028：基于mysqldump备份脚本

MySQL Backup and Recovery

一 MySQL Backup

1.功能

mysqldump全量和增量备份，通过最近一次备份刷新产生binlog来定位执行增量。

• 脚本下载地址

  • github

• 场景一：

  每天执行一次全量备份.

• 场景二：

  每周日执行一次全量备份，然后每天3点执行增量备份.

• 应用场景：

  • 增量备份在周一到周六凌晨3点，会使用mysqlbinlog 导出sql并使用gzip压缩到指定目录

    • mysqlbinlog -vv binlog.000044 binlog.000045 binlog.000046 ..... > |gzip > $INCR_BACKUP_DIR/incr.sql.gz

  • 全量备份则使用mysqldump将所有的数据库导出，每周日凌晨3点执行，并会删除N天之前的目录和文件。参数如下：

    • MYSQLDUMP_OPTION=' --add-drop-table --single-transaction --master-data=2 -E -R --flush-logs --databases'

  • 删除命令

    • (find $BASE_DIR -mtime + $DELETE_DAYS -type d -name "full*" -exec rm -rf {} ;)

2.使用方法

• 脚本需修改参数：

MY_USER="gcdb"                      --备份帐号
MY_PASSWORD="iforgot"               --备份密码
MY_IP="192.168.49.247"              --本机ip，例如从库ip
MY_MASTER_USER="gcdb"               --master帐号
MY_MASTER_PASSWORD="iforgot"        --master密码
MY_MASTER_IP="192.168.49.246"       --指主库ip
BINLOG_FILE=/r2/mysqldata           --binlog文件所在的目录,增量时需要用到
BASE_DIR=/mybak                     --备份基础目录
DELETE_DAYS=15                      --备份保存天数，即删除N天之前的备份，例如一周一个全备、每一天一个增量，该值必须大于配置为7，
FILTER="information_schema|test|mysql|sys|performance_schema" --过滤指定数据库，也就是不备份的数据库（注意mysql库）

• 备份基础目录以/mybak为例，目录的树形结构如下：

[root@node02 scripts]# tree /mybak/
/mybak/
├── full
│   ├── full_20180419
│   │   ├── backup.log
│   │   ├── dbname
│   │   ├── fullbak.sql.gz
│   │   ├── grants.sql
│   │   ├── master_grants.sql
│   │   ├── master_users.sql
│   │   ├── position
│   │   └── users.sql
│   └── full_20180420
│       ├── backup_full.log
│       ├── dbname
│       ├── fullbak.sql.gz
│       ├── grants.sql
│       ├── master_grants.sql
│       ├── master_users.sql
│       ├── position
│       └── users.sql
├── incr
│   ├── incr_20180420130001
│   │   ├── backup_incr.log
│   │   ├── incr.sql.gz
│   │   ├── new_binlogs_list
│   │   ├── old_binlogs_list
│   │   └── tmp_binlog_name
│   └── incr_20180420140001
│       ├── backup_incr.log
│       ├── incr.sql.gz
│       ├── new_binlogs_list
│       ├── old_binlogs_list
│       └── tmp_binlog_name
├── public_backup.log    --记录备份是否成功
└── public_position      --保存最新binlog文件名

2.1 全备

• 备份命令

  • ./bak_mysql.sh full

• 计划任务

  • crontab -e
  • 每天做一次全备，凌晨3点进行全量备份，备份频率可根据项目情况自行调整。
  • 0 3 * * * /bin/sh /scripts/bak_mysql.sh full >/dev/null 2>&1

2.2 增量

• 备份命令

  • ./bak_mysql.sh incr

• 计划任务

  • crontab -e
  • 每个小时(除3点外)进行binglog增量备份,备份频率可根据项目情况自行调整。
  • 0 0-2,4-23 * * * /bin/sh /scripts/bak_mysql.sh incr >/dev/null 2>&1

• 使用参考如下：

+-----------------------------------------------------------------------------+
|Usage : ./bak_mysql.sh  (full|incr|oemu)                                     |
+-----------------------------------------------------------------------------+
|全备              ：./bak_mysql.sh full                                      |
|增量              ：./bak_mysql.sh incr                                      |
|只导出master权限  ：./bak_mysql.sh oemu                                      |
+-----------------------------------------------------------------------------+
计划任务参考
+-----------------------------------------------------------------------------+
|全备  ：30 0 * * *  /bin/sh  /scripts/bak_mysql.sh full >/dev/null 2>&1      |
|增量  ：30 2-23/2 * * *  /bin/sh  /scripts/bak_mysql.sh incr >/dev/null 2>&1 |
+-----------------------------------------------------------------------------+
[root@node02 scripts]#

3.执行备份

3.1 全备执行过程

[root@node01 scripts]# sh bak_mysql.sh full   2>/dev/null
+------------------+
| Backup_Host      |
+------------------+
| node01.mysql.com |
+------------------+
Backup_Host 连接正常
+------------------+
| MY_Host          |
+------------------+
| node01.mysql.com |
+------------------+
192.168.49.245开始导出帐号和权限信息
192.168.49.245成功导出 10 个用户权限
192.168.49.245成功导出 10 个用户帐号
1、20180425 16:51:25 开始备份......
2、备份以下数据库：
 mysql percona
3、20180425 16:51:26 备份成功......
4、备份用时: 1 秒
5、备份数据量大小: 6.9M
6、记录最新的binlog文件名!
+-------------+
| Master_Host |
+-------------+
| slave7      |
+-------------+
master 192.168.101.137开始导出帐号和权限信息
master 192.168.101.137成功导出 9 个用户权限
master 192.168.101.137成功导出 9 个用户帐号
全备成功
[root@node01 scripts]#

3.2 全备执行结果

[root@node01 scripts]# cat /mybak/public_position
binlog.000023

[root@node01 scripts]# cat /mybak/public_backup.log
Backup_Host 连接正常
全备成功
删除 /mybak/full 目录下 7 天之前的备份!
full_bakcup_ok

[root@node01 scripts]# tree /mybak/full/full_20180425/
/mybak/full/full_20180425/
├── backup_full.log
├── dbname            --备份的库名
├── fullbak.sql.gz    --备份文件
├── grants.sql        --本机授权文件(mysql5.7之后权限和帐号分开)
├── master_grants.sql --master授权文件
├── master_users.sql  --master帐号文件
├── position          --GTID和binlog文件名信息
└── users.sql         --本机授权文件(mysql5.7之后权限和帐号分开)

[root@node02 scripts]# cat /mybak/full/full_20180420/position
-- GTID state at the beginning of the backup
SET @@GLOBAL.GTID_PURGED='7debec7f-4797-11e8-9274-0050569d16ce:1-3,
-- CHANGE MASTER TO MASTER_LOG_FILE='binlog.000023', MASTER_LOG_POS=234;

[root@node01 scripts]# cat /mybak/full/full_20180425/position
-- GTID state at the beginning of the backup
SET @@GLOBAL.GTID_PURGED='7debec7f-4797-11e8-9274-0050569d16ce:1-3,
-- CHANGE MASTER TO MASTER_LOG_FILE='binlog.000023', MASTER_LOG_POS=234;

[root@node01 scripts]# cat /mybak/full/full_20180425/backup_full.log
1、20180425 16:51:25 开始备份......
2、备份以下数据库：
 mysql percona
3、20180425 16:51:26 备份成功......
4、备份用时: 1 秒
5、备份数据量大小: 6.9M
6、记录最新的binlog文件名!
binlog.000023
master 192.168.101.137开始导出帐号和权限信息
master 192.168.101.137成功导出 9 个用户权限
master 192.168.101.137成功导出 9 个用户帐号
[root@node01 scripts]#

3.3 增备执行结果

• 执行增量备份之前进行如下操作：

[2018-04-20 15:32:17.838][192.168.49.247-node02][000220][MYSQL]
UPDATE `ttt`.`t1` SET `name` = 'rrrrrssss' WHERE `id` = 3
Time: 0.001s

[2018-04-20 17:10:02.925][192.168.49.246-mycat][016413][MYSQL]
UPDATE `ttt`.`t1` SET `name` = 'xiaowen' WHERE `id` = 3
Time: 0.002s

[2018-04-20 17:11:42.657][192.168.49.246-mycat][016413][MYSQL]
insert into t1 values(8,'xiaomi')
Time: 0.001s

执行中

[root@node02 scripts]# ./bak_mysql.sh incr  2>/dev/null
+------------------+
| Backup_Host      |
+------------------+
| node02.mysql.com |
+------------------+
mysql连接正常
创建INCR_BACKUP_DIR目录
/mybak/incr/incr_20180420171334
创建/mybak/incr/incr_20180420171334/backup_incr.log
000051 : PUBLIC_POSITION 有获取到数值
循环写入binlog名执行成功
mysqlbinlog 执行成功......
写入最新的binlog名到公共文件中
增量备份成功

删除 /mybak/incr 目录下 7 天之前的备份!

3.4 增备执行结果

[root@node02 scripts]# ll /mybak/incr/incr_20180420171334
total 20
-rw-r--r-- 1 root root 2470 Apr 20 17:13 backup_incr.log
-rw-r--r-- 1 root root 1488 Apr 20 17:13 incr.sql.gz
-rw-r--r-- 1 root root  728 Apr 20 17:13 new_binlogs_list
-rw-r--r-- 1 root root  714 Apr 20 17:13 old_binlogs_list
-rw-r--r-- 1 root root   14 Apr 20 17:13 tmp_binlog_name

[root@node02 scripts]# cat /mybak/incr/incr_20180420171334/backup_incr.log
不需要备份，后缀为 000001 binlog文件
不需要备份，后缀为 000002 binlog文件
---------省略---------
不需要备份，后缀为 000049 binlog文件
不需要备份，后缀为 000050 binlog文件  --全量备份到binlog.000050，flush logs生成了binlog.000051
需备份后缀为 000051 binlog文件
mysqlbinlog 执行成功......

[root@node02 scripts]# cat /mybak/public_backup.log
Backup_Host 连接正常
创建 /mybak/incr/incr_20180420171334/backup_incr.log
incr_bakcup_ok
增量备份成功
删除 /mybak/incr 目录下 7 天之前的备份!

[root@node02 scripts]# cat /mybak/incr/incr_20180420153720/tmp_binlog_name
binlog.000051

[root@node02 scripts]# gunzip <  /mybak/incr/incr_20180420171334/incr.sql.gz |more
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=1*/;
/*!50003 SET @OLD_COMPLETION_TYPE=@@COMPLETION_TYPE,COMPLETION_TYPE=0*/;
DELIMITER /*!*/;
# at 4
#180420 17:02:35 server id 49247  end_log_pos 123 CRC32 0xdf164269 	Start: binlog v 4, server v 5.7.18-log created 180420
17:02:35
BINLOG '
q6zZWg9fwAAAdwAAAHsAAAAAAAQANS43LjE4LWxvZwAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
AAAAAAAAAAAAAAAAAAAAAAAAEzgNAAgAEgAEBAQEEgAAXwAEGggAAAAICAgCAAAACgoKKioAEjQA
AWlCFt8=
'/*!*/;
# at 123
#180420 17:02:35 server id 49247  end_log_pos 234 CRC32 0x3cf6d5c4 	Previous-GTIDs
# 8a5dd931-42cc-11e8-aa39-0050569dc4ab:1-4,
# fda7506d-33ea-11e8-b187-000c298b03f2:28759-28763
# at 234
#180420 17:09:48 server id 49246  end_log_pos 299 CRC32 0xe827d4c9 	GTID	last_committed=0	sequence_number=1
SET @@SESSION.GTID_NEXT= 'fda7506d-33ea-11e8-b187-000c298b03f2:28764'/*!*/;
# at 299
#180420 17:09:48 server id 49246  end_log_pos 362 CRC32 0x52db9f29 	Query	thread_id=16413	exec_time=4294967295	error_
code=0
SET TIMESTAMP=1524215388/*!*/;
SET @@session.pseudo_thread_id=16413/*!*/;
SET @@session.foreign_key_checks=1, @@session.sql_auto_is_null=0, @@session.unique_checks=1, @@session.autocommit=1/*!*/;
SET @@session.sql_mode=524288/*!*/;
SET @@session.auto_increment_increment=1, @@session.auto_increment_offset=1/*!*/;
/*!C utf8mb4 *//*!*/;
SET @@session.character_set_client=45,@@session.collation_connection=45,@@session.collation_server=192/*!*/;
SET @@session.lc_time_names=0/*!*/;
SET @@session.collation_database=DEFAULT/*!*/;
BEGIN
/*!*/;
# at 362
#180420 17:09:48 server id 49246  end_log_pos 409 CRC32 0x330344f1 	Table_map: `ttt`.`t1` mapped to number 100172
# at 409
#180420 17:09:48 server id 49246  end_log_pos 473 CRC32 0x7a37b361 	Update_rows: table id 100172 flags: STMT_END_F

BINLOG '
XK7ZWhNewAAALwAAAJkBAAAAAEyHAQAAAAEAA3R0dAACdDEAAgMPAjwAAvFEAzM=
XK7ZWh9ewAAAQAAAANkBAAAAAEyHAQAAAAEAAgAC///8AwAAAAlycnJycnNzc3P8AwAAAAd4aWFv
d2VuYbM3eg==
'/*!*/;
### UPDATE `ttt`.`t1`
### WHERE
###   @1=3 /* INT meta=0 nullable=0 is_null=0 */
###   @2='rrrrrssss' /* VARSTRING(60) meta=60 nullable=1 is_null=0 */
### SET
###   @1=3 /* INT meta=0 nullable=0 is_null=0 */
###   @2='xiaowen' /* VARSTRING(60) meta=60 nullable=1 is_null=0 */
# at 473
#180420 17:09:48 server id 49246  end_log_pos 504 CRC32 0x3b8f41b5 	Xid = 397443
COMMIT/*!*/;
# at 504
#180420 17:09:48 server id 49246  end_log_pos 569 CRC32 0xc05fd7e0 	GTID	last_committed=1	sequence_number=2
SET @@SESSION.GTID_NEXT= 'fda7506d-33ea-11e8-b187-000c298b03f2:28765'/*!*/;
# at 569
#180420 17:09:48 server id 49246  end_log_pos 632 CRC32 0x03de76ae 	Query	thread_id=16413	exec_time=4294967295	error_
code=0
SET TIMESTAMP=1524215388/*!*/;
BEGIN
/*!*/;
# at 632
#180420 17:09:48 server id 49246  end_log_pos 679 CRC32 0xb60b3faa 	Table_map: `ttt`.`t1` mapped to number 100172
# at 679
#180420 17:09:48 server id 49246  end_log_pos 726 CRC32 0xa2092f94 	Write_rows: table id 100172 flags: STMT_END_F

BINLOG '
XK7ZWhNewAAALwAAAKcCAAAAAEyHAQAAAAEAA3R0dAACdDEAAgMPAjwAAqo/C7Y=
XK7ZWh5ewAAALwAAANYCAAAAAEyHAQAAAAEAAgAC//wFAAAABnhpYW9taZQvCaI=
'/*!*/;
### INSERT INTO `ttt`.`t1`
### SET
###   @1=5 /* INT meta=0 nullable=0 is_null=0 */
###   @2='xiaomi' /* VARSTRING(60) meta=60 nullable=1 is_null=0 */
# at 726
#180420 17:09:48 server id 49246  end_log_pos 757 CRC32 0xf7d054a7 	Xid = 397445
COMMIT/*!*/;
# at 757
#180420 17:11:28 server id 49246  end_log_pos 822 CRC32 0xb24fe15b 	GTID	last_committed=2	sequence_number=3
SET @@SESSION.GTID_NEXT= 'fda7506d-33ea-11e8-b187-000c298b03f2:28766'/*!*/;
# at 822
#180420 17:11:28 server id 49246  end_log_pos 885 CRC32 0xdb150b55 	Query	thread_id=16413	exec_time=4294967295	error_
code=0
SET TIMESTAMP=1524215488/*!*/;
BEGIN
/*!*/;
# at 885
#180420 17:11:28 server id 49246  end_log_pos 932 CRC32 0xf91a1b61 	Table_map: `ttt`.`t1` mapped to number 100172
# at 932
#180420 17:11:28 server id 49246  end_log_pos 979 CRC32 0xafce0c68 	Write_rows: table id 100172 flags: STMT_END_F

BINLOG '
wK7ZWhNewAAALwAAAKQDAAAAAEyHAQAAAAEAA3R0dAACdDEAAgMPAjwAAmEbGvk=
wK7ZWh5ewAAALwAAANMDAAAAAEyHAQAAAAEAAgAC//wIAAAABnhpYW9taWgMzq8=
'/*!*/;
### INSERT INTO `ttt`.`t1`
### SET
###   @1=8 /* INT meta=0 nullable=0 is_null=0 */
###   @2='xiaomi' /* VARSTRING(60) meta=60 nullable=1 is_null=0 */
# at 979
#180420 17:11:28 server id 49246  end_log_pos 1010 CRC32 0x6e2f666e 	Xid = 397447
COMMIT/*!*/;
# at 1010
#180420 17:13:34 server id 49247  end_log_pos 1054 CRC32 0x02255fe1 	Rotate to binlog.000052  pos: 4
SET @@SESSION.GTID_NEXT= 'AUTOMATIC' /* added by mysqlbinlog */ /*!*/;
DELIMITER ;
# End of log file
/*!50003 SET COMPLETION_TYPE=@OLD_COMPLETION_TYPE*/;
/*!50530 SET @@SESSION.PSEUDO_SLAVE_MODE=0*/;
[root@node02 scripts]#

3.5 public_position文件是空时，执行增备

• 如果public_position文件是空的，就会从新执行全备

[root@node02 scripts]# > /mybak/public_position    --清空文件

[root@node02 scripts]# ./bak_mysql.sh incr 2>/dev/null
+------------------+
| Backup_Host      |
+------------------+
| node02.mysql.com |
+------------------+
mysql连接正常
创建INCR_BACKUP_DIR目录
/mybak/incr/incr_20180420172634
/mybak/incr/incr_20180420172634/backup_incr.log 不存在，重新创建.
OLD_NUM : PUBLIC_POSITION 没有获取到数值,执行全备                          --这里执行全备
0、(1)成功导出 7 个用户权限
0、(2)成功导出 7 个用户帐号
1、20180420 17:26:35 开始备份......
2、备份以下数据库：
 cmd  ttt
3、20180420 17:28:05 备份成功......
4、备份用时: 90 秒
5、备份数据量大小: 256M
6、记录最新的binlog文件名!
开始导出master帐号和权限信息
+-------------------+
| Master_Host       |
+-------------------+
| mycat01.mysql.com |
+-------------------+
0、(1)master成功导出 8 个用户权限
0、(2)master成功导出 8 个用户帐号
全备执行成功
增量备份失败
增量失败,删除备份目录  --注意：删除刚才创建的增量

[root@node02 scripts]# ll /mybak/incr/incr_20180420172634    --增量目录，显示已被删除
ls: cannot access /mybak/incr/incr_20180420172634: No such file or directory

二 MySQL Recovery

恢复使用全备进行恢复

1.全备目录

[root@node02 scripts]# ls -l /mybak/full/full_20180424/
total 258336
-rw-r--r-- 1 root root       526 Apr 24 12:01 backup_full.log
-rw-r--r-- 1 root root       176 Apr 24 12:00 dbname
-rw-r--r-- 1 root root 264506232 Apr 24 12:01 fullbak.sql.gz
-rw-r--r-- 1 root root      1134 Apr 24 12:00 grants.sql
-rw-r--r-- 1 root root      1200 Apr 24 12:01 master_grants.sql
-rw-r--r-- 1 root root      1630 Apr 24 12:01 master_users.sql
-rw-r--r-- 1 root root       187 Apr 24 12:01 position
-rw-r--r-- 1 root root      1438 Apr 24 12:00 users.sql
[root@node02 scripts]#

2.GTID 模式下恢复

• 恢复从库再重做主从

GTID模式在原机上重做从库，需要reset master，清空master，再导入

• 步骤1

(root@localhost) 16:21:27 [(none)]> stop slave;
Query OK, 0 rows affected, 1 warning (0.00 sec)

(root@localhost) 16:21:58 [(none)]> reset master;
Query OK, 0 rows affected (0.16 sec)

• 步骤2

[root@node02 full_20180424]# cat position
-- GTID state at the beginning of the backup
SET @@GLOBAL.GTID_PURGED='84865d81-b573-11e7-9668-b8ca3a65693c:1-57436835';
-- CHANGE MASTER TO MASTER_LOG_FILE='mysql-bin.000079', MASTER_LOG_POS=194;

[root@node02 full_20180424]# gunzip <fullbak.sql.gz |mysql -uroot -pxxxxxxx
mysql: [Warning] Using a password on the command line interface can be insecure.

• 步骤3 CHANGE MASTER

  • GTID模式执行
    • CHANGE MASTER TO MASTER_HOST='192.168.xxx.xxx',MASTER_USER='repl',MASTER_PASSWORD='XXXXX',MASTER_AUTO_POSITION=1;
  • 非GTID模式执行
    • CHANGE MASTER TO MASTER_LOG_FILE='mysql-bin.000079', MASTER_LOG_POS=194,MASTER_HOST='192.168.101.137',MASTER_PORT=3306,MASTER_USER='repl',MASTER_PASSWORD='XXXXX';

(root@localhost) 16:53:09 [(none)]> CHANGE MASTER TO MASTER_HOST='192.168.101.137',MASTER_USER='repl',MASTER_PASSWORD='XXXXX',MASTER_AUTO_POSITION=1;
Query OK, 0 rows affected, 2 warnings (0.31 sec)

(root@localhost) 16:53:27 [(none)]> start slave ;
Query OK, 0 rows affected (0.00 sec)

(root@localhost) 16:53:34 [(none)]> show slave status G;
*************************** 1. row ***************************
               Slave_IO_State: Queueing master event to the relay log
                  Master_Host: 192.168.101.137
                  Master_User: repl
                  Master_Port: 3306
                Connect_Retry: 60
              Master_Log_File: mysql-bin.000081
          Read_Master_Log_Pos: 348638
               Relay_Log_File: node01-relay-bin.000002
                Relay_Log_Pos: 869569
        Relay_Master_Log_File: mysql-bin.000079
             Slave_IO_Running: Yes
            Slave_SQL_Running: Yes
              Replicate_Do_DB:
          Replicate_Ignore_DB:
           Replicate_Do_Table:
       Replicate_Ignore_Table:
      Replicate_Wild_Do_Table:
  Replicate_Wild_Ignore_Table:
                   Last_Errno: 0
                   Last_Error:
                 Skip_Counter: 0
          Exec_Master_Log_Pos: 869396
              Relay_Log_Space: 29714750
              Until_Condition: None
               Until_Log_File:
                Until_Log_Pos: 0
           Master_SSL_Allowed: No
           Master_SSL_CA_File:
           Master_SSL_CA_Path:
              Master_SSL_Cert:
            Master_SSL_Cipher:
               Master_SSL_Key:
        Seconds_Behind_Master: 49801
Master_SSL_Verify_Server_Cert: No
                Last_IO_Errno: 0
                Last_IO_Error:
               Last_SQL_Errno: 0
               Last_SQL_Error:
  Replicate_Ignore_Server_Ids:
             Master_Server_Id: 1
                  Master_UUID: 84865d81-b573-11e7-9668-b8ca3a65693c
             Master_Info_File: /r2/mysqldata/master.info
                    SQL_Delay: 0
          SQL_Remaining_Delay: NULL
      Slave_SQL_Running_State: update
           Master_Retry_Count: 86400
                  Master_Bind:
      Last_IO_Error_Timestamp:
     Last_SQL_Error_Timestamp:
               Master_SSL_Crl:
           Master_SSL_Crlpath:
           Retrieved_Gtid_Set: 84865d81-b573-11e7-9668-b8ca3a65693c:57436836-57469553
            Executed_Gtid_Set: 84865d81-b573-11e7-9668-b8ca3a65693c:1-57437364
                Auto_Position: 1
         Replicate_Rewrite_DB:
                 Channel_Name:
           Master_TLS_Version:
1 row in set (0.26 sec)

ERROR:
No query specified

3.只导出master用户帐号和权限

[root@node02 scripts]# sh bak_mysql.sh oemu 2>/dev/null
+------------------+
| Backup_Host      |
+------------------+
| node02.mysql.com |
+------------------+
mysql连接正常
master 192.168.49.246开始导出帐号和权限信息
+-------------------+
| Master_Host       |
+-------------------+
| mycat01.mysql.com |
+-------------------+
master 192.168.49.246成功导出 8 个用户权限
master 192.168.49.246成功导出 8 个用户帐号
master 192.168.49.246导出用户帐号和权限成功
[root@node02 scripts]# ll /mybak/
total 20
drwxr-xr-x  4 root root   48 Apr 24 13:46 full
drwxr-xr-x 27 root root 4096 Apr 24 13:46 incr
-rw-r--r--  1 root root 1200 Apr 25 08:40 master_grants.sql  --导出权限
-rw-r--r--  1 root root 1630 Apr 25 08:40 master_users.sql   --导出帐号
-rw-r--r--  1 root root  232 Apr 25 08:40 public_backup.log
-rw-r--r--  1 root root   17 Apr 24 13:46 public_position
[root@node01 scripts]#

4.恢复主库

从库做全备在主库上恢复(步骤同上1-2)，创建帐号和权限（master_users.sql，master_grants.sql）,刷新权限(flush privilegs;),然后做全备，再做从库