[
https://issues.apache.org/jira/browse/SPARK-3039?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14304991#comment-14304991
]
Markus Dale commented on SPARK-3039:
------------------------------------
The Aapche Maven Enforcer plugin has a dependency convergence rule that could
be added to ensure that dependencies/transitive dependencies don't clash. Maybe
this could be added to a "deploy" profile to be checked (initially set fail to
false until all dependency convergence errors are fixed) during release builds
initially. See https://issues.apache.org/jira/browse/SPARK-5584.
For the spark-1.3.0-SNAPSHOT, looks like the fix for "[SPARK-4048] Enhance and
extend hadoop-provided profile" introduced a new scope attribute in the Maven
build:
{code}
grep -R hive.deps.scope *
assembly/pom.xml: <hive.deps.scope>provided</hive.deps.scope>
examples/pom.xml: <hive.deps.scope>provided</hive.deps.scope>
pom.xml: <hive.deps.scope>compile</hive.deps.scope>
pom.xml: <scope>${hive.deps.scope}</scope>
pom.xml: <scope>${hive.deps.scope}</scope>
pom.xml: <scope>${hive.deps.scope}</scope>
pom.xml: <scope>${hive.deps.scope}</scope>
pom.xml: <scope>${hive.deps.scope}</scope>
pom.xml: <scope>${hive.deps.scope}</scope>
pom.xml: <scope>${hive.deps.scope}</scope>
{code}
avro-mapred and hive-exec are marked with that scope so neither library nor
their dependencies will be included in the spark assembly jar. This means that
Spark jobs that want to use Avro-based InputFormat from avro-mapred have to
include their desired avro-mapred version in their jars. So this particular
problem will be gone but still need to prevent this class of problems by
ensuring dependency convergence.
> Spark assembly for new hadoop API (hadoop 2) contains avro-mapred for hadoop
> 1 API
> ----------------------------------------------------------------------------------
>
> Key: SPARK-3039
> URL: https://issues.apache.org/jira/browse/SPARK-3039
> Project: Spark
> Issue Type: Bug
> Components: Build, Input/Output, Spark Core
> Affects Versions: 0.9.1, 1.0.0, 1.1.0, 1.2.0
> Environment: hadoop2, hadoop-2.4.0, HDP-2.1
> Reporter: Bertrand Bossy
> Assignee: Bertrand Bossy
> Priority: Critical
>
> The spark assembly contains the artifact "org.apache.avro:avro-mapred" as a
> dependency of "org.spark-project.hive:hive-serde".
> The avro-mapred package provides a hadoop FileInputFormat to read and write
> avro files. There are two versions of this package, distinguished by a
> classifier. avro-mapred for the new Hadoop API uses the classifier "hadoop2".
> avro-mapred for the old Hadoop API uses no classifier.
> E.g. when reading avro files using
> {code}
> sc.newAPIHadoopFile[AvroKey[SomeClass]],NullWritable,AvroKeyInputFormat[SomeClass]]("hdfs://path/to/file.avro")
> {code}
> The following error occurs:
> {code}
> java.lang.IncompatibleClassChangeError: Found interface
> org.apache.hadoop.mapreduce.TaskAttemptContext, but class was expected
> at
> org.apache.avro.mapreduce.AvroKeyInputFormat.createRecordReader(AvroKeyInputFormat.java:47)
> at
> org.apache.spark.rdd.NewHadoopRDD$$anon$1.<init>(NewHadoopRDD.scala:111)
> at org.apache.spark.rdd.NewHadoopRDD.compute(NewHadoopRDD.scala:99)
> at org.apache.spark.rdd.NewHadoopRDD.compute(NewHadoopRDD.scala:61)
> at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:262)
> at org.apache.spark.rdd.RDD.iterator(RDD.scala:229)
> at org.apache.spark.rdd.MappedRDD.compute(MappedRDD.scala:31)
> at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:262)
> at org.apache.spark.rdd.RDD.iterator(RDD.scala:229)
> at org.apache.spark.rdd.FilteredRDD.compute(FilteredRDD.scala:34)
> at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:262)
> at org.apache.spark.rdd.RDD.iterator(RDD.scala:229)
> at org.apache.spark.rdd.MappedRDD.compute(MappedRDD.scala:31)
> at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:262)
> at org.apache.spark.rdd.RDD.iterator(RDD.scala:229)
> at
> org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:158)
> at
> org.apache.spark.scheduler.ShuffleMapTask.runTask(ShuffleMapTask.scala:99)
> at org.apache.spark.scheduler.Task.run(Task.scala:51)
> at
> org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:187)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
> at java.lang.Thread.run(Thread.java:744)
> {code}
> This error usually is a hint that there was a mix up of the old and the new
> Hadoop API. As a work-around, if avro-mapred for hadoop2 is "forced" to
> appear before the version that is bundled with Spark, reading avro files
> works fine.
> Also, if Spark is built using avro-mapred for hadoop2, it works fine as well.
--
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
