AnsweredAssumed Answered

MR job start: "Failed to create directory /tmp/mapr-hadoop/mapred/local/taskTracker/root ..."

Question asked by chris_almond on Aug 27, 2012
Latest reply on Aug 28, 2012 by chris_almond
I'm trying to start a wordcount example MR job on a healthy M5 cluster from a linux client with the MapR client installed and setup.  I get the same failure mode when trying to run from one of the cluster nodes too.

    
    [root@almond-fedora taskTracker]#  hadoop jar /opt/mapr/hadoop/hadoop-0.20.2/hadoop-0.20.2-dev-examples.jar wordcount /mapr/TestMapRCluster/myvolume/in /mapr/TestMapRCluster/myvolume/out
    12/08/27 14:02:56 INFO fs.JobTrackerWatcher: Current running JobTracker is: vdc-2950-4.irad.net/10.35.225.206:9001
    2012-08-27 14:02:56,2259 ERROR Cidcache fs/client/fileclient/cc/cidcache.cc:117 Thread: 140400142432000 ContainerInfo reported with a server with no hostname for cid: 2074
    2012-08-27 14:02:56,2263 ERROR Cidcache fs/client/fileclient/cc/cidcache.cc:117 Thread: 140400142432000 ContainerInfo reported with a server with no hostname for cid: 2074
    2012-08-27 14:02:56,2264 ERROR Cidcache fs/client/fileclient/cc/cidcache.cc:117 Thread: 140400142432000 ContainerInfo reported with a server with no hostname for cid: 2074
    12/08/27 14:02:56 INFO input.FileInputFormat: Total input paths to process : 1
    12/08/27 14:02:56 INFO util.NativeCodeLoader: Loaded the native-hadoop library
    12/08/27 14:02:56 WARN snappy.LoadSnappy: Snappy native library not loaded
    12/08/27 14:02:56 INFO mapred.JobClient: Creating job's output directory at /mapr/TestMapRCluster/myvolume/out
    12/08/27 14:02:56 INFO mapred.JobClient: Creating job's user history location directory at /mapr/TestMapRCluster/myvolume/out/_logs
    12/08/27 14:02:56 INFO mapred.JobClient: Running job: job_201208060924_0012
    12/08/27 14:02:57 INFO mapred.JobClient:  map 0% reduce 0%
    12/08/27 14:02:57 INFO mapred.JobClient: Task Id : attempt_201208060924_0012_m_000003_0, Status : FAILED on node vdc-2950-5.irad.net
    Error initializing attempt_201208060924_0012_m_000003_0:
    java.io.IOException: Job initialization failed (20)org.apache.hadoop.util.Shell$ExitCodeException: . Output of LinuxTaskController initializeJob: Reading task controller config from /opt/mapr/hadoop/hadoop-0.20.2/conf/taskcontroller.cfg
    number of groups = 10
    main : command provided 0
    main : user is root
    Failed to create directory /tmp/mapr-hadoop/mapred/local/taskTracker/root - No such file or directory
    failed to initialize user directory

        at org.apache.hadoop.mapred.LinuxTaskController.initializeJob(LinuxTaskController.java:195)
        at org.apache.hadoop.mapred.TaskTracker$4.run(TaskTracker.java:1573)
        at java.security.AccessController.doPrivileged(Native Method)
        at javax.security.auth.Subject.doAs(Subject.java:396)
        at org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1110)
        at org.apache.hadoop.mapred.TaskTracker.initializeJob(TaskTracker.java:1548)
        at org.apache.hadoop.mapred.TaskTracker.localizeJob(TaskTracker.java:1446)
        at org.apache.hadoop.mapred.TaskTracker.startNewTask(TaskTracker.java:3705)
        at org.apache.hadoop.mapred.TaskTracker$TaskLauncher.run(TaskTracker.java:3628)

Outcomes