Details
-
Bug
-
Status: Closed
-
Major
-
Resolution: Fixed
-
None
Description
Before HIVE-20723 it was possible to schedule the compaction for each table on specific job queues by putting compactor.mapred.job.queue.name in the table properties.
CREATE TABLE person (name STRING, age INT) STORED AS ORC TBLPROPERTIES( 'transactional'='true', 'compactor.mapred.job.queue.name'='root.user2); ALTER TABLE person COMPACT 'major' WITH OVERWRITE TBLPROPERTIES('compactor.mapred.job.queue.name'='root.user2')
This is no longer possible (after HIVE-20723) and in order to achieve the same effect someone needs to use the compactor.hive.compactor.job.queue.
CREATE TABLE person (name STRING, age INT) STORED AS ORC TBLPROPERTIES( 'transactional'='true', 'compactor.hive.compactor.job.queue'='root.user2); ALTER TABLE person COMPACT 'major' WITH OVERWRITE TBLPROPERTIES('compactor.hive.compactor.job.queue'='root.user2')
Attachments
Issue Links
- is caused by
-
HIVE-20723 Allow per table specification of compaction yarn queue
- Closed
- links to