Hi Morris,

Do you use the pig in trunk ?
If you want to use hbase, you should put hbase configuration in
hbase-site.xml, and put this file on your classpath.


Jeff Zhang


On Thu, Nov 19, 2009 at 8:20 AM, Morris Swertz <[email protected]> wrote:

>
> Hi all,
>
> I try to load data from HBase into pig with HBaseStorage. Something is
> going wrong because no data from HBase (test table) shows up in Pig; only
> errors.
>
> I configured the Hadoop and HBase in Pseudo-Distributed Operation mode.
>
> What follows are the commands that I did and the output it produced.
>
>
> //try with pig in remote mode!
>
> pig -x mapreduce
>
> B = load 'hbase://test' using
> org.apache.pig.backend.hadoop.hbase.HBaseStorage('data') as (col_a);
>
> dump B;
>
> output:
>
> 009-11-19 13:56:02,810 [main] INFO
>  
> org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.MultiQueryOptimizer
> - MR plan size before optimization: 1
>
> 2009-11-19 13:56:02,810 [main] INFO
>  
> org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.MultiQueryOptimizer
> - MR plan size after optimization: 1
>
> 2009-11-19 13:56:04,708 [main] INFO
>  
> org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.JobControlCompiler
> - Setting up single store job
>
> 2009-11-19 13:56:04,729 [main] INFO
>  org.apache.hadoop.metrics.jvm.JvmMetrics - Cannot initialize JVM Metrics
> with processName=JobTracker, sessionId= - already initialized
>
> 2009-11-19 13:56:04,739 [Thread-5] WARN  org.apache.hadoop.mapred.JobClient
> - Use GenericOptionsParser for parsing the arguments. Applications should
> implement Tool for the same.
>
> 2009-11-19 13:56:05,024 [Thread-5] INFO
>  org.apache.pig.backend.hadoop.hbase.HBaseStorage - tablename:
> file:/Users/jorislops/Desktop/pig-0.5.0/test
>
> 2009-11-19 13:56:05,231 [main] INFO
>  
> org.apache.pig.backend.hadoop.executionengine.mapReduceLayer.MapReduceLauncher
> - 0% complete
>
> 2009-11-19 13:56:06,222 [Thread-5] INFO  org.apache.hadoop.ipc.Client -
> Retrying connect to server: localhost/127.0.0.1:60000 <
> http://127.0.0.1:60000>. Already tried 0 time(s).
>
> 2009-11-19 13:56:06,222 [Thread-5] INFO  org.apache.hadoop.ipc.Client -
> Retrying connect to server: localhost/127.0.0.1:60000 <
> http://127.0.0.1:60000>. Already tried 1 time(s).
>
>
>
> //port 60000 is used by a java program
>
>
>
> pig -x local
>
> B = load 'test' using
> org.apache.pig.backend.hadoop.hbase.HBaseStorage('data') as (col_a);
>
> dump B;
>
> output:
>
> 2009-11-19 13:53:18,425 [main] INFO
>  org.apache.pig.backend.local.executionengine.LocalPigLauncher -
> Successfully stored result in: "file:/tmp/temp-1663248768/tmp-1939618752"
>
> 2009-11-19 13:53:18,436 [main] INFO
>  org.apache.pig.backend.local.executionengine.LocalPigLauncher - Records
> written : 0
>
> 2009-11-19 13:53:18,436 [main] INFO
>  org.apache.pig.backend.local.executionengine.LocalPigLauncher - Bytes
> written : 0
>
> 2009-11-19 13:53:18,436 [main] INFO
>  org.apache.pig.backend.local.executionengine.LocalPigLauncher - 100%
> complete!
>
> 2009-11-19 13:53:18,436 [main] INFO
>  org.apache.pig.backend.local.executionengine.LocalPigLauncher - Success!!
>
> //there is nothing in /tmp/temp-1663248768/tmp-1939618752 (it's empty)
>
>
>
> I tried different paths to the HBase table 'hbase://test', 'test',
> hbase://localhost:60000/test
>
>
>
> How I stated the system (Hadoop + HBase) is started and I verified that's
> working as I expected.
>
>
>
> bin/hadoop namenode -format
>
> bin/start-all.sh
>
> //both Namenode and Jobtrackter are running verified by
> http://localhost:50070 and http://localhost:500040
>
>
>
> bin/start-hbase.sh
>
> //both mater and regionserver are running check by localhost:60010
> localhost:20 localhost:30
>
> //also zookeeper Quorum is started at port localhost:2181
>
>
>
> //fill a test table in hbase
>
> hbase-0.20.1/bin/hbase shell
>
> create 'test', 'data'
>
> put 'test', 'row1', 'data', 'value1'
>
> scan 'test'
>
> //localhost:60010 show that the test table is in HBase.
>
>
>
> Hope that someone knows the solution.
>
> Thanks,
>
> Joris
>
>
>
>
>
>
>
>
>
>
>

Reply via email to