ConsumeQueue的作用

上一篇文章讲到Broker在收到消息后，通过MessageStore将消息存储到commitLog中，但是consumer在消费消息的时候是按照topic+queue的维度来拉取消息的。为了方便读取，MessageStore将CommitLog中消息的offset按照topic+queueId划分后，存储到不同的文件中，这就是ConsumeQueue

文件组织方式

回顾一下数据结构图中ConsumeQueue相关的部分。

ConsumeQueue存储结构

ReputMessageService

这个Service是一个单线程的任务，一直循环的调用doReput()方法：

        private boolean isCommitLogAvailable() {
            return this.reputFromOffset < DefaultMessageStore.this.commitLog.getMaxOffset();
        }

        private void doReput() {
            //1、判断commitLog的maxOffset是否比上次读取的offset大
            for (boolean doNext = true; this.isCommitLogAvailable() && doNext; ) {
                if (DefaultMessageStore.this.getMessageStoreConfig().isDuplicationEnable()
                    && this.reputFromOffset >= DefaultMessageStore.this.getConfirmOffset()) {
                    break;
                }
                //2、从上次的结束offset开始读取commitLog文件中的消息
                SelectMappedBufferResult result = DefaultMessageStore.this.commitLog.getData(reputFromOffset);
                if (result != null) {
                    try {
                        this.reputFromOffset = result.getStartOffset();

                        for (int readSize = 0; readSize < result.getSize() && doNext; ) {
                            //3、检查message数据完整性并封装成DispatchRequest
                            DispatchRequest dispatchRequest =
                                DefaultMessageStore.this.commitLog.checkMessageAndReturnSize(result.getByteBuffer(), false, false);
                            int size = dispatchRequest.getMsgSize();

                            if (dispatchRequest.isSuccess()) {
                                if (size > 0) {
                                    //4、分发消息到CommitLogDispatcher，1)构建索引; 2)更新consumeQueue
                                    DefaultMessageStore.this.doDispatch(dispatchRequest);
                                    //5、分发消息到MessageArrivingListener,唤醒等待的PullReqeust接收消息,Only Master?
                                    if (BrokerRole.SLAVE != DefaultMessageStore.this.getMessageStoreConfig().getBrokerRole()
                                        && DefaultMessageStore.this.brokerConfig.isLongPollingEnable()) {
                                        DefaultMessageStore.this.messageArrivingListener.arriving(dispatchRequest.getTopic(),
                                            dispatchRequest.getQueueId(), dispatchRequest.getConsumeQueueOffset() + 1,
                                            dispatchRequest.getTagsCode(), dispatchRequest.getStoreTimestamp(),
                                            dispatchRequest.getBitMap(), dispatchRequest.getPropertiesMap());
                                    }
                                    //5、更新offset
                                    this.reputFromOffset += size;
                                    readSize += size;
                                    if (DefaultMessageStore.this.getMessageStoreConfig().getBrokerRole() == BrokerRole.SLAVE) {
                                        DefaultMessageStore.this.storeStatsService
                                            .getSinglePutMessageTopicTimesTotal(dispatchRequest.getTopic()).incrementAndGet();
                                        DefaultMessageStore.this.storeStatsService
                                            .getSinglePutMessageTopicSizeTotal(dispatchRequest.getTopic())
                                            .addAndGet(dispatchRequest.getMsgSize());
                                    }
                                } else if (size == 0) {
                                    //6、如果读到文件结尾，则切换到新文件
                                    this.reputFromOffset = DefaultMessageStore.this.commitLog.rollNextFile(this.reputFromOffset);
                                    readSize = result.getSize();
                                }
                            } else if (!dispatchRequest.isSuccess()) {
                                //7、解析消息出错，跳过。commitLog文件中消息数据损坏的情况下才会进来
                                if (size > 0) {
                                    log.error("[BUG]read total count not equals msg total size. reputFromOffset={}", reputFromOffset);
                                    this.reputFromOffset += size;
                                } else {
                                    doNext = false;
                                    if (DefaultMessageStore.this.brokerConfig.getBrokerId() == MixAll.MASTER_ID) {
                                        log.error("[BUG]the master dispatch message to consume queue error, COMMITLOG OFFSET: {}",
                                            this.reputFromOffset);

                                        this.reputFromOffset += result.getSize() - readSize;
                                    }
                                }
                            }
                        }
                    } finally {
                        //8、release对MappedFile的引用
                        result.release();
                    }
                } else {
                    doNext = false;
                }
            }
        }
    /**
     * 消息分发
     */
    public void doDispatch(DispatchRequest req) {
        for (CommitLogDispatcher dispatcher : this.dispatcherList) {
            dispatcher.dispatch(req);
        }
    }

• 第1步，每次处理完读取消息后，都将当前已经处理的最大offset记录下来，下次处理从这个offset开始读取消息
• 第2步，从commitLog文件中读取消息详情
• 第4步，分发读取到的消息，MessageStore在初始化的时候会往dispatcherList中添加两个Dispatcher.

this.dispatcherList = new LinkedList<>();
//consumeQueue构建Dispatcher
this.dispatcherList.addLast(new CommitLogDispatcherBuildConsumeQueue());
//索引更新Dispatcher
this.dispatcherList.addLast(new CommitLogDispatcherBuildIndex());

具体Dispatcher的处理逻辑，我们下面详细说

• 第8步，在通过commitLog读取消息时，不会把消息数据复制到堆内存中，只是返回文件映射的byteBuffer，所以MappedFile记录了有多少个引用，在数据使用完后需要释放。

Dispatcher构建ConsumeQueue

CommitLogDispatcherBuildConsumeQueue实现比较简单，直接调用的MessageStore的接口

class CommitLogDispatcherBuildConsumeQueue implements CommitLogDispatcher {

        @Override
        public void dispatch(DispatchRequest request) {
            final int tranType = MessageSysFlag.getTransactionValue(request.getSysFlag());
            switch (tranType) {
                /** 对于非事务消息和commit事务消息 */
                case MessageSysFlag.TRANSACTION_NOT_TYPE:
                case MessageSysFlag.TRANSACTION_COMMIT_TYPE:
                    DefaultMessageStore.this.putMessagePositionInfo(request);
                    break;
                case MessageSysFlag.TRANSACTION_PREPARED_TYPE:
                case MessageSysFlag.TRANSACTION_ROLLBACK_TYPE:
                    break;
            }
        }
    }

MessageStore中的实现：

    public void putMessagePositionInfo(DispatchRequest dispatchRequest) {
        //找到对应的ComsumeQueue文件
        ConsumeQueue cq = this.findConsumeQueue(dispatchRequest.getTopic(), dispatchRequest.getQueueId());
        cq.putMessagePositionInfoWrapper(dispatchRequest);
    }

前面已经讲过consumeQueue的数据存储结构，每个topic+queueId对应一个ConsumeQueue，每个ConsumeQueue包含一系列MappedFile。所以，这里第一步就是获取对应的ConsumeQueue，如果不存在的话就会新建一个。后面就是调用CQ的put方法：

public void putMessagePositionInfoWrapper(DispatchRequest request) {
        //1、写入重试次数，最多30次
        final int maxRetries = 30; 
        //2、判断CQ是否是可写的
        boolean canWrite = this.defaultMessageStore.getRunningFlags().isCQWriteable();
        for (int i = 0; i < maxRetries && canWrite; i++) {
            long tagsCode = request.getTagsCode();
            if (isExtWriteEnable()) {
                //3、如果需要写ext文件，则将消息的tagscode写入
                ConsumeQueueExt.CqExtUnit cqExtUnit = new ConsumeQueueExt.CqExtUnit();
                cqExtUnit.setFilterBitMap(request.getBitMap());
                cqExtUnit.setMsgStoreTime(request.getStoreTimestamp());
                cqExtUnit.setTagsCode(request.getTagsCode());

                long extAddr = this.consumeQueueExt.put(cqExtUnit);
                if (isExtAddr(extAddr)) {
                    tagsCode = extAddr;
                } else {
                    log.warn("Save consume queue extend fail, So just save tagsCode! {}, topic:{}, queueId:{}, offset:{}", cqExtUnit,
                        topic, queueId, request.getCommitLogOffset());
                }
            }
            //4、写入文件
            boolean result = this.putMessagePositionInfo(request.getCommitLogOffset(),
                request.getMsgSize(), tagsCode, request.getConsumeQueueOffset());
            if (result) {
                //5、记录check point
                this.defaultMessageStore.getStoreCheckpoint().setLogicsMsgTimestamp(request.getStoreTimestamp());
                return;
            } else {
                ...
                ...
            }
        }

       ...
        this.defaultMessageStore.getRunningFlags().makeLogicsQueueError();
    }

• 第3步，将tagcode和bitMap记录进CQExt文件中，这个是一个过滤的扩展功能，采用的bloom过滤器先记录消息的bitMap，这样consumer来读取消息时先通过bloom过滤器判断是否有符合过滤条件的消息
• 第4步，将消息offset写入CQ文件中，这边代码如下：

private boolean putMessagePositionInfo(final long offset, final int size, final long tagsCode,
        final long cqOffset) {

        if (offset <= this.maxPhysicOffset) {
            return true;
        }
        //一个CQUnit的大小是固定的20字节
        this.byteBufferIndex.flip();
        this.byteBufferIndex.limit(CQ_STORE_UNIT_SIZE);
        this.byteBufferIndex.putLong(offset);
        this.byteBufferIndex.putInt(size);
        this.byteBufferIndex.putLong(tagsCode);

        final long expectLogicOffset = cqOffset * CQ_STORE_UNIT_SIZE;
        //获取最后一个MappedFile
        MappedFile mappedFile = this.mappedFileQueue.getLastMappedFile(expectLogicOffset);
        if (mappedFile != null) {
            //对新创建的文件，写将所有CQUnit初始化0值
            if (mappedFile.isFirstCreateInQueue() && cqOffset != 0 && mappedFile.getWrotePosition() == 0) {
                this.minLogicOffset = expectLogicOffset;
                this.mappedFileQueue.setFlushedWhere(expectLogicOffset);
                this.mappedFileQueue.setCommittedWhere(expectLogicOffset);
                this.fillPreBlank(mappedFile, expectLogicOffset);
                log.info("fill pre blank space " + mappedFile.getFileName() + " " + expectLogicOffset + " "
                    + mappedFile.getWrotePosition());
            }

            if (cqOffset != 0) {
                long currentLogicOffset = mappedFile.getWrotePosition() + mappedFile.getFileFromOffset();

                if (expectLogicOffset < currentLogicOffset) {
                    log.warn("Build  consume queue repeatedly, expectLogicOffset: {} currentLogicOffset: {} Topic: {} QID: {} Diff: {}",
                        expectLogicOffset, currentLogicOffset, this.topic, this.queueId, expectLogicOffset - currentLogicOffset);
                    return true;
                }

                if (expectLogicOffset != currentLogicOffset) {
                    LOG_ERROR.warn(
                        "[BUG]logic queue order maybe wrong, expectLogicOffset: {} currentLogicOffset: {} Topic: {} QID: {} Diff: {}",
                        expectLogicOffset,
                        currentLogicOffset,
                        this.topic,
                        this.queueId,
                        expectLogicOffset - currentLogicOffset
                    );
                }
            }
            this.maxPhysicOffset = offset;
            //CQUnit写入文件中
            return mappedFile.appendMessage(this.byteBufferIndex.array());
        }
        return false;
    }

写文件的逻辑和写CommitLog的逻辑是一样的，首先封装一个CQUnit，这里面offset占8个字节，消息size占用4个字节，tagcode占用8个字节。然后找最后一个MappedFile，对于新建的文件，会有一个预热的动作，写把所有CQUnit初始化成0值。最后将Unit写入到文件中。

总结

ConsumeQueue文件数据生成的整个步骤就讲到这里了。Consumer来读取文件的时候，只要指定要读的topic和queueId，以及开始offset。因为每个CQUnit的大小是固定的，所以很容易就可以在文件中定位到。找到开始的位置后，只需要连续读取后面指定数量的Unit，然后根据Unit中存的CommitLog的offset就可以到CommitLog中读取消息详情了。