- Users - libvirt List Archives

[libvirt-users] Failed to terminate process X with SIGKILL: Device or resource busy
by sylecn 09 Feb '14

09 Feb '14

Hi, I'm using libvirt/kvm with openstack. When vm has CPU stall, the instance can't be destroyed. I have asked the question on openstack user mailing list and received no answers in weeks. Here is one example: # virsh destroy instance-00000085 error: Failed to destroy domain instance-00000085 error: Failed to terminate process 61222 with SIGKILL: Device or resource busy # lsof -p 61222 # ps -f -p 61222 UID PID PPID C STIME TTY TIME CMD 107 61222 1 68 Jan24 ? 9-23:06:19 [qemu-system-x86] <defunct> The log in /var/log/libvirt/libvirtd.log: 2014-02-08 03:21:31.597+0000: 2177: error : virProcessKillPainfully:315 : Failed to terminate process 61222 with SIGKILL: Device or resource busy Restart libvirt-bin service did not solve the problem. virsh shutdown/reboot just hang. Maybe that is because the CPU in vm is stalled. I wonder if there is some way to destroy (or rebuild) dead instance without rebooting the hypervisor. Any suggestions is welcome. The software I am running: hypervisor os: ubuntu 12.04.3 hypervisor kernel: 3.8.0-35-generic vm os: ubuntu 12.04 vm kernel: 3.2.0-56-virtual openstack: havana libvirt: 1.1.1-0ubuntu8~cloud2 Thanks, Yuanle

1 1

[libvirt-users] Automate Windows 7 installation via Answer file
by RAJESH David John 08 Feb '14

08 Feb '14

Hello, Is there a means to automate windows 7 installation after Running "Virsh create windows7.xml" OS = Centos 6.2 Hypervisor = KVM Any ideas or pointer is greatly appreciated. Thanks Rajesh J

3 3

[libvirt-users] Connecting disks to controller virtio-scsi of qemu/kvm
by chickenmarkus＠freenet.de 08 Feb '14

08 Feb '14

Hello, I try to use the virtio-scsi driver to use my disks. But my setup at first: * in general Debian Wheezy * Kernel: 3.11-0.bpo.2-amd64 (http://packages.debian.org/wheezy-backports/linux-image-3.11-0.bpo.2-amd64) * Qemu-KVM: 1.7.0 (http://packages.debian.org/jessie/qemu-kvm) * Virsh 1.2.1 (http://packages.debian.org/jessie/libvirt-bin) The new kernel and dirty things from Jessie are for discard support (also not working with current configuration despite thin volumes with working discard). Once I have the this "new" qemu version, why not to use virtio... Every disk with target bus "scsi" gets an own scsi-hd device. Libvirt ignores the scsi controller of type virtio-scsi. Here is a XML snippet: <devices> <controller type='scsi' model='virtio-scsi' index="0"/> <disk type="block" device="disk"> <source dev="/dev/ssd0/sarabi"/> <driver discard="unmap"/> <target dev="sda" bus="scsi"/> <address type="drive" controller="0" bus=0/> </disk> </devices> This creates the virtio-scsi controller (verified by lspci in guest) but do not "connect" the disks to it. The resulting command line looks following: kvm [...] -device virtio-scsi-pci,id=scsi0,bus=pci.0,addr=0x5 -drive file=/dev/ssd0/sarabi,if=none,id=drive-scsi0-0-0-0,format=raw,discard=unmap -device scsi-hd,bus=scsi0.0,channel=0,scsi-id=0,lun=0,drive=drive-scsi0-0-0-0,id=scsi0-0-0-0,bootindex=2 [...] How get my disk to run over virtio-scsi? Or did I found something like a bug? Bye Markus

1 2

[libvirt-users] Can I move the disk image of the guest while it is running?
by Gergely Horváth 07 Feb '14

07 Feb '14

Hello guys! We want to move some guests' disk images to different locations (within the same machine, locally), but we do not want to turn off the guests. Can we do that while the guest is running, like live migration? I did not find any related documentation or posts about it. Let's say the guest has a disk at /ssd/image.raw We want to move that to /hdd/image.raw Is there any way to do that without interrupting the guest? -- Üdvözlettel / Best regards Horváth Gergely | gergely.horvath(a)inepex.com IneTrack - Nyomkövetés egyszerűen | Inepex Kft. Ügyfélszolgálat: support(a)inetrack.hu | +36 30 825 7646 | support.inetrack.hu Web: www.inetrack.hu | nyomkovetes-blog.hu | facebook.com/inetrack Inepex - The White Label GPS fleet-tracking platform | www.inepex.com

4 10

[libvirt-users] Question about managing migration transfers
by Joaquim Barrera 07 Feb '14

07 Feb '14

Hi folks, As part of a project I am working on, I need to control every step of a migration, from when I establish the connection to the destination, to when I decide to move data blocks, its size, etc. As I can see in the documentation, tunnelled migration allows some kind of control, as the data goes through libvirt before leaving the host. May someone put more light on the subject? Is there any possibility to achieve such control I am looking for, or libvirt hacking will be needed? Thank you very much.

1 0

[libvirt-users] Extending an existing LVM pool - vgextend and pool XML
by Daniel H Barboza 05 Feb '14

05 Feb '14

Hi, I am trying to figure it out the best way to add a disk/partition to an existing LVM pool. A google search led me to this link (a thread from this same mailing list, September 2013): https://www.redhat.com/archives/libvirt-users/2013-September/msg00136.html The interesting part is this: " On 21/09/13 02:15, McEvoy, James wrote: I looked around but could not find any info on how to expand a libvirt managed LVM storage pool. I do not see any virsh command to do it but I was successful using the vgexpand command to add some more storage once I destroyed the pools and then restarted it. I'd like to verify that this is the proper way to grow the storage in a libvirt LVM storage pool. It's not the proper way in principle, but since libvirt doesn't supportto extendthe volume groups for a logical pool yet, it's the only way now. Onehint though,you can try to refresh the pool by command "virsh pool-refresh" afterthe volume group of the pool is extended. I.e, don't have to restart the pool." I've tried this method and it works, but checking the pool XML using "virsh pool-edit <vg_name>" shows that the XML didn't change. Is it intended? The versions I'm using: # libvirtd --version libvirtd (libvirt) 1.1.3 # virsh --version 1.1.3 Any thoughts/comments are welcome. Thanks!

1 0

[libvirt-users] libvirt-1.2.0
by cool dharma06 04 Feb '14

04 Feb '14

hi, i am using libvirt-1.2.0 which is installed from source. Which i am using to launch instance on xen. its working fine when i am launching vm from the root user. But i am entered as another user means its throwing the error. And libvirt is working fine in that user except the creating vm: Please refer the following things: eucalyptus@cooldharma06:/opt/i-47FD0798$ virsh -c xen:/// WARNING: gnome-keyring:: couldn't connect to: /home/dharmalingams/.cache/keyring-r004e7/pkcs11: Permission denied Welcome to virsh, the virtualization interactive terminal. Type: 'help' for help with commands 'quit' to quit virsh # version Compiled against library: libvirt 1.2.0 Using library: libvirt 1.2.0 Using API: Xen 1.2.0 Running hypervisor: Xen 4.3.0 virsh # list Id Name State ---------------------------------------------------- *virsh # create libvirt_sec.xml * *error: Failed to create domain from libvirt_sec.xml* *error: access denied* >From the libvirtd.log: 2014-01-31 09:46:16.329+0000: 27937: error : virAccessDriverPolkitCheck:176 : access denied: Policy kit denied action org.libvirt.api.domain.write from 28281,1877379: exit status 1 2014-01-31 09:46:16.329+0000: 27937: error : virAccessManagerSanitizeError:203 : access denied 2014-01-31 09:46:16.329+0000: 27937: error : virAccessManagerSanitizeError:203 : access denied Suggest me some solution to solve this one. Thanks & Regards, cooldharma06.

2 1

[libvirt-users] Management of host device for network passthrough
by Richard Gomes 03 Feb '14

03 Feb '14

Hello, I've configured a network passtrough like this: <interface type='direct'> <mac address='52:54:00:ec:e8:d9'/> <source dev='eth1' mode='passthrough'/> <target dev='macvtap0'/> <model type='rtl8139'/> <alias name='net1'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x09' function='0x0'/> </interface> OK. it works :) There's only one small difficulty: I have to make sure the host network interface is down before I start the VM. Then, after the VM starts, I have to bring the host network interface up by hand. If the host interface is up before the VM starts, the VM fails to start. If I do not bring the host network interface up after the VM starts, the VM cannot effectively communicate using that network interface. Does anybody has a "clean" solution for it, I mean: without employing some external "dirty" script or so? Thanks -- Richard Gomes http://rgomes.info http://www.linkedin.com/in/rgomes mobile: +44(77)9955-6813 inum <http://www.inum.net/>: +883(5100)0800-9804 sip:rgomes@ippi.fr

1 0

[libvirt-users] Trouble configuring with macvtap passthrough on Debian Wheezy / Jessie
by Richard Gomes 03 Feb '14

03 Feb '14

( Posting again. Correct subject line now! ) Hello, I'm trying to use macvtap on Debian Wheezy. Actually, I've installed a recent version of libvirt and qemu from Jessie, using wheezy-backports. $ virsh version Compiled against library: libvirt 1.2.1 Using library: libvirt 1.2.1 Using API: QEMU 1.2.1 Running hypervisor: QEMU 1.7.0 I'm trying to configure a macvtap interface like this: <interface type='direct'> <mac address='52:54:00:7b:05:cd'/> <source dev='eth1' mode='passthrough'/> <model type='rtl8139'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x08' function='0x0'/> </interface> The interface eth1 is used exclusively by this VM, so as far as I understand, mode 'passthrough' is enough since I'd like to "plug" the VM straight onto a public static IP. On my /etc/networks/interfaces, I have this: # start :: define eth1 iface eth1 inet manual # end :: define eth1 It does not matter whether eth1 is up or down, when I try to start the VM, I get the following error: Error starting domain: Cannot create macvlan devices on this platform: Function not implemented Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/asyncjob.py", line 45, in cb_wrapper callback(asyncjob, *args, **kwargs) File "/usr/share/virt-manager/virtManager/asyncjob.py", line 66, in tmpcb callback(*args, **kwargs) File "/usr/share/virt-manager/virtManager/domain.py", line 1114, in startup self._backend.create() File "/usr/lib/python2.7/dist-packages/libvirt.py", line 866, in create if ret == -1: raise libvirtError ('virDomainCreate() failed', dom=self) libvirtError: Cannot create macvlan devices on this platform: Function not implemented Trying to solve the issue, I found this: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=737097 I've tried to rebuild the package from Debian sources, but apparently, macvtap is already defined, as I've reported to another mailing list. For your information: > I'm new to building packages from Debian sources, so I followed > instructions from > > https://wiki.debian.org/HowToPackageForDebian#Building_Debian_packages > > $ apt-get -t wheezy-backports source libvirt > $ cd libvirt-0.9.12.3/ > $ sudo apt-get build-dep libvirt > $ debuild -i -us -uc -b > > So... yes, it works and builds without any errors. > > Next step now is applying the changes mentioned. > But I'm confused because /apparently/ debian/rules is already the way > it should be. > > ifneq (,$(findstring $(DEB_HOST_ARCH_OS), linux)) > WITH_STORAGE_LVM = --with-storage-lvm > WITH_STORAGE_ISCSI = --with-storage-iscsi > WITH_STORAGE_DISK = --with-storage-disk > WITH_UDEV = --with-udev --without-hal > WITH_CAPNG = --with-capng > WITH_POLKIT = --with-polkit > * WITH_MACVTAP = --with-macvtap** > ** WITH_NETWORK = --with-network** > * WITH_QEMU = --with-qemu > WITH_OPENVZ = --with-openvz > WITH_NETCF = --with-netcf > ifneq (,$(findstring $(DEB_HOST_ARCH), amd64 i386 ia64 mips mipsel powerpc)) > WITH_NUMA = --with-numactl > else > WITH_NUMA = --without-numactl > endif > ifneq (,$(findstring $(DEB_HOST_ARCH), ia64)) > WITH_LXC = --without-lxc > else > WITH_LXC = --with-lxc > endif > else > WITH_STORAGE_LVM = --without-storage-lvm > WITH_STORAGE_ISCSI = --without-storage-iscsi > WITH_STORAGE_DISK = --without-storage-disk > WITH_UDEV = --without-udev --with-hal > WITH_CAPNG = --without-capng > WITH_POLKIT = --without-polkit > * WITH_MACVTAP = --without-macvtap** > ** WITH_NETWORK = --without-network** > * WITH_QEMU = --without-qemu > WITH_LXC = --without-lxc > WITH_NUMA = --without-numactl > WITH_NETCF = --without-netcf > endif > > > I suppose that nothing has to be changed, correct? ... since the > _/then/_ branch already has the change mentioned and the _/else/_ > branch probably means that nothing related to networking would be > enabled anyway. > > Am I missing something ? Thoughts? Thanks a lot :) -- Richard Gomes http://rgomes.info http://www.linkedin.com/in/rgomes mobile: +44(77)9955-6813 inum <http://www.inum.net/>: +883(5100)0800-9804 sip:rgomes@ippi.fr

1 1

[libvirt-users] intel quad gigabit nic and pci passthrough
by Ivan Kabaivanov 02 Feb '14

02 Feb '14

Hi all I have a very weird case of pci passthrough. I have a machine with 7 network interfaces, all of them intel. Four of them are on one quad giga ethernet device. If I manually unbind the devices and allow qemu to use them, with intel IOMMU working, everything works like a charm. Here's how I do it manually: root@kybrat (x86_64) ~]$ lspci -nn | grep net 00:19.0 Ethernet controller [0200]: Intel Corporation 82579V Gigabit Network Connection [8086:1503] (rev 06) 02:00.0 Ethernet controller [0200]: Intel Corporation 82572EI Gigabit Ethernet Controller (Copper) [8086:107d] (rev 06) 05:00.0 Ethernet controller [0200]: Intel Corporation 82571EB Gigabit Ethernet Controller [8086:10a4] (rev 06) 05:00.1 Ethernet controller [0200]: Intel Corporation 82571EB Gigabit Ethernet Controller [8086:10a4] (rev 06) 06:00.0 Ethernet controller [0200]: Intel Corporation 82571EB Gigabit Ethernet Controller [8086:10a4] (rev 06) 06:00.1 Ethernet controller [0200]: Intel Corporation 82571EB Gigabit Ethernet Controller [8086:10a4] (rev 06) 0c:00.0 Ethernet controller [0200]: Intel Corporation 82572EI Gigabit Ethernet Controller (Copper) [8086:107d] (rev 06) echo "8086 10a4" > /sys/bus/pci/drivers/pci-stub/new_id echo "8086 107d" > /sys/bus/pci/drivers/pci-stub/new_id echo "0000:05:00.0" > /sys/bus/pci/devices/0000\:05\:00.0/driver/unbind echo "0000:05:00.0" > /sys/bus/pci/drivers/pci-stub/bind echo "0000:05:00.1" > /sys/bus/pci/devices/0000\:05\:00.1/driver/unbind echo "0000:05:00.1" > /sys/bus/pci/drivers/pci-stub/bind echo "0000:06:00.0" > /sys/bus/pci/devices/0000\:06\:00.0/driver/unbind echo "0000:06:00.0" > /sys/bus/pci/drivers/pci-stub/bind echo "0000:06:00.1" > /sys/bus/pci/devices/0000\:06\:00.1/driver/unbind echo "0000:06:00.1" > /sys/bus/pci/drivers/pci-stub/bind echo "0000:0c:00.0" > /sys/bus/pci/devices/0000\:0c\:00.0/driver/unbind echo "0000:0c:00.0" > /sys/bus/pci/drivers/pci-stub/bind qemu-system-i386 -enable-kvm \ -m 2048 \ -cpu host \ -machine pc-q35-1.6 \ -drive file=/media/virtual/krym.qcow2,if=virtio \ -net none \ -device pci-assign,host=06:00.1,rombar=0 \ -device pci-assign,host=05:00.0,rombar=0 \ -device pci-assign,host=06:00.0,rombar=0 \ -device pci-assign,host=05:00.1,rombar=0 \ -device pci-assign,host=0c:00.0,rombar=0 \ -cdrom /media/virtual/asg-9.107-33.1.iso \ -boot c \ -vnc :0 \ -runas virtual \ -monitor unix:/media/virtual/control/krym/monitor,server,nowait \ -rtc base=localtime,clock=host,driftfix=none This starts qemu and pci passthrough works nicely. Now, if I try to do the same thing through libvirt, virt-install doesn't like the quad ethernet device pci passthrough and I get the error below: virt-install --name=krym --ram=2048 --vcpus=1 --cpu=host --import --os-type=linux --disk path=/media/virtual/krym.qcow2,format=qcow2,bus=virtio --nonetworks --graphics vnc,port=5900 --virt-type=kvm --machine=q35 --accelerate --host-device=pci_0000_06_00_1 --host-device=pci_0000_05_00_0 --host-device=pci_0000_06_00_0 --host-device=pci_0000_05_00_1 --host-device=pci_0000_0c_00_0 Starting install... ERROR Requested operation is not valid: PCI device 0000:06:00.1 is not assignable Domain installation does not appear to have been successful. If it was, you can restart your domain by running: virsh --connect qemu:///system start krym otherwise, please restart your installation. If I try to use just one NIC (any NIC, as long as it's not the quad NIC), things do work: virt-install --name=krym --ram=2048 --vcpus=1 --cpu=host --import --os-type=linux --disk path=/media/virtual/krym.qcow2,format=qcow2,bus=virtio --nonetworks --graphics vnc,port=5900 --virt-type=kvm --machine=q35 --accelerate --host-device=pci_0000_0c_00_0 Starting install... Creating domain... | 0 B 00:00:01 Cannot open display: Run 'virt-viewer --help' to see a full list of available command line options Domain creation completed. You can restart your domain by running: virsh --connect qemu:///system start krym So I guess libvirt has some problem with the intel quad NIC. I will appreciate any help. Thanks, IvanK.

2 2