Files
tempest/tempest/scenario/test_volume_migrate_attached.py
T
Ghanshyam Mann 2803b57d6c Skip scenario tests early to avoid unnecessary setup
This is change for the volume scenario test to
skip them early.

When we skip the test class using skip_checks(), it check
the conditions and skip the test class at first step
without creating any keystone credentials. But when
tests are skipped with other decorator at test level then
it does create keystone credentials, setup network resources
and service clients.

This will mostly help neutron gate where these volume
tests will be skipped in the initial stage only and will
not create the keystone and network resources.

One good example is TestEncryptedCinderVolumes which is skipped
 - https://zuul.openstack.org/build/babcc06f24764a408ed77702365b4c5b/log/job-output.txt#28695

But it still does the resources setup
- https://zuul.openstack.org/build/babcc06f24764a408ed77702365b4c5b/log/controller/logs/tempest_log.txt#6374-6450

Related-Bug: #2004780
Change-Id: I59cd39c20b995bf2ed2f58f4522743c3ca51b516
2023-08-04 19:38:16 +00:00

219 lines
9.9 KiB
Python

# Licensed under the Apache License, Version 2.0 (the "License"); you may
# not use this file except in compliance with the License. You may obtain
# a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
# WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
# License for the specific language governing permissions and limitations
# under the License.
from oslo_log import log as logging
from tempest.common import utils
from tempest.common import waiters
from tempest import config
from tempest.lib import decorators
from tempest.scenario import manager
CONF = config.CONF
LOG = logging.getLogger(__name__)
class TestVolumeMigrateRetypeAttached(manager.ScenarioTest):
"""This test case attempts to reproduce the following steps:
* Create 2 volume types representing 2 different backends
* Create in Cinder some bootable volume importing a Glance image using
* volume_type_1
* Boot an instance from the bootable volume
* Write to the volume
* Perform a cinder retype --on-demand of the volume to type of backend #2
* Check written content of migrated volume
* Check the type of the volume has been updated.
* Check the volume is still in-use and the migration was successful.
* Check that the same volume is attached to the instance.
"""
credentials = ['primary', 'admin']
@classmethod
def setup_clients(cls):
super(TestVolumeMigrateRetypeAttached, cls).setup_clients()
cls.admin_volumes_client = cls.os_admin.volumes_client_latest
@classmethod
def skip_checks(cls):
super(TestVolumeMigrateRetypeAttached, cls).skip_checks()
if not CONF.service_available.cinder:
raise cls.skipException("Cinder is not available")
if not CONF.volume_feature_enabled.multi_backend:
raise cls.skipException("Cinder multi-backend feature disabled")
if len(set(CONF.volume.backend_names)) < 2:
raise cls.skipException("Requires at least two different "
"backend names")
def _boot_instance_from_volume(self, vol_id, keypair, security_group):
key_name = keypair['name']
security_groups = [{'name': security_group['name']}]
block_device_mapping = [{'device_name': 'vda', 'volume_id': vol_id,
'delete_on_termination': False}]
return self.create_server(image_id='',
key_name=key_name,
security_groups=security_groups,
block_device_mapping=block_device_mapping)
def _create_volume_types(self):
backend_names = CONF.volume.backend_names
backend_source = backend_names[0]
backend_dest = backend_names[1]
source_body = self.create_volume_type(backend_name=backend_source)
dest_body = self.create_volume_type(backend_name=backend_dest)
LOG.info("Created Volume types: %(src)s -> %(src_backend)s, %(dst)s "
"-> %(dst_backend)s", {'src': source_body['name'],
'src_backend': backend_source,
'dst': dest_body['name'],
'dst_backend': backend_dest})
return ({'name': source_body['name'], 'host': backend_source},
{'name': dest_body['name'], 'host': backend_dest})
def _volume_retype_with_migration(self, volume_id, new_volume_type):
# NOTE: The 'on-demand' migration requires admin operation, so
# admin_volumes_client() should be used here.
migration_policy = 'on-demand'
self.admin_volumes_client.retype_volume(
volume_id, new_type=new_volume_type,
migration_policy=migration_policy)
waiters.wait_for_volume_retype(self.volumes_client,
volume_id, new_volume_type)
@decorators.attr(type='slow')
@decorators.idempotent_id('deadd2c2-beef-4dce-98be-f86765ff311b')
@utils.services('compute', 'volume')
def test_volume_retype_attached(self):
LOG.info("Creating keypair and security group")
keypair = self.create_keypair()
security_group = self.create_security_group()
# create volume types
LOG.info("Creating Volume types")
source_type, dest_type = self._create_volume_types()
# create an instance from volume
LOG.info("Booting instance from volume")
volume_id = self.create_volume(imageRef=CONF.compute.image_ref,
volume_type=source_type['name'])['id']
instance = self._boot_instance_from_volume(volume_id, keypair,
security_group)
# write content to volume on instance
LOG.info("Setting timestamp in instance %s", instance['id'])
ip_instance = self.get_server_ip(instance)
timestamp = self.create_timestamp(ip_instance,
private_key=keypair['private_key'],
server=instance)
# retype volume with migration from backend #1 to backend #2
LOG.info("Retyping Volume %s to new type %s", volume_id,
dest_type['name'])
# This method calls for the retype of the volume before calling a
# waiter that asserts that the volume type has changed successfully.
self._volume_retype_with_migration(volume_id, dest_type['name'])
# check the content of written file
LOG.info("Getting timestamp in postmigrated instance %s",
instance['id'])
timestamp2 = self.get_timestamp(ip_instance,
private_key=keypair['private_key'],
server=instance)
self.assertEqual(timestamp, timestamp2)
# Assert that the volume is on the new host, is still in-use and has a
# migration_status of success
volume = self.admin_volumes_client.show_volume(volume_id)['volume']
# dest_type is host@backend, os-vol-host-attr:host is host@backend#type
self.assertIn(dest_type['host'], volume['os-vol-host-attr:host'])
self.assertEqual('in-use', volume['status'])
self.assertEqual('success', volume['migration_status'])
# Assert that the same volume id is attached to the instance, ensuring
# the os-migrate_volume_completion Cinder API has been called.
attached_volumes = self.servers_client.list_volume_attachments(
instance['id'])['volumeAttachments']
self.assertEqual(volume_id, attached_volumes[0]['id'])
@decorators.attr(type='slow')
@decorators.idempotent_id('fe47b1ed-640e-4e3b-a090-200e25607362')
@utils.services('compute', 'volume')
def test_volume_migrate_attached(self):
LOG.info("Creating keypair and security group")
keypair = self.create_keypair()
security_group = self.create_security_group()
LOG.info("Creating volume")
# Create a unique volume type to avoid using the backend default
migratable_type = self.create_volume_type()['name']
volume_id = self.create_volume(imageRef=CONF.compute.image_ref,
volume_type=migratable_type)['id']
volume = self.admin_volumes_client.show_volume(volume_id)
LOG.info("Booting instance from volume")
instance = self._boot_instance_from_volume(volume_id, keypair,
security_group)
# Identify the source and destination hosts for the migration
src_host = volume['volume']['os-vol-host-attr:host']
# Select the first c-vol host that isn't hosting the volume as the dest
# host['host_name'] should take the format of host@backend.
# src_host should take the format of host@backend#type
hosts = self.admin_volumes_client.list_hosts()['hosts']
for host in hosts:
if (host['service'] == 'cinder-volume' and
not src_host.startswith(host['host_name'])):
dest_host = host['host_name']
break
ip_instance = self.get_server_ip(instance)
timestamp = self.create_timestamp(ip_instance,
private_key=keypair['private_key'],
server=instance)
LOG.info("Migrating Volume %s from host %s to host %s",
volume_id, src_host, dest_host)
self.admin_volumes_client.migrate_volume(volume_id, host=dest_host)
# This waiter asserts that the migration_status is success and that
# the volume has moved to the dest_host
waiters.wait_for_volume_migration(self.admin_volumes_client, volume_id,
dest_host)
# check the content of written file
LOG.info("Getting timestamp in postmigrated instance %s",
instance['id'])
timestamp2 = self.get_timestamp(ip_instance,
private_key=keypair['private_key'],
server=instance)
self.assertEqual(timestamp, timestamp2)
# Assert that the volume is in-use
volume = self.admin_volumes_client.show_volume(volume_id)['volume']
self.assertEqual('in-use', volume['status'])
# Assert that the same volume id is attached to the instance, ensuring
# the os-migrate_volume_completion Cinder API has been called
attached_volumes = self.servers_client.list_volume_attachments(
instance['id'])['volumeAttachments']
attached_volume_id = attached_volumes[0]['id']
self.assertEqual(volume_id, attached_volume_id)