Hi all !
A minor incident, (progress bar not working anymore in bootsplash) led me to some awkward manipulations : I first wanted to reinstall all what goes along with bootsplash, and in particular sysv-rc-bootsplash. But since this package provides irreversibled patch on rc, I thought, well, lets reinstall this as well.... plus the dependencies.
Anyway, I ended up removing, to then reinstall, sysv-rc, sysvinit, modutils, sysv-rc, and sysv-rc-bootsplash (I think this last one is still on thought...). And I should have not messed with modutils, because now.... my debian doesn't boot anymore!!! I get this:
And of course, I'm not really sure about the meaning. Is there any way I could get the OS boot again, avoiding complete reinstallation of debian ?
Can I bring it back to life from a live CD; should I recompile the kernel to solve it all ? Any idea and help very welcome !!!

Do you have the old initrd? Try to boot with it.

I suppose I don't. Grubs proposes initrd.img-2-6-18, which is failing,and a recovery mode one, I tried to boot on but fails as well. Booting from a gentoo live cd and mounting the filesystem, I 've reached /boot and found out that there was a initrd.img-2-6-18.bak. I presumed that this was the "recovery mode" one; still I've updated grub's menu.lst to boot on this last one, but same error. I do have a vmlinuz-2.6.17.1,in /boot, that appears in grub's menu.lst as well. I tried to boot on it, but then, nothing happens ! The screen remains black, and there isn't any hardware activity. It just hangs... As a matter of fact, when I check the menu.lst, the part related to 2.6.17.1 doesn't have the initrd flag.

The .bak is the previous initrd that is kept.
I took the asumption that your initrd was regenerated during your problematic update, but was it really?

If I remember well, the steps are:
kernel + initrd loaded
kernel goes to initrd by uncompression initrd to /dev/ram0, a ramdisk
a specific script from initrd is launched, modules are loaded, and the real / is mounted
pivot_root is used to switch from /dev/ram0 to real /
/sbin/init is executed


So you still seem to be in the initramfs, pivot_root seems to have been executed (we see no message of error)
but /sbin/init is not found like if / wasn't found


With your livecd, can you check if you really have /sbin/init?
Also check the file /etc/inittab which is what /sbin/init reads for executing.

Sorry I have no clear idea.
In busybox, try to do some commands like ls, mount,.. to see what could have gone wrong. Not easy..

thanx nx5000 for your answer. If you're still interested in this issue, you can check here
where I'm being "supported".

This seems very similiar:
http://www.linuxquestions.org/questi...d.php?t=519141

You're right nx5000, it just is the same issue... But I've been a bit forward, and since /sbin/init seemed to be missing,I've copied it over from a ubuntu live cd, and I now get:
Then... my root's login and psswrd don't work of course. Should I get back to the same state before the copying of /sbin/init and try HappyTux solution ? I'm gonna do that and see what happens. If you feel inspired, go ahead, let me know, thanx again for your concern.

Well your user/password are not stored in /sbin/init so it should work. The (none) would mean that it was not able to mount a partition and then probably no /etc/passwd file is found.
Yes, use what HappyTux said, its much cleaner than getting "something" from Ubuntu.

You can still boot this broken debian by appending init=/bin/bash in grub menu.lst

I'm now posting in the thread you pointed at me nx5000, thanx again