LVS高可用（七）LVS+keepalived双主
在 LVS高可用（六）LVS+keepalived主从 篇中提到了LVS的高可用及后端的LB，不过一台LVS提供服务，另一台只做为backup，显然是种浪费，实际现网应用中比较多的用法是，两台LVS都作为MASTER节点，互相做为另一台的backup 。具体架构如下：
注：与master-backup 架构相比， master-master架构需要多加一个VIP地址。
一、IP及规划
| 主机名 | IP地址 |
|---|---|
| realserver | 192.168.122.10 |
| realserver | 192.168.122.20 |
| director | 192.168.122.30 |
| director | 192.168.122.40 |
| VIP | 192.168.122.100 |
| VIP | 192.168.122.110 |
同上一篇，两台realserver还是安装httpd ，两台director安装ipvsadm、keepalived 。realserver的回环地址上会配置两个VIP的地址。另外操作前，需要先将几台主机的时间同步掉。
二、director主机配置
MASTER HostA主机的配置如下：
1[root@lvs-dr ~]# cat /etc/keepalived/keepalived.conf
2vrrp_instance bl_one {
3 state MASTER #指定Keepalived的角色，MASTER为主服务器，BACKUP为备用服务器
4 interface eth0 #指定HA监测的接口
5 lvs_sync_daemon_interface eth0
6 virtual_router_id 38 #虚拟路由标识(1-255)，在一个VRRP实例中主备服务器ID必须一样
7 priority 150 #优先级，数字越大越优先，主服务器优先级必须高于备服务器
8 advert_int 3 #设置主备之间同步检查时间间隔，单位秒
9 authentication {
10 auth_type PASS
11 auth_pass 1111
12 }
13 virtual_ipaddress { #定义虚拟IP地址
14 192.168.122.100
15 }
16}
17vrrp_instance bl_two {
18 state BACKUP
19 interface eth0
20 lvs_sync_daemon_interface eth0
21 virtual_router_id 48
22 priority 120
23 advert_int 3
24 authentication {
25 auth_type PASS
26 auth_pass 1111
27 }
28 virtual_ipaddress {
29 192.168.122.110
30 }
31}
32virtual_server 192.168.122.100 80 {
33 delay_loop 3 #设置健康状态检查时间
34 lb_algo rr #设置负载调度算法
35 lb_kind DR #设置LVS实现负载均衡的机制
36 persistence_timeout 50 #会话保持时间
37 protocol TCP
38 real_server 192.168.122.10 80 {
39 weight 1
40 TCP_CHECK {
41 connect_timeout 10 #设置响应超时时间
42 nb_get_retry 3 #设置超时重试次数
43 delay_before_retry 3 #设置超时重试间隔
44 connect_port 80
45 }
46 }
47 real_server 192.168.122.20 80 {
48 weight 1
49 TCP_CHECK {
50 connect_timeout 10
51 nb_get_retry 3
52 delay_before_retry 3
53 connect_port 80
54 }
55 }
56}
57virtual_server 192.168.122.110 80 {
58 delay_loop 3
59 lb_algo rr
60 lb_kind DR
61 persistence_timeout 50
62 protocol TCP
63 real_server 192.168.122.10 80 {
64 weight 1
65 HTTP_GET {
66 url {
67 path /index.html
68 status_code 200
69 }
70 connect_timeout 10
71 nb_get_retry 3
72 delay_before_retry 3
73 connect_port 80
74 }
75 }
76 real_server 192.168.122.20 80 {
77 weight 1
78 HTTP_GET {
79 url {
80 path /index.html
81 status_code 200
82 }
83 connect_timeout 10
84 nb_get_retry 3
85 delay_before_retry 3
86 connect_port 80
87 }
88 }
89}
MASTER HostB主机的配置如下：
1[root@lvs-dr2 ~]# cat /etc/keepalived/keepalived.conf
2vrrp_instance bl_one {
3 state BACKUP
4 interface eth0
5 lvs_sync_daemon_interface eth0
6 virtual_router_id 38
7 priority 120
8 advert_int 3
9 authentication {
10 auth_type PASS
11 auth_pass 1111
12 }
13 virtual_ipaddress {
14 192.168.122.100
15 }
16}
17vrrp_instance bl_two {
18 state MASTER
19 interface eth0
20 lvs_sync_daemon_interface eth0
21 virtual_router_id 48
22 priority 150
23 advert_int 3
24 authentication {
25 auth_type PASS
26 auth_pass 1111
27 }
28 virtual_ipaddress {
29 192.168.122.110
30 }
31}
32virtual_server 192.168.122.100 80 {
33 delay_loop 3
34 lb_algo rr
35 lb_kind DR
36 persistence_timeout 50
37 protocol TCP
38 real_server 192.168.122.10 80 {
39 weight 1
40 TCP_CHECK {
41 connect_timeout 10
42 nb_get_retry 3
43 delay_before_retry 3
44 connect_port 80
45 }
46 }
47 real_server 192.168.122.20 80 {
48 weight 1
49 TCP_CHECK {
50 connect_timeout 10
51 nb_get_retry 3
52 delay_before_retry 3
53 connect_port 80
54 }
55 }
56}
57virtual_server 192.168.122.110 80 {
58 delay_loop 3
59 lb_algo rr
60 lb_kind DR
61 persistence_timeout 50
62 protocol TCP
63 real_server 192.168.122.10 80 {
64 weight 1
65 HTTP_GET {
66 url {
67 path /index.html
68 status_code 200
69 }
70 connect_timeout 10
71 nb_get_retry 3
72 delay_before_retry 3
73 connect_port 80
74 }
75 }
76 real_server 192.168.122.20 80 {
77 weight 1
78 HTTP_GET {
79 url {
80 path /index.html
81 status_code 200
82 }
83 connect_timeout 10
84 nb_get_retry 3
85 delay_before_retry 3
86 connect_port 80
87 }
88 }
89}
以上创建了两个实例lb_one、lb_two ，其中A主机做为VIP1地址的master，B主机做为VIP2地址的master，A、B主机互为backup 。另外还有一个技巧点，在master-backup架构中也可以配置多个VIP地址，只需要在实例的VIP配置中增加地址、并在后面指定对应的虚地址后面的realserver即可。在主备脚架中配置多个VIP的方法如下：
1virtual_ipaddress {
2 192.168.122.100
3 192.168.122.110
4 ………………
5}
三、realserver主机配置
两台realserver主机的配置启停脚本如下：
1[root@lvs01 ~]# cat dr_client.sh
2#!/bin/bash
3VIP1=192.168.122.100
4VIP2=192.168.122.110
5#vip's broadcast
6BROADCAST=192.168.122.255
7. /etc/rc.d/init.d/functions
8case "$1" in
9 start)
10 echo "reparing for Real Server"
11 echo "1" >/proc/sys/net/ipv4/conf/lo/arp_ignore
12 echo "2" >/proc/sys/net/ipv4/conf/lo/arp_announce
13 echo "1" >/proc/sys/net/ipv4/conf/all/arp_ignore
14 echo "2" >/proc/sys/net/ipv4/conf/all/arp_announce
15 ifconfig lo:0 $VIP1 netmask 255.255.255.255 broadcast $BROADCAST up
16 ifconfig lo:1 $VIP2 netmask 255.255.255.255 broadcast $BROADCAST up
17 /sbin/route add -host $VIP1 dev lo:0
18 /sbin/route add -host $VIP2 dev lo:0
19 ;;
20 stop)
21 ifconfig lo:0 down
22 ifconfig lo:1 down
23 echo "0" >/proc/sys/net/ipv4/conf/lo/arp_ignore
24 echo "0" >/proc/sys/net/ipv4/conf/lo/arp_announce
25 echo "0" >/proc/sys/net/ipv4/conf/all/arp_ignore
26 echo "0" >/proc/sys/net/ipv4/conf/all/arp_announce
27 ;;
28 *)
29 echo "Usage: lvs {start|stop}"
30 exit 1
31esac
四、验证
启动服务后，查看两台director和realserver主机的地址信息如下：
master hostA主机的IP信息如下：
1[root@lvs-dr ~]# ip add show
21: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
3 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
4 inet 127.0.0.1/8 scope host lo
5 inet6 ::1/128 scope host
6 valid_lft forever preferred_lft forever
72: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
8 link/ether 52:54:00:39:42:be brd ff:ff:ff:ff:ff:ff
9 inet 192.168.122.30/24 brd 192.168.122.255 scope global eth0
10 inet 192.168.122.100/32 scope global eth0
11 inet6 fe80::5054:ff:fe39:42be/64 scope link
12 valid_lft forever preferred_lft forever
master hostB主机的IP信息如下：
1[root@lvs-dr2 ~]# ip add show
21: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
3 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
4 inet 127.0.0.1/8 scope host lo
5 inet6 ::1/128 scope host
6 valid_lft forever preferred_lft forever
72: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
8 link/ether 52:54:00:7b:f2:1a brd ff:ff:ff:ff:ff:ff
9 inet 192.168.122.40/24 brd 192.168.122.255 scope global eth0
10 inet 192.168.122.110/32 scope global eth0
11 inet6 fe80::5054:ff:fe7b:f21a/64 scope link
12 valid_lft forever preferred_lft forever
realserver主机的IP信息如下：
1[root@lvs01 ~]# ip add show
21: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
3 link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
4 inet 127.0.0.1/8 scope host lo
5 inet 192.168.122.100/32 brd 192.168.122.255 scope global lo:0
6 inet 192.168.122.110/32 brd 192.168.122.255 scope global lo:1
7 inet6 ::1/128 scope host
8 valid_lft forever preferred_lft forever
92: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
10 link/ether 52:54:00:a9:3b:6a brd ff:ff:ff:ff:ff:ff
11 inet 192.168.122.10/24 brd 192.168.122.255 scope global eth0
12 inet6 fe80::5054:ff:fea9:3b6a/64 scope link
13 valid_lft forever preferred_lft forever
以上可以看到，VIP在master节点上可以看到，另一个VIP同样只在该实例的master上可以看到，在两台realserver上的回环地址上有两个VIP地址。
分别到各VIP的主节点上通过ipvadm查看，可以获取到对的连接信息，如下：
1[root@lvs-dr ~]# ipvsadm -Ln
2IP Virtual Server version 1.2.1 (size=4096)
3Prot LocalAddress:Port Scheduler Flags
4 -> RemoteAddress:Port Forward Weight ActiveConn InActConn
5TCP 192.168.122.100:80 rr persistent 1
6 -> 192.168.122.10:80 Route 1 0 12
7 -> 192.168.122.20:80 Route 1 0 11
8TCP 192.168.122.110:80 rr persistent 1
9 -> 192.168.122.10:80 Route 1 0 0
10 -> 192.168.122.20:80 Route 1 0 0
更详细的测试步骤略过 。
捐赠本站(Donate)
如您感觉文章有用，可扫码捐赠本站！(If the article useful, you can scan the QR code to donate))
- Author: shisekong
- Link: https://blog.361way.com/lvs-keepalived-two-master/6625.html
- License: This work is under a 知识共享署名-非商业性使用-禁止演绎 4.0 国际许可协议. Kindly fulfill the requirements of the aforementioned License when adapting or creating a derivative of this work.