Hi all, I'm trying the volume backed live migration via Ceph Block Storage and stunk in the issues of libvirt. The environment of my cluster is as follows: Hosts in cluster: Havana(Controller & compute) and Compute1(pure Compute node) Ceph1, ceph2, ceph3: ceph cluster
1. The ceph cluster: checked by command: ceph health 2. The Openstack cluster: ceph client is installed on each node; Configuration done for ceph in glance, cinder, nova Qemu-kvm, qemu-img, qemu-kvm-tools are installed from ceph source. 3. Image creation correctly; Volume creation correctly; Instance lunched from volume on host havana; 4. Unexpected, the instance migration from Havana to compute1 failed with the following ERROR(I'm struggling to hard work on the issues and it would be much appreciated if any clue!!!): Compute.log on host compute1: 2014-03-25 18:46:44.346 16948 AUDIT nova.compute.manager [req-defe320c-bae6-4d57-817e-1cca779e204a 714bab91932043e98ad2d855a81f19b0 888df5c4bc47459485b96ffa03c671e6] [instance: aa44d1f6-1f4f-434d-b05e-f785f7b0a2a7] Detach volume f5a34a51-b662-43d4-a7d5-4199de8b1d4b from mountpoint vda 2014-03-25 18:46:44.364 16948 WARNING nova.compute.manager [req-defe320c-bae6-4d57-817e-1cca779e204a 714bab91932043e98ad2d855a81f19b0 888df5c4bc47459485b96ffa03c671e6] [instance: aa44d1f6-1f4f-434d-b05e-f785f7b0a2a7] Detaching volume from unknown instance 2014-03-25 18:46:44.375 16948 ERROR nova.compute.manager [req-defe320c-bae6-4d57-817e-1cca779e204a 714bab91932043e98ad2d855a81f19b0 888df5c4bc47459485b96ffa03c671e6] [instance: aa44d1f6-1f4f-434d-b05e-f785f7b0a2a7] Failed to detach volume f5a34a51-b662-43d4-a7d5-4199de8b1d4b from vda 2014-03-25 18:46:44.375 16948 TRACE nova.compute.manager [instance: aa44d1f6-1f4f-434d-b05e-f785f7b0a2a7] Traceback (most recent call last): 2014-03-25 18:46:44.375 16948 TRACE nova.compute.manager [instance: aa44d1f6-1f4f-434d-b05e-f785f7b0a2a7] File "/usr/lib/python2.6/site-packages/nova/compute/manager.py", line 3737, in _detach_volume 2014-03-25 18:46:44.375 16948 TRACE nova.compute.manager [instance: aa44d1f6-1f4f-434d-b05e-f785f7b0a2a7] encryption=encryption) 2014-03-25 18:46:44.375 16948 TRACE nova.compute.manager [instance: aa44d1f6-1f4f-434d-b05e-f785f7b0a2a7] File "/usr/lib/python2.6/site-packages/nova/virt/libvirt/driver.py", line 1202, in detach_volume 2014-03-25 18:46:44.375 16948 TRACE nova.compute.manager [instance: aa44d1f6-1f4f-434d-b05e-f785f7b0a2a7] virt_dom = self._lookup_by_name(instance_name) 2014-03-25 18:46:44.375 16948 TRACE nova.compute.manager [instance: aa44d1f6-1f4f-434d-b05e-f785f7b0a2a7] File "/usr/lib/python2.6/site-packages/nova/virt/libvirt/driver.py", line 3101, in _lookup_by_name 2014-03-25 18:46:44.375 16948 TRACE nova.compute.manager [instance: aa44d1f6-1f4f-434d-b05e-f785f7b0a2a7] raise exception.InstanceNotFound(instance_id=instance_name) 2014-03-25 18:46:44.375 16948 TRACE nova.compute.manager [instance: aa44d1f6-1f4f-434d-b05e-f785f7b0a2a7] InstanceNotFound: Instance instance-0000000c could not be found. 2014-03-25 18:46:44.375 16948 TRACE nova.compute.manager [instance: aa44d1f6-1f4f-434d-b05e-f785f7b0a2a7] Libvirtd.log on host compute1: 2014-03-25 10:32:05.952+0000: 1891: warning : qemuDomainObjTaint:1377 : Domain id=1 name='instance-0000000b' uuid=4ddb08dc-c7ef-4cdf-8108-80a296eaf457 is tainted: high-privileges 2014-03-25 10:32:07.188+0000: 1891: warning : qemuDomainObjEnterMonitorInternal:1005 : This thread seems to be the async job owner; entering monitor without asking for a nested job is dangerous 2014-03-25 10:33:34.411+0000: 1891: warning : qemuDomainObjEnterMonitorInternal:1005 : This thread seems to be the async job owner; entering monitor without asking for a nested job is dangerous 2014-03-25 10:33:34.414+0000: 1891: warning : qemuDomainObjEnterMonitorInternal:1005 : This thread seems to be the async job owner; entering monitor without asking for a nested job is dangerous 2014-03-25 10:33:34.416+0000: 1891: warning : qemuSetupCgroupForVcpu:566 : Unable to get vcpus' pids. 2014-03-25 10:33:34.419+0000: 1891: warning : qemuDomainObjEnterMonitorInternal:1005 : This thread seems to be the async job owner; entering monitor without asking for a nested job is dangerous 2014-03-25 10:33:34.419+0000: 1891: warning : qemuDomainObjEnterMonitorInternal:1005 : This thread seems to be the async job owner; entering monitor without asking for a nested job is dangerous [root@compute1 ~(keystone_admin)]# rpm -qa |grep qemu qemu-img-0.12.1.2-2.415.el6.3ceph.x86_64 qemu-guest-agent-0.12.1.2-2.415.el6.3ceph.x86_64 qemu-kvm-tools-0.12.1.2-2.415.el6.3ceph.x86_64 qemu-kvm-0.12.1.2-2.415.el6.3ceph.x86_64 gpxe-roms-qemu-0.9.7-6.10.el6.noarch [root@compute1 ~(keystone_admin)]# rpm -qa |grep libvirt libvirt-client-0.10.2-29.el6_5.3.x86_64 libvirt-python-0.10.2-29.el6_5.3.x86_64 libvirt-0.10.2-29.el6_5.3.x86_64 _______________________________________________ Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack Post to : openstack@lists.openstack.org Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
