Add an active/standby scenario test

This patch adds a tempest scenario test for the active/standby topology.
It starts an active/standby load balancer, deletes the master amphora,
and validates that the load balancer is still working.

Change-Id: Ibcd5552a67cea650edc72bfaa986357267ca2407
diff --git a/octavia_tempest_plugin/config.py b/octavia_tempest_plugin/config.py
index 7c4ed2f..63c292b 100644
--- a/octavia_tempest_plugin/config.py
+++ b/octavia_tempest_plugin/config.py
@@ -66,7 +66,7 @@
                default=5,
                help='Interval to check for status changes.'),
     cfg.IntOpt('check_timeout',
-               default=60,
+               default=120,
                help='Timeout, in seconds, to wait for a status change.'),
     cfg.BoolOpt('test_with_noop',
                 default=False,
diff --git a/octavia_tempest_plugin/tests/act_stdby_scenario/v2/test_active_standby.py b/octavia_tempest_plugin/tests/act_stdby_scenario/v2/test_active_standby.py
new file mode 100644
index 0000000..612bbe2
--- /dev/null
+++ b/octavia_tempest_plugin/tests/act_stdby_scenario/v2/test_active_standby.py
@@ -0,0 +1,287 @@
+# Copyright 2018 GoDaddy
+# Copyright 2018 Rackspace US Inc.  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.
+
+import requests
+import testtools
+import time
+
+from oslo_log import log as logging
+from tempest import config
+from tempest.lib.common.utils import data_utils
+from tempest.lib import decorators
+from tempest.lib import exceptions
+
+from octavia_tempest_plugin.common import constants as const
+from octavia_tempest_plugin.tests import test_base
+from octavia_tempest_plugin.tests import waiters
+
+CONF = config.CONF
+LOG = logging.getLogger(__name__)
+
+
+@testtools.skipUnless(
+    CONF.validation.run_validation,
+    'Active-Standby tests will not work without run_validation enabled.')
+class ActiveStandbyScenarioTest(test_base.LoadBalancerBaseTestWithCompute):
+
+    @classmethod
+    def resource_setup(cls):
+        """Setup resources needed by the tests."""
+        super(ActiveStandbyScenarioTest, cls).resource_setup()
+
+        # We have to do this here as the api_version and clients are not
+        # setup in time to use a decorator or the skip_checks mixin
+        if not cls.lb_admin_flavor_profile_client.is_version_supported(
+                cls.api_version, '2.3'):
+            return
+
+        lb_name = data_utils.rand_name("lb_member_lb1_actstdby")
+        lb_kwargs = {const.PROVIDER: CONF.load_balancer.provider,
+                     const.NAME: lb_name}
+
+        # TODO(rm_work): Make this work with ipv6 and split this test for both
+        ip_version = 4
+        cls._setup_lb_network_kwargs(lb_kwargs, ip_version)
+
+        lb = cls.mem_lb_client.create_loadbalancer(**lb_kwargs)
+        cls.lb_id = lb[const.ID]
+        cls.addClassResourceCleanup(
+            cls.mem_lb_client.cleanup_loadbalancer,
+            cls.lb_id)
+
+        if CONF.validation.connect_method == 'floating':
+            port_id = lb[const.VIP_PORT_ID]
+            result = cls.lb_mem_float_ip_client.create_floatingip(
+                floating_network_id=CONF.network.public_network_id,
+                port_id=port_id)
+            floating_ip = result['floatingip']
+            LOG.info('lb1_floating_ip: {}'.format(floating_ip))
+            cls.addClassResourceCleanup(
+                waiters.wait_for_not_found,
+                cls.lb_mem_float_ip_client.delete_floatingip,
+                cls.lb_mem_float_ip_client.show_floatingip,
+                floatingip_id=floating_ip['id'])
+            cls.lb_vip_address = floating_ip['floating_ip_address']
+        else:
+            cls.lb_vip_address = lb[const.VIP_ADDRESS]
+
+        waiters.wait_for_status(cls.mem_lb_client.show_loadbalancer,
+                                cls.lb_id, const.PROVISIONING_STATUS,
+                                const.ACTIVE,
+                                CONF.load_balancer.lb_build_interval,
+                                CONF.load_balancer.lb_build_timeout)
+
+        listener_name = data_utils.rand_name("lb_member_listener1_actstdby")
+        listener_kwargs = {
+            const.NAME: listener_name,
+            const.PROTOCOL: const.HTTP,
+            const.PROTOCOL_PORT: '80',
+            const.LOADBALANCER_ID: cls.lb_id,
+        }
+        listener = cls.mem_listener_client.create_listener(**listener_kwargs)
+        cls.listener_id = listener[const.ID]
+        cls.addClassResourceCleanup(
+            cls.mem_listener_client.cleanup_listener,
+            cls.listener_id,
+            lb_client=cls.mem_lb_client, lb_id=cls.lb_id)
+
+        waiters.wait_for_status(cls.mem_lb_client.show_loadbalancer,
+                                cls.lb_id, const.PROVISIONING_STATUS,
+                                const.ACTIVE,
+                                CONF.load_balancer.build_interval,
+                                CONF.load_balancer.build_timeout)
+
+        pool_name = data_utils.rand_name("lb_member_pool1_actstdby")
+        pool_kwargs = {
+            const.NAME: pool_name,
+            const.PROTOCOL: const.HTTP,
+            const.LB_ALGORITHM: const.LB_ALGORITHM_ROUND_ROBIN,
+            const.LISTENER_ID: cls.listener_id,
+        }
+        pool = cls.mem_pool_client.create_pool(**pool_kwargs)
+        cls.pool_id = pool[const.ID]
+        cls.addClassResourceCleanup(
+            cls.mem_pool_client.cleanup_pool,
+            cls.pool_id,
+            lb_client=cls.mem_lb_client, lb_id=cls.lb_id)
+
+        waiters.wait_for_status(cls.mem_lb_client.show_loadbalancer,
+                                cls.lb_id, const.PROVISIONING_STATUS,
+                                const.ACTIVE,
+                                CONF.load_balancer.build_interval,
+                                CONF.load_balancer.build_timeout)
+
+        # Set up Member 1 for Webserver 1
+        member1_name = data_utils.rand_name("lb_member_member1_actstdby")
+        member1_kwargs = {
+            const.POOL_ID: cls.pool_id,
+            const.NAME: member1_name,
+            const.ADMIN_STATE_UP: True,
+            const.ADDRESS: cls.webserver1_ip,
+            const.PROTOCOL_PORT: 80,
+        }
+        if cls.lb_member_1_subnet:
+            member1_kwargs[const.SUBNET_ID] = cls.lb_member_1_subnet[const.ID]
+
+        member1 = cls.mem_member_client.create_member(
+            **member1_kwargs)
+        cls.addClassResourceCleanup(
+            cls.mem_member_client.cleanup_member,
+            member1[const.ID], pool_id=cls.pool_id,
+            lb_client=cls.mem_lb_client, lb_id=cls.lb_id)
+        waiters.wait_for_status(
+            cls.mem_lb_client.show_loadbalancer, cls.lb_id,
+            const.PROVISIONING_STATUS, const.ACTIVE,
+            CONF.load_balancer.check_interval,
+            CONF.load_balancer.check_timeout)
+
+        # Set up Member 2 for Webserver 2
+        member2_name = data_utils.rand_name("lb_member_member2_actstdby")
+        member2_kwargs = {
+            const.POOL_ID: cls.pool_id,
+            const.NAME: member2_name,
+            const.ADMIN_STATE_UP: True,
+            const.ADDRESS: cls.webserver2_ip,
+            const.PROTOCOL_PORT: 80,
+        }
+        if cls.lb_member_2_subnet:
+            member2_kwargs[const.SUBNET_ID] = cls.lb_member_2_subnet[const.ID]
+
+        member2 = cls.mem_member_client.create_member(
+            **member2_kwargs)
+        cls.addClassResourceCleanup(
+            cls.mem_member_client.cleanup_member,
+            member2[const.ID], pool_id=cls.pool_id,
+            lb_client=cls.mem_lb_client, lb_id=cls.lb_id)
+        waiters.wait_for_status(
+            cls.mem_lb_client.show_loadbalancer, cls.lb_id,
+            const.PROVISIONING_STATUS, const.ACTIVE,
+            CONF.load_balancer.check_interval,
+            CONF.load_balancer.check_timeout)
+
+    @testtools.skipIf(CONF.load_balancer.test_with_noop,
+                      'Active/Standby tests will not work in noop mode.')
+    @decorators.idempotent_id('e591fa7a-0eee-485a-8ca0-5cf1a556bdf0')
+    def test_active_standby_vrrp_failover(self):
+        """Tests active/standby VRRP failover
+
+        * Test the load balancer to make sure it is functioning
+        * Identifies the Master and Backup amphora
+        * Deletes the Master amphora
+        * Sends traffic through the load balancer
+        * Validates that the Backup has assumed the Master role
+        """
+        # We have to do this here as the api_version and clients are not
+        # setup in time to use a decorator or the skip_checks mixin
+        if not self.mem_listener_client.is_version_supported(
+                self.api_version, '2.4'):
+            raise self.skipException(
+                'Active/Standby VRRP failover tests require '
+                'Octavia API version 2.3 or newer.')
+
+        session = requests.Session()
+
+        # Send some traffic
+        self.check_members_balanced(self.lb_vip_address)
+
+        # Get the amphorae associated with this load balancer
+        amphorae = self.os_admin.amphora_client.list_amphorae(
+            query_params='{loadbalancer_id}={lb_id}'.format(
+                loadbalancer_id=const.LOADBALANCER_ID,
+                lb_id=self.lb_id))
+
+        # TODO(johnsom): Fix when LB flavors support act/stdby
+        if len(amphorae) < 2:
+            self.skipTest('Load balancer must be using active/standby '
+                          'topology for the VRRP failover test.')
+
+        # Generate traffic on the LB so we can identify the current Master
+        r = session.get('http://{0}'.format(self.lb_vip_address), timeout=2)
+
+        # Cycle through the amps to find the master
+        master_amp = None
+        backup_amp = None
+        start = int(time.time())
+        while True:
+            for amp in amphorae:
+                amphora_stats = self.os_admin.amphora_client.get_amphora_stats(
+                    amp[const.ID])
+                for listener in amphora_stats:
+                    if listener[const.TOTAL_CONNECTIONS] > 0:
+                        master_amp = amp
+                        break
+                # check if we left the listener for loop by finding the master
+                if master_amp:
+                    break
+            # If we found the master and broke out of the amp for loop, break
+            # out of the while loop too.
+            if master_amp:
+                break
+            if int(time.time()) - start >= CONF.load_balancer.check_timeout:
+                message = ('Unable to find Master amphora in {timeout} '
+                           'seconds.'.format(
+                               timeout=CONF.load_balancer.check_timeout))
+                raise exceptions.TimeoutException(message)
+            time.sleep(CONF.load_balancer.check_interval)
+
+        # Find the backup amphora and check it is ready for the test
+        for amp in amphorae:
+            if amp[const.ID] == master_amp[const.ID]:
+                continue
+            else:
+                backup_amp = amp
+        self.assertIsNotNone(backup_amp)
+        amphora_stats = self.os_admin.amphora_client.get_amphora_stats(
+            backup_amp[const.ID])
+        for listener in amphora_stats:
+            self.assertEqual(0, listener[const.TOTAL_CONNECTIONS])
+
+        # Delete the master amphora compute instance
+        self.os_admin_servers_client.delete_server(
+            master_amp[const.COMPUTE_ID])
+
+        # Pass some traffic through the LB
+        # Note: We want this to loop for longer than the heartbeat interval
+        #       to make sure a stats update has come in to the HM
+        for x in range(0, 20):
+            try:
+                r = session.get('http://{0}'.format(self.lb_vip_address),
+                                timeout=1)
+                LOG.info('Got response: %s', r.text)
+            except Exception:
+                LOG.info('Load balancer request failed. Looping')
+            time.sleep(1)
+
+        # Check that the Backup amphora is now Master
+        amphora_stats = self.os_admin.amphora_client.get_amphora_stats(
+            backup_amp[const.ID])
+        connections = 0
+        for listener in amphora_stats:
+            connections += listener[const.TOTAL_CONNECTIONS]
+        self.assertTrue(connections > 0)
+        LOG.info('Backup amphora is now Master.')
+        # Wait for the amphora failover to start
+        waiters.wait_for_status(
+            self.mem_lb_client.show_loadbalancer,
+            self.lb_id, const.PROVISIONING_STATUS,
+            const.PENDING_UPDATE, CONF.load_balancer.check_interval,
+            CONF.load_balancer.check_timeout)
+        # Wait for the load balancer to return to ACTIVE so the
+        # cleanup steps will pass
+        waiters.wait_for_status(
+            self.mem_lb_client.show_loadbalancer,
+            self.lb_id, const.PROVISIONING_STATUS,
+            const.ACTIVE, CONF.load_balancer.lb_build_interval,
+            CONF.load_balancer.lb_build_timeout)
diff --git a/zuul.d/jobs.yaml b/zuul.d/jobs.yaml
index d342aff..aac2460 100644
--- a/zuul.d/jobs.yaml
+++ b/zuul.d/jobs.yaml
@@ -556,7 +556,6 @@
               loadbalancer_topology: ACTIVE_STANDBY
             task_flow:
               engine: parallel
-              devstack_local_conf:
         test-config:
           "$TEMPEST_CONFIG":
             load_balancer:
@@ -608,3 +607,32 @@
     name: octavia-v2-act-stdby-iptables-dsvm-py2-scenario-stable-queens
     parent: octavia-v2-act-stdby-iptables-dsvm-py2-scenario
     override-checkout: stable/queens
+
+- job:
+    name: octavia-v2-act-stdby-dsvm-scenario
+    parent: octavia-dsvm-live-base
+    vars:
+      devstack_local_conf:
+        post-config:
+          $OCTAVIA_CONF:
+            api_settings:
+              api_v1_enabled: False
+            controller_worker:
+              loadbalancer_topology: ACTIVE_STANDBY
+            task_flow:
+              engine: parallel
+      tempest_concurrency: 2
+      tempest_test_regex: ^octavia_tempest_plugin.tests.act_stdby_scenario.v2.test_active_standby\.
+      tox_envlist: all
+
+- job:
+    name: octavia-v2-act-stdby-dsvm-py2-scenario
+    parent: octavia-v2-act-stdby-dsvm-scenario
+    vars:
+      devstack_localrc:
+        USE_PYTHON3: False
+
+- job:
+    name: octavia-v2-act-stdby-dsvm-scenario-stable-stein
+    parent: octavia-v2-act-stdby-dsvm-scenario
+    override-checkout: stable/stein
diff --git a/zuul.d/projects.yaml b/zuul.d/projects.yaml
index 490d68b..4b36f01 100644
--- a/zuul.d/projects.yaml
+++ b/zuul.d/projects.yaml
@@ -38,6 +38,12 @@
             voting: false
         - octavia-v2-act-stdby-iptables-dsvm-py2-scenario-stable-queens:
             voting: false
+        - octavia-v2-act-stdby-dsvm-scenario:
+            voting: false
+        - octavia-v2-act-stdby-dsvm-py2-scenario:
+            voting: false
+        - octavia-v2-act-stdby-dsvm-scenario-stable-stein:
+            voting: false
         - octavia-v2-dsvm-tls-barbican:
             voting: false
         - octavia-v2-dsvm-tls-barbican-stable-stein: