mariadb/sql/ddl_log.h

359 lines
12 KiB
C
Raw Normal View History

/*
Copyright (c) 2000, 2019, Oracle and/or its affiliates.
Revert MDEV-25292 Atomic CREATE OR REPLACE TABLE Specifically: Revert "MDEV-29664 Assertion `!n_mysql_tables_in_use' failed in innobase_close_connection" This reverts commit ba875e939619baefb08936863a889830f595e426. Revert "MDEV-29620 Assertion `next_insert_id == 0' failed in handler::ha_external_lock" This reverts commit aa08a7442a5141750de96353132c7799e10c9302. Revert "MDEV-29628 Memory leak after CREATE OR REPLACE with foreign key" This reverts commit c579d66ba689c8a20a951de5e6be2e1bc255ba7f. Revert "MDEV-29609 create_not_windows test fails with different result" This reverts commit cb583b2f1bf1c910e82c7caee4a53e0f3ff954ba. Revert "MDEV-29544 SIGSEGV in HA_CREATE_INFO::finalize_locked_tables" This reverts commit dcd66c3814991e686615eac2e65d01a7fa80cb66. Revert "MDEV-28933 CREATE OR REPLACE fails to recreate same constraint name" This reverts commit cf6c5176328c8fbfadac80f337ef285732cc8d06. Revert "MDEV-28933 Moved RENAME_CONSTRAINT_IDS to include/sql_funcs.h" This reverts commit f1e1c1335b506905244502ee8584736a17a6ecd0. Revert "MDEV-28956 Locking is broken if CREATE OR REPLACE fails under LOCK TABLES" This reverts commit a228ec80e30cae59c2057b786fd3559171ce794f. Revert "MDEV-25292 gcol.gcol_bugfixes --ps fix" This reverts commit 24fff8267d0722b20ca780cf144ce27560df1394. Revert "MDEV-25292 Disable atomic replace for slave-generated or-replace" This reverts commit 2af15914cb3ae204d3b413995a96c7722ad60d93. Revert "MDEV-25292 backup_log improved" This reverts commit 34398a20b5829f0508c78bc8f765f7bef6e0ad48. Revert "MDEV-25292 Atomic CREATE OR REPLACE TABLE" This reverts commit 93c8252f02faa8ad8dc5f005e52f1990c29d4a0d. Revert "MDEV-25292 Table_name class for (db, table_name, alias)" This reverts commit d145dda9c7b6db394edf7150cc255343a4a7b116. Revert "MDEV-25292 ha_table_exists() cleanup and improvement" This reverts commit 409b8a86de45efc96377c618a99cff1301523150. Revert "MDEV-25292 Cleanups" This reverts commit 595dad83ad3c23879c59deec1173ea16b47871ed. Revert "MDEV-25292 Refactoring: moved select_field_count into Alter_info." This reverts commit f02af1d2295f108d96671b5a4f731ec998002dd5.
2022-10-27 22:18:51 +02:00
Copyright (c) 2010, 2021, MariaDB
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; version 2 of the License.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; if not, write to the Free Software
Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1335 USA
*/
/* External interfaces to ddl log functions */
#ifndef DDL_LOG_INCLUDED
#define DDL_LOG_INCLUDED
enum ddl_log_entry_code
{
/*
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
DDL_LOG_UNKOWN
Here mainly to detect blocks that are all zero
DDL_LOG_EXECUTE_CODE:
This is a code that indicates that this is a log entry to
be executed, from this entry a linked list of log entries
can be found and executed.
DDL_LOG_ENTRY_CODE:
An entry to be executed in a linked list from an execute log
entry.
DDL_LOG_IGNORE_ENTRY_CODE:
An entry that is to be ignored
*/
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
DDL_LOG_UNKNOWN= 0,
DDL_LOG_EXECUTE_CODE= 1,
DDL_LOG_ENTRY_CODE= 2,
DDL_LOG_IGNORE_ENTRY_CODE= 3,
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
DDL_LOG_ENTRY_CODE_LAST= 4
};
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
/*
When adding things below, also add an entry to ddl_log_action_names and
ddl_log_entry_phases in ddl_log.cc
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
*/
enum ddl_log_action_code
{
/*
The type of action that a DDL_LOG_ENTRY_CODE entry is to
perform.
*/
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
DDL_LOG_UNKNOWN_ACTION= 0,
/* Delete a .frm file or a table in the partition engine */
DDL_LOG_DELETE_ACTION= 1,
/* Rename a .frm fire a table in the partition engine */
DDL_LOG_RENAME_ACTION= 2,
/*
Rename an entity after removing the previous entry with the
new name, that is replace this entry.
*/
DDL_LOG_REPLACE_ACTION= 3,
/* Exchange two entities by renaming them a -> tmp, b -> a, tmp -> b */
DDL_LOG_EXCHANGE_ACTION= 4,
/*
log do_rename(): Rename of .frm file, table, stat_tables and triggers
*/
DDL_LOG_RENAME_TABLE_ACTION= 5,
DDL_LOG_RENAME_VIEW_ACTION= 6,
MDEV-24408 Crash-safe DROP DATABASE Description of how DROP DATABASE works after this patch - Collect list of tables - DDL log tables as they are dropped - DDL log drop database - Delete db.opt - Delete data directory - Log either DROP TABLE or DROP DATABASE to binary log - De active ddl log entry This is in line of how things where before (minus ddl logging) except that we delete db.opt file last to not loose it if DROP DATABASE fails. On recovery we have to ensure that all dropped tables are logged in binary log and that they are properly dropped (as with atomic drop table). No new tables be dropped as part of recovery. Recovery of active drop database ddl log entry: - If drop database was logged to ddl log but was not found in the binary log: - drop the db.opt file and database directory. - Log DROP DATABASE to binary log - If drop database was not logged to ddl log - Update binary log with DROP TABLE of the dropped tables. If table list is longer than max_allowed_packet, then the query will be split into multiple DROP TABLE/VIEW queries. Other things: - Added DDL_LOG_STATE and 'current database' as arguments to mysql_rm_table_no_locks(). This was needed to be able to combine ddl logging of DROP DATABASE and DROP TABLE and make the generated DROP TABLE statements shorter. - To make the DROP TABLE statement created by ddl log shorter, I changed the binlogged query to use current directory and omit the directory part for all tables in the current directory. - Merged some DROP TABLE and DROP VIEW code in ddl logger. This was done to be able get separate DROP VIEW and DROP TABLE statements in the binary log. - Added a 'recovery_state' variable to remember the state of dropped tables and views. - Moved out code that drops database objects (stored procedures) from mysql_rm_db_internal() to drop_database_objects() for better code reuse. - Made mysql_rm_db_internal() global so that could be used by the ddl recovery code.
2020-12-20 16:44:11 +01:00
DDL_LOG_DROP_INIT_ACTION= 7,
DDL_LOG_DROP_TABLE_ACTION= 8,
MDEV-24408 Crash-safe DROP DATABASE Description of how DROP DATABASE works after this patch - Collect list of tables - DDL log tables as they are dropped - DDL log drop database - Delete db.opt - Delete data directory - Log either DROP TABLE or DROP DATABASE to binary log - De active ddl log entry This is in line of how things where before (minus ddl logging) except that we delete db.opt file last to not loose it if DROP DATABASE fails. On recovery we have to ensure that all dropped tables are logged in binary log and that they are properly dropped (as with atomic drop table). No new tables be dropped as part of recovery. Recovery of active drop database ddl log entry: - If drop database was logged to ddl log but was not found in the binary log: - drop the db.opt file and database directory. - Log DROP DATABASE to binary log - If drop database was not logged to ddl log - Update binary log with DROP TABLE of the dropped tables. If table list is longer than max_allowed_packet, then the query will be split into multiple DROP TABLE/VIEW queries. Other things: - Added DDL_LOG_STATE and 'current database' as arguments to mysql_rm_table_no_locks(). This was needed to be able to combine ddl logging of DROP DATABASE and DROP TABLE and make the generated DROP TABLE statements shorter. - To make the DROP TABLE statement created by ddl log shorter, I changed the binlogged query to use current directory and omit the directory part for all tables in the current directory. - Merged some DROP TABLE and DROP VIEW code in ddl logger. This was done to be able get separate DROP VIEW and DROP TABLE statements in the binary log. - Added a 'recovery_state' variable to remember the state of dropped tables and views. - Moved out code that drops database objects (stored procedures) from mysql_rm_db_internal() to drop_database_objects() for better code reuse. - Made mysql_rm_db_internal() global so that could be used by the ddl recovery code.
2020-12-20 16:44:11 +01:00
DDL_LOG_DROP_VIEW_ACTION= 9,
DDL_LOG_DROP_TRIGGER_ACTION= 10,
DDL_LOG_DROP_DB_ACTION=11,
DDL_LOG_CREATE_TABLE_ACTION=12,
DDL_LOG_CREATE_VIEW_ACTION=13,
DDL_LOG_DELETE_TMP_FILE_ACTION=14,
DDL_LOG_CREATE_TRIGGER_ACTION=15,
MDEV-25180 Atomic ALTER TABLE MDEV-25604 Atomic DDL: Binlog event written upon recovery does not have default database The purpose of this task is to ensure that ALTER TABLE is atomic even if the MariaDB server would be killed at any point of the alter table. This means that either the ALTER TABLE succeeds (including that triggers, the status tables and the binary log are updated) or things should be reverted to their original state. If the server crashes before the new version is fully up to date and commited, it will revert to the original table and remove all temporary files and tables. If the new version is commited, crash recovery will use the new version, and update triggers, the status tables and the binary log. The one execption is ALTER TABLE .. RENAME .. where no changes are done to table definition. This one will work as RENAME and roll back unless the whole statement completed, including updating the binary log (if enabled). Other changes: - Added handlerton->check_version() function to allow the ddl recovery code to check, in case of inplace alter table, if the table in the storage engine is of the new or old version. - Added handler->table_version() so that an engine can report the current version of the table. This should be changed each time the table definition changes. - Added ha_signal_ddl_recovery_done() and handlerton::signal_ddl_recovery_done() to inform all handlers when ddl recovery has been done. (Needed by InnoDB). - Added handlerton call inplace_alter_table_committed, to signal engine that ddl_log has been closed for the alter table query. - Added new handerton flag HTON_REQUIRES_NOTIFY_TABLEDEF_CHANGED_AFTER_COMMIT to signal when we should call hton->notify_tabledef_changed() during mysql_inplace_alter_table. This was required as MyRocks and InnoDB needed the call at different times. - Added function server_uuid_value() to be able to generate a temporary xid when ddl recovery writes the query to the binary log. This is needed to be able to handle crashes during ddl log recovery. - Moved freeing of the frm definition to end of mysql_alter_table() to remove duplicate code and have a common exit strategy. ------- InnoDB part of atomic ALTER TABLE (Implemented by Marko Mäkelä) innodb_check_version(): Compare the saved dict_table_t::def_trx_id to determine whether an ALTER TABLE operation was committed. We must correctly recover dict_table_t::def_trx_id for this to work. Before purge removes any trace of DB_TRX_ID from system tables, it will make an effort to load the user table into the cache, so that the dict_table_t::def_trx_id can be recovered. ha_innobase::table_version(): return garbage, or the trx_id that would be used for committing an ALTER TABLE operation. In InnoDB, table names starting with #sql-ib will remain special: they will be dropped on startup. This may be revisited later in MDEV-18518 when we implement proper undo logging and rollback for creating or dropping multiple tables in a transaction. Table names starting with #sql will retain some special meaning: dict_table_t::parse_name() will not consider such names for MDL acquisition, and dict_table_rename_in_cache() will treat such names specially when handling FOREIGN KEY constraints. Simplify InnoDB DROP INDEX. Prevent purge wakeup To ensure that dict_table_t::def_trx_id will be recovered correctly in case the server is killed before ddl_log_complete(), we will block the purge of any history in SYS_TABLES, SYS_INDEXES, SYS_COLUMNS between ha_innobase::commit_inplace_alter_table(commit=true) (purge_sys.stop_SYS()) and purge_sys.resume_SYS(). The completion callback purge_sys.resume_SYS() must be between ddl_log_complete() and MDL release. -------- MyRocks support for atomic ALTER TABLE (Implemented by Sergui Petrunia) Implement these SE API functions: - ha_rocksdb::table_version() - hton->check_version = rocksdb_check_versionMyRocks data dictionary now stores table version for each table. (Absence of table version record is interpreted as table_version=0, that is, which means no upgrade changes are needed) - For inplace alter table of a partitioned table, call the underlying handlerton when checking if the table is ok. This assumes that the partition engine commits all changes at once.
2021-03-18 11:41:08 +01:00
DDL_LOG_ALTER_TABLE_ACTION=16,
DDL_LOG_STORE_QUERY_ACTION=17,
DDL_LOG_LAST_ACTION /* End marker */
};
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
/* Number of phases for each ddl_log_action_code */
extern const uchar ddl_log_entry_phases[DDL_LOG_LAST_ACTION];
enum enum_ddl_log_exchange_phase {
EXCH_PHASE_NAME_TO_TEMP= 0,
EXCH_PHASE_FROM_TO_NAME= 1,
MDEV-24408 Crash-safe DROP DATABASE Description of how DROP DATABASE works after this patch - Collect list of tables - DDL log tables as they are dropped - DDL log drop database - Delete db.opt - Delete data directory - Log either DROP TABLE or DROP DATABASE to binary log - De active ddl log entry This is in line of how things where before (minus ddl logging) except that we delete db.opt file last to not loose it if DROP DATABASE fails. On recovery we have to ensure that all dropped tables are logged in binary log and that they are properly dropped (as with atomic drop table). No new tables be dropped as part of recovery. Recovery of active drop database ddl log entry: - If drop database was logged to ddl log but was not found in the binary log: - drop the db.opt file and database directory. - Log DROP DATABASE to binary log - If drop database was not logged to ddl log - Update binary log with DROP TABLE of the dropped tables. If table list is longer than max_allowed_packet, then the query will be split into multiple DROP TABLE/VIEW queries. Other things: - Added DDL_LOG_STATE and 'current database' as arguments to mysql_rm_table_no_locks(). This was needed to be able to combine ddl logging of DROP DATABASE and DROP TABLE and make the generated DROP TABLE statements shorter. - To make the DROP TABLE statement created by ddl log shorter, I changed the binlogged query to use current directory and omit the directory part for all tables in the current directory. - Merged some DROP TABLE and DROP VIEW code in ddl logger. This was done to be able get separate DROP VIEW and DROP TABLE statements in the binary log. - Added a 'recovery_state' variable to remember the state of dropped tables and views. - Moved out code that drops database objects (stored procedures) from mysql_rm_db_internal() to drop_database_objects() for better code reuse. - Made mysql_rm_db_internal() global so that could be used by the ddl recovery code.
2020-12-20 16:44:11 +01:00
EXCH_PHASE_TEMP_TO_FROM= 2,
EXCH_PHASE_END
};
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
enum enum_ddl_log_rename_table_phase {
DDL_RENAME_PHASE_TRIGGER= 0,
DDL_RENAME_PHASE_STAT,
DDL_RENAME_PHASE_TABLE,
MDEV-24408 Crash-safe DROP DATABASE Description of how DROP DATABASE works after this patch - Collect list of tables - DDL log tables as they are dropped - DDL log drop database - Delete db.opt - Delete data directory - Log either DROP TABLE or DROP DATABASE to binary log - De active ddl log entry This is in line of how things where before (minus ddl logging) except that we delete db.opt file last to not loose it if DROP DATABASE fails. On recovery we have to ensure that all dropped tables are logged in binary log and that they are properly dropped (as with atomic drop table). No new tables be dropped as part of recovery. Recovery of active drop database ddl log entry: - If drop database was logged to ddl log but was not found in the binary log: - drop the db.opt file and database directory. - Log DROP DATABASE to binary log - If drop database was not logged to ddl log - Update binary log with DROP TABLE of the dropped tables. If table list is longer than max_allowed_packet, then the query will be split into multiple DROP TABLE/VIEW queries. Other things: - Added DDL_LOG_STATE and 'current database' as arguments to mysql_rm_table_no_locks(). This was needed to be able to combine ddl logging of DROP DATABASE and DROP TABLE and make the generated DROP TABLE statements shorter. - To make the DROP TABLE statement created by ddl log shorter, I changed the binlogged query to use current directory and omit the directory part for all tables in the current directory. - Merged some DROP TABLE and DROP VIEW code in ddl logger. This was done to be able get separate DROP VIEW and DROP TABLE statements in the binary log. - Added a 'recovery_state' variable to remember the state of dropped tables and views. - Moved out code that drops database objects (stored procedures) from mysql_rm_db_internal() to drop_database_objects() for better code reuse. - Made mysql_rm_db_internal() global so that could be used by the ddl recovery code.
2020-12-20 16:44:11 +01:00
DDL_RENAME_PHASE_END
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
};
enum enum_ddl_log_drop_table_phase {
DDL_DROP_PHASE_TABLE=0,
DDL_DROP_PHASE_TRIGGER,
DDL_DROP_PHASE_BINLOG,
DDL_DROP_PHASE_RESET, /* Reset found list of dropped tables */
DDL_DROP_PHASE_END
};
MDEV-24408 Crash-safe DROP DATABASE Description of how DROP DATABASE works after this patch - Collect list of tables - DDL log tables as they are dropped - DDL log drop database - Delete db.opt - Delete data directory - Log either DROP TABLE or DROP DATABASE to binary log - De active ddl log entry This is in line of how things where before (minus ddl logging) except that we delete db.opt file last to not loose it if DROP DATABASE fails. On recovery we have to ensure that all dropped tables are logged in binary log and that they are properly dropped (as with atomic drop table). No new tables be dropped as part of recovery. Recovery of active drop database ddl log entry: - If drop database was logged to ddl log but was not found in the binary log: - drop the db.opt file and database directory. - Log DROP DATABASE to binary log - If drop database was not logged to ddl log - Update binary log with DROP TABLE of the dropped tables. If table list is longer than max_allowed_packet, then the query will be split into multiple DROP TABLE/VIEW queries. Other things: - Added DDL_LOG_STATE and 'current database' as arguments to mysql_rm_table_no_locks(). This was needed to be able to combine ddl logging of DROP DATABASE and DROP TABLE and make the generated DROP TABLE statements shorter. - To make the DROP TABLE statement created by ddl log shorter, I changed the binlogged query to use current directory and omit the directory part for all tables in the current directory. - Merged some DROP TABLE and DROP VIEW code in ddl logger. This was done to be able get separate DROP VIEW and DROP TABLE statements in the binary log. - Added a 'recovery_state' variable to remember the state of dropped tables and views. - Moved out code that drops database objects (stored procedures) from mysql_rm_db_internal() to drop_database_objects() for better code reuse. - Made mysql_rm_db_internal() global so that could be used by the ddl recovery code.
2020-12-20 16:44:11 +01:00
enum enum_ddl_log_drop_db_phase {
DDL_DROP_DB_PHASE_INIT=0,
DDL_DROP_DB_PHASE_LOG,
DDL_DROP_DB_PHASE_END
};
enum enum_ddl_log_create_table_phase {
DDL_CREATE_TABLE_PHASE_INIT=0,
DDL_CREATE_TABLE_PHASE_LOG,
DDL_CREATE_TABLE_PHASE_END
};
enum enum_ddl_log_create_view_phase {
DDL_CREATE_VIEW_PHASE_NO_OLD_VIEW,
DDL_CREATE_VIEW_PHASE_DELETE_VIEW_COPY,
DDL_CREATE_VIEW_PHASE_OLD_VIEW_COPIED,
DDL_CREATE_VIEW_PHASE_END
};
enum enum_ddl_log_create_trigger_phase {
DDL_CREATE_TRIGGER_PHASE_NO_OLD_TRIGGER,
DDL_CREATE_TRIGGER_PHASE_DELETE_COPY,
DDL_CREATE_TRIGGER_PHASE_OLD_COPIED,
DDL_CREATE_TRIGGER_PHASE_END
};
MDEV-25180 Atomic ALTER TABLE MDEV-25604 Atomic DDL: Binlog event written upon recovery does not have default database The purpose of this task is to ensure that ALTER TABLE is atomic even if the MariaDB server would be killed at any point of the alter table. This means that either the ALTER TABLE succeeds (including that triggers, the status tables and the binary log are updated) or things should be reverted to their original state. If the server crashes before the new version is fully up to date and commited, it will revert to the original table and remove all temporary files and tables. If the new version is commited, crash recovery will use the new version, and update triggers, the status tables and the binary log. The one execption is ALTER TABLE .. RENAME .. where no changes are done to table definition. This one will work as RENAME and roll back unless the whole statement completed, including updating the binary log (if enabled). Other changes: - Added handlerton->check_version() function to allow the ddl recovery code to check, in case of inplace alter table, if the table in the storage engine is of the new or old version. - Added handler->table_version() so that an engine can report the current version of the table. This should be changed each time the table definition changes. - Added ha_signal_ddl_recovery_done() and handlerton::signal_ddl_recovery_done() to inform all handlers when ddl recovery has been done. (Needed by InnoDB). - Added handlerton call inplace_alter_table_committed, to signal engine that ddl_log has been closed for the alter table query. - Added new handerton flag HTON_REQUIRES_NOTIFY_TABLEDEF_CHANGED_AFTER_COMMIT to signal when we should call hton->notify_tabledef_changed() during mysql_inplace_alter_table. This was required as MyRocks and InnoDB needed the call at different times. - Added function server_uuid_value() to be able to generate a temporary xid when ddl recovery writes the query to the binary log. This is needed to be able to handle crashes during ddl log recovery. - Moved freeing of the frm definition to end of mysql_alter_table() to remove duplicate code and have a common exit strategy. ------- InnoDB part of atomic ALTER TABLE (Implemented by Marko Mäkelä) innodb_check_version(): Compare the saved dict_table_t::def_trx_id to determine whether an ALTER TABLE operation was committed. We must correctly recover dict_table_t::def_trx_id for this to work. Before purge removes any trace of DB_TRX_ID from system tables, it will make an effort to load the user table into the cache, so that the dict_table_t::def_trx_id can be recovered. ha_innobase::table_version(): return garbage, or the trx_id that would be used for committing an ALTER TABLE operation. In InnoDB, table names starting with #sql-ib will remain special: they will be dropped on startup. This may be revisited later in MDEV-18518 when we implement proper undo logging and rollback for creating or dropping multiple tables in a transaction. Table names starting with #sql will retain some special meaning: dict_table_t::parse_name() will not consider such names for MDL acquisition, and dict_table_rename_in_cache() will treat such names specially when handling FOREIGN KEY constraints. Simplify InnoDB DROP INDEX. Prevent purge wakeup To ensure that dict_table_t::def_trx_id will be recovered correctly in case the server is killed before ddl_log_complete(), we will block the purge of any history in SYS_TABLES, SYS_INDEXES, SYS_COLUMNS between ha_innobase::commit_inplace_alter_table(commit=true) (purge_sys.stop_SYS()) and purge_sys.resume_SYS(). The completion callback purge_sys.resume_SYS() must be between ddl_log_complete() and MDL release. -------- MyRocks support for atomic ALTER TABLE (Implemented by Sergui Petrunia) Implement these SE API functions: - ha_rocksdb::table_version() - hton->check_version = rocksdb_check_versionMyRocks data dictionary now stores table version for each table. (Absence of table version record is interpreted as table_version=0, that is, which means no upgrade changes are needed) - For inplace alter table of a partitioned table, call the underlying handlerton when checking if the table is ok. This assumes that the partition engine commits all changes at once.
2021-03-18 11:41:08 +01:00
enum enum_ddl_log_alter_table_phase {
DDL_ALTER_TABLE_PHASE_INIT,
DDL_ALTER_TABLE_PHASE_RENAME_FAILED,
DDL_ALTER_TABLE_PHASE_INPLACE_COPIED,
DDL_ALTER_TABLE_PHASE_INPLACE,
DDL_ALTER_TABLE_PHASE_PREPARE_INPLACE,
DDL_ALTER_TABLE_PHASE_CREATED,
DDL_ALTER_TABLE_PHASE_COPIED,
DDL_ALTER_TABLE_PHASE_OLD_RENAMED,
DDL_ALTER_TABLE_PHASE_UPDATE_TRIGGERS,
DDL_ALTER_TABLE_PHASE_UPDATE_STATS,
DDL_ALTER_TABLE_PHASE_UPDATE_BINARY_LOG,
DDL_ALTER_TABLE_PHASE_END
};
/*
Flags stored in DDL_LOG_ENTRY.flags
The flag values can be reused for different commands
*/
#define DDL_LOG_FLAG_ALTER_RENAME (1 << 0)
#define DDL_LOG_FLAG_ALTER_ENGINE_CHANGED (1 << 1)
#define DDL_LOG_FLAG_ONLY_FRM (1 << 2)
#define DDL_LOG_FLAG_UPDATE_STAT (1 << 3)
/*
Set when using ALTER TABLE on a partitioned table and the table
engine is not changed
*/
#define DDL_LOG_FLAG_ALTER_PARTITION (1 << 4)
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
/*
Setting ddl_log_entry.phase to this has the same effect as setting
the phase to the maximum phase (..PHASE_END) for an entry.
*/
#define DDL_LOG_FINAL_PHASE ((uchar) 0xff)
typedef struct st_ddl_log_entry
{
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
LEX_CSTRING name;
LEX_CSTRING from_name;
LEX_CSTRING handler_name;
LEX_CSTRING db;
LEX_CSTRING from_db;
LEX_CSTRING from_handler_name;
MDEV-25180 Atomic ALTER TABLE MDEV-25604 Atomic DDL: Binlog event written upon recovery does not have default database The purpose of this task is to ensure that ALTER TABLE is atomic even if the MariaDB server would be killed at any point of the alter table. This means that either the ALTER TABLE succeeds (including that triggers, the status tables and the binary log are updated) or things should be reverted to their original state. If the server crashes before the new version is fully up to date and commited, it will revert to the original table and remove all temporary files and tables. If the new version is commited, crash recovery will use the new version, and update triggers, the status tables and the binary log. The one execption is ALTER TABLE .. RENAME .. where no changes are done to table definition. This one will work as RENAME and roll back unless the whole statement completed, including updating the binary log (if enabled). Other changes: - Added handlerton->check_version() function to allow the ddl recovery code to check, in case of inplace alter table, if the table in the storage engine is of the new or old version. - Added handler->table_version() so that an engine can report the current version of the table. This should be changed each time the table definition changes. - Added ha_signal_ddl_recovery_done() and handlerton::signal_ddl_recovery_done() to inform all handlers when ddl recovery has been done. (Needed by InnoDB). - Added handlerton call inplace_alter_table_committed, to signal engine that ddl_log has been closed for the alter table query. - Added new handerton flag HTON_REQUIRES_NOTIFY_TABLEDEF_CHANGED_AFTER_COMMIT to signal when we should call hton->notify_tabledef_changed() during mysql_inplace_alter_table. This was required as MyRocks and InnoDB needed the call at different times. - Added function server_uuid_value() to be able to generate a temporary xid when ddl recovery writes the query to the binary log. This is needed to be able to handle crashes during ddl log recovery. - Moved freeing of the frm definition to end of mysql_alter_table() to remove duplicate code and have a common exit strategy. ------- InnoDB part of atomic ALTER TABLE (Implemented by Marko Mäkelä) innodb_check_version(): Compare the saved dict_table_t::def_trx_id to determine whether an ALTER TABLE operation was committed. We must correctly recover dict_table_t::def_trx_id for this to work. Before purge removes any trace of DB_TRX_ID from system tables, it will make an effort to load the user table into the cache, so that the dict_table_t::def_trx_id can be recovered. ha_innobase::table_version(): return garbage, or the trx_id that would be used for committing an ALTER TABLE operation. In InnoDB, table names starting with #sql-ib will remain special: they will be dropped on startup. This may be revisited later in MDEV-18518 when we implement proper undo logging and rollback for creating or dropping multiple tables in a transaction. Table names starting with #sql will retain some special meaning: dict_table_t::parse_name() will not consider such names for MDL acquisition, and dict_table_rename_in_cache() will treat such names specially when handling FOREIGN KEY constraints. Simplify InnoDB DROP INDEX. Prevent purge wakeup To ensure that dict_table_t::def_trx_id will be recovered correctly in case the server is killed before ddl_log_complete(), we will block the purge of any history in SYS_TABLES, SYS_INDEXES, SYS_COLUMNS between ha_innobase::commit_inplace_alter_table(commit=true) (purge_sys.stop_SYS()) and purge_sys.resume_SYS(). The completion callback purge_sys.resume_SYS() must be between ddl_log_complete() and MDL release. -------- MyRocks support for atomic ALTER TABLE (Implemented by Sergui Petrunia) Implement these SE API functions: - ha_rocksdb::table_version() - hton->check_version = rocksdb_check_versionMyRocks data dictionary now stores table version for each table. (Absence of table version record is interpreted as table_version=0, that is, which means no upgrade changes are needed) - For inplace alter table of a partitioned table, call the underlying handlerton when checking if the table is ok. This assumes that the partition engine commits all changes at once.
2021-03-18 11:41:08 +01:00
LEX_CSTRING tmp_name; /* frm file or temporary file name */
LEX_CSTRING extra_name; /* Backup table name */
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
uchar uuid[MY_UUID_SIZE]; // UUID for new frm file
ulonglong xid; // Xid stored in the binary log
/*
unique_id can be used to store a unique number to check current state.
Currently it is used to store new size of frm file, link to another ddl log
entry or store an a uniq version for a storage engine in alter table.
For execute entries this is reused as an execute counter to ensure we
don't repeat an entry too many times if executing the entry fails.
*/
ulonglong unique_id;
uint next_entry;
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
uint entry_pos; // Set by write_dll_log_entry()
uint16 flags; // Flags unique for each command
enum ddl_log_entry_code entry_type; // Set automatically
enum ddl_log_action_code action_type;
/*
Most actions have only one phase. REPLACE does however have two
phases. The first phase removes the file with the new name if
there was one there before and the second phase renames the
old name to the new name.
*/
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
uchar phase; // set automatically
} DDL_LOG_ENTRY;
typedef struct st_ddl_log_memory_entry
{
uint entry_pos;
struct st_ddl_log_memory_entry *next_log_entry;
struct st_ddl_log_memory_entry *prev_log_entry;
struct st_ddl_log_memory_entry *next_active_log_entry;
} DDL_LOG_MEMORY_ENTRY;
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
/*
State of the ddl log during execution of a DDL.
A ddl log state has one execute entry (main entry pointing to the first
action entry) and many 'action entries' linked in a list in the order
they should be executed.
One recovery the log is parsed and all execute entries will be executed.
All entries are stored as separate blocks in the ddl recovery file.
*/
typedef struct st_ddl_log_state
{
/* List of ddl log entries */
DDL_LOG_MEMORY_ENTRY *list;
/* One execute entry per list */
DDL_LOG_MEMORY_ENTRY *execute_entry;
MDEV-25180 Atomic ALTER TABLE MDEV-25604 Atomic DDL: Binlog event written upon recovery does not have default database The purpose of this task is to ensure that ALTER TABLE is atomic even if the MariaDB server would be killed at any point of the alter table. This means that either the ALTER TABLE succeeds (including that triggers, the status tables and the binary log are updated) or things should be reverted to their original state. If the server crashes before the new version is fully up to date and commited, it will revert to the original table and remove all temporary files and tables. If the new version is commited, crash recovery will use the new version, and update triggers, the status tables and the binary log. The one execption is ALTER TABLE .. RENAME .. where no changes are done to table definition. This one will work as RENAME and roll back unless the whole statement completed, including updating the binary log (if enabled). Other changes: - Added handlerton->check_version() function to allow the ddl recovery code to check, in case of inplace alter table, if the table in the storage engine is of the new or old version. - Added handler->table_version() so that an engine can report the current version of the table. This should be changed each time the table definition changes. - Added ha_signal_ddl_recovery_done() and handlerton::signal_ddl_recovery_done() to inform all handlers when ddl recovery has been done. (Needed by InnoDB). - Added handlerton call inplace_alter_table_committed, to signal engine that ddl_log has been closed for the alter table query. - Added new handerton flag HTON_REQUIRES_NOTIFY_TABLEDEF_CHANGED_AFTER_COMMIT to signal when we should call hton->notify_tabledef_changed() during mysql_inplace_alter_table. This was required as MyRocks and InnoDB needed the call at different times. - Added function server_uuid_value() to be able to generate a temporary xid when ddl recovery writes the query to the binary log. This is needed to be able to handle crashes during ddl log recovery. - Moved freeing of the frm definition to end of mysql_alter_table() to remove duplicate code and have a common exit strategy. ------- InnoDB part of atomic ALTER TABLE (Implemented by Marko Mäkelä) innodb_check_version(): Compare the saved dict_table_t::def_trx_id to determine whether an ALTER TABLE operation was committed. We must correctly recover dict_table_t::def_trx_id for this to work. Before purge removes any trace of DB_TRX_ID from system tables, it will make an effort to load the user table into the cache, so that the dict_table_t::def_trx_id can be recovered. ha_innobase::table_version(): return garbage, or the trx_id that would be used for committing an ALTER TABLE operation. In InnoDB, table names starting with #sql-ib will remain special: they will be dropped on startup. This may be revisited later in MDEV-18518 when we implement proper undo logging and rollback for creating or dropping multiple tables in a transaction. Table names starting with #sql will retain some special meaning: dict_table_t::parse_name() will not consider such names for MDL acquisition, and dict_table_rename_in_cache() will treat such names specially when handling FOREIGN KEY constraints. Simplify InnoDB DROP INDEX. Prevent purge wakeup To ensure that dict_table_t::def_trx_id will be recovered correctly in case the server is killed before ddl_log_complete(), we will block the purge of any history in SYS_TABLES, SYS_INDEXES, SYS_COLUMNS between ha_innobase::commit_inplace_alter_table(commit=true) (purge_sys.stop_SYS()) and purge_sys.resume_SYS(). The completion callback purge_sys.resume_SYS() must be between ddl_log_complete() and MDL release. -------- MyRocks support for atomic ALTER TABLE (Implemented by Sergui Petrunia) Implement these SE API functions: - ha_rocksdb::table_version() - hton->check_version = rocksdb_check_versionMyRocks data dictionary now stores table version for each table. (Absence of table version record is interpreted as table_version=0, that is, which means no upgrade changes are needed) - For inplace alter table of a partitioned table, call the underlying handlerton when checking if the table is ok. This assumes that the partition engine commits all changes at once.
2021-03-18 11:41:08 +01:00
/*
Entry used for PHASE updates. Normally same as first in 'list', but in
case of a query log event, this points to the main event.
*/
DDL_LOG_MEMORY_ENTRY *main_entry;
uint16 flags; /* Cache for flags */
bool is_active() { return list != 0; }
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
} DDL_LOG_STATE;
/* These functions are for recovery */
bool ddl_log_initialize();
void ddl_log_release();
bool ddl_log_close_binlogged_events(HASH *xids);
int ddl_log_execute_recovery();
/* functions for updating the ddl log */
bool ddl_log_write_entry(DDL_LOG_ENTRY *ddl_log_entry,
DDL_LOG_MEMORY_ENTRY **active_entry);
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
2021-09-27 14:53:52 +02:00
bool ddl_log_write_execute_entry(uint first_entry, uint cond_entry,
DDL_LOG_MEMORY_ENTRY** active_entry);
Revert MDEV-25292 Atomic CREATE OR REPLACE TABLE Specifically: Revert "MDEV-29664 Assertion `!n_mysql_tables_in_use' failed in innobase_close_connection" This reverts commit ba875e939619baefb08936863a889830f595e426. Revert "MDEV-29620 Assertion `next_insert_id == 0' failed in handler::ha_external_lock" This reverts commit aa08a7442a5141750de96353132c7799e10c9302. Revert "MDEV-29628 Memory leak after CREATE OR REPLACE with foreign key" This reverts commit c579d66ba689c8a20a951de5e6be2e1bc255ba7f. Revert "MDEV-29609 create_not_windows test fails with different result" This reverts commit cb583b2f1bf1c910e82c7caee4a53e0f3ff954ba. Revert "MDEV-29544 SIGSEGV in HA_CREATE_INFO::finalize_locked_tables" This reverts commit dcd66c3814991e686615eac2e65d01a7fa80cb66. Revert "MDEV-28933 CREATE OR REPLACE fails to recreate same constraint name" This reverts commit cf6c5176328c8fbfadac80f337ef285732cc8d06. Revert "MDEV-28933 Moved RENAME_CONSTRAINT_IDS to include/sql_funcs.h" This reverts commit f1e1c1335b506905244502ee8584736a17a6ecd0. Revert "MDEV-28956 Locking is broken if CREATE OR REPLACE fails under LOCK TABLES" This reverts commit a228ec80e30cae59c2057b786fd3559171ce794f. Revert "MDEV-25292 gcol.gcol_bugfixes --ps fix" This reverts commit 24fff8267d0722b20ca780cf144ce27560df1394. Revert "MDEV-25292 Disable atomic replace for slave-generated or-replace" This reverts commit 2af15914cb3ae204d3b413995a96c7722ad60d93. Revert "MDEV-25292 backup_log improved" This reverts commit 34398a20b5829f0508c78bc8f765f7bef6e0ad48. Revert "MDEV-25292 Atomic CREATE OR REPLACE TABLE" This reverts commit 93c8252f02faa8ad8dc5f005e52f1990c29d4a0d. Revert "MDEV-25292 Table_name class for (db, table_name, alias)" This reverts commit d145dda9c7b6db394edf7150cc255343a4a7b116. Revert "MDEV-25292 ha_table_exists() cleanup and improvement" This reverts commit 409b8a86de45efc96377c618a99cff1301523150. Revert "MDEV-25292 Cleanups" This reverts commit 595dad83ad3c23879c59deec1173ea16b47871ed. Revert "MDEV-25292 Refactoring: moved select_field_count into Alter_info." This reverts commit f02af1d2295f108d96671b5a4f731ec998002dd5.
2022-10-27 22:18:51 +02:00
inline
bool ddl_log_write_execute_entry(uint first_entry,
DDL_LOG_MEMORY_ENTRY **active_entry)
{
return ddl_log_write_execute_entry(first_entry, 0, active_entry);
}
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
bool ddl_log_disable_execute_entry(DDL_LOG_MEMORY_ENTRY **active_entry);
void ddl_log_complete(DDL_LOG_STATE *ddl_log_state);
bool ddl_log_revert(THD *thd, DDL_LOG_STATE *ddl_log_state);
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
bool ddl_log_update_phase(DDL_LOG_STATE *entry, uchar phase);
MDEV-25180 Atomic ALTER TABLE MDEV-25604 Atomic DDL: Binlog event written upon recovery does not have default database The purpose of this task is to ensure that ALTER TABLE is atomic even if the MariaDB server would be killed at any point of the alter table. This means that either the ALTER TABLE succeeds (including that triggers, the status tables and the binary log are updated) or things should be reverted to their original state. If the server crashes before the new version is fully up to date and commited, it will revert to the original table and remove all temporary files and tables. If the new version is commited, crash recovery will use the new version, and update triggers, the status tables and the binary log. The one execption is ALTER TABLE .. RENAME .. where no changes are done to table definition. This one will work as RENAME and roll back unless the whole statement completed, including updating the binary log (if enabled). Other changes: - Added handlerton->check_version() function to allow the ddl recovery code to check, in case of inplace alter table, if the table in the storage engine is of the new or old version. - Added handler->table_version() so that an engine can report the current version of the table. This should be changed each time the table definition changes. - Added ha_signal_ddl_recovery_done() and handlerton::signal_ddl_recovery_done() to inform all handlers when ddl recovery has been done. (Needed by InnoDB). - Added handlerton call inplace_alter_table_committed, to signal engine that ddl_log has been closed for the alter table query. - Added new handerton flag HTON_REQUIRES_NOTIFY_TABLEDEF_CHANGED_AFTER_COMMIT to signal when we should call hton->notify_tabledef_changed() during mysql_inplace_alter_table. This was required as MyRocks and InnoDB needed the call at different times. - Added function server_uuid_value() to be able to generate a temporary xid when ddl recovery writes the query to the binary log. This is needed to be able to handle crashes during ddl log recovery. - Moved freeing of the frm definition to end of mysql_alter_table() to remove duplicate code and have a common exit strategy. ------- InnoDB part of atomic ALTER TABLE (Implemented by Marko Mäkelä) innodb_check_version(): Compare the saved dict_table_t::def_trx_id to determine whether an ALTER TABLE operation was committed. We must correctly recover dict_table_t::def_trx_id for this to work. Before purge removes any trace of DB_TRX_ID from system tables, it will make an effort to load the user table into the cache, so that the dict_table_t::def_trx_id can be recovered. ha_innobase::table_version(): return garbage, or the trx_id that would be used for committing an ALTER TABLE operation. In InnoDB, table names starting with #sql-ib will remain special: they will be dropped on startup. This may be revisited later in MDEV-18518 when we implement proper undo logging and rollback for creating or dropping multiple tables in a transaction. Table names starting with #sql will retain some special meaning: dict_table_t::parse_name() will not consider such names for MDL acquisition, and dict_table_rename_in_cache() will treat such names specially when handling FOREIGN KEY constraints. Simplify InnoDB DROP INDEX. Prevent purge wakeup To ensure that dict_table_t::def_trx_id will be recovered correctly in case the server is killed before ddl_log_complete(), we will block the purge of any history in SYS_TABLES, SYS_INDEXES, SYS_COLUMNS between ha_innobase::commit_inplace_alter_table(commit=true) (purge_sys.stop_SYS()) and purge_sys.resume_SYS(). The completion callback purge_sys.resume_SYS() must be between ddl_log_complete() and MDL release. -------- MyRocks support for atomic ALTER TABLE (Implemented by Sergui Petrunia) Implement these SE API functions: - ha_rocksdb::table_version() - hton->check_version = rocksdb_check_versionMyRocks data dictionary now stores table version for each table. (Absence of table version record is interpreted as table_version=0, that is, which means no upgrade changes are needed) - For inplace alter table of a partitioned table, call the underlying handlerton when checking if the table is ok. This assumes that the partition engine commits all changes at once.
2021-03-18 11:41:08 +01:00
bool ddl_log_add_flag(DDL_LOG_STATE *entry, uint16 flag);
bool ddl_log_update_unique_id(DDL_LOG_STATE *state, ulonglong id);
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
bool ddl_log_update_xid(DDL_LOG_STATE *state, ulonglong xid);
bool ddl_log_disable_entry(DDL_LOG_STATE *state);
bool ddl_log_increment_phase(uint entry_pos);
void ddl_log_release_memory_entry(DDL_LOG_MEMORY_ENTRY *log_entry);
bool ddl_log_sync();
bool ddl_log_execute_entry(THD *thd, uint first_entry);
void ddl_log_add_entry(DDL_LOG_STATE *state, DDL_LOG_MEMORY_ENTRY *log_entry);
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
void ddl_log_release_entries(DDL_LOG_STATE *ddl_log_state);
bool ddl_log_rename_table(DDL_LOG_STATE *ddl_state,
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
handlerton *hton,
const LEX_CSTRING *org_db,
const LEX_CSTRING *org_alias,
const LEX_CSTRING *new_db,
Revert MDEV-25292 Atomic CREATE OR REPLACE TABLE Specifically: Revert "MDEV-29664 Assertion `!n_mysql_tables_in_use' failed in innobase_close_connection" This reverts commit ba875e939619baefb08936863a889830f595e426. Revert "MDEV-29620 Assertion `next_insert_id == 0' failed in handler::ha_external_lock" This reverts commit aa08a7442a5141750de96353132c7799e10c9302. Revert "MDEV-29628 Memory leak after CREATE OR REPLACE with foreign key" This reverts commit c579d66ba689c8a20a951de5e6be2e1bc255ba7f. Revert "MDEV-29609 create_not_windows test fails with different result" This reverts commit cb583b2f1bf1c910e82c7caee4a53e0f3ff954ba. Revert "MDEV-29544 SIGSEGV in HA_CREATE_INFO::finalize_locked_tables" This reverts commit dcd66c3814991e686615eac2e65d01a7fa80cb66. Revert "MDEV-28933 CREATE OR REPLACE fails to recreate same constraint name" This reverts commit cf6c5176328c8fbfadac80f337ef285732cc8d06. Revert "MDEV-28933 Moved RENAME_CONSTRAINT_IDS to include/sql_funcs.h" This reverts commit f1e1c1335b506905244502ee8584736a17a6ecd0. Revert "MDEV-28956 Locking is broken if CREATE OR REPLACE fails under LOCK TABLES" This reverts commit a228ec80e30cae59c2057b786fd3559171ce794f. Revert "MDEV-25292 gcol.gcol_bugfixes --ps fix" This reverts commit 24fff8267d0722b20ca780cf144ce27560df1394. Revert "MDEV-25292 Disable atomic replace for slave-generated or-replace" This reverts commit 2af15914cb3ae204d3b413995a96c7722ad60d93. Revert "MDEV-25292 backup_log improved" This reverts commit 34398a20b5829f0508c78bc8f765f7bef6e0ad48. Revert "MDEV-25292 Atomic CREATE OR REPLACE TABLE" This reverts commit 93c8252f02faa8ad8dc5f005e52f1990c29d4a0d. Revert "MDEV-25292 Table_name class for (db, table_name, alias)" This reverts commit d145dda9c7b6db394edf7150cc255343a4a7b116. Revert "MDEV-25292 ha_table_exists() cleanup and improvement" This reverts commit 409b8a86de45efc96377c618a99cff1301523150. Revert "MDEV-25292 Cleanups" This reverts commit 595dad83ad3c23879c59deec1173ea16b47871ed. Revert "MDEV-25292 Refactoring: moved select_field_count into Alter_info." This reverts commit f02af1d2295f108d96671b5a4f731ec998002dd5.
2022-10-27 22:18:51 +02:00
const LEX_CSTRING *new_alias);
bool ddl_log_rename_view(DDL_LOG_STATE *ddl_state,
MDEV-23842 Atomic RENAME TABLE - Major rewrite of ddl_log.cc and ddl_log.h - ddl_log.cc described in the beginning how the recovery works. - ddl_log.log has unique signature and is dynamic. It's easy to add more information to the header and other ddl blocks while still being able to execute old ddl entries. - IO_SIZE for ddl blocks is now dynamic. Can be changed without affecting recovery of old logs. - Code is more modular and is now usable outside of partition handling. - Renamed log file to dll_recovery.log and added option --log-ddl-recovery to allow one to specify the path & filename. - Added ddl_log_entry_phase[], number of phases for each DDL action, which allowed me to greatly simply set_global_from_ddl_log_entry() - Changed how strings are stored in log entries, which allows us to store much more information in a log entry. - ddl log is now always created at start and deleted on normal shutdown. This simplices things notable. - Added probes debug_crash_here() and debug_simulate_error() to simply crash testing and allow crash after a given number of times a probe is executed. See comments in debug_sync.cc and rename_table.test for how this can be used. - Reverting failed table and view renames is done trough the ddl log. This ensures that the ddl log is tested also outside of recovery. - Added helper function 'handler::needs_lower_case_filenames()' - Extend binary log with Q_XID events. ddl log handling is using this to check if a ddl log entry was logged to the binary log (if yes, it will be deleted from the log during ddl_log_close_binlogged_events() - If a DDL entry fails 3 time, disable it. This is to ensure that if we have a crash in ddl recovery code the server will not get stuck in a forever crash-restart-crash loop. mysqltest.cc changes: - --die will now replace $variables with their values - $error will contain the error of the last failed statement storage engine changes: - maria_rename() was changed to be more robust against crashes during rename.
2020-10-15 01:25:57 +02:00
const LEX_CSTRING *org_db,
const LEX_CSTRING *org_alias,
const LEX_CSTRING *new_db,
const LEX_CSTRING *new_alias);
bool ddl_log_drop_table_init(DDL_LOG_STATE *ddl_state,
MDEV-24408 Crash-safe DROP DATABASE Description of how DROP DATABASE works after this patch - Collect list of tables - DDL log tables as they are dropped - DDL log drop database - Delete db.opt - Delete data directory - Log either DROP TABLE or DROP DATABASE to binary log - De active ddl log entry This is in line of how things where before (minus ddl logging) except that we delete db.opt file last to not loose it if DROP DATABASE fails. On recovery we have to ensure that all dropped tables are logged in binary log and that they are properly dropped (as with atomic drop table). No new tables be dropped as part of recovery. Recovery of active drop database ddl log entry: - If drop database was logged to ddl log but was not found in the binary log: - drop the db.opt file and database directory. - Log DROP DATABASE to binary log - If drop database was not logged to ddl log - Update binary log with DROP TABLE of the dropped tables. If table list is longer than max_allowed_packet, then the query will be split into multiple DROP TABLE/VIEW queries. Other things: - Added DDL_LOG_STATE and 'current database' as arguments to mysql_rm_table_no_locks(). This was needed to be able to combine ddl logging of DROP DATABASE and DROP TABLE and make the generated DROP TABLE statements shorter. - To make the DROP TABLE statement created by ddl log shorter, I changed the binlogged query to use current directory and omit the directory part for all tables in the current directory. - Merged some DROP TABLE and DROP VIEW code in ddl logger. This was done to be able get separate DROP VIEW and DROP TABLE statements in the binary log. - Added a 'recovery_state' variable to remember the state of dropped tables and views. - Moved out code that drops database objects (stored procedures) from mysql_rm_db_internal() to drop_database_objects() for better code reuse. - Made mysql_rm_db_internal() global so that could be used by the ddl recovery code.
2020-12-20 16:44:11 +01:00
const LEX_CSTRING *db,
const LEX_CSTRING *comment);
bool ddl_log_drop_view_init(DDL_LOG_STATE *ddl_state,
MDEV-24408 Crash-safe DROP DATABASE Description of how DROP DATABASE works after this patch - Collect list of tables - DDL log tables as they are dropped - DDL log drop database - Delete db.opt - Delete data directory - Log either DROP TABLE or DROP DATABASE to binary log - De active ddl log entry This is in line of how things where before (minus ddl logging) except that we delete db.opt file last to not loose it if DROP DATABASE fails. On recovery we have to ensure that all dropped tables are logged in binary log and that they are properly dropped (as with atomic drop table). No new tables be dropped as part of recovery. Recovery of active drop database ddl log entry: - If drop database was logged to ddl log but was not found in the binary log: - drop the db.opt file and database directory. - Log DROP DATABASE to binary log - If drop database was not logged to ddl log - Update binary log with DROP TABLE of the dropped tables. If table list is longer than max_allowed_packet, then the query will be split into multiple DROP TABLE/VIEW queries. Other things: - Added DDL_LOG_STATE and 'current database' as arguments to mysql_rm_table_no_locks(). This was needed to be able to combine ddl logging of DROP DATABASE and DROP TABLE and make the generated DROP TABLE statements shorter. - To make the DROP TABLE statement created by ddl log shorter, I changed the binlogged query to use current directory and omit the directory part for all tables in the current directory. - Merged some DROP TABLE and DROP VIEW code in ddl logger. This was done to be able get separate DROP VIEW and DROP TABLE statements in the binary log. - Added a 'recovery_state' variable to remember the state of dropped tables and views. - Moved out code that drops database objects (stored procedures) from mysql_rm_db_internal() to drop_database_objects() for better code reuse. - Made mysql_rm_db_internal() global so that could be used by the ddl recovery code.
2020-12-20 16:44:11 +01:00
const LEX_CSTRING *db);
bool ddl_log_drop_table(DDL_LOG_STATE *ddl_state,
handlerton *hton,
const LEX_CSTRING *path,
const LEX_CSTRING *db,
Revert MDEV-25292 Atomic CREATE OR REPLACE TABLE Specifically: Revert "MDEV-29664 Assertion `!n_mysql_tables_in_use' failed in innobase_close_connection" This reverts commit ba875e939619baefb08936863a889830f595e426. Revert "MDEV-29620 Assertion `next_insert_id == 0' failed in handler::ha_external_lock" This reverts commit aa08a7442a5141750de96353132c7799e10c9302. Revert "MDEV-29628 Memory leak after CREATE OR REPLACE with foreign key" This reverts commit c579d66ba689c8a20a951de5e6be2e1bc255ba7f. Revert "MDEV-29609 create_not_windows test fails with different result" This reverts commit cb583b2f1bf1c910e82c7caee4a53e0f3ff954ba. Revert "MDEV-29544 SIGSEGV in HA_CREATE_INFO::finalize_locked_tables" This reverts commit dcd66c3814991e686615eac2e65d01a7fa80cb66. Revert "MDEV-28933 CREATE OR REPLACE fails to recreate same constraint name" This reverts commit cf6c5176328c8fbfadac80f337ef285732cc8d06. Revert "MDEV-28933 Moved RENAME_CONSTRAINT_IDS to include/sql_funcs.h" This reverts commit f1e1c1335b506905244502ee8584736a17a6ecd0. Revert "MDEV-28956 Locking is broken if CREATE OR REPLACE fails under LOCK TABLES" This reverts commit a228ec80e30cae59c2057b786fd3559171ce794f. Revert "MDEV-25292 gcol.gcol_bugfixes --ps fix" This reverts commit 24fff8267d0722b20ca780cf144ce27560df1394. Revert "MDEV-25292 Disable atomic replace for slave-generated or-replace" This reverts commit 2af15914cb3ae204d3b413995a96c7722ad60d93. Revert "MDEV-25292 backup_log improved" This reverts commit 34398a20b5829f0508c78bc8f765f7bef6e0ad48. Revert "MDEV-25292 Atomic CREATE OR REPLACE TABLE" This reverts commit 93c8252f02faa8ad8dc5f005e52f1990c29d4a0d. Revert "MDEV-25292 Table_name class for (db, table_name, alias)" This reverts commit d145dda9c7b6db394edf7150cc255343a4a7b116. Revert "MDEV-25292 ha_table_exists() cleanup and improvement" This reverts commit 409b8a86de45efc96377c618a99cff1301523150. Revert "MDEV-25292 Cleanups" This reverts commit 595dad83ad3c23879c59deec1173ea16b47871ed. Revert "MDEV-25292 Refactoring: moved select_field_count into Alter_info." This reverts commit f02af1d2295f108d96671b5a4f731ec998002dd5.
2022-10-27 22:18:51 +02:00
const LEX_CSTRING *table);
bool ddl_log_drop_view(DDL_LOG_STATE *ddl_state,
const LEX_CSTRING *path,
const LEX_CSTRING *db,
const LEX_CSTRING *table);
bool ddl_log_drop_trigger(DDL_LOG_STATE *ddl_state,
const LEX_CSTRING *db,
const LEX_CSTRING *table,
const LEX_CSTRING *trigger_name,
const LEX_CSTRING *query);
bool ddl_log_drop_view(DDL_LOG_STATE *ddl_state,
MDEV-24408 Crash-safe DROP DATABASE Description of how DROP DATABASE works after this patch - Collect list of tables - DDL log tables as they are dropped - DDL log drop database - Delete db.opt - Delete data directory - Log either DROP TABLE or DROP DATABASE to binary log - De active ddl log entry This is in line of how things where before (minus ddl logging) except that we delete db.opt file last to not loose it if DROP DATABASE fails. On recovery we have to ensure that all dropped tables are logged in binary log and that they are properly dropped (as with atomic drop table). No new tables be dropped as part of recovery. Recovery of active drop database ddl log entry: - If drop database was logged to ddl log but was not found in the binary log: - drop the db.opt file and database directory. - Log DROP DATABASE to binary log - If drop database was not logged to ddl log - Update binary log with DROP TABLE of the dropped tables. If table list is longer than max_allowed_packet, then the query will be split into multiple DROP TABLE/VIEW queries. Other things: - Added DDL_LOG_STATE and 'current database' as arguments to mysql_rm_table_no_locks(). This was needed to be able to combine ddl logging of DROP DATABASE and DROP TABLE and make the generated DROP TABLE statements shorter. - To make the DROP TABLE statement created by ddl log shorter, I changed the binlogged query to use current directory and omit the directory part for all tables in the current directory. - Merged some DROP TABLE and DROP VIEW code in ddl logger. This was done to be able get separate DROP VIEW and DROP TABLE statements in the binary log. - Added a 'recovery_state' variable to remember the state of dropped tables and views. - Moved out code that drops database objects (stored procedures) from mysql_rm_db_internal() to drop_database_objects() for better code reuse. - Made mysql_rm_db_internal() global so that could be used by the ddl recovery code.
2020-12-20 16:44:11 +01:00
const LEX_CSTRING *path,
const LEX_CSTRING *db,
const LEX_CSTRING *table);
bool ddl_log_drop_db(DDL_LOG_STATE *ddl_state,
MDEV-24408 Crash-safe DROP DATABASE Description of how DROP DATABASE works after this patch - Collect list of tables - DDL log tables as they are dropped - DDL log drop database - Delete db.opt - Delete data directory - Log either DROP TABLE or DROP DATABASE to binary log - De active ddl log entry This is in line of how things where before (minus ddl logging) except that we delete db.opt file last to not loose it if DROP DATABASE fails. On recovery we have to ensure that all dropped tables are logged in binary log and that they are properly dropped (as with atomic drop table). No new tables be dropped as part of recovery. Recovery of active drop database ddl log entry: - If drop database was logged to ddl log but was not found in the binary log: - drop the db.opt file and database directory. - Log DROP DATABASE to binary log - If drop database was not logged to ddl log - Update binary log with DROP TABLE of the dropped tables. If table list is longer than max_allowed_packet, then the query will be split into multiple DROP TABLE/VIEW queries. Other things: - Added DDL_LOG_STATE and 'current database' as arguments to mysql_rm_table_no_locks(). This was needed to be able to combine ddl logging of DROP DATABASE and DROP TABLE and make the generated DROP TABLE statements shorter. - To make the DROP TABLE statement created by ddl log shorter, I changed the binlogged query to use current directory and omit the directory part for all tables in the current directory. - Merged some DROP TABLE and DROP VIEW code in ddl logger. This was done to be able get separate DROP VIEW and DROP TABLE statements in the binary log. - Added a 'recovery_state' variable to remember the state of dropped tables and views. - Moved out code that drops database objects (stored procedures) from mysql_rm_db_internal() to drop_database_objects() for better code reuse. - Made mysql_rm_db_internal() global so that could be used by the ddl recovery code.
2020-12-20 16:44:11 +01:00
const LEX_CSTRING *db, const LEX_CSTRING *path);
bool ddl_log_create_table(DDL_LOG_STATE *ddl_state,
handlerton *hton,
const LEX_CSTRING *path,
const LEX_CSTRING *db,
const LEX_CSTRING *table,
bool only_frm);
bool ddl_log_create_view(DDL_LOG_STATE *ddl_state,
const LEX_CSTRING *path,
enum_ddl_log_create_view_phase phase);
bool ddl_log_delete_tmp_file(DDL_LOG_STATE *ddl_state,
const LEX_CSTRING *path,
DDL_LOG_STATE *depending_state);
bool ddl_log_create_trigger(DDL_LOG_STATE *ddl_state,
const LEX_CSTRING *db, const LEX_CSTRING *table,
const LEX_CSTRING *trigger_name,
enum_ddl_log_create_trigger_phase phase);
bool ddl_log_alter_table(DDL_LOG_STATE *ddl_state,
MDEV-25180 Atomic ALTER TABLE MDEV-25604 Atomic DDL: Binlog event written upon recovery does not have default database The purpose of this task is to ensure that ALTER TABLE is atomic even if the MariaDB server would be killed at any point of the alter table. This means that either the ALTER TABLE succeeds (including that triggers, the status tables and the binary log are updated) or things should be reverted to their original state. If the server crashes before the new version is fully up to date and commited, it will revert to the original table and remove all temporary files and tables. If the new version is commited, crash recovery will use the new version, and update triggers, the status tables and the binary log. The one execption is ALTER TABLE .. RENAME .. where no changes are done to table definition. This one will work as RENAME and roll back unless the whole statement completed, including updating the binary log (if enabled). Other changes: - Added handlerton->check_version() function to allow the ddl recovery code to check, in case of inplace alter table, if the table in the storage engine is of the new or old version. - Added handler->table_version() so that an engine can report the current version of the table. This should be changed each time the table definition changes. - Added ha_signal_ddl_recovery_done() and handlerton::signal_ddl_recovery_done() to inform all handlers when ddl recovery has been done. (Needed by InnoDB). - Added handlerton call inplace_alter_table_committed, to signal engine that ddl_log has been closed for the alter table query. - Added new handerton flag HTON_REQUIRES_NOTIFY_TABLEDEF_CHANGED_AFTER_COMMIT to signal when we should call hton->notify_tabledef_changed() during mysql_inplace_alter_table. This was required as MyRocks and InnoDB needed the call at different times. - Added function server_uuid_value() to be able to generate a temporary xid when ddl recovery writes the query to the binary log. This is needed to be able to handle crashes during ddl log recovery. - Moved freeing of the frm definition to end of mysql_alter_table() to remove duplicate code and have a common exit strategy. ------- InnoDB part of atomic ALTER TABLE (Implemented by Marko Mäkelä) innodb_check_version(): Compare the saved dict_table_t::def_trx_id to determine whether an ALTER TABLE operation was committed. We must correctly recover dict_table_t::def_trx_id for this to work. Before purge removes any trace of DB_TRX_ID from system tables, it will make an effort to load the user table into the cache, so that the dict_table_t::def_trx_id can be recovered. ha_innobase::table_version(): return garbage, or the trx_id that would be used for committing an ALTER TABLE operation. In InnoDB, table names starting with #sql-ib will remain special: they will be dropped on startup. This may be revisited later in MDEV-18518 when we implement proper undo logging and rollback for creating or dropping multiple tables in a transaction. Table names starting with #sql will retain some special meaning: dict_table_t::parse_name() will not consider such names for MDL acquisition, and dict_table_rename_in_cache() will treat such names specially when handling FOREIGN KEY constraints. Simplify InnoDB DROP INDEX. Prevent purge wakeup To ensure that dict_table_t::def_trx_id will be recovered correctly in case the server is killed before ddl_log_complete(), we will block the purge of any history in SYS_TABLES, SYS_INDEXES, SYS_COLUMNS between ha_innobase::commit_inplace_alter_table(commit=true) (purge_sys.stop_SYS()) and purge_sys.resume_SYS(). The completion callback purge_sys.resume_SYS() must be between ddl_log_complete() and MDL release. -------- MyRocks support for atomic ALTER TABLE (Implemented by Sergui Petrunia) Implement these SE API functions: - ha_rocksdb::table_version() - hton->check_version = rocksdb_check_versionMyRocks data dictionary now stores table version for each table. (Absence of table version record is interpreted as table_version=0, that is, which means no upgrade changes are needed) - For inplace alter table of a partitioned table, call the underlying handlerton when checking if the table is ok. This assumes that the partition engine commits all changes at once.
2021-03-18 11:41:08 +01:00
handlerton *org_hton,
const LEX_CSTRING *db, const LEX_CSTRING *table,
handlerton *new_hton,
handlerton *partition_underlying_hton,
const LEX_CSTRING *new_db,
const LEX_CSTRING *new_table,
const LEX_CSTRING *frm_path,
const LEX_CSTRING *backup_table_name,
const LEX_CUSTRING *version,
ulonglong table_version,
bool is_renamed);
bool ddl_log_store_query(THD *thd, DDL_LOG_STATE *ddl_log_state,
const char *query, size_t length);
2021-09-27 14:53:52 +02:00
bool ddl_log_delete_frm(DDL_LOG_STATE *ddl_state, const char *to_path);
extern mysql_mutex_t LOCK_gdl;
#endif /* DDL_LOG_INCLUDED */