2008-01-08 21:45:15 +01:00
reset master;
2003-09-24 21:25:58 +02:00
set timestamp=1000000000;
2007-01-22 14:52:15 +01:00
drop table if exists t1,t2,t3,t4,t5,t03,t04;
2003-09-24 21:25:58 +02:00
create table t1 (word varchar(20));
create table t2 (id int auto_increment not null primary key);
insert into t1 values ("abirvalg");
insert into t2 values ();
2011-01-10 14:53:09 +01:00
load data infile '../../std_data/words3.dat' into table t1;
load data infile '../../std_data/words3.dat' into table t1;
load data infile '../../std_data/words3.dat' into table t1;
load data infile '../../std_data/words3.dat' into table t1;
load data infile '../../std_data/words3.dat' into table t1;
2003-09-24 21:25:58 +02:00
insert into t1 values ("Alas");
flush logs;
--- Local --
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2004-04-28 12:08:54 +02:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
2005-02-23 19:59:25 +01:00
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
2006-11-28 13:26:15 +01:00
DELIMITER /*!*/;
ROLLBACK/*!*/;
2012-08-24 10:06:16 +02:00
use `test`/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-04 16:47:14 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=0, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1, @@session.sql_if_exists=0, @@session.explicit_defaults_for_timestamp=0/*!*/;
2019-08-30 15:06:54 +02:00
SET @@session.sql_mode=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
2006-11-28 13:26:15 +01:00
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
2010-08-20 04:59:58 +02:00
DROP TABLE IF EXISTS `t1`,`t2`,`t3`,`t4`,`t5`,`t03`,`t04` /* generated by server */
2007-12-17 14:13:25 +01:00
/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
2007-12-17 14:13:25 +01:00
create table t1 (word varchar(20))
/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
2007-12-17 14:13:25 +01:00
create table t2 (id int auto_increment not null primary key)
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2007-12-17 14:13:25 +01:00
insert into t1 values ("abirvalg")
/*!*/;
2009-11-03 20:02:56 +01:00
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
2006-11-28 13:26:15 +01:00
SET INSERT_ID=1/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2007-12-17 14:13:25 +01:00
insert into t2 values ()
/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`word`)
2007-12-17 14:13:25 +01:00
/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`word`)
2007-12-17 14:13:25 +01:00
/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`word`)
2007-12-17 14:13:25 +01:00
/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`word`)
2007-12-17 14:13:25 +01:00
/*!*/;
2009-11-03 20:02:56 +01:00
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2006-11-28 13:26:15 +01:00
DELIMITER ;
2005-09-21 14:27:41 +02:00
# End of log file
2005-09-30 10:58:24 +02:00
ROLLBACK /* added by mysqlbinlog */;
2005-02-23 19:59:25 +01:00
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2003-09-24 21:25:58 +02:00
--- Broken LOAD DATA --
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2004-04-28 12:08:54 +02:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
2005-02-23 19:59:25 +01:00
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
2006-11-28 13:26:15 +01:00
DELIMITER /*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2013-03-26 10:35:34 +01:00
/*!*/;
use `test`/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-04 16:47:14 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=0, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1, @@session.sql_if_exists=0, @@session.explicit_defaults_for_timestamp=0/*!*/;
2019-08-30 15:06:54 +02:00
SET @@session.sql_mode=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
2006-11-28 13:26:15 +01:00
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`word`)
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2007-12-17 14:13:25 +01:00
insert into t1 values ("Alas")
/*!*/;
2009-11-03 20:02:56 +01:00
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2006-11-28 13:26:15 +01:00
DELIMITER ;
2005-09-21 14:27:41 +02:00
# End of log file
2005-09-30 10:58:24 +02:00
ROLLBACK /* added by mysqlbinlog */;
2005-02-23 19:59:25 +01:00
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2003-09-24 21:25:58 +02:00
--- --database --
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2004-04-28 12:08:54 +02:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
2005-02-23 19:59:25 +01:00
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
2006-11-28 13:26:15 +01:00
DELIMITER /*!*/;
ROLLBACK/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2013-03-26 10:35:34 +01:00
/*!*/;
2009-11-10 19:45:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-04 16:47:14 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=0, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1, @@session.sql_if_exists=0, @@session.explicit_defaults_for_timestamp=0/*!*/;
2019-08-30 15:06:54 +02:00
SET @@session.sql_mode=#/*!*/;
2009-11-10 19:45:15 +01:00
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=#/*!*/;
2009-11-10 19:45:15 +01:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-10 19:45:15 +01:00
/*!*/;
2006-11-28 13:26:15 +01:00
SET INSERT_ID=1/*!*/;
2009-11-10 19:45:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-10 19:45:15 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-10 19:45:15 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-10 19:45:15 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-10 19:45:15 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2006-11-28 13:26:15 +01:00
DELIMITER ;
2005-09-21 14:27:41 +02:00
# End of log file
2005-09-30 10:58:24 +02:00
ROLLBACK /* added by mysqlbinlog */;
2005-02-23 19:59:25 +01:00
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2003-09-24 21:25:58 +02:00
2010-02-17 10:18:17 +01:00
--- --start-position --
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2004-04-28 12:08:54 +02:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
2005-02-23 19:59:25 +01:00
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
2006-11-28 13:26:15 +01:00
DELIMITER /*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2013-03-26 10:35:34 +01:00
/*!*/;
use `test`/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-04 16:47:14 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=0, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1, @@session.sql_if_exists=0, @@session.explicit_defaults_for_timestamp=0/*!*/;
2019-08-30 15:06:54 +02:00
SET @@session.sql_mode=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
2006-11-28 13:26:15 +01:00
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
2007-12-17 14:13:25 +01:00
insert into t1 values ("Alas")
/*!*/;
2009-11-03 20:02:56 +01:00
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2006-11-28 13:26:15 +01:00
DELIMITER ;
2005-09-21 14:27:41 +02:00
# End of log file
2005-09-30 10:58:24 +02:00
ROLLBACK /* added by mysqlbinlog */;
2005-02-23 19:59:25 +01:00
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2003-09-24 21:25:58 +02:00
--- Remote --
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2004-04-28 12:08:54 +02:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
2005-02-23 19:59:25 +01:00
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
2006-11-28 13:26:15 +01:00
DELIMITER /*!*/;
ROLLBACK/*!*/;
2012-08-24 10:06:16 +02:00
use `test`/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-04 16:47:14 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=0, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1, @@session.sql_if_exists=0, @@session.explicit_defaults_for_timestamp=0/*!*/;
2019-08-30 15:06:54 +02:00
SET @@session.sql_mode=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
2006-11-28 13:26:15 +01:00
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
2010-08-20 04:59:58 +02:00
DROP TABLE IF EXISTS `t1`,`t2`,`t3`,`t4`,`t5`,`t03`,`t04` /* generated by server */
2007-12-17 14:13:25 +01:00
/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
2007-12-17 14:13:25 +01:00
create table t1 (word varchar(20))
/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
2007-12-17 14:13:25 +01:00
create table t2 (id int auto_increment not null primary key)
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2007-12-17 14:13:25 +01:00
insert into t1 values ("abirvalg")
/*!*/;
2009-11-03 20:02:56 +01:00
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
2006-11-28 13:26:15 +01:00
SET INSERT_ID=1/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2007-12-17 14:13:25 +01:00
insert into t2 values ()
/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`word`)
2007-12-17 14:13:25 +01:00
/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`word`)
2007-12-17 14:13:25 +01:00
/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`word`)
2007-12-17 14:13:25 +01:00
/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`word`)
2007-12-17 14:13:25 +01:00
/*!*/;
2009-11-03 20:02:56 +01:00
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2006-11-28 13:26:15 +01:00
DELIMITER ;
2005-09-21 14:27:41 +02:00
# End of log file
2005-09-30 10:58:24 +02:00
ROLLBACK /* added by mysqlbinlog */;
2005-02-23 19:59:25 +01:00
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2003-09-24 21:25:58 +02:00
--- Broken LOAD DATA --
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2004-04-28 12:08:54 +02:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
2005-02-23 19:59:25 +01:00
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
2006-11-28 13:26:15 +01:00
DELIMITER /*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2013-03-26 10:35:34 +01:00
/*!*/;
use `test`/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-04 16:47:14 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=0, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1, @@session.sql_if_exists=0, @@session.explicit_defaults_for_timestamp=0/*!*/;
2019-08-30 15:06:54 +02:00
SET @@session.sql_mode=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
2006-11-28 13:26:15 +01:00
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`word`)
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2007-12-17 14:13:25 +01:00
insert into t1 values ("Alas")
/*!*/;
2009-11-03 20:02:56 +01:00
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2006-11-28 13:26:15 +01:00
DELIMITER ;
2005-09-21 14:27:41 +02:00
# End of log file
2005-09-30 10:58:24 +02:00
ROLLBACK /* added by mysqlbinlog */;
2005-02-23 19:59:25 +01:00
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2003-09-24 21:25:58 +02:00
--- --database --
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2004-04-28 12:08:54 +02:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
2005-02-23 19:59:25 +01:00
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
2006-11-28 13:26:15 +01:00
DELIMITER /*!*/;
ROLLBACK/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2013-03-26 10:35:34 +01:00
/*!*/;
2009-11-10 19:45:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-04 16:47:14 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=0, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1, @@session.sql_if_exists=0, @@session.explicit_defaults_for_timestamp=0/*!*/;
2019-08-30 15:06:54 +02:00
SET @@session.sql_mode=#/*!*/;
2009-11-10 19:45:15 +01:00
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=#/*!*/;
2009-11-10 19:45:15 +01:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-10 19:45:15 +01:00
/*!*/;
2006-11-28 13:26:15 +01:00
SET INSERT_ID=1/*!*/;
2009-11-10 19:45:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-10 19:45:15 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-10 19:45:15 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-10 19:45:15 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-10 19:45:15 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2006-11-28 13:26:15 +01:00
DELIMITER ;
2005-09-21 14:27:41 +02:00
# End of log file
2005-09-30 10:58:24 +02:00
ROLLBACK /* added by mysqlbinlog */;
2005-02-23 19:59:25 +01:00
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2003-09-24 21:25:58 +02:00
2010-02-17 10:18:17 +01:00
--- --start-position --
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2004-04-28 12:08:54 +02:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
2005-02-23 19:59:25 +01:00
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
2006-11-28 13:26:15 +01:00
DELIMITER /*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2013-03-26 10:35:34 +01:00
/*!*/;
use `test`/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-04 16:47:14 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=0, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1, @@session.sql_if_exists=0, @@session.explicit_defaults_for_timestamp=0/*!*/;
2019-08-30 15:06:54 +02:00
SET @@session.sql_mode=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
2006-11-28 13:26:15 +01:00
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
2007-12-17 14:13:25 +01:00
insert into t1 values ("Alas")
/*!*/;
2009-11-03 20:02:56 +01:00
SET TIMESTAMP=1000000000/*!*/;
COMMIT
/*!*/;
2006-11-28 13:26:15 +01:00
DELIMITER ;
2005-09-21 14:27:41 +02:00
# End of log file
2005-09-30 10:58:24 +02:00
ROLLBACK /* added by mysqlbinlog */;
2005-02-23 19:59:25 +01:00
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2005-02-04 14:29:54 +01:00
--- reading stdin --
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2005-02-04 14:29:54 +01:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
2005-02-23 19:59:25 +01:00
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
2006-11-28 13:26:15 +01:00
DELIMITER /*!*/;
ROLLBACK/*!*/;
SET TIMESTAMP=1108844556/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.pseudo_thread_id=999999999/*!*/;
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
2007-12-17 14:13:25 +01:00
BEGIN
/*!*/;
2012-08-24 14:02:32 +02:00
use `test`/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1108844555/*!*/;
2007-12-17 14:13:25 +01:00
insert t1 values (1)
/*!*/;
2006-11-28 13:26:15 +01:00
DELIMITER ;
2005-09-21 14:27:41 +02:00
# End of log file
2005-09-30 10:58:24 +02:00
ROLLBACK /* added by mysqlbinlog */;
2005-02-23 19:59:25 +01:00
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2005-02-04 14:29:54 +01:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
2005-02-23 19:59:25 +01:00
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
2006-11-28 13:26:15 +01:00
DELIMITER /*!*/;
SET TIMESTAMP=1108844556/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.pseudo_thread_id=999999999/*!*/;
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
2007-12-17 14:13:25 +01:00
BEGIN
/*!*/;
2012-08-24 14:02:32 +02:00
use `test`/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1108844555/*!*/;
2007-12-17 14:13:25 +01:00
insert t1 values (1)
/*!*/;
2006-11-28 13:26:15 +01:00
DELIMITER ;
2005-09-21 14:27:41 +02:00
# End of log file
2005-09-30 10:58:24 +02:00
ROLLBACK /* added by mysqlbinlog */;
2005-02-23 19:59:25 +01:00
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2007-01-22 14:52:15 +01:00
drop table t1,t2;
2006-02-09 15:23:09 +01:00
flush logs;
2006-12-14 11:05:25 +01:00
flush logs;
2006-05-31 12:51:21 +02:00
select * from t5 /* must be (1),(1) */;
a
1
1
2007-01-22 14:52:15 +01:00
drop table t5;
2006-12-07 06:31:53 +01:00
flush logs;
create table t5 (c1 int, c2 varchar(128) character set latin1 not null);
insert into t5 values (1, date_format('2001-01-01','%W'));
set lc_time_names=de_DE;
insert into t5 values (2, date_format('2001-01-01','%W'));
set lc_time_names=en_US;
insert into t5 values (3, date_format('2001-01-01','%W'));
select * from t5 order by c1;
c1 c2
1 Monday
2 Montag
3 Monday
flush logs;
drop table t5;
select * from t5 order by c1;
c1 c2
1 Monday
2 Montag
3 Monday
2007-01-22 14:52:15 +01:00
drop table t5;
2006-11-28 13:26:15 +01:00
drop procedure if exists p1;
flush logs;
create procedure p1()
begin
select 1;
end;
//
flush logs;
call p1();
1
1
drop procedure p1;
call p1();
ERROR 42000: PROCEDURE test.p1 does not exist
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2006-11-28 13:26:15 +01:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
DELIMITER /*!*/;
2012-08-24 10:06:16 +02:00
use `test`/*!*/;
2006-11-28 13:26:15 +01:00
SET TIMESTAMP=1000000000/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-04 16:47:14 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=0, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1, @@session.sql_if_exists=0, @@session.explicit_defaults_for_timestamp=0/*!*/;
2019-08-30 15:06:54 +02:00
SET @@session.sql_mode=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
2006-11-28 13:26:15 +01:00
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
Bug#36570: Parse error of CREATE PROCEDURE stmt with comments on \
slave
The stored-routine code took the contents of the (lowest) parser
and copied it directly to the binlog, which causes problems if there
is a special case of interpretation at the parser level -- which
there is, in the "/*!VER */" comments. The trailing "*/" caused
errors on the slave, naturally.
Now, since by that point we have /properly/ created parse-tree (as
the rest of the server should do!) for the stored-routine CREATE, we
can construct a perfect statement from that information, instead of
writing uncertain information from an unknown parser state.
Fortunately, there's already a function nearby that does exactly
that.
---
Update for Bug#36570. Qualify routine names with db name when
writing to the binlog ONLY if the source text is qualified.
mysql-test/r/binlog_innodb.result:
Offsets changed due to quoting.
---
New offset to account for db-qualified names.
mysql-test/r/ctype_cp932_binlog.result:
Offsets changed due to quoting.
---
Qualify routine names with DB. Offsets change also.
mysql-test/r/mysqlbinlog.result:
Case changed in result due to interpretation of data instead of
literal recitation.
---
Qualify procedure name with db.
mysql-test/r/rpl_sp.result:
Offsets changed due to quoting. Added tests.
---
Qualify routine names with DB if qualified in query. Offsets change also.
mysql-test/t/rpl_sp.test:
Add version-limiting quotes to exercise bug#36570. Test that
backtick-quoted identifiers and labels work also.
---
Use different db to show qualification works. Qualify routine names
with DB if qualified in query.
sql/sp.cc:
In create_string, we may not have a sp_name parameter yet, so
instead pass the char* and length of the only member we'd get out
of it.
Having done that, we can use the same function to write the
CREATE (FUNC|TRIG|PROC) statement to the binlog as we always used
to display the statement to the user.
---
Make the db name part of the CREATE string if it is specified.
Specify it in part of writing to the binlog when creating a new
routine.
sql/sp_head.cc:
Set the sp_head m_explicit_name member as the sp_name member is set.
We can not peek at this later, as the sp_name is gone by then.
sql/sp_head.h:
Add a member to track whether the name is qualified with the
database.
2008-05-16 01:13:24 +02:00
CREATE DEFINER=`root`@`localhost` PROCEDURE `p1`()
2006-11-28 13:26:15 +01:00
begin
select 1;
2007-12-17 14:13:25 +01:00
end
/*!*/;
2006-11-28 13:26:15 +01:00
DELIMITER ;
# End of log file
ROLLBACK /* added by mysqlbinlog */;
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2006-11-28 13:26:15 +01:00
call p1();
1
1
drop procedure p1;
2006-12-14 11:05:25 +01:00
flush logs;
2007-02-28 14:06:57 +01:00
create table t1 (a varchar(64) character set utf8);
2007-12-12 18:19:24 +01:00
load data infile '../../std_data/loaddata6.dat' into table t1;
2007-02-28 14:06:57 +01:00
set character_set_database=koi8r;
2007-12-12 18:19:24 +01:00
load data infile '../../std_data/loaddata6.dat' into table t1;
2007-02-28 14:06:57 +01:00
set character_set_database=latin1;
2007-12-12 18:19:24 +01:00
load data infile '../../std_data/loaddata6.dat' into table t1;
load data infile '../../std_data/loaddata6.dat' into table t1;
2007-02-28 14:06:57 +01:00
set character_set_database=koi8r;
2007-12-12 18:19:24 +01:00
load data infile '../../std_data/loaddata6.dat' into table t1;
2007-02-28 14:06:57 +01:00
set character_set_database=latin1;
2007-12-12 18:19:24 +01:00
load data infile '../../std_data/loaddata6.dat' into table t1;
load data infile '../../std_data/loaddata6.dat' into table t1 character set koi8r;
2007-02-28 14:06:57 +01:00
select hex(a) from t1;
hex(a)
C3BF
D0AA
C3BF
C3BF
D0AA
C3BF
D0AA
drop table t1;
flush logs;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2007-02-28 14:06:57 +01:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
DELIMITER /*!*/;
2012-08-24 10:06:16 +02:00
use `test`/*!*/;
2007-02-28 14:06:57 +01:00
SET TIMESTAMP=1000000000/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-04 16:47:14 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=0, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1, @@session.sql_if_exists=0, @@session.explicit_defaults_for_timestamp=0/*!*/;
2019-08-30 15:06:54 +02:00
SET @@session.sql_mode=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
2007-02-28 14:06:57 +01:00
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=#/*!*/;
BUG#31168: @@hostname does not replicate
Problem: in mixed and statement mode, a query that refers to a
system variable will use the slave's value when replayed on
slave. So if the value of a system variable is inserted into a
table, the slave will differ from the master.
Fix: mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode and produce a warning
in statement mode. There are some exceptions: some variables are actually
replicated. Those should *not* be marked as unsafe.
BUG#34732: mysqlbinlog does not print default values for auto_increment variables
Problem: mysqlbinlog does not print default values for some variables,
including auto_increment_increment and others. So if a client executing
the output of mysqlbinlog has different default values, replication will
be wrong.
Fix: Always print default values for all variables that are replicated.
I need to fix the two bugs at the same time, because the test cases would
fail if I only fixed one of them.
include/m_ctype.h:
Added definition of ILLEGAL_CHARSET_INFO_NUMBER. We just need a symbol
for a number that will never be used by any charset. ~0U should be safe
since charset numbers are sequential, starting from 0.
mysql-test/include/commit.inc:
Upated test to avoid making statements unsafe.
mysql-test/r/commit_1innodb.result:
Updated test needs updated result file.
mysql-test/r/mysqlbinlog.result:
Updated result file.
mysql-test/r/mysqlbinlog2.result:
Updated result file.
mysql-test/r/user_var-binlog.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_base64_flag.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_stm_ctype_ucs.result:
Updated result file.
mysql-test/suite/binlog/r/binlog_unsafe.result:
Modified test file needs modified result file.
mysql-test/suite/binlog/t/binlog_base64_flag.test:
Need to filter out pseudo_thread_id from result since it is
nondeterministic.
mysql-test/suite/binlog/t/binlog_unsafe.test:
Add tests that using variables is unsafe. The 'CREATE VIEW' tests didn't
make sense, so I removed them. SHOW WARNINGS is not necessary either,
because we get warnings for each statement in the result file.
mysql-test/suite/rpl/r/rpl_row_mysqlbinlog.result:
Updated result file.
mysql-test/suite/rpl/r/rpl_skip_error.result:
Updated result file.
mysql-test/suite/rpl/t/rpl_skip_error.test:
The test used @@server_id, which is not safe to replicate, so it would
have given a warning. The way it used @@server_id was hackish (issue a
query on master that removes rows only on master), so I replaced it by a
more robust way to do the same thing (connect to slave and insert the
rows only there).
Also clarified what the test case does.
mysql-test/t/mysqlbinlog2.test:
Use --short-form instead of manually filtering out nondeterministic stuff
from mysqlbinlog (because we added the nondeterministic @@pseudo_thread_id
to the output).
sql/item_func.cc:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/item_func.h:
Added method of Item_func_get_system_var that indicates whether the given
system variable will be written to the binlog or not.
sql/log_event.cc:
- auto_increment_offset was not written to the binlog if
auto_increment_increment=1
- mysqlbinlog did not output default values for some variables
(BUG#34732). In st_print_event_info, we remember for each variable whether
it has been printed or not. This is achieved in different ways for
different variables:
- For auto_increment_*, lc_time_names, charset_database_number,
we set the default values in st_print_event_info to something
illegal, so that it will look like they have changed the first time
they are seen.
- For charset, sql_mode, pseudo_thread_id, we add a flag to
st_print_event_info which indicates whether the variable has been
printed.
- Since pseudo_thread_id is now printed more often, and its value is
not guaranteed to be constant across different runs of the same
test script, I replaced it by a constant if --short-form is used.
- Moved st_print_event_info constructor from log_event.h to log_event.cc,
since it now depends on ILLEGAL_CHARSET_NUMBER, which is defined in
m_ctype.h, which is better to include from a .cc file than from a header
file.
sql/log_event.h:
Added fields to st_print_event_info that indicate whether some of the
variables have been written or not. Since the initialization of
charset_database_number now depends on ILLEGAL_CHARSET_INFO_NUMBER, which
is defined in a header file, which we'd better not include from this
header file -- I moved the constructor from here to log_event.cc.
sql/set_var.cc:
System variables now have a flag binlog_status, which indicates if they
are written to the binlog. If nothing is specified, all variables are
marked as not written to the binlog (NOT_IN_BINLOG) when created. In this
file, the variables that are written to the binlog are marked with
SESSION_VARIABLE_IN_BINLOG.
sql/set_var.h:
Added flag binlog_status to class sys_var. Added a getter and a
constructor parameter that sets it.
Since I had to change sys_var_thd_enum constructor anyways, I simplified
it to use default values of arguments instead of three copies of the
constructor.
sql/sql_yacc.yy:
Mark statements that refer to a system variable as "unsafe",
meaning they will be replicated by row in mixed mode. Added comment to
explain strange piece of code just above.
mysql-test/include/diff_tables.inc:
New auxiliary test file that tests whether two tables (possibly one on
master and one on slave) differ.
mysql-test/suite/rpl/r/rpl_variables.result:
New test case needs new result file.
mysql-test/suite/rpl/r/rpl_variables_stm.result:
New test file needs new result file.
mysql-test/suite/rpl/t/rpl_variables.test:
Test that INSERT of @@variables is replicated correctly (by switching to
row-based mode).
mysql-test/suite/rpl/t/rpl_variables_stm.test:
Test that replication of @@variables which are replicated explicitly works
as expected in statement mode (without giving warnings).
2008-03-07 13:59:36 +01:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
2007-12-17 14:13:25 +01:00
create table t1 (a varchar(64) character set utf8)
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`a`)
2007-12-17 14:13:25 +01:00
/*!*/;
2007-02-28 14:06:57 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2013-03-26 10:35:34 +01:00
SET @@session.collation_database=7/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`a`)
2007-12-17 14:13:25 +01:00
/*!*/;
2007-02-28 14:06:57 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2013-03-26 10:35:34 +01:00
SET @@session.collation_database=DEFAULT/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`a`)
2007-12-17 14:13:25 +01:00
/*!*/;
2007-02-28 14:06:57 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2010-01-20 21:22:43 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`a`)
2007-12-17 14:13:25 +01:00
/*!*/;
2007-02-28 14:06:57 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2013-03-26 10:35:34 +01:00
SET @@session.collation_database=7/*!*/;
2011-01-10 14:53:09 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`a`)
2007-12-17 14:13:25 +01:00
/*!*/;
2007-02-28 14:06:57 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2013-03-26 10:35:34 +01:00
SET @@session.collation_database=DEFAULT/*!*/;
2011-01-10 14:53:09 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`a`)
2007-12-17 14:13:25 +01:00
/*!*/;
2007-02-28 14:06:57 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
2020-07-20 14:27:39 +02:00
START TRANSACTION
2009-11-03 20:02:56 +01:00
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2011-01-10 14:53:09 +01:00
LOAD DATA LOCAL INFILE 'MYSQLTEST_VARDIR/tmp/SQL_LOAD_MB-#-#' INTO TABLE `t1` CHARACTER SET koi8r FIELDS TERMINATED BY '\t' ENCLOSED BY '' ESCAPED BY '\\' LINES TERMINATED BY '\n' (`a`)
2007-12-17 14:13:25 +01:00
/*!*/;
2007-02-28 14:06:57 +01:00
SET TIMESTAMP=1000000000/*!*/;
2009-11-03 20:02:56 +01:00
COMMIT
/*!*/;
SET TIMESTAMP=1000000000/*!*/;
2010-08-20 04:59:58 +02:00
DROP TABLE `t1` /* generated by server */
2007-12-17 14:13:25 +01:00
/*!*/;
2007-02-28 14:06:57 +01:00
DELIMITER ;
# End of log file
ROLLBACK /* added by mysqlbinlog */;
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2007-06-20 23:11:28 +02:00
CREATE TABLE t1 (c1 CHAR(10));
2009-03-03 21:34:18 +01:00
FLUSH LOGS;
2007-06-20 23:11:28 +02:00
INSERT INTO t1 VALUES ('0123456789');
2009-03-03 21:34:18 +01:00
FLUSH LOGS;
2007-06-20 23:11:28 +02:00
DROP TABLE t1;
2013-03-26 10:35:34 +01:00
We expect this value to be 2 (one for the INSERT, one for COMMIT).
2009-02-19 21:37:40 +01:00
The bug being tested was that 'Query' lines were not preceded by '#'
If the line is in the table, it had to have been preceded by a '#'
2013-09-12 20:36:58 +02:00
SELECT COUNT(*) AS `BUG#28293_expect_2` FROM patch WHERE a LIKE '#%Query%';
2013-03-26 10:35:34 +01:00
BUG#28293_expect_2
2
2009-02-19 21:37:40 +01:00
DROP TABLE patch;
2009-03-03 21:34:18 +01:00
FLUSH LOGS;
CREATE TABLE t1(a INT);
INSERT INTO t1 VALUES(connection_id());
FLUSH LOGS;
DROP TABLE t1;
2016-03-25 17:51:22 +01:00
connect con1, localhost, root, , test;
connection con1;
disconnect con1;
connection default;
Fix for bug #29928: INSERT ... VALUES(connection_id(), ...) incorrect
restores from mysqlbinlog out
Problem: using "mysqlbinlog | mysql" for recoveries the connection_id()
result may differ from what was used when issuing the statement.
Fix: if there is a connection_id() in a statement, write to binlog
SET pseudo_thread_id= XXX; before it and use the value later on.
mysql-test/r/mysqlbinlog.result:
Fix for bug #29928: INSERT ... VALUES(connection_id(), ...) incorrect
restores from mysqlbinlog out
- test result.
mysql-test/t/mysqlbinlog.test:
Fix for bug #29928: INSERT ... VALUES(connection_id(), ...) incorrect
restores from mysqlbinlog out
- test case.
sql/item_create.cc:
Fix for bug #29928: INSERT ... VALUES(connection_id(), ...) incorrect
restores from mysqlbinlog out
- set thread_specific_used flag for the connection_id() function.
sql/item_func.cc:
Fix for bug #29928: INSERT ... VALUES(connection_id(), ...) incorrect
restores from mysqlbinlog out
- always return thd->variables.pseudo_thread_id as a connection_id()
result, as it contains a proper value for both master and slave.
sql/log_event.cc:
Fix for bug #29928: INSERT ... VALUES(connection_id(), ...) incorrect
restores from mysqlbinlog out
- set LOG_EVENT_THREAD_SPECIFIC_F event flag if thread_specific_used
is set.
sql/sql_class.cc:
Fix for bug #29928: INSERT ... VALUES(connection_id(), ...) incorrect
restores from mysqlbinlog out
- thd->thread_specific_used introduced, which is set if thread specific
value(s) used in a statement.
sql/sql_class.h:
Fix for bug #29928: INSERT ... VALUES(connection_id(), ...) incorrect
restores from mysqlbinlog out
- thd->thread_specific_used introduced, which is set if thread specific
value(s) used in a statement.
2007-08-01 12:27:03 +02:00
1
2009-03-03 21:34:18 +01:00
DROP TABLE t1;
2007-11-09 13:43:09 +01:00
shell> mysqlbinlog std_data/corrupt-relay-bin.000624 > var/tmp/bug31793.sql
2009-03-06 15:56:17 +01:00
FLUSH LOGS;
Bug#31611 Security risk with BINLOG statement
2007-11-03 01:33:48 +01:00
SET BINLOG_FORMAT=ROW;
CREATE DATABASE mysqltest1;
CREATE USER untrusted@localhost;
GRANT SELECT ON mysqltest1.* TO untrusted@localhost;
SHOW GRANTS FOR untrusted@localhost;
Grants for untrusted@localhost
2019-11-06 12:35:19 +01:00
GRANT USAGE ON *.* TO `untrusted`@`localhost`
GRANT SELECT ON `mysqltest1`.* TO `untrusted`@`localhost`
2007-11-03 01:33:48 +01:00
USE mysqltest1;
CREATE TABLE t1 (a INT, b CHAR(64));
flush logs;
INSERT INTO t1 VALUES (1,USER());
flush logs;
2009-11-03 20:02:56 +01:00
mysqlbinlog var/log/master-bin.000018 > var/tmp/bug31611.sql
2016-03-25 17:51:22 +01:00
connect unsecure,localhost,untrusted,,mysqltest1;
2007-11-03 01:33:48 +01:00
mysql mysqltest1 -uuntrusted < var/tmp/bug31611.sql
2016-03-25 17:51:22 +01:00
connection unsecure;
2007-11-03 01:33:48 +01:00
INSERT INTO t1 VALUES (1,USER());
2022-07-16 14:39:17 +02:00
ERROR 42000: INSERT command denied to user 'untrusted'@'localhost' for table `mysqltest1`.`t1`
2007-11-03 01:33:48 +01:00
SELECT * FROM t1;
a b
1 root@localhost
2016-03-25 17:51:22 +01:00
connection default;
2007-11-03 01:33:48 +01:00
DROP DATABASE mysqltest1;
2007-11-12 11:29:55 +01:00
DROP USER untrusted@localhost;
2009-03-06 15:56:17 +01:00
Bug#32580 mysqlbinlog cannot read binlog event with user variables
2016-03-25 17:51:22 +01:00
connection default;
2007-11-23 14:41:41 +01:00
USE test;
SET BINLOG_FORMAT = STATEMENT;
FLUSH LOGS;
CREATE TABLE t1 (a_real FLOAT, an_int INT, a_decimal DECIMAL(5,2), a_string CHAR(32));
SET @a_real = rand(20) * 1000;
SET @an_int = 1000;
SET @a_decimal = CAST(rand(19) * 999 AS DECIMAL(5,2));
SET @a_string = 'Just a test';
INSERT INTO t1 VALUES (@a_real, @an_int, @a_decimal, @a_string);
FLUSH LOGS;
SELECT * FROM t1;
a_real 158.883
an_int 1000
a_decimal 907.79
a_string Just a test
DROP TABLE t1;
2009-11-03 20:02:56 +01:00
>> mysqlbinlog var/log/master-bin.000020 > var/tmp/bug32580.sql
2007-11-23 14:41:41 +01:00
>> mysql test < var/tmp/bug32580.sql
SELECT * FROM t1;
a_real 158.883
an_int 1000
a_decimal 907.79
a_string Just a test
DROP TABLE t1;
2009-03-05 14:35:03 +01:00
SET @@global.server_id= 4294967295;
RESET MASTER;
FLUSH LOGS;
SELECT
(@a:=LOAD_FILE("MYSQLTEST_VARDIR/tmp/mysqlbinlog_bug37313.binlog"))
IS NOT NULL;
(@a:=LOAD_FILE("MYSQLTEST_VARDIR/tmp/mysqlbinlog_bug37313.binlog"))
IS NOT NULL
2009-02-20 12:55:43 +01:00
1
*** Unsigned server_id 4294967295 is found: 1 ***
2009-03-05 14:35:03 +01:00
SET @@global.server_id= 1;
2009-05-07 16:31:02 +02:00
RESET MASTER;
FLUSH LOGS;
2009-09-30 04:01:52 +02:00
RESET MASTER;
FLUSH LOGS;
#
2010-03-28 13:57:33 +02:00
# Test if the 'BEGIN', 'ROLLBACK' and 'COMMIT' are output if the database specified exists
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2009-09-30 04:01:52 +02:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
DELIMITER /*!*/;
ROLLBACK/*!*/;
2012-08-24 10:06:16 +02:00
use `test`/*!*/;
2009-09-30 04:01:52 +02:00
SET TIMESTAMP=1253783037/*!*/;
2009-09-30 04:31:25 +02:00
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-23 16:38:03 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=1, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1/*!*/;
2009-09-30 04:01:52 +02:00
SET @@session.sql_mode=0/*!*/;
2009-09-30 04:31:25 +02:00
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
2009-09-30 04:01:52 +02:00
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=8/*!*/;
2009-09-30 04:31:25 +02:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
2009-09-30 04:01:52 +02:00
create table t1(a int) engine= innodb
/*!*/;
SET TIMESTAMP=1253783037/*!*/;
BEGIN
/*!*/;
SET TIMESTAMP=1253783037/*!*/;
insert into t1 (a) values (1)
/*!*/;
COMMIT/*!*/;
SET TIMESTAMP=1253783037/*!*/;
create table t3(a int) engine= innodb
/*!*/;
SET TIMESTAMP=1253783037/*!*/;
BEGIN
/*!*/;
SET TIMESTAMP=1253783037/*!*/;
insert into t3 (a) values (2)
/*!*/;
SET TIMESTAMP=1253783037/*!*/;
ROLLBACK
/*!*/;
SET TIMESTAMP=1253783037/*!*/;
create table t5(a int) engine= NDB
/*!*/;
SET TIMESTAMP=1253783037/*!*/;
BEGIN
/*!*/;
SET TIMESTAMP=1253783037/*!*/;
insert into t5 (a) values (3)
/*!*/;
SET TIMESTAMP=1253783037/*!*/;
COMMIT
/*!*/;
DELIMITER ;
# End of log file
ROLLBACK /* added by mysqlbinlog */;
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2009-09-30 04:01:52 +02:00
#
2010-03-28 13:57:33 +02:00
# Test if the 'BEGIN', 'ROLLBACK' and 'COMMIT' are output if the database specified does not exist
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2009-09-30 04:01:52 +02:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
DELIMITER /*!*/;
ROLLBACK/*!*/;
SET TIMESTAMP=1253783037/*!*/;
2009-09-30 04:31:25 +02:00
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-23 16:38:03 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=1, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1/*!*/;
2009-09-30 04:01:52 +02:00
SET @@session.sql_mode=0/*!*/;
2009-09-30 04:31:25 +02:00
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
2009-09-30 04:01:52 +02:00
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=8/*!*/;
2009-09-30 04:31:25 +02:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
2009-09-30 04:01:52 +02:00
BEGIN
/*!*/;
COMMIT/*!*/;
SET TIMESTAMP=1253783037/*!*/;
BEGIN
/*!*/;
SET TIMESTAMP=1253783037/*!*/;
ROLLBACK
/*!*/;
SET TIMESTAMP=1253783037/*!*/;
BEGIN
/*!*/;
SET TIMESTAMP=1253783037/*!*/;
COMMIT
/*!*/;
DELIMITER ;
# End of log file
ROLLBACK /* added by mysqlbinlog */;
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2010-03-28 13:57:33 +02:00
#
# Test if the 'SAVEPOINT', 'ROLLBACK TO' are output if the database specified exists
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2010-03-28 13:57:33 +02:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
DELIMITER /*!*/;
SET TIMESTAMP=1266652094/*!*/;
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-23 16:38:03 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=1, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1/*!*/;
2010-03-28 13:57:33 +02:00
SET @@session.sql_mode=0/*!*/;
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=8/*!*/;
2010-03-28 13:57:33 +02:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
BEGIN
/*!*/;
SET TIMESTAMP=1266652094/*!*/;
SavePoint mixed_cases
/*!*/;
2012-08-24 10:06:16 +02:00
use `db1`/*!*/;
2010-03-28 13:57:33 +02:00
SET TIMESTAMP=1266652094/*!*/;
INSERT INTO db1.t2 VALUES("in savepoint mixed_cases")
/*!*/;
SET TIMESTAMP=1266652094/*!*/;
INSERT INTO db1.t1 VALUES(40)
/*!*/;
SET TIMESTAMP=1266652094/*!*/;
ROLLBACK TO mixed_cases
/*!*/;
SET TIMESTAMP=1266652094/*!*/;
INSERT INTO db1.t2 VALUES("after rollback to")
/*!*/;
SET TIMESTAMP=1266652094/*!*/;
INSERT INTO db1.t1 VALUES(50)
/*!*/;
COMMIT/*!*/;
DELIMITER ;
# End of log file
ROLLBACK /* added by mysqlbinlog */;
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2010-03-28 13:57:33 +02:00
#
# Test if the 'SAVEPOINT', 'ROLLBACK TO' are output if the database specified does not exist
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
2010-03-28 13:57:33 +02:00
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
DELIMITER /*!*/;
SET TIMESTAMP=1266652094/*!*/;
SET @@session.pseudo_thread_id=999999999/*!*/;
2022-07-23 16:38:03 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=1, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1/*!*/;
2010-03-28 13:57:33 +02:00
SET @@session.sql_mode=0/*!*/;
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=8/*!*/;
2010-03-28 13:57:33 +02:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
BEGIN
/*!*/;
SET TIMESTAMP=1266652094/*!*/;
SavePoint mixed_cases
/*!*/;
SET TIMESTAMP=1266652094/*!*/;
ROLLBACK TO mixed_cases
/*!*/;
COMMIT/*!*/;
DELIMITER ;
# End of log file
ROLLBACK /* added by mysqlbinlog */;
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
2012-11-20 13:37:23 +01:00
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2007-02-19 11:57:06 +01:00
End of 5.0 tests
2007-03-07 10:15:45 +01:00
End of 5.1 tests
2021-05-12 19:46:58 +02:00
#
# Expect error for invalid --base64-output argument value.
# MYSQL_BINLOG std_data/master-bin.000001 --base64-output=always 2>&1
Unknown option to base64-output: always
Alternatives are: 'NEVER','AUTO','UNSPEC','DECODE-ROWS'
#
# Expect error for incomplete --base64-output argument.
# MYSQL_BINLOG std_data/master-bin.000001 --base64-output 2>&1
2022-09-23 10:28:12 +02:00
mariadb-binlog: option '--base64-output' requires an argument
2021-05-12 19:46:58 +02:00
#
# Ensure --base64-output=auto outputs the same result as unspecified
# MYSQL_BINLOG -v MYSQLD_DATADIR/master-bin.000001 > MYSQLTEST_VARDIR/tmp/mysqlbinlog_nob64spec.out
# MYSQL_BINLOG --base64-output=auto -v MYSQLD_DATADIR/master-bin.000001 > MYSQLTEST_VARDIR/tmp/mysqlbinlog_b64auto.out
2010-12-29 04:52:57 +01:00
RESET MASTER;
CREATE DATABASE test1;
USE test1;
CREATE TABLE t1(id int);
DROP DATABASE test1;
CREATE DATABASE test1;
USE test1;
CREATE TABLE t1(id int);
DROP TABLE t1;
DROP DATABASE test1;
FLUSH LOGS;
fixes for test failures
and small collateral changes
mysql-test/lib/My/Test.pm:
somehow with "print" we get truncated writes sometimes
mysql-test/suite/perfschema/r/digest_table_full.result:
md5 hashes of statement digests differ, because yacc token codes are different in mariadb
mysql-test/suite/perfschema/r/dml_handler.result:
host table is not ported over yet
mysql-test/suite/perfschema/r/information_schema.result:
host table is not ported over yet
mysql-test/suite/perfschema/r/nesting.result:
this differs, because we don't rewrite general log queries, and multi-statement
packets are logged as a one entry. this result file is identical to what mysql-5.6.5
produces with the --log-raw option.
mysql-test/suite/perfschema/r/relaylog.result:
MariaDB modifies the binlog index file directly, while MySQL 5.6 has a feature "crash-safe binlog index" and modifies a special "crash-safe" shadow copy of the index file and then moves it over. That's why this test shows "NONE" index file writes in MySQL and "MANY" in MariaDB.
mysql-test/suite/perfschema/r/server_init.result:
MariaDB initializes the "manager" resources from the "manager" thread, and starts this thread only when --flush-time is not 0. MySQL 5.6 initializes "manager" resources unconditionally on server startup.
mysql-test/suite/perfschema/r/stage_mdl_global.result:
this differs, because MariaDB disables query cache when query_cache_size=0. MySQL does not
do that, and this causes useless mutex locks and waits.
mysql-test/suite/perfschema/r/statement_digest.result:
md5 hashes of statement digests differ, because yacc token codes are different in mariadb
mysql-test/suite/perfschema/r/statement_digest_consumers.result:
md5 hashes of statement digests differ, because yacc token codes are different in mariadb
mysql-test/suite/perfschema/r/statement_digest_long_query.result:
md5 hashes of statement digests differ, because yacc token codes are different in mariadb
mysql-test/suite/rpl/r/rpl_mixed_drop_create_temp_table.result:
will be updated to match 5.6 when alfranio.correia@oracle.com-20110512172919-c1b5kmum4h52g0ni and anders.song@greatopensource.com-20110105052107-zoab0bsf5a6xxk2y are merged
mysql-test/suite/rpl/r/rpl_non_direct_mixed_mixing_engines.result:
will be updated to match 5.6 when anders.song@greatopensource.com-20110105052107-zoab0bsf5a6xxk2y is merged
2012-09-27 20:09:46 +02:00
include/show_binlog_events.inc
2010-12-29 04:52:57 +01:00
Log_name Pos Event_type Server_id End_log_pos Info
2012-09-13 14:31:29 +02:00
master-bin.000002 # Binlog_checkpoint # # master-bin.000002
2013-03-26 10:35:34 +01:00
master-bin.000002 # Gtid # # GTID #-#-#
2010-12-29 04:52:57 +01:00
master-bin.000002 # Query # # CREATE DATABASE test1
2013-03-26 10:35:34 +01:00
master-bin.000002 # Gtid # # GTID #-#-#
2010-12-29 04:52:57 +01:00
master-bin.000002 # Query # # use `test1`; CREATE TABLE t1(id int)
2013-03-26 10:35:34 +01:00
master-bin.000002 # Gtid # # GTID #-#-#
2010-12-29 04:52:57 +01:00
master-bin.000002 # Query # # DROP DATABASE test1
2013-03-26 10:35:34 +01:00
master-bin.000002 # Gtid # # GTID #-#-#
2010-12-29 04:52:57 +01:00
master-bin.000002 # Query # # CREATE DATABASE test1
2013-03-26 10:35:34 +01:00
master-bin.000002 # Gtid # # GTID #-#-#
2010-12-29 04:52:57 +01:00
master-bin.000002 # Query # # use `test1`; CREATE TABLE t1(id int)
2013-03-26 10:35:34 +01:00
master-bin.000002 # Gtid # # GTID #-#-#
2010-12-29 06:22:52 +01:00
master-bin.000002 # Query # # use `test1`; DROP TABLE `t1` /* generated by server */
2013-03-26 10:35:34 +01:00
master-bin.000002 # Gtid # # GTID #-#-#
2010-12-29 04:52:57 +01:00
master-bin.000002 # Query # # DROP DATABASE test1
2011-03-25 15:16:13 +01:00
RESET MASTER;
USE test;
CREATE TABLE t1 (a INT);
SET GLOBAL SERVER_ID = 2;
DROP TABLE t1;
FLUSH LOGS;
SHOW TABLES IN test;
Tables_in_test
t1
SHOW TABLES IN test;
Tables_in_test
SET GLOBAL SERVER_ID = 1;
2013-09-12 20:36:58 +02:00
#
# MDEV-4645: Incorrect reads of frozen binlog events;
# FDE corrupted in relay log
#
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
DELIMITER /*!*/;
# at 4
2013-09-25 19:41:41 +02:00
#130807 20:29:20 server id 1 end_log_pos 106
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# 4 |20 ae 02 52 |0f |01 00 00 00 |66 00 00 00 |6a 00 00 00 |00 01
#
# 17 04 00 35 2e 31 2e 36 33 2d 67 6f 6f 67 6c 65 2d |..5.1.63-google-|
# 27 64 65 62 75 67 2d 6c 6f 67 00 00 00 00 00 00 00 |debug-log.......|
# 37 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
# 47 00 00 00 00 20 ae 02 52 17 38 0d 00 08 00 12 00 |.... ..R.8......|
# 57 04 04 04 04 12 00 00 53 00 04 1a 08 00 00 00 08 |.......S........|
# 67 08 08 02 |...|
#
2013-09-25 19:41:41 +02:00
# Event: Start: binlog v 4, server v 5.1.63-google-debug-log created 130807 20:29:20 at startup
2013-09-12 20:36:58 +02:00
ROLLBACK/*!*/;
BINLOG '
IK4CUg8BAAAAZgAAAGoAAAAAAQQANS4xLjYzLWdvb2dsZS1kZWJ1Zy1sb2cAAAAAAAAAAAAAAAAA
AAAAAAAAAAAAAAAAAAAgrgJSFzgNAAgAEgAEBAQEEgAAUwAEGggAAAAICAgC
'/*!*/;
# at 106
2013-09-25 19:41:41 +02:00
#130807 20:29:24 server id 1 end_log_pos 207
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# 6a |24 ae 02 52 |02 |01 00 00 00 |65 00 00 00 |cf 00 00 00 |00 00
#
# 81 01 00 00 00 00 00 00 00 00 00 00 1a 00 00 00 40 |...............@|
# 91 00 00 01 00 00 00 00 00 00 00 00 06 03 73 74 64 |.............std|
# a1 04 08 00 08 00 08 00 00 63 72 65 61 74 65 20 74 |........create t|
# b1 61 62 6c 65 20 74 65 73 74 2e 74 31 20 28 69 64 |able test.t1 (id|
# c1 20 69 6e 74 20 6e 6f 74 20 6e 75 6c 6c 29 | int not null)|
#
# Event: Query thread_id=1 exec_time=0 error_code=0
SET TIMESTAMP=1375907364/*!*/;
SET @@session.pseudo_thread_id=1/*!*/;
2022-07-23 16:38:03 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=1, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1/*!*/;
2013-09-12 20:36:58 +02:00
SET @@session.sql_mode=0/*!*/;
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=8/*!*/;
2013-09-12 20:36:58 +02:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
create table test.t1 (id int not null)
/*!*/;
# at 207
2013-09-25 19:41:41 +02:00
#130807 20:29:26 server id 1 end_log_pos 305
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# cf |26 ae 02 52 |02 |01 00 00 00 |62 00 00 00 |31 01 00 00 |00 00
#
# e6 01 00 00 00 00 00 00 00 00 00 00 1a 00 00 00 40 |...............@|
# f6 00 00 01 00 00 00 00 00 00 00 00 06 03 73 74 64 |.............std|
# 106 04 08 00 08 00 08 00 00 69 6e 73 65 72 74 20 69 |........insert i|
# 116 6e 74 6f 20 74 65 73 74 2e 74 31 20 28 69 64 29 |nto test.t1 (id)|
# 126 20 76 61 6c 75 65 73 20 28 31 29 | values (1)|
#
# Event: Query thread_id=1 exec_time=0 error_code=0
SET TIMESTAMP=1375907366/*!*/;
insert into test.t1 (id) values (1)
/*!*/;
# at 305
2013-09-25 19:41:41 +02:00
#130807 20:29:28 server id 1 end_log_pos 386
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# 131 |28 ae 02 52 |02 |01 00 00 00 |51 00 00 00 |82 01 00 00 |00 00
#
# 148 01 00 00 00 00 00 00 00 00 00 00 1a 00 00 00 40 |...............@|
# 158 00 00 01 00 00 00 00 00 00 00 00 06 03 73 74 64 |.............std|
# 168 04 08 00 08 00 08 00 00 64 72 6f 70 20 74 61 62 |........drop tab|
# 178 6c 65 20 74 65 73 74 2e 74 31 |le test.t1|
#
# Event: Query thread_id=1 exec_time=0 error_code=0
SET TIMESTAMP=1375907368/*!*/;
drop table test.t1
/*!*/;
DELIMITER ;
# End of log file
ROLLBACK /* added by mysqlbinlog */;
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
DELIMITER /*!*/;
# at 4
2013-09-25 19:41:41 +02:00
#130807 20:25:35 server id 1 end_log_pos 106
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# 4 |3f ad 02 52 |0f |01 00 00 00 |66 00 00 00 |6a 00 00 00 |00 00
#
# 17 04 00 35 2e 31 2e 36 33 2d 67 6f 6f 67 6c 65 2d |..5.1.63-google-|
# 27 64 65 62 75 67 2d 6c 6f 67 00 00 00 00 00 00 00 |debug-log.......|
# 37 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
# 47 00 00 00 00 3f ad 02 52 1b 38 0d 00 08 00 12 00 |....?..R.8......|
# 57 04 04 04 04 12 00 00 53 00 04 1a 08 00 00 00 08 |.......S........|
# 67 08 08 02 |...|
#
2013-09-25 19:41:41 +02:00
# Event: Start: binlog v 4, server v 5.1.63-google-debug-log created 130807 20:25:35 at startup
2013-09-12 20:36:58 +02:00
ROLLBACK/*!*/;
BINLOG '
P60CUg8BAAAAZgAAAGoAAAAAAAQANS4xLjYzLWdvb2dsZS1kZWJ1Zy1sb2cAAAAAAAAAAAAAAAAA
AAAAAAAAAAAAAAAAAAA/rQJSGzgNAAgAEgAEBAQEEgAAUwAEGggAAAAICAgC
'/*!*/;
# at 106
2013-09-25 19:41:41 +02:00
#130807 20:25:41 server id 1 end_log_pos 211
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# 6a |45 ad 02 52 |02 |01 00 00 00 |69 00 00 00 |d3 00 00 00 |00 00
#
# 85 01 00 00 00 01 00 00 00 00 00 00 1a 00 00 00 40 |...............@|
# 95 00 00 01 00 00 00 00 00 00 00 00 06 03 73 74 64 |.............std|
# a5 04 08 00 08 00 08 00 00 63 72 65 61 74 65 20 74 |........create t|
# b5 61 62 6c 65 20 74 65 73 74 2e 74 31 20 28 69 64 |able test.t1 (id|
# c5 20 69 6e 74 20 6e 6f 74 20 6e 75 6c 6c 29 | int not null)|
#
# Event: Query thread_id=1 exec_time=1 error_code=0
SET TIMESTAMP=1375907141/*!*/;
SET @@session.pseudo_thread_id=1/*!*/;
2022-07-23 16:38:03 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=1, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1/*!*/;
2013-09-12 20:36:58 +02:00
SET @@session.sql_mode=0/*!*/;
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=8/*!*/;
2013-09-12 20:36:58 +02:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
create table test.t1 (id int not null)
/*!*/;
# at 211
2013-09-25 19:41:41 +02:00
#130807 20:25:44 server id 1 end_log_pos 313
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# d3 |48 ad 02 52 |02 |01 00 00 00 |66 00 00 00 |39 01 00 00 |00 00
#
# ee 01 00 00 00 00 00 00 00 00 00 00 1a 00 00 00 40 |...............@|
# fe 00 00 01 00 00 00 00 00 00 00 00 06 03 73 74 64 |.............std|
# 10e 04 08 00 08 00 08 00 00 69 6e 73 65 72 74 20 69 |........insert i|
# 11e 6e 74 6f 20 74 65 73 74 2e 74 31 20 28 69 64 29 |nto test.t1 (id)|
# 12e 20 76 61 6c 75 65 73 20 28 31 29 | values (1)|
#
# Event: Query thread_id=1 exec_time=0 error_code=0
SET TIMESTAMP=1375907144/*!*/;
insert into test.t1 (id) values (1)
/*!*/;
# at 313
2013-09-25 19:41:41 +02:00
#130807 20:25:48 server id 1 end_log_pos 398
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# 139 |4c ad 02 52 |02 |01 00 00 00 |55 00 00 00 |8e 01 00 00 |00 00
#
# 154 01 00 00 00 00 00 00 00 00 00 00 1a 00 00 00 40 |...............@|
# 164 00 00 01 00 00 00 00 00 00 00 00 06 03 73 74 64 |.............std|
# 174 04 08 00 08 00 08 00 00 64 72 6f 70 20 74 61 62 |........drop tab|
# 184 6c 65 20 74 65 73 74 2e 74 31 |le test.t1|
#
# Event: Query thread_id=1 exec_time=0 error_code=0
SET TIMESTAMP=1375907148/*!*/;
drop table test.t1
/*!*/;
DELIMITER ;
# End of log file
ROLLBACK /* added by mysqlbinlog */;
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
DELIMITER /*!*/;
# at 4
2013-09-25 19:41:41 +02:00
#130807 20:20:55 server id 1 end_log_pos 106
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# 4 |27 ac 02 52 |0f |01 00 00 00 |66 00 00 00 |6a 00 00 00 |00 01
#
# 17 04 00 35 2e 31 2e 36 33 2d 67 6f 6f 67 6c 65 2d |..5.1.63-google-|
# 27 64 65 62 75 67 2d 6c 6f 67 00 00 00 00 00 00 00 |debug-log.......|
# 37 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
# 47 00 00 00 00 27 ac 02 52 1f 38 0d 00 08 00 12 00 |....'..R.8......|
# 57 04 04 04 04 12 00 00 53 00 04 1a 08 00 00 00 08 |.......S........|
# 67 08 08 02 |...|
#
2013-09-25 19:41:41 +02:00
# Event: Start: binlog v 4, server v 5.1.63-google-debug-log created 130807 20:20:55 at startup
2013-09-12 20:36:58 +02:00
ROLLBACK/*!*/;
BINLOG '
J6wCUg8BAAAAZgAAAGoAAAAAAQQANS4xLjYzLWdvb2dsZS1kZWJ1Zy1sb2cAAAAAAAAAAAAAAAAA
AAAAAAAAAAAAAAAAAAAnrAJSHzgNAAgAEgAEBAQEEgAAUwAEGggAAAAICAgC
'/*!*/;
# at 106
2013-09-25 19:41:41 +02:00
#130807 20:21:19 server id 1 end_log_pos 215
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# 6a |3f ac 02 52 |02 |01 00 00 00 |6d 00 00 00 |d7 00 00 00 |00 00
#
# 89 01 00 00 00 00 00 00 00 00 00 00 1a 00 00 00 40 |...............@|
# 99 00 00 01 00 00 00 00 00 00 00 00 06 03 73 74 64 |.............std|
# a9 04 08 00 08 00 08 00 00 63 72 65 61 74 65 20 74 |........create t|
# b9 61 62 6c 65 20 74 65 73 74 2e 74 31 20 28 69 64 |able test.t1 (id|
# c9 20 69 6e 74 20 6e 6f 74 20 6e 75 6c 6c 29 | int not null)|
#
# Event: Query thread_id=1 exec_time=0 error_code=0
SET TIMESTAMP=1375906879/*!*/;
SET @@session.pseudo_thread_id=1/*!*/;
2022-07-23 16:38:03 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=1, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1/*!*/;
2013-09-12 20:36:58 +02:00
SET @@session.sql_mode=0/*!*/;
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=8/*!*/;
2013-09-12 20:36:58 +02:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
create table test.t1 (id int not null)
/*!*/;
# at 215
2013-09-25 19:41:41 +02:00
#130807 20:21:31 server id 1 end_log_pos 321
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# d7 |4b ac 02 52 |02 |01 00 00 00 |6a 00 00 00 |41 01 00 00 |00 00
#
# f6 01 00 00 00 00 00 00 00 00 00 00 1a 00 00 00 40 |...............@|
# 106 00 00 01 00 00 00 00 00 00 00 00 06 03 73 74 64 |.............std|
# 116 04 08 00 08 00 08 00 00 69 6e 73 65 72 74 20 69 |........insert i|
# 126 6e 74 6f 20 74 65 73 74 2e 74 31 20 28 69 64 29 |nto test.t1 (id)|
# 136 20 76 61 6c 75 65 73 20 28 31 29 | values (1)|
#
# Event: Query thread_id=1 exec_time=0 error_code=0
SET TIMESTAMP=1375906891/*!*/;
insert into test.t1 (id) values (1)
/*!*/;
# at 321
2013-09-25 19:41:41 +02:00
#130807 20:21:41 server id 1 end_log_pos 410
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# 141 |55 ac 02 52 |02 |01 00 00 00 |59 00 00 00 |9a 01 00 00 |00 00
#
# 160 01 00 00 00 00 00 00 00 00 00 00 1a 00 00 00 40 |...............@|
# 170 00 00 01 00 00 00 00 00 00 00 00 06 03 73 74 64 |.............std|
# 180 04 08 00 08 00 08 00 00 64 72 6f 70 20 74 61 62 |........drop tab|
# 190 6c 65 20 74 65 73 74 2e 74 31 |le test.t1|
#
# Event: Query thread_id=1 exec_time=0 error_code=0
SET TIMESTAMP=1375906901/*!*/;
drop table test.t1
/*!*/;
DELIMITER ;
# End of log file
ROLLBACK /* added by mysqlbinlog */;
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
/*!40019 SET @@session.max_insert_delayed_threads=0*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
DELIMITER /*!*/;
# at 4
2013-09-25 19:41:41 +02:00
#130807 20:38:51 server id 1 end_log_pos 106
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# 4 |5b b0 02 52 |0f |01 00 00 00 |66 00 00 00 |6a 00 00 00 |00 00
#
# 17 04 00 35 2e 31 2e 36 33 2d 67 6f 6f 67 6c 65 2d |..5.1.63-google-|
# 27 64 65 62 75 67 2d 6c 6f 67 00 00 00 00 00 00 00 |debug-log.......|
# 37 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
# 47 00 00 00 00 5b b0 02 52 13 38 0d 00 08 00 12 00 |....[..R.8......|
# 57 04 04 04 04 12 00 00 53 00 04 1a 08 00 00 00 08 |.......S........|
# 67 08 08 02 |...|
#
2013-09-25 19:41:41 +02:00
# Event: Start: binlog v 4, server v 5.1.63-google-debug-log created 130807 20:38:51 at startup
2013-09-12 20:36:58 +02:00
ROLLBACK/*!*/;
BINLOG '
W7ACUg8BAAAAZgAAAGoAAAAAAAQANS4xLjYzLWdvb2dsZS1kZWJ1Zy1sb2cAAAAAAAAAAAAAAAAA
AAAAAAAAAAAAAAAAAABbsAJSEzgNAAgAEgAEBAQEEgAAUwAEGggAAAAICAgC
'/*!*/;
# at 106
2013-09-25 19:41:41 +02:00
#130807 20:38:53 server id 1 end_log_pos 203
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# 6a |5d b0 02 52 |02 |01 00 00 00 |61 00 00 00 |cb 00 00 00 |00 00
#
# 7d 01 00 00 00 00 00 00 00 00 00 00 1a 00 00 00 40 |...............@|
# 8d 00 00 01 00 00 00 00 00 00 00 00 06 03 73 74 64 |.............std|
# 9d 04 08 00 08 00 08 00 00 63 72 65 61 74 65 20 74 |........create t|
# ad 61 62 6c 65 20 74 65 73 74 2e 74 31 20 28 69 64 |able test.t1 (id|
# bd 20 69 6e 74 20 6e 6f 74 20 6e 75 6c 6c 29 | int not null)|
#
# Event: Query thread_id=1 exec_time=0 error_code=0
SET TIMESTAMP=1375907933/*!*/;
SET @@session.pseudo_thread_id=1/*!*/;
2022-07-23 16:38:03 +02:00
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=1, @@session.unique_checks=1, @@session.autocommit=1, @@session.check_constraint_checks=1/*!*/;
2013-09-12 20:36:58 +02:00
SET @@session.sql_mode=0/*!*/;
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
/*!\C latin1 *//*!*/;
[MDEV-30824] Fix binlog to use 'String' for setting 'character_set_client'
Commit a923d6f49c1ad6fd3f4d6ec02e444c26e4d1dfa8 disabled numeric setting
of character_set_* variables with non-default values:
MariaDB [(none)]> set character_set_client=224;
ERROR 1115 (42000): Unknown character set: '224'
However the corresponding binlog functionality still write numeric
values for log event, and this will break binlog replay if the value is
not default. Now make the server use 'String' type for
'character_set_client' when generating binlog events
Before:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=224,@@session.collation_connection=224,@@session.collation_server=33/*!*/;
After:
/*!\C utf8mb4 *//*!*/;
SET @@session.character_set_client=utf8mb4,@@session.collation_connection=33,@@session.collation_server=8/*!*/;
Note: prior to the previous commit, setting with '224' or '45' or
'utf8mb4' have the same effect, as they all set the parameter to
'utf8mb4'.
All new code of the whole pull request, including one or several files
that are either new files or modified ones, are contributed under the
BSD-new license. I am contributing on behalf of my employer Amazon Web
Services, Inc.
2023-03-15 20:14:01 +01:00
SET @@session.character_set_client=latin1,@@session.collation_connection=8,@@session.collation_server=8/*!*/;
2013-09-12 20:36:58 +02:00
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
create table test.t1 (id int not null)
/*!*/;
# at 203
2013-09-25 19:41:41 +02:00
#130807 20:38:55 server id 1 end_log_pos 297
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# cb |5f b0 02 52 |02 |01 00 00 00 |5e 00 00 00 |29 01 00 00 |00 00
#
# de 01 00 00 00 00 00 00 00 00 00 00 1a 00 00 00 40 |...............@|
# ee 00 00 01 00 00 00 00 00 00 00 00 06 03 73 74 64 |.............std|
# fe 04 08 00 08 00 08 00 00 69 6e 73 65 72 74 20 69 |........insert i|
# 10e 6e 74 6f 20 74 65 73 74 2e 74 31 20 28 69 64 29 |nto test.t1 (id)|
# 11e 20 76 61 6c 75 65 73 20 28 31 29 | values (1)|
#
# Event: Query thread_id=1 exec_time=0 error_code=0
SET TIMESTAMP=1375907935/*!*/;
insert into test.t1 (id) values (1)
/*!*/;
# at 297
2013-09-25 19:41:41 +02:00
#130807 20:38:57 server id 1 end_log_pos 374
2013-09-12 20:36:58 +02:00
# Position
# |Timestamp |Type |Master ID |Size |Master Pos |Flags
# 129 |61 b0 02 52 |02 |01 00 00 00 |4d 00 00 00 |76 01 00 00 |00 00
#
# 13c 01 00 00 00 00 00 00 00 00 00 00 1a 00 00 00 40 |...............@|
# 14c 00 00 01 00 00 00 00 00 00 00 00 06 03 73 74 64 |.............std|
# 15c 04 08 00 08 00 08 00 00 64 72 6f 70 20 74 61 62 |........drop tab|
# 16c 6c 65 20 74 65 73 74 2e 74 31 |le test.t1|
#
# Event: Query thread_id=1 exec_time=0 error_code=0
SET TIMESTAMP=1375907937/*!*/;
drop table test.t1
/*!*/;
DELIMITER ;
# End of log file
ROLLBACK /* added by mysqlbinlog */;
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
2022-09-23 10:28:12 +02:00
mariadb-binlog Ver VER for OS at ARCH
2021-02-05 13:57:46 +01:00
#
# Test --rewrite-db
#
RESET MASTER;
CREATE TABLE t1 (a int);
INSERT INTO t1 values(1);
DROP TABLE t1;
FLUSH LOGS;
ERROR: Bad syntax in rewrite-db: missing '->'
ERROR: Bad syntax in rewrite-db: empty TO db
ERROR: Bad syntax in rewrite-db: empty TO db
ERROR: Bad syntax in rewrite-db: empty FROM db
ERROR: Bad syntax in rewrite-db: empty FROM db
