org.apache.hadoop.yarn.api.protocolrecords.FinishApplicationMasterRequest类的使用及代码示例

本文整理了Java中org.apache.hadoop.yarn.api.protocolrecords.FinishApplicationMasterRequest类的一些代码示例，展示了FinishApplicationMasterRequest类的具体用法。这些代码示例主要来源于Github/Stackoverflow/Maven等平台，是从一些精选项目中提取出来的代码，具有较强的参考意义，能在一定程度帮忙到你。FinishApplicationMasterRequest类的具体详情如下：
包路径：org.apache.hadoop.yarn.api.protocolrecords.FinishApplicationMasterRequest
类名称：FinishApplicationMasterRequest

FinishApplicationMasterRequest介绍

[英]The finalization request sent by the ApplicationMaster to inform the ResourceManager about its completion.

The final request includes details such:

• Final state of the ApplicationMaster
• Diagnostic information in case of failure of the ApplicationMaster
• Tracking URL
  [中]ApplicationMaster发送的用于通知ResourceManager其完成情况的终结请求。
  最终请求包括以下详细信息：
  *ApplicationMaster的最终状态
  *ApplicationMaster出现故障时的诊断信息
  *跟踪URL

代码示例

代码示例来源：origin: org.apache.hadoop/hadoop-yarn-server-resourcemanager

public void unregisterAppAttempt(boolean waitForStateRunning)
throws Exception {
final FinishApplicationMasterRequest req =
FinishApplicationMasterRequest.newInstance(
FinalApplicationStatus.SUCCEEDED, "", "");
unregisterAppAttempt(req, waitForStateRunning);
}

代码示例来源：origin: org.apache.hadoop/hadoop-yarn-api

@Public
@Stable
public static FinishApplicationMasterRequest newInstance(
FinalApplicationStatus finalAppStatus, String diagnostics, String url) {
FinishApplicationMasterRequest request =
Records.newRecord(FinishApplicationMasterRequest.class);
request.setFinalApplicationStatus(finalAppStatus);
request.setDiagnostics(diagnostics);
request.setTrackingUrl(url);
return request;
}

代码示例来源：origin: org.apache.apex/apex-engine

private void finishApplication(FinalApplicationStatus finalStatus) throws YarnException, IOException
{
LOG.info("Application completed. Signalling finish to RM");
FinishApplicationMasterRequest finishReq = Records.newRecord(FinishApplicationMasterRequest.class);
finishReq.setFinalApplicationStatus(finalStatus);
if (finalStatus != FinalApplicationStatus.SUCCEEDED) {
String diagnostics = "Diagnostics." + " completed=" + numCompletedContainers.get() + ", allocated=" + allocatedContainers.size() + ", failed=" + numFailedContainers.get();
if (!StringUtils.isEmpty(dnmgr.shutdownDiagnosticsMessage)) {
diagnostics += "\n";
diagnostics += dnmgr.shutdownDiagnosticsMessage;
}
// YARN-208 - as of 2.0.1-alpha dropped by the RM
finishReq.setDiagnostics(diagnostics);
// expected termination of the master process
// application status and diagnostics message are set above
}
LOG.info("diagnostics: " + finishReq.getDiagnostics());
amRmClient.unregisterApplicationMaster(finishReq.getFinalApplicationStatus(), finishReq.getDiagnostics(), null);
}

代码示例来源：origin: org.deeplearning4j/cdh4

public void finishApplication(String diagnostics,
FinalApplicationStatus finishState) throws YarnRemoteException {

if (amResourceManager == null)
throw new IllegalStateException(
"Cannot finish an application without connecting to resource manager!");
FinishApplicationMasterRequest request = Records.newRecord(FinishApplicationMasterRequest.class);
request.setAppAttemptId(appAttemptId);
request.setDiagnostics(diagnostics);
request.setFinishApplicationStatus(finishState);
LOG.info("Sending finish application notification "
+ ", state=" + request.getFinalApplicationStatus()
+ ", diagnostics=" + request.getDiagnostics());

amResourceManager.finishApplicationMaster(request);
}
}

代码示例来源：origin: org.apache.twill/twill-yarn

@Override
public void unregisterApplicationMaster(FinalApplicationStatus appStatus,
String appMessage, String appTrackingUrl) throws YarnRemoteException {
FinishApplicationMasterRequest request = recordFactory
.newRecordInstance(FinishApplicationMasterRequest.class);
request.setAppAttemptId(appAttemptId);
request.setFinishApplicationStatus(appStatus);
if (appMessage != null) {
request.setDiagnostics(appMessage);
}
if (appTrackingUrl != null) {
request.setTrackingUrl(appTrackingUrl);
}
rmClient.finishApplicationMaster(request);
}

代码示例来源：origin: org.apache.hadoop/hadoop-yarn-server-resourcemanager

@Override
public void finishApplicationMaster(
ApplicationAttemptId applicationAttemptId,
FinishApplicationMasterRequest request,
FinishApplicationMasterResponse response) {
RMApp app =
getRmContext().getRMApps().get(applicationAttemptId.getApplicationId());
// For UnmanagedAMs, return true so they don't retry
response.setIsUnregistered(
app.getApplicationSubmissionContext().getUnmanagedAM());
getRmContext().getDispatcher().getEventHandler().handle(
new RMAppAttemptUnregistrationEvent(applicationAttemptId, request
.getTrackingUrl(), request.getFinalApplicationStatus(), request
.getDiagnostics()));
}

代码示例来源：origin: org.springframework.data/spring-yarn-core

finishReq.setFinalApplicationStatus(status);
return rmTemplate.finish(finishReq);

代码示例来源：origin: io.hops/hadoop-yarn-server-nodemanager

/**
* This is called by the AMs started on this node to unregister from the RM.
* This method does the initial authorization and then forwards the request to
* the application instance specific intercepter chain.
*/
@Override
public FinishApplicationMasterResponse finishApplicationMaster(
FinishApplicationMasterRequest request) throws YarnException,
IOException {
LOG.info("Finishing application master. Tracking Url:"
+ request.getTrackingUrl());
RequestInterceptorChainWrapper pipeline =
authorizeAndGetInterceptorChain();
return pipeline.getRootInterceptor().finishApplicationMaster(request);
}

代码示例来源：origin: org.apache.hadoop/hadoop-yarn-server-common

@Override
public FinishApplicationMasterResponse finishApplicationMaster(
FinishApplicationMasterRequest request)
throws YarnException, IOException {
validateRunning();
ApplicationAttemptId attemptId = getAppIdentifier();
LOG.info("Finishing application attempt: " + attemptId);
ApplicationId appId = attemptId.getApplicationId();
if (shouldReRegisterNext) {
String message = "AM is not registered, should re-register.";
LOG.warn(message);
throw new ApplicationMasterNotRegisteredException(message);
}
synchronized (applicationContainerIdMap) {
// Remove the containers that were being tracked for this application
Assert.assertTrue("The application id is NOT registered: " + attemptId,
applicationContainerIdMap.containsKey(appId));
applicationContainerIdMap.remove(appId);
}
return FinishApplicationMasterResponse.newInstance(
request.getFinalApplicationStatus() == FinalApplicationStatus.SUCCEEDED
? true : false);
}

代码示例来源：origin: ch.cern.hadoop/hadoop-yarn-server-resourcemanager

.getTrackingUrl(), request.getFinalApplicationStatus(), request
.getDiagnostics()));

代码示例来源：origin: ch.cern.hadoop/hadoop-sls

finishAMRequest.setFinalApplicationStatus(FinalApplicationStatus.SUCCEEDED);

代码示例来源：origin: org.apache.hadoop/hadoop-yarn-server-nodemanager

/**
* This is called by the AMs started on this node to unregister from the RM.
* This method does the initial authorization and then forwards the request to
* the application instance specific intercepter chain.
*/
@Override
public FinishApplicationMasterResponse finishApplicationMaster(
FinishApplicationMasterRequest request) throws YarnException,
IOException {
LOG.info("Finishing application master. Tracking Url:"
+ request.getTrackingUrl());
RequestInterceptorChainWrapper pipeline =
authorizeAndGetInterceptorChain();
return pipeline.getRootInterceptor().finishApplicationMaster(request);
}

代码示例来源：origin: io.hops/hadoop-yarn-api

@Public
@Stable
public static FinishApplicationMasterRequest newInstance(
FinalApplicationStatus finalAppStatus, String diagnostics, String url) {
FinishApplicationMasterRequest request =
Records.newRecord(FinishApplicationMasterRequest.class);
request.setFinalApplicationStatus(finalAppStatus);
request.setDiagnostics(diagnostics);
request.setTrackingUrl(url);
return request;
}

代码示例来源：origin: ch.cern.hadoop/hadoop-yarn-server-resourcemanager

private void finishApplicationMaster(RMApp rmApp, MockRM rm, MockNM nm,
MockAM am) throws Exception {
final FinishApplicationMasterRequest req =
FinishApplicationMasterRequest.newInstance(
FinalApplicationStatus.SUCCEEDED, "", "");
finishApplicationMaster(rmApp, rm, nm, am, req);
}

代码示例来源：origin: com.github.jiayuhan-it/hadoop-yarn-server-resourcemanager

.getTrackingUrl(), request.getFinalApplicationStatus(), request
.getDiagnostics()));

代码示例来源：origin: ch.cern.hadoop/hadoop-yarn-api

@Public
@Stable
public static FinishApplicationMasterRequest newInstance(
FinalApplicationStatus finalAppStatus, String diagnostics, String url) {
FinishApplicationMasterRequest request =
Records.newRecord(FinishApplicationMasterRequest.class);
request.setFinalApplicationStatus(finalAppStatus);
request.setDiagnostics(diagnostics);
request.setTrackingUrl(url);
return request;
}

代码示例来源：origin: ch.cern.hadoop/hadoop-yarn-server-resourcemanager

public void unregisterAppAttempt(boolean waitForStateRunning)
throws Exception {
final FinishApplicationMasterRequest req =
FinishApplicationMasterRequest.newInstance(
FinalApplicationStatus.SUCCEEDED, "", "");
unregisterAppAttempt(req, waitForStateRunning);
}

代码示例来源：origin: com.github.jiayuhan-it/hadoop-yarn-api

@Public
@Stable
public static FinishApplicationMasterRequest newInstance(
FinalApplicationStatus finalAppStatus, String diagnostics, String url) {
FinishApplicationMasterRequest request =
Records.newRecord(FinishApplicationMasterRequest.class);
request.setFinalApplicationStatus(finalAppStatus);
request.setDiagnostics(diagnostics);
request.setTrackingUrl(url);
return request;
}

代码示例来源：origin: io.hops/hadoop-yarn-client

"AppStatus should not be null.");
FinishApplicationMasterRequest request =
FinishApplicationMasterRequest.newInstance(appStatus, appMessage,
appTrackingUrl);
try {

代码示例来源：origin: org.apache.hadoop/hadoop-yarn-server-resourcemanager

Records.newRecord(FinishApplicationMasterRequest.class);
finishAMRequest
.setFinalApplicationStatus(FinalApplicationStatus.SUCCEEDED);
finishAMRequest.setDiagnostics("diagnostics");
finishAMRequest.setTrackingUrl("url");
rmClient.finishApplicationMaster(finishAMRequest);