- Users - libvirt List Archives

[libvirt-users] Problem after restarting libvirt
by Nikita A Menkovich 09 May '11

09 May '11

After restarting libvirt cgroups for VMs are not recovered. Is there available to restore cgroup state after restart? -- Nikita A Menkovich http://libc6.org/ JID: menkovich(a)gmail.com Tel: +7 (921) 423-96-48

2 1

[libvirt-users] Objective: run commands in the domain. Trouble: setting up a console
by Guilherme Santos 09 May '11

09 May '11

I've installed libvirt and also Virtual Box OSE hypervisor, in order to see if I could properly manage it through virsh. I could easily set up, start, pause, resume, shutdown and destroy the newly created domains, from a XML file. Then I was wondering how could I be able to spawn commands in the domain through virsh, and the command 'console' gave me a the thought that I should connect to the domain's console, so that, once connected, I would be able to "do anything". Am I right? I need to be able to run commands in the virtual machine. If the only way is by accessing the domain's console, how can I properly set up one? I tried several ways, following the examples in the documentation. Problem is that no matter what I wrote in the xml file about serial or console devices, it aways get overriden by null configuration when I start the domain (I can see it by using command 'edit' <domain> ). So then I tried to edit the XML through virsh and insert specifications about console or serial devices, but I never was able to sucessfully edit the file. I always got the same error message: "could not define a domain, rc=80bb0004", although the domain is in fact defined! I tried editing the XML both with the domain running and stopped. My final alternative was using command 'attach-device' passing a xml containing only the specifications of serial or console devices. I aways received an error message saying that memory limit was exceeded, although I strongly suppose that's not the real problem. I am sure that I am configuring something (or many things) wrongly, so that I am getting all those errors... My initial problem is just to be able to run commands in the newly created domain. Anybody can tell me an easier way to do this? If setting up consoles is the only way, what is going on wrongly for me? Sorry if an answer for this issue is already somewhere in the documentation, I tried to find it hard before emailing you guys! Thanks anyway!! -- Guilherme Santos G. Baptista Graduando em Ciência da Computação pela UFCG LSD - Laboratório de Sistemas Distribuídos

2 1

[libvirt-users] kvm libvirt migration problem
by 潘昊 06 May '11

06 May '11

Hi everyone: I am new here I am using kvm and libvirt on my Dell server. Now i am trying to migrate one virtual machine from a physical server to another. However, I failed everytime. In virsh on physicalServer1, I typed: virsh # migrate virtualmachine1 qemu+ssh://username@physicalServer2/system error: operation failed: migration to 'tcp: physicalServer2:49163' failed: migration failed Then I searched FAQ part on libvirt.org http://wiki.libvirt.org/page/FAQ it says: * error: operation failed: migration to '...' failed: migration failed * This is an error often encountered when trying to migrate with QEMU/KVM. This typically happens with plain migration, when the source VM cannot connect to the destination host. You will want to make sure your hosts are properly configured for migration (see the migration section of this FAQ) I managed to ssh physicalServer2 in virtualmachine1 shell. So the above red part did not explain my failure. I also open ports on physicalServer2, iptables -L shows following information: Chain INPUT (policy ACCEPT) target prot opt source destination ACCEPT udp -- anywhere anywhere udp dpt:domain ACCEPT tcp -- anywhere anywhere tcp dpt:domain ACCEPT udp -- anywhere anywhere udp dpt:bootps ACCEPT tcp -- anywhere anywhere tcp dpt:bootps ACCEPT udp -- anywhere anywhere udp dpt:domain ACCEPT tcp -- anywhere anywhere tcp dpt:domain ACCEPT udp -- anywhere anywhere udp dpt:bootps ACCEPT tcp -- anywhere anywhere tcp dpt:bootps ACCEPT tcp -- anywhere anywhere state NEW tcp dpts:49152:49215 Chain FORWARD (policy ACCEPT) target prot opt source destination ACCEPT all -- anywhere 192.168.122.0/24 state RELATED,ESTABLISHED ACCEPT all -- 192.168.122.0/24 anywhere ACCEPT all -- anywhere anywhere REJECT all -- anywhere anywhere reject-with icmp-port-unreachable REJECT all -- anywhere anywhere reject-with icmp-port-unreachable ACCEPT all -- anywhere 192.168.122.0/24 state RELATED,ESTABLISHED ACCEPT all -- 192.168.122.0/24 anywhere ACCEPT all -- anywhere anywhere REJECT all -- anywhere anywhere reject-with icmp-port-unreachable REJECT all -- anywhere anywhere reject-with icmp-port-unreachable Chain OUTPUT (policy ACCEPT) target prot opt source destination The /var/log/libvirt/qemu/virtualmachine1.log on physicalServer2: 2011-05-06 13:37:30.708: starting up LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/usr/bin:/usr/sbin:/sbin:/bin QEMU_AUDIO_DRV=none /usr/bin/kvm -S -M pc-0.14 -enable-kvm -m 2048 -smp 1,sockets=1,cores=1,threads=1 -name openjudge-test -uuid a8c704bc-a4f9-90db-3e57-40e60b00aac1 -nodefconfig -nodefaults -chardev socket,id=charmonitor,path=/var/lib/libvirt/qemu/virtualmachine1.monitor,server,nowait -mon chardev=charmonitor,id=monitor,mode=readline -rtc base=utc -boot c -drive file=/media/nfs/virtualmachine1.img,if=none,id=drive-ide0-0-0,format=raw -device ide-drive,bus=ide.0,unit=0,drive=drive-ide0-0-0,id=ide0-0-0 -drive if=none,media=cdrom,id=drive-ide0-1-0,readonly=on,format=raw -device ide-drive,bus=ide.1,unit=0,drive=drive-ide0-1-0,id=ide0-1-0 -netdev tap,fd=20,id=hostnet0 -device rtl8139,netdev=hostnet0,id=net0,mac=00:16:36:8a:22 :a0,bus=pci.0,addr=0x3 -chardev pty,id=charserial0 -device isa-serial,chardev=charserial0,id=serial0 -usb -vnc 127.0.0.1:2 -vga cirrus -incoming tcp:0.0.0.0:49163 -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x4 char device redirected to /dev/pts/0 2011-05-06 13:37:30.915: shutting down The /var/log/libvirt/qemu/virtualmachine1.log on physicalServer1 is empty. Both physical servers are using Ubuntu 11.04. The libvirt and kvm used are installed by apt-get. The libvirt version is 0.8.8. Is there anyone can help me? Thanks for help! -- Best wishes, -Panda

1 0

[libvirt-users] Error: unable to connect to '/var/run/libvirt/libvirt-sock'
by guan qin 05 May '11

05 May '11

Hi, I meet with the problem that "Error: unable to connect to '/var/run/libvirt/libvirt-sock' ". I do as Error: unable to connect to '/var/run/libvirt/libvirt-sock' <http://wiki.libvirt.org/page/FAQ#Error:_unable_to_connect_to_.27.2Fvar.2Fru…>said but never worked: root@shahe26:~# virsh error: unable to connect to '/var/run/libvirt/libvirt-sock', libvirtd may need to be started: No such file or directory error: failed to connect to the hypervisor root@shahe26:~# ls /usr/lib64 | grep libvirt libvirt libvirt.a libvirt.la libvirt-qemu.a libvirt-qemu.la libvirt-qemu.so libvirt-qemu.so.0 libvirt-qemu.so.0.8.3 libvirt-qemu.so.0.9.0 libvirt.so libvirt.so.0 libvirt.so.0.8.3 libvirt.so.0.9.0 root@shahe26:~# ls /usr/lib32 | grep libvirt root@shahe26:~# /etc/init.d/libvirt-bin restart Restarting libvirt management daemon: /usr/sbin/libvirtd. root@shahe26:~# virsh error: unable to connect to '/var/run/libvirt/libvirt-sock', libvirtd may need to be started: No such file or directory error: failed to connect to the hypervisor root@shahe26:~# I install the libvirt as follows: #./configure --prefix=/usr --libdir=/usr/lib64 #make #make install my OS is debian6.0.1a: root@shahe26:~# uname -a Linux shahe26 2.6.32 #1 SMP Wed Apr 27 21:52:04 PDT 2011 x86_64 GNU/Linux root@shahe26:~# How to solve this problem?thanks. best regards, qinguan

2 2

[libvirt-users] Listing machines as non-root user
by Maciej Gałkiewicz 05 May '11

05 May '11

I have a problem with virsh. I'm using eucalyptus software on top of xen 4 (debian squeeze). The problem is that unix user "eucalyptus" cannot see vm machines when it executes "virsh list". My libvirt version is 0.8.8-3 from debian sid repository. The same problem occurs with older versions. What is more I have tried to change unix_sock_ro_perms to "2777". After libvirt restart it started to work. Even when I reverted this change it was still working. Unfortunately after server restart virsh still does not work (I am not sure but I cannot restart my production servers right now). Everything is fine when I run virsh as root. I have the same issue on 7 machines. regards M. Galkiewicz

2 7

[libvirt-users] Release of libvirt-0.9.1
by Daniel Veillard 05 May '11

05 May '11

As planned and after most of the clang detected problems got fixed (thanks Eric !) the new release is available at: ftp://libvirt.org/libvirt/ It's a mixed release, it includes a number of improvements as well as many bug fixes and a few new features: Features: - support various persistent domain updates (KAMEZAWA Hiroyuki) - improvements on memory APIs (Taku Izumi) - Add virDomainEventRebootNew (Matthias Bolte) - various improvements to libxl driver (Markus Groß) - Spice: support audio, images and stream compression (Michal Privoznik) Documentation: - fix missing VLAN id for Qbg example (Gerhard Stenzel) - docs: Document filesystem tag device (Cole Robinson) - maint: fix comment typos (Eric Blake) - Fix two out-of-date comments in LVM backend (Richard Laager) - docs: Serial and parallel device target ports actually start from 0 (Matthias Bolte) - maint: fix grammar errors (Eric Blake) - docs: document freecell --all (Eric Blake) - docs: remove "returns" word from beginning of lines (Jean-Baptiste Rouault) - docs: add an IPv6 address to network XML examples (Laine Stump) Portability: - build: fix getcwd portability problems (Eric Blake) - build: avoid test warnings on mingw (Eric Blake) - virsh: avoid compiler warning on mingw (Eric Blake) - build: Use pkg-config for libssh2 check (Jiri Denemark) - build: Ignore old audit library (Jiri Denemark) - build: fix 32-bit test failure (Eric Blake) - build: Fix problem of building Python bindings (Osier Yang) - build: fix 32-bit test failure (Eric Blake) - ppc: Enable starting of Qemu VMs on ppc host (Stefan Berger) - Migrate VMs between different-endianess hosts (Stefan Berger) - build: really fix mingw startup (Eric Blake) - build: fix mingw build (Eric Blake) - do not build libvirt_iohelper when building without libvirtd (Wen Congyang) - virsh: fix mingw startup (Eric Blake) - build: avoid compiler warning on cygwin (Eric Blake) - Fix build for older gcc (Jim Fehlig) - Don't try to enable stack protector on Win32 (Daniel P. Berrange) Bug Fixes: - storage: avoid null deref and leak on failure (Eric Blake) - esx: Avoid null dereference on error in esxDomainGetInfo (Matthias Bolte) - remote: avoid null dereference on error (Eric Blake) - cgroup: avoid leaking a file (Eric Blake) - virsh: avoid null pointer dereference (Eric Blake) - pci: fix null pointer dereference (Eric Blake) - qemu: avoid null pointer dereference (Eric Blake) - tests: avoid null pointer dereference (Eric Blake) - Fix disability to run on systems with no PCI bus (Michal Privoznik) - virsh: fix regression in log to file (Supriya Kannery) - nwfilter: Fix memory leak in the ebtables subdriver (Matthias Bolte) - qemu: Fix qemuDomainModifyDeviceFlags leaking the caps bitmap (Matthias Bolte) - Fix memory leak in __virExec (Matthias Bolte) - hash: fix memory leak regression (Eric Blake) - Xen: Do not generate net ifname if domain is inactive (Jim Fehlig) - xen: check if device is assigned to guest before reattaching (Yufang Zhang) - util: Initialize hooks at daemon shutdown if no hooks defined (Osier Yang) - fix virsh's regression (Wen Congyang) - release PCI address only when we have ensured it successfully (Wen Congyang) - free memory properly in cleanup patch (Hu Tao) - free buf content when vsnprintf() failed (Wen Congyang) - esx: Fix dynamic dispatch for CastFromAnyType functions (Matthias Bolte) - Make crash and live flags mutually exclusive in virDomainCoreDump (Mark Wu) - Fix small memory leaks in config parsing related functions (Matthias Bolte) - daemon: Don't try to free an unsigned int in error paths (Matthias Bolte) - Fix QEMU tunnelled migration FD handling (Daniel P. Berrange) - Remove artificial minimum limit for guest memory (Jiri Denemark) - fix tunnelled migration's regression (Wen Congyang) - qemu: fix the check of virDomainObjUnref()'s return value (Wen Congyang) - Fix checking of return codes in dispatcher (Daniel P. Berrange) - qemu: avoid qemu_driver being unlocked twice when virThreadPoolNew() failed (Wen Congyang) - virsh: fix regression in parsing optional integer (Eric Blake) - phyp: avoid memory leaks in command values (Eric Blake) - phyp: avoid memory leak on failure (Eric Blake) - phyp: avoid a logic bug (Eric Blake) - free cpumask of vcpupinDef (Hu Tao) - network: Fix NULL dereference during error recovery (Michal Privoznik) - Add missing checks for QEMU domain state in tunables APIs (Daniel P. Berrange) - phyp: Fix too small buffer allocation in phypAttachDevice (Matthias Bolte) - phyp: Don't overwrite error from virDomainDeviceDefParse by OOM error (Matthias Bolte) - phyp: Don't try to use a string from a failed virAsprintf (Matthias Bolte) - phyp: Remove stack allocating a 4kb volume key and fix related memory leaks (Matthias Bolte) - qemu: fix a dead-lock problem (Hu Tao) - util: Fix crash when removing entries during hash iteration (Jiri Denemark) - Fix possible infinite loop in remote driver (Michal Privoznik) - qemu: Remove the managed state file only if restoring succeeded (Osier Yang) - reattach pci devices when qemuPrepareHostdevPCIDevices() failed (Wen Congyang) - reattach pci device when pciBindDeviceToStub() failed (Wen Congyang) - remove devices from driver activePciHostdevs when qemuPrepareHostdevPCIDevices() failed (Wen Congyang) - pci: avoid invalid free, init path to NULL (Wen Congyang) - qemu: Support for overriding NPROC limit (Jiri Denemark) - qemu: Always reserves slot 0x02 for primary VGA. (Osier Yang) - Fix typo in systemtap tapset directory name (Daniel P. Berrange) - qemu: Ignore unusable binaries (Jiri Denemark) Improvements: - maint: detect clang 2.9 (Eric Blake) - qemu: update qemuCgroupControllerActive signature (Eric Blake) - lxc: report correct error (Eric Blake) - libxl: avoid compiler warning (Eric Blake) - storage: use virCommand to avoid compiler warning (Eric Blake) - tests: Lower stack usage below 4096 bytes (Matthias Bolte) - tests: Update valgrind suppressions file (Matthias Bolte) - tests: simplify common setup (Eric Blake) - network: fix return value of hostsFileWrite (Laine Stump) - libvirt/qemu - support persistent attach/detach disks (KAMEZAWA Hiroyuki) - libvirt/qemu - support persistent modification of devices (KAMEZAWA Hiroyuki) - util: Simplify hash implementation (Jiri Denemark) - tests: More unit tests for internal hash APIs (Jiri Denemark) - Move call to virReportOOMError into virFileBuildPath (Matthias Bolte) - build: use gnulib passfd for simpler SCM_RIGHTS code (Eric Blake) - qemu: Add flags checking in DomainCoreDump (Jiri Denemark) - Make sure DNSMASQ_STATE_DIR exists (Guido Günther) - build: make VIR_FREE do some type checking (Eric Blake) - threads: add one-time initialization support (Eric Blake) - esx: Add a wrapper for shared CURL handles (Matthias Bolte) - esx: Move CURL handling code to it's own type (Matthias Bolte) - daemon: Honor error variable name change in the generator (Matthias Bolte) - libvirt/qemu - clean up UpdateDevice for consolidation. (KAMEZAWA Hiroyuki) - libvirt/qemu - clean up At(De)tachDeviceFlags() for consolidation. (KAMEZAWA Hiroyuki) - libvirt/qemu - Centralize device modification in the more flexible APIs (KAMEZAWA Hiroyuki) - Add support for s390(x) cpu options parsing (Michal Privoznik) - PHYP: Adding reboot domain function (Eduardo Otubo) - nwfilter: no support for direct type of interface (Stefan Berger) - Add cputune support to libxl driver (Markus Groß) - Update and sort msg_gen_function list and mark unmarked messages (Matthias Bolte) - Write error check conditionals in more compact form for dispatcher (Daniel P. Berrange) - Merge all returns paths from dispatcher into single path (Daniel P. Berrange) - Experimental libvirtd upstart job (Alan Pevec) - enhance processWatchdogEvent() (Wen Congyang) - xen: Replace statsErrorFunc with a macro (Matthias Bolte) - tests: Unit tests for internal hash APIs (Jiri Denemark) - build: include esx_vi.generated.* into dist file (Wen Congyang) - tests: test recent virsh option parsing changes (Eric Blake) - virsh: list required options first (Eric Blake) - phyp: use consistent return string handling (Eric Blake) - maint: use lighter-weight function for straight appends (Eric Blake) - libvirt-guests: implement START_DELAY (Alexander Todorov) - network: truncate bridges' dummy tap device names to IFNAMSIZ (15) chars (Laine Stump) - Introduce virDomainChrDefNew() (Michal Novotny) - esx: Make the parsed URI part of the private connection data (Matthias Bolte) - esx: Mark error message in macros for translation (Matthias Bolte) - esx: Extend VI generator to cover managed object types (Matthias Bolte) - esx: Cleanup and refactor CastFromAnyType macros (Matthias Bolte) - esx: Cleanup VI generator code (Matthias Bolte) - phyp: Reduce code duplication in error and success paths (Matthias Bolte) - Replace REMOTE_DEBUG with VIR_DEBUG in daemon dispatcher (Daniel P. Berrange) - Add missing checks for whether the connection is open in dispatcher (Daniel P. Berrange) - PHYP: Adding network interface (Eduardo Otubo) - xen: Remove PATH_MAX sized stack allocation from block stats code (Matthias Bolte) - setmaxmem: add the new options to "virsh setmaxmem" command (Taku Izumi) - maxmem: implement virDomainSetMaxMemory API of the qemu driver (Taku Izumi) - maxmem: introduces VIR_DOMAIN_MEM_MAXIMUM flag (Taku Izumi) - nwfilters: support for TCP flags evaluation (Stefan Berger) - setmem: add --current option to virsh setmem command (Taku Izumi) - setmem: add VIR_DOMAIN_MEM_CURRENT support to qemu (Taku Izumi) - setmem: introduce VIR_DOMAIN_MEM_CURRENT flag (Taku Izumi) - build: Install libxenlight log dir (Jim Fehlig) - Add domainSet/GetSchedulerParameters to libxl driver (Markus Groß) - rename pciUnBindDeviceFromStub() to pciUnbindDeviceFromStub() and float it up (Wen Congyang) - Add domainIsUpdated to libxl driver (Markus Groß) - Change locking for udev monitor and callbacks (Serge Hallyn) - Improve SCSI volume key generation (Daniel P. Berrange) - qemu: Rewrite LOOKUP_PTYS macro into a function (Jiri Denemark) - Enable use of -Wold-style-definition compiler flag (Daniel P. Berrange) - Enable use of -Wmissing-noreturn (Daniel P. Berrange) - Enable -Wmissing-format-attribute warning (Daniel P. Berrange) - Use gnulib's manywarnings and warnings modules (Daniel P. Berrange) - Use virBufferPtr for sexpr2string instead of manual buffer handling (Matthias Bolte) - xend: Remove 4kb stack allocation (Matthias Bolte) - uml: Remove PATH_MAX sized stack allocation from /proc parsing code (Matthias Bolte) - storage: Remove PATH_MAX sized stack allocation from iSCSI backend (Matthias Bolte) - qemu: Remove PATH_MAX sized stack allocation used in commandline building (Matthias Bolte) - Remove PATH_MAX sized stack allocation from virFileOpenTtyAt (Matthias Bolte) - openvz: Remove several larger stack allocations (Matthias Bolte) - daemon: Remove 4kb stack allocation of security label (Matthias Bolte) - virsh: Remove two 4kb stack allocations (Matthias Bolte) - Use virFileAbsPath instead of manually creating the absolute path (Matthias Bolte) - xenxs: Remove PATH_MAX sized stack allocation in XM script parsing (Matthias Bolte) - sasl: Remove stack allocated 8kb temporary buffers (Matthias Bolte) - qemu: Use heap allocated memory to read the monitor greeting (Matthias Bolte) - phyp: Remove 16kb stack allocation (Matthias Bolte) - virt-aa-helper: Remove PATH_MAX sized stack allocations (Matthias Bolte) - ebtables: Remove PATH_MAX sized stack allocation (Matthias Bolte) - pci: Remove PATH_MAX sized stack allocations (Matthias Bolte) - Remove PATH_MAX sized stack allocations related to virFileBuildPath (Matthias Bolte) - vmx: Use case-insensitive compare functions for all content (Matthias Bolte) - vmx: Support persistent CPU shares (Matthias Bolte) - Add autostart support to libxl driver (Markus Groß) - Allow relative path for qemu backing file (Jesse Cook) - build: detect potentential uninitialized variables (Eric Blake) Cleanups: - esx: Disable performance counter queries in esxDomainGetInfo (Matthias Bolte) - esx: Remove dead store in esxUtil_ParseDatastorePath (Matthias Bolte) - util: remove dead assignment (Eric Blake) - qemu: remove dead assignment (Eric Blake) - qemu: silence clang false positives (Eric Blake) - tests: suppress more valgrind situations (Eric Blake) - tests: avoid compiler warning (Eric Blake) - qemu: fix uninitialized variable warning (Christophe Fergeau) - build: fix syntax-check failure (Eric Blake) - virsh: nuke use of TRUE and FALSE (Eric Blake) - phyp: another simplification (Eric Blake) - maint: ignore built file (Eric Blake) - Change some variable names to follow standard in daemon dispatcher (Daniel P. Berrange) - Remove curly braces on all single-line conditional jumps in dispatcher (Daniel P. Berrange) - Remove virConnectPtr from virRaiseErrorFull (Matthias Bolte) - phyp: prefer memcpy over memmove when legal (Eric Blake) - phyp: use consistent style for labels (Eric Blake) - phyp: more return handling cleanup (Eric Blake) - esx: Fix gcc 4.6 warning about initialized but unused variables (Matthias Bolte) - maint: silence cppi warnings (Eric Blake) - Fix gcc 4.6 warnings in vbox_tmpl.c (Christophe Fergeau) - Fix gcc 4.6 warnings (Christophe Fergeau) - Remove C99 variable declare in PHYP network driver (Daniel P. Berrange) - Standard on error variable name in libvirtd dispatcher (Daniel P. Berrange) - Remove all whitespace before function brackets in daemon dispatcher (Daniel P. Berrange) - docs: tweak virsh restore warning (Eric Blake) - setmaxmem: remove the code to invoke virDomainSetMemory in cmdSetmaxmem (Taku Izumi) - build: fix gitignore sorting (Eric Blake) - tests: fix recent test failures (Eric Blake) - libxl: avoid compiler warning (Eric Blake) - Avoid compiler warnings about int to void * casts (Daniel P. Berrange) - Remove acinclude.m4 file (Daniel P. Berrange) - Remove possible uninitialized variable in openvz driver (Daniel P. Berrange) Thanks to everybody who helped with this release in any way, bug reports, documentation, patches or features ! Daniel -- Daniel Veillard | libxml Gnome XML XSLT toolkit http://xmlsoft.org/ daniel(a)veillard.com | Rpmfind RPM search engine http://rpmfind.net/ http://veillard.com/ | virtualization library http://libvirt.org/

1 0

[libvirt-users] virt-install becoming a headache
by Ricard Daniel Santafe Bence 04 May '11

04 May '11

2 1

[libvirt-users] can't up the eth in the vm and access to network
by guan qin 04 May '11

04 May '11

Hi, I encounter a problem that the eth can't be up in the VM by using VT-d. I boot the VM using "virsh create vt_d.xml". The VM could recongize the NIC which was assigned .then I do in VM as follows: #cat /sys/class/net/eth0/operstate down #ifconfig eth0 up #cat /sys/class/net/eth0/operstate down #ficonfig eth 10.100.0.5 #cat /sys/class/net/eth0/operstate down more system information is attached.(uname 、ifconfig 、lspci、cmdline...) I can't figure out the reason why the eth is still down? can anyone help me ? best regards, qinguan vt_d.xml: <domain type='kvm'> <name>test</name> <description>just test using sriov!</description> <os> <type>hvm</type> <boot dev='hd'/> </os> <memory>524288</memory> <currentMemory>524288</currentMemory> <on_poweroff>destroy</on_poweroff> <on_reboot>restart</on_reboot> <on_crash>restart</on_crash> <features><pae/><acpi/><apic/></features> <clock offset='utc'/> <devices> <input type='mouse' bus='usb'/> <disk type='file' device='disk'> <source file='/home/qinguan/exp/ubuntu_1.img'/> <target dev='hda' bus='virtio'/> </disk> <emulator>/usr/bin/kvm</emulator> <graphics type='vnc' port='-1'/> <hostdev mode='subsystem' type='pci' managed='yes'> <source> <address bus='0x0' slot='0x19' function='0x0'/> </source> </hostdev> </devices> </domain>

4 8

[libvirt-users] LibVirtD Networking
by ssaleh＠nrln.us 03 May '11

03 May '11

I'm looking for the best way to setup networking support for VM's. I have a small subnet with 6 routable IP's I can use on the Internet. So, I'd like to setup a host and have VM's on the public IP's. The computer I have has 4 NICs. I'd like to use one for a private 10.x.x.x and the others for the public facing VM's. Can I support 5 VM's with the three raminging NIC's? Should I use virtual IP's for the 5 VM's? Should I use a single NIC for all 5 VM's (similar to what VMWare does)? What is the best way to newtork this solution? Thanks in advance.

2 1

[libvirt-users] [Libvir] libvirt & vde_switch
by Robert Šmol 02 May '11

02 May '11

Hi, I would like to ask about status of VDE and libvirt. I found a thread from http://www.redhat.com/archives/libvir-list/2008-February/msg00030.html. I am not interested in connecting the VDE to the outside world. I have created a FW machine using macvtap. To this machine I would like to attach machines from different zones (dmz, prod, cust). I was thinking on using VDE and vlans (FW would have 1 interface in each vlan - machines also), but I am not aware how to configure this in libvirt. I now assume I can do the same setup with regular bridges for the moment (benefit of VDE would be only 1 switch for many vlans). Regards, Robert -- stereoIT - Open Source Technology Partners http://www.stereoit.com mobile: +420 776 762378

4 9

[libvirt-users] port forwarding
by Ireneusz Szcześniak 28 Apr '11

28 Apr '11

Hi, I'm using CentOS release 5.6 (Final) and Xen 3.1. I'm writing to ask how to configure port forwarding. With virt-manager 0.6.1 I added the machines, but there I don't see an option for port forwarding. I would like to reach the VM on a specific port of the host machine. Once the machinces are running, I can configure iptables so that the port forwarding works, but after host reboots, other rules are inserted (put in front of my rules), which disable my rules. I guess these rules are put by libvirt, and so I'm writing to this list. These are the rules inserted: Chain FORWARD (policy ACCEPT) target prot opt source destination ACCEPT all -- anywhere 192.168.122.0/24 state RELATED,ESTABLISHED ACCEPT all -- 192.168.122.0/24 anywhere ACCEPT all -- anywhere anywhere REJECT all -- anywhere anywhere reject-with icmp-port-unreachable REJECT all -- anywhere anywhere reject-with icmp-port-unreachable I need to have a different rules for 192.168.122.0/24: the states RELATED,ESTABLISHED are fine, but I need the NEW state put there too, so that the new forwards can go through. Thanks, Irek -- Ireneusz (Irek) Szczesniak http://www.irkos.org

3 4

[libvirt-users] Disk not bootable from yesterday resolved
by David Brown 27 Apr '11

27 Apr '11

Just an update and a thanks. The not bootable issue I was having yesterday ended up being for some odd reason the disk type in the domain define xml file had been switched to the disk being "raw" rather than qcow2 format. When I changed:<driver name='qemu' type='raw'/> to <driver name='qemu' type='qcow2'/> It booted without issue. I'm not clear how that got changed on the original domain. It was in use without issue about a week ago. :/ I then ran the clone again and it cloned fine, but had the boot issueagain. I checked the cloned domain.xml file and again it said typeraw. I changed it again to qcow2 and it booted fine. Bug? Shouldn't it keep the type in the config the same? Anyway, thanks for help and especially the quick responses! Dave

2 2

[libvirt-users] problem installing libvirt from source
by Lars Nordin 27 Apr '11

27 Apr '11

I'm trying to compile and install libvirt-0.9.0 on Ubuntu v10.04. The make and install went fine and I don't see any errors but there isn't a libvirt directory created in /usr/local/lib/python2.6/dist-packages. I specified configure --with-vmware -with-esx (and later --with-python), then make, then sudo make install. >From config.log: prefix = /usr/local exec_prefix = ${prefix} pyexecdir = ${exec_prefix}/lib/python2.6/dist-packages pythondir = ${prefix}/lib/python2.6/dist-packages pkgpyexecdir = ${pyexecdir}/libvirt pkgpythondir = ${pythondir}/libvirt

3 3

Re: [libvirt-users] virt-clone: clone not bootable
by David Brown 27 Apr '11

27 Apr '11

It says: booting from hard disk Booting failed device not bootable FATAL: failed to boot from device. Or something very similar. Whit Blauvelt <whit.virt(a)transpect.com> wrote: >On Tue, Apr 26, 2011 at 01:45:23PM -0700, David Brown wrote: >> The clone was made with the following command: virt-clone -o centos -n >> centosclone -f /path/to/new/image.img >> I've attempted it several times and get the same result. >> Has anyone seen this or know how to resolve it? > >I've often cloned CentOS 5.5 KVM VMs with that same invocation, and no >problem. > >When you say it's "not bootable," what exactly do you see at time of >failure? > >Whit

3 2

[libvirt-users] virt-image issue
by Zvi Dubitzky 27 Apr '11

27 Apr '11

Hi Does anybody work with the virt-image utility. - I do not see where/how one specifies the pre-defned image or the CD iso image to produce the image from - what is the .xml file requested for ? I guess it is the output xml file I just used the command with an existing .xml file (thought the utility will modify it according to the specified flags) but got the error message : "ERROR Cannot parse 'vmTest3.xml': Root element is not 'image" Any idea what is missing ? thanks Zvi Dubitzky Email:dubi@il.ibm.com

2 1

[libvirt-users] virt-clone: clone not bootable
by David Brown 26 Apr '11

26 Apr '11

Hi, Not sure if this is the right mailing list, but... I have a Linux VM (CentOS 5.4) VM. I cloned it with virt-clone, but the clone isn't bootable. When I attempt to boot it says: Booting from hard diskBoot failed: not a bootable diskFATAL: No bootable device The clone was made with the following command: virt-clone -o centos -n centosclone -f /path/to/new/image.img I've attempted it several times and get the same result. Has anyone seen this or know how to resolve it? Thanks,Dave

2 1

[libvirt-users] Adding more disks...
by Dale Amon 26 Apr '11

26 Apr '11

I need a VM with about 6-8 qcow2 disk images, but it seems to only allow 4 disks on the ide, like the real hardware. What lines do you need to add on to the xml definition file to get hde, hdf, hdg, etc?

2 2

[libvirt-users] How to providing virtual USB disk (usb-storage for QEMU) devices to guests?
by oPKo o 26 Apr '11

26 Apr '11

Hi, I am trying to provide virtual USB disk devices to guests but without success. With QEMU is simple to plug the virtual USB disk with the command: (qemu) drive_add 0 id=my_usb_disk,if=none,file=usbdisk.img OK (qemu) device_add usb-storage,id=my_usb_disk,drive=my_usb_dis I can do this with libvirt? PK PS: sorry for my poor english

1 0

[libvirt-users] problem with qemu
by Miroslav Ludvik 25 Apr '11

25 Apr '11

Hi All, I use Ubuntu server 10.04 LTS as virtualization platform. Actually running kernel 2.6.32-31-server #61-Ubuntu S root@jupiter:~# uname -a Linux jupiter 2.6.32-31-server #61-Ubuntu SMP Fri Apr 8 19:44:42 UTC 2011 x86_64 GNU/Linux We have on running virtual root@jupiter:~# virsh list --all Id Name State ---------------------------------- 1 kvmtik.4safety.cz running I run script for create virtual system wrote root@jupiter:~# make-hestia.4safety.cz.sh 2011-04-24 21:13:25,613 ERROR : ubuntu-kvm already exists Traceback (most recent call last): File "/usr/bin/vmbuilder", line 24, in <module> cli.main() File "/usr/lib/python2.6/dist-packages/VMBuilder/contrib/cli.py", line 81, in main raise VMBuilderUserError('%s already exists' % destdir) VMBuilder.exception.VMBuilderUserError: ubuntu-kvm already exists /usr/local/bin/make-hestia.4safety.cz.sh: line 29: --libvirt=qemu:///system: No such file or directory I try restart libvirt and system wrote to syslog root@jupiter:~# /etc/init.d/libvirt-bin restart Rather than invoking init scripts through /etc/init.d, use the service(8) utility, e.g. service libvirt-bin restart Since the script you are attempting to invoke has been converted to an Upstart job, you may also use the restart(8) utility, e.g. restart libvirt-bin libvirt-bin start/running, process 25202 root@jupiter:~# tail -f /var/log/syslog Apr 24 20:25:01 jupiter CRON[25031]: (root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1) Apr 24 20:35:01 jupiter CRON[25036]: (root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1) Apr 24 20:45:01 jupiter CRON[25042]: (root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1) Apr 24 20:55:01 jupiter CRON[25048]: (root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1) Apr 24 21:05:01 jupiter CRON[25152]: (root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1) Apr 24 21:15:01 jupiter CRON[25183]: (root) CMD (command -v debian-sa1 > /dev/null && debian-sa1 1 1) Apr 24 21:17:01 jupiter CRON[25189]: (root) CMD ( cd / && run-parts --report /etc/cron.hourly) Apr 24 21:17:58 jupiter libvirtd: 21:17:58.089: warning : qemudDispatchSignalEvent:385 : Shutting down on signal 15 Apr 24 21:17:58 jupiter libvirtd: 21:17:58.617: error : udevStrToLong_ui:73 : Failed to convert '008' to unsigned int#012 Apr 24 21:17:59 jupiter kernel: [106086.771154] lo: Disabled Privacy Extensions Than for your help. Mirek

2 1

[libvirt-users] Media autoremove
by Nikita A Menkovich 25 Apr '11

25 Apr '11

I have an iso, than automate freebsd install, and after installation it reboot VM. I want to remove cdrom drive with ISO from VM config after reboot? How I could do this? -- Nikita A Menkovich http://libc6.org/ JID: menkovich(a)gmail.com

2 1

[libvirt-users] Setup Mulitple VM's
by ssaleh＠nrln.us 24 Apr '11

24 Apr '11

I have a small public IP block (/29) I'd like to populate with VM's from a single physical box running CentOS/5. I'd like to get some thoughts on the best way to setup networking to support this. I'm OK with setting up the virtual hosts. Any advice would be appreciated. TIA...

2 1

[libvirt-users] vvirsh memtune
by Zvi Dubitzky 24 Apr '11

24 Apr '11

Hi Does anybody use the 'virsh memtune' command and can he say what combination of libvirt version and libcgroup does he use thanks Zvi Dubitzky Email:dubi@il.ibm.com

1 0

[libvirt-users] 2 hosts, mirrored VM storage, avoiding conflicting VM instances
by Whit Blauvelt 23 Apr '11

23 Apr '11

Hi, Using libvirt for KVM, experimenting with live migration with the VM on storage shared in a mirrored GlusterFS or an active-active DRBD setup where QEMU/KVM runs on the hosts on both sides of the mirror, the live migration works great. What concerns me is finding a way if possible to prevent a situation where a particular VM is running on both hosts at once. Is there any way within libvirt in this sort of setup to start a VM on mirrored storage from a KVM host on one side of the mirror _only if_ the same VM isn't already running on the other side of the mirror? In a scripting context I can see how to do that check before starting, but not how to prevent virsh or virt-manager from being used to create this bad condition when they're being used directly. Is there some way to set up protection from this within virsh and virt-manager? Or am I into feature request territory? Or is there some other layer at which this can be accomplished? In active-passive DRBD this wouldn't be a danger, since only the active system can see the storage to start the VM. But that wouldn't support live migration. Part of the danger can be avoided by using stonith in failover, but it would be nice to find a more fine-grained, subtle approach. Best, Whit

1 0

[libvirt-users] PCI Pass through of atheros card
by Varun Ganesh 23 Apr '11

23 Apr '11

Hi, I am trying to make a PCI passthrough of the atheros card to a VM running over KVM. am following the steps given at http://docs.fedoraproject.org/en-US/Fedora/13/html/Virtualization_Guide/sec…. but am getting the following errors, 1. Error starting domain: operation failed: failed to retrieve chardev info in qemu with 'info chaqdev' 2. Details: 3. Error starting domain: operation failed: failed to retrieve chardev info in qemu with 'info chaqdev' 4. 5. Traceback (most recent call last): 6. File "/usr/share/virt-manager/virtManager/asyncjob.py", line 45, in cb_wrapper 7. callback(asyncjob, *args, **kwargs) 8. File "/usr/share/virt-manager/virtManager/engine.py", line 959, in asyncfunc 9. vm.startup() 10. File "/usr/share/virt-manager/virtManager/domain.py", line 1114, in startup 11. self._backend.create() 12. File "/usr/lib/python2.7/site-packages/libvirt.py", line 333, in create 13. if ret == -1: raise libvirtError ('virDomainCreate() failed', dom=self) 14. libvirtError: operation failed: failed to retrieve chardev info in qemu with 'info chaqdev' I am using a Dell Optiplex XE system which has VT-d enabled and Atheros AR5001X+ Wireless Network Adapter. How do i rectify this. Thanks, sincerely, Varun Ganesh

1 0

[libvirt-users] domain not present after shutdown
by David White 22 Apr '11

22 Apr '11

I hope this is simple I have one kvm host that lists all my domains even after shut down. 'virsh list --all' shows machines in 'running' and 'shut off' states. If I copy those over to another machine and it only shows the virt machines in running state. vmhost2:~$ virsh list --all Id Name State ---------------------------------- 4 mybox4 running vmhost2:~$ virsh version Compiled against library: libvir 0.8.3 Using library: libvir 0.8.3 Using API: QEMU 0.8.3 Running hypervisor: QEMU 0.12.5 vmhost1:~$ virsh list --all Id Name State ---------------------------------- - mybox3 shut off 2 mybox4 running vmhost1:~$ virsh 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 question -- why doesn't mybox3 show up on vmhost2? where should I start looking? hope this isn't too open ended...

3 6