- Users - libvirt List Archives

[libvirt-users] Error after failed external snapshot.
by Dominique Ramaekers 21 Jan '16

21 Jan '16

Hi, I use this procedure to backup my VM's: http://wiki.libvirt.org/page/Live-disk-backup-with-active-blockcommit It works on my current host: Compiled against library: libvirt 1.2.16 Using library: libvirt 1.2.16 Gebruikte API: QEMU 1.2.16 Draaiende hypervisor: QEMU 2.1.0 Now I'm doing a test on a testing host: Compiled against library: libvirt 1.2.21 Using library: libvirt 1.2.21 Gebruikte API: QEMU 1.2.21 Draaiende hypervisor: QEMU 2.5.0 The first time, the active block commit failed because the block action was stuck on 100% and I got this error: 'block copy still active: disk 'vda' not ready for pivot yet' I did a 'virsh blockjob PCVIRT1 vda -abort' and started 'virsh blockcommit PCVIRT1 vda --active --verbose -pivot' again. This way, the external snapshot was commited ok. Now I get this error when I want to create an external snapshot: fout: internal error: kan QEMU commando 'transaction' niet uitvoeren: The feature 'snapshot' is not enabled I can't find a lot of information about this. Some help will be appreciated. Greetings, Dominique.

1 1

Re: [libvirt-users] channel device "state" value missing within OpenStack Kilo version
by Jean-Pierre Ribeauville 20 Jan '16

20 Jan '16

Hi, Discard this request , I was investigating on the wrong compute node. It works fine. Sorry for incovenience. Regards, J.P. De : Jean-Pierre Ribeauville Envoyé : mercredi 20 janvier 2016 16:18 À : 'libvirt-users(a)redhat.com' Objet : channel device "state" value missing within OpenStack Kilo version Hi, I've installed a OpenStack KILO all-in-one . For private need , I've to add a channel device to all my GUESTS on my compute node. It works fine (e.g. via a virsh command). Problem is that , in the xml describing the Guest , this 'state' value : <target type='virtio' name='agent' state='disconnected'/> showing if the channel is connected on the Guest side is missing. Does it mean that some qemu/KVM components are too old within KILO OpenStack release ? FYI , here are the installed related packages versions coming with KILO: libvirt-daemon-driver-qemu-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-driver-storage-1.2.8-16.el7_1.4.x86_64 libvirt-python-1.2.8-7.el7_1.1.x86_64 libvirt-daemon-1.2.8-16.el7_1.4.x86_64 libvirt-gconfig-0.1.7-3.el7.x86_64 libvirt-client-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-driver-nodedev-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-config-network-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-driver-network-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-driver-nwfilter-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-config-nwfilter-1.2.8-16.el7_1.4.x86_64 libvirt-glib-0.1.7-3.el7.x86_64 libvirt-daemon-driver-interface-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-driver-secret-1.2.8-16.el7_1.4.x86_64 libvirt-gobject-0.1.7-3.el7.x86_64 libvirt-daemon-kvm-1.2.8-16.el7_1.4.x86_64 [root@localhost bin]# rpm -qa | grep kvm qemu-kvm-1.5.3-86.el7_1.8.x86_64 qemu-kvm-common-1.5.3-86.el7_1.8.x86_64 libvirt-daemon-kvm-1.2.8-16.el7_1.4.x86_64 [root@localhost bin]# rpm -qa | grep qemu libvirt-daemon-driver-qemu-1.2.8-16.el7_1.4.x86_64 qemu-guest-agent-2.1.0-4.el7.x86_64 qemu-kvm-1.5.3-86.el7_1.8.x86_64 qemu-img-1.5.3-86.el7_1.6.x86_64 qemu-kvm-common-1.5.3-86.el7_1.8.x86_64 qemu-img-1.5.3-86.el7_1.8.x86_64 ipxe-roms-qemu-20130517-6.gitc4bce43.el7.noarch Thanks for help, Regards, J.P. Ribeauville P: +33.(0).1.47.17.20.49 . Puteaux 3 Etage 5 Bureau 4 jpribeauville(a)axway.com<mailto:jpribeauville@axway.com> http://www.axway.com<http://www.axway.com/> P Pensez à l'environnement avant d'imprimer.

1 0

[libvirt-users] channel device "state" value missing within OpenStack Kilo version
by Jean-Pierre Ribeauville 20 Jan '16

20 Jan '16

Hi, I've installed a OpenStack KILO all-in-one . For private need , I've to add a channel device to all my GUESTS on my compute node. It works fine (e.g. via a virsh command). Problem is that , in the xml describing the Guest , this 'state' value : <target type='virtio' name='agent' state='disconnected'/> showing if the channel is connected on the Guest side is missing. Does it mean that some qemu/KVM components are too old within KILO OpenStack release ? FYI , here are the installed related packages versions coming with KILO: libvirt-daemon-driver-qemu-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-driver-storage-1.2.8-16.el7_1.4.x86_64 libvirt-python-1.2.8-7.el7_1.1.x86_64 libvirt-daemon-1.2.8-16.el7_1.4.x86_64 libvirt-gconfig-0.1.7-3.el7.x86_64 libvirt-client-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-driver-nodedev-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-config-network-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-driver-network-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-driver-nwfilter-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-config-nwfilter-1.2.8-16.el7_1.4.x86_64 libvirt-glib-0.1.7-3.el7.x86_64 libvirt-daemon-driver-interface-1.2.8-16.el7_1.4.x86_64 libvirt-daemon-driver-secret-1.2.8-16.el7_1.4.x86_64 libvirt-gobject-0.1.7-3.el7.x86_64 libvirt-daemon-kvm-1.2.8-16.el7_1.4.x86_64 [root@localhost bin]# rpm -qa | grep kvm qemu-kvm-1.5.3-86.el7_1.8.x86_64 qemu-kvm-common-1.5.3-86.el7_1.8.x86_64 libvirt-daemon-kvm-1.2.8-16.el7_1.4.x86_64 [root@localhost bin]# rpm -qa | grep qemu libvirt-daemon-driver-qemu-1.2.8-16.el7_1.4.x86_64 qemu-guest-agent-2.1.0-4.el7.x86_64 qemu-kvm-1.5.3-86.el7_1.8.x86_64 qemu-img-1.5.3-86.el7_1.6.x86_64 qemu-kvm-common-1.5.3-86.el7_1.8.x86_64 qemu-img-1.5.3-86.el7_1.8.x86_64 ipxe-roms-qemu-20130517-6.gitc4bce43.el7.noarch Thanks for help, Regards, J.P. Ribeauville P: +33.(0).1.47.17.20.49 . Puteaux 3 Etage 5 Bureau 4 jpribeauville(a)axway.com<mailto:jpribeauville@axway.com> http://www.axway.com<http://www.axway.com/> P Pensez à l'environnement avant d'imprimer.

1 0

[libvirt-users] unable to dissect libvirt rpc packets using wireshark plugin
by gowrishankar 20 Jan '16

20 Jan '16

Hi, I am trying libvirt plugin in wireshark to dissect RPC payload in TCP, but finding dissector code not really working. My env is Fedora core 21 (x86_64) and installed packages are as follow: wireshark-1.12.6-1.fc21.x86_64 libvirt-wireshark-1.2.9.3-2.fc21.x86_64 Earlier, just after installation, I noticed libvirt.so available only in /usr/lib64/wireshark/plugins/1.12.5/ . Wireshark could not load libvirt plugin. So, I copied above .so into 1.12.6/ under same plugins folder, following it wireshark could list libvirt as supported protocol. tshark -G protocols | grep libvirt Libvirt libvirt libvirt However, on checking with some pcaps which has libvirt RPC calls captured on wire, wireshark does not list libvirt RPC packets, as I search for "libvirt" protocol in pcap. Have anyone experienced this before or if you have any pointer that I could check in my env, that would be very helpful. -- Regards, Gowrishankar M

2 6

[libvirt-users] Trouble with volume permissions
by Frank Laszlo 20 Jan '16

20 Jan '16

I'm working on an integration that uses QCOW2 volumes with backing stores. I have qemu configured to run as the "qemu" user, which seems to be fine. I now need to make sure my volumes are readable by this user. However, it seems the volume being created is owned by root, instead of "qemu" as I'd expect it to be. I tried various things and was unable to get it to create the file as the correct user. I've supplied a link to the XML I'm using to create the volumes. If anyone has any ideas it would be massively helpful. Thanks Link to XML: http://pastebin.com/kbg9n6kW -- Frank Laszlo Nexcess - Beyond Hosting 21700 Melrose Ave. Southfield, MI. 48075 Cell: +1.248.602.6123 Phone: +1.866.639.2377 Fax: +1.248.281.0473

2 1

[libvirt-users] guest memory monitor
by frankpersist＠gmail.com 20 Jan '16

20 Jan '16

Hi, everyone, My hypervisor is kvm, when I use libvirt-java api to get guest vm memory information, I found what libvirt can not get real memory(within the vm)statistic, libvirt can only get information from .xml file(such as vmName.xml <currentMemory> lable) in host. My question is, does libvirt support memory monitor(within vm)? if it's true, how to get. if false, what is the difficulty? Thank you. frankpersist(a)gmail.com

2 1

[libvirt-users] how to setup a watchdog?
by lejeczek 14 Jan '16

14 Jan '16

hi everybody I'm testing Qemu's watchdog. My understanding was that hardware (here qemu's watchdog) would take action(cold reboot) the system if there is no ping from the OS watchdog, so I thought stopping watchdog service in VM should be a quick test. I have this in the guest: <watchdog model='i6300esb' action='reset'> <address type='pci' domain='0x0000' bus='0x00' slot='0x08' function='0x0'/> </watchdog> and I see /dev/watchdog in my guest. Yet nothing happens. I must be missing something, an expert said it's config problem? many thanks

1 0

[libvirt-users] libvirt + ceph rbd will hang
by 王李明 14 Jan '16

14 Jan '16

hi all： I use openstack icehouse and libvirt 0.10 version use the quem+ceph to save vm disk when I do serval operation for example migrate vm、snapshot vm the libvirtd will be hang I guess whether the ceph rbd will cause this error anyone can help me? Wang Liming

1 0

[libvirt-users] virsh attach-device : Bus 'pci.0' does not support hotplugging.
by Keyur Bhalerao 14 Jan '16

14 Jan '16

Hi , Trying to attach a disk to existing VM using virsh attach-device - XML <disk type='file' device='disk'> <driver name='qemu' type='qcow2' cache='none' io='native'/> <source file='/var/lib/libvirt/images/guest.qcow2'/> <target dev='vdc' bus='virtio'/> </disk> After executing the command I am getting - [root@cent7 ~]# virsh attach-device lib-virt-man-001 newDisk.xml error: Failed to attach device from newDisk.xml error: internal error: unable to execute QEMU command 'device_add': Bus 'pci.0' does not support hotplugging. [root@cent7 ~]# virsh version Compiled against library: libvirt 1.2.17 Using library: libvirt 1.2.17 Using API: QEMU 1.2.17 Running hypervisor: QEMU 1.5.3 While searching on this issue found that some have suggested to install following - modprobe acpiphp modprobe pci_hotplug I am not sure if this is QEMU issue or libvirt issue. Any help in this would be appreciated. Thanks, Keyur Bhalerao

1 0

[libvirt-users] RBD snapshots
by Emmanuel Lacour 13 Jan '16

13 Jan '16

Dear libvirt users, I'm stuck on this subject and would appreciate a working example as this looks a supported feature :) After many tries and packages backports, here is my current setup: qemu 2.5 libvirt 3.0 ceph hammer on Debian jessie. Here is the relevent domain xml parts: <disk type='network' device='disk' snapshot='internal'> <driver name='qemu' type='raw'/> <auth username='libvirt'> <secret type='ceph' uuid='xxxxxxxxxxxxxxxx'/> </auth> <source protocol='rbd' name='libvirt-pool/vm-test'> <host name='192.168.253.1' port='6789'/> <host name='192.168.253.3' port='6789'/> <host name='192.168.253.254' port='6789'/> <snapshot name='backup'/> </source> <backingStore/> <target dev='vda' bus='virtio'/> <alias name='virtio-disk0'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x06' function='0x0'/> </disk> After several tries, I discovered the snapshot option in source tag, and that I need to pre-create this snapshot using rbd so the vm can start. But then I tried: # virsh snapshot-create-as vm-test error: unsupported configuration: internal snapshot for disk vda unsupported for storage type raw # virsh snapshot-create-as vm-test snap1 --memspec file=/tmp/vm-test.mem --diskspec vda,snapshot=internal error: unsupported configuration: internal snapshot for disk vda unsupported for storage type raw # virsh snapshot-create-as vm-test snap1 --disk-only --atomic --diskspec vda,snapshot=internal error: unsupported configuration: active qemu domains require external disk snapshots; disk vda requested internal and some other test, without any success:( What I just would like to achieve is to do a consistent disk snapshot of the VM without VM shutdown (suspend is ok) and using virsh as this seems supported (I can manually run rbd snap create, I know). And if possible whol mem+disk consistent snapshot without downtime :) Any hint to make this working, as anyone out ther having succes in such process?

1 0