- Users - libvirt List Archives

[libvirt-users] Revert to internal snapshot - XML error: Non-empty feature list specified without CPU model
by Kashyap Chamarthy 30 Oct '14

30 Oct '14

Versions: $ rpm -q libvirt-daemon-kvm qemu-system-x86 libvirt-daemon-kvm-1.2.9-3.fc21.x86_64 qemu-system-x86-2.1.2-4.fc21.x86_64 When I attempt to revert to an internal snapshot: $ virsh snapshot-list node1 Name Creation Time State ------------------------------------------------------------ snap1 2014-06-03 14:07:28 -0400 running $ virsh snapshot-revert node1 snap1 error: XML error: Non-empty feature list specified without CPU model Contextual guest XML: $ virsh dumpxml node1 | grep features -A3 <features> <acpi/> <apic/> <pae/> </features> <cpu mode='host-passthrough'> </cpu> <clock offset='utc'> I see it's thrown from here, src/conf/cpu_conf.c, from function: virCPUDefParseXML:371 [. . .] 367 if (n > 0) { 368 if (!def->model && def->mode != VIR_CPU_MODE_HOST_MODEL) { 369 virReportError(VIR_ERR_XML_ERROR, "%s", 370 _("Non-empty feature list specified without " 371 "CPU model")); 372 goto error; 373 } [. . .] I can manually revert this snapshot via `qemu-img`, but wondering if this error from libvirt expected in this case? -- /kashyap

1 1

[libvirt-users] libvirt with VirtualBox - possible to specify path to snapshot folder in domain.xml?
by fdytr ytruy 30 Oct '14

30 Oct '14

Hi! I'm using libvirt withVirtual Box. I have installed libvirt 0.10.2 and Virtual Box 4.1.Is possible to spesify the pat to the folder I want my differencing image created in in my domain.xml file? When I define my vm (virsh domain.xml) with my domain.xml and use the readonly tag, Virual Box creates a differencing image (default in users Vitual Box VMs folder). I want to specify in my domain.xml file what folder this differencing image should be created in (not the default folder). Is this possible? LL

2 1

[libvirt-users] Can't specify filename.xml when using virsh dumpxml
by Allen Qiu 29 Oct '14

29 Oct '14

Hi all, Do you have any idea about this error? virsh # dumpxml csg01win7test > domain.xml error: unexpected data '>' Regards, Allen 2014-10-28 Allen Qiu

2 1

[libvirt-users] problem with libvirt and ovs bridge
by Anna Giannakou 28 Oct '14

28 Oct '14

Hello, I have created an ovs bridge on which i have attached a port. I would like to connect my vm to that port,so i have created an xml defining the network. The xml is: <network> <name>ovs-snort</name> <forward mode='bridge'/> <bridge name='snort'/> <virtualport type='openvswitch'/> </network> but when i do virsh net-define /tmp/ovsnetwork.xml i get the error error: XML error: unexpected virtualport type -1 My libvirt version is 1.2.9 Can you guide me through that?

3 2

Re: [libvirt-users] problem with libvirt and ovs bridge
by Kashyap Chamarthy 28 Oct '14

28 Oct '14

[Please keep such technical discussions on list, so others can correct me or might have better ideas.] On Tue, Oct 28, 2014 at 06:08:02PM +0100, Anna Giannakou wrote: > Thank you for the quick reply. I tried to remove the <virtualport > type='openvswitch'/> line from the network xml and now the network is > created. However when i try to launch the vm with virt-install i get > again the error Unable to add bridge snort port vnet0: Operation not > supported which was what led me to upgrade my libvirt version > > Any other ideas? Sorry, I don't have an immediate test env to try this, maybe you can examine your debug logs to see if you find more meaningful clues log_level = 1 log_outputs="1:file:/var/tmp/libvirtd.log" in /etc/libvirt/libvirtd.conf Restart libvirtd. (And, don't forget to turn it off, once you finish your debugging - it'll fill your disk space pretty quickly, otherwise.) -- /kashyap > ----- Original Message ----- > > From: "Kashyap Chamarthy" <kchamart(a)redhat.com> > > To: "Anna Giannakou" <anna.giannakou(a)inria.fr> > > Cc: libvirt-users(a)redhat.com > > Sent: Tuesday, 28 October, 2014 5:46:46 PM > > Subject: Re: [libvirt-users] problem with libvirt and ovs bridge > > > > On Tue, Oct 28, 2014 at 05:32:58PM +0100, Anna Giannakou wrote: > > > Hello, > > > I have created an ovs bridge on which i have attached a port. I would like > > > to connect my vm to that port,so i have created an xml defining the > > > network. > > > The xml is: > > > <network> > > > <name>ovs-snort</name> > > > <forward mode='bridge'/> > > > <bridge name='snort'/> > > > <virtualport type='openvswitch'/> > > > </network> > > > but when i do virsh net-define /tmp/ovsnetwork.xml i get the error > > > error: XML error: unexpected virtualport type -1 > > > My libvirt version is 1.2.9 > > > Can you guide me through that? > > > > > > > Hmm, I haven't tested this lately, but here's some old notes on Libvirt > > and Open vSwitch: > > > > http://kashyapc.com/2013/07/13/configuring-libvirt-guests-with-an-open-vswi… > > > > And a related thread from libvirt users list: > > > > https://www.redhat.com/archives/libvirt-users/2013-April/msg00044.html > > > > > > -- > > /kashyap > >

1 0

[libvirt-users] wrong time in guest logs
by Lentes, Bernd 28 Oct '14

28 Oct '14

Hi, i have several vm's running on KVM hosts. Recently i found out that the time in the log-files of the guests (/var/log/libvirt/qemu/xxx.log) is wrong. The time on the guest itself is right, just the time in the log-files is one hour back. Windows and linux vm's are affected. Also on another host the same. Changing the clock offset does not have any influence, still the wrong time in the log file. Any ideas ? Bernd -- Bernd Lentes Systemadministration Institut für Entwicklungsgenetik Gebäude 35.34 - Raum 208 HelmholtzZentrum münchen bernd.lentes(a)helmholtz-muenchen.de phone: +49 89 3187 1241 fax: +49 89 3187 2294 http://www.helmholtz-muenchen.de/idg Die Freiheit wird nicht durch weniger Freiheit verteidigt Helmholtz Zentrum München Deutsches Forschungszentrum für Gesundheit und Umwelt (GmbH) Ingolstädter Landstr. 1 85764 Neuherberg www.helmholtz-muenchen.de Aufsichtsratsvorsitzende: MinDir´in Bärbel Brumme-Bothe Geschäftsführer: Prof. Dr. Günther Wess, Dr. Nikolaus Blum, Dr. Alfons Enhsen Registergericht: Amtsgericht München HRB 6466 USt-IdNr: DE 129521671

3 3

[libvirt-users] Generate (vCPU pinning) from host NUMA configuration doesn't act accordingly
by Allen Qiu 28 Oct '14

28 Oct '14

Hi all, It seems no matter how many vCPU I allocated to the VM, the auto-generated vCPU pinning configuration won't cover more than 8 CPUs. Is that normal? Otherwise what should I do? Will vCPU pinning by virsh be more effective? Regards, Allen 2014-10-28 Allen Qiu

1 0

[libvirt-users] Will over-allocation of vCPUs be counterproductive to VM Win7 Ultimate?
by Allen Qiu 28 Oct '14

28 Oct '14

Hi all, Will over-allocation of vCPUs be counterproductive to VM Win7 Ultimate? I allocated 32 (out of 64 physical cores in 8 NUMA cells, and 128G out of 2T physical RAM) to the VM. But the boot process was extremly slow, actually the desktop screen never showed up then the VM just reboot automatically. So is there anything I can do to solve this problem or the truth is Win7 by design just cannot handle that mount of CPUs even with Win7 Ultimate? Regards, Allen 2014-10-28 Allen Qiu

1 0

[libvirt-users] an error caused by lxc-enter-namespace command
by WANG Cheng D 28 Oct '14

28 Oct '14

Dear all, On my last post "Why a "core dump" error occurs when a thread exits in container but it's correct when running in native linux" might be misleading. I did further test and the steps are described below. 1) First, the container is started (we have the first terminal window) 2) Use virsh lxc-enter-namespace command to open another terminal (we have the second terminal) 3) Execute "exit" in the first terminal to quit container 4) Then an error is reported by the second terminal "internal error; child process (XXX) unexpected fatal signal 9" and the terminal is frozen then. How I solved this problems? Best regards, Cheng

1 0

[libvirt-users] Why a "core dump" error occurs when a thread exits in container but it's correct when running in native linux
by WANG Cheng D 28 Oct '14

28 Oct '14

Dear all, When I run my application in a container, sometimes the thread can exit correctly, sometime a "core dump" error occurs when it exits. But the application always exits correctly when it runs in a native fedora. Malloc() and free() functions are used in my application. I wonder if this is the reason for the core dump error. Is the thread exit event treated differently by the container and native OS? Or if I need to tune some options for memory in container's XML configuration file? Best wishes, Cheng

1 0