Cant boot my old pc
 

Hello :),

I have this error showing when booting my debian:

Driver 'pcspkr' is already registered, aborting.

I have read that I should disable snd-pscp module in /etc/modprobe.d/blacklist.conf. But after I did that, I'm still unable to boot my system.

This is what displays:

[10.175214] [drm] nouveau 0000:01:00.0: Setting dpms mode 0 on vga encoder (output 0)
[10.175223] [drm] nouveau 0000:01:00.0: Output VGA-1 is running on CRTC 0 using output 0
[10.181152] Console: switching to colour frame buffer device 240x67
[10.188268] registered panic notifier
[10.188302] [drm] Initialized nouveau 0.0.15 20090420 for 0000:01:00.0 on minor 0
udevadm settle - timeout of 180 seconds reached, the event queue contains:
/sys/devices/pci0000:00/0000:00:1f.1/host0/target0:0:1/0:0:1:0/block/sda (718)
/sys/devices/pci0000:00/0000:00:1f.1/host0/target0:0:1/0:0:1:0/block/sda/sda1 (719)
/sys/devices/pci0000:00/0000:00:1f.1/host0/target0:0:1/0:0:1:0/block/sda/sda2 (720)
/sys/devices/pci0000:00/0000:00:1f.1/host0/target0:0:1/0:0:1:0/block/sda/sda5 (721)
/sys/devices/pci0000:00/0000:00:1f.1/host0/target0:0:1/0:0:1:0/block/sda/sda6 (722)
/sys/devices/pci0000:00/0000:00:1f.1/host1/target1:0:0/1:0:0:0/block/sdb (731)
/sys/devices/pci0000:00/0000:00:1f.1/host1/target1:0:0/1:0:0:0/block/sdb1 (732)
done (timeout).
udevd[294]: worker [327] unexpectedly returned with status 0x0100
udevd[294]: worker [327] failed while loading '/devices/pci0000:00/0000:00:1f.1/host0/target0:0:1/0:0:1:0/block/sda'
udevd[294]: worker [317] unexpectedly returned with status 0x0100
udevd[294]: worker [317] failed while loading '/devices/pci0000:00/0000:00:1f.1/host0/target0:0:1/0:0:1:0/block/sdb'

I can type text on screen, but cannot run any command (nothing happens on enter). I thought it may be problem with my partitions, but I'm able to mount them using live distributions. Anyway, I have checked them using e2fsck and even repaired superblock successfully. But nothing has changed, still got the same error :(

Please, what can cause the problem? What would you do (try) in my situation?
Thank you in advance

What's tha box? Does it really have a 240x67 screen?

I dont know why theres "240x67" ... I have there installed server version of debian without graphic manager. It is connected to monitor with native resolution 1920x1080

I would try booting from an install dvd or cd, and run an fsck on the disks. That failing, you can add

init=/bin/bash

to the boot line, but you halt with a ro / and one terminal and you have to start from there
Then go to /etc/inittab, set the console run level, usually something like

and see if you can boot that. I'm suspicious of some interrupt lockup in the pci bus. Other possible boot options on the kernel command line

noapic
or

pcie_aspm=force

or

pcie_aspm=off

Thank you for helping me,

Unfortunately, I'm not sure if I understand what I should do.

I have hiren's cd and theres Parted Magic live distro (I was able to mount my filesystem using it and backup some files). I have already checked partitions with e2fsck.

So, just to be sure, you want me to:

1. In Grub, edit boot options and add "init=/bin/bash"
2. Rewrite line with "rw /" option to "ro /"
3. edit /etc/inittab and set to "id:3:initdefault:"
4. restart and boot

if this doesnt work, I should edit grub boot options and add "noapic" or "pcie_aspm=force" or "pcie_aspm=off"

Is this right? Thank you

I have eddited /etc/inittab and replaced 2 to 3 so I have there "id:3:initdefault"

After this still unable to boot

So I kept it there and tried to edit grub options and added "pcie_aspm=force"
linux /boot/vmlinuz-2.6.32-5-686 root=UUID=... ro pcie_aspm=force quiet

After this I was still unable to boot, but 2 new lines has appeared:

[12.038926] intel8x0_measure_ac97_clock: measured 55606 usecs (2672 samples)
[12.039006] intel8x0: clocking to 48000

So I tried to replace pcie_aspm=force with pcie_aspm=off
linux /boot/vmlinuz-2.6.32-5-686 root=UUID=... ro pcie_aspm=off quiet

After this, still unable to boot (no new messages on screen)

So I tried "noapic" option
linux /boot/vmlinuz-2.6.32-5-686 root=UUID=... ro quiet noapic

After this still unable to boot and hangs on messages I have written in first post.

What next can I try, should I keep edits made to inittab?

In /etc/inittab, there are usually comments that explain the runlevels It will have a list of numbers, usually 0 to 6 and tell you what they are. You want multiuser mode and not X11. What you provided is not enough for a serious diagnosis. Normally
1. you see grub stuff
2. It loads a kernel
3. It boots the kernel
4. It runs init, and finds the other disks
5. All your startup services are processed.
6. you get a bash console.

Exactly where do you get stuck - 'not boot' isn't enough. We can't fix vague errors, but we can diagnose specific errors. It would be good to see or know as much as you can show us.

I'm sorry I can't tell what stage it does hang on.

Here is my inittab
I will try to describe my problem as clearly as I can:
1. power on machine
2. see grub options
3. choose "Debian GNU/Linux, with Linux 2.6.32-5.686"
4. there are some messages, but I cant rewrite them they are quick
5. Messages from my first post shows, I can type text, but cant run any command (nothing happens after enter (you can see I wrote helo,help,pwd there))

I have made some pics of those quick messages and last screen is what I see and where it stops.

Also, I have uploaded some logs and more pics here http://buksy.netkosice.sk/senior/ I tried to empty logs first and then reboot, but they were still emtpy after reboot, so there are backups

Will this help to determine what stage is the problem one? What more can I do? Thank you

It seems to be the kernel that's running when the error occurs. Kernels usually moan big time and panic and fill your screen with bilge, but this just seems to be hanging. That sequence suggests a hardware problem.

Divide and conquer. Try booting from a live cd, or the like. It's got a different kernel. If it's a really old pc you might have issues with interrupt sharing, because that wasn't always possible. There's even a floppy linux that will boot to a console - and you can make the floppy under m$ windows - get the .zip to do that. To make under linux, use the tgz.
http://www.toms.net/rb

I'm also suspicious of that nouveau module, because all the errors come just after it loads. What exactly is the pc, & how old? Much Ram? etc.

What's the GPU?

It's about 10 years old. Bios writes copyright 1984-2001 :), I'm not sure about the ram, here are the specs:
Intel Celeron 1.3GHz
Ram 3x128Mb or 256MB+64MB (probably this)
Geforce 2 MX 400 (64MB)
HDD 1x20GB & 1x140GB

I found my debian install cd, it's not so old 6.0.3 (current 6.0.7). I was able to run rescue mode
and it has started shell from my linux root partition. So it may be that newer kernel doesn't work.
I'm thinking about downgrading kernel (gotta google it), but, isn't it better to find what kernel
feature is causing the problem? How will I update my system if I won't be able to update kernel?

What should I do, should I try to replace kernel with older one to see if it works and then upgrade again and
try to set the new kernel to work with my system?

EDIT: PS: If I am able to run shell from my partition, isn't it possible to easily install older kernel using apt-get ?

Geforce MX-400 :-))).
I have one of those - working still. Don't use it. They have passed the stage that they could be given away free in breakfast cereal boxes. I got nervous because mesa kept getting updated, and the binary blob overwrites mesa libs (LibGL.so & co) with proprietary ones of unknown vintage.

That aside, on a 10 year old box, you might try one last boot. In the file /etc/modprobe.d/blacklist, add the line
On the kernel command line, put
and try one last time. The fans are probably hard wired, but if the cpu fan is not running, turn it off quickly.

I have tried blacklisting nouveau and adding "noapic acpi=off" to the boot options but it still didn't work. This time, my computer freezes with messages from first picture

If I run my computer just with blacklisted nouveau without editing boot options, it stops with messages from second picture

Last thing I tried was running without blacklisted nouveau and with "noapic acpi=off" boot options set, result is third picture.

This may or may not help, I've had similar happen when a libc update got botched... could type a command, hit enter, but no output. Maybe something to look into.

For some reason that remains unclear to me, your box is not booting. Neither is it complaining of anything except possibly interrupts. The likelihood of a distro kernel going awol like that on even a 10 year old box is low, imho. There are some workarounds for dodgy hardware in the kernel, but you seem to have an Intel chipset, which are normally good enough.

Looking again at post #1, it's possible that udev is acting up. I've never seen bootup errors on udev, so I really don't know what to advise. Is this a fresh install? Does it boot other kernels/OSes? Like does it run from a live cd? Did anything lead up to this happening? What did you fix on it last before the problem started? You want to try for as much information as possible.

No, it is not a fresh install, it has been installed about a year ago. I am able to boot other distros (Parted Magic live, Debian rescue mode), I dont think this is a hardware failure.

I'm using this computer as web server mainly. I had there some dependency problems after installing python, had to fix it with apt-get -f. Later on I think I haddf run "apt-get update" and "apt-get upgrade".

I dont use this computer often, so I cant tell exactly what was the last action taken, but it should be upgrade.

Could you point me to any good tutorial on how to downgrade kernel? Is it necessary to run shell on that computer? Can I replace the kernel by just copying it without any install? What can be influenced and what directories should I back up first. I have backed up /boot


I have booted debian install cd in rescue mode and has run shell from my linux partition. I downloaded 'linux-image-2.6.32-5-686_2.6.32-48squeeze1_i386.deb' and tried to install it with dpkg -i. But I get the following error
dpkg: unable to access dpky status area: no such file or directory