List of usage examples for org.apache.hadoop.yarn.conf YarnConfiguration RM_ADDRESS
String RM_ADDRESS
To view the source code for org.apache.hadoop.yarn.conf YarnConfiguration RM_ADDRESS.
Click Source Link
From source file:cn.edu.buaa.act.petuumOnYarn.Client.java
License:Apache License
/** * Main run function for the client//from w ww . j av a 2 s .c o m * * @return true if application completed successfully * @throws IOException * @throws YarnException */ public boolean run() throws IOException, YarnException { LOG.info("Running Client"); yarnClient.start(); String[] s; s = conf.getStrings(YarnConfiguration.RM_ADDRESS); for (String ss : s) LOG.info("RM address: " + ss); YarnClusterMetrics clusterMetrics = yarnClient.getYarnClusterMetrics(); LOG.info("Got Cluster metric info from ASM" + ", numNodeManagers=" + clusterMetrics.getNumNodeManagers()); List<NodeReport> clusterNodeReports = yarnClient.getNodeReports(NodeState.RUNNING); LOG.info("Got Cluster node info from ASM"); for (NodeReport node : clusterNodeReports) { LOG.info("Got node report from ASM for" + ", nodeId=" + node.getNodeId() + ", nodeAddress" + node.getHttpAddress() + ", nodeRackName" + node.getRackName() + ", nodeNumContainers" + node.getNumContainers() + ", nodeIdHost" + node.getNodeId().getHost()); } QueueInfo queueInfo = yarnClient.getQueueInfo(this.amQueue); LOG.info("Queue info" + ", queueName=" + queueInfo.getQueueName() + ", queueCurrentCapacity=" + queueInfo.getCurrentCapacity() + ", queueMaxCapacity=" + queueInfo.getMaximumCapacity() + ", queueApplicationCount=" + queueInfo.getApplications().size() + ", queueChildQueueCount=" + queueInfo.getChildQueues().size()); List<QueueUserACLInfo> listAclInfo = yarnClient.getQueueAclsInfo(); for (QueueUserACLInfo aclInfo : listAclInfo) { for (QueueACL userAcl : aclInfo.getUserAcls()) { LOG.info("User ACL Info for Queue" + ", queueName=" + aclInfo.getQueueName() + ", userAcl=" + userAcl.name()); } } // Get a new application id YarnClientApplication app = yarnClient.createApplication(); GetNewApplicationResponse appResponse = app.getNewApplicationResponse(); int maxMem = appResponse.getMaximumResourceCapability().getMemory(); LOG.info("Max mem capabililty of resources in this cluster " + maxMem); // A resource ask cannot exceed the max. if (amMemory > maxMem) { LOG.info("AM memory specified above max threshold of cluster. Using max value." + ", specified=" + amMemory + ", max=" + maxMem); amMemory = maxMem; } int maxVCores = appResponse.getMaximumResourceCapability().getVirtualCores(); LOG.info("Max virtual cores capabililty of resources in this cluster " + maxVCores); if (amVCores > maxVCores) { LOG.info("AM virtual cores specified above max threshold of cluster. " + "Using max value." + ", specified=" + amVCores + ", max=" + maxVCores); amVCores = maxVCores; } // set the application name ApplicationSubmissionContext appContext = app.getApplicationSubmissionContext(); ApplicationId appId = appContext.getApplicationId(); appContext.setKeepContainersAcrossApplicationAttempts(keepContainers); appContext.setApplicationName(appName); // set local resources for the application master // local files or archives as needed // In this scenario, the jar file for the application master is part of // the local resources Map<String, LocalResource> localResources = new HashMap<String, LocalResource>(); LOG.info("Copy App Master jar from local filesystem and add to local environment"); // Copy the application master jar to the filesystem // Create a local resource to point to the destination jar path FileSystem fs = FileSystem.get(conf); YarnUtil.copyAndAddToLocalResources(fs, appMasterJar, petuumHDFSPathPrefix, appMasterJarPath, localResources, null); scriptHDFSPath = YarnUtil.copyToHDFS(fs, scriptPath, petuumHDFSPathPrefix, launchPath, null); // Set the log4j properties if needed if (!log4jPropFile.isEmpty()) { YarnUtil.copyAndAddToLocalResources(fs, log4jPropFile, petuumHDFSPathPrefix, log4jPath, localResources, null); } // Set the env variables to be setup in the env where the application // master will be run LOG.info("Set the environment for the application master"); Map<String, String> env = new HashMap<String, String>(); // Add AppMaster.jar location to classpath // At some point we should not be required to add // the hadoop specific classpaths to the env. // It should be provided out of the box. // For now setting all required classpaths including // the classpath to "." for the application jar StringBuilder classPathEnv = new StringBuilder(Environment.CLASSPATH.$$()) .append(ApplicationConstants.CLASS_PATH_SEPARATOR).append("./*"); for (String c : conf.getStrings(YarnConfiguration.YARN_APPLICATION_CLASSPATH, YarnConfiguration.DEFAULT_YARN_CROSS_PLATFORM_APPLICATION_CLASSPATH)) { classPathEnv.append(ApplicationConstants.CLASS_PATH_SEPARATOR); classPathEnv.append(c.trim()); } classPathEnv.append(ApplicationConstants.CLASS_PATH_SEPARATOR).append("./log4j.properties"); // add the runtime classpath needed for tests to work if (conf.getBoolean(YarnConfiguration.IS_MINI_YARN_CLUSTER, false)) { classPathEnv.append(':'); classPathEnv.append(System.getProperty("java.class.path")); } env.put("CLASSPATH", classPathEnv.toString()); // Set the necessary command to execute the application master Vector<CharSequence> vargs = new Vector<CharSequence>(30); // Set java executable command LOG.info("Setting up app master command"); vargs.add(Environment.JAVA_HOME.$$() + "/bin/java"); // Set Xmx based on am memory size vargs.add("-Xmx" + amMemory + "m"); // Set class name vargs.add(appMasterMainClass); // Set params for Application Master vargs.add("--container_memory " + String.valueOf(containerMemory)); vargs.add("--container_vcores " + String.valueOf(containerVirtualCores)); vargs.add("--num_nodes " + String.valueOf(numNodes)); vargs.add("--start_port " + String.valueOf(startPort)); vargs.add("--priority " + String.valueOf(workerPriority)); vargs.add("--script_hdfs_path " + scriptHDFSPath); for (Map.Entry<String, String> entry : shellEnv.entrySet()) { vargs.add("--shell_env " + entry.getKey() + "=" + entry.getValue()); } if (debugFlag) { vargs.add("--debug"); } vargs.add("1>" + ApplicationConstants.LOG_DIR_EXPANSION_VAR + "/AppMaster.stdout"); vargs.add("2>" + ApplicationConstants.LOG_DIR_EXPANSION_VAR + "/AppMaster.stderr"); // Get final commmand StringBuilder command = new StringBuilder(); for (CharSequence str : vargs) { command.append(str).append(" "); } LOG.info("Completed setting up app master command " + command.toString()); List<String> commands = new ArrayList<String>(); commands.add(command.toString()); // Set up the container launch context for the application master ContainerLaunchContext amContainer = ContainerLaunchContext.newInstance(localResources, env, commands, null, null, null); // Set up resource type requirements // For now, both memory and vcores are supported, so we set memory and // vcores requirements Resource capability = Resource.newInstance(amMemory, amVCores); appContext.setResource(capability); // Service data is a binary blob that can be passed to the application // Not needed in this scenario // amContainer.setServiceData(serviceData); // Setup security tokens if (UserGroupInformation.isSecurityEnabled()) { // Note: Credentials class is marked as LimitedPrivate for HDFS and // MapReduce Credentials credentials = new Credentials(); String tokenRenewer = conf.get(YarnConfiguration.RM_PRINCIPAL); if (tokenRenewer == null || tokenRenewer.length() == 0) { throw new IOException("Can't get Master Kerberos principal for the RM to use as renewer"); } // For now, only getting tokens for the default file-system. final Token<?> tokens[] = fs.addDelegationTokens(tokenRenewer, credentials); if (tokens != null) { for (Token<?> token : tokens) { LOG.info("Got dt for " + fs.getUri() + "; " + token); } } DataOutputBuffer dob = new DataOutputBuffer(); credentials.writeTokenStorageToStream(dob); ByteBuffer fsTokens = ByteBuffer.wrap(dob.getData(), 0, dob.getLength()); amContainer.setTokens(fsTokens); } appContext.setAMContainerSpec(amContainer); // Set the priority for the application master Priority pri = Priority.newInstance(amPriority); appContext.setPriority(pri); // Set the queue to which this application is to be submitted in the RM appContext.setQueue(amQueue); // Submit the application to the applications manager // SubmitApplicationResponse submitResp = // applicationsManager.submitApplication(appRequest); // Ignore the response as either a valid response object is returned on // success // or an exception thrown to denote some form of a failure LOG.info("Submitting application to ASM"); yarnClient.submitApplication(appContext); // Monitor the application currentTime = System.currentTimeMillis(); LOG.info("submit AM in " + (currentTime - startTime) + "ms"); return monitorApplication(appId); }
From source file:co.cask.cdap.common.security.YarnTokenUtils.java
License:Apache License
/** * Gets a Yarn delegation token and stores it in the given Credentials. * * @return the same Credentials instance as the one given in parameter. *///from w ww. ja va 2 s .co m public static Credentials obtainToken(YarnConfiguration configuration, Credentials credentials) { if (!UserGroupInformation.isSecurityEnabled()) { return credentials; } try { YarnClient yarnClient = YarnClient.createYarnClient(); yarnClient.init(configuration); yarnClient.start(); try { Text renewer = new Text(UserGroupInformation.getCurrentUser().getShortUserName()); org.apache.hadoop.yarn.api.records.Token rmDelegationToken = yarnClient .getRMDelegationToken(renewer); // TODO: The following logic should be replaced with call to ClientRMProxy.getRMDelegationTokenService after // CDAP-4825 is resolved List<String> services = new ArrayList<>(); if (HAUtil.isHAEnabled(configuration)) { // If HA is enabled, we need to enumerate all RM hosts // and add the corresponding service name to the token service // Copy the yarn conf since we need to modify it to get the RM addresses YarnConfiguration yarnConf = new YarnConfiguration(configuration); for (String rmId : HAUtil.getRMHAIds(configuration)) { yarnConf.set(YarnConfiguration.RM_HA_ID, rmId); InetSocketAddress address = yarnConf.getSocketAddr(YarnConfiguration.RM_ADDRESS, YarnConfiguration.DEFAULT_RM_ADDRESS, YarnConfiguration.DEFAULT_RM_PORT); services.add(SecurityUtil.buildTokenService(address).toString()); } } else { services.add(SecurityUtil.buildTokenService(YarnUtils.getRMAddress(configuration)).toString()); } Token<TokenIdentifier> token = ConverterUtils.convertFromYarn(rmDelegationToken, (InetSocketAddress) null); token.setService(new Text(Joiner.on(',').join(services))); credentials.addToken(new Text(token.getService()), token); // OK to log, it won't log the credential, only information about the token. LOG.info("Added RM delegation token: {}", token); } finally { yarnClient.stop(); } return credentials; } catch (Exception e) { LOG.error("Failed to get secure token for Yarn.", e); throw Throwables.propagate(e); } }
From source file:com.cloudera.kitten.client.service.ApplicationsManagerConnectionFactory.java
License:Open Source License
@Override public ClientRMProtocol connect() { YarnConfiguration yarnConf = new YarnConfiguration(conf); InetSocketAddress rmAddress = NetUtils .createSocketAddr(yarnConf.get(YarnConfiguration.RM_ADDRESS, YarnConfiguration.DEFAULT_RM_ADDRESS)); LOG.info("Connecting to ResourceManager at: " + rmAddress); return ((ClientRMProtocol) rpc.getProxy(ClientRMProtocol.class, rmAddress, conf)); }
From source file:com.datatorrent.stram.StramClient.java
License:Apache License
private void addRMDelegationToken(final String renewer, final Credentials credentials) throws IOException, YarnException { // Get the ResourceManager delegation rmToken final org.apache.hadoop.yarn.api.records.Token rmDelegationToken = yarnClient .getRMDelegationToken(new Text(renewer)); Token<RMDelegationTokenIdentifier> token; // TODO: Use the utility method getRMDelegationTokenService in ClientRMProxy to remove the separate handling of // TODO: HA and non-HA cases when hadoop dependency is changed to hadoop 2.4 or above if (conf.getBoolean(RM_HA_ENABLED, DEFAULT_RM_HA_ENABLED)) { LOG.info("Yarn Resource Manager HA is enabled"); token = getRMHAToken(rmDelegationToken); } else {//from w ww . ja v a 2 s . c o m LOG.info("Yarn Resource Manager HA is not enabled"); InetSocketAddress rmAddress = conf.getSocketAddr(YarnConfiguration.RM_ADDRESS, YarnConfiguration.DEFAULT_RM_ADDRESS, YarnConfiguration.DEFAULT_RM_PORT); token = ConverterUtils.convertFromYarn(rmDelegationToken, rmAddress); } LOG.info("RM dt {}", token); credentials.addToken(token.getService(), token); }
From source file:com.datatorrent.stram.StreamingAppMasterService.java
License:Apache License
/** * Main run function for the application master * * @throws YarnException//from w w w .j a v a2s . c o m */ @SuppressWarnings("SleepWhileInLoop") private void execute() throws YarnException, IOException { LOG.info("Starting ApplicationMaster"); final Credentials credentials = UserGroupInformation.getCurrentUser().getCredentials(); LOG.info("number of tokens: {}", credentials.getAllTokens().size()); Iterator<Token<?>> iter = credentials.getAllTokens().iterator(); while (iter.hasNext()) { Token<?> token = iter.next(); LOG.debug("token: {}", token); } final Configuration conf = getConfig(); long tokenLifeTime = (long) (dag.getValue(LogicalPlan.TOKEN_REFRESH_ANTICIPATORY_FACTOR) * Math .min(dag.getValue(LogicalPlan.HDFS_TOKEN_LIFE_TIME), dag.getValue(LogicalPlan.RM_TOKEN_LIFE_TIME))); long expiryTime = System.currentTimeMillis() + tokenLifeTime; LOG.debug(" expiry token time {}", tokenLifeTime); String hdfsKeyTabFile = dag.getValue(LogicalPlan.KEY_TAB_FILE); // Register self with ResourceManager RegisterApplicationMasterResponse response = amRmClient.registerApplicationMaster(appMasterHostname, 0, appMasterTrackingUrl); // Dump out information about cluster capability as seen by the resource manager int maxMem = response.getMaximumResourceCapability().getMemory(); int maxVcores = response.getMaximumResourceCapability().getVirtualCores(); LOG.info("Max mem {}m and vcores {} capabililty of resources in this cluster ", maxMem, maxVcores); // for locality relaxation fall back Map<StreamingContainerAgent.ContainerStartRequest, MutablePair<Integer, ContainerRequest>> requestedResources = Maps .newHashMap(); // Setup heartbeat emitter // TODO poll RM every now and then with an empty request to let RM know that we are alive // The heartbeat interval after which an AM is timed out by the RM is defined by a config setting: // RM_AM_EXPIRY_INTERVAL_MS with default defined by DEFAULT_RM_AM_EXPIRY_INTERVAL_MS // The allocate calls to the RM count as heartbeat so, for now, this additional heartbeat emitter // is not required. int loopCounter = -1; List<ContainerId> releasedContainers = new ArrayList<ContainerId>(); int numTotalContainers = 0; // keep track of already requested containers to not request them again while waiting for allocation int numRequestedContainers = 0; int numReleasedContainers = 0; int nextRequestPriority = 0; ResourceRequestHandler resourceRequestor = new ResourceRequestHandler(); YarnClient clientRMService = YarnClient.createYarnClient(); try { // YARN-435 // we need getClusterNodes to populate the initial node list, // subsequent updates come through the heartbeat response clientRMService.init(conf); clientRMService.start(); ApplicationReport ar = StramClientUtils.getStartedAppInstanceByName(clientRMService, dag.getAttributes().get(DAG.APPLICATION_NAME), UserGroupInformation.getLoginUser().getUserName(), dag.getAttributes().get(DAG.APPLICATION_ID)); if (ar != null) { appDone = true; dnmgr.shutdownDiagnosticsMessage = String.format( "Application master failed due to application %s with duplicate application name \"%s\" by the same user \"%s\" is already started.", ar.getApplicationId().toString(), ar.getName(), ar.getUser()); LOG.info("Forced shutdown due to {}", dnmgr.shutdownDiagnosticsMessage); finishApplication(FinalApplicationStatus.FAILED, numTotalContainers); return; } resourceRequestor.updateNodeReports(clientRMService.getNodeReports()); } catch (Exception e) { throw new RuntimeException("Failed to retrieve cluster nodes report.", e); } finally { clientRMService.stop(); } // check for previously allocated containers // as of 2.2, containers won't survive AM restart, but this will change in the future - YARN-1490 checkContainerStatus(); FinalApplicationStatus finalStatus = FinalApplicationStatus.SUCCEEDED; final InetSocketAddress rmAddress = conf.getSocketAddr(YarnConfiguration.RM_ADDRESS, YarnConfiguration.DEFAULT_RM_ADDRESS, YarnConfiguration.DEFAULT_RM_PORT); while (!appDone) { loopCounter++; if (UserGroupInformation.isSecurityEnabled() && System.currentTimeMillis() >= expiryTime && hdfsKeyTabFile != null) { String applicationId = appAttemptID.getApplicationId().toString(); expiryTime = StramUserLogin.refreshTokens(tokenLifeTime, "." + File.separator + "tmp", applicationId, conf, hdfsKeyTabFile, credentials, rmAddress, true); } Runnable r; while ((r = this.pendingTasks.poll()) != null) { r.run(); } // log current state /* * LOG.info("Current application state: loop=" + loopCounter + ", appDone=" + appDone + ", total=" + * numTotalContainers + ", requested=" + numRequestedContainers + ", completed=" + numCompletedContainers + * ", failed=" + numFailedContainers + ", currentAllocated=" + this.allAllocatedContainers.size()); */ // Sleep before each loop when asking RM for containers // to avoid flooding RM with spurious requests when it // need not have any available containers try { sleep(1000); } catch (InterruptedException e) { LOG.info("Sleep interrupted " + e.getMessage()); } // Setup request to be sent to RM to allocate containers List<ContainerRequest> containerRequests = new ArrayList<ContainerRequest>(); List<ContainerRequest> removedContainerRequests = new ArrayList<ContainerRequest>(); // request containers for pending deploy requests if (!dnmgr.containerStartRequests.isEmpty()) { StreamingContainerAgent.ContainerStartRequest csr; while ((csr = dnmgr.containerStartRequests.poll()) != null) { if (csr.container.getRequiredMemoryMB() > maxMem) { LOG.warn("Container memory {}m above max threshold of cluster. Using max value {}m.", csr.container.getRequiredMemoryMB(), maxMem); csr.container.setRequiredMemoryMB(maxMem); } if (csr.container.getRequiredVCores() > maxVcores) { LOG.warn("Container vcores {} above max threshold of cluster. Using max value {}.", csr.container.getRequiredVCores(), maxVcores); csr.container.setRequiredVCores(maxVcores); } csr.container.setResourceRequestPriority(nextRequestPriority++); ContainerRequest cr = resourceRequestor.createContainerRequest(csr, true); MutablePair<Integer, ContainerRequest> pair = new MutablePair<Integer, ContainerRequest>( loopCounter, cr); requestedResources.put(csr, pair); containerRequests.add(cr); } } if (!requestedResources.isEmpty()) { //resourceRequestor.clearNodeMapping(); for (Map.Entry<StreamingContainerAgent.ContainerStartRequest, MutablePair<Integer, ContainerRequest>> entry : requestedResources .entrySet()) { if ((loopCounter - entry.getValue().getKey()) > NUMBER_MISSED_HEARTBEATS) { StreamingContainerAgent.ContainerStartRequest csr = entry.getKey(); removedContainerRequests.add(entry.getValue().getRight()); ContainerRequest cr = resourceRequestor.createContainerRequest(csr, false); entry.getValue().setLeft(loopCounter); entry.getValue().setRight(cr); containerRequests.add(cr); } } } numTotalContainers += containerRequests.size(); numRequestedContainers += containerRequests.size(); AllocateResponse amResp = sendContainerAskToRM(containerRequests, removedContainerRequests, releasedContainers); if (amResp.getAMCommand() != null) { LOG.info(" statement executed:{}", amResp.getAMCommand()); switch (amResp.getAMCommand()) { case AM_RESYNC: case AM_SHUTDOWN: throw new YarnRuntimeException("Received the " + amResp.getAMCommand() + " command from RM"); default: throw new YarnRuntimeException("Received the " + amResp.getAMCommand() + " command from RM"); } } releasedContainers.clear(); // Retrieve list of allocated containers from the response List<Container> newAllocatedContainers = amResp.getAllocatedContainers(); // LOG.info("Got response from RM for container ask, allocatedCnt=" + newAllocatedContainers.size()); numRequestedContainers -= newAllocatedContainers.size(); long timestamp = System.currentTimeMillis(); for (Container allocatedContainer : newAllocatedContainers) { LOG.info("Got new container." + ", containerId=" + allocatedContainer.getId() + ", containerNode=" + allocatedContainer.getNodeId() + ", containerNodeURI=" + allocatedContainer.getNodeHttpAddress() + ", containerResourceMemory" + allocatedContainer.getResource().getMemory() + ", priority" + allocatedContainer.getPriority()); // + ", containerToken" + allocatedContainer.getContainerToken().getIdentifier().toString()); boolean alreadyAllocated = true; StreamingContainerAgent.ContainerStartRequest csr = null; for (Map.Entry<StreamingContainerAgent.ContainerStartRequest, MutablePair<Integer, ContainerRequest>> entry : requestedResources .entrySet()) { if (entry.getKey().container.getResourceRequestPriority() == allocatedContainer.getPriority() .getPriority()) { alreadyAllocated = false; csr = entry.getKey(); break; } } if (alreadyAllocated) { LOG.info("Releasing {} as resource with priority {} was already assigned", allocatedContainer.getId(), allocatedContainer.getPriority()); releasedContainers.add(allocatedContainer.getId()); numReleasedContainers++; numRequestedContainers++; continue; } if (csr != null) { requestedResources.remove(csr); } // allocate resource to container ContainerResource resource = new ContainerResource(allocatedContainer.getPriority().getPriority(), allocatedContainer.getId().toString(), allocatedContainer.getNodeId().toString(), allocatedContainer.getResource().getMemory(), allocatedContainer.getResource().getVirtualCores(), allocatedContainer.getNodeHttpAddress()); StreamingContainerAgent sca = dnmgr.assignContainer(resource, null); if (sca == null) { // allocated container no longer needed, add release request LOG.warn("Container {} allocated but nothing to deploy, going to release this container.", allocatedContainer.getId()); releasedContainers.add(allocatedContainer.getId()); } else { AllocatedContainer allocatedContainerHolder = new AllocatedContainer(allocatedContainer); this.allocatedContainers.put(allocatedContainer.getId().toString(), allocatedContainerHolder); ByteBuffer tokens = null; if (UserGroupInformation.isSecurityEnabled()) { UserGroupInformation ugi = UserGroupInformation.getLoginUser(); Token<StramDelegationTokenIdentifier> delegationToken = allocateDelegationToken( ugi.getUserName(), heartbeatListener.getAddress()); allocatedContainerHolder.delegationToken = delegationToken; //ByteBuffer tokens = LaunchContainerRunnable.getTokens(delegationTokenManager, heartbeatListener.getAddress()); tokens = LaunchContainerRunnable.getTokens(ugi, delegationToken); } LaunchContainerRunnable launchContainer = new LaunchContainerRunnable(allocatedContainer, nmClient, sca, tokens); // Thread launchThread = new Thread(runnableLaunchContainer); // launchThreads.add(launchThread); // launchThread.start(); launchContainer.run(); // communication with NMs is now async // record container start event StramEvent ev = new StramEvent.StartContainerEvent(allocatedContainer.getId().toString(), allocatedContainer.getNodeId().toString()); ev.setTimestamp(timestamp); dnmgr.recordEventAsync(ev); } } // track node updates for future locality constraint allocations // TODO: it seems 2.0.4-alpha doesn't give us any updates resourceRequestor.updateNodeReports(amResp.getUpdatedNodes()); // Check the completed containers List<ContainerStatus> completedContainers = amResp.getCompletedContainersStatuses(); // LOG.debug("Got response from RM for container ask, completedCnt=" + completedContainers.size()); for (ContainerStatus containerStatus : completedContainers) { LOG.info("Completed containerId=" + containerStatus.getContainerId() + ", state=" + containerStatus.getState() + ", exitStatus=" + containerStatus.getExitStatus() + ", diagnostics=" + containerStatus.getDiagnostics()); // non complete containers should not be here assert (containerStatus.getState() == ContainerState.COMPLETE); AllocatedContainer allocatedContainer = allocatedContainers .remove(containerStatus.getContainerId().toString()); if (allocatedContainer != null && allocatedContainer.delegationToken != null) { UserGroupInformation ugi = UserGroupInformation.getLoginUser(); delegationTokenManager.cancelToken(allocatedContainer.delegationToken, ugi.getUserName()); } int exitStatus = containerStatus.getExitStatus(); if (0 != exitStatus) { if (allocatedContainer != null) { numFailedContainers.incrementAndGet(); } // if (exitStatus == 1) { // // non-recoverable StreamingContainer failure // appDone = true; // finalStatus = FinalApplicationStatus.FAILED; // dnmgr.shutdownDiagnosticsMessage = "Unrecoverable failure " + containerStatus.getContainerId(); // LOG.info("Exiting due to: {}", dnmgr.shutdownDiagnosticsMessage); // } // else { // Recoverable failure or process killed (externally or via stop request by AM) // also occurs when a container was released by the application but never assigned/launched LOG.debug("Container {} failed or killed.", containerStatus.getContainerId()); dnmgr.scheduleContainerRestart(containerStatus.getContainerId().toString()); // } } else { // container completed successfully numCompletedContainers.incrementAndGet(); LOG.info("Container completed successfully." + ", containerId=" + containerStatus.getContainerId()); } String containerIdStr = containerStatus.getContainerId().toString(); dnmgr.removeContainerAgent(containerIdStr); // record container stop event StramEvent ev = new StramEvent.StopContainerEvent(containerIdStr, containerStatus.getExitStatus()); ev.setReason(containerStatus.getDiagnostics()); dnmgr.recordEventAsync(ev); } if (dnmgr.forcedShutdown) { LOG.info("Forced shutdown due to {}", dnmgr.shutdownDiagnosticsMessage); finalStatus = FinalApplicationStatus.FAILED; appDone = true; } else if (allocatedContainers.isEmpty() && numRequestedContainers == 0 && dnmgr.containerStartRequests.isEmpty()) { LOG.debug("Exiting as no more containers are allocated or requested"); finalStatus = FinalApplicationStatus.SUCCEEDED; appDone = true; } LOG.debug("Current application state: loop=" + loopCounter + ", appDone=" + appDone + ", total=" + numTotalContainers + ", requested=" + numRequestedContainers + ", released=" + numReleasedContainers + ", completed=" + numCompletedContainers + ", failed=" + numFailedContainers + ", currentAllocated=" + allocatedContainers.size()); // monitor child containers dnmgr.monitorHeartbeat(); } finishApplication(finalStatus, numTotalContainers); }
From source file:com.datatorrent.stram.util.ConfigUtils.java
License:Apache License
public static InetSocketAddress getRMAddress(YarnConfiguration conf) { return conf.getSocketAddr(YarnConfiguration.RM_ADDRESS, YarnConfiguration.DEFAULT_RM_ADDRESS, YarnConfiguration.DEFAULT_RM_PORT); }
From source file:com.flyhz.avengers.framework.AvengersClient.java
License:Apache License
AvengersClient(Configuration conf) {
batchId = System.currentTimeMillis();
LOG.info("conf is {},batchId is {}", batchId, conf);
this.conf = conf;
System.setProperty("HADOOP_USER_NAME", "avengers");
yarnClient = YarnClient.createYarnClient();
conf.getSocketAddr(YarnConfiguration.RM_ADDRESS, YarnConfiguration.DEFAULT_RM_ADDRESS,
YarnConfiguration.DEFAULT_RM_PORT);
yarnClient.init(conf);//from w w w . j a v a2 s.com
opts = new Options();
opts.addOption("jar", true, "Jar file containing the application master");
Option all = new Option("all", false, "avengers all:include crawl fetch analyze");
all.setArgs(0);
Option crawl = new Option("crawl", true, "avengers crawl");
crawl.setArgs(1);
crawl.setArgName("batchId");
Option fetch = new Option("fetch", true, "avengers fetch");
fetch.setArgs(1);
fetch.setArgName("batchId");
Option analyze = new Option("analyze", true, "avengers analyze");
analyze.setArgs(1);
analyze.setArgName("batchId");
opts.addOption(all);
opts.addOption(crawl);
opts.addOption(fetch);
opts.addOption(analyze);
avengersProcessCmds = new LinkedHashMap<String, List<String>>();
this.hdfsPath = "avengers/" + batchId + "/avengers.jar";
}
From source file:com.github.sakserv.minicluster.impl.MRLocalCluster.java
License:Apache License
@Override public void configure() throws Exception { // Handle Windows WindowsLibsUtils.setHadoopHome();//from ww w. j av a 2 s. c om configuration.set(YarnConfiguration.RM_ADDRESS, resourceManagerAddress); configuration.set(YarnConfiguration.RM_HOSTNAME, resourceManagerHostname); configuration.set(YarnConfiguration.RM_SCHEDULER_ADDRESS, resourceManagerSchedulerAddress); configuration.set(YarnConfiguration.RM_RESOURCE_TRACKER_ADDRESS, resourceManagerResourceTrackerAddress); configuration.set(YarnConfiguration.RM_WEBAPP_ADDRESS, resourceManagerWebappAddress); configuration.set(JHAdminConfig.MR_HISTORY_ADDRESS, jobHistoryAddress); configuration.set(YarnConfiguration.YARN_MINICLUSTER_FIXED_PORTS, "true"); configuration.set(JHAdminConfig.MR_HISTORY_MINICLUSTER_FIXED_PORTS, "true"); if (getUseInJvmContainerExecutor()) { configuration.set(YarnConfiguration.NM_CONTAINER_EXECUTOR, inJvmContainerExecutorClass); configuration.set("fs.hdfs.impl", org.apache.hadoop.hdfs.DistributedFileSystem.class.getName()); configuration.set("fs.file.impl", org.apache.hadoop.fs.LocalFileSystem.class.getName()); } if (null != hdfsDefaultFs) { configuration.set("fs.defaultFS", hdfsDefaultFs); configuration.set("dfs.replication", "1"); } }
From source file:com.github.sakserv.minicluster.impl.YarnLocalCluster.java
License:Apache License
@Override public void configure() throws Exception { // Handle Windows WindowsLibsUtils.setHadoopHome();/* w w w . j a v a 2 s . com*/ configuration.set(YarnConfiguration.RM_ADDRESS, resourceManagerAddress); configuration.set(YarnConfiguration.RM_HOSTNAME, resourceManagerHostname); configuration.set(YarnConfiguration.RM_SCHEDULER_ADDRESS, resourceManagerSchedulerAddress); configuration.set(YarnConfiguration.RM_RESOURCE_TRACKER_ADDRESS, resourceManagerResourceTrackerAddress); configuration.set(YarnConfiguration.RM_WEBAPP_ADDRESS, resourceManagerWebappAddress); configuration.set(YarnConfiguration.YARN_MINICLUSTER_FIXED_PORTS, "true"); if (getUseInJvmContainerExecutor()) { configuration.set(YarnConfiguration.NM_CONTAINER_EXECUTOR, inJvmContainerExecutorClass); configuration.set("fs.hdfs.impl", org.apache.hadoop.hdfs.DistributedFileSystem.class.getName()); configuration.set("fs.file.impl", org.apache.hadoop.fs.LocalFileSystem.class.getName()); } }
From source file:com.google.mr4c.hadoop.yarn.YarnTestBinding.java
License:Open Source License
private void startMrCluster() throws IOException { Configuration conf = new JobConf(); FileSystem.setDefaultUri(conf, HadoopTestUtils.getTestDFS().getUri()); conf.setBoolean(YarnConfiguration.YARN_MINICLUSTER_FIXED_PORTS, true); conf.setBoolean(JHAdminConfig.MR_HISTORY_MINICLUSTER_FIXED_PORTS, true); String addr = MiniYARNCluster.getHostname() + ":0"; conf.set(YarnConfiguration.RM_ADDRESS, addr); conf.set(JHAdminConfig.MR_HISTORY_ADDRESS, addr); m_mrCluster = MiniMRClientClusterFactory.create(HadoopTestUtils.class, "MR4CTests", 1, // num node managers conf);//from w ww. j a va 2 s . co m // make sure startup is finished for (int i = 0; i < 60; i++) { String newAddr = m_mrCluster.getConfig().get(YarnConfiguration.RM_ADDRESS); if (newAddr.equals(addr)) { s_log.warn("MiniYARNCluster startup not complete"); try { Thread.sleep(1000); } catch (InterruptedException ie) { throw new IOException(ie); } } else { s_log.info("MiniYARNCluster now available at {}", newAddr); return; } } throw new IOException("MiniYARNCluster taking too long to startup"); }