You can use commands like the following in each of the subdirectories.
docker build . -t lumeer/keycloak
docker build . -t lumeer/keycloak-postgres
docker build . -t lumeer/keycloak-ha-postgres
...
By default, Docker containers do not see each other. Instead of linking directly, we can put them on a common network. We then do not need to care about opened ports.
Let's create networks for each server group. keycloak-net for Keycloak server and its database and lumeer-net for the deployed application.
docker network create keycloak-net
docker network create lumeer-net
Keycloak server needs to persist data, so let's start a PostgreSQL database for it first. We will configure database parameters (db. name, user and password)
and we will connect it to the keycloak-net network.
docker run -idt --name postgres -e POSTGRES_USER=keycloak -e POSTGRES_PASSWORD=keycloaksecretpassword -e POSTGRES_DB=keycloak --network keycloak-net postgres
Next we will start Keycloak server connected to the same network so it can access the PostgreSQL database and we'll expose it locally on port 8180 (internally, it runs on 8080 but we do not want to collide with other WildFly instances).
docker run -idt -p 8180:8080 --name keycloak -e POSTGRES_USER=keycloak -e POSTGRES_PASSWORD=keycloaksecretpassword -e POSTGRES_DB=keycloak -e KEYCLOAK_USER=admin -e KEYCLOAK_PASSWORD=admin --network keycloak-net lumeer/keycloak-ha-postgres
To list all running and stopped containers:
docker ps -a
To start/stop a container:
docker run <container id or name>
docker stop <container id or name>
To delete a stopped container:
docker rm <container id or name>
To list and delete a container image:
docker images
docker rmi <image id>
To see what is going on in a running container:
docker exec -it <container id or name> /bin/bash
To attach to the started container. Warning: upon terminating this shell, the container will exit. To detach the tty without exiting the shell, use the escape sequence Ctrl-P + Ctrl-Q.
docker attach <container id or name>