[libvirt-users] libvirt-qemu/kvm passthru w/Intel 82598EB
by laurence.schuler
Hi,
I am trying to configure two identical systems with the following setup.
dom0: Centos 6 OS on Apple Xserve i7 server with PCI cards for 4GB FC
(LSI FC949ES) and 10GB Ethernet (Intel 82598EB)
dom1: Centos 5 OS with passthru mode of PCI 4GB FC and 10GB Ethernet
(ixgbe and mptfc blacklisted in dom0)
Everything starts ok, but networking in dom1 using the 10GB interface
has extremely high rtt w/ping tests (20-30ms avg).
I've tried a few tweaks I've discovered via googling, but nothing seems
to help. (run libvirt as root, set privilege mode, etc)
For validation, I have the same card in Dell hosts running Centos 6
natively and they get an avg rtt of .06ms or better. Furthemore the Dell
will contact the dom1 but the connection has rtt avg times of about 6ms.
With the ethernet interface on the motherboard of the Apple Xserve, I
will see rtt times of .06ms within the dom1's, so I think this points to
a problem with the Intel card passthru usage, perhaps interrupts handling?.
Does anyone know if I need to set any special options for the ixgbe
driver under libvirt?
TIA,
--
Laurence Schuler (Larry) Laurence.Schuler(a)nasa.gov
Systems Support ADNET Systems, Inc
Scientific Visualization Studio http://svs.gsfc.nasa.gov
NASA/Goddard Space Flight Center, Code 606.4 phone: 1-301-286-1799
Greenbelt, MD 20771 fax: 1-301-286-1634
Note: I am not a government employee and have no authority to obligate
any federal, state or local government to perform any action or payment.FC949ES
10 years, 10 months
[libvirt-users] error: virDBusCallMethod
by Thomas Stein
Hello.
Since upgrading to libvirt-1.1.2 i get this error:
2013-09-03 08:29:55.258+0000: 2281: error : virDBusCallMethod:1156 : The name
org.freedesktop.machine1 was not provided by any .service files
Also live migration doesn't seem to work anymore. There is no error except the
above one but the machine is not accessible anymore after migration. The
machine is running if i believe "virsh list" but as i said, no ping no
nothing. How can i debug this further?
thanks and cheers
t.
10 years, 10 months
[libvirt-users] libvirt_lxc heavy cpu usage ?
by zhongjj
I am running lxc via libvirt1.1.1, the cpu usage is heavy of process libvirt_lxc.
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
2384 root 20 0 10480 2048 1200 R 99.6 0.1 1291:39 libvirt_lxc
32157 root 20 0 10448 2032 1200 R 99.3 0.1 1316:02 libvirt_lxc
28228 root 20 0 10448 2064 1216 R 99.0 0.1 5663:43 libvirt_lxc
30461 root 20 0 10480 2048 1200 R 98.7 0.1 5663:15 libvirt_lxc
btw: 0.9.3 is the same ,I googled it , only two reports .
any suggestions ? thx
10 years, 10 months
[libvirt-users] Running libvirt on a virtualized server
by Andre Gustavo Pereira da Silva
Hi everyone
I'm trying to create a virtual machine using OpenNebula [1] on a centos 6.4
server, but i'm facing some erros using kvm / libvirt.
This server is running on a virtual machine created on XCP. Then, my
infrastructure is: XCP 1.6 (hypervisor - dom0) + CentOS 6.4 (running on a
XCP VM) + OpenNebula 4.0.1 (installed on CentOS).
In the host machine, I have the same infrastructure.
The problem I'm facing is that, by following the tutorial [2], i've noticed
that neither the parameter 'vmx' nor 'svm' is listed on the CPU
information provided by the virtual machine where the centOs is running.
However, if i run the same command in XCP hypervisor, the parameter 'vmx'
is listed. That is, the CPU informations provided by XCP are different from
those provided by its virtual machines.
My question is: is it possible to use libvirt in this scenario?
Here are some messages:
- Libvirt log on centoOS (XCP VM):
Aug 27 13:26:21 localhost libvirtd: 9945: info : libvirt version: 0.10.2,
package: 18.el6_4.9 (CentOS BuildSystem <http://bugs.centos.org>,
2013-07-02-11:19:29, c6b8.bsys.dev.centos.org)
Aug 27 13:26:21 localhost libvirtd: 9945: error : udevGetDMIData:1502 :
Failed to get udev device for syspath '/sys/devices/virtual/dmi/id' or
'/sys/class/dmi/id'
Aug 27 13:26:21 localhost kernel: Ebtables v2.0 registered
Aug 27 13:26:22 localhost kernel: lo: Disabled Privacy Extensions
Aug 27 13:26:40 localhost polkitd[10145]: started daemon version 0.96 using
authority implementation `local' version `0.96'
Aug 27 13:28:35 localhost libvirtd: 9937: error : qemuProcessStart:3696 :
unsupported configuration: Domain requires KVM, but it is not available.
Check that virtualization is enabled in the host BIOS, and host
configuration is setup to load the kvm modules.
Aug 27 13:28:35 localhost libvirtd: 9937: error : qemuRemoveCgroup:768 :
internal error Unable to find cgroup for one-8
Aug 27 13:28:35 localhost libvirtd: 9937: warning : qemuProcessStop:4349 :
Failed to remove cgroup for one-8
- On XCP hypervisor:
[root@xenserver2 ~]# egrep '(vmx|svm)' --color=always /proc/cpuinfo
flags: fpu de tsc msr pae mce cx8 apic sep mtrr mca cmov pat clflush acpi
mmx fxsr sse sse2 ss ht nx constant_tsc nonstop_tsc aperfmperf pni vmx est
ssse3 sse4_1 sse4_2 popcnt hypervisor arat tpr_shadow vnmi flexpriority ept
vpid
flags: fpu de tsc msr pae mce cx8 apic sep mtrr mca cmov pat clflush acpi
mmx fxsr sse sse2 ss ht nx constant_tsc nonstop_tsc aperfmperf pni vmx est
ssse3 sse4_1 sse4_2 popcnt hypervisor arat tpr_shadow vnmi flexpriority ept
vpid
flags: fpu de tsc msr pae mce cx8 apic sep mtrr mca cmov pat clflush acpi
mmx fxsr sse sse2 ss ht nx constant_tsc nonstop_tsc aperfmperf pni vmx est
ssse3 sse4_1 sse4_2 popcnt hypervisor arat tpr_shadow vnmi flexpriority ept
vpid
flags: fpu de tsc msr pae mce cx8 apic sep mtrr mca cmov pat clflush acpi
mmx fxsr sse sse2 ss ht nx constant_tsc nonstop_tsc aperfmperf pni vmx est
ssse3 sse4_1 sse4_2 popcnt hypervisor arat tpr_shadow vnmi flexpriority ept
vpid
- On centOS (XCP VM):
[root@centos2 ~]# cat /proc/cpuinfo
processor: 0
vendor_id: GenuineIntel
cpu family: 6
model: 44
model name: Intel(R) Xeon(R) CPU E5606 @ 2.13GHz
stepping: 2
cpu MHz: 2133.486
cache size: 8192 KB
fpu: yes
fpu_exception: yes
cpuid level: 11
wp: yes
flags: fpu de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht
syscall nx lm rep_good aperfmperf unfair_spinlock pni ssse3 cx16 pcid
sse4_1 sse4_2 popcnt hypervisor lahf_lm arat epb dts
bogomips: 4266.97
clflush size: 64
cache_alignment: 64
address sizes: 40 bits physical, 48 bits virtual
thanks
André Gustavo
Recife-PE
Brazil
[1] http://opennebula.org/documentation:archives:rel1.4:libvirtapi
[2]http://www.howtoforge.com/virtualization-with-kvm-on-a-centos-6.4-server
10 years, 11 months
[libvirt-users] How to add uplink in ESXi virtual switch
by varun bhatnagar
Hi,
I have an ESXi server, on top of it I have created one VM whose IP address
is 192.168.79.1.
I want to connect to the VM on ESXi server from my host machine but I am
not able to do it.
I have created a network "networkforSC" and I have pasted my xml file
content below but I am not able to add any uplink (how to add it?) .
Below is the xml file of my node as well as the network which I am creating
on my ESXi server
*<domain type='vmware'>*
*<name>testNode</name>*
*<uuid>50115e16-9bdc-49d7-f171-53c4d7f91710</uuid>*
*<memory>1048576</memory>*
*<currentMemory>1048576</currentMemory>*
*<vcpu>1</vcpu>*
*<console>/dev/console</console>*
*<os>*
*<type arch='x86_64'>hvm</type>*
*</os>*
*<devices>*
*<disk type='file' device='disk'>*
*<source file='[datastore1] testNode/testNode.vmdk'/>*
*<target dev='sda' bus='scsi'/>*
*<address type='drive' controller='0' bus='0' unit='0'/>*
*</disk>*
*<controller type='scsi' index='0' model='lsilogic'/>*
*<!--controller type='scsi' index='0'/-->*
*<interface type='bridge'>*
*<mac address='00:50:56:25:48:c7'/>*
*<source bridge='VM Network'/>*
*</interface>*
*<interface type='bridge'>*
*<source bridge='test'/>*
*</interface>*
*</devices>*
*</domain>*
##########################################
Network File:
*<network>*
*<name>networkforSC</name>*
*<portgroup name='test'>*
*</portgroup>*
*</network>*
I have checked the network which got created on the server.
As I have not attached any uplink (vmnic0) I can not see it (I want to
attach it).
#esxcli network vswitch standard list
*
*
*networkforSC*
*Name: networkforSC*
*Class: etherswitch*
*Num Ports: 128*
*Used Ports: 2*
*Configured Ports: 128*
*MTU: 1500*
*CDP Status: listen*
*Beacon Enabled: false*
*Beacon Threshold: 3*
*Beacon Required By:*
*Uplinks:*
*Portgroups: test*
Can anyone please help me out in adding the uplink and also can anyone tell
me whether I am going in a right direction so that I could ping my virtual
machine created on ESXi server.
Regards,
Varun
10 years, 11 months
[libvirt-users] remote xenserver problem
by linxiao.st＠centerm.com.cn
HI：
I’m a user of your libvirt, and it help me a lot of time. Thank you for your greate work!
However,I found some understand problems,the following is my test environment:
1.I use the version of libvirt is libvirt1.1.1,when I want to use the command like this:
#virsh -c xenapi://<my localip>/?no_verify=1
#virsh :list
#enter username for<my localip>:root
#enter password for<my localip>:
it return correct list,but when i input virsh --all list this
#virsh list --all
it most of time return correct lists,but sometimes returns interal error:TRANSPORT_FAULT:56 why?
With my good wishes!
linxiao.st(a)centerm.com.cn
10 years, 11 months
[libvirt-users] Redhat VM is not getting IP from DHCP
by Praveen Kumar
Hi,
Below is the steps which I performed but not getting the IP of the guest.
* Created a fresh VM (RHEL-6.4-server-x86_64)
* Checked ifconfig output ( Getting IP from DHCP server)
* Took xml dump.
* Changed UUID in the xml
* Changed MAC in the xml
* Changed Name
* defined a another VM using changed xml.
* boot it up and checked the ifconfig status.( Not able to get IP)
If same steps I follow in fedora guest then it's working as expected (getting the IP from DHCP). Any Suggestions?
--
Praveen Kumar
10 years, 11 months
[libvirt-users] how to setup network filter
by Jiaan Zeng
Hi All,
I am new to libvirt and encounter a strange problem to set up network
filter in a NAT network.
I launched VMs in a single host using NAT, i.e. interface
type='network'. Now I want to control the outbound traffic from VM
instance - only allow the VM to asses a set of ip addresses. My
network filter xml is as follows. The problem is once I change the VM
xml, shutdown and start VM, VM cannot get ip address.
/var/log/libvirt/libvirt.log shows " error : virNetDevGetIndex:656 :
Unable to get index for interface vnet2: No such device" error.
But when I remove the drop rule in the filter xml, VM can get IP
address. I even tried the clean-traffic filter shipped with libvirt.
VM throws the same error above.
Any idea why this happens? How can I implement outbound traffic
control in libvirt? Thanks a lot.
<filter name='filter-test'>
<rule action='accept' direction='in' priority='500'>
<tcp dstportstart='22'/>
</rule>
<rule action='accept' direction='out' priority='500'>
<ip dstipaddr='IP1'/>
</rule>
<rule action='accept' direction='out' priority='500'>
<ip dstipaddr='IP2'/>
</rule>
<rule action='drop' direction='out' priority='500'>
<all/>
</rule>
</filter>
The VM network section XML looks like this
<interface type='network'>
<mac address='52:54:00:0d:f1:ce'/>
<source network='default'/>
<filterref filter='filter-test'/>
<address type='pci' domain='0x0000' bus='0x00' slot='0x03'
function='0x0'/>
</interface>
--
Regards,
Jiaan
10 years, 11 months
