Put back a 50 millisecond sleep in too high concurrency situations which I removed in the previous push; count also such sleeping threads to the InnoDB queue in SHOW INNODB STATUS
innobase/srv/srv0srv.c:
Put back a 50 millisecond sleep in too high concurrency situations which I removed in the previous push; count also such sleeping threads to the InnoDB queue in SHOW INNODB STATUS
Fix a benign bug introduced in 4.0.14: InnoDB could complain 'Error: trying to declare trx to enter InnoDB' if several threads tried to init the auto-inc counter for the same table at the same time; in theory, the bug could even lead to a hang of the server, but that shuld be extremely improbable
innobase/srv/srv0srv.c:
Fix a benign bug introduced in 4.0.14: InnoDB could complain 'Error: trying to declare trx to enter InnoDB' if several threads tried to init the auto-inc counter for the same table at the same time; in theory, the bug could even lead to a hang of the server, but that shuld be extremely improbable
Allow also O_DIRECT as innodb_flush_method; it only affects writing to data files
innobase/os/os0file.c:
Allow also O_DIRECT as innodb_flush_method; it only affects writing to data files
innobase/include/os0file.h:
Allow also O_DIRECT as innodb_flush_method; it only affects writing to data files
innobase/include/srv0srv.h:
Allow also O_DIRECT as innodb_flush_method; it only affects writing to data files
innobase/srv/srv0start.c:
Allow also O_DIRECT as innodb_flush_method; it only affects writing to data files
sql/sql_repl.cc:
In my previous change I removed 'no_rotate' from MYSQL_LOG, so this made
LOG_INFO_PURGE_NO_ROTATE useless (an error code which was never returned),
so I remove it of the 'switch'.
sql/log.cc:
Member no_rotate of MYSQL_LOG is useless; this shows it for sure:
[guilhem@gbichot2 mysql-4.0]$ bk -r grep no_rotate
sql/log.cc 1.83 no_rotate(0), need_start_event(1)
sql/log.cc 1.75 if (no_rotate)
sql/log.cc 1.89 if (!no_rotate)
sql/sql_class.h 1.119 bool no_rotate;
i.e. no_rotate is 0 all the time. So we don't need it.
Biggest part of the patch is indentation change.
sql/sql_class.h:
suppress no_rotate
sql/log.cc:
a comment for the future.
sql/mysqld.cc:
if (p) is not needed: fn_ext() always returns a valid pointed; the way
to test if an extension was found is if (*p), not if (p).
But here even if there's no extension, we still want to truncate opt_name
to FN_REFLEN, so we always execute the code.
By design, the test was always 'false' (because we strip the extension before
testing if there's an extension) so log->set_no_rotate never executed.
sql/sql_class.h:
remove set_no_rotate as we don't need it.
port number can be different from 3306 when doing the release builds
with Do-compile, therefore it has to be replaced with the correct
value during the test run using the "--replace_result" function.
mysql-test/r/rpl_error_ignored_table.result:
- replaced hard-coded master port number with MASTER_PORT variable, since
the port number is different when running the test suite during the
release builds with "Do-compile".
mysql-test/t/rpl_error_ignored_table.test:
- replaced hard-coded master port number with MASTER_PORT variable, since
the port number is different when running the test suite during the
release builds with "Do-compile".
ChangeSet was not sufficient to get the full ChangeLog between two
released versions (when using "--changelog=last") as merging between
trees also merges the BK tags. Now we explicitely search for the last
tagged ChangeSet that matches our major+minor version number, which
should match the last ChangeSet used for the previous release.
Build-tools/Bootstrap:
- fixed a bug: simply searching for the last tagged ChangeSet was not
sufficient to get the full ChangeLog between two released versions
(when using "--changelog=last") as merging between trees also merges
the BK tags. Now we explicitely search for the last tagged ChangeSet
that matches our major+minor version number, which should match the
last ChangeSet used for the previous release.
a safer way of initing the mutexes in MYSQL_LOG.
is_open() is now always thread-safe.
See each file for details.
sql/handler.cc:
is_open() with locks
sql/item_func.cc:
is_open() with locks
sql/log.cc:
No more 'inited'.
We now always use is_open() in a thread-safe manner.
This simplifies some functions (no more need to test is_open() twice).
sql/log_event.cc:
is_open() with locks
sql/mysqld.cc:
Init mutexes for the global MYSQL_LOG objects.
We care about no_rotate, because we can't do it in open() anymore (because
we don't have 'inited' anymore).
sql/repl_failsafe.cc:
is_open() with locks
sql/slave.cc:
init pthread objects (mutexes, conds) in the constructor of st_relay_log_info.
Some better locking in rotate_relay_log().
sql/sql_base.cc:
is_open() with locks
sql/sql_class.h:
Before, we inited LOCK_log in MYSQL_LOG::open(), so in other places of the code
when we were never 100% sure that it had been inited. For example, if the server
was running without --log-bin, ::open() was not called so the mutex was not
inited. We could detect it with !inited, but not safely as 'inited' was not
protected by any mutex.
So now:
we *always* init the LOCK_log mutex, even if the log is not used. We can't init
the mutex in MYSQL_LOG's constructor, because for global objects like
mysql_bin_log, mysql_log etc, the constructor is called before MY_INIT(), but
safe_mutex depends on MY_INIT(). So we have a new function MYSQL_LOG::init_pthread_objects
which we call in main(), after MY_INIT().
For the relay log, we call this function in the constructor of
st_relay_log_info, which is called before any function tries to
use the relay log (the relay log is always invoked as rli.relay_log).
So now we should be safe in all cases and we don't need 'inited'.
sql/sql_db.cc:
is_open() with locks
sql/sql_delete.cc:
is_open() with locks
sql/sql_insert.cc:
is_open() with locks
sql/sql_load.cc:
is_open() with locks
sql/sql_parse.cc:
is_open() with locks
sql/sql_rename.cc:
is_open() with locks
sql/sql_repl.cc:
is_open() with locks
sql/sql_table.cc:
is_open() with locks
sql/sql_update.cc:
is_open() with locks
"mysqlhotcopy fails to copy tables but does not indicate a failure"
("does not indicate a failure");
this is about "mysqlhotcopy fails to copy tables".
mysql-test/t/rpl_error_ignored_table-slave.opt:
added a missing newline
scripts/mysqlhotcopy.sh:
Fix for BUG#812.
The problem was that with many tables to copy (10000 in the bug's example),
the generated 'cp' command line was 1MB long, whereas (at least on my Linux) it
should not exceed 128 kB. Testing the 'cp' in a shell terminal gives
"arguments list too long".
So we issue several small (100 kB) 'cp' command lines instead of a big one.
Of course, this will still fail on systems where the limit is below 100 kB.
We now have safe_system() which cuts the command line in pieces,
and calls safe_simple_system() (execution) for each piece.
override CC/CXX when using a special compiler)
- the RPM should now build on AMD64 and IA64, too
support-files/MySQL-shared-compat.spec.sh:
- use %_libdir macro instead of hardcoded value (required on AMD64,
where %_libdir is /usr/lib64 instead)
support-files/mysql.spec.sh:
- removed the GIF Icon (file was not included in the sources anyway)
- removed unused variable %shared_lib_version
- do not run automake before building the standard binary
(should not be necessary)
- add server suffix '-standard' to standard binary (to be in line
with the binary tarball distributions)
- Use more RPM macros (_exec_prefix, _sbindir, _libdir, _sysconfdir,
_datadir, _includedir) throughout the spec file (using _libdir was
required to be used anyway to be able to build on AMD64).
- allow overriding CC and CXX (required when building with other compilers)
Removed double my_thread_end() which caused fatal error on windows if mysqld died on startup
myisam/mi_extra.c:
Fixed bug in ALTER TABLE ... DISABLE/ENABLE KEYS
mysql-test/r/alter_table.result:
Fixed bug in ALTER TABLE ... DISABLE/ENABLE KEYS
mysql-test/t/alter_table.test:
Test DISABLE/ENABLE KEY
sql/ha_myisam.cc:
Fixed bug in ALTER TABLE ... DISABLE/ENABLE KEYS
sql/mysqld.cc:
Removed double my_thread_end() which caused fatal error on windows if mysqld died on startup
sql/sql_table.cc:
Fixed bug in ALTER TABLE ... DISABLE/ENABLE KEYS
sql/table.cc:
Fixed bug in ALTER TABLE ... DISABLE/ENABLE KEYS
sql/table.h:
Fixed bug in ALTER TABLE ... DISABLE/ENABLE KEYS
Undo serg's patch that used my_print_defaults,
and instead simply do not define $opt{user}.
This allows the script to use option files properly.
scripts/mysqlhotcopy.sh:
Undo serg's patch that used my_print_defaults,
and instead simply do not define $opt{user}.
This allows the script to use option files properly.
BUG#797 "If query ignored on slave (replicate-ignore-table) the slave still checks if
the returned error (0) is the same as the one on the master, whereas it shouldn't
test this.
Plus a new test for BUG#797.
include/mysqld_error.h:
New error code (can be sent only to the slave thread, not to normal clients).
sql/log.cc:
removed comment from a previous debugging.
sql/log_event.cc:
ER_SLAVE_IGNORED_TABLE is an ignorable error.
sql/share/czech/errmsg.txt:
for BUG#797
sql/share/danish/errmsg.txt:
for BUG#797
sql/share/dutch/errmsg.txt:
for BUG#797
sql/share/english/errmsg.txt:
for BUG#797
sql/share/estonian/errmsg.txt:
for BUG#797
sql/share/french/errmsg.txt:
for BUG#797
sql/share/german/errmsg.txt:
for BUG#797
sql/share/greek/errmsg.txt:
for BUG#797
sql/share/hungarian/errmsg.txt:
for BUG#797
sql/share/italian/errmsg.txt:
for BUG#797
sql/share/japanese/errmsg.txt:
for BUG#797
sql/share/korean/errmsg.txt:
for BUG#797
sql/share/norwegian-ny/errmsg.txt:
for BUG#797
sql/share/norwegian/errmsg.txt:
for BUG#797
sql/share/polish/errmsg.txt:
for BUG#797
sql/share/portuguese/errmsg.txt:
for BUG#797
sql/share/romanian/errmsg.txt:
for BUG#797
sql/share/russian/errmsg.txt:
for BUG#797
sql/share/slovak/errmsg.txt:
for BUG#797
sql/share/spanish/errmsg.txt:
for BUG#797
sql/share/swedish/errmsg.txt:
for BUG#797
sql/share/ukrainian/errmsg.txt:
for BUG#797
sql/sql_parse.cc:
when in sql_parse in the slave thread we ignore the query because of replicate-do
and replicate-ignore options, we return a specific error to the slave thread,
so that it knows that the query has been ignored (which is different from a
successful query).
A small cleanup (test was done twice).
A cosmetic change: set trx id to zero at creation so that SHOW INNODB STATUS does not print a random value for the id of a transaction object for which the transaction has never been started yet (for example, running SHOW INNODB STATUS as the first command from a mysql prompt printed a random id for the trx object associated with the session itself running the SHOW INNODB STATUS command)
innobase/trx/trx0trx.c:
A cosmetic change: set trx id to zero at creation so that SHOW INNODB STATUS does not print a random value for the id of a transaction object for which the transaction has never been started yet (for example, running SHOW INNODB STATUS as the first command from a mysql prompt printed a random id for the trx object associated with the session itself running the SHOW INNODB STATUS command)
Fix bug: if a DELETE failed in a FOREIGN KEY error and it had already deleted some rows, then MySQL did not roll back the failed SQL statement, and also wrote it to the binlog
sql/sql_delete.cc:
Fix bug: if a DELETE failed in a FOREIGN KEY error and it had already deleted some rows, then MySQL did not roll back the failed SQL statement, and also wrote it to the binlog
FLUSH LOGS now rotates relay logs,
and a new variable max_relay_log_size.
Plus a very small bit of code cleaning.
libmysqld/lib_sql.cc:
open_log has no default arguments anymore.
mysql-test/r/rpl_flush_log_loop.result:
result update now that FLUSH LOGS rotates relay logs.
mysql-test/r/rpl_log.result:
result update now that FLUSH LOGS rotates relay logs.
mysql-test/r/rpl_rotate_logs.result:
result update now that max_binlog_size is 4096.
mysql-test/t/rpl_rotate_logs-master.opt:
now max_binlog_size must be a multiple of 4096 (see change in mysqld.cc)
sql/log.cc:
Got rid of default arguments of various MYSQL_LOG methods (the default arguments
made code reading uneasy).
Set max_size in ::init().
New function set_max_size() to set max_size of a MYSQL_LOG on-the-fly.
More DBUG info.
sql/mysql_priv.h:
no defaults in open_log().
New variables max_relay_log_size.
sql/mysqld.cc:
New variable and option max_relay_log_size.
max_binlog_size and max_relay_log_size are multiples of IO_SIZE.
No more default arguments for log functions.
sql/set_var.cc:
New variable max_relay_log_size.
If it is 0, then max_binlog_size will apply to relay logs.
When one of these variables is changed, fix_max_%log_size is called
to update max_size of the binary and/or relay logs.
sql/slave.cc:
New function rotate_relay_log().
sql/slave.h:
New function rotate_relay_log().
sql/sql_class.h:
New member max_size of MYSQL_LOG (for automatic rotation).
New method set_max_size() for setting on-the-fly.
sql/sql_parse.cc:
Flush the relay log in FLUSH LOGS.
Fix performance bug: if there were thousands of tables, DROP TABLE could take as much as a second CPU time
innobase/row/row0mysql.c:
Fix performance bug: if there were thousands of tables, DROP TABLE could take as much as a second CPU time
In background loop run purge to completion before doing other background operations: it does not make sense to flush buffer pool pages if they are soon modified again by purge
trx0purge.c:
Increase purge_sys->n_pages_handled for every undo log we purge, even if that log would be only a hundred bytes: that way we get the purge batches of 20 pages to set a fresh purge view (limit) more often, and we can reduce the number of old row versions purge has to look at when it decides if it can remove some delete-marked index record
innobase/trx/trx0purge.c:
Increase purge_sys->n_pages_handled for every undo log we purge, even if that log would be only a hundred bytes: that way we get the purge batches of 20 pages to set a fresh purge view (limit) more often, and we can reduce the number of old row versions purge has to look at when it decides if it can remove some delete-marked index record
innobase/srv/srv0srv.c:
In background loop run purge to completion before doing other background operations: it does not make sense to flush buffer pool pages if they are soon modified again by purge
mysql-test/r/handler.result:
test case for the bug #787: HANDLER without INDEX doesn't work with deleted rows
mysql-test/t/handler.test:
test case for the bug #787: HANDLER without INDEX doesn't work with deleted rows
sql/sql_handler.cc:
fix for the bug #787: HANDLER without INDEX doesn't work with deleted rows
