创建RHCS集群环境

创建RHCS集群环境

1.1 问题

准备四台KVM虚拟机，其三台作为集群节点，一台安装luci并配置iSCSI存储服务，实现如下功能：

• 使用RHCS创建一个名为tarena的集群
• 集群中所有节点均需要挂载iSCSI共享存储
• 使用集群中任意节点对iSCSI设置进行分区格式化
• 安装luci的虚拟主机要求额外添加一块20G硬盘
• 物理主机IP地址为192.168.4.1，主机名称为desktop1.example.com

1.2 方案

使用4台虚拟机，1台作为luci和iSCSI服务器、3台作为节点服务器，拓扑结构如图-1所示。

图-1

所有主机的主机名及对应的IP地址如表-1所示。

表－1 主机名称及对应IP地址表

1.3 步骤

实现此案例需要按照如下步骤进行。

步骤一：安装前准备

1）为所有节点配置yum源，注意所有的虚拟主机均需要挂载安装光盘。

1. [[email protected] ~]# mount /dev/cdrom /media
2. [[email protected] ~]# rm -rf /etc/yum.repos.d/*
3. [[email protected] ~]# vim /etc/yum.repos.d/dvd.repo
4. [dvd]
5. name=red hat
6. baseurl=file:///media/
7. enabled=1
8. gpgcheck=0
9. [HighAvailability]
10. name=HighAvailability
11. baseurl=file:///media/HighAvailability
12. enabled=1
13. gpgcheck=0
14. [LoadBalancer]
15. name=LoadBalancer
16. baseurl=file:///media/LoadBalancer
17. enabled=1
18. gpgcheck=0
19. [ResilientStorage]
20. name=ResilientStorage
21. baseurl=file:///media/ResilientStorage
22. enabled=1
23. gpgcheck=0
24. [ScalableFileSystem]
25. name=ScalableFileSystem
26. baseurl=file:///media/ScalableFileSystem
27. enabled=1
28. gpgcheck=0
29. [[email protected] ~]# yum clean all
30. [[email protected] ~]# mount /dev/cdrom /media
31. [[email protected] ~]# rm –rf /etc/yum.repos.d/*
32. [[email protected] ~]# vim /etc/yum.repos.d/dvd.repo
33. [dvd]
34. name=red hat
35. baseurl=file:///media/
36. enabled=1
37. gpgcheck=0
38. [HighAvailability]
39. name=HighAvailability
40. baseurl=file:///media/HighAvailability
41. enabled=1
42. gpgcheck=0
43. [LoadBalancer]
44. name=LoadBalancer
45. baseurl=file:///media/LoadBalancer
46. enabled=1
47. gpgcheck=0
48. [ResilientStorage]
49. name=ResilientStorage
50. baseurl=file:///media/ResilientStorage
51. enabled=1
52. gpgcheck=0
53. [ScalableFileSystem]
54. name=ScalableFileSystem
55. baseurl=file:///media/ScalableFileSystem
56. enabled=1
57. gpgcheck=0
58. [[email protected] ~]# yum clean all
59. [[email protected] ~]# mount /dev/cdrom /media
60. [[email protected] ~]# rm -rf /etc/yum.repos.d/*
61. [[email protected] ~]# vim /etc/yum.repos.d/dvd.repo
62. [dvd]
63. name=red hat
64. baseurl=file:///media/
65. enabled=1
66. gpgcheck=0
67. [HighAvailability]
68. name=HighAvailability
69. baseurl=file:///media/HighAvailability
70. enabled=1
71. gpgcheck=0
72. [LoadBalancer]
73. name=LoadBalancer
74. baseurl=file:///media/LoadBalancer
75. enabled=1
76. gpgcheck=0
77. [ResilientStorage]
78. name=ResilientStorage
79. baseurl=file:///media/ResilientStorage
80. enabled=1
81. gpgcheck=0
82. [ScalableFileSystem]
83. name=ScalableFileSystem
84. baseurl=file:///media/ScalableFileSystem
85. enabled=1
86. gpgcheck=0
87. [[email protected] ~]# yum clean all
88. [[email protected] ~]# mount /dev/cdrom /media
89. [[email protected] ~]# rm -rf /etc/yum.repos.d/*
90. [[email protected] ~]# vim /etc/yum.repos.d/dvd.repo
91. [dvd]
92. name=red hat
93. baseurl=file:///media/
94. enabled=1
95. gpgcheck=0
96. [HighAvailability]
97. name=HighAvailability
98. baseurl=file:///media/HighAvailability
99. enabled=1
100. gpgcheck=0
101. [LoadBalancer]
102. name=LoadBalancer
103. baseurl=file:///media/LoadBalancer
104. enabled=1
105. gpgcheck=0
106. [ResilientStorage]
107. name=ResilientStorage
108. baseurl=file:///media/ResilientStorage
109. enabled=1
110. gpgcheck=0
111. [ScalableFileSystem]
112. name=ScalableFileSystem
113. baseurl=file:///media/ScalableFileSystem
114. enabled=1
115. gpgcheck=0
116. [[email protected] ~]# yum clean all

2）修改/etc/hosts并同步到所有主机。

1. [[email protected] ~]# vim /etc/hosts
2. 192.168.4.1        node1.example.com
3. 192.168.4.2        node2.example.com
4. 192.168.4.3        node3.example.com
5. 192.168.4.4        luci.example.com
6. [[email protected] ~]# for i in {1..3};do scp /etc/hosts 192.168.4.$i:/etc/;done

3）所有节点关闭NetworkManager、SELinux服务。

1. [[email protected] ~]# service NetworkManager stop
2. [[email protected] ~]# chkconfig NetworkManager off
3. [[email protected] ~]# sed -i ‘/SELINUX=/s/enforcing/permissive/‘ /etc/sysconfig/selinux
4. [[email protected] ~]# setenforce 0
5. [[email protected] ~]# iptables -F; service iptables save
6. [[email protected] ~]# service NetworkManager stop
7. [[email protected] ~]# chkconfig NetworkManager off
8. [[email protected] ~]# sed -i ‘/SELINUX=/s/enforcing/permissive/‘ /etc/sysconfig/selinux
9. [[email protected] ~]# setenforce 0
10. [[email protected] ~]# iptables -F; service iptables save
11. [[email protected] ~]# service NetworkManager stop
12. [[email protected] ~]# chkconfig NetworkManager off
13. [[email protected] ~]# sed -i ‘/SELINUX=/s/enforcing/permissive/‘ /etc/sysconfig/selinux
14. [[email protected] ~]# setenforce 0
15. [[email protected] ~]# iptables -F; service iptables save
16. [[email protected] ~]# service NetworkManager stop
17. [[email protected] ~]# chkconfig NetworkManager off
18. [[email protected] ~]# sed -i ‘/SELINUX=/s/enforcing/permissive/‘ /etc/sysconfig/selinux
19. [[email protected] ~]# setenforce 0
20. [[email protected] ~]# iptables -F; service iptables save

步骤二：部署iSCSI服务

1）在luci主机上部署iSCSI服务器，将/dev/sdb使用iSCSI服务共享。

提示：服务器IQN名称为: iqn.2015-06.com.example.luci:cluster。

1. [[email protected] ~]# yum  -y  install  scsi-target-utils        //安装软件
2. .. ..
3. [[email protected] ~]# rpm  -q  scsi-target-utils
4. scsi-target-utils-1.0.24-10.el6.x86_64
5. [[email protected] ~]# vim /etc/tgt/targets.conf
6. <target iqn.2015-06.com.example.luci:cluster>
7. # List of files to export as LUNs
8. backing-store /dev/sdb                            //定义存储设备
9. initiator-address 192.168.4.0/24                    //定义ACL
10. </target>
11. [[email protected] ~]# service  tgtd  start                     //启动服务
12. Starting SCSI target daemon:                               [  OK  ]
13. [[email protected] ~]# chkconfig  tgtd  on

2）所有节点服务器挂载该iSCSI共享。

1. [[email protected] ~]# yum -y install iscsi-initiator-utils    //安装软件
2. [[email protected] ~]# iscsiadm -m discovery -t sendtargets -p 192.168.4.4:3260
3. [[email protected] ~]# iscsiadm -m node -T \
4. >iqn.2015-06.com.example.luci:cluster \
5. >-p 192.168.4.4:3260 -l                                //挂载iSCSI共享
6. [[email protected] ~]# iscsiadm -m node -T \
7. >iqn.2015-06.com.example.luci:cluster \
8. >-p 192.168.4.4:3260 -l
9. [[email protected] ~]# yum -y install iscsi-initiator-utils    //安装软件
10. [[email protected] ~]# iscsiadm -m discovery -t sendtargets -p 192.168.4.4:3260
11. [[email protected] ~]# iscsiadm -m node -T \
12. >iqn.2015-06.com.example.luci:cluster \
13. >-p 192.168.4.4:3260 –l                                //挂载iSCSI共享
14. [[email protected] ~]# iscsiadm -m node -T \
15. >iqn.2015-06.com.example.luci:cluster \
16. >-p 192.168.4.4:3260 -l
17. [[email protected] ~]# yum -y install iscsi-initiator-utils    //安装软件
18. [[email protected] ~]# iscsiadm -m discovery -t sendtargets -p 192.168.4.4:3260
19. [[email protected] ~]# iscsiadm -m node -T \
20. >iqn.2015-06.com.example.luci:cluster \
21. >-p 192.168.4.4:3260 -l                                //挂载iSCSI共享
22. [[email protected] ~]# iscsiadm -m node -T \
23. >iqn.2015-06.com.example.luci:cluster \

1. >-p 192.168.4.4:3260 –l

步骤三：安装集群软件

1）在luci.example.com主机上安装luci，并启动服务。

1. [[email protected] ~]# yum –y install luci
2. [[email protected] ~]# service luci start;chkconfig luci on

2）在所有的集群节点中安装ricci，并启动服务。

1. [[email protected] ~]# yum -y install ricci
2. [[email protected] ~]# echo "11111" |passwd --stdin ricci
3. [[email protected] ~]# service ricci start;chkconfig ricci on
4. [[email protected] ~]# yum -y install ricci
5. [[email protected] ~]# echo "11111" |passwd --stdin ricci
6. [[email protected] ~]# service ricci start;chkconfig ricci on
7. [[email protected] ~]# yum -y install ricci
8. [[email protected] ~]# echo "11111" |passwd --stdin ricci
9. [[email protected] ~]# service ricci start;chkconfig ricci on

步骤四：配置集群

1）浏览器访问luci，任意主机使用浏览器访问即可。

1. [[email protected] ~]# firefox https://luci.example.com:8084

2）创建集群。

使用浏览器访问luici页面后，点击“Manage Clusters“页面，点击”Create“按钮创建一个新的集群，如图-2所示。

图-2

接下来，在弹出的回话框中，输入集群的名称“tarena“，勾选”Download Packages“、“Reboot Nodes Befor Joining Cluster”、“Enable Shared Storage Support”，效果如图-3所示。

图-3

等待所有的节点重启之后，在luci页面将显示如图-4所示的页面，代表所有节点以及加入到了tarena集群中。

图-4

提示：如果节点重启后，有部分节点无法自动加入集群中，可以将正常的节点系统中/etc/cluster/cluster.conf文件同步到其他失败的节点中，并确保失败节点的cman和rgmanager服务为启动状态即可。