仓库源文站点原文


layout: post title: "RocketMQ 位移提交源码分析" categories: RocketMQ tags: offsets 位移提交 源码分析

author: 张乘辉

RocketMQ 每次拉取完消息都会将消息存储到 PullRequest 对象中的 ProcessQueue 中:

org.apache.rocketmq.client.consumer.PullCallback#onSuccess

boolean dispathToConsume = processQueue.putMessage(pullResult.getMsgFoundList());

接着将消息放进消费线程中去执行:

org.apache.rocketmq.client.consumer.PullCallback#onSuccess

DefaultMQPushConsumerImpl.this.consumeMessageService.submitConsumeRequest(//
  pullResult.getMsgFoundList(), //
  processQueue, //
  pullRequest.getMessageQueue(), //
  dispathToConsume);

ConsumeMessageService 类实现消息消费的逻辑,它有两个实现类:

// 并发消息消费逻辑实现类
org.apache.rocketmq.client.impl.consumer.ConsumeMessageConcurrentlyService;
// 顺序消息消费逻辑实现类
org.apache.rocketmq.client.impl.consumer.ConsumeMessageOrderlyService;

这里我们只分析并发消费:

org.apache.rocketmq.client.impl.consumer.ConsumeMessageConcurrentlyService#submitConsumeRequest

ConsumeRequest consumeRequest = new ConsumeRequest(msgThis, processQueue, messageQueue);
try {
  this.consumeExecutor.submit(consumeRequest);
} catch (RejectedExecutionException e) {
    // ...
}

将消息消费任务封装成 ConsumeRequest 对象,然后将其交给消费线程池中去执行。

org.apache.rocketmq.client.impl.consumer.ConsumeMessageConcurrentlyService.ConsumeRequest#run:

if (!processQueue.isDropped()) {
    ConsumeMessageConcurrentlyService.this.processConsumeResult(status, context, this);
} else {
    log.warn("processQueue is dropped without process consume result. messageQueue={}, msgs={}", messageQueue, msgs);
}

ConsumeRequest 是一个实现了 Runnable 的类,因此消息消费的核心逻辑都写在了 run 方法中,如上代码是提交已消费位移的逻辑,当 ProcessQueue 没有被丢弃,则进行已消费位移的提交。

org.apache.rocketmq.client.impl.consumer.ConsumeMessageConcurrentlyService#processConsumeResult

// 移除已消费的消息,并返回已消费的
long offset = consumeRequest.getProcessQueue().removeMessage(consumeRequest.getMsgs());
if (offset >= 0 && !consumeRequest.getProcessQueue().isDropped()) {
    this.defaultMQPushConsumerImpl.getOffsetStore().updateOffset(consumeRequest.getMessageQueue(), offset, true);
}

移除已消费的位移,并返回最小位移量,如果最小位移量大于 0,并且 ProcessQueue 没有被丢弃,则更新本地缓存,

org.apache.rocketmq.client.impl.consumer.ProcessQueue#removeMessage

public long removeMessage(final List<MessageExt> msgs) {
  long result = -1;
  final long now = System.currentTimeMillis();
  try {
    this.lockTreeMap.writeLock().lockInterruptibly();
    this.lastConsumeTimestamp = now;
    try {
      if (!msgTreeMap.isEmpty()) {
        result = this.queueOffsetMax + 1;
        int removedCnt = 0;
                // 移除已消费的消息
        for (MessageExt msg : msgs) {
          MessageExt prev = msgTreeMap.remove(msg.getQueueOffset());
          if (prev != null) {
            removedCnt--;
          }
        }
        // 消息总量累加
        msgCount.addAndGet(removedCnt);
                // 返回消息容器中最小元素 key
        if (!msgTreeMap.isEmpty()) {
          result = msgTreeMap.firstKey();
        }
      }
    } finally {
      this.lockTreeMap.writeLock().unlock();
    }
  } catch (Throwable t) {
    log.error("removeMessage exception", t);
  }

  return result;
}

以上方法就是解答文章开头问题的关键,由于该方法是各个消费线程并发执行,因此需要对其进行加锁操作,msgTreeMap 是 ProcessQueue 的消息容器,它的格式如下:

private final TreeMap<Long, MessageExt> msgTreeMap = new TreeMap<>();

它是一个 TreeMap 结构,key 为消息位移,value 为消息数据,消息容器中,消息可以按照位移进行排序,那也就意味着,当消息消费完,只需要在消息容器中移除即可,然后返回消息容器中最小元素(最小位移),如下:

由于消息是按照位移进行排序,因此我们只需移除已消费的消息,并且确保不会将未消费的位移提交,就可避免了位移大的消息先消费导致消息丢失的问题了。

接下来我们继续看

org.apache.rocketmq.client.consumer.store.RemoteBrokerOffsetStore#updateOffset:

public void updateOffset(MessageQueue mq, long offset, boolean increaseOnly) {
    if (mq != null) {
        AtomicLong offsetOld = this.offsetTable.get(mq);
        if (null == offsetOld) {
            offsetOld = this.offsetTable.putIfAbsent(mq, new AtomicLong(offset));
        }

        if (null != offsetOld) {
            if (increaseOnly) {
                MixAll.compareAndIncreaseOnly(offsetOld, offset);
            } else {
                offsetOld.set(offset);
            }
        }
    }
}

offsetTable 为本地位移缓存容器,它的结构如下:

private ConcurrentMap<MessageQueue, AtomicLong> offsetTable = new ConcurrentHashMap<>();

它是一个 ConcurrentMap,一个线程安全容器,key 为 MessageQueue,value 为当前 MessageQueue 的消费位移,从源码看出,当前消费位移的更新,只能是递增更新。

在更新完本地缓存之后,RocketMQ 是如何将其提交到 broker 的呢?

org.apache.rocketmq.client.impl.factory.MQClientInstance#startScheduledTask:

this.scheduledExecutorService.scheduleAtFixedRate(new Runnable() {
    @Override
    public void run() {
        try {
            MQClientInstance.this.persistAllConsumerOffset();
        } catch (Exception e) {
            log.error("ScheduledTask persistAllConsumerOffset exception", e);
        }
    }
}, 1000 * 10, this.clientConfig.getPersistConsumerOffsetInterval(), TimeUnit.MILLISECONDS);

以上,消费者在启动的时候,开启了一个定时任务,定时将本地缓存提交到broker。

org.apache.rocketmq.client.consumer.store.RemoteBrokerOffsetStore#persistAll:

// 参数mqs是当前分配的队列
public void persistAll(Set<MessageQueue> mqs) {
  if (null == mqs || mqs.isEmpty())
    return;
  final HashSet<MessageQueue> unusedMQ = new HashSet<MessageQueue>();
  if (!mqs.isEmpty()) {
    // 遍历位移缓存容器
    for (Map.Entry<MessageQueue, AtomicLong> entry : this.offsetTable.entrySet()) {
      MessageQueue mq = entry.getKey();
      AtomicLong offset = entry.getValue();
      if (offset != null) {
        // 位移缓存容器包含在当前分配队列,则进行消费位移提交
        if (mqs.contains(mq)) {
          try {
            // 提交消费位移
            this.updateConsumeOffsetToBroker(mq, offset.get());
          } catch (Exception e) {
            log.error("updateConsumeOffsetToBroker exception, " + mq.toString(), e);
          }
        } else {
          unusedMQ.add(mq);
        }
      }
    }
  }

  // 将未分配的队列从位移缓存中移除
  if (!unusedMQ.isEmpty()) {
    for (MessageQueue mq : unusedMQ) {
      this.offsetTable.remove(mq);
      log.info("remove unused mq, {}, {}", mq, this.groupName);
    }
  }
}

最终会调用以上方法,RocketMQ 会从重平衡那里获取当前消费者已分配的队列,如果位移缓存容器包含在当前分配队列,则进行消费位移提交,否则将从位移缓存容器中移除。

broker 端处理:

org.apache.rocketmq.broker.offset.ConsumerOffsetManager#commitOffset

private void commitOffset(final String clientHost, final String key, final int queueId, final long offset) {
  ConcurrentMap<Integer, Long> map = this.offsetTable.get(key);
  if (null == map) {
    map = new ConcurrentHashMap<Integer, Long>(32);
    map.put(queueId, offset);
    this.offsetTable.put(key, map);
  } else {
    Long storeOffset = map.put(queueId, offset);
    if (storeOffset != null && offset < storeOffset) {
      log.warn("[NOTIFYME]update consumer offset less than store. clientHost={}, key={}, queueId={}, requestOffset={}, storeOffset={}", clientHost, key, queueId, offset, storeOffset);
    }
  }
}

以上,offsetTable 为 broker 端的消费位移缓存容器,它的结构如下:

private ConcurrentMap<String/* topic@group */, ConcurrentMap<Integer, Long>> offsetTable =
  new ConcurrentHashMap<>(512);

它同样是一个 ConcurrentMap,一个线程安全容器,key 为的形式为 “topic@group”,value 也是一个 ConcurrentMap 它的 key 为 queueId,value 为位移,它会以 json 的形式持久化到磁盘 ${ROCKETMQ_HOME}/store/config/consumerOffset.json 文件中,具体格式如下:

{
    "offsetTable": {
        "test-topic@test-group": {
            "0": 88526, 
            "1": 88528, 
            "2": 88532, 
            "3": 88537
        }
    }
}