內(nèi)置source和sink

內(nèi)置source包括從文件讀取，從文件夾讀取，從socket中讀取、從集合或者迭代器中讀取。內(nèi)置的sink包括寫文件、控制臺輸出、socket

內(nèi)置connectors

• Apache Kafka?(source/sink)
• Apache Cassandra?(sink)
• Amazon Kinesis Streams?(source/sink)
• Elasticsearch?(sink)
• Hadoop FileSystem?(sink)
• RabbitMQ?(source/sink)
• Apache NiFi?(source/sink)
• Twitter Streaming API?(source)

?HDFS Connector

這個connector提供了一個sink，可以寫分區(qū)到任何一個文件系統(tǒng)（只要支持hadoop filesystem就可以）。

?Kafka Connector

添加依賴

<dependency><groupId>org.apache.flink</groupId><artifactId>flink-connector-kafka_2.11</artifactId><version>${flink.version}</version></dependency>

下載安裝zookeeper

將zookeeper-3.4.5下面的conf文件夾下的zoo_sample.cfg復(fù)制一份zoo.cfg,修改dataDir=/home/vincent/tmp

運(yùn)行zookeeper：./zkServer.sh start?

輸入jps可以看到QuorumPeerMain進(jìn)程說明zk成功啟動了。

下載安裝kafka

• 同樣修改kafka_2.11-2.0.1/config下面的server.properties文件，修改：

? ? ? ? ? ?log.dirs=/home/vincent/tmp/kafka-logs

? ? ? ? ? ?zookeeper.connect=localhost:2181

• 運(yùn)行Kafka：

./bin/kafka-server-start.sh ./config/server.properties

• 輸入jps可以看到Kafka進(jìn)程說明kafka成功啟動了。
• 創(chuàng)建一個topic

vincent@ubuntu:~/kafka_2.11-2.0.1$ ./bin/kafka-topics.sh --create --zookeeper localhost:2181 --replication-factor 1 --partitions 1 --topic mytest

• 創(chuàng)建成功，查看所有topic

vincent@ubuntu:~/kafka_2.11-2.0.1$ ./bin/kafka-topics.sh --list --zookeeper localhost:2181 mytest

• 啟動生產(chǎn)者

vincent@ubuntu:~/kafka_2.11-2.0.1$ ./bin/kafka-console-producer.sh --broker-list localhost:9092 --topic mytest

• 啟動消費(fèi)者

vincent@ubuntu:~/kafka_2.11-2.0.1$ ./bin/kafka-console-consumer.sh --bootstrap-server localhost:9092 --topic mytest

Flink連接Kafka

source 從Kafka中讀取數(shù)據(jù)

Scala：

object KafkaConnectorConsumerApp {def main(args: Array[String]): Unit = {val env = StreamExecutionEnvironment.getExecutionEnvironmentval properties = new Properties()properties.setProperty("bootstrap.servers", "192.168.227.128:9092")properties.setProperty("group.id", "test")env.addSource(new FlinkKafkaConsumer[String]("mytest", new SimpleStringSchema(), properties)).print()env.execute("KafkaConnectorConsumerApp")}

Java：

public class JavaKafkaConsumerApp {public static void main(String[] args) throws Exception {StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();Properties properties = new Properties();properties.setProperty("bootstrap.servers", "192.168.227.128:9092");properties.setProperty("group.id", "test");DataStream<String> stream = env.addSource(new FlinkKafkaConsumer<>("mytest", new SimpleStringSchema(), properties));stream.print();env.execute("JavaKafkaConsumerApp");} }

sink 將數(shù)據(jù)輸出到Kafka中

Scala:

object KafkaConnectorProducerApp {def main(args: Array[String]): Unit = {val env = StreamExecutionEnvironment.getExecutionEnvironment// 從socket接受數(shù)據(jù)，通過Flink，將數(shù)據(jù)Sink到kafkaval data=env.socketTextStream("192.168.227.128", 9999)val properties = new Properties()properties.setProperty("bootstrap.servers", "192.168.227.128:9092")properties.setProperty("group.id", "test")val kafkaSink = new FlinkKafkaProducer[String]("mytest", new KeyedSerializationSchemaWrapper[String](new SimpleStringSchema()), properties)data.addSink(kafkaSink)env.execute("KafkaConnectorProducerApp")} }

Java:

public class JavaKafkaProducerApp {public static void main(String[] args) throws Exception {StreamExecutionEnvironment env = StreamExecutionEnvironment.getExecutionEnvironment();DataStreamSource<String> data = env.socketTextStream("192.168.227.128", 9999);Properties properties = new Properties();properties.setProperty("bootstrap.servers", "192.168.227.128:9092");properties.setProperty("group.id", "test");data.addSink(new FlinkKafkaProducer<String>("192.168.227.128:9092", "mytest", new SimpleStringSchema()));env.execute("JavaKafkaProducerApp");} }

默認(rèn)的flink kafka消費(fèi)策略是setStartFromGroupOffsets(default behaviour)，會自動從上一次未消費(fèi)的數(shù)據(jù)開始

總結(jié)

以上是生活随笔為你收集整理的Apache Flink 零基础入门（二十）Flink kafka connector的全部內(nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯，歡迎將生活随笔推薦給好友。