You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@trafodion.apache.org by "Alice Chen (JIRA)" <ji...@apache.org> on 2015/07/22 20:15:46 UTC

[jira] [Created] (TRAFODION-360) LP Bug: 1328211 - Insert … select’s effect is delayed occasionally

Alice Chen created TRAFODION-360:
------------------------------------

             Summary: LP Bug: 1328211 - Insert … select’s effect is delayed occasionally
                 Key: TRAFODION-360
                 URL: https://issues.apache.org/jira/browse/TRAFODION-360
             Project: Apache Trafodion
          Issue Type: Bug
          Components: dtm
            Reporter: Weishiun Tsai
            Assignee: Apache Trafodion
            Priority: Blocker


The symptom of this problem is similar to another bug report https://bugs.launchpad.net/trafodion/+bug/1305233 “update query's effect is delayed occasionally”.  As shown here, when running all statements in a script, the select count(*) returns only 104 rows right after the insert is done.  But after waiting for a bit, the same select count(*) returns the correct 300 rows.

For a long time, this failure has been attributed to the same problem as BUG#1305233.  But BUG#1305233 is supposed to be fixed in the GIT 0606_0930 build, while this problem is still seen on the same build, especially on a cluster installation (centos-mapr1 in this case).  This report is created to document this problem.

-----------------------------------------------------
Here is the entire script to reproduce this problem (saw this only on a cluster, not on a workstation):

create table t02tabA (a1 int not null, a2 int) store by (a1);

insert into t02tabA values (1,2),(3,4),(5,6);

create table t02tab (a1 int not null, a2 int, v1 int, v2 int) store by (a1);

insert into t02tab (
select * from t02tabA
transpose 0,1,2,3,4,5,6,7,8,9 as v1
transpose 0,1,2,3,4,5,6,7,8,9 as v2
);

select count(*) from t02tab;

-----------------------------------------------------

Here is the execution output, and the second select count(*) that returns the correct result.

>>obey mytest.sql;
>>create table t02tabA (a1 int not null, a2 int) store by (a1);

--- SQL operation complete.
>>
>>insert into t02tabA values (1,2),(3,4),(5,6);

--- 3 row(s) inserted.
>>
>>create table t02tab (a1 int not null, a2 int, v1 int, v2 int) store by (a1);

--- SQL operation complete.
>>
>>insert into t02tab (
+>select * from t02tabA
+>transpose 0,1,2,3,4,5,6,7,8,9 as v1
+>transpose 0,1,2,3,4,5,6,7,8,9 as v2
+>);

--- 300 row(s) inserted.
>>
>>select count(*) from t02tab;

(EXPR)
--------------------

                 104

--- 1 row(s) selected.
>>
>>select count(*) from t02tab;

(EXPR)
--------------------

                 300

--- 1 row(s) selected.
>>
Assigned to LaunchPad User Adriana Fuentes



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)