- Users - libvirt List Archives

[libvirt-users] Managed USB passthrough
by Michael Han 21 Jun '14

21 Jun '14

Does the "managed" attribute of the hostdev element for a USB device make a difference? The Domain XML Format docs has the following sentence for the hostdev element: If managed is omitted or "no", and for USB devices, the user is responsible to call virNodeDeviceDettach (or virsh nodedev-dettach) before starting the guest or hot-plugging the device, and virNodeDeviceReAttach (or virsh nodedev-reattach) after hot-unplug or stopping the guest. Is that still accurate? The device seems to get passed through fine to my LXC container without any dettach/reattach calls. But I'm seeing some other strange USB behaviour in the guest and just want to see if it's at all related to the managed attribute. Cheers.

1 0

[libvirt-users] Access permissions of passed through character devices
by Pickartz, Simon 20 Jun '14

20 Jun '14

Dear all, I am using libvirt in conjunction with LXC. In accordance with http://libvirt.org/formatdomain.html#elementsHostDevCaps I pass an InfiniBand adapter through to the guest container. However, within the container the access permissions are not the same. Within the host they are like this: crw-rw-rw- 1 root root 10, 56 20. Jun 09:40 rdma_cm crw-rw-rw- 1 root root 231, 224 20. Jun 09:41 ucm0 crw-rw-rw- 1 root root 231, 0 20. Jun 09:41 umad0 crw-rw-rw- 1 root root 231, 1 20. Jun 09:41 umad1 crw-rw-rw- 1 root root 231, 192 20. Jun 09:41 uverbs0 Within the guest, only root may read or write from/to the device: crw------- 1 root root 10, 56 20. Jun 09:40 rdma_cm crw------- 1 root root 231, 224 20. Jun 09:41 ucm0 crw------- 1 root root 231, 0 20. Jun 09:41 umad0 crw------- 1 root root 231, 1 20. Jun 09:41 umad1 crw------- 1 root root 231, 192 20. Jun 09:41 uverbs0 Do you have any ideas how I can tell lxc via libvirt to change the permissions in such a way that they correspond to those on the host system? Best regards, Simon

1 0

[libvirt-users] converted VMDK disk iamge and Virtio driver
by Paul Raines 20 Jun '14

20 Jun '14

I have a CentOS 5 VM in vmware I wanted to move to work on KVM/libvirt. I used qemu-img to convert it to a qcow2 image, loaded virt-manager and created a new VM pointed at that qcow2 image. The boot paniced when it tried to mount the root filesystem also complaining about not being able to remount the swap partition. I played around and was able to get it to boot by changing the Virtual disk bus to IDE from Virtio. Searched about Virtio and I got to http://wiki.libvirt.org/page/Virtio and ran the mkinitrd mentioned to force loading of the virtio_pci and virtio_blk drivers. But upon shutting down, configuring the bus back to Virtio and booting it panics in the exact same way. I tried again after adding virtio_blk into /etc/modprobe.conf as scsi_hostadapter and rerunning mkinitrd but it still fails. Any clues what I am missing? --------------------------------------------------------------- Paul Raines http://help.nmr.mgh.harvard.edu MGH/MIT/HMS Athinoula A. Martinos Center for Biomedical Imaging 149 (2301) 13th Street Charlestown, MA 02129 USA The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.

2 3

[libvirt-users] bridge could not be initialized
by abhishek jain 20 Jun '14

20 Jun '14

Hi I'm trying to run VM using libvirt.xml file and getting following error... virsh start instance-00000003 error: Failed to start domain instance-00000003 error: internal error Process exited while reading console log output: failed to launch bridge helper kvm: -netdev bridge,br=qbr1f2191ce-38,id=hostnet0: Device 'bridge' could not be initialized Below is the libvirt.xml file.. domain type="kvm"> <uuid>d9f007b9-584f-41c8-a706-e096f354575b</uuid> <name>instance-00000003</name> <memory>524288</memory> <vcpu>1</vcpu> <os> <type>hvm</type> <kernel>/opt/stack/data/nova/instances/d9f007b9-584f-41c8-a706-e096f354575b/kernel</kernel> <initrd>/opt/stack/data/nova/instances/d9f007b9-584f-41c8-a706-e096f354575b/ramdisk</initrd> <cmdline>root=/dev/vda console=tty0 console=ttyS0</cmdline> </os> <features> <acpi/> <apic/> </features> <clock offset="utc"> <timer name="pit" tickpolicy="delay"/> <timer name="rtc" tickpolicy="catchup"/> </clock> <devices> <disk type="file" device="disk"> <driver name="qemu" type="qcow2" cache="none"/> <source file="/opt/stack/data/nova/instances/d9f007b9-584f-41c8-a706-e096f354575b/disk"/> <target bus="virtio" dev="vda"/> </disk> <interface type="bridge"> <mac address="fa:16:3e:26:55:ff"/> <model type="virtio"/> <source bridge="qbr1f2191ce-38"/> <target dev="tap1f2191ce-38"/> <filterref filter="nova-instance-instance-00000003-fa163e2655ff"/> </interface> <serial type="file"> <source path="/opt/stack/data/nova/instances/d9f007b9-584f-41c8-a706-e096f354575b/console.log"/> </serial> <serial type="pty"/> </devices> </domain> Please help regarding this. Thanks Abhishek Jain

3 2

[libvirt-users] dumpxml removes <source> portion from network descritiption
by Thomas Stein 19 Jun '14

19 Jun '14

Hello. I have a strange error with one of my VMs. If i use dumpxml to save a VM description the source part of the network part is missing. This prevents redifining the machine from the dumped xml file: This is how it looks before dumpxml: <interface type='bridge'> <mac address='52:54:00:34:a1:56'/> <source bridge='br406'/> <model type='virtio'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> And afterwards: <interface type='bridge'> <mac address='52:54:00:34:a1:56'/> <target dev='vnet6'/> <model type='virtio'/> <alias name='net0'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> Someone has an idea whats causing this? Is there a special option maybe? thanks and best regards t.

2 2

[libvirt-users] Use keepalive to monitor health of qemu-kvm process
by Abhijeet Rastogi 19 Jun '14

19 Jun '14

Hi everyone, I've had instances where qemu-kvm process dies and become a zombie. At that state, libvirtd reports the instance to be running and I can't restart the VM because libvirtd fails to kill the qemu-kvm process. Finding the root cause of that is a topic for another day. I also see this line in logs: Jun 15 13:36:57 d12566 libvirtd: 10424: warning : virKeepAliveTimerInternal:156 : No response from client 0x1600420 after 5 keepalive messages in 31 seconds Is it possible to monitor the keepAlive thing via command-line ? (as a nagios check). I know that I can poll for that line in the logs but is there a better way? I couldn't find any option in virsh to do a KeepAlive check. Thanks -- Cheers, Abhijeet Rastogi (shadyabhi)

1 0

[libvirt-users] Live migrate failure
by Tuân Tạ Bá 18 Jun '14

18 Jun '14

Hi everyone, I use KVM on Openstack Cloud. And KVM migrated successfull, but not ping. *Source Host SIDE: *"/var/log/libvirt/libvirtd.log" 2014-06-16 05:15:50.599+0000: 2096: warning : qemuMigrationCancelDriveMirror:1383 : Unable to stop block job on drive-virtio-disk0 2014-06-16 05:15:50.601+0000: 2096: *warning : qemuMigrationCancelDriveMirror:1383 : Unable to stop block job on drive-virtio-disk1* *2014-06-16 05:15:50.651+0000: 2096: error : virCommandWait:2348 : internal error: Child process (ovs-vsctl --timeout=5 get Interface tap6977e4a1-cd external_ids:PortData) unexpected exit status 1: ovs-vsctl: no key "PortData" in Interface record "tap6977e4a1-cd" column external_ids* 2014-06-16 05:15:50.651+0000: 2096: error : virNetDevOpenvswitchGetMigrateData:219 : Unable to run command to get OVS port data for interface tap6977e4a1-cd: Operation not permitted 2014-06-16 05:15:50.651+0000: 2096: error : qemuMigrationCookieNetworkAlloc:367 : Unable to run command to get OVS port data for interface tap6977e4a1-cd: Operation not permitted 2014-06-16 05:15:50.651+0000: 2096: warning : qemuMigrationRun:3262 : Unable to encode migration cookie 2014-06-16 05:15:53.321+0000: 2096: warning : virAuditSend:135 : Failed to send audit message virt=kvm op=stop reason=migrated vm="instance-0001b37c" uuid=631ac4c7-8e84-4cf6-ad98-93877cdbe1bd vm-pid=-1: Operation not permitted *Destination Host SIDE:* "/var/log/libvirt/libvirtd.log" ...... 2014-06-16 05:15:41.419+0000: 2229: warning : virAuditSend:135 : Failed to send audit message virt=kvm resrc=cgroup reason=allow vm="instance-000 1b37c" uuid=631ac4c7-8e84-4cf6-ad98-93877cdbe1bd cgroup="/sys/fs/cgroup/devices/machine/instance-0001b37c.libvirt-qemu/" class=major category=pty maj=88 acl=rw: Operation not permitted 2014-06-16 05:15:41.620+0000: 2229: warning : qemuDomainObjEnterMonitorInternal:1175 : This thread seems to be the async job owner; entering moni tor without asking for a nested job is dangerous 2014-06-16 05:15:43.552+0000: 2229: warning : virAuditSend:135 : Failed to send audit message virt=kvm resrc=vcpu reason=start vm="instance-0001b 37c" uuid=631ac4c7-8e84-4cf6-ad98-93877cdbe1bd old-vcpu=0 new-vcpu=2: Operation not permitted 2014-06-16 05:15:43.552+0000: 2229: warning : virAuditSend:135 :* Failed to send audit message *virt=kvm op=start reason=migrated vm="instance-0001 b37c" uuid=631ac4c7-8e84-4cf6-ad98-93877cdbe1bd vm-pid=21063:* Operation not permitted* 1) I'm using libvirt version 1.1.1, libvirt-bin 1.1.1-0ubuntu8.5~cloud0 libvirt0 1.1.1-0ubuntu8.5~cloud0 python-libvirt 1.1.1-0ubuntu8.5~cloud0 2) #cat /etc/libvirt/qemu.conf ... user = "root" group = "root" I searched GOOGLE, René speaks "downgrading to libvirt 1.1.0 and reinstalling everything. They will ok". But I don't think that. How to resolve this error?. Thanks all! -- TuanTaBa

2 2

[libvirt-users] Libvirt 1.2.5/Qemu woes
by andy zerger 17 Jun '14

17 Jun '14

Hi all - I'm dealing with a libvirt 1.2.5 and Qemu-1.7.1 also Qemu-2.0 (all source compile,) which somehow isn't seeing any hypervisors when connected to via virt-manager. I'm sure I must be doing something wrong but cannot figure out what. My basic config and current state here in pastebin: http://pastebin.com/m5M8PfFC Thanks for any input, this has been baffling. Best Regards, Andy

2 2

[libvirt-users] Fedora 19/20 new install :Unable to create vcpu cgroup
by Patrick Chemla 17 Jun '14

17 Jun '14

Hi, After full new install of Fedora 19 or 20, I am getting this when I want to create a new VMs. Impossible de terminer l'installation : « Unable to create vcpu cgroup for fedora20(vcpu: 0): Aucun fichier ou dossier de ce type » Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/asyncjob.py", line 91, in cb_wrapper callback(asyncjob, *args, **kwargs) File "/usr/share/virt-manager/virtManager/create.py", line 1787, in do_install guest.start_install(meter=meter) File "/usr/share/virt-manager/virtinst/guest.py", line 403, in start_install noboot) File "/usr/share/virt-manager/virtinst/guest.py", line 467, in _create_guest dom = self.conn.createLinux(start_xml or final_xml, 0) File "/usr/lib64/python2.7/site-packages/libvirt.py", line 2897, in createLinux if ret is None:raise libvirtError('virDomainCreateLinux() failed', conn=self) libvirtError: Unable to create vcpu cgroup for fedora20(vcpu: 0): Aucun fichier ou dossier de ce type I saw that there is no cpu mounted, only cpuacct. I saw on other server that cpu is a link to cpu,cpuacct But I can't find how to add this in a config file Thanks for help Patrick

2 12

[libvirt-users] libvirt issue while deploying a VM
by Zvi Dubitzky 16 Jun '14

16 Jun '14

Using libvirt under openstack I encounter a problem when launching a VM with multiple partitions (2 G image) , While a single partition can be launched successfully . The launch is done with Openstack. We are using in libvirtd.conf log_level=1 and log_outputs="1:file:/var/log/libvirt/libvirtd.log" we can see that in /var/log/libvirt/libvirtd.log appear the following error message and the VM is finally shutdown : 2014-06-11 14:47:26.166+0000: 12971: error : qemuMonitorIO:614 : internal error End of file from monitor ( file attached below) ( VM shutdown seen here) The host has enough resources( we see no complains) and the XML of the VM is : The libvirt version is: 0.10.2 The qemu-kvm version: 0.12.1.2 >From the internet we could not figure out a solution to the problem Any idea what is causing this error and how to overcome ? Thx Zvi Dubitzky Email:dubi@il.ibm.com

2 3