Yes that's what scanners are good for they will return all the
columns:lables combos for a row
What does the MR job stats say for rows processed for the maps and reduces?
Billy Pearson
"llpind" <sonny_heer@hotmail.com> wrote in
message news:23967196.post@talk.nabble.com...
>
> also,
>
> I think what we want is a way to wildcard everything after colFam1:
> (e.g.
> colFam1:*). Is there a way to do this in HBase?
>
> This is assuming we dont know the column name, we want them all.
>
>
> llpind wrote:
>>
>> Thanks.
>>
>> Yea I've got that colFam for sure in the HBase table:
>>
>> {NAME => 'tableA', FAMILIES => [{NAME => 'colFam1', VERSIONS => '3',
>> COMPRESSION => 'NONE', LENGTH => '2147483647',
>> TTL => '-1', IN_MEMORY => 'false', BLOCKCACHE => 'false'}, {NAME =>
>> 'colFam2', VERSIONS => '3', COMPRESSION =>
>> 'NONE', LENGTH => '2147483647', TTL => '-1', IN_MEMORY => 'false',
>> BLOCKCACHE => 'false'}]}
>>
>>
>> I've been trying to play with rowcounter, and not having much luck
>> either.
>>
>> I run the command:
>> hadoop19/bin/hadoop org.apache.hadoop.hbase.mapred.Driver rowcounter
>> /home/hadoop/dev/rowcounter7 tableA colFam1:
>>
>>
>> The map/reduce finishes just like it does with my own program, but with
>> all part files empty in /home/hadoop/dev/rowcounter7.
>>
>> Any Ideas?
>>
>>
>
> --
> View this message in context:
> http://www.nabble.com/Help-with-Map-Reduce-program-tp23952252p23967196.html
> Sent from the HBase User mailing list archive at Nabble.com.
>
>
|