mariadb/mysql-test/main/partition_innodb_plugin.test

169 lines
5.4 KiB
Text
Raw Normal View History

Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Problem was still not completely fixed, due to qouting. This is the server side only fix (in explain_filename), the change from filename_to_tablename to use explain_filename in the InnoDB code must be done before the bug is fixed. mysql-test/include/have_not_innodb_plugin.inc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added include file to allow test for only the 'old' built-in innodb engine mysql-test/r/not_true.require: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added require to match 'not' TRUE mysql-test/r/partition_innodb_builtin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the 'old' built-in innodb engine mysql-test/r/partition_innodb_plugin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the new plugin innodb engine mysql-test/t/disabled.def: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Disabling the new test until the fix is included in the InnoDB source too. mysql-test/t/partition_innodb_builtin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the 'old' built-in innodb engine mysql-test/t/partition_innodb_plugin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the new plugin innodb engine sql/mysql_priv.h: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added thd as a parameter to explain_filename to be able to use the correct quote character sql/sql_table.cc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Changed explain_filename, so that it does qouting correctly according to the sessions qoute char.
2009-09-25 11:26:49 +02:00
--source include/have_partition.inc
--source include/have_innodb.inc
let $MYSQLD_DATADIR= `SELECT @@datadir`;
call mtr.add_suppression("InnoDB: Table .* does not exist in the InnoDB internal data dictionary .*");
call mtr.add_suppression("Deadlock found when trying to get lock; try restarting transaction");
--echo #
--echo # Bug#11766879/Bug#60106: DIFF BETWEEN # OF INDEXES IN MYSQL VS INNODB,
--echo # PARTITONING, ON INDEX CREATE
--echo # Bug#12696518: MEMORY LEAKS IN HA_PARTITION (VALGRIND TESTS ON TRUNK)
--echo #
CREATE TABLE t1 (
id bigint NOT NULL AUTO_INCREMENT,
time date,
id2 bigint not null,
PRIMARY KEY (id,time)
) ENGINE=InnoDB DEFAULT CHARSET=utf8
/*!50100 PARTITION BY RANGE(TO_DAYS(time))
(PARTITION p10 VALUES LESS THAN (734708) ENGINE = InnoDB,
PARTITION p20 VALUES LESS THAN MAXVALUE ENGINE = InnoDB) */;
INSERT INTO t1 (time,id2) VALUES ('2011-07-24',1);
INSERT INTO t1 (time,id2) VALUES ('2011-07-25',1);
INSERT INTO t1 (time,id2) VALUES ('2011-07-25',1);
--error ER_DUP_ENTRY
CREATE UNIQUE INDEX uk_time_id2 on t1(time,id2);
SELECT COUNT(*) FROM t1;
SHOW CREATE TABLE t1;
DROP TABLE t1;
2011-12-30 11:22:27 +01:00
call mtr.add_suppression("InnoDB: Error: table `test`.`t1` .* InnoDB internal");
--echo #
--echo # Bug#55091: Server crashes on ADD PARTITION after a failed attempt
--echo #
SET @old_innodb_file_per_table = @@global.innodb_file_per_table;
SET @old_innodb_strict_mode = @@global.innodb_strict_mode;
SET @@global.innodb_file_per_table = ON,
@@global.innodb_strict_mode = ON;
--disable_query_log
SET @save_innodb_read_only_compressed=@@GLOBAL.innodb_read_only_compressed;
SET GLOBAL innodb_read_only_compressed=OFF;
--enable_query_log
--connect(con1,localhost,root,,)
CREATE TABLE t1 (id INT NOT NULL
PRIMARY KEY,
user_num CHAR(10)
) ENGINE = InnoDB
KEY_BLOCK_SIZE=4
PARTITION BY HASH(id) PARTITIONS 1;
--replace_result #p# #P#
--list_files $MYSQLD_DATADIR/test t1*
SHOW CREATE TABLE t1;
SET GLOBAL innodb_file_per_table = OFF;
--disconnect con1
--connect(con2,localhost,root,,)
LOCK TABLE t1 WRITE;
--echo # ALTER fails because COMPRESSED/KEY_BLOCK_SIZE
--echo # are incompatible with innodb_file_per_table = OFF;
Fixed build failures found by buildbot - Added suppression of warnings - Fixed some test cases BUILD/FINISH.sh: Added AM_EXTRA_MAKEFLAGS BUILD/SETUP.sh: Added option --extra-makeflags client/mysqldump.c: Added suppression mysql-test/r/mysql.result: Updated results mysql-test/r/mysql_upgrade.result: Updated results mysql-test/r/partition_innodb_plugin.result: Updated results mysql-test/r/partition_open_files_limit.result: Updated results mysql-test/r/symlink.result: Updated results mysql-test/suite/innodb/r/innodb-create-options.result: Updated results mysql-test/suite/innodb/t/innodb-create-options.test: Don't print error message (as it's varies on different system) mysql-test/t/mysql.test: Don't print error message (as it's varies on different system) mysql-test/t/mysql_upgrade.test: Fixed checking of error number mysql-test/t/partition_innodb_plugin.test: Don't print error message (as it's varies on different system) plugin/semisync/semisync_master.cc: Added suppression sql/ha_partition.cc: Added suppression sql/item_subselect.cc: Added suppression sql/multi_range_read.cc: Added suppression sql/sql_parse.cc: Added suppression sql/sql_select.cc: Added suppression storage/innobase/handler/ha_innodb.cc: Removed not used variable storage/maria/ma_delete.c: Added suppression storage/maria/ma_key_recover.c: Added suppression storage/maria/ma_write.c: Added suppression strings/ctype-ucs2.c: Added suppression support-files/compiler_warnings.supp: Added suppressions unittest/mysys/my_vsnprintf-t.c: Fixed test case with %M to also work on Solaris
2012-06-05 14:09:18 +03:00
--replace_regex / - .*//
--error ER_GET_ERRNO
ALTER TABLE t1 ADD PARTITION PARTITIONS 1;
--replace_result #p# #P#
--list_files $MYSQLD_DATADIR/test t1*
--echo # This SET is not needed to reproduce the bug,
--echo # it is here just to make the test case more realistic
SET innodb_strict_mode = OFF;
ALTER TABLE t1 ADD PARTITION PARTITIONS 2;
--replace_result #p# #P#
--list_files $MYSQLD_DATADIR/test t1*
# really bug#56172
ALTER TABLE t1 REBUILD PARTITION p0;
UNLOCK TABLES;
SHOW CREATE TABLE t1;
DROP TABLE t1;
--disconnect con2
--connection default
SET @@global.innodb_strict_mode = @old_innodb_strict_mode;
SET @@global.innodb_file_per_table = @old_innodb_file_per_table;
--disable_query_log
SET GLOBAL innodb_read_only_compressed=@save_innodb_read_only_compressed;
--enable_query_log
Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Problem was still not completely fixed, due to qouting. This is the server side only fix (in explain_filename), the change from filename_to_tablename to use explain_filename in the InnoDB code must be done before the bug is fixed. mysql-test/include/have_not_innodb_plugin.inc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added include file to allow test for only the 'old' built-in innodb engine mysql-test/r/not_true.require: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added require to match 'not' TRUE mysql-test/r/partition_innodb_builtin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the 'old' built-in innodb engine mysql-test/r/partition_innodb_plugin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the new plugin innodb engine mysql-test/t/disabled.def: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Disabling the new test until the fix is included in the InnoDB source too. mysql-test/t/partition_innodb_builtin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the 'old' built-in innodb engine mysql-test/t/partition_innodb_plugin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the new plugin innodb engine sql/mysql_priv.h: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added thd as a parameter to explain_filename to be able to use the correct quote character sql/sql_table.cc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Changed explain_filename, so that it does qouting correctly according to the sessions qoute char.
2009-09-25 11:26:49 +02:00
#
# Bug#32430 - show engine innodb status causes errors
#
#change to "disable_service_connection" after fix MDEV-29278
--disable_view_protocol
MDEV-24738 Improve the InnoDB deadlock checker A new configuration parameter innodb_deadlock_report is introduced: * innodb_deadlock_report=off: Do not report any details of deadlocks. * innodb_deadlock_report=basic: Report transactions and waiting locks. * innodb_deadlock_report=full (default): Report also the blocking locks. The improved deadlock checker will consider all involved transactions in one loop, even if the deadlock loop includes several transactions. The theoretical maximum number of transactions that can be involved in a deadlock is `innodb_page_size` * 8, limited by the persistent data structures. Note: Similar to mysql/mysql-server@3859219875b62154b921e8c6078c751198071b9c our deadlock checker will consider at most one blocking transaction for each waiting transaction. The new field trx->lock.wait_trx be nullptr if and only if trx->lock.wait_lock is nullptr. Note that trx->lock.wait_lock->trx == trx (the waiting transaction), while trx->lock.wait_trx points to one of the transactions whose lock is conflicting with trx->lock.wait_lock. Considering only one blocking transaction will greatly simplify our deadlock checker, but it may also make the deadlock checker blind to some deadlocks where the deadlock cycle is 'hidden' by the fact that the registered trx->lock.wait_trx is not actually waiting for any InnoDB lock, but something else. So, instead of deadlocks, sometimes lock wait timeout may be reported. To improve on this, whenever trx->lock.wait_trx is changed, we will register further 'candidate' transactions in Deadlock::to_check(), and check for 'revealed' deadlocks as soon as possible, in lock_release() and innobase_kill_query(). The old DeadlockChecker was holding lock_sys.latch, even though using lock_sys.wait_mutex should be less contended (and thus preferred) in the likely case that no deadlock is present. lock_wait(): Defer the deadlock check to this function, instead of executing it in lock_rec_enqueue_waiting(), lock_table_enqueue_waiting(). DeadlockChecker: Complete rewrite: (1) Explicitly keep track of transactions that are being waited for, in trx->lock.wait_trx, protected by lock_sys.wait_mutex. Previously, we were painstakingly traversing the lock heaps while blocking concurrent registration or removal of any locks (even uncontended ones). (2) Use Brent's cycle-detection algorithm for deadlock detection, traversing each trx->lock.wait_trx edge at most 2 times. (3) If a deadlock is detected, release lock_sys.wait_mutex, acquire LockMutexGuard, re-acquire lock_sys.wait_mutex and re-invoke find_cycle() to find out whether the deadlock is still present. (4) Display information on all transactions that are involved in the deadlock, and choose a victim to be rolled back. lock_sys.deadlocks: Replaces lock_deadlock_found. Protected by wait_mutex. Deadlock::find_cycle(): Quickly find a cycle of trx->lock.wait_trx... using Brent's cycle detection algorithm. Deadlock::report(): Report a deadlock cycle that was found by Deadlock::find_cycle(), and choose a victim with the least weight. Altogether, we may traverse each trx->lock.wait_trx edge up to 5 times (2*find_cycle()+1 time for reporting and choosing the victim). Deadlock::check_and_resolve(): Find and resolve a deadlock. lock_wait_rpl_report(): Report the waits-for information to replication. This used to be executed as part of DeadlockChecker. Replication must know the waits-for relations even if no deadlocks are present in InnoDB. Reviewed by: Vladislav Vaintroub
2021-02-17 12:43:33 +02:00
SET @save_detect= @@GLOBAL.innodb_deadlock_detect;
SET @save_report= @@GLOBAL.innodb_deadlock_report;
SET GLOBAL innodb_deadlock_detect=ON;
SET GLOBAL innodb_deadlock_report=BASIC;
Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Problem was still not completely fixed, due to qouting. This is the server side only fix (in explain_filename), the change from filename_to_tablename to use explain_filename in the InnoDB code must be done before the bug is fixed. mysql-test/include/have_not_innodb_plugin.inc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added include file to allow test for only the 'old' built-in innodb engine mysql-test/r/not_true.require: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added require to match 'not' TRUE mysql-test/r/partition_innodb_builtin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the 'old' built-in innodb engine mysql-test/r/partition_innodb_plugin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the new plugin innodb engine mysql-test/t/disabled.def: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Disabling the new test until the fix is included in the InnoDB source too. mysql-test/t/partition_innodb_builtin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the 'old' built-in innodb engine mysql-test/t/partition_innodb_plugin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the new plugin innodb engine sql/mysql_priv.h: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added thd as a parameter to explain_filename to be able to use the correct quote character sql/sql_table.cc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Changed explain_filename, so that it does qouting correctly according to the sessions qoute char.
2009-09-25 11:26:49 +02:00
SET NAMES utf8;
CREATE TABLE `t``\""e` (a INT, PRIMARY KEY (a))
ENGINE=InnoDB
PARTITION BY RANGE (a)
SUBPARTITION BY HASH (a)
(PARTITION `p0``\""e` VALUES LESS THAN (100)
(SUBPARTITION `sp0``\""e`,
SUBPARTITION `sp1``\""e`),
PARTITION `p1``\""e` VALUES LESS THAN (MAXVALUE)
(SUBPARTITION `sp2``\""e`,
SUBPARTITION `sp3``\""e`));
INSERT INTO `t``\""e` VALUES (0), (2), (6), (10), (14), (18), (22);
START TRANSACTION;
connect(con1,localhost,root,,);
SET NAMES utf8;
START TRANSACTION;
connection default;
UPDATE `t``\""e` SET a = 16 WHERE a = 0;
connection con1;
UPDATE `t``\""e` SET a = 8 WHERE a = 22;
let $id_1= `SELECT CONNECTION_ID()`;
SEND;
UPDATE `t``\""e` SET a = 12 WHERE a = 0;
connection default;
let $wait_condition= SELECT COUNT(*)=2 FROM INFORMATION_SCHEMA.INNODB_LOCKS;
Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Problem was still not completely fixed, due to qouting. This is the server side only fix (in explain_filename), the change from filename_to_tablename to use explain_filename in the InnoDB code must be done before the bug is fixed. mysql-test/include/have_not_innodb_plugin.inc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added include file to allow test for only the 'old' built-in innodb engine mysql-test/r/not_true.require: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added require to match 'not' TRUE mysql-test/r/partition_innodb_builtin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the 'old' built-in innodb engine mysql-test/r/partition_innodb_plugin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the new plugin innodb engine mysql-test/t/disabled.def: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Disabling the new test until the fix is included in the InnoDB source too. mysql-test/t/partition_innodb_builtin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the 'old' built-in innodb engine mysql-test/t/partition_innodb_plugin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the new plugin innodb engine sql/mysql_priv.h: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added thd as a parameter to explain_filename to be able to use the correct quote character sql/sql_table.cc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Changed explain_filename, so that it does qouting correctly according to the sessions qoute char.
2009-09-25 11:26:49 +02:00
--source include/wait_condition.inc
--sorted_result
SELECT lock_table, COUNT(*) FROM INFORMATION_SCHEMA.INNODB_LOCKS
GROUP BY lock_table;
set @old_sql_mode = @@sql_mode;
set sql_mode = 'ANSI_QUOTES';
--sorted_result
SELECT lock_table, COUNT(*) FROM INFORMATION_SCHEMA.INNODB_LOCKS
GROUP BY lock_table;
set @@sql_mode = @old_sql_mode;
--error ER_LOCK_DEADLOCK
UPDATE `t``\""e` SET a = 4 WHERE a = 22;
--echo # First table reported in 'SHOW ENGINE InnoDB STATUS'
# RECORD LOCKS space id 0 page no 50 n bits 80 index `PRIMARY` in \
# Database `test`, Table `t1`, Partition `p0`, Subpartition `sp0` \
# trx id 0 775
# NOTE: replace_regex is very slow on match copy/past '(.*)' regex's
# on big texts, removing a lot of text before + after makes it much faster.
#/.*in (.*) trx.*/\1/
--replace_regex /.*RECORD LOCKS space id [0-9]* page no [0-9]* n bits [0-9]* // / trx id .*// /.*index .* in // /trx table locks [0-9]* // /total table locks [0-9]* //
Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Problem was still not completely fixed, due to qouting. This is the server side only fix (in explain_filename), the change from filename_to_tablename to use explain_filename in the InnoDB code must be done before the bug is fixed. mysql-test/include/have_not_innodb_plugin.inc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added include file to allow test for only the 'old' built-in innodb engine mysql-test/r/not_true.require: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added require to match 'not' TRUE mysql-test/r/partition_innodb_builtin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the 'old' built-in innodb engine mysql-test/r/partition_innodb_plugin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the new plugin innodb engine mysql-test/t/disabled.def: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Disabling the new test until the fix is included in the InnoDB source too. mysql-test/t/partition_innodb_builtin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the 'old' built-in innodb engine mysql-test/t/partition_innodb_plugin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the new plugin innodb engine sql/mysql_priv.h: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added thd as a parameter to explain_filename to be able to use the correct quote character sql/sql_table.cc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Changed explain_filename, so that it does qouting correctly according to the sessions qoute char.
2009-09-25 11:26:49 +02:00
SHOW ENGINE InnoDB STATUS;
set @old_sql_mode = @@sql_mode;
set sql_mode = 'ANSI_QUOTES';
--replace_regex /.*RECORD LOCKS space id [0-9]* page no [0-9]* n bits [0-9]* // / trx id .*// /.*index .* in // /trx table locks [0-9]* // /total table locks [0-9]* //
Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Problem was still not completely fixed, due to qouting. This is the server side only fix (in explain_filename), the change from filename_to_tablename to use explain_filename in the InnoDB code must be done before the bug is fixed. mysql-test/include/have_not_innodb_plugin.inc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added include file to allow test for only the 'old' built-in innodb engine mysql-test/r/not_true.require: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added require to match 'not' TRUE mysql-test/r/partition_innodb_builtin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the 'old' built-in innodb engine mysql-test/r/partition_innodb_plugin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the new plugin innodb engine mysql-test/t/disabled.def: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Disabling the new test until the fix is included in the InnoDB source too. mysql-test/t/partition_innodb_builtin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the 'old' built-in innodb engine mysql-test/t/partition_innodb_plugin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the new plugin innodb engine sql/mysql_priv.h: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added thd as a parameter to explain_filename to be able to use the correct quote character sql/sql_table.cc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Changed explain_filename, so that it does qouting correctly according to the sessions qoute char.
2009-09-25 11:26:49 +02:00
SHOW ENGINE InnoDB STATUS;
MDEV-24738 Improve the InnoDB deadlock checker A new configuration parameter innodb_deadlock_report is introduced: * innodb_deadlock_report=off: Do not report any details of deadlocks. * innodb_deadlock_report=basic: Report transactions and waiting locks. * innodb_deadlock_report=full (default): Report also the blocking locks. The improved deadlock checker will consider all involved transactions in one loop, even if the deadlock loop includes several transactions. The theoretical maximum number of transactions that can be involved in a deadlock is `innodb_page_size` * 8, limited by the persistent data structures. Note: Similar to mysql/mysql-server@3859219875b62154b921e8c6078c751198071b9c our deadlock checker will consider at most one blocking transaction for each waiting transaction. The new field trx->lock.wait_trx be nullptr if and only if trx->lock.wait_lock is nullptr. Note that trx->lock.wait_lock->trx == trx (the waiting transaction), while trx->lock.wait_trx points to one of the transactions whose lock is conflicting with trx->lock.wait_lock. Considering only one blocking transaction will greatly simplify our deadlock checker, but it may also make the deadlock checker blind to some deadlocks where the deadlock cycle is 'hidden' by the fact that the registered trx->lock.wait_trx is not actually waiting for any InnoDB lock, but something else. So, instead of deadlocks, sometimes lock wait timeout may be reported. To improve on this, whenever trx->lock.wait_trx is changed, we will register further 'candidate' transactions in Deadlock::to_check(), and check for 'revealed' deadlocks as soon as possible, in lock_release() and innobase_kill_query(). The old DeadlockChecker was holding lock_sys.latch, even though using lock_sys.wait_mutex should be less contended (and thus preferred) in the likely case that no deadlock is present. lock_wait(): Defer the deadlock check to this function, instead of executing it in lock_rec_enqueue_waiting(), lock_table_enqueue_waiting(). DeadlockChecker: Complete rewrite: (1) Explicitly keep track of transactions that are being waited for, in trx->lock.wait_trx, protected by lock_sys.wait_mutex. Previously, we were painstakingly traversing the lock heaps while blocking concurrent registration or removal of any locks (even uncontended ones). (2) Use Brent's cycle-detection algorithm for deadlock detection, traversing each trx->lock.wait_trx edge at most 2 times. (3) If a deadlock is detected, release lock_sys.wait_mutex, acquire LockMutexGuard, re-acquire lock_sys.wait_mutex and re-invoke find_cycle() to find out whether the deadlock is still present. (4) Display information on all transactions that are involved in the deadlock, and choose a victim to be rolled back. lock_sys.deadlocks: Replaces lock_deadlock_found. Protected by wait_mutex. Deadlock::find_cycle(): Quickly find a cycle of trx->lock.wait_trx... using Brent's cycle detection algorithm. Deadlock::report(): Report a deadlock cycle that was found by Deadlock::find_cycle(), and choose a victim with the least weight. Altogether, we may traverse each trx->lock.wait_trx edge up to 5 times (2*find_cycle()+1 time for reporting and choosing the victim). Deadlock::check_and_resolve(): Find and resolve a deadlock. lock_wait_rpl_report(): Report the waits-for information to replication. This used to be executed as part of DeadlockChecker. Replication must know the waits-for relations even if no deadlocks are present in InnoDB. Reviewed by: Vladislav Vaintroub
2021-02-17 12:43:33 +02:00
SET GLOBAL innodb_deadlock_detect= @save_detect;
SET GLOBAL innodb_deadlock_report= @save_report;
Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Problem was still not completely fixed, due to qouting. This is the server side only fix (in explain_filename), the change from filename_to_tablename to use explain_filename in the InnoDB code must be done before the bug is fixed. mysql-test/include/have_not_innodb_plugin.inc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added include file to allow test for only the 'old' built-in innodb engine mysql-test/r/not_true.require: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added require to match 'not' TRUE mysql-test/r/partition_innodb_builtin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the 'old' built-in innodb engine mysql-test/r/partition_innodb_plugin.result: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New result file for partitioning specific to the new plugin innodb engine mysql-test/t/disabled.def: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Disabling the new test until the fix is included in the InnoDB source too. mysql-test/t/partition_innodb_builtin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the 'old' built-in innodb engine mysql-test/t/partition_innodb_plugin.test: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs New test file for partitioning specific to the new plugin innodb engine sql/mysql_priv.h: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Added thd as a parameter to explain_filename to be able to use the correct quote character sql/sql_table.cc: Bug#32430: 'show innodb status' causes errors Invalid (old?) table or database name in logs Changed explain_filename, so that it does qouting correctly according to the sessions qoute char.
2009-09-25 11:26:49 +02:00
set @@sql_mode = @old_sql_mode;
connection con1;
REAP;
ROLLBACK;
disconnect con1;
connection default;
DROP TABLE `t``\""e`;
--enable_view_protocol