Search in sources :

Example 1 with YarnUncaughtExceptionHandler

use of org.apache.hadoop.yarn.YarnUncaughtExceptionHandler in project hadoop by apache.

the class YarnChild method main.

public static void main(String[] args) throws Throwable {
    Thread.setDefaultUncaughtExceptionHandler(new YarnUncaughtExceptionHandler());
    LOG.debug("Child starting");
    final JobConf job = new JobConf(MRJobConfig.JOB_CONF_FILE);
    // Initing with our JobConf allows us to avoid loading confs twice
    Limits.init(job);
    UserGroupInformation.setConfiguration(job);
    // MAPREDUCE-6565: need to set configuration for SecurityUtil.
    SecurityUtil.setConfiguration(job);
    String host = args[0];
    int port = Integer.parseInt(args[1]);
    final InetSocketAddress address = NetUtils.createSocketAddrForHost(host, port);
    final TaskAttemptID firstTaskid = TaskAttemptID.forName(args[2]);
    long jvmIdLong = Long.parseLong(args[3]);
    JVMId jvmId = new JVMId(firstTaskid.getJobID(), firstTaskid.getTaskType() == TaskType.MAP, jvmIdLong);
    CallerContext.setCurrent(new CallerContext.Builder("mr_" + firstTaskid.toString()).build());
    // initialize metrics
    DefaultMetricsSystem.initialize(StringUtils.camelize(firstTaskid.getTaskType().name()) + "Task");
    // Security framework already loaded the tokens into current ugi
    Credentials credentials = UserGroupInformation.getCurrentUser().getCredentials();
    LOG.info("Executing with tokens:");
    for (Token<?> token : credentials.getAllTokens()) {
        LOG.info(token);
    }
    // Create TaskUmbilicalProtocol as actual task owner.
    UserGroupInformation taskOwner = UserGroupInformation.createRemoteUser(firstTaskid.getJobID().toString());
    Token<JobTokenIdentifier> jt = TokenCache.getJobToken(credentials);
    SecurityUtil.setTokenService(jt, address);
    taskOwner.addToken(jt);
    final TaskUmbilicalProtocol umbilical = taskOwner.doAs(new PrivilegedExceptionAction<TaskUmbilicalProtocol>() {

        @Override
        public TaskUmbilicalProtocol run() throws Exception {
            return (TaskUmbilicalProtocol) RPC.getProxy(TaskUmbilicalProtocol.class, TaskUmbilicalProtocol.versionID, address, job);
        }
    });
    // report non-pid to application master
    JvmContext context = new JvmContext(jvmId, "-1000");
    LOG.debug("PID: " + System.getenv().get("JVM_PID"));
    Task task = null;
    UserGroupInformation childUGI = null;
    ScheduledExecutorService logSyncer = null;
    try {
        int idleLoopCount = 0;
        JvmTask myTask = null;
        ;
        // poll for new task
        for (int idle = 0; null == myTask; ++idle) {
            long sleepTimeMilliSecs = Math.min(idle * 500, 1500);
            LOG.info("Sleeping for " + sleepTimeMilliSecs + "ms before retrying again. Got null now.");
            MILLISECONDS.sleep(sleepTimeMilliSecs);
            myTask = umbilical.getTask(context);
        }
        if (myTask.shouldDie()) {
            return;
        }
        task = myTask.getTask();
        YarnChild.taskid = task.getTaskID();
        // Create the job-conf and set credentials
        configureTask(job, task, credentials, jt);
        // log the system properties
        String systemPropsToLog = MRApps.getSystemPropertiesToLog(job);
        if (systemPropsToLog != null) {
            LOG.info(systemPropsToLog);
        }
        // Initiate Java VM metrics
        JvmMetrics.initSingleton(jvmId.toString(), job.getSessionId());
        childUGI = UserGroupInformation.createRemoteUser(System.getenv(ApplicationConstants.Environment.USER.toString()));
        // Add tokens to new user so that it may execute its task correctly.
        childUGI.addCredentials(credentials);
        // set job classloader if configured before invoking the task
        MRApps.setJobClassLoader(job);
        logSyncer = TaskLog.createLogSyncer();
        // Create a final reference to the task for the doAs block
        final Task taskFinal = task;
        childUGI.doAs(new PrivilegedExceptionAction<Object>() {

            @Override
            public Object run() throws Exception {
                // use job-specified working directory
                setEncryptedSpillKeyIfRequired(taskFinal);
                FileSystem.get(job).setWorkingDirectory(job.getWorkingDirectory());
                // run the task
                taskFinal.run(job, umbilical);
                return null;
            }
        });
    } catch (FSError e) {
        LOG.fatal("FSError from child", e);
        if (!ShutdownHookManager.get().isShutdownInProgress()) {
            umbilical.fsError(taskid, e.getMessage());
        }
    } catch (Exception exception) {
        LOG.warn("Exception running child : " + StringUtils.stringifyException(exception));
        try {
            if (task != null) {
                // do cleanup for the task
                if (childUGI == null) {
                    // no need to job into doAs block
                    task.taskCleanup(umbilical);
                } else {
                    final Task taskFinal = task;
                    childUGI.doAs(new PrivilegedExceptionAction<Object>() {

                        @Override
                        public Object run() throws Exception {
                            taskFinal.taskCleanup(umbilical);
                            return null;
                        }
                    });
                }
            }
        } catch (Exception e) {
            LOG.info("Exception cleaning up: " + StringUtils.stringifyException(e));
        }
        // Report back any failures, for diagnostic purposes
        if (taskid != null) {
            if (!ShutdownHookManager.get().isShutdownInProgress()) {
                umbilical.fatalError(taskid, StringUtils.stringifyException(exception));
            }
        }
    } catch (Throwable throwable) {
        LOG.fatal("Error running child : " + StringUtils.stringifyException(throwable));
        if (taskid != null) {
            if (!ShutdownHookManager.get().isShutdownInProgress()) {
                Throwable tCause = throwable.getCause();
                String cause = tCause == null ? throwable.getMessage() : StringUtils.stringifyException(tCause);
                umbilical.fatalError(taskid, cause);
            }
        }
    } finally {
        RPC.stopProxy(umbilical);
        DefaultMetricsSystem.shutdown();
        TaskLog.syncLogsShutdown(logSyncer);
    }
}
Also used : YarnUncaughtExceptionHandler(org.apache.hadoop.yarn.YarnUncaughtExceptionHandler) InetSocketAddress(java.net.InetSocketAddress) UserGroupInformation(org.apache.hadoop.security.UserGroupInformation) ScheduledExecutorService(java.util.concurrent.ScheduledExecutorService) FSError(org.apache.hadoop.fs.FSError) JobTokenIdentifier(org.apache.hadoop.mapreduce.security.token.JobTokenIdentifier) PrivilegedExceptionAction(java.security.PrivilegedExceptionAction) IOException(java.io.IOException) DiskErrorException(org.apache.hadoop.util.DiskChecker.DiskErrorException) Credentials(org.apache.hadoop.security.Credentials)

Example 2 with YarnUncaughtExceptionHandler

use of org.apache.hadoop.yarn.YarnUncaughtExceptionHandler in project hadoop by apache.

the class ContainerLocalizer method main.

public static void main(String[] argv) throws Throwable {
    Thread.setDefaultUncaughtExceptionHandler(new YarnUncaughtExceptionHandler());
    int nRet = 0;
    // LOAD $x/$user/appcache/$appid/appTokens
    try {
        String user = argv[0];
        String appId = argv[1];
        String locId = argv[2];
        InetSocketAddress nmAddr = new InetSocketAddress(argv[3], Integer.parseInt(argv[4]));
        String[] sLocaldirs = Arrays.copyOfRange(argv, 5, argv.length);
        ArrayList<Path> localDirs = new ArrayList<Path>(sLocaldirs.length);
        for (String sLocaldir : sLocaldirs) {
            localDirs.add(new Path(sLocaldir));
        }
        final String uid = UserGroupInformation.getCurrentUser().getShortUserName();
        if (!user.equals(uid)) {
            // TODO: fail localization
            LOG.warn("Localization running as " + uid + " not " + user);
        }
        ContainerLocalizer localizer = new ContainerLocalizer(FileContext.getLocalFSFileContext(), user, appId, locId, localDirs, RecordFactoryProvider.getRecordFactory(null));
        localizer.runLocalization(nmAddr);
        return;
    } catch (Throwable e) {
        // Print traces to stdout so that they can be logged by the NM address
        // space in both DefaultCE and LCE cases
        e.printStackTrace(System.out);
        LOG.error("Exception in main:", e);
        nRet = -1;
    } finally {
        System.exit(nRet);
    }
}
Also used : Path(org.apache.hadoop.fs.Path) YarnUncaughtExceptionHandler(org.apache.hadoop.yarn.YarnUncaughtExceptionHandler) InetSocketAddress(java.net.InetSocketAddress) ArrayList(java.util.ArrayList)

Example 3 with YarnUncaughtExceptionHandler

use of org.apache.hadoop.yarn.YarnUncaughtExceptionHandler in project hadoop by apache.

the class NodeManager method main.

public static void main(String[] args) throws IOException {
    Thread.setDefaultUncaughtExceptionHandler(new YarnUncaughtExceptionHandler());
    StringUtils.startupShutdownMessage(NodeManager.class, args, LOG);
    @SuppressWarnings("resource") NodeManager nodeManager = new NodeManager();
    Configuration conf = new YarnConfiguration();
    new GenericOptionsParser(conf, args);
    nodeManager.initAndStartNodeManager(conf, false);
}
Also used : YarnUncaughtExceptionHandler(org.apache.hadoop.yarn.YarnUncaughtExceptionHandler) Configuration(org.apache.hadoop.conf.Configuration) YarnConfiguration(org.apache.hadoop.yarn.conf.YarnConfiguration) YarnConfiguration(org.apache.hadoop.yarn.conf.YarnConfiguration) GenericOptionsParser(org.apache.hadoop.util.GenericOptionsParser)

Example 4 with YarnUncaughtExceptionHandler

use of org.apache.hadoop.yarn.YarnUncaughtExceptionHandler in project hadoop by apache.

the class JobHistoryServer method launchJobHistoryServer.

static JobHistoryServer launchJobHistoryServer(String[] args) {
    Thread.setDefaultUncaughtExceptionHandler(new YarnUncaughtExceptionHandler());
    StringUtils.startupShutdownMessage(JobHistoryServer.class, args, LOG);
    JobHistoryServer jobHistoryServer = null;
    try {
        jobHistoryServer = new JobHistoryServer();
        ShutdownHookManager.get().addShutdownHook(new CompositeServiceShutdownHook(jobHistoryServer), SHUTDOWN_HOOK_PRIORITY);
        YarnConfiguration conf = new YarnConfiguration(new JobConf());
        new GenericOptionsParser(conf, args);
        jobHistoryServer.init(conf);
        jobHistoryServer.start();
    } catch (Throwable t) {
        LOG.fatal("Error starting JobHistoryServer", t);
        ExitUtil.terminate(-1, "Error starting JobHistoryServer");
    }
    return jobHistoryServer;
}
Also used : YarnUncaughtExceptionHandler(org.apache.hadoop.yarn.YarnUncaughtExceptionHandler) YarnConfiguration(org.apache.hadoop.yarn.conf.YarnConfiguration) JobConf(org.apache.hadoop.mapred.JobConf) GenericOptionsParser(org.apache.hadoop.util.GenericOptionsParser)

Example 5 with YarnUncaughtExceptionHandler

use of org.apache.hadoop.yarn.YarnUncaughtExceptionHandler in project hadoop by apache.

the class TimelineReaderServer method startTimelineReaderServer.

static TimelineReaderServer startTimelineReaderServer(String[] args, Configuration conf) {
    Thread.setDefaultUncaughtExceptionHandler(new YarnUncaughtExceptionHandler());
    StringUtils.startupShutdownMessage(TimelineReaderServer.class, args, LOG);
    TimelineReaderServer timelineReaderServer = null;
    try {
        timelineReaderServer = new TimelineReaderServer();
        ShutdownHookManager.get().addShutdownHook(new CompositeServiceShutdownHook(timelineReaderServer), SHUTDOWN_HOOK_PRIORITY);
        timelineReaderServer.init(conf);
        timelineReaderServer.start();
    } catch (Throwable t) {
        LOG.fatal("Error starting TimelineReaderWebServer", t);
        ExitUtil.terminate(-1, "Error starting TimelineReaderWebServer");
    }
    return timelineReaderServer;
}
Also used : YarnUncaughtExceptionHandler(org.apache.hadoop.yarn.YarnUncaughtExceptionHandler)

Aggregations

YarnUncaughtExceptionHandler (org.apache.hadoop.yarn.YarnUncaughtExceptionHandler)11 YarnConfiguration (org.apache.hadoop.yarn.conf.YarnConfiguration)7 GenericOptionsParser (org.apache.hadoop.util.GenericOptionsParser)5 Configuration (org.apache.hadoop.conf.Configuration)3 InetSocketAddress (java.net.InetSocketAddress)2 Path (org.apache.hadoop.fs.Path)2 JobConf (org.apache.hadoop.mapred.JobConf)2 VisibleForTesting (com.google.common.annotations.VisibleForTesting)1 IOException (java.io.IOException)1 PrivilegedExceptionAction (java.security.PrivilegedExceptionAction)1 ArrayList (java.util.ArrayList)1 ScheduledExecutorService (java.util.concurrent.ScheduledExecutorService)1 FSError (org.apache.hadoop.fs.FSError)1 JobTokenIdentifier (org.apache.hadoop.mapreduce.security.token.JobTokenIdentifier)1 Credentials (org.apache.hadoop.security.Credentials)1 UserGroupInformation (org.apache.hadoop.security.UserGroupInformation)1 DiskErrorException (org.apache.hadoop.util.DiskChecker.DiskErrorException)1 ApplicationAttemptId (org.apache.hadoop.yarn.api.records.ApplicationAttemptId)1 ContainerId (org.apache.hadoop.yarn.api.records.ContainerId)1