1
0
mirror of https://github.com/MariaDB/server.git synced 2025-08-08 11:22:35 +03:00
Files
mariadb/mysql-test/suite/binlog/t/binlog_recover_async_rollback_trx.test
Libing Song 5bbda97111 MDEV-33853 Async rollback prepared transactions during binlog
crash recovery

Summary
=======
When doing server recovery, the active transactions will be rolled
back by InnoDB background rollback thread automatically. The
prepared transactions will be committed or rolled back accordingly
by binlog recovery. Binlog recovery is done in main thread before
the server can provide service to users. If there is a big
transaction to rollback, the server will not available for a long
time.

This patch provides a way to rollback the prepared transactions
asynchronously. Thus the rollback will not block server startup.

Design
======
- Handler::recover_rollback_by_xid()
  This patch provides a new handler interface to rollback transactions
  in recover phase. InnoDB just set the transaction's state to active.
  Then the transaction will be rolled back by the background rollback
  thread.

- Handler::signal_tc_log_recover_done()
  This function is called after tc log is opened(typically binlog opened)
  has done. When this function is called, all transactions will be rolled
  back have been reverted to ACTIVE state. Thus it starts rollback thread
  to rollback the transactions.

- Background rollback thread
  With this patch, background rollback thread is defered to run until binlog
  recovery is finished. It is started by innobase_tc_log_recovery_done().
2024-09-05 21:19:25 +03:00

118 lines
4.2 KiB
Plaintext

################################################################################
# Async rollback prepared transactions during binlog crash recovery
#
# It verifies that binlog recovery just set the prepared transactions to
# active and the background recovery rollback thread will rollback the
# transactions asynchronously.
################################################################################
--source include/have_debug.inc
--source include/have_debug_sync.inc
--source include/have_innodb.inc
--source include/have_binlog_format_row.inc
--source include/have_sequence.inc
CREATE TABLE t1 (id int primary key, data int) ENGINE = InnoDB;
INSERT INTO t1 VALUES (0, 1);
--echo #
--echo # 1. Check DML in prepared state can rollback correctly.
--echo #
--connect(con1, localhost, root,,)
SET debug_sync = "ha_commit_trans_after_prepare SIGNAL prepared1 WAIT_FOR continue";
--send INSERT INTO t1 VALUES(1, 1);
--connect(con2, localhost, root,,)
SET debug_sync = "now WAIT_FOR prepared1";
SET debug_sync = "ha_commit_trans_after_prepare SIGNAL prepared2 WAIT_FOR continue";
--send UPDATE t1 SET data = data + 1 WHERE id = 0
--connection default
SET debug_sync = "now WAIT_FOR prepared2";
--source include/kill_mysqld.inc
--disconnect con1
--disconnect con2
# With the debug option, recovery rollback thread just rolls back the
# first prepared transaction and then goes to sleep.
--source include/start_mysqld.inc
--let $wait_condition= SELECT count(*) = 0 FROM information_schema.innodb_trx
--source include/wait_condition.inc
--echo # Expect (0, 1)
SET SESSION TRANSACTION ISOLATION LEVEL READ UNCOMMITTED;
SELECT * FROM t1;
# The previous INSERT is rolled back, so this INSERT will succeed.
INSERT INTO t1 VALUES(1, 1);
# The previous UPDATE is rolled back, so this UPDATE will succeed.
UPDATE t1 SET data = data + 1 WHERE id = 0;
--echo # Expect (0, 2), (1, 1)
SELECT * FROM t1;
--echo #
--echo # 2. Test that innodb shutdown as expected if any error happens before
--echo # normal rollback task is started. In the situation, rollback task
--echo # should be started at preshutdown accordingly to rollback or
--echo # deregister all recovered active transactions.
--echo #
# Generate an large active transaction
INSERT INTO t1 SELECT seq + 2, 1 FROM seq_1_to_1024;
BEGIN;
UPDATE t1 SET data = 10;
# Make sure above update is persisted.
SET GLOBAL innodb_log_checkpoint_now = 1;
--source include/kill_mysqld.inc
# tc-heuristic-recover triggers an error before innodb rollback task start
# Rollback task will not be started at preshutdown of read only mode. Active
# transactions are not expected to rollback.
--error 1
--exec $MYSQLD_LAST_CMD --tc-heuristic-recover=ROLLBACK --innodb-read-only --log-error=$MYSQLTEST_VARDIR/tmp/log.err
# Rollback task will not be started at preshutdown if recovery mode is greater
# to 2. Active transactions are not expected to rollback.
--error 1
--exec $MYSQLD_LAST_CMD --tc-heuristic-recover=ROLLBACK --innodb-force-recovery=3 --log-error=$MYSQLTEST_VARDIR/tmp/log.err
# Rollback task will be started at preshutdown of fast shutdown if force
# recovery is 2. But the transaction is deregistered instead of rollback.
--error 1
--exec $MYSQLD_LAST_CMD --tc-heuristic-recover=ROLLBACK --innodb-fast-shutdown=1 --innodb-force-recovery=2 --log-error=$MYSQLTEST_VARDIR/tmp/log.err
--let $restart_parameters= --innodb-read-only
--source include/start_mysqld.inc
# Verify that the transaction is still there.
SELECT count(*) FROM information_schema.innodb_trx;
--source include/kill_mysqld.inc
# Rollback task will be started at preshutdown of fast shutdown. The
# active transaction is rolled back.
--error 1
--exec $MYSQLD_LAST_CMD --tc-heuristic-recover=ROLLBACK --innodb-fast-shutdown=1 --log-error=$MYSQLTEST_VARDIR/tmp/log.err
--let $restart_parameters= --innodb-read-only
--source include/start_mysqld.inc
# Verify that the transaction is still there.
SELECT count(*) FROM information_schema.innodb_trx;
--remove_file $MYSQLTEST_VARDIR/tmp/log.err
--let $restart_parameters=
--source include/restart_mysqld.inc
# There should be no any transaction
--let $wait_condition= SELECT count(*) = 0 FROM information_schema.innodb_trx
--source include/wait_condition.inc
# Cleanup.
DROP TABLE t1;