replication causing replication to fail.
In parallel replication, we run transactions from the master in parallel, but
force them to commit in the same order they did on the master. If we force T1
to commit before T2, but T2 holds eg. a row lock that is needed by T1, we get
a deadlock when T2 waits until T1 has committed.
Usually, we do not run T1 and T2 in parallel if there is a chance that they
can have conflicting locks like this, but there are certain edge cases where
it can occasionally happen (eg. MDEV-5914, MDEV-5941, MDEV-6020). The bug was
that this would cause replication to hang, eventually getting a lock timeout
and causing the slave to stop with error.
With this patch, InnoDB will report back to the upper layer whenever a
transactions T1 is about to do a lock wait on T2. If T1 and T2 are parallel
replication transactions, and T2 needs to commit later than T1, we can thus
detect the deadlock; we then kill T2, setting a flag that causes it to catch
the kill and convert it to a deadlock error; this error will then cause T2 to
roll back and release its locks (so that T1 can commit), and later T2 will be
re-tried and eventually also committed.
The kill happens asynchroneously in a slave background thread; this is
necessary, as the reporting from InnoDB about lock waits happen deep inside
the locking code, at a point where it is not possible to directly call
THD::awake() due to mutexes held.
Deadlock is assumed to be (very) rarely occuring, so this patch tries to
minimise the performance impact on the normal case where no deadlocks occur,
rather than optimise the handling of the occasional deadlock.
Also fix transaction retry due to deadlock when it happens after a transaction
already signalled to later transactions that it started to commit. In this
case we need to undo this signalling (and later redo it when we commit again
during retry), so following transactions will not start too early.
Also add a missing thd->send_kill_message() that got triggered during testing
(this corrects an incorrect fix for MySQL Bug#58933).
Analysis: In wsrep case there is two transactions possible trx and
conflicting lock owning transaction. Code was handling trx mutexin
correctly but not for c_lock->trx case.
Fixed by taking clock-trx mutex when needed and releasing trx
mutex if it is taken at lower levels.
Update InnoDB to 5.6.14
Apply MySQL-5.6 hack for MySQL Bug#16434374
Move Aria-only HA_RTREE_INDEX from my_base.h to maria_def.h (breaks an assert in InnoDB)
Fix InnoDB memory leak
IN LOCK_VALIDATE()
rb://917
approved by: Marko Makela
In lock_validate() the limit is used to release the kernel_mutex during
the validation, to obey the latching order.
If we do the limit++ then we are rechecking the same lock most times on
each iteration because limit is being incremented by one and
<space, page_no> will nearly always be > limit. If we set the limit
correctly to (space, page+1) then we are actually making progress
during the iteration.
print page dump
buf_page_print(): Remove the ut_ad(0) from the beginning. Add two flags
(enum buf_page_print_flags) that can be bitwise-ORed together:
BUF_PAGE_PRINT_NO_CRASH:
Do not crash debug builds at the end of buf_page_print().
BUF_PAGE_PRINT_NO_FULL:
Do not print the full page dump. This can be useful when adding
diagnostic printout to flushing or to the doublewrite buffer.
trx_sys_doublewrite_init_or_restore_page(): Replace exit(1) with ut_error,
so that we can get a core dump if this extraordinary condition happens.
rb:924 approved by Sunny Bains
This fix does not remove the underlying cause of the assertion
failure. It just works around the problem, allowing a corrupted
secondary index to be fixed by DROP INDEX and CREATE INDEX (or in the
worst case, by re-creating the table).
ibuf_delete(): If the record to be purged is the last one in the page
or it is not delete-marked, refuse to purge it. Instead, write an
error message to the error log and let a debug assertion fail.
ibuf_set_del_mark(): If the record to be delete-marked is not found,
display some more information in the error log and let a debug
assertion fail.
row_undo_mod_del_unmark_sec_and_undo_update(),
row_upd_sec_index_entry(): Let a debug assertion fail when the record
to be delete-marked is not found.
buf_page_print(): Add ut_ad(0) so that corruption will be more
prominent in stress testing with debug binaries. Add ut_ad(0) here and
there where corruption is noticed.
btr_corruption_report(): Display some data on page_is_comp() mismatch.
btr_assert_not_corrupted(): A wrapper around btr_corruption_report().
Assert that page_is_comp() agrees with the table flags.
rb:911 approved by Inaam Rana
Fix a deadlock in the initial patch. lock_validate() must not hold the
lock system mutex while s-latching a block, because some functions,
such as lock_rec_convert_impl_to_expl(), may be already holding an x-latch
on the block that lock_validate() is interested in while attempting to
acquire the lock system mutex.
This deadlock was not caught by UNIV_SYNC_DEBUG because of
buf_block_dbg_add_level(block, SYNC_NO_ORDER_CHECK).
lock_clust_rec_some_has_impl(), row_get_rec_trx_id(),
lock_rec_queue_validate(), lock_table_other_has_incompatible(),
lock_table_has_to_wait_in_queue(), lock_table_queue_validate():
Add const qualifiers.
row_get_trx_id_offset(): Add const qualifiers. Keep the parameter rec
only in UNIV_DEBUG builds. Inline the function.
lock_rec_validate_page(): Take the buffer block as a parameter, to
avoid a buf_page_get_gen() call in most cases.
lock_rec_validate_page_low(): A version of lock_rec_validate_page()
that assumes that the lock system mutexes are already being held.
lock_rec_get_next_on_page_const(): A const variant of
lock_rec_get_next_on_page().
lock_validate(): Do not release the lock system mutex while
buffer-fixing the block for the lock_rec_validate_page() call.
Releasing the mutex apparently caused the assertion failure.
rb:665 approved by Sunny Bains
It is filling the error log when testing the debug version of the server.
The printout only seems to be useful when debugging a crash, not when
testing an instrumented version of the server.
Fix compiler warning:
lock/lock0lock.c: In function 'lock_print_info_all_transactions':
lock/lock0lock.c:4299:10: error: variable 'page' set but not used [-Werror=unused-but-set-variable]
------------------------------------------------------------
revno: 3495
committer: Marko Mäkelä <marko.makela@oracle.com>
branch nick: 5.1-innodb
timestamp: Wed 2010-06-02 13:37:14 +0300
message:
Bug#53674: InnoDB: Error: unlock row could not find a 4 mode lock on the record
In semi-consistent read, only unlock freshly locked non-matching records.
lock_rec_lock_fast(): Return LOCK_REC_SUCCESS,
LOCK_REC_SUCCESS_CREATED, or LOCK_REC_FAIL instead of TRUE/FALSE.
enum db_err: Add DB_SUCCESS_LOCKED_REC for indicating a successful
operation where a record lock was created.
lock_sec_rec_read_check_and_lock(),
lock_clust_rec_read_check_and_lock(), lock_rec_enqueue_waiting(),
lock_rec_lock_slow(), lock_rec_lock(), row_ins_set_shared_rec_lock(),
row_ins_set_exclusive_rec_lock(), sel_set_rec_lock(),
row_sel_get_clust_rec_for_mysql(): Return DB_SUCCESS_LOCKED_REC if a
new record lock was created. Adjust callers.
row_unlock_for_mysql(): Correct the function documentation.
row_prebuilt_t::new_rec_locks: Correct the documentation.
In semi-consistent read, only unlock freshly locked non-matching records.
Define DB_SUCCESS_LOCKED_REC for indicating a successful operation
where a record lock was created.
lock_rec_lock_fast(): Return LOCK_REC_SUCCESS,
LOCK_REC_SUCCESS_CREATED, or LOCK_REC_FAIL instead of TRUE/FALSE.
lock_sec_rec_read_check_and_lock(),
lock_clust_rec_read_check_and_lock(), lock_rec_enqueue_waiting(),
lock_rec_lock_slow(), lock_rec_lock(), row_ins_set_shared_rec_lock(),
row_ins_set_exclusive_rec_lock(), sel_set_rec_lock(),
row_sel_get_clust_rec_for_mysql(): Return DB_SUCCESS_LOCKED_REC if a
new record lock was created. Adjust callers.
row_unlock_for_mysql(): Correct the function documentation.
row_prebuilt_t::new_rec_locks: Correct the documentation.
------------------------------------------------------------
revno: 3421
revision-id: marko.makela@oracle.com-20100426131029-1ffja69h6n88q6bo
parent: marko.makela@oracle.com-20100426112609-f7lgl8crw4x4sfkk
committer: Marko M?kel? <marko.makela@oracle.com>
branch nick: 5.1-innodb
timestamp: Mon 2010-04-26 16:10:29 +0300
message:
lock_rec_queue_validate(): Disable a bogus check that
a transaction that holds a lock on a clustered index record
also holds a lock on the secondary index record.
modified:
storage/innobase/lock/lock0lock.c 2@cee13dc7-1704-0410-992b-c9b4543f1246:trunk%2Flock%2Flock0lock.c
storage/innodb_plugin/lock/lock0lock.c 2@16c675df-0fcb-4bc9-8058-dcc011a37293:trunk%2Flock%2Flock0lock.c
------------------------------------------------------------
Detailed revision comments:
r6545 | jyang | 2010-02-03 03:57:32 +0200 (Wed, 03 Feb 2010) | 8 lines
branches/5.1: Fix bug #49001, "SHOW INNODB STATUS deadlock info
incorrect when deadlock detection aborts". Print the correct
lock owner when recursive function lock_deadlock_recursive()
exceeds its maximum depth LOCK_MAX_DEPTH_IN_DEADLOCK_CHECK.
rb://217, approved by Marko.
and also applying 5.1-ss6355
Detailed revision comments:
r6324 | jyang | 2009-12-17 06:54:24 +0200 (Thu, 17 Dec 2009) | 8 lines
branches/5.1: Fix bug #47814 - Diagnostics are frequently not
printed after a long lock wait in InnoDB. Separate out the
lock wait timeout check thread from monitor information
printing thread.
rb://200 Approved by Marko.
r6349 | marko | 2009-12-22 11:09:54 +0200 (Tue, 22 Dec 2009) | 3 lines
branches/5.1: lock_print_info_summary(): Remove a reference to
innobase_mysql_end_print_arbitrary_thd() that should have been
removed in r6347 when removing the function.
r6350 | marko | 2009-12-22 11:11:09 +0200 (Tue, 22 Dec 2009) | 1 line
branches/5.1: Remove an obsolete declaration of LOCK_thread_count.
