Commit graph

3812 commits

Author SHA1 Message Date
Sergei Golubchik
0f12ada6b6 Merge remote-tracking branch 'mysql/5.5' into 5.5 2015-04-27 21:04:06 +02:00
Sergei Golubchik
9fd65db329 MDEV-7585 Assertion `thd->is_error() || kill_errno || thd->killed == ABORT_QUERY' failed in ha_rows filesort
if we clear the error status (in THD::clear_error())
make sure to clear the thd->killed == KILL_BAD_DATA too,
because it was caused by the error that we're clearing.
2015-04-27 15:42:12 +02:00
Kristian Nielsen
791b0ab5db Merge 10.0 -> 10.1 2015-04-20 13:21:58 +02:00
Kristian Nielsen
accdabd668 Merge MDEV-7888 and MDEV-7929 into 10.0. 2015-04-08 13:19:22 +02:00
Kristian Nielsen
48c10fb5f7 Merge MDEV-7888 and MDEV-7929 into 10.1. 2015-04-08 11:04:24 +02:00
Kristian Nielsen
3b961347db MDEV-7888, MDEV-7929: Parallel replication hangs sometimes on ANALYZE TABLE or DDL
The hangs occur when the group_commit_orderer object is freed before the last
mark_start_commit() call on it - this loses the wakeup to other waiting worker
threads, causing them to hang until killed manually.

The object was freed because wakeup_subsequent_commits() was called two early
in two places. For MDEV-7888, during ANALYZE TABLE, and for MDEV-7929 during
record_gtid() after processing a DDL event. The group_commit_orderer object
can be freed when its last transaction has called wait_for_prior_commit().

Fix by implementing a suspend/resume mechanism for wakeup_subsequent_commits()
that can be used in places where a transaction is committed without this being
the commit of the actual replication event group.

Also add a protection mechanism (that asserts in debug builds) which can
prevent the too-early free and hang if other similar bugs should remain in
other parts of the code.
2015-04-08 11:01:18 +02:00
Marc Alff
7285b4c495 Bug#20788853 MUTEX ISSUE IN SQL/SQL_SHOW.CC RESULTING IN SIG6. SOURCE LIKELY
FILL_VARIABLES

Prevent mutexes used in SHOW VARIABLES from being locked twice.
2015-04-08 07:01:39 +02:00
Nirbhay Choubey
84a4db263a Merge branch '5.5-galera' into 10.0-galera 2015-03-28 17:29:10 -04:00
Nirbhay Choubey
d7445ea6df MDEV-7194: galera fails to replicate DDL queries when using binlog_checksum
Restore fix for MDEV-4328 (revno: 3391) that got accidentally
overwritten while merging :
http://bazaar.launchpad.net/~codership/codership-mysql/5.5-23/revision/3900

Added a test case.
2015-03-27 20:35:37 -04:00
Alexander Barkov
50eee60504 Preparatory refactoring for:
MDEV-6218 Wrong result of CHAR_LENGTH(non-BMP-character) with 3-byte utf8
- Moving get_text() as a method to Lex_input_stream.
- Moving the unescaping part into a separate function,
  this piece of code will later go to /strings most likely.
- Removing Lex_input_string::yytoklen, as it's not needed any more.
2015-03-26 20:44:12 +04:00
Oleksandr Byelkin
01d7da6785 MDEV-7834: ANALYZE FORMAT=JSON output column should be named ANALYZE 2015-03-25 19:55:11 +01:00
Sergei Petrunia
664ce4c507 Fix linking: move the inline functions 2015-03-24 17:35:29 +03:00
Sergei Petrunia
4106dfe89b Merge branch 'bb-10.1-explain-analyze' into 10.1 2015-03-24 16:26:42 +03:00
Sergei Petrunia
77e16ce7d6 MDEV-7648: Extra data in ANALYZE FORMAT=JSON $stmt
Switch from relying on PERFORMANCE_SCHEMA to using our
own hooks for counting the time spent reading rows from
tables.
2015-03-24 16:17:41 +03:00
Alexander Barkov
0c26c0032c A preparatory patch for MDEV-7284 INDEX: CREATE OR REPLACE.
Removing "bool Key::create_if_not_exists" and deriving Key from
DDL_options instead.
2015-03-20 13:51:41 +04:00
Alexander Barkov
2a2cc16478 Refactoring in sql_yacc.yy. A pre-requirement patch for
MDEV-7801 Unexpected syntax error in ALTER TABLE t1 ADD INDEX TYPE BTREE
MDEV-7284 INDEX: CREATE OR REPLACE
2015-03-20 11:46:44 +04:00
Kristian Nielsen
4ba16ea4bc Merge MDEV-6981 and MDEV-6981 (from danblack) into MariaDB 10.1 2015-03-17 09:24:22 +01:00
Alexander Barkov
e6f67c64cd MDEV-6572 "USE dbname" with a bad sequence erroneously connects to a wrong database 2015-03-16 21:55:10 +04:00
Sergey Vojtovich
6bd24deab4 MDEV-7728 - Improve xid cache scalability by using lock-free hash
XID cache is now based on lock-free hash.
Also fixed lf_hash_destroy() to call alloc destructor.

Note that previous implementation had race condition when thread was accessing
XA owned by different thread. This new implementation doesn't fix it either.
2015-03-16 19:07:51 +04:00
Kristian Nielsen
ef4d8db5ec MDEV-6981: feature request MASTER_GTID_WAIT status variables
Review fixes:
 - Coding style
 - Fix bad .result file
 - Fix test to be tolerant of different timing.
 - Fix test to give better info in case of unexpected timing.
2015-03-16 14:40:29 +01:00
Kristian Nielsen
0e717c5bf4 Merge branch 'mdev-6981-master_gtid_wait-status-variables' of https://github.com/openquery/mariadb-server into danblack
Conflicts:
	sql/mysqld.cc
2015-03-16 13:41:11 +01:00
Kristian Nielsen
184f718fef MDEV-7249: Performance problem in parallel replication with multi-level slaves
Parallel replication (in 10.0 / "conservative" mode) relies on binlog group
commits to group transactions that can be safely run in parallel on the
slave. The --binlog-commit-wait-count and --binlog-commit-wait-usec options
exist to increase the number of commits per group. But in case of conflicts
between transactions, this can cause unnecessary delay and reduced througput,
especially on a slave where commit order is fixed.

This patch adds a heuristics to reduce this problem. When transaction T1 goes
to commit, it will first wait for N transactions to queue up for a group
commit. However, if we detect that another transaction T2 is waiting for a row
lock held by T1, then we will skip the wait and let T1 commit immediately,
releasing locks and let T2 continue.

On a slave, this avoids the unfortunate situation where T1 is waiting for T2
to join the group commit, but T2 is waiting for T1 to release locks, causing
no work to be done for the duration of the --binlog-commit-wait-usec timeout.

(The heuristic seems reasonable on the master as well, so it is enabled for
all transactions, not just replication transactions).
2015-03-13 14:01:52 +01:00
Jan Lindström
8249dcaaeb MDEV-6858: enforce_storage_engine option
Merge from Percona Server enforced use of a specific storage engine
authored by Stewart Smith.

Modified to be session variable and modifiable only by SUPER. Use
similar implementation as default_storage_engine.
2015-03-12 12:17:14 +02:00
Oleksandr Byelkin
a71e2d23bd MDEV-7006 MDEV-7007: SET STATEMENT and slow log
fixed embedded server tests
MDEV-7009: SET STATEMENT min_examined_row_limit has no effect
MDEV-6948:SET STATEMENT gtid_domain_id = ... FOR has no effect (same for gtid_seq_no and server_id)

old values of SET STATENENT variables now saved in its own Query_arena and restored later
2015-03-12 10:19:32 +01:00
Daniel Black
fa5809ce10 Add Master_gtid_wait_{count,time,timeouts} status
MASTER_GTID_WAIT function needs some status to evaluate its use.

master_gtid_wait_count indicates how many times the function is called.
master_gtid_wait_time indicates how much time in microseconds occurred
waiting (or timing out)
master_gtid_timeouts indicates how many time times this function timed
out rather than all successful gtids events being available.
2015-03-12 06:43:38 +11:00
Nirbhay Choubey
a174aaed75 cleanup: remove unused THD::COND_wsrep_thd
Conflicts:
	mysql-test/suite/perfschema/r/all_instances.result
	sql/mysqld.h
2015-02-19 17:45:21 -05:00
Nirbhay Choubey
1e6f46d78f cleanup: remove unused THD::COND_wsrep_thd 2015-02-19 17:28:18 -05:00
Sergey Vojtovich
451e9b7a50 MDEV-7499 - System variables have broken default values on big endian
INFORMATION_SCHEMA.SYSTEM_VARIABLES.DEFAULT_VALUE had broken values on
big endian.

Default value is internally stored as longlong, while I_S references it's
pointer (longlong *) according to variable type (e.g. int, my_bool, etc). This
works well on little endian, but on big endian we always get 0 for such
variables.
2015-02-05 13:54:55 +04:00
Monty
0ee879ff8a Improve performance for calculating memory allocation
Extend interface for 'show variables' with current scope
2015-02-01 15:24:22 +02:00
Sergei Golubchik
6a3fee211f Merge remote-tracking branch 'refs/remotes/github/10.1' into 10.1 2015-02-01 12:39:59 +01:00
Sergei Golubchik
4b21cd21fe Merge branch '10.0' into merge-wip 2015-01-31 21:48:47 +01:00
Oleksandr Byelkin
edf34f38ac MDEV-7024: Assertion `! is_set()' failed in Diagnostics_area::set_eof_status on executing ANALYZE SELECT via PS
select_send::is_result_interceptor() should return FALSE because it return results to the client.
2015-01-29 22:43:07 +01:00
Nirbhay Choubey
47d8001c67 MDEV-7476: Allow SELECT to succeed even when node is not ready
Added a SESSION-only system variable "wsrep_dirty_reads" to allow SELECT
queries to pass even when the node is not prepared to accept queries
(wsrep_ready=OFF). Added a test case.
2015-01-27 16:54:24 -05:00
Nirbhay Choubey
9b7e3801f4 MDEV-7476: Allow SELECT to succeed even when node is not ready
Added a SESSION-only system variable "wsrep_dirty_reads" to allow SELECT
queries to pass even when the node is not prepared to accept queries
(wsrep_ready=OFF). Added a test case.
2015-01-27 16:22:29 -05:00
Nisha Gopalakrishnan
aa1de73728 Bug#20094067: BACKPORT BUG#19683834 TO 5.5 AND 5.6
Backporting the patch and the test case fixed as part
of BUG#16041903 and BUG#19683834 respectively.
2015-01-27 13:13:55 +05:30
Nirbhay Choubey
7cda4bee0e maria-10.0.16 merge
bzr merge -r4588 maria/10.0
2015-01-26 22:54:27 -05:00
Nirbhay Choubey
0105bf349a MDEV-7476: Allow SELECT to succeed even when node is not ready
Added a SESSION-only system variable "wsrep_dirty_reads" to allow SELECT
queries to pass even when the node is not prepared to accept queries
(wsrep_ready=OFF). Added a test case.
2015-01-22 18:00:37 -05:00
Sergei Golubchik
56c323c10f MDEV-6728 KILL QUERY executed on an idle connection can interrupt the next query
reset KILL_QUERY when a new query execution is just about to be started
2015-01-19 14:07:41 +01:00
Sergei Golubchik
3fcbd7cb47 cleanup: remove unused THD::COND_wsrep_thd 2015-01-11 15:49:23 +01:00
Sergey Vojtovich
6dbc48ca79 MDEV-7324 - Lock-free hash for table definition cache 2014-12-28 19:46:18 +04:00
Alexander Barkov
c6d3f8058d MDEV-7112 Split HA_CREATE_INFO 2014-12-08 10:56:08 +04:00
Nirbhay Choubey
3bb02f3e6d bzr merge -rtag:mariadb-10.0.15 maria/10.0 2014-12-05 12:33:02 -05:00
Sergei Golubchik
0438f12540 MDEV-6980 OUT parameters in PREPARE
revert the patch for "out parameters in prepare"
2014-12-04 10:41:52 +01:00
Sergei Golubchik
ec4137c62b Merge branch '10.1' into bb-10.1-merge 2014-12-03 11:37:26 +01:00
Sergei Golubchik
853077ad7e Merge branch '10.0' into bb-10.1-merge
Conflicts:
	.bzrignore
	VERSION
	cmake/plugin.cmake
	debian/dist/Debian/control
	debian/dist/Ubuntu/control
	mysql-test/r/join_outer.result
	mysql-test/r/join_outer_jcl6.result
	mysql-test/r/null.result
	mysql-test/r/old-mode.result
	mysql-test/r/union.result
	mysql-test/t/join_outer.test
	mysql-test/t/null.test
	mysql-test/t/old-mode.test
	mysql-test/t/union.test
	packaging/rpm-oel/mysql.spec.in
	scripts/mysql_config.sh
	sql/ha_ndbcluster.cc
	sql/ha_ndbcluster_binlog.cc
	sql/ha_ndbcluster_cond.cc
	sql/item_cmpfunc.h
	sql/lock.cc
	sql/sql_select.cc
	sql/sql_show.cc
	sql/sql_update.cc
	sql/sql_yacc.yy
	storage/innobase/buf/buf0flu.cc
	storage/innobase/fil/fil0fil.cc
	storage/innobase/include/srv0srv.h
	storage/innobase/lock/lock0lock.cc
	storage/tokudb/CMakeLists.txt
	storage/xtradb/buf/buf0flu.cc
	storage/xtradb/fil/fil0fil.cc
	storage/xtradb/include/srv0srv.h
	storage/xtradb/lock/lock0lock.cc
	support-files/mysql.spec.sh
2014-12-02 22:25:16 +01:00
Sergei Petrunia
3d5f97fd70 Merge ../10.1-explain-json-r4 into 10.1 2014-11-27 00:51:54 +03:00
Kristian Nielsen
d08b893b39 MDEV-6775: Wrong binlog order in parallel replication: Intermediate commit
The code in binlog group commit around wait_for_commit that controls commit
order, did the wakeup of subsequent commits early, as soon as a following
transaction is put into the group commit queue, but before any such commit has
actually taken place. This causes problems with too early wakeup of
transactions that need to wait for prior to commit, but do not take part in
the binlog group commit for one reason or the other.

This patch solves the problem, by moving the wakeup to happen only after the
binlog group commit is completed.

This requires a new solution to ensure that transactions that arrive later
than the leader are still able to participate in group commit. This patch
introduces a flag wait_for_commit::commit_started. When this is set, a waiter
can queue up itself in the group commit queue.

This way, effectively the wait_for_prior_commit() is skipped only for
transactions that participate in group commit, so that skipping the wait is
safe. Other transactions still wait as needed for correctness.
2014-11-13 10:31:20 +01:00
Oleksandr Byelkin
1827d9e6d1 MDEV-5231: Per query variables from Percona Server (rewritten) 2014-11-11 11:48:50 +01:00
Igor Babaev
d24919969a Correction for the fix of the bug mdev-6874. 2014-10-19 09:50:50 -07:00
Igor Babaev
ec89abfef4 Merge branch '10.1' of ../10.1-mdev334 into 10.1 2014-10-17 14:06:54 -07:00