django-azure-deploy-manually

this page contain proper account of deploying A Django project on Azure web service manually without CI/CD pipeline.

• Keep this private it contain secret environment variables;
• we are assuming here that you create Django project with best practices
  • You Django Project is a Repository of GitHub if not make it
  • You install all the python packages required for your project in a virtual environment. If not do it.

Configure Django Project For Deployment

Install additional packages for deployment

• I am assuming here that you already create a virtual environment and install all the package which are required for your Django app to run on the same.
• Install below packages on the same virtual environment

pip install psycopg2
pip install python-dotenv
pip install whitenoise

write deployment settings

• Create a deployment.py file on the directory settings.py present
• this file will treated as settings file in deployment environment
• copy and page the below code on that. change if needed.

import os
from .settings import *
from .settings import BASE_DIR

SECRET_KEY = os.environ['SECRET']
ALLOWED_HOSTS = [os.environ['WEBSITE_HOSTNAME']]
CSRF_TRUSTED_ORIGINS = ['https://' + os.environ['WEBSITE_HOSTNAME']]
DEBUG = False

# WhiteNoise configuration
MIDDLEWARE = [
    'django.middleware.security.SecurityMiddleware',
    'whitenoise.middleware.WhiteNoiseMiddleware',
    'django.contrib.sessions.middleware.SessionMiddleware',
    'django.middleware.common.CommonMiddleware',
    'django.middleware.csrf.CsrfViewMiddleware',
    'django.contrib.auth.middleware.AuthenticationMiddleware',
    'django.contrib.messages.middleware.MessageMiddleware',
    'django.middleware.clickjacking.XFrameOptionsMiddleware',
]

STATICFILES_STORAGE = 'whitenoise.storage.CompressedManifestStaticFilesStorage'
STATIC_ROOT = os.path.join(BASE_DIR, 'staticfiles')

'''
# a way to add database connection
conn_str = os.environ['AZURE_POSTGRESQL_CONNECTIONSTRING']
conn_str_params = {pair.split('=')[0]: pair.split('=')[1] for pair in conn_str.split(' ')}
DATABASES = {
    'default': {
        'ENGINE': 'django.db.backends.postgresql',
        'NAME': conn_str_params['dbname'],
        'HOST': conn_str_params['host'],
        'USER': conn_str_params['user'],
        'PASSWORD': conn_str_params['password'],
    }
}
'''

tell our app when to use deployment settings

• we need to change wsgi.py (present in same dir as settings.py in)
• copy pate the below code on the, don’t forget to change the project_name

import os
from django.core.wsgi import get_wsgi_application

project_name = "my-project-name"

settings_module = '.deployment' if 'WEBSITE_HOSTNAME' in os.environ else '.settings'

os.environ.setdefault('DJANGO_SETTINGS_MODULE', project_name+settings_module)

application = get_wsgi_application()

create requirements.txt

• caution: make sure virtual environment activate, and CMD is on the root directory of project (same on which manage.py present)
• run the below command on CMD to create a requirements.txt

pip freeze > requirements.txt

Add a .production and .gitignore file

• add a .production file in same directory as manage.py in with below code.

[config]
SCM_DO_BUILD_DURING_DEPLOYMENT=true

• add a .gitignore file in same dir as manage.py in with below code.

# Byte-compiled / optimized / DLL files
__pycache__/
*.py[cod]
*$py.class

# Distribution / packaging
.Python
build/
develop-eggs/
dist/
downloads/
eggs/
.eggs/
lib/
lib64/
parts/
sdist/
var/
wheels/
pip-wheel-metadata/
share/python-wheels/
*.egg-info/
.installed.cfg
*.egg
MANIFEST

# PyInstaller
#  Usually these files are written by a python script from a template
#  before PyInstaller builds the exe, so as to inject date/other infos into it.
*.manifest
*.spec

# Installer logs
pip-log.txt
pip-delete-this-directory.txt

# Unit test / coverage reports
htmlcov/
.tox/
.nox/
.coverage
.coverage.*
.cache
nosetests.xml
coverage.xml
*.cover
*.py,cover
.hypothesis/
.pytest_cache/

# Django stuff:
*.log
local_settings.py
db.sqlite3
db.sqlite3-journal

# PEP 582; used by e.g. github.com/David-OConnor/pyflow
__pypackages__/

# Environments
.env
.venv
env/
venv/
ENV/
env.bak/
venv.bak/

Create and Configure Azure resources

Create a App service on Azure

• with Python as a run-time environment. same version as you build your Django app on local machine

Set all necessary environment variables on azure service

you can use the JSON format for doing it faster

[
  {
    "name": "WEBSITE_HTTPLOGGING_RETENTION_DAYS",
    "value": "1",
    "slotSetting": false
  },
  
]

Deploying Project Manually without CI/CD pipeline

Make app service Authentication for GitHub with SSH key

1. Generate SSH-key with Azure App Service SSH

   ssh-keygen -t rsa -b 4096 -C "[email protected]"
   
   

   cat ~/.ssh/id-rsa.pub
   
   

2. Set this SSH-key in GitHub with GUI

   above is the direct link for that same site, better to save with the name of App-service

3. Verify Connection

   ssh -T [email protected]
   
   

Clone the Project Repo and Build it

1. Install git in App service

   apt install git
   
   

2. clone the repo

   cd ~/home/site/wwwroot/ # more to deployment directory
   git clone <ssh-url-of-repo> # remember ssh url and http url
   
   

3. Build it

   here the the build command either you can do it by yourself or just copy past below instruction in chat-gpt by putting your repo-name and github-username on top line and GPT will do give a handy command file for you that you can copy past directly on app-service SSH

   # my repo name is [pur-repo-name-here] my github username is [github-username]
   # re write all these command for me so that I can copy page these in my azure app service cli directly
   cd ~/home/site/wwwroot/ # more to deployment directory
   cd <repo-name> # move into the project
   git checkout <branch-name> # switch to branch you want to deploy
   cd .. # go back
   python -m venv antenv # create python virtual environment with name antenv
   source antenv/bin/activate # activate the virtual environment
   pip install -r <repo-name>/requirements.txt # install the requirement of your django project
   pip install gunicorn # might be gunicorn not in the requirements
   cd <repo-name> # again to into the repo directory
   py manage.py migrate # migrate the migrations
   py manage.py collectstatic --noinput # collect all the static files
   
   

Test if things done correctly or not

1. Open two window of app-service SSH and navigate to repo directory on both.

   cd ~/home/site/wwwroot/my-repo/ # same file where manage.py is present
   
   

2. test server is running properly with | python manage.py runserver

   # on SSH window 1
   py manage.py runserver 1010 # run the runsever on port 1010 as might be 8000 already busy
   # TIP: you can also see in this ouput which setting is working s dettings.py or deployment.py
   
   # on SSH window 2 while server is continuously running on window 1
   curl <http://localhost:1010/> # lisen to the loclhost port 1010
   # if it give no output --> good test passed
   # if it give some html context which show erro code 400 --> test failed something is wrong
   # if it give some html context which do not contain error --> text passed
   
   # now u can stop the server at window 1 by crt+C
   
   

3. test by running gunicorn server

   # on SSH window 1
   gunicorn --bind 0.0.0.0:1010 sharktodo2.wsgi:application --access-logfile - --error-logfile - # run the runsever on port 1010 as might be 8000 already busy
   
   # on SSH window 2 while server is continuously running on window 1
   curl <http://localhost:1010/> # lisen to the loclhost port 1010
   # if it give no output --> good test passed
   # if it give some html context which show erro code 400 --> test failed something is wrong
   # if it give some html context which do not contain error --> text passed
   
   # now u can stop the server at window 1 by crt+C
   
   

Tell app-service to run your Django app on hosting port (8000) not default one.

navigate to app service configuration and set give start-up command there and save it.

just wait for max 5 min and visit your app service URL your Django app will be live there

source antenv/bing/activate
cd [your-repo-name]
gunicorn --bind 0.0.0.0:8000 sharktodo2.wsgi:application --access-logfile - --error-logfile -

Some Debugging Tips

1. you can check Log of of from the app-service sidebar Log Stream
   1. and even you can turn more detailed log from app-service side-bar app-service-logs
2. have doubt my deployment.py settings are running or not
   1. just run python manage.py runserver from the ssh it will show which setting are running