2.【Spring Cloud Alibaba】实现负载均衡-Ribbon
负载均衡的两种方式
如何实现负载均衡
目前已经实现让A总能找到B,如何实现负载均衡
负载均衡的两种方式
- 服务器端负载均衡
- 客户端负载均衡
使用Ribbo实现负载均衡
Ribbon是什么
Netflix开源的客户端侧负载均衡器
引入Ribbon后的架构演进
整合Ribbon实现负载均衡
// 在spring容器中,创建一个对象,类型RestTemplate;名称/ID是:restTemplate
// <bean id="restTemplate" class="xxx.RestTemplate"/>
@Bean
@LoadBalanced
public RestTemplate restTemplate() {
RestTemplate template = new RestTemplate();
return template;
}
@Autowired
private RestTemplate restTemplate;
@GetMapping("/test-rest-template-sentinel/{userId}")
public UserDTO test(@PathVariable Integer userId) {
return this.restTemplate
.getForObject(
"http://user-center/users/{userId}",
UserDTO.class, userId);
}
Ribbon的组成
Ribbon内置的负载均衡规则
细粒度配置自定义
- Java代码配置
- 用配置属性配置
- 最佳实践总结
Java代码配置
@Configuration
@RibbonClient(name = "user-center",configuration = RibbonConfiguration.class)
public class UserCenterRibbonConfiguration {
}
@Configuration
public class RibbonConfiguration {
@Bean
public IRule ribbonRule() {
return new NacosSameClusterWeightedRule();
}
}
用配置属性配置
user-center:
ribbon:
NFLoadBalancerRuleClassName: com.netflix.loadbalancer.RandomRule
两种方式对比
Ribbon全局配置
@Configuration
@RibbonClients(defaultConfiguration = RibbonConfiguration.class)
public class UserCenterRibbonConfiguration {
}
@Configuration
public class RibbonConfiguration {
@Bean
public IRule ribbonRule() {
return new NacosSameClusterWeightedRule();
}
}
Ribbon饥饿加载配置方式
ribbon:
eager-load:
enabled: true
clients: user-center
扩展Ribbon-支持Nacos权重
import com.alibaba.nacos.api.exception.NacosException;
import com.alibaba.nacos.api.naming.NamingService;
import com.alibaba.nacos.api.naming.pojo.Instance;
import com.netflix.client.config.IClientConfig;
import com.netflix.loadbalancer.*;
import lombok.extern.slf4j.Slf4j;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.cloud.alibaba.nacos.NacosDiscoveryProperties;
import org.springframework.cloud.alibaba.nacos.ribbon.NacosServer;
@Slf4j
public class NacosWeightedRule extends AbstractLoadBalancerRule {
@Autowired
private NacosDiscoveryProperties nacosDiscoveryProperties;
@Override
public void initWithNiwsConfig(IClientConfig clientConfig) {
// 读取配置文件,并初始化NacosWeightedRule
}
@Override
public Server choose(Object key) {
try {
BaseLoadBalancer loadBalancer = (BaseLoadBalancer) this.getLoadBalancer();
// log.info("lb = {}", loadBalancer);
// 想要请求的微服务的名称
String name = loadBalancer.getName();
// 拿到服务发现的相关API
NamingService namingService = nacosDiscoveryProperties.namingServiceInstance();
// nacos client自动通过基于权重的负载均衡算法,给我们选择一个实例。
Instance instance = namingService.selectOneHealthyInstance(name);
log.info("选择的实例是:port = {}, instance = {}", instance.getPort(), instance);
return new NacosServer(instance);
} catch (NacosException e) {
return null;
}
}
}
// spring cloud commons --> 定义了标准
// spring cloud loadbalancer --> 没有权重
扩展Ribbon-同一集群优先调用
import com.alibaba.nacos.api.exception.NacosException;
import com.alibaba.nacos.api.naming.NamingService;
import com.alibaba.nacos.api.naming.pojo.Instance;
import com.alibaba.nacos.client.naming.core.Balancer;
import com.netflix.client.config.IClientConfig;
import com.netflix.loadbalancer.AbstractLoadBalancerRule;
import com.netflix.loadbalancer.BaseLoadBalancer;
import com.netflix.loadbalancer.Server;
import lombok.extern.slf4j.Slf4j;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.cloud.alibaba.nacos.NacosDiscoveryProperties;
import org.springframework.cloud.alibaba.nacos.ribbon.NacosServer;
import org.springframework.util.CollectionUtils;
import java.util.ArrayList;
import java.util.List;
import java.util.Objects;
import java.util.stream.Collectors;
@Slf4j
public class NacosSameClusterWeightedRule extends AbstractLoadBalancerRule {
@Autowired
private NacosDiscoveryProperties nacosDiscoveryProperties;
@Override
public void initWithNiwsConfig(IClientConfig clientConfig) {
}
@Override
public Server choose(Object key) {
try {
// 拿到配置文件中的集群名称 BJ
String clusterName = nacosDiscoveryProperties.getClusterName();
BaseLoadBalancer loadBalancer = (BaseLoadBalancer) this.getLoadBalancer();
// 想要请求的微服务的名称
String name = loadBalancer.getName();
// 拿到服务发现的相关API
NamingService namingService = nacosDiscoveryProperties.namingServiceInstance();
// 1. 找到指定服务的所有实例 A
List<Instance> instances = namingService.selectInstances(name, true);
// 2. 过滤出相同集群下的所有实例 B
List<Instance> sameClusterInstances = instances.stream()
.filter(instance -> Objects.equals(instance.getClusterName(), clusterName))
.collect(Collectors.toList());
// 3. 如果B是空,就用A
List<Instance> instancesToBeChosen = new ArrayList<>();
if (CollectionUtils.isEmpty(sameClusterInstances)) {
instancesToBeChosen = instances;
log.warn("发生跨集群的调用, name = {}, clusterName = {}, instances = {}",
name,
clusterName,
instances
);
} else {
instancesToBeChosen = sameClusterInstances;
}
// 4. 基于权重的负载均衡算法,返回1个实例
Instance instance = ExtendBalancer.getHostByRandomWeight2(instancesToBeChosen);
log.info("选择的实例是 port = {}, instance = {}", instance.getPort(), instance);
return new NacosServer(instance);
} catch (NacosException e) {
log.error("发生异常了", e);
return null;
}
}
}
class ExtendBalancer extends Balancer {
public static Instance getHostByRandomWeight2(List<Instance> hosts) {
return getHostByRandomWeight(hosts);
}
}
扩展Ribbon-基于元数据的版本控制
元数据
元数据就是一堆的描述信息,以map存储。举个例子:
spring:
cloud:
nacos:
metadata:
# 自己这个实例的版本
version: v1
# 允许调用的提供者版本
target-version: v1
代码实现
@Slf4j
public class NacosFinalRule extends AbstractLoadBalancerRule {
@Autowired
private NacosDiscoveryProperties nacosDiscoveryProperties;
@Override
public Server choose(Object key) {
// 负载均衡规则:优先选择同集群下,符合metadata的实例
// 如果没有,就选择所有集群下,符合metadata的实例
// 1. 查询所有实例 A
// 2. 筛选元数据匹配的实例 B
// 3. 筛选出同cluster下元数据匹配的实例 C
// 4. 如果C为空,就用B
// 5. 随机选择实例
try {
String clusterName = this.nacosDiscoveryProperties.getClusterName();
String targetVersion = this.nacosDiscoveryProperties.getMetadata().get("target-version");
DynamicServerListLoadBalancer loadBalancer = (DynamicServerListLoadBalancer) getLoadBalancer();
String name = loadBalancer.getName();
NamingService namingService = this.nacosDiscoveryProperties.namingServiceInstance();
// 所有实例
List<Instance> instances = namingService.selectInstances(name, true);
List<Instance> metadataMatchInstances = instances;
// 如果配置了版本映射,那么只调用元数据匹配的实例
if (StringUtils.isNotBlank(targetVersion)) {
metadataMatchInstances = instances.stream()
.filter(instance -> Objects.equals(targetVersion, instance.getMetadata().get("version")))
.collect(Collectors.toList());
if (CollectionUtils.isEmpty(metadataMatchInstances)) {
log.warn("未找到元数据匹配的目标实例!请检查配置。targetVersion = {}, instance = {}", targetVersion, instances);
return null;
}
}
List<Instance> clusterMetadataMatchInstances = metadataMatchInstances;
// 如果配置了集群名称,需筛选同集群下元数据匹配的实例
if (StringUtils.isNotBlank(clusterName)) {
clusterMetadataMatchInstances = metadataMatchInstances.stream()
.filter(instance -> Objects.equals(clusterName, instance.getClusterName()))
.collect(Collectors.toList());
if (CollectionUtils.isEmpty(clusterMetadataMatchInstances)) {
clusterMetadataMatchInstances = metadataMatchInstances;
log.warn("发生跨集群调用。clusterName = {}, targetVersion = {}, clusterMetadataMatchInstances = {}", clusterName, targetVersion, clusterMetadataMatchInstances);
}
}
Instance instance = ExtendBalancer.getHostByRandomWeight2(clusterMetadataMatchInstances);
return new NacosServer(instance);
} catch (Exception e) {
log.warn("发生异常", e);
return null;
}
}
@Override
public void initWithNiwsConfig(IClientConfig iClientConfig) {
}
}
public class ExtendBalancer extends Balancer {
/**
* 根据权重,随机选择实例
*
* @param instances 实例列表
* @return 选择的实例
*/
public static Instance getHostByRandomWeight2(List<Instance> instances) {
return getHostByRandomWeight(instances);
}
}
深入理解Nacos的Namespace
服务调用不能跨namespace
欢迎关注微信或公众号!!!
PS:如果你觉得文章对你有所帮助,别忘了推荐或者分享,因为有你的支持,才是我续写下篇的动力和源泉!
出处: https://www.cnblogs.com/xjknight/
本文版权归作者和博客园共有,欢迎转载,但未经作者同意必须保留此段声明,且在文章页面明显位置给出原文连接,否则保留追究法律责任的权利。