安装nginx: yum install gcc-c++ yum install -y pcre pcre-devel yum install -y zlib zlib-devel yum install -y openssl openssl-devel
Nginx ensure normal access to:
I'm here now 129 146 as follows:
安装keepalived: 1.安装基础必备包. yum install popt-devel 2. 编译. tar -zxvf keepalived-1.2.15.tar.gz -C /usr/local/ cd keepalived-1.2.15/ && ./configure --prefix=/usr/local/keepalived make && make install 3.配置相关文件 mkdir /etc/keepalived cp /usr/local/keepalived/etc/keepalived/keepalived.conf /etc/keepalived/ cp /usr/local/keepalived/etc/rc.d/init.d/keepalived /etc/init.d/ cp /usr/local/keepalived/etc/sysconfig/keepalived /etc/sysconfig/ ln -s /usr/local/sbin/keepalived /usr/sbin/ LN -s / usr / local / keepalived / sbin / keepalived / sbin / 4 . vim / etc / keepalived / keepalived.conf Master: vrrp_script chk_nginx { Script "/etc/keepalived/nginx_check.sh" # run the script, the script content below there is a play nginx down after automatic opening service interval the 2 # detection interval weight -20 # If the conditions are satisfied, then the weights -20 } # define virtual routing, VI_1 virtual route identifier, define their own name {VI_1 vrrp_instance State to determine the master from the mASTER # interface virtual IP network interface eth0 # binding, according to their fill machine virtual_router_id 121 # virtual routing ID, must be provided as two nodes mcast_src_ip 192.168.139.129 # fill local ip priority 100 # node priority, mainly from the node higher priority than nopreempt high priority setting # nopreempt solve the problem again after recovery from the abnormality seize the advert_int 1 # multicast transmission interval, two nodes must be set as the default 1S authentication { PASS AUTH_TYPE AUTH_PASS 1111 } # track_script block will join instance configuration block track_script { chk_nginx # Nginx execution monitoring service } virtual_ipaddress { 192.168.139.110 # virtual ip, ip how to solve that is to write a program ip dead switching can also be extended, Wide range of uses. You can configure multiple top three must be consistent with the server. } } Service keepalived Start start without error keepalived successful installation:
Backup: vrrp_script chk_nginx { Script "/etc/keepalived/nginx_check.sh" # run the script, the script has the following content is played down after a nginx automatically open service interval 2 # detection interval weight -20 # If the conditions are satisfied , then the weight of -20 } # define virtual routing, VI_1 virtual route identifier himself define the name vrrp_instance VI_1 { State the BACKUP # to determine the master from the virtual IP network interface interface eth2 # binding, filled according to their own machine virtual_router_id # 121 virtual routing ID number, set up two nodes must be the same mcast_src_ip 192.168.139.129 # fill the machine ip priority # 100 node priority, mainly from the node higher priority than nopreempt high priority setting # nopreempt resolved abnormal recovery again preemption problem advert_int 1 # multicast transmission interval, two nodes must be set as the default lS authentication { AUTH_TYPE the PASS AUTH_PASS 1111 } # Join the track_script block instance configuration block {track_script chk_nginx # Nginx execution monitoring service } virtual_ipaddress { 192.168.139.110 # virtual ip, ip how to solve that is to write a program ip dead switch, but also scalable, and versatile. You can configure multiple top three must be consistent with the server. } } Keepalived common command -Service keepalived Start -Service keepalived STOP 1. achieve the effect of access to the master node nginx virtual access configuration by vip ip
Master node from the node without vip ip 110
3: Restart the primary site can be seen vip ip need to wait a few seconds
Double-point main completed structures (a master-slave nginx) from the hot spare architecture
longer access Tomcat cluster actually made by vip ip access nginx nginx keeplived generated