|
------------------------------------------------------------
revno: 5585 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6.15-release timestamp: Sun 2013-11-17 18:51:36 +0100 message: Merged Bug #17675622 cset 5632 from mysql-5.6 ------------------------------------------------------------ revno: 5584 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6.15-release timestamp: Fri 2013-11-08 13:55:28 +0100 message: Merge Bug #17738935 cset 5605 from mysql-5.6 ------------------------------------------------------------ revno: 5583 tags: clone-5.6.15-build committer: Thirunarayanan B<thirunarayanan.balathandayuth@oracle.com> branch nick: mysql-5.6_latest timestamp: Mon 2013-11-04 16:21:21 +0530 message: Reverting the patch ------------------------------------------------------------ revno: 5582 committer: Thirunarayanan B<thirunarayanan.balathandayuth@oracle.com> branch nick: mysql-5.6_latest timestamp: Mon 2013-11-04 12:04:11 +0530 message: Problem: There is a difference in behaviour between copy algorithm and inplace algorithm in alter statement for decreasing the auto increment value. In alter table statement, inplace algorithm not able to decrease the auto increment value. Solution: Inplace algorithm should set the maximum of the user defined auto_increment value and maximum, non deleted auto_increment value from the table. ------------------------------------------------------------ revno: 5581 [merge] committer: Tor Didriksen <tor.didriksen@oracle.com> branch nick: 5.6-merge timestamp: Fri 2013-11-01 17:29:04 +0100 message: merge 5.5 => 5.6 ------------------------------------------------------------ revno: 2875.437.284 [merge] tags: clone-5.5.35-build committer: Tor Didriksen <tor.didriksen@oracle.com> branch nick: 5.5-merge timestamp: Fri 2013-11-01 16:52:21 +0100 message: merge 5.1 => 5.5 ------------------------------------------------------------ revno: 2661.876.85 tags: clone-5.1.73-build committer: Tor Didriksen <tor.didriksen@oracle.com> branch nick: 5.1 timestamp: Fri 2013-11-01 16:39:19 +0100 message: Bug#17617945 BUFFER OVERFLOW IN GET_MERGE_MANY_BUFFS_COST WITH SMALL SORT_BUFFER_SIZE get_cost_calc_buff_size() could return wrong value for the size of imerge_cost_buff. ------------------------------------------------------------ revno: 5580 committer: horst.hunger@oracle.com branch nick: mysql-5.6 timestamp: Fri 2013-11-01 10:50:18 +0100 message: Bug #17719504: These test fail sporadically. Disabled until decision how to fix if at all. ------------------------------------------------------------ revno: 5579 [merge] committer: Venkata Sidagam <venkata.sidagam@oracle.com> branch nick: 5.6 timestamp: Fri 2013-11-01 13:51:02 +0530 message: Bug #12917164 DROP USER CAN'T DROP USERS WITH LEGACY UPPER CASE HOST NAME ANYMORE Merging from mysql-5.5 to mysql-5.6 ------------------------------------------------------------ revno: 2875.437.283 committer: Venkata Sidagam <venkata.sidagam@oracle.com> branch nick: 5.5 timestamp: Fri 2013-11-01 13:48:24 +0530 message: Bug #12917164 DROP USER CAN'T DROP USERS WITH LEGACY UPPER CASE HOST NAME ANYMORE Fixing the testcase failure ------------------------------------------------------------ revno: 5578 [merge] committer: Nuno Carvalho <nuno.carvalho@oracle.com> branch nick: mysql-5.6 timestamp: Thu 2013-10-31 23:06:10 +0000 message: BUG#17662398: REMOVE DUPLICATE TEST CASES Merge from mysql-5.5 into mysql-5.6. ------------------------------------------------------------ revno: 2875.437.282 [merge] committer: Nuno Carvalho <nuno.carvalho@oracle.com> branch nick: mysql-5.5 timestamp: Thu 2013-10-31 22:58:07 +0000 message: BUG#17662398: REMOVE DUPLICATE TEST CASES Merge from mysql-5.1 into mysql-5.5. ------------------------------------------------------------ revno: 2661.876.84 committer: Nuno Carvalho <nuno.carvalho@oracle.com> branch nick: mysql-5.1 timestamp: Thu 2013-10-31 22:53:56 +0000 message: BUG#17662398: REMOVE DUPLICATE TEST CASES Remove duplicate test cases. ------------------------------------------------------------ revno: 5577 committer: Joao Gramacho <joao.gramacho@oracle.com> branch nick: mysql-5.6 timestamp: Thu 2013-10-31 21:42:33 +0000 message: Bug#16861624 KILLING WAITING SLAVE SQL THREAD: M_STATUS == DA_ERROR || M_STATUS == DA_OK Problem: ======= An assert will fail when the replication sql thread on a slave server is killed while waiting for commit lock will. Analysis: ======== In the case of a failed commit attempt at Xid_log_event::do_apply_event(), an error message will be reported informing the reason why the commit could not be completed. The error message to be reported is collected using message() at sql_error.h that contains an assert expecting that current m_status must be DA_ERROR or DA_OK before returning the message. The problem happens when the SQL thread is waiting for the commit lock at MDL_context::acquire_lock and is killed. In this case, the resulting wait_status will be MDL_wait::KILLED, and will result in acquire_lock() returning an error but without throwing one (both other cases, MDL_wait::VICTIM and MDL_wait::TIMEOUT, throw errors using my_error()). So, in the case of a thread being deliberate killed while waiting for a lock would lead to a case where no error was flagged, but also no success was flagged, making the assert expecting DA_ERROR or DA_OK to fail. Fix: === The code in MDL_context::acquire_lock in mysql-5.7 already throws an error for MDL_wait::KILLED. This patch just back-ported this behavior. Added also some debug sync points to easy putting sql thread in "Waiting for commit lock" state. ------------------------------------------------------------ revno: 5576 [merge] committer: Venkata Sidagam <venkata.sidagam@oracle.com> branch nick: 5.6 timestamp: Thu 2013-10-31 23:59:36 +0530 message: Bug #12917164 DROP USER CAN'T DROP USERS WITH LEGACY UPPER CASE HOST NAME ANYMORE Merging from mysql-5.5 to mysql-5.6 ------------------------------------------------------------ revno: 2875.437.281 [merge] committer: Venkata Sidagam <venkata.sidagam@oracle.com> branch nick: 5.5 timestamp: Thu 2013-10-31 23:14:33 +0530 message: Bug #12917164 DROP USER CAN'T DROP USERS WITH LEGACY UPPER CASE HOST NAME ANYMORE Merging from mysql-5.1 to mysql-5.5 ------------------------------------------------------------ revno: 2661.876.83 committer: Venkata Sidagam <venkata.sidagam@oracle.com> branch nick: 5.1 timestamp: Thu 2013-10-31 23:02:44 +0530 message: Bug #12917164 DROP USER CAN'T DROP USERS WITH LEGACY UPPER CASE HOST NAME ANYMORE Description: It is not possible to drop users with host names with upper case letters in them. i.e DROP USER 'root'@'Tmp_Host_Name'; is failing with error. Analysis: Since the fix 11748570 we came up with lower case hostnames as standard. But in the current bug the hostname is created by mysql_install_db script is still having upper case hostnames. So, if we have the hostname with upper case letters like(Tmp_Host_Name) then we will have as it is stored in the mysql.user table. In this case if use "'DROP USER 'root'@'Tmp_Host_Name';" it gives error because we do compare with the lower case of hostname since the 11748570 fix. Fix: We need to convert the hostname to lower case before storing into the mysql.user table when we run the mysql_install_db script. ------------------------------------------------------------ revno: 5575 committer: horst.hunger@oracle.com branch nick: mysql-5.6 timestamp: Thu 2013-10-31 11:30:31 +0100 message: Repeat for Bug #17711364: Added the opt files to the tests to run them in binaries. ------------------------------------------------------------ revno: 5574 committer: Saikumar V <saikumar.v@oracle.com> branch nick: mysql-5.6 timestamp: Thu 2013-10-31 13:24:36 +0530 message: Cleaned up .moved folder in internal . ------------------------------------------------------------ revno: 5573 [merge] committer: Saikumar V <saikumar.v@oracle.com> branch nick: mysql-5.6 timestamp: Thu 2013-10-31 09:05:42 +0530 message: Synching rqg grammars/redefines to mysql-repos internal . ------------------------------------------------------------ revno: 2875.437.280 committer: Saikumar V <saikumar.v@oracle.com> branch nick: mysql-5.5 timestamp: Thu 2013-10-31 09:01:09 +0530 message: Synching rqg grammars/redefines to mysql-repos internal. ------------------------------------------------------------ revno: 5572 committer: Jon Olav Hauglid <jon.hauglid@oracle.com> branch nick: mysql-5.6-bug17696705 timestamp: Thu 2013-10-31 00:22:43 +0100 message: Bug#17647863: MYSQL DOES NOT COMPILE ON OSX 10.9 GM Rename test() macro to MY_TEST() to avoid conflict with libc++. ------------------------------------------------------------ revno: 5571 committer: Sivert Sorumgard <sivert.sorumgaard@oracle.com> branch nick: 5.6-13548252 timestamp: Wed 2013-10-30 13:09:43 +0100 message: Bug #13548252: MYSQL ASSUMES ALL ETHERNET DEVICES ARE NAMED ETHX Bug #63055 : mysql assumes all ethernet devices are named ethX The MySQL server is using the first non-zero hardware address of an arbitrary network interface for generating UUIDs. On Linux, the MySQL server is only trying interface names "eth{0-5}". The patch retrieves the list of known interface names rather than assuming names of a certain pattern. Patch contributed by Honza Horak ------------------------------------------------------------ revno: 5570 committer: Jon Olav Hauglid <jon.hauglid@oracle.com> branch nick: mysql-5.6-bug17590095 timestamp: Wed 2013-10-30 09:55:14 +0100 message: Bug#17590095: COVERITY ANALYSIS RESULTS AND PATCHES Bug#70591 : COVERITY ANALYSIS RESULTS AND PATCHES Fix various issues found by Coverity. Some by backporting the fix for Bug#16725945: USE_AFTER_FREE ERRORS - COVERITY SCAN Based on patches contributed by Honza Horak. ------------------------------------------------------------ revno: 5569 committer: Jon Olav Hauglid <jon.hauglid@oracle.com> branch nick: mysql-5.6-bug16725769 timestamp: Wed 2013-10-30 09:30:46 +0100 message: Bug#16725769: VARARG ERRORS - COVERITY RESULTS Bug#68896 : VARARG Errors - Coverity results Add two missing va_end() to UCS2 code. Patch contributed by Jan Stan?k. ------------------------------------------------------------ revno: 5568 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-30 07:47:07 +0100 message: Enabled DWITH_INNODB_MEMCACHED in the spec file ------------------------------------------------------------ revno: 5567 committer: Venkatesh Duggirala<venkatesh.duggirala@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-30 11:04:38 +0530 message: Bug#17280176 TRANSACTIONS SKIPPED ON SLAVE AFTER "STOP/START SLAVE" USING GTID REPLICATION Analysis: Slave updates 'GTID_RETRIEVED' set upon receiving GTID_LOG_EVENT for a particular transaction which is first event in the event group. Say, I/O thread is stopped *after* adding GTID number to 'gtid_trieved' set and *before* it actually retrieves all the events from that GTID event group. Next time when this I/O thread is reconnected, it sends union of GTID_RETRIEVED + GTID_EXECUTED set to master. So Master thinks that slave is having all the events from this GTID set(which includes partially retrieved GTID) and it will not resend them again. Hence slave is missing some events for ever. Fix: It is not easy to find the end of a group of events. So mysql server is unsure whether I/O thread retrieved the last gtid transaction events completely or not (before it is going down because of a crash/normal shutdown/normal stop slave io_thread). It is possible that I/O thread would have retrieved and written only partial transaction events. So Server will request Master to send the last gtid event once again. We do this by removing the last I/O thread retrieved gtid event from "Retrieved_gtid_set". Possible cases: 1) I/O thread would have retrieved full transaction already in the first time itself, but retrieving them again will not cause problem because GTID number is same, Hence SQL thread will not commit it again. 2) I/O thread would have retrieved full transaction already and SQL thread would have already executed it. In that case, We are not going remove last retrieved gtid from "Retrieved_gtid_set" otherwise we will see gaps in "Retrieved set". ------------------------------------------------------------ revno: 5566 [merge] committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-30 04:15:35 +0100 message: Merge 5.5 => 5.6 - Removed non gpl file docs/mysql.info from community package ------------------------------------------------------------ revno: 2875.437.279 [merge] committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.5 timestamp: Wed 2013-10-30 04:12:24 +0100 message: Merge 5.1 => 5.5 ------------------------------------------------------------ revno: 2661.876.82 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.1 timestamp: Wed 2013-10-30 04:07:07 +0100 message: Removed non gpl file docs/mysql.info from community packages ------------------------------------------------------------ revno: 5565 [merge] committer: Tor Didriksen <tor.didriksen@oracle.com> branch nick: 5.6-merge timestamp: Tue 2013-10-29 19:57:24 +0100 message: null merge 5.5 => 5.6 ------------------------------------------------------------ revno: 2875.437.278 [merge] committer: Tor Didriksen <tor.didriksen@oracle.com> branch nick: 5.5-merge timestamp: Tue 2013-10-29 19:55:38 +0100 message: remerge 5.1 => 5.5 ------------------------------------------------------------ revno: 5564 [merge] committer: Tor Didriksen <tor.didriksen@oracle.com> branch nick: 5.6-merge timestamp: Tue 2013-10-29 19:41:28 +0100 message: merge 5.5 => 5.6 ------------------------------------------------------------ revno: 2875.459.1 [merge] committer: Tor Didriksen <tor.didriksen@oracle.com> branch nick: 5.5-merge timestamp: Tue 2013-10-29 17:54:49 +0100 message: Merge 5.1 => 5.5 ------------------------------------------------------------ revno: 2661.876.81 committer: Tor Didriksen <tor.didriksen@oracle.com> branch nick: 5.1 timestamp: Tue 2013-10-29 17:26:20 +0100 message: Bug#17326567 MYSQL SERVER FILESORT IMPLEMENTATION HAS A VERY SERIOUS BUG The filesort implementation needs space for at least 15 records (plus some internal overhead) in its main sort buffer. ------------------------------------------------------------ revno: 5563 [merge] committer: kevin.lewis@oracle.com branch nick: mysql-5.6 timestamp: Tue 2013-10-29 12:58:47 -0500 message: Merge up from mysql5.5 a fix for innodb-alter-restart.test ------------------------------------------------------------ revno: 2875.437.277 committer: kevin.lewis@oracle.com branch nick: mysql-5.5 timestamp: Tue 2013-10-29 11:39:04 -0500 message: Testcase i_innodb.innodb-alter-restart.test fails on some Windows platforms possibly since it was introduced in the patch for Bug#16720368 around 2013-04-30. This fix is simply to adjust the mtr.add_suppression() lines in the testcase and to add a missing "\n" in the error message. Approved by Marko in RB 3746 ------------------------------------------------------------ revno: 5562 committer: Shaohua Wang <shaohua.wang@oracle.com> branch nick: mysql-5.6-bugfix2 timestamp: Tue 2013-10-29 16:27:03 +0800 message: BUG#17607956 - FTS: MORE SQL PARSER PROBLEMS, ASSERTION IN IN FILE PARS0LEX.L LINE 686 The root cause is that innodb parser error occurs when parsing the database name. we assume database name starts by letter, not digit. The solution is quoting fts internal table name with double quote. rb#3742 is approved by Jimmy.Yang ------------------------------------------------------------ revno: 5561 [merge] committer: Mattias Jonsson <mattias.jonsson@oracle.com> branch nick: topush-5.6 timestamp: Mon 2013-10-28 17:25:16 +0100 message: post-push fix for bug#1756588/bug#14621190 ------------------------------------------------------------ revno: 2875.437.276 committer: Mattias Jonsson <mattias.jonsson@oracle.com> branch nick: test-5.5 timestamp: Mon 2013-10-28 17:08:31 +0100 message: post-push fix for bug#17565888/bug#14621190 ------------------------------------------------------------ revno: 5560 committer: Bjorn Munch <bjorn.munch@oracle.com> branch nick: main-56 timestamp: Mon 2013-10-28 15:01:49 +0100 message: RE ticket #46950 main.file_contents failing continuously on PB2 for 5.6 for elx-x86 platform. Fix test by adding glob pattern to also cover ULN rpms ------------------------------------------------------------ revno: 5559 committer: bin.x.su@oracle.com branch nick: mysql-5.6 timestamp: Sun 2013-10-27 11:36:57 +0800 message: BUG#16245498 SERVER CRASH WITH INNODB_BUFFER_POOL_SIZE<8M OR SMALL INNODB_SYNC_ARRAY_SIZE == Analysis == There are two different bugs(defects) here: 1. sync_array_reserve_cell will ut_error 2. sync_array_init will trigger ut_a(n_threads > srv_sync_array_size) == Solution == For 1, sync_array_reserve_cell should return true if a free cell is found, otherwise false, then the callers will choose another sync_array to reserve a free cell For 2, the assertion is too strict. Since the sync_wait_array doesn't require that each array has more than one slot, the assertion can be replaced by ut_a(n_threads > 0), how to use sync_array efficiently depends on users Also, this patch backports the fix for BUG#16884077 - ASSERT FAILURE IN SERVER WHEN LARGE NUMBER OF CLIENT EXECUTES QUERY(rb#2956). After this backport there shall not be any relationship between innodb_buffer_pool_size and number of threads. max_connections is introduced to get the max number of threads. rb#2464, approved by Jimmy. ------------------------------------------------------------ revno: 5558 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-10-25 13:13:15 +0200 message: updating the compatver to latest 5.1.72 ------------------------------------------------------------ revno: 5557 [merge] committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-10-25 11:40:06 +0200 message: - Merge latest repo code from 5.6.14-Fedora branch - Removed mysql-5.1.70.tar.gz from source - Updated CMakeLists.txt to remove mysql-5.1.70.tar.gz ------------------------------------------------------------ revno: 5406.1.14 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6.14-Fedora timestamp: Wed 2013-10-16 12:50:48 +0200 message: Added mysql.init to CMakeLists.txt ------------------------------------------------------------ revno: 5406.1.13 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6.14-Fedora timestamp: Wed 2013-10-16 11:42:54 +0200 message: Fixed mysql_install_db usage and improved handling of plugin directory ------------------------------------------------------------ revno: 5556 committer: Jimmy Yang <jimmy.yang@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-10-25 17:18:42 +0800 message: Fix Bug #17350055 - MEMORY ACCESS VIOLATION FOR A CERTAIN FTS BOOLEAN MODE QUERY rb://3668 approved by Sunny Bains ------------------------------------------------------------ revno: 5555 [merge] committer: Luis Soares <luis.soares@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-10-25 10:07:13 +0100 message: BUG#17460821 Automerged from mysql-5.5 into mysql-5.6. ------------------------------------------------------------ revno: 2875.437.275 committer: Luis Soares <luis.soares@oracle.com> branch nick: mysql-5.5 timestamp: Fri 2013-10-25 09:42:59 +0100 message: BUG#17460821: ASSERTION ERROR WHEN STOPPING SLAVE AFTER SEMI-SYNC ON MASTER IS DISABLED In the test case, make sure that the IO thread is not stopped during the binlog dump request. ------------------------------------------------------------ revno: 5554 [merge] committer: sayantan dutta <sayantan.dutta@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-10-25 11:54:50 +0530 message: 17654275 upmerge 5.5 => 5.6 ------------------------------------------------------------ revno: 2875.437.274 committer: sayantan dutta <sayantan.dutta@oracle.com> branch nick: mysql-5.5 timestamp: Fri 2013-10-25 11:38:34 +0530 message: Bug #17654275 - MTR EXTRACT_WARNING_LINES IS NOT REMOVING TIMESTAMP IN UTC FORMAT WITH TIMEZONES ------------------------------------------------------------ revno: 5553 committer: Shaohua Wang <shaohua.wang@oracle.com> branch nick: mysql-5.6-bugfix1 timestamp: Thu 2013-10-24 18:24:55 +0800 message: BUG#17402002 - FTS: INNODB: TRYING TO TRUNCATE A MISSING INDEX OF TABLE Supress the warning message when it's a fts index. rb#3683 approved by Jimmy.Yang ------------------------------------------------------------ revno: 5552 committer: Tanjot Uppal <tanjot.uppal@oracle.com> branch nick: mysql-5.6 timestamp: Thu 2013-10-24 14:56:21 +0530 message: Backporting from trunk to 5.6 the skipping logic for the below perfschema tests, with Query Cache Enabled. added: mysql-test/include/have_QC_Disabled.inc modified: mysql-test/suite/perfschema/t/aggregate.test mysql-test/suite/perfschema/t/event_aggregate.test mysql-test/suite/perfschema/t/event_aggregate_no_a.test mysql-test/suite/perfschema/t/event_aggregate_no_a_no_h.test mysql-test/suite/perfschema/t/event_aggregate_no_a_no_u.test mysql-test/suite/perfschema/t/event_aggregate_no_a_no_u_no_h.test mysql-test/suite/perfschema/t/event_aggregate_no_h.test mysql-test/suite/perfschema/t/event_aggregate_no_u.test mysql-test/suite/perfschema/t/event_aggregate_no_u_no_h.test mysql-test/suite/perfschema/t/nesting.test mysql-test/suite/perfschema/t/view_table_io.test ------------------------------------------------------------ revno: 5551 committer: Marko M?kel? <marko.makela@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-23 17:26:54 +0300 message: Fix a test failure. ------------------------------------------------------------ revno: 5550 committer: Ashish Agarwal<ashish.y.agarwal@oracle.com> branch nick: bug_16938568 timestamp: Wed 2013-10-23 15:47:54 +0530 message: Bug#16938568: ERROR 1819 (HY000): YOUR PASSWORD DOES NOT SATISFY THE CURRENT POLICY REQUIREMEN PROBLEM: Grant without indentified clause was not allowed even if the user was previously created. This bug is a regression caused by bug#16346443. We missed as case: allow to execute grant statement without identified clause when user exsist. SOLUTION: Allowing grant statement (without identified clause) to execute when validate_password plugin is on. ------------------------------------------------------------ revno: 5549 committer: Marko M?kel? <marko.makela@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-23 10:51:18 +0300 message: Bug#17624926 PREBUILT->TRX->ERROR_KEY_NUM UNINITIALIZED IN ERROR INJECTION ------------------------------------------------------------ revno: 5548 committer: Marko M?kel? <marko.makela@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-23 10:45:20 +0300 message: Bug#17512497 UPDATED ROWS REMAIN AFTER ONLINE DROP+ADD PRIMARY KEY Bug#17625063 REMOVE BOGUS FAULT TOLERANCE FROM ROW_LOG_TABLE_APPLY row_log_table_apply_update(): If the new PRIMARY KEY columns change (but the old PRIMARY KEY columns stay unchanged), always perform a delete followed by an insert. It should be safe to skip the trx_id comparison altogether, because row_merge_read_clustered_index() is performing a REPEATABLE READ when doing the initial copying. row_log_table_apply_convert_mrec(): Remove an outdated comment. We no longer track rollbacks. row_log_table_apply_update(): Remove the parameter trx_id_col. There is no need to compare the trx_id. Throw an error if the record unexpectedly exists or does not exist. row_log_apply_op_low(): Add comments that describe the situations when a record already exists or does not exist. On duplicate key error, abort the log application immediately by flagging DB_DUPLICATE_KEY. row_log_apply(): Do not flag DB_DUPLICATE_KEY, because it will now be flagged in row_log_apply_op_low(). rb#3638 approved by Jimmy Yang ------------------------------------------------------------ revno: 5547 committer: Shaohua Wang <shaohua.wang@oracle.com> branch nick: mysql-5.6-bugfix1 timestamp: Tue 2013-10-22 11:22:38 +0800 message: BUG#17456092 - FAILING ASSERTION: RESULT != FTS_INVALID The root cause is that we remove the savepoint when rollbacking to a savepoint. The fix is to restore the savepoint after the removal. rb#3652 approved by Jimmy.Yang ------------------------------------------------------------ revno: 5546 [merge] committer: Jon Olav Hauglid <jon.hauglid@oracle.com> branch nick: mysql-5.6-test timestamp: Mon 2013-10-21 17:43:15 +0200 message: Merge from mysql-5.5 to mysql-5.6 ------------------------------------------------------------ revno: 2875.437.273 committer: Jon Olav Hauglid <jon.hauglid@oracle.com> branch nick: mysql-5.5-bug17632386 timestamp: Mon 2013-10-21 13:14:46 +0200 message: Bug#17632386: WITH_ASAN CONFLICTS WITH -WL,--NO-UNDEFINED -Wl,--no-undefined (=-z defs) gives linking errors when used with WITH_ASAN. According to the documentation: "When linking shared libraries, the AddressSanitizer run-time is not linked, so -Wl,-z,defs may cause link errors (don?t use it with AddressSanitizer)." This patch turns off -Wl,--no-undefined if WITH_ASAN is used. ------------------------------------------------------------ revno: 5545 [merge] committer: Aditya A <aditya.a@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-10-21 12:14:47 +0530 message: Bug #16051817 GOT ERROR 124 FROM STORAGE ENGINE ON DELETE FROM A PARTITIONED TABLE [Merge from mysql-5.5] ------------------------------------------------------------ revno: 2875.437.272 committer: Aditya A <aditya.a@oracle.com> branch nick: mysql-5.5 timestamp: Mon 2013-10-21 12:07:02 +0530 message: Bug #16051817 GOT ERROR 124 FROM STORAGE ENGINE ON DELETE FROM A PARTITIONED TABLE PROBLEM ------- The user first disables all the non unique indexes in the table and then rebuilds one partition. During rebuild the indexes on that particular partition are enabled. Now when we give a query the optimizer is unaware that on one partition indexes are enabled and if the optimizer selects that index,myisam thinks that the index is not active and gives an error. FIX --- Before rebuilding a partition check whether non unique indexes are disabled on the partitons. If they are disabled then after rebuild disable the index on the partition. [Approved by Mattiasj #rb3469] ------------------------------------------------------------ revno: 5544 committer: Venkatesh Duggirala<venkatesh.duggirala@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-10-21 11:20:23 +0530 message: Bug #17324516 SLAVE ASSERTS AFTER 5.6 UPGRADE Problem: In place upgrade of slave from mysql-5.5 to mysql-5.6.12 fails with assertion (lines >= LINES_IN_RELAY_LOG_INFO_WITH_ID && internal_id == 1). Analysis: mysql-5.5's relay-log.info format is different from mysql-5.6's format. mysql-5.5 version does not have some information like the number of lines in the file(1st line), sql_delay (6th line), number of parallel workers which is used when MTS is enabled (7th line), internal_id which is also used when MTS is enabled (8th line). While a mysql-5.6 slave server is coming up, it parses relay-log.info file (if already available) and regenerates a new content if the format is old (i.e., mysql-5.5's format). In the process of regeneration, a local variable temp_internal_id with a default value 0 is declared. If the file content is new format, then the value was read into temp_internal_id and Rpl_info::internal_id is assigned with this local temp_internal_id. If the file content is old format, then the default value '0' was assigned to Rpl_info::internal_id which is wrong value. Fix: Initialize temp_internal_id with Rpl_info::internal_id which is having the default value '1'. ------------------------------------------------------------ revno: 5543 committer: Jimmy Yang <jimmy.yang@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-10-21 11:56:33 +0800 message: Fix Bug #17430207 FTS: FAILING ASSERTION: SYM_NODE->TABLE != NULL AFTER DISCARDING TABLESPAC rb://3603 approved by Sunny Bains ------------------------------------------------------------ revno: 5542 [merge] committer: Mattias Jonsson <mattias.jonsson@oracle.com> branch nick: topush-5.6 timestamp: Sat 2013-10-19 16:11:14 +0200 message: post-push fix for bug15765888 ------------------------------------------------------------ revno: 2875.437.271 committer: Mattias Jonsson <mattias.jonsson@oracle.com> branch nick: test-5.5 timestamp: Sat 2013-10-19 15:29:51 +0200 message: post-push fix for bug17565888. ------------------------------------------------------------ revno: 5541 [merge] committer: Mattias Jonsson <mattias.jonsson@oracle.com> branch nick: topush-5.6 timestamp: Fri 2013-10-18 23:51:12 +0200 message: post-push fix for bug17565888 ------------------------------------------------------------ revno: 2875.437.270 committer: Mattias Jonsson <mattias.jonsson@oracle.com> branch nick: test-5.5 timestamp: Fri 2013-10-18 23:49:35 +0200 message: post-push fix for bug17565888. Too restrictive assertion, failing during purge ------------------------------------------------------------ revno: 5540 [merge] committer: Mattias Jonsson <mattias.jonsson@oracle.com> branch nick: topush-5.6 timestamp: Fri 2013-10-18 23:40:43 +0200 message: post-push fix for bug17565888. ------------------------------------------------------------ revno: 2875.437.269 committer: Mattias Jonsson <mattias.jonsson@oracle.com> branch nick: test-5.5 timestamp: Fri 2013-10-18 23:39:15 +0200 message: post-push fix for bug17565888. Too restrictive assertion, can fail during purge ------------------------------------------------------------ revno: 5539 [merge] committer: Mattias Jonsson <mattias.jonsson@oracle.com> branch nick: topush-5.6 timestamp: Fri 2013-10-18 19:27:45 +0200 message: manual merge of bug17565888 to mysql-5.6 ------------------------------------------------------------ revno: 2875.437.268 committer: Mattias Jonsson <mattias.jonsson@oracle.com> branch nick: test-5.5 timestamp: Fri 2013-10-18 19:25:59 +0200 message: Bug#17565888: UP TO 75% INCREASE IN DBT3 QUERY (Q7) EXECUTION TIME Regression from bug#14621190 due to disabled optimistic restoration of cursor, which required full key lookup instead of verifying if previously positioned btree cursor could be reused. Fixed by enable optimistic restore and adjust cursor afterward. rb#3324 approved by Marko. ------------------------------------------------------------ revno: 5538 [merge] committer: Anirudh Mangipudi <anirudh.mangipudi@oracle.com> branch nick: 5.6 timestamp: Fri 2013-10-18 17:15:19 +0530 message: Bug#17357535 BACKPORT BUG#16241992 TO 5.5 Null Merge from 5.5 ------------------------------------------------------------ revno: 2875.437.267 committer: Anirudh Mangipudi <anirudh.mangipudi@oracle.com> branch nick: 5.5 timestamp: Fri 2013-10-18 17:14:39 +0530 message: Bug #17357535 BACKPORT BUG#16241992 TO 5.5 Problem: COM_CHANGE_USER allows brute-force attempts to crack a password at a very high rate as it does not cause any significant delay after a login attempt has failed. This issue was reproduced using John-The-Ripper password cracking tool through which about 5000 passwords per second could be attempted. Solution: The non-GA version's solution was to disconnect the connection when a login attempt failed. Now since our aim to to reduce the rate at which passwords are tested, we introduced a sleep(1) after every login attempt failed. This significantly increased the delay with which the password was cracked. ------------------------------------------------------------ revno: 5537 [merge] committer: Luis Soares <luis.soares@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-10-18 11:32:00 +0100 message: BUG#17460821 Automerged mysql-5.5 into latest mysql-5.6. (Empty merge.) ------------------------------------------------------------ revno: 2875.437.266 [merge] committer: Luis Soares <luis.soares@oracle.com> branch nick: mysql-5.5 timestamp: Fri 2013-10-18 10:50:49 +0100 message: BUG#17460821 Automerged from mysql-5.5 bug branch into latest mysql-5.5. ------------------------------------------------------------ revno: 2875.458.1 committer: Luis Soares <luis.soares@oracle.com> branch nick: mysql-5.5 timestamp: Thu 2013-10-17 17:48:26 +0100 message: BUG#17460821: ASSERTION ERROR WHEN STOPPING SLAVE AFTER SEMI-SYNC ON MASTER IS DISABLED The assertion happens when: (i) the master and slave are configured to use the semisync plugin; (ii) the DBA disables semisync on the master; (iii) and he also unsets the option to wait for slaves ACK even if the semisync slave count reaches 0 during the waiting period. This combination of factors makes the server run into an assertion as soon as the last semisync slave disconnects and its dump thread exits. The root of the problem is the fact that when the dump thread disconnects and calls the observer hook transmit_stop, which ends up calling ReplSemiSyncMaster::remove_slave, there is no check whether the master has already disabled semisync or not. If it has, the then a second call to the switch_off member function must be avoided. The quick fix is to avoid calling switch_off if the DBA has disabled the semisync plugin interactively on the master. Also, the switch_off member function should only be called if the plugin has not been switched off already. This is basically the pattern throughout the rest of the semisync plugin and no other calls seem vulnerable to similar crashes/assertions. (This a backport of the patch to 5.5, which is also vulnerable.) ------------------------------------------------------------ revno: 5536 committer: Shaohua Wang <shaohua.wang@oracle.com> branch nick: mysql-5.6-bugfix1 timestamp: Fri 2013-10-18 17:53:12 +0800 message: BUG#17384979 - YIELD ON SCANNER THREAD FOR PARALLEL FTS INDEX CREATE TO MINIMIZE MEMORY USAGE Follow up fix: Modify wrong tokenizer thread exit check condition. Approved by Jimmy.Yang in IM. ------------------------------------------------------------ revno: 5535 [merge] committer: Aditya A <aditya.a@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-10-18 15:02:18 +0530 message: Bug#17559867 AFTER REBUILDING , A MYISAM PARTITION ENDS UP AS A INNODB PARTITTION. [Merged from 5.5] ------------------------------------------------------------ revno: 2875.437.265 [merge] committer: Aditya A <aditya.a@oracle.com> branch nick: mysql-5.5 timestamp: Fri 2013-10-18 13:49:03 +0530 message: Bug#17559867 AFTER REBUILDING , A MYISAM PARTITION ENDS UP AS A INNODB PARTITTION. [Merged from 5.1] ------------------------------------------------------------ revno: 2661.876.80 committer: Aditya A <aditya.a@oracle.com> branch nick: mysql-5.1 timestamp: Fri 2013-10-18 12:26:28 +0530 message: Bug#17559867 AFTER REBUILDING,A MYISAM PARTITION ENDS UP AS A INNODB PARTITTION. PROBLEM ------- The correct engine_type was not being set during rebuild of the partition due to which the handler was always created with the default engine, which is innodb for 5.5+ ,therefore even if the table was myisam, after rebuilding the partitions ended up as innodb partitions. FIX --- Set the correct engine type during rebuild. [Approved by mattiasj #rb3599] ------------------------------------------------------------ revno: 5534 committer: Jimmy Yang <jimmy.yang@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-10-18 14:37:11 +0800 message: Fix a cast issue that breaks windows' build with #17371537 checkin ------------------------------------------------------------ revno: 5533 committer: Shaohua Wang <shaohua.wang@oracle.com> branch nick: mysql-5.6-bugfix1 timestamp: Fri 2013-10-18 13:56:04 +0800 message: BUG#17536995 - ASSERT RBT_EMPTY(INDEX_CACHE->WORDS) IN ALTER TABLE CHANGE COLUMN The root cause is that if a index is to_be_dropped, words is not freed, so that we get assert failure in fts_cache_clear while free_words is false. In the fix, we remove free_words and assert in fts_cache_clear, and also fix a bug related to DICT_TF2_FTS. rb#3623 approved by Jimmy.Yang ------------------------------------------------------------ revno: 5532 committer: Jimmy Yang <jimmy.yang@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-10-18 11:25:17 +0800 message: Fix bug #17371537 TRX_CREATE() AND TRX_FREE() ARE CALLED AT EVERY MEMCACHED GET REQUES rb://3505 approved by Sunny Bains ------------------------------------------------------------ revno: 5531 [merge] committer: Luis Soares <luis.soares@oracle.com> branch nick: mysql-5.6 timestamp: Thu 2013-10-17 16:22:33 +0100 message: BUG#17460821 Automerged mysql-5.6 bug branch into latest mysql-5.6. ------------------------------------------------------------ revno: 5476.1.1 committer: Luis Soares <luis.soares@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-09-30 14:16:56 +0100 message: BUG#17460821: ASSERTION ERROR WHEN STOPPING SLAVE AFTER SEMI-SYNC ON MASTER IS DISABLED The crash happens when: (i) the master and slave are configured to use the semisync plugin; (ii) the DBA disables semisync on the master; (iii) and he also unsets the option to wait for slaves ACK even if the semisync slave count reaches 0 during the waiting period. This combination of factors makes the server run into an assertion and crash as soon as the last semisync slave disconnects and its dump thread exits. The root of the problem is the fact that when the dump thread disconnects and calls the observer hook transmit_stop, which ends up calling ReplSemiSyncMaster::remove_slave, there is no check whether the master has already disabled semisync or not. If it has, the then a second call to the switch_off member function must be avoided. The quick fix is to avoid calling switch_off if the DBA has disabled the semisync plugin interactively on the master. Also, the switch_off member function should only be called if the plugin has not been switched off already. This is basically the pattern throughout the rest of the semisync plugin and no other calls seem vulnerable to similar crashes/assertions. ------------------------------------------------------------ revno: 5530 [merge] committer: Anil Toshniwal <anil.toshniwal@oracle.com> branch nick: mysql-5.6 timestamp: Thu 2013-10-17 18:13:36 +0530 message: Bug#17513737 INTRODUCE CHECK TABLE...QUICK Upmerge from 5.5 -> 5.6 ------------------------------------------------------------ revno: 2875.437.264 committer: Anil Toshniwal <anil.toshniwal@oracle.com> branch nick: mysql-5.5 timestamp: Thu 2013-10-17 18:09:04 +0530 message: Bug#17513737 INTRODUCE CHECK TABLE...QUICK --Implemented CHECK TABLE...QUICK. Introduce CHECK TABLE...QUICK that would skip the btr_validate_index() and btr_search_validate() call, and count the no. of records in each index. Approved by Marko and Kevin. (rb#3567). ------------------------------------------------------------ revno: 5529 [merge] committer: Praveenkumar Hulakund <praveenkumar.hulakund@oracle.com> branch nick: mysql-5.6 timestamp: Thu 2013-10-17 18:01:24 +0530 message: Bug#17474166 - EXECUTING STATEMENT LIKE 'SHOW ENGINE INNODB' Merging from 5.5 to 5.6. ------------------------------------------------------------ revno: 2875.437.263 committer: Praveenkumar Hulakund <praveenkumar.hulakund@oracle.com> branch nick: mysql-5.5 timestamp: Thu 2013-10-17 17:59:45 +0530 message: Bug#17474166 - EXECUTING STATEMENT LIKE 'SHOW ENGINE INNODB' AND 'KILL SESSION' LEAD TO CRASH Fixing pb2 failure on embedded version. ------------------------------------------------------------ revno: 5528 [merge] committer: Luis Soares <luis.soares@oracle.com> branch nick: mysql-5.6 timestamp: Thu 2013-10-17 12:29:17 +0100 message: BUG#17508351 Up merged mysql-5.5 to mysql-5.6. ------------------------------------------------------------ revno: 2875.437.262 [merge] committer: Luis Soares <luis.soares@oracle.com> branch nick: mysql-5.5 timestamp: Thu 2013-10-17 12:27:45 +0100 message: BUG#17508351 Merging mysql-5.5 bug branch into latest mysql-5.5. ------------------------------------------------------------ revno: 2875.457.1 committer: Luis Soares <luis.soares@oracle.com> branch nick: mysql-5.5 timestamp: Tue 2013-10-08 18:37:43 +0100 message: BUG#17508351: REDUNDANT CODE IN REPLSEMISYNCMASTER::UPDATESYNCHEADER ReplSemiSyncMaster::updateSyncHeader contains redundant assignments to the local variable sync. This patch removes them. ------------------------------------------------------------ revno: 5527 committer: Nuno Carvalho <nuno.carvalho@oracle.com> branch nick: mysql-5.6 timestamp: Thu 2013-10-17 10:09:29 +0100 message: WL#6663: Make the MTR binlog suite GTID_MODE agnostic Removed test binlog_tmp_table_gtid, which forced execution with gtid-mode=ON, since we already run all binlog suite with that option. ------------------------------------------------------------ revno: 5526 [merge] committer: Marc Alff <marc.alff@oracle.com> branch nick: mysql-5.6-push2 timestamp: Wed 2013-10-16 23:41:36 +0200 message: Push to mysql-5.6 ------------------------------------------------------------ revno: 5519.1.1 committer: Marc Alff <marc.alff@oracle.com> branch nick: mysql-5.6-bug17542370 timestamp: Wed 2013-10-16 12:28:59 +0200 message: Bug#17542370 SEGMENTATION FAULT WHILE EXECUTING SELECT ON SESSION_CONNECT_ATTRS Before this fix, executing a select on table performance_schema.session_connect_attrs could crash the server, when executed under load. The root cause is that the code that scans records: - does not honor the PFS_thread::m_sesion_lock lock, - does parse the live session attribute data, while it can possibly be modified Race conditions could occur: - when writing session attributes, shortly after a thread connect - when erasing session attributes, shortly before a thread disconnect The fix is to: - use an optimist lock on PFS_thread::m_sesion_lock - make a copy of the session connect attributes, - only when the optimist lock was successful, (meaning the data was not concurently modified), parse the safe copy of the content of it to extract attributes. ------------------------------------------------------------ revno: 5525 committer: Marko M?kel? <marko.makela@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-16 22:17:44 +0300 message: Correct the merge from mysql-5.5 to mysql-5.6 of Bug#17446090 TESTCASE INNODB_BUG16752251.TEST FAILS RANDOMLY ON PB2 dict_stats_update_transient_for_index(): Skip the statistics on secondary indexes if innodb_change_buffering_debug (ibuf_debug) is set. This should get rid of change buffering test failures when using PERSISTENT_STATS=0 on MySQL 5.6 and later. Approved on IM by Kevin Lewis ------------------------------------------------------------ revno: 5524 [merge] committer: Venkatesh Duggirala<venkatesh.duggirala@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-16 22:20:19 +0530 message: Bug#17234370 LAST_INSERT_ID IS REPLICATED INCORRECTLY IF REPLICATION FILTERS ARE USED. Merging fix from mysql-5.5 ------------------------------------------------------------ revno: 2875.437.261 [merge] committer: Venkatesh Duggirala<venkatesh.duggirala@oracle.com> branch nick: mysql-5.5 timestamp: Wed 2013-10-16 22:15:59 +0530 message: Bug#17234370 LAST_INSERT_ID IS REPLICATED INCORRECTLY IF REPLICATION FILTERS ARE USED. Merging fix from mysql-5.1 ------------------------------------------------------------ revno: 2661.876.79 committer: Venkatesh Duggirala<venkatesh.duggirala@oracle.com> branch nick: mysql-5.1 timestamp: Wed 2013-10-16 22:12:23 +0530 message: Bug#17234370 LAST_INSERT_ID IS REPLICATED INCORRECTLY IF REPLICATION FILTERS ARE USED. Problem: When Filtered-slave applies Int_var_log_event and when it tries to write the event to its own binlog, LAST_INSERT_ID value is written wrongly. Analysis: THD::stmt_depends_on_first_successful_insert_id_in_prev_stmt is a variable which is set when LAST_INSERT_ID() is used by a statement. If it is set, first_successful_insert_id_in_ prev_stmt_for_binlog will be stored in the statement-based binlog. This variable is CUMULATIVE along the execution of a stored function or trigger: if one substatement sets it to 1 it will stay 1 until the function/trigger ends, thus making sure that first_successful_insert_id_in_ prev_stmt_for_binlog does not change anymore and is propagated to the caller for binlogging. This is achieved using the following code if(!stmt_depends_on_first_successful_insert_id_in_prev_stmt) { /* It's the first time we read it */ first_successful_insert_id_in_prev_stmt_for_binlog= first_successful_insert_id_in_prev_stmt; stmt_depends_on_first_successful_insert_id_in_prev_stmt= 1; } Slave server, after receiving Int_var_log_event event from master, it is setting stmt_depends_on_first_successful_insert_id_in_prev_stmt to true(*which is wrong*) and not setting first_successful_insert_id_in_prev_stmt_for_binlog. Because of this problem, when the actual DML statement with LAST_INSERT_ID() is parsed by slave SQL thread, first_successful_insert_id_in_prev_stmt_for_binlog is not set. Hence the value zero (default value) is written to slave's binlog. Why only *Filtered slave* is effected when the code is in common place: ------------------------------------------------------- In Query_log_event::do_apply_event, THD::stmt_depends_on_first_successful_insert_id_in_prev_stmt is reset to zero at the end of the function. In case of normal slave (No Filters), this variable will be reset. In Filtered slave, Slave SQL thread defers all IRU events's execution until IRU's Query_log event is received. Once it receives Query_log_event it executes all pending IRU events and then it executes Query_log_event. Hence the variable is not getting reset to 0, causing this bug. Fix: As described above, the root cause was setting THD::stmt_depends_on_first_successful_insert_id_in_prev_stmt when Int_var_log_event was executed by a SQL thread. Hence removing the problematic line from the code. ------------------------------------------------------------ revno: 5523 [merge] committer: Venkata Sidagam <venkata.sidagam@oracle.com> branch nick: 5.6 timestamp: Wed 2013-10-16 14:20:09 +0530 message: Bug#16900358 FIX FOR CVE-2012-5611 IS INCOMPLETE Merging from mysql-5.5 to mysql-5.6 ------------------------------------------------------------ revno: 2875.437.260 [merge] committer: Venkata Sidagam <venkata.sidagam@oracle.com> branch nick: 5.5 timestamp: Wed 2013-10-16 14:16:32 +0530 message: Bug#16900358 FIX FOR CVE-2012-5611 IS INCOMPLETE Merging from mysql-5.1 to mysql-5.5 ------------------------------------------------------------ revno: 2661.876.78 committer: Venkata Sidagam <venkata.sidagam@oracle.com> branch nick: 5.1 timestamp: Wed 2013-10-16 14:14:44 +0530 message: Bug#16900358 FIX FOR CVE-2012-5611 IS INCOMPLETE Description: Fix for bug CVE-2012-5611 (bug 67685) is incomplete. The ACL_KEY_LENGTH-sized buffers in acl_get() and check_grant_db() can be overflown by up to two bytes. That's probably not enough to do anything more serious than crashing mysqld. Analysis: In acl_get() when "copy_length" is calculated it just adding the variable lengths. But when we are using them with strmov() we are adding +1 to each. This will lead to a three byte buffer overflow (i.e two +1's at strmov() and one byte for the null added by strmov() function). Similarly it happens for check_grant_db() function as well. Fix: We need to add "+2" to "copy_length" in acl_get() and "+1" to "copy_length" in check_grant_db(). ------------------------------------------------------------ revno: 5522 [merge] committer: Sujatha Sivakumar <sujatha.sivakumar@oracle.com> branch nick: Bug17429677_mysql-5.6 timestamp: Wed 2013-10-16 11:52:57 +0530 message: Merge from mysql-5.5 to mysql-5.6. ------------------------------------------------------------ revno: 2875.437.259 committer: Sujatha Sivakumar <sujatha.sivakumar@oracle.com> branch nick: Bug17429677_mysql-5.5 timestamp: Wed 2013-10-16 11:49:00 +0530 message: Bug#17429677:LAST ARGUMENT OF LOAD DATA ...SET ...STATEMENT REPEATED TWICE IN BINLOG Problem: ======= If LOAD DATA ... SET ... is used the last argument of SET is repeated twice in replication binlog. Analysis: ======== LOAD DATA statements are reconstructed once again before they are written to the binary log. When SET clauses are specified as part of LOAD DATA statement, these SET clause user command strings need to be stored in order to rebuild the original user command. During parsing each column and the value in the SET command are stored in two differenet lists. All the values are stored in a string list. When SET expression has more than one value as shown in the following example: SET a = @a, b = CONCAT(@b, '| 123456789'); Parser extracts values in the following manner i.e Item name , value string, actual length of the value of the item with in the string. Item a: Value for a:"= @a, b = CONCAT(@b, '| 123456789') str_length = 4 Item b: Value for b:"= CONCAT(@b, '| 123456789') str_length = 27 During reconstructing the LOAD DATA command the above strings are retrived as it is and appended to the LOAD DATA statement. Hence it becomes as shown below. SET `a`= @a, b = CONCAT(@b, '| 123456789'), `b`= CONCAT(@b, '| 123456789') Fix: === During reconstruction of SET command, retrieve exact item value string rather than reading the entire string. ------------------------------------------------------------ revno: 5521 [merge] committer: Sreedhar.S <sreedhar.sreedhargadda@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-16 11:07:24 +0530 message: Upmerge for bug 17584523 ------------------------------------------------------------ revno: 2875.437.258 committer: Sreedhar.S <sreedhar.sreedhargadda@oracle.com> branch nick: mysql-5.5 timestamp: Wed 2013-10-16 11:05:20 +0530 message: Fix for Bug 17584523 - LICENSE AND DOCUMENTATION FILES MISSING FROM WINDOWS MSI PACKAGE ------------------------------------------------------------ revno: 5520 committer: Atanu Ghosh <atanu.ghosh@oracle.com> branch nick: mysql-5.6 timestamp: Tue 2013-10-15 19:08:47 +0530 message: Bug#14040071:RECURSION LEADS TO STACK OVERFLOW WITH XML FUNCTIONS Problem: The problem is that when the XPATH expression exceeds limit, there is a stack overflow. Solution: As a fix, a call to check_stack_overrun was added, which throws the required error on stackoverflow. ------------------------------------------------------------ revno: 5519 committer: Sreedhar.S <sreedhar.sreedhargadda@oracle.com> branch nick: mysql-5.6 timestamp: Tue 2013-10-15 15:34:04 +0530 message: fix to to put mysql_install_db.pl back in windows installation ------------------------------------------------------------ revno: 5518 [merge] committer: Nuno Carvalho <nuno.carvalho@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-10-14 15:57:06 +0100 message: WL#7266: Dump-thread additional concurrency tests Merge from mysql-5.5 into mysql-5.6. ------------------------------------------------------------ revno: 2875.437.257 [merge] committer: Nuno Carvalho <nuno.carvalho@oracle.com> branch nick: mysql-5.5 timestamp: Mon 2013-10-14 15:51:09 +0100 message: WL#7266: Dump-thread additional concurrency tests Merge from mysql-5.1 into mysql-5.5. ------------------------------------------------------------ revno: 2661.876.77 committer: Nuno Carvalho <nuno.carvalho@oracle.com> branch nick: mysql-5.1 timestamp: Mon 2013-10-14 15:45:12 +0100 message: WL#7266: Dump-thread additional concurrency tests This worklog aims at testing the two following scenarios: 1) Whenever the mysql_binlog_send method (dump thread) reaches the end of file when reading events from the binlog, before checking if it should wait for more events, there was a test to check if the file being read was still active, i.e, it was the last known binlog. However, it was possible that something was written to the binary log and then a rotation would happen, after EOF was detected and before the check for active was performed. In this case, the end of the binary log would not be read by the dump thread, and this would cause the slave to lose updates. This test verifies that the problem has been fixed. It waits during this window while forcing a rotation in the binlog. 2) Verify dump thread can send events in active file, correctly after encountering an IO error. ------------------------------------------------------------ revno: 5517 committer: Shaohua Wang <shaohua.wang@oracle.com> branch nick: mysql-5.6-bugfix1 timestamp: Mon 2013-10-14 18:12:00 +0800 message: Follow up fix for bug #17483582 - EXCESSIVE MEMORY USAGE WHEN QUERYING INFORMATION_SCHEMA.INNODB_FT_INDEX." initialize the total_memory in fts index optimization code path. Approved by Jimmy.Yang in IM. ------------------------------------------------------------ revno: 5516 committer: Marc Alff <marc.alff@oracle.com> branch nick: mysql-5.6-push timestamp: Sat 2013-10-12 15:24:01 +0200 message: Build and test cleanup ------------------------------------------------------------ revno: 5515 committer: Shaohua Wang <shaohua.wang@oracle.com> branch nick: mysql-5.6-bugfix1 timestamp: Sat 2013-10-12 11:09:38 +0800 message: BUG#17483582 - EXCESSIVE MEMORY USAGE WHEN QUERYING INFORMATION_SCHEMA.INNODB_FT_INDEX Analysis & Solution: We first fetch all fts index records into memory and then store them into myisam temp table, so it will consumes huge memory when the fts index is big enough. In this patch, we limit the memory usage by INNODB_FT_RESULT_CACHE_LIMIT. When we reach the limit, do more fetches until we have all records rb://3398 approved by Jimmy.Yang ------------------------------------------------------------ revno: 5514 committer: Marc Alff <marc.alff@oracle.com> branch nick: mysql-5.6-push timestamp: Sat 2013-10-12 04:17:21 +0200 message: Fixed number of rwlock, adjusted tests ------------------------------------------------------------ revno: 5513 [merge] committer: Marc Alff <marc.alff@oracle.com> branch nick: mysql-5.6-push timestamp: Sat 2013-10-12 01:22:41 +0200 message: Merge to mysql-5.6 ------------------------------------------------------------ revno: 5510.1.1 committer: Marc Alff <marc.alff@oracle.com> branch nick: mysql-5.6-bug17590161 timestamp: Fri 2013-10-11 17:31:42 +0200 message: Bug#17590161 READ/WRITE MUTEXES ON BINLOG DELEGATE CLASSES ARE NOT COUNTED ON PERF SCHEMA Prior to this fix, the read write lock Delegate::lock was not instrumented for the performance schema. This lock is used in particular for the following classes: - Trans_delegate - Binlog_storage_delegate - Binlog_transmit_delegate - Binlog_relay_IO_delegate With this fix, the read write lock is instrumented. For clarity, a different name is used for each sub class, to have distinct statistics for distinct uses. The instruments added are visible in table performance_schema.setup_instruments as: - wait/synch/rwlock/sql/Trans_delegate::lock - wait/synch/rwlock/sql/Binlog_storage_delegate::lock - wait/synch/rwlock/sql/Binlog_transmit_delegate::lock - wait/synch/rwlock/sql/Binlog_relay_IO_delegate::lock ------------------------------------------------------------ revno: 5512 committer: Marko M?kel? <marko.makela@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-10-11 21:55:53 +0300 message: Bug#16884217 BTR_VALIDATE_LEVEL() FAILS ON THE CHANGE BUFFER TREE ibuf_restore_pos(): If the change buffer cursor restoration fails, do not invoke btr_validate_index() on the change buffer, but instead crash as fast as possible, so that we would have more useful information in the core dump. When WL#5522 (transportable tablespaces) was pushed, it added checks to btr_validate_level(), checking that each page is marked as allocated. Because the allocation of change buffer pages is handled differently from other B-tree pages, these checks would fail on the change buffer. This was the only call to btr_validate_index() on the change buffer. Normally, btr_validate_index() is invoked for indexes of user tables, as part of CHECK TABLE. Approved by Jimmy Yang over IM. ------------------------------------------------------------ revno: 5511 [merge] committer: Evgeny Potemkin <evgeny.potemkin@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-10-11 12:34:14 +0400 message: Auto-merged fix for the bug#17566396. ------------------------------------------------------------ revno: 5500.1.1 committer: Evgeny Potemkin <evgeny.potemkin@oracle.com> branch nick: bug17566396-5.6 timestamp: Fri 2013-10-11 12:26:18 +0400 message: Bug#17566396: MATERIALIZATION IS NOT CHOSEN FOR LONG UTF8 VARCHAR FIELD Semi-join materialization relies on heap tables, and thus have to take into account their limitations. Among other things, heap tables can't handle blob objects, so if a string field would be converted to a blob, semijoin materialization strategy can't be used. To check that, Item::is_blob_field() function is used. It compares item's max_length with CONVERT_IF_BIGGER_TO_BLOB constant. The cause of the bug is that byte length was used in comparison instead of char length. This made sj mat strategy unavailable for varchar fields in utf8 longer than 170 chars. Now Item::is_blob_field() uses char length in comparison. Also create_tmp_table() now uses char length to make decision on whether it should use index for grouping or revert to unique constraint for blobs. ------------------------------------------------------------ revno: 5510 committer: Joao Gramacho <joao.gramacho@oracle.com> branch nick: mysql-5.6 timestamp: Thu 2013-10-10 18:15:38 +0100 message: Bug#17032712 SET @@GLOBAL.GTID_PURGED='UUID:1-1000000000000000000' ASSERTS Problem: ======= The get_string_length() function is failing to assert the correct string length of numbers with 19 digits. Analysis: ======== The get_string_length() function returns the length a string must have in order to store a given number. It is used to calculate the string length of a GTID set. Because this function is static at rpl_gtid_set.cc, it is not trivial to make unit tests to verify its functionality. So, there is an assert executed only in debug enabled versions that checks if the value to be returned is equal to the length of printing the number in a string with snprintf(). Because of the precision of the rpl_gno data type (int64), an auxiliary variable may overflow when the number to be evaluated has 19 digits. The currently implemented code doesn't take into account this overflow possibility, making that every 19 digit number be evaluated as having 20 digits in length. This divergence cause the assert to fail. Fix: === Changed the function code replacing the length calculation algorithm with one that is simple, easy to read and overflow free. ------------------------------------------------------------ revno: 5509 [merge] committer: Satya Bodapati <satya.bodapati@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-09 18:44:35 +0530 message: Merge the testcase fix for BUG#17446090 from mysql-5.5 to mysql-5.6 ------------------------------------------------------------ revno: 2875.437.256 committer: Satya Bodapati <satya.bodapati@oracle.com> branch nick: mysql-5.5 timestamp: Wed 2013-10-09 18:42:27 +0530 message: BUG#17446090 - TESTCASE INNODB_BUG16752251.TEST FAILS RANDOMLY ON PB2 DBUG_SUICIDE() hangs valgrind. Disable the test under valgrind ------------------------------------------------------------ revno: 5508 [merge] committer: Sreedhar.S <sreedhar.sreedhargadda@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-09 17:12:58 +0530 message: Null Merging the changes from 5.5 ------------------------------------------------------------ revno: 2875.437.255 committer: Sreedhar.S <sreedhar.sreedhargadda@oracle.com> branch nick: mysql-5.5 timestamp: Wed 2013-10-09 17:08:45 +0530 message: BUG 17560050 - MYSQL_INSTALL_DB SCRIPT FAILING WITH RESOLVEIP ISSUES. ------------------------------------------------------------ revno: 5507 committer: Murthy Narkedimilli <murthy.narkedimilli@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-09 16:47:06 +0530 message: Bug# 16919882: Changing the copyright header format ------------------------------------------------------------ revno: 5506 committer: Shaohua Wang <shaohua.wang@oracle.com> branch nick: mysql-5.6-bugfix2 timestamp: Wed 2013-10-09 19:00:56 +0800 message: BUG#17458835 - INNODB FULLTEXT SEARCH DOESN'T FIND RECORDS WHEN SAVEPOINTS ARE INVOLVED Analysis & Solution: we simply free a savepoint when releasing a savepoint in which holds all rows including newly inserted rows, so we get rows lost. The solution is putting the last savepoint's rows to top savepoint. rb://3529 approved by Jimmy.Yang ------------------------------------------------------------ revno: 5505 committer: Nisha Gopalakrishnan <nisha.gopalakrishnan@oracle.com> branch nick: mysql-5.6-17024290 timestamp: Wed 2013-10-09 14:19:42 +0530 message: BUG#17024290: RENAMING A COLUMN IN ONLINE DDL REQUIRES TABLE COPY FOR BINARY DATA TYPES Info ---- Backport of the patch provided for BUG#14735373. Note: Have added additional test cases to cover the scenarios mentioned in the report of this bug. ------------------------------------------------------------ revno: 5504 [merge] committer: Praveenkumar Hulakund <praveenkumar.hulakund@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-09 13:34:35 +0530 message: Bug#17474166 - EXECUTING STATEMENT LIKE 'SHOW ENGINE INNODB' Merging from 5.5 to 5.6 ------------------------------------------------------------ revno: 2875.437.254 committer: Praveenkumar Hulakund <praveenkumar.hulakund@oracle.com> branch nick: mysql-5.5 timestamp: Wed 2013-10-09 13:32:31 +0530 message: Bug#17474166 - EXECUTING STATEMENT LIKE 'SHOW ENGINE INNODB' AND 'KILL SESSION' LEAD TO CRASH Analysis: -------- This situation occurs when the connection executes query "show engine innodb status" and this connection is killed by executing statement "kill <con>" by another connection. In function "innodb_show_status", function "stat_print" is called to print the status but return value of function is not checked. After killing connection, if write to connection fails then error is returned and same is set in Diagnostic area. Since FALSE is returned from "innodb_show_status" now, assert to check no error is set in function "set_eof_status" (called from my_eof) is failing. Fix: ---- Changed code to check return value of function "stat_print" in "innodb_show_status". ------------------------------------------------------------ revno: 5503 committer: Chaithra Gopalareddy <chaithra.gopalareddy@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-09 12:54:21 +0530 message: Bug#16894092: PERFORMANCE REGRESSION IN 5.6.6+ FOR INSERT INTO .... SELECT ... FROM Problem: The performance regression is related to writing the temporary table into MyISAM. If we configure tmp_table_size and max_heap_table_size sufficiently high (in this case setting them to 96M) in order to ensure that the temporary table can be stored in the Memory engine, the performance regression is no longer present. Regression source was found to be WL#6071. Analysis: While creating temporary table, a random scan is initialized. This initialization internally frees buffers in case of myisam. As a result writes/reads happen to disk rather than cache. And hence the performance hit when we try to create tmp table in myisam. Solution: Initialize the scan without freeing buffers which can be done by calling ha_rnd_init(1). ------------------------------------------------------------ revno: 5502 [merge] committer: Sreedhar.S <sreedhar.sreedhargadda@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-10-09 11:26:33 +0530 message: upmerge for bug 13878021 ------------------------------------------------------------ revno: 2875.437.253 committer: Sreedhar.S <sreedhar.sreedhargadda@oracle.com> branch nick: mysql-5.5 timestamp: Wed 2013-10-09 11:10:46 +0530 message: Bug 13878021 - WINDOWS PACKAGE THAT INCLUDES .PDB FILES FOR INTERMEDIATE LIBRARIES USED ------------------------------------------------------------ revno: 5501 committer: Marc Alff <marc.alff@oracle.com> branch nick: mysql-5.6-bug70552 timestamp: Tue 2013-10-08 22:48:13 +0200 message: BUG#17576516 HOST_CACHE_SIZE VALUE IS IGNORED Before this fix, starting the server with host_cache_size = N in the my.cnf file had no effect. The internal host cache size was always 128. The root cause is in hostname_cache_init(), which is using 'HOST_CACHE_SIZE' (a #define to 128) instead of 'host_cache_size' (the C global variable that corresponds to the SQL system value named host_cache_size). Overall, the initialization of the host cache is not satisfactory, and needs to be cleaned up. In particular, the fact that there are distinct: - data path (mysqld.cc setting the host_cache_size global variable, to be used by hostname.cc) - control path (mysqld.cc calling hostname_cache_init with no parameters) can only be a source of confusion. This fix: - moves the global variable host_cache_size to mysqld, where it belongs, because this is a server global variable - changes the variable type from ulong to uint, given that the size really is an unsigned int, per hash_filo::m_size - changes hostname_cache_init() to take explicitely a size, for consistency with hostname_cache_resize(), and for a cleaner interface. ------------------------------------------------------------ revno: 5500 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6 timestamp: Tue 2013-10-08 14:36:10 +0200 message: Updated dist version for 5.6.15, fixed rpm-uln advanced build error ------------------------------------------------------------ revno: 5499 committer: Shaohua Wang <shaohua.wang@oracle.com> branch nick: mysql-5.6-bug17384979 timestamp: Tue 2013-10-08 11:22:51 +0800 message: BUG 17384979 - YIELD ON SCANNER THREAD FOR PARALLEL FTS INDEX CREATE TO MINIMIZE MEMORY USAGE Solution: 1. Let scanner thread sleep for a while if fts_doc_list memory exceeds the limit; 2. Free doc_item once it's tokenized(a mutex is created to protect fts_doc_list). rb#3335 approved by Jimmy.Yang ------------------------------------------------------------ revno: 5498 committer: Venkatesh Duggirala<venkatesh.duggirala@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-10-07 21:46:54 +0530 message: Bug#16960133 MYSQLBINLOG VERBOSE OUTPUT, INVALID MEMORY ACCESS IN MY_B_WRITE_QUOTED Analysis: IN RBR mode and binlog_row_image is set to MINIMAL, only the changed columns are logged in binlog. And the meta data (eg: null bytes) will also be written only for these effected columns. `mysqlbinlog` tool in verbose mode (-v) skips these null bytes before reading the actual row. This tool assumes that for each column in the table there is one null bit which is not always true(eg: binlog_row_image=minimal) and calculates the number of total null bytes written in the binlog wrongly (ie., skips wrong number of bytes) and finally ends up in interpreting the row in wrong way. This problem could lead to invalid memory reads. The formula used to calculate the metadata bytes was ((number of columns in the table + 7) /8). For example, if we have 9 columns in a table and if there is a two columns insert/update event, master would have written two bits (i.e., one byte) in row metadata. If we calculate based on number of columns in table, it results 2 bytes instead of 1 byte. Fix: The formula should be based on the number of columns written in the event instead of the total number of columns in the table. Changed new formula is ((number of columns written in the event) + 7)/8) ------------------------------------------------------------ revno: 5497 [merge] committer: Praveenkumar Hulakund <praveenkumar.hulakund@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-10-07 18:19:21 +0530 message: Bug#11745656 - KILL THREAD -> ERROR: "SERVER SHUTDOWN IN PROGRESS" Merging from 5.5 to 5.6 ------------------------------------------------------------ revno: 2875.437.252 committer: Praveenkumar Hulakund <praveenkumar.hulakund@oracle.com> branch nick: mysql_5_5 timestamp: Mon 2013-10-07 18:17:26 +0530 message: Bug#11745656 - KILL THREAD -> ERROR: "SERVER SHUTDOWN IN PROGRESS" Fixing pb2 test failure for 5.5 version on solaris platform. ------------------------------------------------------------ revno: 5496 [merge] committer: Kent Boortz <kent.boortz@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-10-07 13:47:03 +0200 message: Merge ------------------------------------------------------------ revno: 2875.437.251 [merge] committer: Kent Boortz <kent.boortz@oracle.com> branch nick: mysql-5.5 timestamp: Mon 2013-10-07 13:05:49 +0200 message: Corrections of copyright headers (RE-request 45781) ------------------------------------------------------------ revno: 2661.876.76 [merge] committer: Kent Boortz <kent.boortz@oracle.com> branch nick: mysql-5.1 timestamp: Mon 2013-10-07 12:44:38 +0200 message: Corrections of copyright headers (RE-request 45781) ------------------------------------------------------------ revno: 5495 [merge] committer: Kent Boortz <kent.boortz@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-10-07 13:06:48 +0200 message: Corrections of copyright headers (RE-request 45781) ------------------------------------------------------------ revno: 2875.437.250 [merge] committer: Kent Boortz <kent.boortz@oracle.com> branch nick: mysql-5.5 timestamp: Mon 2013-10-07 12:45:05 +0200 message: Corrections of copyright headers (RE-request 45781) ------------------------------------------------------------ revno: 5494 [merge] committer: Kent Boortz <kent.boortz@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-10-07 12:46:18 +0200 message: Corrections of copyright headers (RE-request 45781) ------------------------------------------------------------ revno: 5490.1.1 [merge] committer: Kent Boortz <kent.boortz@oracle.com> branch nick: mysql-5.6 timestamp: Sun 2013-10-06 23:37:40 +0200 message: Corretions of copyright headers (RE-request 45781) ------------------------------------------------------------ revno: 2875.456.1 [merge] committer: Kent Boortz <kent.boortz@oracle.com> branch nick: mysql-5.5 timestamp: Sun 2013-10-06 18:16:09 +0200 message: Corretions of copyright headers (RE-request 45781) ------------------------------------------------------------ revno: 2661.884.1 committer: Kent Boortz <kent.boortz@oracle.com> branch nick: mysql-5.1 timestamp: Fri 2013-10-04 13:51:13 +0200 message: Corretions of copyright headers (RE-request 45781) ------------------------------------------------------------ revno: 5493 [merge] committer: Praveenkumar Hulakund <praveenkumar.hulakund@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-10-07 14:09:29 +0530 message: Bug#11745656 - KILL THREAD -> ERROR: "SERVER SHUTDOWN IN PROGRESS" Merging from 5.5 to 5.6. ------------------------------------------------------------ revno: 2875.437.249 committer: Praveenkumar Hulakund <praveenkumar.hulakund@oracle.com> branch nick: mysql_5_5 timestamp: Mon 2013-10-07 14:03:57 +0530 message: Bug#11745656 - KILL THREAD -> ERROR: "SERVER SHUTDOWN IN PROGRESS" Fixing pb2 test failure. ------------------------------------------------------------ revno: 5492 [merge] committer: Yasufumi Kinoshita <yasufumi.kinoshita@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-10-07 15:19:07 +0900 message: Merge mysql-5.5 to mysql-5.6. ------------------------------------------------------------ revno: 2875.437.248 [merge] committer: Yasufumi Kinoshita <yasufumi.kinoshita@oracle.com> branch nick: mysql-5.5 timestamp: Mon 2013-10-07 15:17:56 +0900 message: Merge mysql-5.1 to mysql-5.5. ------------------------------------------------------------ revno: 2661.876.75 committer: Yasufumi Kinoshita <yasufumi.kinoshita@oracle.com> branch nick: mysql-5.1 timestamp: Mon 2013-10-07 15:16:31 +0900 message: Bug#17431533 : FAILING ASSERTION: INDEX->PAGE != 0XFFFFFFFF AFTER DISCARDING TABLESPACE ha_innobase::records_in_range() should return HA_POS_ERROR for the table during discarded without requesting pages. The later other handler method should treat the error correctly. Approved by Sunny in rb#3433 ------------------------------------------------------------ revno: 5491 [merge] committer: Praveenkumar Hulakund <praveenkumar.hulakund@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-10-07 09:51:28 +0530 message: Bug#11745656 - KILL THREAD -> ERROR: "SERVER SHUTDOWN IN PROGRESS" Merging from 5.5 to 5.6 ------------------------------------------------------------ revno: 2875.437.247 committer: Praveenkumar Hulakund <praveenkumar.hulakund@oracle.com> branch nick: mysql_5_5 timestamp: Mon 2013-10-07 09:49:32 +0530 message: Bug#11745656 - KILL THREAD -> ERROR: "SERVER SHUTDOWN IN PROGRESS" Fixing pb2 test failure for 5.5 version on solaris platform. ------------------------------------------------------------ revno: 5490 [merge] committer: Praveenkumar Hulakund <praveenkumar.hulakund@oracle.com> branch nick: mysql-5.6 timestamp: Sat 2013-10-05 15:29:57 +0530 message: Bug#11745656 - KILL THREAD -> ERROR: "SERVER SHUTDOWN IN PROGRESS" Merging from 5.5 to 5.6 ------------------------------------------------------------ revno: 2875.437.246 committer: Praveenkumar Hulakund <praveenkumar.hulakund@oracle.com> branch nick: mysql_5_5 timestamp: Sat 2013-10-05 15:29:02 +0530 message: Bug#11745656 - KILL THREAD -> ERROR: "SERVER SHUTDOWN IN PROGRESS" Description: ------------ There are 2 issues reported in the bug report, 1. One session running a "long" select, then, from the other session, you kill that first one, while select is running, and it receives that message "Server shutdown in progress". Reported Date: 02-Apr-2006 => Looks like this isuse is already fixed in 2009 by the patch pushed for bug28141. 2. Killing query which goes to filesort, logs error entries like: 120416 9:17:28 [ERROR] mysqld: Sort aborted: Server shutdown in progress 120416 9:18:48 [ERROR] mysqld: Sort aborted: Server shutdown in progress 120416 9:19:39 [ERROR] mysqld: Sort aborted: Server shutdown in progress Reported Date: 16-Apr-2012 => This issue is introduced in 5.5+ versions. Fixing this issue in this patch. Analysis: --------- In function "filesort()", on error we are logging error message. To the error message, the message related THD::killed_errno is also appeneded, if it is set.(THD::kill_errno value is obtained by calling member function THD::killed_errno) In the scenario mentioned in this bug report, when we kill the connection, THD::kill_errno is set to the THD::KILL_CONNECTION. Enum type THD::KILL_CONNECTION corressponds to value ER_SERVER_SHUTDOWN. Because of this, "Server shutdown in ...." is appended to the message logged. Fix: ---- Modified code of "filesort()" function to append "KILL_QUERY" status to error message when thread is killed and server shutdown is not in progress. ------------------------------------------------------------ revno: 5489 committer: Marc Alff <marc.alff@oracle.com> branch nick: mysql-5.6-bug17310878 timestamp: Fri 2013-10-04 22:21:58 +0200 message: Bug#17310878 PFS OVERHEAD ON FREQUENT CONNECT/DISCONNECT This fix is a back port from 5.7 to 5.6 This fix is a performance improvement. The issue was that the call to PSI_THREAD_CALL(delete_current_thread)(), which is executed each time a thread disconnects, was placed inside a critical section involving LOCK_thread_count. The fix is to perform the same call sooner, before entering the LOCK_thread_count critical section. Serializing all calls to pfs_delete_current_thread_v1() was the major cause of performance overhead, for frequent connect/disconnect. ------------------------------------------------------------ revno: 5488 committer: Bjorn Munch <bjorn.munch@oracle.com> branch nick: nostl-56 timestamp: Thu 2013-10-03 21:08:13 +0200 message: Bug #17515220 CANNOT RUN MTR ON DELIVERED SOLARIS PACKAGES UNLESS HAVING SUN STUDIO INSTALLED Problem is that the my_safe_process executable is (unnecessarily) linked with libstlport but this is not in the right relative path. Fix is to remove " -library=stlport4" from flags for this binary We also remove no longer relevant -R link options ------------------------------------------------------------ revno: 5487 committer: Ole John Aske <ole.john.aske@oracle.com> branch nick: mysql-5.6 timestamp: Thu 2013-10-03 14:50:13 +0200 message: Fix for bug#16999886 ORDER BY DOESN'T WORK WITH JOIN Fix for incorrect sort order for *pushed joins* in combination with DESC'ending order by. Thus this fix will only affect *cluster* branches but are pushed to the 5.6 main branch as it affect the common optimizer code. Permission has been granted for 5.6 push by Sveta and Srini. When a DESCending order by is optimized to use an ordered index access, test_if_skip_sort_order() will set the access function pointers to refer join_read_last_key & join_read_prev_same. These will implement reversed read through the ordered index. However, the function pick_table_access_method() later overwrote these access function pointers with similar 'forward read' functions. This fix modifies pick_table_access_methods() such that these functions pointers are not set of already assigned a non-NULL value even for pushed joins. NOTE: This check already existed, but due to an incorrect 'else', didn't have effect for the root of a pushed join. ------------------------------------------------------------ revno: 5486 committer: Annamalai Gurusami <annamalai.gurusami@oracle.com> branch nick: mysql-5.6 timestamp: Thu 2013-10-03 15:17:17 +0530 message: Bug #17253499 INNODB STATS INTERFERES WITH INNODB FORCE RECOVERY AND DROP/CREATE TABLES Problem: During force recovery (SRV_FORCE_NO_TRX_UNDO), create and drop table is not working. Solution: The semantics of the force recovery level SRV_FORCE_NO_TRX_UNDO has been misinterpreted. The force recovery level SRV_FORCE_NO_TRX_UNDO means that rollback of recovered transactions must not be done. This should not prevent other transactions being assigned a rollback segment. So while assigning rollback segments, do not check the level of innodb_force_recovery. Also, when the innodb_force_recovery is greater than SRV_FORCE_NO_TRX_UNDO we now make InnoDB to become read only. rb#3473 approved by Marko ------------------------------------------------------------ revno: 5485 [merge] committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6 timestamp: Tue 2013-10-01 15:24:23 +0200 message: Refresh mysql-install patch and mysqld service naming for repo project ------------------------------------------------------------ revno: 5406.1.12 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6.14-Fedora timestamp: Fri 2013-09-27 18:53:22 +0200 message: Renamed patch file names ------------------------------------------------------------ revno: 5406.1.11 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6.14-Fedora timestamp: Fri 2013-09-27 17:58:55 +0200 message: Refresh mysql-install patch and service renaming ------------------------------------------------------------ revno: 5484 [merge] committer: Marko M?kel? <marko.makela@oracle.com> branch nick: mysql-5.6 timestamp: Tue 2013-10-01 14:24:36 +0300 message: Merge mysql-5.5 to mysql-5.6. ------------------------------------------------------------ revno: 2875.437.245 committer: Marko M?kel? <marko.makela@oracle.com> branch nick: mysql-5.5 timestamp: Tue 2013-10-01 14:17:35 +0300 message: Followup to Bug#17446090 fix: Now that ibuf_change_buffering_debug a.k.a. ibuf_debug should work reliably (no "surprise" merges), make some tests stricter (require that the server crashes where we expect it to). ------------------------------------------------------------ revno: 5483 [merge] committer: Mattias Jonsson <mattias.jonsson@oracle.com> branch nick: topush-5.6 timestamp: Tue 2013-10-01 11:23:10 +0200 message: manual merge of bug#14621190 into mysql-5.6 ------------------------------------------------------------ revno: 2875.437.244 committer: Mattias Jonsson <mattias.jonsson@oracle.com> branch nick: b14621190-55 timestamp: Tue 2013-10-01 10:05:45 +0200 message: Bug#14621190: HA_INNOBASE::INDEX_NEXT SKIPS A RECORD IF PREVIOUS INDEX_READ_MAP HAD NO MATCH If index_read_map is called for exact search and no matching records exists it will position the cursor on the next record, but still having the relative position to BTR_PCUR_ON. This will make a call for index_next to read yet another next record, instead of returning the record the cursor points to. Fixed by setting pcur->rel_pos = BTR_PCUR_BEFORE if an exact [prefix] search is done, but failed. Also avoids optimistic restoration if rel_pos != BTR_PCUR_ON, since btr_cur may be different than old_rec. rb#3324, approved by Marko and Jimmy ------------------------------------------------------------ revno: 5482 committer: Sreedhar.S <sreedhar.sreedhargadda@oracle.com> branch nick: mysql-5.6 timestamp: Tue 2013-10-01 10:54:52 +0530 message: Fix to add mysys_ssl missing pdb file ------------------------------------------------------------ revno: 5481 committer: Sreedhar.S <sreedhar.sreedhargadda@oracle.com> branch nick: mysql-5.6 timestamp: Tue 2013-10-01 10:47:18 +0530 message: Fix to check for file presence before install ------------------------------------------------------------ revno: 5480 [merge] committer: Sreedhar.S <sreedhar.sreedhargadda@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-09-30 19:05:39 +0530 message: Upmerge for the Bugs: 16619754, 16053094 & 13878021 ------------------------------------------------------------ revno: 2875.437.243 committer: Sreedhar.S <sreedhar.sreedhargadda@oracle.com> branch nick: mysql-5.5 timestamp: Mon 2013-09-30 13:01:17 +0530 message: Bug #16619754 - MYSQL_INSTALL_DB.PL FAILS TO CREATE DATABASE ON WINDOWS. As this mysql_install_db.pl file has always generated lots of confusion on Windows. This fix will make sure to get it removed only from Windows ------------------------------------------------------------ revno: 2875.437.242 committer: Sreedhar.S <sreedhar.sreedhargadda@oracle.com> branch nick: mysql-5.5 timestamp: Mon 2013-09-30 12:41:22 +0530 message: Fix for Bug:16053094 Registry redirection was improper and hence not able to pick the value for the OLDERVERSION property. as a result in the install upgrade dialog null value wa being passed. ------------------------------------------------------------ revno: 2875.437.241 committer: Sreedhar.S <sreedhar.sreedhargadda@oracle.com> branch nick: mysql-5.5 timestamp: Mon 2013-09-30 12:29:10 +0530 message: Fix to copy missing pdb files BUG 13878021 ------------------------------------------------------------ revno: 5479 committer: Jorgen Loland <jorgen.loland@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-09-30 12:27:07 +0200 message: BUG#17458273: KEY LENGTHS LEAD TO INCOMPLETE RESULT SETS; UNHINTED SCANS PREFER SUCH INDEXES Consider a table with an index over a prefix of column 'x' and a query with a range condition over 'x': CREATE TABLE (... INDEX (x(4))) SELECT ... WHERE 'foobar2' < x and x < 'foobar5'; The index will not be able to distinguish between rows with x-values like "foobar", "foobaz" or "foobar3" since only the first four characters ("foob") will be stored in the index. Since all rows with an x-value prefix "foob" look equal, the range optimizer has to translate '<' and '>' to '<=' and '>=', respectively, for such indexes. In this bug, only "'foobar2' < x" is correctly translated. Later, when key_and() is executed on the the two ranges "'foob' <= x" and "NULL < x < 'foob'", key_and() correctly finds that these ranges together form an impossible condition. The fix is to also translate '<' to '<=' for prefix indexes so that key_and() can ultimately create the range "'foob' <= x <= 'foob'". ------------------------------------------------------------ revno: 5478 [merge] committer: Yasufumi Kinoshita <yasufumi.kinoshita@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-09-30 15:04:51 +0900 message: merge from mysql-5.5 to mysql-5.6 ------------------------------------------------------------ revno: 2875.437.240 committer: Yasufumi Kinoshita <yasufumi.kinoshita@oracle.com> branch nick: mysql-5.5 timestamp: Mon 2013-09-30 15:02:54 +0900 message: Adjustment for fix for Bug#11758196 log_buffer_extend() should fill the new buffer with 0. ------------------------------------------------------------ revno: 5477 [merge] committer: Yasufumi Kinoshita <yasufumi.kinoshita@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-09-30 13:43:28 +0900 message: Merge fix for Bug#11758196 from mysql-5.5 to mysql-5.6 ------------------------------------------------------------ revno: 2875.437.239 committer: Yasufumi Kinoshita <yasufumi.kinoshita@oracle.com> branch nick: mysql-5.5 timestamp: Mon 2013-09-30 13:41:48 +0900 message: Bug#11758196 : INNODB ASSERTION FAILURE WHEN CONVERTING FROM MYISAM TO INNODB Changed to try to extend log buffer instead of crash, when log size is too large for the size. Approved by Marko in rb#3229 ------------------------------------------------------------ revno: 5476 [merge] committer: Satya Bodapati <satya.bodapati@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-09-27 18:34:37 +0530 message: Merge fix for BUG#17446090 from mysql-5.5 to mysql-5.6 The testcase for this bug fails randomly due to two reasons. 1. Due to ibuf merge happening background 2. Due to dict stats update which brings the evicted page back into buffer pool. Fix ibuf_merge() to not do any merges with ibuf_debug enabled and also changed dict_stats_update() to return fake statistics without brining the secondary index pages into buffer pool. Approved by Marko. rb#3420 ------------------------------------------------------------ revno: 2875.437.238 [merge] committer: Satya Bodapati <satya.bodapati@oracle.com> branch nick: mysql-5.5 timestamp: Fri 2013-09-27 18:32:13 +0530 message: Merge fix for BUG#17446090 from mysql-5.1 to mysql-5.5 The testcase for this bug fails randomly due to two reasons. 1. Due to ibuf merge happening background 2. Due to dict stats update which brings the evicted page back into buffer pool. Fix ibuf_contract_ext() to not do any merges with ibuf_debug enabled and also changed dict_stats_update() to return fake statistics without bringing the secondary index pages into buffer pool. Approved by Marko. rb#3419 ------------------------------------------------------------ revno: 2661.876.74 committer: Satya Bodapati <satya.bodapati@oracle.com> branch nick: mysql-5.1 timestamp: Fri 2013-09-27 18:30:46 +0530 message: BUG#17446090 - TESTCASE INNODB_BUG16752251.TEST FAILS RANDOMLY ON PB2 Enable disabled testcases. ------------------------------------------------------------ revno: 5475 committer: Shivji Kumar Jha <shivji.jha@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-09-27 16:59:11 +0530 message: BUG#16580366- MTR TESTS FAILING SPORADICALLY ON PB2 (5.5, 5.6 AND 5.7) DURING INNODB RECOVERY post push fix ------------------------------------------------------------ revno: 5474 committer: Anitha Gopi <anitha.gopi@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-09-27 08:51:30 +0200 message: default.weekly.basic do not run --big-tests and hence do not need longer timeout ------------------------------------------------------------ revno: 5473 committer: Anitha Gopi <anitha.gopi@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-09-27 06:06:06 +0200 message: Added default.weekly.basic. This defines reduced suite for tier3/3 platforms Following changes to default.weekly: * Improved comments * Fixed typos * Restricted binlog-checksum=CRC32 runs to rpl and binlog suites * Removed run with ps-protocol for rpl and binlog suites. This is already covered by the ps-protocol run of default suites ------------------------------------------------------------ revno: 5472 [merge] committer: Shivji Kumar Jha <shivji.jha@oracle.com> branch nick: mysql-5.6 timestamp: Fri 2013-09-27 01:29:56 +0530 message: BUG#16580366- MTR TESTS FAILING SPORADICALLY ON PB2 (5.5, 5.6 AND 5.7) DURING INNODB RECOVERY bzr merge mysql-5.5->5.6 ------------------------------------------------------------ revno: 2875.437.237 committer: Shivji Kumar Jha <shivji.jha@oracle.com> branch nick: mysql-5.5 timestamp: Fri 2013-09-27 01:24:16 +0530 message: BUG#16580366- MTR TESTS FAILING SPORADICALLY ON PB2 (5.5, 5.6 AND 5.7) DURING INNODB RECOVERY Problem: ======= The connection 'master' is dropped by mysqltest after rpl_end.inc. At this point, dropping temporary tables at the connection 'master' are not synced at slave. So, the temporary tables replicated from master remain on slave leading to an inconsistent close of the test. The following test thus complains about the presence of temporary table(s) left over from the previous test. Fix: === - Put explicit drop commands in replication tests so that the temporary tables are dropped at slave as well. - Added the check for Slave_open_temp_tables in mtr_check.sql to warn about the remaining temporary table, if any, at the close of a test. ------------------------------------------------------------ revno: 5471 committer: Allen lai <zheng.lai@oracle.com> branch nick: mysql-5.6 timestamp: Wed 2013-09-25 10:31:20 +0800 message: Bug#17358875 MEMCACHED: THE BINLOG SEQUENCE IS NOT CORRECT Approved by Jimmy on rb://3213. This bug is because we commit innodb transaction before write binlog. This will cause other thread write its binlog before current thread. The fix patch is moving write binlog before commit transaction. It'll guarantee binlog is written according to transaction commit order. This patch also contains a follow up patch for Bug#16498810 WRONG FILENAME IN ERROR MESSAGE FOR INNODB MEMCACHED CONFIG. ------------------------------------------------------------ revno: 5470 committer: Yasufumi Kinoshita <yasufumi.kinoshita@oracle.com> branch nick: mysql-5.6 timestamp: Tue 2013-09-24 11:52:14 +0900 message: Bug#16870041 I_INNODB.INNODB-ALTER-DEBUG FAILED REGULARLY IN DIFFERENT RUNS WITH TIMEOUT mtr script adjustment and debug build adjustment backporting from mysql-trunk ------------------------------------------------------------ revno: 5469 committer: Aditya A <aditya.a@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-09-23 21:44:36 +0530 message: Bug #17441867 [ERROR] INNODB: FAILED TO SET O_DIRECT ON FILE PROBLEM ------- Setting the O_DIRECT flag on file on tmpfs fails and error messages are printed in error logs. The customer was creating lot of temporary tables in tmpfs and because of which error log was flooded with these error messages. We however ignore these errors and continue to create the tables.These error messages were causing problem with the their monitoring system. FIX --- The fix is to downgrade the error with a warning and printing this warning only once in the error log. ------------------------------------------------------------ revno: 5468 [merge] committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-09-23 13:46:57 +0200 message: Merge from 5.6.14-Fedora for repo project ------------------------------------------------------------ revno: 5406.1.10 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6.14-Fedora timestamp: Mon 2013-09-16 11:47:00 +0200 message: Added fedora and oel to CMakeLists ------------------------------------------------------------ revno: 5406.1.9 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6.14-Fedora timestamp: Mon 2013-09-16 10:51:19 +0200 message: Removed uncessary bak files ------------------------------------------------------------ revno: 5406.1.8 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6.14-Fedora timestamp: Mon 2013-09-16 09:20:37 +0200 message: Build 5.6.14 Fedora/OEL rpms, updated dist version ------------------------------------------------------------ revno: 5406.1.7 author: balasubramanian.kandasamy@oracle.com committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6.14-Fedora timestamp: Fri 2013-09-13 21:50:42 +0200 message: Build 5.6.14 repo rpms ------------------------------------------------------------ revno: 5467 [merge] committer: Sujatha Sivakumar <sujatha.sivakumar@oracle.com> branch nick: Bug17327454_mysql-5.6 timestamp: Mon 2013-09-23 14:02:32 +0530 message: Merge from mysql-5.5 to mysql-5.6 ------------------------------------------------------------ revno: 2875.437.236 committer: Sujatha Sivakumar <sujatha.sivakumar@oracle.com> branch nick: Bug17327454_mysql-5.5 timestamp: Mon 2013-09-23 12:13:37 +0530 message: Bug#17327454:SEMI-SYNC REPLICATION MASTER CRASH WHEN SET RPL_SEMI_SYNC_MASTER_ENABLED OFF. Problem: ======= If master is waiting for a reply from slave, at this time set global rpl_semi_sync_master_enabled=OFF, the master server will crash. Analysis: ======== When master is waiting for a reply from slave, at this time if semi sync is switched off on master, during switch off if active transactions are present the transactions will be cleared and "active_tranxs_" variable will be set to NULL. When the waiting master connection finds that semi sync is switched of it tries to access "active_tranxs_" without checking if the transaction list exists or not. Accessing NULL transaction list causes the crash. Fix: === A check has been added to see a valid list exists before accessing the "active_tranxs_". ------------------------------------------------------------ revno: 5466 committer: viswanatham gudipati <viswanatham.gudipati@oracle.com> branch nick: mysql-5.6 timestamp: Mon 2013-09-23 12:04:52 +0530 message: Bug#16387578: The patch is approved http://rb.no.oracle.com/rb/r/3341/ ------------------------------------------------------------ revno: 5465 [merge] author: sunanda.menon@oracle.com committer: Sunanda Menon<sunanda.menon@oracle.com> branch nick: mysql-5.6 timestamp: Sat 2013-09-21 04:32:44 +0200 message: Merge from mysql-5.6.14-release ------------------------------------------------------------ revno: 5406.1.6 tags: mysql-5.6.14 committer: Balasubramanian Kandasamy <balasubramanian.kandasamy@oracle.com> branch nick: mysql-5.6.14-release timestamp: Tue 2013-09-10 09:25:10 +0200 message: Reverted the changes to spec file, updated the logic to get the correct count of PID files |