filesort() uses file->estimate_rows_upper_bound() call to allocate
internal buffers. If this function returns a value smaller than
a number of row that will be returned later in find_all_keys(),
that can cause server crash.
Fixed by implementing ha_federated::estimate_rows_upper_bound() to
return maximum possible number of rows.
Present estimation for FEDERATED always returns 0 if the linked to the VIEW.
"Federated Denial of Service"
Federated storage engine used to attempt to open connections within
the ::create() and ::open() methods which are invoked while LOCK_open
mutex is being held by mysqld. As a result, no other client sessions
can open tables while Federated is attempting to open a connection.
Long DNS lookup times would stall mysqld's operation and a rogue
connection string which connects to a remote server which simply
stalls during handshake can stall mysqld for a much longer period of
time.
This patch moves the opening of the connection much later, when the
federated actually issues queries, by which time the LOCK_open mutex is
no longer being held.
when creating table
Federated tables had an artificially low maximum of key length,
because the handler failed to implement a method to return it and
the default value is taked from the prototype handler.
Now, implement that method and return the maximum possible key
length, which is that of InnoDB.
"Federared Transactions Failure"
Bug occurs when the user performs an operation which inserts more than
one row into the federated table and the federated table references a
remote table stored within a transactional storage engine. When the
insert operation for any one row in the statement fails due to
constraint violation, the federated engine is unable to perform
statement rollback and so the remote table contains a partial commit.
The user would expect a statement to perform the same so a statement
rollback is expected.
This bug was fixed by implementing bulk-insert handling into the
federated storage engine. This will relieve the bug for most common
situations by enabling the generation of a multi-row insert into the
remote table and thus permitting the remote table to perform
statement rollback when neccessary.
The multi-row insert is limited to the maximum packet size between
servers and should the size overflow, more than one insert statement
will be sent and this bug will reappear. Multi-row insert is disabled
when an "INSERT...ON DUPLICATE KEY UPDATE" is being performed.
The bulk-insert handling will offer a significant performance boost
when inserting a large number of small rows.
This patch builds on Bug29019 and Bug25511
"Federated INSERT failures"
Federated does not correctly handle "INSERT...ON DUPLICATE KEY UPDATE"
However, implementing such support is not reasonably possible without
increasing complexity of the storage engine: checking that constraints
on remote server match local server and parsing error messages.
This patch causes 'ON DUPLICATE KEY' to fail with ER_DUP_KEY message
if a conflict occurs and not to fail silently.
"REPLACE/INSERT IGNORE/UPDATE IGNORE doesn't work"
Federated does not record neccessary HA_EXTRA flags in order to
support REPLACE/INSERT IGNORE/UPDATE IGNORE.
Implement ::extra() to capture flags neccessary for functionality.
New function append_ident() to better escape identifiers consistantly.
(Mostly in DBUG_PRINT() and unused arguments)
Fixed bug in query cache when used with traceing (--with-debug)
Fixed memory leak in mysqldump
Removed warnings from mysqltest scripts (replaced -- with #)
Added HA_NULL_IN_KEY to table flags to allow for nullable unique indexes
and added test to verify
ha_federated.h:
BUG #15133 "unique index with nullable value not accepted in federated table"
added HA_NULL_IN_KEY to table flags to allow for nullable unique indexes
federated.test:
BUG #15133 "unique index with nullable value not accepted in federated table"
New test to show that nullable unique indexes work
federated.result:
BUG #15133 "unique index with nullable value not accepted in federated table"
New results for new test
The idea of the patch
is that every cursor gets its own lock id for table level locking.
Thus cursors are protected from updates performed within the same
connection. Additionally a list of transient (must be closed at
commit) cursors is maintained and all transient cursors are closed
when necessary. Lastly, this patch adds support for deadlock
timeouts to TLL locking when using cursors.
+ post-review fixes.
This is the first changeset of suggested changes recommended in Kostja's
review of my patch, 1.1846, which includes only functionality changes.
Style changes/Documentation patch to follow.
Testing on hp3750 shows these fixes fix the compile problems on HPUX, but I have
a problem where when I run the tests, the test shows that the tables default to MyISAM!
for bugs 8033, 8065, 8535, 8582
This particular changeset contains style changes per code review suggestions
and does not contain bug fixes in of itself.
ha_federated.h:
standardised code style to conform to internals.texi
ha_federated.cc:
more code standardisation to conform to internals.texi.
- casts
- declarations
- comments
- 80 char width
also, append using string1.append(string2) and not
string1.append(string2.c_ptr_quick())
This changeset/patch is on top of changesets 1.1814 and 1.1846
(for bugs 8033 and 8065) and now fixes bug 8535.
These changes have been built and tested successfully on build.mysql.com
handler.cc:
Added hooks for federated_db_init() and federated_db_end(),
as done with ha_archive_db does, per suggestion by Ingo in
code review of patch 1.1846.
ha_federated.h:
declaration of federated_db_init() and federated_db_end()
ha_federated.cc:
- Fixed some indentation problems from indent-ex (mainly to do with
cases where "variablename += value"
- Added federated_db_init() and federated_db_end(), as done with
archive, which also handler more elegantly one of the memory leaks
from bug 8033 where the federated_mutex was not freed
- Removed extrenous debug messages in parse_url()
- Fixed bug 8535, caused by NULL being quoted in write_row. This used to
work (incorrectly) but a recent change was made in the server that
exposed this
This patch contains all that my previous patch (1.1814) contained, with the addition of using cli_fetch_lengths for
handling binary data (Bar noted this on the review of 1.1814, Guilhem suggested using cli_fetch_lenghts by
making available via removal of static in method definition and declaration in mysql.h, but
Konstantin had some reservations, but he said to commit the patch using this anyway,
and I suppose this can be discussed. I abandoned 1.1814 because Monty made a couple
fixes to my code as well as formatting changes, and I thought it would just be easier
to hand-edit my changes into a fresh clone and then make a patch.
The reason for using cli_fetch_lengths is so that I can correctly get the length of
the field I am setting into the field. I was previously using 'strlen' but Bar pointed out this
won't correctly get the length of binary data and is also less effecient. Upon testing,
it was in fact verified that binary data in a blob table was being inserted correctly,
but not being retrieved correctly, all due to not having the correct value for the
field:
(*field)->store(row[x], strlen(row[x]), &my_charset_bin);
was changed to:
(*field)->store(row[x], lengths[x], &my_charset_bin);
lengths being a unsigned long pointer to the values of the field lengths from a
MYSQL_ROW.
Since the server doesn't have the function "mysql_fetch_lengths" available, I tried
to use "result->lengths", but this isn't set, so I finally successfully used
cli_fetch_lenghts, which does give the correct lengths, and now the binary data gets
retrieved correctly.
I've also run the code through indent-ex and am using Brian's vimrc to ensure correct formatting!
This code passes the entire test suite, without any errors or warning on both my
workstation and build.mysql.com
-New tests and results
logging_ok:
Logging to logging@openlogging.org accepted
ha_federated.h:
removed quote_data and type_quote (now in the Field class)
ha_federated.cc:
moved quote_data and type_quote to field class
field.h:
new methods quote_data and needs_quotes declared
field.cc:
new field class methods quote_data and needs_quotes (per Monty's request)
federated.test:
more tests, joins, index tests
have_federated_db.require:
new name of federated system var
federated.result:
new test results for federated handler
have_federated_db.inc:
changed name of variable in test due to change in vars
sql_analyse.cc:
over-ridden append_escaped to take (String *, char *, uint) per requirements of 'create_where_from_key' method in federated handler.
mysql_priv.h:
define over-ridden append_escaped to take arguments from 'create_where_from_key' method in federated handler
ha_federated.cc:
implemented "create_where_from_key" to deal properly with two-byte prefix and multi keys. Initial testing shows it works, but I still need to move quoting to field class and also look at changes per Segei's suggestions.
