[
https://issues.apache.org/jira/browse/SPARK-17842?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Sreelal S L updated SPARK-17842:
--------------------------------
Comment: was deleted
(was: SparkStreaming’s Window internally uses UnionRDD to merge all the
RDDs(created during each batch) in the window time frame.
They have a logic isPartitionListingParallel: Boolean = rdds.length >
conf.getInt("spark.rdd.parallelListingThreshold", 10)
In our case , the rdd count per window is well above 10 , and hence it will
use parallel partition computation.
partitionEvalTaskSupport =new ForkJoinTaskSupport(new ForkJoinPool(8))
This pool is created every slide interval but ForkJoinPool.shutdown() is not
called anywhere.
This can be the reason for scala.concurrent.forkjoin.ForkJoinTask[] to leak .
This is causing a thread leak. )
> Thread and memory leak in WindowDstream (UnionRDD ) when parallelPartition
> computation gets enabled.
> -----------------------------------------------------------------------------------------------------
>
> Key: SPARK-17842
> URL: https://issues.apache.org/jira/browse/SPARK-17842
> Project: Spark
> Issue Type: Bug
> Components: Spark Core, Streaming
> Affects Versions: 2.0.0
> Environment: Yarn cluster, Eclipse Dev Env
> Reporter: Sreelal S L
> Priority: Critical
>
> We noticed a steady increase in ForkJoinTask instances in the driver process
> heap. Found out the root cause to be UnionRDD.
> WindowDstream internally uses UnionRDD which has a parallel partition
> computation logic by using parallel collection with ForkJoinPool task
> support.
> partitionEvalTaskSupport =new ForkJoinTaskSupport(new ForkJoinPool(8))
> The pool is created each time when a UnionRDD is created , but the pool is
> not getting shutdown. This is leaking thread/mem every slide interval of the
> window.
> Easily reproducible with the below code. Just keep a watch on the number of
> threads.
> {code}
> val sparkConf = new
> SparkConf().setMaster("local[*]").setAppName("TestLeak")
> val ssc = new StreamingContext(sparkConf, Seconds(1))
> ssc.checkpoint("checkpoint")
> val rdd = ssc.sparkContext.parallelize(List(1,2,3))
> val constStream = new ConstantInputDStream[Int](ssc,rdd)
> constStream.window(Seconds(20),Seconds(1)).print()
> ssc.start()
> ssc.awaitTermination();
> {code}
> This happens only when the number of rdds to be unioned is above the value
> spark.rdd.parallelListingThreshold (By default 10)
> Currently i'm working around by setting this threshold be a higher value.
>
--
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
