Project Management

This repo provides boilerplate project code to help jumpstart new projects. To date, we have focused on setting up Docker images for projects to help speed up their onboarding process.

Folders

Each folder contains relevant Dockerfiles and code for setting up that particular tool or language within a Docker container. See each folder for more detailed information

Docker Compose

Since many projects may need more than one software, it is often useful to run the containers using a single docker-compose file. Even if you need only one container, you can still use docker-compose. As such, you should usually try running the images using docker-compose as your first approach.

This root folder contains a single docker-compose.yml file that is configured for the structure of this project management repository. To get started using this file, you can copy the file into your project and delete the sections that you don't need. Many of the sections of compose, however, expect this folder structure and expect to find the Dockerfiles that are located in each of the subfolders. When this is the case, look for the 'context' setting, which specifies the subfolder where docker-compose will look for these files. Copy any appropriate Dockerfile into a subfolder of your project, and then edit the context to match the name.

For example, if you wanted to use the Python Docker image, and wanted to store your Python code in a folder of your project called src, you would:

• Copy the Dockerfile and environment.yml from the /python folder into your /src folder

• Edit the relevant portion of docker-compose.yml to look like this:

  sandbox:
      build:
          context: src #used to be python
  

To run the docker-compose file and build/start all the containers listed in it, use:

• docker-compose up -d to start in detached mode (containers run in the background)
• docker-compose stop to end the processes

Notes on Docker

Please make sure to grab the most recent Docker Communitiy Edition for your operating system*.

Linux/Mac

To get started download Docker

Windows

Please use docker toolbox

Logging into Docker Hub

If you don't already have an account you can register for one at Docker Hub.

To login run: docker login and follow the prompts.

Publishing Images

Once you have built the image you would like to publish, run docker publish codefordc2/{youimage}

Docker Toolbox

Windows Home edition does not support docker natively, so VirtualBox is used to create an environment for the docker daemon. Here

Note: The other READMEs will always refer to the IP address to use as localhost or 127.0.0.1. When using docker-machine you will need to use the IP address of the virtual machine. To do so run docker-machine ip

Other useful commands

• docker ps lists all running containers
• docker-compose up starts all containers in the docker-compose.yml file. Add -d to run in detached mode (get your command prompt back)
• docker-compose stop stops running containers
• docker-compose down then docker-compose build rebuilds containers from scratch (useful if one gets corrupted, if you change settings such as in the Python environment.yml file)