介绍：
有了keepalived+Lvs这样的高性能组合，为什么还需keepalived+Nginx呢。keepalived是为了Lvs而设计。Lvs是一个四层的负载均衡设备，虽然有着高性能的优势，但它无后端服务器的健康检查机制。keepalived为lvs提供一系列的健康检查机制，例如：TCP_CHECK,UDP_CHECK,HTTP_GET等。同时lvs也可以自己写健康检查脚脚本。或者结合ldirectory来实现后端健康检测。但LVS始终无法摆脱它是一个四层设备，无法对上层协议进行解析。而Nginx就不一样了，Nginx是一个七层的设备可以对七层协议进行解析，可以对一些请求进行过滤，还可以对请求结果进行缓存。这些都是Nginx独有的优势。但是keepalived并没有为Nginx提供健康检测。需要自己去写一些脚步来进行健康检测。
下面主要讲解Keepalived+Nginx的模式，不包含lvs。如果不是大型负载，一般用不到LVS，当然你也可以参阅：《Keepalived LVS-DR Nginx单网络双活双主配置模式（实战）》篇。
准备四台服务器或虚拟机：
Web Nginx 内网：10.16.8.8/10.16.8.9
Keepalived 内网：10.16.8.10（ka67）/10.16.8.11（ka68）
Keepalived 公网：172.16.8.10/172.16.8.11Keepalived 内网VIP：10.16.8.100/10.16.8.101
Keepalived 公网VIP：172.16.8.100/172.16.8.101OS：CentOS Linux release 7.4.1708 (Core)
先决条件：
安装keepalived。
时间同步。
设置SELinux和防火墙。
互相之间/etc/hosts文件添加对方主机名（可选）。
确认网络接口支持多播（组播）新网卡默认支持。
以上部署请参阅：《keepalived 安装及配置文件讲解》。
1.ka67配置文件
global_defs {
notification_email {
root@localhost
}
notification_email_from ka@localhost
smtp_server 127.0.0.1
smtp_connect_timeout 30
vrrp_mcast_group4 224.0.0.111
}
vrrp_instance External_1 {
state MASTER
interface eth1
virtual_router_id 171
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass renwole0
}
virtual_ipaddress {
10.16.8.100
}
notify_master "/usr/local/keepalived/etc/keepalived/notify.sh master"
notify_backup "/usr/local/keepalived/etc/keepalived/notify.sh backup"
notify_fault "/usr/local/keepalived/etc/keepalived/notify.sh fault"
}
vrrp_instance External_2 {
state BACKUP
interface eth1
virtual_router_id 172
priority 95
advert_int 1
authentication {
auth_type PASS
auth_pass renwole1
}
virtual_ipaddress {
10.16.8.101
}
notify_master "/usr/local/keepalived/etc/keepalived/notify.sh master"
notify_backup "/usr/local/keepalived/etc/keepalived/notify.sh backup"
notify_fault "/usr/local/keepalived/etc/keepalived/notify.sh fault"
}
vrrp_instance Internal_1 {
state MASTER
interface eth0
virtual_router_id 191
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass renwole2
}
virtual_ipaddress {
172.16.8.100
}
notify_master "/usr/local/keepalived/etc/keepalived/notify.sh master"
notify_backup "/usr/local/keepalived/etc/keepalived/notify.sh backup"
notify_fault "/usr/local/keepalived/etc/keepalived/notify.sh fault"
}
vrrp_instance Internal_2 {
state BACKUP
interface eth0
virtual_router_id 192
priority 95
advert_int 1
authentication {
auth_type PASS
auth_pass renwole3
}
virtual_ipaddress {
172.16.8.101
}
notify_master "/usr/local/keepalived/etc/keepalived/notify.sh master"
notify_backup "/usr/local/keepalived/etc/keepalived/notify.sh backup"
notify_fault "/usr/local/keepalived/etc/keepalived/notify.sh fault"
}
2.ka68配置文件
global_defs {
notification_email {
root@localhost
}
notification_email_from ka@localhost
smtp_server 127.0.0.1
smtp_connect_timeout 30
vrrp_mcast_group4 224.0.0.111
}
vrrp_instance External_1 {
state BACKUP
interface eth1
virtual_router_id 171
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass renwole0
}
virtual_ipaddress {
10.16.8.100
}
notify_master "/usr/local/keepalived/etc/keepalived/notify.sh master"
notify_backup "/usr/local/keepalived/etc/keepalived/notify.sh backup"
notify_fault "/usr/local/keepalived/etc/keepalived/notify.sh fault"
}
vrrp_instance External_2 {
state MASTER
interface eth1
virtual_router_id 172
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass renwole1
}
virtual_ipaddress {
10.16.8.101
}
notify_master "/usr/local/keepalived/etc/keepalived/notify.sh master"
notify_backup "/usr/local/keepalived/etc/keepalived/notify.sh backup"
notify_fault "/usr/local/keepalived/etc/keepalived/notify.sh fault"
}
vrrp_instance Internal_1 {
state BACKUP
interface eth0
virtual_router_id 191
priority 95
advert_int 1
authentication {
auth_type PASS
auth_pass renwole2
}
virtual_ipaddress {
172.16.8.100
}
notify_master "/usr/local/keepalived/etc/keepalived/notify.sh master"
notify_backup "/usr/local/keepalived/etc/keepalived/notify.sh backup"
notify_fault "/usr/local/keepalived/etc/keepalived/notify.sh fault"
}
vrrp_instance Internal_2 {
state MASTER
interface eth0
virtual_router_id 192
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass renwole3
}
virtual_ipaddress {
172.16.8.101
}
notify_master "/usr/local/keepalived/etc/keepalived/notify.sh master"
notify_backup "/usr/local/keepalived/etc/keepalived/notify.sh backup"
notify_fault "/usr/local/keepalived/etc/keepalived/notify.sh fault"
}
3.创建检测通用脚本
$ vim /usr/local/keepalived/etc/keepalived/notify.sh
#!/bin/bash
#
contact='root@localhost'
notify() {
local mailsubject="$(hostname) to be $1, vip floating"
local mailbody="$(date +'%F %T'): vrrp transition, $(hostname) changed to be $1"
echo "$mailbody" | mail -s "$mailsubject" $contact
}
case $1 in
master)
notify master
;;
backup)
notify backup
systemctl start nginx # 此处配置后，Nginx服务挂了能自动启动
;;
fault)
notify fault
;;
*)
echo "Usage: $(basename $0) {master|backup|fault}"
exit 1
;;
esac
4.启动keepalived服务并测试
启动ka67后查看其网卡状态：
[root@ka67 ~]# systemctl start keepalived
[root@ka67 ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1
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
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000
link/ether 00:15:5d:ae:02:78 brd ff:ff:ff:ff:ff:ff
inet 172.16.8.10/24 brd 172.16.8.255 scope global eth0
valid_lft forever preferred_lft forever
inet 172.16.8.100/32 scope global eth0
valid_lft forever preferred_lft forever
inet 172.16.8.101/32 scope global eth0
valid_lft forever preferred_lft forever
inet6 fe80::436e:b837:43b:797c/64 scope link
valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000
link/ether 00:15:5d:ae:02:84 brd ff:ff:ff:ff:ff:ff
inet 10.16.8.10/24 brd 10.16.8.255 scope global eth1
valid_lft forever preferred_lft forever
inet 10.16.8.100/32 scope global eth1
valid_lft forever preferred_lft forever
inet 10.16.8.101/32 scope global eth1
valid_lft forever preferred_lft forever
inet6 fe80::1261:7633:b595:7719/64 scope link
valid_lft forever preferred_lft forever
在ka68没有启动时，ka67添加了4个VIP，分别是：
公网eth0：
172.16.8.100/32
172.16.8.101/32
内网eth1：
10.16.8.100/32
10.16.8.101/32
启动ka68后查看其网卡状态：
[root@ka68 ~]# systemctl start keepalived
[root@ka68 ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1
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
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000
link/ether 00:15:5d:ae:02:79 brd ff:ff:ff:ff:ff:ff
inet 172.16.8.11/24 brd 103.28.204.255 scope global eth0
valid_lft forever preferred_lft forever
inet 172.16.8.101/32 scope global eth0
valid_lft forever preferred_lft forever
inet6 fe80::3d2c:ecdc:5e6d:70ba/64 scope link
valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000
link/ether 00:15:5d:ae:02:82 brd ff:ff:ff:ff:ff:ff
inet 10.16.8.11/24 brd 10.16.8.255 scope global eth1
valid_lft forever preferred_lft forever
inet 10.16.8.101/32 scope global eth1
valid_lft forever preferred_lft forever
inet6 fe80::4fb3:d0a8:f08c:4536/64 scope link
valid_lft forever preferred_lft forever
ka68添加了2个VIP，分别是：
公网eth0：
172.16.8.101/32
内网eth1：
10.16.8.101/32
再次查看ka67的网卡状态信息：
[root@ka67 ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1
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
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000
link/ether 00:15:5d:ae:02:78 brd ff:ff:ff:ff:ff:ff
inet 172.16.8.10/24 brd 172.16.8.255 scope global eth0
valid_lft forever preferred_lft forever
inet 172.16.8.100/32 scope global eth0
valid_lft forever preferred_lft forever
inet6 fe80::436e:b837:43b:797c/64 scope link
valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP qlen 1000
link/ether 00:15:5d:ae:02:84 brd ff:ff:ff:ff:ff:ff
inet 10.16.8.10/24 brd 10.16.8.255 scope global eth1
valid_lft forever preferred_lft forever
inet 10.16.8.100/32 scope global eth1
valid_lft forever preferred_lft forever
inet6 fe80::1261:7633:b595:7719/64 scope link
valid_lft forever preferred_lft forever
注意到 172.16.8.101/10.16.8.101 已经被移除了，此时无论停掉任意一台服务器，4个VIP都不会停止通信。
另外可以在ka67/ka68通过如下命令查看组播地址的心跳状态：
[root@ka67 ~]# tcpdump -nn -i eth1 host 224.0.0.111
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on eth1, link-type EN10MB (Ethernet), capture size 262144 bytes 02:00:15.690389 IP 10.16.8.10 > 224.0.0.111: VRRPv2, Advertisement, vrid 171, prio 100, authtype simple, intvl 1s, length 20 02:00:15.692654 IP 10.16.8.11 > 224.0.0.111: VRRPv2, Advertisement, vrid 172, prio 100, authtype simple, intvl 1s, length 20 02:00:16.691552 IP 10.16.8.10 > 224.0.0.111: VRRPv2, Advertisement, vrid 171, prio 100, authtype simple, intvl 1s, length 20 02:00:16.693814 IP 10.16.8.11 > 224.0.0.111: VRRPv2, Advertisement, vrid 172, prio 100, authtype simple, intvl 1s, length 20 02:00:17.692710 IP 10.16.8.10 > 224.0.0.111: VRRPv2, Advertisement, vrid 171, prio 100, authtype simple, intvl 1s, length 20
到目前为止，vrrp的高可用配置&测试已完成，接下来我们继续配置Web Nginx服务。
5.安装并配置Nginx
分别在后端服务器 10.16.8.8/10.16.8.9 安装Nginx：
关于Nginx请参阅：《Centos 7源码编译安装 Nginx》。
或通过以下方式yum安装Nginx；简单快速：
$ yum install epel-release -y $ yum install nginx -y
测试环境为区分机器的不同，故将web页面设置服务器IP地址，但在生产环境中获取的内容是一致的。
分别在10.16.8.8/10.16.8.9执行如下命令：
$ echo "Server 10.16.8.8" > /usr/share/nginx/html/index.html $ echo "Server 10.16.8.9" > /usr/share/nginx/html/index.html
测试是否访问正常：
$ curl //10.16.8.8 Server 10.16.8.8
分别在ka67/ka68上安装Nginx，我这里用yum安装：
$ yum install nginx psmisc -y
说明：psmisc包含了：fuser,killall,pstree等命令。
在ka67/ka68上配置Nginx：
备份默认配置文件：
$ mv /etc/nginx/conf.d/default.conf{,.bak}
$ mv /etc/nginx/nginx.conf{,.bak}
分别在ka67/ka68将nginx主配置文件中添加如下内容：
$ vim /etc/nginx/nginx.conf
user nginx;
worker_processes auto;
error_log /var/log/nginx/error.log;
pid /run/nginx.pid;
include /usr/share/nginx/modules/*.conf;
events {
worker_connections 1024;
}
http {
log_format main '$remote_addr - $remote_user [$time_local] "$request" '
'$status $body_bytes_sent "$http_referer" '
'"$http_user_agent" "$http_x_forwarded_for"';
access_log /var/log/nginx/access.log main;
sendfile on;
tcp_nopush on;
tcp_nodelay on;
keepalive_timeout 65;
types_hash_max_size 2048;
include /etc/nginx/mime.types;
default_type application/octet-stream;
include /etc/nginx/conf.d/*.conf;
upstream webserverapps {
server 10.16.8.8:80;
server 10.16.8.9:80;
#server 127.0.0.1:8080 backup;
}
server {
listen 80;
server_name _;
location / {
proxy_pass //webserverapps;
proxy_redirect off;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
client_max_body_size 10m;
client_body_buffer_size 128k;
proxy_connect_timeout 90;
proxy_send_timeout 90;
proxy_read_timeout 90;
proxy_buffer_size 4k;
proxy_buffers 4 32k;
proxy_busy_buffers_size 64k;
proxy_temp_file_write_size 64k;
add_header Access-Control-Allow-Origin *;
}
}
}
注意：以上配置主要添加了蓝色部分，其他默认，仅为测试使用。生产环境请根据自己需求调整配置。
在ka67/ka68重启Nginx服务：
$ systemctl restart nginx
分别在ka67/ka68上测试：
[root@ka67 ~]# for i in `seq 10`; do curl 10.16.8.10; done Server 10.16.8.8 Server 10.16.8.9 Server 10.16.8.8 Server 10.16.8.9 Server 10.16.8.8 Server 10.16.8.9 Server 10.16.8.8 Server 10.16.8.9 Server 10.16.8.9 Server 10.16.8.9
到目前为止，Nginx反代功能也已实现，下面我们将把Nginx与Keepalived结合起来，使Nginx支持高可用。
6.配置Keepalived Nginx高可用
分别在ka67/ka68配置文件/usr/local/keepalived/etc/keepalived/keepalived.conf的全局配置块global_defs下方添加vrrp_script配置块：
vrrp_script chk_nginx {
script "killall -0 nginx"
interval 2
weight -10
fall 2
rise 2
}
在所有vrrp_instance实例块里，添加track_script块：
track_script {
chk_nginx
}
例如：
...
vrrp_instance External_1 {
state BACKUP
interface eth1
virtual_router_id 171
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass renwole0
}
virtual_ipaddress {
10.16.8.100
}
track_script {
chk_nginx
}
notify_master "/usr/local/keepalived/etc/keepalived/notify.sh master"
notify_backup "/usr/local/keepalived/etc/keepalived/notify.sh backup"
notify_"/usr/local/keepalived/etc/keepalived/notify.sh fault"
}
...
配置完以后，重启ka67/ka68的keepalived服务：
$ systemctl stop keepalived $ systemctl start keepalived
总结：
在配置过程中出现了无法漂移的情况，跨网段问题。解决通道，还是要多看日志，多分析判断，最终还是能解决问题的。无论在何种情况下，既然选择了keepalived，就要坚信自己的初心。
如你在配置过程中出现任何问题，欢迎留言，共同解决问题。