At Quicknode, we understand the importance of ensuring that our customers' applications remain responsive and available at all times. That's why we have implemented a request-per-second (RPS) limit based on your pricing plan -- Pricing page

Exceeding this limit will result in a 429 Error,429 Client Error: Too Many Requests. You can view these errors on your Metrics (Endpoints > endpoint-name-example > Metrics)

Remember that batching also counts as RPS because each request is counted individually. Sending a batch with 26 requests while your RPS is 25 will result in 25 valid responses and one (1) error. The JSON structure will still be valid, but you might run into parsing errors in your code if you only expect valid responses and not 429 errors. 

Preventing the error

You can deal with these 429 errors in many different ways:

1. Monitor the RPS usage: benchmark your code and adjust accordingly;
2. Implement a backoff logic: when experiencing the 429 error, wait a period of time before resuming;
3. Use a third-party package to throttle based on a fixed number, such as
   • https://pypi.org/project/ratelimit/
   • https://www.npmjs.com/package/axios-rate-limit
   • https://github.com/uber-go/ratelimit