Deploying Eden on Kubernetes with Helm

Tags: release

Eden v0.7.0 ships with official Helm charts for Kubernetes. Whether you're running a single-node test cluster or a multi-region production environment, getting Eden up and running is now a few commands away.

Quick Start

# Add the Eden Helm repository
helm repo add eden https://charts.eden.dev
helm repo update

# Install with defaults
helm install eden eden/eden

# Or with your own configuration
helm install eden eden/eden -f values.yaml

That's it. You're running Eden on Kubernetes.

What You Get

The chart deploys Eden with sensible production defaults:

• Deployment with rolling updates
• Services for internal and external access
• Health checks (startup, readiness, liveness)
• Security — non-root containers with dropped capabilities
• Resource limits to play nice with your cluster

Configuration Basics

# values.yaml
replicaCount: 3

image:
  repository: eden/eden-service
  tag: v0.7.0

resources:
  requests:
    cpu: "200m"
    memory: "256Mi"
  limits:
    memory: "1Gi"

Need TLS on your load balancer? AWS ALB annotations? Custom environment variables? The chart supports all of it.

Auto-Scaling

Enable horizontal pod autoscaling to handle traffic spikes:

autoscaling:
  enabled: true
  minReplicas: 2
  maxReplicas: 10
  targetCPUUtilizationPercentage: 70

You can also scale based on custom metrics from Prometheus, like requests per second.

Monitoring with Prometheus

Eden exposes metrics at /metrics. Hook it up to your Prometheus stack:

apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
  name: eden
spec:
  selector:
    matchLabels:
      app: eden
  endpoints:
    - port: http
      path: /metrics
      interval: 15s

Key metrics to watch:

• eden_requests_total — Request count by endpoint and status
• eden_request_duration_seconds — Latency histogram
• eden_connections_active — Current connection count
• eden_migration_progress — How far along your migrations are

Before You Go to Production

A quick checklist:

• [ ] At least 2 replicas for high availability
• [ ] Resource limits based on load testing
• [ ] TLS on public services
• [ ] Prometheus monitoring set up
• [ ] PodDisruptionBudget configured
• [ ] Node affinity for multi-AZ distribution

Full Helm chart documentation →