← Back to list

From Code to Clusters: My Progressive Journey into Kubernetes

In DevOps, there is a distinct “click” moment when you move from running code to orchestrating infrastructure. My journey hasn’t been a…

Isaac Owusu-Ansah · 2026-05-01 06:30 · 1 claps · 1.2 min read
#failing-and-learning #k8s #vpc-endpoint #github
Open on Medium ↗
Wiki topics: FT · Fine-tuning & Adaptation EDU · Education & Learning ☁️ · DevOps & Cloud 🔓 · Open Source 🏃 · Running & Endurance

From Code to Clusters: My Progressive Journey into Kubernetes

In DevOps, there is a distinct “click” moment when you move from running code to orchestrating infrastructure. My journey hasn’t been a straight line it’s been a deliberate transition from local development to a high-availability production environment on Amazon EKS.

The Foundation: Identity and Source Control

It started with the basics: CodeTrack. Before touching a cluster, I had to master the “who” and “where” of engineering. Configuring Git identities and mastering SSH-based workflows weren’t just administrative tasks; they were my first lessons in Access Governance. You cannot manage a cluster if you cannot securely manage your source code.

The Pivot: Learning to Speak “Kubernetes”

Moving into Kubernetes, I realized that manual container management is a recipe for failure. Containers crash, traffic spikes, and IPs change. I transitioned from “telling a server what to do” to “telling a cluster what should be true.”

I learned that stability is an architectural choice. By implementing ReplicaSets, I ensured my application was self-healing. By configuring Readiness and Liveness probes, I taught the cluster how to monitor its own health. I wasn’t just deploying NGINX; I was designing a system that could survive without me.

The Breakthrough: Solving the IAM Puzzle

My most recent project deploying to Amazon EKS presented the “hard thing.” I hit a critical IAM “Access Denied” wall. Instead of pivoting, I chose to fail forward. I performed a deep-dive audit of my IAM principal chain, resolved the credential conflicts, and successfully authorized the provisioning of a multi-AZ cluster.

The Result: Resilience by Design

Today, my NGINX deployment sits behind an AWS Elastic Load Balancer, distributed across multiple nodes. It’s a far cry from my first Git commit.

The Lesson? DevOps isn’t about the tools; it’s about the troubleshooting. I’ve learned that a “Ready” status in the terminal is the reward for rigorous design and a refusal to back down when things get difficult. On to the next challenge.


메타데이터
post_id
f49480c8b1cb
slug
from-code-to-clusters-my-progressive-journey-into-kubernetes-f49480c8b1cb
url
https://medium.com/@oansah.isaac/from-code-to-clusters-my-progressive-journey-into-kubernetes-f49480c8b1cb
canonical_url
https://medium.com/@oansah.isaac/from-code-to-clusters-my-progressive-journey-into-kubernetes-f49480c8b1cb
author_url
https://medium.com/@oansah.isaac
status
ok
fetched_at
2026-06-22 12:55:45