Hi Matthias, This may be https://bugs.launchpad.net/cinder/+bug/1704106. In our deployment it was necessary to disable the the gathering of the provisioning stats and we simply commented it out: #self._get_usage_info() in def _update_volume_stats(self) from cinder/volume/drivers/rbd.py HTH, Arne > On 17 Nov 2017, at 13:55, Matthias Leopold <matthias.leopold at meduniwien.ac.at> wrote: > > Hi, > > we are running a Cinder Pike instance with Ceph (luminous) backend for use with oVirt (not openstack). We are experiencing the "cinder-volume causes high CPU load" issue. In Cinder 11.0.0 we could successfully apply the fix mentioned in https://ask.openstack.org/en/question/110709/ocata-cinder-volume-causes-high-cpu-load-ceph-backend/. In Cinder 11.0.1 /usr/lib/python2.7/site-packages/cinder/volume/drivers/rbd.py changed, but the high CPU load problem reappeared. I tried changing "report_dynamic_total_capacity" (mentioned in the changelog), but it didn't make a difference. Has anyone seen/fixed this? > > thx > matthias > > > _______________________________________________ > Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack > Post to : openstack at lists.openstack.org > Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack -- Arne Wiebalck CERN IT