facing problems while updating rows in hbase

0
votes

I've run samples : SampleUploader,PerformanceEvaluation and rowcount as given in

hadoop wiki: http://wiki.apache.org/hadoop/Hbase/MapReduce

The problem I'm facing is : table1 is my table with the column family column 

>create 'table1','column'
>put 'table1','row1','column:address','SanFrancisco'

hbase(main):020:0> scan 'table1' 
ROW COLUMN+CELL                                                 
row1                        column=column:address, timestamp=1276351974560, value=SanFrancisco

>put 'table1','row1','column:name','Hannah'

hbase(main):020:0> scan 'table1' 

ROW                          COLUMN+CELL                   
row1                       column=column:address,timestamp=1276351974560,value=SanFrancisco   
row1                       column=column:name, timestamp=1276351899573, value=Hannah

I want both the columns to appear in the same row as a different version similary, if i change the name column to sarah, it shows the updated row.... but i want both the old row and the changed row to appear as 2 different versions so that i could make analysis on the data........

what is the mistake im making????

thank u a lot sammy

1
hadoophbase

1 Answers

4
votes

To see multiple versions of the same row, you need to specify a VERSIONS option:

get 'my_table', 'my_row_key', {VERSIONS -> 4}

When the hbase shell prints out 

row1 column=column:address,timestamp=1276351974560,value=SanFrancisco
row1 column=column:name, timestamp=1276351899573, value=Hannah

That's a single row with multiple columns. The text representation just happens to use multiple lines of text, one per column.