Skip to main content

Infrastructure

This document provides technical details about the 23blocks platform infrastructure, including our edge architecture, URL patterns, and environment routing.

Edge Architecture​

Every API request to 23blocks flows through our global edge network:

Your App → CloudFront (400+ edge locations) → Lambda@Edge → Service API

Request Flow​

  1. Edge Location - Request hits the nearest CloudFront edge location (typically < 50ms from any user)
  2. Lambda@Edge - Validates API key and determines routing
  3. Origin Shield - Regional caching layer reduces origin requests
  4. Application Load Balancer - SSL termination and health checking
  5. Service Container - Your request reaches the service API

Per-Service URL Pattern​

Each 23blocks service has its own dedicated URL following this pattern:

https://{service}.api.us.23blocks.com

Complete Service URL List​

ServiceURLDescription
Authhttps://auth.api.us.23blocks.comAuthentication, users, MFA, SSO
Contenthttps://content.api.us.23blocks.comHeadless CMS
Fileshttps://files.api.us.23blocks.comCloud storage & CDN
Formshttps://forms.api.us.23blocks.comDynamic forms
CRMhttps://crm.api.us.23blocks.comCustomer relationships
Wallethttps://wallet.api.us.23blocks.comDigital wallets & payments
Assetshttps://assets.api.us.23blocks.comDigital asset management
Onboardinghttps://onboarding.api.us.23blocks.comUser activation flows
Companieshttps://company.api.us.23blocks.comTeams & organizations
Productshttps://products.api.us.23blocks.comProduct catalog
Saleshttps://sales.api.us.23blocks.comSubscriptions & payments
Rewardshttps://rewards.api.us.23blocks.comLoyalty & gamification
Searchhttps://search.api.us.23blocks.comFull-text search
Universityhttps://university.api.us.23blocks.comLearning management
AI/Jarvishttps://jarvis.api.us.23blocks.comMachine learning features
Geolocationhttps://geolocation.api.us.23blocks.comMaps & locations
Real Timehttps://conversations.api.us.23blocks.comWebSocket & WebRTC

Environment Routing​

API Key-Based Routing​

Use the same URL for all environments. Your API key prefix determines routing:

API Key PrefixEnvironmentUse Case
pk_test_*StagingPublic key for client-side staging
sk_test_*StagingSecret key for server-side staging
pk_live_*ProductionPublic key for client-side production
sk_live_*ProductionSecret key for server-side production

How It Works​

The Lambda@Edge function reads your API key and routes to the appropriate backend:

API Key: pk_test_abc123
         ↓
         Prefix: pk_test_
         ↓
         Route to: Staging environment

Configuration Example​

import { create23BlocksClient } from '@23blocks/sdk';

// Development/Staging
const stagingClient = create23BlocksClient({
  urls: { authentication: 'https://auth.api.us.23blocks.com' },
  apiKey: process.env.BLOCKS_TEST_API_KEY,  // pk_test_* or sk_test_*
});

// Production
const productionClient = create23BlocksClient({
  urls: { authentication: 'https://auth.api.us.23blocks.com' },
  apiKey: process.env.BLOCKS_LIVE_API_KEY,  // pk_live_* or sk_live_*
});

Benefits​

  • Same URL everywhere - No environment-specific URLs to configure
  • Safer deployments - Swap API keys, not infrastructure
  • Consistent testing - Same code paths, same URLs
  • Simple CI/CD - Environment is a configuration value

Health Check Endpoints​

Each service exposes a health check endpoint for monitoring:

GET https://{service}.api.us.23blocks.com/health

Response Format​

{
  "status": "healthy",
  "service": "auth",
  "version": "2.4.1",
  "timestamp": "2026-01-20T10:30:00Z",
  "checks": {
    "database": "healthy",
    "cache": "healthy",
    "dependencies": "healthy"
  }
}

Status Values​

StatusDescription
healthyAll systems operational
degradedPartial functionality, some features may be slow
unhealthyService is down or unresponsive

Using Health Checks​

# Check Auth service health
curl https://auth.api.us.23blocks.com/health

# Check multiple services
for service in auth content files forms; do
  curl -s "https://${service}.api.us.23blocks.com/health" | jq '.status'
done

AWS Infrastructure​

Compute Layer​

  • ECS Fargate - Serverless container orchestration
  • Auto Scaling - Automatic capacity based on demand
  • Multi-AZ - Containers distributed across availability zones

Database Layer​

  • RDS Aurora PostgreSQL - Multi-AZ database clusters
  • Read Replicas - Distributed read capacity
  • Automated Backups - Daily snapshots with 35-day retention

Caching Layer​

  • ElastiCache Redis - In-memory caching
  • Multi-AZ Replication - High availability cache cluster

Storage Layer​

  • S3 - Object storage with 99.999999999% durability
  • CloudFront - Global CDN for file delivery
  • S3 Transfer Acceleration - Fast uploads worldwide

Network Layer​

  • CloudFront - 400+ edge locations
  • Lambda@Edge - Intelligent routing at the edge
  • Application Load Balancer - SSL termination, health checks
  • WAF - Web application firewall protection

Security​

Encryption​

  • In Transit - TLS 1.3 for all connections
  • At Rest - AES-256 encryption for databases and storage
  • Key Management - AWS KMS for key rotation

Network Security​

  • VPC Isolation - Private subnets for services
  • Security Groups - Strict inbound/outbound rules
  • WAF Rules - Protection against common attacks

DDoS Protection​

  • AWS Shield Standard - Always-on DDoS protection
  • CloudFront - Edge-level DDoS mitigation
  • Rate Limiting - Per-API key request limits

Monitoring​

Metrics​

  • Prometheus - Time-series metrics collection
  • Custom Metrics - Request latency, error rates, throughput

Visualization​

  • Grafana - Real-time dashboards
  • Per-Service Dashboards - Service-specific metrics
  • Executive Dashboard - Platform-wide health overview

Logging​

  • Loki - Centralized log aggregation
  • Structured Logs - JSON format for easy querying
  • Log Retention - 90 days for debugging

Alerting​

  • Alert Rules - Threshold-based alerts
  • PagerDuty Integration - On-call engineer notification
  • < 30 Second Alert Time - Fast incident response

SLA & Uptime​

TierUptime SLAResponse Time
Shared Cloud99.5%24-48 hours
Dedicated Cloud99.9%4 hours
Enterprise99.99%1 hour

Incident Communication​

  • Status Page - https://status.23blocks.com
  • Email Notifications - Subscribed customers notified
  • Slack Updates - Real-time incident updates

Regional Availability​

Currently available in:

RegionLocationStatus
usUS East (N. Virginia)Active

Coming Soon​

  • eu - Europe (Frankfurt)
  • ap - Asia Pacific (Singapore)