使用heartbeat和haresources实现Web集群的高可用
实现目标:
heartbeat v2基于haresources实现HA Web
使用heartbeat和haresources实现Web集群的高可用
术语解释:
heartbeat是什么:
Heartbeat 项目是 Linux-HA 工程的一个组成部分，它实现了一个高可用集群系统
为什么要使用heartbeat:
在Linux中实现服务的高可用需要搭建HA集群,HA集群需要提供Messaging Layer和Cluster Resource Manager功能的软件
实现CRM的组件:
heartbeat V2版本中有两个组件可以实现CRM的功能,分别是haresources,crm.在此我们使用haresources来实现对于各节点资源的管理
实验环境:
System: CentOS6.7
host:
Primary: node1.anyisalin.com
Standby: node2.anyisalin.com
Network Configure:
node1: 172.16.100.7/24 eth0
node2: 172.16.100.8/24 eth0
float IP: 172.16.100.1/24#对外提供服务的IP
实验步骤:
1.使用hosts文件实现HA集群中的节点能够解析各节点的主机名
2.使用公钥认证实现主机之间互信通信
3.安装配置heartbeat,httpd
4.测试web服务器是否实现高可用
1.使用hosts文件实现HA集群中的节点能够解析各节点的主机名
# vim /etc/hosts
127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4
::1 localhost localhost.localdomain localhost6 localhost6.localdomain6
172.16.100.7 node1.anyisalin.com
172.16.100.8 node2.anyisalin.com
# scp -p /etc/hosts node1.anyisalin.com:/etc/
2.使用公钥认证实现主机之间互信通信
# ssh-keygen -t rsa -P '' -f ~/.ssh/id_rsa
# ssh-copy-id -i ~/.ssh/id_rsa.pub node1.anyisalin.com
# ssh-keygen -t rsa -P '' -f ~/.ssh/id_rsa
# ssh-copy-id -t ~/.ssh/id_rsa node2.anyisalin.com
3.安装配置heartbeat,httpd
# wget -O /etc/yum.repos.d/centos.repo https://lug.ustc.edu.cn/wiki/_ex ... /centos?codeblock=2##使用USTC镜像源
# yum install epel-rpm-macros.noarch -y
# yum instsall heartbeat httpd -y
# ssh node2.anyisalin.com "yum install -y epel-rpm-macros.noarch"
# ssh node2.anyisalin.com "yum install -y heartbeat httpd"
# cp /usr/share/doc/heartbeat-3.0.4/{authkeys,ha.cf,haresources} /etc/ha.d/
# chmod 600 /etc/ha.d/authkeys
# vim /etc/ha.d/authkeys
##在最后添加两行
auth 1
1 md5 d41d8cd98f00b204e9800998ecf8427e-
##保存退出
# vim /etc/ha.d/ha.cf
##在结尾添加几行
bcast eth0
logfile /var/log/ha-log
keepalive 1
node node1.anyisalin.com #这里的主机名必须和uname -n显示的一致
node node2.anyisalin.com #同上
##保存退出
# vim /etc/ha.d/haresources
##在最后添加一行
node1.anyisalin.com IPaddr::172.16.100.1/24/eth0 httpd
##结束
# scp /etc/ha.d/{authkeys,haresources,ha.cf} node2.anyisalin.com:/etc/ha.d/
# echo "This is HA node1" > /var/www/html/index.html#创建网页文件
# ssh node2.anyisalin.com "echo "This is HA node2" > /var/www/html/index.html"#为node2创建网页文件
# service heartbeat start#启动heartbeat服务
# ssh node2.anyisalin.com "service heartbeat start"#为node2启动heartbeat服务
# tail /var/log/ha-log#监控日志文件
ResourceManager(default):2016/01/02_06:22:00 info: Running /etc/ha.d/resource.d/IPaddr 172.16.100.1/24/eth0 start
IPaddr(IPaddr_172.16.100.1):2016/01/02_06:22:00 INFO: Adding inet address 172.16.100.1/24 with broadcast address 172.16.100.255 to device eth0
IPaddr(IPaddr_172.16.100.1):2016/01/02_06:22:00 INFO: Bringing device eth0 up
IPaddr(IPaddr_172.16.100.1):2016/01/02_06:22:00 INFO: /usr/libexec/heartbeat/send_arp -i 200 -r 5 -p /var/run/resource-agents/send_arp-172.16.100.1 eth0 172.16.100.1 auto not_used not_used
/usr/lib/ocf/resource.d//heartbeat/IPaddr(IPaddr_172.16.100.1):2016/01/02_06:22:00 INFO:Success
ResourceManager(default):2016/01/02_06:22:00 info: Running /etc/init.d/httpdstart
# ip a ##node1为主节点
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 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: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:cf:c0:42 brd ff:ff:ff:ff:ff:ff
inet 172.16.100.7/24 brd 172.16.100.255 scope global eth0
inet 172.16.100.1/24 brd 172.16.100.255 scope global secondary eth0
inet6 fe80::20c:29ff:fecf:c042/64 scope link
valid_lft forever preferred_lft forever
4.测试web服务器是否实现高可用
# curl 172.16.100.1 #测试Web服务器,页面显示为主节点node1的网页文件
This is HA node 1
# /usr/share/heartbeat/hb_standby#使用heartbeat内置脚本模拟FailOver(故障转移)
# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 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: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:cf:c0:38 brd ff:ff:ff:ff:ff:ff
inet 172.16.100.7/24 brd 172.16.100.255 scope global eth0
inet6 fe80::20c:29ff:fecf:c038/64 scope link
valid_lft forever preferred_lft forever
# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 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: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:cf:c0:42 brd ff:ff:ff:ff:ff:ff
inet 172.16.100.8/24 brd 172.16.100.255 scope global eth0
inet 172.16.100.1/24 brd 172.16.100.255 scope global secondary eth0
inet6 fe80::20c:29ff:fecf:c042/64 scope link
valid_lft forever preferred_lft forever
# curl 172.16.100.1#重新发起请求,页面显示为从节点node2的网页文件
This is HA node2
实验总结:
heartbeat和haresources实现Web集群的高可用,在主节点故障的情况下能够通过CRM将资源转移到备用节点上,但是该实验还有很多未考虑的因素,在此不过多进行解释
页:
[1]