Kafka Producer send原理及重试机制浅析(retries/acks如何被使用的)

我们在使用kafka时会在producer端定义这么两个变量的值：retries 和 acks

ack已经介绍过了

retries是kafka producer针对可重试异常自动发起重试的次数

我们在使用kafkaTemplate时发现：

Spring 提供了一个producerListener接口，在发送成功/失败时会调用这个接口的onSend/onError方法，我们只需要实现这个接口就可以在发送成功/失败的时候处理自己的业务逻辑。

		producer.send(producerRecord, new Callback() {

			@Override
			public void onCompletion(RecordMetadata metadata, Exception exception) {
				try {
					if (exception == null) {
						future.set(new SendResult<>(producerRecord, metadata));
						if (KafkaTemplate.this.producerListener != null) {
							KafkaTemplate.this.producerListener.onSuccess(producerRecord, metadata);
						}
						if (KafkaTemplate.this.logger.isTraceEnabled()) {
							KafkaTemplate.this.logger.trace("Sent ok: " + producerRecord + ", metadata: " + metadata);
						}
					}
					else {
						future.setException(new KafkaProducerException(producerRecord, "Failed to send", exception));
						if (KafkaTemplate.this.producerListener != null) {
							KafkaTemplate.this.producerListener.onError(producerRecord, exception);
						}

但是问题来了，我们配的这个ack 和 retries干啥了呢
相关代码：org.apache.kafka.clients.producer

首先介绍一下kafkaProducer的发送流程

1、构建kafkaProducer 时会启动一个新线程Sender,默认情况下只有一个KafkaProducer，
2、将要发送数据按照topic和partition放到队列Deque中，实例化KafkaProducer会实例RecordAccumulator，RecordAccumulator维护了一个Map，key为不同的主题和partition，value为数据队列

private final ConcurrentMap<TopicPartition, Deque<ProducerBatch>> batches;

3、Sender会轮寻这个队列 进行发送
4、如果发送失败，判断是否属于可重试异常，如果可以重试，则放入队列等待再次轮寻，如果不可以重试则抛出异常，由业务方catch处理

KafkaProducer构造函数：

private final Sender sender;
private final Thread ioThread;
private KafkaProducer(ProducerConfig config, Serializer<K> keySerializer, Serializer<V> valueSerializer) {
//省略无关参数
//我们配置的重试次数
int retries = configureRetries(config, transactionManager != null, log);
//acks
short acks = configureAcks(config, transactionManager != null, log);
this.sender = new Sender(logContext,
                    client,
                    this.metadata,
                    this.accumulator,
                    maxInflightRequests == 1,
                    config.getInt(ProducerConfig.MAX_REQUEST_SIZE_CONFIG),
                    acks,
                    retries,
                    metricsRegistry.senderMetrics,
                    Time.SYSTEM,
                    this.requestTimeoutMs,
                    config.getLong(ProducerConfig.RETRY_BACKOFF_MS_CONFIG),
                    this.transactionManager,
                    apiVersions);
            String ioThreadName = NETWORK_THREAD_PREFIX + " | " + clientId;
            //将sender注册进该ioThread
            this.ioThread = new KafkaThread(ioThreadName, this.sender, true);
            //实际启动的是sender线程
            this.ioThread.start();
}

接下来看KafkaProducer的send逻辑

这里介绍一下RecordAccumulator
RecordAccumulator是保存需要发送的消息或者重试消息的核心。发送消息之前先把消息存放在这里，异步线程KafkaThread启动后从这里取消息然后发送到broker。当发送出错且允许重试时，又会把这些需要重试的消息保存到这里再进行重试。

回到Sender，观察run方法

void run(long now) {
        //...省略前面代码
        //发送核心逻辑
        long pollTimeout = sendProducerData(now);
        //发送结果回调
        client.poll(pollTimeout, now);
    }
private long sendProducerData(long now) {
		// get the list of partitions with data ready to send
        RecordAccumulator.ReadyCheckResult result = this.accumulator.ready(cluster, now);
		//从之前的Deque队列中，根据topic partition取数据
		Map<Integer, List<ProducerBatch>> batches = this.accumulator.drain(cluster, result.readyNodes,
                this.maxRequestSize, now);
        
        //发送核心逻辑
        sendProduceRequests(batches, now);
        return pollTimeout;
    }
   
//遍历发送
private void sendProduceRequests(Map<Integer, List<ProducerBatch>> collated, long now) {
        for (Map.Entry<Integer, List<ProducerBatch>> entry : collated.entrySet())
            sendProduceRequest(now, entry.getKey(), acks, requestTimeout, entry.getValue());
    }

构建了一个回调函数，请求

进入回调逻辑

/**
     * Handle a produce response
     */
    private void handleProduceResponse(ClientResponse response, Map<TopicPartition, ProducerBatch> batches, long now) {
        RequestHeader requestHeader = response.requestHeader();
        int correlationId = requestHeader.correlationId();
        if (response.wasDisconnected()) {
            log.trace("Cancelled request with header {} due to node {} being disconnected",
                    requestHeader, response.destination());
            for (ProducerBatch batch : batches.values())
                completeBatch(batch, new ProduceResponse.PartitionResponse(Errors.NETWORK_EXCEPTION), correlationId, now);
        } else if (response.versionMismatch() != null) {
            log.warn("Cancelled request {} due to a version mismatch with node {}",
                    response, response.destination(), response.versionMismatch());
            for (ProducerBatch batch : batches.values())
                completeBatch(batch, new ProduceResponse.PartitionResponse(Errors.UNSUPPORTED_VERSION), correlationId, now);
        } else {
            log.trace("Received produce response from node {} with correlation id {}", response.destination(), correlationId);
            // if we have a response, parse it
            if (response.hasResponse()) {
                ProduceResponse produceResponse = (ProduceResponse) response.responseBody();
                //遍历所有response
                for (Map.Entry<TopicPartition, ProduceResponse.PartitionResponse> entry : produceResponse.responses().entrySet()) {
                    TopicPartition tp = entry.getKey();
                    ProduceResponse.PartitionResponse partResp = entry.getValue();
                    ProducerBatch batch = batches.get(tp);
                    //处理response
                    completeBatch(batch, partResp, correlationId, now);
                }
                this.sensors.recordLatency(response.destination(), response.requestLatencyMs());
            } else {
                // this is the acks = 0 case, just complete all requests
                for (ProducerBatch batch : batches.values()) {
                    completeBatch(batch, new ProduceResponse.PartitionResponse(Errors.NONE), correlationId, now);
                }
            }
        }
    }

进入competeBatch方法

如果不存在异常 则返回成功

如果存在异常，判断是否可以重试
可以重试 则将发送数据重新放入队列等待轮寻

private void reenqueueBatch(ProducerBatch batch, long currentTimeMs) {
        this.accumulator.reenqueue(batch, currentTimeMs);
        this.sensors.recordRetries(batch.topicPartition.topic(), batch.recordCount);
    }
 
 public void reenqueue(ProducerBatch batch, long now) {
        batch.reenqueued(now);
        Deque<ProducerBatch> deque = getOrCreateDeque(batch.topicPartition);
        synchronized (deque) {
            if (transactionManager != null)
                insertInSequenceOrder(deque, batch);
            else
                deque.addFirst(batch);
        }
    }

如果不可重试 则返回错误信息

canRetry方法中，用到了retries

private boolean canRetry(ProducerBatch batch, ProduceResponse.PartitionResponse response) {
        //重试次数小于配置 且 （属于可重试异常或者transactionManager判断可重试）
        return batch.attempts() < this.retries &&
                ((response.error.exception() instanceof RetriableException) ||
                        (transactionManager != null && transactionManager.canRetry(response, batch)));
    }

client.send逻辑中用到了acks

附：打断点跟踪send流程

断点打在Sender类的sendProduceRequest 因为发送一定会走到这个方法

构建了请求参数 及回调函数

调用回调函数的地方：

response就是前文定义的

 private void completeResponses(List<ClientResponse> responses) {
        for (ClientResponse response : responses) {
            try {
                response.onComplete();
            } catch (Exception e) {
                log.error("Uncaught error in request completion:", e);
            }
        }
    }

断点打在回调函数上，定义条件 topic = 自己测试用的

附：KafkaException

来源网上

本机跑的RetriableException的子类
class org.apache.kafka.common.errors.KafkaStorageException
class org.apache.kafka.common.errors.LeaderNotAvailableException
class org.apache.kafka.common.errors.NotCoordinatorForGroupException
class org.apache.kafka.common.errors.UnknownTopicOrPartitionException
class org.apache.kafka.common.errors.DisconnectException
class org.apache.kafka.common.errors.NotEnoughReplicasException
class org.apache.kafka.common.errors.GroupLoadInProgressException
class org.apache.kafka.common.errors.NotCoordinatorException
class org.apache.kafka.common.errors.NotLeaderForPartitionException
class org.apache.kafka.common.errors.NetworkException
class org.apache.kafka.common.errors.CorruptRecordException
class org.apache.kafka.common.errors.CoordinatorLoadInProgressException
class org.apache.kafka.common.errors.TimeoutException
class org.apache.kafka.common.errors.CoordinatorNotAvailableException
class org.apache.kafka.common.errors.NotControllerException
class org.apache.kafka.common.errors.NotEnoughReplicasAfterAppendException
class org.apache.kafka.common.errors.GroupCoordinatorNotAvailableException

send轮寻待发送队列：


在这里插入代码片