From f0eed6e30467d8db640303848f41053c433949a6 Mon Sep 17 00:00:00 2001 From: Tal Sliwowicz Date: Wed, 17 Dec 2014 14:14:22 -0800 Subject: [PATCH] [SPARK-4006] In long running contexts, we encountered the situation of d... ...ouble registe... ...r without a remove in between. The cause for that is unknown, and assumed a temp network issue. However, since the second register is with a BlockManagerId on a different port, blockManagerInfo.contains() returns false, while blockManagerIdByExecutor returns Some. This inconsistency is caught in a conditional statement that does System.exit(1), which is a huge robustness issue for us. The fix - simply remove the old id from both maps during register when this happens. We are mimicking the behavior of expireDeadHosts(), by doing local cleanup of the maps before trying to add new ones. Also - added some logging for register and unregister. This is just like https://github.com/apache/spark/pull/2886 except it's on branch-1.0 Author: Tal Sliwowicz Closes #2914 from tsliwowicz/branch-1.0-block-mgr-removal and squashes the following commits: 1014493 [Tal Sliwowicz] [SPARK-4006] In long running contexts, we encountered the situation of double registe... --- .../storage/BlockManagerMasterActor.scala | 21 ++++++++++++------- 1 file changed, 13 insertions(+), 8 deletions(-) diff --git a/core/src/main/scala/org/apache/spark/storage/BlockManagerMasterActor.scala b/core/src/main/scala/org/apache/spark/storage/BlockManagerMasterActor.scala index 6aed322eeb185..7b908580b7872 100644 --- a/core/src/main/scala/org/apache/spark/storage/BlockManagerMasterActor.scala +++ b/core/src/main/scala/org/apache/spark/storage/BlockManagerMasterActor.scala @@ -208,6 +208,7 @@ class BlockManagerMasterActor(val isLocal: Boolean, conf: SparkConf, listenerBus } } listenerBus.post(SparkListenerBlockManagerRemoved(blockManagerId)) + logInfo(s"Removing block manager $blockManagerId") } private def expireDeadHosts() { @@ -328,16 +329,20 @@ class BlockManagerMasterActor(val isLocal: Boolean, conf: SparkConf, listenerBus private def register(id: BlockManagerId, maxMemSize: Long, slaveActor: ActorRef) { if (!blockManagerInfo.contains(id)) { blockManagerIdByExecutor.get(id.executorId) match { - case Some(manager) => - // A block manager of the same executor already exists. - // This should never happen. Let's just quit. - logError("Got two different block manager registrations on " + id.executorId) - System.exit(1) + case Some(oldId) => + // A block manager of the same executor already exists, so remove it (assumed dead) + logError("Got two different block manager registrations on same executor - " + + s" will replace old one $oldId with new one $id") + removeExecutor(id.executorId) case None => - blockManagerIdByExecutor(id.executorId) = id } - blockManagerInfo(id) = - new BlockManagerInfo(id, System.currentTimeMillis(), maxMemSize, slaveActor) + logInfo("Registering block manager %s with %s RAM, %s".format( + id.hostPort, Utils.bytesToString(maxMemSize), id)) + + blockManagerIdByExecutor(id.executorId) = id + + blockManagerInfo(id) = new BlockManagerInfo( + id, System.currentTimeMillis(), maxMemSize, slaveActor) } listenerBus.post(SparkListenerBlockManagerAdded(id, maxMemSize)) }