[Frans Pop]
> If you would prefer this message to be send somewhere else,
> please send a mail to [EMAIL PROTECTED]
I believe [EMAIL PROTECTED] would be a
better place to send such autogenerated message. It is the address
where all the commit messages are sent.
--
To UNSUBSCRIBE, email to [EMA
[Frans Pop]
> Where do messages send to that address end up?
In the mailbox of all the subscribers. :)
I do not remember how to subscribe, but it should be documented on the
debian web site somewhere.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Conta
[Steinar H. Gunderson]
> get_serial_console() is a function defined further down in the script,
I guess this is the problem. Shell script functions need to be
defined before they are used.
> which appearently can't be run in a subshell in ash like line 7 tries to
> do:
>
> serial="$(get_serial_
[Colin Watson]
> Perhaps this should be asked at critical instead?
I think dual boot scenarios should be considred special enough to
expect people to run on priority high or lower if they want that to
work. Critical should be used for questions where the default will
work for most people, not for
[Colin Watson]
> I appreciate that dual-boot scenarios are special for debian-edu.
> However, I don't think they're special in the majority of
> installations.
I do not know their relative frequency, but I do know that dual-boots
installs are fairly rare in my "enterprise" environment, at the
univ
[Christian Perrier]
> case "$LANGUAGE" in
> ja|ko|ko_KR|el|zh|zh_CN|zh_TW|bg|ar|he|uk|uk_UA)
> # Japanese, Korean, Greek, Chinese, Bulgarian, Arabic, Hebrew,
> # Ukrainian
> apt-install jfbterm || true
> apt-install unifont || true
> ;;
> ru)
>
[Raphael Hertzog]
> Hi everybody,
>
> I wanted to try d-i with qemu but it failed to detect the network card
> (although I made sure that qemu had network support by using the -n
> switch). Selecting the "ne" module by hand doesn't help either (I tried
> several modules related to NE2K).
The vers
Last years d-i devcamp was a great success. The co-location with the
"non-intel linux gathering" in Oldenburg proved to move the d-i
porting a long step forward, and the grouping of d-i people managed to
improve d-i a lot.
Would it be interesting to repeat the success? Who would join? Date
and
[Steve Langasek]
> Except for zh_CN, zh_TW, and pt_BR, surely?
Is there a need to treat them differently in this code? Both zh_*
need the same packages installed, I believe. And pt need the same
extra treatment as pt_BR.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubsc
[Christian Perrier]
> 3 abandoned (no translator registered)
> ga, se, lv
The original translator for 'se' is
"Børre Gaup" <[EMAIL PROTECTED]>.
I believe the translator for 'lv' is
Aigars Mahinovs <[EMAIL PROTECTED]>.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscrib
[Recai Oktas]
> Please apply the patch attached and fix this issue.
Your patch is not correct. I suspect a real fix will involve
detecting the initialy loaded font/acm/encoding, and undo the termwrap
font loading by loading the initialy loaded font at the end.
--
To UNSUBSCRIBE, email to [EMAI
[Eugeniy Meshcheryakov]
> Why do this? This does not needed for installer.
At the moment, termwrap is a generic tool capable of running a program
in the console with the requested locale and console font. And as
such, it need to do stuff before the program is executed, and undo it
after it is don
I believe it is important that the menu entry for the language choice
do not change when a non-english language is selected, to make sure
the person doing the install is able to find the menu entry to make a
different selection when he make an incorrect choice.
--
To UNSUBSCRIBE, email to [EMAI
Package: languagechooser
Version: 1.28
Tags:d-i patch
At the moment, the locale needed in the second stage of the
installation is generated by termwrap before base-config starts. I
believe it would be better if this is done in the prebaseconfig script
in languagechooser. Even better would b
It would be easier to get the console setup right during installation
if all the console configuration stuff worked the same for all.
console-cyrillic seem to be an exception. Why isn't the content and
behaviour of console-cyrillic merged into console-data/console-tools?
--
To UNSUBSCRIBE, ema
[Recai Oktas]
> IMO, the current situation is not wrong, due to its cyrillic
> spefisic content it is something that should be maintained
> seperately.
I don't understand this argument. With the same argument, we should
have console-latin1, console-ascii, console-big5, and so on and so
forth. An
[Gaudenz Steinlin]
> I think this would be a great idea! If the oldenburg meeting is around
> the same time as last year I could probably participate.
OK. 5 people confirmed their interest in a d-i devcamp. That is
enough to make me try to make it happen. I'm not sure if it will be
co-located
[Kurt Gramlich]
> there is the new skolelinux.de testcenter in Guetersloh, perhaps we
> should use this location?
That would work too. But the Oldenburg meeting is packaged with
strange hardware, and debian-installer need more work on ports to
strange hardware, so a co-location there make very mu
[Per Olofsson]
> Obviously, cardmgr should not be started at all. It seems like
> pcmcia-cs does not use invoke-rc.d in the maintainer script. If I
> change it to use invoke-rc.d (which it should anyway), will that make
> it work? Does d-i use policy-rc.d or something to stop init scripts
> from be
[Per Olofsson]
> Isn't policy-rc.d a cleaner solution? AFAIK nothing in policy requires
> a package to use start-stop-daemon, but it does say that maintainer
> scripts [1]should use invoke-rc.d.
Perhaps. But at the moment, start-stop-daemon has the largest chance
of working.
> Maybe
> both techn
[Martin Michlmayr]
> I'd also like to see "#247906: Please don't install boot loaders"
> addressed (at least on mips/mipsel, I cannot speak for the other
> arches with 100% confidence).
As far as I can see, the following archs still install the boot loader
in debootstrap v0.2.40:
alpha: aboot
[Bastian Blank]
> This is horrible broken. You forget the s390 special case. I revert it
> if it is not fixed within the next hours.
The patch in the mail you quote only move the code around. How can it
be more broken that the original?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subj
[Bastian Blank]
> The Makefile decides to install different versions of this file.
Yes. And that should still work for s390? Or am I misunderstanding
something here?
BTW: Perhaps it is a good idea to drop the makefile magic, and merge
the two versions into one file doing runtime detection of
[Christian Gatzemeier]
> Ok, the man page suggests xdebconfigurator is capable to take the
> locales into account when setting up the X keyboard.
The manual page is out of date. This feature was moved to another
package (locale-config-skolelinux), which isn't uploaded into Sarge
yet. Konstantino
[Peter Green]
> or just default to lilo altogether
> unless there is some other major advantage to grub ofc
There are major advantages with grub. It understand several file
systems (no need to update the boot block if the kernel or initrd is
updated), and it contains a simple editor (so one can f
[Martin Schulze]
> You are welcome, of course.
Thank you very much.
The Oldenburg meeting takes place 22th to 26th of September this year.
I suggest we do this similar to last year. Andreas Schuldei accepted
to take care of recites and other paperwork, and I hope Kurt Gramlich
are interested i
[Goswin von Brederlow]
> Wouldn't it be better to ask the debconf question in base-config and
> not install popcon at all then?
Probably not. I suspect it would require base-config to know even
more about the internal logic of popcon, and I believe that internal
information is better kept inside
[Goswin von Brederlow]
> "Install popcon? (yes/no)"
The question would have to explain more about the popcon question. A
simple 'install package foo?' will not do it.
> Where is there internal knowledge?
>
> If base-config asks if popcon should be installed or not then the
> "participate" quest
[Colin Watson]
> grep-dctrl is your friend ...
Yet another tool I don't use very often. :)
But it does not really solve the problem I wanted to solve using
package-versions. grep-dctrl isn't available within d-i, and I
normally use package-version from within d-i.
But I guess we can drop the ca
[Per Olofsson]
> And here's an (untested) patch which also adds policy-rc.d
> support. Furthermore it corrects a problem in your patch: it runs
> cleanup code if apt-get fails.
I submitted a new bug report against di-utils with your patch
included. Someone with more time then me should test it a
-boot
Date: Thu, 29 Jul 2004 18:09:17 +0200
Petter Reinholdtsen wrote:
> Here is a patch to divert start-stop-daemon. Mostly unsteded, but
> half of it is cut-n-paste from my upgrade-testing script.
And here's an (untested) patch which also adds policy-rc.d
support. Furthermore it corre
[John Summerfield]
> A few minutes ago Ibooted d-i off my week-old network boot setup and
> tried to install Woody.
>
> It coukln't find the needed netinstall image from the local official
> mirror.
I'm not too surprised. Woody installs are only tested with CDs, as
far as I know. No-one tested
[Anton Zinoviev]
> I browsed the archives of the mailing list of parted and discovered that
> this is a known bug of parted. There were even flame wars about it.
> This is going to be quiet unfortunate to many people who will try to
> install Debian on dual-boot systems (#258880).
Is is all this
[Tod Hagan]
> I've got a case where debian-installer RC1 fails on reboot. I'm
> writing up my installation report, but first I have a question.
Remember both lspci and lspci -n. :)
> I'm not sure that the necessary modules are loaded to see the RAIDed
> drives holding the install. Would these mod
[Klaus Ade Johnstad]
> I tried setting my hostname to something including the 3 special
> Norwegian characters, and was told that the hostname was invalid (no
> surprise there), with the explanation that the hostname could only
> contain alphanumeric characters and the minussign.
>
> Clearly the 3
[Joshua Kwan]
> We currently accept uppercase/lowercase hostnames. That code has not
> changed in a _long_ time.
As DNS is case insensitive, I believe the comment would do the user a
favour by recommending the use of only lowercase letters.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a s
[Recai Oktas]
> There are some minor problems with Swedish entry which doesn't cause
> harmful results at the moment, but may introduce unexpected bugs in
> future as far as the current format of languagelist and SUPPORTED list
> concerned:
>
> * Fallback locale code (3rd entry 'sv_SE.ISO.8859-
[Christian Perrier]
> After lot of thinking, I have identified two major things to change in
> our infrastructure:
>
> 1) the translations of D-I package split over dozens of files, one per
>package
> 2) the weaknesses of the current "3 stages" system for following
>translation progress
T
[Petter Reinholdtsen]
> I thought there were a bug reported with this idea, but I failed to
> find it in BTS. Joey, do you remember where we wrote down this
> idea?
Found it. Bug #212921, and it is already fixed. Time to use it, I
guess. :)
--
To UNSUBSCRIBE, email to [EMAIL PROTEC
[Christian Perrier]
> Weprobably now have to think how we could use it during a complete
> system install.
>
> My first idea is adding the package to each language task. Maybe
> should we try first with French and Greek?
What about installing it from languagechooser, and add base-config
fragments
[Petter Reinholdtsen]
> Found it. Bug #212921, and it is already fixed. Time to use it, I
> guess. :)
To use it, we would need to call debconf-loadtemplate with --merge.
Here is a patch to change udpkg to do that. This would allow us to
append translations to d-i by making an udeb with t
[Frans Pop]
> The only problem is that I've not yet found/seen/heard anywhere what
> the function is of the pointer in locales to a specific charset
> (i.e. where this is actually used and what goes wrong if you point
> to the wrong charset).
As far as I know, nl_NL is using the EURO, and need to
[Christian Perrier]
> What you're saying here indeed, is that the nl_NL as well as fr_FR and
> so on locales are indeed inconsistent and should *never* be used
Not quite, but almost. The fr_FR listing in SUPPORTED should not be
used, as it uses the ISO-8859-1 charset. But it can not be repla
[Peter Palfrader]
> A better default would be 3128 (squid's default port) or 8080.
I agree. The squid default port 3128 is the proper default port for a
proxy URL.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
[Konstantinos Margaritis]
> Why don't we just use UTF-8? I've been using for Greek for more than a
> year and it's been much easier (and Greek is a difficult case
> anyway). And it solves all these silly problems of @euro and weird
> encodings. Not to mention it's more universal.
Ah, to boldly
[Frans Pop]
> Hmmm. Shouldn't the locale be "ja_JP.EUC-JP"?
> That's the value that's in the languagelist for d-i.
At least EUC-JP is the name of the charset in glibc. The other names
are aliases.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [
[Andreas Schildbach]
> Hello everyone,
>
> does the new Debian Installer somehow support to install Debian on
> encrypted partitions?
No. But I would love to have it. Are you up for the job?
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMA
[Agustin Martin]
> Sumarizing for them, debian-installer/language possible values were
> originally single iso-639 two letters language code, but seems that
> debian-installer/language allows things like the one you used
> (de_DE:de_DE:de:en_GB:en). Since the code for selection of
> ispell/wordlist
[Christian Perrier]
> If you need to get the user's language, you should either get the
> value from the LANG variable, or grab debian-installer/language and
> then "cut -f1 -d: | cut -f1 -d_"
Or perhaps select as default the first available dictionary in the
priority order given in /language.
-
[Herbert]
> Ola
Hei. :)
> fiz a instalacao do Debian basico, usando disquetes 1.44.
>
> No topico - 11.2.3.3.1 Imagens do disquete de drivers: parece que
> existe uma opcao nao necessaria. O arquivo *
> .../current/images-1.44/bf2.4/driver-5.bin nao esta disponivel para
> download e ao iniciar
[Willi Mann]
> I'm seeing a rather strange problem with RC1 and kernel 2.4:
>
> The new installation reboots after exactly 80 seconds.
This sounds like some kind of kernel watchdog trigger a reboot. I've
seen it for some via kernel module, but there might be some other
kernel modules doing this
[Kęstutis Biliūnas]
> Hello,
>
> In attachement there are initial translated lithuanian (lt_LT)
> files for d-i. Could someone commit them for me?
Thank you. :)
I've commited the patch to languagechooser, but do not have time to do
all the translations at this time. If no one beats me to it, I'
The langauge/region choices used in the new installatino system
display a translated language option. But some of these options are
not translated properly. The text was copied from the Woody
installer, and here the text contain a 'press enter' part which is
taken out of the new installer to mak
Good work adding more error detection in udpkg. I look forward to
test it in a low-memory situation. :)
[Matt Kraai]
> - SYSTEM(buf);
> + if (SYSTEM(buf) != 0)
I suspect it would be better to use the logging system() wrappe
Package: anna
Version: 0.038
Severity: serious
Tag: d-i patch
The current anna implementation have special code to detect kernel
module udebs, and make sure to only load modules for the running
kernel version. This code assumes that all kernel module udebs are
named like this: foo-modules
It is correct that autopartkit will fail if the partitions are already
mounted. There is currently no code to detect and umount when this
happen.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
I just managed to install Skolelinux using the current
debian-installer. It is a while since I was able to do that.
I had to install the udeb containing lvm-mod manually, as anna failed
to install the correct kernel modules. (Bug #216433)
I also had to rerun kernel-installer, as the default ker
[Edmund GRIMLEY EVANS]
> Esperanto: Elektu ô ©i tion por daô¡rigi en Esperanto
Thank you. CVS is now updated.
And thank you for all the others checking entries. Still some entries
left. :)
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMA
[Edmund GRIMLEY EVANS]
> Esperanto: Elektu ô ©i tion por daô¡rigi en Esperanto
Thank you. CVS is now updated.
And thank you for all the others checking entries. Still some entries
left. :)
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EM
This is based on my experience with kernel-installer from the
base-installer source directory.
It is currently possible to set the value of a select question to a
option that is _not_ in the list of choices, and get the value back
from debconf. This is annoying.
Given the following template:
[Petter Reinholdtsen]
> Please check the CVS version at
>
>
> http://cvs.debian.org/debian-installer/tools/languagechooser/languagelist.l10n?rev=HEAD&content-type=text/vnd.viewcvs-markup>,
>
> and send updates to [EMAIL PROTECTED] I will remove the
> entries wit
[Matt Kraai]
> Log Message:
> Do not default to a package that provides a virtual package that is
> also provided by a package that is already installed (closes:
> #216787).
Are you sure this work? I tried to test it, and it still selected
lilo-installer after grub-installer completed successful
Autopartkit uses libparted and ped_device_get_next() to locate
available disks and partitions. If this is accessing removable
devices, I suspect the problem is in libparted, and not in
autopartkit.
I'm not sure how libparted detect partitions.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
wit
[Kenshi Muto]
> I'm testing debian-installer, but found some problems about
> localize messages.
I just noticed that no one replied to your message. Some of the
issues you brought up are fixed, as we moved the translation of menu
entries to templates, while others are yet to be solved.
> 4. Buil
Please clean up the conflict you commited, and make it a habit to use
'cvs diff -u' before you commit:
> Index: kernel-installer.postinst
> ===
> RCS file:
> /cvs/debian-boot/debian-installer/tools/base-installer/debian/kernel-instal
[Gaudenz Steinlin]
> Since I didn't hear anything from progeny I have made discover2 packages
> for testing. You can download them at
> http://www.soziologie.ch/~steinlin/d-i/discover2/
Good work. Sorry it took so long before I found time to test it.
Here is the initial test report:
When I try
[Denis Barier]
> Log Message:
> Add some translations of country names
[...]
> msgid "IL"
> -msgstr ""
> +msgstr "Israël"
This is not translating. This is rephrasing.
If the original text is untranslatable, it should be changed to make
it translatable. This will solve the problem for all the t
Package: cdebconf
Version: 0.47
Severity: serious
The current cdebconf is impossible to install in unstable. This stop
the package from entering testing.
The problem is that one of the dependencies for cdebconf depend in
turn on debconf, and cdebconf again conflict on debconf. This make it
i
<[EMAIL PROTECTED]>
Changed-By: Petter Reinholdtsen <[EMAIL PROTECTED]>
Description:
cdebconf - Debian Configuration Management System (C-implementation)
cdebconf-gtk-udeb - Gtk+ frontend for Debian Configuration Management System (udeb)
cdebconf-newt-udeb - Newt frontend for Debian C
[Safir Secerovic]
> I hope that we will get further instructions on how to proceed
You need to update the languagechooser package as well, to get your
language listed as a language option.
Check README.trans for instructions.
You probably also want to make sure the correct keyboard is selected
(
Package: partconf-mkfstab
Version: 0.12
Severity: minor
At the moment, the fstab entry for /cdrom is using the currently
mounted device instead of /dev/cdrom. I believe it should be changed
from for example /dev/hdc on my machine, to /dev/cdrom.
This is the currently generated line:
/dev/h
Package: libdebian-installer4-dev
Version: 0.17
When running 'debuild clean' in the source directory, it starts by
cleaning the source, and then continue by configuring and building the
package.
I'm not sure why, but it should be fixed to only clean the source.
--
To UNSUBSCRIBE, email to [EM
[Falk Hueffner]
> First, IMHO we should drop this stuff and just use malloc. I don't
> see any point in increasing code size and introducing new bugs
> thereby.
I agree too. Unless there is a very good reason to keep it, we should
remove the allocator code and use malloc() instead.
Someone need
[Richard Hirst]
> debug builds of kbd-chooser call di_check_dir() but versions of
> libdebian-installer beyond 0.15 do not provide it.
Checking d-i/doc/libdi-changes.txt, I do not find that function
mentioned at all. I'm not sure what happened to it, nor why it was
removed.
Perhaps kbd-chooser
[Geert Stappers]
> I believe it hides which physical device the cdrom player is.
Yes, that is the nice thing about it. :)
> There needs also at an other place the creation of the link from
> /dev/cdrom to /dev/hdc.
The discover package can handle this.
> Would like to mark this bug as WONTFIX,
Package: autopartkit
Version: 0.63
Severity: important
At the moment, the LVM logical volumes created are gived the minimum
size by default. The reason is simply that the code to scale the
logical volumes relative to the size of the volume group is missing,
and that it was easy to use the mini
Package: autopartkit
Version: 0.63
Severity: minor
There is no progress bar when the LVM logical volumes are created.
There should be.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
> Added missing alignment for array in endianess test.
What about this patch instead? Less code, and checking the endianness
at compile time.
Index: udpkg.c
===
RCS file: /cvs/debian-boot/debian-installer/tools/udpkg/udpkg.c,v
retr
[Joey Hess]
> For the record, this is my first successful install of d-i.
Congratulations. :)
> and it didn't ask me if I wanted grub (which I generally do), and just
> installed lilo.
I would like to switch the selection order of lilo and grub, to get
grub as the default. Any objections?
> He
[Christian Perrier]
> After finding this, I noticed that this letter is never displayed
> wherever it appears in d-i templates.
I suspect that this is because that letter is missing in the templates
used on the boot floppy, and because of this is missing in the reduced
font used by bterm.
The pro
[Daniel Jacobowitz]
> Can you just restart bterm? Yes, I realize that's got some issues
> since you're running inside of it. But it's really not set up to
> re-parse the font.
I would rather not restart debian-installer and main-menu, and it seem
to be a requirement to be able to restart bterm.
> 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 question, not just continue like nothing
happened. Returning OK (0)
Package: autopartkit
Version: 0.63
Severity: important
Tags: d-i
When one of the Skolelinux developers install on his laptop, Dell
Latitude CP M233XT, autopartkit dies with a floating point exception
(SIGFPE). The complete bug report is available from
http://bugs.skolelinux.no/show_bug.cgi
[Debian Installer]
> There are disparities between your recently accepted upload and the
> override file for the following file(s):
>
> grub-installer_0.24_i386.udeb: package says priority is standard,
> override says optional.
The correct priority is standard. We want grub to be available by
de
[Gaudenz Steinlin]
> Please try to build from a clean system without libexpat1-dev and
> without libdiscover2-pic and libdiscover2-dev. Probably it is also a
> good idea to remove all discover1 packages to prevent weird mklibs
> inferences there.
I removed all discover-related packages on my build
[Andreas Schuldei, 2003-10-11]
> i am still missing numberous bills and forms for the developer
> meeting in Oldenburg.
>
> please send them to me soonish.
>
> Andreas Schuldei
> Torekällgatan 35, 3tr
> 15133 Södertälje
> Sweden
I've sent mine. I hope everyone else sent their.
--
To UNSUBSCRI
[Sebastian Ley]
> Even worse, since the udeb will be left unconfigured it will select the
> same entry again. Wit debconf priority=high, you won't have a chance to
> break out, because the item will be selected automatically.
Eh, are you sure? If the postinst script return 0, udpkg will flag it
a
[Alastair McKinstry]
> Do we (expect to) prime the /etc/modules on /target in d-i, in
> general, eg. for network interfaces, or should kbd-chooser add the
> USB modules if it spots a USB keyboard?
My plan is to leave this to the HW detection system (discover), and
make sure it detect and loads the
tags 219464 + pending
thanks
[Santiago Vila]
> This package does not compile under GNU/Hurd because there is no
> PATH_MAX there.
Fixed in CVS.
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
[Kenshi Muto]
> I tried debian-installer beta1 CD (net-install), and it worked on
> first stage. :-)
Sounds good.
> But I noticed second stage (after reboot) had a problem for
> non-Latin users (Japanese, Chinese, Korean, and other non-Latin
> people needs special console).
Hm, I suspected this.
retitle 219558 Crashes when asked to do simple partitioning
thanks
The problem is not with special hardware, it was with the standalone
profile in Skoleliniux. It is triggered by simple partition table
specification. This fail:
/ default 800 5120
swap swap 2
[Branden Robinson]
> Please feel free to go ahead.
>
> Progeny is concentrating on Discover 2.x:
This sounds very good. When will this result in replies regarding the
patches we have been working on in the d-i team to get it to work with
d-i. As far as I know, we are now waiting for Progeny to
[Christian Perrier]
> As discover 1 currently has a FTBFS bug (221302, opened yesterday when
> I tried to build a NMU for it), this is definitely likely.. :-)
Oh. Nto good. NMU required imidiately, I suspect. :)
> But, I guess this switch to discover 2 will occur after sarge
> release ?
I want
[Robert Millan]
> I (as GRUB co-maintainer) have discussed with Jason the possibility
> of GRUB being the default bootloader for i386 in Debian 3.1, and we
> agree that GRUB is now mature enough for that.
Sounds good. I would also use grub as the default. We do it already
in Skolelinux.
> Expec
Package: grub-installer
Version: 0.24
Severity: important
When installing using a serial line console, grub-installer should
enable the serial line console at boot time when installed. This is
from a installation report:
I mentioned that I was on a serial console. I had to use the
instal
[Joey Hess]
> Please stop committing large changes to this very core part of d-i.
I agree to this. We need to be more conservative with the changes to
libd-i.
> These changes will not be accepted into the next beta. The core of
> d-i is frozen.
Some of the changes will have to be accepted. Get
[Jeremie Koenig]
> I think i've read somewhere that the d-i tag wasn't supposed to be
> used on d-i only packages. This seems wrong. Should we tag all d-i
> bugs d-i when we sbumit them ?
I use the d-i tag to list all d-i related bugs. For this to work, all
such packages need the tag, including t
[Geert Stappers]
> Add missing serial line (detection) support to the list.
Are you talking about the missing serial line support in
grub-installer, or some feature of the grub boot loader? The first
should be easily fixable. At least we use grub with serial console
here at the university, and i
[Joey Hess]
> Please can't we change this to use symbolic release names?
Check that it work with CD installs first. I remember some problems
with missing symlinks and / or debootstrap only understanding the code
names. I am not sure if this template is the one used by the
retrievers nor base-ins
At the moment, both kbd-chooser, lilo-installer and grub-installer
need to know if the installation is using a serial console. In
addition, there is code to detect this in rootskel
(/lib/debian-installer.d/S30term-linux) and set TERM_TYPE=serial in
this case. To avoid duplicate code, I suggest u
1 - 100 of 1264 matches
Mail list logo
