if cached query uses many tables
The problem was that query cache would not properly cache
queries which used 256 or more tables but yet would leave
behind query cache blocks pointing to freed (destroyed)
data. Later when invalidating (due to a truncate) query cache
would attempt to grab a lock which resided in the freed data,
leading to hangs or undefined behavior.
This was happening due to a improper return value from the
function responsible for registering the tables used in the
query (so the cache can be invalidated later if one of the
tables is modified). The function expected a return value of
type boolean (char, 8 bits) indicating success (1) or failure
(0) but the number of tables registered (unsigned int, 32 bits)
was being returned instead. This caused the function to return
failure for cases where it had actually succeed because when
a type (unsigned int) is converted to a narrower type (char),
the excess bits on the left are discarded. Thus if the 8
rightmost bits are zero, the return value will be 0 (failure).
The solution is to simply return true (1) only if the number of
registered table is greater than zero and false (0) otherwise.
Fix for this bug and additional improvements/fixes
In detail:
- Remove unicode attribute from several columns
(unicode properties were nowhere needed/tested)
of the table tb3
-> The runnability of these tests depends no more on
the availibility of some optional collations.
- Use a table tb3 with the same layout for all
engines to be tested and unify the engine name
within the protocols.
-> <engine>_trig_<abc>.result have the same content
- Do not load data into tb3 if these rows have no
impact on result sets
- Add tests for NDB (they exist already in 5.1)
- "--replace_result" at various places because
NDB variants of tests failed with "random" row
order in results
This fixes a till now unknown weakness within the
funcs_1 NDB tests existing in 5.1 and 6.0
- Fix the expected result of ndb_trig_1011ext
which suffered from Bug 32656
+ disable this test
- funcs_1 could be executed with the mysql-test-run.pl
option "--reorder", which saves some runtime by
optimizing server restarts.
Runtimes on tmpfs (one attempt only):
with reorder 132 seconds
without reorder 183 seconds
- Adjust two "check" statements within func_misc.test
which were incorrect (We had one run with result set
difference though the server worked good.)
- minor fixes in comments
Upmerge of fix for this bug and a second similar problem
found during experimenting.
This replaces the first fix (already pushed to 5.1
and merged to 6.0) which
- failed in runs with the embedded server
- cannot be ported back to 5.0
Fix for this bug and a second similar problem
found during experimenting.
This replaces the first fix (already pushed to 5.1
and merged to 6.0) which
- failed in runs with the embedded server
- cannot be ported back to 5.0
Bug#36724 - Test funcs_1.<engine>_storedproc_02 needs to be updated
Bug#36726 - Test funcs_1.<engine>_storedproc failing - Needs to be updated on 5.1+
func_view bug: re-records .result files to account for addition of charset and collation data
to SHOW CREATE VIEW output
storedproc bugs: Added expected errors for those storedprocs that use SQLSTATE:00000
in their handlers. re-recorded .result files to account for these
expected errors.
There was a typo in a error check causing wrong thing to be ckecked
and a possible error not being caught.
Fixed by using the correct variable to test for malloc() errors.
The Diagnostic_area caused an assertion failure in debug mode when
the disk was full.
By setting the internal error handler to ignore errors caused by
underlying logging methods, the error is avoided.
The failing test case is depending on unnecessary status variable output
which changes based on build configuration. By reducing the output the test
becomes more stable.
the local tree contains a fix for
Bug#32748 "Inconsistent handling of assignments to
general_log_file/slow_query_log_file",
which changes output of a number of tests.
PREPARE", review fixes:
- make the patch follow the specification of WL#4166 and remove
the new error that was originally introduced.
Now the client never gets an error from reprepare, unless it failed.
I.e. even if the statement at hand returns a completely different
result set, this is not considered a server error.
The C API library, that can not handle this situation, was modified to
return a client error.
Added additional test coverage.
The test is vulnerable because it does not check if slave has stopped at time
of the new session is requested `start slave;'
Fixed with deploying explicitly wait_for_slave_to_stop synchronization macro.
first row or fails with an error:
ERROR 1022 (23000): Can't write; duplicate key in table ''
The server uses intermediate temporary table to store updated
row data. The first column of this table contains rowid.
Current server implementation doesn't reset NULL flag of that
column even if the server fills a column with rowid.
To keep each rowid unique, there is an unique index.
An insertion into an unique index takes into account NULL
flag of key value and ignores real data if NULL flag is set.
So, insertion of actually different rowids may lead to two
kind of problems. Visible effect of each of these problems
depends on an initial engine type of temporary table:
1. If multiupdate initially creates temporary table as
a MyISAM table (a table contains blob columns, and the
create_tmp_table function assumes, that this table is
large), it inserts only one single row and updates
only rows with one corresponding rowid. Other rows are
silently ignored.
2. If multiupdate initially creates MEMORY temporary
table, fills it with data and reaches size limit for
MEMORY tables (max_heap_table_size), multiupdate
converts MEMORY table into MyISAM table and fails
with an error:
ERROR 1022 (23000): Can't write; duplicate key in table ''
Multiupdate has been fixed to update the NULL flag of
temporary table rowid columns.
