[MAPREDUCE-5817] Mappers get rescheduled on node transition even after all reducers are completed

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.3.0
• Fix Version/s: 2.8.0, 2.7.3, 2.6.5, 3.0.0-alpha1
• Component/s: applicationmaster
• Target Version/s: 2.8.0, 2.6.5
• Hadoop Flags: Reviewed
• Assignee: Sangjin Lee
• Reporter: Sangjin Lee
• Created: 31/Mar/14 18:48
• Updated: 11/Jun/18 06:37
• Resolved: 14/Aug/15 19:43

Description

We're seeing a behavior where a job runs long after all reducers were already finished. We found that the job was rescheduling and running a number of mappers beyond the point of reducer completion. In one situation, the job ran for some 9 more hours after all reducers completed!

This happens because whenever a node transition (to an unusable state) comes into the app master, it just reschedules all mappers that already ran on the node in all cases.

Therefore, if any node transition has a potential to extend the job period. Once this window opens, another node transition can prolong it, and this can happen indefinitely in theory.

If there is some instability in the pool (unhealthy, etc.) for a duration, then any big job is severely vulnerable to this problem.

If all reducers have been completed, JobImpl.actOnUnusableNode() should not reschedule mapper tasks. If all reducers are completed, the mapper outputs are no longer needed, and there is no need to reschedule mapper tasks as they would not be consumed anyway.

Issue Links

• is related to: YARN-1996 Provide alternative policies for UNHEALTHY nodes.
• is related to: MAPREDUCE-7109 On completion of shuffle phase in reducers, mappers should not be launched again
• is related to: MAPREDUCE-6870 Add configuration for MR job to finish when all reducers are complete (even with unfinished mappers)