欢迎来到尧图网

客户服务 关于我们

您的位置：首页 > 文旅 > 手游 > Keepalived高可用和Haproxy负载均衡

Keepalived高可用和Haproxy负载均衡

2024/10/23 4:24:12 来源：https://blog.csdn.net/qq_59671942/article/details/142868230  浏览:  次   关键词：Keepalived高可用和Haproxy负载均衡

一、概念

简介

keepalived是集群管理中保证集群高可用的一个服务软件，用来防止单点故障。

工作原理

keepalived是以VRRP协议为实现基础的，N台路由器组成一个路由器组。master上有一个对外提供服务的vip，master会向backup进行发送组播，如果backup未收到vrrp包，就认为master宕掉了，VRRP会根据优先级选取一个backup作为master与之同时会将原master上的vip漂移到新master上

扩展：

脑裂：

如果master和backup之间因为原因无法接收到组播的通知，但是实际两个节点都在正常工作，这两个节点均为master，并且会强行绑定VIP。

解决方式：

1.添加检测手段，给两个主机的网卡做健康检查，ping对方减少脑裂的发生机会

2.设置仲裁机制，依赖第三番检测，启用共享磁盘锁，ping网关等等。

3.将master停掉，检查防火墙等机制，网络之间的通信等等

配置文件

! Configuration File for keepalivedglobal_defs {notification_email {acassen@firewall.locfailover@firewall.locsysadmin@firewall.loc}notification_email_from Alexandre.Cassen@firewall.locsmtp_server 192.168.200.1smtp_connect_timeout 30router_id LVS_DEVELvrrp_skip_check_adv_addrvrrp_strictvrrp_garp_interval 0vrrp_gna_interval 0
}vrrp_instance VI_1 {state MASTERinterface eth0virtual_router_id 51priority 100advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.200.16192.168.200.17192.168.200.18}
}virtual_server 192.168.200.100 443 {delay_loop 6lb_algo rrlb_kind NATpersistence_timeout 50protocol TCPreal_server 192.168.201.100 443 {weight 1SSL_GET {url {path /digest ff20ad2481f97b1754ef3e12ecd3a9cc}url {path /mrtg/digest 9b3a0c85a887a256d6939da88aabd8cd}connect_timeout 3nb_get_retry 3delay_before_retry 3}}
}virtual_server 10.10.10.2 1358 {delay_loop 6lb_algo rrlb_kind NATpersistence_timeout 50protocol TCPsorry_server 192.168.200.200 1358real_server 192.168.200.2 1358 {weight 1HTTP_GET {url {path /testurl/test.jspdigest 640205b7b0fc66c1ea91c463fac6334d}url {path /testurl2/test.jspdigest 640205b7b0fc66c1ea91c463fac6334d}url {path /testurl3/test.jspdigest 640205b7b0fc66c1ea91c463fac6334d}connect_timeout 3nb_get_retry 3delay_before_retry 3}}real_server 192.168.200.3 1358 {weight 1HTTP_GET {url {path /testurl/test.jspdigest 640205b7b0fc66c1ea91c463fac6334c}url {path /testurl2/test.jspdigest 640205b7b0fc66c1ea91c463fac6334c}connect_timeout 3nb_get_retry 3delay_before_retry 3}}
}virtual_server 10.10.10.3 1358 {delay_loop 3lb_algo rrlb_kind NATpersistence_timeout 50protocol TCPreal_server 192.168.200.4 1358 {weight 1HTTP_GET {url {path /testurl/test.jspdigest 640205b7b0fc66c1ea91c463fac6334d}url {path /testurl2/test.jspdigest 640205b7b0fc66c1ea91c463fac6334d}url {path /testurl3/test.jspdigest 640205b7b0fc66c1ea91c463fac6334d}connect_timeout 3nb_get_retry 3delay_before_retry 3}}real_server 192.168.200.5 1358 {weight 1HTTP_GET {url {path /testurl/test.jspdigest 640205b7b0fc66c1ea91c463fac6334d}url {path /testurl2/test.jspdigest 640205b7b0fc66c1ea91c463fac6334d}url {path /testurl3/test.jspdigest 640205b7b0fc66c1ea91c463fac6334d}connect_timeout 3nb_get_retry 3delay_before_retry 3}}
}

tips：

二、keepalived实战操作

安装keepalived

yum -y install keepalived

VIP漂移

机器准备

ip

角色

192.168.252.146

VIP：192.168.252.204

keepalived-master

192.168.252.148

keepalived-backup

编辑配置文件

192.168.252.146：
1.编辑主配置文件
vim /etc/keepalived/keepalived.conf
内容：
! Configuration File for keepalivedglobal_defs {router_id master   
}vrrp_instance VI_1 {state MASTER        interface ens33     virtual_router_id 80  priority 100advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.252.204/24   }
}2.启动：
systemctl start keepalived192.168.252.148：
1.编辑从配置文件
vim /etc/keepalived/keepalived.conf
内容：
! Configuration File for keepalivedglobal_defs {router_id backup   
}vrrp_instance VI_1 {state BACKUP        interface ens33     virtual_router_id 80  priority 50advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.252.204/24   }
}2.启动
systemctl start keepalived

tips:

  1. 这里主节点的优先级更高为100，从节点的优先级为50，这样子VIP就可以漂移向从节点，并且在重启主节点的时候VIP会自动恢复到主节点
  2. virtual_router_id 80 这个选项是虚拟路由两个节点的必须一致才能相互通信
  3. 主从节点的state标识符不能一样route_id也不能一样都是标识这个机子的信息。

测试

没有停掉keepalived之前

192.168.252.146：

有虚拟ip 192.168.252.204

192.168.252.148：

没有虚拟ip

停掉keepalived之后
192.168.252.146:
systemctl stop keepalived

192.168.252.146:

没有了VIP

192.168.252.148：

有了VIP

keepalived+nginx

健康检测nginx

机器准备

ip

角色

192.168.252.146

VIP：192.168.252.204

keepalived-master

nginx

192.168.252.148

keepalived-backup

nginx

编写检查nginx脚本

vim /etc/keepalived/scripts/check_nginx_status.sh
内容：
#!/bin/bash
curl -I http://localhost &>/dev/null
if [ $? -ne 0 ];thensystemctl stop keepalived
fi

编辑配置文件

192.168.252.146：
1.编辑主配置文件
vim /etc/keepalived/keepalived.conf
内容：
! Configuration File for keepalivedglobal_defs {router_id master   
}vrrp_script check_nginx {script "/etc/keepalived/scripts/check_nginx_status.sh"interval 5     
}vrrp_instance VI_1 {state MASTER        interface ens33     virtual_router_id 80  priority 100advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.252.204/24   }track_script {check_nginx}
}2.启动：
systemctl start nginx
systemctl start keepalived192.168.252.148：
1.编辑从配置文件
vim /etc/keepalived/keepalived.conf
内容：
! Configuration File for keepalivedglobal_defs {router_id backup   
}vrrp_instance VI_1 {state BACKUP        interface ens33     virtual_router_id 80  priority 50advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.252.204/24   }
}2.启动
systemctl start nginx
systemctl start keepalived

tips：

  1. 从节点脚本可加可不加
  2. 出现vip不漂移原因：脚本写错、脚本没有执行权限、配置文件错误
  3. nginx要再keepalived之前启动

测试

没有停止nginx之前

停止nginx之后
192.168.252.146：
systemctl stop nginx

keepalived+ipvsadm+nginx(DR)

机器准备

ip

角色

192.168.252.144

VIP[ens33]：192.168.252.200/32

keepalived

ipvsadm

192.168.252.145

VIP[ens33]：192.168.252.200/32

keepalived

ipvsadm

192.168.252.146

VIP[lo]：192.168.252.200/32

nginx

192.168.252.148

VIP[lo]：192.168.252.200/32

nginx

编写检测ipvsadm脚本

vim /etc/keepalived/scripts/ipvsadm.sh
内容：
#!/bin/bash
systemctl status ipvsadm
if [ $? -ne 0 ];thensystemctl stop keepalived
fi

编写配置文件

192.168.252.144：
1.编辑配置文件
vim /etc/keepalived/keepalived.conf
内容：
! Configuration File for keepalivedglobal_defs {router_id lvs-keepalived-master
}vrrp_script ipvsadm_check {script "/etc/keepalived/scripts/ipvsadm.sh"interval 5     #五秒执行一次
}vrrp_instance VI_1 {state MASTERinterface ens33                #VIP绑定接口virtual_router_id 80         	 #VRID 同一组集群，主备一致priority 100                   #本节点优先级，辅助改为50advert_int 1                   #检查间隔，默认为1sauthentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.252.200/32         # 可以写多个vip}track_script {ipvsadm_check}
}virtual_server 192.168.252.200 80 {    #LVS配置delay_loop 3   #故障切换的时间lb_algo rr     #LVS调度算法lb_kind DR     #LVS集群模式（路由模式）net_mask 255.255.255.0protocol TCP      #健康检查使用的协议real_server 192.168.252.146 80 {weight 1inhibit_on_failure   #当该节点失败时，把权重设置为0，而不是从IPVS中删除TCP_CHECK {          #健康检查connect_port 80   #检查的端口connect_timeout 3  #连接超时的时间}}real_server 192.168.252.148 80 {weight 1inhibit_on_failureTCP_CHECK {connect_timeout 3connect_port 80}}
}2.启动
systemctl start ipvsadm
systemctl start keepalived3.查看指向
ipvsadm -ln
回显：
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags-> RemoteAddress:Port           Forward Weight ActiveConn InActConn
TCP  192.168.252.200:80 rr-> 192.168.252.146:80           Route   1      0          0         -> 192.168.252.148:80           Route   1      0          0192.168.252.145：
1.编辑配置文件
vim /etc/keepalived/keepalived.conf
内容：
! Configuration File for keepalivedglobal_defs {router_id lvs-keepalived-backup
}vrrp_instance VI_1 {state BACKUPinterface ens33                #VIP绑定接口nopreemptvirtual_router_id 80         #VRID 同一组集群，主备一致priority 50            #本节点优先级，辅助改为50advert_int 1            #检查间隔，默认为1sauthentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.252.200/32  # 可以写多个vip}
}virtual_server 192.168.252.200 80 {    #LVS配置delay_loop 3   #故障切换的时间lb_algo rr     #LVS调度算法lb_kind DR     #LVS集群模式（路由模式）net_mask 255.255.255.0protocol TCP      #健康检查使用的协议real_server 192.168.252.146 80 {weight 1inhibit_on_failure   #当该节点失败时，把权重设置为0，而不是从IPVS中删除TCP_CHECK {          #健康检查connect_port 80   #检查的端口connect_timeout 3  #连接超时的时间}}real_server 192.168.252.148 80 {weight 1inhibit_on_failureTCP_CHECK {connect_timeout 3connect_port 80}}
}2.启动
systemctl start ipvsadm
systemctl start keepalived3.查看指向
ipvsadm -ln
回显：
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags-> RemoteAddress:Port           Forward Weight ActiveConn InActConn
TCP  192.168.252.200:80 rr-> 192.168.252.146:80           Route   1      0          0         -> 192.168.252.148:80           Route   1      0          0

192.168.252.144

192.168.252.145

服务端添加回环网卡

192.168.252.146:
ip a add dev lo 192.168.252.200/32
192.168.252.148:
ip a add dev lo 192.168.252.200/32

测试

访问虚拟ip

浏览器中输入http://192.168.252.200

刷新之后页面变换

keepalived+mysql(master1)+mysql(master2)

机器准备

ip

角色

192.168.252.146

keepalived-master

mysql-master1

192.168.252.148

keepalived-backup

mysql-master2

做双主复制

此处直接显示结果不做具体详细步骤

192.168.252.146：

192.168.252.148：

编写检测mysql脚本

192.168.252.146：
vim /etc/keepalived/scripts/check_mysqld_status.sh
内容：
#!/bin/bash
mysql -uroot -p"@Syh2025659" -e "show slave status\G" &>/dev/null
if [ $? -ne 0 ];thensystemctl stop keepalived
fi192.168.252.146：
vim /etc/keepalived/scripts/check_mysql_status.sh
内容：
#!/bin/bash
mysql -uroot -p"@Syh2025659" -e "show slave status\G" &>/dev/null
if [ $? -ne 0 ];thensystemctl stop keepalived
fi

编写配置文件

192.168.252.146：
1.编辑配置文件
vim /etc/keepalived/keepalived.conf
内容：
! Configuration File for keepalivedglobal_defs {router_id master
}
vrrp_script check_run {script "/etc/keepalived/scripts/check_mysqld_status.sh"interval 5
}vrrp_instance VI_1 {state MASTERinterface ens33virtual_router_id 80priority 100advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.252.200/24}track_script {check_run}
}2.启动
systemctl start mysqld
systemctl start keepalived192.168.252.148：
1.编辑配置文件
vim /etc/keepalived/keepalived.conf
! Configuration File for keepalivedglobal_defs {router_id backup
}
vrrp_script check_run {script "/etc/keepalived/scripts/check_mysql_status.sh"interval 5
}vrrp_instance VI_1 {state BACKUPinterface ens33virtual_router_id 80priority 50advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.252.200/24}track_script {check_run}
}2.启动
systemctl start mysqld
systemctl start keepalived

测试

mysql启动时

192.168.252.146：

192.168.252.148：

mysql停掉时

192.168.252.146：

192.168.252.148：

三、haproxy

概述

简介

软件，主要是做负载均衡的7层，也可以做4层负载均衡，是一款高性能的负载均衡软件。因为其专注于负载均衡这一些事情，因此与nginx比起来在负载均衡这件事情上做更好，更专业。

特点

  1. 支持tcp / http 两种协议层的负载均衡（支持四层和七层负载均衡）
  2. 支持8种左右的负载均衡算法（有多种负载均衡算法）
  3. 性能好，事件驱动的链接处理模式及单进程处理模式（高性能）
  4. 拥有监控页面
  5. 功能强大的ACL支持（相当于nginx中的location）

常用算法

roundrobin

基于权重进行轮询,每个服务器处理时间相同

static-rr

基于权重进行轮询，根据权重大小分配服务器处理时间

leastconn

新的连接请求被派发至具有最少连接数目的后端服务器

实战操作

安装

yum -y install haproxy

keepalived+haproxy+nginx七层负载均衡

机器准备

ip

角色

192.168.252.144

VIP：192.168.252.205/24

keepalived

haproxy

192.168.252.145

VIP：192.168.252.205/24

keepalived

haproxy

192.168.252.146

nginx

192.168.252.148

nginx

编写检测haproxy脚本
192.168.252.144：
vim  /etc/keepalived/scripts/check_haproxy_status.sh
内容：
#!/bin/bash
curl -I http://localhost &>/dev/null
if [ $? -ne 0 ];thensystemctl stop keepalived
fi192.168.252.145：
vim  /etc/keepalived/scripts/check_haproxy_status.sh
内容：
#!/bin/bash
curl -I http://localhost &>/dev/null
if [ $? -ne 0 ];thensystemctl stop keepalived
fi
编辑haproxy配置文件
两个节点群配置
192.168.252.144：
192.168.252.145：
编辑配置文件
vim /etc/haproxy/haproxy.cfg
内容：
globallog         127.0.0.1 local2 infopidfile     /var/run/haproxy.pidmaxconn     4000   #优先级低user        haproxygroup       haproxydaemon               #以后台形式运行ha-proxynbproc 1                #工作进程数量  cpu内核是几就写几
defaultsmode                    http  #工作模式 http ,tcp 是 4 层,http是 7 层       log                     globalretries                 3   #健康检查。3次连接失败就认为服务器不可用，主要通过后面的check检查option                  redispatch  #服务不可用后重定向到其他健康服务器。maxconn                 4000  #优先级中contimeout              5000  #ha服务器与后端服务器连接超时时间，单位毫秒msclitimeout              50000 #客户端超时srvtimeout              50000 #后端服务器超时
listen statsbind                        *:80stats                       enablestats uri                   /haproxy  #使用浏览器访问 http://192.168.246.169/haproxy,可以看到服务器状态>（uri不可改变）  stats auth                  syh:123  #用户认证，客户端使用elinks浏览器的时候不生效
frontend  webmode                        httpbind                            *:80   #监听哪个ip和什么端口option                  httplog             #日志类别 http 日志格式acl html url_reg  -i  \.html$  #1.访问控制列表名称html。规则要求访问以html结尾的urluse_backend httpservers if  html #2.如果满足acl html规则，则推送给后端服务器httpserversdefault_backend    httpservers   #默认使用的服务器组
backend httpservers    #名字要与上面的名字必须一样balance     roundrobin  #负载均衡的方式server  http1 192.168.252.146:80 maxconn 2000 weight 1  check inter 1s rise 2 fall 2server  http2 192.168.252.148:80 maxconn 2000 weight 1  check inter 1s rise 2 fall 2
编辑keepalived配置文件
192.168.252.144：
1.编辑配置文件
vim /etc/keepalived/keepalived.conf
内容：
! Configuration File for keepalivedglobal_defs {router_id director1
}
vrrp_script check_haproxy {script "/etc/keepalived/scripts/check_haproxy_status.sh"interval 5
}vrrp_instance VI_1 {state MASTERinterface ens33virtual_router_id 80priority 100advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.252.205/24}track_script {check_haproxy}
}2.启动
systemctl start haproxy
systemctl start keepalived192.168.252.145：
1.编辑配置文件
vim /etc/keepalived/keepalived.conf
内容：
! Configuration File for keepalivedglobal_defs {router_id director2
}
vrrp_script check_haproxy {script "/etc/keepalived/scripts/check_haproxy_status.sh"interval 5
}vrrp_instance VI_1 {state BACKUPinterface ens33virtual_router_id 80priority 50advert_int 1authentication {auth_type PASSauth_pass 1111}virtual_ipaddress {192.168.252.205/24}track_script {check_haproxy}
}2.启动
systemctl start haproxy
systemctl start keepalived
测试

haproxy+mysql四层负载均衡

机器准备

ip

角色

192.168.252.144

haproxy

192.168.252.146

mysql

192.168.252.148

mysql

tips：

做haproxy的机器上不能有mysql

加入不同数据库

192.168.252.146：

192.168.252.148：

tips：

要想远程访问需要将用户打开所有ip都能访问

编辑haproxy配置文件
192.168.252.144
vim /etc/haproxy/haproxy.cfg
内容：
globallog         127.0.0.1 local2pidfile     /var/run/haproxy.pidmaxconn     4000user        haproxygroup       haproxydaemonnbproc      1
defaultsmode                    httplog                     globaloption                  redispatchretries                 3maxconn                 4000contimeout              5000clitimeout              50000srvtimeout                  50000
listen statsbind                        *:80stats                       enablestats uri                   /haproxystats auth                  qianfeng:123
frontend  webmode                        httpbind                            *:80option                  httplogdefault_backend    httpservers
backend httpserversbalance     roundrobinserver  http1 192.168.252.146:80 maxconn 2000 weight 1  check inter 1s rise 2 fall 2server  http2 192.168.252.148:80 maxconn 2000 weight 1  check inter 1s rise 2 fall 2listen mysqlbind *:3306mode tcp        #使用TCP四层负载均衡balance roundrobinserver mysql1 192.168.252.146:3306 weight 1  check inter 1s rise 2 fall 2server mysql2 192.168.252.148:3306 weight 1  check inter 1s rise 2 fall 2
测试

再次登录一个

开启日志

1.打开注释
vim /etc/rsyslog.conf 
# Provides UDP syslog reception  #由于haproxy的日志是用udp传输的,所以要启用rsyslog的udp监听
$ModLoad imudp
$UDPServerRun 514
#### RULES #### 
local2.*                       /var/log/haproxy.log
2.重启应用
systemctl restart rsyslog
systemctl restart haproxy

版权声明:

本网仅为发布的内容提供存储空间，不对发表、转载的内容提供任何形式的保证。凡本网注明“来源：XXX网络”的作品，均转载自其它媒体，著作权归作者所有，商业转载请联系作者获得授权，非商业转载请注明出处。

我们尊重并感谢每一位作者，均已注明文章来源和作者。如因作品内容、版权或其它问题，请及时与我们联系，联系邮箱：809451989@qq.com，投稿邮箱：809451989@qq.com

相关资讯

热文排行

最新新闻

推荐新闻

热搜词

JVM、字节码文件介绍 鸿蒙前端-1. 层叠效果 MySQL 存储引擎概述 windows启动qtcreator，任务管理器有，但是界面和任务栏不显示解决方案 【四】企业级JavaScript开发开发者控制台 AIGC：开启智能创造的璀璨新篇章