CODEWITHSUNDEEP
hadoopapache-spark
kostas.kougios
kostas.kougios

Reputation: 993

spark timesout maybe due to binaryFiles() with more than 1 million files in HDFS

I am reading millions of xml files via

val xmls = sc.binaryFiles(xmlDir)

The operation runs fine locally but on yarn it fails with:

 client token: N/A
 diagnostics: Application application_1433491939773_0012 failed 2 times due to ApplicationMaster for attempt appattempt_1433491939773_0012_000002 timed out. Failing the application.
 ApplicationMaster host: N/A
 ApplicationMaster RPC port: -1
 queue: default
 start time: 1433750951883
 final status: FAILED
 tracking URL: http://controller01:8088/cluster/app/application_1433491939773_0012
 user: ariskk
Exception in thread "main" org.apache.spark.SparkException: Application finished with failed status
at org.apache.spark.deploy.yarn.Client.run(Client.scala:622)
at org.apache.spark.deploy.yarn.Client$.main(Client.scala:647)
at org.apache.spark.deploy.yarn.Client.main(Client.scala)
at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:497)
at org.apache.spark.deploy.SparkSubmit$.org$apache$spark$deploy$SparkSubmit$$runMain(SparkSubmit.scala:569)
at org.apache.spark.deploy.SparkSubmit$.doRunMain$1(SparkSubmit.scala:166)
at org.apache.spark.deploy.SparkSubmit$.submit(SparkSubmit.scala:189)
at org.apache.spark.deploy.SparkSubmit$.main(SparkSubmit.scala:110)
at org.apache.spark.deploy.SparkSubmit.main(SparkSubmit.scala)

On hadoops/userlogs logs I am frequently getting these messages:

15/06/08 09:15:38 WARN util.AkkaUtils: Error sending message [message = Heartbeat(1,[Lscala.Tuple2;@2b4f336b,BlockManagerId(1, controller01.stratified, 58510))] in 2 attempts
java.util.concurrent.TimeoutException: Futures timed out after [30 seconds]
at scala.concurrent.impl.Promise$DefaultPromise.ready(Promise.scala:219)
at scala.concurrent.impl.Promise$DefaultPromise.result(Promise.scala:223)
at scala.concurrent.Await$$anonfun$result$1.apply(package.scala:107)
at scala.concurrent.BlockContext$DefaultBlockContext$.blockOn(BlockContext.scala:53)
at scala.concurrent.Await$.result(package.scala:107)
at org.apache.spark.util.AkkaUtils$.askWithReply(AkkaUtils.scala:195)
at org.apache.spark.executor.Executor$$anon$1.run(Executor.scala:427)

I run my spark job via spark-submit and it works for an other HDFS directory that contains only 37k files. Any ideas how to resolve this?

Upvotes: 4

Views: 1399

Answers (1)

kostas.kougios
kostas.kougios

Reputation: 993

Ok after getting some help on sparks mailing list, I found out there were 2 issues:

  1. the src directory, if it is given as /my_dir/ it makes spark fail and creates the heartbeat issues. Instead it should be given as hdfs:///my_dir/*

  2. An out of memory error appears in the logs after fixing #1. This is the spark driver running on yarn running out of memory due to the number of files (apparently it keeps all file info in memory). So I spark-submit'ed the job with --conf spark.driver.memory=8g which fixed the issue.

Upvotes: 4

Related Questions