MySQL的高可用(MHA)

MySQL的高可用(MHA)

MHA简介

MHA:Master High Availability,对主节点进行监控，可实现自动故障转移至其余从节点；经过提高某一从节点为新的主节点，基于主从复制实现，还须要客户端配合实现，目前MHA主要支持一主二从，即一台充当master，一台充当备用master，另一台充当从数据库，出于机器成本的考虑，淘宝进行了改造，目前淘宝TMHA已经一主一从。

MHA架构

MHA的工做原理

MHA是由一台manager服务器远程监控主服务器,当主服务器挂了提高一台从服务器做为主服务器。 
当主节点挂了，manager首先要查看哪台从节点，同步的数据最多，而后提高同步最多的从节点为主节点，再将其他的MySQL服务器对他作从节点。 
若是原主节点没完全死透，manager会让新的主机经过ssh协议远程链接到原先的主节点，拉取二进制日志进行同步。若是主节死透了那就放弃。　　

---

MHA搭建

环境准备

1、准备4台主机，管理节点1台，主节点MySQL服务器1台，从节点MySQL服务器2台

主机	IP
Manager	192.168.73.111
Master	192.168.73.110
Slave1	192.168.73.112
Slave2	192.168.73.113

 

wget https://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm
rpm -ivh epel-release-latest-7.noarch.rpm

 

2、将Manager管理节点配置为时间服务器，向全部MySQL服务器提供时间同步。

1.安装chrony服务

[root@Manager ~]# yum install -y chrony

2.修改chrony配置文件

[root@Manager ~]# vim /etc/chrony.conf 
server 172.22.0.1 iburst
allow 192.168.0.0/16
local stratum 10

3.启动chrony服务

[root@Manager ~]# systemctl start chronyd

4.将MySQL服务器与Manager服务器进行时间同步
4.1在全部MySQL主机上修改配置文件并启动，并启动服务

[root@Master ~]# sed -i '/^server 0/i server 192.168.73.111 iburst' /etc/chrony.conf
[root@Master ~]# systemctl start chronyd

4.2确认时间同步

[root@Master ~]# chronyc sources -v
210 Number of sources = 1

  .-- Source mode  '^' = server, '=' = peer, '#' = local clock.
 / .- Source state '*' = current synced, '+' = combined , '-' = not combined,
| /   '?' = unreachable, 'x' = time may be in error, '~' = time too variable.
||                                                 .- xxxx [ yyyy ] +/- zzzz
||      Reachability register (octal) -.           |  xxxx = adjusted offset,
||      Log2(Polling interval) --.      |          |  yyyy = measured offset,
||                                \     |          |  zzzz = estimated error.
||                                 |    |           \
MS Name/IP address         Stratum Poll Reach LastRx Last sample               
===============================================================================
^* 192.168.73.111                4   6   377    54    +25us[  +41us] +/-  105ms

3、配置ssh为的密钥认证登录

当主节点宕机，manager会让从节点经过ssh协议去尝试链接主节点，并拉取二进制日志，因此要时用密钥的认证方式让从节点登录到主节点拉取数据。
1.在manager服务器上生成私钥文件

[root@Manager ~]# ssh-keygen 
Generating public/private rsa key pair.
Enter file in which to save the key (/root/.ssh/id_rsa): 
Enter passphrase (empty for no passphrase): 
Enter same passphrase again: 
Your identification has been saved in /root/.ssh/id_rsa.
Your public key has been saved in /root/.ssh/id_rsa.pub.
The key fingerprint is:
SHA256:yAvC2PJUlRyAf1udlrVXzmIsUljTdUdW6X6FVpQ3Ajo root@Manager
The key's randomart image is:
+---[RSA 2048]----+
|   ..ooo   ++. +%|
|  .  .o   o oo.=*|
|   ..    E = oo*o|
| + ...... B o B.+|
|o = ..ooS. . =...|
| + . ...       ..|
|  .   .         .|
|                 |
|                 |
+----[SHA256]-----+

2.将公钥文件复制给本身

[root@Manager ~]# ssh-copy-id 127.0.0.1

3.将整个~/.ssh目录复制给全部的MySQL主机

[root@Manager ~]# scp -r ~/.ssh 192.168.73.110:/root

至此全部环境准备完毕

1、配置主从复制

主节点配置

1.修改配置文件

[root@Master ~]# vim /etc/my.cnf
[mysqld]
server-id=1
log-bin
binlog-format=row
skip_name_resolve

2.启动数据库服务

[root@Master ~]# systemctl start mariadb

3.建立主从复制帐号

[root@Master ~]# mysql -e "GRANT REPLICATION SLAVE ON *.* TO 'repluser'@'192.168.73.%' IDENTIFIED BY 'centos';"

4.添加mha的管理帐号，让管理节点远程链接到主机用来设置主从调整

[root@Master ~]# mysql -e "GRANT ALL ON *.* TO 'mhauser'@'192.168.73.%' IDENTIFIED BY 'centos';"

从节点配置

1.修改配置文件

[root@Slave1 ~]# vim /etc/my.cnf
[mysqld]
server-id=2
read-only
log-bin
relay_log_purge=0
skip_name_resolve

2.启动服务

[root@Slave1 ~]# systemctl start mariadb

3.配置CHANGE MASTER TO

MariaDB [(none)]> CHANGE MASTER TO MASTER_HOST='192.168.73.110', MASTER_USER='repluser',MASTER_PASSWORD='centos',MASTER_PORT=3306,MASTER_LOG_FILE='mariadb-bin.000001',MASTER_LOG_POS=245;
Query OK, 0 rows affected (0.00 sec)

4.启动线程

MariaDB [(none)]> START SLAVE;
Query OK, 0 rows affected (0.00 sec)

在Slave2节点上也执行相同的操做，此处步骤省略，须要注意server-id须要修改成和其余主从节点不一样

5.测试 
主节点导入hellodb库

[root@Master ~]# mysql < hellodb_innodb.sql

从节点查看是否同步
slave1

[root@Slave1 ~]# mysql -e "SHOW DATABASES;"
+--------------------+
| Database           |
+--------------------+
| information_schema |
| hellodb            |
| mysql              |
| performance_schema |
| test               |
+--------------------+

Slave2

[root@Slave2 ~]# mysql -e "SHOW DATABASES;"
+--------------------+
| Database           |
+--------------------+
| information_schema |
| hellodb            |
| mysql              |
| performance_schema |
| test               |
+--------------------+

2、配置管理节点及被管理节点

1.在管理节上安装mha4mysql-manager、mha4mysql-node，将两个包放在同一目录下

[root@Manager ~]# yum install *.rpm -y  #这两个包有依赖管理须要一块儿安装

2.在全部被管理节点上安装mha4mysql-node

[root@Master ~]# yum install mha4mysql-node-0.56-0.el6.noarch.rpm -y

[root@Slave1 ~]# yum install mha4mysql-node-0.56-0.el6.noarch.rpm -y

[root@Slave2 ~]# yum install mha4mysql-node-0.56-0.el6.noarch.rpm -y

3.在管理节点上建立配置文件

[root@Manager ~]# vim /etc/mha/aap1.conf

[server default] 
user=mhauser 
password=centos
manager_workdir=/data/mastermha/app1/ 
manager_log=/data/mastermha/app1/manager.log 
remote_workdir=/data/mastermha/app1/ 
ssh_user=root 
repl_user=repluser 
repl_password=centos
ping_interval=1

[server1] 
hostname=192.168.73.110
candidate_master=1 
[server2] 
hostname=192.168.73.112
candidate_master=1 
[server3]
hostname=192.168.73.113
candidate_master=1

4.作检查
4.1检查ssh链接

[root@Manager ~]# masterha_check_ssh --conf=/etc/mha/aap1.conf

4.2检查主从复制

[root@Manager ~]# masterha_check_repl --conf=/etc/mha/aap1.conf

5.以上两项所有成功后启动程序
mha这个程序是跑在前台的，一次性的可使用nohub或screen来解决跑在前台的问题

[root@Manager ~]# masterha_manager --conf=/etc/mha/aap1.conf

3、测试

1.在master上跑个存储过程，导入存储过程

[root@Master ~]# mysql hellodb < testlog.sql

2.调用存储过程

MariaDB [(none)]> USE hellodb
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
MariaDB [hellodb]> call pro_testlog;

3.另起一个主节点窗口将主节点断网

[root@Master ~]# ifdown ens33

4.manager端完成切换退出，查看日志，查看新的主节点是哪台slave

[root@Manager app1]# tail /data/mastermha/app1/manager.log

Started automated(non-interactive) failover.
The latest slave 192.168.73.112(192.168.73.112:3306) has all relay logs for recovery.
Selected 192.168.73.112(192.168.73.112:3306) as a new master.
192.168.73.112(192.168.73.112:3306): OK: Applying all logs succeeded.
192.168.73.113(192.168.73.113:3306): This host has the latest relay log events.
Generating relay diff files from the latest slave succeeded.
192.168.73.113(192.168.73.113:3306): OK: Applying all logs succeeded. Slave started, replicating from 192.168.73.112(192.168.73.112:3306)
192.168.73.112(192.168.73.112:3306): Resetting slave info succeeded.
Master failover to 192.168.73.112(192.168.73.112:3306) completed successfully.
#此处显示最新的主节点为192.168.73.112

因为从节点在配置文件中定义的为read-only，此时被提高为主能执行写操做时应为管理服务器上有管理帐号，他将从节点的服务器全局变量read_only给关闭了

[root@Slave1 ~]# mysql -e "SELECT @@read_only;"
+-------------+
| @@read_only |
+-------------+
|           0 |
+-------------+

为了防止服务服务重启再次变为read-only，此时须要对新主节点的配置文件进行修改将read-only行注释

[mysqld]
server-id=2
#read-only
log-bin
relay_log_purge=0
skip_name_resolve

4、测试新的主节点

1.对hellodb.teachers表插入数据

[root@Slave1 ~]# mysql -e "INSERT hellodb.teachers VALUES(5,'Tang San',30,'M');"

2.Slave2主机上查看是否同步

[root@Slave2 ~]# mysql -e "SELECT * FROM hellodb.teachers;"
+-----+---------------+-----+--------+
| TID | Name          | Age | Gender |
+-----+---------------+-----+--------+
|   1 | Song Jiang    |  45 | M      |
|   2 | Zhang Sanfeng |  94 | M      |
|   3 | Miejue Shitai |  77 | F      |
|   4 | Lin Chaoying  |  93 | F      |
|   5 | Tang San      |  30 | M      |    #已经同步
+-----+---------------+-----+--------+

其余事项

当原主节点被修复后，将其添加为从节点使用。