目录

一 数据库架构图

二 机器配置

三 mysql主主配置

四 Keepalived高可用配置

一 数据库架构图

二 机器配置

master1 172.16.90.180CentOS 7.6Keepalived VIP：172.16.90.186

master2 172.16.90.181 CentOS 7.6Keepalived VIP：172.16.90.186

三 mysql主主配置

首先在两个节点上都下载mysql:

#获取5.7安装包

wget https://dev.mysql.com/get/Downloads/MySQL-5.7/mysql-5.7.36-el7-x86_64.tar.gz

#安装mysql

mkdir -p /app/

tar -xvf mysql-5.7.36-el7-x86_64.tar.gz -C /app/

ln -s /app/mysql-5.7.36-el7-x86_64/ /app/mysql

(ln -s命令可通过linux中ln -s 命令详解_坚持就有希望的博客-CSDN博客_ln- s,详细了解)

cd /app/mysql/

#创建MySQL用户

useradd mysql

#创建数据目录

mkdir /data/mysql -p

#目录授权

chown mysql.mysql -R /data/mysql

chown mysql.mysql -R /app/mysql

#配置环境变量

vim /etc/profile

#末尾添加这行内容

export PATH=/app/mysql/bin:$PATH

source /etc/profile

#初始化数据库

mysqld –initialize-insecure –user=mysql –basedir=/app/mysql –datadir=/data/mysql

#配置mysql主配置文件

cat >> /etc/my.cnf << EOF

[mysqld]

basedir = /app/mysql

datadir = /data/mysql

pid-file = /tmp/mysql.pid

socket = /tmp/mysql.sock

port = 3306

user = mysql

log_error = /data/mysql/mysql-error.log

slow-query-log-file = /data/mysql/mysql-slow.log

log_bin = /data/mysql/mysql-master-bin.log

relay-log = /data/mysql/mysql-relay-bin

server-id = 1

innodb_buffer_pool_size = 1024M

innodb_log_buffer_size = 16M

key_buffer_size = 128M

query_cache_size = 256M

tmp_table_size = 128M

binlog_format = mixed

skip-external-locking

skip-name-resolve

character-set-server = utf8

collation-server = utf8_bin

max_allowed_packet = 16M

thread_cache_size = 256

table_open_cache = 4096

back_log = 1024

max_connect_errors = 100000

interactive_timeout = 1800

wait_timeout = 1800

max_connections = 500

sort_buffer_size = 16M

join_buffer_size = 4M

read_buffer_size = 4M

read_rnd_buffer_size = 16M

binlog_cache_size = 2M

thread_stack = 192K

max_heap_table_size = 128M

myisam_sort_buffer_size = 128M

bulk_insert_buffer_size = 256M

open_files_limit = 65535

query_cache_limit = 2M

slow-query-log

long_query_time = 2

expire_logs_days = 3

max_binlog_size = 1000M

slave_parallel_workers = 4

log-slave-updates

binlog_ignore_db = mysql

replicate_wild_ignore_table = mysql.%

sync_binlog = 1

innodb_file_per_table = 1

innodb_flush_method = O_DIRECT

innodb_buffer_pool_instances = 4

innodb_log_file_size = 512M

innodb_log_files_in_group = 3

innodb_open_files = 4000

innodb_read_io_threads = 8

innodb_write_io_threads = 8

innodb_thread_concurrency = 8

innodb_io_capacity = 2000

innodb_io_capacity_max = 6000

innodb_lru_scan_depth = 2000

innodb_max_dirty_pages_pct = 85

innodb_flush_log_at_trx_commit = 2

sql_mode=NO_ENGINE_SUBSTITUTION,STRICT_TRANS_TABLES

[mysqldump]

quick

quote-names

max_allowed_packet = 16M

[client]

default-character-set = utf8

[mysql]

default-character-set = utf8

[isamchk]

key_buffer = 128M

sort_buffer_size = 4M

read_buffer = 2M

write_buffer = 2M

[myisamchk]

key_buffer = 128M

sort_buffer_size = 4M

read_buffer = 2M

write_buffer = 2M

EOF

#准备服务管理配置文件

cat >> /etc/systemd/system/mysqld.service << EOF

[Unit]

Description=MySQL Server

Documentation=man:mysqld(8)

Documentation=http://dev.mysql.com/doc/refman/en/using-systemd.html

After=network.target

After=syslog.target

[Install]

WantedBy=multi-user.target

[Service]

User=mysql

Group=mysql

ExecStart=/app/mysql/bin/mysqld –defaults-file=/etc/my.cnf

LimitNOFILE = 5000

EOF

#启动mysql服务

systemctl daemon-reload

systemctl start mysqld

systemctl status mysqld

在master1中为mysql从库账户授权：

grant replication slave on . to ‘sync’@’192.168.10.%’identified by ‘syncpwd’;

flush privileges;

show master status; #当前主库状态,即master1

在master2中为mysql从库账户授权：

grant replication slave on . to ‘sync’@’192.168.10.%’ identified by ‘syncpwd’;

flush privileges;

show master status; #当前主库状态,即master2

在maste1中指定master2为主库：

stop slave;

change master to master_host=’172.16.90.181′,master_user=’sync’,master_password=’syncpwd’,master_log_file=’mysql-bin.000004′,master_log_pos=599;

flush privileges;

SET GLOBAL SQL_SLAVE_SKIP_COUNTER=1;

START SLAVE;

start slave;

在maste2中指定master1为主库：

stop slave;

change master to master_host=’172.16.90.180′,master_user=’sync’,master_password=’syncpwd’,master_log_file=’mysql-bin.000004′,master_log_pos=599;

flush privileges;

SET GLOBAL SQL_SLAVE_SKIP_COUNTER=1;

START SLAVE;

start slave;

MySQL双主配置完成，验证配置成功

show slave status\G

如果两个节点下面的属性都是Yes，则说明安装成功

Slave_IO_Running: Yes
Slave_SQL_Running: Yes

笔者在安装的过程中出现了Slave_IO_Running: No，并且报错为：

Got fatal error 1236 from master when reading data from binary log

然后通过这篇博客找到了解决方法，再次记录一下:

Got fatal error 1236 from master when reading data from binary log: ‘Could not find first log file n_君子志邦的博客-CSDN博客

至此，mysql主主已经配置完成。

四 Keepalived高可用配置

首先在两个机器上都安装Keepalived。

安装：yum install -y keepalived

启动：systemctl start keepalived && systemctl enable keepalived

说明： 当两台服务器都正常的时候 用户默认访问服务器A，如果A有异常则访问B服务器。

首先记录一下这里踩的一个坑：yum下载keepalived报错：

然后就更换了yum源，然后就可以正常下载keepalived了，参考博客：

yum 修改源_zb313982521的博客-CSDN博客_修改yum源

下载完keepalived后，vim/etc/keepalived/keepalived.conf

节点1配置文件修改为：

! Configuration File for keepalivedglobal_defs {notification_email {ops@localhost}notification_email_from root@localhostsmtp_server 127.0.0.1smtp_connect_timeout 30router_id mysql-master}vrrp_instance VI_1 {state MASTERinterface ens192virtual_router_id 51priority 100 advert_int 1 authentication { auth_type PASS auth_pass 1111 }virtual_ipaddress {172.16.90.186}}

节点2配置文件修改为：

! Configuration File for keepalivedglobal_defs {notification_email {ops@localhost}notification_email_from root@localhostsmtp_server 127.0.0.1smtp_connect_timeout 30router_id mysql-master}vrrp_instance VI_1 {state BACKUPinterface ens192virtual_router_id 51priority 90advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {172.16.90.186}}

然后直接使用172.16.90.186这个VIP访问数据库。当访问成功，并且插入数据时，两个主节点都更新了数据，则证明mysql主主配置和keepalived配置成功。