tornado celery config文件

tornado+celery的简单使用详解

celery是实现一个简单，灵活可靠的分布式任务队列系统的好选择

首先是celeryconfig文件的编写

celeryconfig.py

#coding=utf-8
from kombu import Queue

CELERY_DEFAULT_EXCHANGE_TYPE = 'direct'

# 时区
CELERY_TIMEZONE = 'Asia/Shanghai'0
# 启动时区
CELERY_ENABLE_UTC = True

# 默认队列
CELERY_DEFAULT_QUEUE = 'mytask0'
# 默认队列的Key
CELERY_DEFAULT_ROUTING_KEY = 'task.mytask0'

# 创建队列，执行异步任务
CELERY_QUEUES = (
  Queue('mytask',  routing_key='task.mytask'),
)

# 创建执行路由
CELERY_ROUTES = {
  'task.mytask': {
    'queue': 'mytask',
    'routing_key': 'task.mytask',
  },
}

# 定时任务
CELERYBEAT_SCHEDULE = {
    'task.mytask': {
        'task': 'task.mytask'',
        'schedule': crontab(minute="1"),
    },
}

task.py

#coding=utf-8
from celery import Celery
from celery.bin import worker as celery_worker
import celeryconfig

broker = 'amqp://'
backend = 'amqp'
app = Celery('task', backend=backend, broker=broker)
app.config_from_object(celeryconfig)

@app.task
def mytask(task_name):
  print "task:%s" %task_name
  return task_name 

def worker_start():
  worker = celery_worker.worker(app=app)
  worker.run(broker=broker, concurrency=4,
        traceback=False, loglevel='INFO')

if __name__ == "__main__":
  worker_start()

执行python task.py将会启动worker

tornado调用celery将阻塞任务变为非阻塞

这会使用到tcelery模块，即tornado下的一个非阻塞的broker实现