- Users - libvirt List Archives

Re: [libvirt-users] snapshot of running vm's
by Lentes, Bernd 03 Dec '15

03 Dec '15

Dominique wrote: > -----Original Message----- > From: Dominique Ramaekers > [mailto:dominique.ramaekers@cometal.be] > Sent: Thursday, December 03, 2015 9:46 AM > To: Lentes, Bernd > Subject: RE: snapshot of running vm's > > > > > -----Oorspronkelijk bericht----- > > Van: Lentes, Bernd [mailto:bernd.lentes@helmholtz-muenchen.de] > > Verzonden: woensdag 2 december 2015 21:22 > > Aan: Dominique Ramaekers; libvirt-ML > > Onderwerp: RE: snapshot of running vm's > > > ... > > > > Hi, > > > > i have inserted: > > > > <channel type='unix'> > > <source mode='bind' > > > path='/var/lib/libvirt/qemu/channel/target/sles11.org.qemu.guest_age > nt > > .0'/ > > > > > <target type='virtio' name='org.qemu.guest_agent.0'/> > > <address type='virtio-serial' controller='0' bus='0' port='2'/> > > </channel> > > > > > > I didn't insert the path, it was added automatically, the same with > > "<address type='virtio-serial' controller='0' bus='0' port='2'/>". > > I tried already port 1 and 2, but get this error: > > > > sunhb58820:~/libvirt_1.2.11 # virsh start sles11 > > error: Failed to start domain sles11 > > error: internal error: process exited while connecting to monitor: > > qemu-kvm: -chardev > > > socket,id=charchannel0,path=/var/lib/libvirt/qemu/channel/target/sles1 > > 1.or > > g.qemu.guest_agent.0,server,nowait: Failed to bind socket: No such > > file or directory > > chardev: opening backend "socket" failed > > Does the path /var/lib/libvirt/qemu/channel/target/ exists? > Is libvirt able to write to this path? > Hi Dominique, no, the path does not exist. But from where does libvirt derive it ? Bernd Helmholtz Zentrum Muenchen Deutsches Forschungszentrum fuer Gesundheit und Umwelt (GmbH) Ingolstaedter Landstr. 1 85764 Neuherberg www.helmholtz-muenchen.de Aufsichtsratsvorsitzende: MinDir'in Baerbel Brumme-Bothe Geschaeftsfuehrer: Prof. Dr. Guenther Wess, Dr. Nikolaus Blum, Dr. Alfons Enhsen Registergericht: Amtsgericht Muenchen HRB 6466 USt-IdNr: DE 129521671

1 0

[libvirt-users] snapshot of running vm's
by Lentes, Bernd 02 Dec '15

02 Dec '15

Hi, i'd like to create snapshots of my running vm's. I have several hosts with SLES11 SP4 64bit. I use libvirt 1.2.5-7.1 . VM's are Windows 7, SLES, Ubuntu, Opensuse. I use raw files for the vm's. I try to orientate myself by http://wiki.libvirt.org/page/Live-disk-backup-with-active-blockcommit . The hosts are backuped every night by a network based backup solution (Legato). My idea is: - delete any existing snapshot - create a new snapshot - of course before Legato arrives - then Legato might come and saves my snapshot - next night the same Questions: When I delete a snapshot, is it merged automatically with its base image ? Or is it something I have to do explicitly (maybe via blockcommit) ? How can I get rid of the unused snapshot afterwards ? just rm ? Because I'm doing a live backup do I need to create an external snapshot ? To create an external snapshot do I have to provide --disk-only ? Do I have to provide --live as a parameter ? I tried --quiesce but got that message: virsh # snapshot-create --domain sles11 --atomic --disk-only --quiesce error: argument unsupported: QEMU guest agent is not configured The system I'm testing with is SLES11 SP4 (host and guest). I installed the guest agent: vm58820-8:~ # rpm -q qemu-guest-agent qemu-guest-agent-2.0.2-1.35 Is there something I have to configure ? Is this way of backup a good solution ? Is there something to improve, something I have to take care of ? Is this backup consistent ? Does that all work with libvirt 1.2.5-7.1 or do I need a more recent version (what I don't like because I tried already to update my libvirt but had tons of dependencies) ? I know. A lot of questions. But it's a backup, and I wanted to be sure that it operates properly. Thanks. Bernd -- Bernd Lentes Systemadministration institute of developmental genetics Gebäude 35.34 - Raum 208 HelmholtzZentrum München bernd.lentes(a)helmholtz-muenchen.de phone: +49 (0)89 3187 1241 fax: +49 (0)89 3187 2294 Wer Visionen hat soll zum Hausarzt gehen Helmut Schmidt Helmholtz Zentrum Muenchen Deutsches Forschungszentrum fuer Gesundheit und Umwelt (GmbH) Ingolstaedter Landstr. 1 85764 Neuherberg www.helmholtz-muenchen.de Aufsichtsratsvorsitzende: MinDir'in Baerbel Brumme-Bothe Geschaeftsfuehrer: Prof. Dr. Guenther Wess, Dr. Nikolaus Blum, Dr. Alfons Enhsen Registergericht: Amtsgericht Muenchen HRB 6466 USt-IdNr: DE 129521671

2 3

[libvirt-users] create a directory storage pool in a random location
by Andrei Perietanu 02 Dec '15

02 Dec '15

Hi all, I am trying to figure out if you can store the definition of a directory storage pool in a custom location. This is how I create the storage pool: xmlDesc = """ <pool type='dir'> <name>guest_images_storage_pool</name> <uuid>8c79f996-cb2a-d24d-9822-ac7547ab2d01</uuid> <capacity unit='bytes'>4306780815</capacity> <allocation unit='bytes'>237457858</allocation> <available unit='bytes'>4069322956</available> <source> </source> <target> <path>/disk1/guest_images</path> <permissions> <mode>0755</mode> <owner>-1</owner> <group>-1</group> </permissions> </target> </pool>""" pool = conn.storagePoolDefineXML(xmlDesc, 0) This creates guest_images_storage_pool.xml in /etc/libvirt/storage. I assume this is the default path libvirt looks for any storage pools on the system The problem is that /etc is mapped to a ramdisk on my system. So after restart anything in there would be wiped. The path /disk1 maps to an actual hdd partition so ideally libvirt sould look for it's storage pools in there. I can't seem to make that happen. I tried adding a source path: <source> <dir path="/disk1/storage_pool"/> </source> But that has no effect. Any ideas? Andrei -- The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you receive this in error please contact the sender and delete the material from any computer immediately. It is the policy of Klas Limited to disavow the sending of offensive material and should you consider that the material contained in the message is offensive you should contact the sender immediately and also your I.T. Manager. Klas Telecom Inc., a Virginia Corporation with offices at 1101 30th St. NW, Washington, DC 20007. Klas Limited (Company Number 163303) trading as Klas Telecom, an Irish Limited Liability Company, with its registered office at Fourth Floor, One Kilmainham Square, Inchicore Road, Kilmainham, Dublin 8, Ireland.

2 1

[libvirt-users] libvirt list --all shows empty list after restart
by Andrei Perietanu 01 Dec '15

01 Dec '15

Hi all, I think I might have a small configuration issue: after restarting my machine, all the VMs previously installed are not available. virsh list --all shows an empty list. The .img files where the VMs were intalled are present in the folder I put them in. Furthermore if I import the VMs (virt-install -import) it all seems to work. It's just that I don't want to have to do this after every restart and can't find where libvirt picks up information about any existing VMs on the system. Thanks, Andrei -- The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of or taking of any action in reliance upon this information by persons or entities other than the intended recipient is prohibited. If you receive this in error please contact the sender and delete the material from any computer immediately. It is the policy of Klas Limited to disavow the sending of offensive material and should you consider that the material contained in the message is offensive you should contact the sender immediately and also your I.T. Manager. Klas Telecom Inc., a Virginia Corporation with offices at 1101 30th St. NW, Washington, DC 20007. Klas Limited (Company Number 163303) trading as Klas Telecom, an Irish Limited Liability Company, with its registered office at Fourth Floor, One Kilmainham Square, Inchicore Road, Kilmainham, Dublin 8, Ireland.

2 1

[libvirt-users] Why does libvirt insert USB controller automagically?
by Sundar Nadathur 01 Dec '15

01 Dec '15

Hello all, I find this curious behavior that libvirt will automatically insert a USB controller and a memballoon virtio driver into the domain's XML, when defineXML is called. IOW, the original XML did not have these 2 devices but they are present right after defineXML is called. Why should libvirt add a USB controller? DETAILS: Here is a little Python snippet to show that behavior: import libvirt conn = libvirt.open('qemu:///system') xml_list = open('/tmp/vm.xml', "r").readlines() in_xml = ''.join(xml_list) <-- This XML string does not have a USB or a memballoon dom = conn.defineXML(in_xml) out_xml = dom.XMLDesc() <-- But this XML string does! print out_xml The file /tmp/vm.xml does not have a memballoon or a USB device. Neither, of course, does in_xml string. But out_xml has: <controller type='usb' index='0'> <address type='pci' domain='0x0000' bus='0x00' slot='0x01' function='0x2'/> </controller> <memballoon model='virtio'> <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0'/> </memballoon> I don't see anything in libvirt.conf that explains this. This is libvirt version: 1.2.8, package: 16.el7_1.5 (Red Hat, Inc....). The host is RHEL 7.1. The memballoon part is understandable once I found: https://www.redhat.com/archives/libvir-list/2010-August/msg00132.html However, why should libvirt add a USB controller gratuitously? Is this controllable from the conf file? Cheers, Sundar ________________________________ Confidentiality Notice. This message may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient, you are hereby notified that any use, disclosure, dissemination, distribution, or copying of this message, or any attachments, is strictly prohibited. If you have received this message in error, please advise the sender by reply e-mail, and delete the message and any attachments. Thank you.

2 1

[libvirt-users] "Failed to start domain..."
by Ken D'Ambrosio 01 Dec '15

01 Dec '15

Sadly, I'm back with another issue. I can do a "system list --all" just fine; however, if I attempt to start the machines, I get back: maas@Bill-MAAS-cc:~$ strace -s 1024 -f -o /tmp/asdfasdf.log virsh -c vbox+ssh://gbadmin@10.20.0.1/system start PXE-client-07 error: Failed to start domain PXE-client-07 error: An error occurred, but the cause is unknown Log files on both client and server are pretty sparse on details of any sort... as, again, is Google. Any ideas? Thanks yet again, -Ken

2 3

[libvirt-users] enabling virtio-scsi-data-plane in libvirt
by Vasiliy Tolstov 30 Nov '15

30 Nov '15

Can somebody knows how to enable virtio-scsi-data-plane in libvirt for specific domain? I know that i need to replace "-device virtio-scsi-pci" with "-object iothread,id=io1 -device virtio-scsi-pci,iothread=io1" in qemu, but how can i do this in libvirt? -- Vasiliy Tolstov, e-mail: v.tolstov(a)selfip.ru

2 7

[libvirt-users] How to disable kvm_steal_time feature
by Piotr Rybicki 30 Nov '15

30 Nov '15

Hi. I would like to workaround a bug, when after live-migration of KVM guest, there is a 100% steal time shown in guest. I've read, that disabling 'kvm_steal_time' feature should workarund this bug, but i can't find a way to disable it in libvirt's domain xml file. Tried in <cpu> section: <feature policy='disable' name='kvm_steal_time'/> but that doesn't work. Also, couldn't find any related information in libvirt documentation. Google helps neither. How can I disable this feature? Thanks in advance. Piotr Rybicki

3 6

[libvirt-users] virsh uses internally qemu-img ?
by Lentes, Bernd 30 Nov '15

30 Nov '15

Hi, i read that virsh uses internally qemu-img (http://serverfault.com/questions/692435/qemu-img-snapshot-on-live-vm) Is that true ? so snapshotting a running vm with virsh or qemu-img is the same ? Bernd -- Bernd Lentes Systemadministration institute of developmental genetics Gebäude 35.34 - Raum 208 HelmholtzZentrum München bernd.lentes(a)helmholtz-muenchen.de phone: +49 (0)89 3187 1241 fax: +49 (0)89 3187 2294 Wer Visionen hat soll zum Hausarzt gehen Helmut Schmidt Helmholtz Zentrum Muenchen Deutsches Forschungszentrum fuer Gesundheit und Umwelt (GmbH) Ingolstaedter Landstr. 1 85764 Neuherberg www.helmholtz-muenchen.de Aufsichtsratsvorsitzende: MinDir'in Baerbel Brumme-Bothe Geschaeftsfuehrer: Prof. Dr. Guenther Wess, Dr. Nikolaus Blum, Dr. Alfons Enhsen Registergericht: Amtsgericht Muenchen HRB 6466 USt-IdNr: DE 129521671

2 1

[libvirt-users] libvirtd doesn't attach Sheepdog storage VDI disk correctly
by Adolf Augustin 30 Nov '15

30 Nov '15

Hi, I am trying to use libvirt with sheepdog. I am using Debian 8 stable with libvirt V1.21.0 I am encountering a Problem which already has been reported. ================================================================= See here: http://www.spinics.net/lists/virt-tools/msg08363.html ================================================================= qemu/libvirtd is not setting the path Variable correctly. Every time I convert an image to sheepdog ============================================= qemu-img convert /home/user/temp/debian-8.2.0-amd64-CD-1.iso sheepdog:debian.iso ============================================ the dump of the image says ============================================= virsh # vol-dumpxml --pool herd debian.iso <volume type='network'> <name>debian.iso</name> <key>herd/debian.iso</key> <source> </source> <capacity unit='bytes'>657457152</capacity> <allocation unit='bytes'>658505728</allocation> <target> <path>debian.iso</path> <format type='unknown'/> </target> </volume> ============================================== It should read =============================================== ... <path>sheepdog:debian.iso</path> ... =============================================== I ungraded the system to Debain testing now running: ================================================ virsh # version Compiled against library: libvirt 1.2.21 Using library: libvirt 1.2.21 Using API: QEMU 1.2.21 Running hypervisor: QEMU 2.4.0 ================================================ The Problem ist still existent. It should have been solved in libvirt 1.2.17 See here: https://libvirt.org/news.html ===================================================== .... update sheepdog client] update sheepdog client path (Vasiliy Tolstov), ..... ===================================================== Nevertheless it doesn't work in Debian 8 testing (libvirt 1.2.21). -- Best regards Adolf Augustin E-mail: adolf.augustin(a)zettamail.de PGP-Key: 0xC4709AFE Fingerprint: 1806 35FA CAE8 0202 B7AF 12B9 5956 5BC0 C470 9AFE

2 3