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
|
|
|
stop slave;
|
|
|
|
|
drop table if exists t1,t2,t3,t4,t5,t6,t7,t8,t9;
|
|
|
|
|
reset master;
|
|
|
|
|
reset slave;
|
|
|
|
|
drop table if exists t1,t2,t3,t4,t5,t6,t7,t8,t9;
|
|
|
|
|
start slave;
|
|
|
|
|
==== Initialization ====
|
|
|
|
|
[on master]
|
|
|
|
|
SET @m_default_week_format= @@global.default_week_format;
|
|
|
|
|
SET @m_init_slave= @@global.init_slave;
|
|
|
|
|
SET @m_lc_time_names= @@global.lc_time_names;
|
|
|
|
|
SET @m_low_priority_updates= @@global.low_priority_updates;
|
|
|
|
|
SET @m_relay_log_purge= @@global.relay_log_purge;
|
|
|
|
|
SET @m_slave_exec_mode= @@global.slave_exec_mode;
|
|
|
|
|
SET @m_sql_mode= @@global.sql_mode;
|
|
|
|
|
SET @m_sync_binlog= @@global.sync_binlog;
|
|
|
|
|
[on slave]
|
|
|
|
|
SET @s_default_week_format= @@global.default_week_format;
|
|
|
|
|
SET @s_init_slave= @@global.init_slave;
|
|
|
|
|
SET @s_lc_time_names= @@global.lc_time_names;
|
|
|
|
|
SET @s_low_priority_updates= @@global.low_priority_updates;
|
|
|
|
|
SET @s_relay_log_purge= @@global.relay_log_purge;
|
|
|
|
|
SET @s_slave_exec_mode= @@global.slave_exec_mode;
|
|
|
|
|
SET @s_sql_mode= @@global.sql_mode;
|
|
|
|
|
SET @s_sync_binlog= @@global.sync_binlog;
|
|
|
|
|
SET @@global.relay_log_purge = OFF;
|
|
|
|
|
SET @@global.sync_binlog = 1000000;
|
|
|
|
|
SET @@global.slave_exec_mode = 'STRICT';
|
|
|
|
|
SET @@sql_big_selects = OFF;
|
|
|
|
|
SET @@last_insert_id = 10;
|
|
|
|
|
SET @@global.low_priority_updates = OFF;
|
|
|
|
|
SET @@local.low_priority_updates = OFF;
|
|
|
|
|
SET @@global.default_week_format = 1;
|
|
|
|
|
SET @@local.default_week_format = 2;
|
|
|
|
|
SET @@global.lc_time_names = 'zh_HK';
|
|
|
|
|
SET @@local.lc_time_names = 'zh_TW';
|
|
|
|
|
SET @@global.sql_mode = 'ALLOW_INVALID_DATES';
|
|
|
|
|
SET @@local.sql_mode = 'ANSI_QUOTES,ERROR_FOR_DIVISION_BY_ZERO,HIGH_NOT_PRECEDENCE';
|
|
|
|
|
SET @user_num = 10;
|
|
|
|
|
SET @user_text = 'Alunda';
|
|
|
|
|
[on master]
|
|
|
|
|
**** Resetting master and slave ****
|
BUG#37975: wait_for_slave_* should increase the timeout
Problem 1: tests often fail in pushbuild with a timeout when waiting
for the slave to start/stop/receive error.
Fix 1: Updated the wait_for_slave_* macros in the following way:
- The timeout is increased by a factor ten
- Refactored the macros so that wait_for_slave_param does the work for
the other macros.
Problem 2: Tests are often incorrectly written, lacking a
source include/wait_for_slave_to_[start|stop].inc.
Fix 2: Improved the chance to get it right by adding
include/start_slave.inc and include/stop_slave.inc, and updated tests
to use these.
Problem 3: The the built-in test language command
wait_for_slave_to_stop is a misnomer (does not wait for the slave io
thread) and does not give as much debug info in case of failure as
the otherwise equivalent macro
source include/wait_for_slave_sql_to_stop.inc
Fix 3: Replaced all calls to the built-in command by a call to the
macro.
Problem 4: Some, but not all, of the wait_for_slave_* macros had an
implicit connection slave. This made some tests confusing to read,
and made it more difficult to use the macro in circular replication
scenarios, where the connection named master needs to wait.
Fix 4: Removed the implicit connection slave from all
wait_for_slave_* macros, and updated tests to use an explicit
connection slave where necessary.
Problem 5: The macros wait_slave_status.inc and wait_show_pattern.inc
were unused. Moreover, using them is difficult and error-prone.
Fix 5: remove these macros.
Problem 6: log_bin_trust_function_creators_basic failed when running
tests because it assumed @@global.log_bin_trust_function_creators=1,
and some tests modified this variable without resetting it to its
original value.
Fix 6: All tests that use this variable have been updated so that
they reset the value at end of test.
mysql-test/extra/rpl_tests/rpl_extraMaster_Col.test:
Replaced wait_for_slave_to_stop by include/wait_for_slave_sql_to_stop.inc
mysql-test/extra/rpl_tests/rpl_extraSlave_Col.test:
Replaced wait_for_slave_to_stop by include/wait_for_slave_sql_to_stop.inc
Added connection slave since includ/wait_for_slave_sql_to_stop.inc
does not do that anymore.
mysql-test/extra/rpl_tests/rpl_log.test:
Replaced start slave+wait_slave_status by start_slave.inc
mysql-test/include/reset_master_and_slave.inc:
replaced start/stop slave by start_slave.inc/stop_slave.inc
mysql-test/include/sync_slave_io_with_master.inc:
Improved comments and error message.
mysql-test/include/wait_for_slave_io_to_stop.inc:
Refactored to use wait_for_slave_param.inc.
Removed connection slave.
mysql-test/include/wait_for_slave_param.inc:
- Improved usage instructions
- Added more debug info in case of timeout
- Added parameters $slave_param_comparison, $slave_timeout,
$slave_keep_connection, $slave_error_message
mysql-test/include/wait_for_slave_sql_error.inc:
Refactored to use wait_for_slave_param.inc.
Removed connection slave.
mysql-test/include/wait_for_slave_sql_to_start.inc:
Refactored to use wait_for_slave_param.inc.
Removed connection slave.
mysql-test/include/wait_for_slave_sql_to_stop.inc:
Refactored to use wait_for_slave_param.inc.
Removed connection slave.
mysql-test/include/wait_for_slave_to_start.inc:
Refactored to use wait_for_slave_param.inc.
Removed connection slave.
mysql-test/include/wait_for_slave_to_stop.inc:
Refactored to use wait_for_slave_param.inc.
Removed connection slave.
mysql-test/include/wait_show_pattern.inc:
Removed unused (and error-prone) file
mysql-test/include/wait_slave_status.inc:
Removed unused (and error-prone) file
mysql-test/suite/binlog/t/binlog_auto_increment_bug33029.test:
Renamed $keep_connection to $slave_keep_connection.
mysql-test/suite/rpl/t/rpl_bug26395.test:
Replace stop slave by stop_slave.inc
mysql-test/suite/rpl/t/rpl_circular_for_4_hosts.test:
Replace start/stop slave by start_slave.inc/stop_slave.inc.
Replace wait_for_slave_param by wait_for_slave_sql_to_stop.inc.
mysql-test/suite/rpl/t/rpl_dual_pos_advance.test:
Renamed $keep_connection to $slave_keep_connection.
mysql-test/suite/rpl/t/rpl_flushlog_loop.test:
Replace wait_slave_status by start_slave.inc
mysql-test/suite/rpl/t/rpl_idempotency.test:
Added connection slave since wait_for_slave_sql_to_stop.inc does not
do that any more.
mysql-test/suite/rpl/t/rpl_incident.test:
Replaced wait_for_slave_to_stop by wait_for_slave_sql_to_stop.inc
mysql-test/suite/rpl/t/rpl_init_slave.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
Replaced save_master_pos;connection slave;sync_with_master by
sync_slave_with_master.
mysql-test/suite/rpl/t/rpl_log_pos.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
Replaced wait_for_slave_param by other wait_for_slave_* macros.
mysql-test/suite/rpl/t/rpl_packet.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_row_inexist_tbl.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_row_stop_middle_update.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_row_until.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
Replaced save_master_pos;connection slave;sync_with_master by
sync_slave_with_master.
mysql-test/suite/rpl/t/rpl_server_id1.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_slave_grp_exec.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_slave_skip.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_slave_status.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_sp.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/suite/rpl/t/rpl_sp_effects.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/suite/rpl/t/rpl_stm_until.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
Replaced save_master_pos;connection slave;sync_with_master by
sync_slave_with_master.
mysql-test/suite/rpl_ndb/t/rpl_ndb_basic.test:
Replaced start slave by start_slave.inc.
Added explicit connection slave since wait_for_slave_sql_to_stop.inc
does not do that anymore.
mysql-test/t/disabled.def:
Disabled failing test.
mysql-test/t/func_time.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/t/grant.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/t/grant2.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/t/innodb_notembedded.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/t/log_bin_trust_function_creators_func.test:
Restore @@global.log_bin_trust_function_creators at end of test.
Clean up at end of test by dropping the created user.
mysql-test/t/query_cache.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/t/query_cache_notembedded.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/t/rpl_init_slave_func.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/t/timezone2.test:
Restore @@global.log_bin_trust_function_creators at end of test.
2008-07-10 18:09:39 +02:00
|
|
|
include/stop_slave.inc
|
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
|
|
|
RESET SLAVE;
|
|
|
|
|
RESET MASTER;
|
BUG#37975: wait_for_slave_* should increase the timeout
Problem 1: tests often fail in pushbuild with a timeout when waiting
for the slave to start/stop/receive error.
Fix 1: Updated the wait_for_slave_* macros in the following way:
- The timeout is increased by a factor ten
- Refactored the macros so that wait_for_slave_param does the work for
the other macros.
Problem 2: Tests are often incorrectly written, lacking a
source include/wait_for_slave_to_[start|stop].inc.
Fix 2: Improved the chance to get it right by adding
include/start_slave.inc and include/stop_slave.inc, and updated tests
to use these.
Problem 3: The the built-in test language command
wait_for_slave_to_stop is a misnomer (does not wait for the slave io
thread) and does not give as much debug info in case of failure as
the otherwise equivalent macro
source include/wait_for_slave_sql_to_stop.inc
Fix 3: Replaced all calls to the built-in command by a call to the
macro.
Problem 4: Some, but not all, of the wait_for_slave_* macros had an
implicit connection slave. This made some tests confusing to read,
and made it more difficult to use the macro in circular replication
scenarios, where the connection named master needs to wait.
Fix 4: Removed the implicit connection slave from all
wait_for_slave_* macros, and updated tests to use an explicit
connection slave where necessary.
Problem 5: The macros wait_slave_status.inc and wait_show_pattern.inc
were unused. Moreover, using them is difficult and error-prone.
Fix 5: remove these macros.
Problem 6: log_bin_trust_function_creators_basic failed when running
tests because it assumed @@global.log_bin_trust_function_creators=1,
and some tests modified this variable without resetting it to its
original value.
Fix 6: All tests that use this variable have been updated so that
they reset the value at end of test.
mysql-test/extra/rpl_tests/rpl_extraMaster_Col.test:
Replaced wait_for_slave_to_stop by include/wait_for_slave_sql_to_stop.inc
mysql-test/extra/rpl_tests/rpl_extraSlave_Col.test:
Replaced wait_for_slave_to_stop by include/wait_for_slave_sql_to_stop.inc
Added connection slave since includ/wait_for_slave_sql_to_stop.inc
does not do that anymore.
mysql-test/extra/rpl_tests/rpl_log.test:
Replaced start slave+wait_slave_status by start_slave.inc
mysql-test/include/reset_master_and_slave.inc:
replaced start/stop slave by start_slave.inc/stop_slave.inc
mysql-test/include/sync_slave_io_with_master.inc:
Improved comments and error message.
mysql-test/include/wait_for_slave_io_to_stop.inc:
Refactored to use wait_for_slave_param.inc.
Removed connection slave.
mysql-test/include/wait_for_slave_param.inc:
- Improved usage instructions
- Added more debug info in case of timeout
- Added parameters $slave_param_comparison, $slave_timeout,
$slave_keep_connection, $slave_error_message
mysql-test/include/wait_for_slave_sql_error.inc:
Refactored to use wait_for_slave_param.inc.
Removed connection slave.
mysql-test/include/wait_for_slave_sql_to_start.inc:
Refactored to use wait_for_slave_param.inc.
Removed connection slave.
mysql-test/include/wait_for_slave_sql_to_stop.inc:
Refactored to use wait_for_slave_param.inc.
Removed connection slave.
mysql-test/include/wait_for_slave_to_start.inc:
Refactored to use wait_for_slave_param.inc.
Removed connection slave.
mysql-test/include/wait_for_slave_to_stop.inc:
Refactored to use wait_for_slave_param.inc.
Removed connection slave.
mysql-test/include/wait_show_pattern.inc:
Removed unused (and error-prone) file
mysql-test/include/wait_slave_status.inc:
Removed unused (and error-prone) file
mysql-test/suite/binlog/t/binlog_auto_increment_bug33029.test:
Renamed $keep_connection to $slave_keep_connection.
mysql-test/suite/rpl/t/rpl_bug26395.test:
Replace stop slave by stop_slave.inc
mysql-test/suite/rpl/t/rpl_circular_for_4_hosts.test:
Replace start/stop slave by start_slave.inc/stop_slave.inc.
Replace wait_for_slave_param by wait_for_slave_sql_to_stop.inc.
mysql-test/suite/rpl/t/rpl_dual_pos_advance.test:
Renamed $keep_connection to $slave_keep_connection.
mysql-test/suite/rpl/t/rpl_flushlog_loop.test:
Replace wait_slave_status by start_slave.inc
mysql-test/suite/rpl/t/rpl_idempotency.test:
Added connection slave since wait_for_slave_sql_to_stop.inc does not
do that any more.
mysql-test/suite/rpl/t/rpl_incident.test:
Replaced wait_for_slave_to_stop by wait_for_slave_sql_to_stop.inc
mysql-test/suite/rpl/t/rpl_init_slave.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
Replaced save_master_pos;connection slave;sync_with_master by
sync_slave_with_master.
mysql-test/suite/rpl/t/rpl_log_pos.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
Replaced wait_for_slave_param by other wait_for_slave_* macros.
mysql-test/suite/rpl/t/rpl_packet.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_row_inexist_tbl.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_row_stop_middle_update.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_row_until.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
Replaced save_master_pos;connection slave;sync_with_master by
sync_slave_with_master.
mysql-test/suite/rpl/t/rpl_server_id1.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_slave_grp_exec.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_slave_skip.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_slave_status.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/suite/rpl/t/rpl_sp.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/suite/rpl/t/rpl_sp_effects.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/suite/rpl/t/rpl_stm_until.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
Replaced save_master_pos;connection slave;sync_with_master by
sync_slave_with_master.
mysql-test/suite/rpl_ndb/t/rpl_ndb_basic.test:
Replaced start slave by start_slave.inc.
Added explicit connection slave since wait_for_slave_sql_to_stop.inc
does not do that anymore.
mysql-test/t/disabled.def:
Disabled failing test.
mysql-test/t/func_time.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/t/grant.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/t/grant2.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/t/innodb_notembedded.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/t/log_bin_trust_function_creators_func.test:
Restore @@global.log_bin_trust_function_creators at end of test.
Clean up at end of test by dropping the created user.
mysql-test/t/query_cache.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/t/query_cache_notembedded.test:
Restore @@global.log_bin_trust_function_creators at end of test.
mysql-test/t/rpl_init_slave_func.test:
Replaced start/stop slave by start_slave.inc/stop_slave.inc.
mysql-test/t/timezone2.test:
Restore @@global.log_bin_trust_function_creators at end of test.
2008-07-10 18:09:39 +02:00
|
|
|
include/start_slave.inc
|
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
|
|
|
[on slave]
|
2008-07-17 18:26:59 +02:00
|
|
|
SET @@global.init_slave = 'SELECT 1';
|
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
|
|
|
[on master]
|
|
|
|
|
CREATE TABLE tstmt (id INT AUTO_INCREMENT PRIMARY KEY,
|
|
|
|
|
truth BOOLEAN,
|
|
|
|
|
num INT,
|
|
|
|
|
text VARCHAR(100));
|
|
|
|
|
CREATE TABLE tproc LIKE tstmt;
|
|
|
|
|
CREATE TABLE tfunc LIKE tstmt;
|
|
|
|
|
CREATE TABLE ttrig LIKE tstmt;
|
|
|
|
|
CREATE TABLE tprep LIKE tstmt;
|
|
|
|
|
CREATE TABLE trigger_table (text CHAR(4));
|
|
|
|
|
==== Insert variables directly ====
|
|
|
|
|
---- global variables ----
|
|
|
|
|
SET @@global.relay_log_purge = ON;
|
|
|
|
|
INSERT INTO tstmt(truth) VALUES (@@global.relay_log_purge);
|
|
|
|
|
SET @@global.relay_log_purge = OFF;
|
|
|
|
|
INSERT INTO tstmt(truth) VALUES (@@global.relay_log_purge);
|
|
|
|
|
SET @@global.sync_binlog = 2000000;
|
|
|
|
|
INSERT INTO tstmt(num) VALUES (@@global.sync_binlog);
|
|
|
|
|
SET @@global.sync_binlog = 3000000;
|
|
|
|
|
INSERT INTO tstmt(num) VALUES (@@global.sync_binlog);
|
2008-07-17 18:26:59 +02:00
|
|
|
SET @@global.init_slave = 'SELECT 2';
|
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
|
|
|
INSERT INTO tstmt(text) VALUES (@@global.init_slave);
|
2008-07-17 18:26:59 +02:00
|
|
|
SET @@global.init_slave = 'SELECT 3';
|
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
|
|
|
INSERT INTO tstmt(text) VALUES (@@global.init_slave);
|
|
|
|
|
SET @@global.slave_exec_mode = 'IDEMPOTENT';
|
|
|
|
|
INSERT INTO tstmt(text) VALUES (@@global.slave_exec_mode);
|
|
|
|
|
SET @@global.slave_exec_mode = 'STRICT';
|
|
|
|
|
INSERT INTO tstmt(text) VALUES (@@global.slave_exec_mode);
|
|
|
|
|
---- session variables ----
|
|
|
|
|
SET @@sql_big_selects = ON;
|
|
|
|
|
INSERT INTO tstmt(truth) VALUES (@@sql_big_selects);
|
|
|
|
|
SET @@sql_big_selects = OFF;
|
|
|
|
|
INSERT INTO tstmt(truth) VALUES (@@sql_big_selects);
|
|
|
|
|
SET @@last_insert_id = 20;
|
|
|
|
|
INSERT INTO tstmt(num) VALUES (@@last_insert_id);
|
|
|
|
|
SET @@last_insert_id = 30;
|
|
|
|
|
INSERT INTO tstmt(num) VALUES (@@last_insert_id);
|
|
|
|
|
---- global and session variables ----
|
|
|
|
|
SET @@global.low_priority_updates = ON;
|
|
|
|
|
SET @@local.low_priority_updates = OFF;
|
|
|
|
|
INSERT INTO tstmt(truth) VALUES (@@global.low_priority_updates);
|
|
|
|
|
INSERT INTO tstmt(truth) VALUES (@@local.low_priority_updates);
|
|
|
|
|
SET @@global.low_priority_updates = OFF;
|
|
|
|
|
SET @@local.low_priority_updates = ON;
|
|
|
|
|
INSERT INTO tstmt(truth) VALUES (@@global.low_priority_updates);
|
|
|
|
|
INSERT INTO tstmt(truth) VALUES (@@local.low_priority_updates);
|
|
|
|
|
SET @@global.default_week_format = 3;
|
|
|
|
|
SET @@local.default_week_format = 4;
|
|
|
|
|
INSERT INTO tstmt(num) VALUES (@@global.default_week_format);
|
|
|
|
|
INSERT INTO tstmt(num) VALUES (@@local.default_week_format);
|
|
|
|
|
SET @@global.default_week_format = 5;
|
|
|
|
|
SET @@local.default_week_format = 6;
|
|
|
|
|
INSERT INTO tstmt(num) VALUES (@@global.default_week_format);
|
|
|
|
|
INSERT INTO tstmt(num) VALUES (@@local.default_week_format);
|
|
|
|
|
SET @@global.lc_time_names = 'sv_SE';
|
|
|
|
|
SET @@local.lc_time_names = 'sv_FI';
|
|
|
|
|
INSERT INTO tstmt(text) VALUES (@@global.lc_time_names);
|
|
|
|
|
INSERT INTO tstmt(text) VALUES (@@local.lc_time_names);
|
|
|
|
|
SET @@global.lc_time_names = 'ar_TN';
|
|
|
|
|
SET @@local.lc_time_names = 'ar_IQ';
|
|
|
|
|
INSERT INTO tstmt(text) VALUES (@@global.lc_time_names);
|
|
|
|
|
INSERT INTO tstmt(text) VALUES (@@local.lc_time_names);
|
|
|
|
|
SET @@global.sql_mode = '';
|
|
|
|
|
SET @@local.sql_mode = 'IGNORE_SPACE,NO_AUTO_CREATE_USER';
|
|
|
|
|
INSERT INTO tstmt(text) VALUES (@@global.sql_mode);
|
|
|
|
|
INSERT INTO tstmt(text) VALUES (@@local.sql_mode);
|
|
|
|
|
SET @@global.sql_mode = 'NO_AUTO_VALUE_ON_ZERO,NO_BACKSLASH_ESCAPES,NO_DIR_IN_CREATE,NO_ENGINE_SUBSTITUTION';
|
|
|
|
|
SET @@local.sql_mode = 'NO_FIELD_OPTIONS,NO_KEY_OPTIONS,NO_TABLE_OPTIONS';
|
|
|
|
|
INSERT INTO tstmt(text) VALUES (@@global.sql_mode);
|
|
|
|
|
INSERT INTO tstmt(text) VALUES (@@local.sql_mode);
|
|
|
|
|
---- user variables ----
|
|
|
|
|
SET @user_num = 20;
|
|
|
|
|
INSERT INTO tstmt(num) VALUES (@user_num);
|
|
|
|
|
SET @user_num = 30;
|
|
|
|
|
INSERT INTO tstmt(num) VALUES (@user_num);
|
|
|
|
|
SET @user_text = 'Bergsbrunna';
|
|
|
|
|
INSERT INTO tstmt(text) VALUES (@user_text);
|
|
|
|
|
SET @user_text = 'Centrum';
|
|
|
|
|
INSERT INTO tstmt(text) VALUES (@user_text);
|
|
|
|
|
==== Insert variables from a stored procedure ====
|
|
|
|
|
CREATE PROCEDURE proc()
|
|
|
|
|
BEGIN
|
|
|
|
|
# GLOBAL
|
|
|
|
|
# boolean
|
|
|
|
|
SET @@global.relay_log_purge = ON;
|
|
|
|
|
INSERT INTO tproc(truth) VALUES (@@global.relay_log_purge);
|
|
|
|
|
SET @@global.relay_log_purge = OFF;
|
|
|
|
|
INSERT INTO tproc(truth) VALUES (@@global.relay_log_purge);
|
|
|
|
|
# numeric
|
|
|
|
|
SET @@global.sync_binlog = 2000000;
|
|
|
|
|
INSERT INTO tproc(num) VALUES (@@global.sync_binlog);
|
|
|
|
|
SET @@global.sync_binlog = 3000000;
|
|
|
|
|
INSERT INTO tproc(num) VALUES (@@global.sync_binlog);
|
|
|
|
|
# string
|
2008-07-17 18:26:59 +02:00
|
|
|
SET @@global.init_slave = 'SELECT 2';
|
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
|
|
|
INSERT INTO tproc(text) VALUES (@@global.init_slave);
|
2008-07-17 18:26:59 +02:00
|
|
|
SET @@global.init_slave = 'SELECT 3';
|
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
|
|
|
INSERT INTO tproc(text) VALUES (@@global.init_slave);
|
|
|
|
|
# enumeration
|
|
|
|
|
SET @@global.slave_exec_mode = 'IDEMPOTENT';
|
|
|
|
|
INSERT INTO tproc(text) VALUES (@@global.slave_exec_mode);
|
|
|
|
|
SET @@global.slave_exec_mode = 'STRICT';
|
|
|
|
|
INSERT INTO tproc(text) VALUES (@@global.slave_exec_mode);
|
|
|
|
|
# SESSION
|
|
|
|
|
# boolean
|
|
|
|
|
SET @@sql_big_selects = ON;
|
|
|
|
|
INSERT INTO tproc(truth) VALUES (@@sql_big_selects);
|
|
|
|
|
SET @@sql_big_selects = OFF;
|
|
|
|
|
INSERT INTO tproc(truth) VALUES (@@sql_big_selects);
|
|
|
|
|
# numeric
|
|
|
|
|
SET @@last_insert_id = 20;
|
|
|
|
|
INSERT INTO tproc(num) VALUES (@@last_insert_id);
|
|
|
|
|
SET @@last_insert_id = 30;
|
|
|
|
|
INSERT INTO tproc(num) VALUES (@@last_insert_id);
|
|
|
|
|
# BOTH
|
|
|
|
|
# boolean
|
|
|
|
|
SET @@global.low_priority_updates = ON;
|
|
|
|
|
SET @@local.low_priority_updates = OFF;
|
|
|
|
|
INSERT INTO tproc(truth) VALUES (@@global.low_priority_updates);
|
|
|
|
|
INSERT INTO tproc(truth) VALUES (@@local.low_priority_updates);
|
|
|
|
|
SET @@global.low_priority_updates = OFF;
|
|
|
|
|
SET @@local.low_priority_updates = ON;
|
|
|
|
|
INSERT INTO tproc(truth) VALUES (@@global.low_priority_updates);
|
|
|
|
|
INSERT INTO tproc(truth) VALUES (@@local.low_priority_updates);
|
|
|
|
|
# numeric
|
|
|
|
|
SET @@global.default_week_format = 3;
|
|
|
|
|
SET @@local.default_week_format = 4;
|
|
|
|
|
INSERT INTO tproc(num) VALUES (@@global.default_week_format);
|
|
|
|
|
INSERT INTO tproc(num) VALUES (@@local.default_week_format);
|
|
|
|
|
SET @@global.default_week_format = 5;
|
|
|
|
|
SET @@local.default_week_format = 6;
|
|
|
|
|
INSERT INTO tproc(num) VALUES (@@global.default_week_format);
|
|
|
|
|
INSERT INTO tproc(num) VALUES (@@local.default_week_format);
|
|
|
|
|
# text
|
|
|
|
|
SET @@global.lc_time_names = 'sv_SE';
|
|
|
|
|
SET @@local.lc_time_names = 'sv_FI';
|
|
|
|
|
INSERT INTO tproc(text) VALUES (@@global.lc_time_names);
|
|
|
|
|
INSERT INTO tproc(text) VALUES (@@local.lc_time_names);
|
|
|
|
|
SET @@global.lc_time_names = 'ar_TN';
|
|
|
|
|
SET @@local.lc_time_names = 'ar_IQ';
|
|
|
|
|
INSERT INTO tproc(text) VALUES (@@global.lc_time_names);
|
|
|
|
|
INSERT INTO tproc(text) VALUES (@@local.lc_time_names);
|
|
|
|
|
# enum
|
|
|
|
|
SET @@global.sql_mode = '';
|
|
|
|
|
SET @@local.sql_mode = 'IGNORE_SPACE,NO_AUTO_CREATE_USER';
|
|
|
|
|
INSERT INTO tproc(text) VALUES (@@global.sql_mode);
|
|
|
|
|
INSERT INTO tproc(text) VALUES (@@local.sql_mode);
|
|
|
|
|
SET @@global.sql_mode = 'NO_AUTO_VALUE_ON_ZERO,NO_BACKSLASH_ESCAPES,NO_DIR_IN_CREATE,NO_ENGINE_SUBSTITUTION';
|
|
|
|
|
SET @@local.sql_mode = 'NO_FIELD_OPTIONS,NO_KEY_OPTIONS,NO_TABLE_OPTIONS';
|
|
|
|
|
INSERT INTO tproc(text) VALUES (@@global.sql_mode);
|
|
|
|
|
INSERT INTO tproc(text) VALUES (@@local.sql_mode);
|
|
|
|
|
# USER
|
|
|
|
|
# numeric
|
|
|
|
|
SET @user_num = 20;
|
|
|
|
|
INSERT INTO tproc(num) VALUES (@user_num);
|
|
|
|
|
SET @user_num = 30;
|
|
|
|
|
INSERT INTO tproc(num) VALUES (@user_num);
|
|
|
|
|
# string
|
|
|
|
|
SET @user_text = 'Bergsbrunna';
|
|
|
|
|
INSERT INTO tproc(text) VALUES (@user_text);
|
|
|
|
|
SET @user_text = 'Centrum';
|
|
|
|
|
INSERT INTO tproc(text) VALUES (@user_text);
|
|
|
|
|
END|
|
|
|
|
|
CALL proc();
|
|
|
|
|
==== Insert variables from a stored function ====
|
|
|
|
|
CREATE FUNCTION func()
|
|
|
|
|
RETURNS INT
|
|
|
|
|
BEGIN
|
|
|
|
|
# GLOBAL
|
|
|
|
|
# boolean
|
|
|
|
|
SET @@global.relay_log_purge = ON;
|
|
|
|
|
INSERT INTO tfunc(truth) VALUES (@@global.relay_log_purge);
|
|
|
|
|
SET @@global.relay_log_purge = OFF;
|
|
|
|
|
INSERT INTO tfunc(truth) VALUES (@@global.relay_log_purge);
|
|
|
|
|
# numeric
|
|
|
|
|
SET @@global.sync_binlog = 2000000;
|
|
|
|
|
INSERT INTO tfunc(num) VALUES (@@global.sync_binlog);
|
|
|
|
|
SET @@global.sync_binlog = 3000000;
|
|
|
|
|
INSERT INTO tfunc(num) VALUES (@@global.sync_binlog);
|
|
|
|
|
# string
|
2008-07-17 18:26:59 +02:00
|
|
|
SET @@global.init_slave = 'SELECT 2';
|
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
|
|
|
INSERT INTO tfunc(text) VALUES (@@global.init_slave);
|
2008-07-17 18:26:59 +02:00
|
|
|
SET @@global.init_slave = 'SELECT 3';
|
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
|
|
|
INSERT INTO tfunc(text) VALUES (@@global.init_slave);
|
|
|
|
|
# enumeration
|
|
|
|
|
SET @@global.slave_exec_mode = 'IDEMPOTENT';
|
|
|
|
|
INSERT INTO tfunc(text) VALUES (@@global.slave_exec_mode);
|
|
|
|
|
SET @@global.slave_exec_mode = 'STRICT';
|
|
|
|
|
INSERT INTO tfunc(text) VALUES (@@global.slave_exec_mode);
|
|
|
|
|
# SESSION
|
|
|
|
|
# boolean
|
|
|
|
|
SET @@sql_big_selects = ON;
|
|
|
|
|
INSERT INTO tfunc(truth) VALUES (@@sql_big_selects);
|
|
|
|
|
SET @@sql_big_selects = OFF;
|
|
|
|
|
INSERT INTO tfunc(truth) VALUES (@@sql_big_selects);
|
|
|
|
|
# numeric
|
|
|
|
|
SET @@last_insert_id = 20;
|
|
|
|
|
INSERT INTO tfunc(num) VALUES (@@last_insert_id);
|
|
|
|
|
SET @@last_insert_id = 30;
|
|
|
|
|
INSERT INTO tfunc(num) VALUES (@@last_insert_id);
|
|
|
|
|
# BOTH
|
|
|
|
|
# boolean
|
|
|
|
|
SET @@global.low_priority_updates = ON;
|
|
|
|
|
SET @@local.low_priority_updates = OFF;
|
|
|
|
|
INSERT INTO tfunc(truth) VALUES (@@global.low_priority_updates);
|
|
|
|
|
INSERT INTO tfunc(truth) VALUES (@@local.low_priority_updates);
|
|
|
|
|
SET @@global.low_priority_updates = OFF;
|
|
|
|
|
SET @@local.low_priority_updates = ON;
|
|
|
|
|
INSERT INTO tfunc(truth) VALUES (@@global.low_priority_updates);
|
|
|
|
|
INSERT INTO tfunc(truth) VALUES (@@local.low_priority_updates);
|
|
|
|
|
# numeric
|
|
|
|
|
SET @@global.default_week_format = 3;
|
|
|
|
|
SET @@local.default_week_format = 4;
|
|
|
|
|
INSERT INTO tfunc(num) VALUES (@@global.default_week_format);
|
|
|
|
|
INSERT INTO tfunc(num) VALUES (@@local.default_week_format);
|
|
|
|
|
SET @@global.default_week_format = 5;
|
|
|
|
|
SET @@local.default_week_format = 6;
|
|
|
|
|
INSERT INTO tfunc(num) VALUES (@@global.default_week_format);
|
|
|
|
|
INSERT INTO tfunc(num) VALUES (@@local.default_week_format);
|
|
|
|
|
# text
|
|
|
|
|
SET @@global.lc_time_names = 'sv_SE';
|
|
|
|
|
SET @@local.lc_time_names = 'sv_FI';
|
|
|
|
|
INSERT INTO tfunc(text) VALUES (@@global.lc_time_names);
|
|
|
|
|
INSERT INTO tfunc(text) VALUES (@@local.lc_time_names);
|
|
|
|
|
SET @@global.lc_time_names = 'ar_TN';
|
|
|
|
|
SET @@local.lc_time_names = 'ar_IQ';
|
|
|
|
|
INSERT INTO tfunc(text) VALUES (@@global.lc_time_names);
|
|
|
|
|
INSERT INTO tfunc(text) VALUES (@@local.lc_time_names);
|
|
|
|
|
# enum
|
|
|
|
|
SET @@global.sql_mode = '';
|
|
|
|
|
SET @@local.sql_mode = 'IGNORE_SPACE,NO_AUTO_CREATE_USER';
|
|
|
|
|
INSERT INTO tfunc(text) VALUES (@@global.sql_mode);
|
|
|
|
|
INSERT INTO tfunc(text) VALUES (@@local.sql_mode);
|
|
|
|
|
SET @@global.sql_mode = 'NO_AUTO_VALUE_ON_ZERO,NO_BACKSLASH_ESCAPES,NO_DIR_IN_CREATE,NO_ENGINE_SUBSTITUTION';
|
|
|
|
|
SET @@local.sql_mode = 'NO_FIELD_OPTIONS,NO_KEY_OPTIONS,NO_TABLE_OPTIONS';
|
|
|
|
|
INSERT INTO tfunc(text) VALUES (@@global.sql_mode);
|
|
|
|
|
INSERT INTO tfunc(text) VALUES (@@local.sql_mode);
|
|
|
|
|
# USER
|
|
|
|
|
# numeric
|
|
|
|
|
SET @user_num = 20;
|
|
|
|
|
INSERT INTO tfunc(num) VALUES (@user_num);
|
|
|
|
|
SET @user_num = 30;
|
|
|
|
|
INSERT INTO tfunc(num) VALUES (@user_num);
|
|
|
|
|
# string
|
|
|
|
|
SET @user_text = 'Bergsbrunna';
|
|
|
|
|
INSERT INTO tfunc(text) VALUES (@user_text);
|
|
|
|
|
SET @user_text = 'Centrum';
|
|
|
|
|
INSERT INTO tfunc(text) VALUES (@user_text);
|
|
|
|
|
RETURN 0;
|
|
|
|
|
END|
|
|
|
|
|
SELECT func();
|
|
|
|
|
func()
|
|
|
|
|
0
|
|
|
|
|
==== Insert variables from a trigger ====
|
|
|
|
|
CREATE TRIGGER trig
|
|
|
|
|
BEFORE INSERT ON trigger_table
|
|
|
|
|
FOR EACH ROW
|
|
|
|
|
BEGIN
|
|
|
|
|
# GLOBAL
|
|
|
|
|
# boolean
|
|
|
|
|
SET @@global.relay_log_purge = ON;
|
|
|
|
|
INSERT INTO ttrig(truth) VALUES (@@global.relay_log_purge);
|
|
|
|
|
SET @@global.relay_log_purge = OFF;
|
|
|
|
|
INSERT INTO ttrig(truth) VALUES (@@global.relay_log_purge);
|
|
|
|
|
# numeric
|
|
|
|
|
SET @@global.sync_binlog = 2000000;
|
|
|
|
|
INSERT INTO ttrig(num) VALUES (@@global.sync_binlog);
|
|
|
|
|
SET @@global.sync_binlog = 3000000;
|
|
|
|
|
INSERT INTO ttrig(num) VALUES (@@global.sync_binlog);
|
|
|
|
|
# string
|
2008-07-17 18:26:59 +02:00
|
|
|
SET @@global.init_slave = 'SELECT 2';
|
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
|
|
|
INSERT INTO ttrig(text) VALUES (@@global.init_slave);
|
2008-07-17 18:26:59 +02:00
|
|
|
SET @@global.init_slave = 'SELECT 3';
|
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
|
|
|
INSERT INTO ttrig(text) VALUES (@@global.init_slave);
|
|
|
|
|
# enumeration
|
|
|
|
|
SET @@global.slave_exec_mode = 'IDEMPOTENT';
|
|
|
|
|
INSERT INTO ttrig(text) VALUES (@@global.slave_exec_mode);
|
|
|
|
|
SET @@global.slave_exec_mode = 'STRICT';
|
|
|
|
|
INSERT INTO ttrig(text) VALUES (@@global.slave_exec_mode);
|
|
|
|
|
# SESSION
|
|
|
|
|
# boolean
|
|
|
|
|
SET @@sql_big_selects = ON;
|
|
|
|
|
INSERT INTO ttrig(truth) VALUES (@@sql_big_selects);
|
|
|
|
|
SET @@sql_big_selects = OFF;
|
|
|
|
|
INSERT INTO ttrig(truth) VALUES (@@sql_big_selects);
|
|
|
|
|
# numeric
|
|
|
|
|
SET @@last_insert_id = 20;
|
|
|
|
|
INSERT INTO ttrig(num) VALUES (@@last_insert_id);
|
|
|
|
|
SET @@last_insert_id = 30;
|
|
|
|
|
INSERT INTO ttrig(num) VALUES (@@last_insert_id);
|
|
|
|
|
# BOTH
|
|
|
|
|
# boolean
|
|
|
|
|
SET @@global.low_priority_updates = ON;
|
|
|
|
|
SET @@local.low_priority_updates = OFF;
|
|
|
|
|
INSERT INTO ttrig(truth) VALUES (@@global.low_priority_updates);
|
|
|
|
|
INSERT INTO ttrig(truth) VALUES (@@local.low_priority_updates);
|
|
|
|
|
SET @@global.low_priority_updates = OFF;
|
|
|
|
|
SET @@local.low_priority_updates = ON;
|
|
|
|
|
INSERT INTO ttrig(truth) VALUES (@@global.low_priority_updates);
|
|
|
|
|
INSERT INTO ttrig(truth) VALUES (@@local.low_priority_updates);
|
|
|
|
|
# numeric
|
|
|
|
|
SET @@global.default_week_format = 3;
|
|
|
|
|
SET @@local.default_week_format = 4;
|
|
|
|
|
INSERT INTO ttrig(num) VALUES (@@global.default_week_format);
|
|
|
|
|
INSERT INTO ttrig(num) VALUES (@@local.default_week_format);
|
|
|
|
|
SET @@global.default_week_format = 5;
|
|
|
|
|
SET @@local.default_week_format = 6;
|
|
|
|
|
INSERT INTO ttrig(num) VALUES (@@global.default_week_format);
|
|
|
|
|
INSERT INTO ttrig(num) VALUES (@@local.default_week_format);
|
|
|
|
|
# text
|
|
|
|
|
SET @@global.lc_time_names = 'sv_SE';
|
|
|
|
|
SET @@local.lc_time_names = 'sv_FI';
|
|
|
|
|
INSERT INTO ttrig(text) VALUES (@@global.lc_time_names);
|
|
|
|
|
INSERT INTO ttrig(text) VALUES (@@local.lc_time_names);
|
|
|
|
|
SET @@global.lc_time_names = 'ar_TN';
|
|
|
|
|
SET @@local.lc_time_names = 'ar_IQ';
|
|
|
|
|
INSERT INTO ttrig(text) VALUES (@@global.lc_time_names);
|
|
|
|
|
INSERT INTO ttrig(text) VALUES (@@local.lc_time_names);
|
|
|
|
|
# enum
|
|
|
|
|
SET @@global.sql_mode = '';
|
|
|
|
|
SET @@local.sql_mode = 'IGNORE_SPACE,NO_AUTO_CREATE_USER';
|
|
|
|
|
INSERT INTO ttrig(text) VALUES (@@global.sql_mode);
|
|
|
|
|
INSERT INTO ttrig(text) VALUES (@@local.sql_mode);
|
|
|
|
|
SET @@global.sql_mode = 'NO_AUTO_VALUE_ON_ZERO,NO_BACKSLASH_ESCAPES,NO_DIR_IN_CREATE,NO_ENGINE_SUBSTITUTION';
|
|
|
|
|
SET @@local.sql_mode = 'NO_FIELD_OPTIONS,NO_KEY_OPTIONS,NO_TABLE_OPTIONS';
|
|
|
|
|
INSERT INTO ttrig(text) VALUES (@@global.sql_mode);
|
|
|
|
|
INSERT INTO ttrig(text) VALUES (@@local.sql_mode);
|
|
|
|
|
# USER
|
|
|
|
|
# numeric
|
|
|
|
|
SET @user_num = 20;
|
|
|
|
|
INSERT INTO ttrig(num) VALUES (@user_num);
|
|
|
|
|
SET @user_num = 30;
|
|
|
|
|
INSERT INTO ttrig(num) VALUES (@user_num);
|
|
|
|
|
# string
|
|
|
|
|
SET @user_text = 'Bergsbrunna';
|
|
|
|
|
INSERT INTO ttrig(text) VALUES (@user_text);
|
|
|
|
|
SET @user_text = 'Centrum';
|
|
|
|
|
INSERT INTO ttrig(text) VALUES (@user_text);
|
|
|
|
|
END|
|
|
|
|
|
INSERT INTO trigger_table VALUES ('bye.');
|
|
|
|
|
==== Insert variables from a prepared statement ====
|
|
|
|
|
PREPARE p1 FROM 'SET @@global.relay_log_purge = ON';
|
|
|
|
|
PREPARE p2 FROM 'INSERT INTO tprep(truth) VALUES (@@global.relay_log_purge)';
|
|
|
|
|
PREPARE p3 FROM 'SET @@global.relay_log_purge = OFF';
|
|
|
|
|
PREPARE p4 FROM 'INSERT INTO tprep(truth) VALUES (@@global.relay_log_purge)';
|
|
|
|
|
PREPARE p5 FROM 'SET @@global.sync_binlog = 2000000';
|
|
|
|
|
PREPARE p6 FROM 'INSERT INTO tprep(num) VALUES (@@global.sync_binlog)';
|
|
|
|
|
PREPARE p7 FROM 'SET @@global.sync_binlog = 3000000';
|
|
|
|
|
PREPARE p8 FROM 'INSERT INTO tprep(num) VALUES (@@global.sync_binlog)';
|
2008-07-17 18:26:59 +02:00
|
|
|
PREPARE p9 FROM 'SET @@global.init_slave = \'SELECT 2\'';
|
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
|
|
|
PREPARE p10 FROM 'INSERT INTO tprep(text) VALUES (@@global.init_slave)';
|
2008-07-17 18:26:59 +02:00
|
|
|
PREPARE p11 FROM 'SET @@global.init_slave = \'SELECT 3\'';
|
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
|
|
|
PREPARE p12 FROM 'INSERT INTO tprep(text) VALUES (@@global.init_slave)';
|
|
|
|
|
PREPARE p13 FROM 'SET @@global.slave_exec_mode = \'IDEMPOTENT\'';
|
|
|
|
|
PREPARE p14 FROM 'INSERT INTO tprep(text) VALUES (@@global.slave_exec_mode)';
|
|
|
|
|
PREPARE p15 FROM 'SET @@global.slave_exec_mode = \'STRICT\'';
|
|
|
|
|
PREPARE p16 FROM 'INSERT INTO tprep(text) VALUES (@@global.slave_exec_mode)';
|
|
|
|
|
PREPARE p17 FROM 'SET @@sql_big_selects = ON';
|
|
|
|
|
PREPARE p18 FROM 'INSERT INTO tprep(truth) VALUES (@@sql_big_selects)';
|
|
|
|
|
PREPARE p19 FROM 'SET @@sql_big_selects = OFF';
|
|
|
|
|
PREPARE p20 FROM 'INSERT INTO tprep(truth) VALUES (@@sql_big_selects)';
|
|
|
|
|
PREPARE p21 FROM 'SET @@last_insert_id = 20';
|
|
|
|
|
PREPARE p22 FROM 'INSERT INTO tprep(num) VALUES (@@last_insert_id)';
|
|
|
|
|
PREPARE p23 FROM 'SET @@last_insert_id = 30';
|
|
|
|
|
PREPARE p24 FROM 'INSERT INTO tprep(num) VALUES (@@last_insert_id)';
|
|
|
|
|
PREPARE p25 FROM 'SET @@global.low_priority_updates = ON';
|
|
|
|
|
PREPARE p26 FROM 'SET @@local.low_priority_updates = OFF';
|
|
|
|
|
PREPARE p27 FROM 'INSERT INTO tprep(truth) VALUES (@@global.low_priority_updates)';
|
|
|
|
|
PREPARE p28 FROM 'INSERT INTO tprep(truth) VALUES (@@local.low_priority_updates)';
|
|
|
|
|
PREPARE p29 FROM 'SET @@global.low_priority_updates = OFF';
|
|
|
|
|
PREPARE p30 FROM 'SET @@local.low_priority_updates = ON';
|
|
|
|
|
PREPARE p31 FROM 'INSERT INTO tprep(truth) VALUES (@@global.low_priority_updates)';
|
|
|
|
|
PREPARE p32 FROM 'INSERT INTO tprep(truth) VALUES (@@local.low_priority_updates)';
|
|
|
|
|
PREPARE p33 FROM 'SET @@global.default_week_format = 3';
|
|
|
|
|
PREPARE p34 FROM 'SET @@local.default_week_format = 4';
|
|
|
|
|
PREPARE p35 FROM 'INSERT INTO tprep(num) VALUES (@@global.default_week_format)';
|
|
|
|
|
PREPARE p36 FROM 'INSERT INTO tprep(num) VALUES (@@local.default_week_format)';
|
|
|
|
|
PREPARE p37 FROM 'SET @@global.default_week_format = 5';
|
|
|
|
|
PREPARE p38 FROM 'SET @@local.default_week_format = 6';
|
|
|
|
|
PREPARE p39 FROM 'INSERT INTO tprep(num) VALUES (@@global.default_week_format)';
|
|
|
|
|
PREPARE p40 FROM 'INSERT INTO tprep(num) VALUES (@@local.default_week_format)';
|
|
|
|
|
PREPARE p41 FROM 'SET @@global.lc_time_names = \'sv_SE\'';
|
|
|
|
|
PREPARE p42 FROM 'SET @@local.lc_time_names = \'sv_FI\'';
|
|
|
|
|
PREPARE p43 FROM 'INSERT INTO tprep(text) VALUES (@@global.lc_time_names)';
|
|
|
|
|
PREPARE p44 FROM 'INSERT INTO tprep(text) VALUES (@@local.lc_time_names)';
|
|
|
|
|
PREPARE p45 FROM 'SET @@global.lc_time_names = \'ar_TN\'';
|
|
|
|
|
PREPARE p46 FROM 'SET @@local.lc_time_names = \'ar_IQ\'';
|
|
|
|
|
PREPARE p47 FROM 'INSERT INTO tprep(text) VALUES (@@global.lc_time_names)';
|
|
|
|
|
PREPARE p48 FROM 'INSERT INTO tprep(text) VALUES (@@local.lc_time_names)';
|
|
|
|
|
PREPARE p49 FROM 'SET @@global.sql_mode = \'\'';
|
|
|
|
|
PREPARE p50 FROM 'SET @@local.sql_mode = \'IGNORE_SPACE,NO_AUTO_CREATE_USER\'';
|
|
|
|
|
PREPARE p51 FROM 'INSERT INTO tprep(text) VALUES (@@global.sql_mode)';
|
|
|
|
|
PREPARE p52 FROM 'INSERT INTO tprep(text) VALUES (@@local.sql_mode)';
|
|
|
|
|
PREPARE p53 FROM 'SET @@global.sql_mode = \'NO_AUTO_VALUE_ON_ZERO,NO_BACKSLASH_ESCAPES,NO_DIR_IN_CREATE,NO_ENGINE_SUBSTITUTION\'';
|
|
|
|
|
PREPARE p54 FROM 'SET @@local.sql_mode = \'NO_FIELD_OPTIONS,NO_KEY_OPTIONS,NO_TABLE_OPTIONS\'';
|
|
|
|
|
PREPARE p55 FROM 'INSERT INTO tprep(text) VALUES (@@global.sql_mode)';
|
|
|
|
|
PREPARE p56 FROM 'INSERT INTO tprep(text) VALUES (@@local.sql_mode)';
|
|
|
|
|
PREPARE p57 FROM 'SET @user_num = 20';
|
|
|
|
|
PREPARE p58 FROM 'INSERT INTO tprep(num) VALUES (@user_num)';
|
|
|
|
|
PREPARE p59 FROM 'SET @user_num = 30';
|
|
|
|
|
PREPARE p60 FROM 'INSERT INTO tprep(num) VALUES (@user_num)';
|
|
|
|
|
PREPARE p61 FROM 'SET @user_text = \'Bergsbrunna\'';
|
|
|
|
|
PREPARE p62 FROM 'INSERT INTO tprep(text) VALUES (@user_text)';
|
|
|
|
|
PREPARE p63 FROM 'SET @user_text = \'Centrum\'';
|
|
|
|
|
PREPARE p64 FROM 'INSERT INTO tprep(text) VALUES (@user_text)';
|
|
|
|
|
EXECUTE p1;
|
|
|
|
|
EXECUTE p2;
|
|
|
|
|
EXECUTE p3;
|
|
|
|
|
EXECUTE p4;
|
|
|
|
|
EXECUTE p5;
|
|
|
|
|
EXECUTE p6;
|
|
|
|
|
EXECUTE p7;
|
|
|
|
|
EXECUTE p8;
|
|
|
|
|
EXECUTE p9;
|
|
|
|
|
EXECUTE p10;
|
|
|
|
|
EXECUTE p11;
|
|
|
|
|
EXECUTE p12;
|
|
|
|
|
EXECUTE p13;
|
|
|
|
|
EXECUTE p14;
|
|
|
|
|
EXECUTE p15;
|
|
|
|
|
EXECUTE p16;
|
|
|
|
|
EXECUTE p17;
|
|
|
|
|
EXECUTE p18;
|
|
|
|
|
EXECUTE p19;
|
|
|
|
|
EXECUTE p20;
|
|
|
|
|
EXECUTE p21;
|
|
|
|
|
EXECUTE p22;
|
|
|
|
|
EXECUTE p23;
|
|
|
|
|
EXECUTE p24;
|
|
|
|
|
EXECUTE p25;
|
|
|
|
|
EXECUTE p26;
|
|
|
|
|
EXECUTE p27;
|
|
|
|
|
EXECUTE p28;
|
|
|
|
|
EXECUTE p29;
|
|
|
|
|
EXECUTE p30;
|
|
|
|
|
EXECUTE p31;
|
|
|
|
|
EXECUTE p32;
|
|
|
|
|
EXECUTE p33;
|
|
|
|
|
EXECUTE p34;
|
|
|
|
|
EXECUTE p35;
|
|
|
|
|
EXECUTE p36;
|
|
|
|
|
EXECUTE p37;
|
|
|
|
|
EXECUTE p38;
|
|
|
|
|
EXECUTE p39;
|
|
|
|
|
EXECUTE p40;
|
|
|
|
|
EXECUTE p41;
|
|
|
|
|
EXECUTE p42;
|
|
|
|
|
EXECUTE p43;
|
|
|
|
|
EXECUTE p44;
|
|
|
|
|
EXECUTE p45;
|
|
|
|
|
EXECUTE p46;
|
|
|
|
|
EXECUTE p47;
|
|
|
|
|
EXECUTE p48;
|
|
|
|
|
EXECUTE p49;
|
|
|
|
|
EXECUTE p50;
|
|
|
|
|
EXECUTE p51;
|
|
|
|
|
EXECUTE p52;
|
|
|
|
|
EXECUTE p53;
|
|
|
|
|
EXECUTE p54;
|
|
|
|
|
EXECUTE p55;
|
|
|
|
|
EXECUTE p56;
|
|
|
|
|
EXECUTE p57;
|
|
|
|
|
EXECUTE p58;
|
|
|
|
|
EXECUTE p59;
|
|
|
|
|
EXECUTE p60;
|
|
|
|
|
EXECUTE p61;
|
|
|
|
|
EXECUTE p62;
|
|
|
|
|
EXECUTE p63;
|
|
|
|
|
EXECUTE p64;
|
|
|
|
|
==== Results ====
|
|
|
|
|
SELECT * FROM tstmt ORDER BY id;
|
|
|
|
|
id truth num text
|
|
|
|
|
1 1 NULL NULL
|
|
|
|
|
2 0 NULL NULL
|
|
|
|
|
3 NULL 2000000 NULL
|
|
|
|
|
4 NULL 3000000 NULL
|
2008-07-17 18:26:59 +02:00
|
|
|
5 NULL NULL SELECT 2
|
|
|
|
|
6 NULL NULL SELECT 3
|
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
|
|
|
7 NULL NULL IDEMPOTENT
|
|
|
|
|
8 NULL NULL STRICT
|
|
|
|
|
9 1 NULL NULL
|
|
|
|
|
10 0 NULL NULL
|
|
|
|
|
11 NULL 20 NULL
|
|
|
|
|
12 NULL 30 NULL
|
|
|
|
|
13 1 NULL NULL
|
|
|
|
|
14 0 NULL NULL
|
|
|
|
|
15 0 NULL NULL
|
|
|
|
|
16 1 NULL NULL
|
|
|
|
|
17 NULL 3 NULL
|
|
|
|
|
18 NULL 4 NULL
|
|
|
|
|
19 NULL 5 NULL
|
|
|
|
|
20 NULL 6 NULL
|
|
|
|
|
21 NULL NULL sv_SE
|
|
|
|
|
22 NULL NULL sv_FI
|
|
|
|
|
23 NULL NULL ar_TN
|
|
|
|
|
24 NULL NULL ar_IQ
|
|
|
|
|
25 NULL NULL
|
|
|
|
|
26 NULL NULL IGNORE_SPACE,NO_AUTO_CREATE_USER
|
|
|
|
|
27 NULL NULL NO_DIR_IN_CREATE,NO_AUTO_VALUE_ON_ZERO,NO_BACKSLASH_ESCAPES,NO_ENGINE_SUBSTITUTION
|
|
|
|
|
28 NULL NULL NO_KEY_OPTIONS,NO_TABLE_OPTIONS,NO_FIELD_OPTIONS
|
|
|
|
|
29 NULL 20 NULL
|
|
|
|
|
30 NULL 30 NULL
|
|
|
|
|
31 NULL NULL Bergsbrunna
|
|
|
|
|
32 NULL NULL Centrum
|
|
|
|
|
Comparing tables master:test.tstmt and master:test.tproc
|
|
|
|
|
Comparing tables master:test.tstmt and master:test.tfunc
|
|
|
|
|
Comparing tables master:test.tstmt and master:test.ttrig
|
|
|
|
|
Comparing tables master:test.tstmt and master:test.tprep
|
|
|
|
|
Comparing tables master:test.tstmt and slave:test.tstmt
|
|
|
|
|
Comparing tables master:test.tstmt and slave:test.tproc
|
|
|
|
|
Comparing tables master:test.tstmt and slave:test.tfunc
|
|
|
|
|
Comparing tables master:test.tstmt and slave:test.ttrig
|
|
|
|
|
Comparing tables master:test.tstmt and slave:test.tprep
|
|
|
|
|
==== Clean up ====
|
|
|
|
|
[on master]
|
|
|
|
|
DROP PROCEDURE proc;
|
|
|
|
|
DROP FUNCTION func;
|
|
|
|
|
DROP TRIGGER trig;
|
|
|
|
|
DROP TABLE tstmt, tproc, tfunc, ttrig, tprep, trigger_table;
|
|
|
|
|
SET @@global.default_week_format= @m_default_week_format;
|
|
|
|
|
SET @@global.init_slave= @m_init_slave;
|
|
|
|
|
SET @@global.lc_time_names= @m_lc_time_names;
|
|
|
|
|
SET @@global.low_priority_updates= @m_low_priority_updates;
|
|
|
|
|
SET @@global.relay_log_purge= @m_relay_log_purge;
|
|
|
|
|
SET @@global.slave_exec_mode= @m_slave_exec_mode;
|
|
|
|
|
SET @@global.sql_mode= @m_sql_mode;
|
|
|
|
|
SET @@global.sync_binlog= @m_sync_binlog;
|
|
|
|
|
[on slave]
|
|
|
|
|
SET @@global.default_week_format= @s_default_week_format;
|
|
|
|
|
SET @@global.init_slave= @s_init_slave;
|
|
|
|
|
SET @@global.lc_time_names= @s_lc_time_names;
|
|
|
|
|
SET @@global.low_priority_updates= @s_low_priority_updates;
|
|
|
|
|
SET @@global.relay_log_purge= @s_relay_log_purge;
|
|
|
|
|
SET @@global.slave_exec_mode= @s_slave_exec_mode;
|
|
|
|
|
SET @@global.sql_mode= @s_sql_mode;
|
|
|
|
|
SET @@global.sync_binlog= @s_sync_binlog;
|