Brightdata Rate Limits

Bright Data Rate Limits

Overview

Handle Bright Data rate limits and concurrent request limits. Unlike traditional API rate limits, Bright Data limits are per-zone and based on concurrent connections and requests per second. The Web Scraper API trigger endpoint is limited to 20 requests/min and 60 requests/hour.

Prerequisites

• Bright Data zone configured
• Understanding of async/await patterns
• p-queue or similar concurrency library

Instructions

Step 1: Understand Bright Data Rate Limits

Step 2: Implement Concurrent Request Limiter

// src/brightdata/limiter.ts
import PQueue from 'p-queue';

// Match concurrency to your Bright Data plan limits
const scrapeQueue = new PQueue({
  concurrency: 10,       // Max concurrent proxy requests
  interval: 1000,        // Per second
  intervalCap: 20,       // Max 20 requests per second
  timeout: 120000,       // Kill after 2 min
  throwOnTimeout: true,
});

export async function queuedScrape(url: string): Promise<string> {
  return scrapeQueue.add(async () => {
    const client = getBrightDataClient();
    const response = await client.get(url);
    return response.data;
  });
}

// Monitor queue health
scrapeQueue.on('active', () => {
  console.log(`Queue: ${scrapeQueue.size} waiting, ${scrapeQueue.pending} active`);
});

Step 3: Exponential Backoff for Proxy Errors

// src/brightdata/backoff.ts
export async function scrapeWithBackoff(
  url: string,
  config = { maxRetries: 5, baseDelay: 2000, maxDelay: 60000 }
): Promise<string> {
  for (let attempt = 0; attempt <= config.maxRetries; attempt++) {
    try {
      const client = getBrightDataClient();
      const response = await client.get(url);
      return response.data;
    } catch (error: any) {
      const status = error.response?.status;
      const luminatiError = error.response?.headers?.['x-luminati-error'];

      // Only retry on transient errors
      const retryable = [502, 503, 429].includes(status)
        || error.code === 'ETIMEDOUT'
        || luminatiError === 'ip_banned';

      if (attempt === config.maxRetries || !retryable) throw error;

      const delay = Math.min(
        config.baseDelay * Math.pow(2, attempt) + Math.random() * 1000,
        config.maxDelay
      );
      console.log(`[${luminatiError || status}] Retry ${attempt + 1} in ${delay.toFixed(0)}ms`);
      await new Promise(r => setTimeout(r, delay));
    }
  }
  throw new Error('Unreachable');
}

Step 4: Web Scraper API Rate Limiter

// src/brightdata/trigger-limiter.ts — 20/min, 60/hr for trigger endpoint
const triggerQueue = new PQueue({
  concurrency: 1,
  interval: 60000,       // Per minute
  intervalCap: 20,       // 20 triggers per minute
});

let hourlyCount = 0;
setInterval(() => { hourlyCount = 0; }, 3600000); // Reset hourly

export async function rateLimitedTrigger(
  datasetId: string,
  urls: string[]
): Promise<any> {
  if (hourlyCount >= 55) { // Leave buffer
    throw new Error('Approaching hourly trigger limit (60/hr). Wait before triggering.');
  }

  return triggerQueue.add(async () => {
    hourlyCount++;
    const response = await fetch(
      `https://api.brightdata.com/datasets/v3/trigger?dataset_id=${datasetId}&format=json`,
      {
        method: 'POST',
        headers: {
          'Authorization': `Bearer ${process.env.BRIGHTDATA_API_TOKEN}`,
          'Content-Type': 'application/json',
        },
        body: JSON.stringify(urls.map(url => ({ url }))),
      }
    );
    return response.json();
  });
}

Step 5: Batch URLs to Minimize Triggers

// Instead of triggering per-URL, batch into single triggers
async function batchTrigger(urls: string[], batchSize = 100) {
  const batches = [];
  for (let i = 0; i < urls.length; i += batchSize) {
    batches.push(urls.slice(i, i + batchSize));
  }

  console.log(`Triggering ${urls.length} URLs in ${batches.length} batches`);

  for (const batch of batches) {
    await rateLimitedTrigger('gd_dataset_id', batch);
  }
}

Output

• Concurrent request limiter matching plan limits
• Exponential backoff handling X-Luminati error headers
• Web Scraper API trigger rate limiter (20/min, 60/hr)
• URL batching to minimize trigger count

Examples

Start a newly approved workload below the documented account limit, persist an idempotency key before dispatch, and honor provider retry signals. On throttling, pause the scoped queue, capture a redacted receipt, and reconcile prior attempts before resuming; do not add zones or increase concurrency simply to evade a limit.

Error Handling

Resources

• Bright Data Proxy Limits
• Web Scraper API Limits
• p-queue Documentation

Next Steps

For security configuration, see brightdata-security-basics.