Commit graph

733 commits

Author SHA1 Message Date
Sergei Golubchik
7842cab8c0 MDEV-34318 post-merge fix 2024-10-17 10:08:24 +02:00
Marko Mäkelä
63913ce5af Merge 10.6 into 10.11 2024-10-03 10:55:08 +03:00
Marko Mäkelä
7e0afb1c73 Merge 10.5 into 10.6 2024-10-03 09:31:39 +03:00
Lena Startseva
0a5e4a0191 MDEV-31005: Make working cursor-protocol
Updated tests: cases with bugs or which cannot be run
with the cursor-protocol were excluded with
"--disable_cursor_protocol"/"--enable_cursor_protocol"

Fix for v.10.5
2024-09-18 18:39:26 +07:00
Oleksandr Byelkin
0fe39d368a Merge branch '10.6' into 10.11 2024-07-22 15:14:50 +02:00
Sergei Golubchik
a6b2f820e0 Merge branch '10.6' into 10.11 2024-05-10 20:02:18 +02:00
Aleksey Midenkov
c477697422 MDEV-29872 MSAN/Valgrind uninitialised value errors in TABLE::vers_switch_partition
Delayed_insert has its own THD (initialized at mysql_insert()) and
hence its own LEX. Delayed_insert initalizes a very few parameters for
LEX and 'duplicates' is not in this list. Now we copy this missing
parameter from parser LEX (as well as sql_command).
2024-04-02 00:11:35 +03:00
Aleksey Midenkov
d966e55c0a MDEV-31903 Server crashes in _ma_reset_history upon UNLOCK table with auto-create history partitions
When INSERT does auto-create for t1 all its handler instances are
closed by alter_close_table(). At this time down the stack
maria_close() clears share->state_history. Later when we unlock the
tables Aria transaction manager accesses old share instance (the one
before t1 was closed) and tries to reset its state_history.

The problem is maria_close() didn't remove table from transaction's
list (used_tables). The fix does _ma_remove_table_from_trnman() which
is triggered by HA_EXTRA_PREPARE_FOR_RENAME.
2024-04-02 00:11:34 +03:00
Marko Mäkelä
64cce8d5bf Merge 10.6 into 10.11 2024-02-14 16:12:53 +02:00
Marko Mäkelä
691f923906 Merge 10.5 into 10.6 2024-02-13 20:42:59 +02:00
Marko Mäkelä
b770633e07 Merge 10.4 into 10.5 2024-02-13 14:25:21 +02:00
Marko Mäkelä
ca88eac835 MDEV-30528 CREATE FULLTEXT INDEX assertion failure WITH SYSTEM VERSIONING
ha_innobase::check_if_supported_inplace_alter(): Require ALGORITHM=COPY
when creating a FULLTEXT INDEX on a versioned table.

row_merge_buf_add(), row_merge_read_clustered_index(): Remove the parameter
or local variable history_fts that had been added in the attempt to fix
MDEV-25004.

Reviewed by: Thirunarayanan Balathandayuthapani
Tested by: Matthias Leich
2024-02-12 16:52:55 +01:00
Sergei Golubchik
fd0b47f9d6 Merge branch '10.6' into 10.11 2023-12-18 11:19:04 +01:00
Sergei Golubchik
e95bba9c58 Merge branch '10.5' into 10.6 2023-12-17 11:20:43 +01:00
Sergei Golubchik
98a39b0c91 Merge branch '10.4' into 10.5 2023-12-02 01:02:50 +01:00
Marko Mäkelä
f87c7d1772 Merge 10.6 into 10.11 2023-11-21 12:47:51 +02:00
Marko Mäkelä
4c16ec3e77 MDEV-32050 fixup: Stabilize tests
In any test that uses wait_all_purged.inc, ensure that InnoDB tables
will be created without persistent statistics.

This is a follow-up to commit cd04673a17
after a similar failure was observed in the innodb_zip.blob test.
2023-11-21 12:42:00 +02:00
Aleksey Midenkov
74883f5e2f MDEV-32082 Server crash in find_field_in_table
Attempt to resolve FOR SYSTEM_TIME expression as field for derived
table is done before derived table is fully prepared, so we fail on
assertion that table_list->table is missing.

Actually Vers_history_point::resolve_unit() is done under the call of
mysql_derived_prepare() itself (sql_derived.cc:824) and the table is
assigned later at 867.

The fix disables unit resolution for field type in FOR SYSTEM_TIME
expression as it does a little sense in any case: making historical
queries based on variable field values produces the result of multiple
time points.

fix_fields_if_needed() in resolve_units() was introduced by 46be31982a
2023-11-10 15:46:15 +03:00
Aleksey Midenkov
e53e7cd134 MDEV-20545 Assertion col.vers_sys_end() in dict_index_t::vers_history_row
Index values for row_start/row_end was wrongly calculated for inplace
ALTER for some layout of virtual fields.

Possible impact

  1. history row is not detected upon build clustered index for
     inplace ALTER which may lead to duplicate key errors on
     auto-increment and FTS index add.
  2. foreign key constraint may falsely fail.
  3. after inplace ALTER before server restart trx-based system
     versioning can cause server crash or incorrect data written upon
     UPDATE.
2023-11-10 15:46:14 +03:00
Oleksandr Byelkin
fecd78b837 Merge branch '10.10' into 10.11 2023-11-08 16:46:47 +01:00
Oleksandr Byelkin
04d9a46c41 Merge branch '10.6' into 10.10 2023-11-08 16:23:30 +01:00
Oleksandr Byelkin
b83c379420 Merge branch '10.5' into 10.6 2023-11-08 15:57:05 +01:00
Oleksandr Byelkin
6cfd2ba397 Merge branch '10.4' into 10.5 2023-11-08 12:59:00 +01:00
Marko Mäkelä
14685b10df MDEV-32050: Deprecate&ignore innodb_purge_rseg_truncate_frequency
The motivation of introducing the parameter
innodb_purge_rseg_truncate_frequency in
mysql/mysql-server@28bbd66ea5 and
mysql/mysql-server@8fc2120fed
seems to have been to avoid stalls due to freeing undo log pages
or truncating undo log tablespaces. In MariaDB Server,
innodb_undo_log_truncate=ON should be a much lighter operation
than in MySQL, because it will not involve any log checkpoint.

Another source of performance stalls should be
trx_purge_truncate_rseg_history(), which is shrinking the history list
by freeing the undo log pages whose undo records have been purged.
To alleviate that, we will introduce a purge_truncation_task that will
offload this from the purge_coordinator_task. In that way, the next
innodb_purge_batch_size pages may be parsed and purged while the pages
from the previous batch are being freed and the history list being shrunk.

The processing of innodb_undo_log_truncate=ON will still remain the
responsibility of the purge_coordinator_task.

purge_coordinator_state::count: Remove. We will ignore
innodb_purge_rseg_truncate_frequency, and act as if it had been
set to 1 (the maximum shrinking frequency).

purge_coordinator_state::do_purge(): Invoke an asynchronous task
purge_truncation_callback() to free the undo log pages.

purge_sys_t::iterator::free_history(): Free those undo log pages
that have been processed. This used to be a part of
trx_purge_truncate_history().

purge_sys_t::clone_end_view(): Take a new value of purge_sys.head
as a parameter, so that it will be updated while holding exclusive
purge_sys.latch. This is needed for race-free access to the field
in purge_truncation_callback().

Reviewed by: Vladislav Lesin
2023-10-25 09:11:58 +03:00
Thirunarayanan Balathandayuthapani
952f06aa8b MDEV-29989 binlog_do_db option breaks versioning table
Problem:
=========
During commit, server calls prepare_commit_versioned to
determine the transaction modified system-versioned data.
Due to binlog_do_db option, we disable the binlog for the
statement. But prepare_commit_versioned() is being
called only when binlog is enabled for the statement.

Fix:
===
prepare_commit_versioned() should happen irrespective
of binlog state. So if the server has any read-write operation
then we should call prepare_commit_versioned().
2023-09-26 10:47:59 +05:30
Oleksandr Byelkin
036df5f970 Merge branch '10.10' into 10.11 2023-08-08 14:57:31 +02:00
Oleksandr Byelkin
ced243a099 Merge branch '10.9' into 10.10 2023-08-05 20:34:09 +02:00
Oleksandr Byelkin
34a8e78581 Merge branch '10.6' into 10.9 2023-08-04 08:01:06 +02:00
Oleksandr Byelkin
6bf8483cac Merge branch '10.5' into 10.6 2023-08-01 15:08:52 +02:00
Aleksey Midenkov
732d1ec3ae MDEV-29873 MSAN uninitialized value errors in bcmp /
prep_alter_part_table upon re-partitioning by system time

memcmp() tries to compare beyond the last member of interval because
sizeof(Vers_part_info::interval) is 80. It is sizeof of variable,
sizeof of type is 76.

Now we compare interval_t struct C++ way.
2023-07-27 19:43:45 +03:00
Aleksey Midenkov
b70d5476b6 MDEV-31244 Assertion "not SELECT" in vers_set_hist_part()
The assertion was to make sure we don't do vers_set_hist_part() for
SELECT (or any non-DML). But actually we must do it if SELECT calls
some function that does DML. Patch moves the assertion to non-routines
only.
2023-07-27 19:43:45 +03:00
Aleksey Midenkov
35533dc0b3 MDEV-29727 ALTER and CREATE with default partitioning
differently react to SQL_MODE => unusable SHOW CREATE

Use abort_on_warning dependent on strict mode over create new table
like it is done for copy data and inplace alter.
2023-07-27 18:12:58 +03:00
Oleksandr Byelkin
f291c3df2c Merge branch '10.4' into 10.5 2023-07-27 15:43:21 +02:00
Oleksandr Byelkin
7564be1352 Merge branch '10.4' into 10.5 2023-07-26 16:02:57 +02:00
Lena Startseva
9854fb6fa7 MDEV-31003: Second execution for ps-protocol
This patch adds for "--ps-protocol" second execution
of queries "SELECT".
Also in this patch it is added ability to disable/enable
(--disable_ps2_protocol/--enable_ps2_protocol) second
execution for "--ps-prototocol" in testcases.
2023-07-26 17:15:00 +07:00
Aleksey Midenkov
14cc7e7d6e MDEV-25644 UPDATE not working properly on transaction precise system versioned table
First UPDATE under START TRANSACTION does nothing (nstate= nstate),
but anyway generates history. Since update vector is empty we get into
(!uvect->n_fields) branch which only adds history row, but does not do
update. After that we get current row with wrong (old) row_start value
and because of that second UPDATE tries to insert history row again
because it sees trx->id != row_start which is the guard to avoid
inserting multiple trx_id-based history rows under same transaction
(because we have same trx_id and we get duplicate error and this bug
demostrates that). But this try anyway fails because PK is based on
row_end which is constant under same transaction, so PK didn't change.

The fix moves vers_make_update() to an earlier stage of
calc_row_difference(). Therefore it prepares update vector before
(!uvect->n_fields) check and never gets into that branch, hence no
need to handle versioning inside that condition anymore.

Now trx->id and row_start are equal after first UPDATE and we don't
try to insert second history row.

== Cleanups and improvements ==

ha_innobase::update_row():

vers_set_fields and vers_ins_row are cleaned up into direct condition
check. SQLCOM_ALTER_TABLE check now is not used as this is dead code,
assertion is done instead.

upd_node->is_delete is set in calc_row_difference() just to keep
versioning code as much in one place as possible. vers_make_delete()
is still located in row_update_for_mysql() as this is required for
ha_innodbase::delete_row() as well.

row_ins_duplicate_error_in_clust():

Restrict DB_FOREIGN_DUPLICATE_KEY to the better conditions.
VERSIONED_DELETE is used specifically to help lower stack to
understand what caused current insert. Related to MDEV-29813.
2023-07-20 18:22:31 +03:00
Aleksey Midenkov
c5a8341115 MDEV-23100 ODKU of non-versioning column inserts history row
Use vers_check_update() to avoid inserting history row for ODKU if now
versioned fields specified in update_fields.
2023-07-20 18:22:30 +03:00
Aleksey Midenkov
fe618de691 MDEV-31313 SYSTEM VERSIONING and FOREIGN KEY CASCADE create orphan rows on replica
Constraints processing row_ins_check_foreign_constraint() was not
called because row_upd_check_references_constraints() didn't see
update as delete: node->is_delete was false.

Since MDEV-30378 we check for TRG_EVENT_DELETE to detect versioned
delete in ha_innobase::update_row().

Now we can use TRG_EVENT_DELETE to set upd_node->is_delete, so
constraints processing is triggered correctly.
2023-07-20 18:22:30 +03:00
Marko Mäkelä
7b17e0d4ac Merge 10.10 into 10.11 2023-04-27 14:03:24 +03:00
Marko Mäkelä
4f16e95310 Merge 10.9 into 10.10 2023-04-27 13:36:36 +03:00
Marko Mäkelä
436d8ef6b5 Merge 10.8 into 10.9 2023-04-27 10:03:26 +03:00
Marko Mäkelä
bbd261bb2e Merge 10.6 into 10.8 2023-04-27 09:53:56 +03:00
Marko Mäkelä
bb1d1dc846 Merge 10.5 into 10.6 2023-04-27 09:48:27 +03:00
Marko Mäkelä
902c622215 Merge 10.4 into 10.5 2023-04-27 09:39:53 +03:00
Brandon Nesterenko
29fb041007 MDEV-30430: Enabling system versioning on tables without primary key breaks replication
When replicating MDL events for a table that uses system versioning
without primary keys, ensure that for data sets with duplicate
records, the updates to these records with duplicates are enacted on
the correct row. That is, there was a bug (reported in MDEV-30430)
such that the function to find the row to update would stop after
finding the first matching record. However, in the absence of
primary keys, the version of the record is needed to compare the row
to ensure we are updating the correct one.

The fix, therefore, updates the record comparison functionality to
use system version columns when there are no primary keys on the
table.

Reviewed By:
============
Andrei Elkin <andrei.elkin@mariadb.com>
2023-04-24 15:09:40 -06:00
Marko Mäkelä
656c2e18b1 Merge 10.10 into 10.11 2023-04-14 13:08:28 +03:00
Marko Mäkelä
a009280e60 Merge 10.9 into 10.10 2023-04-14 12:24:14 +03:00
Marko Mäkelä
44281b88f3 Merge 10.8 into 10.9 2023-04-14 11:32:36 +03:00
Marko Mäkelä
1d1e0ab2cc Merge 10.6 into 10.8 2023-04-12 15:50:08 +03:00
Marko Mäkelä
5bada1246d Merge 10.5 into 10.6 2023-04-11 16:15:19 +03:00