use of org.apache.hadoop.mapreduce.v2.api.records.TaskType in project hadoop by apache.
the class TestCheckpointPreemptionPolicy method testPreemptionContract.
@Test
public void testPreemptionContract() {
final Map<ContainerId, TaskAttemptId> containers = assignedContainers;
AMPreemptionPolicy.Context mPctxt = new AMPreemptionPolicy.Context() {
@Override
public TaskAttemptId getTaskAttempt(ContainerId cId) {
return containers.get(cId);
}
@Override
public List<Container> getContainers(TaskType t) {
List<Container> p = new ArrayList<Container>();
for (Map.Entry<ContainerId, TaskAttemptId> ent : assignedContainers.entrySet()) {
if (ent.getValue().getTaskId().getTaskType().equals(t)) {
p.add(Container.newInstance(ent.getKey(), null, null, contToResourceMap.get(ent.getKey()), Priority.newInstance(0), null));
}
}
return p;
}
};
PreemptionMessage pM = generatePreemptionMessage(preemptedContainers, contToResourceMap, Resource.newInstance(minAlloc, 1), false);
CheckpointAMPreemptionPolicy policy = new CheckpointAMPreemptionPolicy();
policy.init(mActxt);
int supposedMemPreemption = (int) pM.getContract().getResourceRequest().get(0).getResourceRequest().getCapability().getMemorySize() * pM.getContract().getResourceRequest().get(0).getResourceRequest().getNumContainers();
// first round of preemption
policy.preempt(mPctxt, pM);
List<TaskAttemptId> preempting = validatePreemption(pM, policy, supposedMemPreemption);
// redundant message
policy.preempt(mPctxt, pM);
List<TaskAttemptId> preempting2 = validatePreemption(pM, policy, supposedMemPreemption);
// check that nothing got added
assert preempting2.equals(preempting);
// simulate 2 task completions/successful preemption
policy.handleCompletedContainer(preempting.get(0));
policy.handleCompletedContainer(preempting.get(1));
// remove from assignedContainers
Iterator<Map.Entry<ContainerId, TaskAttemptId>> it = assignedContainers.entrySet().iterator();
while (it.hasNext()) {
Map.Entry<ContainerId, TaskAttemptId> ent = it.next();
if (ent.getValue().equals(preempting.get(0)) || ent.getValue().equals(preempting.get(1)))
it.remove();
}
// one more message asking for preemption
policy.preempt(mPctxt, pM);
// triggers preemption of 2 more containers (i.e., the preemption set changes)
List<TaskAttemptId> preempting3 = validatePreemption(pM, policy, supposedMemPreemption);
assert preempting3.equals(preempting2) == false;
}
use of org.apache.hadoop.mapreduce.v2.api.records.TaskType in project hadoop by apache.
the class MRBuilderUtils method newTaskId.
public static TaskId newTaskId(JobId jobId, int id, TaskType taskType) {
TaskId taskId = Records.newRecord(TaskId.class);
taskId.setJobId(jobId);
taskId.setId(id);
taskId.setTaskType(taskType);
return taskId;
}
use of org.apache.hadoop.mapreduce.v2.api.records.TaskType in project hadoop by apache.
the class CLI method run.
public int run(String[] argv) throws Exception {
int exitCode = -1;
if (argv.length < 1) {
displayUsage("");
return exitCode;
}
// process arguments
String cmd = argv[0];
String submitJobFile = null;
String jobid = null;
String taskid = null;
String historyFileOrJobId = null;
String historyOutFile = null;
String historyOutFormat = HistoryViewer.HUMAN_FORMAT;
String counterGroupName = null;
String counterName = null;
JobPriority jp = null;
String taskType = null;
String taskState = null;
int fromEvent = 0;
int nEvents = 0;
int jpvalue = 0;
String configOutFile = null;
boolean getStatus = false;
boolean getCounter = false;
boolean killJob = false;
boolean listEvents = false;
boolean viewHistory = false;
boolean viewAllHistory = false;
boolean listJobs = false;
boolean listAllJobs = false;
boolean listActiveTrackers = false;
boolean listBlacklistedTrackers = false;
boolean displayTasks = false;
boolean killTask = false;
boolean failTask = false;
boolean setJobPriority = false;
boolean logs = false;
boolean downloadConfig = false;
if ("-submit".equals(cmd)) {
if (argv.length != 2) {
displayUsage(cmd);
return exitCode;
}
submitJobFile = argv[1];
} else if ("-status".equals(cmd)) {
if (argv.length != 2) {
displayUsage(cmd);
return exitCode;
}
jobid = argv[1];
getStatus = true;
} else if ("-counter".equals(cmd)) {
if (argv.length != 4) {
displayUsage(cmd);
return exitCode;
}
getCounter = true;
jobid = argv[1];
counterGroupName = argv[2];
counterName = argv[3];
} else if ("-kill".equals(cmd)) {
if (argv.length != 2) {
displayUsage(cmd);
return exitCode;
}
jobid = argv[1];
killJob = true;
} else if ("-set-priority".equals(cmd)) {
if (argv.length != 3) {
displayUsage(cmd);
return exitCode;
}
jobid = argv[1];
try {
jp = JobPriority.valueOf(argv[2]);
} catch (IllegalArgumentException iae) {
try {
jpvalue = Integer.parseInt(argv[2]);
} catch (NumberFormatException ne) {
LOG.info(ne);
displayUsage(cmd);
return exitCode;
}
}
setJobPriority = true;
} else if ("-events".equals(cmd)) {
if (argv.length != 4) {
displayUsage(cmd);
return exitCode;
}
jobid = argv[1];
fromEvent = Integer.parseInt(argv[2]);
nEvents = Integer.parseInt(argv[3]);
listEvents = true;
} else if ("-history".equals(cmd)) {
viewHistory = true;
if (argv.length < 2 || argv.length > 7) {
displayUsage(cmd);
return exitCode;
}
// Some arguments are optional while others are not, and some require
// second arguments. Due to this, the indexing can vary depending on
// what's specified and what's left out, as summarized in the below table:
// [all] <jobHistoryFile|jobId> [-outfile <file>] [-format <human|json>]
// 1 2 3 4 5 6
// 1 2 3 4
// 1 2 3 4
// 1 2
// 1 2 3 4 5
// 1 2 3
// 1 2 3
// 1
// "all" is optional, but comes first if specified
int index = 1;
if ("all".equals(argv[index])) {
index++;
viewAllHistory = true;
if (argv.length == 2) {
displayUsage(cmd);
return exitCode;
}
}
// Get the job history file or job id argument
historyFileOrJobId = argv[index++];
// "-outfile" is optional, but if specified requires a second argument
if (argv.length > index + 1 && "-outfile".equals(argv[index])) {
index++;
historyOutFile = argv[index++];
}
// "-format" is optional, but if specified required a second argument
if (argv.length > index + 1 && "-format".equals(argv[index])) {
index++;
historyOutFormat = argv[index++];
}
// Check for any extra arguments that don't belong here
if (argv.length > index) {
displayUsage(cmd);
return exitCode;
}
} else if ("-list".equals(cmd)) {
if (argv.length != 1 && !(argv.length == 2 && "all".equals(argv[1]))) {
displayUsage(cmd);
return exitCode;
}
if (argv.length == 2 && "all".equals(argv[1])) {
listAllJobs = true;
} else {
listJobs = true;
}
} else if ("-kill-task".equals(cmd)) {
if (argv.length != 2) {
displayUsage(cmd);
return exitCode;
}
killTask = true;
taskid = argv[1];
} else if ("-fail-task".equals(cmd)) {
if (argv.length != 2) {
displayUsage(cmd);
return exitCode;
}
failTask = true;
taskid = argv[1];
} else if ("-list-active-trackers".equals(cmd)) {
if (argv.length != 1) {
displayUsage(cmd);
return exitCode;
}
listActiveTrackers = true;
} else if ("-list-blacklisted-trackers".equals(cmd)) {
if (argv.length != 1) {
displayUsage(cmd);
return exitCode;
}
listBlacklistedTrackers = true;
} else if ("-list-attempt-ids".equals(cmd)) {
if (argv.length != 4) {
displayUsage(cmd);
return exitCode;
}
jobid = argv[1];
taskType = argv[2];
taskState = argv[3];
displayTasks = true;
if (!taskTypes.contains(org.apache.hadoop.util.StringUtils.toUpperCase(taskType))) {
System.out.println("Error: Invalid task-type: " + taskType);
displayUsage(cmd);
return exitCode;
}
if (!taskStates.contains(org.apache.hadoop.util.StringUtils.toLowerCase(taskState))) {
System.out.println("Error: Invalid task-state: " + taskState);
displayUsage(cmd);
return exitCode;
}
} else if ("-logs".equals(cmd)) {
if (argv.length == 2 || argv.length == 3) {
logs = true;
jobid = argv[1];
if (argv.length == 3) {
taskid = argv[2];
} else {
taskid = null;
}
} else {
displayUsage(cmd);
return exitCode;
}
} else if ("-config".equals(cmd)) {
downloadConfig = true;
if (argv.length != 3) {
displayUsage(cmd);
return exitCode;
}
jobid = argv[1];
configOutFile = argv[2];
} else {
displayUsage(cmd);
return exitCode;
}
// initialize cluster
cluster = createCluster();
// Submit the request
try {
if (submitJobFile != null) {
Job job = Job.getInstance(new JobConf(submitJobFile));
job.submit();
System.out.println("Created job " + job.getJobID());
exitCode = 0;
} else if (getStatus) {
Job job = getJob(JobID.forName(jobid));
if (job == null) {
System.out.println("Could not find job " + jobid);
} else {
Counters counters = job.getCounters();
System.out.println();
System.out.println(job);
if (counters != null) {
System.out.println(counters);
} else {
System.out.println("Counters not available. Job is retired.");
}
exitCode = 0;
}
} else if (getCounter) {
Job job = getJob(JobID.forName(jobid));
if (job == null) {
System.out.println("Could not find job " + jobid);
} else {
Counters counters = job.getCounters();
if (counters == null) {
System.out.println("Counters not available for retired job " + jobid);
exitCode = -1;
} else {
System.out.println(getCounter(counters, counterGroupName, counterName));
exitCode = 0;
}
}
} else if (killJob) {
Job job = getJob(JobID.forName(jobid));
if (job == null) {
System.out.println("Could not find job " + jobid);
} else {
JobStatus jobStatus = job.getStatus();
if (jobStatus.getState() == JobStatus.State.FAILED) {
System.out.println("Could not mark the job " + jobid + " as killed, as it has already failed.");
exitCode = -1;
} else if (jobStatus.getState() == JobStatus.State.KILLED) {
System.out.println("The job " + jobid + " has already been killed.");
exitCode = -1;
} else if (jobStatus.getState() == JobStatus.State.SUCCEEDED) {
System.out.println("Could not kill the job " + jobid + ", as it has already succeeded.");
exitCode = -1;
} else {
job.killJob();
System.out.println("Killed job " + jobid);
exitCode = 0;
}
}
} else if (setJobPriority) {
Job job = getJob(JobID.forName(jobid));
if (job == null) {
System.out.println("Could not find job " + jobid);
} else {
if (jp != null) {
job.setPriority(jp);
} else {
job.setPriorityAsInteger(jpvalue);
}
System.out.println("Changed job priority.");
exitCode = 0;
}
} else if (viewHistory) {
// it's a Job ID
if (historyFileOrJobId.endsWith(".jhist")) {
viewHistory(historyFileOrJobId, viewAllHistory, historyOutFile, historyOutFormat);
exitCode = 0;
} else {
Job job = getJob(JobID.forName(historyFileOrJobId));
if (job == null) {
System.out.println("Could not find job " + jobid);
} else {
String historyUrl = job.getHistoryUrl();
if (historyUrl == null || historyUrl.isEmpty()) {
System.out.println("History file for job " + historyFileOrJobId + " is currently unavailable.");
} else {
viewHistory(historyUrl, viewAllHistory, historyOutFile, historyOutFormat);
exitCode = 0;
}
}
}
} else if (listEvents) {
Job job = getJob(JobID.forName(jobid));
if (job == null) {
System.out.println("Could not find job " + jobid);
} else {
listEvents(job, fromEvent, nEvents);
exitCode = 0;
}
} else if (listJobs) {
listJobs(cluster);
exitCode = 0;
} else if (listAllJobs) {
listAllJobs(cluster);
exitCode = 0;
} else if (listActiveTrackers) {
listActiveTrackers(cluster);
exitCode = 0;
} else if (listBlacklistedTrackers) {
listBlacklistedTrackers(cluster);
exitCode = 0;
} else if (displayTasks) {
Job job = getJob(JobID.forName(jobid));
if (job == null) {
System.out.println("Could not find job " + jobid);
} else {
displayTasks(getJob(JobID.forName(jobid)), taskType, taskState);
exitCode = 0;
}
} else if (killTask) {
TaskAttemptID taskID = TaskAttemptID.forName(taskid);
Job job = getJob(taskID.getJobID());
if (job == null) {
System.out.println("Could not find job " + jobid);
} else if (job.killTask(taskID, false)) {
System.out.println("Killed task " + taskid);
exitCode = 0;
} else {
System.out.println("Could not kill task " + taskid);
exitCode = -1;
}
} else if (failTask) {
TaskAttemptID taskID = TaskAttemptID.forName(taskid);
Job job = getJob(taskID.getJobID());
if (job == null) {
System.out.println("Could not find job " + jobid);
} else if (job.killTask(taskID, true)) {
System.out.println("Killed task " + taskID + " by failing it");
exitCode = 0;
} else {
System.out.println("Could not fail task " + taskid);
exitCode = -1;
}
} else if (logs) {
JobID jobID = JobID.forName(jobid);
if (getJob(jobID) == null) {
System.out.println("Could not find job " + jobid);
} else {
try {
TaskAttemptID taskAttemptID = TaskAttemptID.forName(taskid);
LogParams logParams = cluster.getLogParams(jobID, taskAttemptID);
LogCLIHelpers logDumper = new LogCLIHelpers();
logDumper.setConf(getConf());
exitCode = logDumper.dumpAContainersLogs(logParams.getApplicationId(), logParams.getContainerId(), logParams.getNodeId(), logParams.getOwner());
} catch (IOException e) {
if (e instanceof RemoteException) {
throw e;
}
System.out.println(e.getMessage());
}
}
} else if (downloadConfig) {
Job job = getJob(JobID.forName(jobid));
if (job == null) {
System.out.println("Could not find job " + jobid);
} else {
String jobFile = job.getJobFile();
if (jobFile == null || jobFile.isEmpty()) {
System.out.println("Config file for job " + jobFile + " could not be found.");
} else {
Path configPath = new Path(jobFile);
FileSystem fs = FileSystem.get(getConf());
fs.copyToLocalFile(configPath, new Path(configOutFile));
exitCode = 0;
}
}
}
} catch (RemoteException re) {
IOException unwrappedException = re.unwrapRemoteException();
if (unwrappedException instanceof AccessControlException) {
System.out.println(unwrappedException.getMessage());
} else {
throw re;
}
} finally {
cluster.close();
}
return exitCode;
}
use of org.apache.hadoop.mapreduce.v2.api.records.TaskType in project hadoop by apache.
the class TestContainerLauncherImpl method makeTaskAttemptId.
public static TaskAttemptId makeTaskAttemptId(long ts, int appId, int taskId, TaskType taskType, int id) {
ApplicationId aID = ApplicationId.newInstance(ts, appId);
JobId jID = MRBuilderUtils.newJobId(aID, id);
TaskId tID = MRBuilderUtils.newTaskId(jID, taskId, taskType);
return MRBuilderUtils.newTaskAttemptId(tID, id);
}
use of org.apache.hadoop.mapreduce.v2.api.records.TaskType in project hadoop by apache.
the class DefaultSpeculator method maybeScheduleASpeculation.
private int maybeScheduleASpeculation(TaskType type) {
int successes = 0;
long now = clock.getTime();
ConcurrentMap<JobId, AtomicInteger> containerNeeds = type == TaskType.MAP ? mapContainerNeeds : reduceContainerNeeds;
for (ConcurrentMap.Entry<JobId, AtomicInteger> jobEntry : containerNeeds.entrySet()) {
// container prematurely.
if (jobEntry.getValue().get() > 0) {
continue;
}
int numberSpeculationsAlready = 0;
int numberRunningTasks = 0;
// loop through the tasks of the kind
Job job = context.getJob(jobEntry.getKey());
Map<TaskId, Task> tasks = job.getTasks(type);
int numberAllowedSpeculativeTasks = (int) Math.max(minimumAllowedSpeculativeTasks, proportionTotalTasksSpeculatable * tasks.size());
TaskId bestTaskID = null;
long bestSpeculationValue = -1L;
// TODO track the tasks that are potentially worth looking at
for (Map.Entry<TaskId, Task> taskEntry : tasks.entrySet()) {
long mySpeculationValue = speculationValue(taskEntry.getKey(), now);
if (mySpeculationValue == ALREADY_SPECULATING) {
++numberSpeculationsAlready;
}
if (mySpeculationValue != NOT_RUNNING) {
++numberRunningTasks;
}
if (mySpeculationValue > bestSpeculationValue) {
bestTaskID = taskEntry.getKey();
bestSpeculationValue = mySpeculationValue;
}
}
numberAllowedSpeculativeTasks = (int) Math.max(numberAllowedSpeculativeTasks, proportionRunningTasksSpeculatable * numberRunningTasks);
// If we found a speculation target, fire it off
if (bestTaskID != null && numberAllowedSpeculativeTasks > numberSpeculationsAlready) {
addSpeculativeAttempt(bestTaskID);
++successes;
}
}
return successes;
}
Aggregations