Download the docker-compose.yml file in a new folder. Next create a .env file with the following data:
COMPOSE_PROJECT_NAME=clowder
TRAEFIK_HOST=Host:yourmachine.ncsa.illinois.edu;
TRAEFIK_HTTP_PORT=80
TRAEFIK_HTTPS_PORT=443
TRAEFIK_HTTPS_OPTIONS=TLS
TRAEFIK_ACME_ENABLE=true
TRAEFIK_ACME_EMAIL=youremail@ncsa.illinois.edu
TRAEFIK_HTTP_REDIRECT=Redirect.EntryPoint:https
CLOWDER_SSL=true
CLOWDER_ADMINS=youremail@ncsa.illinois.edu
Next create a docker-compose.override.yml file:
version: '3.5'
services:
# point to the PR image (in this case PR-404)
clowder:
image: ghcr.io/clowder-framework/clowder:PR-404
# add any more extractors if you want
# extract preview image
imagepreview:
image: clowder/extractors-image-preview:latest
restart: unless-stopped
networks:
- clowder
depends_on:
rabbitmq:
condition: service_started
environment:
- RABBITMQ_URI=${RABBITMQ_URI:-amqp://guest:guest@rabbitmq/%2F}
# extract image metadata
imagemetadata:
image: clowder/extractors-image-metadata:latest
restart: unless-stopped
networks:
- clowder
depends_on:
rabbitmq:
condition: service_started
environment:
- RABBITMQ_URI=${RABBITMQ_URI:-amqp://guest:guest@rabbitmq/%2F}
# digest
digest:
image: clowder/extractors-digest:latest
restart: unless-stopped
networks:
- clowder
depends_on:
rabbitmq:
condition: service_started
environment:
- RABBITMQ_URI=${RABBITMQ_URI:-amqp://guest:guest@rabbitmq/%2F}
It is best practice to start with a docker-compose pull
to make sure you have all the latest versions of the containers, followed by a docker-compose up -d
. This will start all containers. You should be able to go to https://yourmachine.ncsa.illinois.edu.
If this is the first time running the stack (or if you removed the mongo database), you will need to create the initial user again:
docker run --rm -it \
--network clowder_clowder \
-e "FIRSTNAME=Admin" \
-e "LASTNAME=User" \
-e "ADMIN=true" \
-e "PASSWORD=areallygoodpassword" \
-e "EMAIL_ADDRESS=youremail@ncsa.illinois.edu" \
-e "MONGO_URI=mongodb://mongo:27017/clowder" \
clowder/mongo-init