bondy_k8s

Still in draft, but functional very soon

This repository provides example for depl0ying a Bondy cluster on several Kubernetes services.

The repo uses Kustomize, a tool that introduces a template-free way to customize application configuration that simplifies the use of off-the-shelf applications. Now, built into kubectl.

Directory Layout

.
├── Makefile
├── README.md
├── aws-eks
│   └── kustomization.yaml
├── bases
│   ├── bondy
│   └── kustomization.yaml
└── gke
    ├── bondy
    └── kustomization.yaml

The repository is organised using Kustomize best practice of layering the configuration.

The bases directory contains the skeleton of a Bondy cluster deployment.

The other directories represent the resources to deploy a Bondy cluster in a given K8s service e.g. Google K8s Engine (GKE).

Using

Requirements

• A cluster up and running
• kubectl installed and configured to access the clsuter

Building

For simplicity we have included a Makefile with the following build targets:

• build-gke - builds the K8s manifest for Google K8s Engine (GKE). It outputs a YAML file named gke.yaml. If a file existed with that name, the target removes it before building.
• build-aws-eks - builds the K8s manifest for AWS Elastice K8s Service (EKS). It outputs a YAML file named aws-eks.yaml. If a file existed with that name, the target removes it before building.

Usage

make build-gke

or

make build-aws-eks

Deploy

• deploy-gke - uses kubectl to apply the gke.yaml manifest
• deploy-aws-eks - uses kubectl to apply the aws-eks.yaml manifest

Usage

K8S_CONTEXT=yourCtxt make deploy-gke

or

K8S_CONTEXT=yourCtxt make deploy-aws-eks

where yourCtxt is your cluster kubectl context

Undeploy

• undeploy-gke - uses kubectl to delete all the resources previously deployed uaing the gke.yaml manifest
• undeploy-aws-eks - uses kubectl to delete all the resources previously deployed uaing the aws-eks.yaml manifest

Usage

K8S_CONTEXT=yourCtxt make undeploy-gke

or

K8S_CONTEXT=yourCtxt make undeploy-aws-eks

where yourCtxt is your cluster kubectl context