阿里云RDS数据库利用物理备份搭建从库

（一）、背景介绍
由于数据库采用阿里云的RDS数据库，当所有的读写都在一台服务器上的时候，出现了CPU负载较高，尤其在业务时间段内，并发的访问较多，造成业务压力比较大，经询问可以采用主从复制，RDS数据库只用于写，其他的本地安装的mysql数据库用于读，可以有多台，后台的程序还可以做读写分离，来减轻RDS数据库的压力负载。

安装条件：

1. 本地自建Mysql数据必须和云mysql的大版本保持一致
2. 64位的Linux系统
3. MySQL 5.6及之前（含5.5）的版本需要安装 Percona XtraBackup 2.3 ； MySQL 5.7版本需要安装 Percona XtraBackup2.4
4. 为了快速恢复，本地mysql安装都采用centos的yum安装方式
   5.注意下面没有说明哪个版本执行的就是都需要执行的
   6.搭建复制环境前需要保证ECS能够访问的通RDS，建议用RDS内网地址进行复制搭建
   一般来说我们的数据库版本：mysql 5.6

（二）、安装配置
1、mysql数据库的安装和配置（略），具体参考https://blog.51cto.com/liqingbiao/1692315
2、替换/etc/my.cnf配置内容，其中跳过权限表进行启动

[appuser@inside-nginx152245 ~]$ cat /etc/my.cnf
[mysqld]
innodb_checksum_algorithm=crc32
innodb_data_file_path=ibdata1:200M:autoextend
innodb_log_files_in_group=2
innodb_log_file_size=524288000
innodb_undo_directory=/usr/local/mysql/
basedir=/usr/local/mysql
datadir=/data/mysql
innodb_undo_tablespaces=0
server_id=999999998
log_bin=mysql-log
gtid_mode=on
enforce_gtid_consistency=on
log-slave-updates=1
relay_log=relay-log
sql_mode=''
binlog_format=row
slave_skip_errors = 1062,1032
skip-grant-tables=1

#######注意server_id可以自己改，务必不要和RDS的一样了

2、安装配置Percona
1、安装Percona 仓库
yum install https://repo.percona.com/yum/percona-release-latest.noarch.rpm
2、查看percona库
[appuser@predb ~]$ yum list|grep percona
3、安装percona-xtrabackup-24包
yum install percona-xtrabackup-24

（二）、rpm 安装

2.1、下载percona rpm包
wget https://www.percona.com/downloads/XtraBackup/Percona-XtraBackup-2.4.4/binary/redhat/7/x86_64/percona-xtrabackup-24-2.4.4-1.el7.x86_64.rpm

2.2、通过yum localinstall 进行安装
yum localinstall percona-xtrabackup-24-2.4.4-1.el7.x86_64.rpm    #####5.6版本安装这个
 yum -y install percona-xtrabackup-24.x86_64   #####5.7版本需要安装这个

 2.3、percona卸载
yum remove percona-xtrabackup

3、下载备份文件并恢复。登录云服务器ECS，执行下载命令。wget -c '<数据备份文件外网下载地址>' -O <自定义文件名>.tar.gz -c启用断点续传模式，-O将下载结果保存指定文件。

 [root@inside-nginx152245 data]#wget -c 'http://rxxxxxxxxxxxxxxxxxxxx/custins4543777/hins6010151_data_20190621021827.tar.gz?OSSAccessKeyId=LTAIyKzxtSYNknVO&Expires=1561261206&Signature=KgiZPWl9YlX5kOX0WChx3h6Yw18%3D' -O mysql20190621.tar.gz
[root@inside-nginx152245 data]# tar xf mysql20190621.tar.gz -C mysql
[root@inside-nginx152245 data]# innobackupex --defaults-file=/etc/my.cnf --apply-log /data/mysql20190402/
190621 16:16:41 innobackupex: Starting the apply-log operation

IMPORTANT: Please check that the apply-log run completes successfully.
           At the end of a successful apply-log run innobackupex
           prints "completed OK!".
[root@inside-nginx152245 data]# chown -R mysql:mysql /data/mysql        
[root@inside-nginx152245 data]# /etc/init.d/mysqld restart
Shutting down MySQL....                                    [  OK  ]
Starting MySQL.........                                    [  OK  ]

4、直接进入mysql更新root密码

mysql> update user set password=PASSWORD('xxxxx') where user='root';
Query OK, 2 rows affected (0.00 sec)
Rows matched: 2  Changed: 2  Warnings: 0
mysql> flush privileges;
Query OK, 0 rows affected (0.00 sec)
##############一般使用##########
mysql>update user set host='localhost' where host='::1';
mysql>update user set host='%' where host='127.0.0.1';

5、删除/etc/my.cnf里面的skip-grant-tables=1，然后重启

[root@inside-nginx152245 data]# vim /etc/my.cnf
[root@inside-nginx152245 data]# /etc/init.d/mysqld restart
Shutting down MySQL..                                      [  OK  ]
Starting MySQL.                                            [  OK  ]

6、通过root进行登录，执行如下操作。

mysql>use mysql;                                   ## 5.6执行
mysql>drop table slave_master_info;     ### 5.6执行
mysql>drop table slave_relay_log_info;  ### 5.6执行
mysql>drop table slave_worker_info;      ###5.6执行
mysql>drop table innodb_index_stats;    ### 5.6执行 
mysql>drop table innodb_table_stats;     ###5.6执行
mysql>source  /usr/local/mysql/share/mysql_system_tables.sql;  ##5.6执行，
mysql>exit
/etc/init.d/mysqld restart      ###########5.6版本必须重启，切记执行

备注：如果yum安装的话，可以找安装路径下mysql_system_tables.sql这个文件一般都在share目录下。mysql> source /tmp/mysql-5.6.43-linux-glibc2.12-x86_64/share/mysql_system_tables.sql      #######5.6版本执行。后边的路径是mysql安装的路径

7、查看xtrbackup_slave_info里边的set global gtid_purged语句，把这个copy到后面复制用

[root@inside-nginx152245 share]# cat /data/mysql/xtrabackup_slave_info
SET GLOBAL gtid_purged='5b61ae19-a34d-11e7-810f-5cb901891368:1-3823344, 70f8ce71-31fb-11e8-a373-7cd30ae014ac:1-4846380, 7302d9cf-a34d-11e7-810f-244c075fefa9:1-2828134';
CHANGE MASTER TO MASTER_AUTO_POSITION=1;

3、在数据库中依次执行如下语句

mysql> reset master
    -> ;
Query OK, 0 rows affected (0.02 sec)

mysql> SET GLOBAL gtid_purged='5b61ae19-a34d-11e7-810f-5cb901891368:1-3823344, 70f8ce71-31fb-11e8-a373-7cd30ae014ac:1-4846380, 7302d9cf-a34d-11e7-810f-244c075fefa9:1-2828134';
Query OK, 0 rows affected (0.02 sec)

mysql> reset slave;
ERROR 1198 (HY000): This operation cannot be performed with a running slave; run STOP SLAVE first
mysql> stop slave;
Query OK, 0 rows affected (0.00 sec)

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

mysql>  change master to master_host ='rm-uf64n5y6yet9b22s1.mysql.rds.aliyuncs.com',master_port=3306,master_user='dowadmin',master_password="70537449c232!dow",master_auto_position=1;
Query OK, 0 rows affected, 2 warnings (0.04 sec)
#############注意:上面的master_host  master_port   master_user  master_password 根据RDS主库的信息各自填写自己的信息，不要照着上面执行；
mysql> start slave;
Query OK, 0 rows affected (0.02 sec)

mysql> show slave status \G;
*************************** 1. row ***************************
               Slave_IO_State: Waiting for master to send event
                  Master_Host: rm-uf64n5y6yet9b22s1.mysql.rds.aliyuncs.com
                  Master_User: dowadmin
                  Master_Port: 3306
                Connect_Retry: 60
              Master_Log_File: mysql-bin.001807
          Read_Master_Log_Pos: 906900
               Relay_Log_File: relay-log.000004
                Relay_Log_Pos: 431601
        Relay_Master_Log_File: mysql-bin.001805
             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: 431471
              Relay_Log_Space: 20108553
              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: 35647
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: 459412040
                  Master_UUID: 70f8ce71-31fb-11e8-a373-7cd30ae014ac
             Master_Info_File: /data/mysql/master.info
                    SQL_Delay: 0
          SQL_Remaining_Delay: NULL
      Slave_SQL_Running_State: Reading event from the relay log
           Master_Retry_Count: 86400
                  Master_Bind:
      Last_IO_Error_Timestamp:
     Last_SQL_Error_Timestamp:
               Master_SSL_Crl:
           Master_SSL_Crlpath:
           Retrieved_Gtid_Set: 70f8ce71-31fb-11e8-a373-7cd30ae014ac:4846381-4867752
            Executed_Gtid_Set: 5b61ae19-a34d-11e7-810f-5cb901891368:1-3823344,
70f8ce71-31fb-11e8-a373-7cd30ae014ac:1-4848839,
7302d9cf-a34d-11e7-810f-244c075fefa9:1-2828134
                Auto_Position: 1
1 row in set (0.00 sec)

ERROR:
No query specified
#####搭建完成show slave status\G看到的结果应该如下，slave io/sql线程都是YES状态：

如有问题，请查看阿里云rds数据库文档（https://yq.aliyun.com/articles/689032?spm=a2c4e.11153940.0.0.48991670fK7Lg2）