• 设为首页
  • 点击收藏
  • 手机版
    手机扫一扫访问
    迪恩网络手机版
  • 关注官方公众号
    微信扫一扫关注
    迪恩网络公众号

Java Keys类代码示例

原作者: [db:作者] 来自: [db:来源] 收藏 邀请

本文整理汇总了Java中org.apache.hadoop.mapred.JobHistory.Keys的典型用法代码示例。如果您正苦于以下问题:Java Keys类的具体用法?Java Keys怎么用?Java Keys使用的例子?那么恭喜您, 这里精选的类代码示例或许可以为您提供帮助。



Keys类属于org.apache.hadoop.mapred.JobHistory包,在下文中一共展示了Keys类的14个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于我们的系统推荐出更棒的Java代码示例。

示例1: updateTip

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
private void updateTip(TaskInProgress tip, JobHistory.Task task) {
  long startTime = task.getLong(Keys.START_TIME);
  if (startTime != 0) {
    tip.setExecStartTime(startTime);
  }
  
  long finishTime = task.getLong(Keys.FINISH_TIME);
  // For failed tasks finish-time will be missing
  if (finishTime != 0) {
    tip.setExecFinishTime(finishTime);
  }
  
  String cause = task.get(Keys.TASK_ATTEMPT_ID);
  if (cause.length() > 0) {
    // This means that the this is a FAILED events
    TaskAttemptID id = TaskAttemptID.forName(cause);
    TaskStatus status = tip.getTaskStatus(id);
    synchronized (JobTracker.this) {
      // This will add the tip failed event in the new log
      tip.getJob().failedTask(tip, id, status.getDiagnosticInfo(), 
                              status.getPhase(), status.getRunState(), 
                              status.getTaskTracker());
    }
  }
}
 
开发者ID:Nextzero,项目名称:hadoop-2.6.0-cdh5.4.3,代码行数:26,代码来源:JobTracker.java


示例2: handle

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
public void handle(JobHistory.RecordTypes recType, Map<Keys, String> values)
  throws IOException {
  if (recType.equals(JobHistory.RecordTypes.MapAttempt) || 
      recType.equals(JobHistory.RecordTypes.ReduceAttempt)) {
    if (failureType.equals(values.get(Keys.TASK_STATUS)) ) {
      String hostName = values.get(Keys.HOSTNAME);
      String taskid = values.get(Keys.TASKID); 
      Set<String> tasks = badNodesToNumFailedTasks.get(hostName); 
      if (null == tasks ){
        tasks = new TreeSet<String>(); 
        tasks.add(taskid);
        badNodesToNumFailedTasks.put(hostName, tasks);
      }else{
        tasks.add(taskid);
      }
    }
  }      
}
 
开发者ID:Nextzero,项目名称:hadoop-2.6.0-cdh5.4.3,代码行数:19,代码来源:DefaultJobHistoryParser.java


示例3: getLastSuccessfulTaskAttempt

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
private java.util.Map<JobHistory.Keys, String> getLastSuccessfulTaskAttempt(JobHistory.Task task) {
  
  Map<String, JobHistory.TaskAttempt> taskAttempts = task.getTaskAttempts();
  int size = taskAttempts.size();
  java.util.Iterator<Map.Entry<String, JobHistory.TaskAttempt>> kv = taskAttempts.entrySet().iterator();
  for (int i=0; i<size; i++) {
    // CHECK_IT: Only one SUCCESSFUL TASK ATTEMPT
    Map.Entry<String, JobHistory.TaskAttempt> tae = kv.next();
    JobHistory.TaskAttempt attempt = tae.getValue();
    if (attempt.getValues().get(JobHistory.Keys.TASK_STATUS).equals("SUCCESS")) {
      return attempt.getValues();
    }
  }
  
  return null;
}
 
开发者ID:Nextzero,项目名称:hadoop-2.6.0-cdh5.4.3,代码行数:17,代码来源:JobStatistics.java


示例4: logInited

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
/**
 * Logs launch time of job.
 *
 * @param startTime start time of job.
 * @param totalMaps total maps assigned by jobtracker.
 * @param totalReduces total reduces.
 */
public void logInited(long startTime, int totalMaps, int totalReduces) {
  if (disableHistory) {
    return;
  }

  if (null != writers) {
    log(writers, RecordTypes.Job,
        new Keys[] {Keys.JOBID, Keys.LAUNCH_TIME, Keys.TOTAL_MAPS,
                    Keys.TOTAL_REDUCES, Keys.JOB_STATUS},
        new String[] {jobId.toString(), String.valueOf(startTime),
                      String.valueOf(totalMaps),
                      String.valueOf(totalReduces),
                      Values.PREP.name()});
  }
}
 
开发者ID:rhli,项目名称:hadoop-EAR,代码行数:23,代码来源:CoronaJobHistory.java


示例5: logFailed

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
/**
 * Logs job failed event. Closes the job history log file.
 * @param timestamp time when job failure was detected in ms.
 * @param finishedMaps no finished map tasks.
 * @param finishedReduces no of finished reduce tasks.
 */
public void logFailed(long timestamp, int finishedMaps,
                      int finishedReduces, Counters counters) {
  if (disableHistory) {
    return;
  }

  if (null != writers) {
    log(writers, RecordTypes.Job,
        new Keys[] {Keys.JOBID, Keys.FINISH_TIME,
                    Keys.JOB_STATUS, Keys.FINISHED_MAPS,
                    Keys.FINISHED_REDUCES, Keys.COUNTERS},
        new String[] {jobId.toString(),
                      String.valueOf(timestamp),
                      Values.FAILED.name(),
                      String.valueOf(finishedMaps),
                      String.valueOf(finishedReduces),
                      counters.makeEscapedCompactString()},
                      true);
    closeAndClear(writers);
  }
}
 
开发者ID:rhli,项目名称:hadoop-EAR,代码行数:28,代码来源:CoronaJobHistory.java


示例6: logKilled

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
/**
 * Logs job killed event. Closes the job history log file.
 *
 * @param timestamp
 *          time when job killed was issued in ms.
 * @param finishedMaps
 *          no finished map tasks.
 * @param finishedReduces
 *          no of finished reduce tasks.
 */
public void logKilled(long timestamp, int finishedMaps,
                      int finishedReduces, Counters counters) {
  if (disableHistory) {
    return;
  }

  if (null != writers) {
    log(writers, RecordTypes.Job,
        new Keys[] {Keys.JOBID,
                    Keys.FINISH_TIME, Keys.JOB_STATUS, Keys.FINISHED_MAPS,
                    Keys.FINISHED_REDUCES, Keys.COUNTERS },
        new String[] {jobId.toString(),
                      String.valueOf(timestamp), Values.KILLED.name(),
                      String.valueOf(finishedMaps),
                      String.valueOf(finishedReduces),
                      counters.makeEscapedCompactString()},
                      true);
    closeAndClear(writers);
  }
}
 
开发者ID:rhli,项目名称:hadoop-EAR,代码行数:31,代码来源:CoronaJobHistory.java


示例7: logTaskStarted

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
/**
 * Log start time of task (TIP).
 * @param taskId task id
 * @param taskType MAP or REDUCE
 * @param startTime startTime of tip.
 */
public void logTaskStarted(TaskID taskId, String taskType,
                           long startTime, String splitLocations) {
  if (disableHistory) {
    return;
  }

  JobID id = taskId.getJobID();
  if (!this.jobId.equals(id)) {
    throw new RuntimeException("JobId from task: " + id +
                               " does not match expected: " + jobId);
  }

  if (null != writers) {
    log(writers, RecordTypes.Task,
        new Keys[]{Keys.TASKID, Keys.TASK_TYPE ,
                   Keys.START_TIME, Keys.SPLITS},
        new String[]{taskId.toString(), taskType,
                     String.valueOf(startTime),
                     splitLocations});
  }
}
 
开发者ID:rhli,项目名称:hadoop-EAR,代码行数:28,代码来源:CoronaJobHistory.java


示例8: logTaskFinished

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
/**
 * Log finish time of task.
 * @param taskId task id
 * @param taskType MAP or REDUCE
 * @param finishTime finish timeof task in ms
 */
public void logTaskFinished(TaskID taskId, String taskType,
                            long finishTime, Counters counters) {

  if (disableHistory) {
    return;
  }

  JobID id = taskId.getJobID();
  if (!this.jobId.equals(id)) {
    throw new RuntimeException("JobId from task: " + id +
        " does not match expected: " + jobId);
  }

  if (null != writers) {
    log(writers, RecordTypes.Task,
        new Keys[]{Keys.TASKID, Keys.TASK_TYPE,
                   Keys.TASK_STATUS, Keys.FINISH_TIME,
                   Keys.COUNTERS},
        new String[]{ taskId.toString(), taskType, Values.SUCCESS.name(),
                      String.valueOf(finishTime),
                      counters.makeEscapedCompactString()});
  }
}
 
开发者ID:rhli,项目名称:hadoop-EAR,代码行数:30,代码来源:CoronaJobHistory.java


示例9: logTaskUpdates

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
/**
 * Update the finish time of task.
 * @param taskId task id
 * @param finishTime finish time of task in ms
 */
public void logTaskUpdates(TaskID taskId, long finishTime) {
  if (disableHistory) {
    return;
  }

  JobID id = taskId.getJobID();
  if (!this.jobId.equals(id)) {
    throw new RuntimeException("JobId from task: " + id +
                               " does not match expected: " + jobId);
  }

  if (null != writers) {
    log(writers, RecordTypes.Task,
        new Keys[]{Keys.TASKID, Keys.FINISH_TIME},
        new String[]{ taskId.toString(),
                      String.valueOf(finishTime)});
  }
}
 
开发者ID:rhli,项目名称:hadoop-EAR,代码行数:24,代码来源:CoronaJobHistory.java


示例10: log

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
/**
 * Log a number of keys and values with the record. This method allows to do
 * it in a synchronous fashion
 * @param writers the writers to send the data to
 * @param recordType the type to log
 * @param keys keys to log
 * @param values values to log
 * @param sync if true - will block until the data is written
 */
private void log(ArrayList<PrintWriter> writers, RecordTypes recordType,
      Keys[] keys, String[] values, boolean sync) {
  StringBuffer buf = new StringBuffer(recordType.name());
  buf.append(JobHistory.DELIMITER);
  for (int i = 0; i < keys.length; i++) {
    buf.append(keys[i]);
    buf.append("=\"");
    values[i] = JobHistory.escapeString(values[i]);
    buf.append(values[i]);
    buf.append("\"");
    buf.append(JobHistory.DELIMITER);
  }
  buf.append(JobHistory.LINE_DELIMITER_CHAR);

  for (PrintWriter out : writers) {
    LogTask task = new LogTask(out, buf.toString());
    if (sync) {
      task.run();
    } else {
      fileManager.addWriteTask(task);
    }
  }
}
 
开发者ID:rhli,项目名称:hadoop-EAR,代码行数:33,代码来源:CoronaJobHistory.java


示例11: getLastSuccessfulTaskAttempt

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
private static Map<JobHistory.Keys, String> getLastSuccessfulTaskAttempt(
        JobHistory.Task task) {

    Map<String, JobHistory.TaskAttempt> taskAttempts = task
            .getTaskAttempts();
    int size = taskAttempts.size();
    Iterator<Map.Entry<String, JobHistory.TaskAttempt>> kv = taskAttempts
            .entrySet().iterator();
    for (int i = 0; i < size; i++) {
        // CHECK_IT: Only one SUCCESSFUL TASK ATTEMPT
        Map.Entry<String, JobHistory.TaskAttempt> tae = kv.next();
        JobHistory.TaskAttempt attempt = tae.getValue();
        if (null != attempt && null != attempt.getValues() && attempt.getValues().containsKey(JobHistory.Keys.TASK_STATUS) && attempt.getValues().get(JobHistory.Keys.TASK_STATUS).equals(
                "SUCCESS")) {
            return attempt.getValues();
        }
    }

    return null;
}
 
开发者ID:sigmoidanalytics,项目名称:spork-streaming,代码行数:21,代码来源:HadoopJobHistoryLoader.java


示例12: processTaskAttempt

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
/**
 * Adds a task-attempt in the listener
 */
private void processTaskAttempt(String taskAttemptId, 
                                JobHistory.TaskAttempt attempt) {
  TaskAttemptID id = TaskAttemptID.forName(taskAttemptId);
  
  // Check if the transaction for this attempt can be committed
  String taskStatus = attempt.get(Keys.TASK_STATUS);
  
  if (taskStatus.length() > 0) {
    // This means this is an update event
    if (taskStatus.equals(Values.SUCCESS.name())) {
      // Mark this attempt as hanging
      hangingAttempts.put(id.getTaskID().toString(), taskAttemptId);
      addSuccessfulAttempt(jip, id, attempt);
    } else {
      addUnsuccessfulAttempt(jip, id, attempt);
      numEventsRecovered += 2;
    }
  } else {
    createTaskAttempt(jip, id, attempt);
  }
}
 
开发者ID:thisisvoa,项目名称:hadoop-0.20,代码行数:25,代码来源:JobTracker.java


示例13: updateJob

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
private JobStatusChangeEvent updateJob(JobInProgress jip, 
                                       JobHistory.JobInfo job) {
  // Change the job priority
  String jobpriority = job.get(Keys.JOB_PRIORITY);
  JobPriority priority = JobPriority.valueOf(jobpriority);
  // It's important to update this via the jobtracker's api as it will 
  // take care of updating the event listeners too
  setJobPriority(jip.getJobID(), priority);

  // Save the previous job status
  JobStatus oldStatus = (JobStatus)jip.getStatus().clone();
  
  // Set the start/launch time only if there are recovered tasks
  // Increment the job's restart count
  jip.updateJobInfo(job.getLong(JobHistory.Keys.SUBMIT_TIME), 
                    job.getLong(JobHistory.Keys.LAUNCH_TIME));

  // Save the new job status
  JobStatus newStatus = (JobStatus)jip.getStatus().clone();
  
  return new JobStatusChangeEvent(jip, EventType.START_TIME_CHANGED, oldStatus, 
                                  newStatus);
}
 
开发者ID:thisisvoa,项目名称:hadoop-0.20,代码行数:24,代码来源:JobTracker.java


示例14: checkAndInit

import org.apache.hadoop.mapred.JobHistory.Keys; //导入依赖的package包/类
private void checkAndInit() throws IOException {
  String jobStatus = this.job.get(Keys.JOB_STATUS);
  if (Values.PREP.name().equals(jobStatus)) {
    hasUpdates = true;
    LOG.info("Calling init from RM for job " + jip.getJobID().toString());
    try {
      initJob(jip);
    } catch (Throwable t) {
      LOG.error("Job initialization failed : \n"
          + StringUtils.stringifyException(t));
      jip.status.setFailureInfo("Job Initialization failed: \n"
          + StringUtils.stringifyException(t));
      failJob(jip);
      throw new IOException(t);
    }
  }
}
 
开发者ID:mammothcm,项目名称:mammoth,代码行数:18,代码来源:JobTracker.java



注:本文中的org.apache.hadoop.mapred.JobHistory.Keys类示例整理自Github/MSDocs等源码及文档管理平台,相关代码片段筛选自各路编程大神贡献的开源项目,源码版权归原作者所有,传播和使用请参考对应项目的License;未经允许,请勿转载。


鲜花

握手

雷人

路过

鸡蛋
该文章已有0人参与评论

请发表评论

全部评论

专题导读
上一篇:
Java JcePublicKeyDataDecryptorFactoryBuilder类代码示例发布时间:2022-05-22
下一篇:
Java WorldGenTaiga1类代码示例发布时间:2022-05-22
热门推荐
阅读排行榜

扫描微信二维码

查看手机版网站

随时了解更新最新资讯

139-2527-9053

在线客服(服务时间 9:00~18:00)

在线QQ客服
地址:深圳市南山区西丽大学城创智工业园
电邮:jeky_zhao#qq.com
移动电话:139-2527-9053

Powered by 互联科技 X3.4© 2001-2213 极客世界.|Sitemap