Bug#852779: installation-reports: Mostly successful install amd64 stretch RC1
Google tells me bug 849124 covers the reportbug issues, and installing gir1.2-vte-2.91 did fix the issue. I reckon that is important enough, though, that Stretch should not be released until that fix makes its way into the release candidate. On Fri, Jan 27, 2017 at 10:18 PM, Chris Tillman <toff.till...@gmail.com> wrote: > Package: installation-reports > Severity: normal > Tags: d-i > > Dear Maintainer, > > I used the graphical installer (for the first time). It worked well, but > really provided nothing beyond the curses-based installer. I'd recommend > some artist provide some lovely backgrounds or maybe a Debian history > lesson > playing under the progress bar or something. Just sayin'. > > I Hooked up wifi with no difficulty. Resized my > Windows partition ... took a bit long, could use a progress marker , or > just > a spinning ball or such to let you know it's all still OK. Selected > suggested > scheme for partitions, well done, just what I was planning. > > One minor item, and who knows this may be my hardware. When I was using the > touchpad to try to select the Complete button on any screen, I had a > difficult > time. If I made a small stroke towards the button and lifted my finger, > when > I put the finger down again the cursor would go back to the center of the > screen. I finally learned the way to make it work was to make one > continuous > movement to the Complete button, then I could select it when I got there. > > At the end, when Grub was installing, it did have an incorrect message. It > said it had found a Vista loader when in fact it is Windows 7. After the > install and reboot, turns out it knew this all along ... in the boot menu > it is listed as Windows 7. > > After the boot into the system, I selected reportbug from the menu. It > didn't respond. I then went into a terminal and started it. During setup, > I asked for it to be the graphical version. When it tried to open > graphically, > it gave the attached errors. > > I suspect something wasn't installed during the basic installation to > support > it. I am using XFCE, so maybe not as well supported out of the box. > > Since reportbug seems to work in the text mode, I guess I won't report a > bug > against it ... must be system installation support. > > > > -- Package-specific info: > > Boot method: CD > Image version: Stretch RC1 > Date: > > Machine: HP Probook 4530s > Partitions: > > > Base System Installation Checklist: > [O] = OK, [E] = Error (please elaborate below), [ ] = didn't try it > > Initial boot: [ ] > Detect network card:[ ] > Configure network: [ ] > Detect CD: [ ] > Load installer modules: [ ] > Clock/timezone setup: [ ] > User/password setup:[ ] > Detect hard drives: [ ] > Partition hard drives: [ ] > Install base system:[ ] > Install tasks: [ ] > Install boot loader:[ ] > Overall install:[ ] > > Comments/Problems: > >and ideas you had during the initial install.> > > > -- > > Please make sure that the hardware-summary log file, and any other > installation logs that you think would be useful are attached to this > report. Please compress large files using gzip. > > Once you have filled out this report, mail it to sub...@bugs.debian.org. > > == > Installer lsb-release: > == > DISTRIB_ID=Debian > DISTRIB_DESCRIPTION="Debian GNU/Linux installer" > DISTRIB_RELEASE="9 (stretch) - installer build 20170112" > X_INSTALLATION_MEDIUM=cdrom > > == > Installer hardware-summary: > == > uname -a: Linux ctillman 4.8.0-2-amd64 #1 SMP Debian 4.8.15-2 (2017-01-04) > x86_64 GNU/Linux > lspci -knn: 00:00.0 Host bridge [0600]: Intel Corporation 2nd Generation > Core Processor Family DRAM Controller [8086:0104] (rev 09) > lspci -knn: Subsystem: Hewlett-Packard Company Device [103c:167c] > lspci -knn: 00:01.0 PCI bridge [0604]: Intel Corporation Xeon E3-1200/2nd > Generation Core Processor Family PCI Express Root Port [8086:0101] (rev 09) > lspci -knn: Kernel driver in use: pcieport > lspci -knn: 00:02.0 VGA compatible controller [0300]: Intel Corporation > 2nd Generation Core Processor Family Integrated Graphics Controller > [8086:0116] (rev 09) > lspci -knn: Subsystem: Hewlett-Packard Company Device [103c:167d] > lspci -knn: 00:16.0 Communication controller [0780]: Intel Corporation 6 > Series/C200 Series Chipset Family MEI Controller #1 [8086:1c3a] (rev 04) > lspci -knn: Subsystem: Hewle
Bug#852779: installation-reports: Mostly successful install amd64 stretch RC1
it looks like this is the first time you have used
reportbug, we are configuring its behavior. These settings will be saved to the
file "/home/chris/.reportbugrc", which you will be free to edit further.
Please choose the default operating mode for reportbug.
1 noviceOffer simple prompts, bypassing technical questions.
2 standard Offer more extensive prompts, including asking about things that a
moderately sophisticated user would be expected to know about
Debian.
3 advanced Like standard, but assumes you know a bit more about Debian,
including "incoming".
4 expertBypass most handholding measures and preliminary triage routines.
This mode should not be used by people unfamiliar with Debian's
policies and operating procedures.
Select mode: [novice] 2
Please choose the default interface for reportbug.
1 text A text-oriented console user interface
2 gtk2 A graphical (GTK+) user interface.
Select interface: 2
Will reportbug often have direct Internet access? (You should answer yes to
this question unless you know what you are doing and plan to check whether
duplicate reports have been filed via some other channel.) [Y|n|q|?]? y
What real name should be used for sending bug reports?
[Chris Tillman]>
Which of your email addresses should be used when sending bug reports? (Note
that this address will be visible in the bug tracking system, so you may want
to use a webmail address or another address with good spam filtering
capabilities.)
[chris@ctillman]> toff.till...@gmail.com
Do you have a "mail transport agent" (MTA) like Exim, Postfix or SSMTP
configured on this computer to send mail to the Internet? [y|N|q|?]? n
Please enter the name of your SMTP host. Usually it's called something like
"mail.example.org" or "smtp.example.org". If you need to use a different port
than default, use the : alternative format. Just press ENTER if you
don't have one or don't know, and so a Debian SMTP host will be used.
>
Please enter the name of your proxy server. It should only use this parameter
if you are behind a firewall. The PROXY argument should be formatted as a valid
HTTP URL, including (if necessary) a port number; for example,
http://192.168.1.1:3128/. Just press ENTER if you don't have one or don't know.
>
Default preferences file written. To reconfigure, re-run reportbug with the "--
configure" option.
Traceback (most recent call last):
File "/usr/bin/reportbug", line 2233, in
main()
File "/usr/bin/reportbug", line 1107, in main
return iface.user_interface()
File "/usr/bin/reportbug", line 1225, in user_interface
main()
File "/usr/bin/reportbug", line 1084, in main
if newui.initialize():
File "/usr/lib/python3/dist-packages/reportbug/ui/gtk2_ui.py", line 1580, in
initialize
gi.require_version('Vte', '2.91')
File "/usr/lib/python3/dist-packages/gi/__init__.py", line 118, in
require_version
raise ValueError('Namespace %s not available' % namespace)
ValueError: Namespace Vte not available
chris@ctillman:~$ reportbug
Traceback (most recent call last):
File "/usr/bin/reportbug", line 2233, in
main()
File "/usr/bin/reportbug", line 1084, in main
if newui.initialize():
File "/usr/lib/python3/dist-packages/reportbug/ui/gtk2_ui.py", line 1580, in
initialize
gi.require_version('Vte', '2.91')
File "/usr/lib/python3/dist-packages/gi/__init__.py", line 118, in
require_version
raise ValueError('Namespace %s not available' % namespace)
ValueError: Namespace Vte not available
chris@ctillman:~$
Bug#762316: To avoid installer lock-up, iso-scan should always ask if more than one iso image is found
Package: iso-scan Version: 1.53 Severity: important Tags: d-i Dear Debian Install System Team, I found a bug using iso-scan in the installer image downloaded from http://ftp.citylink.co.nz/debian/dists/testing/main/installer-amd64/current/images/hd-media/initrd.gz dated 2 Aug 2014. The bug is only triggered under the following conditions: 1) debconf priority = high or greater 2) More than one iso image in the folder where iso-scan detects an iso. I marked it important, though, because of the consequences. In my case, I started to see this error in console 2 when trying to ls: ls: /lib/libc.so.6: version 'GLIBC_2.15' not found (required by ls) As you can imagine, this meant I couldn't do ANYTHING after this, other than hold the power button down on the computer for 5 seconds. Every menu item fails, including the Abort menu item. The problem was that I had two iso-scan images on the root of the partition. The first one alphabetically was a CD for wheezy, while the second was the one I wanted ... for jessie. What iso-scan did, under high debconf priority, was just use the first iso, even though it was inappropriate for the jessie installer. I got a warning about no modules, and after that was when the GLIBC error started coming up. It didn't matter what response I selected to the modules error; it was already too late. If instead debconf is set to medium or low, iso-scan asks which iso you want to use. This should be the behaviour, no matter what the debconf level, if more than one iso image is found. Otherwise we go from asking for a scan to complete failure, just for having two isos in the folder. Chris
Bug#760712: WEP vs WPA2
With this installation I was using WPA2-Personal at the access point. I see the log messages look similar as in bug #741622, deauthenticating immediately after connect. For a later install on the same computer, also to USB target, I used WEP in the installer to connect to the access point (installation report #761148). So possibly the problem has to do with WPA2 as opposed to WEP. -- To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/cao299fvp970app22gegx4ev9qw9g8thwf3mdfz6drrfyy6n...@mail.gmail.com
Bug#760712: Further info
On Sun, Sep 7, 2014 at 3:33 PM, Chris Tillman toff.till...@gmail.com wrote: Here is the picture of the Win 8 WLAN properties which I claimed to have attached originally. -- Chris Tillman Developer -- Chris Tillman Developer wlan0-in-windows-conf.pdf Description: Adobe PDF document
Bug#760712: Further info
Here is the picture of the Win 8 WLAN properties which I claimed to have attached originally. -- Chris Tillman Developer -- To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/cao299ft6mybxt3svsucducclhjw-gg8-teguxrkaig7a4kk...@mail.gmail.com
Bug#760014: win32-loader This program doesn't support dialog box hidden by browser window
Package: win32-loader Version: 0.7.5 I downloaded the daily-build jessie iso today for amd64. I double clicked on the downloaded file and Windows 8 opened it as disk F:. I tried double clicking setup.exe on F: from within Windows 8. The Choose Language dialog box appeared, and I clicked OK. The computer beeped, a Debian-installer loader message alert box was displayed, and the file:///F:/README.html page was displayed in the browser. The issue I have is that the web page is displayed _last_. The dialog box displaying the problem was hidden behind the browser window. I didn't notice the dialog box until I went to shut down the computer, after having been confused by the nice README information but assuming there must have been some problem anyway. Is there any way to reliably make the dialog box appear _after_ the browser window comes up? That would make more sense in this instance. -- Chris Tillman Developer -- To UNSUBSCRIBE, email to debian-boot-requ...@lists.debian.org with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/cao299fsqg6tnfv4wutndkmnccjyas3vhlikeh2xeo7evbvu...@mail.gmail.com
Bug#640789: Crash on folder name with spaces
Hi Cyril,
Thanks for your attention. I think that was my patch. (Chris Tillman).
The only line the patch is affecting is the one with find $dir, adding the
quotes to make it find $dir.
The rest of it certainly does look strange, but anyway it works if you just
add the quotes. Maybe another time to refactor it, because we have been
waiting years for this patch already.
I verified today it is still present in the current testing installer.
Chris
On Mon, 7 Jul 2014 03:56:21 +0200 Cyril Brulebois k...@debian.org wrote:
Control: tag -1 -patch
Hi Modestas,
Modestas Vainius mo...@debian.org (2013-12-29):
Control: tags -1 patch
thanks for the patch but I'm not convinced, see below:
--- a/debian/iso-scan.postinst
+++ b/debian/iso-scan.postinst
@@ -162,7 +162,7 @@ scan_device_for_isos() {
elif [ $look_subdirs = 1 ]; then
opt=-type f
fi
- isolist=$(find $dir $opt -name *.iso -o -name *.ISO 2/dev/null)
+ isolist=$(find $dir $opt -name *.iso -o -name *.ISO 2/dev/null)
This part is certainly OK; at least I can't think of a reason why that
wouldn't be a good thing.
TOPLEVEL_DIRS_COUNT=$(($TOPLEVEL_DIRS_COUNT + 1))
for iso in $isolist; do
but then that means we're possibly going to fail here. Example:
kibi@wodi:~/isos$ ls */
baz/:
baz.iso
foo bar/:
foobar.iso
kibi@wodi:~/isos$ isolist=$(find $dir $opt -name *.iso -o -name
*.ISO 2/dev/null)
kibi@wodi:~/isos$ for iso in $isolist; do echo Found ISO $iso; done
Found ISO ./foo
Found ISO bar/foobar.iso
Found ISO ./baz/baz.iso
I guess it would make sense to fix this for real instead of hiding it a
bit further. Unfortunately 4am isn't a great time to set up a reproducer
and to keep on hacking. :/
(Also, sorry for the lag.)
Mraw,
KiBi.
Bug#737170: Potentially duplicate bug
The error shown Jan 30 21:04:35 main-menu[251]: WARNING **: Configuring 'iso-scan' failed with error code 1 Jan 30 21:04:35 main-menu[251]: WARNING **: Menu item 'iso-scan' failed. is the same symptom shown in iso-scan bug 640789, caused by an unquoted $dir variable in a find command within iso-scan.postinst. OP, did the volume you were trying to query have any paths with spaces in it? That would cause this error. -- Chris Tillman Developer
Bug#628991: Potential bug duplicate
Also see bug 640789, which could be a duplicate.
Bug#712675: Potential duplicate bug
The error shown Jul 10 04:51:52 main-menu[213]: WARNING **: Configuring 'iso-scan' failed with error code 1 Jul 10 04:51:52 main-menu[213]: WARNING **: Menu item 'iso-scan' failed. is the same symptom shown in iso-scan bug 640789, caused by an unquoted $dir variable in a find command within iso-scan.postinst. OP, did the volume you were trying to query have any paths with spaces in it? That would cause this error.
Bug#640789: Same patch applies for more bugs
The patch is a simple fix, and should be applied with urgency. It will also fix some other bugs 628...@bugs.debian.org https://bugs.launchpad.net/ubuntu/+source/iso-scan/+bug/838720 Frans Pop seems to have noticed this in bug 440301 and thought he fixed it in 2007: -- P.S. It does look like there is a minor bug in isoscan (which should not affect this case though). The first call in line 111 uses quotes, the second one in line 166 does not; I think this will make the comparison in line 167 always false (if there are multiple DEVS). I've fixed that. -- So, here it is 7 years later. How can this bug still exist? -- Chris Tillman
Bug#571958: hfsplus supported
Hi, I just wanted to throw in here that the iso images are found successfully on an hfsplus drive (PowerMac G4, MacOS Extended partition). I am getting the same issue as in 628991, so will post my full results over there. But as far as hfsplus, no worries: Feb 9 19:02:40 iso-scan: selected_device(s)='/dev/sda9' Feb 9 19:02:40 kernel: [ 351.164265] FAT-fs (sda9): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive! Feb 9 19:02:41 kernel: [ 351.764528] FAT-fs (sda9): utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive! Feb 9 19:02:41 iso-scan: Mounted /dev/sda9 for first pass Feb 9 19:02:41 iso-scan: Found ISO ./debian-7.3.0-powerpc-netinst.iso on /dev/sda9 Feb 9 19:02:41 kernel: [ 352.324499] ISO 9660 Extensions: Microsoft Joliet Level 3 Feb 9 19:02:41 kernel: [ 352.356792] ISO 9660 Extensions: RRIP_1991A Feb 9 19:02:41 iso-scan: Detected ISO with 'stable' (wheezy) distribution Feb 9 19:02:41 iso-scan: Detected ISO with distribution 'stable' (wheezy) Feb 9 19:02:41 iso-scan: Debian ISO ./debian-7.3.0-powerpc-netinst.iso usable Feb 9 19:02:41 iso-scan: Found ISO ./debian-testing-powerpc-netinst.iso on /dev/sda9 Feb 9 19:02:42 kernel: [ 352.530717] ISO 9660 Extensions: Microsoft Joliet Level 3 Feb 9 19:02:42 kernel: [ 352.557330] ISO 9660 Extensions: RRIP_1991A Feb 9 19:02:42 iso-scan: Detected ISO with 'testing' (jessie) distribution Feb 9 19:02:42 iso-scan: Detected ISO with distribution 'testing' (jessie) Feb 9 19:02:42 iso-scan: Debian ISO ./debian-testing-powerpc-netinst.iso usable Feb 9 19:02:42 iso-scan: Found ISO ./Applications/debian-7.3.0-powerpc-netinst.iso on /dev/sda9 Feb 9 19:02:42 kernel: [ 353.105643] ISO 9660 Extensions: Microsoft Joliet Level 3 Feb 9 19:02:42 kernel: [ 353.141321] ISO 9660 Extensions: RRIP_1991A Feb 9 19:02:42 iso-scan: Detected ISO with 'stable' (wheezy) distribution Feb 9 19:02:42 iso-scan: Detected ISO with distribution 'stable' (wheezy) Feb 9 19:02:42 iso-scan: Debian ISO ./Applications/debian-7.3.0-powerpc-netinst.iso usable Feb 9 19:02:42 main-menu[251]: WARNING **: Configuring 'iso-scan' failed with error code 1 -- Chris Tillman Developer
Bug#628991: Repeated, some more info, patch
Tags: patch Hi, I got the same failure when trying to install on PowerPc. I believe it is related to quoting as the submitter suggested. iso-scan first looks at the root level to find iso's, then scans each first-level directory as well, using the BusyBox find command. When parentheses or spaces are included in the folder name passed to the find command, an error results. To make a long story short, here is the patch that fixed it for my installation: --- iso-scan.postinst rev 1298 +++ iso-scan.postinst.patched @@ -165,165, +165,165 @@ - isolist=$(find $dir $opt -name *.iso -o -name *.ISO 2/dev/null) + isolist=$(find $dir $opt -name *.iso -o -name *.ISO 2/dev/null) -- Chris Tillman Developer
Bug#240109: Additional confirmation and info
I have reconfirmed that this bug still occurs using the Official Sarge CD-ROM #1 image built 4/28/2004, downloaded via jigdo. I also tried the same installation using the 4/30/2004 netinst CD from the installer archive folder, and it works as expected. Here are the kernel packages available on CD #1: kernel-package: kernel-package_8.086_all.deb kernel-patch-2.4.20-apus: kernel-headers-2.4.20-apus_2.4.20-1_powerpc.deb kernel-image-2.4.20-apus_2.4.20-1_powerpc.deb kernel-patch-2.4.22-powerpc: kernel-modules-2.4.22-powerpc-small_2.4.22-13_powerpc.deb kernel-modules-2.4.22-powerpc-smp_2.4.22-13_powerpc.deb kernel-modules-2.4.22-powerpc_2.4.22-13_powerpc.deb kernel-patch-2.4.25-apus: kernel-headers-2.4.25-apus_2.4.25-2_powerpc.deb kernel-image-2.4.25-apus_2.4.25-2_powerpc.deb kernel-image-apus_2.4.25-2_powerpc.deb kernel-patch-2.4.25-powerpc: kernel-image-2.4.25-power3-chrp_2.4.25-8_powerpc.deb kernel-image-2.4.25-power4-chrp_2.4.25-8_powerpc.deb kernel-image-2.4.25-powerpc-prep_2.4.25-8_powerpc.deb kernel-image-2.4.25-powerpc-small-pmac_2.4.25-8_powerpc.deb kernel-image-2.4.25-powerpc-smp-prep_2.4.25-8_powerpc.deb kernel-modules-2.4.25-power3_2.4.25-8_powerpc.deb kernel-modules-2.4.25-power4_2.4.25-8_powerpc.deb kernel-modules-2.4.25-powerpc-small_2.4.25-8_powerpc.deb kernel-modules-2.4.25-powerpc-smp_2.4.25-8_powerpc.deb kernel-modules-2.4.25-powerpc_2.4.25-8_powerpc.deb Here are the kernel packages on the netinst CD: kernel-patch-2.4.25-apus: kernel-image-2.4.25-apus_2.4.25-2_powerpc.deb kernel-patch-2.4.25-powerpc: kernel-image-2.4.25-powerpc-chrp-rs6k_2.4.25-8_powerpc.deb kernel-image-2.4.25-powerpc-chrp_2.4.25-8_powerpc.deb kernel-image-2.4.25-powerpc-pmac_2.4.25-8_powerpc.deb kernel-image-2.4.25-powerpc-prep_2.4.25-8_powerpc.deb kernel-image-2.4.25-powerpc-small-pmac_2.4.25-8_powerpc.deb kernel-modules-2.4.25-powerpc-small_2.4.25-8_powerpc.deb kernel-modules-2.4.25-powerpc_2.4.25-8_powerpc.deb Here is the diff between the two: --- netinst.list2004-05-07 20:58:22.0 + +++ sarge-1.list2004-05-07 20:58:15.0 + @@ -1,11 +1,28 @@ +kernel-package: +kernel-package_8.086_all.deb + +kernel-patch-2.4.20-apus: +kernel-headers-2.4.20-apus_2.4.20-1_powerpc.deb +kernel-image-2.4.20-apus_2.4.20-1_powerpc.deb + +kernel-patch-2.4.22-powerpc: +kernel-modules-2.4.22-powerpc-small_2.4.22-13_powerpc.deb +kernel-modules-2.4.22-powerpc-smp_2.4.22-13_powerpc.deb +kernel-modules-2.4.22-powerpc_2.4.22-13_powerpc.deb + kernel-patch-2.4.25-apus: +kernel-headers-2.4.25-apus_2.4.25-2_powerpc.deb kernel-image-2.4.25-apus_2.4.25-2_powerpc.deb +kernel-image-apus_2.4.25-2_powerpc.deb kernel-patch-2.4.25-powerpc: -kernel-image-2.4.25-powerpc-chrp-rs6k_2.4.25-8_powerpc.deb -kernel-image-2.4.25-powerpc-chrp_2.4.25-8_powerpc.deb -kernel-image-2.4.25-powerpc-pmac_2.4.25-8_powerpc.deb +kernel-image-2.4.25-power3-chrp_2.4.25-8_powerpc.deb +kernel-image-2.4.25-power4-chrp_2.4.25-8_powerpc.deb kernel-image-2.4.25-powerpc-prep_2.4.25-8_powerpc.deb kernel-image-2.4.25-powerpc-small-pmac_2.4.25-8_powerpc.deb +kernel-image-2.4.25-powerpc-smp-prep_2.4.25-8_powerpc.deb +kernel-modules-2.4.25-power3_2.4.25-8_powerpc.deb +kernel-modules-2.4.25-power4_2.4.25-8_powerpc.deb kernel-modules-2.4.25-powerpc-small_2.4.25-8_powerpc.deb +kernel-modules-2.4.25-powerpc-smp_2.4.25-8_powerpc.deb kernel-modules-2.4.25-powerpc_2.4.25-8_powerpc.deb The correct kernel for my machine, kernel-image-2.4.25-powerpc-pmac_2.4.25-8_powerpc.deb, is only available on the netinst image. There are really two corrections necessary: 1) CD #1 needs to get up to speed 2) the installer should never install a completely inappropriate kernel Because #2 occurs, I believe the severity of this bug should be moved back to critical. I'm not changing the severity, because I'm not involved with installer development now. But if people around the world try to install Sarge on newworld Macs and their machine becomes a) unbootable and b) unrecoverable with the CD they burned, they are really gonna be pissed. Remember, newworld Macs have no floppies. There is really no recovery path using that machine, if they've formatted the disk; the user would have to burn a better CD on another machine to get his machine back. We should not release an installer that is capable of this. I'll just mention at the end here that the installer is really top notch now, it installed the base system without a hitch and I didn't have one clue that anything was wrong until yaboot installation barfed with 'Failed to mount /proc'. The underlying problem was yabootconfig being unable to find a bootable kernel in /boot. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#247949: main-menu: eject command not available in installer shell
Package: main-menu Version: debian-installer 4/28/2004 Severity: important On powerpc, the eject command is critical to be able to remove CDs from the computer. Many Macs are slot-based, there is no button on the front which will eject the CD. In the system I tested, where an installation had already been completed using Sarge CD #1 built on 4/28/2004, I was unable to eject the CD which had been booted in order to try another CD. The eject command was not available in /bin, /sbin, /usr/bin, or /usr/sbin in the ramdisk system. eject had been installed on the target hard disk, but was not usable. I tried /target/usr/bin/eject from console #2, and it said it couldn't unmount it. I tried unmounting it first, and it gave an ioctl error. I tried chrooting to /target, but then the cdrom mount point (ramdisk /cdrom) was inaccessible. I couldn't eject the disk from the console. This bug is written on main-menu, because boot-floppies had a main menu item which called eject, and I believe that would be an appropriate solution to this issue. By the way, the CD should be ejected by the installer before it reboots into the new system. -- System Information: Debian Release: testing/unstable APT prefers testing APT policy: (990, 'testing') Architecture: powerpc (ppc) -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#240109: install: Installer puts wrong kernel in boot partition [powerpc-newworld]
Package: install Severity: grave Justification: renders package unusable I used the installer downloaded thru jigdo today (3/25/2004) to install to an unused partition on my iMac. It booted fine, ran like a champ, very nice work! It didn't write a bootloader, which I thought was odd but I soon found out why. I booted my old system, took a look at the boot partition on the new system and saw the kernel was an apus kernel. The shortcut was vmlinuz, not vmlinux as it has been in the past, and pointed to the apus vmlinuz kernel which had been installed. I haven't actually tried manually booting this kernel -- it doesn't look right! # cat /proc/cpuinfo cpu : 740/750 temperature : 48-51 C (uncalibrated) clock : 350MHz revision: 2.2 (pvr 0008 0202) bogomips: 697.95 machine : PowerMac2,1 motherboard : PowerMac2,1 MacRISC2 MacRISC Power Macintosh detected as : 66 (iMac FireWire) pmac flags : 0005 L2 cache: 512K unified memory : 128MB pmac-generation : NewWorld -- System Information: Debian Release: testing/unstable APT prefers testing APT policy: (990, 'testing') Architecture: powerpc (ppc) Kernel: Linux 2.4.20-powerpc Locale: LANG=C, LC_CTYPE=C -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#240109: Further info
I was attempting to find a proper kernel on CD #1 which I had downloaded. Here is the list I found: # cd /cdrom/pool/main/k # find . | grep 'kernel-image' ./kernel-image-2.2.10-powerpc-apus ./kernel-image-2.2.10-powerpc-apus/kernel-image-2.2.10-apus_2.2.10-13_powerpc.deb ./kernel-image-2.2.10-powerpc-apus/TRANS.TBL ./kernel-patch-2.4.20-apus/kernel-image-apus_2.4.20-1_powerpc.deb ./kernel-patch-2.4.20-apus/kernel-image-2.4.20-apus_2.4.20-1_powerpc.deb ./kernel-patch-2.4.25-powerpc/kernel-image-2.4.25-powerpc-smp-prep_2.4.25-2_powerpc.deb ./kernel-patch-2.4.25-powerpc/kernel-image-2.4.25-powerpc-small-pmac_2.4.25-2_powerpc.deb ./kernel-patch-2.4.25-powerpc/kernel-image-2.4.25-powerpc-prep_2.4.25-2_powerpc.deb ./kernel-patch-2.4.25-powerpc/kernel-image-2.4.25-power4-chrp_2.4.25-2_powerpc.deb ./kernel-patch-2.4.25-powerpc/kernel-image-2.4.25-power3-chrp_2.4.25-2_powerpc.deb I see there is a kernel-image-2.4.25-powerpc-pmac in the results of apt-cache search, so why didn't it appear on CD 1? In fact, it _does_ appear, as /install/powermac/vmlinux , but not as a package. Did the installer just pick the apus kernel to install because the pmac one wasn't available? It should have thrown its cookies instead. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#240182: After executing a shell, installer tries to resume automatic operation
Package: install Severity: normal After executing a shell in the installer, the installer attempts to pick up after the last step that succeeded. This is counter productive when you just want to go in and copy a file from a to b, and you end up having to say, Go Back, Go Back until it finally gets back to the menu. Since the shell can only be executed from the menu, after exit it should return to the menu. -- System Information: Debian Release: testing/unstable APT prefers testing APT policy: (990, 'testing') Architecture: powerpc (ppc) Kernel: Linux 2.4.20-powerpc Locale: LANG=C, LC_CTYPE=C -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: netcfg debconf templates - please review
Looks good to me ... but then it already did :) On Sat, Dec 13, 2003 at 12:30:32AM +0100, Gaudenz Steinlin wrote: [Please CC me on replies to debian-l10n-english] Hi could someone please review the attached debconf templates. As I'm not a native English speaker there are probably some mistakes. Thanks Gaudenz -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: 1.44 meg boot disk problem
On Mon, Dec 08, 2003 at 10:07:55AM -0500, James Caballero wrote: I downloaded all 20 of the files from the following link: http://ftp.us.debian.org/debian/dists/stable/main/disks-i386/base-images-3.0 .23-2002-07-18/images-1.44/ The installation proceeds fine until I put the Base-18.bin disk in. The installation software informs me that I have put the Base-17.bin disk in, and when I hit enter, I am forced to go back to the beginning of the Base installation section. IMPORTANT: I went back to the link above and made a new floppy, making sure that I downloaded Base-18.bin and went through the entire install again. Again I had the same, exact, problem. Is it possible that the file named Base-18.bin at the above link is mistakenly Base-17.bin? I think it is. No, the base-17 file has 17 of 20 inside it and the 18 one has 18 of 20 (checked with a binary editor). Here are the md5sums I get from the downloaded files: a56b386132512c7e8bdd334d2af2f593 /home/toff/base-17.bin c2d1e669d65a7a4a4b20fbcd5eae4894 /home/toff/base-18.bin -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#222381: installation-reports: Installation Report
On Fri, Nov 28, 2003 at 03:52:09AM -0800, Owner wrote: Package: installation-reports Version: N/A; reported 2003-11-28 Severity: grave Justification: renders package unusable Target: Dell Workstation 400 Dual 9.1 GiB SCSI Hard Drives and Plexor CDR attached to a PCI 2940UW SCSI card 128MiB Ram Single PII @300 MHz Matrox Mill II *The bios does not recognize the scsi as a boot device* This system has only two methods of booting: Floppy or Netboot. 3C905C with PXE Report: Wow, this is a very nice change. I required the 3COM drivers to boot a kernel. With the new installer it was simply a quick job of mounting the initrd, chrooting and doing a udpkg -i net-extras.udeb. Very nice. This sounds like a process that should be documented. Are there other udebs that should or could be handled this way? After things got burning, I found another issue. In my situation, I need to move the initrd as well as the vmlinux to the netboot server. So I tried the default of scp (DOH.. This is an installer), ftp and finally nfs. Shock and amazement, nothing worked. I was saved by doing a chroot to the target and apt-getting ftp. It was very ugly, but it worked. wget is in the installer. It would be nice if, when the user types ftp, the response could be 'Try wget'. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Modconf and 2.6 kernel
On Sat, Nov 29, 2003 at 12:09:34AM +0100, Goswin von Brederlow wrote: Wolf Bergenheim [EMAIL PROTECTED] writes: Hi! Yesterday I hacked modconf to work with my 2.6 kernel. I was just wondering, wether you wouald be interested in a diff? It doesn't work with 2.4 anymore, so I guess it is not that interesting, just an ugly hack... Tha main problem is that the new kernel uses the extension .ko for it's modules... oh well... Didn't someone already do that? I don't think it's being actively maintained, there was no response to the previous post IIRC. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: boot-floppies ; new release for 3.0r2 ?
On Wed, Nov 19, 2003 at 09:32:13PM +, Alastair McKinstry wrote: Hi, There is a fix in the current CVS boot-floppies, which I believe is required for the Arm RiscStation; it sets the keyboard in kbdconfig.c. Can someone confirm that this is needed for RiscStation? Is there going to be a new release of boot-floppies for the upcoming woody release? Regards, Alastair McKinstry I assume not, because there was talk of releasing next week and boot-floppies has to be built by hand on all architectures. Unless it's been built some time ago, but I think not. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Problems with Sarge netinst and Yaboot
On Tue, Nov 18, 2003 at 06:20:53PM +0100, [EMAIL PROTECTED] wrote: Hi, I've just bought a PowerBook G4 12. I have downloaded the Beta1 of Sarge NetInst for PowerPC (November 9th) and I have got a problem at the beginning of the installation. I enter Open Firmware, and then boot with: boot cd:\\yaboot This is not the right command, try boot cd:,\install\powermac\yaboot Then it will locate the yaboot.conf correctly in the same place. And then I get: /[EMAIL PROTECTED]/[EMAIL PROTECTED]/[EMAIL PROTECTED]/[EMAIL PROTECTED]:0,yaboot.conf: Unknown or corrupt filesystem Can't open config file Welcome to yaboot version 1.3.10 Enter help to get some basic usage information boot: Enter Please wait, loading kernel... :0,/vmlinux: Unable to open file, Invalid device -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: cvs commit to debian-installer/main-menu/debian by toff
On Tue, Nov 18, 2003 at 12:40:01PM -0500, Joey Hess wrote:
Chris Tillman wrote:
On Mon, Nov 17, 2003 at 01:07:48PM -0500, Joey Hess wrote:
[EMAIL PROTECTED] wrote:
+- Added ${VERSION} so the version will show in the main menu title
Template: debian-installer/main-menu
Type: select
Choices: ${MENU}
Default: ${DEFAULT}
_Description: Choose the next step in the install process:
+ This is the main menu for the Debian installer, version ${VERSION}.
What sets this VERSION, and to what number? The text implies that it is
the overall version number of the installer, but the closest we
currently have to such a thing is the build date of the installer
images.
--
see shy jo
I saw it in build/debian/rules. I really don't know if it's correctly
set at the time this is built; maybe it is really be putting the
main-menu version in there. I guess we should set a DI_VERSION
variable at the top level build?
Please don't commit stuff to core parts of debian-installer that you
have not even tested to see it if works.
I don't know what top-level build you're taling about either.
--
see shy jo
Sorry, I see it's already been removed.
--
Debian GNU/Linux Operating System
By the People, For the People
Chris Tillman (a people instance)
toff one at cox dot net
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Debconf templates polishing status
On Mon, Nov 17, 2003 at 07:41:06AM +0100, Christian Perrier wrote: Folks, You will find below the status of the debconf templates polishing work for d-i. You will notice that I started to work on 2nd stage packages templates, which are mentioned at the end. It seems that some package (mostly the *-installer for exotic platforms) haven't been released at all...or are less maintained. In fact they aren't even mentioned in doc/translations.txt Sorry for the few french comments in the lines below. They are mostly intended for my own use. Relance means that I send a small mail for asking whether something is planned. Remember that I do not have commit access (mostly for technical reasons...I do not feel comfortable enough with CVS.) and thus need other people to commit the changes. I did commit quite a few of them to cvs last night. I put comments in the change logs so the bugs will be closed automatically when the changes are released. anna ;217018; RESOLVED ;;OK retriever/floppy (floppy-retriever);218207;CLOSED (obsolete);obsolete; tools/aboot-installer ;218208; Not pending 11/13;not tagged d-i, unknown package?; tools/grub-installer ;218209; RESOLVED;; tools/lilo-installer ;218273; Not pending 11/13;relance 11/13; tools/hppa/palo-installer ;218342; Not pending 11/13;relance 11/13; tools/yaboot-installer ;218343; RESOLVED;; tools/sparc/silo-installer ;218567; Not pending 11/13;never uploaded; tools/ia64/elilo-installer ;218568; Not pending 11/13;relance 11/13; tools/mips/arcboot-installer ;218569; RESOLVED;; tools/autopartkit ;218570; RESOLVED;; tools/baseconfig-udeb ;218701; Not pending 11/13;relance+changement 11/13; tools/s390/dasd ;218884; Not pending 11/13;relance 11/13; tools/base-installer ;218992; Not pending 11/13;relance 17/11; tools/cdrom-checker ;219189; RESOLVED;; tools/ddetect ;219469; Not pending 11/13;relance 17/11; tools/kbd-chooser ;219470; RESOLVED;; tools/lvmcfg ;219471; Not pending 11/13;ACK 6/11. Relance 17/11; tools/iso-scan ;219472; Not pending 11/13;relance 17/11; tools/partconf ;219632; Not pending 11/13;relance 17/11; tools/netcfg ;219633; Not pending 11/13;; tools/partitioner ;220026; Not pending 11/13;plus 218912; tools/prebaseconfig ;220030; Not pending 11/13;; utils (di-utils) ;220180; Not pending 11/13;; tools/cdrom-detect ;220180; Not pending 11/13;; main-menu ;220449; Not pending 11/13;; tools/s390/netdevice ;220543; Not pending 11/13;; tools/cdebconf ; NONE ; Already OK;; retriever/choose-mirror ; NONE ; Already OK?;; tools/bugreporter-udeb ; NONE ; Already OK;; base-config;220836;;; shadow;221151 -- -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#221266: 'images-1.44/rescue.bin bug
On Mon, Nov 17, 2003 at 09:13:32AM -0600, Tarun Kundhi wrote: Package: boot-floppies Version: 3.0r1(Disk 2) flavor:vanilla architecture: i386 model: Intel, PIII memory:512 RAM scsi: none cd-rom:TDK, ATAPI network card: Intel on motherboard D815EEA and a Linksys WMP11 pcmcia:none I am attempting to install Debian onto an Western Digital IDE drive (40G). Problem occurs when attempting install the kernel step: Install Kernel and Driver Modules Step 1 - Select the medium to use: cdrom (this is listed) Message: Please place the first Debian CD-ROM in the CD-ROM drive Continue Step 2 - Message: Please select the directory containing a file images-1.44/rescue.bin that you will use to install the kernel and modules. There are two options list: Choose from a list of all likely directories manually: Enter the directory manually Regardless of which method selected and error is returned indicating file is not found. If I allow the program to search for the file the following error message is returned: -- File not found! The installation program couldn't find and directory containing the files rescue.bin, drivers.tgz -- If I try to manually give it the path but it doesn't like it either. I've tried the following: boot boot/ boot/rescue.bin /boot /boot/ /boot/rescue.bin The install program always wants to add /instmnt/ as a prefix. The closest I've come is with 'boot' . The path is then correct but the error message is as follows: -- File not found! /instmnt/boot/ does not contain the file images-1.44/rescue.bin that is needed to install the kernel and modules -- I can't find 'images-1.44/rescue.bin' anywhere on my CD, but I do find 'rescue.bin' under the following path /boot/ It is buried way down in the CD. But if the CD is built correctly, it should find it just by pointing at the CD, because it knows the standard path. For your case, the path is /debian/dists/woody/main/disks-i386/current/images-1.44/rescue.bin This will appear inside the /instmnt directory, because that's where the installer mounts your CD within the ramdisk file system. So, if you're navigating manually, navigate to the 'current' folder; that's what should contain 'images-1.44/rescue.bin'. Searching the web I found reference to a similar bug in 2000 but it appears the problem was solved. There wasn't a work around that I came across. I'm new to linux, if I'm barking up the wrong tree I would be happy to review any documentation that would point me to a solution. So far, I haven't found any. I even thought I a may have pulled down a bad iso image but I've downloaded a second iso image and get the same exact result. Any guidance would be appreciated. There was a bug at one time, but since late 2001 it's been working great. Are you inserting CD #2, not CD #1? CD #2 is only used for _booting_ the vanilla flavor; the kernels are all there on CD #1 though in the right folder structure. Also, I can't find any 'drivers.tgz' file on the iso image. This is another file that is mentioned in the error message if you allow the install program to search for the files itself. The drivers.tgz is in the current folder within the path above. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: cvs commit to debian-installer/main-menu/debian by toff
On Mon, Nov 17, 2003 at 01:07:48PM -0500, Joey Hess wrote:
[EMAIL PROTECTED] wrote:
+- Added ${VERSION} so the version will show in the main menu title
Template: debian-installer/main-menu
Type: select
Choices: ${MENU}
Default: ${DEFAULT}
_Description: Choose the next step in the install process:
+ This is the main menu for the Debian installer, version ${VERSION}.
What sets this VERSION, and to what number? The text implies that it is
the overall version number of the installer, but the closest we
currently have to such a thing is the build date of the installer
images.
--
see shy jo
I saw it in build/debian/rules. I really don't know if it's correctly
set at the time this is built; maybe it is really be putting the
main-menu version in there. I guess we should set a DI_VERSION
variable at the top level build?
--
Debian GNU/Linux Operating System
By the People, For the People
Chris Tillman (a people instance)
toff one at cox dot net
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Realtek RTL8139
On Mon, Nov 17, 2003 at 10:19:39PM -0500, Ed Heaney wrote: I am stooped in setting up linux on a new disk, by a command line request for a Realtek RTL8139 module. I am worn out from searching Google. I am using a Debian GNU/Linux 3.0 r0 Woody system disk. I don't understand. Normally modconf is run to select modules, you would be using a text-driven interface to select some module with 8139 in the name (8139too works well). If you mean what do you type in for parameters? Probably you don't need to type anything. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: powerpc 2.4.22-3 kernel packages ready.
On Fri, Nov 14, 2003 at 10:07:20AM -0800, Matt Kraai wrote: On Fri, Nov 14, 2003 at 10:55:21AM +0100, Sven Luther wrote: I have prepared a new version of the powerpc kernel packages (2.4.22-3). Again, they are still hold up in NEW, until a ftp-master deigns to find time for looking at it or at least provide feedback about what is wrong with them, but i have also made them available in the meantime at : http://people.debian.org/~luther/powerpc/2.4.22 The kernel-image-2.4.22-powerpc-pmac 2.4.22-3 and kernel-modules-2.4.22-powerpc 2.4.22-3 packages work on my iBook2. Installing the kernel package did not require me to upgrade the modules, though, which led to some unsatisfied symbol messages from depmod. Shouldn't kernel-image-2.4.22-powerpc-pmac have a versioned dependency on kernel-modules-2.4.22-powerpc (= 2.4.22-3) Strange. I had noticed after I installed the new kernel, that my boot process was hanging for a few seconds near a line that said insmod parport was failing. I tried modconf, to try to remove that module, but modconf was not on my system. I thought it had been. I reinstalled it, and tried to remove the parport module, but it refused, saying it was in use. After reading this, I first verified that kernel-modules-2.4.22-powerpc was not installed. Then I installed it; it went through the process, but said it was downloading 0 bytes. After it was finished, I re-checked with dpkg -s and it was _still_ not installed. What's going on here? -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#220949: install: Installer overwrites user data without warning
Package: install Version: beta-1 Severity: critical Justification: causes serious data loss Well, this report is perhaps alarmist, but it is true. I was testing whether the installer could be booted manually from the CD on powerpc. I managed to boot it from an HFS partition, where I had placed files from the CD. Once I got the installer booted, I thought I would mount the HFS partition and try removing some of the files from it which I thought were not used, to re-test booting with them removed. Rather than using console 2, I saw the 'Configure and mount partitions' menu item, and chose it (partconf). I selected the hard disk partition I wanted to mount, and told the installer to leave the existing filesystem alone, and mount it on root (the default selection). Well, then I selected Finish, knowing the partition would not really be mounted until I did so. But unfortunately you can imagine what happened next. The installer took off, having its partconf dependency satisfied, and debootstrap started installing debs. I was a little amused, I figured 'OK, I'll have a bunch of unix files on my Mac partition'. it eventually got an error, d-i segfaulted and started respawning, 6000 process numbers used by the time I saw it on console 3, and the shell was almost unusable due to the cpu activity. I did get it rebooted from console 2 eventually. Mounting the disk in MacOS, I see no Unix files. However as I started looking at the Mac files that were there, I realized they had been corrupted. Applications gave error messages when I tried to use them. The file's contents had been overwritten by the debs from debootstrap, using ext2 filesystem semantics I suppose, leaving the HFS desktop file intact but overwriting the files themselves. So this entire partition's contents are trash. I have no idea what might be good or bad. Say, it's a good thing I haven't used MacOS in a long time. But I did have many valuable items there. This disaster could be prevented by refusing to install on an HFS filesystem, or partconf could refuse to mount such filesystems at the standard unix mount points. I think it just points out the kind of dangers we face when we take something that's been manual in the past and automate it. I was still interpreting the Configure and mount partitions step to be like it was in boot-floppies; it would do that and nothing else. Now, it does much more and resulted in the loss of my data. I'm also going to add something in the powerpc boot.msg telling how to set the debconf priority for 'manual mode'. If the priority had been set to medium, I assume the menu would work one step at a time as it did in boot-floppies. -- System Information: Debian Release: testing/unstable Architecture: powerpc Kernel: Linux iMacBlue.cox.net 2.4.22-powerpc #1 Sat Sep 27 04:08:08 CEST 2003 ppc Locale: LANG=C, LC_CTYPE=C -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: dhcp idea
On Fri, Nov 14, 2003 at 12:02:12PM +0100, Gaudenz Steinlin wrote: Am Fre, den 14.11.2003 schrieb Christian Perrier um 06:44: I second your plans. Of course, I hope you'll try to keep as much as possible of the templates tanslations while merging the packages. There is a lot of work in them, though the polishing stuff will fuzzy some strings. atm there are 3 different template files for netcfg (netcfg-common, netcfg-static and netcfg-dhcp). Are the translation for these files in one po-file per language or in 3 different files? What do I have to take care of, when I move these templates to other files? Is it common practice that the one how introduces or changes a template runs debconf-updatepo or is this done by the translators? I'm pretty sure Denis has a script that does debconf-updatepo when you make changes here, except for that the answer would be to always run it after a change. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#220180: cdrom-detect: Debconf templates polishing
) worked is a bit trivial, isn't it? Proofreading needed Much improved. Template: cdrom-detect/no-cd Type: error -_Description: No CD was detected - Please make sure that there is a CD in the CDROM reader. +_Description: No CD detected + Please check that a CD has been inserted in the CD-ROM reader. 1) Title so no phrase required...but is this correct? Yes, your version is a title, the old one is a complete sentence. 2) Minor rephrase Template: cdrom-detect/wrong-cd Type: error -_Description: A CDROM has been detected, but it isn't a Debian CD +_Description: Non-Debian CD-ROM detected + A CDROM has been detected, but it isn't a Debian CD. + . Please make sure that the correct CD is inserted. 1) Title 2) Move the former short desc to extended description Template: cdrom-detect/skip_autodetection Type: boolean Default: false -_Description: Do you want to skip CDROM autodetection ? - You may want to try this if autodetection is causing trouble. Otherwise - you really shouldn't; manual intervention will be required to mount the - CDROM on /cdrom before you can proceed. +_Description: Skip CD-ROM automatic detection? + You should skip the CD-ROM automatic detection only if it is causing trouble. + If you skip it, a manual intervention will be required to mount the + CD-ROM on /cdrom before you can proceed. 1) Removed space and shorten question 2) rephraseproofreading needed I like it, much improved. Template: mirror/distribution Type: select _Choices: woody, sarge, sid Default: sarge -_Description: Select distribution to install - Please select which of the distributions you want to install. +_Description: Distribution branch to install: + Please select which of the Debian distribution branches you want to install: . - woody is the current stable release of Debian - sarge is the testing unreleased version of Debian - sid is unstable and will never be released. + - woody is the current stable release of Debian; + - sarge is the testing unreleased version of Debian; + - sid is unstable and will never be released. Consistency with the similar template in choose-mirror This is the first time I've heard the term 'branch' associated with distributions. Usually each is just termed a 'distribution' which seems right to me; each distribution stands by itself. So, I would remove 'branch' and 'branches', otherwise OK. Template: debian-installer/cdrom-detect/title Type: text # Item in the main menu to select this package -_Description: Detect CDROM devices and mount the CD +_Description: Detect and mount CD-ROM ShorterThis should be enough and easier to translate in 58 chars..:-) -- System Information: Debian Release: testing/unstable Architecture: i386 Kernel: Linux mykerinos 2.4.22 #2 Mon Sep 29 15:12:10 CEST 2003 i686 Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (ignored: LC_ALL set to fr_FR.UTF-8) Sweet! -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#220030: prebaseconfig: Debconf templates polishing
On Mon, Nov 10, 2003 at 07:35:12PM +0100, Christian Perrier wrote: Package: prebaseconfig Severity: normal Well, this one is maybe more important than othersAt least, ALL users of d-i will see it So, please, native English speakers, check my suggestions... Detailed rationale: Template: prebaseconfig/reboot_in_progress Type: note -_Description: The machine will now reboot. - When you reboot the system, you'll need to make sure you are booting +_Description: System reboot + The system will now reboot. + . + Before this reboot occurs, you'll need to make sure you are booting from the right media. . - If you are booting from the local disk, make sure there are no + If you are booting from the local disk, check that there are no floppies in the floppy drives and no CD-ROM in the tray. If you made - a custom boot floppy and want to boot from that, you should put that - floppy in the first floppy drive. Finally, you can restart the - installation system by booting from your install media. + a custom boot floppy and want to boot from it, you should put that + floppy in the first floppy drive. + . + You may also restart the installation system by booting from your + installation media. . Please take care of all that before you continue. 1) Make this look more like a title 2) thus add The system will now reboot as a separated paragraph. system seems better than machine to me... 3) Before this reboot occurs is maybe not the best possible. English speakers needed..:-) 4) s/make sure/check that : look sbetter to me but your mileage may vary 5) boot from it or boot from that? 'it' is correct. Everything looks fine here. I add a paragraph as the suggestion for rebooting from the install(ation) media is another idea and should appear more easily, imho -- System Information: Debian Release: testing/unstable Architecture: i386 Kernel: Linux mykerinos 2.4.22 #2 Mon Sep 29 15:12:10 CEST 2003 i686 Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (ignored: LC_ALL set to fr_FR.UTF-8) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED] -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#219928: debian-installer: build failures on sparc
On Tue, Nov 11, 2003 at 08:52:01PM +0100, Goswin von Brederlow wrote: Joey Hess [EMAIL PROTECTED] writes: Goswin von Brederlow wrote: Then how do you know it is a bug in mklibs and it is not actually a missing symbol? Which was the real reason it failed in every testcase I got my hand on so far. All of those suposedly mklibs weak symbols bugs disapeared when recompiling the faulty binaries against the proper versions of libraries so far. If I read between the lines right, this time it was breakage in brltty on sparc, and I've removed it from the pkg-lists for sparc now. Is any arch still using brltty? Do we care about it? Yes, Mario Lang invested effort on getting it to work with debian-installer; without it blind users cannot install. Does brltty break on i386, m68k or alpha too? Don't know. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Flavors
On Sun, Nov 09, 2003 at 10:59:51PM -0500, Joey Hess wrote: Chris Tillman wrote: In the old installation manual, there were four (no, 5) flavors of kernels for i386. In the current kernel module, only one kernel version is listed. Can we dump all the instructions for users about choosing flavors on i386? (Please say yes). The installer chooses a flavor based on the hardware, so this should not need a great deal of consideration from users. Well, the whole point of the existing discussion is to choose a CD that will boot on one's hardware. All machines will now boot from CD 1? -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Flavors
In the old installation manual, there were four (no, 5) flavors of kernels for i386. In the current kernel module, only one kernel version is listed. Can we dump all the instructions for users about choosing flavors on i386? (Please say yes). -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Installing Debian GNU/Linux 3.0 For Intel x86 - Xwindows config?
On Fri, Nov 07, 2003 at 05:17:39PM -0600, Michael Hamblin wrote: Hey folks, I'm looking around on the Debian website but I'm finding nothing about configuring X during or post-installation. When I try to startx or do anything to configure it, the system complains it can't find /usr/X11R6/bin/X ... even after I reinstalled it, and selected desktop in tasksel. Other X programs seem to be installed, but why not this one? Starting gdm manually starts some sort of configure script that (naturally) fails with similiar results. Maybe I'm just really stupid, but what in the world am I missing here? I don't see it in the docs. XWindows config is a science all its own, due to the great variety of hardware out there. But to start, dpkg-reconfigure xserver-sfree86 Then if that doesn't do it, search at xfree86.org, and post on a list there, or on debian-user. This list is actually for development of the installer, and X configuration is beyond its scope. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#219471: lvmcfg: Debconf templates rewrite
On Thu, Nov 06, 2003 at 08:08:49PM +0100, Christian Perrier wrote:
Package: lvmcfg
Severity: normal
Tags: patch
Template: lvmcfg/activevg
Type: boolean
Default: true
_Description: Activate existing Volume Groups?
Already existing Volume Groups have been found (count: ${COUNT}). Please
${COUNT} existing Volume Groups have been found. Please
indicate whether you want to activate them.
[...]
Template: lvmcfg/vgcreate_nosel
Type: error
_Description: No physical volumes selected
No physical volumes were selected. The creation of a new
Volume Group is aborted.
Volume Group was aborted.
[...]
Template: lvmcfg/vgextend_nosel
Type: error
_Description: No physical volumes selected
No physical volumes were selected. Extension of the Volume Group will be
No physical volumes were selected. Extension of the Volume Group was
aborted.
[...]
Template: lvmcfg/nopartitions
Type: error
_Description: No usable physical volumes found
No physical volume (e.g. partition) were found in your system. All
No physical volumes (i.e. partitions) were found in your system. All
physical volumes may already be in use. You may also need to load
some required kernel modules or re-partition the hard drives.
--
Debian GNU/Linux Operating System
By the People, For the People
Chris Tillman (a people instance)
toff one at cox dot net
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#219633: netcfg: Debconf templates polishing
On Fri, Nov 07, 2003 at 06:46:26PM +0100, Christian Perrier wrote:
Package: netcfg
Severity: normal
Tags: patch d-i
[...]
Template: netcfg/get_nameservers
Type: string
-_Description: Choose the DNS Server Addresses
+_Description: Name servers addresses:
Please enter the IP addresses (not host names) of up to 3 name servers,
separated by spaces. Do not use commas. The servers will be queried in the
- order in which you enter them. If you don't want to use any name servers
+ order you mentioned. If you don't want to use any name server,
just leave this field blank.
I have to disagree with that one. I understand you, but the first
way is much clearer. Or how about:
The first server in the list will be the first to be queried.
1) --prompt
2) looks better to me
Template: netcfg/choose_interface
Type: select
Choices: ${ifchoices}
-_Description: Choose an interface.
- The following interfaces were detected. Choose the type of your primary
- network interface that you will need for installing the Debian system (via
- NFS or HTTP).
+_Description: Interface:
+ Please choose the interface you want to configure. Below are shown
+ all currently detected interfaces on your system. Please choose the
Please choose the interface you want to configure. All currently
detected interfaces on your system are shown below. Please choose the
+ type of your primary network interface. This interface will be used
+ for installing the Debian system (via NFS or HTTP).
1) --prompt
2) Reformulation for more clarity
[...]
Template: netcfg/dhcp_hostname
Type: string
-_Description: What is your dhcp hostname?
- You may need to supply a DHCP host name. If you are using
- a cable modem, you might need to specify an account number
- here. Most other users can just leave this blank.
+_Description: DHCP hostname:
+ Supplying a DHCP host name is sometimes needed. If you are using
+ a cable modem, you might need to specify an account number here.
+ .
+ Most other users can just leave this blank.
Again, short desc--prompt
I reformulate the extended description for improving readability and make it
clearer that most people will leave the field blank.
I think splitting it up is an improvement. But the first sentence is
better in the original, because it is in active voice instead of passive.
Template: netcfg/confirm_dhcp
Type: boolean
Default: true
-_Description: Is this information correct?
+_Description: Are these informations correct?
+ Currently configured network parameters:
+ .
interface = ${interface}
hostname = ${hostname}
domain = ${domain}
The original style looked very bad with the new cdebconf layout.
Above is a proposalthis may be enhanced. The two added strings are
identical with the similar ones in netcfg/confirm_static.
Information is plural; fact is the singular. Informations is not
really an English word. So the original is correct for the prompt.
Template: netcfg/no_dhcp_client
-Type: note
-_Description: No dhcp client found. I cannot continue.
- This package requires pump or dhcp-client.
+Type: error
+_Description: No DHCP client found
+ No DHCP client was found. This package requires pump or dhcp-client.
+ .
+ The DHCP configuration process will be aborted.
That one also was somewhat incorrect:
1) s/note/error
2) make the short desc a real title
3) explain things in the extended desc (and forget about I)
Template: netcfg/dhcp-title
Type: text
@@ -36,4 +41,4 @@
Template: debian-installer/netcfg-dhcp/title
Type: text
# Item in the main menu to select this package
-_Description: Configure the network via DHCP
+_Description: Configure the network using dynamic addressing (DHCP)
Well, THIS IS A PROPOSAL. This may be more clear for newbies (maybe someday,
Debian will at least be recommended to newbies by Linux for dummies-type
books...:-))
However, this won't make the translators work easy, maybe.
Well, I like it, FWIW.
[...]
Template: netcfg/confirm_static
Type: boolean
Default: true
-_Description: Is this configuration correct?
+_Description: Are these informations correct?
+ Currently configured network parameters:
+ .
interface = ${interface}
hostname = ${hostname}
domain= ${domain}
See above for the similar template in DHCP configuration
Is this information correct? can be re-used.
--
Debian GNU/Linux Operating System
By the People, For the People
Chris Tillman (a people instance)
toff one at cox dot net
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#219469: ddetect: Debconf templates polishing
On Thu, Nov 06, 2003 at 05:34:22PM +0100, Christian Perrier wrote:
Package: ddetect
Severity: normal
Tags: patch
As usual, the following changes will make debconf templates more consistent
with other modules templates.
Detailed rationale:
diff -Nru ddetect/debian/ethdetect.templates ddetect.new/debian/ethdetect.templates
--- ddetect/debian/ethdetect.templates2003-11-06 17:10:22.0 +0100
+++ ddetect.new/debian/ethdetect.templates2003-11-06 07:10:07.0 +0100
@@ -2,34 +2,33 @@
Type: select
_Choices: ${CHOICES}, none of the above
Default: none of the above
-_Description: What module does your ethernet card require?
- This is a list of modules that I know about. Choose the module from the
- list that supports your card. If none of the modules do, you may need to
- load additional modules from a driver floppy; choose none of the above.
+_Description: Module needed by your ethernet card:
+ Please choose amon the known modules which one supports your card. If
+ none of the modules do, you may need to load additional modules from
+ a driver floppy. Then choose the none of the above option.
Please choose the module for your card from the list of known
modules. If the correct module is not listed, you may need to load
additional modules from a driver floppy. In that case, choose the
option which indicates that your module is not listed.
--
Debian GNU/Linux Operating System
By the People, For the People
Chris Tillman (a people instance)
toff one at cox dot net
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#219472: iso-scan: Debconf templates polishing
On Thu, Nov 06, 2003 at 06:30:53PM +0100, Christian Perrier wrote:
Package: iso-scan
Severity: normal
Tags: patch
Template: iso-scan/bad-isos
Type: error
-_Description: Failed to find an installer ISO
+_Description: Failed to find an installer ISO image
Of ${ISO_COUNT} possible ISO images found, ${ISO_MOUNT_FAILURES} could
- not be mounted, and an installer ISO was not found. Perhaps the ISO you
- downloaded is corrupt? You'll have to do a network install instead, or
- reboot and fix the ISO.
+ not be mounted, and an installer ISO image was not found. The ISO
+ image you downloaded is maybe corrupted.
+ .
+ Please do a network install instead, or reboot and fix the ISO image.
... image you downloaded may be corrupted.
--
Debian GNU/Linux Operating System
By the People, For the People
Chris Tillman (a people instance)
toff one at cox dot net
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#219189: cdrom-checker: Debconf templates polishing
On Tue, Nov 04, 2003 at 11:00:24PM +0100, Christian Perrier wrote:
Package: cdrom-checker
Severity: normal
Tags: patch
[...]
Template: cdrom-checker/passed
Type: note
-_Description: Integrity test successful!
- The integrity for the CD-ROM was successful. The CD-ROM looks good!
+_Description: Integrity test successful
+ The integrity for the CD-ROM was successful. The CD-ROM looks correct.
The CD-ROM integrity test was successful. The CD-ROM is valid.
Just facts. The check was successful, not need for yelling, this should be
normal after all..:-)
Looks good vs. looks correct : sounds better to me, but I may be wrong. I
want to avoid trivial language.
Template: cdrom-checker/mismatch
-Type: note
-_Description: Integrity test failed!
- The following file has a md5sum mismatch: ${FILE}. Your CD-ROM or
+Type: error
+_Description: Integrity test failed
+ The ${FILE} file failed the MD5 checkum verification. Your CD-ROM or
this file may be corrupt.
checksum
Error again
No yelling again, just facts..:)
The extended description is reworded. But I'm not sure about
verificationprobably too french..:-)
Excellent, IMHO.
corrupt or corrupted??
I like corrupt better with be. Alternatively, may have been corrupted.
Template: cdrom-checker/nextcd
Type: boolean
Default: false
-_Description: Test another CD-ROM
- Would you like to test another CD-ROM?
+_Description: Check another CD-ROM integrity?
The only problem is if you move CD-ROM to the left, it now
has to be possessive: CD-ROM's integrity?
Probably better: Check the integrity of another CD-ROM?
We always checked the CD-ROM, so let's continueExtended description is
just repeating the short one, so let's drop it.
Template: cdrom-checker/firstcd
Type: note
@@ -62,4 +61,4 @@
Template: debian-installer/cdrom-checker/title
Type: text
# Main menu item
-_Description: Verify the cd contents
+_Description: Check the CD-ROM(s) integrity
Check is used internally, so use it in the Main Menu entry also.
Same for CD-ROM and integrity (the latter may be dropped...)
good idea to drop integrity, because of the possessive issue.
I also suggest adding (s) as we allow for checking several CD's.
-- System Information:
Debian Release: testing/unstable
Architecture: i386
Kernel: Linux mykerinos 2.4.22 #2 Mon Sep 29 15:12:10 CEST 2003 i686
Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (ignored: LC_ALL set to fr_FR.UTF-8)
--
Debian GNU/Linux Operating System
By the People, For the People
Chris Tillman (a people instance)
toff one at cox dot net
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#218992: base-installer: Debconf templates polishing
On Mon, Nov 03, 2003 at 07:07:45PM +0100, Christian Perrier wrote:
Package: base-installer
Severity: normal
Tags: patch
Yet another polishing bug report..:-)
Template: kernel-installer/which-kernel
Type: select
Choices: ${KERNELS}
-_Description: Select which kernel to install into /target/
- These are the kernels available from the APT sources. One of them
- needs to be installed to be able to boot the system from hard drive.
+_Description: Kernel to install into /target/:
+ The list shows the kernels available from the APT sources. Please choose
+ one of them in order to make to system bootable from the hard drive.
Consistency with the Style Guide document. Maybe better english is possible?
... make the system bootable ...
--
Debian GNU/Linux Operating System
By the People, For the People
Chris Tillman (a people instance)
toff one at cox dot net
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#218690: partitioner: Unable to start mac-fdisk
Package: partitioner Version: 0.10 Severity: important File: partitioner I was using the Halloween version of the powerpc netinst CD. Partitioner offered the right path to use for my disc, but failed to start the partitioning program. It stated the correct path in the error message too. On console 3, there was only: Nov 2 04:44:17 (none) user.info partitioner.postinst[1569]: INFO: Using architecture depend fdisk configuration! I retried, and the error repeated except now at [1598]. I switched to console 2 and accomplished partitioning with mac-fdisk /dev/ide/host0/bus0/target0/lun0 I suspect it has to do with recent archdetect changes. -- System Information: Debian Release: 3.0 Architecture: powerpc Kernel: Linux ip68-107-137-100.tc.ph.cox.net 2.4.22-powerpc #1 Sat Sep 27 04:08:08 CEST 2003 ppc Locale: LANG=C, LC_CTYPE=C -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#218689: partitioner: Unable to start mac-fdisk
Package: partitioner Version: 0.10 Severity: important File: partitioner I was using the Halloween version of the powerpc netinst CD. When it got to the partitioner step, it correctly listed the only drive connected to the computer. When I selected Continue, it failed, confirming the correct device name in the failure message. there was only an INFO message in console 4: Nov 2 04:44:17 (none) user.info partitioner.postinst[1569]: INFO: Using architecture depend fdisk configuration! When I retried the step, I got the same info, but at [1598]. I suspect the arch-detect thingy is not working right. I was able to partition the disk by going into console 2 and typing mac-fdisk /dev/ide/host0/bus0/target0/lun0/disc (the same path that was displayed in console 1). -- System Information: Debian Release: 3.0 Architecture: powerpc Kernel: Linux ip68-107-137-100.tc.ph.cox.net 2.4.22-powerpc #1 Sat Sep 27 04:08:08 CEST 2003 ppc Locale: LANG=C, LC_CTYPE=C -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#218597: should display a notice when changing the debconf priority level
On Fri, Oct 31, 2003 at 09:35:25PM -0500, Joey Hess wrote: It can also be confusing when things start to go right again, and main-menu whisks you away to high priority land. I think it should display an error when something fails and it lowers the priority, and should display a note before jumping to a higher priority. I couldn't agree more. This would definitely help to avoid the panicky the 'Captain! She's out of control!' feeling. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [WOODY] Need sugestion for Floppy-Install
On Sun, Nov 02, 2003 at 05:04:35PM +0100, Michelle Konzack wrote: Hello, I am working since 03/1999 with Debian successfuly. For some month I have lost my last Windows for Workgroups 3.11 Workstation (I have never had Win32) by a lightning-stroke and now I am M$-Free. I am working on a very difficult Job in near east ans need the Floppy-Installation. For SLINK 2.1r5 it was no problem... 7 Base-Floppys and 73 Packages Floppys. Now I have tried it with WOODY but it does not work because dpkg-scanpackages... The Version of SLINK was supporting the X-Medium Flag and if I have registerd all Floppys apt-get was asking about the numbered Floppy. Now in WOODY it does not work anymore... OK, I can register all Floppys, but when I install somthing, it tell me to put the Floppy in... I habe 140 Floppys - Where ist the *.deb ? The Other thing I have tried is to make the whole Directory tree, a Packages- and Release-File and add the 'X-Medium: ' manualy to it. Unfortunately the 'X-Medium:' is not recognised by WOODY. Question: = Is there a solution like the CD-Rom Installation which ask for a numberd Medium ? or must I port dpkg-scanpackages for WOODY ? Maybe staticaly linked ??`? I was thinking to make for each Floppy a Packages- and Release-File and put all 280 Files in a Package which installs in /var/lib/apt/lists So no one must make the DJ after installing the BASE to register the other 114 Floppys... And I thought you were _kidding_. Wouldn't it be easier to set up an old hard drive with the stuff you need and hook it up to the secondary IDE? Much faster than reading 114 floppies, one at a time. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#218690: Processed: reopen
On Sun, Nov 02, 2003 at 09:25:33PM +0100, Thorsten Sauter wrote: * Debian Bug Tracking System [EMAIL PROTECTED] [2003-11-02 16:33]: | Processing commands for [EMAIL PROTECTED]: | | reopen 218690 | Bug#218690: partitioner: Unable to start mac-fdisk | Bug#218689: partitioner: Unable to start mac-fdisk | Bug reopened, originator not changed. why you have reopend this bug? Please include a short explanation next time. I did, did it not show up? It is a duplicate of 218689. I closed it when i saw I had submitted twice. Then I saw it had been merged with 218689, a better solution; but I had just closed the merged bug. So I reopened it. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [ppc] netinstall image from gluck report
On Thu, Oct 30, 2003 at 10:40:13AM +0100, Arnaud Vandyck wrote: On Wed, 29 Oct 2003 20:51:22 +0100 Thorsten Sauter [EMAIL PROTECTED] wrote: the partitioner doesn't start? no question about the retriever type? Maybe you should try to install with a lower priority. Set DEBCONF_PRIORITY=normal on the openfirmware prompt. I'll try this but how will be the string? DEBCONF_PRIORITY=normal boot cd:,install\powermac\yaboot No, boot cd:,install\powermac\yaboot Then at the boot: prompt, install DEBCONF_PRIORITY=normal -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: sarge netinst with PCMCIA ethernet fails to install modules
On Thu, Oct 30, 2003 at 04:12:29PM -0800, Brian Dessent wrote: Joey Hess wrote: Brian Dessent wrote: Hello, I'm trying to use the sarge netinst mini-CD to install Debian on a laptop that has a PCMCIA cardbus ethernet adaptor. The problem that I'm having is that while the first stage installer can recognise and activate the PCMCIA ethernet card, it fails to install the kernel modules for it, so that after rebooting and loading the new kernel from the HD, the card is not recognised and the network install cannot continue. Our pcmcia support is still very incomplete, so I'm glad to hear it worked at all. The problem with pcmcia not being installed is a known bug, #214492, though it may not include the detail below: The base system that gets installed contains the pcmcia-cs package and the kernel-image-2.4.22-1-386 package (obviously), but NOT the pcmcia-modules-2.4.22-1-386 package. I plan to reogranise the pcmcia support so it fits in better with the rest of the installer, but I may not have a chance to get around to doing this for a while. Well, that's actually a relief to hear as I thought it was something that I was doing wrong. Does the woody installer suffer from the same problem, or is that a different beast? I was thinking of trying a woody netinst and then a dist-upgrade. But if that's known to have issues then I'll just bite the bullet and download the full ISOs with jigdo. AFAIK pcmcia works in woody. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Minimal powerpc kernel, miboot users please test (Was Re: modularized powerpc kernel)
On Wed, Oct 29, 2003 at 08:17:34AM +0100, Sven Luther wrote: On Tue, Oct 28, 2003 at 08:41:18PM -0700, Chris Tillman wrote: On Fri, Oct 24, 2003 at 11:42:22AM +0200, Gaudenz Steinlin wrote: Am Fre, den 24.10.2003 schrieb Chris Tillman um 08:15: On Thu, Oct 23, 2003 at 06:14:29PM +0200, Sven Luther wrote: Aha! I got a working kernel. I guess I know more than I think I do :) Here is the diff for the config, from your -small config. I really don't know which of these were essential, I ended up with only 59k free space on the floppy. But I guess that's enough. For now it is. Altough I think most of the options you enabled are not necessary. Could you test, if the keyboard also works for loading the initrd when you only enable these two settings: -CONFIG_INPUT_KEYBDEV=m -CONFIG_INPUT_MOUSEDEV=m +CONFIG_INPUT_KEYBDEV=y +CONFIG_INPUT_MOUSEDEV=y I tried several times to make kernels witha smaller set of options, but they all failed to boot, or else panicked after booting. I guess I got very lucky the first time. I'm sure we must not need IDE and SCSI drivers, but I didn't have any luck with the them left out. I still had no luck with the 9500, either. The 2.2.20 kernel boots, but all 2.4 kernels give the same result: the video is not switched on in Linux. Many of the 2.4 kernels boot, I can type in my root login and 'reboot' blind, and the computer reboots. But they don't drive video. The 9500 is different from many Macs in that its video board is not built-in, but instead hangs off the PCI bus. This may have something to do with the problem. Can you post a lspci output when running the 2.2.20 kernel ? This sounds suspiciously like a missing fbdev driver for the graphic board. It says it is a Mach64GX, and I have CONFIG_FB_ATY_GX compiled in to all the ones I've tried; I assume that's the one. 00:0b.0 Host bridge: Apple Computer Inc. Bandit PowerPC host bridge (rev 02) 00:0f.0 VGA comppatible controller: ATI Technolgies Inc 210888GX [Mach64 GX] (rev 02) 00:10.0 Class ff00: Apple Computer Inc. Grand Central I/O (rev 02) -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: custom task with bf
On Wed, Oct 29, 2003 at 07:32:04PM +0300, Andrei Smirnov wrote: i made an altered tasksel with dpkg-repack and included it in a mirror, then made a cd. but when i try to install the base from it, dbootstrap says: couldn't download tasksel while debootstrap in a local dir(without bf) works from this cd can anyone help? debootstrap compares the MD5 signature. Did you generate a new MD5sums file also? -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Minimal powerpc kernel, miboot users please test (Was Re: modularized powerpc kernel)
On Fri, Oct 24, 2003 at 11:42:22AM +0200, Gaudenz Steinlin wrote: Am Fre, den 24.10.2003 schrieb Chris Tillman um 08:15: On Thu, Oct 23, 2003 at 06:14:29PM +0200, Sven Luther wrote: Aha! I got a working kernel. I guess I know more than I think I do :) Here is the diff for the config, from your -small config. I really don't know which of these were essential, I ended up with only 59k free space on the floppy. But I guess that's enough. For now it is. Altough I think most of the options you enabled are not necessary. Could you test, if the keyboard also works for loading the initrd when you only enable these two settings: -CONFIG_INPUT_KEYBDEV=m -CONFIG_INPUT_MOUSEDEV=m +CONFIG_INPUT_KEYBDEV=y +CONFIG_INPUT_MOUSEDEV=y I tried several times to make kernels witha smaller set of options, but they all failed to boot, or else panicked after booting. I guess I got very lucky the first time. I'm sure we must not need IDE and SCSI drivers, but I didn't have any luck with the them left out. I still had no luck with the 9500, either. The 2.2.20 kernel boots, but all 2.4 kernels give the same result: the video is not switched on in Linux. Many of the 2.4 kernels boot, I can type in my root login and 'reboot' blind, and the computer reboots. But they don't drive video. The 9500 is different from many Macs in that its video board is not built-in, but instead hangs off the PCI bus. This may have something to do with the problem. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Bug#217503: Evil autopartkit should _NEVER_ _NEVER_ try to overwrite an unknown partition table
On Tue, Oct 28, 2003 at 07:18:30PM +0100, Sven Luther wrote: On Tue, Oct 28, 2003 at 06:55:35PM +0100, Gaudenz Steinlin wrote: I still think that the real problem is in the modularity of debian-installer. It is all nice to have a bunch of modules be used and have each provide a menu entry, but how do you guarantee structure and coherence of the entries ? I would like that someone (maybe me, but i am maybe not the best placed for it) made a analysis of what is really supposed to happen in the install process, and how the tasks are supposed to follow each others. I can make a begining of UML diagrams for it, if you like, altough i am not really aware of all the debain-installer details. This would also be fantastic for install manual writing purposes. I find it very difficult to understand, much less explain, how all the pieces fit together. I'm also attempting to understand the myriad of targets and images that will be available, we'll have to explain to users how to choose the one they need. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#217775: choose-mirror: Debconf templates polishing
On Mon, Oct 27, 2003 at 01:30:37PM +0100, Matthias Urlichs wrote: Hi, IMHO better (most people don't know about network topology and we shouldn't make them feel insecure if they don't): _Description: Use mirror from: Please select a Debian mirror. You should use a mirror in your country or region if you do not know which mirror has the best Internet connection to you. Replacing the as with a semicolon sounds better to me: _Description: Protocol for files download: Please select the file download protocol. If unsure, select http; it is less prone to problems involving firewalls. I second these. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#217775: choose-mirror: Debconf templates polishing
On Mon, Oct 27, 2003 at 11:30:45AM +0100, Christian Perrier wrote: Package: choose-mirror Severity: normal Tags: patch d-i [...] Template: mirror/http/proxy Type: string -_Description: Enter http proxy information, or leave blank for none: - If you need to use a http proxy to access the outside world, enter the +_Description: HTTP proxy information (blank for none): + If you need to use a HTTP proxy to access the outside world, please enter the proxy information here. Otherwise, leave this blank. . - When entering proxy information, use the standard form of - http://[[user][:[EMAIL PROTECTED]:port]/ + The proxy information should be the standard form of + http://[[user][:[EMAIL PROTECTED]:port]/. The proxy information should be given in the standard form of Would it be OK to incorporate small corrections like this when applying the patch? In general I think what you're doing is great and will make the installer much more user friendly. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Minimal powerpc kernel, miboot users please test
On Sat, Oct 25, 2003 at 10:41:13PM +0200, Sven Luther wrote: On Fri, Oct 24, 2003 at 07:53:56AM -0700, Chris Tillman wrote: The reduced kernel does not seem to boot on the 9500. It produces the small penguin icon and loads from the floppy, then blanks the screen and stops. What are you trying to load from the floppy ? I am speaking about the boot floppy. I can hear the kernel being read from it (this process takes a few minutes) as the penguin icon is displayed, and then when one would normally begin to see boot messages and an opportuinty to insert a root floppy, everything stops. I'll try to adjust some modules based on what works on the 9500 from 2.2.20. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Minimal powerpc kernel, miboot users please test (Was Re: modularized powerpc kernel)
On Thu, Oct 23, 2003 at 06:14:29PM +0200, Sven Luther wrote: Mmm, could you look at the /boot/config-2.4.22-powerpc file, and try to adapt it, and build a kernel which does work for you ? Even if this means adding some stuff, we have to isolate where the problem is. I will upload 2.4.22-2 as is though, as it will need to pass the NEW queue anyway, and we can fix the powerpc-small config in a future -3 release. Aha! I got a working kernel. I guess I know more than I think I do :) Here is the diff for the config, from your -small config. I really don't know which of these were essential, I ended up with only 59k free space on the floppy. But I guess that's enough. Also, whoever can work on it to make it part of the d-i build: I put the script to generate the floppy miBoot image (stolen from boot-floppies), and the necessary pieces, plus the resulting floppy image, at http://members.cox.net/toff1/miBoot.tgz The floppy boots and loads the woody root.bin on my clone machine. I'll try to test on a 9500 later if I get a chance. I couldn't test the sarge root.bin yet, the one I had handy (from a last week's CD) was too big for a floppy. --- boot/config-2.4.22-powerpc-smallWed Oct 22 10:43:36 2003 +++ miBoot/config Thu Oct 23 22:52:29 2003 @@ -90,7 +90,7 @@ # CONFIG_PARPORT_SUNBPP is not set # CONFIG_PARPORT_OTHER is not set CONFIG_PARPORT_1284=y -CONFIG_GEN_RTC=m +CONFIG_GEN_RTC=y CONFIG_PPC601_SYNC_FIX=y CONFIG_PROC_DEVICETREE=y CONFIG_PPC_RTAS=y @@ -369,11 +369,11 @@ # # CONFIG_BLK_DEV_HD_IDE is not set # CONFIG_BLK_DEV_HD is not set -CONFIG_BLK_DEV_IDEDISK=m +CONFIG_BLK_DEV_IDEDISK=y # CONFIG_IDEDISK_MULTI_MODE is not set # CONFIG_IDEDISK_STROKE is not set CONFIG_BLK_DEV_IDECS=m -CONFIG_BLK_DEV_IDECD=m +CONFIG_BLK_DEV_IDECD=y CONFIG_BLK_DEV_IDETAPE=m CONFIG_BLK_DEV_IDEFLOPPY=m CONFIG_BLK_DEV_IDESCSI=m @@ -442,12 +442,12 @@ # # SCSI support # -CONFIG_SCSI=m +CONFIG_SCSI=y # # SCSI support type (disk, tape, CD-ROM) # -CONFIG_BLK_DEV_SD=m +CONFIG_BLK_DEV_SD=y CONFIG_SD_EXTRA_DEVS=40 CONFIG_CHR_DEV_ST=m CONFIG_CHR_DEV_OSST=m @@ -558,10 +558,10 @@ CONFIG_SCSI_U14_34F_MAX_TAGS=8 # CONFIG_SCSI_NSP32 is not set CONFIG_SCSI_DEBUG=m -CONFIG_SCSI_MESH=m +CONFIG_SCSI_MESH=y CONFIG_SCSI_MESH_SYNC_RATE=5 CONFIG_SCSI_MESH_RESET_DELAY_MS=4000 -CONFIG_SCSI_MAC53C94=m +CONFIG_SCSI_MAC53C94=y # # PCMCIA SCSI adapter support @@ -626,7 +626,7 @@ # Ethernet (10 or 100Mbit) # CONFIG_NET_ETHERNET=y -CONFIG_MACE=m +CONFIG_MACE=y # CONFIG_MACE_AAUI_PORT is not set CONFIG_BMAC=m CONFIG_GMAC=m @@ -1105,8 +1105,8 @@ # Input core support # CONFIG_INPUT=y -CONFIG_INPUT_KEYBDEV=m -CONFIG_INPUT_MOUSEDEV=m +CONFIG_INPUT_KEYBDEV=y +CONFIG_INPUT_MOUSEDEV=y CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024 CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768 CONFIG_INPUT_JOYDEV=m @@ -1264,7 +1264,7 @@ # CONFIG_AMD7XX_TCO is not set # CONFIG_SCx200_GPIO is not set # CONFIG_AMD_PM768 is not set -CONFIG_NVRAM=m +CONFIG_NVRAM=y # CONFIG_RTC is not set CONFIG_DTLK=m CONFIG_R3964=m @@ -1352,7 +1352,7 @@ # CONFIG_ADFS_FS_RW is not set CONFIG_AFFS_FS=m CONFIG_ASFS_FS=m -CONFIG_HFS_FS=m +CONFIG_HFS_FS=y CONFIG_HFSPLUS_FS=m CONFIG_BEFS_FS=m # CONFIG_BEFS_DEBUG is not set @@ -1753,7 +1753,7 @@ # # CONFIG_CRC32 is not set CONFIG_ZLIB_INFLATE=y -CONFIG_ZLIB_DEFLATE=m +CONFIG_ZLIB_DEFLATE=y # # Kernel hacking -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: HOWTO install Debian from another distribution?
On Fri, Oct 24, 2003 at 12:51:43PM +0200, Arnaud Vandyck wrote: [Note: arch=new powermac] [Sorry for crosspost, I don't know which list is more appropriated] Hi all, I do have some problems to boot with the new debian-installer images. I did also try woody but it fails to recognize my harddisk! I got a GentooLive for powerpc that boots ok. I could partition my hd and now I have yaboot configured (some linux and osX). I did copy the gentoo kernel, modules and modules.conf and told yaboot to boot on it. I did build debootstrap from another powerpc and launched it. It installed some packages but I do not have 'ar'! nor a 'functionnal' Debian system! :( I did not recall the program that is launched at the first installation. Also, is it the good way to install Debian from another distribution? Do you have some more informations about it? What can I launch to re-install debian? how can I build it? Many thanks for some links. Now that I have a running linux on the powerbook, I can try to compile some stuff to install... I assume you are already referring to http://www.debian.org/releases/stable/installmanual for installing via debootstrap? The program to run after booting or chrooting into Debian is /sbin/base-config. Then run tasksel to install major pieces of software, and aptitude to pick individual packages. If you have dpkg running, but it's not registered, yo can use dpkg --install dpkg The same logic works with apt-get --reinstall install apt-get -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Second round of powerpc subarch investigation : boot-loaders.
On Fri, Oct 24, 2003 at 02:07:19PM +0200, Sven Luther wrote: On Fri, Oct 24, 2003 at 01:26:24PM +0200, Freek Dijkstra wrote: Hi Sven, I recently noted a post stating that the bootloader development team did not have access to an oldword powerpc. I have one of those, and can do some testing this weekend, even if it renders my openfirmware settings obsolete. I'm afraid I'm not an expert, but I got some real great help on this list earlier, and am eager to do something in return. So just mail me privately if there is any testing I can do (I've got a 6500 with internal + external monitor card, internal IDE disk + 2 external SCSI disks. I'm afraid I don't get a OpenFirmware prompt though, but I can read it with the BootVars program or nvsetenv). It maybe too early, we have a kernel now which should be able to be mibootable, but not yet the full debian-installer stuff. I will make a new kernel build this afternoon, maybe you can try this one out, and see if you can boot it. I hope once this kernel reaches the archives, we can try building d-i images also or something such, or you could do them by hand. I don't really know how miboot works, so i can't do it myself. You could certainly see if the hfs-bootfloppy.bin image within http://members.cox.net/miBoot.tgz boots. It should ask for a root disk, at which point you'll just have to reboot for now. We'll need help testing the debian-installer very soon! Thanks for your contributions. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Minimal powerpc kernel, miboot users please test
The reduced kernel does not seem to boot on the 9500. It produces the small penguin icon and loads from the floppy, then blanks the screen and stops. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Minimal powerpc kernel, miboot users please test (Was Re: modularized powerpc kernel)
On Thu, Oct 23, 2003 at 07:54:37AM +0200, Goswin von Brederlow wrote: Chris Tillman [EMAIL PROTECTED] writes: On Thu, Oct 23, 2003 at 05:14:21AM +0200, Goswin von Brederlow wrote: Chris Tillman [EMAIL PROTECTED] writes: However, it didn't get much farther when I tried a kernel that did have drivers ... it read the partitions OK, but failed soon after You are trying to read the ramdisk of /dev/ram which can't work. root=/dev/fd0 load_ramdisk=1 would load from floppy respectively. Ah. I tried using a floppy ramdisk with the regular image, and it did load it up. But Sven's image still didn't work. It prompts for the floppy disk, but doesn't continue after I press Enter. I seem to remember a similar problem cropping up with 2.2.17, when powerpc was a young'n. Maybe something to do with the adb keyboard? Does return work at all? It can well be the keyboard is modular like usb on i386. I tried every key on the keyboard, the kernel wasn't listening. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Minimal powerpc kernel, miboot users please test (Was Re: modularized powerpc kernel)
On Thu, Oct 23, 2003 at 11:03:46AM +0200, Sven Luther wrote: On Wed, Oct 22, 2003 at 09:01:25PM -0700, Chris Tillman wrote: On Thu, Oct 23, 2003 at 05:14:21AM +0200, Goswin von Brederlow wrote: Chris Tillman [EMAIL PROTECTED] writes: However, it didn't get much farther when I tried a kernel that did have drivers ... it read the partitions OK, but failed soon after You are trying to read the ramdisk of /dev/ram which can't work. root=/dev/fd0 load_ramdisk=1 would load from floppy respectively. Ah. I tried using a floppy ramdisk with the regular image, and it did load it up. But Sven's image still didn't work. It prompts for the floppy disk, but doesn't continue after I press Enter. I seem to remember a similar problem cropping up with 2.2.17, when powerpc was a young'n. Maybe something to do with the adb keyboard? Well, another problem could be that the initrd is broken, like seems to be the case with the initrd i built the other day from Gowsin's tree. And i guess you need to build the initrd with the right modules or something. Gaudenz can you comment on this ? The initrd I was using was from woody boot-floppies, as I mentioned the regular kernel did load it. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Minimal powerpc kernel, miboot users please test (Was Re: modularized powerpc kernel)
On Thu, Oct 23, 2003 at 11:44:38AM +0200, Gaudenz Steinlin wrote: Sven Luther schrieb: On Wed, Oct 22, 2003 at 09:01:25PM -0700, Chris Tillman wrote: On Thu, Oct 23, 2003 at 05:14:21AM +0200, Goswin von Brederlow wrote: Chris Tillman [EMAIL PROTECTED] writes: Ah. I tried using a floppy ramdisk with the regular image, and it did load it up. But Sven's image still didn't work. It prompts for the floppy disk, but doesn't continue after I press Enter. I seem to remember a similar problem cropping up with 2.2.17, when powerpc was a young'n. Maybe something to do with the adb keyboard? Well, another problem could be that the initrd is broken, like seems to be the case with the initrd i built the other day from Gowsin's tree. And i guess you need to build the initrd with the right modules or something. Gaudenz can you comment on this ? I will try: ADB: The config I sent Sven supports ADB keyboards through the input core (and sends linux keycodes). This works on my powerbook g4. Altough That sounds right. I don't have this special prompt, because I'm not booting from a floppy. Does the kernel detect your keyboard in the boot messages? It doesn't respond to Shift-PgUp either, but this message happened to be just before the root floppy prompt: Detected ADB keyboard, type ANSI. And then it goes on to list the HID IDs for the keyboard and mouse. I'm just not familiar with kernel drivers and configs at all, to know which ones to tell you to turn on. But it definitely needs ide if it's going to need disk access, and some keyboard help of some kind ... -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Minimal powerpc kernel, miboot users please test (Was Re: modularized powerpc kernel)
On Thu, Oct 23, 2003 at 08:08:27PM +0200, Sven Luther wrote: On Thu, Oct 23, 2003 at 06:14:29PM +0200, Sven Luther wrote: I will upload 2.4.22-2 as is though, as it will need to pass the NEW queue anyway, and we can fix the powerpc-small config in a future -3 release. The build failed because of some minor error in the control file, i will have to build again tomorrow, since there is no way i will be able to make the dinstall run of this evening. So there is still time to send me a working .config file for powerpc-small. Could you have a look at this this evening. I'll see, not too likely to succeed I'm afraid. Any ideas what options might get me around the ramdisk prompt where it doesn't accept the Enter key? -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: new build/debian directory
On Thu, Oct 23, 2003 at 06:18:48PM -0400, Joey Hess wrote: [EMAIL PROTECTED] wrote: It fails to build on PowerPC as follows: for udeb in extraudebs/*.udeb ; do \ dpkg --force-overwrite --root=./tmp/floppy/tree/driver-tmp --unpack $udeb; \ done dpkg: error processing extraudebs/*.udeb (--unpack): cannot access archive: No such file or directory Errors were encountered while processing: extraudebs/*.udeb make[2]: *** [floppy-tree-stamp] Error 1 make[2]: Leaving directory `/tmp/build' make[1]: *** [all_images] Error 2 make[1]: Leaving directory `/tmp/build' make: *** [build-stamp] Error 2 This is also not something I have changed in the build tree, nor can I reproduce it (with dash). I suppose the change I just checked in will fix it though. I got a similar error in the same place, I assume after the update. For me, dpkg was complaining about udebs/*.deb (--unpack) instead of extraudebs. Shortly after that, the build stops when it can't find ./tmp/bootfloppy/tree/boot/vmlinux . (Also on powerpc) -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: first nearly successful USB media install
On Wed, Oct 22, 2003 at 10:40:14AM +0200, Martin Sj?gren wrote: ons 2003-10-22 klockan 05.42 skrev Chris Tillman: On Tue, Oct 21, 2003 at 08:56:39PM -0400, Joey Hess wrote: In cfdisk, I told it to change my 128 mb swap partition to ext2, and write and quit. For some reason the next screen (picking partitions to mount?) still identified it as type swap. I went back and checked, it was ext2 according to fdisk. Still swap as far as d-i knew, but I went ahead and told it to mount that as the root partition. Lacking any other partitions, I continued with the install. I think it identifies it as swap because it's looking at the filesystem that's present on the partition, not the partition type. Exactly. Are the partition types really useful anyway? It seeems like a reasonable behavior to me, it's just different. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Section 3.7.5.1 of Install Manual
On Wed, Oct 22, 2003 at 04:28:49PM +1000, Coffey, James A wrote: Hi, I had some difficulties trying to install a kernel using your crossover procedure. The fix I found was that (at least in Red Hat Linux) proc should be mounted with a filesytem of none NOT proc. Once I did this the kernel installed and everything was good. Thanks for the document though, other than this hiccup everything worked first time. Thanks, I added the tip. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: problems with debian
On Wed, Oct 22, 2003 at 10:39:20AM +, david lorenzo wrote: Hello, Iam trying to install debian from a cd i got from a magazine but i am not very good at linux so i dont know if it is me who is not installing it properly. The problem is that i get to install it and when i choose what to install with tasksel i cannot pick any X-Window-System from the menu and so i think i am not installing properly any desktop environment and it never starts with gnome or kde though i know i have configure gnome2 (i upgraded it from another cd). The further i get is that debian tries to run gnome (after upgrading to gnome2) but it says something like X-Window is not configure, do you want to configure now? I pick yes but it doesnt do anything. Another problem i had during installation is that the locales in the cd (i guess they are not very well), i cannot choose anyone of them. So that when i finished installation my keyboard doesnt work properly and debian keeps saying all the time keyboard unknown scancode or something like that. I really appreciate your help because i wrote to the magazine telling them the problems and they havent helped. Thankyou very much for your time. david Please write to [EMAIL PROTECTED] instead. This list is for problems with the installer and development of the new installer, and you are actually beyond the point of system installation. One thing, though: to reconfigure X, the command is dpkg-reconfigure xserver-xfree86 -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [new-powermac] boot on hd but install does not recognize hd!
On Wed, Oct 22, 2003 at 06:49:23PM +0200, Arnaud Vandyck wrote: Hi! I did follow some recommendations and boot on a hfs partition on the harddisk [powerbook 14 with Hitachi IC25N080 ATMR04-0 Media] by copiing files from gluck. Booting OK (with openfirmware). Start the installation OK. But impossible to partition the harddisk!? Does someone know if I have to pass some parameters to the kernel or so? Why does yaboot can load the kernel from the hd, then the kernel loads and start the installer but the installer can't see the harddisk? Thanks for your time, This is woody, right? Do you have a CD in the drive when you boot? There's some bug in the woody installer that prevents it from seeing the hard drive on powerpc, if there is a CD in the drive when booting. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Minimal powerpc kernel, miboot users please test (Was Re: modularized powerpc kernel)
On Wed, Oct 22, 2003 at 11:15:03PM +0200, Sven Luther wrote: On Wed, Oct 22, 2003 at 12:04:36AM +0200, Gaudenz Steinlin wrote: Hi Sven Please have a look at the two packages in : http://people.debian.org/~luther/gaudenz. One is the : kernel-image-2.4.22-powerpc-small-pmac_2.4.22 package, which contain both the vmlinux and the .coff kernel, the other is the : kernel-modules-2.4.22-powerpc-small_2.4.22 package whihc contain the modules built for the above kernel. Try them out, and if you manage to boot them with miboot or something such, then i will make a full build tomorrow again, and upload those, altough they will be hold in NEW for some time. I tried booting this kernel with no success. I tried with miboot and BootX, and both of them seem to get stuck. The video never changes from the last Mac screen after it reboots. I also tried with quik, and there it booted up to the point: NET4: Unix domain sockets 1.0/SMP for Linux NET4.0 adb: starting probe task... EXT2-fs: unable to read superblock cramfs: wrong magic Kernel panic: VFS: Unable to mount root fs on 01:00 I was using root=/dev/ram partition=6 ramdisk=/root.bin append=video=ofonly load_ramdisk=1 ramdisk-size=8192 I guess this is because I'm asking the kernel to read root.bin off the disk partition, and it probably doesn't have the ide drivers compiled in. However, it didn't get much farther when I tried a kernel that did have drivers ... it read the partitions OK, but failed soon after usb.c: registered new driver wacom set_blocksize: b_count 1, dev ramdisk(1,0), block 1, from c0067bd8 Kernel panic: VFS: Unable to mount root fs on 01:00 I suppose this is just because quik doesn't load the ram up ahead of time? -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Proposal for low-mem cdrom install and minimum kernel versions thereof
On Tue, Oct 21, 2003 at 05:32:42AM +0200, Goswin von Brederlow wrote: Hi, debian-installer currently needs a lot of ram to install all the udebs on the cdrom. That was one reason for me to suggest mixing debix and debian-installer together into a live-cd install. So here is how it could work: - 1. A small initrd is booted with enough stuff on it to find and mount the cdrom. 2. On the cdrom a huge initrd image is stored with all the udebs installed that anna would pull in of a cdrom anyway. 3. The huge initrd is loopback mounted and a lvm snapshot device is set over that with a ramdisk/tmpfs as copy-on-write storage. 4. mount the lvm snapshot on /mnt 5. transfere debconf database and other state files to /mnt 6. pivot-root over to the lvm snapshot and exec+chroot main-menu 7. continue as normal What have we won? - The 20-40 MB ramdisk that would normaly be used will be kept on cdrom and only changes to that get stored to ramdisk on a block-by-block basis. No additional udebs would be installed into ram by anna unless choosen manually. The only big file left would be the Packages file which can probably also be preseeded on the cdrom. Apart from the ram, time is also saved. Installing all the udebs from cdrom takes a while on low-mem systems. Consider that an extra bonus. What needs to be changed for this? -- - The device-mapper patch (kernel-patch-device-mapper.deb) would have to be included in a kernel image and the device-mapper tools (dmsetup) packaged as udeb. - When using tmpfs as storage 2.4.22 is the minimum kernel version. The benefit of tmpfs would be swapping capabilities once swap is initialized. Given this is for low-mem that might be nice. Using a ramdisk the lowest version is 2.4.19 I think. - rootskel-low-mem and busybox-low-mem versions would be needed - A pkglist and medialist per architecture. Two if there should be a cdrom wth and without base debs. But they would be just the cdrom-udebs/cdrom-base lists with the important udebs moved into the initrd. Could even be generated by the Makefile so only one list has to be changed in future. Think about it and let me know what you think. Try an d-i install with 64MB, 32MB or even 16 MB ram. This sounds like a great idea to me. I am adding a low-memory test to my list, to see how the current installer does. There are a lot of older machines out there that we're leaving behind if we require 32MB. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: IBM 43P-150 installation
On Tue, Oct 21, 2003 at 07:42:23AM +0200, Sven Luther wrote: On Mon, Oct 20, 2003 at 04:31:08PM -0500, Rolf Brudeseth wrote: Sven, I found this initrd image (9/20/2003) that I am using for now: http://www.soziologie.ch/users/steinlin/d-i/netboot-initrd.gz However, I am having problems NFS mounting the image, something that I have done successfully with my own kernels. Did you enable root NFS? If not, could you please recompile the kernel with this enabled? This is what I added to my kernel to get it to work: CONFIG_NFS_V3=y CONFIG_ROOT_NFS=y Mmm, making the kernel bigger, not smaller :(( BTW, the quik patch that Ben sent to the list takes away our quik size worries. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [ppc] unable to install on new powerbook14
On Tue, Oct 21, 2003 at 11:10:41AM +0200, Arnaud Vandyck wrote: Hi all, First I've burn the image from yesterday and it's still the same problem. I think it's because the automatic build scripts are broken. But (idiot), I do download it and burn it. But, I have a more precise error (I did write it on a paper): modprobe failed to load module usbkbd, keybdev, usb-ohci and usbserial But I WANT DEBIAN ON MY LAPTOP! So I did download the woody image... I've been unable to boot with kernel 2.2: ... ok opening display /[EMAIL PROTECTED]/ATY, [EMAIL PROTECTED]/ATY,[EMAIL PROTECTED] And when booting with install-safe or install24*, he does not find my hard drive! :( Do I have to pass special argument to the kernel during the boot (and how)? For newer Apple hardware, you need to use the newest yaboot version. You can manually boot the installer from the newer version of yaboot placed on your hard disk, along with a modified yaboot.conf, and linux.bin and root.bin. Details of this method of booting are in the installation manual, and the newer version of yaboot is available at penguinppc.org. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: PowerPC CHRP network install
On Tue, Oct 21, 2003 at 11:26:04PM +0200, Gaudenz Steinlin wrote: Am Die, den 21.10.2003 schrieb Rolf Brudeseth um 22:08: I downloaded the debian-installer via cvs and read the documentation. It appears to indicate that there are two supported netinst methods; however, both require a CD? No, you dont' need to burn the image on a CD. Actually, it's very difficult to use the current CD image without a CD. It makes the assumption that the CD is there, and the last time I tried, I couldn't fool it into thinking the hard disk was where it wanted to look. We need to have options available for mirrors on the hard disk. Also, the yaboot.conf is set up for CD usage. We will have to tell the user to modify it for hard disk. Maybe it would be better to have a separate d-i.tgz target intended to be unpacked on the hard disk. On a PowerPC CHRP system, with the boot-floppies installer, I was able to boot the zImage through BOOTP and load the initrd with NFS, and boot and load the zImage.initrd via BOOTP only. I don't know if anyone tried NFS installs, but they should work in theory. zImage.initrd or zImage plus a separate initrd (if this is possible on your arch) over bootp/dhcp/tftp/whatsoever works. We definitely need testing in this area. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Second round of powerpc subarch investigation : boot-loaders.
On Sun, Oct 19, 2003 at 11:29:38AM +0200, Sven Luther wrote: Hello, In order to continue work on making debian-installer subarch friendly, i now launch a second round of investigation concerning the way different subarches boot. The situation as i understand it is : newpmac: uses yaboot (uncompressed kernel + separate initrd) or OF (.coff kernel, don't know about initrd). Debian-installer will boot with yaboot, and the CDrom can be made to be auto-bootable, altough i don't know how. The bootability of the CD is taken care of by mkisofs. It gives the CD an HFS 'blessing' which makes it bootable, and it boots according to the ofboot.b script which yaboot supplies. We have tried to avoid getting users into OF in the past, probably because it's hard to provide Forth-based support in an installation manual. I don't know how .coff booting works either. oldpmac: uses bootx, miboot, quik or serial console OF. OF uses the .coff kernel, don't know about initrd. miboot uses floopy for kernel (compressed) and initrd. bootx can use a kernel and separate initrd, and quik cannot be used for cdrom boot. It is said that it is possible to use miboot to cdrom boot or ssomething such, but i have not confirmed these rumors. debian-installer will be installable from either bootx (uncompressed kernel + separate initrd) altough we will not provide bootx on the CD, or miboot (two or more floppy images). We do provide BootX in the archive now, it should still be on the CD. But it's there as a Stuffit archive; it's not usable directly from the CD and must be installed on the user's hard disk. Someone said mkisofs could put BootX on the CD as a double-clickable MacOS application, but I'm skeptical. If it could, then the kernels could also be placed in the same folder with BootX, and BootX might find them. 3) We build the kernel as part of the debian-installer build system, and can thus use a specific configuration, possibly even many of them, one of them being specialized for miboot and its 1.3Mo kernel size requirement, and build the kernel udebs from that. This is the solution that prep used in boot-floppies. It was problematic to be building kernels within the installer build. I don't have any better ideas though. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: busybox vs busybox-cvs?
On Fri, Oct 17, 2003 at 10:52:09AM +0200, Wouter Verhelst wrote: Op vr 17-10-2003, om 10:07 schreef Bastian Blank: On Thu, Oct 16, 2003 at 08:25:57PM -0600, Erik Andersen wrote: go away I think I must have missed the rest of this conversation... he criticise many of the fundamental changes i implement within the last months. So what? That's probably because he's concerned about debian-installer, and wants it to become a better thing than boot-floppies (no, I'm not saying you don't want that). If you implement numerous fundamental changes without consulting this mailinglist, the error is yours, not his. If he gives feedback on changes you made, you shouldn't see them as critique, but as peer review. That's an important part of the free/open software world. Joey has been helping intensively for the last few weeks, (thanks!) while Bastian has been helping actively for the past several months (thanks, too!). I thought it was probably a little off the mark to lob quit checking in broken code out there. Which of us has never made a mistake? I wonder how many commits Bastian has made on d-i vs. Joey. Also, Joey wasn't here for the thorough bloat removal that Bastian just did with libd-i. So his remarks about bloat, coupled with the inference that it's somehow Bastian's intention to have bloated code (and hey! we're talking about ONE k, folks!) just rubbed Bastian the wrong way. And, since I don't do any coding, just help with the list, I wouldn't even have known about all the effort Bastian put in regarding libd-i except for reading about functions that were getting thrown out, on the list. Why am I writing this? Because I've heard enough from Bastian to know that he knows very little English, and for that reason if no other, I think it's not fair to browbeat him on the list. Just stick to what's wrong with the code, not what's wrong with the coder. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Install report
On Thu, Oct 16, 2003 at 09:54:52AM +0200, Martin Sj?gren wrote: tor 2003-10-16 klockan 07.28 skrev Goswin von Brederlow: When something completes successfully main-menu seems to assume you correct a previous error and resumes automatic install, which just selects the next choice. That combined with the bug of languagechoose failing and then allways being the next choice gives you that effect. Ouch, it doesn't remember WHAT was failing? That's bad... I experienced this, too. It feels like the installer is 'pushing' you along. If everything is going well, this automatic next-step picking is wonderful; it just zooms along and does its thing. But if there has been some problem along the way which hasn't been solved, it should always drop back to the menu and just take orders from the user. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Guide for Oldworld Powermac d-i test using BootX and netinst CD
On Thu, Oct 16, 2003 at 03:59:08PM +0200, Goswin von Brederlow wrote: Sven Luther [EMAIL PROTECTED] writes: On Wed, Oct 15, 2003 at 10:32:35PM -0700, Chris Tillman wrote: I tested d-i using BootX. Greate. Your destined to test disk boots for powerpc then. You make 'em, I break 'em. Cool. Tell me again though, there are some oldworld pmacs which can't do this and need miboot, right ? 1) Burn a CD from http://gluck.debian.org/cdimage/testing/netinst/powerpc/ (I used the sarge-powerpc-netinst iso). Joeyhs loopback detection udeb _will_be_ helpfull there. Just dump the image to harddisk and hope it finds it. :) I'll need to figure out how this works. 2) Boot into MacOS, insert the CD, and copy linux.bin and root.bin from /install/powermac/ to your Linux Kernels folder in the System Folder. Of course, if you already have files there, you might want to rename these. Can they be left on the cdrom? 3) Start BootX. Set the Options to use a root disk, and Browse to the root.bin you just copied in. Make sure the no-video-driver box is _not_ checked (at least on my box). Add the boot argument: And this configured to use them from cdrom? I tried browsing to the root.bin on the CD, and yes that did work. But BootX constructs a menu of kernels from files placed in the Linux Kernels folder, so the kernel has to be copied. modules also :) ... as it was, I just udpkg -i bla.udeb That's what I tried, but AFAICT it installed stuff into my ramdisk, not the target. And it wasn't successful doing modules debs at all. I wasn't sure if maybe I had to --configure after -i. cp -R /lib/modules /target/lib/modules Ah, the joys of busybox with autocompletion, history, and cp -R !! Yes. I loved that too. 7) Use reboot in console 2 to quit the installer. Didn't I hear something about a reboot menu item being created now? Yes, it's coming. 8) There is no quik-installer yet, but you can boot your installed system by using BootX again. Un-check the root disk option, remove init=/linuxrc, and type in the designation of your root partition such as hda6 in the space provided. I actually got it booted; my bogus modules didn't load, no automatic base-config yet; but we're getting there! Can the CDrom contain an icon that will start the installation when clicked? Can BootX be on the CD or is that a system component allways available? No, I don't think so. Well, maybe. The CD is an hfs/iso, so in theory one could put BootX on it (both resource fork and data fork). But in practice, it might not work well. The first issue that arises is: what kind of Linux script can build a CD with the proper MacOS resource fork for the app? There are some ancient apps around that can translate from macbinary, but when I've tried them they don't work well. Also, the BootX preferences are saved on the Mac partition, and the Linux Kernels folder should go in the System Folder ... it's probably best, especially considering the relatively small audience, to just continue to have BootX.sit on the CD and have people unpack it on their MacOS partition. Also keep in mind we should have a floppy-based 'plug-it-in-and-go' solution. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Install report
On Thu, Oct 16, 2003 at 07:46:54PM +0200, Goswin von Brederlow wrote: Joey Hess [EMAIL PROTECTED] writes: Fraser Campbell wrote: Progress bar does not move when downloading packages file. Again my screen goes blank, on tty2 I can see that debootstrap is running though so I wait ... eventually a progress bar does reappear. Package retrieval froze at mawk, no progress seemed to be happening, I killed a few wget and debootstrap processes until I got back to the main menu and then restarted base system install. Second try got past mawk without incident, odd but didn't feel like debugging that one too much. Packages installed, didn't have much trouble then until the bootloader. Hmm, what kind of network uplink do you have? Packages file should download quite quickly, and it's hard to segment it to give progress. I am on dialup and it still downloads quickly. The hang you experienced also rather points to a network problem. Debootstrap fetches the Release file and Packages file and then hangs while probably verifying and parsing the file. Yes, this part has always been slow, since debootstrap was invented. It is running the 'pkgdetails' script for each package in Packages. I looked once to see if some progress could be implemented, but it's subsidiary to debootstrap itself so I couldn't figure it out. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Guide for Oldworld Powermac d-i test using BootX and netinst CD
On Thu, Oct 16, 2003 at 07:33:36PM +0200, Gaudenz Steinlin wrote: Goswin von Brederlow schrieb: Sven Luther [EMAIL PROTECTED] writes: On Wed, Oct 15, 2003 at 10:32:35PM -0700, Chris Tillman wrote: 6) Test installer! The incarnation I used couldn't install the kernel and modules because of the missing ppcdetect. I got around this in console 2 by copying the kernel from /cdrom/install/powermac to /target/boot. If I knew more about udpkg, I could probably install modules also :) ... as it was, I just udpkg -i bla.udeb You fixed this bug in CVS ? ;-) Don't you remeber? udpkg is not meant to install packages into target. Only to install and configure udebs. Use the dpkg on /target for that purpose (or apt-install). So THAT's what they're talking about when they refer to short-term memory loss. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Sarge netinst on Pismo
On Wed, Oct 15, 2003 at 11:57:50AM +0200, Uwe Steinmann wrote: On Wed, Oct 15, 2003 at 11:05:47AM +0200, Gaudenz Steinlin wrote: Uwe Steinmann schrieb: Hi, I just got the latest (14 Oct) sarge installer for powerpc and booted a Apple PowerBook Pismo. It doesn't get far and starts flashing the lcd when it trieds to load some kernel modules. So it booted and you got to the language question first. That's at least something. :-) No, there was no question at all. It failed earlier. All I can see is an error message about a missing usb-ohci kernel module. There are other modules (mostly usb related) missing as well but usb-ohci is the first one. Strange. These modules should not be loaded because usb is compiled into the kernel. The seems to repeat for ever. So you are never thrown back to the main-menu? When some step is failing it should thorw you back to the main-menu where you can select to try it again or to do another step first. As I said. I didn't get as far as a main-menu. What do the logs on tty3 and tty4 say? You can also use a shell on tty2 to debug what's going on. I could switch to tty2 and saw the message about presssing a key to get a shell, but I couldn't see what happened afterwards because of the flashing lcd. I guess the key press had no effect. Sounds like it's segfaulting and restarting? You might be able to stop the flashing using Ctrl-S or Ctrl-C and then switch to console 2 or 3? -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Guide for Oldworld Powermac d-i test using BootX and netinst CD
I tested d-i using BootX. 1) Burn a CD from http://gluck.debian.org/cdimage/testing/netinst/powerpc/ (I used the sarge-powerpc-netinst iso). 2) Boot into MacOS, insert the CD, and copy linux.bin and root.bin from /install/powermac/ to your Linux Kernels folder in the System Folder. Of course, if you already have files there, you might want to rename these. 3) Start BootX. Set the Options to use a root disk, and Browse to the root.bin you just copied in. Make sure the no-video-driver box is _not_ checked (at least on my box). Add the boot argument: init=/linuxrc 4) Save preferences, if you want, and push the Linux button. 5) When the machine reboots, it ejects the CD. The next screen you see after the boot arguments is the Language Chooser; before choosing a language, re-insert the CD. This will let the CD detection code detect it automatically. If you don't insert it now, you will get an opportunity later when it can't find a CD. 6) Test installer! The incarnation I used couldn't install the kernel and modules because of the missing ppcdetect. I got around this in console 2 by copying the kernel from /cdrom/install/powermac to /target/boot. If I knew more about udpkg, I could probably install modules also :) ... as it was, I just cp -R /lib/modules /target/lib/modules Ah, the joys of busybox with autocompletion, history, and cp -R !! 7) Use reboot in console 2 to quit the installer. 8) There is no quik-installer yet, but you can boot your installed system by using BootX again. Un-check the root disk option, remove init=/linuxrc, and type in the designation of your root partition such as hda6 in the space provided. I actually got it booted; my bogus modules didn't load, no automatic base-config yet; but we're getting there! -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#216045: partconf: No column headings for partition selection table
Package: partconf Version: 0.08 Severity: minor File: partconf In my test on powerpc, there were no headings shown for the displayed columns. Headings would be nice. -- System Information: Debian Release: 3.0 Architecture: powerpc Kernel: Linux ip68-107-137-100.tc.ph.cox.net 2.4.22-powerpc #1 Sat Oct 4 08:20:43 CEST 2003 ppc Locale: LANG=C, LC_CTYPE=C -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Proposed patch: ppcdetect - archdetect
Could someone review this patch? I tested the build image yesterday, and one of the problems it ran into was not finding /usr/bin/ppcdetect. I remembered archdetect had replaced ppcdetect, this looks like what is needed to complete the transition for powerpc. Index: tools/base-installer/debian/changelog === RCS file: /cvs/debian-boot/debian-installer/tools/base-installer/debian/changelog,v retrieving revision 1.158 diff -u -r1.158 changelog --- tools/base-installer/debian/changelog 14 Oct 2003 08:52:27 - 1.158 +++ tools/base-installer/debian/changelog 14 Oct 2003 14:24:54 - @@ -7,6 +7,8 @@ * Alastair McKinstry - Add versioned depends on libdebconfclient-dev, to get working debconf_ macros. + * Chris Tillman +- change ppcdetect to archdetect and corresponding CASEs -- Alastair McKinstry [EMAIL PROTECTED] Sun, 12 Oct 2003 22:01:30 +0100 Index: tools/base-installer/debian/kernel-installer.postinst === RCS file: /cvs/debian-boot/debian-installer/tools/base-installer/debian/kernel-installer.postinst,v retrieving revision 1.31 diff -u -r1.31 kernel-installer.postinst --- tools/base-installer/debian/kernel-installer.postinst 7 Oct 2003 01:35:17 - 1.31 +++ tools/base-installer/debian/kernel-installer.postinst 14 Oct 2003 14:24:54 - @@ -116,22 +116,26 @@ ;; powerpc) version=2.4.22 #hack, hack! - MODEL=`/usr/bin/ppcdetect` + MODEL=`/usr/bin/archdetect` case $MODEL in - NewWorld) + powerpc/powermac_newworld) kernel=kernel-image-$version-powerpc ;; - # what's below is probably nonsene, porters have to start building 2.4 kernels for d-i! + # what's below is probably nonsense, porters have to start building 2.4 kernels for d-i! # But at least the infrastructure will be there and these are the latest versions in # the archive. - OldWorld) + powerpc/powermac_oldworld) kernel=kernel-image-$version-powerpc ;; - PReP) + powerpc/prep) kernel=kernel-image-2.2.20-prep ;; - CHRP) + # do we need a special case here for powerpc/chrp_pegasos? + powerpc/chrp) kernel=kernel-image-2.2.20-chrp + ;; + powerpc/amiga) + kernel=kernel-image-2.4.20-apus ;; *) log warning: Unknown powerpc subarchitecture '$MODEL'. Index: tools/partitioner/debian/changelog === RCS file: /cvs/debian-boot/debian-installer/tools/partitioner/debian/changelog,v retrieving revision 1.44 diff -u -r1.44 changelog --- tools/partitioner/debian/changelog 13 Oct 2003 21:05:26 - 1.44 +++ tools/partitioner/debian/changelog 14 Oct 2003 14:24:55 - @@ -7,6 +7,8 @@ - Update pt_BR (Brazilian Portuguese) translation. * Denis Barbier - Add a comment in templates file to flag the main menu item. + * Chris Tillman +- change ppcdetect to archdetect and corresponding CASEs -- Alastair McKinstry [EMAIL PROTECTED] Sun, 12 Oct 2003 22:07:19 +0100 Index: tools/partitioner/scripts/powerpc.sh === RCS file: /cvs/debian-boot/debian-installer/tools/partitioner/scripts/powerpc.sh,v retrieving revision 1.5 diff -u -r1.5 powerpc.sh --- tools/partitioner/scripts/powerpc.sh1 Oct 2003 15:35:14 - 1.5 +++ tools/partitioner/scripts/powerpc.sh14 Oct 2003 14:24:55 - @@ -2,14 +2,23 @@ DISK=$1 [ -z $DISK ] exit 1 -case `/usr/bin/ppcdetect` in -NewWorld PowerMac|OldWorld PowerMac) +case `/usr/bin/archdetect` in +powerpc/powermac_newworld) mac-fdisk $DISK ;; -CHRP Pegasos) +powerpc/powermac_oldworld) + mac-fdisk $DISK + ;; +powerpc/chrp_pegasos) parted $DISK ;; -Amiga) +powerpc/chrp) + fdisk $DISK + ;; +powerpc/prep) + fdisk $DISK + ;; +powerpc/amiga) parted $DISK ;; *) Index: tools/yaboot-installer/debian/isinstallable === RCS file: /cvs/debian-boot/debian-installer/tools/yaboot-installer/debian/isinstallable,v retrieving revision 1.3 diff -u -r1.3 isinstallable --- tools/yaboot-installer/debian/isinstallable 27 Sep 2003 11:12:59 - 1.3 +++ tools/yaboot-installer/debian/isinstallable 14 Oct 2003 14:24:56 - @@ -1,5 +1,5 @@ #!/bin/sh -/usr/bin/ppcdetect | grep -q NewWorld PowerMac +/usr/bin/archdetect | grep -q powermac_newworld exit $? Index: tools/yaboot-installer/debian/changelog === RCS file: /cvs/debian-boot/debian-installer/tools
Install notes
I have to go to work in a few minutes, but I just wanted to pass on a couple things I found in a powerpc install attempt yesterday. * In the partitioner display, there is no column for the names of the partitions. Or maybe the issue is that for powerpc, the names are not detected properly. Column headings would be very nice too. * Also in the partitioner, when I selected partition 10 or 11 to apply a filesystem or swap, the changes were made in the displayed table to partition 1 instead. If those changes were actually applied to partition 1 on a powerpc, it would wipe out the partition table and then you would lose everything on the disk!!! * When switching to console 2, then back to console 1, the display is not refreshed in console 1. So it looks like you haven't switched and can't get back. If you start hitting Tab, buttons show up over the console 2 text. * Maybe related: when I tried the Execute a Shell menu item, the text overwrote the existing screen, and CR's were not displayed so lines came out in a warerfall. * After installation of extra packages completes, the next item offered is Automatic partitioning! I guess it is because of the dependency map, probably there should be nothing depending on automatic partitioning. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: anna's hard-coded priority list of retreivers
On Mon, Oct 13, 2003 at 08:27:47PM +0200, Goswin von Brederlow wrote: Since in your case the detecting of a network card comes up empty neither configuration nor retriever should be triggered. I somehow see this unwanted, unneeded and harmfull triggering of other menu items when selecting something in main-menu as a big problem. I guess there are some bugs in there or the code is broken by design. Like main-menu configuring the kernel-modules when you select execute a shell, or your case (might not neccessarily related but feel the same). Its awfully confusing for the user that selecting A will do some realy wild things B, C, D, E and fail (which is why you selected A in the first place). I agree, it's not good to surprise the user. If we want to warn them that the dependent item has not been configured, and offer to go there, that would be better. Execute a Shell should not be dependent on anything, though. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: apache: call for english review
/templates http://cvs.debian.org/debian-installer/tools/bugreporter-udeb/debian/bugreporter-udeb.templates http://cvs.debian.org/debian-installer/tools/partconf/debian/partconf.templates http://cvs.debian.org/debian-installer/tools/partitioner/debian/templates http://cvs.debian.org/debian-installer/tools/sparc/silo-installer/debian/silo-installer.templates http://cvs.debian.org/debian-installer/utils/debian/di-utils-shell.templates http://cvs.debian.org/debian-installer/utils/debian/di-utils-choosesystem.templates -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [RFC] hard drive / usb storage installs
On Sat, Oct 11, 2003 at 06:41:57PM -0400, Joey Hess wrote: I am working on making d-i support installs from USB storage. Booting is dealt with, but there's still the question of how d-i finds the install media to install additional udebs and/or to get debs from. As I thought about this, I realized that once it's mounted, a USB storage device is essentially just another hard drive, so the design should also allow for installs where a generic hard drive is used as the install media. I am also leaning toward getting this to work with any iso image that has d-i on it, so I don't have to worry about creating my own isos (or other media) too. Amen to that. Hard disk access is needed in many cases. The four scenarios I am envisioning are: 1. Download a debian iso to a partition you will not be putting Debian on. Boot via a batch file that runs linux. May only work on more antique versions of DOS. This also works on powerpc, using yaboot manually from OpenFirmware. 2. Same as #1 except create a floppy image and boot from it. 3. Put a debian iso on a USB storage device, and boot from it[0]. Or just installer files, not in iso format. 4. Boot from a floppy which then detects your USB storage device. In all cases the d-i initrd that is loaded contains drivers for hard disks and usb storage, as well as a udeb (provisionally called hd-media-detect) that tries to mount each hard drive and partition in turn, and looks for iso files on them[1]. Once the d-i iso is found, it loop mounts it to /cdrom, and the rest of the install proceeds more or less as normal, using the cdrom-retreiver to pull files from it, possibly using a modified version of cdrom-checker to validate it first. Sounds good. The only caveat during the install is that it will need to avoid letting the user partition the drive from which the iso is loop mounted, or format the partition the iso is in. This may call for some changes in the partitioner and formating modules. Well, if the iso is dd'd to the partition, it is read-only and has no space available. Not sure if that's the way you want to do it though. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Booting trouble
On Fri, Oct 10, 2003 at 04:21:18PM -0600, Kyle Tamminga wrote: I am having trouble booting my debian iso's on a spare computer i have in my room. I can get to the boot screen on my windows but when i try on my spare computer i get nowhere. I want to set up a samba file server so i want this computer to be a stand alone with just debian. I have made sure that my bios is set to boot from the cds and i still cant get anything. I have also tried to boot from floppies but have failed with this atempt to. I have a P-133 with 64mb of ram. I am not going to install x so i will only have a text screen. I was wondering if there is anything i am doing wrong. Thank you for your time. Kyle You don't mention the different flavors. Have you tried the different CDs which boot different flavor kernels? http://www.debian.org/releases/stable/installmanual -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: How to create a installer with d-i?
On Wed, Oct 08, 2003 at 07:22:19PM -0300, Nelson A. de Oliveira wrote: Hi people Sorry for my stupid question, but how can I create the installer? What do I have to do? I´ve entedered cvs.debian.org, saw README, made a cvs, but on the directory debian-installer that it created, there is no Makefile, etc. Where should I enter to test it? Thank you very much Nelson A. de Oliveira Fot testing purposes, it's much easier to download an iso image from http://gluck.debian.org/cdimage/testing/netinst/ -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: [install-doc] About translations
On Mon, Oct 06, 2003 at 07:38:27AM +0200, Denis Barbier wrote: Hi, I tried to manage the French translation of the new installation manual in DocBook XML format with poxml. It seems to work fine, except for one major problem when tag attributes contain entities (see #213036). I will continue to investigate this issue. Good news! Docstruct.ent must be patched to build foreign documentation, but I do not know how. One solution is to remove the en/ prefix, move this file into en/ and put symlinks ../en/docstruct.ent - . for other languages. Whatever works ... the idea was to provide a roadmap of the document for people that work on it, as well as pulling it all together. Maybe it would be better to make it language-specific. Translators can also work as with boot-floppies, the doc-check script has been patched to work with current docs. Excellent. Chris, are translators supposed to work on the manual now, or should they wait? Here is the status. There are 8 chapters complete: welcome hardware preparing install-methods partitioning post-install appendix administrivia These need to be reviewed to be sure the information is correct, they are mostly just imported and reorganized from the boot-floppies manual. We need to add the new information in the middle on how to use debian-installer. This corresponds to the old sections rescue-boot, kernel-install, and boot-new. We also need to add the tech-info chapter which talks about how to customize for different kernels etc. I think what we have represents about 80% of the final text. I plan to get started on the final sections as soon as I can actually do an install on my powerpc, or hopefully in the next week. :) So, it's up to them. Myself, I think it would be fine to get started, the more eyes we have on it the better. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Bug#211092: question_db_delete
On Mon, Oct 06, 2003 at 08:30:31AM +0200, Geert Stappers wrote: On Sun, Oct 05, 2003 at 07:15:35PM -0700, Chris Tillman wrote: Trying to build d-i tonight, I get a failure in mklibs: No library provides non-weak question_db_delete Congratulations ;-) you have been hit by #211092 Yet, libdebconf.so matches in a grep search. Try the 's/raise/print/' workaround Well, that succeeded in getting me a list of all the symbols with the same problem. Since I don't have a clue how to resolve it, I'm listing them here: question_db_delete question_getvalue strwidth quetion_get_field rfc822_parse_stanza template_fields_list template_db_new template_db_delete question_db_new rfc822_header_lookup In pass 4, it then failed with: Moving libdiscover.so.1.0.0 to libdiscover.so.1. [... moving other libs ...] Command failed with status 1: readelf --all -W ./tmp/floppy/tree/lib/libdiscover.so With output: readelf: Error: Cannot stat input file ./tmp/floppy/tree/lib/libdiscover.so. make: *** [floppy-tree-stamp] Error 1 -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Bug#211092: question_db_delete
On Mon, Oct 06, 2003 at 04:16:06PM +0200, Goswin von Brederlow wrote: [EMAIL PROTECTED] (Geert Stappers) writes: On Sun, Oct 05, 2003 at 07:15:35PM -0700, Chris Tillman wrote: Trying to build d-i tonight, I get a failure in mklibs: No library provides non-weak question_db_delete Congratulations ;-) you have been hit by #211092 Yet, libdebconf.so matches in a grep search. Try the 's/raise/print/' workaround I had a similar problem here after an cvs up. But it actually wasn't a mklibs bug. The above error is, and that seems to be the more common case, also cause simply by unresolvable symbols. Try to rebuild all udebs from cvs. If the error then disapears you have hit a missing versioned (Build-)Depends. Otherwise I would need more info and all cvs versions of your files. I did a cvs up -dPAR just now, and tried again. Same result. So the udebs are not rebuilt when necessary under make build? -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: My first d-i installation of a mips.
On Mon, Oct 06, 2003 at 04:30:50PM +0200, Goswin von Brederlow wrote: That also means you can't skip a step that half worked but said it failed. With boot-floppies you could ignore steps. That was also handy when the installer turned out to be broken; you could manually do a step in console 2 and then go around it. Is there some way we can fool d-i into thinking a step's been done, with a console command, so we can bull ahead if necessary? -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: current install disks modification
On Mon, Oct 06, 2003 at 04:00:41PM -0700, Matt Kraai wrote: On Mon, Oct 06, 2003 at 07:34:14PM -0300, Marcos Dione wrote: what I couldn't find was dbootstrap's source code, nor any doc, scripts, hints, etc about the way to build the floppies. I've heard a urban legend claiming the existence of some 'boot-floppies' packages ore something like that, but it seems that it was a long, logn time ago (in a galaxy far, far away?). I mean, someone has built those floppies, and given that he might do it several time till they got good, must have used some scripts or Makefiles or ... ? so, can you gimme any pointers? I think I'm very close to the end of this, and just need a little push. may be I overread something, yes, is very possible. ah! and a little note: I'm not subscribed to the list, so please cc'me. TIA. You can find the boot-floppies source at http://http.us.debian.org/debian/pool/main/b/boot-floppies/boot-floppies_3.0.23.tar.gz (I suspect it will take more than just changing targets.) Another option, of course, is trying to get the (almost) beta installer for sarge, debian-installer, to work. The advantage being that support is supposedly built in to make installs scriptable. You can see that code at http://cvs.debian.org/debian-installer If you go that way, though, you would want to be subscribed here or at least check the archives at lists.debian.org/debian-boot. Yet another is to see whether fai or another automated boot-floppies installer could switch distributions at a higher level. And, where should applications be sent? ;) That's a lotta developers. -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#214500: ddetect: FTBFS on powerpc: cc1: unknown C standard `c99'
Package: ddetect Version: 0.44 Severity: normal File: ddetect Script started on Mon Oct 6 21:25:06 2003 [EMAIL PROTECTED]:~/debian-installer$ sudo make ./compile.sh ddetect tools /home/toff/debian-installer/dest + NAME=ddetect + DIR=tools/ddetect + DEST=/home/toff/debian-installer/dest/ddetect + mkdir -p /home/toff/debian-installer/dest/ddetect + cd tools/ddetect + dpkg-buildpackage -S dpkg-buildpackage: source package is ddetect dpkg-buildpackage: source version is 0.44 dpkg-buildpackage: source maintainer is Matt Kraai [EMAIL PROTECTED] debian/rules clean make[1]: Entering directory `/home/toff/debian-installer/tools/ddetect' dh_testdir dh_testroot rm -f build-stamp /usr/bin/make clean make[2]: Entering directory `/home/toff/debian-installer/tools/ddetect' rm -f -f *~ rm -f *.o rm -f archdetect make[2]: Leaving directory `/home/toff/debian-installer/tools/ddetect' dh_clean make[1]: Leaving directory `/home/toff/debian-installer/tools/ddetect' dpkg-source -b ddetect dpkg-source: warning: unknown information field Xb-installer-menu-item in input data in package's section of control info file dpkg-source: warning: unknown information field Xb-installer-menu-item in input data in package's section of control info file dpkg-source: warning: source directory `./ddetect' is not sourcepackage-upstreamversion `ddetect-0.44' dpkg-source: building ddetect in ddetect_0.44.tar.gz dpkg-source: building ddetect in ddetect_0.44.dsc dpkg-genchanges -S dpkg-genchanges: warning: unknown information field Xb-installer-menu-item in input data in package's section of control info file dpkg-genchanges: warning: unknown information field Xb-installer-menu-item in input data in package's section of control info file dpkg-genchanges: including full source code in upload dpkg-buildpackage: source only upload: Debian-native package + cd .. + mv ddetect_0.44.dsc ddetect_0.44.tar.gz ddetect_0.44_source.changes /home/toff/debian-installer/dest/ddetect + cd /home/toff/debian-installer/dest/ddetect + dpkg-source -x ddetect_0.44.dsc dpkg-source: extracting ddetect in ddetect-0.44 + cd ddetect-0.44/ + debuild -us -uc dpkg-buildpackage: source package is ddetect dpkg-buildpackage: source version is 0.44 dpkg-buildpackage: source maintainer is Matt Kraai [EMAIL PROTECTED] dpkg-buildpackage: host architecture is powerpc debian/rules clean dh_testdir dh_testroot rm -f build-stamp /usr/bin/make clean make[1]: Entering directory `/home/toff/debian-installer/dest/ddetect/ddetect-0.44' rm -f -f *~ rm -f *.o rm -f archdetect make[1]: Leaving directory `/home/toff/debian-installer/dest/ddetect/ddetect-0.44' dh_clean dpkg-source -b ddetect-0.44 dpkg-source: warning: unknown information field Xb-installer-menu-item in input data in package's section of control info file dpkg-source: warning: unknown information field Xb-installer-menu-item in input data in package's section of control info file dpkg-source: building ddetect in ddetect_0.44.tar.gz dpkg-source: building ddetect in ddetect_0.44.dsc debian/rules build dh_testdir /usr/bin/make make[1]: Entering directory `/home/toff/debian-installer/dest/ddetect/ddetect-0.44' gcc -Wall -O2 -g --std=c99 -DARCH=powerpc -DCPU=powerpc -DCPU_TEXT='powerpc' -DSYSTEM=linux -c archdetect-powerpc-linux.c cc1: unknown C standard `c99' archdetect-powerpc-linux.c: In function `subarch_analyze': archdetect-powerpc-linux.c:56: warning: implicit declaration of function `isblank' make[1]: *** [archdetect-powerpc-linux.o] Error 1 make[1]: Leaving directory `/home/toff/debian-installer/dest/ddetect/ddetect-0.44' make: *** [build-stamp] Error 2 debuild: fatal error at line 456: dpkg-buildpackage failed! make: *** [.stamp-tools_ddetect] Error 9 [EMAIL PROTECTED]:~/debian-installer$ exit Script done on Mon Oct 6 21:25:28 2003 -- System Information: Debian Release: 3.0 Architecture: powerpc Kernel: Linux ip68-107-137-100.tc.ph.cox.net 2.4.22-powerpc #1 Sat Oct 4 08:20:43 CEST 2003 ppc Locale: LANG=C, LC_CTYPE=C -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Bug#214499: libdebian-installer: No rule to make target 'configure'
Package: libdebian-installer Severity: normal The subject error appears when building. -- System Information: Debian Release: 3.0 Architecture: powerpc Kernel: Linux ip68-107-137-100.tc.ph.cox.net 2.4.22-powerpc #1 Sat Oct 4 08:20:43 CEST 2003 ppc Locale: LANG=C, LC_CTYPE=C -- Debian GNU/Linux Operating System By the People, For the People Chris Tillman (a people instance) toff one at cox dot net -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
