Cool, now lets try a full example. Lets say we have a flask webapp and rq as a
background worker.
Lets put this Dockerfile into a new folder:
FROM sherzberg/python-all-dev
RUN apt-get update -qq && apt-get install -y python-pip
RUN pip install flask rq
WORKDIR /app
EXPOSE 5000
VOLUME ["/app"]
This Dockerfile will install the base dependencies for a Flask app. The base image sherzberg/python-all-dev already has the development versions of python installed.
You also need to create a file called app.py in that folder:
from flask import Flask
app = Flask(__name__)
@app.route('/')
def hello_world():
return 'Hello World!'
if __name__ == '__main__':
app.run(host='0.0.0.0')
And lets see if it works:
$ docker build -t workshop/python-full .
...
$ docker run -i -p 5000:5000 -v $(pwd):/app -t workshop/python-full python app.py
If you don't get any errors, navigate to http://your-ip:5000
First, we need redis:
$ docker run --name redis -d redis
Now create tasks.py for the rq worker code.
import os
from redis import Redis
from rq import Queue, Worker, Connection
redis = Redis(os.environ['REDIS_PORT_6379_TCP_ADDR'], os.environ['REDIS_PORT_6379_TCP_PORT'])
q = Queue('default', connection=redis)
def calculate(a, b):
result = a + b
print("{} + {} = {}".format(a, b, result))
return result
if __name__ == '__main__':
with Connection(redis):
worker = Worker(map(Queue, ['default']))
worker.work()
Run the rq worker:
$ sudo docker run -i -p 5000:5000 --link redis:redis -v $(pwd):/app -t workshop/python-full python tasks.py
Try it out:
$ sudo docker run -i --link redis:redis -v $(pwd):/app -t workshop/python-full python
>>> from tasks import q, calculate
>>> q.enqueue(calculate, 1, 3)
Look at the console output of the worker.
calculate on /calculate?a=1&b=3workshop/python-full to allow docker run -t workshop/-full to serve the app.