April 2011 - Users - libvirt List Archives

[libvirt-users] How to providing virtual USB disk (usb-storage for QEMU) devices to guests?
by oPKo o 26 Apr '11

26 Apr '11

Hi, I am trying to provide virtual USB disk devices to guests but without success. With QEMU is simple to plug the virtual USB disk with the command: (qemu) drive_add 0 id=my_usb_disk,if=none,file=usbdisk.img OK (qemu) device_add usb-storage,id=my_usb_disk,drive=my_usb_dis I can do this with libvirt? PK PS: sorry for my poor english

1 0

[libvirt-users] problem with qemu
by Miroslav Ludvik 25 Apr '11

25 Apr '11

Hi All, I use Ubuntu server 10.04 LTS as virtualization platform. Actually running kernel 2.6.32-31-server #61-Ubuntu S root@jupiter:~# uname -a Linux jupiter 2.6.32-31-server #61-Ubuntu SMP Fri Apr 8 19:44:42 UTC 2011 x86_64 GNU/Linux We have on running virtual root@jupiter:~# virsh list --all Id Name State ---------------------------------- 1 kvmtik.4safety.cz running I run script for create virtual system wrote root@jupiter:~# make-hestia.4safety.cz.sh 2011-04-24 21:13:25,613 ERROR : ubuntu-kvm already exists Traceback (most recent call last): File "/usr/bin/vmbuilder", line 24, in <module> cli.main() File "/usr/lib/python2.6/dist-packages/VMBuilder/contrib/cli.py", line 81, in main raise VMBuilderUserError('%s already exists' % destdir) VMBuilder.exception.VMBuilderUserError: ubuntu-kvm already exists /usr/local/bin/make-hestia.4safety.cz.sh: line 29: --libvirt=qemu:///system: No such file or directory I try restart libvirt and system wrote to syslog root@jupiter:~# /etc/init.d/libvirt-bin restart Rather than invoking init scripts through /etc/init.d, use the service(8) utility, e.g. service libvirt-bin restart Since the script you are attempting to invoke has been converted to an Upstart job, you may also use the restart(8) utility, e.g. restart libvirt-bin libvirt-bin start/running, process 25202 root@jupiter:~# tail -f /var/log/syslog Apr 24 20:25:01 jupiter CRON[25031]: (root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1) Apr 24 20:35:01 jupiter CRON[25036]: (root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1) Apr 24 20:45:01 jupiter CRON[25042]: (root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1) Apr 24 20:55:01 jupiter CRON[25048]: (root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1) Apr 24 21:05:01 jupiter CRON[25152]: (root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1) Apr 24 21:15:01 jupiter CRON[25183]: (root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1) Apr 24 21:17:01 jupiter CRON[25189]: (root) CMD ( cd / && run-parts --report /etc/cron.hourly) Apr 24 21:17:58 jupiter libvirtd: 21:17:58.089: warning : qemudDispatchSignalEvent:385 : Shutting down on signal 15 Apr 24 21:17:58 jupiter libvirtd: 21:17:58.617: error : udevStrToLong_ui:73 : Failed to convert '008' to unsigned int#012 Apr 24 21:17:59 jupiter kernel: [106086.771154] lo: Disabled Privacy Extensions Than for your help. Mirek

2 1

[libvirt-users] Media autoremove
by Nikita A Menkovich 25 Apr '11

25 Apr '11

I have an iso, than automate freebsd install, and after installation it reboot VM. I want to remove cdrom drive with ISO from VM config after reboot? How I could do this? -- Nikita A Menkovich http://libc6.org/ JID: menkovich(a)gmail.com

2 1

[libvirt-users] Setup Mulitple VM's
by ssaleh＠nrln.us 25 Apr '11

25 Apr '11

I have a small public IP block (/29) I'd like to populate with VM's from a single physical box running CentOS/5. I'd like to get some thoughts on the best way to setup networking to support this. I'm OK with setting up the virtual hosts. Any advice would be appreciated. TIA...

2 1

[libvirt-users] vvirsh memtune
by Zvi Dubitzky 24 Apr '11

24 Apr '11

Hi Does anybody use the 'virsh memtune' command and can he say what combination of libvirt version and libcgroup does he use thanks Zvi Dubitzky Email:dubi@il.ibm.com

1 0

[libvirt-users] 2 hosts, mirrored VM storage, avoiding conflicting VM instances
by Whit Blauvelt 24 Apr '11

24 Apr '11

Hi, Using libvirt for KVM, experimenting with live migration with the VM on storage shared in a mirrored GlusterFS or an active-active DRBD setup where QEMU/KVM runs on the hosts on both sides of the mirror, the live migration works great. What concerns me is finding a way if possible to prevent a situation where a particular VM is running on both hosts at once. Is there any way within libvirt in this sort of setup to start a VM on mirrored storage from a KVM host on one side of the mirror _only if_ the same VM isn't already running on the other side of the mirror? In a scripting context I can see how to do that check before starting, but not how to prevent virsh or virt-manager from being used to create this bad condition when they're being used directly. Is there some way to set up protection from this within virsh and virt-manager? Or am I into feature request territory? Or is there some other layer at which this can be accomplished? In active-passive DRBD this wouldn't be a danger, since only the active system can see the storage to start the VM. But that wouldn't support live migration. Part of the danger can be avoided by using stonith in failover, but it would be nice to find a more fine-grained, subtle approach. Best, Whit

1 0

[libvirt-users] PCI Pass through of atheros card
by Varun Ganesh 24 Apr '11

24 Apr '11

Hi, I am trying to make a PCI passthrough of the atheros card to a VM running over KVM. am following the steps given at http://docs.fedoraproject.org/en-US/Fedora/13/html/Virtualization_Guide/sec…. but am getting the following errors, 1. Error starting domain: operation failed: failed to retrieve chardev info in qemu with 'info chaqdev' 2. Details: 3. Error starting domain: operation failed: failed to retrieve chardev info in qemu with 'info chaqdev' 4. 5. Traceback (most recent call last): 6. File "/usr/share/virt-manager/virtManager/asyncjob.py", line 45, in cb_wrapper 7. callback(asyncjob, *args, **kwargs) 8. File "/usr/share/virt-manager/virtManager/engine.py", line 959, in asyncfunc 9. vm.startup() 10. File "/usr/share/virt-manager/virtManager/domain.py", line 1114, in startup 11. self._backend.create() 12. File "/usr/lib/python2.7/site-packages/libvirt.py", line 333, in create 13. if ret == -1: raise libvirtError ('virDomainCreate() failed', dom=self) 14. libvirtError: operation failed: failed to retrieve chardev info in qemu with 'info chaqdev' I am using a Dell Optiplex XE system which has VT-d enabled and Atheros AR5001X+ Wireless Network Adapter. How do i rectify this. Thanks, sincerely, Varun Ganesh

1 0

[libvirt-users] domain not present after shutdown
by David White 22 Apr '11

22 Apr '11

I hope this is simple I have one kvm host that lists all my domains even after shut down. 'virsh list --all' shows machines in 'running' and 'shut off' states. If I copy those over to another machine and it only shows the virt machines in running state. vmhost2:~$ virsh list --all Id Name State ---------------------------------- 4 mybox4 running vmhost2:~$ virsh version Compiled against library: libvir 0.8.3 Using library: libvir 0.8.3 Using API: QEMU 0.8.3 Running hypervisor: QEMU 0.12.5 vmhost1:~$ virsh list --all Id Name State ---------------------------------- - mybox3 shut off 2 mybox4 running vmhost1:~$ virsh version Compiled against library: libvir 0.7.5 Using library: libvir 0.7.5 Using API: QEMU 0.7.5 Running hypervisor: QEMU 0.12.3 question -- why doesn't mybox3 show up on vmhost2? where should I start looking? hope this isn't too open ended...

3 6

[libvirt-users] [PATCHv11 0/6] libvirt/qemu - persistent modification of devices
by KAMEZAWA Hiroyuki 22 Apr '11

22 Apr '11

Here is v11. Fixed comments/bugs and updated against the latest libvirt.git. Changes v10->v11: - fixed comments on each patches - fixed cgroup handling in patch 3. - fixed MODIFY_CURRENT handling in patch 4. most of diff comes from refactoring qemu/qemu_driver.c -- conf/domain_conf.c | 40 ++ conf/domain_conf.h | 5 libvirt_private.syms | 3 qemu/qemu_driver.c | 727 +++++++++++++++++++++++++++++++-------------------- 4 files changed, 500 insertions(+), 275 deletions(-) Thanks, -Kame

2 9

[libvirt-users] No IOMMU found. Unable to assign device
by Marco van Beek 22 Apr '11

22 Apr '11

Hi, I am playing around with PCI passthrough and came across some posts that said you could not do PCI passthrough unless you had IOMMU hardware. which it would appear I don't Is this the case? Am I flogging a dead horse? Thanks, Marco -- Marco van Beek ========================================== Supporting Role Ltd. Grove Park Studios, 188-192 Sutton Court Rd, London, W4 3HR ========================================== T: 0870 757 2824 F: 0870 757 2826 M: 0788 770 3604 E: mvanbeek(a)supporting-role.co.uk W: http://www.supporting-role.co.uk/ ========================================== Company Registered in England No. 03658568 ==========================================

3 2