mariadb/mysql-test/suite/rpl/t/rpl_row_find_row.test
Luis Soares 1b27674429 BUG#53893: RBR: nullable unique key can lead to out-of-sync slave
When using Unique Keys with nullable parts in RBR, the slave can
choose the wrong row to update. This happens because a table with
an unique key containing nullable parts cannot strictly guarantee 
uniqueness. As stated in the manual, for all engines, a UNIQUE 
index allows multiple NULL values for columns that can contain 
NULL.

We fix this at the slave by extending the checks before assuming
that the row found through an unique index is is the correct
one. This means that when a record (R) is fetched from the storage
engine and a key that is not primary (K) is used, the server does 
the following: 

 - If K is unique and has no nullable parts, it returns R;
 - Otherwise, if any field in the before image that is part of K
   is null do an index scan;
 - If there is no NULL field in the BI part of K, then return R.

A side change: renamed the existing test case file and added a
test case covering the changes in this patch.
2010-06-02 23:26:12 +01:00

104 lines
2.3 KiB
Text

# BUG#47312: RBR: Disabling key on slave breaks replication:
# HA_ERR_WRONG_INDEX
#
# Description
# ===========
#
# This test case checks whether disabling a key on a slave breaks
# replication or not.
#
# Case #1, shows that while not using ALTER TABLE... DISABLE KEYS and
# the slave has no key defined while the master has one, replication
# won't break.
#
# Case #2, shows that before patch for BUG#47312, if defining key on
# slave table, and later disable it, replication would break. This
# has been fixed.
#
-- source include/master-slave.inc
-- source include/have_binlog_format_row.inc
#
# Case #1: master has key, but slave has not.
# Replication does not break.
#
SET SQL_LOG_BIN=0;
CREATE TABLE t (a int, b int, c int, key(b));
SET SQL_LOG_BIN=1;
-- connection slave
CREATE TABLE t (a int, b int, c int);
-- connection master
INSERT INTO t VALUES (1,2,4);
INSERT INTO t VALUES (4,3,4);
DELETE FROM t;
-- sync_slave_with_master
-- connection master
DROP TABLE t;
-- sync_slave_with_master
#
# Case #2: master has key, slave also has one,
# but it gets disabled sometime.
# Replication does not break anymore.
#
-- source include/master-slave-reset.inc
-- connection master
CREATE TABLE t (a int, b int, c int, key(b));
-- sync_slave_with_master
ALTER TABLE t DISABLE KEYS;
-- connection master
INSERT INTO t VALUES (1,2,4);
INSERT INTO t VALUES (4,3,4);
DELETE FROM t;
-- sync_slave_with_master
-- connection master
DROP TABLE t;
-- sync_slave_with_master
#
# BUG#53893: RBR: nullable unique key can lead to out-of-sync slave
#
#
# We insert two rows. Both with part of UNIQUE KEY set to null.
# Then we update the last row inserted. On master the correct
# row is updated. On the slave the wrong row would be updated
# because the engine would look it up by the NULL Unique KEY.
# As a consquence, the wrong row would be updated.
#
-- connection master
-- source include/master-slave-reset.inc
-- connection master
CREATE TABLE t1 (c1 INT NOT NULL, c2 INT NOT NULL, c3 INT, UNIQUE KEY(c1,c3), KEY(c2));
INSERT INTO t1(c1,c2) VALUES(1,1);
INSERT INTO t1(c1,c2) VALUES(1,2);
UPDATE t1 SET c1=1000 WHERE c2=2;
-- sync_slave_with_master
-- let $diff_table_1=master:test.t1
-- let $diff_table_2=slave:test.t1
-- source include/diff_tables.inc
-- connection master
DROP TABLE t1;
-- sync_slave_with_master