[MAPREDUCE-5401] Succeeded Job details displayed as Error In JobHistoryServer UI when RM issues Reboot to 1st app master.

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Duplicate
• Affects Version/s: 2.0.5-alpha

Environment

Hadoop-2.0.5-alpha cluster

• Assignee: Unassigned
• Reporter: Rohith Sharma K S
• Created: 18/Jul/13 10:44
• Updated: 22/Aug/13 03:13
• Resolved: 22/Aug/13 03:13

Description

RM issues Reboot to 1st attempt app master and lauches 2nd attempt app master in available NodeManager.
1st attempt app master shutdown gracefully , not deleting staging directory which is handled in MAPREDUCE-5086. But problem is 1st attempt app master creates jhist(Error),summary and config file.

2nd attempt app master also create jhist(succeeded) , summary and config file.
While moving summary file, since file is already exist of 1st attempt summary file, 2nd attempt summary remain as summary_tmp

Issue Links

• is duplicated by: MAPREDUCE-5466 Historyserver does not refresh the result of restarted jobs after RM restart