stacktach/worker/worker.py
Monsyne Dragon 3e4eb35653 Fix db reconnect issue under django 1.6+
Django's orm layer will not auto-reconnect after it looses
the connection to the mysql server, until you manually close the
database connection in django 1.6 and above.

(see: https://code.djangoproject.com/ticket/21597)

This is an issue for persistant connections, as MySQL will timeout
inactive connections, and any loss of the db connection will cause
the stacktach worker to simply repeat the error
"MySQL server has gone away" until restarted.

This fix will allow the stacktach worker to properly reconnect.

Change-Id: I0b0bc75b7e21fd183f3b0e7a55d727ff98d6f02b
2014-07-25 19:38:21 +00:00

241 lines
8.5 KiB
Python

# Copyright 2012 - Dark Secret Software 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.
# This is the worker you run in your OpenStack environment. You need
# to set TENANT_ID and URL to point to your StackTach web server.
import datetime
import sys
import time
import signal
import kombu
import kombu.mixins
try:
import ujson as json
except ImportError:
try:
import simplejson as json
except ImportError:
import json
from pympler.process import ProcessMemoryInfo
from django.db import connection as db_connection
from stacktach import db
from stacktach import message_service
from stacktach import stacklog
from stacktach import views
stacklog.set_default_logger_name('worker')
shutdown_soon = False
def _get_child_logger():
return stacklog.get_logger('worker', is_parent=False)
class Consumer(kombu.mixins.ConsumerMixin):
def __init__(self, name, connection, deployment, durable, queue_arguments,
exchange, topics, connect_max_retries=10):
self.connect_max_retries = connect_max_retries
self.retry_attempts = 0
self.connection = connection
self.deployment = deployment
self.durable = durable
self.queue_arguments = queue_arguments
self.name = name
self.last_time = None
self.pmi = None
self.processed = 0
self.total_processed = 0
self.topics = topics
self.exchange = exchange
signal.signal(signal.SIGTERM, self._shutdown)
def _create_exchange(self, name, type, exclusive=False, auto_delete=False):
return message_service.create_exchange(name, exchange_type=type,
exclusive=exclusive,
durable=self.durable,
auto_delete=auto_delete)
def _create_queue(self, name, nova_exchange, routing_key, exclusive=False,
auto_delete=False):
return message_service.create_queue(
name, nova_exchange, durable=self.durable, auto_delete=exclusive,
exclusive=auto_delete, queue_arguments=self.queue_arguments,
routing_key=routing_key)
def get_consumers(self, Consumer, channel):
exchange = self._create_exchange(self.exchange, "topic")
queues = [self._create_queue(topic['queue'], exchange,
topic['routing_key'])
for topic in self.topics]
return [Consumer(queues=queues, callbacks=[self.on_nova])]
def _process(self, message):
routing_key = message.delivery_info['routing_key']
body = str(message.body)
args = (routing_key, json.loads(body))
asJson = json.dumps(args)
# save raw and ack the message
raw, notif = views.process_raw_data(
self.deployment, args, asJson, self.exchange)
self.processed += 1
message.ack()
POST_PROCESS_METHODS[raw.get_name()](raw, notif)
self._check_memory()
def _check_memory(self):
if not self.pmi:
self.pmi = ProcessMemoryInfo()
self.last_vsz = self.pmi.vsz
self.initial_vsz = self.pmi.vsz
utc = datetime.datetime.utcnow()
check = self.last_time is None
if self.last_time:
diff = utc - self.last_time
if diff.seconds > 30:
check = True
if check:
self.last_time = utc
self.pmi.update()
diff = (self.pmi.vsz - self.last_vsz) / 1000
idiff = (self.pmi.vsz - self.initial_vsz) / 1000
self.total_processed += self.processed
per_message = 0
if self.total_processed:
per_message = idiff / self.total_processed
_get_child_logger().debug("%20s %20s %6dk/%6dk ram, "
"%3d/%4d msgs @ %6dk/msg" %
(self.name, self.exchange, diff, idiff, self.processed,
self.total_processed, per_message))
self.last_vsz = self.pmi.vsz
self.processed = 0
def on_nova(self, body, message):
try:
self._process(message)
except Exception, e:
_get_child_logger().debug("Problem: %s\nFailed message body:\n%s" %
(e, json.loads(str(message.body))))
raise
def _shutdown(self, signal, stackframe = False):
global shutdown_soon
self.should_stop = True
shutdown_soon = True
def on_connection_revived(self):
self.retry_attempts = 0
_get_child_logger().debug("The connection to RabbitMQ was revived.")
def on_connection_error(self, exc, interval):
self.retry_attempts += 1
msg = ("RabbitMQ Broker connection error: %r. "
"Trying again in %s seconds.", exc, interval)
if self.retry_attempts >= self.connect_max_retries:
# If we're on the last retry
_get_child_logger().error(*msg)
else:
_get_child_logger().warn(*msg)
def on_decode_error(self, message, exc):
_get_child_logger().exception("Decode Error: %s" % exc)
# do NOT call message.ack(), otherwise the message will be lost
def continue_running():
return not shutdown_soon
def exit_or_sleep(exit=False):
if exit:
sys.exit(1)
#so django 1.6+ will properly reconnect
db_connection.close()
time.sleep(5)
def run(deployment_config, deployment_id, exchange):
name = deployment_config['name']
host = deployment_config.get('rabbit_host', 'localhost')
port = deployment_config.get('rabbit_port', 5672)
user_id = deployment_config.get('rabbit_userid', 'rabbit')
password = deployment_config.get('rabbit_password', 'rabbit')
virtual_host = deployment_config.get('rabbit_virtual_host', '/')
durable = deployment_config.get('durable_queue', True)
queue_arguments = deployment_config.get('queue_arguments', {})
exit_on_exception = deployment_config.get('exit_on_exception', False)
topics = deployment_config.get('topics', {})
logger = _get_child_logger()
deployment = db.get_deployment(deployment_id)
print "Starting worker for '%s %s'" % (name, exchange)
logger.info("%s: %s %s %s %s %s" %
(name, exchange, host, port, user_id, virtual_host))
params = dict(hostname=host,
port=port,
userid=user_id,
password=password,
transport="librabbitmq",
virtual_host=virtual_host)
# continue_running() is used for testing
while continue_running():
try:
logger.debug("Processing on '%s %s'" % (name, exchange))
with kombu.connection.BrokerConnection(**params) as conn:
try:
consumer = Consumer(name, conn, deployment, durable,
queue_arguments, exchange,
topics[exchange])
consumer.run()
except Exception as e:
logger.error("!!!!Exception!!!!")
logger.exception(
"name=%s, exchange=%s, exception=%s. "
"Reconnecting in 5s" % (name, exchange, e))
exit_or_sleep(exit_on_exception)
logger.debug("Completed processing on '%s %s'" %
(name, exchange))
except Exception:
logger.error("!!!!Exception!!!!")
e = sys.exc_info()[0]
msg = "Uncaught exception: deployment=%s, exchange=%s, " \
"exception=%s. Retrying in 5s"
logger.exception(msg % (name, exchange, e))
exit_or_sleep(exit_on_exception)
logger.info("Worker exiting.")
signal.signal(signal.SIGINT, signal.SIG_IGN)
signal.signal(signal.SIGTERM, signal.SIG_IGN)
POST_PROCESS_METHODS = {
'RawData': views.post_process_rawdata,
'GlanceRawData': views.post_process_glancerawdata,
'GenericRawData': views.post_process_genericrawdata
}