February 2012 - Users - libvirt List Archives

[libvirt-users] libvirtd restore error?
by lee 13 Feb '12

13 Feb '12

Hi qemu-kvm-0.12.1.2-2.160.el6_1.2.x86_64 qemu-kvm-tools-0.12.1.2-2.160.el6_1.2.x86_64 libvirt-java-javadoc-0.4.7-1.el6.noarch libvirt-client-0.8.7-18.el6.x86_64 fence-virtd-libvirt-qpid-0.2.1-8.el6.x86_64 ocaml-libvirt-0.6.1.0-6.2.el6.x86_64 libvirt-devel-0.8.7-18.el6.x86_64 libvirt-python-0.8.7-18.el6.x86_64 libvirt-java-0.4.7-1.el6.noarch libvirt-cim-0.5.11-3.el6.x86_64 libvirt-java-devel-0.4.7-1.el6.noarch libvirt-client-0.8.7-18.el6.i686 libvirt-qpid-0.2.22-6.el6.x86_64 fence-virtd-libvirt-0.2.1-8.el6.x86_64 libvirt-0.8.7-18.el6.x86_64 ocaml-libvirt-devel-0.6.1.0-6.2.el6.x86_64 get this error when I try to restore a vm that was 'save'. Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/engine.py", line 909, in run_domain vm.startup() File "/usr/share/virt-manager/virtManager/domain.py", line 1321, in startup self._backend.create() File "/usr/lib64/python2.7/site-packages/libvirt.py", line 330, in create if ret == -1: raise libvirtError ('virDomainCreate() failed', dom=self) libvirtError: cannot send monitor command '{"execute":"qmp_capabilities"}': Connection reset by peer the solution was to delete the /var/lib/libvirt/qemu/save directory. Anyone any idea of why? Lee

2 1

[libvirt-users] virt-manager thru window's remote desktop
by Hisao Taguchi 13 Feb '12

13 Feb '12

Hi, I am trying to operate the remote KVM host's virt-manager thru Terminal Server Client. This KVM hosts is CentOS6.0 and running KVM guest and xrdp is installed from "epel" repo. And I'm using fedora 14's Terminal Server Client. After I started Virtual Machine Manager and try to connect to the hypervisor, following error happened. Is there any way to resolve this error? I know that when I use VNC server , this error does not happened, but I want to use Terminal Server rather than VNC, because this doesn't need extra software for windows client PCs. -- Error snip -- Virtual Machine Manager Connection Failure Unable to open a connection to the libvirt management daemon. Libvirt URI is: qemu:///system Verify that: - The 'libvirtd' daemon has been started Details Unable to open connection to hypervisor URI 'qemu:///system': authentication failed Traceback (most recent call last): File "/usr/share/virt-manager/virtManager/connection.py", line 992, in_try_open None],flags) File "/usr/lib64/python2.6/site-packages/libvirt.py",line 111,in openAuth if ret is None:raise libvirtError('virConnectOpenAuth() failed') libvirtError: authentication failed -- end -- Hisao

2 3

[libvirt-users] java binding
by Maurizio Paone 12 Feb '12

12 Feb '12

Hi all, I'm trying to use the libvirt event system in order to notify the VM status (VM boot, reboot, crash, etc.). I must use the java binding, so I exposed some calls not present in "Libvirt-java" jna interface: virEventRunDefaultImpl() virEventRegisterDefaultImpl() (and several costant). In my java source I tried to intercept the event, using the "Connection.domainEventRegisterAny( ... )" call , but the "answer" was: "this function is not supported by the connection driver: no event support". No errors issued on the virEventRunDefaultImpl and virEventRegisterDefaultImpl (moreover, both the calls return "0" ). I do not know the libvirt internals, so my question is : it is possible to call the libvirt events calls using java binding? Can you tell me the issues involved in this task? I'm thinking to build a native library and use jni or use jython as alternatives. Thanks, Maurizio

1 0

[libvirt-users] Unable to collect VM metrics for ESX using libvirt
by sethu subbiah 11 Feb '12

11 Feb '12

Hi, I am trying to connect to a server running esx and collect system metrics like cpu usage, network and disk iops. I am able to establish the connection successfully and obtain VM information such as ids, domain names, config xml etc. But I am unable to collect statistics. I am using the python interface and if I do , conn=libvirt.openAuth(uri, auth, 0) dom-conn.lookupByID(id) // Where id is the domain id of the vms netstats=dom.interfaceStats(interface) gives an error libvirt.libvirtError: this function is not supported by the connection driver: virDomainInterfaceStats The same goes for block stats. For the cpu, I used, cpu=dom.info()[4] but, that always returns 0 (I thought it was cumulative and returns a huge value) I have used similar scripts with xen and kvm and it worked well. Can someone tell me what mistake I am doing or if the driver doesnt support these functions, is there any other way of getting the stats ?? Thank you. - Regards, Sethuraman Subbiah Graduate Student - NC state University M.S in Computer Science

2 1

[libvirt-users] virsh migrate results in error: Timed out during operation: cannot acquire state change lock
by Nuno Sousa 11 Feb '12

11 Feb '12

Hi all, Hi have one computer hosting several VM, this particular server is in need of maintenance and has to be shutdown. In order to avoid disrupting the services provided by the VM I intended to migrate the VM to another computer. Computer 1 and 2 do not share any storage. So it is necessary to migrate not only the memory but also any data on the hard-drive. Both computers are running debian squeeze. Virsh is version 0.8.3. I issue this command on server one: sudo LIBVIRT_DEBUG=1 virsh migrate --live --copy-storage-all i00cn1011 qemu+ssh://user@computer2/session > debug.log 2>&1 On virt-manager on computer 1, the VM appears for a few moments as paused, and then disappears. On computer 1 I receive the error: Timed out during operation: cannot acquire state change lock Does anyone know what the problem may be and how to solve it? Next I send all the data returned by virsh. Thanks everyone. user@computer1:~$ cat debug.log 13:52:07.911: debug : virInitialize:339 : register drivers 13:52:07.911: debug : virRegisterDriver:927 : registering Test as driver 0 13:52:07.911: debug : virRegisterNetworkDriver:733 : registering Test as network driver 0 13:52:07.911: debug : virRegisterInterfaceDriver:764 : registering Test as interface driver 0 13:52:07.911: debug : virRegisterStorageDriver:795 : registering Test as storage driver 0 13:52:07.911: debug : virRegisterDeviceMonitor:826 : registering Test as device driver 0 13:52:07.911: debug : virRegisterSecretDriver:857 : registering Test as secret driver 0 13:52:07.911: debug : virRegisterNWFilterDriver:888 : registering Test as network filter driver 0 13:52:07.911: debug : virRegisterDriver:927 : registering Xen as driver 1 13:52:07.911: debug : virRegisterDriver:927 : registering OPENVZ as driver 2 13:52:07.911: debug : vboxRegister:122 : VBoxCGlueInit failed, using dummy driver 13:52:07.911: debug : virRegisterDriver:927 : registering VBOX as driver 3 13:52:07.911: debug : virRegisterNetworkDriver:733 : registering VBOX as network driver 1 13:52:07.911: debug : virRegisterStorageDriver:795 : registering VBOX as storage driver 1 13:52:07.911: debug : virRegisterDriver:927 : registering remote as driver 4 13:52:07.911: debug : virRegisterNetworkDriver:733 : registering remote as network driver 2 13:52:07.911: debug : virRegisterInterfaceDriver:764 : registering remote as interface driver 1 13:52:07.911: debug : virRegisterStorageDriver:795 : registering remote as storage driver 2 13:52:07.911: debug : virRegisterDeviceMonitor:826 : registering remote as device driver 1 13:52:07.911: debug : virRegisterSecretDriver:857 : registering remote as secret driver 1 13:52:07.911: debug : virRegisterNWFilterDriver:888 : registering remote as network filter driver 1 13:52:07.911: debug : virConnectOpenAuth:1498 : name=(null), auth=0x7f3c96c6c7a0, flags=0 13:52:07.911: debug : do_open:1206 : no name, allowing driver auto-select 13:52:07.911: debug : do_open:1243 : trying driver 0 (Test) ... 13:52:07.911: debug : do_open:1249 : driver 0 Test returned DECLINED 13:52:07.911: debug : do_open:1243 : trying driver 1 (Xen) ... 13:52:07.911: debug : do_open:1249 : driver 1 Xen returned DECLINED 13:52:07.911: debug : do_open:1243 : trying driver 2 (OPENVZ) ... 13:52:07.911: debug : do_open:1249 : driver 2 OPENVZ returned DECLINED 13:52:07.911: debug : do_open:1243 : trying driver 3 (VBOX) ... 13:52:07.911: debug : do_open:1249 : driver 3 VBOX returned DECLINED 13:52:07.911: debug : do_open:1243 : trying driver 4 (remote) ... 13:52:07.911: debug : remoteOpen:1120 : Auto-probe remote URI 13:52:07.911: debug : doRemoteOpen:564 : proceeding with name = 13:52:07.911: debug : remoteIO:9886 : Do proc=66 serial=0 length=28 wait=(nil) 13:52:07.911: debug : remoteIO:9961 : We have the buck 66 0x7f3c96d01010 0x7f3c96d01010 13:52:07.912: debug : remoteIODecodeMessageLength:9314 : Got length, now need 64 total (60 more) 13:52:07.912: debug : remoteIOEventLoop:9812 : Giving up the buck 66 0x7f3c96d01010 (nil) 13:52:07.912: debug : remoteIO:9990 : All done with our call 66 (nil) 0x7f3c96d01010 13:52:07.912: debug : remoteIO:9886 : Do proc=1 serial=1 length=40 wait=(nil) 13:52:07.912: debug : remoteIO:9961 : We have the buck 1 0xe671b0 0xe671b0 13:52:07.912: debug : remoteIODecodeMessageLength:9314 : Got length, now need 56 total (52 more) 13:52:07.912: debug : remoteIOEventLoop:9812 : Giving up the buck 1 0xe671b0 (nil) 13:52:07.912: debug : remoteIO:9990 : All done with our call 1 (nil) 0xe671b0 13:52:07.912: debug : remoteIO:9886 : Do proc=110 serial=2 length=28 wait=(nil) 13:52:07.912: debug : remoteIO:9961 : We have the buck 110 0xe671b0 0xe671b0 13:52:07.913: debug : remoteIODecodeMessageLength:9314 : Got length, now need 76 total (72 more) 13:52:07.913: debug : remoteIOEventLoop:9812 : Giving up the buck 110 0xe671b0 (nil) 13:52:07.913: debug : remoteIO:9990 : All done with our call 110 (nil) 0xe671b0 13:52:07.913: debug : doRemoteOpen:922 : Auto-probed URI is qemu:///system 13:52:07.913: debug : doRemoteOpen:941 : Adding Handler for remote events 13:52:07.913: debug : doRemoteOpen:948 : virEventAddHandle failed: No addHandleImpl defined. continuing without events. 13:52:07.913: debug : do_open:1249 : driver 4 remote returned SUCCESS 13:52:07.913: debug : do_open:1269 : network driver 0 Test returned DECLINED 13:52:07.913: debug : do_open:1269 : network driver 1 VBOX returned DECLINED 13:52:07.913: debug : do_open:1269 : network driver 2 remote returned SUCCESS 13:52:07.913: debug : do_open:1288 : interface driver 0 Test returned DECLINED 13:52:07.913: debug : do_open:1288 : interface driver 1 remote returned SUCCESS 13:52:07.913: debug : do_open:1308 : storage driver 0 Test returned DECLINED 13:52:07.913: debug : do_open:1308 : storage driver 1 VBOX returned DECLINED 13:52:07.913: debug : do_open:1308 : storage driver 2 remote returned SUCCESS 13:52:07.913: debug : do_open:1328 : node driver 0 Test returned DECLINED 13:52:07.913: debug : do_open:1328 : node driver 1 remote returned SUCCESS 13:52:07.913: debug : do_open:1355 : secret driver 0 Test returned DECLINED 13:52:07.913: debug : do_open:1355 : secret driver 1 remote returned SUCCESS 13:52:07.913: debug : do_open:1375 : nwfilter driver 0 Test returned DECLINED 13:52:07.913: debug : do_open:1375 : nwfilter driver 1 remote returned SUCCESS 13:52:07.913: debug : virDomainLookupByName:2154 : conn=0xe61390, name=i00cn1011 13:52:07.913: debug : remoteIO:9886 : Do proc=23 serial=3 length=44 wait=(nil) 13:52:07.913: debug : remoteIO:9961 : We have the buck 23 0xe671b0 0xe671b0 13:52:07.913: debug : remoteIODecodeMessageLength:9314 : Got length, now need 92 total (88 more) 13:52:07.913: debug : remoteIOEventLoop:9812 : Giving up the buck 23 0xe671b0 (nil) 13:52:07.913: debug : remoteIO:9990 : All done with our call 23 (nil) 0xe671b0 13:52:07.913: debug : virGetDomain:382 : New hash entry 0xe5da00 13:52:07.913: debug : virConnectOpenAuth:1498 : name=qemu+ssh://user@computer2/session, auth=0x7f3c96c6c7a0, flags=0 13:52:07.913: debug : do_open:1204 : name "qemu+ssh://user@computer2/session" to URI components: scheme qemu+ssh opaque (null) authority (null) server computer2 user user port 0 path /session 13:52:07.913: debug : do_open:1243 : trying driver 0 (Test) ... 13:52:07.913: debug : do_open:1249 : driver 0 Test returned DECLINED 13:52:07.913: debug : do_open:1243 : trying driver 1 (Xen) ... 13:52:07.913: debug : do_open:1249 : driver 1 Xen returned DECLINED 13:52:07.913: debug : do_open:1243 : trying driver 2 (OPENVZ) ... 13:52:07.913: debug : do_open:1249 : driver 2 OPENVZ returned DECLINED 13:52:07.913: debug : do_open:1243 : trying driver 3 (VBOX) ... 13:52:07.913: debug : do_open:1249 : driver 3 VBOX returned DECLINED 13:52:07.913: debug : do_open:1243 : trying driver 4 (remote) ... 13:52:07.913: debug : doRemoteOpen:564 : proceeding with name = qemu:///session 13:52:07.914: debug : virExecWithHook:712 : ssh -l user computer2 sh -c 'nc -q 2>&1 | grep -q "requires an argument";if [ $? -eq 0 ] ; then CMD="nc -q 0 -U /var/run/libvirt/libvirt-sock";else CMD="nc -U /var/run/libvirt/libvirt-sock";fi;eval "$CMD";' 13:52:07.914: debug : remoteIO:9886 : Do proc=66 serial=0 length=28 wait=(nil) 13:52:07.914: debug : remoteIO:9961 : We have the buck 66 0xeaabd0 0xeaabd0 13:52:08.006: debug : remoteIODecodeMessageLength:9314 : Got length, now need 36 total (32 more) 13:52:08.006: debug : remoteIOEventLoop:9812 : Giving up the buck 66 0xeaabd0 (nil) 13:52:08.006: debug : remoteIO:9990 : All done with our call 66 (nil) 0xeaabd0 13:52:08.006: debug : remoteIO:9886 : Do proc=1 serial=1 length=56 wait=(nil) 13:52:08.006: debug : remoteIO:9961 : We have the buck 1 0xeaabd0 0xeaabd0 13:52:08.007: debug : remoteIODecodeMessageLength:9314 : Got length, now need 28 total (24 more) 13:52:08.007: debug : remoteIOEventLoop:9812 : Giving up the buck 1 0xeaabd0 (nil) 13:52:08.007: debug : remoteIO:9990 : All done with our call 1 (nil) 0xeaabd0 13:52:08.007: debug : doRemoteOpen:941 : Adding Handler for remote events 13:52:08.007: debug : doRemoteOpen:948 : virEventAddHandle failed: No addHandleImpl defined. continuing without events. 13:52:08.007: debug : do_open:1249 : driver 4 remote returned SUCCESS 13:52:08.007: debug : do_open:1269 : network driver 0 Test returned DECLINED 13:52:08.007: debug : do_open:1269 : network driver 1 VBOX returned DECLINED 13:52:08.007: debug : do_open:1269 : network driver 2 remote returned SUCCESS 13:52:08.007: debug : do_open:1288 : interface driver 0 Test returned DECLINED 13:52:08.007: debug : do_open:1288 : interface driver 1 remote returned SUCCESS 13:52:08.007: debug : do_open:1308 : storage driver 0 Test returned DECLINED 13:52:08.007: debug : do_open:1308 : storage driver 1 VBOX returned DECLINED 13:52:08.007: debug : do_open:1308 : storage driver 2 remote returned SUCCESS 13:52:08.007: debug : do_open:1328 : node driver 0 Test returned DECLINED 13:52:08.007: debug : do_open:1328 : node driver 1 remote returned SUCCESS 13:52:08.007: debug : do_open:1355 : secret driver 0 Test returned DECLINED 13:52:08.007: debug : do_open:1355 : secret driver 1 remote returned SUCCESS 13:52:08.007: debug : do_open:1375 : nwfilter driver 0 Test returned DECLINED 13:52:08.007: debug : do_open:1375 : nwfilter driver 1 remote returned SUCCESS 13:52:08.007: debug : virDomainMigrate:3517 : domain=0xe5da00, dconn=0xe62b00, flags=65, dname=(null), uri=(null), bandwidth=0 13:52:08.008: debug : remoteIO:9886 : Do proc=60 serial=4 length=32 wait=(nil) 13:52:08.008: debug : remoteIO:9961 : We have the buck 60 0xeaac10 0xeaac10 13:52:08.008: debug : remoteIODecodeMessageLength:9314 : Got length, now need 60 total (56 more) 13:52:08.008: debug : remoteIOEventLoop:9812 : Giving up the buck 60 0xeaac10 (nil) 13:52:08.008: debug : remoteIO:9990 : All done with our call 60 (nil) 0xeaac10 13:52:08.008: debug : remoteIO:9886 : Do proc=60 serial=5 length=32 wait=(nil) 13:52:08.008: debug : remoteIO:9961 : We have the buck 60 0xeaac10 0xeaac10 13:52:08.008: debug : remoteIODecodeMessageLength:9314 : Got length, now need 60 total (56 more) 13:52:08.008: debug : remoteIOEventLoop:9812 : Giving up the buck 60 0xeaac10 (nil) 13:52:08.008: debug : remoteIO:9990 : All done with our call 60 (nil) 0xeaac10 13:52:08.008: debug : remoteIO:9886 : Do proc=60 serial=2 length=32 wait=(nil) 13:52:08.008: debug : remoteIO:9961 : We have the buck 60 0xeaac10 0xeaac10 13:52:08.009: debug : remoteIODecodeMessageLength:9314 : Got length, now need 32 total (28 more) 13:52:08.009: debug : remoteIOEventLoop:9812 : Giving up the buck 60 0xeaac10 (nil) 13:52:08.009: debug : remoteIO:9990 : All done with our call 60 (nil) 0xeaac10 13:52:08.009: debug : remoteIO:9886 : Do proc=14 serial=6 length=68 wait=(nil) 13:52:08.009: debug : remoteIO:9961 : We have the buck 14 0xeaac10 0xeaac10 13:52:08.009: debug : remoteIODecodeMessageLength:9314 : Got length, now need 2540 total (2536 more) 13:52:08.009: debug : remoteIOEventLoop:9812 : Giving up the buck 14 0xeaac10 (nil) 13:52:08.010: debug : remoteIO:9990 : All done with our call 14 (nil) 0xeaac10 13:52:08.010: debug : virDomainGetInfo:3017 : domain=0xe5da00, info=0x7fff102893e0 13:52:08.010: debug : remoteIO:9886 : Do proc=16 serial=7 length=64 wait=(nil) 13:52:08.010: debug : remoteIO:9961 : We have the buck 16 0xeaac10 0xeaac10 13:52:08.010: debug : remoteIODecodeMessageLength:9314 : Got length, now need 88 total (84 more) 13:52:08.010: debug : remoteIOEventLoop:9812 : Giving up the buck 16 0xeaac10 (nil) 13:52:08.010: debug : remoteIO:9990 : All done with our call 16 (nil) 0xeaac10 13:52:08.010: debug : remoteIO:9886 : Do proc=108 serial=3 length=2536 wait=(nil) 13:52:08.010: debug : remoteIO:9961 : We have the buck 108 0xeaac10 0xeaac10 13:52:08.291: debug : remoteIODecodeMessageLength:9314 : Got length, now need 68 total (64 more) 13:52:08.291: debug : remoteIOEventLoop:9812 : Giving up the buck 108 0xeaac10 (nil) 13:52:08.291: debug : remoteIO:9990 : All done with our call 108 (nil) 0xeaac10 13:52:08.291: debug : remoteIO:9886 : Do proc=62 serial=8 length=120 wait=(nil) 13:52:08.291: debug : remoteIO:9961 : We have the buck 62 0xeaac10 0xeaac10 13:52:38.000: debug : remoteIODecodeMessageLength:9314 : Got length, now need 208 total (204 more) 13:52:38.000: debug : remoteIOEventLoop:9812 : Giving up the buck 62 0xeaac10 (nil) 13:52:38.000: debug : remoteIO:9990 : All done with our call 62 (nil) 0xeaac10 13:52:38.000: debug : remoteIO:9886 : Do proc=109 serial=4 length=92 wait=(nil) 13:52:38.000: debug : remoteIO:9961 : We have the buck 109 0xeaac10 0xeaac10 13:52:38.211: debug : remoteIODecodeMessageLength:9314 : Got length, now need 120 total (116 more) 13:52:38.211: debug : remoteIOEventLoop:9812 : Giving up the buck 109 0xeaac10 (nil) 13:52:38.211: debug : remoteIO:9990 : All done with our call 109 (nil) 0xeaac10 13:52:38.211: debug : virConnectClose:1524 : conn=0xe62b00 13:52:38.211: debug : virUnrefConnect:294 : unref connection 0xe62b00 1 13:52:38.211: debug : remoteIO:9886 : Do proc=2 serial=5 length=28 wait=(nil) 13:52:38.211: debug : remoteIO:9961 : We have the buck 2 0xeaac10 0xeaac10 13:52:38.212: debug : remoteIODecodeMessageLength:9314 : Got length, now need 28 total (24 more) 13:52:38.212: debug : remoteIOEventLoop:9812 : Giving up the buck 2 0xeaac10 (nil) 13:52:38.212: debug : remoteIO:9990 : All done with our call 2 (nil) 0xeaac10 13:52:38.212: debug : virReleaseConnect:249 : release connection 0xe62b00 13:52:38.212: debug : virDomainFree:2242 : domain=0xe5da00 13:52:38.212: debug : virUnrefDomain:463 : unref domain 0xe5da00 i00cn1011 1 13:52:38.212: debug : virReleaseDomain:416 : release domain 0xe5da00 i00cn1011 4dd247bc-533a-db0d-f1e5-f1310169e62f 13:52:38.213: debug : virReleaseDomain:433 : unref connection 0xe61390 2 error: Timed out during operation: cannot acquire state change lock 13:52:38.213: debug : virConnectClose:1524 : conn=0xe61390 13:52:38.213: debug : virUnrefConnect:294 : unref connection 0xe61390 1 13:52:38.213: debug : remoteIO:9886 : Do proc=2 serial=9 length=28 wait=(nil) 13:52:38.213: debug : remoteIO:9961 : We have the buck 2 0xe671b0 0xe671b0 13:52:38.213: debug : remoteIODecodeMessageLength:9314 : Got length, now need 56 total (52 more) 13:52:38.213: debug : remoteIOEventLoop:9812 : Giving up the buck 2 0xe671b0 (nil) 13:52:38.213: debug : remoteIO:9990 : All done with our call 2 (nil) 0xe671b0 13:52:38.213: debug : virReleaseConnect:249 : release connection 0xe61390

2 3

Re: [libvirt-users] want to access the page file of VM from host on Xen platform
by Eric Blake 10 Feb '12

10 Feb '12

[re-adding the list, so others may benefit from the answers] On 02/09/2012 02:00 PM, Irfan Ahmed wrote: > thanks Eric. Just for clarification, if the page is swaped out from RAM to HDD, will virDomainMemoryPeek functionalso retrive the block of size 'size' from that too or it is only restricted to the pages exist in RAM. virDomainMemoryPeek is limited to reading what is in the guest's RAM. As I said earlier, if you are interested in reading what is in the guest's disk, you might find libguestfs to be a more powerful tool. > > > Is there any example program that uses virDomainMemoryPeek function. Alas, no one has yet wired up virsh to call virDomainMemoryPeek (patches welcome), and I don't personally have any example programs that use it. Rich Jones might have some better ideas (as both author of libguestfs, and as the implementer of virDomainMemoryPeek). -- Eric Blake eblake(a)redhat.com +1-919-301-3266 Libvirt virtualization library http://libvirt.org

1 0

[libvirt-users] want to access the page file of VM from host on Xen platform
by Irfan Ahmed 10 Feb '12

10 Feb '12

Is there a way to access swap memory or page file of a virtual machine from host. I am using Xen and want to access the page file contents of my windows XP SP2 virtual machine from dom0. Please do tell me if there is already a tool that could provide this functionality. Any help is appreciated. Thanks and regards, Irfan

2 1

[libvirt-users] bad IO on VM
by Akendo 09 Feb '12

09 Feb '12

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 I have to create some VM. I using KVM for it, manage by libvirtd. The problem of my is that the IO is very bad. I'm nut sure, but I using currently vmdk. Could this be the problem? My settings for the VM is: vitrio driver vmdk no cache It's so bad that the Oracle DB can't work efficient. http://pastebin.com/raw.php?i=sDyvcJmB so far akendo -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.17 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iQEcBAEBAgAGBQJPM5NGAAoJEARqOhiQNM7ZQKMIAIVo9+SYmSrJAIooB0xObSEz ERmMANdkaA7tLGBWaSvKTh5Gg0d3NdP9P1aCIS38OXBBKxqy2NRcksgMikJmlaVp DitLHsSKKLNpBZpGe9xBvqrNCbJVdFLmWIjYk+WEBWXlxHFYob9FuSdejQ53JKKI mfLbcfhWb/6mEfcgkWVAWh9PumuRKZ8YLwWOfj/qJhkhnJOfsusai8Ns+TfMpO7s M72d2RYknpyXiNEJSwUIDlEl0OHqStG3T4+Tpx8Wmoqfs+cm4Hklca1GFGV/Wao2 aPYdkLYvMamAMyN8tB2I0c9MXXUQ8mxgruJReHM9NUYoX4iIwIazElPcZBp4FWI= =MDJL -----END PGP SIGNATURE-----

3 2

[libvirt-users] serial console with virt-install
by Felix Blanke 08 Feb '12

08 Feb '12

Hello, sorry for bother you again. I'm new to kvm and libvirt, that's why my next question is maybe some sort of noob question :) I have a headless server (no X at all, just ssh) and want to run some virtual machines on it. I have kvm etc. all set up. When doing something like virt-install --connect qemu:///system -n template -r 1024 --disk path=/data/vm/template.qcow2,format=qcow2 -c /data/vm/livedvd-amd64-multilib-2012.iso --os-type linux -v --network=bridge:br0 --virt-type kvm --arch x86_64 --graphics none the vm starts Starting install... Creating domain... Connected to domain template Escape character is ^] That's it, no more output. I see two threads running, each takes up 100% cpu. I tried to help myself via google and read a lot hints like "you need to setup your guest to get a serial console...", but that's not possible when running a fresh vm. All I want to do is getting some sort of console from the vm to start the installation with the given .iso file. Could someone lead me into the right direct? A link to an explanation would be more then enough :) Thanks for your help. Kind regards, Felix Blanke

3 3

[libvirt-users] virt-install: "ERROR Host does not support any virtualization options"
by Felix Blanke 08 Feb '12

08 Feb '12

Hello, I have a problem with virt-install. I want to use it to create my vm and manage them. But whenever I execute the virt-install cmd I get the error msg mentioned in the subject. Here are some more informations: hp ~ # virt-install --virt-type kvm --arch x86_64 --debug Mon, 06 Feb 2012 19:57:29 DEBUG Launched with command line: /usr/bin/virt-install --virt-type kvm --arch x86_64 --debug Mon, 06 Feb 2012 19:57:29 DEBUG Requesting libvirt URI default Mon, 06 Feb 2012 19:57:29 DEBUG Received libvirt URI qemu:///system Mon, 06 Feb 2012 19:57:29 DEBUG Requesting virt method 'default', hv type 'kvm'. Mon, 06 Feb 2012 19:57:29 ERROR Host does not support any virtualization options for arch 'x86_64' Mon, 06 Feb 2012 19:57:29 DEBUG Traceback (most recent call last): File "/usr/bin/virt-install-2.7", line 272, in get_virt_type machine=options.machine) File "/usr/lib64/python2.7/site-packages/virtinst/CapabilitiesParser.py", line 732, in guest_lookup {'virttype' : osstr, 'arch' : archstr}) ValueError: Host does not support any virtualization options for arch 'x86_64' hp ~ # virsh --connect qemu:///system Welcome to virsh, the virtualization interactive terminal. Type: 'help' for help with commands 'quit' to quit virsh # capabilities <capabilities> <host> <uuid>604c8a12-cb5b-d911-985d-5404a67ef15d</uuid> <cpu> <arch>x86_64</arch> <model>Westmere</model> <vendor>Intel</vendor> <topology sockets='1' cores='4' threads='2'/> <feature name='rdtscp'/> <feature name='x2apic'/> <feature name='xtpr'/> <feature name='tm2'/> <feature name='est'/> <feature name='vmx'/> <feature name='ds_cpl'/> <feature name='monitor'/> <feature name='pbe'/> <feature name='tm'/> <feature name='ht'/> <feature name='ss'/> <feature name='acpi'/> <feature name='ds'/> <feature name='vme'/> </cpu> <migration_features> <live/> <uri_transports> <uri_transport>tcp</uri_transport> </uri_transports> </migration_features> </host> </capabilities> I have no clue what is going on. I have kvm compiled into my kernel. Normal "kvm" commands are working. Intel VT is enabled in the bios. Any thoughts on this one? Kind regards, Felix Blanke

4 13

[libvirt-users] Creating network bridge without IP address
by Zoltan Lippai 08 Feb '12

08 Feb '12

Hi everyone! I have the following setup: domain0 |-domU1 |-domU2 |-domU3 I am running this with libvirt on a XEN server. Right now the server is connected to our office network, but later it will be in a hosted environment. I need domU1 and domU3 to have a public IP, and all of the domUs and the dom0 should have an internal IP address to communicate with each other. So this is what I would like to achieve: domain0 192.168.122.1 |-domU1 192.168.122.2 192.168.1.45 (later this will be changed to the public IP) |-domU2 192.168.122.3 |-domU3 192.168.122.4 192.168.1.46 (this will also be changed to a public IP) On domain0 I have the following network interfaces: br0 Link encap:Ethernet HWaddr 54:04:a6:66:c4:be inet addr:192.168.1.44 Bcast:192.168.1.255 Mask:255.255.255.0 virbr0 Link encap:Ethernet HWaddr 52:54:00:fe:88:0c inet addr:192.168.122.1 Bcast:192.168.122.255 Mask:255.255.255.0 This way everything is working fine, except that I don't want domain0 to have a public IP (which is 192.168.1.44 in this case) So is it possible somehow to setup br0 on domain0 without it having an IP address? Thank you, Zoltan Lippai

2 1

[libvirt-users] virsh & vmxnet3 (VMware) VM
by Florent Paillot 07 Feb '12

07 Feb '12

Hi, I try to install a VMware VM via virsh with a vmxnet3 network card. Virsh sends me an error : "Insufficient video RAM. The maximum resolution of the virtual machine will be limited to 1176x885. To use the configured maximum resolution of 2560x1600, increase the amount of video RAM allocated to this virtual machine by setting svga.vramSize="16384000" in the virtual machine's configuration file." If I edit the VM settings (via vsphere) and set the Video card memory size to "Auto-dectect video setting" I get the error message : " The task was canceled by a user" when I try to start the VM. If I compare a .vmx generated by vsphere and .vmx generated by virsh I see more lines in the vsphere generated .vmx Is it normal ? Thanks .

1 0

Re: [libvirt-users] qcow2 performance
by Trey Dockendorf 07 Feb '12

07 Feb '12

On Mon, Feb 6, 2012 at 2:07 AM, Stefan G. Weichinger <lists(a)xunil.at> wrote: > Am 06.02.2012 03:24, schrieb Trey Dockendorf: >> I wouldn't consider this definitive, but I wrote an article doing >> tests on qcow2 with the variables being preallocation and caching. >> Essentially preallocation + no-cache is the fastest. Article here, >> http://itscblog.tamu.edu/improve-disk-io-performance-in-kvm/. As of >> virt-manager-0.9.0 the only way to generate the qcow2 with >> preallocation is via command line. Details of how are in the article. > > Additional question: > > Doesn't pre-allocation lose its meaning as soon as the image-file is > fully allocated (I mean as soon as it is as big as its maximum size was > defined at start)? > > > Sending response back to list incase someone finds it useful. Preallocation works so that when you create a 100GB image it takes up 100GB from the start. This helps performance because without preallocation the disk would only take up space of whatever is stored on it. So each time you add files the filesystem as to grow the image. With qcow2 the preallocation is just metadata. I've found that ls -lh will show the size correctly but something like "df -h" won't reflect the size of those images, nor will "du -h". - Trey

1 0

[libvirt-users] Network questions
by paul.worner＠thomsonreuters.com 06 Feb '12

06 Feb '12

Hi all, When creating a new network: 1) Is there a way to disable automatic spawning of dnsmasq on network creation? I read that leaving out the <DHCP> section should accomplish this, but that is not what I am seeing. 2) Is there a way to disable automatic installation of iptables rules? 3) For that matter, what is the purpose of the default iptables rule set? Doesn't line 3 let all traffic pass anyway? Thanks, Paul Running libvirt 0.9.2 on Ubuntu 11.10 server. Here's the libvirt network config: RAW CONFIG <network> <name>test</name> <bridge name="virbr%d" stp="off" delay="0"/> <forward mode="route"/> <ip address="192.168.0.1" netmask="255.255.255.0"> </ip> </network> COMMAND virsh net-create test.xml RESULTS virsh net-dumpxml test <network> <name>test</name> <uuid>2eff5e7f-847a-1fbf-ec82-01a46ef0f6c2</uuid> <forward mode='route'/> <bridge name='virbr3' stp='off' delay='0' /> <mac address='52:54:00:47:E6:15'/> <ip address='192.168.0.1' netmask='255.255.255.0'> </ip> </network ps aux | grep dns nobody 4391 0.0 0.0 21616 916 ? S 09:45 0:00 dnsmasq --strict-order --bind-interfaces --pid-file=/var/run/libvirt/network/test.pid --conf-file= --except-interface lo --listen-address 192.168.0.1 iptables -L --line-numbers Chain FORWARD (policy ACCEPT) num target prot opt source destination 1 ACCEPT all -- anywhere 192.168.0.0/24 2 ACCEPT all -- 192.168.0.0/24 anywhere 3 ACCEPT all -- anywhere anywhere 4 REJECT all -- anywhere anywhere reject-with icmp-port-unreachable 5 REJECT all -- anywhere anywhere reject-with icmp-port-unreachable

2 1

[libvirt-users] Performance issues with KVM
by Jatin Kumar 05 Feb '12

05 Feb '12

Hello, We are using libvirt over KVM. If we run a single virtual machine on a host then the performance is very good but if more than one vm (with 1:1 provisioning) runs on same host the performance is very poor (specially in case of parallel programming). libvirt version: 0.9.2 qemu-kvm-0.14.1 We are using virtio module also. What could be the possible reasons for this behavior. Please tell me if you need more info. -- Jatin

1 0

[libvirt-users] How to securely license guests?
by David Knierim 05 Feb '12

05 Feb '12

Hi, I have been asked to research and recommend a method to securely license guests on RHEL/KVM 6.2+ hosts. The guests will be running CentOS or RHEL (5.X or 6.2+). The primary driver to license the software is that we will be selling Telecommunications software in countries where it is common to pirate software. As such, we are worried about the software being purchased one time and then multiple instances of the software being used. We plan to use a USB security dongle (I hate that word!) of some kind. If we were running without virtualization, the solution is pretty straight forward. We would install the security dongle on the server and compile the licensing library which accessed the dongle into our code. We understand that this could be circumvented by a dedicated foe, but it would be reasonably secure. However, as soon as the licensed code is running in a guest, things get a lot more complicated. My first thought was that we could use PCI passthrough to map the USB controller to one guest. Unfortunately, they want to have multiple guests access the same security dongle concurrently. I could see using a bridge that is only on the host to allow the guests to communicate with the host which proxies the requests to the security device. However, it would be pretty easy to connect a network tunnel to the bridge which could allow guests at remote sites to access the security dongle and this would allow the licensed software to be replicated. Can anyone recommend a solution or partial solution to this issue? Any thoughts are welcome, David

1 0

[libvirt-users] Deploying OVF on Different virtual platforms
by Nikunj Badjatya 04 Feb '12

04 Feb '12

Hi All, I am new to Virtualization and Libvirt. Pardon if any information is missing.! I am using VMware PowerCLI And Python to automate tasks for managing my Virtual Infrastructure. I have recently come across Libvirt. and have certain questions. Q1. Can Libvirt be used to deploy OVFs on ESX servers.? Q2. If I wrote a script using Libvirt's Python binding ( A simple one, Say turn on/off/suspend VM) , will that script work on different virtualisation platforms.( HyperV, XEN, VMware-ESX ) ? Q3. Is there a suitable reference for Libvirt's Python APIs ? Thanks, Nikunj Bangalore-India

1 0

Re: [libvirt-users] Can a VM tell what host it's on?
by Paul Lussier 04 Feb '12

04 Feb '12

On Thu, Feb 2, 2012 at 2:08 PM, Eric Blake <eblake(a)redhat.com> wrote: > Did you mean for this to go to the list? Yes, sorry :) > On 02/02/2012 12:04 PM, Paul Lussier wrote: >> On Thu, Feb 2, 2012 at 1:52 PM, Eric Blake <eblake(a)redhat.com> wrote: >>> On 02/02/2012 11:33 AM, Whit Blauvelt wrote: >>>> Is there a way internal to a KVM VM to know which host it's running on? >>> >>> No. The ideal hypervisor is one where the guest doesn't even know it is >>> running as a virtual machine. And consider live migration - a guest >>> might not be running on the same host over its lifetime. Therefore, >>> there should be nothing that requires a guest to know which host it is >>> running on. >> >> From a system administration perspective, though, it's imperative to >> know what physical hosts your VMs are running on. Perhaps the VM >> itself doesn't know, but the sysadmin should be able to have some >> means of figuring this out in a dynamic manner, not simply by "keeping >> track" of where VMs are deployed. > > Yes, but that's a different question. It's not the guests' job to know > which host they are running on, rather, it's the management app _outside > of the guests_ that knows which hosts are running which guests. What do you mean by "management app", virt-manager, or something else ? >>> Why do you think you need it? Perhaps if you ask a better question >>> about what you are really trying to solve, we can give a better answer. >> >> Asset tracking, physical host trouble-shooting, etc. If I'm running >> an environment with 2K physical systems, each of which are running 20+ >> VMs, and someone reports a problem with vm-23475, it would be really >> nice to know that I can ask that VM where it is on my network and on >> what physical hosts. Especially if that VM has been around a while >> and possibly migrated to/from several physical systems. > > That's more a question you should be directing to your management app, > not to your guest. Your management app should know which host is > currently running vm-23475; you shouldn't have to directly query > vm-23475 itself (besides, if you treat guests as untrusted code, you > wouldn't want to rely on any answer vm-23475 gave you in the first place). While I agree with you in principle, my impression is that people are not treating guests as untrusted code, but rather, almost exactly like physical hosts. Perhaps I haven't had the luxury of being in a virtual environment where people are doing things the way they were intended :) -- Paul

3 3

[libvirt-users] Can a VM tell what host it's on?
by Whit Blauvelt 03 Feb '12

03 Feb '12

Is there a way internal to a KVM VM to know which host it's running on? It could send a command via ssh to virsh on a host, and learn from that whether the host currently has it running. But is there something within the VM itself which will reveal this? Thanks, Whit

4 6

[libvirt-users] I want to ask How can I get something like id, uuid, its dom0 in the domU ?
by Ming Q 02 Feb '12

02 Feb '12

hello! I'm new here. I want to ask How can I get something like id, uuid, its dom0 in the domU ?

1 0

[libvirt-users] Gigabit network card not fully operational using KVM/Ubuntu
by Josir Cardoso Gomes 02 Feb '12

02 Feb '12

Hi folks, *Host:* Ubuntu 10.04.3 with KVM lspci | grep net 02:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168B PCI Express Gigabit Ethernet controller (rev 06) *Guest:* Ubuntu 10.04.3 lspci: 00:00.0 Host bridge: Intel Corporation 440FX - 82441FX PMC [Natoma] (rev 02) 00:01.0 ISA bridge: Intel Corporation 82371SB PIIX3 ISA [Natoma/Triton II] 00:01.1 IDE interface: Intel Corporation 82371SB PIIX3 IDE [Natoma/Triton II] 00:01.2 USB Controller: Intel Corporation 82371SB PIIX3 USB [Natoma/Triton II] (rev 01) 00:01.3 Bridge: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 03) 00:02.0 VGA compatible controller: Cirrus Logic GD 5446 00:03.0 Ethernet controller: Qumranet, Inc. Virtio network device *Problem:* The virtio network device operates on 100Mbits. How do I know that it operates on 100Mbits ? I copy a single big file via rsh and I can see the real speed. The network is completely dedicated to this operations, that is, there is no other traffic between other machines. *Question:* What do I have to do to use full bandwidth ? If this is not the best forum to ask this question, do you suggest another channel ? Thanks in advance, Josir Gomes

1 0

[libvirt-users] libvirt 0.9.4-23 on CentOS 6.2 with xen support
by Rudolf Ladyzhenskii 01 Feb '12

01 Feb '12

Hi, all I am trying to install xen on CentOS 6.2. (64 bit on Intel CPU). I am at the stage when xen kernel is app and running and xend is running. Next issue is that I am getting following message when tyring to run virt-manager: no connection driver available for No connection for URI xen:/// Some Google search pointed me to the fact that libvirt as installed from CentOS repository is not built with xen support. I found some quides on how to rebuild libvirt from source with patches for xen support and follow them. I am trying to rebuild libvirt version 0.9.4-23, as this is the rpm that was installed on CentOS. I am getting an error as I build the rpm. xencapstest fails. So, there are few questions: 1. Is there an easier way to get libvirt to work with xen? 2. If rebuilding of RPM is the only option, how do I fix xencaps error? Thanks, Rudolf P.S. Guide I am trying to follow is: http://www.howtoforge.com/virtualization-with-xen-on-centos-6.2-x86_64-para…

1 0