FlinkSQL Kafka to Hive

基于CDH平台，Hive2.1.1-CDH6.2.0 ，写入Hive简单操作。

Hive Streaming

 

import java.time.Duration

import com.sm.common.conf.PropManager
import com.sm.constants.Constants
import com.sm.utils.FlinkUtils
import org.apache.flink.streaming.api.{CheckpointingMode, TimeCharacteristic}
import org.apache.flink.streaming.api.environment.{CheckpointConfig, ExecutionCheckpointingOptions}
import org.apache.flink.streaming.api.scala._
import org.apache.flink.table.api.SqlDialect
import org.apache.flink.table.api.bridge.scala.StreamTableEnvironment
import org.apache.flink.table.catalog.hive.HiveCatalog
import org.apache.log4j.Level
import org.slf4j.LoggerFactory

/**
  * flink sql kafka to hive
  *
  * create by LiuJinHe 2020/10/22
  */
object FlinkSqlKafka2Hive {
  private var logger: org.slf4j.Logger = _

  def main(args: Array[String]): Unit = {
    logger = LoggerFactory.getLogger(this.getClass.getSimpleName)
    org.apache.log4j.Logger.getLogger("org.apache.hadoop").setLevel(Level.WARN)
    org.apache.log4j.Logger.getLogger("org.apache").setLevel(Level.INFO)

    // 初始化 stream 环境
    // 本地测试，需要 flink-runtime-web 依赖
    val env = StreamExecutionEnvironment.createLocalEnvironmentWithWebUI()
    //    val env = StreamExecutionEnvironment.getExecutionEnvironment

    //    参数读取
    //    val params = ParameterTool.fromArgs(args)
    //    env.getConfig.setGlobalJobParameters(params)

    // 失败重启,固定间隔,每隔3秒重启1次,总尝试重启10次
    //    env.setRestartStrategy(RestartStrategies.fixedDelayRestart(10, 3))
    // 本地测试线程 1
    env.setParallelism(1)

    // 事件处理的时间，由系统时间决定
    env.setStreamTimeCharacteristic(TimeCharacteristic.ProcessingTime)

    // 创建 streamTable 环境
    val tableEnv: StreamTableEnvironment = StreamTableEnvironment.create(env, FlinkUtils.getSettings)

    // checkpoint 设置
    val tableConfig = tableEnv.getConfig.getConfiguration
    tableConfig.set(ExecutionCheckpointingOptions.CHECKPOINTING_MODE, CheckpointingMode.EXACTLY_ONCE)
    // checkpoint的超时时间周期，1 分钟做一次checkpoint
    tableConfig.set(ExecutionCheckpointingOptions.CHECKPOINTING_INTERVAL, Duration.ofSeconds(30))
    // checkpoint的超时时间, 检查点一分钟内没有完成将被丢弃
    //    tableConfig.set(ExecutionCheckpointingOptions.CHECKPOINTING_TIMEOUT, Duration.ofSeconds(60))
    // checkpoint 最小间隔，两个检查点之间至少间隔 30 秒
    //    tableConfig.set(ExecutionCheckpointingOptions.MIN_PAUSE_BETWEEN_CHECKPOINTS, Duration.ofSeconds(30))
    // 同一时间只允许进行一个检查点
    //    tableConfig.set(ExecutionCheckpointingOptions.MAX_CONCURRENT_CHECKPOINTS, Integer.valueOf(1))
    // 手动cancel时是否保留checkpoint
    //    tableConfig.set(ExecutionCheckpointingOptions.EXTERNALIZED_CHECKPOINT,
    //      CheckpointConfig.ExternalizedCheckpointCleanup.RETAIN_ON_CANCELLATION)

    // 设置状态的最小空闲时间和最大的空闲时间
    //    tableEnv.getConfig.setIdleStateRetentionTime(Time.hours(12), Time.hours(24))

    // 加载配置
    val catalog_name = PropManager.getProp(Constants.CATALOG_NAME)
    val database = PropManager.getProp(Constants.DEFAULT_DATABASE)
    val schemaDataBase = PropManager.getProp(Constants.SCHEMA_DATABASE)

    // 构造 hive catalog
    val hiveCatalog = new HiveCatalog(
      catalog_name,
      database,
      PropManager.getProp(Constants.HIVE_CONF_DIR)
    )

    tableEnv.registerCatalog(catalog_name, hiveCatalog)
    tableEnv.useCatalog(catalog_name)

    // 构造 kafka