Mysql_HA+IP_SAN+Clvmd+GFS2
====================================================================
GFS:全局文件系统,支持DLM锁管理,支持并发写,Centos6不超过16个节点适用于HA集群使用的共享存储适用于LB集群RealServer使用的共享存储
mysql高可用集群架构设计
1、节点主机系统:Centos6.5
2、高可用集群软件:RHCS
3、两台节点主机node1,node2:
Node1: IP:192.168.1.34host name:edu-mysql-master;
Node2: IP:192.168.1.157host name:edu-mysql-slave;
VIP: 192.168.1.166
Service:mysql
4、一台GFS2服务主机:
IP:192.168.1.111
LVM:/dev/mapper/datavg-mysql--lv
共享目录:/data(LVM),/data(mysql数据目录)
5、三台主机相同用户:mysql mysql(组) 3306(uid、gid)
相关准备配置
1、配置IP、安装WEB服务、关闭防火墙;
2、配置各节点名称;
3、建立各节点间的SSH互信通信;
4、各节点间的时间同步;
一、存储配置
存储端(target)
1、安装包
[root@target ~]# yum -y install scsi-target-utils
2、定义target共享
[root@target ~]# vim /etc/tgt/targets.conf
backing-store /dev/sdb
3、启动服务
[root@target ~]# service tgtd start
[root@target ~]# netstat -tnlp |grep :3260
tcp 0 0 0.0.0.0:3260 0.0.0.0:* LISTEN 15135/tgtd
tcp 0 0 :::3260 :::* LISTEN 15135/tgtd
[root@storage ~]# tgt-admin --show
Target 1: iqn.2017-1.com.example:mysql
Backing store path: /dev/sdb
ACL information:
ALL
集群节点initiator端(node1、node2)
1、安装包
[root@node1 ~]# yum -y install iscsi-*
2、发现存储
[root@node1 ~]# iscsiadm -m discovery -t st -p 192.168.1.111
正在启动iscsid: [确定]
192.168.1.111:3260,1 iqn.2017-1.com.example:mysql
3、登入存储
[root@node1 ~]# service iscsi restart
[root@node1 ~]# chkconfig iscsi on
4、查看设备
[root@node1 ~]# fdisk -cul
Disk /dev/sdb: 8589 MB, 8589934592 bytes
64 heads, 32 sectors/track, 8192 cylinders, total 16777216 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x00000000
二、配置集群,并使用CLVMD(node1、node2)
1. 配置集群
a. 创建集群
2. 配置CLVMD
确保node1和node2 clvmd运行
[root@node1 ~]# /etc/init.d/clvmd status
clvmd (pid 1908) 正在运行...
Clustered Volume Groups: (none)
Active clustered Logical Volumes: (none)
node1创建lv
[root@node1 ~]# pvcreate /dev/sdb
[root@node1~]# vgcreate datavg /dev/sdb
[root@node1~]# lvcreate -L 2G -n mysql-lv datavg
[root@node1~]# vgs
VG #PV #LV #SN Attr VSize VFree
datavg 1 1 0 wz--nc8.00g 6.00g
[root@node1~]# lvs
LV VG Attr LSize Pool Origin Data% Move Log Cpy%Sync Convert
mysql-lv datavg -wi-a---- 2.00g
node2查看
[root@node2 ~]# vgs
VG #PV #LV #SN Attr VSize VFree
VolGroup 1 2 0 wz--n- 9.51g 0
gfs-vg 1 1 0 wz--nc 5.00g 3.00g
[root@node2 ~]# lvs
LV VG Attr LSize Pool Origin Data% Move Log Cpy%Sync Convert
lv_root VolGroup -wi-ao--- 8.54g
lv_swap VolGroup -wi-ao--- 992.00m
gfs-lv gfs-vg -wi-a---- 2.00g
3 、创建GFS2文件系统 node1或node2
a、在node1,node2上安装gfs2的相关软件包
[root@node1 ~]# rpm -qa |grep gfs
gfs2-utils-3.0.12.1-78.el6.x86_64//提供mkfs.gfs2命令用于创建gfs2文件系统
[root@node1 ~]# modinfo -F filename gfs2
/lib/modules/2.6.32-358.el6.x86_64/kernel/fs/gfs2/gfs2.ko
b、在任意一个节点上格式化
[root@node1 ~]# cman_tool status
Version: 6.2.0
Config Version: 19
Cluster Name:gfs-cluster
Cluster Id: 1835
Cluster Member: Yes
Cluster Generation: 24
Membership state: Cluster-Member
Nodes: 2
Expected votes: 1
Total votes: 2
Node votes: 1
Quorum: 1
Active subsystems: 9
Flags: 2node
Ports Bound: 0 11 177
Node name: edu-mysql-master
Node ID: 1
Multicast addresses: 239.192.7.50
Node addresses: 192.168.1.34
[root@node1 ~]# mkfs.gfs2 -tgfs-cluster:table1 -p lock_dlm -j 3 -J 128M /dev/datavg/mysql-lv
This will destroy any data on /dev/datavg/mysql-lv.
It appears to contain: symbolic link to `../dm-2'
Are you sure you want to proceed? [y/n] y
Device: /dev/datavg/mysql-lv
Blocksize: 4096
Device Size 2.00 GB (524288 blocks)
Filesystem Size: 2.00 GB (524288 blocks)
Journals: 3
Resource Groups: 8
Locking Protocol: "lock_dlm"
Lock Table: "gfs-cluster:table1"
UUID: 98326a5a-f72c-5a1d-6ccd-bec094caba26
-t clustername:locktablename
-p lock_dlm
-j 指定日志空间个数,此个数一定要大于等于使用gfs2文件系统存储的主机数
-J 指定一个日志空间的大小
c、挂载并测试写入
[root@node1 ~]# vim /etc/fstab
/dev/datavg/mysql-lv /var/www/html gfs2 defaults,_netdev0 0
[root@node1 ~]# mount -a
[root@node1 ~]# df -P
/dev/mapper/datavg-mysql--lv gfs2 2.0G 410M 1.7G 20% /data
GFS扩展知识:
1. 使用非DLM锁挂载GFS2
[root@node2 ~]# mount -a
/sbin/mount.gfs2: node not a member of the default fence domain
/sbin/mount.gfs2: error mounting lockproto lock_dlm
[root@node2 ~]# mount-o lockproto=lock_nolock/dev/datavg/mysql-lv /data
2. 当集群节点扩充时,增加日志的个数
=====================================================================
MySQL初始化数据库:
1. 安装
[root@node1 ~]# tar -zxfmysql.package.tar-C /usr/local
[root@node1 ~]# cd /usr/local/
[root@node1 local]# cd mysql
[root@node1 mysql]# chown -R mysql:mysql ./*
[root@node1 mysql]# ll
[root@node1 mysql]# ./scripts/mysql_install_db --user=mysql --datadir=/data/
node2同上
(以后脚本安装mysql,更新mysql安装脚本相应路径即可)
2. node1初始化
[root@node1 ~]#mkdir /data
[root@node1 ~]#vim /etc/fstab (node2无须操作)
/dev/datavg/mysql-lv /data gfs2 defaults,_netdev 0 0
[root@node1 ~]#mount –a
[root@node1 ~]# mount /dev/datavg/mysql-lv /data
[root@node1 ~]# service mysqld start;chkconfig mysqld off
[root@node1 ~]# service mysqld stop
[root@node1 ~]# umount /data
2.node1初始化
node2同上
(注:启动mysql服务前先停止node1上的mysql服务,并卸载GFS2共享目录;
然后先挂载存储,再在node2上启动mysql服务,进行测试,测试完后记得停止mysql服务,并卸载GFS2共享目录)
[root@node1 ~]#scp -p /etc/my.cnf node2:/etc
[root@node1 ~]# scp -p /etc/init.d/mysqldnode2:/etc/init.d/
此文档需要与https://192.168.1.111:8084/ 端配置同时进行,集群管理断配置见http://www.jianshu.com/p/988e74a26469;
=====================================================================
[root@node2~]#Apache:
[root@node2~]# ip: 192.168.1.100
[root@node2~]# script: /etc/init.d/httpd
[root@node2~]# filesystem: ------ NFS mount (NAS)
[root@node2~]#MySQL:
[root@node2~]# ip: 192.168.1.200
[root@node2~]# script: /etc/init.d/mysqld
[root@node2~]# filesystem: /dev/datavg/mysql-lv /var/lib/mysql (GFS2)
针对存储资源设置
NFS: NFS Mount
EXT4: Filesystem
GFS2: GFS2