k8s via proxmox (2) - worker node & master node setup (en)

1. Creating k8s-ctrlr & k8s-node

Using the template created earlier, generate the k8s-ctrlr and k8s-node.
Wait a moment after booting until the cloud-init configuration is complete.

2. Setting up static IPs

Set static IPs for k8s-ctrlr and k8s-node.
As always, use netplan.
Depending on your setup, it will likely be similar if you're running a Proxmox server with a typical router and a single computer.
Before editing, make a backup just in case:

sudo cp /etc/netplan/01-netcfg.yaml{,.bak}

Edit the /etc/netplan/50-cloud-init.yaml file as follows:

version: 2
ethernets:
  eth0:
    addresses: [192.168.0.180/24]
    routes:
      - to: default
        via: 192.168.0.1
    nameservers:
      addresses: [192.168.0.120, 1.1.1.1]

In my case, I’m running a local DNS server using pi.hole on 192.168.0.120, so I added it as a nameserver.
Set the static IPs as 192.168.0.180 for k8s-ctrlr and 192.168.0.185 for k8s-node.
Apply the changes with:

sudo netplan apply

3. Installing containerd

Install containerd on both k8s-ctrlr and k8s-node.

sudo apt-get update && sudo apt-get install -y containerd

3.1. Configuring containerd

Save the default configuration with the following command:

sudo mkdir /etc/containerd
containerd config default | sudo tee /etc/containerd/config.toml

Then, edit the /etc/containerd/config.toml file as follows:

[plugins."io.containerd.grpc.v1.cri".containerd.runtimes.runc.options]
            BinaryName = ""
            CriuImagePath = ""
            CriuPath = ""
            CriuWorkPath = ""
            IoGid = 0
            IoUid = 0
            NoNewKeyring = false
            NoPivotRoot = false
            Root = ""
            ShimCgroup = ""
            SystemdCgroup = false

Change SystemdCgroup = false to true.

4. Additional Network Settings

Open the /etc/sysctl.conf file and modify it as follows:

# Uncomment the next line to enable packet forwarding for IPv4
# sysctl -w net.ipv4.ip_forward=1

Find the second line above and remove the # comment.

Add the following to the /etc/modules-load.d/k8s.conf file:

br_netfilter

Reboot to apply the updated network settings.

5. Kubernetes Installation

sudo curl -fsSLo /usr/share/keyrings/kubernetes-archive-keyring.gpg https://packages.cloud.google.com/apt/doc/apt-key.gpg

echo "deb [signed-by=/usr/share/keyrings/kubernetes-archive-keyring.gpg] https://apt.kubernetes.io/ kubernetes-xenial main" | sudo tee /etc/apt/sources.list.d/kubernetes.list

sudo apt-get update
sudo apt-get install -y kubelet kubeadm kubectl
sudo apt-mark hold kubelet kubeadm kubectl

The above commands fetch the key, add the Kubernetes apt repository, and install Kubernetes packages.
Check the official documentation for updates and modify accordingly if needed.

6. Creating a k8s-node Template

Let’s create a k8s-node template to generate as many workers as needed.

The following commands resolve the issue where cloned VMs have the same IP due to duplicate machine IDs.

sudo cloud-init clean
 
sudo truncate -s 0 /etc/machine-id
sudo rm /var/lib/dbus/machine-id
sudo ln -s /etc/machine-id /var/lib/dbus/machine-id

Then, power off the VM and click Convert to Template in the web UI.

Once converted to a template, you can create as many workers as needed.
Here, we will create two workers: k8s-node-1 and k8s-node-2.

8. Modifying VM Specifications

Adjust the specifications of k8s-ctrlr, k8s-node-1, and k8s-node-2.
Power off all VMs and modify their specifications in the web UI.

Depending on the server running Proxmox, the recommended specs are:

• k8s-ctrlr: 2 CPUs, 4 GB RAM
• k8s-node-1, k8s-node-2: 2 CPUs, 2 GB RAM

If the server lacks sufficient resources, set all to 2 CPUs and 2 GB RAM, which are the minimum requirements.
Even with high specs, increasing the number of workers is preferred over allocating more RAM.

Once done, power on all VMs.
For VMs cloned from a template, static IPs may revert to DHCP. Change them back to static IPs if desired.

9. Creating a Cluster

Log in to k8s-ctrlr and create a cluster using kubeadm.

sudo kubeadm init --control-plane-endpoint=<vm-ip> \
--node-name <vm-hostname> --pod-network-cidr=10.244.0.0/16

For example, <vm-ip> could be 192.168.0.180, and <vm-hostname> could be k8s-ctrlr.

Next, run the following commands:

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

Then, copy the join command generated below and execute it on other nodes.

kubeadm join 192.168.0.180:6443 --token ~~~~~~

If the command is unavailable, execute the following on k8s-ctrlr to generate a new one:

kubeadm token create --print-join-command

9.1. Creating a Pod Network

On k8s-ctrlr, execute the following command:

kubectl apply -f \
https://raw.githubusercontent.com/flannel-io/flannel/master/Documentation/kube-flannel.yml

Finally, verify that all nodes are successfully created:

kubectl get nodes

If all nodes show as Ready, the setup is complete:

NAME         STATUS   ROLES                  AGE   VERSION
k8s-ctrlr    Ready    control-plane,master   10m   v1.21.1
k8s-node-1   Ready    <none>                 10m   v1.21.1
k8s-node-2   Ready    <none>                 10m   v1.21.1

The cluster creation process ends here.

Using Kubernetes

We've created a cluster with some effort, but what should we do now? Let's deploy a web server using Kubernetes.

1. Deploying nginx pod

Let's create and deploy a single nginx pod. On k8s-ctrlr, create a pod.yaml file with the following content:

apiVersion: v1
kind: Pod
metadata:
  name: nginx-example
  labels:
    app: nginx
spec:
  containers:
    - name: nginx
      image: linuxserver/nginx
      ports:
        - containerPort: 80
          name: "nginx-http"

Then deploy it using the following command:

kubectl apply -f pod.yaml

Let's verify if the pod was created successfully:

kubectl get pods -o wide

On k8s-node-1, we can confirm that nginx is running on the pod network at 10.244.1.2:80.

NAME            READY   STATUS    RESTARTS   AGE    IP           NODE         NOMINATED NODE   READINESS GATES
nginx-example   1/1     Running   0          3h8m   10.244.1.2   k8s-node-1   <none>

You can verify the connection from k8s-ctrlr, k8s-node-1, and k8s-node-2 using the following command:

curl 10.244.1.2:80

2. Creating nginx service

However, directly accessing pods is not recommended. This is because pods get new IPs when they restart or are deleted. Also, while cluster internal access is possible from any node, external access is not. To solve these issues, we create a service. On k8s-ctrlr, create a service-nodeport.yaml file with the following content:

apiVersion: v1
kind: Service
metadata:
  name: nginx-example
spec:
  type: NodePort
  ports:
    - name: http
      port: 80
      nodePort: 30080
      targetPort: nginx-http
  selector:
    app: nginx

Then deploy it using the following command:

kubectl apply -f service-nodeport.yaml

Let's verify if the service was created successfully:

kubectl get svc

If you see the following output, it was created successfully:

NAME            TYPE        CLUSTER-IP     EXTERNAL-IP   PORT(S)        AGE
kubernetes      ClusterIP   10.96.0.1      <none>        443/TCP        3h19m
nginx-example   NodePort    10.109.7.138   <none>        80:30080/TCP   3h16m

You can verify the connection from k8s-ctrlr, k8s-node-1, and k8s-node-2 using the following command:

curl localhost:30080

Additionally, you can access it externally using the IP of any cluster member:

curl 192.168.0.180:30080
curl 192.168.0.185:30080
curl 192.168.0.190:30080

While on AWS you can use loadbalancer type instead of nodeport type to access through a single address, we used nodeport type here because loadbalancer type is not available locally.

Conclusion

In this post, we learned how to install Kubernetes and deploy nginx. In the next post, we'll explore other Kubernetes features. We'll look at ingress, configmap, secret, volume, deployment, statefulset, etc. Also, if possible, we'll try to set up CI/CD using Jenkins.

References

• youtube/How to Build an Awesome Kubernetes Cluster using Proxmox Virtual Environment
• blog/How to Build an Awesome Kubernetes Cluster using Proxmox Virtual Environment