Patrick Wendell created SPARK-5492:
--------------------------------------
Summary: Thread statistics can break with older Hadoop versions
Key: SPARK-5492
URL: https://issues.apache.org/jira/browse/SPARK-5492
Project: Spark
Issue Type: Bug
Components: Spark Core
Reporter: Patrick Wendell
{code}
java.lang.ClassNotFoundException:
org.apache.hadoop.fs.FileSystem$Statistics$StatisticsData
at java.net.URLClassLoader$1.run(URLClassLoader.java:366)
at java.net.URLClassLoader$1.run(URLClassLoader.java:355)
at java.security.AccessController.doPrivileged(Native Method)
at java.net.URLClassLoader.findClass(URLClassLoader.java:354)
at java.lang.ClassLoader.loadClass(ClassLoader.java:425)
at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:308)
at java.lang.ClassLoader.loadClass(ClassLoader.java:358)
at java.lang.Class.forName0(Native Method)
at java.lang.Class.forName(Class.java:191)
at
org.apache.spark.deploy.SparkHadoopUtil.getFileSystemThreadStatisticsMethod(SparkHadoopUtil.scala:180)
at
org.apache.spark.deploy.SparkHadoopUtil.getFSBytesReadOnThreadCallback(SparkHadoopUtil.scala:139)
at
org.apache.spark.rdd.NewHadoopRDD$$anon$1$$anonfun$2.apply(NewHadoopRDD.scala:120)
at
org.apache.spark.rdd.NewHadoopRDD$$anon$1$$anonfun$2.apply(NewHadoopRDD.scala:118)
at scala.Option.orElse(Option.scala:257)
{code}
I think the issue is we need to catch ClassNotFoundException here:
https://github.com/apache/spark/blob/b1b35ca2e440df40b253bf967bb93705d355c1c0/core/src/main/scala/org/apache/spark/deploy/SparkHadoopUtil.scala#L144
However, I'm really confused how this didn't fail our unit tests, since we
explicitly tried to test this.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org
