如何控制 Spark History Server 的内存堆大小？

Question

我们在 YARN 上运行带有 CDH 5.3.2 和 Spark History Server 的 Spark (1.2)。

对于小型作业，历史服务器能够工作，但对于少数大型作业，Spark History Server 无法检索日志/作业历史。并在

中显示以下错误

  2015-04-09 09:50:48,061 WARN     org.eclipse.jetty.servlet.ServletHandler: Error for /history/application_1428034115331_31584
org.spark-project.guava.common.util.concurrent.ExecutionError: java.lang.OutOfMemoryError: Java heap space
at org.spark-project.guava.common.cache.LocalCache$Segment.get(LocalCache.java:2261)
at org.spark-project.guava.common.cache.LocalCache.get(LocalCache.java:4000)
at org.spark-project.guava.common.cache.LocalCache.getOrLoad(LocalCache.java:4004)
at org.spark-project.guava.common.cache.LocalCache$LocalLoadingCache.get(LocalCache.java:4874)
at org.apache.spark.deploy.history.HistoryServer$$anon$1.doGet(HistoryServer.scala:85)
at javax.servlet.http.HttpServlet.service(HttpServlet.java:735)
at javax.servlet.http.HttpServlet.service(HttpServlet.java:848)
at org.eclipse.jetty.servlet.ServletHolder.handle(ServletHolder.java:684)
at org.eclipse.jetty.servlet.ServletHandler.doHandle(ServletHandler.java:501)
at org.eclipse.jetty.server.handler.ContextHandler.doHandle(ContextHandler.java:1086)
at org.eclipse.jetty.servlet.ServletHandler.doScope(ServletHandler.java:428)
at org.eclipse.jetty.server.handler.ContextHandler.doScope(ContextHandler.java:1020)
at org.eclipse.jetty.server.handler.ScopedHandler.handle(ScopedHandler.java:135)
at org.eclipse.jetty.server.handler.ContextHandlerCollection.handle(ContextHandlerCollection.java:255)
at org.eclipse.jetty.server.handler.HandlerWrapper.handle(HandlerWrapper.java:116)
at org.eclipse.jetty.server.Server.handle(Server.java:370)
at org.eclipse.jetty.server.AbstractHttpConnection.handleRequest(AbstractHttpConnection.java:494)
at org.eclipse.jetty.server.AbstractHttpConnection.headerComplete(AbstractHttpConnection.java:971)
at org.eclipse.jetty.server.AbstractHttpConnection$RequestHandler.headerComplete(AbstractHttpConnection.java:1033)
at org.eclipse.jetty.http.HttpParser.parseNext(HttpParser.java:644)
at org.eclipse.jetty.http.HttpParser.parseAvailable(HttpParser.java:235)
at org.eclipse.jetty.server.AsyncHttpConnection.handle(AsyncHttpConnection.java:82)
at org.eclipse.jetty.io.nio.SelectChannelEndPoint.handle(SelectChannelEndPoint.java:667)
at org.eclipse.jetty.io.nio.SelectChannelEndPoint$1.run(SelectChannelEndPoint.java:52)
at org.eclipse.jetty.util.thread.QueuedThreadPool.runJob(QueuedThreadPool.java:608)
at org.eclipse.jetty.util.thread.QueuedThreadPool$3.run(QueuedThreadPool.java:543)
at java.lang.Thread.run(Thread.java:745)
Caused by: java.lang.OutOfMemoryError: Java heap space
at scala.collection.TraversableLike$class.map(TraversableLike.scala:244)
at scala.collection.AbstractTraversable.map(Traversable.scala:105)
at org.json4s.MonadicJValue$$anonfun$org$json4s$MonadicJValue$$findDirectByName$1.apply(MonadicJValue.scala:26)
at org.json4s.MonadicJValue$$anonfun$org$json4s$MonadicJValue$$findDirectByName$1.apply(MonadicJValue.scala:22)
at scala.collection.TraversableLike$$anonfun$flatMap$1.apply(TraversableLike.scala:251)
at scala.collection.TraversableLike$$anonfun$flatMap$1.apply(TraversableLike.scala:251)
at scala.collection.immutable.List.foreach(List.scala:318)
at scala.collection.TraversableLike$class.flatMap(TraversableLike.scala:251)
at scala.collection.AbstractTraversable.flatMap(Traversable.scala:105)
at org.json4s.MonadicJValue.org$json4s$MonadicJValue$$findDirectByName(MonadicJValue.scala:22)
at org.json4s.MonadicJValue.$bslash(MonadicJValue.scala:16)
at org.apache.spark.util.JsonProtocol$.taskInfoFromJson(JsonProtocol.scala:560)
at org.apache.spark.util.JsonProtocol$.taskEndFromJson(JsonProtocol.scala:465)
at org.apache.spark.util.JsonProtocol$.sparkEventFromJson(JsonProtocol.scala:425)
at org.apache.spark.scheduler.ReplayListenerBus$$anonfun$replay$2$$anonfun$apply$1.apply(ReplayListenerBus.scala:71)
at org.apache.spark.scheduler.ReplayListenerBus$$anonfun$replay$2$$anonfun$apply$1.apply(ReplayListenerBus.scala:69)
at scala.collection.Iterator$class.foreach(Iterator.scala:727)
at scala.collection.AbstractIterator.foreach(Iterator.scala:1157)
at org.apache.spark.scheduler.ReplayListenerBus$$anonfun$replay$2.apply(ReplayListenerBus.scala:69)
at org.apache.spark.scheduler.ReplayListenerBus$$anonfun$replay$2.apply(ReplayListenerBus.scala:55)
at scala.collection.IndexedSeqOptimized$class.foreach(IndexedSeqOptimized.scala:33)
at scala.collection.mutable.WrappedArray.foreach(WrappedArray.scala:34)
at org.apache.spark.scheduler.ReplayListenerBus.replay(ReplayListenerBus.scala:55)
at org.apache.spark.deploy.history.FsHistoryProvider$$anonfun$getAppUI$1.apply(FsHistoryProvider.scala:128)
at org.apache.spark.deploy.history.FsHistoryProvider$$anonfun$getAppUI$1.apply(FsHistoryProvider.scala:117)
at scala.Option.map(Option.scala:145)
at org.apache.spark.deploy.history.FsHistoryProvider.getAppUI(FsHistoryProvider.scala:117)
at org.apache.spark.deploy.history.HistoryServer$$anon$3.load(HistoryServer.scala:55)
at org.apache.spark.deploy.history.HistoryServer$$anon$3.load(HistoryServer.scala:53)
at org.spark-project.guava.common.cache.LocalCache$LoadingValueReference.loadFuture(LocalCache.java:3599)
at org.spark-project.guava.common.cache.LocalCache$Segment.loadSync(LocalCache.java:2379)
at org.spark-project.guava.common.cache.LocalCache$Segment.lockedGetOrLoad(LocalCache.java:2342)

我没有找到任何方法为 spark 历史服务器设置堆？ 还是这与 YARN 历史服务器有关？

谢谢

Answer 1

在 spark-env.sh 中设置 SPARK_DAEMON_MEMORY=2g 对我有帮助

Answer 2

我认为原因之一是日志文件比守护进程内存大，以至于历史服务器无法将整个日志读入堆中。

如果您没有指定日志目录，您可以在/tmp/spark-events 中查看日志文件的大小。否则，进入spark.eventLog.dir的指定文件夹。

就我而言，我的日志文件大约为 17g，因此我需要在 {SPARK_HOME}/conf/spark.env.sh 中设置 SPARK_DAEMON_MEMORY=20g。