keepalived+nginx双机热备+负载均衡

keepalived+nginx双机热备+负载均衡

最近因业务扩展,需要将当前的apache 转为nginx(web), 再在web前端放置nginx(负载均衡)。同时结合keepalived 对前端nginx实现HA。
nginx进程基于于Master+Slave(worker)多进程模型,自身具有非常稳定的子进程管理功能。在Master进程分配模式下,Master进程永远不进行业务处理,只是进行任务分发,从而达到Master进程的存活高可靠性,Slave(worker)进程所有的业务信号都 由主进程发出,Slave(worker)进程所有的超时任务都会被Master中止,属于非阻塞式任务模型。
Keepalived是Linux下面实现VRRP 备份路由的高可靠性运行件。基于Keepalived设计的服务模式能够真正做到主服务器和备份服务器故障时IP瞬间无缝交接。二者结合,可以构架出比较稳定的软件lb方案。




准备4台电脑来做这个实验:

192.168.232.132        web服务器
192.168.232.133        web服务器
192.168.232.134        keepalived nginx
192.168.232.135        keepalived nginx

虚拟IP (VIP):192.168.232.16

134\135两个主机配置虚拟IP


下面以135为例:

vi /etc/sysconfig/network-scripts/ifcfg-eth2:0


		
  1. DEVICE=eth2:0
  2. TYPE=Ethernet
  3. ONBOOT=yes
  4. BOOTPROTO=static
  5. DNS1=192.168.232.2
  6. IPADDR=192.168.232.16
  7. NETMASK=255.255.255.0
  8. GETWAY=192.168.232.2



service network restart

使用ifconfig查看效果:

		
  1. eth2 Link encap:Ethernet HWaddr 00:0C:29:49:90:5B
  2. inet addr:192.168.232.135 Bcast:192.168.232.255 Mask:255.255.255.0
  3. inet6 addr: fe80::20c:29ff:fe49:905b/64 Scope:Link
  4. UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
  5. RX packets:66322 errors:0 dropped:0 overruns:0 frame:0
  6. TX packets:31860 errors:0 dropped:0 overruns:0 carrier:0
  7. collisions:0 txqueuelen:1000
  8. RX bytes:67624991 (64.4 MiB) TX bytes:2723877 (2.5 MiB)
  9. Interrupt:19 Base address:0x2000
  10. eth2:0 Link encap:Ethernet HWaddr 00:0C:29:49:90:5B
  11. inet addr:192.168.232.16 Bcast:192.168.232.255 Mask:255.255.255.0
  12. UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
  13. Interrupt:19 Base address:0x2000
  14. lo Link encap:Local Loopback
  15. inet addr:127.0.0.1 Mask:255.0.0.0
  16. inet6 addr: ::1/128 Scope:Host
  17. UP LOOPBACK RUNNING MTU:16436 Metric:1
  18. RX packets:22622 errors:0 dropped:0 overruns:0 frame:0
  19. TX packets:22622 errors:0 dropped:0 overruns:0 carrier:0
  20. collisions:0 txqueuelen:0
  21. RX bytes:1236328 (1.1 MiB) TX bytes:1236328 (1.1 MiB)


          
说明生效了。

134\135两个主机安装keepalived和nginx


nginx安装:

1、导入外部软件库
rpm -Uvh http://dl.iuscommunity.org/pub/ius/stable/Redhat/6/i386/epel-release-6-5.noarch.rpm
rpm -Uvh http://dl.iuscommunity.org/pub/ius/stable/Redhat/6/i386/ius-release-1.0-10.ius.el6.noarch.rpm
rpm -Uvh http://nginx.org/packages/centos/6/noarch/RPMS/nginx-release-centos-6-0.el6.ngx.noarch.rpm
以下添加注释
mirrorlist=http://dmirr.iuscommunity.org/mirrorlist?repo=ius-el6&arch=$basearch
以下删除注释
#baseurl=http://dl.iuscommunity.org/pub/ius/stable/Redhat/5/$basearch
2、yum安装nginx
yum install nginx

keepalived安装:

安装依赖
yum -y install gcc gcc+ gcc-c++
yum install popt-devel openssl openssl-devel libssl-dev libnl-devel popt-devel

安装内核
yum -y install kernel kernel-devel
当前kernel代码建立连接 ln -s /usr/src/kerners/2.6....../ /usr/src/linux

安装keepalived
wget http://www.keepalived.org/software/keepalived-1.2.2.tar.gz
tar -zxvf keepalived-1.2.2.tar.gz  
cd keepalived-1.2.2  
./configure  
make  
make install   

拷贝相应的文件

cp /usr/local/sbin/keepalived /usr/sbin/ 
cp /usr/local/etc/rc.d/init.d/keepalived /etc/init.d/  
cp /usr/local/etc/sysconfig/keepalived /etc/sysconfig/  
cp -r /usr/local/etc/keepalived/ /etc/  

配置keeplived和nginx主机


134/135执行都执行以下操作:
vi /etc/nginx/conf.d/default.conf

		
  1. server {
  2. listen 8088;
  3. server_name localhost;
  4. location / {
  5. root /var/www/html;
  6. index index.html index.htm;
  7. }
  8. error_page 500 502 503 504 /50x.html;
  9. location = /50x.html {
  10. root /usr/share/nginx/html;
  11. }
  12. }

135执行以下操作:
vi /var/www/html/index.html

		
  1. <html>
  2. <head>
  3. <title>Welcome to nginx!</title>
  4. </head>
  5. <body bgcolor="white" text="black">
  6. <center><h1>Welcome to nginx! 192.168.232.135</h1></center>
  7. </body>
  8. </html>

134执行以下操作:
vi /var/www/html/index.html

		
  1. <html>
  2. <head>
  3. <title>Welcome to nginx!</title>
  4. </head>
  5. <body bgcolor="white" text="black">
  6. <center><h1>Welcome to nginx! 192.168.232.134</h1></center>
  7. </body>
  8. </html>

134执行以下操作:
vi /etc/keepalived/keepalived.conf

		
  1. ! Configuration File for keepalived
  2. global_defs {
  3. notification_email {
  4. #acassen@firewall.loc
  5. #failover@firewall.loc
  6. #sysadmin@firewall.loc
  7. }
  8. #notification_email_from Alexandre.Cassen@firewall.loc
  9. #smtp_server 192.168.200.1
  10. #smtp_connect_timeout 30
  11. router_id LVS_DEVEL
  12. }
  13. vrrp_script chk_http_port {
  14. script "</dev/tcp/127.0.0.1/8088"
  15. interval 1
  16. weight -2
  17. }
  18. vrrp_instance VI_1 {
  19. state MASTER
  20. interface eth2
  21. virtual_router_id 51
  22. priority 100
  23. advert_int 1
  24. authentication {
  25. auth_type PASS
  26. auth_pass 1111
  27. }
  28. virtual_ipaddress {
  29. 192.168.232.16
  30. }
  31. track_script {
  32. chk_http_port
  33. }
  34. }

135执行以下操作:
vi /etc/keepalived/keepalived.conf

		
  1. ! Configuration File for keepalived
  2. global_defs {
  3. notification_email {
  4. #acassen@firewall.loc
  5. #failover@firewall.loc
  6. #sysadmin@firewall.loc
  7. }
  8. #notification_email_from Alexandre.Cassen@firewall.loc
  9. #smtp_server 192.168.200.1
  10. #smtp_connect_timeout 30
  11. router_id LVS_DEVEL
  12. }
  13. vrrp_script chk_http_port {
  14. script "</dev/tcp/127.0.0.1/8088"
  15. interval 1
  16. weight -2
  17. }
  18. vrrp_instance VI_1 {
  19. state BACKUP
  20. interface eth2
  21. virtual_router_id 51
  22. priority 99
  23. advert_int 1
  24. authentication {
  25. auth_type PASS
  26. auth_pass 1111
  27. }
  28. virtual_ipaddress {
  29. 192.168.232.16
  30. }
  31. track_script {
  32. chk_http_port
  33. }
  34. }

Tips:

state   参数值:主的是MASTER、备用的是BACKUP
priority 参数值: MASTER > BACKUP
virtual_router_id: 参数值要一样

测试测试:


两台测试机134\135均启动keepalived和nginx
service keepalived restart
service keepalived nginx

验证nginx启动正常:
访问 master:http://192.168.232.134:8088/
访问 backup: http://192.168.232.135:8088/

查看keepalived的日志信息:

134\135均打开日志信息方便查看keepalived动态:
tail -f /var/log/messages

浏览器打开虚拟ip访问:http://192.168.232.16:8080/ ,此时显示IP为192.168.232.134

服务器层的双机热备(比如服务器宕机、keepalived宕了)测试:


kill 192.168.232.134(master) 的keepalived进程
killall keepalived
134的日志信息如下:

		
  1. Jun 11 18:03:10 localhost Keepalived_vrrp: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth1 for 192.168.232.16
  2. Jun 11 18:03:15 localhost Keepalived_vrrp: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth1 for 192.168.232.16
  3. Jun 11 19:30:44 localhost Keepalived: Terminating on signal
  4. Jun 11 19:30:44 localhost Keepalived: Stopping Keepalived v1.2.2 (06/10,2014)
  5. Jun 11 19:30:44 localhost Keepalived_vrrp: Terminating VRRP child process on signal
  6. Jun 11 19:30:44 localhost Keepalived_healthcheckers: Terminating Healthchecker child process on signal


135的日志信息如下:

		
  1. Jun 11 19:30:50 localhost Keepalived_vrrp: VRRP_Instance(VI_1) setting protocol VIPs.
  2. Jun 11 19:30:50 localhost Keepalived_vrrp: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth2 for 192.168.232.16
  3. Jun 11 19:30:50 localhost Keepalived_healthcheckers: Netlink reflector reports IP 192.168.232.16 added
  4. Jun 11 19:30:55 localhost Keepalived_vrrp: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth2 for 192.168.232.16



刷新http://192.168.232.16:8080/ , 此时显示IP为192.168.232.135。

再次启动192.168.232.134的keepalived进程,192.168.232.134会自动接管成为master,192.168.232.135自动转为backup,从测试结果看,备机能成功接管,已经实现了热备。

应用层(web)的双机热备(比如nginx进程被意外kill、web端口不通)试验:


关闭192.168.232.134(master) 的nginx服务:
service nginx stop

134的日志信息如下:

		
  1. Jun 11 19:38:49 localhost Keepalived_vrrp: VRRP_Script(chk_http_port) failed
  2. Jun 11 19:38:51 localhost Keepalived_vrrp: VRRP_Instance(VI_1) Received higher prio advert
  3. Jun 11 19:38:51 localhost Keepalived_vrrp: VRRP_Instance(VI_1) Entering BACKUP STATE
  4. Jun 11 19:38:51 localhost Keepalived_vrrp: VRRP_Instance(VI_1) removing protocol VIPs.
  5. Jun 11 19:38:51 localhost Keepalived_healthcheckers: Netlink reflector reports IP 192.168.232.16 removed

135的日志信息如下:

		
  1. Jun 11 19:38:52 localhost Keepalived_vrrp: VRRP_Instance(VI_1) forcing a new MASTER election
  2. Jun 11 19:38:53 localhost Keepalived_vrrp: VRRP_Instance(VI_1) Transition to MASTER STATE
  3. Jun 11 19:38:54 localhost Keepalived_vrrp: VRRP_Instance(VI_1) Entering MASTER STATE
  4. Jun 11 19:38:54 localhost Keepalived_vrrp: VRRP_Instance(VI_1) setting protocol VIPs.
  5. Jun 11 19:38:54 localhost Keepalived_vrrp: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth2 for 192.168.232.16
  6. Jun 11 19:38:54 localhost Keepalived_healthcheckers: Netlink reflector reports IP 192.168.232.16 added
  7. Jun 11 19:38:59 localhost Keepalived_vrrp: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth2 for 192.168.232.16


刷新http://192.168.232.16:8080/ , 此时显示IP为192.168.232.135。

再次启动192.168.232.134的nginx进程,192.168.232.134会自动接管成为master,192.168.232.135自动转为backup,从测试结果看,备机能成功接管,已经实现了热备。

为什么主备的参数state都是MASTER,对的你没有看错确实要都设置成一样的,不然并不能实现我们想要的VIP漂浮的效果,我测试很久才发现的.state都设置成MASTER后,会根据priority的值大小竞争来决定谁是真正的MASTER,脚本检测也是在失败的时候会把权重减去相应的值,比如原来master(181)的priority=100,如果脚本检测到端口8088无法连接,就会priority-2=98,< S-B(150)的priority(99),此时 S-B(150) 将竞争成为master,这样就实现了web应用的热备。


如果以上实验都没有问题了,那么就该nginx负载均衡的配置了,配置修改参见如下:http://blog.csdn.net/e421083458/article/details/30086413