CUBRID 11.3 Patch 3 has been released.
In this release, issues found after 11.3.2 released have been fixed.
1. Improvement and Fix
- When using a function index like <= or >= in the WHERE clause, the optimizer has been modified to perform a range scan.
- Modified so that "AUTO_INCREMENT" and "DEFAULT" cannot coexist when changing column properties using ALTER TABLE MODIFY.
- Modified to prevent the use of the "FOR UPDATE" syntax for system tables and system view tables (_db_class, db_root, dual, etc.)
- Improved sort-limit optimization to apply even when there are bind variables and calculation formulas in the limit clause.
- Fixed an error where Oracle-style LEFT OUTER JOIN is not rewritten as INNER JOIN when SP call and host variables (bind variables) are used in the WHERE clause
- Fixed an error that occurs when using INNER JOIN and Oracle-style OUTER JOIN together in the WHERE clause, depending on the position of the OUTER JOIN join condition.
- Fixed an error in query execution result of 'INSERT INTO table ... SELECT ... FROM view table ... ON DUPLICATE KEY UPDATE'
- Fixed an issue where the column order number in an analytic function was incorrect when the view included one.
- Improved unloaddb performance.
- Improved query performance when TRACE is enabled.
- Removed unnecessary length checks on string types.
- Improved interrupt handling while running Java SP (e.g. "Ctrl + C" when running csql, termination via killtran utility, etc.).
- Modified loaddb so that if an error occurs, it outputs an error message and stops execution rather than continuing.
- Modified unloaddb so that serial and trigger names are not omitted when executed as the DBA user.
- Fixed an error in unloaddb that occurred when a NULL value was found in the db_auth table.
- Fixed the issue where the archive log volume created when performing Backupdb was not deleted.
- Activated 'monitor_waiting_thread' feature - changed error output level (NOTIFICATION => WARNING)
- Fixed an issue where archive logs replicated from the slave to the replica node were not deleted when ha_replica_delay was set to 60 seconds or more in an HA environment consisting of a master node, slave node, and replica node.
- Modified to prevent situations where data inconsistency with the master node may occur when performing Restoreslave.
- Fixed a memory leak issue that occurred when DML operations were performed with result-cache enabled.
- Fixed a hidden column handling error in scalar subqueries with ORDER BY clause.
- Fixed to restart javasp process when abnormal situation occurs due to system error
- Fixed an error causing incorrect results due to cursor index error in javasp
Please refer to the 11.3.3 change logs(https://github.com/CUBRID/cubrid/releases/tag/v11.3.3) for details.
2. Precautions
- When using heterogeneous(Oracle/MySQL) DBLINK, you must use ODBC Drvier for Unicode of Oracle/MySQL.
3. Environmental Requirements
- Linux: 64bit, kernel 2.4 or higher, Windows: Windows7 or higher (Windows 7 requires the latest security updates)
- JRE 1.8 or higher is required when using JSP or CUBRID Admin/CUBRID Manager
4. Download
- Product download : https://www.cubrid.org/downloads
- Product source code : https://github.com/CUBRID/cubrid/tree/v11.3.3
In this release, issues found after 11.3.2 released have been fixed.
1. Improvement and Fix
- When using a function index like <= or >= in the WHERE clause, the optimizer has been modified to perform a range scan.
- Modified so that "AUTO_INCREMENT" and "DEFAULT" cannot coexist when changing column properties using ALTER TABLE MODIFY.
- Modified to prevent the use of the "FOR UPDATE" syntax for system tables and system view tables (_db_class, db_root, dual, etc.)
- Improved sort-limit optimization to apply even when there are bind variables and calculation formulas in the limit clause.
- Fixed an error where Oracle-style LEFT OUTER JOIN is not rewritten as INNER JOIN when SP call and host variables (bind variables) are used in the WHERE clause
- Fixed an error that occurs when using INNER JOIN and Oracle-style OUTER JOIN together in the WHERE clause, depending on the position of the OUTER JOIN join condition.
- Fixed an error in query execution result of 'INSERT INTO table ... SELECT ... FROM view table ... ON DUPLICATE KEY UPDATE'
- Fixed an issue where the column order number in an analytic function was incorrect when the view included one.
- Improved unloaddb performance.
- Improved query performance when TRACE is enabled.
- Removed unnecessary length checks on string types.
- Improved interrupt handling while running Java SP (e.g. "Ctrl + C" when running csql, termination via killtran utility, etc.).
- Modified loaddb so that if an error occurs, it outputs an error message and stops execution rather than continuing.
- Modified unloaddb so that serial and trigger names are not omitted when executed as the DBA user.
- Fixed an error in unloaddb that occurred when a NULL value was found in the db_auth table.
- Fixed the issue where the archive log volume created when performing Backupdb was not deleted.
- Activated 'monitor_waiting_thread' feature - changed error output level (NOTIFICATION => WARNING)
- Fixed an issue where archive logs replicated from the slave to the replica node were not deleted when ha_replica_delay was set to 60 seconds or more in an HA environment consisting of a master node, slave node, and replica node.
- Modified to prevent situations where data inconsistency with the master node may occur when performing Restoreslave.
- Fixed a memory leak issue that occurred when DML operations were performed with result-cache enabled.
- Fixed a hidden column handling error in scalar subqueries with ORDER BY clause.
- Fixed to restart javasp process when abnormal situation occurs due to system error
- Fixed an error causing incorrect results due to cursor index error in javasp
Please refer to the 11.3.3 change logs(https://github.com/CUBRID/cubrid/releases/tag/v11.3.3) for details.
2. Precautions
- When using heterogeneous(Oracle/MySQL) DBLINK, you must use ODBC Drvier for Unicode of Oracle/MySQL.
3. Environmental Requirements
- Linux: 64bit, kernel 2.4 or higher, Windows: Windows7 or higher (Windows 7 requires the latest security updates)
- JRE 1.8 or higher is required when using JSP or CUBRID Admin/CUBRID Manager
4. Download
- Product download : https://www.cubrid.org/downloads
- Product source code : https://github.com/CUBRID/cubrid/tree/v11.3.3