Alex的博客

本博客的文章均为原创,是本人从事行业多年来所遇见一些小问题的解决心得,希望可以帮助到大家!



mysql5.7开启binlog日志

在my.cnf的mysqld下加入:
server_id = 0
log_bin=/backup/mysql_data/mysql_binlog/mysql-bin
binlog_format           = MIXED

max_binlog_size         = 100m


重启mysql服务

查看是否成功

登陆mysql终端控制台

show variables like 'log_%';

mysql> show variables like 'log_%';
+----------------------------------------+------------------------------------+
| Variable_name                          | Value                              |
+----------------------------------------+------------------------------------+
| log_bin                                | ON                                 |
| log_bin_basename                       | /backup/mysql/mysql-bin       |
| log_bin_index                          | /backup/mysql/mysql-bin.index |
| log_bin_trust_function_creators        | OFF                                |
| log_bin_use_v1_row_events              | OFF                                |
| log_builtin_as_identified_by_password  | OFF                                |
| log_error                              | /backup/mysql/mysqld.log      |
| log_error_verbosity                    | 3                                  |
| log_output                             | FILE                               |
| log_queries_not_using_indexes          | OFF                                |
| log_slave_updates                      | OFF                                |
| log_slow_admin_statements              | OFF                                |
| log_slow_slave_statements              | OFF                                |
| log_statements_unsafe_for_binlog       | ON                                 |
| log_syslog                             | OFF                                |
| log_syslog_facility                    | daemon                             |
| log_syslog_include_pid                 | ON                                 |
| log_syslog_tag                         |                                    |
| log_throttle_queries_not_using_indexes | 0                                  |
| log_timestamps                         | UTC                                |
| log_warnings                           | 2                                  |
+----------------------------------------+------------------------------------+
21 rows in set (0.00 sec)


常用的binlog日志操作命令

1)查看所有binlog日志列表
mysql> show master logs;
+------------------+-----------+
| Log_name | File_size |
+------------------+-----------+
| mysql-bin.000001 | 149 |
| mysql-bin.000002 | 4102 |
+------------------+-----------+
2 rows in set (0.00 sec)

2)查看master状态,即最后(最新)一个binlog日志的编号名称,及其最后一个操作事件pos结束点(Position)值
mysql> show master status;
+------------------+----------+--------------+------------------+
| File | Position | Binlog_Do_DB | Binlog_Ignore_DB |
+------------------+----------+--------------+------------------+
| mysql-bin.000002 | 4102 | | |
+------------------+----------+--------------+------------------+
1 row in set (0.00 sec)

3)flush刷新log日志,自此刻开始产生一个新编号的binlog日志文件
mysql> flush logs;  
Query OK, 0 rows affected (0.13 sec)

mysql> show master logs; 
+------------------+-----------+
| Log_name | File_size |
+------------------+-----------+
| mysql-bin.000001 | 149 |
| mysql-bin.000002 | 4145 |
| mysql-bin.000003 | 106 |
+------------------+-----------+
3 rows in set (0.00 sec)

注意:
每当mysqld服务重启时,会自动执行此命令,刷新binlog日志;在mysqldump备份数据时加 -F 选项也会刷新binlog日志;

4)重置(清空)所有binlog日志
mysql> reset master;
Query OK, 0 rows affected (0.12 sec)

mysql> show master logs; 
+------------------+-----------+
| Log_name | File_size |
+------------------+-----------+
| mysql-bin.000001 | 106 |
+------------------+-----------+
1 row in set (0.00 sec)

查看binlog日志内容,常用有两种方式:
1)使用mysqlbinlog自带查看命令法:
注意:
-->binlog是二进制文件,普通文件查看器cat、more、vim等都无法打开,必须使用自带的mysqlbinlog命令查看
进入配置文件指定的目录 然后输入 mysqlbinlog 名称 就可以查看了
-->在MySQL5.5以下版本使用mysqlbinlog命令时如果报错,就加上 “--no-defaults”选项

查看mysql的数据存放目录,从下面结果可知是/var/lib//mysql
[root@vm-002 ~]# ps -ef|grep mysql
root 9791 1 0 21:18 pts/0 00:00:00 /bin/sh /usr/bin/mysqld_safe --datadir=/var/lib/mysql --socket=/var/lib/mysql/mysql.sock --pid-file=/var/run/mysqld/mysqld.pid --basedir=/usr --user=mysql
mysql 9896 9791 0 21:18 pts/0 00:00:00 /usr/libexec/mysqld --basedir=/usr --datadir=/var/lib/mysql --user=mysql --log-error=/var/log/mysqld.log --pid-file=/var/run/mysqld/mysqld.pid --socket=/var/lib/mysql/mysql.sock
root 9916 9699 0 21:18 pts/0 00:00:00 mysql -px xxxx
root 9919 9715 0 21:23 pts/1 00:00:00 grep --color mysql

[root@vm-002 ~]# cd /var/lib/mysql/
[root@vm-002 mysql]# ls
ibdata1 ib_logfile0 ib_logfile1 mysql mysql-bin.000001 mysql-bin.000002 mysql-bin.index mysql.sock ops test

使用mysqlbinlog命令查看binlog日志内容,下面截取其中的一个片段分析:
[root@vm-002 mysql]# mysqlbinlog mysql-bin.000002
..............
# at 624
#160925 21:29:53 server id 1 end_log_pos 796 Query thread_id=3 exec_time=0 error_code=0
SET TIMESTAMP=1474810193/*!*/;
insert into member(`name`,`sex`,`age`,`classid`) values('wangshibo','m',27,'cls1'),('guohuihui','w',27,'cls2')        #执行的sql语句
/*!*/;
# at 796
#160925 21:29:53 server id 1 end_log_pos 823 Xid = 17                  #执行的时间
.............

解释:
server id 1 : 数据库主机的服务号;
end_log_pos 796: sql结束时的pos节点
thread_id=11: 线程号

2)上面这种办法读取出binlog日志的全文内容比较多,不容易分辨查看到pos点信息
下面介绍一种更为方便的查询命令:
命令格式:
mysql> show binlog events [IN 'log_name'] [FROM pos] [LIMIT [offset,] row_count];
参数解释:
IN 'log_name' :指定要查询的binlog文件名(不指定就是第一个binlog文件)
FROM pos :指定从哪个pos起始点开始查起(不指定就是从整个文件首个pos点开始算)
LIMIT [offset,] :偏移量(不指定就是0)
row_count :查询总条数(不指定就是所有行)

mysql> show master logs;
+------------------+-----------+
| Log_name | File_size |
+------------------+-----------+
| mysql-bin.000001 | 125 |
| mysql-bin.000002 | 823 |
+------------------+-----------+
2 rows in set (0.00 sec)

mysql> show binlog events in 'mysql-bin.000002'\G;
*************************** 1. row ***************************
Log_name: mysql-bin.000002
Pos: 4
Event_type: Format_desc
Server_id: 1
End_log_pos: 106
Info: Server ver: 5.1.73-log, Binlog ver: 4
*************************** 2. row ***************************
Log_name: mysql-bin.000002
Pos: 106
Event_type: Query
Server_id: 1
End_log_pos: 188
Info: use `ops`; drop table customers
*************************** 3. row ***************************
Log_name: mysql-bin.000002
Pos: 188
Event_type: Query
Server_id: 1
End_log_pos: 529
Info: use `ops`; CREATE TABLE IF NOT EXISTS `member` (
`id` int(10) unsigned NOT NULL AUTO_INCREMENT,
`name` varchar(16) NOT NULL,
`sex` enum('m','w') NOT NULL DEFAULT 'm',
`age` tinyint(3) unsigned NOT NULL,
`classid` char(6) DEFAULT NULL,
PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8
*************************** 4. row ***************************
Log_name: mysql-bin.000002
Pos: 529
Event_type: Query
Server_id: 1
End_log_pos: 596
Info: BEGIN
*************************** 5. row ***************************
Log_name: mysql-bin.000002
Pos: 596
Event_type: Intvar
Server_id: 1
End_log_pos: 624
Info: INSERT_ID=1
*************************** 6. row ***************************
Log_name: mysql-bin.000002
Pos: 624
Event_type: Query
Server_id: 1
End_log_pos: 796
Info: use `ops`; insert into member(`name`,`sex`,`age`,`classid`) values('wangshibo','m',27,'cls1'),('guohuihui','w',27,'cls2')
*************************** 7. row ***************************
Log_name: mysql-bin.000002
Pos: 796
Event_type: Xid
Server_id: 1
End_log_pos: 823
Info: COMMIT /* xid=17 */
7 rows in set (0.00 sec)

ERROR: 
No query specified

mysql> 

上面这条语句可以将指定的binlog日志文件,分成有效事件行的方式返回,并可使用limit指定pos点的起始偏移,查询条数!
如下操作示例:
a)查询第一个(最早)的binlog日志:
mysql> show binlog events\G; 

b)指定查询 mysql-bin.000002这个文件:
mysql> show binlog events in 'mysql-bin.000002'\G;

c)指定查询 mysql-bin.000002这个文件,从pos点:624开始查起:
mysql> show binlog events in 'mysql-bin.000002' from 624\G;

d)指定查询 mysql-bin.000002这个文件,从pos点:624开始查起,查询10条(即10条语句)
mysql> show binlog events in 'mysql-bin.000002' from 624 limit 10\G;

e)指定查询 mysql-bin.000002这个文件,从pos点:624开始查起,偏移2行(即中间跳过2个),查询10条
mysql> show binlog events in 'mysql-bin.000002' from 624 limit 2,10\G;

五、利用binlog日志恢复mysql数据

以下对ops库的member表进行操作
mysql> use ops;
mysql> CREATE TABLE IF NOT EXISTS `member` (
-> `id` int(10) unsigned NOT NULL AUTO_INCREMENT,
-> `name` varchar(16) NOT NULL,
-> `sex` enum('m','w') NOT NULL DEFAULT 'm',
-> `age` tinyint(3) unsigned NOT NULL,
-> `classid` char(6) DEFAULT NULL,
-> PRIMARY KEY (`id`)
-> ) ENGINE=InnoDB DEFAULT CHARSET=utf8;
Query OK, 0 rows affected (0.10 sec)

mysql> show tables;
+---------------+
| Tables_in_ops |
+---------------+
| member |
+---------------+
1 row in set (0.00 sec)

mysql> desc member;
+---------+---------------------+------+-----+---------+----------------+
| Field | Type | Null | Key | Default | Extra |
+---------+---------------------+------+-----+---------+----------------+
| id | int(10) unsigned | NO | PRI | NULL | auto_increment |
| name | varchar(16) | NO | | NULL | |
| sex | enum('m','w') | NO | | m | |
| age | tinyint(3) unsigned | NO | | NULL | |
| classid | char(6) | YES | | NULL | |
+---------+---------------------+------+-----+---------+----------------+
5 rows in set (0.00 sec)

事先插入两条数据
mysql> insert into member(`name`,`sex`,`age`,`classid`) values('wangshibo','m',27,'cls1'),('guohuihui','w',27,'cls2');
Query OK, 2 rows affected (0.08 sec)
Records: 2 Duplicates: 0 Warnings: 0
mysql> select * from member;
+----+-----------+-----+-----+---------+
| id | name | sex | age | classid |
+----+-----------+-----+-----+---------+
| 1 | wangshibo | m | 27 | cls1 |
| 2 | guohuihui | w | 27 | cls2 |
+----+-----------+-----+-----+---------+
2 rows in set (0.00 sec)

下面开始进行场景模拟:
1)
ops库会在每天凌晨4点进行一次完全备份的定时计划任务,如下:
[root@vm-002 ~]# crontab -l
0 4 * * * /usr/bin/mysqldump -uroot -p -B -F -R -x --master-data=2 ops|gzip >/opt/backup/ops_$(date +%F).sql.gz

这里手动执行下,将ops数据库备份到/opt/backup/ops_$(date +%F).sql.gz文件中:
[root@vm-002 ~]# mysqldump -uroot -p -B -F -R -x --master-data=2 ops|gzip >/opt/backup/ops_$(date +%F).sql.gz
Enter password: 
[root@vm-002 ~]# ls /opt/backup/
ops_2016-09-25.sql.gz
-----------------
参数说明:
-B:指定数据库
-F:刷新日志
-R:备份存储过程等
-x:锁表
--master-data:在备份语句里添加CHANGE MASTER语句以及binlog文件及位置点信息
-----------------
待到数据库备份完成,就不用担心数据丢失了,因为有完全备份数据在!!

由于上面在全备份的时候使用了-F选项,那么当数据备份操作刚开始的时候系统就会自动刷新log,这样就会自动产生
一个新的binlog日志,这个新的binlog日志就会用来记录备份之后的数据库“增删改”操作
查看一下:
mysql> show master status;
+------------------+----------+--------------+------------------+
| File | Position | Binlog_Do_DB | Binlog_Ignore_DB |
+------------------+----------+--------------+------------------+
| mysql-bin.000003 | 106 | | |
+------------------+----------+--------------+------------------+
1 row in set (0.00 sec)

也就是说, mysql-bin.000003 是用来记录4:00之后对数据库的所有“增删改”操作。

2)
早上9点上班了,由于业务的需求会对数据库进行各种“增删改”操作。
比如:在ops库下member表内插入、修改了数据等等:

先是早上进行插入数据:
mysql> insert into ops.member(`name`,`sex`,`age`,`classid`) values('yiyi','w',20,'cls1'),('xiaoer','m',22,'cls3'),('zhangsan','w',21,'cls5'),('lisi','m',20,'cls4'),('wangwu','w',26,'cls6');
Query OK, 5 rows affected (0.08 sec)
Records: 5 Duplicates: 0 Warnings: 0

mysql> select * from member;
+----+-----------+-----+-----+---------+
| id | name | sex | age | classid |
+----+-----------+-----+-----+---------+
| 1 | wangshibo | m | 27 | cls1 |
| 2 | guohuihui | w | 27 | cls2 |
| 3 | yiyi | w | 20 | cls1 |
| 4 | xiaoer | m | 22 | cls3 |
| 5 | zhangsan | w | 21 | cls5 |
| 6 | lisi | m | 20 | cls4 |
| 7 | wangwu | w | 26 | cls6 |
+----+-----------+-----+-----+---------+
7 rows in set (0.00 sec)

3)
中午又执行了修改数据操作:
mysql> update ops.member set name='李四' where id=4;
Query OK, 1 row affected (0.07 sec)
Rows matched: 1 Changed: 1 Warnings: 0

mysql> update ops.member set name='小二' where id=2;
Query OK, 1 row affected (0.06 sec)
Rows matched: 1 Changed: 1 Warnings: 0

mysql> select * from member;
+----+-----------+-----+-----+---------+
| id | name | sex | age | classid |
+----+-----------+-----+-----+---------+
| 1 | wangshibo | m | 27 | cls1 |
| 2 | 小二 | w | 27 | cls2 |
| 3 | yiyi | w | 20 | cls1 |
| 4 | 李四 | m | 22 | cls3 |
| 5 | zhangsan | w | 21 | cls5 |
| 6 | lisi | m | 20 | cls4 |
| 7 | wangwu | w | 26 | cls6 |
+----+-----------+-----+-----+---------+
7 rows in set (0.00 sec)

4)
在下午18:00的时候,悲剧莫名其妙的出现了!
手贱执行了drop语句,直接删除了ops库!吓尿!
mysql> drop database ops;
Query OK, 1 row affected (0.02 sec)

5)
这种时候,一定不要慌张!!!
先仔细查看最后一个binlog日志,并记录下关键的pos点,到底是哪个pos点的操作导致了数据库的破坏(通常在最后几步);

先备份一下最后一个binlog日志文件:
[root@vm-002 ~]# cd /var/lib/mysql/
[root@vm-002 mysql]# cp -v mysql-bin.000003 /opt/backup/
`mysql-bin.000003' -> `/opt/backup/mysql-bin.000003'
[root@vm-002 mysql]# ls /opt/backup/
mysql-bin.000003 ops_2016-09-25.sql.gz

接着执行一次刷新日志索引操作,重新开始新的binlog日志记录文件。按理说mysql-bin.000003
这个文件不会再有后续写入了,因为便于我们分析原因及查找ops节点,以后所有数据库操作都会写入到下一个日志文件。
mysql> flush logs;
Query OK, 0 rows affected (0.13 sec)

mysql> show master status;
+------------------+----------+--------------+------------------+
| File | Position | Binlog_Do_DB | Binlog_Ignore_DB |
+------------------+----------+--------------+------------------+
| mysql-bin.000004 | 106 | | |
+------------------+----------+--------------+------------------+
1 row in set (0.00 sec)

6)
读取binlog日志,分析问题。
读取binlog日志的方法上面已经说到。
方法一:使用mysqlbinlog读取binlog日志:
[root@vm-002 ~]# cd /var/lib/mysql/
[root@vm-002 mysql]# mysqlbinlog mysql-bin.000003

方法二:登录服务器,并查看(推荐此种方法)
mysql> show binlog events in 'mysql-bin.000003';

+------------------+-----+-------------+-----------+-------------+----------------------------------------------------------------------------------------------------------------------------+
| Log_name | Pos | Event_type | Server_id | End_log_pos | Info |
+------------------+-----+-------------+-----------+-------------+----------------------------------------------------------------------------------------------------------------------------+
| mysql-bin.000003 | 4 | Format_desc | 1 | 106 | Server ver: 5.1.73-log, Binlog ver: 4 |
| mysql-bin.000003 | 106 | Query | 1 | 173 | BEGIN |
| mysql-bin.000003 | 173 | Intvar | 1 | 201 | INSERT_ID=3 |
| mysql-bin.000003 | 201 | Query | 1 | 444 | use `ops`; insert into ops.member(`name`,`sex`,`age`,`gsan','w',21,'cls5'),('lisi','m',20,'cls4'),('wangwu','w',26,'cls6') |
| mysql-bin.000003 | 444 | Xid | 1 | 471 | COMMIT /* xid=66 */ |
| mysql-bin.000003 | 471 | Query | 1 | 538 | BEGIN |
| mysql-bin.000003 | 538 | Query | 1 | 646 | use `ops`; update ops.member set name='李四' where id= |
| mysql-bin.000003 | 646 | Xid | 1 | 673 | COMMIT /* xid=68 */ |
| mysql-bin.000003 | 673 | Query | 1 | 740 | BEGIN |
| mysql-bin.000003 | 740 | Query | 1 | 848 | use `ops`; update ops.member set name='小二' where id= |
| mysql-bin.000003 | 848 | Xid | 1 | 875 | COMMIT /* xid=69 */ |
| mysql-bin.000003 | 875 | Query | 1 |  954 | drop database ops |
| mysql-bin.000003 | 954 | Rotate | 1 | 997 | mysql-bin.000004;pos=4 |
+------------------+-----+-------------+-----------+-------------+----------------------------------------------------------------------------------------------------------------------------+
13 rows in set (0.00 sec)

或者:

mysql> show binlog events in 'mysql-bin.000003'\G;
.........
.........
*************************** 12. row ***************************
Log_name: mysql-bin.000003
Pos: 875
Event_type: Query
Server_id: 1
End_log_pos: 954
Info: drop database ops
*************************** 13. row ***************************
Log_name: mysql-bin.000003
Pos: 954
Event_type: Rotate
Server_id: 1
End_log_pos: 997
Info: mysql-bin.000004;pos=4
13 rows in set (0.00 sec)

通过分析,造成数据库破坏的pos点区间是介于 875--954 之间(这是按照日志区间的pos节点算的),只要恢复到875前就可

7)
先把凌晨4点全备份的数据恢复:
[root@vm-002 ~]# cd /opt/backup/
[root@vm-002 backup]# ls
mysql-bin.000003 ops_2016-09-25.sql.gz
[root@vm-002 backup]# gzip -d ops_2016-09-25.sql.gz 
[root@vm-002 backup]# mysql -uroot -p -v < ops_2016-09-25.sql 
Enter password: 
--------------
/*!40101 SET @OLD_CHARACTER_SET_CLIENT=@@CHARACTER_SET_CLIENT */
--------------

--------------
/*!40101 SET @OLD_CHARACTER_SET_RESULTS=@@CHARACTER_SET_RESULTS */
--------------

.............
.............

--------------
/*!40111 SET SQL_NOTES=@OLD_SQL_NOTES */
--------------

这样就恢复了截至当日凌晨(4:00)前的备份数据都恢复了。

mysql> show databases;                        #发现ops库已经恢复回来了
mysql> use ops;
Reading table information for completion of table and column names
You can turn off this feature to get a quicker startup with -A

Database changed
mysql> show tables;
+---------------+
| Tables_in_ops |
+---------------+
| member |
+---------------+
1 row in set (0.00 sec)

mysql> select * from member;
+----+-----------+-----+-----+---------+
| id | name | sex | age | classid |
+----+-----------+-----+-----+---------+
| 1 | wangshibo | m | 27 | cls1 |
| 2 | guohuihui | w | 27 | cls2 |
+----+-----------+-----+-----+---------+
2 rows in set (0.00 sec)

mysql>

但是这仅仅只是恢复了当天凌晨4点之前的数据,在4:00--18:00之间的数据还没有恢复回来!!
怎么办呢?
莫慌!这可以根据前面提到的mysql-bin.000003的新binlog日志进行恢复。

8)
从binlog日志恢复数据
恢复命令的语法格式:
mysqlbinlog mysql-bin.0000xx | mysql -u用户名 -p密码 数据库名

--------------------------------------------------------
常用参数选项解释:
--start-position=875 起始pos点
--stop-position=954 结束pos点
--start-datetime="2016-9-25 22:01:08" 起始时间点
--stop-datetime="2019-9-25 22:09:46" 结束时间点
--database=zyyshop 指定只恢复zyyshop数据库(一台主机上往往有多个数据库,只限本地log日志)
--------------------------------------------------------  
不常用选项:  
-u --user=name 连接到远程主机的用户名
-p --password[=name] 连接到远程主机的密码
-h --host=name 从远程主机上获取binlog日志
--read-from-remote-server 从某个MySQL服务器上读取binlog日志
--------------------------------------------------------
小结:实际是将读出的binlog日志内容,通过管道符传递给mysql命令。这些命令、文件尽量写成绝对路径;

a)完全恢复(需要手动vim编辑mysql-bin.000003,将那条drop语句剔除掉)
[root@vm-002 backup]# /usr/bin/mysqlbinlog /var/lib/mysql/mysql-bin.000003 | /usr/bin/mysql -uroot -p123456 -v ops

b)指定pos结束点恢复(部分恢复):
--stop-position=471 pos结束节点(按照事务区间算,是471)
注意:
此pos结束节点介于“member表原始数据”与更新“name='李四'”之前的数据,这样就可以恢复到更改“name='李四'”之前的数据了。
操作如下:
[root@vm-002 ~]# /usr/bin/mysqlbinlog --stop-position=471 --database=ops /var/lib/mysql/mysql-bin.000003 | /usr/bin/mysql -uroot -p123456 -v ops

mysql> select * from member;
+----+-----------+-----+-----+---------+
| id | name | sex | age | classid |
+----+-----------+-----+-----+---------+
| 1 | wangshibo | m | 27 | cls1 |
| 2 | guohuihui | w | 27 | cls2 |
| 3 | yiyi | w | 20 | cls1 |
| 4 | xiaoer | m | 22 | cls3 |
| 5 | zhangsan | w | 21 | cls5 |
| 6 | lisi | m | 20 | cls4 |
| 7 | wangwu | w | 26 | cls6 |
+----+-----------+-----+-----+---------+
7 rows in set (0.00 sec)

恢复截止到更改“name='李四'”之间的数据(按照事务区间算,是673)
[root@vm-002 ~]# /usr/bin/mysqlbinlog --stop-position=673 --database=ops /var/lib/mysql/mysql-bin.000003 | /usr/bin/mysql -uroot -p123456 -v ops

mysql> select * from member;
+----+-----------+-----+-----+---------+
| id | name | sex | age | classid |
+----+-----------+-----+-----+---------+
| 1 | wangshibo | m | 27 | cls1 |
| 2 | guohuihui | w | 27 | cls2 |
| 3 | yiyi | w | 20 | cls1 |
| 4 | 李四 | m | 22 | cls3 |
| 5 | zhangsan | w | 21 | cls5 |
| 6 | lisi | m | 20 | cls4 |
| 7 | wangwu | w | 26 | cls6 |
+----+-----------+-----+-----+---------+
7 rows in set (0.00 sec)

c)指定pso点区间恢复(部分恢复):
更新 name='李四' 这条数据,日志区间是Pos[538] --> End_log_pos[646],按事务区间是:Pos[471] --> End_log_pos[673]

更新 name='小二' 这条数据,日志区间是Pos[740] --> End_log_pos[848],按事务区间是:Pos[673] --> End_log_pos[875]

c1)
单独恢复 name='李四' 这步操作,可这样:
按照binlog日志区间单独恢复:
[root@vm-002 ~]# /usr/bin/mysqlbinlog --start-position=538 --stop-position=646 --database=ops /var/lib/mysql/mysql-bin.000003 | /usr/bin/mysql -uroot -p123456 -v ops

按照事务区间单独恢复
[root@vm-002 ~]# /usr/bin/mysqlbinlog --start-position=471 --stop-position=673 --database=ops /var/lib/mysql/mysql-bin.000003 | /usr/bin/mysql -uroot -p123456 -v ops

c2)
单独恢复 name='小二' 这步操作,可这样:
按照binlog日志区间单独恢复:
[root@vm-002 ~]# /usr/bin/mysqlbinlog --start-position=740 --stop-position=848 --database=ops /var/lib/mysql/mysql-bin.000003 | /usr/bin/mysql -uroot -p123456 -v ops

按照事务区间单独恢复
[root@vm-002 ~]# /usr/bin/mysqlbinlog --start-position=673 --stop-position=875 --database=ops /var/lib/mysql/mysql-bin.000003 | /usr/bin/mysql -uroot -p123456 -v ops

c3)
将 name='李四'、name='小二' 多步操作一起恢复,需要按事务区间,可这样:
[root@vm-002 ~]# /usr/bin/mysqlbinlog --start-position=471 --stop-position=875 --database=ops /var/lib/mysql/mysql-bin.000003 | /usr/bin/mysql -uroot -p123456 -v ops

查看数据库:
mysql> select * from member;
+----+-----------+-----+-----+---------+
| id | name | sex | age | classid |
+----+-----------+-----+-----+---------+
| 1 | wangshibo | m | 27 | cls1 |
| 2 | 小二 | w | 27 | cls2 |
| 3 | yiyi | w | 20 | cls1 |
| 4 | 李四 | m | 22 | cls3 |
| 5 | zhangsan | w | 21 | cls5 |
| 6 | lisi | m | 20 | cls4 |
| 7 | wangwu | w | 26 | cls6 |
+----+-----------+-----+-----+---------+
7 rows in set (0.00 sec)

这样,就恢复了删除前的数据状态了!!

-----------------
另外:
也可指定时间节点区间恢复(部分恢复):
除了用pos节点的办法进行恢复,也可以通过指定时间节点区间进行恢复,按时间恢复需要用mysqlbinlog命令读取binlog日志内容,找时间节点。

如上,误删除ops库后:
先进行全备份恢复
[root@vm-002 backup]# mysql -uroot -p -v < ops_2016-09-25.sql 

查看ops数据库
mysql> select * from member;
+----+-----------+-----+-----+---------+
| id | name | sex | age | classid |
+----+-----------+-----+-----+---------+
| 1 | wangshibo | m | 27 | cls1 |
| 2 | guohuihui | w | 27 | cls2 |
+----+-----------+-----+-----+---------+
2 rows in set (0.00 sec)

mysql>

查看mysq-bin00003日志,找出时间节点
[root@vm-002 ~]# cd /var/lib/mysql
[root@vm-002 mysql]# mysqlbinlog mysql-bin.000003 
.............
.............
BEGIN
/*!*/;
# at 173
#160925 21:57:19 server id 1 end_log_pos 201 Intvar
SET INSERT_ID=3/*!*/;
# at 201
#160925 21:57:19 server id 1 end_log_pos 444 Query thread_id=3 exec_time=0 error_code=0
use `ops`/*!*/;
SET TIMESTAMP=1474811839/*!*/;
insert into ops.member(`name`,`sex`,`age`,`classid`) values('yiyi','w',20,'cls1'),('xiaoer','m',22,'cls3'),('zhangsan','w',21,'cls5'),('lisi','m',20,'cls4'),('wangwu','w',26,'cls6')                               #执行的sql语句
/*!*/;
# at 444
#160925 21:57:19 server id 1 end_log_pos 471 Xid = 66    #开始执行的时间
COMMIT/*!*/;
# at 471
#160925 21:58:41 server id 1 end_log_pos 538 Query thread_id=3 exec_time=0 error_code=0    #结束时间
SET TIMESTAMP=1474811921/*!*/;
BEGIN
/*!*/;
# at 538
#160925 21:58:41 server id 1 end_log_pos 646 Query thread_id=3 exec_time=0 error_code=0
SET TIMESTAMP=1474811921/*!*/;
update ops.member set name='李四' where id=4     #执行的sql语句
/*!*/;
# at 646
#160925 21:58:41 server id 1 end_log_pos 673 Xid = 68    #开始执行的时间
COMMIT/*!*/;
# at 673
#160925 21:58:56 server id 1 end_log_pos 740 Query thread_id=3 exec_time=0 error_code=0   #结束时间
SET TIMESTAMP=1474811936/*!*/;
BEGIN
/*!*/;
# at 740
#160925 21:58:56 server id 1 end_log_pos 848 Query thread_id=3 exec_time=0 error_code=0
SET TIMESTAMP=1474811936/*!*/;
update ops.member set name='小二' where id=2      #执行的sql语句
/*!*/;
# at 848
#160925 21:58:56 server id 1 end_log_pos 875 Xid = 69   #开始执行的时间
COMMIT/*!*/;
# at 875
#160925 22:01:08 server id 1 end_log_pos 954 Query thread_id=3 exec_time=0 error_code=0    #结束时间
SET TIMESTAMP=1474812068/*!*/;
drop database ops
/*!*/;
# at 954
#160925 22:09:46 server id 1 end_log_pos 997 Rotate to mysql-bin.000004 pos: 4
DELIMITER ;
# End of log file
ROLLBACK /* added by mysqlbinlog */;
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;

恢复到更改“name='李四'”之前的数据
[root@vm-002 ~]# /usr/bin/mysqlbinlog --start-datetime="2016-09-25 21:57:19" --stop-datetime="2016-09-25 21:58:41" --database=ops /var/lib/mysql/mysql-bin.000003 | /usr/bin/mysql -uroot -p123456 -v ops

mysql> select * from member;
+----+-----------+-----+-----+---------+
| id | name | sex | age | classid |
+----+-----------+-----+-----+---------+
| 1 | wangshibo | m | 27 | cls1 |
| 2 | guohuihui | w | 27 | cls2 |
| 3 | yiyi | w | 20 | cls1 |
| 4 | xiaoer | m | 22 | cls3 |
| 5 | zhangsan | w | 21 | cls5 |
| 6 | lisi | m | 20 | cls4 |
| 7 | wangwu | w | 26 | cls6 |
+----+-----------+-----+-----+---------+
7 rows in set (0.00 sec)

[root@vm-002 ~]# /usr/bin/mysqlbinlog --start-datetime="2016-09-25 21:58:41" --stop-datetime="2016-09-25 21:58:56" --database=ops /var/lib/mysql/mysql-bin.000003 | /usr/bin/mysql -uroot -p123456 -v ops
mysql> select * from member;
+----+-----------+-----+-----+---------+
| id | name | sex | age | classid |
+----+-----------+-----+-----+---------+
| 1 | wangshibo | m | 27 | cls1 |
| 2 | guohuihui | w | 27 | cls2 |
| 3 | yiyi | w | 20 | cls1 |
| 4 | 李四 | m | 22 | cls3 |
| 5 | zhangsan | w | 21 | cls5 |
| 6 | lisi | m | 20 | cls4 |
| 7 | wangwu | w | 26 | cls6 |
+----+-----------+-----+-----+---------+
7 rows in set (0.00 sec)

[root@vm-002 ~]# /usr/bin/mysqlbinlog --start-datetime="2016-09-25 21:58:56" --stop-datetime="2016-09-25 22:01:08" --database=ops /var/lib/mysql/mysql-bin.000003 | /usr/bin/mysql -uroot -p123456 -v ops
mysql> select * from member;
+----+-----------+-----+-----+---------+
| id | name | sex | age | classid |
+----+-----------+-----+-----+---------+
| 1 | wangshibo | m | 27 | cls1 |
| 2 | 小二 | w | 27 | cls2 |
| 3 | yiyi | w | 20 | cls1 |
| 4 | 李四 | m | 22 | cls3 |
| 5 | zhangsan | w | 21 | cls5 |
| 6 | lisi | m | 20 | cls4 |
| 7 | wangwu | w | 26 | cls6 |
+----+-----------+-----+-----+---------+
7 rows in set (0.00 sec)

这样,就恢复了删除前的状态了!

总结:
所谓恢复,就是让mysql将保存在binlog日志中指定段落区间的sql语句逐个重新执行一次而已。

浏览294  评论0  Alex于 2017-9-6 11:43
发言