Part-80: 🟢Kubernetes – Pods & Multi-Container Pods

When working with Kubernetes, everything starts with Pods. They are the most basic building blocks in a Kubernetes cluster. Let’s break this down step by step.

---

🚀 Kubernetes – Pods

• The main goal of Kubernetes is to deploy applications inside containers on worker nodes.
• But Kubernetes does not deploy containers directly on worker nodes.
• Instead, containers are wrapped inside a special Kubernetes object called a Pod.

👉 Key things to know about Pods:

• A Pod is a single instance of an application.
• A Pod is the smallest object you can create in Kubernetes.
• Pods generally follow a one-to-one relationship with containers.
• To scale up, you create more Pods. To scale down, you delete Pods.
• You cannot run multiple containers of the same type in a Pod.

Example ❌: Running two NGINX containers inside the same Pod is not recommended.

✅ So, think of a Pod as a “wrapper” that holds one containerized application instance.

---

🟢 Kubernetes – Multi-Container Pods

Sometimes, a Pod may need more than one container. This is called a Multi-Container Pod.

👉 Rules for Multi-Container Pods:

• You can have multiple containers in one Pod, but they should not be of the same kind.
• Example: You can’t run two identical NGINX containers in one Pod, but you can run an NGINX + helper container together.

📦 Common Helper (Sidecar) Containers

Multi-Container Pods are often used with sidecar containers, which work alongside the main application container:

• Data Pullers → Download data needed by the main container.
• Data Pushers → Collect logs or data from the main container and push them elsewhere.
• Proxies → Write static data using a helper container, then the main container serves it.

🔗 Communication Inside a Pod

• Containers inside the same Pod share the same network space (they can talk to each other over localhost).
• They also share the same storage volumes, making it easy to exchange files/data.

---

⚠️ Important Note

• Multi-Container Pods are considered a rare use-case.
• In most scenarios, you’ll work with single-container Pods, because that’s the core Kubernetes fundamental.

---

✅ Summary

• Pod → Smallest Kubernetes object, usually runs one container.
• Scaling → Add or remove Pods, not containers inside a Pod.
• Multi-Container Pod → Used only when containers must work closely together (sidecars).
• Communication → Containers inside the same Pod share network and storage.