[MAPREDUCE-569] Streaming should produce a short per-job log in a persistent location.

• Type: Improvement
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Component/s: contrib/streaming
• Labels: newbie
• Assignee: Unassigned
• Reporter: arkady borkovsky
• Created: 24/Aug/06 21:50
• Updated: 16/Jan/12 09:55

Description

Streaming should produce a short per-job log in a persistent location.

The log should include (at least, but not limited to):
– the command line that hadoop-streaming is executing
– the list of matching input fragments
– the actual command lines used by the framework for -mapper and for -reducer
– log about all copied files
– the name of the output HDFS directory
– start and end time of the job
– location of stderr logs and other output.