karlio
June 28, 2022, 10:04pm
1
Kind of a newb, here, and lookin’ for help.
I can’t get Ubuntu-based distros to install (Ubuntu, Kubuntu, etc., as well as Mint, Feren, and Pop. They all stall before the the installation finishes. Virt-manager shows them as paused. They will, however, run in a live environment. I think these are the relevant libvirt.log lines:
host doesn't support requested feature: MSR(48BH).vmx-apicv-register [bit 8]
host doesn't support requested feature: MSR(48BH).vmx-apicv-vid [bit 9]
host doesn't support requested feature: MSR(48DH).vmx-posted-intr [bit 7]
Are these kernel parameters?
There’s one line in the output from virt-host-validate that maybe says something relevant:
Enable 'devices' in kernel Kconfig file or mount/enable cgroup controller in your system
The thing is, Endeavour, Debian, and Slackware VM’s work. Windows 10 VM installation throws the same errors.
This all happens on a Dell i5-3470 running Debian 11 stable.
Any assistance is much appreciated.
Yoray
June 28, 2022, 11:23pm
2
Which version of Debian 11 are you using ?
opened 06:43PM - 12 Mar 21 UTC
closed 07:42PM - 24 May 21 UTC
**Distro**: Ubuntu 20.04.2 LTS (Focal Fossa)
**Kernel**: 5.10.23-xanmod1 x86_64…
**Desktop**: KDE Plasma 5.18.5
**CPU**: e3-1275v2
**Motherboard**: Supermicro X9SAE
These features seem to not be working:
MSR(48DH).vmx-posted-intr [bit 7]
MSR(48BH).vmx-apicv-register [bit 8]
MSR(48BH).vmx-apicv-vid [bit 9]
With the research I have done so far I found a related report on the kubevirt project: https://github.com/kubevirt/kubevirt/issues/3587
The user that filled the report said this:
_Adding a CPU model like Haswell--noTSX-IBRS solves the issue. I wonder why qemu does not detect the correct CPU features, though, as it has the cpu set to host-model and does some additional checking of features as well._
My cpu is not a Haswell, its a xeon, however I would like to try adding a CPU model once I figure out where to put this information into virt-manager and I also need to figure out the correct cpu model for my cpu, surely its not just xeon, there are tons of variations of those.
Here is the details log from virt-manager when I was attempting to create a new VM.
```
Unable to complete install: 'internal error: process exited while connecting to monitor: 2021-03-12T18:24:28.949747Z qemu-system-x86_64: warning: host doesn't support requested feature: MSR(48DH).vmx-posted-intr [bit 7]
2021-03-12T18:24:28.950415Z qemu-system-x86_64: warning: host doesn't support requested feature: MSR(48BH).vmx-apicv-register [bit 8]
2021-03-12T18:24:28.950422Z qemu-system-x86_64: warning: host doesn't support requested feature: MSR(48BH).vmx-apicv-vid [bit 9]
2021-03-12T18:24:28.950426Z qemu-system-x86_64: warning: host doesn't support requested feature: MSR(48DH).vmx-posted-intr [bit 7]
2021-03-12T18:24:29.401942Z qemu-system-x86_64: -device vfio-pci,host=0000:02:00.0,id=hostdev2,bus=pci.3,addr=0x0: vfio 0000:02:00.0: group 1 is not viable
Please ensure all devices within the iommu_group are bound to their vfio bus driver.
libusb: error [udev_hotplug_event] ignoring udev action change
libusb: error [udev_hotplug_event] ignoring udev action change'
Traceback (most recent call last):
File "/usr/share/virt-manager/virtManager/asyncjob.py", line 75, in cb_wrapper
callback(asyncjob, *args, **kwargs)
File "/usr/share/virt-manager/virtManager/createvm.py", line 2089, in _do_async_install
guest.installer_instance.start_install(guest, meter=meter)
File "/usr/share/virt-manager/virtinst/install/installer.py", line 542, in start_install
domain = self._create_guest(
File "/usr/share/virt-manager/virtinst/install/installer.py", line 491, in _create_guest
domain = self.conn.createXML(install_xml or final_xml, 0)
File "/usr/lib/python3/dist-packages/libvirt.py", line 4034, in createXML
if ret is None:raise libvirtError('virDomainCreateXML() failed', conn=self)
libvirt.libvirtError: internal error: process exited while connecting to monitor: 2021-03-12T18:24:28.949747Z qemu-system-x86_64: warning: host doesn't support requested feature: MSR(48DH).vmx-posted-intr [bit 7]
2021-03-12T18:24:28.950415Z qemu-system-x86_64: warning: host doesn't support requested feature: MSR(48BH).vmx-apicv-register [bit 8]
2021-03-12T18:24:28.950422Z qemu-system-x86_64: warning: host doesn't support requested feature: MSR(48BH).vmx-apicv-vid [bit 9]
2021-03-12T18:24:28.950426Z qemu-system-x86_64: warning: host doesn't support requested feature: MSR(48DH).vmx-posted-intr [bit 7]
2021-03-12T18:24:29.401942Z qemu-system-x86_64: -device vfio-pci,host=0000:02:00.0,id=hostdev2,bus=pci.3,addr=0x0: vfio 0000:02:00.0: group 1 is not viable
Please ensure all devices within the iommu_group are bound to their vfio bus driver.
libusb: error [udev_hotplug_event] ignoring udev action change
libusb: error [udev_hotplug_event] ignoring udev action change
```
Also incase anyone asks, the pci devices I am trying to passthrough are properly using vfio-pci
sudo lspci -nnv
Kernel driver in use: vfio-pci
EDIT: seems to add the cpu model I just enable xml editing and add it there, currently It is set to this as default:
```
<cpu mode="host-model">
<topology sockets="1" cores="4" threads="1"/>
</cpu>
```
EDIT2: I figured out how to find the cpu model:
sudo virsh -c qemu:///system capabilities
`<model>IvyBridge-IBRS</model>`
EDIT3: I edited the XML to this: (This was unnecessary, you can just uncheck copy host and select your model from the dropdown once you know it.)
```
<cpu mode="custom" match="exact">
<model>IvyBridge-IBRS</model>
<topology sockets="1" cores="4" threads="1"/>
</cpu>
```
The install now has a different error:
```
Unable to complete install: 'internal error: qemu unexpectedly closed the monitor: 2021-03-12T19:14:32.176083Z qemu-system-x86_64: -device vfio-pci,host=0000:02:00.0,id=hostdev2,bus=pci.3,addr=0x0: vfio 0000:02:00.0: group 1 is not viable
Please ensure all devices within the iommu_group are bound to their vfio bus driver.
libusb: error [udev_hotplug_event] ignoring udev action change
libusb: error [udev_hotplug_event] ignoring udev action change'
Traceback (most recent call last):
File "/usr/share/virt-manager/virtManager/asyncjob.py", line 75, in cb_wrapper
callback(asyncjob, *args, **kwargs)
File "/usr/share/virt-manager/virtManager/createvm.py", line 2089, in _do_async_install
guest.installer_instance.start_install(guest, meter=meter)
File "/usr/share/virt-manager/virtinst/install/installer.py", line 542, in start_install
domain = self._create_guest(
File "/usr/share/virt-manager/virtinst/install/installer.py", line 491, in _create_guest
domain = self.conn.createXML(install_xml or final_xml, 0)
File "/usr/lib/python3/dist-packages/libvirt.py", line 4034, in createXML
if ret is None:raise libvirtError('virDomainCreateXML() failed', conn=self)
libvirt.libvirtError: internal error: qemu unexpectedly closed the monitor: 2021-03-12T19:14:32.176083Z qemu-system-x86_64: -device vfio-pci,host=0000:02:00.0,id=hostdev2,bus=pci.3,addr=0x0: vfio 0000:02:00.0: group 1 is not viable
Please ensure all devices within the iommu_group are bound to their vfio bus driver.
libusb: error [udev_hotplug_event] ignoring udev action change
libusb: error [udev_hotplug_event] ignoring udev action change
```
EDIT4: This one I think I can fix, I just double checked and two of my gpu are reporting in t he same IOMMU Group so I will need the ACS patch:
```
IOMMU Group 1 00:01.0 PCI bridge [0604]: Intel Corporation Xeon E3-1200 v2/3rd Gen Core processor PCI Express Root Port [8086:0151] (rev 09)
IOMMU Group 1 00:01.1 PCI bridge [0604]: Intel Corporation Xeon E3-1200 v2/3rd Gen Core processor PCI Express Root Port [8086:0155] (rev 09)
IOMMU Group 1 01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GF104 [GeForce GTX 460] [10de:0e22] (rev a1)
IOMMU Group 1 01:00.1 Audio device [0403]: NVIDIA Corporation GF104 High Definition Audio Controller [10de:0beb] (rev a1)
IOMMU Group 1 02:00.0 VGA compatible controller [0300]: NVIDIA Corporation GP107 [GeForce GTX 1050 Ti] [10de:1c82] (rev a1)
IOMMU Group 1 02:00.1 Audio device [0403]: NVIDIA Corporation GP107GL High Definition Audio Controller [10de:0fb9] (rev a1)
```
EDIT5: xanmod includes the ACS patch by default, which is great for me! I am already up and running, just had to add the instructions to grub:
```
sudo nano /etc/default/grub
GRUB_CMDLINE_LINUX_DEFAULT="splash quiet iommu=1 intel_iommu=on pcie_acs_override=downstream vfio-pci.ids=10de:1c82,10de:0fb9,8086:1e20"
sudo update-initramfs -u
sudo update-grub
sudo reboot
```
try updating if you are on old version, if not you could also check the shared fix.
1 Like
karlio
June 29, 2022, 12:09am
3
Debian 11 (bullseye)
I saw that post, but it seemed a bit over my head. I’ll go through it again and see if I can understand it. I’m not close to trying anything like a pcie passthrough, just need it to get through the install.
karlio
June 30, 2022, 11:15pm
4
Not convinced these errors are the issue. Mainly because they also show up in the log files of successfully installed VM’s. I’m still stumped.
Yoray
June 30, 2022, 11:28pm
5
Okay I checked and yes you are running the latest stable debian, well I have no major experience with debian and haven’t encountered your issue but tried to find some info on their bug tracker -
https://groups.google.com/g/linux.debian.bugs.dist/c/4SHYNSb7cB4/m/8ej0x8fpBAAJ
this is the closest I could find but I don’t really have time to search. Here is the link, try to find something and if you can’t you could also write to them. -
https://groups.google.com/g/linux.debian.bugs.dist/search?q=qemu
Here is how to report a bug -https://www.debian.org/Bugs/Reporting
btw as it seems to be a qemu bug you could also try installing an older package and try using that.
karlio
July 10, 2022, 7:59pm
6
First and foremost let me say, thank you for the help.
This was really bugging me. It just didn’t make sense that some VM’s would install and others wouldn’t. The log file I shared was a red herring. The real problem was that I was using the root directory for the storage pool. And it was presumably running out of space. I moved the existing VM’s to my home partition and now additional VM’s work just fine.
1 Like
