Creating Custom Charts

# Create scaffold
helm create mywebapp

# Generated structure:
mywebapp/
├── .helmignore          # Files to skip when packaging
├── Chart.yaml           # Chart metadata
├── values.yaml          # Default configuration
├── charts/              # Dependencies
└── templates/
    ├── _helpers.tpl     # Named template definitions
    ├── deployment.yaml
    ├── hpa.yaml         # Horizontal Pod Autoscaler
    ├── ingress.yaml
    ├── service.yaml
    ├── serviceaccount.yaml
    ├── NOTES.txt        # Post-install message
    └── tests/
        └── test-connection.yaml

apiVersion: v2
name: mywebapp
description: A production-ready web application chart
type: application          # "application" (installable) or "library" (helpers only)
version: 0.1.0             # Chart version (bump on chart changes)
appVersion: "1.16.0"       # Version of the app being deployed
keywords:
  - web
  - nodejs
maintainers:
  - name: DevOps Team
    email: devops@company.com

replicaCount: 1

image:
  repository: myregistry/mywebapp
  pullPolicy: IfNotPresent
  tag: ""                  # Defaults to appVersion

service:
  type: ClusterIP
  port: 80

ingress:
  enabled: false
  className: ""
  hosts:
    - host: chart-example.local
      paths:
        - path: /
          pathType: ImplementationSpecific

resources:
  requests:
    cpu: 100m
    memory: 128Mi
  limits:
    cpu: 500m
    memory: 256Mi

autoscaling:
  enabled: false
  minReplicas: 1
  maxReplicas: 10
  targetCPUUtilizationPercentage: 80

{{/*
Expand the name of the chart.
*/}}
{{- define "mywebapp.name" -}}
{{- default .Chart.Name .Values.nameOverride | trunc 63 | trimSuffix "-" }}
{{- end }}

{{/*
Common labels
*/}}
{{- define "mywebapp.labels" -}}
helm.sh/chart: {{ include "mywebapp.chart" . }}
{{ include "mywebapp.selectorLabels" . }}
app.kubernetes.io/managed-by: {{ .Release.Service }}
{{- end }}

{{/*
Selector labels
*/}}
{{- define "mywebapp.selectorLabels" -}}
app.kubernetes.io/name: {{ include "mywebapp.name" . }}
app.kubernetes.io/instance: {{ .Release.Name }}
{{- end }}

1. Get the application URL by running these commands:
{{- if .Values.ingress.enabled }}
  http{{ if $.Values.ingress.tls }}s{{ end }}://{{ (index .Values.ingress.hosts 0).host }}
{{- else if contains "NodePort" .Values.service.type }}
  export NODE_PORT=$(kubectl get --namespace {{ .Release.Namespace }} -o jsonpath="{.spec.ports[0].nodePort}" services {{ include "mywebapp.fullname" . }})
  echo http://$NODE_IP:$NODE_PORT
{{- else if contains "ClusterIP" .Values.service.type }}
  kubectl port-forward svc/{{ include "mywebapp.fullname" . }} {{ .Values.service.port }}:{{ .Values.service.port }}
{{- end }}

# Lint for errors
helm lint ./mywebapp
helm lint ./mywebapp -f custom-values.yaml

# Render templates locally (no cluster needed)
helm template myrelease ./mywebapp

# Render specific template
helm template myrelease ./mywebapp -s templates/deployment.yaml

# Dry-run against cluster (validates with K8s API)
helm install myrelease ./mywebapp --dry-run --debug

# Package into .tgz
helm package ./mywebapp
# Output: mywebapp-0.1.0.tgz

# Push to OCI registry (Helm 3.8+)
helm push mywebapp-0.1.0.tgz oci://myregistry.azurecr.io/helm

# Or host a classic chart repository:
# 1. Generate index.yaml
helm repo index . --url https://charts.example.com
# 2. Serve index.yaml + .tgz via web server or GitHub Pages

# Sign a chart package with a GPG key
helm package --sign --key "DevOps Team" --keyring ~/.gnupg/keyring.gpg ./mywebapp
# Creates: mywebapp-0.1.0.tgz + mywebapp-0.1.0.tgz.prov (provenance file)

# Verify a signed chart
helm verify mywebapp-0.1.0.tgz
# "Signed by: DevOps Team" → Trusted

# Install with verification
helm install myapp mywebapp-0.1.0.tgz --verify --keyring ~/.gnupg/keyring.gpg

# Note: OCI registries support cosign for image-style signing:
# cosign sign myregistry.azurecr.io/helm/mywebapp:0.1.0
# cosign verify myregistry.azurecr.io/helm/mywebapp:0.1.0

myoperator/
├── crds/
│   └── myresource-crd.yaml    # Installed BEFORE templates on first install
├── templates/
│   ├── myresource-instance.yaml  # Custom resource using the CRD
│   └── deployment.yaml
└── Chart.yaml

# Lab: Create and deploy a custom chart

# Step 1: Scaffold
helm create demoweb
cd demoweb

# Step 2: Customize values.yaml
# Change image to nginx:alpine, set replicaCount: 2

# Step 3: Add a ConfigMap template
cat > templates/configmap.yaml <<'EOF'
apiVersion: v1
kind: ConfigMap
metadata:
  name: {{ include "demoweb.fullname" . }}-config
  labels:
    {{- include "demoweb.labels" . | nindent 4 }}
data:
  APP_ENV: {{ .Values.appEnv | default "development" | quote }}
  APP_NAME: {{ .Chart.Name | quote }}
EOF

# Step 4: Lint
helm lint .

# Step 5: Render and inspect
helm template demo . | less

# Step 6: Install
helm install demo . -n demo --create-namespace

# Step 7: Verify
kubectl get all -n demo
helm get manifest demo -n demo

# Step 8: Package
cd ..
helm package demoweb/
ls *.tgz

# Cleanup
helm uninstall demo -n demo

# Usually a whitespace/indentation issue in Go templates
# Step 1: Render the template to see raw output
helm template test . -s templates/deployment.yaml 2>&1

# Step 2: Use --debug for more detail
helm template test . --debug

# Common causes:
# - Missing {{- (dash) causing unwanted newlines
# - nindent with wrong number
# - Unclosed {{ if }} blocks

# Another release or manual kubectl create owns that resource
# Options:
# 1. Use a unique fullname (check _helpers.tpl)
# 2. Adopt existing resource:
kubectl annotate configmap existing-cm \
  meta.helm.sh/release-name=myrelease \
  meta.helm.sh/release-namespace=default
kubectl label configmap existing-cm \
  app.kubernetes.io/managed-by=Helm

# helm lint doesn't talk to the cluster, so:
# - Invalid apiVersion won't be caught
# - Missing CRDs won't be caught
# - Resource quotas won't be checked

# Use --dry-run to validate against the cluster API
helm install test . --dry-run --debug

# Also run kubeconform or kubeval for schema validation
helm template test . | kubeconform -strict