- Users - libvirt List Archives

Re: [libvirt-users] [Qemu-devel] About VM fork in QEMU
by Xinyang Ge 02 Nov '13

02 Nov '13

>>> External snapshots (via the blockdev-snapshot-sync QMP command) can be >>> taken in a matter of milliseconds if you only care about disk state. >>> Furthermore, if you want to take a snapshot of both memory and disk >>> state, such that the clone can be resumed from the same time, you can do >>> that with a guest downtime that only lasts as long as the >>> blockdev-snapshot-sync, by first doing a migrate to file then doing the >>> disk snapshot when the VM pauses at the end of migration. Resuming the >>> original guest is fast; resuming from the migration file is a bit >>> longer, but it is still the fastest way possible to resume from a >>> memory+disk snapshot. If you need anything faster, then yes, you would >>> have to write patches to qemu to attempt cloning via fork() that makes >>> sure to modify the active disk in use by the fork child so as not to >>> interfere with the fork parent. >> >> I think migrating memory to file then doing external disk snapshot is >> exactly what we want. Since we are using libvirt to manage different >> VMs, could you give us some specific guides (or references) that how >> we could migrate memory state to file using virsh interfaces and do >> external snapshots? > > virsh snapshot-create-as $dom $name --live --memspec /path/to/memoryfile I have tried this command on libvirt v1.1.3 and it returns "error: invalid argument: qemuDomainSnapshotCreateXML: unsupported flags (0x100)". Looks like --live is not supported yet. Could you let us know which version we should of libvirt we should use in order to use this feature? Thanks, Xinyang -- Xinyang GE Department of Computer Science & Engineering The Pennsylvania State University Homepage: http://www.cse.psu.edu/~xxg113/

2 5

[libvirt-users] Fwd: Help regarding virsh domifstat
by Rohit Bhat 01 Nov '13

01 Nov '13

Hi, I need a small help. I am working on a project where i have to monitor network activity of a VM running on KVM. I am interested in how much data is going into the VM and how much data is coming out of the VM. I checked on the net and found out virsh domifstat is the way to go about it. 1. But looks like these stats also include bytes related to control traffic for the VM. Is there a way to exclude that? I just want the size of actual data transfers. 2. Is there a way by which i can report the data transfer of VM with the outside world (outside hypervisor) only while excluding data transfer with any other VM on the same host or with the host itself? Thanks, Rohit

1 0

[libvirt-users] libvirt_lxc causes cpu 100% usage
by jj zhong 01 Nov '13

01 Nov '13

libvirt_lxc causes cpu 100% usage? libvirt version : 1.1.1 I don't know what happend, any help would be appreciated. thx top - 07:39:44 up 1 day, 2:13, 2 users, load average: 4.72, 4.59, 4.51 Tasks: 243 total, 4 running, 239 sleeping, 0 stopped, 0 zombie %Cpu0 : 33.1 us, 1.3 sy, 0.0 ni, 64.6 id, 0.7 wa, 0.0 hi, 0.3 si, 0.0 st %Cpu1 : 41.1 us, 3.0 sy, 0.0 ni, 55.3 id, 0.7 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu2 : 18.0 us, 2.0 sy, 0.0 ni, 79.7 id, 0.0 wa, 0.0 hi, 0.3 si, 0.0 st %Cpu3 : 94.8 us, 5.2 sy, 0.0 ni, 0.0 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu4 : 55.7 us, 7.5 sy, 0.0 ni, 36.7 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu5 : 66.1 us, 4.3 sy, 0.0 ni, 29.6 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu6 : 16.8 us, 1.0 sy, 0.0 ni, 82.2 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st %Cpu7 : 7.2 us, 0.7 sy, 0.0 ni, 92.1 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st KiB Mem: 1972112 total, 1894912 used, 77200 free, 236576 buffers KiB Swap: 10485744 total, 0 used, 10485744 free, 574544 cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 15750 root 20 0 10448 2080 1216 R 100.3 0.1 1322:38 libvirt_lxc 27517 root 20 0 10448 2032 1200 R 99.3 0.1 1273:44 libvirt_lxc 25250 root 20 0 10480 2048 1200 R 99.0 0.1 1277:31 libvirt_lxc ....

4 7

[libvirt-users] upstart script for virtlockd
by NEVEU Stephane 01 Nov '13

01 Nov '13

Hi all, Trying to test libvirt 1.1.3 with virtlockd locking my qcow2 images on a NFS storage between two kvm hosts. ./configure ... --with-init-script=upstart Libvirtd upstart script is actually well generated but I can't see nothing about virtlockd... or am I blind ? :) Nevertheless, running virtlockd -d && service libvirtd restart works fine. Am I wrong thinking that editing qemu-lockd.conf with my needs and qemu.conf with lock_manager="lockd" was enough ? Thank you for your help.

2 3

[libvirt-users] libvirt_lxc: SELinux MCS
by Matteo Piccinini 01 Nov '13

01 Nov '13

Hello list, my name is Matteo, i'm new on that list. I'm working on a multitenancy platform with linux containers through libvirt on a production system with Red Hat 6.4. Every container run a separate instance of OpenSSH and Apache HTTPd and I need to give root privileges to the developers and I try to configure SELinux using svirt and MCS. I try the secmodel type dynamic and static in the xml file but it didn't work, I received the following error: error : virSecurityLabelDefParseXML:3228 : XML error: security label is missing error : virNetSocketNewConnectUNIX:566 : Failed to connect socket to '/var/run/libvirt/lxc/cntr1.sock': Connection refused I configure the followings secmodel definition and used chcon on the rootfs directory (created with yum) with "system_u:object_r:svirt_lxc_file_t:s0:c30,c50" label: <seclabel type='static' model='selinux' relabel='no'> <label>system_u:system_r:svirt_lxc_net_t:s0:c30,c50</label> </seclabel> or: <seclabel type='dynamic' model='selinux' relabel='yes'> <label>system_u:system_r:svirt_lxc_net_t:s0:c30,c50</label> </seclabel> I try to compile the last version from the master branch of git the result was always the same, the error was related to the SELinux driver not enabled. The output from "virsh -c lxc:/// capabilities" doesn't show the secmodel and doi tag like the qemu/kvm have for the lxc driver the selinux driver. How can I enable the SELinux driver for libvirt lxc in Red Hat 6.4? SELinux is in enforcing mode. Thanks in advance, Matteo

2 3

[libvirt-users] libvirt-lxc driver on armv7l
by Jan Olszak 01 Nov '13

01 Nov '13

Hi there, I'm interested in using libvirt-lxc on armv7l architecture and I have a hard time building libvirt. 1. Are there any prebuild rpms for ARM? 2. I have many dependency issues during build: error: *** Error Summary *** === the following packages failed to build due to missing build dependencies === libvirt: nothing provides hal-devel nothing provides libnl-devel nothing provides libselinux-devel nothing provides radvd nothing provides ebtables nothing provides cyrus-sasl-devel nothing provides /usr/bin/qemu-img nothing provides lvm2 nothing provides iscsi-initiator-utils nothing provides parted-devel nothing provides device-mapper-devel nothing provides libssh2-devel >= 1.3.0 nothing provides libwsman-devel >= 2.2.3 nothing provides nfs-utils nothing provides scrub Where do I define the flags that discard some of the libraries (for example "--without-hal")? Thanks! Jan

2 3

[libvirt-users] libvirt-lxc without libvirt
by Jan Olszak 01 Nov '13

01 Nov '13

Hi there! 1. Is it possible to build libvirt-lxc driver without building libvirt? How do I do it? 2. Do I understand it correctly, that I can use libvirt-lxc as a substitute of lxc? Thanks, Jan

2 1

[libvirt-users] Fwd: libvirt unsupport rbd storage pool? "missing backend for pool type 8"
by 卢惠勇 31 Oct '13

31 Oct '13

I use "virsh pool-define rbd.xml" to create a rbd storage pool,but get this error virsh pool-define /tmp/rbd.xml error: Failed to define pool from /tmp/rbd.xml error: internal error: missing backend for pool type 8 rbd.xml <pool type="rbd"> <name>cloudstack</name> <source> <name>cloudstack</name> <host name='10.20.197.22' port='6789'/> </source> </pool> this is my system information: centos6.4 2.6.32-358.14.2.el6.x86_64 libvirt-1.1.1-1.el6.x86_64 qemu-kvm-0.12.1.2-2.355.el6.2.cuttlefish.x86_64 (download from http://ceph.com/packages/qemu-kvm/centos/x86_64/ ) qemu-img-0.12.1.2-2.355.el6.2.cuttlefish.x86_64 -- ------------------------------ 金山网络运维部-卢惠勇分机:2477 手机:15018838589 qq:31302409

2 1

[libvirt-users] Upgraded to F18: now machines won't start
by Sean Darcy 31 Oct '13

31 Oct '13

We've gone back to using an older machine that used to run libvirt just fine. Updated to F18, libvirt-1.1.0-1.fc18.x86_64. VirtualBox works ok. Now I can't start any image. For one machine i get: .............. File "/usr/lib64/python2.7/site-packages/libvirt.py", line 698, in create if ret == -1: raise libvirtError ('virDomainCreate() failed', dom=self) libvirtError: internal error Cannot find suitable CPU model for given data that's whether I use what we used to use "Phenom" or the selection from allowing virt-manager to choose "Opteron G3". For another machine we get: libvirtError: internal error Process exited while reading console log output: char device redirected to /dev/pts/37 (label charserial0) failed to initialize KVM: Device or resource busy Doing some debugging, I found: 1. On boot the kvm_amd and kvm kernel modules won't/don't load. Nothing in dmesg. Modprobe does load them. lsmod | grep kvm kvm_amd 59982 0 kvm 429349 1 kvm_amd I still can't make anything work with the modules loaded. 2. When I start libvirtd, I get: Oct 30 14:56:29 sixcore systemd[1]: Starting Virtualization daemon... Oct 30 14:56:29 sixcore systemd[1]: Started Virtualization daemon. ............ Oct 30 14:56:31 sixcore kernel: [ 837.705913] kvm: enabling virtualization on CPU5 failed Oct 30 14:56:31 sixcore kernel: [ 837.705920] kvm: enabling virtualization on CPU4 failed Oct 30 14:56:31 sixcore kernel: [ 837.705928] kvm: enabling virtualization on CPU1 failed Oct 30 14:56:31 sixcore kernel: [ 837.705934] kvm: enabling virtualization on CPU2 failed Oct 30 14:56:31 sixcore kernel: [ 837.705946] kvm: enabling virtualization on CPU3 failed Oct 30 14:56:31 sixcore kernel: [ 837.745267] kvm: enabling virtualization on CPU0 failed I looked at the wiki on how to debug, and I seem to pass all the tests. virsh capabilities | grep kvm <domain type='kvm'> <emulator>/usr/bin/qemu-kvm</emulator> <domain type='kvm'> <emulator>/usr/bin/qemu-kvm</emulator> virsh dumpxml 2008R2 | grep kvm <domain type='kvm'> <emulator>/usr/bin/qemu-kvm</emulator> Am I right that some of the entries in the xml file have now changed? Is there a guide to updating them from the prior entries? And, if I try to create a new machine, virt-install just hangs: virt-install --prompt What is the name of your virtual machine? New-2008R2 How much RAM should be allocated (in megabytes)? 1200 What would you like to use as the disk (file path)? /opt/storage/virt/windows/images/2008R2.img ls -l /opt/storage/virt/windows/images/2008R2.img -rw-------. 1 root root 23571726336 Dec 8 2012 /opt/storage/virt/windows/images/2008R2.img I changed the permissions to 666. still hung. Any help appreciated. Sean

1 0

[libvirt-users] virsh esx return "error: internal error: Could not find Datacenter"
by Andrea Buzzi (abuzzi) 30 Oct '13

30 Oct '13

Hello, I’m running virsh v1.1.3 compiled on a ubuntu system with the "--with-esx=yes” support. root@kvm-189:/home/cisco# virsh --version 1.1.3 I also have a vcenter 5.5 environment that has following structure (couple of hosts at datacenter level, others part of cluster group): MIC-VC-ADMIN MIC-DC-ADMIN 10.58.6.110 10.58.6.113 MIC-CL-ADMIN 10.58.6.109 10.58.6.118 10.58.6.126 I can succesfully connect to ESXi host 10.58.6.110: root@kvm-189:/home/cisco# virsh -c esx://10.58.6.110?no_verify=1 list --all Enter username for 10.58.6.110 [root]: Enter root's password for 10.58.6.110: Id Name State ---------------------------------------------------- 250 ssi-vepc1 running - SSI-Template shut off root@kvm-189:/home/cisco# However if I try to connect to vcenter, I get the following: root@kvm-189:/home/cisco# virsh -c esx://10.58.6.110?no_verify=1\&vcenter=* Enter username for 10.58.6.110 [root]: Enter root's password for 10.58.6.110: Enter username for 10.58.6.7 [administrator]: Administrator Enter Administrator's password for 10.58.6.7: error: failed to connect to the hypervisor error: internal error: Could not find Datacenter root@kvm-189:/home/cisco# Any suggestion on how to isolate (and possibly fix) the issue ? Thanks, Andrea -- [Description: Description: Description: http://www.cisco.com/swa/i/logo.gif] Andrea Buzzi TME @ MIBU abuzzi(a)cisco.com<mailto:abuzzi@cisco.com> Phone: +39 039.6295.319 Mobile: +39 348.711.9960 Cisco Systems Italy S.r.l Via Torri Bianche, 8 20871 Vimercate (MB) Italy Cisco.com<http://www.cisco.com/global/IT/> This email may contain confidential and privileged material for the sole use of the intended recipient. Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender by reply email and delete all copies of this message. For corporate legal information go to: http://www.cisco.com/web/about/doing_business/legal/cri/index.html

1 0