heartbeat+mysql双主复制实现高可用

实验环境

一：搭建主主复制环境

1.1实验环境

两台机器事先都已经装好了MySQL单实例。

IP: 10.192.203.201
10.192.203.202

端口都是3307.

二者的端口号需要保持一致，否则在最后用vip连接的时候，不能使用相同端口号连接。

1.2实验步骤

1.2.1修改配置文件

修改master1：

在[mysqld]下面添加：

server-id = 1
relay-log=/data/server/mysql_3307/binlog/ZabbixServer-relay-bin
relay-log-index=/data/server/mysql_3307/binlog/ZabbixServer-relay-bin.index
auto-increment-offset= 1
auto-increment-increment= 2
log-slave-updates=true

修改master2：

在[mysqld]下面添加：

server-id = 3
relay-log=/data/server/mysql/binlog/single-relay-bin
relay-log-index=/data/server/mysql/binlog/single-relay-bin.index
auto-increment-offset= 2
auto-increment-increment= 2
log-slave-updates=true

添加auto-increment-offset那两项,是为了避免在MySQLINSERT时主键冲突。

修改完后记得重启mysql

1.2.2建复制用户

分别在两台mysql上执行

GRANT REPLICATION SLAVE ON *.* TO ‘RepUser‘@‘%‘identified by ‘beijing‘;

1.2.3指向master

两台服务器均为新建立，且无其它写入操作，各服务器只需记录当前自己二进制日志文件及事件位置，以之作为另外的服务器复制起始位置即可。否则，需要先备份主库，在备库进行恢复，从而保持数据一致，然后再指向master。

Master1:

mysql>show master status;

+------------------+----------+--------------+------------------+-------------------+

|File            |Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set |

+------------------+----------+--------------+------------------+-------------------+

|mysql-bin.000001 |     302|             |                 |                  |

+------------------+----------+--------------+------------------+-------------------+

1 row inset (0.00 sec)

Master2:

mysql>show master status;

+------------------+----------+--------------+------------------+-------------------+

|File            |Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set |

+------------------+----------+--------------+------------------+-------------------+

|mysql-bin.000001 |      120|             |                 |                  |

+------------------+----------+--------------+------------------+-------------------+

1 row inset (0.00 sec)

#Master1指向Master2

1. CHANGE MASTER TO MASTER_USER=‘RepUser‘,MASTER_HOST=‘10.192.203.202‘,MASTER_PASSWORD=‘beijing‘,MASTER_PORT=3307,MASTER_LOG_FILE=‘mysql-bin.000001‘,MASTER_LOG_POS=120;

#Master2指向Master1

[1. CHANGE MASTER TO MASTER_USER=‘RepUser‘,MASTER_HOST=‘10.192.203.201‘,MASTER_PASSWORD=‘beijing‘, MASTER_PORT=3307,MASTER_LOG_FILE=‘mysql-bin.000001‘,MASTER_LOG_POS=302;

1.2.4分别启动slave

startslave ;

确保show slave status

Slave_IO_Running:Yes

Slave_SQL_Running:Yes

测试两边是否同步，略。

二 配心跳

每个主机分别带有两块以太网卡，其中一块用于网络通信，另一块用于心跳功能。

本实验都是在Oracle virtualbox虚拟机里做的，故添加一块儿用于内部连接的网卡，用于心跳测试，请参考：http://blog.csdn.net/yabingshi_tech/article/details/51445006

三：安装部署heartbeat

在两台机器上分别做以下操作：

3.1 安装依赖包

yum install PyXML cluster-glue cluster-glue-libs resource-agents –y

3.2 安装heartbeat

wget http://dl.fedoraproject.org/pub/epel/6/x86_64/heartbeat-3.0.4-2.el6.x86_64.rpm

wget http://dl.fedoraproject.org/pub/epel/6/x86_64/heartbeat-libs-3.0.4-2.el6.x86_64.rpm

rpm -ivh heartbeat-*

3.3 配置heartbeat

复制配置文件

cp /usr/share/doc/heartbeat-3.0.4/authkeys  /etc/ha.d/

cp /usr/share/doc/heartbeat-3.0.4/haresources  /etc/ha.d/

cp /usr/share/doc/heartbeat-3.0.4/ha.cf   /etc/ha.d/

3.3.1 配置心跳的加密方式:authkeys

vi /etc/ha.d/authkeys

#如果使用双机对联线(双绞线)，可以配置如下：

auth 1

1 crc

#存盘退出，然后

chmod 600 /etc/ha.d/authkeys

3.3.2 配置心跳的监控：haresources

vi /etc/ha.d/haresources

#各主机这部分应完全相同。

添加：

PC IPaddr::10.192.203.203

#注意，PC这写你的master的主机名，Ipaddr写的是你的VIP地址。

也可设置heartbeat管理的资源或服务：在该目录下存放服务启动脚本（例如：mysql）,将相同脚本名称添到/etc/ha.d/haresources内容中，从而跟随heartbeat启动而启动该脚本。

如：PC IPaddr::10.192.203.203 mysql #

但是，这样当heartbeat关闭的时候，也会关闭mysql，所以这里我就不添加了。

3.3.3 配置心跳的配置文件：ha.cf

主和从机器除了ucast eth1 10.0.0.2这一行不同外，其他都一样。

vi /etc/ha.d/ha.cf

添加：

logfile/var/log/ha_log/ha-log.log   ## ha的日志文件记录位置。如没有该目录，则需要手动添加
bcast eth1     ##使用eht1做心跳监测
ucast eth1 10.0.0.2   ##心跳网卡连接对方心跳地址
keepalive 2    ##设定心跳(监测)时间时间为2秒
warntime 10
deadtime 30
initdead 120
hopfudge 1
udpport 694    ##使用udp端口694 进行心跳监测
auto_failback off
node PC  ##节点1，必须要与 uname -n 指令得到的结果一致。
node slave2  ##节点2
ping 10.192.203.254   ##通过ping 网关来监测心跳是否正常。

3.3.4 创建日志文件路径

mkdir -p /var/log/ha_log

chmod 777 /var/log/ha_log/

3.3.5 设置ipvsadm的巡回监测

ipvsadm -A -t 10.192.203.203:3307 -s rr

ipvsadm -a -t 10.192.203.203:3307 -r 10.192.203.201:3307-m

ipvsadm -a -t 10.192.203.203:3307 -r 10.192.203.202:3307-m

[[email protected] download]# ipvsadm --list

IP Virtual Server version 1.2.1 (size=4096)

Prot LocalAddress:Port Scheduler Flags

-> RemoteAddress:Port          Forward Weight ActiveConn InActConn

TCP bogon:opsession-prxy rr

-> bogon:opsession-prxy        Local   1      0         0

-> bogon:opsession-prxy        Masq    1      0          0

[[email protected] download]# ipvsadm --list

IP Virtual Server version 1.2.1 (size=4096)

Prot LocalAddress:Port Scheduler Flags

-> RemoteAddress:Port          Forward Weight ActiveConn InActConn

TCP 10.192.203.203:opsession-prx rr

-> 10.192.203.201:opsession-prx Masq    1     0          0

-> 10.192.203.202:opsession-prx Local   1     0          0

3.4 开放防火墙端口

heartbeat 默认使用udp 694端口进行心跳监测。 如果系统有使用iptables 做防火墙，应记住把这个端口打开。

vi/etc/sysconfig/iptables

添加：-A INPUT -pudp --dport 694 -j ACCEPT

service iptables restart

3.5 HA服务的启动、关闭以及测试

启动HA: serviceheartbeat start

在主从都启动heartbeat

[[email protected] init.d]# service heartbeat start

Starting High-Availability services:INFO:  Resource is stopped

Done.

[[email protected] ha_log]# service heartbeat status

heartbeat OK [pid 17943 et al] is runningon pc [pc]...

[[email protected] ha_log]# service heartbeatstatus

heartbeat OK [pid 6536 et al] is running onslave2 [slave2]...

在主上看到虚拟IP了：

[[email protected] ha_log]# ip addr

1: lo: <LOOPBACK,UP,LOWER_UP> mtu16436 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: eth0:<BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen1000
   link/ether 08:00:27:04:05:16 brd ff:ff:ff:ff:ff:ff
   inet 10.192.203.201/24 brd 10.192.203.255 scope global eth0
   inet 10.192.203.203/24 brd 10.192.203.255 scope global secondary eth0
   inet6 fe80::a00:27ff:fe04:516/64 scope link tentative dadfailed
      valid_lft forever preferred_lft forever
3: eth1:<BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen1000
   link/ether 08:00:27:3a:ec:3c brd ff:ff:ff:ff:ff:ff
   inet 10.0.0.1/24 brd 10.0.0.255 scope global eth1
   inet6 fe80::a00:27ff:fe3a:ec3c/64 scope link tentative dadfailed
      valid_lft forever preferred_lft forever

在/var/log/ha_log下的日志文件或者/var/log/messages 都可以看到相关信息。

[[email protected] network-scripts]# tail -f/var/log/messages

May 19 01:34:59 PCResourceManager(default)[17985]: info: Running /etc/ha.d/resource.d/IPaddr10.192.203.203 start
May 19 01:35:00 PCIPaddr(IPaddr_10.192.203.203)[18103]: INFO: Adding inet address10.192.203.203/24 with broadcast address 10.192.203.255 to device eth0
May 19 01:35:00 PCIPaddr(IPaddr_10.192.203.203)[18103]: INFO: Bringing device eth0 up
May 19 01:35:00 PCIPaddr(IPaddr_10.192.203.203)[18103]: INFO: /usr/libexec/heartbeat/send_arp -i200 -r 5 -p /var/run/resource-agents/send_arp-10.192.203.203 eth010.192.203.203 auto not_used not_used
May 19 01:35:00 PC/usr/lib/ocf/resource.d//heartbeat/IPaddr(IPaddr_10.192.203.203)[18089]:INFO:  Success
May 19 01:35:00 PCResourceManager(default)[17985]: info: Running /etc/init.d/mysql  start
May 19 01:35:03 PC heartbeat: [17972]:info: local HA resource acquisition completed (standby).
May 19 01:35:03 PC heartbeat: [17943]:info: Standby resource acquisition done [foreign].
May 19 01:35:03 PC heartbeat: [17943]:info: Initial resource acquisition complete (auto_failback)
May 19 01:35:03 PC heartbeat: [17943]:info: remote resource transition completed.

测试：

将主201上的心跳关闭

[[email protected] ha_log]# service heartbeat stop

Stopping High-Availability services: Done.

查看日志：

May 19 01:46:57 PC heartbeat: [18561]: info:Giving up all HA resources.
May 19 01:46:58 PCResourceManager(default)[18574]: info: Releasing resource group: pcIPaddr::10.192.203.203 mysql
May 19 01:46:58 PCResourceManager(default)[18574]: info: Running /etc/init.d/mysql  stop
May 19 01:46:59 PC ResourceManager(default)[18574]:info: Running /etc/ha.d/resource.d/IPaddr 10.192.203.203 stop
May 19 01:46:59 PCIPaddr(IPaddr_10.192.203.203)[18652]: INFO: IP status = ok, IP_CIP=
May 19 01:46:59 PC/usr/lib/ocf/resource.d//heartbeat/IPaddr(IPaddr_10.192.203.203)[18638]:INFO:  Success
May 19 01:46:59 PC heartbeat: [18561]:info: All HA resources relinquished.
May 19 01:47:00 PC heartbeat: [17943]:WARN: 1 lost packet(s) for [slave2] [2777:2779]
May 19 01:47:00 PC heartbeat: [17943]:info: No pkts missing from slave2!
May 19 01:47:01 PC heartbeat: [17943]:info: killing HBWRITE process 17949 with signal 15
May 19 01:47:01 PC heartbeat: [17943]:info: killing HBREAD process 17950 with signal 15
May 19 01:47:01 PC heartbeat: [17943]:info: killing HBWRITE process 17951 with signal 15
May 19 01:47:01 PC heartbeat: [17943]:info: killing HBREAD process 17952 with signal 15
May 19 01:47:01 PC heartbeat: [17943]:info: killing HBFIFO process 17946 with signal 15
May 19 01:47:01 PC heartbeat: [17943]:info: killing HBWRITE process 17947 with signal 15
May 19 01:47:01 PC heartbeat: [17943]:info: killing HBREAD process 17948 with signal 15
May 19 01:47:01 PC heartbeat: [17943]:info: Core process 17951 exited. 7 remaining
May 19 01:47:02 PC heartbeat: [17943]:info: Core process 17946 exited. 6 remaining
May 19 01:47:02 PC heartbeat: [17943]:info: Core process 17947 exited. 5 remaining
May 19 01:47:02 PC heartbeat: [17943]:info: Core process 17948 exited. 4 remaining
May 19 01:47:02 PC heartbeat: [17943]:info: Core process 17949 exited. 3 remaining
May 19 01:47:02 PC heartbeat: [17943]:info: Core process 17950 exited. 2 remaining
May 19 01:47:02 PC heartbeat: [17943]:info: Core process 17952 exited. 1 remaining
May 19 01:47:02 PC heartbeat: [17943]:info: pc Heartbeat shutdown complete.

查看从202的日志：

harc(default)[8578]:         2016/05/19_01:47:00 info: Running /etc/ha.d//rc.d/statusstatus
mach_down(default)[8595]:    2016/05/19_01:47:00 info: Taking overresource group IPaddr::10.192.203.203
ResourceManager(default)[8622]: 2016/05/19_01:47:00 info: Acquiring resourcegroup: pc IPaddr::10.192.203.203 mysql
/usr/lib/ocf/resource.d//heartbeat/IPaddr(IPaddr_10.192.203.203)[8650]: 2016/05/19_01:47:01 INFO:  Resource is stopped
ResourceManager(default)[8622]: 2016/05/19_01:47:01 info: Running/etc/ha.d/resource.d/IPaddr 10.192.203.203 start
IPaddr(IPaddr_10.192.203.203)[8746]:  2016/05/19_01:47:01 INFO: Adding inet address10.192.203.203/24 with broadcast address 10.192.203.255 to device eth0
IPaddr(IPaddr_10.192.203.203)[8746]:  2016/05/19_01:47:01 INFO: Bringing device eth0up
IPaddr(IPaddr_10.192.203.203)[8746]:  2016/05/19_01:47:01 INFO:/usr/libexec/heartbeat/send_arp -i 200 -r 5 -p/var/run/resource-agents/send_arp-10.192.203.203 eth0 10.192.203.203 autonot_used not_used
/usr/lib/ocf/resource.d//heartbeat/IPaddr(IPaddr_10.192.203.203)[8732]: 2016/05/19_01:47:01 INFO:  Success
ResourceManager(default)[8622]: 2016/05/19_01:47:02 info: Running/etc/init.d/mysql  start
mach_down(default)[8595]:    2016/05/19_01:47:05 info: /usr/share/heartbeat/mach_down:nice_failback: foreign resources acquired
mach_down(default)[8595]:    2016/05/19_01:47:05 info: mach_down takeovercomplete for node pc.
May 19 01:47:05 slave2 heartbeat: [6536]:info: mach_down takeover complete.
May 19 01:47:31 slave2 heartbeat: [6536]:WARN: node pc: is dead
May 19 01:47:31 slave2 heartbeat: [6536]:info: Dead node pc gave up resources.
May 19 01:47:31 slave2 heartbeat: [6536]:info: Link pc:eth1 dead.

显示202接管成功了。

在202上能看到vip已经漂移过来：

[[email protected] ha_log]# ip addr

1: lo: <LOOPBACK,UP,LOWER_UP> mtu16436 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: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP>mtu 1500 qdisc pfifo_fast state UP qlen 1000
   link/ether 08:00:27:04:05:16 brd ff:ff:ff:ff:ff:ff
   inet 10.192.203.202/24 brd 10.192.203.255 scope global eth0
   inet 10.192.203.203/24 brd 10.192.203.255 scope global secondary eth0
   inet6 fe80::a00:27ff:fe04:516/64 scope link
      valid_lft forever preferred_lft forever
3: eth1:<BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen1000
   link/ether 08:00:27:3a:ec:3c brd ff:ff:ff:ff:ff:ff
   inet 10.0.0.2/24 brd 10.0.0.255 scope global eth1
   inet6 fe80::a00:27ff:fe3a:ec3c/64 scope link
      valid_lft forever preferred_lft forever

201已经没有vip

[[email protected] ha_log]# ip addr

1: lo: <LOOPBACK,UP,LOWER_UP> mtu16436 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: eth0:<BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen1000
    link/ether08:00:27:04:05:16 brd ff:ff:ff:ff:ff:ff
   inet 10.192.203.201/24 brd 10.192.203.255 scope global eth0
   inet6 fe80::a00:27ff:fe04:516/64 scope link tentative dadfailed
      valid_lft forever preferred_lft forever
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP>mtu 1500 qdisc pfifo_fast state UP qlen 1000
   link/ether 08:00:27:3a:ec:3c brd ff:ff:ff:ff:ff:ff
   inet 10.0.0.1/24 brd 10.0.0.255 scope global eth1
   inet6 fe80::a00:27ff:fe3a:ec3c/64 scope link tentative dadfailed
      valid_lft forever preferred_lft forever

四：heartbeat+mysql实现高可用

heartbeat只检测心跳也就是只检测设备是否宕机,不会检测MySQL服务,所以我们同样要有一个脚本来检测MySQL服务,如果mysql服务宕掉，则kill掉heartbeat进程实现故障转移(和nginx+keepalived原理一致),脚本内容如下:

分别在master1，master2上新建检查mysql脚本

vi /root/check_mysql.sh

MYSQL=/usr/local/mysql/bin/mysql
MYSQL_HOST=localhost
MYSQL_USER=root
[email protected] 

$MYSQL -h $MYSQL_HOST -u $MYSQL_USER-p$MYSQL_PASSWORD -e "show status;" >/dev/null 2>&1
#$mysqlclient --host=$host --port=$port--user=$user --password=$password  -e"show databases;" > /dev/null 2>&1
if [ $? == 0 ]
then
   echo " $host mysql login successfully "
   exit 0
else
   #echo " $host mysql login faild"
   /etc/init.d/heartbeat stop
   exit 2
fi

这个脚本待写一些邮件通知的操作。

chmod +x /root/check_mysql.sh

设置成定时任务，每分钟检查一次：

*/1 * * * * /root/check_mysql.sh >>/root/check_mysql.log

关闭当前主的mysql，验证下vip是否漂移到了从。

本篇文章参考了以下文章：

http://www.linuxidc.com/Linux/2011-11/46764.htm

http://www.codesky.net/article/201111/173710.html

http://blog.chinaunix.net/uid-20639775-id-3337481.html

http://www.oschina.net/question/163914_31896

https://www.linuxzen.com/heartbeatshi-xian-mysqlshuang-ji-gao-ke-yong.html

http://www.it165.net/admin/html/201308/1702.html

http://blog.csdn.net/wyzxg/article/details/7741116

时间： 2024-10-11 22:38:54

heartbeat+mysql双主复制实现高可用的相关文章

通过KeepAlived搭建MySQL双主模式的高可用集群系统

企业级MySQL集群具备高可用.可扩展.易管理.低成本的特点.下面将介绍企业环境中经常应用的一个解决方案,即MySQL的双主互备架构,主要设计思路是通过MySQL Replication技术将两台MySQL Server互相将对方作为自己的Master,自己又同时作为对方的Slave来进行复制.这样就实现了高可用构架中的数据同步功能,同时,将采用KeepAlived来实现Mysql的自动failover.在这个构架中,虽然两台MySQL Server互为主从,但同一时刻只有一个MySQL Ser

MySQL双主+keepalived实现高可用

mysql+keepalived实现高可用+主主复制模式 为了解决mysql的单点故障问题,衍生出很多mysql的高可用方案: keepalived+双主.MHA.PXC.MMM.Hearbeat+DRBD等,比较常用的一般是keepalived+双主,MHA和PXC 在此搭建实验环境,实现keepalived+mysql双主模式. 实验思路: 两台MySQL互为主从关系(双主),通过keepalived配置虚拟vip,实现当其中的一台MySQL数据库宕机后,应用能自动切换到另外一台MySQL数

企业级-Mysql双主互备高可用负载均衡架构（基于GTID主从复制模式）

前言: 原理与思想 这里选用GTID主从复制模式Mysql主从复制模式,是为了更加确保主从复制的正确性.健康性与易配性.这里做的是两服务器A,B各有Mysql实例3310,两个实例间互为主从 主从复制模式采用GTID主从复制模式,在服务器A,B上配置keepalived负载均衡,通过VIP连接数据库,目的是一旦有某数据库宕机,keepalived 就会立即建VIP执行另外一台 健康的数据库实例上,实现快速切换,避免单点故障,从而保证业务的正常运行. 这里只做了 双主+keepalived  ,

Heartbeat MySQL双主复制

目录 一 基础环境 二 实际部署 2.1 安装MySQL 2.2 初始化MySQL 2.3 master01 my.cf配置 2.4 创建账号 2.5 master02 my.cf配置配置 2.6 创建账号 三 启动主从 3.1 手动同步 3.2 启动Master01的slave功能 3.3 启动Master02的slave功能 四 安装Heartbeat 五 配置Heartbeat 5.1 配置authkeys 5.2 配置Heartbeat 5.3 配置haresources 六 验证Hea

keepalived+mysql双主复制高可用方案

MySQL双主复制,即互为Master-Slave(只有一个Master提供写操作),可以实现数据库服务器的热备,但是一个Master宕机后不能实现动态切换.而Keepalived通过虚拟IP,实现了双主对外的统一接口以及自动检查.失败切换机制.联合使用,可以实现MySQL数据库的高可用方案. 实验环境:OS:centos 6.x x86_64系统MySQL版本: :mysql 5.6.22   64 位A: master :192.168.79.3 3306B: slave :192.168.

mysql+keepalived 双主热备高可用

理论介绍:我们通常说的双机热备是指两台机器都在运行,但并不是两台机器都同时在提供服务.当提供服务的一台出现故障的时候,另外一台会马上自动接管并且提供服务,而且切换的时间非常短.MySQL双主复制,即互为Master-Slave(只有一个Master提供写操作),可以实现数据库服务器的热备,但是一个Master宕机后不能实现动态切换.使用Keepalived,可以通过虚拟IP,实现双主对外的统一接口以及自动检查.失败切换机制,从而实现MySQL数据库的高可用方案. 环境说明:版本 version

Mysql+Keepalived双主热备高可用操作步骤详细解析

mysql+keepalived双主热备高可用的介绍: 我们通常说的双机热备是指两台机器都在运行,但并不是两台机器都同时在提供服务.当提供服务的一台出现故障的时候,另外一台会马上自动接管并且提供服务,而且切换的时间非常短.MySQL双主复制,即互为Master-Slave(只有一个Master提供写操作),可以实现数据库服务器的热备,但是一个Master宕机后不能实现动态切换.使用Keepalived,可以通过虚拟IP,实现双主对外的统一接口以及自动检查.失败切换机制,从而实现MySQL数据库的

Mysql+keeoalived双主热备高可用操作记录

我们通常说的双机热备是指两台机器都在运行,但并不是两台机器都同时在提供服务.当提供服务的一台出现故障的时候,另外一台会马上自动接管并且提供服务,而且切换的时间非常短.之前梳理了Mysql主从同步,下面说下Mysql+keeoalived双主热备高可用方案的实施. 1)Keepalived的工作原理是VRRP(Virtual Router Redundancy Protocol)虚拟路由冗余协议.在VRRP中有两组重要的概念:VRRP路由器和虚拟路由器,主控路由器和备份路由器. 2)VRRP路由器

Mysql+Keepalived双主热备高可用操作记录

环境: ubuntu18.04.2 mysql5.7.21 1 #1)安装keepalived并将其配置成系统服务.master1和master2两台机器上同样进行如下操作: 2 apt-get install libssl-dev 3 apt-get install openssl 4 apt-get install libpopt-dev 5 [[email protected] ~]# cd /usr/local/src/ 6 [[email protected] src]# wget h