kvm-84 kernel panic

DISCUSSION Johannes Baumann 3 participants 11 messages Mar 4, 2009 – Mar 6, 2009 2 messages not in the archive KVM

Share & tools

Johannes Baumann

author

2009-03-04 13:10 UTC #0 lore raw

Hi,

i had serveral kernel panics in the last days on a DELL R300. First i had ubuntu 8.10 running with latest kvm compiled. the panic looked like this: http://memic.paniert.org/screen.png

The impi log looks like this

2d | 02/19/2009 | 02:54:31 | Processor #0x0d | Transition to
Non-recoverable

  2e | 02/19/2009 | 02:54:32 | Processor #0x60 | IERR | Asserted
  32 | 02/19/2009 | 14:55:52 | Processor #0x0d | Transition to

Non-recoverable
33 | 02/24/2009 | 23:00:14 | Processor #0x0d | Transition to
Non-recoverable
37 | 02/25/2009 | 22:58:02 | Processor #0x0d | Transition to
Non-recoverable
38 | 03/03/2009 | 13:59:03 | Processor #0x0d | Transition to
Non-recoverable
39 | 03/03/2009 | 17:41:34 | Processor #0x0d | Transition to
Non-recoverable
3a | 03/04/2009 | 07:41:02 | Processor #0x0d | Transition to
Non-recoverable
3b | 03/04/2009 | 11:03:23 | Processor #0x0d | Transition to
Non-recoverable

  3c | 03/04/2009 | 11:03:26 | Processor #0x60 | IERR | Asserted
  3d | 03/04/2009 | 11:08:16 | Processor #0x60 | IERR | Deasserted

Right now i have fedora 10 as host system running, with the newest kvm version. My CPU is Intel(R) Xeon(R) CPU X3363 @ 2.83GHz Quadcore. I have 2 Windows 2003 Enterprise Domains running at almost no load. Im not sure when to panic happens but it think its reproducable if the guest systems are under load.

What can i do to get more debug information?

Thx so far.. Johannes

Johannes Baumann

author

2009-03-04 21:28 UTC #1 lore raw

update: the kernel panics just seem to happen with smp guests, right now i have only non smp guest and everything is fine, even under load. is this a known problem?

Quoted 44 lines · Johannes Baumann

Hi,
i had serveral kernel panics in the last days on a DELL R300. First i had ubuntu 8.10 running with latest kvm compiled. the panic looked like this: http://memic.paniert.org/screen.png
The impi log looks like this
2d | 02/19/2009 | 02:54:31 | Processor #0x0d | Transition to
Non-recoverable
  2e | 02/19/2009 | 02:54:32 | Processor #0x60 | IERR | Asserted
  32 | 02/19/2009 | 14:55:52 | Processor #0x0d | Transition to
Non-recoverable
33 | 02/24/2009 | 23:00:14 | Processor #0x0d | Transition to
Non-recoverable
37 | 02/25/2009 | 22:58:02 | Processor #0x0d | Transition to
Non-recoverable
38 | 03/03/2009 | 13:59:03 | Processor #0x0d | Transition to
Non-recoverable
39 | 03/03/2009 | 17:41:34 | Processor #0x0d | Transition to
Non-recoverable
3a | 03/04/2009 | 07:41:02 | Processor #0x0d | Transition to
Non-recoverable
3b | 03/04/2009 | 11:03:23 | Processor #0x0d | Transition to
Non-recoverable
  3c | 03/04/2009 | 11:03:26 | Processor #0x60 | IERR | Asserted
  3d | 03/04/2009 | 11:08:16 | Processor #0x60 | IERR | Deasserted
Right now i have fedora 10 as host system running, with the newest kvm version. My CPU is Intel(R) Xeon(R) CPU X3363 @ 2.83GHz Quadcore. I have 2 Windows 2003 Enterprise Domains running at almost no load. Im not sure when to panic happens but it think its reproducable if the guest systems are under load.
What can i do to get more debug information?
Thx so far.. Johannes
--
To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Marcelo Tosatti

2009-03-04 23:19 UTC #2 lore raw

In reply to Johannes Baumann

Quoted 28 lines · Johannes Baumann

Hi,
i had serveral kernel panics in the last days on a DELL R300. First i had ubuntu 8.10 running with latest kvm compiled. the panic looked like this: http://memic.paniert.org/screen.png
The impi log looks like this
2d | 02/19/2009 | 02:54:31 | Processor #0x0d | Transition to
Non-recoverable
  2e | 02/19/2009 | 02:54:32 | Processor #0x60 | IERR | Asserted
  32 | 02/19/2009 | 14:55:52 | Processor #0x0d | Transition to
Non-recoverable
33 | 02/24/2009 | 23:00:14 | Processor #0x0d | Transition to
Non-recoverable
37 | 02/25/2009 | 22:58:02 | Processor #0x0d | Transition to
Non-recoverable
38 | 03/03/2009 | 13:59:03 | Processor #0x0d | Transition to
Non-recoverable
39 | 03/03/2009 | 17:41:34 | Processor #0x0d | Transition to
Non-recoverable
3a | 03/04/2009 | 07:41:02 | Processor #0x0d | Transition to
Non-recoverable
3b | 03/04/2009 | 11:03:23 | Processor #0x0d | Transition to
Non-recoverable
  3c | 03/04/2009 | 11:03:26 | Processor #0x60 | IERR | Asserted
  3d | 03/04/2009 | 11:08:16 | Processor #0x60 | IERR | Deasserted

Quoted 7 lines

Right now i have fedora 10 as host system running, with the newest kvm version. My CPU is Intel(R) Xeon(R) CPU X3363 @ 2.83GHz Quadcore. I have 2 Windows 2003 Enterprise Domains running at almost no load. Im not sure when to panic happens but it think its reproducable if the guest systems are under load.
What can i do to get more debug information?

http://en.wikipedia.org/wiki/Machine_Check_Exception

See the "Decoding MCEs" section.

Johannes Baumann

author

2009-03-05 21:51 UTC #3 lore raw

update: its not only happening with smp guest, but with non-smp guest. mcelog is empty. the commandline for a debian lenny im using is:

/usr/src/kvm-84/qemu/x86_64-softmmu/qemu-system-x86_64 -S -M pc -m 512 -smp 1 -name vm03 -uuid 0b5e3a60-9111-0289-84b5-98cdf27450ac -monitor pty -boot c -drive file=/kvm/domains/linux-test/image.img,if=ide,index=0,boot=on -drive file=,if=ide,media=cdrom,index=2 -net nic,macaddr=54:52:00:0b:a1:26,vlan=0 -net tap,fd=12,script=,vlan=0,ifname=vnet0 -serial none -parallel none -usb -vnc 127.0.0.1:0

(i know the path looks weird but its ok, i didnt want to do a make install, the right modules are also loaded)

if i add -no-kvm the host just seem to run fine..

johannes

Quoted 46 lines · Marcelo Tosatti

Quoted 28 lines · Johannes Baumann
Hi,
i had serveral kernel panics in the last days on a DELL R300. First i had ubuntu 8.10 running with latest kvm compiled. the panic looked like this: http://memic.paniert.org/screen.png
The impi log looks like this
2d | 02/19/2009 | 02:54:31 | Processor #0x0d | Transition to
Non-recoverable
  2e | 02/19/2009 | 02:54:32 | Processor #0x60 | IERR | Asserted
  32 | 02/19/2009 | 14:55:52 | Processor #0x0d | Transition to
Non-recoverable
33 | 02/24/2009 | 23:00:14 | Processor #0x0d | Transition to
Non-recoverable
37 | 02/25/2009 | 22:58:02 | Processor #0x0d | Transition to
Non-recoverable
38 | 03/03/2009 | 13:59:03 | Processor #0x0d | Transition to
Non-recoverable
39 | 03/03/2009 | 17:41:34 | Processor #0x0d | Transition to
Non-recoverable
3a | 03/04/2009 | 07:41:02 | Processor #0x0d | Transition to
Non-recoverable
3b | 03/04/2009 | 11:03:23 | Processor #0x0d | Transition to
Non-recoverable
  3c | 03/04/2009 | 11:03:26 | Processor #0x60 | IERR | Asserted
  3d | 03/04/2009 | 11:08:16 | Processor #0x60 | IERR | Deasserted
Quoted 7 lines
Right now i have fedora 10 as host system running, with the newest kvm version. My CPU is Intel(R) Xeon(R) CPU X3363 @ 2.83GHz Quadcore. I have 2 Windows 2003 Enterprise Domains running at almost no load. Im not sure when to panic happens but it think its reproducable if the guest systems are under load.
What can i do to get more debug information?
http://en.wikipedia.org/wiki/Machine_Check_Exception
See the "Decoding MCEs" section.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Ryan Harper

2009-03-05 21:58 UTC #4 lore raw

Quoted 11 lines · Johannes Baumann

update: its not only happening with smp guest, but with non-smp guest. mcelog is empty. the commandline for a debian lenny im using is:
/usr/src/kvm-84/qemu/x86_64-softmmu/qemu-system-x86_64 -S -M pc -m 512 -smp 1 -name vm03 -uuid 0b5e3a60-9111-0289-84b5-98cdf27450ac -monitor pty -boot c -drive file=/kvm/domains/linux-test/image.img,if=ide,index=0,boot=on -drive file=,if=ide,media=cdrom,index=2 -net nic,macaddr=54:52:00:0b:a1:26,vlan=0 -net tap,fd=12,script=,vlan=0,ifname=vnet0 -serial none -parallel none -usb -vnc 127.0.0.1:0

you'll want -L /usr/src/kvm-84/qemu/pc-bios to ensure you use the bios that came with kvm-84 instead of some other location.

Quoted 63 lines

(i know the path looks weird but its ok, i didnt want to do a make install, the right modules are also loaded)
if i add -no-kvm the host just seem to run fine..
johannes
Quoted 37 lines · Marcelo Tosatti
Quoted 28 lines · Johannes Baumann
Hi,
i had serveral kernel panics in the last days on a DELL R300. First i had ubuntu 8.10 running with latest kvm compiled. the panic looked like this: http://memic.paniert.org/screen.png
The impi log looks like this
2d | 02/19/2009 | 02:54:31 | Processor #0x0d | Transition to
Non-recoverable
  2e | 02/19/2009 | 02:54:32 | Processor #0x60 | IERR | Asserted
  32 | 02/19/2009 | 14:55:52 | Processor #0x0d | Transition to
Non-recoverable
33 | 02/24/2009 | 23:00:14 | Processor #0x0d | Transition to
Non-recoverable
37 | 02/25/2009 | 22:58:02 | Processor #0x0d | Transition to
Non-recoverable
38 | 03/03/2009 | 13:59:03 | Processor #0x0d | Transition to
Non-recoverable
39 | 03/03/2009 | 17:41:34 | Processor #0x0d | Transition to
Non-recoverable
3a | 03/04/2009 | 07:41:02 | Processor #0x0d | Transition to
Non-recoverable
3b | 03/04/2009 | 11:03:23 | Processor #0x0d | Transition to
Non-recoverable
  3c | 03/04/2009 | 11:03:26 | Processor #0x60 | IERR | Asserted
  3d | 03/04/2009 | 11:08:16 | Processor #0x60 | IERR | Deasserted
Quoted 7 lines
Right now i have fedora 10 as host system running, with the newest kvm version. My CPU is Intel(R) Xeon(R) CPU X3363 @ 2.83GHz Quadcore. I have 2 Windows 2003 Enterprise Domains running at almost no load. Im not sure when to panic happens but it think its reproducable if the guest systems are under load.
What can i do to get more debug information?
Quoted 9 lines
http://en.wikipedia.org/wiki/Machine_Check_Exception
See the "Decoding MCEs" section.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Signature · 4 lines

Ryan Harper
Software Engineer; Linux Technology Center
IBM Corp., Austin, Tx
[email protected]

Johannes Baumann

author

2009-03-05 22:07 UTC #5 lore raw

Quoted 4 lines · Ryan Harper

Quoted 3 lines · Johannes Baumann
update: its not only happening with smp guest, but with non-smp guest. mcelog is empty. the commandline for a debian lenny im using is:

Quoted 11 lines

Quoted 8 lines
/usr/src/kvm-84/qemu/x86_64-softmmu/qemu-system-x86_64 -S -M pc -m 512 -smp 1 -name vm03 -uuid 0b5e3a60-9111-0289-84b5-98cdf27450ac -monitor pty -boot c -drive file=/kvm/domains/linux-test/image.img,if=ide,index=0,boot=on -drive file=,if=ide,media=cdrom,index=2 -net nic,macaddr=54:52:00:0b:a1:26,vlan=0 -net tap,fd=12,script=,vlan=0,ifname=vnet0 -serial none -parallel none -usb -vnc 127.0.0.1:0
you'll want -L /usr/src/kvm-84/qemu/pc-bios to ensure you use the bios that came with kvm-84 instead of some other location.

i could try but i dont think this is the point since every kvm version had the same problems so far.

Quoted 62 lines

Quoted 61 lines
(i know the path looks weird but its ok, i didnt want to do a make install, the right modules are also loaded)
if i add -no-kvm the host just seem to run fine..
johannes
Quoted 36 lines · Marcelo Tosatti
Quoted 28 lines · Johannes Baumann
Hi,
i had serveral kernel panics in the last days on a DELL R300. First i had ubuntu 8.10 running with latest kvm compiled. the panic looked like this: http://memic.paniert.org/screen.png
The impi log looks like this
2d | 02/19/2009 | 02:54:31 | Processor #0x0d | Transition to
Non-recoverable
  2e | 02/19/2009 | 02:54:32 | Processor #0x60 | IERR | Asserted
  32 | 02/19/2009 | 14:55:52 | Processor #0x0d | Transition to
Non-recoverable
33 | 02/24/2009 | 23:00:14 | Processor #0x0d | Transition to
Non-recoverable
37 | 02/25/2009 | 22:58:02 | Processor #0x0d | Transition to
Non-recoverable
38 | 03/03/2009 | 13:59:03 | Processor #0x0d | Transition to
Non-recoverable
39 | 03/03/2009 | 17:41:34 | Processor #0x0d | Transition to
Non-recoverable
3a | 03/04/2009 | 07:41:02 | Processor #0x0d | Transition to
Non-recoverable
3b | 03/04/2009 | 11:03:23 | Processor #0x0d | Transition to
Non-recoverable
  3c | 03/04/2009 | 11:03:26 | Processor #0x60 | IERR | Asserted
  3d | 03/04/2009 | 11:08:16 | Processor #0x60 | IERR | Deasserted
Quoted 7 lines
Right now i have fedora 10 as host system running, with the newest kvm version. My CPU is Intel(R) Xeon(R) CPU X3363 @ 2.83GHz Quadcore. I have 2 Windows 2003 Enterprise Domains running at almost no load. Im not sure when to panic happens but it think its reproducable if the guest systems are under load.
What can i do to get more debug information?
Quoted 9 lines
http://en.wikipedia.org/wiki/Machine_Check_Exception
See the "Decoding MCEs" section.
--
To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Ryan Harper

2009-03-05 22:09 UTC #6 lore raw

Quoted 17 lines · Johannes Baumann

Quoted 4 lines · Ryan Harper
Quoted 3 lines · Johannes Baumann
update: its not only happening with smp guest, but with non-smp guest. mcelog is empty. the commandline for a debian lenny im using is:
Quoted 11 lines
Quoted 8 lines
/usr/src/kvm-84/qemu/x86_64-softmmu/qemu-system-x86_64 -S -M pc -m 512 -smp 1 -name vm03 -uuid 0b5e3a60-9111-0289-84b5-98cdf27450ac -monitor pty -boot c -drive file=/kvm/domains/linux-test/image.img,if=ide,index=0,boot=on -drive file=,if=ide,media=cdrom,index=2 -net nic,macaddr=54:52:00:0b:a1:26,vlan=0 -net tap,fd=12,script=,vlan=0,ifname=vnet0 -serial none -parallel none -usb -vnc 127.0.0.1:0
you'll want -L /usr/src/kvm-84/qemu/pc-bios to ensure you use the bios that came with kvm-84 instead of some other location.

Quoted 2 lines

i could try but i dont think this is the point since every kvm version had the same problems so far.

if you don't use the right bios that comes with the kernel modules, things _will_ break.

Signature · 4 lines

Ryan Harper
Software Engineer; Linux Technology Center
IBM Corp., Austin, Tx
[email protected]

Johannes Baumann

author

2009-03-05 22:31 UTC #7 lore raw

Quoted 16 lines · Ryan Harper

Quoted 15 lines · Johannes Baumann
Quoted 14 lines · Ryan Harper
Quoted 11 lines · Johannes Baumann
update: its not only happening with smp guest, but with non-smp guest. mcelog is empty. the commandline for a debian lenny im using is:
/usr/src/kvm-84/qemu/x86_64-softmmu/qemu-system-x86_64 -S -M pc -m 512 -smp 1 -name vm03 -uuid 0b5e3a60-9111-0289-84b5-98cdf27450ac -monitor pty -boot c -drive file=/kvm/domains/linux-test/image.img,if=ide,index=0,boot=on -drive file=,if=ide,media=cdrom,index=2 -net nic,macaddr=54:52:00:0b:a1:26,vlan=0 -net tap,fd=12,script=,vlan=0,ifname=vnet0 -serial none -parallel none -usb -vnc 127.0.0.1:0
you'll want -L /usr/src/kvm-84/qemu/pc-bios to ensure you use the bios that came with kvm-84 instead of some other location.

Quoted 5 lines

Quoted 2 lines
i could try but i dont think this is the point since every kvm version had the same problems so far.
if you don't use the right bios that comes with the kernel modules, things _will_ break.

k, trying the -L switch, with vt enabled

root 6265 23.8 25.6 1100904 1039068 ? Sl 23:24 1:29
/usr/src/kvm-84/qemu/x86_64-softmmu/qemu-system-x86_64 -L /usr/src/kvm-84/qemu/pc-bios -S -M pc -m 1000 -smp 1 -name vm01 -uuid 3490fdf2-9550-4300-883a-a10fbf9b58e6 -monitor pty -localtime -boot c -drive file=/kvm/domains/windows2003-enterprise-jb/image.img,if=ide,index=0,boot=on -drive file=,if=ide,media=cdrom,index=2 -net nic,macaddr=00:16:d3:c2:36:ff,vlan=0,model=e1000 -net tap,fd=18,script=,vlan=0,ifname=vnet1 -serial none -parallel none -usb -usbdevice tablet -vnc 127.0.0.1:1

Quoted 2 lines

johannes
-

Marcelo Tosatti

2009-03-06 00:03 UTC #8 lore raw

Quoted 37 lines · Johannes Baumann

Quoted 16 lines · Ryan Harper
Quoted 15 lines · Johannes Baumann
Quoted 14 lines · Ryan Harper
Quoted 11 lines · Johannes Baumann
update: its not only happening with smp guest, but with non-smp guest. mcelog is empty. the commandline for a debian lenny im using is:
/usr/src/kvm-84/qemu/x86_64-softmmu/qemu-system-x86_64 -S -M pc -m 512 -smp 1 -name vm03 -uuid 0b5e3a60-9111-0289-84b5-98cdf27450ac -monitor pty -boot c -drive file=/kvm/domains/linux-test/image.img,if=ide,index=0,boot=on -drive file=,if=ide,media=cdrom,index=2 -net nic,macaddr=54:52:00:0b:a1:26,vlan=0 -net tap,fd=12,script=,vlan=0,ifname=vnet0 -serial none -parallel none -usb -vnc 127.0.0.1:0
you'll want -L /usr/src/kvm-84/qemu/pc-bios to ensure you use the bios that came with kvm-84 instead of some other location.
Quoted 5 lines
Quoted 2 lines
i could try but i dont think this is the point since every kvm version had the same problems so far.
if you don't use the right bios that comes with the kernel modules, things _will_ break.
k, trying the -L switch, with vt enabled
root 6265 23.8 25.6 1100904 1039068 ? Sl 23:24 1:29
/usr/src/kvm-84/qemu/x86_64-softmmu/qemu-system-x86_64 -L /usr/src/kvm-84/qemu/pc-bios -S -M pc -m 1000 -smp 1 -name vm01 -uuid 3490fdf2-9550-4300-883a-a10fbf9b58e6 -monitor pty -localtime -boot c -drive file=/kvm/domains/windows2003-enterprise-jb/image.img,if=ide,index=0,boot=on -drive file=,if=ide,media=cdrom,index=2 -net nic,macaddr=00:16:d3:c2:36:ff,vlan=0,model=e1000 -net tap,fd=18,script=,vlan=0,ifname=vnet1 -serial none -parallel none -usb -usbdevice tablet -vnc 127.0.0.1:1

Quoted 3 lines

2d | 02/19/2009 | 02:54:31 | Processor #0x0d | Transition to
Non-recoverable
2e | 02/19/2009 | 02:54:32 | Processor #0x60 | IERR | Asserted

http://bugs.centos.org/print_bug_page.php?bug_id=2619

Likely to be a HW problem.

Johannes Baumann

author

2009-03-06 05:02 UTC #9 lore raw

Quoted 24 lines · Marcelo Tosatti

Quoted 23 lines · Johannes Baumann
Quoted 16 lines · Ryan Harper
Quoted 15 lines · Johannes Baumann
Quoted 14 lines · Ryan Harper
Quoted 11 lines · Johannes Baumann
update: its not only happening with smp guest, but with non-smp guest. mcelog is empty. the commandline for a debian lenny im using is:
/usr/src/kvm-84/qemu/x86_64-softmmu/qemu-system-x86_64 -S -M pc -m 512 -smp 1 -name vm03 -uuid 0b5e3a60-9111-0289-84b5-98cdf27450ac -monitor pty -boot c -drive file=/kvm/domains/linux-test/image.img,if=ide,index=0,boot=on -drive file=,if=ide,media=cdrom,index=2 -net nic,macaddr=54:52:00:0b:a1:26,vlan=0 -net tap,fd=12,script=,vlan=0,ifname=vnet0 -serial none -parallel none -usb -vnc 127.0.0.1:0
you'll want -L /usr/src/kvm-84/qemu/pc-bios to ensure you use the bios that came with kvm-84 instead of some other location.
Quoted 4 lines
Quoted 2 lines
i could try but i dont think this is the point since every kvm version had the same problems so far.
if you don't use the right bios that comes with the kernel modules, things _will_ break.
k, trying the -L switch, with vt enabled

Quoted 10 lines

Quoted 10 lines
root 6265 23.8 25.6 1100904 1039068 ? Sl 23:24 1:29
/usr/src/kvm-84/qemu/x86_64-softmmu/qemu-system-x86_64 -L /usr/src/kvm-84/qemu/pc-bios -S -M pc -m 1000 -smp 1 -name vm01 -uuid 3490fdf2-9550-4300-883a-a10fbf9b58e6 -monitor pty -localtime -boot c -drive file=/kvm/domains/windows2003-enterprise-jb/image.img,if=ide,index=0,boot=on -drive file=,if=ide,media=cdrom,index=2 -net nic,macaddr=00:16:d3:c2:36:ff,vlan=0,model=e1000 -net tap,fd=18,script=,vlan=0,ifname=vnet1 -serial none -parallel none -usb -usbdevice tablet -vnc 127.0.0.1:1

seems realy be guest bios related, since using the -L switch the host System didnt crash anymore. I got a lot of mcelogs:

MCE 0
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 2 BANK 0 TSC 34186b6364dd
MCG status:
MCi status:
Error overflow
Uncorrected error
Error enabled
Processor context corrupt
MCA:BUS Level-0 Originated-request Generic Memory-access Request-timeout Error
Model:
STATUS f200080010000800 MCGSTATUS 0
MCE 1
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 2 BANK 5 TSC 34186b6373bd
MCG status:
MCi status:
Uncorrected error
Error enabled
Processor context corrupt
MCA:BUS Generic Generic Generic Other-transaction Request-timeout Error Model:
STATUS b200000806000e0f MCGSTATUS 0
MCE 2
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 0 BANK 5 TSC 34186b637434
MCG status:RIPV MCIP
MCi status:
Uncorrected error
Error enabled
Processor context corrupt
MCA:BUS Generic Generic Generic Other-transaction Request-timeout Error Model:Pad state machine

and so on..

Quoted 9 lines

Quoted 5 lines
2d | 02/19/2009 | 02:54:31 | Processor #0x0d | Transition to
Non-recoverable
2e | 02/19/2009 | 02:54:32 | Processor #0x60 | IERR | Asserted
http://bugs.centos.org/print_bug_page.php?bug_id=2619
Likely to be a HW problem.

I wasn able to reproduce without kvm running, still a hw problem?

Quoted 5 lines

--
To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

Johannes

Johannes Baumann

author

2009-03-06 17:54 UTC #12 lore raw

In reply to a message not in the archive

i installed esxi, same error in ipmi after a while, talking to dell..

Quoted 90 lines · Marcelo Tosatti

Quoted 79 lines · Johannes Baumann
Quoted 24 lines · Marcelo Tosatti
Quoted 23 lines · Johannes Baumann
Quoted 16 lines · Ryan Harper
Quoted 15 lines · Johannes Baumann
Quoted 14 lines · Ryan Harper
Quoted 11 lines · Johannes Baumann
update: its not only happening with smp guest, but with non-smp guest. mcelog is empty. the commandline for a debian lenny im using is:
/usr/src/kvm-84/qemu/x86_64-softmmu/qemu-system-x86_64 -S -M pc -m 512 -smp 1 -name vm03 -uuid 0b5e3a60-9111-0289-84b5-98cdf27450ac -monitor pty -boot c -drive file=/kvm/domains/linux-test/image.img,if=ide,index=0,boot=on -drive file=,if=ide,media=cdrom,index=2 -net nic,macaddr=54:52:00:0b:a1:26,vlan=0 -net tap,fd=12,script=,vlan=0,ifname=vnet0 -serial none -parallel none -usb -vnc 127.0.0.1:0
you'll want -L /usr/src/kvm-84/qemu/pc-bios to ensure you use the bios that came with kvm-84 instead of some other location.
Quoted 4 lines
Quoted 2 lines
i could try but i dont think this is the point since every kvm version had the same problems so far.
if you don't use the right bios that comes with the kernel modules, things _will_ break.
k, trying the -L switch, with vt enabled
Quoted 10 lines
Quoted 10 lines
root 6265 23.8 25.6 1100904 1039068 ? Sl 23:24 1:29
/usr/src/kvm-84/qemu/x86_64-softmmu/qemu-system-x86_64 -L /usr/src/kvm-84/qemu/pc-bios -S -M pc -m 1000 -smp 1 -name vm01 -uuid 3490fdf2-9550-4300-883a-a10fbf9b58e6 -monitor pty -localtime -boot c -drive file=/kvm/domains/windows2003-enterprise-jb/image.img,if=ide,index=0,boot=on -drive file=,if=ide,media=cdrom,index=2 -net nic,macaddr=00:16:d3:c2:36:ff,vlan=0,model=e1000 -net tap,fd=18,script=,vlan=0,ifname=vnet1 -serial none -parallel none -usb -usbdevice tablet -vnc 127.0.0.1:1
seems realy be guest bios related, since using the -L switch the host System didnt crash anymore. I got a lot of mcelogs:
MCE 0
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 2 BANK 0 TSC 34186b6364dd
MCG status:
MCi status:
Error overflow
Uncorrected error
Error enabled
Processor context corrupt
MCA:BUS Level-0 Originated-request Generic Memory-access Request-timeout Error
Model:
STATUS f200080010000800 MCGSTATUS 0
MCE 1
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 2 BANK 5 TSC 34186b6373bd
MCG status:
MCi status:
Uncorrected error
Error enabled
Processor context corrupt
MCA:BUS Generic Generic Generic Other-transaction Request-timeout Error Model:
STATUS b200000806000e0f MCGSTATUS 0
MCE 2
HARDWARE ERROR. This is *NOT* a software problem!
Please contact your hardware vendor
CPU 0 BANK 5 TSC 34186b637434
MCG status:RIPV MCIP
MCi status:
Uncorrected error
Error enabled
Processor context corrupt
MCA:BUS Generic Generic Generic Other-transaction Request-timeout Error Model:Pad state machine
and so on..
Quoted 1 line
Quoted 7 lines
Quoted 6 lines
Quoted 3 lines
2d | 02/19/2009 | 02:54:31 | Processor #0x0d | Transition to
Non-recoverable
2e | 02/19/2009 | 02:54:32 | Processor #0x60 | IERR | Asserted
http://bugs.centos.org/print_bug_page.php?bug_id=2619
Likely to be a HW problem.
I wasn able to reproduce without kvm running, still a hw problem?
MCE usually indicate hardware problems. Take this to Dell support.