- Fixed some issues with partitions and connection_string, which also fixed lp:716890 "Pre- and post-recovery crash in Aria"
- Fixed wrong assert in Aria
Now need to merge with latest xtradb before pushing
sql/ha_partition.cc:
Ensure that m_ordered_rec_buffer is not freed before close.
sql/mysqld.cc:
Changed to use opt_stack_trace instead of opt_pstack.
Removed references to pstack
sql/partition_element.h:
Ensure that connect_string is initialized
storage/maria/ma_key_recover.c:
Fixed wrong assert
sql/multi_range_read.cc:
Added printing of error if something goes wrong in get_next()
(Not critical for this bug fix, but this was something that I noticed while testing and found missing)
storage/myisam/mi_rkey.c:
Fixed wrong error number in mi_yield_and_check_if_killed()
- In Maria/MyISAM: Release/re-acquire locks to give queries that wait on them a chance to make progress
- In Maria/MyISAM: Change from numeric constants to ICP_RES values.
- In Maria: Do check index condition in maria_rprev() (was lost in the merge/backport?)
- In Maria/MyISAM/XtraDB: Check if the query was killed, and return immediately if it was.
Added new storage engine error: HA_ERR_ABORTED_BY_USER, for handler to signal that it detected a kill of the query and aborted
Authors: Sergey Petrunia & Monty
include/my_base.h:
Added HA_ERR_ABORTED_BY_USER, for handler to signal that it detected a kill of the query and aborted
include/my_handler.h:
Added comment
mysql-test/r/myisam_icp.result:
Updated test
mysql-test/t/myisam_icp.test:
Drop used tables at start of test
Added test case that can help with manual testing of killing index condition pushdown query.
mysys/my_handler_errors.h:
Text for new storage engine error
sql/handler.cc:
If engine got HA_ERR_ABORTED_BY_USER, send kill message.
sql/multi_range_read.cc:
Return error code
storage/maria/ha_maria.cc:
Added ma_killed_in_mariadb() to detect kill.
Ensure that file->external_ref points to TABLE object.
storage/maria/ma_extra.c:
Dummy test-if-killed for standalone
storage/maria/ma_key.c:
If ma_check_index_cond() fails, set my_errno and info->cur_row.lastpos
storage/maria/ma_rkey.c:
Release/re-acquire locks to give queries that wait on them a chance to make progress
Check if the query was killed, and return immediately if it was
storage/maria/ma_rnext.c:
Check if the query was killed, and return immediately if it was
Added missing fast_ma_writeinfo(info)
storage/maria/ma_rnext_same.c:
Check if the query was killed, and return immediately if it was
Added missing fast_ma_writeinfo(info)
storage/maria/ma_rprev.c:
Check if the query was killed, and return immediately if it was
Added missing fast_ma_writeinfo(info) and ma_check_index_cond()
storage/maria/ma_search.c:
Give error message if we find a wrong key
storage/maria/ma_static.c:
Added pointer to test-if-killed function
storage/maria/maria_def.h:
New prototypes
storage/myisam/ha_myisam.cc:
Added mi_killed_in_mariadb()
Ensure that file->external_ref points to TABLE object.
storage/myisam/mi_extra.c:
Dummy test-if-killed for standalone
storage/myisam/mi_key.c:
If ma_check_index_cond() fails, set my_errno and info->lastpos
storage/myisam/mi_rkey.c:
Ensure that info->lastpos= HA_OFFSET_ERROR in case of error
Release/re-acquire locks to give queries that wait on them a chance to make progress
Check if the query was killed, and return immediately if it was
Reorder code to do less things in case of error.
Added missing fast_mi_writeinfo()
storage/myisam/mi_rnext.c:
Check if the query was killed, and return immediately if it was
Simplify old ICP code
Added missing fast_ma_writeinfo(info)
storage/myisam/mi_rnext_same.c:
Check if the query was killed, and return immediately if it was
Added missing fast_mi_writeinfo(info)
storage/myisam/mi_rprev.c:
Check if the query was killed, and return immediately if it was
Simplify error handling of ICP
Added missing fast_mi_writeinfo(info)
storage/myisam/mi_search.c:
Give error message if we find a wrong key
storage/myisam/mi_static.c:
Added pointer to test-if-killed function
storage/myisam/myisamdef.h:
New prototypes
storage/xtradb/handler/ha_innodb.cc:
Added DB_SEARCH_ABORTED_BY_USER and ha_innobase::is_thd_killed()
Check if the query was killed, and return immediately if it was
storage/xtradb/handler/ha_innodb.h:
Added prototype
storage/xtradb/include/db0err.h:
Added DB_SEARCH_ABORTED_BY_USER
storage/xtradb/include/row0mysql.h:
Added possible ICP errors
storage/xtradb/row/row0sel.c:
Use ICP errors instead of constants.
Detect if killed and return B_SEARCH_ABORTED_BY_USER
The loop that was looping over subqueries' references to outer field used a
local boolean variable to tell whether the field was grouped or not. But the
implementor failed to reset the variable after each iteration. Thus a field
that was not directly aggregated appeared to be.
Fixed by resetting the variable upon each new iteration.
memory reference
There are two issues present here.
1) There is a possibility that we test a byte beyond the
allocated buffer
2) We compare a byte that might never have been
initalized to see if it's 0.
The first issue is not triggered by existing code, but an
ASSERT has been added to safe-guard against introducing
new code that triggers it.
The second issue is what triggers the Valgrind warnings
reported in the bug report. A buffer is allocated in
class String to hold the value. This buffer is populated
by the character data constituting the string, but is not
zero-terminated in most cases. Testing if it is indeed
zero-terminated means that we check a byte that has never
been explicitly set, thus causing Valgrind to trigger.
Note that issue 2 is not a serious problem. The variable
is read, and if it's not zero, we will set it to zero.
There are no further consequences.
Note that this patch does not fix the underlying problems
with issue 1, as it is deemed too risky to fix at this
point (as noted in the bug report). As discussed in
the report, the c_ptr() method should probably be
replaced, but this requires a thorough analysis of the
~200 calls to the method.
sql/set_var.cc:
These two cases have been reported to fail
with Valgrind.
The patch also adjusts several instable test results
to order the result.
Analysis:
The function prev_record_reads() may skip (jump over)
some query plan nodes where record_count < 1. At the
same time, even though get_partial_join_cost() uses
all first N plan nodes after the last constant table,
it may produce a smaller record_count than
prev_record_reads(), because the record count for
some plan nodes may be < 1, and these nodes may not
participate in prev_record_reads.
Solution:
The current solution is to treat the result of
get_partial_join_cost() as the upper bound for the
total number of unique lookup keys.
attempt to create spatial index on char > 31 bytes".
Attempt to create spatial index on char field with length
greater than 31 byte led to assertion failure on server
compiled with safemutex support.
The problem occurred in mi_create() function which was called
to create a new version of table being altered. This function
failed since it detected an attempt to create a spatial key
on non-binary column and tried to return an error.
On its error path it tried to unlock THR_LOCK_myisam mutex
which has not been not locked at this point. Indeed such an
incorrect behavior was caught by safemutex wrapper and caused
assertion failure.
This patch fixes the problem by ensuring that mi_create()
doesn't releases THR_LOCK_myisam mutex on error path if it was
not acquired.
mysql-test/r/gis.result:
Added test for bug @59888 "debug assertion when attempt to
create spatial index on char > 31 bytes".
mysql-test/t/gis.test:
Added test for bug @59888 "debug assertion when attempt to
create spatial index on char > 31 bytes".
storage/myisam/mi_create.c:
Changed mi_create() not to release THR_LOCK_myisam mutex on
error path if it was not acquired.
rw_lock_create_func(): Initialize lock->writer_thread, so that Valgrind
will not complain even when Valgrind instrumentation is not enabled.
Flag lock->writer_thread uninitialized, so that Valgrind can complain
when it is used uninitialized.
rw_lock_set_writer_id_and_recursion_flag(): Revert the bogus Valgrind
instrumentation that was pushed in the first attempt to fix this bug.
This bug extends the fix for LP BUG#715027 to cover one
more case of an Item being transformed, and its property
Item::with_subselect not being updated because
quick_fix_fields doesn't recalculate any properties.
Analysis:
Before calling:
write_record= (select->skip_record(thd) > 0);
the function find_all_keys needs to restore the original read/write
sets of the table that is sorted if the condition select->cond
contains a subquery.
This didn't happen in this test case because the flag "with_subselect"
was not set properly for select->cond.
The reason for the flag not being set properly, was that this condition
was rewritten by add_cond_and_fix() inside make_join_select() by:
/* Add conditions added by add_not_null_conds(). */
if (tab->select_cond)
add_cond_and_fix(thd, &tmp, tab->select_cond);
However, the function add_cond_and_fix() called the shortcut method
Item::quick_fix_field() that didn't update the "with_subselect"
property.
Solution:
Call the complete Item::fix_fields() to update all Item properties,
including "with_subselect".
Analysis:
The failed assert is a result of calling Item_sum_distinct::clear()
on an incomplete object for which Item_sum_distinct::setup() was
not yet called.
The reason is that JOIN::exec for the outer query calls JOIN::reinit()
for all its subqueries, which in turn calls clear() for all aggregate
functions of the subqueries. The call stack is:
mysql_explain_union -> mysql_select -> JOIN::exec -> select_desribe ->
mysql_explain_union -> mysql_select -> JOIN::reinit
This assert doesn't fail in the main 5.3 because constant subqueries
are being executed during the optimize phase of the outer query,
thus the Unique object is created before calling JOIN::exec for the
outer query, and Item_sum_distinct::clear() actually cleans the
Unique object.
Solution:
The best solution is the obvious one - substitute the assert with
a test whether Item_sum_distinct::tree is NULL.
Fixed that connection string is returned for partitioned federated tables.
mysql-test/r/partition_federated.result:
Fixed error message
mysql-test/suite/federated/federated_partition.result:
Added test to show that connection string is returned in 'show create'.
sql/ha_partition.cc:
Fixed a set of bugs introduced by the last federated patch:
- We can't allocate m_ordered_rec_buffer in memroot as it has to survive call to clear_handler_file()
sql/partition_element.h:
Ensure that connect_string is properly initialized.
(This caused crashed in partition tests)
sql/sql_partition.cc:
Print CONNECTION option for federated partitioned tables
Analysis:
The crash in EXPLAIN resulted from an attempt to print the
name of the internal temporary table created to compute
distinct for the innermost subquery of the test case.
Such tables do not have a corresponding TABLE_LIST (table
reference), hence the crash. The reason for this was that
the subquery was executed as part of constant condition
evaluation before EXPLAIN attempts to print the table name.
During the subquery execution, the subquery JOIN_TAB and
its table are substituted by a temporary table in
make_simple_join.
Solution:
Similar to the analogous case for other Items than the
IS NULL function, do not evaluate expensive constant
conditions.
- Fixed Partition engine to store CONNECTION string for partitions.
Removed HA_NO_PARTITION flag from FederatedX.
Added test 'federated_partition' to suite.
- lp:#585688 - maridb crashes in federatedx code
FederatedX handler instances, created on one thread and used on
another thread (via table cache) when "show table status" is executed
crashed because txn member was not initialized for current thread.
Added test 'federated_bug_585688' to suite.
Author for the patch is Antony Curtis
mysql-test/suite/federated/federated_bug_585688.result:
Test for lp:585688
mysql-test/suite/federated/federated_bug_585688.test:
Test for lp:585688
mysql-test/suite/federated/federated_partition-slave.opt:
Test for partition support in federatedx
mysql-test/suite/federated/federated_partition.result:
Test for partition support in federatedx
mysql-test/suite/federated/federated_partition.test:
Test for partition support in federatedx
mysql-test/t/partition_federated.test:
Updated error message
sql/ha_partition.cc:
Added support for connection strings to partitions for federatedx
sql/ha_partition.h:
Added support for connection strings to partitions for federatedx
sql/partition_element.h:
Added support for connection strings to partitions for federatedx
sql/sql_yacc.yy:
Added support for connection strings to partitions for federatedx
storage/federatedx/ha_federatedx.cc:
Added support for partitions.
FederatedX handler instances, created on one thread and used on another thread (via table cache) when "show table status"
is executed crashed because txn member was not initialized for current thread.
- Fixed bug lp:624099 ma_close.c:75: maria_close: Assertion `share->in_trans == 0' failed on UNLOCK TABLES
- Fixed bug that caused table to be marked as not closed (crashed) during recovery testing.
- Use maria_delete_table_files() instead of maria_delete_table() to delete tempoary tables (faster and safer)
- Added checks to ensure that bitmap and internal mutex are always taken in right order.
- For transactional tables, only mark the table as changed before page for table is to be written to disk (and thus the log is flushed).
This speeds up things a bit and fixes a problem where open_count was incremented on disk but there was no log entry to fix it during recovery -> table was crashed.
- Fixed a bug in repair() where table was not automaticly repaired.
- Ensure that state->global_changed, share->changed and share->state.open_count are set and reset properly.
- Added option --ignore-control-file to maria_chk to be able to run maria_chk even if the control file is locked.
mysql-test/suite/maria/r/maria-recover.result:
Test result changed as we now force checkpoint before copying table, which results in pagecache getting flushed and we have more rows to recover.
mysql-test/suite/maria/r/maria.result:
Added new tests
mysql-test/suite/maria/t/maria-recover.test:
Force checkpoint before copying table.
This is needed as now the open-count is increased first when first page is flushed.
mysql-test/suite/maria/t/maria.test:
Added tests to verify fix for lp:624099
storage/maria/ha_maria.cc:
Use table->in_use instead of current_thd (trivial optimization)
Use maria_delete_table_files() instead of maria_delete_table() to delete tempoary tables (faster and safer)
More DBUG_ASSERT()
Reset locked tables count after locked tables have been moved to new transaction. This fixed lp:624099
storage/maria/ma_bitmap.c:
Temporarly unlock bitmap mutex when calling _ma_mark_file_changed() and pagecache_delete_pages() to ensure right mutex lock order.
Call _ma_bitmap_unpin_all() when bitmap->non_flusable is set to 0. This fixed a case when bitmap was not proparly unpinned.
More comments
Added DBUG_ASSERT() for detecting wrong share->bitmap usage
storage/maria/ma_blockrec.c:
More DBUG_ASSERT()
Moved code around in _ma_apply_redo_insert_row_head_or_tail() to make things safer on error conditions.
storage/maria/ma_check.c:
Changed parameter for _ma_set_uuid()
Corrected test for detecting if we lost many rows. This fixed some cases where auto-recovery failed.
share->changed need to be set if state.open_count is changed.
Removed setting of share->changed= 0 as called function sets it.
storage/maria/ma_close.c:
- Added code to properly decrement open_count and have it written by _ma_state_info_write() for transactional tables.
(This is more correct and also saves us one extra write by _ma_decrement_open_count() at close.
- Added DBUG_ASSERT() to detect if open_count is wrong at maria_close().
storage/maria/ma_delete.c:
Updated argument to _ma_mark_file_changed()
storage/maria/ma_delete_all.c:
Updated argument to _ma_mark_file_changed()
For transactional tables, call _ma_mark_file_changed() after log entry has been written (to allow recover to fix open_count)
Reset more needed variables in _ma_reset_status()
storage/maria/ma_delete_table.c:
Moved deletion of Aria files to maria_delete_table_files().
Remove RAID usage (old not working code)
storage/maria/ma_extra.c:
Set share->changed=1 when state needs to be updated on disk.
Don't reset share->changed after call to _ma_state_info_write() as this calls sets share->changed.
Set share->state.open_count to 1 to force table to be auto repaired if drop fails.
Set share->global_changed before call to _ma_flush_table_files() to ensure that we don't try to mark the table changed during flush.
Added DBUG_ENTER
storage/maria/ma_locking.c:
Split _ma_mark_file_changed() into two functions to delay marking transactional tables as changed on disk until first disk write.
Added argument to _ma_decrement_open_count() to tell if we should call ma_lock_database() or not.
Don't decrement open count for transactional tables during _ma_decrement_open_count(). This will be done during close.
Changed parameter for _ma_set_uuid()
storage/maria/ma_open.c:
Set share->open_count_not_zero_on_open if state.open_count is not zero.
This is needed for DBUG_ASSERT() in maria_close() that is there to enforce that open_count is always 0 at close.
This test doesn't however work for tables that had open_count != 0 already on disk (ie, crashed tables).
Enforce right mutex order for share->intern_lock and share->bitmap.bitmap_lock
Don't set share->changed to 0 if share->state.open_count != 0, as state needs to be be written at close
storage/maria/ma_pagecache.c:
Moved a bit of code in find_block() to avoid one if.
More DBUG_ASSERT()
(I left a comment in the code for Sanja to look at; The code probably works but we need to check if it's optimal)
storage/maria/ma_pagecrc.c:
For transactional tables, just before first write to disk, but after log is flushed, mark the file changed.
This fixes some cases where recovery() did not detect that table was marked as changed and could thus not recover the marker.
storage/maria/ma_recovery.c:
Set share->changed when share->global_changed is set.
storage/maria/ma_update.c:
Updated parameter for _ma_mark_file_changed()
storage/maria/ma_write.c:
Updated parameter for _ma_mark_file_changed()
storage/maria/maria_chk.c:
Added option --ignore-control-file to be able to run maria_chk even if the control file is locked.
storage/maria/maria_def.h:
Updated function prototypes.
Added open_count_not_zero_on_open to MARIA_SHARE.
storage/myisam/ha_myisam.cc:
current_thd -> table->in_use
Fixed LP BUG#714808 Assertion `outer_lookup_keys <= outer_record_count'
Analysis:
The function best_access_path() computes the number or records as
follows:
...
if (rec < MATCHING_ROWS_IN_OTHER_TABLE)
rec= MATCHING_ROWS_IN_OTHER_TABLE; // Fix for small tables
...
if (table->quick_keys.is_set(key))
records= (double) table->quick_rows[key];
else
{
/* quick_range couldn't use key! */
records= (double) s->records/rec;
}
Above MATCHING_ROWS_IN_OTHER_TABLE == 10, and s->records == 1,
thus we get an estimated 0.1 records. As a result JOIN::get_partial_join_cost()
for the outer query computes outer_record_count == 0.1 records, which is
meaningless in this context.
Solution:
Round row count estimates that are < 1 to 1.
Test failed on a certain Linux platform in automated environment. It turns out that this platform has an old version of Perl modules DBI and DBD::mysql installed, as well as the OS itself being relatively old.
Allowing error code 11 to be returned from mysqlhotcopy on expected error seems harmless and will make the test pass also with older libraries.
Added --debug-server and use $opt_debug_server where appropriate
Let --debug imply --debug-server
When merging to 5.5, must adapt fix for 59148
Oops, set debug => debug-server too late, fixed
privileges".
The first problem was that DROP USER didn't properly remove privileges
on stored functions from in-memory structures. So the dropped user
could have called stored functions on which he had privileges before
being dropped while his connection was still around.
Even worse if a new user with the same name was created he would
inherit privileges on stored functions from the dropped user.
Similar thing happened with old user name and function privileges
during RENAME USER.
This problem stemmed from the fact that the handle_grant_data() function
which handled DROP/RENAME USER didn't take any measures to update
in-memory hash with information about function privileges after
updating them on disk.
This patch solves this problem by adding code doing just that.
The second problem was that RENAME USER didn't properly update in-memory
structures describing table-level privileges and privileges on stored
procedures. As result such privileges could have been lost after a rename
(i.e. not associated with the new name of user) and inherited by a new
user with the same name as the old name of the original user.
This problem was caused by code handling RENAME USER in
handle_grant_struct() which [sic!]:
a) tried to update wrong (tables) hash when updating stored procedure
privileges for new user name.
b) passed wrong arguments to function performing the hash update and
didn't take into account the way in which such update could have
changed the order of the hash elements.
This patch solves this problem by ensuring that a) the correct hash
is updated, b) correct arguments are used for the hash_update()
function and c) we take into account possible changes in the order
of hash elements.
mysql-test/r/grant.result:
Added test coverage for bug#36544 "DROP USER does not remove stored
function privileges".
mysql-test/suite/funcs_1/r/innodb_storedproc_06.result:
Since after fixing bug#36544 "DROP USER does not remove stored function
privileges" in-memory structures are correctly updated by DROP USER,
DROP FUNCTION performed after DROP USER for its definer no longer
produces unwarranted warning/error messages.
mysql-test/suite/funcs_1/r/memory_storedproc_06.result:
Since after fixing bug#36544 "DROP USER does not remove stored function
privileges" in-memory structures are correctly updated by DROP USER,
DROP FUNCTION performed after DROP USER for its definer no longer
produces unwarranted warning/error messages.
mysql-test/suite/funcs_1/r/myisam_storedproc_06.result:
Since after fixing bug#36544 "DROP USER does not remove stored function
privileges" in-memory structures are correctly updated by DROP USER,
DROP FUNCTION performed after DROP USER for its definer no longer
produces unwarranted warning/error messages.
mysql-test/t/grant.test:
Added test coverage for bug#36544 "DROP USER does not remove stored
function privileges".
sql/sql_acl.cc:
Changed handle_grant_data() to also update hash with function
privileges. This allows DROP/RENAME USER correctly keep this
in-memory structure up-to-date.
To do this extended handle_grant_struct() to support updating of this
hash. In addition fixed code in this function which is responsible for
handling of column and routine hashes during RENAME USER, ensured that
we correctly update these hashes after changing user name and that we
don't skip elements while iterating through the hash and doing updates.
Also fix bug#59110: Memory leak of QUICK_SELECT_I allocated memory.
Includes Jørgen Lølands review comments.
Root cause of these bugs are that test_if_skip_sort_order() decided to
revert the 'skip_sort_order' descision (and use filesort) after the
query plan has been updated to reflect a 'skip' of the sort order.
This might happen in 'check_reverse_order:' if we have a
select->quick which could not be made descending by appending
a QUICK_SELECT_DESC. ().
The original 'save_quick' was then restored after the QEP has been modified,
which caused:
- An incorrect 'precomputed_group_by= TRUE' may have been set,
and not reverted, as part of the already modifified QEP (Bug#59308)
- A 'select->quick' might have been created which we fail to delete (bug#59110).
This fix is a refactorication of test_if_skip_sort_order() where all logic
related to modification of QEP (controlled by argument 'bool no_changes'), is
moved to the end of test_if_skip_sort_order(), and done after *all* 'test_if_skip'
checks has been performed - including the 'check_reverse_order:' checks.
The refactorication above contains now intentional changes to the logic which
has been moved to the end of the function.
Furthermore, a smaller part of the fix address the handling of the
select->quick objects which may already exists when we call
'test_if_skip_sort_order()' (save_quick) -and
new select->quick's created during test_if_skip_sort_order():
- Before new select->quick may be created by calling ::test_quick_select(), we
set 'select->quick= 0' to avoid that ::test_quick_select() prematurely
delete the save_quick's. (After this call we may have both a 'save_quick'
and 'select->quick')
- All returns from ::test_if_skip_sort_order() where we may have both a
'save_quick' and a 'select->quick' has been changed to goto's to the
exit points 'skiped_sort_order:' or 'need_filesort:' where we
decide which of the QUICK_SELECT's to keep, and delete the other.
