溫馨提示×
您好，登錄后才能下訂單哦！
點擊 登錄注冊 即表示同意《億速云用戶服務條款》
您好，登錄后才能下訂單哦！
HA Cluster的目的，為了防止重要的服務器在提供服務時，出現不可抗力的因素，例如硬件故障
自然災害，斷電，軟件bug，操作系統bug等，導致提供服務的主機出現宕機，死機，從而影響正
常業務，因此采用高可用的方案，實現持續性穩定的提供主機服務的方案稱為HA
A=可用
MTBF=平均無故障時間
MTTR=平均修復時長
換算公式
A=MTBF/(MTBF+MTTR)
注意提升A值的方法：
1.分子足夠大，但是不劃算，成本太高
2.降低分母，即降低平均修復時長，成本較低效果出色
如何降低平均修復時長？
建立備用服務器，實現Failover功能。
原理：在主服務器發生故障時，快速轉移IP地址（floating ip），以及快速轉移應用程序至備用服務器
需要相關軟件轉移IP（轉移IP即實現ip地址從新配置），轉移服務（即在備用主機上啟動相關服務應用程序）
總體來說HA 即為在主服務器宕機時，實現備用服務器的快速切換
關鍵點：IP地址轉移，數據共享
HA Cluster （ip，nginx）ip與nginx進程通常稱為HA資源
備用服務器使用“心跳”檢測，向主服務器發起udp報文（UDP報文不需要三次握手等),
根據主服務器的響應來判斷主服務器是否正常工作
關鍵點：響應時間，資源爭用共享存儲
假如鏈接主從服務器的網線連接中斷，則會導致，ip爭奪，而且最主要的是存儲的混亂（同一文件
一邊在增加，另一邊再刪除）
會導致源數據的損壞，損害很大，解決方法，使用爆頭設備（STONITH），在補刀，將未徹底斷電的還在運行設備斷電
當處理完主服務器的故障后，還要將服務器上線，即為Failback
Failover<------->Failback
HA Cluster實現方案
1、vrrp協議的實現
keepalived
2、ais（available Interface standard）：可用接口標準，完備HA集群
RHCS（cmam）
heartbeat
corosync
Keepalived：
vrrp協議：Virtual Redundant Protocol
術語:
虛擬路由：virtual router
虛擬路由器標識：VRID（0-255）
物理路由：
master：主設備
back：備用設備
priority：優先級
VIP：virtual Ip
VMAC:Virtual MAC (00-00-5e-00-01-VRID)
GraciousARP(免費arp)
通告：心跳，優先級等；周期性；
搶占式，非搶占式；
安全工作：
認證：
無認證
簡單字符認證
MD5
工作模式
主/備：單虛擬路由器；
主/主：主/備（虛擬路由器1），備/主（虛擬路徑器2）
特點：
vrrp協議的軟件實現，原生設計的目的為了高可用ipvs服務：
vrrp協議完成地址流動；
為vip地址所在的節點生成ipvs規則（在配置文件中預先定義）；
為ipvs集群的各RS做健康狀態檢測；
基于腳本調用接口通過執行腳本完成腳本中定義的功能，進而影響集群事務；
組件：
核心組件：
vrrp stack
ipvs wrapper
checkers
控制組件：配置文件分析器
IO復用器
內存管理組件
HA Cluster的配置前提：
（1）各節點時間必須同步
（2）確保iptables及selinux不會成為阻礙；
（3）各節點之間可通過主機名互相通信(對KA并非必須)；
建議使用/etc/hosts文件實現；
（4）各節點之間的root用戶可以基于密鑰認證的ssh服務完成互相通信（并非必須）
keepalived安裝配置：
CentOS 6.4+隨base倉庫提供；
1、同步時間
配置chronyd服務器172.18。200.100
yum安裝chrony，并啟動服務
[root@localhost ~]# service chronyd start
Starting chronyd: [ OK ]
使用ntpdate命令，同步172.18.10.10以及172.18.10.11的時間
[root@localhost ~]# ntpdate 172.18.200.100
2、清空iptables和selinux
iptables -F
setenforce 0
3、配置hosts文件（非必須）
4、安裝keepalived
[root@localhost ~]# yum install keepalived
[root@localhost ~]# cd /etc/keepalived/
[root@localhost keepalived]# ls
keepalived.conf
[root@localhost keepalived]# cp keepalived.conf keepalived.conf.bak
[root@localhost keepalived]# ls
keepalived.conf keepalived.conf.bak
[root@localhost keepalived]# vim keepalived.conf
主配置文件：/etc/keepalived/keepalived.conf
配置文件組成部分及相關選項解釋
TOP HIERACHY
GLOBAL CONFIGURATION
Global definitions
Static routes/addresses
VRRPD CONFIGURATION
VRRP synchronization group(s)：vrrp同步組；
VRRP instance(s)：每個vrrp instance即一個vrrp路由器；
LVS CONFIGURATION
Virtual server group(s)
Virtual server(s)：ipvs集群的vs和rs；
global_defs { ###全局定義
notification_email {
acassen@firewall.loc
failover@firewall.loc ####定義出現問題后發送郵箱的地址
sysadmin@firewall.loc
}
notification_email_from Alexandre.Cassen@firewall.loc ##從哪里發過來
smtp_server 192.168.200.1 ###郵件服務器地址
smtp_connect_timeout 30#####超時時間
router_id LVS_DEVEL###路由器IP
vrrp_mcast_group4 224.0.100.5###ipv4多播地址
}
vrrp_instance VI_1 { ##vrrp配置段
state MASTER###表示是主還是從這里顯示主，另一個則為從
interface eth0###表明工作從哪個網卡發出 “多波心跳信息”
virtual_router_id 51###虛擬路由ID
priority 100###主的優先級
advert_int 1 ##通告時間間隔
authentication###認證
auth_type PASS####認證類型：簡單密鑰認證
auth_pass 1111#####認證密碼：最多不能超過8位
}
virtual_ipaddress {##虛擬IP地址配在哪個網卡上
192.168.200.16/24 dev eth0 ##定義配置在哪個網卡的別名上
192.168.200.17
192.168.200.18
}
}
track_interface { ##配置要監控的網絡接口，一旦接口出現故障，則轉為FAULT狀態；即接口跟蹤
eth0
eth2
...
}
nopreempt：定義工作模式為非搶占模式；
preempt_delay 300：搶占式模式下，節點上線后觸發新選舉操作的延遲時長；
5、修改配置文件
[root@localhost keepalived]# vim 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 node1
vrrp_mcast_group4 224.0.100.50
}
vrrp_instance myroute {
state MASTER
interface eth2
virtual_router_id 50
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 123456
}
virtual_ipaddress {
172.18.50.50/16 dev eth2
}
}
6、將配置文件發送到另一臺機器10上
[root@localhost keepalived]# scp keepalived.conf 172.18.10.10:/etc/keepalived/
修改配置文件
[root@localhost keepalived]# vim 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 node2
vrrp_mcast_group4 224.0.100.50
}
vrrp_instance myroute {
state BACKUP
interface eth2
virtual_router_id 50
priority 98
advert_int 1
authentication {
auth_type PASS
auth_pass 123456
}
virtual_ipaddress {
172.18.50.50/16 dev eth2
}
}
7、啟動服務
啟動備用服務器11
[root@localhost ~]# service keepalived start
查看地址
[root@localhost ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:07:27:ff brd ff:ff:ff:ff:ff:ff
inet 172.18.10.10/16 brd 172.18.255.255 scope global eth2
inet 172.18.50.50/16 scope global secondary eth2
inet6 fe80::20c:29ff:fe07:27ff/64 scope link
valid_lft forever preferred_lft forever
發現地址已經添加，這是若開啟主服務器，由于沒有設置搶斷延遲，則會立刻搶斷
8、啟動主服務器
[root@localhost keepalived]# service keepalived start
Starting keepalived: [ OK ]
[root@localhost keepalived]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:99:76:84 brd ff:ff:ff:ff:ff:ff
inet 172.18.10.11/16 brd 172.18.255.255 scope global eth2
inet 172.18.50.50/16 scope global secondary eth2
inet6 fe80::20c:29ff:fe99:7684/64 scope link
valid_lft forever preferred_lft forever
發現地址已經添加
而從服務器11上
[root@localhost ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:07:27:ff brd ff:ff:ff:ff:ff:ff
inet 172.18.10.10/16 brd 172.18.255.255 scope global eth2
inet6 fe80::20c:29ff:fe07:27ff/64 scope link
valid_lft forever preferred_lft forever
IP地址已經刪除
9、使用tcpdump抓包工具查看主從服務器的相應心跳測試
[root@localhost keepalived]# tcpdump -i eth2 host 224.0.100.50 ###在主服務器端抓包
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on eth2, link-type EN10MB (Ethernet), capture size 65535 bytes
16:39:33.357307 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:34.358905 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:35.360605 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:36.362301 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:37.363904 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:38.365658 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:39.367266 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:40.368921 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:41.370599 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
[root@localhost ~]# tcpdump -i eth2 -nn host 224.0.100.50 ###在從服務器端抓包
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on eth2, link-type EN10MB (Ethernet), capture size 65535 bytes
16:39:40.367044 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:41.368741 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:42.370289 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:43.371983 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:44.373750 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:45.375413 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:46.377092 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
16:39:47.378760 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
分析說明。實現簡單的vrrp
即從服務器每隔一秒向主服務器發送1個通報報文。探測主服務器是否存活，實現具體實施軟件keepalived
###############################################################################################################################
雙主模型
1、我們在172.18.10.11上配置了主服務器配置，雙主服務可在配置文件下面繼續添如下內容，配置如下
[root@localhost keepalived]# vim keepalived.conf
vrrp_instance myroute2 {
state BACKUP
interface eth2
virtual_router_id 51
priority 98
advert_int 1
authentication {
auth_type PASS
auth_pass 123457
}
virtual_ipaddress {
172.18.51.51/16 dev eth2
}
}
2、將內容服務至粘貼至172.18.10.10服務器的keepalived.conf配置文件中,然后需要在state和priority上進行相應修改
vrrp_instance myroute2 {
state MASTER
interface eth2
virtual_router_id 51
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 123457
}
virtual_ipaddress {
172.18.51.51/16 dev eth2
}
}
保存并退出，實現雙主模型的設置
3、從啟服務并測試
service keepalived restart
Stopping keepalived: [ OK ]
Starting keepalived: [ OK ]
使用tcpdump抓包，結果如下
172.18.10.11端
[root@localhost keepalived]# tcpdump -i eth2 -nn host 224.0.100.50
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on eth2, link-type EN10MB (Ethernet), capture size 65535 bytes
00:50:20.150330 IP 172.18.10.10 > 224.0.100.50: VRRPv2, Advertisement, vrid 51, prio 100, authtype simple, intvl 1s, length 20
00:50:20.521639 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
00:50:21.151175 IP 172.18.10.10 > 224.0.100.50: VRRPv2, Advertisement, vrid 51, prio 100, authtype simple, intvl 1s, length 20
00:50:21.522539 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
00:50:22.152517 IP 172.18.10.10 > 224.0.100.50: VRRPv2, Advertisement, vrid 51, prio 100, authtype simple, intvl 1s, length 20
00:50:22.523232 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
00:50:23.154334 IP 172.18.10.10 > 224.0.100.50: VRRPv2, Advertisement, vrid 51, prio 100, authtype simple, intvl 1s, length 20
00:50:23.524046 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
172.18.10.10端
[root@localhost keepalived]# tcpdump -i eth2 host 224.0.100.50
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on eth2, link-type EN10MB (Ethernet), capture size 65535 bytes
00:54:01.436075 IP 172.18.10.10 > 224.0.100.50: VRRPv2, Advertisement, vrid 51, prio 100, authtype simple, intvl 1s, length 20
00:54:01.437266 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
00:54:02.437295 IP 172.18.10.10 > 224.0.100.50: VRRPv2, Advertisement, vrid 51, prio 100, authtype simple, intvl 1s, length 20
00:54:02.438831 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
00:54:03.438695 IP 172.18.10.10 > 224.0.100.50: VRRPv2, Advertisement, vrid 51, prio 100, authtype simple, intvl 1s, length 20
00:54:03.439205 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
分析每次都會收到兩次信息，一次發送，一次接收
使用iptable設置規則，拒絕172.18.10.11向224.0.100.50發送通知報文
[root@localhost keepalived]# iptables -A OUTPUT -s 172.18.10.11 -d 224.0.100.50 -j REJECT
在172.18.10.10端使用tcpdump抓包
[root@localhost keepalived]# tcpdump -i eth2 -nn host 224.0.100.50
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on eth2, link-type EN10MB (Ethernet), capture size 65535 bytes
00:50:20.150330 IP 172.18.10.10 > 224.0.100.50: VRRPv2, Advertisement, vrid 51, prio 100, authtype simple, intvl 1s, length 20
00:50:20.521639 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
00:50:21.151175 IP 172.18.10.10 > 224.0.100.50: VRRPv2, Advertisement, vrid 51, prio 100, authtype simple, intvl 1s, length 20
00:50:21.522539 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
00:50:22.152517 IP 172.18.10.10 > 224.0.100.50: VRRPv2, Advertisement, vrid 51, prio 100, authtype simple, intvl 1s, length 20
00:50:22.523232 IP 172.18.10.11 > 224.0.100.50: VRRPv2, Advertisement, vrid 50, prio 100, authtype simple, intvl 1s, length 20
分析發現發送兩條通告，因為172.18.10.11不通告，便認為172.18.10.11掛掉了，因此搶斷，讓自己變為主機。即別人不通告則認為對方掛掉了
可以使用ip a l 查看相應的ip地址獲取：
[root@localhost keepalived]# ip a l
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:07:27:ff brd ff:ff:ff:ff:ff:ff
inet 172.18.10.10/16 brd 172.18.255.255 scope global eth2
inet 172.18.51.51/16 scope global secondary eth2
inet 172.18.50.50/16 scope global secondary eth2
inet6 fe80::20c:29ff:fe07:27ff/64 scope link
valid_lft forever preferred_lft forever
再次在172.18.10.11服務器上，清空iptables規則
[root@localhost keepalived]# iptables -F
再回到172.18.10.10服務器上使用ip a l 查詢
[root@localhost keepalived]# ip a l
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:07:27:ff brd ff:ff:ff:ff:ff:ff
inet 172.18.10.10/16 brd 172.18.255.255 scope global eth2
inet 172.18.51.51/16 scope global secondary eth2
inet6 fe80::20c:29ff:fe07:27ff/64 scope link
valid_lft forever preferred_lft forever
發現地址已經立馬被奪回，是因為工作在搶占模式下。沒有設置preempt_delay 300搶占延遲時間，
結論：實現雙主模型實驗
##################################################################################################################
如何實現自定義通知腳本
一、在172.18.10.11服務器上添加腳本，實現自動發郵件
1.編寫郵件腳本
vim notify.sh
#!/bin/bash
#
contact='root@localhost'
notify() {
mailsubject="vrrp： $(hostname) to be $1"
mailbody="$(hostname) to be $1,vrrp transition, $(date)."
echo "$mailbody" | mail -s "$mailsubject" $contact
}
case $1 in
master)
notify master ;;
backup)
notify backup ;;
fault)
notify fault ;;
*)
echo "Usage: $(basename $0 ) master|backup|fault"
exit 1
;;
esac
2、測試腳本
語法檢測
[root@localhost keepalived]# bash -n notify.sh
運行腳本測試
[root@localhost keepalived]# bash -x notify.sh master
+ contact=root@localhost
+ case $1 in
+ notify master
++ hostname
+ mailsubject='localhost.localdomain to be master'
++ hostname
++ date
+ mailbody='localhost.localdomain to be master,vrrp transition, Mon May 15 01:36:33 CST 2017.'
+ echo 'localhost.localdomain to be master,vrrp transition, Mon May 15 01:36:33 CST 2017.'
+ mail -s mailsubject root@localhost
[root@localhost keepalived]# vim notify.sh
You have mail in /var/spool/mail/root
3、查看收到的郵件
[root@localhost keepalived]# mail
Heirloom Mail version 12.4 7/29/08. Type ? for help.
"/var/spool/mail/root": 1 message 1 new
>N 1 root Mon May 15 01:36 18/696 "mailsubject"
&
Message 1:
From root@localhost.localdomain Mon May 15 01:36:34 2017
Return-Path: <root@localhost.localdomain>
X-Original-To: root@localhost
Delivered-To: root@localhost.localdomain
Date: Mon, 15 May 2017 01:36:33 +0800
To: root@localhost.localdomain
Subject: mailsubject
User-Agent: Heirloom mailx 12.4 7/29/08
Content-Type: text/plain; charset=us-ascii
From: root@localhost.localdomain (root)
Status: R
localhost.localdomain to be master,vrrp transition, Mon May 15 01:36:33 CST 2017.
&
4、將腳本發送至172.18.10.10端
[root@localhost keepalived]# scp notify.sh 172.18.10.10:/etc/keepalived/
root@172.18.10.10's password:
notify.sh 100% 367 0.4KB/s 00:00
5、調用腳本
[root@localhost keepalived]# vim keepalived.conf
在172.18.10.11上的vrrp_instance myrouter1下面添加如下內容,注意是放在vrrp_instance myrouter1上下文中調用
notify_master "/etc/keepalived/notify.sh master"
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
在172.18.10.10上的vrrp_instance myrouter2下面添加如下內容
notify_master "/etc/keepalived/notify.sh master"
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
6，為了實現測試效果，將之前定義的雙主模型刪除，并停止服務（在10.10和10.11上做同樣的操作）
：.,$d 表示從當前行都最后一行全部刪除
[root@localhost keepalived]# service keepalived stop
Stopping keepalived: [ OK ]
7、給之前編寫的腳本加上執行權限
[root@localhost keepalived]# chmod +x notify.sh
[root@localhost keepalived]# ll
total 8
-rw-r--r-- 1 root root 658 May 15 02:01 keepalived.conf
-rwxr-xr-x 1 root root 367 May 15 01:41 notify.sh
8、啟動服務
在172.18.10.11端
[root@localhost keepalived]# service keepalived start
Starting keepalived: [ OK ]
[root@localhost keepalived]# ip a l
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:99:76:84 brd ff:ff:ff:ff:ff:ff
inet 172.18.10.11/16 brd 172.18.255.255 scope global eth2
inet 172.18.50.50/16 scope global secondary eth2
inet6 fe80::20c:29ff:fe99:7684/64 scope link
valid_lft forever preferred_lft forever
[root@localhost keepalived]# mail
Heirloom Mail version 12.4 7/29/08. Type ? for help.
"/var/spool/mail/root": 3 messages 2 unread
1 root Mon May 15 01:36 19/707 "mailsubject"
>U 2 root Mon May 15 11:03 19/735 "vrrp: localhost.localdomain to be master"
&
9、啟動172.18.10.10端的keepalived，并且再次到172.18.10.11端查看郵件
[root@localhost ~]# mail
Heirloom Mail version 12.4 7/29/08. Type ? for help.
"/var/spool/mail/root": 7 messages 5 new 7 unread
U 1 root Mon May 15 11:09 19/735 "vrrp: localhost.localdomain to be backup"
U 2 root Mon May 15 11:11 19/735 "vrrp: localhost.localdomain to be backup"
>N 3 root Mon May 15 11:11 18/725 "vrrp: localhost.localdomain to be master"
N 4 root Mon May 15 11:11 18/725 "vrrp: localhost.localdomain to be backup"
N 5 root Mon May 15 11:26 18/725 "vrrp: localhost.localdomain to be backup"
N 6 root Mon May 15 11:26 18/725 "vrrp: localhost.localdomain to be master"
N 7 root Mon May 15 11:26 18/725 "vrrp: localhost.localdomain to be backup"
&
結論：通知腳本功能實現
######################################################################################################
如何實現 keepalived 高可用LVS （重點）
實驗準備：4臺虛擬主機
其中172.18.10.10和172.18.10.11做為VS端分別為VS2和VS1
172.18.200.100和172.18.249.57做為RS分別為RS1和RS2
首先分別再RS1和RS2端安裝httpd
1、進行如下配置
[root@localhost ~]# cat /var/www/html/index.html
<h2>RS1:172.18.200.100</h2>
[root@localhost ~]# cat /var/www/html/index.html
<h2>RS2:172.18.249.57</h2>
2、編寫VIP配置腳本
vim setparam.sh
#!/bin/bash
#
vip='172.18.50.50'
netmask='255.255.255.255'
iface='lo:0'
case $1 in
start)
echo 1 > /pro/sys/net/ipv4/conf/all/arp_ignore
echo 1 > /pro/sys/net/ipv4/conf/lo/arp_ignore
echo 2 > /pro/sys/net/ipv4/conf/all/arp_ignore
echo 2 > /pro/sys/net/ipv4/conf/lo/arp_ignore
ifconfig $iface $vip netmask $netmask broadcast $vip up
route add -host $vip dev $iface
;;
stop)
ifconfig $iface down
echo 0 > /pro/sys/net/ipv4/conf/all/arp_ignore
echo 0 > /pro/sys/net/ipv4/conf/lo/arp_ignore
echo 0 > /pro/sys/net/ipv4/conf/all/arp_ignore
echo 0 > /pro/sys/net/ipv4/conf/lo/arp_ignore
;;
esac
3、測試腳本
[root@localhost ~]# bash -n setparam.sh
[root@localhost ~]# bash -x setparam.sh start
+ vip=172.18.50.50
+ netmask=255.255.255.255
+ iface=lo:0
+ case $1 in
+ echo 1
setparam.sh: line 9: /pro/sys/net/ipv4/conf/all/arp_ignore: No such file or directory
+ echo 1
setparam.sh: line 10: /pro/sys/net/ipv4/conf/lo/arp_ignore: No such file or directory
+ echo 2
setparam.sh: line 11: /pro/sys/net/ipv4/conf/all/arp_announce: No such file or directory
+ echo 2
setparam.sh: line 12: /pro/sys/net/ipv4/conf/lo/arp_announce: No such file or directory
+ ifconfig lo:0 172.18.50.50 netmask 255.255.255.255 broadcast 172.18.50.50 up
+ route add -host 172.18.50.50 dev lo:0
4、使用scp將腳本分發至RS2
[root@localhost ~]# scp setparam.sh 172.18.249.57:/root
root@172.18.249.57's password:
setparam.sh 100% 610 0.6KB/s 00:00
5、在RS2端執行腳本，并查看是否生成VIP
[root@localhost ~]# bash -x setparam.sh start
+ vip=172.18.50.50
+ netmask=255.255.255.255
+ iface=lo:0
+ case $1 in
+ echo 1
setparam.sh: line 9: /pro/sys/net/ipv4/conf/all/arp_ignore: No such file or directory
+ echo 1
setparam.sh: line 10: /pro/sys/net/ipv4/conf/lo/arp_ignore: No such file or directory
+ echo 2
setparam.sh: line 11: /pro/sys/net/ipv4/conf/all/arp_announce: No such file or directory
+ echo 2
setparam.sh: line 12: /pro/sys/net/ipv4/conf/lo/arp_announce: No such file or directory
+ ifconfig lo:0 172.18.50.50 netmask 255.255.255.255 broadcast 172.18.50.50 up
+ route add -host 172.18.50.50 dev lo:0
[root@localhost ~]# ip a
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
inet 127.0.0.1/8 scope host lo
inet 172.18.50.50/32 brd 172.18.50.50 scope global lo:0
inet6 ::1/128 scope host
valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
link/ether 00:0c:29:b2:ca:ea brd ff:ff:ff:ff:ff:ff
inet 172.18.249.57/16 brd 172.18.255.255 scope global eth0
inet6 fe80::20c:29ff:feb2:caea/64 scope link
valid_lft forever preferred_lft forever
6、啟動RS1和RS2的httpd服務，并查看端口，兩端都要查看，這里只演示一端的
[root@localhost ~]# service httpd start
[root@localhost ~]# ss -tnl
State Recv-Q Send-Q Local Address:Port Peer Address:Port
LISTEN 0 128 :::80 :::*
LISTEN 0 128 :::22 :::*
LISTEN 0 128 *:22 *:*
LISTEN 0 100 ::1:25 :::*
LISTEN 0 100 127.0.0.1:25
7、在兩個前段節點生成ipvs規則
在VS2端
停止keepalived服務
配置sorry server頁面
vim /var/www/html/index.html
Director2 sorry server2
啟動httpd服務
[root@localhost ~]# service httpd start
在VS1端
首先停止keepalived服務
[root@localhost ~]# service keepalived stop
vim /var/www/html/index.html
Director1
啟動httpd服務
[root@localhost ~]# service httpd start
在VS1端編輯keepalived配置文件，添加如下內容：
virtual_server 172.18.50.50 80 {
delay_loop 6
lb_algo wrr
lb_kind DR
persistence_timeout 0
protocol TCP
real_server 172.18.10.11 80 {
weight 1
HTTP_GET {
persistence_timeout 0
protocol TCP
sorry_server 127.0.0.1 80
real_server 172.18.10.11 80 {
weight 1
HTTP_GET {
url {
path /
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
real_server 172.18.10.10 80 {
weight 1
HTTP_GET {
url {
path /
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
在VS2端，同樣編輯keepalived.conf文件，添加如下內容
virtual_server 172.18.50.50 80 {
delay_loop 6
lb_algo wrr
lb_kind DR
persistence_timeout 0
protocol TCP
sorry_server 127.0.0.1 80
real_server 172.18.10.11 80 {
weight 1
HTTP_GET {
persistence_timeout 0
protocol TCP
real_server 172.18.10.11 80 {
weight 1
HTTP_GET {
url {
path /
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
real_server 172.18.10.10 80 {
weight 1
HTTP_GET {
url {
path /
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
在VS2上啟動keepalived服務
[root@localhost ~]# service keepalived start
Starting keepalived: [ OK ]
[root@localhost ~]# ipvsadm -ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 172.18.50.50:80 wrr
-> 172.18.200.100:80 Route 1 0 0
-> 172.18.249.57:80 Route 1 0 0
在客戶端使用curl進行訪問測試（配置完有一定延遲，稍等片刻在訪問）
[root@localhost ~]# curl http://172.18.50.50
<h2>RS2:172.18.249.57</h2>
[root@localhost ~]# curl http://172.18.50.50
<h2>RS1:172.18.200.100</h2>
[root@localhost ~]# curl http://172.18.50.50
<h2>RS2:172.18.249.57</h2>
[root@localhost ~]# curl http://172.18.50.50
<h2>RS1:172.18.200.100</h2>
[root@localhost ~]# curl http://172.18.50.50
<h2>RS2:172.18.249.57</h2>
[root@localhost ~]# curl http://172.18.50.50
<h2>RS1:172.18.200.100</h2>
[root@localhost ~]# curl http://172.18.50.50
<h2>RS2:172.18.249.57</h2>
[root@localhost ~]# curl http://172.18.50.50
<h2>RS1:172.18.200.100</h2>
在172.18.200.100端停止httpd服務
[root@localhost ~]# service httpd stop
Stopping httpd: [ OK ]
在VS2端使用ipvsadm觀察
[root@localhost keepalived]# ipvsadm -ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 172.18.50.50:80 wrr
-> 172.18.249.57:80 Route 1 0 2
在172.18.200.100端停止httpd服務
[root@localhost ~]# service httpd start
在VS2端使用ipvsadm觀察
[root@localhost keepalived]# ipvsadm -ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 172.18.50.50:80 wrr
-> 172.18.200.100:80 Route 1 0 0
-> 172.18.249.57:80 Route 1 0 0
啟動VS1上的keepalived服務，并且關閉VS2，客戶端使用curl測試發現，仍然能夠訪問
[root@localhost keepalived]# curl http://172.18.50.50
<h2>RS2:172.18.249.57</h2>
[root@localhost keepalived]# curl http://172.18.50.50
<h2>RS1:172.18.200.100</h2>
[root@localhost keepalived]# curl http://172.18.50.50
<h2>RS2:172.18.249.57</h2>
[root@localhost keepalived]# curl http://172.18.50.50
<h2>RS1:172.18.200.100</h2>
更改配置文件，將之前刪除的雙主內容添加進去
VS1端
vrrp_instance myroute2 {
state BACKUP
interface eth2
virtual_router_id 51
priority 98
advert_int 1
authentication {
auth_type PASS
auth_pass 123457
}
virtual_ipaddress {
172.18.51.51/16 dev eth2
}
}
VS2端
vrrp_instance myroute2 {
state MASTER
interface eth2
virtual_router_id 51
priority 98
advert_int 1
authentication {
auth_type PASS
auth_pass 123457
}
virtual_ipaddress {
172.18.51.51/16 dev eth2
}
}
重啟keepalived服務,相當重要，，，，，不重啟不會有效果，這就是個坑
總結VS端
VS2端keepalived配置
! 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 node2
vrrp_mcast_group4 224.0.100.50
}
vrrp_instance myroute1 {
state BACKUP
interface eth2
virtual_router_id 50
priority 98
advert_int 1
authentication {
auth_type PASS
auth_pass 123456
}
virtual_ipaddress {
172.18.50.50/16 dev eth2
}
notify_master "/etc/keepalived/notify.sh master"
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
}
vrrp_instance myroute2 {
state MASTER
interface eth2
virtual_router_id 51
priority 98
advert_int 1
authentication {
auth_type PASS
auth_pass 123457
}
virtual_ipaddress {
172.18.51.51/16 dev eth2
}
}
virtual_server 172.18.50.50 80 {
delay_loop 6
lb_algo wrr
lb_kind DR
persistence_timeout 0
protocol TCP
sorry_server 127.0.0.1 80
real_server 172.18.200.100 80 {
weight 1
HTTP_GET {
url {
path /
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
real_server 172.18.249.57 80 {
weight 1
HTTP_GET {
url {
path /
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
virtual_server 172.18.51.51 80 {
delay_loop 6
lb_algo wrr
lb_kind DR
persistence_timeout 0
protocol TCP
sorry_server 127.0.0.1 80
real_server 172.18.200.100 80 {
weight 1
HTTP_GET {
url {
path /
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
real_server 172.18.249.57 80 {
weight 1
HTTP_GET {
url {
path /
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
VS1端keepalived配置
! 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 node1
vrrp_mcast_group4 224.0.100.50
}
vrrp_instance myroute1 {
state MASTER
interface eth2
virtual_router_id 50
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 123456
}
virtual_ipaddress {
172.18.50.50/16 dev eth2
}
notify_master "/etc/keepalived/notify.sh master"
notify_backup "/etc/keepalived/notify.sh backup"
notify_fault "/etc/keepalived/notify.sh fault"
}
vrrp_instance myroute2 {
state BACKUP
interface eth2
virtual_router_id 51
priority 98
advert_int 1
authentication {
auth_type PASS
auth_pass 123457
}
virtual_ipaddress {
172.18.51.51/16 dev eth2
}
}
virtual_server 172.18.50.50 80 {
delay_loop 6
lb_algo wrr
lb_kind DR
persistence_timeout 0
protocol TCP
sorry_server 127.0.0.1 80
real_server 172.18.200.100 80 {
weight 1
HTTP_GET {
url {
path /
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
real_server 172.18.249.57 80 {
weight 1
HTTP_GET {
url {
path /
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
virtual_server 172.18.51.51 80 {
delay_loop 6
lb_algo wrr
lb_kind DR
persistence_timeout 0
protocol TCP
sorry_server 127.0.0.1 80
real_server 172.18.200.100 80 {
weight 1
HTTP_GET {
url {
path /
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
real_server 172.18.249.57 80 {
weight 1
HTTP_GET {
url {
path /
status_code 200
}
connect_timeout 3
nb_get_retry 3
delay_before_retry 3
}
}
}
VIP配置腳本 （由用戶是雙主模型因此VIP有兩個）
#!/bin/bash
#
vip='172.18.50.50'
vip2='172.18.51.51'
netmask='255.255.255.255'
iface='lo:0'
iface2='lo:1'
case $1 in
start)
echo 1 > /pro/sys/net/ipv4/conf/all/arp_ignore
echo 1 > /pro/sys/net/ipv4/conf/lo/arp_ignore
echo 2 > /pro/sys/net/ipv4/conf/all/arp_announce
echo 2 > /pro/sys/net/ipv4/conf/lo/arp_announce
ifconfig $iface $vip netmask $netmask broadcast $vip up
ifconfig $iface2 $vip2 netmask $netmask broadcast $vip2 up
route add -host $vip dev $iface
;;
stop)
ifconfig $iface down
ifconfig $iface2 down
echo 0 > /pro/sys/net/ipv4/conf/all/arp_ignore
echo 0 > /pro/sys/net/ipv4/conf/lo/arp_ignore
echo 0 > /pro/sys/net/ipv4/conf/all/arp_announce
echo 0 > /pro/sys/net/ipv4/conf/lo/arp_announce
;;
esac
實驗結論：實現keepalived 高可用lvs負載均衡
免責聲明：本站發布的內容（圖片、視頻和文字）以原創、轉載和分享為主，文章觀點不代表本網站立場，如果涉及侵權請聯系站長郵箱：is@yisu.com進行舉報，并提供相關證據，一經查實，將立刻刪除涉嫌侵權內容。
