本文介绍了keepalived，它基于VRRP协议实现高可用架构，与lvs、nginx等配合，有健康检查和失败切换两大核心功能。还阐述了VRRP协议原理。此外，详细说明了高可用负载集群的配置过程，包括主机部署、keepalived主机设置、配置文件修改等，并进行了故障模拟测试。
一.keepalived的简单介绍
1.keepalived
keepalived是基于VRRP协议实现高可用架构，VRRP是路由灾备的实现基础。他与其他负载均衡技术例如lvs、nginx等一起工作达到集群高可用
keepalived作用是检验调度器的状态，在负载均衡时，所有的数据包请求均需要经过调度器进行调度转发。因此当调度器万一发生故障则整个集群系统都会崩溃，所以使用keeplived实现集群的高可用，部署两台或者多台调度器，仅一台作为主调度器，其他调度器作为备用，当主调度器发生故障时，keepalived可以自动将备用调度器升为主调度器，最终实现集群的高负载高可用
健康检查和失败切换是keepalived的两大核心功能。 所谓的健康检查， 就是采用tcp三次握手， icmp请求， http请求， udp echo请求等方式对负载均衡器后面的实际的服务器(通常是承载真实业务的服务器)进行保活；而失败切换主要是应用于配置了主备模式的负载均衡器， 利用VRRP 维持主备负载均衡器的心跳， 当主负载均衡器出现问题时， 由备负载均衡器承载对应的业务， 从而在最大限度上减少流量损失， 并提供服务的稳定性。
2.VRRP协议
VRRP全称 Virtual Router Redundancy Protocol，即虚拟路由冗余协议。它是实现路由器高可用的容错协议，即将N台提供相同功能的路由器组成一个路由器组(Router Group)，这个组里面有一个master和多个backup，但在外界看来就像一台一样，构成虚拟路由器，拥有一个虚拟IP（vip，也就是路由器所在局域网内其他机器的默认路由），占有这个IP的master实际负责ARP相应和转发IP数据包，组中的其它路由器作为备份的角色处于待命状态。master会发组播消息，当backup在超时时间内收不到vrrp包时就认为master宕掉了，这时就需要根据VRRP的优先级来选举一个backup当master，保证路由器的高可用。
二.高可用负载集群的配置
1.实验主机部署
keepalived主机1：172.25.4.111
keepalived主机2：172.25.4.112
后端服务器1：172.25.4.113
后端服务器2：172.25.4.114
2.keeplived两台主机的设置
server1
[root@server1 ~]# cd /mnt
[root@server1 mnt]# ls
keepalived-2.0.6.tar.gz ##源码包先编译再安装
[root@server1 mnt]# tar zxf keepalived-2.0.6.tar.gz ##解压源码包
[root@server1 mnt]# ls
keepalived-2.0.6 keepalived-2.0.6.tar.gz
[root@server1 mnt]# cd keepalived-2.0.6
[root@server1 keepalived-2.0.6]# ls
aclocal.m4 compile depcomp keepalived missing
ar-lib configure doc keepalived.spec.in README.md
AUTHOR configure.ac genhash lib snap
bin_install CONTRIBUTORS INSTALL Makefile.am TODO
ChangeLog COPYING install-sh Makefile.in
[root@server1 keepalived-2.0.6]# yum install gcc openssl-devel -y ##安装对源码包进行编译的辅助软件
[root@server1 keepalived-2.0.6]# ./configure --prefix=/usr/local/keepalived --with-init=systemd ##源码包进行编译
[root@server1 keepalived-2.0.6]# make && make install ##源码包进行安装
[root@server1 keepalived-2.0.6]# yum install ipvsadm -y
server2
[root@server2 ~]# cd /mnt
[root@server2 mnt]# ls
keepalived-2.0.6.tar.gz
[root@server2 mnt]# tar zxf keepalived-2.0.6.tar.gz
[root@server2 mnt]# ls
keepalived-2.0.6 keepalived-2.0.6.tar.gz
[root@server2 mnt]# cd keepalived-2.0.6
[root@server2 keepalived-2.0.6]# ls
aclocal.m4 compile depcomp keepalived missing
ar-lib configure doc keepalived.spec.in README.md
AUTHOR configure.ac genhash lib snap
bin_install CONTRIBUTORS INSTALL Makefile.am TODO
ChangeLog COPYING install-sh Makefile.in
[root@server2 keepalived-2.0.6]# yum install gcc openssl-devel -y
[root@server2 keepalived-2.0.6]# ./configure --prefix=/usr/local/keepalived --with-init=systemd
[root@server2 keepalived-2.0.6]# make && make install
[root@server2 keepalived-2.0.6]# yum install ipvsadm -y ##安装管理工具仅安装不编辑策略
3.设置链接方便访问keepalived配置文件
server1
[root@server1 keepalived-2.0.6]# cd /usr/local/keepalived/
[root@server1 keepalived]# ls
bin etc sbin share
[root@server1 keepalived]# ln -s /usr/local/keepalived/etc/sysconfig/keepalived /etc/sysconfig/
[root@server1 keepalived]# ln -s /usr/local/keepalived/etc/keepalived/ /etc
[root@server1 keepalived]# ln -s /usr/local/keepalived/sbin/keepalived /sbin/
server2
[root@server2 keepalived-2.0.6]# ln -s /usr/local/keepalived/etc/sysconfig/keepalived /etc/sysconfig/
[root@server2 keepalived-2.0.6]# ln -s /usr/local/keepalived/etc/keepalived/ /etc
[root@server2 keepalived-2.0.6]# ln -s /usr/local/keepalived/sbin/keepalived /sbin/
4.keepalived配置文件的修改
server1
[root@server1 keepalived]# cd /etc/keepalived/
[root@server1 keepalived]# ls
keepalived.conf samples
[root@server1 keepalived]# vim keepalived.conf
1 ! Configuration File for keepalived
2
3 global_defs {
4 notification_email {
5 root@localhost ##节点宕机了将会接收到异常邮件的主机
6 }
7 notification_email_from keepalived@localhost ##邮件发送人
8 smtp_server 127.0.0.1 ##发送的服务器
9 smtp_connect_timeout 30 #指定连接超时时间
10 router_id LVS_DEVEL
11 vrrp_skip_check_adv_addr
12 # vrrp_strict
13 vrrp_garp_interval 0
14 vrrp_gna_interval 0
15 }
16
17 vrrp_instance VI_1 {
18 state MASTER ##主节点
19 interface eth0
20 virtual_router_id 14
21 priority 100 ##权重
22 advert_int 1 ##检查间隔时间为1s
23 authentication {
24 auth_type PASS ##认证方式
25 auth_pass 1111 ##认证密码
26 }
27 virtual_ipaddress {
28 172.25.4.100 ##虚拟ip
29 }
30 }
31
32 virtual_server 172.25.4.100 80 {
33 delay_loop 6 ##连接失败6次以后发送邮件
34 lb_algo rr ##lvs的调度算法为rr
35 lb_kind DR ##lvs的调度方法为DR轮询
36 # persistence_timeout 50
37 protocol TCP ##端口
38
39 real_server 172.25.4.113 80 {
40 TCP_CHECK {
41 weight 1
42 connect_port 80
43 connect_timeout 3
44 }
45 }
46
47 real_server 172.25.4.114 80 {
48 TCP_CHECK {
49 weight 1
50 connect_port 80
51 connect_timeout 3
52 }
53 }
54 }
[root@server1 keepalived]# systemctl start keepalived
[root@server1 keepalived]# systemctl enable keepalived
server2
1 ! Configuration File for keepalived
2
3 global_defs {
4 notification_email {
5 root@localhost
6 }
7 notification_email_from keepalived@localhost
8 smtp_server 127.0.0.1
9 smtp_connect_timeout 30
10 router_id LVS_DEVEL
11 vrrp_skip_check_adv_addr
12 # vrrp_strict
13 vrrp_garp_interval 0
14 vrrp_gna_interval 0
15 }
16
17 vrrp_instance VI_1 {
18 state BACKUP ##主节点
19 interface eth0
20 virtual_router_id 14
21 priority 50 ##权重小
22 advert_int 1
23 authentication {
24 auth_type PASS
25 auth_pass 1111
26 }
27 virtual_ipaddress {
28 172.25.4.100
29 }
30 }
31
32 virtual_server 172.25.4.100 80 {
33 delay_loop 6
34 lb_algo rr
35 lb_kind DR
36 # persistence_timeout 50
37 protocol TCP
38
39 real_server 172.25.4.113 80 {
40 TCP_CHECK {
41 weight 1
42 connect_port 80
43 connect_timeout 3
44 }
45 }
46
47 real_server 172.25.4.114 80 {
48 TCP_CHECK {
49 weight 1
50 connect_port 80
51 connect_timeout 3
52 }
53 }
54 }
[root@server2 keepalived]# systemctl start keepalived
[root@server2 keepalived]# systemctl enable keepalived
5.后端服务器的设置
（1）apache的安装
server3
[root@server3 ~]# yum install -y httpd
[root@server3 ~]# vim /var/www/html/index.html
[root@server3 ~]# cat /var/www/html/index.html
server3
[root@server3 ~]# systemctl start httpd
[root@server3 ~]# systemctl enable httpd
server4
[root@server4 ~]# yum install httpd -y
[root@server4 ~]# vim /var/www/html/index.html
[root@server4 ~]# cat /var/www/html/index.html
server4
[root@server4 ~]# systemctl start httpd
[root@server4 ~]# systemctl enable httpd
（2）虚拟vip的添加
[root@server3 ~]# ip addr add 172.25.4.100/24 dev eth0
[root@server4 ~]# ip addr add 172.25.4.100/24 dev eth0
（3）火墙策略的设置
[root@server3 ~]# yum install arptables -y
[root@server3 ~]# arptables -A INPUT -d 172.25.4.100 -j DROP ##不回应对自己的请求
[root@server3 ~]# arptables -A OUTPUT -s 172.25.4.100 -j mangle --mangle-ip-s 172.25.4.113 ##将发送出去的数据包的源ip修改成VIP
[root@server3 ~]# arptables -L
Chain INPUT (policy ACCEPT)
-j DROP -d server3
Chain OUTPUT (policy ACCEPT)
-j mangle -s server3 --mangle-ip-s server3
Chain FORWARD (policy ACCEPT)
[root@server4 ~]# yum install arptables -y
[root@server4 ~]# arptables -A INPUT -d 172.25.4.100 -j DROP
[root@server4 ~]# arptables -A OUTPUT -s 172.25.4.100 -j mangle --mangle-ip-s 172.25.4.114
[root@server4 ~]# arptables -L
Chain INPUT (policy ACCEPT)
-j DROP -d server4
Chain OUTPUT (policy ACCEPT)
-j mangle -s server4 --mangle-ip-s server4
Chain FORWARD (policy ACCEPT)
6.在两个调度器上查看生成的策略
[root@server1 keepalived]# ipvsadm
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP server1:http rr
[root@server1 keepalived]# ipvsadm
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP server1:http rr
-> server3:http Masq 1 0 0
-> server4:http Masq 1 0 0
[root@server2 keepalived]# ipvsadm
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 172.25.4.100:http rr
-> server3:http Masq 1 0 0
-> server4:http Masq 1 0 0
7.物理机测试
[root@foundation4 kiosk]# curl 172.25.4.100
server4
[root@foundation4 kiosk]# curl 172.25.4.100
server3
[root@foundation4 kiosk]# curl 172.25.4.100
server4
[root@foundation4 kiosk]# curl 172.25.4.100
server3
[root@foundation4 kiosk]# curl 172.25.4.100
server4
[root@foundation4 kiosk]# curl 172.25.4.100
server3
8.模拟故障
（1）当后端服务器中的其中一台httpd损坏
[root@server3 ~]# systemctl stop httpd
此时调度器状态
[root@server1 keepalived]# ipvsadm
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP server1:http rr
-> server4:http Route 1 0 0 ##被关闭的服务器自动从调度器列表中移出
You have new mail in /var/spool/mail/root
物理机测试
[root@foundation4 kiosk]# curl 172.25.4.100
server4
[root@foundation4 kiosk]# curl 172.25.4.100
server4
[root@foundation4 kiosk]# curl 172.25.4.100
server4
（2）将损坏的后端服务器恢复
[root@server3 ~]# systemctl start httpd
此时调度器状态
[root@server1 keepalived]# ipvsadm
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP server1:http rr
-> server3:http Route 1 0 0 ##恢复的后端服务器被自动重新加入策略中
-> server4:http Route 1 0 4
You have new mail in /var/spool/mail/root
物理机测试
[root@foundation4 kiosk]# curl 172.25.4.100
server3
[root@foundation4 kiosk]# curl 172.25.4.100
server4
[root@foundation4 kiosk]# curl 172.25.4.100
server3
[root@foundation4 kiosk]# curl 172.25.4.100
server4
[root@foundation4 kiosk]# curl 172.25.4.100
server3
[root@foundation4 kiosk]# curl 172.25.4.100
server4
扫一扫
432
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
