public virtual void TestMapRedExecutionEnv()
{
// test if the env variable can be set
try
{
// Application environment
IDictionary <string, string> environment = new Dictionary <string, string>();
string setupHadoopHomeCommand = Shell.Windows ? "HADOOP_COMMON_HOME=C:\\fake\\PATH\\to\\hadoop\\common\\home"
: "HADOOP_COMMON_HOME=/fake/path/to/hadoop/common/home";
MRApps.SetEnvFromInputString(environment, setupHadoopHomeCommand, conf);
// Add the env variables passed by the admin
MRApps.SetEnvFromInputString(environment, conf.Get(MRJobConfig.MapredAdminUserEnv
, MRJobConfig.DefaultMapredAdminUserEnv), conf);
string executionPaths = environment[Shell.Windows ? "PATH" : "LD_LIBRARY_PATH"];
string toFind = Shell.Windows ? "C:\\fake\\PATH\\to\\hadoop\\common\\home\\bin" :
"/fake/path/to/hadoop/common/home/lib/native";
// Ensure execution PATH/LD_LIBRARY_PATH set up pointing to hadoop lib
NUnit.Framework.Assert.IsTrue("execution path does not include the hadoop lib location "
+ toFind, executionPaths.Contains(toFind));
}
catch (Exception e)
{
Sharpen.Runtime.PrintStackTrace(e);
NUnit.Framework.Assert.Fail("Exception in testing execution environment for MapReduce task"
);
TearDown();
}
// now launch a mapreduce job to ensure that the child
// also gets the configured setting for hadoop lib
try
{
JobConf conf = new JobConf(mr.GetConfig());
// initialize input, output directories
Path inDir = new Path("input");
Path outDir = new Path("output");
string input = "The input";
// set config to use the ExecutionEnvCheckMapClass map class
Configure(conf, inDir, outDir, input, typeof(TestMiniMRChildTask.ExecutionEnvCheckMapClass
), typeof(IdentityReducer));
LaunchTest(conf, inDir, outDir, input);
}
catch (Exception e)
{
Sharpen.Runtime.PrintStackTrace(e);
NUnit.Framework.Assert.Fail("Exception in testing propagation of env setting to child task"
);
TearDown();
}
}
public static void SetVMEnv(IDictionary <string, string> environment, Task task)
{
JobConf conf = task.conf;
// Add the env variables passed by the user
string mapredChildEnv = GetChildEnv(conf, task.IsMapTask());
MRApps.SetEnvFromInputString(environment, mapredChildEnv, conf);
// Set logging level in the environment.
// This is so that, if the child forks another "bin/hadoop" (common in
// streaming) it will have the correct loglevel.
environment["HADOOP_ROOT_LOGGER"] = MRApps.GetChildLogLevel(conf, task.IsMapTask(
)) + ",console";
// TODO: The following is useful for instance in streaming tasks. Should be
// set in ApplicationMaster's env by the RM.
string hadoopClientOpts = Runtime.Getenv("HADOOP_CLIENT_OPTS");
if (hadoopClientOpts == null)
{
hadoopClientOpts = string.Empty;
}
else
{
hadoopClientOpts = hadoopClientOpts + " ";
}
environment["HADOOP_CLIENT_OPTS"] = hadoopClientOpts;
// setEnvFromInputString above will add env variable values from
// mapredChildEnv to existing variables. We want to overwrite
// HADOOP_ROOT_LOGGER and HADOOP_CLIENT_OPTS if the user set it explicitly.
IDictionary <string, string> tmpEnv = new Dictionary <string, string>();
MRApps.SetEnvFromInputString(tmpEnv, mapredChildEnv, conf);
string[] keys = new string[] { "HADOOP_ROOT_LOGGER", "HADOOP_CLIENT_OPTS" };
foreach (string key in keys)
{
if (tmpEnv.Contains(key))
{
environment[key] = tmpEnv[key];
}
}
// Add stdout/stderr env
environment[MRJobConfig.StdoutLogfileEnv] = GetTaskLogFile(TaskLog.LogName.Stdout
);
environment[MRJobConfig.StderrLogfileEnv] = GetTaskLogFile(TaskLog.LogName.Stderr
);
}
/// <exception cref="System.IO.IOException"/>
public virtual ApplicationSubmissionContext CreateApplicationSubmissionContext(Configuration
jobConf, string jobSubmitDir, Credentials ts)
{
ApplicationId applicationId = resMgrDelegate.GetApplicationId();
// Setup resource requirements
Resource capability = recordFactory.NewRecordInstance <Resource>();
capability.SetMemory(conf.GetInt(MRJobConfig.MrAmVmemMb, MRJobConfig.DefaultMrAmVmemMb
));
capability.SetVirtualCores(conf.GetInt(MRJobConfig.MrAmCpuVcores, MRJobConfig.DefaultMrAmCpuVcores
));
Log.Debug("AppMaster capability = " + capability);
// Setup LocalResources
IDictionary <string, LocalResource> localResources = new Dictionary <string, LocalResource
>();
Path jobConfPath = new Path(jobSubmitDir, MRJobConfig.JobConfFile);
URL yarnUrlForJobSubmitDir = ConverterUtils.GetYarnUrlFromPath(defaultFileContext
.GetDefaultFileSystem().ResolvePath(defaultFileContext.MakeQualified(new Path(jobSubmitDir
))));
Log.Debug("Creating setup context, jobSubmitDir url is " + yarnUrlForJobSubmitDir
);
localResources[MRJobConfig.JobConfFile] = CreateApplicationResource(defaultFileContext
, jobConfPath, LocalResourceType.File);
if (jobConf.Get(MRJobConfig.Jar) != null)
{
Path jobJarPath = new Path(jobConf.Get(MRJobConfig.Jar));
LocalResource rc = CreateApplicationResource(FileContext.GetFileContext(jobJarPath
.ToUri(), jobConf), jobJarPath, LocalResourceType.Pattern);
string pattern = conf.GetPattern(JobContext.JarUnpackPattern, JobConf.UnpackJarPatternDefault
).Pattern();
rc.SetPattern(pattern);
localResources[MRJobConfig.JobJar] = rc;
}
else
{
// Job jar may be null. For e.g, for pipes, the job jar is the hadoop
// mapreduce jar itself which is already on the classpath.
Log.Info("Job jar is not present. " + "Not adding any jar to the list of resources."
);
}
// TODO gross hack
foreach (string s in new string[] { MRJobConfig.JobSplit, MRJobConfig.JobSplitMetainfo })
{
localResources[MRJobConfig.JobSubmitDir + "/" + s] = CreateApplicationResource(defaultFileContext
, new Path(jobSubmitDir, s), LocalResourceType.File);
}
// Setup security tokens
DataOutputBuffer dob = new DataOutputBuffer();
ts.WriteTokenStorageToStream(dob);
ByteBuffer securityTokens = ByteBuffer.Wrap(dob.GetData(), 0, dob.GetLength());
// Setup the command to run the AM
IList <string> vargs = new AList <string>(8);
vargs.AddItem(MRApps.CrossPlatformifyMREnv(jobConf, ApplicationConstants.Environment
.JavaHome) + "/bin/java");
Path amTmpDir = new Path(MRApps.CrossPlatformifyMREnv(conf, ApplicationConstants.Environment
.Pwd), YarnConfiguration.DefaultContainerTempDir);
vargs.AddItem("-Djava.io.tmpdir=" + amTmpDir);
MRApps.AddLog4jSystemProperties(null, vargs, conf);
// Check for Java Lib Path usage in MAP and REDUCE configs
WarnForJavaLibPath(conf.Get(MRJobConfig.MapJavaOpts, string.Empty), "map", MRJobConfig
.MapJavaOpts, MRJobConfig.MapEnv);
WarnForJavaLibPath(conf.Get(MRJobConfig.MapredMapAdminJavaOpts, string.Empty), "map"
, MRJobConfig.MapredMapAdminJavaOpts, MRJobConfig.MapredAdminUserEnv);
WarnForJavaLibPath(conf.Get(MRJobConfig.ReduceJavaOpts, string.Empty), "reduce",
MRJobConfig.ReduceJavaOpts, MRJobConfig.ReduceEnv);
WarnForJavaLibPath(conf.Get(MRJobConfig.MapredReduceAdminJavaOpts, string.Empty),
"reduce", MRJobConfig.MapredReduceAdminJavaOpts, MRJobConfig.MapredAdminUserEnv);
// Add AM admin command opts before user command opts
// so that it can be overridden by user
string mrAppMasterAdminOptions = conf.Get(MRJobConfig.MrAmAdminCommandOpts, MRJobConfig
.DefaultMrAmAdminCommandOpts);
WarnForJavaLibPath(mrAppMasterAdminOptions, "app master", MRJobConfig.MrAmAdminCommandOpts
, MRJobConfig.MrAmAdminUserEnv);
vargs.AddItem(mrAppMasterAdminOptions);
// Add AM user command opts
string mrAppMasterUserOptions = conf.Get(MRJobConfig.MrAmCommandOpts, MRJobConfig
.DefaultMrAmCommandOpts);
WarnForJavaLibPath(mrAppMasterUserOptions, "app master", MRJobConfig.MrAmCommandOpts
, MRJobConfig.MrAmEnv);
vargs.AddItem(mrAppMasterUserOptions);
if (jobConf.GetBoolean(MRJobConfig.MrAmProfile, MRJobConfig.DefaultMrAmProfile))
{
string profileParams = jobConf.Get(MRJobConfig.MrAmProfileParams, MRJobConfig.DefaultTaskProfileParams
);
if (profileParams != null)
{
vargs.AddItem(string.Format(profileParams, ApplicationConstants.LogDirExpansionVar
+ Path.Separator + TaskLog.LogName.Profile));
}
}
vargs.AddItem(MRJobConfig.ApplicationMasterClass);
vargs.AddItem("1>" + ApplicationConstants.LogDirExpansionVar + Path.Separator + ApplicationConstants
.Stdout);
vargs.AddItem("2>" + ApplicationConstants.LogDirExpansionVar + Path.Separator + ApplicationConstants
.Stderr);
Vector <string> vargsFinal = new Vector <string>(8);
// Final command
StringBuilder mergedCommand = new StringBuilder();
foreach (CharSequence str in vargs)
{
mergedCommand.Append(str).Append(" ");
}
vargsFinal.AddItem(mergedCommand.ToString());
Log.Debug("Command to launch container for ApplicationMaster is : " + mergedCommand
);
// Setup the CLASSPATH in environment
// i.e. add { Hadoop jars, job jar, CWD } to classpath.
IDictionary <string, string> environment = new Dictionary <string, string>();
MRApps.SetClasspath(environment, conf);
// Shell
environment[ApplicationConstants.Environment.Shell.ToString()] = conf.Get(MRJobConfig
.MapredAdminUserShell, MRJobConfig.DefaultShell);
// Add the container working directory at the front of LD_LIBRARY_PATH
MRApps.AddToEnvironment(environment, ApplicationConstants.Environment.LdLibraryPath
.ToString(), MRApps.CrossPlatformifyMREnv(conf, ApplicationConstants.Environment
.Pwd), conf);
// Setup the environment variables for Admin first
MRApps.SetEnvFromInputString(environment, conf.Get(MRJobConfig.MrAmAdminUserEnv),
conf);
// Setup the environment variables (LD_LIBRARY_PATH, etc)
MRApps.SetEnvFromInputString(environment, conf.Get(MRJobConfig.MrAmEnv), conf);
// Parse distributed cache
MRApps.SetupDistributedCache(jobConf, localResources);
IDictionary <ApplicationAccessType, string> acls = new Dictionary <ApplicationAccessType
, string>(2);
acls[ApplicationAccessType.ViewApp] = jobConf.Get(MRJobConfig.JobAclViewJob, MRJobConfig
.DefaultJobAclViewJob);
acls[ApplicationAccessType.ModifyApp] = jobConf.Get(MRJobConfig.JobAclModifyJob,
MRJobConfig.DefaultJobAclModifyJob);
// Setup ContainerLaunchContext for AM container
ContainerLaunchContext amContainer = ContainerLaunchContext.NewInstance(localResources
, environment, vargsFinal, null, securityTokens, acls);
ICollection <string> tagsFromConf = jobConf.GetTrimmedStringCollection(MRJobConfig
.JobTags);
// Set up the ApplicationSubmissionContext
ApplicationSubmissionContext appContext = recordFactory.NewRecordInstance <ApplicationSubmissionContext
>();
appContext.SetApplicationId(applicationId);
// ApplicationId
appContext.SetQueue(jobConf.Get(JobContext.QueueName, YarnConfiguration.DefaultQueueName
));
// Queue name
// add reservationID if present
ReservationId reservationID = null;
try
{
reservationID = ReservationId.ParseReservationId(jobConf.Get(JobContext.ReservationId
));
}
catch (FormatException)
{
// throw exception as reservationid as is invalid
string errMsg = "Invalid reservationId: " + jobConf.Get(JobContext.ReservationId)
+ " specified for the app: " + applicationId;
Log.Warn(errMsg);
throw new IOException(errMsg);
}
if (reservationID != null)
{
appContext.SetReservationID(reservationID);
Log.Info("SUBMITTING ApplicationSubmissionContext app:" + applicationId + " to queue:"
+ appContext.GetQueue() + " with reservationId:" + appContext.GetReservationID(
));
}
appContext.SetApplicationName(jobConf.Get(JobContext.JobName, YarnConfiguration.DefaultApplicationName
));
// Job name
appContext.SetCancelTokensWhenComplete(conf.GetBoolean(MRJobConfig.JobCancelDelegationToken
, true));
appContext.SetAMContainerSpec(amContainer);
// AM Container
appContext.SetMaxAppAttempts(conf.GetInt(MRJobConfig.MrAmMaxAttempts, MRJobConfig
.DefaultMrAmMaxAttempts));
appContext.SetResource(capability);
appContext.SetApplicationType(MRJobConfig.MrApplicationType);
if (tagsFromConf != null && !tagsFromConf.IsEmpty())
{
appContext.SetApplicationTags(new HashSet <string>(tagsFromConf));
}
return(appContext);
}
