1 基本樣例
來(lái)源 ：官方文檔
在基本樣例中我們提供如下的功能場(chǎng)景：
- 使用RocketMQ發(fā)送三種類型的消息：同步消息、異步消息和單向消息脐供。其中前兩種消息是可靠的，因?yàn)闀?huì)有發(fā)送是否成功的應(yīng)答卖陵。
- 使用RocketMQ來(lái)消費(fèi)接收到的消息儡嘶。
1.1 加入依賴：
maven:
<dependency>
<groupId>org.apache.rocketmq</groupId>
<artifactId>rocketmq-client</artifactId>
<version>4.3.0</version>
</dependency>
gradle
compile 'org.apache.rocketmq:rocketmq-client:4.3.0'
1.2 消息發(fā)送
1、Producer端發(fā)送同步消息
這種可靠性同步地發(fā)送方式使用的比較廣泛芜飘，比如：重要的消息通知务豺，短信通知。
public class SyncProducer {
public static void main(String[] args) throws Exception {
// 實(shí)例化消息生產(chǎn)者Producer
DefaultMQProducer producer = new DefaultMQProducer("please_rename_unique_group_name");
// 設(shè)置NameServer的地址
producer.setNamesrvAddr("localhost:9876");
// 啟動(dòng)Producer實(shí)例
producer.start();
for (int i = 0; i < 100; i++) {
// 創(chuàng)建消息嗦明，并指定Topic笼沥，Tag和消息體
Message msg = new Message("TopicTest" /* Topic */,
"TagA" /* Tag */,
("Hello RocketMQ " + i).getBytes(RemotingHelper.DEFAULT_CHARSET) /* Message body */
);
// 發(fā)送消息到一個(gè)Broker
SendResult sendResult = producer.send(msg);
// 通過(guò)sendResult返回消息是否成功送達(dá)
System.out.printf("%s%n", sendResult);
}
// 如果不再發(fā)送消息，關(guān)閉Producer實(shí)例娶牌。
producer.shutdown();
}
}
2奔浅、發(fā)送異步消息
異步消息通常用在對(duì)響應(yīng)時(shí)間敏感的業(yè)務(wù)場(chǎng)景，即發(fā)送端不能容忍長(zhǎng)時(shí)間地等待Broker的響應(yīng)诗良。
public class AsyncProducer {
public static void main(String[] args) throws Exception {
// 實(shí)例化消息生產(chǎn)者Producer
DefaultMQProducer producer = new DefaultMQProducer("please_rename_unique_group_name");
// 設(shè)置NameServer的地址
producer.setNamesrvAddr("localhost:9876");
// 啟動(dòng)Producer實(shí)例
producer.start();
producer.setRetryTimesWhenSendAsyncFailed(0);
for (int i = 0; i < 100; i++) {
final int index = i;
// 創(chuàng)建消息汹桦，并指定Topic，Tag和消息體
Message msg = new Message("TopicTest",
"TagA",
"OrderID188",
"Hello world".getBytes(RemotingHelper.DEFAULT_CHARSET));
// SendCallback接收異步返回結(jié)果的回調(diào)
producer.send(msg, new SendCallback() {
@Override
public void onSuccess(SendResult sendResult) {
System.out.printf("%-10d OK %s %n", index,
sendResult.getMsgId());
}
@Override
public void onException(Throwable e) {
System.out.printf("%-10d Exception %s %n", index, e);
e.printStackTrace();
}
});
}
// 如果不再發(fā)送消息鉴裹，關(guān)閉Producer實(shí)例舞骆。
producer.shutdown();
}
}
3、單向發(fā)送消息
這種方式主要用在不特別關(guān)心發(fā)送結(jié)果的場(chǎng)景径荔，例如日志發(fā)送督禽。
public class OnewayProducer {
public static void main(String[] args) throws Exception{
// 實(shí)例化消息生產(chǎn)者Producer
DefaultMQProducer producer = new DefaultMQProducer("please_rename_unique_group_name");
// 設(shè)置NameServer的地址
producer.setNamesrvAddr("localhost:9876");
// 啟動(dòng)Producer實(shí)例
producer.start();
for (int i = 0; i < 100; i++) {
// 創(chuàng)建消息，并指定Topic总处，Tag和消息體
Message msg = new Message("TopicTest" /* Topic */,
"TagA" /* Tag */,
("Hello RocketMQ " + i).getBytes(RemotingHelper.DEFAULT_CHARSET) /* Message body */
);
// 發(fā)送單向消息狈惫，沒(méi)有任何返回結(jié)果
producer.sendOneway(msg);
}
// 如果不再發(fā)送消息，關(guān)閉Producer實(shí)例鹦马。
producer.shutdown();
}
}
1.3 消費(fèi)消息
public class Consumer {
public static void main(String[] args) throws InterruptedException, MQClientException {
// 實(shí)例化消費(fèi)者
DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("please_rename_unique_group_name");
// 設(shè)置NameServer的地址
consumer.setNamesrvAddr("localhost:9876");
// 訂閱一個(gè)或者多個(gè)Topic胧谈，以及Tag來(lái)過(guò)濾需要消費(fèi)的消息
consumer.subscribe("TopicTest", "*");
// 注冊(cè)回調(diào)實(shí)現(xiàn)類來(lái)處理從broker拉取回來(lái)的消息
consumer.registerMessageListener(new MessageListenerConcurrently() {
@Override
public ConsumeConcurrentlyStatus consumeMessage(List<MessageExt> msgs, ConsumeConcurrentlyContext context) {
System.out.printf("%s Receive New Messages: %s %n", Thread.currentThread().getName(), msgs);
// 標(biāo)記該消息已經(jīng)被成功消費(fèi)
return ConsumeConcurrentlyStatus.CONSUME_SUCCESS;
}
});
// 啟動(dòng)消費(fèi)者實(shí)例
consumer.start();
System.out.printf("Consumer Started.%n");
}
}
2 順序消息樣例
消息有序指的是可以按照消息的發(fā)送順序來(lái)消費(fèi)(FIFO)玖院。RocketMQ可以嚴(yán)格的保證消息有序，可以分為分區(qū)有序或者全局有序第岖。
順序消費(fèi)的原理解析难菌，在默認(rèn)的情況下消息發(fā)送會(huì)采取Round Robin輪詢方式把消息發(fā)送到不同的queue(分區(qū)隊(duì)列)；而消費(fèi)消息的時(shí)候從多個(gè)queue上拉取消息蔑滓，這種情況發(fā)送和消費(fèi)是不能保證順序郊酒。但是如果控制發(fā)送的順序消息只依次發(fā)送到同一個(gè)queue中，消費(fèi)的時(shí)候只從這個(gè)queue上依次拉取键袱，則就保證了順序燎窘。當(dāng)發(fā)送和消費(fèi)參與的queue只有一個(gè)，則是全局有序蹄咖；如果多個(gè)queue參與褐健，則為分區(qū)有序，即相對(duì)每個(gè)queue澜汤，消息都是有序的蚜迅。
下面用訂單進(jìn)行分區(qū)有序的示例。一個(gè)訂單的順序流程是：創(chuàng)建俊抵、付款谁不、推送、完成徽诲。訂單號(hào)相同的消息會(huì)被先后發(fā)送到同一個(gè)隊(duì)列中刹帕，消費(fèi)時(shí)，同一個(gè)OrderId獲取到的肯定是同一個(gè)隊(duì)列谎替。
2.1 順序消息生產(chǎn)
package org.apache.rocketmq.example.order2;
import org.apache.rocketmq.client.producer.DefaultMQProducer;
import org.apache.rocketmq.client.producer.MessageQueueSelector;
import org.apache.rocketmq.client.producer.SendResult;
import org.apache.rocketmq.common.message.Message;
import org.apache.rocketmq.common.message.MessageQueue;
import java.text.SimpleDateFormat;
import java.util.ArrayList;
import java.util.Date;
import java.util.List;
/**
* Producer偷溺，發(fā)送順序消息
*/
public class Producer {
public static void main(String[] args) throws Exception {
DefaultMQProducer producer = new DefaultMQProducer("please_rename_unique_group_name");
producer.setNamesrvAddr("127.0.0.1:9876");
producer.start();
String[] tags = new String[]{"TagA", "TagC", "TagD"};
// 訂單列表
List<OrderStep> orderList = new Producer().buildOrders();
Date date = new Date();
SimpleDateFormat sdf = new SimpleDateFormat("yyyy-MM-dd HH:mm:ss");
String dateStr = sdf.format(date);
for (int i = 0; i < 10; i++) {
// 加個(gè)時(shí)間前綴
String body = dateStr + " Hello RocketMQ " + orderList.get(i);
Message msg = new Message("TopicTest", tags[i % tags.length], "KEY" + i, body.getBytes());
SendResult sendResult = producer.send(msg, new MessageQueueSelector() {
@Override
public MessageQueue select(List<MessageQueue> mqs, Message msg, Object arg) {
Long id = (Long) arg; //根據(jù)訂單id選擇發(fā)送queue
long index = id % mqs.size();
return mqs.get((int) index);
}
}, orderList.get(i).getOrderId());//訂單id
System.out.println(String.format("SendResult status:%s, queueId:%d, body:%s",
sendResult.getSendStatus(),
sendResult.getMessageQueue().getQueueId(),
body));
}
producer.shutdown();
}
/**
* 訂單的步驟
*/
private static class OrderStep {
private long orderId;
private String desc;
public long getOrderId() {
return orderId;
}
public void setOrderId(long orderId) {
this.orderId = orderId;
}
public String getDesc() {
return desc;
}
public void setDesc(String desc) {
this.desc = desc;
}
@Override
public String toString() {
return "OrderStep{" +
"orderId=" + orderId +
", desc='" + desc + '\'' +
'}';
}
}
/**
* 生成模擬訂單數(shù)據(jù)
*/
private List<OrderStep> buildOrders() {
List<OrderStep> orderList = new ArrayList<OrderStep>();
OrderStep orderDemo = new OrderStep();
orderDemo.setOrderId(15103111039L);
orderDemo.setDesc("創(chuàng)建");
orderList.add(orderDemo);
orderDemo = new OrderStep();
orderDemo.setOrderId(15103111065L);
orderDemo.setDesc("創(chuàng)建");
orderList.add(orderDemo);
orderDemo = new OrderStep();
orderDemo.setOrderId(15103111039L);
orderDemo.setDesc("付款");
orderList.add(orderDemo);
orderDemo = new OrderStep();
orderDemo.setOrderId(15103117235L);
orderDemo.setDesc("創(chuàng)建");
orderList.add(orderDemo);
orderDemo = new OrderStep();
orderDemo.setOrderId(15103111065L);
orderDemo.setDesc("付款");
orderList.add(orderDemo);
orderDemo = new OrderStep();
orderDemo.setOrderId(15103117235L);
orderDemo.setDesc("付款");
orderList.add(orderDemo);
orderDemo = new OrderStep();
orderDemo.setOrderId(15103111065L);
orderDemo.setDesc("完成");
orderList.add(orderDemo);
orderDemo = new OrderStep();
orderDemo.setOrderId(15103111039L);
orderDemo.setDesc("推送");
orderList.add(orderDemo);
orderDemo = new OrderStep();
orderDemo.setOrderId(15103117235L);
orderDemo.setDesc("完成");
orderList.add(orderDemo);
orderDemo = new OrderStep();
orderDemo.setOrderId(15103111039L);
orderDemo.setDesc("完成");
orderList.add(orderDemo);
return orderList;
}
}
2.2 順序消費(fèi)消息
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.common.message.MessageExt;
import java.util.List;
package org.apache.rocketmq.example.order2;
import org.apache.rocketmq.client.consumer.DefaultMQPushConsumer;
import org.apache.rocketmq.client.consumer.listener.ConsumeOrderlyContext;
import org.apache.rocketmq.client.consumer.listener.ConsumeOrderlyStatus;
import org.apache.rocketmq.client.consumer.listener.MessageListenerOrderly;
import org.apache.rocketmq.common.consumer.ConsumeFromWhere;
import org.apache.rocketmq.common.message.MessageExt;
import java.util.List;
import java.util.Random;
import java.util.concurrent.TimeUnit;
/**
* 順序消息消費(fèi)，帶事務(wù)方式（應(yīng)用可控制Offset什么時(shí)候提交）
*/
public class ConsumerInOrder {
public static void main(String[] args) throws Exception {
DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("please_rename_unique_group_name_3");
consumer.setNamesrvAddr("127.0.0.1:9876");
/**
* 設(shè)置Consumer第一次啟動(dòng)是從隊(duì)列頭部開(kāi)始消費(fèi)還是隊(duì)列尾部開(kāi)始消費(fèi)<br>
* 如果非第一次啟動(dòng)钱贯，那么按照上次消費(fèi)的位置繼續(xù)消費(fèi)
*/
consumer.setConsumeFromWhere(ConsumeFromWhere.CONSUME_FROM_FIRST_OFFSET);
consumer.subscribe("TopicTest", "TagA || TagC || TagD");
consumer.registerMessageListener(new MessageListenerOrderly() {
Random random = new Random();
@Override
public ConsumeOrderlyStatus consumeMessage(List<MessageExt> msgs, ConsumeOrderlyContext context) {
context.setAutoCommit(true);
for (MessageExt msg : msgs) {
// 可以看到每個(gè)queue有唯一的consume線程來(lái)消費(fèi), 訂單對(duì)每個(gè)queue(分區(qū))有序
System.out.println("consumeThread=" + Thread.currentThread().getName() + "queueId=" + msg.getQueueId() + ", content:" + new String(msg.getBody()));
}
try {
//模擬業(yè)務(wù)邏輯處理中...
TimeUnit.SECONDS.sleep(random.nextInt(10));
} catch (Exception e) {
e.printStackTrace();
}
return ConsumeOrderlyStatus.SUCCESS;
}
});
consumer.start();
System.out.println("Consumer Started.");
}
}
3 延時(shí)消息樣例
3.1 啟動(dòng)消費(fèi)者等待傳入訂閱消息
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.common.message.MessageExt;
import java.util.List;
public class ScheduledMessageConsumer {
public static void main(String[] args) throws Exception {
// 實(shí)例化消費(fèi)者
DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("ExampleConsumer");
// 訂閱Topics
consumer.subscribe("TestTopic", "*");
// 注冊(cè)消息監(jiān)聽(tīng)者
consumer.registerMessageListener(new MessageListenerConcurrently() {
@Override
public ConsumeConcurrentlyStatus consumeMessage(List<MessageExt> messages, ConsumeConcurrentlyContext context) {
for (MessageExt message : messages) {
// Print approximate delay time period
System.out.println("Receive message[msgId=" + message.getMsgId() + "] " + (System.currentTimeMillis() - message.getStoreTimestamp()) + "ms later");
}
return ConsumeConcurrentlyStatus.CONSUME_SUCCESS;
}
});
// 啟動(dòng)消費(fèi)者
consumer.start();
}
}
3.2 發(fā)送延時(shí)消息
import org.apache.rocketmq.client.producer.DefaultMQProducer;
import org.apache.rocketmq.common.message.Message;
public class ScheduledMessageProducer {
public static void main(String[] args) throws Exception {
// 實(shí)例化一個(gè)生產(chǎn)者來(lái)產(chǎn)生延時(shí)消息
DefaultMQProducer producer = new DefaultMQProducer("ExampleProducerGroup");
// 啟動(dòng)生產(chǎn)者
producer.start();
int totalMessagesToSend = 100;
for (int i = 0; i < totalMessagesToSend; i++) {
Message message = new Message("TestTopic", ("Hello scheduled message " + i).getBytes());
// 設(shè)置延時(shí)等級(jí)3,這個(gè)消息將在10s之后發(fā)送(現(xiàn)在只支持固定的幾個(gè)時(shí)間,詳看delayTimeLevel)
message.setDelayTimeLevel(3);
// 發(fā)送消息
producer.send(message);
}
// 關(guān)閉生產(chǎn)者
producer.shutdown();
}
}
3.3 驗(yàn)證
您將會(huì)看到消息的消費(fèi)比存儲(chǔ)時(shí)間晚10秒挫掏。
3.4 延時(shí)消息的使用場(chǎng)景
比如電商里，提交了一個(gè)訂單就可以發(fā)送一個(gè)延時(shí)消息喷舀，1h后去檢查這個(gè)訂單的狀態(tài)砍濒，如果還是未付款就取消訂單釋放庫(kù)存。
3.5 延時(shí)消息的使用限制
// org/apache/rocketmq/store/config/MessageStoreConfig.java
private String messageDelayLevel = "1s 5s 10s 30s 1m 2m 3m 4m 5m 6m 7m 8m 9m 10m 20m 30m 1h 2h";
現(xiàn)在RocketMq并不支持任意時(shí)間的延時(shí)硫麻，需要設(shè)置幾個(gè)固定的延時(shí)等級(jí)爸邢，從1s到2h分別對(duì)應(yīng)著等級(jí)1到18
消息消費(fèi)失敗會(huì)進(jìn)入延時(shí)消息隊(duì)列，消息發(fā)送時(shí)間與設(shè)置的延時(shí)等級(jí)和重試次數(shù)有關(guān)拿愧，詳見(jiàn)代碼SendMessageProcessor.java
4 批量消息樣例
批量發(fā)送消息能顯著提高傳遞小消息的性能杠河。限制是這些批量消息應(yīng)該有相同的topic，相同的waitStoreMsgOK，而且不能是延時(shí)消息券敌。此外唾戚，這一批消息的總大小不應(yīng)超過(guò)4MB。
4.1 發(fā)送批量消息
如果您每次只發(fā)送不超過(guò)4MB的消息待诅，則很容易使用批處理叹坦，樣例如下：
String topic = "BatchTest";
List<Message> messages = new ArrayList<>();
messages.add(new Message(topic, "TagA", "OrderID001", "Hello world 0".getBytes()));
messages.add(new Message(topic, "TagA", "OrderID002", "Hello world 1".getBytes()));
messages.add(new Message(topic, "TagA", "OrderID003", "Hello world 2".getBytes()));
try {
producer.send(messages);
} catch (Exception e) {
e.printStackTrace();
//處理error
}
4.2 消息列表分割
復(fù)雜度只有當(dāng)你發(fā)送大批量時(shí)才會(huì)增長(zhǎng)，你可能不確定它是否超過(guò)了大小限制（4MB）卑雁。這時(shí)候你最好把你的消息列表分割一下：
public class ListSplitter implements Iterator<List<Message>> {
private final int SIZE_LIMIT = 1024 * 1024 * 4;
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; // 增加日志的開(kāi)銷20字節(jié)
if (tmpSize > SIZE_LIMIT) {
//單個(gè)消息超過(guò)了最大的限制
//忽略,否則會(huì)阻塞分裂的進(jìn)程
if (nextIndex - currIndex == 0) {
//假如下一個(gè)子列表沒(méi)有元素,則添加這個(gè)子列表然后退出循環(huán),否則只是退出循環(huán)
nextIndex++;
}
break;
}
if (tmpSize + totalSize > SIZE_LIMIT) {
break;
} else {
totalSize += tmpSize;
}
}
List<Message> subList = messages.subList(currIndex, nextIndex);
currIndex = nextIndex;
return subList;
}
}
//把大的消息分裂成若干個(gè)小的消息
ListSplitter splitter = new ListSplitter(messages);
while (splitter.hasNext()) {
try {
List<Message> listItem = splitter.next();
producer.send(listItem);
} catch (Exception e) {
e.printStackTrace();
//處理error
}
}
5 過(guò)濾消息樣例
在大多數(shù)情況下募书，TAG是一個(gè)簡(jiǎn)單而有用的設(shè)計(jì)，其可以來(lái)選擇您想要的消息测蹲。例如：
DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("CID_EXAMPLE");
consumer.subscribe("TOPIC", "TAGA || TAGB || TAGC");
消費(fèi)者將接收包含TAGA或TAGB或TAGC的消息莹捡。但是限制是一個(gè)消息只能有一個(gè)標(biāo)簽，這對(duì)于復(fù)雜的場(chǎng)景可能不起作用扣甲。在這種情況下篮赢，可以使用SQL表達(dá)式篩選消息。SQL特性可以通過(guò)發(fā)送消息時(shí)的屬性來(lái)進(jìn)行計(jì)算琉挖。在RocketMQ定義的語(yǔ)法下启泣，可以實(shí)現(xiàn)一些簡(jiǎn)單的邏輯。下面是一個(gè)例子：
------------
| message |
|----------| a > 5 AND b = 'abc'
| a = 10 | --------------------> Gotten
| b = 'abc'|
| c = true |
------------
------------
| message |
|----------| a > 5 AND b = 'abc'
| a = 1 | --------------------> Missed
| b = 'abc'|
| c = true |
------------
5.1 基本語(yǔ)法
RocketMQ只定義了一些基本語(yǔ)法來(lái)支持這個(gè)特性粹排。你也可以很容易地?cái)U(kuò)展它种远。
- 數(shù)值比較涩澡，比如：>顽耳，>=，<妙同，<=射富，BETWEEN，=粥帚；
- 字符比較胰耗，比如：=，<>芒涡，IN柴灯；
- IS NULL 或者 IS NOT NULL；
- 邏輯符號(hào) AND费尽，OR赠群，NOT；
常量支持類型為：
- 數(shù)值旱幼，比如：123查描，3.1415；
- 字符，比如：'abc'冬三，必須用單引號(hào)包裹起來(lái)匀油；
- NULL，特殊的常量
- 布爾值勾笆，TRUE 或 FALSE
只有使用push模式的消費(fèi)者才能用使用SQL92標(biāo)準(zhǔn)的sql語(yǔ)句敌蚜，接口如下：
public void subscribe(finalString topic, final MessageSelector messageSelector)
5.2 使用樣例
1、生產(chǎn)者樣例
發(fā)送消息時(shí)窝爪，你能通過(guò)putUserProperty來(lái)設(shè)置消息的屬性
DefaultMQProducer producer = new DefaultMQProducer("please_rename_unique_group_name");
producer.start();
Message msg = new Message("TopicTest",
tag,
("Hello RocketMQ " + i).getBytes(RemotingHelper.DEFAULT_CHARSET)
);
// 設(shè)置一些屬性
msg.putUserProperty("a", String.valueOf(i));
SendResult sendResult = producer.send(msg);
producer.shutdown();
2钝侠、消費(fèi)者樣例
用MessageSelector.bySql來(lái)使用sql篩選消息
DefaultMQPushConsumer consumer = new DefaultMQPushConsumer("please_rename_unique_group_name_4");
// 只有訂閱的消息有這個(gè)屬性a, a >=0 and a <= 3
consumer.subscribe("TopicTest", MessageSelector.bySql("a between 0 and 3");
consumer.registerMessageListener(new MessageListenerConcurrently() {
@Override
public ConsumeConcurrentlyStatus consumeMessage(List<MessageExt> msgs, ConsumeConcurrentlyContext context) {
return ConsumeConcurrentlyStatus.CONSUME_SUCCESS;
}
});
consumer.start();
6 消息事務(wù)樣例
事務(wù)消息共有三種狀態(tài)，提交狀態(tài)酸舍、回滾狀態(tài)帅韧、中間狀態(tài)：
- TransactionStatus.CommitTransaction: 提交事務(wù)，它允許消費(fèi)者消費(fèi)此消息啃勉。
- TransactionStatus.RollbackTransaction: 回滾事務(wù)忽舟，它代表該消息將被刪除，不允許被消費(fèi)淮阐。
- TransactionStatus.Unknown: 中間狀態(tài)叮阅，它代表需要檢查消息隊(duì)列來(lái)確定狀態(tài)。
6.1 發(fā)送事務(wù)消息樣例
1泣特、創(chuàng)建事務(wù)性生產(chǎn)者
使用 TransactionMQProducer類創(chuàng)建生產(chǎn)者浩姥，并指定唯一的 ProducerGroup，就可以設(shè)置自定義線程池來(lái)處理這些檢查請(qǐng)求状您。執(zhí)行本地事務(wù)后勒叠、需要根據(jù)執(zhí)行結(jié)果對(duì)消息隊(duì)列進(jìn)行回復(fù)「嗝希回傳的事務(wù)狀態(tài)在請(qǐng)參考前一節(jié)眯分。
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.common.message.MessageExt;
import java.util.List;
public class TransactionProducer {
public static void main(String[] args) throws MQClientException, InterruptedException {
TransactionListener transactionListener = new TransactionListenerImpl();
TransactionMQProducer producer = new TransactionMQProducer("please_rename_unique_group_name");
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;
}
});
producer.setExecutorService(executorService);
producer.setTransactionListener(transactionListener);
producer.start();
String[] tags = new String[] {"TagA", "TagB", "TagC", "TagD", "TagE"};
for (int i = 0; i < 10; i++) {
try {
Message msg =
new Message("TopicTest1234", 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();
}
}
2、實(shí)現(xiàn)事務(wù)的監(jiān)聽(tīng)接口
當(dāng)發(fā)送半消息成功時(shí)柒桑，我們使用 executeLocalTransaction 方法來(lái)執(zhí)行本地事務(wù)弊决。它返回前一節(jié)中提到的三個(gè)事務(wù)狀態(tài)之一。checkLocalTranscation 方法用于檢查本地事務(wù)狀態(tài)魁淳，并回應(yīng)消息隊(duì)列的檢查請(qǐng)求飘诗。它也是返回前一節(jié)中提到的三個(gè)事務(wù)狀態(tài)之一。
public class TransactionListenerImpl implements TransactionListener {
private AtomicInteger transactionIndex = new AtomicInteger(0);
private ConcurrentHashMap<String, Integer> localTrans = new ConcurrentHashMap<>();
@Override
public LocalTransactionState executeLocalTransaction(Message msg, Object arg) {
int value = transactionIndex.getAndIncrement();
int status = value % 3;
localTrans.put(msg.getTransactionId(), status);
return LocalTransactionState.UNKNOW;
}
@Override
public LocalTransactionState checkLocalTransaction(MessageExt msg) {
Integer status = localTrans.get(msg.getTransactionId());
if (null != status) {
switch (status) {
case 0:
return LocalTransactionState.UNKNOW;
case 1:
return LocalTransactionState.COMMIT_MESSAGE;
case 2:
return LocalTransactionState.ROLLBACK_MESSAGE;
}
}
return LocalTransactionState.COMMIT_MESSAGE;
}
}
6.2 事務(wù)消息使用上的限制
- 事務(wù)消息不支持延時(shí)消息和批量消息界逛。
- 為了避免單個(gè)消息被檢查太多次而導(dǎo)致半隊(duì)列消息累積昆稿，我們默認(rèn)將單個(gè)消息的檢查次數(shù)限制為 15 次，但是用戶可以通過(guò) Broker 配置文件的
transactionCheckMax參數(shù)來(lái)修改此限制仇奶。如果已經(jīng)檢查某條消息超過(guò) N 次的話（ N =transactionCheckMax） 則 Broker 將丟棄此消息貌嫡，并在默認(rèn)情況下同時(shí)打印錯(cuò)誤日志比驻。用戶可以通過(guò)重寫(xiě)AbstractTransactionCheckListener類來(lái)修改這個(gè)行為。 - 事務(wù)消息將在 Broker 配置文件中的參數(shù) transactionMsgTimeout 這樣的特定時(shí)間長(zhǎng)度之后被檢查岛抄。當(dāng)發(fā)送事務(wù)消息時(shí)别惦，用戶還可以通過(guò)設(shè)置用戶屬性 CHECK_IMMUNITY_TIME_IN_SECONDS 來(lái)改變這個(gè)限制，該參數(shù)優(yōu)先于
transactionMsgTimeout參數(shù)夫椭。 - 事務(wù)性消息可能不止一次被檢查或消費(fèi)掸掸。
- 提交給用戶的目標(biāo)主題消息可能會(huì)失敗，目前這依日志的記錄而定蹭秋。它的高可用性通過(guò) RocketMQ 本身的高可用性機(jī)制來(lái)保證扰付，如果希望確保事務(wù)消息不丟失、并且事務(wù)完整性得到保證仁讨，建議使用同步的雙重寫(xiě)入機(jī)制羽莺。
- 事務(wù)消息的生產(chǎn)者 ID 不能與其他類型消息的生產(chǎn)者 ID 共享。與其他類型的消息不同洞豁，事務(wù)消息允許反向查詢盐固、MQ服務(wù)器能通過(guò)它們的生產(chǎn)者 ID 查詢到消費(fèi)者。
7 Logappender樣例
RocketMQ日志提供log4j丈挟、log4j2和logback日志框架作為業(yè)務(wù)應(yīng)用刁卜，下面是配置樣例
7.1 log4j樣例
按下面樣例使用log4j屬性配置
log4j.appender.mq=org.apache.rocketmq.logappender.log4j.RocketmqLog4jAppender
log4j.appender.mq.Tag=yourTag
log4j.appender.mq.Topic=yourLogTopic
log4j.appender.mq.ProducerGroup=yourLogGroup
log4j.appender.mq.NameServerAddress=yourRocketmqNameserverAddress
log4j.appender.mq.layout=org.apache.log4j.PatternLayout
log4j.appender.mq.layout.ConversionPattern=%d{yyyy-MM-dd HH:mm:ss} %-4r [%t] (%F:%L) %-5p - %m%n
按下面樣例使用log4j xml配置來(lái)使用異步添加日志
<appender name="mqAppender1"class="org.apache.rocketmq.logappender.log4j.RocketmqLog4jAppender">
<param name="Tag" value="yourTag" />
<param name="Topic" value="yourLogTopic" />
<param name="ProducerGroup" value="yourLogGroup" />
<param name="NameServerAddress" value="yourRocketmqNameserverAddress"/>
<layout class="org.apache.log4j.PatternLayout">
<param name="ConversionPattern" value="%d{yyyy-MM-dd HH:mm:ss}-%p %t %c - %m%n" />
</layout>
</appender>
<appender name="mqAsyncAppender1"class="org.apache.log4j.AsyncAppender">
<param name="BufferSize" value="1024" />
<param name="Blocking" value="false" />
<appender-ref ref="mqAppender1"/>
</appender>
7.2 log4j2樣例
用log4j2時(shí)，配置如下曙咽，如果想要非阻塞蛔趴，只需要使用異步添加引用即可
<RocketMQ name="rocketmqAppender" producerGroup="yourLogGroup" nameServerAddress="yourRocketmqNameserverAddress"
topic="yourLogTopic" tag="yourTag">
<PatternLayout pattern="%d [%p] hahahah %c %m%n"/>
</RocketMQ>
7.3 logback樣例
<appender name="mqAppender1"class="org.apache.rocketmq.logappender.logback.RocketmqLogbackAppender">
<tag>yourTag</tag>
<topic>yourLogTopic</topic>
<producerGroup>yourLogGroup</producerGroup>
<nameServerAddress>yourRocketmqNameserverAddress</nameServerAddress>
<layout>
<pattern>%date %p %t - %m%n</pattern>
</layout>
</appender>
<appender name="mqAsyncAppender1"class="ch.qos.logback.classic.AsyncAppender">
<queueSize>1024</queueSize>
<discardingThreshold>80</discardingThreshold>
<maxFlushTime>2000</maxFlushTime>
<neverBlock>true</neverBlock>
<appender-ref ref="mqAppender1"/>
</appender>
8 OpenMessaging樣例
OpenMessaging旨在建立消息和流處理規(guī)范，以為金融例朱、電子商務(wù)孝情、物聯(lián)網(wǎng)和大數(shù)據(jù)領(lǐng)域提供通用框架及工業(yè)級(jí)指導(dǎo)方案。在分布式異構(gòu)環(huán)境中茉继，設(shè)計(jì)原則是面向云咧叭、簡(jiǎn)單、靈活和獨(dú)立于語(yǔ)言烁竭。符合這些規(guī)范將幫助企業(yè)方便的開(kāi)發(fā)跨平臺(tái)和操作系統(tǒng)的異構(gòu)消息傳遞應(yīng)用程序。提供了openmessaging-api 0.3.0-alpha的部分實(shí)現(xiàn)吉挣，下面的示例演示如何基于OpenMessaging訪問(wèn)RocketMQ派撕。
8.1 OMSProducer樣例
下面的示例演示如何在同步、異步或單向傳輸中向RocketMQ代理發(fā)送消息睬魂。
import io.openmessaging.Future;
import io.openmessaging.FutureListener;
import io.openmessaging.Message;
import io.openmessaging.MessagingAccessPoint;
import io.openmessaging.OMS;
import io.openmessaging.producer.Producer;
import io.openmessaging.producer.SendResult;
import java.nio.charset.Charset;
import java.util.concurrent.CountDownLatch;
public class SimpleProducer {
public static void main(String[] args) {
final MessagingAccessPoint messagingAccessPoint =
OMS.getMessagingAccessPoint("oms:rocketmq://localhost:9876/default:default");
final Producer producer = messagingAccessPoint.createProducer();
messagingAccessPoint.startup();
System.out.printf("MessagingAccessPoint startup OK%n");
producer.startup();
System.out.printf("Producer startup OK%n");
{
Message message = producer.createBytesMessage("OMS_HELLO_TOPIC", "OMS_HELLO_BODY".getBytes(Charset.forName("UTF-8")));
SendResult sendResult = producer.send(message);
//final Void aVoid = result.get(3000L);
System.out.printf("Send async message OK, msgId: %s%n", sendResult.messageId());
}
final CountDownLatch countDownLatch = new CountDownLatch(1);
{
final Future<SendResult> result = producer.sendAsync(producer.createBytesMessage("OMS_HELLO_TOPIC", "OMS_HELLO_BODY".getBytes(Charset.forName("UTF-8"))));
result.addListener(new FutureListener<SendResult>() {
@Override
public void operationComplete(Future<SendResult> future) {
if (future.getThrowable() != null) {
System.out.printf("Send async message Failed, error: %s%n", future.getThrowable().getMessage());
} else {
System.out.printf("Send async message OK, msgId: %s%n", future.get().messageId());
}
countDownLatch.countDown();
}
});
}
{
producer.sendOneway(producer.createBytesMessage("OMS_HELLO_TOPIC", "OMS_HELLO_BODY".getBytes(Charset.forName("UTF-8"))));
System.out.printf("Send oneway message OK%n");
}
try {
countDownLatch.await();
Thread.sleep(500); // 等一些時(shí)間來(lái)發(fā)送消息
} catch (InterruptedException ignore) {
}
producer.shutdown();
}
}
8.2 OMSPullConsumer
用OMS PullConsumer 來(lái)從指定的隊(duì)列中拉取消息
import io.openmessaging.Message;
import io.openmessaging.MessagingAccessPoint;
import io.openmessaging.OMS;
import io.openmessaging.OMSBuiltinKeys;
import io.openmessaging.consumer.PullConsumer;
import io.openmessaging.producer.Producer;
import io.openmessaging.producer.SendResult;
public class SimplePullConsumer {
public static void main(String[] args) {
final MessagingAccessPoint messagingAccessPoint =
OMS.getMessagingAccessPoint("oms:rocketmq://localhost:9876/default:default");
messagingAccessPoint.startup();
final Producer producer = messagingAccessPoint.createProducer();
final PullConsumer consumer = messagingAccessPoint.createPullConsumer(
OMS.newKeyValue().put(OMSBuiltinKeys.CONSUMER_ID, "OMS_CONSUMER"));
messagingAccessPoint.startup();
System.out.printf("MessagingAccessPoint startup OK%n");
final String queueName = "TopicTest";
producer.startup();
Message msg = producer.createBytesMessage(queueName, "Hello Open Messaging".getBytes());
SendResult sendResult = producer.send(msg);
System.out.printf("Send Message OK. MsgId: %s%n", sendResult.messageId());
producer.shutdown();
consumer.attachQueue(queueName);
consumer.startup();
System.out.printf("Consumer startup OK%n");
// 運(yùn)行直到發(fā)現(xiàn)一個(gè)消息被發(fā)送了
boolean stop = false;
while (!stop) {
Message message = consumer.receive();
if (message != null) {
String msgId = message.sysHeaders().getString(Message.BuiltinKeys.MESSAGE_ID);
System.out.printf("Received one message: %s%n", msgId);
consumer.ack(msgId);
if (!stop) {
stop = msgId.equalsIgnoreCase(sendResult.messageId());
}
} else {
System.out.printf("Return without any message%n");
}
}
consumer.shutdown();
messagingAccessPoint.shutdown();
}
}
8.3 OMSPushConsumer
以下示范如何將 OMS PushConsumer 添加到指定的隊(duì)列终吼，并通過(guò) MessageListener 消費(fèi)這些消息。
import io.openmessaging.Message;
import io.openmessaging.MessagingAccessPoint;
import io.openmessaging.OMS;
import io.openmessaging.OMSBuiltinKeys;
import io.openmessaging.consumer.MessageListener;
import io.openmessaging.consumer.PushConsumer;
public class SimplePushConsumer {
public static void main(String[] args) {
final MessagingAccessPoint messagingAccessPoint = OMS
.getMessagingAccessPoint("oms:rocketmq://localhost:9876/default:default");
final PushConsumer consumer = messagingAccessPoint.
createPushConsumer(OMS.newKeyValue().put(OMSBuiltinKeys.CONSUMER_ID, "OMS_CONSUMER"));
messagingAccessPoint.startup();
System.out.printf("MessagingAccessPoint startup OK%n");
Runtime.getRuntime().addShutdownHook(new Thread(new Runnable() {
@Override
public void run() {
consumer.shutdown();
messagingAccessPoint.shutdown();
}
}));
consumer.attachQueue("OMS_HELLO_TOPIC", new MessageListener() {
@Override
public void onReceived(Message message, Context context) {
System.out.printf("Received one message: %s%n", message.sysHeaders().getString(Message.BuiltinKeys.MESSAGE_ID));
context.ack();
}
});
consumer.startup();
System.out.printf("Consumer startup OK%n");
}
}
