is 2.4.15 really available at www.kernel.org?

is 2.4.15 really available at www.kernel.org?

[rpjday]

  i just tried to download linux-2.4.15.tar.bz2 and, far from being a 22M
new kernel, it's a 155312 byte file that i suspect is really just a patch
file, so i suspect someone mislabelled the link there.  any hints?

rday

Re: is 2.4.15 really available at www.kernel.org?

[Keith Owens]

On Fri, 23 Nov 2001 03:58:34 -0500 (EST), 
rpjday <rpjday@mindspring.com> wrote:
>  i just tried to download linux-2.4.15.tar.bz2 and, far from being a 22M
>new kernel, it's a 155312 byte file that i suspect is really just a patch
>file, so i suspect someone mislabelled the link there.  any hints?

Works for me.

-rw-r--r--    1 kaos     ocs      23748963 Nov 23 17:39 linux-2.5.0.tar.bz2
-rw-r--r--    1 kaos     ocs           248 Nov 23 17:39 linux-2.5.0.tar.bz2.sign

Re: is 2.4.15 really available at www.kernel.org?

[rpjday]

On Fri, 23 Nov 2001, Keith Owens wrote:

> On Fri, 23 Nov 2001 03:58:34 -0500 (EST), 
> rpjday <rpjday@mindspring.com> wrote:
> >  i just tried to download linux-2.4.15.tar.bz2 and, far from being a 22M
> >new kernel, it's a 155312 byte file that i suspect is really just a patch
> >file, so i suspect someone mislabelled the link there.  any hints?
> 
> Works for me.
> 
> -rw-r--r--    1 kaos     ocs      23748963 Nov 23 17:39 linux-2.5.0.tar.bz2
> -rw-r--r--    1 kaos     ocs           248 Nov 23 17:39 linux-2.5.0.tar.bz2.sign

you're referring to the 2.5.0 kernel, i'm talking about the 2.4.15
kernel.

i swear, i am not making this up.  i just tried again, through mozilla,
to download the file 
www.kernel.org/pub/linux/kernel/v2.4/linux-2.4.15.tar.bz2, and it 
completed after downloading *exactly* 155312 bytes, just as before.

getting it via ftp works fine -- it's http that's giving me this
weird problem.   is it just me?

rday

Re: is 2.4.15 really available at www.kernel.org?

[Jochen Striepe]

        Hi,

On 23 Nov 2001, rpjday <rpjday@mindspring.com> wrote:
> 
> i swear, i am not making this up.  i just tried again, through mozilla,
> to download the file 
> www.kernel.org/pub/linux/kernel/v2.4/linux-2.4.15.tar.bz2, and it 
> completed after downloading *exactly* 155312 bytes, just as before.
> 
> getting it via ftp works fine -- it's http that's giving me this
> weird problem.   is it just me?

Getting it via http using wget worked fine here:

-rw-------    1 jochen   users    23747061 Nov 23 07:18 /tmp/linux-2.4.15.tar.bz2


I am *much* more irritated by:

$ uname -r
2.4.15-greased-turkey


So long,

Jochen.

-- 
Think of the mess on the carpet. Sensible people do all their
demon-summoning in the garage, which you can just hose down afterwards.
        -- damerell@chiark.greenend.org.uk

Re: is 2.4.15 really available at www.kernel.org?

[Rik van Riel]

On Fri, 23 Nov 2001, Jochen Striepe wrote:

> I am *much* more irritated by:
>
> $ uname -r
> 2.4.15-greased-turkey

Would you really have preferred 2.4.15-sitting-duck  ?

cheers,

Rik
-- 
Shortwave goes a long way:  irc.starchat.net  #swl

http://www.surriel.com/		http://distro.conectiva.com/

Re: is 2.4.15 really available at www.kernel.org?

[Oliver.Neukum]

On Fri, 23 Nov 2001, Rik van Riel wrote:

> On Fri, 23 Nov 2001, Jochen Striepe wrote:
>
> > I am *much* more irritated by:
> >
> > $ uname -r
> > 2.4.15-greased-turkey
>
> Would you really have preferred 2.4.15-sitting-duck  ?

speedful-penguin ?

	Regards
		Oliver

Re: is 2.4.15 really available at www.kernel.org?

[Cameron Simpson]

On Fri, Nov 23, 2001 at 07:48:13AM -0200, Rik van Riel <riel@conectiva.com.br> wrote:
| On Fri, 23 Nov 2001, Jochen Striepe wrote:
| > I am *much* more irritated by:
| > $ uname -r
| > 2.4.15-greased-turkey
| 
| Would you really have preferred 2.4.15-sitting-duck  ?

Given what just happened with inode caches in 2.4.15, maybe more apt :-(
-- 
Cameron Simpson, DoD#743        cs@zip.com.au    http://www.zip.com.au/~cs/

Ride to not crash. Dress to crash. Live to ride to not crash again.
	- Lawrence Smith, DoD#i, lawrence@msc.cornell.edu

Re: is 2.4.15 really available at www.kernel.org?

[Flavio Stanchina]

On Friday 23 November 2001 10:43, Jochen Striepe wrote:

> I am *much* more irritated by:
>
> $ uname -r
> 2.4.15-greased-turkey

So I guess you are vegetarian. Try changing to "2.4.15-tasteful-salad".

-- 
Ciao,
    Flavio Stanchina
    Trento - Italy

"The best defense against logic is ignorance."
http://spazioweb.inwind.it/fstanchina/

Re: is 2.4.15 really available at www.kernel.org?

[Michael H. Warfield]

On Fri, Nov 23, 2001 at 11:33:38AM +0100, Flavio Stanchina wrote:
> On Friday 23 November 2001 10:43, Jochen Striepe wrote:
> 
> > I am *much* more irritated by:
> >
> > $ uname -r
> > 2.4.15-greased-turkey

> So I guess you are vegetarian. Try changing to "2.4.15-tasteful-salad".

	Point is that it BROKE some things....  Like "make install" on
RedHat installed the damn thing as /boot/vmlinuz-2.4.15-greased-turkey,
breaking the lilo settings if you set an image for "vmlinuz-2.4.15"
like you expected it to be.  Not funny.  Just had three freeswan
kinstall builds blow up because of that.

	Now got to go back and fix it and rebuild.

> -- 
> Ciao,
>     Flavio Stanchina
>     Trento - Italy

> "The best defense against logic is ignorance."
> http://spazioweb.inwind.it/fstanchina/

	Mike
-- 
 Michael H. Warfield    |  (770) 985-6132   |  mhw@WittsEnd.com
  /\/\|=mhw=|\/\/       |  (678) 463-0932   |  http://www.wittsend.com/mhw/
  NIC whois:  MHW9      |  An optimist believes we live in the best of all
 PGP Key: 0xDF1DD471    |  possible worlds.  A pessimist is sure of it!

Re: is 2.4.15 really available at www.kernel.org?

[Jeff Garzik]

On Fri, 23 Nov 2001, Michael H. Warfield wrote:
> 	Point is that it BROKE some things....  Like "make install" on
> RedHat installed the damn thing as /boot/vmlinuz-2.4.15-greased-turkey,
> breaking the lilo settings if you set an image for "vmlinuz-2.4.15"
> like you expected it to be.  Not funny.  Just had three freeswan
> kinstall builds blow up because of that.

Life is rough...

	Jeff

Re: is 2.4.15 really available at www.kernel.org?

[Michael H. Warfield]

On Fri, Nov 23, 2001 at 10:27:45AM -0600, Jeff Garzik wrote:
> On Fri, 23 Nov 2001, Michael H. Warfield wrote:
> > 	Point is that it BROKE some things....  Like "make install" on
> > RedHat installed the damn thing as /boot/vmlinuz-2.4.15-greased-turkey,
> > breaking the lilo settings if you set an image for "vmlinuz-2.4.15"
> > like you expected it to be.  Not funny.  Just had three freeswan
> > kinstall builds blow up because of that.

> Life is rough...

	Yes, this is true.  Rough enough as it is.  As one of the (minor)
device driver maintainers, I expect to test some of these versions (some
of which may be unexpectedly radioactive) as quick as I can and I keep
backs so my systems can be booted when a new version is unbootable (very
rare now days, thankfully).  This did NOT help that cause.  Making life
MORE difficult is not the way to get people to TEST things.

> 	Jeff

	Mike
-- 
 Michael H. Warfield    |  (770) 985-6132   |  mhw@WittsEnd.com
  /\/\|=mhw=|\/\/       |  (678) 463-0932   |  http://www.wittsend.com/mhw/
  NIC whois:  MHW9      |  An optimist believes we live in the best of all
 PGP Key: 0xDF1DD471    |  possible worlds.  A pessimist is sure of it!

Re: is 2.4.15 really available at www.kernel.org?

[Stephen Frost]

[-- Attachment #1: Type: text/plain, Size: 658 bytes --]

* Michael H. Warfield (mhw@wittsend.com) wrote:
> On Fri, Nov 23, 2001 at 10:27:45AM -0600, Jeff Garzik wrote:
> > On Fri, 23 Nov 2001, Michael H. Warfield wrote:
> > > 	Point is that it BROKE some things....  Like "make install" on
> > > RedHat installed the damn thing as /boot/vmlinuz-2.4.15-greased-turkey,
> > > breaking the lilo settings if you set an image for "vmlinuz-2.4.15"
> > > like you expected it to be.  Not funny.  Just had three freeswan
> > > kinstall builds blow up because of that.
> 
	Uh, so don't make assumptions on what the kernel rev. is going
	to be?  It's not that hard to figure it out from the Makefile.

		Stephen

[-- Attachment #2: Type: application/pgp-signature, Size: 232 bytes --]

Re: is 2.4.15 really available at www.kernel.org?

[J Sloan]

Stephen Frost wrote:

> * Michael H. Warfield (mhw@wittsend.com) wrote:
> > On Fri, Nov 23, 2001 at 10:27:45AM -0600, Jeff Garzik wrote:
> > > On Fri, 23 Nov 2001, Michael H. Warfield wrote:
> > > >   Point is that it BROKE some things....  Like "make install" on
> > > > RedHat installed the damn thing as /boot/vmlinuz-2.4.15-greased-turkey,
> > > > breaking the lilo settings if you set an image for "vmlinuz-2.4.15"
> > > > like you expected it to be.  Not funny.  Just had three freeswan
> > > > kinstall builds blow up because of that.
> >
>         Uh, so don't make assumptions on what the kernel rev. is going
>         to be?  It's not that hard to figure it out from the Makefile.
>
>                 Stephen

Or, better yet, a quick glance at /lib/modules?

cu

jjs

Re: is 2.4.15 really available at www.kernel.org?

[Mike Fedyk]

On Fri, Nov 23, 2001 at 11:05:05AM -0500, Michael H. Warfield wrote:
> On Fri, Nov 23, 2001 at 11:33:38AM +0100, Flavio Stanchina wrote:
> > On Friday 23 November 2001 10:43, Jochen Striepe wrote:
> > 
> > > I am *much* more irritated by:
> > >
> > > $ uname -r
> > > 2.4.15-greased-turkey
> 
> > So I guess you are vegetarian. Try changing to "2.4.15-tasteful-salad".
> 
> 	Point is that it BROKE some things....  Like "make install" on
> RedHat installed the damn thing as /boot/vmlinuz-2.4.15-greased-turkey,
> breaking the lilo settings if you set an image for "vmlinuz-2.4.15"
> like you expected it to be.  Not funny.  Just had three freeswan
> kinstall builds blow up because of that.
> 
> 	Now got to go back and fix it and rebuild.

OMFG!

How can you *not* point to the /boot/vmlinuz symlink?!!!  It points directly
to the latest kernel. And, /boot/vmlinuz.old points to the previous kernel.

Here are some examples:  This is *just too simple*!!!

Lilo:
image=/boot/vmlinuz
	label=vmlinuz
	read-only
#	restricted
	alias=1

image=/boot/vmlinuz.old
	label=vmlinuz-old
	read-only
	optional
	append="single"
#	restricted
	alias=3
			
Grub:
title           Debian GNU/Linux, Latest Kernel
root            (hd0,0)
kernel          /boot/vmlinuz root=/dev/hda1 ro vga=extended

title           Debian GNU/Linux, Latest Kernel (recovery mode)
root            (hd0,0)
kernel          /boot/vmlinuz root=/dev/hda1 ro single vga=extended

title           Debian GNU/Linux, Previous Kernel
root            (hd0,0)
kernel          /boot/vmlinuz.old root=/dev/hda1 ro vga=extended

title           Debian GNU/Linux, Previous Kernel (recovery mode)
root            (hd0,0)
kernel          /boot/vmlinuz.old root=/dev/hda1 ro single vga=extended

Re: is 2.4.15 really available at www.kernel.org?

[Michael H. Warfield]

On Fri, Nov 23, 2001 at 01:08:20PM -0800, Mike Fedyk wrote:
> On Fri, Nov 23, 2001 at 11:05:05AM -0500, Michael H. Warfield wrote:
> > On Fri, Nov 23, 2001 at 11:33:38AM +0100, Flavio Stanchina wrote:
> > > On Friday 23 November 2001 10:43, Jochen Striepe wrote:
> > > 
> > > > I am *much* more irritated by:
> > > >
> > > > $ uname -r
> > > > 2.4.15-greased-turkey
> > 
> > > So I guess you are vegetarian. Try changing to "2.4.15-tasteful-salad".
> > 
> > 	Point is that it BROKE some things....  Like "make install" on
> > RedHat installed the damn thing as /boot/vmlinuz-2.4.15-greased-turkey,
> > breaking the lilo settings if you set an image for "vmlinuz-2.4.15"
> > like you expected it to be.  Not funny.  Just had three freeswan
> > kinstall builds blow up because of that.
> > 
> > 	Now got to go back and fix it and rebuild.

> OMFG!

> How can you *not* point to the /boot/vmlinuz symlink?!!!  It points directly
> to the latest kernel. And, /boot/vmlinuz.old points to the previous kernel.

	Clue alert...

	The PRIMARY link goes to vmlinuz.  The backup links go to the
specific versions (the install script even facilitates this for you
by installing it there).  That way, when you end up with a radioactive
version, you can boot your prior version and recover from the disaster.

> Here are some examples:  This is *just too simple*!!!

	I typically keep 4 to six fall back versions in each of the
2.2 and 2.4 lines active and want (or occasionally need) to target specific
versions, especially when I'm testing preX kernels and my device driver.
You are way TOO simple.

	Mike
-- 
 Michael H. Warfield    |  (770) 985-6132   |  mhw@WittsEnd.com
  /\/\|=mhw=|\/\/       |  (678) 463-0932   |  http://www.wittsend.com/mhw/
  NIC whois:  MHW9      |  An optimist believes we live in the best of all
 PGP Key: 0xDF1DD471    |  possible worlds.  A pessimist is sure of it!

Re: is 2.4.15 really available at www.kernel.org?

[Kai Henningsen]

mhw@wittsend.com (Michael H. Warfield)  wrote on 23.11.01 in <20011123185407.A3499@alcove.wittsend.com>:

> 	I typically keep 4 to six fall back versions in each of the
> 2.2 and 2.4 lines active and want (or occasionally need) to target specific
> versions, especially when I'm testing preX kernels and my device driver.
> You are way TOO simple.

I keep more (though I really don't need that many) ... and I *do* add text  
to kernel names myself.

So I wrote a (very quick-and-dirty) little Perl script. Maybe a variant of  
that works for other people, too.

Features: label is (hopefully sensibly) shortened image name. Also, a  
number is used as an alias; it's easier to select "1" than some lengthy  
string. Kernels are (hopefully) sorted chronologically (this doesn't work  
if EXTRAVERSION starts with a letter).

WARNING: this makes some assumptions about my system. You need to adapt  
that part.

WARNING: the sort routine only works on a Debian system. If you live on  
something else, adapt the sorter.

The script asks before overwriting your lilo.conf and keeps backups, so  
you have a chance of looking at the result and tweaking the script before  
committing to it.

License: public domain.

make-lilo.conf.pl:
#! /usr/bin/perl -w

use strict;

open LILO, "> /etc/lilo.conf.gen" or die $!;
print LILO <<headend;
# LILO configuration created by $0 @{[scalar localtime]}

linear
boot = /dev/sda
compact
delay = 100	# optional, for systems that boot very quickly
#vga = normal	# force sane state
vga = ask
root = current	# use "current" root
#root = /dev/sdc1
#other = /dev/sda1
#  table = /dev/sda
#  label = dos

headend

my $sorter = sub {
	my ($aa, $bb) = ($a, $b);
	$aa =~ tr/+/-/;
	$bb =~ tr/+/-/;
	$aa eq $bb? 0:
	system('/usr/bin/dpkg', '--compare-versions', $aa, 'lt', $bb)? -1: 1;
};

opendir BOOT, "/boot/" or die $!;
my @kernels = sort $sorter grep m/linu/i, readdir BOOT;
close BOOT;

my $n = 0;

for my $kernel (@kernels) {
	my ($version) = ($kernel =~ m/^[-a-z]*(.*)$/);
	$version =~ s/.*(.{15})$/$1/ if length($version) > 15;
	$n++;
	if ($n > 9) {
		print "Ignoring $kernel ($n)\n";
		next;
	}
	print LILO <<imageend;
	
image = /boot/$kernel
  label = $version
  alias = $n
  append = " hisax=3,2,10, "

imageend
}

close LILO;

system('/bin/mv', '-vib', '/etc/lilo.conf.gen', '/etc/lilo.conf');

system('/sbin/lilo');


MfG Kai

Re: is 2.4.15 really available at www.kernel.org?

[Keith Owens]

On 24 Nov 2001 12:01:00 +0200, 
kaih@khms.westfalen.de (Kai Henningsen) wrote:
>mhw@wittsend.com (Michael H. Warfield)  wrote on 23.11.01 in <20011123185407.A3499@alcove.wittsend.com>:
>> 	I typically keep 4 to six fall back versions in each of the
>> 2.2 and 2.4 lines active and want (or occasionally need) to target specific
>> versions, especially when I'm testing preX kernels and my device driver.
>> You are way TOO simple.
>
>I keep more (though I really don't need that many) ... and I *do* add text  
>to kernel names myself.
>
>So I wrote a (very quick-and-dirty) little Perl script. Maybe a variant of  
>that works for other people, too.

kbuild 2.5 has standard support for running user specific install
scripts after installing the bootable kernel and modules.  That is, the
"update my bootloader" phase can be automated and will propagate from
one .config to the next when you make oldconfig.

bool 'Run a post-install script or command' CONFIG_INSTALL_SCRIPT
if [ "$CONFIG_INSTALL_SCRIPT" = "y" ]; then
  string '  Post-install script or command name' CONFIG_INSTALL_SCRIPT_NAME ""
fi

$(CONFIG_INSTALL_SCRIPT_NAME) is run with several environment variables
set, including the kernel release.  There is a sample install script in
scripts/lilo_new_kernel which will satisfy most people, if not you can
copy and edit it to suit.

#!/bin/sh
#
#  This is a sample script to add a new kernel to /etc/lilo.conf.  If it
#  does not do what you want, copy this script to somewhere outside the
#  kernel, change the copy and point your .config at the modified copy.
#  Then you do not need to change the script when you upgrade your kernel.
#

label=$(echo "$KERNELRELEASE" | cut -c1-15)
if ! grep "label=$label\$" /etc/lilo.conf > /dev/null
then
  ed /etc/lilo.conf > /dev/null 2>&1 <<EODATA 
/^image/
i
image=$CONFIG_INSTALL_PREFIX_NAME$CONFIG_INSTALL_KERNEL_NAME
	label=$label
	optional

.
w
q
EODATA
  if [ ! $? ]
  then
    echo edit of /etc/lilo.conf failed
    exit 1
  fi
fi
lilo

Re: is 2.4.15 really available at www.kernel.org?

[Christoph Hellwig]

In article <2450.1006608941@ocs3.intra.ocs.com.au> you wrote:
> kbuild 2.5 has standard support for running user specific install
> scripts after installing the bootable kernel and modules.  That is, the
> "update my bootloader" phase can be automated and will propagate from
> one .config to the next when you make oldconfig.

Never 2.4 kernels already try to excecute ~/bin/installkernel in the
'make install' pass on i386.

My personal tip for people keeping lots of kernels around is grub, though.
No need for a menu entry, one can just boot all kernel on the accessible
filesystems.

Together with the above "~/bin/installkernel" option I put my kernels always
into /lib/modules/<version>/vmlinux so I can find them easily (IMHO this
should be default in 2.5), so even lilo-using people could write simple
scripts to add all kernels present in /lib/modules/ to their config.
This does of course make the path '/lib/modules/' grossly misnamed, maybe
we could change it into /kernel in 2.5 :)

	Christoph

-- 
Of course it doesn't work. We've performed a software upgrade.

Re: is 2.4.15 really available at www.kernel.org?

[H. Peter Anvin]

Followup to:  <200111241356.fAODuIb30257@ns.caldera.de>
By author:    Christoph Hellwig <hch@ns.caldera.de>
In newsgroup: linux.dev.kernel
>
> In article <2450.1006608941@ocs3.intra.ocs.com.au> you wrote:
> > kbuild 2.5 has standard support for running user specific install
> > scripts after installing the bootable kernel and modules.  That is, the
> > "update my bootloader" phase can be automated and will propagate from
> > one .config to the next when you make oldconfig.
> 
> Never 2.4 kernels already try to excecute ~/bin/installkernel in the
> 'make install' pass on i386.
> 

Or you can just put in your /sbin/installkernel:

if [ -x $HOME/bin/installkernel ]; then
  exec $HOME/bin/installkernel "$@"
fi

	-hpa
-- 
<hpa@transmeta.com> at work, <hpa@zytor.com> in private!
"Unix gives you enough rope to shoot yourself in the foot."
http://www.zytor.com/~hpa/puzzle.txt	<amsp@zytor.com>

Re: is 2.4.15 really available at www.kernel.org?

[Keith Owens]

On Sat, 24 Nov 2001 14:56:18 +0100, 
Christoph Hellwig <hch@ns.caldera.de> wrote:
>In article <2450.1006608941@ocs3.intra.ocs.com.au> you wrote:
>> kbuild 2.5 has standard support for running user specific install
>> scripts after installing the bootable kernel and modules.  That is, the
>> "update my bootloader" phase can be automated and will propagate from
>> one .config to the next when you make oldconfig.
>
>Never 2.4 kernels already try to excecute ~/bin/installkernel in the
>'make install' pass on i386.

I know.  kbuild 2.5 goes further and gives the user a choice about
(a) whether to run a script on install and (b) what the script name is,
instead of hard coding it.

>Together with the above "~/bin/installkernel" option I put my kernels always
>into /lib/modules/<version>/vmlinux so I can find them easily (IMHO this
>should be default in 2.5)

Architecture dependent.  In kbuild 2.5 for most architectures the
default location for the kernel, System.map and .config is in
/lib/modules.

string 'Where to install the kernel' CONFIG_INSTALL_KERNEL_NAME "/lib/modules/KERNELRELEASE/vmlinuz"
bool 'Install System.map' CONFIG_INSTALL_SYSTEM_MAP
if [ "$CONFIG_INSTALL_SYSTEM_MAP" = "y" ]; then
  string '  Where to install System.map' CONFIG_INSTALL_SYSTEM_MAP_NAME "/lib/modules/KERNELRELEASE/System.map"
fi
bool 'Install .config' CONFIG_INSTALL_CONFIG
if [ "$CONFIG_INSTALL_CONFIG" = "y" ]; then
  string '  Where to install .config' CONFIG_INSTALL_CONFIG_NAME "/lib/modules/KERNELRELEASE/.config"
fi

Users with special requirements (old BIOS, small /lib etc.) can
configure their install to put the kernel where they like.  At least
one architecture (ia64) mandates that bootable images live in a
separate partition, the firmware on ia64 requires this, so the default
for vmlinuz is different.

string 'Where to install the kernel' CONFIG_INSTALL_KERNEL_NAME "/boot/efi/vmlinuz-KERNELRELEASE"

>so even lilo-using people could write simple
>scripts to add all kernels present in /lib/modules/ to their config.
>This does of course make the path '/lib/modules/' grossly misnamed, maybe
>we could change it into /kernel in 2.5 :)

I was tempted, but the number of things that would break ... shudder.

Re: is 2.4.15 really available at www.kernel.org?

[Kai Henningsen]

kaos@ocs.com.au (Keith Owens)  wrote on 25.11.01 in <9705.1006646279@ocs3.intra.ocs.com.au>:

> On Sat, 24 Nov 2001 14:56:18 +0100,
> Christoph Hellwig <hch@ns.caldera.de> wrote:

> >so even lilo-using people could write simple
> >scripts to add all kernels present in /lib/modules/ to their config.
> >This does of course make the path '/lib/modules/' grossly misnamed, maybe
> >we could change it into /kernel in 2.5 :)
>
> I was tempted, but the number of things that would break ... shudder.

Just put a symlink there?


MfG Kai

Re: is 2.4.15 really available at www.kernel.org?

[Mike Fedyk]

[-- Attachment #1: Type: text/plain, Size: 16403 bytes --]

On Fri, Nov 23, 2001 at 06:54:07PM -0500, Michael H. Warfield wrote:
> On Fri, Nov 23, 2001 at 01:08:20PM -0800, Mike Fedyk wrote:
> > On Fri, Nov 23, 2001 at 11:05:05AM -0500, Michael H. Warfield wrote:
> > > On Fri, Nov 23, 2001 at 11:33:38AM +0100, Flavio Stanchina wrote:
> > > > On Friday 23 November 2001 10:43, Jochen Striepe wrote:
> > > > 
> > > > > I am *much* more irritated by:
> > > > >
> > > > > $ uname -r
> > > > > 2.4.15-greased-turkey
> > > 
> > > > So I guess you are vegetarian. Try changing to "2.4.15-tasteful-salad".
> > > 
> > > 	Point is that it BROKE some things....  Like "make install" on
> > > RedHat installed the damn thing as /boot/vmlinuz-2.4.15-greased-turkey,
> > > breaking the lilo settings if you set an image for "vmlinuz-2.4.15"
> > > like you expected it to be.  Not funny.  Just had three freeswan
> > > kinstall builds blow up because of that.
> > > 
> > > 	Now got to go back and fix it and rebuild.
> 
> > OMFG!
> 
> > How can you *not* point to the /boot/vmlinuz symlink?!!!  It points directly
> > to the latest kernel. And, /boot/vmlinuz.old points to the previous kernel.
> 
> 	Clue alert...
> 
> 	The PRIMARY link goes to vmlinuz.  The backup links go to the
> specific versions (the install script even facilitates this for you
> by installing it there).  That way, when you end up with a radioactive
> version, you can boot your prior version and recover from the disaster.
>

Lets see...

/boot/vmlinuz -> vmlinuz-2.4.15-pre7
/boot/vmlinuz.old -> vmlinuz-2.4.15-pre7.old

That looks like new, and old to me...

> > Here are some examples:  This is *just too simple*!!!
> 
> 	I typically keep 4 to six fall back versions in each of the
> 2.2 and 2.4 lines active and want (or occasionally need) to target specific
> versions, especially when I'm testing preX kernels and my device driver.
> You are way TOO simple.
>

$ ls /boot|grep -c vmlinuz
39

All right...

I sortened my answer to fix your specific problem, but grub is working great
for this too...

There's also a nice util provided by the debian grub package that keeps this
up to date.  All you have to do is modify your shutdown scripts, or modify
/sbin/installkernel

My response was terse, I admit.  I must ask you though, why were you
complaining about boot loader settings that you say is already setup
correctly?

/boot/grub/menu.lst:
# By default, boot the first entry after five seconds.
default         0
timeout         5

# Pretty colours
color cyan/blue white/blue

#
# examples
#
# title		Windows 95/98/NT/2000
# root		(hd0,0)
# makeactive
# chainloader	+1
#
# title		Linux
# root		(hd0,1)
# kernel	/vmlinuz root=/dev/hda2 ro
#

title           Debian GNU/Linux, Latest Kernel
root            (hd0,0)
kernel          /boot/vmlinuz root=/dev/hda1 ro vga=extended

title           Debian GNU/Linux, Latest Kernel (recovery mode)
root            (hd0,0)
kernel          /boot/vmlinuz root=/dev/hda1 ro single vga=extended

title           Debian GNU/Linux, Previous Kernel
root            (hd0,0)
kernel          /boot/vmlinuz.old root=/dev/hda1 ro vga=extended

title           Debian GNU/Linux, Previous Kernel (recovery mode)
root            (hd0,0)
kernel          /boot/vmlinuz.old root=/dev/hda1 ro single vga=extended

# rootflags=data=journal
### BEGIN AUTOMAGIC KERNELS LIST
# lines between the AUTOMAGIC KERNELS LIST markers will be modified
# by the debian update-grub script except for the default optons below
# DO NOT UNCOMMENT THEM, Just edit them to your needs

# ## Start Default Options ##
# default kernel options
# kopt=root=/dev/hda1 ro vga=extended
# default grub root device
# groot=(hd0,0)
# default should update-grub add recovery options to menu
# recovery=true
# ## End Default Options ##

title		Debian GNU/Linux, kernel 2.4.14-pre8+netdev_random-p7+xsched+ext3_0.9.14-2414p8+elevator
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre8+netdev_random-p7+xsched+ext3_0.9.14-2414p8+elevator root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.14-pre8+netdev_random-p7+xsched+ext3_0.9.14-2414p8+elevator (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre8+netdev_random-p7+xsched+ext3_0.9.14-2414p8+elevator root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.14-pre8+netdev_random-p7+xsched+ext3_0.9.14-2414p8
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre8+netdev_random-p7+xsched+ext3_0.9.14-2414p8 root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.14-pre8+netdev_random-p7+xsched+ext3_0.9.14-2414p8 (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre8+netdev_random-p7+xsched+ext3_0.9.14-2414p8 root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.14-pre8+netdev_random-p7+xsched+elevator
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre8+netdev_random-p7+xsched+elevator root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.14-pre8+netdev_random-p7+xsched+elevator (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre8+netdev_random-p7+xsched+elevator root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.14-pre8+netdev_random-p7+ext3_0.9.14-2414p8+elevator+preempt-p7
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre8+netdev_random-p7+ext3_0.9.14-2414p8+elevator+preempt-p7 root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.14-pre8+netdev_random-p7+ext3_0.9.14-2414p8+elevator+preempt-p7 (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre8+netdev_random-p7+ext3_0.9.14-2414p8+elevator+preempt-p7 root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.14-pre8+netdev_random-p7+elevator+preempt-p7+xsched
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre8+netdev_random-p7+elevator+preempt-p7+xsched root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.14-pre8+netdev_random-p7+elevator+preempt-p7+xsched (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre8+netdev_random-p7+elevator+preempt-p7+xsched root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.14-pre7+xsched+ext3_0.9.14-2414p5+netdev_random.old
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre7+xsched+ext3_0.9.14-2414p5+netdev_random.old root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.14-pre7+xsched+ext3_0.9.14-2414p5+netdev_random.old (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre7+xsched+ext3_0.9.14-2414p5+netdev_random.old root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.14-pre7+xsched+ext3_0.9.14-2414p5+netdev_random
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre7+xsched+ext3_0.9.14-2414p5+netdev_random root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.14-pre7+xsched+ext3_0.9.14-2414p5+netdev_random (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre7+xsched+ext3_0.9.14-2414p5+netdev_random root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.14-pre6+xsched
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre6+xsched root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.14-pre6+xsched (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre6+xsched root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.14-pre6+preempt+netdev_random+ext3_0.9.14-2414p5
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre6+preempt+netdev_random+ext3_0.9.14-2414p5 root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.14-pre6+preempt+netdev_random+ext3_0.9.14-2414p5 (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre6+preempt+netdev_random+ext3_0.9.14-2414p5 root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.14-pre6+ext3_0.9.14-2414p5
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre6+ext3_0.9.14-2414p5 root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.14-pre6+ext3_0.9.14-2414p5 (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre6+ext3_0.9.14-2414p5 root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.14-pre6
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre6 root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.14-pre6 (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-pre6 root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.14-ext3-2.4-0.9.14-2414p8
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-ext3-2.4-0.9.14-2414p8 root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.14-ext3-2.4-0.9.14-2414p8 (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14-ext3-2.4-0.9.14-2414p8 root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.14
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14 root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.14 (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.14 root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.13freeswan-1.91+ac5+preempt+netdev_random+vm_freeswap
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.13freeswan-1.91+ac5+preempt+netdev_random+vm_freeswap root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.13freeswan-1.91+ac5+preempt+netdev_random+vm_freeswap (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.13freeswan-1.91+ac5+preempt+netdev_random+vm_freeswap root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.12-ac5+acct-entropy+preempt+netdev-ramdom+vm-free-swapcache.old
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.12-ac5+acct-entropy+preempt+netdev-ramdom+vm-free-swapcache.old root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.12-ac5+acct-entropy+preempt+netdev-ramdom+vm-free-swapcache.old (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.12-ac5+acct-entropy+preempt+netdev-ramdom+vm-free-swapcache.old root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.12-ac5+acct-entropy+preempt+netdev-ramdom+vm-free-swapcache
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.12-ac5+acct-entropy+preempt+netdev-ramdom+vm-free-swapcache root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.12-ac5+acct-entropy+preempt+netdev-ramdom+vm-free-swapcache (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.12-ac5+acct-entropy+preempt+netdev-ramdom+vm-free-swapcache root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.12-ac3+netdev_ramdom+preempt+vm_hogstop2.old
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.12-ac3+netdev_ramdom+preempt+vm_hogstop2.old root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.12-ac3+netdev_ramdom+preempt+vm_hogstop2.old (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.12-ac3+netdev_ramdom+preempt+vm_hogstop2.old root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.12-ac3+netdev_ramdom+preempt+vm_hogstop2+account-rand-cleanup
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.12-ac3+netdev_ramdom+preempt+vm_hogstop2+account-rand-cleanup root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.12-ac3+netdev_ramdom+preempt+vm_hogstop2+account-rand-cleanup (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.12-ac3+netdev_ramdom+preempt+vm_hogstop2+account-rand-cleanup root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.12-ac3+netdev_ramdom+preempt+vm_hogstop2
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.12-ac3+netdev_ramdom+preempt+vm_hogstop2 root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.12-ac3+netdev_ramdom+preempt+vm_hogstop2 (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.12-ac3+netdev_ramdom+preempt+vm_hogstop2 root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.10-ac11+smp+preempt+vm_hogstop.old
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac11+smp+preempt+vm_hogstop.old root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.10-ac11+smp+preempt+vm_hogstop.old (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac11+smp+preempt+vm_hogstop.old root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.10-ac11+smp+preempt+vm_hogstop
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac11+smp+preempt+vm_hogstop root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.10-ac11+smp+preempt+vm_hogstop (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac11+smp+preempt+vm_hogstop root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.10-ac10-smp-preempt.old
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac10-smp-preempt.old root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.10-ac10-smp-preempt.old (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac10-smp-preempt.old root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.10-ac10-smp-preempt
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac10-smp-preempt root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.10-ac10-smp-preempt (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac10-smp-preempt root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.10-ac10
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac10 root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.10-ac10 (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac10 root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.10-ac4.old
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac4.old root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.10-ac4.old (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac4.old root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.10-ac4
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac4 root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.10-ac4 (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac4 root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.4.10-ac3
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac3 root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.4.10-ac3 (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.4.10-ac3 root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.2.20pre10_raid-2219A1_ext3-007a_eide-05042001
root		(hd0,0)
kernel		/boot/vmlinuz-2.2.20pre10_raid-2219A1_ext3-007a_eide-05042001 root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.2.20pre10_raid-2219A1_ext3-007a_eide-05042001 (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.2.20pre10_raid-2219A1_ext3-007a_eide-05042001 root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.2.19.old
root		(hd0,0)
kernel		/boot/vmlinuz-2.2.19.old root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.2.19.old (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.2.19.old root=/dev/hda1 ro vga=extended rootflags=data=journal single

title		Debian GNU/Linux, kernel 2.2.19
root		(hd0,0)
kernel		/boot/vmlinuz-2.2.19 root=/dev/hda1 ro vga=extended rootflags=data=journal

title		Debian GNU/Linux, kernel 2.2.19 (recovery mode)
root		(hd0,0)
kernel		/boot/vmlinuz-2.2.19 root=/dev/hda1 ro vga=extended rootflags=data=journal single

### END DEBIAN AUTOMAGIC KERNELS LIST

[-- Attachment #2: update-grub --]
[-- Type: text/plain, Size: 9929 bytes --]

#!/bin/sh
#
# Insert a list of installed kernels in a grub menu.lst file
#   Copyright 2001 Wichert Akkerman <wichert@linux.com>
#
# This file is free software; you can redistribute it and/or modify it
# under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
# General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software
# Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
#
# Contributors:
#	Jason Thomas <jason@debian.org>
#	David B.Harris <dbarclay10@yahoo.ca>
#	Marc Haber <mh@zugschlus.de>
#	Crispin Flowerday <crispin@zeus.com>

## StartOPTIONS
# name of file menu is stored in
menufile="menu.lst"

# directory's to look for the grub installation and the menu file
grubdirs="/boot/grub /boot/boot/grub"

# Default kernel options, overidden by the kopt statement in the menufile.
kopt="root=/dev/hda1 ro"

# Drive(in GRUB terms) where the kernel is located. If a seperate
# partition, this would be mounted under /boot, overridden by the kopt statement in menufile
groot="(hd0,0)"

# should grub create the alternative boot options in the menu
alternative="true"

# should grub lock the alternative boot options in the menu
lockalternative="false"

# options to use with the alternative boot options
altoptions="(recovery mode) single"

# Default options to use in a new menu.lst . This will only be used if menu.lst
# doesn't already exist. Only edit the lines between the two "EOF"s. The others are
# part of the script.
newtemplate=$(tempfile)
cat > "$newtemplate" <<EOF
## default num
# Set the default entry to the entry number NUM. Numbering starts from 0, and
# the entry number 0 is the default if the command is not used.
#
# You can specify 'saved' instead of a number. In this case, the default entry
# is the entry saved with the command 'savedefault'.           
default		0

## timeout sec
# Set a timeout, in SEC seconds, before automatically booting the default entry
# (normally the first entry defined).
timeout		5

# Pretty colours
color cyan/blue white/blue

## password ['--md5'] passwd
# If used in the first section of a menu file, disable all interactive editing
# control (menu entry editor and command-line)  and entries protected by the
# command 'lock'
# e.g. password topsecret
#      password --md5 \$1\$gLhU0/\$aW78kHK1QfV3P2b2znUoe/
# password topsecret

#
# examples
#
# title		Windows 95/98/NT/2000
# root		(hd0,0)
# makeactive
# chainloader	+1
#
# title		Linux
# root		(hd0,1)
# kernel	/vmlinuz root=/dev/hda2 ro
#

EOF
## End OPTIONS

# Make sure we use the standard sorting order
LC_COLLATE=C
# Magic markers we use
start="### BEGIN AUTOMAGIC KERNELS LIST"
end="### END DEBIAN AUTOMAGIC KERNELS LIST"

# Abort on errors
set -e

abort() {
	echo >&2
	echo "$@" >&2
	echo >&2
	exit 1
}

# Compares two version strings A and B
# Returns -1 if A<B
#          0 if A==B
#          1 if A>B
# This compares version numbers of the form
# 2.4.14-random > 2.4.14-ac10 > 2.4.14 > 2.4.14-pre2 > 
# 2.4.14-pre1 > 2.4.13-ac99
CompareVersions()
{
	# First split the version number and remove any '.' 's or dashes
	v1=$(echo $1 | sed -e 's![^0-9]\+! & !g' -e 's![\.\-]!!g')
	v2=$(echo $2 | sed -e 's![^0-9]\+! & !g' -e 's![\.\-]!!g')

	# also convert ac -> 50, pre -> -50 and anything else to 99
	v1=$(echo $v1 | sed -e 's!ac!50!g' -e 's!pre!-50!g' -e 's![^-0-9 ]\+!99!g')
	v2=$(echo $v2 | sed -e 's!ac!50!g' -e 's!pre!-50!g' -e 's![^-0-9 ]\+!99!g')

	result=0; v1finished=0; v2finished=0;
	while [ $result -eq 0 -a $v1finished -eq 0 -a $v2finished -eq 0 ];
	do
		if [ "$v1" = "" ]; then
			v1comp=0; v1finished=1
		else
			set -- $v1; v1comp=$1; shift; v1=$*
		fi

		if [ "$v2" = "" ]; then
			v2comp=0; v2finished=1
		else
			set -- $v2; v2comp=$1; shift; v2=$*
		fi

		if   [ $v1comp -gt $v2comp ]; then result=1
		elif [ $v1comp -lt $v2comp ]; then result=-1
		fi
	done

	# finally return the result
	echo $result
}

echo  -n "Searching for GRUB installation directory ... "

for d in $grubdirs ; do
	if [ -d "$d" ] ; then
		dir="$d"
		break
	fi
done

if [ -z "$dir" ] ; then
	abort "GRUB is not installed. To install grub, you may use the 'grub-install' command"
else
	echo "found: $dir ."
fi

echo -n "Testing for an existing GRUB menu.list file... "

# Test if our menu file exists
if [ -f "$dir/$menufile" ] ; then
	menu="$dir/$menufile"
	unset newtemplate
	echo "found: $menu ."
else
	# if not ask user if they want us to create one
	menu="$dir/$menufile"
	echo
	echo
	echo -n "Could not find $menu file. "
	echo -n "Would you like one generated for you? "
	echo -n "(y/N) "
	read answer
	case "$answer" in
		y* | Y*)
		cat "$newtemplate" > $menu
		unset newtemplate
		;;
		*)
		abort "Not creating menu.lst as you wish"
		;;
	esac
fi

# Extract the kernel options to use
tmp=$(sed -ne 's/# kopt=\(.*\)/\1/p' $menu)
[ -z "$tmp" ] || kopt="$tmp"

# Extract the grub root
tmp=$(sed -ne 's/# groot=\(.*\)/\1/p' $menu)
[ -z "$tmp" ] || groot="$tmp"

# Extract the old recovery value
tmp=$(sed -ne 's/# recovery=\(.*\)/\1/p' $menu)
[ -z "$tmp" ] || alternative="$tmp"

# Extract the alternative value
tmp=$(sed -ne 's/# alternative=\(.*\)/\1/p' $menu)
[ -z "$tmp" ] || alternative="$tmp"

# Extract the lockalternative value
tmp=$(sed -ne 's/# lockalternative=\(.*\)/\1/p' $menu)
[ -z "$tmp" ] || lockalternative="$tmp"

# Generate the menu options we want to insert
buffer=$(tempfile)
echo $start >> $buffer
echo "## lines between the AUTOMAGIC KERNELS LIST markers will be modified" >> $buffer
echo "## by the debian update-grub script except for the default optons below" >> $buffer
echo >> $buffer
echo "## DO NOT UNCOMMENT THEM, Just edit them to your needs" >> $buffer
echo >> $buffer
echo "## ## Start Default Options ##" >> $buffer

echo "## default kernel options" >> $buffer
echo "## e.g. kopt=\"root=/dev/hda1 ro\"" >> $buffer
echo "# kopt=$kopt" >> $buffer
echo >> $buffer

echo "## default grub root device" >> $buffer
echo "## e.g. groot=(hd0,0)" >> $buffer
echo "# groot=$groot" >> $buffer
echo >> $buffer

echo "## should update-grub create alternative boot options" >> $buffer
echo "## e.g. alternative=true" >> $buffer
echo "##      alternative=false" >> $buffer
echo "# alternative=$alternative" >> $buffer
echo >> $buffer

echo "## should update-grub lock alternative boot options" >> $buffer
echo "## e.g. lockalternative=true" >> $buffer
echo "##      lockalternative=false" >> $buffer
echo "# lockalternative=$lockalternative" >> $buffer
echo >> $buffer

echo "## altoption boot targets option" >> $buffer
echo "## multiple altoptions lines are allowed" >> $buffer
echo "## e.g. altoptions=(extra menu suffix) extra boot options" >> $buffer
echo "##      altoptions=(recovery mode) single" >> $buffer

if ! grep -q "^# altoptions" $menu ; then
	echo "# altoptions=$altoptions" >> $buffer
else
	grep "^# altoptions" $menu >> $buffer
fi

echo >> $buffer

echo "## ## End Default Options ##" >> $buffer
echo >> $buffer

sortedKernels=""
for kern in $(/bin/ls -1vr /boot/vmlinuz-*) ; do
	# found a kernel
	newerKernels=""
	for i in $sortedKernels ; do
		res=`CompareVersions "$kern" "$i"`;
		if [ "$kern" != "" -a $res -gt 0 ] ; then
			newerKernels="$newerKernels $kern $i"
			kern=""
		else
			newerKernels="$newerKernels $i"
		fi
	done
	if [ "$kern" != "" ] ; then
		newerKernels="$newerKernels $kern"
	fi
	sortedKernels="$newerKernels"
done

for kern in $sortedKernels ; do
	kernelName=$(basename $kern)
	kernelVersion=$(echo $kernelName | sed -e 's/vmlinuz-//')

	if mount | grep -qs "on /boot " > /dev/null 2>&1 ; then
		kernel=/$kernelName
		initrd=/initrd-$kernelVersion
	else
		kernel=/boot/$kernelName
		initrd=/boot/initrd-$kernelVersion
	fi

	echo "title		Debian GNU/Linux, kernel $kernelVersion" >> $buffer
	echo "root		$groot" >> $buffer
	echo "kernel		$kernel $kopt"  >> $buffer
	if [ -e /boot/initrd-$kernelVersion ]; then
		echo "initrd		$initrd" >> $buffer
	fi
	echo "savedefault" >> $buffer
	echo >> $buffer

	# insert the alternative boot options
	if test ! x"$alternative" = x"false" ; then
		# for each altoptions line do this stuff
		sed -ne 's/# altoptions=\(.*\)/\1/p' $buffer | while read line; do
			descr=$(echo $line | sed -ne 's/\(([^)]*)\)[[:space:]]\(.*\)/\1/p')
			suffix=$(echo $line | sed -ne 's/\(([^)]*)\)[[:space:]]\(.*\)/\2/p')
			echo "title		Debian GNU/Linux, kernel $kernelVersion $descr" >> $buffer
			# lock the alternative options
			if test x"$lockalternative" = x"true" ; then
				echo "lock" >> $buffer
			fi
			echo "root		$groot" >> $buffer
			echo "kernel		$kernel $kopt $suffix"  >> $buffer
			if [ -e /boot/initrd-$kernelVersion ]; then
				echo "initrd		$initrd" >> $buffer
			fi
			echo "savedefault" >> $buffer
			echo >> $buffer
		done
	fi
done

echo $end >> $buffer

echo -n "Updating $menu ... "
# Insert the new options into the menu
if ! grep -q "^$start" $menu ; then
	cat $buffer >> $menu
else
	umask 077
	sed -e "/^$start/,/^$end/{
	/^$start/r $buffer
	d
	}
	" $menu > $menu.new
	cat $menu.new > $menu
	rm -f $buffer $menu.new
fi
echo "done"
echo

if [ ! -z $answer ]; then
cat <<EOF

Please note that configuration parameters for GRUB are stored in
$menu . You must edit this file in order to set the options
which GRUB passes to the kernel, as well as the drive which GRUB looks in to
for the kernel.

Everything on the line after "kopt=" is passed to the kernel as parameters,
and "groot=" must be set to the partition(in GRUB terms, such as "(hd0,0)")
which GRUB will load the kernel from.

After you have edited $menu , please re-run 'update-grub'.
EOF
fi

Re: is 2.4.15 really available at www.kernel.org?

[Christian Haugan Toldnes]

rpjday <rpjday@mindspring.com> writes:

 
> i swear, i am not making this up.  i just tried again, through mozilla,
> to download the file 
> www.kernel.org/pub/linux/kernel/v2.4/linux-2.4.15.tar.bz2, and it 
> completed after downloading *exactly* 155312 bytes, just as before.
> 
> getting it via ftp works fine -- it's http that's giving me this
> weird problem.   is it just me?
> 
> rday

I experienced no problems at all:
(first one with ftp, second with mozilla)

23747061 Nov 23 07:18 linux-2.4.15.tar.bz2
23747061 Nov 23 10:45 linux-2.4.15.tar.bz2-2


chr

-- 
Christian H. Toldnes

Re: is 2.4.15 really available at www.kernel.org?

[rpjday]

On 23 Nov 2001, Christian Haugan Toldnes wrote:

> rpjday <rpjday@mindspring.com> writes:
> 
>  
> > i swear, i am not making this up.  i just tried again, through mozilla,
> > to download the file 
> > www.kernel.org/pub/linux/kernel/v2.4/linux-2.4.15.tar.bz2, and it 
> > completed after downloading *exactly* 155312 bytes, just as before.
> > 
> > getting it via ftp works fine -- it's http that's giving me this
> > weird problem.   is it just me?
> > 
> > rday
> 
> I experienced no problems at all:
> (first one with ftp, second with mozilla)
> 
> 23747061 Nov 23 07:18 linux-2.4.15.tar.bz2
> 23747061 Nov 23 10:45 linux-2.4.15.tar.bz2-2
> 
then i'm just plain baffled.  using mozilla, i've tried downloading both 
2.4.15 and 2.5.0, from the main www.kernel.org page, and from the kernel
subpage.  in *every* case, the download window starts off fine with
"0K of 28716K", so it knows the right size at the beginning.

the download progresses until it reads 115K of ...K, there is a several
second pause, a brief flurry of activity, and the download terminates.
in *every* case, the final downloaded file is 155312 bytes long.

as i said, i can ftp just fine, but it sure is puzzling me why mozilla
is doing this.

ok, i'll shut up now.

rday

Re: is 2.4.15 really available at www.kernel.org?

[Stefan Smietanowski]

Hi.


> then i'm just plain baffled.  using mozilla, i've tried downloading both 
> 2.4.15 and 2.5.0, from the main www.kernel.org page, and from the kernel
> subpage.  in *every* case, the download window starts off fine with
> "0K of 28716K", so it knows the right size at the beginning.
> 
> the download progresses until it reads 115K of ...K, there is a several
> second pause, a brief flurry of activity, and the download terminates.
> in *every* case, the final downloaded file is 155312 bytes long.
> 
> as i said, i can ftp just fine, but it sure is puzzling me why mozilla
> is doing this.
> 
> ok, i'll shut up now.

Just to be on the safe side,m you're not out of disk space are you ? :))

// Stefan

Re: is 2.4.15 really available at www.kernel.org?

[rpjday]

On Fri, 23 Nov 2001, Stefan Smietanowski wrote:

> Just to be on the safe side,m you're not out of disk space are you ? :))
> 

not even close -- still 2.5 G to go under /home.

rday

Re: is 2.4.15 really available at www.kernel.org?

[kees]

Hi

Is it possible that the _temporary_ file is on another filesystem which
is low in space?

Kees


On Fri, 23 Nov 2001, rpjday wrote:

> On Fri, 23 Nov 2001, Stefan Smietanowski wrote:
>
> > Just to be on the safe side,m you're not out of disk space are you ? :))
> >
>
> not even close -- still 2.5 G to go under /home.
>
> rday
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>

Re: is 2.4.15 really available at www.kernel.org?

[Michael H. Warfield]

On Fri, Nov 23, 2001 at 08:30:06AM -0500, rpjday wrote:
> On Fri, 23 Nov 2001, Stefan Smietanowski wrote:

> > Just to be on the safe side,m you're not out of disk space are you ? :))
> > 

> not even close -- still 2.5 G to go under /home.

	Mozilla downloads to a temporary directory before saving it.
Could you be low on space in /?

> rday

	Mike
-- 
 Michael H. Warfield    |  (770) 985-6132   |  mhw@WittsEnd.com
  /\/\|=mhw=|\/\/       |  (678) 463-0932   |  http://www.wittsend.com/mhw/
  NIC whois:  MHW9      |  An optimist believes we live in the best of all
 PGP Key: 0xDF1DD471    |  possible worlds.  A pessimist is sure of it!

Re: is 2.4.15 really available at www.kernel.org?

[Ross Vandegrift]

> then i'm just plain baffled.  using mozilla, i've tried downloading both 
> 2.4.15 and 2.5.0, from the main www.kernel.org page, and from the kernel
> subpage.  in *every* case, the download window starts off fine with
> "0K of 28716K", so it knows the right size at the beginning.

Turn your MTU down to 1490, maybe smaller.  There is a broken TCP/IP stack 
or switch between you and kernel.org.

Ross Vandegrift
ross@willow.seitz.com

Re: is 2.4.15 really available at www.kernel.org?

[H. Peter Anvin]

Followup to:  <20011123151053.B13009@willow.seitz.com>
By author:    Ross Vandegrift <ross@willow.seitz.com>
In newsgroup: linux.dev.kernel
>
> > then i'm just plain baffled.  using mozilla, i've tried downloading both 
> > 2.4.15 and 2.5.0, from the main www.kernel.org page, and from the kernel
> > subpage.  in *every* case, the download window starts off fine with
> > "0K of 28716K", so it knows the right size at the beginning.
> 
> Turn your MTU down to 1490, maybe smaller.  There is a broken TCP/IP stack 
> or switch between you and kernel.org.
> 

I was just going to ask... is there a problem with either (a) Path MTU
Discovery or (b) Explicit Congestion Notification?

	-hpa
-- 
<hpa@transmeta.com> at work, <hpa@zytor.com> in private!
"Unix gives you enough rope to shoot yourself in the foot."
http://www.zytor.com/~hpa/puzzle.txt	<amsp@zytor.com>

Re: is 2.4.15 really available at www.kernel.org?

[Kristofer T. Karas]

rpjday wrote:

> [...]www.kernel.org/pub/linux/kernel/v2.4/linux-2.4.15.tar.bz2, and it
> completed after downloading *exactly* 155312 bytes, just as before.
>
> getting it via ftp works fine -- it's http that's giving me this
> weird problem.   is it just me?

You probably have a proxy server in your path that doesn't re-request partially-sent
files.  I have your problem too, using squid-cache as the proxy.  Here's what
happens:

The http server (kernel.org) sends the file to the proxy, which sends it to you.  But
because the net connection from kernel.org to the proxy is faster than it is from
proxy to you, the tcp window between proxy and kernel.org goes to zero.  When it
opens up again, kernel.org closes the connection even though the file is only
partially received.  The proxy, failing to detect that the file is smaller than the
Content-Length header said it would be, simply sends the truncated contents  your
way.

I "fixed" this bug by using 'wget' to retrieve the files for me.  It has the same
problem, obtaining a truncated file; but it is smart enough to re-request missing
byte-ranges of the file, ultimately obtaining the entire thing.

Kris