尝试在 emr 上使用 oozie 执行 spark 提交时出现以下错误

Question

我运行处于集群模式。 apacheds-kerberos-codec-2.0.0-M15.jar 存在于 oozie/share/lib/lib*/spark 和 oozie/share/lib/lib*/oozie 的多个位置。这是环境问题吗？

ava.lang.IllegalArgumentException: Attempt to add (hdfs://ip-172-20-10-53.ec2.internal:8020/user/oozie/share/lib/lib_20170208121307/oozie/apacheds-kerberos-codec-2.0.0-M15.jar) multiple times to the distributed cache.
    at org.apache.spark.deploy.yarn.Client$$anonfun$prepareLocalResources$$anonfun$apply.apply(Client.scala:608)
    at org.apache.spark.deploy.yarn.Client$$anonfun$prepareLocalResources$$anonfun$apply.apply(Client.scala:599)
    at scala.collection.mutable.ArraySeq.foreach(ArraySeq.scala:74)
    at org.apache.spark.deploy.yarn.Client$$anonfun$prepareLocalResources.apply(Client.scala:599)
    at org.apache.spark.deploy.yarn.Client$$anonfun$prepareLocalResources.apply(Client.scala:598)
    at scala.collection.immutable.List.foreach(List.scala:381)
    at org.apache.spark.deploy.yarn.Client.prepareLocalResources(Client.scala:598)
    at org.apache.spark.deploy.yarn.Client.createContainerLaunchContext(Client.scala:868)
    at org.apache.spark.deploy.yarn.Client.submitApplication(Client.scala:170)
    at org.apache.spark.deploy.yarn.Client.run(Client.scala:1154)
    at org.apache.spark.deploy.yarn.Client$.main(Client.scala:1213)
    at org.apache.spark.deploy.yarn.Client.main(Client.scala)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:498)
    at org.apache.spark.deploy.SparkSubmit$.org$apache$spark$deploy$SparkSubmit$$runMain(SparkSubmit.scala:738)
    at org.apache.spark.deploy.SparkSubmit$.doRunMain(SparkSubmit.scala:187)
    at org.apache.spark.deploy.SparkSubmit$.submit(SparkSubmit.scala:212)
    at org.apache.spark.deploy.SparkSubmit$.main(SparkSubmit.scala:126)
    at org.apache.spark.deploy.SparkSubmit.main(SparkSubmit.scala)
    at org.apache.oozie.action.hadoop.SparkMain.runSpark(SparkMain.java:338)
    at org.apache.oozie.action.hadoop.SparkMain.run(SparkMain.java:257)
    at org.apache.oozie.action.hadoop.LauncherMain.run(LauncherMain.java:60)
    at org.apache.oozie.action.hadoop.SparkMain.main(SparkMain.java:78)
    at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
    at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
    at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke(Method.java:498)
    at org.apache.oozie.action.hadoop.LauncherMapper.map(LauncherMapper.java:232)
    at org.apache.hadoop.mapred.MapRunner.run(MapRunner.java:54)
    at org.apache.hadoop.mapred.MapTask.runOldMapper(MapTask.java:455)
    at org.apache.hadoop.mapred.MapTask.run(MapTask.java:344)
    at org.apache.hadoop.mapred.LocalContainerLauncher$EventHandler.runSubtask(LocalContainerLauncher.java:380)
    at org.apache.hadoop.mapred.LocalContainerLauncher$EventHandler.runTask(LocalContainerLauncher.java:301)
    at org.apache.hadoop.mapred.LocalContainerLauncher$EventHandler.access0(LocalContainerLauncher.java:187)
    at org.apache.hadoop.mapred.LocalContainerLauncher$EventHandler.run(LocalContainerLauncher.java:230)
    at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
    at java.util.concurrent.FutureTask.run(FutureTask.java:266)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
    at java.lang.Thread.run(Thread.java:745)

Answer 1

oozie sharelib 和 spark sharelib 目录似乎共享相同的 jar，运行使用 spark 工作流导入这两个目录，hadoop-3 代码库不喜欢。

我不得不重新组织 oozie sharelib 目录以仅包含 oozie 特定的 jar，这样 oozie 和 spark sharelib 目录之间就没有重复项：

export HADOOP_USER_NAME=oozie
hdfs dfs -mv /user/oozie/share/lib/lib_20170222143042/oozie /user/oozie/share/lib/lib_20170222143042/oozie.old
hdfs dfs -mkdir /user/oozie/share/lib/lib_20170222143042/oozie
hdfs dfs -cp /user/oozie/share/lib/lib_20170222143042/oozie.old/oozie-hadoop-utils-hadoop-2-4.3.0.jar /user/oozie/share/lib/lib_20170222143042/oozie
hdfs dfs -cp /user/oozie/share/lib/lib_20170222143042/oozie.old/oozie-sharelib-oozie-4.3.0.jar /user/oozie/share/lib/lib_20170222143042/oozie

这解决了能够从 oozie 运行激发工作流的直接问题，但我不确定这是否会影响非激发工作流。

Answer 2

我有一个启动 Spark 作业的 oozie 作业，运行在 Amazon EMR 中。当 EMR Hadoop 设置有一个主实例和一个从属实例时，我遇到了同样的错误。当我将奴隶的实例数量增加到两个时，一切都按预期进行。

尝试在 emr 上使用 oozie 执行 spark 提交时出现以下错误

I am getting the below error while trying to execute spark submit using oozie on emr

emr

oozie

apache-spark

spark-submit