Keepalived 高可用配置

Linux › Keepalived
,
  • Keepalived 是一种高性能的服务器高可用或热备解决方案，Keepalived 可以用来防止服务器单点故障的发生，通过配合 Nginx 可以实现 Web 前端服务的高可用

1. 安装 Keepalived

1
2
3
4
5
6
7
8
9
$ sudo apt install keepalived

$ tar -zxvf keepalived-2.0.18.tar.gz
$ cd keepalived-2.0.18
$ ./configure --prefix=/usr/local/keepalived
$ make && make install
$ cp keepalived/etc/keepalived/keepalived.conf /etc/keepalived/
$ cp keepalived/etc/init.d/keepalived /etc/init.d/
$ cp keepalived/etc/sysconfig/keepalived /etc/sysconfig/

2. 准备检查脚本

(1) check_nginx.sh

1
2
3
4
5
6
7
8
9
10
$ sudo vi /etc/keepalived/check_nginx.sh
#!/bin/bash
COUNT=`ps -C nginx --no-header |wc -l`
if [ $COUNT -eq 0 ]; then
    echo 'nginx server is died'
    sudo systemctl stop keepalived.service  # 需要停止keepalived服务才能漂移到其他节点
    #killall keepalived
fi

$ sudo chmod +x /etc/keepalived/check_nginx.sh

(2) check_nacos.sh

1
2
3
4
5
6
7
8
9
$ sudo vi /etc/keepalived/check_nacos.sh
#!/bin/bash
COUNT=$(ps -ef | grep nacos |grep -v "check_nacos" | grep -v "grep" | wc -l )
if [ $COUNT -eq 0 ]; then
  echo 'nacos server is died'
  sudo systemctl stop keepalived.service
fi

$ sudo chmod +x /etc/keepalived/check_nacos.sh

3. 配置 Keepalived

  • MASTER
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
$ sudo vi /etc/keepalived/keepalived.conf
global_defs {
    router_id debian101
}

vrrp_script check_nginx_alive {
    script "/etc/keepalived/check_nginx.sh"
    interval 3
    weight -10
}

vrrp_instance VI_1 {
    state MASTER
    interface ens32
    mcast_src_ip 10.4.7.101
    virtual_router_id 51
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1643
    }
    virtual_ipaddress {
        10.4.7.100
    }
    track_script {
        check_nginx_alive
    }
}

vrrp_script check_nacos_alive {
    script "/etc/keepalived/check_nacos.sh"
    interval 2
}

vrrp_instance VI_2 {
    state MASTER
    interface ens32
    mcast_src_ip 10.4.7.101
    virtual_router_id 52
    priority 103
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 16434
    }
    track_script {
        check_nacos_alive
    }
    virtual_ipaddress {
        10.4.7.100
    }
}
  • BACKUP
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
$ sudo vi /etc/keepalived/keepalived.conf
vrrp_script check_nginx_alive {
    script "/etc/keepalived/check_nginx.sh"
    interval 3
    weight -10
}

global_defs {
    router_id debian102
}

vrrp_instance VI_1 {
    state BACKUP
    interface ens32
    mcast_src_ip 10.4.7.102
    unicast_src_ip 10.4.7.102
    virtual_router_id 51
    priority 90
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 1643
    }
    virtual_ipaddress {
        10.4.7.100
    }
    track_script {
        check_nginx_alive
    }
}

vrrp_script check_nacos_alive {
    script "/etc/keepalived/check_nacos.sh"
    interval 2
}

vrrp_instance VI_2 {
    state BACKUP
    interface ens32
    mcast_src_ip 10.4.7.102
    unicast_src_ip 10.4.7.102
    virtual_router_id 52
    priority 100
    advert_int 1
    authentication {
        auth_type PASS
        auth_pass 16434
    }
    virtual_ipaddress {
        10.4.7.100
    }
    track_script {
        check_nacos_alive
    }
}
  • 重启服务
1
2
$ sudo systemctl restart keepalived
$ sudo systemctl enable keepalived

4. 配置文件详解

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
global_defs {
    # keepalived自带的邮件提醒需要开启sendmail服务。建议用独立的监控或第三方SMTP
    notification_email {
        xxx@xxx.xxx
    }
    notification_email_from xxx@xxx.xxx
    smtp_server 127.0.0.1
    smtp_connect_timeout 30
    # 标识本节点的字条串，通常为hostname
    router_id LVS_DEVEL
}

# keepalived会定时执行脚本并对执行结果进行分析，并动态调整vrrp_instance的优先级priority
# 如果脚本执行结果为0，并且weight配置的值大于0，则优先级相应的增加
# 如果脚本执行结果非0，并且weight配置的值小于0，则优先级相应的减少
# 其他情况，维持原本配置的优先级，即配置文件中priority对应的值
vrrp_script check_nginx_alive {
    # 检测服务状态的脚本路径
    script "/etc/keepalived/check_nginx.sh"
    # 检测时间间隔
    interval 3
    # 如果条件成立，权重-10
    weight -10
}

# 定义虚拟路由，VI_1为虚拟路由的标示符，可自己定义
vrrp_instance VI_1 {
    # 主节点为MASTER，对应的备份节点为BACKUP
    state MASTER
    # 绑定虚拟IP的网络接口，与本机IP地址所在的网络接口相同
    interface ens32
    # 本机IP地址
    mcast_src_ip 10.4.7.101
    # 虚拟路由的ID号，多个节点设置必须一样，可选IP最后一段使用。相同的VRID为一个组，它将决定多播的MAC地址
    virtual_router_id 51
    # 节点优先级，值范围0-254，MASTER要比BACKUP高
    priority 100
    # 组播信息发送间隔，多个节点设置必须一样，默认1s
    advert_int 1
    # 设置验证信息，多个节点必须一致
    authentication {
        auth_type PASS
        # 生产环境注意强度
        auth_pass 1643
    }
    # 虚拟IP池，两个节点设置必须一样
    virtual_ipaddress {
        # 虚拟ip，可以定义多个
        10.4.7.100
    }
    # 将track_script块加入instance配置块
    track_script {
        # 执行监控的服务
        check_nginx_alive
    }
}

参考