- Users - libvirt List Archives

[libvirt-users] correct way to hot-add cdrom ?
by Alexandr 06 Dec '13

06 Dec '13

Good day to all. i have problems with cdrom hot adding code. currently i using virDomainAttachDevice with type=file, device=cdrom, dev=hdc, this code works for machine with one ide hdd and one ide cdrom, but this not work for machine with only one ide hdd, and i looking for solution to hot add cdrom to machine independent of existing devices or i need way to determinate which target device can be attached to vm. thanks in advance for any help.

3 3

[libvirt-users] how to add/remove emulated usb removable devices to guests
by Alexandr 06 Dec '13

06 Dec '13

Good day, i was not able to find api to add/remove usb removable devices to guest, i have windows machines and need to plug usb stick with some fs (fat or ntfs) which is image on host, i was found what it is possible and implemented in libvirt https://bugzilla.redhat.com/show_bug.cgi?id=922495, but i can't find how to use this.

2 1

[libvirt-users] libvirt, Open vSwitch and iptables
by Yoann Juet 06 Dec '13

06 Dec '13

Hi all, We're using since a long time libvirt with KVM guest machines and linux bridges. Firewall rules based on iptables and defined on the host server control inbound/outbound traffic to/from each VM. In order to improve remote administration facility and get extra services, it makes sense for us to replace linux bridges with Open vSwitch. However, the side effect is the solution's inability to filter VM traffic since it's impossible to set-up iptables rules with ovs bridges. OpenStack/Quantum circumvents this problem (no talking about performance) by setting an extra linux bridge and veth pair between the guest TAP and ovs. Is there {a simple|an alternative} solution to achieve it without installing the OpenStack/Quantum layer ? Thanks, Regards, -- Université de Nantes - Direction des Systèmes d'Information IM jabber: yoann.juet(a)univ-nantes.fr

1 0

Re: [libvirt-users] Virsh snapshots
by Juraj Melo 05 Dec '13

05 Dec '13

Thanks for your answer, I am trying to make a snapshot of whole virtual machine (disk, CPU state, memory). But I need to make this snapshot in matter of seconds. I have already try to create snapshot of disk, which is not problem. I use qcow2 format, and create new disk image using original disk image as backing file. But still I am not able to assign state of vm with new disk. I have found some commands for creating snapshots in libvirt API, so in my opinion one way how to accomplish my task is to create utility using theese commands, and create snapshot of vm in RAM - I hope it would be faster. But I wonder whether virsh contains similar functionality, so I won't need to programm it again. Dne 2.12.2013 19:32, Eric Blake napsal(a): On 12/02/2013 09:04 AM, Juraj Melo wrote: > Hello, > > I am working on my PhD thesis and it would be really helpfull if someone > could advise me, whether can Virsh create snapshots of VMs using > copy-on-write. What are you trying to copy-on-write? Is it just disk state, or VM memory state? Do you have corresponding qemu commands that you are trying to figure out if virsh maps to those same commands? There's a lot of flexibility in the virsh snapshot XML, but not all of it has been wired up yet (in part because it's still a moving target for what upstream qemu supports), so knowing more details about what you are trying to do will help us better answer whether it can be done now and if so with which commands.

3 2

[libvirt-users] libvirt, lvm thin provisioning
by Michael Mol 04 Dec '13

04 Dec '13

I know that lvm supports thin provisioning, and I think I have a pretty good grasp on how that works. Does libvirt support lvm thin provisioning and thin snapshots? I know that in order to set up lvm thin provisioning by hand, I have to create a thin-provisioning pool within the volume group, and then thin-provisioned logical volumes and thin-provisioned snapshots can be created within that thin-provision pool. I also know that it's possible to have more than one thin-provisioning pool within the same volume group, which tells me that in order to properly set up any lvm-aware application to use thin-provisioning, I may need to inform it which thin-provisioning pool it should base LVs and snapshots from. When I look at the example here: http://libvirt.org/storage.html#StorageBackendLogical I don't see anything particular which tells me it supports building off a thin-provisioning pool if I specify such by name. Checking here: http://libvirt.org/formatstorage.html#StoragePool I again don't see anything specific about logical thick vs thin provisioning. I would expect to see *something*, as there are performance consequences when making such choices, so I'm again not sure libvirt recognizes a difference. So is this something that libvirt can do? And is there good documentation somewhere on the subject?

3 5

[libvirt-users] cputune shares with multiple cpu and pinning
by Edoardo Comar 04 Dec '13

04 Dec '13

Hi, I have found the cpu time partitioning based on cpu shares weights not very intuitive. On RHEL64, I deployed two qemu/kvm VMs VM1 with 1 vcpu and 512 cpu shares VM2 with 2 vcpus and 1024 cpu shares I pinned their vcpus to specific host pcpus: VM1 vcpu 0 to host pcpu1 VM2 vcpu 0 to host pcpu1, VM2 vcpu 1 to host pcpu2 I executed inside the VMs a simple process that consume all available cpu, eg # cat /dev/zero > /dev/null on the host, using 'top', the reported cpu usage per qemu process is : with 1 process in VM1 and 1 process on vcpu1 in VM2 VM1 = 100%, VM2=100% explanation - without contention for pcpu2 shares are to be irrelevant (that's ok!) with 1 process in VM1 and 1 process on vcpu0 in VM2 VM1=33%, VM2=66% explanation - with contention on pcpu2, host cpu usage is partitioned according to shares (that's ok!) with 1 process in VM1, and 2 processes in VM2 (one on vcpu0 and on vcpu1 - launched with taskset) VM1=50%, VM2=150% This result was a bit unexpected to me: adding load on VM2 resulted in more cpu time allowed for VM1 - can anyone please explain the logic ? Changing the pinning so that both VMs can use the host pcpuset 1-2 then the CPU usage is 100%/100% when VM2 is executing one task only ( on whatever vcpu ) 50%/150% when VM2 is executing two tasks. again, not intuitive just by looking at the shares weights, I think. -------------------------------------------------- Edoardo Comar WebSphere Application Service Platform for Networks (ASPN) ecomar(a)uk.ibm.com +44 (0)1962 81 5576 IBM UK Ltd, Hursley Park, SO21 2JN Unless stated otherwise above: IBM United Kingdom Limited - Registered in England and Wales with number 741598. Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU

1 0

[libvirt-users] virsh iface-list fail
by Slash Sda 04 Dec '13

04 Dec '13

Hi all, I'm using libvirt 1.1.1 on my ubuntu 13.10 minimal. When I try a virsh iface-list i get this message: error: Failed to list interfaces error: internal error: failed to get number of host interfaces I create a BR0 thru my ETH0. On my virt-manager i don't see the computer interfaces What could it be? Thanks

1 0

[libvirt-users] error: Failed to start domain
by Stefan G. Weichinger 04 Dec '13

04 Dec '13

On a gentoo server with libvirt-1.1.3 I get problems with starting VMs. When I do: # virsh start vm180 error: Failed to start domain vm180 error: Input/output error This happens with 1.1.3 and 1.1.4 (I rebuilt the packages and restarted the libvirtd.service). # journalctl -f -u libvirtd.service shows: Dec 03 17:32:56 jupiter libvirtd[24020]: Input/output error Dec 03 17:33:47 jupiter libvirtd[24020]: Input/output error The log for the specific VM shows: 2013-12-03 16:33:47.004+0000: starting up LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin QEMU_AUDIO_DRV=none /usr/bin/qemu-kvm -name vm180 -S -machine pc-i440fx-1.5,accel=kvm,usb=off -cpu Opteron_G5,+bmi1,+perfctr_nb,+perfctr_core,+topoext,+nodeid_msr,+tce,+lwp,+wdt,+skinit,+ibs,+osvw,+cr8legacy,+extapic,+cmp_legacy,+fxsr_opt,+mmxext,+osxsave,+monitor,+ht,+vme -m 3954 -realtime mlock=off -smp 4,sockets=4,cores=1,threads=1 -uuid 646f757f-2ab7-4fdf-2140-a79396200c6f -no-user-config -nodefaults -chardev socket,id=charmonitor,path=/var/lib/libvirt/qemu/vm180.monitor,server,nowait -mon chardev=charmonitor,id=monitor,mode=control -rtc base=localtime,clock=vm,driftfix=slew -no-kvm-pit-reinjection -no-hpet -no-shutdown -device piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 -drive file=/dev/VG01/vm180_disk0,if=none,id=drive-virtio-disk0,format=raw,cache=none,aio=threads -device virtio-blk-pci,scsi=off,bus=pci.0,addr=0x7,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1 -drive file=/mnt/tests/grml64-full_2013.02.iso,if=none,id=drive-ide0-1-0,readonly=on,format=raw -device ide-cd,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0 -netdev tap,fd=27,id=hostnet0,vhost=on,vhostfd=28 -device virtio-net-pci,netdev=hostnet0,id=net0,mac=52:54:00:8d:3b:af,bus=pci.0,addr=0x3 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -vnc 127.0.0.1:5 -vga cirrus -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x6 2013-12-03 16:33:47.013+0000: shutting down libvirt: error : libvirtd quit during handshake: Input/output error What can I do about this issue? Thanks for any pointers, Stefan

1 2

[libvirt-users] help about migration with libvirt
by lei yang 04 Dec '13

04 Dec '13

Hi experts I want to use libvirt doing the migration migration the guest from core 0 to core 2, Is there a way to do this ? any points? Lei

3 4

[libvirt-users] layer2 tunnel between guests across hosts
by sujay g 04 Dec '13

04 Dec '13

Hi, Is it possible to setup layer2 tunnels between guests across hosts? Using vrbr? -- i.e the Guest is unaware. On Host 1: guest (a) --- vrbr0 ------- | | On Host 2: | guest(b) -- vrbr0 ------ Has there been any work done around it? TIA, -Sujay

2 1