You can not change the partition columns. I would use a dynamic partition insert to select all the data from the original table into the new table.
On 6/19/12, Tim Havens <[email protected]> wrote: > So...I have a table that has thousands of files, and Billions of rows > related it. > > Lets make this a simple table: > > CREATE TABLE test_table ( > ts BIGINT, > exec_time DOUBLE, > domain_id BIGINT, > domain_name STRING, > ) > PARTITIONED BY (logdate STRING, source STRING, datacenter STRING, > hostname STRING) > ROW FORMAT DELIMITED > FIELDS TERMINATED BY '\t' > COLLECTION ITEMS TERMINATED BY '\001' > MAP KEYS TERMINATED BY '\002' > LINES TERMINATED BY '\n' > STORED AS TextFile; > > So...what I need to do is ADD a partition to the PARTITIONED BY spec > above.... > > The partitioned by column I want to add is 'loghour STRING'. > > I can't seem to find any way to accomplish actually adding a NEW > PARTITION COLUMN in the 'PARTITIONED BY' spec, without completely > recreating and reloading the table. > > What's the correct way of adding to the partition schema and new > column like 'loghour STRING'. > > I'm not trying to add an entry into the table DATA, I'm trying to add > a completely new PARTITIONED BY Column... >
