[Openstack-operators] Guest crash and KVM unhandled rdmsr

Arne Wiebalck Arne.Wiebalck at cern.ch
Wed Oct 18 06:43:14 UTC 2017


Blair,

We’ve seen these errors in our deployment as well, on CentOS 7.3 with 3.10 kernels, when looking into instance
issues. So far we’ve always discarded them as not relevant to the problems observed, so I’d be very interested if
it turns out that these should better not be ignored.

Cheers,
 Arne


On 17 Oct 2017, at 18:52, George Mihaiescu <lmihaiescu at gmail.com<mailto:lmihaiescu at gmail.com>> wrote:

Hi Blair,

We had a few cases of compute nodes hanging with the last log in syslog being related to "rdmsr", and requiring hard reboots:
 kvm [29216]: vcpu0 unhandled rdmsr: 0x345

The workloads are probably similar to yours (SGE workers doing genomics) with CPU mode host-passthrough, on top of Ubuntu 16.04 and kernel 4.4.0-96-generic.

I'm not sure the "rdmsr" logs are relevant though, because we see them on other  compute nodes that have no issues.

Did you find anything that might indicate what the root cause is?

Cheers,
George


On Thu, Oct 12, 2017 at 5:26 PM, Blair Bethwaite <blair.bethwaite at gmail.com<mailto:blair.bethwaite at gmail.com>> wrote:
Hi all,

Has anyone seen guest crashes/freezes associated with KVM unhandled rdmsr messages in dmesg on the hypervisor?

We have seen these messages before but never with a strong correlation to guest problems. However over the past couple of weeks this is happening almost daily with consistent correlation for a set of hosts dedicated to a particular HPC workload. So far as I know the workload has not changed, but we have just recently moved the hypervisors to Ubuntu Xenial (though they were already on the Xenial kernel previously) and done minor guest (CentOS7) updates. CPU mode is host-passthrough. Currently trying to figure out if the CPU flags in the guest have changed since the host upgrade...

Cheers,

_______________________________________________
OpenStack-operators mailing list
OpenStack-operators at lists.openstack.org<mailto:OpenStack-operators at lists.openstack.org>
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators


_______________________________________________
OpenStack-operators mailing list
OpenStack-operators at lists.openstack.org<mailto:OpenStack-operators at lists.openstack.org>
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-operators

--
Arne Wiebalck
CERN IT

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-operators/attachments/20171018/8b4bf37d/attachment.html>


More information about the OpenStack-operators mailing list