I've been trying to run some Hadoop jobs, but not having much luck. I'm using the hadoop streaming example from the Wiki.
The command is: hadoopstream -input /crud/20417-8.txt -mapper /bin/cat -reducer /bin/wc -output /user/hadoop/results/stream2 with hadoopstream being an alias "/home/hadoop/hadoop/bin/hadoop jar /home/hadoop/hadoop/contrib/hadoop-0.15.3-streaming.jar" It runs, but the job fails. There's a bunch of FileNotFoundExceptions from the job tracker URL: http://<server>:50030/jobtasks.jsp?jobid=job_200807301636_0001&type=map& pagenum=1 <http://%3cserver%3e:50030/jobtasks.jsp?jobid=job_200807301636_0001&type =map&pagenum=1> Example Exception: java.io.FileNotFoundException: /home/hadoop/hadoop/bin/../logs/userlogs/task_200807301636_0001_m_000000 _0/stdout (No such file or directory) at java.io.FileOutputStream.openAppend(Native Method) at java.io.FileOutputStream.(FileOutputStream.java:177) at org.apache.hadoop.mapred.TaskRunner.copyStream(TaskRunner.java:451) at org.apache.hadoop.mapred.TaskRunner.runChild(TaskRunner.java:483) at org.apache.hadoop.mapred.TaskRunner.run(TaskRunner.java:343) It looks like it's looking for a local path, not hdfs.
