hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Amandeep Khurana <ama...@gmail.com>
Subject Re: count from shell failed
Date Sun, 22 Feb 2009 22:00:48 GMT
Ok, I'll do this.

No, I havent tried the hadoop 0.19.1 yet. However, I tried to pull data from
an Oracle database directly into Hbase rather than having it go into a
flatfile before that. I was able to go upto 4M rows before one of the nodes
crashed.

I plan to get the 0.19.1 release and work on in the next couple of days.

Amandeep


Amandeep Khurana
Computer Science Graduate Student
University of California, Santa Cruz


On Sun, Feb 22, 2009 at 1:58 PM, stack <stack@duboce.net> wrote:

> Enable DEBUG in your log4j.  Doing this, shell emits the names of regions
> as
> it loads them.  Before you hit the NPE, you'll see the name of the
> problematic region.  Take the name found and then using the UI figure which
> host is supposedly carrying it.  Grep its logs for the problematic region
> to
> see if you can figure what its recent history has been.
>
> Did you retry yesterday's loadings with haoop 0.19.1 release candidate?
>
> St.Ack
>
> On Sun, Feb 22, 2009 at 12:51 PM, Amandeep Khurana <amansk@gmail.com>
> wrote:
>
> > I had given a count command for a table from the shell. It counted for a
> > while after which it failed with the following error:
> >
> >
> > Current count: 2078000, row: 605428239
> >
> > NativeException: java.lang.NullPointerException: null
> >    from java.lang.String:-1:in `<init>'
> >    from org/apache/hadoop/hbase/util/Bytes.java:92:in `toString'
> >    from
> org/apache/hadoop/hbase/client/RetriesExhaustedException.java:50:in
> > `getMessage'
> >
> >
> > What could be going wrong?
> >
> > Amandeep
> >
> > Amandeep Khurana
> > Computer Science Graduate Student
> > University of California, Santa Cruz
> >
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message