On a terasort job a task attempt failed during the commit phase. Another attempt was rescheduled, but when it tried to commit it failed.
attempt_1329019187148_0083_r_000586_0 already given a go for committing the task output, so killing attempt_1329019187148_0083_r_000586_1
The job hung as new attempts kept getting scheduled only to fail during commit.
- is related to
MAPREDUCE-5255 Reduce task preemption results in task never completing , incomplete fix to MAPREDUCE-3858 ?