MDEV-7106: Sporadic test failure in multi_source.gtid
MDEV-7153: Yet another sporadic failure of multi_source.gtid in buildbot
This patch fixes three races in the multi_source.gtid test case that could
cause sporadic failures:
1. Do not put SHOW ALL SLAVES STATUS in the output, the output is not stable.
2. Ensure that slave1 has replicated as far as expected, before stopping its
connection to master1 (otherwise the following wait will time out due to rows
not replicated from master1).
3. Ensure that slave2 has replicated far enough before connecting slave1 to it
(otherwise we get an error during connect that slave1 is ahead of slave2).
Problem is on test it tried to verify that no files were left
on test database.
Fix: There's no need to list other file types, it can only
list *.par files
I saw two test failures in rpl.rpl_gtid_crash where we get this in the error
log:
141123 12:47:54 [Note] InnoDB: Restoring possible half-written data pages
141123 12:47:54 [Note] InnoDB: from the doublewrite buffer...
InnoDB: Warning: database page corruption or a failed
InnoDB: file read of space 6 page 3.
InnoDB: Trying to recover it from the doublewrite buffer.
141123 12:47:54 [Note] InnoDB: Recovered the page from the doublewrite buffer.
This test case deliberately crashes the server, and if this crash happens
right in the middle of writing a buffer pool page to disk, it is not
unexpected that we can get a half-written page. The page is recovered
correctly from the doublewrite buffer.
So this patch adds a suppression for this warning in the error log for this
test case.
When a master slave restarts, it logs a special restart format description
event in its binlog. When the slave sees this event, it knows it needs to roll
back any active partial transaction, in case the master crashed previously in
the middle of writing such transaction to its binlog.
However, there was a bug where this rollback did not reset rgi->pending_gtid.
This caused the @@gtid_slave_pos to be updated incorrectly with the GTID of
the partial transaction that was rolled back.
Fix this by always clearing rgi->pending_gtid in cleanup_context(), hopefully
preventing similar bugs from turning up in other special cases where a
transaction is rolled back during replication.
Thanks to Pavel Ivanov for tracking down the issue and providing a test case.
after Operating system error number 36 in a file operation.
Analysis: os_file_get_status did not handle error ENAMETOOLONG
correctly.
Fix: Add correct handling for error ENAMETOOLONG. Note that on InnoDB
case the error is not passed all the way up to server. That would
be bigger rewamp.
innodb.innodb_stats_drop_locked fail and
innodb.innodb_stats_fetch_nonexistent fails in buildbot on Windows
Analysis: Problem is that innodb_stats_create_on_corrupted
test renames mysql.innodb.index_stats and all the rest
are dependend on this table.
Fix: After rename back to original, restart mysqld to
make sure that table is correct.
- Drop all tables in explain_json.test
- Tabular form should print ref='' when type='fulltext' (another peculiarity
of the traditional EXPLAIN format)
- String_list::append_str should allocate memory for \0, too
- Some temporary code for EXPLAIN JSON and join buffering.
* adjust viossl.c to take account the new code
(SSL_get_error is used now, cannot simply remap it)
* remove unnecessary version check
* update the test to 10.0
innodb_stats_sample_pages
Analysis: If you set the number of analyzed pages
to very low number compared to actual pages on
that table/index it randomly pics those pages
(default 8 pages), this leads to fact that query
after analyze table returns different results. If
the index tree is small, smaller than 10 *
n_sample_pages + total_external_size, then the
estimate is ok. For bigger index trees it is
common that we do not see any borders between
key values in the few pages we pick. But still
there may be n_sample_pages different key values,
or even more. And it just tries to
approximate to n_sample_pages (8).
Fix: (1) Introduced new dynamic configuration variable
innodb_stats_sample_traditional that retains
the current design. Default false.
(2) If traditional sample is not used we use
n_sample_pages = max(min(srv_stats_sample_pages,
index->stat_index_size),
log2(index->stat_index_size)*
srv_stats_sample_pages);
(3) Introduced new dynamic configuration variable
stat_modified_counter (default = 0) if set
sets lower bound for row updates when statistics is re-estimated.
If user has provided upper bound for how many rows needs to be updated
before we calculate new statistics we use minimum of provided value
and 1/16 of table every 16th round. If no upper bound is provided
(srv_stats_modified_counter = 0, default) then calculate new statistics
if 1 / 16 of table has been modified
since the last time a statistics batch was run.
We calculate statistics at most every 16th round, since we may have
a counter table which is very small and updated very often.
@param t table
@return true if the table has changed too much and stats need to be
recalculated
*/
#define DICT_TABLE_CHANGED_TOO_MUCH(t) \
((ib_int64_t) (t)->stat_modified_counter > (srv_stats_modified_counter ? \
ut_min(srv_stats_modified_counter, (16 + (t)->stat_n_rows / 16)) : \
16 + (t)->stat_n_rows / 16))
followup:
* explicitly disable SSLv2 and SSLv3, keep other protocols enabled
* fix a compiler warning
* rename the test and combinations to avoid confusion
vio/viossl.c:
fix a compiler warning
When the optimizer considers an option to use Loose Scan, it should
still consider UNIQUE keys (Previously, MDEV-4120 disabled loose scan
for all kinds of unique indexes. That was wrong)
However, we should not use Loose Scan when trying to satisfy
"SELECT DISTINCT col1, col2, .. colN"
when using an index defined as UNIQU(col1, col2, ... colN).
Problem is that page compressed tables currently require atomic_blobs and
that feature is not availabe currently for row_format=redundant.
Fix: Disallow page compressed create option if table row_format=redundant.
The problem is that the binlog position is updated before
Executed_log_entries and Slave_SQL_State. So, it's possible to hit
the moment when MASTER_POS_WAIT (and hence sync_with_master) already
returned success, but Slave_SQL_State and Executed_log_entries were not
modified yet.
Fixing it by adding a wait on the expected Executed_log_entries value.
in mysql_upgrade: do FLUSH PRIVILEGES at the end, not together with
mysql_fix_privilege_tables
mysql-test/t/mysql_upgrade-6984.opt:
use a dummy second option to force server restart after the test
use the same restriction for character_set_client on the command line
and from SQL.
Also: remove strange hack from thd_init_client_charset() that contradicted
the manual (collation_connection and character_set_result were not always set)
ALTER TABLE: don't fill default values per row, do it once.
And do it in two places - for copy_data_between_tables() and for online ALTER.
Also, run function_defaults test both for MyISAM and for InnoDB.
change SSL methods to be SSLv23 (according to openssl manpage:
"A TLS/SSL connection established with these methods may understand
the SSLv2, SSLv3, TLSv1, TLSv1.1 and TLSv1.2 protocols") from
TLSv1 methods, that go back to the initial SSL implementation
in MySQL in 2001.
OpenSSL default ciphers are different if TLSv1.2 is enabled,
so tests need to take this into account.
MDEV-6789 segfault in Item_func_from_unixtime::get_date on updating table with virtual columns
* prohibit VALUES in partitioning expression
* prohibit user and system variables in virtual column expressions
* fix Item_func_date_format to cache locale (for %M/%W to return the same as MONTHNAME/DAYNAME)
* fix Item_func_from_unixtime to cache time_zone directly, not THD (and not to crash)
* added tests for other incorrectly allowed (in vcols) functions to see that they don't crash
A "field" could be either an Item_field or
(if loading into a view) an Item_direct_ref that references Item_field.
Also: when iterating fields, use fields of the TABLE_LIST (table or view),
not fields of a TABLE (actual underlying table - might have more columns).
MDEV-6971 Bad results with joins comparing TIME and DOUBLE/DECIMAL columns
Disallow using indexes on non-temporal columns to optimize
ref access, range access and table elimination when the counterpart's
cmp_type is TIME_RESULT, e.g.:
SELECT * FROM t1 WHERE indexed_int_column=time_expression;
Only index on a temporal column can be used to optimize temporal comparison
operations.
The test case rpl.rpl_parallel_temptable deliberately crashes the master
server as part of the testing. This makes it unsuitable for Valgrind
testing. So make sure that it will be skipped when testing with Valgrind.
Merge Facebook commit cd063ab930
authored by Peng Tian from https://github.com/facebook/mysql-5.6
Introduced a new configuration variable innodb_fatal_semaphore_wait_threshold,
it makes the fatal semaphore timeout configurable. Modified original commit
so that no MariaDB server files are changed, instead introduced a new
InnoDB/XtraDB configuration variable.
Its default/min/max vlaues are 600/1/2^32-1 in seconds (it was hardcoded
as 600, now its default value is 600, so the default behavior of this diff
should be no change).
The real problem here was inconsistent handling of entry->commit_errno in
MYSQL_BIN_LOG::write_transaction_or_stmt(). Some return paths were setting it
to the value of errno, some where not. And the setting was redundant anyway,
as it is set consistently by the caller.
Fix by consistently setting it in the caller, and not in each return path in
the function.
The test failure happened because a DBUG_EXECUTE_IF() used in the test case
set an entry->commit_errno that was immediately overwritten in the caller with
whatever happened to be the value of errno. This could lead to different error
message in the .result file.
reset default fields not for every modified row, but only once,
at the beginning, as the set of modified fields doesn't change.
exception: INSERT ... ON DUPLICATE KEY UPDATE - the set of fields
does change per row and in that case we reset default fields per row.
This bug was seen when parallel replication experienced a deadlock between
transactions T1 and T2, where T2 has reached the commit phase and is waiting
for T1 to commit first. In this case, the deadlock is broken by sending a kill
to T2; that kill error is then later detected and converted to a deadlock
error, which causes T2 to be rolled back and retried.
The problem was that the kill caused ha_commit_trans() to errorneously call
wakeup_subsequent_commits() on T3, signalling it to abort because T2 failed
during commit. This is incorrect, because the error in T2 is only a temporary
error, which will be resolved by normal transaction retry. We should not
signal error to the next transaction until we have executed the code that
handles such temporary errors.
So this patch just removes the calls to wakeup_subsequent_commits() from
ha_commit_trans(). They are incorrect in this case, and they are not needed in
general, as wakeup_subsequent_commits() must in any case be called in
finish_event_group() to wakeup any transactions that may have started to wait
after ha_commit_trans(). And normally, wakeup will in fact have happened
earlier, either from the binlog group commit code, or (in case of no
binlogging) after the fast part of InnoDB/XtraDB group commit.
The symptom of this bug was that replication would break on some transaction
with "Commit failed due to failure of an earlier commit on which this one
depends", but with no such failure of an earlier commit visible anywhere.
The retry of an event group in parallel replication set the wrong value for
the end log position of the event that was retried
(qev->future_event_relay_log_pos). It was too large by the size of the event,
so it pointed into the middle of the following event.
If the retry happened in the very last event of the event group, _and_ the SQL
thread was stopped just after successfully retrying that event, then the SQL
threads's relay log position would be left incorrect. Restarting the SQL
thread could then try to read events from a garbage offset in the relay log,
usually leading to an error about not being able to read the event.
The code in binlog group commit around wait_for_commit that controls commit
order, did the wakeup of subsequent commits early, as soon as a following
transaction is put into the group commit queue, but before any such commit has
actually taken place. This causes problems with too early wakeup of
transactions that need to wait for prior to commit, but do not take part in
the binlog group commit for one reason or the other.
This patch solves the problem, by moving the wakeup to happen only after the
binlog group commit is completed.
This requires a new solution to ensure that transactions that arrive later
than the leader are still able to participate in group commit. This patch
introduces a flag wait_for_commit::commit_started. When this is set, a waiter
can queue up itself in the group commit queue.
This way, effectively the wait_for_prior_commit() is skipped only for
transactions that participate in group commit, so that skipping the wait is
safe. Other transactions still wait as needed for correctness.
setting of innodb_io_capacity_max
(a) Changed the behaviour so that if you set innodb_io_capacity to a
value > innodb_io_capacity_max that the value is accepted AND
that innodb_io_capacity_max = innodb_io_capacity * 2.
(b) If someone wants to reduce innodb_io_capacity_max and
reduce it below innodb_io_capacity then innodb_io_capacity
should be reduced to the same level as innodb_io_capacity_max.
In both cases give a warning to user.
main.information_schema: added a condition to the query to exclude perfschema tables
main.information_schema_all_engines: added a call to the include file to check for the presence of perfschema
When parsing a field declaration, grab type information from LEX before it's overwritten
by further rules. Pass type information through the parser stack to the rule that needs it.
The test complains that the server failed to disappear upon shutdown /
wait_for_disconnect. Trying to solve the probably race condition by
adding a wait before restart.
The test case had a classic mistake: SET DEBUG_SYNC='now SIGNAL xxx'
followed immediately by SET DEBUG_SYNC='RESET'. This makes it
possible for the waiter to miss the signal, if it does not manage
to wake up prior to the RESET.
The test cases had some --replace_result $USER USER. The problem is that the
value of $USER can be anything, depending on the name of the unix account that
runs the test suite. So random parts of the result can be errorneously
replaced, causing test failures.
Fix by making the replacements more specific, so they will match only the
intended stuff regardless of the value of $USER.
Merged Facebook commit 617aef9f911d825e9053f3d611d0389e02031225
authored by Inaam Rana to InnoDB storage engine (not XtraDB)
from https://github.com/facebook/mysql-5.6
WL#7047 - Optimize buffer pool list scans and related batch processing
Reduce excessive scanning of pages when doing flush list batches. The
fix is to introduce the concept of "Hazard Pointer", this reduces the
time complexity of the scan from O(n*n) to O.
The concept of hazard pointer is reversed in this work. Academically
hazard pointer is a pointer that the thread working on it will declar
such and as long as that thread is not done no other thread is allowe
do anything with it.
In this WL we declare the pointer as a hazard pointer and then if any
thread attempts to work on it, it is allowed to do so but it has to a
the hazard pointer to the next valid value. We use hazard pointer sol
reverse traversal of lists within a buffer pool instance.
Add an event to control the background flush thread. The background f
thread wait has been converted to an os event timed wait so that it c
signalled by threads that want to kick start a background flush when
buffer pool is running low on free/dirty pages.
Merge Facebook commit 154c579b828a60722a7d9477fc61868c07453d08
and e8f0052f9b112dc786bf9b957ed5b16a5749f7fd authored
by Steaphan Greene from https://github.com/facebook/mysql-5.6
Optimize prefix index queries to skip cluster index lookup when possible.
Currently InnoDB will always fetch the clustered index (primary key
index) for all prefix columns in an index, even when the value of a
particular record is smaller than the prefix length. This change
optimizes that case to use the record from the secondary index and avoid
the extra lookup.
Also adds two status vars that track how effective this is:
innodb_secondary_index_triggered_cluster_reads:
Times secondary index lookup triggered cluster lookup.
innodb_secondary_index_triggered_cluster_reads_avoided:
Times prefix optimization avoided triggering cluster lookup.
The bug is not very important per se, but it was helpful to move
Item_func_strcmp out of Item_bool_func2 (to Item_int_func),
for the purposes of "MDEV-4912 Add a plugin to field types (column types)".
The test runs a query in one thread, then in another queries the processlist
and expects to find the first thread in the COM_SLEEP state. The problem is
that the thread signals completion to the client before changing to COM_SLEEP
state, so there is a window where the other thread can see the wrong state.
A previous attempt to fix this was ineffective. It set a DEBUG_SYNC to handle
proper waiting, but unfortunately that DEBUG_SYNC point ended up triggering
already at the end of SET DEBUG_SYNC=xxx, so the wait was ineffective.
Fix it properly now (hopefully) by ensuring that we wait for the DEBUG_SYNC
point to trigger at the end of the SELECT SLEEP(), not just at the end of
SET DEBUG_SYNC=xxx.
Merge Facebook commit 4f3e0343fd2ac3fc7311d0ec9739a8f668274f0d
authored by Steaphan Greene from https://github.com/facebook/mysql-5.6
Adds innodb_idle_flush_pct to enable tuning of the page flushing rate
when the system is relatively idle. We care about this, since doing
extra unnecessary flash writes shortens the lifespan of the flash.
Merge Facebook commit ca40b4417fd224a68de6636b58c92f133703fc68
authored by Steaphan Greene from https://github.com/facebook/mysql-5.6
Change the default value for innodb_log_compressed_pages to false
Logging these pages is a waste. We don't want this to be enabled.
One caution here: If the zlib version used by innodb is changed, but
the running version is still the previous version, and the running
version crashes, it is possible crash recovery could fail.
When crash recovery uses a zlib version at all different than the
version used by the crashed instance, it is possible that a redone
compression could fail, where the original did not, because the new
zlib version compresses the same data to a slightly larger size.
Because of the nature of compression, this is even possible when
upgrading to a version of zlib which actually peforms overall better
compression than the previous version.
If this happens, mysql will fail to recover, since a page split can
not be safely triggered during crash recovery.
So, either the exact zlib version must be controlled between builds,
or these rare recovery failures must be accepted. The cost of
logging these pages is quite high, so we consider this limitation to
be worthwhile.
This failure scenario can not happen if there was a clean shutdown.
This is only relevant to restarting crashed instances, or starting an
instance built via a hot backup too (XtraBackup).
New generation hard drives, SSDs and NVM devices support 4K
sector size. Supported sector size can be found using fstatvfs()
or GetDiskFreeSpace() functions.
(Backport to 5.3)
(Attempt #2)
- Don't attempt to use BKA for materialized derived tables. The
table is neither filled nor fully opened yet, so attempt to
call handler->multi_range_read_info() causes crash.
(Backport to 5.3)
(variant #2, with fixed coding style)
- Make Mrr_ordered_index_reader::resume_read() restore index position
only if it was saved before with Mrr_ordered_index_reader::interrupt_read().
- TABLE::create_key_part_by_field() should not set PART_KEY_FLAG in field->flags
= The reason is that it is used by hash join code which calls it to create a hash
table lookup structure. It doesn't create a real index.
= Another caller of the function is TABLE::add_tmp_key(). Made it to set the flag itself.
- The differences in join_cache.result could also be observed before this patch: one
could put "FLUSH TABLES" before the queries and get exactly the same difference.
Merged Facebook commit dd2d11be7aaf3be270e740fb95cbc4eacb52f4d7
authored by Rongrong Zhong from https://github.com/facebook/mysql-5.6
This fixes MySQL Bug #68220 innodb_rows_updated is misleading on slave
http://bugs.mysql.com/bug.php?id=68220
Added innodb_system_rows_read/inserted/updated/deleted counters
that are the equivalent of innodb_rows_* but that only account for
changes made to system databases (mysql, information_schame and
preformance_schema). These counters will be used on slaves to
differentiated the updates made on system databases from those made on
user databases.
innodb_rows_* status counters are not updated when innodb_system_rows_*
are updated.
dd2d11be7a
Merged Facebook commit ecff018632c6db49bad73d9233c3cdc9f41430e9
authored by Steaphan Greene from https://github.com/facebook/mysql-5.6
This change is to fix: http://bugs.mysql.com/62534
This makes innodb_max_dirty_pages_pct a double with min,default,max values
0.001, 75, 99.999.
This also makes innodb_max_dirty_pages_pct_lwm and adaptive_flushing_lwm
doubles, as these sysvars are inter-dependent.
Added more to the BUFFER POOL AND MEMORY section of SHOW INNODB STATUS:
Percent pages dirty: X.X
This is all n_dirty_pages / used_pages
Percent all pages dirty: X.X
This is all n_dirty_pages / all-pages
Max dirty pages percent: X.X
This is innodb_max_dirty_pages_pct
Also changed all of buf from 2 to 3 digits of precision (%.2f -> %.3f).
(Attempt #2)
- Don't attempt to use BKA for materialized derived tables. The
table is neither filled nor fully opened yet, so attempt to
call handler->multi_range_read_info() causes crash.
The method subselect_union_engine::no_rows() must take
into account the fact that now unit->fake_select_lex is
NULL for for select_union_direct objects.
