{"id":60746,"date":"2024-03-14T13:58:10","date_gmt":"2024-03-14T08:28:10","guid":{"rendered":"https:\/\/www.tothenew.com\/blog\/?p=60746"},"modified":"2024-03-19T14:08:25","modified_gmt":"2024-03-19T08:38:25","slug":"k8s-cluster-setup-using-kubeadm-with-bare-minimum-configurations","status":"publish","type":"post","link":"https:\/\/www.tothenew.com\/blog\/k8s-cluster-setup-using-kubeadm-with-bare-minimum-configurations\/","title":{"rendered":"K8s Cluster Setup using Kubeadm with Bare-minimum Configurations"},"content":{"rendered":"

Introduction<\/b><\/h2>\n

Kubernetes is a powerful tool for managing containerized applications. It helps us to automate <\/span>the deployment, scaling, and management of our applications in containers, which are like <\/span>lightweight, standalone packages for software.<\/span><\/p>\n

In the market, multiple tools are available to set up Kubernetes clusters locally (on our host <\/span>machines), like <\/span>Minikube, Microk8s, Kind (Kubernetes in Docker), KubeSpray, <\/b>etc. <\/span>One of them is <\/span>Kubeadm, <\/b>and we will be going to learn how to set up a <\/span>Kubeadm<\/b> cluster on an AWS <\/span>EC2 machine.<\/span><\/p>\n

Kubernetes Architecture<\/b><\/h2>\n

\"\"<\/p>\n

Control Plane Components:<\/span><\/h3>\n
    \n
  • Etcd: <\/b>Cluster Data Store\u200a\u2014\u200aKubernetes uses \u201cetcd.\u201d This is a strong, consistent, and highly-available key value store that Kubernetes uses for persistent storage of all API objects.\u00a0<\/span><\/li>\n
  • API Server<\/b>: The API server exposes a restful Kubernetes API and consumes JSON manifest files.<\/span><\/li>\n
  • Controller Manager<\/b>: Known as the \u201ckube-controller manager,\u201d this runs all the controllers that handle routine tasks in the cluster.\u00a0<\/span><\/li>\n
  • Scheduler<\/b>: The scheduler watches for newly-created pods (groups of one or more containers) and assigns them to nodes.<\/span><\/li>\n
  • Cloud Controller Manager: <\/b>If we are using Kubernetes as a cloud provider service like EKS (AWS\u2019s Kubernetes service), then the whole master node is managed by the AWS in the backend using this component<\/span><\/li>\n<\/ul>\n

    Slave Node Components:<\/span><\/h3>\n
      \n
    • Kubelet: <\/b>An agent that runs on each node and makes sure that containers are running in a pod<\/span><\/li>\n
    • Kube Proxy<\/b>: <\/span>An agent that runs on each node and makes sure that containers are running in a pod<\/span><\/li>\n<\/ul>\n

      About Kubeadm<\/b><\/h2>\n
        \n
      • A tool to create a Kubernetes cluster<\/span><\/li>\n
      • Different from <\/span>Microk8s, Minikube, Kind, <\/b>etc., Kubeadm sets<\/b>\u00a0up the whole cluster from scratch. I.e., master node and slave nodes, respectively, with each component like <\/span>etcd<\/span>, <\/span>scheduler<\/span>, <\/span>controller manager<\/span>, <\/span>api-server<\/span>, etc. However, Microk8s, Minikube, etc.,<\/span>\u00a0are mainly used to test our deployments and pods.<\/span><\/li>\n
      • Kubeadm<\/b> is versatile, i.e., it allows us to create clusters on our own hardware or cloud servers.<\/span><\/li>\n
      • We have more control over the cluster’s configuration, which is useful for custom setups.<\/span><\/li>\n<\/ul>\n

        Benefits of Kubeadm<\/b><\/h2>\n
          \n
        • Kubeadm<\/b> is more versatile and flexible, suitable for various scenarios, including production.<\/span><\/li>\n
        • MicroK8s<\/b> and <\/span>Minikube<\/b> are designed for simplicity and are excellent for beginners or local development.<\/span><\/li>\n
        • Kind<\/b> is great for lightweight, multi-node cluster simulation during development and testing.<\/span><\/li>\n
        • The choice depends on our needs i.e., if we want production-level control, then kubeadm is best suited. If we want a quick and easy setup, <\/span>MicroK8s<\/b> or <\/span>Minikube<\/b> might be better.<\/span><\/li>\n
        • Kubeadm<\/b> offers more customization and is best for advanced use cases, while <\/span>MicroK8s<\/b>, <\/span>Minikube<\/b>, and <\/span>Kind<\/b> are simpler tools tailored for specific purposes, like quick local setups or lightweight development and testing environments.<\/span><\/li>\n<\/ul>\n

          Prerequisites<\/b><\/h2>\n
            \n
          • Two EC2 instances (1 for master node setup and 1 for slave node setup)<\/span><\/li>\n
          • Security group rules for master (control-plane) node and slave node:<\/span><\/li>\n<\/ul>\n


            \n\"\"
            \n<\/span><\/p>\n

            Setup<\/b><\/h2>\n
              \n
            • Setting up the master node by running the following script:<\/span><\/li>\n<\/ul>\n
              #! \/bin\/bash\r\n\r\nset -eo pipefail\r\n\r\n# disable swap\u00a0\r\n\r\nsudo swapoff -a\r\n\r\n# keeps the swaf off during reboot\r\n\r\nsudo sed -i '\/ swap \/ s\/^\\(.*\\)$\/#\\1\/g' \/etc\/fstab\r\n\r\n# Updating the system\r\n\r\nsudo apt-get update -y\r\n\r\nsudo apt-get install -y \\\r\n\r\n\u00a0\u00a0\u00a0\u00a0apt-transport-https \\\r\n\r\n\u00a0\u00a0\u00a0\u00a0ca-certificates \\\r\n\r\n\u00a0\u00a0\u00a0\u00a0curl \\\r\n\r\n\u00a0\u00a0\u00a0\u00a0gnupg \\\r\n\r\n\u00a0\u00a0\u00a0\u00a0lsb-release\r\n\r\n# Process to install Docker\r\n\r\ncurl -fsSL https:\/\/download.docker.com\/linux\/ubuntu\/gpg | sudo gpg --dearmor -o \/usr\/share\/keyrings\/docker-archive-keyring.gpg\r\n\r\necho \\\r\n\r\n\u00a0\u00a0\"deb [arch=amd64 signed-by=\/usr\/share\/keyrings\/docker-archive-keyring.gpg] https:\/\/download.docker.com\/linux\/ubuntu \\\r\n\r\n\u00a0\u00a0$(lsb_release -cs) stable\" | sudo tee \/etc\/apt\/sources.list.d\/docker.list > \/dev\/null\r\n\r\nsudo apt-get update -y\r\n\r\nsudo apt-get install docker-ce docker-ce-cli containerd.io -y\r\n\r\n# Setting CGroup to systemd for Docker\r\n\r\ncat <<EOF | sudo tee \/etc\/docker\/daemon.json\r\n\r\n{\r\n\r\n\u00a0\u00a0\"exec-opts\": [\"native.cgroupdriver=systemd\"],\r\n\r\n\u00a0\u00a0\"log-driver\": \"json-file\",\r\n\r\n\u00a0\u00a0\"log-opts\": {\r\n\r\n\u00a0\u00a0\u00a0\u00a0\"max-size\": \"100m\"\r\n\r\n\u00a0\u00a0},\r\n\r\n\u00a0\u00a0\"storage-driver\": \"overlay2\"\r\n\r\n}\r\n\r\nEOF\r\n\r\nsudo systemctl enable docker\r\n\r\nsudo systemctl daemon-reload\r\n\r\nsudo systemctl restart docker\r\n\r\necho \"Docker Runtime Configured Successfully\"\r\n\r\n# Installing net-tools, curl, kubernetes certs to install kubectl, kubeadm, and kubelet\r\n\r\nsudo apt-get update\r\n\r\nsudo apt-get install -y apt-transport-https ca-certificates curl\r\n\r\nsudo curl -fsSLo \/usr\/share\/keyrings\/kubernetes-archive-keyring.gpg https:\/\/packages.cloud.google.com\/apt\/doc\/apt-key.gpg\r\n\r\necho \"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\r\n\r\nsudo apt-get update -y\r\n\r\nsudo apt-get install -y kubelet kubeadm kubectl\r\n\r\n# Pausing the updates for kubelet, kubeadm, and kubectl\r\n\r\nsudo apt-mark hold kubelet kubeadm kubectl\r\n\r\n# Initialize the cluster\r\n\r\nsudo kubeadm init\r\n\r\nmkdir -p $HOME\/.kube\r\n\r\nsudo cp -i \/etc\/kubernetes\/admin.conf $HOME\/.kube\/config\r\n\r\nsudo chown $(id -u):$(id -g) $HOME\/.kube\/config\r\n\r\n# Creating pod network (without setting this network, the deployments and pods will not come up....they will always be in not-ready state)\r\n\r\ncurl https:\/\/docs.projectcalico.org\/manifests\/calico.yaml -O\r\n\r\nkubectl apply -f calico.yaml<\/pre>\n
              The above script does the following:<\/span><\/p>\n
                \n
              • \n
                  \n
                • sudo swapoff -a<\/b>: Disables the system’s swap space. Kubernetes typically requires a swap to be disabled.<\/span><\/li>\n
                • sudo sed -i ‘\/ swap \/ s\/^\\(.*\\)$\/#\\1\/g’ \/etc\/fstab<\/b>: This line comments out the swap entry in the \/etc\/fstab file, ensuring that swap remains disabled even after a system reboot.<\/span><\/li>\n
                • Updating the system and then installing Docker<\/span><\/li>\n
                • Configuring the Docker to use <\/span>systemd <\/b>as the <\/span>cgroup<\/b> driver and setting up some logging and storage options<\/span><\/li>\n
                • Installing additional packages required for Kubernetes installation, including curl and certificates, and then importing GPG keys<\/span><\/li>\n
                • Now, the most important part of installing the Kubernetes components: <\/span>kubelet <\/b>(responsible for managing containers that are part of the pod)<\/span>, kubeadm, <\/b>and <\/span>kubectl <\/b>(cli to interact with Kube’s api-server)<\/span><\/li>\n
                • sudo apt-mark hold kubelet kubeadm kubectl: <\/b>Pausing the updates for these components<\/span><\/li>\n
                • Initializing the Kubernetes cluster by running <\/span>sudo kubeadm init<\/b>. This command sets up the control plane on the current node.<\/span>\n
                    \n
                  • The components of the control plane are the <\/span>scheduler, api-server, etcd, and the control manager.<\/b><\/li>\n<\/ul>\n<\/li>\n
                  • Setting up the Kubernetes configuration for the user by creating a ~\/.kube\/config file.<\/span>\n
                      \n
                    • This file is responsible for accessing the Kubernetes cluster<\/span><\/li>\n<\/ul>\n<\/li>\n
                    • Now, we are setting up a <\/span>calico<\/b> pod network.<\/b>\n
                        \n
                      • Calico is a commonly used network plugin for Kubernetes that allows pods to communicate with each other<\/span><\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n
                        We are all set now. To check whether our cluster is running or not, just ssh inside the master node and run the following Kubernetes commands:
                        \n<\/span><\/span><\/p>\n
                        kubectl get all; kubectl get all -n kube-system; kubectl get nodes\u2026\u2026etc<\/em><\/p>\n
                        \"\"<\/p>\n
                        NOTE: <\/b>By default, the cluster is accessible via the <\/span>master node <\/b>only because in <\/span>$HOME\/.kube\/config<\/span> this file, there will be a private IP listed for the cluster, and to access this cluster from our local machine, we need to add public IP here (not secure). <\/span>The more secure option to access the cluster is using a VPN.<\/span><\/p>\n
                        Now, it’s time to join the slave nodes <\/b>in our cluster. To join slave nodes, run the below script in the slave nodes:<\/p>\n
                        #! \/bin\/bash\r\n\r\nset -eo pipefail\r\n\r\n# disable swap\r\n\r\nsudo swapoff -a\r\n\r\n# keeps the swaf off during reboot\r\n\r\nsudo sed -i '\/ swap \/ s\/^\\(.*\\)$\/#\\1\/g' \/etc\/fstab\r\n\r\nsudo apt-get update -y\r\n\r\nsudo apt-get install -y \\\r\n\r\n\u00a0\u00a0\u00a0\u00a0apt-transport-https \\\r\n\r\n\u00a0\u00a0\u00a0\u00a0ca-certificates \\\r\n\r\n\u00a0\u00a0\u00a0\u00a0curl \\\r\n\r\n\u00a0\u00a0\u00a0\u00a0gnupg \\\r\n\r\n\u00a0\u00a0\u00a0\u00a0lsb-release\r\n\r\ncurl -fsSL https:\/\/download.docker.com\/linux\/ubuntu\/gpg | sudo gpg --dearmor -o \/usr\/share\/keyrings\/docker-archive-keyring.gpg\r\n\r\necho \\\r\n\r\n\u00a0\u00a0\"deb [arch=amd64 signed-by=\/usr\/share\/keyrings\/docker-archive-keyring.gpg] https:\/\/download.docker.com\/linux\/ubuntu \\\r\n\r\n\u00a0\u00a0$(lsb_release -cs) stable\" | sudo tee \/etc\/apt\/sources.list.d\/docker.list > \/dev\/null\r\n\r\nsudo apt-get update -y\r\n\r\nsudo apt-get install docker-ce docker-ce-cli containerd.io -y\r\n\r\n# Setting up CGroup to systemd in docker runtime\r\n\r\ncat <<EOF | sudo tee \/etc\/docker\/daemon.json\r\n\r\n{\r\n\r\n\u00a0\u00a0\"exec-opts\": [\"native.cgroupdriver=systemd\"],\r\n\r\n\u00a0\u00a0\"log-driver\": \"json-file\",\r\n\r\n\u00a0\u00a0\"log-opts\": {\r\n\r\n\u00a0\u00a0\u00a0\u00a0\"max-size\": \"100m\"\r\n\r\n\u00a0\u00a0},\r\n\r\n\u00a0\u00a0\"storage-driver\": \"overlay2\"\r\n\r\n}\r\n\r\nEOF\r\n\r\nsudo systemctl enable docker\r\n\r\nsudo systemctl daemon-reload\r\n\r\nsudo systemctl restart docker\r\n\r\nsudo apt-get update\r\n\r\nsudo apt-get install -y apt-transport-https ca-certificates curl\r\n\r\nsudo curl -fsSLo \/usr\/share\/keyrings\/kubernetes-archive-keyring.gpg https:\/\/packages.cloud.google.com\/apt\/doc\/apt-key.gpg\r\n\r\necho \"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\r\n\r\nsudo apt-get update -y\r\n\r\nsudo apt-get install -y kubelet kubeadm kubectl\r\n\r\nsudo apt-mark hold kubelet kubeadm kubectl<\/pre>\n
                        The above script installs and configures the same things that are in the master node except for <\/span>initializing the cluster.<\/b><\/p>\n
                          \n
                        • <\/b>Once we have installed all the required components via the above script, we need to ssh into the master node to generate the Token <\/b>and Hash, <\/b>which the slave nodes will use to join the cluster. The below command will generate the token and hash:<\/li>\n<\/ul>\n
                          sudo kubeadm token create --print-join-command<\/pre>\n
                            \n
                          • Once we get the command to join the slave nodes to the cluster, we ssh into the slave nodes and run that command:<\/span><\/li>\n<\/ul>\n
                            sudo kubeadm join <master node IP>:6443 --token <token value> --discovery-token-ca-cert-hash sha256:<hash value><\/pre>\n
                              \n
                            • Once this is successful, you can ssh in the master node and check the status of the attached node using the below command:<\/span><\/span><\/li>\n<\/ul>\n
                              Kubectl get nodes<\/p>\n
                              \"\"<\/span><\/span><\/p>\n
                                \n
                              • Now, we have a running Kubernetes cluster with master and slave nodes.<\/span><\/li>\n<\/ul>\n
                                Conclusion<\/b><\/h2>\n
                                So, in this blog, we have learned about the basics of Kubernetes, which includes its architecture and comp<\/span>onents, as well as the use of each element. We have also checked multiple <\/span>approaches for creating Kubernetes clusters using different tools and how to use <\/span>Kubeadm <\/b>to <\/span>set up a Kubernetes cluster from scratch, including master and slave nodes.<\/span><\/p>\n
                                <\/div>","protected":false},"excerpt":{"rendered":"
                                Introduction Kubernetes is a powerful tool for managing containerized applications. It helps us to automate the deployment, scaling, and management of our applications in containers, which are like lightweight, standalone packages for software. In the market, multiple tools are available to set up Kubernetes clusters locally (on our host machines), like Minikube, Microk8s, Kind (Kubernetes […]<\/p>\n","protected":false},"author":1644,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"iawp_total_views":28},"categories":[2348],"tags":[1883,5700,3965],"aioseo_notices":[],"_links":{"self":[{"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/posts\/60746"}],"collection":[{"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/users\/1644"}],"replies":[{"embeddable":true,"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/comments?post=60746"}],"version-history":[{"count":2,"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/posts\/60746\/revisions"}],"predecessor-version":[{"id":60867,"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/posts\/60746\/revisions\/60867"}],"wp:attachment":[{"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/media?parent=60746"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/categories?post=60746"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.tothenew.com\/blog\/wp-json\/wp\/v2\/tags?post=60746"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}