RocketMQ-消费重试机制
介绍：
RocketMQ的消息重试及时分为两种，一种是Producer端重试，一种是Consume端重试。
1、Producer端重试 ：
1.1消息发没发成功，默认情况下是3次重试。
2、Consumer端重试：
2.1 exception的情况，一般重复16次 10s、30s、1mins、2mins、3mins等。注意reconsumeTimes这个参数；
2.2 超时情况，这种情况MQ会无限制的发送给消费端。这种情况就是Consumer端没有返回ConsumeConcurrentlyStatus.CONSUME_SUCCESS，也没有返回ConsumeConcurrentlyStatus.RECONSUME_LATER。Consumner超时的情况我们还分为一个Producer和一个Consumer的场景和一个Producer和多个Consumer（Consumer集群）的场景。下面的实例中我们会详细的做实验。
问题，Consumer的默认超时时间是多少？
示例：
Producer端重试
producer端重试的机制比较简单，我们通过看一下源码可以发现，通过设置retryTimesWhenSendFailed定义重试次数，通过设置sendMsgTimeout来定义超时时间
producer.setRetryTimesWhenSendFailed(3); //默认是2
producer.setSendMsgTimeout(6000); //默认是3000
package org.hope.lee.producer; import com.alibaba.rocketmq.client.exception.MQBrokerException;
import com.alibaba.rocketmq.client.exception.MQClientException;
import com.alibaba.rocketmq.client.producer.DefaultMQProducer;
import com.alibaba.rocketmq.client.producer.SendCallback;
import com.alibaba.rocketmq.client.producer.SendResult;
import com.alibaba.rocketmq.common.message.Message;
import com.alibaba.rocketmq.common.message.MessageQueue;
import com.alibaba.rocketmq.remoting.exception.RemotingException; public class Producer {
public static void main(String[] args) {
DefaultMQProducer producer = new DefaultMQProducer("push_consumer");
// producer.setNamesrvAddr("192.168.31.176:9876");
producer.setNamesrvAddr("192.168.31.176:9876;192.168.31.165:9876");
try {
// 设置实例名称
producer.setInstanceName("quick_start_producer");
// 设置重试次数,默认2
producer.setRetryTimesWhenSendFailed(3);
//设置发送超时时间，默认是3000
producer.setSendMsgTimeout(6000);
// 开启生产者
producer.start();
// 创建一条消息
Message msg = new Message("PushTopic_tt1", "TagB", "OrderID0034", "uniform_just_for_test".getBytes());
SendResult send = producer.send(msg);
System.out.println("id:--->" + send.getMsgId() + ",result:--->" + send.getSendStatus());
} catch (MQClientException e) {
e.printStackTrace();
} catch (RemotingException e) {
e.printStackTrace();
} catch (MQBrokerException e) {
e.printStackTrace();
} catch (InterruptedException e) {
e.printStackTrace();
}
producer.shutdown();
}
}
Consumer端exception的情况。
Producer端发送消息：
package org.hope.lee.producer; import com.alibaba.rocketmq.client.exception.MQBrokerException;
import com.alibaba.rocketmq.client.exception.MQClientException;
import com.alibaba.rocketmq.client.producer.DefaultMQProducer;
import com.alibaba.rocketmq.client.producer.SendCallback;
import com.alibaba.rocketmq.client.producer.SendResult;
import com.alibaba.rocketmq.common.message.Message;
import com.alibaba.rocketmq.common.message.MessageQueue;
import com.alibaba.rocketmq.remoting.exception.RemotingException; public class ProducerException {
public static void main(String[] args) {
DefaultMQProducer producer = new DefaultMQProducer("push_consumer");
producer.setNamesrvAddr("192.168.31.176:9876;192.168.31.165:9876");
try {
producer.setInstanceName("quick_start_producer");
producer.setRetryTimesWhenSendFailed(3);
producer.start();
for(int i = 0; i < 10; i++) {
Message msg = new Message("PushTopic_tt1", "TagA", "OrderID0034", ("message" + i).getBytes());
//目前发现3.2.6版本没有这个方法，3.5.3版本有这个方法，并且必须要设置为false否则会报错
// producer.setVipChannelEnabled(false);
SendResult send = producer.send(msg);
System.out.println("id:--->" + send.getMsgId() + ",result:--->" + send.getSendStatus());
} } catch (MQClientException e) {
e.printStackTrace();
} catch (RemotingException e) {
e.printStackTrace();
} catch (MQBrokerException e) {
e.printStackTrace();
} catch (InterruptedException e) {
e.printStackTrace();
}
producer.shutdown();
}
}
Consumer端接收消息
package org.hope.lee.consumer; import java.util.List; import com.alibaba.rocketmq.client.consumer.DefaultMQPushConsumer;
import com.alibaba.rocketmq.client.consumer.listener.ConsumeConcurrentlyContext;
import com.alibaba.rocketmq.client.consumer.listener.ConsumeConcurrentlyStatus;
import com.alibaba.rocketmq.client.consumer.listener.MessageListener;
import com.alibaba.rocketmq.client.consumer.listener.MessageListenerConcurrently;
import com.alibaba.rocketmq.client.exception.MQClientException;
import com.alibaba.rocketmq.client.producer.DefaultMQProducer;
import com.alibaba.rocketmq.common.consumer.ConsumeFromWhere;
import com.alibaba.rocketmq.common.message.MessageExt;
import com.sun.org.apache.xpath.internal.SourceTree; /**
* 消费端重试的情况 :异常情况
*/
public class ConsumerException {
public static void main(String[] args) throws MQClientException {
DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("push_consumer");
consumer.setNamesrvAddr("192.168.31.176:9876;192.168.31.165:9876");
// 批量消费,每次拉取10条
consumer.setConsumeMessageBatchMaxSize(10);// ①
// consumer.setInstanceName("quick_start_consumer");
// 3.2.6这个版本没有这个方法，3.5.3版本要设置这个方法为false,否则取不到topic
// consumer.setVipChannelEnabled(false); // 程序第一次启动从消息队列头取数据
// 如果非第一次启动，那么按照上次消费的位置继续消费
consumer.setConsumeFromWhere(ConsumeFromWhere.CONSUME_FROM_FIRST_OFFSET);
// 订阅PushTopic下Tag为push的消息
consumer.subscribe("PushTopic_tt1", "*");
consumer.registerMessageListener(new MessageListenerConcurrently() {
public ConsumeConcurrentlyStatus consumeMessage(List<MessageExt> msgs, ConsumeConcurrentlyContext context) { MessageExt msg = msgs.get(0);
try {
String topic = msg.getTopic();
String msgBody = new String(msg.getBody(), "utf-8");
String tags = msg.getTags();
System.out.println("收到消息:topic:" + topic + ",tags:" + tags + ",msg:" + msg + "msgBody:" + msgBody);
if ("message4".equals(msgBody)) {
System.out.println("====失败消息开始=====");
System.out.println("msg:" + msg);
System.out.println("msgBody:" + msgBody);
System.out.println("====失败消息结束=====");
// 发生异常
int i = 1 / 0;
System.out.println(i);
}
} catch (Exception e) {
e.printStackTrace();
// ②如果重试了三次就返回成功
if (msg.getReconsumeTimes() == 3) {
return ConsumeConcurrentlyStatus.CONSUME_SUCCESS;
}
return ConsumeConcurrentlyStatus.RECONSUME_LATER;
} return ConsumeConcurrentlyStatus.CONSUME_SUCCESS;
}
});
consumer.start();
System.out.println("Consumer Started.");
// consumer.suspend(); }
}
我们先启动ConsumerException再启动ProducerException。我们看到在①处发生了异常，这个时候消息就会进行重试，我们看看Consumer端的输出来分析一下。
Consumer Started.
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=0, storeSize=149, queueOffset=4, sysFlag=0, bornTimestamp=1515158266232, bornHost=/192.168.31.38:7309, storeTimestamp=1515158293401, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000001603, commitLogOffset=5635, bodyCRC=886509244, reconsumeTimes=0, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={TAGS=TagA, KEYS=OrderID0034, WAIT=true, MAX_OFFSET=5, MIN_OFFSET=0}, body=8]]msgBody:message0
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=1, storeSize=149, queueOffset=3, sysFlag=0, bornTimestamp=1515158266249, bornHost=/192.168.31.38:7309, storeTimestamp=1515158293411, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000001698, commitLogOffset=5784, bodyCRC=1137720874, reconsumeTimes=0, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={TAGS=TagA, KEYS=OrderID0034, WAIT=true, MAX_OFFSET=4, MIN_OFFSET=0}, body=8]]msgBody:message1
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=3, storeSize=149, queueOffset=2, sysFlag=0, bornTimestamp=1515158266267, bornHost=/192.168.31.38:7309, storeTimestamp=1515158293428, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F00000000000017C2, commitLogOffset=6082, bodyCRC=769548038, reconsumeTimes=0, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={TAGS=TagA, KEYS=OrderID0034, WAIT=true, MAX_OFFSET=3, MIN_OFFSET=0}, body=8]]msgBody:message3
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=2, storeSize=149, queueOffset=2, sysFlag=0, bornTimestamp=1515158266259, bornHost=/192.168.31.38:7309, storeTimestamp=1515158293420, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F000000000000172D, commitLogOffset=5933, bodyCRC=1524199312, reconsumeTimes=0, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={TAGS=TagA, KEYS=OrderID0034, WAIT=true, MAX_OFFSET=3, MIN_OFFSET=0}, body=8]]msgBody:message2
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=0, storeSize=149, queueOffset=5, sysFlag=0, bornTimestamp=1515158266276, bornHost=/192.168.31.38:7309, storeTimestamp=1515158293445, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000001857, commitLogOffset=6231, bodyCRC=867879589, reconsumeTimes=0, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={TAGS=TagA, KEYS=OrderID0034, WAIT=true, MAX_OFFSET=6, MIN_OFFSET=0}, body=8]]msgBody:message4
====失败消息开始=====
msg:MessageExt [queueId=0, storeSize=149, queueOffset=5, sysFlag=0, bornTimestamp=1515158266276, bornHost=/192.168.31.38:7309, storeTimestamp=1515158293445, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000001857, commitLogOffset=6231, bodyCRC=867879589, reconsumeTimes=0, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={TAGS=TagA, KEYS=OrderID0034, WAIT=true, MAX_OFFSET=6, MIN_OFFSET=0}, body=8]]
msgBody:message4
====失败消息结束=====
java.lang.ArithmeticException: / by zero
at org.hope.lee.consumer.ConsumerException$1.consumeMessage(ConsumerException.java:55)
at com.alibaba.rocketmq.client.impl.consumer.ConsumeMessageConcurrentlyService$ConsumeRequest.run(ConsumeMessageConcurrentlyService.java:142)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)
at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:334)
at java.util.concurrent.FutureTask.run(FutureTask.java:166)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:722)
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=1, storeSize=149, queueOffset=4, sysFlag=0, bornTimestamp=1515158266292, bornHost=/192.168.31.38:7309, storeTimestamp=1515158293452, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F00000000000018EC, commitLogOffset=6380, bodyCRC=1153301043, reconsumeTimes=0, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={TAGS=TagA, KEYS=OrderID0034, WAIT=true, MAX_OFFSET=5, MIN_OFFSET=0}, body=8]]msgBody:message5
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=2, storeSize=149, queueOffset=3, sysFlag=0, bornTimestamp=1515158266300, bornHost=/192.168.31.38:7309, storeTimestamp=1515158293482, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000001A9D, commitLogOffset=6813, bodyCRC=1572121481, reconsumeTimes=0, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={TAGS=TagA, KEYS=OrderID0034, WAIT=true, MAX_OFFSET=4, MIN_OFFSET=0}, body=8]]msgBody:message6
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=3, storeSize=149, queueOffset=3, sysFlag=0, bornTimestamp=1515158266328, bornHost=/192.168.31.38:7309, storeTimestamp=1515158293491, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000001B32, commitLogOffset=6962, bodyCRC=716413727, reconsumeTimes=0, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={TAGS=TagA, KEYS=OrderID0034, WAIT=true, MAX_OFFSET=4, MIN_OFFSET=0}, body=8]]msgBody:message7
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=0, storeSize=149, queueOffset=6, sysFlag=0, bornTimestamp=1515158266337, bornHost=/192.168.31.38:7309, storeTimestamp=1515158293497, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000001BC7, commitLogOffset=7111, bodyCRC=973899406, reconsumeTimes=0, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={TAGS=TagA, KEYS=OrderID0034, WAIT=true, MAX_OFFSET=7, MIN_OFFSET=0}, body=8]]msgBody:message8
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=1, storeSize=149, queueOffset=5, sysFlag=0, bornTimestamp=1515158266348, bornHost=/192.168.31.38:7309, storeTimestamp=1515158293508, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000001C5C, commitLogOffset=7260, bodyCRC=1292613144, reconsumeTimes=0, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={TAGS=TagA, KEYS=OrderID0034, WAIT=true, MAX_OFFSET=6, MIN_OFFSET=0}, body=8]]msgBody:message9
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=0, storeSize=285, queueOffset=7, sysFlag=0, bornTimestamp=1515158266276, bornHost=/192.168.31.38:7309, storeTimestamp=1515158303482, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000001CF1, commitLogOffset=7409, bodyCRC=867879589, reconsumeTimes=1, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={DELAY=3, ORIGIN_MESSAGE_ID=C0A81FB000002A9F0000000000001857, REAL_TOPIC=%RETRY%push_consumer, TAGS=TagA, WAIT=false, KEYS=OrderID0034, RETRY_TOPIC=PushTopic_tt1, MAX_OFFSET=8, MIN_OFFSET=0, REAL_QID=0}, body=8]]msgBody:message4
====失败消息开始=====
msg:MessageExt [queueId=0, storeSize=285, queueOffset=7, sysFlag=0, bornTimestamp=1515158266276, bornHost=/192.168.31.38:7309, storeTimestamp=1515158303482, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000001CF1, commitLogOffset=7409, bodyCRC=867879589, reconsumeTimes=1, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={DELAY=3, ORIGIN_MESSAGE_ID=C0A81FB000002A9F0000000000001857, REAL_TOPIC=%RETRY%push_consumer, TAGS=TagA, WAIT=false, KEYS=OrderID0034, RETRY_TOPIC=PushTopic_tt1, MAX_OFFSET=8, MIN_OFFSET=0, REAL_QID=0}, body=8]]
msgBody:message4
====失败消息结束=====
java.lang.ArithmeticException: / by zero
at org.hope.lee.consumer.ConsumerException$1.consumeMessage(ConsumerException.java:55)
at com.alibaba.rocketmq.client.impl.consumer.ConsumeMessageConcurrentlyService$ConsumeRequest.run(ConsumeMessageConcurrentlyService.java:142)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)
at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:334)
at java.util.concurrent.FutureTask.run(FutureTask.java:166)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:722)
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=0, storeSize=285, queueOffset=8, sysFlag=0, bornTimestamp=1515158266276, bornHost=/192.168.31.38:7309, storeTimestamp=1515158333701, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000001F2A, commitLogOffset=7978, bodyCRC=867879589, reconsumeTimes=2, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={DELAY=4, ORIGIN_MESSAGE_ID=C0A81FB000002A9F0000000000001857, REAL_TOPIC=%RETRY%push_consumer, TAGS=TagA, WAIT=false, KEYS=OrderID0034, RETRY_TOPIC=PushTopic_tt1, MAX_OFFSET=9, MIN_OFFSET=0, REAL_QID=0}, body=8]]msgBody:message4
====失败消息开始=====
msg:MessageExt [queueId=0, storeSize=285, queueOffset=8, sysFlag=0, bornTimestamp=1515158266276, bornHost=/192.168.31.38:7309, storeTimestamp=1515158333701, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000001F2A, commitLogOffset=7978, bodyCRC=867879589, reconsumeTimes=2, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={DELAY=4, ORIGIN_MESSAGE_ID=C0A81FB000002A9F0000000000001857, REAL_TOPIC=%RETRY%push_consumer, TAGS=TagA, WAIT=false, KEYS=OrderID0034, RETRY_TOPIC=PushTopic_tt1, MAX_OFFSET=9, MIN_OFFSET=0, REAL_QID=0}, body=8]]
msgBody:message4
====失败消息结束=====
java.lang.ArithmeticException: / by zero
at org.hope.lee.consumer.ConsumerException$1.consumeMessage(ConsumerException.java:55)
at com.alibaba.rocketmq.client.impl.consumer.ConsumeMessageConcurrentlyService$ConsumeRequest.run(ConsumeMessageConcurrentlyService.java:142)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)
at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:334)
at java.util.concurrent.FutureTask.run(FutureTask.java:166)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:722)
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=0, storeSize=285, queueOffset=9, sysFlag=0, bornTimestamp=1515158266276, bornHost=/192.168.31.38:7309, storeTimestamp=1515158394718, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000002163, commitLogOffset=8547, bodyCRC=867879589, reconsumeTimes=3, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={DELAY=5, ORIGIN_MESSAGE_ID=C0A81FB000002A9F0000000000001857, REAL_TOPIC=%RETRY%push_consumer, TAGS=TagA, WAIT=false, KEYS=OrderID0034, RETRY_TOPIC=PushTopic_tt1, MAX_OFFSET=10, MIN_OFFSET=0, REAL_QID=0}, body=8]]msgBody:message4
====失败消息开始=====
msg:MessageExt [queueId=0, storeSize=285, queueOffset=9, sysFlag=0, bornTimestamp=1515158266276, bornHost=/192.168.31.38:7309, storeTimestamp=1515158394718, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F0000000000002163, commitLogOffset=8547, bodyCRC=867879589, reconsumeTimes=3, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={DELAY=5, ORIGIN_MESSAGE_ID=C0A81FB000002A9F0000000000001857, REAL_TOPIC=%RETRY%push_consumer, TAGS=TagA, WAIT=false, KEYS=OrderID0034, RETRY_TOPIC=PushTopic_tt1, MAX_OFFSET=10, MIN_OFFSET=0, REAL_QID=0}, body=8]]
msgBody:message4
====失败消息结束=====
java.lang.ArithmeticException: / by zero
at org.hope.lee.consumer.ConsumerException$1.consumeMessage(ConsumerException.java:55)
at com.alibaba.rocketmq.client.impl.consumer.ConsumeMessageConcurrentlyService$ConsumeRequest.run(ConsumeMessageConcurrentlyService.java:142)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)
at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:334)
at java.util.concurrent.FutureTask.run(FutureTask.java:166)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:722)
收到消息:topic:PushTopic_tt1,tags:TagA,msg:MessageExt [queueId=0, storeSize=285, queueOffset=10, sysFlag=0, bornTimestamp=1515158266276, bornHost=/192.168.31.38:7309, storeTimestamp=1515158514742, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F000000000000239C, commitLogOffset=9116, bodyCRC=867879589, reconsumeTimes=4, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={DELAY=6, ORIGIN_MESSAGE_ID=C0A81FB000002A9F0000000000001857, REAL_TOPIC=%RETRY%push_consumer, TAGS=TagA, WAIT=false, KEYS=OrderID0034, RETRY_TOPIC=PushTopic_tt1, MAX_OFFSET=11, MIN_OFFSET=0, REAL_QID=0}, body=8]]msgBody:message4
====失败消息开始=====
msg:MessageExt [queueId=0, storeSize=285, queueOffset=10, sysFlag=0, bornTimestamp=1515158266276, bornHost=/192.168.31.38:7309, storeTimestamp=1515158514742, storeHost=/192.168.31.176:10911, msgId=C0A81FB000002A9F000000000000239C, commitLogOffset=9116, bodyCRC=867879589, reconsumeTimes=4, preparedTransactionOffset=0, toString()=Message [topic=PushTopic_tt1, flag=0, properties={DELAY=6, ORIGIN_MESSAGE_ID=C0A81FB000002A9F0000000000001857, REAL_TOPIC=%RETRY%push_consumer, TAGS=TagA, WAIT=false, KEYS=OrderID0034, RETRY_TOPIC=PushTopic_tt1, MAX_OFFSET=11, MIN_OFFSET=0, REAL_QID=0}, body=8]]
msgBody:message4
====失败消息结束=====
java.lang.ArithmeticException: / by zero
at org.hope.lee.consumer.ConsumerException$1.consumeMessage(ConsumerException.java:55)
at com.alibaba.rocketmq.client.impl.consumer.ConsumeMessageConcurrentlyService$ConsumeRequest.run(ConsumeMessageConcurrentlyService.java:142)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)
at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:334)
at java.util.concurrent.FutureTask.run(FutureTask.java:166)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:722)
java.lang.ArithmeticException: / by zero
at org.hope.lee.consumer.ConsumerException$1.consumeMessage(ConsumerException.java:55)
at com.alibaba.rocketmq.client.impl.consumer.ConsumeMessageConcurrentlyService$ConsumeRequest.run(ConsumeMessageConcurrentlyService.java:142)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)
at java.util.concurrent.FutureTask$Sync.innerRun(FutureTask.java:334)
at java.util.concurrent.FutureTask.run(FutureTask.java:166)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:722)
我们看到最初消费到第四条数据的时候抛出异常，然后继续消费了剩下的6条消息，过了一段时间MQ又重试发送消息给Consumer端，又抛出异常，过一段时间MQ还是会发送消息过来。我么恩看一下用蓝色标注的字体，这个就是记录重试次数的字段。我们可以利用这个字段进行业务逻辑处理，比如重试了3次就返回SUCCESS了，就不再重试了。比如Consumern代码中的②处
customer端发生超时的情况讨论。
在消费端，我们有两个Consumer来组成一个集群。ConsumerClusterMember1中我们设置一个Thread.sleep()来模拟一个消费超时的状态。
实验的步骤如下：
一、先启动ConsumerClusterMember1
二、再启动ConsumerClusterMember2。
三、然后启动ProducerCluster来发送一条消息。
四、这个时候ConsumerClusterMember1可能会先接收到这条消息，然后sleep等待了。这个时候我们停掉它的JVM
五、过了一段时间，ConsumerClusterMember2就会重新收到Producer端发过来的消息。完成了Consumer端集群的消息超时。
下面我们看看代码
Produder端：
package org.hope.lee.producer; import com.alibaba.rocketmq.client.exception.MQBrokerException;
import com.alibaba.rocketmq.client.exception.MQClientException;
import com.alibaba.rocketmq.client.producer.DefaultMQProducer;
import com.alibaba.rocketmq.client.producer.SendCallback;
import com.alibaba.rocketmq.client.producer.SendResult;
import com.alibaba.rocketmq.common.message.Message;
import com.alibaba.rocketmq.common.message.MessageQueue;
import com.alibaba.rocketmq.remoting.exception.RemotingException;
/**
* 配合ConsumerClusterMember1,2做测试
*/
public class ProducerCluster {
public static void main(String[] args) {
DefaultMQProducer producer = new DefaultMQProducer("producer_cluster");
producer.setNamesrvAddr("192.168.31.176:9876;192.168.31.165:9876");
try {
// 设置实例名称
producer.setInstanceName("producer_cluster_name");
// 开启生产者
producer.start();
// 创建一条消息
Message msg = new Message("cluster_timeout_test", "TagA", "OrderID0034", "customer_cluster_test".getBytes());
SendResult send = producer.send(msg);
System.out.println("id:--->" + send.getMsgId() + ",result:--->" + send.getSendStatus());
} catch (MQClientException e) {
e.printStackTrace();
} catch (RemotingException e) {
e.printStackTrace();
} catch (MQBrokerException e) {
e.printStackTrace();
} catch (InterruptedException e) {
e.printStackTrace();
}
producer.shutdown();
}
}
消费端集群：
package org.hope.lee.consumer; import java.util.List; import com.alibaba.rocketmq.client.consumer.DefaultMQPushConsumer;
import com.alibaba.rocketmq.client.consumer.listener.ConsumeConcurrentlyContext;
import com.alibaba.rocketmq.client.consumer.listener.ConsumeConcurrentlyStatus;
import com.alibaba.rocketmq.client.consumer.listener.MessageListenerConcurrently;
import com.alibaba.rocketmq.client.exception.MQClientException;
import com.alibaba.rocketmq.common.consumer.ConsumeFromWhere;
import com.alibaba.rocketmq.common.message.MessageExt; public class ConsumerClusterMember1 {
public static void main(String[] args) throws MQClientException {
DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("consumer_cluster");
consumer.setNamesrvAddr("192.168.31.176:9876;192.168.31.165:9876");
// 批量消费,每次拉取10条
consumer.setConsumeMessageBatchMaxSize(10);
// 如果非第一次启动，那么按照上次消费的位置继续消费
consumer.setConsumeFromWhere(ConsumeFromWhere.CONSUME_FROM_FIRST_OFFSET);
// 订阅PushTopic下Tag为push的消息
consumer.subscribe("cluster_timeout_test", "TagA || Tag B || Tage C");
consumer.registerMessageListener(new MqMessageListener1());
consumer.start();
System.out.println("Consumer Started."); }
}
class MqMessageListener1 implements MessageListenerConcurrently{
public ConsumeConcurrentlyStatus consumeMessage(List<MessageExt> msgs, ConsumeConcurrentlyContext context) {
try {
MessageExt msg = msgs.get(0);
System.out.println("------消息处理中--------");
Thread.sleep(60000);
String msgBody = new String(msg.getBody(), "utf-8");
System.out.println("msgBody:" + msgBody);
} catch(Exception e) {
e.printStackTrace();
return ConsumeConcurrentlyStatus.RECONSUME_LATER;
}
return ConsumeConcurrentlyStatus.CONSUME_SUCCESS;
} }
package org.hope.lee.consumer; import java.util.List; import com.alibaba.rocketmq.client.consumer.DefaultMQPushConsumer;
import com.alibaba.rocketmq.client.consumer.listener.ConsumeConcurrentlyContext;
import com.alibaba.rocketmq.client.consumer.listener.ConsumeConcurrentlyStatus;
import com.alibaba.rocketmq.client.consumer.listener.MessageListenerConcurrently;
import com.alibaba.rocketmq.client.exception.MQClientException;
import com.alibaba.rocketmq.common.consumer.ConsumeFromWhere;
import com.alibaba.rocketmq.common.message.MessageExt; public class ConsumerClusterMember2 {
public static void main(String[] args) throws MQClientException {
DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("consumer_cluster");
consumer.setNamesrvAddr("192.168.31.176:9876;192.168.31.165:9876");
// 批量消费,每次拉取10条
consumer.setConsumeMessageBatchMaxSize(10);
// 如果非第一次启动，那么按照上次消费的位置继续消费
consumer.setConsumeFromWhere(ConsumeFromWhere.CONSUME_FROM_FIRST_OFFSET);
// 订阅PushTopic下Tag为push的消息
consumer.subscribe("cluster_timeout_test", "TagA || Tag B || Tage C");
consumer.registerMessageListener(new MqMessageListener2());
consumer.start();
System.out.println("Consumer Started."); }
} class MqMessageListener2 implements MessageListenerConcurrently{
public ConsumeConcurrentlyStatus consumeMessage(List<MessageExt> msgs, ConsumeConcurrentlyContext context) {
try {
MessageExt msg = msgs.get(0);
String msgBody = new String(msg.getBody(), "utf-8");
System.out.println("msgBody:" + msgBody);
} catch(Exception e) {
e.printStackTrace();
return ConsumeConcurrentlyStatus.RECONSUME_LATER;
}
return ConsumeConcurrentlyStatus.CONSUME_SUCCESS;
} }
https://gitee.com/huayicompany/RocketMQ-learn/tree/master/rocketmq-api
参考：
[1]博客，http://blog.csdn.net/u010634288/article/details/56049305
RocketMQ-消费重试机制的更多相关文章
- Spring-Kafka —— 消费重试机制实现
消息处理问题 在从Kafka主题接收消息之后立即处理消息的消费者的实现非常简单.不幸的是,现实要复杂得多,并且由于各种原因,消息处理可能会失败.其中一些原因是永久性问题,例如数据库约束失败或消息格式无 ...
- RocketMQ(5)---RocketMQ重试机制
RocketMQ重试机制 消息重试分为两种:Producer发送消息的重试 和 Consumer消息消费的重试. 一.Producer端重试 Producer端重试是指: Producer往MQ上发消 ...
- RocketMQ重试机制和消息幂等
一.重试机制 由于MQ经常处于复杂的分布式系统中,考虑网络波动,服务宕机,程序异常因素,很有可能出现消息发送或者消费失败的问题.因此,消息的重试就是所有MQ中间件必须考虑到的一个关键点.如果没有消息重 ...
- Rocket重试机制，消息模式，刷盘方式
一.Consumer 批量消费(推模式) 可以通过 consumer.setConsumeMessageBatchMaxSize(10);//每次拉取10条 这里需要分为2种情况 Consumer端先 ...
- RocketMQ源码 — 八、 RocketMQ消息重试
RocketMQ的消息重试包含了producer发送消息的重试和consumer消息消费的重试. producer发送消息重试 producer在发送消息的时候如果发送失败了,RocketMQ会自动重 ...
- 要做重试机制，就只能选择 DelayQueue ？其实 RabbitMQ 它上它也行！
原文链接:要做重试机制,就只能选择 DelayQueue ?其实 RabbitMQ 它上它也行! 一.场景 最近研发一个新功能,后台天气预警:后台启动一条线程,定时调用天气预警 API,查询现有城市的 ...
- RabbitMq手动确认时的重试机制
本文转载自RabbitMq手动确认时的重试机制 消息手动确认模式的几点说明 监听的方法内部必须使用channel进行消息确认,包括消费成功或消费失败 如果不手动确认,也不抛出异常,消息不会自动重新推送 ...
- ENode 1.0 - 消息的重试机制的设计思路
项目开源地址:https://github.com/tangxuehua/enode 上一篇文章,简单介绍了enode框架中消息队列的设计思路,本文介绍一下enode框架中关系消息的重试机制的设计思路 ...
- RocketMQ消费批拉超过32不生效
由于一些原因,我需要RocketMQ消费的时候,一批拉400条,一批处理400条.设置如下: 为了简单验证是否正确,消费如下: 直接通过打印msgs.size()观察情况即可. 现象 实验的topic ...
- 【Dubbo 源码解析】07_Dubbo 重试机制
Dubbo 重试机制 通过前面 Dubbo 服务发现&引用 的分析,我们知道,Dubbo 的重试机制是通过 com.alibaba.dubbo.rpc.cluster.support.Fail ...
随机推荐
- js 数组API之filter的用法
filter 查找数组中满足条件的元素,返回新数组:原数组不变 var subArr = arr.filter(function(value, index, array){ return 条件 }) ...
- jdk8新特性（文章推荐）
文章推荐 jdk9都已经出来了,虽然很多项目都已经使用jdk8,但是很少会用到jdk8中的新特性.本人经常用的到也就是使用Stream,Lambda,但也仅仅是使用,基本不知道什么Function,C ...
- springboot 静态方法注入service
p.p1 { margin: 0.0px 0.0px 0.0px 0.0px; line-height: 16.0px; font: 14.0px Arial; color: #3f3f3f; bac ...
- Vue2.0 相对于Vue1.0的变化：生命周期
1.生命周期 根据提供的生命周期钩子说明Vue.js实例各个阶段的情况,Vue.js 2.0对不少钩子进行了修改,以下说明:(灰色字体代表是 2.0已经废弃或者被更名的钩子,黑色字体代表1.0 和2. ...
- css的学习笔记
CSS3有哪些新特性? 1. CSS3实现圆角(border-radius),阴影(box-shadow), 2. 对文字加特效(text-shadow.),线性渐变(gradient),旋转(tra ...
- 基于 Webpack 引入 jquery 插件的笔记
如果都是基于 webpack(npm 上有包),那就非常顺利: import $ from 'jquery' import 'jquery-modal/jquery.modal.min.css' im ...
- Java DB访问之 JDBC
项目说明 项目采用 maven 组织 ,jdbc 唯一的依赖就是 mysql-connector-java pom 依赖如下: <dependency> <groupId>my ...
- TCP/IP（三）数据链路层~1
前言 其实前面一堆讲的物理层的概念,会感觉特别的难理解,因为这是一个非常强大的计算机网络体系的底层知识,没有关系!我们大致了解一下就行了. 一.数据链路层概述 这是百度的简介 看图:理解一下,数据链路 ...
- CTF---安全杂项入门第二题 A记录
A记录分值:20 来源: sammie 难度:中 参与人数:2255人 Get Flag:566人 答题人数:621人 解题通过率:91% 他在看什么视频,好像很好看,不知道是什么网站的. 还好我截取 ...
- 【Java数据结构学习笔记之一】线性表的存储结构及其代码实现
应用程序后在那个的数据大致有四种基本的逻辑结构: 集合:数据元素之间只有"同属于一个集合"的关系 线性结构:数据元素之间存在一个对一个的关系 树形结构:数据元素之间存在一个对多个关 ...