Advanced Kubernetes 1 | CNCF Basics and Kubeadm

1. CNCF Basics

(1) CNCF Trail Map

(2) Topics Overview

kubeadm: installing kubernetes on-prem
K8s Operators: file, block, and object storage
cert-manager: managing ssl
Dex: managing LDAP
Istio: service mesh, load balancing
Calico: networking
Vault: secret store
Openshift: PaaS

2. Environment Configuration

I’m on a M2 Macbook with Parallel as VM. So I use Parallel to start a Ubuntu 22.04 VM. When the VM is up, install the network tools and use ifconfig to find the IP address

$ sudo apt install net-tools
$ ifconfig

Then ssh through the terminal from the macbook outside the VM.

Also, install git with,

$ sudo apt-get update
$ sudo apt install git

3. Build 1-Node Environment with Kubeadm

First, git clone from the libraray,

$ cd /home/parallels/
$ git clone https://github.com/Sadamingh/on-prem-or-cloud-agnostic-kubernetes.git

Then go to the directory by,

$ sudo su -
$ cd /home/parallels/on-prem-or-cloud-agnostic-kubernetes

Install docker, kubelet, kubeadm, kubectl with scripts/install-node.sh,

$ source scripts/install-node.sh

The let’s bring up kubelet service by,

$ swapoff -a
$ sudo systemctl restart kubelet.service

After installation, configure kubeadm through,

$ source scripts/config-kubeadm.sh 

Note that the configuration step is supposed to be run only on the master node. For the other working nodes, we only have to run install-node.sh.

Also note that podSubnet value in config-kubeadm.sh is originally set to be 10.211.0.0/16. Please make sure this value aligns to the IP address.

Finally, let’s create a new user. First we have to go with the root permission. And execute the create-user.sh script.

$ source scripts/create-user.sh

Then for the newly created user ubuntu, we can reset the password by passwd,

$ passwd ubuntu

4. Test

We can test the cluster through the following commands and the following results are expected.

$ kubectl get nodes
NAME                         STATUS   ROLES           AGE     VERSION
ubuntu-linux-22-04-desktop   Ready    control-plane   3h28m   v1.26.2

$ kubectl get pods -A
NAMESPACE     NAME                                                 READY   STATUS    RESTARTS   AGE
kube-system   calico-kube-controllers-57b57c56f-4kb6g              1/1     Running   0          3h28m
kube-system   canal-vpb26                                          2/2     Running   0          3h28m
kube-system   coredns-787d4945fb-6kvh2                             1/1     Running   0          3h28m
kube-system   coredns-787d4945fb-t9cqf                             1/1     Running   0          3h28m
kube-system   etcd-ubuntu-linux-22-04-desktop                      1/1     Running   0          3h28m
kube-system   kube-apiserver-ubuntu-linux-22-04-desktop            1/1     Running   0          3h28m
kube-system   kube-controller-manager-ubuntu-linux-22-04-desktop   1/1     Running   0          3h28m
kube-system   kube-proxy-dcw8p                                     1/1     Running   0          3h28m
kube-system   kube-scheduler-ubuntu-linux-22-04-desktop            1/1     Running   0          3h28m

5. Debug

After rebooting the VM, the connection to kubernetes will be lost. Logs like

$ kubectl get nodes
E0313 10:34:40.260814 2711656 mencache.g:238] couldn't get current server API g roup list: Get "http://localhost: 8080/api?timeout=32s": dial tcp 127.0.0.1:8080:connect: connection refused

In this case, rerun install-node.sh by,

$ source scripts/install-node.sh
$ swapoff -a
$ sudo systemctl restart kubelet.service