Installation

open-plan-tool runs already live at open-plan-tool https://open-plan.rl-institut.de and the simulation server at https://mvs-open-plan.rl-institut.de/ However it is possible to install both of them locally on your machine or on your own server

Prior to the deploy you should find yourself at the root of the open-plan-tool GUI repository

git clone https://github.com/open-plan-tool/gui.git open-plan-tool-GUI
cd open-plan-tool-GUI

Deploy GUI locally using and using our open plan MVS server

Prior to be able to develop locally, you might need to install postgres, simply google install postgres followed by your os name (linux/mac/windows)

  1. Create a virtual environment

  2. Activate your virtual environment

  3. Install the dependencies with pip install -r app/requirements/postgres.txt

  4. Install extra local development dependencies with pip install -r app/dev_requirements.txt

  5. Move to the app folder with cd app

  6. Create environment variables (only replace content surrounded by <>)

SQL_ENGINE=django.db.backends.postgresql
SQL_DATABASE=<your db name>
SQL_USER=<your user name>
SQL_PASSWORD=<your password>
SQL_HOST=localhost
SQL_PORT=5432
DEBUG=(True|False)
  1. Add an environment variable MVS_HOST_API and set the url of the simulation server you wish to use for your models (https://mvs-open-plan.rl-institut.de/ if you wish to use our simulation server). You can deploy your own simulation server locally if you need

  2. Execute the local_setup.sh file (. local_setup.sh on linux/mac bash local_setup.sh on windows) you might have to make it executable first. Answer yes to the question

  3. Start the local server with python manage.py runserver

  4. You can then login with testUser and ASas12,. or create your own account

Deploy using Docker Compose

The following commands should get everything up and running, using the web based version of the MVS API.

You need to be able to run docker-compose inside your terminal. If you can’t you should install Docker desktop first.

  • Clone the repository locally git clone --single-branch --branch main https://github.com/open-plan-tool/gui.git open_plan_gui

  • Move inside the created folder (cd open_plan_gui)

  • Edit the .envs/epa.postgres and .envs/db.postgres environment files

    • Change the value assigned to EPA_SECRET_KEY with a randomly generated one

    • Make sure to replace dummy names with you preferred names

    • The value assigned to the variables POSTGRES_DB, POSTGRES_USER, POSTGRES_PASSWORD in .envs/db.postgres should match the ones of the variables SQL_DATABASE, SQL_USER, SQL_PASSWORD in .envs/epa.postgres, respectively

    • Define an environment variable MVS_HOST_API in .envs/epa.postgres and set the url of the simulation server you wish to use for your models (for example MVS_API_HOST="<url to your favorite simulation server>"), you can deploy your own simulation server locally if you need

    • Assign the domain of your website (without http:// or https://) to TRUSTED_HOST , see https://docs.djangoproject.com/en/4.2/ref/settings/#csrf-trusted-origins for more information

Next you can either provide the following commands inside a terminal (with ubuntu you might have to prepend sudo)

docker-compose --file=docker-compose-postgres.yml up -d --build

(you can replace postgres by mysql if you want to use mysql)

docker-compose --file=docker-compose-postgres.yml exec -u root app_pg sh initial_setup.sh

(this will also load a default testUser account with sample scenario).

Or you can run a python script with the following command

python deploy.py -db postgres

Finally, open a browser and navigate to http://localhost:8080 (or to http://localhost:8090 if you chose to use mysql instead of postgres): you should see the login page of the open-plan-tool app. You can then login with testUser and ASas12,. or create your own account.

Proxy settings (optional)

If you use a proxy you will need to set USE_PROXY=True and edit PROXY_ADDRESS=http://proxy_address:port with your proxy settings in .envs/epa.postgres.

Note

If you wish to use mysql instead of postgres, simply replace postgres by mysql and app_pg by app in the above commands or filenames

Note

Grab a cup of coffee or tea for this…

Test Account

You can access a preconfigured project using the following login credentials: testUser:ASas12,.


Tear down (uninstall) docker containers

To remove the application (including relevant images, volumes etc.), one can use the following commands in terminal:

docker-compose down --file=docker-compose-postgres.yml -v

you can add --rmi local if you wish to also remove the images (this will take you a long time to rebuild the docker containers from scratch if you want to redeploy the app later then)

Or you can run a python script with the following command

python deploy.py -db postgres --down