本文整理汇总了Java中org.apache.hadoop.mapred.TaskAttemptID.forName方法的典型用法代码示例。如果您正苦于以下问题:Java TaskAttemptID.forName方法的具体用法?Java TaskAttemptID.forName怎么用?Java TaskAttemptID.forName使用的例子?那么, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类org.apache.hadoop.mapred.TaskAttemptID
的用法示例。
在下文中一共展示了TaskAttemptID.forName方法的9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,您的评价将有助于系统推荐出更棒的Java代码示例。
示例1: initStdOut
import org.apache.hadoop.mapred.TaskAttemptID; //导入方法依赖的package包/类
/**
* clean previous std error and outs
*/
private void initStdOut(JobConf configuration) {
TaskAttemptID taskId = TaskAttemptID.forName(configuration
.get(MRJobConfig.TASK_ATTEMPT_ID));
File stdOut = TaskLog.getTaskLogFile(taskId, false, TaskLog.LogName.STDOUT);
File stdErr = TaskLog.getTaskLogFile(taskId, false, TaskLog.LogName.STDERR);
// prepare folder
if (!stdOut.getParentFile().exists()) {
stdOut.getParentFile().mkdirs();
} else { // clean logs
stdOut.deleteOnExit();
stdErr.deleteOnExit();
}
}
示例2: readStdOut
import org.apache.hadoop.mapred.TaskAttemptID; //导入方法依赖的package包/类
private String readStdOut(JobConf conf) throws Exception {
TaskAttemptID taskId = TaskAttemptID.forName(conf
.get(MRJobConfig.TASK_ATTEMPT_ID));
File stdOut = TaskLog.getTaskLogFile(taskId, false, TaskLog.LogName.STDOUT);
return readFile(stdOut);
}
示例3: open
import org.apache.hadoop.mapred.TaskAttemptID; //导入方法依赖的package包/类
/**
* create the temporary output file for hadoop RecordWriter.
* @param taskNumber The number of the parallel instance.
* @param numTasks The number of parallel tasks.
* @throws java.io.IOException
*/
@Override
public void open(int taskNumber, int numTasks) throws IOException {
// enforce sequential open() calls
synchronized (OPEN_MUTEX) {
if (Integer.toString(taskNumber + 1).length() > 6) {
throw new IOException("Task id too large.");
}
TaskAttemptID taskAttemptID = TaskAttemptID.forName("attempt__0000_r_"
+ String.format("%" + (6 - Integer.toString(taskNumber + 1).length()) + "s", " ").replace(" ", "0")
+ Integer.toString(taskNumber + 1)
+ "_0");
this.jobConf.set("mapred.task.id", taskAttemptID.toString());
this.jobConf.setInt("mapred.task.partition", taskNumber + 1);
// for hadoop 2.2
this.jobConf.set("mapreduce.task.attempt.id", taskAttemptID.toString());
this.jobConf.setInt("mapreduce.task.partition", taskNumber + 1);
this.context = new TaskAttemptContextImpl(this.jobConf, taskAttemptID);
this.outputCommitter = this.jobConf.getOutputCommitter();
JobContext jobContext = new JobContextImpl(this.jobConf, new JobID());
this.outputCommitter.setupJob(jobContext);
this.recordWriter = this.mapredOutputFormat.getRecordWriter(null, this.jobConf, Integer.toString(taskNumber + 1), new HadoopDummyProgressable());
}
}
示例4: Application
import org.apache.hadoop.mapred.TaskAttemptID; //导入方法依赖的package包/类
/**
* Start the child process to handle the task for us.
* @param conf the task's configuration
* @param recordReader the fake record reader to update progress with
* @param output the collector to send output to
* @param reporter the reporter for the task
* @param outputKeyClass the class of the output keys
* @param outputValueClass the class of the output values
* @throws IOException
* @throws InterruptedException
*/
Application(JobConf conf,
RecordReader<FloatWritable, NullWritable> recordReader,
OutputCollector<K2,V2> output, Reporter reporter,
Class<? extends K2> outputKeyClass,
Class<? extends V2> outputValueClass
) throws IOException, InterruptedException {
serverSocket = new ServerSocket(0);
Map<String, String> env = new HashMap<String,String>();
// add TMPDIR environment variable with the value of java.io.tmpdir
env.put("TMPDIR", System.getProperty("java.io.tmpdir"));
env.put(Submitter.PORT,
Integer.toString(serverSocket.getLocalPort()));
//Add token to the environment if security is enabled
Token<JobTokenIdentifier> jobToken = TokenCache.getJobToken(conf
.getCredentials());
// This password is used as shared secret key between this application and
// child pipes process
byte[] password = jobToken.getPassword();
String localPasswordFile = new File(".") + Path.SEPARATOR
+ "jobTokenPassword";
writePasswordToLocalFile(localPasswordFile, password, conf);
env.put("hadoop.pipes.shared.secret.location", localPasswordFile);
List<String> cmd = new ArrayList<String>();
String interpretor = conf.get(Submitter.INTERPRETOR);
if (interpretor != null) {
cmd.add(interpretor);
}
String executable = DistributedCache.getLocalCacheFiles(conf)[0].toString();
if (!FileUtil.canExecute(new File(executable))) {
// LinuxTaskController sets +x permissions on all distcache files already.
// In case of DefaultTaskController, set permissions here.
FileUtil.chmod(executable, "u+x");
}
cmd.add(executable);
// wrap the command in a stdout/stderr capture
// we are starting map/reduce task of the pipes job. this is not a cleanup
// attempt.
TaskAttemptID taskid =
TaskAttemptID.forName(conf.get(MRJobConfig.TASK_ATTEMPT_ID));
File stdout = TaskLog.getTaskLogFile(taskid, false, TaskLog.LogName.STDOUT);
File stderr = TaskLog.getTaskLogFile(taskid, false, TaskLog.LogName.STDERR);
long logLength = TaskLog.getTaskLogLength(conf);
cmd = TaskLog.captureOutAndError(null, cmd, stdout, stderr, logLength,
false);
process = runClient(cmd, env);
clientSocket = serverSocket.accept();
String challenge = getSecurityChallenge();
String digestToSend = createDigest(password, challenge);
String digestExpected = createDigest(password, digestToSend);
handler = new OutputHandler<K2, V2>(output, reporter, recordReader,
digestExpected);
K2 outputKey = (K2)
ReflectionUtils.newInstance(outputKeyClass, conf);
V2 outputValue = (V2)
ReflectionUtils.newInstance(outputValueClass, conf);
downlink = new BinaryProtocol<K1, V1, K2, V2>(clientSocket, handler,
outputKey, outputValue, conf);
downlink.authenticate(digestToSend, challenge);
waitForAuthentication();
LOG.debug("Authentication succeeded");
downlink.start();
downlink.setJobConf(conf);
}
示例5: Application
import org.apache.hadoop.mapred.TaskAttemptID; //导入方法依赖的package包/类
/**
* Start the child process to handle the task for us.
* @param conf the task's configuration
* @param recordReader the fake record reader to update progress with
* @param output the collector to send output to
* @param reporter the reporter for the task
* @param outputKeyClass the class of the output keys
* @param outputValueClass the class of the output values
* @throws IOException
* @throws InterruptedException
*/
Application(JobConf conf,
RecordReader<FloatWritable, NullWritable> recordReader,
OutputCollector<K2,V2> output, Reporter reporter,
Class<? extends K2> outputKeyClass,
Class<? extends V2> outputValueClass
) throws IOException, InterruptedException {
serverSocket = new ServerSocket(0);
Map<String, String> env = new HashMap<String,String>();
// add TMPDIR environment variable with the value of java.io.tmpdir
env.put("TMPDIR", System.getProperty("java.io.tmpdir"));
env.put("hadoop.pipes.command.port",
Integer.toString(serverSocket.getLocalPort()));
//Add token to the environment if security is enabled
Token<JobTokenIdentifier> jobToken = TokenCache.getJobToken(conf
.getCredentials());
// This password is used as shared secret key between this application and
// child pipes process
byte[] password = jobToken.getPassword();
String localPasswordFile = new File(".") + Path.SEPARATOR
+ "jobTokenPassword";
writePasswordToLocalFile(localPasswordFile, password, conf);
env.put("hadoop.pipes.shared.secret.location", localPasswordFile);
List<String> cmd = new ArrayList<String>();
String interpretor = conf.get("hadoop.pipes.executable.interpretor");
if (interpretor != null) {
cmd.add(interpretor);
}
String executable = DistributedCache.getLocalCacheFiles(conf)[0].toString();
if (!new File(executable).canExecute()) {
// LinuxTaskController sets +x permissions on all distcache files already.
// In case of DefaultTaskController, set permissions here.
FileUtil.chmod(executable, "u+x");
}
cmd.add(executable);
// wrap the command in a stdout/stderr capture
TaskAttemptID taskid = TaskAttemptID.forName(conf.get("mapred.task.id"));
// we are starting map/reduce task of the pipes job. this is not a cleanup
// attempt.
File stdout = TaskLog.getTaskLogFile(taskid, false, TaskLog.LogName.STDOUT);
File stderr = TaskLog.getTaskLogFile(taskid, false, TaskLog.LogName.STDERR);
long logLength = TaskLog.getTaskLogLength(conf);
cmd = TaskLog.captureOutAndError(null, cmd, stdout, stderr, logLength,
false);
process = runClient(cmd, env);
clientSocket = serverSocket.accept();
String challenge = getSecurityChallenge();
String digestToSend = createDigest(password, challenge);
String digestExpected = createDigest(password, digestToSend);
handler = new OutputHandler<K2, V2>(output, reporter, recordReader,
digestExpected);
K2 outputKey = (K2)
ReflectionUtils.newInstance(outputKeyClass, conf);
V2 outputValue = (V2)
ReflectionUtils.newInstance(outputValueClass, conf);
downlink = new BinaryProtocol<K1, V1, K2, V2>(clientSocket, handler,
outputKey, outputValue, conf);
downlink.authenticate(digestToSend, challenge);
waitForAuthentication();
LOG.debug("Authentication succeeded");
downlink.start();
downlink.setJobConf(conf);
}
示例6: Application
import org.apache.hadoop.mapred.TaskAttemptID; //导入方法依赖的package包/类
/**
* Start the child process to handle the task for us.
* @param conf the task's configuration
* @param recordReader the fake record reader to update progress with
* @param output the collector to send output to
* @param reporter the reporter for the task
* @param outputKeyClass the class of the output keys
* @param outputValueClass the class of the output values
* @throws IOException
* @throws InterruptedException
*/
Application(JobConf conf,
RecordReader<FloatWritable, NullWritable> recordReader,
OutputCollector<K2,V2> output, Reporter reporter,
Class<? extends K2> outputKeyClass,
Class<? extends V2> outputValueClass
) throws IOException, InterruptedException {
serverSocket = new ServerSocket(0);
Map<String, String> env = new HashMap<String,String>();
// add TMPDIR environment variable with the value of java.io.tmpdir
env.put("TMPDIR", System.getProperty("java.io.tmpdir"));
env.put("hadoop.pipes.command.port",
Integer.toString(serverSocket.getLocalPort()));
List<String> cmd = new ArrayList<String>();
String interpretor = conf.get("hadoop.pipes.executable.interpretor");
if (interpretor != null) {
cmd.add(interpretor);
}
String executable = DistributedCache.getLocalCacheFiles(conf)[0].toString();
FileUtil.chmod(executable, "a+x");
cmd.add(executable);
// wrap the command in a stdout/stderr capture
TaskAttemptID taskid = TaskAttemptID.forName(conf.get("mapred.task.id"));
File stdout = TaskLog.getTaskLogFile(taskid, TaskLog.LogName.STDOUT);
File stderr = TaskLog.getTaskLogFile(taskid, TaskLog.LogName.STDERR);
long logLength = TaskLog.getTaskLogLength(conf);
cmd = TaskLog.captureOutAndError(null, cmd, stdout, stderr, logLength,
false);
process = runClient(cmd, env);
clientSocket = serverSocket.accept();
handler = new OutputHandler<K2, V2>(output, reporter, recordReader);
K2 outputKey = (K2)
ReflectionUtils.newInstance(outputKeyClass, conf);
V2 outputValue = (V2)
ReflectionUtils.newInstance(outputValueClass, conf);
downlink = new BinaryProtocol<K1, V1, K2, V2>(clientSocket, handler,
outputKey, outputValue, conf);
downlink.start();
downlink.setJobConf(conf);
}
示例7: Application
import org.apache.hadoop.mapred.TaskAttemptID; //导入方法依赖的package包/类
/**
* Start the child process to handle the task for us.
* @param conf the task's configuration
* @param recordReader the fake record reader to update progress with
* @param output the collector to send output to
* @param reporter the reporter for the task
* @param outputKeyClass the class of the output keys
* @param outputValueClass the class of the output values
* @throws IOException
* @throws InterruptedException
*/
Application(JobConf conf,
RecordReader<FloatWritable, NullWritable> recordReader,
OutputCollector<K2,V2> output, Reporter reporter,
Class<? extends K2> outputKeyClass,
Class<? extends V2> outputValueClass
) throws IOException, InterruptedException {
serverSocket = new ServerSocket(0);
Map<String, String> env = new HashMap<String,String>();
// add TMPDIR environment variable with the value of java.io.tmpdir
env.put("TMPDIR", System.getProperty("java.io.tmpdir"));
env.put("hadoop.pipes.command.port",
Integer.toString(serverSocket.getLocalPort()));
List<String> cmd = new ArrayList<String>();
String interpretor = conf.get("hadoop.pipes.executable.interpretor");
if (interpretor != null) {
cmd.add(interpretor);
}
String executable = DistributedCache.getLocalCacheFiles(conf)[0].toString();
FileUtil.chmod(executable, "a+x");
cmd.add(executable);
// wrap the command in a stdout/stderr capture
TaskAttemptID taskid = TaskAttemptID.forName(conf.get("mapred.task.id"));
File stdout = TaskLog.getTaskLogFile(taskid, TaskLog.LogName.STDOUT);
File stderr = TaskLog.getTaskLogFile(taskid, TaskLog.LogName.STDERR);
long logLength = TaskLog.getTaskLogLength(conf);
cmd = TaskLog.captureOutAndError(cmd, stdout, stderr, logLength);
process = runClient(cmd, env);
clientSocket = serverSocket.accept();
handler = new OutputHandler<K2, V2>(output, reporter, recordReader);
K2 outputKey = (K2)
ReflectionUtils.newInstance(outputKeyClass, conf);
V2 outputValue = (V2)
ReflectionUtils.newInstance(outputValueClass, conf);
downlink = new BinaryProtocol<K1, V1, K2, V2>(clientSocket, handler,
outputKey, outputValue, conf);
downlink.start();
downlink.setJobConf(conf);
}
示例8: Application
import org.apache.hadoop.mapred.TaskAttemptID; //导入方法依赖的package包/类
/**
* Start the child process to handle the task for us.
* @param conf the task's configuration
* @param recordReader the fake record reader to update progress with
* @param output the collector to send output to
* @param reporter the reporter for the task
* @param outputKeyClass the class of the output keys
* @param outputValueClass the class of the output values
* @throws IOException
* @throws InterruptedException
*/
Application(JobConf conf,
RecordReader<FloatWritable, NullWritable> recordReader,
OutputCollector<K2,V2> output, Reporter reporter,
Class<? extends K2> outputKeyClass,
Class<? extends V2> outputValueClass
) throws IOException, InterruptedException {
serverSocket = new ServerSocket(0);
Map<String, String> env = new HashMap<String,String>();
// add TMPDIR environment variable with the value of java.io.tmpdir
env.put("TMPDIR", System.getProperty("java.io.tmpdir"));
env.put(Submitter.PORT,
Integer.toString(serverSocket.getLocalPort()));
//Add token to the environment if security is enabled
Token<JobTokenIdentifier> jobToken = TokenCache.getJobToken(conf
.getCredentials());
// This password is used as shared secret key between this application and
// child pipes process
byte[] password = jobToken.getPassword();
String localPasswordFile = new File(".") + Path.SEPARATOR
+ "jobTokenPassword";
writePasswordToLocalFile(localPasswordFile, password, conf);
env.put("hadoop.pipes.shared.secret.location", localPasswordFile);
List<String> cmd = new ArrayList<String>();
String interpretor = conf.get(Submitter.INTERPRETOR);
if (interpretor != null) {
cmd.add(interpretor);
}
String executable = DistributedCache.getLocalCacheFiles(conf)[0].toString();
if (!new File(executable).canExecute()) {
// LinuxTaskController sets +x permissions on all distcache files already.
// In case of DefaultTaskController, set permissions here.
FileUtil.chmod(executable, "u+x");
}
cmd.add(executable);
// wrap the command in a stdout/stderr capture
// we are starting map/reduce task of the pipes job. this is not a cleanup
// attempt.
TaskAttemptID taskid =
TaskAttemptID.forName(conf.get(MRJobConfig.TASK_ATTEMPT_ID));
File stdout = TaskLog.getTaskLogFile(taskid, false, TaskLog.LogName.STDOUT);
File stderr = TaskLog.getTaskLogFile(taskid, false, TaskLog.LogName.STDERR);
long logLength = TaskLog.getTaskLogLength(conf);
cmd = TaskLog.captureOutAndError(null, cmd, stdout, stderr, logLength,
false);
process = runClient(cmd, env);
clientSocket = serverSocket.accept();
String challenge = getSecurityChallenge();
String digestToSend = createDigest(password, challenge);
String digestExpected = createDigest(password, digestToSend);
handler = new OutputHandler<K2, V2>(output, reporter, recordReader,
digestExpected);
K2 outputKey = (K2)
ReflectionUtils.newInstance(outputKeyClass, conf);
V2 outputValue = (V2)
ReflectionUtils.newInstance(outputValueClass, conf);
downlink = new BinaryProtocol<K1, V1, K2, V2>(clientSocket, handler,
outputKey, outputValue, conf);
downlink.authenticate(digestToSend, challenge);
waitForAuthentication();
LOG.debug("Authentication succeeded");
downlink.start();
downlink.setJobConf(conf);
}
示例9: Application
import org.apache.hadoop.mapred.TaskAttemptID; //导入方法依赖的package包/类
/**
* Start the child process to handle the task for us.
* @param conf the task's configuration
* @param recordReader the fake record reader to update progress with
* @param output the collector to send output to
* @param reporter the reporter for the task
* @param outputKeyClass the class of the output keys
* @param outputValueClass the class of the output values
* @param runOnGPU
* @throws IOException
* @throws InterruptedException
*/
Application(JobConf conf,
RecordReader<FloatWritable, NullWritable> recordReader,
OutputCollector<K2,V2> output, Reporter reporter,
Class<? extends K2> outputKeyClass,
Class<? extends V2> outputValueClass,
boolean runOnGPU
) throws IOException, InterruptedException {
serverSocket = new ServerSocket(0);
Map<String, String> env = new HashMap<String,String>();
// add TMPDIR environment variable with the value of java.io.tmpdir
env.put("TMPDIR", System.getProperty("java.io.tmpdir"));
env.put("hadoop.pipes.command.port",
Integer.toString(serverSocket.getLocalPort()));
List<String> cmd = new ArrayList<String>();
String interpretor = conf.get("hadoop.pipes.executable.interpretor");
if (interpretor != null) {
cmd.add(interpretor);
}
// Check whether the applicaiton will run on GPU
int i = runOnGPU ? 1 : 0;
String executable = DistributedCache.getLocalCacheFiles(conf)[i].toString();
FileUtil.chmod(executable, "a+x");
cmd.add(executable);
// wrap the command in a stdout/stderr capture
TaskAttemptID taskid = TaskAttemptID.forName(conf.get("mapred.task.id"));
File stdout = TaskLog.getTaskLogFile(taskid, TaskLog.LogName.STDOUT);
File stderr = TaskLog.getTaskLogFile(taskid, TaskLog.LogName.STDERR);
long logLength = TaskLog.getTaskLogLength(conf);
cmd = TaskLog.captureOutAndError(cmd, stdout, stderr, logLength);
process = runClient(cmd, env);
clientSocket = serverSocket.accept();
handler = new OutputHandler<K2, V2>(output, reporter, recordReader);
K2 outputKey = (K2)
ReflectionUtils.newInstance(outputKeyClass, conf);
V2 outputValue = (V2)
ReflectionUtils.newInstance(outputValueClass, conf);
downlink = new BinaryProtocol<K1, V1, K2, V2>(clientSocket, handler,
outputKey, outputValue, conf);
downlink.start();
downlink.setJobConf(conf);
}