| 服务器IP | hostname | 节点说明 | 端口 | 管控台地址 | 账号 | 密码 |
|---|---|---|---|---|---|---|
| 192.168.0.115 | mq-01 | rabbitmq master | 5672 | http://192.168.0.115:15672 | guest | guest |
| 192.168.0.117 | mq-02 | rabbitmq slave | 5672 | http://192.168.0.117:15672 | guest | guest |
| 192.168.0.118 | mq-03 | rabbitmq slave | 5672 | http://192.168.0.118:15672 | guest | guest |
| 192.168.0.119 | hk-01 | haproxy+keepalived | 8100 | http://192.168.0.119:8100/rabbitmq-stats | admin | 123456 |
| 192.168.0.120 | hk-02 | haproxy+keepalived | 8100 | http://192.168.0.120:8100/rabbitmq-stats | admin | 123456 |
| 命令 | 说明 |
|---|---|
| sudo service keepalived start | 启动keepalived 服务 |
| sudo service keepalived stop | 停止keepalived 服务 |
| sudo service keepalived restart | 重新启动keepalived 服务 |
| sudo service keepalived status | 查看keepalived 服务运行状态 |
| sudo chkconfig keepalived on | keepalived 服务开机启动 |
| 命令 | 说明 |
|---|---|
| sudo systemctl start keepalived.service | 启动keepalived 服务 |
| sudo systemctl stop keepalived.service | 停止keepalived 服务 |
| sudo systemctl restart keepalived.service | 重新启动keepalived 服务 |
| sudo systemctl status keepalived.service | 查看keepalived 服务运行状态 |
| sudo systemctl enable keepalived.service | keepalived 服务开机启动 |
| sudo systemctl disable keepalived.service | keepalived 取消服务开机启动 |
接上一篇：RabbitMQ+haproxy+keeplived 高可用负载均衡+镜像集群模式_集成负载均衡组件 Ha-Proxy_02
文章目录
- 一、Keepalived简介
- 二、Keepalived 安装实战
- 2.1. 安装所需软件包
- 2.2. 下载keepalived 软件包
- 2.3. 同步keepalived 软件包
- 2.4. 解压、编译、安装 keepalived
- 三、将keepalived安装成Linux系统服务
- 3.1. 首先创建文件夹，将keepalived配置文件进行复制
- 3.2. 然后复制 keepalived 脚本文件
- 3.3. keepalived 服务设置开机启动
- 四、配置+修改 Keepalived配置文件
- 4.1. 创建并编辑keepalived.conf文件
- 4.2. 119 服务器配置
- 4.3. 同步+修改 keepalived.conf 配置文件
- 4.4. 执行脚本编写
- 4.5. 脚本说明：
- 4.6. 执行脚本赋权
- 五、启动 keepalived 服务
- 5.1. 查看haproxy启运行状态
- 5.2. 启动keepalived
- 5.3. 查看keepalived运行状态
- 六、测试+验证 keepalived 单点故障转移
- 6.1. 正常场景测试
- 6.2. 主节点出现单点故障 测试
- 6.3. 主节点出现单点故障+重新启动主节点 测试
一、Keepalived简介
Keepalived，它是一个高性能的服务器高可用或热备解决方案，Keepalived主要来防止服务器单点故障的发生问题，可以通过其与Nginx、Haproxy等反向代理的负载均衡服务器配合实现web服务端的高可用。Keepalived以VRRP协议为实现基础，用VRRP协议来实现高可用性（HA）.VRRP（Virtual Router Redundancy Protocol）协议是用于实现路由器冗余的协议，VRRP协议将两台或多台路由器设备虚拟成一个设备，对外提供虚拟路由器IP（一个或多个）。
二、Keepalived 安装实战
PS:下载地址：
http://www.keepalived.org/download.html
2.1. 安装所需软件包
yum install -y openssl openssl-devel
2.2. 下载keepalived 软件包
wget https://www.keepalived.org/software/keepalived-2.0.20.tar.gz
2.3. 同步keepalived 软件包
为了节省时间，将此软件包同步120服务器
scp keepalived-2.0.20.tar.gz root@192.168.0.120:/app/software
2.4. 解压、编译、安装 keepalived
# 解压keepalived
tar -zxf keepalived-2.0.20.tar.gz -C /app/
# 编译、安装 keepalived
cd keepalived-2.0.20/ && ./configure --prefix=/app/keepalived
make && make install
三、将keepalived安装成Linux系统服务
将keepalived安装成Linux系统服务，因为没有使用keepalived的默认安装路径（默认路径：/usr/local）,安装完成之后，需要做一些修改工作
3.1. 首先创建文件夹，将keepalived配置文件进行复制
# 创建文件夹
mkdir /etc/keepalived
# 将keepalived配置文件进行复制
cp /app/keepalived/etc/keepalived/keepalived.conf /etc/keepalived/
3.2. 然后复制 keepalived 脚本文件
cp /app/keepalived-2.0.20/keepalived/etc/init.d/keepalived /etc/init.d/
cp /app/keepalived/etc/sysconfig/keepalived /etc/sysconfig/
# 建立软连接
ln -s /app/sbin/keepalived /usr/sbin/
# 由于系统默认建立软连接，因此需要先删除默认的
rm -f /sbin/keepalived
# 和自己指定安装的脚本文件建立软连接
ln -s /app/keepalived/sbin/keepalived /sbin/
3.3. keepalived 服务设置开机启动
可以设置开机启动：systemctl enable keepalived.service，到此我们安装完毕!
systemctl enable keepalived.service
四、配置+修改 Keepalived配置文件
PS:修改keepalived.conf配置文件
4.1. 创建并编辑keepalived.conf文件
vim /etc/keepalived/keepalived.conf
4.2. 119 服务器配置
! Configuration File for keepalivedglobal_defs {router_id hk-01 ##标识节点的字符串，通常为hostname}vrrp_script chk_haproxy {script "/etc/keepalived/haproxy_check.sh" ##执行脚本位置interval 2 ##检测时间间隔weight -20 ##如果条件成立则权重减20
}
#监测haproxy进程状态，每2秒执行一次
vrrp_instance VI_1 {state MASTER ## 主节点为MASTER，备份节点为BACKUPinterface ens33 ## 绑定虚拟IP的网络接口（网卡），与本机IP地址所在的网络接口相同（我这里是eth0）virtual_router_id 119 ## 虚拟路由ID号（主备节点一定要相同）mcast_src_ip 192.168.0.119 ## 本机ip地址priority 100 ##优先级配置（0-254的值）nopreemptadvert_int 1 ## 组播信息发送间隔，俩个节点必须配置一致，默认1s
authentication { ## 认证匹配auth_type PASSauth_pass ncl@1234}track_script {chk_haproxy}virtual_ipaddress {192.168.0.112 ## 虚拟ip，可以指定多个}
}
4.3. 同步+修改 keepalived.conf 配置文件
将此配置文件同步120服务器
scp keepalived.conf root@192.168.0.120:/etc/keepalived/
修改第一处：将router_id 修改为120 服务器hostname
修改第二处：mcast_src_ip 修改为120 本机ip地址
修改第三处：priority 修改为 90 ## 主节点 100 从节点 90
修改第四处：state 将 MASTER 修改为BACKUP
120服务器配置：
! Configuration File for keepalivedglobal_defs {router_id hk-02 ##标识节点的字符串，通常为hostname}
#监测haproxy进程状态，每2秒执行一次
vrrp_script chk_haproxy {script "/etc/keepalived/haproxy_check.sh" ##执行脚本位置interval 2 ##检测时间间隔weight -20 ##如果条件成立则权重减20
}vrrp_instance VI_1 {state BACKUP ## 主节点为MASTER，备份节点为BACKUPinterface ens33 ## 绑定虚拟IP的网络接口（网卡），与本机IP地址所在的网络接口相同（我这里是eth0）virtual_router_id 119 ## 虚拟路由ID号（主备节点一定要相同）mcast_src_ip 192.168.0.120 ## 本机ip地址priority 90 ##优先级配置（0-254的值）nopreemptadvert_int 1 ## 组播信息发送间隔，俩个节点必须配置一致，默认1s
authentication { ## 认证匹配auth_type PASSauth_pass ncl@1234}track_script {chk_haproxy}virtual_ipaddress {192.168.0.112 ## 虚拟ip，可以指定多个}
}
4.4. 执行脚本编写
PS:添加文件位置为/etc/keepalived/haproxy_check.sh（119、120两个节点文件内容一致即可）
vim /etc/keepalived/haproxy_check.sh
#!/bin/bash
COUNT=`ps -C haproxy --no-header |wc -l`
if [ $COUNT -eq 0 ];then/app/haproxy/sbin/haproxy -f /etc/haproxy/haproxy.cfgsleep 2if [ `ps -C haproxy --no-header |wc -l` -eq 0 ];thenkillall keepalivedfi
fi
4.5. 脚本说明：
- 监测haproxy进程状态，每2秒执行一次
- 如果存在，则跳过逻辑
- 如果不存在，重新启动 haproxy 服务
4.6. 执行脚本赋权
PS:haproxy_check.sh脚本授权,赋予可执行权限.
chmod +x /etc/keepalived/haproxy_check.sh
五、启动 keepalived 服务
PS:启动keepalived之前先检查haproxy启运行状态
5.1. 查看haproxy启运行状态
命令查看方式：
ps -ef | grep haproxy
浏览器查看方式：
从上面可以看出haproxy 已经正常运行
5.2. 启动keepalived
PS:当我们启动俩个haproxy节点以后，我们可以启动keepalived服务程序
# 启动两台机器的keepalived
service keepalived start
5.3. 查看keepalived运行状态
ps -ef | grep keepalived
六、测试+验证 keepalived 单点故障转移
6.1. 正常场景测试
测试结果预测:
keepalived 服务正常启动，虚拟ip在 主节点的服务器上(119 服务器)
在119服务器 查看虚拟IP
ip a
在120服务器上是没有虚拟IP的(除非主节点119 keepalived服务停止)
在119服务器 查看虚拟IP
ip a
6.2. 主节点出现单点故障 测试
测试结果预测:
1. keepalived 服务正常启动，虚拟ip在 主节点的服务器上(119 服务器)
2. 当主节点出现单点故障后，虚拟ip会漂移到BACKUP节点
模拟虚拟IP漂移到 120服务器场景
停止主节点119 keepalived服务停止
service keepalived stop
再次查看119 是否存在虚拟IP
ip a
再次查看120 虚拟IP是否漂移成功
ip a
从上面模拟测试结果，符合咱们的预测！！！
6.3. 主节点出现单点故障+重新启动主节点 测试
1. keepalived 服务正常启动，虚拟ip在 主节点的服务器上(119 服务器)
2. 当主节点出现单点故障后，虚拟ip会漂移到BACKUP节点
3. 当主节点单点故障修复后，由于咱们设置了权重，主节点权重比从节点权重大，因此，虚拟IP会重新回到主节点服务器上
二次模拟测试，权重是否设置正常
119 权重权重 100
120 设置权重 90
yuce测试结果：
当主节点119 keepalived节点再次启动，虚拟ip又会回到主节点服务器上
主节点119服务器再次启动 keepalived 服务，进行模拟测试
[root@hk-01 keepalived]# service keepalived start
Starting keepalived (via systemctl): [ OK ]
[root@hk-01 keepalived]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00inet 127.0.0.1/8 scope host lovalid_lft forever preferred_lft foreverinet6 ::1/128 scope host valid_lft forever preferred_lft forever
2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000link/ether 00:0c:29:2a:fc:5d brd ff:ff:ff:ff:ff:ffinet 192.168.0.119/24 brd 192.168.0.255 scope global noprefixroute dynamic ens33valid_lft 70561sec preferred_lft 70561secinet 192.168.0.112/32 scope global ens33valid_lft forever preferred_lft foreverinet6 fe80::ac32:9647:2dd9:bed5/64 scope link noprefixroute valid_lft forever preferred_lft forever
[root@hk-01 keepalived]#
120 服务器 查看测试结果
[root@hk-02 keepalived]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00inet 127.0.0.1/8 scope host lovalid_lft forever preferred_lft foreverinet6 ::1/128 scope host valid_lft forever preferred_lft forever
2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000link/ether 00:0c:29:f9:d8:e3 brd ff:ff:ff:ff:ff:ffinet 192.168.0.120/24 brd 192.168.0.255 scope global noprefixroute dynamic ens33valid_lft 70574sec preferred_lft 70574secinet6 fe80::ac32:9647:2dd9:bed5/64 scope link tentative noprefixroute dadfailed valid_lft forever preferred_lft foreverinet6 fe80::c92d:38e3:9ea0:a936/64 scope link noprefixroute valid_lft forever preferred_lft forever
[root@hk-02 keepalived]#
