實(shí)驗(yàn)環(huán)境

server1 和 server2 是調(diào)度器，server3 和 server4 是服務(wù)器

一、調(diào)度器中進(jìn)行配置 haproxy+keepalived

server1 與server2 均要配置

【1】下載keepalived

yum install -y keepalived

【2】server1 中修改配置文件

[root@server1 keepalived]# cat keepalived.conf ! Configuration File for keepalivedglobal_defs {notification_email {root@localhost}notification_email_from keepalived@localhostsmtp_server 127.0.0.1smtp_connect_timeout 30router_id LVS_DEVELvrrp_skip_check_adv_addr#vrrp_strictvrrp_garp_interval 0vrrp_gna_interval 0 }vrrp_script check_haproxy {#script "killall -0 haproxy"script "/opt/check_haproxy.sh" #可以使用腳本，也可以直接killallinterval 2weight 0 }vrrp_instance VI_1 {state MASTERinterface eth0virtual_router_id 65priority 100advert_int 1authentication {auth_type PASSauth_pass 1111}track_script {check_haproxy}virtual_ipaddress {172.25.0.100} }

【3】server2 中修改配置文件（server2是backup，且優(yōu)先級低于server1）

[root@server2 keepalived]# cat keepalived.conf ! Configuration File for keepalivedglobal_defs {notification_email {root@localhost}notification_email_from keepalived@localhostsmtp_server 127.0.0.1smtp_connect_timeout 30router_id LVS_DEVELvrrp_skip_check_adv_addr#vrrp_strictvrrp_garp_interval 0vrrp_gna_interval 0 }vrrp_script check_haproxy {#script "killall -0 haproxy"script "/opt/check_haproxy.sh" #可以使用腳本，也可以直接killallinterval 2weight 0}vrrp_instance VI_1 {state BACKUPinterface eth0virtual_router_id 65priority 50 advert_int 1authentication {auth_type PASSauth_pass 1111}track_script {check_haproxy}virtual_ipaddress {172.25.0.100} }

二、編寫腳本

記得給可執(zhí)行權(quán)限
兩臺(tái)虛擬機(jī)中都需要?jiǎng)?chuàng)建此腳本

[root@server1 opt]# vim /opt/check_haproxy.sh #!/bin/bash systemctl status haproxy &> /dev/null || systemctl restart haproxy &> /dev/null killall -0 haproxyif [ $? -ne 0 ];thensystemctl stop keepalived fi[root@server1 opt]# chmod +x /opt/check_haproxy.sh #可執(zhí)行權(quán)限

三、haproxy

可以先修改好一臺(tái)虛擬機(jī)中的/etc/haproxy/haproxy.cfg文件，然后使用scp命令發(fā)送到另一臺(tái)虛擬機(jī)上

# use_backend static if url_staticacl read_request method GETacl read_request method HEADacl write_request method PUTacl write_request method POSTuse_backend static if read_requestuse_backend app if write_requestdefault_backend static # default_backend app #--------------------------------------------------------------------- # static backend for serving up images, stylesheets and such #--------------------------------------------------------------------- backend staticbalance roundrobinserver static 172.25.15.3:80 check #--------------------------------------------------------------------- # round robin balancing between the various backends #--------------------------------------------------------------------- backend appbalance roundrobin# balance source# balance static-rrserver app1 172.25.15.4:80 checkserver backup 127.0.0.1:80 backup

四、測試

【1】初始狀態(tài)
兩臺(tái)虛擬機(jī)中的haproxy和keepalived都是打開狀態(tài)。此時(shí)172.25.15.100是在server1上（因?yàn)閟erver1的是master優(yōu)先級高），server2沒有172.25.15.100

【2】驗(yàn)證高可用

在server1上：systemctl stop haproxy

此時(shí)172.25.15.100不再server1上了
關(guān)閉haproxy后172.25.15.100直接遷移到server2中了，實(shí)現(xiàn)了高可用

【3】驗(yàn)證優(yōu)先級
重新打開server1中的haproxy之后，ip又會(huì)遷移回來。這是因?yàn)閟erver1中的優(yōu)先級高于server2

總結(jié)

以上是生活随笔為你收集整理的高可用集群下的负载均衡（7）：haproxy + keepalived 实现高可用负载均衡的全部內(nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。