Thanks for your answer, below the details. I am running Hive and Impala on CDH 6.1.1 I've set those parameters: SET hive.support.concurrency = true;
SET hive.exec.dynamic.partition.mode = nonstrict;
set hive.compactor.initiator.on = true;
set hive.compactor.worker.threads = 1;
set hive.txn.manager = org.apache.hadoop.hive.ql.lockmgr.DbTxnManager; Then I've created a sample table: CREATE TABLE MYDIM (key int, name string, zip string, is_current boolean)
CLUSTERED BY (key) INTO 3 BUCKETS
STORED AS ORC TBLPROPERTIES ('transactional'='true'); I've loaded some data INSERT INTO MYDIM VALUES
(1, 'bob', '95136', true),
(2, 'joe', '70068', true),
(3, 'steve', '22150', true); I run a select in HIVE and records are returned SELECT * FROM MYDIM;
mydim.key mydim.name mydim.zip mydim.is_current
1 bob 95136 true
2 joe 70068 true
3 steve 22150 true I can update & delete UPDATE MYDIM SET NAME = 'svasi' WHERE KEY=3;
SELECT * FROM MYDIM;
mydim.key mydim.name mydim.zip mydim.is_current
1 bob 95136 true
2 joe 70068 true
3 svasi 22150 true Table is available in Impala but no results are returned when I run the select * from mydim I've tried to invalidate metadata but still no records are read from Impala query Thanks for your help
... View more
