[MAPREDUCE-5862] Line records longer than 2x split size aren't handled correctly

• Type: Bug
• Status: Closed
• Priority: Critical
• Resolution: Fixed
• Affects Version/s: 2.3.0
• Fix Version/s: 2.5.0
• Hadoop Flags: Reviewed
• Assignee: bc Wong
• Reporter: bc Wong
• Created: 27/Apr/14 01:02
• Updated: 03/Sep/14 20:33
• Resolved: 28/May/14 19:54

Description

Suppose this split (100-200) is in the middle of a record (90-240):

   0              100            200             300
   |---- split ----|---- curr ----|---- split ----|
                 <------- record ------->
                 90                     240

Currently, the first split would read the entire record, up to offset 240, which is good. But the 2nd split has a bug in producing a phantom record of (200, 240).