A Kubernetes cluster is made of control plane nodes and worker nodes. And the nodes are made up of a number of components with specific functionalities 🌊
🧠 Control plane or the master node
These nodes handle and manage the Kubernetes cluster. They don’t run any containers. The main components of a control plane node 👇
👉 API Server
👉 Scheduler
👉 Controller-manager
👉 Storage or etcd
Let’s explain these components one by one.
💠 API Server
It is the only entry point to a Kubernetes cluster that you directly interact with. User commands are sent to the API server, and we receive responses from the API server.
💠 Scheduler
The scheduler chooses which nodes(can be physical servers, VMs, cloud instances, anything) to run the user application on.
💠 Controller-manager
Allows Kubernetes to integrate with cloud services, such as storage and load-balancers.
💠 Storage or etcd
It’s where the state of the cluster and all the applications are stored. Kubernetes uses etcd as its database. Because Kubernetes is a distributed system, it needs a distributed database like etcd.
Worker nodes or just nodes are where user applications run. The main components of worker nodes 👇
👉 Kubelet
👉 Kube-proxy
👉 Container Runtime
Let’s explain these components one by one.
💠 Kubelet
An agent that runs on each worker node of the cluster. It communicates with the control plane reporting on the status of containers and tasks.
💠 Kube-proxy
Responsible for local cluster networking. It ensures each node gets its own unique IP address and implements rules to handle routing and load-balancing of traffic.
💠 Container Runtime
Container runtime is the software for running containers.
That’s all for this thread on components of Kubernetes nodes 🧡
If you enjoyed it, follow @SignozHQ for more such content 👋🤖
• • •
Missing some Tweet in this thread? You can try to
force a refresh