Source code for ceilometer.notification

#
# Copyright 2017 Red Hat, Inc.
# Copyright 2012-2013 eNovance <licensing@enovance.com>
#
# 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 itertools
import threading
import time
import uuid

from concurrent import futures
import cotyledon
from futurist import periodics
from oslo_config import cfg
from oslo_log import log
import oslo_messaging
from stevedore import named
from tooz import coordination

from ceilometer.i18n import _
from ceilometer import messaging
from ceilometer import utils


LOG = log.getLogger(__name__)


OPTS = [
    cfg.IntOpt('pipeline_processing_queues',
               deprecated_for_removal=True,
               default=10,
               min=1,
               help='Number of queues to parallelize workload across. This '
                    'value should be larger than the number of active '
                    'notification agents for optimal results. WARNING: '
                    'Once set, lowering this value may result in lost data.'),
    cfg.BoolOpt('ack_on_event_error',
                default=True,
                help='Acknowledge message when event persistence fails.'),
    cfg.BoolOpt('workload_partitioning',
                deprecated_for_removal=True,
                default=False,
                help='Enable workload partitioning, allowing multiple '
                     'notification agents to be run simultaneously.'),
    cfg.MultiStrOpt('messaging_urls',
                    default=[],
                    secret=True,
                    help="Messaging URLs to listen for notifications. "
                         "Example: rabbit://user:pass@host1:port1"
                         "[,user:pass@hostN:portN]/virtual_host "
                         "(DEFAULT/transport_url is used if empty). This "
                         "is useful when you have dedicate messaging nodes "
                         "for each service, for example, all nova "
                         "notifications go to rabbit-nova:5672, while all "
                         "cinder notifications go to rabbit-cinder:5672."),
    cfg.IntOpt('batch_size',
               default=100, min=1,
               help='Number of notification messages to wait before '
               'publishing them. Batching is advised when transformations are '
               'applied in pipeline.'),
    cfg.IntOpt('batch_timeout',
               default=5,
               help='Number of seconds to wait before publishing samples '
               'when batch_size is not reached (None means indefinitely)'),
    cfg.IntOpt('workers',
               default=1,
               min=1,
               deprecated_group='DEFAULT',
               deprecated_name='notification_workers',
               help='Number of workers for notification service, '
               'default value is 1.'),
    cfg.MultiStrOpt('pipelines',
                    default=['meter', 'event'],
                    help="Select which pipeline managers to enable to "
                    " generate data"),
]


EXCHANGES_OPTS = [
    cfg.MultiStrOpt('notification_control_exchanges',
                    default=['nova', 'glance', 'neutron', 'cinder', 'heat',
                             'keystone', 'sahara', 'trove', 'zaqar', 'swift',
                             'ceilometer', 'magnum', 'dns', 'ironic', 'aodh'],
                    deprecated_group='DEFAULT',
                    deprecated_name="http_control_exchanges",
                    help="Exchanges name to listen for notifications."),
]


[docs]class NotificationService(cotyledon.Service):
    """Notification service.

    When running multiple agents, additional queuing sequence is required for
    inter process communication. Each agent has two listeners: one to listen
    to the main OpenStack queue and another listener(and notifier) for IPC to
    divide pipeline sink endpoints. Coordination should be enabled to have
    proper active/active HA.
    """

    NOTIFICATION_NAMESPACE = 'ceilometer.notification.v2'

    def __init__(self, worker_id, conf, coordination_id=None):
        super(NotificationService, self).__init__(worker_id)
        self.startup_delay = worker_id
        self.conf = conf

        self.periodic = None
        self.shutdown = False
        self.listeners = []
        # NOTE(kbespalov): for the pipeline queues used a single amqp host
        # hence only one listener is required
        self.pipeline_listener = None

        if self.conf.notification.workload_partitioning:
            # XXX uuid4().bytes ought to work, but it requires ascii for now
            coordination_id = (coordination_id or
                               str(uuid.uuid4()).encode('ascii'))
            self.partition_coordinator = coordination.get_coordinator(
                self.conf.coordination.backend_url, coordination_id)
            self.partition_set = list(range(
                self.conf.notification.pipeline_processing_queues))
            self.group_state = None
        else:
            self.partition_coordinator = None

[docs]    def get_targets(self):
        """Return a sequence of oslo_messaging.Target

        This sequence is defining the exchange and topics to be connected.
        """
        topics = (self.conf.notification_topics
                  if 'notification_topics' in self.conf
                  else self.conf.oslo_messaging_notifications.topics)
        return [oslo_messaging.Target(topic=topic, exchange=exchange)
                for topic in set(topics)
                for exchange in
                set(self.conf.notification.notification_control_exchanges)]

    def _log_missing_pipeline(self, names):
        LOG.error(_('Could not load the following pipelines: %s'), names)

[docs]    def run(self):
        # Delay startup so workers are jittered
        time.sleep(self.startup_delay)

        super(NotificationService, self).run()
        self.coord_lock = threading.Lock()

        self.managers = [ext.obj for ext in named.NamedExtensionManager(
            namespace='ceilometer.notification.pipeline',
            names=self.conf.notification.pipelines, invoke_on_load=True,
            on_missing_entrypoints_callback=self._log_missing_pipeline,
            invoke_args=(self.conf,
                         self.conf.notification.workload_partitioning))]

        self.transport = messaging.get_transport(self.conf)

        if self.conf.notification.workload_partitioning:
            self.partition_coordinator.start(start_heart=True)
        else:
            # FIXME(sileht): endpoint uses the notification_topics option
            # and it should not because this is an oslo_messaging option
            # not a ceilometer. Until we have something to get the
            # notification_topics in another way, we must create a transport
            # to ensure the option has been registered by oslo_messaging.
            messaging.get_notifier(self.transport, '')

        self._configure_main_queue_listeners()

        if self.conf.notification.workload_partitioning:
            # join group after all manager set up is configured
            self.hashring = self.partition_coordinator.join_partitioned_group(
                self.NOTIFICATION_NAMESPACE)

            @periodics.periodic(spacing=self.conf.coordination.check_watchers,
                                run_immediately=True)
            def run_watchers():
                self.partition_coordinator.run_watchers()
                if self.group_state != self.hashring.ring.nodes:
                    self.group_state = self.hashring.ring.nodes.copy()
                    self._refresh_agent()

            self.periodic = periodics.PeriodicWorker.create(
                [], executor_factory=lambda:
                futures.ThreadPoolExecutor(max_workers=10))
            self.periodic.add(run_watchers)
            utils.spawn_thread(self.periodic.start)

    def _configure_main_queue_listeners(self):
        endpoints = []
        for pipe_mgr in self.managers:
            endpoints.extend(pipe_mgr.get_main_endpoints())
        targets = self.get_targets()

        urls = self.conf.notification.messaging_urls or [None]
        for url in urls:
            transport = messaging.get_transport(self.conf, url)
            # NOTE(gordc): ignore batching as we want pull
            # to maintain sequencing as much as possible.
            listener = messaging.get_batch_notification_listener(
                transport, targets, endpoints, allow_requeue=True)
            listener.start(
                override_pool_size=self.conf.max_parallel_requests
            )
            self.listeners.append(listener)

    def _refresh_agent(self):
        with self.coord_lock:
            if self.shutdown:
                # NOTE(sileht): We are going to shutdown we everything will be
                # stopped, we should not restart them
                return
            self._configure_pipeline_listener()

    def _configure_pipeline_listener(self):
        partitioned = list(filter(
            self.hashring.belongs_to_self, self.partition_set))

        endpoints = []
        for pipe_mgr in self.managers:
            endpoints.extend(pipe_mgr.get_interim_endpoints())

        targets = []
        for mgr, hash_id in itertools.product(self.managers, partitioned):
            topic = '-'.join([mgr.NOTIFICATION_IPC, mgr.pm_type, str(hash_id)])
            LOG.debug('Listening to queue: %s', topic)
            targets.append(oslo_messaging.Target(topic=topic))

        if self.pipeline_listener:
            self.kill_listeners([self.pipeline_listener])

        self.pipeline_listener = messaging.get_batch_notification_listener(
            self.transport, targets, endpoints, allow_requeue=True,
            batch_size=self.conf.notification.batch_size,
            batch_timeout=self.conf.notification.batch_timeout)
        # NOTE(gordc): set single thread to process data sequentially
        # if batching enabled.
        batch = (1 if self.conf.notification.batch_size > 1
                 else self.conf.max_parallel_requests)
        self.pipeline_listener.start(override_pool_size=batch)

[docs]    @staticmethod
    def kill_listeners(listeners):
        # NOTE(gordc): correct usage of oslo.messaging listener is to stop(),
        # which stops new messages, and wait(), which processes remaining
        # messages and closes connection
        for listener in listeners:
            listener.stop()
            listener.wait()

[docs]    def terminate(self):
        self.shutdown = True
        if self.periodic:
            self.periodic.stop()
            self.periodic.wait()
        if self.partition_coordinator:
            self.partition_coordinator.stop()
        with self.coord_lock:
            if self.pipeline_listener:
                self.kill_listeners([self.pipeline_listener])
            self.kill_listeners(self.listeners)

        super(NotificationService, self).terminate()