Instructions

This is a demonstration of a Multi Docker Compose.

The purpose of this repositoy is ongoing research on "Docker compose" architecture design for easier and elegant control over the services, without duplication.

Create a new project environment aka sample platform

To create a new environment

mkdir platform-new
touch platform-new/.env
ln -s docker-compose.base.yml platform-new/docker-compose.yml
touch platform-new/docker-compose.override.yml

Tree view visualizing of the filesystem

platform-new/
├── data
│ └── backend1
│ └── config
│ └── main.yml
├── docker-compose.override.yml
└── docker-compose.yml -> ../docker-compose.base.yml

In short:

• We link the original docker-compose.base.yml.

• We create new .env file.

  • Each .env can configure the inside/outside ports and other user configurable options.

Each override has the example structure:

version: '3'services:
backend1:
# add content herevolumes:
- ./data/backend3/config:/config

Note:

• Each platform environment will not to be committed to GIT.

• The platform environments will be created on the dev machine or production.

• There can be an infinite number of environments on one machine/node as long as the ports are different in each .env file.

• Having a data directory which can be mounted/binded to particular location inside the target service.

Environment keys:

• PORT_BACKEND1 is the outside port

• PORT_BACKEND1_IN is the inside port

• BACKEND1_PORT is the listen port for the service

Commands

Rebuild

cd platform-dev # sample env
docker-compose build
docker-compose up -d

Test: http://localhost:8003

Status

cd platform-dev # sample env
docker-compose ps

Features (desired)

• Multi environment docker compose

• Building an image instructions

• Controlling inside env vars

• Controlling outside env vars in docker -compose

• Structuring environment variables for easier control and maintaince

• Multi stage Dockerfile builds

Reference

• https://pspdfkit.com/blog/2018/how-to-manage-multiple-system-configurations-using-docker-compose/

• https://github.com/docker/awesome-compose/ (for the Go sample insipiration)

TODOS

Please check the Issues section for a complete list of pending features here.