[TRAFODION-2766] HIVE: Inserted row into a hive table has gone missing

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 2.3
• Fix Version/s: 2.3
• Component/s: sql-exe
• Assignee: Sandhya Sundaresan
• Reporter: Sandhya Sundaresan
• Created: 05/Oct/17 16:05
• Updated: 05/Sep/18 21:39
• Resolved: 07/Dec/17 21:44

Description

An inserted row into a hive table would go missing in various circumstances. As shown in the following occurrence, the first 2 inserts were executed fine. Select afterwards also showed that the rows were there:

1st insert: (C_CHAR10, P_CHAR10) = ('STR_1_01', 'STR_1_01')
2nd insert: (C_CHAR10, P_CHAR10) = (NULL, 'STR_1_02')

Then the 3rd insert inserted a new row, which was inserted fine too:

3rd insert: (C_CHAR10, P_CHAR10) = ('STR_1_03', NULL)

But the select statement afterwards only got 2 rows back. The row from the 2nd insert (C_CHAR10, P_CHAR10) = (NULL, 'STR_1_02') has gone missing after the 3rd insert took place. This doesn't appear to be a select problem, since a select from the hive shell in the end doesn't show the row from the 2nd insert either.

Simpler testcase :
cqd hive_max_string_length_in_bytes '10';
process hive statement 'drop table t031hive1';
process hive statement 'create table t031hive1 (a int, b timestamp, c string)';

insert into hive.hive.t031hive1 values ('1', '2017-01-01 10:10:10', 2);
--select * from hive.hive.t031hive1;

insert into hive.hive.t031hive1 values ('2', '2017-01-02 11:11:11', 3),
('3', '2017-01-03 11:11:11', 4),
(4, timestamp '2017-01-04 11:11:11', '5');

--select * from hive.hive.t031hive1;

insert into hive.hive.t031hive1 values (2, '2017-01-02 11:11:11', 'a'),
(111111111111, '2017-01-03 11:11:11', 'b');

select * from hive.hive.t031hive1;
>>process hive statement 'create table t031hive1 (a int, b timestamp, c string)';

— SQL operation complete.
>>insert into hive.hive.t031hive1 values ('1', '2017-01-01 10:10:10', 2);

— 1 row(s) inserted.
>>insert into hive.hive.t031hive1 values ('2', '2017-01-02 11:11:11', 3),
+> ('3', '2017-01-03 11:11:11', 4),
+> (4, timestamp '2017-01-04 11:11:11', '5');

— 3 row(s) inserted.
>>
>>-- this insert should return overflow error
>>insert into hive.hive.t031hive1 values (2, '2017-01-02 11:11:11', 'a'),
+> (111111111111, '2017-01-03 11:11:11', 'b');

• • • ERROR[8411] A numeric overflow occurred during an arithmetic computation or data conversion. Conversion of Source Type:LARGEINT(REC_BIN64_SIGNED) Source Value:111111111111 to Target Type:INTEGER SIGNED(REC_BIN32_SIGNED).

— 0 row(s) inserted.
>>
>>select * from hive.hive.t031hive1;

A B C
----------- -------------------------- ----------

1 2017-01-01 10:10:10.000000 2

— 1 row(s) selected.
>>

Issue Links

• links to: GitHub Pull Request #1259