环境:
K1:192.168.11.26
K2:192.168.11.28
R1:192.168.11.21
R2:192.168.11.30
VIP:192.168.11.99
一、K1配置
[[email protected] ~]# cd /etc/keepalived/ [[email protected] keepalived]# cat keepalived.conf #查看配置文件 ! Configuration File for keepalived global_defs { #定义全局信息 notification_email { [email protected] } notification_email_from [email protected] #定义发送者邮箱,这里随意填 smtp_server 127.0.0.1 smtp_connect_timeout 30 #连接超时30秒 router_id LVS_DEVEL #定义id,我们使用默认就好 # vrrp_mcast_group4 224.0.100.18 #定义组播地址,建议最好使用 } vrrp_instance VI_1 { #定义虚拟路由地址 state MASTER #我们K1为MASTER interface eth0 #在那口网卡,Centos 7 是enoxxxx virtual_router_id 51 #虚拟路由ID,必须一样主从 priority 100 #优先级 advert_int 1 #默认1秒就好, authentication { #认真 auth_type PASS #keepalived只支持俩中认证,这里使用第二种简单密码认证,也就是预认证 auth_pass 6Nb6zjWB #使用openssh rant -base64 6 获取随机密码,最长支持八位所以也就是后面使用6的原因 } virtual_ipaddress { #定义虚拟IP地址 192.168.11.17 } notify_master "/etc/keepalived/notify.sh master" #我们直接写的脚本通知 notify_backup "/etc/keepalived/notify.sh backup" notify_fault "/etc/keepalived/notify.sh fault" } virtual_server 192.168.11.17 80 { #定义keepalived集群VIP地址 delay_loop 6 #检查周期6秒 lb_algo rr #调度算法 lb_kind DR #调度模型 protocol TCP #只支持TCP,所以想想就好 sorry_server 127.0.0.1 80 #用于集群挂掉了,提示的友好界面,比如提示:服务器正在维护等等友好信息... real_server 192.168.11.30 80 { #定义R1 weight 1 #权重,我们是rr调度,权重可以忽略 HTTP_GET { #这里我们使用的HTTP 80端口 url { #定义要请求的url地址 path / #请求地址为/ ,也就是/indexx.html或者index.php等等 status_code 200 #返回状态码为200表示OK } connect_timeout 3 #连接超时时长,3秒 nb_get_retry 3 #重试3次 delay_before_retry 3 #每次重试的时间间隔3秒 } } real_server 192.168.11.21 80 { #定义R2 weight 1 #权重 HTTP_GET { #这里我们使用的HTTP 80端口 url { #定义要请求的url地址 path / #请求地址为/ ,也就是/indexx.html或者index.php等等 status_code 200 #返回状态码为200表示OK } connect_timeout 3 #连接超时时长,3秒 nb_get_retry 3 #重试3次 delay_before_retry 3 #每次重试的时间间隔3秒 } } }
二、K2配置
[[email protected] keepalived]# cat keepalived.conf ! Configuration File for keepalived global_defs { notification_email { [email protected] } notification_email_from [email protected] smtp_server 127.0.0.1 smtp_connect_timeout 30 router_id LVS_DEVEL # vrrp_mcast_group4 224.0.100.18 } vrrp_instance VI_1 { state BACKUP #这里不同,是BACKUP interface eth0 virtual_router_id 51 priority 98 #这里不同,优先级是98,下面其他都一样 advert_int 1 authentication { auth_type PASS auth_pass 6Nb6zjWB } virtual_ipaddress { 192.168.11.17 } notify_master "/etc/keepalived/notify.sh master" notify_backup "/etc/keepalived/notify.sh backup" notify_fault "/etc/keepalived/notify.sh fault" } virtual_server 192.168.11.17 80 { delay_loop 6 lb_algo rr lb_kind DR protocol TCP sorry_server 127.0.0.1 80 real_server 192.168.11.30 80 { weight 1 HTTP_GET { url { path / status_code 200 } connect_timeout 3 nb_get_retry 3 delay_before_retry 3 } } real_server 192.168.11.21 80 { weight 1 HTTP_GET { url { path / status_code 200 } connect_timeout 3 nb_get_retry 3 delay_before_retry 3 } } }
三、R2配置
[[email protected] ~]# cat skp.sh #查看脚本内容 #!/bin/bash # vip=192.168.11.17 case $1 in start) echo 1 > /proc/sys/net/ipv4/conf/all/arp_ignore echo 1 > /proc/sys/net/ipv4/conf/lo/arp_ignore echo 2 > /proc/sys/net/ipv4/conf/all/arp_announce echo 2 > /proc/sys/net/ipv4/conf/lo/arp_announce ifconfig lo:0 $vip netmask 255.255.255.255 broadcast $vip route add -host $vip dev lo:0 ;; stop) echo 0 > /proc/sys/net/ipv4/conf/all/arp_ignore echo 0 > /proc/sys/net/ipv4/conf/lo/arp_ignore echo 0 > /proc/sys/net/ipv4/conf/all/arp_announce echo 0 > /proc/sys/net/ipv4/conf/lo/arp_announce ifconfig lo:0 del $vip ;; esac [[email protected] ~]# sh skp.sh start #执行脚本,记得传参进去start [[email protected] ~]# ip add sh #查看,配置成功:inet 192.168.11.17/32 brd 192.168.11.17 scope global lo:0 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 valid_lft forever preferred_lft forever inet 192.168.11.17/32 brd 192.168.11.17 scope global lo:0 valid_lft forever preferred_lft forever inet6 ::1/128 scope host valid_lft forever preferred_lft forever 2: eno16780032: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000 link/ether 00:0c:29:c8:1b:d5 brd ff:ff:ff:ff:ff:ff inet 192.168.11.30/24 brd 192.168.11.255 scope global eno16780032 valid_lft forever preferred_lft forever inet6 fe80::20c:29ff:fec8:1bd5/64 scope link valid_lft forever preferred_lft forever [[email protected] ~]# yum install httpd -y #安装httpd服务 [[email protected] ~]# cat /var/www/html/index.html #标记index.html内容,以做区别,在生产机千万要别这样区别,不然用户会迷茫的!~ <h1>===30===</h1> [[email protected] ~]# systemctl start httpd #启动httpd服务,这台是Centos 7 启动方式可能有所不同
四、R1配置
[[email protected] ~]# yum install httpd -y #安装httpd服务 [[email protected] ~]# sh skp.sh start #执行脚本,记得传参进去 [[email protected] ~]# ip add sh #查看,配置成功:inet 192.168.11.17/32 brd 192.168.11.17 scope global lo:0 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 inet 192.168.11.17/32 brd 192.168.11.17 scope global lo:0 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:29:4e:ce:b9 brd ff:ff:ff:ff:ff:ff inet 192.168.11.21/24 brd 192.168.11.255 scope global eth0 inet6 fe80::20c:29ff:fe4e:ceb9/64 scope link valid_lft forever preferred_lft forever [[email protected] ~]# cat /var/www/html/index.html #标记index.html内容,以做区别,在生产机千万要别这样区别,不然用户会迷茫的!~ <h1>===21===</h1> [[email protected] ~]# service httpd start #启动服务,这是Centos 6 很熟悉吧 ^-^
五、在K2启动,这台是初始是BACKUP优先级是98
[[email protected] keepalived]# service keepalived start #启动keepalived,提示OK启动成功 Starting keepalived: [ OK ] [[email protected] keepalived]# ipvsadm -Ln #查看是否成功,发现是OK的 IP Virtual Server version 1.2.1 (size=4096) Prot LocalAddress:Port Scheduler Flags -> RemoteAddress:Port Forward Weight ActiveConn InActConn TCP 192.168.11.17:80 rr -> 192.168.11.21:80 Route 1 0 0 -> 192.168.11.30:80 Route 1 0 0 [[email protected] ~]# tail -f /var/log/messages #这是在启动前,在另外的一个session执行的命令,有兴趣可以看看哦,后面就示例啦 Oct 21 22:58:34 redis_master Keepalived[22646]: Starting Keepalived v1.2.13 (03/19,2015) Oct 21 22:58:34 redis_master Keepalived[22647]: Starting Healthcheck child process, pid=22648 Oct 21 22:58:34 redis_master Keepalived_healthcheckers[22648]: Netlink reflector reports IP 192.168.11.28 added Oct 21 22:58:34 redis_master Keepalived[22647]: Starting VRRP child process, pid=22649 Oct 21 22:58:34 redis_master Keepalived_healthcheckers[22648]: Netlink reflector reports IP fe80::20c:29ff:fe2d:bab0 added Oct 21 22:58:34 redis_master Keepalived_healthcheckers[22648]: Registering Kernel netlink reflector Oct 21 22:58:34 redis_master Keepalived_healthcheckers[22648]: Registering Kernel netlink command channel Oct 21 22:58:34 redis_master Keepalived_vrrp[22649]: Netlink reflector reports IP 192.168.11.28 added Oct 21 22:58:34 redis_master Keepalived_vrrp[22649]: Netlink reflector reports IP fe80::20c:29ff:fe2d:bab0 added Oct 21 22:58:34 redis_master Keepalived_vrrp[22649]: Registering Kernel netlink reflector Oct 21 22:58:34 redis_master Keepalived_vrrp[22649]: Registering Kernel netlink command channel Oct 21 22:58:34 redis_master Keepalived_vrrp[22649]: Registering gratuitous ARP shared channel Oct 21 22:58:34 redis_master Keepalived_vrrp[22649]: Opening file ‘/etc/keepalived/keepalived.conf‘. Oct 21 22:58:34 redis_master Keepalived_healthcheckers[22648]: Opening file ‘/etc/keepalived/keepalived.conf‘. Oct 21 22:58:34 redis_master Keepalived_vrrp[22649]: Configuration is using : 63319 Bytes Oct 21 22:58:34 redis_master Keepalived_vrrp[22649]: Using LinkWatch kernel netlink reflector... Oct 21 22:58:34 redis_master Keepalived_healthcheckers[22648]: Configuration is using : 17292 Bytes Oct 21 22:58:34 redis_master Keepalived_vrrp[22649]: VRRP_Instance(VI_1) Entering BACKUP STATE Oct 21 22:58:34 redis_master Keepalived_vrrp[22649]: VRRP sockpool: [ifindex(2), proto(112), unicast(0), fd(10,11)] Oct 21 22:58:34 redis_master Keepalived_healthcheckers[22648]: Using LinkWatch kernel netlink reflector... Oct 21 22:58:34 redis_master Keepalived_healthcheckers[22648]: Activating healthchecker for service [192.168.11.30]:80 Oct 21 22:58:34 redis_master Keepalived_healthcheckers[22648]: Activating healthchecker for service [192.168.11.21]:80 Oct 21 22:58:38 redis_master Keepalived_vrrp[22649]: VRRP_Instance(VI_1) Transition to MASTER STATE Oct 21 22:58:39 redis_master Keepalived_vrrp[22649]: VRRP_Instance(VI_1) Entering MASTER STATE Oct 21 22:58:39 redis_master Keepalived_vrrp[22649]: VRRP_Instance(VI_1) setting protocol VIPs. Oct 21 22:58:39 redis_master Keepalived_vrrp[22649]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 192.168.11.17 Oct 21 22:58:39 redis_master Keepalived_healthcheckers[22648]: Netlink reflector reports IP 192.168.11.17 added Oct 21 22:58:44 redis_master Keepalived_vrrp[22649]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 192.168.11.17 [[email protected] ~]# curl http://192.168.11.17 #使用其他的机制访问,OK也是我们所设定的rr调度算法 <h1>===21===</h1> [[email protected] ~]# curl http://192.168.11.17 <h1>===30===</h1> [[email protected] ~]# curl http://192.168.11.17 <h1>===21===</h1> [[email protected] ~]# curl http://192.168.11.17 <h1>===30===</h1>
六、启动K1,由于是MASTER 优先级是100 所以VIP地址资源会被抢占到本机
[[email protected] keepalived]# ip add sh #我们发现,VIP地址被抢过来了 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:ad:98:5c brd ff:ff:ff:ff:ff:ff inet 192.168.11.26/24 brd 192.168.11.255 scope global eth0 inet 192.168.11.17/32 scope global eth0 inet6 fe80::20c:29ff:fead:985c/64 scope link valid_lft forever preferred_lft forever [[email protected] keepalived]# ipvsadm -Ln #服务也是正常的,记得K2使用ipvsadm -Ln 看到的也是一样的 IP Virtual Server version 1.2.1 (size=4096) Prot LocalAddress:Port Scheduler Flags -> RemoteAddress:Port Forward Weight ActiveConn InActConn TCP 192.168.11.17:80 rr -> 192.168.11.21:80 Route 1 0 0 -> 192.168.11.30:80 Route 1 0 0 [[email protected] ~]# curl http://192.168.11.17 #使用其他的机制访问,OK也是我们所设定的rr调度算法 <h1>===21===</h1> [[email protected] ~]# curl http://192.168.11.17 <h1>===30===</h1> [[email protected] ~]# curl http://192.168.11.17 <h1>===21===</h1> [[email protected] ~]# curl http://192.168.11.17 <h1>===30===</h1>
七、停掉R1/R2服务,看看是否会是我们所希望的那样提示sorry_server的友好界面
[[email protected] keepalived]# cat /var/www/html/index.html #K1的友好界面提示,为了实验区分,特地+K1
<h1>===localhost--K1==</h1>
[[email protected] keepalived]# cat /var/www/html/index.html #K2的友好界面提示,为了实验区分,特地+K2
<h1>===localhost--K2===</h1>
[[email protected] ~]# systemctl stop httpd #停掉R2的httpd服务
[[email protected] keepalived]# ipvsadm -Ln #K1上查看,192.168.11.30:80的被自动去掉了
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 192.168.11.17:80 rr
-> 192.168.11.21:80 Route 1 0 0
[[email protected] keepalived]# ipvsadm -Ln #K2上查看,192.168.11.30:80的被自动去掉了,更K1是一样的
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 192.168.11.17:80 rr
-> 192.168.11.21:80 Route 1 0 0
[[email protected] ~]# service httpd stop #停掉R1的httpd服务
Stopping httpd: [ OK ]
[[email protected] keepalived]# ipvsadm -Ln #使用了我们的定义的sorry_server
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 192.168.11.17:80 rr
-> 127.0.0.1:80 Local 1 0 0
[[email protected] ~]# curl http://192.168.11.17 #如我们所要的结果
<h1>===localhost--K1==</h1>
[[email protected] keepalived]# service keepalived stop #停掉K1的keepalived
停止 keepalived: [确定]
[[email protected] ~]# curl http://192.168.11.17 #如我们所要的结果,VIP跑去了K2,至此keepalived搭建的简单httpd高可用就搭建OK啦 ^-^
<h1>===localhost--K2===</h1>