Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.loadforge.com/llms.txt

Use this file to discover all available pages before exploring further.

This guide shows how to handle API rate limits with simple backoff strategies. Perfect for APIs that return 429 status codes when limits are exceeded.

​
Use Cases

  • Test API rate limiting behavior
  • Handle 429 responses gracefully
  • Implement basic backoff strategies
  • Monitor rate limit headers

​
Simple Implementation

from locust import task, HttpUser
import time
import random

class RateLimitUser(HttpUser):
    def on_start(self):
        self.backoff_until = 0  # Track when to resume requests
        self.rate_limit_count = 0

    @task(5)
    def test_api_with_rate_limits(self):
        """Test API endpoint that has rate limits"""
        
        # Check if we're in backoff period
        if time.time() < self.backoff_until:
            print("Skipping request - in backoff period")
            return
        
        with self.client.get("/api/data", name="Rate Limited API") as response:
            if response.status_code == 200:
                print("Request successful")
                
                # Check rate limit headers
                remaining = response.headers.get("X-RateLimit-Remaining")
                if remaining:
                    print(f"Requests remaining: {remaining}")
                    
                    # Warn if getting close to limit
                    if int(remaining) < 5:
                        print("WARNING: Approaching rate limit!")
                        
            elif response.status_code == 429:
                # Rate limited - implement backoff
                print("Rate limited! Backing off...")
                self.rate_limit_count += 1
                
                # Get retry-after header or use default
                retry_after = response.headers.get("Retry-After", "60")
                backoff_time = int(retry_after)
                
                # Set backoff period
                self.backoff_until = time.time() + backoff_time
                print(f"Backing off for {backoff_time} seconds")
                
                response.failure(f"Rate limited - backing off for {backoff_time}s")
            else:
                response.failure(f"Request failed: {response.status_code}")

    @task(2)
    def test_search_api(self):
        """Test search API with rate limits"""
        
        if time.time() < self.backoff_until:
            return
        
        search_terms = ["test", "product", "user", "data"]
        query = random.choice(search_terms)
        
        with self.client.get(f"/api/search?q={query}", name="Search API") as response:
            if response.status_code == 200:
                print(f"Search for '{query}' successful")
            elif response.status_code == 429:
                print(f"Search rate limited")
                
                # Simple backoff
                self.backoff_until = time.time() + 30  # 30 second backoff
                response.failure("Search rate limited")
            else:
                response.failure(f"Search failed: {response.status_code}")

    @task(1)
    def check_rate_limit_status(self):
        """Check current rate limit status"""
        
        if time.time() < self.backoff_until:
            remaining_backoff = self.backoff_until - time.time()
            print(f"In backoff for {remaining_backoff:.0f} more seconds")
        else:
            print("No active backoff - ready for requests")
            
        if self.rate_limit_count > 0:
            print(f"Total rate limits hit: {self.rate_limit_count}")

​
Setup Instructions

  1. Replace /api/data and /api/search with your actual API endpoints
  2. Adjust backoff times based on your API’s requirements
  3. Customize rate limit header names for your API
  4. Test with different request volumes to trigger rate limits

​
What This Tests

  • Rate Limit Detection: Handles 429 status codes
  • Backoff Strategy: Waits before retrying after rate limits
  • Header Monitoring: Checks remaining request counts
  • Recovery: Resumes requests after backoff period

​
Rate Limit Headers

Common headers to check:
  • X-RateLimit-Remaining: Requests left in current window
  • Retry-After: Seconds to wait before retrying
  • X-RateLimit-Reset: When the limit resets

​
Simple Backoff Strategy

When rate limited (429 response):
  1. Stop making requests
  2. Wait for the time specified in Retry-After header
  3. Resume requests after backoff period
  4. Track how many times you hit limits

​
Best Practices

  • Always respect the Retry-After header
  • Monitor remaining requests to avoid hitting limits
  • Use exponential backoff for repeated rate limits
  • Don’t retry immediately after being rate limited