コード例 #1
0
ファイル: fabfile.py プロジェクト: japaniel/CloudFerry
def create_filters(name_config=None, filter_folder=DEFAULT_FILTERS_FILES,
                   images_date='2000-01-01'):
    """Generates filter files for CloudFerry based on the schedule prepared by
    condensation/grouping."""
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    make_filters.make(filter_folder, images_date)
コード例 #2
0
ファイル: fabfile.py プロジェクト: cburns-mirantis/CloudFerry
def evacuate(name_config=None, debug=False, iteration=False):
    if debug:
        utils.configure_logging("DEBUG")

    try:
        iteration = int(iteration)
    except ValueError:
        LOG.error("Invalid value provided as 'iteration' argument, it must be "
                  "integer")
        return
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    utils.init_singletones(cfglib.CONF)
    env.key_filename = cfglib.CONF.migrate.key_filename
    cloud = cloud_ferry.CloudFerry(cfglib.CONF)
    LOG.info("running evacuation")
    evacuation_chain.process_chain(cloud, iteration)

    freed_nodes = get_freed_nodes(iteration)

    if not freed_nodes:
        LOG.warning("Evacuation cannot be completed, because there are no "
                    "available compute nodes, that should be freed")
        return

    LOG.info(
        "Following nodes will be freed once in-cloud migration finishes, "
        "and can be moved from source to destination: %s", freed_nodes)
コード例 #3
0
ファイル: fabfile.py プロジェクト: japaniel/CloudFerry
def evacuate(name_config=None, debug=False, iteration=False):
    if debug:
        utils.configure_logging("DEBUG")

    try:
        iteration = int(iteration)
    except ValueError:
        LOG.error("Invalid value provided as 'iteration' argument, it must be "
                  "integer")
        return
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    utils.init_singletones(cfglib.CONF)
    env.key_filename = cfglib.CONF.migrate.key_filename
    cloud = cloud_ferry.CloudFerry(cfglib.CONF)
    LOG.info("running evacuation")
    evacuation_chain.process_chain(cloud, iteration)

    freed_nodes = get_freed_nodes(iteration)

    if not freed_nodes:
        LOG.warning("Evacuation cannot be completed, because there are no "
                    "available compute nodes, that should be freed")
        return

    LOG.info("Following nodes will be freed once in-cloud migration finishes, "
             "and can be moved from source to destination: %s", freed_nodes)
コード例 #4
0
ファイル: fabfile.py プロジェクト: cburns-mirantis/CloudFerry
def create_filters(name_config=None,
                   filter_folder=DEFAULT_FILTERS_FILES,
                   images_date='2000-01-01'):
    """Generates filter files for CloudFerry based on the schedule prepared by
    condensation/grouping."""
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    make_filters.make(filter_folder, images_date)
コード例 #5
0
ファイル: fabfile.py プロジェクト: cburns-mirantis/CloudFerry
def condense(config=None, vm_grouping_config=None, debug=False):
    """
    When migration is done in-place (there's no spare hardware), cloud
    migration admin would want to free as many hardware nodes as possible. This
    task handles that case by analyzing source cloud and rearranging source
    cloud load and generating migration scenario.

    Condensation is a process of:
     1. Retrieving groups of connected VMs (see `get_groups`)
     2. Rearrangement of source cloud load in order to free up as many
        hardware nodes as possible.
     3. Generation filter files for CloudFerry, which only contain groups of
        VMs identified in step 1 in the order identified in step 2.

    Method arguments:
     :config: - path to CloudFerry configuration file (based on
                `configs/config.ini`)
     :vm_grouping_config: - path to grouping config file (based on
                `configs/groups.yaml`)
     :debug: - boolean value, enables debugging messages if set to `True`
    """
    if debug:
        utils.configure_logging("DEBUG")
    cfglib.collector_configs_plugins()
    cfglib.init_config(config)
    data_storage.check_redis_config()

    LOG.info("Retrieving flavors, VMs and nodes from SRC cloud")
    flavors, vms, nodes = nova_collector.get_flavors_vms_and_nodes(cfglib.CONF)

    if cfglib.CONF.condense.keep_interim_data:
        condense_utils.store_condense_data(flavors, nodes, vms)

    LOG.info("Retrieving groups of VMs")

    # get_groups stores results in group_file_path config
    get_groups(config, vm_grouping_config)
    groups = condense_utils.read_file(cfglib.CONF.migrate.group_file_path)
    if groups is None:
        message = ("Grouping information is missing. Make sure you have "
                   "grouping file defined in config.")

        LOG.critical(message)
        raise RuntimeError(message)

    LOG.info("Generating migration schedule based on grouping rules")
    process.process(nodes=nodes, flavors=flavors, vms=vms, groups=groups)

    LOG.info("Starting generation of filter files for migration")
    create_filters(config)

    LOG.info("Migration schedule generated. You may now want to start "
             "evacuation job: 'fab evacuate'")

    LOG.info("Condensation process finished. Checkout filters file: %s.",
             DEFAULT_FILTERS_FILES)
コード例 #6
0
ファイル: fabfile.py プロジェクト: japaniel/CloudFerry
def condense(config=None, vm_grouping_config=None, debug=False):
    """
    When migration is done in-place (there's no spare hardware), cloud
    migration admin would want to free as many hardware nodes as possible. This
    task handles that case by analyzing source cloud and rearranging source
    cloud load and generating migration scenario.

    Condensation is a process of:
     1. Retrieving groups of connected VMs (see `get_groups`)
     2. Rearrangement of source cloud load in order to free up as many
        hardware nodes as possible.
     3. Generation filter files for CloudFerry, which only contain groups of
        VMs identified in step 1 in the order identified in step 2.

    Method arguments:
     :config: - path to CloudFerry configuration file (based on
                `configs/config.ini`)
     :vm_grouping_config: - path to grouping config file (based on
                `configs/groups.yaml`)
     :debug: - boolean value, enables debugging messages if set to `True`
    """
    if debug:
        utils.configure_logging("DEBUG")
    cfglib.collector_configs_plugins()
    cfglib.init_config(config)
    data_storage.check_redis_config()

    LOG.info("Retrieving flavors, VMs and nodes from SRC cloud")
    flavors, vms, nodes = nova_collector.get_flavors_vms_and_nodes(cfglib.CONF)

    if cfglib.CONF.condense.keep_interim_data:
        condense_utils.store_condense_data(flavors, nodes, vms)

    LOG.info("Retrieving groups of VMs")

    # get_groups stores results in group_file_path config
    get_groups(config, vm_grouping_config)
    groups = condense_utils.read_file(cfglib.CONF.migrate.group_file_path)
    if groups is None:
        message = ("Grouping information is missing. Make sure you have "
                   "grouping file defined in config.")

        LOG.critical(message)
        raise RuntimeError(message)

    LOG.info("Generating migration schedule based on grouping rules")
    process.process(nodes=nodes, flavors=flavors, vms=vms, groups=groups)

    LOG.info("Starting generation of filter files for migration")
    create_filters(config)

    LOG.info("Migration schedule generated. You may now want to start "
             "evacuation job: 'fab evacuate'")

    LOG.info("Condensation process finished. Checkout filters file: %s.",
             DEFAULT_FILTERS_FILES)
コード例 #7
0
ファイル: fabfile.py プロジェクト: paulmathews/CloudFerry
def migrate(name_config=None, name_instance=None):
    """
        :name_config - name of config yaml-file, example 'config.yaml'
    """
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    utils.init_singletones(cfglib.CONF)
    env.key_filename = cfglib.CONF.migrate.key_filename
    cloud = cloud_ferry.CloudFerry(cfglib.CONF)
    cloud.migrate(Scenario())
コード例 #8
0
ファイル: fabfile.py プロジェクト: wzbbbb/CloudFerry
def migrate(name_config=None, name_instance=None):
    """
        :name_config - name of config yaml-file, example 'config.yaml'
    """
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    utils.init_singletones(cfglib.CONF)
    env.key_filename = cfglib.CONF.migrate.key_filename
    cloud = cloud_ferry.CloudFerry(cfglib.CONF)
    cloud.migrate()
コード例 #9
0
def migrate(name_config=None, name_instance=None, debug=False):
    """
        :name_config - name of config yaml-file, example 'config.yaml'
    """
    if debug:
        utl.configure_logging("DEBUG")
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    utils.init_singletones(cfglib.CONF)
    env.key_filename = cfglib.CONF.migrate.key_filename
    cloud = cloud_ferry.CloudFerry(cfglib.CONF)
    cloud.migrate(Scenario())
コード例 #10
0
ファイル: fabfile.py プロジェクト: andybondar/CloudFerry
def migrate(name_config=None, name_instance=None, debug=False):
    """
        :name_config - name of config yaml-file, example 'config.yaml'
    """
    if debug:
        utils.configure_logging("DEBUG")
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    utils.init_singletones(cfglib.CONF)
    env.key_filename = cfglib.CONF.migrate.key_filename
    cloud = cloud_ferry.CloudFerry(cfglib.CONF)
    cloud.migrate(Scenario(path_scenario=cfglib.CONF.migrate.scenario,
                           path_tasks=cfglib.CONF.migrate.tasks_mapping))
コード例 #11
0
ファイル: fabfile.py プロジェクト: cburns-mirantis/CloudFerry
def migrate(name_config=None, name_instance=None, debug=False):
    """
        :name_config - name of config yaml-file, example 'config.yaml'
    """
    if debug:
        utils.configure_logging("DEBUG")
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    utils.init_singletones(cfglib.CONF)
    env.key_filename = cfglib.CONF.migrate.key_filename
    env.connection_attempts = cfglib.CONF.migrate.ssh_connection_attempts
    cloud = cloud_ferry.CloudFerry(cfglib.CONF)
    cloud.migrate(
        Scenario(path_scenario=cfglib.CONF.migrate.scenario,
                 path_tasks=cfglib.CONF.migrate.tasks_mapping))
コード例 #12
0
ファイル: fabfile.py プロジェクト: andybondar/CloudFerry
def get_groups(name_config=None, group_file=None, cloud_id='src',
               validate_users_group=False):
    """
    Function to group VM's by any of those dependencies (f.e. tenants,
    networks, etc.).

    :param name_config: name of config ini-file, example 'config.ini',
    :param group_file: name of groups defined yaml-file, example 'groups.yaml',
    :param validate_users_group: Remove dublicate id's and check if valid
           VM id specified. Takes more time because of nova API multiple calls
    :return: yaml-file with tree-based groups defined based on grouping rules.
    """
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    group = grouping.Grouping(cfglib.CONF, group_file, cloud_id)
    group.group(validate_users_group)
コード例 #13
0
ファイル: fabfile.py プロジェクト: cburns-mirantis/CloudFerry
def get_groups(name_config=None,
               group_file=None,
               cloud_id='src',
               validate_users_group=False,
               debug=False):
    """
    Function to group VMs by any of those dependencies (f.e. tenants,
    networks, etc.).

    :param name_config: name of config ini-file, example 'config.ini',
    :param group_file: name of groups defined yaml-file, example 'groups.yaml',
    :param validate_users_group: Remove duplicate IDs and check if valid
           VM id specified. Takes more time because of nova API multiple calls
    :return: yaml-file with tree-based groups defined based on grouping rules.
    """

    if debug:
        utils.configure_logging("DEBUG")

    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    group = grouping.Grouping(cfglib.CONF, group_file, cloud_id)
    group.group(validate_users_group)
コード例 #14
0
ファイル: fabfile.py プロジェクト: VitaliiYerys/CloudFerry
def nova_collector(name_config=None):
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    nova_collector_module.run_it(cfglib.CONF)
コード例 #15
0
ファイル: fabfile.py プロジェクト: cburns-mirantis/CloudFerry
def get_condensation_info(name_config=None):
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    nova_collector.run_it(cfglib.CONF)
コード例 #16
0
ファイル: fabfile.py プロジェクト: VitaliiYerys/CloudFerry
def create_filters(name_config=None, filter_folder='configs/filters',
                   images_date='2000-01-01'):
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    make_filters.make(filter_folder, images_date)
コード例 #17
0
ファイル: fabfile.py プロジェクト: VitaliiYerys/CloudFerry
def condense(name_config=None, debug=False):
    if debug:
        utl.configure_logging("DEBUG")
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    process.process()
コード例 #18
0
ファイル: fabfile.py プロジェクト: japaniel/CloudFerry
def get_condensation_info(name_config=None):
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    nova_collector.run_it(cfglib.CONF)
コード例 #19
0
ファイル: fabfile.py プロジェクト: Arvinder86/CloudFerry
def load_config(name_config):
    cfglib.collector_configs_plugins()
    cfglib.init_config(name_config)
    utils.init_singletones(cfglib.CONF)
    if cfglib.CONF.migrate.hide_ssl_warnings:
        warnings.simplefilter("ignore")