Feign timeout doesn't work properly, the response time is far more bigger (approx 1 sec) than the configured timeout (20 ms)

Hello,

I am using AsyncFeign in a project

Resilience4j version: 1.7.1 or 2.0.2
Spring Boot Version: 3.0.0
Java version: 17
spring-cloud-dependencies: 2022.0.0-RC3
OpenFeign: 4.0.0-RC3
io.github.openfeign:feign-hc5:12.1

We have set the Feign Connection timeout to 10ms and Read timeout to 20 ms. The timeout works and we get a

feign.RetryableException: 20 MILLISECONDS executing GET http://localhost:8082/unstable

but the response is not returned until approx 1030 ms afterward (after more than 1 second in total).

You can see an example of the code in this repo, under the module graphql-java-kickstart-example (ProductClient), it's calling an unstable service also included under the module unstable-service.

This is the current configuration

Thank you in advance

Reproduce the timeout behavior using the linked graphql-java-kickstart-example, starting with ProductClient and its application.yml configuration, then inspect the linked unstable-service. Compare the configured 10 ms connection and 20 ms read timeouts with the observed exception and total response time. Done means identifying and correcting the source of the extra delay, with the timeout returning within the configured bounds.