[HADOOP-1300] deletion of excess replicas does not take into account 'rack-locality'

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 0.13.0
• Fix Version/s: 0.14.0
• Assignee: Hairong Kuang
• Reporter: Koji Noguchi
• Created: 26/Apr/07 21:04
• Updated: 08/Jul/09 16:42
• Resolved: 15/Jun/07 21:45

Description

One rack went down today, resulting in one missing block/file.
Looking at the log, this block was originally over-replicated.
3 replicas on one rack and 1 replica on another.

Namenode decided to delete the latter, leaving 3 replicas on the same rack.

It'll be nice if the deletion is also rack-aware.

Issue Links

• is blocked by: HDFS-66 Develop a unit test to test deletion of excess replicas in DFS