python manage.py runserver sleeps forever because of threading lock

[Krish]

I have upgraded Django 1.6 o 1.7.7. I was run my project using 'python manage.py runserver' command. It went well in test server. However, when I tried same in production server, it sleeps forever. Detail Server info is attached below. Doing trackback, I figure out problem, It lies on django.apps.registry.populate function. Trace back steps as follows:

1. Command line execution starts.

   django.core.management:: utility.execute()
   

2. It calls application setup process, where it loads enabled Apps

   django.core.management ::jango.setup()
   django.setup:: apps.populate(settings.INSTALLED_APPS)
   

3. Thread starts threading lock process, so that same application not loaded twice.

   django.apps.registry::with self._lock:
   

4. It goes through for each app,

   django.apps.registry::app_config.ready()
   ...
   # after many lines of codes
   ...
   for app_config in self.get_app_configs():
   app_config.ready()
   

5. When it matches <AdminConfig: admin>, it calls to

   django.contrib.admin.apps:: self.module.autodiscover()
   django.contrib.admin:: autodiscover_modules('admin', register_to=site)
   

6. autodiscover_modules agains calls django.setup(). I don't have full traceback. But it goes like:

   django.core.management ::django.setup()
   

   django.setup::apps.populate(settings.INSTALLED_APPS)
   

   django.apps.registry::with self._lock:
   

Application stuck here in django.apps.registry::with self._lock: line because the threading lock has been already acquired and not released.

Python:
Python 2.7.6 on both server

Dev Server:
Distributor ID: Ubuntu
Description: Ubuntu 14.04.1 LTS
Release: 14.04
Codename: trusty
CPU: 2.40GHz x 1
MemTotal: 1017880kb

Production Server:
Distributor ID: Ubuntu
Description: Ubuntu 14.04 LTS
Release: 14.04
Codename: trusty
CPU: 2.00GHz x 24
MemTotal: 4942810