lvs-nat模式的优点配置简单,缺点是请求和响应都必须经过ds,容易称为性能瓶颈
希望有这样的模式,请求的时候使用input链进行负载均衡,响应的时候就不要经过ds,直接由rs响应给客户端
在nat模式的时候,请求vip,接收vip的响应
构想 请求vip,接受rip响应,这是不允许lvs-dr模式
NAT脚本
bash
# ds脚本
#!/bin/bash
#配置网卡
echo TYPE="Ethernet" >> /etc/sysconfig/network-
scripts/ifcfg-ens36
echo BOOTPROTO="none" >> /etc/sysconfig/network-
scripts/ifcfg-ens36
read -p "router name:" router_name
echo NAME='"$rount_name"' >> /etc/sysconfig/network-
scripts/ifcfg-ens36
uuidkey=$( uuidgen )
echo UUID='"$uuidkey"' >> /etc/sysconfig/network-
scripts/ifcfg-ens36 >> /etc/sysconfig/network-
scripts/ifcfg-ens36
echo DEVICE='"$rount_name"' >> /etc/sysconfig/network-
scripts/ifcfg-ens36
echo ONBOOT="yes" >> /etc/sysconfig/network-scripts/ifcfg-
ens36
echo IPADDR=192.168.10.100 >> /etc/sysconfig/network-
scripts/ifcfg-ens36
systemctl restart network
#安装ipvsadm
yum list installed|grep ipvsadm
if[ $? -ne 0 ];then
yum -y install ipvsadm
fi
#配置规则
read -p "vip:" vip
read -p "port:" port
read -p "rule:" s
ipvsadm -A -t $vip:$port -s $s
# ip forward
echo "net.ipv4.ip_forward=1" >/etc/sysctl.conf
sysctl -p
# rs脚本
#!/bin/bash
read -p "dip:" dip
# 设置网关
route del default
route add defualt gw $dip
DR模式
1.性能更优,回路不再经过ds
2.ds和rs为了保证用户响应,都要求配置统一的vip
3.由于rs是直接响应client,网关一定不能设置为ds的dip
4.对rs的vip进行抑制,让ds的vip接收请求,rs的vip不接受请求
5.rs的vip绑定点lo回路网卡上
1.在ds的ens33上挂一个vip 10.1.1.102
cpp
[root@dr03 ~]# ifconfig
ens33: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 10.1.1.101 netmask 255.255.255.0 broadcast 10.1.1.255
[root@dr03 ~]# ifconfig ens33:0 10.1.1.102 broadcast
10.1.1.102 netmask 255.255.255.255 up
[root@dr03 ~]# route add -host 10.1.1.102 dev ens33:0
# 10.1.1.101 dip
# 10.1.1.102 vip 在rs上的vip和这个vip相同
2.设置规范
bash
# 安装ipvsadm
yum -y install ipvsadm
$ 设置规则
ipvsadm -A -t 10.1.1.102:80 -s rr
ipvsadm -a -t 10.1.1.102:80 -r 10.1.1.200 -g
ipvsadm -a -t 10.1.1.102:80 -r 10.1.1.201 -g
# rs不在需要指定端口,dr不支持端口映射,vip上是80端口,最终就是80端
口
# -m nat -g gateway
设置RS主机
1.在lo接口上绑定vip
cs
[root@web01 ~]# ifconfig lo:0 10.1.1.102 broadcast
10.1.1.102 netmask 255.255.255.255 up
2.设置主机路由
cs
[root@web01 ~]# route add -host 10.1.1.102 dev lo:0
3.抑制rs接收请求
bash
echo 1 > /proc/sys/net/ipv4/conf/lo/arp_ignore
echo 2 > /proc/sys/net/ipv4/conf/lo/arp_announce
echo 1 > /proc/sys/net/ipv4/conf/all/arp_ignore
echo 2 > /proc/sys/net/ipv4/conf/all/arp_announce
4.生成脚本,对web02使用
bash
ifconfig lo:0 10.1.1.102 broadcast 10.1.1.102 netmask
255.255.255.255 up
route add -host 10.1.1.102 dev lo:0
echo 1 > /proc/sys/net/ipv4/conf/lo/arp_ignore
echo 2 > /proc/sys/net/ipv4/conf/lo/arp_announce
echo 1 > /proc/sys/net/ipv4/conf/all/arp_ignore
echo 2 > /proc/sys/net/ipv4/conf/all/arp_announce
5.测试使用,查看状态
cpp
[root@dr03 ~]# ipvsadm -Ln --stats
DR模式的脚本
1.ds脚本
bash
#! /bin/bash
#在ens33上挂载一个ip地址
read -p "vip:" vip
read -p "mac:" mac
read -p "num" num
ifconfig $mac: $num $vip broadcast $vip netmask
255.255.255.255
# 主机路由
route add -host $vip dev $mac:$num
#安装ipvsadm
yum list installed|grep ipvsadm
if [ $? -ne 0 ] ; then
yum -y install ipvsadm
fi
#配置规则(不需要设置ip_forword)
ipvsadm -C
read -p "rule:" rule
read -p "port:" port
ipvsadm -A -t $vip:$port -s $rule
read -p "rip1:" rip1
ipvsadm -a -t $vip:$port -r $rip1 -g
read -p "rip2:" rip2
ipvsadm -a -t $vip:$port -r $rip2 -g
2. rs脚本
bash
#!/bin/bash
#在ens33上挂载一个ip地址
read -p "vip:" vip
read -p "mac:" mac
read -p "num" num
ifconfig $mac:$num $vip broadcast $vip netmask
255.255.255.255
# 主机路由
route add -host $vip dev $mac:$num
echo 1 > /proc/sys/net/ipv4/conf/lo/arp_ignore
echo 2 > /proc/sys/net/ipv4/conf/lo/arp_announce
echo 1 > /proc/sys/net/ipv4/conf/all/arp_ignore
echo 2 > /proc/sys/net/ipv4/conf/all/arp_announce