AnsweredAssumed Answered

Jobtracker does not start any tasks

Question asked by feeblefakie on Feb 22, 2012
Latest reply on Feb 22, 2012 by nabeel
I installed the M3 version and tried a example mapreduce program as in the document.

$ hadoop jar /opt/mapr/hadoop/hadoop-0.20.2/hadoop-0.20.2-dev-examples.jar wordcount /myvolume/in /myvolume/out

But, JobTracker seems not starting any tasks.
The jobtracker log only says like below.

Job job_201202231457_0003 initialized successfully with 1 map tasks and 1 reduce tasks.

So, the tasks are initialized but not started.
The programs stays "0% reduce 0%" like the following.
$ hadoop jar /opt/mapr/hadoop/hadoop-0.20.2/hadoop-0.20.2-dev-examples.jar wordcount /myvolume/in /myvolume/out
12/02/23 15:55:06 INFO fs.JobTrackerWatcher: Current running JobTracker is: myhost/
12/02/23 15:55:06 INFO input.FileInputFormat: Total input paths to process : 1
12/02/23 15:55:06 WARN snappy.LoadSnappy: Snappy native library is not available
12/02/23 15:55:06 INFO util.NativeCodeLoader: Loaded the native-hadoop library
12/02/23 15:55:06 WARN snappy.LoadSnappy: Snappy native library not loaded
12/02/23 15:55:06 INFO mapred.JobClient: Running job: job_201202231457_0004
12/02/23 15:55:07 INFO mapred.JobClient:  map 0% reduce 0%

The web says "Queued Jobs 1" in dashboard right pane, and<br/>
"There are currently no known active Task Trackers." in JobTracker pane.<br/>
(JobTracker thinks no active tasktrackers ?<br/>
TaskTracker seems running but inactive ? <br/> If so, Is there any way to activate TaskTracker ?)

Does anyone tell me what's going on ?<br/>
(I have been using hadoop a lot but, first time using mapr.)