How to deploy FlexMeasures
Here you can learn how to get FlexMeasures onto a server.
FlexMeasures can be deployed via Docker. Read more at Running via Docker.
On your own computer,
flexmeasures run is a nice way to start FlexMeasures. On a production web server, you want it done the WSGI way. Here is an example how to serve FlexMeasures as WSGI app:
# This file contains the WSGI configuration required to serve up your
# web application.
# It works by setting the variable 'application' to a WSGI handler of some description.
# The crucial part are the last two lines. We add some ideas for possible other logic.
project_home = u'/path/to/your/code/flexmeasures'
# use this if you want to load your own ``.env`` file.
from dotenv import load_dotenv
# use this if you run from source
if project_home not in sys.path:
sys.path = [project_home] + sys.path
# adapt PATH to find our LP solver if it is installed from source
os.environ["PATH"] = os.environ.get("PATH") + ":/home/seita/Cbc-2.9/bin"
# create flask app - the name "application" has to be passed to the WSGI server
from flexmeasures.app import create as create_app
application = create_app()
The web server is told about the WSGI script, but also about the object which represents the application. For instance, if this script is called
wsgi.py, then the relevant argument to the gunicorn server is
Keep in mind that FlexMeasures is based on Flask, so almost all knowledge on the web on how to deploy a Flask app also helps with deploying FlexMeasures.
Cbc needs to be present on the server where FlexMeasures runs, under the
You can install it on Debian like this:
apt-get install coinor-cbc
If you can’t use the package manager on your host, the solver has to be installed from source.
We provide an example script in
ci/install-cbc-from-source.sh to do that, where you can also
pass a directory for the installation.
In case you want to install a later version, adapt the version in the script.