Thanks for the reply Allen.  I fixed it, though I am not sure of the exact cause as I found more than one thing wrong.  There were some directories owned by root instead of hadoop that I think were the main issue.

Allen Wittenauer wrote:

On 1/7/10 8:34 AM, "Scott" <skes...@weather.com> wrote:
  
 WARN hdfs.DFSClient: DataStreamer Exception:
org.apache.hadoop.ipc.RemoteException: java.io.IOException: File
/user/hadoop/ads3x11-1256301562.log.lzo could only be replicated to 0
nodes, instead of 1
    

This almost always means your HDFS is in safemode and/or has no live
datanodes.

  
I have checked quotas and found none.  I have also tried other users,
including the hadoop user, and get the same result.  Any ideas?
    

How is the namenode heap?

Are you out of physical space?

What does hadoop fsck / say?

  

Reply via email to