mariadb/mysql-test/suite/galera/t/galera_replica_no_gtid.test
Kristian Nielsen 0a68328673 MDEV-34705: Binlog-in-engine: Protect against concurrent RESET MASTER and dump threads
This is actually an existing problem in the old binlog implementation, and
this patch is applicable to old binlog also. The problem is that RESET
MASTER can run concurrently with binlog dump threads / connected slaves.
This will remove the binlog from under the feet of the reader, which can
cause all sorts of strange behaviour.

This patch fixes the problem by disallowing to run RESET MASTER when dump
threads (or other RESET MASTER or SHOW BINARY LOGS) are running. An error is
thrown in this case, user must stop slaves and/or kill dump threads to make
the RESET MASTER go through. A slave that connects in the middle of RESET
MASTER will wait for it to complete.

Fix a lot of test cases to kill any lingering dump threads before doing
RESET MASTER, mostly just by sourcing include/kill_binlog_dump_threads.inc.

Signed-off-by: Kristian Nielsen <knielsen@knielsen-hq.org>
2025-06-11 11:32:10 +02:00

125 lines
4 KiB
Text

#
# Test Galera as a replica to a MySQL async replication
#
# The galera/galera_2node_slave.cnf describes the setup of the nodes
#
--source include/force_restart.inc
--source include/galera_cluster.inc
--source include/have_innodb.inc
--source include/have_sequence.inc
# As node #3 is not a Galera node, and galera_cluster.inc does not open connetion to it
# we open the node_3 connection here
--connect node_3, 127.0.0.1, root, , test, $NODE_MYPORT_3
create user repl@'%' identified by 'repl';
grant all on *.* to repl@'%';
flush privileges;
ALTER TABLE mysql.gtid_slave_pos ENGINE=InnoDB;
--let $node_1 = node_1
--let $node_2 = node_2
--source include/auto_increment_offset_save.inc
--connection node_2
--disable_query_log
--eval CHANGE MASTER TO master_host='127.0.0.1', master_user='repl', master_password='repl', master_ssl_verify_server_cert=0, master_port=$NODE_MYPORT_3, master_use_gtid=slave_pos;
--enable_query_log
START SLAVE;
--connection node_3
CREATE TABLE t1 (id bigint primary key, msg varchar(100)) engine=innodb;
--disable_query_log
INSERT INTO t1 SELECT seq, 'test' from seq_1_to_10000;
--enable_query_log
SELECT COUNT(*) AS EXPECT_10000 FROM t1;
--connection node_2
--let $wait_condition = SELECT COUNT(*) = 1 FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_NAME = 't1';
--source include/wait_condition.inc
--let $wait_condition = SELECT COUNT(*) = 10000 FROM t1;
--source include/wait_condition.inc
#
# Node_2 is slave so mysql.gtid_slave_pos table is also replicated
#
SELECT COUNT(*) > 0 AS EXPECT_1 FROM mysql.gtid_slave_pos;
SELECT COUNT(*) AS EXPECT_10000 FROM t1;
--connection node_1
--let $wait_condition = SELECT COUNT(*) = 1 FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_NAME = 't1';
--source include/wait_condition.inc
--let $wait_condition = SELECT COUNT(*) = 10000 FROM t1;
--source include/wait_condition.inc
#
# mysql-gtid_slave_pos table should not be replicated by Galera
#
SELECT COUNT(*) AS EXPECT_0 FROM mysql.gtid_slave_pos;
SELECT COUNT(*) AS EXPECT_10000 FROM t1;
--connection node_2
--echo # Verify that graceful shutdown succeeds.
--source include/shutdown_mysqld.inc
--echo # Force SST
--remove_file $MYSQLTEST_VARDIR/mysqld.2/data/grastate.dat
--connection node_1
--echo # Waiting until node_2 is not part of cluster anymore
--let $wait_condition = SELECT VARIABLE_VALUE = 1 FROM INFORMATION_SCHEMA.GLOBAL_STATUS WHERE VARIABLE_NAME = 'wsrep_cluster_size';
--source include/wait_condition.inc
--let $wait_condition = SELECT VARIABLE_VALUE = 'Primary' FROM INFORMATION_SCHEMA.GLOBAL_STATUS WHERE VARIABLE_NAME = 'wsrep_cluster_status';
--source include/wait_condition.inc
--connection node_2
--echo # Start node_2 again
--source include/start_mysqld.inc
--echo ¤ Wait until node_2 is back on cluster
--let $wait_condition = SELECT VARIABLE_VALUE = 'Primary' FROM INFORMATION_SCHEMA.GLOBAL_STATUS WHERE VARIABLE_NAME = 'wsrep_cluster_status';
--source include/wait_condition.inc
--let $wait_condition = SELECT VARIABLE_VALUE = 2 FROM INFORMATION_SCHEMA.GLOBAL_STATUS WHERE VARIABLE_NAME = 'wsrep_cluster_size';
--source include/wait_condition.inc
--let $wait_condition = SELECT VARIABLE_VALUE = 'ON' FROM INFORMATION_SCHEMA.GLOBAL_STATUS WHERE VARIABLE_NAME = 'wsrep_ready';
--source include/wait_condition.inc
--connection node_2
call mtr.add_suppression("Slave: Operation CREATE USER failed for ");
SELECT COUNT(*) AS EXPECT_0 FROM mysql.gtid_slave_pos;
SELECT COUNT(*) AS EXPECT_10000 FROM t1;
--connection node_1
SELECT COUNT(*) AS EXPECT_0 FROM mysql.gtid_slave_pos;
SELECT COUNT(*) AS EXPECT_10000 FROM t1;
--connection node_3
SELECT COUNT(*) AS EXPECT_10000 FROM t1;
#
# Cleanup
#
--connection node_2
STOP SLAVE;
RESET SLAVE ALL;
--connection node_3
--source include/kill_binlog_dump_threads.inc
RESET MASTER;
drop table t1;
--connection node_2
DROP TABLE t1;
--connection node_1
--let $wait_condition = SELECT COUNT(*) = 0 FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_NAME = 't1';
--source include/wait_condition.inc
--connection node_1
--disconnect node_3
--source include/auto_increment_offset_restore.inc
--source include/galera_end.inc
--echo # End of test