CentOS 6.4 Heartbeat+httpd+nfs实现高可用的web服务器

43r21

一、Heartbeat概念
Heartbeat是Linux-HA项目中的一个组件，也是目前开源HA项目中最成功的一个例子, Linux-HA的全称是High-Availability Linux,这个开源项目的目标是：通过社区开发者的共同努力，提供一个增强linux可靠性（reliability)、可用性（availability）和可服务性（serviceability）（RAS）的群集解决方案.Heartbeat提供了所有 HA 软件所需要的基本功能，比如心跳检测和资源接管、监测群集中的系统服务、在群集中的节点间转移共享 IP 地址的所有者等.
Heartbeat官方站点：
* http://www.linux-ha.org
* http://hg.linux-ha.org

二、准备工作
1、Heartbeat网络架构


2、操作系统

1 2 3

CentOS 6.4 X86-64 最小化安装 由于用源码编译安装heartbeat一直没有通过，所以没办法只能采用yum安装。 heartbeat v3

3、地址规划

1 2 3 4

node1 192.168.0.101 255.255.255.0 192.168.0.1 node1.test.com eth1 Active node2 192.168.0.102 255.255.255.0 192.168.0.1 node2.test.com eth1 Passive node3 192.168.0.103 255.255.255.0 192.168.0.1 node3.test.com eth1 nfs vip   192.168.0.200 255.255.255.0

4、主机名解析

1 2 3 4 5 6 7

[iyunv@node1 ~]# uname -n node1.test.com [iyunv@node1 ~]# cat /etc/hosts 127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4 ::1         localhost localhost.localdomain localhost6 localhost6.localdomain6 192.168.1.196 node1.test.com node1 192.168.1.197 node2.test.com node2

1 2 3 4 5 6 7

[iyunv@node2 ~]# uname -n node2.test.com [iyunv@node2 ~]# cat /etc/hosts 127.0.0.1   localhost localhost.localdomain localhost4 localhost4.localdomain4 ::1         localhost localhost.localdomain localhost6 localhost6.localdomain6 192.168.1.196 node1.test.com node1 192.168.1.197 node2.test.com node2

5、双机互信

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

[iyunv@node1 ~]# ssh-keygen -t rsa -f ~/.ssh/id_rsa -P '' Generating public/private rsa key pair. Created directory '/root/.ssh'. Your identification has been saved in /root/.ssh/id_rsa. Your public key has been saved in /root/.ssh/id_rsa.pub. The key fingerprint is: ce:f3:d7:63:10:9b:d2:86:f8:8a:5a:ee:41:d8:d2:01 root@node1.test.com The key's randomart image is: +--[ RSA 2048]----+ |    E            | |     .           | |      .          | |     + .    .    | |    o + S. o +   | |     o o. o *    | |      o +. o o   | |     o o o. . +  | |    .o+ .... . . | +-----------------+ [iyunv@node1 ~]# ssh-copy-id -i .ssh/id_rsa.pub root@node2.test.com The authenticity of host 'node2.test.com (192.168.0.102)' can't be established. RSA key fingerprint is 46:b9:7c:11:db:75:93:ad:f1:26:f0:a7:4d:00:40:20. Are you sure you want to continue connecting (yes/no)? yes Warning: Permanently added 'node2.test.com,192.168.0.102' (RSA) to the list of known hosts. root@node2.test.com's password: Now try logging into the machine, with "ssh 'root@node2.test.com'", and check in:   .ssh/authorized_keys to make sure we haven't added extra keys that you weren't expecting.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

[iyunv@node2 ~]# ssh-keygen -t rsa -f ~/.ssh/id_rsa -P '' Generating public/private rsa key pair. Your identification has been saved in /root/.ssh/id_rsa. Your public key has been saved in /root/.ssh/id_rsa.pub. The key fingerprint is: c4:e3:71:f8:82:09:f0:42:9c:e7:20:db:db:ce:dc:0b root@node2.test.com The key's randomart image is: +--[ RSA 2048]----+ | .o.             | |..+o.  . .       | | +.+o   * .      | |. .... = =       | |   o  o S .      | |  . .    .       | |   +E.           | |    +..          | |      ..         | +-----------------+ [iyunv@node2 ~]# ssh-copy-id -i .ssh/id_rsa.pub root@node1.test.com The authenticity of host 'node1.test.com (192.168.0.101)' can't be established. RSA key fingerprint is 46:b9:7c:11:db:75:93:ad:f1:26:f0:a7:4d:00:40:20. Are you sure you want to continue connecting (yes/no)? yes Warning: Permanently added 'node1.test.com,192.168.0.101' (RSA) to the list of known hosts. root@node1.test.com's password: Now try logging into the machine, with "ssh 'root@node1.test.com'", and check in:   .ssh/authorized_keys to make sure we haven't added extra keys that you weren't expecting.

6、时间同步

1 2	# yum -y install ntpdate # ntpdate asia.pool.ntp.org

7、关闭防火墙

1 2 3 4

# getenforce Disabled # /etc/init.d/iptables status iptables：未运行防火墙。

三、安装heartbeat包
1、node1和node2节点安装epel源

1 2 3

# wget http://dl.fedoraproject.org/pub/ ... ease-6-8.noarch.rpm # wget http://rpms.famillecollet.com/enterprise/remi-release-6.rpm # rpm -Uvh remi-release-6*.rpm epel-release-6*.rpm

2、修改epel源的配置文件

1 2	# sed -i 's/#baseurl/baseurl/g' /etc/yum.repos.d/epel.repo # sed -i 's/mirrorlist/#mirrorlist/' /etc/yum.repos.d/epel.repo

3、安装heartbeat包

1	# yum install heartbeat heartbeat-libs

4、查看heartbeat所依赖的包




四、配置Heartbeat服务
1、heartbeat配置文件的介绍

1 2 3 4

heartbeat3个配置文件     authkeys    #节点之间认证的秘钥key文件，权限为600     ha.cf       #heartbeat服务核心配置文件     aresources  #集群资源管理配置文件(haresource | crm)

2、拷贝heartbeat初始配置文件

1	[iyunv@node1 ~]# cp /usr/share/doc/heartbeat-3.0.4/{ha.cf,authkeys,haresources} /etc/ha.d/

3、编辑authkeys文件

1 2 3 4 5 6 7 8 9 10

[iyunv@node1 ~]# dd if=/dev/random bs=512 count=1 | openssl md5  #生成密钥随机数 记录了0+1 的读入 记录了0+1 的写出 72字节(72 B)已复制，4.8467e-05 秒，1.5 MB/秒 (stdin)= acf7401e6b20d4cec482ba1160eb8efe [iyunv@node1 ~]# vim /etc/ha.d/authkeys #注释：末尾添加以下两行 auth 1 1 md5 acf7401e6b20d4cec482ba1160eb8efe [iyunv@node1 ~]# chmod 600 /etc/ha.d/authkeys

4、编辑ha.cf主配置文件

1 2 3 4 5 6 7

[iyunv@node1 ha.d]# grep -v '^#' ha.cf |sed '/^$/d' 注释：主要修改两处，其它的都可以默认 logfacility local0 mcast eth1 225.100.100.100 694 1 0     #修改心跳信息的传播方式|组播 auto_failback on node node1.test.com    #配置集群中的节点数 node node2.test.com    #配置集群中的节点数

5、编辑haresources配置文件

1 2	[iyunv@node1 ha.d]# grep -v '^#' /etc/ha.d/haresources node1.test.com IPaddr::192.168.0.200/24/eth1 httpd

6、拷贝配置文件到node2节点

1	[iyunv@node1 ~]# scp /etc/ha.d/{ha.cf,haresources,authkeys} root@node2.test.com:/etc/ha.d/

五、节点提供httpd服务
1、安装httpd包

1	[iyunv@node1 ~]# yum -y install httpd

2、提供测试页面

1	[iyunv@node1 ~]# echo "node1.test.com" > /var/www/html/index.html

3、启动httpd服务

1	[iyunv@node1 ~]# service httpd start

4、浏览器访问web页面

注释：测试完成后关闭服务，并让其开机不启动，httpd由heartbeat管理
5、停止httpd服务，设置开机不启动httpd服务

1 2 3 4 5

[iyunv@node1 ~]# service httpd stop 停止 httpd：                                               [确定] [iyunv@node1 ~]# chkconfig httpd off [iyunv@node1 ~]# chkconfig --list httpd httpd             0:关闭    1:关闭    2:关闭    3:关闭    4:关闭    5:关闭    6:关闭

6、节点2同上操作

1 2 3 4 5

[iyunv@node2 ~]# yum -y install httpd [iyunv@node2 ~]# echo "node2.test.com" > /var/www/html/index.html [iyunv@node2 ~]# service httpd start [iyunv@node2 ~]# service httpd stop [iyunv@node2 ~]# chkconfig httpd off

7、访问节点2的httpd服务测试页面




六、启动heartbeat服务
1、启动heartbeat服务

1 2 3 4 5 6 7

[iyunv@node1 ~]# /etc/init.d/heartbeat start Starting High-Availability services: INFO:  Resource is stopped Done. [iyunv@node1 ~]# ssh node2 "/etc/init.d/heartbeat start" Starting High-Availability services: 2014/12/25_21:09:12 INFO:  Resource is stopped Done.

2、查看heartbeat日志

1	[iyunv@node1 ~]# tail -f /var/log/message

3、查看vip信息

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

[iyunv@node1 ~]# ip addr 1: lo:  mtu 16436 qdisc noqueue state UNKNOWN     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00     inet 127.0.0.1/8 scope host lo     inet6 ::1/128 scope host        valid_lft forever preferred_lft forever 2: eth1: mtu 1500 qdisc pfifo_fast state UP qlen 1000     link/ether 00:0c:29:c7:14:97 brd ff:ff:ff:ff:ff:ff     inet 192.168.0.101/24 brd 192.168.0.255 scope global eth1     inet 192.168.0.200/24 scope global eth1    #vip已经成功绑定在eth1的网卡上     inet6 fe80::20c:29ff:fec7:1497/64 scope link        valid_lft forever preferred_lft forever

4、查看httpd服务是否被heartbeat接管

1 2	[iyunv@node1 ~]# netstat -tnlpu |grep httpd tcp        0      0 :::80                       :::*                        LISTEN      2140/httpd

5、浏览器访问测试

6、将node1节点由Active切换成Passive

1 2	[iyunv@node1 ~]# sh /usr/share/heartbeat/hb_standby Going standby [all].

7、查看node1节点的日志信息

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

[iyunv@node1 ~]# tail -f /var/log/messages Dec 25 21:36:11 node1 heartbeat: [1255]: info: node1.test.com wants to go standby [all] Dec 25 21:36:11 node1 heartbeat: [1255]: info: standby: node2.test.com can take our all resources Dec 25 21:36:11 node1 heartbeat: [1701]: info: give up all HA resources (standby). Dec 25 21:36:11 node1 ResourceManager(default)[1714]: info: Releasing resource group: node1.test.com IPaddr::192.168.0.200/24/eth1 httpd Dec 25 21:36:11 node1 ResourceManager(default)[1714]: info: Running /etc/init.d/httpd  stop Dec 25 21:36:11 node1 ResourceManager(default)[1714]: info: Running /etc/ha.d/resource.d/IPaddr 192.168.0.200/24/eth1 stop Dec 25 21:36:11 node1 IPaddr(IPaddr_192.168.0.200)[1789]: INFO: IP status = ok, IP_CIP= Dec 25 21:36:11 node1 /usr/lib/ocf/resource.d//heartbeat/IPaddr(IPaddr_192.168.0.200)[1763]: INFO:  Success Dec 25 21:36:11 node1 heartbeat: [1701]: info: all HA resource release completed (standby). Dec 25 21:36:11 node1 heartbeat: [1255]: info: Local standby process completed [all]. Dec 25 21:36:12 node1 heartbeat: [1255]: WARN: 1 lost packet(s) for [node2.test.com] [425:427] Dec 25 21:36:12 node1 heartbeat: [1255]: info: remote resource transition completed. Dec 25 21:36:12 node1 heartbeat: [1255]: info: No pkts missing from node2.test.com! Dec 25 21:36:12 node1 heartbeat: [1255]: info: Other node completed standby takeover of all resources.

8、注释说明

1	node1节点由Active切换到Passive后，httpd服务停止，vip有node1转移到node2上

9、查看node2节点

1 2 3 4 5 6 7 8 9 10 11 12 13 14

[iyunv@node2 ~]# ip addr 1: lo:  mtu 16436 qdisc noqueue state UNKNOWN     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00     inet 127.0.0.1/8 scope host lo     inet6 ::1/128 scope host        valid_lft forever preferred_lft forever 2: eth1: mtu 1500 qdisc pfifo_fast state UP qlen 1000     link/ether 00:0c:29:ad:9f:36 brd ff:ff:ff:ff:ff:ff     inet 192.168.0.102/24 brd 192.168.0.255 scope global eth1     inet 192.168.0.200/24 brd 192.168.0.255 scope global secondary eth1     inet6 fe80::20c:29ff:fead:9f36/64 scope link        valid_lft forever preferred_lft forever [iyunv@node2 ~]# netstat -tnlp |grep httpd tcp        0      0 :::80                       :::*                        LISTEN      2709/httpd

10、vip转移后再次访问

到此处，最基本最简单的heartbeat服务的高可用就完成了。

七、Heartbeat的共享存储
1、配置node3的NFS服务

1 2 3 4 5 6 7 8 9 10

[iyunv@node3 ~]# yum -y install nfs-utils rpcbind [iyunv@node3 ~]# mkdir /web/htdocs -p [iyunv@node3 ~]# cat /etc/exports /web/htdocs 192.168.0.0/24(ro)   [iyunv@node3 ~]# /etc/init.d/rpcbind start [iyunv@node3 ~]# /etc/init.d/nfs start   [iyunv@node3 ~]# showmount -e '192.168.0.103' Export list for 192.168.0.103: /web/htdocs 192.168.0.0/24 [iyunv@node3 ~]# echo "node3 nfs server" > /web/htdocs/index.html

2、节点挂载测试

node1

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

[iyunv@node1 ~]# mount -t nfs 192.168.0.103:/web/htdocs /mnt/ [iyunv@node1 ~]# ll /mnt/ 总用量 4 -rw-r--r-- 1 nobody nobody 26 12月 25 21:53 index.html [iyunv@node1 ~]# cat /mnt/index.html node3 nfs server [iyunv@node1 ~]# df 文件系统             1K-块      已用      可用 已用% 挂载点 /dev/mapper/VolGroup-lv_root                       16134560   1395740  13919212  10% / tmpfs                   247208         0    247208   0% /dev/shm /dev/sda1               495844     32418    437826   7% /boot 192.168.0.103:/web/htdocs                       16134560   1302528  14012416   9% /mnt [iyunv@node1 ~]# umount /mnt/

node2

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

[iyunv@node2 ~]# mount -t nfs 192.168.0.103:/web/htdocs /mnt/ [iyunv@node2 ~]# ll /mnt/ 总用量 4 -rw-r--r-- 1 nobody nobody 26 12月 25 21:53 index.html [iyunv@node2 ~]# cat /mnt/index.html node3 nfs server [iyunv@node2 ~]# df 文件系统             1K-块      已用      可用 已用% 挂载点 /dev/mapper/VolGroup-lv_root                       16134560   1416796  13898156  10% / tmpfs                   247208         0    247208   0% /dev/shm /dev/sda1               495844     32418    437826   7% /boot 192.168.0.103:/web/htdocs                       16134560   1302528  14012416   9% /mnt [iyunv@node2 ~]# umount /mnt/

3、停止node1和node2节点的heartbeat服务

1 2 3 4 5

[iyunv@node1 ~]# ssh node2 'service heartbeat stop' Stopping High-Availability services: Done. [iyunv@node1 ~]# service heartbeat stop Stopping High-Availability services: Done.

4、修改haresource配置文件

1 2	[iyunv@node1 ~]# vim /etc/ha.d/haresources node1.test.com IPaddr::192.168.0.200/24/eth1 Filesystem::192.168.0.103:/web/htdocs::/var/www/html::nfs httpd

5、拷贝修改后的haresource配置文件到node2

1	[iyunv@node1 ~]# scp /etc/ha.d/haresources root@node2.test.com:/etc/ha.d/

6、启动节点的heartbeat服务

1 2 3 4 5 6 7

[iyunv@node1 ~]# service heartbeat start Starting High-Availability services: INFO:  Resource is stopped Done. [iyunv@node1 ~]# ssh node2 "service heartbeat start" Starting High-Availability services: 2014/12/25_22:01:40 INFO:  Resource is stopped Done.

7、浏览器测试访问

8、查看node1节点信息

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

[iyunv@node1 ~]# ip addr 1: lo:  mtu 16436 qdisc noqueue state UNKNOWN     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00     inet 127.0.0.1/8 scope host lo     inet6 ::1/128 scope host        valid_lft forever preferred_lft forever 2: eth1: mtu 1500 qdisc pfifo_fast state UP qlen 1000     link/ether 00:0c:29:c7:14:97 brd ff:ff:ff:ff:ff:ff     inet 192.168.0.101/24 brd 192.168.0.255 scope global eth1     inet 192.168.0.200/24 brd 192.168.0.255 scope global secondary eth1     inet6 fe80::20c:29ff:fec7:1497/64 scope link        valid_lft forever preferred_lft forever [iyunv@node1 ~]# netstat -tnlp |grep httpd tcp        0      0 :::80                       :::*                        LISTEN      3301/httpd          [iyunv@node1 ~]# df 文件系统             1K-块      已用      可用 已用% 挂载点 /dev/mapper/VolGroup-lv_root                       16134560   1395756  13919196  10% / tmpfs                   247208         0    247208   0% /dev/shm /dev/sda1               495844     32418    437826   7% /boot 192.168.0.103:/web/htdocs                       16134560   1302528  14012416   9% /var/www/html [iyunv@node1 ~]# cat /var/www/html/index.html node3 nfs server

9、停止node1节点的heartbeat服务，进行切换，并查看vip信息，再次访问vip地址

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

[iyunv@node1 ~]# service heartbeat stop Stopping High-Availability services: Done. [iyunv@node1 ~]# ip addr 1: lo:  mtu 16436 qdisc noqueue state UNKNOWN     link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00     inet 127.0.0.1/8 scope host lo     inet6 ::1/128 scope host        valid_lft forever preferred_lft forever 2: eth1: mtu 1500 qdisc pfifo_fast state UP qlen 1000     link/ether 00:0c:29:c7:14:97 brd ff:ff:ff:ff:ff:ff     inet 192.168.0.101/24 brd 192.168.0.255 scope global eth1     inet6 fe80::20c:29ff:fec7:1497/64 scope link        valid_lft forever preferred_lft forever [iyunv@node1 ~]# netstat -tnlp |grep httpd

到此，heartbeat的web高可用就完成了。