- Users - libvirt List Archives

[libvirt-users] Unable to provision VM attaching it directly to a OVS bridge
by Lucas Brasilino 29 Oct '13

29 Oct '13

Reposting from virt-tools mailing list: Hi! I'm facing a problem that could be triggered by some lacking of support from libvirt on Open vSwitch (or could be my mistake). I have interests in researching on virtual networks and SDN. To keep things simple, I've decided to use libvirt/virt-tools to manage VM's since my focus is on the network, instead of using a full feature system like OpenStack. I'm quite new with libvirt/virt-tools, but I have a good experience with openvswitch and other virtualizations technologies (which I dropped in libvirt/kvm favor). I'm using Fedora 19 packages openvswitch (1.11.0-1.fc19.x86_64) libvirt (1.0.5.6-3.fc19.x86_64) virt-install (0.10.0-4.fc19.noarch) I've created an OVS bridge (databr0) outside libvirt, and then defined it inside libvirt with: <network> <name>databr0</name> <forward mode='bridge'/> <bridge name='databr0'/> <virtualport type='openvswitch'/> </network> And then 'net-autostart' and 'net-start' it with virsh. Now I've got: # virsh net-list Name State Autostart Persistent ---------------------------------------------------------- databr0 active yes yes When I try to provision an VM, if I use the virt-install option "--network=bridge:databr0,model=virtio,mac=00:00:00:00:00:03" I got the following error: # virt-install --connect qemu:///system --virt-type kvm --name vm2 --ram 768 --disk path=/home/lucas/local/vm/images/vm2.img --vnc --cdrom /home/lucas/local/vm/fc19-x86_64.iso --network=bridge:databr0,model=virtio,mac=00:00:00:00:00:03 --os-type=linux --os-variant fedora19 Starting install... ERROR Unable to add bridge databr0 port vnet0: Operation not supported Domain installation does not appear to have been successful. If it was, you can restart your domain by running: virsh --connect qemu:///system start vm2 otherwise, please restart your installation. I just managed to create a VM when I use '--nonetworks' option and after I do a 'virtsh edit vm2' and add: <interface type='bridge'> <mac address='00:00:00:00:00:03'/> <source bridge='databr0'/> <virtualport type='openvswitch'/> <model type='virtio'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0'/> </interface> Well, I read elsewhere that openvswitch bridging isn't fully supported. Is it the case or I'm facing another kind of problem ? regards Att Lucas Brasilino MSc Student @ Federal University of Pernambuco (UFPE) twitter: @lucas_brasilino

1 1

[libvirt-users] Is it possible to define serial port speed?
by Marco André Dinis 28 Oct '13

28 Oct '13

Hi I would like to know if it's possible to define serial port speed using libvirt. I'm trying to connect the guest with the host for communication. Usually only small messages are transferred, but i might need to transfer huge files, so i would like to know if i can use SerialPort interface to do that. Thanks anyway, Marco Dinis

2 1

[libvirt-users] how can I use libvirt to connect vbox
by hackxay 28 Oct '13

28 Oct '13

Hi. I want to connect hypervisor vbox using command of "virsh connect vbox:///session". But there is an error occurred: [root@vbox ~]# virsh connect vbox:///session error: Failed to connect to the hypervisor error: internal error unable to initialize VirtualBox driver API And my os is Fedora 19. The VirtualBox is 4.3. Some other info is there: [root@vbox ~]# virsh version Compiled against library: libvirt 1.0.5 Using library: libvirt 1.0.5 Using API: QEMU 1.0.5 Running hypervisor: QEMU 1.4.2 Thanks.

1 0

[libvirt-users] virsh migrate fails with Input/output error
by Tim 27 Oct '13

27 Oct '13

Hey guys, I'm running the following setup on a few hosts: ~ # virsh version Compiled against library: libvir 0.9.8 Using library: libvir 0.9.8 Using API: QEMU 0.9.8 Running hypervisor: QEMU 1.0.0 Online Migrations with local storage are working smooth, Host OS is always Ubuntu 12.04, CPUs are Intel E3-1245 V2, 32GB Ram. Now I installed the same software on different hardware: Intel E5-2620, ~128GB Ram The problem: Online migrations from an E5-2620 to another E5-2620 aren't possible: ~ # virsh --debug=1 migrate --verbose my-virtual-maschine qemu+ssh://virt@192.168.99.16:222/system migrate: verbose(bool): (none) migrate: domain(optdata): vm-42800 migrate: desturi(optdata): qemu+ssh://virt@192.168.99.16:222/system migrate: found option <domain>: vm-42800 migrate: found option <domain>: vm-42800 error: End of file while reading data: : Input/output error error: Reconnected to the hypervisor I can't find anything related to this in the libvirt logfile or via an strace. Let me know if you have any ideas or if you need more information.

1 1

[libvirt-users] Automated hashdeep and upgrade for VMs
by Sean Alexandre 25 Oct '13

25 Oct '13

Are there any utilities or scripts that automate the process of upgrading VMs, along with other maintenance such as doing a hashdeep check of system files? The key features I'm looking for would do something like: - A: Shut down VM. - B: Do a hashdeep check of system files. - C: Optionally restore VM from a previous backup. - D: Restart VM. - E: Upgrade packages. - F: Shut down VM. - G: Backup VM (for next "C".) - H: Run hashdeep on system files (for next "B".) - I: Restart VM. Other features, but less important, would backup/restore application files as well.

1 0

[libvirt-users] libvirt_lxc causes cpu 100% usage
by 止语 24 Oct '13

24 Oct '13

libvirt_lxc causes cpu 100% usage? libvirt version : 1.1.1 I don't know what happend, any help would be appreciated. thx top - 07:39:44 up 1 day, 2:13, 2 users, load average: 4.72, 4.59, 4.51 Tasks: 243 total, 4 running, 239 sleeping, 0 stopped, 0 zombie %Cpu0 : 33.1 us, 1.3 sy, 0.0 ni, 64.6 id, 0.7 wa, 0.0 hi, 0.3 si, 0.0 st %Cpu1 : 41.1 us, 3.0 sy, 0.0 ni, 55.3 id, 0.7 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu2 : 18.0 us, 2.0 sy, 0.0 ni, 79.7 id, 0.0 wa, 0.0 hi, 0.3 si, 0.0 st %Cpu3 : 94.8 us, 5.2 sy, 0.0 ni, 0.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu4 : 55.7 us, 7.5 sy, 0.0 ni, 36.7 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu5 : 66.1 us, 4.3 sy, 0.0 ni, 29.6 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu6 : 16.8 us, 1.0 sy, 0.0 ni, 82.2 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu7 : 7.2 us, 0.7 sy, 0.0 ni, 92.1 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st KiB Mem: 1972112 total, 1894912 used, 77200 free, 236576 buffers KiB Swap: 10485744 total, 0 used, 10485744 free, 574544 cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 15750 root 20 0 10448 2080 1216 R 100.3 0.1 1322:38 libvirt_lxc 27517 root 20 0 10448 2032 1200 R 99.3 0.1 1273:44 libvirt_lxc 25250 root 20 0 10480 2048 1200 R 99.0 0.1 1277:31 libvirt_lxc ....

1 0

[libvirt-users] cannot attach device using libvirt
by song 23 Oct '13

23 Oct '13

Hello guys: I am using libvirt 0.9.4 on a CentOS 6.3 server.The operation of attach device won't work.It states "error:Operation failed.Device hotplug is not supported." When issue attach device command, there's nothing happen in guest dmesg log. Anybody came across this problem? Any help will be appreciated!

2 1

Re: [libvirt-users] kvm/libvirt can't use vmdk files? qemu: 'vmdk' invalid format
by Hammad Siddiqi 22 Oct '13

22 Oct '13

Dear All, While trying to run kvm VM using vmdk disk on 6.4 , I got into the following error, error: Failed to create domain from vmdkvm.xml error: internal error process exited while connecting to monitor: qemu-kvm: -drive file=/mnt/vmdkTOKVM/vmdkTOKVM.vmdk,if=none,id=drive-ide0-0-0,format=vmdk,cache=none: 'vmdk' invalid format upon further digging I came across this thread ( https://www.redhat.com/archives/libvirt-users/2011-November/msg00044.html) also qemu-kvm shows that vmdk is supported but all in vain Even if I try running with command line instead of XML I get error though it says it is supported [root@vmdkTOKVM]# /usr/libexec/qemu-kvm -drive file=/mnt/vmdkTOKVM/vmdkTOKVM.vmdk,format=? Supported formats: raw cow qcow vdi vmdk cloop dmg bochs vpc vvfat qcow2 qed parallels nbd blkdebug host_cdrom host_floppy host_device file [root@vmdkTOKVM]# /usr/libexec/qemu-kvm -drive file=/mnt/vmdkTOKVM/vmdkTOKVM.vmdk,format=vmdk qemu-kvm: -drive file=/mnt/XOcom-prx-kvm/XOcom-prx-kvm.vmdk,format=vmdk: 'vmdk' invalid format The packages being used are: qemu-img-0.12.1.2-2.355.0.1.el6_4.9.x86_64 qemu-kvm-0.12.1.2-2.355.0.1.el6_4.9.x86_64 qemu-kvm-tools-0.12.1.2-2.355.0.1.el6_4.9.x86_64 libvirt-0.10.2-18.el6_4.14.x86_64 Any help in this regard will highly be apprectiated. Thank you Hammad Siddqi

1 0

[libvirt-users] hotplug is not supported
by song 22 Oct '13

22 Oct '13

Hello everybody: I am using libvirt 0.9.4 on centos 6.4,when I am trying to attach disk to a running vm, it shows : error: operation failed: drive hotplug is not supported Any body ever come across this problem?

1 0

[libvirt-users] Exploring linbvirt.
by cooldharma06 22 Oct '13

22 Oct '13

hi, i am started to exploring libvirt. i am using the version libvirt-1.1.3 and i want to know (or) find the path how the libvirt is creating VM with xen (libxl) in the source. Any guide or reference is available for this. How i can find the path of calling sequence for other operations(api). Guide me to get some idea on debugging the libvirt. Regards, cooldharma06

1 0