一、环境准备
flink 1.14写入Kafka,首先在pom.xml文件中导入相关依赖
java
<properties>
<project.build.sourceEncoding>UTF-8</project.build.sourceEncoding>
<flink.version>1.14.6</flink.version>
<spark.version>2.4.3</spark.version>
<hadoop.version>2.8.5</hadoop.version>
<hbase.version>1.4.9</hbase.version>
<hive.version>2.3.5</hive.version>
<java.version>1.8</java.version>
<scala.version>2.11.8</scala.version>
<mysql.version>8.0.22</mysql.version>
<scala.binary.version>2.11</scala.binary.version>
<maven.compiler.source>${java.version}</maven.compiler.source>
<maven.compiler.target>${java.version}</maven.compiler.target>
</properties>
java
<dependency>
<groupId>org.apache.flink</groupId>
<artifactId>flink-connector-kafka_2.12</artifactId>
<version>${flink.version}</version>
</dependency>
二、Flink将Socket数据写入Kafka(精准一次)
注意:如果要使用 精准一次 写入 Kafka,需要满足以下条件,缺一不可
1、开启 checkpoint
2、设置事务前缀
3、设置事务超时时间: checkpoint 间隔 < 事务超时时间 < max 的 15 分钟
java
package com.flink.DataStream.Sink;
import org.apache.flink.api.common.serialization.SimpleStringSchema;
import org.apache.flink.connector.base.DeliveryGuarantee;
import org.apache.flink.connector.kafka.sink.KafkaRecordSerializationSchema;
import org.apache.flink.connector.kafka.sink.KafkaSink;
import org.apache.flink.streaming.api.CheckpointingMode;
import org.apache.flink.streaming.api.datastream.DataStreamSource;
import org.apache.flink.streaming.api.environment.StreamExecutionEnvironment;
import java.util.Properties;
public class flinkSinkKafka {
public static void main(String[] args) throws Exception {
StreamExecutionEnvironment streamExecutionEnvironment = StreamExecutionEnvironment.getExecutionEnvironment();
streamExecutionEnvironment.setParallelism(1);
// 如果是精准一次,必须开启 checkpoint
streamExecutionEnvironment.enableCheckpointing(2000, CheckpointingMode.EXACTLY_ONCE);
DataStreamSource<String> streamSource = streamExecutionEnvironment.socketTextStream("localhost", 8888);
/**
* TODO Kafka Sink
* TODO 注意:如果要使用 精准一次 写入 Kafka,需要满足以下条件,缺一不可
* 1、开启 checkpoint
* 2、设置事务前缀
* 3、设置事务超时时间: checkpoint 间隔 < 事务超时时间 < max 的 15 分钟
*/
Properties properties=new Properties();
properties.put("transaction.timeout.ms",10 * 60 * 1000 + "");
KafkaSink<String> kafkaSink = KafkaSink.<String>builder()
// 指定 kafka 的地址和端口
.setBootstrapServers("localhost:9092")
// 指定序列化器:指定Topic名称、具体的序列化(产生方需要序列化,接收方需要反序列化)
.setRecordSerializer(KafkaRecordSerializationSchema
.<String>builder()
.setTopic("testtopic01")
// 指定value的序列化器
.setValueSerializationSchema(new SimpleStringSchema())
.build()
)
// 写到 kafka 的一致性级别: 精准一次、至少一次
.setDeliverGuarantee(DeliveryGuarantee.EXACTLY_ONCE)
// 如果是精准一次,必须设置 事务的前缀
.setTransactionalIdPrefix("flinkkafkasink-")
// 如果是精准一次,必须设置 事务超时时间: 大于 checkpoint间隔,小于 max 15 分钟
.setKafkaProducerConfig(properties)
//.setProperty(ProducerConfig.TRANSACTION_TIMEOUT_CONFIG, 10 * 60 * 1000 + "")
.build();
streamSource.sinkTo(kafkaSink);
streamExecutionEnvironment.execute();
}
}
查看ProduceerConfig配置
三、启动Zookeeper、Kafka
powershell
#启动zookeeper
${ZK_HOME}/bin/zkServer.sh start
#查看zookeeper状态
${ZK_HOME}/bin/zkServer.sh status
#启动kafka
${KAFKA_HOME}/bin/kafka-server-start.sh ${KAFKA_HOME}/config/server.properties
#查看topic
${KAFKA_HOME}/bin/kafka-topics.sh --list --zookeeper localhost:2181
#创建topic
${KAFKA_HOME}/bin/kafka-topics.sh --zookeeper localhost:2181 --create --topic testtopic02 --partitions 2 --replication-factor 1
#删除topic
${KAFKA_HOME}/bin/kafka-topics.sh --zookeeper localhost:2181 --delete --topic testtopic02
#生产消息
${KAFKA_HOME}/bin/kafka-console-producer.sh --broker-list localhost:9092 --topic testtopic01
#消费消息
${KAFKA_HOME}/bin/kafka-console-consumer.sh --bootstrap-server localhost:9092 --topic testtopic01 --from-beginning
通过socket模拟数据写入Flink之后,Flink将数据写入Kafka