Multi-Environment Kubernetes Deployments with GitOps, Helm & ArgoCD

This is a starter post — replace it with your own write-up. It exists to show the MDX blog pipeline works end-to-end: frontmatter, rendering, and per-post metadata.

The problem

Deploying the same application to multiple environments by hand is error-prone: config drifts, secrets get copied incorrectly, and "it worked in dev" becomes a recurring incident. I wanted a single source of truth where Git is the desired state and the cluster continuously reconciles toward it.

The approach

• Helm to template the manifests once and parameterize per environment.
• Environment-specific values (values-dev.yaml, values-prod.yaml) for ConfigMaps, Secrets, resource limits, and health probes.
• ArgoCD to watch the repo and sync each environment automatically.

# argocd/app-prod.yaml
apiVersion: argoproj.io/v1alpha1
kind: Application
metadata:
  name: webapp-prod
spec:
  project: default
  source:
    repoURL: https://github.com/Saadnadeem07/dockerized-webapp-ci-cd
    path: charts/webapp
    helm:
      valueFiles:
        - values-prod.yaml
  destination:
    server: https://kubernetes.default.svc
    namespace: prod
  syncPolicy:
    automated:
      prune: true
      selfHeal: true

What I learned

1. selfHeal: true turns drift into a non-issue — manual kubectl edits get reverted to match Git.
2. Keeping probes and resource limits in values files made the prod/dev difference explicit and reviewable in a pull request.
3. GitOps makes rollbacks trivial: revert the commit, ArgoCD does the rest.

Want the full walkthrough? Reach out via the contact section.