Rate Limiting

Overview

Rate limiting is a critical security and performance mechanism that controls the number of requests a client can make to the Strike Auth Service within a specified time window. It protects against abuse, prevents DDoS attacks, ensures fair resource usage, and maintains service quality for all users.

Rate Limiting Strategies

Fixed Window Rate Limiting

The simplest approach that resets the counter at fixed intervals.

class FixedWindowLimiter {
  constructor(maxRequests, windowMs) {
    this.maxRequests = maxRequests;
    this.windowMs = windowMs;
    this.requests = new Map();
  }
  
  isAllowed(key) {
    const now = Date.now();
    const windowStart = Math.floor(now / this.windowMs) * this.windowMs;
    const windowKey = `${key}:${windowStart}`;
    
    const currentCount = this.requests.get(windowKey) || 0;
    
    if (currentCount >= this.maxRequests) {
      return {
        allowed: false,
        remaining: 0,
        resetTime: windowStart + this.windowMs
      };
    }
    
    this.requests.set(windowKey, currentCount + 1);
    
    return {
      allowed: true,
      remaining: this.maxRequests - currentCount - 1,
      resetTime: windowStart + this.windowMs
    };
  }
}

Sliding Window Rate Limiting

More accurate approach that maintains a rolling window of requests.

class SlidingWindowLimiter {
  constructor(maxRequests, windowMs) {
    this.maxRequests = maxRequests;
    this.windowMs = windowMs;
    this.requests = new Map();
  }
  
  isAllowed(key) {
    const now = Date.now();
    const userRequests = this.requests.get(key) || [];
    
    // Remove requests outside the current window
    const validRequests = userRequests.filter(
      timestamp => now - timestamp < this.windowMs
    );
    
    if (validRequests.length >= this.maxRequests) {
      const oldestRequest = Math.min(...validRequests);
      return {
        allowed: false,
        remaining: 0,
        resetTime: oldestRequest + this.windowMs
      };
    }
    
    validRequests.push(now);
    this.requests.set(key, validRequests);
    
    return {
      allowed: true,
      remaining: this.maxRequests - validRequests.length,
      resetTime: now + this.windowMs
    };
  }
}

Token Bucket Rate Limiting

Allows burst traffic while maintaining average rate limits.

class TokenBucketLimiter {
  constructor(capacity, refillRate, refillPeriod) {
    this.capacity = capacity;
    this.refillRate = refillRate;
    this.refillPeriod = refillPeriod;
    this.buckets = new Map();
  }
  
  isAllowed(key, tokens = 1) {
    const now = Date.now();
    let bucket = this.buckets.get(key) || {
      tokens: this.capacity,
      lastRefill: now
    };
    
    // Refill tokens based on time elapsed
    const timePassed = now - bucket.lastRefill;
    const tokensToAdd = Math.floor(timePassed / this.refillPeriod) * this.refillRate;
    
    bucket.tokens = Math.min(this.capacity, bucket.tokens + tokensToAdd);
    bucket.lastRefill = now;
    
    if (bucket.tokens >= tokens) {
      bucket.tokens -= tokens;
      this.buckets.set(key, bucket);
      
      return {
        allowed: true,
        remaining: bucket.tokens,
        resetTime: null
      };
    }
    
    this.buckets.set(key, bucket);
    
    return {
      allowed: false,
      remaining: bucket.tokens,
      resetTime: bucket.lastRefill + (this.refillPeriod * Math.ceil((tokens - bucket.tokens) / this.refillRate))
    };
  }
}

Implementation with Express.js

Basic Rate Limiting Middleware

const rateLimit = require('express-rate-limit');

// General API rate limiting
const createRateLimiter = (options) => {
  return rateLimit({
    windowMs: options.windowMs || 15 * 60 * 1000, // 15 minutes
    max: options.max || 100,
    message: options.message || 'Too many requests from this IP',
    standardHeaders: true,
    legacyHeaders: false,
    keyGenerator: options.keyGenerator || ((req) => req.ip),
    skip: options.skip || (() => false),
    onLimitReached: options.onLimitReached || (() => {}),
    ...options
  });
};

// Authentication endpoints - strict limits
const authLimiter = createRateLimiter({
  windowMs: 15 * 60 * 1000, // 15 minutes
  max: 5, // 5 attempts per window
  message: {
    error: 'Too many authentication attempts',
    retryAfter: 15 * 60 // seconds
  },
  skipSuccessfulRequests: true
});

// General API endpoints
const apiLimiter = createRateLimiter({
  windowMs: 15 * 60 * 1000,
  max: 100,
  message: {
    error: 'Rate limit exceeded',
    retryAfter: 15 * 60
  }
});

// Apply to routes
app.use('/auth/login', authLimiter);
app.use('/auth/signup', authLimiter);
app.use('/auth/recover', authLimiter);
app.use('/api/', apiLimiter);

Advanced Rate Limiting Strategies

User-Based Rate Limiting

// Rate limiting based on authenticated user
const userBasedLimiter = createRateLimiter({
  windowMs: 15 * 60 * 1000,
  max: (req) => {
    // Different limits based on user role
    if (req.user?.role === 'premium') return 1000;
    if (req.user?.role === 'admin') return 10000;
    return 100; // default for regular users
  },
  keyGenerator: (req) => {
    // Use user ID if authenticated, otherwise IP
    return req.user?.id || req.ip;
  },
  skip: (req) => {
    // Skip rate limiting for admin users
    return req.user?.role === 'admin';
  }
});

Endpoint-Specific Rate Limiting

// Different limits for different endpoints
const endpointLimiters = {
  login: createRateLimiter({
    windowMs: 15 * 60 * 1000,
    max: 5,
    message: 'Too many login attempts'
  }),
  
  signup: createRateLimiter({
    windowMs: 60 * 60 * 1000, // 1 hour
    max: 3,
    message: 'Too many signup attempts'
  }),
  
  passwordReset: createRateLimiter({
    windowMs: 60 * 60 * 1000, // 1 hour
    max: 3,
    message: 'Too many password reset requests'
  }),
  
  emailVerification: createRateLimiter({
    windowMs: 5 * 60 * 1000, // 5 minutes
    max: 3,
    message: 'Too many verification emails sent'
  })
};

// Apply specific limiters
app.use('/auth/login', endpointLimiters.login);
app.use('/auth/signup', endpointLimiters.signup);
app.use('/auth/recover', endpointLimiters.passwordReset);
app.use('/auth/resend', endpointLimiters.emailVerification);

Redis-Based Rate Limiting

For distributed systems, use Redis to store rate limiting data.

Redis Implementation

const redis = require('redis');
const client = redis.createClient();

class RedisRateLimiter {
  constructor(maxRequests, windowMs, keyPrefix = 'rate_limit') {
    this.maxRequests = maxRequests;
    this.windowMs = windowMs;
    this.keyPrefix = keyPrefix;
  }
  
  async isAllowed(identifier) {
    const key = `${this.keyPrefix}:${identifier}`;
    const now = Date.now();
    const windowStart = Math.floor(now / this.windowMs) * this.windowMs;
    const windowKey = `${key}:${windowStart}`;
    
    try {
      // Use Redis pipeline for atomic operations
      const pipeline = client.pipeline();
      pipeline.incr(windowKey);
      pipeline.expire(windowKey, Math.ceil(this.windowMs / 1000));
      
      const results = await pipeline.exec();
      const currentCount = results[0][1];
      
      if (currentCount > this.maxRequests) {
        return {
          allowed: false,
          remaining: 0,
          resetTime: windowStart + this.windowMs,
          totalHits: currentCount
        };
      }
      
      return {
        allowed: true,
        remaining: this.maxRequests - currentCount,
        resetTime: windowStart + this.windowMs,
        totalHits: currentCount
      };
    } catch (error) {
      console.error('Redis rate limiter error:', error);
      // Fail open - allow request if Redis is down
      return { allowed: true, remaining: this.maxRequests, resetTime: now + this.windowMs };
    }
  }
}

// Usage with Express
const redisLimiter = new RedisRateLimiter(100, 15 * 60 * 1000);

const redisRateLimitMiddleware = async (req, res, next) => {
  const identifier = req.user?.id || req.ip;
  const result = await redisLimiter.isAllowed(identifier);
  
  // Set rate limit headers
  res.set({
    'X-RateLimit-Limit': redisLimiter.maxRequests,
    'X-RateLimit-Remaining': result.remaining,
    'X-RateLimit-Reset': new Date(result.resetTime).toISOString()
  });
  
  if (!result.allowed) {
    return res.status(429).json({
      error: 'Rate limit exceeded',
      retryAfter: Math.ceil((result.resetTime - Date.now()) / 1000)
    });
  }
  
  next();
};

Sliding Window with Redis

class RedisSlidingWindowLimiter {
  constructor(maxRequests, windowMs, keyPrefix = 'sliding_limit') {
    this.maxRequests = maxRequests;
    this.windowMs = windowMs;
    this.keyPrefix = keyPrefix;
  }
  
  async isAllowed(identifier) {
    const key = `${this.keyPrefix}:${identifier}`;
    const now = Date.now();
    const windowStart = now - this.windowMs;
    
    try {
      const pipeline = client.pipeline();
      
      // Remove old entries
      pipeline.zremrangebyscore(key, 0, windowStart);
      
      // Count current entries
      pipeline.zcard(key);
      
      // Add current request
      pipeline.zadd(key, now, `${now}-${Math.random()}`);
      
      // Set expiry
      pipeline.expire(key, Math.ceil(this.windowMs / 1000));
      
      const results = await pipeline.exec();
      const currentCount = results[1][1];
      
      if (currentCount >= this.maxRequests) {
        // Remove the request we just added since it's not allowed
        await client.zrem(key, `${now}-${Math.random()}`);
        
        return {
          allowed: false,
          remaining: 0,
          resetTime: now + this.windowMs
        };
      }
      
      return {
        allowed: true,
        remaining: this.maxRequests - currentCount - 1,
        resetTime: now + this.windowMs
      };
    } catch (error) {
      console.error('Redis sliding window limiter error:', error);
      return { allowed: true, remaining: this.maxRequests, resetTime: now + this.windowMs };
    }
  }
}

Rate Limiting Patterns

Hierarchical Rate Limiting

Implement multiple layers of rate limiting for comprehensive protection.

class HierarchicalRateLimiter {
  constructor() {
    this.limiters = {
      // Global rate limiting
      global: new RedisRateLimiter(10000, 60 * 1000), // 10k requests per minute globally
      
      // Per-IP rate limiting
      ip: new RedisRateLimiter(100, 15 * 60 * 1000), // 100 requests per 15 minutes per IP
      
      // Per-user rate limiting
      user: new RedisRateLimiter(1000, 15 * 60 * 1000), // 1000 requests per 15 minutes per user
      
      // Per-endpoint rate limiting
      endpoint: new Map([
        ['/auth/login', new RedisRateLimiter(5, 15 * 60 * 1000)],
        ['/auth/signup', new RedisRateLimiter(3, 60 * 60 * 1000)],
        ['/auth/recover', new RedisRateLimiter(3, 60 * 60 * 1000)]
      ])
    };
  }
  
  async checkLimits(req) {
    const checks = [];
    
    // Global check
    checks.push({
      name: 'global',
      result: await this.limiters.global.isAllowed('global')
    });
    
    // IP-based check
    checks.push({
      name: 'ip',
      result: await this.limiters.ip.isAllowed(req.ip)
    });
    
    // User-based check (if authenticated)
    if (req.user?.id) {
      checks.push({
        name: 'user',
        result: await this.limiters.user.isAllowed(req.user.id)
      });
    }
    
    // Endpoint-specific check
    const endpointLimiter = this.limiters.endpoint.get(req.path);
    if (endpointLimiter) {
      const identifier = req.user?.id || req.ip;
      checks.push({
        name: 'endpoint',
        result: await endpointLimiter.isAllowed(identifier)
      });
    }
    
    // Find the most restrictive limit
    const failedCheck = checks.find(check => !check.result.allowed);
    
    if (failedCheck) {
      return {
        allowed: false,
        reason: failedCheck.name,
        ...failedCheck.result
      };
    }
    
    // Return the most restrictive remaining count
    const minRemaining = Math.min(...checks.map(check => check.result.remaining));
    
    return {
      allowed: true,
      remaining: minRemaining,
      resetTime: Math.max(...checks.map(check => check.result.resetTime))
    };
  }
}

// Middleware implementation
const hierarchicalLimiter = new HierarchicalRateLimiter();

const hierarchicalRateLimitMiddleware = async (req, res, next) => {
  const result = await hierarchicalLimiter.checkLimits(req);
  
  res.set({
    'X-RateLimit-Remaining': result.remaining,
    'X-RateLimit-Reset': new Date(result.resetTime).toISOString()
  });
  
  if (!result.allowed) {
    return res.status(429).json({
      error: `Rate limit exceeded (${result.reason})`,
      retryAfter: Math.ceil((result.resetTime - Date.now()) / 1000)
    });
  }
  
  next();
};

Adaptive Rate Limiting

Dynamically adjust rate limits based on system load and user behavior.

class AdaptiveRateLimiter {
  constructor() {
    this.baseLimits = {
      login: 5,
      signup: 3,
      api: 100
    };
    
    this.systemLoad = 0; // 0-1 scale
    this.suspiciousIPs = new Set();
  }
  
  async getAdjustedLimit(endpoint, identifier, req) {
    let baseLimit = this.baseLimits[endpoint] || this.baseLimits.api;
    
    // Adjust based on system load
    const loadMultiplier = Math.max(0.1, 1 - this.systemLoad);
    baseLimit = Math.floor(baseLimit * loadMultiplier);
    
    // Reduce limits for suspicious IPs
    if (this.suspiciousIPs.has(req.ip)) {
      baseLimit = Math.floor(baseLimit * 0.5);
    }
    
    // Increase limits for trusted users
    if (req.user?.trusted) {
      baseLimit = Math.floor(baseLimit * 2);
    }
    
    // Reduce limits during peak hours
    const hour = new Date().getHours();
    if (hour >= 9 && hour <= 17) { // Business hours
      baseLimit = Math.floor(baseLimit * 0.8);
    }
    
    return Math.max(1, baseLimit); // Ensure at least 1 request is allowed
  }
  
  async updateSystemLoad() {
    // Monitor system metrics and update load
    // This could integrate with monitoring systems
    const metrics = await this.getSystemMetrics();
    this.systemLoad = (metrics.cpu + metrics.memory + metrics.responseTime) / 3;
  }
  
  markSuspiciousIP(ip) {
    this.suspiciousIPs.add(ip);
    // Remove after 1 hour
    setTimeout(() => this.suspiciousIPs.delete(ip), 60 * 60 * 1000);
  }
}

Rate Limiting Headers

Implement standard rate limiting headers for client transparency.

function setRateLimitHeaders(res, limiter, result) {
  const headers = {
    'X-RateLimit-Limit': limiter.maxRequests,
    'X-RateLimit-Remaining': result.remaining,
    'X-RateLimit-Reset': Math.ceil(result.resetTime / 1000),
    'X-RateLimit-Reset-After': Math.ceil((result.resetTime - Date.now()) / 1000)
  };
  
  // Add retry-after header if rate limited
  if (!result.allowed) {
    headers['Retry-After'] = Math.ceil((result.resetTime - Date.now()) / 1000);
  }
  
  res.set(headers);
}

// Enhanced middleware with proper headers
const rateLimitWithHeaders = (limiter) => {
  return async (req, res, next) => {
    const identifier = req.user?.id || req.ip;
    const result = await limiter.isAllowed(identifier);
    
    setRateLimitHeaders(res, limiter, result);
    
    if (!result.allowed) {
      return res.status(429).json({
        error: 'Rate limit exceeded',
        message: 'Too many requests, please try again later',
        retryAfter: Math.ceil((result.resetTime - Date.now()) / 1000)
      });
    }
    
    next();
  };
};

Monitoring and Analytics

Rate Limiting Metrics

class RateLimitingMetrics {
  constructor() {
    this.metrics = {
      totalRequests: 0,
      blockedRequests: 0,
      byEndpoint: new Map(),
      byIP: new Map(),
      byUser: new Map()
    };
  }
  
  recordRequest(endpoint, identifier, blocked = false) {
    this.metrics.totalRequests++;
    
    if (blocked) {
      this.metrics.blockedRequests++;
    }
    
    // Track by endpoint
    const endpointStats = this.metrics.byEndpoint.get(endpoint) || { total: 0, blocked: 0 };
    endpointStats.total++;
    if (blocked) endpointStats.blocked++;
    this.metrics.byEndpoint.set(endpoint, endpointStats);
    
    // Track by identifier
    const identifierStats = this.metrics.byIP.get(identifier) || { total: 0, blocked: 0 };
    identifierStats.total++;
    if (blocked) identifierStats.blocked++;
    this.metrics.byIP.set(identifier, identifierStats);
  }
  
  getMetrics() {
    return {
      ...this.metrics,
      blockRate: this.metrics.totalRequests > 0 
        ? (this.metrics.blockedRequests / this.metrics.totalRequests) * 100 
        : 0
    };
  }
  
  getTopBlockedIPs(limit = 10) {
    return Array.from(this.metrics.byIP.entries())
      .sort((a, b) => b[1].blocked - a[1].blocked)
      .slice(0, limit);
  }
}

// Integration with rate limiting middleware
const metrics = new RateLimitingMetrics();

const monitoredRateLimit = (limiter) => {
  return async (req, res, next) => {
    const identifier = req.user?.id || req.ip;
    const result = await limiter.isAllowed(identifier);
    
    metrics.recordRequest(req.path, identifier, !result.allowed);
    
    setRateLimitHeaders(res, limiter, result);
    
    if (!result.allowed) {
      return res.status(429).json({
        error: 'Rate limit exceeded',
        retryAfter: Math.ceil((result.resetTime - Date.now()) / 1000)
      });
    }
    
    next();
  };
};

Best Practices

Configuration Management

Environment-Based Limits

Configure different rate limits for development, staging, and production

Graceful Degradation

Fail open when rate limiting infrastructure is unavailable

Clear Error Messages

Provide helpful error messages with retry information

Monitoring Integration

Track rate limiting metrics and set up alerts

Implementation Guidelines

// Environment-based configuration
const getRateLimitConfig = () => {
  const env = process.env.NODE_ENV;
  
  const configs = {
    development: {
      auth: { max: 100, windowMs: 15 * 60 * 1000 },
      api: { max: 1000, windowMs: 15 * 60 * 1000 }
    },
    staging: {
      auth: { max: 10, windowMs: 15 * 60 * 1000 },
      api: { max: 200, windowMs: 15 * 60 * 1000 }
    },
    production: {
      auth: { max: 5, windowMs: 15 * 60 * 1000 },
      api: { max: 100, windowMs: 15 * 60 * 1000 }
    }
  };
  
  return configs[env] || configs.production;
};

// Graceful error handling
const createResilientLimiter = (config) => {
  return async (req, res, next) => {
    try {
      const result = await rateLimiter.isAllowed(req.ip);
      
      if (!result.allowed) {
        return res.status(429).json({
          error: 'Rate limit exceeded',
          message: 'Please wait before making another request',
          retryAfter: Math.ceil((result.resetTime - Date.now()) / 1000)
        });
      }
      
      next();
    } catch (error) {
      console.error('Rate limiter error:', error);
      // Log error but allow request to proceed
      next();
    }
  };
};

Testing Rate Limits

// Jest test for rate limiting
describe('Rate Limiting', () => {
  test('should block requests after limit exceeded', async () => {
    const limiter = new FixedWindowLimiter(3, 60000); // 3 requests per minute
    
    // First 3 requests should be allowed
    for (let i = 0; i < 3; i++) {
      const result = limiter.isAllowed('test-ip');
      expect(result.allowed).toBe(true);
    }
    
    // 4th request should be blocked
    const blockedResult = limiter.isAllowed('test-ip');
    expect(blockedResult.allowed).toBe(false);
    expect(blockedResult.remaining).toBe(0);
  });
  
  test('should reset after window expires', async () => {
    const limiter = new FixedWindowLimiter(1, 100); // 1 request per 100ms
    
    // First request allowed
    expect(limiter.isAllowed('test-ip').allowed).toBe(true);
    
    // Second request blocked
    expect(limiter.isAllowed('test-ip').allowed).toBe(false);
    
    // Wait for window to reset
    await new Promise(resolve => setTimeout(resolve, 150));
    
    // Request should be allowed again
    expect(limiter.isAllowed('test-ip').allowed).toBe(true);
  });
});

Rate limiting is essential for maintaining service stability, preventing abuse, and ensuring fair resource allocation. The Strike Auth Service implements multiple rate limiting strategies to provide comprehensive protection while maintaining a good user experience.

Get Started

Core Concepts

Guides

Overview

Rate Limiting Strategies

Fixed Window Rate Limiting

Sliding Window Rate Limiting

Token Bucket Rate Limiting

Implementation with Express.js

Basic Rate Limiting Middleware

Advanced Rate Limiting Strategies

User-Based Rate Limiting

Endpoint-Specific Rate Limiting

Redis-Based Rate Limiting

Redis Implementation

Sliding Window with Redis

Rate Limiting Patterns

Hierarchical Rate Limiting

Adaptive Rate Limiting

Rate Limiting Headers

Monitoring and Analytics

Rate Limiting Metrics

Best Practices

Configuration Management

Environment-Based Limits

Graceful Degradation

Clear Error Messages

Monitoring Integration

Implementation Guidelines

Testing Rate Limits

Get Started

Core Concepts

Guides

​Overview

​Rate Limiting Strategies

​Fixed Window Rate Limiting

​Sliding Window Rate Limiting

​Token Bucket Rate Limiting

​Implementation with Express.js

​Basic Rate Limiting Middleware

​Advanced Rate Limiting Strategies

​User-Based Rate Limiting

​Endpoint-Specific Rate Limiting

​Redis-Based Rate Limiting

​Redis Implementation

​Sliding Window with Redis

​Rate Limiting Patterns

​Hierarchical Rate Limiting

​Adaptive Rate Limiting

​Rate Limiting Headers

​Monitoring and Analytics

​Rate Limiting Metrics

​Best Practices

​Configuration Management

Environment-Based Limits

Graceful Degradation

Clear Error Messages

Monitoring Integration

​Implementation Guidelines

​Testing Rate Limits

Overview

Rate Limiting Strategies

Fixed Window Rate Limiting

Sliding Window Rate Limiting

Token Bucket Rate Limiting

Implementation with Express.js

Basic Rate Limiting Middleware

Advanced Rate Limiting Strategies

User-Based Rate Limiting

Endpoint-Specific Rate Limiting

Redis-Based Rate Limiting

Redis Implementation

Sliding Window with Redis

Rate Limiting Patterns

Hierarchical Rate Limiting

Adaptive Rate Limiting

Rate Limiting Headers

Monitoring and Analytics

Rate Limiting Metrics

Best Practices

Configuration Management

Implementation Guidelines

Testing Rate Limits