Mysql双主热备+读写分离

 

#双主热备=互相配slave

vi /etc/my.cnf
[mysqld]
server-id = 1
log-bin = mysql-bin
binlog-ignore-db = mysql,information_schema
binlog_format = mixed
auto-increment-increment = 2
auto-increment-offset = 1
or
[mysqld]
log-bin=mysql-bin
server-id=1

#为所有从服务器授权所有数据库
grant replication slave on *.* to 'dd'@'192.168.55.%' identified by '123456';

#dd授权dd0的账户
mysql -uroot -p

grant replication slave on *.* to 'dd'@'192.168.55.130' identified by '123456';
flush privileges;
show master status;
#dd0授权dd的账户
mysql -uroot -p
grant replication slave on *.* to 'dd'@'1

92.168.55.138' identified by '123456';
flush privileges;
show master status;
#dd0设dd为主服务器
change master to master_host='192.168.55.138', master_user='dd', master_password='123456', master_log_file='mysql-bin.000001', ma

ster_log_pos=328;
start slave
show slave status \G;
#dd设dd0为主服务器
change master to master_host='192.168.55.139', master_user='dd', master_password='123456', master_log_file='mysql-bin.000002', master_log_pos=329;
start slave
show slave status \G;

create table dd(id int primary key auto_increment,name varchar(20));
--------------------------------------------------------------------------------
#读写分离
#dd授权dd2 dd3的账户
#授权
mysql -uroot -p
grant replication slave on *.* to 'dd'@'192.168.55.130' identified by '123456';
flush privileges;
show master status;
#设主
mysql -uroot -p
change master to master_host='192.168.55.138', master_user='dd', master_password='123456', master_log_file='mysql-bin.000001', master_log_pos=658;
start slave;
show slave status \G;
#主建表从查看测试
create database dd;
use dd;
create table xq(id int(10),name varchar(10),address varchar(20));
insert into xq values('1','xq01','this_is_master');
----------------------------------------
错误： Slave_IO_Running: Yes
Slave_SQL_Running: Connecting
解决： show variables like 'datadir';
select uuid();
restart mariadb/mysqld
----------------------------------------
#dd0安装amoeba
#dd dd2 dd3
grant all privileges on *.* to 'root'@'192.168.55.%' identified by 'root';
flush privileges;
--------------------------------------------------------------------------------
#dd0
tar /usr/local/amoeba
chmod -R 755 /usr/local/amoeba/
cd /usr/local/amoeba
vi conf/amoeba.xml
#amoeba账户密码
user root
password root
最后三个pool
master
master
slaves
--------------------------------------------------------------------------------
vi conf/dbServers.xml (用户密码的地方去掉注释)
#之前授权的账户密码
root
root

dd

dd2

dd3

dbServer name="slaves" virtual="true"
slave0,slave1

vi bin/amoeba
DEFAULT_OPTS="-server -Xms1024m -Xmx1024m -Xss256k"

/usr/local/amoeba/bin/amoeba start
--------------------------------------------------------------------------------
#dd1连接dd0测试
mysql -u root -proot -h dd0 -P8066

错误提示：
Could not create a validated object, cause: ValidateObject failed
依据 conf/dbServers.xml
test
主建test数据库

mysql -uroot -proot -h 192.168.55.130 -P8066
use test;
create table dd(id int primary key auto_increment,name varchar(20));
create table xq(id int(10),name varchar(10),address varchar(20));
insert into xq values('1','xq01','this_is_master');
select * from xq;

#查看dd dd2 dd3 数据库数据
--------------------------------------------------------------------------------

以优化的方式配置主从
#主
[mysqld]
vi /etc/my.cnf
log-bin=master-bin
log-slave-updates
server-id=1
innodb_flush_log_at_trx_commit=2
binlog-format=ROW
binlog_cache_size = 4M
max_binlog_size = 1G
max_binlog_cache_size = 2G
skip-name-resolve
#for 5.6
master_info_repository=table
relay_log_info_repository=TABLE
gtid-mode=ON
enforce-gtid-consistency=true

注解：
binlog-format=ROW ##二进制文件的格式
gtid-mode=ON ##开启 GTID 模式
enforce-gtid-consistency=true ##开启 GTID 的一些安全限制
binlog_cache_size = 4M ##单个事务的二进制日志占用内存的最大值
max_binlog_size = 1G ##二进制文件最大的大小
max_binlog_cache_size = 2G ##二进制文件占用内存的最大值

从
[mysqld]
relay-log=relay1-log-bin
relay-log-index=slave-relay1-bin.index
server-id=2
innodb_flush_log_at_trx_commit=2
slave_parallel_workers=16
binlog-format=ROW
binlog-row-image = minimal
log-bin=slave1-bin
log-bin-index=slave1-log-bin.index
binlog_cache_size = 4M
max_binlog_size = 1G
max_binlog_cache_size = 2G
slave-sql-verify-checksum=1
binlog-rows-query-log_events=1
log-slave-updates
relay_log_purge = 1
relay_log_recovery = 1
skip-name-resolve
#for 5.6
master_info_repository=table
relay_log_info_repository=TABLE
gtid-mode=ON
enforce-gtid-consistency=true
slave-parallel-type=LOGICAL_CLOCK

注解：
binlog-format=ROW ##二进制文件的格式
binlog-row-image = minimal ##使 mysql 复制中的行体积更小
log-bin=slave1-bin ##二进制日志文件的名称
log-bin-index=slave1-log-bin.index ##二进制日志文件的索引文件
gtid-mode=ON ##开启 GTID
enforce-gtid-consistency=true ##开启 GTID 的一些安全限制
binlog_cache_size = 4M ##单个事务的二进制日志占用内存的最大值
max_binlog_size = 1G ##二进制日志文件的最大大小
max_binlog_cache_size = 2G ##二进制日志文件占用内存的最大值
slave-sql-verify-checksum=1 ##判断从库 SQL Threa 线程从 relay log 中读出
的事件是否正常,默认设置为 1
binlog-rows-query-log_events=1 ##优化主从性能
log-slave-updates ##开启 slave 角色的日志更新
relay_log_purge = 1 ##不需要中继日志时自动清空
relay_log_recovery = 1 ##当slave从库宕机后，假如relay-log损坏了，
导致一部分中继日志没有处理，则自动放弃所有未执行的 relay-log，并且重新从 master 上
获取日志，这样就保证了 relay-log 的完整性。

#dd授权dd0 dd1的账户
mysql -uroot -p
grant replication slave on *.* to 'dd'@'192.168.55.139' identified by '123456';
flush privileges;
show master status;
grant replication slave on *.* to 'dd'@'192.168.55.140' identified by '123456';
flush privileges;
show master status;
#dd0 dd1设dd为主服务器
change master to master_host='192.168.55.138', master_user='dd', master_password='123456', master_log_file='mysql-bin.000001', master_log_pos=328;
start slave;
show slave status \G;

create table xq(id int(10),name varchar(10),address varchar(20));
insert into xq values(1,'xq','this is xq');

报错：Unable to load replication GTID slave state from mysql.gtid_slave_pos: Table 'mysql.gtid_slave_pos' doesn't exist Can't find rec
use mysql;
create table `gtid_slave_pos` (
`domain_id` int(10) unsigned NOT NULL,
`sub_id` bigint(20) unsigned NOT NULL,
`server_id` int(10) unsigned NOT NULL,
`seq_no` bigint(20) unsigned NOT NULL,
PRIMARY KEY (`domain_id`,`sub_id`)
) ENGINE=MyISAM DEFAULT CHARSET=utf8 COMMENT='Replication slave GTID state';
stop slave;
start slave;
show slave status \G;

报错：Got fatal error 1236
flush logs;
change master to master_log_file='mysql-bin.000010', master_log_pos=387;
show slave status \G;

转载于:https://www.cnblogs.com/feizhuanye/p/10395142.html