Mandrake 9.2 - dead hard disk
 

Hi all,

Ok heres my problem. I have a server running mandrake 9.2 a hard disk died with the swap partition on. I got another hard disk, stuck it in, booted into rescue and put a new swap partition on there.

I reboot, both in normal and failsafe and this happens,

Normal startup until it comes to a message saying something about /dev/sdc1 (the swap partition) having errors, do you want to check Y/N?

I select N and it freezes there, if I select yes, messages appear on the screen quicker than I can take them down and then it stops with

Code: Bad EIP value.
<0>Kernel Panic: Aiee, killing interupt handler!
In interupt handler, not syncing

Sorry for being a bit vague, if there is any logs you want just tell me where to find them and ill get them for you.

The setup is as follows;

4GB hard disk / (dev/sda)
36GB hard disk /music (dev/sdb)
4GB hard disk swap partition 100meg and the rest empty (dev/sdc)
10GB hard disk /docs (dev/hdd)
CD-ROM (dev/hda)

I want to try and fix the system without reinstalling as I have alot setup on there.

Thanks in advance

Mike

Which is the HD that died?
If you created a new swap partition, I guess that you have declared it to the system in the /etc/fstab file, haven't you?
Because if you haven't, then the kernel, at boot time, will try to mount the dead swap partition.
You can find the main .log files in /var/log/

if you have enought ram to boot without the swap disk, you can
comment out the line for loading it in /etc/fstab.

fdisk -l will list all your partitions as linux sees them.

mkswap /dev/??? makes the swap partition swappable.

swapon /dev/??? starts it up. thats how you can troubleshoot it.

when you get something working right, you can change what you
need in fstab.

thought I could help... but couldn't

Thank you for all of your help, I just commented out the line for the swap and now I get the following error, I am not sure if this is to do with the swap disk dying or whether its a totally different unrelated error; During boot it fails and comes up with this,

getpwnam failed for root
grsec: attempted resouce overstep by requesting 4096 for RLIMIT_CORE against limit 0 by (pam_console_app:404) uid/euid:0/0 gid/egid:0/0, parent (rc.sysinit:20) uid/euid:0/0 gid/egid:0/0
/etc/rc.d rc.sysinit: line 874:
404 Segmentation Fault
/sbin/pam_console_apply -r
chgrp: invalid group name 'utmp'

The hard disk that died was /dev/sdc1 and ONLY had a swap partition on it the rest was unformatted.

I really want to revive this server without reinstalling linux on it, if you can help me to do that, it would be fantastic.

Thanks so far

Mike

I left the server for a while after that and it continued loading, after half an hour of loading (and still hadnt finished) it gave up after about 25 errors.

I think it might be corrupt files from the server crashing as the hard disk died.

I was going to reinstall mandrake 9.2, but instead have opted for slackware, I have built a computer with this before and it was a good distro, so ill give it a test run on the server. I have been using mandrake since 6, so I am used to text based installs and this looks like it could make an excellent server from the onset.

Michael_Mead,

I was thinking that if you ran the mandrake install while using the option to upgrade, you should repair the system and preserve your settings/partitions (depending on your choices in diskdrake).

This isn't much you learn from this solution, but you already know its the failed HD at fault.

Take this info with a large pinch of salt. I'm still really green with linux stuff (newbie ;-)

LG

Ps. You may want to double-check with some 'Guru' about this suggestion.