def deploy_cluster_with_reboot_plugin(self):
"""Add pre-deployment reboot task to nailgun via plugin.
Scenario:
1. Revert snapshot with 5 nodes
2. Download and install fuel-plugin-builder
3. Create plugin with reboot task
4. Build plugin and copy it in var directory
5. Install plugin to fuel
6. Create cluster and enable plugin
7. Provision nodes
8. Collect timestamps from nodes
9. Deploy cluster
10. Check if timestamps are changed
Duration 40m
"""
# define some plugin related variables
plugin_name = 'reboot_plugin'
source_plugin_path = os.path.join('/root/', plugin_name)
plugin_path = '/var'
tasks_path = os.path.dirname(os.path.abspath(__file__))
tasks_file = 'reboot_tasks.yaml'
self.show_step(1, initialize=True)
self.env.revert_snapshot("ready_with_5_slaves")
# let's get ssh client for the master node
# initiate fuel plugin builder instance
self.show_step(2)
fpb = FuelPluginBuilder()
# install fuel_plugin_builder on master node
fpb.fpb_install()
# create plugin template on the master node
self.show_step(3)
fpb.fpb_create_plugin(source_plugin_path)
fpb.fpb_update_release_in_metadata(source_plugin_path)
# replace plugin tasks with our file
fpb.fpb_replace_plugin_content(
os.path.join(tasks_path, tasks_file),
os.path.join(source_plugin_path, 'tasks.yaml'))
# build plugin
self.show_step(4)
packet_name = fpb.fpb_build_plugin(source_plugin_path)
fpb.fpb_copy_plugin(os.path.join(source_plugin_path, packet_name),
plugin_path)
self.show_step(5)
utils.install_plugin_check_code(ip=self.ssh_manager.admin_ip,
plugin=os.path.join(
plugin_path, packet_name))
self.show_step(6)
# create cluster
cluster_id = self.fuel_web.create_cluster(
name=self.__class__.__name__,
mode=DEPLOYMENT_MODE,
)
# get plugins from fuel and enable our one
msg = "Plugin couldn't be enabled. Check plugin version. Test aborted"
asserts.assert_true(
self.fuel_web.check_plugin_exists(cluster_id, plugin_name), msg)
options = {'metadata/enabled': True}
self.fuel_web.update_plugin_data(cluster_id, plugin_name, options)
logger.info('Cluster is {!s}'.format(cluster_id))
self.fuel_web.update_nodes(
cluster_id, {
'slave-01': ['controller', 'ceph-osd'],
'slave-02': ['compute', 'ceph-osd'],
'slave-03': ['compute'],
'slave-04': ['ceph-osd']
})
# firstly, let's provision nodes
self.show_step(7)
self.fuel_web.provisioning_cluster_wait(cluster_id)
# after provision is done, collect timestamps from nodes
old_timestamps = {}
nodes = {
'slave-01': True,
'slave-02': True,
'slave-03': False,
'slave-04': True
}
self.show_step(8)
for node in nodes:
logger.debug(
"Get init object creation time from node {0}".format(node))
cmd = 'stat --printf=\'%Y\' /proc/1'
with self.fuel_web.get_ssh_for_node(node) as node_ssh:
old_timestamps[node] = node_ssh.execute(cmd)['stdout'][0]
# start deploying nodes
# here nodes with controller and ceph roles should be rebooted
self.show_step(9)
self.fuel_web.deploy_cluster_wait_progress(cluster_id, 30)
# collect new timestamps and check them
self.show_step(10)
for node in nodes:
logger.debug(
"Get init object creation time from node {0}".format(node))
cmd = 'stat --printf=\'%Y\' /proc/1'
with self.fuel_web.get_ssh_for_node(node) as node_ssh:
new_timestamp = node_ssh.execute(cmd)['stdout'][0]
# compute node without ceph role shouldn't reboot
if not nodes[node]:
asserts.assert_equal(
new_timestamp, old_timestamps[node],
'The new timestamp {0} is not equal to old one {1}, '
'but it shouldn\'t for {2} node'.format(
new_timestamp, old_timestamps[node], node))
else:
# other nodes should be rebooted and have new timestamps
# greater than old
asserts.assert_true(
new_timestamp > old_timestamps[node],
'The new timestamp {0} is not greater than old one {1} '
'but it should for node {2}'.format(
new_timestamp, old_timestamps[node], node))
def deploy_cluster_with_reboot_plugin(self):
"""Add pre-deployment reboot task to nailgun via plugin.
Scenario:
1. Revert snapshot with 5 nodes
2. Download and install fuel-plugin-builder
3. Create plugin with reboot task
4. Build plugin and copy it in var directory
5. Install plugin to fuel
6. Create cluster and enable plugin
7. Provision nodes
8. Collect timestamps from nodes
9. Deploy cluster
10. Check if timestamps are changed
Duration 40m
"""
# define some plugin related variables
plugin_name = 'reboot_plugin'
source_plugin_path = os.path.join('/root/', plugin_name)
plugin_path = '/var'
tasks_path = os.path.dirname(os.path.abspath(__file__))
tasks_file = 'reboot_tasks.yaml'
self.show_step(1, initialize=True)
self.env.revert_snapshot("ready_with_5_slaves")
# let's get ssh client for the master node
# initiate fuel plugin builder instance
self.show_step(2)
fpb = FuelPluginBuilder()
# install fuel_plugin_builder on master node
fpb.fpb_install()
# create plugin template on the master node
self.show_step(3)
fpb.fpb_create_plugin(source_plugin_path)
fpb.fpb_update_release_in_metadata(source_plugin_path)
# replace plugin tasks with our file
fpb.fpb_replace_plugin_content(
os.path.join(tasks_path, tasks_file),
os.path.join(source_plugin_path, 'tasks.yaml'))
# build plugin
self.show_step(4)
packet_name = fpb.fpb_build_plugin(source_plugin_path)
fpb.fpb_copy_plugin(
os.path.join(source_plugin_path, packet_name), plugin_path)
self.show_step(5)
utils.install_plugin_check_code(
ip=self.ssh_manager.admin_ip,
plugin=os.path.join(plugin_path, packet_name))
self.show_step(6)
# create cluster
cluster_id = self.fuel_web.create_cluster(
name=self.__class__.__name__,
mode=DEPLOYMENT_MODE,
)
# get plugins from fuel and enable our one
msg = "Plugin couldn't be enabled. Check plugin version. Test aborted"
asserts.assert_true(
self.fuel_web.check_plugin_exists(cluster_id, plugin_name),
msg)
options = {'metadata/enabled': True}
self.fuel_web.update_plugin_data(cluster_id, plugin_name, options)
logger.info('Cluster is {!s}'.format(cluster_id))
self.fuel_web.update_nodes(
cluster_id,
{'slave-01': ['controller', 'ceph-osd'],
'slave-02': ['compute', 'ceph-osd'],
'slave-03': ['compute'],
'slave-04': ['ceph-osd']}
)
# firstly, let's provision nodes
self.show_step(7)
self.fuel_web.provisioning_cluster_wait(cluster_id)
# after provision is done, collect timestamps from nodes
old_timestamps = {}
nodes = {
'slave-01': True,
'slave-02': True,
'slave-03': False,
'slave-04': True
}
self.show_step(8)
for node in nodes:
logger.debug(
"Get init object creation time from node {0}".format(node))
cmd = 'stat --printf=\'%Y\' /proc/1'
with self.fuel_web.get_ssh_for_node(node) as node_ssh:
old_timestamps[node] = node_ssh.execute(cmd)['stdout'][0]
# start deploying nodes
# here nodes with controller and ceph roles should be rebooted
self.show_step(9)
self.fuel_web.deploy_cluster_wait_progress(cluster_id, 30)
# collect new timestamps and check them
self.show_step(10)
for node in nodes:
logger.debug(
"Get init object creation time from node {0}".format(node))
cmd = 'stat --printf=\'%Y\' /proc/1'
with self.fuel_web.get_ssh_for_node(node) as node_ssh:
new_timestamp = node_ssh.execute(cmd)['stdout'][0]
# compute node without ceph role shouldn't reboot
if not nodes[node]:
asserts.assert_equal(
new_timestamp, old_timestamps[node],
'The new timestamp {0} is not equal to old one {1}, '
'but it shouldn\'t for {2} node'
.format(new_timestamp, old_timestamps[node], node)
)
else:
# other nodes should be rebooted and have new timestamps
# greater than old
asserts.assert_true(
new_timestamp > old_timestamps[node],
'The new timestamp {0} is not greater than old one {1} '
'but it should for node {2}'
.format(new_timestamp, old_timestamps[node], node)
)
def deploy_cluster_with_reboot_plugin_timeout(self):
"""Check deployment is failed by reboot task plugin.
Scenario:
1. Revert snapshot with 3 nodes
2. Download and install fuel-plugin-builder
3. Create plugin with reboot task,
set timeout for reboot task as 1 second
4. Build plugin
5. Install plugin to fuel
6. Create cluster and enable plugin
7. Provision nodes
8. Deploy cluster
9. Check deployment was failed by reboot task
10. Check error msg at the logs
Duration 15m
"""
# define some plugin related variables
plugin_name = 'timeout_plugin'
source_plugin_path = os.path.join('/root/', plugin_name)
plugin_path = '/var'
tasks_path = os.path.dirname(os.path.abspath(__file__))
tasks_file = 'reboot_tasks.yaml'
# start reverting snapshot
self.show_step(1, initialize=True)
self.env.revert_snapshot("ready_with_3_slaves")
# let's get ssh client for the master node
self.show_step(2)
# initiate fuel plugin builder instance
fpb = FuelPluginBuilder()
# install fuel_plugin_builder on master node
fpb.fpb_install()
# change timeout to a new value '1'
fpb.put_value_to_local_yaml(os.path.join(tasks_path, tasks_file),
os.path.join('/tmp/', tasks_file),
[1, 'parameters', 'timeout'], 1)
self.show_step(3)
# create plugin template on the master node
fpb.fpb_create_plugin(source_plugin_path)
fpb.fpb_update_release_in_metadata(source_plugin_path)
# replace plugin tasks with our file
fpb.fpb_replace_plugin_content(
os.path.join('/tmp/', tasks_file),
os.path.join(source_plugin_path, 'tasks.yaml'))
# build plugin
self.show_step(4)
packet_name = fpb.fpb_build_plugin(source_plugin_path)
# copy plugin archive file
# to the /var directory on the master node
fpb.fpb_copy_plugin(os.path.join(source_plugin_path, packet_name),
plugin_path)
# let's install plugin
self.show_step(5)
utils.install_plugin_check_code(ip=self.ssh_manager.admin_ip,
plugin=os.path.join(
plugin_path, packet_name))
# create cluster
self.show_step(6)
cluster_id = self.fuel_web.create_cluster(
name=self.__class__.__name__,
mode=DEPLOYMENT_MODE,
)
# get plugins from fuel and enable it
msg = "Plugin couldn't be enabled. Check plugin version. Test aborted"
asserts.assert_true(
self.fuel_web.check_plugin_exists(cluster_id, plugin_name), msg)
options = {'metadata/enabled': True}
self.fuel_web.update_plugin_data(cluster_id, plugin_name, options)
logger.info('Cluster is {!s}'.format(cluster_id))
self.fuel_web.update_nodes(cluster_id,
{'slave-01': ['controller', 'ceph-osd']})
self.show_step(7)
self.fuel_web.provisioning_cluster_wait(cluster_id)
logger.info('Start cluster #%s deployment', cluster_id)
self.show_step(8)
task = self.fuel_web.client.deploy_nodes(cluster_id)
self.show_step(9)
self.fuel_web.assert_task_failed(task)
msg = 'Time detection (1 sec) for node reboot has expired'
cmd = 'grep "{0}" /var/log/astute/astute.log'.format(msg)
self.show_step(10)
self.ssh_manager.execute_on_remote(
ip=self.ssh_manager.admin_ip,
cmd=cmd,
err_msg='Failed to find reboot plugin warning message in logs')
def deploy_cluster_with_reboot_plugin_timeout(self):
"""Check deployment is failed by reboot task plugin.
Scenario:
1. Revert snapshot with 3 nodes
2. Download and install fuel-plugin-builder
3. Create plugin with reboot task,
set timeout for reboot task as 1 second
4. Build plugin
5. Install plugin to fuel
6. Create cluster and enable plugin
7. Provision nodes
8. Deploy cluster
9. Check deployment was failed by reboot task
10. Check error msg at the logs
Duration 15m
"""
# define some plugin related variables
plugin_name = 'timeout_plugin'
source_plugin_path = os.path.join('/root/', plugin_name)
plugin_path = '/var'
tasks_path = os.path.dirname(os.path.abspath(__file__))
tasks_file = 'reboot_tasks.yaml'
# start reverting snapshot
self.show_step(1, initialize=True)
self.env.revert_snapshot("ready_with_3_slaves")
# let's get ssh client for the master node
self.show_step(2)
# initiate fuel plugin builder instance
fpb = FuelPluginBuilder()
# install fuel_plugin_builder on master node
fpb.fpb_install()
# change timeout to a new value '1'
fpb.put_value_to_local_yaml(os.path.join(tasks_path, tasks_file),
os.path.join('/tmp/', tasks_file),
[1, 'parameters', 'timeout'],
1)
self.show_step(3)
# create plugin template on the master node
fpb.fpb_create_plugin(source_plugin_path)
fpb.fpb_update_release_in_metadata(source_plugin_path)
# replace plugin tasks with our file
fpb.fpb_replace_plugin_content(
os.path.join('/tmp/', tasks_file),
os.path.join(source_plugin_path, 'tasks.yaml'))
# build plugin
self.show_step(4)
packet_name = fpb.fpb_build_plugin(source_plugin_path)
# copy plugin archive file
# to the /var directory on the master node
fpb.fpb_copy_plugin(
os.path.join(source_plugin_path, packet_name),
plugin_path)
# let's install plugin
self.show_step(5)
utils.install_plugin_check_code(
ip=self.ssh_manager.admin_ip,
plugin=os.path.join(plugin_path, packet_name))
# create cluster
self.show_step(6)
cluster_id = self.fuel_web.create_cluster(
name=self.__class__.__name__,
mode=DEPLOYMENT_MODE,
)
# get plugins from fuel and enable it
msg = "Plugin couldn't be enabled. Check plugin version. Test aborted"
asserts.assert_true(
self.fuel_web.check_plugin_exists(cluster_id, plugin_name),
msg)
options = {'metadata/enabled': True}
self.fuel_web.update_plugin_data(cluster_id, plugin_name, options)
logger.info('Cluster is {!s}'.format(cluster_id))
self.fuel_web.update_nodes(
cluster_id,
{'slave-01': ['controller', 'ceph-osd']}
)
self.show_step(7)
self.fuel_web.provisioning_cluster_wait(cluster_id)
logger.info('Start cluster #%s deployment', cluster_id)
self.show_step(8)
task = self.fuel_web.client.deploy_nodes(cluster_id)
self.show_step(9)
self.fuel_web.assert_task_failed(task)
msg = 'Time detection (1 sec) for node reboot has expired'
cmd = 'grep "{0}" /var/log/astute/astute.log'.format(msg)
self.show_step(10)
self.ssh_manager.execute_on_remote(
ip=self.ssh_manager.admin_ip, cmd=cmd,
err_msg='Failed to find reboot plugin warning message in logs')
