RabbitMQ 延迟队列实现订单支付结果异步阶梯性通知

在第三方支付中，例如支付宝、或者微信，对于订单请求，第三方支付系统采用的是消息同步返回、异步通知+主动补偿查询的补偿机制。

由于互联网通信的不可靠性，例如双方网络、服务器、应用等因素的影响，不管是同步返回、异步通知、主动查询报文都可能出现超时无响应、报文丢失等情况，所以像支付业务，对结果的通知一般采用几种方案结合的补偿机制，不能完全依赖某一种机制。
例如一个支付结果的通知，一方面会在支付页面跳转时候返回支付结果（一般只用作前端展示使用，非最终状态），同时会采用后台异步通知机制（有前台、后台通知的，以后台异步通知结果为准），但由于前台跳转、后台结果通知都可能失效，因此还以定时补单+请求方主动查询接口作为辅助手段。
常见的补单操作，任务调度策略一般设定30秒、60秒、3分钟、6分钟、10分钟调度多次（以自己业务需要），如果调度接收到响应确认报文，补单成功，则中止对应订单的调度任务；如果超过补单上限次数，则停止补单，避免无谓的资源浪费。请求端随时可以发起请求报文查询对应订单的状态。

在日常开发中，对于网站前端来说，支付计费中心对于订单请求信息的处理也是通过消息同步返回、异步通知+主动补偿查询相结合的机制，其中对于订单的异步通知，目前的通知策略为3s、30s、60s、120s、180、300s的阶梯性通知。返回成功情况下就不继续通知了，本来打算使用将失败的消息写到数据库等待发送，然后每秒查询数据库获取消息通知前端。但觉得这样的处理方式太粗暴。存在以下缺点：

1 、每秒请求有点儿浪费资源； 2 、通知方式不稳定； 3 、无法承受大数据量等等

所以最终打算使用rabbitmq的消息延迟+死信队列来实现。消息模型如下：

producer发布消息，通过exchangeA的消息会被分发到QueueA,Consumer监听queueA，一旦有消息到来就被消费，这边的消费业务就是通知前端，如果通知失败，就创建一个延迟队列declareQueue，设置每个消息的ttl然后通过declare_exchange将消息分发到declare_queue,因为declare_queue没有consumer并且declare_queue中的消息设置了ttl，当ttl到期后，将通过DEX路由到queueA,被重新消费。

代码如下：DeclareQueue.java

    
    
package org.delayQueue;

import com.rabbitmq.client.BuiltinExchangeType;
import com.rabbitmq.client.Channel;
import com.rabbitmq.client.Connection;
import com.rabbitmq.client.ConnectionFactory;

public class DeclareQueue {
public static String EXCHANGE_NAME = "notifyExchange";

public static void init() {
ConnectionFactory factory = new ConnectionFactory();
factory.setHost( "localhost");
factory.setPort( 5672);

Connection connection = null;
try {
connection = factory.newConnection();
Channel channel = connection.createChannel();
channel.exchangeDeclare(EXCHANGE_NAME, BuiltinExchangeType.TOPIC);
String routingKey = "AliPaynotify";
String message = "http://localhost:8080/BossCenter/payGateway/notifyRecv.jsp?is_success=T¬ify_id=4ab9bed148d043d0bf75460706f7774a¬ify_time=2014-08-29+16%3A22%3A02¬ify_type=trade_status_sync&out_trade_no=1421712120109862&total_fee=424.42&trade_no=14217121201098611&trade_status=TRADE_SUCCESS";
channel.basicPublish(EXCHANGE_NAME, routingKey, null, message.getBytes());
System.out.println( " [x] Sent :" + message);
} catch (Exception e) {
// TODO Auto-generated catch block
e.printStackTrace();
} finally {
if (connection != null) {
try {
connection.close();
} catch (Exception ignore) {
}
}
}
}

public static void main(String args[]) {
init();
}

}
   
   

DeclareConsumer.java

    
    
package org.delayQueue;

import java.io.BufferedReader;
import java.io.IOException;
import java.io.InputStreamReader;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.Map.Entry;

import org.apache.http.HttpResponse;
import org.apache.http.client.ClientProtocolException;
import org.apache.http.client.HttpClient;
import org.apache.http.client.methods.HttpPost;
import org.apache.http.impl.client.DefaultHttpClient;

import com.rabbitmq.client.AMQP;
import com.rabbitmq.client.Channel;
import com.rabbitmq.client.Connection;
import com.rabbitmq.client.ConnectionFactory;
import com.rabbitmq.client.Consumer;
import com.rabbitmq.client.DefaultConsumer;
import com.rabbitmq.client.Envelope;

public class DeclareConsumer {
public static String EXCHANGE_NAME = "notifyExchange";
public static String QU_declare_15S = "Qu_declare_15s";
public static String EX_declare_15S = "EX_declare_15s";
public static String ROUTINGKEY = "AliPaynotify";
public static Connection connection = null;
public static Channel channel = null;
public static Channel DECLARE_15S_CHANNEL = null;
public static String declare_queue = "init";
public static String originalExpiration = "0";
public static void init() throws Exception {
ConnectionFactory factory = new ConnectionFactory();
factory.setHost("localhost");
factory.setPort(5672);
connection = factory.newConnection();
channel = connection.createChannel();
DECLARE_15S_CHANNEL = connection.createChannel();
}

public static void consume() {
try {
channel.exchangeDeclare(EXCHANGE_NAME, "topic");
final String queueName = channel.queueDeclare().getQueue();

channel.queueBind(queueName, EXCHANGE_NAME, ROUTINGKEY);
System.out.println(" [*] Waiting for messages. To exit press CTRL+C");
final Consumer consumer = new DefaultConsumer(channel) {
@Override
public void handleDelivery(String consumerTag, Envelope envelope, AMQP.BasicProperties properties, byte[] body) throws IOException {
String message = new String(body, "UTF-8");
Map <String, Object> headers = properties.getHeaders();
if (headers != null) {
List <Map<String, Object>> xDeath = (List <Map<String, Object>>) headers.get("x-death");
System.out.println("xDeath--- > " + xDeath);
if (xDeath != null && !xDeath.isEmpty()) {
Map <String, Object> entrys = xDeath.get(0);
// for(Entry <String, Object>
// entry:entrys.entrySet()){
// System.out.println(entry.getKey()+":"+entry.getValue());
// }
originalExpiration = entrys.get("original-expiration").toString();
}
}
System.out.println(" [x] Received '" + envelope.getRoutingKey() + "':'" + message + "'" + "time" + System.currentTimeMillis());
HttpClient httpClient = new DefaultHttpClient();
HttpPost post = new HttpPost(message);
HttpResponse response = httpClient.execute(post);
BufferedReader inreader = null;
if (response.getStatusLine().getStatusCode() == 200) {
inreader = new BufferedReader(new InputStreamReader(response.getEntity().getContent(), "UTF-8"));
StringBuffer responseBody = new StringBuffer();
String line = null;
while ((line = inreader.readLine()) != null) {
responseBody.append(line);
}
if (!responseBody.equals("success")) {
// putDeclre15s(message);
if (originalExpiration.equals("0")) {
putDeclreQueue(message, 3000, QU_declare_15S);
}
if (originalExpiration.equals("3000")) {
putDeclreQueue(message, 30000, QU_declare_15S);
}
if (originalExpiration.equals("30000")) {
putDeclreQueue(message, 60000, QU_declare_15S);
}
if (originalExpiration.equals("60000")) {
putDeclreQueue(message, 120000, QU_declare_15S);
}
if (originalExpiration.equals("120000")) {
putDeclreQueue(message, 180000, QU_declare_15S);
}
if (originalExpiration.equals("180000")) {
putDeclreQueue(message, 300000, QU_declare_15S);
}
if (originalExpiration.equals("300000")) {
// channel.basicConsume(QU_declare_300S,true, this);
System.out.println("finish notify");
}
}
} else {
System.out.println(response.getStatusLine().getStatusCode());
}
}
};

channel.basicConsume(queueName, true, consumer);
} catch (Exception e) {
e.printStackTrace();
} finally {
}
}



static Map <String, Object> xdeathMap = new HashMap <String, Object>();
static List <Map<String, Object>> xDeath = new ArrayList <Map<String, Object>>();
static Map <String, Object> xdeathParam = new HashMap <String, Object>();

public static void putDeclre15s(String message) throws IOException {
channel.exchangeDeclare(EX_declare_15S, "topic");
Map <String, Object> args = new HashMap <String, Object>();
args.put("x-dead-letter-exchange", EXCHANGE_NAME);// 死信exchange
AMQP.BasicProperties.Builder builder = new AMQP.BasicProperties.Builder();
builder.expiration("3000").deliveryMode(2);// 设置消息TTL
AMQP.BasicProperties properties = builder.build();
channel.queueDeclare(QU_declare_15S, false, false, false, args);
channel.queueBind(QU_declare_15S, EX_declare_15S, ROUTINGKEY);
channel.basicPublish(EX_declare_15S, ROUTINGKEY, properties, message.getBytes());
System.out.println("send message in QA_DEFERRED_15S" + message + "time" + System.currentTimeMillis());
}

public static void putDeclreQueue(String message, int mis, String queue) throws IOException {
channel.exchangeDeclare(EX_declare_15S, "topic");
Map <String, Object> args = new HashMap <String, Object>();
args.put("x-dead-letter-exchange", EXCHANGE_NAME);// 死信exchange
AMQP.BasicProperties.Builder builder = new AMQP.BasicProperties.Builder();
builder.expiration(String.valueOf(mis)).deliveryMode(2);// 设置消息TTL
AMQP.BasicProperties properties = builder.build();
channel.queueDeclare(queue, false, false, false, args);
channel.queueBind(queue, EX_declare_15S, ROUTINGKEY);
channel.basicPublish(EX_declare_15S, ROUTINGKEY, properties, message.getBytes());
System.out.println("send message in " + queue + message + "time============" + System.currentTimeMillis());
}

public static void main(String args[]) throws Exception {
init();
consume();
}
}
   
   

消息 通过dlx转发的情况下，header头部会带有x-death的一个数组，里面包含消息的各项属性，比如说消息成为死信的原因reason，original-expiration这个字段表示消息在原来队列中的过期时间，根据这个值来确定下一次通知的延迟时间应该是多少秒。

运行结果如下：