Docker Samples :: Avatars 🤪

Avatars is a small sample project for Docker demonstration purposes.

It consists of a Python web API backend to generate avatars and a Javascript SPA (single page app) frontend. If you are not a Python or Javascript guru, don't panic!

Usage

To get started, clone this repository locally:

git clone https://github.com/dockersamples/avatars.git

All subsequent commands assume they are run from the root repository directory (i.e. avatars/ folder).

Run make help to see a list of available targets.

Tilt

You'll need to first install Tilt and prerequisites (Docker + development Kubernetes cluster).

Run tilt up or make tilt. All resources will be deployed to the namespace and cluster as determined by the active Kubernetes config context.

Tilt web interface: http://localhost:10350/
Avatars frontend: http://localhost:5735/
Avatars API: curl http://localhost:5734/api/avatar/spec

Docker Compose with file watch

Run make compose.

Avatars frontend: http://localhost:5735/

Compose launches the services and attaches in file watch mode. Run docker compose -p avatars logs -f from a separate terminal to stream logs.

Kubernetes (Docker Desktop)

ℹ️ If you haven't already, open the Docker Desktop dashboard and enable Kubernetes from Settings > Kubernetes > ☑️ Enable Kubernetes and apply.

Run make kubernetes. All resources will be deployed to the avatars namespace in the docker-desktop Kubernetes cluster.

Avatars frontend: http://localhost:5735/

Buildx builds images and Kubernetes creates Deployments and Services. The images are not pushed to a registry, as Docker Desktop's Kubernetes integration runs the images directly from the Docker Engine image store using cri-dockerd (formerly known as dockershim).

Run `kubectl --context=docker-desktop --namespace=avatars logs -l='app.kubernetes.io/part-of=dockersamples_avatars' from a separate terminal to stream logs.

Simple Compose Watch Test

git clone https://github.com/dockersamples/avatars
cd avatars
docker compose up -d

Initiate the Watch

docker compose alpha watch

Generate a graphviz graph from your compose file

docker compose alpha viz

worthwhileindustries/avatars