- Users - libvirt List Archives

[libvirt-users] Regarding persistence of VM's after live migration (virDomainMigrateToURI() problem)
by Coding Geek 03 May '12

03 May '12

Hello I am working with 3 host machines each running xen with shared NFS storage. I am working on automatic load balancing if one host is over utilized and another is under utilized by measuring the utilization from xentop. I am facing a problem after migration of VM. I am setting the flags ( 1| 8| 16) in order to do live migration, persist VM on destination, undefine host from source. After migration if I shut off the migrated VM on destination host it does not persist. I am using the migrateToURI() API for migration. Please help how to make VM persist on destination. Thank you, Tasvinder Singh -- -- "To Iterate is Human, To Recurse is Divine"

3 6

[libvirt-users] Snooze energy-efficient cloud manager is available as open-source!
by Eugen Feller 03 May '12

03 May '12

Dear all, herewith we proudly announce the first public release of Snooze. Snooze is an open-source scalable, autonomic, and energy-efficient virtual machine (VM) management framework for private clouds based on libvirt. Similarly to other VM management frameworks such as Nimbus, OpenNebula, Eucalyptus, and OpenStack it allows to build compute infrastructures from virtualized resources. Particularly, once installed and configured users can submit and control the life-cycle of a large number of VMs. However, contrary to existing frameworks for scalability and fault tolerance, Snooze employs a self-organizing and healing hierarchical architecture. Moreover, it performs distributed VM management and is designed to be energy efficient. Therefore, it implements features to monitor and estimate VM resource (CPU, memory, network Rx, network Tx) demands, detect and resolve overload/underload situations, perform dynamic VM consolidation through live migration, and finally power management to save energy. Last but not least, it integrates a generic scheduler which allows to implement any VM placement algorithms. The system can be either used to manage production data centers or as an experimental testbed for advanced (i.e. requiring live migration support) VM placement algorithms. Please see our website "http://snooze.inria.fr" for more information. We are looking forward for your feedback and contributions! Best regards, Eugen -- Eugen Feller PhD student, University of Rennes I INRIA Rennes Bretagne Atlantique - MYRIADS project Tel: +33 (0) 2 99 84 72 68 Web: http://www.irisa.fr/myriads/members/efeller

2 1

[libvirt-users] Python Libvirt Bindings
by John Wayne 02 May '12

02 May '12

Hi, Can anyone please share sample Python code for using qemu+tcp with SASL or qemu+tls with SASL? (i.e., connecting to a hypervisor using qemu+tls or qemu+tcp when a username/password using SASL has been defined on the hypervisor) I am having some trouble figuring out how to use callbacks for authenticating from with my Python program. I assume I will have to use http://libvirt.org/html/libvirt-libvirt.html#virConnectCredentialType, but I can't quite translate that to Python from C. Thanks

2 1

[libvirt-users] serial console locking for a domain (concurrent access = garbled output)
by Dusty Mabe 02 May '12

02 May '12

Hi, I am seeing an issue in libvirt-0.9.4-23.el6_2.7.x86_64.rpm where my virsh console output for a domain gets garbled because I have virt-manager running and it is also trying to access the serial device for the guest domain. Basically there is concurrent serial access and half the text ends up on one end while the rest ends up on the other end. Closing virt-manager does not help. Disconnecting and reconnecting using virsh console does not help. The only way to recover is to destroy the domain and then start the guest again and never start virt-manager. Note that the domain does not have VNC graphics so virt-manager defaults to try to access the serial port when started for a particular domain. Should there be locking within libvirt to only allow for a single client to access a specific serial port of a domain? Thanks, Dusty Mabe

2 2

[libvirt-users] Console to RHEL6.1 container
by Sukadev Bhattiprolu 02 May '12

02 May '12

We are trying to create RHEL6.1 container and having trouble getting a proper console to the container. I have used lxc containers in the past and lxc has a script (lxc-fedora) to setup a basic Fedora container. Is there something similar or docs with virtmgr ? We started out making a copy of the root fs and using that copy as the root for the container. I made a couple of tweaks to the container rootfs: 1. ensured that /dev/ptmx is a symlink to pts/ptmx 2. Added 'newinstance' option to the devpts mount in fstab The container xml file is attached. If devpts entry in container's fstab has the 'newinstance' option, I get a brief "domain test01 started" message on the stdout. When I run 'virsh consle test01', I only see a few messages --- Setting hostname node03: [ OK ] Setting up Logical Volume Management: No volume groups found [ OK ] Checking filesystems [ OK ] mount: can't find / in /etc/fstab or /etc/mtab Mounting local filesystems: [ OK ] Enabling local filesystem quotas: [ OK ] Enabling /etc/fstab swaps: [ OK ] --- and then a hang wihtout a shell. At ths point sshd or any of the other services don't appear to be running. If I remove the 'newinstance' entry in fstab and run 'virsh start test01' the container starts up ok. I get all "console" messages on whichever is my current /dev/pts/0 ! At this point sshd and many other services are running and I can ssh in. Even in this case, though 'virsh console test01' shows the same _few_ messages as above and no login shell. Back to the newinstance case, by putting some debug messages in the container's /etc/rc.sysinit, I see that devpts is already mounted with newinstance. With lxc, in setup_pts(), we unmount devpts in the container and then mount with newinstance. I wonder what that sequence is with virtmgr. I double checked that /dev/ptmx in container is a symlink to pts/ptmx and that CONFIG_DEVPTS_MULTIPLE_INSTANCES=y. Appreciate any tips on how I can get access to the console of the RHEL6.1 container. Suka Suka

1 1

[libvirt-users] question of iSCSI pool on the same iSCSI target
by david_zhu 28 Apr '12

28 Apr '12

3 2

[libvirt-users] Is there a way for telling libvirt to auto define an xml under /etc/libvirt/qemu?
by Raoul Scarazzini 27 Apr '12

27 Apr '12

Hi everybody, like in the subject: is there a way for telling libvirt to auto define an xml under /etc/libvirt/qemu? I mean, sometimes it happens to have the /etc/libvirt/qemu shared between nodes, say with a NFS share, since I can set a new xml on a node and I want o see the virtual machine on all the others, is there a way to automate this or am I obligated to launch virsh define <xml> on EACH node? Thank you so much, -- Raoul Scarazzini Solution Architect MMUL: Niente è impossibile da realizzare, se lo pensi bene! +39 3281776712 rasca(a)mmul.it http://www.mmul.it

4 3

[libvirt-users] Installing the python libvirt bindings in a virtualenv
by Daniel Gonzalez 26 Apr '12

26 Apr '12

Hi, I have a virtualenv which I am using for production (I need python 2.7 and certain libraries which can not be easily installed on the host environment). Now I am trying to install also libvirt, but I have not succeeded yet. The problem that I have now looks quite difficult to solve. This is the script I am using to install libvirt (bash script): install_libvirt_ ( ) { local myvirtualenv="$1" local libvirt_tag=v0.9.9 cd $TMP_DIR git clone git://libvirt.org/libvirt.git cd libvirt git checkout $libvirt_tag mkdir -p $VIRTUALENVS_DIR/$myvirtualenv/usr sudo apt-get install autopoint # configure: error: You must install the GnuTLS library in order to compile and run libvirt sudo apt-get install -y gnutls-bin gnutls-dev # configure: error: You must install device-mapper-devel/libdevmapper >= 1.0.0 to compile libvirt sudo apt-get install -y libdevmapper-dev libdevmapper # configure: error: You must install python-devel to build Python bindings sudo apt-get install -y python-all-dev ./autogen.sh --prefix=$VIRTUALENVS_DIR/$myvirtualenv/usr --enable-compile-warnings=error make make install } But this is failing with error message: checking for python script directory... ${prefix}/lib/python2.7/site-packages checking for python extension module directory... ${exec_prefix}/lib/python2.7/site-packages configure: error: You must install python-devel to build Python bindings python-devel does not exist. I have used python-all-dev, but it has not solved the problem. Has somebody succeeded in installing libvirt-python inside a virtualenv? Thanks, Daniel Gonzalez

2 2

[libvirt-users] Snapshot system: really confusing.
by NoxDaFox 25 Apr '12

25 Apr '12

Greetings, I am developing a platform for test and debug use. The typical scenario I want to realize is the following: I run a .qcow2 image via QEMU-KVM, the running OS is not relevant. I want to take several snapshots of both Disk and RAM status; same file, separate files doesn't matter. I just need a way to have consistent information about them (better with XML description) and the data as well. I want to store these snapshots on my storage devices to use them whenever and wherever I want. Is it possible to store a single snapshot providing both the memory and disks state in a file (maybe a .qcow2 file)? Is there any way to get a unique interface which handles my snapshots? I was used to use the virDomainSnapshotCreateXML() defining the destination file in the XML with <disk> fields. After updating libvirt it was not working anymore, I thought was a bug but then I realized it was intentional. The function complains about the fact that the <disk> parameter is not accepted anymore. So I started guessing how to solve reading the API documentation and I fall in a completely nebulous world. For what I got: - virDomainSnapshotCreateXML(): According to flags can take system checkpoints (really useful) and disks snapshots. System checkpoints: What I need but I didn't find any way to retrieve the storage file; I'm only able to get the snapshot pointer, quite useless as from its pointer I can only print the XML description. Disk snapshot: here through the XML description I can specify a target file where to write the information. But I still need the memory status. - virDomainSaveFlags(): No matter how I set the flags, it halts the guest; I need it still running afterward and I think that reverting the domain and restarting for my scenarios is unaffordable. - virDomainCoreDump(): Does what I need for the memory, but doesn't give any XML description and doesn't supply any useful interface to handle it, I just get back an int that represent the exit status of the routine. (other functions really similar) The question is: why all this confusion? I absolutely understand the problematic that realizing a multiplatform snapshots management raises; but I think that for an API purpose what is implemented here is completely confusing the developer. Regards, NoxDaFox.

2 1

[libvirt-users] Non-Root not able to use openReadOnly()
by vipul borikar 24 Apr '12

24 Apr '12

Hi, I have a code : import libvirt con = libvirt.openReadOnly(None) print con.listDomainsID() When i run the above code using root i get the proper output... but when i use non-root user it givers be a empty list [ ] If i change the 2nd line to con = libvirt.open(qemu:///system) It works. I am using RHEL 6.1 with default kvm and libvirt. unix_sock_group = "virt" in libvirtd.conf where virt is a group of non-root user group = "virt" in qemu.conf is set. Any Ideas???? -- Thanks Vipul Borikar "Our task must be to free ourselves...by widening our circle of compassion to embrace all living creatures and the whole of nature and its beauty."

2 1