码迷,mamicode.com
首页 > 数据库 > 详细

RHEL6+GFS2+MYSQL高可用

时间:2019-12-02 11:50:16      阅读:155      评论:0      收藏:0      [点我收藏+]

标签:sed   fail   rpo   一个   dia   inux   amp   服务   style   

RHCS集群安装部署

组件介绍:

luci: rhcs集群管理工具-主控端

ricci: rhcs集群管理工具-受控端

rgmanager: 高可用性服务管理

cman: 集群管理

gfs2: 集群文件系统

iscsi-target-utils: iscsi服务端

iscsi-initiator: iscsi发起端

 

 

主机名

IP

software

csnode01

192.168.2.10

Ricci,luci,rgmanager,cman

csnode02

192.168.2.20

Ricci,rgmanager,cman

csnode03

192.168.2.30

Ricci,rgmanager,cman

csnode04

192.168.2.40

 

 

 

 

 

 

 

 

 

 

 

 

 

 

实现过程

 

配置yum仓库,通过httpd实现:

 

在csnode04上:

挂载OS镜像

 

启动httpd服务:

#service httpd start

#chkconfig httpd on

#cd /var/www/html

[root@csnode04 html]# mkdir rhel/6/os/x86_64/ -pv

[root@csnode04 html]# mount /dev/sr0 rhel/6/os/x86_64/

mount: block device /dev/sr0 is write-protected, mounting read-only

 

 

在csnode01,cdnode02,cdnode03上配置以下:

#vim /etc/yum.repo.d/base.repo

[rhel-source]

name=Red Hat Enterprise Linux $releasever - $basearch - Source

baseurl=http://192.168.2.40/rhel/6/os/x86_64/

enabled=1

gpgcheck=0

 

[HighAvailability]

name=Red Hat Enterprise Linux $releasever - $basearch - Source

baseurl=http://192.168.2.40/rhel/6/os/x86_64/HighAvailability

enabled=1

gpgcheck=0

 

[LoadBalancer]

name=Red Hat Enterprise Linux $releasever - $basearch - Source

baseurl=http://192.168.2.40/rhel/6/os/x86_64/LoadBalancer

enabled=1

gpgcheck=0

 

[ResilientStorage]

name=Red Hat Enterprise Linux $releasever - $basearch - Source

baseurl=http://192.168.2.40/rhel/6/os/x86_64/ResilientStorage

enabled=1

gpgcheck=0

 

[ScalableFileSystem]

name=Red Hat Enterprise Linux $releasever - $basearch - Source

baseurl=http://192.168.2.40/rhel/6/os/x86_64/ScalableFileSystem

enabled=1

gpgcheck=0

 

4台服务器配置hosts名称解析:

#vim /etc/hosts

127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4

::1         localhost localhost.localdomain localhost6 localhost6.localdomain6

192.168.2.10 csnode01.example.com csnode01

192.168.2.20 csnode02.example.com csnode02

192.168.2.30 csnode03.example.com csnode03

192.168.2.40 csnode04.example.com csnode04

 

禁用防火墙:

#service iptables stop

#chkconfig iptables off

关闭网络服务:

#service NetworkManager stop

#chkconfig NetworkManager off

关闭Selinux:

#sed -i ‘s/^SELINUX=.*/SELINUX=disabled/g‘ /etc/selinux/config

 

安装RHCS工具:

管理服务器安装luci:

 

[root@csnode01 yum.repos.d]# yum install luci -y

[root@csnode01 yum.repos.d]# service luci start

[root@csnode01 yum.repos.d]# chkconfig luci on

 

浏览器访问 http://10.0.0.80:8084 使用root帐号密码即可登陆管理界面

 

集群节点csnode01,csnode02,csnode03安装ricci rgmanager cman

[root@csnode01 yum.repos.d]#yum install ricci rgmanager cman -y

 

从红帽企业版 Linux 6.1 开始,您在任意节点中使用 ricci 推广更新的集群配置时要求输入密码。所以在前面两台主机安装完成ricci后需要修改ricci用户的密码,这个是后面节点设置的密码,并且要作时间同步;然后启动安装的服务

 

修改密码:

[root@csnode01 yum.repos.d]# echo 123456 | passwd --stdin ricci

永久启动服务:

[root@csnode01 yum.repos.d]# service ricci start;service rgmanager start ;service cman start

[root@csnode01 yum.repos.d]# chkconfig ricci on ; chkconfig rgmanager on ; chkconfig cman on

 

安装好RHCS,在启动cman的时候会提示如下错误:

Starting cman... xmlconfig cannot find/etc/cluster/cluster.conf    [Faile]

是由于cluster.conf文件没有创建,可以安装luci图形界面配置。

 

访问图形管理界面,配置RHCS:

登入:https://192.168.2.10:8084

 

配置Cluster:

 

Manag Clusters -> Create (其中password填写 123456 用户的密码,同时建议将”Reboot Nodes Before Joining Cluster“推荐勾选上)

 

 

 

 

 

添加故障转移域:FailOverDomains -> Add ,添加转移域

Prioritized:故障转移时选择优先级高

 

Restricted:服务只运行在指定的节点上

 

No Failback:故障节点恢复后,服务不用切换回去

 

 

 

 

服务安装

给csnode04添加一块硬盘,将/dev/sdb(新添加的硬盘)共享给csnode01、csnode02和csnode03

 

安装服务端程序:

[root@csnode04 yum.repos.d]# yum install scsi-target-utils -y

 

创建分区:

[root@csnode04 yum.repos.d]# fdisk /dev/sdb

Device contains neither a valid DOS partition table, nor Sun, SGI or OSF disklabel

Building a new DOS disklabel with disk identifier 0xe747f71d.

Changes will remain in memory only, until you decide to write them.

After that, of course, the previous content won‘t be recoverable.

 

Warning: invalid flag 0x0000 of partition table 4 will be corrected by w(rite)

 

WARNING: DOS-compatible mode is deprecated. It‘s strongly recommended to

         switch off the mode (command ‘c‘) and change display units to

         sectors (command ‘u‘).

 

Command (m for help): n

Command action

   e   extended

   p   primary partition (1-4)

p

Partition number (1-4): 1

First cylinder (1-20480, default 1):

Using default value 1

Last cylinder, +cylinders or +size{K,M,G} (1-20480, default 20480):

Using default value 20480

 

Command (m for help): w

The partition table has been altered!

 

Calling ioctl() to re-read partition table.

Syncing disks.

 

让内核重读分区表:

 

[root@csnode04 yum.repos.d]# partx -a /dev/sdb

BLKPG: Device or resource busy

error adding partition 1

[root@csnode04 yum.repos.d]# partx -a /dev/sdb

BLKPG: Device or resource busy

error adding partition 1

 

 

配置服务端,添加配置文件:

 

[root@csnode04 yum.repos.d]# vim /etc/tgt/targets.conf

<target iqn.2019-11.com.example:server.target1>

backing-store /dev/sdb1

initiator-address 192.168.2.10

initiator-address 192.168.2.20

initiator-address 192.168.2.30

</target>

 

启动tgtd服务:

[root@csnode04 yum.repos.d]# service tgtd start

 

查看共享的设备信息:

 

[root@csnode04 yum.repos.d]# tgtadm -L iscsi -m target -o show

Target 1: iqn.2019-11.com.example:server.target1

    System information:

        Driver: iscsi

        State: ready

    I_T nexus information:

    LUN information:

        LUN: 0

            Type: controller

            SCSI ID: IET     00010000

            SCSI SN: beaf10

            Size: 0 MB, Block size: 1

            Online: Yes

            Removable media: No

            Prevent removal: No

            Readonly: No

            Backing store type: null

            Backing store path: None

            Backing store flags:

        LUN: 1

            Type: disk

            SCSI ID: IET     00010001

            SCSI SN: beaf11

            Size: 21475 MB, Block size: 512

            Online: Yes

            Removable media: No

            Prevent removal: No

            Readonly: No

            Backing store type: rdwr

            Backing store path: /dev/sdb1

            Backing store flags:

    Account information:

    ACL information:

        192.168.2.10

        192.168.2.20

        192.168.2.30

 

 

csnode01、csnode02和csnode03安装客户端程序:

[root@csnode01 yum.repos.d]# yum install iscsi-initiator-utils –y

启动服务:

[root@csnode01 yum.repos.d]# service iscsi start;service iscsid start

 

csnode01、csnode02和csnode03发现共享设备:

[root@csnode01 yum.repos.d]# iscsiadm -m discovery -t st -p 192.168.2.40

Starting iscsid:                                           [  OK  ]

192.168.2.40:3260,1 iqn.2019-11.com.example:server.target1

 

csnode01、csnode02和csnode03登入设备:

[root@csnode01 yum.repos.d]# iscsiadm -m node -T iqn.2019-11.com.example:server.target1 -p 192.168.2.40 -l

Logging in to [iface: default, target: iqn.2019-11.com.example:server.target1, portal: 192.168.2.40,3260] (multiple)

Login to [iface: default, target: iqn.2019-11.com.example:server.target1, portal: 192.168.2.40,3260] successful.

 

此时使用fdisk -l可以看到/dev/sdb设备

 

 

csnode01、csnode02和csnode03安装集群文件系统程序:

[root@csnode01 yum.repos.d]# yum install gfs2-utils –y

创建分区:

[root@csnode01 ~]# fdisk /dev/sdb

Device contains neither a valid DOS partition table, nor Sun, SGI or OSF disklabel

Building a new DOS disklabel with disk identifier 0x20e43e5e.

Changes will remain in memory only, until you decide to write them.

After that, of course, the previous content won‘t be recoverable.

 

Warning: invalid flag 0x0000 of partition table 4 will be corrected by w(rite)

 

WARNING: DOS-compatible mode is deprecated. It‘s strongly recommended to

         switch off the mode (command ‘c‘) and change display units to

         sectors (command ‘u‘).

 

Command (m for help): n

Command action

   e   extended

   p   primary partition (1-4)

p

Partition number (1-4): 1

First cylinder (1-20479, default 1):

Using default value 1

Last cylinder, +cylinders or +size{K,M,G} (1-20479, default 20479):

Using default value 20479

 

Command (m for help): w

The partition table has been altered!

 

Calling ioctl() to re-read partition table.

Syncing disks.

 

任意节点格式化集群文件系统:

[root@csnode01 ~]# mkfs.gfs2 -j 3 -p lock_dlm -t csCluster:sdb1 /dev/sdb1

This will destroy any data on /dev/sdb1.

It appears to contain: data

 

Are you sure you want to proceed? [y/n] y

 

Device:                    /dev/sdb1

Blocksize:                 4096

Device Size                20.00 GB (5242620 blocks)

Filesystem Size:           20.00 GB (5242617 blocks)

Journals:                  3

Resource Groups:           80

Locking Protocol:          "lock_dlm"

Lock Table:                "csCluster:sdb1"

UUID:                      4b12a97a-c23c-998a-9ac3-8a579deb5391

 

 

csnode01、csnode02和csnode03安装mysql服务:

[root@csnode01 ~]# yum install mysql-server –y

挂载刚才的gfs2文件系统给mysql:

[root@csnode01 ~]# mount -t gfs2 /dev/sdb1 /var/lib/mysql/

 

测试:

Csnode01创建一个目录haha:

[root@csnode01 ~]# cd /var/lib/mysql/

[root@csnode01 mysql]# mkdir haha

[root@csnode01 mysql]# ls

Haha

 

Csnode02挂载:

[root@csnode02 ~]# mount -t gfs2 /dev/sdb1 /var/lib/mysql/

invalid device path "/dev/sdb1"

发现有问题:应让内核重读分区表:

[root@csnode02 ~]# partx -a /dev/sdb

[root@csnode02 ~]# mount -t gfs2 /dev/sdb1 /var/lib/mysql/

[root@csnode02 ~]# cd /var/lib/mysql/

[root@csnode02 mysql]# ls

Haha

在csnode02上也看到了刚才看到的haha目录

 

在csnode01上删除haha,在csnode02上看不到haha目录,实现同步操作。

 

此时gfs2配置完成,卸载csnode02上的/var/lib/mysql,在luci界面上配置gfs2资源,使其自动开启服务,并实现数据共享:

 

[root@csnode02 ~]# umount /var/lib/mysql/

 

 

其实就是将资源添加之后,新建资源组,谁拿到资源组谁就可以挂载共享磁盘,打开服务,获得数据库中数据。

 

新建资源组,按照IP,文件管理器,服务的顺序加入资源组中:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

加入资源组,确认之后,重新刷新luci页面,查看效果。配置完成如下:

 

 

自动挂载到了csnode01,mysql服务自动开启:

[root@csnode01 ~]# cd /var/lib/mysql/

[root@csnode01 mysql]# ls

ibdata1  ib_logfile0  ib_logfile1  mysql  mysql.sock  test

 

[root@csnode01 mysql]# clustat

Cluster Status for csCluster @ Sun Dec  1 14:47:05 2019

Member Status: Quorate

 

 Member Name                                                  ID   Status

 ------ ----                                                  ---- ------

 csnode01.example.com                                             1 Online, Local, rgmanager

 csnode02.example.com                                             2 Online, rgmanager

 csnode03.example.com                                             3 Online, rgmanager

 

 Service Name                                        Owner (Last)                                        State

 ------- ----                                        ----- ------                                        -----

 service:mysqlha                                     csnode01.example.com 

 

 

 

切换到csnode02上:

 

 

查看csnode02上的相关服务:

[root@csnode01 ~]# clustat

Cluster Status for csCluster @ Sun Dec  1 14:49:49 2019

Member Status: Quorate

 

 Member Name                                                  ID   Status

 ------ ----                                                  ---- ------

 csnode01.example.com                                             1 Online, Local, rgmanager

 csnode02.example.com                                             2 Online, rgmanager

 csnode03.example.com                                             3 Online, rgmanager

 

 Service Name                                        Owner (Last)                                        State

 ------- ----                                        ----- ------                                        -----

 service:mysqlha                                     csnode02.example.com                                starting

 

 

 

[root@csnode02 ~]# df

Filesystem           1K-blocks      Used Available Use% Mounted on

/dev/mapper/system-lv_root

                      36950724   2882816  32190916   9% /

tmpfs                  1962828     31884   1930944   2% /dev/shm

/dev/sda1               198337     32970    155127  18% /boot

/dev/sdb1             20967856    418908  20548948   2% /var/lib/mysql

 

 

[root@csnode02 ~]# mysql

Welcome to the MySQL monitor.  Commands end with ; or \g.

Your MySQL connection id is 2

Server version: 5.1.66 Source distribution

 

Copyright (c) 2000, 2012, Oracle and/or its affiliates. All rights reserved.

 

Oracle is a registered trademark of Oracle Corporation and/or its

affiliates. Other names may be trademarks of their respective

owners.

 

Type ‘help;‘ or ‘\h‘ for help. Type ‘\c‘ to clear the current input statement.

 

mysql> show databases;

+--------------------+

| Database           |

+--------------------+

| information_schema |

| mysql              |

| test               |

+--------------------+

3 rows in set (0.00 sec)

 

mysql> create databases test1;

mysql> create database test1;

Query OK, 1 row affected (0.01 sec)

 

 

重启csnode02,发现服务重新启动在csnode01上:

 

止此,rhcs+gfs2+mysql搭建完成。

RHEL6+GFS2+MYSQL高可用

标签:sed   fail   rpo   一个   dia   inux   amp   服务   style   

原文地址:https://www.cnblogs.com/mark-dq/p/11969523.html

(0)
(0)
   
举报
评论 一句话评论(0
登录后才能评论!
© 2014 mamicode.com 版权所有  联系我们:gaon5@hotmail.com
迷上了代码!