9.5 18.1-18.5

18.1 集群介绍

关于linux集群

分类：

高可用：

通常是两台机器，其中一台机器提供服务，另一台机器备用；

当一台机器宕机，另一台机器会接替继续提供服务；

对于一些关键角色（机器），需要保证其健壮性（高可用）；

高可用衡量标准9999，即在一年时间里99.99%的时间机器都要在线且正常提供服务；

实现高可用的开源软件：heartbeat、keepalived

Heartbeat在centos6下有bug，且长期未更新新的版本，不推荐使用

keepalived有高可用和负载均衡两个功能

负载均衡：

可以将用户的请求分发到后端的多台服务器上分别处理，此时需要有一台服务器作为分发器；

此时至少有两台服务器在后端提供服务；

实现负载均衡的开源软件：LVS、Keepalived、haproxy、nginx

实现负载均衡的商用设备：F5、Netscaler

一般商用负载均衡设备成本较高，但可以胜任更大的并发量，稳定性较好；

使用开源软件搭建的负载均衡稳定性取决于物理服务器的稳定性；

18.2 keepalived介绍

Keepalived通过VRRP协议实现高可用；

VRRP：

该协议在网络中用于实现路由器冗余，在这里用于实现服务器高可用；

由多台机器组成一个组，一个组内包含多台机器，其中包含了一个master角色和多个backup角色（backup至少为1个）；

VRRP工作过程：

正常的master会以组播形式每隔一段时间向所有backup发送数据包，当backup在超过一定时间仍未收到master的数据包时，backup会认为master已宕机

此时会根据优先级从backup中选出新的master，如果只有一个backup则唯一的backup直接启动成为新的master

Keepalived的三个模块：

core

keepalived的核心，负责主进程启动、维护及全局配置文件加载、解析

check

负责健康检查

Vrrp

用于实现vrrp协议

18.3 用keepalived配置高可用集群（上）

配置前准备：

准备两台机器，一个作为master，一个作为backup；

在两台机器上安装keepalived软件：

[root@hyc-01-01 ~]# yum install -y keepalived

给两台服务器均安装nginx作为高可用的对象；

[root@hyc-01-01 ~]# rpm -qa|grep nginx

用于检查是否yum安装过某个软件，非yum安装可能检测不到

[root@hyc-01 ~]# yum install -y epel-release

[root@hyc-01 ~]# yum install -y nginx

很多企业将nginx用作负载均衡器，若负载均衡器故障，则即使后端服务器正常运行也无法正常提供服务

配置keepalived配置文件（主机器上）：

[root@hyc-01-01 ~]# vim /etc/keepalived/keepalived.conf

[root@hyc-01-01 ~]# > !$ 清空keepalived配置文件

> /etc/keepalived/keepalived.conf

将提前编辑好的配置拷贝进配置文件：

[root@hyc-01-01 ~]# vim /etc/keepalived/keepalived.conf

这里配置脚本在代码托管平台复制（视频中提供）

global_defs {

notification_email {

[email protected]

}

notification_email_from [email protected]

smtp_server 127.0.0.1

smtp_connect_timeout 30

router_id LVS_DEVEL

}

vrrp_script chk_nginx {

script "/usr/local/sbin/check_ng.sh" 执行编写好的检测脚本

interval 3 每隔3s检测一次

}

vrrp_instance VI_1 {

state MASTER 定义角色为master，若为从此处为BACKUP

interface ens33 定义组播发送vrrp数据包的网卡

virtual_router_id 51

priority 100 权重值，主和备的权重值不能一样

advert_int 1

authentication {

auth_type PASS 定义认证类型为密码认证

auth_pass aminglinux>com 密码

}

virtual_ipaddress {

192.168.188.100

}

track_script {

chk_nginx 前面定义了检测脚本，这里需要加载脚本

}

global_defs 全局定义参数

notification_email 出现问题时，给特定的邮箱发送邮件

vrrp_script chk_nginx 检测服务是否正常，这里检测nginx服务，若不正常需要将服务启动起来

vrrp_instance VI_1 用于定义master相关参数，若为backup，此处定义backup相关参数

virtual_router_id 51 同组机器的id需要保持一致

authentication 认证相关信息

virtual_ipaddress

定义virtual ip，主从均可使用该ip对外提供服务；

主存活时由主绑定该ip，主宕机后由从绑定该ip，无论主或从提供服务，均将域名解析到该ip上，这样主备切换对用户是透明的；

编辑nginx检测脚本：

按keepalived.conf文件中定义的检测脚本路径、名称编辑

[root@hyc-01-01 ~]# vim /usr/local/sbin/check_ng.sh

#!/bin/bash

#时间变量，用于记录日志

d=`date --date today +%Y%m%d_%H:%M:%S`

#计算nginx进程数量

n=`ps -C nginx --no-heading|wc -l`

#如果进程为0，则启动nginx，并且再次检测nginx进程数量，

#如果还为0，说明nginx无法启动，此时需要关闭keepalived

if [ $n -eq "0" ]; then

/etc/init.d/nginx start

n2=`ps -C nginx --no-heading|wc -l`

if [ $n2 -eq "0" ]; then

echo "$d nginx down,keepalived will stop" >> /var/log/check_ng.log

#若nginx无法启动成功，则keepalived失去存在意义，应该被杀掉，主上的keepalived被杀掉后从的keepalived会自动启动

systemctl stop keepalived

修改检测脚本文件权限为755，使所有用户均有权执行该脚本：

[root@hyc-01-01 sbin]# chmod 755 /usr/local/sbin/check_ng.sh

[root@hyc-01-01 sbin]# ls -l /usr/local/sbin/check_ng.sh

-rwxr-xr-x 1 root root 567 9月 6 07:39 /usr/local/sbin/check_ng.sh

脑裂：

若主上无法正常启动nginx，则需要启动从的keepalived；、

若主和从的keepalived同时存在则两者会争抢资源，比如virtual ip；

若两台机器都使用该virtual ip则在域名解析时会出现混乱；

18.4 用keepalived配置高可用集群（中）

启动主上的keepalived：

[root@hyc-01-01 sbin]# systemctl start keepalived

[root@hyc-01-01 sbin]# ps aux|grep keep

root 5008 0.0 0.0 118652 1400 ? Ss 07:45 0:00 /usr/sbin/keepalived -D

root 5009 0.0 0.1 127516 3288 ? S 07:45 0:00 /usr/sbin/keepalived -D

root 5010 0.0 0.1 127456 2840 ? S 07:45 0:00 /usr/sbin/keepalived -D

root 5059 0.0 0.0 112720 980 pts/1 S+ 07:46 0:00 grep --color=autokeep

[root@hyc-01-01 sbin]# ps aux|grep nginx

root 4415 0.0 0.0 46040 1264 ? Ss 06:11 0:00 nginx: master process /usr/local/nginx/sbin/nginx -c /usr/local/nginx/conf/nginx.conf

nobody 4416 0.0 0.2 48528 3908 ? S 06:11 0:00 nginx: worker process

nobody 4417 0.0 0.2 48528 3908 ? S 06:11 0:00 nginx: worker process

root 5097 0.0 0.0 112720 976 pts/1 R+ 07:46 0:00 grep --color=autonginx

测试停掉的nginx服务会不会自动加载：

[root@hyc-01-01 sbin]# /etc/init.d/nginx stop 停止nginx服务

Stopping nginx (via systemctl): [ 确定 ]

[root@hyc-01-01 sbin]# ps aux|grep nginx nginx又启动了

root 5462 0.0 0.0 46040 1264 ? Ss 07:49 0:00 nginx: master process /usr/local/nginx/sbin/nginx -c /usr/local/nginx/conf/nginx.conf

nobody 5464 0.0 0.2 48528 3904 ? S 07:49 0:00 nginx: worker process

nobody 5465 0.0 0.2 48528 3904 ? S 07:49 0:00 nginx: worker process

root 5469 0.0 0.0 112720 976 pts/1 R+ 07:49 0:00 grep --color=autonginx

查看keepalived日志：

[root@hyc-01-01 sbin]# less /var/log/messages

…

r 192.168.188.100

Sep 6 07:45:55 hyc-01-01 Keepalived_vrrp[5010]: Sending gratuitous ARP on ens33 for 192.168.188.100

Sep 6 07:46:00 hyc-01-01 Keepalived_vrrp[5010]: Sending gratuitous ARP on ens33 for 192.168.188.100

Sep 6 07:46:00 hyc-01-01 Keepalived_vrrp[5010]: VRRP_Instance(VI_1) Sending/queueing gratuitous ARPs on ens33 for 192.168.188.100

Sep 6 07:46:00 hyc-01-01 Keepalived_vrrp[5010]: Sending gratuitous ARP on ens33 for 192.168.188.100

Sep 6 07:49:12 hyc-01-01 systemd: Stopping SYSV: http service....

Sep 6 07:49:12 hyc-01-01 nginx: Stopping Nginx: [ 确定 ]

Sep 6 07:49:12 hyc-01-01 systemd: Stopped SYSV: http service..

Sep 6 07:49:14 hyc-01-01 systemd: Starting SYSV: http service....

Sep 6 07:49:14 hyc-01-01 nginx: Starting Nginx: [ 确定 ]

Sep 6 07:49:14 hyc-01-01 systemd: Started SYSV: http service..

Virtual ip无法使用ifconfig查看，必须使用ip add

为保证实验成功，需要检查主和从的防火墙和selinux，（关闭selinux和防火墙）：

防火墙没有策略可以不关闭，有策略时需要关闭

[root@hyc-01-01 sbin]# iptables -nvL

Chain INPUT (policy ACCEPT 0 packets, 0 bytes)

pkts bytes target prot opt in out source destination

Chain FORWARD (policy ACCEPT 0 packets, 0 bytes)

pkts bytes target prot opt in out source destination

Chain OUTPUT (policy ACCEPT 0 packets, 0 bytes)

pkts bytes target prot opt in out source destination

[root@hyc-01-01 sbin]# systemctl stop firewalld

[root@hyc-01-01 sbin]# getenforce

Disabled

在从上配置keepalived：

[root@hyc-01 ~]# > /etc/keepalived/keepalived.conf 清空配置文件原有配置

编辑从上的keepalived配置文件：

[root@hyc-01 ~]# vim /etc/keepalived/keepalived.conf

global_defs {

notification_email {

[email protected]

}

notification_email_from [email protected]

smtp_server 127.0.0.1

smtp_connect_timeout 30

router_id LVS_DEVEL

}

vrrp_script chk_nginx {

script "/usr/local/sbin/check_ng.sh"

interval 3

}

vrrp_instance VI_1 {

state BACKUP

interface ens33

virtual_router_id 51

priority 90 从的优先级必须低于主

advert_int 1

authentication {

auth_type PASS

auth_pass aminglinux>com

}

virtual_ipaddress {

192.168.31.100 主和从的虚拟ip保持一致

}

track_script {

chk_nginx

}

配置从上keepalived执行的监控脚本：

[root@hyc-01 ~]# vim /usr/local/sbin/check_ng.sh

#时间变量，用于记录日志

d=`date --date today +%Y%m%d_%H:%M:%S`

#计算nginx进程数量

n=`ps -C nginx --no-heading|wc -l`

#如果进程为0，则启动nginx，并且再次检测nginx进程数量，

#如果还为0，说明nginx无法启动，此时需要关闭keepalived

if [ $n -eq "0" ]; then

systemctl start nginx

主和从启动nginx的命令不同，主上的nginx为自定义脚本编译安装，可以使用/etc/init.d/nginx start命令启动；而从是yum安装，需要使用systemctl start nginx命令启动

n2=`ps -C nginx --no-heading|wc -l`

if [ $n2 -eq "0" ]; then

echo "$d nginx down,keepalived will stop" >> /var/log/check_ng.log

systemctl stop keepalived

主从的监控脚本内容基本一致

[root@hyc-01 ~]# chmod 755 /usr/local/sbin/check_ng.sh 为保证监控脚本可以正常执行，赋予文件755权限

启动从上的keepalived：

[root@hyc-01 ~]# systemctl start keepalived

[root@hyc-01 ~]# ps aux|grep keep

root 4642 0.0 0.1 118608 1384 ? Ss 21:33 0:00 /usr/sbin/keepalived -D

root 4643 0.0 0.3 127468 3280 ? S 21:33 0:00 /usr/sbin/keepalived -D

root 4644 0.2 0.2 127408 2804 ? S 21:33 0:00 /usr/sbin/keepalived -D

root 4679 0.0 0.0 112676 980 pts/1 R+ 21:33 0:00 grep --color=autokeep

测试nginx：

主上nginx默认虚拟主机的默认主页：

[root@hyc-01-01 b]# cat index.html

this is nginx master from keepalived

从上nginx默认虚拟主机的默认主页：

[root@hyc-01 html]# cat index.html

hello world

访问虚拟地址192.168.31.100：

当访问虚拟地址时访问的是主上的内容

查看主和从的ip地址：

主上：

[root@hyc-01-01 b]# ip addr

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1

link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00

inet 127.0.0.1/8 scope host lo

valid_lft forever preferred_lft forever

inet6 ::1/128 scope host

valid_lft forever preferred_lft forever

2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:4d:9d:95 brd ff:ff:ff:ff:ff:ff

inet 192.168.31.129/24 brd 192.168.31.255 scope global ens33

valid_lft forever preferred_lft forever

inet 192.168.31.100/32 scope global ens33

valid_lft forever preferred_lft forever

inet6 fe80::526f:cdd5:5651:5dcc/64 scope link

valid_lft forever preferred_lft forever

inet6 fe80::d46b:4589:4da1:2f34/64 scope link tentative dadfailed

valid_lft forever preferred_lft forever

3: ens37: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:4d:9d:9f brd ff:ff:ff:ff:ff:ff

inet 192.168.100.1/24 brd 192.168.100.255 scope global ens37

valid_lft forever preferred_lft forever

inet6 fe80::27ac:760a:76cf:bb27/64 scope link

valid_lft forever preferred_lft forever

从上：

[root@hyc-01 html]# ip addr

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1

link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00

inet 127.0.0.1/8 scope host lo

valid_lft forever preferred_lft forever

inet6 ::1/128 scope host

valid_lft forever preferred_lft forever

2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:5b:59:7c brd ff:ff:ff:ff:ff:ff

inet 192.168.31.128/24 brd 192.168.31.255 scope global ens33

valid_lft forever preferred_lft forever

inet6 fe80::d46b:4589:4da1:2f34/64 scope link

valid_lft forever preferred_lft forever

此时虚拟ip在主的网卡上

18.5 用keepalived配置高可用集群（下）

测试高可用：

主上的防火墙规则DROP出向的vrrp数据包：

[root@hyc-01-01 ~]# iptables -I OUTPUT -p vrrp -j DROP

[root@hyc-01-01 ~]# iptables -nvL

Chain INPUT (policy ACCEPT 135 packets, 6236 bytes)

pkts bytes target prot opt in out source destination

Chain FORWARD (policy ACCEPT 0 packets, 0 bytes)

pkts bytes target prot opt in out source destination

Chain OUTPUT (policy ACCEPT 20 packets, 1684 bytes)

pkts bytes target prot opt in out source destination

148 5920 DROP 112 -- * * 0.0.0.0/0 0.0.0.0/0

主上的日志：

[root@hyc-01-01 ~]# tail /var/log/messages

Sep 8 15:22:13 hyc-01-01 Keepalived_vrrp[30534]: Sending gratuitous ARP on ens33 for 192.168.31.100

Sep 8 15:22:13 hyc-01-01 Keepalived_vrrp[30534]: VRRP_Instance(VI_1) Received advert with lower priority 90, ours 100, forcing new election

Sep 8 15:22:13 hyc-01-01 Keepalived_vrrp[30534]: Sending gratuitous ARP on ens33 for 192.168.31.100

Sep 8 15:22:13 hyc-01-01 Keepalived_vrrp[30534]: VRRP_Instance(VI_1) Sending/queueing gratuitous ARPs on ens33 for 192.168.31.100

Sep 8 15:22:13 hyc-01-01 Keepalived_vrrp[30534]: Sending gratuitous ARP on ens33 for 192.168.31.100

从上的日志：

[root@hyc-01 ~]# tail /var/log/messages

Sep 8 15:19:36 hyc-01 Keepalived_vrrp[43613]: Sending gratuitous ARP on ens33 for 192.168.31.100

Sep 8 15:19:41 hyc-01 Keepalived_vrrp[43613]: Sending gratuitous ARP on ens33 for 192.168.31.100

Sep 8 15:19:41 hyc-01 Keepalived_vrrp[43613]: VRRP_Instance(VI_1) Sending/queueing gratuitous ARPs on ens33 for 192.168.31.100

Sep 8 15:19:41 hyc-01 Keepalived_vrrp[43613]: Sending gratuitous ARP on ens33 for 192.168.31.100

主上的ip：

[root@hyc-01-01 ~]# ip addr

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1

link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00

inet 127.0.0.1/8 scope host lo

valid_lft forever preferred_lft forever

inet6 ::1/128 scope host

valid_lft forever preferred_lft forever

2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:4d:9d:95 brd ff:ff:ff:ff:ff:ff

inet 192.168.31.129/24 brd 192.168.31.255 scope global ens33

valid_lft forever preferred_lft forever

inet 192.168.31.100/32 scope global ens33

valid_lft forever preferred_lft forever

inet6 fe80::526f:cdd5:5651:5dcc/64 scope link

valid_lft forever preferred_lft forever

inet6 fe80::d46b:4589:4da1:2f34/64 scope link tentative dadfailed

valid_lft forever preferred_lft forever

3: ens37: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:4d:9d:9f brd ff:ff:ff:ff:ff:ff

inet 192.168.100.1/24 brd 192.168.100.255 scope global ens37

valid_lft forever preferred_lft forever

inet6 fe80::27ac:760a:76cf:bb27/64 scope link

valid_lft forever preferred_lft forever

从上的ip：

[root@hyc-01 ~]# ip addr

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1

link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00

inet 127.0.0.1/8 scope host lo

valid_lft forever preferred_lft forever

inet6 ::1/128 scope host

valid_lft forever preferred_lft forever

2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:5b:59:7c brd ff:ff:ff:ff:ff:ff

inet 192.168.31.128/24 brd 192.168.31.255 scope global ens33

valid_lft forever preferred_lft forever

inet 192.168.31.100/32 scope global ens33

valid_lft forever preferred_lft forever

inet6 fe80::d46b:4589:4da1:2f34/64 scope link

valid_lft forever preferred_lft forever

此时虚拟ip同时在主和从上产生，发生脑裂现象

直接封掉主发出的vrrp包无法彻底切换主从

[root@hyc-01-01 ~]# iptables –F 清空主上的防火墙规则

关停主的keepalived服务：

[root@hyc-01-01 ~]# systemctl stop keepalived

主上的ip：

[root@hyc-01-01 ~]# ip addr

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1

link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00

inet 127.0.0.1/8 scope host lo

valid_lft forever preferred_lft forever

inet6 ::1/128 scope host

valid_lft forever preferred_lft forever

2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:4d:9d:95 brd ff:ff:ff:ff:ff:ff

inet 192.168.31.129/24 brd 192.168.31.255 scope global ens33

valid_lft forever preferred_lft forever

inet6 fe80::526f:cdd5:5651:5dcc/64 scope link

valid_lft forever preferred_lft forever

inet6 fe80::d46b:4589:4da1:2f34/64 scope link tentative dadfailed

valid_lft forever preferred_lft forever

3: ens37: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:4d:9d:9f brd ff:ff:ff:ff:ff:ff

inet 192.168.100.1/24 brd 192.168.100.255 scope global ens37

valid_lft forever preferred_lft forever

inet6 fe80::27ac:760a:76cf:bb27/64 scope link

valid_lft forever preferred_lft forever

从上的ip：

[root@hyc-01 ~]# ip addr

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1

link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00

inet 127.0.0.1/8 scope host lo

valid_lft forever preferred_lft forever

inet6 ::1/128 scope host

valid_lft forever preferred_lft forever

2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:5b:59:7c brd ff:ff:ff:ff:ff:ff

inet 192.168.31.128/24 brd 192.168.31.255 scope global ens33

valid_lft forever preferred_lft forever

inet 192.168.31.100/32 scope global ens33

valid_lft forever preferred_lft forever

inet6 fe80::d46b:4589:4da1:2f34/64 scope link

valid_lft forever preferred_lft forever

此时主上的虚拟ip（VIP）被彻底释放且从上拥有VIP

从上的日志：

[root@hyc-01 ~]# tail -40 /var/log/messages

Sep 8 15:19:26 hyc-01 Keepalived_healthcheckers[43612]: Opening file '/etc/keepalived/keepalived.conf'.

Sep 8 15:19:32 hyc-01 Keepalived_vrrp[43613]: SECURITY VIOLATION - scripts are being executed but script_security not enabled.

Sep 8 15:19:32 hyc-01 Keepalived_vrrp[43613]: VRRP_Instance(VI_1) removing protocol VIPs.

Sep 8 15:19:32 hyc-01 Keepalived_vrrp[43613]: Using LinkWatch kernel netlink reflector...

Sep 8 15:19:32 hyc-01 Keepalived_vrrp[43613]: VRRP_Instance(VI_1) Entering BACKUP STATE

Sep 8 15:19:32 hyc-01 Keepalived_vrrp[43613]: VRRP sockpool: [ifindex(2), proto(112), unicast(0), fd(10,11)]

Sep 8 15:19:32 hyc-01 Keepalived_vrrp[43613]: VRRP_Script(chk_nginx) succeeded

Sep 8 15:19:35 hyc-01 Keepalived_vrrp[43613]: VRRP_Instance(VI_1) Transition to MASTER STATE

Sep 8 15:19:36 hyc-01 Keepalived_vrrp[43613]: VRRP_Instance(VI_1) Entering MASTER STATE

Sep 8 15:19:36 hyc-01 Keepalived_vrrp[43613]: VRRP_Instance(VI_1) setting protocol VIPs.

Sep 8 15:19:36 hyc-01 Keepalived_vrrp[43613]: Sending gratuitous ARP on ens33 for 192.168.31.100

Sep 8 15:19:36 hyc-01 Keepalived_vrrp[43613]: VRRP_Instance(VI_1) Sending/queueing gratuitous ARPs on ens33 for 192.168.31.100

Sep 8 15:19:36 hyc-01 Keepalived_vrrp[43613]: Sending gratuitous ARP on ens33 for 192.168.31.100

…

重新启动主的keepalived服务：

[root@hyc-01-01 ~]# systemctl start keepalived

[root@hyc-01-01 ~]# ip addr

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN qlen 1

link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00

inet 127.0.0.1/8 scope host lo

valid_lft forever preferred_lft forever

inet6 ::1/128 scope host

valid_lft forever preferred_lft forever

2: ens33: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:4d:9d:95 brd ff:ff:ff:ff:ff:ff

inet 192.168.31.129/24 brd 192.168.31.255 scope global ens33

valid_lft forever preferred_lft forever

inet 192.168.31.100/32 scope global ens33

valid_lft forever preferred_lft forever

inet6 fe80::526f:cdd5:5651:5dcc/64 scope link

valid_lft forever preferred_lft forever

inet6 fe80::d46b:4589:4da1:2f34/64 scope link tentative dadfailed

valid_lft forever preferred_lft forever

3: ens37: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:4d:9d:9f brd ff:ff:ff:ff:ff:ff

inet 192.168.100.1/24 brd 192.168.100.255 scope global ens37

valid_lft forever preferred_lft forever

inet6 fe80::27ac:760a:76cf:bb27/64 scope link

valid_lft forever preferred_lft forever

启动主上的keepalived后，虚拟ip立即切换回主上

实现mysql高可用时需要保证主和所有从的数据保持一致，否则会出问题

猜你喜欢