Restoring Boot Loader 1

[Niclov]

Hey all
Apparently the power was interrupted at my work today, causeing all the machines to reboot. On one of the Linux servers though (Running Caldera OpenLinux 2.2), it seems LILO has been removed from the MBR and it won't boot up, stalling at LIL-. Is there any way to replace LILO without re-installing Linux? There's a fair about of information on the HDD that I'd rather not loose, if possible. Thanks,

Nic

 

[RhythmAce]

If you still have a config file (/etc/lilo.conf), then you should be able to just type lilo in your terminal and it should rewrite it. You can find out more by typing "man lilo" from your terminal. Hope this fixes your problem.

 

[krischrist]

Hi,
Better try upgrading Linux without disturbing your partitions.This should solve your problem.

regards
krischrist

http://krischrist.tripod.com

http://krischrist.xaper.com

 

[Nostradamus]

The problem is that you can´t get that far, right?

Try booting from a boot-disc and then type /sbin/lilo

Have you done any work in lilo.conf recently?

Do you have any other operating system on the machine?
You can restore the windows boot-manager by inserting a windows boot-disc and type fdisk /mbr. Thus deleting lilo from MBR. /Sören

 

[ifincham]

Hi,

That particular LILO error (stopping at LIL) means : "The second stage boot loader has been started, but it can't load the descriptor table from the map file. This is typically caused by a media failure or by a geometry mismatch."

So you have a corrupted disk and it could be a tricky one. You would need to boot somehow (using a boot floppy or CD) and have access to fsck, etc., because you probably couldn't mount any of the partitions to run the utilities already on the damaged system.

You could try -->

http://www.toms.net/rb/

and get hold of tom's root/boot floppy disk package which is designed for disaster recovery. The faq are at -->

http://www.toms.net/rb/tomsrtbt.FAQ

. You can download & create from dos if necessary. Once you've got a root console you'll first want to run fsck against all your partitions - something like this for each one :

fsck -t ext2 /dev/sda1 (example device only - use your real partitions)

Good luck!!

 

[Niclov]

Thanks for the help, got up and running again with the boot disk. When I run fsck though (on hda1) I get a warning that it can cause SEVERE filesystem damage. That expected, or are them some steps I should take first?

Also, something in mysql seems to have gotten corrupted too, know where I can find the mysql error logs so I can see what's going on? (installed --prefix=/usr/local/mysql).

-Nic

 

[ifincham]

Hi,

I'd guess you got that warning if you tried to run fsck with the drive mounted as r/w. This is not a good idea. Ordinarily during the boot sequence the system will mount partitions r/o, run fsck depending on mount count, then re-mount as r/w and continue. If you've got the system booting again in multi-user (normal) mode then you've got past that hurdle already. If you do want to run fsck manually you would go to runlevel one (either 'boot:linux 1' from a cold boot or 'init 1' from within linux) and run from there.

I think you'll find Mysql error logs would be in /usr/local/mysql/data.

Rgds