- Users - libvirt List Archives

[libvirt-users] API to set cpuset.cpu_exclusive flag
by Peeyush Gupta 24 Jul '13

24 Jul '13

Hi all, I have been trying to set cpu_exclusive flag. Now I can do it using "echo". I want to know is there any other way (an API) to set this flag? Is it possible to set this flag using API? Thanks. ~Peeyush Gupta

2 2

[libvirt-users] How to monitor a lxc container started by libvirt_lxc from inside ?
by hzguanqiang 24 Jul '13

24 Jul '13

Hi Guys, When I created a lxc container by libvirt, I logged into the lxc container and noticed that info under /proc/ dir did not match the lxc container resource. Is the /proc dir in lxc container just showing the same thing as the lxc host? If I want to monitor the realtime resource usage inside the lxc container, What should I do? -------------- Best regards! GuanQiang 2013-07-23

5 9

[libvirt-users] Is there any document about the feauture and usage of lxc supported by libvirt?
by hzguanqiang 23 Jul '13

23 Jul '13

Hi Guys, Recently, I'm doing some tests on openstack with lxc virtualization supported by libvirt. I need some recently new documents about libvirt lxc. Anything related is just ok. Thanks! -------------- Best regards! GuanQiang 2013-07-23

3 2

[libvirt-users] Resize errors with virt-resize/vgchange
by Alex 22 Jul '13

22 Jul '13

Hi, I have an fc18 system and trying to resize an LVM partition with an ext4 filesystem and receiving the following message from virt-resize: # virt-resize -d --expand /dev/sda1 --LV-expand /dev/mapper/prop-home prop-1.img prop-expand.img command line: virt-resize -d --expand /dev/sda1 --LV-expand /dev/mapper/prop-home prop-1.img prop-expand.img Examining prop-1.img ... libguestfs: trace: add_drive "prop-1.img" "readonly:true" libguestfs: trace: add_drive = 0 libguestfs: trace: add_drive "prop-expand.img" "readonly:false" libguestfs: trace: add_drive = 0 libguestfs: trace: launch libguestfs: trace: get_tmpdir libguestfs: trace: get_tmpdir = "/tmp" libguestfs: trace: get_cachedir libguestfs: trace: get_cachedir = "/var/tmp" libguestfs: trace: disk_format "/var/lib/libvirt/images/prop-expand.img" libguestfs: trace: disk_format = "raw" libguestfs: trace: get_cachedir libguestfs: trace: get_cachedir = "/var/tmp" [...] 100% --:-- libguestfs: trace: launch = 0 libguestfs: trace: lvm_set_filter "/dev/sda" libguestfs: trace: lvm_set_filter = -1 (error) Fatal error: exception Guestfs.Error("lvm_set_filter: vgchange: Couldn't find device with uuid zouQ8X-qxqJ-mp6p-pzg3-mi2i-K9YM-A763Kc. Refusing activation of partial LV home. Use --partial to override. Refusing activation of partial LV swap. Use --partial to override. R libguestfs: trace: close libguestfs: trace: internal_autosync libguestfs: trace: internal_autosync = 0 I don't understand this error message. I also see that vgchange doesn't even have a 'partial' option, so I'm not sure how to troubleshoot it. Here is the filesystem layout for this system: # df -h Filesystem Size Used Avail Use% Mounted on devtmpfs 7.9G 0 7.9G 0% /dev tmpfs 7.9G 0 7.9G 0% /dev/shm tmpfs 7.9G 3.9M 7.9G 1% /run tmpfs 7.9G 0 7.9G 0% /sys/fs/cgroup /dev/mapper/prop-root 15G 7.4G 6.3G 55% / tmpfs 7.9G 0 7.9G 0% /tmp /dev/mapper/prop-boot 477M 95M 358M 21% /boot /dev/mapper/prop-home 222G 212G 9.9G 96% /home The image is the 222G /home partition only. The other partitions are on another image. Any ideas how to troubleshoot this would be greatly appreciated. Thanks, Alex

2 1

[libvirt-users] Libvirt-lxc and systemd question
by Matt Hicks 22 Jul '13

22 Jul '13

Warning - I'm fairly new to libvirt, lxc and systemd so there is a good chance I'm doing something terribly wrong here. However, instead of continuing to struggle, I figured I would mail the list for some advice. What I'm trying to accomplish is a libvirt-lxc, systemd-based container running on my system (Fedora 19). I've read that sharing the underlying OS filesystem with the containers doesn't work, so I've installed a minimal Fedora 19 install in /srv/mycontainer. Everything seems to work okay but what I'm struggling with is how to setup the initial accounts. I've tried to attach to the container using 'nsenter' (entering all the namespaces) but it doesn't appear that the bind mounts are in place. For example, I see the /etc/passwd for my host OS, not the container. Is there a better way to setup the initial accounts on the container? Here is what I have installed: $ rpm -qa | grep lxc libvirt-daemon-driver-lxc-1.0.5.2-1.fc19.x86_64 libvirt-daemon-lxc-1.0.5.2-1.fc19.x86_64 $ rpm -qa | grep systemd systemd-libs-204-9.fc19.x86_64 systemd-python-204-9.fc19.x86_64 systemd-sysv-204-9.fc19.x86_64 systemd-libs-204-9.fc19.i686 systemd-204-9.fc19.x86_64 Here is the scenario I'm trying to go through: $ export LIBVIRT_DEFAULT_URI=lxc:/// $ getenforce Enforcing $ sudo yum -y --releasever=19 --nogpg --installroot=/srv/mycontainer --disablerepo='*' --enablerepo=fedora install systemd passwd yum fedora-release vim-minimal ... lots of output $ ls /srv/mycontainer/ bin boot dev etc home lib lib64 media mnt opt proc root run sbin srv sys tmp usr var $ cat test2.xml <domain type='lxc'> <name>test2</name> <memory>102400</memory> <os> <type arch='x86_64'>exe</type> <init>/bin/systemd</init> </os> <devices> <console type='pty'/> <filesystem type='mount'> <source dir='/srv/mycontainer'/> <target dir='/'/> </filesystem> </devices> </domain> $ virsh define test2.xml Domain test2 defined from test2.xml $ virsh start test2 Domain test2 started # Attach to container to set account passwords $ sudo nsenter -m -u -i -n -p -t `pgrep -f test2` [sudo] password for mhicks: [root@localhost /]# diff -q /srv/mycontainer/etc/passwd /etc/passwd Files /srv/mycontainer/etc/passwd and /etc/passwd differ Any ideas? -Matt

2 3

[libvirt-users] Clipboard
by Dean Hunter 22 Jul '13

22 Jul '13

I am unable to cut and paste between a Fedora 19 host and a Fedora 19 guest. Am I doing something wrong or is this a known problem that I have not been able to find?

2 4

[libvirt-users] Any API in Java Binding for retrieving the actual Memory and CPU usage of a domain?
by V Plus 22 Jul '13

22 Jul '13

I tried some like 'domain.getInfo().memory;', but it is not the actual usage..

1 0

[libvirt-users] CfP 8th Workshop on Virtualization in High-Performance Cloud Computing (VHPC '13)
by VHPC 12 20 Jul '13

20 Jul '13

we apologize if you receive multiple copies of this CfP ================================================================= CALL FOR PAPERS 8th Workshop on Virtualization in High-Performance Cloud Computing (VHPC '13) as part of SC 13, Denver, Colorado | sponsored by ACM sighpc ================================================================= Date: November 22, 2013 Workshop URL: http://vhpc.org Paper Submission Deadline: September 23, 2013 CALL FOR PAPERS Virtualization has become a common abstraction layer in modern data centers, enabling resource owners to manage complex infrastructures independently of their applications. Conjointly, virtualization is becoming a driving technology for a manifold of industry grade IT services. The cloud concept includes the notion of a separation between resource owners and users, adding services such as hosted application frameworks and queueing. Utilizing the same infrastructure, clouds carry significant potential for use in cpu-intensive or data-intensive computing. The ability of clouds to provide for requests and releases of vast computing resources dynamically and close to the marginal cost of providing the services is unprecedented in the history of scientific and commercial computing. This workshop aims to bring together industrial providers with the application community in order to foster discussion, collaboration and mutual exchange of knowledge and experience. The workshop will be one day in length, composed of 20 min paper presentations, each followed by 10 min discussion sections. Lightning talks are limited to 5 minutes. Presentations may be accompanied by interactive demonstrations. TOPICS Topics of interest include, but are not limited to: - Management, deployment and monitoring of VM-based environments - VM-cloud performance monitoring - VM cloud topology management and optimization - Operating systems virtualization supportpptimization - VM-based cloud performance modelling - Network virtualization for VM-environments - Data virtualization - Cloudbursting - Evolved grid architectures including such based on network virtualization - Workload characterization for VM-based environments - Optimized communication libraries/protocols in the cloud - System and process/bytecode VM convergence - Cloud frameworks and APIs - GPU Virtualization architectures and APIs - Checkpointing/migration of large compute jobs - Instrumentation interfaces and languages - VMM performance (auto-)tuning on various load types - Cloud reliability, fault-tolerance, and security - Heterogeneous virtualized environments - Paravirtualized I/O - Services in cloud HPC - Research and education use cases - Virtualization in cloud, cluster and grid environments - Cross-layer VM optimizations - Cloud HPC use cases including optimizations - Energy-aware virtualization - Performance and cost modelling - QoS and and service levels - Languages for describing highly-distributed compute jobs - VM cloud and cluster distribution algorithms, load balancing - Instrumentation interfaces and languages - Hypervisor extensions and tools for cluster and grid computing - Virtual machine monitor platforms - Cluster provisioning in the cloud Important Dates: Rolling Paper registration September 23, 2013 - Full paper submission October 21, 2013 - Acceptance notification November 8, 2013 - Camera-ready version due Lightning Talks: August 9, 2013 - Deadline for lightning talk abstracts September 2, 2013 - Lightning talk notification November 22, 2013 - Workshop Date TPC CHAIR Michael Alexander (chair), TU Wien, Austria Gianluigi Zanetti (co-chair), CRS4, Italy Anastassios Nanos (co-chair), NTUA, Greece PROGRAM COMMITTEE Costas Bekas, IBM, Switzerland Jakob Blomer, CERN Giovanni Busonera, CRS4, Italy Roberto Canonico, University of Napoli Federico II, Italy Simon Crosby, Bromium, USA Tommaso Cucinotta, Alcatel-Lucent Bell Labs, Ireland Casimer DeCusatis, IBM, USA William Gardner, University of Guelph, USA Marcus Hardt, Forschungszentrum Karlsruhe, Germany Sverre Jarp, CERN, Switzerland Xuxian Jiang, NC State, USA Krishna Kant, George Mason University, USA Romeo Kinzler, IBM, Switzerland Nectarios Koziris, National Technical University of Athens, Greece Simone Leo, CRS4, Italy Jean-Marc Menaud, Ecole des Mines de Nantes, France Dimitrios Nikolopoulos, Queen's University of Belfast, UK Josh Simons, VMWare, USA Borja Sotomayor, University of Chicago, USA Yoshio Turner, HP Labs, USA Kurt Tutschku, Blekinge Institute of Technology, Sweden Chao-Tung Yang, Tunghai University, Taiwan PAPER SUBMISSION-PUBLICATION Papers submitted to the workshop will be reviewed by at least two members of the program committee and external reviewers. Submissions should include abstract, key words, the e-mail address of the corresponding author, and must not exceed 8 pages, including tables and figures at a main font size no smaller than 11 point. Submission of a paper should be regarded as a commitment that, should the paper be accepted, at least one of the authors will register and attend the conference to present the work. Accepted papers will be published in the ACM International Conference Proceedings Series. The format must be according to the ACM SIG style. Initial submissions are in PDF; authors of accepted papers will be requested to provide source files. Format Guidelines: http://www.acm.org/sigs/publications/sigguide-v2.2sp Style template: http://www.acm.org/sigs/publications/proceedings-templates Abstract Submission Link: http://edas.info/newPaper.php?c=14791 LIGHTNING TALKS Lightning Talks are non-paper track synoptical in nature that are strictly limited to 5 minutes. They can be used to gain early feedback on ongoing research, for demonstrations, to present research results, early research ideas, perspectives and positions of interest to the community. GENERAL INFORMATION The workshop will be held as part of SC’13, Denver, Colorado. SC 2013: http://sc13.supercomputing.org/

1 0

[libvirt-users] Changing CPU cache size of guest
by Peter Matulis 19 Jul '13

19 Jul '13

Hi everybody, Using libvirt 1.0.0 how does one change the L2 cache size of the virtual CPU? I can change the CPU model but the cache size never changes: <cpu mode='host-passthrough'/> <model>Nehalem</model> OR <cpu match='exact' mode='custom'/> <model>Nehalem</model> Is there a qemu requirement? -- peter matulis

2 1

[libvirt-users] How to handle IP-based Networkfilters
by Matthias Babisch 19 Jul '13

19 Jul '13

Hello People. We are currently exploring the possibility to use libvirt and kvm/quemu for production purposes. The general stability seems good enough and the performance is great. There are some issues we do not understand here yet. For security reasons we are considering the extensive use of Networkfilters for virtual machines. But we found some simple scheme for a test-server not to be working as we expected. It might well be that we misunderstand something here, so I am hoping someone could point out to us, where either we or perhaps libvirt failed in this example. We are using an ubuntu 13.04 Server running the provided "1.0.2-0ubuntu11.13.04.2" libvirt-bin using amd64-architecture. The type of VM should not be relevant for this problem. Its a linux-based xmpp-Server which uses ucarp. I reduced the used filter-file just so i could prove my point. It contains: <filter name='linux-based-xmpp-server' chain='root'> <uuid>fb539996-eed5-11e2-8bd3-00e081e0f040</uuid> <rule action='accept' direction='in' priority='999'> <tcp state='NEW' dstportstart='5222'/> </rule> <rule action='accept' direction='in' priority='999'> <tcp state='NEW' dstportstart='5269'/> </rule> <rule action='accept' direction='inout' priority='999'> <ip dstipaddr='224.0.0.18' proto='112'/> </rule> <rule action='reject' direction='inout' priority='999'> <all/> </rule> </filter> Practically it should allow TCP-traffic on Ports 5222,5269 incoming and incoming and outgoing traffic for ip protocol 112 to destination ip 224.0.0.18 (VRRP used by ucarp). All other traffic should be rejected. There is only one VM on the system and the VM has this ruleset attached. Note: It is clear to me that this example won't work as areal world example, because packets of the state ESTABLISHED,RELATED are not allowed through the firewall. I removed these rules because they where in a filter-file i referenced. After reloading the libvirt-bin i do get part of the rules in would expect in iptables: Chain FI-vnet0 (1 references) target prot opt source destination REJECT all -- 0.0.0.0/0 0.0.0.0/0 reject-with icmp-port-unreachable Chain FO-vnet0 (1 references) target prot opt source destination ACCEPT tcp -- 0.0.0.0/0 0.0.0.0/0 tcp dpt:5222 state NEW ACCEPT tcp -- 0.0.0.0/0 0.0.0.0/0 tcp dpt:5269 state NEW REJECT all -- 0.0.0.0/0 0.0.0.0/0 reject-with icmp-port-unreachable Chain HI-vnet0 (1 references) target prot opt source destination REJECT all -- 0.0.0.0/0 0.0.0.0/0 reject-with icmp-port-unreachable What is missing is any reference to the rule for ucarp (protocol 112). Please note though that removing the protocol and just allowing any ip traffic to 224.0.0.18 as a rule, does not appear in the iptables either. Am i misunderstanding anything here? Is there a bug in libvirt? How do you interpret this? Do you know of any other way to achieve the simple ruleset intended? I am hoping to get more information from this list. If you are replying, please cc me (matthias.babisch(a)bmiag.de) because i receive this list as a digest. Sincerely Matthias Babisch IT/Organisation *b+m Informatik AG* Rotenhofer Weg 20 24109 Melsdorf T +49 4340/404-1444 F +49 4340/404-111 M +49 160/8866426 matthias.babisch(a)bmiag.de Aktuelle Informationen unter www.bmiag.de <%5C%22http://www.bmiag.de%5C%22> Die b+m Informatik AG ist ein Unternehmen der Allgeier Gruppe <%5C%22http://www.allgeier-holding.de%5C%22> Vorsitzender des Aufsichtsrates: Dr. Marcus Goedsche Vorstand: Dipl-Ing. Frank Mielke Amtsgericht Kiel, HRB 5526

2 1