Production Readiness Checklist for Helm Values

Production Readiness Validation

Before deploying a Helm chart to production, the values.yaml should be reviewed against a comprehensive checklist. The Helm Values Validator helps catch the most common production readiness issues.

Production-Ready values.yaml

replicaCount: 3

image:
  repository: my-registry.io/my-app
  pullPolicy: IfNotPresent
  tag: "2.1.0"

imagePullSecrets:
  - name: registry-credentials

resources:
  limits:
    cpu: 500m
    memory: 512Mi
  requests:
    cpu: 250m
    memory: 256Mi

autoscaling:
  enabled: true
  minReplicas: 3
  maxReplicas: 20
  targetCPUUtilizationPercentage: 75

podSecurityContext:
  runAsNonRoot: true
  runAsUser: 1000
  fsGroup: 2000

securityContext:
  allowPrivilegeEscalation: false
  readOnlyRootFilesystem: true
  capabilities:
    drop:
      - ALL

nodeSelector:
  kubernetes.io/os: linux

Production Checklist

Category	Check	Status
Image	Specific tag (not "latest")	Critical
Image	pullPolicy: IfNotPresent	Recommended
Resources	Both limits and requests set	Critical
Scaling	replicaCount >= 2 or HPA enabled	Critical
Security	runAsNonRoot: true	Recommended
Security	readOnlyRootFilesystem	Recommended
Security	Drop all capabilities	Recommended
Networking	Ingress TLS configured	Recommended
Secrets	No hardcoded credentials	Critical

What the Validator Catches

• image.tag set to "latest" (warning)
• Missing resources.limits or resources.requests (warning/info)
• autoscaling.minReplicas > autoscaling.maxReplicas (error)
• replicaCount set while autoscaling is enabled (info)
• Invalid service.type or image.pullPolicy values (error)
• Type mismatches on known fields (warning)

Beyond the Validator

Some production checks require runtime context that a static validator cannot assess:

• Health check probes (liveness, readiness, startup)
• Pod disruption budgets
• Network policies
• RBAC configuration
• Persistent volume claims