LinuxQuestions.org - LILO reinstall on dual disk system

- Slackware (https://www.linuxquestions.org/questions/slackware-14/)

- - LILO reinstall on dual disk system (https://www.linuxquestions.org/questions/slackware-14/lilo-reinstall-on-dual-disk-system-485442/)

LILO reinstall on dual disk system

error: LILO halts at "L"

I ran lilo as I was trying to help someone else and wanted to make sure I had all of the steps right and, ironically, borked my boot.

Config: Dual-boot desktop with two drives - W2K on /dev/hda, Slackware on /dev/hdb2 (swap is hdb1)

Nothing has been changed on this box in forever - no new kernels or changes to the configuration. I've tried fixing this for over a day and I'm waving a white flag.

chrooting into the existing Slack install, using either the Slack install disk or Knoppix, and running lilo gives this result:

Code:

bash-3.00# lilo

Warning: '/proc/partitions' does not exist, disk scan bypassed

Warning: The boot sector and map file are on different disks.

Added Slackware10.2 *

Added Slack2.4

Added Win2K

note: Binding proc to the chroot made the /proc warning go away, but the symptom is the same.

Here's lilo.conf:

Code:

# Start LILO global section

lba32 # Allow booting past 1024th cylinder with a recent BIOS

boot = /dev/hda

#compact # faster, but won't work on all systems.

prompt

timeout = 50

# VESA framebuffer console @ 1024x768x256

vga = 791

# ramdisk = 0 # paranoia setting

# End LILO global section

# Linux bootable partition config begins

 

image = /boot/vmlinuz

  initrd = /boot/initrd.gz

  root = /dev/hdb2

  label = Slackware10.2

  read-only

#  append="acpi=off"

 

image = /boot/vmlinuz-ide-2.4.31

#  initrd = /boot/initrd2.4.gz

  root = /dev/hdb2

  label = Slack2.4

  read-only

 

# Non-UMSDOS filesystems should be mounted read-only for checking

# Linux bootable partition config ends

# Windows bootable partition config begins

 

other = /dev/hda1

label = Win2K

table = /dev/hda

 

# Windows bootable partition config ends

Contents of /boot:

Code:

drwxr-xr-x  3 root root    592 Sep 20 10:04 .

drwxr-xr-x  20 root root    464 Sep 20 05:53 ..

lrwxrwxrwx  1 root root      37 Sep 29  2005 README.initrd -> /usr/doc/mkinitrd-1.0.1/README.initrd

lrwxrwxrwx  1 root root      25 Sep 29  2005 System.map -> System.map-generic-2.6.13

-rw-r--r--  1 root root  870274 Sep  3  2005 System.map-generic-2.6.13

-rw-r--r--  1 root root  608033 Jun  5  2005 System.map-ide-2.4.31

-rw-r--r--  1 root root    512 Mar  6  2005 boot.0300

-rw-r--r--  1 root root    512 Sep 19 03:55 boot.0340

lrwxrwxrwx  1 root root      21 Sep 29  2005 config -> config-generic-2.6.13

-rw-r--r--  1 root root  54250 Sep  3  2005 config-generic-2.6.13

-rw-r--r--  1 root root  41760 Jun  5  2005 config-ide-2.4.31

-rw-r--r--  1 root root    5032 May 21  2004 diag1.img

drwxr-xr-x  8 root root    304 Sep 29  2005 initrd-tree

-rw-------  1 root root  405639 Sep 29  2005 initrd.gz

-rw-------  1 root root  49664 Sep 20 10:04 map

lrwxrwxrwx  1 root root      22 Sep 29  2005 vmlinuz -> vmlinuz-generic-2.6.13

-rw-r--r--  1 root root 1492552 Sep  3  2005 vmlinuz-generic-2.6.13

-rw-r--r--  1 root root 1256566 Jun  5  2005 vmlinuz-ide-2.4.31

I'd rather not use GRUB. GRUB is not installed in any case, so the ideal situation would be to fix lilo and worry about GRUB later.

I also redid my initrd, just in case (using the 2.6.13 generic with Reiser).

Thoughts?

maybe you have used the lilo.conf of the boot-cd
did you try
lilo -C /yourlinuxdisk/etc/lilo.conf

campher

Thanks for the suggestion. I am using the correct one, though. I even broke down and installed grub, using the 0.97 package from the 10.2/extra CD.

Grub hangs on stage1_5 with no errors. /boot on /hdb2 is part of the root directory, using Reiser. Grub install confirms that it's emdedding reiserfs_stage1_5 in the MBR on hda, but it's not working.

reiserfsck --check returns "No corruptions found"

The lilo.conf looks ok to me. According to the man page for lilo, the "L" during boot means that the second stage boot loader has received control. Can you not boot any of the three options? Try using lilo -v -b /dev/hda and see what that does. If that doesn't work, please post the output of lilo with the -v option specifed and the output of fdisk -l.

It's working, and I can only speculate on why.

Chrooting into the existing install from the generic 2.4.31 kernel from the Slack 10.2 CD1 failed repeatedly - no errors, but hung at L.

Same with a Knoppix 3.6 CD (old 2.4 kernel). Trying to run lilo.real (not chrooted) from either of two Knoppix CDs errored with IOCTL and kernel, map issues. Unbuntu CD didn't have lilo installed, so I installed it and tried to run lilo from the Ubuntu system. Same errors.

I finally tried chroot from the Knoppix 3.9 (2.6.11, I think) ran lilo one more time and it worked.

Speculation from the folks helping me out on IRC is that I had a kernel/lilo mismatch. If I were a scientist, I't rerun lilo from the native, running machine to make sure it still worked. But, that's what started this whole mess. I'm just hoping that the upgrade to 11.0 goes well.

Thanks for the replies, folks. I appreciate it.

Glad you got it fixed. Too many mysteries in the Linux world lately :mad: