- Users - libvirt List Archives

[libvirt-users] LXC + USB passthrough = Operation not permitted
by Filip Maj 18 Apr '14

18 Apr '14

Hi! First post, kind of a noobie. I've been working with LXC and libvirt for a few months now. Trying to do some interesting things with containers and Android devices :D I'm running ubuntu 13.10 with LXC 1.0.1 and tried both libvirt 1.1.1 and 1.2.2 (backported from ubuntu-trusty), but with either version of libvirt am getting issues as soon as I try to get access to USB devices inside the container. Relevant versions of stuff: $ dpkg -l | grep -i lxc ii liblxc1 1.0.1-0ubuntu1~ubuntu13.10.1~ppa1 i386 Linux Containers userspace tools (library) ii lxc 1.0.1-0ubuntu1~ubuntu13.10.1~ppa1 i386 Linux Containers userspace tools ii lxc-templates 1.0.1-0ubuntu1~ubuntu13.10.1~ppa1 i386 Linux Containers userspace tools (templates) ii python3-lxc 1.0.1-0ubuntu1~ubuntu13.10.1~ppa1 i386 Linux Containers userspace tools (Python 3.x bindings) $ dpkg -l | grep libvirt ii libvirt-bin 1.2.2-0ubuntu12 i386 programs for the libvirt library ii libvirt0 1.2.2-0ubuntu12 i386 library for interfacing with different virtualization systems ii python-libvirt 1.2.2-0ubuntu1 i386 libvirt Python bindings Here's my entire domain definition: <domain type='lxc'> <name>oshi32134</name> <uuid>xxxxx</uuid> <memory unit='KiB'>3145728</memory> <currentMemory unit='KiB'>3145728</currentMemory> <vcpu placement='static'>1</vcpu> <resource> <partition>/machine</partition> </resource> <os> <type arch='i686'>exe</type> <init>/sbin/init</init> </os> <clock offset='utc'/> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>destroy</on_crash> <devices> <emulator>/usr/lib/libvirt/libvirt_lxc</emulator> <filesystem type='mount' accessmode='passthrough'> <source dir='/some/valid/filesystem/location'/> <target dir='/'/> </filesystem> <filesystem type='mount' accessmode='passthrough'> <source dir='/another/valid/filesystem/location'/> <target dir='/mnt/android'/> </filesystem> <interface type='bridge'> <mac address='xx:xx:xx:xx:xx:xx'/> <source bridge='br1'/> </interface> <console type='pty'> <target type='lxc' port='0'/> </console> <hostdev mode='capabilities' type='misc'> <source> <char>/dev/kvm</char> </source> </hostdev> <hostdev mode='subsystem' type='usb' managed='yes'> <source> <vendor id='0x04e8'/> <product id='0x6860'/> </source> </hostdev> </devices> </domain> Everything worked fine until I added the USB <hostdev> element. I'm essentially trying to get access to a physical Android device connected to the host from inside a container. When I go to start the container, I get an error about Operation not permitted. Here's the relevant bits from /var/log/libvirt/lxc/machine.log: 2014-04-11 22:46:40.491+0000: starting up PATH=/usr/local/sbin:/usr/local/bin:/usr/bin:/usr/sbin:/sbin:/bin LIBVIRT_DEBUG=3 LIBVIRT_LOG_OUTPUTS=3:stderr /usr/lib/libvirt/libvirt_lxc --name oshi32134 --console 24 --security=none --handshake 27 --background --veth vnet1 2014-04-11 22:46:40.597+0000: 685: info : libvirt version: 1.2.2 2014-04-11 22:46:40.597+0000: 685: error : virLXCControllerSetupHostdevSubsysUSB:1390 : Unable to create device //var/run/libvirt/lxc/oshi32134.dev/bus/usb//002//003: Operation not permitted Unable to create device //var/run/libvirt/lxc/oshi32134.dev/bus/usb//002//003: Operation not permitted I get the same output above when I use libvirt 1.1.1. Tracing down through the libvirt code it looks like it fails when calling `mknod`. It's clear that libvirt converts the product/vendor ids I specified in the domain definition file to the appropriate USB bus (/dev/bus/usb/002/003, verified by comparing against output of `usb-devices`). In a parallel effort, a colleague of mine set up a vanilla LXContainer on a new install of the same version of ubuntu, and successfully was able to communicate with the Android device from a container. The only configuration tweaks he made was make sure to whitelist the cgroup with the appropriate major/minor device number (associated with the Android device) inside the LXC configuration file. From there the container was able to appropriately create the device file as you attach/detach the Android device. Hopefully someone can shed some light, and appreciate your patience with me as I am learning a lot of this stuff as I go :) Thanks in advance for any help! Fil

2 7

[libvirt-users] virsh blockcommit hangs at 100%
by zyro 18 Apr '14

18 Apr '14

hi, i am trying to blockcommit a external qcow2 snapshot back into its raw backing file like: sudo virsh blockcommit myDomainName vda --wait --verbose the output just shows: Block Commit: [100 %] and stays like that (i waited about an hour or so...). No "Commit complete" message. - explicitly passing --base and/or --top arguments makes no difference. - did not find a related log entry in syslog and consorts env: - ubuntu server 14.04 x64 - libvirt-bin 1.2.2-0ubuntu13 i doubt that it just takes hours to complete after "100%", right? any ideas? kind of a dead-end for me :/ thanks, zyro

2 1

[libvirt-users] virtual subnet
by Lucio Crusca 18 Apr '14

18 Apr '14

Hello all, I need to setup a virtual subnet to create a test AD domain (server + clients): guests need to talk to each other, they need to reach the internet through a virtual router with nat (virbr0?), broadcasts must not reach the host nor the physical network (because of conflicts with the existing real AD domain) and libvirt must NOT offer a dhcp service inside the virtual subnet. What kind of networking model should I look at?

2 2

[libvirt-users] What's the meaning of sub-element <ip address='X.X.X.X'> in <interface type='bridge'> of domain xml?
by Jianwei Hu 18 Apr '14

18 Apr '14

Hi guys, I saw this sub-element in http://libvirt.org/firewall.html, there is some confusion, what's the meaning of sub-element <ip address='X.X.X.X'> in <interface type='bridge'> of domain xml? The detail <interface> in domain xml as below: <interface type='bridge'> <mac address='52:54:00:56:44:32'/> <source bridge='br1'/> <ip address='10.X.X.X'/> <===it's my question <target dev='vnet0'/> <model type='virtio'/> <filterref filter='clean-traffic'/> </interface> Is it static IP(specified by customer) in guest OS? or a IP of interface(e.g. eno1) in "br1" on host machine? what's definition about it? [root@localhost src]# brctl show bridge name bridge id STP enabled interfaces br1 8000.24be051881ce no eno1 If you know how to use it, please show me a detail scenario. Thanks, Jianwei

3 2

[libvirt-users] KVM guests unusable after install
by Jay Vyas 17 Apr '14

17 Apr '14

I'm finding that my KVM images are not bootable after i reboot my host, and i've detailed my problem here.... http://serverfault.com/questions/589806/kvm-machines-unbootable-after-host-… Any thoughts on why? Here is a repaste of the problem: **************************************************************************************** Im finding that my KVM guests are unusable after i reboot my host. Specifically, I always get this error: Booting from hard disk Boot failed: not a bootable disk What causes this error? I've tried toggling the VM disk type between qcow2 and raw, because a few articles online mentioned that this can happen if virt doesnt know the type of image to load, but that did not help. ** some more details ** I create My VMs from source using a command somewhat like this: base="http://mirror.pnl.gov/fedora/linux/releases/20/Fedora/x86_64/os/" nohup sudo virt-install --hvm --name $vm_name$i --ram 4000 \ --disk path=/VirtualMachines/$vm_name$i,size=30 \ --location $base -x "ks=http://xxx.os21.ks$kx"& My virsh file looks like this: 1 <domain type='kvm'> 2 <name>testFriApr11092714EDT2014_1</name> 3 <uuid>d4c86892-fb15-04ed-f4ab-7e3c76974ffc</uuid> 4 <memory unit='KiB'>4096000</memory> 5 <currentMemory unit='KiB'>4096000</currentMemory> 6 <vcpu placement='static'>1</vcpu> 7 <os> 8 <type arch='x86_64' machine='rhel6.4.0'>hvm</type> 9 <boot dev='hd'/> 10 </os> 11 <features> 12 <acpi/> 13 <apic/> 14 <pae/> 15 </features> 16 <clock offset='utc'/> 17 <on_poweroff>destroy</on_poweroff> 18 <on_reboot>restart</on_reboot> 19 <on_crash>restart</on_crash> 20 <devices> 21 <emulator>/usr/libexec/qemu-kvm</emulator> 22 <disk type='file' device='disk'> 23 <driver name='qemu' type='raw' cache='none'/> 24 <source file='/VirtualMachines/testFriApr11092714EDT2014_1'/> 25 <target dev='hda' bus='ide'/> 26 <address type='drive' controller='0' bus='0' target='0' unit='0'/> 27 </disk> 28 <controller type='usb' index='0'> 29 <address type='pci' domain='0x0000' bus='0x00' slot='0x01' function='0x2'/> 30 </controller> 31 <controller type='ide' index='0'> 32 <address type='pci' domain='0x0000' bus='0x00' slot='0x01' function='0x1'/> 33 </controller> 34 <interface type='network'> -- Jay Vyas http://jayunit100.blogspot.com

1 0

[libvirt-users] "failed to send audit message operation not permitted"
by Pasquale Dir 17 Apr '14

17 Apr '14

Hello, I have a remote ubuntu server on which I have a kvm hypervisor installed. Now, on my ubuntu desktop, I create a new connection with virt-manager and start a remote guest. The guest successfully starts, however I can't see it in the console and, on client side, I get that remote user closed it. Looking at the /var/log/libvirtd on the server side I get that error: "failed to send audit message operation not permitted" I tried to call "virsh console myDomain" on the remote server and what I get is that the console freezes. The log is always the same, operation not permitted. Here's the complete one: 2014-04-17 18:26:26.532+0000: 2985: warning : virAuditSend:135 : Failed to send audit message virt=kvm op=stop reason=destroyed vm="dhcpVPN" uuid=929bac48-b87c-9f6f-156a-5c066bf36085 vm-pid=-1: Operation not permitted Where the problem could be? Pasquale

1 0

[libvirt-users] libvirt with glusterfs problem
by lyz_pro 17 Apr '14

17 Apr '14

Hi I have two node: hcg3, hcg4 which run glusterfs Also run libvirtd on each node. The problem cames when I use `virsh attach-device` to attach a glusterfs type disk . My device xml like: <disk type="network" device="disk"> <driver name="qemu" type="raw" cache="none"/> <source protocol="gluster" name="h2312v3/volume-257e07bd-9e80-4d06-8d6d-5f83979de1fd"> <host name="hcg4"/> </source> <target bus="virtio" dev="vdb"/> <serial>257e07bd-9e80-4d06-8d6d-5f83979de1fd</serial> </disk> when I attach a gluster disk to a domain running on hcg4. the above xml work great. BUT when I try to change the tag `host` name attribute to hcg3, the attach action will fail with output: error: Requested operation is not valid: domain is not running Also: I can use `qemu-kvm` command to start vm successful with gluster://hcg3 or gluster://hcg4. I have stop both nodes' iptables. and my libvirt and qemu version is : Compiled against library: libvirt 1.1.4 Using library: libvirt 1.1.4 Using API: QEMU 1.1.4 Running hypervisor: QEMU 1.6.2 glusterfs-3.4.1-3.el6.x86_64 -------------- lyz_pro 2014-04-11

2 2

[libvirt-users] Snapshots
by lukass.vacek＠seznam.cz 17 Apr '14

17 Apr '14

Hello, I would like to ask what I am doing wrong and what is the best approach to resolve this situation: First of all I start virtual machine by virDomainCreateXML. In my application I use libvncclient library that sends mouse events to this running domain. I had some troubles with coordinates of mouse cursor but all problem was solved when I added <input type='tablet' bus='usb'/> to xml configuration. So now I have a running domain and I want to create snapshot and then after some operations I want to restore this snapshot. I use virDomainSave and virDomainRestore that work fine. The only problem is, that after creating and restoring snapshot I have the same problem with coordinates of mouse. I need to create and restore snapshot also with <input type='tablet' bus= 'usb'/> somewhere specified. I also tried methods virDomainSaveFlags with xml configuration string(see below) as parameter *dxml, but method fails with: libvir: Domain Config error : XML error: domainsnapshot.Unable to create snapshot. What should I add to xml configuration? I look at http://libvirt.org/ formatsnapshot.html on example but I dont know what paths I should use in my program. I dont understand much snapshots and similar stuff so could you give me example how could I edit my xml configuration? Or what I should do to make virDomainSave and virDomainRestore work? xml configuration: <domain type='kvm'> <name>Name</name> <uuid>c7a5fdbd-cdaf-9455-926a-d65c16db1810</uuid> <memory unit='KiB'>524288</memory> <currentMemory unit='KiB'>524288</currentMemory> <vcpu>2</vcpu> <os> <type arch='x86_64' machine='pc'>hvm</type> <boot dev='hd'/> </os> <devices> <input type='tablet' bus='usb'/> <emulator>/usr/bin/qemu-system-x86_64</emulator> <disk type='file' device='disk'> <source file='/home/lukas/Desktop/Image.img'/> <target dev='hda'/> </disk> <interface type='network'> <source network='default'/> </interface> <graphics type='vnc' port='5901'> <listen type='address' address='127.0.0.1'/> </graphics> </devices> </domain>

2 1

Re: [libvirt-users] [virt-tools-list] live-migrating a VM whose current host's storage is fubar'ed
by Cole Robinson 16 Apr '14

16 Apr '14

On 04/15/2014 05:16 PM, Digimer wrote: > Hi all, > > For those on the linux-cluster list, this is a similar question to my > post[1] there. > > I recently had a client destroy their node's RAID array (they yanked hard > drives until the array was destroyed). The VMs running on the node continued > to operate for another hour, impressively. Their storage was based on DRBD, so > the VM's storage just started going to/from the other node's storage. > > The problem was when I tried to migrate them off of the damaged node to a > healthy node. Without being able to write to the logs (I assume), the > migration failed. In the end, I had to force-off the damaged host and reboot > the VMs on the backup node. This caused an outage which, I think, could have > been avoided. > > So my question is; Can VMs be migrated if/when the host's FS is > read-only/destroyed? If so, how? > > Thanks! > > 1. https://www.redhat.com/archives/linux-cluster/2014-April/msg00010.html > Probably a better question for libvirt lists, CCd, but I don't think anyone tests a scenario like that so there's bound to be problems. - Cole

2 1

[libvirt-users] LXC vCPU controlling problem
by Xingjun Wang 16 Apr '14

16 Apr '14

Hi,guys! I've been using libvirt-1.1.1 with LXC1.0.0 to do my CS homework.But I met a problem when I tried to set vcpu for a LXC.The /sys/fs/cgroup/cpuset/machine/cpuset.cpus always show 0-3 no matter how many vcpu I defined in the XML setting file. But I found the command "virsh -c lxc:/// domaininfo lxc_name" got the right number as it is defined. What's the problem there? Is it a bug in libvirt-1.1.1? Thank you!

1 0