1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
|
+++
title = "Git – how to setup your own server pt 2"
author = "Danilo M."
type = "tech"
date = "2018-12-17T07:39:08+00:00"
excerpt = "followup on my previous article regarding GIT and how to automatically deploy python flask scripts to serve a web app with apache."
draft = true
image = "/uppies/2018/12/gitout.jpg"
categories = [ "DIY", "Code"]
tags = [ "git", "linux", "python", "howto", "automation", "ssh", "flask", "do it yourself"]
+++
This is a followup on [my previous article about how to setup your server][1] to handle a GIT repository and deploy to a web server like apache.
Since I started experimenting with python and web publishing with it I wanted a way to push all the changes to my codebase directly to a web server, in a similar way as to what I already do with php and apache.
> I want to do it myself, the way I like it..
>
> — danix
Serving python scripts is slightly different than serving html or php, here's a scheme that shows what happens when using python (flask in this example) to serve some web content.
As you can see the web server is acting as a reverse proxy to the [WSGI][2] server Gunicorn, Apache receives a request from a client, forwards the request to Gunicorn which in turn asks the python web service, in our case our flask web app. Other than as a reverse proxy, apache is used also to serve static content which is not originated in python, like images, css or js scripts.
In order to deploy a python web app I had to install the mod_wsgi extension for apache, then I decided to use a folder in my system to install the virtual environment for python to run my web app and point to that directory in my vhost configuration file inside apache. After that I created a directory where to store all files for my web app and instructed apache to serve them using the virtualenv I created earlier. This is more or less what I did:
```bash
# create the virtual environment for python inside
# the folder /usr/local/virtualenvs/my-app:
virtualenv /usr/local/virtualenvs/my-app
# now create the directory inside the apache root
# to store our web app
mkdir /var/www/htdocs/my-app
```
Now if we put all the py files for our flask web app we can run it inside apache, this is the configuration for the vhost I used:
```apache
<VirtualHost *:80>
ServerName my-app.mysite.ext
ErrorLog "/var/log/httpd/my-app.error_log"
CustomLog "/var/log/httpd/my-app.access_log" common
WSGIDaemonProcess my-app user=apache group=apache threads=5 python-home=/usr/local/virtualenvs/my-app
WSGIScriptAlias / /var/www/htdocs/my-app/my-app.wsgi
WSGIProcessGroup my-app
Alias "/static/" "/var/www/htdocs/my-app/static/"
<Directory "/var/www/htdocs/my-app/static/">
Order allow,deny
Allow from all
</Directory>
<Directory "/var/www/htdocs/my-app/">
Require all granted
</Directory>
</VirtualHost>
```
With this setting I can now reach my web app and test it after restarting the web server.
[1]: https://danix.xyz/2018/07/git-setup-own-server/
[2]: https://en.wikipedia.org/wiki/Web_Server_Gateway_Interface
|
