looks like this is dealt in https://review.openstack.org/#/c/505771/
I could get the same error with initial cold migration and then tried to live migrate an instance back with AUTO_SCHEDULE (only two computes in my test) ** Changed in: nova Status: Expired => New -- You received this bug notification because you are a member of Yahoo! Engineering Team, which is subscribed to OpenStack Compute (nova). https://bugs.launchpad.net/bugs/1678577 Title: nova live migration failed in some case Status in OpenStack Compute (nova): New Bug description: env: nova 15.0.2 + libvirt + kvm + centos in some situation, nova request spec become {"nova_object.version": "1.8", "nova_object.changes": ["instance_uuid", "requested_destination", "retry", "num_instances", "pci_requests", "limits", "availability_zone", "force_nodes", "image", "instance_group", "force_hosts", "numa_topology", "flavor", "project_id", "scheduler_hints", "ignore_hosts"], "nova_object.name": "RequestSpec", "nova_object.data": {"requested_destination": null, "instance_uuid": "ca01b22b-d2d4-4291-96bd-ff6111f1f88b", "retry": {"nova_object.version": "1.1", "nova_object.changes": ["num_attempts", "hosts"], "nova_object.name": "SchedulerRetries", "nova_object.data": {"num_attempts": 1, "hosts": {"nova_object.version": "1.16", "nova_object.changes": ["objects"], "nova_object.name": "ComputeNodeList", "nova_object.data": {"objects": [{"nova_object.version": "1.16", "nova_object.changes": ["host", "hypervisor_hostname"], "nova_object.name": "ComputeNode", "nova_object.data": {"host": "control01", "hypervisor_hostname": "control01"}, "nova_object.namespace": "nova"}]}, "nova_object.namespace": "nova"}}, "nova_object.namespace": "nova"}, "num_instances": 1, "pci_requests": {"nova_object.version": "1.1", "nova_object.name": "InstancePCIRequests", "nova_object.data": {"instance_uuid": "ca01b22b-d2d4-4291-96bd-ff6111f1f88b", "requests": []}, "nova_object.namespace": "nova"}, "limits": {"nova_object.version": "1.0", "nova_object.changes": ["memory_mb", "vcpu", "disk_gb", "numa_topology"], "nova_object.name": "SchedulerLimits", "nova_object.data": {"vcpu": null, "memory_mb": 245427, "disk_gb": 8371, "numa_topology": null}, "nova_object.namespace": "nova"}, "availability_zone": null, "force_nodes": null, "image": {"nova_object.version": "1.8", "nova_object.changes": ["min_disk", "container_format", "min_ram", "disk_format", "properties"], "nova_object.name": "ImageMeta", "nova_object.data": {"min_disk": 1, "container_format": "bare", "min_ram": 0, "disk_format": "raw", "properties": {"nova_object.version": "1.16", "nova_object.name": "ImageMetaProps", "nova_object.data": {}, "nova_object.namespace": "nova"}}, "nova_object.namespace": "nova"}, "instance_group": null, "force_hosts": null, "numa_topology": null, "ignore_hosts": null, "flavor": {"nova_object.version": "1.1", "nova_object.name": "Flavor", "nova_object.data": {"disabled": false, "root_gb": 1, "name": "m1.tiny", "flavorid": "a70249ef-5ea9-49cb-b35f-ab4732064981", "deleted": false, "created_at": "2017-03-22T08:13:48Z", "ephemeral_gb": 0, "updated_at": null, "memory_mb": 256, "vcpus": 1, "extra_specs": {}, "swap": 0, "rxtx_factor": 1.0, "is_public": true, "deleted_at": null, "vcpu_weight": 0, "id": 119}, "nova_object.namespace": "nova"}, "project_id": "f3c6d500b267432c858c588800b49653", "scheduler_hints": {}}, "nova_object.namespace": "nova"} check the retry part retry": {"nova_object.version": "1.1", "nova_object.changes": ["num_attempts", "hosts"], "nova_object.name": "SchedulerRetries", "nova_object.data": {"num_attempts": 1, "hosts": {"nova_object.version": "1.16", "nova_object.changes": ["objects"], "nova_object.name": "ComputeNodeList", "nova_object.data": {"objects": [{"nova_object.version": "1.16", "nova_object.changes": ["host", "hypervisor_hostname"], "nova_object.name": "ComputeNode", "nova_object.data": {"host": "control01", "hypervisor_hostname": "control01"}, "nova_object.namespace": "nova"}]} it has control01 as host even it is in control02 when live migrate this vm from controll02 to control01, get error in "migration-list", after check the nova-scheduler logs, got 2017-04-02 14:01:47.010 6 DEBUG nova.filters [req-191c8f6e-010b-42f6-acc6-c84c689f649c 2442cfcb9d5c4daf8d90af8bcfe30df7 8eb03bbcdfd84f68b88a7fbaa74e2327 - - -] Starting with 1 host(s) get_filtered_objects /var/lib/kolla/venv/lib/python2.7/site-packages/nova/filters.py:70 2017-04-02 14:01:47.010 6 INFO nova.scheduler.filters.retry_filter [req-191c8f6e-010b-42f6-acc6-c84c689f649c 2442cfcb9d5c4daf8d90af8bcfe30df7 8eb03bbcdfd84f68b88a7fbaa74e2327 - - -] Host [u'control01', u'control01'] fails. Previously tried hosts: [[u'control01', u'control01']] I think the root cause is the retry part, and still do not know how it happen. To manage notifications about this bug go to: https://bugs.launchpad.net/nova/+bug/1678577/+subscriptions -- Mailing list: https://launchpad.net/~yahoo-eng-team Post to : yahoo-eng-team@lists.launchpad.net Unsubscribe : https://launchpad.net/~yahoo-eng-team More help : https://help.launchpad.net/ListHelp