feat(kubernetes): complete checkpoint 30 - scaling deployments

[wendev27]

Summary

Completed Checkpoint 30 by learning how Kubernetes scales Deployments using ReplicaSets.

Implemented and practiced:

• Scaling Deployments using kubectl scale
• Observing Pod creation in real time
• Understanding the relationship between Deployments, ReplicaSets, and Pods
• Verifying Deployment and ReplicaSet status
• Learning Kubernetes' desired state reconciliation model

---

What went wrong

At first, it was unclear which Kubernetes resource was actually responsible for creating new Pods. It was easy to assume the Deployment directly managed Pods, when in reality the ReplicaSet performs that responsibility.

---

What surprised me

I was surprised that scaling only required changing a single number (--replicas=3), and Kubernetes automatically created the additional Pods without any manual container management.

---

How I fixed it

By observing the Deployment, ReplicaSet, and Pods together using:

kubectl get deployments
kubectl get replicasets
kubectl get pods -w

I was able to see the full control flow and understand how each resource works together.

---

What I would do differently

Next time I would inspect the resources in a more systematic order:

1. Deployment
2. ReplicaSet
3. Pods

This makes it much easier to understand Kubernetes' control hierarchy.

---

Lessons learned

• Deployments manage ReplicaSets.
• ReplicaSets manage Pods.
• Scaling changes the desired state, not the Pods directly.
• Kubernetes continuously reconciles the cluster until the desired state is achieved.
• Horizontal scaling is one of Kubernetes' core strengths for building highly available applications.

[wendev27]

before scaling deployment

after scaling delployment