Mutex order violation when wsrep bf thread kills a conflicting trx,
the stack is
wsrep_thd_LOCK()
wsrep_kill_victim()
lock_rec_other_has_conflicting()
lock_clust_rec_read_check_and_lock()
row_search_mvcc()
ha_innobase::index_read()
ha_innobase::rnd_pos()
handler::ha_rnd_pos()
handler::rnd_pos_by_record()
handler::ha_rnd_pos_by_record()
Rows_log_event::find_row()
Update_rows_log_event::do_exec_row()
Rows_log_event::do_apply_event()
Log_event::apply_event()
wsrep_apply_events()
and mutexes are taken in the order
lock_sys->mutex -> victim_trx->mutex -> victim_thread->LOCK_thd_data
When a normal KILL statement is executed, the stack is
innobase_kill_query()
kill_handlerton()
plugin_foreach_with_mask()
ha_kill_query()
THD::awake()
kill_one_thread()
and mutexes are
victim_thread->LOCK_thd_data -> lock_sys->mutex -> victim_trx->mutex
This patch is the plan D variant for fixing potetial mutex locking
order exercised by BF aborting and KILL command execution.
In this approach, KILL command is replicated as TOI operation.
This guarantees total isolation for the KILL command execution
in the first node: there is no concurrent replication applying
and no concurrent DDL executing. Therefore there is no risk of
BF aborting to happen in parallel with KILL command execution
either. Potential mutex deadlocks between the different mutex
access paths with KILL command execution and BF aborting cannot
therefore happen.
TOI replication is used, in this approach, purely as means
to provide isolated KILL command execution in the first node.
KILL command should not (and must not) be applied in secondary
nodes. In this patch, we make this sure by skipping KILL
execution in secondary nodes, in applying phase, where we
bail out if applier thread is trying to execute KILL command.
This is effective, but skipping the applying of KILL command
could happen much earlier as well.
This also fixed unprotected calls to wsrep_thd_abort
that will use wsrep_abort_transaction. This is fixed
by holding THD::LOCK_thd_data while we abort transaction.
Reviewed-by: Jan Lindström <jan.lindstrom@mariadb.com>
This patch is the plan D variant for fixing potetial mutex locking
order exercised by BF aborting and KILL command execution.
In this approach, KILL command is replicated as TOI operation.
This guarantees total isolation for the KILL command execution
in the first node: there is no concurrent replication applying
and no concurrent DDL executing. Therefore there is no risk of
BF aborting to happen in parallel with KILL command execution
either. Potential mutex deadlocks between the different mutex
access paths with KILL command execution and BF aborting cannot
therefore happen.
TOI replication is used, in this approach, purely as means
to provide isolated KILL command execution in the first node.
KILL command should not (and must not) be applied in secondary
nodes. In this patch, we make this sure by skipping KILL
execution in secondary nodes, in applying phase, where we
bail out if applier thread is trying to execute KILL command.
This is effective, but skipping the applying of KILL command
could happen much earlier as well.
This patch also fixes mutex locking order and unprotected
THD member accesses on bf aborting case. We try to hold
THD::LOCK_thd_data during bf aborting. Only case where it
is not possible is at wsrep_abort_transaction before
call wsrep_innobase_kill_one_trx where we take InnoDB
mutexes first and then THD::LOCK_thd_data.
This will also fix possible race condition during
close_connection and while wsrep is disconnecting
connections.
Added wsrep_bf_kill_debug test case
Reviewed-by: Jan Lindström <jan.lindstrom@mariadb.com>
Incorrect processing of an auto-incrementing field in the
WSREP-related code during applying transactions results in
a duplicate key being created. This is due to the fact that
at the beginning of the write_row() and update_row() functions,
the values of the auto-increment parameters are used, which
are read from the parameters of the current thread, but further
along the code other values are used, which are read from global
variables (when applying a transaction). This can happen when
the cluster configuration has changed while applying a transaction
(for example in the high_priority_service mode for Galera 4).
Further during IST processing duplicating key is detected, and
processing of the DB_DUPLICATE_KEY return code (inside innodb,
in the write_row() handler) results in a call to the
wsrep_thd_self_abort() function.
Some DML operations on tables having unique secondary keys cause scanning
in the secondary index, for instance to find potential unique key violations
in the seconday index. This scanning may involve GAP locking in the index.
As this locking happens also when applying replication events in high priority
applier threads, there is a probabality for lock conflicts between two wsrep
high priority threads.
This PR avoids lock conflicts of high priority wsrep threads, which do
secondary index scanning e.g. for duplicate key detection.
The actual fix is the patch in sql_class.cc:thd_need_ordering_with(), where
we allow relaxed GAP locking protocol between wsrep high priority threads.
wsrep high priority threads (replication appliers, replayers and TOI processors)
are ordered by the replication provider, and they will not need serializability
support gained by secondary index GAP locks.
PR contains also a mtr test, which exercises a scenario where two replication
applier threads have a false positive conflict in GAP of unique secondary index.
The conflicting local committing transaction has to replay, and the test verifies
also that the replaying phase will not conflict with the latter repllication applier.
Commit also contains new test scenario for galera.galera_UK_conflict.test,
where replayer starts applying after a slave applier thread, with later seqno,
has advanced to commit phase. The applier and replayer have false positive GAP
lock conflict on secondary unique index, and replayer should ignore this.
This test scenario caused crash with earlier version in this PR, and to fix this,
the secondary index uniquenes checking has been relaxed even further.
Now innodb trx_t structure has new member: bool wsrep_UK_scan, which is set to
true, when high priority thread is performing unique secondary index scanning.
The member trx_t::wsrep_UK_scan is defined inside WITH_WSREP directive, to make
it possible to prepare a MariaDB build where this additional trx_t member is
not present and is not used in the code base. trx->wsrep_UK_scan is set to true
only for the duration of function call for: lock_rec_lock() trx->wsrep_UK_scan
is used only in lock_rec_has_to_wait() function to relax the need to wait if
wsrep_UK_scan is set and conflicting transaction is also high priority.
Reviewed-by: Jan Lindström <jan.lindstrom@mariadb.com>
If log_slave_updates==OFF, wsrep applier threads used to be configured
with option: thd->variables.option_bits&= ~(OPTION_BIN_LOG);
(i.e. like sql_log_bin=ON). And this was regardless of log-bin configuration.
With this, having configuration of: --log-bin && --log-slave-updates=OFF,
local threads used binlogging, but applier threads did not. And further:
local threads went through binlog group commit, while applier threads did
direct commits. This resulted in situation, where applier threads entered
earlier in wsrep XID checkpointing, and could sync their wsrep XID out of order.
Later local thread commit would see that higher seqno was already checkpointed,
and fire an assert because of this.
As a fix, applier threads are now forced to enable binlogging regardless of
log-slave-updates configuration.
This PR comes with new mtr test: galera.MDEV-24327, which causes a scenario
where applier transaction is applied and committed while earlier local transaction
is parked before commit order monitor enter. A buggy mariadb versoin would fail
for assertion because of wsrep XID checkpoint order violation.
Reviewed-by: Jan Lindström <jan.lindstrom@mariadb.com>
The reason for the failure is that
thd->mdl_context.release_transactional_locks()
was called after commit & rollback even in cases where the current
transaction is still active.
For 10.2, 10.3 and 10.4 the fix is simple:
- Replace all calls to thd->mdl_context.release_transactional_locks() with
thd->release_transactional_locks(). The thd function will only call
the mdl_context function if there are no active transactional locks.
In 10.6 we will better fix where we will change the return value for
some trans_xxx() functions to indicate if transaction did close the
transaction or not. This will avoid the need of the indirect call.
Other things:
- trans_xa_commit() and trans_xa_rollback() will automatically
call release_transactional_locks() if the transaction is closed.
- We can't do that for the other functions as the caller of many of these
are doing additional work (like close_thread_tables) before calling
release_transactional_locks().
- Added missing abort_result_set() and missing DBUG_RETURN in
select_create::send_eof()
- Fixed wrong indentation in injector::transaction::commit()
Remove incorrect BF (brute force) handling from lock_rec_has_to_wait_in_queue
and move condition to correct callers. Add a function to report
BF lock waits and assert if incorrect BF-BF lock wait happens.
wsrep_report_bf_lock_wait
Add a new function to report BF lock wait.
wsrep_assert_no_bf_bf_wait
Add a new function to check do we have a
BF-BF wait and if we have report this case
and assert as it is a bug.
lock_rec_has_to_wait
Use new wsrep_assert_bf_wait to check BF-BF wait.
lock_rec_create_low
lock_table_create
Use new function to report BF lock waits.
lock_rec_insert_by_trx_age
lock_grant_and_move_on_page
lock_grant_and_move_on_rec
Assert that trx is not Galera as VATS is not compatible
with Galera.
lock_rec_add_to_queue
If there is conflicting lock in a queue make sure that
transaction is BF.
lock_rec_has_to_wait_in_queue
Remove incorrect BF handling. If there is conflicting
locks in a queue all transactions must wait.
lock_rec_dequeue_from_page
lock_rec_unlock
If there is conflicting lock make sure it is not
BF-BF case.
lock_rec_queue_validate
Add Galera record locking rules comment and use
new function to report BF lock waits.
All attempts to reproduce the original assertion have been
failed. Therefore, there is no test case on this commit.
Backported the support for aborting and replaying stored procedure and fix for trigger
key assigments from 10.4 version.
Backported also two mtr tests: wsrep_sp_bf_abort and MDEV-20225
If async replication slave thread conflicts with cluster replication,
then the async slave transaction should be BF aborted, and depending on the
state of async slave transaction execution, potentially also replayed.
There were problems in such BF abort implementation and the replaying was not
started.
This pull request contains fixes which make sure that if async slave thread is
marked to abort and replay, it will complete carry out the rollback and
release all locks and resources before starting the replaying. After replaying,
async slave transactions is treated as successful, so the slave thread will
continue as usual, handling next replication event.
There is also new mtr test: galera.galera_slave_replay, which stresses both a
certification failure for async slave thread and a successful BF abort
followed by replaying.
For release builds, do not declare unused variables.
unpack_row(): Omit a debug-only variable from WSREP diagnostic message.
create_wsrep_THD(): Fix -Wmaybe-uninitialized for the PSI_thread_key.
Galera threads were not registered to performance schema and
used pthread_create when mysql_thread_create should have been
used.
Added test case to verify current galera performance schema
instrumentation does work.
Problem was that tests select INFORMATION_SCHEMA.PROCESSLIST processes
from user system user and empty state. Thus, there is not clear
state for slave threads.
Changes:
- Added new status variables that store current amount of applier threads
(wsrep_applier_thread_count) and rollbacker threads
(wsrep_rollbacker_thread_count). This will make clear how many slave threads
of certain type there is.
- Added THD state "wsrep applier idle" when applier slave thread is
waiting for work. This makes finding slave/applier threads easier.
- Added force-restart option for mtr to always restart servers between tests
to avoid race on start of the test
- Added wait_condition_with_debug to wait until the passed statement returns
true, or the operation times out. If operation times out, the additional error
statement will be executed
Changes to be committed:
new file: mysql-test/include/force_restart.inc
new file: mysql-test/include/wait_condition_with_debug.inc
modified: mysql-test/mysql-test-run.pl
modified: mysql-test/suite/galera/disabled.def
modified: mysql-test/suite/galera/r/MW-336.result
modified: mysql-test/suite/galera/r/galera_kill_applier.result
modified: mysql-test/suite/galera/r/galera_var_slave_threads.result
new file: mysql-test/suite/galera/t/MW-336.cnf
modified: mysql-test/suite/galera/t/MW-336.test
modified: mysql-test/suite/galera/t/galera_kill_applier.test
modified: mysql-test/suite/galera/t/galera_parallel_autoinc_largetrx.test
modified: mysql-test/suite/galera/t/galera_parallel_autoinc_manytrx.test
modified: mysql-test/suite/galera/t/galera_var_slave_threads.test
modified: mysql-test/suite/wsrep/disabled.def
modified: mysql-test/suite/wsrep/r/variables.result
modified: mysql-test/suite/wsrep/t/variables.test
modified: sql/mysqld.cc
modified: sql/wsrep_mysqld.cc
modified: sql/wsrep_mysqld.h
modified: sql/wsrep_thd.cc
modified: sql/wsrep_var.cc
This reverts commit 21b2fada7a
and commit 81d71ee6b2.
The MDEV-18464 change introduces a few data race issues. Contrary to
the documentation, the field trx_t::victim is not always being protected
by lock_sys_t::mutex and trx_t::mutex. Most importantly, it seems
that KILL QUERY could wrongly avoid acquiring both mutexes when
invoking lock_trx_handle_wait_low(), in case another thread had
already set trx->victim=true.
We also revert MDEV-12009, because it should depend on the MDEV-18464
fix being present.
As noted on kill_one_thread SUPER should be able to kill even
system threads i.e. threads/query flagged as high priority or
wsrep applier thread. Normal user, should not able to kill
threads/query flagged as high priority (BF) or wsrep applier
thread.
This patch contains a fix for the MDEV-17262/17243 issues and
new mtr test.
These issues (MDEV-17262/17243) have two reasons:
1) After an intermediate commit, a transaction loses its status
of "transaction that registered in the MySQL for 2pc coordinator"
(in the InnoDB) due to the fact that since version 10.2 the
write_row() function (which located in the ha_innodb.cc) does
not call trx_register_for_2pc(m_prebuilt->trx) during the processing
of split transactions. It is necessary to restore this call inside
the write_row() when an intermediate commit was made (for a split
transaction).
Similarly, we need to set the flag of the started transaction
(m_prebuilt->sql_stat_start) after intermediate commit.
The table->file->extra(HA_EXTRA_FAKE_START_STMT) called from the
wsrep_load_data_split() function (which located in sql_load.cc)
will also do this, but it will be too late. As a result, the call
to the wsrep_append_keys() function from the InnoDB engine may be
lost or function may be called with invalid transaction identifier.
2) If a transaction with the LOAD DATA statement is divided into
logical mini-transactions (of the 10K rows) and binlog is rotated,
then in rare cases due to the wsrep handler re-registration at the
boundary of the split, the last portion of data may be lost. Since
splitting of the LOAD DATA into mini-transactions is technical,
I believe that we should not allow these mini-transactions to fall
into separate binlogs. Therefore, it is necessary to prohibit the
rotation of binlog in the middle of processing LOAD DATA statement.
https://jira.mariadb.org/browse/MDEV-17262 and
https://jira.mariadb.org/browse/MDEV-17243
Problem was that we skipped background persistent statistics calculation
on applier nodes if thread is marked as high priority (a.k.a BF).
However, on applier nodes all DDL which is replicate will be executed
as high priority i.e BF.
Fixed by allowing background persistent statistics calculation on
applier nodes even when thread is marked as BF. This could lead
BF lock waits but for queries on that node needs that statistics.
If we have a 2+ node cluster which is replicating from an async master
and the binlog_format is set to STATEMENT and multi-row inserts are executed
on a table with an auto_increment column such that values are automatically
generated by MySQL, then the server node generates wrong auto_increment
values, which are different from what was generated on the async master.
In the title of the MDEV-9519 it was proposed to ban start slave on a Galera
if master binlog_format = statement and wsrep_auto_increment_control = 1,
but the problem can be solved without such a restriction.
The causes and fixes:
1. We need to improve processing of changing the auto-increment values
after changing the cluster size.
2. If wsrep auto_increment_control switched on during operation of
the node, then we should immediately update the auto_increment_increment
and auto_increment_offset global variables, without waiting of the next
invocation of the wsrep_view_handler_cb() callback. In the current version
these variables retain its initial values if wsrep_auto_increment_control
is switched on during operation of the node, which leads to inconsistent
results on the different nodes in some scenarios.
3. If wsrep auto_increment_control switched off during operation of the node,
then we must return the original values of the auto_increment_increment and
auto_increment_offset global variables, as the user has set. To make this
possible, we need to add a "shadow copies" of these variables (which stores
the latest values set by the user).
https://jira.mariadb.org/browse/MDEV-9519
If we have a 2+ node cluster which is replicating from an async master
and the binlog_format is set to STATEMENT and multi-row inserts are executed
on a table with an auto_increment column such that values are automatically
generated by MySQL, then the server node generates wrong auto_increment
values, which are different from what was generated on the async master.
In the title of the MDEV-9519 it was proposed to ban start slave on a Galera
if master binlog_format = statement and wsrep_auto_increment_control = 1,
but the problem can be solved without such a restriction.
The causes and fixes:
1. We need to improve processing of changing the auto-increment values
after changing the cluster size.
2. If wsrep auto_increment_control switched on during operation of
the node, then we should immediately update the auto_increment_increment
and auto_increment_offset global variables, without waiting of the next
invocation of the wsrep_view_handler_cb() callback. In the current version
these variables retain its initial values if wsrep_auto_increment_control
is switched on during operation of the node, which leads to inconsistent
results on the different nodes in some scenarios.
3. If wsrep auto_increment_control switched off during operation of the node,
then we must return the original values of the auto_increment_increment and
auto_increment_offset global variables, as the user has set. To make this
possible, we need to add a "shadow copies" of these variables (which stores
the latest values set by the user).
https://jira.mariadb.org/browse/MDEV-9519
If we have a 2+ node cluster which is replicating from an async master
and the binlog_format is set to STATEMENT and multi-row inserts are executed
on a table with an auto_increment column such that values are automatically
generated by MySQL, then the server node generates wrong auto_increment
values, which are different from what was generated on the async master.
The causes and fixes:
1. We need to improve processing of changing the auto-increment values
after changing the cluster size.
2. If wsrep auto_increment_control switched on during operation of
the node, then we should immediately update the auto_increment_increment
and auto_increment_offset global variables, without waiting of the next
invocation of the wsrep_view_handler_cb() callback. In the current version
these variables retain its initial values if wsrep_auto_increment_control
is switched on during operation of the node, which leads to inconsistent
results on the different nodes in some scenarios.
3. If wsrep auto_increment_control switched off during operation of the node,
then we must return the original values of the auto_increment_increment and
auto_increment_offset global variables, as the user has set. To make this
possible, we need to add a "shadow copies" of these variables (which stores
the latest values set by the user).
These test can sporadically show mutex deadlock warnings between LOCK_wsrep_thd
and LOCK_thd_data mutexes. This means that these mutexes can be locked in opposite
order by different threads, and thus result in deadlock situation.
To fix such issue, the locking policy of these mutexes should be revised and
enforced to be uniform. However, a quick code review shows that the number of
lock/unlock operations for these mutexes combined is between 100-200, and all these
mutex invocations should be checked/fixed.
On the other hand, it turns out that LOCK_wsrep_thd is used for protecting access to
wsrep variables of THD (wsrep_conflict_state, wsrep_query_state), whereas LOCK_thd_data
protects query, db and mysys_var variables in THD. Extending LOCK_thd_data to protect
also wsrep variables looks like a viable solution, as there should not be a use case
where separate threads need simultaneous access to wsrep variables and THD data variables.
In this commit LOCK_wsrep_thd mutex is refactored to be replaced by LOCK_thd_data.
By bluntly replacing LOCK_wsrep_thd by LOCK_thd_data, will result in double locking
of LOCK_thd_data, and some adjustements have been performed to fix such situations.
The problem was that the introduction of max-thread-mem-used can cause
an allocation error very early, even before mysql_parse() is called.
As mysql_parse() calls thd->reset_for_next_command(), which called
clear_error(), the error number was lost.
Fixed by adding an option to have unique messages for each KILL
signal and change max-thread-mem-used to use this new feature.
This removes a lot of problems with the original approach, where
one could get errors signaled silenty almost any time.
ixed by moving clear_error() from reset_for_next_command() to
do_command(), before any memory allocation for the thread.
Related changes:
- reset_for_next_command() now have an optional parameter if we should
call clear_error() or not. By default it's called, but not anymore from
dispatch_command() which was the original problem.
- Added optional paramater to clear_error() to force calling of
reset_diagnostics_area(). Before clear_error() only called
reset_diagnostics_area() if there was no error, so we normally
called reset_diagnostics_area() twice.
- This change removed several duplicated calls to clear_error()
when starting a query.
- Reset max_mem_used on COM_QUIT, to protect against kill during
quit.
- Use fatal_error() instead of setting is_fatal_error (cleanup)
- Set fatal_error if max_thead_mem_used is signaled.
(Same logic we use for other places where we are out of resources)
Do not silence uncertain cases, or fix any bugs.
The only functional change should be that ha_federated::extra()
is not calling DBUG_PRINT to report an unhandled case for
HA_EXTRA_PREPARE_FOR_DROP.
Merge feature into 10.2 from feature branch.
Delayed replication adds an option
CHANGE MASTER TO master_delay=<seconds>
Replication will then delay applying events with that many
seconds. This creates a replication slave that reflects the state of
the master some time in the past.
Feature is ported from MySQL source tree.
Signed-off-by: Kristian Nielsen <knielsen@knielsen-hq.org>
- popping PS reprepare observer before BF aborted PS replaying begins
dangling observer will cause failure in open_table() ater on
- test case for this anomaly
Transaction replay causes the THD to re-apply the replication
events from execution, using the same path appliers do. While
applying the log events, the THD's timestamp is set to the
timestamp of the event.
Setting the timestamp explicitly causes function NOW() to
always the timestamp that was set. To avoid this behavior we
reset the timestamp after replaying is done.
mysqld maintains a list of TABLE objects for all temporary
tables created within a session in THD. Here each table is
represented by a TABLE object.
A query referencing a particular temporary table for more
than once, however, failed with ER_CANT_REOPEN_TABLE error
because a TABLE_SHARE was allocate together with the TABLE,
so temporary tables always had only one TABLE per TABLE_SHARE.
This patch lift this restriction by separating TABLE and
TABLE_SHARE objects and storing TABLE_SHAREs for temporary
tables in a list in THD, and TABLEs in a list within their
respective TABLE_SHAREs.
different fix. remove old ones, wait for THD to be fully
initialized before continuing with the server startup process.
This reverts commits db2e21b, 13615c5, 3f515a0, 70113ee.
