mirror of
https://github.com/MariaDB/server.git
synced 2025-02-02 20:11:42 +01:00
39a1a50dfb
"HANDLER statements within a transaction might lead to deadlocks". Introduce a notion of a sentinel to MDL_context. A sentinel is a ticket that separates all tickets in the context into two groups: before and after it. Currently we can have (and need) only one designated sentinel -- it separates all locks taken by LOCK TABLE or HANDLER statement, which must survive COMMIT and ROLLBACK and all other locks, which must be released at COMMIT or ROLLBACK. The tricky part is maintaining the sentinel up to date when someone release its corresponding ticket. This can happen, e.g. if someone issues DROP TABLE under LOCK TABLES (generally, see all calls to release_all_locks_for_name()). MDL_context::release_ticket() is modified to take care of it. ****** A fix and a test case for Bug#46224 "HANDLER statements within a transaction might lead to deadlocks". An attempt to mix HANDLER SQL statements, which are transaction- agnostic, an open multi-statement transaction, and DDL against the involved tables (in a concurrent connection) could lead to a deadlock. The deadlock would occur when HANDLER OPEN or HANDLER READ would have to wait on a conflicting metadata lock. If the connection that issued HANDLER statement also had other metadata locks (say, acquired in scope of a transaction), a classical deadlock situation of mutual wait could occur. Incompatible change: entering LOCK TABLES mode automatically closes all open HANDLERs in the current connection. Incompatible change: previously an attempt to wait on a lock in a connection that has an open HANDLER statement could wait indefinitely/deadlock. After this patch, an error ER_LOCK_DEADLOCK is produced. The idea of the fix is to merge thd->handler_mdl_context with the main mdl_context of the connection, used for transactional locks. This makes deadlock detection possible, since all waits with locks are "visible" and available to analysis in a single MDL context of the connection. Since HANDLER locks and transactional locks have a different life cycle -- HANDLERs are explicitly open and closed, and so are HANDLER locks, explicitly acquired and released, whereas transactional locks "accumulate" till the end of a transaction and are released only with COMMIT, ROLLBACK and ROLLBACK TO SAVEPOINT, a concept of "sentinel" was introduced to MDL_context. All locks, HANDLER and others, reside in the same linked list. However, a selected element of the list separates locks with different life cycle. HANDLER locks always reside at the end of the list, after the sentinel. Transactional locks are prepended to the beginning of the list, before the sentinel. Thus, ROLLBACK, COMMIT or ROLLBACK TO SAVEPOINT, only release those locks that reside before the sentinel. HANDLER locks must be released explicitly as part of HANDLER CLOSE statement, or an implicit close. The same approach with sentinel is also employed for LOCK TABLES locks. Since HANDLER and LOCK TABLES statement has never worked together, the implementation is made simple and only maintains one sentinel, which is used either for HANDLER locks, or for LOCK TABLES locks.
242 lines
6.7 KiB
C++
242 lines
6.7 KiB
C++
/* Copyright (C) 2006 MySQL AB
|
|
|
|
This program is free software; you can redistribute it and/or modify
|
|
it under the terms of the GNU General Public License as published by
|
|
the Free Software Foundation; version 2 of the License.
|
|
|
|
This program is distributed in the hope that it will be useful,
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
GNU General Public License for more details.
|
|
|
|
You should have received a copy of the GNU General Public License
|
|
along with this program; if not, write to the Free Software
|
|
Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA */
|
|
|
|
#include "mysql_priv.h"
|
|
#include "rpl_injector.h"
|
|
#include "transaction.h"
|
|
|
|
/*
|
|
injector::transaction - member definitions
|
|
*/
|
|
|
|
/* inline since it's called below */
|
|
inline
|
|
injector::transaction::transaction(MYSQL_BIN_LOG *log, THD *thd)
|
|
: m_state(START_STATE), m_thd(thd)
|
|
{
|
|
/*
|
|
Default initialization of m_start_pos (which initializes it to garbage).
|
|
We need to fill it in using the code below.
|
|
*/
|
|
LOG_INFO log_info;
|
|
log->get_current_log(&log_info);
|
|
/* !!! binlog_pos does not follow RAII !!! */
|
|
m_start_pos.m_file_name= my_strdup(log_info.log_file_name, MYF(0));
|
|
m_start_pos.m_file_pos= log_info.pos;
|
|
|
|
trans_begin(m_thd);
|
|
|
|
thd->set_current_stmt_binlog_row_based();
|
|
}
|
|
|
|
injector::transaction::~transaction()
|
|
{
|
|
if (!good())
|
|
return;
|
|
|
|
/* Needed since my_free expects a 'char*' (instead of 'void*'). */
|
|
char* const the_memory= const_cast<char*>(m_start_pos.m_file_name);
|
|
|
|
/*
|
|
We set the first character to null just to give all the copies of the
|
|
start position a (minimal) chance of seening that the memory is lost.
|
|
All assuming the my_free does not step over the memory, of course.
|
|
*/
|
|
*the_memory= '\0';
|
|
|
|
my_free(the_memory, MYF(0));
|
|
}
|
|
|
|
int injector::transaction::commit()
|
|
{
|
|
DBUG_ENTER("injector::transaction::commit()");
|
|
m_thd->binlog_flush_pending_rows_event(true);
|
|
/*
|
|
Cluster replication does not preserve statement or
|
|
transaction boundaries of the master. Instead, a new
|
|
transaction on replication slave is started when a new GCI
|
|
(global checkpoint identifier) is issued, and is committed
|
|
when the last event of the check point has been received and
|
|
processed. This ensures consistency of each cluster in
|
|
cluster replication, and there is no requirement for stronger
|
|
consistency: MySQL replication is asynchronous with other
|
|
engines as well.
|
|
|
|
A practical consequence of that is that row level replication
|
|
stream passed through the injector thread never contains
|
|
COMMIT events.
|
|
Here we should preserve the server invariant that there is no
|
|
outstanding statement transaction when the normal transaction
|
|
is committed by committing the statement transaction
|
|
explicitly.
|
|
*/
|
|
trans_commit_stmt(m_thd);
|
|
if (!trans_commit(m_thd))
|
|
{
|
|
close_thread_tables(m_thd);
|
|
m_thd->mdl_context.release_transactional_locks();
|
|
}
|
|
DBUG_RETURN(0);
|
|
}
|
|
|
|
|
|
int injector::transaction::use_table(server_id_type sid, table tbl)
|
|
{
|
|
DBUG_ENTER("injector::transaction::use_table");
|
|
|
|
int error;
|
|
|
|
if ((error= check_state(TABLE_STATE)))
|
|
DBUG_RETURN(error);
|
|
|
|
server_id_type save_id= m_thd->server_id;
|
|
m_thd->set_server_id(sid);
|
|
error= m_thd->binlog_write_table_map(tbl.get_table(),
|
|
tbl.is_transactional());
|
|
m_thd->set_server_id(save_id);
|
|
DBUG_RETURN(error);
|
|
}
|
|
|
|
|
|
int injector::transaction::write_row (server_id_type sid, table tbl,
|
|
MY_BITMAP const* cols, size_t colcnt,
|
|
record_type record)
|
|
{
|
|
DBUG_ENTER("injector::transaction::write_row(...)");
|
|
|
|
if (int error= check_state(ROW_STATE))
|
|
DBUG_RETURN(error);
|
|
|
|
server_id_type save_id= m_thd->server_id;
|
|
m_thd->set_server_id(sid);
|
|
m_thd->binlog_write_row(tbl.get_table(), tbl.is_transactional(),
|
|
cols, colcnt, record);
|
|
m_thd->set_server_id(save_id);
|
|
DBUG_RETURN(0);
|
|
}
|
|
|
|
|
|
int injector::transaction::delete_row(server_id_type sid, table tbl,
|
|
MY_BITMAP const* cols, size_t colcnt,
|
|
record_type record)
|
|
{
|
|
DBUG_ENTER("injector::transaction::delete_row(...)");
|
|
|
|
if (int error= check_state(ROW_STATE))
|
|
DBUG_RETURN(error);
|
|
|
|
server_id_type save_id= m_thd->server_id;
|
|
m_thd->set_server_id(sid);
|
|
m_thd->binlog_delete_row(tbl.get_table(), tbl.is_transactional(),
|
|
cols, colcnt, record);
|
|
m_thd->set_server_id(save_id);
|
|
DBUG_RETURN(0);
|
|
}
|
|
|
|
|
|
int injector::transaction::update_row(server_id_type sid, table tbl,
|
|
MY_BITMAP const* cols, size_t colcnt,
|
|
record_type before, record_type after)
|
|
{
|
|
DBUG_ENTER("injector::transaction::update_row(...)");
|
|
|
|
if (int error= check_state(ROW_STATE))
|
|
DBUG_RETURN(error);
|
|
|
|
server_id_type save_id= m_thd->server_id;
|
|
m_thd->set_server_id(sid);
|
|
m_thd->binlog_update_row(tbl.get_table(), tbl.is_transactional(),
|
|
cols, colcnt, before, after);
|
|
m_thd->set_server_id(save_id);
|
|
DBUG_RETURN(0);
|
|
}
|
|
|
|
|
|
injector::transaction::binlog_pos injector::transaction::start_pos() const
|
|
{
|
|
return m_start_pos;
|
|
}
|
|
|
|
|
|
/*
|
|
injector - member definitions
|
|
*/
|
|
|
|
/* This constructor is called below */
|
|
inline injector::injector()
|
|
{
|
|
}
|
|
|
|
static injector *s_injector= 0;
|
|
injector *injector::instance()
|
|
{
|
|
if (s_injector == 0)
|
|
s_injector= new injector;
|
|
/* "There can be only one [instance]" */
|
|
return s_injector;
|
|
}
|
|
|
|
void injector::free_instance()
|
|
{
|
|
injector *inj = s_injector;
|
|
|
|
if (inj != 0)
|
|
{
|
|
s_injector= 0;
|
|
delete inj;
|
|
}
|
|
}
|
|
|
|
|
|
injector::transaction injector::new_trans(THD *thd)
|
|
{
|
|
DBUG_ENTER("injector::new_trans(THD*)");
|
|
/*
|
|
Currently, there is no alternative to using 'mysql_bin_log' since that
|
|
is hardcoded into the way the handler is using the binary log.
|
|
*/
|
|
DBUG_RETURN(transaction(&mysql_bin_log, thd));
|
|
}
|
|
|
|
void injector::new_trans(THD *thd, injector::transaction *ptr)
|
|
{
|
|
DBUG_ENTER("injector::new_trans(THD *, transaction *)");
|
|
/*
|
|
Currently, there is no alternative to using 'mysql_bin_log' since that
|
|
is hardcoded into the way the handler is using the binary log.
|
|
*/
|
|
transaction trans(&mysql_bin_log, thd);
|
|
ptr->swap(trans);
|
|
|
|
DBUG_VOID_RETURN;
|
|
}
|
|
|
|
int injector::record_incident(THD *thd, Incident incident)
|
|
{
|
|
Incident_log_event ev(thd, incident);
|
|
if (int error= mysql_bin_log.write(&ev))
|
|
return error;
|
|
mysql_bin_log.rotate_and_purge(RP_FORCE_ROTATE);
|
|
return 0;
|
|
}
|
|
|
|
int injector::record_incident(THD *thd, Incident incident, LEX_STRING const message)
|
|
{
|
|
Incident_log_event ev(thd, incident, message);
|
|
if (int error= mysql_bin_log.write(&ev))
|
|
return error;
|
|
mysql_bin_log.rotate_and_purge(RP_FORCE_ROTATE);
|
|
return 0;
|
|
}
|