RocketMQ-client 批处理、过滤、事务消息
简单研究下rockermq 批量消息、过滤消息、事务消息的使用。
1. 批量消息
- 简介
批量消息的发送能提升投递小消息的性能。但是批量消息有一些限制,一批投递的消息应该有相同的主题、具有相同的刷盘策略、不支持延时消息与事务型消息。
另外,生产者发送消息的大小有一些限制。默认不超过1MB 的消息。如果超出可以将批量消息进行拆分。或者通过修改配置。
生产者发送的消息结构如下:
生产者发送的消息并不是将消息直接序列化后发送到网络上的,而是通过Message 对象生成了一个字符串发送出去的。这个字符串包含:Topic、消息body、消息日志(20字节)、以及用于描述消息属性的key-value 结构,这些属性包括生产者地址、生产时间、以及要发送的queueId 等。最终写入到broker 中消息单元中的数据都来自于这些属性。
- 批量消费消息
org.apache.rocketmq.client.consumer.listener.MessageListenerConcurrently#consumeMessage 方法的第一个参数为消息列表,但是默认情况下每次只能消费一条。其规则简单理解是:一次性拉取 pullBatchSize(默认32) 参数指定的消息数量,然后按照consumeMessageBatchMaxSize (默认为1)指定的大小分发到多个消费线程。
- 代码
(1)生产者代码
package com.zd.bx.rocketmq.batch; import org.apache.rocketmq.client.producer.DefaultMQProducer; import org.apache.rocketmq.client.producer.SendResult; import org.apache.rocketmq.common.message.Message; import java.util.ArrayList; import java.util.Iterator; import java.util.List; import java.util.Map; public class Producer2 { public static void main(String[] args) throws Exception { DefaultMQProducer producer = new DefaultMQProducer("batchProducer"); producer.setNamesrvAddr("192.168.13.111:9876"); producer.start(); // 构造消息 String topic = "batchTest"; List<Message> messages = new ArrayList<>(); for (int index = 0; index < 100000; index++) { messages.add(new Message(topic, "TagA", "OrderI" + index, ("Hello world " + index).getBytes())); } // redirectSend(producer, messages); // 批量发送 ListSplitter splitter = new ListSplitter(messages); int sendNum = 0; while (splitter.hasNext()) { System.out.println("sendNum: " + (++sendNum)); try { List<Message> listItem = splitter.next(); producer.send(listItem); } catch (Exception e) { e.printStackTrace(); } } producer.shutdown(); } private static void redirectSend(DefaultMQProducer producer, List<Message> messages) { try { // 直接发送会报错: org.apache.rocketmq.client.exception.MQClientException: CODE: 13 DESC: the message body size over max value, MAX: 4194304 SendResult send = producer.send(messages); System.out.println(send); } catch (Exception e) { e.printStackTrace(); } } } class ListSplitter implements Iterator<List<Message>> { private final int SIZE_LIMIT = 1 * 1024 * 1024; private final List<Message> messages; private int currIndex; public ListSplitter(List<Message> messages) { this.messages = messages; } @Override public boolean hasNext() { return currIndex < messages.size(); } @Override public List<Message> next() { int nextIndex = currIndex; int totalSize = 0; for (; nextIndex < messages.size(); nextIndex++) { Message message = messages.get(nextIndex); int tmpSize = message.getTopic().length() + message.getBody().length; Map<String, String> properties = message.getProperties(); for (Map.Entry<String, String> entry : properties.entrySet()) { tmpSize += entry.getKey().length() + entry.getValue().length(); } tmpSize = tmpSize + 20; //for log overhead if (tmpSize > SIZE_LIMIT) { //it is unexpected that single message exceeds the SIZE_LIMIT //here just let it go, otherwise it will block the splitting process if (nextIndex - currIndex == 0) { //if the next sublist has no element, add this one and then break, otherwise just break nextIndex++; } break; } if (tmpSize + totalSize > SIZE_LIMIT) { break; } else { totalSize += tmpSize; } } List<Message> subList = messages.subList(currIndex, nextIndex); currIndex = nextIndex; return subList; } }
(2)批量消费者
package com.zd.bx.rocketmq.batch; import org.apache.rocketmq.client.consumer.DefaultMQPushConsumer; import org.apache.rocketmq.client.consumer.listener.ConsumeConcurrentlyContext; import org.apache.rocketmq.client.consumer.listener.ConsumeConcurrentlyStatus; import org.apache.rocketmq.client.consumer.listener.MessageListenerConcurrently; import org.apache.rocketmq.client.exception.MQClientException; import org.apache.rocketmq.common.message.MessageExt; import java.util.List; public class PushConsumer { public static void main(String[] args) throws InterruptedException, MQClientException { // 定义一个push消费者 DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("myTestConsumerGroup"); // 指定nameserver consumer.setNamesrvAddr("192.168.13.111:9876"); // 指定消费的topic与tag consumer.subscribe("batchTest", "*"); // 指定使用 广播模式进行消费,默认为集群模式 // consumer.setMessageModel(MessageModel.BROADCASTING); // 修改一次拉取的最大值(默认32),可以理解为一次拉取回来之后然后按下面的参数consumeMessageBatchMaxSize,分发到多个线程组中进行消费 consumer.setPullBatchSize(60); // 指定每次可以消费100条消息(可以理解是下面监听器中集合的大小),默认为1,值范围是[1, 1024] consumer.setConsumeMessageBatchMaxSize(3); consumer.registerMessageListener(new MessageListenerConcurrently() { @Override public ConsumeConcurrentlyStatus consumeMessage(List<MessageExt> msgs, ConsumeConcurrentlyContext context) { System.out.println(Thread.currentThread().getName() + " msgs.size(): " + msgs.size()); for (MessageExt msg : msgs) { System.out.printf(Thread.currentThread().getName() + " %s Receive New Messages: %s body: %s %n", Thread.currentThread().getName(), msgs, new String(msg.getBody())); } return ConsumeConcurrentlyStatus.CONSUME_SUCCESS; } }); consumer.start(); System.out.printf("Consumer Started.%n"); } }
2. 过滤消息
在接收消息的时候除了可以指定消息的Topic外,还可以对指定Topic的消息进行过滤。可以根据Tag 进行过滤,也可以根据sql 过滤。sql过滤是对消息中属性进行筛选过滤,只有push模式的消费者才能使用sql过滤。
sql过滤表达式中支持多种常量类型与运算符。支持的常量类型:
- 数值:比如1,2,3
- 字符:必须用单引号包裹,比如'abc'
- 布尔:TRUE或者FALSE
支持的运算符有:
- 数值比较:>,<,>=,<=,BETWEEN,=
- 字符比较:=,<>,IN
- 逻辑比较:AND,OR,NOT
- NULL判断: IS NULL 或者 IS NOT NULL
默认情况下Broker没有开启sql过滤功能,需要在Broker加载的配置文件中添加如下属性:(单机在broker.conf 文件中)
enablePropertyFilter=true
测试代码:
(1)生产者:自定义属性
package com.zd.bx.rocketmq.filter; import org.apache.rocketmq.client.exception.MQBrokerException; import org.apache.rocketmq.client.exception.MQClientException; import org.apache.rocketmq.client.producer.DefaultMQProducer; import org.apache.rocketmq.common.message.Message; import org.apache.rocketmq.remoting.exception.RemotingException; public class Producer { public static void main(String[] args) throws MQClientException, MQBrokerException, RemotingException, InterruptedException { DefaultMQProducer producer = new DefaultMQProducer("batchProducer"); producer.setNamesrvAddr("192.168.13.111:9876"); producer.start(); String[] tags = new String[]{"TagA", "TagB", "TagC"}; String[] sexs = new String[]{"男", "女"}; // 构造消息 String topic = "filterTopic"; for (int index = 0; index < 20; index++) { byte[] body = ("Hi, " + index).getBytes(); String tag = tags[index % tags.length]; Message message = new Message(topic, tag, body); // 存一个自定义属性age, 一个自定义属性sex message.putUserProperty("age", index + ""); message.putUserProperty("sex", sexs[index % sexs.length]); System.out.println(producer.send(message)); } producer.shutdown(); } }
(2)消费者
消费者一:按标签进行过滤
package com.zd.bx.rocketmq.filter; import org.apache.rocketmq.client.consumer.DefaultMQPushConsumer; import org.apache.rocketmq.client.consumer.listener.ConsumeConcurrentlyContext; import org.apache.rocketmq.client.consumer.listener.ConsumeConcurrentlyStatus; import org.apache.rocketmq.client.consumer.listener.MessageListenerConcurrently; import org.apache.rocketmq.client.exception.MQClientException; import org.apache.rocketmq.common.message.MessageExt; import java.util.List; public class Consumer1 { public static void main(String[] args) throws MQClientException { // 定义一个push消费者 DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("myTestConsumerGroup"); // 指定从第一条消息开始消费 // 指定nameserver consumer.setNamesrvAddr("192.168.13.111:9876"); // 指定消费的topic与tag. "TagA || TagB" 等价于 MessageSelector.byTag("TagA || TagB") consumer.subscribe("filterTopic", "TagA || TagB"); consumer.registerMessageListener(new MessageListenerConcurrently() { @Override public ConsumeConcurrentlyStatus consumeMessage(List<MessageExt> msgs, ConsumeConcurrentlyContext context) { for (MessageExt msg : msgs) { System.out.printf("%s Receive Messages: %s, property: %s %n", Thread.currentThread().getName(), new String(msg.getBody()), msg.getProperties()); } return ConsumeConcurrentlyStatus.CONSUME_SUCCESS; } }); consumer.start(); System.out.printf("Consumer Started.%n"); } }
消费者二:按sql条件进行过滤(接收age大于并且sex为男的)
package com.zd.bx.rocketmq.filter; import org.apache.rocketmq.client.consumer.DefaultMQPushConsumer; import org.apache.rocketmq.client.consumer.MessageSelector; import org.apache.rocketmq.client.consumer.listener.ConsumeConcurrentlyContext; import org.apache.rocketmq.client.consumer.listener.ConsumeConcurrentlyStatus; import org.apache.rocketmq.client.consumer.listener.MessageListenerConcurrently; import org.apache.rocketmq.client.exception.MQClientException; import org.apache.rocketmq.common.message.MessageExt; import java.util.List; public class Consumer2 { public static void main(String[] args) throws MQClientException { // 定义一个push消费者 DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("myTestConsumerGroup"); // 指定nameserver consumer.setNamesrvAddr("192.168.13.111:9876"); // 指定消费的topic与tag consumer.subscribe("filterTopic", MessageSelector.bySql("age > 5 and sex = '男'")); consumer.registerMessageListener(new MessageListenerConcurrently() { @Override public ConsumeConcurrentlyStatus consumeMessage(List<MessageExt> msgs, ConsumeConcurrentlyContext context) { for (MessageExt msg : msgs) { System.out.printf("%s Receive New Messages: %s, property: %s %n", Thread.currentThread().getName(), new String(msg.getBody()), msg.getProperties()); } return ConsumeConcurrentlyStatus.CONSUME_SUCCESS; } }); consumer.start(); System.out.printf("Consumer Started.%n"); } }
3. 事务消息
事务型消息可以确保执行本地事务和发送消息保证原子性。这里需要注意是执行本地事务和发送消息是在一个事务里面。
- 举一个场景:工行用户A向建行用户B转账一万元。同步消息处理该需求的场景:
这里有个问题:若第三步骤的扣款失败,消息已经到达Broker。对于MQ来说,只要消息写入MQ就可以被消费。此时建行系统增加了1万元。此时就出现了数据不一致的问题。
- 解决思路:
使用MQ的事务消息。让1、2、3具有原子性,这里就是用到事务消息。但是对于4、5操作,是不在该事务中的,也就是事务消息只能保证本地事务和发送消息在是一致的。解决示意图如下:
解释:第3步发送的消息是半事务消息(预提交到Broker),消费者是不能消费该消息的;第6步发送扣款执行结果,实际是向TC汇报本地事务的执行状态(LocalTransactionState里面的状态)。
- 事务消息的使用限制:
(1)不支持定时和批处理消息
(2)事务消息要做好幂等性检查,因为事务消息可能不止一次被消费(存在回滚后再提交的情况)
- 基本概念
(1) 分布式事务:可以理解为多个本地事务的集合体,多个事务直接构成一个大的分布式事务。要么全失败,要么全成功。
(2)事务消息:RocketMQ提供了类似于X/OpenXA的分布式事务功能,通过事务消息能达到分布式事务的最终一致性。XA是一种分布式事务解决方案,一种分布式事务处理模式。
(3)半事务消息:暂不能投递的消息, 发送方已经成功地将消息发送到了Broker,但是Broker未收到最终确认指令,该消息被标记成"暂不能投递"状态,即不能为消费者看到。出于该状态下的消息成为半事务消息。
(4)本地事务状态:Producer回调执行的结果为本地事务状态,其会发给TC(事务控制器),而TC再发给TM(事务管理器)。TM根据TC送来的本地事务状态来决定全局事务确认指令。
package org.apache.rocketmq.client.producer; public enum LocalTransactionState { COMMIT_MESSAGE, // 提交事务,表示允许消费者消费该消息 ROLLBACK_MESSAGE, // 消息回滚,类似于预提交的消息撤回 UNKNOW, // 未知,执行回调 }
(5)消息回查:重新查询本地事务的执行状态。关于消息回查的设置:
transactionTimeout=20, 指定TM应该在20s内将消息发送给TC,否则引发消息回查。默认60s。
transactionCheckMax=5,指定最多回查次数,超过后将丢弃消息并且记录日志。默认15次。
transactionCheckInterval=10,指定设置的多次的消息回查的时间间隔为10s。默认为60s。
- 测试代码
(1)监听器
package com.zd.bx.rocketmq.tx; import org.apache.commons.lang3.StringUtils; import org.apache.rocketmq.client.producer.LocalTransactionState; import org.apache.rocketmq.client.producer.TransactionListener; import org.apache.rocketmq.common.message.Message; import org.apache.rocketmq.common.message.MessageExt; public class Listener implements TransactionListener { // 回调操作方法 // 消息预提交成功就会触发该方法的执行,用于完成本地事务 @Override public LocalTransactionState executeLocalTransaction(Message msg, Object arg) { System.out.println("预提交消息成功:" + msg); // 假设接收到TagA的消息就表示扣款操作成功,TagB的消息表示扣款失败, // TagC表示扣款结果不清楚,需要执行消息回查 if (StringUtils.equals("TagA", msg.getTags())) { return LocalTransactionState.COMMIT_MESSAGE; } else if (StringUtils.equals("TagB", msg.getTags())) { return LocalTransactionState.ROLLBACK_MESSAGE; } else if (StringUtils.equals("TagC", msg.getTags())) { return LocalTransactionState.UNKNOW; } return LocalTransactionState.UNKNOW; } // 消息回查方法 // 引发消息回查的原因最常见的有两个: // 1)回调操作返回UNKNWON // 2)TC没有接收到TM的最终全局事务确认指令 @Override public LocalTransactionState checkLocalTransaction(MessageExt msg) { System.out.println("执行消息回查: body" + new String(msg.getBody()) + " tag: " + msg.getTags() + " txId: " + msg.getTransactionId()); return LocalTransactionState.COMMIT_MESSAGE; } }
(2)生产者
package com.zd.bx.rocketmq.tx; import org.apache.rocketmq.client.exception.MQClientException; import org.apache.rocketmq.client.producer.SendResult; import org.apache.rocketmq.client.producer.TransactionMQProducer; import org.apache.rocketmq.common.message.Message; import org.apache.rocketmq.remoting.common.RemotingHelper; import java.io.UnsupportedEncodingException; import java.util.concurrent.*; public class Producer { public static void main(String[] args) throws InterruptedException, MQClientException { TransactionMQProducer producer = new TransactionMQProducer("transactionProducerGroup"); producer.setNamesrvAddr("192.168.13.111:9876"); ExecutorService executorService = new ThreadPoolExecutor(2, 5, 100, TimeUnit.SECONDS, new ArrayBlockingQueue<Runnable>(2000), new ThreadFactory() { @Override public Thread newThread(Runnable r) { Thread thread = new Thread(r); thread.setName("client-transaction-msg-check-thread"); return thread; } }); // 为生产者指定线程池和监听器 producer.setExecutorService(executorService); producer.setTransactionListener(new Listener()); producer.start(); String[] tags = new String[]{"TagA", "TagB", "TagC"}; for (int i = 0; i < 10; i++) { try { Message msg = new Message("txTopic", tags[i % tags.length], "KEY" + i, ("Hello RocketMQ " + i).getBytes(RemotingHelper.DEFAULT_CHARSET)); SendResult sendResult = producer.sendMessageInTransaction(msg, null); System.out.printf("%s%n", sendResult); Thread.sleep(10); } catch (MQClientException | UnsupportedEncodingException e) { e.printStackTrace(); } } for (int i = 0; i < 100000; i++) { Thread.sleep(1000); } producer.shutdown(); } }
日志如下:
预提交消息成功:Message{topic='txTopic', flag=0, properties={KEYS=KEY0, TRAN_MSG=true, UNIQ_KEY=7F0000015BC818B4AAC2793C9CF80000, WAIT=true, PGROUP=transactionProducerGroup, TAGS=TagA}, body=[72, 101, 108, 108, 111, 32, 82, 111, 99, 107, 101, 116, 77, 81, 32, 48], transactionId='7F0000015BC818B4AAC2793C9CF80000'} SendResult [sendStatus=SEND_OK, msgId=7F0000015BC818B4AAC2793C9CF80000, offsetMsgId=null, messageQueue=MessageQueue [topic=txTopic, brokerName=broker-a, queueId=2], queueOffset=40] 预提交消息成功:Message{topic='txTopic', flag=0, properties={KEYS=KEY1, TRAN_MSG=true, UNIQ_KEY=7F0000015BC818B4AAC2793C9D170001, WAIT=true, PGROUP=transactionProducerGroup, TAGS=TagB}, body=[72, 101, 108, 108, 111, 32, 82, 111, 99, 107, 101, 116, 77, 81, 32, 49], transactionId='7F0000015BC818B4AAC2793C9D170001'} SendResult [sendStatus=SEND_OK, msgId=7F0000015BC818B4AAC2793C9D170001, offsetMsgId=null, messageQueue=MessageQueue [topic=txTopic, brokerName=broker-a, queueId=3], queueOffset=41] 预提交消息成功:Message{topic='txTopic', flag=0, properties={KEYS=KEY2, TRAN_MSG=true, UNIQ_KEY=7F0000015BC818B4AAC2793C9D280002, WAIT=true, PGROUP=transactionProducerGroup, TAGS=TagC}, body=[72, 101, 108, 108, 111, 32, 82, 111, 99, 107, 101, 116, 77, 81, 32, 50], transactionId='7F0000015BC818B4AAC2793C9D280002'} SendResult [sendStatus=SEND_OK, msgId=7F0000015BC818B4AAC2793C9D280002, offsetMsgId=null, messageQueue=MessageQueue [topic=txTopic, brokerName=broker-a, queueId=0], queueOffset=42] 预提交消息成功:Message{topic='txTopic', flag=0, properties={KEYS=KEY3, TRAN_MSG=true, UNIQ_KEY=7F0000015BC818B4AAC2793C9D3D0003, WAIT=true, PGROUP=transactionProducerGroup, TAGS=TagA}, body=[72, 101, 108, 108, 111, 32, 82, 111, 99, 107, 101, 116, 77, 81, 32, 51], transactionId='7F0000015BC818B4AAC2793C9D3D0003'} SendResult [sendStatus=SEND_OK, msgId=7F0000015BC818B4AAC2793C9D3D0003, offsetMsgId=null, messageQueue=MessageQueue [topic=txTopic, brokerName=broker-a, queueId=1], queueOffset=43] 预提交消息成功:Message{topic='txTopic', flag=0, properties={KEYS=KEY4, TRAN_MSG=true, UNIQ_KEY=7F0000015BC818B4AAC2793C9D510004, WAIT=true, PGROUP=transactionProducerGroup, TAGS=TagB}, body=[72, 101, 108, 108, 111, 32, 82, 111, 99, 107, 101, 116, 77, 81, 32, 52], transactionId='7F0000015BC818B4AAC2793C9D510004'} SendResult [sendStatus=SEND_OK, msgId=7F0000015BC818B4AAC2793C9D510004, offsetMsgId=null, messageQueue=MessageQueue [topic=txTopic, brokerName=broker-a, queueId=2], queueOffset=44] 预提交消息成功:Message{topic='txTopic', flag=0, properties={KEYS=KEY5, TRAN_MSG=true, UNIQ_KEY=7F0000015BC818B4AAC2793C9D650005, WAIT=true, PGROUP=transactionProducerGroup, TAGS=TagC}, body=[72, 101, 108, 108, 111, 32, 82, 111, 99, 107, 101, 116, 77, 81, 32, 53], transactionId='7F0000015BC818B4AAC2793C9D650005'} SendResult [sendStatus=SEND_OK, msgId=7F0000015BC818B4AAC2793C9D650005, offsetMsgId=null, messageQueue=MessageQueue [topic=txTopic, brokerName=broker-a, queueId=3], queueOffset=45] 执行消息回查: bodyHello RocketMQ 2 tag: TagC txId: 7F0000015BC818B4AAC2793C9D280002 执行消息回查: bodyHello RocketMQ 5 tag: TagC txId: 7F0000015BC818B4AAC2793C9D650005
(3)消费者
package com.zd.bx.rocketmq.tx; import org.apache.rocketmq.client.consumer.DefaultMQPushConsumer; import org.apache.rocketmq.client.consumer.listener.ConsumeConcurrentlyContext; import org.apache.rocketmq.client.consumer.listener.ConsumeConcurrentlyStatus; import org.apache.rocketmq.client.consumer.listener.MessageListenerConcurrently; import org.apache.rocketmq.client.exception.MQClientException; import org.apache.rocketmq.common.message.MessageExt; import java.util.List; public class Consumer { public static void main(String[] args) throws MQClientException { // 定义一个push消费者 DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("myTestConsumerGroup"); // 指定nameserver consumer.setNamesrvAddr("192.168.13.111:9876"); consumer.subscribe("txTopic", "*"); consumer.registerMessageListener(new MessageListenerConcurrently() { @Override public ConsumeConcurrentlyStatus consumeMessage(List<MessageExt> msgs, ConsumeConcurrentlyContext context) { for (MessageExt msg : msgs) { System.out.printf("%s Receive Messages: %s, property: %s %n", Thread.currentThread().getName(), new String(msg.getBody()), msg.getProperties()); } return ConsumeConcurrentlyStatus.CONSUME_SUCCESS; } }); consumer.start(); System.out.printf("Consumer Started.%n"); } }
日志如下:
ConsumeMessageThread_myTestConsumerGroup_5 Receive Messages: Hello RocketMQ 0, property: {MIN_OFFSET=0, REAL_TOPIC=txTopic, MAX_OFFSET=4, KEYS=KEY0, TRAN_MSG=true, CONSUME_START_TIME=1648098015500, UNIQ_KEY=7F0000015BC818B4AAC2793C9CF80000, CLUSTER=DefaultCluster, PGROUP=transactionProducerGroup, WAIT=true, TAGS=TagA, REAL_QID=2} ConsumeMessageThread_myTestConsumerGroup_6 Receive Messages: Hello RocketMQ 3, property: {MIN_OFFSET=0, REAL_TOPIC=txTopic, MAX_OFFSET=3, KEYS=KEY3, TRAN_MSG=true, CONSUME_START_TIME=1648098015559, UNIQ_KEY=7F0000015BC818B4AAC2793C9D3D0003, CLUSTER=DefaultCluster, PGROUP=transactionProducerGroup, WAIT=true, TAGS=TagA, REAL_QID=1} ConsumeMessageThread_myTestConsumerGroup_7 Receive Messages: Hello RocketMQ 5, property: {TRANSACTION_CHECK_TIMES=1, TRAN_MSG=true, CONSUME_START_TIME=1648098030724, MIN_OFFSET=0, REAL_TOPIC=txTopic, MAX_OFFSET=4, KEYS=KEY5, UNIQ_KEY=7F0000015BC818B4AAC2793C9D650005, CLUSTER=DefaultCluster, PGROUP=transactionProducerGroup, WAIT=true, TAGS=TagC, REAL_QID=3} ConsumeMessageThread_myTestConsumerGroup_8 Receive Messages: Hello RocketMQ 2, property: {TRANSACTION_CHECK_TIMES=1, TRAN_MSG=true, CONSUME_START_TIME=1648098030733, MIN_OFFSET=0, REAL_TOPIC=txTopic, MAX_OFFSET=4, KEYS=KEY2, UNIQ_KEY=7F0000015BC818B4AAC2793C9D280002, CLUSTER=DefaultCluster, PGROUP=transactionProducerGroup, WAIT=true, TAGS=TagC, REAL_QID=0}
转载:https://www.cnblogs.com/qlqwjy/p/16029672.html