Listener EventLoggingListener threw an exception java.io.IOException

2018年7月11日 0 条评论 4.71k 次阅读 2 人点赞

问题来源

最近在做实时流计算，Coding过程中需要对写入到Redis中的数据进行Dump，经过综合考量决定采用Spark Core进行数据操作，并将结果数据Dump到HDFS上。Spark程序操作HDFS的工具类中，手动获取了FileSystem，执行完HDFS相关操作后对FileSystem进行close：

Scala
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
object HdfsHandler {
 
  @transient val logger = LoggerFactory.getLogger(this.getClass)
  val config: Configuration = new Configuration()
  /**
    * Get the storage path for current batch data.
    *
    * @param dateTime DateTime
    * @return storage path in String
    */
  def getCurrentPath(dateTime: DateTime): String = {
    var fs: FileSystem = null
    try {
      fs = FileSystem.get(config)
    } catch {
      case e: Exception => {
        logger.error("Get fs:FileSystem error!")
        e.printStackTrace()
      }
    }
    val dateDir: String = getDateDir(dateTime)
    val hour = dateTime.toString(AppConfig.HDFS_DIR_DATE_HOUR_FORMATTER)
    val pathStr = dateDir
      .concat(File.separator)
      .concat(hour)
    try {
      delHdfsFile(fs, pathStr)
    } catch {
      case e: Exception => {
        logger.error("Delete files from the directory to which the data will be storaged to failed!")
        e.printStackTrace()
      }
    }finally{
        fs.close()
   }
    pathStr
  }
.......
  /**
    * Delete file on HDFS.
    *
    * @param fs   FileSystem
    * @param elem string path
    */
  private def delHdfsFile(fs: FileSystem, elem: String): Unit = {
    try {
      val path = new Path(elem)
      if (fs.exists(path)) {
        fs.delete(path, true)
      }
    } catch {
      case e: Exception => {
        e.printStackTrace()
        new Throwable("Delete HDFS files Error! codeLocation: HdfsHandler.delHdfsFile")
      }
    }
  }
}

本来是标准资源释放流程，但是却抛出了一个大大的异常：

Shell
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
18/07/10 15:42:18 ERROR LiveListenerBus: Listener EventLoggingListener threw an exception
java.io.IOException: Filesystem closed
at org.apache.hadoop.hdfs.DFSClient.checkOpen(DFSClient.java:798)
at org.apache.hadoop.hdfs.DFSOutputStream.flushOrSync(DFSOutputStream.java:1985)
at org.apache.hadoop.hdfs.DFSOutputStream.hflush(DFSOutputStream.java:1946)
at org.apache.hadoop.fs.FSDataOutputStream.hflush(FSDataOutputStream.java:130)
at org.apache.spark.scheduler.EventLoggingListener$$anonfun$logEvent$3.apply(EventLoggingListener.scala:140)
at org.apache.spark.scheduler.EventLoggingListener$$anonfun$logEvent$3.apply(EventLoggingListener.scala:140)
at scala.Option.foreach(Option.scala:257)
at org.apache.spark.scheduler.EventLoggingListener.logEvent(EventLoggingListener.scala:140)
at org.apache.spark.scheduler.EventLoggingListener.onJobStart(EventLoggingListener.scala:163)
at org.apache.spark.scheduler.SparkListenerBus$class.doPostEvent(SparkListenerBus.scala:37)
at org.apache.spark.scheduler.LiveListenerBus.doPostEvent(LiveListenerBus.scala:36)
at org.apache.spark.scheduler.LiveListenerBus.doPostEvent(LiveListenerBus.scala:36)
at org.apache.spark.util.ListenerBus$class.postToAll(ListenerBus.scala:63)
at org.apache.spark.scheduler.LiveListenerBus.postToAll(LiveListenerBus.scala:36)
at org.apache.spark.scheduler.LiveListenerBus$$anon$1$$anonfun$run$1$$anonfun$apply$mcV$sp$1.apply$mcV$sp(LiveListenerBus.scala:94)
at org.apache.spark.scheduler.LiveListenerBus$$anon$1$$anonfun$run$1$$anonfun$apply$mcV$sp$1.apply(LiveListenerBus.scala:79)
at org.apache.spark.scheduler.LiveListenerBus$$anon$1$$anonfun$run$1$$anonfun$apply$mcV$sp$1.apply(LiveListenerBus.scala:79)
at scala.util.DynamicVariable.withValue(DynamicVariable.scala:58)
at org.apache.spark.scheduler.LiveListenerBus$$anon$1$$anonfun$run$1.apply$mcV$sp(LiveListenerBus.scala:78)
at org.apache.spark.util.Utils$.tryOrStopSparkContext(Utils.scala:1245)
at org.apache.spark.scheduler.LiveListenerBus$$anon$1.run(LiveListenerBus.scala:77)

解决方法

由于Spark程序是以Spark-On-Yarn模式运行，Spark会把job日志是写在HDFS上。他们使用的是同一个FileSystem，从而导致异常。将fs.close()去掉就会恢复正常。

© 著作权归作者所有

勇敢，和生活的艰难无关。

文章评论（0）