引言

Nginx是一个高性能的代理服务器,单台Nginx容易出现单点故障,使用keepalived可以实现Nginx的故障转移,保证了网站的高可用性

一、使用Nginx+keepalived的两种方案

1、主从模式

使用一个VIP,前端有2台服务器,一主一从,正常情况下是主服务器提供服务只有当主服务器不能正常提供服务之后,从服务器才提供服务,此时总会有一台服务器是空闲状态。

2、双主模式

使用两个VIP,前段有2台服务器,互为主从,两台服务器同时工作,不存在资源浪费情况。同时在前段的DNS服务器对网站做多条A记录,实现了Nginx
的负载均衡,当一台服务器故障时候,资源会转移到另一台服务器,继续提供服务,在大型的网站中多数都使用此种架构。在此使用主主模式配置
Nginx+keepalived的高可用性。

二、准备实验环境

1、服务器IP地址规划

VIP:172.16.10.8

VIP:172.16.10.9

Keepalived1:172.16.10.1

Keepalived2:172.16.10.2

2、服务器操作系统

Keepalived1:Centos 6.4 x86_64

Keepalived2:Centos 6.4 x86_64

3、网络拓扑图

4、修改主机名以及hosts文件keepalived1

1
2
3
4
5
6
7
8
9
10
11
12
13
####keepalived1 server############
sed -i 's@\(HOSTNAME=\).*@\1keepalived1@g'/etc/sysconfig/network
hostname keepalived1
[root@keepalived1 ~]# echo "172.16.10.1 keepalived1">> /etc/hosts
[root@keepalived1 ~]# echo "172.16.10.2 keepalived2">> /etc/hosts
[root@keepalived1 ~]# ssh-keygen -t rs
[root@keepalived1 ~]# ssh-copy-id -i .ssh/id_rsa.pub keepalived2
[root@keepalived1 ~]# scp /etc/hosts keepalived1:/etc/
####keepalived2 server############
sed -i 's@\(HOSTNAME=\).*@\1keepalived2@g'/etc/sysconfig/network
hostname keepalived2
[root@keepalived2 ~]# ssh-keygen -t rsa
[root@keepalived2 ~]# ssh-copy-id -i .ssh/id_rsa.pub keepalived1

三、编译安装Nginx

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
[root@keepalived1 ~]# yum install openssl-devel pcre-devel gcc -y
[root@keepalived1 ~]# tar xf nginx-1.4.2.tar.gz -C /usr/local/
[root@keepalived1 ~]# cd /usr/local/
[root@keepalived1 local]# groupadd -r nginx
[root@keepalived1 local]# useradd -r -g nginx nginx
[root@keepalived1 local]# cd nginx-1.4.2/
[root@keepalived1 nginx-1.4.2]# ./conf
conf/      configure
[root@keepalived1 nginx-1.4.2]# ./configure \
> --prefix=/usr \
>    --sbin-path=/usr/sbin/nginx \
>    --conf-path=/etc/nginx/nginx.conf \
>    --error-log-path=/var/log/nginx/error.log \
>    --http-log-path=/var/log/nginx/access.log \
>    --pid-path=/var/run/nginx/nginx.pid  \
>    --lock-path=/var/lock/nginx.lock \
>    --user=nginx \
>    --group=nginx \
>    --with-http_ssl_module \
>    --with-http_flv_module \
>    --with-http_stub_status_module \
>    --with-http_gzip_static_module \
>    --http-client-body-temp-path=/var/tmp/nginx/client/ \
>    --http-proxy-temp-path=/var/tmp/nginx/proxy/ \
>    --http-fastcgi-temp-path=/var/tmp/nginx/fcgi/ \
>    --http-uwsgi-temp-path=/var/tmp/nginx/uwsgi \
>    --http-scgi-temp-path=/var/tmp/nginx/scgi \
>    --with-pcre
[root@keepalived1 nginx-1.4.2]# make && make install

注意 在此只上传了keepalived1的代码,keepalived2也需要同样的操作

1、修改默认网页以方便后期测试

1
2
3
4
###############keepalived1######################
[root@keepalived1 ~]# echo "<h1>keepalived1</h1>" > /usr/html/index.html
###############keepalived2######################
[root@keepalived2 ~]# echo "<h1>keepalived2</h1>" > /usr/html/index.html

四、 安装与配置keepalived

1、安装keepalived

1
2
3
4
###############keepalived1######################
[root@keepalived1 ~]# yum install keepalived -y
###############keepalived2######################
[root@keepalived2 ~]# yum install keepalived -y

2、修改配置文件

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
[root@keepalived1 keepalived]# grep -v "#" /etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs {
   notification_email {
     root@localhost
   }
   notification_email_from Alexandre.Cassen@localhost
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id LVS_DEVEL
}
 vrrp_script chk_nginx {        #监控nginx脚本
    script "killall -0 nginx"   #监控nginx进程
    interval 1                  #监控间隔
    weight -2                   #优先级-2
}
vrrp_instance VI_1 {
    state MASTER                 #主server 
    interface eth0             
    virtual_router_id 80
      priority 100               #优先级
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        172.16.10.8            #定义vip
    }
    track_script {
    chk_nginx                 #跟踪脚本
}
    notify_master "/etc/keepalived/notify8.sh master"  #定义邮件通知
    notify_backup "/etc/keepalived/notify8.sh backup"
    notify_fault "/etc/keepalived/notify8.sh fault"
}
                                                                                                                                                              
vrrp_instance VI_2 {
    state BACKUP            #从server
    interface eth0
    virtual_router_id 81
    priority 99
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        172.16.10.9
    }
    track_script {
    chk_nginx
}
    notify_master "/etc/keepalived/notify9.sh master" #定义邮件通知
    notify_backup "/etc/keepalived/notify9.sh backup"
    notify_fault "/etc/keepalived/notify9.sh fault"
}
[root@keepalived1 keepalived]#

3、编辑邮件通知脚本(notify8.sh notify9.sh)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
#####################notify8.sh##############
[root@keepalived1 keepalived]# cat notify8.sh
#!/bin/bash
# Author: xiaodong <xiaodong@foxmail.com>
# description: An example of notify script
#
vip=172.16.10.8
contact='root@localhost'
notify() {
    mailsubject="`hostname` to be $1: $vip floating"
    mailbody="`date '+%F %H:%M:%S'`: vrrp transition, `hostname` changed to be $1"
    echo $mailbody | mail -s "$mailsubject" $contact
}
case "$1" in
    master)
        notify master
        /etc/rc.d/init.d/nginx start
        exit 0
    ;;
    backup)
        notify backup
        /etc/rc.d/init.d/nginx stop
       exit 0
    ;;
    fault)
        notify fault
        exit 0
    ;;
    *)
        echo 'Usage: `basename $0` {master|backup|fault}'
        exit 1
    ;;
esac
####################notfiy9.sh#################
[root@keepalived1 keepalived]# cat notify9.sh
#!/bin/bash
# Author: xiaodong <xiaodong@foxmail.com>
# description: An example of notify script
#
vip=172.16.10.9
contact='root@localhost'
notify() {
    mailsubject="`hostname` to be $1: $vip floating"
    mailbody="`date '+%F %H:%M:%S'`: vrrp transition, `hostname` changed to be $1"
    echo $mailbody | mail -s "$mailsubject" $contact
}
case "$1" in
    master)
        notify master
         exit 0
    ;;
    backup)
        notify backup
       exit 0
    ;;
    fault)
        notify fault
        exit 0
    ;;
    *)
        echo 'Usage: `basename $0` {master|backup|fault}'
        exit 1
    ;;
esac
[root@keepalived1 keepalived]# chmod +x notify8.sh
[root@keepalived1 keepalived]# chmod +x notify9.sh

4、复制配置文件到keepalived2,并做修改.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
[root@keepalived1 keepalived]# scp -p  keepalived.conf notify8.sh notify9.sh keepalived2:/etc/keepalived/
[root@keepalived2 keepalived]# grep -v "#" /etc/keepalived/keepalived.conf
! Configuration File for keepalived
global_defs {
   notification_email {
     root@localhost
   notification_email_from Alexandre.Cassen@localhost
   smtp_server 127.0.0.1
   smtp_connect_timeout 30
   router_id LVS_DEVEL
}
  vrrp_script chk_nginx {
   script "killall -0 nginx "
    interval 1
    weight -2
}
vrrp_instance VI_1 {
    state BACKUP                   #改为backup
    interface eth0
    virtual_router_id 80
    priority 99                    #改为99
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        172.16.10.8
    }
    track_script {
        chk_nginx
}
    notify_master "/etc/keepalived/notify.sh master"
    notify_backup "/etc/keepalived/notify.sh backup"
    notify_fault "/etc/keepalived/notify.sh fault"
}
vrrp_instance VI_2 {
    state MASTER              #改为MASTER
    interface eth0
    virtual_router_id 81
    priority 100              #改为100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1111
    }
    virtual_ipaddress {
        172.16.10.9
    }
    track_script {
        chk_nginx
}
    notify_master "/etc/keepalived/notify9.sh master"
    notify_backup "/etc/keepalived/notify9.sh backup"
    notify_fault "/etc/keepalived/notify9.sh fault"
}

注释:此处使用本地的邮件服务器接受邮件,如果需要用其它邮件服务器请修改contact='root@localhost'

5、启动keepalived服务

1
2
3
4
###############keepalived1######################
[root@keepalived1 ~]# service keepalived start
###############keepalived2######################
[root@keepalived2 ~]# service keepalived start

6、查看两个节点的vip是否启动正常

五、测试nginx+keepalived的高可用性

1、使用游览器访问测试

 

2、模拟节点出现故障,nginx服务器是否能自动转移

1
[root@keepalived1 keepalived]# service keepalived stop

 

通过以上测试,节点出现故障的时候,服务可以自动转移到备用节点上

3、测试主节点服务down掉之后,备用节点服务是否能正常运行

1
2
[root@keepalived1 keepalived]# service keepalived start
[root@keepalived1 keepalived]# killall nginx

通过以上测试,实现了Nginx的高可用性,但是,运维人员是否能第一时间得知服务器出现故障,这时候就需要查看邮件了

4、查看邮件是否收到信息

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
[root@keepalived1 keepalived]# mail   #查看邮件命令
Heirloom Mail version 12.4 7/29/08.  Type ? for help.
"/var/spool/mail/root": 1 message 1 new
>N  1 root                  Wed Sep 25 20:15  18/728   "keepalived1 to be backup: 172.16.10.8 floating"
& 1
Message  1:
From root@keepalived1.localdomain  Wed Sep 25 20:15:46 2013
Return-Path: <root@keepalived1.localdomain>
X-Original-To: root@localhost
Delivered-To: root@localhost.localdomain
Date: Wed, 25 Sep 2013 20:15:46 +0800
To: root@localhost.localdomain
Subject: keepalived1 to be backup: 172.16.10.8 floating
User-Agent: Heirloom mailx 12.4 7/29/08
Content-Type: text/plain; charset=us-ascii
From: root@keepalived1.localdomain (root)
Status: R
2013-09-25 20:15:46: vrrp transition, keepalived1 changed to be backup
& quit                               #退出邮件

5、当nginx服务启动之后,主节点恢复

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
[root@keepalived1 keepalived]# service nginx start
[root@keepalived1 keepalived]# mail
Heirloom Mail version 12.4 7/29/08.  Type ? for help.
"/var/spool/mail/root": 2 messages 1 unread
    1 root                  Wed Sep 25 20:15  19/739   "keepalived1 to be backup: 172.16.10.8 floating"
>U  2 root                  Wed Sep 25 20:16  19/738   "keepalived1 to be master: 172.16.10.8 floating"
&
Message  2:
From root@keepalived1.localdomain  Wed Sep 25 20:16:22 2013
Return-Path: <root@keepalived1.localdomain>
X-Original-To: root@localhost
Delivered-To: root@localhost.localdomain
Date: Wed, 25 Sep 2013 20:16:22 +0800
To: root@localhost.localdomain
Subject: keepalived1 to be master: 172.16.10.8 floating
User-Agent: Heirloom mailx 12.4 7/29/08
Content-Type: text/plain; charset=us-ascii
From: root@keepalived1.localdomain (root)
Status: RO
2013-09-25 20:16:22: vrrp transition, keepalived1 changed to be master

Nginx+keepalived的高可用负载均衡配置完成,本博客至此结束,如有任何疑问请留言!

 

Nginx+keepalived构建双主负载均衡代理服务器的更多相关文章

  1. Nginx keepalived实现高可用负载均衡详细配置步骤

    Keepalived是一个免费开源的,用C编写的类似于layer3, 4 & 7交换机制软件,具备我们平时说的第3层.第4层和第7层交换机的功能.主要提供loadbalancing(负载均衡) ...

  2. Centos7.2下基于Nginx+Keepalived搭建高可用负载均衡(一.基于Keepalived搭建HA体系)

    说明 本文只为方便日后查阅,不对一些概念再做赘述,网上都有很多明确的解释,也请大家先了解相关概念. 两台搭建HA的服务器是华为云上的ECS(不要忘记开通VPC,保证我们的服务器都处在一个内网环境),由 ...

  3. Nginx+Keepalived+Tomcat高可用负载均衡,Zookeeper集群配置,Mysql(MariaDB)搭建,Redis安装,FTP配置

    JDK 安装步骤 下载 http://www.oracle.com/technetwork/java/javase/downloads/jdk8-downloads-2133151.html rpm ...

  4. nginx+keepalived高可用web负载均衡

    一:安装环境 准备2台虚拟机,都安装好环境 centos 7keepalived:vip: 192.168.1.112192.168.1.110 nginxip 192.168.1.109 maste ...

  5. 基于Haproxy+Keepalived构建高可用负载均衡集群

    实验环境: 主机名 IP地址 VIP 192.168.200.254 Haproxy-1 192.168.200.101 Haproxy-2 192.168.200.102 Nginx1 192.16 ...

  6. Haproxy+Keepalived构建高可用负载均衡集群

    实验环境: 主机名 IP地址 VIP 192.168.200.254 Haproxy-1 192.168.200.101 Haproxy-2 192.168.200.102 Nginx1 192.16 ...

  7. nginx+keepalived(双主)

    一.环境 nginx1   192.168.40.211 nginx2   192.168.40.132 vip1     192.168.40.223  主为keep1,从为keep2 vip2   ...

  8. Nginx+Keepalived搭建高可用负载均衡集群

    本文的重点是Keepalived的配置,Nginx的配置就简略带过.软件:CentOS 7.2 / Nginx 1.12.2 / Keepalived 1.3.9 ha-01:192.168.1.97 ...

  9. HAProxy+Keepalived构建高可用负载均衡

    http://www.linuxidc.com/Linux/2012-03/55880.htm web1 IP 192.168.0.47 web2 IP 192.168.0.48 haproxy_ma ...

随机推荐

  1. javaweb基础(11)_cookie的会话管理

    一.会话的概念 会话可简单理解为:用户开一个浏览器,点击多个超链接,访问服务器多个web资源,然后关闭浏览器,整个过程称之为一个会话. 有状态会话:一个同学来过教室,下次再来教室,我们会知道这个同学曾 ...

  2. 转 Hystrix入门指南 Introduction

    https://www.cnblogs.com/gaoyanqing/p/7470085.html

  3. Angular - angularjs2 一些报错的概览(数据为json格式)

    {"Unterminated string literal.": "未终止的字符串文本.","Identifier expected.": ...

  4. Windows 10+Ubuntu双系统修复Ubuntu启动引导

    U盘启动,联网 $ sudo su sudo add-apt add-apt-repository ppa:yannubuntu/boot-repair apt-get update apt-get ...

  5. Python自定义一个数组类,支持数组之间的四则运算和其他常见方法

    class MyArray: '''保证输入的内容是整型.浮点型''' def ___isNumber(self, num): if not isinstance(num, (int,float)): ...

  6. matplotlib学习记录 五

    # 绘制电影票房竖条形图 from matplotlib import pyplot as plt a = ["战狼2","速度与激情8","功夫瑜伽 ...

  7. 我的Python分析成长之路6

    模块:本质就是.py结尾的文件.从逻辑上组织python代码. 包: 本质就是一个目录,带有__init__.py文件,从逻辑上组织模块. 模块的分类: 1.标准库(内置的模块) 2.开源库(第三方库 ...

  8. 通过SWD J-Link使用J-Link RTT Viewer来查看打印日志

    详细的说明可以参考:https://www.cnblogs.com/iini/p/9279618.html sdk版本: 15.2.0 例程目录:\nRF5_SDK_15.2.0_9412b96\ex ...

  9. Educational Codeforces Round 41 E. Tufurama (961E)

    [题解] 第一眼看题飞快地想到一种做法,然后假掉了. 这道题其实是主席树的模板题来着.但是也有别的水法. 我们可以发现每个位置的查询区间是[1,min(a[i],i-1)],所以我们可以把查询区间按照 ...

  10. Docker容器技术的核心原理

    目录 1 前言 2 docker容器技术 2.1 隔离:Namespace 2.2 限制:Cgroup 2.3 rootfs 2.4 镜像分层 3 docker容器与虚拟机的对比 1 前言 上图是百度 ...