Gunicorn Environment Variable Setting - django

Question

I'm currently having difficulty passing environment variables into Gunicorn for my Django project. I'm on the latest 19.1 version. I have a wsgi.py file like so:
import os
import sys
from django.core.wsgi import get_wsgi_application
BASE_DIR = os.path.dirname(os.path.abspath(__file__))
PROJECT_DIR = os.path.abspath(os.path.join(BASE_DIR, '..'))
sys.path.append(PROJECT_DIR)
os.environ.setdefault("DJANGO_SETTINGS_MODULE", "app.settings")
def application(environ, start_response):
_application = get_wsgi_application()
os.environ['SERVER_ENV'] = environ['SERVER_ENV']
os.environ['SERVER_ID'] = environ['SERVER_ID']
return _application(environ, start_response)
When I run gunicorn from the command line as:
SERVER_ENV=TEST SERVER_ID=TEST gunicorn -b 127.0.0.1:8080 --error-logfile - --access-logfile - app.wsgi:application
and I then pass a request to gunicorn I keep getting:
2014-08-01 08:39:17 [21462] [ERROR] Error handling request
Traceback (most recent call last):
File "/opt/virtualenv/python-2.7.5/django-1.5.5/lib/python2.7/site-packages/gunicorn/workers/sync.py", line 93, in handle
self.handle_request(listener, req, client, addr)
File "/opt/virtualenv/python-2.7.5/django-1.5.5/lib/python2.7/site-packages/gunicorn/workers/sync.py", line 134, in handle_request
respiter = self.wsgi(environ, resp.start_response)
File "/opt/sites/itracker/wsgi.py", line 18, in application
os.environ['SERVER_ENV'] = environ['SERVER_ENV']
KeyError: 'SERVER_ENV'
Printing out the environ values confirms that the environment variables are not being passed in. I've tried setting the environment variables in the virtualenv activation script and also in a dedicated gunicorn shell script and also by setting the environment variables using the --env flag but nothing seems to work.
Any ideas?

Answer 1

I got a similar problem when deploying gunicorn.service. And I passed environment variables to it by a file:
<on gunicorn.service>
[Service]
...
EnvironmentFile=/pathto/somefilewith_secrets
...
For example (cat /etc/systemd/system/gunicorn.service)
[Unit]
Description=gunicorn daemon
After=network.target
[Service]
User=ubuntu
Group=ubuntu
WorkingDirectory=/home/ubuntu/10008/digichainOpen
EnvironmentFile=/home/ubuntu/10008/digichainOpen/.env
ExecStart=/home/ubuntu/.local/share/virtualenvs/digichainOpen-Zk2Jnvjv/bin/gunicorn \
--worker-class=gevent --workers 4 \
--bind unix:/home/ubuntu/10008/digichainOpen/gunicorn.sock digichainOpen.wsgi:application
[Install]
WantedBy=multi-user.target
and the .env file can be:
my_var=someValue
some_secret=secretvalue
another_secret=blah

Answer 2

You just have to export your environment variable.
export SERVER_ENV=TEST
export SERVER_ID=TEST
gunicorn -b 127.0.0.1:8080 --error-logfile - --access-logfile - app.wsgi:application
And in your code you can get them like that
os.getenv('SERVER_ENV')

Answer 3

If you want to run Django using gunicorn config file:
Write a config.py file
command = 'venv/bin/gunicorn'
pythonpath = 'venv'
bind = '127.0.0.1:8000'
workers = 2
raw_env = ["VARIABLE_HERE=VARIABLE_VALUE_HERE"]
wsgi_app = "project.wsgi"
Run it like this:
From inside the project directory
gunicorn -c config.py

Answer 4

I don't quite understand what you are trying to do here. If you pass environment variables in the bash command line, they are already in os.environ: there is no need to get them from anywhere else. The environ dictionary is made up of elements passed from the request, not the shell.