[Users] TSC clocksource gets lost after live migration

Darrell Budic darrell.budic at zenfire.com
Sat Apr 5 17:09:11 UTC 2014


Whups, should note that I don’t see the CPU utilization issues, just the clock source errors. This is on a 3.3.4 install with centos 6.4 guests.

On Apr 4, 2014, at 6:19 PM, Darrell Budic <darrell.budic at zenfire.com> wrote:

> I see this on some guests as well, possibly relating to moving between hosts with the same family of CPU but different absolute CPU speeds?
> 
>   -Darrell
> 
> On Apr 4, 2014, at 8:33 AM, Michal Skrivanek <michal.skrivanek at redhat.com> wrote:
> 
>> Hi,
>> this is more for the KVM folks I suppose…can you get the qemu process cmdline please?
>> 
>> Thanks,
>> michal
>> 
>> On 3 Apr 2014, at 12:13, Markus Stockhausen wrote:
>> 
>>> Hello,
>>> 
>>> we have an up to date ovirt 3.4 installation. Inside we are running SLES11 SP3
>>> VMs (Kernel 3.0.76-0.11). After live migration of these VMs they all of a sudden
>>> do not react any longer and CPU usage of the VM goes to 100%.
>>> 
>>> We identified kvm-clock source to be the culprit and therefore switched to another
>>> clocksource. We ended with hpet but are not happy with that as our inital goal
>>> was to use the more simple designed TSC clocksoure. 
>>> 
>>> The reason behind that is the question I have for you experts.
>>> 
>>> Our hosts all have the constant_tsc CPU flag available. Just to mention these
>>> are not identical hosts. We have a mix of Xeon 5500 and 5600 machines. E.G.
>>> [root at colovn01 ~]# cat /proc/cpuinfo | grep constant_tsc | wc -l
>>> 8
>>> 
>>> When we start the VM the client sees TSC as available clocksource:
>>> 
>>> colvm53:~ # cat /sys/devices/system/clocksource/clocksource0/available_clocksource
>>> kvm-clock tsc hpet acpi_pm
>>> 
>>> After the first live migration to another host that also has constant_tsc (see above)
>>> that flag is lost inside the VM.
>>> 
>>> colvm53:~ # cat /sys/devices/system/clocksource/clocksource0/available_clocksource
>>> kvm-clock hpet acpi_pm
>>> 
>>> Any ideas?
>>> 
>>> Markus
>>> 
>>> 
>>> <InterScan_Disclaimer.txt>_______________________________________________
>>> Users mailing list
>>> Users at ovirt.org
>>> http://lists.ovirt.org/mailman/listinfo/users
>> 
>> _______________________________________________
>> Users mailing list
>> Users at ovirt.org
>> http://lists.ovirt.org/mailman/listinfo/users
> 
> _______________________________________________
> Users mailing list
> Users at ovirt.org
> http://lists.ovirt.org/mailman/listinfo/users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ovirt.org/pipermail/users/attachments/20140405/e70726fd/attachment-0001.html>


More information about the Users mailing list