[SPARK-23894] Flaky Test: BucketedWriteWithoutHiveSupportSuite

• Type: Improvement
• Status: In Progress
• Priority: Minor
• Resolution: Unresolved
• Affects Version/s: 3.1.0
• Component/s: SQL
• Assignee: Unassigned
• Reporter: Imran Rashid
• Created: 08/Apr/18 00:57
• Updated: 25/Feb/22 10:19

Description

Flaky test observed here: https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/88991/

I'll attach a snippet of the unit-tests logs, for this suite and the preceeding one. Here's a snippet of the exception.

08:36:34.694 Executor task launch worker for task 436 ERROR Executor: Exception in task 0.0 in stage 402.0 (TID 436)
java.lang.IllegalStateException: LiveListenerBus is stopped.
        at org.apache.spark.scheduler.LiveListenerBus.addToQueue(LiveListenerBus.scala:97)
        at org.apache.spark.scheduler.LiveListenerBus.addToStatusQueue(LiveListenerBus.scala:80)
        at org.apache.spark.sql.internal.SharedState.<init>(SharedState.scala:93)
        at org.apache.spark.sql.SparkSession$$anonfun$sharedState$1.apply(SparkSession.scala:117)
        at org.apache.spark.sql.SparkSession$$anonfun$sharedState$1.apply(SparkSession.scala:117)
        at scala.Option.getOrElse(Option.scala:121)
        at org.apache.spark.sql.SparkSession.sharedState$lzycompute(SparkSession.scala:117)
        at org.apache.spark.sql.SparkSession.sharedState(SparkSession.scala:116)
        at org.apache.spark.sql.internal.BaseSessionStateBuilder.build(BaseSessionStateBuilder.scala:286)
        at org.apache.spark.sql.test.TestSparkSession.sessionState$lzycompute(TestSQLContext.scala:42)
        at org.apache.spark.sql.test.TestSparkSession.sessionState(TestSQLContext.scala:41)
        at org.apache.spark.sql.SparkSession$$anonfun$1$$anonfun$apply$1.apply(SparkSession.scala:92)
        at org.apache.spark.sql.SparkSession$$anonfun$1$$anonfun$apply$1.apply(SparkSession.scala:92)
        at scala.Option.map(Option.scala:146)
        at org.apache.spark.sql.SparkSession$$anonfun$1.apply(SparkSession.scala:92)
        at org.apache.spark.sql.SparkSession$$anonfun$1.apply(SparkSession.scala:91)
        at org.apache.spark.sql.internal.SQLConf$.get(SQLConf.scala:110)
        at org.apache.spark.sql.types.DataType.sameType(DataType.scala:84)
        at org.apache.spark.sql.catalyst.analysis.TypeCoercion$$anonfun$1.apply(TypeCoercion.scala:105)
        at org.apache.spark.sql.catalyst.analysis.TypeCoercion$$anonfun$1.apply(TypeCoercion.scala:86)

I doubt this is actually because of BucketedWriteWithoutHiveSupportSuite. I think it has something more to do with SparkSession 's lazy evaluation of SharedState doing something funny with the way we setup the test spark context etc ... though I don't really understand it yet.

Issue Links

• blocks: SPARK-24139 Fix Flaky Tests
• is related to: SPARK-22938 Assert that SQLConf.get is accessed only on the driver.
• relates to: SPARK-38328 SQLConf.get flaky causes NON-default spark session settings being lost
• links to: [Github] Pull Request #21185 (squito)