You can find a dashboard with experiments with Kubernetes (k8s) to observe it's behaviour in real scenarios in this project. You should have a k8s cluster up and running to deploy the tools from this repository.
kube-alive is tested on the following platforms, with Kubernetes version 1.8 and 1.9:
- Minikube (see official guide)
- Microsoft Azure Container Service AKS (see official guide)
- Google Kubernetes Engine GKE (see official guide)
- Bare metal arm32v7-cluster based on Raspberry Pi 3 (see a guide by Scott Hanselman)
- Bare metal arm64v8-cluster on https://www.packet.net/ (Type 2A on location EWR1 - thanks to packet.net for sponsoring hardware to help port kube-alive)
Prerequisites:
- You need to have
kubectl
installed and configured to point at a running cluster that you have admin access to. - If you are running on Google Kubernetes Engine, your default account does not have the privilege to grant more privileges. To circumvent this, you can grant yourself cluster-admin rights:
kubectl create clusterrolebinding cluster-admin-binding \ --clusterrole cluster-admin --user $(gcloud config get-value account)
You can deploy kube-alive to your local cluster with this single command:
curl -sSL https://raw.githubusercontent.com/daniel-kun/kube-alive/master/deploy.sh | bash
This will install the required resources into the kube-alive
Kubernetes namespace.
Behaviours of k8s that can be observed "live":