1. 基本概念

  消费kafka消息的客户端称为consumer,consumer负责订阅kafka的topic,并从该topic上拉取消息。除了consumer本身,kafka还有一个消费组(consumer group)的概念。每个kafka consumer都属于一个消费者组,一条消息只会被一个消费者组内的一个消费者消费,因此一个消费者组内消费者的数量一般不会超过分区的数量。例如,某个主题共有4个分区:P0, P1, P2, P3;有两个消费者组A, B;A中有四个消费者(C0, C1, C2, C3),B中有两个消费者(C4, C5),则A中每个消费者分配到1个分区,B中每个消费者分配到两个分区,消费分配情况如下:

发送拉取请求: 


public synchronized int sendFetches() {
    // Update metrics in case there was an assignment change
    sensors.maybeUpdateAssignment(subscriptions);

    Map<Node, FetchSessionHandler.FetchRequestData> fetchRequestMap = prepareFetchRequests(); // 选取没有in-flight request的leader, 包装成map<node, requests>的形式
    for (Map.Entry<Node, FetchSessionHandler.FetchRequestData> entry : fetchRequestMap.entrySet()) {
        final Node fetchTarget = entry.getKey();
        final FetchSessionHandler.FetchRequestData data = entry.getValue();
        final short maxVersion;
        if (!data.canUseTopicIds()) {
            maxVersion = (short) 12;
        } else {
            maxVersion = ApiKeys.FETCH.latestVersion();
        }
        final FetchRequest.Builder request = FetchRequest.Builder
                .forConsumer(maxVersion, this.maxWaitMs, this.minBytes, data.toSend())
                .isolationLevel(isolationLevel)
                .setMaxBytes(this.maxBytes)
                .metadata(data.metadata())
                .removed(data.toForget())
                .replaced(data.toReplace())
                .rackId(clientRackId);

        if (log.isDebugEnabled()) {
            log.debug("Sending {} {} to broker {}", isolationLevel, data.toString(), fetchTarget);
        }
        // 调用client.send(),发送请求。client.send为异步接口,不做实际的发送动作,底层仍然是调用client.poll()
        RequestFuture<ClientResponse> future = client.send(fetchTarget, request);
        // 注册回调函数
        this.nodesWithPendingFetchRequests.add(entry.getKey().id());
        future.addListener(new RequestFutureListener<ClientResponse>() {
            @Override
            public void onSuccess(ClientResponse resp) {
                synchronized (Fetcher.this) {
                    try {
                        FetchResponse response = (FetchResponse) resp.responseBody();
                        FetchSessionHandler handler = sessionHandler(fetchTarget.id());
                        if (handler == null) {
                            log.error("Unable to find FetchSessionHandler for node {}. Ignoring fetch response.",
                                    fetchTarget.id());
                            return;
                        }
                        if (!handler.handleResponse(response, resp.requestHeader().apiVersion())) {
                            if (response.error() == Errors.FETCH_SESSION_TOPIC_ID_ERROR) {
                                metadata.requestUpdate();
                            }
                            return;
                        }
                        // 将response封装成map<TopicPartition, PartitionData>的格式
                        Map<TopicPartition, FetchResponseData.PartitionData> responseData = response.responseData(handler.sessionTopicNames(), resp.requestHeader().apiVersion());
                        Set<TopicPartition> partitions = new HashSet<>(responseData.keySet());
                        FetchResponseMetricAggregator metricAggregator = new FetchResponseMetricAggregator(sensors, partitions);

                        for (Map.Entry<TopicPartition, FetchResponseData.PartitionData> entry : responseData.entrySet()) {
                            TopicPartition partition = entry.getKey();
                            FetchRequest.PartitionData requestData = data.sessionPartitions().get(partition);
                            if (requestData == null) {
                                String message;
                                if (data.metadata().isFull()) {
                                    message = MessageFormatter.arrayFormat(
                                            "Response for missing full request partition: partition={}; metadata={}",
                                            new Object[]{partition, data.metadata()}).getMessage();
                                } else {
                                    message = MessageFormatter.arrayFormat(
                                            "Response for missing session request partition: partition={}; metadata={}; toSend={}; toForget={}; toReplace={}",
                                            new Object[]{partition, data.metadata(), data.toSend(), data.toForget(), data.toReplace()}).getMessage();
                                }

                                // Received fetch response for missing session partition
                                throw new IllegalStateException(message);
                            } else {
                                long fetchOffset = requestData.fetchOffset;
                                FetchResponseData.PartitionData partitionData = entry.getValue();

                                log.debug("Fetch {} at offset {} for partition {} returned fetch data {}",
                                        isolationLevel, fetchOffset, partition, partitionData);

                                Iterator<? extends RecordBatch> batches = FetchResponse.recordsOrFail(partitionData).batches().iterator();
                                short responseVersion = resp.requestHeader().apiVersion();
                                // 结果添加到completedFetches
                                completedFetches.add(new CompletedFetch(partition, partitionData,
                                        metricAggregator, batches, fetchOffset, responseVersion));
                            }
                        }

                        sensors.fetchLatency.record(resp.requestLatencyMs());
                    } finally {
                        nodesWithPendingFetchRequests.remove(fetchTarget.id());
                    }
                }
            }

            @Override
            public void onFailure(RuntimeException e) {
                synchronized (Fetcher.this) {
                    try {
                        FetchSessionHandler handler = sessionHandler(fetchTarget.id());
                        if (handler != null) {
                            handler.handleError(e);
                        }
                    } finally {
                        nodesWithPendingFetchRequests.remove(fetchTarget.id());
                    }
                }
            }
        });

    }
    return fetchRequestMap.size();
}

View Code

获取拉取响应:


public Fetch<K, V> collectFetch() {
    Fetch<K, V> fetch = Fetch.empty();
    Queue<CompletedFetch> pausedCompletedFetches = new ArrayDeque<>();
    int recordsRemaining = maxPollRecords;

    try {
        while (recordsRemaining > 0) {
            if (nextInLineFetch == null || nextInLineFetch.isConsumed) {
                CompletedFetch records = completedFetches.peek();
                if (records == null) break;

                if (records.notInitialized()) {
                    try {
                        // 将CompletedFetch包装成nextInLineFetch
                        nextInLineFetch = initializeCompletedFetch(records);
                    } catch (Exception e) {
                        // Remove a completedFetch upon a parse with exception if (1) it contains no records, and
                        // (2) there are no fetched records with actual content preceding this exception.
                        // The first condition ensures that the completedFetches is not stuck with the same completedFetch
                        // in cases such as the TopicAuthorizationException, and the second condition ensures that no
                        // potential data loss due to an exception in a following record.
                        FetchResponseData.PartitionData partition = records.partitionData;
                        if (fetch.isEmpty() && FetchResponse.recordsOrFail(partition).sizeInBytes() == 0) {
                            completedFetches.poll();
                        }
                        throw e;
                    }
                } else {
                    nextInLineFetch = records;
                }
                completedFetches.poll();
            } else if (subscriptions.isPaused(nextInLineFetch.partition)) {
                // when the partition is paused we add the records back to the completedFetches queue instead of draining
                // them so that they can be returned on a subsequent poll if the partition is resumed at that time
                log.debug("Skipping fetching records for assigned partition {} because it is paused", nextInLineFetch.partition);
                pausedCompletedFetches.add(nextInLineFetch);
                nextInLineFetch = null;
            } else {
                Fetch<K, V> nextFetch = fetchRecords(nextInLineFetch, recordsRemaining);
                recordsRemaining -= nextFetch.numRecords();
                fetch.add(nextFetch);
            }
        }
    } catch (KafkaException e) {
        if (fetch.isEmpty())
            throw e;
    } finally {
        // add any polled completed fetches for paused partitions back to the completed fetches queue to be
        // re-evaluated in the next poll
        completedFetches.addAll(pausedCompletedFetches);
    }

    return fetch;
}

View Code

KafkaConsumer和Fetcher类的交互过程如下图所示:

 

图2 kafka consumer和Fetcher的交互过程

整个流程为:KafkaConsumer.poll() -> Fetcher.collectFetch() -> 有数据,直接返回/没数据,继续调用 -> Fetcher.sendFetches() -> client.poll()/client.send() -> 异步数据接收并填充到Fetcher.completedFetches。这样下一次调用到collectFetch时就有数据了。

4. 位移提交

  位移提交,就是把当前消费到的消息的offset,反馈给服务端,记录消费的结果。一般情况下,我们都使用kafka的自动位移提交。当然KafkaConsumer也提供了手动提交位移的函数,分别是:

// 异步提交offset
public void commitAsync(final Map<TopicPartition, OffsetAndMetadata> offsets, OffsetCommitCallback callback)

// 同步提交offset
public void commitSync(final Map<TopicPartition, OffsetAndMetadata> offsets, final Duration timeout)

位移提交实际上是调用GroupCoordinator的相关函数,具体细节在GroupCoordinator介绍。

5. 总结

kafka consumer整体使用流程较为简单,需要重点注意以下几点:

(1)kafka consumer不是线程安全的,不能多线程共享一个consumer。因此在编写consumer对应的代码时,需要设计好对应的消费模型,见第二节的讲述。

(2)kafka的consumer支持使用subscribe()订阅主题,同时默认是自动rebalance。但是如果通过assign()指定partition进行消费,则不能进行rebalance。rebalance受到很多kafka配置参数的影响,实际使用时,可能导致频繁的rebalance,因此有些场合也会选择手动执行rebalance,并添加对于partition消费的lag的监控。

(3)一般情况下,一个partition只会被同一个group内的一个消费者消费,但是这个限制也可以打破,比如通过assign将多个消费者指定到了同一个partition。这种消费场景一定要注意消费位移的顺序提交问题,实际实现顺序提交消费位移是非常困难的,因此在实际使用中非常不建议这种方式。在大流量的场景下,最常用的消费模型是多个消费者线程+多个消息处理线程,如下:

图3 大流量常用的消费模型

使用这种消费模型消费者和处理线程解耦,可以横向拓展消息处理能力,不受partition数量的影响。但是这种模型是异步提交消费位移,所以当MessageHandler出错时,需要设计一套额外的机制,实现错误重试。

 

 

原文地址:http://www.cnblogs.com/yuanwebpage/p/16898687.html

1. 本站所有资源来源于用户上传和网络,如有侵权请邮件联系站长! 2. 分享目的仅供大家学习和交流,请务用于商业用途! 3. 如果你也有好源码或者教程,可以到用户中心发布,分享有积分奖励和额外收入! 4. 本站提供的源码、模板、插件等等其他资源,都不包含技术服务请大家谅解! 5. 如有链接无法下载、失效或广告,请联系管理员处理! 6. 本站资源售价只是赞助,收取费用仅维持本站的日常运营所需! 7. 如遇到加密压缩包,默认解压密码为"gltf",如遇到无法解压的请联系管理员! 8. 因为资源和程序源码均为可复制品,所以不支持任何理由的退款兑现,请斟酌后支付下载 声明:如果标题没有注明"已测试"或者"测试可用"等字样的资源源码均未经过站长测试.特别注意没有标注的源码不保证任何可用性