Example usage for org.apache.hadoop.fs FileSystem getHomeDirectory

List of usage examples for org.apache.hadoop.fs FileSystem getHomeDirectory

Introduction

In this page you can find the example usage for org.apache.hadoop.fs FileSystem getHomeDirectory.

Prototype

public Path getHomeDirectory() 

Source Link

Document

Return the current user's home directory in this FileSystem.

Usage

From source file:com.flyhz.avengers.framework.AvengersClient.java

License:Apache License

/**
 * Main run function for the client/*from   www . j  ava  2  s . c o m*/
 * 
 * @return true if application completed successfully
 * @throws IOException
 * @throws YarnException
 */
private boolean run(String appName, List<String> commands) throws IOException, YarnException {

    LOG.info("Running Client");

    yarnClient.start();

    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());
    }

    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();
    // if needed
    // If we do not have min/max, we may not be able to correctly request
    // the required resources from the RM for the app master
    // Memory ask has to be a multiple of min and less than max.
    // Dump out information about cluster capability as seen by the resource
    // manager
    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;
    }

    // set the application name
    ApplicationSubmissionContext appContext = app.getApplicationSubmissionContext();
    ApplicationId appId = appContext.getApplicationId();
    appContext.setApplicationName(appName);

    // Set up the container launch context for the application master
    ContainerLaunchContext amContainer = Records.newRecord(ContainerLaunchContext.class);

    // 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>();

    FileSystem fs = DistributedFileSystem.get(conf);
    Path src = new Path(appJar);
    Path dst = new Path(fs.getHomeDirectory(), "avengers/" + batchId + "/avengers.jar");
    if (copy) {
        LOG.info("copy local jar to hdfs");
        fs.copyFromLocalFile(false, true, src, dst);
        copy = false;
    }
    this.hdfsPath = dst.toUri().toString();
    LOG.info("hdfs hdfsPath = {}", dst);
    FileStatus destStatus = fs.getFileStatus(dst);
    LocalResource amJarRsrc = Records.newRecord(LocalResource.class);

    amJarRsrc.setType(LocalResourceType.FILE);
    amJarRsrc.setVisibility(LocalResourceVisibility.APPLICATION);
    LOG.info("YarnURLFromPath ->{}", ConverterUtils.getYarnUrlFromPath(dst));
    amJarRsrc.setResource(ConverterUtils.getYarnUrlFromPath(dst));
    amJarRsrc.setTimestamp(destStatus.getModificationTime());
    amJarRsrc.setSize(destStatus.getLen());
    localResources.put("avengers.jar", amJarRsrc);

    // Set the log4j properties if needed
    if (!log4jPropFile.isEmpty()) {
        Path log4jSrc = new Path(log4jPropFile);
        Path log4jDst = new Path(fs.getHomeDirectory(), "log4j.props");
        fs.copyFromLocalFile(false, true, log4jSrc, log4jDst);
        FileStatus log4jFileStatus = fs.getFileStatus(log4jDst);
        LocalResource log4jRsrc = Records.newRecord(LocalResource.class);
        log4jRsrc.setType(LocalResourceType.FILE);
        log4jRsrc.setVisibility(LocalResourceVisibility.APPLICATION);
        log4jRsrc.setResource(ConverterUtils.getYarnUrlFromURI(log4jDst.toUri()));
        log4jRsrc.setTimestamp(log4jFileStatus.getModificationTime());
        log4jRsrc.setSize(log4jFileStatus.getLen());
        localResources.put("log4j.properties", log4jRsrc);
    }

    // The shell script has to be made available on the final container(s)
    // where it will be executed.
    // To do this, we need to first copy into the filesystem that is visible
    // to the yarn framework.
    // We do not need to set this as a local resource for the application
    // master as the application master does not need it.

    // Set local resource info into app master container launch context
    amContainer.setLocalResources(localResources);

    // Set the necessary security tokens as needed
    // amContainer.setContainerTokens(containerToken);

    // 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>();
    StringBuilder classPathEnv = new StringBuilder(Environment.CLASSPATH.$()).append(File.pathSeparatorChar);
    for (String c : conf.getStrings(YarnConfiguration.YARN_APPLICATION_CLASSPATH,
            YarnConfiguration.DEFAULT_YARN_APPLICATION_CLASSPATH)) {
        classPathEnv.append(File.pathSeparatorChar);
        classPathEnv.append(c.trim());
    }
    classPathEnv.append(File.pathSeparatorChar).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"));
    }
    LOG.info("CLASSPATH -> " + classPathEnv);
    env.put("CLASSPATH", classPathEnv.toString());

    amContainer.setEnvironment(env);

    for (String cmd : commands) {
        LOG.info("run command {},appId {}", cmd, appId.getId());
    }

    amContainer.setCommands(commands);

    // Set up resource type requirements
    // For now, only memory is supported so we set memory requirements
    Resource capability = Records.newRecord(Resource.class);
    capability.setMemory(amMemory);
    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()) {
        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 = Records.newRecord(Priority.class);
    pri.setPriority(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);

    // Try submitting the same request again
    // app submission failure?

    // Monitor the application
    return monitorApplication(appId);

}

From source file:com.github.hdl.tensorflow.yarn.app.Client.java

License:Apache License

private String copyLocalFileToDfs(FileSystem fs, String appId, String srcFilePath, String dstFileName)
        throws IOException {
    String suffix = TFYarnConstants.APP_NAME + "/" + appId + "/" + dstFileName;
    Path dst = new Path(fs.getHomeDirectory(), suffix);
    if (srcFilePath != null) {
        fs.copyFromLocalFile(new Path(srcFilePath), dst);
    }//  w  w  w.j a v  a 2  s  .  c om
    LOG.info("Copy " + srcFilePath + " to " + dst.toString());
    return dst.toString();
}

From source file:com.github.hdl.tensorflow.yarn.app.TFContainer.java

License:Apache License

public void addToLocalResources(FileSystem fs, String fileSrcPath, String fileDstPath, String appId,
        Map<String, LocalResource> localResources, String resources) throws IOException {

    execCmd("pwd");
    execCmd("ls -l");
    String suffix = appName + "/" + appId + "/" + fileDstPath;
    Path dst = new Path(fs.getHomeDirectory(), suffix);
    LOG.info("copy: " + fileSrcPath + " ===> " + dst.toString());
    if (fileSrcPath == null) {
        FSDataOutputStream ostream = null;
        try {//from w  w  w.  ja v a2  s  .  co  m
            ostream = FileSystem.create(fs, dst, new FsPermission((short) 0710));
            ostream.writeUTF(resources);
        } finally {
            IOUtils.closeQuietly(ostream);
        }
    } else {
        fs.copyFromLocalFile(new Path(fileSrcPath), dst);
    }

    FileStatus scFileStatus = fs.getFileStatus(dst);
    LocalResource scRsrc = LocalResource.newInstance(URL.fromURI(dst.toUri()), LocalResourceType.FILE,
            LocalResourceVisibility.APPLICATION, scFileStatus.getLen(), scFileStatus.getModificationTime());
    localResources.put(fileDstPath, scRsrc);
}

From source file:com.github.seqware.queryengine.plugins.runners.hbasemr.MRHBasePluginRunner.java

License:Open Source License

/**
 * //from   w ww.  j  av a  2 s . co m
 * @param mapReducePlugin the particular plugin to instantiate and run
 * @param reference a reference (has to be provided in lieu of a feature set) 
 * @param inputSet a set of feature sets to operate on
 * @param parameters an arbitrary number of external parameters for plugin developers to provide to their plugins
 */
public MRHBasePluginRunner(MapReducePlugin mapReducePlugin, Reference reference, List<FeatureSet> inputSet,
        Object... parameters) {
    // handle null inputSet
    if (inputSet == null) {
        inputSet = new ArrayList<FeatureSet>();
    }
    // we should either have a reference or more than one input set
    assert (reference != null || inputSet.size() > 0);
    // all feature sets should have the same reference
    if (inputSet.size() > 0) {
        SGID ref = inputSet.iterator().next().getReference().getSGID();
        for (FeatureSet set : inputSet) {
            assert (set.getReferenceID().equals(ref));
        }
    }

    SGID referenceSGID = reference != null ? reference.getSGID() : inputSet.iterator().next().getReferenceID();

    this.mapReducePlugin = mapReducePlugin;
    try {
        CreateUpdateManager manager = SWQEFactory.getModelManager();
        //outputSet should attach to the original reference
        this.outputSet = manager.buildFeatureSet().setReferenceID(referenceSGID).build();
        manager.close();

        // do setup for Map/Reduce from the HBase API
        String tableName = generateTableName(outputSet);
        String destTableName = generateTableName(outputSet);

        Configuration conf = new Configuration();
        HBaseStorage.configureHBaseConfig(conf);
        HBaseConfiguration.addHbaseResources(conf);

        // we need to pass the parameters for a featureset, maybe we can take advantage of our serializers
        byte[][] sSet = new byte[inputSet.size()][];//SWQEFactory.getSerialization().serialize(inputSet);
        for (int i = 0; i < sSet.length; i++) {
            sSet[i] = SWQEFactory.getSerialization().serialize(inputSet.get(i));
        }
        byte[] dSet = SWQEFactory.getSerialization().serialize(outputSet);

        String[] str_params = serializeParametersToString(parameters, mapReducePlugin, sSet, dSet);

        File file = new File(new URI(Constants.Term.DEVELOPMENT_DEPENDENCY.getTermValue(String.class)));
        if (file.exists()) {
            conf.setStrings("tmpjars", Constants.Term.DEVELOPMENT_DEPENDENCY.getTermValue(String.class));
        }
        conf.setStrings(EXT_PARAMETERS, str_params);
        conf.set("mapreduce.map.java.opts", "-Xmx4096m  -verbose:gc");
        conf.set("mapreduce.reduce.java.opts", "-Xmx4096m  -verbose:gc");
        conf.set("mapreduce.map.ulimit", "4194304");
        conf.set("mapreduce.reduce.ulimit", "4194304");
        conf.set("mapreduce.map.memory.mb", "4096");
        conf.set("mapreduce.reduce.memory.mb", "4096");
        conf.set("mapreduce.map.memory.physical.mb", "4096");
        conf.set("mapreduce.reduce.memory.physical.mb", "4096");

        conf.set("mapred.job.map.memory.mb", "4096");
        conf.set("mapred.job.reduce.memory.mb", "4096");

        // the above settings all seem to be ignored by hboot
        // TODO: only this one works, but as far I know, we're using mapreduce not mapred.
        // Strange
        conf.set("mapred.child.java.opts", "-Xmx2048m -verbose:gc");

        this.job = new Job(conf, mapReducePlugin.getClass().getSimpleName());

        Scan scan = new Scan();
        scan.setMaxVersions(); // we need all version data
        scan.setCaching(500); // 1 is the default in Scan, which will be bad for MapReduce jobs
        scan.setCacheBlocks(false); // don't set to true for MR jobs
        for (FeatureSet set : inputSet) {
            byte[] qualiferBytes = Bytes.toBytes(set.getSGID().getUuid().toString());
            scan.addColumn(HBaseStorage.getTEST_FAMILY_INBYTES(), qualiferBytes);
        }
        // this might be redundant, check this!!!! 
        // scan.setFilter(new QualifierFilter(CompareFilter.CompareOp.EQUAL, new BinaryComparator(qualiferBytes)));

        // handle the part that changes from job to job
        // pluginInterface.performVariableInit(tableName, destTableName, scan);
        TableMapReduceUtil.initTableMapperJob(tableName, // input HBase table name
                scan, // Scan instance to control CF and attribute selection
                PluginRunnerMapper.class, // mapper
                mapReducePlugin.getMapOutputKeyClass(), // mapper output key 
                mapReducePlugin.getMapOutputValueClass(), // mapper output value
                job);
        TableMapReduceUtil.initTableReducerJob(tableName, PluginRunnerReducer.class, job);

        if (mapReducePlugin.getOutputClass() != null) {
            job.setOutputFormatClass(mapReducePlugin.getOutputClass());
        }
        job.setReducerClass(MRHBasePluginRunner.PluginRunnerReducer.class); // reducer class

        if (mapReducePlugin.getResultMechanism() == PluginInterface.ResultMechanism.FILE) {
            FileContext fileContext = FileContext.getFileContext(this.job.getConfiguration());
            FileSystem fs = FileSystem.get(job.getConfiguration());
            Path path = new Path(fs.getHomeDirectory(),
                    new BigInteger(20, new SecureRandom()).toString(32) + mapReducePlugin.toString());
            path = fileContext.makeQualified(path);
            TextOutputFormat.setOutputPath(job, path); // adjust directories as required
        }

        job.setJarByClass(MRHBasePluginRunner.class);
        TableMapReduceUtil.addDependencyJars(job);
        TableMapReduceUtil.addDependencyJars(conf, MRHBasePluginRunner.class,
                MRHBasePluginRunner.PluginRunnerMapper.class, MRHBasePluginRunner.PluginRunnerReducer.class);
        // submit the job, but do not block
        job.submit();
    } catch (URISyntaxException ex) {
        Logger.getLogger(MRHBasePluginRunner.class.getName()).fatal(null, ex);
    } catch (InterruptedException ex) {
        Logger.getLogger(MRHBasePluginRunner.class.getName()).fatal(null, ex);
    } catch (ClassNotFoundException ex) {
        Logger.getLogger(MRHBasePluginRunner.class.getName()).fatal(null, ex);
    } catch (IOException ex) {
        Logger.getLogger(MRHBasePluginRunner.class.getName()).fatal(null, ex);
    }
}

From source file:com.github.seqware.queryengine.tutorial.Poster.java

License:Open Source License

private void recordSpace(String key) throws IOException {
    try {/*w ww .j a  v  a  2 s.  co m*/
        Configuration conf = new Configuration();
        HBaseStorage.configureHBaseConfig(conf);
        HBaseConfiguration.addHbaseResources(conf);
        FileSystem fs = FileSystem.get(conf);
        Path homeDirectory = fs.getHomeDirectory();
        Path root = homeDirectory.getParent().getParent();
        Path hbase = new Path(root, "hbase");
        ContentSummary contentSummary = fs.getContentSummary(hbase);
        long spaceConsumedinGB = convertToGB(contentSummary);
        keyValues.put(key + "-total-space-in-GB", Long.toString(spaceConsumedinGB));

        /**
         *
         * if (spaceConsumedinGB > CUT_OFF){ return; }
         *
         */
        Path featureTable = new Path(hbase,
                Constants.Term.NAMESPACE.getTermValue(String.class) + ".hbaseTestTable_v2.Feature." + HG_19);
        contentSummary = fs.getContentSummary(featureTable);
        spaceConsumedinGB = convertToGB(contentSummary);
        keyValues.put(key + "-feature-space-in-GB", Long.toString(spaceConsumedinGB));
    } catch (FileNotFoundException e) {
        /**
         * throw away, this is ok the first time *
         */
    }

}

From source file:com.inforefiner.hdata.SubmitClient.java

License:Apache License

/**
 * Main run function for the client//www .j  a v  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();

    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());
    }

    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());
        }
    }

    if (domainId != null && domainId.length() > 0 && toCreateDomain) {
        prepareTimelineDomain();
    }

    // Get a new application id
    YarnClientApplication app = yarnClient.createApplication();
    GetNewApplicationResponse appResponse = app.getNewApplicationResponse();
    // TODO get min/max resource capabilities from RM and change memory ask if needed
    // If we do not have min/max, we may not be able to correctly request
    // the required resources from the RM for the app master
    // Memory ask has to be a multiple of min and less than max.
    // Dump out information about cluster capability as seen by the resource manager
    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);

    if (attemptFailuresValidityInterval >= 0) {
        appContext.setAttemptFailuresValidityInterval(attemptFailuresValidityInterval);
    }

    // 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);
    addToLocalResources(fs, appMasterJar, appMasterJarPath, appId.toString(), localResources, null);

    // Set the log4j properties if needed
    if (!log4jPropFile.isEmpty()) {
        addToLocalResources(fs, log4jPropFile, log4jPath, appId.toString(), localResources, null);
    }

    // The shell script has to be made available on the final container(s)
    // where it will be executed.
    // To do this, we need to first copy into the filesystem that is visible
    // to the yarn framework.
    // We do not need to set this as a local resource for the application
    // master as the application master does not need it.
    String hdfsShellScriptLocation = "";
    long hdfsShellScriptLen = 0;
    long hdfsShellScriptTimestamp = 0;
    if (!shellScriptPath.isEmpty()) {
        Path shellSrc = new Path(shellScriptPath);
        String shellPathSuffix = appName + "/" + appId.toString() + "/" + SCRIPT_PATH;
        Path shellDst = new Path(fs.getHomeDirectory(), shellPathSuffix);
        fs.copyFromLocalFile(false, true, shellSrc, shellDst);
        hdfsShellScriptLocation = shellDst.toUri().toString();
        FileStatus shellFileStatus = fs.getFileStatus(shellDst);
        hdfsShellScriptLen = shellFileStatus.getLen();
        hdfsShellScriptTimestamp = shellFileStatus.getModificationTime();
    }

    if (!shellCommand.isEmpty()) {
        addToLocalResources(fs, null, shellCommandPath, appId.toString(), localResources, shellCommand);
    }

    if (shellArgs.length > 0) {
        addToLocalResources(fs, null, shellArgsPath, appId.toString(), localResources,
                StringUtils.join(shellArgs, " "));
    }

    // Set the necessary security tokens as needed
    //amContainer.setContainerTokens(containerToken);

    // 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>();

    // put location of shell script into env
    // using the env info, the application master will create the correct local resource for the
    // eventual containers that will be launched to execute the shell scripts
    env.put(DSConstants.DISTRIBUTEDSHELLSCRIPTLOCATION, hdfsShellScriptLocation);
    env.put(DSConstants.DISTRIBUTEDSHELLSCRIPTTIMESTAMP, Long.toString(hdfsShellScriptTimestamp));
    env.put(DSConstants.DISTRIBUTEDSHELLSCRIPTLEN, Long.toString(hdfsShellScriptLen));
    if (domainId != null && domainId.length() > 0) {
        env.put(DSConstants.DISTRIBUTEDSHELLTIMELINEDOMAIN, domainId);
    }

    // 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_containers " + String.valueOf(numContainers));
    if (null != nodeLabelExpression) {
        appContext.setNodeLabelExpression(nodeLabelExpression);
    }
    vargs.add("--priority " + String.valueOf(shellCmdPriority));

    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
    // TODO - what is the range for priority? how to decide?
    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);

    // TODO
    // Try submitting the same request again
    // app submission failure?
    Thread t = new Thread(new LogReceiver());
    t.start();
    // Monitor the application
    return monitorApplication(appId);
}

From source file:com.inmobi.grid.fs.s4fs.NativeS4FileSystem.java

License:Apache License

/**
 * fs.default.name in conf is the HDFS store which has credential file for
 * s3n in /user/<name> with bucket name.crd
 * the .crd file contains access:secret in a single line.
 *//*from ww  w  . j ava 2  s  .  co m*/
@Override
public void initialize(URI uri, Configuration conf) throws IOException {

    this.uri = uri;

    if (new Path(conf.get("fs.default.name")).toUri().getScheme().equals("s4")) {
        // currently illegal to set fs.default.name to s4;
        // without this, below code causes recursive call.
        return;
    }

    FileSystem fs = FileSystem.get(conf);
    Path nnWorkingDir = fs.getHomeDirectory();

    if (!fs.exists(nnWorkingDir)) {
        throw new IOException("Users home directory does not exist: " + fs.getWorkingDirectory());
    }

    String scheme = uri.getScheme();
    String bucket = uri.getAuthority();

    Path credFile = new Path(nnWorkingDir, bucket + ".crd");
    if (!fs.exists(credFile)) {
        throw new IOException(credFile.toString() + " does not exists");
    }

    StringBuilder sb = new StringBuilder(getCredentialFromFile(fs, credFile)).append("@").append(bucket);
    String bucketWithAccess = uri.toString().replaceFirst(scheme, "s3n");
    bucketWithAccess = bucketWithAccess.replaceFirst(bucket, sb.toString());

    super.initialize(new Path(bucketWithAccess).toUri(), conf);

}

From source file:com.ruizhan.hadoop.hdfs.Trash.java

License:Apache License

/**
 * Construct a trash can accessor for the FileSystem provided.
 * @param fs the FileSystem/*from  w ww  . j  ava2 s  .  c o m*/
 * @param conf a Configuration
 */
public Trash(FileSystem fs, Configuration conf) throws IOException {
    super(conf);
    trashPolicy = TrashPolicy.getInstance(conf, fs, fs.getHomeDirectory());
}

From source file:com.scistor.dshell.ScistorClient.java

License:Apache License

/**
 * Main run function for the client/*www. ja  v a  2  s  .  c om*/
 * 
 * @return true if application completed successfully
 * @throws IOException
 * @throws YarnException
 */
public boolean run() throws IOException, YarnException {

    LOG.info("Running Client");
    yarnClient.start();

    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());
    }

    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();
    // TODO get min/max resource capabilities from RM and change memory ask
    // if needed
    // If we do not have min/max, we may not be able to correctly request
    // the required resources from the RM for the app master
    // Memory ask has to be a multiple of min and less than max.
    // Dump out information about cluster capability as seen by the resource
    // manager
    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 up the container launch context for the application master
    ContainerLaunchContext amContainer = Records.newRecord(ContainerLaunchContext.class);

    // 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);
    addToLocalResources(fs, appMasterJar, appMasterJarPath, appId.toString(), localResources, null);

    // Set the log4j properties if needed
    if (!log4jPropFile.isEmpty()) {
        addToLocalResources(fs, log4jPropFile, log4jPath, appId.toString(), localResources, null);
    }

    // The shell script has to be made available on the final container(s)
    // where it will be executed.
    // To do this, we need to first copy into the filesystem that is visible
    // to the yarn framework.
    // We do not need to set this as a local resource for the application
    // master as the application master does not need it.
    String hdfsShellScriptLocation = "";
    long hdfsShellScriptLen = 0;
    long hdfsShellScriptTimestamp = 0;
    if (!shellScriptPath.isEmpty()) {
        Path shellSrc = new Path(shellScriptPath);
        String shellPathSuffix = appName + "/" + appId.toString() + "/" + SCRIPT_PATH;
        Path shellDst = new Path(fs.getHomeDirectory(), shellPathSuffix);
        fs.copyFromLocalFile(false, true, shellSrc, shellDst);
        hdfsShellScriptLocation = shellDst.toUri().toString();
        FileStatus shellFileStatus = fs.getFileStatus(shellDst);
        hdfsShellScriptLen = shellFileStatus.getLen();
        hdfsShellScriptTimestamp = shellFileStatus.getModificationTime();
    }

    if (!shellCommand.isEmpty()) {
        addToLocalResources(fs, null, shellCommandPath, appId.toString(), localResources, shellCommand);
    }

    if (shellArgs.length > 0) {
        addToLocalResources(fs, null, shellArgsPath, appId.toString(), localResources,
                StringUtils.join(shellArgs, " "));
    }
    // Set local resource info into app master container launch context
    amContainer.setLocalResources(localResources);

    // Set the necessary security tokens as needed
    // amContainer.setContainerTokens(containerToken);

    // 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>();

    // put location of shell script into env
    // using the env info, the application master will create the correct
    // local resource for the
    // eventual containers that will be launched to execute the shell
    // scripts
    env.put(ScistorDSConstants.DISTRIBUTEDSHELLSCRIPTLOCATION, hdfsShellScriptLocation);
    env.put(ScistorDSConstants.DISTRIBUTEDSHELLSCRIPTTIMESTAMP, Long.toString(hdfsShellScriptTimestamp));
    env.put(ScistorDSConstants.DISTRIBUTEDSHELLSCRIPTLEN, Long.toString(hdfsShellScriptLen));

    // ========================================jar?
    if (containerJarPaths.length != 0) {
        for (int i = 0; i < containerJarPaths.length; i++) {
            String hdfsJarLocation = "";
            String[] jarNameSplit = containerJarPaths[i].split("/");
            String jarName = jarNameSplit[jarNameSplit.length - 1];

            long hdfsJarLen = 0;
            long hdfsJarTimestamp = 0;
            if (!containerJarPaths[i].isEmpty()) {
                Path jarSrc = new Path(containerJarPaths[i]);
                String jarPathSuffix = appName + "/" + appId.toString() + "/" + jarName;
                Path jarDst = new Path(fs.getHomeDirectory(), jarPathSuffix);
                fs.copyFromLocalFile(false, true, jarSrc, jarDst);
                hdfsJarLocation = jarDst.toUri().toString();
                FileStatus jarFileStatus = fs.getFileStatus(jarDst);
                hdfsJarLen = jarFileStatus.getLen();
                hdfsJarTimestamp = jarFileStatus.getModificationTime();
                env.put(ScistorDSConstants.DISTRIBUTEDJARLOCATION + i, hdfsJarLocation);
                env.put(ScistorDSConstants.DISTRIBUTEDJARTIMESTAMP + i, Long.toString(hdfsJarTimestamp));
                env.put(ScistorDSConstants.DISTRIBUTEDJARLEN + i, Long.toString(hdfsJarLen));
            }
        }
    }
    // ========================================jar?

    // 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());

    amContainer.setEnvironment(env);

    // 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_containers " + String.valueOf(numContainers));
    vargs.add("--priority " + String.valueOf(shellCmdPriority));

    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());
    amContainer.setCommands(commands);

    // Set up resource type requirements
    // For now, both memory and vcores are supported, so we set memory and
    // vcores requirements
    Resource capability = Records.newRecord(Resource.class);
    capability.setMemory(amMemory);
    capability.setVirtualCores(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()) {
        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 = Records.newRecord(Priority.class);
    // TODO - what is the range for priority? how to decide?
    pri.setPriority(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);

    // TODO
    // Try submitting the same request again
    // app submission failure?

    // Monitor the application
    return monitorApplication(appId);

}

From source file:com.sogou.dockeronyarn.client.DockerClient.java

License:Apache License

/**
 * Main run function for the client//from w  ww  .ja va  2  s .co  m
 * @return true if application completed successfully
 * @throws IOException
 * @throws YarnException
 */
public ApplicationId run() throws IOException, YarnException {

    LOG.info("Running Client");
    yarnClient.start();

    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());
    }

    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();
    // TODO get min/max resource capabilities from RM and change memory ask if needed
    // If we do not have min/max, we may not be able to correctly request 
    // the required resources from the RM for the app master
    // Memory ask has to be a multiple of min and less than max. 
    // Dump out information about cluster capability as seen by the resource manager
    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);
    addToLocalResources(fs, appMasterJar, appMasterJarPath, appId.toString(), localResources, null);

    // Set the log4j properties if needed 
    if (!log4jPropFile.isEmpty()) {
        addToLocalResources(fs, log4jPropFile, log4jPath, appId.toString(), localResources, null);
    }

    // The shell script has to be made available on the final container(s)
    // where it will be executed. 
    // To do this, we need to first copy into the filesystem that is visible 
    // to the yarn framework. 
    // We do not need to set this as a local resource for the application 
    // master as the application master does not need it.       
    String hdfsShellScriptLocation = "";
    long hdfsShellScriptLen = 0;
    long hdfsShellScriptTimestamp = 0;
    //if (!shellScriptPath.isEmpty()) {
    // Path shellSrc = new Path(fs.getHomeDirectory(), SCRIPT_PATH);
    String shellPathSuffix = SCRIPT_PATH;
    Path shellDst = new Path(fs.getHomeDirectory(), shellPathSuffix);
    //fs.copyFromLocalFile(false, true, shellSrc, shellDst);
    hdfsShellScriptLocation = shellDst.toUri().toString();
    FileStatus shellFileStatus = fs.getFileStatus(shellDst);
    hdfsShellScriptLen = shellFileStatus.getLen();
    hdfsShellScriptTimestamp = shellFileStatus.getModificationTime();
    //}

    if (shellArgs.length > 0) {
        addToLocalResources(fs, null, shellArgsPath, appId.toString(), localResources,
                StringUtils.join(shellArgs, " "));
    }

    // Set the necessary security tokens as needed
    //amContainer.setContainerTokens(containerToken);

    // 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>();

    // put location of shell script into env
    // using the env info, the application master will create the correct local resource for the 
    // eventual containers that will be launched to execute the shell scripts
    env.put(DSConstants.DISTRIBUTEDSHELLSCRIPTLOCATION, hdfsShellScriptLocation);
    env.put(DSConstants.DISTRIBUTEDSHELLSCRIPTTIMESTAMP, Long.toString(hdfsShellScriptTimestamp));
    env.put(DSConstants.DISTRIBUTEDSHELLSCRIPTLEN, Long.toString(hdfsShellScriptLen));

    // 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(File.pathSeparatorChar)
            .append("./*");

    //    StringBuilder classPathEnv = new StringBuilder(Environment.CLASSPATH.$$())
    //      .append(ApplicationConstants.CLASS_PATH_SEPARATOR).append("./*");

    for (String c : conf.getStrings(YarnConfiguration.YARN_APPLICATION_CLASSPATH,
            YarnConfiguration.DEFAULT_YARN_APPLICATION_CLASSPATH)) {
        classPathEnv.append(File.pathSeparatorChar);
        classPathEnv.append(c.trim());
    }
    classPathEnv.append(File.pathSeparatorChar).append("./log4j.properties");

    //    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");
    //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_containers " + String.valueOf(numContainers));
    vargs.add("--priority " + String.valueOf(shellCmdPriority));
    vargs.add("--container_retry " + String.valueOf(this.container_retry));

    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
    // TODO - what is the range for priority? how to decide? 
    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);

    // TODO
    // Try submitting the same request again
    // app submission failure?

    // Monitor the application
    return appId;

}