Django+Gunicorn+nginx Internal Server Error, where is the error and how to fix it?

Question

So I am running pretty standart setup, I followed the same tutorial I did before but now it doesn't work for unknown reasons.

When I run ./manage.py runserver my_ip:8000 it works fine. When I run my gunicorn script via bin/gunicorn_start it works fine and creates the sock file However when I run gunicorn script via supervisor, and nginx, it results with Internal Server Error, and no info in error logs. What am I doing wrong? I guess it's gunicorn or permission problems?

bin/gunicorn_start

#!/bin/bash
 
NAME="today" # Name of the application
DJANGODIR=~/deployment/today_project/
SOCKFILE=~/deployment/run/gunicorn.sock # we will communicte using this unix socket
USER=ferski # the user to run as
GROUP=ferski # the group to run as
NUM_WORKERS=3 # how many worker processes should Gunicorn spawn
DJANGO_SETTINGS_MODULE=today_project.settings # which settings file should Django use
DJANGO_WSGI_MODULE=today_project.wsgi # WSGI module name
 
echo "Starting $NAME as `whoami`"
 
# Activate the virtual environment
cd $DJANGODIR
source ../bin/activate
export DJANGO_SETTINGS_MODULE=$DJANGO_SETTINGS_MODULE
export PYTHONPATH=$DJANGODIR:$PYTHONPATH
 
# Create the run directory if it doesn't exist
RUNDIR=$(dirname $SOCKFILE)
test -d $RUNDIR || mkdir -p $RUNDIR
 
# Start your Django Unicorn
# Programs meant to be run under supervisor should not daemonize themselves (do not use --daemon)
exec ../bin/gunicorn ${DJANGO_WSGI_MODULE}:application \
--name $NAME \
--workers $NUM_WORKERS \
--user=$USER --group=$GROUP \
--log-level=debug \
--bind=unix:$SOCKFILE

/etc/supervisord.conf (just uncomented files)

[unix_http_server]
file=/tmp/supervisor.sock   ; (the path to the socket file)
...
[supervisorctl]
serverurl=unix:///tmp/supervisor.sock ; use a unix:// URL  for a unix socket
...
[program:today]
command = /home/ferski/deployment/bin/gunicorn_start
user = ferski
stdout_logfile = /home/ferski/deployment/logs/gunicorn_supervisor.log
redirect_stderr = true
...

And finally nginx.conf

upstream today_app_server {
        server unix:/home/ferski/deployment/run/gunicorn.sock fail_timeout=0;
}

server {
        listen 80;
        server_name haxelita.pl;

        client_max_body_size 4G;
        access_log /home/ferski/deployment/logs/nginx-access.log;
        error_log /home/ferski/deployment/logs/nginx-error.log;

        location /static/ {
                alias /home/ferski/deployment/today_project/today/static/;
        }

        location / {

                proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
                proxy_set_header Host $http_host;
                proxy_redirect off;

                if (!-f $request_filename) {
                        proxy_pass http://today_app_server;
                        break;
                }
        }

         # Error pages
        error_page 500 502 503 504 /500.html;
        location = /500.html {
                root /home/ferski/deployment/today_project/today/static/;
        }
}

Whats wrong here? Is it permission problems?

Answer 1

You can't activate a virtualenv environment through source in a script that gets started with supervisor. Use directory=/home/ferski/deployment/today_project/ in your supervisord.conf.

See Supervising virtualenv django app via supervisor

Answer 2

This problem confused the hell out of me. But I figured it out, I thought i would post it for future reference, since this is the top google search for this error.

    uwsgi --socket /path/to/sock --chdir /path/to/django/project/ --module project_name.wsgi --chmod-socket=664

Then you can add in the number of cores and processors but appending this (change 4 to the number of core in your application).

    --master --processes 4 --async 4 --ugreen