Need to find source of hard crash that caused LILO damage
 

I built a computer for someone and have been running it for a week until she picks it up, just to be sure it all works OK. She's never used Linux so I gave her Mandrake 9.1, since I've known it to be very stable.

I just experienced a HARD system crash, and worse. Keyboard and mouse totally froze. I had to hit the reset button to turn it off and reboot. When it rebooted, LILO gave a second-stage error message "LI 99 99 99..." indicating the boot map was corrupted in some way. I could only boot up using an emergency boot floppy. Once up, I ran LILO from the command line, rebooted again, and everything "seems" fine again.

I used components identical or similar to my own box because they've always been reliable. Nothing cutting-edge, external hardware serial modem, PCI soundcard, PS2 mouse and keyboard, Antec True Power 400 Watt PS, running VERY cool...
LILO is configured with acpi=off.

At the time it crashed, I was in X (KDE) and online (KPPP), using Opera and playing .wav files from the harddrive using XMMS. That's ALL that was running. I have aRts setup to NOT run at startup, since that's the only way to get sound right in certain games (oddly enough though, the login music always plays when I start KDE.)
XMMS only works with the OSS sound driver, although I use my PCI soundcard with the default ALSA drivers.

I really want to know what happened so I can prevent it from happening again. Is it a problem with XMMS, or Supermount, or the setpci command (see below)? This woman has hardly used a computer and I want it to be as trouble-free as possible for her. Here's some info but please tell me if I need to post something else. Thanks for taking the time to read this. Siri Amrit

Mandrake 9.1 using kernel 2.4.21-0.13mdk and ReiserFS.
Duron 1.4 processor
PC-Chips M811LU MoBo with 256 MB Crucial PC-2100 RAM.

Here's the output from lspci:

[root@localhost norma]# lspci
00:00.0 Host bridge: VIA Technologies, Inc. VT8366/A/7 [Apollo KT266/A/333]
00:01.0 PCI bridge: VIA Technologies, Inc. VT8366/A/7 [Apollo KT266/A/333 AGP]
00:0c.0 Multimedia audio controller: C-Media Electronics Inc CM8738 (rev 10)
00:10.0 USB Controller: VIA Technologies, Inc. USB (rev 80)
00:10.1 USB Controller: VIA Technologies, Inc. USB (rev 80)
00:10.2 USB Controller: VIA Technologies, Inc. USB (rev 80)
00:10.3 USB Controller: VIA Technologies, Inc. USB 2.0 (rev 82)
00:11.0 ISA bridge: VIA Technologies, Inc. VT8235 ISA Bridge
00:11.1 IDE interface: VIA Technologies, Inc. VT82C586/B/686A/B PIPC Bus Master IDE (rev 06)
01:00.0 VGA compatible controller: nVidia Corporation NV17 [GeForce4 MX 440] (rev a3)

Here's the /etc/fstab:

/dev/hda1 / reiserfs notail 1 1
none /dev/pts devpts mode=0620 0 0
/dev/hda5 /home reiserfs notail 1 2
none /mnt/cdrom supermount dev=/dev/scd0,fs=auto,ro,--,iocharset=iso8859-1,codepage=850,umask=0 0 0
none /mnt/floppy supermount dev=/dev/fd0,fs=auto,--,iocharset=iso8859-1,sync,codepage=850,umask=0 0 0
none /proc proc defaults 0 0
/dev/hda2 swap swap defaults 0 0
/dev/scsi/host0/bus0/target0/lun0/cd /mnt/cdrecorder auto ro,noauto,user,exec 0 0


FWIW, I put the following Athlon/Duron Power-Saving command in /etc/rc.d/rc.local:

#Athlon/Duron Power-Saving with KT266A Chipset:

setpci -v -H1 -s 0:0.0 92=$(printf %x $((0x$(setpci -H1 -s 0:0.0 92) | 0x80)))

setpci -v -H1 -s 0:0.0 95=$(printf %x $((0x$(setpci -H1 -s 0:0.0 95) | 0x02)))

At the time it crashed, the CPU was running at 25 degrees C (less than 75F.)

Here are some things I found in /var/log/syslog just before the system crashed:

Jan 31 08:20:33 localhost kernel: DROPPED IN=ppp0 OUT= MAC= SRC=216.203.251.26 DST=216.203.252.210 LEN=48 TOS=0x00 PREC=0x00 TTL=125 ID=8867 DF PROTO=TCP SPT=3327 DPT=135 SEQ=1395750106 ACK=0 WINDOW=8160 RES=0x00 SYN URGP=0 OPT (0204055001010402)
Jan 31 08:20:33 localhost kernel: ABORTED IN=ppp0 OUT= MAC= SRC=64.73.96.21 DST=216.203.252.210 LEN=40 TOS=0x00 PREC=0x00 TTL=46 ID=2675 DF PROTO=TCP SPT=80 DPT=1321 SEQ=212465311 ACK=0 WINDOW=24840 RES=0x00 RST URGP=0
Jan 31 08:20:34 localhost kernel: sr0: CDROM not ready. Make sure there is a disc in the drive.
Jan 31 08:20:57 localhost last message repeated 23 times
Jan 31 08:20:57 localhost kernel: DROPPED IN=ppp0 OUT= MAC= SRC=216.200.6.3 DST=216.203.252.210 LEN=92 TOS=0x00 PREC=0x00 TTL=114 ID=28556 PROTO=ICMP TYPE=8 CODE=0 ID=512 SEQ=13008
Jan 31 08:20:58 localhost kernel: sr0: CDROM not ready. Make sure there is a disc in the drive.
Jan 31 08:21:29 localhost last message repeated 31 times
Jan 31 08:21:32 localhost last message repeated 3 times
Jan 31 08:21:32 localhost kernel: DROPPED IN=ppp0 OUT= MAC= SRC=216.205.78.234 DST=216.203.252.210 LEN=92 TOS=0x00 PREC=0x00 TTL=112 ID=13528 PROTO=ICMP TYPE=8 CODE=0 ID=512 SEQ=20968
Jan 31 08:21:33 localhost kernel: sr0: CDROM not ready. Make sure there is a disc in the drive.
Jan 31 08:22:04 localhost last message repeated 31 times
Jan 31 08:23:05 localhost last message repeated 61 times
Jan 31 08:23:37 localhost last message repeated 32 times
Jan 31 08:23:37 localhost kernel: ABORTED IN=ppp0 OUT= MAC= SRC=64.254.14.170 DST=216.203.252.210 LEN=40 TOS=0x00 PREC=0x00 TTL=41 ID=0 DF PROTO=TCP SPT=80 DPT=1340 SEQ=3123938492 ACK=0 WINDOW=0 RES=0x00 RST URGP=0
Jan 31 08:23:38 localhost kernel: sr0: CDROM not ready. Make sure there is a disc in the drive.
Jan 31 08:23:39 localhost kernel: sr0: CDROM not ready. Make sure there is a disc in the drive.
Jan 31 08:23:39 localhost kernel: ABORTED IN=ppp0 OUT= MAC= SRC=64.254.14.170 DST=216.203.252.210 LEN=40 TOS=0x00 PREC=0x00 TTL=41 ID=0 DF PROTO=TCP SPT=80 DPT=1340 SEQ=3123938493 ACK=0 WINDOW=0 RES=0x00 RST URGP=0
Jan 31 08:23:40 localhost kernel: sr0: CDROM not ready. Make sure there is a disc in the drive.
Jan 31 08:24:11 localhost last message repeated 31 times
Jan 31 08:25:12 localhost last message repeated 61 times
Jan 31 08:26:13 localhost last message repeated 61 times
Jan 31 08:27:14 localhost last message repeated 61 times
Jan 31 08:28:15 localhost last message repeated 61 times
Jan 31 08:28:26 localhost last message repeated 11 times
Jan 31 08:28:26 localhost kernel: DROPPED IN=ppp0 OUT= MAC= SRC=216.205.87.164 DST=216.203.252.210 LEN=92 TOS=0x00 PREC=0x00 TTL=112 ID=47018 PROTO=ICMP TYPE=8 CODE=0 ID=512 SEQ=44802
Jan 31 08:28:27 localhost kernel: sr0: CDROM not ready. Make sure there is a disc in the drive.
Jan 31 08:28:58 localhost last message repeated 31 times

I'm afraid i can't help with the crash, but:
do you mean OSS drivers for XMMS (not OGG)?
if you do, there is an ALSA plugin for XMMS that gives you the option of using ALSA for output.

btw, if you have a corrupted boot map, that suggests that you're mounting your /boot directory at boot time. Don't. Put the /boot directory in a separate partition (if you haven't already done so) and add the option "noauto" in /etc/fstab. This will protect your boot directory from damage.

You're right, I meant OSS. I'll edit that, and get the ALSA plugin.

I don't have a separate /boot partition, only / and /home. /boot is inside /. Could you explain the steps for doing this a little more fully, please?I've never made new partitions and moved contents into them on a running system before; have only done it when installing a new system. I have read the man pages for lilo and lilo.conf but it's still confusing.

Here's /etc/lilo.conf:

boot=/dev/hda
map=/boot/map
vga=normal
default="LinuxMandrake"
keytable=/boot/us.klt
prompt
nowarn
timeout=300
message=/boot/message
menu-scheme=wb:bw:wb:bw
image=/boot/vmlinuz

label="LinuxMandrake"
root=/dev/hda1
initrd=/boot/initrd.img
append="quiet devfs=mount hdc=ide-scsi acpi=off"
vga=788
read-only
image=/boot/vmlinuz

label="linux-nonfb"
root=/dev/hda1
initrd=/boot/initrd.img
append="devfs=mount hdc=ide-scsi acpi=off"
read-only
image=/boot/vmlinuz

label="failsafe"
root=/dev/hda1
initrd=/boot/initrd.img
append="failsafe devfs=nomount hdc=ide-scsi acpi=off"
read-only

other=/dev/fd0
label="floppy"
unsafe

Thanks!

Your lilo.conf looks fine. The only change you should make is for readability. Put a blank line before each instance of <label="xxx">. This will allow you to see the boot options more clearly if you ever need to edit them.

I have never repartitioned the hard drive in a running system, but I know it can be done without damaging the data. There was a thread recently about a program called "parted" which can apparently do useful things. Your hard drive should look something like this, once you've repartitioned:
/dev/hda1 mounted on /boot, size 15MB, filesystem ext2
/dev/hda2 mounted on /, size whatever you want, filesystem reiserfs (or ext3)
/dev/hda3 mounted on /home, size whatever you want, filesystem reiserfs (or ext3)
/dev/hda4 size twice your RAM, no filesystem, no mount point

(the order of the partitions doesn't have to be the same, but your /boot partition should be first)
You will then need to change /etc/fstab (although from memory, you might be able to use diskdrake in the Control Center) to reflect the changes. Make sure the /boot partition has the option "noauto". This will prevent it from being mounted auotomatically, which will make its data safe in the event of a system crash.

Thanks again!
So, once I make the /boot partition, I take the data from /boot that's now inside of / and just plop it in there? And, I make /boot bootable? I know this sounds dumb to ask, but I spent a long time tweaking this system, doing all the configurations for this woman, and I don't want to screw up and have to start all over.

****************************************************
Some scientists claim that hydrogen, because it is so plentiful, is the basic building block of the universe. I dispute that. I say there is more stupidity than hydrogen, and that is the basic building block of the universe.

Frank Zappa
****************************************************

Basically, yes. But if you install LILO in the MBR (which is the default, I think) you don't need to make /boot bootable. But don't forget to:
1. add "noauto" to the boot partition line in /etc/fstab
2. You should probably run "lilo" again once you've finished to make sure it recognises the changes.