[MAPREDUCE-870] Clean up the job Retire code

• Type: Sub-task
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Fix Version/s: 0.21.0
• Hadoop Flags: Reviewed
• Release Note: Removed the Job Retire thread and the associated configuration parameters. Job is purged from memory as soon as the history file is copied to HDFS. Only JobStatus object is retained in the retired jobs cache.
• Assignee: Sharad Agarwal
• Reporter: Sharad Agarwal
• Created: 14/Aug/09 08:16
• Updated: 24/Aug/10 21:15
• Resolved: 21/Aug/09 11:48

Description

Currently completed job's full data structures are kept in memory based on mapred.jobtracker.completeuserjobs.maximum, mapred.jobtracker.retirejob.interval.min, mapred.jobtracker.retirejob.interval and mapred.jobtracker.retirejob.check settings. These controls are not much useful now since MAPREDUCE-817 introduced a cache for keeping just the very basic info of the completed job. These settings should be removed and the job should be purged as soon as the history files are available in HDFS.
Going forward, clients can read the history files if they need to drill down into more information (MAPREDUCE-864).

Issue Links

• relates to: MAPREDUCE-1920 Job.getCounters() returns null when using a cluster