Ribbon细粒度配置

1、java代码配置

单独建包，不然会产生spring父子上下文重叠，导致该配置为全局配置

package configuration;

import com.netflix.loadbalancer.IRule;
import com.netflix.loadbalancer.RandomRule;
import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.Configuration;

/**
 * @author: yangchun
 * @description:
 * @date: Created in 2020-02-19 7:49
 */
@Configuration
public class RibbonConfiguration {
    @Bean
    public IRule ribbonRule(){
        return new RandomRule();
    }
}

package com.fj.xiaofeiyang.pay.api.configuration;

import configuration.RibbonConfiguration;
import org.springframework.cloud.netflix.ribbon.RibbonClient;
import org.springframework.context.annotation.Configuration;

/**
 * @author: yangchun
 * @description:
 * @date: Created in 2020-02-19 7:51
 */
@RibbonClient(name = "pay",configuration = RibbonConfiguration.class)
public class PayRibbonConfiguration {
}

2、配置文件配置

pay:
  ribbon:
    NFLoadBalancerRuleClassName: com.netflix.loadbalancer.RandomRule

3、最佳实践总结

配置方式	优点	缺点
代码配置	基于代码，更加灵活	有小坑(父子上下文)，线上修改得重新打包、发包
属性配置	易上手，配置更加直观，线上修改无需重新打包发布，优先级更高	极端场景下没有代码配置灵活

 

 

尽量使用属性配置，属性配置不了采用代码配置，属性配置实现不了的采用代码配置。

4、全局配置

1、让前面的configuration包可以被应用扫描到，让父子上下文重叠

2、第二种如下

package com.fj.xiaofeiyang.pay.api.configuration;

import org.springframework.cloud.netflix.ribbon.RibbonClient;
import org.springframework.cloud.netflix.ribbon.RibbonClients;

/**
 * @author: yangchun
 * @description:
 * @date: Created in 2020-02-19 7:51
 */
@RibbonClients(defaultConfiguration = configuration.RibbonConfiguration.class)
public class RibbonConfiguration {}
5、支持的配置项

接口	作用	默认值
IClientConfig	读取配置	DefaultClientConfigImpl
IRule	负载均衡规则，选择实例	ZoneAvoidanceRule
IPing	筛选Ping不通的实例	DummyPing
ServerList<Server>	交给Ribbon的实例列表	Ribbon：ConfigurationBasedServerList,Spring Cloud Alibaba :NacosServerList
ServerListFilter<Server>	过滤掉不符合条件的实例	ZonePreferenceServerListFilter
ILoadBalancer	Ribbon的入口	ZoneAwareLoadBalancer
ServerListUpdater	更新交给Ribbon的List的策略	PollingServerListUpdater

java代码配置如下

@Bean
public IRule ribbonRule(){
    return new RandomRule();
}

属性配置如下
<clientName>.ribbon.如下属性：
NFLoadBalancerClassName：ILoaderBalancer实现类

NFLoadBalancerRuleClassName：IRule实现类

NFLoadBalancerPingClassName：IPing实现类

NIWSServerListClassName：ServerList实现类

NIWSServerListFilterClassName：ServerListFilter实现类

6、ribbon开启饥饿加载

ribbon:
  eager-load:
    enabled: true
    clients: pay
7、扩展ribbon支持权重选择
spring cloud commons --定义了标准
spring cloud loadbalancer --没有定义权重

 

 先将两个服务实例权重进行修改，

public class NacosWeightRule extends AbstractLoadBalancerRule {
    @Autowired
    private NacosDiscoveryProperties nacosDiscoveryProperties;
    @Override
    public void initWithNiwsConfig(IClientConfig iClientConfig) {

    }

    @Override
    public Server choose(Object key) {
        BaseLoadBalancer baseLoadBalancer = (BaseLoadBalancer)this.getLoadBalancer();
        String name = baseLoadBalancer.getName();
        NamingService nameingService = nacosDiscoveryProperties.namingServiceInstance();
        try {
            Instance instance= null;
            instance = nameingService.selectOneHealthyInstance(name);
            return new NacosServer(instance);
        } catch (NacosException e) {
            return null;
        }
    }
}
然后采用属性

pay:
  ribbon:
    NFLoadBalancerRuleClassName: com.fj.xiaofeiyang.pay.api.configuration.NacosWeightRule

配置的方式
扩展同一个集群下的服务器实例同时选择

package com.fj.xiaofeiyang.pay.api.configuration;

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 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;

/**
 * @author: yangchun
 * @description:
 * @date: Created in 2020-02-19 9:09
 */
public class NacosClusterRule extends AbstractLoadBalancerRule {
    @Autowired
    private NacosDiscoveryProperties nacosDiscoveryProperties;
    @Override
    public void initWithNiwsConfig(IClientConfig iClientConfig) {

    }

    @Override
    public Server choose(Object key) {
        BaseLoadBalancer baseLoadBalancer = (BaseLoadBalancer)this.getLoadBalancer();
        String name = baseLoadBalancer.getName();
        String clusterName = nacosDiscoveryProperties.getClusterName();
        NamingService namingService = nacosDiscoveryProperties.namingServiceInstance();
        try {
            List<Instance> instances= namingService.selectInstances(name,true);
            List<Instance> sameClusterInstances = instances.stream().filter(instance -> Objects.equals(instance.getClusterName(),clusterName))
                    .collect(Collectors.toList());
            List<Instance> instancesToBchosen = new ArrayList<>();
            if(CollectionUtils.isEmpty(sameClusterInstances)){
                instancesToBchosen = instances;
            }else {
                instancesToBchosen = sameClusterInstances;
            }
            Instance instance= ExtendBalancer.getHostByRandomWeight2(instancesToBchosen);
            return new NacosServer(instance);
        } catch (NacosException e) {
            return null;
        }
    }
}
class ExtendBalancer extends Balancer{
    public static Instance getHostByRandomWeight2(List<Instance> instances){
        return getHostByRandomWeight(instances);
    }
}

然后采用属性

pay:
  ribbon:
    NFLoadBalancerRuleClassName: com.fj.xiaofeiyang.pay.api.configuration.NacosClusterRule 

扩展通过元数据版本来实现

package com.fj.xiaofeiyang.pay.api.configuration;

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.AbstractLoadBalancerRule;
import com.netflix.loadbalancer.BaseLoadBalancer;
import com.netflix.loadbalancer.Server;
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;

/**
 * @author: yangchun
 * @description:
 * @date: Created in 2020-02-19 9:09
 */
public class NacosVersionRule extends AbstractLoadBalancerRule {
    @Autowired
    private NacosDiscoveryProperties nacosDiscoveryProperties;
    @Override
    public void initWithNiwsConfig(IClientConfig iClientConfig) {

    }

    @Override
    public Server choose(Object key) {
        BaseLoadBalancer baseLoadBalancer = (BaseLoadBalancer)this.getLoadBalancer();
        String name = baseLoadBalancer.getName();
        String targetVesion = nacosDiscoveryProperties.getMetadata().get("target-version");
        NamingService namingService = nacosDiscoveryProperties.namingServiceInstance();
        try {
            List<Instance> instances= namingService.selectInstances(name,true);
            List<Instance> sameClusterInstances = instances.stream().filter(instance -> Objects.equals(instance.getMetadata().get("version"),targetVesion))
                    .collect(Collectors.toList());
            List<Instance> instancesToBchosen = new ArrayList<>();
            if(CollectionUtils.isEmpty(sameClusterInstances)){
                instancesToBchosen = instances;
            }else {
                instancesToBchosen = sameClusterInstances;
            }
            Instance instance= ExtendBalancer.getHostByRandomWeight2(instancesToBchosen);
            return new NacosServer(instance);
        } catch (NacosException e) {
            return null;
        }
    }

}