Before this fix, the server could crash during shutdown,
due to race conditions, that occured when killing the server.
In particular, the performance schema instrumentation handle,
PSI_server, and the performance schema itself would be cleaned up
too soon, causing race conditions with a running kill server thread.
The specifics of the race condition found are that:
the main thread executing "PSI_server= NULL" can cause crashes in
other threads still running, which are executing
"if (PSI_server != NULL) PSI_server->xxx()"
as part of the performance schema instrumentation.
While the bug was reported for the kill server thread,
in theory the same crash could happen with the signal thread,
as found by code analysis.
The correct fix would be to only shutdown the performance schema
and set PSI_server to NULL after every other thread is guaranteed
to be completed, including the kill_server_thread.
However, due to the existing mysqld server design, this is not the case.
See in particular bug number 56666.
The work around used to fix this race condition is to simply not
perform the call to shutdown_performance_schema() when the server exits,
and to keep the PSI_server pointer unchanged.
This will cause memory leaks to be reported by tools like valgrind,
but no memory leak actually happen because the process is about to exit().
As a result, the file mysql-test/valgrind.supp has been updated
to filter out these false positive messages.
This code has been tested with running in a loop the following
tests in parallel, which have been known to fail with race conditions
in the past:
- rpl_change_master
- binlog_max_extension
- events_restart
- rpl_heartbeat_basic
and no crash of test failure has been seen with the changed code.
Fixed typo that caused warnings from mysql-test-run
mysql-test/mysql-test-run.pl:
Fixed typo
sql/mysqld.cc:
Give a more precise warning why something fails.
Fixed test failures in buildbot
Don't write errors when failing to send ok packet
mysql-test/suite/pbxt/r/range.result:
Don't write number of rows as it varies.
mysql-test/suite/pbxt/t/range.test:
Don't write number of rows as it varies.
sql/mysqld.cc:
Don't write errors when failing to send ok packet
storage/maria/ma_bitmap.c:
Added DBUG_ASSERT to detect wrong bitmap pages
storage/maria/ma_blockrec.c:
Don't reset BLOCKUSED_USE_ORG_BITMAP flag. This fixed a bug where bitmap could be wrong after UNDO of row with blobs
Fixed bug in Aria when replacing short keys with long keys and a key tree both overflow and underflow at same time.
Fixed several bugs when generating recovery logs when using RGQ with replacing long keys with short keys and vice versa.
Lots of new DBUG_ASSERT()'s
Added more information to recovery log to make it easier to know from where log entry orginated.
Introduced MARIA_PAGE->org_size that tells what the size of the page was in last log entry. This allows us to find out if all key changes for index page was logged.
Small code cleanups:
- Introduced _ma_log_key_changes() to log crc of key page changes
- Added share->max_index_block_size as max size of data one can put in key block (block_size - KEYPAGE_CHECKSUM_SIZE)
This will later simplify adding a directory to index pages.
- Write page number instead of page postition to DBUG log
mysql-test/lib/v1/mysql-test-run.pl:
Use --general-log instead of --log to disable warning when using RQG
sql/mysqld.cc:
If we have already sent ok to client when we get an error, log this to stderr
Don't disable option --log-output if CSV engine is not supported.
storage/maria/ha_maria.cc:
Log queries to recovery log also in LOCK TABLES
storage/maria/ma_check.c:
If param->max_trid is set, use this value instead of max_trid_in_system().
This is used by recovery to set max_trid to max seen trid so far.
keyinfo->block_length - KEYPAGE_CHECKSUM_SIZE -> max_index_block_size (Style optimization)
storage/maria/ma_delete.c:
Mark tables crashed early
Write page number instead of page position to debug log.
Added parameter to ma_log_delete() and ma_log_prefix() that is logged so that we can find where wrong log entries where generated.
Fixed bug where a page was not proplerly written when same key tree had both an overflow and underflow when deleting a key.
keyinfo->block_length - KEYPAGE_CHECKSUM_SIZE => max_index_block_size (Style optimization)
ma_log_delete() now has extra parameter of how many bytes from end of page should be appended to log for page (for page overflows)
storage/maria/ma_key_recover.c:
Added extra parameter to ma_log_prefix() to indicate what caused log entry.
Update MARIA_PAGE->org_size when logging info about page.
Much more DBUG_ASSERT()'s.
Fix some bugs in maria_log_add() to handle page overflows.
Added _ma_log_key_changes() to log crc of key page changes.
If EXTRA_STORE_FULL_PAGE_IN_KEY_CHANGES is defines, log the resulting pages to log so one can trivally
see how the resulting page should have looked like (for errors in CRC values)
storage/maria/ma_key_recover.h:
Added _ma_log_key_changes() which is only called if EXTRA_DEBUG_KEY_CHANGES is defined.
Updated function prototypes.
storage/maria/ma_loghandler.h:
Added more values to en_key_debug, to get more exact location where things went wrong when logging to recovery log.
storage/maria/ma_open.c:
Initialize share->max_index_block_size
storage/maria/ma_page.c:
Added updating and testing of MARIA_PAGE->org_size
Write page number instead of page postition to DBUG log
Generate error if we read page with wrong data.
Removed wrong assert: key_del_current != share->state.key_del.
Simplify _ma_log_compact_keypage()
storage/maria/ma_recovery.c:
Set param.max_trid to max seen trid before running repair table (used for alter table to create index)
storage/maria/ma_rt_key.c:
Update call to _ma_log_delete()
storage/maria/ma_rt_split.c:
Use _ma_log_key_changes()
Update MARIA_PAGE->org_size
storage/maria/ma_unique.c:
Remove casts
storage/maria/ma_write.c:
keyinfo->block_length - KEYPAGE_CHECKSUM_SIZE => share->max_index_block_length.
Updated calls to _ma_log_prefix()
Changed code to use _ma_log_key_changes()
Update ma_page->org_size
Fixed bug in _ma_log_split() for pages that overflow
Added KEY_OP_DEBUG logging to functions
Log KEYPAGE_FLAG in all log entries
storage/maria/maria_def.h:
Added SHARE->max_index_block_size
Added MARIA_PAGE->org_size
storage/maria/trnman.c:
Reset flags for new transaction.
thread-specific variables weren't set when we load error message files.
per-file comments:
libmysqld/lib_sql.cc
Bug#53251 mysql_library_init fails on second execution with embedded library
we need to call my_thread_init() once more. Normally it's called at the my_init()
stage but that doesn't happen on the second my_init() call.
sql/derror.cc
Bug#53251 mysql_library_init fails on second execution with embedded library
use default errors for the embedded server.
sql/mysqld.cc
Bug#53251 mysql_library_init fails on second execution with embedded library
unregister server errors in clean_up(). Without it the error list contains
that on the second mysql_server_init() which is not good.
sql/set_var.cc
Bug#53251 mysql_library_init fails on second execution with embedded library
sys_var::cleanup() call instead of the destructor
sql/set_var.h
Bug#53251 mysql_library_init fails on second execution with embedded library
sys_var::cleanup() introduced instead of the destructor
sql/sys_vars.h
Bug#53251 mysql_library_init fails on second execution with embedded library
Sys_var_charptr::cleanup() implemented
- Changed to still use bcmp() in certain cases becasue
- Faster for short unaligneed strings than memcmp()
- Bettern when using valgrind
- Changed to use my_sprintf() instead of sprintf() to get higher portability for old systems
- Changed code to use MariaDB version of select->skip_record()
- Removed -%::SCCS/s.% from Makefile.am:s to remove automake warnings
Before this fix, the ha_read_last_count status variable was defined and
updated internally, for never exposed as a system variable.
This fix exposes the system variable as "Handler_read_last",
for completness of the Handler_read_* system variables interface.
Adjusted tests results accordingly.
Added open log file with FILE_SHARE_DELETE flag on Windows.
sql/log.cc:
added reopen_fstreams();
modified redirect_std_streams(): call to sequence of freopen()
replaced to reopen_fstreams();
modified flush_error_log(): removed file rename for flushed
error log file.
sql/mysqld.cc:
modified main() and init_server_components(): do open log error file
over call to reopen_fstreams().
temp table
This patch introduces two key changes in the replication's behavior.
Firstly, it reverts part of BUG#51894 which puts any update to temporary tables
into the trx-cache. Now, updates to temporary tables are handled according to
the type of their engines as a regular table.
Secondly, an unsafe mixed statement, (i.e. a statement that access transactional
table as well non-transactional or temporary table, and writes to any of them),
are written into the trx-cache in order to minimize errors in the execution when
the statement logging format is in use.
Such changes has a direct impact on which statements are classified as unsafe
statements and thus part of BUG#53259 is reverted.
Fixed LP#613418 (M)aria recovery failure: ma_key_recover.c:981: _ma_apply_redo_index: Assertion `check_page_length == page_length' failed
include/my_sys.h:
Added my_disable_sync
mysql-test/mysql-test-run.pl:
Added --sync-sys=0 to run test suite faster
mysys/my_static.c:
Added my_disable_sync
mysys/my_sync.c:
Added my_disable_sync
sql/mysqld.cc:
Added -sync-sys=0 option for mysqld to skip sync() calls for faster testing
storage/maria/ma_key_recover.c:
More DBUG_ASSERT()
Added logging of KEY_OP_DEBUG to make examening of logs easier
Fixed testing of page length in recovery to ensure we don't overwrite checksum (previous tests was too relaxed)
Fixed bug in recovery logging of split pages which caused failure during recovery:
- Length was not adjusted properly for pages to be split
- Added KEY_OP_MAX_PAGELENGTH to tell recovery that page is now full length
- This fixed LP#613418
storage/maria/ma_key_recover.h:
Changed prototype for ma_log_change() for KEY_OP_DEBUG
storage/maria/ma_loghandler.h:
Added new enums for better debugging of recovery logs
storage/maria/ma_rt_index.c:
Added debugging information to calls to ma_log_change()
storage/maria/ma_write.c:
Added debugging information to calls to ma_log_change() and ma_log_split()
TABLES <list> WITH READ LOCK are incompatible".
The problem was that FLUSH TABLES <list> WITH READ LOCK
which was issued when other connection has acquired global
read lock using FLUSH TABLES WITH READ LOCK was blocked
and has to wait until global read lock is released.
This issue stemmed from the fact that FLUSH TABLES <list>
WITH READ LOCK implementation has acquired X metadata locks
on tables to be flushed. Since these locks required acquiring
of global IX lock this statement was incompatible with global
read lock.
This patch addresses problem by using SNW metadata type of
lock for tables to be flushed by FLUSH TABLES <list> WITH
READ LOCK. It is OK to acquire them without global IX lock
as long as we won't try to upgrade those locks. Since SNW
locks allow concurrent statements using same table FLUSH
TABLE <list> WITH READ LOCK now has to wait until old
versions of tables to be flushed go away after acquiring
metadata locks. Since such waiting can lead to deadlock
MDL deadlock detector was extended to take into account
waits for flush and resolve such deadlocks.
As a bonus code in open_tables() which was responsible for
waiting old versions of tables to go away was refactored.
Now when we encounter old version of table in open_table()
we don't back-off and wait for all old version to go away,
but instead wait for this particular table to be flushed.
Such approach supported by deadlock detection should reduce
number of scenarios in which FLUSH TABLES aborts concurrent
multi-statement transactions.
Note that active FLUSH TABLES <list> WITH READ LOCK still
blocks concurrent FLUSH TABLES WITH READ LOCK statement
as the former keeps tables open and thus prevents the
latter statement from doing flush.
mysql-test/include/handler.inc:
Adjusted test case after changing status which is set
when FLUSH TABLES waits for tables to be flushed from
"Flushing tables" to "Waiting for table".
mysql-test/r/flush.result:
Added test which checks that "flush tables <list> with
read lock" is compatible with active "flush tables with
read lock" but not vice-versa. This test also covers
bug #52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES
<list> WITH READ LOCK are incompatible".
mysql-test/r/mdl_sync.result:
Added scenarios in which wait for table to be flushed
causes deadlocks to the coverage of MDL deadlock detector.
mysql-test/suite/perfschema/r/dml_setup_instruments.result:
Adjusted test results after removal of COND_refresh
condition variable.
mysql-test/suite/perfschema/r/server_init.result:
Adjusted test and its results after removal of COND_refresh
condition variable.
mysql-test/suite/perfschema/t/server_init.test:
Adjusted test and its results after removal of COND_refresh
condition variable.
mysql-test/t/flush.test:
Added test which checks that "flush tables <list> with
read lock" is compatible with active "flush tables with
read lock" but not vice-versa. This test also covers
bug #52044 "FLUSH TABLES WITH READ LOCK and FLUSH TABLES
<list> WITH READ LOCK are incompatible".
mysql-test/t/kill.test:
Adjusted test case after changing status which is set
when FLUSH TABLES waits for tables to be flushed from
"Flushing tables" to "Waiting for table".
mysql-test/t/lock_multi.test:
Adjusted test case after changing status which is set
when FLUSH TABLES waits for tables to be flushed from
"Flushing tables" to "Waiting for table".
mysql-test/t/mdl_sync.test:
Added scenarios in which wait for table to be flushed
causes deadlocks to the coverage of MDL deadlock detector.
sql/ha_ndbcluster.cc:
Adjusted code after adding one more parameter for
close_cached_tables() call - timeout for waiting for
table to be flushed.
sql/ha_ndbcluster_binlog.cc:
Adjusted code after adding one more parameter for
close_cached_tables() call - timeout for waiting for
table to be flushed.
sql/lock.cc:
Removed COND_refresh condition variable. See comment
for sql_base.cc for details.
sql/mdl.cc:
Now MDL deadlock detector takes into account information
about waits for table flushes when searching for deadlock.
To implement this change:
- Declaration of enum_deadlock_weight and
Deadlock_detection_visitor were moved to mdl.h header
to make them available to the code in table.cc which
implements deadlock detector traversal through edges
of waiters graph representing waiting for flush.
- Since now MDL_context may wait not only for metadata
lock but also for table to be flushed an abstract
Wait_for_edge class was introduced. Its descendants
MDL_ticket and Flush_ticket incapsulate specifics
of inspecting waiters graph when following through
edge representing wait of particular type.
We no longer require global IX metadata lock when acquiring
SNW or SNRW locks. Such locks are needed only when metadata
locks of these types are upgraded to X locks. This allows
to use SNW locks in FLUSH TABLES <list> WITH READ LOCK
implementation and keep the latter compatible with global
read lock.
sql/mdl.h:
Now MDL deadlock detector takes into account information
about waits for table flushes when searching for deadlock.
To implement this change:
- Declaration of enum_deadlock_weight and
Deadlock_detection_visitor were moved to mdl.h header
to make them available to the code in table.cc which
implements deadlock detector traversal through edges
of waiters graph representing waiting for flush.
- Since now MDL_context may wait not only for metadata
lock but also for table to be flushed an abstract
Wait_for_edge class was introduced. Its descendants
MDL_ticket and Flush_ticket incapsulate specifics
of inspecting waiters graph when following through
edge representing wait of particular type.
- Deadlock_detection_visitor now has m_table_shares_visited
member which allows to support recursive locking for
LOCK_open. This is required when deadlock detector
inspects waiters graph which contains several edges
representing waits for flushes or needs to come through
the such edge more than once.
sql/mysqld.cc:
Removed COND_refresh condition variable. See comment
for sql_base.cc for details.
sql/mysqld.h:
Removed COND_refresh condition variable. See comment
for sql_base.cc for details.
sql/sql_base.cc:
Changed approach to how threads are waiting for table
to be flushed. Now thread that wants to wait for old
table to go away subscribes for notification by adding
Flush_ticket to table's share and waits using
MDL_context::m_wait object. Once table gets flushed
(i.e. all tables are closed and table share is ready
to be destroyed) all such waiters are notified
individually.
Thanks to this change MDL deadlock detector can take
such waits into account.
To implement this/as result of this change:
- tdc_wait_for_old_versions() was replaced with
tdc_wait_for_old_version() which waits for individual
old share to go away and which is called by open_table()
after finding out that share is outdated. We don't
need to perform back-off before such waiting thanks
to the fact that deadlock detector now sees such waits.
- As result Open_table_ctx::m_mdl_requests became
unnecessary and was removed. We no longer allocate
copies of MDL_request objects on MEM_ROOT when
MYSQL_OPEN_FORCE_SHARED/SHARED_HIGH_PRIO flags are
in effect.
- close_cached_tables() and tdc_wait_for_old_version()
share code which implements waiting for share to be
flushed - the both use TABLE_SHARE::wait_until_flush()
method. Thanks to this close_cached_tables() supports
timeouts and has extra parameter for this.
- Open_table_context::OT_MDL_CONFLICT enum element was
renamed to OT_CONFLICT as it is now also used in cases
when back-off is required to resolve deadlock caused
by waiting for flush and not metadata lock.
- In cases when we discover that current connection tries
to open tables from different generation we now simply
back-off and restart process of opening tables. To
support this Open_table_context::OT_REOPEN_TABLES enum
element was added.
- COND_refresh condition variable became unnecessary and
was removed.
- mysql_notify_thread_having_shared_lock() no longer wakes
up connections waiting for flush as all such connections
can be waken up by deadlock detector if necessary.
sql/sql_base.h:
- close_cached_tables() now has one more parameter -
timeout for waiting for table to be flushed.
- Open_table_context::OT_MDL_CONFLICT enum element was
renamed to OT_CONFLICT as it is now also used in cases
when back-off is required to resolve deadlock caused
by waiting for flush and not metadata lock.
Added new OT_REOPEN_TABLES enum element to be used in
cases when we need to restart open tables process even
in the middle of transaction.
- Open_table_ctx::m_mdl_requests became unnecessary and
was removed.
sql/sql_class.h:
Added assert ensuring that we won't use LOCK_open mutex
with THD::enter_cond(). Otherwise deadlocks can arise in
MDL deadlock detector.
sql/sql_parse.cc:
Changed FLUSH TABLES <list> WITH READ LOCK to take SNW
metadata locks instead of X locks on tables to be flushed.
Since we no longer require global IX lock to be taken
when SNW locks are taken this makes this statement
compatible with FLUSH TABLES WITH READ LOCK statement.
Since SNW locks allow other connections to have table
opened FLUSH TABLES <list> WITH READ LOCK now has to
wait during open_tables() for old version to go away.
Such waits can lead to deadlocks which will be detected
by MDL deadlock detector which now takes waits for table
to be flushed into account.
Also adjusted code after adding one more parameter for
close_cached_tables() call - timeout for waiting for
table to be flushed.
sql/sql_yacc.yy:
FLUSH TABLES <list> WITH READ LOCK now needs only SNW
metadata locks on tables.
sql/sys_vars.cc:
Adjusted code after adding one more parameter for
close_cached_tables() call - timeout for waiting for
table to be flushed.
sql/table.cc:
Implemented new approach to how threads are waiting for
table to be flushed. Now thread that wants to wait for
old table to go away subscribes for notification by
adding Flush_ticket to table's share and waits using
MDL_context::m_wait object. Once table gets flushed
(i.e. all tables are closed and table share is ready
to be destroyed) all such waiters are notified
individually. This change allows to make such waits
visible inside of MDL deadlock detector.
To do it:
- Added list of waiters/Flush_tickets to TABLE_SHARE
class.
- Changed free_table_share() to postpone freeing of
share memory until last waiter goes away and to
wake up subscribed waiters.
- Added TABLE_SHARE::wait_until_flushed() method which
implements subscription to the list of waiters for
table to be flushed and waiting for this event.
Implemented interface which allows to expose waits for
flushes to MDL deadlock detector:
- Introduced Flush_ticket class a descendant of
Wait_for_edge class.
- Added TABLE_SHARE::find_deadlock() method which allows
deadlock detector to find out what contexts are still
using old version of table in question (i.e. to find
out what contexts are waited for by owner of
Flush_ticket).
sql/table.h:
In order to support new strategy of waiting for table flush
(see comment for table.cc for details) added list of
waiters/Flush_tickets to TABLE_SHARE class.
Implemented interface which allows to expose waits for
flushes to MDL deadlock detector:
- Introduced Flush_ticket class a descendant of
Wait_for_edge class.
- Added TABLE_SHARE::find_deadlock() method which allows
deadlock detector to find out what contexts are still
using old version of table in question (i.e. to find
out what contexts are waited for by owner of
Flush_ticket).
Fix warnings flagged by the new warning option -Wunused-but-set-variable
that was added to GCC 4.6 and that is enabled by -Wunused and -Wall. The
option causes a warning whenever a local variable is assigned to but is
later unused. It also warns about meaningless pointer dereferences.
client/mysql.cc:
Meaningless pointer dereferences.
client/mysql_upgrade.c:
Check whether reading from the file succeeded.
extra/comp_err.c:
Unused.
extra/yassl/src/yassl_imp.cpp:
Skip instead of reading data that is discarded.
include/my_pthread.h:
Variable is only used in debug builds.
include/mysys_err.h:
Add new error messages.
mysys/errors.c:
Add new error message for permission related functions.
mysys/mf_iocache.c:
Variable is only checked under THREAD.
mysys/my_copy.c:
Raise a error if chmod or chown fails.
mysys/my_redel.c:
Raise a error if chmod or chown fails.
regex/engine.c:
Use a equivalent variable for the assert.
server-tools/instance-manager/instance_options.cc:
Unused.
sql/field.cc:
Unused.
sql/item.cc:
Unused.
sql/log.cc:
Do not ignore the return value of freopen: only set buffer if
reopening succeeds.
Adjust doxygen comment to the right function.
Pass message lenght to log function.
sql/mysqld.cc:
Do not ignore the return value of freopen: only set buffer if
reopening succeeds.
sql/partition_info.cc:
Unused.
sql/slave.cc:
No need to set pointer to the address of '\0'.
sql/spatial.cc:
Unused. Left for historical purposes.
sql/sql_acl.cc:
Unused.
sql/sql_base.cc:
Pointers are always set to the same variables.
sql/sql_parse.cc:
End statement if reading fails.
Store the buffer after it has actually been updated.
sql/sql_repl.cc:
No need to set pointer to the address of '\0'.
sql/sql_show.cc:
Put variable under the same ifdef block.
sql/udf_example.c:
Set null pointer flag appropriately.
storage/csv/ha_tina.cc:
Meaningless dereferences.
storage/example/ha_example.cc:
Return the error since it's available.
storage/myisam/mi_locking.c:
Remove unused and dead code.
- Lots of TODO comments
- add mrr_sort_keys flag to @@optimizer_switch
- [from Igor] SQL layer part passes HA_MRR_MATERIALIZED_KEYS flag
- Don't call rnd_pos() many times in a row if sorted rowid buffer
has the same rowid value for multiple consequive (rowid, range_id) pairs.
- Let "mysqld --help --verbose" list all optimizer options
- Make it possible to add new @@optimizer_switch flags w/o causing .result
changes all over the testsuite:
= Remove "select @@optimizer_switch" from tests that do not need all switches
= Move @@optimizer_switch-specific tests to t/optimizer_switch.test
compiler problem
GCC-style inline assembly is not supported by the Sun Studio
compilers prior to version 12.
Added a check for the Sun Studio version to avoid using
_FPU_GETCW() / _FPU_SETCW() when inline assembly is
unsupported. This can lead to some differences in floating
point calculations on Solaris 8/x86 which, however, is not worth
bothering with Sun-style assembly .il templates.
