upgrading lock, even with low_priority_updates
The problem is that there is no mechanism to control whether a
delayed insert takes a high or low priority lock on a table.
The solution is to modify the delayed insert thread ("handler")
to take into account the global value of low_priority_updates
when taking table locks. The value of low_priority_updates is
retrieved when the insert delayed thread is created and will
remain the same for the duration of the thread.
If delayed insert fails to upgrade the lock it was not
freeing the temporary memory storage used to keep
newly constructed blob values in memory.
Fixed by iterating over the remaining rows in the delayed
insert rowset and freeing the blob storage for each row.
No test suite because it involves concurrent delayed inserts
on a table and cannot easily be made deterministic.
Added a correct valgrind suppression for Fedora 9.
The failure was caused by executing a CREATE-SELECT statement that creates a
table in another database than the current one. In row-based logging, the
CREATE statement was written to the binary log without the database, hence
creating the table in the wrong database, causing the following inserts to
fail since the table didn't exist in the given database.
Fixed the bug by adding a parameter to store_create_info() that will make
the function print the database name before the table name and used that
in the calls that write the CREATE statement to the binary log. The database
name is only printed if it is different than the currently selected database.
The output of SHOW CREATE TABLE has not changed and is still printed without
the database name.
Concurrent inserts produce valgrind error messages.
The reason is that the query cache is invalidated after the target table object
is closed.
Since the delayed insert thread already takes care of invalidating the query
cache there is no need to try to synchronize an extra cache invalidation call.
The fix is to remove the query_cache_invalidate3 call altogether.
If a delayed insert thread was aborted by a concurrent 'truncate table'
statement, then the diagnostic area would fail with an assert in a debug build
because no actual error message was pushed on the stack despite a thread
being killed.
This patch adds an error message to the stack.
in open_table()
Problem: repeating "CREATE... ( AUTOINCREMENT) ... SELECT" may lead to
an assertion failure.
Fix: reset table->auto_increment_field_not_null after each record
writing.
The assert is about binlogging must have been activated, but it was
not actually according to the reported how-to-repeat instuctions.
Analysis revealed that binlog_start_trans_and_stmt() was called
without prior testing if binlogging is ON.
Fixed with avoing entering binlog_start_trans_and_stmt() if binlog is
not activated.
In order to handle CHAR() fields, 8 bits were reserved for
the size of the CHAR field. However, instead of denoting the
number of characters in the field, field_length was used which
denotes the number of bytes in the field.
Since UTF-8 fields can have three bytes per character (and
has been extended to have four bytes per character in 6.0),
an extra two bits have been encoded in the field metadata
work for fields of type Field_string (i.e., CHAR fields).
Since the metadata word is filled, the extra bits have been
encoded in the upper 4 bits of the real type (the most
significant byte of the metadata word) by computing the
bitwise xor of the extra two bits. Since the upper 4 bits
of the real type always is 1111 for Field_string, this
means that for fields of length <256, the encoding is
identical to the encoding used in pre-5.1.26 servers, but
for lengths of 256 or more, an unrecognized type is formed,
causing an old slave (that does not handle lengths of 256
or more) to stop.
Problem was an unclear error message since it could suggest that
MyISAM did not support INSERT DELAYED.
Changed the error message to say that DELAYED is not supported by the
table, instead of the table's storage engine.
The confusion is that a partitioned table is in somewhat sense using
the partitioning storage engine, which in turn uses the ordinary
storage engine. By saying that the table does not support DELAYED we
do not give any extra informantion about the storage engine or if it
is partitioned.
Among two claimed artifacts the critical one is in that the Table map of
a query following the failing with a duplicate key error CREATE-SELECT is skipped from
instantionating (and thus binlogging). That leads to sending a "chopped" group of the data
row-events without the table map head to the slave.
The slave can not apply the only data row events.
It's not easy to force the slave to react with an error in such a case (the second complaint
on the bug report), because the lack of a table Rows_log_event::do_apply_event the data row event
handler is a common situation which normally designates the event has to be filtered out
basing on the repliation do/ingore rules decision.
Fixed: table map creating and binlogging is restored via deploying the standard cleanup call in
select_create::abort().
No error is reported if by chance the table map was not been binlogged.
Leaving this out to resolve with considering how to combine the do/ingore rules with the situation
when erronoulsy the Table_map is not written to binlog.
The bug allow multiple executing transactions working with non-transactional
to interfere with each others by interleaving the events of different trans-
actions.
Bug is fixed by writing non-transactional events to the transaction cache and
flushing the cache to the binary log at statement commit. To mimic the behavior
of normal statement-based replication, we flush the transaction cache in row-
based mode when there is no committed statements in the transaction cache,
which means we are committing the first one. This means that it will be written
to the binary log as a "mini-transaction" with just the rows for the statement.
Note that the changes here does not take effect when building the server with
HAVE_TRANSACTIONS set to false, but it is not clear if this was possible before
this patch either.
For row-based logging, we also have that when AUTOCOMMIT=1, the code now always
generates a BEGIN/COMMIT pair for single statements, or BEGIN/ROLLBACK pair in the
case of non-transactional changes in a statement that was rolled back. Note that
for the case where changes to a non-transactional table causes a rollback due
to error, the statement will now be logged with a BEGIN/ROLLBACK pair, even
though some changes has been committed to the non-transactional table.
Bug#21413
"Engine table handler used by multiple threads in REPLACE DELAYED"
When executing a REPLACE DELAYED statement, the storage engine
::extra() method was invoked by a different thread than the thread
which has acquired the handler instance.
This did not cause problems within the current server and with
the current storage engines.
But it has the potential to confuse future storage engines.
Added code to avoid surplus calls to extra() method in case of DELAYED
which avoids calling storage engine from a different thread than
expected.
No test case.
This change does not change behavior in conjunction with current
storage engines. So it cannot be tested by the regression test suite.
binlog_format=mixed
Statement-based replication of DELETE ... LIMIT, UPDATE ... LIMIT,
INSERT ... SELECT ... LIMIT is not safe as order of rows is not
defined.
With this fix, we issue a warning that this statement is not safe to
replicate in statement mode, or go to row-based mode in mixed mode.
Note that we may consider a statement as safe if ORDER BY primary_key
is present. However it may confuse users to see very similiar statements
replicated differently.
Note 2: regular UPDATE statement (w/o LIMIT) is unsafe as well, but
this patch doesn't address this issue. See comment from Kristian
posted 18 Mar 10:55.
sporadically
Under some circumstances, the mysql_insert_id() value after SELECT ...
INSERT could return a wrong value. This could happen when the last
SELECT ... INSERT did not involve an AUTO_INCREMENT column, but the
value of mysql_insert_id() was changed by some previous statements.
Fixed by checking the value of thd->insert_id_used in
select_insert::send_eof() and returning 0 for mysql_insert_id() if it
is not set.
a SELECT doesn't cause ROLLBACK of statem".
The idea of the fix is to ensure that we always commit the current
statement at the end of dispatch_command(). In order to not issue
redundant disc syncs, an optimization of the two-phase commit
protocol is implemented to bypass the two phase commit if
the transaction is read-only.
value when inserting into a view.
The mysql_prepare_insert function checks all fields of the target table that
directly or indirectly (through a view) are specified in the INSERT
statement to have a default value. This check can be skipped if the INSERT
statement doesn't mention any insert fields. In case of a view this allows
fields that aren't mentioned in the view to bypass the check.
Now fields of the target table are always checked to have a default value
when insert goes into a view.
cause ROLLBACK of statement", part 1. Review fixes.
Do not send OK/EOF packets to the client until we reached the end of
the current statement.
This is a consolidation, to keep the functionality that is shared by all
SQL statements in one place in the server.
Currently this functionality includes:
- close_thread_tables()
- log_slow_statement().
After this patch and the subsequent patch for Bug#12713, it shall also include:
- ha_autocommit_or_rollback()
- net_end_statement()
- query_cache_end_of_result().
In future it may also include:
- mysql_reset_thd_for_next_command().
insert ... select.
The 5.0 manual page for mysql_insert_id() does not mention anything
about INSERT ... SELECT, though its current behavior is incosistent
with what the manual says about the plain INSERT.
Fixed by changing the AUTO_INCREMENT and mysql_insert_id() handling
logic in INSERT ... SELECT to be consistent with the INSERT behavior,
the manual, and the changes in 5.1 introduced by WL3146:
- mysql_insert_id() now returns the first automatically generated
AUTO_INCREMENT value that was successfully inserted by INSERT ... SELECT
- if an INSERT ... SELECT statement is executed, and no automatically
generated value is successfully inserted, mysql_insert_id() now returns
the ID of the last inserted row.
