Question 1

How does Limiters compare to other Go rate limiting libraries like tollbooth?

Accepted Answer

Limiters focuses on distributed rate limiting with pluggable backends and multiple algorithms, making it better for microservices needing consistency across instances. In contrast, tollbooth is simpler and more HTTP-centric, suited for basic API rate limiting without distributed requirements.

Question 2

Can I use Limiters with a standard HTTP server, not just gRPC?

Accepted Answer

Yes, Limiters is algorithm-agnostic and can be integrated into any Go application. You can wrap HTTP handlers with limit checks similar to the gRPC interceptor example provided in the README, using the same limiter instances for request validation.

Question 3

What happens if the distributed lock service (like etcd) goes down?

Accepted Answer

If the lock service fails, algorithms requiring locks might not function correctly, leading to potential rate limit breaches or errors. The library includes error handling in examples, but developers must monitor and handle such failures gracefully, possibly with fallbacks or alerts.

Question 4

How to choose between Token Bucket and Sliding Window for API rate limiting?

Accepted Answer

Token Bucket allows bursts and precise control but requires locks, while Sliding Window smooths bursts and is stricter at boundaries but uses more memory. For APIs with variable traffic, Sliding Window might be better to prevent boundary exploits, as explained in the algorithm descriptions.

Question 5

Is there support for rate limiting based on user IP or other attributes?

Accepted Answer

Yes, the README shows an example of IP-based rate limiting in the gRPC examples. You can create separate limiter instances per attribute or use keys in the backend to differentiate limits, enabling flexible per-client or per-resource rate limiting.

Question 6

How to set up DynamoDB backend for Limiters?

Accepted Answer

You need to create a DynamoDB table with a string partition key, and for some algorithms, a sort key and TTL. Use the LoadDynamoDBTableProperties function to validate the table schema, as described in the DynamoDB section, ensuring compatibility with the chosen limiter backend.

limiters

What is limiters?

Overview

Use Cases

Best For

Related Projects

Found a gem we're missing?

Not Ideal For

Pros & Cons

Pros

Cons

Frequently Asked Questions