Install k8s on Raspberry pi 3
===
2019/4/27

# Introduction

PC: Master
Raspberry: Worker

## Prepaer Tools
* flash tool [Etcher](https://www.balena.io/etcher/)
* Image [Raspbian Stretch Lite](https://www.raspberrypi.org/downloads/raspbian/)
* Install k8s script [alexellis/k8s-on-raspbian](https://raw.githubusercontent.com/alexellis/k8s-on-raspbian/master/script/prep.sh)
---

# build k8s cluster step

## Flash image

The flash step that using **etcher** :

1. Download **Etcher** and install it.
2. Connect an SD card reader with the SD card inside.
3. Open Etcher and select from your hard drive the Raspberry Pi `.img` or `.zip` file you wish to write to the SD card.
4. Select the SD card you wish to write your image to.
5. Review your selections and click 'Flash!' to begin writing data to the SD card.

Default Account
User: `pi`
password: `raspberry`

> default enabled `ssh`
> >write ssh file to boot volume
> >ex: touch /Volumes/boot/ssh

ref:
* [Installation Guide](https://www.raspberrypi.org/documentation/installation/installing-images/README.md)

## Configure WiFi

setup `ssh`, `language` and `WiFi` by `raspi-config`

`sudo raspi-config`

### enable SSH
1. 5 Interfacing Options
2. P2 SSH
3. YES
4. reboot

### setting language
1. 4 Localisation Options
2. I1 Change Locale
3. select `[*] zh_TW.UTF-8 UTF-8`
4. NONE

### setting wifi
1. 2 Network Options
2. N2 Wi-fi
3. input your SSID
4. input your password

## Install docker, kubeadm

curl https://raw.githubusercontent.com/alexellis/k8s-on-raspbian/master/script/prep.sh > ~/install_script.sh
sudo bash ~/install_script.sh
sudo reboot

ref:
* [alexellis/k8s-on-raspbian](https://github.com/alexellis/k8s-on-raspbian)

## Create k8s cluster in Master PC

* network using Weave(or flannel)

1. `sudo kubeadm init`

Your Kubernetes master has initialized successfully!

To start using your cluster, you need to run the following as a regular user:

  mkdir -p $HOME/.kube
  sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
  sudo chown $(id -u):$(id -g) $HOME/.kube/config

You should now deploy a pod network to the cluster.
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:
  /docs/concepts/cluster-administration/addons/

You can now join any number of machines by running the following on each node
as root:

  kubeadm join <master-ip>:<master-port> --token <token> --discovery-token-ca-cert-hash sha256:<hash>

2. `kubectl apply -f "https://cloud.weave.works/k8s/net?k8s-version=$(kubectl version | base64 | tr -d '\n')"`

P.S. this is my using script

sudo swapoff -a && sudo kubeadm init --pod-network-cidr=192.168.0.0/16 --apiserver-cert-extra-sans=sian-home.ddns.net

ref:
* [create-cluster-kubeadm](https://kubernetes.io/docs/setup/independent/create-cluster-kubeadm/)

## Join Raspberry Pi to Closter

1. `kubeadm join <master-ip>:<master-port> --token <token> --discovery-token-ca-cert-hash sha256:<hash>`

## Completed

* label rapsberry to worker
`kubectl label node raspberrypi node-role.kubernetes.io/worker=worker`

* take off `taint`
`kubectl taint nodes --all node-role.kubernetes.io/master-`

* check nodes are `ready` state
`kubectl get nodes -o wide`

NAME          STATUS   ROLES    AGE   VERSION   INTERNAL-IP      EXTERNAL-IP   OS-IMAGE                         KERNEL-VERSION      CONTAINER-RUNTIME
raspberrypi   Ready    worker   9h    v1.14.1   192.168.1.179    <none>        Raspbian GNU/Linux 9 (stretch)   4.14.98-v7+         docker://18.9.0
sian          Ready    master   9h    v1.14.1   36.236.142.127   <none>        Ubuntu 16.04.6 LTS               4.15.0-47-generic   docker://18.6.2

* check pods are `Running`
`kubectl get pods --all-namespaces`

NAMESPACE     NAME                           READY   STATUS    RESTARTS   AGE   IP               NODE          NOMINATED NODE   READINESS GATES
kube-system   coredns-fb8b8dccf-9nxf9        1/1     Running   0          9h    10.244.0.3       sian          <none>           <none>
kube-system   coredns-fb8b8dccf-dlc7v        1/1     Running   0          9h    10.244.0.2       sian          <none>           <none>
kube-system   etcd-sian                      1/1     Running   0          9h    36.236.142.127   sian          <none>           <none>
kube-system   kube-apiserver-sian            1/1     Running   0          9h    36.236.142.127   sian          <none>           <none>
kube-system   kube-controller-manager-sian   1/1     Running   0          9h    36.236.142.127   sian          <none>           <none>
kube-system   kube-flannel-ds-amd64-6kgjq    1/1     Running   0          9h    36.236.142.127   sian          <none>           <none>
kube-system   kube-flannel-ds-arm-jzlz5      1/1     Running   1          9h    192.168.1.179    raspberrypi   <none>           <none>
kube-system   kube-proxy-lbmwn               1/1     Running   1          9h    192.168.1.179    raspberrypi   <none>           <none>
kube-system   kube-proxy-pgz42               1/1     Running   0          9h    36.236.142.127   sian          <none>           <none>
kube-system   kube-scheduler-sian            1/1     Running   0          9h    36.236.142.127   sian          <none>           <none>

# TBD
* [wireguard VPN](https://www.wireguard.com/quickstart/)

# Troubleshooting

## coredns crash

Q: Check nameserver is `127.0.0.1` in `/etc/resolv.conf`
A: `vi /etc/resolv.conf`

# Dynamic resolv.conf(5) file for glibc resolver(3) generated by resolvconf(8)
#     DO NOT EDIT THIS FILE BY HAND -- YOUR CHANGES WILL BE OVERWRITTEN
nameserver 8.8.8.8

###### tags: `Raspberry Pi` `k8s` `IoT`