Skip to content

Based on Docker

Deploy a Dockerize Taipy application with Heroku


1. Create your Dockerfile

Along with your application, you must create a Dockerfile that will allow Docker to build your container.

Here is an example with as the entry point of your application and requirements.txt file with all your dependencies (Taipy included):

# Your Python version
FROM python:3.9

# Install your application
COPY . /app
RUN pip install -r requirements.txt

# Startup command
CMD python


This Dockerfile is provided as an example and tested without security consideration, do not use it for a production environment.

2. Build and test your Docker

At the same location as your Dockerfile, run the following command to create a Docker image named : docker build . -t <my-taipy-app>

You can now run it with the command line: docker run -p 5000:5000 -d --name <my-taipy-app> <my-taipy-app>

Open a browser and go to http://localhost:5000 -- there may be some startup latency --.

3. Deployment

In our example, we use the name <my-taipy-app> for our application. On Heroku, this name should be unique. So you should replace it consistently with a custom value.

heroku login
heroku create <my-taipy-app>
heroku container:login
heroku container:push web -a <my-taipy-app>
heroku config:set CLIENT_URL="https://<my-taipy-app>" -a <my-taipy-app>
heroku container:release web -a <my-taipy-app>

4. Check your deployment

You can go to the url https://<my-taipy-app> with your browser or run heroku open -a <my-taipy-app>. Your application should be deployed correctly.

5. Clean up your resources

Stop the local docker container: docker stop <my-taipy-app>

Remove the local docker container: docker rm <my-taipy-app>

Remove the local docker image: docker rmi <my-taipy-app>

Remove the Heroku application: heroku apps:destroy <my-taipy-app> --confirm <my-taipy-app>