[SPARK-12940] Partition field in Spark SQL WHERE clause causing Exception

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Cannot Reproduce
• Affects Version/s: 1.5.2, 1.6.0
• Component/s: SQL

Environment

AWS EMR 4.2, OSX

• Assignee: Unassigned
• Reporter: Brian Wheeler
• Created: 20/Jan/16 22:21
• Updated: 12/Dec/22 18:10
• Resolved: 10/Jan/17 07:41

Description

I have partitioned Parquet that I am trying to query with Spark SQL. When I involve a partition column in the WHERE clause when using OR I get an exception.

I have had this issue when using spark-submit on a cluster when the Parquet was created externally and registered with Hive JDBC-backed metastore externally. I can also duplicate this behavior with a simplified example in the spark shell. I will include the simplified example. Note that I am using my hive-site.xml when I launch the spark-shell so the metastore is set up the same way.

I also tried this locally with the same results on a Mac laptop with 1.6.0.

Create some partitioned parquet:

case class Hit(meta_ts_unix_ms: Long, username: String, srclatitude: Double, srclongitude: Double, srccity: String, srcregion: String, srccountrycode: String, metaclass: String)

val rdd = sc.parallelize(Array(Hit(34L, "user1", 45.2, 23.2, "city1", "state1", "US", "blah, other"), Hit(35L, "user1", 53.2, 11.2, "city2", "state2", "US", "blah")))

sqlContext.createDataFrame(rdd).registerTempTable("test_table")

sqlContext.sql("select * from test_table where meta_ts_unix_ms = 35").write.parquet("file:///tmp/year=2015/month=12/day=4/hour=1/")
sqlContext.sql("select * from test_table where meta_ts_unix_ms = 34").write.parquet("file:///tmp/year=2015/month=12/day=3/hour=23/")

Create an external table from the parquet:

sqlContext.createExternalTable("test_table2", "file:///tmp/year=2015/", "parquet")

If I understand correctly the partitions were discovered automatically because they show up in the describe command even though they were not part of the schema generated from the case classes:

+---------------+---------+-------+
|       col_name|data_type|comment|
+---------------+---------+-------+
|meta_ts_unix_ms|   bigint|       |
|       username|   string|       |
|    srclatitude|   double|       |
|   srclongitude|   double|       |
|        srccity|   string|       |
|      srcregion|   string|       |
| srccountrycode|   string|       |
|      metaclass|   string|       |
|           year|      int|       |
|          month|      int|       |
|            day|      int|       |
|           hour|      int|       |
+---------------+---------+-------+

This query:

sqlContext.sql("SELECT meta_ts_unix_ms,username,srclatitude,srclongitude,srccity,srcregion,srccountrycode FROM test_table2 WHERE meta_ts_unix_ms IS NOT NULL AND username IS NOT NULL AND metaclass like '%blah%' OR hour = 1").show()

Throws this exception:

16/01/20 21:36:46 WARN TaskSetManager: Lost task 0.0 in stage 13.0 (TID 84, ip-192-168-111-222.ec2.internal): org.apache.spark.sql.catalyst.errors.package$TreeNodeException: Binding attribute, tree: metaclass#53
	at org.apache.spark.sql.catalyst.errors.package$.attachTree(package.scala:49)
	at org.apache.spark.sql.catalyst.expressions.BindReferences$$anonfun$bindReference$1.applyOrElse(BoundAttribute.scala:86)
	at org.apache.spark.sql.catalyst.expressions.BindReferences$$anonfun$bindReference$1.applyOrElse(BoundAttribute.scala:85)
	at org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$3.apply(TreeNode.scala:227)
	at org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$3.apply(TreeNode.scala:227)
	at org.apache.spark.sql.catalyst.trees.CurrentOrigin$.withOrigin(TreeNode.scala:51)
	at org.apache.spark.sql.catalyst.trees.TreeNode.transformDown(TreeNode.scala:226)
	at org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$transformDown$1.apply(TreeNode.scala:232)
	at org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$transformDown$1.apply(TreeNode.scala:232)
	at org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$4.apply(TreeNode.scala:249)
	at scala.collection.Iterator$$anon$11.next(Iterator.scala:328)
	at scala.collection.Iterator$class.foreach(Iterator.scala:727)
	at scala.collection.AbstractIterator.foreach(Iterator.scala:1157)
	at scala.collection.generic.Growable$class.$plus$plus$eq(Growable.scala:48)
	at scala.collection.mutable.ArrayBuffer.$plus$plus$eq(ArrayBuffer.scala:103)
	at scala.collection.mutable.ArrayBuffer.$plus$plus$eq(ArrayBuffer.scala:47)
	at scala.collection.TraversableOnce$class.to(TraversableOnce.scala:273)
	at scala.collection.AbstractIterator.to(Iterator.scala:1157)
	at scala.collection.TraversableOnce$class.toBuffer(TraversableOnce.scala:265)
	at scala.collection.AbstractIterator.toBuffer(Iterator.scala:1157)
	at scala.collection.TraversableOnce$class.toArray(TraversableOnce.scala:252)
	at scala.collection.AbstractIterator.toArray(Iterator.scala:1157)
	at org.apache.spark.sql.catalyst.trees.TreeNode.transformChildren(TreeNode.scala:279)
	at org.apache.spark.sql.catalyst.trees.TreeNode.transformDown(TreeNode.scala:232)
	at org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$transformDown$1.apply(TreeNode.scala:232)
	at org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$transformDown$1.apply(TreeNode.scala:232)
	at org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$4.apply(TreeNode.scala:249)
	at scala.collection.Iterator$$anon$11.next(Iterator.scala:328)
	at scala.collection.Iterator$class.foreach(Iterator.scala:727)
	at scala.collection.AbstractIterator.foreach(Iterator.scala:1157)
	at scala.collection.generic.Growable$class.$plus$plus$eq(Growable.scala:48)
	at scala.collection.mutable.ArrayBuffer.$plus$plus$eq(ArrayBuffer.scala:103)
	at scala.collection.mutable.ArrayBuffer.$plus$plus$eq(ArrayBuffer.scala:47)
	at scala.collection.TraversableOnce$class.to(TraversableOnce.scala:273)
	at scala.collection.AbstractIterator.to(Iterator.scala:1157)
	at scala.collection.TraversableOnce$class.toBuffer(TraversableOnce.scala:265)
	at scala.collection.AbstractIterator.toBuffer(Iterator.scala:1157)
	at scala.collection.TraversableOnce$class.toArray(TraversableOnce.scala:252)
	at scala.collection.AbstractIterator.toArray(Iterator.scala:1157)
	at org.apache.spark.sql.catalyst.trees.TreeNode.transformChildren(TreeNode.scala:279)
	at org.apache.spark.sql.catalyst.trees.TreeNode.transformDown(TreeNode.scala:232)
	at org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$transformDown$1.apply(TreeNode.scala:232)
	at org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$transformDown$1.apply(TreeNode.scala:232)
	at org.apache.spark.sql.catalyst.trees.TreeNode$$anonfun$4.apply(TreeNode.scala:249)
	at scala.collection.Iterator$$anon$11.next(Iterator.scala:328)
	at scala.collection.Iterator$class.foreach(Iterator.scala:727)
	at scala.collection.AbstractIterator.foreach(Iterator.scala:1157)
	at scala.collection.generic.Growable$class.$plus$plus$eq(Growable.scala:48)
	at scala.collection.mutable.ArrayBuffer.$plus$plus$eq(ArrayBuffer.scala:103)
	at scala.collection.mutable.ArrayBuffer.$plus$plus$eq(ArrayBuffer.scala:47)
	at scala.collection.TraversableOnce$class.to(TraversableOnce.scala:273)
	at scala.collection.AbstractIterator.to(Iterator.scala:1157)
	at scala.collection.TraversableOnce$class.toBuffer(TraversableOnce.scala:265)
	at scala.collection.AbstractIterator.toBuffer(Iterator.scala:1157)
	at scala.collection.TraversableOnce$class.toArray(TraversableOnce.scala:252)
	at scala.collection.AbstractIterator.toArray(Iterator.scala:1157)
	at org.apache.spark.sql.catalyst.trees.TreeNode.transformChildren(TreeNode.scala:279)
	at org.apache.spark.sql.catalyst.trees.TreeNode.transformDown(TreeNode.scala:232)
	at org.apache.spark.sql.catalyst.trees.TreeNode.transform(TreeNode.scala:217)
	at org.apache.spark.sql.catalyst.expressions.BindReferences$.bindReference(BoundAttribute.scala:85)
	at org.apache.spark.sql.catalyst.expressions.InterpretedPredicate$.create(predicates.scala:31)
	at org.apache.spark.sql.execution.SparkPlan.newPredicate(SparkPlan.scala:281)
	at org.apache.spark.sql.execution.Filter$$anonfun$4.apply(basicOperators.scala:114)
	at org.apache.spark.sql.execution.Filter$$anonfun$4.apply(basicOperators.scala:113)
	at org.apache.spark.rdd.RDD$$anonfun$mapPartitions$1$$anonfun$apply$17.apply(RDD.scala:710)
	at org.apache.spark.rdd.RDD$$anonfun$mapPartitions$1$$anonfun$apply$17.apply(RDD.scala:710)
	at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:38)
	at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:300)
	at org.apache.spark.rdd.RDD.iterator(RDD.scala:264)
	at org.apache.spark.rdd.MapPartitionsRDD.compute(MapPartitionsRDD.scala:38)
	at org.apache.spark.rdd.RDD.computeOrReadCheckpoint(RDD.scala:300)
	at org.apache.spark.rdd.RDD.iterator(RDD.scala:264)
	at org.apache.spark.scheduler.ResultTask.runTask(ResultTask.scala:66)
	at org.apache.spark.scheduler.Task.run(Task.scala:88)
	at org.apache.spark.executor.Executor$TaskRunner.run(Executor.scala:214)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
	at java.lang.Thread.run(Thread.java:745)
Caused by: java.lang.RuntimeException: Couldn't find metaclass#53 in [meta_ts_unix_ms#45L,username#46,srclatitude#47,srclongitude#48,srccity#49,srcregion#50,srccountrycode#51]
	at scala.sys.package$.error(package.scala:27)
	at org.apache.spark.sql.catalyst.expressions.BindReferences$$anonfun$bindReference$1$$anonfun$applyOrElse$1.apply(BoundAttribute.scala:92)
	at org.apache.spark.sql.catalyst.expressions.BindReferences$$anonfun$bindReference$1$$anonfun$applyOrElse$1.apply(BoundAttribute.scala:86)
	at org.apache.spark.sql.catalyst.errors.package$.attachTree(package.scala:48)
	... 77 more

This query works fine and returns expected results but it does not involve any of the partition columns in the OR portion of the WHERE clause:

sqlContext.sql("SELECT meta_ts_unix_ms,username,srclatitude,srclongitude,srccity,srcregion,srccountrycode FROM test_table2 WHERE meta_ts_unix_ms IS NOT NULL AND username IS NOT NULL AND metaclass like '%other%' OR metaclass = 'blah'").show()

Issue Links

• is related to: SPARK-26041 catalyst cuts out some columns from dataframes: org.apache.spark.sql.catalyst.errors.package$TreeNodeException: Binding attribute