feat: Phase 1 Production Validation Infrastructure

.env.example

@@ -183,8 +183,30 @@ STREAM_POLL_INTERVAL_MS=2000
 STREAM_MAX_RETRIES=5
 NOTIFY_DEDUP_TTL_SECONDS=86400
 
-# Database pool size
-DATABASE_POOL_SIZE=10
+# =============================================================================
+# Database Connection Pool
+# Recommended: min=10, max=30 for concurrent workers
+# =============================================================================
+DATABASE_POOL_MIN=10
+DATABASE_POOL_MAX=30
+DATABASE_IDLE_TIMEOUT=30000
+DATABASE_CONNECTION_TIMEOUT=10000
+
+# =============================================================================
+# Worker Configuration
+# =============================================================================
+# Graceful shutdown timeout (milliseconds)
+GRACEFUL_SHUTDOWN_TIMEOUT=30000
+# Worker type identifier (set by docker-compose)
+WORKER_TYPE=
+
+# =============================================================================
+# Sentry Error Tracking (optional but recommended for production)
+# Get DSN from: https://sentry.io
+# =============================================================================
+SENTRY_DSN=https://xxx@xxx.ingest.sentry.io/xxx
+SENTRY_RELEASE=pricehawk@0.1.0
+SENTRY_TRACES_SAMPLE_RATE=0.1
 
 # =============================================================================
 # Cron & Admin Security

docker-compose.yml

@@ -7,17 +7,36 @@ services:
       POSTGRES_USER: pricehawk
       POSTGRES_PASSWORD: password
       POSTGRES_DB: pricehawk
+      # Connection pool settings (recommended: max_connections = max_pool * num_workers + overhead)
+      POSTGRES_INITDB_ARGS: "--data-checksums"
     ports:
       - "5432:5432"
     volumes:
       - postgres_data:/var/lib/postgresql/data
+    healthcheck:
+      test: ["CMD-SHELL", "pg_isready -U pricehawk -d pricehawk"]
+      interval: 10s
+      timeout: 5s
+      retries: 5
+      start_period: 10s
 
   redis:
     image: redis:7-alpine
     ports:
       - "6379:6379"
     volumes:
       - redis_data:/data
+    # Enable AOF persistence for BullMQ job recovery
+    # AOF fsync policy: everysec provides good balance of durability and performance
+    # Memory policy: allkeys-lru evicts least recently used keys when memory limit reached
+    # This ensures job queue reliability even under memory pressure
+    command: redis-server --appendonly yes --appendfsync everysec --maxmemory 256mb --maxmemory-policy allkeys-lru
+    healthcheck:
+      test: ["CMD", "redis-cli", "ping"]
+      interval: 10s
+      timeout: 5s
+      retries: 5
+      start_period: 5s
 
   app:
     build:
@@ -29,11 +48,21 @@ services:
       REDIS_HOST: redis
       REDIS_PORT: 6379
       NEXT_PUBLIC_APP_URL: http://localhost:3000
+      DATABASE_POOL_MIN: "10"
+      DATABASE_POOL_MAX: "30"
     ports:
       - "3000:3000"
     depends_on:
-      - postgres
-      - redis
+      postgres:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+    healthcheck:
+      test: ["CMD-SHELL", "curl -f http://localhost:3000/api/health || exit 1"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 40s
 
   # Separate service for Scraper because it needs Playwright browsers
   # We use the official Playwright image to ensure browsers are there
@@ -49,9 +78,21 @@ services:
       DATABASE_URL: postgres://pricehawk:password@postgres:5432/pricehawk
       REDIS_HOST: redis
       REDIS_PORT: 6379
+      WORKER_TYPE: scraper
+      GRACEFUL_SHUTDOWN_TIMEOUT: "30000"
     depends_on:
-      - postgres
-      - redis
+      postgres:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+    healthcheck:
+      test: ["CMD-SHELL", "pgrep -f 'tsx src/worker.ts' || exit 1"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 60s
+    # Graceful shutdown: allow worker to finish in-flight jobs
+    stop_grace_period: 30s
 
   # Other workers can use the base App image (lighter) or the same playwright image if code is shared
   worker-validator:
@@ -61,9 +102,20 @@ services:
       DATABASE_URL: postgres://pricehawk:password@postgres:5432/pricehawk
       REDIS_HOST: redis
       REDIS_PORT: 6379
+      WORKER_TYPE: anomaly-validator
+      GRACEFUL_SHUTDOWN_TIMEOUT: "30000"
     depends_on:
-      - postgres
-      - redis
+      postgres:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+    healthcheck:
+      test: ["CMD-SHELL", "pgrep -f 'anomaly-validator' || exit 1"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 30s
+    stop_grace_period: 30s
 
   worker-notify:
     build: .
@@ -72,9 +124,20 @@ services:
       DATABASE_URL: postgres://pricehawk:password@postgres:5432/pricehawk
       REDIS_HOST: redis
       REDIS_PORT: 6379
+      WORKER_TYPE: notification-sender
+      GRACEFUL_SHUTDOWN_TIMEOUT: "30000"
     depends_on:
-      - postgres
-      - redis
+      postgres:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+    healthcheck:
+      test: ["CMD-SHELL", "pgrep -f 'notification-sender' || exit 1"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 30s
+    stop_grace_period: 30s
 
   worker-social:
     build: .
@@ -83,9 +146,20 @@ services:
       DATABASE_URL: postgres://pricehawk:password@postgres:5432/pricehawk
       REDIS_HOST: redis
       REDIS_PORT: 6379
+      WORKER_TYPE: social-poster
+      GRACEFUL_SHUTDOWN_TIMEOUT: "30000"
     depends_on:
-      - postgres
-      - redis
+      postgres:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+    healthcheck:
+      test: ["CMD-SHELL", "pgrep -f 'social-poster' || exit 1"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 30s
+    stop_grace_period: 30s
 
   worker-verifier:
     build: .
@@ -94,9 +168,64 @@ services:
       DATABASE_URL: postgres://pricehawk:password@postgres:5432/pricehawk
       REDIS_HOST: redis
       REDIS_PORT: 6379
+      WORKER_TYPE: deal-verifier
+      GRACEFUL_SHUTDOWN_TIMEOUT: "30000"
     depends_on:
-      - postgres
-      - redis
+      postgres:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+    healthcheck:
+      test: ["CMD-SHELL", "pgrep -f 'deal-verifier' || exit 1"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 30s
+    stop_grace_period: 30s
+
+  worker-newsletter:
+    build: .
+    command: tsx src/workers/newsletter-digest.ts
+    environment:
+      DATABASE_URL: postgres://pricehawk:password@postgres:5432/pricehawk
+      REDIS_HOST: redis
+      REDIS_PORT: 6379
+      WORKER_TYPE: newsletter-digest
+      GRACEFUL_SHUTDOWN_TIMEOUT: "30000"
+    depends_on:
+      postgres:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+    healthcheck:
+      test: ["CMD-SHELL", "pgrep -f 'newsletter-digest' || exit 1"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 30s
+    stop_grace_period: 30s
+
+  worker-telegram:
+    build: .
+    command: tsx src/workers/telegram-bot.ts
+    environment:
+      DATABASE_URL: postgres://pricehawk:password@postgres:5432/pricehawk
+      REDIS_HOST: redis
+      REDIS_PORT: 6379
+      WORKER_TYPE: telegram-bot
+      GRACEFUL_SHUTDOWN_TIMEOUT: "30000"
+    depends_on:
+      postgres:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+    healthcheck:
+      test: ["CMD-SHELL", "pgrep -f 'telegram-bot' || exit 1"]
+      interval: 30s
+      timeout: 10s
+      retries: 3
+      start_period: 30s
+    stop_grace_period: 30s
 
 volumes:
   postgres_data: