kill_galera.inc can no longer rely on wait_until_connected_again.inc.
This is because wait_until_connected_again now tries to make sure
that the server it is connected eventually transition to ready
state. Whereas some tests may need to kill galera while the server
is in a non-primary view.
Test galera_3nodes.galera_pc_weight started to fail because it
expects to use wait_until_connected_again while remaining in
non-primary view. Hopefully this is the only test which makes
this assumption, and fortunately those wait_until_connected_again
seem unnecessary, so this patch removes them.
wait_until_connected_again issues 'SHOW STATUS' query repeatedly
until mysqld replies without errors.
However, SHOW STATUS is treated specially by wsrep in that it is
allowed to proceed even if wsrep is not yet in ready state. As a
consequence, after returning from wait_until_connected_again,
wsrep may not be ready yet and subsequent queries may fail with
error "1047 WSREP has not yet prepared node for application use".
To avoid those errors, the patch includes wait_wsrep_ready.inc at
the end of the wait_until_connected_again.
Encountered illegal value '' when converting to DECIMAL
The issue was that EITS data was allocated but then not read for some reason (one being to avoid a deadlock),
then the optimizer was using these bzero'ed buffers as EITS statistics.
This should not be allowed, we should use statistcs for a table only when we have successfully loaded/read
the stats from the statistical tables.
This patch makes two changes:
* It replaces `--sleep 1` with appropriate wait_conditions, and
removes another `--sleep 1` which is no longer necessary, after
MDEV-14144 has been fixed.
* It moves the `RESET MASTER` from node_1 at the very end, when it
it sure that the other nodes have applied the the final cleanup
`DROP TABLE t1,t2.`
This solves the problem with the `DROP TABLE` not being replicated
to the asynchronous slaves, which would make the test fail with:
`Timeout in wait_condition.inc for SELECT COUNT(*) = 0 FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_NAME = 't1';`
Test galera_suspend_slave checks that after suspending a galera
node, like this:
```
--perl
my $pid_filename = $ENV{'NODE_2_PIDFILE'};
my $mysqld_pid = `cat $pid_filename`;
chomp($mysqld_pid);
system("kill -SIGSTOP $mysqld_pid");
exit(0);
EOF
```
the remaining one node cluster is no longer able to successfully
INSERT new rows:
```
--error ER_UNKNOWN_COM_ERROR,ER_LOCK_WAIT_TIMEOUT,ER_LOCK_DEADLOCK,ER_ERROR_DURING_COMMIT
INSERT INTO t1 VALUES (1);
```
On rare occasions when the system is overloaded, it appears that
suspending the process with ```system("kill -SIGSTOP $mysqld_pid")```
takes some time, enough for the subsequent INSERT to succeed. In which
case the test fails because it rightly expects the INSERT to fail.
To fix the problem, the patch makes sure that the cluster has shrinked
to one node, before trying to INSERT the new value.
For this case we have a view that is mergeable but we are not able to merge it in the
parent select because that would exceed the maximum tables allowed in the join list, so we
materialise this view
TABLE_LIST::dervied is NULL for such views, it is only set for views which have ALGORITHM=TEMPTABLE
Fixed by making sure TABLE_LIST::derived is set for views that could not be merged
MDEV--15609 engines/funcs.crash_manytables_number crashes with error 24
(too many open files)
MDEV-10286 Adjustment of table_open_cache according to system limits
does not work when open-files-limit option is provided
Fixed by adjusting tc_size downwards if there is not enough file
descriptors to use.
Other changes:
- Ensure that there is 30 (was 10) extra file descriptors for other usage
- Decrease TABLE_OPEN_CACHE_MIN to 200 as it's better to have a smaller
table cache than getting error 24
- Increase minimum of max_connections and table_open_cache from 1 to 10
as 1 is not usable for any real application, only for testing.
galera SST tests have a debug part, but we don't want to limit them
to fulltest2 builder. So, add support for test files that
have a debug part:
* add maybe_debug.inc and maybe_debug.combinations
* 'debug' combination is run when debug is available
* 'release' combination is run otherwise
* test wraps debug parts in if($with_debug) { ... }
* and creates ,debug.rdiff for debug results
* make galera.galera_sst_xtrabackup* not big
* auto-select between socat and nc, whatever available
* auto-skip xtrabackup tests if no xtrabackup or neither socat nor nc
fix galera.galera_sst_mysqldump test to work:
* must connect to 127.0.0.1, where mysqld is listening
* disable wsrep_sync_wait in wsrep_sst_mysqldump, otherwise
sst can deadlock
* allow 127.0.0.1 for bind_address and wsrep_sst_receive_address.
(it's useful in tests, or when two nodes are on the same box,
or when nodes are on different boxes, but the connection is
tunelled, or whatever. Don't judge user's setup). MDEV-14070
* don't wait for client connections to die when doing
mysqldump sst. they'll die in a due time, and if needed mysql
will wait on locks until they do. MDEV-14069
Also don't mark it big, to make sure it's sufficiently tested
It doesn't make sense to allow selects from I_S but disallow selects
that don't use any tables at all, because any (disallowed) select that
doesn't use tables can be made allowed by adding
"FROM I_S.COLLATIONS LIMIT 1" to the end.
And it break mysql-test rather badly, even check-testcase.test
fails on its first `SELECT '$tmp' = 'No such row'`
This reverts 9a89614857, c5dd2abf4c, and 33028f7c4b:
Refs: MW-245 - changed logic so that in non primary node it is possible to do SET + SHOW + SELECT from information and pfs schema, when dirty reads are not enabled - however, non table selects are not allowed (e.g. SELECT 1)
Refs MW-245 - logic was wrong in detecting if queries are allowed in non primary node. it allowed select with no table list to execute even if dirty reads was not specified
Refs: MW-245 - Adjust tests to account for the new behavior.
Conversion of a subquery to a semi-join is blocked when we have an
IN subquery predicate in the on_expr of an outer join. Currently this
scenario is handled but the cases when an IN subquery predicate is wrapped
inside a Item_in_optimizer item then this blocking is not done.
For the query having an IN subquery with no tables, we were converting the subquery with an expression between
the left part and the select list of the subquery . This can give incorrect results when we have a condition
in the subquery with a dual table (as this is treated as a no table).
The fix is that we don't do this conversion when we have conds in the subquery with a dual table.
fil_iterate(), fil_tablespace_iterate(): Replace os_file_read()
with os_file_read_no_error_handling().
os_file_read_func(), os_file_read_no_error_handling_func():
Do not retry partial reads. There used to be an infinite amount
of retries. Because InnoDB extends both data and log files upfront,
partial reads should be impossible during normal operation.
assert on UTF-8 columns
Problem:
=======
(1) Multi-byte character cases are not considered during prefix index
cluster optimization check. It leads to fetch of improper results during
read operation.
(2) Strict assert in row_sel_field_store_in_mysql_format_func and it asserts
for prefix index record to mysql conversion.
Solution:
========
(1) Consider the case of multi-byte character during prefix index
cluster optimization check.
(2) Relax the assert in row_sel_field_store_in_mysql_format_func to allow
prefix index record to mysql format conversion.
The patch is taken from
1eee538087
cmp_item_sort_string::store_value() did not cache the string returned
from item->val_str(), whose result can point to various private members
such as Item_char_typecast::tmp_value.
- cmp_item_sort_string::store_value() remembered the pointer returned
from item->val_str() poiting to tmp_value into cmp_item_string::value_res.
- Later, cmp_item_real::store_value() was called, which called
Item_str_func::val_real(), which called Item_char_typecast::val_str(&tmp)
using a local stack variable "String tmp". Item_char_typecast::tmp_value
was overwritten and become a link to "tmp":
tmp_value.Ptr freed its own buffer and set to point to the buffer
owned by "tmp".
- On return from Item_str_func::val_real(), "String tmp" was destructed,
but "tmp_value" still pointed to the buffer owned by "tmp",
So tmp_value.Ptr became invalid.
- Then cmp_item_sort_string() passed cmp_item_string::value_res to sortcmp().
At this point, value_res still pointed to an invalid value of
Item_char_typecast::tmp_value.
Fix:
changing cmp_item_sort_string::store_value() to force copying
to cmp_item_string::value if item->val_str(&value) returned
a different pointer (instead of &value).
To disallow equality propagation for DATETIME with non-zero YYYYMMDD part we were setting null_value to true.
This caused issues when we were calculating selectivity for a condition as this returned IMPOSSIBLE WHERE.
The issue is resolved by not setting null_value to true for DATETIME with non-zero YYYYMMDD.
Two changes were made to the test:
1) Suppress warning "Refusing exit for the last slave thread."
This warning was already suppressed, but on the wrong node.
2) The test occasionally fails because it expects that the
number of applier threads changes immediately after
changing the value of ```variable wsrep_slave_threads```.
Which is not true. This patch turns snippets like this:
```
SET GLOBAL wsrep_slave_threads = x;
SELECT COUNT(*) = x FROM INFORMATION_SCHEMA.PROCESSLIST WHERE USER = 'system user';
```
Into proper wait_conditions:
```
SET GLOBAL wsrep_slave_threads = x;
let $wait_condition = SELECT COUNT(*) = x FROM ...;
--source include/wait_condition.inc
```
Test galera.MW-366 is not deterministic and depends on timing assumptions.
The test occasionally fails after checking the number of 'system user'
processes in processlist after changing the value of variable global
wsrep_slave_threads, like this:
```
SET GLOBAL wsrep_slave_threads = x;
--sleep 0.5
SELECT COUNT(*) = x FROM INFORMATION_SCHEMA.PROCESSLIST WHERE USER = 'system user';
```
The problem is that the number of slave threads is internally adjusted
'asynchronously', and it may take some time to spawn/kill new threads,
especially in a heavily loaded system.
This patch removes the '--sleep 0.5' statements from the test and replaces
those with appropriate wait conditions, like this:
```
SET GLOBAL wsrep_slave_threads = x;
let $wait_condition = SLECT COUNT(*) = x FROM ...;
--source include/wait_condition.inc
```
This patch re-enables test galera.galera_var_max_ws_rows.
The test did not work because there were two distinct places where
the server was incrementing member THD::wsrep_affected_rows before
enforcing wsrep_max_ws_rows. Essentially, the test would fail because
every inserted row was counted twice.
The patch removes the extra code.
replicate_events_marked_for_skip=FILTER_ON_MASTER
[Note this is a cherry-pick from 10.2 branch.]
When events of a big transaction are binlogged offsetting over 2GB from
the beginning of the log the semisync master's dump thread
lost such events.
The events were skipped by the Dump thread that found their skipping
status erroneously.
The current fixes make sure the skipping status is computed correctly.
The test verifies them simulating the 2GB offset.
This test failed to work properly because the fixes it came
with were not merged from upstream.
The test would fail with a spurious ER_LOCK_DEADLOCK error
for a conflict that happened earlier in the test execution,
while wsrep is disabled.
The original fix was to set THD::wsrep_conflict_state only
if wsrep is enabled (see wsrep_thd_set_conflict_state() in
sql/wsrep_mysqld.cc)
InnoDB in Debian uses utf8mb4 as default character set since
version 10.0.20-2. This leads to major pain due to keys longer
than 767 bytes.
MariaDB 10.2 (and MySQL 5.7) introduced the setting
innodb_default_row_format that is DYNAMIC by default. These
versions also changed the default values of the parameters
innodb_large_prefix=ON and innodb_file_format=Barracuda.
This would allow longer column index prefixes to be created.
The original purpose of these parameters was to allow InnoDB
to be downgraded to MySQL 5.1, which is long out of support.
Every InnoDB version since MySQL 5.5 does support operation
with the relaxed limits.
We backport the parameter innodb_default_row_format to
MariaDB 10.1, but we will keep its default value at COMPACT.
This allows MariaDB 10.1 to be configured so that CREATE TABLE
is less likely to encounter a problem with the limitation:
loose_innodb_large_prefix=ON
loose_innodb_default_row_format=DYNAMIC
(Note that the setting innodb_large_prefix was deprecated in
MariaDB 10.2 and removed in MariaDB 10.3.)
The only observable difference in the behaviour with the default
settings should be that ROW_FORMAT=DYNAMIC tables can be created
both in the system tablespace and in .ibd files, no matter what
innodb_file_format has been assigned to. Unlike MariaDB 10.2,
we are not changing the default value of innodb_file_format,
so ROW_FORMAT=COMPRESSED tables cannot be created without
changing the parameter.
This performance regression was introduced in the MariaDB 10.1
file format incompatibility bug fix MDEV-11623 (MariaDB 10.1.21
and MariaDB 10.2.4) and partially fixed in MariaDB 10.1.25 in
MDEV-12610 without adding a regression test case.
On a normal startup (without crash recovery), InnoDB should not read
every .ibd data file, because this is slow. Like in MySQL, for now,
InnoDB will still open every data file (without reading), and it
will read every .ibd file for which an .isl file exists, or the
DATA DIRECTORY attribute has been specified for the table.
The test case shuts down InnoDB, moves data files, replaces them
with garbage, and then restarts InnoDB, expecting no messages to
be issued for the garbage files. (Some messages will for now be
issued for the table that uses the DATA DIRECTORY attribute.)
Finally, the test shuts down the server, restores the old data files,
and restarts again to drop the tables.
fil_open_single_table_tablespace(): Remove the condition on flags,
and only call fsp_flags_try_adjust() if validate==true
(reading the first page has been requested). The only caller with
validate==false is at server startup when we are processing all
records from SYS_TABLES. The flags passed to this function are
actually derived from SYS_TABLES.TYPE and SYS_TABLES.N_COLS,
and there never was any problem with SYS_TABLES in MariaDB 10.1.
The problem that MDEV-11623 was that incorrect tablespace flags
were computed and written to FSP_SPACE_FLAGS.
my_safe_alloca()/my_safe_afree() work as alloca() or malloc()/free()
depending on the memory size to allocate, that is, depending on
reclength here. They only work correctly if reclength doesn't
change in the middle.
lock_trx_release_locks(): Relax a debug assertion to allow
recovered TRX_STATE_COMMITTED_IN_MEMORY transactions.
trx_commit_in_memory(): Add DEBUG_SYNC instrumentation.
trx_undo_insert_cleanup(): Skip persistent changes if innodb_read_only
is set. This should only happen when a recovered committed transaction
would be cleaned up at shutdown.
PREVIOUS TO MYSQL 8.0
Description :
-------------
The mysqld--defaults-file test fails when the test suite is run from a
non-canonical path, which happens when the current working directory
when mysql-test-run.pl is started contains a symbolic link.
The problem is that this test case uses --replace-result with
$MYSQL_TEST_DIR. This variable is a potentially non-canonical path
based on the current working directory when mtr is started. However,
the path in the expected error message from mysqld contains a
canonical path. This means it does not contain $MYSQL_TEST_DIR if
mtr's working directory is not the canonical path of the working
directory.
Because other tests produce output that may contain non-canonical
paths, making $MYSQL_TEST_DIR always canonical is not a fix.
Fix :
-----
Introduced a new environment variable '$ABS_MYSQL_TEST_DIR' which will
contin the canonical path to the test directory and replaced
$MYSQL_TEST_DIR with the new variable in main.mysqld--defaults-file
test file.
This is a back-port of BUG#24579973.
Change-Id: I3b8df6f2d7ce2b04e188a896d76250cc1addbbc1
PageConverter::adjust_cluster_record(): Instead of writing
the invalid value DB_ROLL_PTR=0, write a value that indicates
a fresh insert, that is, prevents the DB_ROLL_PTR from being
dereferenced in any circumstances.
It can be argued that IMPORT TABLESPACE should actually
update the dict_index_t::trx_id to prevent older transactions
from accessing the table, similar to what I did on table
rebuild in MySQL 5.6.6 in
03f81a55f2
Suppress some messages that are emitted rarely
(when the FIL_PAGE_FILE_FLUSH_LSN of the first page of ibdata1
does not match the latest redo log checkpoint).
Counter for select numbering made stored with the statement (before was global)
So now it does have always accurate value which does not depend on
interruption of statement prepare by errors like lack of table in
a view definition.
The thd->lex->part_info should be kept intact during PS
execution. Or the second execution gets that modified part_info.
Let's modify ths->work_part_info instead.
Item_xml_str_func::fix_fields() used a local "String tmp" as a buffer
for args[1]->val_str(). "tmp" was freed at the end of fix_fields(),
while Items created during my_xpath_parse() still pointed to its fragments.
Adding a new member Item_xml_str_func::m_xpath_query and store the result
of args[1]->val_str() into it.
To disable debug instrumentation, save and restore the original value
of the variable DEBUG_DBUG. Assigning -d,... will enable the output of
a lot of unrelated DBUG messages to the server error log.
To disable debug instrumentation, save and restore the original value
of the variable DEBUG_DBUG. Assigning -d,... will enable the output of
a lot of unrelated DBUG messages to the server error log.
Since 736afe8680 the
mysql_system_tables_data.sql only installs the anonymous
user if skip_auth_anonymous IS NULL.
Its installation is prevented by setting this to 1.
remove HA_EXTRA_PREPARE_FOR_RENAME - neither OPTIMIZE nor REPAIR need it
(was introduced in b58e79566c when replacing remove_table_from_cache()
with wait_while_table_is_used() even though remove_table_from_cache()
did not have it).
May also fix: MDEV-14970 "MariaDB crashed with signal 11 and Aria table"
I am not able to reproduce a crash, however there was no protection in
print_keydup_error() if the storage engine reported the wrong key number.
This patch adds such a protection and should stop any further crashes
in this case.
Other things:
- Added extra protection in Aria to not set errkey to more than number of
keys. (Don't think this is cause of this crash, but better safe than
sorry)
- Extend test_if_equal_repl_errors() to handle different cases of
ER_DUP_ENTRY. This is just mainly precaution for the future.
MDEV-14957: JOIN::prepare gets unusable "conds" as argument
Do not touch merged derived (it is irreversible)
Fix first argument of in_optimizer for calls possible before fix_fields()
escape all charecters less or equal 0x1F (control symbols)
(shorter sequence are not used to make code simple, long encoding is always legal according to the rfc4627)
Problem:-
If we create table using myisam/aria then this crashes the server.
CREATE TABLE t1(a bit(1), b int auto_increment , index(a,b));
insert into t1 values(1,1);
Or this query
CREATE TABLE t1 (b BIT(1), pk INTEGER AUTO_INCREMENT PRIMARY KEY);
ALTER TABLE t1 ADD INDEX(b,pk);
INSERT INTO t1 VALUES (1,b'1');
ALTER TABLE t1 DROP PRIMARY KEY;
Reason:-
The reason for this is
1st- find_ref_key() finds what key an auto_increment field belongs to by
comparing key_part->offset and field->ptr. But BIT fields might have
zero length in the record, so a key might have many key parts with the
same offset. That is, comparing offsets cannot uniquely identify the
correct key part.
2nd- Since next_number_key_offset is zero it myisam/aria will think that
auto_increment is in first part of key.
3nd- myisam/aria will call retrieve_auto_key which will see first key_part
field as a bit field and call assert(0)
Solution:-
Many key parts might have the same offset, but BIT fields do not
support auto_increment. So, we can skip all key parts over BIT fields,
and then comparing offsets will be unambiguous.
The assertion failure was caused by an incorrectly set read_set for
functions in the ORDER BY clause in part of a union, when we are using
a mergeable view and the order by clause can be skipped (removed).
An order by clause can be skipped if it's part of one part of the UNION as
the result set is not meaningful when multiple SELECT queries are UNIONed. The
server is aware of this optimization and tries to remove the order by
clause before JOIN::prepare. The problem is that we need to throw an
error when the ORDER BY clause contains invalid columns. To do this, we
attempt resolving the ORDER BY expressions, then subsequently drop them
if resolution succeeded. However, ORDER BY resolution had the side
effect of adding the expressions to the all_fields list, which is used
to construct temporary tables to store the result. We may be ignoring
the ORDER BY statement, but the tmp table still tried to compute the
values for the expressions, even if the columns are never used.
The assertion only shows itself if the order by clause contains members
which were not previously in the select list, and are part of a
function.
There is an additional question as to why this only manifests when using
VIEWS and not when using a regular table. The difference lies with the
"reset" of the read_set for the temporary table during
SELECT_LEX::update_used_tables() in JOIN::optimize(). The changes
introduced in fdf789a7ea cleared the
read_set when a mergeable view is encountered in the TABLE_LIST
defintion.
Upon initial order_list resolution, the table's read_set is updated
correctly. JOIN::optimize() will only reset the read_set if it
encounters a VIEW. Since we no longer have ORDER BY clause in
JOIN::optimize() we never get to correctly update the read_set again.
Other relevant commit by Timour, which first introduced the order
resolution when we "can_skip_sort_order":
883af99e7d
Solution:
Don't add the resolved ORDER BY elements to all_fields. We only resolve
them to check if an error should be returned for the query. Ignore them
completely otherwise.
It has its limitations, e.g. it assumes that there's only one
gdb and only one valgrind process is running. And a hard-coded
one-second delay might be too short for slow machines.
Still, it's better than "doesn't work at all"
InnoDB limited the maximum number of bytes per character to 4.
But, the filename character set that was introduced in MySQL 5.1
uses up to 5 bytes per character.
To allow InnoDB tables to be created with wider characters, let
us split the mbminmaxlen fields into mbminlen, mbmaxlen, and increase
the limit to 7 bytes per character. This will increase the payload size
of dtype_t and dict_col_t by one bit. The storage size will be unchanged
(54 bits and 77 bits will use the same number of bytes as the
previous sizes 53 and 76 bits).
In this case we were using the optimization derived_with_keys but we could not create a key
because the length of the key was greater than the max allowed(MI_MAX_KEY_LENGTH).
To do the join we needed to create a hash join key instead, but in the explain output it
showed that we were still referring to derived keys which were created but not used.
In the function JOIN::shrink_join_buffers the iteration over joined
tables was organized in a wrong way. This could cause a crash if
the optimizer chose to materialize a semi-join that used join caches
for which the sizes must be adjusted.
optimizer_switch
For DATE and DATETIME columns defined as NOT NULL,
"date_notnull IS NULL" has to be modified to:
"date_notnull IS NULL OR date_notnull == 0"
if date_notnull is from an inner table of outer join);
"date_notnull == 0" - otherwise.
This must hold for such columns of mergeable views and derived
tables as well. So far the code did the above re-writing only
for columns of base tables and temporary tables.
The function trans_rollback_to_savepoint(), unlike trans_savepoint(),
did not allow xa_state=XA_ACTIVE, so an attempt to do ROLLBCK TO SAVEPOINT
inside an XA transaction incorrectly returned an error
"...command cannot be executed ... in the ACTIVE state...".
Partially merging a MySQL patch:
7fb5c47390311d9b1b5367f97cb8fedd4102dd05
This is WL#7193 (Decouple THD and st_transactions)...
The currently merged part includes these changes:
- Introducing st_xid_state::check_has_uncommitted_xa()
- Reusing it in both trans_rollback_to_savepoint() and trans_savepoint(),
so now both allow XA_ACTIVE.
The problem was in such scenario:
T1 - starts registering query and locked QC
T2 - starts disabling QC and wait for UNLOCK
T1 - unlock QC
T2 - disable QC and destroy signals without waiting for query unlock
T1 a) - not yet unlocked query in qc and crash on attempt to unlock because
QC signals are destroyed
b) if above was done before destruction, it execute end_of results first
time at exit on after try_lock which see QC disables and return TRUE.
But it do not reset query_cache_tls->first_query_block which lead to
second call of end_of_result when diagnostic arena has already
inappropriate status (not is_eof()).
Fix is:
1) wait for all queries unlocked before destroying them by locking and
unlocking
2) remove query_cache_tls->first_query_block if QC disabled
with joins, SQ, ORDER BY, semijoin=on
A bug in get_sort_by_table() could mislead the function
setup_semijoin_dups_elimination(). As a result the optimizer
could produce invalid execution plans for queries with ORDER BY
and subquery predicates that could be converted to semi-joins.
Remove non prepared (and so belonging to removed clauses FT functions) from the list.
in later version it will be fixed by building the list during preparation.
This bug happens when locking the same Aria "transactional" table
(page format) more then once with LOCK TABLES and inserting into one
of them with INSERT ... SELECT when the table is empty.
Fixed by ensuring we don't use fast bulk insert if table is opened
twice with LOCK TABLES (as this changes table->s->state)
Code changes:
- Added use_count to MARIA_USED_TABLES to be able to check if
table is opened twice for a statement/lock table
- Don't clear history or reset info->start_state if we
don't have versioning. One reason for the bug was
was that info->start_state was set to point to different
states for the two tables. If there is no versioning
info->start_state should always point to info->s->state.common.
Other things:
- Fixed also some typos that was noticed while scanning the code
- More DBUG_PRINT
The XtraDB option innodb_track_changed_pages causes
the function log_group_read_log_seg() to be invoked
even when recv_sys==NULL, leading to the SIGSEGV.
This regression was caused by
MDEV-11027 InnoDB log recovery is too noisy
innodb/buf_LRU_get_free_block
Add debug instrumentation to produce error message about
no free pages. Print error message only once and do not
enable innodb monitor.
xtradb/buf_LRU_get_free_block
Add debug instrumentation to produce error message about
no free pages. Print error message only once and do not
enable innodb monitor. Remove code that does not seem to
be used.
innodb-lru-force-no-free-page.test
New test case to force produce desired error message.
dict_foreign_find_index(): Ignore incompletely created indexes.
After a failed ADD UNIQUE INDEX, an incompletely created index
could be left behind until the next ALTER TABLE statement.
This bug affects both writing and reading encrypted redo log in
MariaDB 10.1, starting from version 10.1.3 which added support for
innodb_encrypt_log. That is, InnoDB crash recovery and Mariabackup
will sometimes fail when innodb_encrypt_log is used.
MariaDB 10.2 or Mariabackup 10.2 or later versions are not affected.
log_block_get_start_lsn(): Remove. This function would cause trouble if
a log segment that is being read is crossing a 32-bit boundary of the LSN,
because this function does not allow the most significant 32 bits of the
LSN to change.
log_blocks_crypt(), log_encrypt_before_write(), log_decrypt_after_read():
Add the parameter "lsn" for the start LSN of the block.
log_blocks_encrypt(): Remove (unused function).
debug_key_management
encrypt_and_grep
innodb_encryption
If real table count is different from what is expected by the test, it
just hangs on waiting to fulfill hardcoded number. And then exits with
**failed** after 10 minutes of wait: quite unfriendly and hard to
figure out what's going on.
* don't use Env module in tests, use $ENV{xxx} instead
* collateral changes:
** $file in the error message was unset
** $file in the other error message was unset too :)
** source file arguments are conventionally upper-cased
** abort the test (die) on error, don't just echo/exit
If translation table present when we materialize the derived table then
change it to point to the materialized table.
Added debug info to see really what happens with what derived.
Problem:- Gtid are not transferred in Galera Cluster.
Solution:- We need to transfer gtid in the case on either when cluster is
slave/master in async replication. In normal Gtid replication gtid are generated on
recieving node itself and it is always on sync with other nodes. Because galera keeps
node in sync , So all nodes get same no of event groups. So the issue arises when
say galera is slave in async replication.
A
| (Async replication)
D <-> E <-> F {Galera replication}
So what should happen is that all node should apply the master gtid but this does
node happen, becuase node E, F does not recieve gtid from D in write set , So what E(or F)
does is that it applies wsrep_gtid_domain_id, D server-id , E gtid next seq no. This
generated gtid does not always work when say A has different domain id.
So In this commit, on galera node when we see that this event is recieved from master
we simply write Gtid_Log_Event in write_set and send it to other nodes.
Sometimes, the test would fail with a result difference for
the READ UNCOMMITTED read, because the incremental backup
would finish before redo log was written for all the rows
that were inserted in the second batch.
To fix that, cause a redo log write by creating another
transaction. The transaction rollback (which internally does commit)
will be flushed to the redo log, and before that, all the preceding
changes will be flushed to the redo log as well.
row_log_table_apply_insert_low(), row_log_table_apply_update():
When reporting the error_key_num, only count the clustered index
if it corresponds to a key in the SQL layer.
The assertion failure was probably introduced by the (incomplete)
MySQL 5.6.28 bug fix
Bug #21364096 THE BOGUS DUPLICATE KEY ERROR IN ONLINE DDL
WITH INCORRECT KEY NAME
which we are improving.
Side note: the fix was incorrectly merged to MySQL 5.7.10;
incorrect key names will continue to be reported in MySQL 5.7.
In the function make_sortkey a tmp buffer was defined and in the absence of
param->tmp_buffer, tmp buffer used the sort_keys buffer. sort_keys buffer
has a length defined in sort_field->length, while param->tmp_buffer is
stored in param->rec_length. Make sure to use the appropriate length
based on which buffer we are using otherwise we'll overflow.
Also added a type cast to size_t during the calculation of the sort keys
buffer size to avoid an oveflow if the buffer size exceeds 32 bits.
find_type_or_exit() client helper did exit(1) on error, exit(1) moved to
clients.
mysql_read_default_options() did exit(1) on error, error is passed through and
handled now.
my_str_malloc_default() did exit(1) on error, replaced my_str_ allocator
functions with normal my_malloc()/my_realloc()/my_free().
sql_connect.cc did many exit(1) on hash initialisation failure. Removed error
check since my_hash_init() never fails.
my_malloc() did exit(1) on error. Replaced with abort().
my_load_defaults() did exit(1) on error, replaced with return 2.
my_load_defaults() still does exit(0) when invoked with --print-defaults.
Whenever we call merge_role_privileges on a role, we make use of
the role->counter variable to check if all it's children have had their
privileges merged. Only if all children have had their privileges merged,
do we update the privileges on parent. This is done to prevent extra work.
The same idea is employed during flush privileges. You only begin merging
from "leaf" roles. The recursive calls will merge their parents at some point.
A problem arises when we try to "re-merge" a parent. Take the following graph:
{noformat}
A (0) ---- C (2) ---- D (2) ---- USER
/ /
B (0) ----/ /
/
E (0) --------------/
{noformat}
In parentheses we have the "counter" value right before we start to iterate
through the roles hash and propagate values. It represents the number of roles
granted to the current role. The order in which we iterate through the roles
hash is alphabetical.
* First merge A, which leads to decreasing the counter for C to 1. Since C is
not 0, we don't proceed with merging into C.
* Second we merge B, which leads to decreasing the counter for C to 0. Now
we proceed with merging into C. This leads to reducing the counter for D to 1
as part of C merge process.
* Third as we iterate through the hash, we see that C has counter 0, thus we
start the merge process *again*. This leads to reducing the counter for
D to 0! We then attempt to merge D.
* Fourth we start merging E. When E sees D as it's parent (according to the code)
it attempts to reduce D's counter, which leads to overflow. Now D's counter is
a very large number, thus E's privileges are not forwarded to D yet.
To correct this behavior we must make sure to only start merging from initial
leaf nodes.
When granting a role to another role, DB privileges get propagated. If
the grantee had no previous DB privileges, an extra ACL_DB entry is created to
house those "indirectly received" privileges. If, afterwards, DB
privileges are granted to the grantee directly, we must make sure to not
create a duplicate ACL_DB entry.
row_undo_step(), trx_rollback_active(): Abort the rollback of a
recovered ordinary transaction if fast shutdown has been initiated.
trx_rollback_resurrected(): Convert an aborted-rollback transaction
into a fake XA PREPARE transaction, so that fast shutdown can proceed.
trx_rollback_resurrected(): If shutdown was initiated, fake all
remaining active transactions to XA PREPARE state, so that shutdown
can proceed. Also, make the parameter "all" an output that will be
assigned to FALSE in this case.
trx_rollback_or_clean_recovered(): Remove the shutdown check
(it was moved to trx_rollback_resurrected()).
trx_undo_free_prepared(): Relax assertions.
This was missing bug fix from MySQL wsrep i.e. Galera.
Problem was that if stored procedure declares a handler that
catches deadlock error, then the error may have been
cleared in method sp_rcontext::handle_sql_condition().
Use wsrep_conflict_state correctly to determine is the
error already sent to client.
Add test case for both this bug and MDEV-12837: WSREP: BF
lock wait long. Test requires both fixes to pass.
This is 10.1 version where no merge error exists.
wsrep_on_check
New check function. Galera can't be enabled
if innodb-lock-schedule-algorithm=VATS.
innobase_kill_query
In Galera async kill we could own lock mutex.
innobase_init
If Variance-Aware-Transaction-Sheduling Algorithm (VATS) is
used on Galera we refuse to start InnoDB.
Changed innodb-lock-schedule-algorithm as read-only parameter
as it was designed to be.
lock_rec_other_has_expl_req,
lock_rec_other_has_conflicting,
lock_rec_lock_slow
lock_table_other_has_incompatible
lock_rec_insert_check_and_lock
Change pointer to conflicting lock to normal pointer as this
pointer contents could be changed later.
Problem was that Binlog_checkpoint can happen at random times.
Fixed by not write binlog_checkpoint for the rpl_log test.
Other things:
- Removed not used variable "$keep_gtid_events"
- Added option for show_binlog_events to skip binlog_checkpoint
ROOT
DESCRIPTION
===========
If the .pid file is created at a world-writable location,
it can be compromised by replacing the server's pid with
another running server's (or some other non-mysql process)
PID causing abnormal behaviour.
ANALYSIS
========
In such a case, user should be warned that .pid file is
being created at a world-writable location.
FIX
===
A new function is_file_or_dir_world_writable() is defined
and it is called in create_pid_file() before .pid file
creation. If the location is world-writable, a relevant
warning is thrown.
NOTE
====
1. PID file is always created with permission bit 0664, so
for outside world its read-only.
2. Ignoring the case when permission is denied to get the
dir stats since the .pid file creation would fail anyway in
such a case.
For BIT field null_bit is not set to 0 even for a field defined as NOT NULL.
So now in the function TABLE::create_key_part_by_field, if the bit field is not nullable
then the null_bit is explicitly set to 0
