Resolving Bad Request (400) when using Django on an EC2 instance DEBUG

Django EC2 Bad Request (400)

This gist pertains to a Bad Request(400) error when trying to access an EC2 instance running django . . . 

I set up an EC2 server, installed django, started runserver and tested. The result was

Bad Request (400)

When using Django 1.5 and higher, if you specify DEBUG=False in the settings.py file, you must also specify ALLOWED_HOSTS 
or you will recieve a "Bad Request (400)" error message when attempting to retrieve a page.

On an Amazone EC2 instance, if you are ***testing*** with the public DNS, then you should include the public DNS as one of the allowed hosts.

--- example ---

DEBUG = False

ALLOWED_HOSTS = ['.yourdomain.com', 'ec2-XX-XX-XX-XXX.us-west-2.compute.amazonaws.com',]

--- end ---

If have SSH'ed into the server and wish to perform on-server testing you should include 'localhost' as well: 
python manage.py runserver 0:8000
then . . .
curl http://localhost:8000/  
or . . .
telnet localhost 8000 (then type GET / and hit enter twice)

If that does not work add 'localhost' to your ALLOWED_HOSTS file.

ALLOWED_HOSTS = ['.yourdomain.com', 'ec2-XX-XX-XX-XXX.us-west-2.compute.amazonaws.com', 'localhost',]

========

NOTE WELL: If you reboot your EC2 instance, when it reboots it will have a different public DNS and you 
will need to update your ALLOWED_HOSTS accordingly.


See the 1.5 release notes for more: 
https://docs.djangoproject.com/en/dev/releases/1.5/#allowed-hosts-required-in-production

+1

I too had this issue and solved it by adding following to wsgi.py

import sys, os
sys.path.insert(0, '/opt/python/current/app')   # I use ElasticBeanstalk, modify to your project path
os.environ.setdefault('DJANGO_SETTINGS_MODULE', 'your.settings')

I got 400 error, because I forget to import a model class in views.

+1

+1

+1

+1

+1

+1

+1

+1