參考了這篇文章的一些內容:
http://xm-king.iteye.com/blog/770721
記住以下這張表:

我在springdemo庫里面建了一個表:
CREATE TABLE `tx` (
`id` bigint(11) NOT NULL auto_increment,
`num` bigint(11) default 0 COMMENT '用戶名',
PRIMARY KEY (`id`)
) ENGINE=InnoDB AUTO_INCREMENT=1 DEFAULT CHARSET=utf8 COMMENT='事務隔離級別測試表';
Mysql通過以下語句可以查詢
SELECT @@tx_isolation;
SELECT @@global.tx_isolation;
SELECT @@session.tx_isolation;
默認的是 REPEATABLE-READ
可以通過以下方式修改隔離級別:
set tx_isolation='read-committed';
mysql> select @@tx_isolation;
+-----------------+
| @@tx_isolation |
+-----------------+
| REPEATABLE-READ |
+-----------------+
1 row in set (0.00 sec)
mysql> set tx_isolation='read-committed';
Query OK, 0 rows affected (0.00 sec)
mysql> select @@tx_isolation;
+----------------+
| @@tx_isolation |
+----------------+
| READ-COMMITTED |
+----------------+
1 row in set (0.00 sec)
但是,注意以上的改動,只能針對當前會話。
另起一個客戶端,輸入
select @@tx_isolation;
+-----------------+
| @@tx_isolation |
+-----------------+
| REPEATABLE-READ |
+-----------------+
————————————————————————————————
以下是各種情況的測試:
READ-UNCOMMITTED
update tx set num=11 where id=1;
ERROR 1665 (HY000): Cannot execute statement: impossible to write to binary log since BINLOG_FORMAT = STATEMENT and at least one table uses a storage engine limited to row-based logging. InnoDB is limited to row-logging when transaction isolation level is READ COMMITTED or READ UNCOMMITTED.
READ-COMMITTED
update tx set num=11 where id=1;
ERROR 1665 (HY000): Cannot execute statement: impossible to write to binary log since BINLOG_FORMAT = STATEMENT and at least one table uses a storage engine limited to row-based logging. InnoDB is limited to row-logging when transaction isolation level is READ COMMITTED or READ UNCOMMITTED.
REPEATABLE-READ
A:
mysql> start transaction; Query OK, 0 rows affected (0.00 sec) mysql> select * from tx; (B更新前) +----+------+ | id | num | +----+------+ | 1 | 1 | | 2 | 2 | | 3 | 3 | +----+------+ 3 rows in set (0.00 sec) mysql> select * from tx; (B更新后) +----+------+ | id | num | +----+------+ | 1 | 1 | | 2 | 2 | | 3 | 3 | +----+------+ 3 rows in set (0.00 sec) mysql> commit; Query OK, 0 rows affected (0.00 sec)
mysql> select * from tx;
+----+------+
| id | num |
+----+------+
| 1 | 10 |
| 2 | 2 |
| 3 | 3 |
+----+------+
可以看出,A的事務中,看不到B對數據的更新。
同樣是 REPEATABLE-READ
mysql> start transaction; Query OK, 0 rows affected (0.00 sec) mysql> select * from tx; +----+------+ | id | num | +----+------+ | 1 | 10 | | 2 | 2 | | 3 | 3 | +----+------+ 3 rows in set (0.00 sec) mysql> select * from tx; +----+------+ | id | num | +----+------+ | 1 | 10 | | 2 | 2 | | 3 | 3 | +----+------+ 3 rows in set (0.00 sec)
mysql> rollback;
Query OK, 0 rows affected (0.00 sec)
mysql> select * from tx;
+----+------+
| id | num |
+----+------+
| 1 | 10 |
| 2 | 2 |
| 3 | 3 |
| 4 | 4 |
+----+------+
4 rows in set (0.00 sec)
另一個客戶端,insert 了一條記錄.
對於正常的repeatable-read級別,是有可能出現幻讀的情況,也就是說,第二遍的時候,A能夠讀到新插入的數據。
但是,InnoDB和Falcon存儲引擎通過多版本並發控制(MVCC,Multiversion Concurrency Control)機制解決了該問題。
所以我上面用InnoDB引擎創建的數據庫,就沒有出現幻讀的情況。
SERIALIZABLE
這里注意,只需要對使用事務Transaction的客戶端設置SERIALIZABLE ,其他客戶端的級別是什么都行,比如REPEATABLE-READ .
A
mysql> select @@tx_isolation; +----------------+ | @@tx_isolation | +----------------+ | SERIALIZABLE | +----------------+ 1 row in set (0.00 sec) mysql> start transaction; Query OK, 0 rows affected (0.00 sec) mysql> select * from tx; +----+------+ | id | num | +----+------+ | 1 | 10 | | 2 | 2 | | 3 | 3 | | 4 | 4 | | 5 | 5 | +----+------+ 5 rows in set (0.00 sec) mysql> rollback; Query OK, 0 rows affected (0.00 sec)
B
mysql> select @@tx_isolation; +-----------------+ | @@tx_isolation | +-----------------+ | REPEATABLE-READ | +-----------------+ 1 row in set (0.00 sec) mysql> insert into tx values(5,5); Query OK, 1 row affected (0.09 sec) mysql> insert into tx values(6,6); Query OK, 1 row affected (31.60 sec)
這里,有2點需要注意:
1. A沒有使用(select)表tx的時候,B仍然能夠向tx表中插入數據;
2. A使用(select)表tx之后,B再進行insert操作,就會hang住,直到A表transaction結束。所以可以看到B的insert操作耗時31秒。當然了,也可能超時失敗。
比如:
mysql> insert into tx values(10,10);
ERROR 1205 (HY000): Lock wait timeout exceeded; try restarting transaction
以上這條語句,看起來是鎖了整個表。但是其實深層次的目的,是為了保證事務的完整性,以及讓A兩次操作的REPEATABLE_READ;
比如 使用 select * from tx limit 1; 后面就仍然可以insert。因為不改變之前select的結果。
如果 select * from tx limit 10; 而最終只檢出8條。這時候就不能insert,因為insert了的話,下次同樣select得到的結果就不一樣。
同理,如果select了某條記錄,那么update同一條記錄就不行,update其他的記錄就可以。而且在事務中能夠讀到新更新的數據。
所以記住,事務的隔離級別的要求,只對事務過程中已經獲取過的數據有關。跟沒獲取過、其他不可見的數據,無關。
