Fedora reboots when using libvirt

[Marco André Dinis]

Hi

This is my system:
Fedora 19
libvirtd (libvirt) 1.0.5.7
( I downgraded to this kernel version trying to solve the problem, but no
luck.)

-------------
This is the error message:
> [  536.083922] Disabling lock debugging due to kernel taint
>
> [  536.089320] mce: [Hardware Error]: CPU 3: Machine Check Exception: 4
> > Bank 5: be00000000800400
>
> [  536.097942] mce: [Hardware Error]: TSC 1343d498e08 ADDR 3fbbf66e4a78
> > MISC 1
>
> [  536.105139] mce: [Hardware Error]: PROCESSOR 0:20652 TIME 1386007650
> > SOCKET 0 APIC 5 microcode 9
>
> [  536.114021] mce: [Hardware Error]: Run the above through 'mcelog
> > --ascii'
>
> [  536.120889] mce: [Hardware Error]: CPU 1: Machine Check Exception: 4
> > Bank 5: be00000000800400
>
> [  536.129505] mce: [Hardware Error]: TSC 1343d4b4948 ADDR 3fbbf66e4a78
> > MISC 1
>
> [  536.136700] mce: [Hardware Error]: PROCESSOR 0:20652 TIME 1386007650
> > SOCKET 0 APIC 4 microcode 9
>
> [  536.145582] mce: [Hardware Error]: Run the above through 'mcelog
> > --ascii'
>
> [  536.152444] mce: [Hardware Error]: Machine check: Processor context
> > corrupt
>
> [  536.159483] Kernel panic - not syncing: Fatal Machine check
>
> [  536.165127] drm_kms_helper: panic occurred, switching back to text
> > console
>
> [  536.174297] Rebooting in 30 seconds..
>
> [  566.288306] ACPI MEMORY or I/O RESET_REG.

I was running a virtual machine using qemu-kvm.

What i already tried:
- memtest -> no error.
- formatted and installed only the base system (no DE).
- downgraded kernel: 3.9.5-301.fc19.x86
- swapped the disk with another server (no error with kernel 3.9.5 but i
could only test during some hours, i dont have an extra machine anymore)

Any suggestions?
Thanks, and sorry my not-good english

[Cole Robinson]

On 12/02/2013 01:28 PM, Marco André Dinis wrote:
> Hi
>
> This is my system:
> Fedora 19
> libvirtd (libvirt) 1.0.5.7
> ( I downgraded to this kernel version trying to solve the problem, but no luck.)
>
> This is the error message:
>     [  536.083922] Disabling lock debugging due to kernel taint

This says your kernel is tainted, are you using any proprietary drivers?
>     [  536.089320] mce: [Hardware Error]: CPU 3: Machine Check Exception:
>     4 Bank 5: be00000000800400
>
>     [  536.097942] mce: [Hardware Error]: TSC 1343d498e08 ADDR
>     3fbbf66e4a78 MISC 1 
>
>     [  536.105139] mce: [Hardware Error]: PROCESSOR 0:20652 TIME
>     1386007650 SOCKET 0 APIC 5 microcode 9
>
>     [  536.114021] mce: [Hardware Error]: Run the above through 'mcelog
>     --ascii'
>
>     [  536.120889] mce: [Hardware Error]: CPU 1: Machine Check Exception:
>     4 Bank 5: be00000000800400
>
>     [  536.129505] mce: [Hardware Error]: TSC 1343d4b4948 ADDR
>     3fbbf66e4a78 MISC 1 
>
>     [  536.136700] mce: [Hardware Error]: PROCESSOR 0:20652 TIME
>     1386007650 SOCKET 0 APIC 4 microcode 9
>
>     [  536.145582] mce: [Hardware Error]: Run the above through 'mcelog
>     --ascii'
>
>     [  536.152444] mce: [Hardware Error]: Machine check: Processor context
>     corrupt
>
>     [  536.159483] Kernel panic - not syncing: Fatal Machine check
>
>     [  536.165127] drm_kms_helper: panic occurred, switching back to text
>     console
>
>     [  536.174297] Rebooting in 30 seconds..
>
>     [  566.288306] ACPI MEMORY or I/O RESET_REG.
>
>
> I was running a virtual machine using qemu-kvm.
>
> What i already tried:
> memtest -> no error.
> formatted and installed only the base system (no DE).
> downgraded kernel: 3.9.5-301.fc19.x86
> swapped the disk with another server (no error with kernel 3.9.5 but i could
>
> only test during some hours, i dont have an extra machine anymore)
>
> Any suggestions?
> Thanks, and sorry my not-good english

Make sure you try the latest kernel from rawhide:

sudo yum install fedora-release-rawhide
sudo yum update kernel --enablerepo=rawhide

If you can still reproduce there, please file a bug against the kernel
component in fedora. Please CC me on the bug report, and report the bug number
here after filing so anyone else on the mailing list can follow along there.

Thanks,
Cole

[James Harrison]

Try turning off ACPI on the grub kernel line.
Thanks
James

[Marco André Dinis]

Using: 3.13.0-0.rc2.git0.1.fc21.x86_64
So far, so good.
I'll check again on Thursday. I let it running a virtual machine

On Tue, Dec 3, 2013 at 9:07 AM, James Harrison <[e-mail]
> wrote:
> Try turning off ACPI on the grub kernel line.
> Thanks
> James
>
> From: * Marco André Dinis ;
>
> To: * ;
>
> Subject: * [fedora-virt] Fedora reboots when using libvirt
>
> Sent: * Mon, Dec 2, 2013 6:28:51 PM
>
> Hi
>
> This is my system:
> Fedora 19
> libvirtd (libvirt) 1.0.5.7
> ( I downgraded to this kernel version trying to solve the problem, but no
> luck.)
>
> This is the error message:
> > [  536.083922] Disabling lock debugging due to kernel taint
> >
> > [  536.089320] mce: [Hardware Error]: CPU 3: Machine Check Exception: 4
> > > Bank 5: be00000000800400
> >
> > [  536.097942] mce: [Hardware Error]: TSC 1343d498e08 ADDR 3fbbf66e4a78
> > > MISC 1
> >
> > [  536.105139] mce: [Hardware Error]: PROCESSOR 0:20652 TIME 1386007650
> > > SOCKET 0 APIC 5 microcode 9
> >
> > [  536.114021] mce: [Hardware Error]: Run the above through 'mcelog
> > > --ascii'
> >
> > [  536.120889] mce: [Hardware Error]: CPU 1: Machine Check Exception: 4
> > > Bank 5: be00000000800400
> >
> > [  536.129505] mce: [Hardware Error]: TSC 1343d4b4948 ADDR 3fbbf66e4a78
> > > MISC 1
> >
> > [  536.136700] mce: [Hardware Error]: PROCESSOR 0:20652 TIME 1386007650
> > > SOCKET 0 APIC 4 microcode 9
> >
> > [  536.145582] mce: [Hardware Error]: Run the above through 'mcelog
> > > --ascii'
> >
> > [  536.152444] mce: [Hardware Error]: Machine check: Processor context
> > > corrupt
> >
> > [  536.159483] Kernel panic - not syncing: Fatal Machine check
> >
> > [  536.165127] drm_kms_helper: panic occurred, switching back to text
> > > console
> >
> > [  536.174297] Rebooting in 30 seconds..
> >
> > [  566.288306] ACPI MEMORY or I/O RESET_REG.
>
> I was running a virtual machine using qemu-kvm.
>
> What i already tried:
> memtest -> no error.
> formatted and installed only the base system (no DE).
> downgraded kernel: 3.9.5-301.fc19.x86
> swapped the disk with another server (no error with kernel 3.9.5 but i
>
> could only test during some hours, i dont have an extra machine anymore)
>
> Any suggestions?
> Thanks, and sorry my not-good english

[Marco André Dinis]

Using this kernel: 3.13.0-0.rc2.git0.1.fc21.x86_64
and turning ACPI off solved my problem.
Thanks :)

On Tue, Dec 3, 2013 at 6:52 PM, Marco André Dinis <[e-mail]
> wrote:
> Using: 3.13.0-0.rc2.git0.1.fc21.x86_64
> So far, so good.
> I'll check again on Thursday. I let it running a virtual machine
>
> On Tue, Dec 3, 2013 at 9:07 AM, James Harrison <
> [e-mail]> wrote:
> > Try turning off ACPI on the grub kernel line.
> > Thanks
> > James
> >
> > From: * Marco André Dinis ;
> >
> > To: * ;
> >
> > Subject: * [fedora-virt] Fedora reboots when using libvirt
> >
> > Sent: * Mon, Dec 2, 2013 6:28:51 PM
> >
> > Hi
> >
> > This is my system:
> > Fedora 19
> > libvirtd (libvirt) 1.0.5.7
> > ( I downgraded to this kernel version trying to solve the problem, but no
> > luck.)
> >
> > This is the error message:
> > > [  536.083922] Disabling lock debugging due to kernel taint
> > >
> > > [  536.089320] mce: [Hardware Error]: CPU 3: Machine Check Exception: 4
> > > > Bank 5: be00000000800400
> > >
> > > [  536.097942] mce: [Hardware Error]: TSC 1343d498e08 ADDR 3fbbf66e4a78
> > > > MISC 1
> > >
> > > [  536.105139] mce: [Hardware Error]: PROCESSOR 0:20652 TIME 1386007650
> > > > SOCKET 0 APIC 5 microcode 9
> > >
> > > [  536.114021] mce: [Hardware Error]: Run the above through 'mcelog
> > > > --ascii'
> > >
> > > [  536.120889] mce: [Hardware Error]: CPU 1: Machine Check Exception: 4
> > > > Bank 5: be00000000800400
> > >
> > > [  536.129505] mce: [Hardware Error]: TSC 1343d4b4948 ADDR 3fbbf66e4a78
> > > > MISC 1
> > >
> > > [  536.136700] mce: [Hardware Error]: PROCESSOR 0:20652 TIME 1386007650
> > > > SOCKET 0 APIC 4 microcode 9
> > >
> > > [  536.145582] mce: [Hardware Error]: Run the above through 'mcelog
> > > > --ascii'
> > >
> > > [  536.152444] mce: [Hardware Error]: Machine check: Processor context
> > > > corrupt
> > >
> > > [  536.159483] Kernel panic - not syncing: Fatal Machine check
> > >
> > > [  536.165127] drm_kms_helper: panic occurred, switching back to text
> > > > console
> > >
> > > [  536.174297] Rebooting in 30 seconds..
> > >
> > > [  566.288306] ACPI MEMORY or I/O RESET_REG.
> >
> > I was running a virtual machine using qemu-kvm.
> >
> > What i already tried:
> > memtest -> no error.
> > formatted and installed only the base system (no DE).
> > downgraded kernel: 3.9.5-301.fc19.x86
> > swapped the disk with another server (no error with kernel 3.9.5 but i
> >
> > could only test during some hours, i dont have an extra machine anymore)
> >
> > Any suggestions?
> > Thanks, and sorry my not-good english