MySQL SQL更新鎖定

SQL/">MySQL資料庫鎖是實現資料一致性，解決併發問題的重要手段。資料庫是一個多使用者共享的資源，當出現併發的時候，就會導致出現各種各樣奇怪的問題，就像程式程式碼一樣，出現多執行緒併發的時候，如果不做特殊控制的話，就會出現意外的事情，比如“髒“資料、修改丟失等問題。所以資料庫併發需要使用事務來控制，事務併發問題需要資料庫鎖來控制，所以資料庫鎖是跟併發控制和事務聯絡在一起的。本文主要描述基於更新SQL語句來理解MySQL鎖定。

一、構造環境

(root@localhost) [user]> show variables like 'version';
+---------------+------------+
| Variable_name | Value |
+---------------+------------+
| version | 5.7.23-log |
+---------------+------------+

(root@localhost) [user]> desc t1;
+-------------+--------------+------+-----+---------+----------------+
| Field | Type | Null | Key | Default | Extra |
+-------------+--------------+------+-----+---------+----------------+
| id | int(11) | NO | PRI | NULL | auto_increment |
| n | int(11) | YES | | NULL | |
| table_name | varchar(64) | YES | | NULL | |
| column_name | varchar(64) | YES | | NULL | |
| pad | varchar(100) | YES | | NULL | |
+-------------+--------------+------+-----+---------+----------------+

(root@localhost) [user]> select count(*) from t1;
+----------+
| count(*) |
+----------+
| 3406 |
+----------+

(root@localhost) [user]> create unique index idx_t1_pad on t1(pad);
Query OK, 0 rows affected (0.35 sec)
Records: 0 Duplicates: 0 Warnings: 0

(root@localhost) [user]> create index idx_t1_n on t1(n);
Query OK, 0 rows affected (0.03 sec)
Records: 0 Duplicates: 0 Warnings: 0
(root@localhost) [user]> show index from t1;
+-------+------------+------------+--------------+-------------+-----------+-------------+------+------------+
| Table | Non_unique | Key_name | Seq_in_index | Column_name | Collation | Cardinality | Null | Index_type |
+-------+------------+------------+--------------+-------------+-----------+-------------+------+------------+
| t1 | 0 | PRIMARY | 1 | id | A | 3462 | | BTREE |
| t1 | 0 | idx_t1_pad | 1 | pad | A | 3406 | YES | BTREE |
| t1 | 1 | idx_t1_n | 1 | n | A | 12 | YES | BTREE |
+-------+------------+------------+--------------+-------------+-----------+-------------+------+------------+
select 'Leshami' author,'http://blog.csdn.net/leshami' Blog;
+---------+------------------------------+
| author | Blog |
+---------+------------------------------+
| Leshami | http://blog.csdn.net/leshami |
+---------+------------------------------+

二、基於主鍵更新

(root@localhost) [user]> start transaction;
Query OK, 0 rows affected (0.00 sec)

(root@localhost) [user]> update t1 set table_name='t1' where id=1299;
Query OK, 1 row affected (0.00 sec)
Rows matched: 1 Changed: 1 Warnings: 0

SELECT trx_id,
trx_state,
trx_started,
trx_mysql_thread_id,
trx_tables_locked,
trx_rows_locked,
trx_rows_modified,
trx_isolation_level
FROM INFORMATION_SCHEMA.INNODB_TRX \G

-- 從下面的結果可知，trx_rows_locked，一行被鎖定
*************************** 1. row ***************************
trx_id: 6349647
trx_state: RUNNING
trx_started: 2018-11-06 16:54:12
trx_mysql_thread_id: 2
trx_tables_locked: 1
trx_rows_locked: 1
trx_rows_modified: 1
trx_isolation_level: REPEATABLE READ

(root@localhost) [user]> rollback;
Query OK, 0 rows affected (0.01 sec)

三、基於二級唯一索引

(root@localhost) [user]> start transaction;
Query OK, 0 rows affected (0.00 sec)

(root@localhost) [user]> update t1 set table_name='t2' where pad='4f39e2a03df3ab94b9f6a48c4aecdc0b';
Query OK, 1 row affected (0.00 sec)
Rows matched: 1 Changed: 1 Warnings: 0

SELECT trx_id,
trx_state,
trx_started,
trx_mysql_thread_id,
trx_tables_locked,
trx_rows_locked,
trx_rows_modified,
trx_isolation_level
FROM INFORMATION_SCHEMA.INNODB_TRX \G

-- 從下面的查詢結果可知，trx_rows_locked，2行被鎖定
*************************** 1. row ***************************
trx_id: 6349649
trx_state: RUNNING
trx_started: 2018-11-06 16:55:22
trx_mysql_thread_id: 2
trx_tables_locked: 1
trx_rows_locked: 2
trx_rows_modified: 1
trx_isolation_level: REPEATABLE READ

(root@localhost) [user]> rollback;
Query OK, 0 rows affected (0.00 sec)

三、基於二級非唯一索引

(root@localhost) [user]> start transaction;
Query OK, 0 rows affected (0.00 sec)

(root@localhost) [user]> update t1 set table_name='t3' where n=8;
Query OK, 350 rows affected (0.01 sec)
Rows matched: 351 Changed: 351 Warnings: 0

SELECT trx_id,
trx_state,
trx_started,
trx_mysql_thread_id,
trx_tables_locked,
trx_rows_locked,
trx_rows_modified,
trx_isolation_level
FROM INFORMATION_SCHEMA.INNODB_TRX \G

--從下面的查詢結果可知，703行被鎖定
*************************** 1. row ***************************
trx_id: 6349672
trx_state: RUNNING
trx_started: 2018-11-06 17:06:53
trx_mysql_thread_id: 2
trx_tables_locked: 1
trx_rows_locked: 703
trx_rows_modified: 351
trx_isolation_level: REPEATABLE READ 

(root@localhost) [user]> rollback;
Query OK, 0 rows affected (0.00 sec)

四、無索引更新

(root@localhost) [user]> start transaction;
Query OK, 0 rows affected (0.00 sec)

(root@localhost) [user]> update t1 set table_name='t4' where column_name='id';
Query OK, 26 rows affected (0.00 sec)
Rows matched: 26 Changed: 26 Warnings: 0

SELECT trx_id,
trx_state,
trx_started,
trx_mysql_thread_id,
trx_tables_locked,
trx_rows_locked,
trx_rows_modified,
trx_isolation_level
FROM INFORMATION_SCHEMA.INNODB_TRX \G

-- 從下面的查詢結果可知，trx_rows_locked，3429行被鎖定，而被更新的僅僅為26行
-- 而且這個結果超出了表上的總行數3406
*************************** 1. row ***************************
trx_id: 6349674
trx_state: RUNNING
trx_started: 2018-11-06 17:09:41
trx_mysql_thread_id: 2
trx_tables_locked: 1
trx_rows_locked: 3429
trx_rows_modified: 26
trx_isolation_level: REPEATABLE READ 

(root@localhost) [user]> rollback;
Query OK, 0 rows affected (0.00 sec)

-- 也可以通過show engine innodb status進行觀察

show engine innodb status\G

------------
TRANSACTIONS
------------
Trx id counter 6349584
Purge done for trx's n:o < 0 undo n:o < 0 state: running but idle
History list length 0
LIST OF TRANSACTIONS FOR EACH SESSION:
---TRANSACTION 421943222819552, not started
0 lock struct(s), heap size 1136, 0 row lock(s)
---TRANSACTION 6349583, ACTIVE 2 sec
2 lock struct(s), heap size 1136, 1 row lock(s), undo log entries 1


------------
TRANSACTIONS
------------
Trx id counter 6349586
Purge done for trx's n:o < 6349585 undo n:o < 0 state: running but idle
History list length 1
LIST OF TRANSACTIONS FOR EACH SESSION:
---TRANSACTION 421943222819552, not started
0 lock struct(s), heap size 1136, 0 row lock(s)
---TRANSACTION 6349585, ACTIVE 8 sec
3 lock struct(s), heap size 1136, 2 row lock(s), undo log entries 1
MySQL thread id 2, OS thread handle 140467640694528, query id 29 localhost root

五、鎖相關查詢SQL

1：檢視當前的事務
SELECT * FROM INFORMATION_SCHEMA.INNODB_TRX;

2：檢視當前鎖定的事務

SELECT * FROM INFORMATION_SCHEMA.INNODB_LOCKS;

3：檢視當前等鎖的事務
SELECT * FROM INFORMATION_SCHEMA.INNODB_LOCK_WAITS;

SELECT trx_id,
trx_state,
trx_started,
trx_mysql_thread_id thr_id,
trx_tables_locked tb_lck,
trx_rows_locked rows_lck,
trx_rows_modified row_mfy,
trx_isolation_level is_lvl
FROM INFORMATION_SCHEMA.INNODB_TRX;

SELECT r.`trx_id` waiting_trx_id,
r.`trx_mysql_thread_id` waiting_thread,
r.`trx_query` waiting_query,
b.`trx_id` bolcking_trx_id,
b.`trx_mysql_thread_id` blocking_thread,
b.`trx_query` block_query
FROM information_schema.`INNODB_LOCK_WAITS` w
INNER JOIN information_schema.`INNODB_TRX` b
ON b.`trx_id` = w.`blocking_trx_id`
INNER JOIN information_schema.`INNODB_TRX` r
ON r.`trx_id` = w.`requesting_trx_id`;

六、小結

1、MySQL表更新時，對記錄的鎖定根據更新時where謂詞條件來確定鎖定範圍

2、對於聚簇索引過濾，由於索引即資料，因為僅僅鎖定更新行，這是由聚簇索引的性質決定的

3、對於非聚簇唯一索引過濾，由於需要回表，因此鎖定為唯一索引過濾行數加上回錶行數

4、對於非聚簇非唯一索引過濾，涉及到了間隙鎖，因此鎖定的記錄數更多

5、如果過濾條件無任何索引或無法使用到索引，則鎖定整張表上所有資料行