Comprehensive Guide to Alibaba Sentinel: Installation, Configuration, Flow Control, Degrade Rules, Hotspot Parameters, Persistence, Cluster Flow Control and More

1. Introduction

This article introduces Alibaba's open‑source traffic‑guard Sentinel, a mature project refined through nearly ten years of Double‑11 traffic peaks.

2. What is Sentinel?

Sentinel, literally "sentinel", is a traffic‑defense guard for microservices. It protects service stability from traffic control, circuit breaking, and system load perspectives.

Key features include:

Rich application scenarios : handles flash sales, message throttling, cluster traffic control, real‑time circuit breaking, etc.

Real‑time monitoring : per‑machine second‑level metrics and cluster‑wide summaries.

Broad open‑source ecosystem : integrates with Spring Cloud, Dubbo, gRPC, Quarkus, and supports Java/Go/C++.

Complete SPI extension mechanism : custom rule management, dynamic data source adaptation.

3. Sentinel vs. Hystrix

Sentinel outperforms Hystrix in many aspects; the article provides a visual comparison.

4. Selecting Sentinel Version

When using Spring Cloud Alibaba, the version aligns with spring-cloud-alibaba-dependencies 2.2.1.RELEASE, so Sentinel 1.7.1 is recommended.

5. Installing the Sentinel Dashboard

Download the appropriate JAR from the GitHub releases page and run:

java -Dserver.port=8080 -Dcsp.sentinel.dashboard.server=localhost:8080 -Dproject.name=sentinel-dashboard -jar sentinel-dashboard-1.7.1.jar

Key parameters:

-Dserver.port : dashboard port (default 8080).

-Dproject.name : service name.

-Dcsp.sentinel.dashboard.server : dashboard address for registration.

Default login: sentinel/sentinel . JDK >= 1.8 is required.

6. Integrating Microservices with Sentinel

Register the service with Nacos, add the Sentinel starter dependency, and configure the dashboard address:

spring:
  cloud:
    sentinel:
      transport:
        dashboard: localhost:8080

Create a test controller:

@RestController
@RequestMapping("/sentinel")
public class FlowLimitController {
    @GetMapping("/test")
    public String test() {
        return "Received a message--------";
    }
}

Start the microservice (e.g., port 9008) and verify it appears in the dashboard.

7. Configuring Flow Control

Flow control monitors QPS or concurrent threads and limits traffic when thresholds are reached.

Each rule consists of:

resource : target name.

count : threshold.

grade : 1 for QPS, 0 for threads.

limitApp : caller origin (default default ).

strategy : 0 direct, 1 associated, 2 chain.

controlBehavior : 0 fast‑fail, 1 warm‑up, 2 queue‑waiting.

Three flow‑control effects:

Fast Fail

Requests exceeding the threshold are immediately rejected with a FlowException .

Warm‑Up

Gradually ramps up traffic using a token‑bucket algorithm; implemented by WarmUpController .

Queue‑Waiting

Uses a leaky‑bucket algorithm to queue requests; implemented by RateLimiterController .

8. Degrade (Circuit‑Breaker) Rules

Three degrade strategies:

Average response time (RT) – triggers when the average RT exceeds count ms within 1 s.

Exception ratio – triggers when the exception ratio exceeds count (0.0‑1.0) after at least 5 QPS.

Exception count – triggers when the number of exceptions in the last minute exceeds count .

Example: a 3‑second sleep endpoint exceeds a 200 ms RT threshold and is circuit‑broken.

9. Hotspot Parameter Limiting

Limits frequently accessed parameter values (e.g., product IDs) using LRU + token bucket. Only the specified parameter index is subject to limiting.

10. System Adaptive Flow Control

Protects the whole system based on load, CPU usage, average RT, entry QPS, or concurrent threads. Example shown for entry‑QPS.

11. Custom Block Handler

Use @SentinelResource with blockHandler (or blockHandlerClass ) to return custom messages when a rule blocks a request.

12. Exception Fallback (Degrade) Handling

Define fallback , fallbackClass , or defaultFallback in @SentinelResource to handle business exceptions and degrade gracefully.

13. Black/White List

Configure AuthorityRule to allow or deny requests based on origin (e.g., IP). Implement RequestOriginParser to extract the origin.

14. Rule Persistence

Sentinel stores rules in memory by default. Use Nacos as a configuration center with the sentinel-datasource-nacos dependency. Example application.yml snippet:

spring:
  cloud:
    sentinel:
      datasource:
        ds-flow:
          nacos:
            server-addr: 127.0.0.1:8848
            dataId: ${spring.application.name}-flow
            groupId: DEFAULT_GROUP
            data-type: json
            rule-type: flow
        ds-degrade:
          nacos:
            server-addr: 127.0.0.1:8848
            dataId: ${spring.application.name}-degrade
            groupId: DEFAULT_GROUP
            data-type: json
            rule-type: degrade

JSON formats for flow, degrade, hotspot, system, and authority rules are provided in the article.

15. Push Mode Synchronization

Push mode automatically pushes Nacos changes to the dashboard. To push dashboard changes back to Nacos, modify Sentinel’s dashboard source code (replace test‑scope dependencies, copy Nacos rule provider/publisher to main, adjust @Qualifier beans, and configure NacosConfigUtil ).

16. Cluster Flow Control

Two roles: Token Client and Token Server. Modes: Alone (stand‑alone server) and Embedded (server inside each instance). Example: three instances (ports 9009, 9011, 9013) with one designated as token server via the dashboard.

17. Gateway Flow Control

Configuration details are omitted; refer to Sentinel’s official gateway documentation.

18. OpenFeign Integration for Fault Tolerance

See the previous OpenFeign article for detailed steps.

Conclusion

If you find this tutorial helpful, please like, share, comment, and bookmark. The full source code is available on GitHub; reply with keyword 9528 to the public account "码猿技术专栏" to obtain it.