قواطع الدائرة باستخدام Resilience4j

Microservices are great, but they also bring new challenges. When one service fails, it shouldn't take down the whole system. This is where resilience comes in.

Building resilient systems means they can recover gracefully from failures and continue to function, even if in a degraded mode. It's crucial for maintaining service availability and user experience.

Imagine a chain of services: Service A calls B, which calls C. If Service C becomes slow or unresponsive, Service B will wait, consuming resources. Then Service A will wait for B, and so on.

Eventually, all services in the chain might exhaust their resources (like threads or connections) and fail, leading to a complete system outage. This is a cascading failure, a common problem in distributed systems.

The circuit breaker pattern helps prevent cascading failures. It's like an electrical circuit breaker in your house: if there's an overload, it 'trips' and cuts off the power to prevent damage.

In software, a circuit breaker wraps a function call to a potentially failing service. If calls to that service repeatedly fail, the circuit breaker 'opens,' preventing further calls to the failing service and returning an error immediately.

Resilience4j is a lightweight, easy-to-use fault-tolerance library for Java and Kotlin. It provides several patterns to make your applications more resilient, including:

• Circuit Breaker: Prevents repeated calls to failing services.
• Retry: Automatically retries failed operations.
• Rate Limiter: Controls the rate of requests to a service.

In this lesson, we'll focus specifically on the Circuit Breaker pattern.

First, let's add the necessary dependency to our Spring Boot project. For Maven, you'd add this to your pom.xml:

<dependency>
<groupId>io.github.resilience4j</groupId>
<artifactId>resilience4j-spring-boot2</artifactId>
<version>2.2.0</version> <!-- Use latest stable -->
</dependency>

This dependency brings in the core Resilience4j modules and its Spring Boot integration for easy configuration.

We can configure circuit breakers in our application.yml or application.properties. Here's a basic example for a service named 'myExternalService':

resilience4j.circuitbreaker:
instances:
myExternalService:
failureRateThreshold: 50
waitDurationInOpenState: 5s
slidingWindowType: COUNT_BASED
slidingWindowSize: 10

• failureRateThreshold: If 50% of calls fail, the circuit opens.
• waitDurationInOpenState: How long the circuit stays open (5 seconds).
• slidingWindowType & slidingWindowSize: Metrics are collected over the last 10 calls.

Now, let's protect a service call using the @CircuitBreaker annotation. This tells Spring to apply the configured circuit breaker named "myExternalService" to this method.

Try running this example. If callExternalService() throws an exception frequently (simulated here), the circuit will eventually open and reject calls immediately.

A circuit breaker operates in three main states:

• CLOSED: Normal operation. Calls go through. If failures exceed a threshold, it transitions to OPEN.
• OPEN: Calls are immediately rejected with an error. After a waitDurationInOpenState, it transitions to HALF_OPEN.
• HALF_OPEN: A limited number of test calls are allowed through. If these succeed, it transitions back to CLOSED. If they fail, it returns to OPEN.

This cycle prevents overwhelming a struggling service while allowing it to recover.

When a circuit breaker is OPEN, or if a call fails for other reasons, we don't want to just return a generic error. We can provide a fallback method to return a default response or perform alternative logic.

Resilience4j's @CircuitBreaker annotation allows you to specify a fallback method that will be called if the primary method fails or the circuit is open. This ensures a more graceful degradation of service.

Let's enhance our previous example with a fallback method. When the circuit is open or callExternalService() fails, fallbackMethod() will be executed instead.

Run this and try to trigger the circuit breaker. Notice how the fallback message is returned when the external service fails or the circuit is open, providing a better user experience.

Consider a circuit breaker configured with failureRateThreshold: 70 and slidingWindowSize: 10. If 8 out of the last 10 consecutive calls fail while the circuit is CLOSED, what is the most likely immediate next state of the circuit breaker?

We've learned how circuit breakers, especially with Resilience4j, are vital for building resilient microservices. They prevent cascading failures by stopping repeated calls to failing services.

Key takeaways:

• Circuit breakers protect services from overload and unresponsiveness.
• They operate in CLOSED, OPEN, and HALF_OPEN states.
• Fallbacks provide graceful degradation when a service is unavailable.

Next, we'll explore other resilience patterns like implementing fallbacks and timeouts more deeply.