use of org.apache.hadoop.yarn.api.records.LocalResource in project hadoop by apache.
the class Client method addToLocalResources.
private void addToLocalResources(FileSystem fs, String fileSrcPath, String fileDstPath, String appId, Map<String, LocalResource> localResources, String resources) throws IOException {
String suffix = appName + "/" + appId + "/" + fileDstPath;
Path dst = new Path(fs.getHomeDirectory(), suffix);
if (fileSrcPath == null) {
FSDataOutputStream ostream = null;
try {
ostream = FileSystem.create(fs, dst, new FsPermission((short) 0710));
ostream.writeUTF(resources);
} finally {
IOUtils.closeQuietly(ostream);
}
} else {
fs.copyFromLocalFile(new Path(fileSrcPath), dst);
}
FileStatus scFileStatus = fs.getFileStatus(dst);
LocalResource scRsrc = LocalResource.newInstance(URL.fromURI(dst.toUri()), LocalResourceType.FILE, LocalResourceVisibility.APPLICATION, scFileStatus.getLen(), scFileStatus.getModificationTime());
localResources.put(fileDstPath, scRsrc);
}
use of org.apache.hadoop.yarn.api.records.LocalResource in project hadoop by apache.
the class Client method run.
/**
* Main run function for the client
* @return true if application completed successfully
* @throws IOException
* @throws YarnException
*/
public boolean run() throws IOException, YarnException {
LOG.info("Running Client");
yarnClient.start();
YarnClusterMetrics clusterMetrics = yarnClient.getYarnClusterMetrics();
LOG.info("Got Cluster metric info from ASM" + ", numNodeManagers=" + clusterMetrics.getNumNodeManagers());
List<NodeReport> clusterNodeReports = yarnClient.getNodeReports(NodeState.RUNNING);
LOG.info("Got Cluster node info from ASM");
for (NodeReport node : clusterNodeReports) {
LOG.info("Got node report from ASM for" + ", nodeId=" + node.getNodeId() + ", nodeAddress=" + node.getHttpAddress() + ", nodeRackName=" + node.getRackName() + ", nodeNumContainers=" + node.getNumContainers());
}
QueueInfo queueInfo = yarnClient.getQueueInfo(this.amQueue);
LOG.info("Queue info" + ", queueName=" + queueInfo.getQueueName() + ", queueCurrentCapacity=" + queueInfo.getCurrentCapacity() + ", queueMaxCapacity=" + queueInfo.getMaximumCapacity() + ", queueApplicationCount=" + queueInfo.getApplications().size() + ", queueChildQueueCount=" + queueInfo.getChildQueues().size());
List<QueueUserACLInfo> listAclInfo = yarnClient.getQueueAclsInfo();
for (QueueUserACLInfo aclInfo : listAclInfo) {
for (QueueACL userAcl : aclInfo.getUserAcls()) {
LOG.info("User ACL Info for Queue" + ", queueName=" + aclInfo.getQueueName() + ", userAcl=" + userAcl.name());
}
}
if (domainId != null && domainId.length() > 0 && toCreateDomain) {
prepareTimelineDomain();
}
// Get a new application id
YarnClientApplication app = yarnClient.createApplication();
GetNewApplicationResponse appResponse = app.getNewApplicationResponse();
// TODO get min/max resource capabilities from RM and change memory ask if needed
// If we do not have min/max, we may not be able to correctly request
// the required resources from the RM for the app master
// Memory ask has to be a multiple of min and less than max.
// Dump out information about cluster capability as seen by the resource manager
long maxMem = appResponse.getMaximumResourceCapability().getMemorySize();
LOG.info("Max mem capability of resources in this cluster " + maxMem);
// A resource ask cannot exceed the max.
if (amMemory > maxMem) {
LOG.info("AM memory specified above max threshold of cluster. Using max value." + ", specified=" + amMemory + ", max=" + maxMem);
amMemory = maxMem;
}
int maxVCores = appResponse.getMaximumResourceCapability().getVirtualCores();
LOG.info("Max virtual cores capability of resources in this cluster " + maxVCores);
if (amVCores > maxVCores) {
LOG.info("AM virtual cores specified above max threshold of cluster. " + "Using max value." + ", specified=" + amVCores + ", max=" + maxVCores);
amVCores = maxVCores;
}
// set the application name
ApplicationSubmissionContext appContext = app.getApplicationSubmissionContext();
ApplicationId appId = appContext.getApplicationId();
appContext.setKeepContainersAcrossApplicationAttempts(keepContainers);
appContext.setApplicationName(appName);
if (attemptFailuresValidityInterval >= 0) {
appContext.setAttemptFailuresValidityInterval(attemptFailuresValidityInterval);
}
Set<String> tags = new HashSet<String>();
if (flowName != null) {
tags.add(TimelineUtils.generateFlowNameTag(flowName));
}
if (flowVersion != null) {
tags.add(TimelineUtils.generateFlowVersionTag(flowVersion));
}
if (flowRunId != 0) {
tags.add(TimelineUtils.generateFlowRunIdTag(flowRunId));
}
appContext.setApplicationTags(tags);
// set local resources for the application master
// local files or archives as needed
// In this scenario, the jar file for the application master is part of the local resources
Map<String, LocalResource> localResources = new HashMap<String, LocalResource>();
LOG.info("Copy App Master jar from local filesystem and add to local environment");
// Copy the application master jar to the filesystem
// Create a local resource to point to the destination jar path
FileSystem fs = FileSystem.get(conf);
addToLocalResources(fs, appMasterJar, appMasterJarPath, appId.toString(), localResources, null);
// Set the log4j properties if needed
if (!log4jPropFile.isEmpty()) {
addToLocalResources(fs, log4jPropFile, log4jPath, appId.toString(), localResources, null);
}
// The shell script has to be made available on the final container(s)
// where it will be executed.
// To do this, we need to first copy into the filesystem that is visible
// to the yarn framework.
// We do not need to set this as a local resource for the application
// master as the application master does not need it.
String hdfsShellScriptLocation = "";
long hdfsShellScriptLen = 0;
long hdfsShellScriptTimestamp = 0;
if (!shellScriptPath.isEmpty()) {
Path shellSrc = new Path(shellScriptPath);
String shellPathSuffix = appName + "/" + appId.toString() + "/" + SCRIPT_PATH;
Path shellDst = new Path(fs.getHomeDirectory(), shellPathSuffix);
fs.copyFromLocalFile(false, true, shellSrc, shellDst);
hdfsShellScriptLocation = shellDst.toUri().toString();
FileStatus shellFileStatus = fs.getFileStatus(shellDst);
hdfsShellScriptLen = shellFileStatus.getLen();
hdfsShellScriptTimestamp = shellFileStatus.getModificationTime();
}
if (!shellCommand.isEmpty()) {
addToLocalResources(fs, null, shellCommandPath, appId.toString(), localResources, shellCommand);
}
if (shellArgs.length > 0) {
addToLocalResources(fs, null, shellArgsPath, appId.toString(), localResources, StringUtils.join(shellArgs, " "));
}
// Set the necessary security tokens as needed
//amContainer.setContainerTokens(containerToken);
// Set the env variables to be setup in the env where the application master will be run
LOG.info("Set the environment for the application master");
Map<String, String> env = new HashMap<String, String>();
// put location of shell script into env
// using the env info, the application master will create the correct local resource for the
// eventual containers that will be launched to execute the shell scripts
env.put(DSConstants.DISTRIBUTEDSHELLSCRIPTLOCATION, hdfsShellScriptLocation);
env.put(DSConstants.DISTRIBUTEDSHELLSCRIPTTIMESTAMP, Long.toString(hdfsShellScriptTimestamp));
env.put(DSConstants.DISTRIBUTEDSHELLSCRIPTLEN, Long.toString(hdfsShellScriptLen));
if (domainId != null && domainId.length() > 0) {
env.put(DSConstants.DISTRIBUTEDSHELLTIMELINEDOMAIN, domainId);
}
// Add AppMaster.jar location to classpath
// At some point we should not be required to add
// the hadoop specific classpaths to the env.
// It should be provided out of the box.
// For now setting all required classpaths including
// the classpath to "." for the application jar
StringBuilder classPathEnv = new StringBuilder(Environment.CLASSPATH.$$()).append(ApplicationConstants.CLASS_PATH_SEPARATOR).append("./*");
for (String c : conf.getStrings(YarnConfiguration.YARN_APPLICATION_CLASSPATH, YarnConfiguration.DEFAULT_YARN_CROSS_PLATFORM_APPLICATION_CLASSPATH)) {
classPathEnv.append(ApplicationConstants.CLASS_PATH_SEPARATOR);
classPathEnv.append(c.trim());
}
classPathEnv.append(ApplicationConstants.CLASS_PATH_SEPARATOR).append("./log4j.properties");
// add the runtime classpath needed for tests to work
if (conf.getBoolean(YarnConfiguration.IS_MINI_YARN_CLUSTER, false)) {
classPathEnv.append(':');
classPathEnv.append(System.getProperty("java.class.path"));
}
env.put("CLASSPATH", classPathEnv.toString());
// Set the necessary command to execute the application master
Vector<CharSequence> vargs = new Vector<CharSequence>(30);
// Set java executable command
LOG.info("Setting up app master command");
vargs.add(Environment.JAVA_HOME.$$() + "/bin/java");
// Set Xmx based on am memory size
vargs.add("-Xmx" + amMemory + "m");
// Set class name
vargs.add(appMasterMainClass);
// Set params for Application Master
vargs.add("--container_memory " + String.valueOf(containerMemory));
vargs.add("--container_vcores " + String.valueOf(containerVirtualCores));
vargs.add("--num_containers " + String.valueOf(numContainers));
if (null != nodeLabelExpression) {
appContext.setNodeLabelExpression(nodeLabelExpression);
}
vargs.add("--priority " + String.valueOf(shellCmdPriority));
for (Map.Entry<String, String> entry : shellEnv.entrySet()) {
vargs.add("--shell_env " + entry.getKey() + "=" + entry.getValue());
}
if (debugFlag) {
vargs.add("--debug");
}
vargs.addAll(containerRetryOptions);
vargs.add("1>" + ApplicationConstants.LOG_DIR_EXPANSION_VAR + "/AppMaster.stdout");
vargs.add("2>" + ApplicationConstants.LOG_DIR_EXPANSION_VAR + "/AppMaster.stderr");
// Get final commmand
StringBuilder command = new StringBuilder();
for (CharSequence str : vargs) {
command.append(str).append(" ");
}
LOG.info("Completed setting up app master command " + command.toString());
List<String> commands = new ArrayList<String>();
commands.add(command.toString());
// Set up the container launch context for the application master
ContainerLaunchContext amContainer = ContainerLaunchContext.newInstance(localResources, env, commands, null, null, null);
// Set up resource type requirements
// For now, both memory and vcores are supported, so we set memory and
// vcores requirements
Resource capability = Resource.newInstance(amMemory, amVCores);
appContext.setResource(capability);
// Setup security tokens
if (UserGroupInformation.isSecurityEnabled()) {
// Note: Credentials class is marked as LimitedPrivate for HDFS and MapReduce
Credentials credentials = new Credentials();
String tokenRenewer = YarnClientUtils.getRmPrincipal(conf);
if (tokenRenewer == null || tokenRenewer.length() == 0) {
throw new IOException("Can't get Master Kerberos principal for the RM to use as renewer");
}
// For now, only getting tokens for the default file-system.
final Token<?>[] tokens = fs.addDelegationTokens(tokenRenewer, credentials);
if (tokens != null) {
for (Token<?> token : tokens) {
LOG.info("Got dt for " + fs.getUri() + "; " + token);
}
}
DataOutputBuffer dob = new DataOutputBuffer();
credentials.writeTokenStorageToStream(dob);
ByteBuffer fsTokens = ByteBuffer.wrap(dob.getData(), 0, dob.getLength());
amContainer.setTokens(fsTokens);
}
appContext.setAMContainerSpec(amContainer);
// Set the priority for the application master
// TODO - what is the range for priority? how to decide?
Priority pri = Priority.newInstance(amPriority);
appContext.setPriority(pri);
// Set the queue to which this application is to be submitted in the RM
appContext.setQueue(amQueue);
// Submit the application to the applications manager
// SubmitApplicationResponse submitResp = applicationsManager.submitApplication(appRequest);
// Ignore the response as either a valid response object is returned on success
// or an exception thrown to denote some form of a failure
LOG.info("Submitting application to ASM");
yarnClient.submitApplication(appContext);
// Monitor the application
return monitorApplication(appId);
}
use of org.apache.hadoop.yarn.api.records.LocalResource in project hadoop by apache.
the class TestContainerManager method testChangeContainerResource.
@Test
public void testChangeContainerResource() throws Exception {
containerManager.start();
File scriptFile = Shell.appendScriptExtension(tmpDir, "scriptFile");
PrintWriter fileWriter = new PrintWriter(scriptFile);
// Construct the Container-id
ContainerId cId = createContainerId(0);
if (Shell.WINDOWS) {
fileWriter.println("@ping -n 100 127.0.0.1 >nul");
} else {
fileWriter.write("\numask 0");
fileWriter.write("\nexec sleep 100");
}
fileWriter.close();
ContainerLaunchContext containerLaunchContext = recordFactory.newRecordInstance(ContainerLaunchContext.class);
URL resource_alpha = URL.fromPath(localFS.makeQualified(new Path(scriptFile.getAbsolutePath())));
LocalResource rsrc_alpha = recordFactory.newRecordInstance(LocalResource.class);
rsrc_alpha.setResource(resource_alpha);
rsrc_alpha.setSize(-1);
rsrc_alpha.setVisibility(LocalResourceVisibility.APPLICATION);
rsrc_alpha.setType(LocalResourceType.FILE);
rsrc_alpha.setTimestamp(scriptFile.lastModified());
String destinationFile = "dest_file";
Map<String, LocalResource> localResources = new HashMap<String, LocalResource>();
localResources.put(destinationFile, rsrc_alpha);
containerLaunchContext.setLocalResources(localResources);
List<String> commands = Arrays.asList(Shell.getRunScriptCommand(scriptFile));
containerLaunchContext.setCommands(commands);
StartContainerRequest scRequest = StartContainerRequest.newInstance(containerLaunchContext, createContainerToken(cId, DUMMY_RM_IDENTIFIER, context.getNodeId(), user, context.getContainerTokenSecretManager()));
List<StartContainerRequest> list = new ArrayList<>();
list.add(scRequest);
StartContainersRequest allRequests = StartContainersRequest.newInstance(list);
containerManager.startContainers(allRequests);
// Make sure the container reaches RUNNING state
BaseContainerManagerTest.waitForNMContainerState(containerManager, cId, org.apache.hadoop.yarn.server.nodemanager.containermanager.container.ContainerState.RUNNING);
// Construct container resource increase request,
List<Token> increaseTokens = new ArrayList<>();
// Add increase request.
Resource targetResource = Resource.newInstance(4096, 2);
Token containerToken = createContainerToken(cId, DUMMY_RM_IDENTIFIER, context.getNodeId(), user, targetResource, context.getContainerTokenSecretManager(), null);
increaseTokens.add(containerToken);
IncreaseContainersResourceRequest increaseRequest = IncreaseContainersResourceRequest.newInstance(increaseTokens);
IncreaseContainersResourceResponse increaseResponse = containerManager.increaseContainersResource(increaseRequest);
Assert.assertEquals(1, increaseResponse.getSuccessfullyIncreasedContainers().size());
Assert.assertTrue(increaseResponse.getFailedRequests().isEmpty());
// Check status
List<ContainerId> containerIds = new ArrayList<>();
containerIds.add(cId);
GetContainerStatusesRequest gcsRequest = GetContainerStatusesRequest.newInstance(containerIds);
ContainerStatus containerStatus = containerManager.getContainerStatuses(gcsRequest).getContainerStatuses().get(0);
// Check status immediately as resource increase is blocking
assertEquals(targetResource, containerStatus.getCapability());
// Simulate a decrease request
List<org.apache.hadoop.yarn.api.records.Container> containersToDecrease = new ArrayList<>();
targetResource = Resource.newInstance(2048, 2);
org.apache.hadoop.yarn.api.records.Container decreasedContainer = org.apache.hadoop.yarn.api.records.Container.newInstance(cId, null, null, targetResource, null, null);
containersToDecrease.add(decreasedContainer);
containerManager.handle(new CMgrDecreaseContainersResourceEvent(containersToDecrease));
// Check status with retry
containerStatus = containerManager.getContainerStatuses(gcsRequest).getContainerStatuses().get(0);
int retry = 0;
while (!targetResource.equals(containerStatus.getCapability()) && (retry++ < 5)) {
Thread.sleep(200);
containerStatus = containerManager.getContainerStatuses(gcsRequest).getContainerStatuses().get(0);
}
assertEquals(targetResource, containerStatus.getCapability());
}
use of org.apache.hadoop.yarn.api.records.LocalResource in project hadoop by apache.
the class TestContainerManager method testContainerLaunchAndStop.
//@Test
public void testContainerLaunchAndStop() throws IOException, InterruptedException, YarnException {
containerManager.start();
File scriptFile = Shell.appendScriptExtension(tmpDir, "scriptFile");
PrintWriter fileWriter = new PrintWriter(scriptFile);
File processStartFile = new File(tmpDir, "start_file.txt").getAbsoluteFile();
// ////// Construct the Container-id
ContainerId cId = createContainerId(0);
if (Shell.WINDOWS) {
fileWriter.println("@echo Hello World!> " + processStartFile);
fileWriter.println("@echo " + cId + ">> " + processStartFile);
fileWriter.println("@ping -n 100 127.0.0.1 >nul");
} else {
// So that start file is readable by the test
fileWriter.write("\numask 0");
fileWriter.write("\necho Hello World! > " + processStartFile);
fileWriter.write("\necho $$ >> " + processStartFile);
fileWriter.write("\nexec sleep 100");
}
fileWriter.close();
ContainerLaunchContext containerLaunchContext = recordFactory.newRecordInstance(ContainerLaunchContext.class);
URL resource_alpha = URL.fromPath(localFS.makeQualified(new Path(scriptFile.getAbsolutePath())));
LocalResource rsrc_alpha = recordFactory.newRecordInstance(LocalResource.class);
rsrc_alpha.setResource(resource_alpha);
rsrc_alpha.setSize(-1);
rsrc_alpha.setVisibility(LocalResourceVisibility.APPLICATION);
rsrc_alpha.setType(LocalResourceType.FILE);
rsrc_alpha.setTimestamp(scriptFile.lastModified());
String destinationFile = "dest_file";
Map<String, LocalResource> localResources = new HashMap<String, LocalResource>();
localResources.put(destinationFile, rsrc_alpha);
containerLaunchContext.setLocalResources(localResources);
List<String> commands = Arrays.asList(Shell.getRunScriptCommand(scriptFile));
containerLaunchContext.setCommands(commands);
StartContainerRequest scRequest = StartContainerRequest.newInstance(containerLaunchContext, createContainerToken(cId, DUMMY_RM_IDENTIFIER, context.getNodeId(), user, context.getContainerTokenSecretManager()));
List<StartContainerRequest> list = new ArrayList<>();
list.add(scRequest);
StartContainersRequest allRequests = StartContainersRequest.newInstance(list);
containerManager.startContainers(allRequests);
int timeoutSecs = 0;
while (!processStartFile.exists() && timeoutSecs++ < 20) {
Thread.sleep(1000);
LOG.info("Waiting for process start-file to be created");
}
Assert.assertTrue("ProcessStartFile doesn't exist!", processStartFile.exists());
// Now verify the contents of the file
BufferedReader reader = new BufferedReader(new FileReader(processStartFile));
Assert.assertEquals("Hello World!", reader.readLine());
// Get the pid of the process
String pid = reader.readLine().trim();
// No more lines
Assert.assertEquals(null, reader.readLine());
// Now test the stop functionality.
// Assert that the process is alive
Assert.assertTrue("Process is not alive!", DefaultContainerExecutor.containerIsAlive(pid));
// Once more
Assert.assertTrue("Process is not alive!", DefaultContainerExecutor.containerIsAlive(pid));
List<ContainerId> containerIds = new ArrayList<>();
containerIds.add(cId);
StopContainersRequest stopRequest = StopContainersRequest.newInstance(containerIds);
containerManager.stopContainers(stopRequest);
BaseContainerManagerTest.waitForContainerState(containerManager, cId, ContainerState.COMPLETE);
GetContainerStatusesRequest gcsRequest = GetContainerStatusesRequest.newInstance(containerIds);
ContainerStatus containerStatus = containerManager.getContainerStatuses(gcsRequest).getContainerStatuses().get(0);
int expectedExitCode = ContainerExitStatus.KILLED_BY_APPMASTER;
Assert.assertEquals(expectedExitCode, containerStatus.getExitStatus());
// Assert that the process is not alive anymore
Assert.assertFalse("Process is still alive!", DefaultContainerExecutor.containerIsAlive(pid));
}
use of org.apache.hadoop.yarn.api.records.LocalResource in project hadoop by apache.
the class TestContainerManager method testIncreaseContainerResourceWithInvalidResource.
@Test
public void testIncreaseContainerResourceWithInvalidResource() throws Exception {
containerManager.start();
File scriptFile = Shell.appendScriptExtension(tmpDir, "scriptFile");
PrintWriter fileWriter = new PrintWriter(scriptFile);
// Construct the Container-id
ContainerId cId = createContainerId(0);
if (Shell.WINDOWS) {
fileWriter.println("@ping -n 100 127.0.0.1 >nul");
} else {
fileWriter.write("\numask 0");
fileWriter.write("\nexec sleep 100");
}
fileWriter.close();
ContainerLaunchContext containerLaunchContext = recordFactory.newRecordInstance(ContainerLaunchContext.class);
URL resource_alpha = URL.fromPath(localFS.makeQualified(new Path(scriptFile.getAbsolutePath())));
LocalResource rsrc_alpha = recordFactory.newRecordInstance(LocalResource.class);
rsrc_alpha.setResource(resource_alpha);
rsrc_alpha.setSize(-1);
rsrc_alpha.setVisibility(LocalResourceVisibility.APPLICATION);
rsrc_alpha.setType(LocalResourceType.FILE);
rsrc_alpha.setTimestamp(scriptFile.lastModified());
String destinationFile = "dest_file";
Map<String, LocalResource> localResources = new HashMap<String, LocalResource>();
localResources.put(destinationFile, rsrc_alpha);
containerLaunchContext.setLocalResources(localResources);
List<String> commands = Arrays.asList(Shell.getRunScriptCommand(scriptFile));
containerLaunchContext.setCommands(commands);
StartContainerRequest scRequest = StartContainerRequest.newInstance(containerLaunchContext, createContainerToken(cId, DUMMY_RM_IDENTIFIER, context.getNodeId(), user, context.getContainerTokenSecretManager()));
List<StartContainerRequest> list = new ArrayList<>();
list.add(scRequest);
StartContainersRequest allRequests = StartContainersRequest.newInstance(list);
containerManager.startContainers(allRequests);
// Make sure the container reaches RUNNING state
BaseContainerManagerTest.waitForNMContainerState(containerManager, cId, org.apache.hadoop.yarn.server.nodemanager.containermanager.container.ContainerState.RUNNING);
// Construct container resource increase request,
List<Token> increaseTokens = new ArrayList<>();
// Add increase request. The increase request should fail
// as the current resource does not fit in the target resource
Token containerToken = createContainerToken(cId, DUMMY_RM_IDENTIFIER, context.getNodeId(), user, Resource.newInstance(512, 1), context.getContainerTokenSecretManager(), null);
increaseTokens.add(containerToken);
IncreaseContainersResourceRequest increaseRequest = IncreaseContainersResourceRequest.newInstance(increaseTokens);
IncreaseContainersResourceResponse increaseResponse = containerManager.increaseContainersResource(increaseRequest);
// Check response
Assert.assertEquals(0, increaseResponse.getSuccessfullyIncreasedContainers().size());
Assert.assertEquals(1, increaseResponse.getFailedRequests().size());
for (Map.Entry<ContainerId, SerializedException> entry : increaseResponse.getFailedRequests().entrySet()) {
if (cId.equals(entry.getKey())) {
Assert.assertNotNull("Failed message", entry.getValue().getMessage());
Assert.assertTrue(entry.getValue().getMessage().contains("The target resource " + Resource.newInstance(512, 1).toString() + " is smaller than the current resource " + Resource.newInstance(1024, 1)));
} else {
throw new YarnException("Received failed request from wrong" + " container: " + entry.getKey().toString());
}
}
}
Aggregations