Traditionally, I have configured the UWSGI configuration file to call an application like below:
mydirectory/uwsgi_application.ini
...
#python module to import
app = run_web
module = %(app)
callable = app
...
,
mydirectory/run_web.py
from ersapp import app
if __name__ == "__main__":
app.run()
,
mydirectory/ersapp/__init__.py
...
app = Flask('ersapp')
...
But now, I am following Miguel Grinberg's Flask book and here he uses an application factory like below
mydirectory/ersapp/__init__.py
...
def create_app(config_name):
webapp = Flask(__name__)
...
return webapp
with a "manager" (see flask-script Manager)
mydirectory/manage.py
from webapp import create_app
...
webapp = create_app(os.getenv('FLASK_CONFIG') or 'default')
manager = Manager(webapp)
...
if __name__ == '__main__':
manager.run()
In this configuration, I trigger my development server with $ python manage.py runserver
whereas before I triggered it with $ python run_web.py
Now, I am struggling with what to put in the uwsgi configuration file to allow this app to be deployed via UWSGI. Specifically, the app
, module
, and callable
variables.
The error I am getting in my logs is:
...
--- no python application found, check your startup logs for errors ---
...
You don't use Flask-Script with uWSGI. You point it at the app directly. Or in your case, you point it at a call to the app factory. The simplest example is:
uwsgi --module 'myapp:create_app()'
I am also using the flask-script
manager , so what i did was
created a
wsgi.py
.which was like.from app import * application = create_app("development")
then
uwsgi --wsgi-file wsgi.py --callable application
Note: the callable is the flask object name in the wsgi.py
.
来源:https://stackoverflow.com/questions/30721269/deploying-flask-app-with-uwsgi-and-flask-script-manager