- Users - libvirt List Archives

[libvirt-users] ballooning option xml
by udo1 01 Jun '10

01 Jun '10

Hi everybody, I need the ballooning option with kvm using libvirt xml file and virsh (virsh create domain.xml) kvm line must be appended with the -balloon virtio switch to enable ballooning. (can be disabled if there no option or with -balloon none ) I currently using Ubuntu Lucid How can I define this in the domain xml file, or what need I to do to get this option? is there any option like enabling with <balloon>virtio</balloon> or disabling with <balloon>none</balloon> Thanks in Advance Udo Attila Fischer

3 6

[libvirt-users] Java Bindings against RHEL5.5 KVM
by James Hogarth 01 Jun '10

01 Jun '10

Hi all, I'm doing some dev work against the native RHEL/centos 5.5 KVM installation for maintenance, creation and deletion of virtual guests using the libvirt API vai the java bindings from a java based web app. Things in general seem to be working - I can connect to and query the kvm instance using the libvirt interface and get a list of guests, pull information about them and stop/start/restart guests. I'm hitting a bit of an issue with creation of a guest however.... When calling domainDefineXML with XML following the domain XML specification (with the XML as per a dumpxml on the relevant box) I get an error creating the guest of: "internal error unknown lifecycle action restart " The virsh version information for the host is: Compiled against library: libvir 0.6.3 Using library: libvir 0.6.3 Using API: QEMU 0.6.3 Running hypervisor: QEMU 0.10.5 I have tried this against my ubuntu system as well with the same error - that has 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 I'm currently using jna 3.2.5 with libvirt 0.4.5 bindings. The example XML being used is: <domain type="kvm"> <name> newguestfomwebapp </name> <memory> 512 </memory> <vcpu> 1 </vcpu> <os> <type arch="x86_64" machine="rhel5.4.0"> hvm </type> <boot dev="hd"/> </os> <features> <acpi/> <apic/> <pae/> </features> <clock offset="utc"/> <on_poweroff> destroy </on_poweroff> <on_reboot> restart </on_reboot> <on_crash> restart </on_crash> <devices> <emulator> /usr/libexec/qemu-kvm </emulator> <disk type="file" device="cdrom"> <driver name="qemu" type="raw"/> <source file="/var/lib/libvirt/images/ISOs/centos5.5-ks.iso"/> <target dev="hdc" bus="ide"/> <readonly/> </disk> <interface type="network"> <source network="default"/> <model type="virtio"/> </interface> <input type="mouse" bus="ps2"/> <graphics type="vnc" port="-1" autoport="yes"/> </devices> </domain> If I remove the lifecycle information from the xml I get: "unknown OS type hvm " The xml works from the command line with virsh define ... Anyone seen anything similar or have any ideas? Regards, James

1 1

[libvirt-users] Update Libvirt
by Rafael Braga 31 May '10

31 May '10

Hi, I installed the opensuse 11.0 and libvirt version installed is 0.4. But, Nimbus need the version 0.6.3. How do I upgrade? I can't by YAST... thanks! -- Rafael Braga

2 2

[libvirt-users] monitoring kvm machines with libvirt
by Karsten Knese 31 May '10

31 May '10

Hey folks. Currently, i am working on my bachelorthesis. Subject of this thesis is to setup an accounting-solution for the Private Cloud with Eucalyptus. Eucalyptus works with libvirt providing kvm instances. my question is, if i can grep facts like general uptimes, cpu-peaks and disk i/o of each vm with libvirt-tools?! I just found one called virt-top, which monitors each vm, but unfortunately it produces to much overhead. If you should be no tool, can anybody give me a hint, where i best look at to grep these facts? First try is to grep the uptime of each vm by writing a little deamon, checking one time per minute via virsh, if the vm is in state running. But i don't know, if this is best. thanks for your support! cheers, Karsten

1 0

[libvirt-users] VM parameters best practice?
by Guido Winkelmann 31 May '10

31 May '10

Hi, When creating many virtual machines from a template (disk image), you often need to set some different parameters for each vm. Those usually include root passwords and/or ssh public keys for login, ssh Host keys (so that a) not all the VMs have the same host key and b) the user can know the real host key for his vm before connecting for the first time) plus any number of user-defined parameters. IP addresses and hostnames can be set via DHCP. Obviously, some of those parameters need to be treated as confidential, so you cannot just grab them from some FTP server somewhere. In fact, as long as you haven't set any authentication credentials (like an SSL Client Certificate for example) that the vm can use to uniquely and securely identify itself to some server, you cannot securely get any of these parameters over the network. How do people usually solve that sort of problem? What is considered best practice here? Currently, my own plans are to write these parameters to a floppy image, use Unix file permissions to make sure nobody else can read the image, and then hook up that image to a virtual floppy drive in the respective vm. A shell script in the vm template can then read and apply all those parameter at some early stage during the very first boot. Unfortunately, this means mounting the image in the host system, writing the parameters and then unmounting it again - which seems sort of error prone to me in a fully automated environment. (What if two vms are being created simultaneously? What if the script crashes or hangs somewhere between mounting and unmounting the image?) Regards Guido Winkelmann

1 0

[libvirt-users] Cannot create qcow2 images with libvirt 0.8.1
by Guido Winkelmann 31 May '10

31 May '10

Hi, After upgrading to libvirt 0.8.1, I can no longer create empty volumes with no backing store and an explicit format of qcow2. This XML volume definition: <volume> <name>testserverb-data2.img</name> <allocation>0</allocation> <capacity units='G'>20</capacity> <target> <format type='qcow2'/> <permissions> <owner>107</owner> <group>107</group> <mode>0744</mode> </permissions> </target> </volume> yields the error "Cannot run /usr/bin/qemu-img to create /data/testserverb- data.img: Invalid argument" In /var/log/messages on the qemu host I can find the following snippet after that: error : virRunWithHook:933 : internal error '/usr/bin/qemu-img create -f qcow2 /data/testserverb-data.img 0K' exited with non-zero status 1 and signal 0: qemu-img: Image creation needs a size parameter#012 error : virStorageBackendCreateExecCommand:537 : Cannot run /usr/bin/qemu-img to create /data/testserverb-data.img: Invalid argument error : qemudClientReadBuf:1653 : gnutls_record_recv: A TLS packet with unexpected length was received. If I leave out the <format> element and/or specify a <backingStore> element for the volume, it works again. (I suppose the backing store makes it work because in that case qemu-img does not need a size parameter...) The host system is a vanilla Fedora Core 12 system (aside from the upgraded libvirt). Regards, Guido Winkelmann

2 3

[libvirt-users] Hot-add interface crashes libvirt
by steven765＠yahoo.com 27 May '10

27 May '10

Hi Everyone, After upgrading FC12 and now FC13 whenever I try to attach a network device using attach-device it promptly crashes libvirtd. Is there an option that changed in the network interface xml spec I missed? I really need this feature or at least the ability to connect and disconnect interfaces if that's possible? I've tried defining my network card in several ways: First the one that worked previously: <interface type="network"> <mac address="52:54:00:5c:fe:a8"/> <source network="fedora"/> </interface> Second copying what appeared in the xml after using the add hardware through the VMM: <interface type='network'> <mac address='52:54:00:5c:fe:a8'/> <source network='fedora'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x06' function='0x0'/> </interface> Both ways fail now. Please help. Thanks, Steve

2 1

[libvirt-users] How to obtain rights to extend libvirt documentation pages
by Марк Коренберг 25 May '10

25 May '10

For example, http://libvirt.org/formatdomain.html#elementsDisks does not contain <serial>xxxx</serial> tag. (https://bugzilla.redhat.com/show_bug.cgi?id=593266) I wrote to community (users and developers). No one has added this information to the site... -- Segmentation fault

3 2

[libvirt-users] NAT-based network 2x-4x faster than routed network?
by Kenni Lund 22 May '10

22 May '10

Hi Is it expected behaviour that a libvirt NAT-based network is faster than a libvirt routed network? I would guess that the routed network would be the fastest one in all cases, but I'm seeing the opposite in my setup. On a 100mbit internet connection, the routed network tops at 4mbyte/sec, while the nat-based network goes all the way to 10mbyte/sec. I've spend the last 6 hours on trying to figure out the cause of this, changing settings, reinstalling the system, but the problem persists and I can't find the cause of it. It is a clean and fairly simple setup: - Clean installation of CentOS 5.5 - Installation of kvm+libvirt+virt-manager+xorg - iptables service disabled - 1 public IP for the host (xxx.yyy.zzz.215) on a /32 subnet with xxx.yyy.zzz.193 as gateway - 3 public IP for the routed network (xxx.yyy.zzz.251-253) on a /26 subnet with the same gateway as the host - 1 virtual machine with either the NAT or the routed libvirt network assigned I have no networking connection errors or similar, it works perfectly with both the NAT and the routed network, but the routed network is just 2-4 times slower than the NAT one. Do you have any suggestions of what the problem might be? When I look at the iptables rules generated by libvirt, and the routing table, everything looks fine AFAICT. Traceroute from guest doesn't reveal anything either. Ping times from the two networks are similar, so the main issue seems to be throughput. Any help or pointers to what I should look at, is highly appreciated... :) I've copy pasted some of my relevant system configuration into pastebin: http://pastebin.com/jtTrHLqA Thank you, Kenni

1 0

[libvirt-users] Update of libvirt website required. not all information about XML tags
by Марк Коренберг 21 May '10

21 May '10

The problem desribed here: https://bugzilla.redhat.com/show_bug.cgi?id=593266 Please update http://www.libvirt.org/formatdomain.html and describe NEW xml tags, like <serial>xxxx</serial> for hard drive and all cpu tags, other if any. -- Segmentation fault

4 8