When I try to boot after the install I get this error "Kernel panic: Attempted
to kill init" This error occurs at the end of the USB check/install/whatever at
the first boot after the install..
Ignore the line below..
(FFS OMFG GENTOO TOOK HOURS TO INSTALL AND NOW IT DOESNT WORK? WTF?!)
Steps to Reproduce:
Worked fine (letting me into Gentoo)
what kernel did you use ? we cant debug your problem w/out any information
I use the gentoo kernel.
I have encountered the same issue. Using genkernel 1.4 to build gentoo-sources (kernel-2.4.20-gentoo-r5), it builds perfectly. I then add the image to LILO using the procedure described both in the install.html and genkernel's README file. Upon a reboot, it finishes the usb-detection with finding usb storage hardware, then gives the "Kernel panic: attempted to kill init!" with my caps and scroll lock lights blinking on the kb.
My system is a Shuttle SN41G2; shuttle nforce2-based mobo, using the on-board video/audio/net etc.
That is exactly the same situation as I get!
Though I dont got nForce mobo..
Would be great if someone could come up with an solution..
See http://forums.gentoo.org/viewtopic.php?t=75643 for more info; I have not tried this myself to verify that it fixes this problem. I was able to get around it by removing the append= and initrd= lines from lilo.conf completely.
Same problem here. Architecture athlon. Followed install guide exactly, did emerge of hotplug, genkernel on gentoo-sources. Plugged provided names into lilo. On reboot - ended up with kernel panic "attempted to kill init" and flashing numlock and capslock lights. There are large numbers of people reporting this problem, which seems to have something to do with the append = line in lilo.conf. Changing the initrd to /dev/hdax rather than /dev/ram0 seems to help some people, but kind of defeats the whole purpose of using an initrd and doing hotplug detection, doesn't it?
Can you please post your /usr/src/linux/.config ? Things tend to get fixed much faster if we can see that.
I also run Gentoo 1.4 on a Shuttle SN41G2 and I was getting the Kernel Panic following the USB detect section also.
I removed the root=/dev/ram0 part of the append line in lilo.conf and it booted without an issue - seems the RAM problems listed in the docs with GRUB may be coming into play here on the nForce 2 chipset.
I'll post the .config as soon as I get the system up on its feet :)
Sven, have we resolved this?
I know I'm not Sven, but - we have a workaround rather than a fix. The USB autodetection causes a kernel panic when you use a RAM disk, although quite why we aren't sure. Using the gs-sources appears to fix this too. I'm toying with the idea of putting an entire Gentoo Linux installation on one of those USB key thingies, so this could well be an issue for me (which is why I'm particularly interested...!).
OK, this sounds like another "Gentoo-sources is out of date and our users are suffering" issue. Assigning to pfeifer.
the same for me. failed to boot with kernel generated by genkernel on gentoo-sources using lilo. installing kernel by hands (make menuconfig, ..., make bzImage and copying it to /boot worked fine). System is Intel Celeron 1.7 GHz with 845 chipset.
I'm not sure if it's related to usb autodetection. I've tried running with
nousb as kernel argument and i still get this error.
I've tried with gentoo-2.4.20-r7 and gentoo-test-2.4.22-r0.
If somebody is interested in solving this bug I'll try to be of as good help
as possible. I'll attach my .config below.
Created attachment 19274 [details]
This is my config file as it was saved by genkernel.
This happens on the amd64 beta liveCD (kernel 2.6).
and on ALL gentoo-dev-sources built with that config.
(and modified configs with ps2 keyboard/mouse drivers excluded).
Workaround was to DISABLE "legacy" USB support
side effect : cant use USB keyboard to nagigate grub menu
I've seen similar errors when using ck-sources for example, when generated with genkernel, and it would work with a hand compile, thus I don't see this as a kernel bug: I can confirm that using a hand-compiled kernel seems to work and comment #12 does as well. Can other folks confirm this?
Reassigning to Daniel as this seems to be something with the initrd voodoo going bad.
use genkernel 3.x closing