點擊▲關注 “爪哇筆記” 給公衆號標星置頂
更多精彩 第一時間直達
前言
妹子圖架構發佈以後，不少小夥伴在後臺問路由網關、鑑權、限流怎麼搞得？這不熱乎乎的教程馬上就出來了。
架構
架構基本這麼一個組合，如果是個人開發或者小團隊項目，其實各個組件服務完全可以單機部署，因爲集羣費銀子也沒必要。
基礎組件
路由網關：
GateWay註冊中心：
Nacos負載均衡：
Ribbon熔斷器：
Hystrix
配置
網關 pom.xml 引入：
<dependencies>
<dependency>
<groupId>org.springframework.cloud</groupId>
<artifactId>spring-cloud-starter-gateway</artifactId>
</dependency>
<dependency>
<groupId>org.springframework.cloud</groupId>
<artifactId>spring-cloud-starter-netflix-hystrix</artifactId>
</dependency>
<dependency>
<groupId>com.alibaba.cloud</groupId>
<artifactId>spring-cloud-starter-alibaba-nacos-discovery</artifactId>
</dependency>
<dependency>
<groupId>com.alibaba.cloud</groupId>
<artifactId>spring-cloud-starter-alibaba-nacos-config</artifactId>
</dependency>
</dependencies>
注意 Gateway 默認使用的是 webflux，不要引入 web，否則啓動會報錯。
啓動配置 bootstrap.yml，請自行安裝 Nacos：
server:
port: 8080
spring:
application:
name: tools-gateway
cloud:
nacos:
config:
server-addr: 127.0.0.1:8848
discovery:
server-addr: 127.0.0.1:8848
gateway:
routes:
- id: sys
uri: lb://tools-sys
predicates:
- Path=/api/sys/**
filters:
- StripPrefix=2
- name: Hystrix
args:
name: fallback
fallbackUri: forward:/fallback # 熔斷回調
- id: weChat
uri: lb://tools-meizi
predicates:
- Path=/api/meizi/**
filters:
- StripPrefix=2
# 跨域請求
filter:
remove-hop-by-hop:
headers:
- trailer
- te
- keep-alive
- transfer-encoding
- upgrade
- proxy-authenticate
- connection
- proxy-authorization
- x-application-context
- access-control-allow-credentials
- access-control-allow-headers
- access-control-allow-methods
- access-control-allow-origin
- access-control-max-age
- vary
globalcors:
corsConfigurations:
'[/**]':
allowCredentials: true
allowedHeaders: '*'
allowedMethods: '*'
allowedOrigins: '*'
maxAge: 3628800
# 熔斷
hystrix:
command:
default:
circuitBreaker:
enabled: true
errorThresholdPercentage: 50
forceClosed: false
forceOpen: false
requestVolumeThreshold: 4
sleepWindowInMilliseconds: 10000
execution:
isolation:
semaphore:
maxConcurrentRequests: 2
strategy: SEMAPHORE
thread:
timeoutInMilliseconds: 3000
metrics:
healthSnapshot:
intervalInMilliseconds: 500
rollingPercentile:
bucketSize: 100
enabled: true
numBuckets: 6
timeInMilliseconds: 60000
rollingStats:
numBuckets: 10
timeInMilliseconds: 5000
requestCache:
enabled: false
requestLog:
enabled: false
shareSecurityContext: true
threadpool:
default:
coreSize: 1
maxQueueSize: 200
queueSizeRejectionThreshold: 2
鑑權
/**
* 鑑權
* @Author 小柒2012
* @Date 2020/6/12
*/
@Configuration
public class AuthFilter implements GlobalFilter, Ordered {
private static final Logger logger = LoggerFactory.getLogger(AuthFilter.class);
@Override
public Mono<Void> filter(ServerWebExchange exchange, GatewayFilterChain chain) {
String token = exchange.getRequest().getHeaders().getFirst("token");
String path = exchange.getRequest().getPath().toString();
/**
* 排除
*/
if(patterns.contains(path)){
return chain.filter(exchange);
}
/**
* 判斷 token 是否爲空
*/
if (StringUtils.isBlank(token)) {
logger.info( "token is empty..." );
exchange.getResponse().setStatusCode(HttpStatus.UNAUTHORIZED);
return exchange.getResponse().setComplete();
}else{
/**
* 驗證真僞
*/
CheckResult checkResult = JwtUtils.validateJWT(token);
if (!checkResult.isSuccess()) {
logger.info( "token is error..." );
exchange.getResponse().setStatusCode(HttpStatus.UNAUTHORIZED);
return exchange.getResponse().setComplete();
}
}
return chain.filter(exchange);
}
@Override
public int getOrder() {
return -100;
}
private static List<String> patterns =
Arrays.asList(new String[] {"/api/sys/login","/error","/api/sys/v2/api-docs"});
}
熔斷
一般是指軟件系統中，由於某些原因使得服務出現了過載現象，爲防止造成整個系統故障，從而採用的一種保護措施，所以很多地方把熔斷亦稱爲過載保護。
適用場景
防止應用程序直接調用那些很可能會調用失敗的遠程服務或共享資源。
服務降級
當服務器壓力劇增的情況下，根據當前業務情況及流量對一些服務和頁面有策略的降級，以此釋放服務器資源以保證核心任務的正常運行。
核心配置：
spring:
cloud:
nacos:
discovery:
server-addr: 127.0.0.1:8848
gateway:
routes:
- id: sys
uri: lb://tools-meizi
predicates:
- Path=/api/meizi/**
filters:
- StripPrefix=2
- name: Hystrix
args:
name: fallback
fallbackUri: forward:/fallback # 熔斷回調
主要參數：
# 熔斷
hystrix:
command:
default:
circuitBreaker:
enabled: true
errorThresholdPercentage: 50
forceClosed: false
forceOpen: false
requestVolumeThreshold: 4
sleepWindowInMilliseconds: 10000
execution:
isolation:
semaphore:
maxConcurrentRequests: 2
strategy: SEMAPHORE
thread:
timeoutInMilliseconds: 3000
metrics:
healthSnapshot:
intervalInMilliseconds: 500
rollingPercentile:
bucketSize: 100
enabled: true
numBuckets: 6
timeInMilliseconds: 60000
rollingStats:
numBuckets: 10
timeInMilliseconds: 5000
requestCache:
enabled: false
requestLog:
enabled: false
shareSecurityContext: true
threadpool:
default:
coreSize: 1
maxQueueSize: 200
queueSizeRejectionThreshold: 2
核心代碼 DefaultHystrixController：
/**
* 降級處理
*/
@RestController
public class DefaultHystrixController {
@RequestMapping("/fallback")
public Map<String,String> fallback(){
Map<String,String> map = new HashMap<>(8);
map.put("code","fail");
map.put("msg","服務異常");
return map;
}
}
小結
以上基本能滿足項目的常用需求，如果想更加靈活的使用，建議使用 Alibaba Sentinel，阿里巴巴團隊開源的一款熔斷限流神器，配合控制檯更加靈活實用。
▲掃一掃回覆【妹子圖】獲取源碼
你點的每個在看，我都認真當成了喜歡