mysql集群安装指南

环境说明

系统环境：CentOS 6.5 mini(64bit)
软件版本：mysql-cluster-advanced-7.3.7-linux-glibc2.5-x86_64.tar.gz

IP地址：
MGM：192.168.100.211

NDBD1：192.168.100.212

NDBD2：192.168.100.213

SQL1：192.168.100.214
SQL2：192.168.100.215


一、Cluster安装（五个节点）
安装相关依赖包
# yum -y install perl libaio ntpdate

关闭防火墙
# service iptables stop

删除系统自带mysql：
# rpm -qa | grep mysql
# rpm -e --allmatches --nodeps mysql-libs-5.1.71-1.el6.x86_64

解压缩软件
上传软件到/main目录
# cd /main
# tar zxvf mysql-cluster-advanced-7.3.7-linux-glibc2.5-x86_64.tar.gz
# mv mysql-cluster-advanced-7.3.7-linux-glibc2.5-x86_64 mysql

新建mysql用户及用户组
# groupadd mysql
# useradd -g mysql -s /usr/sbin/nologin mysql
# chown -R mysql:mysql /main/mysql

安装Cluster
# /main/mysql/scripts/mysql_install_db --user=mysql --basedir=/main/mysql --datadir=/main/mysql/data


二、节点配置
1.配置 MGM （MGM 节点）
删除自带的my.cnf文件
必须要删除自带的/etc/my.cnf，如果没有就不用删除
# rm -rf /etc/my.cnf

新建配置文件存放目录
# mkdir /main/mysql/etc
# chown mysql:mysql /main/mysql/etc

创建配置文件config.ini
# vi /main/mysql/etc/config.ini
[ndb_mgmd default]
datadir = /main/mysql/data

[ndbd default]
NoOfReplicas = 2
DataMemory = 80M
IndexMemory = 18M
datadir = /main/mysql/data

[ndb_mgmd]
NodeId = 1
HostName = 192.168.100.211

[ndbd]
NodeId = 11 
HostName = 192.168.100.212
[ndbd]
NodeId = 12
HostName = 192.168.100.213

[mysqld]
NodeId = 81
HostName = 192.168.100.214
[mysqld] 
NodeId = 82
HostName = 192.168.100.215


2.配置数据节点（NDBD1和NDBD2节点）
创建配置文件my.cnf
# vi /etc/my.cnf
[mysqld]
ndbcluster
ndb-connectstring = 192.168.100.211
[mysql_cluster]
ndb-connectstring = 192.168.100.211
3.配置SQL节点(SQL1和SQL2节点)
拷贝文件
# cp -p /main/mysql/support-files/mysql.server /etc/init.d/mysqld

创建配置文件my.cnf
# vi /etc/my.cnf
[mysqld]
basedir = /main/mysql
datadir = /main/mysql/data
ndbcluster
ndb-connectstring = 192.168.100.211
[mysql_cluster]
ndb-connectstring = 192.168.100.211


三、启动节点
启动需按如下顺序进行：MGM > NDBD > SQL

如下的启动方式只适合首次启动，或者备份、恢复或配置变化后重启时，否则将清空所有数据
1：启动 MGM （MGM 节点）
# mkdir -p /usr/local/mysql/mysql-cluster
# ndb_mgmd -f /main/mysql/etc/config.ini --initial

查看集群状态
# ndb_mgm -e show

1 2 3 4 5 6 7 8 9 10 11 12 13

Connected to Management Server at : localhost:1186 Cluster Configuration --------------------- [ndbd(NDB)]     2 node(s) id=11 ( not connected, accepting connect from 192.168.100.212) id=12 ( not connected, accepting connect from 192.168.100.213)   [ndb_mgmd(MGM)] 1 node(s) id=1    @192.168.100.211  (mysql-5.6.21 ndb-7.3.7)   [mysqld(API)]   2 node(s) id=81 ( not connected, accepting connect from 192.168.100.214) id=82 ( not connected, accepting connect from 192.168.100.215)

2：启动NDBD（NDBD1和NDBD2节点）
# ndbd --initial

3：启动SQL (SQL1和SQL2节点)

# service mysqld start

 

四、关闭集群

在管理节点切换到 /usr/local/mysql/bin目录

#./ndb_mgm -e shutdown

 

五、非初次启动

--启动管理节点
[root@localhost config]# /cluster/bin/ndb_mgmd -f /cluster_data/config/config.ini
MySQL Cluster Management Server mysql-5.6.29 ndb-7.4.11

--启动数据节点
[root@localhost mysql-cluster-gpl-7.4.11]# /cluster/bin/ndbd 
2016-05-15 06:41:36 [ndbd] INFO     -- Angel connected to '192.168.78.141:1186'
2016-05-15 06:41:36 [ndbd] INFO     -- Angel allocated nodeid: 2

--重启SQL节点
[root@localhost mysql-cluster-gpl-7.4.11]# /cluster/bin/mysqladmin -usystem -p shutdown
[root@localhost mysql-cluster-gpl-7.4.11]# /cluster/bin/mysqld_safe --defaults-file=/etc/my.cnf

 

转载于:https://www.cnblogs.com/zyzloner/p/6656754.html