Python学习web框架开发

Flask：在Flask中使用Celery实现异步任务

2021-07-24  本文已影响0人  xiaogp

摘要：Flask，Celery

Celery简介

Celery是一个专注于实时处理和任务调度的分布式任务队列，使用Celery的常见场景：

Celery组件

（1）Celery包含的组件如下

Celery的各组件的工作流程如下


（2）消息代理的选择

Celery推荐使用RabbitMQ，Redis，如果选择Redis则存在断电停机数据丢失的问题

（3）数据序列化

数据在消息队列的传输需要序列化和反序列化，Celery支持json，yaml，msgpack，默认是json，其中msgpack是一个二进制类似json的序列化方案，比json数据结构更小更快

快速开始：Flask + Celery异步发送邮件

（1）安装准备

安装准备，以redis作为消息代理和结果存储，在安装celery，flask

pip install celery
pip install redis

版本如下

celery --version
5.1.2 (sun-harmonics)

flask --version
Python 3.7.6
Flask 1.1.1
Werkzeug 1.0.0
（2）Flask应用脚本

构建一个Web应用实现给输入的邮箱发送邮件，发送邮件采用Python自带模块smtplib和email，前端使用HTML渲染，Flask后台完成POST表单获取邮箱数据发送邮件，分别对比同步执行和使用Celery异步执行的响应的时间

import os
import sys
import traceback
import smtplib
from email.mime.multipart import MIMEMultipart
from email.mime.text import MIMEText
from email.header import Header

sys.path.append(os.path.dirname(os.path.abspath(__file__)))

from flask import Flask, render_template, request, redirect, url_for, session
from celery import Celery

from settings import Config

app = Flask(__name__)
app.config.from_object(Config)
app.config['CELERY_BROKER_URL'] = 'redis://localhost:6379/0'  # 消息代理
app.config['CELERY_RESULT_BACKEND'] = 'redis://localhost:6379/0'  # 任务结果写入redis
app.config['CELERY_TASK_SERIALIZER'] = 'msgpack'  # 序列化方式
app.config['CELERY_ACCEPT_CONTENT'] = ['msgpack', 'json']  # 如果指定了msgpack序列化方式，需要增加msgpack为可接受

celery = Celery(app.name, broker=app.config['CELERY_BROKER_URL'])
celery.conf.update(app.config)  # 使用flask app的config

@app.route("/", methods=["GET", "POST"])
def index():
    if request.method == "GET":
        return render_template("index.html")
    email = request.form.get("email")
    if request.form.get("submit") == "Send":
        # send_async_email.delay("直接发送", email)
        import time
        t = time.time()
        # send_async_email("直接发送", email)
        send_async_email.delay("直接发送", email)
        # TODO somethine
        print(time.time() - t)
    elif request.form.get("submit") == "Send in 1 minute":
        send_async_email.apply_async(["延迟1分分钟发送", email], countdown=60)
    return redirect(url_for('index'))


@celery.task(name='main.send_async_email')
def send_async_email(message, to_email):
    send_mail(message, to_email)


def send_mail(message, to_email):
    conn = None
    try:
        conn = smtplib.SMTP_SSL(app.config["SMTP_HOST"], app.config["SMTP_PORT"])
        conn.login(app.config["FROM_EMAIL_ACCOUNT"], app.config["FROM_EMAIL_PASSWORD"])
        msg = MIMEMultipart()
        subject = Header('测试邮件', 'utf-8').encode()
        msg['Subject'] = subject
        msg['From'] = app.config["FROM_EMAIL_ACCOUNT"]
        msg['To'] = to_email
        text = MIMEText(message, 'plain', 'utf-8')
        msg.attach(text)
        conn.sendmail(app.config["FROM_EMAIL_ACCOUNT"], to_email, msg.as_string())
    except Exception as e:
        traceback.print_exc()
    finally:
        if conn:
            conn.quit()


if __name__ == '__main__':
    app.run("0.0.0.0", "5010")

脚本主要有2个地方加入和调用了Celery生成的任务

下一步启动Flask，最终的Web界面显示效果如下


（3）启动Celery消费者

只有启动了消费者，脚本中需要被传输给中间件的任务才会被执行，下一步启动Celery的消费者，在项目根目录下命令行中输入

root@ubuntu:~/myproject/CELERY_TEST# celery -A celery_test.main:celery worker -l info
/opt/anaconda3/lib/python3.7/site-packages/celery/platforms.py:835: SecurityWarning: You're running the worker with superuser privileges: this is
absolutely not recommended!

Please specify a different user using the --uid option.

User information: uid=0 euid=0 gid=0 egid=0

  uid=uid, euid=euid, gid=gid, egid=egid,
 
 -------------- celery@ubuntu (sun-harmonics)
--- ***** ----- 
-- ******* ---- Linux-4.15.0-54-generic-x86_64-with-debian-buster-sid 2021-07-22 19:57:25
- *** --- * --- 
- ** ---------- [config]
- ** ---------- .> app:         celery_test.main:0x7f6430716690
- ** ---------- .> transport:   redis://localhost:6379/0
- ** ---------- .> results:     redis://localhost:6379/0
- *** --- * --- .> concurrency: 4 (prefork)
-- ******* ---- .> task events: OFF (enable -E to monitor tasks in this worker)
--- ***** ----- 
 -------------- [queues]
                .> celery           exchange=celery(direct) key=celery
                

[tasks]
  . main.send_async_email

[2021-07-22 19:57:25,863: WARNING/MainProcess] Please run `celery upgrade settings path/to/settings.py` to avoid these warnings and to allow a smoother upgrade to Celery 6.0.
[2021-07-22 19:57:26,144: INFO/MainProcess] Connected to redis://localhost:6379/0
[2021-07-22 19:57:26,169: INFO/MainProcess] mingle: searching for neighbors
[2021-07-22 19:57:27,192: INFO/MainProcess] mingle: all alone
[2021-07-22 19:57:27,231: INFO/MainProcess] celery@ubuntu ready.

出现最后一行ready表示消费者启动成功，同时可以看到celery配置的相关参数：

Celery可以采用后台运行的方式，并且将日志写入指定目录，方法是使用celery multi来管理任务的启动和停止等，指定logfile日志目录和pidfile的pid目录，以一个自定义的命名来定义任务名，比如web

root@ubuntu# celery multi start web -A celery_test.main:celery -l info --logfile=logs/celery_%n.log --pidfile=celery_%n.pid
celery multi v5.1.2 (sun-harmonics)
> Starting nodes...
    > web@ubuntu: OK(sun-harmonics)
> Starting nodes...
    > web@ubuntu: OK
root@ubuntu:~/myproject
root@ubuntu:~/myproject/CELERY_TEST# ls
celery_test  celery_web.pid  logs  __pycache__  settings.py
root@ubuntu:~/myproject/CELERY_TEST/logs# ls
celery_web.log

任务在后台启动，并在在目录下生成了pid文件也log目录

（4）异步任务测试

在界面输入邮箱地址发送邮件，点击send发送，此时是采用delay方式执行函数，采用异步发送邮件，视图函数中使用time进行计时，假设在发送邮件下面还有其他任务，记录这些总共花费的时间
异步方式执行完发送邮件和其他任务总计花费8ms

127.0.0.1 - - [23/Jul/2021 14:32:44] "POST / HTTP/1.1" 302 -
127.0.0.1 - - [23/Jul/2021 14:32:44] "GET / HTTP/1.1" 200 -
0.009320974349975586
127.0.0.1 - - [23/Jul/2021 14:32:46] "POST / HTTP/1.1" 302 -
127.0.0.1 - - [23/Jul/2021 14:32:46] "GET / HTTP/1.1" 200 -
0.008226156234741211
127.0.0.1 - - [23/Jul/2021 14:32:48] "POST / HTTP/1.1" 302 -
127.0.0.1 - - [23/Jul/2021 14:32:48] "GET / HTTP/1.1" 200 -
0.007288694381713867

然后使用原始的Python函数不调用delay（send_async_email("直接发送", email)），需要800ms左右，并且网页端明显感觉有卡顿

0.8022255897521973
127.0.0.1 - - [23/Jul/2021 14:35:00] "POST / HTTP/1.1" 302 -
127.0.0.1 - - [23/Jul/2021 14:35:00] "GET / HTTP/1.1" 200 -
0.7504754066467285
127.0.0.1 - - [23/Jul/2021 14:35:02] "POST / HTTP/1.1" 302 -
127.0.0.1 - - [23/Jul/2021 14:35:02] "GET / HTTP/1.1" 200 -
127.0.0.1 - - [23/Jul/2021 14:35:04] "POST / HTTP/1.1" 302 -
127.0.0.1 - - [23/Jul/2021 14:35:04] "GET / HTTP/1.1" 200 -
0.8088760375976562
0.8098959922790527

进一步修改html加入计算机的功能，将执行函数改为直接sleep 10秒，可以在Web界面发现提交之后等10秒才能得到计算结果

@celery.task(name='main.send_async_email')
def send_async_email(message, to_email):
    import time
    time.sleep(10)

@app.route("/", methods=["GET", "POST"])
def index():
    if request.method == "GET":
        return render_template("index.html")
    email = request.form.get("email")
    val1 = request.form.get("value1")
    val2 = request.form.get("value2")
    if request.form.get("submit") == "Send":
        # send_async_email.delay("直接发送", email)
        import time
        t = time.time()
        send_async_email("直接发送", email)
        # send_async_email.delay("异步发送", email)
        value3 = float(val1) + float(val2)
    return render_template("index.html", **locals())
（5）观察Redis中存储的队列和处理结果

修改消息队列的broker地址为redis库1，结果处理地址为redis库2，查看redis的数据

root@ubuntu:~# redis-cli 
127.0.0.1:6379> select 1;
(error) ERR invalid DB index
127.0.0.1:6379> select 1
OK
127.0.0.1:6379[1]> keys *
1) "_kombu.binding.celery.pidbox"
2) "_kombu.binding.celeryev"
3) "_kombu.binding.celery"
127.0.0.1:6379[1]> select 2
OK
127.0.0.1:6379[2]> keys *
1) "celery-task-meta-4b64592e-bcc5-4dcf-b838-e0960b7c8e4b"
2) "celery-task-meta-80cb2a66-23f0-4e22-80af-7faa39f12094"
127.0.0.1:6379[2]> get celery-task-meta-80cb2a66-23f0-4e22-80af-7faa39f12094
"{\"status\": \"SUCCESS\", \"result\": null, \"traceback\": null, \"children\": [], \"date_done\": \"2021-07-23T07:19:06.224197\", \"task_id\": \"80cb2a66-23f0-4e22-80af-7faa39f12094\"}"

其中kombu是Celery使用kombu 维护消息队列存储的相关数据，celery-task-meta相关的数据是存储结果数据，json类型，包含处理状态，时间，处理结果，任务id等

上一篇 下一篇

猜你喜欢

热点阅读