GROUPING BY FUNCTIONS.... (PART
The bug was introduced in a patch for bug 49897.
Problem: The assertion inserted by the original patch to guard against
zero-lenght sort keys during merge phase triggers also when the whole
set fits in memory.
Fix: Move assert so that it does not trigger if the whole set is in
memory.
mysql-test/r/group_by.result:
Add test for bug#11765254
mysql-test/t/group_by.test:
Add test for bug#11765254
sql/filesort.cc:
Move assertion
Converting the number zero to binary and back yielded the number zero,
but with no digits, i.e. zero precision.
This made the multiply algorithm go haywire in various ways.
include/decimal.h:
Document struct st_decimal_t
mysql-test/r/type_newdecimal.result:
New test case (valgrind warnings)
mysql-test/t/type_newdecimal.test:
New test case (valgrind warnings)
sql/my_decimal.h:
Remove the HAVE_purify enabled/disabled code.
strings/decimal.c:
Make a proper zero, with non-zero precision.
The fix of Bug#12612184 broke crash recovery. When a record that
contains off-page columns (BLOBs) is updated, we must first write redo
log about the BLOB page writes, and only after that write the redo log
about the B-tree changes. The buggy fix would log the B-tree changes
first, meaning that after recovery, we could end up having a record
that contains a null BLOB pointer.
Because we will be redo logging the writes off the off-page columns
before the B-tree changes, we must make sure that the pages chosen for
the off-page columns are free both before and after the B-tree
changes. In this way, the worst thing that can happen in crash
recovery is that the BLOBs are written to free pages, but the B-tree
changes are not applied. The BLOB pages would correctly remain free in
this case. To achieve this, we must allocate the BLOB pages in the
mini-transaction of the B-tree operation. A further quirk is that BLOB
pages are allocated from the same file segment as leaf pages. Because
of this, we must temporarily "hide" any leaf pages that were freed
during the B-tree operation by "fake allocating" them prior to writing
the BLOBs, and freeing them again before the mtr_commit() of the
B-tree operation, in btr_mark_freed_leaves().
btr_cur_mtr_commit_and_start(): Remove this faulty function that was
introduced in the Bug#12612184 fix. The problem that this function was
trying to address was that when we did mtr_commit() the BLOB writes
before the mtr_commit() of the update, the new BLOB pages could have
overwritten clustered index B-tree leaf pages that were freed during
the update. If recovery applied the redo log of the BLOB writes but
did not see the log of the record update, the index tree would be
corrupted. The correct solution is to make the freed clustered index
pages unavailable to the BLOB allocation. This function is also a
likely culprit of InnoDB hangs that were observed when testing the
Bug#12612184 fix.
btr_mark_freed_leaves(): Mark all freed clustered index leaf pages of
a mini-transaction allocated (nonfree=TRUE) before storing the BLOBs,
or freed (nonfree=FALSE) before committing the mini-transaction.
btr_freed_leaves_validate(): A debug function for checking that all
clustered index leaf pages that have been marked free in the
mini-transaction are consistent (have not been zeroed out).
btr_page_alloc_low(): Refactored from btr_page_alloc(). Return the
number of the allocated page, or FIL_NULL if out of space. Add the
parameter "mtr_t* init_mtr" for specifying the mini-transaction where
the page should be initialized, or if this is a "fake allocation"
(init_mtr=NULL) by btr_mark_freed_leaves(nonfree=TRUE).
btr_page_alloc(): Add the parameter init_mtr, allowing the page to be
initialized and X-latched in a different mini-transaction than the one
that is used for the allocation. Invoke btr_page_alloc_low(). If a
clustered index leaf page was previously freed in mtr, remove it from
the memo of previously freed pages.
btr_page_free(): Assert that the page is a B-tree page and it has been
X-latched by the mini-transaction. If the freed page was a leaf page
of a clustered index, link it by a MTR_MEMO_FREE_CLUST_LEAF marker to
the mini-transaction.
btr_store_big_rec_extern_fields_func(): Add the parameter alloc_mtr,
which is NULL (old behaviour in inserts) and the same as local_mtr in
updates. If alloc_mtr!=NULL, the BLOB pages will be allocated from it
instead of the mini-transaction that is used for writing the BLOBs.
fsp_alloc_from_free_frag(): Refactored from
fsp_alloc_free_page(). Allocate the specified page from a partially
free extent.
fseg_alloc_free_page_low(), fseg_alloc_free_page_general(): Add the
parameter "mtr_t* init_mtr" for specifying the mini-transaction where
the page should be initialized, or NULL if this is a "fake allocation"
that prevents the reuse of a previously freed B-tree page for BLOB
storage. If init_mtr==NULL, try harder to reallocate the specified page
and assert that it succeeded.
fsp_alloc_free_page(): Add the parameter "mtr_t* init_mtr" for
specifying the mini-transaction where the page should be initialized.
Do not allow init_mtr == NULL, because this function is never to be
used for "fake allocations".
mtr_t: Add the operation MTR_MEMO_FREE_CLUST_LEAF and the flag
mtr->freed_clust_leaf for quickly determining if any
MTR_MEMO_FREE_CLUST_LEAF operations have been posted.
row_ins_index_entry_low(): When columns are being made off-page in
insert-by-update, invoke btr_mark_freed_leaves(nonfree=TRUE) and pass
the mini-transaction as the alloc_mtr to
btr_store_big_rec_extern_fields(). Finally, invoke
btr_mark_freed_leaves(nonfree=FALSE) to avoid leaking pages.
row_build(): Correct a comment, and add a debug assertion that a
record that contains NULL BLOB pointers must be a fresh insert.
row_upd_clust_rec(): When columns are being moved off-page, invoke
btr_mark_freed_leaves(nonfree=TRUE) and pass the mini-transaction as
the alloc_mtr to btr_store_big_rec_extern_fields(). Finally, invoke
btr_mark_freed_leaves(nonfree=FALSE) to avoid leaking pages.
buf_reset_check_index_page_at_flush(): Remove. The function
fsp_init_file_page_low() already sets
bpage->check_index_page_at_flush=FALSE.
There is a known issue in tablespace extension. If the request to
allocate a BLOB page leads to the tablespace being extended, crash
recovery could see BLOB writes to pages that are off the tablespace
file bounds. This should trigger an assertion failure in fil_io() at
crash recovery. The safe thing would be to write redo log about the
tablespace extension to the mini-transaction of the BLOB write, not to
the mini-transaction of the record update. However, there is no redo
log record for file extension in the current redo log format.
rb:693 approved by Sunny Bains
Suppress the known warnings generated by filesort().
The real fix belongs to worklog 1509:
Pack values of non-sorted fields in the sort buffer
(which is basically the same issue, but in an optimization context:
We are writing the entire sort buffer to disk,
including un-used space for varchar columns.)
mysql-test/valgrind.supp:
Add new Memcheck suppressions for filesort.
sql/filesort.cc:
Remove the ifdef HAVE_purify/bzero code, use valgrind suppressions instead.
PARTITONING, ON INDEX CREATE
If the first partition succeeded in adding a index, but a successive partition failed,
then the first partition had still the new index.
The fix reverts the added indexes from previous partitions on failure.
CRASHES SERVER
Flushing of MERGE table or one of its child tables, which was
locked by flushing thread using LOCK TABLES, might have caused
crashes or assertion failures if the thread failed to reopen
child or parent table.
Particularly, this might have happened when another connection
killed this FLUSH TABLE statement/connection.
Also this problem might have occurred when we failed to reopen
MERGE table or one of its children when executing DDL statement
under LOCK TABLES.
The problem was caused by the fact that reopen_tables() might
have failed to reopen child table but still tried to reopen,
reattach children for and re-lock its parent. Vice versa it
might have failed to reopen parent but kept references from
children to parent around. Since reopen_tables() closes table
it has failed to reopen and therefore frees all associated
memory such dangling references led to crashes when followed.
This patch solves this problem by ensuring that we always close
parent table and all its children if we fail to reopen this
table or one of its children. Same happens if we fail to reattach
children to parent.
Affects 5.1 only.
mysql-test/r/merge.result:
A test case for BUG#11763712.
mysql-test/t/merge.test:
A test case for BUG#11763712.
sql/sql_base.cc:
When flushing tables under LOCK TABLES, all locked
and flushed tables are released and then reopened.
It may happen that we failed to reopen some tables,
in this case we reopen as much tables as possible.
If it was not possible to reopen MERGE child, MERGE
parent is unusable and must be removed from thread
open tables list.
If it was not possible to reopen MERGE parent, all
MERGE child table objects are unusable as well, at
least because their locks are handled by MERGE parent.
They must also be removed from thread open tables
list.
In other words if it was impossible to reopen any
object of a MERGE table or reattach child tables,
all objects of this MERGE table must be considered
unusable and closed.
for compressed InnoDB tables
ha_innodb::info_low(): For calculating data_length or index_length,
use the compressed page size for compressed tables instead of UNIV_PAGE_SIZE.
rb:714 approved by Sunny Bains
There is an optimization of DISTINCT in JOIN::optimize()
which depends on THD::used_tables value. Each SELECT statement
inside SP resets used_tables value(see mysql_select()) and it
leads to wrong result. The fix is to replace THD::used_tables
with LEX::used_tables.
mysql-test/r/sp.result:
test case
mysql-test/t/sp.test:
test case
sql/sql_base.cc:
THD::used_tables is replaced with LEX::used_tables
sql/sql_class.cc:
THD::used_tables is replaced with LEX::used_tables
sql/sql_class.h:
THD::used_tables is replaced with LEX::used_tables
sql/sql_insert.cc:
THD::used_tables is replaced with LEX::used_tables
sql/sql_lex.cc:
THD::used_tables is replaced with LEX::used_tables
sql/sql_lex.h:
THD::used_tables is replaced with LEX::used_tables
sql/sql_prepare.cc:
THD::used_tables is replaced with LEX::used_tables
sql/sql_select.cc:
THD::used_tables is replaced with LEX::used_tables
The problem is that TIME_FUZZY_DATE is explicitly used for get_arg0_date()
function in Item_date_typecast::get_date method. The fix is to use real
fuzzy_date value.
mysql-test/r/func_time.result:
test case
mysql-test/t/func_time.test:
test case
sql/item_timefunc.cc:
use real fuzzy_date value
The buffer was simply too small.
In 5.5 and trunk, the size is 311 + 31,
in 5.1 and below, the size is 331
client/sql_string.cc:
Increase buffer size in String::set(double, ...)
include/m_string.h:
Increase FLOATING_POINT_BUFFER
mysql-test/r/type_float.result:
New test cases.
mysql-test/t/type_float.test:
New test cases.
sql/sql_string.cc:
Increase buffer size in String::set(double, ...)
sql/unireg.h:
Move definition of FLOATING_POINT_BUFFER
The server crashes if it processes table map events that are
corrupted, especially if they map different tables to the same
identifier. This could happen, for instance, due to BUG 56226.
We fix this by checking whether the table map has already been
mapped before actually applying the event. If it has been mapped
with different settings an error is raised and the slave SQL
thread stops. If it has been mapped with same settings the event
is skipped. If the table is set to be ignored by the filtering
rules, there is no change in behavior: the event is skipped and
ids are not checked.
mysql-test/suite/rpl/t/rpl_row_corruption.test:
Added a simple test case that checks both cases:
- multiple table maps with the same identifier
- multiple table maps with the same identifier, but only one
is processed (the others are filtered out)
When CREATE TABLE wasn't given ENGINE=... it would determine
the default ENGINE at parse-time rather than at execution
time, leading to incorrect behaviour (namely, later changes
to the default engine being ignore) when calling CREATE TABLE
from a stored procedure.
We now defer working out the default engine till execution of
CREATE TABLE.
mysql-test/r/sp_trans.result:
results!
mysql-test/t/sp_trans.test:
Show that CREATE TABLE (called from store routine) heeds
any changes after CREATE SP / parse-time. Show that explicitly
requesting an ENGINE still works.
sql/sql_parse.cc:
If no ENGINE=... was given at parse-time, determine default
engine at execution time of CREATE TABLE.
sql/sql_yacc.yy:
If CREATE TABLE is not given ENGINE=..., don't bother
figuring out the default engine during parsing; we'll
do it at execution time instead to be aware of the
latest updates.
We must allocate a larger ref_pointer_array. We failed to account for extra
items allocated here:
#0 find_order_in_list
uint el= all_fields.elements;
all_fields.push_front(order_item); /* Add new field to field list. */
ref_pointer_array[el]= order_item;
order->item= ref_pointer_array + el;
#1 setup_order
#2 setup_without_group
#3 JOIN::prepare
mysql-test/r/order_by.result:
New test case.
mysql-test/r/union.result:
New test case.
mysql-test/t/order_by.test:
New test case.
mysql-test/t/union.test:
New test case.
sql/sql_lex.cc:
find_order_in_list() may need some extra space, so multiply og_num by two.
sql/sql_union.cc:
For UNION, the 'n_sum_items' are accumulated in the "global_parameters" select_lex.
This number must be propagated to setup_ref_array()
When preparing a 'fake_select_lex' we need to use global_parameters->order_list
rather than fake_select_lex->order_list (see comments inside st_select_lex_unit::cleanup)
bug. It added this assert;
ut_ad(ind_field->prefix_len);
before a section of code that assumes there is a prefix_len.
The patch replaced code that explicitly avoided this with a check for
prefix_len. It turns out that the purge thread can get to that assert
without a prefix_len because it does not use a row_ext_t* .
When UNIV_DEBUG is not defined, the affect of this is that the purge thread
sets the dfield->len to zero and then cannot find the entry in the index to
purge. So secondary index entries remain unpurged.
This patch does not do the assert. Instead, it uses
'if (ind_field->prefix_len) {...}'
around the section of code that assumes a prefix_len. This is the way the
patch I provided to Marko did it.
The test case is simply modified to do a sleep(10) in order to give the
purge thread a chance to run. Without the code change to row0row.c, this
modified testcase will assert if InnoDB was compiled with UNIV_DEBUG.
I tried to sleep(5), but it did not always assert.
SYNTAX TRIGGERS IN ANY WAY
Table with triggers which were using deprecated (5.0-only) syntax became
unavailable for any DML and DDL after upgrade to 5.1 version of server.
Attempt to execute any statement on such a table resulted in parsing
error reported. Since this included DROP TRIGGER and DROP TABLE
statements (actually, the latter was allowed but was not functioning
properly for such tables) it was impossible to fix the problem without
manual operations on .TRG and .TRN files in data directory.
The problem was that failure to parse trigger body (due to 5.0-only
syntax) when opening trigger file for a table prevented the table
from being open. This made all operations on the table impossible
(except DROP TABLE which due to peculiarity in its implementation
dropped the table but left trigger files around).
This patch solves this problem by silencing error which occurs when
we parse trigger body during table open. Error message is preserved
for the future use and table is marked as having a broken trigger.
We also try to analyze parse tree to recover trigger name, which
will be needed in order to drop the broken trigger. DML statements
which invoke triggers on the table marked as having broken trigger
are prohibited and emit saved error message. The same happens for
DDL which change triggers except DROP TRIGGER and DROP TABLE which
try their best to do what was requested. Table becomes no longer
marked as having broken trigger when last such trigger is dropped.
mysql-test/r/trigger-compat.result:
Add results for test case for bug#45235
mysql-test/t/trigger-compat.test:
Add test case for bug#45235.
sql/sp_head.cc:
Added protection against MEM_ROOT double restoring to
sp_head::restore_thd_mem_root() method. Since this
method can be sometimes called twice during parsing
of stored routine (the first time during normal flow
of parsing, and the second time when a syntax error
is detected) we need to shortcut execution of the
method to avoid damaging MEM_ROOT by the second
consecutive call to this method.
sql/sql_trigger.cc:
Added error handler Deprecated_trigger_syntax_handler to
catch non-OOM errors during parsing of trigger body.
Added handling of parse errors into method
Table_triggers_list::check_n_load().
sql/sql_trigger.h:
Added new members to handle broken triggers and error messages.
THE EVENT STATUS.
Any ALTER EVENT statement on a disabled event enabled it back
(unless this ALTER EVENT statement explicitly disabled the event).
The problem was that during processing of an ALTER EVENT statement
value of status field was overwritten unconditionally even if new
value was not specified explicitly. As a consequence this field
was set to default value for status which corresponds to ENABLE.
The solution is to check if status field was explicitly specified in
ALTER EVENT statement before assigning new value to status field.
mysql-test/r/events_bugs.result:
test's result for Bug#11764334 was added.
mysql-test/t/events_bugs.test:
new test for Bug#11764334 was added.
sql/event_db_repository.cc:
mysql_event_fill_row() was modified: set value for status field
in events tables only in case if statement CREATE EVENT
is being processed or if this value was set in ALTER EVENT
statement.
Event_db_repository::create_event was modified: removed redundant
setting of status field after return from call to mysql_event_fill_row().
sql/event_parse_data.h:
Event_parse_data structure was modified: added flag
status_changed that is set to true if status's value
was changed in ALTER EVENT statement.
sql/sql_yacc.yy:
Set flag status_changed if status was set in ALTER EVENT
statement.
SEEMS TO BE 'LEAKING' INTO THE SCHEMA NAME SPACE)
and bug#12428824 (Parser stack overflow and crash in sp_add_used_routine
with obscure query).
The first problem was that attempts to call a stored function by
its fully qualified name ended up with unwarranted error "ERROR 1305
(42000): FUNCTION someMixedCaseDb.my_function_name does not exist"
if this function belonged to a schema that had uppercase letters in
its name AND --lower_case_table_names was equal to either 1 or 2.
The second problem was that 5.5 version of MySQL server might have
crashed when a user tried to call stored function with too long name
or too long database name (i.e if a function and database name combined
occupied more than 2*3*64 bytes in utf8). This issue didn't affect
versions of server < 5.5.
The first problem was caused by the fact that in cases when a stored
function was called by its fully qualified name we didn't lowercase
name of its schema before performing look up of the function in
mysql.proc table even although lower_case_table_names mode was on.
As result we were unable to find this function since during its
creation we store lowercased version of schema name in the system
table in this mode and field for schema name uses binary collation.
Calls to stored functions were unaffected by this problem since for
them schema name is converted to lowercase as necessary.
The reason for the second bug was that MySQL Server didn't check length
of function name and database name before proceeding with execution of
stored function. As a consequence too long database name or function
name caused buffer overruns in places where the code assumes that their
length is within fixed limits, like mdl_key_init() in 5.5.
Again this issue didn't affect calls to stored procedures as for them
length of schema name and procedure name are properly checked.
This patch fixes both these bugs by adding calls to check_db_name()
and check_routine_name() to grammar rule which corresponds to a call
to a stored function. These functions ensure that length of database
name and function name for routine called is within standard limit.
Moreover call to check_db_name() handles conversion of database name
to lowercase if --lower_case_table_names mode is on.
Note that even although the second issue seems to be only reproducible
in 5.5 we still add code fixing it to 5.1 to be on the safe side (and
make code a bit more robust against possible future changes).
mysql-test/r/sp-error.result:
Added testcase results for bug#12428824.
mysql-test/r/sp.result:
Added testcase result for bug#11840395.
mysql-test/t/sp-error.test:
Added testcase for bug#12428824.
mysql-test/t/sp.test:
Added testcase for bug#11840395.
sql/sql_yacc.yy:
Modified 'function_call_generic' rule to call check_db_name() and
check_routine_name() in order to ensure that lengths of database name
and function name are within limits. check_db_name() is also responsible
for normalizing function's database name for lookup in cases when
lowercase_table_names mode is on.
Problem: in case of wrong data insert into indexed GEOMETRY fields
(e.g. NULL value for a not NULL field) MyISAM reported
"ERROR 126 (HY000): Incorrect key file for table; try to repair it"
due to misuse of the key deletion function.
Fix: always use R-tree key functions for R-tree based indexes
and B-tree key functions for B-tree based indexes.
mysql-test/r/gis-rtree.result:
Bug#11764487: myisam corruption with insert ignore and invalid spatial data
- test result.
mysql-test/t/gis-rtree.test:
Bug#11764487: myisam corruption with insert ignore and invalid spatial data
- test case.
storage/myisam/mi_update.c:
Bug#11764487: myisam corruption with insert ignore and invalid spatial data
- handling update errors check for HA_ERR_NULL_IN_SPATIAL as well to be
consistent with mi_write();
- always use keyinfo->ck_delete()/ck_insert() instead of _mi_ck_delete()/_mi_ck_write()
to handle index properly, as it may be of B-tree or R-tree type.
storage/myisam/mi_write.c:
Bug#11764487: myisam corruption with insert ignore and invalid spatial data
- always use keyinfo->ck_delete() instead of _mi_ck_delete() to handle
index properly, as it may be of B-tree or R-tree type.
row_build_index_entry(): In innodb_file_format=Barracuda
(ROW_FORMAT=DYNAMIC or ROW_FORMAT=COMPRESSED), a secondary index on a
full column can refer to a field that is stored off-page in the
clustered index record. Take that into account.
rb:692 approved by Jimmy Yang
BOGUS "THE TABLE MYSQL.PROC IS MISSING,..."
There was a race condition between loading a stored routine
(function/procedure/trigger) specified by fully qualified name
SCHEMA_NAME.PROC_NAME and dropping the stored routine database.
The problem was that there is a window for race condition when one server
thread tries to load a stored routine being executed and the other thread
tries to drop the stored routine schema.
This condition race window exists in implementation of function
mysql_change_db() called by db_load_routine() during loading of stored
routine to cache. Function mysql_change_db() calls check_db_dir_existence()
that might failed because specified database was dropped during concurrent
execution of DROP SCHEMA statement. db_load_routine() calls mysql_change_db()
with flag 'force_switch' set to 'true' value so when referenced db is not found
then my_error() is not called and function mysql_change_db() returns ok.
This shadows information about schema opening error in db_load_routine().
Then db_load_routine() makes attempt to parse stored routine that is failed.
This makes to return error to sp_cache_routines_and_add_tables_aux() but since
during error generation a call to my_error wasn't made and hence
THD::main_da wasn't set we set the generic "mysql.proc table corrupt" error
when running sp_cache_routines_and_add_tables_aux().
The fix is to install an error handler inside db_load_routine() for
the mysql_op_change_db() call, and check later if the ER_BAD_DB_ERROR
was caught.
sql/sql_db.cc:
Added synchronization point "before_db_dir_check" to emulate a race condition during
processing of CALL/DROP SCHEMA.
TO POSITION FIRST CAN CAUSE DATA TO BE CORRUPTED".
ALTER TABLE MODIFY/CHANGE ... FIRST did nothing except renaming
columns if new version of the table had exactly the same
structure as the old one (i.e. as result of such statement, names
of columns changed their order as specified but data in columns
didn't). The same thing happened for ALTER TABLE DROP COLUMN/ADD
COLUMN statements which were supposed to produce new version of
table with exactly the same structure as the old version of table.
I.e. in the latter case the result was the same as if old column
was renamed instead of being dropped and new column with default
as value being created.
Both these problems were caused by the fact that ALTER TABLE
implementation incorrectly interpreted both these situations as
simple renaming of columns and assumed that in-place ALTER TABLE
algorithm could have been used for them.
This patch fixes this problem by ensuring that in cases when some
column is moved to the first position or some column is dropped
the default ALTER TABLE algorithm involving table copying is
always used. This is achieved by detecting such situations in
mysql_prepare_alter_table() and setting Alter_info::change_level
to ALTER_TABLE_DATA_CHANGED for them.
mysql-test/r/alter_table.result:
Added test for bug #12652385 - "61493: REORDERING COLUMNS TO
POSITION FIRST CAN CAUSE DATA TO BE CORRUPTED".
mysql-test/t/alter_table.test:
Added test for bug #12652385 - "61493: REORDERING COLUMNS TO
POSITION FIRST CAN CAUSE DATA TO BE CORRUPTED".
sql/sql_table.cc:
Changed mysql_prepare_alter_table() to detect situations in
which we some column moved to the first position or some column
is dropped and ensure that such ALTER TABLE statements won't
be carried out using in-place algorithm. The latter could have
happened before this patch if new version of table had the same
structure as the old one (except the column names).
FAIL IN EMBEDDED SERVER
FreeBSD 64 bit needs the FP_X_DNML to fpsetmask() to prevent exceptions from
propagating into mysql (as a threaded application).
However fpsetmask() itself is deprecated in favor of fedisableexcept().
1. Fixed the #ifdef to check for FP_X_DNML instead of i386.
2. Added a configure.in check for fedisableexcept() and, if present,
this function is called insted of the fpsetmask().
No need for new tests, as the existing tests cover this already.
Removed the affected tests from the experimental list.
