On Wed, 25 Oct 2000, Clayton Stapleton wrote:
> Hi Debian's;
> Have installed Debian 2.2 (potato) using 3 CD's from CheapBytes.
> Things are going ok except that when I run "uname -a" the return
> is "2.2.15-4mdk" whereas was expecting "2.2.2.17pre6.deb".
>
> My system is a Pentium 166MMX, 64MB RAM, cd-rom, floppy,
> 2 hard drives 8.1G and 9.1G. Mandrake is installed on /dev/hda1
> (8.1G) with lilo using the MBR for startup. It also provides for
> starting alternate OS which I use. Mandrake was installed before
> I installed Debian.
>
> Used Mandrake to fdisk /dev/hdb so that there is 4 partitions on
> /dev/hdb. "dev/hdb1", "/dev/hdb5", "/dev/hdb6" and "/dev/hdb7".
> Used Debian to activate "dev/hdb1" as "/", "dev/hdb5" as "/usr",
> "dev/hdb6" as "/var" and "dev/hdb7" as "/home".
>
> Why didn't Debian use the kernel version 2.2.17pre6 that is on
> the CD's instead of using 2.2.15-4mdk?
>
> TIA
> Clay Stapleton
Funny. But the following post appears to answer your question. Can you
"modprobe" properly? If you can't, then your problem is probably a
misconfigured boot manager (lilo or grub?). From the suffix alone, 2.2.15-4mdk
appears to be a Mandrake kernel. If Ethan's solution doesn't work (I haven't
tried it yet), try mine (boot Debian as a non-Linux OS).
Re: Debian and Red Hat togheter
From: Ethan Benson <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Cc: debian-user@lists.debian.org
On Wed, Oct 25, 2000 at 04:49:29AM +0800, [EMAIL PROTECTED] wrote:
> I've installed multiple Linuses in combinations like Mandrake + Debian,
> Mandrake + Storm, Redhat + Debian. The biggest problem for me was the module
> loading. modprobe produces error messages about not finding the module to be
> loaded.
>
> But this may be because in LILO I installed Debian as just another Linux OS.
> In a recent reinstall I configured Debian to boot as an alien ("other")
> operating system (like winblows). Now I'm able to modprobe without too much
> fuss.
sounds like you weren't configuring lilo correctly, i assume you have
seperate / filesystems for each distro, and you did NOT have seperate
/boot partitions, you probably had a lilo.conf like this:
image=/boot/vmlinuz
root=/dev/hda1 ## redhat
read-only
image=/boot/vmlinuz
root=/dev/hda2 ## debian
read-only
on your redhat /etc and ran lilo from redhat. this would boot your
REDHAT kernel using your Debian root filesystem (and thus debian
modules which don't match the redhat kernel)
what you should have done instead is:
image=/boot/vmlinuz
root=/dev/hda1 ## redhat
read-only
image=/debian/boot/vmlinuz
root=/dev/hda2 ## debian
read-only
and mounted your debian / on /debian under redhat, THEN run lilo, and
your debian kernel will be loaded for debian and your redhat kernel
for redhat.
remember that lilo does not read /etc/lilo.conf at boot time, it only
uses it at install time to create a blocklist for the specified
kernel. the root= is not used by lilo except as an argument it passes
to the kernel. the first lilo.conf just creates one blocklist for the
redhat kernel and passes it a different root= argument.
the latter creates two blocklists, one for redhat one for debian
kernels.
