kunkler99 Posted June 3, 2015 Share Posted June 3, 2015 I have having problems creating a windows 8 virtual machine. I installed windows and as soon as it got to the point where it was installing the Apps it froze shutdown and now when I try and turn it back on I get this error "Error: internal error: process exited while connecting to monitor: 2015-06-03T20:00:58.271644Z qemu-system-x86_64: -drive file=/mnt/cache/VM/HTPC/vdisk1.img,if=none,id=drive-virtio-disk2,format=raw,cache=none,aio=native: Could not open '/mnt/cache/VM/HTPC/vdisk1.img': Permission denied" I have not passed through the video card as I was getting another error doing that, so I figured I would try to install it with out the pass through and then add it later. Any ideas on how to fix this? Thanks Link to comment
jonp Posted June 3, 2015 Share Posted June 3, 2015 I have having problems creating a windows 8 virtual machine. I installed windows and as soon as it got to the point where it was installing the Apps it froze shutdown and now when I try and turn it back on I get this error "Error: internal error: process exited while connecting to monitor: 2015-06-03T20:00:58.271644Z qemu-system-x86_64: -drive file=/mnt/cache/VM/HTPC/vdisk1.img,if=none,id=drive-virtio-disk2,format=raw,cache=none,aio=native: Could not open '/mnt/cache/VM/HTPC/vdisk1.img': Permission denied" I have not passed through the video card as I was getting another error doing that, so I figured I would try to install it with out the pass through and then add it later. Any ideas on how to fix this? Thanks I'm not sure what is causing this permissions issue. Are you using any plugins on your system by any chance? You can try to manually adjust the permissions by logging into your server via SSH or Telnet and typing this command: chown nobody:users /mnt/cache/VMVM/HTPC/vdisk1.img Please report back on plugin usage as well if the above resolves your issue. Link to comment
kunkler99 Posted June 3, 2015 Author Share Posted June 3, 2015 No plug-ins just a fresh install of r3. I do have a 180gig ssd as my cache drive would that matter? also the command you sent did not fix anything Link to comment
kunkler99 Posted June 3, 2015 Author Share Posted June 3, 2015 here is the log file if that helps 2015-06-03 19:42:50.938+0000: starting up libvirt version: 1.2.15, qemu version: 2.3.0 Domain id=2 is tainted: high-privileges Domain id=2 is tainted: host-cpu char device redirected to /dev/pts/0 (label charserial0) qemu: terminating on signal 15 from pid 2153 2015-06-03 19:43:17.625+0000: shutting down 2015-06-03 19:43:34.293+0000: starting up libvirt version: 1.2.15, qemu version: 2.3.0 Domain id=3 is tainted: high-privileges Domain id=3 is tainted: host-cpu char device redirected to /dev/pts/0 (label charserial0) qemu-system-x86_64: block.c:3689: bdrv_error_action: Assertion `error >= 0' failed. 2015-06-03 19:56:12.741+0000: shutting down 2015-06-03 19:58:55.836+0000: starting up libvirt version: 1.2.15, qemu version: 2.3.0 Domain id=4 is tainted: high-privileges Domain id=4 is tainted: host-cpu char device redirected to /dev/pts/0 (label charserial0) 2015-06-03T19:58:55.847382Z qemu-system-x86_64: -drive file=/mnt/cache/VM/HTPC/vdisk1.img,if=none,id=drive-virtio-disk2,format=raw,cache=none,aio=native: Could not open '/mnt/cache/VM/HTPC/vdisk1.img': Permission denied 2015-06-03 19:58:56.041+0000: shutting down and the permissions on the file ls -l /mnt/cache/VM/HTPC/vdisk1.img -rw-r--r-- 1 root users 64424509440 Jun 3 15:53 /mnt/cache/VM/HTPC/vdisk1.img Link to comment
jonp Posted June 3, 2015 Share Posted June 3, 2015 I typed the command wrong for the folder path. So if you type this: chown nobody:users /mnt/cache/VM/HTPC/vdisk1.img Does nothing change with the file permission? Link to comment
kunkler99 Posted June 4, 2015 Author Share Posted June 4, 2015 I saw that you added an extra VM so I removed it. I installed a standard hard drive in place of the SSD and reinstalled windows 8. It now works fine through VNC. Now to my original problem, when I try and pass through my Nvidia GeForce 8400 I receive this error "Error: internal error: early end of file from monitor: possible problem: 2015-06-03T21:46:30.907179Z qemu-system-x86_64: -device vfio-pci,host=02:00.0,bus=root.1,addr=00.0,multifunction=on,x-vga=on: vfio: error, group 1 is not viable, please ensure all devices within the iommu_group are bound to their vfio bus driver. 2015-06-03T21:46:30.907195Z qemu-system-x86_64: -device vfio-pci,host=02:00.0,bus=root.1,addr=00.0,multifunction=on,x-vga=on: vfio: failed to get group 1 2015-06-03T21:46:30.907202Z qemu-system-x86_64: -device vfio-pci,host=02:00.0,bus=root.1,addr=00.0,multifunction=on,x-vga=on: Device initialization failed 2015-06-03T21:46:30.907209Z qemu-system-x86_64: -device vfio-pci,host=02:00.0,bus=root.1,addr=00.0,multifunction=on,x-vga=on: Device 'vfio-pci' could not be initialized" here is the log file 2015-06-03 21:46:30.432+0000: starting up libvirt version: 1.2.15, qemu version: 2.3.0 Domain id=4 is tainted: high-privileges Domain id=4 is tainted: custom-argv Domain id=4 is tainted: host-cpu char device redirected to /dev/pts/0 (label charserial0) 2015-06-03T21:46:30.907179Z qemu-system-x86_64: -device vfio-pci,host=02:00.0,bus=root.1,addr=00.0,multifunction=on,x-vga=on: vfio: error, group 1 is not viable, please ensure all devices within the iommu_group are bound to their vfio bus driver. 2015-06-03T21:46:30.907195Z qemu-system-x86_64: -device vfio-pci,host=02:00.0,bus=root.1,addr=00.0,multifunction=on,x-vga=on: vfio: failed to get group 1 2015-06-03T21:46:30.907202Z qemu-system-x86_64: -device vfio-pci,host=02:00.0,bus=root.1,addr=00.0,multifunction=on,x-vga=on: Device initialization failed 2015-06-03T21:46:30.907209Z qemu-system-x86_64: -device vfio-pci,host=02:00.0,bus=root.1,addr=00.0,multifunction=on,x-vga=on: Device 'vfio-pci' could not be initialized 2015-06-03 21:46:30.954+0000: shutting down I have a Asrock z97 extreme6 motherboard with a Intel® Core™ i5-4590S CPU @ 3.00GHz processor. I have all of the virtualization stuff turned on in the bios and I had this working fine with my old Xen setup so I know it works. Any thoughts? Link to comment
jonp Posted June 4, 2015 Share Posted June 4, 2015 Yes. You are attempting to use a device that is part of an iommu group on your chipset for which another device also belongs and is already bound to an in-use driver on the system. The risk (and this existed with Xen as well) is hardware level peer to peer dma occurring between two devices in the same group when assigned to different VM guests or host-assigned services. The pcie acs override setting under vm settings in the webgui allows you to override the hardware enforced safety protocol that prevents the isolated use of independent PCI devices within a group by remapping the groups to one per device. This doesn't solve for the potential of peer to peer dma, but depending on the devices in the group, this risk may be acceptable. We are working on a way to help folks better understand the potential risks here, but its going to be very hardware-specific. The simpler route to understanding right now is that of you are getting this error, please try moving the GPU to a different PCI slot in your motherboard (with the system powered off and power cable detached). Link to comment
kunkler99 Posted June 4, 2015 Author Share Posted June 4, 2015 Thanks for the quick reply, I have removed all of the other card except my nvidia card and I get this error now. "Error: Hook script execution failed: internal error: Child process (LC_ALL=C PATH=/bin:/sbin:/usr/bin:/usr/sbin HOME=/ /etc/libvirt/hooks/qemu HTPC start begin -) unexpected exit status 1: Failed to bind device 02:00.0 to vfio-pci driver" Link to comment
jonp Posted June 4, 2015 Share Posted June 4, 2015 Hmm, do you have an on board graphics chip in addition to the GPU you are trying to assign to the VM? We may need to bind the driver to the pci stub driver before unraid boots. Need to look into this. Link to comment
kunkler99 Posted June 4, 2015 Author Share Posted June 4, 2015 I do have an onboard video card that I use for unriad Link to comment
bungee91 Posted June 4, 2015 Share Posted June 4, 2015 We are working on a way to help folks better understand the potential risks here, but its going to be very hardware-specific. The simpler route to understanding right now is that of you are getting this error, please try moving the GPU to a different PCI slot in your motherboard (with the system powered off and power cable detached). I believe it would be very beneficial to have a discussion involving this, as it is very unclear in the research I have done so far. For instance, I have no idea why by default every one of my Pcie slots are grouped to the same c group!.. This basically makes any use of vt-d useless, and without the downstream flag the inability to assign a card to a Vm. However, this seems like it is up to the motherboard manufacturer to address, and this is certainly not stated in any manual that I've found with a motherboard. I'd really look forward to this discussion as I have seen plenty of instabilities because of this (or so I think) enough to where I'd just prefer to buy a new motherboard that properly supports having Pcie slots in separate groups by default. ----- Edit: I started a thread to discuss, sorry for the derailment http://lime-technology.com/forum/index.php?topic=40409.msg380239#new Link to comment
jonp Posted June 4, 2015 Share Posted June 4, 2015 Kunkler, do you have any plugins installed such as libvirt support or anything else? It seems like you may have a modified libvirt configuration from what we use by default which may be causing the issue. Link to comment
Recommended Posts
Archived
This topic is now archived and is closed to further replies.