Projects / LILO / Comments

Comments for LILO

20 Feb 2003 10:13 zentara

Re: encryption of passwords


> Read my initial reply about:
>
> password = ""

Doh!!! Yes, I see how it works now. It's very good.
Thank you for your patience. (Now I can tell all my
grub friends, that lilo dosn't have a weakness. )

20 Feb 2003 10:04 johncoffman

Re: encryption of passwords

>
> % Passwords have been secured with the
> SHS
> % hash since 31-Oct-2001 (version 22.1)
>
> Sorry, I should have been more specific,
> I meant in lilo.conf.


Read my initial reply about:

password = ""


20 Feb 2003 00:16 snodx

Re: Primary HDD error
Dear John,
Sorry for the delay in responding. In any case the hardware "expert" to whom I had given the hard disk, simply threw away this hard disk and replaced it with a brand new one. So I cannot try out that rescue procedure. Thanx anyway.

SNODX


> This sounds much more like a power
> glitch wiping critical disk records.
> You'll have to boot from a rescue floppy
> or CD and see what is left of your
> disks. --John Coffman


19 Feb 2003 14:39 zentara

Re: encryption of passwords

> Passwords have been secured with the SHS
> hash since 31-Oct-2001 (version 22.1)

Sorry, I should have been more specific, I meant in lilo.conf.

19 Feb 2003 13:53 johncoffman

Re: encryption of passwords
Passwords have been secured with the SHS hash since 31-Oct-2001 (version 22.1), or obscured with a CRC digest since 29-Aug-2001 (version 22, initial release). If using version 22 or later, are you aware of the configuration file option:

password = ""

(empty double quote), which will require interactive entry and confirmation of the password?

19 Feb 2003 05:48 zentara

encryption of passwords
Is there any chance that lilo passwords will be encrypted anytime soon, like they do in grub?
It's the 1 feature I would like to see.

13 Feb 2003 19:09 johncoffman

Re: Clarification on -r option


> But what
> drive will get the boot sector updated
> when I remove the -t option? Since
> lilo.conf has boot=/dev/hda I'm afraid
> that running lilo, even with the -r
> option, will update the boot sector on
> the wrong drive (my Pentium's main
> drive, /dev/hda).


That is exactly where the configuration file specifies for it to be placed. You want to place it on the /dev/hdc drive; viz., boot=/dev/hdc.

However, this will not be enough to make the drive bootable when it is moved to the other system and installed as /dev/hda. Simply stated. the bios device codes will change.

You will also need the following in the global section (top) of the configuration file:

disk=/dev/hdc bios=0x80

This is a lie now; but it will be the truth when the drive is switched to be the primary drive on the other system. Once the drive has been moved, the configuration file will have to be updated, to reflect the new drive placement; i.e., references to hdc will need to be changed to hda.

13 Feb 2003 12:58 wtrenker

Clarification on -r option
(First, I've searched the docs and gone hunting with Google but am still stuck on this question. I also tried to find a mailing list for LILO but this is all I could find.)

I have an old 486 with a 425Meg hard drive. I've got the drive temporarily hooked up as a secondary master on my Pentium machine as /dev/hdc. My goal is to use the tools on the Pentium to configure the 425Meg drive as the Linux boot drive for the 486 where it will be /dev/hda.

The 425 drive has a single reiserfs partition (temporarily as /mnt/hdc1 on the Pentium) with a small Linux distro installed, including a /etc/lilo.conf file and a /boot directory with the usual files. The lilo.conf file is very simple:
boot=/dev/hda
map=/boot/map
install=/boot/boot.b
prompt
timeout=50
image=/boot/vmlinuz
label=Linux
root=/dev/hda1
read-only

On the Pentium I set up the 425Meg drive for booting like this:
#lilo -M /dev/hdc
#lilo -A /dev/hdc 1

Next I wanted to test the map file generation, like this:
#lilo -r /mnt/hdc1 -t
Warning: LBA32 addressing assumed
Added Linux*
The boot sector and the map file have *NOT* been altered.

The LBA warning lead me to use the -g option, like this:
#lilo -r /mnt/hdc1 -t -g
Added Linux*
The boot sector and the map file have *NOT* been altered.

Now, at last, for my question. If I remove the -t option I understand that a map file will be generated in the /boot directory of the /mnt/hdc1 drive (because of the -r option). But what drive will get the boot sector updated when I remove the -t option? Since lilo.conf has boot=/dev/hda I'm afraid that running lilo, even with the -r option, will update the boot sector on the wrong drive (my Pentium's main drive, /dev/hda).

I would appreciate clarification from those who know far more than me on this topic.

Thanks for your patience,
Bill

06 Feb 2003 07:04 johncoffman

Re: Primary HDD error
This sounds much more like a power glitch wiping critical disk records. You'll have to boot from a rescue floppy or CD and see what is left of your disks. --John Coffman

06 Feb 2003 07:00 johncoffman

Re: installed lilo, but windows drive can't be accessed from LILO
It sounds like the Windows boot record on /dev/hda1 has been overwritten, probably with an earlier "lilo.conf" file that specified boot=/dev/hda1 (not 'hda', as it is now). LILO creates backups of boot records when they are first overwritten, so you may be able to use the LILO uninstall (-u switch) feature to recover /dev/hda1. See if you have a file named /boot/boot.0301. This would be the original Windows boot record. --John Coffman

Screenshot

Project Spotlight

ReciJournal

An open, cross-platform journaling program.

Screenshot

Project Spotlight

Veusz

A scientific plotting package.