Can navigate via the mapred ui to find the task that failed, navigate to the
machine it was running on and then look at all logs for that failed task.
You might clue as to why StandarAnalyzer failed creation (Maybe its not on
the HADOOP_CLASSPATH?  Is the lucene jar present?).

St.Ack

On Sun, Apr 26, 2009 at 6:30 AM, Puri, Aseem <[email protected]>wrote:

> Hi
>        During while running BuildTableIndex I got this exception. Can
> anybody tell about how I can solve this exception?
>
> 09/04/26 16:24:23 WARN mapred.JobClient: Use GenericOptionsParser for
> parsing the arguments. Applications should implement Tool for the same.
> 09/04/26 16:24:24 INFO mapred.TableInputFormatBase: split: 0->master:,
> 09/04/26 16:24:25 INFO mapred.JobClient: Running job:
> job_200904261454_0018
> 09/04/26 16:24:26 INFO mapred.JobClient:  map 0% reduce 0%
> 09/04/26 16:24:41 INFO mapred.JobClient:  map 100% reduce 0%
> 09/04/26 16:25:03 INFO mapred.JobClient: Task Id :
> attempt_200904261454_0018_r_000000_0, Status : FAILED
> java.io.IOException: Error in creating an analyzer object
> org.apache.lucene.analysis.standard.StandardAnalyzer
>        at
> org.apache.hadoop.hbase.mapred.IndexOutputFormat.getRecordWriter(IndexOu
> tputFormat.java:75)
>        at org.apache.hadoop.mapred.ReduceTask.run(ReduceTask.java:404)
>        at org.apache.hadoop.mapred.Child.main(Child.java:158)
>
>
> Thanks & Regards
> Aseem Puri
>
> -----Original Message-----
> From: Stack [mailto:[email protected]]
> Sent: Wednesday, April 22, 2009 9:32 PM
> To: [email protected]
> Subject: Re: search on HBase
>
> See the build table index mr job in the hbase mapred package
>
> Stack
>
>
>
> On Apr 22, 2009, at 8:21, "Puri, Aseem" <[email protected]>
> wrote:
>
> > Hi
> >
> >            I am a new HBase user. I have a table in which I am storing
> > data. Can anybody tell me how should I search values in table with
> > help
> > of map reduce. If somebody have some code snippet then please share
> > with
> > me.
> >
> >
> >
> > Thanks & Regards
> >
> > Aseem Puri
> >
> >
> >
>

Reply via email to