vampire-online

This is a web interface for vampire.

Frontend

React application created with create-react-app.

Uses .env files to store environment (local = .env.local| prod = .env.production) variables. (These files should be placed in the root folder of the frontend and NOT be checked in git, they are present in the .gitignore)

There is a special config to allow websockets in the nginx/nginx.conf file. Update the proxy_pass field under location /solver to point to the backend solver.

Template files are provided.

Set the following variables and remove _template from the filename

• REACT_APP_API_HOST - URL to backend
• REACT_APP_API_DEFAULT_TOKEN - JWT for userType="any" (when the server start it will print a frontend token, copy that one)

Has default react-app readme in folder.

Backend

NodeJs/Express thin server making calls to a vampire binary.
Has detailed readme in folder.

Deployment

Both parts are deployed using docker containers.

Prerequisites

• Access to a container registry (e.g. dockerhub) where you can push and/or pull images.

Local

• Docker daemon

Remote/Deploy Machine

• Docker daemon
• Docker compose
• problem_library and tutorial content (please see Deployment notes in /backend for details)

Steps

If the images you wish to deploy are already available on a remote repository you may skip steps 1 and 2.

Assuming both frontend and backend will be deployed on the same machine.

1. Build and tag the images

   docker build -t <image_name>:<release_tag> ./<frontend|backend>
   

2. Push the images to the container registry

   docker push -t <image_name>:<release_tag>
   

3. Update docker-compose to use the name and tag from build and transfer it to the remote machine
4. Spin up the containers on the remote machine

   docker-compose up -d
   

   To stop an already running version and remove the images used for that use

   docker-compose down --rmi all
   

The deploy script will take care of all those steps for you. Make sure to update the details of the deployment machine and that you have ssh access to it.

If they should be deployed separately split the docker-compose in two (one for each service) and transfer them to the desired machines.