Search in sources :

Example 96 with Task

use of org.apache.hadoop.mapreduce.v2.app.job.Task in project hadoop by apache.

the class MockJobs method newTasks.

public static Map<TaskId, Task> newTasks(JobId jid, int n, int m, boolean hasFailedTasks) {
    Map<TaskId, Task> map = Maps.newHashMap();
    for (int i = 0; i < n; ++i) {
        Task task = newTask(jid, i, m, hasFailedTasks);
        map.put(task.getID(), task);
    }
    return map;
}
Also used : Task(org.apache.hadoop.mapreduce.v2.app.job.Task) TaskId(org.apache.hadoop.mapreduce.v2.api.records.TaskId)

Example 97 with Task

use of org.apache.hadoop.mapreduce.v2.app.job.Task 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;
}
Also used : Path(org.apache.hadoop.fs.Path) TaskAttemptID(org.apache.hadoop.mapreduce.TaskAttemptID) JobPriority(org.apache.hadoop.mapreduce.JobPriority) AccessControlException(org.apache.hadoop.security.AccessControlException) IOException(java.io.IOException) LogParams(org.apache.hadoop.mapreduce.v2.LogParams) JobStatus(org.apache.hadoop.mapreduce.JobStatus) FileSystem(org.apache.hadoop.fs.FileSystem) LogCLIHelpers(org.apache.hadoop.yarn.logaggregation.LogCLIHelpers) Counters(org.apache.hadoop.mapreduce.Counters) Job(org.apache.hadoop.mapreduce.Job) RemoteException(org.apache.hadoop.ipc.RemoteException) JobConf(org.apache.hadoop.mapred.JobConf) JobID(org.apache.hadoop.mapreduce.JobID)

Example 98 with Task

use of org.apache.hadoop.mapreduce.v2.app.job.Task in project hadoop by apache.

the class ClientServiceDelegate method getLogFilePath.

public LogParams getLogFilePath(JobID oldJobID, TaskAttemptID oldTaskAttemptID) throws IOException {
    org.apache.hadoop.mapreduce.v2.api.records.JobId jobId = TypeConverter.toYarn(oldJobID);
    GetJobReportRequest request = recordFactory.newRecordInstance(GetJobReportRequest.class);
    request.setJobId(jobId);
    JobReport report = ((GetJobReportResponse) invoke("getJobReport", GetJobReportRequest.class, request)).getJobReport();
    if (EnumSet.of(JobState.SUCCEEDED, JobState.FAILED, JobState.KILLED, JobState.ERROR).contains(report.getJobState())) {
        if (oldTaskAttemptID != null) {
            GetTaskAttemptReportRequest taRequest = recordFactory.newRecordInstance(GetTaskAttemptReportRequest.class);
            taRequest.setTaskAttemptId(TypeConverter.toYarn(oldTaskAttemptID));
            TaskAttemptReport taReport = ((GetTaskAttemptReportResponse) invoke("getTaskAttemptReport", GetTaskAttemptReportRequest.class, taRequest)).getTaskAttemptReport();
            if (taReport.getContainerId() == null || taReport.getNodeManagerHost() == null) {
                throw new IOException("Unable to get log information for task: " + oldTaskAttemptID);
            }
            return new LogParams(taReport.getContainerId().toString(), taReport.getContainerId().getApplicationAttemptId().getApplicationId().toString(), NodeId.newInstance(taReport.getNodeManagerHost(), taReport.getNodeManagerPort()).toString(), report.getUser());
        } else {
            if (report.getAMInfos() == null || report.getAMInfos().size() == 0) {
                throw new IOException("Unable to get log information for job: " + oldJobID);
            }
            AMInfo amInfo = report.getAMInfos().get(report.getAMInfos().size() - 1);
            return new LogParams(amInfo.getContainerId().toString(), amInfo.getAppAttemptId().getApplicationId().toString(), NodeId.newInstance(amInfo.getNodeManagerHost(), amInfo.getNodeManagerPort()).toString(), report.getUser());
        }
    } else {
        throw new IOException("Cannot get log path for a in-progress job");
    }
}
Also used : GetTaskAttemptReportRequest(org.apache.hadoop.mapreduce.v2.api.protocolrecords.GetTaskAttemptReportRequest) IOException(java.io.IOException) GetJobReportRequest(org.apache.hadoop.mapreduce.v2.api.protocolrecords.GetJobReportRequest) JobReport(org.apache.hadoop.mapreduce.v2.api.records.JobReport) GetJobReportResponse(org.apache.hadoop.mapreduce.v2.api.protocolrecords.GetJobReportResponse) LogParams(org.apache.hadoop.mapreduce.v2.LogParams) AMInfo(org.apache.hadoop.mapreduce.v2.api.records.AMInfo) TaskAttemptReport(org.apache.hadoop.mapreduce.v2.api.records.TaskAttemptReport) GetTaskAttemptReportResponse(org.apache.hadoop.mapreduce.v2.api.protocolrecords.GetTaskAttemptReportResponse)

Example 99 with Task

use of org.apache.hadoop.mapreduce.v2.app.job.Task in project hadoop by apache.

the class TestHsWebServicesAttempts method testTaskAttemptId.

@Test
public void testTaskAttemptId() throws JSONException, Exception {
    WebResource r = resource();
    Map<JobId, Job> jobsMap = appContext.getAllJobs();
    for (JobId id : jobsMap.keySet()) {
        String jobId = MRApps.toString(id);
        for (Task task : jobsMap.get(id).getTasks().values()) {
            String tid = MRApps.toString(task.getID());
            for (TaskAttempt att : task.getAttempts().values()) {
                TaskAttemptId attemptid = att.getID();
                String attid = MRApps.toString(attemptid);
                ClientResponse response = r.path("ws").path("v1").path("history").path("mapreduce").path("jobs").path(jobId).path("tasks").path(tid).path("attempts").path(attid).accept(MediaType.APPLICATION_JSON).get(ClientResponse.class);
                assertEquals(MediaType.APPLICATION_JSON_TYPE + "; " + JettyUtils.UTF_8, response.getType().toString());
                JSONObject json = response.getEntity(JSONObject.class);
                assertEquals("incorrect number of elements", 1, json.length());
                JSONObject info = json.getJSONObject("taskAttempt");
                verifyHsTaskAttempt(info, att, task.getType());
            }
        }
    }
}
Also used : ClientResponse(com.sun.jersey.api.client.ClientResponse) Task(org.apache.hadoop.mapreduce.v2.app.job.Task) JSONObject(org.codehaus.jettison.json.JSONObject) TaskAttemptId(org.apache.hadoop.mapreduce.v2.api.records.TaskAttemptId) WebResource(com.sun.jersey.api.client.WebResource) TaskAttempt(org.apache.hadoop.mapreduce.v2.app.job.TaskAttempt) Job(org.apache.hadoop.mapreduce.v2.app.job.Job) JobId(org.apache.hadoop.mapreduce.v2.api.records.JobId) Test(org.junit.Test)

Example 100 with Task

use of org.apache.hadoop.mapreduce.v2.app.job.Task in project hadoop by apache.

the class TestHsWebServicesAttempts method testTaskAttemptIdErrorGeneric.

private void testTaskAttemptIdErrorGeneric(String attid, String error) throws JSONException, Exception {
    WebResource r = resource();
    Map<JobId, Job> jobsMap = appContext.getAllJobs();
    for (JobId id : jobsMap.keySet()) {
        String jobId = MRApps.toString(id);
        for (Task task : jobsMap.get(id).getTasks().values()) {
            String tid = MRApps.toString(task.getID());
            try {
                r.path("ws").path("v1").path("history").path("mapreduce").path("jobs").path(jobId).path("tasks").path(tid).path("attempts").path(attid).accept(MediaType.APPLICATION_JSON).get(JSONObject.class);
                fail("should have thrown exception on invalid uri");
            } catch (UniformInterfaceException ue) {
                ClientResponse response = ue.getResponse();
                assertResponseStatusCode(Status.NOT_FOUND, response.getStatusInfo());
                assertEquals(MediaType.APPLICATION_JSON_TYPE + "; " + JettyUtils.UTF_8, response.getType().toString());
                JSONObject msg = response.getEntity(JSONObject.class);
                JSONObject exception = msg.getJSONObject("RemoteException");
                assertEquals("incorrect number of elements", 3, exception.length());
                String message = exception.getString("message");
                String type = exception.getString("exception");
                String classname = exception.getString("javaClassName");
                WebServicesTestUtils.checkStringMatch("exception message", error, message);
                WebServicesTestUtils.checkStringMatch("exception type", "NotFoundException", type);
                WebServicesTestUtils.checkStringMatch("exception classname", "org.apache.hadoop.yarn.webapp.NotFoundException", classname);
            }
        }
    }
}
Also used : ClientResponse(com.sun.jersey.api.client.ClientResponse) Task(org.apache.hadoop.mapreduce.v2.app.job.Task) UniformInterfaceException(com.sun.jersey.api.client.UniformInterfaceException) JSONObject(org.codehaus.jettison.json.JSONObject) WebResource(com.sun.jersey.api.client.WebResource) Job(org.apache.hadoop.mapreduce.v2.app.job.Job) JobId(org.apache.hadoop.mapreduce.v2.api.records.JobId)

Aggregations

Task (org.apache.hadoop.mapreduce.v2.app.job.Task)157 Test (org.junit.Test)153 Job (org.apache.hadoop.mapreduce.v2.app.job.Job)150 JobId (org.apache.hadoop.mapreduce.v2.api.records.JobId)107 TaskAttempt (org.apache.hadoop.mapreduce.v2.app.job.TaskAttempt)94 TaskId (org.apache.hadoop.mapreduce.v2.api.records.TaskId)79 TaskAttemptId (org.apache.hadoop.mapreduce.v2.api.records.TaskAttemptId)73 Configuration (org.apache.hadoop.conf.Configuration)68 ClientResponse (com.sun.jersey.api.client.ClientResponse)56 WebResource (com.sun.jersey.api.client.WebResource)56 TaskAttemptEvent (org.apache.hadoop.mapreduce.v2.app.job.event.TaskAttemptEvent)52 JSONObject (org.codehaus.jettison.json.JSONObject)46 AppContext (org.apache.hadoop.mapreduce.v2.app.AppContext)25 ApplicationId (org.apache.hadoop.yarn.api.records.ApplicationId)23 Path (org.apache.hadoop.fs.Path)22 MapTaskAttemptImpl (org.apache.hadoop.mapred.MapTaskAttemptImpl)20 ContainerId (org.apache.hadoop.yarn.api.records.ContainerId)20 HashMap (java.util.HashMap)19 ApplicationAttemptId (org.apache.hadoop.yarn.api.records.ApplicationAttemptId)19 JobConf (org.apache.hadoop.mapred.JobConf)16