Flink比较完美的支持kerberos读写Kafka.

1 flink-conf.yaml  , 将如下涉及kerberos的配置注解去除

2 代码读写kafka示例如下：

def main(args: Array[String]): Unit = {
    val params: ParameterTool = ParameterTool.fromArgs(args)

    // set up execution environment
    val env = StreamExecutionEnvironment.getExecutionEnvironment
    // make parameters available in the web interface
    env.getConfig.setGlobalJobParameters(params)
    env.setStreamTimeCharacteristic(TimeCharacteristic.EventTime)
    env.enableCheckpointing(10*1000,CheckpointingMode.EXACTLY_ONCE)

    val properties = new Properties()
    properties.setProperty("bootstrap.servers", "192.168.142.13:9092,192.168.142.11:9092,192.168.142.12
    properties.setProperty("group.id", "test")
    properties.setProperty("security.protocol","SASL_PLAINTEXT")
    properties.setProperty("sasl.mechanism","GSSAPI")
    properties.setProperty("sasl.kerberos.service.name","kafka")

    //读取kafka -source
    val stream = env
      .addSource(new FlinkKafkaConsumer[String]("test", new SimpleStringSchema(), properties))
    // .print()  

    // 写入kafka -sink
    val myProducer = new FlinkKafkaProducer("sink-topic",new SimpleStringSchema(),properties)
    stream.addSink(myProducer).setParallelism(1)
    env.execute()
  }

3 在kafka consolei新建2各topic: test 用于消费  sink-topic用于写入。

在控制台手动生产数据，手动启动消费sink-topic控制台。部署flink后观察sink-topic是否写入成功即可。