[HADOOP-4246] Reduce task copy errors may not kill it eventually

• Type: Bug
• Status: Closed
• Priority: Blocker
• Resolution: Fixed
• Affects Version/s: 0.19.0
• Fix Version/s: 0.19.0
• Assignee: Amareshwari Sriramadasu
• Reporter: Amareshwari Sriramadasu
• Created: 23/Sep/08 03:48
• Updated: 08/Jul/09 16:53
• Resolved: 03/Oct/08 15:11

Description

maxFetchRetriesPerMap in reduce task can be zero some times (when maxMapRunTime is less than 4 seconds or mapred.reduce.copy.backoff is less than 4). This will not count reduce task copy errors to kill it eventually.