Hi,

I tried bulk uploading and it ran well with TSV files, we first ran importtsv 
and then completebulkload, after doing these two steps I can scan my HBase 
table and see the data. I can also see the data when I traverse HDFS of my 
Hadoop cluster using web browser.

But when I try to upload my CSVs in a folder, I get bad lines for all the lines 
of my CSV files. I use following command to upload my CSVs on my local file 
system to HDFS, 

HADOOP_CLASSPATH=`hbase classpath` $HADOOP_HOME/bin/hadoop jar 
/hbase_home/hbase-0.92.0/hbase-0.92.0.jar importtsv  
-Dimporttsv.bulk.output=/my_output_dir 
-Dimporttsv.columns=HBASE_ROW_KEY,SerialNumber,Column1,Column2 my_table 
file:/my_csv/data.txt '-Dimporttsv.separator=,'

my csv file is of following format

1,data11,data12
2,data21,data22
3,data31,data32
.....
.....

And my HBase table has 3 columns


Please let me know what is the exact problem and how this can be resolved?

Kind regards,
Keshav



-----Original Message-----
From: Savant, Keshav 
Sent: Friday, March 02, 2012 7:02 PM
To: [email protected]
Cc: '[email protected]'
Subject: RE: Inserting Data from CSV into HBase

Hi Harsh,

Thanks for your response, I don't get any error using the code mentioned in 
that URL. I will get back to you after analyzing the tools suggested by you.
Thanks again.


Kind regards,
Keshav C Savant 

-----Original Message-----
From: Harsh J [mailto:[email protected]]
Sent: Friday, March 02, 2012 6:51 PM
To: [email protected]
Subject: Re: Inserting Data from CSV into HBase

Hi,

You may use the importtsv tool and the bulk-load utilities in HBase to achieve 
this fast-and-easy.

This is detailed at http://hbase.apache.org/bulk-loads.html (See section about 
importtsv along the bottom) and also under section "Using the importtsv tool" 
on Page 460 of Lars George's "HBase: The Definitive Guide" (O'Reilly).

Also when you say something didn't work, please also supply any errors you 
encountered and the configuration you used. Its hard to help without those.

On Fri, Mar 2, 2012 at 6:24 PM, Savant, Keshav <[email protected]> 
wrote:
> Hi All,
>
> I am looking for a way so that I can map my existing CSV file to HBase table, 
> basically for each column family I want only one value (just like RDBMS).
>
> Just to illustrate more suppose I define a HBase table as
>
> create 'inventory', 'item', 'supplier', 'quantity'
> (here table name is inventory and it has three columns named as item, 
> supplier and quantity)
>
> Now I want to load my N number of CSVs in following format into this 
> HBase table
>
> Burger,abc confectionary,100
> Pizza,xyz bakers,50
> ...
> ...
> ...
>
> Here I want to put the data of CSV into my inventory table on HBase, the 
> number of lines in a CSV and even number of CSVs are dynamic, and this will 
> be a continuous process.
>
> What I want to know that, do we have any way by which we can achieve above 
> goal, I tried SampleUploader as specified on 
> http://svn.apache.org/repos/asf/hbase/trunk/src/examples/mapreduce/org/apache/hadoop/hbase/mapreduce/SampleUploader.java,
>  but it did not worked and data does not gets populated in HBase table though 
> the program ran successfully.
>
> Please suggest on this, any help is appreciated.
>
> Kind regards,
> Keshav C Savant
>
> _____________
> The information contained in this message is proprietary and/or confidential. 
> If you are not the intended recipient, please: (i) delete the message and all 
> copies; (ii) do not disclose, distribute or use the message in any manner; 
> and (iii) notify the sender immediately. In addition, please be aware that 
> any message addressed to our domain is subject to archiving and review by 
> persons other than the intended recipient. Thank you.



--
Harsh J

_____________
The information contained in this message is proprietary and/or confidential. 
If you are not the intended recipient, please: (i) delete the message and all 
copies; (ii) do not disclose, distribute or use the message in any manner; and 
(iii) notify the sender immediately. In addition, please be aware that any 
message addressed to our domain is subject to archiving and review by persons 
other than the intended recipient. Thank you.

Reply via email to