​

Version 2.12.0 - Enhanced Security & Performance

​

🎯 Overview

​

🚀 Key Features

​

Redis-Based Rate Limiting with Intelligent Fallback

• Advanced DDoS Protection: Implemented Redis-based rate limiting for all API endpoints
• Intelligent Fallback: In-memory fallback mechanism ensures rate limiting continues even during Redis failures
• Tiered Limits: Different rate limits for authentication (5/15min), API (60/min), and sensitive operations
• IP-Based Tracking: Removed user-agent from rate limit keys to prevent bypass attempts

​

LRU Cache Implementation

• Memory-Efficient Caching: New Least Recently Used (LRU) cache with automatic eviction
• Configurable TTL: Cache time-to-live configurable via RAG_CACHE_TTL_MS environment variable
• Automatic Cleanup: Periodic cleanup of expired entries every minute
• Prevents Memory Leaks: Maximum cache size limits with intelligent eviction strategies

​

Enhanced Security Features

• Password Security: Upgraded to Bcrypt cost factor 14 (16,384 iterations) for stronger password hashing
• Dynamic CSP Nonces: Cryptographically secure Content Security Policy nonces generated per request
• Security Headers: Comprehensive headers including HSTS, X-Frame-Options, X-Content-Type-Options
• Timing Attack Prevention: Fixed delay added to failed authentication attempts

​

Dual Storage Display

• Comprehensive Storage View: Library page now shows both file storage and RAG vector storage usage
• Vector Storage Metrics: Display embedding dimensions, chunk counts, and estimated storage in MB
• Real-Time Updates: Cache invalidation on document operations for accurate storage display

​

Performance Optimizations

• Static Imports: Replaced dynamic imports with static imports for faster build times
• Build Performance: Significant reduction in Next.js build times
• Database Query Optimization: Improved query patterns for storage calculations

​

🔧 Technical Improvements

​

Code Quality

• Eliminated Duplication: Created shared utilities (rag-storage-utils.ts) for consistent storage calculations
• Type Safety: Fixed all TypeScript errors and improved type definitions
• Error Handling: Enhanced error handling with proper fallback mechanisms

​

Infrastructure

• Redis Integration: Full Redis support with connection pooling and retry strategies
• Environment Variables: New configuration options for cache TTL and Redis connection
• Docker Optimization: Improved container builds with better dependency management

​

🐛 Bug Fixes

• Fixed storage validation bug using wrong property (usage vs totalUsage)
• Resolved duplicate generateNonce function definitions
• Fixed TypeScript errors in AI documents route
• Corrected type assertions in OpenSource component
• Fixed user_id format mismatch causing 0 bytes display in RAG storage

​

📝 Configuration Changes

​

New Environment Variables

# Redis Configuration (Optional)
REDIS_URL=redis://localhost:6379

# Cache Configuration (Optional)
RAG_CACHE_TTL_MS=60000  # Default: 1 minute

​

Updated Dependencies

• Added ioredis for Redis support
• Security dependency updates

​

🔄 Migration Notes

​

For Existing Installations

1. Optional Redis Setup: Redis is optional but recommended for production
2. No Database Changes: No migrations required
3. Backward Compatible: All changes are backward compatible

​

Upgrade Steps

# Pull latest changes
git pull origin main

# Install new dependencies
pnpm install

# Build application
pnpm build

# Restart services
pnpm start

​

🔐 Security Advisory

• Rate limiting prevents brute force attacks
• Enhanced password hashing increases resistance to rainbow table attacks
• CSP nonces prevent XSS attacks
• Security headers protect against various web vulnerabilities

​

📊 Performance Metrics

• Build Time: ~30% faster with static imports
• Cache Hit Rate: ~80% for RAG storage queries
• Memory Usage: Stable with LRU cache eviction
• Rate Limiting: less than 1 ms overhead per request

​

🙏 Acknowledgments

​

📚 Documentation

• Security Overview
• Rate Limiting Configuration
• Cache Management

​

🚀 What’s Next

• Implement distributed caching with Redis Cluster
• Add more granular rate limiting rules
• Enhance monitoring and alerting capabilities
• Continue security hardening efforts