原理简述
本篇主要学习keepalived配合nginx实现nginx的高可用, 也就是需要keepalived检测到nginx宕机时停用keepalived, 备用keepalived会自动接收过来.
简单的原理(如下图), 主备服务器会配置相同的vip(虚拟ip), 谁的优先级高谁来接收vip的请求, 然后nginx和keepalived部署在同一个服务器上面, keeplived控制机器接收到vip的请求, 交给了nginx来处理请求. nginx的功能主要是负责负载均衡, nginx的安装配置在此不再赘述, 可以参考这个: ngix安装与使用
keepalived功能有很多, 此篇只是最简单的配合ngxin实现高可用的demo.
安装
-
安装常用的的指令包:
yum install -y curl gcc openssl-devel libnl3-devel net-snmp-devel -
安装:
yum install -y keepalived -
启动:
systemctl start keepalived -
重启:
systemctl restart keepalived -
关闭:
systemctl stop keepalived -
开机自启:
systemctl enable keepalived -
修改配置文件:
vim /etc/keepalived/keepalived.conf! Configuration File for keepalived # 定义虚拟路由, 必须叫VI_1 vrrp_instance VI_1 { state MASTER #设置为主服务器, 备份服务器设置为BACKUP interface enp0s3 #监控的网络接口(ifconfig或者ip addr指令找出网卡) priority 100 #(优先级, 主机大一点, 备份机小一点) virtual_router_id 99 #同一个vrrp_instance下routerId必须是一致的 authentication { auth_type PASS #vrrp认证方式主备必须一致 auth_pass 12345 #密码 } virtual_ipaddress { 192.168.0.99 #虚拟ip, 主从一致, 可配置多个 } } -
另外一台机相同方法, 相同配置(state改成
BACKUP, priority调整调一下, 此例中是80)
vrrp 的主从并不是通过stat配置的MASTER和BACKUP决定的, 是通过优先级决定的
- 默认日志位置:
/var/log/message位置修改参考: keepalived 配置日志方法
参考2: Keepalived原理介绍和配置实践
参考3: keepalived介绍、安装及配置详解
参考4: https://codor.lanzoue.com/b012qnsvc 密码:1i77
检查是否脑裂
-
使用
tcpdump -i enp0s3 -nn host 224.0.0.18或者tcpdump -i enp0s3 | grep VRRP进行查看, 默认的广播通道为224.0.0.18(我把时间删除了, 内容是我改的)192.168.0.111 > 224.0.0.18: VRRPv2, Advertisement, vrid 99, prio 100, authtype simple, intvl 1s, length 20 192.168.0.117 > 224.0.0.18: VRRPv2, Advertisement, vrid 99, prio 80, authtype simple, intvl 1s, length 20 192.168.0.111 > 224.0.0.18: VRRPv2, Advertisement, vrid 99, prio 100, authtype simple, intvl 1s, length 20 192.168.0.117 > 224.0.0.18: VRRPv2, Advertisement, vrid 99, prio 80, authtype simple, intvl 1s, length 20 192.168.0.111 > 224.0.0.18: VRRPv2, Advertisement, vrid 99, prio 100, authtype simple, intvl 1s, length 20 192.168.0.117 > 224.0.0.18: VRRPv2, Advertisement, vrid 99, prio 80, authtype simple, intvl 1s, length 20 -
如果结果如上, 说明出现了脑裂(主备都向外宣誓我是老大),
出现这种情况的原因是防火墙或者iptables拦截了vrrp请求, 进行放行即可.
防火墙(推荐):
firewall-cmd --add-rich-rule='rule protocol value="vrrp" accept' --permanent firewall-cmd --reloadiptables:
iptables -A INPUT -s 192.168.1.0/24 -d 224.0.0.18 -j ACCEPT iptables -A INPUT -s 192.168.1.0/24 -p vrrp -j ACCEPT不存在可以进行安装,
yum install iptables-services -
最后附上正常运行结果, 即只有100或者80优先级的机器来广播自己是老大
09:26:55.782258 IP 192.168.0.111 > 224.0.0.18: VRRPv2, Advertisement, vrid 99, prio 100, authtype simple, intvl 1s, length 20 09:26:56.782910 IP 192.168.0.111 > 224.0.0.18: VRRPv2, Advertisement, vrid 99, prio 100, authtype simple, intvl 1s, length 20 09:26:57.783787 IP 192.168.0.111 > 224.0.0.18: VRRPv2, Advertisement, vrid 99, prio 100, authtype simple, intvl 1s, length 20 09:26:58.784709 IP 192.168.0.111 > 224.0.0.18: VRRPv2, Advertisement, vrid 99, prio 100, authtype simple, intvl 1s, length 20 09:26:59.784792 IP 192.168.0.111 > 224.0.0.18: VRRPv2, Advertisement, vrid 99, prio 100, authtype simple, intvl 1s, length 20 09:27:00.785171 IP 192.168.0.111 > 224.0.0.18: VRRPv2, Advertisement, vrid 99, prio 100, authtype simple, intvl 1s, length 20
测试ip漂移
ip漂移: 就是主备切换过程成, vip漂到真实ip上的过程. 也称为主备切换.
测试过程就是停用master机器上面的keepalived或者关机master机器, 查看backup机器是否正常接过来, 一般1s左右可以切换过去. 当出现脑裂情况的时候切换过程也能实现, 只是很慢大约7s左右. 具体原因未深究.
漂移过程可以通过抓包实现, 也可以通过两给ngxin转发到不同tomcat中的项目或网页, 或者修改ngxin的默认网页进行测试都可.
至此位置简单使用就完成了, 下面介绍几个功能配置
VRRP脚本
-
签到keepalived的配置文件夹:
cd /etc/keepalived/ -
创建一个脚本文件:
vim nginx_check.sh#!/bin/bash count=`ps -C nginx --no-header |wc -l` if [ $count -eq 0 ];then killall keepalived fi -
赋予执行权限:
chmode +x nginx_check.sh -
引入脚本:
vim keepalived.conf与
vrrp_instance同级, 其中-
chk_nginx: 脚本名称, 自定义的 -
script: 脚本位置 -
interval: 执行间隔 -
weight: 权重, 如果是负数, 当执行失败时候会影响vrrp_instance中的优先级priority, 因为主备切换是通过优先级的高低的进行切换的, 所以也可以通过这个优先级来进行主动控制主备切换. 而脚本中的内容可以很灵活地实现很多功能. 此个demo中只是简单实现检测到ngxin关闭后自动关闭keepalived, 也可以实现检测启动后进行开启, 然后延迟2s后查看是否启动成功, 未成功再进行关闭keepalived或者降低优先级(配合右键通知).vrrp_script chk_nginx { script "/etc/keepalived/nginx_check.sh" interval 2 #weight -30 }
-
-
配置到vrrp_instance中, 与authentication和virtual_ipaddress同级
track_script { chk_nginx } -
修改后的配置文件
! Configuration File for keepalived vrrp_script chk_nginx { script "/etc/keepalived/nginx_check.sh" interval 2 #weight -30 } vrrp_instance VI_1 { state MASTER interface enp0s3 priority 100 advert_int 1 virtual_router_id 99 authentication { auth_type PASS auth_pass 221531 } track_script { chk_nginx } virtual_ipaddress { 192.168.0.99 } } -
测试
正常启动时候, 手动关闭nginx, 查看keepalived的状态.
-
参考:
-
参考2: Keepalived 的主备切换怎么做
邮件配置
邮件功能是linux上面的mail指令.
-
安装
mail:yum -y install mailx -
编辑配置文件(设置发送人信息):
vim /etc/mail.rc, 在末尾处添加set from=xxx@163.com set smtp=smtp.163.com set smtp-auth-user=xxx@qq.com set smtp-auth-password=KJFHTOSXZQPNFAIU #邮箱需要开启POP3/SMTP服务并设置密钥 set smtp-auth=login set ssl-verify=ignore -
测试mail功能:
echo test mail | mail -s testa 收件人id@qq.com-
-s后面是主题的意思 -
echo test maill中的test mail 是邮件正文. -
最后跟着收件人
-
-
配置到keepalived中, 方法1
-
创建脚本
vim mail_send.sh(记得赋予执行权限)可以使用
./mail_send.sh master进行测试#!/bin/bash contact='收件人邮箱@qq.com' notify() { mailsubject="$(hostname) to be $1, vip 转移" mailbody="$(date +'%F %T'): vrrp transition, $(hostname) changed to be $1" echo "$mailbody" | mail -s "$mailsubject" $contact } case $1 in master) notify master ;; backup) notify backup ;; fault) notify fault ;; *) echo "Usage: $(basename $0) {master|backup|fault}" exit 1 ;; esac -
修改配置文件:
vim keepalived.confvrrp_instance下与authentication同级处
notify_master "/etc/keepalived/mail_send.sh master" notify_backup "/etc/keepalived/mail_send.sh backup" notify_fault "/etc/keepalived/mail_send.sh fault" -
整体配置文件
! Configuration File for keepalived vrrp_script chk_nginx { script "/etc/keepalived/nginx_check.sh" interval 2 #weight -30 } vrrp_instance VI_1 { state MASTER interface enp0s3 priority 100 advert_int 1 virtual_router_id 99 # 当进入master/backup/fault状态时触发脚本, 可携带参数 notify_master "/etc/keepalived/mail_send.sh master" notify_backup "/etc/keepalived/mail_send.sh backup" notify_fault "/etc/keepalived/mail_send.sh fault" authentication { auth_type PASS auth_pass 221531 } track_script { chk_nginx } virtual_ipaddress { 192.168.0.99 } }
-
-
配置到keepalived中, 方法2
-
脚本内容, 下面这个是漂移到master时, 另外创建backup和fault
#!/bin/bash contacts='收件人邮箱1, 收件人邮箱2' ip a > ipa_temp.txt echo "$(date +'%F %T'): Keepalived instance I became MASTER on $(hostname). --- from master" | mail -s "Master Keepalived notification" -a ipa_temp.txt "$contacts" -
修改配置文件:
vim keepalived.confvrrp_instance下与authentication同级处, 后面的
root是执行人和所在组notify_master /etc/keepalived/mail_send_master.sh root root notify_backup /etc/keepalived/mail_send_backup.sh root root notify_fault /etc/keepalived/mail_send_fault.sh root root
-
-
测试状态转移时有没有邮箱接收到即可, 通过重启, 停用
-
参考:
-
参考2: keepalived邮件通知
-
参考3: Keepalived故障切换时的邮件通知
-
参考4: mail指令同时发送给多个用户