Hi there,
I have the following table created, with indexes.
{NAME => 'foo', IS_ROOT => 'false', IS_META => 'false', FAMILIES => [{NAME
=> 'entry', BLOOMFILTER => 'false',
COMPRESSION => 'NONE', VERSIONS => '3', LENGTH => '2147483647', TTL => '-1',
IN_MEMORY => 'false', BLOCKCACHE
=> 'false'}], INDEXES => [ID => hostname, ID => msg, ID => char_count, ID =>
ip]}
Now this in turn will create new 'tables' in HBase:
hbase(main):033:0> list
foo-char_count
foo-hostname
foo-msg
foo-ip
hbase(main):037:0> describe 'foo-ip'
{NAME => 'foo-ip', IS_ROOT => 'false', IS_META => 'false', FAMILIES =>
[{NAME => 'entry', BLOOMFILTER => 'false
', COMPRESSION => 'NONE', VERSIONS => '3', LENGTH => '2147483647', TTL =>
'-1', IN_MEMORY => 'false', BLOCKCAC
HE => 'false'}, {NAME => '__INDEX__', BLOOMFILTER => 'false', COMPRESSION =>
'NONE', VERSIONS => '3', LENGTH =
> '2147483647', TTL => '-1', IN_MEMORY => 'false', BLOCKCACHE => 'false'}],
INDEXES => []}
Great. So this works so far. I then insert a few thousand rows into the
'foo' table and then run a normal scan:
Scanner search = hbaseTableName.getScanner(searchString, "", Long.MAX_VALUE,
filter);
...
This works fine, but when I try an indexed scan i receive back no results.
Does another process need to encourage the indexes to function?
Thanks in advance, and apologies if this is something simple...big learning
curve
-sd
--
Sasha Dolgy
[email protected]
