Mysql 事務的隔離級別

開發工做中咱們會使用到事務，那大家知道事務又分哪幾種嗎？

MYSQL標準定義了4類隔離級別，用來限定事務內外的哪些改變是可見的，哪些是不可見的。

低的隔離級通常支持更高的併發處理，並擁有更低的系統開銷。
隔離級別由低到高：Read Uncommitted < Read Committed < Repeatable Read < Serializable。

Read Uncommitted（讀取未提交內容）

在該隔離級別，全部事務均可以看到其餘未提交(commit)事務的執行結果。
本隔離級別不多用於實際應用，由於它的性能也不比其餘級別好多少。
讀取未提交的數據，也被稱之爲髒讀（Dirty Read）。

[窗口A]:

mysql> set GLOBAL tx_isolation='READ-UNCOMMITTED';
Query OK, 0 rows affected (0.00 sec)

mysql> quit;
Bye

[root@vagrant-centos65 ~]# mysql -uroot -pxxxx(從新登陸)

mysql> SELECT @@tx_isolation;
+------------------+
| @@tx_isolation   |
+------------------+
| READ-UNCOMMITTED |
+------------------+
1 row in set (0.00 sec)

mysql> use test;
Database changed
mysql> begin;
Query OK, 0 rows affected (0.00 sec)

mysql> select * from user;
+----+------+
| id | name |
+----+------+
|  1 | a    |
|  2 | b    |
+----+------+
2 rows in set (0.00 sec)

[窗口B]:
mysql> select @@tx_isolation;
+------------------+
| @@tx_isolation   |
+------------------+
| READ-UNCOMMITTED |
+------------------+
1 row in set (0.00 sec)

mysql> begin;
Query OK, 0 rows affected (0.00 sec)

mysql> insert into test.user values (3, 'c');
Query OK, 1 row affected (0.00 sec)

mysql> select * from user;
+----+------+
| id | name |
+----+------+
|  1 | a    |
|  2 | b    |
|  3 | c    |
+----+------+
3 rows in set (0.00 sec)

//目前爲止，窗口B並未commit;

[窗口A]:
mysql> select * from user ;
+----+------+
| id | name |
+----+------+
|  1 | a    |
|  2 | b    |
|  3 | c    |
+----+------+
3 rows in set (0.00 sec)

Read Committed（讀取提交內容）

這是大多數數據庫系統的默認隔離級別（但不是MySQL默認的）。
它知足了隔離的簡單定義：一個事務只能看見已經提交事務所作的改變。
這種隔離級別 也支持所謂的不可重複讀（NonrepeatableRead），由於同一事務的其餘實例在該實例處理其間可能會有新的commit，因此同一 select 可能返回不一樣結果。

[窗口A]：

mysql> SET GLOBAL tx_isolation='READ-COMMITTED';
Query OK, 0 rows affected (0.00 sec)

mysql> quit;
Bye

[root@vagrant-centos65 ~]# mysql -uroot -pxxxx(從新登陸)

mysql> SELECT @@tx_isolation;
+----------------+
| @@tx_isolation |
+----------------+
| READ-COMMITTED |
+----------------+
1 row in set (0.00 sec)

mysql> begin;
Query OK, 0 rows affected (0.00 sec)

mysql> select * from test.user;
+----+------+
| id | name |
+----+------+
|  1 | a    |
|  2 | b    |
+----+------+
2 rows in set (0.00 sec)


[窗口B]:

mysql> SELECT @@tx_isolation;
+----------------+
| @@tx_isolation |
+----------------+
| READ-COMMITTED |
+----------------+
1 row in set (0.00 sec)

mysql> begin;
Query OK, 0 rows affected (0.00 sec)

mysql> select * from test.user;
+----+------+
| id | name |
+----+------+
|  1 | a    |
|  2 | b    |
+----+------+
2 rows in set (0.00 sec)

mysql> delete from test.user where id=1;
Query OK, 1 row affected (0.00 sec)

mysql> select * from test.user;
+----+------+
| id | name |
+----+------+
|  2 | b    |
+----+------+
1 row in set (0.00 sec)

[窗口A]:

mysql> select * from test.user;
+----+------+
| id | name |
+----+------+
|  1 | a    |
|  2 | b    |
+----+------+
2 rows in set (0.00 sec)

[窗口B]:

mysql> commit;
Query OK, 0 rows affected (0.02 sec)

[窗口A]:

mysql> select * from test.user;
+----+------+
| id | name |
+----+------+
|  2 | b    |
+----+------+
1 row in set (0.00 sec)

Repeatable Read（可重讀）

這是MySQL的默認事務隔離級別，它確保同一事務的多個實例在併發讀取數據時，會看到一樣的數據行。
不過理論上，這會致使另外一個棘手的問題：幻讀 （Phantom Read）。
簡單的說，幻讀指當用戶讀取某一範圍的數據行時，另外一個事務又在該範圍內插入了新行，當用戶再讀取該範圍的數據行時，會發現有新的「幻影」 行。
InnoDB和Falcon存儲引擎經過多版本併發控制（MVCC，Multiversion Concurrency Control）機制解決了該問題。

[窗口A]:

mysql> SET GLOBAL tx_isolation='REPEATABLE-READ';
Query OK, 0 rows affected (0.00 sec)

mysql> quit;
Bye

[root@vagrant-centos65 ~]# mysql -uroot -pxxxx(從新登陸)

mysql> SELECT @@tx_isolation;
+-----------------+
| @@tx_isolation  |
+-----------------+
| REPEATABLE-READ |
+-----------------+
1 row in set (0.00 sec)

mysql> begin;
Query OK, 0 rows affected (0.00 sec)

[窗口B]:

mysql> quit;
Bye

[root@vagrant-centos65 ~]# mysql -uroot -pxxxx(從新登陸)

mysql> SELECT @@tx_isolation;
+-----------------+
| @@tx_isolation  |
+-----------------+
| REPEATABLE-READ |
+-----------------+
1 row in set (0.00 sec)

mysql> insert into test.user values (4, 'd');
Query OK, 1 row affected (0.00 sec)

mysql> select * from test.user;
+----+------+
| id | name |
+----+------+
|  2 | b    |
|  4 | d    |
+----+------+
2 rows in set (0.00 sec)

[窗口A]:

mysql> select * from test.user;
+----+------+
| id | name |
+----+------+
|  2 | b    |
+----+------+
1 rows in set (0.00 sec)

mysql> commit;
Query OK, 0 rows affected (0.00 sec)

mysql> select * from test.user;
+----+------+
| id | name |
+----+------+
|  2 | b    |
|  4 | d    |
+----+------+
2 rows in set (0.00 sec)

Serializable（序列化執行）

這是最高的隔離級別，它經過強制事務排序，使之不可能相互衝突，從而解決幻讀問題。
簡言之，它是在每一個讀的數據行上加上共享鎖。在這個級別，可能致使大量的超時現象和鎖競爭。

[窗口A]:

mysql> SET GLOBAL tx_isolation='SERIALIZABLE';
Query OK, 0 rows affected (0.00 sec)

mysql> quit;
Bye

[root@vagrant-centos65 ~]# mysql -uroot -pxxxx(從新登陸)

mysql> SELECT @@tx_isolation;
+----------------+
| @@tx_isolation |
+----------------+
| SERIALIZABLE   |
+----------------+
1 row in set (0.00 sec)

mysql> select * from test.user;
+----+------+
| id | name |
+----+------+
|  2 | b    |
|  4 | d    |
+----+------+
2 rows in set (0.00 sec)

mysql> begin;
Query OK, 0 rows affected (0.00 sec)

mysql> insert into test.user values (5, 'e');
Query OK, 1 row affected (0.00 sec)

[窗口B]:

mysql> quit;
Bye

[root@vagrant-centos65 ~]# mysql -uroot -pxxxx(從新登陸)

mysql> SELECT @@tx_isolation;
+----------------+
| @@tx_isolation |
+----------------+
| SERIALIZABLE   |
+----------------+
1 row in set (0.00 sec)

mysql> select * from test.user;
ERROR 1205 (HY000): Lock wait timeout exceeded; try restarting transaction

[窗口A]:

mysql> commit;
Query OK, 0 rows affected (0.01 sec)

[窗口B]:

mysql> mysql> select * from test.user;
+----+------+
| id | name |
+----+------+
|  2 | b    |
|  4 | d    |
|  5 | e    |
+----+------+
3 rows in set (0.00 sec)