1
0
mirror of https://github.com/MariaDB/server.git synced 2026-01-06 05:22:24 +03:00
Commit Graph

172325 Commits

Author SHA1 Message Date
Vladislav Vaintroub
cd35dd6a05 Windows : Do not use CRT routine to dump memory leaks.
Its output is useless,and, in case of large output, it also may prevent
with search_pattern_in_file.inc from working.
2017-08-24 15:49:50 +00:00
Vladislav Vaintroub
dd229430a9 Windows compile : make compilation fail on "uninitialized variable used" warning C4700
This is a genuine error, and will crash debug buildd in runtime checks if not fixed.
it is better to fail during compile.
2017-08-24 08:05:11 +00:00
Vladislav Vaintroub
7aa846e9e3 CONNECT engine: install ha_connect.lib 2017-08-23 23:30:51 +00:00
Vladislav Vaintroub
7b36395ee9 MDEV-13630 : dont install connect-specific JAR files if connect is not built. 2017-08-23 23:29:59 +00:00
Vladislav Vaintroub
db51ad1e01 Remove workaround for ancient and already fixed CMake bug in MSI creation.
The workaround, an extra cmake calls, somehow makes the
connect/cmake_install.cmake to lose installation of connect-engine's
specific jar files.
2017-08-23 18:11:24 +00:00
Marko Mäkelä
b8b3ba632b MDEV-13606 XA PREPARE transactions should survive innodb_force_recovery=1 or 2
When MySQL 5.0.3 introduced InnoDB support for two-phase commit,
it also introduced the questionable logic to roll back XA PREPARE
transactions on startup when innodb_force_recovery is 1 or 2.

Remove this logic in order to avoid unwanted side effects when
innodb_force_recovery is being set for other reasons. That is,
XA PREPARE transactions will always remain in that state until
InnoDB receives an explicit XA ROLLBACK or XA COMMIT request
from the upper layer.

At the time the logic was introduced in MySQL 5.0.3, there already
was a startup parameter that is the preferred way of achieving
the behaviour: --tc-heuristic-recover=ROLLBACK.
2017-08-23 13:03:13 +03:00
Jan Lindström
ce6c0e584e MDEV-8960: Can't refer the same column twice in one ALTER TABLE
Problem was that if column was created in alter table when
it was refered again it was not tried to find from list
of current columns.

mysql_prepare_alter_table:
  There is two cases
    (1) If alter table adds a new column and then later alter
        changes the field definition, there was no check from
	list of new columns, instead an incorrect error was given.
    (2) If alter table adds a new column and then later alter
        changes the default, there was no check from list of
	new columns, instead an incorrect error was given.
2017-08-18 10:11:54 +03:00
Sergei Golubchik
7581fb23e2 compilation fix for SLES 11 SP4
also fix innodb
2017-08-14 18:37:53 +02:00
Sergei Golubchik
fc556a8d94 compilation fix for SLES 11 SP4 2017-08-13 23:37:21 +02:00
Marko Mäkelä
cb9648a6b5 Revert an InnoDB Memcached plugin fix that was merged from MySQL 5.6.37
Revert the following change, because Memcached is not present
in MariaDB Server. We had better avoid adding dead code.

commit d9bc5e03d788b958ce8c76e157239953db60adb2
Author: Aakanksha Verma <aakanksha.verma@oracle.com>
Date:   Thu May 18 14:31:01 2017 +0530

    Bug #24605783 MYSQL GOT SIGNAL 6 ASSERTION FAILURE
2017-08-09 14:29:22 +03:00
Daniel Bartholomew
ef2e51c396 bump the VERSION 2017-08-07 10:09:23 -04:00
Vicențiu Ciorbaru
a346a5613e Merge branch 'bb-10.0-vicentiu' into 10.0
Includes Percona XtraDB and TokuDB 5.6.36-82.1
mariadb-10.0.32
2017-08-04 09:32:40 +02:00
Vicențiu Ciorbaru
d2a19082c0 Disable rpl_tokudb_row_img_* tests
We need to pull rpl config files from MySQL before these tests can be
run.
2017-08-04 04:27:33 +03:00
Elena Stepanova
d85d6c9c41 Remove extra trailing spaces from the result file 2017-08-04 02:11:10 +03:00
Elena Stepanova
e1aded1420 Additions to 10.0.32 unstable test list after TokuDB merge 2017-08-04 02:10:19 +03:00
Vicențiu Ciorbaru
8b0edbfb5e Merge branch 'merge-tokudb-5.6' into 10.0 2017-08-04 00:56:16 +03:00
Vicențiu Ciorbaru
d5164569e4 Move tokudb_rpl to rpl-tokudb suite 2017-08-03 22:15:19 +03:00
Monty
e35670f498 MDEV-12972 Random and Frequent Segfault
Problem was a memory overflow in MRR

Reviewed by Sergei Petruna. Testcase by Elena
2017-08-03 19:28:05 +03:00
Vicențiu Ciorbaru
13e4bf674e Fix tokudb compilation post merge 2017-08-03 12:22:35 +03:00
Vicențiu Ciorbaru
40e213f210 Merge branch 'merge-tokudb-5.6' into 10.0 2017-08-03 12:11:31 +03:00
Vicențiu Ciorbaru
b1a2031ff9 5.6.36-82.1 2017-08-03 11:48:44 +03:00
Vicențiu Ciorbaru
b278c02e18 Merge branch 'merge-xtradb-5.6' into 10.0 2017-08-02 12:15:58 +03:00
Vicențiu Ciorbaru
04ae1207ed 5.6.36-82.1 2017-08-02 12:11:06 +03:00
Vicențiu Ciorbaru
60c9485a8e Merge connect/10.0 into 10.0
Squash all connect/10.0's commits into one and keep only
changes from storage/connect directory.
2017-08-01 11:39:10 +03:00
Elena Stepanova
31f4df4505 MDEV-12915 ST_Centroid does not return the same result than MySQL
Fix storage_engine test results
2017-08-01 09:52:59 +02:00
Vicențiu Ciorbaru
e075c5b491 Update myisam_debug test result post print_lock_error change 2017-08-01 09:52:59 +02:00
Vicențiu Ciorbaru
3772516391 Update innodb test result post print_lock_error change 2017-08-01 09:52:58 +02:00
Elena Stepanova
fd47b9639d List of unstable tests for 10.0.32 release 2017-08-01 09:52:58 +02:00
Sergei Golubchik
e6e23c0d49 MDEV-9739 Assertion `m_status == DA_ERROR || m_status == DA_OK' failed in Diagnostics_area::message() ; connect.xml* tests fail in buildbot 2017-08-01 09:52:57 +02:00
Sergei Golubchik
60343871a7 MDEV-8453 Alter table not returning engine errors
remove ~15 years old print_lock_error() function, use
handler::print_error() instead

Backport from 10.1
2017-08-01 09:52:57 +02:00
Alexey Botchkov
3d2067460e MDEV-12915 ST_Centroid does not return the same result than MySQL.
Test fixed for IA64.
2017-08-01 09:52:56 +02:00
Alexey Botchkov
7f5a8f176a MDEV-12915 ST_Centroid does not return the same result than MySQL.
Calculation of the polygon's centroid fixed.
2017-08-01 09:52:56 +02:00
Vicențiu Ciorbaru
f8736063de Merge branch 'merge-pcre' into 10.0 2017-07-30 11:58:56 +03:00
Vicențiu Ciorbaru
dba454ef54 8.41 2017-07-30 11:53:36 +03:00
Vicențiu Ciorbaru
051f8bc8e3 Merge branch 'merge-perfschema-5.6' into 10.0 2017-07-30 11:36:13 +03:00
Vicențiu Ciorbaru
900e3218cf Bump innodb version post merge 2017-07-30 11:35:48 +03:00
Vicențiu Ciorbaru
e1b5e98648 5.6.37 (no-changes) 2017-07-30 11:33:55 +03:00
Vicențiu Ciorbaru
226dd699df Merge remote-tracking branch 'merge/merge-innodb-5.6' into 10.0
Following merge from 5.6.36, this merge also rejects changes that
collided with the rejection of 6ca4f693c1ce472e2b1bf7392607c2d1124b4293.

We initially rejected 6ca4f693c1ce472e2b1bf7392607c2d1124b4293 because
it was introducing a new storage engine API method.
2017-07-29 20:51:29 +03:00
Vicențiu Ciorbaru
2160a85184 Update subselect test results to pass on 10.0
One query was removed due to inconsistent behaviour with subquery cache
enabled vs disabled. MDEV-13399 has been created to track this change
and fix.
2017-07-29 20:01:20 +03:00
Vicențiu Ciorbaru
d36bd69790 5.6.37 2017-07-29 19:59:51 +03:00
Vicențiu Ciorbaru
af40426fcd Fix purge_relay_logs post merge
slave_skip_counter must not be reset to 0 during purge_relay_logs. See
MDEV-4937 as to when this change happened.
2017-07-27 16:22:37 +03:00
Sergei Golubchik
7134afa22e MYSQL_BIN_LOG::open/close must be under LOCK_log 2017-07-27 12:43:45 +02:00
Sergei Golubchik
c784277590 move the error message where it belongs 2017-07-27 12:43:03 +02:00
Sergei Golubchik
496cea45e2 update error messages for 10.0 2017-07-27 12:42:40 +02:00
Sergei Golubchik
2804a3fac4 memory leak: add a missing end_relay_log_info() 2017-07-27 12:42:21 +02:00
Vicențiu Ciorbaru
786ad0a158 Merge remote-tracking branch 'origin/5.5' into 10.0 2017-07-25 00:41:54 +03:00
Sergei Golubchik
2ce3f95151 Connect engine is Gamma 2017-07-22 13:10:10 +02:00
Jan Lindström
92cbe388b6 MDEV-13325: InnoDB assert dict_sys->size > 0 during ALTER TABLE
Problem was that dict_sys->size tries to maintain used memory
occupied by the data dictionary table and index objects.
However at least on table objects table->heap size can increase
between when table object is inserted to dict_sys and when
it is removed from dict_sys causing inconsistency on amount
of memory added to and removed from dict_sys->size variable.

Removed unnecessary dict_sys:size variable as it is really
used only for status output.

Introduced dict_sys_get_size function to calculate memory
occupied by the data dictionary table and index objects
that is then used on show engine innodb output.

dict_table_add_to_cache(),
dict_table_rename_in_cache(),
dict_table_remove_from_cache_low(),
dict_index_remove_from_cache_low(),
	Remove size calculation.

srv_printf_innodb_monitor(): Use dict_sys_get_size function to
get dictionary memory allocated.

xtradb_internal_hash_tables_fill_table(): Use dict_sys_get_size
function to get dictionary memory allocated.
2017-07-21 16:05:20 +03:00
Sergei Golubchik
0ec147b617 fix main.func_regexp_pcre on jessie-ppc64le
apparently it uses a smaller per-thread stack
2017-07-20 22:26:08 +02:00
Sergei Golubchik
3ef5596505 MDEV-13175 Adding a new enum value at the end of a list triggers a table rebuild
Backport of 7e29f2d64f from 10.1.

Create_field does not set BINARY_FLAG, so the check didn't work at all.
Also, character sets were already compared, so this check would've been
redundant (if it would've worked).
2017-07-20 20:13:28 +02:00