[MAPREDUCE-3240] NM should send a SIGKILL for completed containers also

• Type: Bug
• Status: Closed
• Priority: Blocker
• Resolution: Fixed
• Affects Version/s: 0.23.0
• Fix Version/s: 0.23.0
• Component/s: mrv2, nodemanager
• Hadoop Flags: Reviewed
• Assignee: Hitesh Shah
• Reporter: Vinod Kumar Vavilapalli
• Created: 21/Oct/11 14:38
• Updated: 15/Nov/11 00:48
• Resolved: 27/Oct/11 12:04

Description

This is to address the containers which exit properly after spawning sub-processes themselves. We don't want to leave these sub-process-tree or else they can pillage the NM's resources.

Today, we already have code to send SIGKILL to the whole process-trees (because of single sessionId resulting from setsid) when the container is alive. We need to obtain the PID of the containers when they start and use that PID to send signal for completed containers' case also.

Issue Links

• duplicates: MAPREDUCE-3260 Yarn app stuck in KILL_WAIT state
• duplicates: MAPREDUCE-3084 race when KILL_CONTAINER is received for a LOCALIZED container