Am Do, den 18.12.2003 schrieb Alastair McKinstry um 12:45:
> >Quoting Joey Hess ([EMAIL PROTECTED]):
> >> Having this ugly debconf priority question right after choosing the
> >> language is annoying. Do we really need it there?
> >
> >I haven't tested a full d-i for a while now, but I think I can
Am Do, den 20.11.2003 schrieb Karsten Merker um 19:31:
> Although Thiemo is not a Debian developer, he was of much help with
> Debian/MIPS issues and the work on d-i. He also took part in the
> Oldenburg developers meeting and worked on porting d-i there.
> I advocate his application for cvs commi
Am Mi, den 19.11.2003 schrieb John Lewis um 13:49:
> Jonathan mentions that by choosing the UK Engkish language you get by
> default a dvorak keyboard layout. Since this must only be correct for
> a minority of users should this not be changed to default to the
> standard UK 102 key qwerty keyb
Am Di, den 18.11.2003 schrieb Geert Stappers um 11:15:
> > Well, udpkg does not even handle versioned Depends... So as the
> > situation is now we do only care about names. Hence my proposal would
> > work. However we should investigate if it would be a good idea to
> > implement version tracking
On Mon, Nov 17, 2003 at 10:27:50PM +, Alastair McKinstry wrote:
> I recommend we insert the question now; if we get bug reports in beta2
> reporting it as an issue, I will change the kbd-chooser priority.
>
> I'm sending this mail to debian-boot, and particularly to Joey Hess, as
> I do not w
Am Mo, den 17.11.2003 schrieb Colin Watson um 13:30:
> > I once proposed that all library udebs should Provide the name of the
> > "original" library. Then we will be able to use shlibs to resolve
> > dependencies also for udebs.
>
> That'll only do you any good if udpkg implements versioned prov
Am So, den 16.11.2003 schrieb Denis Barbier um 22:29:
> The problem with partconf may be solved by RESETting partconf/partitions
> when its value is Abort.
That is not enough: If the user dived deeper into partconf before
Aborting, there are other partconf questions marked as seen. They won't
be
Am Mo, den 17.11.2003 schrieb Goswin Brederlow um 10:45:
> Partconf Build-Depends on "libparted1.6-dev | libparted-dev" and its
> binary packages "Depends: ${shlibs:Depends}". That results in
> "Depends: libc6 (>= 2.3.2.ds1-4), libdebconfclient0,
> libdebian-installer4 (>= 0.16), libparted1.6-0 (>
Am Mo, den 17.11.2003 schrieb Goswin Brederlow um 10:45:
> Partconf Build-Depends on "libparted1.6-dev | libparted-dev" and its
> binary packages "Depends: ${shlibs:Depends}". That results in
> "Depends: libc6 (>= 2.3.2.ds1-4), libdebconfclient0,
> libdebian-installer4 (>= 0.16), libparted1.6-0 (>
Am So, den 16.11.2003 schrieb Steinar H. Gunderson um 19:29:
> Something like "partconf/userspace-packages/ext3 = e2fsprogs", perhaps? This
> would of course have to be a documented hook. :-)
Yes, in case of reiserfs, the template would come with mkreiserfs-udeb.
I neither don't know what provide
Am So, den 16.11.2003 schrieb Steinar H. Gunderson um 18:24:
> For reiserfs we have progsreiserfs and reiserfsprogs (yes, two different
> packages with different maintainers), for xfs we have xfsprogs, for ext2 and
> ext3 we have e2fsprogs, for jfs jfsutils etc.. Should we simply make a table
> of
Am So, den 16.11.2003 schrieb Marco Bascietto um 15:33:
> The installation was not successfull.
> The installation steps untill reboot were ok, the laptop rebooted and
> reentered the installation process. After that it automatically
> rebooted, no matter what it was doing, or I was typing. In f
reassign 220868 cdebconf
thanks
The problem encountered with partconf is cdebconf related. In short this
is what happens:
If you back out to the main-menu from a menu item, cdebconf will set the
seen flags for all the questions it has seen. However, the package will
be stl marked "unconfigured" a
I thought that a
"debconf_fset(debconf, "partconf/partitions", "seen", "false");"
after an Abort would do the trick but that was obviously wrong.
With this line added, I hit Abort for the first time and was thrown to
the main-menu. Good.
Then I tried to execute partconf again, which did not wor
Package: partconf
Severity: important
Tags: d-i
If in prio=high mode and selecting "Abort" in the main menu of
partitioner, it will bring me to the main-menu. However if I decide to
run partconf again later, it seems to use the previous answer "Abort"
automatically and so drops me to the main-menu
Am Sa, den 15.11.2003 schrieb Colin Watson um 01:03:
> According to the Installed-Size: fields, base is more like 86Mb. I'd be
> astonished if it was over 256Mb. Perhaps something else is going on?
coyote:/home/sl# debootstrap sarge test http://www.kawo1.rwth-aachen.de/debian
[...]
coyote:/home/s
Am Fr, den 14.11.2003 schrieb Andre Sodermans um 22:42:
> Everything worked fine, except I ran out of space on
> my harddisk (not your problem, I know), which is a
> 256MB SSD. I was wondering if there is a way I can
> choose packages like in woody so I can install the
> bare minimum.
Debian-Inst
Am Do, den 13.11.2003 schrieb Craig Ruefenacht um 23:38:
> Howdy,
>
> I just tried installing sarge on a Intel P4, using an ISO I downloaded from
> ftp.hsn.hu, dated November 2, 2003, and I ran into a problem.
This is hopelessly outdated, a lot has happened since then. I even do
not know if the i
Package: base-installer
Severity: important
Tags: d-i
Debootstrap is not idempotent, it will fail if told to bootstrap a
partition where already an instalation of linux resides.
Base-installer should be aware of the fact and try to rule out problems
before debootstrap runs.
Idea:
Try to detect i
Am Do, den 13.11.2003 schrieb Doug du Boulay um 01:42:
> Unfortunately at the final stage of the install of the base system
> I got the error:
>
> debootstrap exited with an error (return value 1)
>
> from /target/var/log/debootstrap.log
> I see it dies on :
>
> Setting up base-files (3.0.12
Am Do, den 13.11.2003 schrieb Glenn McGrath um 00:16:
> Im sure there are a lot of users who want things to work automatically,
> but not everyone does, and it makes problems more difficult to fix.
>
> Unless the 2nd question asks the level of questioning the only way the
> user effecting the inst
Am Di, den 11.11.2003 schrieb Joey Hess um 23:37:
> How's this for an idea for dhcp. If priority is high, then run dhcp, but
> in the background, and only for 30 seconds (or some better chosen time
> that is long enough for most dhcp servers to reply, and short enough to
> not annoy people with no
Package: partconf
Severity: important
Tags: d-i
Partconf should make sure that appropriate userspace tools for created
filesystem are installed to /target. Especially fsck.$FILESYSTEM is
needed.
-- System Information:
Debian Release: testing/unstable
Architecture: i386
Kernel: Linux coyote 2.4.22
Am Di, den 11.11.2003 schrieb Joey Hess um 20:17:
> > As for the normal boot:
> > Initial boot worked:[X]
> > Configure network HW: [X]
> > Config network: [ ] I have no DHCP server
> > Detect CD: [ ]
> > Load installer modules: [X]
> > Detect hard drives: [X]
> >
On Tue, Nov 11, 2003 at 03:21:03PM +0100, Gaudenz Steinlin wrote:
> > I do not know if we should think of an entire different solution to
> > handle that but for now we could add some code to main-menu that assures
> > that the priority will not be set >= high until the offending menu point
> > ac
Hello,
it seems that we are locking out people from testing by using isolinux
on our netinst/businesscard CDs, see #220139 and #220139.
Since we do not need any of isolinux's features I propose to use
syslinux on the CDs.
Regards,
Sebastian
--
PGP-Key: http://www.mmweg.rwth-aachen.de/~sebastian
On Tue, Nov 11, 2003 at 11:55:11AM +0100, Martin Pitt wrote:
> None unfortunately. On both machines and with all three CD drives, the
> BIOS tries to boot from CD-ROM, cannot do it and boots from HDD
> instead.
> [...]
>
> However, I just tried to boot the very same CD on the laptop of a
> frien
Hello,
in my opinion the priority handling on errors in main-menu needs still
some work. Presently the logic is simple: Have an initial priority. If a
menu item fails, lower the priority by one step. If a menu item suceeds,
pump up the priority by one but not greater than the initial priority.
Ho
On Mon, Nov 10, 2003 at 01:03:52PM +0100, Kilian Krause wrote:
> sorry, but with the BETA1 this is not fixed.. (or i can't find it for
> some reason)
Yes, unfortunatly it did not made it into the first beta. The bug is
fixed and the new choose-mirror will be part of the next beta.
signature.as
Architecture: source i386
Version: 0.021
Distribution: unstable
Urgency: low
Maintainer: Debian Install System Team <[EMAIL PROTECTED]>
Changed-By: Sebastian Ley <[EMAIL PROTECTED]>
Description:
choose-mirror - Choose mirror to install from (udeb)
Closes: 219503 219716
Changes:
c
Package: kbd-chooser
Severity: minor
Tags: d-i
If I select german (Germany) the default for kbd-chooser will be Swiss
German. I don't know if there are any differences at all to a German
keyboard. If not just close this bug.
-- System Information:
Debian Release: testing/unstable
Architecture: i3
Package: choose-mirror
Severity: important
Tags: d-i
Choose-mirror now asks for country and mirror on priority medium,
defaulting to a country with a corresponding language, selcted in
languagechooser.
1) I do not think it is a good idea to select the mirror automatically
on ddefault priority. Ma
reassign 213314 reiserfsprogs
tags 213314 + patch
thanks
The attached patch implemets Joey's idea for calling mkreiserfs
interactively. It introoduces a simple wrapper, that calls mkreiserfs
with -q.
--
PGP-Key: http://www.mmweg.rwth-aachen.de/~sebastian.ley/public.key
Fingerprint: A46A 753F AED
Am Fr, den 07.11.2003 schrieb Petter Reinholdtsen um 10:52:
> > On user selecting GOBACK out of the module, exit 0, not 30. Its not
> > an error and main-menu shouldn't lower the debconf priority.
>
> How do you expect this to work? If the user chooses GOBACK, he expect
> to get to the previous
* Dominic-Luc Webb wrote:
> [bla]
Stop whining, start reading. It seems you need a little hand in
orientating on our website, so here we go:
1) Start you favorite Browser
2) Enter "http://www.debian.org"; in the addressfield
3) Hit enter
4) You'll see our website with an area to the left called
partconf-mkfstab
Architecture: source i386
Version: 0.10
Distribution: unstable
Urgency: low
Maintainer: Debian Install System Team <[EMAIL PROTECTED]>
Changed-By: Sebastian Ley <[EMAIL PROTECTED]>
Description:
partconf - Configure and Mount Partitions (udeb)
partconf-find-partiti
i386
Version: 0.017
Distribution: unstable
Urgency: low
Maintainer: Debian Install System Team <[EMAIL PROTECTED]>
Changed-By: Sebastian Ley <[EMAIL PROTECTED]>
Description:
choose-mirror - Choose mirror to install from (udeb)
Closes: 211223
Changes:
choose-mirror (0.017) unstable;
Am Mo, den 13.10.2003 schrieb Paul E Condon um 20:21:
> But then I tried 'enter information manually' again and, this time, entered
> http://ftp.us.debian.org/
You must enter a hostname, not an URL. That means you have to enter
'ftp.us.debian.org' for the above mentiones example. The protocol is
On Mon, Oct 13, 2003 at 11:05:15AM +0200, Christian Perrier wrote:
> Retrieving the module modutils-basic failed for unknown reasons.
> Aborting.
>
> Then, I'm bringed back to main-menu, but no other option works
> anymore
>
> This is probably a known bug, but who knows ? If this isn't, agai
On Mon, Oct 13, 2003 at 09:29:59AM +0200, Petter Reinholdtsen wrote:
> We should have a working CD image stored somewhere, if I remember
> correctly.
You do. The image is stored on gluck, where previously the daily
builds were. The daily builds have been moved to a subdir. I agreed
with manty tha
Package: installation
Severity: grave
File: installation
Tags: d-i
Justification: renders package unusable
I use a recent daily build of the businesscard image (12.10.2003). After
I choose my language, d-i loads some kernel modules and then fires up
anna, which fails at a point, leaving the follow
The boot screen that currently is displayed with the testing cd-images
on gluck (iirc more or less a copy of the b-f's bootscreen) is in my
opinion not well suited.
Without special localized builds, the text diplayed there will always be
english, which is IMO a pretty bad idea.
What I would like
Am So, den 12.10.2003 schrieb Goswin von Brederlow um 14:00:
> for (j = 0; retrievers_after[i]; i++)
> I notice two things:
>
> 1. The indentation suggests a missing {} for the first loop
> 2. The second loop might want to test "retrievers_after[j]"
3. The second loop might want to increme
Am Do, den 09.10.2003 schrieb Martin Sjögren um 20:21:
> I know that using a select question in a loop like this in partconf is
> hideous, but I don't see what other option there is.
Just get rid of th OK button in select questions. It is redundant and
misleading. It lead to the usability issues
Am Do, den 09.10.2003 schrieb Steinar H. Gunderson um 18:08:
> Hm? I'm not sure I understand you. I was talking about this particular
> selection box, not selection boxes in general :-)
Yeah, but the frontend code does not know anything about the things it
should display. It always shows the ques
: source i386
Version: 0.25
Distribution: unstable
Urgency: low
Maintainer: Debian Install System Team <[EMAIL PROTECTED]>
Changed-By: Sebastian Ley <[EMAIL PROTECTED]>
Description:
netcfg-dhcp - Configure the network via DHCP (udeb)
netcfg-static - Configure a static network (udeb)
Cl
Am Do, den 09.10.2003 schrieb Steinar H. Gunderson um 15:12:
> What about simply dropping the "Finish" and "Abort" selections? They should
> be the "Finish" and "Back" button anyhow, not entries in a selection box.
And what will you do with that button in languagechooser? Or in the
filesystem que
Package: cdebconf-newt-udeb
Severity: normal
Tags: d-i
I don't know if it is a problem with the german translation "Weiter" for
the OK button, but since two of my (german) friends ran into this usability
issue I believe it is worth reporting.
In partitioner and partconf you have a select question
Am So, den 05.10.2003 schrieb Gaudenz Steinlin um 14:38:
> The image target (which builds a floppy image) is not meaningfull for
> every arch and every boot method. That i386 emulates floppys for cd
> booting is a special case, not a general way of booting from cds. So if
> we want to have a deb f
Am So, den 05.10.2003 schrieb Goswin von Brederlow um 12:14:
> I was going to put the images in the changes file with full path and
> fix the upload scripts to put the files there.
Yeah, that is another solution. But it is up to other poeple to decide
whether they like that or not.
Sebastian
--
Am So, den 05.10.2003 schrieb Geert Stappers um 09:59:
> $ fakeroot debian_proposed/rules binary
> for TYPE in i386 ; do \
> make TYPE=$TYPE image ; \
> done
That won't work, as all scripts will try to access the files in debian
directiry nonetheless. You have to copy debian_proposed to d
Hello,
on the way to a build system that enables us to use the current buildd
infrastructure for d-i images, I checked in a preliminary proposal into
cvs.
I placed it under build/debian_proposed because it discards most of the
old build system. It is quite simple:
- Stick with the idea of havin
Package: rootskel-bootfloppy
Severity: grave
File: rootskel-bootfloppy
Tags: d-i
Justification: renders package unusable
I suppose that with the recent busybox and rootskel upload everything
was in place to build working floppy images again. And in fact, building
of the bootfloppy and floppy targe
Am Fr, den 03.10.2003 schrieb Anthony Towns um 19:26:
> On Fri, Oct 03, 2003 at 03:54:10PM +0200, Sebastian Ley wrote:
> > It would perhaps be a good idea to let udebs propagate that path the
> > same way the debs do and build official testing images only from
> > udebs
Hello FTP Masters, Release Manger, d-i Team,
at the d-i debcamp in Oldenburg the debian-installer team had a
discussion about how d-i will fit into the archive. Presently we are
just pushing our udebs into unstable where the build processes and the
net-install images fetch from. Images are build b
: source i386
Version: 0.19
Distribution: unstable
Urgency: low
Maintainer: Debian Install System Team <[EMAIL PROTECTED]>
Changed-By: Sebastian Ley <[EMAIL PROTECTED]>
Description:
grub-installer - Install GRUB on a hard disk (udeb)
Closes: 187135
Changes:
grub-installer (0.19) unstable;
Package: pcmcia-udeb
Severity: important
File: pcmcia-udeb
Tags: d-i
pcmcia-udeb and ethdetect both provide ethernet-card-detection. Since
main-menu will skip packages who provide something that is already
provided by an installed package ethdetect will never be selected (be
default).
On non pcmc
Am Fr, den 03.10.2003 schrieb Matt Kraai um 00:32:
> On Thu, Oct 02, 2003 at 05:03:36PM -0500, Debian Bug Tracking System wrote:
> > Searching for the reason why grub-installer fails to install grub, I had
> > a look at partconf-mkfstab.
>
> Does this also fix the grub-installer problem, bug 18713
The bug is fixed with a new upload, I just forgot to add myself to
uploaders:
>From the changelog:
- Use proc as the device for the proc filesystem and fix the pass
values. (Closes: #213291)
--
PGP-Key: http://www.mmweg.rwth-aachen.de/~sebastian.ley/public.key
Fingerprint: A46A 753F AEDC 2C01 B
partconf partconf-find-partitions
Architecture: source i386
Version: 0.07
Distribution: unstable
Urgency: low
Maintainer: Debian Install System Team <[EMAIL PROTECTED]>
Changed-By: Sebastian Ley <[EMAIL PROTECTED]>
Description:
partconf - Configure and Mount Partitions (udeb)
p
Version: 0.041
Distribution: unstable
Urgency: low
Maintainer: Debian Install System Team <[EMAIL PROTECTED]>
Changed-By: Sebastian Ley <[EMAIL PROTECTED]>
Description:
main-menu - Debian installer main menu (udeb)
Closes: 213360
Changes:
main-menu (0.041) unstable; urgency=low
.
The fact that the postinst works if invoked manually makes me believe
that at least this is main-menu's fault and that partonf-mkfs is not
executed at all when selected in the menu.
The issue with the wrong proc entry remains of course.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a sub
Package: installation
Severity: important
Tags: d-i
No matter if I choose the defualt keyboard layout for US or German
(Germany), after running base-install the keymap changes to an awerty
keyboard, which is extremely annoying...
-- System Information:
Debian Release: testing/unstable
Architectu
Hello,
as most of you have noticed, we took the opportunity on the small d-i
debcamp to push some necessary major changes into d-i which results now
in unsuable images. I prepared a list of showstoppers and their
resolution:
- The E: unimplemented function bug
This is a incompatiblility of the ma
Am Do, den 02.10.2003 schrieb Robert Millan um 01:08:
> When running the latest d-i test iso image, I get a weird "Unimplemented
> function" error.
This is a known (though not reported) bug. The main-menu module on the
image depends on a newer version of cdebconf than is installed on the
image. T
Am Di, den 30.09.2003 schrieb Geert Stappers um 23:26:
> > - Anna should be able to fetch from stable/testing/sid. That should be
> > a question with priority low, with defaulting to the suite the image
> > was build with.
>
> Should I read stable/testing/sid
> as
>stable or testing or si
Am Di, den 30.09.2003 schrieb Matt Kraai um 23:21:
> * the d-i images in unstable will be rebuilt whenever a udeb is
>uploaded and
>
> * the d-i images in testing will be rebuilt whenever a udeb
>propagates to testing,
Currently not of course but I'd like to see exactly that happen. So
Am Di, den 30.09.2003 schrieb Joey Hess um 18:24:
> What's the advantage of a -src package instead of a source package?
Hm, good question... What will the buildds do with non .deb packages? I
had in mind to to something similar like the debian-cd package...
Perhaps you can elaborate a bit on this
Hi,
on debcamp we had a discussion about how we are ensuring proper
testing versions of debian-installer. The main problem was the build
process of the images. If we had a package in the archive that builds
the installer, which files should it create?
I propose the following:
Make a package buil
On Tue, Sep 30, 2003 at 10:40:54AM +0200, Martin Sjögren wrote:
> These templates can even be supplied by the mkfs udebs, they don't have
> to ship with partconf (they SHOULDN'T ship with partconf!)
>
> Does this sound sane?
Yes, this will play nice with modularity!
Sebastian
--
PGP-Key: http
Package: partconf-mkfstab
Severity: grave
Tags: d-i
Justification: renders package unusable
Searching for the reason why grub-installer fails to install grub, I had
a look at partconf-mkfstab.
When I just run through the normal install process and select to use
this program to create the fstab (i
Package: partconf
Severity: normal
Tags: d-i
The way partconf mkreiserfs will cause mkreiserfs to ask questions,
which will freeze partconf.
-- System Information:
Debian Release: testing/unstable
Architecture: i386
Kernel: Linux coyote 2.4.22-1-k7 #4 Sun Sep 28 13:36:00 EST 2003 i686
Locale: [EM
Package: partconf
Severity: normal
Tags: d-i
With switching to 2.4.22 we have everything in place to create and mount
reiser filesystems. However although the userlandtools are present and
the kernel modules are in place, partconf does not load the kernel
modules and hence does not display the opt
retitle 187135 grub-installer: Fails to install grub
severity 187135 grave
thanks
Given the fact that I encounter the same error as Petter with having
only a / and a swap partition I conclude that grub-installer is broken.
Error message:
Could not find device for /boot: Not found or not a block d
Am Mo, 2003-09-22 um 10.45 schrieb Geert Stappers:
> About the network infrastructure:
>
> May I run a DHCP server on the network?
> The DHCPD should tell which image to transfer by tftp.
As Joey mentioned earlier, we have to build the LAN ourselves. I intend
to bring two 24 Port switches (10 Mb
Am Sa, 2003-09-20 um 16.58 schrieb Gaudenz Steinlin:
> gluck.d.o[1] only contains images for i386 and alpha. If someone tells
> me where to find the scripts that build these images I will have a look
> what needs to be done to adopt these scripts for ppc builds.
> [1] http://gluck.debian.org/cd
Am Do, 2003-09-18 um 02.49 schrieb Geert Stappers:
> At the debian day on Linuxtag 2003 there was a talk about D-I.
>
> Where can I find the source of the PDF that where presented?
http://madkiss.org/div
Sebastian
--
PGP-Key: http://www.mmweg.rwth-aachen.de/~sebastian.ley/public.key
Fingerpri
Version: 0.038
Distribution: unstable
Urgency: low
Maintainer: Debian Install System Team <[EMAIL PROTECTED]>
Changed-By: Sebastian Ley <[EMAIL PROTECTED]>
Description:
main-menu - Debian installer main menu (udeb)
Closes: 210920
Changes:
main-menu (0.038) unstable; urgency=low
.
Am Di, 2003-09-16 um 01.58 schrieb Daniel Silverstone:
> I fired up a bochs machine and booted the netinst CD
>
> It sat looping saying "unable to load vesafb.o"
Fixed in cvs.
> Firstly, the framebuffer console is incredibly slow. But I can cope with
> this...
This is a problem of vmware. It i
Hello,
I send this mail to all people who showed porting activities for d-i in
the past (at least those who are named on the porters site).
Since aj has ambitious plans regarding the sarge release, we should
definitly put some work into ports to other architectures now. Looking
at http://www.debi
Please check, if this bug still occurs with our new d-i images. Two
links for your convenience:
Netinst CD image here:
http://gluck.debian.org/cdimage/testing/netinst/i386/
Floppy images here:
http://people.debian.org/~sjogren/d-i/images/daily/
(you need floppy-image.img and one of net_drivers or
Please check, if this bug still occurs with our new d-i images. Two
links for your convenience:
Floppy images here:
http://people.debian.org/~sjogren/d-i/images/daily/
(you need floppy-image.img and one of net_drivers or cd_drivers)
Netinst CD image here:
http://gluck.debian.org/cdimage/testing/n
Please try to reproduce the bug with the new netinst cd-images. The full
netinst image found here,
http://gluck.debian.org/cdimage/testing/netinst/i386/sarge-i386-netinst.iso
should let you install the base system from cd.
If you still have problems please add more information.
signature.asc
De
tags 207409 + fixed
thanks
I dropped usb-discover and usb-modules from the floppy disks. That makes
them fit onto a 1.44 MB image again. This is meant as a temporary
solution.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Am Fr, 2003-09-12 um 11.08 schrieb Urbani Jacopo:
> Can I test the graphical installer? Is it finish?
No, and considering the tight sarge schedule there will not a graphical
installer for the next release. Priorities have shifted, we need a
stable working beta.
> what's its status?
Some fr
Am Di, 2003-09-09 um 16.11 schrieb Martin Sjögren:
> I think we should reserve a special exit code
> (anna uses 20) to mean "backup" and main-menu should detect this and
> silently ignore it but keeping anna unconfigured.
>
> Does that sound reasonable?
Yeah, sounds so to me. If no one objects,
Package: anna
Version: unavailable; reported 2003-09-09
Severity: normal
1) Select "Load installer modules"
2) Select "Go Back" from the upcoming retriever question.
-> anna exits with an error.
-- System Information:
Debian Release: testing/unstable
Architecture: i386
Kernel: Linux coyote 2.4.2
Package: partitioner
Version: 0.07 (not installed)
Severity: grave
Justification: renders package unusable
After finally fixing the dependencies for cfdisk it still does not
work. Firing up partitioner shows cfdisk black on black. Since I am
able to quit with "q" I suppose it really is cfdisk I am
Am Mo, 2003-09-08 um 17.33 schrieb Martin Michlmayr:
> - The screen on F4 doesn't show /var/log/syslog. Instead, it shows:
> tail: /var/log/syslog: No such file or directory
> tail: no files
Hm, I can not reproduce this with the new netinst images from gluck.
> | db_subst cdrom-detect/success
Hello,
to get rid of the annoying (OLD) entries in main-menu, I want to file a
bug for removing the corresponding modules from the archive. These would
be:
di-utils-mkfs
di-utils-mkswap
di-utils-mount-partitions
di-utils-partitioner
What is with the other packages from di-utils? It seems the onl
Package: base-config
Version: 1.70
Severity: normal
Tags: d-i
After an install with d-i (language set to german) base-config looked rather
ugly when started e.g. the borders were missing. This is most probably not
a base-config bug but since I don't know where to start, you might have an
idea and
Package: cdebconf-newt-udeb
Version: 0.41 (not installed)
Severity: normal
During an install I noticed that the newt frontend does not linewrap
correctly in all cases. Especially the progressbar gets fucked up when
it come to package names or during hardware detection.
-- System Information:
Debi
Am Fr, 2003-09-05 um 00.48 schrieb [EMAIL PROTECTED]:
> This seems like a great approach. My only concern is that
> DEBIAN_FRONTEND may not always be set. If there's only one frontend
> available, and it gets used, does DEBIAN_FRONTEND get set? What's the
> current behavior?
It is always set by
Hello,
for various reasons [1] it is useful that we can switch the frontend of
cdebconf during the install process. I have a patch attached that
enhances cdebconf with this capability. Usage is simple, before each GO,
cdebconf checks whether the value of a question debconf/frontend differs
from th
Package: busybox-cvs-net-udeb
Version: unavailable; reported 2003-09-04
Followup-For: Bug #208685
This is the output when i run the floppy image with busybox-cvs-net-udeb
under bochs:
-->8--
RAMDISK driver initialized: 16 RAM disks of 8192K
Package: busybox-cvs-net-udeb
Version: 0.60.99.cvs20030819-2 (not installed)
Severity: grave
Justification: renders package unusable
The floppy image uses the busybox-cvs-net-udeb. That image fails to
boot, an error occurs while executing the last line of linuxrc (the
chroot command).
Unfortunatl
Am Fr, 2003-08-29 um 13.46 schrieb Davide Viti:
> I tried to run "make main-menu" in the chroot sid environment and had
> problems with gcc3.3.something
> inside woody (gcc 2.95.4) I had to fix a couple of include paths and I
> managed to get the object files but had problems when linking because
Am Fr, 2003-08-29 um 10.27 schrieb Davide Viti:
> I thought it would be very instructive if anyone would be so kind to send me
> a log of a typical gdb session...
Hm, I do not have such a log...
> last night a managed to attach gdb to the runninf process but had
> problems with symbols and so
Am Do, 2003-08-28 um 17.35 schrieb Bastian Blank:
> i commit the changes today.
>
> the following sources depends on libdebian-installer and needs updates:
> [...]
We should not upload udebs that depend on the new libdebian-installer
until we can upload them all at once. Otherwise we get two ver
Package: install
Version: unavailable; reported 2003-08-26
Severity: important
The image built for the floppy type is too large:
System stats for floppy
-
Installed udebs: rootskel main-menu cdebconf-udeb udpkg anna di-utils-shell
kernel-image-2.4.20-1-386-udeb floppy-re
1 - 100 of 168 matches
Mail list logo
