(Note: All information is from the internet)
A personal note about installing k8s on Raspberry Pi4
![]() |
| A messy setup before got the cluster case.. |
A) On Master Node
1.# curl -
After install the docker, update the line /lib/
ExecStart=/usr/bin/dockerd -H fd:// --containerd=/run/containerd/containerd.sock --exec-opt native.cgroupdriver=systemd
- I use
-
-
#
Further connection has to jump from other 10.10.10.x nodes (via 192.168.1.11 / 10.10.10.11 to 10.10.10.10)
3.
4. Aim to use flannel as CNI, so start with
$
5.
sudo chown $(id -u):$(id -g) $HOME/.kube/config
6. Next keep the message of installation below:
---
You should now deploy a pod network to the cluster.
Run "
https://kubernetes.io/docs/concepts/cluster-administration/addons/
You can now join any number of control-plane nodes by copying certificate authorities
--
--
Then you can join any number of worker nodes by running the following on each as root:
--
---
7.
Since we need to connect to
One of the easier
$
clusterrole.rbac.authorization.k8s.io/flannel created
clusterrolebinding.rbac.authorization.k8s.io/flannel created
Check there is a new network interface create
8. Check the installation
pi@rbpi4-n0
NAME STATUS ROLES AGE VERSION
rbpi4-n0 Ready master 8m34s v1.16.2
pi@rbpi4-n0
9. Setup Worker nodes in section (B)
10. Setup dashboard
$ kube apply -f kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/v2.0.0-beta5/aio/deploy/recommended.yaml
11. Check status
pi@rbpi4-n0
NAMESPACE NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
pi@rbpi4-n0
B) On Worker Node
1) Reset and start again
pi@rbpi4-n1
[
[
[
W1028 18:51:58.108270 24830
[
[
[
[
[
[
[
To avoid complication, advise to reboot the node and start again. Clean the /etc/cabinet files except the manifest
2) Join the cluster as worker node (not control plane node)
$
# root@rbpi4-n1
> --
[
[
[
[
[
[
[
[
This node has joined the cluster:
* The Kubelet was informed of the new secure connection details.
Run '
(Repeat the same for all worker
