my blog my blog

Category: Python
配置Django的Celery异步之路踩坑

终于的终于,奶牛还是踏上了新的踩坑之路。

人生苦短,我用python。

看到这句话的时候,感觉可能确实是很深得人心,不过每每想学学,就又止步,年纪大了,感觉学什么东西都很慢,很难,精神啊注意力啊思维啊都跟不上。今天奶牛来分享自己今天踩的一个坑。

先说说配置过程吧,初学Django,啥都不懂,当然,python也很水,啥东西都得现查现用。Django安装还是很简单的。

apt-get install python3
pip3 install django

嗯,就是两条命令的事儿。

再说celery的安装:

pip3 install celery
pip3 install redis==2.10.6

目前奶牛所在的时间redis for python的版本是redis-3.0.1,为什么要用2.10.6呢?因为3.0.1压根配置就无法运行!!!

继续安装redis server

apt-get install redis
service redis start

然后就可以按照celery的官方教程走了,放个URL:http://docs.celeryproject.org/en/latest/django/index.html

python3 manage.py startproject nenew
cd nenew
python3 manage.py startapp nenewapp
touch ./nenew/celery.py
touch ./nenewapp/tasks.py

然后增加nenew/nenew/celery.py内容为

from __future__ import absolute_import, unicode_literals
import os
from celery import Celery

# set the default Django settings module for the 'celery' program.
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'nenew.settings')

app = Celery('nenew')

# Using a string here means the worker doesn't have to serialize
# the configuration object to child processes.
# - namespace='CELERY' means all celery-related configuration keys
#   should have a `CELERY_` prefix.
app.config_from_object('django.conf:settings', namespace='CELERY')

# Load task modules from all registered Django app configs.
app.autodiscover_tasks()


@app.task(bind=True)
def debug_task(self):
    print('Request: {0!r}'.format(self.request))

增加nenew/nenew/__init__.py的内容

from __future__ import absolute_import, unicode_literals

# This will make sure the app is always imported when
# Django starts so that shared_task will use this app.
from .celery import app as celery_app

__all__ = ('celery_app',)

增加nenew/nenewtest/tasks.py的内容

# Create your tasks here
from __future__ import absolute_import, unicode_literals
from celery import shared_task


@shared_task
def add(x, y):
    return x + y


@shared_task
def mul(x, y):
    return x * y


@shared_task
def xsum(numbers):
    return sum(numbers)

在nenew/nenew/settings.py中增加和修改

...

ALLOWED_HOSTS = ['*']
....

INSTALLED_APPS = [
...
    'nenewtest',
]

...

CELERY_BROKER_URL = 'redis://localhost:6379/1'
CELERY_RESULT_BACKEND = ‘redis://localhost:6379/0'

在nenew/nenewtest/views.py中增加或修改为

from django.shortcuts import render
from django.http import HttpResponse
from .tasks import add
# Create your views here.
def nenewtest(request):
    result = add.delay('2','2')
    result.ready()
    return HttpResponse('nenew Django Celery worker run !')

最后把views添加到nenew/nenew/urls.py中

from django.contrib import admin
from django.urls import path
from nenewtest import views

urlpatterns = [
    path('admin/', admin.site.urls),
    path('test/', views.nenewtest),
]

然后在项目目录nenew执行

celery -A nenew worker -l info

这时候worker启动正常会显示:

-------------- [email protected] v4.2.1 (windowlicker)
---- **** -----
--- * ***  * -- Linux-4.15.0-39-generic-x86_64-with-Ubuntu-18.04-bionic 2018-11-23 17:31:25
-- * - **** ---
- ** ---------- [config]
- ** ---------- .> app:         nenew:0x7fdc5a155cc0
- ** ---------- .> transport:   redis://localhost:6379/1
- ** ---------- .> results:     redis://localhost:6379/0
- *** --- * --- .> concurrency: 1 (prefork)
-- ******* ---- .> task events: OFF (enable -E to monitor tasks in this worker)
--- ***** -----
 -------------- [queues]
                .> celery           exchange=celery(direct) key=celery

这样类似的信息,然后我们启动项目,这里需要新开一个shell:

python3 manage.py runserver 0:80

这样我们就可以通过80端口直接访问我们的web了。地址是http://locahost/test

当我们这里访问正常后,在worker界面会有

[2018-11-23 18:09:19,469: INFO/MainProcess] Received task: nenewtest.tasks.add[35faa0fe-dd48-4f8d-9559-516556a93a40]
[2018-11-23 18:09:19,470: INFO/ForkPoolWorker-1] Task nenewtest.tasks.add[35faa0fe-dd48-4f8d-9559-516556a93a40] succeeded in 0.00031037399821798317s: '22'

如下语句表示执行成功,这样子就表示通过Django的网页我们对celery任务的异步执行成功。

当然,按照我的方法是可以一步成功的,因为奶牛已经踩了一整天的坑了,被一个错误郁闷得不要不要的。

AttributeError: 'float' object has no attribute 'items'

就是这个错误,查遍国内的所有网站都没有结果,然后就去bing的国际版查,然后发现果然是有bug在啊,奶牛这一天浪费得可真是够了。

这是celery的一个issue,在地址https://github.com/celery/celery/issues/5175 ,issue里面提及在2018/11/22日 it’s fixed in kombu and celery master。问题的根源是celery对redis3的支持不好,补救方法是

Solution: Roll back redis with pip: pip install redis==2.10.6

然后在commit里面找到是对requirements/extras/redis.txt文件进行版本限定。

-redis>=2.10.5
+redis>=2.10.5,<3

好了,踩坑总算结束了。毕竟是新手,很多问题可能就是潜意识认为是自己的代码有问题,实际。。。实际可能并不是这样子,得多关注源码的更迭和问题处理才好。