Fix fur BUG#13348: "multi-table updates and deletes are not logged if no rows were affected".
Not fixed in 4.1 as not critical. Also I'm correcting error checking of multi-UPDATE/DELETE when it comes to binlogging, to make it consistent with when we rollback the statement. mysql-test/r/rpl_multi_delete.result: result update mysql-test/r/rpl_multi_update.result: result update mysql-test/t/rpl_multi_delete.test: test for BUG#13348 mysql-test/t/rpl_multi_update.test: test of BUG#13348 sql/sql_delete.cc: We now binlog multi-DELETE even if no row was updated (like we do for DELETE). I'm also correcting some error checking (< instead of <=), basing myself on when we rollback. sql/sql_update.cc: we now binlog multi-UPDATE even if no row was updated (like we do for UPDATE). Adding to existing tests to test new behaviour. I'm also correcting some error checking (< instead of <=), basing myself on when we rollback.
This commit is contained in:
unknown
parent
2b24a2b5b5
commit
03045ee70a
@ -19,4 +19,13 @@ a
|
||||
select * from t2;
|
||||
a
|
||||
1
|
||||
delete from t1;
|
||||
delete from t2;
|
||||
insert into t1 values(1);
|
||||
insert into t2 values(1);
|
||||
DELETE t1.*, t2.* from t1, t2;
|
||||
select * from t1;
|
||||
a
|
||||
select * from t2;
|
||||
a
|
||||
drop table t1,t2;
|
||||
|
||||
@ -24,3 +24,16 @@ a b
|
||||
1 0
|
||||
2 1
|
||||
UPDATE t1, t2 SET t1.b = t2.b WHERE t1.a = t2.a;
|
||||
delete from t1;
|
||||
delete from t2;
|
||||
insert into t1 values(1,1);
|
||||
insert into t2 values(1,1);
|
||||
update t1 set a=2;
|
||||
UPDATE t1, t2 SET t1.a = t2.a;
|
||||
select * from t1;
|
||||
a b
|
||||
1 1
|
||||
select * from t2;
|
||||
a b
|
||||
1 1
|
||||
drop table t1, t2;
|
||||
|
||||
@ -16,10 +16,26 @@ sync_with_master;
|
||||
select * from t1;
|
||||
select * from t2;
|
||||
|
||||
# End of 4.1 tests
|
||||
|
||||
# Check if deleting 0 rows is binlogged (BUG#13348)
|
||||
|
||||
connection master;
|
||||
delete from t1;
|
||||
delete from t2;
|
||||
|
||||
connection slave;
|
||||
# force a difference to see if master's multi-DELETE will correct it
|
||||
insert into t1 values(1);
|
||||
insert into t2 values(1);
|
||||
|
||||
connection master;
|
||||
DELETE t1.*, t2.* from t1, t2;
|
||||
|
||||
sync_slave_with_master;
|
||||
select * from t1;
|
||||
select * from t2;
|
||||
|
||||
connection master;
|
||||
drop table t1,t2;
|
||||
save_master_pos;
|
||||
connection slave;
|
||||
sync_with_master;
|
||||
|
||||
# End of 4.1 tests
|
||||
sync_slave_with_master;
|
||||
|
||||
@ -24,3 +24,26 @@ connection slave;
|
||||
sync_with_master;
|
||||
|
||||
# End of 4.1 tests
|
||||
|
||||
# Check if updating 0 rows is binlogged (BUG#13348)
|
||||
|
||||
connection master;
|
||||
delete from t1;
|
||||
delete from t2;
|
||||
insert into t1 values(1,1);
|
||||
insert into t2 values(1,1);
|
||||
|
||||
connection slave;
|
||||
# force a difference to see if master's multi-UPDATE will correct it
|
||||
update t1 set a=2;
|
||||
|
||||
connection master;
|
||||
UPDATE t1, t2 SET t1.a = t2.a;
|
||||
|
||||
sync_slave_with_master;
|
||||
select * from t1;
|
||||
select * from t2;
|
||||
|
||||
connection master;
|
||||
drop table t1, t2;
|
||||
sync_slave_with_master;
|
||||
|
||||
@ -258,19 +258,12 @@ cleanup:
|
||||
|
||||
delete select;
|
||||
transactional_table= table->file->has_transactions();
|
||||
/*
|
||||
We write to the binary log even if we deleted no row, because maybe the
|
||||
user is using this command to ensure that a table is clean on master *and
|
||||
on slave*. Think of the case of a user having played separately with the
|
||||
master's table and slave's table and wanting to take a fresh identical
|
||||
start now.
|
||||
error < 0 means "really no error". error <= 0 means "maybe some error".
|
||||
*/
|
||||
if ((deleted || (error < 0)) && (error <= 0 || !transactional_table))
|
||||
/* See similar binlogging code in sql_update.cc, for comments */
|
||||
if ((error < 0) || (deleted && !transactional_table))
|
||||
{
|
||||
if (mysql_bin_log.is_open())
|
||||
{
|
||||
if (error <= 0)
|
||||
if (error < 0)
|
||||
thd->clear_error();
|
||||
Query_log_event qinfo(thd, thd->query, thd->query_length,
|
||||
transactional_table, FALSE);
|
||||
@ -718,6 +711,9 @@ bool multi_delete::send_eof()
|
||||
/* Does deletes for the last n - 1 tables, returns 0 if ok */
|
||||
int local_error= do_deletes(); // returns 0 if success
|
||||
|
||||
/* compute a total error to know if something failed */
|
||||
local_error= local_error || error;
|
||||
|
||||
/* reset used flags */
|
||||
thd->proc_info="end";
|
||||
|
||||
@ -730,19 +726,11 @@ bool multi_delete::send_eof()
|
||||
query_cache_invalidate3(thd, delete_tables, 1);
|
||||
}
|
||||
|
||||
/*
|
||||
Write the SQL statement to the binlog if we deleted
|
||||
rows and we succeeded, or also in an error case when there
|
||||
was a non-transaction-safe table involved, since
|
||||
modifications in it cannot be rolled back.
|
||||
Note that if we deleted nothing we don't write to the binlog (TODO:
|
||||
fix this).
|
||||
*/
|
||||
if (deleted && ((error <= 0 && !local_error) || normal_tables))
|
||||
if ((local_error == 0) || (deleted && normal_tables))
|
||||
{
|
||||
if (mysql_bin_log.is_open())
|
||||
{
|
||||
if (error <= 0 && !local_error)
|
||||
if (local_error == 0)
|
||||
thd->clear_error();
|
||||
Query_log_event qinfo(thd, thd->query, thd->query_length,
|
||||
transactional_tables, FALSE);
|
||||
|
||||
@ -475,11 +475,20 @@ int mysql_update(THD *thd,
|
||||
query_cache_invalidate3(thd, table_list, 1);
|
||||
}
|
||||
|
||||
if ((updated || (error < 0)) && (error <= 0 || !transactional_table))
|
||||
/*
|
||||
error < 0 means really no error at all: we processed all rows until the
|
||||
last one without error. error > 0 means an error (e.g. unique key
|
||||
violation and no IGNORE or REPLACE). error == 0 is also an error (if
|
||||
preparing the record or invoking before triggers fails). See
|
||||
ha_autocommit_or_rollback(error>=0) and DBUG_RETURN(error>=0) below.
|
||||
Sometimes we want to binlog even if we updated no rows, in case user used
|
||||
it to be sure master and slave are in same state.
|
||||
*/
|
||||
if ((error < 0) || (updated && !transactional_table))
|
||||
{
|
||||
if (mysql_bin_log.is_open())
|
||||
{
|
||||
if (error <= 0)
|
||||
if (error < 0)
|
||||
thd->clear_error();
|
||||
Query_log_event qinfo(thd, thd->query, thd->query_length,
|
||||
transactional_table, FALSE);
|
||||
@ -1439,16 +1448,14 @@ bool multi_update::send_eof()
|
||||
/*
|
||||
Write the SQL statement to the binlog if we updated
|
||||
rows and we succeeded or if we updated some non
|
||||
transacational tables.
|
||||
Note that if we updated nothing we don't write to the binlog (TODO:
|
||||
fix this).
|
||||
transactional tables.
|
||||
*/
|
||||
|
||||
if (updated && (local_error <= 0 || !trans_safe))
|
||||
if ((local_error == 0) || (updated && !trans_safe))
|
||||
{
|
||||
if (mysql_bin_log.is_open())
|
||||
{
|
||||
if (local_error <= 0)
|
||||
if (local_error == 0)
|
||||
thd->clear_error();
|
||||
Query_log_event qinfo(thd, thd->query, thd->query_length,
|
||||
transactional_tables, FALSE);
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user