python-ganttclient/nova/tests/test_migrations.py

# vim: tabstop=4 shiftwidth=4 softtabstop=4

# Copyright 2010-2011 OpenStack, LLC
# All Rights Reserved.
#
#    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.

"""
Tests for database migrations. This test case reads the configuration
file test_migrations.conf for database connection settings
to use in the tests. For each connection found in the config file,
the test case runs a series of test cases to ensure that migrations work
properly both upgrading and downgrading, and that no data loss occurs
if possible.

There are also "opportunistic" tests for both mysql and postgresql in here,
which allows testing against all 3 databases (sqlite in memory, mysql, pg) in
a properly configured unit test environment.

For the opportunistic testing you need to set up a db named 'openstack_citest'
with user 'openstack_citest' and password 'openstack_citest' on localhost.
The test will then use that db and u/p combo to run the tests.

For postgres on Ubuntu this can be done with the following commands:

sudo -u postgres psql
postgres=# create user openstack_citest with createdb login password
      'openstack_citest';
postgres=# create database openstack_citest with owner openstack_citest;

"""

import collections
import commands
import ConfigParser
import os
import urlparse

from migrate.versioning import repository
import sqlalchemy

import nova.db.migration as migration
import nova.db.sqlalchemy.migrate_repo
from nova.db.sqlalchemy.migration import versioning_api as migration_api
from nova.openstack.common import log as logging
from nova import test


LOG = logging.getLogger(__name__)


def _get_connect_string(backend,
                        user="openstack_citest",
                        passwd="openstack_citest",
                        database="openstack_citest"):
    """
    Try to get a connection with a very specific set of values, if we get
    these then we'll run the tests, otherwise they are skipped
    """
    if backend == "postgres":
        backend = "postgresql+psycopg2"
    elif backend == "mysql":
        backend = "mysql+mysqldb"

    return ("%(backend)s://%(user)s:%(passwd)s@localhost/%(database)s"
            % locals())


def _is_backend_avail(backend,
                      user="openstack_citest",
                      passwd="openstack_citest",
                      database="openstack_citest"):
    try:
        if backend == "mysql":
            connect_uri = _get_connect_string("mysql",
                user=user, passwd=passwd, database=database)
        elif backend == "postgres":
            connect_uri = _get_connect_string("postgres",
                user=user, passwd=passwd, database=database)
        engine = sqlalchemy.create_engine(connect_uri)
        connection = engine.connect()
    except Exception:
        # intentionally catch all to handle exceptions even if we don't
        # have any backend code loaded.
        return False
    else:
        connection.close()
        engine.dispose()
        return True


def _have_mysql():
    present = os.environ.get('NOVA_TEST_MYSQL_PRESENT')
    if present is None:
        return _is_backend_avail('mysql')
    return present.lower() in ('', 'true')


def get_table(engine, name):
    """Returns an sqlalchemy table dynamically from db.

    Needed because the models don't work for us in migrations
    as models will be far out of sync with the current data."""
    metadata = sqlalchemy.schema.MetaData()
    metadata.bind = engine
    return sqlalchemy.Table(name, metadata, autoload=True)


class TestMigrations(test.TestCase):
    """Test sqlalchemy-migrate migrations."""

    DEFAULT_CONFIG_FILE = os.path.join(os.path.dirname(__file__),
                                       'test_migrations.conf')
    # Test machines can set the NOVA_TEST_MIGRATIONS_CONF variable
    # to override the location of the config file for migration testing
    CONFIG_FILE_PATH = os.environ.get('NOVA_TEST_MIGRATIONS_CONF',
                                      DEFAULT_CONFIG_FILE)
    MIGRATE_FILE = nova.db.sqlalchemy.migrate_repo.__file__
    REPOSITORY = repository.Repository(
                                os.path.abspath(os.path.dirname(MIGRATE_FILE)))

    def setUp(self):
        super(TestMigrations, self).setUp()

        self.snake_walk = False
        self.test_databases = {}

        # Load test databases from the config file. Only do this
        # once. No need to re-run this on each test...
        LOG.debug('config_path is %s' % TestMigrations.CONFIG_FILE_PATH)
        if os.path.exists(TestMigrations.CONFIG_FILE_PATH):
            cp = ConfigParser.RawConfigParser()
            try:
                cp.read(TestMigrations.CONFIG_FILE_PATH)
                defaults = cp.defaults()
                for key, value in defaults.items():
                    self.test_databases[key] = value
                self.snake_walk = cp.getboolean('walk_style', 'snake_walk')
            except ConfigParser.ParsingError, e:
                self.fail("Failed to read test_migrations.conf config "
                          "file. Got error: %s" % e)
        else:
            self.fail("Failed to find test_migrations.conf config "
                      "file.")

        self.engines = {}
        for key, value in self.test_databases.items():
            self.engines[key] = sqlalchemy.create_engine(value)

        # We start each test case with a completely blank slate.
        self._reset_databases()

    def tearDown(self):
        # We destroy the test data store between each test case,
        # and recreate it, which ensures that we have no side-effects
        # from the tests
        self._reset_databases()
        super(TestMigrations, self).tearDown()

    def _reset_databases(self):
        def execute_cmd(cmd=None):
            status, output = commands.getstatusoutput(cmd)
            LOG.debug(output)
            self.assertEqual(0, status)
        for key, engine in self.engines.items():
            conn_string = self.test_databases[key]
            conn_pieces = urlparse.urlparse(conn_string)
            engine.dispose()
            if conn_string.startswith('sqlite'):
                # We can just delete the SQLite database, which is
                # the easiest and cleanest solution
                db_path = conn_pieces.path.strip('/')
                if os.path.exists(db_path):
                    os.unlink(db_path)
                # No need to recreate the SQLite DB. SQLite will
                # create it for us if it's not there...
            elif conn_string.startswith('mysql'):
                # We can execute the MySQL client to destroy and re-create
                # the MYSQL database, which is easier and less error-prone
                # than using SQLAlchemy to do this via MetaData...trust me.
                database = conn_pieces.path.strip('/')
                loc_pieces = conn_pieces.netloc.split('@')
                host = loc_pieces[1]
                auth_pieces = loc_pieces[0].split(':')
                user = auth_pieces[0]
                password = ""
                if len(auth_pieces) > 1:
                    if auth_pieces[1].strip():
                        password = "-p\"%s\"" % auth_pieces[1]
                sql = ("drop database if exists %(database)s; "
                       "create database %(database)s;") % locals()
                cmd = ("mysql -u \"%(user)s\" %(password)s -h %(host)s "
                       "-e \"%(sql)s\"") % locals()
                execute_cmd(cmd)
            elif conn_string.startswith('postgresql'):
                database = conn_pieces.path.strip('/')
                loc_pieces = conn_pieces.netloc.split('@')
                host = loc_pieces[1]

                auth_pieces = loc_pieces[0].split(':')
                user = auth_pieces[0]
                password = ""
                if len(auth_pieces) > 1:
                    password = auth_pieces[1].strip()
                # note(krtaylor): File creation problems with tests in
                # venv using .pgpass authentication, changed to
                # PGPASSWORD environment variable which is no longer
                # planned to be deprecated
                os.environ['PGPASSWORD'] = password
                os.environ['PGUSER'] = user
                # note(boris-42): We must create and drop database, we can't
                # drop database which we have connected to, so for such
                # operations there is a special database template1.
                sqlcmd = ("psql -w -U %(user)s -h %(host)s -c"
                                     " '%(sql)s' -d template1")
                sql = ("drop database if exists %(database)s;") % locals()
                droptable = sqlcmd % locals()
                execute_cmd(droptable)
                sql = ("create database %(database)s;") % locals()
                createtable = sqlcmd % locals()
                execute_cmd(createtable)
                os.unsetenv('PGPASSWORD')
                os.unsetenv('PGUSER')

    def test_walk_versions(self):
        """
        Walks all version scripts for each tested database, ensuring
        that there are no errors in the version scripts for each engine
        """
        for key, engine in self.engines.items():
            self._walk_versions(engine, self.snake_walk)

    def test_mysql_connect_fail(self):
        """
        Test that we can trigger a mysql connection failure and we fail
        gracefully to ensure we don't break people without mysql
        """
        if _is_backend_avail('mysql', user="openstack_cifail"):
            self.fail("Shouldn't have connected")

    def test_mysql_opportunistically(self):
        # Test that table creation on mysql only builds InnoDB tables
        if not _is_backend_avail('mysql'):
            self.skipTest("mysql not available")
        # add this to the global lists to make reset work with it, it's removed
        # automatically in tearDown so no need to clean it up here.
        connect_string = _get_connect_string("mysql")
        engine = sqlalchemy.create_engine(connect_string)
        self.engines["mysqlcitest"] = engine
        self.test_databases["mysqlcitest"] = connect_string

        # build a fully populated mysql database with all the tables
        self._reset_databases()
        self._walk_versions(engine, False, False)

        connection = engine.connect()
        # sanity check
        total = connection.execute("SELECT count(*) "
                                   "from information_schema.TABLES "
                                   "where TABLE_SCHEMA='openstack_citest'")
        self.assertTrue(total.scalar() > 0, "No tables found. Wrong schema?")

        noninnodb = connection.execute("SELECT count(*) "
                                       "from information_schema.TABLES "
                                       "where TABLE_SCHEMA='openstack_citest' "
                                       "and ENGINE!='InnoDB' "
                                       "and TABLE_NAME!='migrate_version'")
        count = noninnodb.scalar()
        self.assertEqual(count, 0, "%d non InnoDB tables created" % count)
        connection.close()

    def test_postgresql_connect_fail(self):
        """
        Test that we can trigger a postgres connection failure and we fail
        gracefully to ensure we don't break people without postgres
        """
        if _is_backend_avail('postgresql', user="openstack_cifail"):
            self.fail("Shouldn't have connected")

    def test_postgresql_opportunistically(self):
        # Test postgresql database migration walk
        if not _is_backend_avail('postgres'):
            self.skipTest("postgresql not available")
        # add this to the global lists to make reset work with it, it's removed
        # automatically in tearDown so no need to clean it up here.
        connect_string = _get_connect_string("postgres")
        engine = sqlalchemy.create_engine(connect_string)
        self.engines["postgresqlcitest"] = engine
        self.test_databases["postgresqlcitest"] = connect_string

        # build a fully populated postgresql database with all the tables
        self._reset_databases()
        self._walk_versions(engine, False, False)

    def _walk_versions(self, engine=None, snake_walk=False, downgrade=True):
        # Determine latest version script from the repo, then
        # upgrade from 1 through to the latest, with no data
        # in the databases. This just checks that the schema itself
        # upgrades successfully.

        # Place the database under version control
        migration_api.version_control(engine, TestMigrations.REPOSITORY,
                                     migration.INIT_VERSION)
        self.assertEqual(migration.INIT_VERSION,
                migration_api.db_version(engine,
                                         TestMigrations.REPOSITORY))

        migration_api.upgrade(engine, TestMigrations.REPOSITORY,
                              migration.INIT_VERSION + 1)

        LOG.debug('latest version is %s' % TestMigrations.REPOSITORY.latest)

        for version in xrange(migration.INIT_VERSION + 2,
                               TestMigrations.REPOSITORY.latest + 1):
            # upgrade -> downgrade -> upgrade
            self._migrate_up(engine, version, with_data=True)
            if snake_walk:
                self._migrate_down(engine, version)
                self._migrate_up(engine, version)

        if downgrade:
            # Now walk it back down to 0 from the latest, testing
            # the downgrade paths.
            for version in reversed(
                xrange(migration.INIT_VERSION + 2,
                       TestMigrations.REPOSITORY.latest + 1)):
                # downgrade -> upgrade -> downgrade
                self._migrate_down(engine, version)
                if snake_walk:
                    self._migrate_up(engine, version)
                    self._migrate_down(engine, version)

    def _migrate_down(self, engine, version):
        migration_api.downgrade(engine,
                                TestMigrations.REPOSITORY,
                                version)
        self.assertEqual(version,
                         migration_api.db_version(engine,
                                                  TestMigrations.REPOSITORY))

    def _migrate_up(self, engine, version, with_data=False):
        """migrate up to a new version of the db.

        We allow for data insertion and post checks at every
        migration version with special _prerun_### and
        _check_### functions in the main test.
        """
        # NOTE(sdague): try block is here because it's impossible to debug
        # where a failed data migration happens otherwise
        try:
            if with_data:
                data = None
                prerun = getattr(self, "_prerun_%d" % version, None)
                if prerun:
                    data = prerun(engine)

                migration_api.upgrade(engine,
                                          TestMigrations.REPOSITORY,
                                          version)
                self.assertEqual(
                    version,
                    migration_api.db_version(engine,
                                             TestMigrations.REPOSITORY))

            if with_data:
                check = getattr(self, "_check_%d" % version, None)
                if check:
                    check(engine, data)
        except Exception:
            LOG.error("Failed to migrate to version %s on engine %s" %
                      (version, engine))
            raise

    # migration 146, availability zone transition
    def _prerun_146(self, engine):
        data = {
            'availability_zone': 'custom_az',
            'aggregate_name': 1,
            'name': 'name',
            }

        aggregates = get_table(engine, 'aggregates')
        result = aggregates.insert().values(data).execute()
        # NOTE(sdague) it's important you don't insert keys by value in
        # postgresql, because it's autoincrement counter won't get updated
        data['id'] = result.inserted_primary_key[0]
        return data

    def _check_146(self, engine, data):
        aggregate_md = get_table(engine, 'aggregate_metadata')
        md = aggregate_md.select(
            aggregate_md.c.aggregate_id == data['id']).execute().first()
        self.assertEqual(data['availability_zone'], md['value'])

    # migration 147, availability zone transition for services
    def _prerun_147(self, engine):
        az = 'test_zone'
        host1 = 'compute-host1'
        host2 = 'compute-host2'
        # start at id == 2 because we already inserted one
        data = [
            {'id': 1, 'host': host1,
             'binary': 'nova-compute', 'topic': 'compute',
             'report_count': 0, 'availability_zone': az},
            {'id': 2, 'host': 'sched-host',
             'binary': 'nova-scheduler', 'topic': 'scheduler',
             'report_count': 0, 'availability_zone': 'ignore_me'},
            {'id': 3, 'host': host2,
             'binary': 'nova-compute', 'topic': 'compute',
             'report_count': 0, 'availability_zone': az},
            ]

        services = get_table(engine, 'services')
        engine.execute(services.insert(), data)
        return data

    def _check_147(self, engine, data):
        aggregate_md = get_table(engine, 'aggregate_metadata')
        aggregate_hosts = get_table(engine, 'aggregate_hosts')
        # NOTE(sdague): hard coded to id == 2, because we added to
        # aggregate_metadata previously
        for item in data:
            md = aggregate_md.select(
                aggregate_md.c.aggregate_id == 2).execute().first()
            if item['binary'] == "nova-compute":
                self.assertEqual(item['availability_zone'], md['value'])

        host = aggregate_hosts.select(
            aggregate_hosts.c.aggregate_id == 2
            ).execute().first()
        self.assertEqual(host['host'], data[0]['host'])

        # NOTE(sdague): id 3 is just non-existent
        host = aggregate_hosts.select(
            aggregate_hosts.c.aggregate_id == 3
            ).execute().first()
        self.assertEqual(host, None)

    # migration 149, changes IPAddr storage format
    def _prerun_149(self, engine):
        provider_fw_rules = get_table(engine, 'provider_fw_rules')
        data = [
            {'protocol': 'tcp', 'from_port': 1234,
             'to_port': 1234, 'cidr': "127.0.0.1"},
            {'protocol': 'tcp', 'from_port': 1234,
             'to_port': 1234, 'cidr': "255.255.255.255"},
            {'protocol': 'tcp', 'from_port': 1234,
             'to_port': 1234, 'cidr': "2001:db8::1:2"},
            {'protocol': 'tcp', 'from_port': 1234,
             'to_port': 1234, 'cidr': "::1"}
            ]
        engine.execute(provider_fw_rules.insert(), data)
        return data

    def _check_149(self, engine, data):
        provider_fw_rules = get_table(engine, 'provider_fw_rules')
        result = provider_fw_rules.select().execute()

        iplist = map(lambda x: x['cidr'], data)

        for row in result:
            self.assertIn(row['cidr'], iplist)

    # migration 152 - convert deleted from boolean to int
    def _prerun_152(self, engine):
        host1 = 'compute-host1'
        host2 = 'compute-host2'
        # NOTE(sdague): start at #4 because services data already in table
        # from 147
        services_data = [
            {'id': 4, 'host': host1, 'binary': 'nova-compute',
             'report_count': 0, 'topic': 'compute', 'deleted': False},
            {'id': 5, 'host': host1, 'binary': 'nova-compute',
             'report_count': 0, 'topic': 'compute', 'deleted': True}
            ]
        volumes_data = [
            {'id': 'first', 'host': host1, 'deleted': False},
            {'id': 'second', 'host': host2, 'deleted': True}
            ]

        services = get_table(engine, 'services')
        engine.execute(services.insert(), services_data)

        volumes = get_table(engine, 'volumes')
        engine.execute(volumes.insert(), volumes_data)
        return dict(services=services_data, volumes=volumes_data)

    def _check_152(self, engine, data):
        services = get_table(engine, 'services')
        service = services.select(services.c.id == 4).execute().first()
        self.assertEqual(0, service.deleted)
        service = services.select(services.c.id == 5).execute().first()
        self.assertEqual(service.id, service.deleted)

        volumes = get_table(engine, 'volumes')
        volume = volumes.select(volumes.c.id == "first").execute().first()
        self.assertEqual("", volume.deleted)
        volume = volumes.select(volumes.c.id == "second").execute().first()
        self.assertEqual(volume.id, volume.deleted)

    # migration 153, copy flavor information into system_metadata
    def _prerun_153(self, engine):
        fake_types = [
            dict(id=10, name='type1', memory_mb=128, vcpus=1,
                 root_gb=10, ephemeral_gb=0, flavorid="1", swap=0,
                 rxtx_factor=1.0, vcpu_weight=1, disabled=False,
                 is_public=True),
            dict(id=11, name='type2', memory_mb=512, vcpus=1,
                 root_gb=10, ephemeral_gb=5, flavorid="2", swap=0,
                 rxtx_factor=1.5, vcpu_weight=2, disabled=False,
                 is_public=True),
            dict(id=12, name='type3', memory_mb=128, vcpus=1,
                 root_gb=10, ephemeral_gb=0, flavorid="3", swap=0,
                 rxtx_factor=1.0, vcpu_weight=1, disabled=False,
                 is_public=False),
            dict(id=13, name='type4', memory_mb=128, vcpus=1,
                 root_gb=10, ephemeral_gb=0, flavorid="4", swap=0,
                 rxtx_factor=1.0, vcpu_weight=1, disabled=True,
                 is_public=True),
            dict(id=14, name='type5', memory_mb=128, vcpus=1,
                 root_gb=10, ephemeral_gb=0, flavorid="5", swap=0,
                 rxtx_factor=1.0, vcpu_weight=1, disabled=True,
                 is_public=False),
            ]

        fake_instances = [
            dict(uuid='m153-uuid1', instance_type_id=10),
            dict(uuid='m153-uuid2', instance_type_id=11),
            dict(uuid='m153-uuid3', instance_type_id=12),
            dict(uuid='m153-uuid4', instance_type_id=13),
            # NOTE(danms): no use of type5
            ]

        instances = get_table(engine, 'instances')
        instance_types = get_table(engine, 'instance_types')
        engine.execute(instance_types.insert(), fake_types)
        engine.execute(instances.insert(), fake_instances)

        return fake_types, fake_instances

    def _check_153(self, engine, data):
        fake_types, fake_instances = data
        # NOTE(danms): Fetch all the tables and data from scratch after change
        instances = get_table(engine, 'instances')
        instance_types = get_table(engine, 'instance_types')
        sys_meta = get_table(engine, 'instance_system_metadata')

        # Collect all system metadata, indexed by instance_uuid
        metadata = collections.defaultdict(dict)
        for values in sys_meta.select().execute():
            metadata[values['instance_uuid']][values['key']] = values['value']

        # Taken from nova/compute/api.py
        instance_type_props = ['id', 'name', 'memory_mb', 'vcpus',
                               'root_gb', 'ephemeral_gb', 'flavorid',
                               'swap', 'rxtx_factor', 'vcpu_weight']

        for instance in fake_instances:
            inst_sys_meta = metadata[instance['uuid']]
            inst_type = fake_types[instance['instance_type_id'] - 10]
            for prop in instance_type_props:
                prop_name = 'instance_type_%s' % prop
                self.assertIn(prop_name, inst_sys_meta)
                self.assertEqual(str(inst_sys_meta[prop_name]),
                                 str(inst_type[prop]))