You are viewing a plain text version of this content. The canonical link for it is here.
Posted to hdfs-user@hadoop.apache.org by Scott <sk...@weather.com> on 2010/01/07 17:34:19 UTC

Cant get HDFS to load more than 1Gig total data

Hi, new to hadoop/HDFS.  We have a 4 node test cluster, and I can't 
write more than approximately 1 Gig of total data.  Any additional file 
puts error out with the following:

 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

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

Thanks,
Scott


Re: Cant get HDFS to load more than 1Gig total data

Posted by Allen Wittenauer <aw...@linkedin.com>.


On 1/7/10 8:34 AM, "Scott" <sk...@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?