diff options
author | Sandy Ryza <sandy@cloudera.com> | 2014-12-09 11:02:43 -0800 |
---|---|---|
committer | Andrew Or <andrew@databricks.com> | 2014-12-09 11:02:43 -0800 |
commit | 912563aa3553afc0871d5b5858f533aa39cb99e5 (patch) | |
tree | 092241ac4c78deef8053f5095bc9680b3c8532cd /yarn | |
parent | 383c5555c9f26c080bc9e3a463aab21dd5b3797f (diff) | |
download | spark-912563aa3553afc0871d5b5858f533aa39cb99e5.tar.gz spark-912563aa3553afc0871d5b5858f533aa39cb99e5.tar.bz2 spark-912563aa3553afc0871d5b5858f533aa39cb99e5.zip |
SPARK-4338. [YARN] Ditch yarn-alpha.
Sorry if this is a little premature with 1.2 still not out the door, but it will make other work like SPARK-4136 and SPARK-2089 a lot easier.
Author: Sandy Ryza <sandy@cloudera.com>
Closes #3215 from sryza/sandy-spark-4338 and squashes the following commits:
1c5ac08 [Sandy Ryza] Update building Spark docs and remove unnecessary newline
9c1421c [Sandy Ryza] SPARK-4338. Ditch yarn-alpha.
Diffstat (limited to 'yarn')
-rw-r--r-- | yarn/README.md | 12 | ||||
-rw-r--r-- | yarn/alpha/pom.xml | 35 | ||||
-rw-r--r-- | yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/Client.scala | 145 | ||||
-rw-r--r-- | yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala | 139 | ||||
-rw-r--r-- | yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocationHandler.scala | 229 | ||||
-rw-r--r-- | yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClientImpl.scala | 118 | ||||
-rw-r--r-- | yarn/pom.xml | 129 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/deploy/yarn/ApplicationMaster.scala (renamed from yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ApplicationMaster.scala) | 0 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/deploy/yarn/ApplicationMasterArguments.scala (renamed from yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ApplicationMasterArguments.scala) | 0 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/deploy/yarn/Client.scala (renamed from yarn/stable/src/main/scala/org/apache/spark/deploy/yarn/Client.scala) | 0 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/deploy/yarn/ClientArguments.scala (renamed from yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientArguments.scala) | 36 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/deploy/yarn/ClientBase.scala (renamed from yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientBase.scala) | 0 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/deploy/yarn/ClientDistributedCacheManager.scala (renamed from yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientDistributedCacheManager.scala) | 0 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala (renamed from yarn/stable/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala) | 0 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnableUtil.scala (renamed from yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnableUtil.scala) | 7 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocationHandler.scala (renamed from yarn/stable/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocationHandler.scala) | 0 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocator.scala (renamed from yarn/common/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocator.scala) | 0 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClient.scala (renamed from yarn/common/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClient.scala) | 0 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClientImpl.scala (renamed from yarn/stable/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClientImpl.scala) | 0 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnSparkHadoopUtil.scala (renamed from yarn/common/src/main/scala/org/apache/spark/deploy/yarn/YarnSparkHadoopUtil.scala) | 10 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/scheduler/cluster/YarnClientClusterScheduler.scala (renamed from yarn/common/src/main/scala/org/apache/spark/scheduler/cluster/YarnClientClusterScheduler.scala) | 0 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/scheduler/cluster/YarnClientSchedulerBackend.scala (renamed from yarn/common/src/main/scala/org/apache/spark/scheduler/cluster/YarnClientSchedulerBackend.scala) | 0 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/scheduler/cluster/YarnClusterScheduler.scala (renamed from yarn/common/src/main/scala/org/apache/spark/scheduler/cluster/YarnClusterScheduler.scala) | 0 | ||||
-rw-r--r-- | yarn/src/main/scala/org/apache/spark/scheduler/cluster/YarnClusterSchedulerBackend.scala (renamed from yarn/common/src/main/scala/org/apache/spark/scheduler/cluster/YarnClusterSchedulerBackend.scala) | 0 | ||||
-rw-r--r-- | yarn/src/test/resources/log4j.properties (renamed from yarn/stable/src/test/resources/log4j.properties) | 0 | ||||
-rw-r--r-- | yarn/src/test/scala/org/apache/spark/deploy/yarn/ClientBaseSuite.scala (renamed from yarn/common/src/test/scala/org/apache/spark/deploy/yarn/ClientBaseSuite.scala) | 0 | ||||
-rw-r--r-- | yarn/src/test/scala/org/apache/spark/deploy/yarn/ClientDistributedCacheManagerSuite.scala (renamed from yarn/common/src/test/scala/org/apache/spark/deploy/yarn/ClientDistributedCacheManagerSuite.scala) | 0 | ||||
-rw-r--r-- | yarn/src/test/scala/org/apache/spark/deploy/yarn/YarnAllocatorSuite.scala (renamed from yarn/common/src/test/scala/org/apache/spark/deploy/yarn/YarnAllocatorSuite.scala) | 0 | ||||
-rw-r--r-- | yarn/src/test/scala/org/apache/spark/deploy/yarn/YarnClusterSuite.scala (renamed from yarn/stable/src/test/scala/org/apache/spark/deploy/yarn/YarnClusterSuite.scala) | 0 | ||||
-rw-r--r-- | yarn/src/test/scala/org/apache/spark/deploy/yarn/YarnSparkHadoopUtilSuite.scala (renamed from yarn/common/src/test/scala/org/apache/spark/deploy/yarn/YarnSparkHadoopUtilSuite.scala) | 0 | ||||
-rw-r--r-- | yarn/stable/pom.xml | 95 |
31 files changed, 85 insertions, 870 deletions
diff --git a/yarn/README.md b/yarn/README.md deleted file mode 100644 index 65ee85447e..0000000000 --- a/yarn/README.md +++ /dev/null @@ -1,12 +0,0 @@ -# YARN DIRECTORY LAYOUT - -Hadoop Yarn related codes are organized in separate directories to minimize duplicated code. - - * common : Common codes that do not depending on specific version of Hadoop. - - * alpha / stable : Codes that involve specific version of Hadoop YARN API. - - alpha represents 0.23 and 2.0.x - stable represents 2.2 and later, until the API changes again. - -alpha / stable will build together with common dir into a single jar diff --git a/yarn/alpha/pom.xml b/yarn/alpha/pom.xml deleted file mode 100644 index 40e9e99c6f..0000000000 --- a/yarn/alpha/pom.xml +++ /dev/null @@ -1,35 +0,0 @@ -<?xml version="1.0" encoding="UTF-8"?> -<!-- - ~ Licensed to the Apache Software Foundation (ASF) under one or more - ~ contributor license agreements. See the NOTICE file distributed with - ~ this work for additional information regarding copyright ownership. - ~ The ASF licenses this file to You under the Apache License, Version 2.0 - ~ (the "License"); you may not use this file except in compliance with - ~ the License. You may obtain a copy of the License at - ~ - ~ http://www.apache.org/licenses/LICENSE-2.0 - ~ - ~ Unless required by applicable law or agreed to in writing, software - ~ distributed under the License is distributed on an "AS IS" BASIS, - ~ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - ~ See the License for the specific language governing permissions and - ~ limitations under the License. - --> -<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd"> - <modelVersion>4.0.0</modelVersion> - <parent> - <groupId>org.apache.spark</groupId> - <artifactId>yarn-parent_2.10</artifactId> - <version>1.3.0-SNAPSHOT</version> - <relativePath>../pom.xml</relativePath> - </parent> - <properties> - <sbt.project.name>yarn-alpha</sbt.project.name> - </properties> - - <groupId>org.apache.spark</groupId> - <artifactId>spark-yarn-alpha_2.10</artifactId> - <packaging>jar</packaging> - <name>Spark Project YARN Alpha API</name> - -</project> diff --git a/yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/Client.scala b/yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/Client.scala deleted file mode 100644 index 73b705ba50..0000000000 --- a/yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/Client.scala +++ /dev/null @@ -1,145 +0,0 @@ -/* - * Licensed to the Apache Software Foundation (ASF) under one or more - * contributor license agreements. See the NOTICE file distributed with - * this work for additional information regarding copyright ownership. - * The ASF licenses this file to You under the Apache License, Version 2.0 - * (the "License"); you may not use this file except in compliance with - * the License. You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package org.apache.spark.deploy.yarn - -import java.nio.ByteBuffer - -import org.apache.hadoop.conf.Configuration -import org.apache.hadoop.io.DataOutputBuffer -import org.apache.hadoop.security.UserGroupInformation -import org.apache.hadoop.yarn.api._ -import org.apache.hadoop.yarn.api.protocolrecords._ -import org.apache.hadoop.yarn.api.records._ -import org.apache.hadoop.yarn.client.YarnClientImpl -import org.apache.hadoop.yarn.conf.YarnConfiguration -import org.apache.hadoop.yarn.util.Records - -import org.apache.spark.{Logging, SparkConf} -import org.apache.spark.deploy.SparkHadoopUtil - -/** - * Version of [[org.apache.spark.deploy.yarn.ClientBase]] tailored to YARN's alpha API. - */ -@deprecated("use yarn/stable", "1.2.0") -private[spark] class Client( - val args: ClientArguments, - val hadoopConf: Configuration, - val sparkConf: SparkConf) - extends YarnClientImpl with ClientBase with Logging { - - def this(clientArgs: ClientArguments, spConf: SparkConf) = - this(clientArgs, SparkHadoopUtil.get.newConfiguration(spConf), spConf) - - def this(clientArgs: ClientArguments) = this(clientArgs, new SparkConf()) - - val yarnConf: YarnConfiguration = new YarnConfiguration(hadoopConf) - - /* ------------------------------------------------------------------------------------- * - | The following methods have much in common in the stable and alpha versions of Client, | - | but cannot be implemented in the parent trait due to subtle API differences across | - | hadoop versions. | - * ------------------------------------------------------------------------------------- */ - - /** Submit an application running our ApplicationMaster to the ResourceManager. */ - override def submitApplication(): ApplicationId = { - init(yarnConf) - start() - - logInfo("Requesting a new application from cluster with %d NodeManagers" - .format(getYarnClusterMetrics.getNumNodeManagers)) - - // Get a new application from our RM - val newAppResponse = getNewApplication() - val appId = newAppResponse.getApplicationId() - - // Verify whether the cluster has enough resources for our AM - verifyClusterResources(newAppResponse) - - // Set up the appropriate contexts to launch our AM - val containerContext = createContainerLaunchContext(newAppResponse) - val appContext = createApplicationSubmissionContext(appId, containerContext) - - // Finally, submit and monitor the application - logInfo(s"Submitting application ${appId.getId} to ResourceManager") - submitApplication(appContext) - appId - } - - /** - * Set up a context for launching our ApplicationMaster container. - * In the Yarn alpha API, the memory requirements of this container must be set in - * the ContainerLaunchContext instead of the ApplicationSubmissionContext. - */ - override def createContainerLaunchContext(newAppResponse: GetNewApplicationResponse) - : ContainerLaunchContext = { - val containerContext = super.createContainerLaunchContext(newAppResponse) - val capability = Records.newRecord(classOf[Resource]) - capability.setMemory(args.amMemory + amMemoryOverhead) - containerContext.setResource(capability) - containerContext - } - - /** Set up the context for submitting our ApplicationMaster. */ - def createApplicationSubmissionContext( - appId: ApplicationId, - containerContext: ContainerLaunchContext): ApplicationSubmissionContext = { - val appContext = Records.newRecord(classOf[ApplicationSubmissionContext]) - appContext.setApplicationId(appId) - appContext.setApplicationName(args.appName) - appContext.setQueue(args.amQueue) - appContext.setAMContainerSpec(containerContext) - appContext.setUser(UserGroupInformation.getCurrentUser.getShortUserName) - appContext - } - - /** - * Set up security tokens for launching our ApplicationMaster container. - * ContainerLaunchContext#setContainerTokens is renamed `setTokens` in the stable API. - */ - override def setupSecurityToken(amContainer: ContainerLaunchContext): Unit = { - val dob = new DataOutputBuffer() - credentials.writeTokenStorageToStream(dob) - amContainer.setContainerTokens(ByteBuffer.wrap(dob.getData())) - } - - /** - * Return the security token used by this client to communicate with the ApplicationMaster. - * If no security is enabled, the token returned by the report is null. - * ApplicationReport#getClientToken is renamed `getClientToAMToken` in the stable API. - */ - override def getClientToken(report: ApplicationReport): String = - Option(report.getClientToken).map(_.toString).getOrElse("") -} - -object Client { - def main(argStrings: Array[String]) { - if (!sys.props.contains("SPARK_SUBMIT")) { - println("WARNING: This client is deprecated and will be removed in a " + - "future version of Spark. Use ./bin/spark-submit with \"--master yarn\"") - } - println("WARNING: Support for YARN-alpha API's will be removed in Spark 1.3 (see SPARK-3445)") - - // Set an env variable indicating we are running in YARN mode. - // Note that any env variable with the SPARK_ prefix gets propagated to all (remote) processes - System.setProperty("SPARK_YARN_MODE", "true") - val sparkConf = new SparkConf - - val args = new ClientArguments(argStrings, sparkConf) - new Client(args, sparkConf).run() - } -} diff --git a/yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala b/yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala deleted file mode 100644 index 7023a11706..0000000000 --- a/yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala +++ /dev/null @@ -1,139 +0,0 @@ -/* - * Licensed to the Apache Software Foundation (ASF) under one or more - * contributor license agreements. See the NOTICE file distributed with - * this work for additional information regarding copyright ownership. - * The ASF licenses this file to You under the Apache License, Version 2.0 - * (the "License"); you may not use this file except in compliance with - * the License. You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package org.apache.spark.deploy.yarn - -import java.net.URI -import java.nio.ByteBuffer -import java.security.PrivilegedExceptionAction - -import scala.collection.JavaConversions._ -import scala.collection.mutable.HashMap - -import org.apache.hadoop.conf.Configuration -import org.apache.hadoop.io.DataOutputBuffer -import org.apache.hadoop.net.NetUtils -import org.apache.hadoop.security.UserGroupInformation -import org.apache.hadoop.yarn.api._ -import org.apache.hadoop.yarn.api.records._ -import org.apache.hadoop.yarn.api.protocolrecords._ -import org.apache.hadoop.yarn.conf.YarnConfiguration -import org.apache.hadoop.yarn.ipc.YarnRPC -import org.apache.hadoop.yarn.util.{Apps, ConverterUtils, Records, ProtoUtils} - -import org.apache.spark.{SecurityManager, SparkConf, Logging} -import org.apache.spark.network.util.JavaUtils - -@deprecated("use yarn/stable", "1.2.0") -class ExecutorRunnable( - container: Container, - conf: Configuration, - spConf: SparkConf, - masterAddress: String, - slaveId: String, - hostname: String, - executorMemory: Int, - executorCores: Int, - appAttemptId: String, - securityMgr: SecurityManager) - extends Runnable with ExecutorRunnableUtil with Logging { - - var rpc: YarnRPC = YarnRPC.create(conf) - var cm: ContainerManager = _ - val sparkConf = spConf - val yarnConf: YarnConfiguration = new YarnConfiguration(conf) - - def run = { - logInfo("Starting Executor Container") - cm = connectToCM - startContainer - } - - def startContainer = { - logInfo("Setting up ContainerLaunchContext") - - val ctx = Records.newRecord(classOf[ContainerLaunchContext]) - .asInstanceOf[ContainerLaunchContext] - - ctx.setContainerId(container.getId()) - ctx.setResource(container.getResource()) - val localResources = prepareLocalResources - ctx.setLocalResources(localResources) - - val env = prepareEnvironment - ctx.setEnvironment(env) - - ctx.setUser(UserGroupInformation.getCurrentUser().getShortUserName()) - - val credentials = UserGroupInformation.getCurrentUser().getCredentials() - val dob = new DataOutputBuffer() - credentials.writeTokenStorageToStream(dob) - ctx.setContainerTokens(ByteBuffer.wrap(dob.getData())) - - val commands = prepareCommand(masterAddress, slaveId, hostname, executorMemory, executorCores, - appAttemptId, localResources) - logInfo("Setting up executor with commands: " + commands) - ctx.setCommands(commands) - - ctx.setApplicationACLs(YarnSparkHadoopUtil.getApplicationAclsForYarn(securityMgr)) - - // If external shuffle service is enabled, register with the Yarn shuffle service already - // started on the NodeManager and, if authentication is enabled, provide it with our secret - // key for fetching shuffle files later - if (sparkConf.getBoolean("spark.shuffle.service.enabled", false)) { - val secretString = securityMgr.getSecretKey() - val secretBytes = - if (secretString != null) { - // This conversion must match how the YarnShuffleService decodes our secret - JavaUtils.stringToBytes(secretString) - } else { - // Authentication is not enabled, so just provide dummy metadata - ByteBuffer.allocate(0) - } - ctx.setServiceData(Map[String, ByteBuffer]("spark_shuffle" -> secretBytes)) - } - - // Send the start request to the ContainerManager - val startReq = Records.newRecord(classOf[StartContainerRequest]) - .asInstanceOf[StartContainerRequest] - startReq.setContainerLaunchContext(ctx) - cm.startContainer(startReq) - } - - def connectToCM: ContainerManager = { - val cmHostPortStr = container.getNodeId().getHost() + ":" + container.getNodeId().getPort() - val cmAddress = NetUtils.createSocketAddr(cmHostPortStr) - logInfo("Connecting to ContainerManager at " + cmHostPortStr) - - // Use doAs and remoteUser here so we can add the container token and not pollute the current - // users credentials with all of the individual container tokens - val user = UserGroupInformation.createRemoteUser(container.getId().toString()) - val containerToken = container.getContainerToken() - if (containerToken != null) { - user.addToken(ProtoUtils.convertFromProtoFormat(containerToken, cmAddress)) - } - - val proxy = user - .doAs(new PrivilegedExceptionAction[ContainerManager] { - def run: ContainerManager = { - rpc.getProxy(classOf[ContainerManager], cmAddress, conf).asInstanceOf[ContainerManager] - } - }) - proxy - } - -} diff --git a/yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocationHandler.scala b/yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocationHandler.scala deleted file mode 100644 index abd37834ed..0000000000 --- a/yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocationHandler.scala +++ /dev/null @@ -1,229 +0,0 @@ -/* - * Licensed to the Apache Software Foundation (ASF) under one or more - * contributor license agreements. See the NOTICE file distributed with - * this work for additional information regarding copyright ownership. - * The ASF licenses this file to You under the Apache License, Version 2.0 - * (the "License"); you may not use this file except in compliance with - * the License. You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package org.apache.spark.deploy.yarn - -import java.util.concurrent.CopyOnWriteArrayList -import java.util.concurrent.atomic.AtomicInteger - -import scala.collection.JavaConversions._ -import scala.collection.mutable.{ArrayBuffer, HashMap} - -import org.apache.spark.{SecurityManager, SparkConf} -import org.apache.spark.scheduler.SplitInfo - -import org.apache.hadoop.conf.Configuration -import org.apache.hadoop.yarn.api.AMRMProtocol -import org.apache.hadoop.yarn.api.records._ -import org.apache.hadoop.yarn.api.protocolrecords.AllocateRequest -import org.apache.hadoop.yarn.util.Records - -/** - * Acquires resources for executors from a ResourceManager and launches executors in new containers. - */ -private[yarn] class YarnAllocationHandler( - conf: Configuration, - sparkConf: SparkConf, - resourceManager: AMRMProtocol, - appAttemptId: ApplicationAttemptId, - args: ApplicationMasterArguments, - preferredNodes: collection.Map[String, collection.Set[SplitInfo]], - securityMgr: SecurityManager) - extends YarnAllocator(conf, sparkConf, appAttemptId, args, preferredNodes, securityMgr) { - - private val lastResponseId = new AtomicInteger() - private val releaseList: CopyOnWriteArrayList[ContainerId] = new CopyOnWriteArrayList() - - override protected def allocateContainers(count: Int, pending: Int): YarnAllocateResponse = { - var resourceRequests: List[ResourceRequest] = null - - logDebug("asking for additional executors: " + count + " with already pending: " + pending) - val totalNumAsk = count + pending - if (count <= 0) { - resourceRequests = List() - } else if (preferredHostToCount.isEmpty) { - logDebug("host preferences is empty") - resourceRequests = List(createResourceRequest( - AllocationType.ANY, null, totalNumAsk, YarnSparkHadoopUtil.RM_REQUEST_PRIORITY)) - } else { - // request for all hosts in preferred nodes and for numExecutors - - // candidates.size, request by default allocation policy. - val hostContainerRequests: ArrayBuffer[ResourceRequest] = - new ArrayBuffer[ResourceRequest](preferredHostToCount.size) - for ((candidateHost, candidateCount) <- preferredHostToCount) { - val requiredCount = candidateCount - allocatedContainersOnHost(candidateHost) - - if (requiredCount > 0) { - hostContainerRequests += createResourceRequest( - AllocationType.HOST, - candidateHost, - requiredCount, - YarnSparkHadoopUtil.RM_REQUEST_PRIORITY) - } - } - val rackContainerRequests: List[ResourceRequest] = createRackResourceRequests( - hostContainerRequests.toList) - - val anyContainerRequests: ResourceRequest = createResourceRequest( - AllocationType.ANY, - resource = null, - totalNumAsk, - YarnSparkHadoopUtil.RM_REQUEST_PRIORITY) - - val containerRequests: ArrayBuffer[ResourceRequest] = new ArrayBuffer[ResourceRequest]( - hostContainerRequests.size + rackContainerRequests.size + 1) - - containerRequests ++= hostContainerRequests - containerRequests ++= rackContainerRequests - containerRequests += anyContainerRequests - - resourceRequests = containerRequests.toList - } - - val req = Records.newRecord(classOf[AllocateRequest]) - req.setResponseId(lastResponseId.incrementAndGet) - req.setApplicationAttemptId(appAttemptId) - - req.addAllAsks(resourceRequests) - - val releasedContainerList = createReleasedContainerList() - req.addAllReleases(releasedContainerList) - - if (count > 0) { - logInfo("Allocating %d executor containers with %d of memory each.".format(totalNumAsk, - executorMemory + memoryOverhead)) - } else { - logDebug("Empty allocation req .. release : " + releasedContainerList) - } - - for (request <- resourceRequests) { - logInfo("ResourceRequest (host : %s, num containers: %d, priority = %s , capability : %s)". - format( - request.getHostName, - request.getNumContainers, - request.getPriority, - request.getCapability)) - } - new AlphaAllocateResponse(resourceManager.allocate(req).getAMResponse()) - } - - override protected def releaseContainer(container: Container) = { - releaseList.add(container.getId()) - } - - private def createRackResourceRequests(hostContainers: List[ResourceRequest]): - List[ResourceRequest] = { - // First generate modified racks and new set of hosts under it : then issue requests - val rackToCounts = new HashMap[String, Int]() - - // Within this lock - used to read/write to the rack related maps too. - for (container <- hostContainers) { - val candidateHost = container.getHostName - val candidateNumContainers = container.getNumContainers - assert(YarnSparkHadoopUtil.ANY_HOST != candidateHost) - - val rack = YarnSparkHadoopUtil.lookupRack(conf, candidateHost) - if (rack != null) { - var count = rackToCounts.getOrElse(rack, 0) - count += candidateNumContainers - rackToCounts.put(rack, count) - } - } - - val requestedContainers: ArrayBuffer[ResourceRequest] = - new ArrayBuffer[ResourceRequest](rackToCounts.size) - for ((rack, count) <- rackToCounts){ - requestedContainers += - createResourceRequest(AllocationType.RACK, rack, count, - YarnSparkHadoopUtil.RM_REQUEST_PRIORITY) - } - - requestedContainers.toList - } - - private def createResourceRequest( - requestType: AllocationType.AllocationType, - resource:String, - numExecutors: Int, - priority: Int): ResourceRequest = { - - // If hostname specified, we need atleast two requests - node local and rack local. - // There must be a third request - which is ANY : that will be specially handled. - requestType match { - case AllocationType.HOST => { - assert(YarnSparkHadoopUtil.ANY_HOST != resource) - val hostname = resource - val nodeLocal = createResourceRequestImpl(hostname, numExecutors, priority) - - // Add to host->rack mapping - YarnSparkHadoopUtil.populateRackInfo(conf, hostname) - - nodeLocal - } - case AllocationType.RACK => { - val rack = resource - createResourceRequestImpl(rack, numExecutors, priority) - } - case AllocationType.ANY => createResourceRequestImpl( - YarnSparkHadoopUtil.ANY_HOST, numExecutors, priority) - case _ => throw new IllegalArgumentException( - "Unexpected/unsupported request type: " + requestType) - } - } - - private def createResourceRequestImpl( - hostname:String, - numExecutors: Int, - priority: Int): ResourceRequest = { - - val rsrcRequest = Records.newRecord(classOf[ResourceRequest]) - val memCapability = Records.newRecord(classOf[Resource]) - // There probably is some overhead here, let's reserve a bit more memory. - memCapability.setMemory(executorMemory + memoryOverhead) - rsrcRequest.setCapability(memCapability) - - val pri = Records.newRecord(classOf[Priority]) - pri.setPriority(priority) - rsrcRequest.setPriority(pri) - - rsrcRequest.setHostName(hostname) - - rsrcRequest.setNumContainers(java.lang.Math.max(numExecutors, 0)) - rsrcRequest - } - - private def createReleasedContainerList(): ArrayBuffer[ContainerId] = { - val retval = new ArrayBuffer[ContainerId](1) - // Iterator on COW list ... - for (container <- releaseList.iterator()){ - retval += container - } - // Remove from the original list. - if (!retval.isEmpty) { - releaseList.removeAll(retval) - logInfo("Releasing " + retval.size + " containers.") - } - retval - } - - private class AlphaAllocateResponse(response: AMResponse) extends YarnAllocateResponse { - override def getAllocatedContainers() = response.getAllocatedContainers() - override def getAvailableResources() = response.getAvailableResources() - override def getCompletedContainersStatuses() = response.getCompletedContainersStatuses() - } - -} diff --git a/yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClientImpl.scala b/yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClientImpl.scala deleted file mode 100644 index e342cc82f4..0000000000 --- a/yarn/alpha/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClientImpl.scala +++ /dev/null @@ -1,118 +0,0 @@ -/* - * Licensed to the Apache Software Foundation (ASF) under one or more - * contributor license agreements. See the NOTICE file distributed with - * this work for additional information regarding copyright ownership. - * The ASF licenses this file to You under the Apache License, Version 2.0 - * (the "License"); you may not use this file except in compliance with - * the License. You may obtain a copy of the License at - * - * http://www.apache.org/licenses/LICENSE-2.0 - * - * Unless required by applicable law or agreed to in writing, software - * distributed under the License is distributed on an "AS IS" BASIS, - * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - * See the License for the specific language governing permissions and - * limitations under the License. - */ - -package org.apache.spark.deploy.yarn - -import scala.collection.{Map, Set} -import java.net.URI - -import org.apache.hadoop.net.NetUtils -import org.apache.hadoop.yarn.api._ -import org.apache.hadoop.yarn.api.records._ -import org.apache.hadoop.yarn.api.protocolrecords._ -import org.apache.hadoop.yarn.conf.YarnConfiguration -import org.apache.hadoop.yarn.ipc.YarnRPC -import org.apache.hadoop.yarn.util.{ConverterUtils, Records} - -import org.apache.spark.{Logging, SecurityManager, SparkConf} -import org.apache.spark.scheduler.SplitInfo -import org.apache.spark.util.Utils - -/** - * YarnRMClient implementation for the Yarn alpha API. - */ -private class YarnRMClientImpl(args: ApplicationMasterArguments) extends YarnRMClient with Logging { - - private var rpc: YarnRPC = null - private var resourceManager: AMRMProtocol = _ - private var uiHistoryAddress: String = _ - private var registered: Boolean = false - - override def register( - conf: YarnConfiguration, - sparkConf: SparkConf, - preferredNodeLocations: Map[String, Set[SplitInfo]], - uiAddress: String, - uiHistoryAddress: String, - securityMgr: SecurityManager) = { - this.rpc = YarnRPC.create(conf) - this.uiHistoryAddress = uiHistoryAddress - - synchronized { - resourceManager = registerWithResourceManager(conf) - registerApplicationMaster(uiAddress) - registered = true - } - - new YarnAllocationHandler(conf, sparkConf, resourceManager, getAttemptId(), args, - preferredNodeLocations, securityMgr) - } - - override def getAttemptId() = { - val envs = System.getenv() - val containerIdString = envs.get(ApplicationConstants.AM_CONTAINER_ID_ENV) - val containerId = ConverterUtils.toContainerId(containerIdString) - val appAttemptId = containerId.getApplicationAttemptId() - appAttemptId - } - - override def unregister(status: FinalApplicationStatus, diagnostics: String = "") = synchronized { - if (registered) { - val finishReq = Records.newRecord(classOf[FinishApplicationMasterRequest]) - .asInstanceOf[FinishApplicationMasterRequest] - finishReq.setAppAttemptId(getAttemptId()) - finishReq.setFinishApplicationStatus(status) - finishReq.setDiagnostics(diagnostics) - finishReq.setTrackingUrl(uiHistoryAddress) - resourceManager.finishApplicationMaster(finishReq) - } - } - - override def getAmIpFilterParams(conf: YarnConfiguration, proxyBase: String) = { - val proxy = YarnConfiguration.getProxyHostAndPort(conf) - val parts = proxy.split(":") - val uriBase = "http://" + proxy + proxyBase - Map("PROXY_HOST" -> parts(0), "PROXY_URI_BASE" -> uriBase) - } - - override def getMaxRegAttempts(conf: YarnConfiguration) = - conf.getInt(YarnConfiguration.RM_AM_MAX_RETRIES, YarnConfiguration.DEFAULT_RM_AM_MAX_RETRIES) - - private def registerWithResourceManager(conf: YarnConfiguration): AMRMProtocol = { - val rmAddress = NetUtils.createSocketAddr(conf.get(YarnConfiguration.RM_SCHEDULER_ADDRESS, - YarnConfiguration.DEFAULT_RM_SCHEDULER_ADDRESS)) - logInfo("Connecting to ResourceManager at " + rmAddress) - rpc.getProxy(classOf[AMRMProtocol], rmAddress, conf).asInstanceOf[AMRMProtocol] - } - - private def registerApplicationMaster(uiAddress: String): RegisterApplicationMasterResponse = { - val appMasterRequest = Records.newRecord(classOf[RegisterApplicationMasterRequest]) - .asInstanceOf[RegisterApplicationMasterRequest] - appMasterRequest.setApplicationAttemptId(getAttemptId()) - // Setting this to master host,port - so that the ApplicationReport at client has some - // sensible info. - // Users can then monitor stderr/stdout on that node if required. - appMasterRequest.setHost(Utils.localHostName()) - appMasterRequest.setRpcPort(0) - // remove the scheme from the url if it exists since Hadoop does not expect scheme - val uri = new URI(uiAddress) - val authority = if (uri.getScheme == null) uiAddress else uri.getAuthority - appMasterRequest.setTrackingUrl(authority) - resourceManager.registerApplicationMaster(appMasterRequest) - } - -} diff --git a/yarn/pom.xml b/yarn/pom.xml index bba73648c7..d7579bf962 100644 --- a/yarn/pom.xml +++ b/yarn/pom.xml @@ -25,9 +25,9 @@ </parent> <groupId>org.apache.spark</groupId> - <artifactId>yarn-parent_2.10</artifactId> - <packaging>pom</packaging> - <name>Spark Project YARN Parent POM</name> + <artifactId>spark-yarn_2.10</artifactId> + <packaging>jar</packaging> + <name>Spark Project YARN</name> <properties> <sbt.project.name>yarn</sbt.project.name> </properties> @@ -59,6 +59,12 @@ <artifactId>hadoop-client</artifactId> </dependency> <dependency> + <groupId>org.apache.hadoop</groupId> + <artifactId>hadoop-yarn-server-tests</artifactId> + <classifier>tests</classifier> + <scope>test</scope> + </dependency> + <dependency> <groupId>org.scalatest</groupId> <artifactId>scalatest_${scala.binary.version}</artifactId> <scope>test</scope> @@ -70,41 +76,54 @@ </dependency> </dependencies> + <!-- + See SPARK-3710. hadoop-yarn-server-tests in Hadoop 2.2 fails to pull some needed + dependencies, so they need to be added manually for the tests to work. + --> <profiles> <profile> - <id>yarn-alpha</id> - <build> - <plugins> - <plugin> - <artifactId>maven-antrun-plugin</artifactId> - <executions> - <execution> - <phase>validate</phase> - <goals> - <goal>run</goal> - </goals> - <configuration> - <tasks> - <echo>*******************************************************************************************</echo> - <echo>***WARNING***: Support for YARN-alpha API's will be removed in Spark 1.3 (see SPARK-3445).*</echo> - <echo>*******************************************************************************************</echo> - </tasks> - </configuration> - </execution> - </executions> - </plugin> - </plugins> - </build> - <modules> - <module>alpha</module> - </modules> - </profile> - - <profile> - <id>yarn</id> - <modules> - <module>stable</module> - </modules> + <id>hadoop-2.2</id> + <properties> + <jersey.version>1.9</jersey.version> + </properties> + <dependencies> + <dependency> + <groupId>org.mortbay.jetty</groupId> + <artifactId>jetty</artifactId> + <version>6.1.26</version> + <exclusions> + <exclusion> + <groupId>org.mortbay.jetty</groupId> + <artifactId>servlet-api</artifactId> + </exclusion> + </exclusions> + <scope>test</scope> + </dependency> + <dependency> + <groupId>com.sun.jersey</groupId> + <artifactId>jersey-core</artifactId> + <version>${jersey.version}</version> + <scope>test</scope> + </dependency> + <dependency> + <groupId>com.sun.jersey</groupId> + <artifactId>jersey-json</artifactId> + <version>${jersey.version}</version> + <scope>test</scope> + <exclusions> + <exclusion> + <groupId>stax</groupId> + <artifactId>stax-api</artifactId> + </exclusion> + </exclusions> + </dependency> + <dependency> + <groupId>com.sun.jersey</groupId> + <artifactId>jersey-server</artifactId> + <version>${jersey.version}</version> + <scope>test</scope> + </dependency> + </dependencies> </profile> </profiles> @@ -125,38 +144,6 @@ </configuration> </plugin> <plugin> - <groupId>org.codehaus.mojo</groupId> - <artifactId>build-helper-maven-plugin</artifactId> - <executions> - <execution> - <id>add-scala-sources</id> - <phase>generate-sources</phase> - <goals> - <goal>add-source</goal> - </goals> - <configuration> - <sources> - <source>src/main/scala</source> - <source>../common/src/main/scala</source> - </sources> - </configuration> - </execution> - <execution> - <id>add-scala-test-sources</id> - <phase>generate-test-sources</phase> - <goals> - <goal>add-test-source</goal> - </goals> - <configuration> - <sources> - <source>src/test/scala</source> - <source>../common/src/test/scala</source> - </sources> - </configuration> - </execution> - </executions> - </plugin> - <plugin> <groupId>org.scalatest</groupId> <artifactId>scalatest-maven-plugin</artifactId> <configuration> @@ -169,12 +156,6 @@ <outputDirectory>target/scala-${scala.binary.version}/classes</outputDirectory> <testOutputDirectory>target/scala-${scala.binary.version}/test-classes</testOutputDirectory> - - <resources> - <resource> - <directory>../common/src/main/resources</directory> - </resource> - </resources> </build> </project> diff --git a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ApplicationMaster.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ApplicationMaster.scala index 987b3373fb..987b3373fb 100644 --- a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ApplicationMaster.scala +++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ApplicationMaster.scala diff --git a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ApplicationMasterArguments.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ApplicationMasterArguments.scala index d76a63276d..d76a63276d 100644 --- a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ApplicationMasterArguments.scala +++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ApplicationMasterArguments.scala diff --git a/yarn/stable/src/main/scala/org/apache/spark/deploy/yarn/Client.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/Client.scala index addaddb711..addaddb711 100644 --- a/yarn/stable/src/main/scala/org/apache/spark/deploy/yarn/Client.scala +++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/Client.scala diff --git a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientArguments.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ClientArguments.scala index 4d859450ef..c439969510 100644 --- a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientArguments.scala +++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ClientArguments.scala @@ -178,21 +178,25 @@ private[spark] class ClientArguments(args: Array[String], sparkConf: SparkConf) private def getUsageMessage(unknownParam: List[String] = null): String = { val message = if (unknownParam != null) s"Unknown/unsupported param $unknownParam\n" else "" - message + - "Usage: org.apache.spark.deploy.yarn.Client [options] \n" + - "Options:\n" + - " --jar JAR_PATH Path to your application's JAR file (required in yarn-cluster mode)\n" + - " --class CLASS_NAME Name of your application's main class (required)\n" + - " --arg ARG Argument to be passed to your application's main class.\n" + - " Multiple invocations are possible, each will be passed in order.\n" + - " --num-executors NUM Number of executors to start (Default: 2)\n" + - " --executor-cores NUM Number of cores for the executors (Default: 1).\n" + - " --driver-memory MEM Memory for driver (e.g. 1000M, 2G) (Default: 512 Mb)\n" + - " --executor-memory MEM Memory per executor (e.g. 1000M, 2G) (Default: 1G)\n" + - " --name NAME The name of your application (Default: Spark)\n" + - " --queue QUEUE The hadoop queue to use for allocation requests (Default: 'default')\n" + - " --addJars jars Comma separated list of local jars that want SparkContext.addJar to work with.\n" + - " --files files Comma separated list of files to be distributed with the job.\n" + - " --archives archives Comma separated list of archives to be distributed with the job." + message + """ + |Usage: org.apache.spark.deploy.yarn.Client [options] + |Options: + | --jar JAR_PATH Path to your application's JAR file (required in yarn-cluster + | mode) + | --class CLASS_NAME Name of your application's main class (required) + | --arg ARG Argument to be passed to your application's main class. + | Multiple invocations are possible, each will be passed in order. + | --num-executors NUM Number of executors to start (Default: 2) + | --executor-cores NUM Number of cores for the executors (Default: 1). + | --driver-memory MEM Memory for driver (e.g. 1000M, 2G) (Default: 512 Mb) + | --executor-memory MEM Memory per executor (e.g. 1000M, 2G) (Default: 1G) + | --name NAME The name of your application (Default: Spark) + | --queue QUEUE The hadoop queue to use for allocation requests (Default: + | 'default') + | --addJars jars Comma separated list of local jars that want SparkContext.addJar + | to work with. + | --files files Comma separated list of files to be distributed with the job. + | --archives archives Comma separated list of archives to be distributed with the job. + """ } } diff --git a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientBase.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ClientBase.scala index f95d723791..f95d723791 100644 --- a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientBase.scala +++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ClientBase.scala diff --git a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientDistributedCacheManager.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ClientDistributedCacheManager.scala index c592ecfdfc..c592ecfdfc 100644 --- a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ClientDistributedCacheManager.scala +++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ClientDistributedCacheManager.scala diff --git a/yarn/stable/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala index fdd3c2300f..fdd3c2300f 100644 --- a/yarn/stable/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala +++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnable.scala diff --git a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnableUtil.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnableUtil.scala index 88dad0febd..22d73ecf6d 100644 --- a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnableUtil.scala +++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/ExecutorRunnableUtil.scala @@ -93,12 +93,13 @@ trait ExecutorRunnableUtil extends Logging { /* else { // If no java_opts specified, default to using -XX:+CMSIncrementalMode - // It might be possible that other modes/config is being done in spark.executor.extraJavaOptions, - // so we dont want to mess with it. + // It might be possible that other modes/config is being done in + // spark.executor.extraJavaOptions, so we dont want to mess with it. // In our expts, using (default) throughput collector has severe perf ramnifications in // multi-tennent machines // The options are based on - // http://www.oracle.com/technetwork/java/gc-tuning-5-138395.html#0.0.0.%20When%20to%20Use%20the%20Concurrent%20Low%20Pause%20Collector|outline + // http://www.oracle.com/technetwork/java/gc-tuning-5-138395.html#0.0.0.%20When%20to%20Use + // %20the%20Concurrent%20Low%20Pause%20Collector|outline javaOpts += " -XX:+UseConcMarkSweepGC " javaOpts += " -XX:+CMSIncrementalMode " javaOpts += " -XX:+CMSIncrementalPacing " diff --git a/yarn/stable/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocationHandler.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocationHandler.scala index 2bbf5d7db8..2bbf5d7db8 100644 --- a/yarn/stable/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocationHandler.scala +++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocationHandler.scala diff --git a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocator.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocator.scala index b32e15738f..b32e15738f 100644 --- a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocator.scala +++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnAllocator.scala diff --git a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClient.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClient.scala index 2510b9c9ce..2510b9c9ce 100644 --- a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClient.scala +++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClient.scala diff --git a/yarn/stable/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClientImpl.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClientImpl.scala index 8d4b96ed79..8d4b96ed79 100644 --- a/yarn/stable/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClientImpl.scala +++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnRMClientImpl.scala diff --git a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/YarnSparkHadoopUtil.scala b/yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnSparkHadoopUtil.scala index 7d453ecb79..d7cf904db1 100644 --- a/yarn/common/src/main/scala/org/apache/spark/deploy/yarn/YarnSparkHadoopUtil.scala +++ b/yarn/src/main/scala/org/apache/spark/deploy/yarn/YarnSparkHadoopUtil.scala @@ -48,15 +48,17 @@ class YarnSparkHadoopUtil extends SparkHadoopUtil { dest.addCredentials(source.getCredentials()) } - // Note that all params which start with SPARK are propagated all the way through, so if in yarn mode, this MUST be set to true. + // Note that all params which start with SPARK are propagated all the way through, so if in yarn + // mode, this MUST be set to true. override def isYarnMode(): Boolean = { true } - // Return an appropriate (subclass) of Configuration. Creating config can initializes some hadoop subsystems - // Always create a new config, dont reuse yarnConf. + // Return an appropriate (subclass) of Configuration. Creating a config initializes some Hadoop + // subsystems. Always create a new config, dont reuse yarnConf. override def newConfiguration(conf: SparkConf): Configuration = new YarnConfiguration(super.newConfiguration(conf)) - // add any user credentials to the job conf which are necessary for running on a secure Hadoop cluster + // Add any user credentials to the job conf which are necessary for running on a secure Hadoop + // cluster override def addCredentials(conf: JobConf) { val jobCreds = conf.getCredentials() jobCreds.mergeAll(UserGroupInformation.getCurrentUser().getCredentials()) diff --git a/yarn/common/src/main/scala/org/apache/spark/scheduler/cluster/YarnClientClusterScheduler.scala b/yarn/src/main/scala/org/apache/spark/scheduler/cluster/YarnClientClusterScheduler.scala index 254774a6b8..254774a6b8 100644 --- a/yarn/common/src/main/scala/org/apache/spark/scheduler/cluster/YarnClientClusterScheduler.scala +++ b/yarn/src/main/scala/org/apache/spark/scheduler/cluster/YarnClientClusterScheduler.scala diff --git a/yarn/common/src/main/scala/org/apache/spark/scheduler/cluster/YarnClientSchedulerBackend.scala b/yarn/src/main/scala/org/apache/spark/scheduler/cluster/YarnClientSchedulerBackend.scala index 2923e6729c..2923e6729c 100644 --- a/yarn/common/src/main/scala/org/apache/spark/scheduler/cluster/YarnClientSchedulerBackend.scala +++ b/yarn/src/main/scala/org/apache/spark/scheduler/cluster/YarnClientSchedulerBackend.scala diff --git a/yarn/common/src/main/scala/org/apache/spark/scheduler/cluster/YarnClusterScheduler.scala b/yarn/src/main/scala/org/apache/spark/scheduler/cluster/YarnClusterScheduler.scala index 4157ff95c2..4157ff95c2 100644 --- a/yarn/common/src/main/scala/org/apache/spark/scheduler/cluster/YarnClusterScheduler.scala +++ b/yarn/src/main/scala/org/apache/spark/scheduler/cluster/YarnClusterScheduler.scala diff --git a/yarn/common/src/main/scala/org/apache/spark/scheduler/cluster/YarnClusterSchedulerBackend.scala b/yarn/src/main/scala/org/apache/spark/scheduler/cluster/YarnClusterSchedulerBackend.scala index b1de81e6a8..b1de81e6a8 100644 --- a/yarn/common/src/main/scala/org/apache/spark/scheduler/cluster/YarnClusterSchedulerBackend.scala +++ b/yarn/src/main/scala/org/apache/spark/scheduler/cluster/YarnClusterSchedulerBackend.scala diff --git a/yarn/stable/src/test/resources/log4j.properties b/yarn/src/test/resources/log4j.properties index 9dd05f17f0..9dd05f17f0 100644 --- a/yarn/stable/src/test/resources/log4j.properties +++ b/yarn/src/test/resources/log4j.properties diff --git a/yarn/common/src/test/scala/org/apache/spark/deploy/yarn/ClientBaseSuite.scala b/yarn/src/test/scala/org/apache/spark/deploy/yarn/ClientBaseSuite.scala index 17b79ae1d8..17b79ae1d8 100644 --- a/yarn/common/src/test/scala/org/apache/spark/deploy/yarn/ClientBaseSuite.scala +++ b/yarn/src/test/scala/org/apache/spark/deploy/yarn/ClientBaseSuite.scala diff --git a/yarn/common/src/test/scala/org/apache/spark/deploy/yarn/ClientDistributedCacheManagerSuite.scala b/yarn/src/test/scala/org/apache/spark/deploy/yarn/ClientDistributedCacheManagerSuite.scala index 80b57d1355..80b57d1355 100644 --- a/yarn/common/src/test/scala/org/apache/spark/deploy/yarn/ClientDistributedCacheManagerSuite.scala +++ b/yarn/src/test/scala/org/apache/spark/deploy/yarn/ClientDistributedCacheManagerSuite.scala diff --git a/yarn/common/src/test/scala/org/apache/spark/deploy/yarn/YarnAllocatorSuite.scala b/yarn/src/test/scala/org/apache/spark/deploy/yarn/YarnAllocatorSuite.scala index 8d184a09d6..8d184a09d6 100644 --- a/yarn/common/src/test/scala/org/apache/spark/deploy/yarn/YarnAllocatorSuite.scala +++ b/yarn/src/test/scala/org/apache/spark/deploy/yarn/YarnAllocatorSuite.scala diff --git a/yarn/stable/src/test/scala/org/apache/spark/deploy/yarn/YarnClusterSuite.scala b/yarn/src/test/scala/org/apache/spark/deploy/yarn/YarnClusterSuite.scala index d79b85e867..d79b85e867 100644 --- a/yarn/stable/src/test/scala/org/apache/spark/deploy/yarn/YarnClusterSuite.scala +++ b/yarn/src/test/scala/org/apache/spark/deploy/yarn/YarnClusterSuite.scala diff --git a/yarn/common/src/test/scala/org/apache/spark/deploy/yarn/YarnSparkHadoopUtilSuite.scala b/yarn/src/test/scala/org/apache/spark/deploy/yarn/YarnSparkHadoopUtilSuite.scala index 2cc5abb3a8..2cc5abb3a8 100644 --- a/yarn/common/src/test/scala/org/apache/spark/deploy/yarn/YarnSparkHadoopUtilSuite.scala +++ b/yarn/src/test/scala/org/apache/spark/deploy/yarn/YarnSparkHadoopUtilSuite.scala diff --git a/yarn/stable/pom.xml b/yarn/stable/pom.xml deleted file mode 100644 index 8b6521ad7f..0000000000 --- a/yarn/stable/pom.xml +++ /dev/null @@ -1,95 +0,0 @@ -<?xml version="1.0" encoding="UTF-8"?> -<!-- - ~ Licensed to the Apache Software Foundation (ASF) under one or more - ~ contributor license agreements. See the NOTICE file distributed with - ~ this work for additional information regarding copyright ownership. - ~ The ASF licenses this file to You under the Apache License, Version 2.0 - ~ (the "License"); you may not use this file except in compliance with - ~ the License. You may obtain a copy of the License at - ~ - ~ http://www.apache.org/licenses/LICENSE-2.0 - ~ - ~ Unless required by applicable law or agreed to in writing, software - ~ distributed under the License is distributed on an "AS IS" BASIS, - ~ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. - ~ See the License for the specific language governing permissions and - ~ limitations under the License. - --> -<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd"> - <modelVersion>4.0.0</modelVersion> - <parent> - <groupId>org.apache.spark</groupId> - <artifactId>yarn-parent_2.10</artifactId> - <version>1.3.0-SNAPSHOT</version> - <relativePath>../pom.xml</relativePath> - </parent> - <properties> - <sbt.project.name>yarn-stable</sbt.project.name> - </properties> - - <groupId>org.apache.spark</groupId> - <artifactId>spark-yarn_2.10</artifactId> - <packaging>jar</packaging> - <name>Spark Project YARN Stable API</name> - - <dependencies> - <dependency> - <groupId>org.apache.hadoop</groupId> - <artifactId>hadoop-yarn-server-tests</artifactId> - <classifier>tests</classifier> - <scope>test</scope> - </dependency> - </dependencies> - - <!-- - See SPARK-3710. hadoop-yarn-server-tests in Hadoop 2.2 fails to pull some needed - dependencies, so they need to be added manually for the tests to work. - --> - <profiles> - <profile> - <id>hadoop-2.2</id> - <properties> - <jersey.version>1.9</jersey.version> - </properties> - <dependencies> - <dependency> - <groupId>org.mortbay.jetty</groupId> - <artifactId>jetty</artifactId> - <version>6.1.26</version> - <exclusions> - <exclusion> - <groupId>org.mortbay.jetty</groupId> - <artifactId>servlet-api</artifactId> - </exclusion> - </exclusions> - <scope>test</scope> - </dependency> - <dependency> - <groupId>com.sun.jersey</groupId> - <artifactId>jersey-core</artifactId> - <version>${jersey.version}</version> - <scope>test</scope> - </dependency> - <dependency> - <groupId>com.sun.jersey</groupId> - <artifactId>jersey-json</artifactId> - <version>${jersey.version}</version> - <scope>test</scope> - <exclusions> - <exclusion> - <groupId>stax</groupId> - <artifactId>stax-api</artifactId> - </exclusion> - </exclusions> - </dependency> - <dependency> - <groupId>com.sun.jersey</groupId> - <artifactId>jersey-server</artifactId> - <version>${jersey.version}</version> - <scope>test</scope> - </dependency> - </dependencies> - </profile> - </profiles> - -</project> |