and clarifies the invariant in dict_table_get_on_id().
In Mar 2007 Marko observed a crash during recovery, the crash resulted from
an UNDO operation on a system table. His solution was to acquire an X lock on
the data dictionary, this in hindsight was an overkill. It is unclear what
caused the crash, current hypothesis is that it was a memory corruption.
The X lock results in performance issues by when undoing changes due to
rollback during normal operation on regular tables.
Why the change is safe:
======================
The InnoDB code has changed since the original X lock change was made. In the
new code we always lock the data dictionary in X mode during startup when
UNDOing operations on the system tables (this is a given). This ensures that
the crash Marko observed cannot happen as long as all transactions that update
the system tables follow the standard rules by setting the appropriate DICT_OP
flag when writing the log records when they make the changes.
If transactions violate the above mentioned rule then during recovery (at
startup) the rollback code (see trx0roll.c) will not acquire the X lock
and we will see the crash again. This will however be a different bug.
During creation of the table list of
processed tables hidden I_S table 'VARIABLES'
is erroneously added into the table list.
it leads to ER_UNKNOWN_TABLE error in
TABLE_LIST::add_table_to_list() function.
The fix is to skip addition of hidden I_S
tables into the table list.
DML flow and SAVEPOINT
The problem was that replication could break if a transaction involving
both transactional and non-transactional tables was rolled back to a
savepoint. It broke if a concurrent connection tried to drop a
transactional table which was locked after the savepoint was set.
This DROP TABLE completed when ROLLBACK TO SAVEPOINT was executed as the
lock on the table was dropped by the transaction. When the slave later
tried to apply the binlog, it would fail as the table would already
have been dropped.
The reason for the problem is that transactions involving both
transactional and non-transactional tables are written fully to the
binlog during ROLLBACK TO SAVEPOINT. At the same time, metadata locks
acquired after a savepoint, were released during ROLLBACK TO SAVEPOINT.
This allowed a second connection to drop a table only used between
SAVEPOINT and ROLLBACK TO SAVEPOINT. Which caused the transaction binlog
to refer to a non-existing table when it was written during ROLLBACK
TO SAVEPOINT.
This patch fixes the problem by not releasing metadata locks when
ROLLBACK TO SAVEPOINT is executed if binlogging is enabled.
and clarifies the invariant in dict_table_get_on_id().
In Mar 2007 Marko observed a crash during recovery, the crash resulted from
an UNDO operation on a system table. His solution was to acquire an X lock on
the data dictionary, this in hindsight was an overkill. It is unclear what
caused the crash, current hypothesis is that it was a memory corruption.
The X lock results in performance issues by when undoing changes due to
rollback during normal operation on regular tables.
Why the change is safe:
======================
The InnoDB code has changed since the original X lock change was made. In the
new code we always lock the data dictionary in X mode during startup when
UNDOing operations on the system tables (this is a given). This ensures that
the crash Marko observed cannot happen as long as all transactions that update
the system tables follow the standard rules by setting the appropriate DICT_OP
flag when writing the log records when they make the changes.
If transactions violate the above mentioned rule then during recovery (at
startup) the rollback code (see trx0roll.c) will not acquire the X lock
and we will see the crash again. This will however be a different bug.
Problem: SQL and IO thread were racing for the IO_CACHE. The former to
flush it, the latter to close it. In some cases this would cause the
SQL thread to lock an invalid IO_CACHE mutex (it had been destroyed by
IO thread). This would happen when SQL thread was initializing the
master.info
Solution: We solve this by locking the log and checking if it is
hot. If it is we keep the log while seeking. Otherwise we release it
right away, because a log can get from hot to cold, but not from cold
to hot.
require O(#scans) memory
When an index merge operation was restarted, it would
re-allocate the Unique object controlling the duplicate row
ID elimination. Fixed by making the Unique object a member
of QUICK_INDEX_MERGE_SELECT and thus reusing it throughout
the lifetime of this object.
------------------------------------------------------------
revno: 3523
revision-id: marko.makela@oracle.com-20100624104620-pklunowaigv7quu9
parent: jimmy.yang@oracle.com-20100624021010-oh2hnp8e1xbaax6u
committer: Marko Mäkelä <marko.makela@oracle.com>
branch nick: 5.1-innodb
timestamp: Thu 2010-06-24 13:46:20 +0300
message:
Bug#54679: alter table causes compressed row_format to revert to compact
ha_innobase::create(): Add the local variable row_type = form->s->row_type.
Adjust it to ROW_TYPE_COMPRESSED when ROW_FORMAT is not specified or inherited
but KEY_BLOCK_SIZE is. Observe the inherited ROW_FORMAT even when it is not
explicitly specified.
innodb_bug54679.test: New test, to test the bug and to ensure that there are
no regressions. (The only difference in the test result without the patch
applied is that the first ALTER TABLE changes ROW_FORMAT to Compact.)
ha_innobase::create(): Add the local variable row_type = form->s->row_type.
Adjust it to ROW_TYPE_COMPRESSED when ROW_FORMAT is not specified or inherited
but KEY_BLOCK_SIZE is. Observe the inherited ROW_FORMAT even when it is not
explicitly specified.
innodb_bug54679.test: New test, to test the bug and to ensure that there are
no regressions. (The only difference in the test result without the patch
applied is that the first ALTER TABLE changes ROW_FORMAT to Compact.)
file .\filesort.cc, line 149 (part II)
Problem: the server didn't disregard sort order
for some zero length tuples.
Fix: skip sort order in such a case
(zero length NOT NULL string functions).
Accidental change in compile-time definitions for FreeBSD
Revert the accidental setting of "HAVE_BROKEN_REALPATH"
on current versions of FreeBSD,
do it for both autotools ("configure.in")
and cmake ("cmake/os/FreeBSD.cmake").
cmake/build_configurations/mysql_release.cmake
- Corrected spelling ENABLE_LOCAL_INFILE => ENABLED_LOCAL_INFILE
- In addition to "RelWithDebInfo", set target "Release" and "Debug"
- Set Debug flags
- Enabled SSL on Mac OS X
- For gcc builds, set RELEASE and DEBUG flags as well
- For g++ builds, added "-fno-implicit-templates"
- Use "-O" (gcc -O1) for optimized binaries, as "DEBUG" in out case
is more about enabling trace support to the server, no optimization
makes binaries too slow to be practical to reproduce problems
cmake/os/WindowsCache.cmake
- Removed unused HAVE_SYS_IOCTL
config.h.cmake
- Added header checks and missing defines
- Removed unused HAVE_SYS_IOCTL
- Grouped and uncommented some HAVE_* that are really not
defines, but internal variables used in the CMake setup,
- Added hard coded flags for HP-UX and Mac OS X
configure.cmake
- Added header checks and missing defines
- Removed unused HAVE_SYS_IOCTL
- "sys/dir.h" test needs "sys/types.h"
- Corrected syntax for "sys/ptem.h" test
- Don't exclude test for some types if Mac OS X, harmless
to do the test and we want the HAVE_<type> settings
- Added hard coded flags for HP-UX and Mac OS X
extra/yassl/CMakeLists.txt
extra/yassl/taocrypt/CMakeLists.txt
- Added missing source file "template_instnt.cpp"
------------------------------------------------------------
revno: 3520
committer: Jimmy Yang <jimmy.yang@oracle.com>
branch nick: mysql-5.1-innodb
timestamp: Tue 2010-06-22 19:04:31 -0700
message:
Fix bug #54044, Create temporary tables and using innodb crashes. Screen
out NULL type columns, and return without creating the table.
rb://378 approved by Marko
------------------------------------------------------------
