Comment 14 for bug 1836754

Eyeballing the logstash query [1], the majority look to occur in the DeleteServersTestJSON [2]. There is a test in there that deletes a server while it is building.

But it looks like there are cases where we fail to delete a server after it goes into VERIFY_RESIZE state [3][4]:

Traceback (most recent call last):
  File "/opt/stack/tempest/tempest/api/compute/servers/test_delete_server.py", line 106, in test_delete_server_while_in_verify_resize_state
    waiters.wait_for_server_termination(self.client, server['id'])
  File "/opt/stack/tempest/tempest/common/waiters.py", line 124, in wait_for_server_termination
    raise lib_exc.DeleteErrorException(
tempest.lib.exceptions.DeleteErrorException: Resource %(resource_id)s failed to delete and is in ERROR status
Details: Server 2f13b96e-201b-49a1-b2e1-08c12b8527b1 failed to delete and is in ERROR status

In this case, I don't yet understand how the conflict is occurring.

[1] http://logstash.openstack.org/#/dashboard/file/logstash.json?query=message:%5C%22Unable%20to%20delete%20allocation%20for%20instance%5C%22%20AND%20message:%5C%22%5C%5C%5C%22code%5C%5C%5C%22:%20%5C%5C%5C%22placement.concurrent_update%5C%5C%5C%22%5C%22%20AND%20tags:%5C%22screen-n-api.txt%5C%22%20AND%20voting:1%20AND%20build_status:%5C%22FAILURE%5C%22&from=864000s
[2] https://github.com/openstack/tempest/blob/8e76271b5cf031901c29bea851eec8b44d1e40f3/tempest/api/compute/servers/test_delete_server.py#L28
[3] http://storage.gra.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_474/802056/12/check/nova-ceph-multistore/4747c0b/testr_results.html
[4] https://github.com/openstack/tempest/blob/8e76271b5cf031901c29bea851eec8b44d1e40f3/tempest/api/compute/servers/test_delete_server.py#L99