Commit graph

178946 commits

Author SHA1 Message Date
Marko Mäkelä
e9bc0f75ef MDEV-5834 cleanup: Inline two tiny functions 2017-12-06 10:32:24 +02:00
Marko Mäkelä
1d526f31fb Merge 10.1 into 10.2 2017-12-05 14:23:57 +02:00
Marko Mäkelä
63cbb98275 MDEV-14587 dict_stats_process_entry_from_defrag_pool() fails to call dict_table_close() when index==NULL
dict_stats_process_entry_from_defrag_pool(): Simplify the logic,
and always call dict_table_close() when dict_table_open() returned
a non-NULL handle.
2017-12-05 13:25:09 +02:00
Marko Mäkelä
d1ab89037a MDEV-13670/MDEV-14550 Error log flood : "InnoDB: page_cleaner: 1000ms intended loop took N ms. The settings might not be optimal."
Silence the error log output that was introduced in MySQL 5.7
(MariaDB 10.2.2) if log_warnings=2 or less.

We should still figure out what these messages really indicate
and how to solve the problems.

pc_sleep_if_needed(): Add a parameter for the current time,
so that there will be fewer successive calls to ut_time_ms()
with no I/O between them.

buf_flush_page_cleaner_coordinator(): Exit the first loop
whenever shutdown has been requested. At the start of the loop,
call ut_time_ms() only once. Do not display the message if
log_warnings=2 or less.
2017-12-05 12:58:09 +02:00
Marko Mäkelä
8be7548085 Follow-up to MDEV-12698: Adjust some comments
The function dict_stats_update_if_needed() replaced
row_update_statistics_if_needed(). Adjust the comments accordingly.
2017-12-04 13:43:02 +02:00
Marko Mäkelä
bd8fd3b7c3 Remove references to UNIV_SYNC_DEBUG which was merged with UNIV_DEBUG 2017-12-04 11:48:12 +02:00
Monty
d7b0b8ddac MDEV-10688 rpl.rpl_row_log_innodb failed in buildbot
Problem was that Binlog_checkpoint can happen at random times.
Fixed by not write binlog_checkpoint for the rpl_log test.

Other things:
- Removed not used variable "$keep_gtid_events"
- Added option for show_binlog_events to skip binlog_checkpoint
2017-12-03 15:21:53 +02:00
Monty
d9188adae5 resolve_stack_dump updated to match latest stack trace format
Originally only symbols withing [] where resolved. Now we resolve
symbols also withing (+...)

To make it easier to see where the resolved symbol comes from, we
resolve the symbol 'in place' instead of just printing the resolved
symbol alone.
2017-12-03 12:45:54 +02:00
Marko Mäkelä
40bf5c951b Fix a Bison warning about semantic type clash in default action 2017-12-01 15:28:48 +02:00
Marko Mäkelä
8d24bef640 MDEV-14080 InnoDB shutdown sometimes hangs
srv_purge_wakeup(): If thd_destructor_proxy has initiated the first
step of shutdown, ensure that all purge threads terminate.

logs_empty_and_mark_files_at_shutdown(): Add a debug assertion.
(The purge threads should have been shut down already before this step.)
2017-12-01 15:02:04 +02:00
Vladislav Vaintroub
f59a1826f8 MDEV-14536 : during backup, retry read of log blocks, if there is
(possibly intermittent) checksum mismatch.
2017-11-30 16:30:40 +00:00
Vladislav Vaintroub
3fe261bd2b Merge remote-tracking branch 'origin/10.1' into 10.2 2017-11-29 15:08:04 +00:00
Vladislav Vaintroub
bf6d11c4d6 MDEV-14536 : In mariabackup, reread redo log blocks , if checksum mismatch
is detected.

The checksum mismatch can be due to partial write, thus retry the read
2017-11-29 14:53:12 +00:00
Vicențiu Ciorbaru
8cee2f136d MDEV-13384: "window" seems like a reserved column name but it's not listed as one
Window is a reserved keyword according to SQL Standard 2016. However, we
can make the grammar slightly flexible by allowing WINDOW keyword everywhere
except table aliases. Change yacc grammar to separate between all keywords
and table_alias keywords.
2017-11-29 09:39:31 +02:00
Daniel Bartholomew
b65fd73bb1 bump the VERSION 2017-11-28 15:59:36 -05:00
Marko Mäkelä
23d2dae5f0 Fix some integer type mismatch warnings 2017-11-28 18:29:20 +02:00
Vladislav Vaintroub
e02b860861 Windows, generic threadpool cleanups
- get rid of casts between int and HANDLE.
- store file descriptor in TP_connection_generic to get
rid of multiple mysql_socket_getfd(c->thd->net.vio->mysql_socket))

Also support named pipes (no reason not to support, since it is easy)
2017-11-28 13:23:51 +00:00
Elena Stepanova
77872e4519 Additions to the list of unstable tests for 10.2.11 2017-11-27 12:04:51 +02:00
Vladislav Vaintroub
414d3a3e17 Fix warning. 2017-11-24 17:33:53 +00:00
Sergey Vojtovich
4cc20c88df Fixed build failure with PFS disabled
mariabackup fails to builds with PFS disabled. The reason was missing include,
which was included by PFS otherwise.
2017-11-24 17:30:40 +00:00
Vladislav Vaintroub
9cefffdab1 Merge remote-tracking branch 'origin/10.1' into 10.2
# Conflicts:
#	storage/innobase/include/os0file.h
2017-11-24 17:30:26 +00:00
Vladislav Vaintroub
40756c9151 Fix Windows build with -DPLUGIN_PERFSCHEMA=NO 2017-11-24 16:55:20 +00:00
Marko Mäkelä
f1cc6e3874 Merge 10.1 into 10.2 2017-11-24 17:17:16 +02:00
Marko Mäkelä
6979d20426 MDEV-14499 Mariabackup 10.2 fails to back up a multi-file InnoDB system tablespace
When Mariabackup is invoked on an instance that uses a multi-file
InnoDB system tablespace, it may fail to other files of the system
tablespace than the first one.
This was revealed by the MDEV-14447 test case.
The offending code is assuming that the first page of each data file
is page 0. But, in multi-file system tablespaces that is not the case.

xb_fil_cur_open(): Instead of re-reading the first page of the file,
rely on the fil_space_t metadata that already exists in memory.

xb_get_space_flags(): Remove.
2017-11-24 17:12:38 +02:00
Vladislav Vaintroub
59150361c3 Update C/C to fix unit tests with clang 2017-11-24 14:39:38 +01:00
Vladislav Vaintroub
c5fffb33a4 MDEV-14483 Export _mysl_client_plugin_declaration from auth_gssapi_client.so
Update C/C to include fix for this bug.
2017-11-24 00:30:51 +00:00
Vladislav Vaintroub
316f0d8fe3 MDEV-14447 mariabackup incremental incorrectly extends system tablespace
for multi-file innodb_data_file_path.

Use fil_extend_space_to_desired_size() to correctly extend system
tablespace. Make sure to get tablespace size from the first tablespace
part.
2017-11-24 00:01:29 +00:00
Elena Stepanova
e6d6b0c08c Updated list of unstable tests for 10.2.11 2017-11-24 01:38:38 +02:00
Andrei Elkin
c666ca7b1b MDEV-12012. Post-push attempt to catch failure in rpl_gtid_delete_domain failing on P8. The test is made more verbose. 2017-11-23 22:10:31 +02:00
Marko Mäkelä
e2dd4e3206 MDEV-4697 UPDATE_TIME field for InnoDB
Import and adjust the MySQL 5.7 tests
innodb.update_time innodb.update_time_wl6658 into MariaDB.

The functionality is present since MariaDB 10.2.2 merged
InnoDB from MySQL 5.7.9. It was implemented in MySQL 5.7.2.
2017-11-23 08:56:06 +02:00
Marko Mäkelä
187a5bbf28 Adjust the tests for MariaDB, and optimize them 2017-11-23 08:55:51 +02:00
Marko Mäkelä
f8bc799a89 Import WL#6658 update_time tests from MySQL 5.7 2017-11-23 08:55:51 +02:00
Sergei Golubchik
0055e1a57f Merge branch 'connect/10.2' into 10.2 2017-11-22 17:38:30 +01:00
Aleksey Midenkov
37dd3cf424 MDEV-13550 Copy ctor instread of memcpy() in partition_info::get_clone() (#436)
List<>::last is wrong after memcpy(). Doing it on constructed objects is bad practice.
2017-11-22 12:52:16 +01:00
Sergei Golubchik
3422ceb10c remove dead code 2017-11-22 12:52:16 +01:00
David Carlier
f86413ecc1 Fix the build on OpenBSD (#488)
* rocksdb fails without timer_delete() - only use it when it exists
2017-11-22 12:52:12 +01:00
Vladislav Vaintroub
12840f97cd Fix typo, and disable own dtrace proibes on Solaris, really. 2017-11-21 16:20:08 -05:00
wlad
83eb14ff65 Fix compile error. 2017-11-21 21:30:45 +01:00
wlad
b6d72ed44d MDEV-14283 : Fix Solaris 10 build.
- introduce system check for posix_memalign (not available on Solaris 10)
- Disable dtrace probes, to fix weird link errors in mariabackup
2017-11-21 21:14:06 +01:00
Sergei Golubchik
14c2a9a52e don't git submodule update from rocksdb/CMakeLists.txt (#492)
we now have cmake/submodules.cmake that updates all submodules
2017-11-21 20:03:57 +01:00
Sergei Golubchik
8c422bf48d MDEV-14256 MariaDB 10.2.10 can't SST with xtrabackup-v2
another followup for 4c2c057d40.

there are six possible cases:
--port can be set or not.
--address can be set, not set, or set but without a port number

The correct behavior is:
1 both --port and --address have a port number
  - use it if it's the same, otherwise an error
2 only --port has the number (--address isn't set)
  - use the value from --port
3 only --port has the number (--address is set, but has no port)
  - use the value from --port
4 --port is unset, --address has the port number
  - use the value from --address
5 --port is unset, --address has no port number
  - use the value from --address, that is, port is empty string
6 --port is unset, --address is unset
  - port is unset (an error somewhere later)

case 5 wasn't handled correctly
2017-11-21 20:03:57 +01:00
Sergei Golubchik
a963cb95db MDEV-14268 Executable C source code files? 2017-11-21 20:03:57 +01:00
Sergei Golubchik
5dd505b709 MDEV-14428 main.cte_nonrecursive failed in --embedded
move the privilege related test to main.cte_grant
2017-11-21 20:03:57 +01:00
Sergei Golubchik
7f1900705b Merge branch '10.1' into 10.2 2017-11-21 19:47:46 +01:00
Vladislav Vaintroub
7c4f859384 MDEV-14283 : Fix compilation of mariabackup for gcc3.x 2017-11-21 17:19:32 +00:00
Marko Mäkelä
375caf99c4 Adjust an imported test 2017-11-21 18:53:00 +02:00
Marko Mäkelä
9405fdeb9e MDEV-13201 Assertion srv_undo_sources || ... failed on shutdown during DDL operation
dict_stats_exec_sql(): Refuse the operation if shutdown has been
initiated.

The real fix would be to update the persistent statistics as part
of the data dictionary transactions. To do this, we should move the
storage of InnoDB persistent statistics to the InnoDB data files,
and maybe also remove the InnoDB data dictionary.
2017-11-21 17:56:31 +02:00
Marko Mäkelä
18cdc254b7 MDEV-13626 Merge InnoDB test cases from MySQL 5.7 (part 3) 2017-11-20 16:06:17 +02:00
Marko Mäkelä
f233c9778e Adjust the MySQL 5.7 tests for MariaDB 10.2 2017-11-20 16:05:41 +02:00
Marko Mäkelä
55a94ef1cf Import WL#7277 bulk insert creation tests from MySQL 5.7 2017-11-20 13:28:34 +02:00