lvs dr keepalive
Lvs linux virtual server linux虚拟服务器，是个网络第四层的负载均衡群集的调度器，是个软件实现。lvs实现在linux内核中，其模块名为ipvs，用户管理命令为ipvsadm(rpm软件包ipvsadm提供），ipvs工作在netfiler的input链上，有经过input链的且符合定义群集服务的流量，强行改变其方向，根据调度算法调度到响应real server上去。
Lvs类型有四种：
1.nat 2.dr 3.tun 4.fullnat
Lvs-Nat主要特点：1.vip为公网地址，dip、rip为私网地址，且dip、rip必须在同一个物理网络里，且每个real server节点的网关必须指向dip. 2. Cip的请求报文及rip的响应报文都经过director,所以director很容易成为网络瓶颈。3.nat 其实就是多目标的dnat,所以支持端口映射。4.director主机的操作系统一般为linux,而real server 主机操作系统，可以为提供群集服务的任何操作系统。
Lvs-Dr的主要特点：1. Vip为公网的地址，dip、rip可以是公网地址，也可以是私网地址。2.cip的请求报文经过director调度，但rip的响应报文不经过director,real server的网关一定不能指向dir,所以此种调度方式，效率极高。
3.由于调度时通过修改mac地址进行，不修改第三，四层信息，所以不支持端口映射。4.dip与rip必须在同一个物理网络里。5.director的操作系统一般为linux,而real server 的操作系统，必须支持lo网卡别名，管理arp响应级别和通告级别的机制。
Tun的主要特点：1. Vip dip rip都是公网地址2.不修改报文的ip首部，而是在client 的请求报文之外再封装一个ip首部 ，源ip是dip,目标ip是rip 3.rs的网关一定不能指向dip 4.不支持端口映射5.rs必须支持隧道功能。
Lvs-Fullnat 的主要特点：1.同时修改请求报文的源ip,目标ip cip改为dip, vip改为rip,因此director与rs可以用路由器分割跨网，其他特点以lvs-nat一样。
Lvs scheduler十种调度方法：
根据其调度时是否考虑个各RS当前的负载状态，可分为静态和动态调度方法：
根据算法本身静态调度有四种：
1. rr---简单的轮循调度，起点公平
2. Wrr-----加权重的轮循调度，根据权重比例来轮循调度
3. Sh----原地址hash 根据原地址生成一个原地址hash表进行调度，在记录的有效的时间内，相同原地址始终发往同一个real server ,可以在第四层进行会话绑定。
4. Dh ----目标地址hash，一般用在正向代理，根据相同目标地址始终发往同一个real server.
根据RS的实时的负载进行动态调度有六种：
1.lc------最少连接 谁连接最少，最先被选中。计算方法：Overhead=activeconns*256 +inactivconns
2.Wlc--加权重的最少连接数,计算方法：Overhead=(activeconns*256+inactivconns)/weighted
3.Sed--最短的期望延迟，让性能最好的real server 来响应。计算方法：Overhead=(activeconns+1)*256/weighted
4.Nq:nerver queue 不能出现有real server 有队列，而有的real server 无请求。
5.LbLC：动态的dh算法。Locality-based least connections 基于本地的最少连接
6.LBLCR：LBLC with Replication
十种算法的默认算法是wlc
四种lvs类型的默认类型是dr
Lvs dr keepalive 实现一个高可用的lvs 的基于web服务的群集实验
实验环境：
四台主机: 两台主机centos7.2A centos7.2B 做基于keepAlived 高可用的Lvs dr
另两台主机：centos6.7A centos6.7B 各做一个网站
在网段172.16.0.0/16网络已有dhcp server 172.16.0.1/16（172.16.0.0网段的）
Ntp server 172.16.0.1
Centos7.2A 一块网卡 桥接模式 ip 172.16.254.79/16
Centos7.2B 一块网卡 桥接模式 ip 172.16.250.205/16
Cenots6.7A 一块网卡 桥接模式 ip 172.16.251.237/16
Centos 6.7B 一块网卡 桥接模式 ip 172.16.251.178/16
为了做实验的方便，关闭掉所有主机的防火墙及selinux：
Centos7.2A centos7.2B systemctl stop firewalld setenforce 0
Centoos6.7A centos6.7B service iptables stop setenforce 0
Vim /etc/keepalived/keepalived.conf
global_defs {
notification_email {
root@localhost
}
notification_email_from keepalived@localhost
smtp_server 127.0.0.1
smtp_connect_timeout 30
router_id router1
vrrp_mcast_group4 224.10.10.10
}
vrrp_instance VI_1 {
state MASTER
interface eno16777728
virtual_router_id 22
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 11112222
}
virtual_ipaddress {
172.16.100.100/32 dev eno16777728 brd 172.16.100.100 label eno16777728:0
}
Notify_master “/scripts/vrrpstatuschange.sh master”
Notify_backup “/scripts/vrrpstatuschange.sh backup”
Notify_fault “/scripts/vrrpstatuschange.sh fault”
}
virtual_server 172.16.100.100 80 {
delay_loop 3
lb_algo rr
lb_kind DR
nat_mask 255.255.0.0
# persistence_timeout 50
protocol TCP
sorry_server 127.0.0.1 80
real_server 172.16.251.237 80 {
weight 1
HTTP_GET {
url {
path /index.html
# digest ff20ad2481f97b1754ef3e12ecd3a9cc
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
real_server 172.16.251.178 80 {
weight 1
HTTP_GET {
url {
path /index.html
# digest ff20ad2481f97b1754ef3e12ecd3a9cc
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
Sorry server 网页
Vim /var/www/html/index.html
<h1> sorry server centos6.7A </h1>
Vim /scripts/vrrpstatuschange.sh
#!/bin/bash
recv='root@localhost'
notify() {
mailsubject="$(hostname) to be $1,vip floating"
mailbody="$(date +'%F %T'):vrrp virtual address 172.16.100.100 trans , $(hostname) change to be $1"
echo "$mailbody" | mail -s "$mailsubject" $recv
}
case $1 in
master)
notify master
;;
backup)
notify backup
;;
fault)
notify fault
;;
*)
echo "error Usage: $(basename $0) master|backup|fault"
esac
Systemctl start keepalived
Ipvsadm -L -n
[root@localhost Desktop]# ipvsadm -L -n
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 172.16.100.100:80 rr
-> 172.16.251.178:80 Route 1 0 0
-> 172.16.251.237:80 Route 1 0 0
把后面两个网站都关闭：
Service httpd stop
Ipvsadm -L -n [root@localhost Desktop]# ipvsadm -L -n
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 172.16.100.100:80 rr
-> 127.0.0.1:80 Route 1 0 0 ----- 调度sorry server
[root@localhost Desktop]# curl http://172.16.100.100
<h1> say sorry 7.2A </h1>
Cenots7.2A yum -y install ipvsadm keepalived httpd
ntpdate 172.16.0.1
Sorry server
Vim /var/www/html/index.html
<h1> sorry server centos7.2B</h1>
Vim /etc/keepalive/keepalived.conf
! Configuration File for keepalived
global_defs {
notification_email {
root@localhost
}
notification_email_from keepalived@localhost
smtp_server 127.0.0.1
smtp_connect_timeout 30
router_id router2
vrrp_mcast_group4 224.10.10.10
}
vrrp_instance VI_1 {
state BACKUP
interface eno16777728
virtual_router_id 22
priority 90
advert_int 1
authentication {
auth_type PASS
auth_pass 11112222
}
virtual_ipaddress {
172.16.100.100/32 dev eno16777728 brd 172.16.100.100 label eno16777728:0
}
Notify_master “/scripts/vrrpstatuschange.sh master”
Notify_backup “/scripts/vrrpstatuschange.sh backup”
Notify_fault “/scripts/vrrpstatuschange.sh fault”
}
virtual_server 172.16.100.100 80 {
delay_loop 3
lb_algo rr
lb_kind DR
nat_mask 255.255.0.0
# persistence_timeout 50
protocol TCP
sorry_server 127.0.0.1 80
real_server 172.16.251.237 80 {
weight 1
HTTP_GET {
url {
path /index.html
# digest ff20ad2481f97b1754ef3e12ecd3a9cc
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
real_server 172.16.251.178 80 {
weight 1
HTTP_GET {
url {
path /index.html
# digest ff20ad2481f97b1754ef3e12ecd3a9cc
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
Vim /scripts/vrrpstatuschange.sh
#!/bin/bash
recv='root@localhost'
notify() {
mailsubject="$(hostname) to be $1,vip floating"
mailbody="$(date +'%F %T'):vrrp virtual address 172.16.100.100 trans , $(hostname) change to be $1"
echo "$mailbody" | mail -s "$mailsubject" $recv
}
case $1 in
master)
notify master
;;
backup)
notify backup
;;
fault)
notify fault
;;
*)
echo "error Usage: $(basename $0) master|backup|fault"
esac
Cenots6.7A 配置：
[root@www ~]# cat /scripts/lvsdr.sh
#!/bin/bash
vip=172.16.100.100
netmask=255.255.255.255
a=`ifconfig | grep lo:0 | awk '{print $1}'`
case $1 in
enable)
if [ "$a" == lo:0 ];then
echo "lvsdr is already enable"
else
echo 1 > /proc/sys/net/ipv4/conf/lo/arp_ignore
echo 1 > /proc/sys/net/ipv4/conf/all/arp_ignore
echo 2 > /proc/sys/net/ipv4/conf/lo/arp_announce
echo 2 > /proc/sys/net/ipv4/conf/all/arp_announce
ifconfig lo:0 $vip netmask $netmask broadcast $vip up
route add -host $vip dev lo:0
fi
;;
disable)
if [ "$a" == lo:0 ] ;then
echo 0 > /proc/sys/net/ipv4/conf/lo/arp_ignore
echo 0 > /proc/sys/net/ipv4/conf/all/arp_ignore
echo 0 > /proc/sys/net/ipv4/conf/lo/arp_announce
echo 0 > /proc/sys/net/ipv4/conf/all/arp_announce
ifconfig lo:0 down
else
echo "lvsdr is already disable"
fi
;;
*)
echo "error:Usage:$(basename $0) enable|disable"
;;
Esac
/scirpts/lvsdr.sh enable
显示结果：
Ifconfig
lo:0 Link encap:Local Loopback
inet addr:172.16.100.100 Mask:255.255.255.255
UP LOOPBACK RUNNING MTU:65536 Metric:1
Route -n
[root@www ~]# route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
172.16.100.100 0.0.0.0 255.255.255.255 UH 0 0 0 lo
172.16.0.0 0.0.0.0 255.255.0.0 U 1 0 0 eth0
0.0.0.0 172.16.0.1 0.0.0.0 UG 0 0 0 eth0
网站首页：
Vim /var/www/html/index.html
<h1> centos6.7A</h1>
Centos6.7B 配置： [root@www ~]# cat /scripts/lvsdr.sh
#!/bin/bash
vip=172.16.100.100
netmask=255.255.255.255
a=`ifconfig | grep lo:0 | awk '{print $1}'`
case $1 in
enable)
if [ "$a" == lo:0 ];then
echo "lvsdr is already enable"
else
echo 1 > /proc/sys/net/ipv4/conf/lo/arp_ignore
echo 1 > /proc/sys/net/ipv4/conf/all/arp_ignore
echo 2 > /proc/sys/net/ipv4/conf/lo/arp_announce
echo 2 > /proc/sys/net/ipv4/conf/all/arp_announce
ifconfig lo:0 $vip netmask $netmask broadcast $vip up
route add -host $vip dev lo:0
fi
;;
disable)
if [ "$a" == lo:0 ] ;then
echo 0 > /proc/sys/net/ipv4/conf/lo/arp_ignore
echo 0 > /proc/sys/net/ipv4/conf/all/arp_ignore
echo 0 > /proc/sys/net/ipv4/conf/lo/arp_announce
echo 0 > /proc/sys/net/ipv4/conf/all/arp_announce
ifconfig lo:0 down
else
echo "lvsdr is already disable"
fi
;;
*)
echo "error:Usage:$(basename $0) enable|disable"
;;
Esac
/scirpts/lvsdr.sh enable
显示结果：
Ifconfig
lo:0 Link encap:Local Loopback
inet addr:172.16.100.100 Mask:255.255.255.255
UP LOOPBACK RUNNING MTU:65536 Metric:1
Route -n
[root@www ~]# route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
172.16.100.100 0.0.0.0 255.255.255.255 UH 0 0 0 lo
172.16.0.0 0.0.0.0 255.255.0.0 U 1 0 0 eth0
0.0.0.0 172.16.0.1 0.0.0.0 UG 0 0 0 eth0
网站首页：
Vim /var/www/html/index.html
<h1> centos6.7B</h1>
访问四个网站：
ocalhost Desktop]# curl http://172.16.254.79
<h1> say sorry 7.2A </h1>
[root@localhost Desktop]# curl http://172.16.250.205
<h1> say sorry 7.2B </h1>
[root@localhost Desktop]# curl http://172.16.251.237
<h1> centos6.7A</h1>
[root@localhost Desktop]# curl http://172.16.251.178
<h1> centos6.7B</h1>
当centos7.2A 作为lvs master时：
[root@www Desktop]# curl http://172.16.100.100
<h1> centos6.7A</h1>
[root@www Desktop]# curl http://172.16.100.100
<h1> centos6.7B</h1>
[root@www Desktop]# curl http://172.16.100.100
<h1> centos6.7A</h1>
[root@www Desktop]# curl http://172.16.100.100
<h1> centos6.7B</h1>
[root@www Desktop]# curl http://172.16.100.100
<h1> centos6.7A</h1>
两个站点轮循访问
关闭centos6.7A 站点：
Service httpd stop
[root@www Desktop]# curl http://172.16.100.100
<h1> centos6.7B</h1>
[root@www Desktop]# curl http://172.16.100.100
<h1> centos6.7B</h1>
[root@www Desktop]# curl http://172.16.100.100
<h1> centos6.7B</h1>
[root@www Desktop]# curl http://172.16.100.100
<h1> centos6.7B</h1>
只能访问centos6.7B 站点了
把centos7.2A 的keepalvied 关闭：
Systemctl keepalived stop[root@localhost Desktop]# ipvsadm -L -n
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 172.16.100.100:80 rr
-> 172.16.251.178:80 Route 1 0 0
Centos 7.2B keepalived 开始变成master[root@www Desktop]# curl http://172.16.100.100
<h1> centos6.7B</h1>
[root@www Desktop]# curl http://172.16.100.100
<h1> centos6.7B</h1>
[root@www Desktop]# curl http://172.16.100.100
<h1> centos6.7B</h1>
[root@www Desktop]# curl http://172.16.100.100
<h1> centos6.7B</h1>
Centos 6.7B 站点照样可以访问：
Keepalived服务达到了lvs dr 高可用。
在关闭centos6.7B 的站点
Service httpd stop
Centos7.2B
[root@localhost Desktop]# ipvsadm -L -n
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 172.16.100.100:80 rr
-> 127.0.0.1:80 Route 1 0 0
[root@www Desktop]# curl http://172.16.100.100
<h1> say sorry 7.2B </h1>----------------sorry server 访问正常
