Multiple Guests Randomly Rebooting - Windows Bug Check Error

[hamburgertime]

Hey all,

Having a bit of an issue with multiple Windows VMs running in vbox. They were all previously working.
Am running Fedora 32, Vbox 6.1. I believe there was an update for VirtualBox that came with some other os updates I got a week or two ago then this issue started.

Randomly, windows server 2016, 2019 and windows 10vms reboot for no reason now, usually within 10 min of starting up, sometimes even during boot, nothing triggers it specifically, just on its own. I've even tried re-installing fresh just to make sure. The VMs work fine on a centos host. Tried to re-install VirtualBox, update dependencies, re-install guest additions, changed graphics driver, network adapter as well as I thought this could be do to windows driver issues. Linux VMs run fine.

I found a couple other mentions of this in back in Jan/Feb 2020 but no luck with a resolution.

Event logs in windows say Windows Bug Check Error - error says related to drivers, can't see what is causing it though.

Vbox logs dont say anything.

Any ideas? Anyone else experience this before?

Thanks!

[scottgus1]

Am running Fedora 32, Vbox 6.1. ... an update for VirtualBox that came with some other os updates

Virtualbox does not feed updates with the host OS's system updates.

Unless...

you are running the Fedora repo version of Virtualbox, then Fedora might update its fork of Virtualbox. If you are running Fedora's Virtualbox rather than official Virtualbox from https://www.virtualbox.org/wiki/Linux_Downloads, then we are probably not going to be able to help. Forks have to be supported by the forkers, since they know what they have changed and we don't.

Randomly, windows server 2016, 2019 and windows 10vms reboot for no reason ... Event logs in windows say Windows Bug Check Error

Generally, if the Virtualbox (official, that is) window stays open and the guest BSODs, then the problem should be troubleshot in the guest OS. Of course, if you did not have these troubles before the Fedora update, then the thing to check is the Fedora update.

Or install official Virtualbox, then we could try to see what is happening if the problems continue.

[hamburgertime]

Hey there, thanks for the reply, I am running the VirtualBox version, not a fedora repo of VirtualBox, I should have clarified there were system updates to Fedora, I have tried running an older Kernel as well, doesn't solve the issue either. It definitely has to be on the Fedora side though I can't narrow down what else to check.

[fth0]

Virtualbox does not feed updates with the host OS's system updates.

Unless...

you are running the Fedora repo version of Virtualbox

Your assumption is not necessarily correct:

There are two methods to install the official VirtualBox versions on Linux hosts. One method begins with downloading a VirtualBox installer package for the Linux distribution in use, the other method begins with integrating the official VirtualBox package source into the Linux package management software in use. If you use the second (preferred) method, any later VirtualBox update looks like a Linux update, if you don't look closely enough.

[fth0]

Vbox logs dont say anything.

Maybe we could also have a look?

Please post a zipped VBox.log file from a complete VM run where the problem occurred, and tell us when the Bug Check occurred, if possible.

[scottgus1]

integrating the official VirtualBox package source into the Linux package management software

So if the 'repo path' for Virtualbox is pointed at official Virtualbox, then when the devs put up a new version, the Linux OS will tell the user about it? (or something like that?)

I learn something new every day!

[hamburgertime]

Vbox logs dont say anything.

Maybe we could also have a look?

Please post a zipped VBox.log file from a complete VM run where the problem occurred, and tell us when the Bug Check occurred, if possible.

Hey there - sorry there was a couple things in the logs but weren't overly specific.
So between 00:00:18.4***** and 00:02:10.4**** on a brand new installed vm I watched live with tail via command line and confirmed in event log once rebooted this was the bug check error. I've attached the whole log file via zip.

But also a quick snippit at line 1258 in the zipped log file

00:00:18.405136 NAT: DNS#0: 192.168.0.3
00:00:18.405140 NAT: DNS#1: 192.168.0.4
00:00:18.405144 NAT: DHCP offered IP address 10.0.2.15
------ reset here - bugcheck error ------
00:02:10.426398 AHCI#0: Reset the HBA
00:02:10.426454 VD#0: Cancelling all active requests
00:02:10.426821 AHCI#0: Port 0 reset
00:02:10.427898 VD#0: Cancelling all active requests
00:02:20.799182 ACPI: Reset initiated by ACPI
00:02:20.799274 Changing the VM state from 'RUNNING' to 'RESETTING'
00:02:20.799648 PIT: mode=3 count=0x10000 (65536) - 18.20 Hz (ch=0)
00:02:20.808068 AHCI#0: Reset the HBA
00:02:20.808088 VD#0: Cancelling all active requests
00:02:20.808148 HDA: Codec reset
00:02:20.808155 HDA: Reset
00:02:20.808505 VMMDevNotifyGuest: fAddEvents=0x1 ignored because enmVMState=8
00:02:20.874895 Changing the VM state from 'RESETTING' to 'RUNNING'
00:02:20.875803 OHCI: root hub reset completed with VINF_SUCCESS

Any help is appreciated, thanks guys!

[fth0]

So if the 'repo path' for Virtualbox is pointed at official Virtualbox, then when the devs put up a new version, the Linux OS will tell the user about it? (or something like that?)

Yes. As an example, see the following screenshot of the Linux Mint 19.3 Update Manager from one of my hosts (coincidentally, there is a VirtualBox update waiting ):

Linux Mint 19.3 Update Manager Screenshot (German)

Screenshot.png (91.7 KiB) Viewed 6422 times

The column labelled Herkunft (German for Origin) shows 5 different repositories as the respective origin of the offered updates. Linux Mint is based on Ubuntu, so the Linux Mint and Ubuntu repositories are in the majority. A PPA (Personal Package Archive) is a special kind of repository if you need newer versions of a software than your Linux distribution provides. And Oracle Corporation speaks for itself.

[fth0]

I don't see anything special in the log file either, but I also don't see other expected information. Let me ask you a few questions:

00:02:23.396552 VMMDev: Guest Log: BIOS: Booting from CD-ROM...

Are you deliberately booting from the ISO image?

Did you install the VirtualBox Guest Additions in your guests?

Does your host PC have an NVIDIA graphics card? If yes, what's the driver name and version?

Did you deliberately configure System > Acceleration > Paravirtualization Interface to None? If you configure it to Default, does the problem persist?

What information do you have about the Bug Check Error?

The following additional question may sound strange:
Are you working on your host PC remotely (like in a home office situation common these days)?

[hamburgertime]

I don't see anything special in the log file either, but I also don't see other expected information. Let me ask you a few questions:

00:02:23.396552 VMMDev: Guest Log: BIOS: Booting from CD-ROM...

Are you deliberately booting from the ISO image?

Did you install the VirtualBox Guest Additions in your guests?

Does your host PC have an NVIDIA graphics card? If yes, what's the driver name and version?

Did you deliberately configure System > Acceleration > Paravirtualization Interface to None? If you configure it to Default, does the problem persist?

What information do you have about the Bug Check Error?

The following additional question may sound strange:
Are you working on your host PC remotely (like in a home office situation common these days)?

I had just left the iso in following a re-install just wanted to show a fresh log, it's the same thing on vms that have been setup for awhile. I had since removed the iso, same thing.
I have tried with and without Guest additions
Not sure why acceleration is showing as none, it is set to default and I have not changed it.
Pc is using intel iris graphics
Bug check error inside windows event logs says it could be related to driver, but super unspecific when i search the error code, I can assure you it is not on the guest side, the vms work fine on other hosts and previously worked fine on this one too.
The vms are local to the host yes , not remoting
I've also tried loading up older linux kernels but no luck there either

[fth0]

What information do you have about the Bug Check Error?

Bug check error inside windows event logs says it could be related to driver, but super unspecific when i search the error code, I can assure you it is not on the guest side, the vms work fine on other hosts and previously worked fine on this one too.

We seem to have a little communication problem here:

I've been asking about information regarding the Bug Check error, and you tell me that you searched the error code, but provide neither the error code nor the available additional information to me. How do you expect me (or other people) to help you in this case? Or did you take my question too literally?

Please provide all information you have about the Bug Check error, which may provide a starting point for what specifically to concentrate on when looking at a VBox.log file. And provide a zipped VBox.log file of a VM run that doesn't start from the ISO file, but includes the VirtualBox Guest Additions, which may also provide important information in the VBox.log file.

Not sure why acceleration is showing as none, it is set to default and I have not changed it.

Thank you for this information. I double checked and noticed that indeed the Windows 2019 and Windows 2016 presets behave differently than the Windows 2012 and the Windows desktop presets. Please try if setting System > Acceleration > Paravirtualization Interface to Hyper-V (the expected outcome of Default) makes a difference.

Are you working on your host PC remotely (like in a home office situation common these days)?

The vms are local to the host yes , not remoting

There's maybe a misunderstanding here: I meant to ask if you and the host PC (not the guests) are at different locations ...

[hamburgertime]

What information do you have about the Bug Check Error?

Bug check error inside windows event logs says it could be related to driver, but super unspecific when i search the error code, I can assure you it is not on the guest side, the vms work fine on other hosts and previously worked fine on this one too.

We seem to have a little communication problem here:

I've been asking about information regarding the Bug Check error, and you tell me that you searched the error code, but provide neither the error code nor the available additional information to me. How do you expect me (or other people) to help you in this case? Or did you take my question too literally?

Please provide all information you have about the Bug Check error, which may provide a starting point for what specifically to concentrate on when looking at a VBox.log file. And provide a zipped VBox.log file of a VM run that doesn't start from the ISO file, but includes the VirtualBox Guest Additions, which may also provide important information in the VBox.log file.

Not sure why acceleration is showing as none, it is set to default and I have not changed it.

Thank you for this information. I double checked and noticed that indeed the Windows 2019 and Windows 2016 presets behave differently than the Windows 2012 and the Windows desktop presets. Please try if setting System > Acceleration > Paravirtualization Interface to Hyper-V (the expected outcome of Default) makes a difference.

Are you working on your host PC remotely (like in a home office situation common these days)?

The vms are local to the host yes , not remoting

There's maybe a misunderstanding here: I meant to ask if you and the host PC (not the guests) are at different locations ...

As you have requested, this is from one of my three existing VMS. The VM I posted a log from originally is not booting from an ISO it simply had an ISO image attached as I had recently re-installed a VM for testing purposes to show this issue was happening with a brand new installed VM. LIke I said, I can use these VMS on other hosts without issue.

These Guests are installed on my laptop, the laptop is the host, there is no remote sessions for clarification. I use these VMs for management purposes.

Anyways from the Event Log in Windows, from an existing VM with VirtualBox Guest Additions installed, trying to run with Hyper-V selected instead of Default (I tried this previously also with no luck )

The computer has rebooted from a bugcheck. The bugcheck was: 0x0000003d (0xffffb50d86e29388, 0xffffb50d86e28bd0, 0x0000000000000000, 0x0000000000000000). A dump was saved in: C:\Windows\MEMORY.DMP. Report Id: 53fd9fff-5748-421a-abc6-d563a2256f48.

From the Dump file within the Window VM there are two filenames highlighted - ntoskrnl.exe and tcpip.sys that are perhaps related to the BugCheck.

Please see the attached log as requested, hopefully this all clarifies everything Sorry for any confusion there!
00:02:04.639238 is where the crash begins

Any help would be fantastic thanks all!

[fth0]

00:02:04.639238 GIM: HyperV: Guest indicates a fatal condition! P0=0x1e P1=0xffffffff80000002 P2=0xfffff806114d9e98 P3=0x0 P4=0x1
00:02:04.639387 GIMHv: BugCheck 1e {ffffffff80000002, fffff806114d9e98, 0, 1}
00:02:04.639388 KMODE_EXCEPTION_NOT_HANDLED
00:02:04.639388 P1: ffffffff80000002 - exception code - STATUS_DATATYPE_MISALIGNMENT
00:02:04.639388 P2: fffff806114d9e98 - EIP/RIP
00:02:04.639388 P3: 0000000000000000 - Xcpt param #0
00:02:04.639389 P4: 0000000000000001 - Xcpt param #1

This Bug Check and the one you've posted both indicate problems with a kernel level driver inside the Windows guest OS. Further details could be investigated using a debugger in the Windows guest, but that is something I cannot really help you with.

You already suspected an update on the Linux guest. My primary suspect would be the graphics driver in this case. You could also investigate which Linux packages got updated for further candidates.

[hamburgertime]

00:02:04.639238 GIM: HyperV: Guest indicates a fatal condition! P0=0x1e P1=0xffffffff80000002 P2=0xfffff806114d9e98 P3=0x0 P4=0x1
00:02:04.639387 GIMHv: BugCheck 1e {ffffffff80000002, fffff806114d9e98, 0, 1}
00:02:04.639388 KMODE_EXCEPTION_NOT_HANDLED
00:02:04.639388 P1: ffffffff80000002 - exception code - STATUS_DATATYPE_MISALIGNMENT
00:02:04.639388 P2: fffff806114d9e98 - EIP/RIP
00:02:04.639388 P3: 0000000000000000 - Xcpt param #0
00:02:04.639389 P4: 0000000000000001 - Xcpt param #1

This Bug Check and the one you've posted both indicate problems with a kernel level driver inside the Windows guest OS. Further details could be investigated using a debugger in the Windows guest, but that is something I cannot really help you with.

You already suspected an update on the Linux guest. My primary suspect would be the graphics driver in this case. You could also investigate which Linux packages got updated for further candidates.

Yeah quite unfortunate. I have now also tried:

Re-installing Fedora. Tried RPM fusion and VirtualBox versions of VBox.
I can install Windows VMs in both kvm and vmware just fine without any issues. Unfortunately I have so little to go on with the bug check error it really doesn't push me in the right direction. There were some package updates to for gcc, dkms etc. Was hoping that would fix it. No luck though.

[Macy]

I confirm this exact issue occurs on Ubuntu Linux 20.04.1 with hwe kernel 5.8.0-36-generic and Virtualbox 6.1.16-140961~Ubuntu~eoan installed through the official virtualbox focal apt repository

This problem occurs even with a virgin installation of Windows 10 (guest), so it is not attributable to 3rd party drivers. It does not occur to me with the non-hwe Ubuntu 20.04.1 kernel 5.4.

My error log:

Code: Select all

00:21:26.789581 GIM: HyperV: Guest indicates a fatal condition! P0=0x1e P1=0xffffffff80000002 P2=0xfffff8065acd7d93 P3=0xffffe0077a4663bb P4=0x7010008004002001
00:21:26.789648 GIMHv: BugCheck 1e {ffffffff80000002, fffff8065acd7d93, ffffe0077a4663bb, 7010008004002001}
00:21:26.789649 KMODE_EXCEPTION_NOT_HANDLED
00:21:26.789649 P1: ffffffff80000002 - exception code - STATUS_DATATYPE_MISALIGNMENT
00:21:26.789649 P2: fffff8065acd7d93 - EIP/RIP
00:21:26.789649 P3: ffffe0077a4663bb - Xcpt param #0
00:21:26.789649 P4: 7010008004002001 - Xcpt param #1
00:21:30.726903 VMMDev: vmmDevHeartbeatFlatlinedTimer: Guest seems to be unresponsive. Last heartbeat received 4 seconds ago