def negative_manual_cic_maintenance_mode(self): """Check negative scenario for manual maintenance mode Scenario: 1. Revert snapshot 2. Disable UMM 3. Switch in maintenance mode 4. Check the controller not switching in maintenance mode 5. Check the controller become available Duration 45m """ self.env.revert_snapshot('cic_maintenance_mode') cluster_id = self.fuel_web.get_last_created_cluster() # Select a non-primary controller regular_ctrl = self.fuel_web.get_nailgun_node_by_name("slave-02") dregular_ctrl = self.fuel_web.get_devops_node_by_nailgun_node( regular_ctrl) _ip = regular_ctrl['ip'] _id = regular_ctrl['id'] asserts.assert_true('True' in check_available_mode(_ip), "Maintenance mode is not available") self.ssh_manager.execute_on_remote( ip=_ip, cmd="umm disable") asserts.assert_false('True' in check_available_mode(_ip), "Maintenance mode should not be available") logger.info('Try to execute maintenance mode ' 'for node-{0}'.format(_id)) self.ssh_manager.execute_on_remote( ip=_ip, cmd="umm on", assert_ec_equal=[1]) # If we don't disable maintenance mode, # the node would have gone to reboot, so we just expect time.sleep(30) asserts.assert_true( self.fuel_web.get_nailgun_node_by_devops_node(dregular_ctrl) ['online'], 'Node-{0} should be online after command "umm on"'. format(_id)) try: self.fuel_web.run_ostf(cluster_id, test_sets=['ha', 'smoke', 'sanity']) except AssertionError: logger.debug("Test failed from first probe," " we sleep 300 second try one more time" " and if it fails again - test will fails ") time.sleep(300) self.fuel_web.run_ostf(cluster_id, test_sets=['ha', 'smoke', 'sanity'])
def negative_manual_cic_maintenance_mode(self): """Check negative scenario for manual maintenance mode Scenario: 1. Revert snapshot 2. Disable UMM 3. Switch in maintenance mode 4. Check the controller not switching in maintenance mode 5. Check the controller become available Duration 45m """ self.env.revert_snapshot('cic_maintenance_mode') cluster_id = self.fuel_web.get_last_created_cluster() # Select a non-primary controller regular_ctrl = self.fuel_web.get_nailgun_node_by_name("slave-02") dregular_ctrl = self.fuel_web.get_devops_node_by_nailgun_node( regular_ctrl) _ip = regular_ctrl['ip'] _id = regular_ctrl['id'] asserts.assert_true('True' in check_available_mode(_ip), "Maintenance mode is not available") self.ssh_manager.execute_on_remote(ip=_ip, cmd="umm disable") asserts.assert_false('True' in check_available_mode(_ip), "Maintenance mode should not be available") logger.info('Try to execute maintenance mode ' 'for node-{0}'.format(_id)) self.ssh_manager.execute_on_remote(ip=_ip, cmd="umm on", assert_ec_equal=[1]) # If we don't disable maintenance mode, # the node would have gone to reboot, so we just expect time.sleep(30) asserts.assert_true( self.fuel_web.get_nailgun_node_by_devops_node( dregular_ctrl)['online'], 'Node-{0} should be online after command "umm on"'.format(_id)) try: self.fuel_web.run_ostf(cluster_id, test_sets=['ha', 'smoke', 'sanity']) except AssertionError: logger.debug("Test failed from first probe," " we sleep 300 second try one more time" " and if it fails again - test will fails ") time.sleep(300) self.fuel_web.run_ostf(cluster_id, test_sets=['ha', 'smoke', 'sanity'])
def manual_cic_maintenance_mode(self): """Check manual maintenance mode for controller Scenario: 1. Revert snapshot 2. Switch in maintenance mode 3. Wait until controller is rebooting 4. Exit maintenance mode 5. Check the controller become available Duration 155m """ self.env.revert_snapshot('cic_maintenance_mode') cluster_id = self.fuel_web.get_last_created_cluster() # Select a non-primary controller regular_ctrl = self.fuel_web.get_nailgun_node_by_name("slave-02") dregular_ctrl = self.fuel_web.get_devops_node_by_nailgun_node( regular_ctrl) _ip = regular_ctrl['ip'] _id = regular_ctrl['id'] logger.info('Maintenance mode for node-{0}'.format(_id)) asserts.assert_true('True' in check_available_mode(_ip), "Maintenance mode is not available") self.ssh_manager.check_call(ip=_ip, command="umm on", expected=[-1]) self.fuel_web.wait_node_is_offline(dregular_ctrl) asserts.assert_true( checkers.check_ping(self.env.get_admin_node_ip(), _ip, deadline=600), "Host {0} is not reachable by ping during 600 sec".format(_ip)) asserts.assert_true('True' in check_auto_mode(_ip), "Maintenance mode is not switched on") self.ssh_manager.check_call(ip=_ip, command="umm off") self.fuel_web.wait_node_is_online(dregular_ctrl) # Wait until Cinder services UP on a controller self.fuel_web.wait_cinder_is_up([dregular_ctrl.name]) # Wait until RabbitMQ cluster is UP wait_pass(lambda: self.fuel_web.run_single_ostf_test( cluster_id, test_sets=['ha'], test_name=ostf_test_mapping.OSTF_TEST_MAPPING.get( 'RabbitMQ availability')), timeout=1500) logger.info('RabbitMQ cluster is available') wait_pass(lambda: self.fuel_web.run_single_ostf_test( cluster_id, test_sets=['sanity'], test_name=ostf_test_mapping.OSTF_TEST_MAPPING.get( 'Check that required services are running')), timeout=1500) logger.info("Required services are running") # TODO(astudenov): add timeout_msg try: self.fuel_web.run_ostf(cluster_id, test_sets=['smoke', 'sanity', 'ha']) except AssertionError: logger.debug("Test failed from first probe," " we sleep 600 second try one more time" " and if it fails again - test will fails ") time.sleep(600) self.fuel_web.run_ostf(cluster_id, test_sets=['smoke', 'sanity', 'ha'])
def negative_auto_cic_maintenance_mode(self): """Check negative scenario for auto maintenance mode Scenario: 1. Revert snapshot 2. Disable UMM 3. Change UMM.CONF 4. Unexpected reboot 5. Check the controller not switching in maintenance mode 6. Check the controller become available Duration 85m """ self.env.revert_snapshot('cic_maintenance_mode') cluster_id = self.fuel_web.get_last_created_cluster() # Select a non-primary controller regular_ctrl = self.fuel_web.get_nailgun_node_by_name("slave-02") dregular_ctrl = self.fuel_web.get_devops_node_by_nailgun_node( regular_ctrl) _ip = regular_ctrl['ip'] _id = regular_ctrl['id'] asserts.assert_true('True' in check_available_mode(_ip), "Maintenance mode is not available") logger.info('Disable UMM on node-{0}'.format(_id)) change_config(_ip, umm=False, reboot_count=0) asserts.assert_false('True' in check_available_mode(_ip), "Maintenance mode should not be available") logger.info('Unexpected reboot on node-{0}'.format(_id)) self.ssh_manager.check_call(ip=_ip, command='reboot >/dev/null & ') wait( lambda: not checkers.check_ping(self.env.get_admin_node_ip(), _ip), timeout=60 * 10, timeout_msg='Node {} still responds to ping'.format( dregular_ctrl.name)) # Node don't have enough time for set offline status # after reboot # Just waiting asserts.assert_true( checkers.check_ping(self.env.get_admin_node_ip(), _ip, deadline=600), "Host {0} is not reachable by ping during 600 sec".format(_ip)) self.fuel_web.wait_node_is_online(dregular_ctrl) logger.info('Check that node-{0} not in maintenance mode after' ' unexpected reboot'.format(_id)) wait(lambda: tcp_ping(_ip, 22), timeout=60 * 10, timeout_msg='Node {} still is not available by SSH'.format( dregular_ctrl.name)) asserts.assert_false('True' in check_auto_mode(_ip), "Maintenance mode should not switched") # Wait until MySQL Galera is UP on some controller self.fuel_web.wait_mysql_galera_is_up([dregular_ctrl.name]) # Wait until Cinder services UP on a controller self.fuel_web.wait_cinder_is_up([dregular_ctrl.name]) # Wait until RabbitMQ cluster is UP wait_pass(lambda: self.fuel_web.run_single_ostf_test( cluster_id, test_sets=['ha'], test_name=ostf_test_mapping.OSTF_TEST_MAPPING.get( 'RabbitMQ availability')), timeout=1500) logger.info('RabbitMQ cluster is available') # TODO(astudenov): add timeout_msg wait_pass(lambda: self.fuel_web.run_single_ostf_test( cluster_id, test_sets=['sanity'], test_name=ostf_test_mapping.OSTF_TEST_MAPPING.get( 'Check that required services are running')), timeout=1500) logger.info("Required services are running") try: self.fuel_web.run_ostf(cluster_id, test_sets=['smoke', 'sanity', 'ha']) except AssertionError: logger.debug("Test failed from first probe," " we sleep 600 second try one more time" " and if it fails again - test will fails ") time.sleep(600) self.fuel_web.run_ostf(cluster_id, test_sets=['smoke', 'sanity', 'ha'])
def auto_cic_maintenance_mode(self): """Check auto maintenance mode for controller Scenario: 1. Revert snapshot 2. Unexpected reboot 3. Wait until controller is switching in maintenance mode 4. Exit maintenance mode 5. Check the controller become available Duration 155m """ self.env.revert_snapshot('cic_maintenance_mode') cluster_id = self.fuel_web.get_last_created_cluster() # Select a non-primary controller regular_ctrl = self.fuel_web.get_nailgun_node_by_name("slave-02") dregular_ctrl = self.fuel_web.get_devops_node_by_nailgun_node( regular_ctrl) _ip = regular_ctrl['ip'] _id = regular_ctrl['id'] asserts.assert_true('True' in check_available_mode(_ip), "Maintenance mode is not available") change_config(_ip, reboot_count=0) logger.info('Change UMM.CONF on node-{0}'.format(_id)) logger.info('Unexpected reboot on node-{0}'.format(_id)) command = 'reboot --force >/dev/null & ' self.ssh_manager.execute_on_remote(ip=_ip, cmd=command) wait( lambda: not checkers.check_ping(self.env.get_admin_node_ip(), _ip), timeout=60 * 10, timeout_msg='Node {} still responds to ping'.format( dregular_ctrl.name)) self.fuel_web.wait_node_is_offline(dregular_ctrl) logger.info('Check that node-{0} in maintenance mode after' ' unexpected reboot'.format(_id)) asserts.assert_true( checkers.check_ping(self.env.get_admin_node_ip(), _ip, deadline=600), "Host {0} is not reachable by ping during 600 sec".format(_ip)) asserts.assert_true('True' in check_auto_mode(_ip), "Maintenance mode is not switched on") logger.info('turn off Maintenance mode') self.ssh_manager.execute_on_remote(ip=_ip, cmd="umm off") time.sleep(30) change_config(_ip) self.fuel_web.wait_node_is_online(dregular_ctrl) # Wait until MySQL Galera is UP on some controller self.fuel_web.wait_mysql_galera_is_up([dregular_ctrl.name]) # Wait until Cinder services UP on a controller self.fuel_web.wait_cinder_is_up([dregular_ctrl.name]) # Wait until RabbitMQ cluster is UP wait_pass(lambda: self.fuel_web.run_single_ostf_test( cluster_id, test_sets=['ha'], test_name=ostf_test_mapping.OSTF_TEST_MAPPING.get( 'RabbitMQ availability')), timeout=1500) logger.info('RabbitMQ cluster is available') # Wait until all Openstack services are UP wait_pass(lambda: self.fuel_web.run_single_ostf_test( cluster_id, test_sets=['sanity'], test_name=ostf_test_mapping.OSTF_TEST_MAPPING.get( 'Check that required services are running')), timeout=1500) logger.info("Required services are running") try: self.fuel_web.run_ostf(cluster_id, test_sets=['smoke', 'sanity', 'ha']) except AssertionError: logger.debug("Test failed from first probe," " we sleep 600 second try one more time" " and if it fails again - test will fails ") time.sleep(600) self.fuel_web.run_ostf(cluster_id, test_sets=['smoke', 'sanity', 'ha'])
def manual_cic_maintenance_mode(self): """Check manual maintenance mode for controller Scenario: 1. Revert snapshot 2. Switch in maintenance mode 3. Wait until controller is rebooting 4. Exit maintenance mode 5. Check the controller become available Duration 155m """ self.env.revert_snapshot('cic_maintenance_mode') cluster_id = self.fuel_web.get_last_created_cluster() # Select a non-primary controller regular_ctrl = self.fuel_web.get_nailgun_node_by_name("slave-02") dregular_ctrl = self.fuel_web.get_devops_node_by_nailgun_node( regular_ctrl) _ip = regular_ctrl['ip'] _id = regular_ctrl['id'] logger.info('Maintenance mode for node-{0}'.format(_id)) asserts.assert_true('True' in check_available_mode(_ip), "Maintenance mode is not available") self.ssh_manager.execute_on_remote( ip=_ip, cmd="umm on") self.fuel_web.wait_node_is_offline(dregular_ctrl) asserts.assert_true( checkers.check_ping(self.env.get_admin_node_ip(), _ip, deadline=600), "Host {0} is not reachable by ping during 600 sec" .format(_ip)) asserts.assert_true('True' in check_auto_mode(_ip), "Maintenance mode is not switched on") self.ssh_manager.execute_on_remote( ip=_ip, cmd="umm off") self.fuel_web.wait_node_is_online(dregular_ctrl) # Wait until Cinder services UP on a controller self.fuel_web.wait_cinder_is_up( [dregular_ctrl.name]) # Wait until RabbitMQ cluster is UP wait_pass(lambda: self.fuel_web.run_single_ostf_test( cluster_id, test_sets=['ha'], test_name=ostf_test_mapping.OSTF_TEST_MAPPING.get( 'RabbitMQ availability')), timeout=1500) logger.info('RabbitMQ cluster is available') wait_pass(lambda: self.fuel_web.run_single_ostf_test( cluster_id, test_sets=['sanity'], test_name=ostf_test_mapping.OSTF_TEST_MAPPING.get( 'Check that required services are running')), timeout=1500) logger.info("Required services are running") # TODO(astudenov): add timeout_msg try: self.fuel_web.run_ostf(cluster_id, test_sets=['smoke', 'sanity', 'ha']) except AssertionError: logger.debug("Test failed from first probe," " we sleep 600 second try one more time" " and if it fails again - test will fails ") time.sleep(600) self.fuel_web.run_ostf(cluster_id, test_sets=['smoke', 'sanity', 'ha'])
def negative_auto_cic_maintenance_mode(self): """Check negative scenario for auto maintenance mode Scenario: 1. Revert snapshot 2. Disable UMM 3. Change UMM.CONF 4. Unexpected reboot 5. Check the controller not switching in maintenance mode 6. Check the controller become available Duration 85m """ self.env.revert_snapshot('cic_maintenance_mode') cluster_id = self.fuel_web.get_last_created_cluster() # Select a non-primary controller regular_ctrl = self.fuel_web.get_nailgun_node_by_name("slave-02") dregular_ctrl = self.fuel_web.get_devops_node_by_nailgun_node( regular_ctrl) _ip = regular_ctrl['ip'] _id = regular_ctrl['id'] asserts.assert_true('True' in check_available_mode(_ip), "Maintenance mode is not available") logger.info('Disable UMM on node-{0}'.format(_id)) change_config(_ip, umm=False, reboot_count=0) asserts.assert_false('True' in check_available_mode(_ip), "Maintenance mode should not be available") command = 'reboot --force >/dev/null & ' logger.info('Unexpected reboot on node-{0}' .format(_id)) self.ssh_manager.execute_on_remote( ip=_ip, cmd=command) wait(lambda: not checkers.check_ping(self.env.get_admin_node_ip(), _ip), timeout=60 * 10, timeout_msg='Node {} still responds to ping'.format( dregular_ctrl.name)) # Node don't have enough time for set offline status # after reboot --force # Just waiting asserts.assert_true( checkers.check_ping(self.env.get_admin_node_ip(), _ip, deadline=600), "Host {0} is not reachable by ping during 600 sec" .format(_ip)) self.fuel_web.wait_node_is_online(dregular_ctrl) logger.info('Check that node-{0} not in maintenance mode after' ' unexpected reboot'.format(_id)) wait(lambda: tcp_ping(_ip, 22), timeout=60 * 10, timeout_msg='Node {} still is not available by SSH'.format( dregular_ctrl.name)) asserts.assert_false('True' in check_auto_mode(_ip), "Maintenance mode should not switched") # Wait until MySQL Galera is UP on some controller self.fuel_web.wait_mysql_galera_is_up( [dregular_ctrl.name]) # Wait until Cinder services UP on a controller self.fuel_web.wait_cinder_is_up( [dregular_ctrl.name]) # Wait until RabbitMQ cluster is UP wait_pass(lambda: self.fuel_web.run_single_ostf_test( cluster_id, test_sets=['ha'], test_name=ostf_test_mapping.OSTF_TEST_MAPPING.get( 'RabbitMQ availability')), timeout=1500) logger.info('RabbitMQ cluster is available') # TODO(astudenov): add timeout_msg wait_pass(lambda: self.fuel_web.run_single_ostf_test( cluster_id, test_sets=['sanity'], test_name=ostf_test_mapping.OSTF_TEST_MAPPING.get( 'Check that required services are running')), timeout=1500) logger.info("Required services are running") try: self.fuel_web.run_ostf(cluster_id, test_sets=['smoke', 'sanity', 'ha']) except AssertionError: logger.debug("Test failed from first probe," " we sleep 600 second try one more time" " and if it fails again - test will fails ") time.sleep(600) self.fuel_web.run_ostf(cluster_id, test_sets=['smoke', 'sanity', 'ha'])