Configuration Examples

Overview

This page provides production-ready configuration examples for common deployment scenarios. Copy and adapt these configurations for your infrastructure.

Microservices

Load balance across microservice instances

API Gateway

API gateway configuration

High Availability

Multi-region high-availability setup

Development

Local development configuration

Common Scenarios

Simple Web Application

Basic setup for a web application with 3 backend servers:

listen_address: "0.0.0.0"
listen_port: 80
algorithm: "round-robin"

backends:
  - host: "web1.internal"
    port: 8080
  - host: "web2.internal"
    port: 8080
  - host: "web3.internal"
    port: 8080

health_check:
  enabled: true
  interval_ms: 5000
  path: "/health"

logging:
  level: "info"
  format: "json"

API Gateway

Load balancer as an API gateway with multiple services:

config-api-gateway.yaml

listen_address: "0.0.0.0"
listen_port: 443
algorithm: "least-connections"
max_connections: 50000

# User service backends
backends:
  - host: "user-service-1.prod.internal"
    port: 8080
    weight: 100
  - host: "user-service-2.prod.internal"
    port: 8080
    weight: 100

health_check:
  enabled: true
  interval_ms: 3000
  timeout_ms: 1500
  max_failures: 3
  path: "/api/health"

timeout:
  connect_ms: 3000
  request_ms: 30000
  idle_ms: 60000

tls:
  enabled: true
  cert_path: "/etc/ssl/certs/api.example.com.crt"
  key_path: "/etc/ssl/private/api.example.com.key"
  min_version: "1.3"

logging:
  level: "info"
  format: "json"
  output: "/var/log/ultrabalancer/api-gateway.log"

metrics:
  enabled: true
  endpoint: "/metrics"
  prometheus_endpoint: "/prometheus"

Microservices Architecture

Configuration for microservices with different weights:

config-microservices.yaml

listen_address: "0.0.0.0"
listen_port: 8080
algorithm: "weighted"
workers: auto

# Mix of different instance types
backends:
  # High-performance instances
  - host: "ms-large-1.internal"
    port: 8080
    weight: 300              # 3x capacity
    max_connections: 3000

  - host: "ms-large-2.internal"
    port: 8080
    weight: 300
    max_connections: 3000

  # Standard instances
  - host: "ms-medium-1.internal"
    port: 8080
    weight: 200              # 2x capacity
    max_connections: 2000

  - host: "ms-medium-2.internal"
    port: 8080
    weight: 200
    max_connections: 2000

  # Small instances
  - host: "ms-small-1.internal"
    port: 8080
    weight: 100              # 1x capacity
    max_connections: 1000

health_check:
  enabled: true
  interval_ms: 5000
  max_failures: 3
  path: "/actuator/health"

  circuit_breaker:
    enabled: true
    failure_threshold: 5
    success_threshold: 2
    timeout_seconds: 30

logging:
  level: "info"
  format: "json"
  output: "/var/log/ultrabalancer/microservices.log"

Session Affinity (Sticky Sessions)

Use IP hash for session persistence:

config-sticky-sessions.yaml

listen_address: "0.0.0.0"
listen_port: 80
algorithm: "ip-hash"         # Same client → same backend

backends:
  - host: "app1.internal"
    port: 8080
  - host: "app2.internal"
    port: 8080
  - host: "app3.internal"
    port: 8080

health_check:
  enabled: true
  interval_ms: 5000
  path: "/health"

timeout:
  connect_ms: 5000
  request_ms: 60000          # Longer for sessions
  idle_ms: 120000

logging:
  level: "info"
  format: "json"

High-Availability Setup

Multi-region HA configuration with circuit breaker:

config-ha.yaml

listen_address: "0.0.0.0"
listen_port: 80
algorithm: "least-connections"
workers: 16
max_connections: 100000

backends:
  # Primary region (us-east-1)
  - host: "app-useast1a-1.prod.internal"
    port: 8080
    weight: 150
    max_connections: 5000

  - host: "app-useast1a-2.prod.internal"
    port: 8080
    weight: 150
    max_connections: 5000

  - host: "app-useast1b-1.prod.internal"
    port: 8080
    weight: 150
    max_connections: 5000

  # Failover region (us-west-2)
  - host: "app-uswest2a-1.prod.internal"
    port: 8080
    weight: 100
    max_connections: 3000

  - host: "app-uswest2a-2.prod.internal"
    port: 8080
    weight: 100
    max_connections: 3000

health_check:
  enabled: true
  interval_ms: 2000          # Fast detection
  timeout_ms: 1000
  max_failures: 2

  circuit_breaker:
    enabled: true
    failure_threshold: 5
    success_threshold: 3
    timeout_seconds: 60
    half_open_requests: 5

timeout:
  connect_ms: 3000
  request_ms: 30000
  idle_ms: 60000

logging:
  level: "info"
  format: "json"
  output: "/var/log/ultrabalancer/ha.log"
  max_size_mb: 500
  max_files: 30

metrics:
  enabled: true

Environment-Specific Configurations

Development

config-dev.yaml

listen_address: "127.0.0.1"
listen_port: 8080
algorithm: "round-robin"

backends:
  - host: "localhost"
    port: 3001
  - host: "localhost"
    port: 3002

health_check:
  enabled: true
  interval_ms: 10000         # Less aggressive
  max_failures: 5            # More tolerant

timeout:
  connect_ms: 10000          # Generous timeouts
  request_ms: 60000

logging:
  level: "debug"             # Verbose logging
  format: "text"             # Human-readable
  output: "stdout"

Staging

config-staging.yaml

listen_address: "0.0.0.0"
listen_port: 8080
algorithm: "least-connections"
workers: 4

backends:
  - host: "backend1.staging.internal"
    port: 8080
  - host: "backend2.staging.internal"
    port: 8080

health_check:
  enabled: true
  interval_ms: 5000
  path: "/health"

timeout:
  connect_ms: 5000
  request_ms: 30000

logging:
  level: "info"
  format: "json"
  output: "/var/log/ultrabalancer/staging.log"

metrics:
  enabled: true

Production

config-prod.yaml

listen_address: "0.0.0.0"
listen_port: 80
algorithm: "least-connections"
workers: auto
max_connections: 50000

backends:
  - host: "backend1.prod.internal"
    port: 8080
    weight: 100
    max_connections: 2000

  - host: "backend2.prod.internal"
    port: 8080
    weight: 100
    max_connections: 2000

  - host: "backend3.prod.internal"
    port: 8080
    weight: 100
    max_connections: 2000

health_check:
  enabled: true
  interval_ms: 3000
  timeout_ms: 1500
  max_failures: 3
  path: "/health"

  circuit_breaker:
    enabled: true
    failure_threshold: 5
    success_threshold: 2
    timeout_seconds: 60

timeout:
  connect_ms: 5000
  request_ms: 30000
  idle_ms: 60000

logging:
  level: "info"
  format: "json"
  output: "/var/log/ultrabalancer/prod.log"
  max_size_mb: 100
  max_files: 30

metrics:
  enabled: true

Cloud Provider Configurations

AWS

EC2 Instances
ECS Tasks
Lambda Integration

config-aws-ec2.yaml

listen_address: "0.0.0.0"
listen_port: 80
algorithm: "least-connections"

backends:
  # Private subnet instances
  - host: "10.0.1.10"     # us-east-1a
    port: 8080
    weight: 100

  - host: "10.0.1.11"     # us-east-1a
    port: 8080
    weight: 100

  - host: "10.0.2.10"     # us-east-1b
    port: 8080
    weight: 100

health_check:
  enabled: true
  interval_ms: 5000
  path: "/health"

logging:
  level: "info"
  format: "json"
  output: "/var/log/ultrabalancer/access.log"

config-aws-ecs.yaml

listen_address: "0.0.0.0"
listen_port: 8080
algorithm: "least-connections"

# Service discovery backends
backends:
  - host: "service.prod.local"  # ECS Service Discovery
    port: 8080
    weight: 100

health_check:
  enabled: true
  interval_ms: 5000
  path: "/health"

logging:
  level: "info"
  format: "json"
  output: "stdout"  # CloudWatch Logs

config-aws-lambda.yaml

listen_address: "0.0.0.0"
listen_port: 80
algorithm: "round-robin"

backends:
  # Lambda Function URLs
  - host: "abc123.lambda-url.us-east-1.on.aws"
    port: 443

  - host: "def456.lambda-url.us-east-1.on.aws"
    port: 443

health_check:
  enabled: true
  interval_ms: 10000
  path: "/health"

tls:
  enabled: true  # HTTPS to Lambda

Google Cloud Platform

config-gcp.yaml

listen_address: "0.0.0.0"
listen_port: 80
algorithm: "least-connections"

backends:
  # Compute Engine instances
  - host: "10.128.0.2"        # us-central1-a
    port: 8080

  - host: "10.128.0.3"        # us-central1-a
    port: 8080

  - host: "10.132.0.2"        # us-central1-b
    port: 8080

health_check:
  enabled: true
  interval_ms: 5000
  path: "/health"

logging:
  level: "info"
  format: "json"
  output: "stdout"  # Cloud Logging

Azure

config-azure.yaml

listen_address: "0.0.0.0"
listen_port: 80
algorithm: "least-connections"

backends:
  # VM Scale Set instances
  - host: "10.0.1.4"          # Availability Zone 1
    port: 8080

  - host: "10.0.1.5"          # Availability Zone 1
    port: 8080

  - host: "10.0.2.4"          # Availability Zone 2
    port: 8080

health_check:
  enabled: true
  interval_ms: 5000
  path: "/api/health"

logging:
  level: "info"
  format: "json"
  output: "/var/log/ultrabalancer/access.log"

DigitalOcean

config-digitalocean.yaml

listen_address: "0.0.0.0"
listen_port: 80
algorithm: "round-robin"

backends:
  # Droplets in same VPC
  - host: "10.116.0.2"
    port: 8080

  - host: "10.116.0.3"
    port: 8080

  - host: "10.116.0.4"
    port: 8080

health_check:
  enabled: true
  interval_ms: 5000
  path: "/health"

logging:
  level: "info"
  format: "json"

Special Use Cases

WebSocket Support

config-websocket.yaml

listen_address: "0.0.0.0"
listen_port: 8080
algorithm: "ip-hash"         # Sticky sessions for WebSocket

backends:
  - host: "ws1.internal"
    port: 8080
  - host: "ws2.internal"
    port: 8080

timeout:
  connect_ms: 5000
  request_ms: 300000         # 5 minutes for long-lived connections
  idle_ms: 600000            # 10 minutes idle timeout

health_check:
  enabled: true
  interval_ms: 10000
  path: "/ws/health"

gRPC Services

config-grpc.yaml

listen_address: "0.0.0.0"
listen_port: 50051
algorithm: "least-connections"

backends:
  - host: "grpc-service-1.internal"
    port: 50051
  - host: "grpc-service-2.internal"
    port: 50051

health_check:
  enabled: true
  interval_ms: 5000
  path: "/grpc.health.v1.Health/Check"

timeout:
  connect_ms: 3000
  request_ms: 30000

Container Orchestration

Docker Compose
Kubernetes
Nomad

docker-compose.yml

version: '3.8'

services:
  ultrabalancer:
    image: ultrabalancer/ultrabalancer:latest
    ports:
      - "80:8080"
    volumes:
      - ./config.yaml:/etc/ultrabalancer/config.yaml
    command: --config /etc/ultrabalancer/config.yaml

  backend1:
    image: myapp:latest
    expose:
      - "8080"

  backend2:
    image: myapp:latest
    expose:
      - "8080"

  backend3:
    image: myapp:latest
    expose:
      - "8080"

config.yaml

listen_address: "0.0.0.0"
listen_port: 8080
algorithm: "round-robin"

backends:
  - host: "backend1"
    port: 8080
  - host: "backend2"
    port: 8080
  - host: "backend3"
    port: 8080

k8s-config.yaml

apiVersion: v1
kind: ConfigMap
metadata:
  name: ultrabalancer-config
data:
  config.yaml: |
    listen_address: "0.0.0.0"
    listen_port: 8080
    algorithm: "least-connections"

    backends:
      - host: "backend-service"
        port: 8080

    health_check:
      enabled: true
      interval_ms: 5000
      path: "/health"

    logging:
      level: "info"
      format: "json"

nomad-job.hcl

job "ultrabalancer" {
  datacenters = ["dc1"]

  group "lb" {
    count = 1

    task "ultrabalancer" {
      driver = "docker"

      config {
        image = "ultrabalancer/ultrabalancer:latest"
        port_map {
          http = 8080
        }
      }

      template {
        data = <<EOF
listen_address: "0.0.0.0"
listen_port: 8080
algorithm: "round-robin"

backends:
  {{- range service "backend" }}
  - host: "{{ .Address }}"
    port: {{ .Port }}
  {{- end }}
EOF
        destination = "local/config.yaml"
      }

      env {
        ULTRA_CONFIG = "local/config.yaml"
      }
    }
  }
}

Blue-Green Deployment

config-blue-green.yaml

listen_address: "0.0.0.0"
listen_port: 80
algorithm: "weighted"

backends:
  # Blue environment (current production)
  - host: "blue-1.internal"
    port: 8080
    weight: 100              # 100% traffic

  - host: "blue-2.internal"
    port: 8080
    weight: 100

  # Green environment (new version, no traffic)
  - host: "green-1.internal"
    port: 8080
    weight: 0                # 0% traffic initially

  - host: "green-2.internal"
    port: 8080
    weight: 0

health_check:
  enabled: true
  interval_ms: 3000

Switching traffic to green:

# Update weights and reload
backends:
  - host: "blue-1.internal"
    weight: 0                # 0% traffic
  - host: "blue-2.internal"
    weight: 0
  - host: "green-1.internal"
    weight: 100              # 100% traffic
  - host: "green-2.internal"
    weight: 100

# Reload configuration
kill -HUP $(pidof ultrabalancer)

Canary Deployment

config-canary.yaml

listen_address: "0.0.0.0"
listen_port: 80
algorithm: "weighted"

backends:
  # Stable version (95% traffic)
  - host: "stable-1.internal"
    port: 8080
    weight: 475

  - host: "stable-2.internal"
    port: 8080
    weight: 475

  # Canary version (5% traffic)
  - host: "canary-1.internal"
    port: 8080
    weight: 50

health_check:
  enabled: true
  interval_ms: 3000

Configuration Overview

Learn about all configuration methods

CLI Reference

Command-line interface documentation

File Config

Complete YAML/TOML reference

Deployment

Deploy UltraBalancer in production

Getting Started

Concepts

Configuration

Deployment

Testing

API

Advanced

Overview

Microservices

API Gateway

High Availability

Development

Common Scenarios

Simple Web Application

API Gateway

Microservices Architecture

Session Affinity (Sticky Sessions)

High-Availability Setup

Environment-Specific Configurations

Development

Staging

Production

Cloud Provider Configurations

AWS

Google Cloud Platform

Azure

DigitalOcean

Special Use Cases

WebSocket Support

gRPC Services

Container Orchestration

Blue-Green Deployment

Canary Deployment

Configuration Overview

CLI Reference

File Config

Deployment

Getting Started

Concepts

Configuration

Deployment

Testing

API

Advanced

​Overview

Microservices

API Gateway

High Availability

Development

​Common Scenarios

​Simple Web Application

​API Gateway

​Microservices Architecture

​Session Affinity (Sticky Sessions)

​High-Availability Setup

​Environment-Specific Configurations

​Development

​Staging

​Production

​Cloud Provider Configurations

​AWS

​Google Cloud Platform

​Azure

​DigitalOcean

​Special Use Cases

​WebSocket Support

​gRPC Services

​Container Orchestration

​Blue-Green Deployment

​Canary Deployment

​Related Topics

Configuration Overview

CLI Reference

File Config

Deployment

Overview

Common Scenarios

Simple Web Application

API Gateway

Microservices Architecture

Session Affinity (Sticky Sessions)

High-Availability Setup

Environment-Specific Configurations

Development

Staging

Production

Cloud Provider Configurations

AWS

Google Cloud Platform

Azure

DigitalOcean

Special Use Cases

WebSocket Support

gRPC Services

Container Orchestration

Blue-Green Deployment

Canary Deployment

Related Topics