PhEMA Workbench Deployment Config

Docker resources to set up a PhEMA Workbench instance.

Prerequisites

You'll need Docker and Docker Compose installed.

Local Deploy

To run the workbench locally, first clone this repo:

git clone https://github.com/PheMA/phema-workbench-deploy.git && cd phema-workbench-deploy

Then spin up the containers with Compose:

PHEMA_WORKBENCH_CERTS_PATH=./certs PHEMA_WORKBENCH_HTPASSWD_PATH=./htpasswd docker compose up

The environment variables provided point to local, testing versions of the .htpasswd file and SSL certificates. None of these should be used in a publicly-facing environment.

💡 You may have to run this a few times until it works, since container dependencies are not currently set up properly. Once Postgres is initialized it should start up first time every time.

Access

Once the containers are all running, the following URLs should be accessible:

https://workbench.local.phema.science:4321/ - The PhEMA Workbench app
https://workbench.local.phema.science:4321/fhir - The FHIR server base URL
https://workbench.local.phema.science:4321/hapi - The HAPI FHIR web overlay

💡 When running CQL, use the PhEMA Workbench CQL Executor server.

Workbench Credentials

For the local instance, the default credentials are:

Login: test
Password: test

Production Deploy

We have the ability to perform a production deploy protected by basic auth (for now). To do this, you need to set the following environment variables before running Compose:

PHEMA_WORKBENCH_HTPASSWD_PATH: The path to the htpasswd file to use.
PHEMA_WORKBENCH_ENV: Set this to prod to use the nginx-prod.conf file, or create a new Nginx config file called nginx-whatever.conf and set the variable to whatever.

For example:

PHEMA_WORKBENCH_HTPASSWD_PATH=/opt/phema/workbench/.htpasswd PHEMA_WORKBENCH_ENV=prod docker-compose up