spring-kafka消费者配置

     <dependency>
            <groupId>org.springframework.kafka</groupId>
            <artifactId>spring-kafka</artifactId>
            <version>${spring-kafka.version}</version>
        </dependency>

目前kafka版本更新的有快,0.10版本刚用稳定,1.0都出来了，0.11版本重启的时候会偶尔会报日志文件未正常结束，如果用0.10就基本不会出现这个问题，这里主要是在开发环境会经常关闭再开，

这里spring-kafka的版本为1.2.2.RELEASE版本

上consumer配置文件

<?xml version="1.0" encoding="UTF-8"?>
<beans xmlns="http://www.springframework.org/schema/beans"
       xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
       xmlns:int="http://www.springframework.org/schema/integration"
       xmlns:int-kafka="http://www.springframework.org/schema/integration/kafka"
       xmlns:context="http://www.springframework.org/schema/context"
       xmlns:task="http://www.springframework.org/schema/task"
       xsi:schemaLocation="http://www.springframework.org/schema/integration/kafka
        http://www.springframework.org/schema/integration/kafka/spring-integration-kafka.xsd
        http://www.springframework.org/schema/integration
        http://www.springframework.org/schema/integration/spring-integration.xsd
        http://www.springframework.org/schema/beans
        http://www.springframework.org/schema/beans/spring-beans.xsd http://www.springframework.org/schema/context http://www.springframework.org/schema/context/spring-context.xsd http://www.springframework.org/schema/task http://www.springframework.org/schema/task/spring-task.xsd">

    <bean id="messagingMessageConverter" class="org.springframework.kafka.support.converter.MessagingMessageConverter"/>

    <!-- 定义consumer的参数 -->
    <bean id="consumerProperties" class="java.util.HashMap">
        <constructor-arg>
            <map>
                <entry key="bootstrap.servers" value="${kafka.broker.address}"/>
                <entry key="group.id" value="${kafka.broker.groupid}"/>
                <entry key="enable.auto.commit" value="false"/>
                <entry key="auto.commit.interval.ms" value="1000"/>
                <entry key="session.timeout.ms" value="15000"/>
                <entry key="key.deserializer" value="org.apache.kafka.common.serialization.StringDeserializer"/>
                <entry key="value.deserializer" value="org.apache.kafka.common.serialization.StringDeserializer"/>
            </map>
        </constructor-arg>
    </bean>

    <!-- 创建consumerFactory bean -->
    <bean id="consumerFactory" class="org.springframework.kafka.core.DefaultKafkaConsumerFactory">
        <constructor-arg>
            <ref bean="consumerProperties"/>
        </constructor-arg>
    </bean>

    <!-- 消费者容器配置信息 -->
    <bean id="containerProperties_flowevent" class="org.springframework.kafka.listener.config.ContainerProperties">
        <constructor-arg value="${kafka.topic}"/>
        <property name="messageListener" ref="eventAsyncConsumerService"/>
        <property name="AckMode" value="MANUAL"/>
    </bean>

    <bean id="concurrentMessageListenerContainer" class="org.springframework.kafka.listener.ConcurrentMessageListenerContainer">
        <constructor-arg ref="consumerFactory"/>
        <constructor-arg ref="containerProperties_flowevent"/>
        <property name="Concurrency" value="2"/>
    </bean>

</beans>

如果需要手动提交offset,那么调用acknowledgment的ack方法。但这里有一个问题，不知道是不是项目版本的问题，
我之前的一个项目仅仅只配置enable.auto.commit为false，然后在配合acknowledgment使用是没任何问题的，到现在项目都在跑，

@Component
public class EventAsyncConsumerService implements AcknowledgingMessageListener<String, String> {
    private final static Logger log = LoggerFactory.getLogger(EventAsyncConsumerService.class);

    @Override
    public void onMessage(ConsumerRecord<String, String> consumerRecord, Acknowledgment acknowledgment) {
        if (log.isDebugEnabled()) {
            log.debug("value:" + consumerRecord.value());
            log.debug("topic:" + consumerRecord.topic());
            log.debug("partition:" + consumerRecord.partition());
            log.debug("offset:" + consumerRecord.offset());
        }

        try {
            EventMsg message=JSONSerializer.deserialize(consumerRecord.value(), EventMsg.class);
            EventInterface eventModel = ApplicationContextHelper.getBean(message.getEventName());
            Map result= eventModel.execute(message.getParams(),message.getPrevResult());
            // 最后 调用acknowledgment的ack方法，提交offset
            acknowledgment.acknowledge();
        } catch (Exception ex) {
            log.error("EventAsyncConsumer==Exception==>"+ex.getMessage()+ex.getStackTrace());
        }
    }
}

这几天升级了spring框架版本，发现不管用了。于是看了下spring-kafka源码，发现需要配置   <property name="AckMode" value="MANUAL"/>
就ok了

由于kafka官方最近升级很快，开发者使用的速度有点跟不上了，1.0版本在windows环境中启动会报错，这让人情何以堪，难道是必须要放到linux环境下使用的吗