optikon-api

The Optikon API makes it easy to watch multiple Kubernetes clusters, and to orchestrate Helm charts on multiple clusters. It was originally designed as an Edge Computing application manager, for easy monitoring and deployment across clusters.

Components

The Optikon API wraps two APIs:

1. the Kubernetes Cluster Registry, deployed onto one "central" cluster.
2. the Helm client -- and facilitates Helm chart CRUD operations for multiple Helm tillers, allowing one-touch deployment of a Helm chart to multiple clusters at once.

See the docs/ folder in this repo for an architecture diagram.

Installation

These examples shows how to run the API using mock handlers. See the optikon-vagrant docs for information on deploying the Optikon API in kubernetes, next to a live Cluster Registry API.

Run in Docker

1. Deploy the API:

docker run intelligentedgeadmin/optikon-api:latest --scheme http --port 9000 --mock-base-path ./api/v0/mock/api`

2. You should now be able to curl http://127.0.0.1:9000/v0/clusters and get back a list of clusters:

~> curl http://127.0.0.1:9000/v0/clusters
[{"metadata":{"annotations":{"Health":"Green","Lat":"55.6899699",...

Run from source

1. clone this repo

2. make (from repo root)

3. ./bin/optikon-api --scheme http --port 9000 --mock-base-path ./api/v0/mock/api