Details
-
Bug
-
Status: Open (View Workflow)
-
Major
-
Resolution: Unresolved
-
10.11.4, 10.11.6
-
None
-
10.11.4-MariaDB-1:10.11.4+maria~deb11 (mariadb.org binary distribution)
Linux xxxx.local 5.10.0-23-amd64 #1 SMP Debian 5.10.179-2 (2023-07-14) x86_64 GNU/Linux
Description
The insert of mariadb-dump sql file ends up with error because ROW_START > ROW_END in some records. The dump created on 10.11.4 and import is performed on 10.11.6 server versions.
No errors in server log. We assume the application updates the records with the same value, perhaps in a transaction.
(No more info on it )
Here are the phantom records:
MariaDB [mcatu001]> select gsid, ROW_START, ROW_END from grid_show FOR SYSTEM_TIME ALL WHERE gsid in (25119, 25120) ORDER BY ROW_START;
|
+-------+----------------------------+----------------------------+
|
| gsid | ROW_START | ROW_END |
|
+-------+----------------------------+----------------------------+
|
| 25119 | 2024-01-03 12:08:45.804315 | 2024-01-03 17:08:08.453210 |
|
| 25120 | 2024-01-03 12:08:45.805367 | 2024-01-03 17:08:08.454756 |
|
| 25119 | 2024-01-03 17:08:08.453210 | 2024-01-03 17:08:08.370360 |
|
| 25120 | 2024-01-03 17:08:08.454756 | 2024-01-03 17:08:08.370360 |
|
+-------+----------------------------+----------------------------+
|
4 rows in set (0.001 sec)
|
(see the attached text file as well).
In "normal" query the rows where row_start > row_end are invisible FOR any particular system time.
The main issue with it is the requirement for dump sql file editing before import.
Thx