YARN-6943. Update Yarn to YARN in documentation. Contributed by Chetna Chaudhari.
This commit is contained in:
Akira Ajisaka
parent
4a87773718
commit
c9b525bd94
|
|
@ -2208,7 +2208,7 @@ public class TestKMS {
|
|||
"hadoop.kms.authentication.delegation-token.renew-interval.sec", "5");
|
||||
writeConf(confDir, conf);
|
||||
|
||||
// Running as a service (e.g. Yarn in practice).
|
||||
// Running as a service (e.g. YARN in practice).
|
||||
runServer(null, null, confDir, new KMSCallable<Void>() {
|
||||
@Override
|
||||
public Void call() throws Exception {
|
||||
|
|
@ -2223,7 +2223,7 @@ public class TestKMS {
|
|||
final InetSocketAddress kmsAddr =
|
||||
new InetSocketAddress(getKMSUrl().getHost(), getKMSUrl().getPort());
|
||||
|
||||
// Job 1 (e.g. Yarn log aggregation job), with user DT.
|
||||
// Job 1 (e.g. YARN log aggregation job), with user DT.
|
||||
final Collection<Token<?>> job1Token = new HashSet<>();
|
||||
doAs("client", new PrivilegedExceptionAction<Void>() {
|
||||
@Override
|
||||
|
|
@ -2268,7 +2268,7 @@ public class TestKMS {
|
|||
});
|
||||
Assert.assertFalse(job1Token.isEmpty());
|
||||
|
||||
// job 2 (e.g. Another Yarn log aggregation job, with user DT.
|
||||
// job 2 (e.g. Another YARN log aggregation job, with user DT.
|
||||
doAs("client", new PrivilegedExceptionAction<Void>() {
|
||||
@Override
|
||||
public Void run() throws Exception {
|
||||
|
|
|
|||
|
|
@ -21,7 +21,7 @@ Hadoop Archive Logs Guide
|
|||
Overview
|
||||
--------
|
||||
|
||||
For clusters with a lot of Yarn aggregated logs, it can be helpful to combine
|
||||
For clusters with a lot of YARN aggregated logs, it can be helpful to combine
|
||||
them into hadoop archives in order to reduce the number of small files, and
|
||||
hence the stress on the NameNode. This tool provides an easy way to do this.
|
||||
Aggregated logs in hadoop archives can still be read by the Job History Server
|
||||
|
|
@ -50,7 +50,7 @@ How to Archive Logs
|
|||
to be eligible (default: 20)
|
||||
-noProxy When specified, all processing will be
|
||||
done as the user running this command (or
|
||||
the Yarn user if DefaultContainerExecutor
|
||||
the YARN user if DefaultContainerExecutor
|
||||
is in use). When not specified, all
|
||||
processing will be done as the user who
|
||||
owns that application; if the user
|
||||
|
|
@ -86,7 +86,7 @@ The tool works by performing the following procedure:
|
|||
its aggregated log files with the resulting archive.
|
||||
|
||||
The ``-noProxy`` option makes the tool process everything as the user who is
|
||||
currently running it, or the Yarn user if DefaultContainerExecutor is in use.
|
||||
currently running it, or the YARN user if DefaultContainerExecutor is in use.
|
||||
When not specified, all processing will be done by the user who owns that
|
||||
application; if the user running this command is not allowed to impersonate that
|
||||
user, it will fail. This is useful if you want an admin user to handle all
|
||||
|
|
|
|||
|
|
@ -12,10 +12,10 @@
|
|||
limitations under the License. See accompanying LICENSE file.
|
||||
-->
|
||||
|
||||
Yarn Scheduler Load Simulator (SLS)
|
||||
YARN Scheduler Load Simulator (SLS)
|
||||
===================================
|
||||
|
||||
* [Yarn Scheduler Load Simulator (SLS)](#Yarn_Scheduler_Load_Simulator_SLS)
|
||||
* [YARN Scheduler Load Simulator (SLS)](#Yarn_Scheduler_Load_Simulator_SLS)
|
||||
* [Overview](#Overview)
|
||||
* [Overview](#Overview)
|
||||
* [Goals](#Goals)
|
||||
|
|
@ -39,11 +39,11 @@ Overview
|
|||
|
||||
### Overview
|
||||
|
||||
The Yarn scheduler is a fertile area of interest with different implementations, e.g., Fifo, Capacity and Fair schedulers. Meanwhile, several optimizations are also made to improve scheduler performance for different scenarios and workload. Each scheduler algorithm has its own set of features, and drives scheduling decisions by many factors, such as fairness, capacity guarantee, resource availability, etc. It is very important to evaluate a scheduler algorithm very well before we deploy in a production cluster. Unfortunately, currently it is non-trivial to evaluate a scheduler algorithm. Evaluating in a real cluster is always time and cost consuming, and it is also very hard to find a large-enough cluster. Hence, a simulator which can predict how well a scheduler algorithm for some specific workload would be quite useful.
|
||||
The YARN scheduler is a fertile area of interest with different implementations, e.g., Fifo, Capacity and Fair schedulers. Meanwhile, several optimizations are also made to improve scheduler performance for different scenarios and workload. Each scheduler algorithm has its own set of features, and drives scheduling decisions by many factors, such as fairness, capacity guarantee, resource availability, etc. It is very important to evaluate a scheduler algorithm very well before we deploy in a production cluster. Unfortunately, currently it is non-trivial to evaluate a scheduler algorithm. Evaluating in a real cluster is always time and cost consuming, and it is also very hard to find a large-enough cluster. Hence, a simulator which can predict how well a scheduler algorithm for some specific workload would be quite useful.
|
||||
|
||||
The Yarn Scheduler Load Simulator (SLS) is such a tool, which can simulate large-scale Yarn clusters and application loads in a single machine.This simulator would be invaluable in furthering Yarn by providing a tool for researchers and developers to prototype new scheduler features and predict their behavior and performance with reasonable amount of confidence, thereby aiding rapid innovation.
|
||||
The YARN Scheduler Load Simulator (SLS) is such a tool, which can simulate large-scale YARN clusters and application loads in a single machine.This simulator would be invaluable in furthering YARN by providing a tool for researchers and developers to prototype new scheduler features and predict their behavior and performance with reasonable amount of confidence, thereby aiding rapid innovation.
|
||||
o
|
||||
The simulator will exercise the real Yarn `ResourceManager` removing the network factor by simulating `NodeManagers` and `ApplicationMasters` via handling and dispatching `NM`/`AMs` heartbeat events from within the same JVM. To keep tracking of scheduler behavior and performance, a scheduler wrapper will wrap the real scheduler.
|
||||
The simulator will exercise the real YARN `ResourceManager` removing the network factor by simulating `NodeManagers` and `ApplicationMasters` via handling and dispatching `NM`/`AMs` heartbeat events from within the same JVM. To keep tracking of scheduler behavior and performance, a scheduler wrapper will wrap the real scheduler.
|
||||
|
||||
The size of the cluster and the application load can be loaded from configuration files, which are generated from job history files directly by adopting [Apache Rumen](../hadoop-rumen/Rumen.html).
|
||||
|
||||
|
|
@ -74,7 +74,7 @@ The following figure illustrates the implementation architecture of the simulato
|
|||
|
||||
|
||||
|
||||
The simulator takes input of workload traces, or synthetic load distributions and generaters the cluster and applications information. For each NM and AM, the simulator builds a simulator to simulate their running. All NM/AM simulators run in a thread pool. The simulator reuses Yarn Resource Manager, and builds a wrapper out of the scheduler. The Scheduler Wrapper can track the scheduler behaviors and generates several logs, which are the outputs of the simulator and can be further analyzed.
|
||||
The simulator takes input of workload traces, or synthetic load distributions and generaters the cluster and applications information. For each NM and AM, the simulator builds a simulator to simulate their running. All NM/AM simulators run in a thread pool. The simulator reuses YARN Resource Manager, and builds a wrapper out of the scheduler. The Scheduler Wrapper can track the scheduler behaviors and generates several logs, which are the outputs of the simulator and can be further analyzed.
|
||||
|
||||
### Usecases
|
||||
|
||||
|
|
@ -110,9 +110,9 @@ The following sections will describe how to use the simulator step by step. Befo
|
|||
|
||||
### Step 1: Configure Hadoop and the simulator
|
||||
|
||||
Before we start, make sure Hadoop and the simulator are configured well. All configuration files for Hadoop and the simulator should be placed in directory `$HADOOP_ROOT/etc/hadoop`, where the `ResourceManager` and Yarn scheduler load their configurations. Directory `$HADOOP_ROOT/share/hadoop/tools/sls/sample-conf/` provides several example configurations, that can be used to start a demo.
|
||||
Before we start, make sure Hadoop and the simulator are configured well. All configuration files for Hadoop and the simulator should be placed in directory `$HADOOP_ROOT/etc/hadoop`, where the `ResourceManager` and YARN scheduler load their configurations. Directory `$HADOOP_ROOT/share/hadoop/tools/sls/sample-conf/` provides several example configurations, that can be used to start a demo.
|
||||
|
||||
For configuration of Hadoop and Yarn scheduler, users can refer to Yarn’s website (<http://hadoop.apache.org/docs/current/hadoop-yarn/hadoop-yarn-site/>).
|
||||
For configuration of Hadoop and YARN scheduler, users can refer to Yarn’s website (<http://hadoop.apache.org/docs/current/hadoop-yarn/hadoop-yarn-site/>).
|
||||
|
||||
For the simulator, it loads configuration information from file `$HADOOP_ROOT/etc/hadoop/sls-runner.xml`.
|
||||
|
||||
|
|
@ -244,7 +244,7 @@ The simulator supports two types of input files: the rumen traces and its own in
|
|||
Metrics
|
||||
-------
|
||||
|
||||
The Yarn Scheduler Load Simulator has integrated [Metrics](http://metrics.codahale.com/) to measure the behaviors of critical components and operations, including running applications and containers, cluster available resources, scheduler operation timecost, et al. If the switch `yarn.sls.runner.metrics.switch` is set `ON`, `Metrics` will run and output it logs in `--output-dir` directory specified by users. Users can track these information during simulator running, and can also analyze these logs after running to evaluate the scheduler performance.
|
||||
The YARN Scheduler Load Simulator has integrated [Metrics](http://metrics.codahale.com/) to measure the behaviors of critical components and operations, including running applications and containers, cluster available resources, scheduler operation timecost, et al. If the switch `yarn.sls.runner.metrics.switch` is set `ON`, `Metrics` will run and output it logs in `--output-dir` directory specified by users. Users can track these information during simulator running, and can also analyze these logs after running to evaluate the scheduler performance.
|
||||
|
||||
### Real-time Tracking
|
||||
|
||||
|
|
@ -320,7 +320,7 @@ Appendix
|
|||
|
||||
### Resources
|
||||
|
||||
[YARN-1021](https://issues.apache.org/jira/browse/YARN-1021) is the main JIRA that introduces Yarn Scheduler Load Simulator to Hadoop Yarn project.
|
||||
[YARN-1021](https://issues.apache.org/jira/browse/YARN-1021) is the main JIRA that introduces YARN Scheduler Load Simulator to Hadoop YARN project.
|
||||
[YARN-6363](https://issues.apache.org/jira/browse/YARN-6363) is the main JIRA that introduces the Synthetic Load Generator to SLS.
|
||||
|
||||
### SLS JSON input file format
|
||||
|
|
|
|||
|
|
@ -55,7 +55,7 @@ Modules
|
|||
-------
|
||||
YARN consists of multiple modules. The modules are listed below as per the directory structure:
|
||||
|
||||
hadoop-yarn-api - Yarn's cross platform external interface
|
||||
hadoop-yarn-api - YARN's cross platform external interface
|
||||
|
||||
hadoop-yarn-common - Utilities which can be used by yarn clients and server
|
||||
|
||||
|
|
|
|||
|
|
@ -367,7 +367,7 @@ public abstract class ApplicationReport {
|
|||
* Get the AMRM token of the application.
|
||||
* <p>
|
||||
* The AMRM token is required for AM to RM scheduling operations. For
|
||||
* managed Application Masters Yarn takes care of injecting it. For unmanaged
|
||||
* managed Application Masters YARN takes care of injecting it. For unmanaged
|
||||
* Applications Masters, the token must be obtained via this method and set
|
||||
* in the {@link org.apache.hadoop.security.UserGroupInformation} of the
|
||||
* current user.
|
||||
|
|
|
|||
|
|
@ -53,7 +53,7 @@ import org.apache.hadoop.yarn.util.Records;
|
|||
* <li>
|
||||
* maxAppAttempts. The maximum number of application attempts.
|
||||
* It should be no larger than the global number of max attempts in the
|
||||
* Yarn configuration.
|
||||
* YARN configuration.
|
||||
* </li>
|
||||
* <li>
|
||||
* attemptFailuresValidityInterval. The default value is -1.
|
||||
|
|
|
|||
|
|
@ -21,7 +21,7 @@ package org.apache.hadoop.yarn.exceptions;
|
|||
import org.apache.hadoop.classification.InterfaceAudience.LimitedPrivate;
|
||||
import org.apache.hadoop.classification.InterfaceStability.Unstable;
|
||||
|
||||
/** Base Yarn Exception.
|
||||
/** Base YARN Exception.
|
||||
*
|
||||
* NOTE: All derivatives of this exception, which may be thrown by a remote
|
||||
* service, must include a String only constructor for the exception to be
|
||||
|
|
|
|||
|
|
@ -35,7 +35,7 @@ import com.google.common.annotations.VisibleForTesting;
|
|||
* NMTokenCache manages NMTokens required for an Application Master
|
||||
* communicating with individual NodeManagers.
|
||||
* <p>
|
||||
* By default Yarn client libraries {@link AMRMClient} and {@link NMClient} use
|
||||
* By default YARN client libraries {@link AMRMClient} and {@link NMClient} use
|
||||
* {@link #getSingleton()} instance of the cache.
|
||||
* <ul>
|
||||
* <li>
|
||||
|
|
|
|||
|
|
@ -229,7 +229,7 @@ public abstract class YarnClient extends AbstractService {
|
|||
* Get the AMRM token of the application.
|
||||
* <p>
|
||||
* The AMRM token is required for AM to RM scheduling operations. For
|
||||
* managed Application Masters Yarn takes care of injecting it. For unmanaged
|
||||
* managed Application Masters YARN takes care of injecting it. For unmanaged
|
||||
* Applications Masters, the token must be obtained via this method and set
|
||||
* in the {@link org.apache.hadoop.security.UserGroupInformation} of the
|
||||
* current user.
|
||||
|
|
|
|||
|
|
@ -31,7 +31,7 @@ import org.apache.hadoop.util.ShutdownHookManager;
|
|||
* This class is intended to be installed by calling
|
||||
* {@link Thread#setDefaultUncaughtExceptionHandler(UncaughtExceptionHandler)}
|
||||
* In the main entry point. It is intended to try and cleanly shut down
|
||||
* programs using the Yarn Event framework.
|
||||
* programs using the YARN Event framework.
|
||||
*
|
||||
* Note: Right now it only will shut down the program if a Error is caught, but
|
||||
* not any other exception. Anything else is just logged.
|
||||
|
|
|
|||
|
|
@ -38,7 +38,7 @@ import org.apache.hadoop.yarn.api.records.ApplicationId;
|
|||
import org.apache.hadoop.yarn.exceptions.YarnRuntimeException;
|
||||
|
||||
/**
|
||||
* Yarn internal application-related utilities
|
||||
* YARN internal application-related utilities
|
||||
*/
|
||||
@Private
|
||||
public class Apps {
|
||||
|
|
|
|||
|
|
@ -27,7 +27,7 @@ import org.apache.hadoop.conf.Configured;
|
|||
import org.apache.hadoop.yarn.api.records.ApplicationId;
|
||||
|
||||
/**
|
||||
* Plugin to derive a tracking URL from a Yarn Application ID
|
||||
* Plugin to derive a tracking URL from a YARN Application ID
|
||||
*
|
||||
*/
|
||||
@InterfaceAudience.LimitedPrivate({"MapReduce"})
|
||||
|
|
|
|||
|
|
@ -38,8 +38,8 @@ public class YarnVersionInfo extends VersionInfo {
|
|||
super("yarn");
|
||||
}
|
||||
/**
|
||||
* Get the Yarn version.
|
||||
* @return the Yarn version string, eg. "0.6.3-dev"
|
||||
* Get the YARN version.
|
||||
* @return the YARN version string, eg. "0.6.3-dev"
|
||||
*/
|
||||
public static String getVersion() {
|
||||
return YARN_VERSION_INFO._getVersion();
|
||||
|
|
@ -62,7 +62,7 @@ public class YarnVersionInfo extends VersionInfo {
|
|||
}
|
||||
|
||||
/**
|
||||
* The date that Yarn was compiled.
|
||||
* The date that YARN was compiled.
|
||||
* @return the compilation date in unix date format
|
||||
*/
|
||||
public static String getDate() {
|
||||
|
|
@ -78,14 +78,14 @@ public class YarnVersionInfo extends VersionInfo {
|
|||
}
|
||||
|
||||
/**
|
||||
* Get the subversion URL for the root Yarn directory.
|
||||
* Get the subversion URL for the root YARN directory.
|
||||
*/
|
||||
public static String getUrl() {
|
||||
return YARN_VERSION_INFO._getUrl();
|
||||
}
|
||||
|
||||
/**
|
||||
* Get the checksum of the source files from which Yarn was
|
||||
* Get the checksum of the source files from which YARN was
|
||||
* built.
|
||||
**/
|
||||
public static String getSrcChecksum() {
|
||||
|
|
@ -102,7 +102,7 @@ public class YarnVersionInfo extends VersionInfo {
|
|||
|
||||
public static void main(String[] args) {
|
||||
LOG.debug("version: "+ getVersion());
|
||||
System.out.println("Yarn " + getVersion());
|
||||
System.out.println("YARN " + getVersion());
|
||||
System.out.println("Subversion " + getUrl() + " -r " + getRevision());
|
||||
System.out.println("Compiled by " + getUser() + " on " + getDate());
|
||||
System.out.println("From source with checksum " + getSrcChecksum());
|
||||
|
|
|
|||
|
|
@ -21,7 +21,7 @@
|
|||
* <ul>
|
||||
* <li>
|
||||
* The {@link org.apache.hadoop.registry.server.services.RegistryAdminService}
|
||||
* extends the shared Yarn Registry client with registry setup and
|
||||
* extends the shared YARN Registry client with registry setup and
|
||||
* (potentially asynchronous) administrative actions.
|
||||
* </li>
|
||||
* <li>
|
||||
|
|
|
|||
|
|
@ -80,7 +80,7 @@ public final class FederationPolicyUtils {
|
|||
* and configuration as fallback.
|
||||
*
|
||||
* @param queue the queue of the application
|
||||
* @param conf the Yarn configuration
|
||||
* @param conf the YARN configuration
|
||||
* @param federationFacade state store facade
|
||||
* @return SubClusterPolicyConfiguration recreated
|
||||
*/
|
||||
|
|
@ -139,7 +139,7 @@ public final class FederationPolicyUtils {
|
|||
*
|
||||
* @param queue the queue of the application
|
||||
* @param oldPolicy the previous policy instance (can be null)
|
||||
* @param conf the Yarn configuration
|
||||
* @param conf the YARN configuration
|
||||
* @param federationFacade state store facade
|
||||
* @param homeSubClusterId home sub-cluster id
|
||||
* @return FederationAMRMProxyPolicy recreated
|
||||
|
|
|
|||
|
|
@ -187,7 +187,7 @@ public class MockResourceManagerFacade implements ApplicationClientProtocol,
|
|||
final private AtomicInteger applicationCounter = new AtomicInteger(0);
|
||||
|
||||
// True if the Mock RM is running, false otherwise.
|
||||
// This property allows us to write tests for specific scenario as Yarn RM
|
||||
// This property allows us to write tests for specific scenario as YARN RM
|
||||
// down e.g. network issue, failover.
|
||||
private boolean isRunning;
|
||||
|
||||
|
|
|
|||
|
|
@ -1716,7 +1716,7 @@ public class ContainerManagerImpl extends CompositeService implements
|
|||
* @param containerId Container Id.
|
||||
* @param autoCommit Auto Commit flag.
|
||||
* @param reInitLaunchContext Target Launch Context.
|
||||
* @throws YarnException Yarn Exception.
|
||||
* @throws YarnException YARN Exception.
|
||||
*/
|
||||
public void reInitializeContainer(ContainerId containerId,
|
||||
ContainerLaunchContext reInitLaunchContext, boolean autoCommit)
|
||||
|
|
@ -1743,7 +1743,7 @@ public class ContainerManagerImpl extends CompositeService implements
|
|||
* Rollback the last reInitialization, if possible.
|
||||
* @param containerId Container ID.
|
||||
* @return Rollback Response.
|
||||
* @throws YarnException Yarn Exception.
|
||||
* @throws YarnException YARN Exception.
|
||||
*/
|
||||
@Override
|
||||
public RollbackResponse rollbackLastReInitialization(ContainerId containerId)
|
||||
|
|
@ -1764,7 +1764,7 @@ public class ContainerManagerImpl extends CompositeService implements
|
|||
* Commit last reInitialization after which no rollback will be possible.
|
||||
* @param containerId Container ID.
|
||||
* @return Commit Response.
|
||||
* @throws YarnException Yarn Exception.
|
||||
* @throws YarnException YARN Exception.
|
||||
*/
|
||||
@Override
|
||||
public CommitResponse commitLastReInitialization(ContainerId containerId)
|
||||
|
|
|
|||
|
|
@ -370,7 +370,7 @@ class CGroupsHandlerImpl implements CGroupsHandler {
|
|||
}
|
||||
|
||||
// We are working with a pre-mounted contoller
|
||||
// Make sure that Yarn cgroup hierarchy path exists
|
||||
// Make sure that YARN cgroup hierarchy path exists
|
||||
initializePreMountedCGroupController(controller);
|
||||
}
|
||||
|
||||
|
|
@ -378,9 +378,9 @@ class CGroupsHandlerImpl implements CGroupsHandler {
|
|||
* This function is called when the administrator opted
|
||||
* to use a pre-mounted cgroup controller.
|
||||
* There are two options.
|
||||
* 1. Yarn hierarchy already exists. We verify, whether we have write access
|
||||
* 1. YARN hierarchy already exists. We verify, whether we have write access
|
||||
* in this case.
|
||||
* 2. Yarn hierarchy does not exist, yet. We create it in this case.
|
||||
* 2. YARN hierarchy does not exist, yet. We create it in this case.
|
||||
* @param controller the controller being initialized
|
||||
* @throws ResourceHandlerException yarn hierarchy cannot be created or
|
||||
* accessed for any reason
|
||||
|
|
|
|||
|
|
@ -118,7 +118,7 @@ public class TestCGroupsHandlerImpl {
|
|||
|
||||
/**
|
||||
* Create configuration where the cgroups are premounted.
|
||||
* @param myHierarchy Yarn cgroup
|
||||
* @param myHierarchy YARN cgroup
|
||||
* @return configuration object
|
||||
*/
|
||||
private Configuration createNoMountConfiguration(String myHierarchy) {
|
||||
|
|
@ -396,7 +396,7 @@ public class TestCGroupsHandlerImpl {
|
|||
File mtab = createPremountedCgroups(parentDir, false);
|
||||
File mountPoint = new File(parentDir, "cpu");
|
||||
|
||||
// Initialize Yarn classes
|
||||
// Initialize YARN classes
|
||||
Configuration confNoMount = createNoMountConfiguration(myHierarchy);
|
||||
CGroupsHandlerImpl cGroupsHandler = new CGroupsHandlerImpl(confNoMount,
|
||||
privilegedOperationExecutorMock, mtab.getAbsolutePath());
|
||||
|
|
@ -555,7 +555,7 @@ public class TestCGroupsHandlerImpl {
|
|||
assertTrue("Could not create dirs",
|
||||
new File(newMountPoint, "cpu").mkdirs());
|
||||
|
||||
// Initialize Yarn classes
|
||||
// Initialize YARN classes
|
||||
Configuration confMount = createMountConfiguration();
|
||||
confMount.set(YarnConfiguration.NM_LINUX_CONTAINER_CGROUPS_MOUNT_PATH,
|
||||
parentDir.getAbsolutePath() + Path.SEPARATOR + newMountPointDir);
|
||||
|
|
|
|||
|
|
@ -129,7 +129,7 @@ public abstract class SchedulerNode {
|
|||
* Typically this is the 'hostname' reported by the node, but it could be
|
||||
* configured to be 'hostname:port' reported by the node via the
|
||||
* {@link YarnConfiguration#RM_SCHEDULER_INCLUDE_PORT_IN_NODE_NAME} constant.
|
||||
* The main usecase of this is Yarn minicluster to be able to differentiate
|
||||
* The main usecase of this is YARN minicluster to be able to differentiate
|
||||
* node manager instances by their port number.
|
||||
* @return Name of the node for scheduling matching decisions.
|
||||
*/
|
||||
|
|
|
|||
|
|
@ -22,7 +22,7 @@ import org.apache.hadoop.util.ProgramDriver;
|
|||
import org.apache.hadoop.yarn.server.resourcemanager.recovery.TestZKRMStateStorePerf;
|
||||
|
||||
/**
|
||||
* Driver for Yarn tests.
|
||||
* Driver for YARN tests.
|
||||
*
|
||||
*/
|
||||
public class YarnTestDriver {
|
||||
|
|
|
|||
|
|
@ -212,7 +212,7 @@ public class FederationClientInterceptor
|
|||
}
|
||||
|
||||
/**
|
||||
* Yarn Router forwards every getNewApplication requests to any RM. During
|
||||
* YARN Router forwards every getNewApplication requests to any RM. During
|
||||
* this operation there will be no communication with the State Store. The
|
||||
* Router will forward the requests to any SubCluster. The Router will retry
|
||||
* to submit the request on #numSubmitRetries different SubClusters. The
|
||||
|
|
@ -431,7 +431,7 @@ public class FederationClientInterceptor
|
|||
}
|
||||
|
||||
/**
|
||||
* The Yarn Router will forward to the respective Yarn RM in which the AM is
|
||||
* The YARN Router will forward to the respective YARN RM in which the AM is
|
||||
* running.
|
||||
*
|
||||
* Possible failures and behaviors:
|
||||
|
|
@ -496,7 +496,7 @@ public class FederationClientInterceptor
|
|||
}
|
||||
|
||||
/**
|
||||
* The Yarn Router will forward to the respective Yarn RM in which the AM is
|
||||
* The YARN Router will forward to the respective YARN RM in which the AM is
|
||||
* running.
|
||||
*
|
||||
* Possible failure:
|
||||
|
|
|
|||
|
|
@ -228,7 +228,7 @@ public class FederationInterceptorREST extends AbstractRESTRequestInterceptor {
|
|||
}
|
||||
|
||||
/**
|
||||
* Yarn Router forwards every getNewApplication requests to any RM. During
|
||||
* YARN Router forwards every getNewApplication requests to any RM. During
|
||||
* this operation there will be no communication with the State Store. The
|
||||
* Router will forward the requests to any SubCluster. The Router will retry
|
||||
* to submit the request on #numSubmitRetries different SubClusters. The
|
||||
|
|
@ -497,7 +497,7 @@ public class FederationInterceptorREST extends AbstractRESTRequestInterceptor {
|
|||
}
|
||||
|
||||
/**
|
||||
* The Yarn Router will forward to the respective Yarn RM in which the AM is
|
||||
* The YARN Router will forward to the respective YARN RM in which the AM is
|
||||
* running.
|
||||
* <p>
|
||||
* Possible failure:
|
||||
|
|
@ -552,7 +552,7 @@ public class FederationInterceptorREST extends AbstractRESTRequestInterceptor {
|
|||
}
|
||||
|
||||
/**
|
||||
* The Yarn Router will forward to the respective Yarn RM in which the AM is
|
||||
* The YARN Router will forward to the respective YARN RM in which the AM is
|
||||
* running.
|
||||
* <p>
|
||||
* Possible failures and behaviors:
|
||||
|
|
@ -606,7 +606,7 @@ public class FederationInterceptorREST extends AbstractRESTRequestInterceptor {
|
|||
}
|
||||
|
||||
/**
|
||||
* The Yarn Router will forward the request to all the Yarn RMs in parallel,
|
||||
* The YARN Router will forward the request to all the YARN RMs in parallel,
|
||||
* after that it will group all the ApplicationReports by the ApplicationId.
|
||||
* <p>
|
||||
* Possible failure:
|
||||
|
|
@ -615,8 +615,8 @@ public class FederationInterceptorREST extends AbstractRESTRequestInterceptor {
|
|||
* <p>
|
||||
* Router: the Client will timeout and resubmit the request.
|
||||
* <p>
|
||||
* ResourceManager: the Router calls each Yarn RM in parallel by using one
|
||||
* thread for each Yarn RM. In case a Yarn RM fails, a single call will
|
||||
* ResourceManager: the Router calls each YARN RM in parallel by using one
|
||||
* thread for each YARN RM. In case a YARN RM fails, a single call will
|
||||
* timeout. However the Router will merge the ApplicationReports it got, and
|
||||
* provides a partial list to the client.
|
||||
* <p>
|
||||
|
|
@ -692,14 +692,14 @@ public class FederationInterceptorREST extends AbstractRESTRequestInterceptor {
|
|||
return null;
|
||||
}
|
||||
|
||||
// Merge all the application reports got from all the available Yarn RMs
|
||||
// Merge all the application reports got from all the available YARN RMs
|
||||
|
||||
return RouterWebServiceUtil.mergeAppsInfo(apps.getApps(),
|
||||
returnPartialReport);
|
||||
}
|
||||
|
||||
/**
|
||||
* The Yarn Router will forward to the request to all the SubClusters to find
|
||||
* The YARN Router will forward to the request to all the SubClusters to find
|
||||
* where the node is running.
|
||||
* <p>
|
||||
* Possible failure:
|
||||
|
|
@ -779,7 +779,7 @@ public class FederationInterceptorREST extends AbstractRESTRequestInterceptor {
|
|||
}
|
||||
|
||||
/**
|
||||
* The Yarn Router will forward the request to all the Yarn RMs in parallel,
|
||||
* The YARN Router will forward the request to all the YARN RMs in parallel,
|
||||
* after that it will remove all the duplicated NodeInfo by using the NodeId.
|
||||
* <p>
|
||||
* Possible failure:
|
||||
|
|
@ -788,8 +788,8 @@ public class FederationInterceptorREST extends AbstractRESTRequestInterceptor {
|
|||
* <p>
|
||||
* Router: the Client will timeout and resubmit the request.
|
||||
* <p>
|
||||
* ResourceManager: the Router calls each Yarn RM in parallel by using one
|
||||
* thread for each Yarn RM. In case a Yarn RM fails, a single call will
|
||||
* ResourceManager: the Router calls each YARN RM in parallel by using one
|
||||
* thread for each YARN RM. In case a YARN RM fails, a single call will
|
||||
* timeout. However the Router will use the NodesInfo it got, and provides a
|
||||
* partial list to the client.
|
||||
* <p>
|
||||
|
|
@ -850,7 +850,7 @@ public class FederationInterceptorREST extends AbstractRESTRequestInterceptor {
|
|||
}
|
||||
|
||||
// Delete duplicate from all the node reports got from all the available
|
||||
// Yarn RMs. Nodes can be moved from one subclusters to another. In this
|
||||
// YARN RMs. Nodes can be moved from one subclusters to another. In this
|
||||
// operation they result LOST/RUNNING in the previous SubCluster and
|
||||
// NEW/RUNNING in the new one.
|
||||
|
||||
|
|
|
|||
|
|
@ -57,7 +57,7 @@ public class MockDefaultRequestInterceptorREST
|
|||
LoggerFactory.getLogger(MockDefaultRequestInterceptorREST.class);
|
||||
final private AtomicInteger applicationCounter = new AtomicInteger(0);
|
||||
// True if the Mock RM is running, false otherwise.
|
||||
// This property allows us to write tests for specific scenario as Yarn RM
|
||||
// This property allows us to write tests for specific scenario as YARN RM
|
||||
// down e.g. network issue, failover.
|
||||
private boolean isRunning = true;
|
||||
private HashSet<ApplicationId> applicationMap = new HashSet<>();
|
||||
|
|
|
|||
|
|
@ -105,11 +105,11 @@ import static org.apache.hadoop.yarn.server.resourcemanager.resource.TestResourc
|
|||
|
||||
/**
|
||||
* <p>
|
||||
* Embedded Yarn minicluster for testcases that need to interact with a cluster.
|
||||
* Embedded YARN minicluster for testcases that need to interact with a cluster.
|
||||
* </p>
|
||||
* <p>
|
||||
* In a real cluster, resource request matching is done using the hostname, and
|
||||
* by default Yarn minicluster works in the exact same way as a real cluster.
|
||||
* by default YARN minicluster works in the exact same way as a real cluster.
|
||||
* </p>
|
||||
* <p>
|
||||
* If a testcase needs to use multiple nodes and exercise resource request
|
||||
|
|
|
|||
|
|
@ -263,7 +263,7 @@ public final class HBaseTimelineStorageUtils {
|
|||
}
|
||||
|
||||
/**
|
||||
* @param conf Yarn configuration. Used to see if there is an explicit config
|
||||
* @param conf YARN configuration. Used to see if there is an explicit config
|
||||
* pointing to the HBase config file to read. It should not be null
|
||||
* or a NullPointerException will be thrown.
|
||||
* @return a configuration with the HBase configuration from the classpath,
|
||||
|
|
|
|||
|
|
@ -613,7 +613,7 @@ public class WebAppProxyServlet extends HttpServlet {
|
|||
* <p>
|
||||
* <I>Do not remove</I>
|
||||
* <p>
|
||||
* Yarn isn't currently serializing this class, but findbugs
|
||||
* YARN isn't currently serializing this class, but findbugs
|
||||
* complains in its absence.
|
||||
*
|
||||
*
|
||||
|
|
|
|||
|
|
@ -15,9 +15,9 @@
|
|||
limitations under the License.
|
||||
-->
|
||||
|
||||
# Yarn UI
|
||||
# YARN UI
|
||||
|
||||
The Yarn UI is an Ember based web-app that provides visualization of the applications running on the Apache Hadoop YARN framework.
|
||||
The YARN UI is an Ember based web-app that provides visualization of the applications running on the Apache Hadoop YARN framework.
|
||||
|
||||
## Configurations
|
||||
|
||||
|
|
@ -49,7 +49,7 @@ You will need the following things properly installed on your computer.
|
|||
|
||||
**Warning: Do not edit the _package.json_ or _bower.json_ files manually. This could make them out-of-sync with the respective lock or shrinkwrap files.**
|
||||
|
||||
Yarn UI has replaced NPM with Yarn package manager. And hence Yarn would be used to manage dependencies defined in package.json.
|
||||
YARN UI has replaced NPM with Yarn package manager. And hence Yarn would be used to manage dependencies defined in package.json.
|
||||
|
||||
* Please use the Yarn and Bower command-line tools to add new dependencies. And the tool version must be same as those defined in Prerequisites section.
|
||||
* Once any dependency is added:
|
||||
|
|
|
|||
|
|
@ -21,7 +21,7 @@
|
|||
<head>
|
||||
<meta charset="utf-8">
|
||||
<meta http-equiv="X-UA-Compatible" content="IE=edge">
|
||||
<title>YarnUi</title>
|
||||
<title>YARN</title>
|
||||
<meta name="description" content="">
|
||||
<meta name="viewport" content="width=device-width, initial-scale=1">
|
||||
|
||||
|
|
|
|||
|
|
@ -16,7 +16,7 @@
|
|||
* limitations under the License.
|
||||
*/
|
||||
|
||||
module.exports = { // Yarn UI App configurations
|
||||
module.exports = { // YARN UI App configurations
|
||||
hosts: {
|
||||
localBaseAddress: "",
|
||||
timelineWebAddress: "localhost:8188",
|
||||
|
|
|
|||
Loading…
Reference in New Issue