- Users - libvirt List Archives

[libvirt-users] Libvirt exiting and respawning
by Stephen Eilert 06 Nov '11

06 Nov '11

Hi, I'm trying to use Libvirt with LXC, under Ubuntu 11.04, but I can't get it to work. It keeps exiting and being restarted by upstart. Log file follows. It appears that it is complaining, among other things, of QEMU. I am not running QEMU and would like to disable everything except for LXC, but I am not sure it would fix my issue. I've set the log level to debug, even, but didn't see anything out of ordinary. Even resorted to strace, but I don't know enough about libvirtd to try and understand its behavior. Anyone got any clue about what's causing it to decide to exit? # libvirtd --version libvirtd (libvirt) 0.9.2 # uname -a Linux CI-Server-3-Backend-2 2.6.35.4-rscloud #2 SMP Sat Oct 29 21:16:49 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux # lxc-checkconfig --- Namespaces --- Namespaces: enabled Utsname namespace: enabled Ipc namespace: enabled Pid namespace: enabled User namespace: enabled Network namespace: enabled Multiple /dev/pts instances: enabled --- Control groups --- Cgroup: enabled Cgroup namespace: enabled Cgroup device: enabled Cgroup sched: enabled Cgroup cpu account: enabled Cgroup memory controller: enabled Cgroup cpuset: enabled --- Misc --- Veth pair device: enabled Macvlan: enabled Vlan: enabled File capabilities: enabled # cat /var/log/syslog Nov 5 18:42:50 CI-Server-3-Backend-2 libvirtd: 18:42:50.696: 25820: info : libvirt version: 0.9.2 Nov 5 18:42:50 CI-Server-3-Backend-2 libvirtd: 18:42:50.696: 25820: info : networkReloadIptablesRules:1410 : Reloading iptab les rules Nov 5 18:42:50 CI-Server-3-Backend-2 libvirtd: 18:42:50.776: 25820: error : virCommandWait:1393 : internal error Child proce ss (/sbin/iptables --table mangle --delete POSTROUTING --out-interface virbr0 --protocol udp --destination-port 68 --jump CHE CKSUM --checksum-fill) status unexpected: exit status 1 Nov 5 18:42:50 CI-Server-3-Backend-2 libvirtd: 18:42:50.817: 25820: error : virCommandWait:1393 : internal error Child proce ss (/sbin/iptables --table mangle --insert POSTROUTING --out-interface virbr0 --protocol udp --destination-port 68 --jump CHE CKSUM --checksum-fill) status unexpected: exit status 1 Nov 5 18:42:50 CI-Server-3-Backend-2 libvirtd: 18:42:50.818: 25820: warning : networkAddGeneralIptablesRules:1199 : Could not add rule to fixup DHCP response checksums on network 'default'. Nov 5 18:42:50 CI-Server-3-Backend-2 libvirtd: 18:42:50.818: 25820: warning : networkAddGeneralIptablesRules:1200 : May need to update iptables package & kernel to support CHECKSUM rule. Nov 5 18:42:50 CI-Server-3-Backend-2 libvirtd: 18:42:50.888: 25820: error : udevDeviceMonitorStartup:1616 : Failed to initialize libpciaccess: No such file or directory Nov 5 18:42:50 CI-Server-3-Backend-2 libvirtd: 18:42:50.889: 25820: error : virStateInitialize:830 : Initialization of udev state driver failed Nov 5 18:42:50 CI-Server-3-Backend-2 libvirtd: 18:42:50.930: 25820: info : qemudStartup:525 : Unable to create cgroup for driver: Invalid argument Nov 5 18:42:50 CI-Server-3-Backend-2 libvirtd: 18:42:50.931: 25820: info : qemudLoadDriverConfig:315 : Configured cgroup controller 'cpu' Nov 5 18:42:50 CI-Server-3-Backend-2 libvirtd: 18:42:50.931: 25820: info : qemudLoadDriverConfig:315 : Configured cgroup controller 'memory' Nov 5 18:42:50 CI-Server-3-Backend-2 libvirtd: 18:42:50.931: 25820: info : qemudLoadDriverConfig:315 : Configured cgroup controller 'devices' Nov 5 18:42:50 CI-Server-3-Backend-2 libvirtd: 18:42:50.931: 25820: info : qemudLoadDriverConfig:315 : Configured cgroup controller 'blkio' Nov 5 18:42:51 CI-Server-3-Backend-2 libvirtd: 18:42:51.100: 25820: info : virDomainLoadAllConfigs:9950 : Scanning for configs in /var/run/libvirt/qemu Nov 5 18:42:51 CI-Server-3-Backend-2 libvirtd: 18:42:51.100: 25820: info : virDomainLoadAllConfigs:9950 : Scanning for configs in /etc/libvirt/qemu Nov 5 18:42:51 CI-Server-3-Backend-2 kernel: [ 442.486147] lo: Disabled Privacy Extensions Nov 5 18:42:51 CI-Server-3-Backend-2 libvirtd: 18:42:51.319: 25820: info : virDomainLoadAllConfigs:9950 : Scanning for configs in /var/run/libvirt/lxc Nov 5 18:42:51 CI-Server-3-Backend-2 libvirtd: 18:42:51.319: 25820: info : virDomainLoadAllConfigs:9950 : Scanning for configs in /etc/libvirt/lxc Nov 5 18:42:51 CI-Server-3-Backend-2 libvirtd: 18:42:51.320: 25820: info : umlStartup:431 : Adding inotify watch on /var/run/libvirt/uml-guest Nov 5 18:42:51 CI-Server-3-Backend-2 libvirtd: 18:42:51.321: 25820: info : virDomainLoadAllConfigs:9950 : Scanning for configs in /etc/libvirt/uml Nov 5 18:42:51 CI-Server-3-Backend-2 libvirtd: 18:42:51.321: 25820: error : main:3390 : Driver state initialization failed Nov 5 18:42:51 CI-Server-3-Backend-2 libvirtd: 18:42:51.321: 25821: warning : qemudDispatchSignalEvent:424 : Shutting down on signal 3 -- Stephen *"Kids these days.* *Whatever happened to hard work?"* -- Joel Spolsky, The perils of javaschools

1 0

[libvirt-users] no connection driver available for openvz:///system
by 张立洪 05 Nov '11

05 Nov '11

dears. I can not solved the problem: vzctl works. But how do I connect to Virtuozzo? Using the openvz >> > driver >> > it doesn't work out of the box: >> > virsh # connect openvz:///system >> > error: Failed to connect to the hypervisor >> > error: no connection driver available for openvz:///system Why can not I connect to the openvz??? please help me , thanks.

4 5

[libvirt-users] Event&Callback in QEMU
by huachao yao 04 Nov '11

04 Nov '11

Hi buddy, I wanna to monitor a domain's life cycle event, so i register a event an a virConnectDomainEventCallback, the code segment is as below: ------------------------ int DomainEventCallback(virConnectPtr conn, virDomainPtr dom, int event, int detail, void * opaque){ /*My code */ ....... } int main(){ /* .....*/ virDomainPtr dom = domain.getDomains("xp"); virFreeCallback freecb = NULL; void *opaque =NULL; virConnectDomainEventCallback cb =DomainEventCallback; virConnectDomainEventRegisterAny(conn, dom, VIR_DOMAIN_EVENT_ID_LIFECYCLE, VIR_DOMAIN_EVENT_CALLBACK(cb),opaque,freecb); /* .....*/ } but such a error happened: libvir: Remote error : this function is not supported by the connection driver: no event support and the libvirt version is 0.9; the driver is QEMU-KVM ,version is 0.14.0. I want to know why this Error will happen? Is the QEMU not support the LIFECYCLE Event ? If it is , how can i monitor the domain's life cycle event( such as started,defined and stopped..) Pls help~~

2 1

[libvirt-users] Cannot start VM after restart libvirt daemon
by Jean-Luc Derrien 04 Nov '11

04 Nov '11

Hello all, I'm using libvirt 0.8.3 on Debian Lenny (package provided in lenny-backports). After restarting libvirtd, I cannot start VM anymore. The daemon is restarted, but each time I try to start a domain, I get this message : virsh # start vm2 error: Failed to start domain vm2 error: internal error Process exited while reading console log output: When daemon is started with -v argument : 18:20:26.142: warning : qemudParsePCIDeviceStrs:1422 : Unexpected exit status '1', qemu probably failed 18:20:26.149: error : virExecDaemonize:782 : internal error Intermediate daemon process exited with status 0. 18:20:26.230: error : qemudReadLogOutput:2118 : internal error Process exited while reading console log output: In /var/log/syslog : Nov 3 18:20:26 ksm03 kernel: device vnet0 entered promiscuous mode Nov 3 18:20:26 ksm03 kernel: br0: port 2(vnet0) entering learning state Nov 3 18:20:26 ksm03 kernel: br0: port 2(vnet0) entering learning state Nov 3 18:20:26 ksm03 kernel: libvirtd[13599]: segfault at 41882000 ip 000000000043ac8b sp 000000004187e3b0 error 4 in libvirtd[400000+db000] Nov 3 18:20:26 ksm03 kernel: br0: port 2(vnet0) entering learning state Nov 3 18:20:26 ksm03 kernel: device vnet0 left promiscuous mode Nov 3 18:20:26 ksm03 kernel: br0: port 2(vnet0) entering disabled state When I reboot the server it works. Help welcome. Thanks in advance. Regards, Jean-Luc

1 0

[libvirt-users] general protection errors
by Alex 03 Nov '11

03 Nov '11

Hi, Some time ago I sent a message about some general protection errors, and didn't receive a response. I hoped I could post again and that someone might have some ideas. I really don't think it is a hardware problem producing these errors, but am otherwise out of ideas. I'm having a problem with a fedora15 x86_64 host with a fedora15 x86_64 guest on a Xeon E3-1240 with 8GB RAM on the guest. This is the only guest. The server is configured as a mail server, using spamassassin, postfix, amavisd, and clamav. Have there been significant changes in the current version from 0.8.8 that is shipped with fedora15? It was previously running without error on different hardware. I migrated the disks to a more powerful server due to load. It's a new server, and has been stress tested with memtest86, mprime, and passmark's Linux test for more than 24 hours without producing a problem. However, when I run the kvm vm, I receive a bunch of errors similar to these on the guest: [169245.360511] clamscan[27448] general protection ip:7f125f2e6ffb sp:7fff117566f0 error:0 in libclamav.so.6.1.11[7f125f229000+9ce000] [29016.445470] clamd[1110] general protection ip:30df2c3981 sp:7fffa08f4fe0 error:0 in libclamav.so.6.1 .11[30df200000+9ce000] There are other applications, such as pyzor, that produce similar errors. Is this something that could be a result of some software configuration error I've made? Is there something I can do to debug this further? There also aren't any errors on the host indicating a kernel problem or hardware issue. Is it possible this is a known error? I was previously trying to improve IO performance by using 'block' instead of 'file' for the device, but had to abandon that idea because it didn't work and I had to concentrate on this problem. Could disabling the cache (cache=none) cause this? I've included some info below, in hopes someone could guide me towards a solution. # rpm -qva|egrep 'kvm|libvirt|kernel|qemu' libvirt-python-0.8.8-7.fc15.x86_64 kernel-2.6.40.6-0.fc15.x86_64 qemu-img-0.14.0-8.fc15.x86_64 gpxe-roms-qemu-1.0.1-4.fc15.noarch kernel-2.6.40.4-5.fc15.x86_64 qemu-system-x86-0.14.0-8.fc15.x86_64 libvirt-0.8.8-7.fc15.x86_64 libvirt-client-0.8.8-7.fc15.x86_64 qemu-common-0.14.0-8.fc15.x86_64 kernel-headers-2.6.40.6-0.fc15.x86_64 qemu-kvm-0.14.0-8.fc15.x86_64 I've included my domain config below in hopes it helps. <domain type='kvm'> <name>mail02</name> <uuid>ec4f3cf5-2f27-fb3e-72f6-3fa3176b13b6</uuid> <memory>6291456</memory> <currentMemory>6291456</currentMemory> <vcpu>4</vcpu> <os> <type arch='x86_64' machine='pc-0.14'>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/bin/qemu-kvm</emulator> <disk type='file' device='disk'> <driver name='qemu' type='raw' cache='none'/> <source file='/var/lib/libvirt/images/mail02.img'/> <target dev='vda' bus='virtio'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x05' function='0x0'/> </disk> <disk type='block' device='cdrom'> <driver name='qemu' type='raw'/> <target dev='hdc' bus='ide'/> <readonly/> <address type='drive' controller='0' bus='1' unit='0'/> </disk> <controller type='ide' index='0'> <address type='pci' domain='0x0000' bus='0x00' slot='0x01' function='0x1'/> </controller> <interface type='bridge'> <mac address='52:54:00:67:2c:4c'/> <source bridge='br0'/> <model type='virtio'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x03' function='0x0'/> </interface> <serial type='pty'> <target port='0'/> </serial> <console type='pty'> <target type='serial' port='0'/> </console> <input type='mouse' bus='ps2'/> <graphics type='vnc' port='-1' autoport='yes'/> <video> <model type='cirrus' vram='9216' heads='1'/> <address type='pci' domain='0x0000' bus='0x00' slot='0x02' function='0x0'/> </video> <memballoon model='virtio'> <address type='pci' domain='0x0000' bus='0x00' slot='0x06' function='0x0'/> </memballoon> </devices> </domain> Thanks, Alex

1 0

[libvirt-users] auto-set /cgroup/*/libvirt/qemu/<vm name>/ directory permissions on creation
by Ben Clay 02 Nov '11

02 Nov '11

Is there a way to automatically set the permissions of cgroups directories created by libvirt when a VM is launched? Although it looks like I can set permissions on top-level /cgroup directories via cgconfig.conf, the cgconfig.conf manpage says "Permissions are related only to enclosing control group and are not inherited by subgroups.". From this, I believe that when libvirt creates VMs, the new directories under /cgroup/*/libvirt/qemu/<vm name>/ will not receive the cgconfig.conf permissions. Where in the libvirt config can I set this? We need users other than root to write to specific subsystems. I can ask on the libcg mailing list as well, but I thought I'd try here first since libvirt is dynamically creating / destroying these groups. Thanks. Ben Clay <mailto:rbclay@ncsu.edu> rbclay(a)ncsu.edu

2 1

[libvirt-users] CentOS5: libvirtd filling up /var/lib/libvirt/qemu/save
by Ben Clay 02 Nov '11

02 Nov '11

How can I disable libvirtd from attempting to suspend VMs on poweroff / reboot? It is slowly filling up our /var/ filesystem with save points at /var/lib/libvirt/qemu/save/<vm name>.save, even though I'm not initiating any suspend or restore functionality. I checked /etc/libvirt/libvirtd.conf and /etc/libvirt/qemu.conf but could not find anything. This is CentOS 5 with libvirt 0.8.2-22 and kvm-83-239. Thanks. Ben Clay <mailto:rbclay@ncsu.edu> rbclay(a)ncsu.edu

2 1

[libvirt-users] Network filesystem pool mount options
by Joe Auty 02 Nov '11

02 Nov '11

Hello, What are the default NFS mount options used by libvirt's network filesystem pool storage management, and can these be modified? -- Joe Auty, NetMusician NetMusician helps musicians, bands and artists create beautiful, professional, custom designed, career-essential websites that are easy to maintain and to integrate with popular social networks. www.netmusician.org <http://www.netmusician.org> joe(a)netmusician.org <mailto:joe@netmusician.org>

1 0

[libvirt-users] Error on the site
by Alexander Biryukov 02 Nov '11

02 Nov '11

Good afternoon. I apologize for my English. Error on page: http://libvirt.org/formatdomain.html To be more precise in the line "<target *def*='hdc' bus='ide'/>" -- Biryukov Alexander ICQ: 467093457 Skype: reddog_gad

2 1

[libvirt-users] Hard limit for the cpu usage of a VM
by sethuraman subbiah 01 Nov '11

01 Nov '11

Hi , I was previously using xen and currently moved to KVM. I am using libvirt to manage these VMs. In den's credit scheduler , I had the ability to set a cap on the cpu usage for a VM. But I was not able to find a similar substitute in KVM. I find that we can use cgroups to provide shares for VM but that will be more like weight based and it doesn't set a hard cap for that VM. I tried using cpulimit but I find it inaccurate and we can give values only between 0-100. Thus, I think it cannot support multi core environments. Can any one suggest a method to set a hard limit on a VM's cpu usage? Thank you. - Regards, Sethuraman Subbiah Graduate Student - NC state University M.S in Computer Science

3 9