聊聊RibbonLoadBalancerClient的choose方法

序

本文主要研究一下RibbonLoadBalancerClient的choose方法

RibbonLoadBalancerClient.choose

spring-cloud-netflix-ribbon-2.0.0.RC1-sources.jar!/org/springframework/cloud/netflix/ribbon/RibbonLoadBalancerClient.java

public ServiceInstance choose(String serviceId) {
        Server server = getServer(serviceId);
        if (server == null) {
            return null;
        }
        return new RibbonServer(serviceId, server, isSecure(server, serviceId),
                serverIntrospector(serviceId).getMetadata(server));
    }

    protected ILoadBalancer getLoadBalancer(String serviceId) {
        return this.clientFactory.getLoadBalancer(serviceId);
    }

    protected Server getServer(String serviceId) {
        return getServer(getLoadBalancer(serviceId));
    }

    protected Server getServer(ILoadBalancer loadBalancer) {
        if (loadBalancer == null) {
            return null;
        }
        return loadBalancer.chooseServer("default"); // TODO: better handling of key
    }

能夠看到，choose方法傳入serviceId，而後經過SpringClientFactory獲取ILoadBalancer，最後調用ILoadBalancer.chooseServer方法選取Server
這裏的key，2.0.0.RC1版本直接寫死default了。

ILoadBalancer.chooseServer

ribbon-loadbalancer-2.2.5-sources.jar!/com/netflix/loadbalancer/ILoadBalancer.java

/**
     * Choose a server from load balancer.
     * 
     * @param key An object that the load balancer may use to determine which server to return. null if 
     *         the load balancer does not use this parameter.
     * @return server chosen
     */
    public Server chooseServer(Object key);

ZoneAwareLoadBalancer

ribbon-loadbalancer-2.2.5-sources.jar!/com/netflix/loadbalancer/ZoneAwareLoadBalancer.java

/**
 * Load balancer that can avoid a zone as a whole when choosing server. 
 *<p>
 * The key metric used to measure the zone condition is Average Active Requests,
which is aggregated per rest client per zone. It is the
total outstanding requests in a zone divided by number of available targeted instances (excluding circuit breaker tripped instances).
This metric is very effective when timeout occurs slowly on a bad zone.
<p>
The  LoadBalancer will calculate and examine zone stats of all available zones. If the Average Active Requests for any zone has reached a configured threshold, this zone will be dropped from the active server list. In case more than one zone has reached the threshold, the zone with the most active requests per server will be dropped.
Once the the worst zone is dropped, a zone will be chosen among the rest with the probability proportional to its number of instances.
A server will be returned from the chosen zone with a given Rule (A Rule is a load balancing strategy, for example {@link AvailabilityFilteringRule})
For each request, the steps above will be repeated. That is to say, each zone related load balancing decisions are made at real time with the up-to-date statistics aiding the choice.

 * @author awang
 *
 * @param <T>
 */
public class ZoneAwareLoadBalancer<T extends Server> extends DynamicServerListLoadBalancer<T> {

    private static final DynamicBooleanProperty ENABLED = DynamicPropertyFactory.getInstance().getBooleanProperty("ZoneAwareNIWSDiscoveryLoadBalancer.enabled", true);

    //......
    public Server chooseServer(Object key) {
        if (!ENABLED.get() || getLoadBalancerStats().getAvailableZones().size() <= 1) {
            logger.debug("Zone aware logic disabled or there is only one zone");
            return super.chooseServer(key);
        }
        Server server = null;
        try {
            LoadBalancerStats lbStats = getLoadBalancerStats();
            Map<String, ZoneSnapshot> zoneSnapshot = ZoneAvoidanceRule.createSnapshot(lbStats);
            logger.debug("Zone snapshots: {}", zoneSnapshot);
            if (triggeringLoad == null) {
                triggeringLoad = DynamicPropertyFactory.getInstance().getDoubleProperty(
                        "ZoneAwareNIWSDiscoveryLoadBalancer." + this.getName() + ".triggeringLoadPerServerThreshold", 0.2d);
            }

            if (triggeringBlackoutPercentage == null) {
                triggeringBlackoutPercentage = DynamicPropertyFactory.getInstance().getDoubleProperty(
                        "ZoneAwareNIWSDiscoveryLoadBalancer." + this.getName() + ".avoidZoneWithBlackoutPercetage", 0.99999d);
            }
            Set<String> availableZones = ZoneAvoidanceRule.getAvailableZones(zoneSnapshot, triggeringLoad.get(), triggeringBlackoutPercentage.get());
            logger.debug("Available zones: {}", availableZones);
            if (availableZones != null &&  availableZones.size() < zoneSnapshot.keySet().size()) {
                String zone = ZoneAvoidanceRule.randomChooseZone(zoneSnapshot, availableZones);
                logger.debug("Zone chosen: {}", zone);
                if (zone != null) {
                    BaseLoadBalancer zoneLoadBalancer = getLoadBalancer(zone);
                    server = zoneLoadBalancer.chooseServer(key);
                }
            }
        } catch (Exception e) {
            logger.error("Error choosing server using zone aware logic for load balancer={}", name, e);
        }
        if (server != null) {
            return server;
        } else {
            logger.debug("Zone avoidance logic is not invoked.");
            return super.chooseServer(key);
        }
    }
}

這裏ZoneAwareNIWSDiscoveryLoadBalancer.enabled默認爲true
若是關閉這裏ZoneAwareNIWSDiscoveryLoadBalancer，或者只有單個zone的話，則走super.chooseServer(key)
ZoneAwareLoadBalancer繼承了DynamicServerListLoadBalancer，而DynamicServerListLoadBalancer繼承了BaseLoadBalancer，而BaseLoadBalancer又繼承了AbstractLoadBalancer。chooseServer方法在BaseLoadBalancer中有實現。

BaseLoadBalancer.chooseServer

ribbon-loadbalancer-2.2.5-sources.jar!/com/netflix/loadbalancer/BaseLoadBalancer.java

public Server chooseServer(Object key) {
        if (counter == null) {
            counter = createCounter();
        }
        counter.increment();
        if (rule == null) {
            return null;
        } else {
            try {
                return rule.choose(key);
            } catch (Exception e) {
                logger.warn("LoadBalancer [{}]:  Error choosing server for key {}", name, key, e);
                return null;
            }
        }
    }

這裏能夠看到最後是調用rule.choose(key)，這個rule默認是ZoneAvoidanceRule

RibbonClientConfiguration

spring-cloud-netflix-ribbon-2.0.0.RC1-sources.jar!/org/springframework/cloud/netflix/ribbon/RibbonClientConfiguration.java

@SuppressWarnings("deprecation")
@Configuration
@EnableConfigurationProperties
//Order is important here, last should be the default, first should be optional
// see https://github.com/spring-cloud/spring-cloud-netflix/issues/2086#issuecomment-316281653
@Import({HttpClientConfiguration.class, OkHttpRibbonConfiguration.class, RestClientRibbonConfiguration.class, HttpClientRibbonConfiguration.class})
public class RibbonClientConfiguration {
    //......
    @Bean
    @ConditionalOnMissingBean
    public IRule ribbonRule(IClientConfig config) {
        if (this.propertiesFactory.isSet(IRule.class, name)) {
            return this.propertiesFactory.get(IRule.class, config, name);
        }
        ZoneAvoidanceRule rule = new ZoneAvoidanceRule();
        rule.initWithNiwsConfig(config);
        return rule;
    }
}

能夠看到這裏指定了ZoneAvoidanceRule

PredicateBasedRule.choose

ribbon-loadbalancer-2.2.5-sources.jar!/com/netflix/loadbalancer/PredicateBasedRule.java

/**
 * A rule which delegates the server filtering logic to an instance of {@link AbstractServerPredicate}.
 * After filtering, a server is returned from filtered list in a round robin fashion.
 * 
 * 
 * @author awang
 *
 */
public abstract class PredicateBasedRule extends ClientConfigEnabledRoundRobinRule {
   
    /**
     * Method that provides an instance of {@link AbstractServerPredicate} to be used by this class.
     * 
     */
    public abstract AbstractServerPredicate getPredicate();
        
    /**
     * Get a server by calling {@link AbstractServerPredicate#chooseRandomlyAfterFiltering(java.util.List, Object)}.
     * The performance for this method is O(n) where n is number of servers to be filtered.
     */
    @Override
    public Server choose(Object key) {
        ILoadBalancer lb = getLoadBalancer();
        Optional<Server> server = getPredicate().chooseRoundRobinAfterFiltering(lb.getAllServers(), key);
        if (server.isPresent()) {
            return server.get();
        } else {
            return null;
        }       
    }
}

能夠看到這裏是調用getPredicate().chooseRoundRobinAfterFiltering方法

AbstractServerPredicate.chooseRoundRobinAfterFiltering

ribbon-loadbalancer-2.2.5-sources.jar!/com/netflix/loadbalancer/AbstractServerPredicate.java

/**
     * Choose a server in a round robin fashion after the predicate filters a given list of servers and load balancer key. 
     */
    public Optional<Server> chooseRoundRobinAfterFiltering(List<Server> servers, Object loadBalancerKey) {
        List<Server> eligible = getEligibleServers(servers, loadBalancerKey);
        if (eligible.size() == 0) {
            return Optional.absent();
        }
        return Optional.of(eligible.get(incrementAndGetModulo(eligible.size())));
    }

    /**
     * Get servers filtered by this predicate from list of servers. 
     */
    public List<Server> getEligibleServers(List<Server> servers, Object loadBalancerKey) {
        if (loadBalancerKey == null) {
            return ImmutableList.copyOf(Iterables.filter(servers, this.getServerOnlyPredicate()));            
        } else {
            List<Server> results = Lists.newArrayList();
            for (Server server: servers) {
                if (this.apply(new PredicateKey(loadBalancerKey, server))) {
                    results.add(server);
                }
            }
            return results;            
        }
    }

而chooseRoundRobinAfterFiltering調用的是getEligibleServers(servers, loadBalancerKey)方法
這裏loadBalancerKey不爲null，走的是else裏頭的apply方法。

CompositePredicate.getEligibleServers

ribbon-loadbalancer-2.2.5-sources.jar!/com/netflix/loadbalancer/CompositePredicate.java

@Override
    public boolean apply(@Nullable PredicateKey input) {
        return delegate.apply(input);
    }

    /**
     * Get the filtered servers from primary predicate, and if the number of the filtered servers
     * are not enough, trying the fallback predicates  
     */
    @Override
    public List<Server> getEligibleServers(List<Server> servers, Object loadBalancerKey) {
        List<Server> result = super.getEligibleServers(servers, loadBalancerKey);
        Iterator<AbstractServerPredicate> i = fallbacks.iterator();
        while (!(result.size() >= minimalFilteredServers && result.size() > (int) (servers.size() * minimalFilteredPercentage))
                && i.hasNext()) {
            AbstractServerPredicate predicate = i.next();
            result = predicate.getEligibleServers(servers, loadBalancerKey);
        }
        return result;
    }

CompositePredicate重寫了getEligibleServers方法，能夠看到先調用了父類AbstractServerPredicate的getEligibleServers，而後再對fallback等邏輯進行處理

ZoneAvoidanceRule

ribbon-loadbalancer-2.2.5-sources.jar!/com/netflix/loadbalancer/ZoneAvoidanceRule.java

/**
 * A rule that uses the a {@link CompositePredicate} to filter servers based on zone and availability. The primary predicate is composed of
 * a {@link ZoneAvoidancePredicate} and {@link AvailabilityPredicate}, with the fallbacks to {@link AvailabilityPredicate}
 * and an "always true" predicate returned from {@link AbstractServerPredicate#alwaysTrue()} 
 * 
 * @author awang
 *
 */
public class ZoneAvoidanceRule extends PredicateBasedRule {
    //......
    private CompositePredicate compositePredicate;
    
    public ZoneAvoidanceRule() {
        super();
        ZoneAvoidancePredicate zonePredicate = new ZoneAvoidancePredicate(this);
        AvailabilityPredicate availabilityPredicate = new AvailabilityPredicate(this);
        compositePredicate = createCompositePredicate(zonePredicate, availabilityPredicate);
    }
    
    private CompositePredicate createCompositePredicate(ZoneAvoidancePredicate p1, AvailabilityPredicate p2) {
        return CompositePredicate.withPredicates(p1, p2)
                             .addFallbackPredicate(p2)
                             .addFallbackPredicate(AbstractServerPredicate.alwaysTrue())
                             .build();
        
    }

    @Override
    public AbstractServerPredicate getPredicate() {
        return compositePredicate;
    }
}

這裏的getPredicate返回的是CompositePredicate，是組合Predicate，ZoneAvoidancePredicate及AvailabilityPredicate，若是兩者and不知足，則fallback到AvailabilityPredicate，再不知足，則fallback到AbstractServerPredicate.alwaysTrue()，即返回true，不過濾。

小結

• RibbonLoadBalancerClient.choose方法裏頭，調用loadBalancer.chooseServer("default")，寫死了loadBalancerKey。
• loadBalancer.chooseServer方法最後是委託給IRule的choose方法，默認是使用ZoneAvoidanceRule，其內部又委託給AbstractServerPredicate的chooseRoundRobinAfterFiltering方法進行過濾。
• 最後調用的是CompositePredicate的apply進行篩選，若是ZoneAvoidancePredicate與AvailabilityPredicate的and不經過，則看AvailabilityPredicate是否經過，若是再不經過則返回true，不進行過濾。

doc

• Client Side Load Balancer: Ribbon