Requirements

sudo apt install docker.io curl

Add user to docker group.

sudo usermod -aG docker $USER

Remember you have to logout and re-login for changes to take effect

Kubernetes install

git clone https://github.com/Mirantis/kubeadm-dind-cluster.git  
cd kubeadm-dind-cluster  
./fixed/dind-cluster-v1.9.sh up

config.sh for node count and networking
image/kubelet.service has eviction policies. #Haven't tried it yet

Install kubectl

Easiest way to install kubectl in Ubuntu is using snap packages.

snap install kubectl --classic

Api server configuration

docker exec -it kube-master /bin/bash  
apt install vim-tiny  
vim.tiny /etc/kubeadm.conf

Add following to the file

apiServerExtraArgs:
  runtime-config: "api/all=true"
controllerManagerExtraArgs:
  horizontal-pod-autoscaler-use-rest-clients: "true"
  horizontal-pod-autoscaler-sync-period: "10s"
  node-monitor-grace-period: "10s"

Restart apiserver using kubeadm

kubeadm config upload from-file --config /etc/kubeadm.conf  
kubeadm upgrade apply 1.9.1

Metrics-server install

git clone https://github.com/kubernetes-incubator/metrics-server.git  
cd metrics-server  
kubectl create -f deploy/1.8+/

Check metrics-server

kubectl api-versions

Check if following api endpoints exist at the output

metrics.k8s.io/v1beta1  
autoscaling/v2beta1

Check if metrics-server getting any data

kubectl get --raw "/apis/metrics.k8s.io/v1beta1/nodes"  
kubectl get --raw "/apis/metrics.k8s.io/v1beta1/pods"

Monitoring

git clone https://github.com/coreos/prometheus-operator.git  
cd contrib/kube-prometheus/  
hack/cluster-monitoring/deploy

Prometheus UI on node port 30900
Alertmanager UI on node port 30903
Grafana (Dashboard) on node port 30902

Dashboard link

http://localhost:8080/api/v1/namespaces/kube-system/services/kubernetes-dashboard:/proxy

Error: Path is mounted on but it is not a shared or slave mount.

You have to execute this command on the nodes

mount --make-shared PATH

Some useful commands

These are useful for kubectl get
-o wide shows IP address and node name for pods
-n namespace for example -n kube-system for system pods or -n monitoring for prometheus.

kubectl describe resource/RESOURCE_NAME is essential for debugging.

For getting inside of a container docker exec -it CONTAINER_ID /bin/bash or kubectl exec -it POD_NAME /bin/bash
For learning ip address of docker container

docker inspect CONTAINER_ID | grep '"IPAddress"' | head -n 1

You can put these into your .bashrc like

dip() {
  docker inspect $1 | grep '"IPAddress"' | head -n 1
}

dbash() {
  docker exec -it $1 /bin/bash
}

Go install

sudo apt install golang-1.9

Then add these to .profile

export GOPATH="$HOME/workspace/go"
PATH="$PATH:/usr/lib/go-1.9/bin/:$GOPATH:$GOPATH/bin"

Benchmark tool

go get -u github.com/rakyll/hey  
hey -z 5m url

Throwaway docker hub account

Above installs a prometheus with dashboard but for custom metrics re-read below pages

https://github.com/stefanprodan/k8s-prom-hpa
https://github.com/luxas/kubeadm-workshop
https://github.com/DirectXMan12/k8s-prometheus-adapter
https://docs.bitnami.com/kubernetes/how-to/configure-autoscaling-custom-metrics/

ugurcan377 / k8s-resources