[HADOOP-2119] JobTracker becomes non-responsive if the task trackers finish task too fast

• Type: Bug
• Status: Closed
• Priority: Critical
• Resolution: Fixed
• Affects Version/s: 0.16.0
• Fix Version/s: 0.17.0
• Hadoop Flags: Reviewed

Release Note

This removes many inefficiencies in task placement and scheduling logic. The JobTracker would perform linear scans of the list of submitted tasks in cases where it did not find an obvious candidate task for a node. With better data structures for managing job state, all task placement operations now run in constant time (in most cases). Also, the task output promotions are batched.

• Assignee: Amar Kamat
• Reporter: Runping Qi
• Created: 29/Oct/07 20:43
• Updated: 02/May/13 02:29
• Resolved: 26/Mar/08 22:18

Description

I ran a job with 0 reducer on a cluster with 390 nodes.
The mappers ran very fast.
The jobtracker lacks behind on committing completed mapper tasks.
The number of running mappers displayed on web UI getting bigger and bigger.
The jos tracker eventually stopped responding to web UI.

No progress is reported afterwards.

Job tracker is running on a separate node.
The job tracker process consumed 100% cpu, with vm size 1.01g (reach the heap space limit).

Issue Links

• blocks: MAPREDUCE-315 Bias the decision of task scheduling (both for not-running and running) on node metrics (load, processing rate etc).
• blocks: MAPREDUCE-262 Optimize finding of speculative tasks
• depends upon: HADOOP-1985 Abstract node to switch mapping into a topology service class used by namenode and jobtracker
• is related to: MAPREDUCE-93 Job Tracker should prefer input-splits from overloaded racks
• relates to: HADOOP-2790 TaskInProgress.hasSpeculativeTask is very inefficient