CentOS 7 部署LVS集群（DR模式、NAT模式，LVS+keepalived）
1.简介
linux虚拟服务器，是一个虚拟的服务器集群系统，可以在Unix和linux平台下实现负载均衡集群的功能。
& nbsp; Linux Virtual Servers, a virtual server cluster system that can achieve load-balanced cluster functions under Unix and Linux platforms.
1.2LVS与nginx的对比
1）工作在网络模型的7层，可以针对http应用做一些分流的策略，nginx单凭这点可利用的场合远多于LVS。
& nbsp; 1) works on the 7th floor of the network model, which allows for some sort of diversion strategy for http applications, and nginx can use more than LVS alone.
2）最新版的nginx也支持4层TCP负载，这曾是LVS的优势。
andnbsp; 2) the latest edition of nginx also supports 4 floor TCP loads, which used to be the advantage of LVS.
3）nginx对网络稳定性的依赖小，相反LVS对网络稳定性依赖大。
andnbsp; 3) nginx relies less on network stability than LVS relies heavily on network stability.
4）nginx安装和配置比较简单，测试起来比较方便，基本上能把错误用日志打印出来。LVS的配置、测试就要花比较长的时间，对网络依赖较大。
4) nginx installation and configuration is simpler and easier to test and basically print errors in logs. LVS configuration and testing takes longer and depends more on the network.
1.3搭建LVS的原因
1.3 The reason why LVS was built
1）日1000-2000W PV或并发请求1万以下都可考虑nginx
andnbsp; 1) 1000-2000W PV or as many as 10,000 requests can be considered ginx
2）大型门户网站、电商网站需要用到LVS。
andnbsp; 2) Large portals, power provider websites need to be used on LVS.
1.4LVS集群的工作模式
1）DR模式：通过改写请求报文的目标MAC地址，将请求发送给真实服务器，而真实服务器将相应后的处理结果直接返还给客户端用户。极大地提高集群的伸缩性，但LB与RS必须在同一局域网环境。
& nbsp; 1) DR mode: send the request to the real server by rewriting the target MAC address of the request, and the real server returns the processing results directly to the client user. Greatly improves the scalability of the cluster, but LB and RS must be in the same local area network environment.
2）NAT模式：
andnbsp; 2) NAT mode:
通过网络地址转换，LB重写请求报文的目标地址，根据预设的调度算法，将请求分派给后端的真实服务器，真实服务器的响应报文处理之后，返回时必须通过LB，经过LB时报文的源地址被重写，再返回给客户。
Rewrites the target address of the request via a web address, LB rewrites the request to the backend real server according to a predefined schedule algorithm. After the real server's response is processed, the return must be through LB and the source address of the LB time report is rewritten and returned to the client.
/span>
3）隧道模式：LB把请求的报文通过IP隧道转发至真实服务器，而真实服务器将响应处理后直接返回给客户端用户。
3) tunnel mode: LB forwards the requested message through the IP tunnel to the real server, which returns the response directly to the client user.
4）FULLNAT模式：数据包进入时，除了做DNAT，还做SNAT，从而实现LVS-RealServer间可以跨vlan通讯，RealServer只需要连接到内网。
4) FILLNAT mode: When data packages enter, they do SNAT in addition to DNA T so that the LVS-RealServer can cross the vlan, and RealServer just connects to the intranet.
2.环境准备
strong>2. Environmental readiness
1）关闭防火墙和Selinux
& nbsp; 1) Close firewalls and Selinux
2）主机名及IP地址关系如下：
2) hostname and IP address relationship as follows:
lb01 10.0.0.5
lb042 10.0.0.6
web03 10.0.0.17
web04 10.0.0.18
3）web03和web04安装Tomcat软件，并满足下面条件：
& nbsp; 3) web03 and web04 install Tomcat software and meet the following conditions:
curl http://10.0.0.17 页面底部得到结果为web03
andbsp; curlhttp://10.0.0.17 curl http://10.0.0.18 页面底部得到结果为web04
http://10.0.0.18 3.ipvsadm管理LVS负载集群（DR模式）
[root@lb01 ~]# yum install -y ipvsadm
说明：默认没有加载模块，需要安装管理工具才能激活
3.2查看LVS状态，并激活LVS内核模块
3.2 views the LVS state and activates the LVS kernel module
[root@lb01 ~]# ipvsadm
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
[root@lb01 ~]# lsmod | grep ip_vs
ip_vs_rr 12600 0
ip_vs 140944 2 ip_vs_rr
nf_conntrack 105745 1 ip_vs
libcrc32c 12644 2 xfs,ip_vs
3.3在eth0网卡绑定VIP地址（只在lb01操作）
3.3 Bind the VIP address on the eth0 net card (Ib01 only)
[root@lb01 ~]# ip addr add 10.0.0.13/24 dev eth0
[root@lb01 ~]# ip a s eth0
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:a6:ba:9e brd ff:ff:ff:ff:ff:ff
inet 10.0.0.5/24 brd 10.0.0.255 scope global eth0
valid_lft forever preferred_lft forever
inet 10.0.0.13/24 scope global secondary eth0
valid_lft forever preferred_lft forever
inet6 fe80::20c:29ff:fea6:ba9e/64 scope link
valid_lft forever preferred_lft forever
3.4清除所有LVS规则（只在lb01操作）
3.4 Clears all LVS rules (Ib01 only)
[root@lb01 ~]# ipvsadm -C
3.5设置tcp、tcpfin、udp连接超时时间
3.5 sets tcp, tcpfin, udp connection timeout
[root@lb01 ~]# ipvsadm --set 30 5 60
3.6添加虚拟服务（只在lb01操作）
[root@lb01 ~]# ipvsadm -A -t 10.0.0.13:80 -s wrr -p 20
3.7将虚拟服务关联到真实服务上（只在lb01操作）
[root@lb01 ~]# ipvsadm -a -t 10.0.0.13:80 -r 10.0.0.17:80 -g -w 1
[root@lb01 ~]# ipvsadm -a -t 10.0.0.13:80 -r 10.0.0.18:80 -g -w 1
[root@lb01 ~]# ipvsadm -ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 10.0.0.13:80 wrr persistent 20
-> 10.0.0.17:80 Route 1 0 0
-> 10.0.0.18:80 Route 1 0 0
3.8在lo网卡绑定VIP地址（web03、web04同时操作）
ip addr add 10.0.0.13/24 dev lo
3.9修改内核参数抑制ARP响应（web03、web04同时操作）
3.9 Modification of kernel parameters to inhibit ARP response (web03, web04 simultaneous operation)
cat /etc/sysctl.conf
net.ipv4.conf.all.arp_ignore=1
net.ipv4.conf.all.arp_announce=2
net.ipv4.conf.lo.arp_ignore=1
net.ipv4.conf.lo.arp_announce=2
3.10浏览器访问http://10.0.0.13
http://10.0.0.13;
4.LVS+keepalived(DR模式)
yum install -y keepalived
4.2修改lb01的keepalived配置文件
4.2 Modifys the lb01's kepalived profile
[root@lb01 ~]# cat /etc/keepalived/keepalived.conf
global_defs {
router_id LVS_01
}
vrrp_instance VI_1 {
state MASTER
interface eth0
virtual_router_id 51
priority 150
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
10.0.0.13/24
}
}
virtual_server 10.0.0.13 80 {
delay_loop 6
lb_algo wrr
lb_kind DR
nat_mask 255.255.255.0
persistence_timeout 50
protocol TCP
real_server 10.0.0.17 80 {
weight 1
TCP_CHECK {
connect_timeout 8
nb_get_retry 3
delay_before_retry 3
connect_port 80
}
}
real_server 10.0.0.18 80 {
weight 1
TCP_CHECK {
connect_timeout 8
nb_get_retry 3
delay_before_retry 3
connect_port 80
}
}
}
4.3修改lb02的keepalived配置文件
4.3 Modifys the lb02's kepalived profile
[root@lb02 ~]# cat /etc/keepalived/keepalived.conf
global_defs {
router_id LVS_02
}
vrrp_instance VI_1 {
state BACKUP
interface eth0
virtual_router_id 51
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
10.0.0.13/24
}
}
virtual_server 10.0.0.13 80 {
delay_loop 6
lb_algo wrr
lb_kind DR
nat_mask 255.255.255.0
persistence_timeout 50
protocol TCP
real_server 10.0.0.17 80 {
weight 1
TCP_CHECK {
connect_timeout 8
nb_get_retry 3
delay_before_retry 3
connect_port 80
}
}
real_server 10.0.0.18 80 {
weight 1
TCP_CHECK {
connect_timeout 8
nb_get_retry 3
delay_before_retry 3
connect_port 80
}
}
}
4.4启动keepalived
4.4 Starts keepived
systemctl start keepalived.service
4.5测试keepalived高可用，故障转移
4.5 test for high serviceability of keepived, failure transfer
4.6在lo网卡绑定VIP地址（web03、web04同时操作）
ip addr add 10.0.0.13/24 dev lo
4.7修改内核参数抑制ARP响应（web03、web04同时操作）
cat /etc/sysctl.conf
net.ipv4.conf.all.arp_ignore=1
net.ipv4.conf.all.arp_announce=2
net.ipv4.conf.lo.arp_ignore=1
net.ipv4.conf.lo.arp_announce=2
4.8浏览器访问http://10.0.0.13（测试keepalived健康检查）
4.8 browser access http://10.0.0.13 (test keepalived health check)
5.ipvsadm管理LVS负载集群（NAT模式）
[root@lb01 ~]# yum install -y ipvsadm
说明：默认没有加载模块，需要安装管理工具才能激活
5.2查看LVS状态，并激活LVS内核模块
[root@lb01 ~]# ipvsadm
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
[root@lb01 ~]# lsmod | grep ip_vs
ip_vs_rr 12600 0
ip_vs 140944 2 ip_vs_rr
nf_conntrack 105745 1 ip_vs
libcrc32c 12644 2 xfs,ip_vs
5.3在eth0网卡绑定VIP地址（只在lb01操作）
[root@lb01 ~]# ip addr add 10.0.0.13/24 dev eth0
[root@lb01 ~]# ip a s eth0
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:a6:ba:9e brd ff:ff:ff:ff:ff:ff
inet 10.0.0.5/24 brd 10.0.0.255 scope global eth0
valid_lft forever preferred_lft forever
inet 10.0.0.13/24 scope global secondary eth0
valid_lft forever preferred_lft forever
inet6 fe80::20c:29ff:fea6:ba9e/64 scope link
valid_lft forever preferred_lft forever
5.4清除所有LVS规则（只在lb01操作）
[root@lb01 ~]# ipvsadm -C
5.5设置tcp、tcpfin、udp连接超时时间（只在lb01操作）
5.5 sets tcp, tcpfin, udp connection timeout (lb01 only)
[root@lb01 ~]# ipvsadm --set 30 5 60
5.6添加虚拟服务（只在lb01操作）
[root@lb01 ~]# ipvsadm -A -t 10.0.0.13:80 -s rr -p 20
5.7将虚拟服务关联到真实服务上（只在lb01操作）
ipvsadm -a -t 10.0.0.13:80 -r 172.16.1.203:80 -m
ipvsadm -a -t 10.0.0.13:80 -r 172.16.1.204:80 -m
5.9开启路由转发功能（只在lb01操作）
5.9 Start route forwarding (lb01 only)
[root@lb01 ~]# echo 'net.ipv4.ip_forward=1' >> /etc/sysctl.conf
[root@lb01 ~]# sysctl -p
5.10清空iptables规则（只在lb01操作）
5.10 Clears the iptables (Ib01 only)
[root@lb01 ~]# iptables -F
[root@lb01 ~]# iptables -F -t nat
5.11添加nat转换规则（只在lb01操作）
[root@lb01 ~]# iptables -t nat -A POSTROUTING -s 172.16.1.0/24 -j MASQUERADE
5.12RS服务器内网网卡添加虚拟网关（web03、web04同时操作）
echo "GATEWAY=172.16.1.13" >> /etc/sysconfig/network-scripts/ifcfg-eth1
5.13浏览器访问http://10.0.0.15
6.LVS+keepalived（NAT模式）
yum install -y keepalived
6.2修改lb01的keepalived配置文件
[root@lb01 ~]# cat /etc/keepalived/keepalived.conf
global_defs {
router_id LVS_DEVEL
}
vrrp_instance VI_1 {
state MASTER
interface eth0
virtual_router_id 51
priority 150
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
10.0.0.13/24 dev eth0 label eth0:1
}
}
vrrp_instance VI_2 {
state MASTER
interface eth1
virtual_router_id 52
priority 150
advert_int 1
authentication {
auth_type PASS
auth_pass 1112
}
virtual_ipaddress {
172.16.1.13/24 dev eth1 label eth1:1
}
}
virtual_server 10.0.0.13 80 {
delay_loop 6
lb_algo rr
lb_kind NAT
nat_mask 255.255.255.0
persistence_timeout 50
protocol TCP
real_server 172.16.1.17 80 {
weight 1
TCP_CHECK {
connect_timeout 8
nb_get_retry 3
delay_before_retry 3
connect_port 80
}
}
real_server 172.16.1.18 80 {
weight 1
TCP_CHECK {
connect_timeout 8
nb_get_retry 3
delay_before_retry 3
connect_port 80
}
}
}
6.3修改lb02的keepalived配置文件
6.3 Modify the lb02's kepalived profile
[root@lb02 ~]# cat /etc/keepalived/keepalived.conf
global_defs {
router_id LVS_DEVEL1
}
vrrp_instance VI_1 {
state BACKUP
interface eth0
virtual_router_id 51
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
10.0.0.13/24 dev eth0 label eth0:1
}
}
vrrp_instance VI_2 {
state BACKUP
interface eth1
virtual_router_id 52
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 1112
}
virtual_ipaddress {
172.16.1.13/24 dev eth1 label eth1:1
}
}
virtual_server 10.0.0.13 80 {
delay_loop 6
lb_algo rr
lb_kind NAT
nat_mask 255.255.255.0
persistence_timeout 50
protocol TCP
real_server 172.16.1.17 80 {
weight 1
TCP_CHECK {
connect_timeout 8
nb_get_retry 3
delay_before_retry 3
connect_port 80
}
}
real_server 172.16.1.18 80 {
weight 1
TCP_CHECK {
connect_timeout 8
nb_get_retry 3
delay_before_retry 3
connect_port 80
}
}
}
6.4开启路由转发功能（lb01、lb02同时操作）
6.4 Turns on the transmission function (lb01, lb02 simultaneous operation)
echo 'net.ipv4.ip_forward=1' >> /etc/sysctl.conf
sysctl -p
6.5清空iptables规则（lb01、lb02同时操作）
6.5 Clears the iptables rules (lb01, lb02 simultaneous operation)
iptables -F
iptables -F -t nat
6.6添加nat转换规则（lb01、lb02同时操作）
iptables -t nat -A POSTROUTING -s 172.16.1.0/24 -j MASQUERADE
6.7防止keepalived脑裂（lb01、lb02同时操作）
iptables -A INPUT -i eth0 -p vrrp -j ACCEPT
6.8启动keepalived
systemctl start keepalived.service
6.8RS服务器内网网卡添加虚拟网关（web03、web04同时操作）
echo "GATEWAY=172.16.1.13" >> /etc/sysconfig/network-scripts/ifcfg-eth1
6.9浏览器访问http://10.0.0.13
6.9 browser visit http://10.0.0.13
注册有任何问题请添加 微信：MVIP619 拉你进入群
打开微信扫一扫
添加客服
进入交流群
1.本站遵循行业规范，任何转载的稿件都会明确标注作者和来源；2.本站的原创文章，请转载时务必注明文章作者和来源，不尊重原创的行为我们将追究责任；3.作者投稿可能会经我们编辑修改或补充。
