Homework: Templating Fundamentals

Task 1: Multi-Service ConfigMap with Pipelines

Create a Helm chart called hw-templating with a ConfigMap template that reads a list of services from values.yaml and generates environment variables for each service.

values.yaml

services:
  - name: api-gateway
    port: 8080
    protocol: https
  - name: auth-service
    port: 3000
    protocol: http
  - name: worker
    port: 9090
    protocol: grpc

defaults:
  timeout: "30s"
  retries: 3

Requirements

• Use range to iterate over .Values.services
• For each service, generate two data entries: <NAME>_URL and <NAME>_PORT
• The name key must be uppercased and have hyphens replaced with underscores (use upper and replace)
• URL format: <protocol>://<name>:<port> (use printf and quote)
• Include .Values.defaults.timeout and .Values.defaults.retries with default fallbacks
• All string values must be quoted

Expected output (partial)

data:
  API_GATEWAY_URL: "https://api-gateway:8080"
  API_GATEWAY_PORT: "8080"
  AUTH_SERVICE_URL: "http://auth-service:3000"
  AUTH_SERVICE_PORT: "3000"
  WORKER_URL: "grpc://worker:9090"
  WORKER_PORT: "9090"
  DEFAULT_TIMEOUT: "30s"
  DEFAULT_RETRIES: "3"

Self-Check

• ✓ helm template test . renders without errors
• ✓ All 6 service entries appear with correct uppercased, underscore-separated names
• ✓ URLs follow the protocol://name:port format
• ✓ All string values are double-quoted in the output
• ✓ Default timeout and retries appear in the data section

Task 2: Conditional Deployment with Resource Toggles

Extend your chart with a Deployment template that uses conditionals, with, and range to handle optional sections.

Add to values.yaml

replicaCount: 3
image:
  repository: myapp
  tag: "2.1.0"

probes:
  liveness:
    enabled: true
    path: /healthz
    port: 8080
  readiness:
    enabled: true
    path: /ready
    port: 8080

nodeSelector:
  disktype: ssd
  region: us-east-1

tolerations:
  - key: "dedicated"
    operator: "Equal"
    value: "gpu"
    effect: "NoSchedule"

Requirements

• Use if to conditionally include liveness and readiness probes based on .enabled
• Use with for the nodeSelector block (skip entirely if empty)
• Use toYaml | nindent for tolerations
• Use required for .Values.image.tag so the template fails if tag is missing
• All trim markers must be correct — no blank lines in rendered output
• Access $.Release.Name from inside a with block using $

Self-Check

• ✓ helm template test . renders a valid Deployment with both probes
• ✓ Setting --set probes.liveness.enabled=false removes the livenessProbe block
• ✓ Setting --set nodeSelector=null removes the entire nodeSelector section
• ✓ Setting --set image.tag="" causes a render failure with your custom error message
• ✓ No spurious blank lines in the rendered YAML
• ✓ The tolerations section renders correctly indented

Task 3: Release-Aware Annotations with Capabilities

Create a Service template that uses .Release and .Capabilities objects for metadata and version-aware configuration.

Requirements

• Name the Service using .Release.Name
• Add annotations that include:
  • meta.helm.sh/release-name set to {{ .Release.Name }}
  • meta.helm.sh/release-namespace set to {{ .Release.Namespace }}
  • deploy-type set to "install" or "upgrade" based on .Release.IsInstall (use ternary)
  • kube-version set to {{ .Capabilities.KubeVersion.Version }}
• Use range to iterate over .Values.services and create port entries
• Each port should have name, port, and targetPort

Self-Check

• ✓ helm template test . --namespace production shows release-namespace: production
• ✓ The deploy-type annotation renders as "install" (template always simulates an install)
• ✓ The kube-version annotation contains a valid Kubernetes version string
• ✓ All three services from values appear as port entries
• ✓ All annotations are properly quoted