Example usage for org.apache.hadoop.mapred JobClient JobClient

List of usage examples for org.apache.hadoop.mapred JobClient JobClient

  1. HOME
  2. Java
  3. org.apache.hadoop
  4. org.apache.hadoop.mapred.*
  5. JobClient
  6. JobClient

Introduction

In this page you can find the example usage for org.apache.hadoop.mapred JobClient JobClient.

Prototype

public JobClient(Configuration conf) throws IOException

Source Link

Document

Build a job client with the given Configuration , and connect to the default cluster

Usage

From source file:DataJoinJob.java

License:Apache License

/**
 * Submit/run a map/reduce job.//from   w  w  w  .ja  v  a 2s .c  om
 * 
 * @param job
 * @return true for success
 * @throws IOException
 */
public static boolean runJob(JobConf job) throws IOException {
    JobClient jc = new JobClient(job);
    boolean sucess = true;
    RunningJob running = null;
    try {
        running = jc.submitJob(job);
        JobID jobId = running.getID();
        System.out.println("Job " + jobId + " is submitted");
        while (!running.isComplete()) {
            System.out.println("Job " + jobId + " is still running.");
            try {
                Thread.sleep(60000);
            } catch (InterruptedException e) {
            }
            running = jc.getJob(jobId);
        }
        sucess = running.isSuccessful();
    } finally {
        if (!sucess && (running != null)) {
            running.killJob();
        }
        jc.close();
    }
    return sucess;
}

From source file:Text2FormatStorageMR.java

License:Open Source License

@SuppressWarnings("deprecation")
public static void main(String[] args) throws Exception {

    if (args.length != 2) {
        System.out.println("FormatFileMR <input> <output>");
        System.exit(-1);//from ww  w.j  a v  a2 s.c  o m
    }

    JobConf conf = new JobConf(FormatStorageMR.class);

    conf.setJobName("Text2FormatMR");

    conf.setNumMapTasks(1);
    conf.setNumReduceTasks(4);

    conf.setOutputKeyClass(LongWritable.class);
    conf.setOutputValueClass(Unit.Record.class);

    conf.setMapperClass(TextFileTestMapper.class);
    conf.setReducerClass(FormatFileTestReducer.class);

    conf.setInputFormat(TextInputFormat.class);
    conf.setOutputFormat(FormatStorageOutputFormat.class);
    conf.set("mapred.output.compress", "flase");

    Head head = new Head();
    initHead(head);

    head.toJobConf(conf);

    FileInputFormat.setInputPaths(conf, args[0]);
    Path outputPath = new Path(args[1]);
    FileOutputFormat.setOutputPath(conf, outputPath);

    FileSystem fs = outputPath.getFileSystem(conf);
    fs.delete(outputPath, true);

    JobClient jc = new JobClient(conf);
    RunningJob rj = null;
    rj = jc.submitJob(conf);

    String lastReport = "";
    SimpleDateFormat dateFormat = new SimpleDateFormat("yyyy-MM-dd hh:mm:ss,SSS");
    long reportTime = System.currentTimeMillis();
    long maxReportInterval = 3 * 1000;
    while (!rj.isComplete()) {
        try {
            Thread.sleep(1000);
        } catch (InterruptedException e) {
        }

        int mapProgress = Math.round(rj.mapProgress() * 100);
        int reduceProgress = Math.round(rj.reduceProgress() * 100);

        String report = " map = " + mapProgress + "%,  reduce = " + reduceProgress + "%";

        if (!report.equals(lastReport) || System.currentTimeMillis() >= reportTime + maxReportInterval) {

            String output = dateFormat.format(Calendar.getInstance().getTime()) + report;
            System.out.println(output);
            lastReport = report;
            reportTime = System.currentTimeMillis();
        }
    }

    System.exit(0);

}

From source file:Text2ColumntStorageMR.java

License:Open Source License

@SuppressWarnings("deprecation")
public static void main(String[] args) throws Exception {

    if (args.length != 3) {
        System.out.println("Text2ColumnStorageMR <input> <output> <columnStorageMode>");
        System.exit(-1);//from ww w. ja v  a 2  s  . c  o  m
    }

    JobConf conf = new JobConf(Text2ColumntStorageMR.class);

    conf.setJobName("Text2ColumnStorageMR");

    conf.setNumMapTasks(1);
    conf.setNumReduceTasks(4);

    conf.setOutputKeyClass(LongWritable.class);
    conf.setOutputValueClass(Unit.Record.class);

    conf.setMapperClass(TextFileMapper.class);
    conf.setReducerClass(ColumnStorageReducer.class);

    conf.setInputFormat(TextInputFormat.class);
    conf.setOutputFormat((Class<? extends OutputFormat>) ColumnStorageHiveOutputFormat.class);
    conf.set("mapred.output.compress", "flase");

    Head head = new Head();
    initHead(head);

    head.toJobConf(conf);

    int bt = Integer.valueOf(args[2]);

    FileInputFormat.setInputPaths(conf, args[0]);
    Path outputPath = new Path(args[1]);
    FileOutputFormat.setOutputPath(conf, outputPath);

    FileSystem fs = outputPath.getFileSystem(conf);
    fs.delete(outputPath, true);

    JobClient jc = new JobClient(conf);
    RunningJob rj = null;
    rj = jc.submitJob(conf);

    String lastReport = "";
    SimpleDateFormat dateFormat = new SimpleDateFormat("yyyy-MM-dd hh:mm:ss,SSS");
    long reportTime = System.currentTimeMillis();
    long maxReportInterval = 3 * 1000;
    while (!rj.isComplete()) {
        try {
            Thread.sleep(1000);
        } catch (InterruptedException e) {
        }

        int mapProgress = Math.round(rj.mapProgress() * 100);
        int reduceProgress = Math.round(rj.reduceProgress() * 100);

        String report = " map = " + mapProgress + "%,  reduce = " + reduceProgress + "%";

        if (!report.equals(lastReport) || System.currentTimeMillis() >= reportTime + maxReportInterval) {

            String output = dateFormat.format(Calendar.getInstance().getTime()) + report;
            System.out.println(output);
            lastReport = report;
            reportTime = System.currentTimeMillis();
        }
    }

    System.exit(0);

}

From source file:FormatStorage2ColumnStorageMR.java

License:Open Source License

@SuppressWarnings("deprecation")
public static void main(String[] args) throws Exception {

    if (args.length != 2) {
        System.out.println("FormatStorage2ColumnStorageMR <input> <output>");
        System.exit(-1);//from   w  ww  . j a v  a  2  s  .c  om
    }

    JobConf conf = new JobConf(FormatStorageMR.class);

    conf.setJobName("FormatStorage2ColumnStorageMR");

    conf.setNumMapTasks(1);
    conf.setNumReduceTasks(4);

    conf.setOutputKeyClass(LongWritable.class);
    conf.setOutputValueClass(Unit.Record.class);

    conf.setMapperClass(FormatStorageMapper.class);
    conf.setReducerClass(ColumnStorageReducer.class);

    conf.setInputFormat(FormatStorageInputFormat.class);
    conf.set("mapred.output.compress", "flase");

    Head head = new Head();
    initHead(head);

    head.toJobConf(conf);

    FileInputFormat.setInputPaths(conf, args[0]);
    Path outputPath = new Path(args[1]);
    FileOutputFormat.setOutputPath(conf, outputPath);

    FileSystem fs = outputPath.getFileSystem(conf);
    fs.delete(outputPath, true);

    JobClient jc = new JobClient(conf);
    RunningJob rj = null;
    rj = jc.submitJob(conf);

    String lastReport = "";
    SimpleDateFormat dateFormat = new SimpleDateFormat("yyyy-MM-dd hh:mm:ss,SSS");
    long reportTime = System.currentTimeMillis();
    long maxReportInterval = 3 * 1000;
    while (!rj.isComplete()) {
        try {
            Thread.sleep(1000);
        } catch (InterruptedException e) {
        }

        int mapProgress = Math.round(rj.mapProgress() * 100);
        int reduceProgress = Math.round(rj.reduceProgress() * 100);

        String report = " map = " + mapProgress + "%,  reduce = " + reduceProgress + "%";

        if (!report.equals(lastReport) || System.currentTimeMillis() >= reportTime + maxReportInterval) {

            String output = dateFormat.format(Calendar.getInstance().getTime()) + report;
            System.out.println(output);
            lastReport = report;
            reportTime = System.currentTimeMillis();
        }
    }

    System.exit(0);

}

From source file:azkaban.jobtype.javautils.AbstractHadoopJob.java

License:Apache License

public void run() throws Exception {
    JobConf conf = getJobConf();/*  www  .  j a  va 2 s  .com*/

    if (System.getenv(HADOOP_TOKEN_FILE_LOCATION) != null) {
        conf.set(MAPREDUCE_JOB_CREDENTIALS_BINARY, System.getenv(HADOOP_TOKEN_FILE_LOCATION));
    }

    jobClient = new JobClient(conf);
    runningJob = jobClient.submitJob(conf);
    logger.info("See " + runningJob.getTrackingURL() + " for details.");
    jobClient.monitorAndPrintJob(conf, runningJob);

    if (!runningJob.isSuccessful()) {
        throw new Exception("Hadoop job:" + getJobName() + " failed!");
    }

    // dump all counters
    Counters counters = runningJob.getCounters();
    for (String groupName : counters.getGroupNames()) {
        Counters.Group group = counters.getGroup(groupName);
        logger.info("Group: " + group.getDisplayName());
        for (Counter counter : group)
            logger.info(counter.getDisplayName() + ":\t" + counter.getValue());
    }
    updateMapReduceJobState(conf);
}

From source file:azkaban.jobtype.SecurePigWrapper.java

License:Apache License

public static void main(final String[] args) throws IOException, InterruptedException {
    final Logger logger = Logger.getRootLogger();
    final Properties p = System.getProperties();
    final Configuration conf = new Configuration();

    SecurityUtils.getProxiedUser(p, logger, conf).doAs(new PrivilegedExceptionAction<Void>() {
        @Override/*ww  w. j av  a  2 s  . co  m*/
        public Void run() throws Exception {
            prefetchToken();
            org.apache.pig.Main.main(args);
            return null;
        }

        // For Pig jobs that need to do extra communication with the
        // JobTracker, it's necessary to pre-fetch a token and include it in
        // the credentials cache
        private void prefetchToken() throws InterruptedException, IOException {
            String shouldPrefetch = p.getProperty(OBTAIN_BINARY_TOKEN);
            if (shouldPrefetch != null && shouldPrefetch.equals("true")) {
                logger.info("Pre-fetching token");
                Job job = new Job(conf, "totally phony, extremely fake, not real job");

                JobConf jc = new JobConf(conf);
                JobClient jobClient = new JobClient(jc);
                logger.info("Pre-fetching: Got new JobClient: " + jc);
                Token<DelegationTokenIdentifier> mrdt = jobClient.getDelegationToken(new Text("hi"));
                job.getCredentials().addToken(new Text("howdy"), mrdt);

                File temp = File.createTempFile("mr-azkaban", ".token");
                temp.deleteOnExit();

                FileOutputStream fos = null;
                DataOutputStream dos = null;
                try {
                    fos = new FileOutputStream(temp);
                    dos = new DataOutputStream(fos);
                    job.getCredentials().writeTokenStorageToStream(dos);
                } finally {
                    if (dos != null) {
                        dos.close();
                    }
                    if (fos != null) {
                        fos.close();
                    }
                }
                logger.info("Setting " + MAPREDUCE_JOB_CREDENTIALS_BINARY + " to " + temp.getAbsolutePath());
                System.setProperty(MAPREDUCE_JOB_CREDENTIALS_BINARY, temp.getAbsolutePath());
            } else {
                logger.info("Not pre-fetching token");
            }
        }
    });
}

From source file:azkaban.security.commons.SecurityUtils.java

License:Apache License

public static synchronized void prefetchToken(final File tokenFile, final Props p, final Logger logger)
        throws InterruptedException, IOException {

    final Configuration conf = new Configuration();
    logger.info("Getting proxy user for " + p.getString(TO_PROXY));
    logger.info("Getting proxy user for " + p.toString());

    getProxiedUser(p.toProperties(), logger, conf).doAs(new PrivilegedExceptionAction<Void>() {
        @Override//  w w w .jav a  2s.  c  o m
        public Void run() throws Exception {
            getToken(p);
            return null;
        }

        private void getToken(Props p) throws InterruptedException, IOException {
            String shouldPrefetch = p.getString(OBTAIN_BINARY_TOKEN);
            if (shouldPrefetch != null && shouldPrefetch.equals("true")) {
                logger.info("Pre-fetching token");

                logger.info("Pre-fetching fs token");
                FileSystem fs = FileSystem.get(conf);
                Token<?> fsToken = fs.getDelegationToken(p.getString("user.to.proxy"));
                logger.info("Created token: " + fsToken.toString());

                Job job = new Job(conf, "totally phony, extremely fake, not real job");
                JobConf jc = new JobConf(conf);
                JobClient jobClient = new JobClient(jc);
                logger.info("Pre-fetching job token: Got new JobClient: " + jc);
                Token<DelegationTokenIdentifier> mrdt = jobClient.getDelegationToken(new Text("hi"));
                logger.info("Created token: " + mrdt.toString());

                job.getCredentials().addToken(new Text("howdy"), mrdt);
                job.getCredentials().addToken(fsToken.getService(), fsToken);

                FileOutputStream fos = null;
                DataOutputStream dos = null;
                try {
                    fos = new FileOutputStream(tokenFile);
                    dos = new DataOutputStream(fos);
                    job.getCredentials().writeTokenStorageToStream(dos);
                } finally {
                    if (dos != null) {
                        dos.close();
                    }
                    if (fos != null) {
                        fos.close();
                    }
                }
                logger.info("Loading hadoop tokens into " + tokenFile.getAbsolutePath());
                p.put("HadoopTokenFileLoc", tokenFile.getAbsolutePath());
            } else {
                logger.info("Not pre-fetching token");
            }
        }
    });
}

From source file:azkaban.security.HadoopSecurityManager_H_1_0.java

License:Apache License

@Override
public synchronized void prefetchToken(final File tokenFile, final String userToProxy, final Logger logger)
        throws HadoopSecurityManagerException {

    logger.info("Getting hadoop tokens for " + userToProxy);

    try {//from ww  w  . j  av a 2  s  .c o  m
        getProxiedUser(userToProxy).doAs(new PrivilegedExceptionAction<Void>() {
            @Override
            public Void run() throws Exception {
                getToken(userToProxy);
                return null;
            }

            private void getToken(String userToProxy)
                    throws InterruptedException, IOException, HadoopSecurityManagerException {

                FileSystem fs = FileSystem.get(conf);
                // check if we get the correct FS, and most importantly, the conf
                logger.info("Getting DFS token from " + fs.getCanonicalServiceName() + fs.getUri());
                Token<?> fsToken = fs.getDelegationToken(userToProxy);
                if (fsToken == null) {
                    logger.error("Failed to fetch DFS token for ");
                    throw new HadoopSecurityManagerException("Failed to fetch DFS token for " + userToProxy);
                }
                logger.info("Created DFS token: " + fsToken.toString());
                logger.info("Token kind: " + fsToken.getKind());
                logger.info("Token id: " + fsToken.getIdentifier());
                logger.info("Token service: " + fsToken.getService());

                JobConf jc = new JobConf(conf);
                JobClient jobClient = new JobClient(jc);
                logger.info("Pre-fetching JT token: Got new JobClient: " + jc);

                Token<DelegationTokenIdentifier> mrdt = jobClient.getDelegationToken(new Text("mr token"));
                if (mrdt == null) {
                    logger.error("Failed to fetch JT token for ");
                    throw new HadoopSecurityManagerException("Failed to fetch JT token for " + userToProxy);
                }
                logger.info("Created JT token: " + mrdt.toString());
                logger.info("Token kind: " + mrdt.getKind());
                logger.info("Token id: " + mrdt.getIdentifier());
                logger.info("Token service: " + mrdt.getService());

                jc.getCredentials().addToken(mrdt.getService(), mrdt);
                jc.getCredentials().addToken(fsToken.getService(), fsToken);

                FileOutputStream fos = null;
                DataOutputStream dos = null;
                try {
                    fos = new FileOutputStream(tokenFile);
                    dos = new DataOutputStream(fos);
                    jc.getCredentials().writeTokenStorageToStream(dos);
                } finally {
                    if (dos != null) {
                        dos.close();
                    }
                    if (fos != null) {
                        fos.close();
                    }
                }
                // stash them to cancel after use.
                logger.info("Tokens loaded in " + tokenFile.getAbsolutePath());
            }
        });
    } catch (Exception e) {
        e.printStackTrace();
        throw new HadoopSecurityManagerException(
                "Failed to get hadoop tokens! " + e.getMessage() + e.getCause());

    }
}

From source file:azkaban.security.HadoopSecurityManager_H_1_0.java

License:Apache License

private void cancelMRJobTrackerToken(final Token<? extends TokenIdentifier> t, String userToProxy)
        throws HadoopSecurityManagerException {
    try {// w ww .j  a  va  2 s  .  c om
        getProxiedUser(userToProxy).doAs(new PrivilegedExceptionAction<Void>() {
            @SuppressWarnings("unchecked")
            @Override
            public Void run() throws Exception {
                cancelToken((Token<DelegationTokenIdentifier>) t);
                return null;
            }

            private void cancelToken(Token<DelegationTokenIdentifier> jt)
                    throws IOException, InterruptedException {
                JobConf jc = new JobConf(conf);
                JobClient jobClient = new JobClient(jc);
                jobClient.cancelDelegationToken(jt);
            }
        });
    } catch (Exception e) {
        e.printStackTrace();
        throw new HadoopSecurityManagerException("Failed to cancel Token. " + e.getMessage() + e.getCause());
    }
}

From source file:azkaban.security.HadoopSecurityManager_H_1_0.java

License:Apache License

@Override
public synchronized void prefetchToken(final File tokenFile, final Props props, final Logger logger)
        throws HadoopSecurityManagerException {

    final String userToProxy = props.getString(USER_TO_PROXY);

    logger.info("Getting hadoop tokens for " + userToProxy);

    final Credentials cred = new Credentials();

    if (props.getBoolean(OBTAIN_HCAT_TOKEN, false)) {
        try {//from   w  w w. j ava2 s. c o  m
            logger.info("Pre-fetching Hive MetaStore token from hive");

            HiveConf hiveConf = new HiveConf();
            logger.info("HiveConf.ConfVars.METASTOREURIS.varname "
                    + hiveConf.get(HiveConf.ConfVars.METASTOREURIS.varname));
            logger.info("HIVE_METASTORE_SASL_ENABLED " + hiveConf.get(HIVE_METASTORE_SASL_ENABLED));
            logger.info("HIVE_METASTORE_KERBEROS_PRINCIPAL " + hiveConf.get(HIVE_METASTORE_KERBEROS_PRINCIPAL));
            logger.info("HIVE_METASTORE_LOCAL " + hiveConf.get(HIVE_METASTORE_LOCAL));

            HiveMetaStoreClient hiveClient = new HiveMetaStoreClient(hiveConf);
            String hcatTokenStr = hiveClient.getDelegationToken(userToProxy,
                    UserGroupInformation.getLoginUser().getShortUserName());
            Token<DelegationTokenIdentifier> hcatToken = new Token<DelegationTokenIdentifier>();
            hcatToken.decodeFromUrlString(hcatTokenStr);
            logger.info("Created hive metastore token: " + hcatTokenStr);
            logger.info("Token kind: " + hcatToken.getKind());
            logger.info("Token id: " + hcatToken.getIdentifier());
            logger.info("Token service: " + hcatToken.getService());
            cred.addToken(hcatToken.getService(), hcatToken);
        } catch (Exception e) {
            e.printStackTrace();
            logger.error("Failed to get hive metastore token." + e.getMessage() + e.getCause());
        } catch (Throwable t) {
            t.printStackTrace();
            logger.error("Failed to get hive metastore token." + t.getMessage() + t.getCause());
        }
    }

    try {
        getProxiedUser(userToProxy).doAs(new PrivilegedExceptionAction<Void>() {
            @Override
            public Void run() throws Exception {
                getToken(userToProxy);
                return null;
            }

            private void getToken(String userToProxy)
                    throws InterruptedException, IOException, HadoopSecurityManagerException {
                logger.info("Here is the props for " + OBTAIN_NAMENODE_TOKEN + ": "
                        + props.getBoolean(OBTAIN_NAMENODE_TOKEN));
                if (props.getBoolean(OBTAIN_NAMENODE_TOKEN, false)) {
                    FileSystem fs = FileSystem.get(conf);
                    // check if we get the correct FS, and most importantly, the
                    // conf
                    logger.info("Getting DFS token from " + fs.getUri());
                    Token<?> fsToken = fs.getDelegationToken(userToProxy);
                    if (fsToken == null) {
                        logger.error("Failed to fetch DFS token for ");
                        throw new HadoopSecurityManagerException(
                                "Failed to fetch DFS token for " + userToProxy);
                    }
                    logger.info("Created DFS token: " + fsToken.toString());
                    logger.info("Token kind: " + fsToken.getKind());
                    logger.info("Token id: " + fsToken.getIdentifier());
                    logger.info("Token service: " + fsToken.getService());
                    cred.addToken(fsToken.getService(), fsToken);
                }

                if (props.getBoolean(OBTAIN_JOBTRACKER_TOKEN, false)) {
                    JobClient jobClient = new JobClient(new JobConf());
                    logger.info("Pre-fetching JT token from JobTracker");

                    Token<DelegationTokenIdentifier> mrdt = jobClient.getDelegationToken(new Text("mr token"));
                    if (mrdt == null) {
                        logger.error("Failed to fetch JT token");
                        throw new HadoopSecurityManagerException("Failed to fetch JT token for " + userToProxy);
                    }
                    logger.info("Created JT token: " + mrdt.toString());
                    logger.info("Token kind: " + mrdt.getKind());
                    logger.info("Token id: " + mrdt.getIdentifier());
                    logger.info("Token service: " + mrdt.getService());
                    cred.addToken(mrdt.getService(), mrdt);
                }
            }
        });

        FileOutputStream fos = null;
        DataOutputStream dos = null;
        try {
            fos = new FileOutputStream(tokenFile);
            dos = new DataOutputStream(fos);
            cred.writeTokenStorageToStream(dos);
        } finally {
            if (dos != null) {
                dos.close();
            }
            if (fos != null) {
                fos.close();
            }
        }

        // stash them to cancel after use.
        logger.info("Tokens loaded in " + tokenFile.getAbsolutePath());

    } catch (Exception e) {
        e.printStackTrace();
        throw new HadoopSecurityManagerException(
                "Failed to get hadoop tokens! " + e.getMessage() + e.getCause());
    } catch (Throwable t) {
        t.printStackTrace();
        throw new HadoopSecurityManagerException(
                "Failed to get hadoop tokens! " + t.getMessage() + t.getCause());
    }
}