[MAPREDUCE-6485] MR job hanged forever because all resources are taken up by reducers and the last map attempt never get resource to run

• Type: Bug
• Status: Resolved
• Priority: Critical
• Resolution: Fixed
• Affects Version/s: 2.4.1, 2.6.0, 2.7.1, 3.0.0-alpha1
• Fix Version/s: 2.8.0, 3.0.0-alpha1
• Component/s: applicationmaster
• Hadoop Flags: Reviewed
• Assignee: Xianyin Xin
• Reporter: Bob.zhao
• Due: 19/Sep/15
• Created: 19/Sep/15 06:59
• Updated: 30/Apr/20 15:27
• Resolved: 02/Oct/15 15:08

Description

The scenarios is like this:
With configuring mapreduce.job.reduce.slowstart.completedmaps=0.8, reduces will take resource and start to run when all the map have not finished.
But It could happened that when all the resources are taken up by running reduces, there is still one map not finished.
Under this condition , the last map have two task attempts .
As for the first attempt was killed due to timeout(mapreduce.task.timeout), and its state transitioned from RUNNING to FAIL_CONTAINER_CLEANUP then to FAILED, but failed map attempt would not be restarted for there is still one speculate map attempt in progressing.
As for the second attempt which was started due to having enable map task speculative is pending at UNASSINGED state because of no resource available. But the second map attempt request have lower priority than reduces, so preemption would not happened.
As a result all reduces would not finished because of there is one map left. and the last map hanged there because of no resource available. so, the job would never finish.

Issue Links

• causes: MAPREDUCE-7278 Speculative execution behavior is observed even when mapreduce.map.speculative and mapreduce.reduce.speculative are false