AWS Builder Center
Kubernetes Package Management with Helm on Amazon EKS

Kubernetes Package Management with Helm on Amazon EKS

Deploy, upgrade, rollback - managing applications the right way

Head of DevOps
Managing Kubernetes applications with raw YAML files works — until it does not. When you have five YAML files per application, multiple environments, and a team making changes, things break quickly. Values get hardcoded, files go out of sync, and rollbacks mean manually reverting each file one by one.
Helm solves all of this. It is the package manager for Kubernetes — the same way  npm  manages Node packages or  apt  manages Ubuntu packages, Helm manages Kubernetes applications as versioned, configurable packages called charts.
In this demo, I created a custom Helm chart for Nginx, deployed it on EKS, upgraded it with new configuration, and rolled it back — all through clean, repeatable Helm commands.



What Helm Provides

Without Helm:
kubectl apply -f deployment.yaml
kubectl apply -f service.yaml
kubectl apply -f hpa.yaml
kubectl apply -f configmap.yaml
(every file managed separately)

With Helm:
helm install nginx-release ./nginx-chart
(everything deployed in one command)

Upgrade:
helm upgrade nginx-release ./nginx-chart --set replicaCount=4

Rollback:
helm rollback nginx-release 1

Helm Chart Structure

Running  helm create nginx-chart  generates a standard chart skeleton:
nginx-chart/
├── Chart.yaml ← chart metadata (name, version, description)
├── values.yaml ← default configuration values
└── templates/
├── deployment.yaml ← uses {{ .Values.x }} for dynamic config
├── service.yaml
├── hpa.yaml
└── _helpers.tpl ← reusable template functions
The power is in the templates — they reference values from  values.yaml  using Go template syntax. This means the same chart can deploy to dev with 1 replica and production with 10, just by changing a single value.



Chart.yaml

apiVersion: v2
name: nginx-chart
description: AIOps Demo Nginx Helm Chart
type: application
version: 1.0.0 # chart version — bump on chart changes
appVersion: "1.25.0" # application version being packaged
values.yaml — The Configuration Layer
replicaCount: 2

image:
repository: nginx
tag: "1.25-alpine"
pullPolicy: IfNotPresent

service:
type: LoadBalancer
port: 80
annotations:
service.beta.kubernetes.io/aws-load-balancer-type: "external"
service.beta.kubernetes.io/aws-load-balancer-scheme: "internet-facing"

resources:
requests:
cpu: "100m"
memory: "128Mi"
limits:
cpu: "200m"
memory: "256Mi"

autoscaling:
enabled: true
minReplicas: 2
maxReplicas: 8
targetCPUUtilizationPercentage: 30
Every value here can be overridden at install or upgrade time — no YAML editing required.



Template Example — Deployment

apiVersion: apps/v1
kind: Deployment
metadata:
name: {{ include "nginx-chart.fullname" . }}
namespace: {{ .Values.namespace }}
spec:
{{- if not .Values.autoscaling.enabled }}
replicas: {{ .Values.replicaCount }}
{{- end }}
template:
spec:
containers:
- name: {{ .Chart.Name }}
image: "{{ .Values.image.repository }}:{{ .Values.image.tag }}"
resources:
{{- toYaml .Values.resources | nindent 12 }}
 {{ .Values.image.tag }}  pulls from  values.yaml . Override it at deploy time and the template renders with the new value — no file editing needed.




Step 1 — Lint Before Deploy

helm lint nginx-chart/
# ==> Linting nginx-chart/
# 1 chart(s) linted, 0 chart(s) failed ✅
Always lint before deploying — catches template errors before they reach the cluster.
Common lint issue encountered:  helm create  generates templates for  serviceaccount.yaml ,  ingress.yaml , and  httproute.yaml  by default, but these reference  values.serviceAccount ,  values.ingress , and  values.httpRoute  which do not exist in the custom  values.yaml . This causes a nil pointer error during lint.
Fix: delete the unused templates:
rm nginx-chart/templates/serviceaccount.yaml
rm nginx-chart/templates/ingress.yaml
rm nginx-chart/templates/httproute.yaml
Step 2 — Install
kubectl create namespace helm-demo

helm install nginx-release nginx-chart/ \
--namespace helm-demo

# Verify
helm list -n helm-demo
# NAME NAMESPACE REVISION STATUS CHART
# nginx-release helm-demo 1 deployed nginx-chart-1.0.0 ✅

kubectl get all -n helm-demo
# Pods, Service, HPA all running ✅
Step 3 — Upgrade (Override Values)
# Bump replicas to 4 — no YAML editing
helm upgrade nginx-release nginx-chart/ \
--namespace helm-demo \
--set replicaCount=4

# Upgrade to newer Nginx image
helm upgrade nginx-release nginx-chart/ \
--namespace helm-demo \
--set image.tag=1.26-alpine
Each upgrade creates a new revision tracked by Helm automatically.




Step 4 — History

helm history nginx-release -n helm-demo

# REVISION STATUS CHART DESCRIPTION
# 1 superseded nginx-chart-1.0.0 Install complete
# 2 superseded nginx-chart-1.0.0 Upgrade complete
# 3 deployed nginx-chart-1.0.0 Upgrade complete
Every change is recorded. This is what makes rollbacks reliable.




Step 5 — Rollback

# Roll back to revision 1
helm rollback nginx-release 1 -n helm-demo

# Verify
helm history nginx-release -n helm-demo
# REVISION STATUS DESCRIPTION
# 1 superseded Install complete
# 2 superseded Upgrade complete
# 3 superseded Upgrade complete
# 4 deployed Rollback to 1 ✅
One command. Previous state restored. No manual YAML reverting needed.



Step 6 — Uninstall

helm uninstall nginx-release -n helm-demo
# All Kubernetes resources removed ✅
# Helm release record removed ✅
Helm vs Raw kubectl

Key Lessons

Always lint before install.  helm lint  catches template errors locally before they reach your cluster. A failed lint is far less painful than a failed production deployment.
 helm create  generates more than you need. The default skeleton includes ServiceAccount, Ingress, and HTTPRoute templates that reference values not present in the default  values.yaml . Delete unused templates or add the corresponding values — otherwise lint will fail with nil pointer errors.
Helm tracks every change as a revision. Every install and upgrade is versioned. This is what makes  helm rollback  reliable — you are not guessing what the previous state was, Helm knows exactly.
 --set  overrides are powerful but  values.yaml  files are better for teams. For quick demos,  --set image.tag=1.26  works well. In production, maintain separate  values-dev.yaml  and  values-prod.yaml  files and deploy with  helm install -f values-prod.yaml  — this keeps configuration in version control.



Full Source Code

All Helm chart files are available on GitHub:
Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.
Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article