Fresh instal: kernel panic not syncing attempted kill init

Drone4four · Apprentice Joined: 09 May 2006 Posts: 247

Following the Gentoo handbook installation guide closely, I just completed a fresh install of Gentoo as a guest VM. The host is Ubuntu 14.04. I powered off the VM. I removed the live iso media from within the VirtualBox setttings menu. I start the VM. Grub is called at boot as it should. I select Gentoo GNU/Linux. Three seconds into the initramfs and my kernel panics. And here is a screenshot of the contents of my fstab and output of fdisk -l /dev/sda. That last screenshot was taken when I reached the end of the Gentoo handbook before rebooting my fresh install for the first time.

I queried Google for 'kernel panic not syncing attempted to kill init' which turned up some pretty ancient forum posts from 10 years ago with forum members discussing issues involving a possibly toasted physical (rather than virtual as in my case) IDE hard drive. I added the word 'gentoo' to that Google query which turned up this thread which is exceedingly frustrating to read. Another Gentoo forum member who goes by the name of senz found a solution (dating back to 2008) which I don't understand. senz writes:

comprookie2000 · Posted: Sun Nov 29, 2015 3:48 am Post subject:

I can't be of much help because I have never done an install in vm but just to double check you did follow the wiki guide?
https://wiki.gentoo.org/wiki/VirtualBox#Linux_guest-related
_________________
http://dev.gentoo.org/~dabbott/

NeddySeagoon · Posted: Sun Nov 29, 2015 10:27 am Post subject:

Drone4four,

Post the output of lspci from within your VM so we can see the hardware Virtualbox is pretending you have.
That's what you need to build Gentoo for.

Also, provide the content of grub.cfg from your install in the VM.

wgetpaste is your friend.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.

Drone4four · Apprentice Joined: 09 May 2006 Posts: 247

NeddySeagoon · Posted: Mon Nov 30, 2015 8:19 pm Post subject:

Drone4four,

Well, bpaste works here. You grub.cfg looks good provided that root=UUID=a1f6d306-391c-4ac8-b6c0-91e1086b0384 is correct for your filesystem on /dev/sda4.
Check with blkid. Be aware that the partition will have a UUID whicd is different to the UUID of the filesystem on the partition. Its the filesystem UUID that needs to match.

That leaves two things, your emulated hardware and your kernel. I suspect that the issue is a mismatch between the two.

Booting an old Gentoo DVD in Virtualbox and running lspci,

Drone4four · Apprentice Joined: 09 May 2006 Posts: 247

You've already seen it, Neddy, but for those who haven't seen my grub.cfg who might be viewing this thread now, here it is: https://bpaste.net/show/b04e586dbc0f

If grub.cfg is referring to my / partition using a UUID, then should UUIDs be also found in my fstab? The contents of my fstab can be found here: https://bpaste.net/show/ace822c716dc

In my fstab there are no traces of a UUID. The / partition is referred to as /dev/sda4. Does this mean I have to replace all instances of the root UUID in my grub.cfg with /dev/sda4?

Here is my output of lspci: https://bpaste.net/show/ab5ccf201125

Neddy, here are the segments you drew attention to in your previous post:

NeddySeagoon · Posted: Tue Dec 01, 2015 9:33 pm Post subject:

Drone4four,

With genkernel you may use the filesystem UUID and the partition /dev entry interchangeably.
The kernel can deal with /dev/sd.. on its own. UUIDs need the userspace mount program, which is included in the initrd, so it can be used to mount root.

genkernel should build support for both the SATA and IDE chip sets.
In your Virtual Machine set up in Virtual Box, which controller is your emulated HDD attached to?
You could try the other one.

Post your /usr/src/linux/.config to a pasebin and I'll look it over.

You could also try genkernel-next in place of genkernel.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.

Drone4four · Apprentice Joined: 09 May 2006 Posts: 247

Here are the contents of my /usr/src/linux/.config: https://bpaste.net/show/939698f27b16

The Controller for my storage tree in VirtualBox is IDE for my live iso and SATA for Gentoo.vdi. Does that answer your question? Its also worth noting that under the attributes option to the right, it says:

Drone4four · Apprentice Joined: 09 May 2006 Posts: 247

I unmerged the genkernel and successfully emerged genkernl-next. I rebooted, same kernel panic.

It's also worth noting that I emerged the VBox guest additions. Kernel panics with the same error.
_________________
My rig:
IBM Personal System/2 Model 30-286 - - Intel 80286 (16 bit) 10 Mhz - - 1MB DRAM - - Integrated VGA Display adapter
1.44MB capacity Floppy Disk - - PS/2 keyboard (no mouse)

NeddySeagoon · Posted: Wed Dec 02, 2015 7:38 pm Post subject:

Drone4four,

I've just done a fairly minimal VBox install, following the handbook (almost) , to attempt to recreate your issue.
I even used genkernel and grub2, much against my better judgement. I failed. It booted flawlessly.

I left everything possible set to the defaults.

My VBox setup is at https://bpaste.net/show/6ee781796125
That can be found in ~/VirtualBox VMs/<VM_NAME>/<VM_NAME>.vbox. Its XML. but diff may give some hints.
If you post your .vbox file I'll look at it.

My root .bash_history from inside the chroot, for the install is https://bpaste.net/show/45687b458e8e
There are a few typos, which you can see and I omitted to mount /boot before entering the chroot the first time.

I suspect its a VBox set up issue.

For completeness lspci https://bpaste.net/show/7bac1fdbbd1d
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.

Drone4four · Apprentice Joined: 09 May 2006 Posts: 247

Thank you NeddySeagoon for your kindness in helping troubleshoot my issue.

Here is my Gentoo.vbox xml file (with nice syntax highlighting) : http://pastebin.com/4hZkQYxp

Here is my root bash history file: https://bpaste.net/show/51237227d878
What's interesting is yesterday I installed the X server and e16 in my chroot environment. I tried starting e16 and it choked. But I don't see any evidence of that procedure in my root bash history file. Strange. Today when I got home from work I powered off the VM without shutting down gently because I couldn't escape the idle black screen by moving the mouse or entering random keystrokes. I know powering off the VM as such wasn't best practices or the wisest thing to do but I couldn't figure out how to get the terminal back. Could this be why there are no traces of the X server / enlightenment installation in my root bash history log file?

Here is my lspci too:

NeddySeagoon · Posted: Fri Dec 04, 2015 6:51 pm Post subject:

Drone4four,

There is nothing stands out in your root .bash_history. A few things about .bash_history.
.bash_history is only saved when the shell exits, if you open several concurrent shells, only one .bash_history is saved.
The commands in the others are lost.

You have created a normal user. Did you test X as your normal user?
If so the command will be in the users .bash_history.

Drone4four · Apprentice Joined: 09 May 2006 Posts: 247

Here is the output of uname -a:

NeddySeagoon · Posted: Sun Dec 06, 2015 10:56 am Post subject:

Drone4four,

HerbMillerJW · n00b Joined: 16 Feb 2012 Posts: 37

Do you have 3d-accelleration enabled in the VM? Try turning that off and seeing if it boots. When using VirtualBox, always disable that as a troubleshooting step. It can cause all kinds of random fits in the guest.

Drone4four · Apprentice Joined: 09 May 2006 Posts: 247

NeddySeagoon · Posted: Mon Dec 07, 2015 5:12 pm Post subject:

Drone4four,

Drone4four · Apprentice Joined: 09 May 2006 Posts: 247

My system clock has been working as it should. The problem is not with my clock but with my fast and loose approach to following the Gentoo Handbook, specifically the kernel installation section. I apologize for this. So I've been reviewing the genkernel installation section carefully tonight.

Its also very important that I explain now that I do recall reformatting /dev/sda1(a fat32 EFI partition) and /dev/sda2 (/boot) at some point a few days AFTER initially setting them up. Yes, the /boot file system was recreated. I forget exactly what I was doing or why I was reformatting these parts so late into the Handbook but this could very well be the reason why the source of my problem that we're troubleshooting here is my /boot mount point. I really should have reported this earlier, but it didn't occur to me. I apologize for my carelessness again.

NeddySeagoon · Posted: Tue Dec 08, 2015 4:35 pm Post subject:

Drone4four,

genkernel is old a and difficult to maintain ... to the point where its been rewritten as genkernel-next.
Both nominally do the same thing but genkernel-next is under activie development. Portage can tell them apart.
I expect they both support the same commands too.

/mnt/gentoo/boot being empty means that you did not mess up a kernel install by installing it to the wrong boot.
You have a boot partition and a /boot directory on the root partition. When /boot is mounted, it looks like they are the same place.
When grub runs at boot, it matters as they are different partitions.

Your genkernel-next build and install looks good.

Drone4four · Apprentice Joined: 09 May 2006 Posts: 247

The kernel panics at about the same time (about three and a half seconds into the boot sequence) but the kernel panic doesn't attempt to kill init now. Now its not syncing with a "Fatal exception in interrupt". Although it does indicate something about "snbep_uncore_msr_init_box+0x3c/0c50" which seems init related. See here: http://i.imgur.com/SF7EKpg.png

I Googled, snbep_uncore_msr_init_box which turned up a detailed VirtualBox bug report dated September of this year with a number of people discussing my issue with hardware similar to mine and with the same 4.1 kernel in the comment section.
_________________
My rig:
IBM Personal System/2 Model 30-286 - - Intel 80286 (16 bit) 10 Mhz - - 1MB DRAM - - Integrated VGA Display adapter
1.44MB capacity Floppy Disk - - PS/2 keyboard (no mouse)

NeddySeagoon · Posted: Wed Dec 09, 2015 4:45 pm Post subject:

Drone4four,

Try the 4.3 kernel or even the 3.x version that is reported to work.
That bug was closed as WONTFIX, so nothing was changed in virtualbox
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.

Drone4four · Apprentice Joined: 09 May 2006 Posts: 247

I just spent the better part of the last hour ripping my hair out following the kernel configuration section of the Gentoo Handbook. It was such a trivial struggle to find File systems in the menuconfig. I finally found it after much frustration. So I'm ready to build my kernel and then I realize that I'm configuring 4.1.12 which is what genkernel-next is based on which is the kernel that is problematic in my case as documented in this forum thread. Is there a way to use the menuconfig I just made with 4.3.x? If so, how? The Gentoo Handbook just tells you to emerge sys-kernel/gentoo-sources which right now produces the sources necessary to build 4.1.12. The Book doesn't indicate how to download anything else. Little guidance?
_________________
My rig:
IBM Personal System/2 Model 30-286 - - Intel 80286 (16 bit) 10 Mhz - - 1MB DRAM - - Integrated VGA Display adapter
1.44MB capacity Floppy Disk - - PS/2 keyboard (no mouse)

NeddySeagoon · Posted: Sat Dec 12, 2015 10:37 am Post subject:

Drone4four,

genkernel(-next) works with the kernel pointed to by the /usr/src/linux symlink.

Drone4four · Apprentice Joined: 09 May 2006 Posts: 247

I borked my native Ubuntu host. I updated my kernel, nvdia drivers and Plymouth. Ugh. I'm troubleshooting my new issue elsewhere. And these next few days I have x-mas related things to attend to. So it might be a few days before I report back here when I have my Gentoo guest accessible again. Thanks for your help so far, Neddy. Your kindness and knowledge sharing is wholeheartedly appreciated.
_________________
My rig:
IBM Personal System/2 Model 30-286 - - Intel 80286 (16 bit) 10 Mhz - - 1MB DRAM - - Integrated VGA Display adapter
1.44MB capacity Floppy Disk - - PS/2 keyboard (no mouse)

Drone4four · Apprentice Joined: 09 May 2006 Posts: 247

I'm back!

Yes, I was aware of / to invoke find in menuconfig. It was my futile attempt at using the find function that was contributing to my frustration. I felt that it wasn't helping at all. Anyways, I am happy with the way my Makefile turned out ok. I backed it up into a safe place and am ready to download either 4.2.x or 4.3.x sources, whether they be -next or something else. My problem now is figuring out how to use, eselect. According to the Handbook, eselect can be used to present news about package updates in portage. eselect can also be used to switch between portage profiles. Mine is set to [1] default/linux/amd64/13.0 *

I gather that eselect can declare what kernel symlink to use. The output of eselect kernel list is: [1] linux-4.1.12 gentoo *

How do I change that to something newer? I don't see options for 4.2.x or 4.3.x. How do I choose a newer kernel?

Thanks.
_________________
My rig:
IBM Personal System/2 Model 30-286 - - Intel 80286 (16 bit) 10 Mhz - - 1MB DRAM - - Integrated VGA Display adapter
1.44MB capacity Floppy Disk - - PS/2 keyboard (no mouse)