Commands

Cluster Setup and Initial Status Checks

• Minikube Cluster and kubectl:

  • Minikube cluster is used as the Kubernetes environment.

  • kubectl (command-line tool) is essential for interacting with the cluster.

• Basic Status Commands:

  • kubectl get nodes:

    • Shows cluster nodes.

    • In Minikube, a single master node is used.

  • kubectl get pods:

    • Lists current pods.

    • Returns “No resources” if none exist.

  • kubectl get services:

    • Displays running services.

    • Shows default service if no additional ones are created.

---

Creating Kubernetes Resources

• kubectl create:

  • Used for creating Kubernetes components.

  • Running kubectl create --help shows available resource types.

  • Note: Direct pod creation is not listed.

• Pod Abstraction via Deployment:

  • In Kubernetes, pods are typically not created directly.

  • Instead, use deployments, which manage pods via replica sets.

---

Creating a Deployment (Example: nginx)

• Command Syntax:

  kubectl create deployment nginx-deployment --image=nginx

  • Creates a deployment named nginx-deployment.

  • Downloads the latest nginx image from Docker Hub.

• Deployment Verification:

  • kubectl get deployments: Checks deployment status.

  • kubectl get pods: Shows auto-generated pod with status like ContainerCreating → Running.

• Architecture Layers:

  • Deployment → ReplicaSet → Pod → Container (Image).

  • Pod names reflect hierarchy (deployment name, replica set hash, pod-specific hash).

---

Working with ReplicaSets

• Understanding ReplicaSets:

  • Created automatically when a deployment is created.

  • Manage pod replication.

  • Not typically managed directly by users.

• Viewing ReplicaSets:

  kubectl get replicasets

  • Shows replica set linked to the deployment.

---

Editing Deployments (Example: Set Specific nginx Version)

• Edit Command:

  kubectl edit deployment nginx-deployment

• Changes Made:

  • Scroll to the image field.

  • Update image version (e.g., nginx:1.16).

  • Save changes.

• Observed Behavior:

  • Old pod terminates.

  • New pod is created with updated image.

  • ReplicaSet is updated automatically.

---

Debugging and Logging

• Logs:

  kubectl logs <pod-name>

  • Returns output of container logs inside the pod.

  • Useful for debugging application behavior.

• Example with MongoDB Deployment:

  kubectl create deployment mongodb-deployment --image=mongo
  kubectl logs <mongodb-pod-name>

  • MongoDB logs are displayed once the container starts.

• Describing Pod State:

  kubectl describe pod <pod-name>

  • Provides detailed state changes: image pulled, container created, container started.

---

Interactive Debugging with exec

• Command for Terminal Access:

  kubectl exec -it <pod-name> -- /bin/bash

  • Opens terminal in the container as root user.

  • Enables execution of internal commands.

  • Critical for real-time troubleshooting.

---

Deleting Resources

• Deleting Deployments:

  kubectl delete deployment <deployment-name>

  • Automatically deletes associated pods and replica sets.

• Example:

  • Delete MongoDB deployment:

    kubectl delete deployment mongodb-deployment

  • ReplicaSet and pods associated with MongoDB are removed.

---

Managing Complexity with Configuration Files

• Command-Line Limitation:

  • Command-line creation (e.g., kubectl create deployment) requires all options to be specified inline.

  • Becomes impractical for complex configurations.

• Using YAML Configuration Files:

  • More scalable and manageable method.

  • Define:

    • Resource type

    • Name

    • Image

    • Additional configuration

  • Apply using:

    kubectl apply -f <filename>.yaml

  • f flag specifies the configuration file to execute.