DF-BPMN provides a Docker image to run the BPMN modeler as a Container in Docker or Kubernetes.
The DF-BPMN Docker image is based on the official NodeJS image (node:16-buster). The container image contains a prebuild appliction and exposes the port 3000
In the Dockerfile we are using start script as the entrypoint:
ENTRYPOINT [ "/usr/src/app/start.sh" ]
Note: The start script is setting the environment param --hostname=0.0.0.0 which is important to allow access form outside the container. Find more details also here and here.
To build the docker image from sources run:
$ docker build . -t alinoureldin/df-bpmn --build-arg OPENAI_KEY=<open-api-key>
To run the docker image locally run:
$ docker run --name="df-bpmn" --rm -p 3000:3000 -p 3001:3001 alinoureldin/df-bpmn --env OPENAI_KEY=<open-api-key>
After starting the container the applicaiton is available on
http://localhost:3000/#/usr/src/app/workspace
To stop the container run:
$ docker stop df-bpmn ## Workspace For the integration with Bonita studio, you should add `-v` to mount the workspace of bonita project within the docker image
$ docker run --name="df-bpmn" --rm -p 3000:3000 -p 3001:3001 --env OPENAI_KEY=<open-api-key> -it -v /path/to/bonita-workspace:/usr/src/app/bonita alinoureldin/df-bpmn
When you need to export a DF-BPMN project to Bonita proc, you should add the /bonita-project-name/ and the BDM of bonita will automatically imported, and the export of DF-BPMN will be within download within your Bonita project.
To push the image manually to a docker repo:
$ docker build . -t alinoureldin/df-bpmn:latest
$ docker push alinoureldin/df-bpmn:latest