Re: Documenting the generation of the live images in Debian

2020-03-21 Thread Eduard Bloch
Hallo,
* Roland Clobus [Sat, Mar 21 2020, 05:27:55PM]:

> * To reduce the amount of downloads, I am using a slightly patched
> version of apt-cacher-ng, in order to be able to use the installer. See
> my bug report [4]
>   The live-wrapper script downloads the installer, but apt-cacher-ng
> rejects a path ending in a slash. The following command simulates this
> download:
> 
>   wget -S
> "http://localhost:3142/deb.debian.org/debian/dists/buster/main/installer-amd64/current/images/cdrom/;
> 

I cannot accept your patch as-is. Because it's brute-force, a directory
is not neccessarily a non-volatile file. Can you give me any hints how
to distinguish between volatile and non-volatile directories? Maybe the
same as it's done for for other d-i files, by regex, so

"|/dists/.*/installer-[^/]+/[0-9][^/]+/images/.*" // d-i stuff 
with revision

means frozen contents and

"|/dists/.*/installer-[^/]+/[^0-9][^/]+/images/.*" // d-i stuff 
but not containing a date (year number) in the revision directory (like 
"current", "beta", ...)

is volatile? (those regexps are already defined, I just need to change
your patch to make them applicable to directories)

Also, what's your deadline? I was planing to release a new version of
ACNG in a couple of days anyway. Do you also need it in backports?

Best regards,
Eduard.



Please clean up the list of netinst images

2002-12-01 Thread Eduard Bloch
Hello,

could you please check the http://www.debian.org/CD/netinst/ page? The
_first_ netinst link is dead, and the last one is points to something
for the obsolete Potato 2.2r4 (http://markybobdeb.sourceforge.net/elf/).

Gruss/Regards,
Eduard.
-- 
As of next week, passwords will be entered in Morse code.
  Frank Knappe in debian-user-de


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]




Re: Installing Debian 3.0

2002-08-13 Thread Eduard Bloch

#include hallo.h
* Markus Bollinger [Tue, Aug 13 2002, 11:04:26AM]:
 I tried to boot CD's 1, 2, and 3, but no success.
 (The CD's are booting on other machines with IDE CD rom drive)
 
 Any ideas ?

Yes, try reading the boot-screens on the first CD.

Gruss/Regards,
Eduard.
-- 
Merkt euch, wenn Recht zu Unrecht wird, 
wird der Widerstand zur Pflicht


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]




Re: Sorting boot images into the beginning of the CD.

2002-06-27 Thread Eduard Bloch

#include hallo.h
Philip Hands wrote on Wed Jun 26, 2002 um 03:25:30PM:

 As far as I can see, now that I'm checking the sorted vs. unsorted
 images, the boot images are at the start of the ISO image anyway, so no
 patching is required.

Nice. Okay, if you think, the current scheme works okay, then I am happy
with it.

 Match of `dists/woody/main/disks-i386/3.0.22-2002-04-03/images-1.44/bf2.4/root.bin' 
at offset 3559424
 Match of 
`dists/woody/main/disks-i386/3.0.22-2002-04-03/images-1.44/compact/root.bin' at 
offset 5036032
 Match of `dists/woody/main/disks-i386/3.0.22-2002-04-03/images-1.44/idepci/root.bin' 
at offset 6531072
 Match of `dists/woody/main/disks-i386/3.0.22-2002-04-03/images-1.44/root.bin' at 
offset 8019968
 Match of `dists/woody/Contents-i386.gz' at offset 9498624

I guess, it is early enough.

 I think we've misinterpreted Eduard's original query, which on
 re-reading seems to be implying that there's a problem with Knopix, or
 there was, or something like that.

There was a problem with remastered Knoppix, according to the postings
there. And there was a similar problem, reported to me by some people.
So I assumed the problem may be caused by missing -sort.

Gruss/Regards,
Eduard.
-- 
Zomb.oO(Die 3 F's fon Debian: Find the bug, Fix the bug, Forward to
  maintainer)
  -- Zomb in #debian.de



msg04088/pgp0.pgp
Description: PGP signature


Sorting boot images into the beginning of the CD.

2002-06-24 Thread Eduard Bloch

Hello people,

in the past, I received problem reports from different people, saying
the same thing: Debian-CD does not boot on some old machine, some other
Debian CD does boot, and another one does not - but RedHat/SuSE CD did
boot fine on this machine.

Now, I came across a similar issue on the KNOPPIX mailing list.
According to Christian Perle, one of the guys hacking privately on
KNOPPIX, the image must be somewhere in the beginning of the CD.
Otherwise, there are lots of broken BIOSes (especially some Award models
from around 1997), and he recommends -sort option of mkisofs.

It looks that he's right, syslinux manpage also recomends to put the
boto image into beginning of the CD.

So any opinions? Cc' when the message should reach me, I am not
subscribed.

Gruss/Regards,
Eduard.
-- 
Expect the unexpected and expect the expected to be late.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]




Re: Sorting boot images into the beginning of the CD.

2002-06-24 Thread Eduard Bloch

|Otherwise, if we have to use the -sort argument, some tries may need to
|be done ... i'm not sure if I understand what to put in the sort file.

 -sort sort file 
 Sort file locations on the media. Sorting is controlled by a file
 that contains pairs of filenames and sorting offset weighting.  If
 the weighting is higher, the file will be located closer to the
 beginning of the media, if the weighting is lower, the  file will
 be located closer to the end of the media.  There must be only one
 space or tabs


The order of given filenames/directories does not change much if I
understand mkisofs correctly. It writes the contents in random order,
and the path table is sorted in ASCII order afterwards.

Gruss/Regards,
Eduard.
-- 
Weasel: schreib mal da hinein, daß ich im IRC immer so schroff bin und nur auf
die Türen zeige, öffnen und durchgehen müssen die Leute schon selbst.  Und wenn
sie nicht tun, was ich ihnen sage, endet mein Support an der Stelle.
  -- Joey in #debian.de


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]




Re: Bug#145904: install-doc: Document isolinux multiboot failure with SCSI cdrom (i386 only)

2002-05-05 Thread Eduard Bloch

#include hallo.h
Raphael Hertzog wrote on Sun May 05, 2002 um 06:43:23PM:

 We realized that the isolinux multiboot method seems to fail on most
 SCSI cdrom readers. So, that means that SCSI users should directly use

Hm. There was no problem with Tekram controllers.

Gruss/Regards,
Eduard.
-- 
Ich bin drin!


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]




Re: Bug#142305: i386 - borders gone on vanilla flavor

2002-04-16 Thread Eduard Bloch

#include hallo.h
Adam Di Carlo wrote on Mon Apr 15, 2002 um 09:45:26PM:

  If tweaking TERM makes it work, that's great and we should probably make
  that change for 3.0r1.  For the initial release of woody, we should just
  leave it alone.  It certainly isn't release critical: the installer
  works perfectly well without the borders, just doesn't look so pretty.
 
 Ok, can someone find a way to do this that isn't too gross?  Eduard?
 Philip?

Well, debian-cd did not release afaik, so they still could set this
argument at least for the multiboot CD. Others require the change in BFs
and a new release, of course.

Gruss/Regards,
Eduard.
-- 
Sie haben soeben ihren Mauszeiger erfolgreich bewegt.
Sie müssen Windows neu starten damit die Änderungen wirksam werden.
Wollen Sie Windows jetzt neu starten?


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]




Re: frame-buffer on vanilla?

2002-04-11 Thread Eduard Bloch

#include hallo.h
Adam Di Carlo wrote on Thu Apr 11, 2002 um 01:51:10AM:

 Eduard, there's no point in debating this on this list.  kernel-image
 changes must be approved by the maintainer of that package.  Whether
 the changed package (and i386 b-f rebuild) is accepted in woody is up
 to the release maintainers.

Err, I though we are talking about the change in debian-cd to use bf2.4
or isolinux (kernel-chooser).

Gruss/Regards,
Eduard.
-- 
The early bird gets the worm. If you want something else for
breakfast, get up later.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]




Re: Bug#142305: i386 - borders gone on vanilla flavor

2002-04-11 Thread Eduard Bloch

#include hallo.h
Richard Hirst wrote on Thu Apr 11, 2002 um 11:15:32AM:

  I observed this as well today.  I tested the vanilla kernel, there
  were no borders on the first few boxes (I didn't progress as far as
  the driver config, so I can't confirm problems there.)
 
 Booting with linux TERM=vt102 gives you the borders.  TERM=linux only
 seems to display borders under framebuffer/bterm now.

What about adding this option to isolinux.cfg (for vanilla only)? I do
not see how it would hurt, and it is better to show nice dialog boxes if
someone boots the vanilla flavor. This would also help the default
vanilla boot floppies, though I doubt that we will get a chance to make
a new release.

Gruss/Regards,
Eduard.
-- 
If a train station is where the train stops, what's a work station?
 -- Roger B. Dannenberg 


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]




Re: solution for base-config/cat:: problem

2002-04-09 Thread Eduard Bloch

#include hallo.h
Adam Di Carlo wrote on Tue Apr 09, 2002 um 08:55:22AM:

  Chris: Please, use ALL arguments from particular syslinux.cfg files. I
  was able to reproduce this bug. IMHO caused by missing ram_size
  specification, so the ramdisk-filesystem gets random IO errors, so the
  dbootstrap_settings file is not written correctly (contains cat:
  input/outpur error) and base-config dies while parsing it.
  
  I extracted the rootfs from your image, built an own CD with it (rest
  was from bf2.4 flavor), and it worked fine.
 
 That's strange, Raphael had told me before that they used the
 syslinux.cfg files as is... 

No idea which file you mean. isolinux.cfg is a syslinux.cfg similar
file, adapted by Chris to load different kernels. Compare attached
files.

Gruss/Regards,
Eduard.
-- 
Flüssigseifenbenutzer!
Fristeneinhalter!
Geschenkpapierbügler!
Auslandstanker!


DEFAULT /install/linpci
APPEND root=/dev/ram ro initrd=idepci.bin flavor=idepci
LABEL linux
  kernel /install/linpci
LABEL idepci
  kernel /install/linpci
LABEL compact
  kernel /install/lincompt
  append root=/dev/ram ro initrd=compact.bin flavor=compact
LABEL vanilla
  kernel /install/linux
  append root=/dev/ram ro initrd=root.bin
LABEL bf24
  kernel /install/lin24
  append root=/dev/ram ro initrd=bf24.bin flavor=bf2.4
LABEL rescue
  kernel /install/linpci
  append noinitrd
LABEL resccomp
  kernel /install/lincompt
  append noinitrd
LABEL rescvanl
  kernel /install/linux
  append noinitrd
LABEL rescbf24
  kernel /install/lin24
  append noinitrd
DISPLAY isolinux.txt
TIMEOUT 0
PROMPT 1
F1 f1.txt
F2 f2.txt
F3 f3.txt
F4 f4.txt
F5 f5.txt
F6 f6.txt
F7 f7.txt
F8 f8.txt
F9 f9.txt
F0 f10.txt


FONT font.psf
# see /usr/doc/syslinux/syslinux.doc.gz for file format description
DEFAULT linux.bin
APPEND vga=normal noinitrd load_ramdisk=1 prompt_ramdisk=1 ramdisk_size=16384 
root=/dev/fd0 disksize=1.44 flavor=bf2.4
TIMEOUT 0
DISPLAY debian.txt
PROMPT 1
F1 f1.txt
F2 f2.txt
F3 f3.txt
F4 f4.txt
F5 f5.txt
F6 f6.txt
F7 f7.txt
F8 f8.txt
F9 f9.txt
F0 f10.txt
LABEL linux
KERNEL linux.bin
APPEND vga=normal noinitrd load_ramdisk=1 prompt_ramdisk=1 ramdisk_size=16384 
root=/dev/fd0 disksize=1.44 flavor=bf2.4
LABEL initrd
KERNEL linux.bin
APPEND vga=normal noinitrd load_ramdisk=1 prompt_ramdisk=1 ramdisk_size=16384 
root=/dev/fd0 disksize=1.44 flavor=bf2.4
LABEL ramdisk
KERNEL linux.bin
APPEND vga=normal noinitrd load_ramdisk=1 prompt_ramdisk=1 ramdisk_size=16384 
root=/dev/fd0 disksize=1.44 flavor=bf2.4
LABEL ramdisk0
KERNEL linux.bin
APPEND vga=normal noinitrd load_ramdisk=1 prompt_ramdisk=1 ramdisk_size=16384 
root=/dev/fd0 disksize=1.44 flavor=bf2.4
LABEL ramdisk1
KERNEL linux.bin
APPEND vga=normal noinitrd load_ramdisk=1 prompt_ramdisk=1 ramdisk_size=16384 
root=/dev/fd1 disksize=1.44 flavor=bf2.4
LABEL floppy
KERNEL linux.bin
APPEND vga=normal noinitrd root=/dev/fd0 disksize=1.44 flavor=bf2.4
LABEL floppy0
KERNEL linux.bin
APPEND vga=normal noinitrd root=/dev/fd0 disksize=1.44 flavor=bf2.4
LABEL floppy1
KERNEL linux.bin
APPEND vga=normal noinitrd root=/dev/fd1 disksize=1.44 flavor=bf2.4
LABEL rescue
KERNEL linux.bin
APPEND vga=normal noinitrd



msg03299/pgp0.pgp
Description: PGP signature


Re: frame-buffer on vanilla?

2002-04-06 Thread Eduard Bloch

#include hallo.h
Herbert Xu wrote on Sat Apr 06, 2002 um 08:53:49PM:

  Enabling VGA16 as a module will be done in the next release.
 
 Actually it won't be since vga16 is only modularised in 2.4.

Okay, let's make a summary (+ is pro, - is contra):

 - vanilla (2.2.20)
   + known as stable
   - no framebuffer, so no lang chooser 
 (adding requires changes in kernel and evtl. in bfs)
   - no support for modern hardware (harddisks controllers, network cards)
   - no support for modern filesystems
 - bf2.4
   + framebuffer, lang-chosser possible
   +/- relatively stable, probably less stable as 2.2.x if you use new features
 - compact
   + framebuffer, lang-chosser possible
   + (see vanilla)
   -- (see vanilla)
   - too few drivers for a default install, not even sound cards
 - idepci (2.2.20)
   ++--- (see compact)
   - almost limited to IDE and PCI stuff, but more stable on such
 machines. Best/Only useable as fallback option for seldom broken
 hardware, IMHO.

Change to bf2.4 on the first CD requires a swap of few bytes in
debian-cd, which needs a new release anyhow.

These are not real arguments, but a simple calculation. Not using bf2.4
on the first CD (*) is a shame.

(*) I do not say default. Default implies the best working choice
for most people, which is not true if you have (in useability) equal
options.

Gruss/Regards,
Eduard.
-- 
weasel for i in *; do if ( cat $i | grep -q hallo ); then echo $i; fi; done
[einen Tag später]
-:- Topic (#debian.de): changed by _shorty_: Award fuer useless use of cat an
 Weasel verliehen. :)


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]




Re: 2.4 kernel as default boot kernel on CD #1 ??

2002-04-05 Thread Eduard Bloch

#include hallo.h
Jim Westveer wrote on Thu Apr 04, 2002 um 05:24:24PM:
  And last, woody ought to be 2.2 based and not 2.4 based ... but since I
  also find it lame to use 2.2.x by default nowadays ...
 
 
 I disagree,  2.4.x kernels have been out for over 2 years.we should
 not release a new version of Debian with such an outdated kernel !
 [appologies in advance to the boot-floppy group for all their hard work]

Agreed. The decission, which flavor has to be on the first CD, is done
by debian-cd's scripts, see
/usr/share/debian-cd/tools/boot/woody/boot-i386. CD distributors can
change it if they like.
IMHO bf2.4 is stable enough to be used as installation
kernel and is used by most people (Impression, looking at submitter's
data of recent bug/success reports).

But who makes this decission for the CD Nr.1? DPL?

BTW: I personaly think that compact on the second CD does not make
sence, its primarily goal is smart floppy-based installation. 

Gruss/Regards,
Eduard.
-- 
Beim Reparieren von Elektrogeräten kann einem schlagartig
   klar werden, wo der Fehler liegt.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]




Re: 2.4 kernel as default boot kernel on CD #1 ??

2002-04-05 Thread Eduard Bloch

#include hallo.h
Anthony Towns wrote on Fri Apr 05, 2002 um 07:00:08PM:
 On Fri, Apr 05, 2002 at 10:07:38AM +0200, Eduard Bloch wrote:
  #include hallo.h
  Jim Westveer wrote on Thu Apr 04, 2002 um 05:24:24PM:
And last, woody ought to be 2.2 based and not 2.4 based ... but since I
also find it lame to use 2.2.x by default nowadays ...
   I disagree,  2.4.x kernels have been out for over 2 years.we should
   not release a new version of Debian with such an outdated kernel !
   [appologies in advance to the boot-floppy group for all their hard work]
  Agreed. The decission, which flavor has to be on the first CD, is done
  by debian-cd's scripts, see
  /usr/share/debian-cd/tools/boot/woody/boot-i386. CD distributors can
  change it if they like.
 
 GUYS, _STOP IT_.
 
 NOW IS _NOT_ THE TIME TO MAKE CHANGES TO THE WAY THE INSTALL WORKS.
 
 2.4 is not the default kernel for woody installs, 2.2 is. Yes, it sucks,

What is _default_? What is _option_? Saying that something is default
implies that you stick to this thing from the installation time until
you install another (optional) thing manually. This is not the case
here. You can completely install and work with bf2.4 unless you decide
to change to optimised 2.4.x kernels or 2.2.x.

 and I don't care. If anyone asks why, blame me. DO NOT MAKE FURTHER
 CHANGES TO THE WAY DEFAULT INSTALLS WORK. Not if it makes a billion more
 people able to use Debian. Not if it's so fundamentally offensive that
 it makes you want to disembowel yourself with a screwdriver.

The problem is, many people download only the first CD and install the
rest from the net. Or they even do not look on other CDs, boot with the
first one and are disappointed (no lang chooser, missing drivers for
various hardware, harddisk controllers, USB keyboards, etc. etc.).

This is not a technical change. It is a political one. debian-cd still
needs at least one release before Woody. IMHO enough time to change.

 debian-cd has exactly one goal right now: make sure we're ready to produce
 official CDs that include the entire archive and let you get started on
 an install.

As said before, I know enough people that run their harddisks on Promise
or Highpoint controllers. With the vanilla flavor, Debian would be
completely useless for them.

Gruss/Regards,
Eduard.
-- 
Ich glaube nicht, daß man dieses Stück in Software umgesetzte Scheiße über-
haupt mieser machen kann, als es sowieso schon ist. Das dürfte das einzige
Programm sein, das vom Verhalten und seinen Anwendern her schlimmer als XP
auf einem Amiga ist. - Manuel Richardt in ka.talk ueber Outlook Express



msg03184/pgp0.pgp
Description: PGP signature


Re: Bug#123948: can someone test the patch?

2002-03-02 Thread Eduard Bloch

#include hallo.h
Philip Blundell wrote on Sat Mar 02, 2002 um 10:48:41AM:
 Who creates the existing .disk/info file?  I thought that was
 debian-cd.  The new files Eduard is talking about can be done in exactly
 the same way, as far as I can tell.

Buxy admited already to enable this in debian-cd, so it is only a
question of time...

Gruss/Regards,
Eduard.
-- 
Actually, Microsoft is sort of a mixture between the Borg and the Ferengi.
Das stimmt so nicht - die Borg wissen wenigstens, wie ein Netzwerk
auszusehen hat.


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]




Re: Bug#123948: can someone test the patch?

2002-03-01 Thread Eduard Bloch

#include hallo.h
Phil Blundell wrote on Fri Mar 01, 2002 um 12:23:12PM:

 How about a compromise.  If an attractive-looking CD is detected in the
 drive, pop up a single dialogue box at the beginning along the lines of
 I see you have a Debian CD-ROM.  Do you want to use this as the source
 medium for all the installation files?.  If the user answers yes,

A good compromise. I suggest two kinds of flags on the CD-ROM:

.disk/kernel_installable
.disk/base_installable

The first marks the disk containing kernel and modules, the second the
one with the basedebs.

 behave as if they had booted with the cdrom argument and suppress all
 the media-selection menus; if they answer no, prompt as usual.

I looked trough the source, and I would like to rewrite most of the
cruft, beginning with is_cdrom_image().

Gruss/Regards,
Eduard.
-- 
shorty_ Fuchur: Finger weg von Inn, damit kannn man sich schneller als
einem lieb ist in den Fuss schiessen.
shorty_ Fuchur: Vertraue mir.
Joey Hmm, shorty kling so als koenne er die Loecher in seinen Fuessen
nicht mehr zaehlen...
  -- #debian.de


-- 
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]




Re: Bug#109826: base floppy images on the Debian CD

2001-08-24 Thread Eduard Bloch

#include hallo.h
Ethan Benson wrote on Fri Aug 24, 2001 um 04:49:19AM:

 for the bazillionth time now, basedebs.tgz is NOT NOT NOT a root
 filesystem.  its a tarball of a bunch of .deb files.  

Exactly. What is the reason for having basedebs.tgz? We cannot have the
tarball INSTEAD of the deb files (or can we?) Who needs basedebs.tgz? We
could put DOS versions of tar and gzip on the CD plus a BATch file, so
people can make their own basedebs.tgz if they want. Puting the same
data in different forms is wastage of disc space.

Gruss/Regards,
Eduard.
-- 
...mit VIM kein Problem!
:wq

 PGP signature