Running Flask App with HttpPlatformHandler in Azure App Services(Windows)

Article
09/01/2017

Flask is a micro web framework written in Python and based on the Werkzeug toolkit and Jinja2 template engine.

Below are list of steps we would follow

Create Sample Project
Create Azure WebApp and Use Site Extension to Upgrade Python
Create and Change Deployment script
Adding web.config (for production app with waitress)
Publish App

You can find a Sample Python Flask project with above operations @ GitHub Link

Create Sample Project in local environment

Create index.py file with below content

 from flask import Flask  
app = Flask(__name__)

@app.route("/")
def hello():  
    return "Hello World!"

if __name__ == "__main__":  
    app.run()

Create requirements.txt file with below content

 Flask==0.12

Install dependencies listed in requirements.txt file using below command.

 pip install --upgrade -r requirements.txt

Run app in local environment using below command

 $ python index.py 
 * Running on https://127.0.0.1:5000/ (Press CTRL+C to quit)

Navigate to https://127.0.0.1:5000/ and you should see app up and running

Create Azure WebApp and Use Site Extension to Upgrade Python

Navigate to Azure portal

Create a new web app
Setup Continuous Deployment
Navigate to your App Service blade, select Extensions and then Add.
From the list of extensions, scroll down until you spot the Python logos, then choose the version you need

For this blog I'm choosing Python 2.7.12 x64, It would install new version of python @ D:\home\Python27

If you choose Python 3.5.2 x64, It would install new version of python @ D:\home\Python35

Create and Change Deployment script

I have explained more on how to utilize deployment script and what it does @ link. you can use this script to run commands on Azure App Services after moving code to it.

Install Azure Cli using below command

 npm install azure-cli -g

Use Below command to create deployment script.

 azure site deploymentscript --python

Above step would create below two files

.deployment
deploy.cmd

Note: If you haven't used nodejs/npm earlier, Create above files manually and copy content for these files from https://github.com/prashanthmadi/azure-flask-httpplatformhandler

Replace content of deploy.cmd file with content at Link

We are changing default deployment script(removing virtual environment creation and others) and adding below content to install packages listed in requirements.txt

 :: 2. Install packages
echo Pip install requirements.  
D:\home\Python27\python.exe -m pip install --upgrade -r requirements.txt  
IF !ERRORLEVEL! NEQ 0 goto error

If you are using Python35 site extension change path from D:\home\Python27\python.exe to D:\home\Python35\python.exe in deploy.sh

 D:\home\Python35\python.exe -m pip install --upgrade -r requirements.txt

Adding web.config(Production Mode - Using Waitress)

Waitress is meant to be a production-quality pure-Python WSGI server with very acceptable performance

Create web.config file with below content

 <?xml version="1.0" encoding="UTF-8"?>  
<configuration>  
 <system.webServer>
  <handlers>
   <add name="httpPlatformHandler" path="*" verb="*" 
         modules="httpPlatformHandler" resourceType="Unspecified" />
  </handlers>

  <httpPlatform processPath="D:\home\Python27\python.exe"
 arguments="run_waitress_server.py" requestTimeout="00:04:00" startupTimeLimit="120" startupRetryCount="3" stdoutLogEnabled="true">
   <environmentVariables>
    <environmentVariable name="PYTHONPATH" value="D:\home\site\wwwroot" />
    <environmentVariable name="PORT" value="%HTTP_PLATFORM_PORT%" />
   </environmentVariables>
  </httpPlatform>
 </system.webServer>
</configuration>

If you observe carefully, we have

httpplatform arguments section to run run_waitress_server.py file
Added new evnironmentVariable PORT to pass the internal port number we use in waitress

If you use Python35 site extension, change web.config content to use D:\home\Python35\python.exe for processPath

  <httpPlatform processPath="D:\home\Python35\python.exe"
 arguments="run_waitress_server.py" requestTimeout="00:04:00" startupTimeLimit="120" startupRetryCount="3" stdoutLogEnabled="true">

Add below line at the end in requirements.txt file. This would install waitress module during deployment

 waitress==1.0.1

Create a new file run_waitress_server.py with below content

 import os  
from waitress import serve  
from index import app

serve(app,host="0.0.0.0",port=os.environ["PORT"])

Publish App

Navigate to your root folder and commit your changes to WEB_APP_GIT_URL

 git init  
git add .  
git commit -m "initial commit"  
git remote add sampleflaskapp WEB_APP_GIT_URL  
git push sampleflaskapp master

Here is my App on Azure After publish

Django Sample App Azure