[HADOOP-1331] Multiple entries for 'dfs.client.buffer.dir'

• Type: Improvement
• Status: Closed
• Priority: Minor
• Resolution: Duplicate
• Assignee: Dhruba Borthakur
• Reporter: Koji Noguchi
• Created: 04/May/07 22:16
• Updated: 08/Jul/09 16:42
• Resolved: 18/Jul/07 18:05

Description

If the (DFS) client host has multiple drives, I'd like the different dfs -put calls to utilize these drives.

Also,

• It might be helpful when we have multiple reducers writing to dfs.
• If we want datanode/tasktracker to skip dead drive, we probably need this?

Issue Links

• is duplicated by: HADOOP-1372 DFS Clients should start using the org.apache.hadoop.fs.LocalDirAllocator