty1 I get two error messages like:
/target/bin/sed: error in loading shared libraries: /target/bin/sed:
undefined symbol: _IO_stdout_
3. After I get the "Select Category" and I choose any category, I get the
same errors ("/target/bin/sed: error in loading shared libraries:
/target/bin/sed: undefined symbol: _IO_stdout_") uncountable many times, and
the category sub-menu contains lines like:
"- (No description available)"
Regards,
csani@lme.linux.hu
---------------------------------------
Received: (at 60180-done) by bugs.debian.org; 23 Apr 2000 19:48:52 +0000
Received: (qmail 26124 invoked from network); 23 Apr 2000 19:48:49 -0000
Received: from amsterdam.interport.net (199.184.165.19)
by master.debian.org with SMTP; 23 Apr 2000 19:48:49 -0000
Received: from arroz.fake (postfix@209-122-226-214.s468.tnt1.nyw.ny.dialup.rcn.com [209.122.226.214])
by amsterdam.interport.net (8.8.5/8.8.5) with ESMTP id PAA12814;
Sun, 23 Apr 2000 15:48:21 -0400 (EDT)
Received: by arroz.fake (Postfix, from userid 421)
id E8EBD93802; Sun, 23 Apr 2000 15:48:14 -0400 (EDT)
Sender: apharris@arroz.fake
To: 60180-done@bugs.debian.org, 57686-done@bugs.debian.org,
61926-done@bugs.debian.org, 58989-done@bugs.debian.org,
54660-done@bugs.debian.org, 58563-done@bugs.debian.org,
58779-done@bugs.debian.org, 60182-done@bugs.debian.org,
61679-done@bugs.debian.org, 61861-done@bugs.debian.org,
62333-done@bugs.debian.org
Subject: bugs closed in boot-floppies 2.2.11
From: Adam Di Carlo <adam@onshore.com>
Date: 23 Apr 2000 15:48:14 -0400
Message-ID: <oapurgd2f5.fsf@arroz.fake>
Lines: 116
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: quoted-printable
Thank you for your bug report. This bug has been fixed in
boot-floppies 2.2.11, which is not part of the frozen or 'potato'
distribution of Debian. The changelog for this version has been
included.
--=20
.....Adam Di Carlo....adam@onShore.com.....<URL:http://www.onShore.com/>
boot-floppies (2.2.11) frozen unstable; urgency=3Dlow
* Ben Collins <bcollins@debian.org>
- fdflush.c is_root_a_floppy.c lists.c nfsroot.c no_match.c:
merged into util.c; these were all single function files, no need=
to
have the spread around like that
- compile warning cleanups
- fixed logic for cdrom-install detection
- rootdisk/rcS: perform the ramdisk mount for /tmp for anytime we a=
re
on a readonly root. TODO: make it so the cdrom's don't have to do
anything special. Take care of it all here.
- more changes to support read-only root fs; trying to move stuff f=
rom
debian-cd to here
- library reduction appears to work on sparc now (removed extraneous
-lc from dbootstrap's LIBS and it fixed the problem)
- dbootstrap/util.c: (getif_start)(getif)(getif_end): new functions
that get the actual list of available interfaces, even allow for
filtering to only the configured (up) interfaces. (is_network_up)
returns number of active (up) interfaces. (is_interface_up) takes
the name of an interface as an argument and returns true if the
interface is active (up), 0 if not, and -1 if error (doesn't exis=
t).
- changed most calls from is_network_configured() to is_network_up(=
) so
that we get a real feel of if the network is actually running (as
opposed to just having the config files laying around from a
previous install attempt)
- main_menu.c: in addition to checking is_network_configured() also
check is_network_up() to make sure it is actually running
- final updates to support read-only filesystems. TODO, make this t=
he
default setup, since it works both ways. Just with read-only, we
need to mount a ramdisk on /tmp
- syslogd.c: use realpath() instead of readlink()
- dbootstrap: removed kernel_image_filename and drivers_filename, a=
nd
simply use the #define's throughout (always the same value anyway)
- dbootstrap/bootconfig.c: silo image requires the Arch2 extension
- mount_any():make sure we create all directories up to the mountpo=
int
- Allow the hostname to be configured by itself
- scripts/rescue/install.sh: install config.gz if it exists to
/boot/config-$version (the same as the kernel-image package would)
- rescue.sh, kernel.sh, Makefile: create, install, and dep on confi=
g.gz
for all archs. This creates a config.gz on the rescue floppy
(stripped of blank lines and comments for compactness) so that it
can then be installed on the base system.
* Risko Gergely <risko@njszki.hu>: Hungarian updates
* Ji=F8=ED Ma=B9=EDk <masik@debian.cz>: Czech updates
* Yoshizumi Endo <y-endo@ceres.dti.ne.jp>: Japanese updates
* Adam Di Carlo <aph@debian.org>
- root.bin: add /dev/md* (closes part of Bug#61065);
- root.bin make /dev/sr# symlink to /dev/scd# if the former doesn't=
exist
(closes: Bug#61679)
- dbootstrap: fix minor bug where bmap was called bmp in err msg
- dbootstrap: start more verbose syslogging, i.e., when we mount and
format stuff; consistency of message box titles; fix the "touch: =
too
many levels of symbolic links" problem when configuring drivers
- documentation/Makefile: fix deps for creating docs
- README-Users: some updates, spell fixes, and clarifications
- documentation/*: some unfortunately rather minor updates; expand =
the
flavors documentation; mention that ARM is supported
- documentation/defaults.ent: turn on FIXME for now
* Daniel Jacobowitz <dan@debian.org>
- kbdconfig.c: fix keyboard maps for PowerMac
- add miBoot for oldworld powermac floppy support
- add hfsutils build dependency on powerpc
- always download APUS kernel on powerpc
- fix tarball organization on powerpc
- fix a dbootstrap segfault on powerpc
- tell dbootstrap that /vmlinux (as opposed to /vmlinuz) might be a
successfully installed kernel
- making the hard drive bootable has a long, long way to go - and n=
ewt
is corrupting the display - but otherwise, installation on a powe=
rmac
now succeeds! Needs more documentation... (closes: Bug#58779)
- Enable PowerMac PCMCIA support.
* Erik Andersen <andersen@lineo.com>: busybox fixes (closes: Bug#6233=
3)
* Vincent Renardias <vincent@debian.org>:
- French updates
- dbootstrap typo fixes
- reordered the keymaps
* Van Buggenhaut <wana5703@wanadoo.be>: French updates
* Nick Holgate <holgate@debian.org>:
- VME porting -- fix MVME detection (closes: Bug#54660)
- m68k porting; m68k no longer builds 720k disks
* Marcin Owsiany <porridge@pandora.info.bielsko.pl>: Polish updates
* Marcel Harkema <marcel@debian.org>:
- http-fetch fixes relevant to proxies; virtual hosts now supported
* Randolph Chung <tausq@debian.org>
- fix an interaction with modconf, fixing the "wrong tty" problem
(closes: Bug#60182)
- i386 flavors documentation added
* Christian T. Steigies <cts@debian.org>
- m68k porting: fixes in release.sh and keymaps
* Pedro Guerreiro <digito@bigfoot.com>
- re-enable Portugues keymaps (closes: Bug#58563)
- re-enable Greek keymaps
* Edmund GRIMLEY EVANS <edmundo@rano.org>:
- fixes to 'bogl' stuff
* David Huggins-Daines <dhd@debian.org>:
- Alpha porting:
+ experimental support for installing APB and kernel on a FAT
partition
+ workaround MILO problems by using 1024 byte blocks in mke2fs
+ 'make bootable' should work now
* Ross Boylan <RossBoylan@stanfordalumni.org>
- README-Overview updated (then updated more by Adam)
* Mark van Walraven <markv@wave.co.nz>
- make libfdisk ignore partitions that don't appear in /proc/partit=
ions,
which previously caused assertion 'partition->name' (closes: Bug#=
61861)
-- Daniel Jacobowitz <dan@debian.org> Fri, 21 Apr 2000 11:12:40 -0400
Reply to: