Limitador is a generic rate-limiter written in Rust. It can be used as a library, as an HTTP service, or as a GRPC service that implements the Envoy Rate Limit protocol.
Status: Experimental
Add this to your Cargo.toml:
toml
[dependencies]
limitador = { version = "0.1.0" }
To use limitador in a project that compiles to WASM, there are some features
that need to be disabled. Add this to your Cargo.toml instead:
toml
[dependencies]
limitador = { version = "0.1.0", default-features = false }
The OpenAPI spec of the service is here.
Run with Docker (replace latest with the version you want):
bash
docker run --rm --net=host -it quay.io/3scale/limitador:latest http-server
To use Redis, specify the URL with REDIS_URL:
bash
docker run -e REDIS_URL=redis://127.0.0.1:6379 --rm --net=host -it quay.io/3scale/limitador:latest http-server
You can also run the service locally:
bash
cargo run --release --bin http-server
You can change the host and port with the HOST and PORT envs.
To run Limitador, you need to provide a YAML file with the limits. There's an
example file that allows 10 requests per minute and per
user_id when the HTTP method is "GET" and 5 when it is a "POST". You can run it
with Docker (replace latest with the version you want):
bash
docker run -e LIMITS_FILE=/home/limitador/my_limits.yaml --rm --net=host -it -v $(pwd)/examples/limits.yaml:/home/limitador/my_limits.yaml:ro quay.io/3scale/limitador:latest envoy-rls
To use Redis, specify the URL with REDIS_URL:
bash
docker run -e LIMITS_FILE=/home/limitador/my_limits.yaml -e REDIS_URL=redis://127.0.0.1:6379 --rm --net=host -it -v $(pwd)/examples/limits.yaml:/home/limitador/my_limits.yaml:ro quay.io/3scale/limitador:latest envoy-rls
You can also run the service locally:
bash
LIMITS_FILE=./examples/limits.yaml cargo run --release --bin envoy-rls
There's a minimal Envoy config to try limitador here. The config forwards the "userid" header and the request method to Limitador. It assumes that there's an upstream API deployed in the port 1323. You can use echo, for example.
You can change the host and port with the HOST and PORT envs.
Limitador can store its limits and counters in memory or in Redis. In memory is faster, but the limits are applied per instance. When using Redis, multiple instances of limitador can share the same limits, but it's slower.
bash
cargo build
Some tests need a redis deployed in localhost:6379. You can run it in Docker with:
bash
docker run --rm --net=host -it redis
Then, run the tests:
bash
cargo test