slack 12 post install boot problem (yes another)

jaguarrh · 07-28-2007, 02:57 AM

Firstly let me say i have read ftp://slackware.mirrors.tds.net/pub/..._AND_HINTS.TXT and such, but am still having problems with booting.

my system:

an old compaq presario 1240 laptop.
pentium mmx 266mhz
96mb ram
4gb ide hd.

i was previously running a trimmed down version of slack 11 with fluxbox using the 2.4 kernel, had the install size down to about 1.6gb and it was very stable.

the install:

initially i tried the huge-smp.s kernel as recommended by RW in changes and hints, but the warning about old machines in the boot message was relevant and the smp kernel would not boot. so, as instructed, i tried the huge.s kernel to run the first disc and i was able to boot the cd

i was hoping to do a similar install with slack 12 and use fluxbox or a similar wm so i repeated the install process i used for slack 11. I selected the /a, /ap, /d, /k, /l, /n, /tcl, /x, and /xap series and then did a full install to hda2 as a single / partition (ext2 format). hda1 was a 192mb swap partition.

the install seemed to go ok, i completed the post install config scripts and installed lilo to the mbr of hda. then , as usual i was instructed to exit setup and reboot with 3 fingers.

first boot

lilo started as expected and i chose the only entry... Linux

the boot process then started ok, but at the point where it...

Quote:

triggering udev events

i get the

Quote:

kobject_add failed for ohci-hcd with EEXIST, don't try to register things with the same name in the same directory

error

followed by lots of

Quote:

ata: 0x1F0 IDE port busy

type messages. this continues for some time and the machine never boots

ok, i thought, i have read about this in changes and hints...

Quote:

As stated earlier, it is recommended that you use one of the generic kernels
rather than the huge kernels; the huge kernel is primarily intended as
an "installer" and "emergency" kernel in case you forget to make an initrd.
However, if you do use one of the huge kernels, you will likely encounter
errors like this:
kobject_add failed for uhci_hcd with -EEXIST, don't try to register
These occur because the respective drivers are compiled statically into the
huge kernels but udev tries to load them anyway. These errors should be safe
to ignore, but if you really don't want them to appear, you can blacklist the
modules that try to load in /etc/modprobe.d/blacklist. However, make sure you
remove from the blacklist if you ever decide to use the (recommended)
generic kernels.

so all i need to do is boot using the cd and install a generic kernel, right?

if i boot using the cd following these instructions from "booting.txt"

Quote:

Section 1: What to do if after the installation process, the machine will
not boot.

This problem is not uncommon. The reason for it is that the kernel your system
tries to boot after installation is not the same as the one used to install (the
bootdisk kernel). Typically the kernel that is installed to your hard drive
will contain more device drivers, and sometimes one of them won't like some
piece of hardware in your machine.

If you installed using the Slackware CDROM, you should boot from this CDROM
again and then load the kernel you used while installing, with additional
parameters to make it mount your hard drive's root partition. Let's assume
your root Linux partition is on /dev/hda2 and you used the 'sata.i' kernel.
To boot the system, you'd enter this command at the "boot: " prompt:

sata.i root=/dev/hda2 noinitrd ro

which in my case means entering

Code:

huge.s  root=/dev/hda2 noinitrd ro

this brings me to the installation login prompt (after selecting the keyboard map). from here i have been able to mount hda2 and can see the files that have been installed.

what i cannot figure out is how to install the generic kernel.

after i login as root at the installer login, i seem to have very few commands available.

can anyone help me out here...?

i have questions about mkinitrd and other topics but i need to get the system booted.

any help would be greatly appreciated.

XGizzmo · 07-28-2007, 05:48 AM

Boot off the install media with huge.s root=/dev/hda2 rdinit= ro
This will have your system running with out the need to mount and chroot
then you can installpkg the kernel. Make sure you create an initrd and edit and run lilo .

Vincent_Vega · 07-28-2007, 06:38 AM

Will you actually need an initrd for the huge kernel? I think all filesystems are built into that kernel, and therefore eliminating the need. Or am I wrong about that?

jaguarrh · 07-28-2007, 06:54 PM

to xgizzmo...

i meant to add this to my first post, but when i try the line u mentioned

Code:

huge.s root=/dev/hda2 rdinit= ro

as instructed at the boot prompt, it still gives the EEXIST error just the same as if i was trying to boot from the hd.

any other ideas?

dennisk · 07-28-2007, 10:52 PM

If you installed everything the generic kernels are already installed and in /boot.

You will need a initrd with the generic kernel

Quote:

cd /boot

then (assuming ext3 is your file system)

Quote:

mkinitrd -c -m ext3

followed by

Quote:

lilo -v

will do the job.

Dennisk

jaguarrh · 07-29-2007, 03:33 AM

forgive me dennisk, but don't all of ur instructions assume that i can boot into my new install, to issue those commands?

i am unable to boot anything but the cd installer login prompt...

should i mount and chroot?

could you explain in more detail pls?

Karu · 07-29-2007, 04:02 AM

Quote:

Originally Posted by jaguarrh

but don't all of ur instructions assume that i can boot into my new install, to issue those commands?

Yes you can !

Quote:

Originally Posted by jaguarrh

i am unable to boot anything but the cd installer login prompt...

Exactly ! and there above a install login promt you should see
a hint how you can login in you freshly installed Slackware:

Code:

boot: hugesmp.s root=/dev/hda1 rdinit= ro

adapt root= to you situation and you are booting !

jaguarrh · 07-29-2007, 04:52 AM

as i have previously posted, i have tried the command you suggest

boot: huge.s root=/dev/hda2 rdinit= ro (in my case, as i initially installed with the huge.s kernel)

and my machine does not boot. it fails with the EEXIST error.

all of this is in my previous posts

onebuck · 07-29-2007, 06:55 AM

Quote:

Originally Posted by jaguarrh

as i have previously posted, i have tried the command you suggest

boot: huge.s root=/dev/hda2 rdinit= ro (in my case, as i initially installed with the huge.s kernel)

and my machine does not boot. it fails with the EEXIST error.

all of this is in my previous posts

Hi,

First, as stated the 'EEXIST' is not a error that will prevent you from operating. Just an attempt to load an already existing module. You can safely ignore them.

The easiest way would be to use the install cd1 to boot the system as if you were going to install.

Since you say that you cannot boot into your installation then just use the install cd1 to boot the system then login (hint; press return).

From the cli (command line);

Code:

~#mkdir /slacktemp                   #temporary mount point
~#mount /dev/your_device /slacktemp  #this is the device you installed to
~#chroot /slacktemp                  #change to yours
~#cd /slacktemp/etc                  #change to directory with lilo.conf 
~#vi lilo.conf                       #edit lilo.conf, if need be
~#lilo -v -t -b /dev/your_device     #sda, hda this will only test 
~#lilo -v -b /dev/your_device        #this will write MBR to your_device

You should now be able to re-boot your system.