MariaDB 10.4 was crashing when thread-handling was set to
pool-of-threads and wsrep was enabled.
There were two apparent reasons for the crash:
- Connection handling in threadpool_common.cc was missing calls to
control wsrep client state.
- Thread specific storage which contains thread variables (THR_KEY_mysys)
was not handled appropriately by wsrep patch when pool-of-threads
was configured.
This patch addresses the above issues in the following way:
- Wsrep client state open/close was moved in thd_prepare_connection() and
end_connection() to have common handling for one-thread-per-connection
and pool-of-threads.
- Thread local storage handling in wsrep patch was reworked by introducing
set of wsrep_xxx_threadvars() calls which replace calls to
THD store_globals()/reset_globals() and deal with thread handling
specifics internally.
Wsrep-lib was updated to version which relaxes internal concurrency
related sanity checks.
Rollback code from wsrep_rollback_process() was extracted to separate calls
for better readability.
Post rollback thread was removed as it was completely unused.
* Collect and pass apply error data to provider
* Rollback failed transaction and continue operation if provider returns
SUCCESS
* MTR tests for inconsistency voting
The extensive usage of stack space, especially when used with ASan
(AdressSanitizer) of mysql_execute_command caused the test
rpl.rpl_row_sp011 to fail because it did run out of stack. In this
test case mysql_execute_command is called recursively for each
function all.
Changes done:
- Changed a few functions that used big local variables to be marked
__attribute__ ((noinline))
- Moved sub parts that used big local variables to external functions.
- Changed wsrep_commit_empty() from inline to normal function as this used
more than 1K of stack space and because there is no reason for this
rarely used function to be inline.
End result (with gcc 7.4.1 on Intel Xeon):
Starting point for stack space usage:
gcc -O: 7800
gcc with -fsanitize=address -O (ASan) : 27240
After this patch:
gcc -O: 1160
gcc -O0 (debug build) 1584
gcc with -fsanitize=address -O (ASan): 4424
gcc with -fsanitize=address -O2 (ASan): 3874
A 6x improvement and will allow us to run all mtr tests with ASan.
Problem was that wsrep thread argument was deleted on wrong
place. Furthermore, scan method incorrectly used unsafe c_ptr().
Finally, fixed wsrep thread initialization to correctly set
up thread_id and pass correct argument to functions and
fix signess problem causing compiler errors.
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
Code properly integrated into callers instead.
do_handle_one_connection(): no need to reset thd and thd->thread_stack
as they're not updated by cache_thread().
Part of MDEV-19515 - Improve connect speed
It was meaningful only for one-thread-per-connection scheduler anyway,
so call init_new_connection_handler_thread() directly from
handle_one_connection().
It was somewhat harmful for no-threads scheduler, because it'd attempt to
detach main thread on AIX_3_2 and Siemens unix (and DEC OSF/1 3.2 too).
Also it does duplicate my_thread_init() call, which may produce a warning
under EXTRA_DEBUG_THREADS.
start_wsrep_THD() shouldn't have called it at all: it already detached
and called my_thread_init().
Part of MDEV-19515 - Improve connect speed
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.
* MDEV-16509 Improve wsrep commit performance with binlog disabled
Release commit order critical section early after trx_commit_low() if
binlog is not transaction coordinator. In order to avoid two phase commit,
binlog_hton is not registered for THD during IO_CACHE population.
Implemented a test which verifies that the transactions release
commit order early.
This optimization will change behavior during recovery as the commit
is not two phase when binlog is off. Fixed and recorded wsrep-recover-v25
and wsrep-recover to match the behavior.
* MDEV-18730 Ordering for wsrep binlog group commit
Previously out of order execution was allowed for wsrep commits.
Established proper ordering by populating wait_for_commit
for every wsrep THD and making group commit leader to wait for
prior commits before proceeding to trx_group_commit_leader().
* MDEV-18730 Added a test case to verify correct commit ordering
* MDEV-16509, MDEV-18730 Review fixes
Use WSREP_EMULATE_BINLOG() macro to decide if the binlog_hton
should be registered. Whitespace/syntax fixes and cleanups.
* MDEV-16509 Require binlog for galera_var_innodb_disallow_writes test
If the commit to InnoDB is done in one phase, the native InnoDB behavior
is that the transaction is committed in memory before it is persisted to
disk. This means that the innodb_disallow_writes=ON may not prevent
transaction to become visible to other readers before commit is completely
over. On the other hand, if the commit is two phase (as it is with binlog),
the transaction will be blocked in prepare phase.
Fixed the test to use binlog, which enforces two phase commit, which
in turn makes commit to block before the changes become visible to
other connections. This guarantees that the test produces expected
result.
Removed redundant initialisation in unireg_init(): already done by
mysql_init_variables().
Slave threads already check THD::killed, which eliminates the need to
check abort_loop.
Removed unused wsrep_kill_mysql().
Refactored wsrep patch to not use LOCK_thread_count and COND_thread_count anymore.
This has partially been replaced by using old LOCK_wsrep_slave_threads mutex.
For slave thread count change waiting, new COND_wsrep_slave_threads signal has been added
Added LOCK_wsrep_cluster_config mutex to control that cluster address change cannot happen in parallel
Protected wsrep_slave_threads variable changes with LOCK_cluster_config mutex
This is for avoiding concurrent slave thread count and cluster joining operations to happen
Fixes according to Teemu's review
Global variable wsrep_debug now can be used to filter wsrep-lib messages based on debug level provided.
Type of wsrep_debug is now set to be unsigned int, so tests and configuration files changed accordingly.
If wsrep_load_data_splitting is configured, change streaming replication
parameters internally to match the original behavior, i.e. replicate
on every 10000 rows. After load data is over, restore original
streaming replication settings.
Removed redundant wsrep_tc_log_commit().
Signal handler is now respoinsible for setting abort_loop and breaking
poll() in main thread. The rest is handled by main thread itself.
Removed redundant LOCK_error_log init/destroy wrappers.
Removed redundant unireg_end(): it is trivial and it has only one caller.
Removed unused ready_to_exit from PFS.
Removed kill_in_progress: duplicates abort_loop.
Removed shutdown_in_progress: duplicates abort_loop.
Removed ready_to_exit: was used to make sure main thread waits for
cleanups, which are now done by main thread itself.
Removed SIGNALS_DONT_BREAK_READ, MAYBE_BROKEN_SYSCALL,
kill_broken_server: never defined/used.
Make clean_up() static.
In contrast to thread_count, which is decremented by THD destructor,
this one was most probably intended to be decremented after all THD
destructors are done.
THD_count class was added to achieve similar effect with thread_count.
Aim is to reduce usage of LOCK_thread_count and COND_thread_count.
Part of MDEV-15135.
Implemented and integrated THD_list as a replacement for the global
thread list. It uses own mutex instead of LOCK_thread_count for THD
list protection.
Removed unused first_global_thread() and next_global_thread().
delayed_insert_threads is now protected by LOCK_delayed_insert. Although
this patch doesn't fix very wrong synchronization of this variable.
After this patch there are only 2 legitimate uses of LOCK_thread_count
left, both in mysqld.cc: thread_count and ready_to_exit.
Aim is to reduce usage of LOCK_thread_count and COND_thread_count.
Part of MDEV-15135.
The second line of changes related to replacing xtrabackup with
mariabackup:
1) All unnecessary references to xtrabackup are removed from
the documentation, from some comments, from the control files
that are used to prepare the packages.
2) Made corrections of the tests from the galera_3nodes suite
that mentioned xtrabackup or the old (associated with xtrabackup)
version of innobackupex.
3) Fixed flaws in the galera_3nodes mtr suite control scripts,
because of which they could not work with mariabackup.
4) Fixed numerous bugs in the SST scripts and in the mtr test
files (galera_3nodes mtr suite) that prevented the use of Galera
with IPv6 addresses.
5) Fixed flaws in tests for rsync and mysqldump (for galera_3nodes
mtr tests suite). These tests were not performed successfully without
these fixes.
https://jira.mariadb.org/browse/MDEV-17835
Part of MDEV-5336 Implement LOCK FOR BACKUP
- Changed check of Global_only_lock to also include BACKUP lock.
- We store latest MDL_BACKUP_DDL lock in thd->mdl_backup_ticket to be able
to downgrade lock during copy_data_between_tables()
Problem was that controlling connection i.e. connection that
executed the query SET GLOBAL wsrep_reject_queries = ALL_KILL;
was also killed but server would try to send result from that
query to controlling connection resulting a assertion
mysqld: /home/jan/mysql/10.2-sst/include/mysql/psi/mysql_socket.h:738: inline_mysql_socket_send: Assertion `mysql_socket.fd != -1' failed.
as socket was closed when controlling connection was closed.
wsrep_close_client_connections()
Do not close controlling connection and instead of
wsrep_close_thread() we do now soft kill by THD::awake
wsrep_reject_queries_update()
Call wsrep_close_client_connections using current thd.
Problem was that controlling connection i.e. connection that
executed the query SET GLOBAL wsrep_reject_queries = ALL_KILL;
was also killed but server would try to send result from that
query to controlling connection resulting a assertion
mysqld: /home/jan/mysql/10.2-sst/include/mysql/psi/mysql_socket.h:738: inline_mysql_socket_send: Assertion `mysql_socket.fd != -1' failed.
as socket was closed when controlling connection was closed.
wsrep_close_client_connections()
Do not close controlling connection and instead of
wsrep_close_thread() we do now soft kill by THD::awake
wsrep_reject_queries_update()
Call wsrep_close_client_connections using current thd.
main.derived_cond_pushdown: Move all 10.3 tests to the end,
trim trailing white space, and add an "End of 10.3 tests" marker.
Add --sorted_result to tests where the ordering is not deterministic.
main.win_percentile: Add --sorted_result to tests where the
ordering is no longer deterministic.
