Mysql+keepalive高可用自动切换

最新推荐文章于 2023-11-14 15:36:54 发布
最新推荐文章于 2023-11-14 15:36:54 发布
阅读量261 收藏
点赞数
文章标签： 数据库 运维 网络
原文链接：http://www.cnblogs.com/zwj-linux/p/11552364.html
版权

Keepalive

Keepalived是HA cluster的一种实现方案，其以vrrp协议(Virtual Router Redundancy Protocol，即虚拟路由冗余协议)为实现基础，在N台具有相同功能的路由器或服务器中选出一个master或多个backup，这些路由器或服务器会共用一个对外提供服务的VIP。在此HA cluster集群中，master主机会发送组播来跟backup主机进行通信确认双方是否存活。当backup主机在指定的时间内没有收到vrrp组播报文则认为master主机宕机了，此时就会在所有的backup主机中根据vrrp协议来选举出新的master主机来保证服务的高可用性。

1、域名指向，域名只能指向一个固定IP，如果该IP绑定的机器宕机了，那么这个域名便无法访问了；
2、数据库服务指向，应用是通过一个固定的IP连接到数据库，如果该IP宕机，也无法提供服务；

一般主要解决以上两个问题，即keepalived+lvs+nginx和keepalive+mysql等等


keepalived主要有三个模块，分别是ipvs wrapper、checkers和vrrp stack。

  • ipvs wrapper模块为keepalived的核心，负责主进程的启动、维护以及全局配置文件的加载和解析。
  • checkers负责健康检查，包括常见的各种检查方式。
  • vrrp stack模块是来实现VRRP协议的。
  • IPVS wrapper 用来发送设定的规则到内核IPVS代码。
  • Netlink Reflector 用来设定 vrrp 的vip地址等。

 

1、安装

keepalived源码包下载地址：http://www.keepalived.org/download.html

说明文档：https://www.keepalived.org/LVS-NAT-Keepalived-HOWTO.html

安装环境

# yum install -y openssl-devel popt-devel  curl gcc  libnl3-devel net-snmp-devel  libnfnetlink-devel

下载安装包，主机备机都要安装

wget https://www.keepalived.org/software/keepalived-2.0.0.tar.gz

解压缩、编译、安装，过程中提示缺什么库就安装什么库，然后再重新编译安装。

# tar -zxf keepalived-2.0.0.tar.gz 
# cd keepalived-2.0.0
#  ./configure --prefix=/usr/local/keepalived
# make && make install
# echo $?
0
#
# cp /usr/local/keepalived/sbin/keepalived /usr/sbin/
# cp /usr/local/keepalived/etc/sysconfig/keepalived /etc/sysconfig/
# cp ./keepalived/etc/init.d/keepalived /etc/init.d/
# mkdir /etc/keepalived
# cp /usr/local/keepalived/etc/keepalived/keepalived.conf /etc/keepalived/

2、配置

keepalived 只有一个配置文件 keepalived.conf，其中包含了三类配置区域。

  • 全局配置 Global Configuration，包含了 global_defs（全局配置标识）、static_ipaddress（静态地址）、static_routes（静态路由） 区域配置项。

  • VRRP 配置，包含了 vrrp_instance（VRRP 实例）、vrrp_sync_group（VRRP 同步组）、vrrp_script（VRRP 脚本） 、virtual_server 区域配置项。

    • vrrp_instance 用来定义对外提供服务的 VIP 区域及其相关属性。
    • vrrp_rsync_group 用来定义 vrrp_intance 组，使得这个组内成员动作一致。例如两个 vrrp_instance 同属于一个 vrrp_rsync_group，那么其中一个vrrp_instance 发生故障切换时，另一个 vrrp_instance 也会跟着切换（即使这个 instance 没有发生故障）

  • vrrp_script 区域

    • 用来做健康检查，当时检查失败时会将 vrrp_instance 的 priority 减少相应的值。
    • 首先在 vrrp_script 区域定义脚本名字和脚本执行的间隔和脚本执行的优先级变更。然后在实例 vrrp_instance 里面引用，有点类似脚本里面的函数引用一样：先定义，后引用函数名。

 master

[root@nginx ~]# cat /etc/keepalived/keepalived.conf
global_defs {
    notification_email {                  #故障发生时给谁发邮件通知。
        root@localhost
        root@192.168.10.18
    }
    notification_email_from keepalived@localhost      #通知邮件的发出地址。
    smtp_server 127.0.0.1         #发现Email的smtp地址
    smtp_connect_timeout 30       #超时时间
    router_id node1               #运行keepalived的机器标识号，主从机必须不同
}

vrrp_instance VI_1 {
    state MASTER               #可以是 MASTER 或 BACKUP，不过当其他节点 keepalived 启动时
                               #会将 Priority（优先级）比较大的节点选举为 MASTER，因此该项其实没有实质用途。
    interface eth1             #本机物理网卡，节点固有 IP（非 VIP）的网卡，用来发 VRRP 包。
    lvs_sync_daemon_inteface eth1    #绑定 lvs syncd 的网卡。
    virtual_router_id 151            #虚拟路由，主从机必需一致。设置 VRID，取值在 0-255 之间，用来区分多个 instance 的 VRRP 组播
                                     #（同一网段中 virtual_router_id 的值不能重复，否则会出错）。
                                     #将决定多播的 MAC 地址
    priority 100          #用来选举 master，要成为 master，这个选项的值最好高于其他机器 50 个点，
                          #该项取值范围是 1-255（在此范围之外会被识别成默认值 100）。
    advert_int 5          #发 VRRP 包的时间间隔，即多久进行一次 master 选举（可以认为是健康查检时间间隔，默认为 1 秒）。
    authentication {       #认证区域，保持master和backup一致
        auth_type PASS
        auth_pass 123456
    }
    virtual_ipaddress {          #VIP（虚拟 IP 地址），随着 state 的变化而增加删除，
                                 #当 state 为 master 的时候就添加，当 state 为 backup 的时候删除，
                                 #主要由优先级来决定的，和 state 设置的值没有多大关系，可以设置多个 IP 地址。
        192.168.100.100 dev eth1   
    }
}

virtual_server 192.168.100.100 3306 {           #LVS配置，VIP
    delay_loop 6                                #服务轮询的时间间隔
    lb_algo wrr                                 #LVS 调度算法
    lb_kind DR　　　　　　　　　　　　　　　　　　　　 #LVS 集群算法
    persistence_timeout 60                       #同一IP的链接50秒内被分配到同一台realserver
    protocol TCP                                 #用TCP协议检查realserver状态
    real_server 192.168.10.18 3306 {        #实际服务器的IP和端口
        weight 100
        notify_down /data/sh/mysql.sh
        TCP_CHECK {
        connect_timeout 10
        nb_get_retry 3
        delay_before_retry 3
        connect_port 3306
        }
    }

}
[root@nginx ~]#

backup

[root@mysql ~]# 
[root@mysql ~]# cat /etc/keepalived/keepalived.conf
global_defs {
    notification_email {
        root@localhost
        root@192.168.10.18
    }
    notification_email_from keepalived@localhost
    smtp_server 127.0.0.1
    smtp_connect_timeout 30
    router_id node2
}

vrrp_instance VI_1 {
    state BACKUP
    interface eth1
    lvs_sync_daemon_inteface eth1
    virtual_router_id 151
    priority 90
    advert_int 5
    authentication {
        auth_type PASS
        auth_pass 123456
    }
    virtual_ipaddress {
        192.168.100.100 dev eth1
    }
}

virtual_server 192.168.100.100 3306 {
    delay_loop 6
    lb_algo wrr
    lb_kind DR
    persistence_timeout 60
    protocol TCP
    real_server 192.168.10.23 3306 {
        weight 100
        notify_down /data/sh/mysql.sh
        TCP_CHECK {
        connect_timeout 10
        nb_get_retry 3
        delay_before_retry 3
        connect_port 3306
        }
    }

}
[root@mysql ~]#

在两端/data/sh/mysql.sh建立关闭keepalived的脚本

[root@mysql ~]# cat /data/sh/mysql.sh 
pkill keepalived
[root@mysql ~]#

3、启动

master和backup都启动keepalived服务

[root@nginx ~]# service keepalived start
Starting keepalived:                                       [  OK  ]
[root@nginx ~]# 
[root@nginx ~]# ps -ef | grep keep 
root     127942      1  0 12:27 ?        00:00:00 keepalived -D
root     127944 127942  0 12:27 ?        00:00:00 keepalived -D
root     127945 127942  0 12:27 ?        00:00:00 keepalived -D
root     130062 130014  0 14:27 pts/0    00:00:00 grep keep
[root@nginx ~]# 
[root@nginx ~]# 
[root@nginx ~]# netstat -lantup | grep keep
[root@nginx ~]# 
[root@mysql ~]# /etc/init.d/keepalived restart
Stopping keepalived:                                       [  OK  ]
Starting keepalived:                                       [  OK  ]
[root@mysql ~]# ps -ef | grep keep
root      49218      1  0 12:01 ?        00:00:00 keepalived -D
root      49220  49218  0 12:01 ?        00:00:00 keepalived -D
root      49221  49218  0 12:01 ?        00:00:00 keepalived -D
root      49223  48581  0 12:01 pts/3    00:00:00 grep keep
[root@mysql ~]# 
[root@mysql ~]# netstat -lantup | grep keep
tcp        0      1 192.168.10.23:58138         192.168.33.23:3306          SYN_SENT    49220/keepalived    
[root@mysql ~]#

查看两端的虚拟IP启动情况，系统日志

Master端

[root@nginx ~]# ip addr
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: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:69:6e:c1 brd ff:ff:ff:ff:ff:ff
    inet 192.168.10.18/24 brd 192.168.10.255 scope global eth1
    inet 192.168.100.100/32 scope global eth1
    inet6 fe80::20c:29ff:fe69:6ec1/64 scope link 
       valid_lft forever preferred_lft forever
[root@nginx ~]#

虚拟IP：192.168.100.100已经成功启动

[root@nginx ~]# tail -fn 10 /var/log/messages6
Sep 20 15:13:22 nginx Keepalived_healthcheckers[419]: TCP connection to [192.168.10.18]:tcp:3306 success.
Sep 20 15:13:32 nginx Keepalived_vrrp[420]: (VI_1) Entering MASTER STATE
Sep 20 15:13:32 nginx Keepalived_vrrp[420]: (VI_1) setting protocol VIPs.
Sep 20 15:13:32 nginx Keepalived_vrrp[420]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:13:32 nginx Keepalived_vrrp[420]: (VI_1) Sending/queueing gratuitous ARPs on eth1 for 192.168.100.100
Sep 20 15:13:32 nginx Keepalived_vrrp[420]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:13:32 nginx Keepalived_vrrp[420]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:13:32 nginx Keepalived_vrrp[420]: Sending gratuitous ARP on eth1 for 192.168.100.100

说明已经成功作为MASTER启动，且发送ARP包传输

backup端

[root@mysql ~]# tail -fn 10 /var/log/messages
Sep 20 15:25:37 mysql Keepalived_healthcheckers[52339]: WARNING - default user 'keepalived_script' for script execution does not exist - please create.
Sep 20 15:25:37 mysql Keepalived_healthcheckers[52339]: Unknown keyword 'nb_get_retry'
Sep 20 15:25:37 mysql Keepalived_healthcheckers[52339]: Please add a #! shebang to script /data/sh/mysql.sh
Sep 20 15:25:37 mysql Keepalived_vrrp[52340]: (VI_1) removing protocol VIPs.
Sep 20 15:25:37 mysql Keepalived_vrrp[52340]: (VI_1) Entering BACKUP STATE (init)
Sep 20 15:25:37 mysql Keepalived_healthcheckers[52339]: SECURITY VIOLATION - check scripts are being executed but script_security not enabled.
Sep 20 15:25:37 mysql Keepalived_healthcheckers[52339]: Gained quorum 1+0=1 <= 100 for VS [192.168.100.100]:tcp:3306
Sep 20 15:25:37 mysql Keepalived_healthcheckers[52339]: Activating healthchecker for service [192.168.10.23]:tcp:3306 for VS [192.168.100.100]:tcp:3306
Sep 20 15:25:37 mysql Keepalived_vrrp[52340]: VRRP sockpool: [ifindex(2), proto(112), unicast(0), fd(8,9)]
Sep 20 15:25:43 mysql Keepalived_healthcheckers[52339]: TCP connection to [192.168.10.23]:tcp:3306 success.
 (VI_1) Entering BACKUP STATE (init)已作为backup机启用
[root@mysql ~]# ip addr
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: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:ad:ad:54 brd ff:ff:ff:ff:ff:ff
    inet 192.168.10.23/24 brd 192.168.10.255 scope global eth1
    inet6 fe80::20c:29ff:fead:ad54/64 scope link 
       valid_lft forever preferred_lft forever
[root@mysql ~]#

backup的IP情况

 

4、测试

当Master的MySQL服务意外挂掉后，会自动启用backup的Mysql继续为系统提供支持。

[root@nginx ~]# service mysqld stop
Stopping mysqld:                                           [  OK  ]
[root@nginx ~]# 
[root@nginx ~]# 
[root@nginx ~]# ip addr
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: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:69:6e:c1 brd ff:ff:ff:ff:ff:ff
    inet 192.168.10.18/24 brd 192.168.10.255 scope global eth1
    inet6 fe80::20c:29ff:fe69:6ec1/64 scope link 
       valid_lft forever preferred_lft forever
[root@nginx ~]# 
[root@nginx ~]# 
[root@nginx ~]# 
[root@nginx ~]# tail -fn 10 /var/log/messages
Sep 20 15:36:44 nginx Keepalived_healthcheckers[419]: Error processing RCPT cmd on SMTP server [127.0.0.1]:25. SMTP status code = 501
Sep 20 15:36:44 nginx Keepalived_healthcheckers[419]: Can not read data from remote SMTP server [127.0.0.1]:25.
Sep 20 15:36:44 nginx Keepalived[417]: Stopping
Sep 20 15:36:44 nginx kernel: IPVS: __ip_vs_del_service: enter
Sep 20 15:36:44 nginx Keepalived_vrrp[420]: (VI_1) sent 0 priority
Sep 20 15:36:44 nginx Keepalived_vrrp[420]: (VI_1) removing protocol VIPs.
Sep 20 15:36:44 nginx Keepalived_healthcheckers[419]: Shutting down service [192.168.10.18]:tcp:3306 from VS [192.168.100.100]:tcp:3306
Sep 20 15:36:44 nginx Keepalived_healthcheckers[419]: Stopped
Sep 20 15:36:45 nginx Keepalived_vrrp[420]: Stopped
Sep 20 15:36:45 nginx Keepalived[417]: Stopped Keepalived v2.0.0 (05/27,2018)
^C
[root@nginx ~]# ps -ef  | grep keep
root       1127 130014  0 15:40 pts/0    00:00:00 grep keep
[root@nginx ~]#

Master 测试关闭mysql服务，虚拟IP停止发送ARP包，keepalived服务终止

[root@mysql ~]# ip addr
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: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:ad:ad:54 brd ff:ff:ff:ff:ff:ff
    inet 192.168.10.23/24 brd 192.168.10.255 scope global eth1
    inet 192.168.100.100/32 scope global eth1
    inet6 fe80::20c:29ff:fead:ad54/64 scope link 
       valid_lft forever preferred_lft forever
[root@mysql ~]# 
[root@mysql ~]# tail -fn 20 /var/log/messages
Sep 20 15:25:37 mysql Keepalived_healthcheckers[52339]: SECURITY VIOLATION - check scripts are being executed but script_security not enabled.
Sep 20 15:25:37 mysql Keepalived_healthcheckers[52339]: Gained quorum 1+0=1 <= 100 for VS [192.168.100.100]:tcp:3306
Sep 20 15:25:37 mysql Keepalived_healthcheckers[52339]: Activating healthchecker for service [192.168.10.23]:tcp:3306 for VS [192.168.100.100]:tcp:3306
Sep 20 15:25:37 mysql Keepalived_vrrp[52340]: VRRP sockpool: [ifindex(2), proto(112), unicast(0), fd(8,9)]
Sep 20 15:25:43 mysql Keepalived_healthcheckers[52339]: TCP connection to [192.168.10.23]:tcp:3306 success.
Sep 20 15:36:44 mysql Keepalived_vrrp[52340]: (VI_1) Backup received priority 0 advertisement
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: (VI_1) Entering MASTER STATE
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: (VI_1) setting protocol VIPs.
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: (VI_1) Sending/queueing gratuitous ARPs on eth1 for 192.168.100.100
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:50 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:50 mysql Keepalived_vrrp[52340]: (VI_1) Sending/queueing gratuitous ARPs on eth1 for 192.168.100.100
Sep 20 15:36:50 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:50 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:50 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:50 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
ps^C
[root@mysql ~]# 
[root@mysql ~]# ps -ef | grep keepalived
root      52337      1  0 15:25 ?        00:00:00 keepalived -D
root      52339  52337  0 15:25 ?        00:00:00 keepalived -D
root      52340  52337  0 15:25 ?        00:00:00 keepalived -D
root      52760  51101  0 15:42 pts/0    00:00:00 grep keepalived
[root@mysql ~]# 
(VI_1) Entering MASTER STATE，backup机成为Master，继续为系统提供mysql服务，测试成功。
重启Master主机的mysql服务后，Master主机重新成为master，备机重新成为备机
[root@nginx ~]# 
[root@nginx ~]# service keepalived restasrt
Usage: /etc/init.d/keepalived {start|stop|reload|restart|condrestart|status}
[root@nginx ~]# service keepalived restart
Stopping keepalived:                                       [FAILED]
Starting keepalived:                                       [  OK  ]
[root@nginx ~]# 
[root@nginx ~]# 
[root@nginx ~]# 
[root@nginx ~]# tail -fn 10 /var/log/messages
Sep 20 15:44:35 nginx Keepalived_healthcheckers[1409]: Gained quorum 1+0=1 <= 100 for VS [192.168.100.100]:tcp:3306
Sep 20 15:44:35 nginx Keepalived_vrrp[1410]: Unknown keyword 'lvs_sync_daemon_inteface'
Sep 20 15:44:35 nginx Keepalived_vrrp[1410]: Assigned address 192.168.10.18 for interface eth1
Sep 20 15:44:35 nginx Keepalived_healthcheckers[1409]: Activating healthchecker for service [192.168.10.18]:tcp:3306 for VS [192.168.100.100]:tcp:3306
Sep 20 15:44:35 nginx Keepalived_vrrp[1410]: Assigned address fe80::20c:29ff:fe69:6ec1 for interface eth1
Sep 20 15:44:35 nginx Keepalived_vrrp[1410]: Registering gratuitous ARP shared channel
Sep 20 15:44:35 nginx Keepalived_vrrp[1410]: (VI_1) removing protocol VIPs.
Sep 20 15:44:35 nginx Keepalived_vrrp[1410]: (VI_1) Entering BACKUP STATE (init)
Sep 20 15:44:35 nginx Keepalived_vrrp[1410]: VRRP sockpool: [ifindex(2), proto(112), unicast(0), fd(8,9)]
Sep 20 15:44:36 nginx Keepalived_vrrp[1410]: (VI_1) received lower priority (90) advert from 192.168.10.23 - discarding
Sep 20 15:44:41 nginx Keepalived_vrrp[1410]: (VI_1) received lower priority (90) advert from 192.168.10.23 - discarding
Sep 20 15:44:41 nginx Keepalived_healthcheckers[1409]: TCP connection to [192.168.10.18]:tcp:3306 success.
Sep 20 15:44:46 nginx Keepalived_vrrp[1410]: (VI_1) received lower priority (90) advert from 192.168.10.23 - discarding
Sep 20 15:44:50 nginx Keepalived_vrrp[1410]: (VI_1) Entering MASTER STATE
Sep 20 15:44:50 nginx Keepalived_vrrp[1410]: (VI_1) setting protocol VIPs.
Sep 20 15:44:50 nginx Keepalived_vrrp[1410]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:44:50 nginx Keepalived_vrrp[1410]: (VI_1) Sending/queueing gratuitous ARPs on eth1 for 192.168.100.100
Sep 20 15:44:50 nginx Keepalived_vrrp[1410]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:44:50 nginx Keepalived_vrrp[1410]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:44:50 nginx Keepalived_vrrp[1410]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:44:50 nginx Keepalived_vrrp[1410]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:44:55 nginx Keepalived_vrrp[1410]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:44:55 nginx Keepalived_vrrp[1410]: (VI_1) Sending/queueing gratuitous ARPs on eth1 for 192.168.100.100
Sep 20 15:44:55 nginx Keepalived_vrrp[1410]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:44:55 nginx Keepalived_vrrp[1410]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:44:55 nginx Keepalived_vrrp[1410]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:44:55 nginx Keepalived_vrrp[1410]: Sending gratuitous ARP on eth1 for 192.168.100.100
^C
[root@nginx ~]# ip addr
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: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:69:6e:c1 brd ff:ff:ff:ff:ff:ff
    inet 192.168.10.18/24 brd 192.168.10.255 scope global eth1
    inet 192.168.100.100/32 scope global eth1
    inet6 fe80::20c:29ff:fe69:6ec1/64 scope link 
       valid_lft forever preferred_lft forever
[root@nginx ~]# 
[root@mysql ~]# 
[root@mysql ~]# tail -fn 20 /var/log/messages
Sep 20 15:25:37 mysql Keepalived_vrrp[52340]: VRRP sockpool: [ifindex(2), proto(112), unicast(0), fd(8,9)]
Sep 20 15:25:43 mysql Keepalived_healthcheckers[52339]: TCP connection to [192.168.10.23]:tcp:3306 success.
Sep 20 15:36:44 mysql Keepalived_vrrp[52340]: (VI_1) Backup received priority 0 advertisement
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: (VI_1) Entering MASTER STATE
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: (VI_1) setting protocol VIPs.
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: (VI_1) Sending/queueing gratuitous ARPs on eth1 for 192.168.100.100
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:45 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:50 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:50 mysql Keepalived_vrrp[52340]: (VI_1) Sending/queueing gratuitous ARPs on eth1 for 192.168.100.100
Sep 20 15:36:50 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:50 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:50 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:36:50 mysql Keepalived_vrrp[52340]: Sending gratuitous ARP on eth1 for 192.168.100.100
Sep 20 15:44:50 mysql Keepalived_vrrp[52340]: (VI_1) Master received advert from 192.168.10.18 with higher priority 100, ours 90
Sep 20 15:44:50 mysql Keepalived_vrrp[52340]: (VI_1) Entering BACKUP STATE
Sep 20 15:44:50 mysql Keepalived_vrrp[52340]: (VI_1) removing protocol VIPs.
^C
[root@mysql ~]# ip addr
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: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:ad:ad:54 brd ff:ff:ff:ff:ff:ff
    inet 192.168.10.23/24 brd 192.168.10.255 scope global eth1
    inet6 fe80::20c:29ff:fead:ad54/64 scope link 
       valid_lft forever preferred_lft forever
[root@mysql ~]#

 



转载于:https://www.cnblogs.com/zwj-linux/p/11552364.html

Q350121248
关注
  • 0
    点赞
  • 踩
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
mysql启用keepalive_mysql主从之keepalive+MySQL高可用
weixin_39852688的博客
12-19 387
一 keepalive介绍1.1 keepalived 是什么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。1.2 keepalived 工作原理keepalived 是以VRRP 协议为实现基础的，VRRP 全称VirtualRouter Redundancy Protocol，即虚拟路由冗余协议。虚拟路由冗余协议，可以认为是实现路由器高可用的协议，即将N 台...
mysql keepalive设置_keepalive配置mysql自動故障轉移
weixin_39880301的博客
02-04 213
本文先配置了一個雙master環境，互為主從，然后通過Keepalive配置了一個虛擬IP，客戶端通過虛擬IP連接master1,當master1宕機，自動切換到master2。一次只能連接其中一個master進行讀寫，所以是active-passive模式。一 Mysql主主復制搭建1.1 實驗環境兩台機器事先都已經裝好了mysql單實例。 二者的端口號需要保持一致，否則在最后用vip連接的時候...
云原生|docker|基于docker部署高可用keepalived集群
zsk_john的技术分享专用博客
12-07 2477
那么，docker有一个理念是：一次构建，随处运行，通过这次的build示例，将这个理念是体现的非常好了。 本文将在cento7操作系统下利用dockerfile制作keepalived镜像，然后在服务器上组建一个简单的keepalived集群，也就是抛砖引玉吧。
centos7 keepalived+mysql双主高可用部署
hfaul的博客
09-02 1659
centos7 keepalived+mysql双主高可用部署 一、简介 为什么要使用双主部署呢？ 在一台mysql主机情况下，如果该主机存在宕机或者其他什么故障，直接会影响整个服务的不可用。为了确保整体服务的高可用性，不仅要在服务层实现多节点的集群部署，还需要在数据库层实现多节点集群部署。如有两台mysql数据库服务器，若其中有一台mysql服务器故障后，另一台可以立马接替工作。 Keepalived+mysql双主高可用方案原理？ Keepalived+mysql双主来实现MySQL-HA，我
keekeepalived 报错 Keepalived_healthcheckers[1127] Unknown keyword 'nb_get_retry'
qq_36801585的博客
03-27 4702
keepalived 报错： Keepalived_healthcheckers[1127]: Unknown keyword ‘nb_get_retry’ 完整的报错如下 Mar 26 21:09:19 director Keepalived_healthcheckers[1127]: Unknown keyword 'nb_get_retry' 而我只在tcp_check 用到了nb_get...
keepalived配置抢占模式
MssGuo的博客
10-15 2273
keepalived的抢占模式
mysql+lvs+keepalive+mha高可用
07-19
`MHA`是一个用于MySQL集群的高可用性管理工具，它能实时监控MySQL主从复制的状态，并在主服务器故障时自动选择最佳的备份服务器升级为主，同时处理数据同步，确保数据一致性。MHA不仅可以检测并处理主服务器的故障，...
Centos下Mysql+Keepalived双主热备高可用环境完整部署记录（个人珍藏版）
08-29
在本篇文章中，我们将深入探讨如何在CentOS操作系统上配置MySQL与Keepalived实现双主热备高可用环境。MySQL的双主复制意味着两个主服务器互相复制数据，任何一台服务器上的更新都会被复制到另一台服务器，从而确保...
Keepalive+mysql主主同步.pdf
09-04
Keepalive+mysql主主同步是指在高可用性环境中，使用Keepalive和MySQL主主同步技术来实现数据库的高可用性和负载均衡。下面是相关的知识点： 一、Keepalive简介 Keepalive是一个高可用性解决方案，主要用于实现...
MySQL 高可用性—keepalived+mysql双主.docx
06-27
MySQL的高可用性是数据库系统的关键特性，确保在单个实例故障时，服务能够无缝切换到备份节点，维持业务连续性。在这个场景中，我们利用Keepalived和MySQL的双主复制来构建高可用集群。Keepalived是一个开源工具，...
MySQL LVS+Keepalived+MHA 高可用群集 应用部署操作手册
10-02
MHA（Master High Availability）目前在...在MySQL故障切换过程中，MHA能做到0~30秒之内自动完成数据库的故障切换操作，并且在进行故障切换的过程中，MHA能最大程度上保证数据库的一致性，以达到真正意义上的高可用。
Keepalived入门
青晨的歌
11-19 2117
Keepalived入门1 安装和基本操作1.1 准备1.2 安装1.3 配置1.4 启动停止2 常用配置2.1 非抢占模式2.2 密码验证2.3 监控脚本2.4 网卡监控2.5 通知脚本  　Keepalived通过VRRP协议（Virtual Router Redundancy Protocol，虚拟路由器冗余协议）实现高可用功能，解决静态路由单点故障问题，保证个别节点宕机时，整个网络能不间断...
负载均衡之LVS+keepalived
Grit_ICPC的博客
04-26 685
负载均衡是为了解决某台服务器压力大而产生。 下面针对这种问题引出LVS的使用： LVS需要基础设备：一台负载均衡服务器，两台真实服务器（A,B）       就像我们去访问淘宝网站，一致输入都是www.taobao.com进入网站，为了应对大量用户同时访问该网站 ，淘宝的服务器肯定不可能是一台，而是若干台，但是输入网址后他都能够进入该网站中的某台服务器，那么 我们的负载均衡服务器就是这个
Mysql-4 keepalive
weixin_44641254的博客
05-27 269
** 一：配置环境 ** 1： 服务器端： 客户端： 两台机器都： 2： 客户端（192.168.142.146）： （这样做，相当于把第二台机器（客户端）当作服务器了，效果都一样） 客户端授权： 服务器端（192.168.142.145）： 机器146： 机器145： 3：生成master.info文件 机器146： 机器145： 机器146： 机器146和机器145都： yes,表示同步成功。 3：进行测试 机器145： 机器146： 机器146数据库里有了aa,
keepalived源码解析 —— vrrp_state_master_rx()
傲然君
02-07 2526
功能： 1、分析收到的 vrrp 包； 2、若包分析返回结果：VRRP_PACKET_OK，则直接返回 false； 3、若满足以下任一条件： 1、master 收到一个 priority = 0 的通告； 2、配置文件指定了：higher_prio_send_advert（若 master 收到一个更高优先级的 vrrp，则自身在变成 backup 之前发送一个通告） 且 对端 vrrp 的优先...
keepalived 的安装部署及使用详细完整版
最新发布
不积跬步，无以至千里；不积小流，无以成江海。
11-14 525
这部分主要用来设置Keepalived的故障通知机制和Router ID标识mcast_src_ip 192.124.107.129 #当前机器192.124.126.109 #另一台机器192.124.91.241/17 dev eth0 # 设置虚拟ip，主节点自动添加如果两节点的上联交换机禁用了组播，则只能采用vrrp单播通告的方式mcast_src_ip 192.124.107.129 #当前机器192.124.126.109 #另一台机器定义实例名称为VI_1。
Keepalived实现VIP+Nginx实现反向代理
LCK 博客
06-23 1198
一、实验环境 Keepalived+Nginx 服务器：10.0.0.11（keepalived-1）、10.0.0.12（keepalived-2） nginx服务器：10.0.0.13（nginx-web-1）、10.0.0.14（nginx-web-2） VIP：10.0.0.100
双A复制mysql_MySQL 双主复制 + keepalived
weixin_32946543的博客
01-26 555
1.双主规划架构：双主复制+keepalivedmysql01: 192.168.10.161 3306mysql02: 192.168.10.162 3306keepalived: 192.168.10.160操作系统： CentOS 7 x64数据库版本：mysql-5.7.28备注：MySQL 安装过程省略。2.修改参数#mysql01[mysqld]user = mysql...
LVS+keepalived搭建负载均衡学习手札（二）
朱智文的专栏
03-01 590
上次做了LVS+keepalived的负载均衡，效果还不错，但是没有做主备，这次补上： 安装环境：环境 centos4.4 一共准备四台机器： 负载均衡机器： 主：192.168.2.158              #安装lvs+keepalived 备：192.168.2.159              #安装lvs+keepalived VIP:19
Nginx+Keepalive主从双机热备与自动切换技术详解
本文档主要探讨的是如何在IT领域实现高效的 nginx 主从双机热备架构，并且结合 keepalive 技术以及自动切换功能，确保高可用性和服务连续性。以下是关键知识点的详细介绍： 1. **Nginx 安装基础**： Nginx 是一个...
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值