Opened 9 years ago

Closed 8 years ago

#13506 closed defect (obsolete)

Kernel Panic in supdrv

Reported by: jbr Owned by:
Component: other Version: VirtualBox 4.3.16
Keywords: Cc:
Guest type: Windows Host type: Linux


Hi there,

On a big'ish machine, 16 cores, I'm running VirtualBox. At the time of the kernel panic, which was not the first of its kind (although I don't have pictures from previous kernel panics), I was running roughly 30 Windows XP Virtual Machines. The machine has Debian as host. VBoxManage returns version 4.3.16r95972.

The Windows XP Virtual Machines are constantly (every few minutes) stopped, reverted to a snapshot, and resumed. Hence deadlocks are not unheard of, I guess. Unfortunately I don't have logfiles as the VMs were located in a tmpfs mount.

Please let me know if I can be of any more assistence. If I get any more kernel panics I'll update this issue.

Regards, Jurriaan

Attachments (1)

kernel_panic_supdrv.jpg (426.7 KB ) - added by jbr 9 years ago.

Download all attachments as: .zip

Change History (7)

by jbr, 9 years ago

Attachment: kernel_panic_supdrv.jpg added

comment:1 by Frank Mehnert, 9 years ago

Does this also happen if you add nmi_watchdog=0 to the kernel command line of your host kernel? You are running Debian 7.0 as host, correct?

comment:2 by jbr, 9 years ago

Hi Frank,

Thanks, trying now with nmi_watchdog=0. This is Debian 7.6, btw. Let's give it a few days :-)


comment:3 by jbr, 9 years ago

Hi Frank,

So far so good! Do you mind elaborating a little bit as to why this is causing problems or what the NMI watchdog is actually doing? Or some link with explanation that you know of, perhaps.

There's a little bit here,, but not really how that affects VirtualBox (of course..)

Thanks in advance!

Regards, Jurriaan

comment:4 by Frank Mehnert, 9 years ago

The NMI watchdog checks if the kernel code inhibits hardware interrupts for longer time periods. I don't know the details of your box but running that many virtual machines on it causes a big load with increased runtime inside critical kernel sections. From the screenshot it's visible that the panic is triggered inside the VirtualBox vboxdrv code. The VBoxDrvLinuxIOCtl/supdrvIOCtlFast functions in the stack trace show that this happens when executing guest code. It would be still interesting to see a VBox.log file of one of your VMs.

comment:5 by Frank Mehnert, 9 years ago

Actually I'm not sure if this is a VBox bug at all. See this Ubuntu ticket. Could you try to blacklist the hpwdt module like suggested there and check if this resolves your problem as well?

comment:6 by Frank Mehnert, 8 years ago

Resolution: obsolete
Status: newclosed
Note: See TracTickets for help on using tickets.

© 2023 Oracle
ContactPrivacy policyTerms of Use