On Sun, Apr 3, 2016 at 10:07 PM, gregor <gregor_forum@catrix.at> wrote:
Update: The problem occur when a VM reboots.
When I change the CPU Type from default "Intel Haswell-noTSX" to
"Westmere" the error is gone.

The error ""kvm ... vcpu0 unhandled rdmsr ..." is quite harmless. 
I assume you are running the latest qemu/kvm packages.
Can you ensure NX is enabled on your host?
In any case, this is most likely a qemu/kvm issue - the command line of the VM and information regarding the qemu packages and host versions will be needed.
Y.
 

But which CPU type is now the best so I don't lose performance.

Host CPU: Intel(R) Xeon(R) CPU E5-2603 v3 @ 1.60GHz

regards
gregor

On 03/04/16 20:36, gregor wrote:
> Hi,
>
> on one Host I get very often the message
>
> "kvm ... vcpu0 unhandled rdmsr ..."
>
> When this occurs some VM's are stuck. Really bad is this for an Windows
> Server 2012 R2 VM which stucks so heavy that the VM is getting corrupt
> and the VM is unable to boot anymore and a Windows Recovery in any way
> didn't help. Therefor I had to reinstall the VM, this works for some
> day's but now the VM is still damaged. So I can't use Windows Server
> 2012 R2 on this machine but the customer needs it and I have some days
> to ship it to my customer. So I have to decide to stay on oVirt or use
> another product. Besides, oVirt run on my others hosts (without Windows
> VM) very well since a long time.
> On a CentOS 7 VM I have similar problems where the NIC is getting
> offline sometime + the XFS filesystems get some errors and I have to fix
> this in recovery mode.
>
> oVirt: 3.6.4.1-1.el7.centos
> machine: HP ProLiant ML110 Gen9
> VM's: 3 CentOS 7 and one Windows Server 2012 R2
>
> I hope somebody can help.
>
> regards
> gregor
> _______________________________________________
> Users mailing list
> Users@ovirt.org
> http://lists.ovirt.org/mailman/listinfo/users
>
_______________________________________________
Users mailing list
Users@ovirt.org
http://lists.ovirt.org/mailman/listinfo/users