Simple Django project to practice. Creates list of products with ability to create/delete/modify list items. Deployed on Heroku: https://murmuring-ravine-49991.herokuapp.com/

• Install a relatively fresh version of Python. I used Python 3.6.8.
• Clone the repo or download the sources as ZIP.
• Create venv and install packages:

Linux:

# Ubuntu/Debian need installation of package for venv creation
sudo apt-get install python3-venv
python3 -m venv venv
source venv/bin/activate
pip install -r requirements/development.txt

Windows:

python3 -m venv venv
venv/Scripts/activate.bat
pip install -r requirements/development.txt

• Apply migrations: python manage.py migrate
• Run Django-server: python manage.py runserver 127.0.0.1:8000

Used OS:

• Host OS: Windows 10 Professional
• Guest OS: Ubuntu 18.04 Server

Used software:

• VirtualBox (Host OS) - software for VM management.
• Vagrant (Host OS) - useful tool to create VMs from config files (e.g. Vagrantfiles) . I used this configuration.
• Gunicorn (Guest OS) - HTTP WSGI server for Python (based on Ruby's Unicorn server). Hidden "behind" Nginx server: gets requests and responses to nginx, but does not communitate with clients directly. Nginx takes care of that.
• Nginx (Guest OS) - reverse proxy-server which is being used to enhance speed of request handling.
• Supervisor (Guest OS) - tool which is being used to restart Gunicorn if it crashes for some reason.

Note! Deployment algorithm can vary from one OS to another, so be careful when trying to copy-paste my implementation.

• Check interfaces possible for bridged connection:

cd C:/Program Files/Oracle/VirtualBox
VBoxManage list bridgedifs

• Go to Vagrantfile and place selected interface after the bridge. For instance:

config.vm.network "public_network", bridge: "Intel(R) Ethernet Connection (2) I219-V"

• Run: vagrant run. You will notice the following information: Adapter 1: nat and Adapter 2: bridged. Vagrant sets eth0 as NAT - this is fundamental requirement as noted here. So eth1 will be bridged interface.
• Login. Login: vagrant. Password: vagrant.
• Add PROD_LOCAL environment variable in the .bashrc, so Django will import correct settings needed for local deployment: sudo nano ~/.bashrc.
• Clone the repo: git clone https://github.com/K9173A/ProductCatalog.git
• Create virtual enviroment:

cd ProductCatalog
python3 -m venv venv
source venv/bin/activate
pip install -r requirements/production-local.txt

• Go to the 01-netcfg.yaml:

cd /etc/netplan
sudo nano 01-netcfg.yaml

• Paste the following config. We manually set static IP and switch off DHCP on eth1.

network:
  version: 2
  renderer: networkd
  ethernets:
    eth1:
      dhcp4: no
      dhcp6: no
      addresses: [192.168.0.98/24]
      gateway4: 192.168.0.1
      nameservers:
        addresses: [8.8.8.8, 192.168.0.1]

• Apply settings: sudo netplan apply
• Check the IP: ifconfig -a. It should be 192.168.0.98.

• Check Postgres activity: sudo systemctl status postgresql. It should be active.
• Enter the psql: sudo -u postgres psql.
• Create a new database:

CREATE DATABASE productcatalog;
CREATE USER admin with PASSWORD 'admin';
GRANT ALL PRIVILEGES ON DATABASE geekplanner TO admin;
ALTER ROLE admin SET CLIENT_ENCODING TO 'UTF8';
ALTER ROLE admin SET default_transaction_isolation TO 'READ COMMITTED';
ALTER ROLE admin SET TIME ZONE 'Europe/Moscow';

• Go to the pb_hba.conf:

cd /etc/postgresql/10/main
sudo nano pg_hba.conf

• Add host all all 192.168.0.0/24 md5 after IPv4 local connections line.
• Go to the postgresql.conf:

cd /etc/postgresql/10/main
sudo nano postgresql.conf

• Find title CONNETIONS AND AUTHENTICATION and set: listen_addresses = '*'
• Restart PostgreSQL and check its status:

sudo systemctl restart postgresql
sudo systemctl status postgresql

• Go to default file:

sudo nano /etc/nginx/sites-available/default

• Paste the following settings:

server {
  listen 80;
  server_name 192.168.0.98;
  access_log /var/log/nginx/example.log;

  location / {
    include proxy_params;
    proxy_pass http://127.0.0.1:8000;
  }

  location /static/ {
    root /home/vagrant/ProductCatalog;
  }
}

• Restart Nginx and check its status:

sudo systemctl restart nginx
sudo systemctl status nginx

• Check configuration file: sudo nginx -t.
• Enable FireWall and add a rule for Nginx:

sudo ufw enable
sudo ufw allow 'Nginx Full'

• Create Supervisor config:

cd /etc/supervisor/conf.d/
touch productcatalog.conf
sudo nano productcatalog.conf

• Paste the following settings:

[program:productcatalog]
command=/home/vagrant/ProductCatalog/venv/bin/gunicorn ProductCatalog.wsgi:application -c /home/vagrant/ProductCatalog/deploy/gunicorn.conf.py
directory=/home/vagrant/ProductCatalog
user=vagrant
autorestart=true
redirect_stderr=true

• Enable and start Supervisor:

sudo update-rc.d supervisor enable
sudo service supervisor start

• Apply migrations:

cd ProductCatalog/
source venv/bin/activate
python manage.py migrate
python manage.py createsuperuser

• Now you can test website from the host OS: http://192.168.0.98/.

Note! Heroku requires requirements.txt in the root directory even if we have separate files for development and production environment. Hence, we need requirements.txt to point to the right file (requirements/production-heroku.txt) if we want build to succeed.

• Install Heroku: sudo snap install heroku --classic.
• Login: heroku login.
• Clone the repo: git clone https://github.com/K9173A/ProductCatalog.
• Go to the root folder: cd ProductCatalog.
• Install and create virtual environment with requirements (@see "Running Locally").
• Create Heroku app: heroku create.
• Install PostgreSQL addon: heroku addons:create heroku-postgresql:hobby-dev.
• Push data to Heroku: git push heroku master.
• Apply migrations: heroku run python manage.py migrate --no-input.
• Create superuser: heroku run python manage.py createsuperuser.

ProductCatalog is released under the terms of the MIT license. See MIT and LICENSE for more information.