RocketMQ的前提回顧
RocketMQ是一款分布式、隊(duì)列模型的消息中間件，具有以下特點(diǎn)：
- 能夠保證嚴(yán)格的消息順序
- 提供豐富的消息拉取模式
- 高效的訂閱者水平擴(kuò)展能力
- 實(shí)時(shí)的消息訂閱機(jī)制
- 億級(jí)消息堆積能力
為什么使用RocketMQ
- 強(qiáng)調(diào)集群無單點(diǎn)，可擴(kuò)展，任意一點(diǎn)高可用、水平可擴(kuò)展
- 海量消息堆積能力，消息堆積后寫入低延遲
- 支持上萬個(gè)隊(duì)列
- 消息失敗重試機(jī)制
- 消息可查詢
- 開源社區(qū)活躍
- 成熟度已經(jīng)經(jīng)過淘寶雙十一的考驗(yàn)
RocketMQ的發(fā)展變化
RocketMQ開源是使用文件作為持久化工具，阿里內(nèi)部未開源的性能會(huì)更高，使用oceanBase作為持久化工具。
在RocketMQ1.x和2.x使用zookeeper管理集群，3.x開始使用nameserver代替zk，更輕量級(jí)，此外RocketMQ的客戶端擁有兩種的操作方式：DefaultMQPushConsumer和DefaultMQPullConsumer。
DefaultMQPushConsumer的Maven配置
org.apache.rocketmq rocketmq-client 4.3.0 DefaultMQPushConsumer使用示例
- CONSUME_FROM_LAST_OFFSET：第一次啟動(dòng)從隊(duì)列最后位置消費(fèi)，后續(xù)再啟動(dòng)接著上次消費(fèi)的進(jìn)度開始消費(fèi)
- CONSUME_FROM_FIRST_OFFSET：第一次啟動(dòng)從隊(duì)列初始位置消費(fèi)，后續(xù)再啟動(dòng)接著上次消費(fèi)的進(jìn)度開始消費(fèi)
- CONSUME_FROM_TIMESTAMP：第一次啟動(dòng)從指定時(shí)間點(diǎn)位置消費(fèi)，后續(xù)再啟動(dòng)接著上次消費(fèi)的進(jìn)度開始消費(fèi)
以上所說的第一次啟動(dòng)是指從來沒有消費(fèi)過的消費(fèi)者，如果該消費(fèi)者消費(fèi)過，那么會(huì)在broker端記錄該消費(fèi)者的消費(fèi)位置，如果該消費(fèi)者掛了再啟動(dòng)，那么自動(dòng)從上次消費(fèi)的進(jìn)度開始
public class MQPushConsumer { public static void main(String[] args) throws MQClientException { String groupName = "rocketMqGroup1"; // 用于把多個(gè)Consumer組織到一起，提高并發(fā)處理能力 DefaultMQPushConsumer consumer = new DefaultMQPushConsumer(groupName); // 設(shè)置nameServer地址，多個(gè)以;分隔 consumer.setNamesrvAddr("name-serverl-ip:9876;name-server2-ip:9876"); consumer.setConsumeFromWhere(ConsumeFromWhere.CONSUME_FROM_FIRST_OFFSET); consumer.setMessageModel(MessageModel.BROADCASTING); // 訂閱topic，可以對(duì)指定消息進(jìn)行過濾，例如："TopicTest","tagl||tag2||tag3",*或null表示topic所有消息 consumer.subscribe("order-topic", "*"); consumer.registerMessageListener(new MessageListenerConcurrently() { @Override public ConsumeConcurrentlyStatus consumeMessage(List mgs, ConsumeConcurrentlyContext consumeconcurrentlycontext) { System.out.println(Thread.currentThread().getName()+"Receive New Messages:"+mgs); return ConsumeConcurrentlyStatus.CONSUME_SUCCESS; } }); consumer.start(); }} - CLUSTERING：默認(rèn)模式，同一個(gè)ConsumerGroup(groupName相同)每個(gè)consumer只消費(fèi)所訂閱消息的一部分內(nèi)容，同一個(gè)ConsumerGroup里所有的Consumer消息加起來才是所
- 訂閱topic整體，從而達(dá)到負(fù)載均衡的目的
- BROADCASTING：同一個(gè)ConsumerGroup每個(gè)consumer都消費(fèi)到所訂閱topic所有消息，也就是一個(gè)消費(fèi)會(huì)被多次分發(fā)，被多個(gè)consumer消費(fèi)。
ConsumeConcurrentlyStatus.RECONSUME_LATER boker會(huì)根據(jù)設(shè)置的messageDelayLevel發(fā)起重試，默認(rèn)16次。
DefaultMQPushConsumerImpl中各個(gè)對(duì)象的主要功能如下：
RebalancePushImpl：主要負(fù)責(zé)決定，當(dāng)前的consumer應(yīng)該從哪些Queue中消費(fèi)消息；
- 1）PullAPIWrapper：長連接，負(fù)責(zé)從broker處拉取消息，然后利用ConsumeMessageService回調(diào)用戶的Listener執(zhí)行消息消費(fèi)邏輯；
- 2）ConsumeMessageService：實(shí)現(xiàn)所謂的"Push-被動(dòng)"消費(fèi)機(jī)制；從Broker拉取的消息后，封裝成ConsumeRequest提交給ConsumeMessageSerivce，此service負(fù)責(zé)回調(diào)用戶的Listener消費(fèi)消息；
- 3）OffsetStore：維護(hù)當(dāng)前consumer的消費(fèi)記錄（offset）；有兩種實(shí)現(xiàn)，Local和Rmote，Local存儲(chǔ)在本地磁盤上，適用于BROADCASTING廣播消費(fèi)模式；而Remote則將消費(fèi)進(jìn)度存儲(chǔ)在Broker上，適用于CLUSTERING集群消費(fèi)模式；
- 4）MQClientFactory：負(fù)責(zé)管理client（consumer、producer），并提供多中功能接口供各個(gè)Service（Rebalance、PullMessage等）調(diào)用；大部分邏輯均在這個(gè)類中完成；
consumer.registerMessageListener執(zhí)行過程：
/** * Register a callback to execute on message arrival for concurrent consuming. * @param messageListener message handling callback. */ @Override public void registerMessageListener(MessageListenerConcurrently messageListener) { this.messageListener = messageListener; this.defaultMQPushConsumerImpl.registerMessageListener(messageListener); }通過源碼可以看出主要實(shí)現(xiàn)過程在DefaultMQPushConsumerImpl類中consumer.start后調(diào)用DefaultMQPushConsumerImpl的同步start方法
public synchronized void start() throws MQClientException { switch (this.serviceState) { case CREATE_JUST: log.info("the consumer [{}] start beginning. messageModel={}, isUnitMode={}", this.defaultMQPushConsumer.getConsumerGroup(), this.defaultMQPushConsumer.getMessageModel(), this.defaultMQPushConsumer.isUnitMode()); this.serviceState = ServiceState.START_FAILED; this.checkConfig(); this.copySubscription(); if (this.defaultMQPushConsumer.getMessageModel() == MessageModel.CLUSTERING) { this.defaultMQPushConsumer.changeInstanceNameToPID(); } this.mQClientFactory = MQClientManager.getInstance().getAndCreateMQClientInstance(this.defaultMQPushConsumer, this.rpcHook); this.rebalanceImpl.setConsumerGroup(this.defaultMQPushConsumer.getConsumerGroup()); this.rebalanceImpl.setMessageModel(this.defaultMQPushConsumer.getMessageModel()); this.rebalanceImpl.setAllocateMessageQueueStrategy(this.defaultMQPushConsumer.getAllocateMessageQueueStrategy()); this.rebalanceImpl.setmQClientFactory(this.mQClientFactory); this.pullAPIWrapper = new PullAPIWrapper( mQClientFactory, this.defaultMQPushConsumer.getConsumerGroup(), isUnitMode()); this.pullAPIWrapper.registerFilterMessageHook(filterMessageHookList); if (this.defaultMQPushConsumer.getOffsetStore() != null) { this.offsetStore = this.defaultMQPushConsumer.getOffsetStore(); } else { switch (this.defaultMQPushConsumer.getMessageModel()) { case BROADCASTING: this.offsetStore = new LocalFileOffsetStore(this.mQClientFactory, this.defaultMQPushConsumer.getConsumerGroup()); break; case CLUSTERING: this.offsetStore = new RemoteBrokerOffsetStore(this.mQClientFactory, this.defaultMQPushConsumer.getConsumerGroup()); break; default: break; } this.defaultMQPushConsumer.setOffsetStore(this.offsetStore); } this.offsetStore.load(); if (this.getMessageListenerInner() instanceof MessageListenerOrderly) { this.consumeOrderly = true; this.consumeMessageService = new ConsumeMessageOrderlyService(this, (MessageListenerOrderly) this.getMessageListenerInner()); } else if (this.getMessageListenerInner() instanceof MessageListenerConcurrently) { this.consumeOrderly = false; this.consumeMessageService = new ConsumeMessageConcurrentlyService(this, (MessageListenerConcurrently) this.getMessageListenerInner()); } this.consumeMessageService.start(); boolean registerOK = mQClientFactory.registerConsumer(this.defaultMQPushConsumer.getConsumerGroup(), this); if (!registerOK) { this.serviceState = ServiceState.CREATE_JUST; this.consumeMessageService.shutdown(); throw new MQClientException("The consumer group[" + this.defaultMQPushConsumer.getConsumerGroup() + "] has been created before, specify another name please." + FAQUrl.suggestTodo(FAQUrl.GROUP_NAME_DUPLICATE_URL), null); } mQClientFactory.start(); log.info("the consumer [{}] start OK.", this.defaultMQPushConsumer.getConsumerGroup()); this.serviceState = ServiceState.RUNNING; break; case RUNNING: case START_FAILED: case SHUTDOWN_ALREADY: throw new MQClientException("The PushConsumer service state not OK, maybe started once, " + this.serviceState + FAQUrl.suggestTodo(FAQUrl.CLIENT_SERVICE_NOT_OK), null); default: break; } this.updateTopicSubscribeInfoWhenSubscriptionChanged(); this.mQClientFactory.checkClientInBroker(); this.mQClientFactory.sendHeartbeatToAllBrokerWithLock(); this.mQClientFactory.rebalanceImmediately(); }通過mQClientFactory.start();發(fā)我們發(fā)現(xiàn)他調(diào)用
public void start() throws MQClientException { synchronized (this) { switch (this.serviceState) { case CREATE_JUST: this.serviceState = ServiceState.START_FAILED; // If not specified,looking address from name server if (null == this.clientConfig.getNamesrvAddr()) { this.mQClientAPIImpl.fetchNameServerAddr(); } // Start request-response channel this.mQClientAPIImpl.start(); // Start various schedule tasks this.startScheduledTask(); // Start pull service this.pullMessageService.start(); // Start rebalance service this.rebalanceService.start(); // Start push service this.defaultMQProducer.getDefaultMQProducerImpl().start(false); log.info("the client factory [{}] start OK", this.clientId); this.serviceState = ServiceState.RUNNING; break; case RUNNING: break; case SHUTDOWN_ALREADY: break; case START_FAILED: throw new MQClientException("The Factory object[" + this.getClientId() + "] has been created before, and failed.", null); default: break; } } }在這個(gè)方法中有多個(gè)start，我們主要看pullMessageService.start();通過這里我們發(fā)現(xiàn)RocketMQ的Push模式底層其實(shí)也是通過pull實(shí)現(xiàn)的，下面我們來看下pullMessageService處理了哪些邏輯：
private void pullMessage(final PullRequest pullRequest) { final MQConsumerInner consumer = this.mQClientFactory.selectConsumer(pullRequest.getConsumerGroup()); if (consumer != null) { DefaultMQPushConsumerImpl impl = (DefaultMQPushConsumerImpl) consumer; impl.pullMessage(pullRequest); } else { log.warn("No matched consumer for the PullRequest {}, drop it", pullRequest); } }我們發(fā)現(xiàn)其實(shí)他還是通過DefaultMQPushConsumerImpl類的pullMessage方法來進(jìn)行消息的邏輯處理.
pullRequest拉取方式
PullRequest這里說明一下，上面我們已經(jīng)提了一下rocketmq的push模式其實(shí)是通過pull模式封裝實(shí)現(xiàn)的，pullrequest這里是通過長輪詢的方式達(dá)到push效果。
長輪詢方式既有pull的優(yōu)點(diǎn)又有push模式的實(shí)時(shí)性有點(diǎn)。
push方式是server端接收到消息后，主動(dòng)把消息推送給client端，實(shí)時(shí)性高。弊端是server端工作量大，影響性能，其次是client端處理能力不同且client端的狀態(tài)不受server端的控制，如果client端不能及時(shí)處理消息容易導(dǎo)致消息堆積已經(jīng)影響正常業(yè)務(wù)等。
- pull方式是client循環(huán)從server端拉取消息，主動(dòng)權(quán)在client端，自己處理完一個(gè)消息再去拉取下一個(gè)，缺點(diǎn)是循環(huán)的時(shí)間不好設(shè)定，時(shí)間太短容易忙等，浪費(fèi)CPU資源，時(shí)間間隔太長client的處理能力會(huì)下降，有時(shí)候有些消息會(huì)處理不及時(shí)。
長輪詢的方式可以結(jié)合兩者優(yōu)點(diǎn)
- 檢查PullRequest對(duì)象中的ProcessQueue對(duì)象的dropped是否為true（在RebalanceService線程中為topic下的MessageQueue創(chuàng)建拉取消息請(qǐng)求時(shí)要維護(hù)對(duì)應(yīng)的ProcessQueue對(duì)象，若Consumer不再訂閱該topic則會(huì)將該對(duì)象的dropped置為true）；若是則認(rèn)為該請(qǐng)求是已經(jīng)取消的，則直接跳出該方法；
- 更新PullRequest對(duì)象中的ProcessQueue對(duì)象的時(shí)間戳（ProcessQueue.lastPullTimestamp）為當(dāng)前時(shí)間戳；
- 檢查該Consumer是否運(yùn)行中，即DefaultMQPushConsumerImpl.serviceState是否為RUNNING;若不是運(yùn)行狀態(tài)或者是暫停狀態(tài)（DefaultMQPushConsumerImpl.pause=true），則調(diào)用PullMessageService.executePullRequestLater(PullRequest pullRequest, long timeDelay)方法延遲再拉取消息，其中timeDelay=3000；該方法的目的是在3秒之后再次將該P(yáng)ullRequest對(duì)象放入PullMessageService. pullRequestQueue隊(duì)列中；并跳出該方法；
- 進(jìn)行流控。若ProcessQueue對(duì)象的msgCount大于了消費(fèi)端的流控閾值（DefaultMQPushConsumer.pullThresholdForQueue，默認(rèn)值為1000），則調(diào)用PullMessageService.executePullRequestLater方法，在50毫秒之后重新該P(yáng)ullRequest請(qǐng)求放入PullMessageService.pullRequestQueue隊(duì)列中；并跳出該方法；
- 若不是順序消費(fèi)（即DefaultMQPushConsumerImpl.consumeOrderly等于false），則檢查ProcessQueue對(duì)象的msgTreeMap:TreeMap
- 以PullRequest.messageQueue對(duì)象的topic值為參數(shù)從RebalanceImpl.subscriptionInner: ConcurrentHashMap, SubscriptionData>中獲取對(duì)應(yīng)的SubscriptionData對(duì)象，若該對(duì)象為null，考慮到并發(fā)的關(guān)系，調(diào)用executePullRequestLater方法，稍后重試；并跳出該方法；
- 若消息模型為集群模式（RebalanceImpl.messageModel等于CLUSTERING），則以PullRequest對(duì)象的MessageQueue變量值、type =READ_FROM_MEMORY（從內(nèi)存中獲取消費(fèi)進(jìn)度offset值）為參數(shù)調(diào)用DefaultMQPushConsumerImpl. offsetStore對(duì)象（初始化為RemoteBrokerOffsetStore對(duì)象）的readOffset(MessageQueue mq, ReadOffsetType type)方法從本地內(nèi)存中獲取消費(fèi)進(jìn)度offset值。若該offset值大于0 則置臨時(shí)變量commitOffsetEnable等于true否則為false；該offset值作為pullKernelImpl方法中的commitOffset參數(shù)，在Broker端拉取消息之后根據(jù)commitOffsetEnable參數(shù)值決定是否用該offset更新消息進(jìn)度。該readOffset方法的邏輯是：以入?yún)essageQueue對(duì)象從RemoteBrokerOffsetStore.offsetTable:ConcurrentHashMap
- 當(dāng)每次拉取消息之后需要更新訂閱關(guān)系（由DefaultMQPushConsumer. postSubscriptionWhenPull參數(shù)表示，默認(rèn)為false）并且以topic值參數(shù)從RebalanceImpl.subscriptionInner獲取的SubscriptionData對(duì)象的classFilterMode等于false（默認(rèn)為false），則將sysFlag標(biāo)記的第3個(gè)字節(jié)置為1，否則該字節(jié)置為0；
- 該sysFlag標(biāo)記的第1個(gè)字節(jié)置為commitOffsetEnable的值；第2個(gè)字節(jié)（suspend標(biāo)記）置為1；第4個(gè)字節(jié)置為classFilterMode的值；
- 初始化匿名內(nèi)部類PullCallback，實(shí)現(xiàn)了onSucess/onException方法； 該方法只有在異步請(qǐng)求的情況下才會(huì)回調(diào)；
- 調(diào)用底層的拉取消息API接口：
PullAPIWrapper.pullKernelImpl
PullAPIWrapper.pullKernelImpl(MessageQueue mq, String subExpression, long subVersion,long offset, int maxNums, int sysFlag,long commitOffset,long brokerSuspendMaxTimeMillis, long timeoutMillis, CommunicationMode communicationMode, PullCallback pullCallback)方法進(jìn)行消息拉取操作。
將回調(diào)類PullCallback傳入該方法中，當(dāng)采用異步方式拉取消息時(shí)，在收到響應(yīng)之后會(huì)回調(diào)該回調(diào)類的方法。
public void pullMessage(final PullRequest pullRequest) { final ProcessQueue processQueue = pullRequest.getProcessQueue(); if (processQueue.isDropped()) { log.info("the pull request[{}] is dropped.", pullRequest.toString()); return; } pullRequest.getProcessQueue().setLastPullTimestamp(System.currentTimeMillis()); try { this.makeSureStateOK(); } catch (MQClientException e) { log.warn("pullMessage exception, consumer state not ok", e); this.executePullRequestLater(pullRequest, PULL_TIME_DELAY_MILLS_WHEN_EXCEPTION); return; } if (this.isPause()) { log.warn("consumer was paused, execute pull request later. instanceName={}, group={}", this.defaultMQPushConsumer.getInstanceName(), this.defaultMQPushConsumer.getConsumerGroup()); this.executePullRequestLater(pullRequest, PULL_TIME_DELAY_MILLS_WHEN_SUSPEND); return; } long cachedMessageCount = processQueue.getMsgCount().get(); long cachedMessageSizeInMiB = processQueue.getMsgSize().get() / (1024 * 1024); if (cachedMessageCount > this.defaultMQPushConsumer.getPullThresholdForQueue()) { this.executePullRequestLater(pullRequest, PULL_TIME_DELAY_MILLS_WHEN_FLOW_CONTROL); if ((queueFlowControlTimes++ % 1000) == 0) { log.warn( "the cached message count exceeds the threshold {}, so do flow control, minOffset={}, maxOffset={}, count={}, size={} MiB, pullRequest={}, flowControlTimes={}", this.defaultMQPushConsumer.getPullThresholdForQueue(), processQueue.getMsgTreeMap().firstKey(), processQueue.getMsgTreeMap().lastKey(), cachedMessageCount, cachedMessageSizeInMiB, pullRequest, queueFlowControlTimes); } return; } if (cachedMessageSizeInMiB > this.defaultMQPushConsumer.getPullThresholdSizeForQueue()) { this.executePullRequestLater(pullRequest, PULL_TIME_DELAY_MILLS_WHEN_FLOW_CONTROL); if ((queueFlowControlTimes++ % 1000) == 0) { log.warn( "the cached message size exceeds the threshold {} MiB, so do flow control, minOffset={}, maxOffset={}, count={}, size={} MiB, pullRequest={}, flowControlTimes={}", this.defaultMQPushConsumer.getPullThresholdSizeForQueue(), processQueue.getMsgTreeMap().firstKey(), processQueue.getMsgTreeMap().lastKey(), cachedMessageCount, cachedMessageSizeInMiB, pullRequest, queueFlowControlTimes); } return; } if (!this.consumeOrderly) { if (processQueue.getMaxSpan() > this.defaultMQPushConsumer.getConsumeConcurrentlyMaxSpan()) { this.executePullRequestLater(pullRequest, PULL_TIME_DELAY_MILLS_WHEN_FLOW_CONTROL); if ((queueMaxSpanFlowControlTimes++ % 1000) == 0) { log.warn( "the queues messages, span too long, so do flow control, minOffset={}, maxOffset={}, maxSpan={}, pullRequest={}, flowControlTimes={}", processQueue.getMsgTreeMap().firstKey(), processQueue.getMsgTreeMap().lastKey(), processQueue.getMaxSpan(), pullRequest, queueMaxSpanFlowControlTimes); } return; } } else { if (processQueue.isLocked()) { if (!pullRequest.isLockedFirst()) { final long offset = this.rebalanceImpl.computePullFromWhere(pullRequest.getMessageQueue()); boolean brokerBusy = offset < pullRequest.getNextOffset(); log.info("the first time to pull message, so fix offset from broker. pullRequest: {} NewOffset: {} brokerBusy: {}", pullRequest, offset, brokerBusy); if (brokerBusy) { log.info("[NOTIFYME]the first time to pull message, but pull request offset larger than broker consume offset. pullRequest: {} NewOffset: {}", pullRequest, offset); } pullRequest.setLockedFirst(true); pullRequest.setNextOffset(offset); } } else { this.executePullRequestLater(pullRequest, PULL_TIME_DELAY_MILLS_WHEN_EXCEPTION); log.info("pull message later because not locked in broker, {}", pullRequest); return; } } final SubscriptionData subscriptionData = this.rebalanceImpl.getSubscriptionInner().get(pullRequest.getMessageQueue().getTopic()); if (null == subscriptionData) { this.executePullRequestLater(pullRequest, PULL_TIME_DELAY_MILLS_WHEN_EXCEPTION); log.warn("find the consumers subscription failed, {}", pullRequest); return; } final long beginTimestamp = System.currentTimeMillis(); PullCallback pullCallback = new PullCallback() { @Override public void onSuccess(PullResult pullResult) { if (pullResult != null) { pullResult = DefaultMQPushConsumerImpl.this.pullAPIWrapper.processPullResult(pullRequest.getMessageQueue(), pullResult, subscriptionData); switch (pullResult.getPullStatus()) { case FOUND: long prevRequestOffset = pullRequest.getNextOffset(); pullRequest.setNextOffset(pullResult.getNextBeginOffset()); long pullRT = System.currentTimeMillis() - beginTimestamp; DefaultMQPushConsumerImpl.this.getConsumerStatsManager().incPullRT(pullRequest.getConsumerGroup(), pullRequest.getMessageQueue().getTopic(), pullRT); long firstMsgOffset = Long.MAX_VALUE; if (pullResult.getMsgFoundList() == null || pullResult.getMsgFoundList().isEmpty()) { DefaultMQPushConsumerImpl.this.executePullRequestImmediately(pullRequest); } else { firstMsgOffset = pullResult.getMsgFoundList().get(0).getQueueOffset(); DefaultMQPushConsumerImpl.this.getConsumerStatsManager().incPullTPS(pullRequest.getConsumerGroup(), pullRequest.getMessageQueue().getTopic(), pullResult.getMsgFoundList().size()); boolean dispatchToConsume = processQueue.putMessage(pullResult.getMsgFoundList()); DefaultMQPushConsumerImpl.this.consumeMessageService.submitConsumeRequest( pullResult.getMsgFoundList(), processQueue, pullRequest.getMessageQueue(), dispatchToConsume); if (DefaultMQPushConsumerImpl.this.defaultMQPushConsumer.getPullInterval() > 0) { DefaultMQPushConsumerImpl.this.executePullRequestLater(pullRequest, DefaultMQPushConsumerImpl.this.defaultMQPushConsumer.getPullInterval()); } else { DefaultMQPushConsumerImpl.this.executePullRequestImmediately(pullRequest); } } if (pullResult.getNextBeginOffset() < prevRequestOffset || firstMsgOffset < prevRequestOffset) { log.warn( "[BUG] pull message result maybe data wrong, nextBeginOffset: {} firstMsgOffset: {} prevRequestOffset: {}", pullResult.getNextBeginOffset(), firstMsgOffset, prevRequestOffset); } break; case NO_NEW_MSG: pullRequest.setNextOffset(pullResult.getNextBeginOffset()); DefaultMQPushConsumerImpl.this.correctTagsOffset(pullRequest); DefaultMQPushConsumerImpl.this.executePullRequestImmediately(pullRequest); break; case NO_MATCHED_MSG: pullRequest.setNextOffset(pullResult.getNextBeginOffset()); DefaultMQPushConsumerImpl.this.correctTagsOffset(pullRequest); DefaultMQPushConsumerImpl.this.executePullRequestImmediately(pullRequest); break; case OFFSET_ILLEGAL: log.warn("the pull request offset illegal, {} {}", pullRequest.toString(), pullResult.toString()); pullRequest.setNextOffset(pullResult.getNextBeginOffset()); pullRequest.getProcessQueue().setDropped(true); DefaultMQPushConsumerImpl.this.executeTaskLater(new Runnable() { @Override public void run() { try { DefaultMQPushConsumerImpl.this.offsetStore.updateOffset(pullRequest.getMessageQueue(), pullRequest.getNextOffset(), false); DefaultMQPushConsumerImpl.this.offsetStore.persist(pullRequest.getMessageQueue()); DefaultMQPushConsumerImpl.this.rebalanceImpl.removeProcessQueue(pullRequest.getMessageQueue()); log.warn("fix the pull request offset, {}", pullRequest); } catch (Throwable e) { log.error("executeTaskLater Exception", e); } } }, 10000); break; default: break; } } } @Override public void onException(Throwable e) { if (!pullRequest.getMessageQueue().getTopic().startsWith(MixAll.RETRY_GROUP_TOPIC_PREFIX)) { log.warn("execute the pull request exception", e); } DefaultMQPushConsumerImpl.this.executePullRequestLater(pullRequest, PULL_TIME_DELAY_MILLS_WHEN_EXCEPTION); } }; boolean commitOffsetEnable = false; long commitOffsetValue = 0L; if (MessageModel.CLUSTERING == this.defaultMQPushConsumer.getMessageModel()) { commitOffsetValue = this.offsetStore.readOffset(pullRequest.getMessageQueue(), ReadOffsetType.READ_FROM_MEMORY); if (commitOffsetValue > 0) { commitOffsetEnable = true; } } String subExpression = null; boolean classFilter = false; SubscriptionData sd = this.rebalanceImpl.getSubscriptionInner().get(pullRequest.getMessageQueue().getTopic()); if (sd != null) { if (this.defaultMQPushConsumer.isPostSubscriptionWhenPull() && !sd.isClassFilterMode()) { subExpression = sd.getSubString(); } classFilter = sd.isClassFilterMode(); } int sysFlag = PullSysFlag.buildSysFlag( commitOffsetEnable, // commitOffset true, // suspend subExpression != null, // subscription classFilter // class filter ); try { // 下面我們看繼續(xù)跟進(jìn)這個(gè)方法，這個(gè)方法已經(jīng)就是客戶端如何拉取消息 this.pullAPIWrapper.pullKernelImpl( pullRequest.getMessageQueue(), subExpression, subscriptionData.getExpressionType(), subscriptionData.getSubVersion(), pullRequest.getNextOffset(), this.defaultMQPushConsumer.getPullBatchSize(), sysFlag, commitOffsetValue, BROKER_SUSPEND_MAX_TIME_MILLIS, CONSUMER_TIMEOUT_MILLIS_WHEN_SUSPEND, // 消息的通信方式為異步 CommunicationMode.ASYNC, pullCallback ); } catch (Exception e) { log.error("pullKernelImpl exception", e); this.executePullRequestLater(pullRequest, PULL_TIME_DELAY_MILLS_WHEN_EXCEPTION); } }發(fā)送遠(yuǎn)程請(qǐng)求拉取消息
在MQClientAPIImpl.pullMessage方法中，根據(jù)入?yún)ommunicationMode的值分為異步拉取和同步拉取方式兩種。
無論是異步方式拉取還是同步方式拉取，在發(fā)送拉取請(qǐng)求之前都會(huì)構(gòu)造一個(gè)ResponseFuture對(duì)象，以請(qǐng)求消息的序列號(hào)為key值，存入NettyRemotingAbstract.responseTable:ConcurrentHashMap, ResponseFuture>變量中，對(duì)該變量有幾種情況會(huì)處理：
- 發(fā)送失敗后直接刪掉responseTable變量中的相應(yīng)記錄；
- 收到響應(yīng)消息之后，會(huì)以響應(yīng)消息中的序列號(hào)（由服務(wù)端根據(jù)請(qǐng)求消息的序列號(hào)原樣返回）從responseTable中查找ResponseFuture對(duì)象，并設(shè)置該對(duì)象的responseCommand變量。若是同步發(fā)送會(huì)喚醒等待響應(yīng)的ResponseFuture.waitResponse方法；若是異步發(fā)送會(huì)調(diào)用ResponseFuture.executeInvokeCallback()方法完成回調(diào)邏輯處理；
- 在NettyRemotingClient.start()啟動(dòng)時(shí)，也會(huì)初始化定時(shí)任務(wù)，該定時(shí)任務(wù)每隔1秒定期掃描responseTable列表，遍歷該列表中的ResponseFuture對(duì)象，檢查等待響應(yīng)是否超時(shí)，若超時(shí)，則調(diào)用ResponseFuture. executeInvokeCallback()方法，并將該對(duì)象從responseTable列表中刪除；
public PullResult pullMessage( final String addr, final PullMessageRequestHeader requestHeader, final long timeoutMillis, final CommunicationMode communicationMode, final PullCallback pullCallback ) throws RemotingException, MQBrokerException, InterruptedException { RemotingCommand request = RemotingCommand.createRequestCommand(RequestCode.PULL_MESSAGE, requestHeader); switch (communicationMode) { case ONEWAY: assert false; return null; case ASYNC: this.pullMessageAsync(addr, request, timeoutMillis, pullCallback); return null; case SYNC: return this.pullMessageSync(addr, request, timeoutMillis); default: assert false; break; } return null; }同步拉取
對(duì)于同步發(fā)送方式，調(diào)用MQClientAPIImpl.pullMessageSync(String addr, RemotingCommand request, long timeoutMillis)方法,大致步驟如下：
- 調(diào)用RemotingClient.invokeSync(String addr, RemotingCommand request, long timeoutMillis)方法：
- 獲取Broker地址的Channel信息。根據(jù)broker地址從RemotingClient.channelTables:ConcurrentHashMap, ChannelWrapper>變量中獲取ChannelWrapper對(duì)象并返回該對(duì)象的Channel變量；若沒有ChannelWrapper對(duì)象則與broker地址建立新的連接并將連接信息存入channelTables變量中，便于下次使用；
- 若NettyRemotingClient.rpcHook:RPCHook變量不為空（該變量在應(yīng)用層初始化DefaultMQPushConsumer或者DefaultMQPullConsumer對(duì)象傳入該值），則調(diào)用RPCHook.doBeforeRequest(String remoteAddr, RemotingCommand request)方法；
- 調(diào)用NettyRemotingAbstract.invokeSyncImpl(Channel channel, RemotingCommand request, long timeoutMillis)方法，該方法的邏輯如下：
- A）使用請(qǐng)求的序列號(hào)（opaue）、超時(shí)時(shí)間初始化ResponseFuture對(duì)象；并將該ResponseFuture對(duì)象存入NettyRemotingAbstract.responseTable: ConcurrentHashMap變量中；
- B）調(diào)用Channel.writeAndFlush(Object msg)方法將請(qǐng)求對(duì)象RemotingCommand發(fā)送給Broker；然后調(diào)用addListener(GenericFutureListener> listener)方法添加內(nèi)部匿名類：該內(nèi)部匿名類實(shí)現(xiàn)了ChannelFutureListener接口的operationComplete方法，在發(fā)送完成之后回調(diào)該監(jiān)聽類的operationComplete方法，在該方法中，首先調(diào)用ChannelFuture. isSuccess()方法檢查是否發(fā)送成功，若成功則置ResponseFuture對(duì)象的sendRequestOK等于true并退出此回調(diào)方法等待響應(yīng)結(jié)果；若不成功則置ResponseFuture對(duì)象的sendRequestOK等于false，然后從NettyRemotingAbstract.responseTable中刪除此請(qǐng)求序列號(hào)（opaue）的記錄，置ResponseFuture對(duì)象的responseCommand等于null，并喚醒ResponseFuture.waitResponse(long timeoutMillis)方法的等待；
- C）調(diào)用ResponseFuture.waitResponse(long timeoutMillis)方法等待響應(yīng)結(jié)果；在發(fā)送失敗或者收到響應(yīng)消息（詳見5.10.3小節(jié)）或者超時(shí)的情況下會(huì)喚醒該方法返回ResponseFuture.responseCommand變量值；
- D）若上一步返回的responseCommand值為null，則拋出異常:若ResponseFuture.sendRequestOK為true，則拋出RemotingTimeoutException異常，否則拋出RemotingSendRequestException異常；
- E）若上一步返回的responseCommand值不為null，則返回responseCommand變量值；
- 若NettyRemotingClient.rpcHook: RPCHook變量不為空，則調(diào)用RPCHook.doAfterResponse(String remoteAddr, RemotingCommand request)方法；
- 以上一步的返回值RemotingCommand對(duì)象為參數(shù)調(diào)用MQClientAPIImpl. processPullResponse (RemotingCommand response)方法將返回對(duì)象解析并封裝成PullResultExt對(duì)象然后返回給調(diào)用者，響應(yīng)消息的結(jié)果狀態(tài)轉(zhuǎn)換如下：
- 若RemotingCommand對(duì)象的Code等于SUCCESS，則PullResultExt.pullStatus=FOUND；
- 若RemotingCommand對(duì)象的Code等于PULL_NOT_FOUND，則PullResultExt.pullStatus= NO_NEW_MSG；
- 若RemotingCommand對(duì)象的Code等于PULL_RETRY_IMMEDIATELY，則PullResultExt.pullStatus= NO_MATCHED_MSG；
- 若RemotingCommand對(duì)象的Code等于PULL_OFFSET_MOVED，則PullResultExt.pullStatus= OFFSET_ILLEGAL；
@Override public RemotingCommand invokeSync(String addr, final RemotingCommand request, long timeoutMillis) throws InterruptedException, RemotingConnectException, RemotingSendRequestException, RemotingTimeoutException { long beginStartTime = System.currentTimeMillis(); final Channel channel = this.getAndCreateChannel(addr); if (channel != null && channel.isActive()) { try { if (this.rpcHook != null) { this.rpcHook.doBeforeRequest(addr, request); } long costTime = System.currentTimeMillis() - beginStartTime; if (timeoutMillis < costTime) { throw new RemotingTimeoutException("invokeSync call timeout"); } RemotingCommand response = this.invokeSyncImpl(channel, request, timeoutMillis - costTime); if (this.rpcHook != null) { this.rpcHook.doAfterResponse(RemotingHelper.parseChannelRemoteAddr(channel), request, response); } return response; } catch (RemotingSendRequestException e) { log.warn("invokeSync: send request exception, so close the channel[{}]", addr); this.closeChannel(addr, channel); throw e; } catch (RemotingTimeoutException e) { if (nettyClientConfig.isClientCloseSocketIfTimeout()) { this.closeChannel(addr, channel); log.warn("invokeSync: close socket because of timeout, {}ms, {}", timeoutMillis, addr); } log.warn("invokeSync: wait response timeout exception, the channel[{}]", addr); throw e; } } else { this.closeChannel(addr, channel); throw new RemotingConnectException(addr); } }getMQClientAPIImpl().pullMessage最終通過channel寫入并刷新隊(duì)列中。然后在消息服務(wù)端大體的處理邏輯是服務(wù)端收到新消息請(qǐng)求后，如果隊(duì)列中沒有消息不急于返回，通過一個(gè)循環(huán)狀態(tài)，每次waitForRunning一段時(shí)間默認(rèn)5秒，然后再check，如果broker一直沒有新新消息，第三次check的時(shí)間等到時(shí)間超過SuspendMaxTimeMills就返回空，如果在等待過程中收到了新消息直接調(diào)用notifyMessageArriving函數(shù)返回請(qǐng)求結(jié)果。“長輪詢”的核心是，Broker端HOLD住客戶端過來的請(qǐng)求一小段時(shí)間，在這個(gè)時(shí)間內(nèi)有新消息到達(dá)，就利用現(xiàn)有的連接立刻返回消息給 Consumer 。長輪詢的主動(dòng)權(quán)掌握在consumer中，即使broker有大量的消息堆積也不會(huì)主動(dòng)推送給consumer。
